WHITE PAPER

Recommendations for titration methods

validation
The objective of validation of an analytical proce- dures: Text and Methodology [1] and in USP General
dure is to demonstrate that it is suitable for its Chapter <1225> Validation of Compendial Proce-
intended purpose. Recommendations for the vali- dures [2]. The objective of this paper is to provide
dation of analytical methods can be found in ICH some recommendations for the validation of titra-
Guidance Q2(R1) Validation of Analytical Proce- tion methods.

1
 STANDARDIZATION SPECIFICITY

For method validation in titration, titrant standardiza- Specificity is the ability to assess the analyte without
tion is the first step to obtaining the most reliable any interference from other components that might
results. Dilution and weighing errors when preparing be present in the sample. Other components could
a titrant can lead to deviations between the nominal include impurities, excipients, or degradation prod-
titrant concentration and the exact titrant concentra- ucts. It is therefore necessary to show that the analyt-
tion. Furthermore, all titrants (including commercially ical procedure is not affected by such compounds. This
available titrants) will age over time, leading to a can be achieved by spiking the sample with impurities
change in the titrant concentration. Titrant standard- or excipients and demonstrating that the result is
ization is therefore paramount, even if commercially unaffected.
available titrants are used. Additionally, the result of
the standardization can be used to assess the system For titration, this means that either the found equiv-
suitability. alence point (EP) is not shifted by the added impurities
or excipients, or if it is shifted, that a second EP can be
For standardization, either a primary standard or a observed that corresponds to these added compo-
pre-standardized titrant is used. In either case, the nents when using a potentiometric sensor for indica-
standardization step needs to be performed at the tion. If color indicators are used for end point indica-
same temperature as the sample titration, since the tion and a shift is observed, demonstration of the
temperature influences titrant density. Titrants expand specificity can be achieved by a second titration with
in volume at higher temperatures, and thus their titer another suitable color indicator.
factor decreases accordingly. Standardization proce-
dures for the various titrants are described in the In some cases, titration is not specific. An example is
Volumetric Solution section of the USP - NF [3]. when the assay of a substance is done by non-aqueous
titration, and impurities or degradation products have
a similar pKa value to the substance of interest. In such
Primary standards fulfill several criteria, which makes cases, specificity needs to be complemented by other
them ideal for the standardization of titrants. Primary techniques.
standards are of:
Using the assay of potassium bicarbonate by titration
− High purity and stability
with hydrochloric acid [4] as an example, the expected
− Low hygroscopy (to minimize weight changes)
impurity is potassium carbonate. The pKb values for
− High molecular weight (to minimize weighing
potassium carbonate are at approximately 8.3 and
errors)
3.69, meaning it is possible to separate both species
during an acid-base titration. To demonstrate this,
Additionally, they are traceable to standard reference pure potassium bicarbonate as well as a sample spiked
materials (e.g., NIST traceable). with potassium carbonate were titrated with 1 N
hydrochloric acid VS. Figure 1 shows a curve overlay

Figure 1. Curve overlay of the specificity test using 1 g KHCO3 with and without 0.5 g K2CO3 (green and orange = no K2CO3 added; blue
and yellow = K2CO3 added).

2
comparing the titration curves of potassium bicar- range from 50% to 150%, the assay of potassium
bonate both with and without added potassium bicarbonate by titration with hydrochloric acid is
carbonate impurity. The titration curves for potassium highly linear.
bicarbonate alone clearly exhibit only one EP for
potassium bicarbonate, while the titration curves for
16
the solution with potassium bicarbonate and potas-
14 y = 10.055x + 0.1221
sium carbonate have two EPs. The first equivalence 12 R² = 0.9999

EP volume (mL)
point corresponds to the added potassium carbonate, 10
while the second one corresponds to the sum of potas- 8
sium bicarbonate and potassium carbonate. 6
4
2
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
LINEARITY
Sample size (g)

The results of a linear analytical procedure are propor- Figure 2. Linear regression curve for the assay of potassium bicar-
tional to the concentration of the analyte, either bonate.
directly or by a well-defined mathematical transfor-
mation within a given range. As titration is an absolute
method, the linearity can usually be obtained directly.
For this, at least five different concentrations are ACCURACY AND PRECISION
titrated and a linear regression of the sample size
versus the consumed titration volume is established. Accuracy is defined as the closeness of the result to
To evaluate the linearity, the coefficient of determi- the true value. The accuracy contains the information
nation (R2) is used. The recommendation is to use a of the bias of a method and should be established
concentration range from 80% to 120% of the over the complete determination range. Also, the
intended assay weight [2]. accuracy determination of assays is different from
impurity tests. For assays, a reference substance of
For the potassium bicarbonate example, five different known purity is analyzed, while for impurity tests, the
weights ranging from 50% to 150% of the assay sample is spiked with known quantities of the impu-
weight were analyzed in duplicate. The results are rity. The accuracy is then calculated from the recovery
listed in Table 1, and the linear regression plot is of the analyte.
shown in Figure 2. With an R2 of 0.9999 over a weight
Precision contains the information regarding how well
the individual results agree within an analysis of a
Table 1. Linearity determination for the assay of potassium
bicarbonate.
homogeneous sample. The precision is usually
expressed as standard deviation (SD) or relative stan-
Sample Equivalence dard deviation (RSD). Precision is evaluated in three
Sample Assay
weight (%) Point volume levels: repeatability, intermediate precision, and repro-
weight (g) (%)
for linearity (mL) ducibility. Repeatability refers to the precision
50 0.5022 5.1897 102.21 obtained by a single analyst for the same sample in a
50 0.5023 5.1482 101.37 short period of time using the same equipment for all
75 0.7520 7.7571 102.03 determinations. Intermediate precision can be deter-
mined by the analysis of the same sample on different
75 0.7506 7.6197 100.41
days, by different analysts and different equipment,
100 1.0012 10.1627 100.40
if possible, within the same laboratory. Reproducibility
100 1.0026 10.1881 100.51
refers to the precision obtained by analysis of the same
125 1.2599 12.8030 100.51 sample across different laboratories. The reproduc-
125 1.2534 12.7439 100.57 ibility is usually obtained by performing inter-labora-
150 1.5030 15.1888 99.95 tory studies (ILS). For the precision determination, it
150 1.5007 15.2459 100.48 is important that not only the analysis itself but also
all sample preparation steps are done independently
for each analysis.

3
The determination of both accuracy and precision is titrated in triplicate. The results are shown in Table 2.
required because only the combination of both factors With a relative standard deviation of 0.43% over all
ensures that correct results are obtained (Figure 3). nine determinations, the required precision is met. The
obtained assay is close to the manufacturer’s certifi-
cate of analysis (99.9%) with a mean value of 100.40%,
fulfilling the requirements for accuracy.
High Precision
To determine the intermediate precision for the potas-
sium bicarbonate assay, the precision and accuracy
determinations were repeated on a different day on
different equipment. The results are shown in Table 3.
Low Precision The required precision is met, with a relative standard
deviation of 0.43% over all nine determinations.

Low Accuracy High Accuracy

SUMMARY

Figure 3. Only when both precision and accuracy are high can
correct results be obtained, as high precision does not necessarily Method validation of a titration ensures that the
mean good accuracy, and vice versa. selected titration method and parameters will provide
a reliable and robust result. Before the method vali-
dation, it is necessary to standardize the titrant in
For titration, accuracy and repeatability are usually order to achieve accurate results. Method validation
determined together. At least nine determinations at for titration should include determination of the spec-
three different concentration levels are recommended ificity, linearity, accuracy, and precision to obtain a
for determination of both parameters. In the case of complete picture of the suitability of the method for
potassium bicarbonate, 80%, 100%, and 120% of the the analysis of the analyte of interest. In cases where
sample weight suggested in the assay were used. A specificity cannot be met with titration, it is necessary
potassium bicarbonate sample of known purity was to complement the titration by other techniques.
used for the analysis, and each sample weight was

Table 2. Results of the accuracy and precision determinations for potassium bicarbonate.

Sample weight (%) Equivalence Point

Determination Sample weight (g) Assay (%)
for linearity volume (mL)
1 80 0.8066 8.1476 99.91
2 80 0.8092 8.2710 101.10
3 80 0.8069 8.1987 100.50
4 100 0.9978 10.1069 100.19
5 100 0.9953 10.1654 101.02
6 100 1.0072 10.1869 100.04
7 120 1.2094 12.3060 100.64
8 120 1.2067 12.2102 100.08
9 120 1.2171 12.3238 100.15
Mean 100.40
SD 0.436
RSD 0.43%

4
 Table 3. Results of the intermediate precision determination for potassium bicarbonate.

Sample weight (%) Equivalence Point

Determination Sample weight (g) Assay (%)
for linearity volume (mL)
1 80 0.8023 8.2194 101.33
2 80 0.8011 8.1468 100.59
3 80 0.8051 8.1834 100.54
4 100 1.0065 10.1819 100.06
5 100 1.0103 10.2640 100.49
6 100 1.0344 10.5486 100.87
7 120 1.2030 12.1670 100.03
8 120 1.2023 12.1789 100.19
9 120 1.2039 12.1819 100.08
Mean 100.46
SD 0.433
RSD 0.43%

References
[1] ICH Guidance Q2(R1) – Validation of Analytical Procedures: Text and Methodology, ICH, 2005. [Link]

[2] USP. <1225> Validation of Compendial Procedures. In: USP 43 – NF 38. Rockville, MD: USP; 2020.

[3] USP. Volumetric Solutions. In: USP 43 – NF 38, First Supplement. Rockville, MD: USP; 2020.

[4] USP. Potassium Bicarbonate. In: USP 43 – NF 38, First Supplement. Rockville, MD: USP; 2020.

Authors
Subject to change | Layout by RTS Rieger Team, printed in Switzerland by Metrohm AG, CH-9101 Herisau
WP-068EN – 2021-04 © Metrohm AG. All trademarks are the property of Metrohm and its subsidiaries.

Margareth R. C. Marques1, Horacio Pappa1, Michael Chang1, Lori Spafford2, Michael Klein3, Lucia Meier3

1
U. S. Pharmacopeia
2
Metrohm USA
3
Metrohm International Headquarters, Switzerland

Contact
Lucia Meier
Metrohm International Headquarters; Herisau, Switzerland

info@[Link]