Chapter R.

11: PBT/vPvB assessment

1 Version 4.0 – December 2023

GUIDAN CE

Guidance on Information Requirements

and Chemical Safety Assessment

Chapter R.11: PBT/vPvB assessment

Version 4.0

December 2023
 Chapter R.11: PBT/vPvB assessment

2 Version 4.0 – December 2023

Legal notice

This document aims to assist users in complying with their obligations under the REACH
Regulation. However, users are reminded that the text of the REACH Regulation is the only
authentic legal ref erence and that the inf ormation in this document does not constitute legal
advice. Usage of the inf ormation remains under the sole responsibility of the user. The
European Chemicals Agency does not accept any liability with regard to the use that may be
made of the inf ormation contained in this document.

Guidance on Information Requirements and Chemical Safety Assessment

Chapter R.11: PBT/vPvB Assessment

Reference: ECHA-23-H-10-EN
Cat. Number: ED-02-23-137-EN-N
ISBN: 978-92-9468-326-7
DOI: 10.2823/312974
Publication date: December 2023
Language: EN

© European Chemicals Agency, 2023

If you have questions or comments in relation to this document please send them (quote the
ref erence and issue date) using the inf ormation request f orm. The inf ormation request f orm
can be accessed via the Contact ECHA page at: [Link]

European Chemicals Agency

Mailing address: P.O. Box 400, FI-00121 Helsinki, Finland

Visiting address: Telakkakatu 6, Helsinki, Finland
 Chapter R.11: PBT/vPvB assessment

3 Version 4.0 – December 2023

Preface

This document describes the inf ormation requirements under the REACH Regulation with
regard to substance properties, exposure, use and risk management measures, and the
chemical saf ety assessment. It is part of a series of guidance documents that are aimed to
help all stakeholders with their preparation f or f ulf illing their obligations under the REACH
Regulation. These documents cover detailed guidance f or a range of essential REACH
processes as well as f or some specif ic scientif ic and/or technical methods that industry or
authorities need to make use of under the REACH Regulation.

The original versions of the guidance documents were drafted and discussed within the REACH
Implementation Projects (RIPs) led by the European Commission services, involving
stakeholders f rom Member States, industry and non-governmental organisations. Af ter
acceptance by the Member States competent authorities the guidance documents had been
handed over to ECHA f or publication and f urther maintenance. Any updates of the guidance
are draf ted by ECHA and are then subject to a consultation procedure, involving stakeholders
f rom Member States, industry and non-governmental organisations. For details of the
consultation procedure, please see:

[Link]

Consultation procedure f or Guidance [PDF]

The guidance documents can be obtained via the website of the European Che micals Agency
at:

[Link]

Further guidance documents will be published on this website when they are f ina lised or
updated.

This document relates to the REACH Regulation (EC) No 1907/2006 of the European
1
Parliament and of the Council of 18 December 2006 .

1
Regulation (EC) No 1907/2006 of the European Parliament and of the Council of 18 December 2006 concerning the
Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH), establishing a European Chemicals
Agency, amending Directive 1999/45/EC and repealing Council Regulation (EEC) No 793/93 and Commission
Regulation (EC) No 1488/94 as well as Council Directive 76/769/EEC and Commission Directives 91/155/EEC,
93/67/EEC, 93/105/EC and 2000/21/EC (OJ L 396, 30.12.2006, p.1; corrected by OJ L 136, 29.5.2007, p.3).
 Chapter R.11: PBT/vPvB assessment

4 Version 4.0 – December 2023

Document History

Version Changes Date

Version 1 First edition May 2008

Version 1.2 Corrigendum: November 2012

(i) replacing references to DSD/DPD by references

to CLP;

(ii) further minor editorial changes/corrections.

Version 2.0 Second edition. Full revision of this document was November 2014
necessary to take into account the amendment of
Annex XIII to REACH (according to Commission
Regulation (EU) No 253/2011 of 15 March 2011, OJ
L 69 7 16.3.2011). Main changes in the guidance
document include the following:

• Chapter R.11 title has been changed to

“PBT/vPvB assessment”;

• Chapter R.11 has been re-structured to

differentiate more clearly between the
obligations of the registrant arising directly
from the legal text (Section R.11.3) and the
description of the scientific method to
assess PBT/vPvB properties (Section
R.11.4);

• Description of the registrant’s obligations in

Section R.11.3 has been expanded upon;

• The description of the scope of PBT/vPvB

assessment regarding relevant
constituents/impurities/additives and
transformation/degradation products has
been expanded upon and divided into two
Sections: Section R.[Link] for legal
aspects and Section R.11.4 for the aspects
related to assessment;

• The different steps of the PBT/vPvB

assessment process, in particular the first
step of comparison with the PBT and vPvB
criteria, and the subsequent conclusions
and consequences for the registrant have
been refined to take account of the case
where the registrant concludes that further
information is needed but he decides not to
generate additional information by
 Chapter R.11: PBT/vPvB assessment

5 Version 4.0 – December 2023

considering the substance “as if it is a

PBT/vPvB”;

• The number of conclusions deriving from

the first Step of the PBT/vPvB assessment
process has been reduced from four to
three in Section R.[Link] “Conclusions on
PBT or vPvB properties”;

• Consequences for the registrant of the

conclusions deriving from the first Step of
the PBT/vPvB assessment process are
described in the new Section R.11.3.2.

• Section R.[Link] is new and describes the

situation of substances concluded as being
PBT/vPvB by ECHA’s Member State
Committee in relation to the inclusion in the
Candidate List of Substances of Very High
Concern;

• The basic approach to bioaccumulation

assessment described in Section R.[Link]
has been slightly extended to reflect in
particular the revised OECD test guideline
305 and the possibility to take other
bioaccumulation information into account.
The molecular length screening threshold
value has been removed;

• As the screening threshold values for

PBT/vPvB assessment are part of the
scientific methodology and not part of legal
text, they are now presented in relevant
parts of Section R.11.4 only.

• The document has been re-formatted to

ECHA new corporate identity.

Version 3.0 Revision of this document was necessary in sections June 2017
related to the scientific assessment approach to
take into account recent scientific and technical
developments in the field, including recently
adopted or revised OECD TGs. Main changes in the
guidance document include the following:

• Update of Section R.[Link] on

“Persistence assessment” and of the
corresponding Integrated Testing Strategy
described in Section R.[Link].1 and
Figure R.11-3;

• Update of Section R.[Link] on

“Bioaccumulation assessment” and of the
corresponding Integrated Testing Strategy
described in Section R.[Link].1 and
Figure R.11-4;
 Chapter R.11: PBT/vPvB assessment

6 Version 4.0 – December 2023

• Update of Section R.[Link] on

“Conclusions on PBT or vPvB properties”;

• Revision of Section R.[Link] on

“Assessment of substances containing
multiple constituents, impurities and/or
additives”;

• Update of Appendix R.11—3 on “PBT

assessment of UVCB petroleum
substances”;

• Update of cross references and links to the

revised sections of Chapters R.7b and R.7c.

Version 4.0 Revision of this document was necessary in sections December 2023
related to the scientific assessment approach to
take into account recent scientific and technical
developments in the field, including recently
adopted or revised OECD TGs. Main changes in the
guidance document include the following:

• Update of the R.[Link] to provide

clarification on the procedure for the
inclusion of substances in the Candidate List
of Substances of Very High Concern.

• Revision of the R.11.4.1 on “Standard

approach” including the Weight-of-Evidence
determination, Benchmarking and Reporting
of the analytical methods as part of the
environmental fate studies.

• R.[Link] on “Persistence assessment” and

of the corresponding Integrated Testing
Strategy described in Section R.[Link].1
and Figure R.11-3

• Update of the R.[Link].3 regarding the

Degradation half-life (DegT50) derivation

• Update of Section R.[Link] on

“Bioaccumulation assessment” and of the
corresponding Integrated Testing Strategy
described in Section R.[Link].1 and
Figure R.11-4;

• Introduction in section R.[Link].2 of text

regarding the use of aquatic invertebrate
tests (Hyalella azteca bioconcentration test
(HYBIT))

• Introduction of new section R.[Link].4 on

In vitro biotransformation data and in vitro-
in vivo extrapolation (IVIVE)
 Chapter R.11: PBT/vPvB assessment

7 Version 4.0 – December 2023

• Introduction of new section R.[Link].8

regarding the assessment of
bioaccumulation in air-breathing organisms

• Introduction of the new section

R.[Link].3 regarding volatile substances.

• Introduction of the new section

R.[Link].4 regarding substances with
nanoforms.

• Update of the R.[Link].3 regarding the

toxicity assessment for UVCBs.

• Deletion of the section of the Appendix

R.11—1 regarding Indicators for limited
bioconcentration for PBT assessment

• Deletion of the Appendix R.11—1 Annex 3

(Examples-Use of the indicators for limited
bioaccumulation)

• Deletion of the Appendix R.11—2

(Assessment of substances requiring special
consideration during Testing)

• Introduction of the new Appendix R.11-4

(Approach on NER quantification and
characterisation in persistence assessment)

• Introduction of the new Appendix R.11- 5

(Comparison of HPLC and KowWIN v1.68
QSAR generated log K ow values)

• Introduction of the new Appendix R.11- 6

(Relevance and scientific justification of
correction for growth dilution when deriving
BCF)

• Introduction of the new Appendix R.11-7

(Volatilisation correction approaches for the
kinetic analysis of simulation studies)

• Update of cross references and links to the

revised sections of Chapters R.7b, R.7c and
the Appendices for substance in nanoform.
 Chapter R.11: PBT/vPvB assessment

8 Version 4.0 – December 2023

Convention for citing the REACH Regulation

Where the REACH Regulation is cited literally, this is indicated by text in italics between
quotes, or text in blue boxes.

Table of Terms and Abbreviations

See Chapter R.20.

Pathfinder

The f igure below indicates the location of Chapter R.11 within the Guidance Document:

Information: available - required/needed

Hazard Assessment (HA) Emission Characterisation

Exposure Scenario Building

R11 R11.3.4

No Yes
Article 14(4) Risk Characterisation (RC)
Stop criteria?

Document Yes No Iteration

Risk
in CSR controlled?

Communicate
ES via SDS
 Chapter R.11: PBT/vPvB assessment

9 Version 4.0 – December 2023

Table of Contents

R.11 PBT and vPvB Assessment ........................................................... 14

R.11.1 Introduction.......................................................................................14

R.11.2 Overview of Annex XIII to the REACH Regulation ...................................17

R.11.2.1 Elements and terminology of Annex XIII to the REACH Regulation................ 17
R.11.2.2 PBT and vPvB criteria and inf ormation listed in Annex XIII to the REACH
Regulation .................................................................................................. 20

R.11.3 Duties of the registrant........................................................................23

R.11.3.1 Objective and overview of the PBT/vPvB assessment process ...................... 23
R.11.3.2 Comparison with the criteria (Step 1) ................................................... 26
R.[Link] Scope of the PBT and vPvB assessment (relevant constituents,
transformation/degradation products) ............................................................... 27
R.[Link] Specif ic cases: substances f ulf illing the PBT/vPvB criteria in relation to the
inclusion of substances in the Candidate List of Substances of Very High Concern ........ 28
R.11.3.3 Consequences of Step 1 .................................................................... 29
R.[Link] No consequences......................................................................... 29
R.[Link] Conduct emission characterisation and risk characterisation ................... 29
R.[Link] Generate relevant additional inf ormation (including, where relevant,
submission of a testing proposal)..................................................................... 30
R.[Link] Treat the substance “as if it is a PBT or vPvB” ..................................... 31
R.11.3.4 Emission characterisation, risk characterisation and risk management measures .
.................................................................................................. 32
R.[Link] Emission characterisation .............................................................. 33
R.[Link] Risk characterisation and risk management measures f or “PBT or vPvB
Substances” .............................................................................................. 34
R.[Link].1 Options and measures to minimise emissions and exposure .............. 35
R.[Link].2 Risk Characterisation f or humans in cases of direct exposure to “PBT or
vPvB substances” ..................................................................................... 36
R.11.3.5 Documentation of the PBT/vPvB assessment........................................... 37
R.11.3.6 Documentation of the risk characterisation and communication of measures ... 38

R.11.4 Assessment of PBT/vPvB properties – the scientific method.....................40

R.11.4.1 Standard approach .......................................................................... 40
R.[Link] Persistence assessment (P and vP)................................................... 50
R.[Link].1 Integrated assessment and testing strategy (ITS) f or persistence
assessment ......................................................................................... 50
R.[Link].2 Introduction to persistence assessment ....................................... 58
R.[Link].3 Test data on biodegradation ..................................................... 61
R.[Link].4 Assessment based on estimation models (QSAR, SAR)..................... 84
R.[Link].5 Field studies for persistence...................................................... 88
R.[Link].6 Monitoring data ..................................................................... 89
R.[Link] Bioaccumulation assessment (B and vB) ............................................ 90
R.[Link].1 Integrated Assessment and Testing Strategy (ITS) ......................... 90
 Chapter R.11: PBT/vPvB assessment

10 Version 4.0 – December 2023

R.[Link].2 Experimental aquatic bioconcentration factor (BCF) data .................. 96

R.[Link].3 Dietary Exposure Bioaccumulation Fish Test................................ 103
R.[Link].4 In vitro biotransformation data and in vitro-in vivo extrapolation...... 107
R.[Link].5 Experimental sediment and soil bioaccumulation data.................... 109
R.[Link].6 Field data and biomagnif ication ............................................... 113
R.[Link].7 Addressing uncertainty of f ield data in bioaccumulation screening and
assessment ....................................................................................... 118
R.[Link].8 Bioaccumulation in air-breathing organisms ................................ 122
R.[Link].9 Other testing data ................................................................ 127
R.[Link].10 Further data...................................................................... 128
R.[Link].11 Use of a fugacity approach for bioaccumulation assessment .......... 131
R.[Link] Toxicity assessment (T) .............................................................. 132
R.[Link].1 Integrated testing and assessment strategy (ITS) f or T-testing in support
of PBT assessment for the aquatic environment .............................................. 132
R.[Link].2 The toxicity criterion ............................................................. 135
R.[Link].3 Use of QSAR data ................................................................ 137
R.[Link].4 Screening information and screening threshold values ................... 138
R.[Link].5 Water accommodated fraction (WAF) ........................................ 139
R.[Link].6 Use of non-testing data ......................................................... 140
R.[Link] Conclusions on PBT or vPvB properties ............................................ 141
R.[Link].1 (i) The substance does not f ulf il the PBT and vPvB criteria. The available
inf ormation show that the properties of the substance do not meet the specif ic criteria
provided in REACH Annex XIII Section 1, or if the inf ormation does not allow a direct
comparison with all the criteria there is no indication of P or B properties based on
screening information or other information. ................................................... 142
R.[Link].2 (ii) The substance f ulf ils the PBT and/or vPvB criteria. The available
inf ormation show that the properties of the substance meet the specific criteria detailed
in REACH Annex XIII Section 1 based on a Weight-of -Evidence determination using
expert judgement comparing all relevant and available inf ormation listed in Section 3.2
of Annex XIII to REACH with the criteria (f or more specif ic terminology, also used in
IUCLID, please, see subsection “Terminology”)............................................... 143
R.[Link].3 (iii) The available inf ormation does not allow to conclude (i) or (ii). The
substance may have PBT or vPvB properties. Further inf ormation f or the PBT/vPvB
assessment is needed. ............................................................................. 144
R.11.4.2 Assessment of PBT/vPvB properties – consideration of specif ic substance
properties ................................................................................................ 147
R.[Link] Assessment of substances requiring special considerations with regard to
testing ............................................................................................ 147
R.[Link].1 Substances with very high sorptivity ......................................... 147
R.[Link].2 Substances with low solubility in octanol and water....................... 149
R.[Link].3 Volatile substances ............................................................... 151
R.[Link].4 Substances with nanoforms .................................................... 155
R.[Link] Assessment of substances containing multiple constituents, impurities and/or
additives ............................................................................................ 157
R.[Link].1 Initial prof iling of the substance composition ............................... 158
R.[Link].2 Assessment approaches......................................................... 159
R.[Link].3 Specif ic aspects ................................................................... 166
R.[Link].4 Test items.......................................................................... 169
 Chapter R.11: PBT/vPvB assessment

11 Version 4.0 – December 2023

R.11.5 References ....................................................................................... 170

Table of Figures
Figure R.11—1: Overview of the conclusions f rom Step 1 (“Comparison with the criteria”)
and their consequences. ...................................................... 24
Figure R.11—2: Overview of the PBT/vPvB assessment process for the registrant........ 25
Figure R.11—3: Integrated Assessment and Testing Strategy f or persistence assessment –
maximising data use and targeting testing. .............................. 51
Figure R.11—4: Integrated assessment and testing strategy for B-assessment. .......... 94
Figure R.11—5: Schematic overview on bioaccumulation metrics: laboratory
bioconcentration f actor (lab BCF), laboratory biomagnification f actor (lab
BMF), f ield biomagnif ication f actor (f ield BMF), bioaccumulation f actor
(BAF) ref erenced to water, biota sediment or soil accumulation f actor
(BSAF), and trophic magnification factor (TMF). TMF is obtained from the
slope of a plot of concentration versus trophic level. Please note that lipid
content of the organisms need to be considered in the assessment for
substances that partition into lipids. For non-lipophilic substances,
another ref erence parameter than lipid content may be considered for
normalisation, e.g. dry weight or protein content. Red bullets ref er to
chemicals, green lines ref er to sediment and orange lines ref er to soil.
................................................................................... 117
Figure R.11—6: T testing in support of PBT assessment for the aquatic environment. . 133
Figure R.11—7: Example of the f irst assessment tier of a UVCB substance f or which f raction
prof iling has been applied .................................................. 163
Figure R.11—8: Log BCF v calculated Log Kow. ................................................ 187
Figure R.11—9: LogBCF v measured log Kow. .................................................. 188
Figure R.11—10: LogBCF derived from feeding studies versus calculated Log Kow. .... 189
Figure R.11—11: Relationship between lipid and organic carbon normalised BSAF values and
Log Kow as indicator for the B and vB criterion. ....................... 198
Figure R. 11—12: Approach for the quantification of the total extractable and non-extractable
residues (NER) and the characterisation of the NER types in
environmental matrices (water/soil/sediment) (adapted f rom Kästner et
al. 2018). ...................................................................... 201
Figure R. 11—13: Assessment scheme f or Step 1: Quantif ication of the total extractable and
non-extractable residues (NER) in environmental matrices
(water/soil/sediment). ...................................................... 204
Figure R. 11—14: Assessment scheme for Step 2: Quantification and characterisation of Type
I NER by Silylation or EDTA extraction of the matrix. ................ 206
Figure R. 11—15: Assessment scheme for Step 3: Quantification of parent substance in Type
I NER............................................................................ 209
Figure R. 11—16: Predictivity of the OECD TG 117 HPLC method f or generating log Kow
estimates using 450 substances with HPLC generated log Kow values in
IUCLID (January 2021) f or which experimental data on log Kow is also
available. The linear regression equation is given on top with its R2 value
as a measure of the variance explained, and plotted as the dotted orange
line. ............................................................................. 213
 Chapter R.11: PBT/vPvB assessment

12 Version 4.0 – December 2023

Figure R. 11—17: Predictivity of the KowWIN v1.68 QSAR estimating log Kow f or the 450
substances for which HPLC-derived log Kow values were given in IUCLID,
and f or which also experimentally determined log Kow values were
available. The linear regression equation is given on top with its R2 value
as a measure of the variance explained and plotted as the dotted blue
line. ............................................................................. 214
Figure R. 11—18: HPLC generated log Kow estimates in IUCLID f or which no experimental
data was available, compared to the KowWIN v1.68 estimate of log Kow.
The linear regression equation is given on top with its R2 value as a
measure of the variance explained and plotted as the dotted blue line.
................................................................................... 215
Figure R. 11—19: Relationships of oxygen consumption and feeding rate with growth rate,
based on data presented in Gobas and Lee (2019). .................. 219
Figure R. 11—20: Compartment model f or the parent in total water-sediment system
(C_TOT) including volatilisation (C_VOL) and sink (elimination
compartment) implemented in ModelMaker (Figure [Link]-6 in the
Annex I of the CLP report on EC 254-938-2). .......................... 224
Figure R. 11—21: Structure of the model used in the CAKE tool. Parent: concentration of
parent substance measured in soil; A1: sum of extractable
transf ormation/degradation products; B1: volatilised parent substance.
................................................................................... 225

Tables
Table R.11—1: PBT and vPvB criteria according to Section 1 of Annex XIII to the REACH
Regulation. ...................................................................... 20
Table R.11—2: Screening inf ormation as listed in Section 3.1 of Annex XIII to the REACH
Regulation. ...................................................................... 21
Table R.11—3: Assessment information according to Section 3.2 of Annex XIII to the REACH
Regulation. ...................................................................... 22
Table R.11—4: Screening information for P and vP. ............................................. 60
Table R.11—5: Persistence (P/vP) criteria according to Annex XIII to the REACH Regulation
and related simulation tests.................................................. 61
Table R.11—6: Scenarios f or deriving DegT50 for substances with incomplete mass balance.
..................................................................................... 69
Table R.11—7: Screening threshold values for toxicity........................................ 138
Table R.11—8: Solubility of some pigments and comparison of their Co/Cw values with
estimated Kows (US EPA, 2000b)......................................... 150
 Chapter R.11: PBT/vPvB assessment

13 Version 4.0 – December 2023

Appendices
Appendix R.11-1.1: Development of a log Kow cut-of f value for the B-criterion in the PBT-
assessment .................................................................... 187
Appendix R.11-1.2:Graphic def initions for the molecular dimensions...................... 190
Appendix R.11-2: PBT assessment of UVCB petroleum substances....................... 193
Appendix R.11-3: Bioconcentration studies with benthic and terrestrial invertebrate species
(BSAF).......................................................................... 198
Appendix R.11-4: Approach on non-extractable residues (NER) quantif ication and
characterisation in persistence assessment............................. 200
Appendix R.11-5: Comparison of HPLC and KOWWIN v1.68 QSAR generated log Kow values
................................................................................... 212
Appendix R.11-6: Relevance and scientif ic justif ication of correction f or growth dilution
when deriving BCF ........................................................... 217
Appendix R.11-7: Volatilisation correction approaches f or the kinetic analysis of simulation
studies.......................................................................... 222
 Chapter R.11: PBT/vPvB assessment

14 Version 4.0 – December 2023

R.11 PBT and vPvB Assessment

R.11.1 Introduction
According to Section 4 of Annex I to the REACH Regulation the objective of the persistent,
bioaccumulative and toxic (PBT) and very persistent and very bioaccumulative (vPvB)
assessment is to determine if the substance assessed in Chemical Saf ety Assessment (CSA)
f ulf ils the criteria set out in Annex XIII. It f urthermore states that a conventional hazard
assessment of the long-term effects and the estimation of the long-term exposure cannot be
carried out with suf f icient reliability f or the purpose of assessing the saf ety of substances
satisf ying the PBT and vPvB criteria in Annex XIII. Theref ore a PBT and vPvB assessment is
required to be carried out f or all substances f or which CSA is carried out.

This guidance document contains a description of scientif ic principles f or the PBT and vPvB
assessment in accordance with Section 4 of Annex I to the REACH Regulation, and a
description of the obligations of the registrant in carrying out a PBT and vPvB assessment as
part of CSA.

PBT substances are substances that are persistent, bioaccumulative and toxic, while vPvB
substances are characterised by a particular high persistence in combination with a high
tendency to bioaccumulate, which may, based on experience f rom the past with such
substances, lead to toxic ef fects and have an impact in a manner which is dif f icult to predict
and prove by testing, regardless of whether there are specif ic ef f ects already known or not.
These properties are defined by the criteria laid down in Section 1 of Annex XIII to the REACH
Regulation (CRITERIA FOR THE IDENTIFICATION OF PERSISTENT, BIOACCUMULATIVE AND
TOXIC SUBSTANCES, AND VERY PERSISTENT AND VERY BIOACCUMULATIVE SUBSTANCES,
hencef orth “the PBT and vPvB criteria”).
2
A PBT/vPvB assessment is required for all substances f or which a CSA must be conducted and
reported in the chemical saf ety report (CSR). These are, according to Article 14(1) of the
REACH Regulation, in general all substances manuf actured or imported in amounts of 10 or
more tonnes per year that are not exempted f rom the registration requirement under the
Regulation. However, some f urther exemptions apply as described in Article 14(2), e.g. f or
substances present in a mixture if the concentration is less than 0.1% weight by weight (w/w),
f or on-site or transported isolated intermediates, and f or substances used f or Product and
Process Oriented Research and Development (f or f urther inf ormation see the Guidance on
Registration). Theref ore, this guidance is mainly targeted at registrants manuf acturing or
importing a substance in amounts of 10 or more tonnes per year and to downstream users
who have an obligation to conduct their own CSA. This guidance is also relevant f or ECHA and
f or Member State competent authorities who carry out PBT/vPvB assessment related tasks
under REACH.

Experience with PBT/vPvB substances has shown that they can give rise to specif ic concerns
that may arise due to their potential to accumulate in parts of the environment and

2
The term “PBT/vPvB assessment” is applied in this document to denote “PBT and vPvB assessment” and co vers both
“screening” and “assessment” as described in the following sections.
 Chapter R.11: PBT/vPvB assessment

15 Version 4.0 – December 2023

• that the ef f ects of such accumulation are unpredictable in the long-term;

• such accumulation is in practice dif f icult to reverse as cessation of emission will not
necessarily result in a reduction in substance concentration.

Furthermore, PBT or vPvB substances may have the potential to contaminate remote areas
that should be protected from f urther contamination by hazardous substances resulting f rom
human activity because the intrinsic value of pristine environments should be protected.

These specific concerns occur particularly with substances that can be shown both to persist
f or long periods and to bioaccumulate in biota and which can give rise to toxic ef f ects af ter a
longer time and over a greater spatial scale than substances without these properties. These
ef fects may be dif ficult to detect at an early stage because of long-term exposures at normally
low concentration levels and long lif e-cycles of species at the top of the f ood chain. In the case
of vPvB substances, there is concern that even if no toxicity is demonstrated in laboratory
testing, long-term ef f ects might be possible since high but unpredictable levels may be
reached in man or the environment over extended time periods.

The properties of the PBT/vPvB substances lead to an increased uncertainty in the estimation
of risk to human health and the environment when applying quantitative risk assessment
methodologies. For PBT and vPvB substances a “safe” concentration in the environment cannot
be established using the methods currently available with suff icient reliability f or an acceptable
3
risk to be determined in a quantitative way . Theref ore, a separate PBT/vPvB assessment is
required according to Article 14(3)(d) of the REACH Regulation in order to take these specif ic
concerns into account. Registrants are required to perf orm this specific PBT/vPvB assessment in
the context of their CSA.

According to Section 4 of Annex I to the REACH Regulation, the objective of the PBT/vPvB
assessment is to determine if the substance f ulf ils the criteria given in Annex XIII to the
REACH Regulation (“Step 1: Comparison with the Criteria”), and if so, to characterise the
potential emissions of the substance to the dif f erent environmental compartments during all
activities carried out by the registrant and all identif ied uses (“Step 2: Emission
characterisation”). In addition, in the latter step it is also necessary to identif y the likely routes
by which humans and the environment are exposed to the substance. According to Section 6.5
of Annex I to the REACH Regulation the registrant then needs to use the inf ormation obtained
during the emission characterisation step, when implementing on his site, and recommending
to downstream users, risk management measures (RMMs) which minimise emissions and
subsequent exposures of humans and the environment throughout the lif e-cycle of the
substance that results f rom manufacture or identif ied uses. The authorities may f urther subject
substances with PBT or vPvB properties to restrictions or the authorisation requirement, with
substitution of the substance as objective in the latter case where economically and technically
viable.

The registrant’s process for assessing the substance and the registrant’s obligations resulting
f rom the conclusions are outlined in detail in Section R.11.3. Guidance on scientif ic methods

3
It should be noted that over the last years a number of methods have been proposed in the scientific literature that
could eventually be used to reduce the uncertainty in the risk estimation (on either the exposure or effects side) of
PBTs and vPvBs and hence may lead to a better understanding of the level of risk associated with these substances, in
particular in a comparative sense.
 Chapter R.11: PBT/vPvB assessment

16 Version 4.0 – December 2023

that can be used f or carrying out Step 1 is given in Section R.11.4 of this Chapter. The sub-
sections of Section R.11.4 on the assessment of the P, B and T properties of a substance
provide guidance on how a registrant or an authority can make best use of the dif f erent types
of inf ormation available in order to conclude with least ef forts on the PBT/vPvB–properties of
the substance. These sub-sections also contain guidance on specif ic assessment and testing
strategies f or substances that are dif ficult to test, including adaptation of tests, specif ic rules
f or interpretation of results, consideration of monitoring data and cut-of f criteria.

The guidance explains how all available evidence can be considered in order to decide with
suf ficient certainty whether the PBT/vPvB criteria are f ulf illed or not without always requiring
the generation of such types of data that numerically match with the Annex XIII criteria.
Generating such data may f or instance not be possible because the properties of the substance
do not permit the respective tests to be conducted. In these cases a conclusion may need to
be drawn on the basis of screening inf ormation and all f urther evidence available. In many
cases f urther inf ormation may need to be generated bef ore it can be judged whether the
substance f ulfils the Annex XIII criteria, and the guidance provides detailed testing strategies
that the registrant should use f or each endpoint in Section R.11.4.

Substances are considered as PBT or vPvB substances when they f ulfil the criteria f or all three
inherent properties P, B and T or both of the inherent properties vP and vB, respectively. It is
the task of the registrant to assess if the inf ormation that is available and/or produced is
suf f icient to assess whether the substance is a PBT or a vPvB substance or not.

It is to be noted that this guidance is not meant to guide authorities directly in identif ying
substances f ulf illing the criteria of Article 57(f ) of the REACH Regulation (substances of
equivalent level of concern). However, this guidance may in such cases be used as one
ref erence f or understanding what indications may be needed to identif y a substance to be of
equivalent level of concern to PBT or vPvB substances.
 Chapter R.11: PBT/vPvB assessment

17 Version 4.0 – December 2023

R.11.2 Overview of Annex XIII to the REACH Regulation

The purpose of this section is to introduce the content and terminology of Annex XIII to the
REACH Regulation. The interpretation of the content is presented mainly f rom Section R.11.3
onwards. Only some key clarif ications of the legal text are included in this section.

R.11.2.1 Elements and terminology of Annex XIII to the REACH Regulation

The introductory section of Annex XIII to the REACH Regulation def ines the PBT/vPvB
assessment scope regarding substance groups:

Introductory Section of Annex XIII to REACH

[…] This Annex shall apply to all organic substances, including organo -metals.

Annex XIII to the REACH Regulation is generally applicable to any substance containing an
organic moiety. Based on the common definition of an organic substance in chemistry, PBT and
vPvB criteria are not applicable to inorganic substances.

The PBT/vPvB criteria as set out in Annex XIII to the REACH Regulation are presented in
Section R.11.2.2, Table R.11—1.

Annex XIII def ines two levels of assessment within the PBT/vPvB assessment (“screening”
and “assessment”) and two sets of inf ormation (“screening information” and “assessment
information”). The two sets of inf ormation are presented in Table R.11—2 and Table R.11—3,
respectively. The differentiation of the two assessment levels within the PBT/vPvB assessment
is mainly designed to help the registrant identify his obligations specif ically with respect to the
PBT/vPvB assessment.

The combination of several passages of extracts of the text of Annex XIII, as cited below,
stipulate that all relevant and available “assessment inf ormation” and “screening
inf ormation” must be used in the PBT/vPvB assessment:

Introductory Section of Annex XIII to REACH

[…] For the identification of PBT substances and vPvB substances a weight-of-evidence determination
using expert judgement shall be applied, by comparing all relevant and available information listed in
Section 3.2 with the criteria set out in Section 1. […]
Section 2.1 of Annex XIII to REACH

For the identification of PBT and vPvB substances in the registration dossier, the registrant shall
consider the information as described in Annex I and in Section 3 of this Annex. […]
Section 2.2 of Annex XIII to REACH

For dossiers for the purposes of identifying substances referred to in Article 57(d) and Article 57(e),
relevant information from the registration dossiers and other available information as described in
Section 3 shall be considered. […]
 Chapter R.11: PBT/vPvB assessment

18 Version 4.0 – December 2023

Recital 5 of Commission Regulation (EU) No 253/2011

Experience shows that, for the adequate identification of PBT and vPvB substances, all relevant
information should be used in an integrated manner and applying a weight-of-evidence approach by
comparing the information to the criteria set out in Section 1 of Annex XIII.

Screening inf ormation cannot be directly (numerically) compared with the PBT/vPvB criteria,
i.e. the screening inf ormation does not contain measured degradation half -lif e values or BCF
values, which could be directly compared with the criteria. Screening inf ormation can inf orm
on whether a substance does not have or potentially has a particular property. Screening
inf ormation involves simple data, typically inf ormation f rom Annex VII to VIII endpoints, that
must be used to assess whether f urther inf ormation is needed. Relevant and available
assessment inf ormation involves data, typically f rom Annex IX and X endpoints, of higher
weight in the Weight-of Evidence. Sections 3.2.1.(d), 3.2.2.(b) and 3.2.3(f ) of Annex XIII to
the REACH Regulation allow also “other inf ormation” to be used as assessment inf ormation,
provided that its suitability and reliability can be reasonably demonstrated.

A Weight-of-Evidence determination by expert judgment must be used in the PBT/vPvB

assessment (see the blue boxes below). It is def ined as f ollows:

Introductory Section of Annex XIII to REACH

[…]
A weight-of-evidence determination means that all available information bearing on the identification
of a PBT or a vPvB substance is considered together, such as the results of monitoring and modelling,
suitable in vitro tests, relevant animal data, infor mation from the application of the category approach
(grouping, read-across), (Q)SAR results, human experience such as occupational data and data from
accident databases, epidemiological and clinical studies and well documented case reports and
observations. The quality and consistency of the data shall be given appropriate weight. The available
results regardless of their individual conclusions shall be assembled together in a single weight -of-
evidence determination. […]

The Weight-of-Evidence determination by expert judgement enables the use of all (screening
and assessment) inf ormation types listed in Section 3 of Annex XIII to the REACH Regulation
in the PBT/vPvB assessment f or comparing with the criteria, although not all of these
inf ormation types can be directly (numerically) compared with the criteria.

Examples and principles of Weight-of-Evidence determination f or the PBT/vPvB assessment

f urther applying the introductory section of Annex XIII to the REACH Regulation are provided
in Section R.11.4. In addition, the Practical Guide on “How to use alternatives to animal testing
to fulfil your information requirements for REACH registration” provides a general scheme f or
building a Weight-of-Evidence approach.

As regards the registrants’ specific duties for the PBT/vPvB assessment, the f ollowing
provision of Annex XIII to the REACH Regulation must be considered f urther to Annex I:
 Chapter R.11: PBT/vPvB assessment

19 Version 4.0 – December 2023

Section 2.1 of Annex XIII to REACH

[…] If the technical dossier contains for one or more endpoints only information as required in
Annexes VII and VIII, the registrant shall consider information relevant for screening for P, B, or T
properties in accordance with Section 3.1 of this Annex. If the result from the screening tests or other
information indicate that the substance may have PBT or vPvB properties, the registrant shall
generate relevant additional information as set out in Section 3.2 of this Annex. In case the
generation of relevant additional information would require information listed in Annexes IX or X, the
registrant shall submit a testing proposal. Where the process and use conditions of the substance
meet the conditions as specified in Section 3.2(b) or (c) of Annex XI the a dditional information may be
omitted, and subsequently the substance is considered as if it is a PBT or vPvB in the registration
dossier. No additional information needs to be generated for the assessment of PBT/vPvB properties if
there is no indication of P or B properties following the result from the screening test or other
information.

When f ulf illing the data requirements of Annexes IX and X to the REACH Regulation,
adaptations according to Column 2 and Annex XI should be applied wherever possible to
minimise testing on animals, which must be only as a last resort under REACH (see REACH
recital 47 and Articles 13(1) and 25 and Step 4 of REACH Annex VI).

In addition, the f ollowing principles must be applied while perf orming a PBT/vPvB
assessment:

Introductory Section of Annex XIII to REACH

[…] The information used for the purposes of assessment of the PBT/vPvB properties shall be based
on data obtained under relevant conditions. […]

By “relevant conditions”, relevant environmental conditions and relevant testing conditions are
generally meant. These are f urther discussed in Section R.11.4.

Introductory Section of Annex XIII to REACH

[…] The identification shall also take account of the PBT/vPvB properties of relevant constituents of a
substance and relevant transformation and/or degradation products. […]

The term “constituent” refers to the main constituents, impurities and additives of substances
of well-defined composition and constituents of UVCB substances as def ined in the Guidance
for identification and naming of substances under REACH and CLP. The implication in terms of
PBT/vPvB assessment requirement f or the registrant is described in Section R.[Link] and
f urther guidance on what should be considered as relevant constituents is provided in
Section R.11.4.1.
 Chapter R.11: PBT/vPvB assessment

20 Version 4.0 – December 2023

R.11.2.2 PBT and vPvB criteria and information listed in Annex XIII to the
REACH Regulation

The f ollowing tables (Table R.11—1, Table R.11—2, and Table R.11—3) summarise the PBT
and vPvB criteria given in accordance with Section 1 of Annex XIII to REACH and the relevant
inf ormation to be used f or the PBT/vPvB assessment as provided in Sections 3.1 and 3.2 of
Annex XIII to the REACH Regulation.

Table R.11—1: PBT and vPvB criteria according to Section 1 of Annex XIII to the
REACH Regulation.

Property PBT criteria vPvB criteria

Persistence A substance fulfils the persistence criterion (P) in A substance fulfils the “very
any of the following situations: persistent” criterion (vP) in
(a) the degradation half-life in marine water is any of the following
higher than 60 days; situations:

(b) the degradation half-life in fresh or estuarine (a) the degradation half-life
water is higher than 40 days; in marine, fresh or estuarine
water is higher than 60 days;
(c) the degradation half-life in marine sediment
is higher than 180 days; (b) the degradation half-life
in marine, fresh or estuarine
(d) the degradation half-life in fresh or estuarine
water sediment is higher than
water sediment is higher than 120 days;
180 days;
(e) the degradation half-life in soil is higher than
(c) the degradation half-life in
120 days.
soil is higher than 180 days.

Bioaccumulation A substance fulfils the bioaccumulation criterion A substance fulfils the “very
(B) when the bioconcentration factor in aquatic bioaccumulative” criterion
species is higher than 2000. (vB) when the
bioconcentration factor in
aquatic species is higher than
5000.

Toxicity* A substance fulfils the toxicity criterion (T) in

any of the following situations**:
(a) the long-term no-observed effect
concentration (NOEC) or EC10 for marine or
freshwater organisms is less than 0.01 mg/L;
(b) the substance meets the criteria for
classification as carcinogenic (category 1A or
1B), germ cell mutagenic (category 1A or 1B),
or toxic for reproduction (category 1A, 1B or 2)
according to Regulation EC No 1272/2008;
(c) there is other evidence of chronic toxicity, as
identified by the substance meeting the criteria
for classification: specific target organ toxicity
after repeated exposure (STOT RE category 1 or
2) according to Regulation EC No 1272/2008.

* EC10 preferred over NOEC (see further explanation in Section R.[Link]).

** Annex I (Part 4) to CLP Regulation (EC) No 1272/2008 was amended to include Classification criteria
for PBT. These criteria include a possibility to conclude a substance as T based on classification as
endocrine disruptor (category 1) for humans or the environment (19 Dec 2022).
 Chapter R.11: PBT/vPvB assessment

21 Version 4.0 – December 2023

Table R.11—2: Screening information as listed in Section 3.1 of Annex XIII to the
REACH Regulation.

Indication of P and vP properties (a) Results from tests on ready biodegradation in accordance with
Section [Link] of Annex VII;
(b) Results from other screening tests (e.g. enhanced ready test,
tests on inherent biodegradability);
(c) Results obtained from biodegradation (Q)SAR models in
accordance with Section 1.3 of Annex XI;
(d) Other information provided that its suitability and reliability
can be reasonable demonstrated.

Indication of B and vB properties (a) Octanol-water partitioning coefficient experimentally

determined in accordance with Section 7.8 of Annex VII to
REACH or estimated by (Q)SAR models in accordance with
Section 1.3 of Annex XI;
(b) Other information provided that its suitability or reliability can
be reasonably demonstrated.

Indication of T properties* (a) Short-term aquatic toxicity in accordance with Section 9.1 of
Annex VII to REACH and Section 9.1.13 of Annex VIII;
(b) Other information provided that its suitability or reliability can
be reasonably demonstrated.

* Acute or short-term aquatic toxicity data are considered to be screening information (Annex XIII,
Section 3.1) and may be used as an indication that the substance may fulfil the T criterion. However,
when acute/short-term aquatic toxicity data show that the substance is very toxic (L(E)C50 < 0.01
mg/L), a definitive conclusion can be drawn that the substance fulfils the T criterion and no further
testing is necessary. Acute data cannot be used for concluding definitively “not T”. If long-term or chronic
aquatic toxicity data are available, a definitive assessment can be made.
 Chapter R.11: PBT/vPvB assessment

22 Version 4.0 – December 2023

Table R.11—3: Assessment information according to Section 3.2 of Annex XIII to the
REACH Regulation.
Assessment of P or vP (a) Results from simulation testing on degradation in surface water;
properties (b) Results from simulation testing on degradation in soil;
(c) Results from simulation testing on degradation in sediment;
(d) Other information, such as information from field studies or monitoring
studies, provided that its suitability and reliability can be reasonably
demonstrated.

Assessment of B or vB (a) Results from a bioconcentration or bioaccumulation study in aquatic

properties* species;
(b) Other information on the bioaccumulation potential provided that its
suitability and reliability can be reasonably demonstrated, such as:
- Results from a bioaccumulation study in terrestrial species;
- Data from scientific analysis of human body fluids or tissues, such
as blood, milk, or fat;
- Detection of elevated levels in biota, in particular in endangered
species or in vulnerable populations, compared to levels in their
surrounding environment;
- Results from a chronic toxicity study on animals;
- Assessment of the toxicokinetic behaviour of the substance;
(c) Information on the ability of the substance to biomagnify in the food
chain, where possible expressed by biomagnification factors or trophic
magnification factors.

Assessment of T (a) Results from long-term toxicity testing on invertebrates as set out in
properties Section 9.1.5 of Annex IX;
(b) Results from long-term toxicity testing on fish as set out in Section
9.1.6 of Annex IX;
(c) Results from growth inhibition study on aquatic plants as set out in
Section 9.1.2 of Annex VII;
(d) The substance meeting the criteria for classification as carcinogenic in
Category 1A and 1B (assigned hazard phrases: H350 or H350i), germ
cell mutagenic in Category 1A or 1B (assigned hazard phrase: H340),
toxic for reproduction in Category 1A, 1B and/or 2 (assigned hazard
phrases: H360,H360F, H360D, H360FD, H360Fd, H360 fD, H361,
H361f, H361d or H361fd), specific target organ toxic after repeated
dose in Category 1 or 2 (assigned hazard phrase: H372 or H373),
according to Regulation EC No 1272/2008;
(e) Results from long-term or reproductive toxicity testing with birds as
set out in Section 9.6.1 of Annex X;
(f) Other information provided that its suitability and reliability can be
reasonably demonstrated.

* At present, there is no guidance on how to apply in the PBT/vPvB assessment the information coming
from:
- data from scientific analysis of human body fluids or tissues, such as blood, milk, or fat; or
- the detection of elevated levels in biota, in particular in endangered species or in vulnerable
populations, compared to levels in their surrounding environment.
Such guidance needs to be developed in the future.
 Chapter R.11: PBT/vPvB assessment

23 Version 4.0 – December 2023

R.11.3 Duties of the registrant

The purpose of this section is to delineate the obligations of the registrant within the PBT/vPvB
assessment workflow. For f urther details, the registrant may ref er to the recommendations
provided in Section R.11.4.

R.11.3.1 Objective and overview of the PBT/vPvB assessment process

Section 4.0.1 of Annex I to the REACH Regulation def ines the objective of the PBT/vPvB
assessment:

Annex I to REACH

[…]
4. PBT AND VPVB ASSESSMENT
4.0. Introduction
4.0.1. The objective of the PBT/vPvB assessment shall be to determine if the substance fulfils the
criteria given in Annex XIII and if so, to characterise the potential emissions of the substance. […]

It f urthermore states that a hazard assessment and exposure assessment f or CSA cannot be
carried out with suf f icient reliability f or substances satisfying the PBT or vPvB criteria and that
theref ore a separate PBT/vPvB assessment is required.

According to Section 4.0.2 of Annex I to the REACH Regulation, the process of the PBT/vPvB
assessment consists of the f ollowing two steps: Step 1: “Comparison with the criteria” and
Step 2: “Emission characterisation”. Section 6.5 of Annex I to the REACH Regulation
requires the registrant to implement f or PBT/vPvB substances risk management measures
which minimise exposures and emission to humans and the environment, throughout the
lif ecycle of the substance that result f rom manufacture and identif ied uses. The obligations of
the registrant f or carrying out the PBT/vPvB assessment are def ined more in detail in Section
2.1 of Annex XIII to the REACH Regulation. In the f ollowing paragraphs the main assessment
steps are described.

Step 1 comprises a scientif ic PBT/vPvB assessment where the relevant available inf ormation
must be compared with the PBT/vPvB criteria (f or detailed guidance on this step, see Section
R.11.4). In Step 1 the registrant must come to one of the conclusions presented in Figure
R.11—1. Each conclusion leads to specif ic consequences, which the registrant must comply
with. The conclusions are described in more detail in Section R.[Link] and consequences in
Section R.11.3.3.
 Chapter R.11: PBT/vPvB assessment

24 Version 4.0 – December 2023

Conclusion (i): The substance does not fulfil the PBT and vPvB criteria.
For screening assessment: there is no indication of P or B properties.

• No consequences for the registrant. The PBT/vPvB assessment stops.

Conclusion (ii): The substance fulfils the PBT or vPvB criteria.

• The registrant must carry out emission characterisation and ensure minimisation of exposures and
emissions throughout the life-cycle of the substance that results from manufacture and identified
uses.

Conclusion (iii): The available information does not allow to conclude

(i) or (ii). The substance may have PBT or vPvB properties.
Further information for the PBT/vPvB assessment is needed.

• The registrant must generate relevant additional information (including, where necessary,
submission of a testing proposal) and carry out Step 1 again, OR
• The registrant must treat the substance as if it is a PBT or vPvB.

Figure R.11—1: Overview of the conclusions from Step 1 (“Comparison with the
criteria”) and their consequences.
The registrant is only allowed to f inalise Step 1 of the assessment process if he is able to reach
4
an unequivocal conclusion on the PBT or vPvB properties (conclusion (i) or conclusion (ii) ).

Conclusion (iii) is an interim conclusion in Step 1. This conclusion triggers the requirement f or
the registrant to generate all necessary additional inf ormation and to continue in Step 1 until
the available inf ormation allows a def initive conclusion. Section 2.1 of Annex XIII to the REACH
Regulation requires inf ormation to be generated by the registrant irrespective of the standard
inf ormation requirements of the registrant. This may require several iterative steps of
acquisition of f urther inf ormation, testing and assessment. Alternatively, the registrant can
decide af ter conclusion (iii) to apply an exemption from the requirement to generate additional
data by considering the substance “as if it is a PBT or vPvB”. This is only allowed if the
registrant applies specific exposure based adaptation conditions as specif ied in Section 3.2(b)
or (c) of Annex XI to the REACH Regulation.

The consequences of each conclusion f or the registrant are described in more detail in Section
R.11.3.3. Figure R.11—2 provides an overview of the PBT/vPvB assessment process of the
registrant as a f lowchart. Relevant constituents, impurities, additives,
transf ormation/degradation products must also be encompassed in this process.

4
Conclusion (i) and (ii) are either based on a) data directly comparable with the PBT/vPvB criteria or b) based on
Weight-of-Evidence expert judgement of information which is not directly (numerically) comparable with the PBT/vPvB
criteria or c) a combination of both situations a) and b).
 Chapter R.11: PBT/vPvB assessment

25 Version 4.0 – December 2023

Step 1: Compare all

relevant and available
information with the
PBT/vPvB criteria

- beyond the standard

Generate further information requirements,
relevant if necessary for the
information PBT/vPvB assessment
Registrant must
(including, where
draw one of the
relevant,
following three
submission of a
conclusions
testing proposal)

(iii) Further
information is
needed

Registrant must
choose one of
the following
two options

If specific exposure-based
adaptation conditions are (ii) PBT/vPvB (i) PBT/vPvB
met1, the substance can be criteria are criteria are not
considered as if it is a fulfilled2 fulfilled
PBT/vPvB

Step 2: Emission The PBT/vPvB

characterisation assessment can be
stopped

Minimise exposures3 and

emissions to humans and the
environment

Communicate the outcome of the

PBT/vPvB assessment and risk
management measures within
the supply chain
1
Please refer to the conditions as specified in Section 3.2(b) or (c) of Annex XI to the REACH Regulation.
2
Normally not applicable if only screening information is available.
3
For further information on exposure minimisation please refer to Section R.[Link].

Figure R.11—2: Overview of the PBT/vPvB assessment process for the registrant.
 Chapter R.11: PBT/vPvB assessment

26 Version 4.0 – December 2023

R.11.3.2 Comparison with the criteria (Step 1)

In the f ollowing Sections the f ormal obligations f or Step 1 (“Comparison with the criteria”) of
the PBT/vPvB assessment are described.

In Step 1 of the PBT/vPvB assessment, the standard inf ormation requirements are f irst applied
by the registrant as described in the Guidance on Information Requirements & Chemical Safety
Assessment (IR&CSA). It should be noted that any data adaptations according to Column 2 of
Annexes VII to X or Annex XI to the REACH Regulation should be justif ied according to the
relevant ECHA documents (e.g. Practical Guides on “How to use and report (Q)SARs” and on
“How to use alternatives to animal testing to fulfil your information requirements for REACH
registration”, and Chapter 5 and Chapter 6 of the Guidance on IR&CSA,). The inf ormation
included in the registration dossier as a result of adaptations of standard inf ormation
requirements and their justif ications are part of the available inf ormation f or the PBT/vPvB
assessment, where relevant. The PBT and vPvB assessment must initially be based on all the
relevant inf ormation available which is as a minimum the inf ormation as listed in Annexes VII
and VIII to the REACH Regulation. This inf ormation normally corresponds to PBT/vPvB
screening inf ormation as listed in Section R.11.2.2.

The registrant must conclude Step 1 by selecting one of the three conclusions presented in
Figure R.11—1 and Figure R.11—2. If conclusion (iii) “The available data inf ormation does not
allow to conclude (i) or (ii)” applies, Step 1 continues af ter the necessary new inf ormation has
been generated (see more details in Section R.11.3.3).

In cases where only screening inf ormation as listed in Section R.11.2.2 is available f or one or
more endpoints, Step 1 of the PBT/vPvB assessment implies f irst that the registrant is not able
to compare the inf ormation directly (numerically) with the PBT/vPvB criteria. Although it might
be theoretically possible to calculate degradation half-life values or BCF values f rom screening
inf ormation, such values must not be directly compared with the criteria. At this stage, the
registrant is required to analyse whether the inf ormation indicates that the substance may
meet the PBT/vPvB criteria, in which case the registrant must draw conclusion (iii) “The
available data inf ormation does not allow to conclude (i) or (ii)”, or whether the inf ormation
shows that there is no indication on P or B properties, in which case the conclusion (i) “The
substance does not f ulf il the PBT and vPvB criteria” applies. In Section R.11.4 several
screening threshold values and conditions f or applying them are described, which the
registrant should consider while drawing a conclusion f or screening. The screening threshold
values are indicative and the registrant must use all relevant pieces of inf ormation on his
substance to justif y his conclusion. Also, where only screening inf ormation is available, the
choice of the conclusion should be based on a Weight-of-Evidence consideration by expert
judgement where all relevant and available data f or all endpoints are considered in
conjunction.

If only screening inf ormation is available, it is normally not possible to conclude (ii) (“The
substance f ulf ils the PBT or vPvB criteria”) due to the uncertainties related to screening
inf ormation. However, if scientif ically justified, it is in principle possible to draw conclusion (ii)
based on screening inf ormation. In Section R.11.4 f ew such exceptional cases are described,
where the registrant may make use of screening inf ormation f or concluding (ii).

The conclusion of Step 1 should be derived by the registrant taking into account also all
aspects as described in Section R.[Link].
 Chapter R.11: PBT/vPvB assessment

27 Version 4.0 – December 2023

The consequences of the individual conclusions to the registrant are described in more detail in
Section R.11.3.3.

R.[Link] Scope of the PBT and vPvB assessment (relevant constituents,

transformation/degradation products)

For the purpose of this Guidance it should be noted that the term “constituent” as mentioned
in Annex XIII to the REACH Regulation ref ers to constituents and impurities of well-def ined
substances, constituents of UVCB substances, and additives to all substances.

The PBT/vPvB assessment must, according to Annex XIII to the REACH Regulation, take
account of the PBT/vPvB properties of relevant constituents and relevant transf ormation
and/or degradation products of organic substances (including organo-metals).

Generally, the PBT/vPvB assessment obligations as described in Sections R.11.3.1 and

R.11.3.2 have to be applied f or relevant constituents, impurities, additives and
transf ormation/degradation products. The registrant cannot stop the PBT/vPvB assessment if
there is not enough inf ormation available to take into account the PBT/vPvB properties of
relevant constituents, impurities, additives and transf ormation/degradation products. This
means that if there is not enough inf ormation available on the PBT/vPvB properties of relevant
constituents, impurities, additives and transformation/degradation products to derive f or the
registrant’s substance either conclusion (i) (“The substance does not f ulf il the PBT and vPvB
criteria”) or conclusion (ii) (”The substance f ulf ils the PBT or vPvB criteria”), the registrant
must generate the necessary f urther inf ormation on the PBT/vPvB properties of the relevant
constituents, impurities, additives and transformation/degradation products until one of these
two def initive conclusions can be achieved. The other option, as provided in Sections R.11.3.1
and R.11.3.3 is to treat the substance “as if it is a PBT or vPvB”.

If the registrant deems as a result of the PBT/vPvB assessment an uncharacterized

constituent, impurity, additive or transf ormation/degradation product relevant f or the
PBT/vPvB assessment, the registrant must characterize its substance identity as required in
the Guidance for identification and naming of substances under REACH and CLP .

The interpretation of the term “relevant” constituent, impurity, additive,

transf ormation/degradation product, is described in Section R.11.4.1. It is recommended that
the registrant f ollows this interpretation in the PBT/vPvB assessment, in def ining which
constituents, impurities, additives, transf ormation or degradation products are relevant.

The registrant must show in the PBT/vPvB assessment that he has taken into account the
relevant constituents, impurities and additives. This is normally possible only if he includes in
the PBT/vPvB assessment appropriate justif ications f or all constituents, impurities and
additives or f or all f ractions/blocks of the substance composition on why these are considered
to be relevant or judged to be not relevant f or the PBT/vPvB assessment, regardless of
5
whether the substance identity of these could be ultimately determined or not . The registrant
may derive such reasoning quantitatively or qualitatively, by using the PBT/vPvB assessment

5
The PBT/vPvB assessment of short-chain chlorinated paraffins (EC 287-476-5) used for the identification of the
substance to the Candidate List is one of the examples where the constituents were not characterized ultimately. See
related Member State Committee SVHC Support Document at [Link]
56a1-4b0c-bb0f-a6c40e74ece2.
 Chapter R.11: PBT/vPvB assessment

28 Version 4.0 – December 2023

principles as described in Section R.11.4. This also applies to the transf ormation/degradation
products. It should be noted that also Section 9.2.3 of Annex IX to the REACH Regulation
requires identif ication of degradation products.

R.[Link] Specific cases: substances fulfilling the PBT/vPvB criteria in relation

to the inclusion of substances in the Candidate List of Substances of Very High
Concern

According to REACH Article 59, ECHA includes substances into the Candidate List of Substances
of Very High Concern (SVHC) if they f ulf il the PBT and/or vPvB criteria. Where ECHA takes the
decision that the substance f ulfils the PBT/ vPvB criteria, the justif ication f or such decision is
f ound in the support document. The support document is published on ECHA’s website. When
comments are received, unanimous agreement f rom ECHA’s Member State Committee (MSC)
is needed bef ore a decision can be taken whether a substance f ulf ils the PBT/vPvB criteria.
Where, due to lack of unanimous agreement at the MSC, the European Commission takes the
decision that a substance f ulf ils the PBT/ vPvB criteria, the justif ication is f ound in the
European Commission decision published in the Of f icial Journal of the EU. Such Commission
decision may ref er to the reasoning contained in the majority opinion of the MSC. So f ar no
such Commission decision has been taken f or the identif ication of a substance as a PBT/ vPvB.
These decisions and the justif ications why these substances meet the PBT/vPvB criteria are
published on ECHA’s website. If a registrant’s substance has been included in the Candidate
List as a PBT/vPvB substance, the registrant must align his PBT/vPvB assessment and
conclusion with the PBT/vPvB assessment which was the basis of the relevant identif ication
decision. This PBT/vPvB assessment is usually reported in a support document of the decision
on inclusion of the substance in the Candidate List and is available on ECHA’s website. In such
cases, it is appropriate to replace in the CSR the documentation of Step (1) of the PBT/vPvB
assessment with a ref erence to the relevant ECHA or European Commission decision. If the
registrant has new inf ormation available which was not referred to in the support document of
the relevant ECHA or European Commission decision, the registrant must include the new
inf ormation in the registration dossier and may reflect his opinion of the relevance of the new
inf ormation to the conclusion in the CSR. If the registrant would in this case present in the CSR
the opinion that the new inf ormation would trigger another conclusion than the one drawn by
ECHA or the European Commission, the registrant is obliged to apply the conclusion of ECHA or
the European Commission in his CSR. In case ECHA’s Committee f or Risk Assessment provides
an opinion to the European Commission recommending the restriction of a substance because
it meets PBT/vPvB criteria, it is strongly recommended that the registrant(s) recognise and
implement the PBT/vPvB status of the substance in their dossiers, minimise releases and
exposures in their activities and inf orm their downstream users about the PBT/vPVB status,
even bef ore the European Commission has taken the restriction decision.

If a registered substance contains a relevant constituent, impurity or additive or

transf orms/degrades to a substance on the Candidate List that meets the PBT and/or vPvB
criteria, the registrant must conclude that its substance meets the PBT or vPvB criteria
accordingly. To help the registrant, Section R.11.4 provides def initions on what are relevant
constituents, impurities, additives and relevant transf ormation and degradation products.

There are several substances on the Candidate List which have been identif ied as f ulf illing PBT
or vPvB criteria because their constituents or transformation/degradation products fulfil PBT or
 Chapter R.11: PBT/vPvB assessment

29 Version 4.0 – December 2023

6
vPvB criteria . The support documents of ECHA decisions on the Candidate List inclusion
identif y in these cases the constituents or transformation/degradation products of concern and
contain a PBT/vPvB assessment of them. If a registered substance contains one of these as
constituent, impurity, additive, or transf orms/degrades into one of these substances, the
registrant should ref lect the conclusion presented in such support documents in his own
PBT/vPvB assessment. This applies by analogy also to any f uture cases where inclusion to the
Candidate List was due to PBT/vPvB properties of impurities or additives.

R.11.3.3 Consequences of Step 1

The three conclusions f rom Step 1: “Comparison with the criteria” trigger f our dif f erent
consequences f or the registrant (see Figure R.11—1 and Figure R.11—2). These are:

• No consequences: af ter conclusion (i)

• Conduct emission characterisation and risk characterisation: af ter conclusion (ii)

• Generate relevant additional inf ormation (including, where relevant, submission of

testing proposal) and continue under Step 1: af ter conclusion (iii) or Treat the
substance “as if it is a PBT or vPvB”: af ter conclusion (iii)

In the f ollowing sections the consequences are described more in detail.

R.[Link] No consequences

If the registrant concludes (i): The substance does not fulfil the PBT and vPvB criteria,
this is the end of the PBT/vPvB assessment process. In this case, the general obligation of
REACH Article 22 to take into account relevant new inf ormation or relevant changes in the
substance composition applies f or triggering the need to revise the PBT/vPvB assessment.

R.[Link] Conduct emission characterisation and risk characterisation

If the registrant concludes (ii): The substance fulfils the PBT or vPvB criteria, he must
carry out an emission characterisation and implement and recommend such risk management
measures which minimise emissions and subsequent exposures of humans and the
environment f rom manuf acture and identif ied uses (see Section R.11.3.4).

Also substances concluded according to the principles described in Section R.[Link] as

f ulf illing PBT or vPvB criteria because their constituents, impurities, additives or
transf ormation/degradation products f ulf il the PBT or vPvB criteria must be subjected to
emission characterisation and minimisation of releases f or their whole lif e-cycle.

It should be noted that if the registrant draws this conclusion within his CSA, it does not
automatically lead to initiation of the REACH Article 59 process for inclusion of the substance in
the Candidate List but the registrant has the primary responsibility to implement the necessary
risk management measures f or minimisation of the exposure and emissions.

6
Such substances are for example: Coal tar pitch, high temperature (EINECS No: 266-028-2) and
Bis(pentabromophenyl) ether (EC 214-604-9).
 Chapter R.11: PBT/vPvB assessment

30 Version 4.0 – December 2023

R.[Link] Generate relevant additional information (including, where relevant,

submission of a testing proposal)

If the registrant concludes (iii): The available information does not allow to conclude (i)
or (ii), the registrant must generate relevant additional inf ormation and continue the
PBT/vPvB assessment Step 1 until the comparison with the criteria can be reliably done and a
f inal conclusion (i) “The substance does not f ulf il the PBT and vPvB criteria” or (ii) “The
substance f ulfils the PBT or vPvB criteria” can be unequivocally drawn (see Figure R.11—1 in
Section R.11.3.1). The obligation of the registrant to generate relevant additional inf ormation
f or the PBT/vPvB assessment concerns also relevant constituents, impurities, additives and
transf ormation/degradation products. This means that if there is not enough inf ormation
available on the PBT/vPvB properties of relevant constituents, impurities, additives and
transf ormation/degradation products to derive f or the registrant’s substance either conclusion
(i) or conclusion (ii), the registrant must generate the necessary f urther inf ormation on the
PBT/vPvB properties of the relevant constituents, impurities, additives and
transf ormation/degradation products until one of these two def initive conclusions can be
arrived at.

This obligation to generate relevant additional inf ormation is valid regardless of whether the
registrant’s dossier contains experimental inf ormation on the registered substance f or all
standard inf ormation requirements or whether he has made use of the data adaptation
possibilities of Annex XI and Column 2 of Annexes VII to X to the REACH Regulation. In certain
cases this may mean that the adaptation the registrant originally made (or planned to make)
in the registration needs to be replaced by results f rom a study which needs to be carried out
f or the purpose of the PBT/vPvB assessment as required in Section 2.1 of Annex XIII to the
REACH Regulation. Especially f or such Column 2 waivers of Annexes VII to X to the REACH
Regulation which are based on limited or unlikely exposure, it is important to note that the
registrant, if not able to conclude (i) (“The substance does not f ulfil the PBT or vPvB criteria” ),
may need to carry out the tests he originally wished to waive in order to be able to conclude
the PBT/vPvB assessment ultimately either by conclusion (i) or (ii), unless he decides to treat
the substance “as if it is a PBT or vPvB” (see next Section). For example, a registrant may
apply the Column 2 adaptation rule “The study need not be conducted if direct and indirect
exposure of the aquatic compartment is unlikely” f or the testing requirement (bioaccumulation
in aquatic species) of Section 9.3.2 of Annex IX to the REACH Regulation. If he concludes the
PBT/vPvB assessment with the conclusion (iii) (“The available data inf ormation does not allow
to conclude (i) or (ii)”) because the substance f ulfils the P or vP criteria and due to a Log Kow >
4.5 potentially f ulf ils the B/vB criteria, he must either carry out the bioaccumulation test he
originally wished to waive or he must treat the substance “as if it is a PBT or vPvB” (see next
Section).

The additional relevant inf ormation needed to be generated by the registrant must be
identif ied by the registrant in the technical dossier and CSR. This additional inf ormation can
relate to one or several tests as listed in Annexes IX or X to the REACH Regulation. The
additional relevant inf ormation can also be an “other type” of inf ormation, which the registrant
considers to be optimal f or the PBT/vPvB assessment, as Section 3.2 or Annex XIII to the
REACH Regulation allows the use of such other inf ormation. The other type of inf ormation can
be experimental inf ormation not f alling under Annex IX or X, but it may also be a combination
of experimental research inf ormation and monitoring research or solely research based on
monitoring/measured f ield data. Section R.11.4 provides guidance to the registrant f or
deciding which inf ormation could be necessary in pursuing an unequivocal conclusion (i) or (ii).
 Chapter R.11: PBT/vPvB assessment

31 Version 4.0 – December 2023

The additional inf ormation can be generated by the registrant in a tiered way by means of a
testing strategy, if this is deemed necessary. Elements of such testing strategies include
avoiding unnecessary animal or other testing and ensuring ef f icient use of resources while
optimising the generation of data that can be used to reach def initive conclusion (i) or (ii).

If the registrant, based on the PBT/vPvB assessment, identifies that inf ormation listed in Annex
IX or X to the REACH Regulation is needed, he must submit appropriate testing proposal(s).
Such testing proposals are subject to the normal testing proposal evaluation process of REACH.

If the registrant is using his right to generate f or the purpose of the PBT/vPvB assessment an
“other type” of inf ormation as described above, testing proposals cannot be submitted. The
registrant should, however, inf orm ECHA about his plans to generate any such other
inf ormation by specifying in the CSR to the degree of detail possible an appropriate inf ormation
gathering or testing strategy and an estimated time needed to update the PBT/vPvB
assessment and the registration dossier. This is the only way the registrant can inf orm ECHA
that he is using this possibility f or complying with the data generation obligation in his
PBT/vPvB assessment.

The registrant should strive to plan generation of f urther relevant inf ormation in a way that
leads to submission of a minimum number of updates of the PBT assessment and technical
dossier. However, it is recognized that PBT assessment can be challenging and the inf ormation
generated may sometimes provide results which indicate that f urther inf ormation not initially
f oreseen by the registrant needs to be generated to come to f inal conclusion (i) or (ii). In such
cases the registrant is obliged to update the registration dossier (including the CSR) without
delay each time new inf ormation becomes available. Hence, the registration dossier may in the
most complex cases need to be updated several times before the PBT assessment Step 1 can
be concluded.

Section 0.5 of Annex I to the REACH Regulation, requires of the registrant that: “[…] While
waiting f or results of f urther testing, he shall record in his chemical saf ety report, and include
in the exposure scenario developed, the interim risk management measures that he has put in
place and those he recommends to downstream users intended to manage the risks being
explored.” It is thus the duty of the registrant to identif y appropriate interim risk management
measures.

Section 2.1 of Annex XIII to the REACH Regulation requires relevant further inf ormation to be
generated regardless of the tonnage band f or the substance of the registrant conducting the
PBT/vPvB assessment. This obligation is illustrated by the f ollowing example: a registrant with
a tonnage band f or a substance of 10-100 t/y identif ies that more inf ormation is needed and
that (a) degradation simulation test(s) would be the f irst test(s) needed, f ollowed by a f ish
bioaccumulation test if the substance is deemed persistent af ter simulation testing. He must
submit a testing strategy and testing proposals, even though the degradation simulation test
and the f ish bioaccumulation test are not listed as standard inf ormation requirements f or 10-
100 t/y registrations.

R.[Link] Treat the substance “as if it is a PBT or vPvB”

If the registrant arrives at the conclusion (iii): The available information does not allow to
conclude (i) or (ii), he can also decide - based on REACH Annex XIII, Section 2.1 - not to
generate f urther inf ormation, if he f ulfils the conditions of exposure based adaptation of Annex
XI, Section 3.2(b) and (c). Uniquely to the PBT assessment, the registrant must additionally
 Chapter R.11: PBT/vPvB assessment

32 Version 4.0 – December 2023

consider the substance “as if it is a PBT or vPvB”, i.e. state that he wishes to regard the
substance as a PBT/vPvB without having all necessary inf ormation f or f inalising the PBT/vPvB
assessment. This option has exactly the same consequences f or the registrant and his supply
chain, as if the substance had been identif ied as PBT or vPvB based on a completed PBT/vPvB
assessment. This includes the obligation that if a substance is considered “as if it is a PBT or
vPvB”, the registrant must compile and provide recipients with a Saf ety Data Sheet (SDS) in
accordance with REACH Article 31 even if the substance does not already meet the criteria in
Article 31(1)(b) f or supply of an SDS. It is important that the registrant clearly f lags in the
registration dossier and in the supply chain communication that the substance is considered
“as if it is a PBT or vPvB”.

R.11.3.4 Emission characterisation, risk characterisation and risk

management measures
7
The registrant must develop f or a “PBT or vPvB substance” exposure assessments including
the generation of Exposure Scenario(s) (ES(s)) f or manufacturing and all identif ied uses as f or
any other substance meeting the criteria f or classif ication f or any of the hazard classes or
8
categories of Article 14(4) of the REACH Regulation .

Whereas for substances meeting the classification criteria f or Article 14(4) hazard classes or
categories the objective of an exposure assessment is to make qualitative or quantitative
estimates of the dose/concentration of the substance to which humans and the environment
are or may be exposed, the main objective of the emission characterisation f or “a PBT or vPvB
substance” is to estimate the amounts of the substance released to the dif f erent environmental
compartments during all activities carried out by the registrant and during all identif ied uses .

Additionally, f or a substance to be considered “as if it is a PBT/vPvB” (i.e., the substance is

regarded as a PBT/vPvB without f inalising the PBT/vPvB assessment), appropriate parts of the
CSR and the technical dossier must clearly demonstrate that the registrant f ulf ils the
conditions f or exposure based adaptation. This is the prerequisite as defined by Section 2.1 of
Annex XIII to the REACH Regulation f or avoiding the f urther inf ormation needed to f inalise the
PBT assessment Step 1. All use and exposure related inf ormation of the registration dossier
must in this case be in line with the specif ic conditions f or exposure based adaptation as
stipulated in Section 3.2(b) and (c) of Annex XI to to the REACH Regulation. For a description

7
For the purpose of this section including the sub-sections, it is noted, that when reference to a “PBT or vPvB
substance(s)” in italics is made, this covers both the case that the substance has been concluded to fulfil the PBT/vPvB
criteria and the case that the registrant considers the substance “as if it is a PBT/vPvB” (for when these terms apply,
see Section R.[Link]). However, it is noted, that the registrant needs to clearly flag in the technical dossier, CSR and
Safety Data Sheet which of the two cases applies to his substance.
8
i.e.:
• hazard classes 2.1 to 2.4, 2.6 and 2.7, 2.8 types A and B, 2.9, 2.10, 2.12, 2.13 categories 1 and 2, 2.14
categories 1 and 2, 2.15 types A to F
• hazard classes 3.1 to 3.6, 3.7 adverse effects on sexual function and fertility or on development, 3.8 effects
other than narcotic effects, 3.9 and 3.10
• hazard class 4.1
• hazard class 5.1
 Chapter R.11: PBT/vPvB assessment

33 Version 4.0 – December 2023

of the required conditions please ref er to the Guidance on intermediates and Chapter R.5:
Adaptation of information requirements of the Guidance on IR&CSA.

The subsequent risk characterisation f or “PBT or vPvB substances” requires a registrant to use
the inf ormation obtained in the emission characterisation step to implement on his site, or to
recommend to his downstream users, Risk Management Measures (RMM) and Operational
Conditions (OC) which minimise emissions and subsequent exposure of humans and the
environment throughout the lif e-cycle of the substance that results f rom manuf acture or
identif ied uses (Section 6.5 of Annex I to the REACH Regulation). RMMs and OCs are
documented in an ES(s).

R.[Link] Emission characterisation

The objective of the emission characterisation is:

• to identif y and estimate the amount of releases of a “PBT or vPvB-substance” to the

environment; and

• to identif y exposure routes by which humans and the environment are exposed to a “
PBT or vPvB-substance”.

The principal tool to achieve this objective is exposure scenarios. Part D and Chapters R.12 to
R.18 of the Guidance on IR&CSA provide guidance on how to develop exposure scenarios f or
substances in general. Parts of the exposure assessment guidance are relevant also f or “PBT or
vPvB substances” (i.e. emission estimation and assessment of chemical f ate and pathways).
However, since the objectives are not the same, the general scheme f or exposure assessment
needs to be adapted to the requirements of emission characterisation f or “PBT or vPvB
substances”. Guidance is given below on some issues where special considerations are needed
f or “PBT or vPvB substances”.

Throughout the development of an ES f or a particular use, the objective of the risk

characterisation f or “PBT or vPvB substances”, namely the minimisation of emissions and
(subsequent) exposures of humans and the environment that results f rom that use, needs to
be considered. Hence the need or a potential to (f urther) minimise emissions may be
recognised at any point in the development of the ES. In this case, the appropriate RMMs or
OCs must be included in the risk management f ramework and their effectiveness be assessed.
In particular, f or a substance to be considered “as if it is a PBT or vPvB”, the exposure
scenarios must be in line with the f act that the adaptation criteria of REACH Annex XI Section
3.2(b) and/or (c) are f ulf illed. The f inal ES, or ES(s) in case of dif f erent uses, must be
presented under the relevant heading of the chemical safety report, and included in an annex
to the SDS. It must describe the required OCs and RMMs in a way that downstream users can
check which measures they have to implement in order to minimise emissions or exposures of
humans and the environment.

It should be noted that a registrant has to take care of his own tonnage (manuf actured and
imported). In co-operation with his downstream users the registrant has to cover, where
relevant, his own uses and all identif ied uses including all resulting lif e-cycle stages. However,
it can be usef ul to consider on a voluntary basis exposure resulting f rom emissions of the same
substance manufactured or imported by other registrants (i.e. the overall estimated market
volume), c.f . Part A.2.1.
 Chapter R.11: PBT/vPvB assessment

34 Version 4.0 – December 2023

As “PBTs or vPvB substances” are substances of very high concern, the registrant must pay
attention to the level of detail of his assessment as well as to whether its accuracy and
reliability is suf ficient f or a “PBT or vPvB substance”. Where generic scenarios and assumptions
may be suf ficient f or exposure assessment of non PBT/vPvB-substances, specific scenarios and
data will be needed throughout an emission characterisation for “PBT or vPvB substances”. The
emission characterisation must, in particular be specif ic in the use description and concerning
RMMs, and must f urthermore contain an estimation of the release rate (e.g. kg/year) to the
dif f erent environmental compartments during all activities carried out during manuf acture or
identif ied uses. Emissions and losses may e.g. be addressed by performing mass balances. The
total amount of a substance going to each identif ied use must be accounted f or and the whole
use-specific lif e-cycles be covered. This can, f or instance, be done by perf orming a substance
f low analysis covering manufacture, all identif ied uses, emissions, recovery, disposal, etc. of
the substance. If the total amount of the substance cannot be accounted f or, the identif ication
of emission sources should be ref ined. All ef f ort necessary should be made to acquire f or
manuf acture and any identif ied use throughout the lif e-cycle, site- and product-specif ic
inf ormation on emissions and likely routes by which humans and the environment are exposed
to the substance. However, inf ormation on environmental concentrations is normally not
needed because minimisation of emissions and exposure is required f or “PBT or vPvB
substances” (data on environmental concentrations, if available, may however be usef ul in the
assessment and should be considered). Gathering of the mentioned inf ormation is not required
f or uses that are advised against as mentioned under heading 2.3 of the CSR and in Section
1.2 of the SDS.

R.[Link] Risk characterisation and risk management measures for “PBT or

vPvB Substances”

According to REACH, the objective of a risk characterisation f or PBTs or vPvBs is to minimise

emissions and subsequent exposure to these substances. Section 6.5 of Annex I to the REACH
Regulation f urther requires that: “For substances satisfying the PBT and vPvB criteria the
manufacturer or importer shall use the information as obtained in Section 5, Step 2 when
implementing on its site, and recommending for downstream users, RMM which minimise
exposures and emissions to humans and the environment, throughout the life-cycle of the
substance that results from manufacture or identified uses. ”

Risk characterisation f or PBT/vPvB substances includes, as f or other hazardous substances, the

consideration of dif f erent risks. These are:

• Risks f or the environment

• Risks f or dif ferent human populations (exposed as workers, consumers or indirectly via
the environment and if relevant a combination thereof )

• Risks due to the physico-chemical properties of a substance.

For the assessment of the likelihood and severity of an event occurring due to the physico -
chemical properties of a PBT/vPvB substance, the same approach f or risk characterisation
applies as f or any other substance (see Section R.7.1 of Chapter R.7a of the Guidance on
IR&CSA).
 Chapter R.11: PBT/vPvB assessment

35 Version 4.0 – December 2023

The estimation of emissions to the environment and exposure of humans perf ormed in the
emission characterisation provides the basis f or risk characterisation and risk management of
PBT/vPvB substances.

R.[Link].1 Options and measures to minimise emissions and exposure

A registrant has to generate ES(s) which describe how emissions and exposures to PBT/vPvB
substances are controlled. These ES(s) have to cover manufacturing, registrants’ own uses, all
other identif ied uses and lif e-cycle stages resulting f rom manuf acturing and identif ied uses.
Lif e-cycle stages resulting f rom the manufacture and identif ied uses include, where relevant,
service-life of articles and waste. The registrants are advised to consider at an early stage
which uses they wish to cover in their CSR. Obviously, if the registrant substitutes a PBT/vPvB
substance in his own uses or he decides to stop supplying f or certain downstream uses, he
does not need to cover these uses in his CSR. Supply chain communication is of high relevance
f or such cases.

For the uses the registrant decides to include in his CSA and theref ore develops ES(s) f or,
supply chain communication can be crucial f or getting detailed enough inf ormation on
conditions of use applied in practice. The registrant can conclude on the basis of the ES(s) he
develops that he is not able to demonstrate that emissions can be minimised f rom a specif ic
use. He must list any such uses as ‘uses advised against’ under heading 2.3 of the CSR.
Furthermore, this inf ormation has also be documented under heading 3.7 of the technical
dossier and communicated to the downstream users in Section 1.2 of the SDS.

The registrant has to implement the risk management measures and operational conditions
described in the f inal ES(s) f or manufacture and his own uses. He has to communicate as an
annex to the SDS the relevant ES(s) f or his downstream users. The downstream users have to
implement the recommended ES(s) or alternatively prepare a downstream user CSR.

One possibility to develop ES(s) that minimise emissions and exposure is to use a similar
approach as f or isolated intermediates (outlined below, for f urther details see the Guidance on
intermediates).

Rigorous containment of the substance

The “PBT or vPvB substance” must be rigorously contained by technical means during its whole
lif e-cycle. This covers all steps in the manuf acturing of the substance itself as well as all its
identif ied uses. It f urther includes cleaning and maintenance, sampling, analysis, loading and
unloading of equipment/vessels, waste disposal, packaging, storage and transport. This
containment may only become unnecessary from a step in the lif e-cycle on f or which it can be
demonstrated that the substance is being transf ormed to (an)other substance(s) without
PBT/vPvB properties or that the substance is included into a matrix f rom which it or any of its
breakdown products with PBT/vPvB properties will not be released during the entire lif e-cycle
of the matrix including the waste lif e stage. Note however that residues of the original “PBT or
vPvB substance” in the matrix or impurities with PBT/vPvB properties resulting f rom side-
reactions must additionally be considered (see Section R.[Link]).

Application of procedural and control technologies

Ef f icient procedural and/or control technologies must on the one hand be used to control and
minimise emissions and resulting exposure when emissions have been identif ied. For example,
 Chapter R.11: PBT/vPvB assessment

36 Version 4.0 – December 2023

in case of emissions to wastewater (including during cleaning and maintenance processes), it

will be considered that the substance is rigorously contained if the registrant can prove that
techniques are used that give virtually no emissions. The same applies to emissions to air or
disposal of wastes where technologies are used to minimise potential exposure of humans and
the environment. It is important to consider that RMM which protect humans, f or instance from
direct exposure at the workplace, can in some cases lead to emissions to the environment
(e.g. ventilation without f iltration of exhaust air). For a “PBT or vPvB substance”, such a
measure is insuf f icient as exposure of both humans and the environment must be minimised
(ventilation plus f iltration of exhaust air may thus be an option in the case of the example) .

On the other hand, procedural and/or control technologies must also be implemented to
guarantee safe use, i.e. to prevent accidents or to mitigate their consequences. Regarding this,
the clarif ications according to the Directive 2012/18/EU on the control of major-accident
hazards involving dangerous substances and the Directive 2014/34/EU concerning equipment
and protective systems intended f or use in potentially explosive atmospheres might be
consulted.

Handling of the substance by trained personnel

In order to minimise emissions and any resulting exposure, it is important that only traine d
personnel handle “PBT or vPvB substances” or mixtures. From this perspective any consumer
use of these substances on their own or in mixtures is probably inappropriate, because in
these cases suf f icient control of the emissions is in practice dif f icult to ensure.

R.[Link].2 Risk Characterisation for humans in cases of direct exposure to

“PBT or vPvB substances”

Although quantitative risk assessment methodologies can, due to the associated high
uncertainties regarding the extent of long-term exposure and effects, generally not be used f or
estimating the risk posed by “PBT or vPvB substances” to the environment or to humans via
the environment (indirect exposure of humans), it may be possible to use the quantitative
approach f or assessing the risk f or workers caused by direct exposure to the substance at the
workplace, because in this case exposure under the controlled conditions of the working
environment is predictable. A quantitative approach can only be applied to characterise the
risk f or workers resulting f rom direct exposure.

In case of assessing exposure at the workplace the quantitative approach (i.e.

Exposure / DNEL) must be used, wherever possible, to demonstrate that workplace exposure
does not result in health risks. If a DNEL cannot be derived (e.g. f or substances f or which effect
thresholds cannot be established), the respective approach for assessing the health risk posed
9
by non-threshold substances must be applied . The overall risk f or workers (resulting f rom all
types and routes of exposure) can normally only be assessed in qualitative terms and in doing
so the increased uncertainty in estimating the risk via indirect exposure through the
environment must be taken into due consideration. As a consequence, the application of a
higher margin of saf ety (i.e. a risk quotient Workplace Exposure / DNEL << 1) than usually
applied to non-“PBT or vPvB substances” may be required to account f or this increased

9
Note that, apart from predictable exposure, a further prerequisite for quantitative assessment of risk is the possibility
to derive the no-effect level for humans with an appropriate level of certainty.
 Chapter R.11: PBT/vPvB assessment

37 Version 4.0 – December 2023

uncertainty and to consider workplace exposure as saf e. Guidance on risk assessment f or

human health is given in Chapter R.8 of the Guidance on IR&CSA.

It should f urther be noted that even if a quantitative assessment of health risks at the
workplace would indicate low risks, this does not imply that the RMM and the OC at the
workplace can be considered sufficient where it is technically and practically possible to f urther
minimise emissions and exposure at the workplace.

R.11.3.5 Documentation of the PBT/vPvB assessment

The documentation of the PBT/vPvB assessment in the registration dossier consists of several
elements depending on the outcome. Section 8 of the CSR and Section 2.3 “PBT assessment”
10
of the technical dossier generated in IUCLID should be provided by all registrants who need
to conduct a CSA. Furthermore, f or substances with conclusion (iii) “The available data
inf ormation does not allow to conclude (i) or (ii)”, the registrant must identif y the additional
inf ormation needed in the CSA and in the technical dossier . These elements are described
f urther in the f ollowing.

When the registrant conducts a CSA and submits a CSR he needs to conduct the PBT/vPvB
assessment based on the relevant and available data (Step 1). This should be reported in
detail in Section 8.1 “Assessment of PBT/vPvB properties” of the CSR. One of the three
conclusion options described in Section R.[Link] must be recorded in this chapter as well.
Furthermore, if the registrant as the result of conclusion (iii) “The available data inf ormation
does not allow to conclude (i) or (ii)” considers his substance “as if it is a PBT or vPvB”, this
must be recorded in Section 8.1 as well.

If the registrant concludes that the substance f ulf ils the PBT/vPvB criteria or considers the
substance “as if it is a PBT or vPvB”, emission characterisation and risk characterisation shall
be conducted and the CSR must contain also a section “Emission characterisation”, reported as
Section 8.2 of the CSR. It is noted, that the CSR-plugin of IUCLID automatically creates these
two section titles. It is recommended that the registrant lists in Section 8.2 all relevant
sections of the CSR (Sections 9 and 10), including the details of the emission characterisation
elements.

All available relevant data must be recorded in the technical dossier in relevant endpoint study
records and those relevant to the PBT/vPvB assessment must be reflected in the CSR, Section
8.1. Furthermore, the conclusions of the PBT/vPvB assessment including brief justif ication
should be recorded in IUCLID Section 2.3. Support on how to f ill in the inf ormation in Section
2.3 “PBT assessment” of IUCLID in practice is given in the IUCLID End-User Manual. In this
section, it is possible to create one endpoint summary and several endpoint records. Note that
the objective of the PBT Section 2.3 in IUCLID is not to repeat inf ormation already provided in
other IUCLID sections. A ref erence to other IUCLID sections can be made.

10
The IUCLID software is downloadable from the IUCLID website at [Link] for free by all parties, if used for
non-commercial purposes.
 Chapter R.11: PBT/vPvB assessment

38 Version 4.0 – December 2023

If the conclusion (iii): “The available data inf ormation does not allow to conclude (i) or (ii)” is
drawn in the PBT assessment Step 1 the registrant must as part of the technical dossier submit
testing proposals, if the inf ormation needed is listed in Annex IX or X to the REACH Regulation.
Instructions f or recording the testing proposals in the technical dossier are provided in Data
Submission Manual 5. If the additional inf ormation needed to f inalise the PBT assessment Step
1 is not listed in Annex IX or X, the registrant cannot submit a testing proposal as testing
proposals on other items than those listed in Annex IX or X will be rejected by ECHA. If the
additional inf ormation is not listed in Annex IX or X, the registrant should describe in his CSR,
Section 8.1 what inf ormation is envisaged to be generated. In this case the CSR should also
contain the estimated timeline.

Af ter relevant studies have been conducted, the PBT/vPvB assessment must be updated. The
same applies to the CSR and the technical dossier including endpoint study records f or newly
generated inf ormation. The tasks of generation of f urther inf ormation and subsequent updating
of the CSR and the technical dossier should ideally be carried out in one step. However, it is
recognised that PBT/vPvB assessment sometimes may be a challenging task where several
updates and cycles of generation of additional inf ormation may be needed until the PBT/vPvB
assessment can be f inalised by the registrant.

Furthermore, the registrant must dif ferentiate in the registration dossier, CSR and Saf ety Data
Sheet between the status of a substance f ulf illing the PBT/vPvB criteria and a substance
considered “as if it is a PBT or vPvB”. This ensures that the downstream user receives enough
inf ormation to be able to make use of his rights and obligations under Article 37 of REACH.
Furthermore, this requirement is consistent with the purpose of the SDS, as stated in Section
0.2.1 of Annex II to to the REACH Regulation: ‘The safety data sheet shall enable users to take
the necessary measures relating to protection of human health and safety at the workplace,
and protection of the environment (…) a safety data sheet must inform its audience of the
hazards of a substance or a mixture and provide information on the safe storage, handling and
disposal of the substance or mixture’. Correct inf ormation on the hazard is provided when
there is a dif f erentiation between substances which meet the PBT/vPvB criteria based on data
and those which are treated "as if it is a PBT or vPvB".

If a registrant’s substance is included in the Candidate List as a PBT or vPvB substance, please,
see also Section R.[Link].

R.11.3.6 Documentation of the risk characterisation and communication of

measures
11
Given the potential risk exerted by “PBT or vPvB substances” , the descriptions of the
implemented or recommended RMMs and OCs in an ES need to be suf f iciently detailed to
demonstrate rigorous control of the substance and to allow examination and assessment of
their ef f iciency by authorities. The level of detail communicated in the ES attached to the
Saf ety Data Sheet must f urther permit downstream users to check that their use(s) are

11
“PBT or vPvB substance(s)” covers both the case that the substance has been concluded to fulfil the PBT/vPvB
criteria and the case that the registrant considers the substance “as if it is a PBT/vPvB” (for when these terms apply,
see Section R.[Link]).
 Chapter R.11: PBT/vPvB assessment

39 Version 4.0 – December 2023

covered by the ES developed by their supplier and that they have implemented the
recommended RMMs and OCs correctly.

The risk characterisation f or all ESs developed f or the identif ied uses of the “PBT or vPvB
substance” have to be documented under heading 10 of the CSR. The registrant is obliged
according to REACH Article 14 to keep his CSR available and up to date. It should be f urther
noted that any update or amendment of the CSR will require an update of the registration by
the registrant without undue delay.

If the registrant concludes based on available inf ormation (ii) “The substance f ulfils the PBT or
vPvB criteria” or he considers the substance “as if it is a PBT or vPvB”, this triggers the
obligation to generate a Saf ety Data Sheet according to REACH Article 31. For both cases, the
general obligations of Article 31 apply. Furthermore, the registrant must dif f erentiate in the
Saf ety Data Sheet which of the two cases applies f or his substance. This dif f erentiation is
necessary in order to provide the downstream users the possibility to take own action f or
assessing f urther the PBT/vPvB properties of the substance.
 Chapter R.11: PBT/vPvB assessment

40 Version 4.0 – December 2023

R.11.4 Assessment of PBT/vPvB properties – the scientific method

This section describes the method f or comparison of the available inf ormation with the criteria,
which f or the registrant is Step 1 of the PBT/vPvB assessment process. It should be noted that
this section is not meant to set obligations/requirements f or the registrant, but the registrant
should nonetheless use this part of the guidance f or pursuing the overall requirement to clarif y
unequivocally whether a substance f ulf ils the PBT or vPvB criteria or not. The method is the
same as used by authorities f or PBT/vPvB assessments, e.g., f or identif ying a substance as
“Substance of Very High Concern” f or the ECHA Candidate List according to REACH Article 59.
The method has been developed on a scientif ic basis and as such lays out the rules of
convention.

As in several areas of PBT/vPvB assessment scientific development activities are on-going, it is

underlined that the assessor has the responsibility to critically scrutinize and apply in the
PBT/vPvB assessment any relevant new scientif ic developments.

Sections R.[Link], R.[Link] and R.[Link] contain an assessment and testing strategy at
the beginning of those sections. It should be noted that there is a high number of dif f erent
combinations of property–specif ic conclusions, which a registrant may reach af ter the
assessment. Due to the high number of the possible outcomes, they are not presented in this
section. However, Section R.[Link] (conclusion (iii)) provides an overview of the dif f erent
situations that may arise f or which f urther inf ormation is needed.

Bef ore starting the assessment at the level of individual properties, it is recommended to
become f amiliarised with Section R.[Link]. Any substance containing multiple constituents,
impurities and/or additives should be assessed according to that section.

R.11.4.1 Standard approach

The PBT/vPvB assessment must cover a consideration of each property persistence,

bioaccumulation and toxicity against each respective criterion (P or vP, B or vB, and T) in order
to arrive at an inf ormed decision on the properties of a substance or of its relevant individual
constituents, impurities, additives or transf ormation/degradation products. In principle,
substances are considered as f ulfilling the PBT or vPvB criteria when they are deemed to f ulf il
the criteria P, B and T or vP and vB, respectively.

The assessment strategies set out in this section and Section R.11.4.2 should normally be
f ollowed and f urther inf ormation be searched f or or generated, if necessary. In deciding which
inf ormation is required on persistence, bioaccumulation or toxicity in order to arrive at an
unequivocal conclusion, care must be taken to avoid vertebrate animal testing when possible.
This implies that, when f or several properties f urther inf ormation is needed, the assessment
should normally f ocus on clarifying the potential f or persistence f irst. When it is clear that the P
criterion is f ulf illed, a stepwise approach should be f ollowed to elucidate whether the B
criterion is f ulf illed, eventually f ollowed by toxicity testing to clarif y the T criterion.

It should be noted that f or some elements of the PBT/vPvB assessment there may be , f or the
purpose of a particular PBT/vPvB assessment, a need to take the recent scientif ic
developments into account although they have not yet been implemented in this guidance. In
such a case the assessor should duly justif y the reasons f or deviation f rom, or extension of ,
the approach presented in this document.
 Chapter R.11: PBT/vPvB assessment

41 Version 4.0 – December 2023

Weight-of -Evidence determination

As described in Section R.11.2.1, a Weight-of-Evidence determination using expert judgement
is to be applied in the PBT/vPvB assessment. This applies f or all assessment situations
employing screening and/or assessment inf ormation. In order to decide whether the substance
must be considered as a potential PBT/vPvB substance based on screening inf ormation or as a
substance meeting the PBT or vPvB criteria, all relevant available inf ormation must be taken
into account.

The requirement to use a Weight-of-Evidence approach using expert judgement implies,

according to the introductory section of Annex XIII to the REACH Regulation, that “The
available results regardless of their individual conclusions shall be assembled together in a
single Weight-of-Evidence determination”. This normally means that the individual pieces of
data available do not need to be compared individually to each of the P, B, T or vP, vB criteria
but all inf ormation are assembled together f or each of the properties, respectively, f or the
purpose of a single comparison with the respective criteria. This does not exclude the option to
compare inf ormation directly with each of the P, B, T or vP, vB criteria to support the
assessment, where appropriate. It should be noted that Weight-of-Evidence determination is
not a mechanism to justif y disregarding valid, standard test data. The quality and consistency
of the data should be given appropriate weight.

In the Weight-of-Evidence assessment the derivation of a conclusion property by property

needs expert judgement, especially when very different types of inf ormation are available and
12
when the inf ormation cannot be directly (numerically) compared with the criteria .
13
The template and related background document f or Weight-of-Evidence /Uncertainty in
hazard assessment contains instructions on problem f ormulation, collection and documentation
of inf ormation, reporting of evidence, assessment of the quality of individual evidence,
integration and weighing of evidence, and uncertainty analysis. The Weight-of-Evidence
approach does not mean averaging results. It should rather help to structure the evaluation
and integration of all relevant and available inf ormation. The Practical Guide on “How to use
alternatives to animal testing to fulfil your information requirements for REACH registration”
provides a general scheme f or building a Weight-of-Evidence approach.

An essential prerequisite f or applying a Weight-of-Evidence approach is that the reliability and

relevance of experimental studies and non-experimental data are evaluated according to
Chapters R.4, R.7b and R.7c of the Guidance on IR&CSA. The reliability and relevance of
inf ormation to the PBT/vPvB assessment is f urther described in the f ollowing sub-sections. The
available inf ormation should be adequate f or purpose and evaluated in relation to the level of
certainty and accuracy needed to meet the regulatory requirement and this evaluation must be
well documented in the assessment report.

For particular cases, f urther described in Section R.[Link], the Weight-of-Evidence

determination should consider all three properties (i.e. persistence, bioaccumulation and
toxicity) in conjunction. In particular, if f or one or more of these properties only screening

12
In particular, it should be noted that although it might be theoretically possible to calculate degradation half -life
values or BCF values from screening information, such values must not be directly compared with the criteria.
13
ECHA Weight-of-Evidence/Uncertainty Template: [Link]
implementation/formats
 Chapter R.11: PBT/vPvB assessment

42 Version 4.0 – December 2023

inf ormation is available and screening threshold values as provided in the f ollowing sub-
sections are applied to draw a conclusion, all three properties must be considered in
conjunction.

Some examples of Weight-of-Evidence approaches used in the PBT/vPvB assessments f or

concluded assessment cases are listed at ECHA’s website

Relevant constituents, impurities, additives and transf ormation/degradation products

The PBT/vPvB assessment should be perf ormed on each relevant constituent, impurity,
additive and transf ormation/degradation product. It is not possible to draw an overall
conclusion if , e.g., the assessment of persistence has been concluded f or one constituent and
the assessment of bioaccumulation or toxicity f or another constituent.

Constituents, impurities and additives should normally be considered relevant for the PBT/vPvB
assessment when they are present in concentration of ≥ 0.1% (w/w). This limit of 0.1% (w/w)
is set based on a well-established practice recognised in European Union legislation to use this
14
limit as a generic limit . Individual concentrations < 0.1% (w/w) normally need not be
considered.

In practice, this means that the registrant should carry out a comparison of the available data
with the criteria f or all constituents, impurities and additives present in concentration of ≥
0.1% (w/w). Alternatively, the registrant should provide a justif ication in the CSR f or why he
considers certain constituents, impurities or additives present in concentration of ≥ 0.1%
15
(w/w) or certain constituent fractions/blocks as not relevant f or the PBT/vPvB assessment.

It may not always be possible or even necessary to f ully characterize and identif y f or the
purpose of the PBT/vPvB assessment UVCBs (substances of Unknown or Variable composition,
Complex reaction products or Biological materials) or fractions of impurities based on the
inf ormation given in Section 2 of Annex VI to the REACH Regulation f or substance
identif ication. This is because (i) the number of constituents/impurities may be relatively large
and/or (ii) the composition may, to a signif icant part, be unknown, and/or (iii) the variability of
composition may be relatively large or poorly predictable. Regardless of whether full
substance identification is possible or not for the whole composition, the registrant
should make efforts for carrying out a PBT/vPvB assessment for all constituents,
impurities and additives present in concentrations ≥ 0.1% (w/w). Section R.[Link]

14
The limit of 0.1% (w/w) is indicated in the European Union legislation, where there is no specific reason (e.g., based
on toxicity) to establish a concentration limit specific to the case. Examples of this generic concentration limit are, i.a.,
another category of substances of very high concern according to Article 57 of REACH, where the default concentration
of Carcinogenic/Mutagenic (category 1A/1B) ingredients in a mixture requiring a Carcinogen/Mutagen (1A/1B)
classification of the mixture under Regulation (EC) No 1272/2008 is 0.1% (w/w). Furthermore, Articles 14(2)(b),
31(3)(b) and 56(6)(a) of REACH apply a similar principle and the same concentration limit for PBT/vPvB substances in
mixtures regarding some obligations under REACH. Additionally, the Judgments of the General Court (Seventh
Chamber, extended composition) of 7 March 2013 in cases T-93/10, T-94/10, T-95/10 and T-96/10 (see in particular
paragraphs 117 to 121) confirmed the validity of this approach for PBT/vPvB constituents of a substance.

15
The terms “constituent fractions” refer to a situation where for a UVCB substance not all its constituents can be
identified individually and the substance identity needs then to be based on its fractions/groups of constituents.
“Block” is a term analogous to fraction/group and is used in the hydrocarbon block–approach (see Section R.[Link]).
 Chapter R.11: PBT/vPvB assessment

43 Version 4.0 – December 2023

provides further insight into how to carry out PBT/vPvB assessment for fractions of
the substance that cannot be fully identified by the registrant. For an example of a

application of this recommendation in a specif ic industry sector, please see the Environmental
16
assessment guidance on essential oils .

In specif ic cases it may be considered, f or the sake of proportionality of assessment efforts and
the level of risk being considered, to elevate or reduce the threshold value above or below
0.1% (w/w) f or the PBT/vPvB assessment. Account could be taken of , e.g. the use pattern of
the substance and the potential emissions of the constituents, impurities or additives having
PBT or vPvB properties. Careful consideration should be given especially when uses are known
or anticipated to cause signif icant emissions.

An elevated threshold value should not exceed 10% (w/w) f or the total amount of all
constituents, impurities and additives f or which conclusion on PBT/vPvB properties can not be
reached, and the total amount of these within the manuf actured/imported substance should in
no case exceed 1 t/year. A reduced threshold might be necessary to derive inf ormation
relevant f or PBT/vPvB assessment, e.g. f or very toxic substances, and the inf ormation on the
toxicity derived f or the classif ication and labelling purposes could be used f or def ining such a
lower concentration limit f or PBT/vPvB assessment.

Especially f or very complex UVCBs it is possible that individual constituents are present in
concentrations <0.1% (w/w) and that these have not been characterised by chemical analysis
individually. For UVCBs even the whole substance may consist of individual constituents only
present in such low concentrations. The f act that all individual constituents of a UVCB-
substance are present in concentration <0.1% (w/w) does not automatically exempt the
registrant f rom the obligation to carry out the PBT/vPvB assessment. A close structural
similarity of individual constituents within a f raction of a UVCB substance, i.e. constituents with
the same carbon number, chain lengths, degree and/or site of branching or stereoisomers,
triggers the need to sum up the concentrations of these constituents and to compare the total
concentration with the limit of 0.1% (w/w) in order to determine whether these constituents
need to be covered in the PBT/vPvB assessment. Criteria f or grouping or read across, as
mentioned in the Advice on using read-across f or UVCB substances, Practical Guide on “How to
use alternatives to animal testing to fulfil your information requirements for REACH
registration” and the “Introductory note to the illustrative example of a grouping of substances
and read-across approach” , should be applied to the determination and justif ication of such
f raction and (an) appropriate approach(es) as provided in Section R.[Link] should be applied
f or the PBT/vPvB assessment.

Similarly, a UVCB substance which contains constituents in concentrations well above 0.1%
(w/w) each, but also (a) large f raction(s) where constituents are individually <0.1% (w/w),
cannot be concluded as “not PBT/vPvB” unless it can be justif ied with suf f icient reliability that
none of the constituents and f ractions of minor constituents would cause a concern. For
example, a UVCB-substance may contain ten constituents, present in a total concentration of
60% (w/w) and the remaining 40% of the composition consists of not f ully identif ied
constituents. All latter minor constituents are individually present in concentration of <0.1%

16
Environmental assessment guidance on essential oils: [Link]
identification/sector-specific-support-for-substance-identification/essential-oils
 Chapter R.11: PBT/vPvB assessment

44 Version 4.0 – December 2023

(w/w) but are expected to be similar to each other structurally and hence expected to have
similar degradation, bioaccumulation and toxicity-properties. Not only the ten constituents
making the largest part of the substance, but also the remaining 40% of the composition
would need to be assessed using the appropriate approach provided in Section R.[Link] and
testing, where necessary.

The same principles, as described in the two previous paragraphs above f or UVCB-substances,
apply also to the constituents of well-defined substances and their impurity f ractions.

It should be noted in this connection that in cases where large f ractions of unidentif ied
constituents are present at <0.1% w/w, the assessment efforts need to remain proportionate.

A close structural similarity of individual constituents within a f raction, determined by criteria

of grouping or read across as mentioned above, means that the concentrations of constituents
with P, B and T (or vP and vB) properties should normally be summed up in order to compare
with the threshold of 0.1% (w/w). Structural similarity of the constituents justify assessing the
constituents as if they were one substance in terms of their physico-chemical, degradation and
bioaccumulative properties and ef fects. This recommendation relies on the assumption that the
mode of action of similar constituents is the same and the f ate properties are very similar,
hence causing an exposure which triggers ef f ects in humans and the environment as if the
exposure were to one substance. This understanding of aggregated exposure (aggregated
concentration) leading to corresponding aggregated ef f ects draws f rom the same s cientif ic
basis as the concept of additivity (“joint action”, “dose additivity”, “concentration additivity”,
“additivity of toxicity”), used in many regulatory activities, e.g. in the CLP-Regulation (EC,
2012; ECB, 2003; Feron et al., 2002). However, it should be noted, that if the criteria f or read
across are not f ulfilled f or degradation, bioaccumulation and (eco)toxicity in PBT-assessment
and f or the f irst two properties in the vPvB-assessment, such summing up is not applicable
and the normal 0.1% (w/w) threshold should be applied.

Similar arguments apply to relevant transformation/degradation products. However,

there is no set w/w threshold concentration f or transf ormation/degradation products. A
transf ormation/degradation product can be considered relevant f or example if the
concentration is continuously increasing or it seems to be stable during a degradation study.
The PBT/vPvB assessment should normally be carried out f or each relevant transf ormation or
degradation product (see also “Assessment of relevant transf ormation/degradation products”
under Section R.[Link].3).

It is not possible to draw an overall conclusion f or the substance if the assessment of

persistence has been concluded f or one transf ormation/degradation product and the
assessment of bioaccumulation or toxicity f or another transf ormation/degradation product.

The registrant should endeavour to carry out a comparison of the relevant available data with
the PBT/vPvB criteria f or each relevant transformation/degradation product (or in case those
cannot be ultimately identif ied: f or each group or block of transf ormation or degradation
products), respectively. If the registrant considers transf ormation/degradation products that
are f ormed (or groups/blocks of them) as not relevant f or the PBT/vPvB assessment, he should
also clearly explain in the PBT/vPvB assessment the reasons why they are not relevant.

If the available and relevant screening and other inf ormation allows the registrant to conclude
that the substance is not persistent using the screening threshold values as provided in Table
R.11—2, then it may normally be assumed that the substance is mineralized quickly and is not
 Chapter R.11: PBT/vPvB assessment

45 Version 4.0 – December 2023

likely to f orm transf ormation/degradation products relevant f or the PBT/vPvB assessment.

However, the available relevant screening or other inf ormation (including inf ormation f rom
hydrolysis tests and f ield data) may indicate that transf ormation or degradation products
relevant f or the PBT/vPvB assessment are indeed f ormed. These indications should be
addressed in the registrant’s PBT/vPvB assessment either qualitatively or quantitatively.

Following the obligation of the registrant under Article 13(3) of REACH in the situation where
new degradation simulation testing is necessary, the transformation and degradation products
relevant f or the registrant’s own PBT/vPvB assessment are those products, which must be
identif ied in tests C.23, C.24 and C.25 carried out in accordance with Council Regulation No
440/2008 of 30 May 2008 laying down test methods pursuant to Regulation No 1907/2006
(REACH) (“Test Methods Regulation”). It should be mentioned in particular that guideline C.24
requires that “…in general transformation products detected at ≥ 10% of the applied
radioactivity in the total water-sediment system at any sampling time should be identified
unless reasonably justified otherwise. Transformation products for which concentrations are
continuously increasing during the study should also be considered for identification, even if
their concentrations do not exceed the limits given above, as this may indicate persistence.
The latter should be considered on a case by case basis ...”. Identif ication of continuously
increasing transformation/degradation products with accompanying justification always applies
when the registrant is in the situation of generating new degradation simulation data f or the
purpose of the PBT/vPvB assessment because he will have previously concluded that the
substance may have PBT/vPvB properties.

For the situation where inf ormation f rom tests comparable to the standard degradation
simulation tests mentioned above are already available to the registrant or the registrant
considers it more appropriate to generate new degradation inf ormation in accordance with
Section 2.1 of Annex XIII to the REACH Regulation other than degradation simulation test data
(see Section R.[Link] f or the other possibilities), the principles of the standard test
guidelines mentioned above f or identif ying relevant transformation and degradation products
should be applied by analogy.

It should be noted that authorities are not bound under the REACH Substance Evaluation and
SVHC-identif ication processes to the stipulations of the Test Methods Regulation or other
standards f or def ining what is a relevant transf ormation/degradation product but have the
possibility to use other types of justified (concentration or f ormation rate) limits to def ine on a
case-by-case basis which transformation/degradation products are relevant f or their PBT/vPvB
assessment (e.g, see the Support Document of the Decision to identif y Bis(pentabromophenyl)
17
ether as Substance of Very High Concern ). Guidance is given in Section R.11.4.2 on the
assessment and testing strategy f or substances with specif ic substance properties such as
UVCBs or multi-constituent substances with several constituents, in relation to
transf ormation/degradation products, and f or substances with low water solubility, high
adsorption or volatility requiring deviations f rom the standard PBT/vPvB assessment.

Benchmarking as a part of Weight-of-Evidence

Benchmarking can be used as a part of Weight-of-Evidence approach. The benchmarking
approach is def ined as the approach where the f ate or behaviour of a substance is measured

17
[Link]
 Chapter R.11: PBT/vPvB assessment

46 Version 4.0 – December 2023

relative to the f ate or the behaviour of a well-described substance(s), the so called benchmark
chemical(s) (Adolfsson-Erici, et al. 2012). Benchmarking relies on the similarity between the
f ate or the behaviour of the benchmark chemical and the substance of interest in the
environmental system being studied (McLachlan, et al., 2017).

The benchmarking approach is a comparative approach and as such the available data on the
benchmark substance and the substance of interest determine the reliability of the conclusion
drawn f rom the benchmarking approach. The available data need to be comparable in terms of
(1) test conditions and test set up, (2) tested organisms (species, age, size) and (3) data
analysis and interpretation. Similarity of the benchmark chemical in aspects other than the one
investigated, e.g., similar physico-chemical partitioning behavior in a degradation study, can
lead to higher conf idence in the benchmark approach. It is however noted that the benchmark
approach is dif f erent f rom read-across, where structural similarity of the substances is
important. The benchmark approach is relatively new, and it is recommended to f ollow
developments in the f ield, e.g. via the ECHA website.

The benchmark chemical can be chosen to represent the worst-case benchmark, an equivalent
case benchmark or a saf e case benchmark in regard to the f ate or behaviour of a chemical. For
example, the benchmarking approach can be considered as a worst-case f or bioaccumulation
when the “well-characterised substance” is a benchmark substance that will not be eliminated
by excretion, biotransformation, or respiration in the duration of a BCF experiment (Adolfsson-
Erici, 2012). Respectively, an “equivalent case benchmark” or a “safe case benchmark” would
be a benchmark substance that will have similar behaviour or a behaviour below a well-defined
saf e level (bioaccumulation criteria not met).

In the case of persistence, there is currently little experience of the approach. McLachlan, et al.
(2017) provide some examples, e.g. measurement of the degradation rate of the chemical of
interest relative to the degradation rate of another (benchmark) chemical with an already well
characterized degradation behaviour.

If benchmarking is used to assess toxicity, it is recommended to report concentrations based

on mass (e.g. mg/l), but also on molar basis (e.g. mmol/l). Comparison of toxicity on molar
basis prevents bias f rom molecular weight dif ferences (e.g. a low molecular weight substance
would appear more toxic compared to a high molecular weight substance and toxicity of larger
molecules may be underestimated, when compared only on mass per litre basis , even if the
same amount of molecules are present per litre). This could be especially usef ul f or MOCS and
halogenated substances.

In the case of bioaccumulation, Inoue et al., (2012) conducted a benchmarking exercise to

compare the f ish dietary BMF values measured f or nine substances with their measured f ish
BCF values.

Co-exposure to a benchmark chemical with known bioaccumulation behaviour could provide

additional indication of the bioaccumulation potential of the test substance. For example, if the
derived BCF is well below or above that derived f or a benchmark chemical with known
bioaccumulation behaviour (e.g., PCBs) of similar hydrophobicity, this may f urther indicate
that the test substance is less bioaccumulative (or non-bioaccumulative) or more
bioaccumulative, respectively. The use of non-metabolisable benchmark chemicals in
laboratory bioaccumulation tests can also f acilitate the derivation of the other toxicokinetic
parameters, including whole-body biotransformation rate constants (Lo et al. 2015; Saunders
et al. 2020). DiMauro, 2018 conducted OECD TG 305 BCF tests using the minimised test
 Chapter R.11: PBT/vPvB assessment

47 Version 4.0 – December 2023

design with f our substances. Each substance was tested both alone and in the presence of four
non-biotransformed reference substances. Comparing the treatments with the test chemicals
on their own and the test chemicals in the presence of ref erence substances, the author
concluded that the presence of conservative ref erence substances did not af f ect the
biotransf ormation rate constant and BCF of the test chemicals.

When testing more than one substance in a bioaccumulation test, test substances should not
interact with one another. The combined dose should be below the combined concentration
which might cause toxicity and the potential f or interactive ef f ects, such as ef f ects on
metabolism should be taken into account (OECD, 2012a).

Furthermore, multimedia bioaccumulation f actors (mmBAFs) have been used to support

benchmarking f or cyclic volatile methylsiloxanes when compared to a known strongly
bioaccumulative substance, PCB 180 (Kierkegaard, 2011). The benchmarking approach has
been used as part of a Weight-of-Evidence for comparing the depuration rate constant (k 2) and
BMF derived f rom an experimental dietary BMF with other substances with known
bioaccumulation potential. Such an exercise provides useful evidence f or the bioaccumulation
assessment.

For some examples f or benchmarking in regulatory context see e.g. benchmarking to support
bioaccumulation assessment in the Support Document of the Decision to identif y Medium-
18 19
Chain Chlorinated Paraffins (MCCP) , Dodecamethylcyclohexasiloxane (D6) and “Dechlorane
20
Plus”™ as Substances of Very High Concern. More endpoint specific benchmarking examples
are provided under the endpoint specif ic assessment sections.

Reporting of the analytical methods as part of the environmental f ate studies

OECD Test Guidelines (TGs) f or simulation and bioaccumulation testing require the use of
appropriate methods to sample, extract, and analyse the test substance and its
transf ormation/degradation products (as well as metabolites) in environmental matrices
(water, soil and sediments, suspended matter, animal tissues).

Respective OECD TGs specif y the inf ormation that is required to be reported on the applied
analytical methods. Inf ormation provided must be detailed enough to allow the independent
assessment of the methods used and be reproducible by the description provided. Analytical
method description shall include adequate details regarding each step necessary to perf orm
the analyses such as sampling, pre-treatment, extraction, clean-up procedures, instrumental,
quantif ication and data processing methods (e.g. peak smoothing, area reject, integration
parameters). In general, the most appropriate method shall be selected and applied, and
reasonable ef f ort should be put in development and validation of such a method.

In cases where an appropriate method cannot be developed f or the target substance, there
must be justif ication with supporting evidence to show why it is technically not possible to do
so. All the ef forts made to develop a substance specific method should be reported considering
appropriate f it-f or-purpose approaches and state of the art techniques and methods,

18
[Link]
19
[Link]
20
[Link]
 Chapter R.11: PBT/vPvB assessment

48 Version 4.0 – December 2023

emphasising which specific issues do not allow to conduct the study and achieve its objectives.
Such justif ication should at least include descriptions of test system, sample stability, the
applied instrumental techniques and extraction- and clean-up methods with the results
achieved. Absence of an existing standard method f or a specific substance cannot be used as a
justif ication f or the study being technically not f easible.

Further inf ormation on analytical methods can be f ound f or example f rom Vogel’s Textbook of
Quantitative Chemical Analysis (Vogel et al., 2006) that gives an overview about available
quantitative methods. Critical literature review of analytical methods applicable to
environmental f ate studies published on ECHA website 21 provides an overview of available
techniques (chapter 5), guidance on validation criteria concerning analytical methods, such as
LOD and LOQ (chapter 6.6) and a detailed description of commonly used extraction techniques
(chapter 7). In addition, ECETOC (2013) provides inf ormation on relationships between
extraction techniques and bioavailability. However, please note that development of analytical
methods is a continuously evolving area of science, thus no recommendation on a most
suitable analytical method(s) f or a specific substance can be provided in the present guidance
document.

Development of workflows for the quantif ication of non-target data f rom High-Resolution Mass
Spectrometry (HRMS) is ongoing. Quantif ication of non-target data was demonstrated in a
study using LC-HRMS based on the ionisation ef f iciencies of different substances (Liigand et al.
2020). However, data analysis workflows are not yet standardised (Hollender et al. 2019), but
may potentially be considered as supporting analysis of environmental f ate studies f or example
in prioritisation and environmental monitoring (e.g. McCord et al., 2022).

To ensure availability of adequate inf ormation of the analytical methods used, the reporting
requirements described in the respective OECD TGs must be f ollowed and the relevant
inf ormation must be provided. Inf ormation to be included in the registration dossier can be
f ound in the ECHA’s Practical guide 3 (How to report robust study summaries) (Version 3.0,
June 2023), sections 5.2.3 and 5.2.4 f or biodegradation and bioaccumulation tests,
respectively. A review of current analytical methods applicable to environmental f ate studies 21
provides an overview of the requirements to be f ulf illed f or the specif ic simulation studies in
Chapter 3 (Tables 3.5-3.7) (Peter Fisk Associates, 2021).

Key inf ormation (not exhaustive list) to be reported where applicable includes details about the
analytical methods used and inf ormation regarding the validity of the method:

• Sample preparation procedure, type of the extraction, clean-up method, pre-

concentration step used, including solvents/reagents, sample storage stability and
instrumentation used (e.g. solid phase extraction (SPE), pressurized solvent extraction
(PLE), solvent extraction with extraction kits (QuEChERS))

21
This consultancy service report provides a critical literature review of feasibility of available analytical methods
applicable to environmental fate studies in the context of the PBT/vPvB assessment. It summarises recent scientific
developments, describes pros and cons of available analytical tools and provides guidance to improve the quality of
the data obtained from environmental fate studies. Available online:
[Link] _en.pdf/b3a7e562-bf9c-ef02-
948f-eaf1b8f89e3f
 Chapter R.11: PBT/vPvB assessment

49 Version 4.0 – December 2023

• Materials, ref erence standards, methods, instrumentation, calculations

• Test material chemical and radiochemical purity and stability

• Ef f iciency of the extraction method (the efficiency of the extraction should be tested in
a pre-study bef ore perf orming the f ate study); total recovery of the method (%).

• Demonstration of the stability of the analyte (and transformation/degradation products,

if applicable) in relation to the extraction method used (also to be assessed as part of a
preliminary study)

• Instrumentation used (e.g. HPLC-UV, LC-MS/MS, GC-MS, HRMS, NMR, LSC etc.)

• Representative chromatograms/spectra, quantitative reports (controls, blanks, spikes,

standards, samples)

• Repeatability and sensitivity including the limit of detection -and quantif ication
(LOD/LOQ) and working range

• Conclusions (applicability)
 Chapter R.11: PBT/vPvB assessment

50 Version 4.0 – December 2023

R.[Link] Persistence assessment (P and vP)

R.[Link].1 Integrated assessment and testing strategy (ITS) for persistence

assessment

A strategy f or degradation assessment and testing in the context of PBT/vPvB assessment is

proposed in Figure R.11—3. A tiered approach to assessment and testing is necessary until a
def initive conclusion on persistence can be drawn.

Available data consisting solely of screening inf ormation can be employed to derive a
conclusion mainly f or “not P and not vP” or “may f ulf il the P or vP criteria”. If screening
inf ormation indicates that the substance “may f ulfil the P or vP criteria”, higher tier inf ormation
generally needs to be made available.

Appropriate data need to be available to conclude the P/vP-assessment with a conclusion “not
P/vP” on all three compartments (or f ive, with marine compartments): water (marine water),
sediment (marine sediment) and soil. Either the available data, including in normal case
simulation test data f rom one or two compartments, can be interpreted so that a conclusion
can be derived on the remaining compartment(s) f or which no higher tier data are available, or
data need to be available directly on all compartments, or there is another justification f or why
a conclusion does not need to be drawn f or all three (f ive) compartments .

In the opposite situation, if a conclusion “P” or “vP” is reached f or one compartment, no f urther
testing or assessment of persistence of other environmental compartments is normally
necessary. In certain cases it may be possible to draw a conclusion “P” or “vP” based on
screening inf ormation (e.g. tests on inherent biodegradation) combined with other usef ul
inf ormation in a Weight-of-Evidence approach, as described later in this section and indicated
in the ITS in Figure R.11—3.

Screening test(s) indicating “not P and not vP” may not be suf f icient to conclude that a
substance is not persistent, if there is contradicting inf ormation on persistence. In this case,
additional inf ormation (e.g. QSAR, literature studies, read across) and quality of the screening
tests should be considered in a Weight-of-Evidence assessment (see also Section R.[Link] in
Chapter R.7b of the Guidance on IR&CSA). The results of the higher tier degradation
simulation studies (Annex XIII, Section 3.2: assessment inf ormation) are to be given more
weight in the Weight-of-Evidence assessment than the screening studies (Annex XIII, section
3.1: screening inf ormation). In the presence of a reliable higher tier study, it is not necessary
to analyse in detail the reasons f or potentially inconsistent outcomes of the screening tests.
The outcomes of a reliable and relevant higher tier study, supersede the screening tests.

For substances containing multiple constituents, impurities and/or additives, the guidance
provided below apply to that/those “part(s)” of the substance, which is/are the target(s) of the
assessment and testing. The criteria f or selecting an appropriate assessment approach is
provided in Section R.[Link].
 Chapter R.11: PBT/vPvB assessment

51 Version 4.0 – December 2023

1. Substance is readily
biodegradable?
yes Not P/vP*

no

yes

2. Screening information (Table R.11—4):

Positive enhanced ready

Specific inherent test positive
biodegradation test and other Not P and not vP*
with non-adapted inoculum
data supporting?

Negative enhanced ready Specific inherent

Potentially P and vP
biodegradation test? biodegradation test negative

3. Other information useful in a weight-of-evidence approach:

In situ/field Other (testing

Abiotic Applicable Monitoring Simulation test
degradation and non-testing
degradation QSARs data results
study results information)

no

4. Potentially P/vP: Further information needed if

substance also potentially B**. Develop a testing
strategy for simulation testing

OECD TG 309 technically feasible?

no
yes

4.1 Is there compartment specific concern for soil or sediment?

- water compartm ent is not at all relevant (based on fate and compartment of
release),
no - persistence criteria are most likely to be exceeded in sediment or soil,
- High hydrolysis rate etc...

Start with
OECD TG 309 4.2 Specify/justify the test compartment
(freshwater or (O ECD TGs 308, 307, or other) and test conditions
yes
- technical aspects
marine water) - compartment of concern aspects

5. Further information needed for P-

assessment for the chosen compartment yes
(e.g. on specific degradation products)?

ves
no

7. Conclude P & vP assessment for the

compartment

6. Can the result(s) be used, together Choose the next

with other relevant available information, compartment for
no
to conclude persistence in the remaining testing
compartment(s)?

yes

7. Conclude P and/or vP assessment

for all compartments

*This conclusion (not P/vP) can not be drawn if the Weight-of-Evidence assessment indicates a P concern.
** In the context of the Biocidal Product Regulation (BPR), it is worth noting that the P-criteria has to be assessed also when the T-criterion is
(potentially) fulfilled.

Figure R.11—3: Integrated Assessment and Testing Strategy for persistence

assessment – maximising data use and targeting testing.
 Chapter R.11: PBT/vPvB assessment

52 Version 4.0 – December 2023

Integrated assessment and testing of Persistence - Explanatory Notes to Figure R.11—

3.

1. Evidence of ready biodegradation

If the substance is readily biodegradable, or if the criteria f or ready biodegradability are
f ulf illed with the exception of the 10-day window, there is normally no reason to perf orm
f urther biodegradation tests f or the PBT/vPvB assessment. The conclusion is that the
substance is generally not regarded as f ulf illing the criteria f or Persistence (P or vP) (see
Sections R.7.9.4 and R.7.9.5 in Chapter R.7b of the Guidance on IR&CSA, and f or multi-
constituent substances see Section R.[Link]). However, based on a Weight-of-Evidence
assessment, screening inf ormation indicating “not P and not vP” may not always exclude the
substance f rom being persistent or even very persistent. If assessment inf ormation as listed in
Section 3.2 of Annex XIII is available, then this inf ormation should be compared in a Weight-
of-Evidence approach against the Annex XIII criteria rather than only relying on screening
22
inf ormation . Additionally, it should be ensured that the respective inf ormation is relevant and
reliable.

2. Other screening information (Table R.11—4)

Following the ITS, and based on the screening inf ormation, the substance can be concluded as
potentially P/vP or not P/vP according to the criteria and conditions described in Table R.11—4
and Sections R.7.9.4 and R.7.9.5 of Chapter R.7b of the Guidance on IR&CSA. Af ter
consideration of the Explanatory Notes bulleted below, and bef ore concluding that a substance
is “not P" or "not vP”, it should be caref ully examined if counter -evidence to that conclusion
exists, e.g. f rom monitoring data or other available inf ormation (see Points 3-7 below f or more
inf ormation). When combined with all available inf ormation on persistence in a Weight-of-
Evidence, the conclusion on persistence may cover one or multiple environmental
compartments.
If the substance is conf irmed to degrade in other biodegradation screening tests than the tests
f or ready biodegradability, the results may be used to indicate that the substance will not
persist in the environment. Specific enhancement conditions described in Sections R.7.9.4 and
R.7.9.5 of Chapter R.7b of the Guidance on IR&CSA can be used f or this purpose. For example,
a result of more than 60% ultimate biodegradability (ThOD, CO 2 evolution) or 70% ultimate
biodegradability (DOC removal) obtained under the conditions specif ied in Chapter R.7b in an
enhanced ready biodegradability test may be used to indicate that the criteria f or P are not
f ulf illed (see Sections R.7.9.4 and R.7.9.5 in Chapter R.7b of the Guidance on IR&CSA. The
enhancements may also be applied to standardised marine biodegradability tests (OECD TG
306, Marine CO2 Evolution test, Marine BODIS test, and the Marine CO 2 Headspace test).
• Assessment of inherent biodegradation test data - Results of a Zahn-Wellens test
(OECD TG 302B) or MITI II test (OECD TG 302C) only (not SCAS-test) may be used to
conf irm that the substance does not f ulfil the criteria f or P provided that certain additional
conditions are f ulf illed. In the Zahn-Wellens test, a level of 70% mineralisation (DOC

22
Judgment of the General Court of 9 June 2021, Exxonmobil v. ECHA, T‑177/19, not published in European Court
Reports, EU:T:2021:336, paragraphs 167-174 at paragraph 167.
 Chapter R.11: PBT/vPvB assessment

53 Version 4.0 – December 2023

removal) must be reached within 7 days, the log phase should be no longer than 3 days,
and the percentage removal in the test bef ore degradation occurs should be below 15%
(pre-adaptation of the inoculum is not allowed). In the MITI II test, a level of 70%
mineralization (O2 uptake) must be reached within 14 days, and the log phase should be no
longer than 3 days (pre-adaptation of the inoculum is not allowed). A lack of degradation in
an inherent biodegradation test (≤20%) can provide evidence that degradation in the
environment would be slow (see f urther consideration under “Tests on inherent
biodegradation in the main text). It should however be noted that the very low solubility of
many PBT/vPvB substances may reduce their availability and hence their degradability in
the test. The lack of degradation in an inherent test does not always imply that the
substance is intrinsically persistent and in some cases f urther testing might be needed.
• Enhanced ready biodegradability tests (screening tests)– Positive results f rom
enhanced biodegradability tests may be used together with other supporting inf ormation to
conclude that the substance is not P/vP. Enhanced ready biodegradability tests are allowed
when there is a need to compensate the poor bioavailability i.e. the substance is poorly
soluble and/or adsorptive. Only one enhancement is allowed at a time. The prolongation of
the test duration up to 60 days should only be considered if some initial, slow but steady,
biodegradation was observed not reaching a plateau by the end of the ready
biodegradability test, i.e. af ter 28 days. However, it is important that the f ollowing
conditions are met: 1) the enhancements should only be about an extended test duration or
an increased test vessel size, 2) the test should be perf ormed with non-pre-adapted/non
pre-exposed inocula, 3) the test duration should never be extended beyond 60 days, and 4)
the test criteria set f or ready biodegradability tests should be applied, i.e. 60% or 70%
degradation, depending on analyte, without the 10-day window. If the results are negative,
then it is generally not possible to def initively conclude on the absence of persistence of the
substance and f urther testing will be needed. More inf ormation on enhanced screening tests
can be f ound in Sections R.7.9.4 and R.7.9.5 of Chapter R.7b of the Guidance on IR&CSA.

3. Other information useful for a Weight-of-Evidence approach (not exhaustive)

All available inf ormation on (bio)degradation, including testing, non-testing and monitoring
data, should be considered. The overall evaluation could either show that the inf ormation
available coherently provides proof of (non-)persistence and is suf f icient to allow concluding
the P/vP assessment, or indicate that f urther testing is needed. If f urther testing is needed a
testing strategy should be developed f ollowing the ITS starting f rom step 4 below.
• Use of (Q)SAR (both QSARs and SARs) estimates – Ref er to Section R.[Link].4
below on “Assessment based on estimation models (QSAR, SAR)”, which describes QSARs
appropriate f or specif ic P/vP screening.
• Use of pure culture data – The data derived f rom studies with pure culture(s), single
species or mixture of species, cannot be used on their own within persistence assessment
but should be considered as part of a Weight-of-Evidence approach.
• Use of information on anaerobic degradation – The data derived f rom anaerobic
degradation studies cannot be used on their own within persistence assessment but should
be considered as a part of a Weight-of-Evidence approach.
• Use of information on any other degradation studies – The data derived f rom
degradation studies other than those described above cannot be used on their own within
 Chapter R.11: PBT/vPvB assessment

54 Version 4.0 – December 2023

persistence assessment but should be considered as a part of a Weight-of-Evidence

approach (e.g. OECD TG 314).
• Abiotic degradation – Concern f or P/vP screening cannot be removed by signif icant and
substantial loss of the parent substance by hydrolysis alone. Caref ul consideration of the
hydrolysis test is required (f or example mass balance is needed to address concerns f or
losses by volatilisation or adsorption to glassware). Rapid hydrolysis also needs to be shown
across all environmentally relevant pH. Additional evidence is also needed to examine
whether the f ate properties of the substance would cause attenuation of the hydrolysis rate
in sediment or soil, or whether DOC would similarly af fect the rate in aquatic media such as
river or sea water. Additional studies, e.g. examining the inf luence of dissolved organic
carbon / adsorption processes on hydrolysis rates, may be necessary f or this. In
biodegradation studies, including sterile controls as a part of study design is usef ul to
indicate abiotic degradation in relevant conditions (see Sections R.[Link].2 and R.[Link].
in Chapter R.7b of the Guidance on IR&CSA f or more details). The degradation half -lives
obtained in a hydrolysis test cannot be directly compared to the persistence criteria of
Annex XIII. As abiotic degradation is primary degradation, careful consideration will need to
be given to the potential f ormation of stable degradation products with PBT/vPvB properties.
Hydrolysis products should be identif ied in accordance with the recommendations contained
in the test guidelines (e.g. OECD TG 111).
• Use of other abiotic data – Data derived f rom other abiotic studies (e.g.
photodegradation, oxidation, reduction) cannot be used on their own within persistence
assessment, but may be used as part of a Weight-of-Evidence approach. Due to the large
variation in the light available in dif f erent environmental compartments, the use of
photolysis data is not generally recognised for persistence assessment. This is discussed in
more details in the Chapter R.7b of the Guidance on IR&CSA.
• Field studies – Data derived f rom field studies (e.g. mesocosm) may be used as part of a
Weight-of-Evidence approach. This is discussed in more detail in Section R.[Link].5 below
named “Field studies f or persistence”.
• Monitoring data – If monitoring data, used as part of a Weight-of-Evidence analysis, show
that a substance is present in remote areas (i.e. long distance f rom populated areas and
known point sources, e.g. Arctic sea or sub-Arctic/Arctic lakes in Scandinavia), it may be
possible to conclude a substance as P or vP. Monitoring data obtained in areas closer to the
sources may also be usef ul f or P/vP assessment and can be used as one line of evidence for
supporting the conclusions(in both directions: P/vP or not P/vP). Use of monitoring data in
P/vP-assessment encompasses several uncertainties and conclusions should be drawn on
the basis of monitoring data only when there is suf f icient understanding of the substance
distribution and transport behaviour and under the condition that the uncertainties in the
monitoring data presented are adequately addressed. The lack of detection of a substance
in monitoring data should be considered caref ully as it does not necessarily mean that a
substance is not persistent (e.g. shortcomings in analytical methods may af f ect monitoring
of substances in the environment). If monitoring data show that the substance levels in
environmental media or biota are rising, the reasons f or such a time trend should be
assessed very carefully against the inf ormation on the time trends of vo lumes, uses and
releases. Archived samples from environmental specimen banks, dated sediments cores and
ice cores can be used to gain understanding on temporal changes. Where monitoring data
clearly indicate that the substance f ulfils the vP-criterion or, depending on the case, that the
P criterion is f ulf illed in addition to other supporting inf ormation (and without any conf licting
 Chapter R.11: PBT/vPvB assessment

55 Version 4.0 – December 2023

data), it may not be necessary to generate simulation degradation data. In the latter case,
conclusions on the f ulf ilment of the P/vP criteria may be drawn based on the monitoring
data, the inf ormation on the substance distribution/transport behaviour, in addition to other
supporting inf ormation used as part of a Weight-of-Evidence analysis.

4. Further information needed to conclude on P/vP – Testing strategy to be

developed as described below
If f urther degradation testing is needed based on steps 1 to 3 of the ITS, a testing strategy on
persistence should be developed. The testing strategy should aim to conclude on persistence
with the least possible ef f orts in testing and at the same time cover the assessment of
persistence in all environmental compartments (marine water, f resh or estuarine water,
marine sediment, f resh or estuarine sediment and soil).
4.1. Identification of any specific environmental compartment(s) of concern
This paragraph describes the part of the ITS where the need f or f urther testing has been
identif ied and there is a need to make a decision on the test compartment(s).
In general, it is recommended to start testing with the OECD TG 309 if it is technically f easible.
However, if there is evidence that the OECD TG 309 does not provide means to ref lect the
persistence of the substance in the environment, other environmental compartments may be
considered as f irst test environment. For example, in case a P/vP criterion is expected to be
exceeded in (a) compartment(s) other than water or if the substance hydrolyses f ast in
environmentally relevant conditions, this should be taken into account in the testing strategy.
If , based on the f ate and release(s) of the substance, it is considered that water compartment
is not a relevant environmental compartment at all, this should also be taken into account in
the testing strategy.
If the OECD TG 309 is not technically f easible, selection of the most relevant environmental
compartment to test f irst should be justif ied (Step 4.2).
OECD TG 309 should be pref erred f or the f ollowing reasons:
• Firstly the aquatic compartment is considered to be a relevant environmental compartment
due to the large global volume of water: by default water compartment receives signif icant
amount of emissions directly or indirectly, and transports/distributes the substance through
e.g. deposition and run-off (unless based on the f ate and release(s) of the substance, it is
considered that the water compartment is not a relevant environmental compartment at
all). Once entering water, a substance may stay there for very long time and be spread over
long distances bef ore it reaches other environmental compartments (via environmental
transport, partitioning and distribution processes) such as sediments or (via air) the soil
compartment;
• Particularly f or lower water solubility substances which tend to be adsorptive, the OECD TG
309 (with a def ault concentration of suspended solids of 15 mgdw /L, see section below on
OECD TG 309) minimizes potential NER f ormation. If NER is f ormed at signif icant levels in
the OECD TGs 307 and 308 studies, this can be dif f icult to interpret and compare with
degradation half -lives criteria of Annex XIII to the REACH Regulation;
• OECD TG 309 is conducted under aerobic condition (there is no “anaerobic” option). This is
considered as a relevant test condition as P assessment should f irst consider aerobic
degradation. In general, a test using exclusively anaerobic conditions is not required as a
 Chapter R.11: PBT/vPvB assessment

56 Version 4.0 – December 2023

f irst step. For f urther inf ormation, see Section R.[Link].3 below, under “aerobic and
anaerobic conditions”.
It should be noted that, at this step, considerations of complete absence of uses/releases, and
thereby exclusion of the need to test a certain environmental compartment, is not discussed
under this section. Further inf ormation on exposure-based exclusion of testing may be f ound in
this Guidance under Section R.11.3 and Figure R.11—2.
Inf ormation on degradation and f rom environmental monitoring data, emissions estimated in
the CSR, distribution modelling data (e.g. Mackay Level III and SimpleTreat v4.0) and physico-
chemical inf ormation should be assessed to determine whether there is an environmental
compartment (pelagic surface water, pelagic marine, sediment, marine sediment or soil) of
specif ic concern f or persistence. The driving f actor f or the assessment is that a con clusion
needs to be derived f or all three (f ive) environmental compartments with the least possible
testing ef f orts. The specific concern f or persistence is normally present f or the environmental
compartment f or which the P/vP criteria are most likely to be exceeded or where the expected
degradation half -lif e is the closest to the criteria. Consideration of the environmental
compartment(s) of most relevant exposure may also play a role in the identif ication of the
specif ic environmental compartment f or testing. Absence of exposure in a specif ic
environmental compartment may, in exceptional cases, be acceptable to exclude certain
compartments f rom the P/vP assessment.
The f ollowing pieces of evidence may help in the identif ication of the potential environmental
compartment of specif ic concern:
• Any available inf ormation suggests that (abiotic and bio-) degradation rates/half -lives are
expected to meet the P/vP criteria f or a specif ic environmental compartment;
• Environmental monitoring data suggesting persistence is likely in a particular environmental
compartment f or a substance;
• Direct discharge to an environmental compartment is expected to occur;
• The lif e-cycle is well characterised and the environmental emission and exposure
assessment (including environmental f ate, modelling and/or monitoring data) show that a
specif ic environmental compartment is exposed.
If any environmental compartment other than surf ace water is chosen f or simulation
degradation testing, a justif ication should be provided (see step 4.2 below).
4.2. Specify/justify the test compartment
As explained above (step 4.1) the OECD TG 309 is the pref erred test. If another test is
selected f or f urther testing, this should be justif ied. Possible reasons are listed below:
• OECD TG 309 is typically perf ormed at concentrations between 1 and 100 μg/L and
pref erably in the range of <1-10 μg/L (to ensure that biodegradation f ollows f irst order
kinetics).
• Generally, when water solubility of a substance is very low (typically below 1 μg/L), testing
on sediment (OECD TG 308) and/or soil (OECD TG 307) may be needed instead of a pelagic
test (OECD TG 309). The detection limit(s) of analytical methods of quantif ication needs to
be taken into account when designing the test setup.
• Aquatic testing is not technically f easible. Technically f easible means that it ha s been
impossible, with allocation of reasonable efforts, to develop suitable analytical methods and
 Chapter R.11: PBT/vPvB assessment

57 Version 4.0 – December 2023

other test procedures to accomplish testing in surf ace water so that reliable results can be
generated. Appropriate analytical methods should have a suitable sensitivity and be able to
detect relevant changes in concentration (including that of transf ormation/degradation
products).
• Indications f rom available data (e.g. literature) suggest that persistence is likely to occur in
a dif f erent environmental compartment (i.e. in soil or sediment), including evidence of
direct or indirect exposure.
• The substance is a multi-constituent / UVCB which af fects the test substance concentration
at which the test can be perf ormed (i.e. due to dif f erent multiple water solubilities of the
individual constituents).
Please see also f urther considerations on the simulation testing strategy in Section
R.[Link].3 below.

5. Is there further information needed to conclude on persistence for the tested

environmental compartment?
The inf ormation obtained f rom the perf ormed tests should be assessed and the results
compared with the REACH Annex XIII criteria f or P/vP:
• If the substance or its degradation products are concluded to be persistent or very
persistent, there is no need f or f urther testing f or persistence assessment.
• If the substance and its degradation products are concluded to be non-persistent in the
tested environmental compartment it should be verif ied that there is no concern in
remaining compartments (see step 6).

6. Remaining concern in untested environmental compartments

It should be considered whether the available inf ormation is adequate to conclude persistence
assessment f or all or some of the remaining environmental compartments for which there are
no testing data. If it can be concluded that the P and/or vP criteria are f ulf illed in one
environmental compartment, then no f urther inf ormation is needed f or the other
compartments (see above step 5).
In general, results of a single simulation degradation study demonstrating non persistence in
one compartment cannot be directly extrapolated to other non-tested environmental
compartments. However, the results could be suf f icient to conclude on persistence or very
persistence in other compartments, provided that the relevant test conditions have been
applied and the interpretation of the results/bridging is backed by proper justif ications.
Availability or generation of multiple simulation test data may allow more Weight-of-Evidence
based conclusions to be drawn by expert judgement regarding environmental degradation half -
lives f or one or more environmental compartments. At this point of the f low chart, a de cision
on whether the data cover one, two or all f ive environmental compartments should be made
on a case-by-case basis.
It should be highlighted that the requirement is to draw a conclusion f or all three (f ive)
environmental compartments (see REACH Annex I, Section 3.0.2). If f or the f irst tested
compartment a conclusion “not P” could be derived, but the available data are not suf f icient for
drawing conclusions in (an)other compartment(s), f urther data generation is necessary to
 Chapter R.11: PBT/vPvB assessment

58 Version 4.0 – December 2023

complete the assessment f or the compartments f or which a conclusion could not be drawn.
Exclusion of (a) certain environmental compartment(s) f rom the P/vP assessment based on
absence of exposure may be acceptable only in very exceptional cases and upon justif ication.
A justif ication of absence of exposure in (a) certain environmental compartment(s) is dif f erent
f rom a justif ication f or the purpose of normal quantitative risk assessment, because f or
(potential) PBT/vPvB substances, and hence f or the PBT/vPvB assessment, distribution over a
very long timespan would need to be considered as well.

7. Evaluation versus the P and vP criteria

The half -lif e (lives) obtained f rom the simulation data is evaluated against the criteria of Annex
XIII to the REACH Regulation f or the three (f ive) environmental compartments to determine
whether the P or vP criteria are met or not. Bef ore f inally concluding that a substance is “not
P" or "not vP”, it should be caref ully examined if there exists conf licting evidence f rom
monitoring data, either f rom national monitoring programmes of Member States (e.g. Swedish
23
national monitoring data collection ), f rom European monitoring programmes (e.g. NORMAN
24
Network ) or internationally acknowledged organisations (such as OSPAR or the Danube
Convention). For example, f indings of signif icant concentrations of the substance under
consideration in remote and pristine environments such as the Arctic sea or Alpine lakes need
to be scrutinized caref ully as there may be evidence of high persistence. Also, signif icant
concentrations of the substance in higher levels of the f ood chain in unpolluted areas may
indicate high persistence (beside a potential to bioaccumulate). If such evidence indicates that
the substance may be persistent, f urther investigations are required.

R.[Link].2 Introduction to persistence assessment

When assessing data concerning the persistence of a substance and, if necessary, determining
the next steps of the assessment, there are a number of stages to go through. The f irst part of
the assessment should address the extent to which available data enable an unequivocal
assessment to be made. These data may comprise simple screening biodegradation tests (e.g.
OECD TG 301C ready biodegradability MITI I test) or complex, high-tier simulation tests (e.g.
OECD TG 308 aerobic and anaerobic transformation test in aquatic sediment systems). At this
stage, it is only necessary to assess the strength of the data in one direction or another. Thus,
f or example, when an OECD TG 301 study indicates that the substance is readily
biodegradable, the decision that a substance is not P could normally be taken. However, , a
positive (meeting the set criteria) ready biodegradation test does not exclude that, in some
25
cases, f urther studies may be required . This is the case when based on the Weight-of
Evidence there are indications of potential persistence. For example, if a simulation test is
available, this data should be used in the assessment rather than only relying on screening
inf ormation. If a reliable simulation test with a degradation half -life f ulfillling the P/vP criteria is
available, this would be suf ficient to decide that the substance meets the P and vP criteria..
However, as described in Section R.7.9 of Chapter R.7b of the Guidance on IR&CSA, a negative

23
[Link]
24
[Link]
25
Judgment of the General Court of 9 June 2021, Exxonmobil v. ECHA, T‑177/19, not published in European Court
Reports, EU:T:2021:336, paragraphs 167-174.
 Chapter R.11: PBT/vPvB assessment

59 Version 4.0 – December 2023

result in a test f or ready biodegradability does not necessarily mean that the substance will not
be degraded under relevant environmental conditions and persist in the environment. Indeed,
there are some cases where ready biodegradation tests have been reported to underestimate
the potential f or degradation in real environmental conditions (Guhl and Steber, 2006). A f ailed
ready biodegradability test may indicate the need f or further testing under less stringent test
conditions (e.g. enhanced biodegradation tests, simulation tests…). In addition, all relevant
degradation pathways (biotic, abiotic, aerobic, anaerobic conditions) need to be considered
with regard to the relevant route of exposure bef ore concluding on persistence.

Of ten, biodegradation data are not so clear-cut, and f requently they are dif f erent and/or
contradictory. Therefore careful consideration is needed before a decision is taken in order to
avoid a f alse negative or f alse positive conclusion. The strategy outlined in this section is a
recommendation and is not intended to be an explicit prescriptive description of the sequence
of steps to be taken. Ultimately the actual route taken will depend upon the data available and
the physico-chemical properties of the substance being assessed. As a minimum, and where
possible and technically f easible, inf ormation on vapour pressure, water solubility,
octanol/water partition coeff icient (Kow ), other partition coef f icients (such as the octanol-air
partition coef ficient (Koa) and organic carbon normalised adsorption coef f icient (K oc)), basic
dissociation behaviour (if relevant), surf ace active properties (if relevant) and Henry's law
constant must be available. The impact of these data on the test design and data
interpretation should be considered.

With regard to persistence, it is insuf ficient to consider removal alone where this may simply
represent the transfer of a substance f rom one environmental compartment to another (e.g.
f rom the water phase to the sediment). Degradation may be biotic and/or abiotic (e.g.
hydrolysis) and result in complete mineralisation, or simply in the transformation of the parent
substance (primary degradation). Where only primary degradation is observed, it is necessary
to identif y the degradation products and to assess whether they possess PBT/vPvB properties.
Other properties of transf ormation/degradation products of specif ic concern (e.g. ED
26
properties ) may also be relevant to be f urther considered. In addition to the substance
intrinsic properties, its transf ormation and/or degradation is dependent on the surrounding
environment.

The f ollowing sections give guidance on how to address data f rom biodegradation studies,
abiotic degradation studies and inf ormation available f rom estimation models (QSARs/SARs). A
subsequent section addresses inf ormation generation and particularly how to choose the
correct compartment f or f urther testing. As mentioned above, the sequence in which the
subjects of these sections are addressed will depend upon the data available . Furthermore,
most of the inf ormation reported in this guidance is f urther developed under the endpoint-
specif ic guidance on degradation, which should also be consulted (see Section R.7.9 in Chapter
R.7b of the Guidance on IR&CSA).

In case only screening inf ormation is available, screening threshold values listed in Table
R.11—4 can be used to judge whether an ultimate conclusion on the persistence of a

26
Annex I (Part 4) to CLP Regulation (EC) No 1272/2008 was amended to include Classification criteria for PBT/vPvB.
These criteria include that a substance classified under the CLP as endocrine disuptor for human health (Cat. 1) or
endocrine disruptor for the environment (Cat. 1) fulfills the T criterion under the CLP hazard class 'Persistent,
Bioaccumulative and Toxic or Very Persistent, Very Bioaccumulative properties’ (19 Dec 2022).
 Chapter R.11: PBT/vPvB assessment

60 Version 4.0 – December 2023

substance can be made or whether f urther inf ormation is needed. It should be noted that
screening criteria can only be applied as provided. The triggers were originally derived f or
drawing only those conclusions indicated in Table R.11—4 and are not recommended to be
used to draw other conclusions. (However, it should be noted that these criteria are indicative
and the assessor should consider the relevance of any other indications e.g. other
experimental data than standard simulation and screening tests, other QSARs, monitoring data
and expert judgement, bef ore drawing a conclusion.)

Table R.11—4: Screening information for P and vP.

Screening information Conclusion

Persistence

Biowin 2 (non-linear model Does not biodegrade fast (probability < 0.5)* Potentially P or vP
prediction) and Biowin 3 and ultimate biodegradation timeframe
(ultimate biodegradation time) prediction: ≥ months (value < 2.25 (to
2.75)**)
or or
Biowin 6 (MITI non-linear model Does not biodegrade fast (probability < 0.5)*
Potentially P or vP
prediction) and Biowin 3 and ultimate biodegradation timeframe
(ultimate biodegradation time) prediction: ≥ months (value < 2.25 (to
2.75)**)

or or
other models * Model specific values Potentially P or vP

Ready biodegradability test ≥70% biodegradation measured as DOC Not P and not vP
(including modifications allowed removal (OECD TGs 301A, 301E and 306) or
in the respective TGs) ≥60% biodegradation measured as ThCO 2
(OECD TG 301B) or ThOD (OECD TGs 301C,
301D, 301F, 306 and 310)***
<70% biodegradation measured as DOC Potentially P or vP
removal (OECD TGs 301A, 301E and 306) or
<60% biodegradation measured as ThCO 2
(OECD TG 301 B) or ThOD (OECD TGs 301C,
301D, 301F,306 and 310)
Enhanced ready biodegradability Biodegradable (pass levels as for ready Not P and not vP
tests**** biodegradability tests above) **** Potentially P or vP
Not biodegradable****
Specified tests on inherent
biodegradability:

- Zahn-Wellens (OECD TG 302B) ≥70 % mineralisation (DOC removal) within 7 Not P and not vP
d; log phase no longer than 3d; removal
before degradation occurs below 15%; no
pre-adapted inoculum
Potentially P or vP
Any other result*****
- MITI II test (OECD TG 302C) ≥70% mineralisation (O2 uptake) within 14 Not P and not vP
days; log phase no longer than 3d; no pre -
adapted inoculum
Any other result***** Potentially P or vP

* The probability is low that it biodegrades fast (see Section R.[Link] in Chapter R.7b of the Guidance on IR&CSA).
Other models are described in Section R.[Link] of Chapter R.7b of the Guidance on IR&CSA and in this section below.
** For substances fulfilling this but BIOWIN 3 indicates a value between 2.25 and 2.75 more degradation relevant
information is generally warranted.
 Chapter R.11: PBT/vPvB assessment

61 Version 4.0 – December 2023

*** These pass levels have to be reached within the 28-day period of the test. The conclusions on the P or vP
properties can be based on these pass levels only (not necessarily achieved within the 10-d window) for
monoconstituent substances. For multi-constituents substances and UVCBs these data have to be used with care as
detailed in Section R.[Link] of this Guidance. “Not P/vP” conclusion may not be possible if WoE assessment indicates
potential for persistence.
**** See Sections R.7.9.4 and R.7.9.5 in Chapter R.7b of the Guidance on IR&CSA. Expert judgement and or use of
Weight-of-Evidence also employing other information may be required to reach a conclusion (i.e. concerning
« biodegradable/ not biodegradable »). “Not P/vP” conclusion may not be possible if WoE assessment indicates
potential for persistence.
*****see section R.[Link].3 for concluding ultimately on persistence in particular cases (in particular “Tests on
inherent biodegradation”). “Not P/vP” conclusion may not be possible if WoE assessment indicates potential for
persistence.

In the ITS f or persistence assessment (Figure R.11—3), the types of simulation degradation
tests that should be considered is indicated. The inf ormation in Table R.11—5 below presents
the criteria f or the assessment of persistence (P/vP) and identif ies relevant test systems f or
determining environmental degradation half -lives.

Table R.11—5: Persistence (P/vP) criteria according to Annex XIII to the REACH
Regulation and related simulation tests.

According to REACH, Annex According to REACH, Annex Biodegradation simulation

XIII, a substance fulfils the P XIII, a substance fulfils the tests from which relevant
criterion when: vP criterion when: data may be obtained
include:
The degradation half-life in The degradation half-life in OECD TG 309: Simulation test –
marine water is higher than 60 marine, fresh- or estuarine aerobic mineralisation in surface
days, or water is higher than 60 days, or water
The degradation half-life in
fresh- or estuarine water is
higher than 40 days, or
The degradation half-life in The degradation half-life in OECD TG 308: Aerobic and
marine sediment is higher than marine, fresh- or estuarine anaerobic transformation in
180 days, or sediment is higher than 180 aquatic sediment systems
The degradation half-life in days, or
fresh- or estuarine water
sediment is higher than 120
days, or
The degradation half-life in soil The degradation half-life in soil OECD TG 307: Aerobic and
is higher than 120 days is higher than 180 days anaerobic transformation in soil

R.[Link].3 Test data on biodegradation

In principle, there are three types of tests that measure biological degradation:

1. Tests on ready biodegradation (e.g. OECD TG 301 series, OECD TG 306, OECD TG 310
and enhanced ready test)
2. Tests on inherent biodegradation (OECD TG 302 series)
3. Tests on simulation degradation and transformation (OECD TG 309 surf ace water, OECD
TG 308 sediment or OECD TG 307 soil)
Tests on ready and inherent biodegradability contribute inf ormation at a screening level whilst
simulation tests are adequate to assess degradation kinetics, degradation half -lives,
inf ormation about mineralisation, non-extractable residues (NERs) and
transf ormation/degradation products (extracted residues).
 Chapter R.11: PBT/vPvB assessment

62 Version 4.0 – December 2023

In order to select the appropriate test type, caref ul consideration of the physico -chemical
properties and the environmental behaviour of a substance is required, which is discussed later
on in this section.

For f urther inf ormation on test descriptions ref er to the degradation guidance ( see Sections
R.7.9.3 and R.7.9.4 in Chapter R.7b of the Guidance on IR&CSA).

Tests on ready biodegradation

Tests on ready biodegradation are described in OECD TG 301 A-F and OECD TG 310.
Biodegradability in Seawater test (OECD TG 306) can also be used to describe the ready
biodegradability in sea water. Degradation is f ollowed by determination of parameters such as
dissolved organic carbon (DOC), CO2 production or oxygen uptake. The parameter measures
the mineralisation and the pass level is set to 60% (ThOD or ThCO 2) or 70% f or DOC removal
assuming that the yield f or growth of the microbial biomass is 30-40%. In the context of ready
biodegradability, test substance-specif ic analysis can also be used and primary degradation
and f ormation of any transformation/degradation products can be assessed. Measurement of
primary degradation is however a requirement only in the MITI I test (OECD TG 301C).

Due to the f act that the test methodology f or the screening tests on ready biodegradability is
stringent, a negative result does not necessarily mean that the substance will not be degraded
relatively f ast under environmental conditions. A lack of biodegradability may f or example be
caused by toxicity of the substance towards microorganisms due to the very high
concentration employed in ready biodegradability tests compared with lower , environmentally
relevant concentrations. Another reason f or negative outcomes in ready biodegradability tests
can be low water solubility of the test substance. A low solubility could constitute the rate
limiting step f or degradation at the environmentally unrealistic high test substance
concentrations and not the intrinsic recalcitrance towards microbial transf ormation. ISO
method 10634 and Annex III of OECD TG 301 also describe options to address poorly soluble
substances.

Given the time, costs and, in some cases, practical dif ficulties associated with conducting and
interpreting a simulation degradation test, an enhanced ready biodegradation test design
of fers a cost-ef f ective intermediate screening test in those cases where persistence in the
environment is not expected although (a) standard ready biodegradation test(s) give(s) the
result “not readily biodegradable”. The purpose of enhancements should only be to
compensate the poor bioavailability to the degrading microorganisms due to poor solubility
and/or adsorptive properties of the substance, but should not be used to induce additional
adaptation of the inoculum. If suf f icient degradation is shown in an enhanced ready
biodegradation test, i.e. the pass level as given in the test guidelines f or ready biodegradation
is reached, the substance can be considered as “not P”. It should be noted that, in this case,
the 10-day window indicated in the corresponding test guideline does not need to be f ulf illed.
According to Gartiser et al. (2022), similar to inherent biodegradation tests, degradation below
20% in enhanced ready biodegradability tests could be used as part of Weight-of-Evidence to
indicate persistence. In general, under PBT assessment the conclusion that a substance is P/vP
cannot only be based on screening level inf ormation (including enhanced tests). However, if
based on the structure of the substance (e.g. perf luorinated substances with covalent C-F
bonds) it is known to be resistant towards degradation based on scientif ic evidence, screening
level inf ormation would be adequate to conclude a substance as P/vP , unless higher tier
studies indicate non-persistence. Furthermore, as explained in the section below, under some
 Chapter R.11: PBT/vPvB assessment

63 Version 4.0 – December 2023

conditions results of inherent biodegradability test may provide suf f icient inf ormation to
conf irm that the P-criteria are f ulf illed without the need f or f urther simulation testing.

More inf ormation on modif icationns of screening tests such as ready biodegradability tests or
enhanced ready biodegradability tests is contained in Sections R.7.9.4 and R.7.9.5 of Chapter
R.7b of the Guidance on IR&CSA. Please note that these tests are ref erred to as “enhanced
ready biodegradability tests”. Allowed enhancements are prolongation of the test duration and
testing in larger vessels. Only one enhancement is allowed at a time. The prolongation of the
test duration up to 60 days should only be considered if some initial, slow but steady,
biodegradation was observed not reaching a plateau by the end of the ready biodegradability
test, i.e. af ter 28 days.

The variation in the degradation potential of dif f erent constituents must be considered when
conducting screening tests with multi-constituent or UVCB substances (see further inf ormation
in Section R.[Link]). In general, f easibility of testing the whole substance to demonstrate
degradation of all of its constituents must be evaluated.

Tests on inherent biodegradation

Tests on inherent biodegradability are useful to give an indication of biological degradability on
a screening level. Inherent tests are similar to ready biodegradability tests as they usually
measure sum parameters and are conducted with a high test substance concentration and an
even higher microbial concentration. In general, they use more f avourable, if not optimal,
conditions than ready biodegradability tests (e.g. with increased biomass to test substance
ratio and allowing pre-adaptation of the microbial inoculum), and are hence designed to show
whether a potential f or degradation exists.

Due to the more f avourable conditions of an inherent test, results need to meet specific criteria
(specif ied in Table R.11—4 above and Section R.[Link] “Data on degradation/biodegradation”
of Chapter R.7b of the Guidance on IR&CSA) in order f or a substance to be considered as not
P/vP.

Lack of degradation (<20% degradation) in an inherent biodegradability test equivalent to the

OECD TG 302 series may provide suf f icient inf ormation to conf irm that the P-criteria are
f ulf illed without the need f or f urther simulation testing f or the purpose of PBT/vPvB
assessment. Additionally, in specif ic cases it may be possible to conclude that the vP-criteria
are f ulf illed with this result if there is additional specif ic inf ormation supporting it (e.g., specific
stability of the chemical bonds). The tests provide optimum conditions to stimulate adaptation
of the micro-organisms thus increasing the biodegradation potential, compared to natural
environments. A lack of degradation theref ore provides evidence that degradation in the
environment would be slow. Care should be taken in the interpretation of such tests, however,
since, f or example, a very low water solubility of a test substance may reduce the availability
of the substance in the test medium. These issues are discussed in more detail in Sections
R.7.9.4 and R.7.9.5 of Chapter R.7b of the Guidance on IR&CSA.

Tests on simulation of biodegradation

In principle, degradation simulation studies performed in appropriate environmental media and
at environmentally relevant conditions are the only tests that can provide a def initive
degradation half -lif e that can be compared directly to the persistence criteria as def ined in
REACH Annex XIII. Such tests allow both biotic and abiotic degradation processes to operate.
The simulation tests as described in OECD TGs 307, 308 and 309 address the f ate and
 Chapter R.11: PBT/vPvB assessment

64 Version 4.0 – December 2023

behaviour of a substance as it may be expected in the environment including inf ormation

about partitioning in the test system, primary or complete degradation, adsorption behaviour
and route(s) of transf ormation/degradation (transf ormation/degradation products). The
endpoints that need to be addressed are primary or ultimate degradation rate and degradation
half -lif e (DegT50) or dissipation half -lif e (DT50) f or the compartments included in the test
system as well as the route of degradation, transf ormation/degradation products and non-
extractable residues. In addition, a mass balance is included in these tests and theref ore
possible losses f rom the test system during the test period can also be quantif ied. An
incomplete mass balance will introduce severe uncertainty to interpretation of data. This in
turn can ultimately impede that a substance can be assessed with suf f icient certainty and
possibly cause the need to repeat the respective test and/or to give a low weight to the test
and its results in the P/vP assessment as part of a weight-of-evidence approach. A simulation
study should be perf ormed using a radio-labelled molecule, whenever f easible.

In order to evaluate the outcome of a simulation test, the reporting of the results should f ollow
the respective test guideline(s). Tests should report the degradation rate (or degradation half -
lif e) in each medium determined through mineralisation, e.g. volatile 14C-CO2, and/or direct
substance analysis. An option, if measuring mineralisation, is to measure the mineralisation
rate f or the whole system: if the mineralisation half -lif e f or the whole system is below the
respective half -lif e –value of P/vP criteria, it has been shown that the substance is not
persistent in the tested environmental compartment (surf ace water, sediment or soil).
However, investigation of degradation pathways/transf ormation/degradation products would
be needed since it cannot be excluded that a second transformation route f orms a persistent
transf ormation/degradation product in concentrations relevant f or the P assessment. When the
mineralisation half -life for the whole system is above the P criterion, f urther analysis of the
data is needed. A f ull mass balance of the substance and any transf ormation/degradation
products should be determined (or justification provided if this is not technically f easible), and
a determination of the level of non-extractable residues should be included. In general,
determination of non-extractable residues is recommended in soil and water-sediment studies
(OECD TG 307, OECD TG 308 and Kästner et al., 2014). Determination of non-extractable
residues is also recommended in surf ace water simulation degradation studies (OECD TG 309)
especially when relevant for mass balance calculations and derivation of degradation half -lif e.
In all cases, the extraction method f or NER and the choice of extraction solvents should be
justif ied (see more detailed inf ormation on NER below in this section). Where primary
degradation is observed, the identity of possible relevant transformation/degradation products
must also be determined and/or evaluated as regards their possible PBT/vPvB -properties.
Where only degradation of the parent substance is monitored, this does not address all the
concerns and f urther assessment of the transformation/degradation products may be required
in order to complete the PBT/vPvB assessment (see Sections R.7.9.4 and R.7.9.5 in C hapter
R.7b of the Guidance on IR&CSA).
27
Degradation half -lif e (DegT50 ) derivation

For direct comparison to the P/vP criteria only estimates of degradation half -lif e (DegT50) are
appropriate. Such estimates are mostly based on data derived f rom simulation biodegradation

27
DegT50 abbreviation is used only for the purpose of guidance documents published by ECHA to describe the
degradation half-life and may differ from abbreviations of half-life used in other guidance documents.
 Chapter R.11: PBT/vPvB assessment

65 Version 4.0 – December 2023

tests. Degradation half -lif e (DegT50), is the time taken f or 50% degradation of a test
substance when the degradation can be described by (pseudo-) f irst-order kinetics, i.e where
the degradation rate constant is independent of concentration and time . The methods f or
estimating degradation half -life are described in the paragraphs below. It is important to note
that a dissipation half -lif e (DT50) is ref erring to the overall process leading to the
disappearance of the test substance f rom the test system (or one compartment of the
system). Dissipation comprises two main types of processes: degradation processes (such as
microbial degradation, hydrolysis and/or photolysis transforming substances into degradation
products) and transfer processes (such as volatilisation, and adsorption). Both processes may
af fect the disappearance of the test substance f rom the system. If transf er processes have
occurred simultaneously with degradation, the DT50 value is not representative of the DegT50
value.

In the f ollowing paragraphs advice is provided on the degradation half -lif e estimation, the
volatilisation correction of the data as well as normalisation approaches in the cases when an
incomplete mass balance is observed. Further inf ormation on the degradation kinetic models,
the data handling, assessment of the goodness of f it and general recommendations on t he
kinetic analysis can be f ound in the Generic Guidance Document f or Estimating Persistence and
Degradation Kinetics f rom Environmental Fate Studies on Pesticides in EU Registration
(FOCUS, 2014).

Lag phase of degradation could be occasionally observed in simulation studies. A lag phase
describes the phase when microbes are ‘adjusting’ to the new substrate (f ood source) and/or
new environment conditions and depends on the cell density in tests, the possible pre -
adaptation of the inoculum and the total amount of specif ically degrading bacteria (Ingerslev
et al., 2000). In the OECD 309, lag phase is def ined as the time f rom the start of a test until
adaptation of the degrading micro-organisms is achieved and the biodegradation degree of a
chemical substance has increased to a detectable level. OECD TG 309 includes also more
specif ic advice indicating that the lag phase duration is estimated f rom the degradation curve
(semi-logarithmic plot) by extrapolating its linear part to zero degradation or alternatively by
determining the time f or approximately 10% degradation.

In simulation studies (OECD TG 309, OECD TG 308 and OECD TG 307), the possibility to
detect a lag phase and to accurately def ine its duration depends on the lowest degradation
(e.g. a decrease in test substance concentration, or amount of CO2 produced) that is
detectable by the corresponding test method (inf luenced e.g by the test substance
concentration and the analytical method used, on the number and f requency of measurements
and the degradation rate). Therefore, in simulation testing, even not detectable biodegradation
of the test substance might occur already during the lag phase. When a lag phase occurs in
simulation tests the estimated length of the lag phase should be reported, together with the
explanation how it is determined (e.g. based on detection limit of the method or another
def inition, or whether the value is derived f rom data analysis sof tware). In addition, ef f orts
should be made to distinguish whether the observed lag phase can be attributed to any
experimental artefacts. Justification f or the treatment of the lag phase length in the DegT50
derivation should be provided. When the lag phase is attributed to experimental artef act the
validity of the study needs to be assessed carefully as this might indicate issues related to the
test design and perf ormance.
 Chapter R.11: PBT/vPvB assessment

66 Version 4.0 – December 2023

The kinetic model that best f its and/or most appropriately describes the experimental data
28
should be used f or estimating the degradation half -lif e . Prior to degradation kinetic analysis, a
qualitative assessment should be made whether the degradation pattern observed f rom the
experimental data is representative of the degradation of the substance under the test
conditions and not the result of experimental artef acts. In general, ef f orts to identif y
experimental artefacts in laboratory studies must be made. In the case experimental artef acts
(e.g. adsorption to test vessels) are observed the acceptability of the study must be caref ully
investigated. For DegT50 derivation model f its should initially always include all the data (f or
data sets with a lag phase this does not always apply; see FOCUS 2014 and OECD TG 309
paragraph 44 f or f urther inf ormation). After the initial f it, any outlier data points could then be
excluded, and the f it should be repeated. The decision to exclude any data point as an outlier
is usually based on a clear deviation f rom the values (or the trend of the values) of the other
data points in the same data series and/or f rom the f itted curve, or inf ormation on issues
related to the perf ormance of the test. The use of statistical methods to identif y outliers is
desirable, to remove ambiguity. In any case a justif ication should be provided f or excluding
any data point.

The selection of a degradation kinetic model should be based on the assessment of the metrics
f or determining the “goodness of f it” which include visual assessment of goodness of f it, χ2
error and t-test statistical metric. Detailed description f or the criteria f or the acceptability of
the f it is included in FOCUS guidance (2014). Regardless of the kinetic model selected f or
describing the data, a justif ication should be provided with adequate and reliable
documentation on the applied method. Estimated DegT50 can depend on the model used to f it
the experimental data.

When the kinetic of decline is f irst-order and no lag phase occurs, the degradation half -lif e
predicted by single-f irst order kinetic model (SFO) can be used f or direct comparison with the
P/vP criteria. Degradation can f ollow different kinetic patterns which are inf luenced by the test
conditions (e.g. matrix properties, temperature, microbial composition) and the test substance.
The use of the SFO model f or all types of kinetic pattern might lead to under - or
overestimation of the derived DegT50 and thus introduce bias in the persistence assessment.

When the kinetics of decline are bi-phasic (f ast initial decrease in test substance concentration
29
f ollowed by a slower decline), the best-fit model (e.g. DFOP, HS) should be selected and used
f or predicting a degradation half -life DegT50. Bi-phasic kinetics (f ast phase f ollowed by slow
phase) are predominantly the result of the limited bioavailability of the substance in the
system due to its sorption to the solid matrices (soil and sediment or suspended matter) which
might lead to non-extractable residues (NERs) f ormation. When DFOP or the HS kinetic model
(both models allow deriving slow phase DegT50) is selected as the best f itting model, the
degradation half -lif e (DegT50) predicted f rom the slow phase should be pref erred f or

28
In the context of the Plant Protection Products legislation (EC 1107/2009) and specifically within the FOCUS
guidance (2014) a distinction is made between trigger and modelling endpoints, for the purpose of the REACH
legislation and the PBT assessment according to Annex XIII this distinction does not apply and as described above the
kinetic model that most appropriately describes the observed data should be used.

29
The DFOP and HS biphasic kinetic models are based on first order degradation kinetics. The DFOP model (Double-
First-Order in Parallel model) consist of two SFO models in parallel (the sum of two first order equations), and the HS
model (Hockey-Stick model) consists of two SFO models in series (two sequential first order curves).
 Chapter R.11: PBT/vPvB assessment

67 Version 4.0 – December 2023

comparison with the P/vP criteria. Use of bi-phasic kinetic models is recommended to cases
where an acceptable single-first order (SFO) f itting is not possible. In any case, a justif ication
f or the selection of a model should be provided with adequate and reliable documentation such
as the key parameters of the kinetic analysis and assessment of the goodness of f it.

DegT50 based on the slow phase represents a conservative approach compared to overall
DegT50. If the slow phase DegT50 is above the P and/or vP it can be expected that the
concentration of the substance in the environment would increase over time and that the
substance would persist in the environment in the long term. When such substances are
released to the environment, their accumulation in the environment is in practice dif f icult to
reverse as cessation of emission will not necessarily result in a reduc tion in substance
concentration. The slow degradation phase may represent a small f raction of the degradation
of the substance under the degradation simulation test conditions. However, it is considered
that the degradation rate at the slow phase better describes the degradation rate in the
environment over time due to the concentration build up and partitioning properties of PBT
and/or vPvB substances. For these reasons, when the kinetics of decline are bi-phasic and the
DFOP or HS model is used, the more conservative DegT50 slow phase should be pref erred
(over the f ast phase DegT50 or overall study period DegT50) and compared to the P/vP criteria
of REACH Annex XIII.

When there is no signif icant measurable degradation observed during the test and the kinetic
model indicates that the relevant rate constant is not signif icantly dif f erent f rom zero the
calculated degradation half -lives should be interpreted with care. In such a case it is still
possible to reach a conclusion on persistence as demostrated in the cases of substances
included in the candidate list (e.g. Melamine, EC 203-615-4 and 1,4-dioxane EC 204-661-8).

The First Order Multi-Compartment (FOMC) model, also mentioned in the FOCUS guidance is a
bi-phasic mechanistic model based on the soil heterogeneous nature (FOCUS, 2014). The
model divides the soil into large number of sub-compartments each with dif f erent f irst order
degradation rate constants. The use of the DegT50 derived f rom the FOMC model can be
considered in a Weight-of-evidence approach only if the other models do not f it the data
adequately. Furthermore, when the FOMC model is the best f it model, the derivation of a
pseudo-DegT50 f rom a DegT90/3.32 is used f or specif ic purposes in the FOCUS guidance
(FOCUS, 2014). The denominator value of 3.32 is the mathematical ratio between DegT90 and
DegT50 derived f rom single f irst order kinetics modelling. The recalculated pseudo -DegT50 is
longer than the overall DegT50 and might be highly uncertain. The uncertainties associated
with the pseudo-DegT50 are associated with the f act that the mathematical ratio of 3.32 is
derived f rom empirical model (SFO) while the FOMC, as stated above, is a mechanistic model
and that it is not representative to the actual mathematical ratio between DegT90 and DegT50
of the FOMC model. Furthermore, the FOMC model was developed f or the soil compartment
and thus its application in the context of the OECD 309 and OECD 308 should be considered
caref ully (Gustaf son and Holden, 1990). Considering the uncertainties around the DegT50
values derived using the FOMC model, this model is the less pref erred one to be used f or the
PBT assessment.

The mass balance/recovery f or a substance should be provided and deviations f rom the
recommended mass balance/recovery, as they are described in the corresponding testing
guidelines (OECD TG 309, OECD TG 308 and OECD TG 307) should be reported and justif ied.
 Chapter R.11: PBT/vPvB assessment

68 Version 4.0 – December 2023

In case the mass balance decreases over the study, it could suggest possible losses of the test
substance or transformation/degradation products f rom the test system by volatilisation. Other
possible reasons f or an incomplete mass balance could be e.g. losses due to adsorption or
during sampling/sample treatment. In the absence of inf ormation on the cause of incomplete
mass balance, the unknown fraction is considered non-degraded parent substance f or deriving
the DegT50. If incomplete mass balance is due to experimental errors (e.g. f ailed
sampling/measurement) and it is observed f or f ew data points, these data points should be
30
excluded f rom the kinetic analysis and the DegT50 derivation. In any case a justif ication
should be provided f or excluding any data point.

In case of incomplete mass balance the test should be treated with caution, nevertheless the
f ollowing three scenarios exist f or deriving the DegT50 (Table R.11—6). If none of these
scenarios applies the study should be considered inconclusive.

30
A robust DegT50 derivation is only possible if the number of observations is appreciably larger than the number of
model parameters. However, the number of data points should not be a sole reason for rejecting a study prior to
kinetic analysis.
 Chapter R.11: PBT/vPvB assessment

69 Version 4.0 – December 2023

Table R.11—6: Scenarios for deriving DegT50 for substances with incomplete mass
balance.
Scenarios/method of
Description Conclusion
calculations

1/ DegT50 without This scenario assumes that the mass balance P

normalisation loss is degraded parent substance. It is
vP
therefore the less conservative approach. As this
is the less conservative option, the DegT50 If you cannot conclude P
obtained without normalisation can be used to or vP run scenarios 2
and/or 3
conclude P or vP.
As this option is not conservative enough, it
must not be used to conclude a substance as not
P or as not vP.

2/DegT50 with This scenario can be used under the following P

31 two circumstances:
normalisation vP
i. The substance
concentration/radioactivity measured at If you cannot conclude P
the 1st time point (zero time point) or vP, or the substance is
differs from the initial applied P but it is unclear if vP (in
concentration/applied radioactivity of the particular for substances
substance and the mass balance remains having vB properties) run
constant over the testing period. scenario 3
ii. The decline rate of the mass balance is
constant over the testing period.
As this option is not conservative enough, it
must not be used to conclude a substance as not
P or as not vP.
3/ DegT50 with In the absence of information on the cause of Not P
unknown fraction incomplete mass balance, the unknown fraction
considered as non- is considered non-degraded parent substance for Not vP
degraded parent
32 deriving the DegT50 (representing a worst-case
P/vP as part of a WoE only
scenario).
This option can be used to conclude a substance
as not P or not vP as this is the most
conservative approach.

When dissipation through volatilisation is observed, correction procedures can be applied (see
also paragraph on Volatile substances below in this Section, Section R.[Link].3 and related
Appendix R.11-7). For this purpose, correction procedures are described f or the parent
substance f or the SFO kinetics model which could be also applied in the case of HS and DFOP
kinetic model. The correction procedures assume that the volatile losses of parent substance
were adequately identif ied and quantif ied in the volatile trapping systems. Each time these
correction procedures are used, a justif ication should be made f or their use. These approaches

31
DegT50 with normalisation: for each data point the residual parent substance can be calculated either as:
(i) a percentage of the initial measured concentration (or the sum of parent substance and transformation
products) or as a percentage of the initial measured applied radioactivity
(ii) as a percentage of the total recovered measured applied radioactivity at each time point.
32
DegT50 with unknown fraction: for each data point the unknown fraction is considered as non-degraded parent
and is expressed as a percentage of the initial nominal radioactivity applied.
 Chapter R.11: PBT/vPvB assessment

70 Version 4.0 – December 2023

may also apply f or transformation/degradation products, though a justif ication f or its use must
be made which accounts for when the transformation/degradation products are f ormed in an
experiment. Further advice on how to handle data on non-extractable residues in simulation
studies is provided in Appendix R.11-4. Advice on sterile controls, which are also relevant f or
the estimation of DegT50, is provided Section R.[Link] in Chapter R.7b of the Guidance on
IR&CSA.

Specific advice on DegT50 derivation for transformation/degradation products

When a degradation study is perf ormed on a transf ormation/degradation product of a

substance to be assessed, the data should be analysed as described f or the parent substance
above. When a study is perf ormed on a parent substance and transf orm ation/degradation
products are f ormed, the pathway model approach should be used as described below. For
deriving a reliable degradation half -lif e f or a transf ormation/degradation product, a good
knowledge of the degradation pathway up to this transf ormation/degradation product is
essential.

The initial approach f or determining the half -lives f or transf ormation/degradation products is
f itting the data with the stepwise pathway approach model, in which parent and
transf ormation/degradation data are assessed together. In a sequential and stepwise manner,
each transformation/degradation product is added to the model and the parameters f or the
newly added transformation/degradation product are f itted while the parameters f or the other
previously f itted parent or transf ormation/degradation product are f ixed to their estimated
value. The kinetic model used f or f itting the parent substance data should be considered
together with the number of the f itting parameters of the transformation/degradation product
kinetic model. In addition, the kinetic model should cover both, the f ormation phase, which
accounts f or when the transformation/degradation product is f ormed, and the decline phase.
The stepwise approach may be the preferred way in cases where the degradation pathway is
very complex and includes many transf ormation/degradation products.

Alternatively, one can proceed straight to the complete pathway model approach (cf .
simultaneously f itting the parent and all transf ormation/degradation product data), especially
when the pathway is well understood and/or less complex (e.g., only f ew
transf ormation/degradation products are to be considered) and when the complete pathway
model f it generates a good representation of the experimental data, a good visual f it, and
reasonable parameters f or all transf ormation/degradation products.

Another option is to evaluate the transf ormation/degradation products data individually by

using only the decline phase (Decline model). This option should be used only if the pathway
f it does not appropriately describe the data. It is important to note that the degradation half -
lives f rom a transformation/degradation product decline model f it are more conservative than
the degradation half-lives f rom a pathway model. This is because in a pathway approach model
the transf ormation/degradation product is still being f ormed af ter the peak is reached.

Considerations f or simulation testing strategy

Annex IX to the REACH Regulation lists three simulation degradation tests as standard
endpoints f or the CSA (which, according to Annex I to the REACH Regulation, includes the
quantitative risk assessment and the PBT/vPvB assessment).

The P/vP assessment should cover all three (f ive) environmental compartments (water, marine
water, sediment, marine sediment, soil). However, a substance can already be concluded as P
 Chapter R.11: PBT/vPvB assessment

71 Version 4.0 – December 2023

or vP if the criteria are f ulfilled f or one compartment only. For the purpose of reducing ef f orts
of testing, the test should be selected in such a way that it ref lects the worst case of
persistence potential (f or which the degradation half -life is close or is more likely to exceed the
P/vP criteria ). This would also ideally be the environmental compartment with the best
possibility to use the results f or concluding the P/vP-assessment (as being “worst case”).

The inf luence of the relevant environmental compartment(s) in terms of exposure potential
based on f ate properties, the identif ied uses and release patterns to the order of testing also
need to be considered. In some cases, it may be necessary, and hence acceptable, to choose
an environmental compartment f or simulation degradation testing other than the one normally
considered as the f irst pref erence (see discussion below).

The f urther elements to be considered when choosing the environmental compartment(s) f or

testing are described in the context of the ITS (Figure R.11—3).

Bef ore testing, the simulation test(s) that is(are) the most appropriate f or addressing
degradation should be identif ied. This is f urther discussed below.

Simulation studies on ultimate degradation in surf ace water are warranted unless the
substance is highly insoluble in water. If a substance is highly insoluble in water it may not be
technically possible to conduct a simulation study that provides reliable results, and at very low
concentrations technical issues may make it very dif f icult to establish a reliable degradation
curve in the study. Therefore, depending on the substance physico-chemical properties and the
availability of good quality analytical methods f or identif ication and quantif ication, it may not
be possible to conduct this study if the water solubility of the substance is very low (typically
<1 µg/L). The surf ace water transf ormation test (OECD TG 309) recommends using a test
substance concentration f or the kinetic part of the study in a range which is environmentally
realistic, i.e. in a range of “less than 1 to 100 µg/L”. The part of the study perf ormed to
identif y degradation pathways may usea a higher test substance concentration to ease the
analytical identif ication and characterisation of the transf ormation/degradation products.
Further considerations on the OECD TG 309 study are provided below.

Testing in the aquatic compartment (OECD TG 309) is the pref erred f irst step when there is a
need f or f urther inf ormation on persistence in the environment, considering the f ollowing
reasons:

• Firstly, the aquatic compartment is considered to be a relevant environmental

compartment f or persistence assessment because the criteria f or B/vB and T are mainly
based on tests perf ormed in this compartment. In addition, by def ault, water
compartment receives a signif icant amount of emissions, directly or indirectly, and
transports/distributes the substance through e.g. deposition and run-of f (unless
evidence f rom substance emission data suggests otherwise). Once entering water, a
substance may reside there f or very long time and be spread over long distances before
it reaches other environmental compartments (via environmental transport, partitioning
and distribution processes) such as sediments or (via air) the soil compartment.

• The OECD TG 309 minimises potential NER f ormation. If NER is f ormed at signif icant
levels in OECD TGs 307 and 308 tests, this can be dif f icult to interpret and compare
with the degradation half -lif e criteria of Annex XIII to the REACH Regulation.

Reasons to deviate f rom this general approach can be that:

 Chapter R.11: PBT/vPvB assessment

72 Version 4.0 – December 2023

• The substance is a multi-constituent / UVCB substance, which affects the concentration

at which the test can be perf ormed (due to dif f erent multiple water solubilities of the
individual components).

• Indications f rom available data (e.g. literature) suggest that persistence is likely to
occur in a dif f erent environmental compartment (i.e. in soil or sediment), including
evidence of strong sorption potential and direct or indirect emission to specif ic
compartment.

• Aquatic testing is not technically f easible, i.e. it has been impossible, with allocation of
reasonable efforts, to develop suitable analytical methods and other test procedures to
accomplish testing in surf ace water so that reliable results can be generated. This may
be the case in particular if the water solubility of the test substance is very low.
Appropriate analytical methods should have suitable sensitivity to detect relevant
changes in concentration (including transformation/degradation products). Please see
also substance type specif ic considerations in Section R.[Link].

• OECD TG 309 should be perf ormed at concentrations between 1 and 100 μg/L and
pref erably in the range of <1-10 μg/L (to ensure that biodegradation f ollows f irst order
kinetics). Generally, when water solubility of a substance is very low (typically <1
μg/L), testing on sediment and/or soil will be pref erred, if aquatic simulation
degradation testing is not technically f easible due to analytical limitations and low
solubility of the test substance. Pre-test is recommended to assess technical f easibility
of the study if substance properties indicate the need.

• High sorption af f inity of cationic substances to particles may lead to reduced

degradation rate in sediment and soil. Therefore, OECD TG 309, in which particles are
sparse, may not be the worst-case test f or cationic substances and OECD TG 308 or
OECD TG 307 could be a better option as a f irst simulation test to be conducted (Claßen
et al. 2021 and Holzmann et al. 2022).

Soil/sediment simulation degradation testing may be warranted as a f irst test in the above-
listed cases. In addition, as described in the ITS (Figure R.11—3), the soil and sediment
degradation simulation tests may be needed when results f rom simulation tests in water do
not exceed the P/vP criteria but there are indications that the substance or its degradation
products could persist in soil and sediment, meeting the respective P criteria.

Bef ore performing a soil or a sediment simulation degradation test, it is worth noting that f or
the purpose of quantitative risk assessment and f or adsorptive substances, a simulation test in
soil (OECD TG 307) could be more relevant than a simulation test in sediment (OECD TG
33
308) . Degradation rates/half-lives f rom simulation tests in soil can be used instead of generic
values f or the assessment of PECsoil. While degradation rates/half-lives f rom simulation tests
in sediment can be taken into account f or the calculation of the PECregional, in practice this
would have only a negligible inf luence on risk assessment.

33
Removal of the substance during the WWTP process may be taken into account when considering the emissions
through WWTP in relation to the relevance of the simulation test compartment (e.g. incineration of the sludge and
removal or degradation during the water treatment process).
 Chapter R.11: PBT/vPvB assessment

73 Version 4.0 – December 2023

Once the appropriate simulation test(s) have been identif ied and conducted, data need to be
interpreted to determine environmental degradation half -lives. A prerequisite f or data
interpretation is that exhaustive extraction methods are used to ensure that suitable data are
generated. Guidance on how to conduct the test and interpret data f rom a simulation test is
available in the present Guidance document and in Section R.7.9.4 of Chapter R.7b of the
Guidance on IR&CSA.

OECD TG 309

OECD TG 309 should be perf ormed at concentrations between 1 and 100 µg/L and pref erably
in the range of <1-10 μg/L (to ensure that biodegradation f ollows f irst order kinetics).
However, f or low solubility substances, even if their water solubility is within this range, it is
acknowledged that the f easibility of the test depends, inter alia, on the possibility to develop
with reasonable ef f orts appropriate analytical methods with suitable sensitivity to detect
relevant changes in concentration (including transf ormation/degradation products).

According to OECD TG 309, the test is applicable f or non-volatile and slightly volatile
substances. The f easibility of the surface water simulation test f or volatile substances depends
on the possibility to minimise volatilisation and assure that the test substance is maintained in
the water phase, and thus, is available f or microorganisms to the extent that a reliable
degradation half -lif e can be determined. It is also important to ensure that the volatilised
f raction is adequately trapped and quantif ied in order to be able to interpret the results
reliably . Further inf ormation on how to address volatilisation in simulation testing can be f ound
in the paragraph of “Volatile substances” below and in section R.[Link].3.

Also hydrophobic substances with low water solubility are challenging to add to the test
system. OECD TG 309 allows using an organic co-solvent f or adding poorly water-soluble
substances but requests that steps are taken to minimize actual co-solvent concentrations in
the test. Passive dosing and microvolume spiking have recently been proposed as approaches
to avoid or at least minimize co-solvent addition (Birch et al, 2023). However, currently there
is still limited experience in using such methods in regulatory persistence assessment and
f urther developments and discussions are thus f [Link] TG 309 uses as a def ault one
matrix sample, which is in contrast to the soil (4 soils) and sediment (2 sediments) simulation
studies. Nothing prevents registrants from employing or authorities f rom requesting simulation
degradation testing in more than one surf ace water. It is generally recommended to consider
perf orming the test with more than one water source.
In OECD TG 309, there are options to perf orm the test as a ‘pelagic test’ or as a ‘suspended
sediment test’. In both cases, the coarse particles are removed f rom the water sample, f or
example by f iltration through a f ilter with 100 µm mesh size or with a coarse paper f ilter, or by
sedimentation. For the ’suspended sediment test’, surf ace sediment is added af terwards to
obtain a suspension.

Use of sterile controls is required in the OECD TG 309. More guidance on the sterile controls
and suitable sterilisation methods are provided in the Section R.[Link]. in Chapter R.7b of the
Guidance on IR&CSA and in ECHA Note on Sterile controls available on ECHA Website.

For the PBT/vPvB assessment, the amount of suspended matter in the pelagic test should be
representative of the level of suspended solids in EU surf ace water. For large rivers, the
concentration of suspended matter (SPM) is reasonably constant and a n EU def ault of
15 mgdw /L has been proposed, e.g. f or the implementation of the Water Framework Directive
 Chapter R.11: PBT/vPvB assessment

74 Version 4.0 – December 2023

(European Commission, 2018) or in EUSES. For marine waters, a def ault SPM concentration of
3 mgdw /L has been proposed f or the Water Framework Directive. Similarly a n SPM
concentration of 5 mgdw /L has been implemented in EUSES f or marine waters. For REACH,
using natural surf ace water containing between 10 and 20 mg dw /L SPM f or simulation tests in
f reshwater and ca. 5 mgdw /L f or simulation tests in marine water is considered acceptable.
Further details are available in Section R.[Link] of Chapter R.7b of the Guidance on IR&CSA.

Even if more than one water source is used in the assessment, it is recommended that the
amount of suspended solids still ref lects realistic concentrations f or the EU surf ace waters (e.g.
samples f rom dif ferent water bodies or to ref lect seasonal variations in the concentration of the
suspended solids). When the test concentration is well under the water solubility limit of the
substance, one might consider testing several water sources instead of testing two
concentrations of the test substance. In any case, a ref erence substance should be used to
demonstrate the viability of the system.

According to the OECD TG the ’suspended sediment test’ can be used to simulate surface
water free of coarse particles or turbid surface water (which might exist near the water-
sediment interface). Ingerslev and Nyholm (2000) f urther indicate that conducting the tests
with added suspended sediment signif icantly enhance the biodegradability of some of the test
substances. However, this test design is generally not recommended f or P testing purposes as
such highly sediment particle loaded surface water systems are not the most prevailing ones.
There is also a high probability that increasing the suspended solids concentration will increase
the potential f or NER f ormation and to avoid this the pelagic test without artif icially added
particular material/sediment particles is pref erred. In specif ic cases where there is a need to
address the inf luence of the suspended solids to the abiotic degradation rate in the surf ace
waters, the addition of suspended solids may be justified. If suspended solids are added, it is
recommended that a magnetic stirrer bar should not be used f or agitation as it may grind the
solids/sediment and result in increased levels of NER. Other methods are reco mmended
instead, e.g. shaking of test vessels (Shrestha et al., 2016). According to OECD TG 309,
agitation should be continuous and as gentle as possible, while still maintaining a
homogeneous suspension.

Following attempts have been reported to decrease variability in OECD TG 309 test system by
increasing the amount suspended solids in the test media. It has been demonstrated that 10-
f old higher sediment content (10 g/L) compared to allowed in standard suspended sediment
OECD TG 309 test reduced interreplicate variability (Seller et at. 2020). Tian et al. (2023)
reported that in modif ied OECD TG 309 with sediment and water samples f rom wastewater
impacted rivers having even higher suspended solids concentration (50 g of wet solids/L) and
mixture of substances, low spiking concentrations (0.5 µg/L) produced degradation rate
constants comparable to those derived f rom the non-spiked test. However, half -lives obtained
f rom studies with suspended solids concentration much higher than allowed in the respective
TG or studies with water and sediment with previous exposure of the substance cannot be
directly compared with Annex XIII P criteria.

In order to minimise the f ormation of NERs, it is recommended that the pelagic test be
considered f irst before conducting any other simulation tests. However, it is worth noting that
even f or the ‘pelagic test’, the test water will contain suspended matter onto which the test
substance and/or its transf ormation/degradation products can adsorb. Theref ore, the
f ormation of NERs may be signif icant in the ‘pelagic test’ too. It is thus necessary f or this test
 Chapter R.11: PBT/vPvB assessment

75 Version 4.0 – December 2023

as well to quantif y the NERs and to explain and scientif ically justif y the extraction procedure
and solvent used (see also Section R.[Link]. in Chapter R.7b of the Guidance on IR&CSA).

Unless there is a specif ic concern for the marine compartment, f or the REACH PBT assessment,
generally the OECD TG 309 would be perf ormed using a f reshwater rather than salt water
media. However, the degradation in marine compartment should always be considered in PBT
assessment. It should theref ore be assessed if the inf ormation on degradation in f reshwater
may be used to extrapolate the degradation rate in marine environment.

The OECD TG 309 proposes quite a f lexible f ramework f or designing the test. The registrants
should provide justifications in the robust study summaries and/or test plan proposals f or the
dif f erent options taken with regard to, f or example the type and characteristics of the water
used, whether suspended sediments were added, whether shaking or stirring of the test
vessels was used, whether the test was perf ormed in the dark or with dif f use light.
The role of NER in P/vP assessment is discussed f urther in the section on “Non-extractable
residues” below.

OECD TG 308 & TG 307

Testing on sediment (OECD TG 308) or soil (OECD TG 307) may be needed instead of a pelagic
test (OECD TG 309) if the latter is not technically f easible; i.e. if the water solubility of the test
substance is very low (typically below 1 µg/L) or if it is not possible with reasonable ef f orts to
develop a suitable analytical method or other test procedures f or conducting the test in water.
Besides, in some situations it can be anticipated that the simulation test in water will not be a
worst case and that the persistence criteria will possibly be exceeded in sediment and/or soil
but not necessarily in water. This may be the case f or example f or strongly sorbing or
hydrolysable substances, as hydrolysis may be hindered by adsorption onto sediment and soil.
Another example is cationic substances f or which data suggests that degradation generally
seems to be f aster in aquatic than in sediment or soil systems (Claßen et al., 2021; Holzmann
et al., 2022). For volatile substances that have high adsorption potential, adsorption to
sediment or soil may lead to less volatilisation f rom the test matrices in OECD TG 308 and 307
than in OECD TG 309 test where volatilisation f rom the water phase could be higher. Testing on
sediment and/or soil may also have to be conducted in addition to the test in water, to
demonstrate that the substance is persistent in none of the compartments relevant f or the
PBT/vPvB assessment. Testing in the sediment and/or soil compartments should be considered
in particular if there is a specif ic concern f or this compartment, f or example, if direct or indirect
releases to these compartments are likely, or if the substance is predicted to accumulate in
these compartments. Koc or Kd values can be used as an indicator of whether the substance is
likely to be of concern f or the sediment and soil compartments. As a rule of thumb, substances
with log Koc > 4 are generally regarded as highly adsorptive and likely to distribute in sediment
and soil.

Guidance on how to apply sterile controls is provided in the Section R.[Link]. in Chapter R.7b
of the Guidance on IR&CSA. There are many benefits of the sterile controls in biodegradation
test described R.[Link]. Among others, sterile controls inf orm on maintenance of the test
material in the test system and, in particular, indicate potential technical problems due to the
physicochemical properties of the substance (adsorption, low solubility, volatility). The use of
sterile controls is already described in OECD TG 307 (including in the list of results to be
reported). This guidance recommends the use of sterile controls f or both OECD TG 307 and
OECD TG 308 tests.
 Chapter R.11: PBT/vPvB assessment

76 Version 4.0 – December 2023

For the PBT/vPvB assessment, a half -lif e in sediment should be estimated. However, f rom
OECD TG 308 simulation tests, the half -lives calculated f or the sediment phase and the water
phase separately are less reliable than the half -lif e calculated f or the total water-sediment
system. Still, because of the low volume and depth of water relative to the volume of sediment
and the surf ace of the water-sediment interf ace used in OECD TG 308, even moderately
adsorptive substances will tend to rapidly partition f rom the water phase to the sediment
phase. Therefore, f or adsorptive substances, the half -lif e in the sediment can reasonably be
estimated f rom the half -lif e f or the total water-sediment system. This approach avoids the
34
need to determine specif ic half -lives f or each phase separately (Honti and Fenner, 2015) .
However, the parent substance may degrade to more soluble and less adsorptive degradation
products that can be released f rom the sediment to the water phase. This should be taken into
account in the assessment.

OECD TG 308 outcome can be af f ected both by test vessel and system geometry and the
associated water-sediment interf ace size. Headspace volume and height of the water and
sediment columns can inf luence the partitioning and consequently degradation of the test
substance (Hennecke et al., 2014; Shrestha et al., 2016), especially f or volatile substances.
There is no specif ication of the vessel size or geometry in the test guideline , but the system
geometry should be consistent with the range indicated in the OECD TG 308 (i.e.
water:sediment volume ratio between 3:1 and 4:1, height of 2.5 cm (±0.5) layer and
minimum weight of 50g of the sediment). The dimensions of the test vessel should be included
in the robust study summary. Fenner et al. (2017) recommend several qualitative and
quantitative parameters to be considered f or interpretation of the OECD TG 308 results
including among others Koc, TOCw , f raction of organic carbon in sediment, test vessel
geometry, height of water and sediment layers and sediment porosity.

The OECD 308 TG states on several occasions that “disturbance of the sediment is undesirable
and should be avoided as f ar as possible” e.g. when introducing the test substance, during
aeration and sampling. Deviating f rom the TG recommendation, sediment spiking instead of
addition of the test substance via water may in some cases be necessary to ensure realistic
exposure of sediment in the test. This may be the case e.g. f or substances which would
transf er signif icantly quicker to the atmospheric compartment via volatilisation compared to
transf er to the sediment compartment. Pre-tests as well as f ugacity modelling (see also
paragraph on Volatile substances below in this Section, Sections R.[Link].3 and R.[Link].4
on Multi-media modelling) can provide useful inf ormation to identif y the proper test setup f or
the f inal test. Currently, however, deviations f rom the test guideline may only be justif ied on a
case-by-case basis as there are no generic criteria available when spiking would be
appropriate. If sediment spiking is undertaken, the overall half -life f rom such a test should be
assumed to be the sediment half -lif e (unless there is signif icant desorption, which seems
unlikely in the case of most potential PBT substances). Sediment spiking methods have been
developed so f ar f or the purpose of sediment toxicity testing. These are mentioned in Section
R.[Link] “Laboratory data on toxicity to sediment organisms” in Chapter R.7b of the

34
Part of LRI ECO18 – “Improved strategy to assess chemical persistence at the water-sediment interface”
[Link]
sediment-interface/
 Chapter R.11: PBT/vPvB assessment

77 Version 4.0 – December 2023

Guidance on IR&CSA. However, none can currently be directly recommended f or the purpose
of sediment simulation testing but f urther approach development would be necessary.

According to the OECD TG 308, the aerobic test simulates an aerobic water column over an
aerobic sediment layer that is underlain with an anaerobic gradient. Aeration of the test
system is needed in order to maintain aerobic conditions throughout the study. OECD TG 308
recommends aeration by gentle bubbling or by passing air over and gently stirring the water
surf ace in open test vessels (f or non-volatile substances), and by gentle stirring of the water
surf ace in closed, biometer type systems (f or slightly volatile substances). When testing
volatile substances in closed systems, regular exchange of headspace gas is necessary to
maintain aerobic test conditions. It should also be ensured that the oxygen is distributed f rom
the headspace to the water layer. However, any aeration method should disturb as little as
possible the sediment layer and its stratif ication. For example, visual assessment of
resuspension/turbidity of the overlying water is one indication of disturbed sediment.

In the OECD TG 308 agitation may modif y the stratif ication of the sediment, af f ecting the
maintenance of the anaerobic layer, and therefore, may have an inf luence on the degradation
process in the sediment simulation test. While, in the OECD TG 309 continuos agitation is
required to maintain particles and microorganisms in suspension and f acilitate the aerobic
conditions, such aeration should not be used in OECD TG 308. Aeration methods recommended
in the OECD TG 308 should be applied instead. If any other method is used, it must be
demonstrated that sediment stratif ication is maintained troughout the study.

Multiple simulation test results

A substance can be concluded to be not-P only if it can be demonstrated that it is not

persistent in any of the environmental compartments relevant f or the PBT/vPvB assessment,
i.e. water, sediment and soil.

Generally, f or substances registered under REACH, the likelihood of having more than f our
dif f erent results f rom the same environmental compartment is deemed to be limited. For
determining transformation rates, OECD TGs 307, 308 and 309 recommend respectively that
at least f our dif f erent soils and two dif f erent sediments and one type of water should be
tested.

For the same environmental compartment, when f our or less results are available, the most
stringent result should be used with respect to the PBT assessment.

Where more than f our results are available f or the same compartment, the f irst step is to
assess the validity of the data and whether the dif f erent tests are equivalent (f or example
temperature, pH, organic carbon content, microbial biomass, test design etc). Only test results
corresponding to equivalent test conditions can be compared and average/mean derived using
all available DegT50 values is generally not applicable. In all cases, the approach should be
well justif ied and documented and should be supported by the Weight-of-Evidence analysis.
This should include a discussion of outlying results. In particular, the representativeness of the
test conditions should be caref ully assessed f or each test result. Particular scrutiny should be
given if results f rom the tests are close to P or vP threshold.

Aerobic and anaerobic conditions

The f ollowing options are available in the environmental simulation test guidelines:
 Chapter R.11: PBT/vPvB assessment

78 Version 4.0 – December 2023

• OECD TG 307 – Aerobic and Anaerobic Transf ormation in Soil: The test is usually
conducted under aerobic conditions. The test can be perf ormed also under partial or
strict anaerobic conditions.

• OECD TG 308 – Aerobic and Anaerobic Transf ormation in Aquatic Sediment Systems:
The normally employed test includes aerobic and anaerobic sub-compartments. The
test can be perf ormed also under strict anaerobic conditions.

• OECD TG 309 – Aerobic Mineralisation in Surf ace Water – Simulation Biodegradation

Test; There is no “anaerobic” option.

In the anaerobic OECD TG 307 study, the anaerobic conditions can be achieved by covering
the soil with water, i.e. mimicking a f looded f ield, in the absence of oxygen (the soil is purged
with nitrogen and oxygen excluded f or the test duration). A f urther option is a f looded soil but
without the specif ic exclusion of oxygen (paddy f ield simulation). Anaerobic degradation in soil
may also have inf luence on the results in some study cases, f or example, if water covered soil
environments are studied in the f ield. However, f or REACH PBT/vPvB assessments, neither of
the solely anaerobic test conditions are considered to be especially relevant scenarios for the P
assessment in the EU. Nevertheless, if anaerobic soil data are available, they may be used as
part of a Weight-of-Evidence approach in the P assessment.

The OECD TG 309 is an aerobic test. There is no anaerobic option in the test guideline - this
would ef f ectively be stagnant water. The main discussion on simulation tests aerobic and
anaerobic conditions f ocuses on OECD TG 308. The “aerobic” OECD TG 308 is a mixture of
aerobic and anaerobic sediment. The OECD TG states that the “aerobic test simulates an
aerobic water column over an aerobic sediment layer that is underlain with an anaerobic
gradient”. By comparison, the anaerobic test “simulates a completely anaerobic water-
sediment system”.

It is not recommended to judge whether a substance has an environmental half -life exceeding
the P and/or vP thresholds using only anaerobic simulation data. Generally it would be
expected that an anaerobic half-life would be greater than an aerobic half -life where the main
35
route of degradation is aerobic, i.e. if there is no oxygen, degradation will be hindered . Care
should also be taken where the anaerobic data show rapid degradation of a substance. This is
because there is generally no immediate discharge of a substance to anaerobic sediment or
soil. Instead, the substance will usually need to cross an aerobic zone bef ore reaching the
anaerobic zone. This means it is important in persistence assessment to understand the
36
degradation rate across that aerobic zone before the substance reaches the anaerobic zone .

Where anaerobic data are already available, these might be usef ul as part of a Weight-of-
Evidence of whether the P or vP thresholds are met. For example the presence of oxygen may
be less relevant if the primary degradation step is hydrolysis.

Sediment core data might provide some indication of anaerobic degradation capacity. However
some caution should be exercised as the initial starting concentration is rarely known.

35
For example, some widely degradable materials may take considerably longer to degrade under anaerobic
conditions such as newspapers in landfill waste sites.
36
New information on anaerobic degradation may be needed in specific cases to understand the degradation dynamics
in the sediment compartment.
 Chapter R.11: PBT/vPvB assessment

79 Version 4.0 – December 2023

Therefore any derived degradation kinetics estimating a half -lif e will have uncertainty due to
the assumptions required. The history of any local emissions and contamination at the sample
site also provides useful inf ormation to help interpret the data. It is more likely that core data
can be used in an evidence base for anaerobic degradation, as part of a broader Weight-of -
Evidence in the persistence assessment.

When new sediment simulation testing is assessed to be required f or P/vP characterisation,

37 38
transf ormation route prediction or prior knowledge should be used to judge whether
additional inf ormation will be gained f rom perf orming the anaerobic-only test. Exploring an
anaerobic route of degradation may be usef ul in specif ic cases where a
transf ormation/degradation product may be of concern. However, in general a test using
exclusively anaerobic conditions is not required. For the OECD TG 308 sediment simulation test
the “aerobic”’ test will include anaerobic sediment. Theref ore, if a substance is expected to
degrade under anaerobic conditions f aster than under aerobic conditions, an OECD TG 308
may not be the most suitable test to assess the persistence of the substance. Even in the
aerobic version of the OECD TG 308 a large part of the sediment is anaerobic. The substa nces
that degrade only anaerobically may degrade in an OECD TG 308 study but not in an OECD TG
309 study. This has been shown f or example with nitro-containing substances, like musk
xylene. OECD TG 308 might theref ore overestimate the degradation rate in the aerobic
environment f or some substances. If only an OECD TG 308 study is conducted, the conclusion
drawn on persistence may be less conservative compared to the situation where testing is
conducted in other environmental compartments. In such cases, to exclude potential f alse
negative results in relation to the P/vP assessment, strictly aerobic degradation should also be
assessed if technically f easible, i.e. the surf ace water simulation degradation tests with its
strictly aerobic conditions.

Sterile controls in simulation studies

Considerations on the use of sterile controls as part of the degradation tests and ster ilisation
methods can be f ound in Section R.[Link] in Chapter R.7b of the Guidance on IR&CSA and in
ECHA Note on Sterile controls available on ECHA Website.

In simulation tests it is important to dif f erentiate degradation and disappearance of the test
substance due to other dissipation processes. An important advantage of using sterile controls
in degradation studies is that it allows to estimate to what extent abiotic processes (abiotic
degradation and non-degradative dissipation) impact the disappearance of the substance. In
addition, sterile controls can be helpf ul in verif ying the exposure level of the test material in
the test system and f or the determination of a mass balance. Sterile controls are always highly
recommended to be included to the test setup, even if not required by the respective test
guideline.

For the PBT assessment there is normally no need to dif f erentiate the biotic and abiotic
degradation occurring in simulation studies.

37
E.g. with the EAWAG-BBD Pathway Prediction System ([Link]
38
For example consider the application of substance – an anti-oxidant would be expected to be affected by oxygen
and therefore aerobic degradation is likely to be more relevant.
 Chapter R.11: PBT/vPvB assessment

80 Version 4.0 – December 2023

Volatile substances

In simulation tests it is important to dif f erentiate degradation and disappearance of the test
substance due to other dissipation processes. Volatilisation of the test substance makes the
interpretation of the study more dif f icult and increase uncertainty of the persistence
assessment. Pre-test is always recommended to ensure f easibility of the simulation test. Based
on the outcome of that pre-test, necessary modifications to the test design and set-up may be
determined in order to minimise volatilisation or technical inf easibility demonstrated.
Modif ications should not compromise the relevance of the study and theref ore, any
modif ication of the test design or set-up should be consistent with the OECD TGs conditions
and the validity criteria and recommendations of the guidelines should be f ulf illed.
As vapour pressure or Henry´s Law constant do not solely allow predicting the volatilisation
rates in simulation tests, a case-by-case assessment of potential volatilisation taking into
account vapour pressure, HLC, distribution modelling and additional f actors such as water
solubility, phase partitioning and adsorption is recommended. Additionally, experience and
inf ormation f rom other existing studies, e.g. volatility observed in ecotoxicity tests, can be
usef ul. Properties of both the parent substance and potential transf ormation/degradation
products should be considered when designing the study as the transf ormation/degradation
products may be more volatile than the parent substance.
Further inf ormation on options to address volatilisation of test substances in OECD TGs 307,
308 and 309 is available under Section R.[Link].3 and how to consider volatilisation in the
kinetic analyses can be f ound in the Appendix R.11-7 and in ECHA note on Volatile substances
available on ECHA Website. See also Section R.[Link] in Chapter R.7b of the Guidance on
IR&CSA.

Test temperature

Guidance on test temperature f or the simulation test(s) is provided in Section R.[Link] of

Chapter R.7b of the Guidance on IR&CSA. The reference temperature f or providing results on
higher tier tests (and carrying out tests, where relevant) is 12°C f or surf ace water , soil and
f resh or estuarine sediment environments and 9°C f or marine environment.

Non-extractable residues

Non-extractable residues (NER) in soils are def ined as species originating f rom chemicals, that
remain un-extracted by methods which do not signif icantly change the chemical nature of
these residues (Roberts, 1984). With regard to evaluation of water, soil or sediment simulation
degradation test results (according to OECD TGs 307, 308 or 309), it is important f or
persistence assessment to consider the amount and type of NER besides the measured
degradation half -lives of a substance. The amount of NER is driven by substance properties as
well as test conditions. While NER f ormation is expected mainly in the soil or sediment
simulation tests, it cannot be excluded in a surf ace water simulation test. The f ormation of NER
in a surf ace water simulation depends on the Suspended Particulate Matter (SPM)
concentration (including its OC content). The f ormation of NER should not be conf used with the
degradation phenomenon.

Inf ormation on the total amount of NERs (Total NER) is needed f or the mass balance
calculations which, in turn, are essential f or interpretability and reliability of data. To complete
the mass balance as a part of the simulation test with radio-labelled substances ( 14C), the
amount of Total NER must be determined. Quantif ication and f urther characterisation of NERs
 Chapter R.11: PBT/vPvB assessment

81 Version 4.0 – December 2023

is highly recommended to ref ine the persistence assessment, when relevant and f easible. A
tiered extraction scheme f or quantifying Total NER and characterising the dif f erent NER types
is described in Appendix R.11-4. For example, if the substance is concluded to be very
persistent (vP) by taking into account only the extractable parent, f urther ref inement is not
relevant. When the substance is concluded to be persistent (P) or very persistent (vP) based
on the amount of extractable parent and the Total NER, ref inement the assessment by
characterisation of NER is relevant.

In the context of persistence assessment, the Total NER is considered as a non-degraded

parent substance unless f urther characterisation of the Total NER is perf ormed. Determination
and characterisation of the Total NER with the methods described in Appendix R.11-4, rely on
the use of radio-labelled substance. Further characterisation of the Total NER is considered
technically f easible if the Total NER is ≥10 % of the total applied radioactivity (at least on one
of the sampling days) and it should be attempted to be perf ormed f or all sampling points.
When the Total NER is <10 %, the possibility f or f urther characterisation could still be explored
as the technical f easibility of the characterisation might also depend on the test methodology
(spiking radioactivity) and availability of analytical methods.

The NER should ideally be dif f erentiated in remobilisable (Type I) and irreversibly bound
f ractions (Type II and III). While the irreversibly (e.g. covalently bound) and the biogenically
bound part can be assessed as a potential removal pathway, the remobilisable f raction
(strongly sorbed, physically entrapped) pose a potential risk f or the environment.

The f ollowing types of NER in environmental matrices can be considered (Löf f ler et al. 2022,
ECHA, 2019, Kästner et al., 2018):

• NER Type I (strongly sorbed and physically entrapped): strongly sorbed or

physically entrapped into the matrix, contain the parent substance,
transf ormation/degradation products or both. NER Type I have the potential to be
remobilised, and should be considered f or assessment of persistence. If chemical
analyses are conducted, it may be possible to distinguish whether NER Type I consist of
unmodif ied parent substance or of transf ormation/degradation products.

• NER Type II (covalently bound): residues that are covalently bound to the matrix
(e.g. to humic matter) in surf ace water, soils or sediments and that are considered to
have low remobilisation rates. Unless there are indications from the available literature
or monitoring data regarding their potential remobilisation, covalently bound res idues
may be regarded as irreversibly bound. .

• NER Type III (bioNER): incorporated into biomass (biogenic NER, also called bioNER),
NER Type III result f rom the anabolic f ormation of biomolecules (amino acids,
phospholipids, and other biomass compounds) f rom the degradation products of the
parent substance. Dead biomass, and therefore biogenic NER, are eventually f ixed in
organic matter derived from decaying microbial biomass. NER Type III are considered to
be of no concern.

A lack of degradation of the parent substance may be assumed if f ast NER f ormation (with
extensive NER f ormation in several days without any degradation observed) is f ollowed by a
period of relative constant levels of NER. This might indicate the f act that the parent substance
has become non-extractable, and thus is not readily available to degradation.
 Chapter R.11: PBT/vPvB assessment

82 Version 4.0 – December 2023

Inf ormation obtained by comparing results f rom the NER f ormation in sterile and non-sterile
water-sediments/soils can sometimes provide insight into the processes involved in NER
f ormation. If the NER is only f ormed at high levels in non-sterile soils/sediments, this may
indicate degradation of the parent substance and biogenic NER f ormation (bioNER). In this
case, the f ormed NER in the non-sterile soil/sediment is unlikely to consist of the parent
substance. However, when interpreting the results it should be kept in mind that the
sterilisation methods may inf luence the NER f ormation (see Chapter R.7b of the Guidance on
IR&CSA and Kästner et al., 2018). More guidance on the sterilisation methods is provided in
the Section R.[Link]. in Chapter R.7b of the Guidance on IR&CSA and in ECHA Note on Sterile
controls available on ECHA Website.

In the Appendix R.11-4 a detailed description of the test procedures f or the extraction and the
characterisation of the Total NER is provided. Quantif ication and characterisation of the Total
NER f ollows a stepwise approach. For ref ining the DegT50 based on the Total NER
quantif ication and characterisation adequate number of sampling points should be analysed to
allow kinetic analysis. As a f irst step, the Total NER are quantif ied under a tiered extraction
scheme. The Total NER are considered as non-degraded parent substance in degradation half -
lif e derivation unless f urther characterisation of the Total NER is perf ormed. As a second step,
characterisation of the Total NER should be perf ormed when relevant (Total NER ≥10%) and
f easible. The aim of this second step is the quantif ication of the Type I NER in the Total NER.
Type I NER must be considered as non-degraded parent in the degradation half -life derivation.
As a third step, characterisation of the amount of parent substance in the Type I NER can be
perf ormed and subsequently used to derive the degradation half -lif e.

Assessment of relevant transf ormation/degradation products

Where a substance is degraded by abiotic means or partly biodegraded, it may be necessary to

consider whether there are any transf ormation/degradation products f ormed that could be
potential PBTs/vPvBs. Where the original substance f orms a transf ormation/degradation
product that could be PBT/vPvB, there should be an assessment to compare the amount of this
transf ormation/degradation product with the parent substance. In relation to degradation
testing results, including those f rom simulation degradation tests which also include
investigation of degradation pathways (OECD TGs 307, 308 and 309), there are of ten practical
constraints to the analytical identif ication of transf ormation/degradation products.
Biotransformation/ degradation pathways may be complex and many dif f erent degradation
products may be f ormed and some only in small amounts or at slow rates. Practical constraints
in relation to analytical methodologies f or identif ication of degradation products may thus limit
the possibility f or identif ying them chemically, when they occur in very small concentrations. In
the simulation degradation test guidelines f or soil, water -sediment and surf ace water,
transf ormation/degradation products detected at ≥10% of the applied concentration of the
parent substance at any sampling time (principal transf ormation/degradation products) should
at least be identif ied unless reasonably justif ied otherwise. The test guidelines f urthermore
stipulate that transf ormation/degradation products detected at values even lower than 10%
may be may need to be identif ied depending on the specif ic case. However,
transf ormation/degradation products f or which concentrations are continuously increasing or
seem to be stable during the study should also be considered f or identif ication, even if their
concentrations do not exceed the general limit given above, as this may indicate persistence.
The need f or quantif ication and identif ication of transformation/degradation products should be
considered on a case-by-case basis with justif ications. See also the def inition of relevant
transf ormation/degradation products in Section R.11.4.1.
 Chapter R.11: PBT/vPvB assessment

83 Version 4.0 – December 2023

It should be noted that neither a readily biodegradable substance (based on ultimate

degradation) nor its transformation/degradation products will normally need to be assessed
because any transf ormation/degradation products can be assumed to be minimal and
transient. This is the case when based on the Weight-of Evidence there are no indications on
potential persistence.

To assess whether the transf ormation/degradation products may be potential PBT or vPvB
substances, the f ollowing approaches may be helpf ul:

• Based on the structure of the parent molecule, predictions of the structures of the
transf ormation/degradation products may be made. These can be based on QSAR
models/expert systems e.g. the f reely available EAWAG-BBD Pathway Prediction
System (available at: [Link] KEGG
biodegradation database/prediction tool, the OECD QSAR Tool Box (see microbial
metabolism f unctionality) or the commercial CATALOGIC (successor of CATABOL) or
Multicase modelling tools, and by use of expert judgement, supported by appropriate
substance-relevant scientif ic documentation.

For f urther PBT/vPvB assessment of the relevant transf ormation/degradation products

products, the normal PBT/vPvB assessment approach and data generation principles apply, as
described in this Guidance document. See also the def inition of and discussion on relevant
transf ormation/degradation products products in Section R.11.4.1.

Assessment of abiotic degradation data

Abiotic degradation tests are not required in a P assessment f or readily biodegradable
substances, or f or substances shown to be (ultimately) degraded in “enhanced” biodegradation
tests and modif ied ready biodegradability tests, or f or substances with a degradation half -lif e
in a simulation test not f ulf illing the P-criterion. If abiotic degradation tests are available, there
is a need to assess the properties of abiotic degradation products against the screening P , B
and T criteria (see Sections R.7.9.4. and R.7.9.5 in Chapter R.7b of the Guidance on IR&CSA).

It should be noted that the abiotic degradation processes typically concern only primary
degradation. Hence, when assessing such data f or PBT/vPvB characterisation, the identif ication
of the transf ormation product(s) should be perf ormed.

There are several abiotic degradation/transf ormation processes in the environment to be

considered, including e.g. hydrolysis, direct and indirect photodegradation,
oxidation/reduction, surface-controlled catalytic reactions, molecular internal conversions etc.
The most important of these processes is usually hydrolysis, which is relatively independent
f rom the mode of entry of the substance into the environment. Hydrolysis may proceed
ef fectively in aquatic, sediment and soil compartments but it is however noted that there are
substances reaching rapid hydrolysis rates which are well known to be persistent in soil and/or
sediment, e.g. endosulfan and Octamethylcyclotetrasiloxane (D4) (ECHA, 2015) Theref ore,
rapid hydrolysis rates cannot alone lead to concluding that a substance is not persistent. Test
results showing rapid hydrolysis rates always need to be evaluated caref ully in context with
other inf ormation on the substance, such as partitioning and ionogenic properties both of
which may signif icantly inf luence the extent and strength of sorption to soil and sediment.
Hydrolysis also needs to be consistently rapid across the range of environmentally relevant pH.
To provide conf idence in the hydrolysis results, analytical data identif ying
transf ormation/degradation products to provide a mass balance are also needed. These both
 Chapter R.11: PBT/vPvB assessment

84 Version 4.0 – December 2023

demonstrate that primary degradation has occurred, and allow subsequent PBT assessment of
the transf ormation/degradation products.

There is currently no cut of f f or hydrolysis rate, which could alone be used as justif ication to
conclude that a substance is not persistent. Hydrolysis data always need to be considere d in
connection with the other properties, such as partitioning properties and the knowledge on the
abiotic and biotic degradation pathways.

Due to the number of f actors that affect photodegradation rates, this process is not generally
considered in the persistence assessment f or substances registered under REACH.
Nevertheless, recent research has investigated the possibility to derive photodegradation half -
lives taking into account many of those factors (latitude, season, depth, water DOC) (Lin and
Emberger, 2017; Lin et al. 2019, 2022). Further discussion on photodegradation is provided in
Chapter R.7b of the Guidance on IR&CSA.

According to Castro-Jiménez and Van de Meent (2011), light absorption in natural water is
signif icantly slower than measured in laboratory water with photo degradation occurring
around 30 times more slowly f or typical f resh water, 400 times more slowly f or typical coastal
sea water, and 500 times more slowly f or ocean water. These authors also conclude that the
“contribution of photodegradation in water to overall degradation is significant only for
substances that reside in water to a considerable extent”. They highlight that many substances
reside in sediment and soil, rather than in water. They give as an example bromophenyl
ethers, which are “photochemically labile in water” but only slowly photodegrade in the
environment. The relative importance of direct photolysis versus the indirect process varies
and is dependent both on the composition of the substance as the prevailing conditions of the
media. Indirect photodegradation is stimulated in natural environmental waters by the
presence of dissolved organic matter (which is not present in pure laboratory water).

The tests used and their interpretation are discussed in Sections R.7.9.4 and R.7.9.5 of
Chapter R.7b of the Guidance on IR&CSA.

R.[Link].4 Assessment based on estimation models (QSAR, SAR)

The use of QSAR and SAR predictions f or identif ying substances f or persistence (P and vP)
might be used at the screening level, as described below and in detail in Sections R.7.9.4 and
R.7.9.5 of Chapter R.7b of the Guidance on IR&CSA.

Biodegradation QSAR models – screening

Generally, it is recommended to consider both the validation status of any QSAR model and
whether the substance f or which predictions are made may be regarded as being within the
applicability domain of the model (see Section R.6.1 in Chapter R.6 of the Guidance on
IR&CSA).

(Q)SAR estimates may be used f or a preliminary identif ication of substances with a potential
f or persistence. For this purpose, it is recommended to use combined results f rom three
estimation models in the EPI Suite™ (US EPA, 2012, as described above in the Table R.11—4).

Degradation half -lives based on QSAR models using data f rom ready biodegradation tests
should not be used f or comparison with the P/vP criteria.
 Chapter R.11: PBT/vPvB assessment

85 Version 4.0 – December 2023

Other QSAR approaches

Pavan and Worth (2006) describe a number of models and approaches that specifically address
the issue of identif ying structures that meet or do not meet the P criteria.

In the same way, Nendza et al. (2013) provide an inventory of in silico screening tools that
could be used f or the assessment of the degradation potential of substances under the REACH
Regulation. Such estimates may be used f or preliminary identif ication of substances with a
potential f or persistence (see also Section above). The combined results of the three f reely
available estimation models BIOWIN 2, 6 and 3 in the EPI Suite (US EPA, 2012) may be used
as f ollows:

• Non-linear model prediction (BIOWIN 2): does not biodegrade f ast (probability < 0.5)
and ultimate biodegradation timef rame prediction (BIOWIN 3): ≥ months (value <
2.25), or

• MITI non-linear model prediction (BIOWIN 6): does not biodegrade f ast (probability <
0.5) and ultimate biodegradation timeframe prediction (BIOWIN 3): ≥ months (value <
2.25).

QSAR predictions can be used as part of a Weight-of-Evidence approach: predictions that the
substance is not rapidly degradable would support the conclusion that the substance is
potentially P/vP. In the contrary situation, predictions indicating that the substance could
degrade rapidly would support the conclusion that the substance is not persistent. However,
QSAR results alone are in most cases not suf ficient to conclude on non-persistence but should
be supported by additional inf ormation. In every case, it should be verif ied that the QSAR
model and predictions are reliable and applicable to the substance. While the QSAR predictions
using these models are reliable and the estimation results clearly indicate that the substance is
not persistent, all other available inf ormation should still be taken into account together with
QSAR estimation(s) in order to be able to consider the substance as not f ulf illing the criteria for
P. Borderline cases should be caref ully examined, e.g. when the estimate of the ultimate
degradation time predicted by BIOWIN 3 gives a result in the range of 2.25 to 2.75 (see
Sections R.7.9.4 and R.7.9.5 in Chapter R.7b of the Guidance on IR&CSA). Note however that,
in any case, all other existing and reliable QSAR predictions, read across and test data
inf ormation should be considered f or deriving a conclusion regarding the persistence status of
the substance.

The use of QSAR model predictions are of particular relevance and interest when test data are
lacking and when assessing multi-constituent substances f or which it may of ten be dif f icult to
f ind or even to generate test data on relevant individual constituents (including impurities) due
to analytical, technical, practical and cost implications (see Section R.[Link]).

Abiotic degradation models

There are very f ew sof tware models available f or predicting hydrolytic degradation,
atmospheric and hydrolysis or aquatic photodegradation (e.g. AOPWIN and HYDROWIN
modelss are f reely available in Estimations Programs Interface f or Windows (EPI Suite), and a
f ew published models (Peijnenburg et al., 1992, Stegeman et al., 1993). These are reviewed in
Section R.7.9.4 of Chapter R.7b of the Guidance on IR&CSA.
 Chapter R.11: PBT/vPvB assessment

86 Version 4.0 – December 2023

Other modelling data

Another useful source of inf ormation is programmes that predict metabolic pathways f or the
degradation of a substance. These can be useful f or exploring likely routes of degradation as
well as f or helping identif y potential transf ormation/degradation products (both f or analysis
and evaluation). One programme is the EAWAG-BBD Pathway Prediction System (f ormally
f rom the University of Minnesota), which can be f ound at:[Link]
[Link]/predict/[Link].

Multi-media modelling
Results f rom multi-media modelling (e.g. Mackay level III model as this is included in the
EPIWIN QSAR package) could also be explored in order to evaluate the environmental
exposure and compartment(s) of specific concern. Typically, the results used f rom such models
are the relative (%) mass of the substance (in a steady state situation with continuous
environmental release) in each environmental compartment, in a simple “Unit world”
consisting of air, surf ace water, sediment and soil.

Typically, the def ault situation is assumption of an environmental release pattern with equal
release to air, surf ace water and soil (see the def ault settings in the Mackay level III part of
the EPIWIN). It should be noted that the results of such models should be regarded as
qualitative or at most semi-quantitative as they strongly depend on the relative size of the
environmental compartments, the emission pattern (see below) and partitioning and
transf ormation parameters employed in the modelling.

Mackay level III modelling includes environmental inputs and provides inf ormation on the likely
relative importance of intercompartmental transport rates and environmental properties. It is
also dependent on the release pattern (f raction of emission between air, water, soil) and thus
also on the use of the substance. Mackay level III model can be considered suitable to support
the relevance assessment of the compartment(s) f or testing.

If a more relevant /realistic release pattern than equal emission rate to air, water and soil can
be assumed based on knowledge about use of the substance , the Mackay level III model
should be run with an appropriately changed release pattern (f or example, this can easily be
done in the EPIWIN model package). Typically, but depending on the use prof ile of the
substance, it is relevant to run such models assuming the def ault environmental risk
assessment emission pattern, e.g. release to water only. Alternative and f reely available
models exist beside that included in EPIWIN, e.g. EQC (Mackay et al., 1996; see also
[Link]
SIMPLEBOX (Schoorl et al., 2016; see also
[Link]/en/Topics/S/Soil_and_water/SimpleBox).

Another option is to consider comparing the results of the modelling with the normally
employed environmental exposure assessment where emission normally takes place via
39
emission to STP, such as SimpleTreat v4.0 model . SimpleTreat v4.0, which is incorporated
in EUSES v2.2, is an assessment tool f or the f ate and distribution of substances in an STP
where the f ractions at steady state are presented: volatilisation to air, adsorption to STP-
sludge, STP-degradation and the emission f raction to surface water. A process f or volatilisation

39
Available at: [Link] (last accessed: November 2023)
 Chapter R.11: PBT/vPvB assessment

87 Version 4.0 – December 2023

f rom the aeration tank (Mikkelsen, 1995) is incorporated in the current version and elimination
due to volatilisation f rom the sewage system is also estimated. Ionisable organics are present
as neutral and charged species in f raction according to input substance parameters such as
apparent octanol-water partition coef f icient (Dow) and acid dissociation constant (pKa) at
environmental characteristic pH. Such models also typically employ the f ugacity concept. The
f raction sorbed to STP sludge is normally assumed to be disposed of on soil and hence indirect
exposure of the soil compartment has to be assumed.

For some substances which have distinctive use patterns and pulsed releases into the
environment, more specif ic models could be considered, e.g. the FOCUS models f or
agrochemicals. The FOCUS modelling f ramework relies on mechanistic process-based models
to predict the exposure from substances, either directly applied in agricultural areas or driven
by weather-related compartmental transfer processes such as run-of f and drainage. FOCUS
models can thus be used to identif y the relevant compartment(s) to which agrochemicals will
partition, taking into account the specif ic use and release patterns of those substances.

Multimedia f ate modelling, especially the Mackay fugacity level I modelling, can be used in
screening to assess whether volatilisation of a substance is expected in simulation testing. This
model is a steady state calculation with no inf low, outf low, nor intermedia transport.
Degradation reactions are not considered either. This type of model predicts the f ate and
environmental distribution of neutral substances based on calculations considering melting
point (MP), VP, HLC, water solubility (WS), log K ow and Koc. The results give an indication on
where a substance is likely to partition and in which environmental media the concentrations
are likely to be highest (i.e. the f ugacity capacity is largest). Hence, Level I models could be
suitable f or predicting partitioning of substances in a closed system like the closed test vessels
used f or volatile substances in simulation tests. Therefore, results of Level I models could be
used when deciding on the appropriate test system f or simulation testing of volatile substances
or whether some modif ications are needed in the test design to minimise volatilisation (see
R.[Link].3 and in ECHA note on Volatile substances available on ECHA Website. Mackay
f ugacity level 1 modelling can not be used on its own to predict f ate and environmental
distribution. The results of the SimpleTreat model may also give some insight on the possible
volatilisation during simulation testing as it seems to of ten give relatively similar results as the
Level I modelling. However, volatilisation in a simulation test cannot always be excluded even
in those cases where distribution to air is predicted to be low by these models. If the model
overestimates partitioning to the aquatic and/or sediment compartment, partitioning to air will
be underestimated (Guidance on IR&CSA, Chapter R.16).

Finally, f reely available multi-media models f ocussing on the potential f or long-range

environmental (mainly air) transport also exist like the OECD Pov and LRTP Screening Tool
(OECD, 2006) (f or further inf ormation on models predicting long-range transport see Section
III of the draf t document on long-range environmental transport prepared by the POP Review
Committee; UNEP, 2022). They could be employed f or considering possible relevance of
certain environmental compartments of concern f or simulation degradation testing, in
particular whether or not pristine environmental compartments (e.g. open sea) may be
exposed to a signif icant extent.
 Chapter R.11: PBT/vPvB assessment

88 Version 4.0 – December 2023

With respect to the results of the distribution modelling results, they should only be regarded
40
as qualitative or semi-quantitative and a case-by-case evaluation of the results is needed.

R.[Link].5 Field studies for persistence

If f ield studies are available, they are an option to additionally assess the persistence of
substances under realistic outdoor conditions. In contrast to laboratory studies that of ten
include artif icial elements such as drying and sieving of soils (e.g. OECD TG 307 study) it is
possible to study the degradation of a substance under natural conditions in the undisturbed
environment. One of the most important advantages of f ield studies over laboratory studies is
the option to run them over long periods up to several years. There is no risk that the system
gets exhausted as what happens with longer-lasting laboratory studies where the
microbiological activity might signif icantly decrease if the study period needs to be extended to
derive reliable half -lives. With f ield studies, it is also possible to study the accumulation
potential of substances over several years. However, compared to laboratory studies, f ield
studies are semi-controlled with a range of varying environmental f actors. These f actors and
uncertainties derived therein should be taken into account in the assessment.

Reliable f ield studies can be used in a Weight-of-Evidence approach. PBT assessment is

normally not bound to local conditions whereas f ield studies are particularly dependent on local
conditions. Therefore, results f rom f ield studies are not directly comparable with one another,
laboratory tests or P/vP criteria.

When including f ield studies in the Weight-of-Evidence, the varying temperature conditions
should be taken into account (if available). Consideration should be given to whether
temperature correction should be applied. Guidance on test temperature is provided in Section
R.[Link] of Chapter R.7b of the Guidance on IR&CSA.

In general, f ield studies can be carried out f or the dif ferent compartments of interest. For the
soil compartment several guidance documents exist on how to conduct terrestrial f ield
dissipation studies. These guidance documents were mainly developed f or Plant Protection
Products (PPPs) but can also be used f or any other chemical substance. The NAFTA guidance
(2006) is based on the degradation behaviour of substances under realistic exposure
conditions considering all possible dissipation and degradation pathways. The use of a
conceptual model of the substance behaviour that would depend on results f rom laboratory
studies should be supported and the results conf irmed by dif ferent modules of the f ield study.
In addition to the environmental conditions, the mass balance and potential NER f ormation are
to be taken into account to determine in which extent the DegT50 of f ield studies would be
comparable to the half -lif e criteria.

EFSA developed a guidance (EFSA, 2014) f ocused on biodegradation in the soil matrix. It
describes how surface processes such as volatilization and photolysis as well as dissipation by

40
This should include consideration of the values of water solubility, octanol -water and organic-carbon partitioning
coefficients, vapour pressure and half-life coefficients used in the modelling, since these values may be predicted by
the model, even if measured values have been used as input values in the model. A robust study summary should be
provided giving sufficient information on the modelling. (i.e. default assumptions and input parameters of the model).
Finally, consideration on how the substance is likely to be released to the environment should be made. This is
important to understand which fugacity model may be most appropriate – for example 100% release to water, soil etc.
A sense check should also be made to review whether the predictions seem reasonable.
 Chapter R.11: PBT/vPvB assessment

89 Version 4.0 – December 2023

leaching to deeper soil layers are taken into account in order to get a DegT50 value that can
be used in exposure modelling. In order to avoid surf ace processes , it is recommended f or
instance to mix the substance with the topsoil layer of the f ield or to cover the f ield af ter
substance application with a sand layer. For mobile substances that can be leached down to
deeper soil layers during the course of the study, the EFSA guidance requires sampling down
to a depth were no substance can be f ound anymore to account f or all residues.

The OECD Guidance document 232 (OECD, 2016) considers aspects f rom both the NAFTA
(2006) and the EFSA guidance (2014) and is the most recent guidance document. It should be
used f or the conduct of f ield degradation studies.

Lysimeter studies, which are of ten carried out with radiolabelled substances (OECD, 2000a),
can also provide useful inf ormation about the degradation behaviour of a substance to be used
in the context of the P-assessment. Guidance f or deriving DegT50 values f rom lysimeter
studies is provided in FOCUS (2014).

For studying the behaviour of a substance in water or sediment, less guidance is available.
However, meso- or macrocosm studies, which are sometimes used in ecotoxicology, can in
general be used to provide valuable inf ormation on the f ate of the substance, e.g. on the
partition behaviour of the substances. In some cases, if dissipation e.g. due to volatilisation
f rom soil, leaching, surf ace run-off or uptake into plants can be excluded, mesocosm or f ield
studies may be used to derive DegT50 (EFSA, 2014). Guidance on how to derive DegT50
values f rom meso- or macrocosm studies is provided in Deneer et al. (2015).

For f urther ref erences, please see Section R.[Link] in Chapter R.7b of the Guidance on
IR&CSA.

R.[Link].6 Monitoring data

Monitoring data in themselves cannot demonstrate persistence because the presence of a

substance in the environment is dependent on a range of f actors other than degradation rates ,
namely emission and distribution rates. Potential sources, trends of volume, uses and releases
should be considered when evaluating the suitability of monitoring data in the P/vP
assessment. Nevertheless, if monitoring data as a part of a Weight-of-Evidence analysis show
that a substance is present in remote areas (i.e. long distance f rom populated areas and
known point sources, e.g. arctic sea or Alpine lakes), it may be possible to conclude a
substance as P or vP. Monitoring data obtained in areas closer to the sources may also be
usef ul f or P/vP assessment and can be used as one line of evidence f or suppor ting the
conclusions (in both directions: P/vP or not P/vP). Use of monitoring data in P/vP-assessment
encompasses several uncertainties and conclusions should be drawn on the basis of monitoring
data only when there is suf f icient understanding of the substance distribution and transport
behaviour and under the condition that the uncertainties in the monitoring data presented are
adequately addressed. The lack of detection of a substance in monitoring data should be
considered carefully as it does not necessarily mean that a substance is not persistent (e.g.
shortcomings in analytical methods may affect monitoring of substances in the environment).
If monitoring data show that the substance levels in environmental media or biota are rising,
the reasons f or such a time trend should be assessed very carefully against the inf ormation on
the time trends of volumes, uses and releases. Where monitoring data clearly indicate that the
substance f ulfils the vP-criterion or, depending on the case, that the P criterion is f ulf illed in
addition to other supporting inf ormation (and without conf licting data), it may not be
 Chapter R.11: PBT/vPvB assessment

90 Version 4.0 – December 2023

necessary to generate simulation degradation data. In the latter case, conclusions on the
f ulf ilment of the P/vP criteria may be drawn based on the monitoring data, the inf ormation on
the substance distribution/transport behaviour, in addition to other supporting inf ormation
used as part of a Weight-of-Evidence analysis.

R.[Link] Bioaccumulation assessment (B and vB)

This section deals with assessment of bioaccumulation data accepted f or use in the PBT and
vPvB assessment and f urther provides guidance on how to evaluate whether a substance
meets the B or the vB criteria. For this puspose, the section comprises a decision scheme on
how to use data of dif f erent experimental tests as well as non-testing inf ormation. For a B and
vB assessment all available relevant inf ormation should be taken into account. In accordance
with Annex XIII all available inf ormation/evidence on bioaccumulation must be considered in a
Weight-of-Evidence approach. This comprises results f rom bioaccumulation experiments,
monitoring data f rom the f ield and toxicokinetic inf ormation f rom toxicity studies on
accumulation as well as other testing and non-testing indications of bioaccumulation. The order
of data types presented in the below ITS and in the f ollowing subsections are not meant to
def ine the order of importance or weight of individual data types. The data types are presented
so that the experimental data providing inf ormation on bioaccumulation directly comparable
with the B/vB criteria are described f irst and other data relevant f or the assessment as last.

Guidance on the evaluation and validation of both testing data and non-testing inf ormation can
be f ound in Section R.7.10 of Chapter R.7c of the Guidance on IR&CSA.

For substances containing multiple constituents, impurities and/or additives, the guidance
provided below applies to that/those “part(s)” of the substance, which is/are the target of
assessment and testing. The criteria f or selecting an appropriate assessment approach are
provided in Section R.[Link].

41
R.[Link].1 Integrated Assessment and Testing Strategy (ITS)

If a substance is imported or produced in an amount of more than 100 t/y, inf ormation to f ulf il
REACH Annex IX, 9.3.2. standard inf ormation requirement is mandatory. The option of waiving
the bioaccumulation test according to Column 2 of REACH Annex IX can only be taken if the
inf ormation f rom the experimental test is not required f or the conclusion on the PBT/vPvB-
properties (see also Section R.11.3.3). Similarly, the standard aquatic bioaccumulation test
requirement cannot be adapted according to REACH Annex XI, if the PBT/vPvB assessment
shows that a bioaccumulation test in aquatic species is necessary (and it is technically
f easible). However, it is noted that the possibility to use inf ormation ref erred to in REACH
Annex XI should be investigated in the f rame of the PBT/vPvB assessment f irst bef ore
proposing a bioaccumulation test. In that case the evaluation of the B and vB criteria f or the
PBT and vPvB assessment should be performed simultaneously with the assessment of the BCF

41
The mitigating factors that are listed below only refer to the assessment of the B and vB criteria in the context of
the PBT and vPvB assessment. If bioaccumulation appears to be a critical parameter in the risk assessment process, it
could still be necessary to perform a bioaccumulation test, although this may not be needed from the perspective of
the PBT and vPvB assessment.
 Chapter R.11: PBT/vPvB assessment

91 Version 4.0 – December 2023

value. Detailed guidance regarding an ITS f or BCF assessment is presented in Section R.7.10
of Chapter R.7c of the Guidance on IR&CSA.

Figure R.11—4 in this section should be seen as a detailed scheme of the B-assessment block
within the ITS.

If the tonnage produced or imported is below 100 t/y, normally a bioaccumulation test is not
required and theref ore a BCF value may not be available. In that case it should be f irst
considered if the available testing and non-testing data are suf f icient to conclude on the B-
properties for those substances produced or imported at <100 t/y or if bioaccumulation testing
is needed and hence required to draw a reliable conclusion.

A substance meets the B or vB criterion if it is considered bioaccumulative or very

bioaccumulative in one or more of the relevant f ood chains or receptors, e.g. the aquatic
environment, the terrestrial environment or wildlif e or humans. To determine these
classif ications, all reliable and relevant inf ormation on the bioaccumulation potential of a
substance has to be gathered by the registrant and considered in the CSA, including the
PBT/vPvB assessment. If available, such inf ormation might be suf f icient to conclude whether
the substance is vB, B, or not B.

• If the substance has a log K ow lower than 4.5 and no specif ic mechanism of uptake
apart f rom hydrophobic partitioning is known and the possibility f or accumulation in
other f ood chains than the aquatic f ood chain can be ruled out, then the substance can
be considered as not B and not vB. In such a case f urther evaluation of the B and vB
criteria is not necessary. A partitioning process other than lipophilic partitioning could
f or example be the binding to proteins. The possibility of a substance to accumulate in
air-breathing organisms instead of aquatic organisms is indicated by the combination of
a log Koa > 5 with a log Kow >2. A high metabolism rate for the substance could mitigate
such a potential f or bioaccumulation in air-breathing organisms.

• If the substance has very limited potential to be taken up by biota, this might be
indicated by several f actors based on substance properties listed below. These
indicators should be conf irmed by other inf ormation to exclude the possibility of a high
bioaccumulation potential. If such a lack of signif icant uptake is proven, the substance
can be considered as not B and not vB. In such a case, f urther evaluation of the B and
vB criteria is not necessary. It should be noted that the only conclusion drawn based on
this inf ormation is that the substance is not (very) bioaccumulative, and not that the
substance can’t be taken up at all. A substance is unlikely to meet the B criterion (i.e.
unlikely to have a BCF > 2,000) if some or all of the f ollowing indicators are met:

1. an average maximum diameter (D max aver) of greater than 1.7 nm (Further

information is provided in section R.[Link].10)53
2. octanol-water partition coefficient as Log10 (Log K ow) > 10 (calculated
value, preferably by several estimation programs, for substances for
which Log Kow can be calculated and the model is reliable)
3. a measured octanol solubility (mg/L) < 0.002 mmol/L × MW (g/mol)
(without observed toxicity or other indicators of bioaccumulation)

Indicator 1. recommended here as non-testing inf ormation inf luences uptake and
distribution of substances. The log Kow (Indicator 2.) is a general indicator f or uptake,
distribution and excretion whereas the octanol solubility (Indicator 3.) ref lects the
 Chapter R.11: PBT/vPvB assessment

92 Version 4.0 – December 2023

potential f or mass storage, which might f urther prevent uptake in signif icant amounts in
the organism.

The supplementary inf ormation to confirm this limited uptake may comprise data f rom
a chronic toxicity study with mammals (≥ 90 days, showing no toxicity), a toxicokinetic
study with mammals or birds, a bioconcentration study with invertebrates, or reliable
read-across f rom a structurally similar substance (all showing no uptake). These types
of inf ormation should be examined in a Weight-of-Evidence approach together with the
non-testing inf ormation on the substance to conclude whether the B or vB criteria are
met. Evidence of signif icant uptake of a substance in vertebrates af ter prolonged
exposure is a contra-indication to using the above indicators.

• If there is a reliable aqueous bioaccumulation study available, such as an aqueous

exposure OECD TG 305 study, or a bioaccumulation study with Hyalella azteca (OECD
draf t test guideline under revision; OECD, 2023), or a reliable standard bioaccumulation
study with another aquatic invertebrate such as mussels, the result can be directly
related to the criteria f or B and vB. If the BCF is higher than 2000 or 5000 the
substance can be assigned to be B or vB. If a reliable BCF is lower than the B criterion
(BCF < 2000), this is an indication of reduced uptake or metabolism f or hydrophobic
substances with a Log K ow > 4.5. Rapid metabolisation of a substance may lead to a
lower BCF value. In vitro methods according to OECD TG 319A (Determination of in
vitro intrinsic clearance using cryopreserved rainbow trout hepatocytes (RT-HEP)) and
OECD TG 319B (Determination of in vitro intrinsic clearance using rainbow trout liver S9
sub-cellular f raction (RT-S9)) provide inf ormation on biotransformation in the organism,
and thus have potential to support the assessment of bioaccumulation. Further research
in f uture may increase the predictive capacities of in vitro methods. Reduced uptake
and metabolism will most likely also mitigate the bioaccumulation potential in general.
If there are no other indications f or accumulation outside the pelagic f ood chain, such
as elevated concentrations in terrestrial and air-breathing organisms, the substance can
be considered as not B and not vB. Such a conclusion could also be drawn f or
substances having log Kow < 4.5. However, in that case additional consideration should
be given to the possibility of accumulation in f ood chains containing air -breathing
organisms or humans.

• The results of a dietary bioaccumulation study with f ish, such as an OECD TG 305
dietary exposure study, can be used in a similar way to that described a bove to
conclude on the B and vB criterion. The pref erred endpoint f rom the OECD TG 305
dietary exposure test is the BCF value estimated f rom experimentally derived
elimination rate constant, which can be directly compared to the REACH Annex XIII
criteria, unless it can be demonstrated that the uptake rate constant cannot be reliably
estimated with the available methods. In that case other methods (direct application of
k2, or using a correlation of dietary BMF and BCF results to interpolate other BMF
results) as describe in the Guidance document on aspects of OECD TG 305 (OECD,
2017) shall be explored. For very hydrophobic substances, k1 estimates may become
increasingly uncertain.

• In some cases, a conclusion can be drawn f rom additional inf ormation only. This could
be inf ormation f rom f ield studies showing clear accumulation in a f ood chain, or long
half -lif es f rom monitoring studies in humans or wildlif e. If this type of inf ormation is
 Chapter R.11: PBT/vPvB assessment

93 Version 4.0 – December 2023

assessed as reliable, this could be considered as suf f icient inf ormation to draw a
conclusion on the bioaccumulation potential as part of a Weight-of-Evidence approach.

In any other case, no conclusion on the bioaccumulation potential can be drawn and the B and
vB properties should be evaluated in more detail. Based on the above described inf ormation,
this ref ers to the f ollowing cases:
• No direct inf ormation on bioaccumulation (e.g. BCF, BAF or BMF data) are available and
the substance has a Log K ow higher than 4.5, or the partitioning process into aquatic
organisms is not driven by lipophilicity.

• Inf ormation on bioaccumulation is available f or aquatic compartment indicating that

substance is not B, but the screening inf ormation indicates potential bioaccumulation in
air-breathing organisms and no conclusion could be derived f or them based on available
data. In this case new inf ormation may need to be generated on bioaccumulation
potential in air-breathing organisms (mammals), e.g. by appropriate testing or by
generating suitable biomonitoring data, based on a case-by-case assessment of the
needs.

• Direct data on bioaccumulation are available but these data are not reliable and/or
consistent to a degree suf f icient to conclude whether the B or vB criteria are met.
 Chapter R.11: PBT/vPvB assessment

94 Version 4.0 – December 2023

Start B
assessment

Screening Screening criteria

criteria for for air-breathing Conclude
Non-lipophilic
aquatic No organisms: No No
organisms: Log Kow > 2 and
bioaccumulation Not B1
Log Kow > 4.5 log K oa > 5

Yes
Yes Yes

- physicochemical properties
- read-across with structurally similar substance Physicochemical indicators for hindered uptake:
- in vitro data on metabolism in combination with kinetics due to molecular size
of uptake and depuration • Dmax aver > 17.4 Å OR
- uptake and absorption efficiency uptake and distribution in general
- absence or presence of chronic toxicity • Log KOW > 10 OR
- terrestrial or benthic accumulation studies for low potential mass storage
- aquatic bioconcentration or bioaccumulation studies Octanol solubility [mg/L ] <0.002 [mM] x MW
(fish BCF, fish dietary test results, Hyalella azteca BCF, [g/mol]
mussels BCF)
- toxicokinetic information on (laboratory) mam mals, AND :
humans, aquatic organisms, birds; results of assessment Experimental indicators for hindrance of uptake
of toxicokinetic behaviour • No chronic toxicity for m ammals and birds
- detection of elevated levels in biota • No uptake in mammalian toxicokinetic study
- field data concerning biomagnification and • Very low uptake after chronic exposure
bioaccumulation

Yes No

The substance is potentially B/vB.

Conclude the B-assessment More data should be generated if the substance fulfils the P or vP
(B, vB or not B). criteria.
If it can be concluded that the substance What type of further data are needed, depends on the screening criteria and
is not B, no further information for a other available information.
definitive P assessment needs to be Consider whether other type of data, e.g. aquatic invertebrate tests, provide
generated.
sufficient information, before conducting an in vivo fish BCF test.

Any compartment/taxa of specific

concern based on available Generate the further
information? (highest likelihood of information based on
Yes
bioaccumulation in a specific case-by-case
compartment/taxa or non-lipid assessment
binding)

No

OECD TG 305, (aqueous exposure test preferred, if

feasible) or invertebrate studies (e.g. HYBIT) following
standard TGs or OECD TG 417 (air breathing organisms).

1
This conclusion can not be drawn if there is other relevant information available, e.g. as listed in the subsequent text box, that
does not support the not B conclusion.

Figure R.11—4: Integrated assessment and testing strategy for B-assessment.

 Chapter R.11: PBT/vPvB assessment

95 Version 4.0 – December 2023

Integrated assessment and testing of Bioaccumulation - Explanatory Notes to Figure

R.11-4.
Bioaccumulation is most of ten driven by partitioning to storage lipid. In these cases, a log
Kow greater than 4.5 is used as screening criterion for aquatic organisms, and a log Kow greater
than 2 together with a log KOA greater than 5 as screening criteria f or air-breathing organisms.
If log Kow is less than 2, the substance can normally be regarded as not f ulf illing the B/vB
criteria. If the substance has a log Kow between 2 and 4.5, but log Koa is below 5, then it can be
expected that the substance is neither hydrophobic enough to bioaccumulate in aquatic
species, nor that it is bioaccumulating in air-breathing species, because it can be eliminated
rapidly enough by exhalation. Because of the importance of log Kow as a screening criterion, its
uncertainty needs to be taken into account, especially when close to the threshold. Guidance
on the derivation of log Kow is given in chapter R.[Link].10 and in the Appendix R.11-5

For some groups of substances, such as organometals, ionisable substances and surface
active substances, log Kow is not a valid descriptor f or assessing the bioaccumulation
potential (Armitage et al., 2017, Hodges et al., 2019). Inf ormation on bioaccumulation of such
substances should theref ore take account of other descriptors or mechanisms than
hydrophobicity. Guidance on consideration f or bioaccumulation assessment of ionisable and
surf ace active substances is given in Appendix R.7.10 3 of Guidance on IR&CSA, Chapter R.7c.

Furthermore, specific binding to proteins instead of lipids might result in an erroneously low
BCF value if this value is estimated f rom log K ow . Per- and polyf luoroalkyl substances (PFASs)
are examples of such partitioning behaviour, of which perfluorooctane sulphonic acid (PFOS) is
a well-known example (Kelly et al., 2009). Guidance on consideration f or bioaccumulation
assessment of organic substances that do not partition to lipid is given in Appendix R.7.10 3 of
Guidance on IR&CSA, Chapter R.7c.

When assessing the bioaccumulation potential, all available relevant inf ormation has to be
considered in a Weight-of-Evidence approach. Inf ormation to be considered includes:

- physicochemical properties;

- read-across with structurally similar substance (Further data R.[Link].10);

- in vitro data on metabolism in combination with kinetics of uptake and depuration

(R.[Link].4);

- uptake and absorption ef f iciency (R.[Link].3);

- absence or presence of chronic toxicity (R.[Link].9);

- terrestrial or benthic accumulation studies (R.[Link].5);

- aquatic bioconcentration or bioaccumulation studies, such as f ish bioaccumulation studies

with aqueous or dietary exposure, Hyalella azteca bioconcentration test (OECD, 2023), mussel
bioconcentration test (ASTM, 2003)) (R.[Link].2, R.[Link].3);

- toxicokinetic inf ormation f rom dietary exposure Bioaccumulation Fish test (R.[Link].3) and
toxicokinetic inf ormation f rom studies with mammals (R.[Link].8);

- detection of elevated levels in biota and f ield data concerning biomagnif ication and
bioaccumulation (R.[Link].6);
 Chapter R.11: PBT/vPvB assessment

96 Version 4.0 – December 2023

- indication of hindered uptake by physicochemical parameters, together with observed

absence of chronic toxicity in experimental studies or very low uptake after chronic exposure in
mammals and birds; physicochemical parameters comprise molecular size, logKow and octanol
solubility; molecular weight as parameter to indicate hindered uptake is no longer supported
(R.[Link].10 and Appendix R.11-1.1).

The Bioaccumulation Assessment Tool (BAT), accompanied by guiding principles in the BAT
manual (Armitage et al., 2021), promotes standardised recording and evaluation of various
lines of evidence related to the endpoint bioaccumulation. When integrating and weighing
inf ormation, reliable evidence of bioaccumulation cannot be outweighed by inf ormation
showing no bioaccumulation. (see also Chapters R.4 of the Guidance on IR&CSA and Section
R.11.4.1). See also ECHA Weight-of-Evidence templates13

If the Weight-of-Evidence approach described under "Conclusions on the Endpoint" is not

suf ficient to draw a conclusion, the perf ormance of an experimental bioaccumulation test or
generation of other appropriate bioaccumulation inf ormation is required. However, before such
a study is conducted f or assessing the B and vB criteria, the P criterion should be investigated
in order to prevent unnecessary testing of animals. Further generation of inf ormation on
bioaccumulation is only necessary, if the P criterion has been conf irmed to be f ulf illed f or the
substance.

If generation of f urther bioaccumulation data is necessary, there are several options f or the
most appropriate strategy. Additional data should always be generated in a tiered way
revisiting the B-assessment af ter each time new data are made available. In normal case it
may be possible to conclude on the B/vB properties af ter one study, but in specif ic cases
several bioaccumulation studies may be needed.

The available data def ine the choice of the study/test. Hereby, the understanding of in which
type of species/compartment the bioaccumulation potential seems highest is crucial f or the
choice of the test. In very specific cases, the most relevant compartment(s) of exposure may
also inf luence the choice of the study.

R.[Link].2 Experimental aquatic bioconcentration factor (BCF) data

It should be noted that the greatest weight under PBT assessment f or REACH is placed on a
valid aquatic BCF test due to the current understanding that BCF is the most representative
way of reflecting the bioaccumulation potential of a substance, where aquatic bioaccumulation
is relevant. If BCF-values are inconsistent with other data types, it is very important to address
the reasons f or such inconsistency and discuss caref ully about the plausibility of the BCF -
values in this context. If a substance has a valid and plausible aquatic BCF > 2000 or 5000
(indicating a signif icant accumulation in the test organism), the substance is def ined as B or vB
regardless of whether f urther indications f or biomagnif ication or trophic magnif ication exist.

Aqueous Exposure Bioaccumulation Fish Test

In most cases where experimental inf ormation on bioaccumulation in aquatic species is
needed, a f low-through aqueous exposure bioconcentration f ish test according to OECD TG
305-I (or OECD TG 305-II) gives reliable results f or comparison with the B/vB criteria.
Whenever possible, the f reshwater amphipod Hyalella azteca bioconcentration test (HYBIT)
(OECD draf t TG under revision; OECD, 2023) should be considered to avoid vertebrate testing
(See section Aquatic invertebrate tests: Hyalella azteca bioconcentration test (HYBIT) , page
 Chapter R.11: PBT/vPvB assessment

97 Version 4.0 – December 2023

100). Only in specif ic cases, described in f ollowing subsections, other study/test types may be
warranted as the option f or generating f urther inf ormation.

In line with Annex 1 of the OECD TG 305, the f ollowing def initions are used in this guidance:

• The bioconcentration f actor (BCF) at any time during the uptake phase of this
accumulation test is the concentration of test substance in/on the f ish or specif ied
tissues thereof (Cf as mg/kg) divided by the concentration of the substance in the
surrounding medium (Cw as mg/L). BCF is expressed in L·kg-1. Please note that
corrections f or growth and/or a standard lipid content are not accounted f or in this
def inition of the BCF.

• The steady-state bioconcentration f actor (BCF SS) does not change signif icantly over a
prolonged period of time, the concentration of the test substance in the surrounding
medium being constant during this period.

• The kinetic bioconcentration f actor (BCFK ) is the ratio of the uptake rate constant, k 1, to
the depuration rate constant, k 2 (i.e. k1/k2 – see corresponding definitions in Annex 1 of
the OECD TG 305). In principle the value should be comparable to the BCF SS (see
def inition above), but deviations may occur if steady-state was uncertain or if
corrections f or growth have been applied to the kinetic BCF.

• The lipid normalised kinetic bioconcentration f actor (BCFKL) is normalised to a f ish with
a 5% lipid content.

• The lipid normalised, growth corrected kinetic bioconcentration f actor (BCF KgL) is a
kinetic BCF which is corrected f or fish growth observed during the study period and is
subsequently normalised to a f ish with a 5% lipid content as described in Annex 5 of
the OECD TG 305 (see also Appendix R.11-6).

Bioconcentration data f rom controlled laboratory experiments can be used in assessing the
bioaccumulation potential of a substance. For example, OECD TG 305-I: Aqueous Exposure
Bioconcentration Fish Test (OECD, 2012a) or an equivalent test protocol f or f ish can be
f ollowed f or producing experimental bioconcentration data. Valid BCF results f rom the
Bioconcentration Fish Test can be used directly f or comparison with the B and vB criteria.
Nevertheless, it is underlined, that in addition to BCF values, other relevant inf ormation should
be considered. The REACH Annex XIII Introduction requires all other available bioaccumulation
data to be taken into account in an integrated manner and applying a Weight-of-Evidence
approach using expert judgement to derive the conclusion. If BCFs seem not coherent with
other data or there are very dif f erent BCF-values available, it is important to address the
reasons f or inconsistency and discuss in which way this inconsistency impacts the overall
conclusions on bioaccumulation potential.

The use of other taxonomic groups than f ish (e.g. Hyalella azteca, OECD, 2023; mussel
bioconcentration test, ASTM, 2003) is possible f or measuring bioconcentration in the aquatic
environment. Valid BCFs determined in tests with other taxonomic groups can be used in
assessing whether or not the B/vB criteria are met. Furthermore, in case a Log Kow as
screening inf ormation is considered likely to be reliable f or estimating the bioaccumulation
potential of a substance while still some experimental inf ormation is needed to ref ute or
conf irm this assumption, the OECD TG 305-II: Minimised Aqueous Exposure Fish Test may also
be used to assess B or vB, provided that the f inal results will most likely not result in
borderline cases of meeting either the B or vB criterion. This should be investigated before the
 Chapter R.11: PBT/vPvB assessment

98 Version 4.0 – December 2023

test is initiated, e.g. by the use of QSARs, to avoid the results of the test being insuf f icient f or
the B assessment after the test has been completed. Conditions f or selecting the minimised
OECD TG 305-II instead of the OECD TG 305-I are described in the OECD TG 305 and it should
be noted that the OECD TG 305-II test conducted within those conditions can be used f or the
bioaccumulation assessment to minimise animal use. Whether minimised tests should be
carried out depends on a range of f actors including the required level of precision of the
determination of the BCF value f or a particular substance. For instance, if it is estimated that
the BCF-value may be close to the threshold values of either 2000 f or 'B' or 5000 f or 'vB', the
BCF determination by OECD TG 305-II is not warranted because the result may be associated
with too much uncertainty. In such a case an OECD TG 305-I test would be appropriate.

Bioconcentration can be tested experimentally f or substances that are water soluble to an

extent allowing that the exposure concentration(s) can be maintained constant throughout the
uptake phase of the test, as demonstrated by regular analytical verif ication of the exposure
concentrations. A proper analytical method should be available to measure the test substance
concentration in water at the used test concentrations that should always be below the water
solubility limit of the substance (taking into account the ef fect that the test media composition
may have an ef fective solubility). In addition, an analytical method is required to measure the
substance accumulated in the animal tissues. In bioconcentration tests, the water phase must
be the only route of exposure and exposure via f eed must be avoided.

The aim of the bioconcentration testing is to produce a reliable estimate of how much
substance could concentrate f rom the aquatic compartment (Cw ) to f ish (Cf) so that a
bioconcentration f actor (BCF SS) can be calculated by using ratio C f/Cw at steady-state.
However, a BCF k value is pref erred, which is calculated as the ratio of the uptake rate constant
(k1) and the depuration rate constant (k 2). This approach is especially usef ul in those cases in
which steady-state is not reached during the uptake phase, as BCF k in these cases can still be
estimated. If uptake f ollows first order kinetics and steady-state was reached, calculation of
BCF k and BCF ss should in principle lead to the same result. However, f or bioaccumulative
substances a real steady-state is of ten not attained during the uptake phase, and the
conclusion of steady-state f rom the concentrations in f ish at three consecutive time points
could be erroneous. If the BCF k based on f irst order kinetics is signif icantly dif f erent f rom the
BCF SS, this is a clear indication that steady-state has not been attained in the uptake phase.

Besides that, the BCF ss cannot be corrected f or the growth of f ish as no agreed method is
available to correct BCF SS f or growth. The increase in f ish mass during the test will result in a
decrease of the test substance concentration in the growing f ish (= growth dilution) and thus
the BCF may be underestimated if no correction is made. To avoid complications of
interpretation due to growth dilution, the f eeding rate should be selected such that f ast growth
and large increase of lipid content are avoided. If possible, avoid testing f ish species during a
(juvenile) lif e-stage with rapid growth (OECD TG 305). Although there has been some criticism
on the growth correction recently (Gobas and Lee, 2019), this is considered insuf f icient
justif ication to refrain f rom growth correction. Therefore, growth correction should be applied
in the B assessment. Further explanations on correction f or growth dilution are given in
Appendix R.11-6. To avoid uncertainty caused by growth correction, non-growing adult f ish are
pref erred f or testing.

Growth dilution may af fect both BCF SS and BCF K and theref ore the BCF K should be calculated
and corrected f or growth dilution, BCF kg, if f ish growth is signif icant during the test (this is
especially important f or f ast growing juvenile f ish, such as juvenile rainbow trout, bluegill
 Chapter R.11: PBT/vPvB assessment

99 Version 4.0 – December 2023

sunf ish and carp). OECD TG 305 (Annex 5) contains two dif f erent methods f or growth dilution
correction. For bioaccumulative substances the kinetics of bioaccumulation are slow and
growth dilution may have a major impact on the BCF. In conclusion, BCF Kg is pref erred f or PBT
substances due to i) the slow kinetics possibly leading to non-equilibrium within the timef rame
of the experimental bioaccumulation test, and especially ii) the correction f or growth dilution,
which is not included in the BCF SS. More emphasis on BCF Kg is also given in OECD TG 305.

For older f ish bioaccumulation studies, inf ormation on growth may not be available. In this
case, an assessment of the likely signif icance of growth on the results should be made to
determine what weight should be given to the study in the Weight-of-Evidence assessment. As
noted in the OECD TG 305 (paragraph 32), juvenile f ish may be f ast growing at the lif e-stage
(and size) they are tested in the OECD TG 305. Small rainbow trout (O. mykiss) are an
example of this. In contrast, f ish such as zebrafish (D. rerio) are usually adults and theref ore
signif icantly slower growing (for example see an analysis in Brooke and Crookes, 2012). In the
absence of growth data, the uncertainty in a BCF value derived f rom a f ast-growing f ish will be
greater than a slow growing f ish, which is important f or results near a regulatory threshold.
Overall, any approach to using f ish bioaccumulation data where growth data are not available
needs to be considered on a case-by-case basis with justif ication f or the conclusion drawn. It
should be noted that apart f rom growth dilution, several other f actors have been suggested to
potentially inf luence test results, f or example water -to-f ish-ratios, temperature, sex
dif f erences, f eeding procedure and slight variances in water chemistry and dis solved oxygen
concentrations (Wassenaar et al., 2019). Most of these, and other variables can inf luence the
metabolic capacity of the test animals and/or are directly related to changes in activity or
oxygen consumption.

The pref erred way to derive k 1 and k2 is, in most cases, to f it both parameters simultaneously
by non-linear regression to the data f or both, the uptake phase and the depuration phase (see
Annex 5 of the OECD TG 305 (OECD, 2012a), and Guidance document on aspects of OECD TG
305 (OECD, 2017)). This procedure represents the best way to f it both parameters to all
available data and yields a consistent f it f or the uptake and depuration phase. Another way to
derive k 1 and k2 is to use a sequential f itting procedure and f ind values of k 1 and k2
independently. This may sometimes lead to a gap in the f it between the uptake and depuration
phase. However, a benef it of sequential f itting is that k 2 is f itted f irst, and is theref ore
unaf f ected by the uptake phase, which might be relevant if kinetics of both phases are
signif icantly dif ferent due to the increased f ish size during the test duration. The depuration
rate constant, k 2, is the parameter of most interest in a bioaccumulation test given that the
uncertainties in its derivation are understood and can be addressed. As recommended in
Annex 5 of OECD TG 305, visual inspection of the modelled uptake and depuration curves
when plotted against the measured sample data can be used to assess and compare the
goodness of f it of both methods. This is a reporting requirement of OECD TG 305.

The data could be transformed by taking the natural logarithms, if this transformation reduces
the variation in the replicates and/or leads to a better f it of the data. However, care must be
taken as such a transf ormation could give too much weight to very low concentrations
observed at the end of the depuration phase, leading to a worse f it towards the end of the
uptake phase and beginning of the depuration phase. If f ish concentrations are log normal-
transf ormed, a geometric mean f or the water concentration should be used instead of an
 Chapter R.11: PBT/vPvB assessment

100 Version 4.0 – December 2023

42
arithmetic mean. The f itting routine bcmfR R-package (OECD-TG305 R-Package) , provided by
the OECD to support the OECD TG 305, allows also box-cox transf ormation next to
untransformed and ln-transformed data. This box-cox transformation can be considered as an
intermediate between the untransformed and ln-transformed data. Next to the f itting itself , it
also determines which transf ormation f its the best f rom a statistical point of view.

Normally, the concentration of the test substance in f ish tissues should be lipid normalised.
Suitable methods should be used f or determination of lipid content (Schlechtriem et al. 2012).
A lipid normalisation to a f ish with a 5% lipid content as recommended in OECD TG 305 should
be perf ormed unless it is evident that the substance does not primarily accumulate in lipid
tissues. The resulting BCF that is pref erred f or a comparison with the bioaccumulation criteria
is the kinetic growth corrected and subsequently lipid normalised (to 5% lipids) BCF value
(BCF KgL). A justif ication is needed in case no normalisation is carried out.

Aquatic invertebrate tests: Hyalella azteca bioconcentration test (HYBIT)

Bioaccumulation of aquatic species, pref erably f ish, is a standard inf ormation requirement
under REACH Annex IX. The criteria f or (very) bioaccumulative substances under REACH
Annex XIII are based on bioconcentration f actors in aquatic species. Fish as well as other
aquatic species, including aquatic invertebrates, may be used f or bioaccumulation assessment
(Section R.[Link] of Chapter R.7c of the Guidance on IR&CSA). Inf ormation on
bioaccumulation in aquatic invertebrates has been used as part of the weight-of -evidence in
43
B/vB identif ication .

Hyalella azteca is an epibenthic amphipod which is widespread in North and Central America
and commonly used f or ecotoxicity studies (Environment Canada 2013; US EPA 2000a; ASTM
International 2020). The f reshwater amphipods can be easily cultured in the laboratory and
are available during the entire year. Due to their high reproduction rate a nd f ast growth,
experimental organisms can be raised within a f ew weeks to adult size to meet the need f or a
high number of large organisms required f or bioaccumulation testing (Schlechtriem et al.
2019).

A draf t OECD TG f or the Hyalella azteca bioconcentration test (HYBIT) is under revision (OECD
draf t TG under revision; OECD, 2023). Current regulation on bioaccumulation f ocuses on the
bioconcentration f actor (BCF) f or f ish. The draf t OECD TG provides a non-vertebrate test to
derive BCF values, which allow f or the assessment of the bioconcentration potential of
substances. BCF values f or lipophilic substances determined with H. azteca show a strong
correlation with BCFs that have been determined according to the OECD TG 305 when BCF
values are normalised with a def ault 5 % lipid content (based on whole body wet weight),
which is equivalent to the lipid content commonly used for the normalisation of f ish BCF values
(Schlechtriem et al. 2019). H. azteca can reach total lipid concentrations of up to 20 % on a
dry weight basis (Cavaletto and Gardner, 1999) which is, assuming a dry to wet weight ratio of

42
Accessible at [Link]
[Link] (last accessed: October 2022)
43
MCCPs [Link] Methoxychlor
[Link] Chlorpyrifos
[Link]
 Chapter R.11: PBT/vPvB assessment

101 Version 4.0 – December 2023

0.25 (Othman and Pascoe, 2001), equivalent to the def ault 5 % lipid content (w/w). However,
the lipid content of lab-grown and f ield-caught H. azteca is usually lower than 5%, mostly
ranging f rom about 1-3 % (w/w) (Schlechtriem et al. 2019, Kosf eld et al. 2020; Arts et al.
1995; Huf f Hartz et al. 2021). A normalisation of H. azteca BCF values to a def ault value of 3
% lipid content (w/w) typical f or H. azteca is thus recommended.

H. azteca has the ability to metabolise substances, but the biotransf ormation reactions can be
dif f erent f rom f ish species (Fu et al, 2021; Kosf eld et al., 2020). Comparison between the
metabolic rate of H. azteca with f ish in vitro has shown that f ish tend to have higher metabolic
activity (Kosf eld et al., 2020). Since metabolism rates inf luence the BCF, this may explain why
H. azteca tends to have higher BCFs than f ish when normalised to a def ault 5 % lipid content
(Schlechtriem et al. 2019).

Bioconcentration f actors (BCFs) f rom H. azteca bioconcentration tests can be compared against
the REACH Annex XIII criteria on B and vB properties. Consequently, it could become an
alternative test to the bioaccumulation test with f ish (OECD TG 305). This could help to avoid
unnecessary vertebrate testing using f ish.

The f ollowing def initions are used in the OECD draf t TG on HYBIT (OECD draf t TG under
revision; OECD, 2023).

• BCF: The bioconcentration f actor at any time during the uptake phase of this
accumulation test is the concentration of test substance in/on H. azteca (CH as mg/kg)
divided by the concentration of the chemical in the surrounding medium (Cw as mg/L).
BCF is expressed in L·kg-1. Please note that correction f or a standard lipid content are
not accounted f or.

• BCF K: The kinetic bioconcentration f actor is the ratio of the uptake rate constant, k1, to
the depuration rate constant, k 2 (i.e., k1/k2 – cf. respective definitions). In principle the
value should be comparable to the BCF SS (cf. respective definition), but deviations may
occur if steady-state was uncertain.

• BCF KL: The lipid normalised kinetic bioconcentration f actor is normalised to H. azteca
tissue with a 3 % lipid content (w/w).

• BCF SS: The steady-state bioconcentration f actor does not change signif icantly over a
prolonged period of time, the concentration of the test substance in the surrounding
medium being constant during this period of time (cf. Def inition of steady-state).

• BCF SSL: The lipid normalised steady-state bioconcentration f actor is normalised to H.

azteca tissue with a 3 % lipid content (w/w).

• Bioconcentration: Bioconcentration is the increase in concentration of the test

substance in or on an organism (or specif ied tissues thereof ) relative to the
concentration of test substance in the surrounding medium.

• Depuration: The depuration or post-exposure (loss) phase is the time, f ollowing the
transf er of the test organism f rom a medium containing test substance to a medium
f ree of that substance, during which the depuration (or the net loss) of the substance
f rom the test organism (or specif ied tissue thereof ) is studied.
 Chapter R.11: PBT/vPvB assessment

102 Version 4.0 – December 2023

• k1: The uptake rate constant is the numerical value def ining the rate of increase in the
concentration of test substance in/on test organism (or specified tissues thereof ) when
44
exposed to that chemical (k 1 is expressed in L kg-1 day-1) .

• k2: The depuration (loss) rate constant is the numerical value def ining the rate of
reduction in the concentration of the test substance in the test organism (or specif ied
tissues thereof) f ollowing the transfer of the test organism f rom a medium containing
the test substance to a medium f ree of that substance (k 2 is expressed in day -1).

The small size of H. azteca enables reduction of the size of the test system and can lead to
substantial savings of test substance as steady state is reached faster and the test duration is
shorter than in f ish. Especially f or test substances with high log Kow , this can be an advantage.
H. azteca have a larger surface/volume ratio compared with that f or larger organisms such as
f ish. This can theoretically lead to higher estimates of bioconcentration due to adsorption of
chemicals to their body surf ace. However, an apparent deviation f rom f irst order kinetics as a
result of potential adsorption processes has not been observed f or hydrophobic organic
substances. Apart f rom the established f low-through regime commonly applied in
bioconcentration studies, semi-static regimes can be used in studies carried out according to
the OECD draf t TG f or HYBIT (OECD draft TG under revision; OECD, 2023). Both regimes have
been validated as part of an international ring trial. For inf ormation on the test design, f urther
ref erence can be made to Kosf eld et al. (2020) and Schlechtriem et al. (2019). An appropriate
analytical method of known accuracy, precision and sensitivity should be available f or the
quantif ication of the substance in the test solutions and in biological material. The test is
applicable to neutral and ionised organic substances. However, there is no inf ormation on its
applicability to surf ace-active substances. Testing of very hydrophobic substances via aqueous
exposure may be dif f icult, e.g. due to challenges in establishing stable exposure
concentrations. A preliminary experiment should be conducted to optimise the test conditions
of the def initive test, such as selection of test substance concentration(s), duration of the
uptake and depuration phases.

As f or f ish, the appropriate duration of the uptake phase is dependent on the hydrophobicity of
the test substance. Schlechtriem et al. (2019) recommended that f or substances having log
KOW <4, an exposure period of 2 days is suf ficient for the uptake period. For substances with
log KOW of 4-6, exposure of at least 4 days is needed to ensure that steady-state conditions are
reached. For substances with log KOW >6, exposure periods of >12 days are required and only
the BCF K rather than BCF SS should be calculated.

H. azteca BCF results converted to 3 % lipid (BCF KL, 3 %) are pref erred f or a comparison
against the REACH Annex XIII criteria on B and vB properties, and deviations should be
justif ied. If a substance has a valid and plausible H. azteca BCF (3%, w/w) >2000 or >5000
(indicating a signif icant accumulation in the test organism), the substance is def ined as ‘B’ or
‘vB’, respectively. A H. azteca BCF (3%, w/w) <1200 and <3000 indicates ‘not B’ and ‘not vB’
f or the aquatic compartment, because even with a lipid normalisation to 5 % (w/w) as applied
f or f ish, the threshold values of 2000 and 5000 f or ‘B’ and ‘vB’ would not be passed,

44
To note that k1 is the initial rate of uptake, as during the uptake phase the concentration increase already levels off
due to elimination.
 Chapter R.11: PBT/vPvB assessment

103 Version 4.0 – December 2023

respectively. A ‘not B’ and ‘not vB’ conclusion f or the aquatic compartment can only be drawn
if there is no other relevant and reliable inf ormation indicating the contrary.

For lipid normalised H. azteca BCF values between 1200 and 2000 and between 3000 and
5000, it cannot be excluded that due to the lower lipid content of the amphipods (3 %, w/w)
the bioaccumulation potential of a substance may be underestimated compared to f ish (5 %,
w/w). In such a situation, f urther investigations are required to allow a clear “B” and “vB”
conclusion. Further inf ormation, such as an in vitro test according to OECD TG 319 could help
to clarif y the bioaccumulation potential of the test substances. Estimated reliable BCF values <
2000 and < 5000 obtained f rom the OECD TG 319 could be used to conf irm the ‘not B’ and
‘not vB’ conclusions f or the aquatic based on the H. azteca BCF, respectively. If no conclusion
can be made with the available data, and uncertainties remain about the bioaccumulation
potential on aquatic organisms, f urther testing with f ish should be considered as a last resort,
if justif ied.

The use of H. azteca to assess bioaccumulation is based on current knowledge and experience.
Registrants are advised to f ollow recent and f uture developments in the f ield, e.g. via the PBT
expert group website.

R.[Link].3 Dietary Exposure Bioaccumulation Fish Test

A dietary exposure test, preferably OECD TG 305-III: Dietary Exposure Bioaccumulation Fish
Test, should be considered f or substances f or which it is not possible to maintain and measure
aqueous concentrations reliably. For strongly hydrophobic substances (log Kow > 5 and a water
solubility below ~ 0.01-0.1 mg/L), testing via aqueous exposure may become increasingly
dif f icult. A high Koc usually indicates a strong dissipation of substances f rom water to organic
matter. However, an aqueous exposure test is still pref erred f or substances that have a high
log Kow as long as they have an appreciable water solubility with respect to the sensitivity of
available analytical techniques, and the maintenance of the aqueous concentration as well as
the analysis of these concentrations do not pose any constraints.

The use of solvents and dispersants (solubilising agents) may help to produce suitably
concentrated stock solutions, but is not generally recommended according to OECD TG 305
and every effort should be made to minimise the use of such materials. The use of technical
devices such as solid-phase desorption and passive dosing systems (Schlechtriem et al. 2017;
Wang et al. 2022) should be considered, as they are suitable to provide stable aqueous
concentrations of strongly hydrophobic substances and allow the testing of such substances in
f ish BCF studies, according to OECD 305, without using solubilising agents. A preliminary
experiment to optimise the test conditions of the def initive test should be conducted. Improved
analytical techniques or the use of a radiolabelled substance should be considered f irst to
improve the detection limit of strongly hydrophobic substances in water bef ore deciding on
whether a dietary test is indeed the only f easible option.

Nevertheless, if the expected f ish concentration (body burden) f ollowing aqueous exposure is
expected to be below the detection limit within 60 days, the dietary test may provide an option
to achieve body burdens that exceed the detection limits f or the substance. The endpoint f or a
dietary study is a dietary biomagnif ication f actor (dietary BMF), which is the concentration of a
substance in predator (i.e. f ish) relative to the concentration in the prey (i.e. f ood) at steady
state. The dietary test also provides valuable toxicokinetics data including the dietary chemical
absorption ef f iciency, the whole body elimination rate constant (k 2) and the half -lif e f or
 Chapter R.11: PBT/vPvB assessment

104 Version 4.0 – December 2023

substances f or which obtaining aquatic BCF data is technically not f easible. From the
elimination rate constant, an aquatic BCF can be estimated using the Dietary Exposure Test
45
Spreadsheet of OECD 305 TG . Several methods are described in Annex 8 of OECD TG 305
(OECD, 2012a) and the Guidance Document on Aspects of OECD TG 305 (OECD 2017).

The f ollowing def initions are used in this guidance:

• The dietary biomagnif ication f actor (dietary BMF) is the term used in OECD TG 305 to
describe the result of a dietary exposure test, in which exposure via the aqueous phase
is caref ully avoided.

• The BMF value f rom a f ield study (f ield BMF) is the concentration of a substance in a
predator relative to the concentration in the predator’s prey (or f ood) originating f rom
the same ecosystem at steady-state and in which both, water and dietary exposure
may be combined.

The dietary BMF dif f ers f rom a f ield BMF in terms of exposure routes. Furthermore, the
laboratory dietary study is usually not perf ormed using environmentally relevant
concentrations but uses high concentrations in f ood to dose the organism quickly to a level
suf ficient to assess the depuration. However, the test concentration in f ood should not lead to
tissue concentrations/body burden that may cause adverse ef f ects in the test f ish. Another
important dif ference that can occur between the dietary BMF and the f ield BMF f or substances
with biomagnif ication potential, is the variability of f ish growth rates under laboratory and f ield
conditions, but this dif f erence should be covered by growth correction of the kinetic data f rom
juvenile f ish in the dietary exposure test. It is possible to simulate f ield BMFs f rom lab BMFs to
address these two dif f erences using mass balance toxicokinetics (bioaccumulation) models.

Use of the results f rom a dietary exposure study in the B assessment

The dietary bioaccumulation approach results in a BMF rather than a BCF, which is commonly
used f or bioaccumulation assessment. However, Annex 8 of the OECD TG 305 summarises
approaches currently available to estimate kinetic BCFs f rom data collected in the dietary
exposure study. Guidance document on aspects of OECD TG 305 (OECD, 2017) gives f urther
details in chapter 4.6.3, and summarises three available approaches that can be used to
estimate a BCF f rom dietary study data: 1) Uptake rate constant estimation method, 2)
Relating depuration rate constant directly to BCF and 3) Correlating dietary BMF with BCF.

The calculation f or the uptake rate constant estimation method (Method 1) is based on a
model predicted uptake rate constant (k1) and the depuration rate constant (k 2) determined
f rom the dietary bioaccumulation study. In this way, it is possible to use the dietary
experimental data to estimate BCFs, which allow f or a comparison against the BCF criteria f or
PBT assessment outlined in Annex XIII. It should be noted that these calculated BCFs may be
more uncertain than experimental BCFs due to the uncertainty in the k1 prediction. In
particular, k1 is a f unction of chemical properties relating to the chemical transf er ef f iciency
f rom water (e.g., membrane permeation or absorption ef f iciency), the physiology of the f ish
(body size, respiration rate), the experimental conditions (e.g., dissolved oxygen
concentrations, water temperature, gill water pH f or ionic substances) and the
interdependence of these parameters. Several models are available to estimate a k1 value

45
[Link]
 Chapter R.11: PBT/vPvB assessment

105 Version 4.0 – December 2023

needed to calculate an aqueous BCF f rom a dietary bioaccumulation study ( OECD, 2017).
Results f or k1 must be used with ref erence to the models’ assumed applicability domains (e.g.,
mostly restricted to neutral organic substances with log K ow above 3.5 but including some
weakly acidic or basic substances as well). Uptake and elimination processes are dif f erent f or
ions compared to neutral chemicals (Rendal et al., 2011) and ionic substances thus need to be
discussed separately. For poorly soluble non-polar organic substances, f irst order uptake and
depuration kinetics is assumed. More complex kinetic models should be used only f or
substances that do not f ollow f irst order kinetics. Generally, estimates of k 1 should be derived
according to all the models available to give a range of BCFs. These results should be used in a
Weight-of-Evidence approach f or the assessment of bioaccumulation, possibly together with
other inf ormation on bioaccumulation. The estimation of k 1 may be less reliable f or large or
bulky molecules, log Kow above ca 9 and/or low assimilation efficiency (see paragraph 253 of
OECD, 2017). For very hydrophobic substances, k1 estimates may become increasingly
uncertain.

A f ield BMF > 1 indicates that biomagnif ication of a substance occurs. The dietary BMF
however dif fers f rom the f ield BMF, because exposure is through a combination of water and
f ood in the f ield situation, while in the dietary exposure study, the exposure through the water
phase is excluded under controlled conditions. This leads to dietary BMF values that are
generally lower than f ield BMF values. For very bioaccumulative substances (BCF > 5000) such
as the of ten used reference substance hexachlorobenzene, the BMF values sometimes have
been even below one (e.g. Hashizume et al 2018). In a study by Inoue et al. (2012) with carp,
only two of the f ive substances that had a BCF value higher than 5000 L/kg, had a BMF value
in excess of 1. In a study by Martin et al. (2003 a, b) with perf luorinated substances, one of
the three substances with a BCF > 2000 had a BMF of 1.0, while the two others had
substantially lower BMF values. Therefore, a laboratory dietary BMF below 1 cannot be used to
conclude on no B concern and it should be f irst assessed if the bioaccumulation potential can
be concluded based on the estimated BCF, which can be directly compared to the criteria.

The lipid and growth-corrected kinetic BMF (BMFkgL) is the pref erred endpoint in the OECD TG
305-III (OECD, 2012a). However, there have been several observations indicating that a lipid
normalised dietary BMF is unsuitable as an endpoint f or bioaccumulation f rom a regulatory
perspective. For instance, in a study examining biomagnification of hexachlorobenzene in carp
using dif f erent f eeding regimes, it was shown that this BMF kgL did not result in a constant value
f or dif ferent experimental conditions (Hashizume et al. 2018). In particular, the BMF kg did not
correlate well with dietary lipid content (L diet) but was signif icantly correlated with the lipid
content of the f ish (Lfish). Normalising the BMF kg to the lipid content of both f ish and diet
(BMF kgL), yielded increasing BMF kgL values with increasing Ldiet. The same f inding was f ound in
the interlaboratory ring test of the OECD TG 305 dietary study (OECD, 2012b: page 142).
Based on this f inding, Hashizume et al. (2018) recommended to normalise the value of the
BMF kg to a f ish with a 5% lipid content in an ef f ort to standardise dietary BMFs (BMF 5% ).

Gobas et al. (2021) evaluated these results. The presented modelling of these data showed an
increasing lipid normalised dietary BMF L (i.e. BMF divided by Lfish and multiplied by Ldiet) with
increasing lipid content in the diet too, while the BMF 5% was shown to be relatively constant
with Lfish. Thus, also f rom a theoretical point of view, normalising the BMF to both, the lipid
content of the f ish and the lipid content of the diet, makes the resulting BMF L dependent on
f eeding conditions, while reducing variability if results are only normalised to the f ish lipid
content.
 Chapter R.11: PBT/vPvB assessment

106 Version 4.0 – December 2023

Environment Agency (2022) noted that the change in lipid content as the f ood is digested is
important, rather than the dietary lipid content itself , and f urther, concluded that the dietary
BMF deviated in this respect from the f ield BMF. It was reasoned that a dietary BMF that does
not depend on the lipid content of the diet is not inconsistent with f ield BMFs that are
dependent on the lipid content of both predator and its prey. Gobas et al. (2021) stated that
the lipid normalised BMF L is most meaningf ul f rom the point of view of f ugacity increase.
Indeed, this is a very relevant metric for f ield BMFs. However, the observations of low dietary
BMF L f or bioaccumulative compounds and the variability with lipid normalisation make a lipid
normalised dietary BMF L >1 unsuitable as an endpoint f or bioaccumulation f rom regulatory
perspective.

It is noted that Gobas et al. (2021) showed that BMF 5% is on average lower than BMF L, and
highlighted the potential f or miscategorising bioaccumulative chemicals as non-
bioaccumulative using the BMF 5% metric. Because of the uncertainties of dietary BMFs, it
should f irst be assessed if the bioaccumulation potential can be concluded based on the
estimated BCF, which can be directly compared to the criteria. In case this is not possible, the
BMF 5% may be useful to compare results f rom different studies (Environment Agency, 2022).
For any use of the BMF kgL, it is important that the dietary lipid content and the f eeding rate are
reported alongside the value. BMF5% and BMFkgL could be used in a benchmarking exercise.

Besides the calculation of a BCF f rom the depuration phase, the dietary BMF derived f rom the
OECD TG 305-III test can be compared with laboratory BMF values f or substances with known
bioaccumulation potential in a benchmarking exercise (see Correlating dietary BMF with
BCF (Method 3) in OECD, 2017). For example, such an approach has been described f or
dietary bioaccumulation studies with carp (Inoue et al., 2012). Based on a regression between
BCF L and BMF kgL f or nine substances tested in this set-up, it was shown that a BCF L value of
5000 L/kg, normalised to a lipid content of 5%, corresponds to a lipid corrected BMF kgL f rom
the dietary test of 0.31 kg f ood lipids/kg f ish lipids, and a BCF L of 2000 L/kg corresponds to a
BMF kgL of 0.10 kg f ood lipids/kg f ish lipids.

A dif f erent benchmarking could be obtained f rom aqueous and dietary bioaccumulation studies
f or perf luorinated substances with rainbow trout (Martin et al., 2003a, b). These studies
emphasise the f act that even if a BMF f rom an OECD TG 305 dietary bioaccum ulation study is
f ound to be <1, it cannot be considered as a good discriminator f or concluding substances not
to be (very) bioaccumulative according to the BCF criteria of Annex XIII. If benchmarking is
used f or comparing dietary BMF values with BMF values f or substances with a known
bioaccumulation potential, it must be ensured that these BMF values were obtained under
similar conditions (i.e. f ish species, f ish weight/size, diet lipid content, f eeding rate, f ish lipid
content and temperature).

Another endpoint f rom the dietary OECD 305 test is the elimination rate constant. The
elimination rate constant has been proposed as an endpoint f or the bioaccumulation
assessment (e.g. Brooke and Crookes, 2012; Goss et al. 2013, Goss et al. 2018). For example,
Brooke and Crookes (2012) presented lipid normalised depuration rate constants of 0.181 and
0.085 d-1 as critical values f or lipid normalised BCF values of 2000 and 5000. Relating
depuration rate constant directly to BCF is described as Method 2 in Guidance document
on aspects of OECD TG 305 (OECD, 2017). The depuration rate constant is a usef ul metric f or
assessing bioaccumulation. However, it should be noted that the kinetics of uptake and
depuration are still dependent on other f actors, f or example the size of the f ish (e.g. Barber
2008; Brooke and Crookes, 2012). Indeed, f rom the analysis f rom Brooke and Crookes (2012)
 Chapter R.11: PBT/vPvB assessment

107 Version 4.0 – December 2023

there is considerable scatter around the regression line between log BCF L and log k 2 (lipid
normalised), which may be caused by the variability in f ish weight used in the underlying
studies, at least partly. This implies that it is not possible to set one value f or the depuration
rate constant f or dif ferent organisms. If aqueous bioconcentration is considered, an uptake
rate constant of 520 L/kg/d could be estimated f or f ish with a weight of 1 g (Sijm et al., 1995).
The depuration rate constants that lead to bioconcentration f actors of 2000 and 5000 could
thus be estimated to be 0.26 d-1 and 0.10 d-1. For f ish weighing ten grams these values would
be approximately half of these values (0.12 d -1 and 0.05 d-1).

No agreed criterion is available f or the dietary BMF to indicate B or vB. Moreover, the dietary
BMF seems dependent on both the f ood characteristics (type of f ood, lipid content, f eeding
rate) and on f ish characteristics (as also observed f or BCF itself). This will hamper the f urther
derivation of a single value of BMF as a criterion. Similarly, f or the depuration rate constant,
no agreed criterion is available to assess bioaccumulation. The depuration rate constant is also
dependent on other f actors, for example the f ish size, which makes it dif f icult to set a criterion
f or this endpoint too. Therefore, in the B assessment, the pref erred endpoint f rom the OECD
TG 305 dietary exposure test is the BCF value estimated f rom an experimentally derived
elimination rate constant, which can be directly compared to the criteria, unless it can be
demonstrated that the uptake rate constant (k1) cannot be reliably estimated with the
available methods. This would also be consistent with the data treatment of the OECD TG
319A/B in vitro tests, in which experimental data are only available f or the depuration rate
constant. In both cases (OECD 305-III dietary test and OECD TG 319 A/B in vitro tests), the
estimation of the BCF f rom the depuration rate constant f ollows the same principle.

R.[Link].4 In vitro biotransformation data and in vitro-in vivo extrapolation

In vitro methods such as f ish liver S9 and primary hepatocyte assays provide inf ormation on
biotransformation in the organism (OECD TG 319B and OECD TG 319A, respectively). Because
biotransformation is considered to be the dominant mechanism of elimination of hydrophobic
substances, such in vitro tests have the potential to support the assessment of
bioaccumulation in a Weight-of -Evidence approach and may contribute to a reduction in (or
ref inement of ) animal testing. Specifically, in vitro biotransf ormation tests could serve as an
interim step between screening based on physico-chemical properties and in vivo tests.
Kosf eld, et al. (2020) have shown that IVIVE (in vitro–in vivo extrapolation) BCF estimation via
rainbow trout hepatocytes delivers plausible result ranges f or lipophilic organic substances, but
recommend f urther investigations with a broader range of substances. Experience with in vitro
data and IVIVE is still limited, and theref ore it is recommended to f ollow developments in the
f ield, e.g. via the ECHA website.

In vitro biotransformation assays were first developed in the pharmaceutical sector and later
on adapted f or f ish and the use in bioaccumulation assessment (Nichols et al. 2006). Over the
years, the methods were refined and validated (Fay et al. 2014; Nichols et al. 2018) and f inally
adopted in the two OECD test guidelines 319 A/B (OECD 2018b; OECD 2018c), accompanied
46
by a guidance document (OECD 2018a) and excel spreadsheets f or IVIVE calculations . For

46
[Link] GD No 280:
Hepatocytes: [Link] S9-mix:
[Link]
 Chapter R.11: PBT/vPvB assessment

108 Version 4.0 – December 2023

ionisable substances, the OECD TG 319 may apply, however, the currently available in vitro-in
vivo extrapolation models may not always apply to all (types of ) ionisable substances and
adaptation may be needed (Regnery et al. 2022).

The OECD test guidelines 319 A/B (OECD 2018b; OECD 2018c) describe the use of either
cryopreserved rainbow trout hepatocytes or of liver S9 subcellular f ractions f or determining in
vitro biotransformation kinetics in a detailed manner. In brief , the test chemical is inc ubated
together with either hepatocytes or S9 f raction and substrate depletion is monitored over the
duration of the experiment (maximum 4 h). From the measured substrate depletion curve, the
in vitro biotransformation kinetics can be determined. The methods were successf ully applied
f or a variety of chemicals, f or example PAHs (Lo et al. 2015; Nichols et al. 2018; Trowell et al.
2018), f ragrance chemicals (Laue et al. 2014; Laue et al. 2020; Weeks et al. 2020a,b),
sunscreen agents (Saunders et al. 2020), pharmaceuticals (Gomez et al. 2010; Regnery et al.
2022) and ionisable chemicals (Ribbenstedt et al. 2022).

For highly hydrophobic or highly volatile chemicals, performance of in vitro biotransf ormation
assays becomes increasingly challenging and modif ications, f or example , passive dosing
methods or the use of closed vials, are discussed f or these cases (Schug et al. 2018, 2019; see
Guidance Document (OECD 2018a) f or details). Many slowly metabolised substances cannot be
accurately assessed with these in vitro test systems due to their relatively short working
lif etimes (2-4 h) and theref ore the depletion of such chemicals can not be conf idently
quantif ied. The activity of a trout liver S9 substrate depletion assay has been shown to decline
over time, presumably due to proteolytic degradation of biotransf ormation enzymes. To
address this problem, protease inhibitors (i.e., phenylmethylsulfonyl f luoride) have been added
to homogenisation buffers and/or reaction mixtures which can increase the working lif etime of
these assays and theref ore could improve the detection of slow in vitro clearance rates
(Nichols et al. 2021).

To make use of OECD TG 319 data f or bioaccumulation assessment, the application of in vitro–
in vivo extrapolation (IVIVE) bioaccumulation models is needed. Essentially, these models
accomplish two tasks:

i) Converting in vitro biotransf ormation data to corresponding in vivo biotransf ormation

estimates, taking into account (a) dif f erences in binding in vitro and in vivo and (b)
dif f erences in the amounts of metabolically active components (e.g, hepatocytes or S9)
in both cases.

ii) Calculating kinetic BCFs f rom a combination of rate constants f or gill uptake, gill
elimination, f ecal egestion and biotransf ormation.

Over the past years, several IVIVE-bioaccumulation models have become available (Krause
and Goss 2020; Nichols et al. 2013; Trowell et al. 2018). Recent ref inements concerning the
models are the use of the revised in vitro-in vivo extrapolation f ormalism (Krause and Goss
2018) and the use of composition-based binding algorithms (Krause and Goss 2021; Lee et al.
2017; Saunders et al. 2020), rejecting the assumption that binding in vitro and in vivo is the
same (“f u = 1”), which is is especially important in case of hydrophobic organic chem icals.
These aspects should be considered in all IVIVE-bioaccumulation models. Accordingly,
dif f erences between existing models primarily concern the way in which the various rate
constants are combined and the BCFs are calculated. In an evaluation by Krause and Goss
(2020) it was f ound that results with multi-compartment models were similar to results with a
simple one-compartment model, except in cases with f ast biotransformation, in which the BCFs
 Chapter R.11: PBT/vPvB assessment

109 Version 4.0 – December 2023

were underestimated with the one-compartment model. Dif f erent complexities are possible
regarding the models, but in most cases the use of a very simple approach (one-compartment
model) may suf f ice.

The f ollowing inf ormation should be documented and provided in an IVIVE-based

bioaccumulation assessment:

• in vitro test conditions (measured test chemical concentration, number of time points,
species f rom which in vitro material originated, S9/hepatocyte concentration, total
assay volume, open or closed system, assay duration, characterisation of in vitro
material (EROD, GST activities etc.), incubation temperature)

• evidence that the depletion f ollows f irst-order kinetics or that the chemical starting
concentration is below the Michaelis-Menten constant; and documentation of the
behaviour of the negative control (if the negative control shows signif icant losses, the
test should not be used)

• determined in vitro biotransformation kinetics (rate constants or clearances with units)

• estimated in vivo biotransf ormation kinetics (with units) and used extrapolation
f ormalism (with ref erence)

• used IVIVE-bioaccumulation model (with ref erence)

Conceptually, the in vitro methods f all into the class of experimental inf ormation on
bioaccumulation as f ollows: the f low-through tests OECD TG 305-I and OECD TG 305-II
provide experimental inf ormation on uptake and on elimination, the OECD TG 305-III test
provides experimental inf ormation on elimination while using estimates of uptake f or BCF
calculation, and the in vitro methods provide experimental inf ormation on biotransformation as
a central part of elimination and use estimated parameter f or uptake and other elimination
pathways f or BCF calculation. The in vitro assays hardly require animals compared to the other
methods and can also be adapted f or use with air-breathing animals (using appropriate in vitro
material and IVIVE bioaccumulation models), thus also providing a potentially usef ul method
f or terrestrial bioaccumulation assessment.

R.[Link].5 Experimental sediment and soil bioaccumulation data

Experimental sediment bioaccumulation data (experimental Bioaccumulation Factors BAF and

BSAF f or sediment)
In most cases where experimental inf ormation on bioaccumulation is needed, a
bioconcentration test with aquatic species is pref erred due to the better possibilities of
comparing the results f rom such test with the B/vB criteria. However, there may be some very
specif ic cases, where a f ish (or aquatic invertebrate) bioaccumulation test is not expected to
ref lect suf f iciently the bioaccumulation potential but testing of bioaccumulation potential in
sediment might provide the necessary inf ormation f or deriving conclusions on the B/vB -
assessment in a Weight-of-evidence approach.

In line with Annex 1 of the OECD TG 315, the f ollowing def initions are used in this guidance:

• The bioaccumulation f actor (BAF) at any time during the uptake phase of this
bioaccumulation test is the concentration of test substance in/on the test organism (Ca
in g·kg-1 wet or dry weight) divided by the concentration of the substance in the
 Chapter R.11: PBT/vPvB assessment

110 Version 4.0 – December 2023

surrounding medium (Cs as g·kg-1 of wet or dry weight of sediment). In order to ref er
to the units of Ca and Cs, the BAF has the units of kg sediment·kg-1worm.

• The steady state bioaccumulation f actor (BAFss) is the BAF at steady state and does not
change signif icantly over a prolonged period of time, the concentration of the test
substance in the surrounding medium (Cs as g·kg-1 of wet or dry weight of sediment)
being constant during this period of time.

• Bioaccumulation f actors calculated directly f rom the ratio of the sediment uptake rate
constant divided by the elimination constant kinetic rate constants (ks and ke,
respectively - see Annex 1 of the OECD TG 315) are termed kinetic biota-sediment
accumulation f actor (BAF K ).

• The biota-sediment accumulation f actor (BSAF) is the lipid-normalised steady state

concentration of test substance in/on the test organism divided by the organic carbon-
normalised concentration of the substance in the sediment at steady state. Ca is then
expressed as g·kg-1 lipid content of the organism, and Cs as g·kg-1 organic content of
the sediment. BSAF is expressed in kgsediment OC ·kg-1worm lipid content. To reduce variability in
test results f or organic substances with high lipophilicity, the BSAF should be reported
(OECD, 2008).

The units of the concentration values used f or the calculations must all be related either to dry
weight or to wet weight. The unit used should be reported. Optimally, calculations based on
both the wet and the dry weights are presented.

It should be noted that the term biota-sediment accumulation factor (BSAF) has been used in
the literature to ref er to bioaccumulation f actors in sediment which have not been normalised
to organism lipid and sediment total organic carbon content. Care should be taken to ensure it
is clear what the reported value ref ers to.

Bioaccumulation studies on sediment dwelling organisms can be used both f or the screening
and as part of the Weight-of-Evidence assessment of bioaccumulation properties. It should be
considered that (soil or sediment) invertebrate species may have a lower metabolic capacity
than f ish species, e.g. as is the case f or polycyclic aromatic hydrocarbons (Bleeker and
Verbruggen, 2009). Bioaccumulation in these invertebrates may theref ore be higher than in
f ish under the same exposure conditions. No systematic analysis on a comparison with f ish
bioaccumulation data has been conducted, therefore data on sediment dwelling invertebrates
should be considered in a Weight-of-Evidence approach.

The OECD TG 315 Bioaccumulation in Sediment-dwelling Benthic Oligochaetes is the pref erred
method f or generating additional inf ormation. The recommended oligochaeta species are
Tubifex tubifex (Tubif icidae) and Lumbriculus variegatus (Lumbriculidae). The species
Branchiura sowerbyi (Tubificidae) is also indicated but it should be noted that it has not been
validated in ring tests at the time of writing. The bioaccumulation factor (expressed in kg wet
(or dry) sediment·kg-1 wet (or dry) worm) is the main relevant outcome and can be reported
as a steady state biota-sediment accumulation f actor BAF ss or as the kinetic biota-sediment
accumulation f actor (BAFK ). In both cases the sediment uptake rate constant k s (expressed in
kg wet (or dry) sediment·kg-1 of wet (or dry) worm d-1), and elimination rate constant k e
(expressed in d-1) should be reported as well. The normalised biota-sediment accumulation
f actor (BSAF) should be additionally reported for highly lipophilic substances. It is important to
 Chapter R.11: PBT/vPvB assessment

111 Version 4.0 – December 2023

consider the implications of the worm gut contents when interpreting the study results (Mount
et al, 1999; OECD TG 315).

OECD TG 315 recommends the use of artif icial sediment. If natural sediments are used, the
sediment characteristics should be specif ically reported as described in the test guideline. For
lipophilic substances, BSAFs of ten vary with the organic carbon content of the sediment.
Typically a substance will have greater availability to the organism when the sediment OC is
low, compared to a higher OC. It should be considered to test at least two natural sediments
with dif f erent organic matter content, and the characteristics of the organic matter, in
particular the content of black carbon, should be reported. To ensure comparability of results
between dif ferent sediments, BSAF normalised to organism lipid and sediment total organic
carbon content is used. This allows tests on the same substance and tests on dif f erent
substances to be comparable. The load rate should be as low as possible and well below the
expected toxicity, however it should be suf f icient f or ensuring that the concentrations in the
sediment and in the organisms are above the detection limit throughout the test.

The relevance of bioavailability of the substance f or the test organism should also be
considered and if relevant and possible, bioaccumulation could be expressed as a BCF between
organism and dissolved pore water concentrations.

It should be noted that it is not possible to give any threshold values f or using sediment BSAF
values in PBT assessment. Therefore, this type of test does not directly result in endpoints that
can be compared to the REACH Annex XIII bioaccumulation criteria although the BSAF in
combination with Kow /Koc can provide evidence of high bioaccumulation potential (Appendix
R.11-3). BCF values can be calculated based on measured or estimated pore water
concentrations according to Appendix R.11-3. If BCF values are normalised to a lipid content of
5%, they can be considered as a conservative estimate f or f ish, because metabolism is
generally much lower in invertebrates than in f ish. A case-by-case assessment based on expert
judgement of the reliability and relevance of the available inf ormation is required in order to be
able to give BSAF values an appropriate weight in the B and vB assessment.

Other indications of a high bioaccumulation potential, such as a bioaccumulation process not

reaching the steady state at the end of the exposure period of OECD TG 315 test or a low
depuration rate, both representing slow kinetics, are relevant parts of a Weight-of-Evidence
approach when considering whether B or vB criteria are f ulfilled. Substances with background
sediment concentrations and potentially adaptable uptake mechanisms need caref ul
consideration because sediment-dwelling organisms may have adapted to such substances,
potentially af f ecting the bioaccumulation process.

Experimental soil bioaccumulation data (experimental Bioaccumulation Factor BAF and BSAF
f or soil)
The Bioaccumulation in Terrestrial Oligochaetes test according to OECD TG 317 measures the
residues of a test substance in earthworms af ter exposure to the test substance mixed with
soil. In those cases where the bioaccumulation concern is f or bioaccumulation in air -breathing
species and not so much f or aquatic organisms, a bioaccumulation test in soil organisms like
the earthworm (Eisenia fetida) will not be suf ficient to disprove concern for bioaccumulation in
vertebrates/mammals. The test is usef ul f or aspects of secondary poisoning, and may support
the bioaccumulation assessment in a Weight-of-Evidence approach. The earthworm may be
more in equilibrium with the pore water and its diet is not comparable to a vertebrate diet.
 Chapter R.11: PBT/vPvB assessment

112 Version 4.0 – December 2023

Bioaccumulation studies with terrestrial invertebrates (e.g. according to OECD TG 317) are
insuf f icient to assess the bioaccumulation potential in terrestrial mammals and other air -
breathing organisms.

In line with Annex 1 of the OECD TG 317, the f ollowing def initions are used in this guidance:

• The bioaccumulation f actor (BAF) at any time during the uptake phase of this
bioaccumulation test is the concentration of test substance in/on the test organism (C a
in g·kg-1 dry (or wet) weight of worm) divided by the concentration of the substance in
the surrounding medium (Cs as g·kg-1 of dry (or wet) weight of soil); the BAF has the
units of kg dry (or wet) soil·kg-1 dry (or wet) worm.

• The steady state bioaccumulation f actor (BAFss) is the BAF at steady state and does not
change signif icantly over a prolonged period of time, the concentration of the test
substance in the surrounding medium (C s as g·kg-1 of dry (or wet) weight of soil) being
constant during this period of time.

• Bioaccumulation f actors calculated directly f rom the ratio of the soil uptake rate
constant and the elimination rate constant (ks and ke,) are termed kinetic biota-soil
accumulation f actor (BAF K ).

• The biota-soil accumulation f actor (BSAF) is the lipid-normalised concentration of the

test substance in/on the test organism divided by the organic carbon-normalised
concentration of the test substance in the soil at steady state. Ca is then expressed as
g·kg-1 lipid content of the organism, and Cs as g·kg-1 organic content of the soil; the
BSAF has the units of kgsoil OC ·kg-1lipid.

The units of the concentration values used f or the calculations must be all related either to dry
weight or to wet weight. The unit used should be reported. Optimally, calculations based on
both the wet and the dry weights are presented.

It should be noted that the term biota-soil accumulation f actor (BSAF) has been used in the
literature to ref er to bioaccumulation f actors in soil which have not been normalised to
organism lipid and soil total organic carbon content. Care should be taken to ensure it is clear
what the reported value ref ers to.

Earthworms and enchytraeids are the recommended taxonomic groups to be tested according
to OECD TG 317. To ensure comparability of results between different soils, BSAF normalised
to organism lipid and soil total organic carbon content is used. The dependence of these values
on the concentrations of the substance in soil, and when relevant, the soil characteristics
should be specif ically reported.

The bioaccumulation of ten varies with the organic carbon content of the soil. Typically a
substance will have greater availability to the organism when the soil organic carbon content is
low, compared to a higher OC. To ensure comparability of results between dif f erent soils, a
BSAF should be derived by normalising the results both to the soil organic carbon content and
the lipid content of the organisms employed. The load rate should be as low as possible and
well below the expected toxicity, however it should be suf f icient to ensure that the
concentrations in the soil and in the organisms are above the detection limit throughout the
test.
 Chapter R.11: PBT/vPvB assessment

113 Version 4.0 – December 2023

The relevance of bioavailability of test substances potentially containing irreversibly bound

f ractions in soil should also be considered and, if relevant and possible, the BSAF should be
corrected f or the bioavailable f raction.

It should be noted that it is not possible to give any threshold values f o r BSAF in soil.
Therefore, this type of test does not directly result in endpoints that can be compared to the
REACH Annex XIII bioaccumulation criteria although the BSAF in combination with K ow /Koc can
provide evidence of high bioaccumulation potential (Appendix R.11-3). BCF values can be
calculated based on measured or estimated pore water concentrations according to Appendix
R.11-3. If BCF values are normalised to a lipid content of 5%, they can be considered as a
conservative estimate f or f ish, because metabolism is generally much lower in invertebrates
than in f ish. A case-by-case assessment based on expert judgement of the r eliability and
relevance of the available inf ormation is required in order to be able to give BSAF values an
appropriate weight in the B and vB assessment.

Other indications of a high bioaccumulation potential such as a bioaccumulation process not

reaching the steady state at the end of the exposure period of an OECD TG 317 study or a low
depuration rate, both representing slow kinetics, are relevant parts of a Weight-of-Evidence
approach when considering whether the B or vB criteria are f ulf illed. It should be noted that
organo-metals and other substances with background soil concentrations and potentially
adaptable uptake mechanisms require particularly caref ul consideration, as the soil-dwelling
organisms may have adapted to such substances which potentially affects the bioaccumulation
process.

Some additional parameters relevant to bioaccumulation that can potentially be used f or

screening or in a Weight-of-Evidence approach, may be derived f rom other invertebrate
studies. For the OECD TG 222 earthworm reproduction test, in which earthworms are exposed
f or 28 days to a test substance spiked into soil, it has been demonstrated that at test end
(provided that the relevant analytical procedures are available) the concentration of the test
substance in the adult worms can give an indication of uptake into the organism (Kinney et al.,
2012). Care must be taken that the assessment of uptake is perf ormed at a non-toxic test
concentration (i.e. at which less than 10% mortality and no signif icant loss of body weight
compared to control occurs over the 28d test period). It must also be noted that only uptake is
measured at test termination and that elimination of the substance is not considered. As such,
the results of this test should be interpreted with caution, but it can provide valuable screening
inf ormation on substance accumulation that can help as preliminary inf ormation for considering
whether more specific testing f or bioaccumulation according to OECD TG 317 is needed. The
same approach could potentially be usef ul f or other guideline studies on invertebrate species
as well, such as the 21 day larval survival test on dung beetles (OECD GD 122), the
developmental test with dipteran f lies (OECD TG 228) or the collembolan reproduction test
(OECD TG 232), depending on the expected route of exposure. However, measuring tissue
residues in these studies could be hampered by the small size of the test organisms (Hoke et
al., 2015).

R.[Link].6 Field data and biomagnification

Field bioaccumulation f actors (Field BAF calculated f rom monitoring data, f ield measurements
or measurements in mesocosms) or specif ic accumulation in f ood chains/webs expressed as
biomagnif ication f actors (BMFs) or trophic magnif ication f actors (TMFs) can provide
supplementary inf ormation indicating that the substance does or does not have
 Chapter R.11: PBT/vPvB assessment

114 Version 4.0 – December 2023

bioaccumulation potential. Furthermore, the same inf ormation may be used to support the
assessment of persistence, in particular f or possible long-range transport, if signif icant
concentrations are f ound in biota in remote areas or in locations distant f rom known point
sources. If f ield data indicate that a substance is ef fectively transferred in the f ood chain, this
is a strong indication that it is taken up f rom f ood in an ef ficient way and that the substance is
not easily eliminated (e.g. excreted and/or metabolized) by the organism (this principle is also
used in the f ish f eeding test f or bioaccumulation), which will lead to biomagnif ication f rom prey
to predator (trophic magnification). A reliable f ield BMF or TMF value higher than 1 (see also
Section R.[Link] in Chapter R.7c of the Guidance on IR&CSA) can also be considered as an
indication of very high bioaccumulation. For aquatic organisms, this value indicates an
enhanced accumulation due to additional uptake of a substance f rom f ood along with direct
accumulation f rom water. However, as dietary and trophic biomagnification represent different
processes than bioconcentration in aquatic organisms, f ield BMF and/or TMF values <1 cannot
be directly used to disregard a valid assessment based on reliable BCF data f ulf illing the
numerical B/vB criteria in Annex XIII to the REACH Regulation, but in this kind of case all
available data need to be considered together in a Weight-of-Evidence approach.

The quality of f ield data needs to be assessed and interpreted correctly. Respective guidance
documents and recommendations f or assessing the quality of biomonitoring data including
interpretation of wildlif e biomonitoring has been elaborated by the EU project LIFE APEX
(Badry et al., 2022a; Badry et al., 2022b; Treu et al., 2022a) and Guidance Document No. 32
on Biota Monitoring prepared under the Water Framework Directive 2000/60/EC (European
Commission, 2014).

Types of field/biomonitoring data and objectives

Field or biomonitoring data can be used as supporting evidence that substances are taken up
by organisms in the environment. Indeed, the substance concentrations in wildlif e samples
show that uptake is actually taking place f rom the environment. Care should be taken if gut
content and adsorption to skin contribute signif icantly to the measured concentration.

At present, an increasing number of biota monitoring data of high quality are generated under
dif f erent European and global initiatives. The European Commission (EC) has made signif icant
ef forts to make (biota) monitoring data better accessible and comparable, e.g. via the platf orm
IPCHeM ([Link] or the NORMAN network ([Link]
[Link]/apex/). These open access data bases can be used to screen f or presence and
concentrations of contaminants in wildlif e and humans.

BCFs, BAFs express ratios of substance concentrations in biota to water, while BMFs and TMFs
ref lect ratios of substance concentrations in predator–prey relationships (Burkhard et al.,
2012). Field BAF or f ield BMF of a substance may be greater than what is estimated based on
BCF and BMF f rom laboratory experiments. This is because in the laboratory tests f ish are
exposed either via water or via f ood, while under f ield conditions organisms are exposed to
substances via all exposure routes depending on where they live (terrestrial or aquatic) and
which taxa they belong to (air-breathers or water-breathers like f ish). Furthermore, apex (top)
predators ref lect biomagnif ication over the whole f ood chain while laboratory tests usually
include only one trophic level in the biomagnif ication process f rom diet to test organism. This
will ultimately lead to higher bioaccumulation in wild organisms f eeding at higher trophic levels
compared to the laboratory experiments for substances that are not rapidly metabolized and
eliminated. Furthermore, the duration of exposure is expected to be substantially longer in wild
 Chapter R.11: PBT/vPvB assessment

115 Version 4.0 – December 2023

animals as compared to the laboratory tests, which can play a substantial role in long-lived
species such as many apex predators that accumulate hydrophobic substances over a lif etime.
Bioaccumulation measurements of very hydrophobic, persistent substances that have not
approached steady-state in a f ield study, are considered to be underestimations (Burkhard et
al., 2012). Despite this, wildlif e monitoring data can give valuable indication of an increased
bioaccumulation potential particularly f or dif f icult to test chemicals.

Kelly et al. (2007) explained that apart f rom low rate of respiratory elimination to air, higher
biomagnif ication of certain organic substances in air-breathing organisms is due to the greater
ability to absorb and digest their diet, which is related to dif f erences in digestive tract
physiology and body temperature. In this context, f ield data on bioaccumulation and
magnif ication in air-breathing biota again can provide valuable inf ormation f or identif ying
substances that accumulate in wildlif e and in human f ood webs (Czub and McLachlan, 2004).

Field bioaccumulation metrics

Following types of biota f ield data can generally be considered in the context of
bioaccumulation screening and assessment:

• Bioaccumulation factors (BAFs/BSAFs)

Field studies can be used to derive bioaccumulation f actors (Field BAFs). They usually have
been done with f ish or aquatic invertebrates such as mussels and crustaceans and have been
used to develop water quality standards (Moermond and Verbruggen, 2013). The
bioaccumulation f actor (BAF) can be expressed f or simplicity as the steady -state (equilibrium)
ratio of the substance concentration in an organism to the concentration in the surrounding
medium (e.g. water in natural ecosystems). Similarly, f ield measured biota-sediment or biota-
soil accumulation f actors (BSAF) are derived by the concentration of a substance in biota
divided by the concentration in the sediment or soil (Burkhard et al., 2010).

Substance concentrations in biota that are orders of magnitude higher than those in water and
air are important f or several reasons. Such large concentrations may adversely af f ect
organisms across f ood webs, if internal concentrations reach toxic levels (Mackay et al., 2018).

Field BAF values are largely variable due to site-specif ic environmental conditions af f ecting
their determination and less available than BCF values (Arnot and Gobas 2006; Weisbrod et al.
2009; Costanza et al. 2012). The basis f or the f ield BAF value is the ratio of the concentration
in wet weight (ww) of the organism divided by the water concentration. The unit of the f ield
BAF is L·kgww -1. It is recommended that the f ield BAF is reported in terms of wet weight as
well as dry weight and is also normalised to lipid weight, with an explanation of how the
normalisation was perf ormed (European Commission, 2018).

If f ield BAF values (based on reliable inf ormation) are above the criteria f or B or vB it should
be considered whether this inf ormation is suf ficient to conclude that the substance meets the B
or vB criteria as part of the Weight-of-Evidence approach. For comparison of a f ish f ield BAF
with the Annex XIII criteria, BAF values should be on wet weight basis and f or whole body and
also lipid normalised to 5%. Care should be taken that the exposures f rom all relevant routes
and compartments are considered when f ield BAF values are evaluated.
 Chapter R.11: PBT/vPvB assessment

116 Version 4.0 – December 2023

• Field biomagnification

BMFs describe the increase in concentrations f rom prey to predator. For f ield data, BMF values
are related to BAF values as both prey and predator are f rom the same environment (BMF
prey-predator = BAFpredator/BAF prey).Food chain transfer and secondary poisoning are basic
concerns in relation to PBT and vPvB substances, and theref ore an indication of a
biomagnif ication potential (BMF and/or TMF > 1) can on its own be considered as a basis to
conclude that a substance meets the B or vB criteria. However, absence of such a
biomagnif ication potential cannot be used to conclude that these criteria are not f ulf illed. This
is because a f ield BMF only represents the degree of biomagnif ication in the specif ic
predator/prey relationship f or which it was measured. Biomagnif ication will vary be tween
predator/prey relationships, so a low f ield BMF in one does not mean that it will be low in other
predator/prey relationship. Evidence of high biomagnif ication in one predator/prey relationship
is an indication that biomagnif ication may also occur in other (unmeasured) predator/prey
relationships.

Substances that partition into lipids should, as f ar as possible, be lipid normalised to account
f or dif ferences in lipid content between prey and predator. It allows f or a comparison of f ield
BMF values in a direct and objective manner. It should however be noted that non-lipophilic
substances such as PFAS may bioaccumulate by other mechanisms than partitioning/binding to
lipids. In such a case, another reference parameter than lipid content may be considered f or
normalisation, e.g. dry weight or protein content. Normalisation of measured data with respect
to lipid and dry weight content is described in Guidance Document No. 32 on Biota Monitoring
prepared under the Water Framework Directive 2000/60/EC (European Commission, 2014).

• Trophic magnification factor (TMF)

TMF can be used to understand the biomagnification potential of a substance as it represents

the average increase or decrease of concentration levels in a f ood web per trophic level (TL): a
TMF higher than 1 indicates that the substance biomagnifies in the f ood web (i.e. concentration
increases with each trophic level) and thus can on its own be considered as a basis to conclude
that a substance meets the B or vB criteria; a TMF lower than 1 indicates that the substance
undergoes trophic dilution (Weisbrod et al. 2009). The TMF f or a f ood web is calculated as the
exponent of the slope of the natural logarithm transformed concentrations f or organisms in the
f ood chain as a f unction of the trophic level of these organisms. As such, the TMF represents
the average biomagnif ication per trophic level within that f ood web. Currently, there is no
standard procedure f or studying TMFs. Hence, the conductance and sampling may vary
considerably between dif ferent studies. The validity of the TMF is strongly dependent on the
spatial and temporal scales over which the samples were retrieved. TMF can show variability
related to ecosystem characteristics, organism biology and ecology, study design, and the
statistical methods used f or TMF calculation (Kidd et al. 2019). More reliable TMFs may be
derived f rom data f or non-migratory species originating f rom a conf ined area and sampled in
the same period, or f rom f ood chains f or which low variability in time and space can be
assumed (e.g. f or vast remote areas). See also publications f rom Borgå et al., 2012; Kidd et
al., 2019; Kosf eld et al., 2021; Rüdel et al., 2020, ECETOC (2014), Burkhard et al., 2013,
Guidance Document No. 27 f or deriving Environmental Quality Standards prepared under the
Water Framework Directive 2000/60/EC (European Commission, 2018) and Kidd et al. (2019)
f or discussion on uncertainties.
 Chapter R.11: PBT/vPvB assessment

117 Version 4.0 – December 2023

As f or the BMF, f or the assessment of substances that partition into lipids the TMF should be
derived f rom lipid-normalised biota concentrations versus TL. For non-lipophilic substances,
another reference parameter than lipid content may be considered f or normalisation, e.g. dry
weight or protein content. Normalisation of measured data with respect to lipid and dry weight
content is described in Guidance Document No. 32 on Biota Monitoring prepared under the
Water Framework Directive 2000/60/EC (European Commission, 2014).

The way data, on the basis of which the TMF values are calculated, are treated has a great
impact on the outcome of the TMF value. Not only the magnitude of the TMF value can be
impacted, but also whether biomagnification or biodilution occurs. In addition, the setup of the
f ield study could have an inf luence on the resulting TMF values as well. These aspects cover
both spatial and temporal variability in sampling, but also the selection of species belonging to
the f ood web (Kosfeld et al., 2021). Spatial variability can lead to dif f erent organisms being
exposed to dif ferent environmental concentrations (Kim et al. (2016). Temporal dif f erences
could have a strong impact on trophic magnif ication as well. Such temporal variability f urther
complicates the interpretation of the observed TMF values. If there are large dif f erences in
space or time with regard to the samples within a f ood chain, the results of such a f ood web
study should be considered with care. Further, it appears that TMF values could be strongly
dependent on the inclusion or exclusion of certain species and on which part of the f ood chain
is considered, for example pelagic species only or the benthopelagic f ood web. Apart f rom that,
even f rom similar f ood webs widely varying results can be obtained f or the TMF (Houde et al.,
2008). The Figure R.11—5 illustrates the relevant metrics used in the B/vB assessment.

Figure R.11—5: Schematic overview on bioaccumulation metrics: laboratory

bioconcentration factor (lab BCF), laboratory biomagnification factor (lab BMF), field
biomagnification factor (field BMF), bioaccumulation factor (BAF) referenced to
water, biota sediment or soil accumulation factor (BSAF), and trophic magnification
factor (TMF). TMF is obtained from the slope of a plot of concentration versus trophic
level. Please note that lipid content of the organisms need to be considered in the
assessment for substances that partition into lipids. For non-lipophilic substances,
another reference parameter than lipid content may be considered for normalisation,
 Chapter R.11: PBT/vPvB assessment

118 Version 4.0 – December 2023

e.g. dry weight or protein content. Red bullets refer to chemicals, green lines refer to
sediment and orange lines refer to soil.

• Detection of substances in wildlife

The detection of substances in wild biota (concentration or occurrence data), in particular in

apex species (top predators), provides a clear indication that it has been taken up by that
organism. Care should be taken if gut content and adsorption to skin contribute signif icantly to
the measured concentration. These data could be used at the screening level or within a
Weight-of-Evidence approach to assess bioaccumulation of a substance case by case
(depending on the statistical power, quality and standardisation of the study). However, a
detection of a substance as such does not necessarily mean that signif icant bioconcentration or
bioaccumulation has occurred since exposure level f rom the surrounding media and/or diet
would be needed f or such an assessment. Thus, concentrations measured in prey species or
water in the surrounding media can be helpf ul to identif y cases where bioaccumulation
occurred in wild organisms. Furthermore, data f rom different time points as well as regions can
give indications on temporal and spatial trends.

In cases where no data is available on sources and contemporary exposure levels, a high
f requency of appearance of a substance in several biota species across different compartments
could indicate bioaccumulation potential. In such cases, other available evidence of the
substance’s bioaccumulation potential should be thoroughly examined bef ore reaching a
conclusion.

Detection of elevated levels of a substance in biota compared to levels in their surrounding

environment indicates an increased concern f or bioaccumulation. Reliable monitoring data can
be used as line of evidence that the substance meets the B/vB criteria.

Concentrations in biota increasing with age due to exposure and accumulation over lif e-time,
particularly in long-lived apex species (top predators), indicate an increased concern f or
bioaccumulation.

Finally, it is is important that the quality of monitoring data (detection or quantif ication of a
substance in biota) needs to be assessed and interpreted correctly.

R.[Link].7 Addressing uncertainty of field data in bioaccumulation screening

and assessment

The uncertainties related to f ield data apply to all f ield metrics described above. If f ield data
are available, these should be considered in the assessment. In particular, if the number of
f ield studies is not very high and do not cover a range of conditions and/or species, a
comprehensive discussion on the uncertainties is required. Generally, bioaccumulation and
biomagnif ication are inf luenced by an interplay between physicochemical properties of a
substance, source distribution, trophic interaction, species biology and many other biotic and
abiotic f actors which are not f ully understood yet. This makes the interpretation of ten dif f icult.
The f ollowing elements are essential to be discussed f or each study (where relevant) and when
compiling the inf ormation f rom the studies together to draw an overall conclusion f rom the
f ield studies:
 Chapter R.11: PBT/vPvB assessment

119 Version 4.0 – December 2023

Sampling:

• Quality assurance and reporting quality criteria throughout the sampling, sample
treatment, storage and analysis (including (f ield-) blanks and spiked samples) of biota
samples; detailed guidance can be f ound at the Lif e Apex project dissemination website
as well as in Badry et al. (2022a and b).

• Sampling design (site selection, spatial resolution, f requency of determination, etc.)

and details of the sampling methodology.

• Dividing the concentration of a substance in a predator by that in a prey implies that

this prey is the sole f ood source. However, the f ood sources may be diverse.
Additionally, there is no standard procedure so f ar how to conduct such f ield studies,
and dif f erent study designs may therefore have an inf luence. The uncertainties of f ield
studies have been addressed and discussed by Borga et al. (2012) and Kidd et al.,
(2019).

• Problems arise with increasing body size of predators because analysis is based on
tissue or serum samples. This is especially true f or organisms at the higher trophic
levels (e.g., polar bears), while it is f easible to measure the whole -body on smaller
species at lower trophic levels. Whole-body analysis is not f easible f or ethical reasons,
and due to the challenging logistics with respect to sampling and laboratory constraints.
Therefore, in many cases the derived f ield BMF-values are restricted to certain tissue
samples rather than whole body samples.

• It is worth noting that Jurgens et al. (2013) discusses dif f erences in contaminant
content in whole f ish versus dissected tissues, and points out that “although
concentrations of hydrophobic substances tended to be higher in the liver than in the
rest of the f ish, the dif f erence largely disappeared when the results were lipid-
normalised”. This suggests that regimes that use (or have used) liver sampling can
deliver results that are comparable to whole f ish sampling if the data are lipid-
normalised.

• Tissue-to-whole body extrapolations of measured concentrations, where this cannot be

avoided, introduce additional uncertainties which need to be addressed.

• The number of organisms sampled at each point of the f ood web.

Species ecology:

• Interspecies dif f erences in gut physiology, diet pref erence, f oraging strategies,
environmental interactions, mobility and migration, physiological dif ferences, and other
species-specif ic ecological traits can have important consequences f or substance
exposure, uptake and metabolism as recently reviewed e.g. f or birds (Kuo et al., 2022).

• Therefore, the inf luence of sampling location(s) and timing(s), concentration gradients
and migration behaviour need to be considered. In particular , migratory behaviour
might strongly impact exposure levels. Care should be taken that the samples used to
derive bioaccumulation and biomagnification/trophic magnification factors are collected
at the same time and f rom the same location, and suf f icient details are provided.

• Details of the organisms being analysed, including species, sex, size, weight, lipid
content, lif e history, and ecological traits (e.g. migration, diet, and f ood web structure
 Chapter R.11: PBT/vPvB assessment

120 Version 4.0 – December 2023

which may be determined using measurements on nitrogen or carbon isotopes, and

composition). For resident species, the sample collection should be f airly
straightf orward. Migratory species may present special challenges in determining which
f ood, sediment, or water sample should be used to calculate the f ield biomagnif ication
f actor.

• Dif f erence between poikilotherms and homeotherms (cold and warm blooded) : An
investigation of an Arctic f ood web revealed the unequal magnif ication behaviour of
POPs within both thermal groups (Hop, 2002). These results may be explained by a
higher f ood intake, caused by a higher energy demand, and a longer lif e span of birds
and mammals. Intrinsic dif ferences in gastrointestinal absorption mechanisms have also
been suggested as an explanation f or these dif f erences between homeotherms and
aquatic poikilotherms (Drouillard, 2000). Theref ore, when the trophic magnif ication
potential of a substance is determined via a single regression f or the overall f ood web,
the magnif ication in poikilotherms may be overestimated and the magnif ication in
homeotherms, in particular apex predators, may be underestimated (Fisk et al., 2001).

• Inf luence of species physiological characteristics (e.g. typical lipid content, whether air -
breather or water respirer); Inf luence of digestion rate/diet energy content, size and
growth, ability to metabolise, sex, age;The inf luence of habitat and exposure to
potential point pollution (e.g. avoidance of human settlements especially in the case of
hemerophile species);Inf luence of diet (generalist vs specialist), which might be
controlled by analysing stable isotope values of predators and potential prey species
(d15N, d13C); challenges and alternative methods like Sulphur stable isotope analysis
are discussed in Elliott et al. (2021).

Uptake routes and analytical considerations:

• Data on biomagnif ication (TMF, BMF or BAF-values) should be calculated based on lipid-
normalised concentrations (unless lipid is not important in the partitioning process, e.g.
f or protein binding substances). Normalisation of measured data with respect to lipid
content is described in Guidance Document No. 32 on Biota Monitoring prepared under
the Water Framework Directive 2000/60/EC (European Commission, 2014).

• Opportunistic f eeders vary their diet and point sources may inf luence observed BMFs
and TMFs. Additionally, apart f rom the diet, direct uptake of a substance by an apex
predator occurs, e.g. f rom contaminated air or water. The relative importance of f ood
versus e.g. water exposure therefore inf luences the magnitude of the TMF. It has to be
noted that f or typical PBT substances the relative contribution of uptake via air and
(drinking) water is negligible compared with f ood.

• Data on the concentrations in the surrounding environment and on the tem poral trend
of environmental inputs are of ten missing, which complicates the bioaccumulation
assessment when only using monitoring data.

• Selck et al. (2012) showed that at lower trophic levels (mayf ly and polychaete),
variability in bioaccumulation of benthic organisms is mainly driven by sediment
composition and substance partitioning to sediment components, which is in turn
 Chapter R.11: PBT/vPvB assessment

121 Version 4.0 – December 2023

47
dominated by the inf luence of black carbon . At higher trophic levels (yellow perch and
the little owl), f ood web structure (i.e., diet composition and abundance) and substance
concentration in the diet became more important particularly f or the most persistent
substance, PCB-153. These results suggest that variation in bioaccumulation
assessment can most effectively be reduced by improved identif ication of f ood sources
as well as by accounting f or the substance bioavailability within the f ood web.

• Although tissues of top predators can be used to monitor contaminant levels in the
environment, variation in diet can conf ound the interpretation of the results (Braune et
al., 2014a; Braune et al., 2014b; Hebert et al., 2000). Theref ore, an accurate B-
assessment f or higher trophic level predator species such as raptors or marine
mammals requires knowledge on the substance concentration in potential prey species.

• Stable isotope (SI) analysis is an integrated approach to improve the overall

understanding of f eeding behaviour and f ood web ecology and to unravelling animal
diets, as stable isotopes in animal tissues and excreta reflect dietary pref erences, and
yield insight into the environmental conditions experienced by the animal (e.g. West et
al., 2006, review by Kelly, 2011). Thorough elucidation of the f ood-web structure
(f eeding ecology; determination of the trophic level) and identif ication of the position in
the f ood web is important f or interpreting BMFs/TMFs. Nitrogen stable isotopes are
commonly used as proxies f or estimating the trophic position of animals since
consumers get enriched with 15N in relation to 14N (ratio expressed as δ15N) by
compared to their prey (Vanderklif t and Ponsard, 2003), whereas measuring the
enrichment of 13C in relation to 12C (ratio expressed as δ13C) can be used to
distinguish dif f erent carbon sources of terrestrial (depleted δ13C) and aquatic (enriched
δ13C) environments (Kelly, 2000). From the literature it seems that an average
dif f erence in δ15N of 3.4‰ between adjacent trophic levels has been suggested to be
remarkably constant among different types of consumers in aquatic f ood chains (Post,
2002; Vander Zanden and Fetzer, 2007).For f reshwater f oodchains, the trophic level is
calculated assuming an enrichment of 2 to 5‰ (usually 3.4 or 3.8‰) f or δ15N (based
on stable nitrogen isotope ratios) per trophic level (European Commission, 2018). The
δ13C is a usef ul marker to distinguish benthic, benthopelagic and pelagic carbon
sources. Although all three isotopes are associated with habitat, δ13C also varies
systematically with trophic position and δ18O ref lects variable contributions f rom diet.
In contrast, δ34S may be a particularly usef ul dietary tracer of spatial origin as δ34S
varies little f rom source to consumer (Florin et al., 2011). The combined use of multiple
isotopes could provide a more nuanced description of f ood web structure in the context
of environmental pollution. In particular, the use of amino acid specif ic δ15N and δ13C
values, including both those that change f or predator to prey (‘trophic’ and ‘non-
essential’ amino acids, respectively, f or δ15N and δ13C) and those that do not (‘source’
and ‘essential’), may additionally ref ine the dietary estimate (Elliott et al., 2021).

• It is worth noting that the relative abundance of these isotopes and thus the
determination of the trophic level and TMF is inf luenced by the physiology of the
organism and its lif e trait history. Rapid growth with a higher pr otein demand f or new

47
Black carbon is formed by the incomplete combustion of fossil fuels, wood and other fuels and has a strong binding
capacity for organic chemicals.
 Chapter R.11: PBT/vPvB assessment

122 Version 4.0 – December 2023

tissue leads to lower enrichment f actors than those with slower growth rates.
Insuf ficient f ood supply and f asting and starvation leads to catabolism of body proteins
and an increase of 15N in organisms relative to those organisms with adequate f ood
supply.

• An alternative approach to the use of stable isotopes in order to elucidate the diet
composition of the animals can be the analysis of the diet with hard component analysis
(i.e. bones…) of the stomach content or metabarcoding of the gut content.

• Many f ood webs are complex, and a single metric (f ood chain length) is unable to
represent all variation in relationships (Elliott et al., 2021). Contaminant levels of ten
vary among habitats due to processes such as long-range transport, point source
pollution, microbial degradation of organic substances, and variation in processes at the
base of f ood webs might be as important as biomagnif ication f or understanding
contaminant levels in higher trophic level species (‘habitat variation hypothesis’; (Elliott
and Elliott, 2016; Lavoie et al., 2015). Additional dietary tracers associated with the
spatial origin of diet (δ13C, δ18O, δ34S) are recommended to ref ine diet reconstruction
(Elliott et al., 2021; Elliott and Elliott, 2016; Hobson et al., 1993).

Also where a high number of f ield studies are available, the discussion on uncertainties
mentioned above may support the assessment. It should also be noted that f ield studies of ten
sample vertebrate species. Theref ore, as Annex XI to the REACH Regulation requires
vertebrate testing to be the last resort, the need f or additional f ield studies requires caref ul
consideration f or whether alternative sources (e.g., already existing stored samples f rom
specimen banks) could provide the same inf ormation, particularly in the light of uncertainties
stated above.

Further considerations on f ield evaluation of bioaccumulation (with particular f ocus on

terrestrial bioaccumulation) can be f ound in Van den Brink et al. (2016). Furthermore, a recent
report on the use of biomonitoring f rom apex predators to support the screening f or
bioaccumulative chemicals was developed by the LIFE APEX project (Treu et. al., 2022b, Badry
et al. (2022a and b)).

R.[Link].8 Bioaccumulation in air-breathing organisms

Although f or many substances the assessment of bioaccumulation in aquatic species is

suf ficient, some substances like endosulf an, beta-hexachlorocyclohexane, many perfluorinated
alkyl substances or highly lipophilic substances may accumulate more than expected in air-
breathing organisms and are not recognised as highly bioaccumulative if only aquatic data are
used in the assessment (Kelly and Gobas, 2001; Kelly and Gobas, 2003; Czub and McLachlan,
2004). One reason may be the ability of gill-breathing organisms to eliminate substances into
the water that cannot be eliminated by air-breathing organisms by respiration as they are not
volatile. For mammals and birds, bioaccumulation essentially occurs through the dietary route,
associated with elimination via urination and gastrointestinal tract, metabolism, exhalation and
growth (dilution) (Kelly and Gobas, 2003; Kelly et al., 2007). In this context, air-breathing
organisms also include marine mammals. The main concern of bioaccumulation is that
concentrations in an organism reach levels that lead to adverse ef f ects, especially in apex
predators at the top of the f ood chain.

In a tiered approach, the f irst tier can be used f or de-/prioritisation, while the higher tiers
include a testing approach to allow a conclusion upon B and vB properties according to REACH
 Chapter R.11: PBT/vPvB assessment

123 Version 4.0 – December 2023

Annex XIII, 3.2.2 (b), and 3.2.2 (c), where possible expressed by biomagnif ication f actors or
trophic magnif ication f actors (REACH Annex XIII). Relevant assessment endpoints are the
biomagnif ication f actor (BMF), the whole-body total (or terminal) elimination rate and the
biotransformation rate. This guidance refers mainly to neutral organic substances; for ionisable
and surf ace-active substances, it may be possible to apply the same principles, taking into
account specif icities as described in Appendix R.7.10-3 in Chapter R.7c of the Guidance on
IR&CSA).

The BMF is the steady-state ratio between the chemical concentration in an organism (C org;
mg/kg) and its diet (Cdiet; mg/kg), but it can also be expressed using rate constants describing
competing rates of chemical uptake (k ) and elimination (k
dietary uptake ) in an organism:
elimination

The chemical uptake is the product of the absorption efficiency of the chemical f rom the f ood
(E ; unitless) and the ingestion rate (I; kg /k
diet /d). The BMF is inversely related to the
diet body weight

whole-body total (or terminal) elimination rate constant and is proportionally related to the
elimination half -lif e.

For neutral hydrophobic organic chemicals, the BMF can be expressed as the ratio of the lipid
normalised concentrations in the organism and in its diet f rom the same f ood web and same
time period, i.e., units kg-lipid/kg-lipid (Gobas et al., 2009; Mackay et al., 2013), if ref erring
to steady state conditions.

Normalisation of contaminant concentrations to the lipid content is usually applied to

concentrations along trophic levels in a f ood web, to assess the thermodynamic activity (or
f ugacity) of a chemical in a consumer organism compared to that in its diet (Debruyn and
Gobas 2006). Increase in concentrations along the f ood chain or f ood web occurs when a
chemical is taken up in the process of f ood digestion (food components are assimilated, which
is leading to a change in composition and reduction in volume, and the capacity of the f ood f or
the chemical reduces), and its elimination rate is slower than the uptake rate by f ood
ingestion. The elimination rate is inversely related to the lipid content of the organism apart
f rom confounding f actors such as metabolism, so the higher the lipid content, the lower the
kelimination. Lipid normalisation is relevant f or lipophilic organic substances in a f ood web that
partition to f atty tissue. Other substances such as PFAS may partition to other body
compartments, thus it has to be carefully considered if lipid normalisation is applicable ( Gobas
et al., 2015). If lipid normalisation is not considered appropriate, it should be examined
whether normalisation to another compartment, e.g. proteins is applicable or if non-normalised
data are pref erred.

The discussion paper “Bioaccumulation assessment of air-breathing mammals” available at the

ECHA website (ECHA Working group on Toxicokinetics, 2022) gives details on the scientif ic
background.

Tier 1: Screening assessment:

For the assessment of bioaccumulation in mammals and other air -breathing organisms at
screening level, the f ollowing properties of a substance are assessed:

• Volatility; chemicals that are sufficiently volatile will be readily eliminated by exhalation.
 Chapter R.11: PBT/vPvB assessment

124 Version 4.0 – December 2023

• Hydrophilicity; chemicals that are sufficiently water soluble will be readily eliminated by
urinary excretion.

These properties can be estimated or measured and compared with threshold values, i.e. log
Kow > 2 and logKoa > 5, that are meant to separate chemicals that are potentially
bioaccumulative f rom those that clearly will not be able to bioaccumulate.

As screening criteria f or substances that might bioaccumulate or biomagnif y in air -breathing

organisms, a combination of the octanol-water partition coefficient Kow and octanol-air partition
coef f icient Koa have been identif ied (Gobas et al., 2003). Baskaran et al. (2021a) have
compiled all Koa values reported in the published literature. Their dataset includes more than
2500 experimentally derived values and more than 10 000 estimated values f or K oa, in total
covering over 1500 distinct molecules. Furthermore, the ability of f ive techniques, specif ically
polyparameter linear f ree energy relationships (ppLFERs) with either experimental or predicted
solute descriptors, EPI Suite's KOAWIN, COSMOtherm, and OPERA, to predict the K oa of
organic substances, was assessed by comparison with available mea sured Koa values. The
ppLFER equation using experimental solute descriptors predicted the K oa the best and was
described by Baskaran et al. (2021b) as the pref erred method. The ppLFERs can be obtained
f rom the UFZ-LSER data base (Ulrich et al., 2017), and used f or the estimation of K oa. Koa can
f urthermore be calculated based on the inf ormation available in the registration dossier: K ow
and Henry’s Law Constant (H) (Meylan and Howard, 2005). In case H is also unavailable, H
can be estimated based on water solubility (WS), vapour pressure (VP), and molecular weight
(MW) (see equation R.16-4 of Chapter R.16 in the Guidance on IR&CSA). Sander (2015)
published a compilation of 17350 Henry's law constants f or 4632 organic and inorganic species
48
in water, collected f rom 689 ref erences, with f urther inf ormation made available online .

An ef ficiently absorbed, non-biotransf ormed neutral organic substance with a log K oa ≥ 5 in

combination with a log Kow ≥ 2 has the potential to biomagnif y in vertebrates of the terrestrial
f ood chains and air-breathing marine wildlif e as well as in humans, while the substances with
log Kow < 2 have a reduced gastrointestinal uptake or are ef f iciently excreted in urine , and
therefore do not biomagnif y even though their Koa is high (Armitage and Gobas, 2007; Kelly et
al., 2007; Gobas et al., 2009; McLachlan et al., 2011; Goss et al., 2013).

The precise values f or the K ow and Koa values indicated in the ITS are a f unction of the
modelled organisms, f ood webs and environments used to obtain these values (e.g., Kelly et
al., 2007; Armitage and Gobas, 2007). It should be f urther noted that these screening criteria
are only applicable to neutral organic substances that primarily partition into lipids. For other
substances, log Koa and log K ow are not suitable predictors f or the distribution coef f icients
between organisms and the environment. In such cases, it should be investigated whether
there is a concern f or bioaccumulation. In some cases, e.g., f or ionogenic substances, the use
of artif icial liposome-water partition coef f icients could give additional inf ormation on the
bioaccumulation potential.

Tier 2: Intermediate assessment:

A screening based solely on K ow and Koa would identif y a large proportion of substances as
potentially bioaccumulative, because the f raction of chemicals that can be judged as

48
Available under: [Link] (last accessed: 28 October 2022).
 Chapter R.11: PBT/vPvB assessment

125 Version 4.0 – December 2023

suf f iciently volatile and/or suf f iciently water soluble f or rapid respiratory and urinary
elimination based on the partitioning properties predicted f or their neutral f orm is relatively
small. On the other hand, substances that are biotransformed rapidly will be readily eliminated
and therefore not bioaccumulate (Wania et al., 2021). Note that this assumption is only true
f or the majority of substances where the metabolites are more water soluble and less
bioaccumulative than the parent. For some substances, the metabolite may be mor e persistent
and more bioaccumulative than the parent, e.g., DDT and its metabolite DDE. Thus, the
transf ormation/degradation products should be considered separately in the assessment, in
agreement with the general procedure f or the whole PBT assessment.

Biotransf ormation rates can be estimated f rom in vivo studies, in vitro bioassays and by
quantitative structure activity relationships (QSARs). Bef ore perf orming animal testing, a
search f or existing reliable quality measured in vivo data should be conducted. If such data are
available this could f acilitate a def initive assessment (tier 3) to conclude whether a substance
is bioaccumulative in air-breathing organisms or not. Relevant databases of measured
49
toxicokinetic (TK) data in air-breathing organisms include the OECD QSAR Toolbox , EAS-E
50 51
Suite and MamTKDB . The OECD QSAR Toolbox presently contains a database with whole
body terminal elimination half -lif e (HLT) data and will soon be expanded with HLT data f rom
animals through addition of MamTKDB data. It is also possible to search publications f or TK
data f or structurally similar substances. Existing rat or other mammalian absorption,
distribution, metabolism and excretion (ADME) / TK data should be evaluated f or relevance
and reliability. Empirical in vivo HLT data have been collected f or several hundred chemicals in
humans (Arnot et al., 2014, Obach et al., 2008, Lombardo et al., 2018), other mammalian
species (Hof er et al., 2021), and birds (Kuo et al., 2022). Databases to obtain in vivo HLT data
parameters are also emerging, e.g., [Link] and the Mammalian toxicokinetic
database (MamTKDB).

QSAR models (need for further development)

QSAR models f or the estimation of biotransformation rates of substances in humans have been
developed by Arnot et al. (2014) and Papa et al. (2018), f urther models are under
development. More good quality inf ormation on biotransf ormation rates of substance s in
terrestrial biota is needed to f urther develop and improve the predictability of QSAR models
(Gobas et al., 2023). Hence, QSAR predictions can be used f or prioritisation, but cannot be
recommended yet to be used as stand-alone inf ormation to justif y that there is no
bioaccumulation concern f or air-breathers. Future improvements of the proposed procedure
depend largely on the development and validation of prediction methods f or the
biotransf ormation kinetics in air-breathing organisms and f or the potential f or renal
reabsorption (Wania et al., 2021).

49 [Link], last accessed: September 2022

50
[Link], last accessed: September 2022

51
Mammalian toxicokinetic database (MamTKDB) 1.0 - Data Europa EU, last accessed: September 2022
 Chapter R.11: PBT/vPvB assessment

126 Version 4.0 – December 2023

In-vitro testing (need for further development)

While an OECD test guideline is available f or f ish hepatocytes and S9 mix (OECD TG 319A and
B), this is not the case f or mammalian cells yet, although similar tests have been in use by
industry f or other purposes. The use of mammalian in vitro tests f or the estimation of
biotransf ormation rates that could then be extrapolated to an in vivo half -lif e f or
biotransformation, which can be used in bioaccumulation assessment, is a promising pathway
to be developed further (Goss et al., 2018). IVIVE (in vitro-in vivo extrapolation) methods are
available and may need to be/have been adapted to air -breathing organisms (see discussion
paper “Bioaccumulation assessment of air-breathing mammals” (ECHA Working group on
Toxicokinetics, 2022) available at the ECHA website and CEFIC LRi ECO41: Enhanced screening
methods to determine bioaccumulatin potential of chemicals in air -breathing species). Since
current in vitro tests f or bioaccumulation assessment measure depletion of the parent
substance only, metabolites are not usually identif ied or quantif ied. Metabolites can be taken
into account by additional experimental measurements of their concentrations in the in vitro
tests. Also, since very hydrophobic substances may sorb to any surf aces in the test vials or
volatilise f rom the test system, it must be ensured that losses due to sorption or volatilisation
are not erroneously attributed to biotransf ormation.

Tier 3: Def initive assessment: In vivo testing

Pref erred options f or testing

If the screening assessment and/or intermediate tier assessment point to possible

bioaccumulation in air-breathing organisms, and the substance cannot be concluded to be B or
vB through a bioaccumulation assessment using aquatic organisms, an in vivo test may have
to be performed. The preferred test guideline would be the OECD 417 on Toxicokinetics, and
the pref erred organisms the rat. If no in vivo test has been perf ormed yet, the f ollowing
considerations should be taken into account f or the test to best serve bioaccumulation
assessment:

TG 417 of fers quite some flexibility in study design to accommodate f or dif f erent regulatory
needs, but it does not include guidance on how to assess accumulation. Several f actors will
inf luence the clearance rate (or the corresponding elimination half -lif e) thus it is not a f ixed
value but relates to the test conditions, rat strain, animal age (f at content), etc. In repeated
daily administration studies, clearance rates are pref erably measured af ter steady state
conditions have been reached, when the administration is stopped. The time to establish a
steady state will dif f er depending on substance and dose. Repeated (compared to single)
dosing should better ascertain a high radiolabelled substance load into peripheral organ/tissue
compartments and establishment of steady state. This is because some large and/or deep
organs or tissues may have slow inf lux rates due to little blood perf usion, unf avourable
partitioning, low active or passive transport through the cell membrane or else. So-called
preconditioning studies (repeated dosing with unlabelled substance f ollowed by a single
radiolabelled dose the last day (TG 417 §57) to investigate enzyme induction/inhibition,
appear not appropriate f or bioaccumulation assessment since the last administered
radiolabelled dose (measured) will not be present at steady state conditions, and be small in
comparison to repeated administration using a radiolabelled substance (Hof er et al., 2021).

The terminal half -lif e is the time required f or the concentration to f all by 50% during the
terminal phase studied. A f ield BMF of 1 can be translated into a whole-body, terminal
 Chapter R.11: PBT/vPvB assessment

127 Version 4.0 – December 2023

elimination half -lif e of about 4 days in rat, and/or about 50 days in humans(for background on
derivation of these thresholds see ECHA Working group on Toxicokinetics (2022)). If the
terminal elimination half -lives are assessed to be longer than these, taking into account the
considerations described above, then this is an indication that the substance has vB properties.
Tissue, organ, or body f luid specific elimination half -lives may be shorter than the whole body
terminal elimination half -lif e and therefore should be interpreted with care. Measurements of
declining concentrations in organs/tissues are of ten more relevant than in blood
plasma/serum, which of ten underrepresents elimination half -lives in organs/tissues (Hof er
[Link]. 2021). Elimination in blood is relevant f or substances with a high blood distribution such
as PFAS. If whole-body terminal elimination half -lives are between 2.5 and 4 days in rat,
and/or 20 and 50 days in human, the assessment of the B property should be accompanied by
a T assessment (PBT concern). It is noted that the derived elimination half -lif e thresholds f or
rat and human are tentative. There may be exceptional cases where the derived elimination
half -lif e threshold values in rats or humans cannot be used as an indicator of vB, f or example
where there is very low dietary absorption ef f iciency. Such cases require an individual
assessment to determine whether the substance is vB or not.

The discussion paper “Bioaccumulation assessment of air-breathing mammals” which is

available at the ECHA website was developed by the ECHA Working group on Toxicokinetics
(2022), describes and discusses the tiered approach, scientif ic developments and
recommendations f or f urther work, and provides background inf ormation to the guidance given
in this section.

The use of toxicokinetic data in B-assessment is under scientif ic development and the
recommendations above are based on current knowledge and experience. Registrants are
advised to f ollow-up recent and f uture developments in the f ield, e.g. via the ECHA website.

R.[Link].9 Other testing data

In the f ollowing section other testing inf ormation which may be relevant f or the
bioaccumulation assessment is discussed. It should be noted f rom the outset that this other
inf ormation does not override valid inf ormation on aquatic bioaccumulation of the substance if
the aquatic data indicate high bioaccumulation potential.

Chronic toxicity studies with mammals

If chronic toxicity studies with mammals are available, the complete absence of any ef f ects in
the long-term is an indication that the substance is either chronically non-toxic and/or that it is
not taken up to a signif icant extent. Although this is only indirect inf ormation on the uptake of
a substance, it may be used together with other indicators, e.g. ref erring to non-testing
inf ormation, to conclude in a Weight-of-Evidence approach that a substance is likely to be not
bioaccumulative.

Particular attention should be drawn to the toxicokinetic studies considered to be included in

the PBT/vPvB-assessment. For f urther inf ormation, see Sections R.7.10.14 and R.7.12 in
Chapter R.7c of the Guidance on IR&CSA. In f uture assessments it might be advisable to
combine a toxicokinetic assessment with toxicity testing if there is a concern f or
bioaccumulation.
 Chapter R.11: PBT/vPvB assessment

128 Version 4.0 – December 2023

R.[Link].10 Further data

In this section, several types of non-animal data are discussed that can be used in a Weight-
of-Evidence approach f or the B and vB assessment. If average molecular size, log Kow , and
octanol solubility are above or below certain values (as described below), they may indicate a
52
limited bioaccumulation potential due to the lack of uptake . However, these parameters
should never be used on their own to conclude that a substance is not bioaccumulative. The
inf ormation f rom these parameters should be accompanied by other inf ormation confirming the
low uptake of the substance in living organisms, e.g. by read-across with similar substances,
absence of toxicity or lack of uptake in toxicokinetic studies with mammals. Evidence of
signif icant uptake in f ish or mammals af ter long-term exposure implies that the indicators
above will likely underestimate the real bioaccumulative potential of the substance and thus
these indicator values should be considered unreliable f or assessing the bioaccumulation
potential.

Some studies have proposed a reduced uptake based on experimental bioconcentration

studies. The reduced uptake then usually refers to reduced uptake via the f ish gills. This does
not imply that there will be reduced or no uptake possible via the gut, i.e. f rom f ood, where
other uptake mechanisms may play a role (Larisch and Goss, 2018; Li et al., 2014). The extent
to which those additional uptake mechanisms play a role in bioaccumulation, however, is
inadequately quantif ied f or f ish and aquatic invertebrates. There is evidence, however, f or
certain highly persistent and super hydrophobic substances, that signif icant accumulation via
the f ood chain takes place (e.g. chlorinated paraf f ins (Ding et al., 2021, Zhou et al., 2020),
chlorinated f lame retardants (Wu et al., 2010)).

Other methods such as in vitro methods or biomimetic extraction procedures may also be
usef ul and are mentioned brief ly at the end of the section.

(Q)SAR models
BCF-QSARs and other computer models may be used to address aquatic bioconcentration,
provided that the model is appropriate f or the chemical class (see Section R.[Link] in
Chapter R.7c of the Guidance on IR&CSA and in Appendix R.11-1.1 and Appendix R.11-1.2 of
this guidance document). For very hydrophobic substances (substances with logK ow above 5-
6), f ish BCF predictions may not be indicative of low bioaccumulation potential due to a risk of
underestimation, e.g. when predicted BCF is not based on f reely dissolved chemical
concentration in water (Glüge, et al. 2022, Böhm et al. 2016, Ehrlich et al., 2011).

Read-across with other substances

If a valid and reliable BCF value f or a structurally closely-related substance is available, read-
across can be applied. When applying read-across data in bioaccumulation assessment, two

52
The original document on parameters indicating hindered uptake was drafted as part of an ECETOC report on the
use of alternatives in assessing the environmental safety of substances (ECETOC, 2005). A revised document was
included in the R.11 guidance published in 2017. For the current version, the molecular weight parameter has been
removed as indicator. Relevant information from Appendix R.11—1: Indicators for limited bioconcentration for PBT
assessment has been integrated into the current version, and the Appendix 1 as such has been removed.
 Chapter R.11: PBT/vPvB assessment

129 Version 4.0 – December 2023

generally important aspects have to be considered in addition to the normal criteria of read-
across applicability: hydrophobicity and the likelihood f or metabolisation of both substances. ,

Molecular size
Inf ormation on molecular size can be an indicator to strengthen the evidence f or a limited
bioaccumulation potential of a substance. One parameter f or molecular size is the maximum
molecular length of a substance, which is def ined as the diameter of the smallest sphere into
which the molecule would reside, as written, i.e. not accounting f or conformers. From a certain
minimum length upwards it may be assumed that the substance disturbs the structure of the
lipid bilayer of cell membranes and therefore does not accumulate to a signif icant extent, i.e.
has a BCF value lower than 2000. Folding of long linear structures may alter the ef f ective
length of the molecule of the substance, which renders it more easily transf erable across cell
membranes. Therefore, the criterion f or molecular length should only be used in a Weight-of-
Evidence approach together with other inf ormation as described under "c onclusion on the
endpoint". In conclusion, an assessor may justif y that, in certain cases when inf ormation on
the ef f ective length and other inf ormation indicating a low bioaccumulation potential is
available, the criterion f or B and hence also f or vB is not met. It is noted that currently there is
no agreed cut-off criterion f or molecular length and theref ore the use of molecular length as
one indicator of low bioaccumulation potential needs to be well justif ied. An earlier threshold of
a maximum molecular length (MML) of greater than 4.3 nm as indicator f or hindered uptake
was based on a small dataset and cannot be recommended in this Guidance as agreed by the
Partner Expert Group consulted during the f irst revision of this Guidance (v2.0 – Nov 2014).

A parameter that directly reflects the molecular size of a substance is the average maximum
diameter (Dmax aver). Dmaxaver is def ined as the diameter of the smallest sphere into which the
molecule may be placed. Of ten this will be the same as the maximum molecular length,
especially f or rigid molecules. However, when f lexible molecules are assessed, energetically
reasonable conformers could be present for which these parameters might dif fer substantially.
In this document the average value f or this Dmax f or “energetically stable” conformers is used,
i.e. Dmaxaver. Very bulky molecules will less easily pass through the cell membranes. This
results in a reduced BCF of the substance. Consistently with this notion, one study of a diverse
53
set of substances showed that f or substances with a Dmax aver larger than 1.7 nm the BCF
value will be less than 2000 (see Environment Agency, 2009)). However, the applicability of a
numeric cut-off should be considered on a case-by-case basis. Also, it should be noted that the
estimate of molecular size depends on the substance conformation as well as the method used.
More inf ormation on the background of the molecular size indicator is given in ECETOC (2005).

The def initions of maximum molecular length and Dmaxaver are shown graphically in Appendix
R.11-1.2 together with examples of sof tware that may be used f or their calculations.

Log Kow
For the B/vB assessment f or aquatic organism, a screening threshold value has been
established, which is log Kow greater than 4.5. The assumption behind this is that the uptake of
an organic substance in aquatic organisms is driven by its hydrophobicity. For organic

53
Please note that the indicator value of 1.7 nm for the average maximum diameter was derived using the descriptor
Dmax from OASIS. However, it appears from the Environment Agency (2009) that the use of different software tools
could lead to variable results for the same substance.
 Chapter R.11: PBT/vPvB assessment

130 Version 4.0 – December 2023

substances with a log Kow value below 4.5 it is assumed that the B criterion, i.e. a BCF value of
2000 (based on wet weight of the organism, which ref ers to f ish in most cases) , is not
exceeded.

At log Kow values between 4 and 5, Log BCF increases linearly with log Kow , if the substance is
absorbed at the same rate and it is not biotransformed. This linear relationship is the basis f or
the B screening threshold value of log Kow > 4.5. However, at very high log Kow (>6), the
relationship between these two parameters becomes progressively non-linear (Nendza, 1991).
Apart f rom experimental errors in the determination of BCF values f or these very hydrophobic
substances, reduced uptake due to the increasing molecular size may play a role as well.
Moreover, the experimental determination of log Kow f or very hydrophobic substances is
normally also very uncertain due to experimental dif f iculties. Reverse phase HPLC method
enables partition coefficients to be estimated in the log Kow range between 0 and 6, but can
be expanded to cover the log Kow range between 6 and 10 in exceptional cases (OECD,
2022a). It is advised to always generate QSAR estimations of log Kow together with HPLC
generated estimates, especially if the HPLC generated estimate of log Kow is in the range of
one log unit below or above the screening value of log Kow = 4.5. Especially ionising
substances, salts, metal complexes and complex mixtures would benef it f rom additional
estimated data (next to HPLC generated data) f rom a variety of log Kow QSAR models and a
Weight-of -Evidence evaluation (Appendix R.11-5). The reliability of measured and modelled
log Kow values > about 8 is of ten lower than the reliability of measured and modelled log Kow
values < about 8. Ideally the results of several model predictions f or log Kow should be
considered.

The aquatic BCF of a substance is probably lower than 2000 if the calculated log K ow is higher
than 10. Given that none of the models have experimental inf ormation in this range, more
than one model should be used to estimate the log Kow value and the results evaluated by
expert judgement. If a log Kow value indicates that the substance screens as B/vB, but a
registrant concludes it is not B/vB based on other data (see R.[Link].10 “Further data”),
there should be specif ic reference to the REACH guidance indicating how such a conclusion was
drawn. It should be noted that neither a high K oc value nor low water solubility value can be
used to argue that a substance lacks signif icant bioaccumulation potential. Instead these
properties may inf luence the f orm of PBT testing required.

Octanol solubility
Octanol is of ten used as a surrogate f or fish lipids. With a low solubility in octanol, the Log Kow
and hence the BCF can be either high or low, depending on the water solubility of the
substance. Therefore, the solubility in n-octanol is not a parameter that is directly related to
the BCF value. However, if the solubility of a substance in octanol is so low that the maximum
concentration levels that can be attained in organisms do not reach levels suf f icient to elicit
any toxic ef f ects, it can be reasoned that such accumulation would not be of concern. The
concentration of a substance at which the occurrence of toxic effects normally can be excluded
is 0.002 mmol/L in n-octanol (measured octanol solubility (mg/L) < 0.002 mmol/L × MW
(g/mol)). Furthermore, octanol solubility is only an indicator f or substances accumulating in
f atty tissues and certain substances may bind to proteins instead of partition into lipids.
Finally, inf ormation on octanol solubility should in particular be accompanied and
complemented by inf ormation on mammalian toxicity or toxicokinetics to conf irm the absence
of uptake and/or chronic toxicity. More inf ormation on the background to this indicator is given
in ECETOC (2005).
 Chapter R.11: PBT/vPvB assessment

131 Version 4.0 – December 2023

Biomimetic extraction procedures

Biomimetic extraction procedures with semi-permeable membrane devices (SPMD) and solid
phase micro extraction (SPME) are used to mimic the way organisms extract substances f rom
water. These types of methods are at the moment only well described f or hydrophobic
substances. For more detailed inf ormation, see Section R.[Link] in Chapter R.7c of the
Guidance on IR&CSA.

R.[Link].11 Use of a fugacity approach for bioaccumulation assessment

The use of f ugacity ratios has been proposed as method to convert laboratory and f ield
bioaccumulation metrics into a common f ugacity ratio scale to f acilitate the interpretation of
dif f erent sources of bioaccumulation data (Burkhard et al., 2012, Mackay et al.,2013, Armitage
et al., 2021). The ratio between the f ugacity in the organism divided by the f ugacity in the diet
expresses the increase in f ugacity from f ood to organism. Mackay et al. (2013) proposed the
BMF as the lipid-normalized ratio of the concentration in the predator to that of the diet with
BMFs exceeding 1.0 indicating an increase in lipid concentration and thus also an increase in
thermodynamic potential or f ugacity with ascending position in the f ood chain or f ood web.
This approach converts a wet weight based BMF, which can be def ined by the concentration in
the organism divided by the concentration in the diet, into a f ugacity based BMF f or lipid
sorbing substances.

The calculation of a f ugacity ratio is an approximation based on certain assumptions. One of

the assumptions of ten made is that the partitioning to lipids is equal to the octanol-water
partitioning and this may not always be the case. The f ugacity capacity of an organism can be
derived f rom partition coefficients that are not estimated f rom K ow (f rom ppLFERs). Fugacity
capacity can also be calculated based on partitioning dif f erent tissue components (storage
lipid, membrane lipid, albumin, structural protein) if the partition coef f icients f or these
components are known. An example of this approach was recently described by Fremlin et al.
(2021).

However, there is a lack of agreement on how to interpret f ugacity ratios and the method ha s
not yet been validated suf ficiently, f or example with existing POP and PBT substances. That is
why, the f ugacity approach in bioaccumulation assessment under REACH cannot be
recommended at this stage.

Apart f rom these considerations, it must be realised that the use of f ugacity ratios is only
justif ied in cases of thermodynamic equilibrium between the dif f erent compartments that an
organism is exposed to. When applied to f ield studies, this is seldom the case. If f or example a
ratio between biota and sediment is used as basis f or the f ugacity ratio the assessment might
be strongly hampered by strong sorption to the sediment and consequently very slow
depuration of the substance f rom the sediment into (pore-)water. In such cases, which f or
example could be expected for many well-known PBT substances, the f ugacity ratio between
biota and sediment will be low, while the f ugacity ratio between biota and the depleted pore -
water could be high. However, also in laboratory studies, thermodynamic equilibrium between
dif f erent exposure media (water and f ood) is even prevented. In both the aqueous and dietary
OECD TG 305 studies, f ish are exposed to only one exposure route, either water or diet. The
consequence is that the remaining medium to which f ish are ex posed simultaneously have
arbitrarily a very low f ugacity compared to f ish and the exposure medium. Further, variable
results in laboratory settings were obtained when normalising dif ferent f eeding regimes to the
 Chapter R.11: PBT/vPvB assessment

132 Version 4.0 – December 2023

lipid content of the f ood (Hashizume et al., 2018). It appeared that the f ugacity in an organism
is rather dependent of the dose delivered with the f ood than of the f ugacity in the f ood itself .
This shows that f or the same organism, f ugacity increase is not a constant f actor .

The f ugacity ratio only considers a substance of concern f or bioaccumulation if there is an

increase in f ugacity, i.e. biomagnification occurs. Indeed, if biomagnifications is conf irmed this
is a clear indication of bioaccumulative properties of a substance (Gobas et al., 2009).
Nevertheless, the bioaccumulative properties of substances that do not biomagnif y could be
considered of concern as well. Polycyclic aromatic hydrocarbons (PAHs) could be considered as
an example of this concern. These substances are very ef f iciently tak en up in invertebrates
with very high bioaccumulation f actors. However, they are not biomagnif ied in higher trophic
levels, such as f ish. Still, the additional uptake due to the consumption of high concentrations
in invertebrates can lead to signif icantly higher bioaccumulation f actors in the f ield (Khairy et
al., 2014) than would be predicted based on laboratory bioconcentration data. This example
illustrates that high bioaccumulation in a part of the f ood chain may have unpredictable effects
throughout other parts of the f ood chain as well.

Even though the f ugacity approach in bioaccumulation assessment under REACH cannot be
recommended at this stage, it is noted that the approach allows various lines of evidence to be
put into a consistent f ramework to apply a quantitative Weight-of-Evidence determination as to
whether or not a substance biomagnif ies.

R.[Link] Toxicity assessment (T)

R.[Link].1 Integrated testing and assessment strategy (ITS) for T-testing in

support of PBT assessment for the aquatic environment

In this section guidance on the recommended testing and assessment strategy is provided as
an annotated f low chart (Figure R.11—6). The strategy is based on the T criteria (Table R.11—
1), which state that the T criterion is f ulf illed if at least one of the data types listed in the
criteria is f ulf illed. If P and B criteria are f ulfilled, inf ormation would need to be generated until
f or each (eco)toxicity data type it is clear whether the criterion is f ulf illed or not.
 Chapter R.11: PBT/vPvB assessment

133 Version 4.0 – December 2023

Start

Step 1
Classified H350,
yes H340, H372, H373,
T H350i, H360, H361 or
Acute data E(L)C5 0 <
0.01 mg/L

no

Step 2
Acute data yes
E(L)C50 Potentially T
< 0.1 mg/L ?

no
Step 5

yes
Carry out screening
No further assessment for P
assessment and B: Log Kow <
necessary 4.5 or readily
biodegradable ?

no

Step 6 Step 3

yes Other convincing no no No further

P and B
Not T evidence chronic assessment
T > 0.01 mg/L ? confirmed ?
necessary

Step 4 yes
Chronic T studies (order and
selection of test organisms,
see main text)

no NOEC or EC10
Not T
< 0.01 mg/L ?

yes

PBT

Figure R.11—6: T testing in support of PBT assessment for the aquatic environment.
 Chapter R.11: PBT/vPvB assessment

134 Version 4.0 – December 2023

According to Article 14 of REACH, PBT assessment is required f or substances registered at

quantities of ≥ 10 t/y (it is assumed that at least acute algae, daphnia and f ish data are
available):

Step 1: Assessment of mammalian toxicity data and acute aquatic toxicity data;
• IF classif ied or likely to be classif ied as carcinogenic (cat. 1A or 1B), germ cell
mutagenic (cat. 1 or 1B) or toxic to reproduction (class 1A, 1B or 2) or STOT RE 1,
STOT RE 2 or any EC50 or LC50 < 0.01 mg/L, THEN def ine the substance as T and stop
assessment

• IF not classif ied or likely to be classif ied as carcinogenic (cat. 1A or 1B), germ cell
mutagenic (cat. 1A or 1B) or toxic to reproduction (cat. 1A, 1B or 2) or STOT RE 1, or
STOT RE 2 or any EC50 or LC50 ≥ 0.01 mg/L, THEN move to step 2.

Step 2: Assessment of acute aquatic toxicity data;

• IF any EC50 or LC50 < 0.1 mg/L, THEN the substance is a Potential T candidate. Move to
step 3.

• IF all EC50 or LC50 ≥ 0.1 mg/L, THEN it needs to be conf irmed that this is not a f alse
negative (i.e. a substance with possibly a high chronic toxicity). Move to step 5.

Step 3: Consider outcome of P and B assessment* (Note.: it is considered good practice to

assess P, B and T in that order)
• IF P and B conf irmed, THEN proceed to Step 4 (chronic T testing) **

• IF conf irmed not P or not B, THEN STOP

Step 4: Chronic T testing (on f ish, daphnids, algae). The approach here is that chronic aquatic
toxicity testing should be f irstly carried out on non-vertebrate species, unless there
are indications that f ish is the most sensitive group (NB: it is not def ined in this ITS
how to rank the sensitivities). If the T-criterion is f ulf illed by the chronic algae or
Daphnia data, a chronic f ish test is not necessary f or PBT assessment, even if it would
be the most sensitive species.
• IF NOEC or EC10 < 0.01 mg/L, THEN PBT conf irmed

• IF NOEC or EC10 ≥ 0.01 mg/L, THEN not T, and STOP

Step 5: Screening of the substance f or P and B *

• IF Log Kow ≤ 4.5*** or other B-cut-of f criteria met, and no other indications are
available that the substance might bioaccumulate in other ways than by absorption to
lipids, then not B and STOP.

• IF substance is readily biodegradable, then not P and STOP

• IF Log Kow > 4.5 AND not readily biodegradable, THEN move to step 6

Step 6: Other long term T-evidence (e.g. by means of read across and Weight-of-Evidence or
group approach)
• IF chronic toxicity cannot be excluded, THEN move to step 3 (P & B conf irmation)
 Chapter R.11: PBT/vPvB assessment

135 Version 4.0 – December 2023

• IF strong evidence f or non-T properties, THEN STOP.

* For specific guidance on the identification of P & B substances, please refer to Section R.[Link] for
persistence and Section R.[Link] for bioaccumulation

** If B is likely but vB is not and a reliable BCF is not available, consider conducting tests on
invertebrates to check the T status for these organisms before co nsidering tests on fish (either for
chronic toxicity or for obtaining a BCF).

*** Care must be taken in case a substance is known to bioaccumulate by a mechanism other than
passive diffusion driven by hydrophobicity; e.g. specific binding to proteins instead of lipids might
result in an erroneously low bioaccumulation potential if it is estimated from Log K ow .

Care must also be taken for substances classified as polar non-volatiles (with low Log K ow and high
Log K oa). This group of substances has a low bioaccumulation potential in aquatic organisms but a high
bioaccumulation potential in air-breathing organisms (unless they are rapidly metabolised).

R.[Link].2 The toxicity criterion

According to Section 1.1.3 of Annex XIII to REACH, a substance is considered to f ulf il the
54
toxicity criterion (T) when :

• the long-term no-observed ef f ect concentration (NOEC) or EC10 f or marine or

f reshwater organisms is less than 0.01 mg/L; or

• the substance meets the criteria for classification as carcinogenic (category 1A or 1B),
germ cell mutagenic (category 1A or 1B), or toxic f or reproduction (category 1A, 1B or
2) according to the CLP Regulation; or

• there is other evidence of chronic toxicity, as identif ied by the substance meeting the
criteria f or classif ication: STOT RE 1, or STOT RE 2 according to the CLP Regulation.

For the assessment of aquatic toxicity, EC10 values are preferred compared to NOEC values for
55
deriving long-term toxicity to marine or f reshwater organisms .

The evidence of CMR and chronic toxicity specif ied above does not only ref er to substances
that are already classified accordingly (i.e. CLP hazard statements H350, H340, H372, H373,
56
H350i, H360 and H361 ) but also implies an obligation to check whether the criteria f or
assigning the respective classifications are f ulfilled in accordance with the provisions of Annex I
to REACH (Section 1.3 Step 3: Classification and Labelling). If any classif ication criterion
leading to the assignment of the mentioned classif ications is met, the substance f ulf ils the T
criterion and there is no need to perf orm any f urther aquatic studies f or T assessment. If data
are available f or birds these cannot be directly (numerically) compared with the T criterion

54
Annex I (Part 4) to CLP Regulation (EC) No 1272/2008 was amended to include Classification criteria for PBT. These
criteria include a possibility to conclude a substance as T based on classification as endocrine disruptor (category 1) for
humans or the environment ([Link]
55
An OECD workshop (OECD, 1998) recommended that the NOEC should be phased out from international standard.
Indeed, concerns were expressed about deciding to abandon the NOEC since it may not be sufficiently protective
because of the danger of false negatives. According to the Report of the OECD Workshop on Statistical Analysis of
Aquatic Toxicity Data (OECD, 1998), NOECs are leading to misunderstandings, misinterpretations and NOECs are
statistically unfounded.
56
H360 and H361 here include also all the possible combinations (e.g H360F, H360FD, etc).
 Chapter R.11: PBT/vPvB assessment

136 Version 4.0 – December 2023

(see Section 1.1.3 to Annex XIII). However, reprotoxicity studies or other chronic data on
birds, if they exist, should be used in conjunction with other evidence of toxicity as part of a
Weight-of-Evidence determination to conclude on the substance toxicity (a NOEC of  30
mg/kg f ood in a long term bird study should in this context be considered as strong indicator
f or f ulf illing the T criterion).

The rest of this document is limited to testing of the T criterion on the basis of evidence f rom
aquatic tests.

Due to animal welf are concerns, the general scheme of testing is sequentially f irst P, B and
then T if there are no specif ic reasons f or deviation f rom that sequence. Furthermore,
vertebrate animal testing should be generally minimised by f irst testing non-vertebrate species
as data f rom invertebrates are equivalent to vertebrate data in the context of the PBT/vPvB-
assessment. For determination of whether a substance f ulf ils the criteria f or aquatic toxicity,
and in the absence of any long-term ecotoxicity data on aquatic species, a 21-d Daphnia
reproduction test (OECD TG 211) would normally be the pref erred test to perf orm with the f ew
exceptions described later in this section where the results f rom short-term tests can already
lead to concluding that the criteria are f ulfilled. Under most circumstances, the T criterion of
0.01 mg/L (NOEC or EC10) can be compared to results f rom tests listed in REACH annexes VII
to X. Existing data f rom other equivalent test methods must be assessed on a case by case
basis based on the recommendations described in the ef f ects assessment methodology.

As the aquatic T criterion is based on a NOEC or EC10 f or pelagic organisms, the standardised
chronic tests on f ish, daphnids and algae are pref erred to assess the NOEC or EC10. However,
f or poorly water-soluble substances, the f easibility of perf orming a test via the water phase
needs to be considered carefully. Such a study may be technically dif f icult to perf orm as the
substance will partition out of solution, especially if it is known to partition strongly to
sediment and suspended solids. In such cases, it may be both impractical and uninf ormative to
test pelagic species via the water phase. Tests with sediment dwelling species may provide
more useful inf ormation on the toxicity of the substance in the compartment in which it will be
mainly f ound. However, the T criteria do not include a chronic value f or sediment as only NOEC
or EC10 values related to pelagic toxicity are accounted f or in Annex XIII. A possible way to
determine whether a substance has equivalent toxicity in sediment to that in the water column
could be to extrapolate the sediment toxicity value (e.g. NOEC) to a pelagic toxicity value by
assuming that sediment toxicity occurs mainly through the pore water and using the
equilibrium partitioning method (EPM). The EPM method is normally used to calculate a
PNECsediment f rom a pelagic PNECwater (see Section R.7.8 in Chapter R.7b of the Guidance on
IR&CSA).

However, the EPM method may also be used to back-calculate a NOEC or EC10 value of an
existing sediment test to a corresponding pelagic NOEC or EC10. The pelagic NOEC or EC10
derived can then be compared with the T criterion of 0.01 mg/L given in Annex XIII. The
sediment concentration equivalent to a pelagic NOEC or EC10 value of 0.01 mg/L increases
linearly with the suspended matter-water partitioning coefficient (see Section R.7.8 in Chapter
R.7b of the Guidance on IR&CSA).
 Chapter R.11: PBT/vPvB assessment

137 Version 4.0 – December 2023

To check whether the T criterion of 0.01 mg/L is f ulf illed, the equation f or the equilibrium
partitioning method used in order to calculate the PNEC sediment is slightly revised:
𝑵𝑶𝑬𝑪(𝑬𝑪𝟏𝟎)𝒔𝒆𝒅,,𝒅𝒘
𝑵𝑶𝑬𝑪(𝑬𝑪𝟏𝟎)𝒑𝒐𝒓𝒆𝒘𝒂𝒕𝒆𝒓 = Equation 11-1
𝑲𝒑 𝒔𝒖𝒔𝒑

NOEC(EC10)porewater (mg.L-1)
Kpsusp ([Link]-1 dw)
NOEC(EC10)sed dw ([Link] -1)

Kpsusp ([Link]-1 dw) can be estimated f rom the Koc of the substance as Kpsusp= Focsusp x Koc where
Focsusp is the mass f raction of organic carbon in dry suspended matter.

It should be noted that NOECsed derived f rom experimental studies are given in dry weight (as
mg/kg dw).

As the equilibrium between sediment and water is inf luenced by the suspended solid-water
partition coef f icient (Kpsusp), it is necessary to calculate the T criterion f or each substance,
using its own partitioning coef f icient.

For substances with water solubility below 0.01 mg/L, a chronic limit test (Csed,lim) can be
perf ormed at the spiked sediment concentration that is calculated to be at equilibrium with the
water solubility limit of the test substance.

C sed,lim = C [Link] susp Equation 11-2

Cwatersol (mg.L-1)
Kpsusp ([Link]-1 dw)
Csed,lim ([Link]-1 dw)

If no chronic ef fects are f ound f rom this limit test, the result can be regarded as experimental
evidence that the substance does not meet the pelagic T criterion f or invertebrates provided
that the equilibrium partitioning theory holds in the particular case (f or guidance on the
limitations of the equilibrium partitioning method, see Section R.[Link] in Chapter R.7b of
the Guidance on IR&CSA). However no f inal conclusion on pelagic toxicity can be drawn if no
f urther reliable toxicity data on f ish and algae are available. If chronic ef f ects are f ound then
this is an indicator that T could be met in a pelagic test and consideration should be given to
f urther testing (although care has to be taken at high spiking concentrations that the test
substance does not cause indirect ef f ects, e.g. by oxygen depletion as a result of
biodegradation).

R.[Link].3 Use of QSAR data

Only a f ew QSAR models predicting chronic aquatic toxicity are available but f urther research
on the QSAR prediction of chronic toxicity may increase their predictive capacities. Therefore at
the current state of the art, QSAR models generally seem not to be applicable f or an
unequivocal assessment of the T criterion. However, it should be noted that the registrant is,
within the f rame of Annex XI to REACH, allowed to make use of QSARs when they are
applicable.
 Chapter R.11: PBT/vPvB assessment

138 Version 4.0 – December 2023

R.[Link].4 Screening information and screening threshold values

If only screening inf ormation is available f or the PBT/vPvB assessment, screening criteria listed
in Table R.11—7 can be used f or screening. It should be noted that these criteria are indicative
and f urther description on the application of these criteria is provided below.

Table R.11—7: Screening threshold values for toxicity.

Screening information*** Conclusion

Toxicity
Short-term aquatic toxicity EC50 or LC50 < 0.01 mg/L**** T, criterion considered to be
(algae, daphnia, fish)* definitely fulfilled
Short-term aquatic toxicity EC50 or LC50 < 0.1 mg/L**** Potentially T
(algae, daphnia, fish)**

* From acute tests.

** From acute tests or valid/applicable QSARs.
*** The screening assignments should always be considered together for P, B and T to decide if the
substance may be a potential PBT/ vPvB candidate .
**** These threshold values only apply for the aquatic compa rtment.

A substance is considered to potentially meet the criteria f or T when an acute E(L)C50 value
f rom a standard E(L)C50 toxicity test (REACH Annexes VII to X) is less than 0.1 mg/ L. In
addition to data f rom standard toxicity tests, data f rom reliable non-standard tests and non-
testing methods may also be used if available. These data should be particularly assessed f or
their reliability, adequacy, relevance and completeness (see Chapter R.4 of the Guidance on
IR&CSA).

The toxicity criterion (T) f or PBT assessment cannot be decided upon the basis of acute studies
alone. If the screening threshold value is met, the substance is ref erred to T testing and
chronic studies are needed unless E(L)C50 < 0.01 mg/L. Normally, the testing order f or
57
conclusion on T based on chronic data is Daphnia and then f ish . If the T-criterion is f ulf illed
by the chronic algae or Daphnia data, a chronic f ish test is not necessary and should theref ore
not be carried out as it would be an unnecessary vertebrate animal test.

For certain lipophilic substances (with a Log Kow > 4) acute toxicity may not occur at the limit
of the water solubility of the substance tested (or the highest concentration tested). In such
situations, chronic toxicity with a NOEC/EC10 < 0.01 mg/L cannot be excluded. Theref ore, it
may not be possible to draw a screening conclusion f or T (see decision tree f or aquatic
endpoints, steps 2, 5 and 6, and Figure R.11—6).

In the absence of conclusive inf ormation on T, f or substances with very high lipophilicity, a
Weight-of-Evidence or grouping approach f or long-term toxicity may be used to predict
whether long-term effects are likely to occur. If convincing evidence is available that aquatic
toxicity is not expected to occur at < 0.01 mg/L, chronic testing may not be required. Such
evidence should be based on expert judgement and Weight-of-Evidence of data including
reliable QSAR predictions/read-across/grouping approaches indicating a narcotic mode of

57
Algae are not mentioned here because chronic algae data (i.e. 72h NOEC) normally will be available, as it can be
easily obtained from the same 72h standard test from which the acute endpoint (72h EC50) is derived.
 Chapter R.11: PBT/vPvB assessment

139 Version 4.0 – December 2023

action together with measured low chronic f ish toxicity f rom a related substance. Supporting
inf ormation could be chronic data on aquatic species such as, e.g., daphnids, algae or
sediment dwelling species and/or low acute or chronic mammalian and avian toxicity.
If data f rom this approach provide insufficient evidence that toxicity will not occur in a chronic
test a conclusion on the P and B properties should be drawn bef ore f urther T-testing is
considered. If the substance is f ound to be both P and B, a chronic study is required (testing
order see above).
In choosing the appropriate test organism, the data f rom the available base set of toxicity
tests f or algae (acute / chronic), Daphnia (acute) and f ish (acute) should be evaluated under
consideration of the possible hydrophobic properties of the test substance, and hence the
expected time to steady-state. Any specific mode of action of the test substance also needs to
be considered.
If it can be concluded that one taxonomic group is signif icantly more sensitive than the others,
e.g. because there is evidence f or a specif ic mode of action, this sensitive group should be
58
chosen f or chronic testing and conclusion on the T-properties . If no conclusive evidence f or
signif icant dif f erences in sensitivity between the groups can be f ound the testing order as
mentioned above applies.
If the relevant test species is selected in accordance with the suggested approach in the
paragraph above, lack of toxicity at or below the T criterion f or the tested species is evidence
that f urther studies on T are not necessary. If however a long-term test on Daphnia or algae
provides a NOEC close to but above 0.01 mg/L, a long-term f ish study is likely to be needed to
conf irm “not T” unless, taking into consideration the above-mentioned approach, convincing
evidence exists that the f ish NOEC will be higher than 0.01 mg/L. Supporting evidence in such
considerations could be an acute f ish value that is a f actor of 10 or more greater than that of
the other two trophic levels under the provision that the acute daphnid test showed toxicity at
least one order of magnitude lower than the limit of solubility.
Certain chemical characteristics (such as high adsorption or extremely low solubility) are likely
to make any toxicity testing extremely laborious if not technically impossible. Guidance has
59
been developed by OECD on toxicity testing of dif f icult substances (OECD, 2019) . Some
examples together with recommendations to overcome the technical dif ficulties are provided in
the chapter on assessment of problematic substances (see Chapter R.7b of the Guidance on
IR&CSA).

R.[Link].5 Water accommodated fraction (WAF)

For any substance with very low water solubility, all ef forts should f irst be made to produce a
reliable and stable test concentration. For a UVCB substance which is only partially soluble in
water, it may not be f easible to conduct the test f or individual constituents/blocks of
constituents due to the properties of the substance or due to disproportionate ef f orts. In this
case the water accommodated f raction (WAF) can be considered as last resort to generate

58
This could mean that no further testing is necessary if it is concluded that algae are significantly more sensitive than
daphnids or fish and the available chronic algae data are well above a NOEC of 0.01 mg/L.
59
The OECD "Guidance document on aqueous-phase aquatic toxicity testing of difficult test chemicals" has introduced
additional recommendations for poorly water-soluble chemicals, and in particular with regard to the use of liquid/liquid
saturator units and of passive dosing.
 Chapter R.11: PBT/vPvB assessment

140 Version 4.0 – December 2023

exposure in a test (OECD, 2019; Girling et al., 1992, see also Appendix R.7.8-1 in Chapter
R.7b of the Guidance on IR&CSA). Test results are expressed as a lethal or ef f ective loading
that causes a given adverse ef f ect af ter a specif ied exposure period. For complex multi-
constituent substances, the principal advantage of this test procedure is that the observed
aquatic toxicity ref lects the multi-component dissolution behaviour of the constituents at a
given substance to water loading. Expressing aquatic toxicity in terms of lethal loading enables
multi-constituent substances comprised primarily of constituents that are not toxic to aquatic
organisms at their water solubility limits to be distinguished f rom substances that are more
soluble and which may elicit aquatic toxicity. As a consequence, this test procedure provides a
basis f or assessing the toxicity of UVCB substances as a whole. Ef f ect concentrations in tests
based upon WAFs can be calculated f rom (1) the loading rates and are identif ied as either LL 50
or EL50 values and/or (2) the measured mass of test substance in the WAF and are identif ied
as either LC50 or EC50 values. LL50 or EL50 values are comparable to LC 50 or EC50 values
determined only f or pure (i.e. mono-constituent) substances tested within their solubility
range. Similarly the NOEC (No Observable Ef f ect Concentration) becomes the NOELR (No
Observable Ef f ect Loading Rate). The statistical methods used to determine LL 50, EL50 and
NOELR values are the same as those used to determine LC 50, EC50 and NOEC values.

The WAF procedure has been adopted for use in environmental hazard classification (f or acute
and long-term hazard classification) (OECD, 2019; UNECE, 2003). Poorly soluble substances
that exhibit no observed chronic toxicity at a substance loading of 1 mg/L indicate that the
respective constituents do not pose long term hazards to the aquatic environment and,
accordingly, do not require hazard classification (CONCAWE, 2001; UNECE 2003). Using WAF
data f or PBT assessment may be more complex as PBT assessment is required f or all
constituents of the substance f ulf illing P and B criteria and by its nature the WAF-method is
testing several constituents. Where toxicity is exhibited by a test substance containing several
constituents, interpretation of the test results can be problematic if the toxicity cannot be
allocated to specif ic constituents directly. In such case, interpretation of the results (given that
use of WAF is the last resort) should be supported by use of other data, such as QSAR –values
or read-across values from a structurally similar substances. Generally, the loading rate cannot
be directly compared to the Annex XIII toxicity criterion as true ECx value may be
overestimated. To f acilitate the interpretation of such data, analytical verif ication of the
constituents present in the test solution must be provided. Only in the case of analytical
verif ication of the water-soluble f raction(s), this type of tests might be used in the PBT
assessment.

R.[Link].6 Use of non-testing data

At preliminary stages in the assessment, in cases where no acute or chronic toxicity data are
available, the assessment of the T criterion at a screening level can be perf ormed using data
obtained f rom quantitative structure activity relationships (QSARs) f or acute aquatic toxicity as
described in Table R.11—7. In order to be suitable, the QSAR prediction should comply with
the general principles described in Chapter R.6.1 of the Guidance on IR&CSA. Long-term
testing is required if QSAR estimations indicate that the substance f ulf ils the screening
threshold values f or T (EC50 or LC50 < 0.1 mg/L). It may, on a case by-case-basis, be decided
whether confirmatory chronic testing on f ish is necessary if v alid QSAR prediction indicates
that the acute E(L)C50 is < 0.01 mg/L. Alternatively either first an acute f ish toxicity limit test
could be perf ormed to check whether the acute toxicity is below 0.1 mg/ L or the QSAR-
prediction could be accepted as providing sufficient evidence of the T criterion being f ulf illed.
 Chapter R.11: PBT/vPvB assessment

141 Version 4.0 – December 2023

If the substance is conf irmed to f ulfil the P and B criteria, testing on long-term toxicity should
be perf ormed to determine whether the substance meets the criteria f or T. Alternatively,
QSARs f or chronic toxicity, if applicable, may be used by the registrant to conclude that the
substance f ulfils the T criterion, but normally, due to the uncertainties of the present QSAR-
models, these cannot be used to conclude “not T”.

When considering the use of non-testing data, it is important f or substances containing

multiple constituents, impurities and/or additives, to consider f irst the appropriate assessment
approach provided in Section R.[Link].

R.[Link] Conclusions on PBT or vPvB properties

A detailed analysis of the Persistence, Bioaccumulation and Toxicity should be brought

together into a clear overall conclusion. Three conclusions f or the comparison of the relevant
available inf ormation on the PBT properties with the criteria listed in REACH Annex XIII Section
1 are possible.
(i) The substance does not fulfil the PBT and vPvB criteria. The available
inf ormation show that the properties of the substance do not meet the specif ic
criteria provided in REACH Annex XIII Section 1, or if the inf ormation does not allow
a direct comparison with all the criteria there is no indication of P or B properties
based on screening inf ormation or other inf ormation.
(ii) The substance fulfils the PBT or vPvB criteria. The available inf ormation show
that the properties of the substance meet the specif ic criteria detailed in REACH
Annex XIII Section 1 based on a Weight-of-Evidence determination using expert
judgement comparing all relevant and available inf ormation listed in Section 3.2 of
Annex XIII to REACH with the criteria.
(iii) The available data information does not allow to conclude (i) or (ii). The
substance may have PBT or vPvB properties. Further inf ormation f or the PBT/vPvB
assessment is needed.
The sub-chapters below provide more details on the circumstances that would lead to each of
these conclusions. The consequences of each conclusion f or the registrants are described in
Section R.11.3.
The prerequisite f or drawing a correct overall conclusion is that the endpoint –assessments
described in Sections R.[Link], R.[Link] and R.[Link] are carried out and concluded
correctly. Additionally, the assessment described in Section R.[Link] f or substances
containing multiple constituents, impurities and/or additives needs to be carried out in such
manner that the principles f or choosing an approach are f ulf illed (see Section R.[Link] f or
details). A very high number (tens) of combinations of end-point conclusions is possible. . If a
substance contains multiple relevant constituents, impurities and/or additives, the overall
picture may be highly complex. In such cases the overall conclusion(s) can be best presented
by providing conclusion tables f or all relevant constituents, impurities and/or additives (or
f ractions, where relevant).
 Chapter R.11: PBT/vPvB assessment

142 Version 4.0 – December 2023

R.[Link].1 (i) The substance does not fulfil the PBT and vPvB criteria . The
available information show that the properties of the substance do
not meet the specific criteria provided in REACH Annex XIII Section
1, or if the information does not allow a direct comparison with all
the criteria there is no indication of P or B properties based on
screening information or other information.

This would be the case if , as a result of an analysis of existing data, or of data generated af ter
conclusion (iii) any one of the parameters, i.e. environmental degradation half -lif e in an
appropriate environmental compartment, the BCF f or aquatic species (or a comparable metric)
or, in the case of a decision on PBT, long-term aquatic toxicity and the appropriate human
health hazard classif ication do not meet the criteria in Annex XIII.

In many cases, the inf ormation available, while not allowing a direct comparison with the
criteria in Annex XIII, can be considered suf f icient f or a decision to be made, by applying
Weight-of-Evidence based expert judgement, that the substance is not PBT/vPvB. Such would
f or instance be the case if the screening threshold values as provided in Section R.11.4 were
not met f or any particular endpoint based on screening inf ormation. Furthermore, w hen the
screening threshold values f or persistence or bioaccumulation as def ined in the f ollowing sub-
sections are not f ulf illed, f urther PBT/vPvB assessment can stop when there is a well justif ied
lack of counter evidence which would raise concern f or the substance to ha ve PBT or vPvB
properties. In this case, the registrant can also draw the conclusion (i).

It has to be kept in mind that the f act that a substance does not meet the T criterion is not a
suf ficient basis on which to stop the evaluation of the remaining endpoints in the PBT/vPvB
screening step.

Where supplementary inf ormation is available, such as suf f icient evidence based on
monitoring data, that indicates that a particular property, such as persistence or high
bioaccumulation may in f act be present, a cautious approach should be f ollowed and
conclusion (iii) may need to be drawn (see below).

When drawing conclusion (i), the registrant should show in the PBT/vPvB assessment that
there is no indication that the relevant constituents, impurities, additives or
transf ormation/degradation products have PBT or vPvB properties.

It should be noted that where toxicity is a critical parameter f or PBT assessment, i.e. the
substance is persistent and bioaccumulative but there are insuf ficient (only acute valid) toxicity
data, it will be necessary to conduct f urther testing (unless the registrant decides to treat the
substance “as if it is a PBT or vPvB”). In such cases, the assessor must choose conclusion (iii)
instead of conclusion (i).
 Chapter R.11: PBT/vPvB assessment

143 Version 4.0 – December 2023

R.[Link].2 (ii) The substance fulfils the PBT and/or vPvB criteria. The
available information show that the properties of the substance
meet the specific criteria detailed in REACH Annex XIII Section 1
based on a Weight-of-Evidence determination using expert
judgement comparing all relevant and available information listed
in Section 3.2 of Annex XIII to REACH with the criteria (for more
specific terminology, also used in IUCLID, please, see subsection
“Terminology”).

In principle, substances are only considered as PBT or vPvB when they are deemed to f ulf il the
PBT or vPvB criteria f or all inherent properties. This would be the case if , as a result of an
analysis of existing data, or of data generated af ter concluding that f urther inf ormation is
needed (conclusion iii), the environmental degradation half -lif e in an appropriate
environmental compartment, the BCF f or aquatic species or a comparable metric and, in the
case of a decision on PBT, long-term aquatic toxicity or an appropriate human health hazard
classif ication show the criteria to be met. The data must show that all three criteria are met in
the case of PBT, or both vP and vB criteria in the case of vPvB. In this context it is important to
note that even where one criterion is marginally not f ulf illed but the others are exceeded
considerably, the assessor may, based on a justif ication relying on the available evidence and
considering weigh-of-evidence, conclude in specif ic cases that the substance f ulf ils the Annex
XIII criteria.

If a constituent, impurity or additive of a substance f ulfils the PBT/vPvB properties (based on

the assessment of the registrant or of ECHA), a ≥0.1 % (w/w) threshold applies f or concluding
the substance as f ulfilling the same PBT or vPvB criteria. For substances containing PBT/vPvB
constituents, impurities or additives in individual amounts <0.1 % (w/w) of the substance, the
same conclusion need not normally be drawn. This is in line with the threshold used f or
considering PBT and vPvB substances in mixtures (Article 14(2)(f ) of REACH).

Furthermore, where a substance contains a high number of constituents, impurities or

additives in individual amounts <0.1% (w/w) which are structurally similar and theref ore can
be considered together as a f raction, the concentration limit is considered to apply f or the
f raction. This in particular applies to highly complex substances where all or most individual
constituents are present in concentration <0.1 % (w/w) but also to other substances
containing blocks of similar constituents whereby the assessment ef f orts should remain
proportionate (f or f urther details, please, see Section R.11.4.1 on “Relevant constituents,
impurities, additives and transf ormation/degradation products” and Section R.[Link]).

Additionally, there may be other particular cases f or which specification of percentages below
0.1% is required. This requirement is then driven by the toxicological prof ile of the constituent,
impurity or additive (e.g. high potency carcinogenic, mutagenic or reprotoxic (CMR) and the
provisions f or classification and labelling) and not by the f act that the respective constituent is
concomitantly a PBT/vPvB. If a substance (its constituents, impurities or additives) degrades
or is transformed into transformation/degradation products which f ulfil the PBT or vPvB criteria
(based on the assessment of the registrant or of ECHA) and if these are f ormed in relevant
amounts, the substance is concluded to f ulf il the PBT or vPvB criteria . The def inition of
“relevant” transf ormation/degradation product f or the registrant’s substance is provided in
Section R.11.4.1. Authorities should justif y case by case what they consider as relevant
transf ormation/degradation in their PBT/vPvB assessments. Terminology provided at the end
of this section must be applied in the registration dossier to the substance subject to PBT/vPvB
assessment to distinguish which of the cases above the substance represents.
 Chapter R.11: PBT/vPvB assessment

144 Version 4.0 – December 2023

Overview of case types of conclusion (ii)

The f ollowing dif f erentiation is used f or substances which have to be concluded to f ulf il the PBT
and/or vPvB criteria:

• The substance is PBT/vPvB. This conclusion is drawn because this is a mono-constituent

substance and it has a main constituent present at a concentration of 80% or more with
PBT and/or vPvB properties;

• The substance is PBT/vPvB. This conclusion is drawn because this is a mono-

constituent substance, well-def ined multi-constituent substance or UVCB substance.
60 61
and it contains one or more relevant (group(s) of ) constituent(s) which f ulfil the PBT
62
and/or vPvB criteria ;

• The substance is PBT/vPvB. This conclusion is drawn because one or more (group(s) of)
constituent(s), impurity or additive of the substance degrade(s) or is/are transf ormed
into substance(s) which f ulfil the PBT and/or vPvB criteria and these transf ormation or
60
degradation products are f ormed in “relevant” amounts.

• Combination of two or all of the above types.

It should be noted that there is no dif f erence in risk management between the dif ferent types.
The consequences of conclusion (ii) f or the registrant are described in Section R.11.3.

R.[Link].3 (iii) The available information does not allow to conclude (i) or (ii).
The substance may have PBT or vPvB properties. Further
information for the PBT/vPvB assessment is needed.

The consequences of this conclusion f or the registrant are described in Section R.11.3.3.

This conclusion is derived when one or more of the f ollowing combinations of endpoint–specific
conclusions apply:

Potential P/vP + Potential B/vB + any T -conclusion

Potential P/vP + B but not vB + Potential Teco
Potential P/vP + B but not vB + Potential Thh
Potential P/vP + B but not vB + Teco
Potential P/vP + B but not vB + Thh
Potential P/vP + vB + any T -conclusion
Potential P/vP + B/potential vB + any T -conclusion

P/potential vP + Potential B/vB + any T -conclusion

60
“Relevant” is defined in section R.11.4.1.
61
“Constituent” as referred to in Annex XIII of REACH means “constituent”, “impurity” or “additive” as described in
the Guidance for identification and naming of substances under REACH and CLP.
62
The terminology corresponds with IUCLID 6 section 2.3 terminology. The constituent(s) or constituent group(s)
fulfilling the PBT/vPvB criteria should be specified in specific endpoint study records in section 2.3 of IUCLID.
 Chapter R.11: PBT/vPvB assessment

145 Version 4.0 – December 2023

P/potential vP + B but not vB + Potential Teco

P/potential vP + B but not vB + Potential Thh
P/potential vP + vB + any T -conclusion
P/potential vP + B/potential vB + any T -conclusion

P but not vP + Potential B/vB + Potential Teco

P but not vP + Potential B/vB + Potential Thh
P but not vP + Potential B/vB + Teco
P but not vP + Potential B/vB + Thh

P but not vP + B/vB + Potential Teco

P but not vP + B/vB + Potential Thh

vP + Potential B/vB + Any T-conclusion

vP + B + Potential Teco
vP + B + Potential Thh

Where the data on the PBT properties of a substance do not allow a direct (numerical)
comparison with the criteria specif ied in Annex XIII, but there are nevertheless indications
f rom other data such as screening data, that the substance may be PBT/vPvB, then it is
necessary to consider which inf ormation is needed to draw a f inal conclusion.

Where it is concluded that f urther inf ormation is needed, consideration should f irst be given to
clarif ying the persistence of the substance since persistence is a critical property in
determining PBT/vPvB properties and since degradation testing does not involve the use of
63
vertebrate animals .

Once the new inf ormation is available, comparison with the criteria in Annex XIII should be
carried out according to the principles described above and a decision be taken on whether the
substance f alls under conclusion (i) (is not a PBT/vPvB) or (ii) (i.e. is a PBT/vPvB). In certain
cases the revised assessment may again lead to the conclusion that f urther inf ormation still
needs to be generated. If f or one of the relevant constituents, impurities, additives or
transf ormation/degradation products there is indication that it may have P and B properties,
the registrant should draw conclusion (iii) and generate the necessary additional inf ormation
until the available inf ormation allows to draw one of the two ultimate conclusions in relation to
the whole composition (see Section R.11.4.1 f or description of “relevant” and Section
R.[Link] f or the relevant assessment approaches).

There may be cases where a clear decision on the properties of a substance cannot be made ,
but there are indications f rom available inf ormation that the substance may f ulf il the PBT or
vPvB criteria. In these cases conclusion (iii) applies. For instance, where there is a reason to

63
Depending on the substance properties it may, however, be appropriate to consider bioaccumulation testing first.
Guidance on the general approach to P, B and T testing is given in Section R.11.4.
 Chapter R.11: PBT/vPvB assessment

146 Version 4.0 – December 2023

expect that a substance may contain a known PBT constituent , impurity or additive (or
f ractions thereof) but it is not possible to characterise a substance identity to an extent that
will allow the registrant to state with enough conf idence that his substance does not contain
PBT/vPvB constituents/impurities/additives or that it does not generate
transf ormation/degradation products with PBT/vPvB properties above the relevant threshold
levels as specif ied in Section R.11.4.1.

Finally, there may be cases where it is simply technically not possible to conduct testing, either
at screening or at conf irmatory level and therefore not possible to derive conclusion (i) or (ii).
If there are no indications or justif ication which would exclude the possibility that the
substance could potentially f ulf il the criteria, conclusion (iii) should be drawn.
 Chapter R.11: PBT/vPvB assessment

147 Version 4.0 – December 2023

R.11.4.2 Assessment of PBT/vPvB properties – consideration of specific

substance properties

R.[Link] Assessment of substances requiring special considerations with

regard to testing

For substances that have exceptional properties (e.g. very high sorptivity, very low water
solubility, or high volatility), or which consist of multiple constituents, test guidelines used to
determine persistence, bioaccumulation and toxicity in the PBT/vPvB assessment may not be
directly applicable. Instead specif ic testing and assessment strategies may be warranted.

R.[Link].1 Substances with very high sorptivity

The assessment strategy should be applicable to strongly sorbing substances in general.

General considerations
In R.[Link].10 “Further data” indicators f or limited bioaccumulation are described. For
substances with very high calculated Log Kow , e.g. > 10, reduced bioaccumulation is expected.
Log Kow values > 8 cannot be measured reliably due to technical issues and need theref ore to
be calculated by property estimation methods based on the concept of Linear Free Energy
Relationship (LFER). Before using a specif ic LFER method the extent to which the structural
elements of the substance under consideration are covered by the applicability domain of the
LFER needs to be checked. For example, organometallic substances like tin organics may not
be covered whereas the corresponding carbon analogue of the substance is.

It is very important to realise that the calculated Log Kow values > 10 are used simply to
indicate a degree of hydrophobicity that is extreme. Such values should not be used in a
quantitative manner.

Assessment steps
STEP 1 Calculated / measured Log Kow

Check/generate the calculated / measured Log Kow of the substance of interest.

STEP 2 Assessment type to be applied

If the Log Kow is < 10 an assessment of P, B and T should f ollow the standard approach as
described in Section R.11.4.1.

If the Log Kow is > 10 it should be checked if available ecotoxicity and / or mammalian data
do not meet the T criteria. If the T criteria are not met, a specif ic vPvB assess ment might be
applicable as described below.

If f or a substance with Log Kow > 10 data are available demonstrating toxicity in accordance
with the T criteria f or PBT substances, then a standard PBT assessment as described in Section
R.11.4.1 is warranted.
 Chapter R.11: PBT/vPvB assessment

148 Version 4.0 – December 2023

STEP 3 vPvB Assessment for substances with Log Kow > 10

Step 3a Persistence check

Substances with transformation potential

If the substance can be transf ormed abiotically or biotically (e.g. when it has structural
moieties like ester groups, phosphites or phosphonites) it should be checked if a specif ic
biodegradation test at low concentrations and specific analysis or a specific hydrolysis test (see
Section R.7.9.4 in Chapter R.7b of the Guidance on IR&CSA) could be carried out to
demonstrate transf ormation with a primary half -lif e of < 40 d. In such circumstances, the
transf ormation/degradation products will need to be checked to ensure they do not have PBT
or vPvB properties. If the substance is transf ormed into substances not having PBT or vPvB
properties it can be considered not to f ulf il the vPvB criteria. In this case Step 3b can be
omitted.

Substances with limited transformation potential

If a substance may not be easily transf ormed based on the structure (e.g. it has no ester
f unctions or the transformation rate is limited by very low (bio)availability) it is nevertheless
recommended to estimate the metabolic pattern, using e.g. CATALOGIC, which is the
successor of Catabol (Mekenyan, 2006). For all relevant transformation/degradation products it
must be checked that they do not f ulf il the criteria f or PBT or vPvB substances. For these
substances Step 3b is mandatory.

Step 3b Bioaccumulation check for substances with limited transformation potential

Claims of a low bioaccumulation potential indicated by the Log Kow > 10 should be supported
by additional inf ormation indicating limited bioaccumulation as described in R.[Link].10
“Further data”). This inf ormation may comprise results f rom an animal study (mammalian or
f ish) conf irming no or low bioaccumulation.

Log Kow >10 and at least one additional indicator for limited bioaccumulation
If f or a substance with Log Kow > 10 at least one additional indicator (1. or 2.) mentioned
above is f ulf illed the substance should not be considered as vPvB, provided that potential
metabolites are themselves not PBT or vPvB.

Log Kow >10 and no additional indicator for limited bioaccumulation

If none of the additional indicators (1. or 2.) mentioned under Step 3b is met, then an
appropriate test as described in Section R.[Link] is warranted.

STEP 4 Overall conclusions

Log Kow >10 and ready biodegradability in a specific biodegradation confirmed

No f urther investigation necessary, if transformation/degradation products are neither PBT nor
vPvB. In this case the (parent) substance is not vPvB.

Log Kow >10 and no ready biodegradability confirmed

If at least one additional indicator f or limited bioaccumulation is f ulf illed and potential
transf ormation/degradation products are not PBT or vPvB, then the substance is not vPvB.
 Chapter R.11: PBT/vPvB assessment

149 Version 4.0 – December 2023

If no additional indicator f or limited bioaccumulation is f ulf illed a standard vPvB assessment as

described in Section R.11.4.1 is warranted.

R.[Link].2 Substances with low solubility in octanol and water

The assessment strategy should be applicable to substances with low solubility in octanol and
water and f or which lipid is the target compartment f or accumulation in organisms. For
illustrative purposes certain organic pigments are used as examples (see List of Pigments, in
Table R.11—8) .

It should be noted that these examples are presented under the assumption that the named
pigments would not have specif ic nanof orm -related properties. Whether the assumption is
correct or not is not relevant f or the purpose of the examples.

General considerations
1) Critical body burden (CBB) concept and octanol solubility
In R.[Link].10 “Further data” it is described how octanol solubility could be used in the B
assessment (Critical Body Burden approach) as well as the limits of the approach.

As octanol is a reasonable surrogate f or f ish lipid, a low substance concentration in octanol

may indicate reduced bioconcentration / bioaccumulation potential. The concept is based on
available measurements f or substances using a saf ety f actor of 10 f or the uncertainty of the
available CBB measurements. It is proposed that where a substance shows no specific mode of
action and has a

Coctanol [mg/L] < 0.002 [mMol/L] x Mol weight (g/Mol) Equation 11-3
it can be assumed that the substance has only a limited potential to establish high body
burdens and to bioaccumulate. If it does bioaccumulate, it would be unlikely to rise to levels in
biota that would cause signif icant ef f ects.

2) Octanol water partitioning

For substances with very low solubility specif ic methods exist to derive a K ow , e.g. OECD TG
123 slow stirring method. However, this method is not always applicable due to experimental
constraints caused e.g. by the low solubility and the available analytical methods.

Kow values derived f rom f ragment based LFER methods like KOWWin (US EPA, 2000b) of ten
overestimate the actual Kow of such substances e.g. organic pigments (Table R.11—8). In order
to overcome the dif ficulties in measuring the Kow , the solubility in octanol (Co) and water (Cw )
may be determined separately. With these solubilities the quotient Log Co/Cw can be
calculated. This quotient is not exactly identical to Log Kow , as the latter is related to the
partitioning of the substance in water-saturated octanol and octanol-saturated water. For
Pigment Yellow 12, Log Co/Cw as well as Log Kow (f rom solubility measurements using water-
saturated octanol and octanol-saturated water) have been determined as 2.1 and 1.8, and
hence being in the same order of magnitude (see Table R.11—8). This single comparison
between Log Co/Cw and Log Kow needs f urther verification but the f igures available f or Pigment
Yellow 12 can be interpreted as f ollows: as water saturation in octanol diminishes the octanol
solubility of the substance and octanol saturation in water enhances the water solub ility, the
Log Kow of the substance should normally be smaller than Log Co/Cw (see values f or Pigment
 Chapter R.11: PBT/vPvB assessment

150 Version 4.0 – December 2023

Yellow 12, Table R.11—8). A measured Log Co/Cw = 4.5 would mean that the measured Log
Kow should be < 4.5.

In Table R.11—8 solubility data are given f or some other organic pigments as well. The
comparison of the measured quotient Log Co/Cw with estimated Log Kow using KOWWIN (US
EPA, 2000b) shows that the estimated K ow exceeds Co/Cw by between 1 and 8 orders of
magnitude.

Table R.11—8: Solubility of some pigments and comparison of their Co/Cw values
with estimated Kows (US EPA, 2000b)
Co (µg/L) Cw (µg/L)
Colour Index Mol weight Log K ow
at ambient at ambient Log Co/Cw
Name (g/Mol) (KOWWin)
temperature temperature

48* 0.8 1.8*

630 7,1
Pigment Yellow 12 0.4 2.1
50

Pigment Red 122 340 600 19,6 1,5 2,5

Pigment Red 168 464 124 10,8 1,1 7,1

Pigment Red 176 573 15 1,9 0,9 7,3

Pigment Violet 23 589 330 25 1,1 9,4

* values relating to saturated solvents = water saturated octanol, octanol saturated water, this Log
Co/Cw corresponds to Log K ow .

3) Additional Indicators to be used for the ‘B’ Assessment

As described in R.[Link].10 “Further data”, additional indicators f or low bioaccumulation
potential, such as results f rom an animal study (mammalian or f ish) conf irming no or low
uptake into the organism, might also be applicable f or substances with low solubility in octanol
and water.

Assessment steps
STEP 1 Solubility measurements for Substances with low Octanol & Water Solubility
For the determination of the water solubility the column elution method and the f lask method
exist (OECD TG 105) but it needs to be checked which one is the most appropriate (Section
R.7.1.7 in Chapter R.7a of the Guidance on IR&CSA). No OECD Guideline exists f or the
measurement of the octanol solubility but in principle the OECD TG 105 methods may be used
in adapted f orm.

STEP 2 B and T Assessment

The octanol solubility of the substance is compared with the critical body burden (CBB)
according to equation (11-3) given above using the Mol weight of the substance.

Result 2A: Co < CBB

 Chapter R.11: PBT/vPvB assessment

151 Version 4.0 – December 2023

If the octanol solubility is below the CBB, the maximum uptake of the substance can be
expected to be below the CBB and toxicity is not likely.

Animal studies should, in addition, be checked to conf irm reduced uptake and low toxicity. In
this case the substance has low bioaccumulation potential and low toxicity.

Result 2B: Co > CBB and Log Co /Cw ≤ 4.5

If the octanol solubility is above the CBB a build-up to a critical concentration of the substance
in lipid cannot be excluded and additional inf ormation on adsorption is required. If the quotient
Log Co/Cw of measured solubilities is ≤ 4.5 (if measurable / available) a reduced uptake is
expected as well. Animal studies should, in addition, be assessed to conf irm reduced uptake
and low toxicity. In this case the substance can be considered to have low bioaccumulation
potential.

Result 2C: Co > CBB and Log Co /Cw > 4.5

For this substance a standard approach of P, B and T assessment as described in Section
R.11.4.1 must be applied. No conclusion on B and T can be drawn.

In addition indicators like molecular weight and average size of the molecule and reduced
uptake in mammalian studies should be checked f or f urther evidence, if necessary, and be
used in a Weight-of-Evidence approach.

STEP 3 Weight-of-Evidence approach for Results 2A & 2B

Based on the results of Step 2 (2A and 2B) a Weight-of-Evidence approach with the elements
Co, CBB, Log Co/Cw, possibly molecular weight and Dmax (size) as well as ecotoxicity and
uptake behaviour in animal studies, is warranted to demonstrate that the substance is not a
vPvB or PBT substance.

R.[Link].3 Volatile substances

This section describes dif f erent aspects related to degradation simulation testing of volatile
substances and their data treatment.

In simulation tests it is important to dif f erentiate degradation and disappearance of the test
substance due to volatilisation. Volatilisation of the test substance makes the interpretation of
the study more dif ficult and increase uncertainty of the persistence assessment. In order to
achive acceptable mass balance, use of appropriate traps f or the parent and potential
transf ormation/degradation is essential.

If the substance properties or properties of the potential transformation/degradation products

indicate potential f or volatilisation during the degradation test, this needs to be taken into
account as part of the degradation test design. Henry´s Law constant (HLC) >1.0 Pa m3/mol
or vapour pressure (VP) above 300 Pa may be used as such indicators f or volatility even if
these do not solely allow predicting the volatilisation rates in simulation tests. A case-by-case
assessment of potential volatilisation is needed, and pre-study is always recommended to
 Chapter R.11: PBT/vPvB assessment

152 Version 4.0 – December 2023

64
ensure f easibility of the simulation test. The ECHA note on Volatile substances provides
f urther considerations on identif ying volatile substances and their persistence assessment.

In addition, the Level I f ugacity models, based on steady-state calculations, may provide
inf ormation on partitioning in a closed system like the closed bottles of simulation tests.
Another suitable model, SimpleTreat v4.0 predicts the volatilisaton process f rom the aeration
tank (Mikkelsen, 1995) of a wastewater treatment plant. Level I and Simple Treat models
seem to of ten predict relatively similar distributions between air and the other compartments
(see ECHA note on Volatile substances available on ECHA Website f or f urther details).
However, volatilisation in a simulation test cannot always be excluded even in those cases
where distribution to air is predicted to be low by the models (i.e. Level I and Simple Treat).
For some substances with high WS and/or high K oc, the models may overestimate the
partitioning to the aquatic and sediment compartment, respectively, and underestimate the
partitioning to air (ECHA Guidance on IR&CSA Chapter R.16).

Several options to address volatilization of substances in OECD TGs 307, 308 and 309 are
described below. As a general rule, any modif ication of the test design or set-up should be
consistent with the OECD TGs conditions and the validity criteria of the guidelines should be
f ulf illed. Minimising the losses in surf ace water simulation test due to volatilisation Birch et al.
(2023) recommend handling aqueous solutions using gas tight syringes, conduct the studies in
gas tight vials and apply automated analysis directly on unopened test vials.

A pre-test or pilot study (OECD TGs 307, 308 and 309) could be necessary in order to
determine what modif ications to the guideline are necessary and f easible (e. g. whether
biometer type or closed test vessels should be used).

Whenever volatility is suspected, biometer-type-flasks or closed test vessels should be

used to ensure adequately high recovery rates. Following considerations should be taken into
accont regarding the biometer-type and closed vessels design:

• Closed vessels with minimised headspace (OECD TGs 307, 308 and 309). For example
volume ranging f rom 100-125 ml have been used (Shrestha et al, 2019; OECD TG 309
65
with EC 202-046-9 ).

• Maintaining aerobic conditions in closed test systems (OECD TGs 307, 308 and 309) to
compensate the oxygen consumption by the biomass. A gentle stirring of the water
surf ace is recommended in the TGs 309 and 308. According to the OECD TG 307 if
biometer-type f lasks are used, exchange of air is maintained by dif f usion. However, the
guideline indicates that under aerobic conditions, the soil moisture content should be
adjusted to maintain adequate aeration and nutrition of soil microf lora. Care should be
given to prevent or minimise losses of test substance and/or
transf ormation/degradation products by volatilisation during moisture addition in soil.

64
ECHA note: Options to assess persistence of volatile substances in regulatory PBT assessment (2022) available at
[Link]
e5d0-3eae3e1c26dc?t=1669388686441

65
[Link]
446e-89b2-9d8d33361290
 Chapter R.11: PBT/vPvB assessment

153 Version 4.0 – December 2023

• System geometry (OECD TG 308). In OECD TG 308 a system geometry with a larger
diameter to increase surface area at the air-water and water-sediment interf aces has
been used to improve oxygenation (Shrestha et al., 2016). Increasing the water-air
interf ace may also lead to more losses due to volatilisation. Headspace volume and
height of the water and sediment columns can inf luence the partitioning and
consequently degradation of the test substance (Hennecke et al., 2014; Shrestha et al.,
2016), especially f or volatile substances. The test design and any modif ication proposed
in the system geometry should be consistent with the range indicated in the standard
OECD TGs conditions f or water-sediment volume ratio and height and weight of the
sediment layer.

• Inf luence of the organic carbon (OC) content (OECD TGs 307, 308 and 309). In
general, higher volatilization is expected for systems with low organic carbon content as
adsorption of the substance to organic material may hinder volatilisation. In sediment
simulation tests, the OC content of sediments should be within the range indicated in
the OECD TG 308. The amount of suspended solids in the water phase of the OECD TG
309 pelagic tests should be representative of the level of suspended solids in the EU
surf ace water, c.a. 15 mg dw/L. Testing natural surf ace waters containing between 10
and 20 mg SPM dw/L is considered acceptable in the OECD TG 309 (pelagic test). In the
case of testing volatile substances in the OECD TG 307 it is recommended to include at
least one soil with high organic carbon content, e.g. > 10%, which is the highest
organic carbon content range f or soils recommended f or soil adsorption/desorption
studies in the Final Report of the OECD Workshop on Selection of Soils/Sediments
(OECD, 1995), which is ref erenced in the OECD TG 307.

• Absorbing surf aces (all TGs). Use of tubing, plastic and rubber components or other
absorbing surfaces in respirometer systems should be minimised. These components
have been observed to lead to increased losses of volatile substances f rom test
66
systems (Brown et al., 2020) resulting in lower bioavailability, lower biodegradation
observed and lower mass balance recoveries. Special attention should be paid to
components located in the headspace. Stainless steel, plastic-free bottle caps and lock
systems (Swagelok connections, Hamlet valves and f itting) have been used to
completely close the test setup (Shrestha et al 2019, 2020).

Increase number of samplings to monitor substance concentrations in water,

sediment, soil and air. Additional sampling points during the initial period of the study may
be needed in order to determine the rate of distribution between the dif f erent phases of the
test system (water, sediment, soil and air) f or a successf ul kinetic modelling.

Adsorption traps. In f low-through systems, the setup should be permanently connected to a

series of dif f erent trapping solutions in order to passively trap the volatilized f raction, e.g.,
NaOH and ethylene glycol or tenax f lasks to trap mineralized 14CO2 and volatile
parent/degradation products, respectively. In biometer type f lasks, systems f or adsorbing
volatile parent/degradation products, e.g. coated glass wool or polyurethane f oam, and CO2,
e.g. soda lime, can be used. Other methods such as using purge and trap gas

66
Biodegradation in water and sediment: simulation tests of Di -tert-butyl 3,3,5-trimethylcyclohexylidene diperoxide
(Trigonox 29) (EC number: 229-782-3 | CAS number: 6731-36-8) . [Link]
/registered-dossier/13187/5/3/3/?documentUUID=8cb36c04-6283-4938-a9f1-70fafd154f4e
 Chapter R.11: PBT/vPvB assessment

154 Version 4.0 – December 2023

chromatography/mass spectrometry with evaluation of mass balance at each time point may
also be applicable f or volatile substances.

Preparation of test solutions (All TGs). During the preparation of the stock solution and
the diluted test solutions f or exposure, f lasks should be sealed and the headspace kept to a
minimum or eliminated (OECD GD 23). The combination of hydrophobicity and volatility makes
testing challenging especially to dissolve a substance in water and avoid evaporative losses.
The use of a co-solvent could be needed but it can decrease the oxygen concentration in the
water phase due to the oxygen consumption by microorganisms degrading the solvent
(Shrestha et al. 2020). Theref ore, the use of co-solvent should be avoided or minimised as
much as possible, and, if needed, a solvent with slow degradation rate should be chosen.

Methods for substance application. In the case of volatile substances, the application
solution can be spiked directly on the sediment matrix and the biometer-f lask immediately
closed using an insert cap to avoid losses due to volatilization. Sediment spiking is currently
not mentioned as an option in OECD TG 308 but it has been used in water -sediment system
studies f or highly volatile substances and the results have been f ound suitable f or regulatory
67
persistence assessment, e.g. sediment simulation study with decamethylcyclopentasiloxane
(EC 208-764-9)).

Monitoring oxygen levels. It is recommended to monitor the aerobic conditions inside the
closed f lasks, including the water phase and headspace, pref erably by external oxygen
measurements, thereby minimizing test substance losses due to opening of the vessel. This is
not necessary if it can be reliably demonstrated that O2 concentration remains sufficiently high
until the end of the study. Samples with high ThOD e.g. ref erence substance and solvent (if
applied) may be used to indicate that the O2 was suf ficient f or the period of degradation of the
ref erence substance, although it may not inf orm on O2 level at the end of the study.

Sterile controls. Sterile controls are always recommended to determine to what extent the
decrease of the test substance is due to biotransf ormation or to potential abiotic losses (e.g.
adsorption to caps and leaking out of the test system). Test conditions and operational
conditions in sterile controls should be the same as in active vessels (see also Section
R.[Link].3 on “Sterile controls” and Section R.[Link] in Chapter R.7 of the Guidance on
IR&CSA f or additional inf ormation). Abiotic losses can be counted via concentration ratios
between the test system and sterile controls (Birch et al. 2023).

Use of 14C radiolabelled material. The simulation study must be performed preferably using
a radiolabelled test material and a mass balance has to be included in reporting. Use of 14C
labelled test substance f acilitates the monitoring of the test substance and verif ying
degradation kinetics, obtaining the complete mass balance and dif f erentiate mineralization,
NER f ormation and volatilization. For highly volatile substances this is especially important
because decreasing mass balance during the study can indicate that there may have been
leakage of the test substance f rom the test system. If radiolabelling is not f easible, a
systematic and reliable monitoring of the test substance and transf ormation/degradation

67
[Link]
4b4c-9cca-a2f7e256db6d
 Chapter R.11: PBT/vPvB assessment

155 Version 4.0 – December 2023

products should be perf ormed in all compartments of the test system (water/sediment or soil
and air) and reported during the whole simulation study.

Data treatment and interpretation issues

There are several volatilisation correction approaches available f or the kinetic analysis of
simulation study results. More detailed description of the f ollowing procedures is included in
Appendix R.11-7 and in ECHA note on Volatile substances available on ECHA Website.

A. Separate f itting of data on total dissipation and volatile traps

B. Simultaneous f itting of data on parent substance in water-sediment and volatile traps to
SFO kinetics in OECD 308 using suitable sof tware (e.g., CAKE)
C. Simultaneous f itting of data on parent substance in soil and volatile traps to SFO
kinetics in OECD 307 using suitable sof tware (e.g., CAKE)

In all correction procedures it is assumed that the volatile losses of parent substance are
adequately identif ied and quantif ied in the volatile trapping systems. Losses through other
processes, such as adsorption, increase uncertainties in the kinetic analyses.

R.[Link].4 Substances with nanoforms

When a substance f ulfils the criteria of the nanof orm definition, specif ic considerations on the
testing and assessment apply. The term nanof orm ref ers to a nanof orm or a set of similar
nanof orms as described in REACH Annex VI. According to REACH Annex I, PBT and vPvB
assessment shall address all nanof orms covered in the substanc e registration.

Definition of a nanoform and a set of similar nanoforms (REACH Annex VI)

On the basis of the Commission Recommendation of 18 October 2011 the definition of nanomaterial,
a nanoform is a form of a natural or manufactured substance containing particles, in an unbound
state or as an aggregate or as an agglomerate and where, for 50 % or more of the particles in the
number size distribution, one or more external dimensions is in the size range 1 nm -100 nm,
including also by derogation fullerenes, graphene flakes and single wall carbon nanotubes with one or
more external dimensions below 1 nm.

The term ‘nanoform’, refers to a nanoform or a set of similar nanoforms, when one has been defined .
A ‘set of similar nanoforms’ is a group of characterised nanoforms where the clearly defined
boundaries in the parameters of the individual nanoforms within the set still allow to conclude that
the hazard assessment, exposure assessment and risk assessment of these nanoforms can be
performed jointly.

Updated Commission Recommendation (10 June 2022)

’Nanomaterial’ means a natural, incidental or manufactured material consisting of solid particles that
are present, either on their own or as identifiable constituent particles in aggregates or agglomerates,
and where 50 % or more of these particles in the number -based size distribution fulfil at least one of
the following conditions: (a) one or more external dimensions of the particle are in the size range 1
nm to 100 nm;(b) the particle has an elongated shape, such as a rod, fibre or tube, where two
external dimensions are smaller than 1 nm and the othe r dimension is larger than 100 nm; (c) the
particle has a plate-like shape, where one external dimension is smaller than 1 nm and the other
dimensions are larger than 100 nm. In the determination of the particle number -based size
distribution, particles with at least two orthogonal external dimensions larger than 100 μm need not
be considered. However, a material with a specific surface area by volume of < 6 m 2/cm 3 shall not be
considered a nanomaterial. [Link]
content/EN/TXT/PDF/?uri=CELEX:32022H0614(01)&from=EN
 Chapter R.11: PBT/vPvB assessment

156 Version 4.0 – December 2023

According to REACH Annex VI, each nanof orm needs to be characterised. Furthermore, Annex
VII to X inf ormation requirements (i.e. fate and (eco)toxicity data) apply individually to each
nanof orm of a substance. For example, degradation assessment of nanof orms that are not
soluble, nor have high dissolution rate, shall consider also morphological transf ormation (e.g.
irreversible changes in particle size, shape and surf ace properties, loss of coating), chemical
transf ormation (e.g. oxidation, reduction) and other abiotic degradation (e.g. photolysis).

Hazard properties of a nanof orm may differ between dif f erent nanof orms and f rom its non-
nanof orm and, in general, testing of these properties require specif ic attention and advice.
Theref ore, PBT/vPvB assessment has to be carried out f or each relevant organic and
organometallic nanoform of a substance. For nanoforms the surf ace treatment agent must also
be considered. Currently, there is no nano-specif ic Appendix f or ECHA Guidance IR&CSA,
Chapter R.11: PBT/vPvB assessment. Appendices for nanoforms applicable to ECHA Guidance
on IR&CSA Chapters R.7a, R.7b and R.7c provide advice for hazard identif ication of nanoforms.

Where it can be demonstrated that a nanof orm behaves in the same way as a non-nanoform of
the substance, the ECHA Guidance on IR&CSA Chapters R.7a, b and c and R.11 may in general
apply. This is the case when a nanof orm is considered to be highly soluble i.e. no detectable
particles are present. According to the Appendix for ECHA Guidance IR&CSA on nanoforms
applicable to Chapter R.7a, Section 1.2.1, nanof orm is highly soluble if its water solubility is
measured to be above 100 mg/L in less than 24h (according to OECD GD 318). It is expected
that such nanof orms will elicit the same response as the soluble non-nanof orms of the
substance.

If the water solubility of a nanof orm is less than 100 mg/L, within 24h, then PBT assessment
has to be perf ormed with specif ic attention to the specif ic behaviour of the nanof orms by
considering the corresponding appendices f or nanof orms (solid or undissolved f raction)
applicable to ECHA Guidance on IR&CSA Chapters R.7a, R.7b and R.7c. For the dissolved
f raction ECHA Guidance IR&CSA, Chapter R.11 can be f ollowed.
 Chapter R.11: PBT/vPvB assessment

157 Version 4.0 – December 2023

R.[Link] Assessment of substances containing multiple constituents ,

impurities and/or additives

Annex XIII to the REACH Regulation requires that relevant constituents are taken into account
in the PBT/vPvB assessment. Section R.[Link] describes registrants’ obligations in this
matter and Section R.11.4.1 (under “Relevant constituents, impurities, additives and
transf ormation/degradation products”) provides ECHA’s interpretation of the term “relevant”.

This section gives recommendations on how to assess a substance containing several/many

constituents, impurities and/or additives. In the f ollowing the term “constituent” is used to
cover all these, in line with the legal text. A particular emphasis is given to UVCB substances,
68
but the guidance should be applied by analogy f or those well-def ined substances which
contain several/many relevant constituents.

The assessment stages, listed briefly below, are the same as f or assessing pure (i.e. mono-
constituent) substances but contain some additional f eatures due to the complexity of
assessment. The additional features are highlighted in bold and discussed in the
corresponding subsections. The purpose of these additional f eatures is to enhance the
assessment efficiency by showing ways to use the limited inf ormation normally available on
dif f erent constituents and to help in building an ef f ective strategy f or generating f urther
inf ormation, where needed. Ultimately this helps to avoid the elaborate option of taking into
account – i.e. assessing – all relevant constituents individually.

• Gathering of available inf ormation: similar requirements as f or any substance under

REACH apply. However, f or substances containing multiple constituents specif ic
attention needs to be paid that all relevant inf ormation on identity and properties of the
constituents and on the whole substance is gathered. In addition, specif ic attention
needs to be paid that all relevant inf ormation on the test item identity/composition is
gathered in order to be able to assess to which extent the gathered data actually
represents the registered substance.

• Assessment:

o Initial profiling of the substance composition f or the purpose of the PBT/vPvB

assessment, including prof iling of the unidentif ied constituents/constituent f ractions
using available inf ormation on substance identity

o Assessment using one or more of the assessment approaches described below. If the
approaches and principles def ined in this section are correctly applied, guidance in
sections R.[Link], R.[Link] and R.[Link] can be applied to the target “entities” of
assessment and testing but additionally also taking into account specific aspects of
assessing substances containing multiple constituents.

▪ If necessary, generation of f urther inf ormation: For the purpose of f urther

specif ication of identity of specific constituents or f ractions of constituents. It should
be noted that the PBT/vPvB assessment may eventually require characterisation of
constituents or f ractions of constituents to a level beyond what is normally

68
For definition of UVCBs, well-defined multi-constituent and mono-constituent substances, please see the Guidance
for identification and naming of substances under REACH and CLP.
 Chapter R.11: PBT/vPvB assessment

158 Version 4.0 – December 2023

suf f icient and necessary to identif y constituents of the registered s ubstance

according to section 2 of Annex VI to the REACH Regulation. However, the level of
detail to be pursued is also dependent on the f easibility and proportionality of
ef f orts and is theref ore case dependent.

▪ Testing selected constituent(s)/fractions of constituents (or in well justif ied cases

the whole substance) f or necessary properties. For substances containing various
constituents the choice of appropriate test items is essential. Furthermore, the
order in the normal tiered testing strategy (P f irst, then B, then T) may in some
cases be changed, depending upon the ease and cost of generating such data and
animal welf are considerations. Testing process may, e.g. start af ter a P and B –
screening assessment with B–testing of the most relevant f ractions with appropriate
analytical characterisation of all constituents. Based on these results the specif ic
f ractions to be tested in degradation and ecotoxicity tests could be narrowed
f urther. Due to animal welf are considerations such reverse order of testing should,
however, only be carried out when it is likely that B-testing will anyway be needed
and that the reverse order does in no case lead to more vertebrate testing than
what would be the case when starting with degradation testing.

o Next tier of the assessment will include change/modif ication of the assessment
approach, where needed, and repetition of the previous steps, if needed.

o Conclusion (see Section R.[Link]).

R.[Link].1 Initial profiling of the substance composition

The complexity of the composition dif fers greatly between substances. Even f or some UVCBs,
the composition may be f ully known. For other UVCBs as well as f or large f ra ctions of
impurities of well-def ined substances knowledge of the exact composition may be limited.

The Guidance for identification and naming of substances under REACH and CLP prescribes
that unknown constituents are reported as f ar as possible by a generic description of their
chemical nature f or the identif ication of a substance. This description must be f it-f or-purpose
in light of determining the properties of the substance. For the PBT/vPvB -assessment, the
description of these unknown constituents needs to be provided to the level of detail making
screening PBT/vPvB -assessment possible and f easible. Type and expected variation of
constituents (in terms of chemical groups or classes) will determine the level of detail. For
example, f or petroleum substances it would be hydrocarbon class, like mono-aromatics, n-
alkanes, etc... For UVCB substances of botanical origin (e.g. essential oils) it could be
terpenoid blocks, such as "monoterpene" and "sesquiterpene", subdivided by the appropriate
f unctional descriptors "hydrocarbon", "alcohol", "ketone", etc and/or carbon skeletons
69
"acyclic", "monocyclic", "bicyclic", etc… The limitations of the analytical methods and
proportionality of efforts to make other related inf ormation available may define the achievable
level of detail and are case dependent. Therefore, the level of detail to be used to describe the

69
For further guidance provided by the fragrance industry, please, see: [Link]
identification/sector-specific-support-for-substance-identification/essential-oils
 Chapter R.11: PBT/vPvB assessment

159 Version 4.0 – December 2023

constituents will vary f rom substance to substance and is case dependent. However, the level
of available detail should allow def ining chemical classes/functions present or modelling of the
individual structures present.

Descriptors such as identity of the chemical f unctionalities present, molecular weight range,
carbon number range, etc. may be useful as specifications. In some cases, these co nstituents
may be best reported as a group (e.g. ‘alkanes, C10-13, chloro’ or “sesquiterpene
hydrocarbons, C15H24”). Raw material(s) and manuf acturing process details may help in
generating the necessary inf ormation on substance composition. Prof iling of the composition
with new methods, e.g, as reviewed by Dimitrov et al. (2015) is recommended f or the purpose
of f illing the data gaps at screening level.

An example of an initial prof iling strategy of a f raction of unidentif ied constituents is given
below:

1. Assess the available data that is used to characterise/describe the substance.

Inf ormation derived by chemical identity characterisation is of highest value, but if such
cannot be derived f or technical f easibility reasons, other inf ormation sources can also be
used. For example boiling point range is typically one of the main descriptors of
petroleum substances and, if used combined with other more specif ic manuf acturing
inf ormation, it can be used to generate a list of structures that could reasonably be
predicted to be present in the substance. For example with petroleum substances this
would probably be hydrocarbon classes within specif ied chain lengths, degree of
branching, and content of (iso)alkane, cyclic and aromatic constituents. For other
classes of similar substances that are also UVCB (e.g. many surf actants, essential oils,
halogenated mineral oil derived UVCBs) the composition could potentially be described as
the distribution of non-polar and polar f unctional groups, as a f unction of molecular
weight or chain length. Halogenated UVCBs could be described based on the nature of
halogenation, chain length, degree of branching, saturation, cyclic and aromatic
constituents and degree and nature of halogenation. Whatever approach is used to
characterise the composition of the UVCB substance, a scientif ic and technical
justif ication should be provided.

2. Determine the structures that are to be used as representative structures of each

f raction f or which f ull analytical identif ication is not available, detailing why these
structures are regarded as representative and, if possible, give the approximate
concentrations of the f raction f or which they are considered representative.

3. In general it would not be necessary to generate representative structures if it were

possible to demonstrate that the f raction for any representative structure were present
at less than 0.1% (w/w). In practice this may be dif f icult to achieve.

R.[Link].2 Assessment approaches

Below the approaches which are recommended to be applied are described. These approaches
are based on the idea that dif f erent “parts” (i.e. constituents or constituent f ractions) of the
 Chapter R.11: PBT/vPvB assessment

160 Version 4.0 – December 2023

70
substances are assessed separately (see the concept of “Assessment entity” ), unless the
whole manuf actured/imported substance is consisting of such similar constituents, that read
across criteria can be applied amongst them f or the purpose of the PBT/vPvB assessment.
Whichever approach is considered suitable f or a particular substance, the assessment
document should contain a clear justif ication for the choice. Issues related to f easibility and /or
proportionality of efforts may play a role in the choice of the assessment approach in addition
to the technical elements listed under each approach. These should also be duly described in
the assessment document, where appropriate.

The approaches described below do not necessarily cover all possible cases exhaustively,
hence there may be situations where a dif f erent approach, not described below, could be
justif ied.

“Known constituents” approach

This can be applied when a substance is “a priori” known to contain specif ic constituents at
relevant concentrations, these constituents are suspected based on available inf ormation to
represent the worst case of the (v)P, (v)B and T properties of all constituents of the substance,
and these specif ic constituents can be isolated or separately manuf actured or otherwise
acquired f or the purpose of testing.

In this approach, the known constituents of the substance are f irst subjected to screening
assessment individually. Hereby assessment approaches applied to pure (i.e. mono-
constituent) substances can be applied (e.g. using experimental data, read across, QSARs).
Specif ic constituents that are considered to be (the most) suspected ones with regard to the
PBT/vPvB properties are targeted in the f urther steps. Testing, if necessary, is done by using
individual constituents (or their surrogates) as test items. Each selected constituent is
assessed f or its P, B and T status, on its own, using available data on that c onstituent (or on
read across–substances, if justif ied). The f act that a constituent can be more easily isolated or
manuf actured than another constituent may play a role in the choice of the constituent f or
assessment and testing but that should not be taken as the main criterion to test this specif ic
constituent. The need to test a constituent should be driven by its relevance and
representativeness f or the overall PBT assessment of the substance (or f raction addressed).

In this approach known constituents present at ≥0.1 % w/w concentration in the substance
should normally be considered as relevant (see section R.11.4.1 f or f urther discussion on the
concentration limit). The substance can be deemed as “not PBT/vPvB” if none of the relevant
constituents individually is PBT or vPvB. This does not mean that all known constituents need
to be tested but step-wise assessment and testing is crucial f or f ocussing on the known
constituents which represent the worst case in relation to the PBT/vPvB properties among all
constituents of the substance.

In the opposite situation, if at least one of the relevant constituents meets the combination of
P, B and T or vP and vB screening criteria, the assessment needs to progress to testing of
those individual constituents f ollowing the normal P-, B- and then T-testing strategy. If one or
more of the constituents are proven to f ulf il either the vPvB or PBT criteria, the entire

70
Presentation by Magaud H et al. at SETAC Europe 25th Annual Meeting (3-7 May 2015 - Barcelona, Spain): Abstract
311 available at: [Link]
 Chapter R.11: PBT/vPvB assessment

161 Version 4.0 – December 2023

(registered) substance must be concluded as “The substance f ulf ils the PBT and/or vPvB
criteria” and the (group(s) of ) constituent(s) causing this conclusion must be specif ied in the
dossier.

This approach has been applied, e.g., in the SVHC identif ication of substances originating f rom
coal tar distillation (e.g., coal tar pitch, high temperature; anthracene oil). It was also applied
e.g. f or phenol, styrenated (EC 262-975-0) under Substance evaluation.

Advantages of the known constituents-approach are, i.a.:

• Actual tests are perf ormed on a pure (i.e. mono-constituent) discrete organic
substance, and are easy to perf orm and interpret;

• In addition to being the pref erred option, this approach may be the most ef f icient
option in cases where substances contain constituents with diverse properties ;

• It may in some cases require less ef f ort to characterise the composition of the
substance than the f raction prof iling approach described below;

• The specif ic constituents may in some cases already be known f or their properties and
hence assessment ef f ort can be reduced.

Disadvantages of the known constituents -approach are, i.a.:

• In many situations requires greater analytical ability to characterise the composition of

the substance at the start of the PBT/vPvB assessment than the “whole substance -
approach” described below;

• May require synthesis or other type of generation of specific constituent(s) f or testing, if

not otherwise available (e.g., f rom commercial providers of laboratory grade
standards). Isolation or synthesis of relevant constituent(s) may not be technically
f easible;.

• May require more than one test f or each P, B, T –endpoint. This might raise testing
costs and needs f or vertebrate testing;

• Requires justif ication that any representative constituent chosen f or testing is a

reasonable worst case.

“Fraction prof iling” (or “block prof iling”) approach

This approach is applied when, due to the complexity of the substance, it is not f easible to f ully
identif y, assess or isolate single constituents but the substance can be divided into
f ractions/blocks, in which the constituents are structurally similar or in which the constituents
are to such extent similar that their degradation, bioaccumulation and toxicity properties can
be predicted to f ollow a regular predictable pattern(e.g., C14 chlorinated n-alkane with a
71
chlorine content of 50-52 % by weight ). A prerequisite f or application of this approach is that
the PBT/vPvB-properties are assumed to be the same in the f raction (in this case the f raction
should behave with regard to the PBT/vPvB-concern as if it were a single constituent or in a

71
See for example this decision on substance evaluation: [Link]
chemicals/evaluation/community-rolling-action-plan/corap-table/-/dislist/details/0b0236e1807e3841
 Chapter R.11: PBT/vPvB assessment

162 Version 4.0 – December 2023

predictable manner relative to the single constituents) or to f ollow a regular – predictable -

pattern. The assessment report should justif y why the constituents in the blocks can be
considered to be suf f iciently similar f or the purpose of the PBT/vPvB assessment. For the
purpose of testing, an actual physical f ractionation or separate manufacturing of a f raction of
the substance may be carried out to derive appropriate test substance(s) (for more details, see
the subsection R.[Link].4 below).

A usef ul way to approach and document the assessment of the dif f erent f ractions is via a
matrix of the dif f erent blocks vs. P, B and T properties.

Two possible variations of this approach are described below:

i) The substance is conceptually divided into f ractions containing similar constituents based
on structural f ragments and/ or other relevant molecular descriptors. The f raction itself is
the main target of the testing and assessment, not individual (or surrogate) constituents
therein, as is the case in the method described below in (ii).

ii) This approach can be applied in particular to complex UVCBs, however, application to
other UVCBs or large impurity f ractions of well-def ined substances may also in some
cases be appropriate. This approach has been used in the PBT assessment of, e.g. EC no
293-728-5 under the previous legislation and is applied in several ongoing PBT/vPvB
assessments of the MSCAs (e.g., “tetrabutane”, EC 292-461-1; medium chain
chlorinated paraf f ins, EC 287-477-0).

iii) One example of this approach is where the substance is conceptually divided into
f ractions containing constituents having the same degradation behaviour (e.g. based on
ready biodegradation tests). For these f ractions the P assessment is clarif ied. The
f ractions identif ied as potentially P/vP may then b e divided f u r t h e r into f ractions
containing similar constituents and assessed and tested in the same way as above.

iv) The so-called block method: this method is applied when a substance can be divided
conceptually into f ractions containing constituents which are very similar with regard to
the properties to be assessed. Within a f raction read-across criteria can be applied
among the constituents. For each of the f ractions one or more representative
constituent(s) is/are chosen f or which testing and assessment is carried out. The
constituent can be selected based on several considerations, e.g. that it can be easily
retrieved f or testing, there are already data on that constituent available or that it
represents the worst case PBT-properties of the f raction (in case the constituents in the
f raction are expected to exhibit a pattern of P, B, and/or T -properties within the
boundaries of read across).

In all these variations of the “f raction prof iling approach” f ractions present at ≥0.1% w/w
concentration in the UVCB are normally considered as relevant.

Advantages of the “f raction prof iling approach” are, i.a.:

• More targeted and ref ined assessment compared to the “whole substance approach”

• Assessment of a complex substance f raction-wise allows ef f icient targeting of testing;

• May be the only practical option f or some very complex UVCBs ;

 Chapter R.11: PBT/vPvB assessment

163 Version 4.0 – December 2023

• Provides a ref inement option if the “known constituents approach” is not f easible.

• Disadvantages of the “f raction prof iling approach” are, i.a.:

• May require in some cases greater analytical ef f ort to characterise the substance
composition at start of PBT assessment than the “whole substance approach”;

• May requires synthesis or other type of generation of specif ic substance/test item f or

testing, if not otherwise available (e.g. raw material may in some situations be used as
representative of a f raction which consists of unreacted raw material);

• May require more than one test f or each P, B, T endpoint. This might increase needs f or
vertebrate testing.

Requires demonstration that any test item chosen f or testing is a reasonable worst case .
Figure R.11—7 below shows an anonymised example of the f irst assessment tier of a UVCB
substance f or which f raction prof iling has been applied.

Figure R.11—7: Example of the first assessment tier of a UVCB substance for which fraction
profiling has been applied

Whole substance approach

The substance is considered to be one chemical substance for the purpose of the assessment
and testing. This is possible, if all the constituents therein can be justif ied to be very similar
with regard to the PBT-properties relevant f or the assessment based on inf ormation on, e.g.
manuf acturing method, raw materials and/or chemical composition/analyses.
 Chapter R.11: PBT/vPvB assessment

164 Version 4.0 – December 2023

Due to the disadvantages and limitations, the application of the “whole substance” approach
may only be possible in certain limited cases f or the complete PBT/vPvB assessment of a
substance. If one of the above mentioned approaches is f easible, these should be used instead
of the ‘whole substance’ -approach as they are generally more transparent and regarded as
providing inf ormation of higher certainty. For certain tests and f or certain endpoint-specif ic
assessments it may be possible to address the substance as a whole despite some slight
dif f erences in the properties of the constituents. For example, if it is known or can be
reasonably assumed (e.g. based on the known chemical composition and/or relevant
description of raw materials and production process but in addition also relative to the known
or likely chemical identity of constituents) that (all) the constituents are structurally similar
and therefore can be expected to have a reasonably similar PBT-properties, using the whole
substance as test item may be considered – especially if such an analysis can be supported by
non-testing or experimental data.

In cases where “not PBT/vPvB” is concluded based on results f rom tests with the whole
substance, there should be a clear case made in the assessment f or why all constituents are
structurally suf f iciently similar and hence also similar with regard to the PBT properties to
justif y such a conclusion. Same principles in establishing similarity of constituents apply f or
monoconstituent, multiconstituent and UVCB substances. For such similarity criteria, please
ref er to Chapter R.6 of the Guidance on IR&CSA, Read-Across Assessment Framework (RAAF)
and Advice on using read-across f or UVCB substances.

The “whole substance approach” is of ten applied by the registrants. It has been observed that
the use of this approach should be better justif ied in the CSRs. Such justif ication and rationale
supporting similar P, B and T properties between constituents may include f or example
analytical data, QSAR data, physical-chemical inf ormation, bioavailability considerations, mode
of action etc..

Advantages of the “whole substance approach” are, i.a.:

• The registered substance itself is used f or testing and thus there is no need f or
generation of new material;

• It may be the only option if it is technically not f easible within reasonable ef f orts to
establish the exact identity of the constituents in the registered substance to the level
needed;

• In some cases the analytical requirements f or whole substance identif ication may be
simpler than f or identif ication of individual constituents.

Some disadvantages and considerations of situations where the “whole substance approach”
should not be applied are described below:

• Conclusion provides a single prof ile f or the whole substance. This may be too inaccurate
in some cases. Test results may not be representative of all constituents: Possible risk
of miss-screening, f or instance using a single log Kow value to represent a range of
constituents or assuming ready biodegradability f or a UVCB, where constituents are not
suf f iciently similar in reality.

• Some tests using the whole substance as test item may not produce reliable results
(e.g. if physico-chemical properties of the constituents vary signif icantly, the exposure
 Chapter R.11: PBT/vPvB assessment

165 Version 4.0 – December 2023

concentrations cannot in some cases be maintained in such way that the test would be
considered valid according to the test guideline);

• Available whole substance test data may not be relevant and/or may be unreliable
and/or be dif f icult to interpret (either due to dif f erences of physico-chemical properties
between constituents or because the composition may be partly unknown/uncertain
/vary, and hence data may not be shown to be representative enough f or the registered
substance);

• May trigger the need f or the water accommodated f raction (WAF) approach f or
ecotoxicity testing (see discussion in Section R.[Link]).

Combination of more/several approaches described above

It may be most ef f icient with regard to resources and time needed to combine se veral
approaches in the assessment of one substance. E.g., for a complex UVCB it may be necessary
to carry out an assessment of certain known constituents always present in the substance, but
also to carry out a prof iling f raction-wise f or the remaining parts of the composition of the
substance, if the remaining parts are anticipated to be so dif f erent f rom the known
constituents that they may make a dif f erence f or the assessment conclusion.

CONCAWE (2001) has used an approach which combines inf ormation f rom tests where the
whole substance has been tested and inf ormation f rom tests utilising the block approach. This
approach is presented in Appendix R.11—2.

Dif f erent approaches may also be applied at dif f erent stages of the assessment, e.g. if
inf ormation and knowledge on the substance increases during the assessment.

A particular example is that f or bioaccumulation, simultaneous testing at low concentration of

several constituents each below its water solubility and sampling and analysis of their
concentration in water and in the organism (f ish), if technically f easible, may be a cost ef ficient
testing option. The approach may also be applied in the dietary bioaccumulation study. It may
be employed on separate f ractions or blocks – or in some cases even on the whole substance.
A prerequisite f or obtaining reliable results is that the co-occurrence of each constituent does
not interf ere with the bioaccumulation behaviour of other constituents also being tested (e.g.
through enzyme induction, etc.)

Choice of the assessment approach

Finally, taking all the above into account including the substance composition and technical
f easibility and suitability of the presented approaches f or a particular substance the most
appropriate approach f or the assessment should be chosen. The choice of the assessment
approach may be dependent on the data already available. In any case, r esults f rom relevant
studies carried out by using the whole (registered) substance as test item should always be
included into the dataset, where these are already available, regardless of the assessment
approach chosen. Such results may in some cases support prof iling of the substance, even in
such cases where the “whole substance approach” will not be chosen as the main assessment
approach f or the case. Additionally, readily available test results on individual constituents
need to be taken into account in the assessment even if the “whole substance approach” is
applied. In such cases the results on individual constituents need to support the choice of the
 Chapter R.11: PBT/vPvB assessment

166 Version 4.0 – December 2023

“whole substance approach”. If they do not support the use of the “whole substance
approach”, another approach would need to be considered.

R.[Link].3 Specific aspects

The PBT/vPvB properties of a substance are evaluated in a stepwise approach, starting with P -
assessment. Depending on the assessment approach chosen, the PBT/vPvB assessment
generally f ocusses on individual constituents or groups of constituents/f ractions (see Section
R.[Link].2).

When assessing P, B and T it is important to understand that there is a dif f erence in testing
and interpretation of the data, that relates to the concentration of the test substance and that
this has consequences f or the assessment of substances containing various constituents. For
degradation (hence persistence) and bioaccumulation, the concentration of the substance in
the test vessel is not included within the measure of the endpoint (Mackay et al., 2001). This is
not the case f or toxicity which is expressed in terms of concentration. The impact this has
when assessing P, B and T is discussed under each of the endpoints below.

When evaluating P, B and T -related studies it is important to pay attention to the available
physico-chemical data and its representativeness. For example, a water solubility or K ow test
carried out with the whole substance where whole substance –related analytics has been
f ollowed does not give inf ormation on the specif ic water solubility or K ow of individual
constituents, in case these have dif ferent properties (due to structural dif ferences). Theref ore,
the basic physico-chemical data may also need to be generated f or the constituents or
constituent f ractions depending on the assessment approach chosen, before other results can
be evaluated or f urther testing decided.

QSARs-prof iling, where applicable, is of ten crucial f or the screening assessment of the
potential properties of expected constituents and hence f or the search f or the worst case
f ractions/constituents which can be targeted f or f urther assessment and testing. QSAR results
of P, B, T and relevant physico-chemical properties of the expected constituents or
representatives of f ractions of ten have important role in justif ying selected assessment
approach and test items. It should be remembered that individual QSAR-model predictions are
not normally able to accommodate the multi-constituent nature of a substance, but they
represent the results f or a particular chemical structure (i.e, f or one selected constituent at a
time). Otherwise, f or the use of QSARs in the assessment of constituents the same principles
apply as f or the use of QSARs in the assessment of pure (i.e. mono-constituent) chemical
substances.

The f ollowing specif ic considerations on data interpretation take as prerequisite that there is
dif f erentiation between the test item and the registered substance (of course, in the whole
substance-approach these are the one and same).

Where new data are generated f or a f raction profiling or known constituent-approach, it should
be kept in mind that the most persistent constituent may not be the most bioaccumulative or
toxic – and vice versa.

i) Persistence

One cannot easily assess the persistence of complex substances that contain many
constituents using biodegradation testing methods that measure parameters (e.g. CO 2
 Chapter R.11: PBT/vPvB assessment

167 Version 4.0 – December 2023

evolution), since these tests measure the properties of the whole substance but do not provide
inf ormation on the individual constituents.

If the selected test item consists of suf f iciently similar structures and is shown to meet the
stringent ultimate ready biodegradation test criterion (>60% in 28 days), it can be concluded
that the underlying constituents comprising the complex substances are not expected to be
persistent (OECD, 2001).

If the test item composition does not consist of similar structures or is not well characterised,
it may still contain a certain amount of constituents that are persistent although the amount of
easily degradable constituents is high enough to lead to an overall degradation percentage
suf f icient to meet the criteria f or ready biodegradation.

ii) Potential f or Bioaccumulation

Similar dif f iculties apply to bioaccumulation assessment.

Estimates f or the individual constituents based on Kow , QSARs or other methods may be used.
Also multi-component measuring techniques such as SPME or HPLC could be usef ul to give an
initial estimate of bioaccumulation potential. For example, if all the peaks in the HPLC
chromatogram have a log K ow <4.5, it may be assumed that all constituents of the substance
have log Kow < 4.5. For interpretation of such results and estimates, please see Section
R.[Link].

iii) Toxicity

Extrapolating toxicity data f rom the whole substance to its constituents is of ten challenging
and may lead to under- or overestimation of the toxicity of an individual constituent. Toxicity is
def ined via a concentration response and is dependent on the bioavailability. If the tested
substance contains many constituents having dif f erences in the response and bioavailability,
this makes the interpretation very dif ficult. For example, the physical f orm may prevent the
dissolution of the individual constituents of such a substance to any signif icant extent where
the whole substance is applied directly, as required in normal ecotoxicity test guidelines, to the
test medium. The apparent exposure concentration(s)/dose(s) in the test system may lead to
incorrect interpretation on toxicity of individual constituents. Theref ore, care should be taken
to interpret the observed (lack of ) ef f ect(s) in relation to actual exposure
concentration(s)/dose(s) of individual constituents.

In the event that a constituent/f raction is concluded as P and subsequently as B, then a

toxicity assessment must be performed on that same constituent/f raction f or which P and B
are also evaluated.

Bef ore starting toxicity testing on relevant constituents/fractions of the substance, all available
inf ormation, including results of testing with whole substance, should be gathered and
assessed if either T or not T conclusion can be reached in Weight-of-evidence approach for the
relevant constituents/fractions. Weight-of-evidence to support the assessment may include
assessment of all available in vivo and in vitro (e.g. mutagenicity) inf ormation on the
substance and individual constituent(s), similarity of the toxic ef f ect among constituents,
physical-chemical properties, similar substances including read-across, other substances
containing the same constituent(s), structural alerts f or (eco)toxicity, mode of action and
QSAR.
 Chapter R.11: PBT/vPvB assessment

168 Version 4.0 – December 2023

When (eco)toxicity inf ormation is available f or a whole substance containing multiple

constituents (f or instance via WAF f or aquatic toxicity, see Section R.[Link].5), the results
may not be adequate f or the PBT-assessment of individual constituents/fraction. During whole
substance testing, f or example the solubility and bioavailability of the constituent/f raction of
interest may have been inf luenced by other constituents in the substance and the bioavailable
composition may vary over time during the exposure period, resulting in additional
interpretational uncertainty. In cases where a whole substance approach f or (eco)toxicity is
justif ied, whole substance data can be directly compared to the T-criteria, and can be used to
derive conclusions on the T properties f or PBT-assessment purposes.

(Eco)toxicity data f or a whole susbtance can be used when it can be determined with sufficient
certainty which constituent(s)/f raction(s) (and to which extent) contribute to the observed
toxicity. Assessment must be accompanied with adequate justif ication to link both the
presence or absence of the toxicity of the whole substance to the constituent/f raction.
Inf ormation on whole substance f or concluding T of constituents has been used f or example for
EC 401-850-9 .
72

If no def initive conclusion f or the constituent/f raction can be drawn, f rom the available
inf ormation on the whole substance or other available inf ormation, f urther constituent specif ic
experimental inf ormation may be needed. The generation of new relevant inf ormation should
be started whenever possible by testing with non-vertebrate organisms. Generation of new
inf ormation on vertebrate toxicity f or the constituent must be avoided, if possible.

If there is a need f or new experimental inf ormation, it is recommended to reflect if the aquatic
toxicity is relevant to be tested f irst taking into account indications f or low aquatic toxicity or
concern f or relevant human health endpoints. F or aquatic toxicity testing the procedures
described in Section R.[Link] can be f ollowed. As described in the ITS f or aquatic toxicity,
vertebrate animal testing should be generally minimised by f irst testing non-vertebrate
species, if relevant.

To avoid unnecessary animal testing, no f urther constituent specific inf ormation on mammalian
toxicity is needed f or the PBT assessment if the constituent is concluded vPvB or if the
constituent is concluded T based on inf ormation on ecotoxicity or if based on case -by-case
Weight-of-Evidence assessment as described above the conclusion on T can be reached. If
based on the Weight-of-Evidence assessment a conclusion on T is still not possible to reach f or
a constituent f ulfilling P and B criteria, the need of generating new inf ormation on vertebrate
toxicity f or that constituent must be caref ully considered.

In case the whole substance is classif ied as carcinogenic (1A or 1B), mutagenic (1A or 1B),
toxic to reproduction (1A, 1B or 2) or STOT RE (1 or 2) but the contribution of the PB-
constituent on the classif ication cannot be conf irmed in the Weight-of-Evidence assessment,
the Registrant could also consider treating the substance as if it is a PBT/vPvB substance in
accordance with the conditions set out in Annex XIII, section 2.1.

72
SVHC Support document – EC 410-850-9
 Chapter R.11: PBT/vPvB assessment

169 Version 4.0 – December 2023

R.[Link].4 Test items

If new testing is considered necessary, the set of tests, test sequence and test item(s) should
be determined so that the results serve in the most ef f icient way the assessment with the
chosen approach.

The test items are allowed to deviate f rom the registered UVCB substance, if that is justif ied by
the selected assessment approach. It should be noted, that the test item(s) may
itself /themselves be UVCB(s), well-defined multi-constituent substance(s) or mono-constituent
substance(s), depending on the case and purpose.

The choice of the test item(s) is always dependent on the type of the substance but also on
the case-specific understanding of which testing strategy is most ef f icient to conclude on the
PBT/vPvB properties. Furthermore, f easibility and proportionality of efforts may also play a role
in selecting the test item. It may in some cases be necessary to run a test on a particular
property, e.g., simulation degradation test, f or several test items, where one or more test
items per f raction are used in parallel or in sequential tests.

In the “known constituent–approach” the test item consists of a single chemical structure. It
can be extracted f rom the substance itself or be a separately synthesised as surrogate f or a
constituent (a similar chemical substance to the constituent). In block method the test item(s)
per block targeted f or testing and assessment may consist of one or more substances which
are present as constituent(s) in the block or surrogate substances. Test item of a block may
also be the whole block or similar multi-constituent substance. In the other f raction prof iling
approaches, the test item is either the whole f raction itself or a f raction of the f raction hence
always consisting of multiple constituents. In that case also, the test item can be extracted
f rom the substance or be separately synthesised. Similarly, also in f raction prof iling, the test
item may be a representative multi-constituent substance/mixture, if no extraction or
synthesis of the target f raction of the registered substance is f easible.

Justif ication of test item selection should also be documented in the CSR or authority’s
assessment report.

The choice of the assessment approach and the test item may in some cases also af f ect the
selection of the test method. For instance an aqueous BCF study can only in practice be
perf ormed with a substance where exposure concentration of constituents can be verif ied by
measurements. Any uncertainty due dif ferences in constituent properties of a test item (e.g.,
such as increased leaching of test substance f rom f ood pellets due to variation in physchem
properties) need to be considered when interpreting the results. For this purpose a GC-
characterisation of the test substance in the test system and/or in dif f erent test system
matrixes bef ore, during and af ter the test has been conducted might be usef ul.
 Chapter R.11: PBT/vPvB assessment

170 Version 4.0 – December 2023

R.11.5 References
Adolf sson-Erici M, Kerman GA, and McLachlan, MS (2012) Measuring bioconcentration f actors
in f ish using exposure to multiple chemicals and internal benchmarking to correct f or growth
dilution. Environ Toxicol Chem 31:1853–1860.
Armitage JM, Erickson RJ, Luckenbach T, Ng CA, Prosser RS, Arnot JA, Schirmer K, Nichols JW
(2017) Assessing the bioaccumulation potential of ionizable organic compounds: current
knowledge and research priorities. Environmental Toxicology and Chemistry, p. 882–897,
2017.
Armitage JM, Toose L, Embry M, Foster KL, Hughes L, Sangion A, Arnot JA. 2021. The
Bioaccumulation Assessment Tool (BAT) Version 2.02. ARC Arnot Research and Consulting
Inc., Toronto, ON, Canada.
Arnot JA and Gobas FAPC (2006) A review of bioconcentration f actor (BCF) and
bioaccumulation f actor (BAF) assessments f or organic chemicals in aquatic organisms .
Environmetnal Reviews 14(4): 257-297. [Link]
Arnot JA, Brown TN and Wania F (2014) Estimating screening-level organic chemical half -lives
in humans. Environ Sci Technol 48:723-730.
Arts MT, Ferguson ME, Glozier NE, Robarts RD, Donald DB (1995) Spatial and temporal
variability in lipid dynamics of common amphipods: assessing the potential f or uptake of
lipophilic contaminants. Ecotoxicology 4(2):91-113.
ASTM (2003) E1022-94. Standard Guide f or Conducting Bioconcentration Tests with Fishes and
Saltwater Bivalve Mollusks. ASTM International, West Conshohocken, PA, USA.
ASTM International (2020) Standard Test Method f or Measuring the Toxicity of Sediment -
Associated Contaminants with Freshwater Invertebrates E1706-20.
Badry, A., Rüdel, H., Göckener, B., Nika, M.-C., Alygizakis, N., Gkotsis, G., Thomaidis, N. S.,
Treu, G., Dekker, R. W. R. J., Movalli, P., Walker, L. A., Potter, E. D., Cincinelli, A., Martellini,
T., Duke, G., Slobodnik, J., & Koschorreck, J. (2022). Making use of apex predator sample
collections: an integrated workflow f or quality assured sample processing, analysis and digital
sample f reezing of archived samples. Chemosphere, 309, 136603.
[Link]

Badry, A., Rüdel, H., Göckener, B., Nika, M.-C., Alygizakis, N., Gkotsis, G., Thomaidis, N. S.,
Treu, G., Dekker, R. W. R. J., Movalli, P., Walker, L. A., Potter, E. D., Cincinelli, A., Martellini,
T., Duke, G., Slobodnik, J., & Koschorreck, J. (2022a). Making use of apex predator sample
collections: an integrated workflow f or quality assured sample processing, analysis and digital
sample f reezing of archived samples. Chemosphere, 309, 136603.
[Link]
Badry, A., Slobodnik, J., Alygizakis, N., Bunke, D., Cincinelli, A., Claßen, D., Dekker, R. W. R.
J., Duke, G., Dulio, V., Göckener, B., Gkotsis, G., Hanke, G., Jartun, M., Movalli, P., Nika, M.-
C., Rüdel, H., Thomaidis, N. S., Tarazona, J. v, Tornero, V., Koschorreck, J. (2022b). Using
environmental monitoring data f rom apex predators f or chemicals management: towards
harmonised sampling and processing of archived wildlif e samples to increase the regulatory
uptake of monitoring data in chemicals management. Environmental Sciences Europe, 34(1),
81. [Link]
 Chapter R.11: PBT/vPvB assessment

171 Version 4.0 – December 2023

Barber MC (2008) Dietary uptake models used f or modeling the bioaccumulation of organic
contaminants in f ish. Environ Toxicol Chem 27(4):755-77.

Baskaran S, Lei YD, and Wania F (2021a) A database of experimentally derived and estimated
octanol-air partition ratios (KOA). J. Phys. Chem. Ref . Data 50, 043101 (2021).
[Link]
Baskaran S, Lei YD, and Wania F (2021b) Reliable prediction of the octanol-air partition ratio.
Environmental Toxicology and Chemistry, 40(11), 3166–3180.
[Link]
Birch H, Hammershøj R, Torsbjerg Møller M, Mayer P (2023) Technical guidance on
biodegradation testing of difficult substances and mixtures in surf ace water. MethodsX, 10:
102138. [Link]
Bleeker EAJ and Verbruggen EMJ (2009) Bioaccumulation of polycyclic aromatic hydrocarbons
in aquatic organisms. RIVM report 601779002/2009. National Institute f or Public Health and
the Environment, Bilthoven, The Netherlands.
Böhm L, Schlechtriem C, Düring RA (2016) Sorption of Highly Hydrophobic Organic Chemicals
to Organic Matter Relevant f or Fish Bioconcentration Studies. Environmental Science &
Technology 2016 50 (15), 8316-8323. DOI: 10.1021/[Link].6b01778
Borga K, Kidd KA, Muir DCG, Berglund O, Conder JM, Gobas FAPC, Kucklick J, Malm O and
Powell DE (2012) Trophic magnification factors: Considerations of ecology, ecosystems, and
study design. Integr Environ Assess Manag 8:64–84.
Braune BM, Gaston AJ, Elliott KH, Provencher JF, Woo KJ, Chambellant M, Ferguson SH,
Letcher RJ. Organohalogen contaminants and total mercury in f orage f ish preyed upon by
thick-billed murres in northern Hudson Bay. Mar Pollut Bull. 2014 Jan 15;78(1-2):258-66. doi:
10.1016/[Link].2013.11.003. Epub 2013 Nov 20. PMID: 24268851.
Braune BM, Gaston AJ, Hobson KA, Gilchrist HG, Mallory ML (2014b) Changes in f ood web
structure alter trends of mercury uptake at two seabird colonies in the Canadian Arctic.
Environmental Science and Technology 48: 13246-13252.
Brooke D and Crookes M (2012) Depuration rate constant: growth correction and use as an
indicator of bioaccumulation potential. UK Environment Agency, Bristol, UK. Product code LIT
7371. ISBN: 978-1-84911-283-3.
Brown DM, Lyon D, Saunders DMV, Hughes CB, Wheeler JR, Shen H, Whale G (2020)
Biodegradability assessment of complex, hydrophobic substances: Insights f rom gas -to-liquid
(GTL) f uel and solvent testing. Science of the Total Environm ent 727 (2020) 138528.
Burkhard LP, Arnot JA, Embry MR, Farley KJ, Hoke RA, Kitano M, Leslie HA, Lotuf o GR,
Parkerton TF, Sappington KG, Tomy GT, Woodburn KB. (2012) Comparing laboratory and f ield
measured bioaccumulation endpoints. Integr Environ Assess Manag. 8(1):17-31. doi:
10.1002/ieam.260.
Burkhard LP, Borgå K, Powell DE, Leonards P, Muir DCG, Parkerton TF, Woodburn KB (2013)
Improving the quality and scientif ic understanding of Trophic magnif ication f actors (TMFs).
Env. Sci. technol. 47: 1186-1187.
Burkhard LP, Cook PM and Lukasewycz MT (2010) Direct application of biota-sediment
accumulation f actors. Environmental Toxicology and Chemistry; 29: 230-236.
 Chapter R.11: PBT/vPvB assessment

172 Version 4.0 – December 2023

Castro-Jiménez J and Van de Meent D (2011) Accounting f or photodegradation in P -

assessment of chemicals. Reports Environmental Science no 381. Radboud University
Nijmegen. Radboud University Nijmegen. The Netherlands. 45 pp.
Cavaletto JF, Gardner WS. (1999) Seasonal Dynamics of Lipids in Freshwater Benthic
Invertebrates. In: Arts MT, Wainman BC (eds) Lipids in Freshwater Ecosystems. Springer, New
York, NY.
Claßen, D., Ackermann, J., & Schaef f er, A. (2021) Fate and behavior of 14C-labelled ionic
compounds in a soil simulation test. Science of The Total Environment, 768: 144970.
CONCAWE (2001) Environmental classification of petroleum substances - summary data and
rationale report no. 01/54, Prepared by CONCAWE Petroleum Products Ecology Group
CONCAWE, Brussels, Oct 2001. [Link]
[Link]
Corbin, M, Eckel, W., Ruhman, M., Spatz, D., Thurman, N., Gangaraju, R., Kuchnicki, T.
Mathew, R. & Nicholson, I. 2006. Costanza J, Lynch DG, Boethling RS and Arnot JA (2012) Use
of the bioaccumulation f actor to screen chemicals f or bioaccumulation potential. Environmental
Toxicology and Chemistry 31:2261-2268.
Czub G and McLachlan MS (2004) Bioaccumulation potential of persistent organic chemicals in
humans. Environmental Science and Technology 38: 2406-2412.
Debruyn AMH and Gobas FAPC (2006) A bioenergetic biomagnif ication model f or the animal
kingdom. Environ Sci Technol 40:1581-7.
Deneer JW, Adriaanse PI, van Griethuysen C and Boesten JJTI (2015) Estimation of
degradation rates in cosm water. Guidance f or inverse modelling using TOXSWA. Alterra
Wageningen. Alterra report 2679, ISSN 1566-7197. Available at: [Link]
Dimitrov SD, Georgieva DG, Pavlov TS, Karakolev YH, Karamertzanis PG, Rasenberg M and
Mekenyan OG (2015) UVCB substances: methodology f or structural description and application
to f ate and hazard assessment. Environ Toxicol Chem 34:2450-62.
Ding L, Zhang S, Zhu Y, Zhao N, Yan W, Li Y. Overlooked long-chain chlorinated paraf f in
(LCCP) contamination in f oodstuf f f rom China. Sci Total Environ. 2021 Dec 20;801:149775.
doi: 10.1016/[Link].2021.149775
EC (2012) Commission Communication of 31.5.2012: The combination ef f ects of chemicals,
Chemical mixtures, COM(2012) 252 f inal, available at:
[Link]
ECB (2003) Technical Guidance Document on Risk Assessment in support of Commission
Directive 93/67/EEC on Risk Assessment f or new notified substances, Commission Regulation
(EC) No 1488/94 on Risk Assessment f or existing substances, Directive 98/8/EC of the
European Parliament and of the Council concerning the placing of biocidal products on the
market, Part II. Available at:
[Link]
ECETOC (2005) Alternative Testing Approaches in Environmental Safety Assessment, Technical
Report No 97, European Centre f or Ecotoxicology and Toxicology of Chemicals, Brussels,
Belgium. Available at: [Link]
ECETOC (2013) Understanding the Relationship between Extraction Technique and
Bioavailability, Technical Report No.117. 87 p. Available at: [Link]
 Chapter R.11: PBT/vPvB assessment

173 Version 4.0 – December 2023

content/uploads/2021/10/ECETOC-TR-117-Understanding-the-relationship-between-
[Link]
ECETOC (2014) Inf ormation to be considered in a weight-of -evidence-based PBT/vPvB
assessment of chemicals (Annex XIII of REACH). Special report no. 18. 158 p. Available at:
[Link]
ECHA (2015) Member State Committee Opinion on persistency and bioaccumulation of
Octamethylcyclotetrasiloxane (D4) EC Number: 209-136-7 CAS Number: 556-67-2 and
Decamethylcyclopentasiloxane (D5) EC Number: 208-764-9 CAS Number: 541-02-6. Available
at: [Link]
3c_msc_opinion_on_d4_and_d5_20150422_en.pdf
EFSA (2014) EFSA Guidance Document f or evaluating laboratory and f ield dissipation studies to
obtain DegT50 values of active substances of plant protection products and transf ormation
products of these active substances in soil. EFSA Journal 2014; 12(5):3662, 67 pp. Available at:
[Link] [Link]/f r/ef sajournal/pub/3662
Elliott KH and Elliott JE (2016) Origin of sulf ur in diet drives spatial and temporal mercury
trends in seabird eggs f rom pacific Canada 1968−2015. Environmental Science and Technology
50: 13380-13386.
Elliott KH, Braune BM and Elliott JE (2021) Beyond bulk δ15N: Combining a suite of stable
isotopic measures improves the resolution of the f ood webs mediating contaminant signals
across space, time and communities. Environment International 148: 106370.
Environment Agency (2009) Calculation of molecular dimensions related to indicators f or low
bioaccumulation potential, Environment Agency Science Report SCHO0109BPGT-E-E. ISBN
978-1-84432-978-6. Available at: [Link]
[Link]/pdf /[Link]
Environment Agency (2022) Lipid 173ormalization in the OECD 305 dietary test. Environment
Agency, Bristol.
Environment Canada (2013) Biological Test Method: Test f or Survival and Growth in Sediment
and Water Using the Freshwater Amphipod Hyalella azteca EPS 1/RM/33 Second Edition.
European Commission (2014) Common implementation strategy f or the Water f ramework
Directive (2000/60/EC), Guidance document No. 32 on biota monitoring (the implementation
of EQSbiota) under the Water Framework Directive. Technical Report - 2014 – 083. ISBN 978-
92-79-44634-4 doi: 10.2779/833200.
European Commission (2018) Guidance document No. 27, updated version 2018. Technical
Guidance f or Deriving Environmental Quality Standards. Document endorsed by EU Water
Directors at their meeting in Sof ia on 11-12 June 2018. Available at:
[Link]
9964bbe8312d/library/ba6810cd-e611-4f 72-9902-f 0d8867a2a6b/details?download=true
Fay KA, Mingoia RT, Goeritz I, Nabb DL, Hof f man AD, Ferrell BD, Peterson HM, Nichols JW,
Segner H, Han X. Intra- and interlaboratory reliability of a cryopreserved trout hepatocyte
assay f or the prediction of chemical bioaccumulation potential. Environ Sci Technol. 2014 Jul
15;48(14):8170-8. doi: 10.1021/es500952a. Epub 2014 Jul 2. PMID: 24941452.
Fenner K, Stamm C, Bischof f F, Honti M, Varga L (2017) Suitability of laboratory simulation
tests f or the identif ication of persistence in surf ace waters. Final Report. Environmental
Research of the Federal Ministry f or the Environment, Nature Conservation, Building and
 Chapter R.11: PBT/vPvB assessment

174 Version 4.0 – December 2023

Nuclear Saf ety, TEXTE 40/2017. ISSN 1862-4804

[Link] ault/f iles/medien/1410/publikationen/2017-05-
19_texte_40-2017_p-[Link] .
Feron VJ and Groten JP (2002) Toxicological evaluation of chemical mixtures. Food Chem
Toxicol 40:825–39.
Fisk AT, Hobson KA, Norstrom RJ (2001) Inf luence of chemical and biological f actors on trophic
transf er of persistent organic pollutants in the northwater polynya marine f ood web. Environ.
Sci. Technol. 35: 732–738.
Florin ST, Felicetti LA and Robbins CT (2011) The biological basis f or understanding and
predicting dietary-induced variation in nitrogen and sulphur isotope ratio discrimination.
Functional Ecology 25: 519-526.
FOCUS (2014). Generic guidance f or Estimating Persistence and Degradation Kinetics f rom
Environmental Fate Studies on Pesticides in EU Registration. Report of the FOCUS Work Group
on Degradation Kinetics. 18 December 2014. Version 1.1, 440 pp. Available at:
[Link]
[Link]
Fremlin KM, Elliott JE, Martin PA, Harner T, Saini A, Gobas FAPC (2021) Fugacity-Based Trophic
Magnif ication Factors Characterize Bioaccumulation of Cyclic Methyl Siloxanes within an Urban
Terrestrial Avian Food Web: Importance of Organism Body Temperatur e and Composition.
Environ. Sci. Technol. 55:20, 13932-13941.
Fu Q, Scheidegger A, Laczko E, and Hollender J. (2021) Metabolomic Prof iling and
Toxicokinetics Modeling to Assess the Ef fects of the Pharmaceutical Diclof enac in the Aquatic
Invertebrate Hyalella azteca. Environmental Science & Technology 55 (12), 7920-7929. DOI:
10.1021/[Link].0c07887
Gartiser S, Brunswick-Titze A, Flach F, Junker T, Sättler, S and Jöhncke U. (2022) Enhanced
ready biodegradability screening tests for the evaluation of potential PBT substances. Science
of the Total Environment 833:155134. [Link]
Girling, A. E., Markarian, R. K., & Bennett, D. (1992). Aquatic toxicity testing of oil products -
some recommendations. Chemosphere, 24(10), 1469–1472. [Link]
6535(92)90268-V
Glüge J, Escher BI, Scheringer M (2022) How error-prone bioaccumulation experiments af f ect
the risk assessment of hydrophobic chemicals and what could be improved. Integr Environ
Assess Manag. [Link]
Gobas FAPC, Kelly BC, Arnot JA. (2003). Quantitative Structure Activity Relationships f or
Predicting the Bioaccumulation of POPs in Terrestrial Food-Webs. QSAR Comb. Sci. 22: 346-
351.
Gobas FA, de Wolf W, Burkhard LP, Verbruggen E and Plotzke K (2009) Revisiting
bioaccumulation criteria f or POPs and PBT assessments. Integr Environ Assess Manag 4:624-
37.
Gobas FAPC, Burkhard LP, Doucette WJ, Sappington KG, Verbruggen EMJ, Hope BK, Bonnell
MA, Arnot JA, Tarazona JV (2015) Review of existing terrestrial bioaccumulation models and
terrestrial bioaccumulation modeling needs f or organic chemicals. Health & Ecological Risk
Assessment 12(1): 123-134.
 Chapter R.11: PBT/vPvB assessment

175 Version 4.0 – December 2023

Gobas FAPC, Lee YS (2019) Growth-Correcting the Bioconcentration Factor and

Biomagnif ication Factor in Bioaccumulation Assessments. Environ Toxicol Chem. 38(9):2065-
2072. doi: 10.1002/etc.4509
Gobas FAPC, Lee YS, Arnot JA (2021) Normalizing the biomagnification factor. Environ Toxicol
Chem 40, 1204-1211.
Gobas FAPC, Lee YS, Fremlin KM, Stelmachuk SC, Redman AD. (2023) Methods f or assessing
the bioaccumulation of hydrocarbons and related substances in terrestrial organisms: A critical
review. Integr Environ Assess Manag. doi: 10.1002/ieam.4756. Epub ahead of print. PMID:
36880196.
Gomez CF, Constantine L, Huggett DB (2010) The inf luence of gill and liver metabolism on the
predicted bioconcentration of three pharmaceuticals in f ish. Chemosphere 81(10):1189-1195
doi:[Link]

Goss KU, Brown TN and Endo S (2013) Elimination half -lif e as a metric f or the bioaccumulation
potential of chemicals in aquatic and terrestrial f ood chains. Environ Toxicol Chem 32:1663-71.
Goss, KU, Linden L, Ulrich N, and Schlechtriem C (2018). Revisiting elimination half live as an
indicator f or bioaccumulation in f ish and terrestrial mammals. Chemosphere, 210, 341–346.
[Link]
Guhl W. and Steber J, (2006). The value of biodegradation screening test results f or predicting
the elimination of chemicals’ organic carbon in waste water treatment plants . Chemosphere,
63(1):9-16. [Link]
Gustaf son, D. I., & Holden, L. R. (1990). Nonlinear pesticide dissipation in soil: a new model
based on spatial variability. Environmental Science & Technology, 24(7), 1032–1038.
[Link]
Hashizume N, Inoue Y, Suzuki Y, Murakami H, Sumi S, Ishibashi T, Yoshida T (2018)
Comparison of laboratory-derived biomagnification factors f or hexachlorobenzene in common
carp conducted under 9 test conditions. Environ Toxicol Chem. 37(4):1032-1039. doi:
10.1002/etc.4030.
Hebert CE, Hobson KA, Shutt JL (2000) Changes in f ood web structure af f ect rate of PCB
decline in herring gull (Larus argentatus) eggs. Environmental Science and Technology 34:
1609-1614.
Hennecke D, Bauer A, Shrestha P, Junker T, Fenner, K (2014) Improved test system to
determine chemical degradation in laboratory water/sediment systems – experimental results.
SETAC Europe 24th Annual Meeting, May 2014, Basel, Switzerland. [Link]
content/uploads/2014/03/[Link]
Hobson KA, Alisauskas RT, Clark RG (1993) Stable-nitrogen isotope enrichment in avian
tissues due to f asting and nutritional stress: Implications f or isotopic analyses of diet. Condor
95: 388-394.
Hodges G, Eadsf orth C, Bossuyt B, Bouvy A, Enrici MH, Geurts M, Kotthof f M, Michie E, Miller
D, Müller J, Oetter G, Roberts J, Schowanek D, Sun P, and Venzmer J (2019). A comparison of
log K ow (n-octanol–water partition coef f icient) values f or non-ionic, anionic, cationic and
amphoteric surf actants determined using predictions and experimental methods.
Environmental Sciences Europe, 31(1), 1–18. [Link]
 Chapter R.11: PBT/vPvB assessment

176 Version 4.0 – December 2023

Hof er T; Myhre O; Peltola-Thies J; Hirmann D (2021) Analysis of elimination half -lives in

MamTKDB 1.0 related to bioaccumulation: Requirement of repeated administration and blood
plasma values underrepresent tissues. Environment international 155, 106592. DOI:
[Link]
Hoke RA, Huggett D, Brasfield S, Brown B, Embry M, Fairbrother A, Kivi M, Leon Paumen M,
Prosser R, Salvito D and Scroggins R (2015) Review of Laboratory -Based Terrestrial
Bioaccumulation Assessment Approaches f or Organic Chemicals: Current Status and Future
Possibilities. Integr Environ Assess Manag 12:109-22.
Hollender J, Bavel Bv, Dulio V, Farmen E, Furtmann K. Koschorrek J, Kunkel U, Krauss M,
Munthe J, Schlabach M, Slobodnik J, Stroomberg G, Termes T, Thomaidis NS, Togola A,
Tornero V (2019) High resolution mass spectrometry-based non-target screening can support
regulatory environmental monitoring and chemicals management. Environ Sci Eur 31:42.
[Link]
Holzmann, H, Claßen, D, Ackermann, J, Schäffer, A. (2022) Fate of 14c-Labelled Ionic Organic
Chemicals in a Water-Sediment System and Surf ace Water. Chemosphere 303, 134885.
[Link]
Honti and Fenner (2015) Deriving Persistence Indicators f rom Regulatory Water -Sediment
Studies − Opportunities and Limitations in OECD 308 Data. Environ Sci Technol 49:5879−86.
Houde M, Muir DCG, Kidd KA, Guildf ord S, Drouillard K, Evans MS, Wang X, Whittle DM,
Haf f ner D and Kling H (2008) Inf luence of lake characteristics on the biomagnif ication of
persistent organic pollutants in lake trout f ood webs. Environ Toxicol Chem 27:2169-78.
Huf f Hartz K, Weston D, Johanif N, Poynton H, Connon R, Lydy M (2021) Pyrethroid
bioaccumulation in f ield-collected insecticide-resistant Hyalella azteca. Ecotoxicology 30:514–
523.
Ingerslev F and Nyholm N (2000) Shake-Flask Test f or Determination of Biodegradation Rates
of 14C-Labeled Chemicals at Low Concentrations in Surf ace Water Systems. Ecotoxicol Environ
Saf 45:274-83.
Ingerslev, F., Toräng, L., & Nyholm, N. (2000). Importance of the test volume on the lag
phase in biodegradation studies. Environmental Toxicology and Chemistry, 19(10), 2443–
2447. [Link]
Inoue Y, Hashizume N, Yoshida T, Murakami H, Suzuki Y, Koga Y, Takeshige R, Kik ushima E,
Yakata N, Otsuka M (2012) Comparison of Bioconcentration and Biomagnif ication Factors f or
Poorly Water-Soluble Chemicals Using Common Carp (Cyprinus carpio L.) Arch Environ Contam
Toxicol 63:241–248 DOI 10.1007/s00244-012-9761-8.
IPCS (2010) Characterization and application of physiologically based pharmacokinetic models
in risk assessment. Harmonization project document no. 9. WHO Press, Geneva, Switzerland.
ISBN 978-92-4-150090-6 (available at
[Link] )
Jurgens M, Johnson A, Jones K, Hughes D and Lawlor A (2013) The presence of EU priority
substances mercury, hexachlorobenzene, hexachlorobutadiene and PBDEs in wild f ish f rom
f our English rivers. Science of the Total Environment 461-462, 441-452.
Kelly BC, Gobas FAPC. 2001. Bioaccumulation of persistent organic pollutants in lichen-
caribou-wolf f ood chains of Canada's Central and Western Arctic. Environ Sci Technol 35:325–
334.
 Chapter R.11: PBT/vPvB assessment

177 Version 4.0 – December 2023

Kelly BC, Gobas FAPC. 2003. An arctic terrestrial f ood-chain bioaccumulation model f or
persistent organic pollutants. Environ Sci Technol 37:2966–2974.
Kelly BC, Ikonomou MG, Blair JD, Morin AE and Gobas FAPC (2007) Food web-specif ic
biomagnif ication of persistent organic pollutants. Science 317: 236-239.
Kelly BC, Ikonomou MG, Blair JD, Surridge B, Hoover D, Grace R, and Gobas FAPC (2009)
Perf luoroalkyl Contaminants in an Arctic Marine Food Web: Trophic Magnif ication and Wildlif e
Exposure Environmental Science & Technology 43 (11), 4037-4043. DOI: 10.1021/es9003894
Kelly JF (2000) Stable isotopes of carbon and nitrogen in the study of avian and mammalian
trophic ecology. Canadian journal of zoology 78: 1-27.
Khairy M, Weinstein MP and Lohmann R (2014) Trophodynamic behavior of hydrophobic
organic contaminants in the aquatic f ood web of a tidal river. Environ Sci Technol
48:12533−42.
Kidd KA, Burkhard LP, Babut M, Borgå K, Muir DCG, Perceval O, Ruedel H, Woodburn K and
Embry MR (2019) Practical advice f or selecting or determining trophic magnification f actors f or
application under the European Union Water Framework Directive. Integrated Environmental
Assessment and Management 15: 266-277. DOI: 10.1002/ieam.4102
Kierkegaard A, van Egmond R, and McLachlan MS (2011) Cyclic volatile methylsiloxane
bioaccumulation in f lounder and ragworm in the humber estuary. Environmental Science and
Technology 45:5936–5942. [Link]
Kim, J., Gobas, F. A. P. C., Arnot, J. A., Powell, D. E., Seston, R. M., & Woodburn, K. B.
(2016). Evaluating the roles of biotransformation, spatial concentration dif f erences, organism
home range, and f ield sampling design on trophic magnif ication f actors. Science of The Total
Environment, 551–552, 438–451. [Link]
Kinney CA, Campbell BR, Thompson R, Furlong ET, Kolpin DW, Burkhardt MR, Zaugg SD,
Werner SL and Hay AG (2012) Earthworm bioassays and seedling emergence f or monitoring
toxicity, aging and bioaccumulation of anthropogenic waste indicator compounds in biosolids-
amended soil. Sci Total Environ 433:507-15.
Kosf eld V, Fu Q, Ebersbach I, Esser D, Schauerte A, Bischof I, Hollender J, Schlechtriem C.
(2020) Comparison of Alternative Methods f or Bioaccumulation Assessment: Scope and
Limitations of In Vitro Depletion Assays with Rainbow Trout and Bioconcentration Tests in the
Freshwater Amphipod Hyalella Azteca. Environmental Toxicology and Chemistry 39 (9) 1813–
1825.
Kosf eld V, Rüdel H, Schlechtriem C, Rauert C, Koschorreck J (2021) Food web on ice: a
pragmatic approach to investigate the trophic magnif ication of chemicals of concern.
Environmental Sciences Europe 33: 93.
Krause S, Goss K-U (2018) In Vitro–in Vivo Extrapolation of Hepatic Metabolism f or Dif f erent
Scenarios-a Toolbox. Chemical Research in Toxicology 31(11):1195-1202.
doi:10.1021/[Link].8b00187
Krause S, Goss K-U (2020) Comparison of a simple and a complex model f or BCF prediction
using in vitro biotransf ormation data. Chemosphere 256:127048
doi:[Link]
 Chapter R.11: PBT/vPvB assessment

178 Version 4.0 – December 2023

Krause S, Goss K-U (2021) Prediction of Unbound Fractions f or in Vitro–in Vivo Extrapolation of
Biotransf ormation Data. Chemical Research in Toxicology 34(1):7-11.
doi:10.1021/[Link].0c00349
Kuo, D.T.F., Rattner, B.A., Marteinson, S.C. et al. Correction: A Critical Review of
Bioaccumulation and Biotransf ormation of Organic Chemicals in Birds. Reviews
[Link] (f ormerly:Residue Reviews) 260, 7 (2022). [Link]
022-00008-8
Larisch W, Goss KU (2018) Modelling oral up-take of hydrophobic and super-hydrophobic
chemicals in f ish. Environ Sci Process Impacts 24;20(1):98-104. doi: 10.1039/c7em00495h
Laue H, Gf eller H, Jenner KJ, Nichols JW, Kern S, Natsch A (2014) Predicting the
bioconcentration of fragrance ingredients by rainbow trout using measured rates of in vitro
intrinsic clearance. Environmental science & technology 48(16):9486-9495.
Laue, H., Hostettler, L., Badertscher, R. P., Jenner, K. J., Sanders, G., Arnot, J. A., & Natsch,
A. (2020). Examining Uncertainty in In Vitro–In Vivo Extrapolation Applied in Fish
Bioconcentration Models. Environmental Science & Technology, 54(15), 9483–9494.
[Link]
Lavoie RA, Kyser TK, Friesen VL, Campbell LM (2015) Tracking overwintering areas of f ish-
eating birds to identif y mercury exposure. Environmental Science and Technology 49: 863-
872.
Lee Y-S, Lo JC, Otton SV, Moore MM, Kennedy CJ, Gobas FAPC (2017) In vitro to in vivo
extrapolation of biotransformation rates f or assessing bioaccumulation of hydrophobic organic
chemicals in mammals. Environmental Toxicology and Chemistry 36(7):1934-1946.
doi:10.1002/etc.3718
Liigand J., Wang T., Kellogg J., Smedsgaard J., Cech N., Kruve A. (2020) Quantif ication f or
non-targeted LC/MS screening without standard substances. Sci Rep 10:5808.
[Link]
Lin J, Apell J, McNeill K, Emberger M, Ciraulo V, Gimeno S (2019) A streamlined workf low to
study direct photodegradation kinetics and transformation products f or persistence assessment
of a f ragrance ingredient in natural waters. Environmental Science: Processes & Impacts 21:
1713-1721. [Link]

Lin J, Ciraulo V, Gimeno S (2022) Investigations on the direct photodegradation of a f ragrance

ingredient f or its persistence assessment in worldwide natural waters. Environmental Advances
9, 100292. [Link]

Lin J, Emberger M (2017) Study of the photodegradation of a f ragrance ingredient f or aquatic

environmental f ate assessment. Chemosphere 173: 485-493.
[Link]
Lo JC, Allard GN, Otton SV, Campbell DA, Gobas FAPC (2015) Concentration dependence of
biotransf ormation in f ish liver S9: Optimizing substrate concentrations to estimate hepatic
clearance f or bioaccumulation assessment. Environmental Toxicology and Chemistry
34(12):2782-2790. [Link]
Lombardo F, Berellini G, Obach RS (2018) Trend Analysis of a Database of Intravenous
Pharmacokinetic Parameters in Humans f or 1352 Drug Compounds. Drug Metabolism and
Disposition 46, (11), 1466-1477.
 Chapter R.11: PBT/vPvB assessment

179 Version 4.0 – December 2023

Mackay D, Celsie AKD, Powell DE and Parnis JM (2018) Bioconcentration, bioaccumulation,

biomagnif ication and trophic magnification: a modelling perspective. Environmental Science:
Processes & Impacts 20: 72-85.
Mackay D, Arnot JA, Gobas FAPC and Powell DE (2013) Mathematical relationshipos between
metrics of chemical bioaccumulation in f ish. Environ Toxicol Chem 32:1459-66.
Mackay, D., McCarty, L. S., & MacLeod, M. (2001). On the validity of classif ying chemicals f or
persistence, bioaccumulation, toxicity, and potential f or long-range transport. Environmental
Toxicology and Chemistry, 20(7), 1491–1498. [Link]
Mackay D, Paterson S, Di Guardo A, Cowan CE (1996) Evaluating the Environmental Fate of a
Variety of Types of Chemicals Using the EQC Model. Environ Toxicol Chem 15:1627- 37.
Martin JW, Mabury SA, Solomon KR and Muir DCG (2003a) Dietary accumulation of
perf luorinated acids in juvenile rainbow trout (Oncorhynchus mykiss). Environ Toxicol Chem
22:189-95.
Martin JW, Mabury SA, Solomon KR and Muir DCG (2003b) Bioconcentration and tissue
distribution of perf luorinated acids in rainbow trout (Oncorhynchus mykiss). Environ Toxicol
Chem 22:196-204.
McCord JP, Grof f LC, Sobus JR (2022) Quantitative non-targeted analysis: Bridging the gap
between contaminant discovery and risk characterization. Environment International
158:107011. [Link]
McLachlan MS, Czub G, Macleod M and Arnot J (2011) Bioaccumulation of organic
contaminants in humans: a multimedia perspective and the importance of biotransf ormation.
Environ Sci Technol 45:197–202.
McLachlan MS, Zou H, and Gouin T (2017) Using benchmarking to strengthen the assessment
of persistence. Environ Sci and Technol 51:4-11. [Link]
Mekenyan O (2006) Oasis Catabol Sof tware, Laboratory of Mathematical Science, Bourgas,
Bulgaria ([Link]
Meylan WM and Howard PH (2005) Estimating octanol–air partition coef f icients with octanol–
water partition coef f icients and Henry’s law constants, Chemosphere 61 (5), 640-644.
[Link] .
Mikkelsen J (1995) Fate Model f or Organic Chemicals in an Activated Sludge Wastewater
Treatment Plant - Modif ication of SimpleTreat. Technical Report no.2 to “Environmental
Exposure Assessment of Chemicals”. Environmental project No.308. Danish Environmental
Protection Agency. ISBN 87-7810-454-8.
Moermond CT and Verbruggen EM (2013) An evaluation of bioaccumulation data f or
hexachlorobenzene to derive water quality standards according to the EU-WFD methodology.
Integr Environ Assess Manag 9: 87-97.
Mount DR, Dawson TD, Burkhard LP, (1999). Implications of gut purging f or tissue residues
determined in bioaccumulation testing of sediment with Lumbriculus variegatus. Environmental
Toxicology and Chemistry (18),1244–1249.
NAFTA Guidance Document f or Conducting Terrestrial Field Dissipation Studies (2006).
Available at: [Link] ta-
guidance-document-conducting-terrestrial.
 Chapter R.11: PBT/vPvB assessment

180 Version 4.0 – December 2023

Nichols, J., Fay, K., Bernhard, M. J., Bischof , I., Davis, J., Halder, M., Hu, J., Johanning, K.,
Laue, H., Nabb, D., Schlechtriem, C., Segner, H., Swintek, J., Weeks, J., & Embry, M. (2018).
Reliability of In Vitro Methods Used to Measure Intrinsic Clearance of Hydrophobic Organic
Chemicals by Rainbow Trout: Results of an International Ring Trial. Toxicological sciences : an
of f icial journal of the Society of Toxicology, 164(2), 563–575.
[Link]
Nichols JW, Hof f man A D, Swintek J A, Droge STJ, Fitzsimmons PN (2021) Addition of
phenylmethylsulf onylf luoride increases theworking lif etime of the trout liver S9 substrate
depletion assay, resulting in improved detection of low intrinsic clearance rates. Environ
Toxicol Chem, 40, 148–161. [Link]

Nichols JW, Huggett DB, Arnot JA, Fitzsimmons PN, Cowan-Ellsberry CE (2013) Toward
improved models f or predicting bioconcentration of well-metabolized compounds by rainbow
trout using measured rates of in vitro intrinsic clearance. Environmental Toxicology and
Chemistry 32(7):1611-22. doi:10.1002/etc.2219
Nichols JW, Schultz IR, Fitzsimmons PN (2006) In vitro–in vivo extrapolation of quantitative
hepatic biotransf ormation data f or f ish: I. A review of methods, and strategies f or
incorporating intrinsic clearance estimates into chemical kinetic models. Aquatic Toxicology
78(1):74-90. doi:[Link]
Obach RS, Lombardo F, Waters NJ (2008) Trend Analysis of a Database of Intravenous
Pharmacokinetic Parameters in Humans f or 670 Drug Compounds. Drug Metabolism and
Disposition 36, (7), 1385-1405.
OECD (1998) Report of the OECD workshop on statistical analysis of aquatic toxicity data.
Series on testing and assessment, N°10. Environmental Health and Saf ety Publications. Series
on testing and Assessment (ENV/MC/CHEM(98)18). Available at:
[Link]
OECD (2000a) OECD Series on Testing and Assessment, Number 22, Guidance Document f or
the Performance of Out-door Monolith Lysimeter Studies (ENV/JM/MONO(2000)8). Available
at: [Link] ety/testing/series-testing-assessment-publications-
[Link]
OECD (2001) OECD Series on Testing and Assessment, Number 27, Guidance Document on
the Use of the Harmonised System for the Classif ication of Chemicals which are Hazardous f or
the Aquatic Environment (ENV/JM/MONO(2001)8). Available at:
[Link] ety/testing/series-testing-assessment-publications-
[Link]
OECD (2002) OECD Series on Testing and Assessment, Number 36, Report of the OECD/UNEP
workshop on the use of multimedia models f or estimating overall environmental persistence
and long-range transport in the context of PBTs/POPs assessment (ENV/JM/MONO(2002)15).
Available at: [Link] ety/testing/series-testing-assessment-
[Link]
OECD (2006) The OECD Pov and LRTP Screening Tool 2.0. Sof tware and Manual, OECD, Paris.
Available at: [Link] ety/risk-
assessment/[Link]
 Chapter R.11: PBT/vPvB assessment

181 Version 4.0 – December 2023

OECD (2008) OECD Guidelines f or the testing of chemicals, TG 315, Bioaccumulation in

Sediment-dwelling Benthic Oligochaetes. Organisation f or Economic Co-operation and
Development, Paris, France.
OECD (2010) OECD Guidelines f or the testing of chemicals, TG 317, Bioaccumulation in
Terrestrial Oligochaetes. Organisation f or Economic Co-operation and Development, Paris,
France.
OECD (2012a), Test No. 305: Bioaccumulation in Fish: Aqueous and Dietary Exposure, OECD
Guidelines f or the Testing of Chemicals, Section 3, OECD Publishing, Paris,
[Link]
OECD (2012b) OECD Series on Testing and Assessment, Numer 175, Validation report of a ring
test f or the OECD 305 dietary exposure bioaccumulation f ish test (Part 1) with additional
report including comparative analysis of trout and carp results (Part II). Available at:
[Link] orenviron
mentalf [Link]
OECD (2016) OECD Series on Testing and Assessment Number 232, Series on Pesticides
Number 82. Guidance Document f or Conducting Pesticide Terrestrial Field Dissipation Studies
(ENV/JM/MONO(2016)6). Available at: [Link] ety/testing/series-
[Link]
OECD (2017) OECD Series on Testing and Assessment Number 264, Guidance document on
aspects of OECD TG 305 on Fish bioaccumulation. OECD Environmental Health and Saf ety
Publications, Series on Testing and Assessment, Paris, France. Available at:
[Link] ety/testing/series-testing-assessment-publications-
[Link]
OECD (2018a) Guidance document on the determination of in vitro intrinsic clearance using
cryopreserved hepatocytes (RT-HEP) or liver S9 sub-cellular f ractions (RT-S9) f rom rainbow
trout and extrapolation to in vivo clearances
OECD (2018b) Test No. 319A: Determination of in vitro intrinsic clearance using cryopreserved
rainbow trout hepatocytes (RT-HEP). doi:[Link]

OECD (2018c) Test No. 319B: Determination of in vitro intrinsic clearance using rainbow trout
liver S9 sub-cellular f raction (RT-S9). doi:[Link]
OECD (2019) Guidance Document on Aqueous-Phase Aquatic Toxicity Testing of Dif f icult test
Chemicals, OECD Series on Testing and Assessment No 23 (second edition), OECD Publishing,
Paris. [Link] 88e-en
OECD (2023) Draf t Test Guideline Hyalella azteca Bioconcentration Test (HYBIT) available at:
[Link] ety/test-guidelines-f or-comments-section3-degradation-
[Link].
Papa E, Sangion A, Arnot JA, Gramatica P (2018) Development of human biotransf ormation
QSARs and application f or PBT assessment ref inement. Food and Chemical Toxicology 112,
535-543.
Pavan M and Worth AP (2006) Review of QSAR Models f or Biodegradation, EUR 22355 EN.
Available at: [Link] (last accessed:
December 2023)
 Chapter R.11: PBT/vPvB assessment

182 Version 4.0 – December 2023

Peijnenburg WJGM, de Beer KGM, de Haan MWA, den Hollander HA, Stegeman MHL and
Verboom H (1992) Development of a structure-reactivity relationship f or the photohydrolysis
of substituted aromatic halides. Environ Sci Technol 11:2116-21.
Peter Fisk Associates Limited (2021). Critical literature review of analytical methods applicable
to environmental f ate studies. European Chemicals Agency (ECHA).
[Link]
/b3a7e562-bf 9c-ef 02-948f -eaf 1b8f 89e3f ?t=1616414618970
Post, D. M. (2002) Using stable isotopes to estimate trophic position: models, methods, and
assumptions. Ecology 83: 703-718.
Regnery, J., Riegraf , C., Jacob, S., & Friesen, A. (2022). New insights on in vitro
biotransf ormation of anticoagulant rodenticides in f ish. Chemosphere, 294, 133727.
[Link]
Rendal, C.; Kusk, K. O.; Trapp, S. (2011) Optimal choice of pH f or toxicity and
bioaccumulation studies of ionizing organic chemicals. Environmental Toxicology and
Chemistry 30: 2395 – 2406.
Ribbenstedt, A., Armitage, J. M., Günther, F., Arnot, J. A., Droge, S. T. J., & McLachlan, M. S.
(2022). In Vivo Bioconcentration of 10 Anionic Surf actants in Rainbow Trout Explained by In
Vitro Data on Partitioning and S9 Clearance. Environmental Science and Technology, 56 (10),
6305–6314. [Link]
Roberts TR. (1984). Non-extractable pesticide residues in soils and plants. Pure and Applied
Chemistry 56: 945-956.
Sander R (2015) Compilation of Henry's law constants (version 4.0) f or water as solvent,
Atmos. Chem. Phys., 15, 4399-4981.
Saunders LJ, Hof f man AD, Nichols JW, Gobas FAPC. 2020. Dietary bioaccumulation and
biotransformation of hydrophobic organic sunscreen agents in rainbow trout. Environ Toxicol
Chem: 39:574-586. [Link]
Schlechtriem C, Böhm L, Bebon R, Bruckert HJ, Düring R (2017) Fish bioconcentration studies
with column‑generated analyte concentrations of highly hydrophobic organic chemicals.
Environmental Toxicology and Chemistry 36:4, 906–916.
Schlechtriem C, Fliedner A and Schäf ers C (2012) Determination of lipid content in f ish
samples f rom bioaccumulation studies: Contributions to the revision of OECD Test Guideline
305. Environmental Sciences Europe 24:13.
Schlechtriem C, Kampe S, Bruckert HJ, Bischof I, Ebersbach I, Kosf eld V, Kotthoff M, Schäf ers
C, L’Haridon J. (2019). Bioconcentration studies with the f reshwater amphipod Hyalella azteca:
are the results predictive of bioconcentration in f ish? Environ Sci Pollut Res 26, 1628–1641.
Schoorl M, Hollander A, van de Meent D (2016) SimpleBox 4.0: A multimedia mass balance
model f or evaluating the f ate of chemical substances. RIVM report 2015-0161, 90 pp. Available
at: [Link]
Schug H, Begnaud F, Debonneville C, Berthaud F, Gimeno S, Schrimer K (2018) TransFEr: a
new device to measure the transfer of volatile and hydrophobic organic chemicals across an in
vitro intestinal f ish cell barrier. Analytical Methods 10: 4394-4403. doi 10.1039/C8AY01253A
Schug H, Maner J, Begnaud F, Berthaud F, Gimeno S, Schirmer K, Zupanic A (2019) Intestinal
f ish cell barrier model to assess transf er of organic chemicals in vitro: an experimental and
 Chapter R.11: PBT/vPvB assessment

183 Version 4.0 – December 2023

computational study. Environmental Science and Technology 53(20): 12062-12070.

doi:10.021/[Link].9b04281
Selck, H., Drouillard, K., Eisenreich, K., Koelmans, A. A., Palmqvist, A., Ruus, A., Salvito, D.,
Schultz, I., Stewart, R., Weisbrod, A., van den Brink, N. W., & van den Heuvel-Greve, M.
(2012). Explaining dif f erences between bioaccumulation measurements in laboratory and f ield
data through use of a probabilistic modeling approach. Integrated environmental assessment
and management, 8(1), 42–63. [Link]
Shrestha P, Junker T, Fenner K, Hahn S, Honti M, Bakkour R, Diaz C and Hennecke D (2016)
Simulation Studies to Explore Biodegradation in Water-Sediment Systems: From OECD 308 to
OECD 309. Environ Sci Technol 50:6856-64.
Shrestha P, Meisterjahn B, Klein M, Mayer P, Birch H, Hughes CB, Hennecke D (2019)
Biodegradation of volatile chemicals in soil: separating volatilisation and degradation in an
improved test setup (OECD 307). Environ. Sci. Technol. 53 (1), 20e28.
[Link]
Sijm DTHM, Verberne ME, De Jonge WJ, Pärt P and Opperhuizen A (1995) Allometry in the
uptake of hydrophobic chemicals determined in vivo and in isolated perf used gills. Toxicol
Applied Pharmacol 131:130-5.
Stegeman MHL, Peijnenburg WJGM and Verboom H (1993) A quantitative structure-activity
relationship f or the direct photohydrolysis of meta-substituted halobenzene derivatives in
water. Chemosphere 5:837-49.
Tian R, Posselt M, Fenner K, and McLachlan (2023) Increasing the Environmental Relevance of
Biodegradation Testing by Focusing on Initial Biodegradation Kinetics and Employing Low-Level
Spiking. Environ Sci Technol Lett 10:40-45. [Link]
Treu G, Slobodnik J, Alygizakis N, Badry A, Bunke D, Cincinelli A, Claßen D, Dekker RWRJ,
Göckener B, Gkotsis G, Hanke G, Duke G, Jartun M, Movalli P, Nika M-C, Rüdel H, Tarazona JV,
Thomaidis NS, Tornero V, Vorkamp K, Walker LA, Koschorreck J, Dulio V (2022a) Using
environmental monitoring data f rom apex predators f or chemicals management: Towards
better use of monitoring data f rom apex predators in support of prioritisation and risk
assessment of chemicals in [Link] Sciences Europe 34, 82.
[Link]
Treu G, Badry A, Bauer K, Claßen D ,Drost, W, Rüdel H (2022b). Deliverable B5.5 A draf t
guideline f or assessment of bioaccumulation of relevant pollutants in AP&P samples, LIFE
APEX-LIFE17 ENV/SK/000355. [Link]
content/uploads/2022/11/Deliverable_B5.5_1_Draft-guideline_B_assessment_final.pdf
Trowell JJ, Gobas FAPC, Moore MM, Kennedy CJ (2018) Estimating the Bioconcentration
Factors of Hydrophobic Organic Compounds f rom Biotransformation Rates Using Rainbow Trout
Hepatocytes. Archives of Environmental Contamination and Toxicology 75(2):295-305.
doi:10.1007/s00244-018-0508-z
Ulrich N, Endo S , Brown T, Bronner G, Abraham MH, and Goss KU (2017). UFZ‑LSER database
v 3.2. [Link] [Link]/lserd
UNECE (2003) Globally Harmonised System of Classif ication and Lableeling of Chemicals
(GHS), United Nations. Availabe at: [Link]
UNEP (2022) Draf t document on long-range environmental transport. United Nations,
Stockholm Convention on Persistent Organic Pollutants, Persistent Organic Pollutants Review
 Chapter R.11: PBT/vPvB assessment

184 Version 4.0 – December 2023

Committee, eighteenth meeting, Rome, 23 September 2022. UNEP/POPS/POPRC.18/INF/21.

Inf ormation document available at:
[Link]
/9165/Def [Link]
US EPA (2000a) Methods f or Measuring the Toxicity and Bioaccumulation of Sediment-
Associated Contaminants with Freshwater Invertebrates, Second Edition.
US EPA (2000b) Quantitative Property Estimation f or K ow (KOWWin).
US EPA (2012) Estimation Programs Interface Suite™ f or Microsoft® Windows, v 4.11. United
States Environmental Protection Agency, Washington, DC, USA.

van den Brink NW, Arblaster JA, Bowman SR, Condor JM, Elliott JE, Johnson MS, Muir DCG,
Natal-da-Luz T, Rattner BA, Sample BE and Shore RF (2016) Use of terrestrial f ield studies in
the derivation of bioaccumulation potential of chemicals. Integr Environ Assess Manag 12:135-
45.
Vander Zanden MJ and Fetzer WW (2007). Global patterns of aquatic f ood chain length. Oikos
116: 1378–1388. ISSN 0030-1299.
Vanderklift MA, Ponsard S (2003) Sources of variation in consumer-diet δ 15 N enrichment: a
meta-analysis. Oecologia 136: 169-182.
Vogel A., Denney RC, Barnes JD, Thomas MJK (2006) Vogel’s Textbook of Quantitative
Chemical Analysis, Prentice Hall; 6th edition, Addison Wesley Publishing Co.
Wang H, Birch H, Sjøholm KK, Xia X, and Mayer P (2022) In-Tube Passive Dosing of
Hydrophobic Organic Chemicals: Controlling Freely Dissolved Concentrations in Flow -Through
and Large-Volume Experiments. Environ. Sci. Technol. Lett. 9:4, 339-344.
Wania F, Lei YD, Baskaran S, Sangion E (2021) Identif ying organic chemicals not subject to
bioaccumulation in air‑breathing organisms using predicted partitioning and biotransf ormation
properties; Integrated Environmental Assessment and Management 18(5): 1297-1312.
[Link]
Wassenaar PNH, Verbruggen EMJ, Cieraad E, Peijnenburg WJGM, Vijver MG. Variability in f ish
bioconcentration f actors: Inf luences of study design and consequences f or regulation.
Chemosphere. 2020;239:124731. [Link]
Weeks J, Guiney P, Johanning K (2020a) In vitro and in vivo metabolic stability of various
f ragrance materials and insect repellent in rainbow trout (Oncorhynchus mykiss). Journal of
Applied Toxicology 40(6):763-779. doi:10.1002/jat.3942
Weeks J, Li A, Doshi U, Johanning K, Guiney P (2020b) In vitro evaluation of the metabolic
stability of nine f ragrance chemicals in trout and humany hepatocytes. Journal of Applied
Toxicology 40(10): 1421-1434. doi:10.1002/jat.3995
Weisbrod, Annie V, Kent B Woodburn, Albert A Koelmans, Thomas F Parkerton, Anne E
McElroy, und Katrine Borgå (2009) Evaluation of Bioaccumulation Using In Vivo Laboratory and
Field Studies“. Integrated Environmental Assessment and Management 5(4): 598–623.
[Link]
Wu JP, Zhang Y, Luo XJ, Wang J, Chen SJ, Guan YT, Mai BX (2010) Isomer-specif ic
bioaccumulation and trophic transfer of Dechlorane Plus in the f reshwater f ood web f rom a
highly contaminated site, South China. Environ Sci Technol. 15;44(2):606-11. doi:
10.1021/es902744b
 Chapter R.11: PBT/vPvB assessment

185 Version 4.0 – December 2023

Zhou Y, Yuan B, Nyberg E, Yin G, Bignert A, Glynn A, Odland JØ, Qiu Y, Sun Y, Wu Y, Xiao Q,
Yin D, Zhu Z, Zhao J, Bergman Å. (2020) Chlorinated Paraffins in Human Milk f rom Urban Sites
in China, Sweden, and Norway. Environ Sci Technol 54:7, 4356-4366. doi:
10.1021/[Link].9b06089
 Chapter R.11: PBT/vPvB assessment

186 Version 4.0 – December 2023

Appendices
Appendix R.11-1.1: Development of a log Kow cut-of f value f or the B-criterion in the
PBT-assessment

Appendix R.11-1.2: Graphic def initions f or the molecular dimensions

Appendix R.11-2: PBT assessment of UVCB petroleum substances

Appendix R.11-3: Bioconcentration studies with benthic and terrestrial invertebrate

species (BSAF)

Appendix R.11-4: Non-extractable residues (NER)

Appendix R.11-5: Comparison of HPLC and KOWWIN v1.68 QSAR generated log Kow
values

Appendix R.11-6: Relevance and scientif ic justif ication of correction f or growth

dilution when deriving BCF

Appendix R.11-7: Volatilisation correction approaches f or the kinetic analysis of

simulation studies
 Chapter R.11: PBT/vPvB assessment

187 Version 4.0 – December 2023

Appendix R.11—1.1: Development of a log Kow cut-off value for the B-criterion in the
PBT-assessment (developed based on fish BCF)
The f ollowing assessment was based on the same data set used f or development of the D max ave
indicators (Dimitrov et al., 2005). Since publication the data set has been extended by
Dimitrov, S. This was the dataset used f or this exercise. With respect to the database used f or
the development of the cut-off value it is important to realize that the database comprises two
data sets obtained f rom ExxonMobil and MITI. A quality assessment was made of the MITI
data (as described in Dimitrov et al. 2005) and consequently the assessed data does not
contain all the MITI data and may contain values that may not be considered as reliable. The
experimental data f rom ExxonMobil are generated f rom f ish-f eeding studies, but only cover
substances with Log Kow values of < 7. For these reasons, it is recommended that this indicator
be re-evaluated.

The f itted lines in Figure R.11—8, Figure R.11—9 and Figure R.11—10 are based on subsets of
the BCF-dataset and are use to illustrate a limited bioconcentration potential f or substances
with high Kow -values. However, they are not to be used as a QSAR to estimate BCF f rom Log
Kow (see Section R.7.10 in Chapter R.7c of the Guidance on IR&CSA).

For substances with a Log K ow higher than 9.3 (based on CLogP) it was estimated that the
maximum BCF value is equal to 2000. The 95% conf idence interval f or this exercise is 9.5
(Figure R.11—8).

5.5
5.0
4.5
4.0
3.5
3.0
2.5
log BCF

2.0
1.5
1.0
0.5
0.0
-0.5
-1.0
-1.5
-5.0 -2.5 0.0 2.5 5.0 7.5 10.0 12.5 15.0
log Kow (ClogP)

Figure R.11—8: Log BCF v calculated Log Kow.

Figure R.11—9 plots the available BCF data against measured Log Kow values. No experimental
were available above Log Kow of 8.5 apart f rom estimates by HPLC. This supports the belief
that this is the limit of current state-of-the-art techniques f or the determination of Log Kow (i.e.
slow-stirring and column elution).
 Chapter R.11: PBT/vPvB assessment

188 Version 4.0 – December 2023

5.5
5.0
4.5
4.0
3.5
3.0
2.5
log BCF

2.0
1.5
1.0
0.5
0.0
-0.5
-1.0
-1.5
-3 -2 -1 0 1 2 3 4 5 6 7 8 9 10
log Kow (experimental)

Figure R.11—9: LogBCF v measured log Kow.

The relevance and experimental dif f iculties of conducting aqueous exposure on substances
with very high Log Kow must be questioned. Therefore it was decided to repeat the calculation
with the BCFs f rom f eeding experiments only (Figure R.11—10). The data f or very hydrophobic
substances are limited and there were 15 values f or substances with calculated Log Kow values
above 7. None of these 15 reached the same level of BCF as the highest BCFs between Log Kow
values of 6.5 and 7.0 when compared to the parabolic relationship in Figure R.11—9. Of these
15, three substances had calculated Log Kow values above 8, one is a vB substance and one is
a B substance (very close to vB).
 Chapter R.11: PBT/vPvB assessment

189 Version 4.0 – December 2023

6.5
6.0
5.5
5.0
log BCF (feeding test only)

4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0

1 2 3 4 5 6 7 8 9
log Kow (ClogP)

Figure R.11—10: LogBCF derived from feeding studies versus calculated Log Kow.
Summarized, the results of Figure R.11—8 to Figure R.11—10 suggest that the B-criterion is
unlikely to be triggered f or substances with a Log Kow higher than 10. As with the other
indicators f or hindered uptake, a Log Kow -value higher than 10 should be used in a Weight-of-
Evidence approach in combination with the other indicators.

References

Dimitrov S, Dimitrova N, Parkerton T, Comber M, Bonnell M, Mekenyan O (2005). Base -line

model f or identif ying the bioaccumulation potential of chemicals. SAR QSAR Environ Res
16(6):531-54.
 Chapter R.11: PBT/vPvB assessment

190 Version 4.0 – December 2023

Appendix R.11—1.2 Graphic definitions for the molecular dimensions used in the
main paper
• Maximum molecular length (MML) – the diameter of the smallest sphere into which the
molecule would reside, as written, i.e. not accounting f or conf ormers

• Maximum diameter, Dmax – the diameter of the smallest sphere into which the molecule
may be placed. Of ten this will be the same as the MML, especially f or rigid molecules.
However, when f lexible molecules are assessed, energetically reasonable conformers could
be present f or which this is very dif f erent. The average value of D max f or “energetically
stable” conf ormers is used, i.e. D max ave.

• (Maximum) Cross-sectional diameter – the diameter of the smallest cylinder into which the
molecule may be placed. Again dif f erent conf ormers will have dif f erent cross-sectional
diameters.

Conf ormer 1 (Ho = -84.5 kcal/mol), Dmax = 21.4; Def f = 4.99; Dmin = 4.92

Conf ormer 2 (Ho = -71.8 kcal/mol), Dmax = 19.8; Def f = 6.63; Dmin = 5.12
 Chapter R.11: PBT/vPvB assessment

191 Version 4.0 – December 2023

Conf ormer 3 (Ho = -68.5 kcal/mol), Dmax = 14.0; Def f = 11.5; Dmin = 5.52

Example Softwares:

• OASIS

To calculate Dmax ave conf ormational analysis of the molecule needs to be conducted. This is
done by estimating Dmax of each conf ormers and then the average Dmax values across the
conf ormers. An OASIS sof tware module is used to generate the energetically stable conformers
representing conformational space of the molecules. The method is based on genetic algorithm
(GA) generating a f inal number of structurally diverse conf ormers to best represent
conf ormational space of the molecules (Mekenyan et al., 1999 and 2005). For this purpose the
algorithm minimizes 3D similarity among the generated conf ormers. The application of GA
makes the problem computationally f easible even f or large, f lexible molecules, at the cost of
non-deterministic character of the algorithm. In contrast to traditional GA, the f itness of a
conf ormer is not quantif ied individually, but only in conjunction with the population it belongs
to. The approach handles the f ollowing stereochemical and conf ormational degrees of f reedom:

• rotation around acyclic single and double bonds,

• inversion of stereocenters,
• f lip of f ree corners in saturated rings,
• ref lection of pyramids on the junction of two or three saturated rings.

The latter two were introduced to encompass structural diversity of polycyclic structures. When
strained conf ormers are obtained by any of the algorithms the possible violations of imposed
geometric constraints are corrected with a strain-relief procedure (pseudo molecular
mechanics; PMM) based on a truncated f orce f ield energy-like f unction, where the electrostatic
terms are omitted (Ivanov et al., 1994). Geometry optimization is f urther completed by
quantum-chemical methods. MOPAC 93 (Stewart, 1990 and 1993) is employed by making use
of the AM1 Hamiltonian. Next, the conf ormers are screened to eliminate those, whose heat of
f ormation, DHf o, is greater f rom the DHf o associated with the conf ormer with absolute energy
minimum by user def ined threshold - to be within the range of 20 kcal/Mol (or 15 kcal/mol)
threshold f rom the low(est) energy conf ormers (Wiese and Brooks, 1994). Subsequently,
conf ormational degeneracy, due to molecular symmetry and geometry convergence is detected
within a user def ined torsion angle resolution.
 Chapter R.11: PBT/vPvB assessment

192 Version 4.0 – December 2023

A f reely available version of this Dmax calculation is available in the OECD QSAR Toolbox
under 3D parameters (Dimitrov et al, 2002; Mekenyan et al, 2005).

• Calculation of the 3D Dimension of a Molecule

A molecular modelling program, e.g. Molecular Modelling Pro, uses a 2D molecular structure as
a starting point f or the calculation. In the 1st step the program calculates the least strained 3D
conf ormer using e.g. MOLY Minimizer as built in the Molecular Modelling Pro. Normally this
minimizing of strain requires multiple steps. If the strain energy is minimized the program
calculates the 2nd step the 3D molecular dimensions (x length, y width, z depth) e.g. in
Angstrom. Based on these x,y,z dimensions Molecular Modelling Pro is able to calculate a
global maximum and minimum which can be used a Dmax. Another example of a sof tware
which may be used to estimate Dmax is Marvin Sketch.

References

Dimitrov SD, Dimitrova NC, Walker JD, Veith GD and Mekenyan OG (2002) Predicting
bioconcentration f actors of highly hydrophobic chemicals: eff ects of molecular size. Pure and
Applied Chemistry 74:1823-30.
Ivanov JM, Karabunarliev SH, Mekenyan OG (1994) 3DGEN: A system For an Exhaustive 3D
Molecular Design. J Chem Inf Comput Sci 34:234-43.

Mekenyan OG, Dimitrov D, Nikolova N and Karabunarliev S (1999) Conf ormational Coverage
by a Genetic Algorithm. Chem Inf Comput Sci 39:997-1016.

Mekenyan OG, Pavlov T, Grancharov V, Todorov M, Schmieder P and Veith G (2005) 2D -3D
Migration of Large Chemical Inventories with Conf ormational Multiplication. Application of the
Genetic Algorithm. J Chem Inf Model 45:283 -92.

Stewart JJP (1990) MOPAC: A semi empirical molecular orbital program. J Comput-Aided Mol
Des 4:1-105.

Stewart JJP (1993) MOPAC 93. Fujitsu Limited, 9-3, Nakase 1-Chome, Mihama-ku, Chiba-city,
Chiba 261, Japan, and Stewart Computational Chemistry. 15210 Paddington Circle, Colorado
Springs, Colorado 80921, USA.

Wiese T and Brooks SC (1994) Molecular modelling of steroidal estrogens: novel conformations
and their role in biological activity. J Steroid Biochem Molec Biol 50:61-72.
 Chapter R.11: PBT/vPvB assessment

193 Version 4.0 – December 2023

Appendix R.11—2: PBT assessment of UVCB petroleum substances.

UVCB petroleum substances are assessed using the same principles as other UVCBs, as
introduced in Section R.[Link]. However, at the time of developing PBT assessment
principles f or UVCBs the available knowledge on the composition and behaviour of petroleum
substances was broader than the knowledge available on other types of UVCBs, thereby
warranting the development of a specif ic methodology to assess petroleum substances. The
f ollowing subsections introduce how such knowledge can be used. The specif ic assessment
path presented is called the hydrocarbon block method, developed by CONCAWE (2001). An
analogous assessment path may be used f or other UVCB categories, if appropriate.

Step 1: Characterisation of the petroleum substance

Due to their derivation f rom natural crude oils and the ref ining processes used in their
production, petroleum substances are complex mixtures of hydrocarbons, of ten of variable
composition. Many petroleum substances are produced in very high tonnages to a range of
technical specif ications, with the precise chemical composition of particular substances, rarely
if ever characterized. Since these substances are typically separated on the basis of distillation,
the technical specif ications usually include a boiling range. These boiling ranges correlate with
carbon number ranges, while the nature of the original crude oil and subsequence ref inery
processing inf luence the types of hydrocarbon structures present. The CAS name def initions
established f or the various petroleum substance streams generally ref lect this, including f inal
ref inery process; boiling range; carbon number range and predominant hydro carbon types
present.

For most petroleum substances, the complexity of the chemical composition is such that it is
beyond the capability of routine analytical methodology to obtain complete characterisation.
Typical substances may consist of predominantly mixtures of straight and branched chain
alkanes, single and multiple naphthenic ring structures (of ten with alkyl side chains), single
and multiple aromatic ring structures (often with alkyl side chains). As the molecular weights
of the constituent hydrocarbons increase, the number and complexity of possible structures
(isomeric f orms) increases exponentially.

For the purposes of a PBT assessment of petroleum substances, when required, it is suggested
that an analytical approach using GCxGC is used when f easible. This method of f ers a high
resolution that may also be helpf ul in being more precise as to the exact type of structures
present, (Forbes et al., 2006), in contrast to more generic methods based on Total Petroleum
Hydrocarbon (e.g. TNRCC Method 1005, 2001). Still other methods could be used to
characterize the composition of petroleum substances as the GCxGC method has the caveat
that it can only be used f or carbon numbers up to around C 30.

The outcome of this step should be a matrix of hydrocarbon blocks, containing the %
contribution of the block to the petroleum substance. With GCxGC this characterisation will be
extended to include broad descriptions of structures including alkanes, isoalkanes,
naphthenics, aromatics, etc.

Step 2: Assessment

The next step is to collate the available inf ormation on persistence, bioaccumulation and
toxicity of the petroleum substance(s) being assessed. Where this is done as part of a
 Chapter R.11: PBT/vPvB assessment

194 Version 4.0 – December 2023

category, there will be need f or a good justif ication, which could also include analytical
characterisation of a category. The assessment of the data will f ollow similar lines as f or any
data examination, including the extent to which the petroleum substances were characterised
or described, the type of protocol f ollowed and the quality of the inf ormation obtained f or the
respective endpoints.

Persistence (P)

The f irst part of the P assessment would be to examine the available data, and in particular
attempt to identif y whether the data on the petroleum substance(s) under investigation can be
considered representative f or the whole composition. The principles as provided f or applying
the “whole substance” approach as specified in Section R.[Link] and elements as discussed
in Section R.[Link] (Persistence) need to be considered. Where there is convincing evidence
of ready biodegradation of the whole substance under these principles, it can be reasonably
assumed that the individual components are unlikely to be persistent.

If there is insuf ficient evidence f or ready biodegradation or the substance composition is not
suf f iciently homogenous (i.e. the known or assumed constituents are structurally too
dissimilar) to interpret data on the whole substance, then the assessment should proceed to
the next stage. This involves generating typical structures either f rom the chemical analysis
conducted or f rom other sources of inf ormation relevant to the petroleum substances being
assessed. For example, Redman et al. (2012, 2014) describe how a set of over 1500
structures are available f or assessing hydrocarbon blocks of petroleum substances. The
structures cover a wide range of hydrocarbon types including isoparaf f inic, normal paraf f inic,
mono-naphthenic (1-ring cycloalkanes), di-naphthenic (2-ring cycloalkanes) and poly-
naphthenic, mono-aromatic, di-aromatic and aromatic (3 to 6-ring cycloalkanes) classes and
mixed aromatic/napthenic hydrocarbons. By correlating the predicted boiling point of these
structures to the available analytical inf ormation, a series of blocks can be generated in which
these structures are representative of the types potentially present in the petroleum
substance.

The assessment can then proceed by evaluating available degradation half -lif e inf ormation on
any known individual constituents, e.g. benzene, hexane, pristane etc. This inf ormation will in
every case be insuf ficient for the assessment of petroleum UVCB substances due to the wide
range of potential structures and the relatively limited inf ormation currently available on most
of the individual structures that have normally not been tested, as they are rarely isolated or
manuf actured. Consequently, the inf ormation will need to be supplemented with data f rom
predictive models.

For hydrocarbons, there are two QSAR models that could be considered f or assessing
environmental degradation half -lif es and a third that could be used f or assessing potential
transf ormation/degradation products:

• Howard et al. (2005) describe a model that predicts the degradation half -lif e of a
hydrocarbon in the environment. The model is well described, including inf ormation on
the test/training sets. In using the model it would be advisable to assess the training
and tests sets to ensure suitable coverage of the structures being assessed. Some data
used in BIOHCWIN are based partially on half -lif es obtained f or single substances
studied as multi-constituent substances, e.g. cycloalkanes, thus the predicted half -lif es
can be af f ected by co-metabolism and theref ore may overestimate the rate of
 Chapter R.11: PBT/vPvB assessment

195 Version 4.0 – December 2023

degradation in such cases (see Section R [Link] in Chapter R.7c of the Guidance on
IR&CSA). BIOHCWIN half -lifes which are below the P/vP criteria should not be used to
support a conclusion "not P/vP". Care must also be taken with branched substances, as
it appears that their environmental half -lives may be underestimated by BIOHCWIN
(Rorije et al., 2012). This model is f reely available in EPI Suite as BIOHCWIN (US EPA,
2012).

• Dimitrov et al. (2007) describe a model that combines CATABOL (Jaworska et al., 2002)
with assumptions of f irst order catabolic transf ormations. The training and test sets
include inf ormation of petroleum substances as well as observed catabolic pathways
compiled f rom various sources including public web sites such as EAWAG BBD
([Link]

• Finally, f or demonstrating that there are no concerns, caused by potential

transf ormation/degradation products (the previous assessments are all addressing
primary biodegradation), it is recommended that available inf ormation is collected and
predictions made of relevant PBT properties of potential transf ormation/degradation
products. CATABOL (successor CATALOGIC) is an example of integration of such an
approach in a commercial modelling system (Jawoska et al., 2002).

If these assessments indicate that there are structures or blocks that are of concern, the
assessment can either proceed to the generation of new inf ormation as described in the main
report, or conclude that the assessed blocks can be considered persistent and proceed to the
bioaccumulation assessment.

Bioaccumulation (B)

The B assessment essentially f ollows the same process as that described f or the P assessment
except that it is highly unlikely that there will be good quality experimental data on petroleum
UVCB substances. Instead the B assessment is more likely to address the individual structures
f or their potential to bioaccumulate. This, as with the P assessment, will start with addressing
where there is available experimental evidence to be able to draw a conclusion on the B
properties of blocks or individual constituents.

Where there are insuf f icient experimental data to be able to make a judgement there are
several QSAR models available f or continuing the process. These are discussed in Section
R.[Link] of Chapter R.7c of the Guidance on IR&CSA and Appendix R.11—1.1 of this
Guidance document. An assessment of the predictions f rom these QSAR models, with available
experimental inf ormation should lead to the identif ication of those blocks where there are
concerns f or their potential (or realised, if specif ic structures are assessed) ability to
bioconcentrate. The use of experimental f ish or aquatic invertebrate bioaccumulation data from
tests f ollowing standard test methods/guidelines is pref erred over that f rom other sources,
including invertebrates, because f ish bioaccumulation data are generally more reliable as
standard test methods/guidelines are used to determine them. Fish bioaccumulation data
include the ef f ect of biotransformation in f ish which can be substantial f or some hydrocar bons.
Such data also provide indications of whether the potential f or f ood-chain magnif ication at
higher trophic levels exists. This type of data, with f urther inf ormation on trophic level
biomagnif ication or dilution, can be used in a Weight-of-Evidence approach to demonstrate
whether the longer term uncertainties associated with bioaccumulation of constituents may
exist.
 Chapter R.11: PBT/vPvB assessment

196 Version 4.0 – December 2023

Toxicity (T)

Assessment of the toxicity of all individual constituents within a petroleum substance would in
many cases be extremely dif f icult or practically impossible. While the whole substance
assessment using the Water Accommodated Fraction (WAF) methodology has been accepted
f or classification purposes (OECD, 2001), the use of this inf ormation f or the T assessment is
problematic.

For petroleum substances, a model, PETROTOX, has been developed (Redman et al., 2012),
based on previous work assuming a non-polar narcosis mode of action (i.e. baseline toxicity,
McGrath et al., 2004, 2005). The equations underlying the hazard portion of this model, which
was developed to predict the acute and chronic ecotoxicity of petroleum substances and
hydrocarbon blocks, may be used to address the predicted baseline toxicity of individual
structures when no experimental data are available.

It should be noted that f or the ultimate conclusion on the T property, long-term toxicity test
results are generally necessary as, at present, no appropriate prediction tools f or long-term
ecotoxicity are available. The prediction tools may, however, be used as supporting tools f or
designing tests and f or the interpretation of experimental results. Bef ore initiating
experimental f ish toxicity tests it should be considered whether data exist allowing a robust
conclusion to be drawn on whether the substance f ulf ils the Tmammalian criteria (see Section
R.[Link]).

How to proceed further

Where there are constituents or blocks that show a concern f or both P and B properties, there
is a need to generate f urther higher tier inf ormation on these properties. Exceptions to this
conclusion might be in case there are sufficient ecotoxicological data on specif ic constituents or
representative structures in the blocks that demonstrate no concern f or the T criterion and
where the P and B properties are concluded not to indicate vPvB-properties.

References

Dimitrov S, Pavlov T, Nedelcheva D, Reuschenbach P, Silvani M, Bias R, Comber M, Low L, Lee

C, Parkerton T and Mekenyan O (2007) A Kinetic Model f or Predicting Biodegradation. SAR
QSAR Environ Res 18:443-57.

Forbes S, Eadsforth C, Dmytrasz B, Comber M and King D (2006) Application of comprehensive

two-dimensional gas chromatography (GCxGC) f or the detailed compositional analysis of gas-
oils and kerosines – SETAC Den Haag.

Howard P, Meylan W, Aronson D, Stiteler W, Tunkel J, Comber M, Parkerton T (2005) A new

biodegradation prediction model specif ic to petroleum hydrocarbons, Environ Toxicol Chem
24:1847-60.

Jaworska J, Dimitrov S, Nikolova N and Mekenyan O (2002) Probabilistic assessment of

biodegradability based on metabolic pathways: catabol system. SAR QSAR Environ Res
13:307-23.
 Chapter R.11: PBT/vPvB assessment

197 Version 4.0 – December 2023

McGrath JA, Parkerton TF and Di Toro DM (2004) Application of the narcosis target lipid model
to algal toxicity and deriving predicted no ef f ect concentrations. Environ Toxicol Chem
23:2503–17.

McGrath JA, Parkerton TF, Hellweger FL and Di Toro DM (2005) Validation of the narcosis
target lipid model f or petroleum products: gasoline as a case study. Environ Toxicol Chem
24:2382–94.

OECD (2001) OECD Series on Testing and Assessment, Number 27, Guidance Document on
the Use of the Harmonised System for the Classif ication of Chemicals which are Hazardous f or
the Aquatic Environment. ENV/JM/MONO(2001)8 (available at:
[Link]
&doclanguage=en).

Redman AD, Parkerton TF, Comber MH, Paumen ML, Eadsf orth CV, Dmytrasz B, King D,
Warren CS, den Haan K, Djemel N (2014) PETRORISK: a risk assessment f ramework f or
petroleum substances. Integr Environ Assess Manag. 10(3):437-48.

Redman AD, Parkerton TF, McGrath JA and Di Toro DM (2012) PETROTOX: An aquatic toxicity
model f or petroleum substances. Environ Toxicol Chem 31:2498–2506.

Rorije,E., Verbruggen,E.M.J., DeKnecht,J.A. Framework Contract No ECHA/2008. (2012)

Service Request on a critical review of the environmental and physicochemical methodologies
commonly employed in the environmental risk assessment of petroleum substances in the
context of REACH registrations. Available at:
[Link]
physicochemical_methodol_en.pdf .

TNRCC (2001) TNRCC Method 1005, TOTAL PETROLEUM HYDROCARBONS, Texas Natural
Resource Conservation Commission, Revision 03, June 1, 2001.

US EPA (2012) Estimation Programs Interface Suite™ f or Microsoft® Windows, v 4.11. United
States Environmental Protection Agency, Washington, DC, USA.
 Chapter R.11: PBT/vPvB assessment

198 Version 4.0 – December 2023

Appendix R.11—3: Bioconcentration studies with benthic and terrestrial invertebrate

species (BSAF).
In case data are available f rom bioconcentration studies on benthic and terrestrial invertebrate
species they may be used as indicator f or a high bioaccumulation potential. Results of these
studies are expressed as biota-to soil/sediment accumulation f actor (BSAF). In order to
compare BSAF with BCF values care must be taken if a species with a very low lipid content
was used because BCF values are normaly reported on a wet weight basis. Lipid normalisation
(to 5% lipid content) should therefore always be performed, whenever possible f or substance
that are lipid binding.

The relationship between BSAF and BCF is expressed in the f ollowing equation, in which BCF
could be replaced by the criterion f or B or vB.

BCF (lipid ) 2000 / 0.05 5000 / 0.05

BSAF = = for indication of B or for indication of vB
K oc K oc K oc

A terrestrial or benthic (lipid and organic carbon normalized) BSAF value f or a substance with a
Log Kow of 4.5 that exceeds the value of 2 is an indication of a BCF of 2000 and higher, based
on pore water concentration (Figure R.11—11). Similar f or a substance with a Log Kow of 4.5 a
BSAF value higher than 5 is an indication that the BCF exceeds the value of 5000, based on
pore water concentration.
5

3
BSAF

0
4 4.5 5 5.5 6 6.5 7
log K ow

Figure R.11—11: Relationship between lipid and organic carbon normalised BSAF
values and Log Kow as indicator for the B and vB criterion.
The solid line is calculated with a BCF value (5% lipids) f rom pore water of 2000, the dotted
line is calculated with a BCF value of 5000. The Log Koc has been calculated according to the
equation Log Koc = Log Kow - 0.21 by Karickhof f et al. (1979).

Due to increasing sorption with Log Kow , the BSAF values f or calculated BCF values of 2000 and
5000 rapidly decrease. Therefore, f or a substance exceeding Log Kow of 5.5, a BSAF value in
the order of 0.5 and above indicates high bioaccumulation potential.
 Chapter R.11: PBT/vPvB assessment

199 Version 4.0 – December 2023

However, lower BSAF values are dif ficult to interpret in the context of the B and vB assessment
due to several conf ounding f actors. Sorption and bioconcentration increase with
hydrophobicity, and as it is not necessarily in the same manner, sorption is an important
parameter dependent on soil and substance properties. Bioconcentration might be reduced
compared to what is expected f rom Log Kow value but even low BSAF values of 0.1 and lower
do not necessarily mean that the BCF value based on pore water concentration do not exceed
5000, because of the strongly increased sorption f or highly hydrophobic substances. Moreover,
sorption might be higher than what is expected from Log Kow because sorption to carbonaceous
materials may play an important role. Besides that, f or these low BSAF values it is of ten
dif f icult to distinguish between real uptake and adsorption to the organisms or interf erence of
gut content in the determination of the BSAF values.

In conclusion, lipid and organic carbon normalized BSAF values of 0.5 and higher are an
indication of high bioaccumulation. In some cases these values might be considered to be
enough evidence in itself to assess the substance as B and vB, especially if reliable
experimental data on pore water concentrations are available and the system is in equilibrium.
However, lower BSAF values should not be used to the contrary, because low uptake f rom
sediment or soil does not imply a low aquatic BCF value.

Reference

Karickhoff SW, Brown DS and Scott TA (1979) Sorption of hydrophobic pollutants on natural
sediments. Water Res 13:241-8.
 Chapter R.11: PBT/vPvB assessment

200 Version 4.0 – December 2023

Appendix R.11-4: Approach on non-extractable residues (NER) quantification and

characterisation in persistence assessment
To complete the mass balance as a part of a simulation test with radiolabelled substances
(14C), quantif ication of the amount of total non-extractable residues (Total NER) is necessary.
The assessment may be ref ined by characterising and quantif ying the dif f erent NER types. This
Appendix provides a stepwise approach f or characterising and quantif ying NER (based on
Kästner et al. 2018). A tiered extraction scheme f or quantif ication of Total NER (Step 1) and
the characterisation of the dif f erent NER types (Step 2 and 3) is described below (Figure R.
11—12).

If any other method is used, the relevance and adequacy of the method needs to be
demonstrated. Determination and characterisation of the Total NER with the methods
described in this Appendix, rely on the use of radio-labelled substance. In the absence of any
inf ormation on the nature/types of NER, the residues should be regarded as non-degraded
parent substance.

Types of NER in environmental matrices (Löf f ler et al., 2022, ECHA, 2019, Kästner et al.,
2018):

• NER Type I (strongly sorbed and physically entrapped): strongly sorbed or

physically entrapped into the matrix, contain the parent substance,
transf ormation/degradation products or both. NER Type I have the potential to be
remobilised, and should be considered f or assessment of persistence. If chemical
analyses are conducted, it may be possible to distinguish whether NER Type I consist of
unmodif ied parent substance or of transf ormation/degradation products.

• NER Type II (covalently bound): residues that are covalently bound to the matrix
(e.g. to humic matter) in surf ace water, soils or sediments and that are considered to
have low remobilisation rates. Unless there are indications from the available literature
or monitoring data regarding their potential remobilisation, covalently bound res idues
may be regarded as irreversibly bound.

• NER Type III (bioNER): incorporated into biomass (biogenic NER, also called bioNER),
NER Type III result f rom the anabolic f ormation of biomolecules (amino acids,
phospholipids, and other biomass compounds) f rom the degradation products of the
parent substance. Dead biomass, and therefore biogenic NER, are eventually f ixed in
organic matter derived from decaying microbial biomass. NER Type III are considered to
be of no concern.

The sum of NER Type I and II can be also ref erred as XenoNER. XenoNER can be roughly
estimated as Total NER - NER Type III (bioNER).
 Chapter R.11: PBT/vPvB assessment

201 Version 4.0 – December 2023

Figure R. 11—12: Approach for the quantification of the total extractable and non-
extractable residues (NER) and the characterisation of the NER types in
environmental matrices (water/soil/sediment) (adapted from Kästner et al. 2018).
ASE: Accelerated Solvent Extraction; PLE: Pressurised Liquid Extraction at elevate d pressure and temperature
(equivalent to ASE); MTB approach: Microbial Turnover to Biomass approach; EDTA: Ethylenediaminetetraacetic acid.

The Figure R. 11—12 provides an overview approach f or the quantif ication of the total
extractable and non-extractable residues (NER) and the characterisation of the NER types in
environmental matrices. Detailed description of each step is provided in the sections below and
in Figure R. 11—13, Figure R. 11—14 and Figure R. 11—15.

Quantification of NER: extraction methodologies (Step 1)

This step allows the distinction into the extractable fraction and the non-extractable residues,
the latter representing the substance or its transf ormation/degradation products remaining in
the matrix (soil, sediment, suspended matter). The extraction efficiency needs to be optimized
f or the test substance and/or its potential transf ormation/degradation products. An ef f icient
extraction procedure f or the test substance may not be equally ef f icient f or its
transf ormation/degradation products.

Total NER are def ined as the residues remaining in the matrix af ter def ined exhaustive
extractions steps described below (one standardised extraction procedure with a ternary
solvent mixture or three consecutive extraction steps). The amount of NER is operationally
def ined by the extraction method employed. Dif f erent extraction regimes (i.e. dif f erent
solvents and dif f erent extraction conditions) are possible f or the same substance.
 Chapter R.11: PBT/vPvB assessment

202 Version 4.0 – December 2023

The f ollowing extraction procedures ref er to Step 1 (A or B) of Figure R. 11—12.

The selection of the most appropriate extraction strategy to def ine the amount of extractable
residues in simulation test is case specif ic and it is af fected by the substance itself and the test
matrix. The stepwise extraction procedure below (Step 1: A.1, B.1, B.2 and B.3) describes a
way to release all the extractable fractions. Each of these f ractions can be used to analyse the
amounts and to identif y the structures of the extractable residues. The residues remaining in
the matrix af ter those three steps are def ined as Total NER.

The choice of extraction procedure depends on the applicability and ef ficiency of the extraction
procedures. The extraction method should not destroy the parent substance and/or the
transf ormation/degradation products as quantif ication and characterisation could be
subsequently performed. Control experiments should be conducted to assess whether or not
the parent substance and/or the transformation/degradation product(s) are stable under the
extraction conditions (e.g. ef f ects of high temperature). Selection of the most appropriate
approach should be evaluated in a pre-study and justif ication provided.

A.1 The pressurized liquid extraction (PLE) (also called Accelerated Solvent Extraction (ASE))
with a standard solvent mixture (MeOH, acetone, water 50/25/25, v, v, v) at 100°C and 100
bar with 3 extraction cycles is one of the two recommended extraction procedures f or
determination of Total NER. Alternatively, the sample can also undergo a stepwise solvent
extraction f irst (B.1 and B.2) to be f ollowed by PLE, both with a substance specif ic solvent
(B.3).

The aqueous extraction (B.1) is intended to be a surrogate of the pore water in order to
determine the ‘easily extractable’ f raction that is bioavailable f or organisms living in the
matrix, e.g. soil or sediment. For the aqueous extraction at every sampling day, the sample
can be extracted by shaking f or 24 hours with CaCl2 solution (0.01M) (Houba et al., 2000;
Peijnenburg et al., 2007; Beulke and van Beinum (2012)). Settling of the solids during shaking
should be avoided. The soil (or sediment):solution ratio should be the same at every sampling
time point. The solid and liquid phases should be separated by centrif ugation and the
concentration of parent substance and potential transformation product(s) in the liquid phase
analysed. The liquid should be recovered from the sample as much as possible if consecutive
aqueous and solvent extractions are performed on the same sample. Other solutions of salts
like calcium or sodium nitrate, ammonium acetate or nitrate etc. have also been used f or this
purpose.

B.2 includes extraction with organic solvent mixtures which are intended to be a surrogate f or
73
the potentially bioaccessible or desorbable f raction f rom the matrix). The remaining residues
f rom step B.1 could be sequentially extracted at ambient temperature with caref ully selected
aqueous:organic solvent mixtures (e.g. 50:50 or 20:80 water:acetonitrile; v/v), which at times
may be modif ied with minute amounts (0.1-2.5 % v/v) of f ormic acid, acetic acid and/or
ammonia in order to enhance the solubility of the substance and/or its
transf ormation/degradation product(s).

73
Bioaccessibility of residues in organisms encompasses what is directly bioavailable and what may potentially become
bioavailable, for instance after release from a surrounding matrix or the matrix degradation.
 Chapter R.11: PBT/vPvB assessment

203 Version 4.0 – December 2023

Samples should be extracted f or prolonged time periods (4 h – 24 h) using physical agitation,

(e.g., an overhead shaker or a horizontal shaker) ensuring complete suspension of the matrice
(e.g. soil, sediment) that is being investigated. Shorter extractions times could be appropriate
and should be evaluated on a case by case basis. Ultra-sonication may enhance the extraction
ef ficiency, but the temperature of the sample should then be monitored. During studies with
radiolabelled substances, sequential extractions with the same solvent should be perf ormed
until <5 % of the radioactivity released f rom the f irst extraction is obtained. This usua lly
occurs between three and f ive extractions with one solvent system (Nießner and Schäf f er,
2017).

The selection of the proper organic solvents is a critical step. The physico-chemical properties
of the analyte, i.e., its volatility, water solubility, the solubility in the organic solvent to be
used, the pKa, and the stability, as well as the test matrix properties (such as the moisture
and organic matter content of soils and sediments), must be considered (ECETOC, 2013).
Properties of some extraction solvents and their relations to properties of analytes are
provided in Kästner, et al., 2018.

Pure organic solvents should be avoided in the f irst extraction steps because molecules
distributed in the interlayers of clay particles in soil may be entrapped by shrinking of the clay
when water is removed. Therefore, in the f irst extraction steps water miscible organic solvents
should be used mixed with low volumes of water, f ollowed by exhaustive extraction with the
use of pure solvents (or solvent mixtures).

Since transf ormation/degradation product(s) usually dif fer f rom the parent substance in terms
of polarity (most often more polar, sometimes less polar) and of chemical reactivity, extraction
procedures have to be developed during the course of a degradation study. An ef f ective
extraction solvent for the parent substance may not be ef fective f or transformation product(s).
Polar solvents or solvent mixtures should be tested f or extraction depending on the nature of
the residues.

B.3 includes exhaustive extractions which are intended to be a surrogate f or the slowly
desorbable f raction. Compared to the extraction methods described in the previous steps
(Steps B.1 and B.2), exhaustive extractions are conducted under harsher conditions (e.g.
elevated temperature and/or elevated pressure). The exhaustive extraction will release part of
the molecules strongly sorbed to the matrix and thus provide inf ormation on the total
extractable f raction. Remaining residues af ter exhaustive extractions will def ine the total
amount of NER.

The recommended method f or the last exhaustive extraction step is Pressurised Liquid
Extraction (PLE) also called Accelerated Solvent Extraction (ASE). If the ef f iciency of the PLE
method is not satisf actory other harsh extraction procedures (i.e Soxhlet, Microwave Assisted
Extraction or Supercritical f luid extraction) with solvent variations can be applied.

The results f rom Step 1 can be used f or DegT50 derivation as shown in the Figure R. 11—13.
The parent in total extractable f raction is considered f irst f or deriving the DegT50 and is
compared with the vP criterion. If the vP criterion is f ulf illed no f urther characterisation of NER
is needed. If the vP criterion is not f ulf illed a DegT50 based on the sum of the total extractable
f raction and the Total NER is derived and compared with the P/vP criterion. If the criterion is
f ulf illed the substance could be considered as P or vP or if the Total NER is ≥ 10 %
quantif ication of the NER types should be considered to ref ine the assessment (Step 2).
 Chapter R.11: PBT/vPvB assessment

204 Version 4.0 – December 2023

Figure R. 11—13: Assessment scheme for Step 1: Quantification of the total

extractable and non-extractable residues (NER) in environmental matrices
(water/soil/sediment).
* Conclusion on persistence in the absence of information on the amount of Type I in Total NER.

Characterisation of NER (Steps 2 and 3)

Step 2 describes the characterisation of the NER into potentially remobilisable f raction (NER
type I) and NER remaining in the matrix (NER type II and bioNER/NER Type III) (based on
Kästner et al. 2018, Löf f ler et al. 2020). Estimation on bioNER (Type III) may also be
considered if relevant. Subsequently Step 3 allows the quantif ication of the parent substance in
Type I NER f or all sampling days/points.

The identif ication and quantif ication of the parent substance and transf ormation/degradation
products should be carried out both in the extractable f raction and in the NER type I f raction
(obtained by silylation/EDTA extraction) in all sampling days/points. If it is not possible to
dif f erentiate parent f rom transf ormation/degradation products, the total radioactivity
measured in this f raction shall be considered as parent.
 Chapter R.11: PBT/vPvB assessment

205 Version 4.0 – December 2023

The dif f erentiation of NER type I and II does not need to be perf ormed when this inf ormation is
not expected to change the conclusion f or the persistence assessment, f or example when
based on all available inf ormation it is already possible to conclude .

Step 2: Determination of NER type I by silylation or EDTA extraction.

As stated above Step 2 allows the dif f erentiation of the Total NER into potentially remobilisable
f raction (NER type I) and NER remaining in the matrix (NER type II and bioNER/Type III).
Characterisation of NER in all sampling days/points is considered relevant when the amount of
Total NER is ≥10 % of the total applied radioactivity (AR) at least at one of the sampling days.
When the total NER is <10 %, the possibility f or f urther characterisation could still be explored
as the technical f easibility of the characterisation might also depend on the test methodology
(spiking radioactivity) and availability of analytical methods. Determination and
characterisation of the Total NER described in this Appendix rely on the use of radiolabelled
substances.

The potentially remobilisable f raction (NER type I) can be quantif ied by silylation or EDTA
extraction of the extraction residues (i.e. sample containing Total NER). The extraction
ef f iciency and the technical f easibility in relation to the test substance and
transf ormation/degradation products should be considered when selecting the method, i.e. the
extraction procedure shall not destroy or react with the test substance. This should also be
checked in a preliminary study. NER remaining in the matrix (soil, sediment, suspended
matter) after silylation/EDTA extraction have a low remobilisation potential, it does not need to
be considered in the persistence assessment and can be seen as a removal pathway (Kästner
et al., 2018).

Step 2 (Figure R. 11—14) describes quantif ication and characterisation of the NERType I by
Silylation or EDTA extraction of the matrix.
 Chapter R.11: PBT/vPvB assessment

206 Version 4.0 – December 2023

Figure R. 11—14: Assessment scheme for Step 2: Quantification and characterisation

of Type I NER by Silylation or EDTA extraction of the matrix.
* Conclusion on persistence in the absence of information on the amount or parent and degradation product in NER I.

Silylation is the introduction of a substituted silyl group (R3Si-) to a molecule carrying

f unctional groups with exchangeable protons. Silylation will replace the exchangeable hydrogen
atoms of f unctional groups in soil organic matter - e.g. carboxylic, hydroxy or amino groups -
with trimethylsilyl groups (Haider et al., 2000). The majority of the NER entrapped in the
humic matrix (type I NER) are released af ter silylation, while NER f ormed by covalent binding
(type II NER) remain bound to the matrix. In addition, both f ractions derived f rom this
procedure may also contain bioNER. The released and the remaining f ractions can be
quantif ied when radioactively labelled chemicals have been used. However, it is recommended
that subsequent analyses f or released parent substances or transf ormation product(s) are
perf ormed in addition in order to quantif y specifically type I NER, e.g. by mass spectrometric
approaches. For additional structural inf ormation, size exclusion chromatography or
spectroscopic methods like nuclear magnetic resonance (NMR) can be applied. Alternatively,
stable isotope labelling (13C, 15N) can be applied with subsequent GC-MS or LC-MS analyses.
Berns et al. (2005), Wang et al. (2017), Löf f ler et al. (2020) and Hennecke et al., 2023
provide a detailed description of the silylation procedure.

In general, similarly to the extraction method mentioned above the characterisation methods
should not chemically alter the parent substance or its transf ormation/degradation products.
 Chapter R.11: PBT/vPvB assessment

207 Version 4.0 – December 2023

Limitations of the silylation method (including f ollowing analytical methods):

• The silylation method has not been standardised.

• Silylation may also destroy the target molecules (parent substance and/or
transf ormation product(s)) if it contains the f unctional groups modified by the silylation
procedure. Therefore, appropriate control extraction treatments should always be run
while perf orming the silylation.

• Both the released f raction and the remaining f raction derived from this procedure may
also contain bioNER. Theref ore in order to quantif y specif ically the amount of type I
NER, it is necessary either to analyse f or released parent substances or transf ormation
product(s) e.g. by mass spectrometric approaches, or to quantif y the amount of bioNER
(see Step 2.2 below)

• Analyses of parent substance and transf ormation product(s) in the releas ed f raction
might be challenging in particular if there are many matrix components. By def ault,
NER type I should be considered to be parent substance if the analysis cannot be
perf ormed.

• Structural identif ication of type I residues using stable isotope labelling f or subsequent
stable isotope NMR (13C, 15N) is otherwise possible but has the disadvantage that high
concentrations of the test substance have to be used due to the low sensitivity of the
NMR method. Theref ore, f or such experiments, control experiments analysing the
degradation kinetics at low concentrations should be perf ormed.

EDTA extraction destabilizes the soil matrix by complexing with the cations in the soil e.g.
Ca2+ . The physically trapped substance is then subsequently released which is c onsidered to be
comprised of mainly type I NER. It is noted that EDTA extraction and silylation are releasing
dif f erent substances/f ractions as the release mechanism is dif f erent. Löf f ler et. al. (2020)
provides a detailed description of the EDTA extraction procedure.

The results f rom Step 2 can then be used in DegT50 derivation. The total extractable parent
f raction together with the Type I NER are considered f or deriving the DegT50. If the criterion is
f ulf illed the substance could be considered as P or vP. In this case f urther ref inement of the
assessment could be perf ormed by quantif ying the parent substance concentration in the Type
I NER (Step 3).

In case neither silylation nor EDTA extraction (Step 2) are f easible, normally the Total NER
should be regarded as non-degraded parent substance and taken into account as such f or
DegT50 derivation. However, f urther investigation of the extent of the bioNER f ormation could
be perf ormed using the Microbial Turnover to Biomass (MTB) – model and/or the amino-acid
extraction method. MTB could clarif y whether additional experimental testing to estimate the
amount of bioNER would be relevant. For example, if the amount of bioNER is predicted to be
negligible, then extraction of amino acids, would likely not have additional value to the
assessment. If the amount of bioNER is predicted to be signif icant this might indicate that a
large f raction of the Total NER might be associated with the bioNER and thus extraction of
amino acids could provide some additional inf ormation. The impact of the measured bioNER
f raction to the DegT50 could then need to be investigated. In this case the DegT50 can be
estimated on the basis of the total extractable f raction and the XenoNER (Total NER- NER Type
III (bioNER)). The quantif ication of the bioNER using the extraction of amino acids method is
 Chapter R.11: PBT/vPvB assessment

208 Version 4.0 – December 2023

highly uncertain and thus the results and the DegT50 estimated f rom such investigation can
only be considered undet Weight-of -Evidence approach.

MTB - Microbial Turnover to Biomass

Microbial Turnover to Biomass (MTB) approach is an in silico screening tool and may be used
f or assessing whether type III (bioNER) is likely to be f ormed. It can clarif y whether additional
testing f or the evaluation of bioNER is relevant.

The MTB approach is based on the carbon conversion to microbial biomass and estimates the
microbial yield and bioNER f ormation potential of microbes growing on a specif ic carbon source
(chemical). The estimation of bioNER f ormation with the MTB method is a two-steps process
described in Trapp et al., 2018 and Trapp et al., 2022.

Limitations of the MTB approach:

• Predicted range of the amount of bioNER may deviate f rom the experimental va lues in
e.g. f ollowing situations:

o degradation is incomplete (transf ormation product(s) accumulate),

o degradation is partially or f ully anaerobic,

o signif icant amount of C f rom the substance is stored within carbonates,

o used inoculum does not contain microbes (enzymes) capable of ef f icient

degradation,

o substrate is toxic and inhibits the degradation, or

o the model does not usually take into account the position of labelling. Depending
on the labelling position, the measurement of mineralisation ( 14CO2 evolution)
and of NER can dif f er signif icantly f or the same substance. In particular, position
of the 13C- or 14C label in carboxylic moieties or other highly oxidised positions of
the parent substance should be avoided.

If both the MTB method and other inf ormation (e.g. evolved CO2) indicate that bioNER may be
f ormed and that the quantif ication of the bioNER may be relevant f or the assessment,
extraction of labelled amino acids as described below could be perf ormed.

Extraction of amino acids

The amount of bioNER can be experimentally estimated by f ull hydrolysis (acid hydrolysis with
6 M HCl at 105 °C in closed glass vessels f or 24 hours) of the matrix using isotope labelled
substances with subsequent amino acid analysis. The amount of bioNER can be back-calculated
f rom the amount of amino-acids extracted and identif ied (Kästner et al. 2018).

Both labelling with radioactive ( 14C) (Possberg et al., 2016) and stable isotopes ( 13C, 15N)
(Kästner et al., 2014; Nowak et al., 2011; Wang et al., 2016; Wang et al., 2017.) have been
applied to quantif y the amounts of type III NER using the amino acids extraction method.

Limitations of the extraction of amino acids (including f ollowing analytical methods):

• The method to identif y bioNER type III has not been standardised but rather
represents a method derived f rom basic research.
 Chapter R.11: PBT/vPvB assessment

209 Version 4.0 – December 2023

• Structural identif ication of the amino-acids in the type III NER using stable isotope
labelling f or subsequent stable isotope NMR ( 13C, 15N) is possible but has the
disadvantage that high concentrations of the test substance have to be used due to
the low sensitivity of the NMR method.

Step 3: Quantif ication of parent on Type I NER.

Step 3 (Figure R. 11—15) allows the quantif ication of the parent substance in Type I NER f or
all sampling days/points. The results f rom Step 3 can then be used f or DegT50 derivation. The
total extractable parent f raction together with the parent Type I NER are considered f or
deriving the DegT50. If the criterion is f ulf illed the substance could be considered as P or vP.

Figure R. 11—15: Assessment scheme for Step 3: Quantification of parent substance

in Type I NER.

References

Berns A, Vinken R, Bertmer M, Breitschwerdt A and Schaf f er A (2005) Use of N-15-depleted

artif icial compost in bound residue studies. Chemosphere 2005; 59: 649-658.

Beulke S and van Beinum W (2012) Guidance on how aged sorption studies f or pesticides
should be conducted, analysed and used in regulatory assessments. The Food and
Environment Research Agency. Revised version, July 2012.

ECETOC (2013) Technical Report 117, Understanding the relationship between the extraction
technique and bioavailability.

ECHA (2019) Options to address non-extractable residues in regulatory persistence

assessment. Available at: [Link]
extractable_residues.pdf/e88d4fc6-a125-efb4-8278-d58b31a5d342 ECHA Working group on
Toxicokinetics (2022) Bioaccumulation assessment of air -breathing mammals. Discussion
paper. November 2022.
[Link]
als_en.pdf/56de6276-06e9-9eed-a7dd-a75336fda71b?t=1669388928484 (last access: January 2023)
 Chapter R.11: PBT/vPvB assessment

210 Version 4.0 – December 2023

Haider K, Spiteller M, Dec J and Schäffer A (2000) Silylation of soil organic matter: Extraction
of humic compounds and soil bound residues. In: Bollag JM, Stotzky G, editors. Soil
biochemistry. 10. Marcel Dekker, New York, 2000, pp. 139-170.

Hennecke, D., Kruse, M., Bräutigam, J., Meisterjahn, B., Klein, J., Claßen, D., Trapp, S.,
Kästner, M., Brock, A. L., and Schäf f er, A. (2023) Silylation: a reproducible method f or
characterization of non-extractable residues (NER) of organic chemicals in the assessment of
persistence. Environmental Science: Advances. [Link]

Houba VJG, Temminghoff EJM, Gaikhorst GA and van Vark W (2000) Soil analysis proce dures
using 0.01 M calcium chloride as extraction reagent. Communications in Soil Science and Plant
Analysis 2000; 31: 1299-1396.

[Link]

Kästner M, Nowak KM, Miltner A, Trapp S and Schäffer A (2014) Classif ication and Mo delling of
Nonextractable Residue (NER) Formation of Xenobiotics in Soil - A Synthesis. Critical Reviews
in Environmental Science and Technology 2014; 44: 2107-2171.

Kästner M, Trapp S, and Schäffer A (2018) Consultancy services to support ECHA in improving
the interpretation of Non-Extractable Residues (NER) in degradation assessment. Discussion
paper - f inal report. June 2018.
[Link]
fad2-8ddb-85c09a560f7c

Löf fler D., Ternes T., Martin,A., Albrecht D., Fligg M. and Hogeback J. (2022): Are substances
more persistent than test systems lead to believe? Non-extractable residues: experimental
examination of suitable extraction methods in view of a long-term risk f or the environment.
UBA TEXTE 133/2022, Report No. FB000708/ENG. Wiemann, A. [ed.], ISSN 1862-4804,
German Environment Agency (UBA). Available at:
[Link]

Löf f ler, D., Hatz, A., Albrecht, D., Fligg, M., Hogeback, J., and Ternes, T. A. (2020)
Determination of non-extractable residues in soils: Towards a standardised approach.
Environmental Pollution, 259, 113826. [Link]

Nießner R and Schäffer A (2017) Organic Trace Analysis (Chapter 6). Berlin: De Gruyter, 2017.

Nowak KM, Miltner A, Gehre M, Schaffer A and Kastner M (2011) Formation and Fate of Bound
Residues f rom Microbial Biomass during 2,4-D Degradation in Soil. Environmental Science &
Technology 2011; 45: 999-1006.

Peijnenburg W, Zablotskaja M and Vijver MG (2007) Monitoring metals in terrestrial

environments within a bioavailability f ramework and a f ocus on soil extraction. Ecotoxicology
and Environmental Saf ety 2007; 67: 163-179.

Possberg C, Schmidt B, Nowak K, Telscher M, Lagojda A and Schaef f er A (2016) Quantitative

Identif ication of Biogenic Nonextractable Pesticide Residues in Soil by C-14-Analysis.
Environmental Science & Technology 2016; 50: 6415-6422.

Trapp S, Brock AL, Kästner M, Schäf f er A, and Hennecke H (2022) Critical evaluation of the
microbial turnover to biomass approach for the estimation of biogenic non-extractable residues
(NER). Environmental Sciences Europe 34 (1), 1-13.
 Chapter R.11: PBT/vPvB assessment

211 Version 4.0 – December 2023

Trapp S, Brock AL, Nowak K and Kästner M (2018) Prediction of the f ormation of biogenic non-
extractable residues during degradation of environmental chemicals f rom biomass yields.
Environmental Science & Technology 2018;52(2): 663-672.

Wang S, Seiwert B, Kästner M, Miltner A, Schäf f er A, Reemtsma T, Yang Q, and Nowak M.

(2016) (Bio)degradation of glyphosate in water-sediment microcosms - A stable isotope co-
labeling approach. Water Research 2016; 99: 91-100.

Wang SZ, Miltner A, Kästner M, Schäf f er A and Nowak KM (2017) Transf ormation of
metamitron in water-sediment systems: Detailed insight into the biodegradation processes.
Science of the Total Environment 2017; 578: 100-108.
Wang YF, Xu J, Shan J, Ma YN and Ji R (2017) Fate of phenanthrene and mineralization of its
non-extractable residues in an oxic soil. Environmental Pollution 2017; 224: 377-383.
 Chapter R.11: PBT/vPvB assessment

212 Version 4.0 – December 2023

Appendix R.11- 5: Comparison of HPLC and KOWWIN v1.68 QSAR generated log K ow
values
The log Kow generated by the HPLC-method according to OECD TG 117 (OECD, 2022a) is an
estimation method that is equivalent to theoretical models using descriptive inf ormation (like
chemical structure, i.e. QSARs) to estimate the log K ow . A f irst attempt to compare the
perf ormance of the HPLC method against QSAR estimation of log K ow using 450 HPLC
generated log Kow entries in IUCLID (January 2021) f or which experimental, curated log K ow
data was available, shows that the HPLC method is less accurate in predicting these 450
substances (R2 of 0.87 f or HPLC method vs 0.95 f or KOWWIN v1.68 QSAR, Figure R. 11—16
and Figure R. 11—17) as well as slightly biased (given the y-axis of f set of 0.18 log units f or
HPLC vs 0.03 log units f or the KOWWIN v1.698 QSAR). This is expected to give slight over -
prediction of log Kow values f or substances with a low log K ow , and under-prediction of
substances with log Kow above 4.0. The under-prediction will however be very small close to
log Kow of 4, and increase with increasing log K ow . HPLC and KOWWIN v1.68 QSAR are however
very close to each other in predictivity. This is conf irming observations on the precision of both
HPLC and QSAR generated log K ow estimates in literature; Finizio et al. (1997) showed f or 87
pesticides that HPLC gives good results, whereas the ClogP QSAR (BioByte Corp.) gives
excellent results. It was concluded that QSAR (in this case ClogP) seemed to be the only
reliable method f or substances with log K ow > 5. Hodges et al. (2019) compared HPLC and
QSAR generated log Kow values f or non-ionic, anionic, cationic and amphoteric surf actants.
HPLC generated consistently higher log Kow estimates than the slow-stirring method f or non-
ionic surf actants (OECD, 2022b). QSAR gave very dif f erent results than HPLC f or ionized
surf actants. The ECETOC technical report nr.9 (ECETOC, 1983) concluded that HPLC has been
adequately validated. A method of calculating partition coefficients f rom molecular f ragmental
constants was also noted (already in 1983) as being (as) usef ul (Rekker, 1977; Hansch and
Leo, 1979). For suf ficiently soluble non-polar substances HPLC results are generally within 1
log unit, with the applicability domain in the range of log K ow 0-6. For the extremes (log Kow <0
or >6) it is concluded that the molecular f ragmental constants method (QSAR) is more
trustworthy.

When comparing KOWWIN v1.68 QSAR vs HPLC generated log K ow estimates f or the larger
dataset of 2077 substances in IUCLID f or which no experimental log K ow is available (see
Figure R.11—8) it becomes clear that there can be huge dif f erences between the two
estimation methods. Bicherel et al. (2016) f ound a relatively poor relationship between the log
Kow based on HPLC data and the water solubility compared to shake-f lask and slow stir data,
conf irming limitations of the HPLC derived logK ow data. One of the reasons f or the dif f erences
may be intramolecular hydrogen bonds. The f ormation of intramolecular hydrogen bonds
(IMHBs) competes with intermolecular hydrogen bonding and thus increases the chemical
potential of substances in solvents where intermolecular hydrogen bonds can be f ormed
(water). IMHBs may impact the log K ow by several orders of magnitude. Since EPI Suite does
not consider the potential f ormation of IMHBs, the estimates f or such substances are less
reliable (see e.g. Wang et al., 2011, Buser et al., 2013).

When trying to establish specif ic groups of substances f or which the HPLC and QSAR method
are disagreeing most it turns out that specif ically ionogenic substances and salts are predicted
very dif ferently – probably due to partly neglecting ionization in the QSAR estimates, but also
due to the choice of column in the HPLC method not being ‘f it f or purpose’ of measuring
ionogenic substances.
 Chapter R.11: PBT/vPvB assessment

213 Version 4.0 – December 2023

It is hard to identif y clearly specif ic f unctional groups that are consistently over - or
underpredicted by HPLC or a specif ic QSAR method. Theref ore, taking into account multiple
sources of (predicted) log Kow inf ormation, and weight-of -evidence becomes essential when
estimates are within a log unit below or above the screening value of 4.5.

Therefore, it is advised to always generate QSAR estimations of log K ow together with HPLC
generated estimates, especially if the HPLC generated estimate of log K ow is in the range of one
log unit below or above the screening value of log K ow = 4.5. Given the large scatter observed
when comparing HPLC generated values to KOwWIN v1.68 QSAR generated values (see Figure
R. 11—18), one should always try to include QSAR (including other QSARs than KOWWIN
v1.68) values in the Weight-of-Evidence approach when interpreting an HPLC generated log
Kow estimate. Especially ionising substances, salts, metal complexes and complex mixtures
would benef it f rom additional estimated data (next to HPLC generated data) f rom a variety of
log Kow QSAR models and a Weight-of-Evidence evaluation.

Figure R. 11—16: Predictivity of the OECD TG 117 HPLC method for generating log
Kow estimates using 450 substances with HPLC generated log K ow values in IUCLID
(January 2021) for which experimental data on log K ow is also available. The linear
regression equation is given on top with its R 2 value as a measure of the variance
explained, and plotted as the dotted orange line.
 Chapter R.11: PBT/vPvB assessment

214 Version 4.0 – December 2023

Figure R. 11—17: Predictivity of the KowWIN v1.68 QSAR estimating log K ow for the
450 substances for which HPLC-derived log Kow values were given in IUCLID, and
for which also experimentally determined log K ow values were available. The linear
regression equation is given on top with its R2 value as a measure of the variance
explained and plotted as the dotted blue line.
 Chapter R.11: PBT/vPvB assessment

215 Version 4.0 – December 2023

Figure R. 11—18: HPLC generated log Kow estimates in IUCLID for which no
experimental data was available, compared to the KowWIN v1.68 estimate of log
Kow. The linear regression equation is given on top with its R2 value as a measure of
the variance explained and plotted as the dotted blue line.

References

Bicherel P, Thomas P, Sahigara F (2016) High-accuracy water solubility determination using

logKOW f rom different methods. Poster presented at SETAC Europe 26th Annual Meeting 2016,
Nantes, France.

Buser AM, Schenker S, Hungerbühler K (2013) Comparing the Perf ormance of Computational
Estimation Methods f or Physicochemical Properties of Dimethylsiloxanes and Selected
Siloxanols. Journal of Chemical & Engineering Data 58:11, 3170-3178.
ECETOC (1983) Assessment of reverse-phase chromatographic methods f or determining
partition coef ficients, Technical Report No 9, European Centre f or Ecotoxicology and Toxicology
of Chemicals, Brussels, Belgium. Available at: [Link]
Finizio A, Vighi M, Sandroni D (1997) Determination of n-octanol/water partition coef f icient
(Kow ) of pesticide critical review and comparison of methods. Chemosphere 34(1), 131-161.
[Link]
Hansch, C. and Leo, a. (1979). Substituent constants f or correlation analysis in chemistry and
biology. J. Wiley and Sons, New York.
OECD (2022a), Test No. 117: Partition Coef f icient (n-octanol/water), HPLC Method, OECD
Guidelines f or the Testing of Chemicals, Section 1, OECD Publishing, Paris,
[Link]
OECD (2022b), Test No. 123: Partition Coef f icient (1-Octanol/Water): Slow-Stirring Method,
OECD Guidelines f or the Testing of Chemicals, Section 1, OECD Publishing, Paris,
[Link]
 Chapter R.11: PBT/vPvB assessment

216 Version 4.0 – December 2023

Rekker, R.F. (1977). The hydrophobic f ragmental constant, its derivation and application. A
means of characterizing membrane systems. In: Pharmacochemistry Library, Volume 1., Ed.,
Nauda, [Link]. and Rekker, R.F., Elsevier Scientif ic Publishing Co, Amsterdam.
Wang Z, MacLeod M, Cousins IT, Scheringer M, Hungerbühler K (2011) Using COSMOtherm to
predict physicochemical properties of poly- and perf luorinated alkyl substances
(PFASs). Environmental Chemistry 8, 389-398.
 Chapter R.11: PBT/vPvB assessment

217 Version 4.0 – December 2023

Appendix R.11- 6: Relevance and scientific justification of correction for growth

dilution when deriving BCF
Overview of used terminology

The terminology of the OECD 305 and REACH guidance documents is f ollowed below:

• k2 is the overall elimination rate constant, i.e. the elimination rate constant f itted
directly f rom the experimental data
• kV is the elimination rate constant due to passive dif fusion by gill ventilation. This value
is denoted as k 2 in e.g. Gobas & Lee (2019)
• kE is the elimination rate constant due to f ecal egestion
• kmet is the elimination rate constant due to metabolic transf ormation
• kT is the overall elimination rate constants of these three routes of actual elimination
f rom the body of the f ish (k V+kE+kmet); k2 considers also growth
• kG is the exponential growth rate of the f ish
• k1 is the overall uptake rate constant f rom the aqueous phase into the f ish
• I is the ingestion rate
• α is the assimilation ef f iciency
• Cwater is the concentration in the exposure water
• Cfood is the concentration in the administered f ood
• Cfish is the concentration in f ish
• Vfish is the volume of f ish
• mfish is the mass of f ish
• m is the mass of the substance

Theoretical background

In steady-state, the concentration of a substance in f ish (Cfish) remains constant over time (t).
This is expressed by the f ollowing f ormula:
𝑑𝐶𝑓𝑖𝑠ℎ
=0 (1)
𝑑𝑡

If f ish is considered as a one-compartment model and f ish is not growing, this means that the
sum of all depuration mass f lows is equal to the sum of all uptake mass f lows (either f rom the
aqueous phase or via dietary uptake):

𝑘 1 ∙ 𝐶𝑤𝑎𝑡𝑒𝑟 = 𝑘 𝑇 ∙ 𝐶𝑓𝑖𝑠ℎ (2)

𝛼 ∙ 𝐼 ∙ 𝐶𝑓𝑜𝑜𝑑 = 𝑘 𝑇 ∙ 𝐶𝑓𝑖𝑠ℎ (3)

However, f or growing f ish, the volume of the compartment increases over time and to
maintain a constant concentration (i.e. steady-state), there must be a net mass inf low in the
f ish.
𝑚
𝑑
𝑑𝐶𝑓𝑖𝑠ℎ 𝑉𝑓𝑖𝑠ℎ 𝑑𝑉𝑓𝑖𝑠ℎ 𝑑𝑚
= = 0, 𝑤𝑖𝑡ℎ > 0 𝑎𝑛𝑑 𝑡ℎ𝑢𝑠 >0 (4)
𝑑𝑡 𝑑𝑡 𝑑𝑡 𝑑𝑡

Hence, to maintain steady-state, and to conserve the mass balance, the mass inf low resulting
f rom the uptake f rom the exposure media (water and/or f ood) must equal the sum of the mass
outf low due to total elimination f rom the f ish body and the mass f low compensating f or the
growing f ish. To include f ish growth, the equation f or steady-state is extended to
 Chapter R.11: PBT/vPvB assessment

218 Version 4.0 – December 2023

𝑘 1 ∙ 𝐶𝑤𝑎𝑡𝑒𝑟 = 𝑘 𝑇 ∙ 𝐶𝑓𝑖𝑠ℎ + 𝑘 𝐺 ∙ 𝐶𝑓𝑖𝑠ℎ (5)

It is noted that the rate constant k G does not represent a substance mass transport.

This equation is based on concentrations of the substance in the f ish, e.g. in µg/kgfish/d. If both
sides of the equation are multiplied by the mass of f ish (m fish [kg]), this mass balance (in units
of mass of the substance per day, e.g. µg/d) can be written as:

𝑘 1 ∙ 𝐶𝑤𝑎𝑡𝑒𝑟 ∙ 𝑚𝑓𝑖𝑠ℎ = 𝑘 𝑇 ∙ 𝐶𝑓𝑖𝑠ℎ ∙ 𝑚𝑓𝑖𝑠ℎ + 𝑘𝐺 ∙ 𝐶𝑓 ∙ 𝑚𝑓𝑖𝑠ℎ (6)

With

k1 = (respiratory) uptake rate constant f rom water [L/kg fish/d]

kT = overall depuration rate constant [/d] f rom real depuration processes

Cwater = concentration of the substance in water [mg/L]

Cfish = concentration of the substance in f ish [mg/kg fish]

mfish = mass of the f ish [kgfish]

kG = exponential growth rate constant [/d]

In the one compartment model that describes bioaccumulation in f ish, the parameters k 1 and
kT ref er to mass transf er processes in and out of the f ish and k G describes the increasing
compartment volume in terms of mass of f ish. Thus, to obtain a constant concentration in
growing f ish, k 1 · Cwater has to be greater than k T · Cfish. In steady state, this means that the
apparent bioconcentration f actor (C fish/Cwater) can be described as:
𝐶𝑓𝑖𝑠ℎ 𝑘1 𝑘1
= = (7)
𝐶𝑤𝑎𝑡𝑒𝑟 𝑘𝑇 +𝑘𝐺 𝑘2

This is the equation that f orms the basis f or the growth corrected bioconcentration f actor,
which is included in the OECD technical guideline 305. This equation correctly describes the
steady state situation f or growing f ish expressed on the basis of (substance) mass f lows .

Elimination processes and their relation to fish growth

The purpose of the growth correction is to obtain an estimate of the steady state BCF under
non-growing conditions. It should thus be investigated, whether the rate constants k 1 and kT,
or more precisely k 1/kT, f or the growing f ish are representative f or non-growing f ish too. To
examine this, the parameter k T, which represents the sum of all real elimination processes of
substance mass out of the f ish, should be f urther investigated. k T consists of three separate
elimination processes, being elimination by passive dif f usion by ventilation through the gills
(kV), f ecal egestion (k E) and metabolism of the substance (k met).

𝑘 𝑇 = 𝑘 𝑉 + 𝑘 𝐸 + 𝑘 𝑚𝑒𝑡 (8)

In the case of bioconcentration, uptake is solely through ventilation via the gills. The
ventilation rate is determining the uptake rate constant. It is plausible that this ventilation rate
will be higher in growing f ish due to an increased metabolism (Gobas & Lee, 2019). However,
the ventilation rate is not only determining the uptake rate constant in a BCF test, but also the
elimination rate constant via the gills. In other words, the ratio of the respiratory uptake and
elimination rates is independent of an increased ventilation rate due to growing of the f ish.
 Chapter R.11: PBT/vPvB assessment

219 Version 4.0 – December 2023

From the model f or f ish bioaccumulation, which f orms the basis f or BCFBAF f rom Epi Suite of
the U.S. EPA (Arnot and Gobas 2003), it appears that f or small f ish (1 g) with a lipid content of
5%, the elimination via the gills is a more important route than f ecal egestion f or substances
with a log Kow up to 6.7, emphasizing the importance of elimination through the gills.

Also the f eeding rate is linked to the growth rate, but in the dietary bioaccumulation
experiment this is reversed: the f eeding rate is a controlled parameter, consequently the
growth rate is dependent on the f eeding rate. For bioconcentration experiments with aqueous
exposure this is dif f erent, as the ventilation rate is not a controlled parameter, but rather a
result of the metabolism of the f ish. In addition to the f eeding rate, the f ecal egestion rate
could be assumed to be higher f or growing f ish too. Indeed, in the BCFBAF model (Arnot and
Gobas 2003), the f ecal egestion rate is directly coupled to the dietary uptake rate.

From the limited data presented by Gobas and Lee (2019) on growth rate in relation to f eeding
rate and oxygen consumption, originating f rom dif f erent tests with dif f erent organisms
(rainbow trout and sockeye salmon), it appears that both processes are more or less
proportionally related to growth rate (Figure R. 11—19). In other words, if f ish f eed at a higher
rate, their ventilation rate will increase proportionally. Also, as described above, the f ecal
excretion rate will increase in a more or less similar proportional manner. Depending on the
nature of the substance, one of these two elimination processes is more important (i.e.
elimination via the gills or f ecal egestion), but f or dependence of the overall elimination rate on
growth rate, it seems not to matter too much which of the two processes is dominant.
Oxygen consumption [mg O2/kgfish/d

Feeding rate [kgfeed/kgfish/d]

20 0.04
Feeding rate
Oxygen consumption
15 0.03

10 0.02

5 0.01

0 0.00
0.00 0.01 0.02 0.03 0.04
Growth rate [d-1]

Figure R. 11—19: Relationships of oxygen consumption and feeding rate with growth
rate, based on data presented in Gobas and Lee (2019).
The ef f ect of growth on the third parameter in k T, the metabolism rate constant k met, is largely
unknown. However, if general metabolism is higher in f ast growing f ish, this might have
consequences f or the metabolic transf ormation of xenobiotic substances as well. It seems
reasonable to assume that growth has also an inf luence on the metabolic transf ormation rate
as well. However, in general, the substances f or which growth rate plays an important role on
the BCF are usually not the ones that are metabolized in signif icant amounts.
 Chapter R.11: PBT/vPvB assessment

220 Version 4.0 – December 2023

In general, it can be concluded that with f ast growing fish, not only the uptake parameters are
af fected but the elimination f rom the f ish as well (i.e. k T). For gill ventilation, it is reasonable to
state that the ratio of k 1 and kV is the same f or growing and non-growing f ish. If egestion is
taken into account as well, it might be assumed that the ratio between k1 and the sum of k V
and k E is still rather similar f or growing and non-growing f ish, because both ventilation rate
and f eeding rate seem to be proportionally dependent on growth rate (Figure R. 11—19). For
metabolic transformation, the relationship with growth rate is least well-known, but it seems
reasonable to consider an equal dependence on growth rate as well. In summary the ratio,
between k 1 and k T seems rather independent of the growth of f ish. Theref ore, the ratio
between k 1 and k T can be considered as the estimate of the kinetic BCF f or non-growing f ish.

Summary and implications for bioaccumulation assessment

In conclusion, the kinetic rate constants f or growing f ish will generally be higher than f or non-
growing f ish. The increase in uptake rate constant and elimination rate constants f or mass
transf er in and out of f ish is proportional. Growth is not a mass transfer out of the f ish, but an
increase of the volume of compartment (f ish). To keep the f ish concentration equal (steady
state), a part of the mass of the substance taken up by the f ish is needed to compensate f or
the increasing volume of the f ish. This will result in a shorter period to reach steady-state and
a steady-state situation with overall lower concentrations f or growing f ish than f or non-growing
f ish (and thus a lower apparent steady-state BCF f or growing f ish). Growth correction by
subtracting k G f rom the observed k 2, indicated by k 2G, yields the sum of the overall elimination
rates that denote real elimination f rom the f ish (k T=kV+kE+kmet). This value is almost
proportionally dependent on growth as k 1 and theref ore the overall growth corrected BCF,
which is equal to k 1/(kV+kE+kmet), seems to be a reasonable estimate of the kinetic BCF f or
non-growing f ish. This leads to the conclusion that growth correction should be applied to the
kinetic BCF in case of growing f ish, to have a reasonable estimate f or non-growing f ish.

It should be noted that not only growth and an increased metabolism has its inf luence on the
ventilation rate. Also temperature is a very important parameter f or the ventilation rate.
However, nearly all models that are developed to estimate the respiratory upta ke rate constant
k1 do only account f or the weight of f ish and possibly log Kow . If the models that exist to
estimate a k 1 value are not suitable f or f ast-growing f ish, they are likely to underestimate the
k1 f or f ast-growing f ish. The BCF estimated f rom a dietary OECD 305 test might theref ore be
underestimated if an estimated k 1 value is combined with the high k T value determined
experimentally f or f ast growing f ish.

Verif ication of the procedure to estimate k T as k2G is provided by Brooke and Crookes (2012).
In this report an alternative method was tested in which the mass of the substance in each
individual f ish was f itted against time rather than concentrations. It appeared that there was
an almost one-to-one relationship between k T that was estimated f rom the mass of the
substance in individual f ish and k 2G that was calculated by subtracting the overall growth rate
f rom k2 that was derived f rom the concentrations of the individual f ish. Only at very low values
of k2, when k T approaches zero, the subtraction of the growth rate yields slightly
underestimated values compared to the alternative method. This method of alternativ ely
determining k T to estimate the BCF is also included in the OECD TG 305 (OECD, 2012) and the
Guidance Document to the OECD TG 305 (OECD, 2017).
 Chapter R.11: PBT/vPvB assessment

221 Version 4.0 – December 2023

References

Arnot JA, Gobas FAPC (2003): A Generic QSAR f or Assessing the Bioaccumulation Potential of
Organic Chemicals in Aquatic Food Webs. QSAR & Combinatorial Science 22, 337–345.

Brooke D and Crookes M (2012) Depuration rate constant: growth correction and use as an
indicator of bioaccumulation potential. UK Environment Agency, Bristol, UK. Product code LIT
7371. ISBN: 978-1-84911-283-3
Gobas FAPC, Lee YS (2019) Growth-Correcting the Bioconcentration Factor and
Biomagnif ication Factor in Bioaccumulation Assessments. Environ Toxicol Chem. 2019
Sep;38(9):2065-2072. doi: 10.1002/etc.4509.

OECD (2012), Test No. 305: Bioaccumulation in Fish: Aqueous and Dietary Exposure, OECD
Guidelines f or the Testing of Chemicals, Section 3, OECD Publishing, Paris,
[Link]
OECD (2017) OECD Series on Testing and Assessment Number 264, Guidance document on
aspects of OECD TG 305 on Fish bioaccumulation. OECD Environmental Health and Saf ety
Publications, Series on Testing and Assessment, Paris, France. Available at:
[Link] ety/testing/series-testing-assessment-publications-
[Link]
 Chapter R.11: PBT/vPvB assessment

222 Version 4.0 – December 2023

Appendix R.11-7: Volatilisation correction approaches for the kinetic analysis of

simulation studies
Three dif ferent volatilisation correction approaches f or the kinetic analysis of simulation studies
are presented below. In all of these procedures it is assumed that the volatile losses of parent
are adequately identif ied and quantif ied in the volatile trapping systems. If part of the
volatilised test substance was not trapped and was lost f rom the test system, the results of the
kinetic analysis have more uncertainty. Also losses through other processes, such as
adsorption, increase uncertainties in the kinetic analyses. If repartitioning f rom the headspace
to water (or water-sediment) or soil phase occur, e.g. if the test substance is not trapped
during the test and is lost f rom the test system only when the test bottle is acidif ied and
purged f or CO2 analysis, the volatilisation correction approaches presented below should not be
used as they consider the volatilised f raction as a sink.

A. Separate fitting of data on total dissipation and volatile traps (Verbruggen, E., personal
communication, 16 March 2022) – instead of Appendix 11 of FOCUS guidance (2014)

B. Simultaneous f itting of data on parent substance in water-sediment and volatile traps to

SFO kinetics in OECD 308 using ModelMaker (Jene, 2007b in Annex I of the CLP report
74
on EC 254-938-2 (2018))

C. Simultaneous f itting of data on parent substance in soil and volatile traps to SFO
kinetics in OECD 307 using CAKE (Shrestha et al. 2019)

A. Separate fitting of data on total dissipation and volatile traps

The Appendix 11 of the FOCUS Generic guidance f or Estimating Persistence and Degradation
Kinetics f rom Environmental Fate Studies on Pesticides in EU Registration (2014) includes
correction procedures to account f or volatilisation in the kinetic analysis. However, this
approach where k vol is derived f rom the f it of the amount accumulated in the volatile traps is
incorrect since k vol should give a result close to the k tot f itted f rom the dissipation of the
substance f rom the test system. Deriving k vol f rom the amount accumulated in the volatile
traps may lead to cases where kinetics calculations end up with even slightly negative
degradation rates, because the overall dissipation rate f itted in one way is subtracted from the
overall dissipation rate f itted in another way. Theref ore, a proposal to substitute the FOCUS
Guidance Appendix 11 regarding estimation of k deg is presented in this section. The approach
applies f or cases where both the total dissipation and volatilisation f ollow f irst order kinetics.

In a simulation test, the decrease in concentration of the parent substance in the test system
ref lects the overall dissipation of the test substance in the test system. From the
concentrations of parent substance in the test system no distinction can be made between the
dif f erent processes that have caused this dissipation if the processes occur concurrently. The
pattern of decrease f ollows thus the overall kinetics of the simultaneous processes.

If the two simultaneous processes (degradation and volatilisation) are occurring with the same
kinetic pattern (e.g. f irst-order degradation of the parent substance will be accompanied by a
f irst-order volatilisation of the parent substance), the amount dissipated due to one of the two

74
Annex I to the CLH report (EC 254-938-2)
[Link]
ab04f42ff9e8
 Chapter R.11: PBT/vPvB assessment

223 Version 4.0 – December 2023

processes is a constant f raction of the total dissipation. This holds also true f or the amount
accumulated in the volatile traps. The amount of volatiles trapped at any time is assumed to
be a constant f raction of the amount dissipated f rom the test system. This means that the
accumulated amount f ollows the overall dissipation. However, data should be checked to verify
that this assumption is met.

Hence, f or f irst-order kinetics, if k tot = kvol + kdeg, the pattern of volatilisation accumulated in
the volatile traps thus f ollows k tot and not k vol. That is because the amount that is degraded is
also not available anymore f or volatilisation. Therefore, it is not correct to derive k vol f rom the
f it of the data of the volatile traps.

In the case of f irst-order kinetics, the f ormula f or the total overall dissipation losses can be
expressed, as:

ktot kdeg + kvol

with

ktot is f irst-order overall dissipation rate

kvol is f irst-order overall volatilisation rate

kdeg is f irst-order overall degradation rate

If the starting mass in the test system is m tot, then in principle with a complete mass balance
(e.g. f or radioactivity), the amount of parent substance degraded if all the substance has
dissipated is:

mtot = m deg(∞) + m vol(∞)

mdeg(∞)= m tot - mvol(∞)

where,

mdeg (∞), mass degraded in the test system when all the substance has dissipated,

mtot, starting mass in the test system and,

mvol (∞), mass volatilised and accumulated in the traps when all the substance has dissipated.

The mass of the parent substance volatilised and accumulated in the gas trap can be f itted as:

mvol(t) = m vol(∞)*(1-exp(-ktot*t)

Thus, the amount m vol(∞) can be derived f rom the f it of the volatilised mass in the gas traps.
Because m tot and m vol(∞) are known, m deg(∞) can be calculated.

Because k deg / ktot = m deg(∞) / m tot, then k deg can be calculated as:

kdeg = ktot * (m tot - mvol(∞)) / m tot

In summary, to be able to estimate k deg, what is needed is the overall dissipation rate constant
(ktot), the starting mass of radioactivity (m tot) and the f itted amount of volatilised radioactivity
if all substance has dissipated (m vol(∞)).
 Chapter R.11: PBT/vPvB assessment

224 Version 4.0 – December 2023

For additional clarif ication, intermediate equations and explanations are provided in the Annex
75
I of the ECHA note on Volatile substances available on ECHA Website.

B. Simultaneous fitting of data on parent substance in water-sediment and volatile

traps to SFO kinetics in OECD 308 using ModelMaker
In the OECD TG 308 study (STUDY CA [Link]/4) included in the CLH report of pendimethalin,
ModelMaker (v3.0.4) was used to simultaneously f it the total residue data of the whole system
and the cumulative volatilisation data in order to derive DegT50 that described the volatilisation
corrected total degradation of the substance.

A compartment model was set up to describe the total dissipation indicated by the total
dissipation rate kTOT as sum of the degradation and the volatilisation indicated by the
degradation rate kDEG and the volatilisation rate kVOL. A schematic diagram of the model is
shown below (Figure R. 11—20). The model was implemented in ModelMaker (v4.0) and the Chi2
error level was calculated using the FOCUS kinetics tool FOCUS_DEGKIN_v2.

Figure R. 11—20: Compartment model for the parent in total water-sediment system
(C_TOT) including volatilisation (C_VOL) and sink (elimination compartment)
implemented in ModelMaker (Figure [Link]-6 in the Annex I of the CLP report on EC
254-938-2).

The underlying dif f erential equation system is given by:

Where,

CTOT = total measured concentration in water-sediment system

75
ECHA note: Options to assess persistence of volatile substances in regulatory PBT assessment (2022)
[Link]
e5d0-3eae3e1c26dc?t=1669388686441
 Chapter R.11: PBT/vPvB assessment

225 Version 4.0 – December 2023

CVOL = cumulative volatilisation

CSINK = cumulative degradation products (and other elimination processes, e.g. NER)

kDEG = degradation rate of the system

kVOL = volatilisation rate of the system

kTOT = total dissipation rate of the system

C. Simultaneous fitting of data on parent substance in soil and volatile traps to SFO
kinetics in OECD 307 using CAKE
Shrestha et al. (2019) presented an extended kinetic modelling to enable considering
volatilisation in the modelling of degradation kinetics in OECD TG 307 tests. In the model, the
volatilisation losses are considered as an additional product that neither decline nor repartition
into the soil. The volatilisation is thus treated as a separate sink f or the parent substance, and
it is considered to occur in parallel to the biodegradation. Theref ore, in this extended model
the degradation and the volatilisation of the substance were considered as two processes and
separated so that individual rate constants could be calculated f or the volatilisation process as
well as the degradation process (Figure R. 11—21).

Figure R. 11—21: Structure of the model used in the CAKE tool. Parent: concentration
of parent substance measured in soil; A1: sum of extractable
transformation/degradation products; B1: volatilised parent substance.

In general, the model assumes f irst order kinetics with k as an overall dissipation rate and c
the concentration of test substance according to f ollowing equation:

𝑑𝑐
= −𝑘𝑐
𝑑𝑡
However, f or the extended model it is assumed that k consists of two rate constants kV
(volatilisation rate) and kT (transf ormation rate):

k = kT + kV
 Chapter R.11: PBT/vPvB assessment

226 Version 4.0 – December 2023

To describe the ratio of the two parallel processes the model internally uses “f ractions” F V
(volatilisation f raction) and F T (transformation fraction) which can be calculated based on the
individual rates f or volatilisation and transformation together with the overall decline rate as
f ollows:

𝑘𝑇
𝐹𝑇 =
𝑘
𝑘𝑉
𝐹𝑉 =
𝑘
When the model is run in CAKE, the values f or the two f ractions (F T and F V) are estimated by
the tool. They describe how the optimization tool evaluate the importance of the respective
processes, transf ormation and volatilisation, in the experiment. Based on the f raction f or
volatilisation and the overall DT50 estimated by CAKE, half -lives f or volatilisation (DT50,vol:
half -lif e due to the volatilisation of the substance) and f or transf ormation (DegT50: half -lif e
due to all (primary) transformation processes) can be calculated using the f ollowing equations:

𝐷𝑇 50
𝐷𝑇50,𝑣𝑜𝑙 =
𝐹𝑉
𝐷𝑇50
𝐷𝑒𝑔𝑇50 =
1 − 𝐹𝑉

It is noted that this DegT50 does not only include the f ormation of degradation products but
also other processes (e.g., formation of NER or loss of the substance through adsorption to
test vessel, leakage etc).

References

FOCUS (2014) Appendix 11: Correction procedures to account f or dissipation by volatilisation.

In Generic guidance f or Estimating Persistence and Degradation Kinetics f rom Environmental
Fate Studies on Pesticides in EU Registration.
[Link] vc1.1D
[Link]

Shrestha P, Meisterjahn B, Klein M, Mayer P, Birch H, Hughes CB, Hennecke D (2019)

Biodegradation of volatile chemicals in soil: separating volatilisation and degradation in an
improved test setup (OECD 307). Environ. Sci. Technol. 53 (1), 20e28.
[Link]
EUROPEAN CHEMICALS AGENCY
TELAKKAKATU 6, P.O. BOX 400,
FI-00121 HELSINKI, FINLAND
[Link]