Energy & Buildings 330 (2025) 115336

Contents lists available at ScienceDirect

Energy & Buildings

journal homepage: [Link]/locate/enb

Prospective life cycle analysis of a BIPV façade − Life cycle assessment of

greenhouse gas emissions using future projections for a case study☆
Christian Popp * , Clara Uhlig , Michael Engelmann
Institute of Building Construction, TUD Dresden University of Technology, August-Bebel-Straße 30 01219 Dresden, Germany

A R T I C L E I N F O A B S T R A C T

Keywords: BIPV façades activate unused surfaces for energy generation and save conventional façade cladding and therefore
Prospective life cycle analysis case study of resources, emissions and energy. However, these advantages can only be recognised by analysing the entire life
BIPV façade cycle of BIPV facades and comparing their life cycle performance to facades without photovoltaics. In principle,
Prospective LCA of BIPV systems considering
life cycle analyses are based on static boundary conditions and all assumptions are based on current technology,
dynamic grid mix
Prognosis of greenhouse gas emissions
which can clearly limit the results. In this study, scenario-based prospective life cycle analyses are carried out in
compensated by BIPV over lifetime order to forecast the future impact of BIPV façades. Two temporal scenarios (2023 and 2033) and three regional
BIPV as action for climate change mitigation scenarios for the entire PV module production chain (Germany, China and mixed Chinese-German) are
considered. The prospective approach leads to differentiated results in the context of the underlying boundary
conditions and projections compared to a static state of the art approach. The considered progresses from 2023 to
2033 lead to reductions in embodied greenhouse gas emissions of 37.8% (DE), 32.3% (CN-DE) and 30.6% (CN).
The reduction of a German compared to a Chinese production rises from 23.5% in 2023 to 37.8% in 2033. In all
the scenarios analysed, more greenhouse gas emissions are reduced over the life cycle than are emitted. BIPV
façades perform significantly better in comparison with an energetic inactive fibre cement façade.

1. Introduction façades and their often unexploited potential. Forecasting their future
benefits will make it possible to identify so far unused potentials of BIPV
1.1. Motivation and goal façades over conventional – non-energy-generating – façade solutions in
order to integrate BIPV solutions into the planning process at an early
The use of photovoltaics is a key component of green building due to stage, make informed planning decisions and scale up the installation of
the local supply of renewable and low-emission energy, which reduces BIPV façades.
the greenhouse gas emissions1 of buildings over their life cycle. The
application of BIPV2 offers further potential, such as the activation of 1.2. Life cycle analysis of (BI)PV-Systems
unused areas and the substitution of conventional building components.
Calculating the environmental impacts and benefits of BIPV modules can LCA3 is the commonly used method for analysing environmental
help substantiate the advantages of the technology and thereby advance impacts and CO2 eq.-emissions of systems over the whole life cycle.
its implementation. Goal of this study is to find the environmental Different studies have focussed on the factors influencing the green­
footprint of a BIPV façade in comparison with a non-energy-generating house gas emission balance over the life cycle of PV modules and PV
façade system over the whole life cycle and under consideration of the systems. [1] identifies a lack of up to date LCI4-data from pv industry
expected future developments. The findings of the study will help to gain and shows the significant and negative effect of outdated data on the
a better understanding and picture of the life cycle benefits of BIPV results of life cycle analysis of pv systems. By updating the LCI-data, the

☆
This article is part of a special issue entitled: ‘Photovoltaics Built Environment’ published in Energy & Buildings.
* Corresponding author.
E-mail address: christian.popp1@[Link] (C. Popp).
1
Hereafter referred to as CO2 eq.-emissions.
2
Building integrated photovoltaics.
3
Life cycle assessment.
4
Life cycle inventory.

[Link]
Received 31 March 2024; Received in revised form 6 January 2025; Accepted 15 January 2025
Available online 21 January 2025
0378-7788/© 2025 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license ([Link]
C. Popp et al. Energy & Buildings 330 (2025) 115336

global warming potential of the electricity production of a pv system of the PVPS IEA [17].
reduced by a factor of 2 or even more. [2], [3] and [4] have identified Challenging boundary conditions during conducting prospective and
the electricity mix used in production respectively the production of the dynamic LCA are the data availability, data quality and scaling respec­
preliminary products, the service life and the electricity yield during the tively forecasting of data, as well as maintaining comparability and
utilisation phase as the main factors influencing the life cycle CO2 eq.- limiting uncertainties [18]. An exact transfer of the production of a
emissions. Fig. 1 illustrates the different phases in the context of the system into the future requires a high level of effort, as all production
life cycle of a BIPV façade system, highlights the most addressed phases processes of the system as well as the production of all (auxiliary) ma­
and shows the main influences on CO2 eq.-emissions. terials, resources and preliminary products used have to be projected.
Furthermore, [5] and [6] show that the background data in the life However, projecting future developments is always connected with the
cycle analysis of BIPV systems have a significant influence on the results. risk of using data, which could be outdated soon. Due to the lack of data
For example, in a German production scenario, switching the database and the fact, that an assessment of the accuracy and fitting of the chosen
from GaBi to ecoinvent5 leads to a difference in results of 15 % for the projections and data is not possible at the time of the study, uncertainties
climate change indicator and to an increase of 10 % to 169 % for other have to be considered. Furthermore, updating all these influencing
commonly used LCA indicators [5]. It was concluded that the climate variables can only be implemented by editing the background data in
change indicator produces comparable and reliable results and that the the LCI database, which leads to a level of complexity that cannot be
use of the ecoinvent database leads to more conservative assessments. practically realised with conventional LCA tools [19]. For this reason,
This shows that uncertainties can already arise through the choice of this study focuses on the most energy-intensive production processes,
database with its associated background data and with the indicators components and influencing factors by developing individual unit pro­
analysed. As this study focuses on the greenhouse gas emissions of BIPV cesses for these in which the projections are implemented. Updates of
façade systems and uses the CO2 eq.-emissions as target indicator, the the electricity mix, the production technologies and the secondary
higher deviations of other indicators are not relevant. fraction of materials are based on recent projections on grid mix
development, technology trend reports and assumptions based on the
1.3. Dynamic and prospective life cycle analysis current state-of-the-art. The remaining production processes as well as
substance and material flows are mapped with ecoinvent 3.8. The
Most LCAs set a static point of view on the subject of the study implementation of future efficient production processes results in lower
assuming technologies and boundary conditions will remain fixed from material, resource and energy requirements per functional unit pro­
time of the study into the future. Future developments are not taken into duced. Assuming a correct projection of future electricity mixes and
account. However, since a LCA can require an observation period of production processes, the approach represents a conservative forecast
many years, depending on the system being analysed, and technologies due to the unchanged material flows but maps the sensitivity of the
and boundary conditions – such as the electricity mix – are constantly model and the influencing factors.
developing, disregarding changes during the observation period is a The approach includes four models:
clear limitation of such contemporary LCAs. To analyse the impact of
time related aspects prospective, retrospective, ex-ante, ex-post, lab- 1. Model 1 − base model according IEA data (approx. 2012–2018) − static
scale, anticipatory, attributional, consequential and dynamic LCAs state of the art analysis
have been used in literature [7]. The term prospective LCA describes an LCA models for all components were developed on the basis of the
analysis that models a system at a future point in time while dynamic LCI data from IEA. These models and the results of the LCIA8 were
LCA models a system along the timeline from past to present to future compared to literature to validate the models.
[[7]. 2. Model 2 − updated model 1: Production and installation in 2023 − static
Objective of this study is to analyse BIPV façade systems and their state of the art analysis
special context as building component under consideration of the The base models were updated to represent a production in current
changes and developments in future by LCA. This means considering the state of the art. The analysis of the life cycle was carried out in a static
present situation of production, scenarios for future technological de­ approach without adjustments in future processes.
velopments and future developments in the power supply through the 3. Model 3 − updated model 2: Production and installation in 2023 −
grid and their influence on the life cycle performance of these systems. prospective analysis of currently beginning life cycle
Previous and recent studies already addressed the research gap on future To model future developments, the updated model 1 with the
oriented and dynamic LCA of energy and PV systems [1,8–15]. How­ static state of the art approach was adapted. The production
ever, no studies have yet analysed this issue in the applied context of remained unchanged with the electricity mix and technological de­
BIPV façades. velopments of 2023, while the following processes along the life
cycle were modelled taking into account the projections. Therefore,
2. Methodology the electricity mix of 2038 and the predicted material savings were
used as the basis for the production of the new inverter that replaces
2.1. Scope and case study the initial one after 15 years. Furthermore, the electricity mix of
2053 was used for modelling the end-of-life processes. In addition,
The methodological boundary conditions are based on the Method­ the electricity yield of the BIPV façade during operation and the
ology Guidelines on Life Cycle Assessment of Photovoltaic of Task 12 PV electricity feed-in credits were dynamically assessed by applying an
Sustainability of the PVPS6 of the IEA7 [16] and on the prEN electricity mix projection.
15978–1:2021 Sustainability of construction works − Methodology for the 4. Model 4 − projected model: Production and installation in 2033 −
assessment of performance of buildings − Part 1: Environmental Perfor­ prospective analysis of life cycle beginning in future
mance. The used database is ecoinvent v3.8 and LCI–data bases on the Life Analog to the updated model, all processes along the life cycle
Cycle Inventories and Life Cycle Assessment of Photovoltaics Systems 2020 were modelled with technology standards and the electricity mix of
trend predictions for the projected model. The production was set 10
years into the future to 2033 and was modelled with the corre­
5
Both are common and widely used LCA-databases. GaBi is also a LCA- sponding electricity mix. The replacement of the inverters takes
software.
6
Photovoltaic Power Systems Programme.
7 8
International Energy Agency. Life cycle impact assessment.

2
C. Popp et al. Energy & Buildings 330 (2025) 115336

Fig. 1. Life cycle of BIPV façade systems. The greyed-out life cycle stages are not analysed in this study.

place after 15 years, in 2048 and the end-of-life treatment of the neutral society and economy in Germany from 2045 with climate-
modules after 30 years in 2063. neutral production chains, the potential benefit that can be achieved
through recycling in 2053 resp. 2063 is assumed to be close to zero
The analysed BIPV façade consists of a rear-ventilated façade con­ anyway. For the production (A1-A3) in 2023 and 2033 the secondary
struction made up of stainless steel wall brackets, aluminium support fractions of recyclable materials like steel, copper and aluminium are
profiles and aluminium aperture profiles as well as single crystalline updated resp. projected, as an increasing amount of secondary material
glass-glass PV modules. Based on an internally prepared non-published production significantly reduces its CO2-eq. emissions. This is an
analysis of 23 BIPV façade projects, 4.5 kg/m2 aluminium and 0.96 kg/ essential element of the industry’s transformation processes towards
m2 stainless steel are used as material quantities for the substructure. circular economy and greenhouse gas-neutral production. In the pro­
The functional unit is one square meter (m2) of the BIPV façade. An jected scenarios yearly grid emission factors are used for all energy-
assessment time of 30 years is assumed. related emissions in module B6 and module D2.
The investigation covers the entire life cycle − from cradle to grave.
The analysed life cycle stages are shown in Fig. 1. A German, a Chinese
and a mixed Chinese-German production chain of PV modules is 2.2. Projection of future developments
assumed. The substructure, inverter and electric installation is based on
European production. The yields and transports result from the instal­ In a first step, the basic model is updated to current technology and
lation on the south façade of the case study building located in Dresden. manufacturing standards and electricity mix representing 2023. In a
Effects due to age-related replacement are summarized over the entire second step, a projection of the material and energy consumption in the
life cycle and reported in module B4. The assumed lifetimes of the production processes, as well as the used electricity mix are updated for
components are 30 years for the PV modules, 15 years for the inverters, the 2033 scenario. Therefore, the LCI data from IEA is adapted according
60 years for the substructure and 30 years for the cabling and electrical to [17,21,22] to represent the 2023 and 2033 situation for PV module
connection components [16]. The electricity yield of the BIPV façade production chain. The energy consumption of the polysilicon produc­
during use is determined by simulation software and reported in module tion, the silicon demand per wafer, the amount of silver and aluminum
B6 (self-consumption) or module D29 (grid feed-in) depending on the in the metallization pastes and the hydrochloric acid consumption are
yield utilisation. The night-time power consume of the inverters is re­ modified for the PV modules. For the inverter as well as the substructure,
ported in module B6. At the end of life, dismantling (C1), transport (C2), the secondary fraction of aluminum is projected to increase from 55 % in
waste treatment (C3) and waste disposal (C4) of all components are 2023 [23] to 70 % in 2033 to 85 % in 2038 and to 100 % in 2048 as an
taken into account. Outside the system boundaries, in module D only the assumption in this study. Furthermore, an increase in power density is
potential benefits from electricity export (D2) are shown.10 Possible assumed by Chinese Photovoltaic Industry Association (CPIA) from 2.39
benefits through recycling (D1) are not reported in this study, by kW/kg in 2021 to 3.79 kW/kg in 2030 for string inverters [22]. Inter­
applying the cut-off approach, which is in line with the strong sustain­ nally prepared non-published market research, however, resulted in 81
ability concept [20]. Based on the simplified assumption of a climate- data points of state-of-the-art string inverters and an average power
density of 0.34 kW/kg. For this reason, the increase in power density in
this study was assumed to be 0.43 kW/kg in 2023, 0.6 kW/kg in 2033,
9
0.75 kW/kg in 2038 and 1.5 kW/kg in 2048. For this study the mass
For precise reporting module D is split into D1 – potential recycling benefits
ratio of the single components of the inverters is assumed to be constant
and D2 – potential benefits due to energy export.
10 with increasing power density. For stainless steel the secondary fraction
No clear framework is formulated in the currently valid DIN EN 15978:
2012–10 for the balancing of exported energy. The new draft standard E DIN in 2023 is 90 % [24] and is projected in this study to reach 100 % until
EN 15978:2021–09 describes the calculation based on the current, average 2033. For the electric installation an increase of secondary copper from
annual electricity mix as the standard assumption (static approach). The 40 % in 2023 [25] is projected in this study to increase to 75 % in 2033.
assumption of a dynamic change in the electricity mix is included as an addi­ All the chosen assumptions respectively the base for the projections are
tional option for balancing electricity exports. developed from official projection reports. The choice of the projections

3
C. Popp et al. Energy & Buildings 330 (2025) 115336

and its influence on the results is discussed in the conclusion section. A the environmental impact through insulation and heating is not re­
detailed summary of the modelled trends and scenarios is presented in ported. All further positive and negative contributions of the façade to
Table 2 (supplementary material). the CO2 eq.-emissions of the building respectively to the global envi­
The basic model and literature review revealed main factors influ­ ronment over the whole life cycle are reported in the following ways:
encing the environmental impact of PV modules and systems. Due to the
electricity intensive production chain, electricity consumption is a key • “Impact on building”: embodied impacts of the BIPV system are
driver for production of crystalline PV modules. Main factors that in­ allocated based on the energy use (feed-in: 42 % − self-consumption:
fluence these emissions are the electricity mix, the energy demand 58 %), benefits for the building through self-consumption are re­
within the silicon production chain and the amount of silicon needed per ported in B6, benefits in the grid through feed-in are not reported.
PV module. Additionally, the electronic components of the inverter and • “Global impact”: no allocation is carried out, all embodied and
the metal consumption of the inverter and the substructure contribute operational emissions but also the benefits for the building (B6) and
significantly to the CO2 eq.–emissions of the production of a BIPV sys­ outside the system boundaries (feed-in D2) are reported.
tem. For these energy intensive production processes projected elec­
tricity mixes are implemented for the respective years of production. 3. Results and discussion
Several projection studies and reports were used to forecast scenarios
regarding the development of the electricity mix. Fig. 2 summarizes 3.1. Influence of prospective approach on embodied CO2 eq.-emissions
relevant studies for the German electricity mix [26–29]. Results vary
significantly in their projection between an optimistic scenario by UBA Comparing the resulting embodied CO2 eq.-emissions by static state
and a more pessimistic option by Ökobaudat − SMS. of the art (model 2) and prospective (model 3) approach illustrates the
The projections used in this study are based on the MMS scenario of impact of the prospective approach (Fig. 5 for German production chain
the projection report 2023 of the Umweltbundesamt (UBA) for Germany in 2023). Regarding production phase (A1-A3), the prospective
[26] (yellow line in Fig. 2) and the IEA’s Energy Outlook 2018 for China approach has no influence, as the boundary conditions are identical for
(sustainable development scenario) [30]. Both projections contain both scenarios. The influence of the construction phase (A4-A5) itself
detailed data on the mix of domestic electricity generation. When and the application of the prospective approach in the construction
choosing the data, free accessibility and the differentiated composition phase is negligible. When replacing the inverter (B4), with the pro­
of the energy sources of electricity generation were prioritised over the spective approach, the GWP of the inverter is reduced by 76 % in both
currency of the data set. This is discussed in the conclusion section. Both 2023 and 2033 (Fig. 5 – red mark). The total amount of embedded
chosen scenarios project a fast transmission from fossil fuels to renew­ emissions of the BIPV façade system is reduced by 7 % in 2023 and by
able energies, whereby this takes place over a longer period of time in 5.5 % in 2033. The lower reduction in the 2033 scenario can be
China. The future composition of the electricity mix of the two countries explained by the fact that the electricity mix from 2033 onwards expe­
(Fig. 3) is derived from these reports. By modelling the electricity mix in riences a lower reduction in emissions compared to the period from
detail, taking into account the regionality of the electricity generation 2023 onwards (see Fig. 2). In the case of end of life (C1-C4), the influ­
capacities and the voltage levels (low, medium and high voltage) with ence is small, as the waste treatment processes primarily contain ma­
the respective shares and losses, the domestic net electricity generation terial emissions, which were not projected with the present approach.
is mapped for the respective years according to Itten & Fischknecht [31] A longer observation period of more than 30 years will additionally
and Khan et al. [3] based on the forecast scenarios. lead to a replacement of the PV modules, which will emphasise the
impact of the prospective approach. The emissions from module pro­
2.3. Comparison with building component duction in 2033 can be used to estimate the effect. The module
replacement will thus be balanced with 96.78 kg CO2 eq./m2 in the
In order to give the study an application-specific character, a typical static approach and 62.18 kg CO2 eq./m2 in the prospective approach,
small office building according to Klauß [32] is considered as a case which corresponds to a reduction of around 36 %.
study building in Dresden, Germany (Fig. 4). Passive house standard
with an average U–value of 0.13 W/(m2 K) and heating by means of 3.2. Influence of prospective approach on whole life CO2 eq.-emissions
heat pumps and geothermal probes is taken as a basis. The gross floor
area is 1,972 m2. The resulting final energy requirement for heating, If the operational emissions and potential emission reductions over
without cooling, is 18.6 kWh/(m2 a). The energy requirement for the life cycle are analysed using a static and a prospective approach,
lighting and user electricity is 35 kWh/(m2 a). there is a significant difference visible (Fig. 6). Due to the projected
Assuming a PV coverage of 90 % of the opaque area of the south regressive course of the emissions of the electricity mix, the potential
façade, an active PV area of around 350 m2 results. No shading and a emission reductions in the building (B6) and in the electricity grid (D2)
linear degradation in efficiency of − 0.7 %/a are taken into account. through the PV yield decrease over time. A static approach of the
Depending on the efficiency, this results in 77 kWp for sc-Si modules for emission factor of the electricity mix therefore strongly overestimates
2023 (225.6 Wp/m2, eta = 22.8 %) and 83.79 kWp for 2033 (245.5 Wp/ the benefits. The prospective approach captures the dynamic develop­
m2, eta = 24.5 %). Five string inverters with an output of 15 kW each ment of the grid mix over the life cycle within the context of the un­
and a night-time power requirement of less than 5 W are planned. derlying projection data and avoids an overestimation. The reference
Table 1 shows the energy yields over 30 years. façade is only shown once, as the prospective approach has almost no
The BIPV façade is analysed in the specific building context of the visible effects because there are no replacements to account.
typical small office building with the described energy demand. The
BIPV façade system is compared to a conventional rear-ventilated fibre 3.3. Influence of projections on embodied CO2 eq.-emissions
cement façade to give a realistic context with reference to the decision-
making in the planning process of buildings. This reference façade was In 2023 (model 3), production of the BIPV façade system with the full
chosen because of similar construction system, performance in building German PV module production chain achieves emission reductions of
physics and aesthetic design. The environmental impacts are reported 23.5 % compared with production in China (Fig. 7). If only the PV
for every component and all relevant life cycle phases shown in Fig. 1
except module D1. The focus is on the façade – the environmental per­
formance of the rest of the building is not reported. The energetic per­
formance and the insulation of both façades is identical for which reason

4
C. Popp et al. Energy & Buildings 330 (2025) 115336

Fig. 2. Projections of electricity grid mix of Germany until 2050. From 2050 onwards the reached value is assumed to be constant for this study.

Fig. 3. Projection of domestic net electricity mix of Germany and China based on UBA and IEA. Diamond shapes show the fossil content of the electricity mix. Data
for China is extrapolated from 2040 onwards. Data for Germany is assumed to be constant from 2050 onwards.

modules are considered, the savings are 26.6 %.11 In 2033 (model 4), the 29.5 % in 2033.
savings of the overall system are 37.8 % and those of the modules 37.5 The relative distribution of embedded emissions of the individual
%. This shows that, according to the underlying projections, the trans­ components barely changes between 2023 and 2033. The PV modules
formation of the electricity mix in Germany is already further advanced account for the largest share with 65 % (DE) to 72 % (CN) in 2023 and
and will achieve a faster and higher reduction in emissions in the period 63 % (DE) to 73 % (CN) in 2033. The substructure accounts for 16 %
up to 2033 thanks to a higher proportion of renewable energies than in (DE) to 13 % (CN) of the embedded emissions and shows no significant
China. Compared with mixed production chain (CN-DE) a complete change between 2023 and 2033. The embedded emissions of the in­
German production chain reaches reductions of 19.0 % in 2023 and verters account for 11 % (DE) to 9 % (CN) in 2023 and reduce their share
to 8 % (DE) to 6 % (CN) by 2033. Cabling contributes 9 % (DE) to 7 %
(CN) to the embedded emissions in 2023. As only one electricity-based
11 production processes was updated for cabling together with the sec­
This is in line with the reduction of around 28 % by German production
determined by Fraunhofer ISE for the years 2019/2020 [32].
ondary share of copper, the emission reductions up to 2033 are lower

5
C. Popp et al. Energy & Buildings 330 (2025) 115336

and the share in the overall system in 2033 increases to 13 % (DE) to 9 %

(CN). The ratio in the CN-DE scenario is almost identical to the CN
scenario. In principle, the values are plausible in the context of the un­
derlying projection assumptions and modelling. The considered pro­
gresses in the next ten years lead to reductions in embodied CO2 eq.-
emissions of 37.8 % (DE), 32.3 % (CN-DE) and 30.6 % (CN). The
reduction of a German production compared to a Chinese production
rises from 23.5 % in 2023 to 37.8 % in 2033.

3.4. Projections of whole life CO2 eq.-emissions

The embodied emissions from the production of the BIPV façade (A1-
Fig. 4. South-west elevation of case study building [31, S. 98]. A5) allocated to the building according to the PV self-consumption are
1.7 times higher than the fibre cement façade with the German PV
module production chain in 2023 and 1.4 times higher in 2033. Using
Table 1 Chinese PV modules the factor increases to 2.2 in 2023 and 1.9 in 2033.
PV yield and use of energy in the case study building over 30 years. In the mixed production scenario the BIPV façade comes with 2.1 and
Time of installation / Total yield Feed-in Consumption 1.8 times higher embodied CO2 eq.-emissions in 2023 and 2033. How­
model [kWh] [kWh] [kWh] ever, the advantages of BIPV are realised through the PV yield during the
2023 / updated model 1,459,235 609,209 850,025 use phase, which represents a positive counterbalance to the emissions
2033 / projected model 1,503,765 668,645 835,119 caused in the production phase. Depending on the amount of emissions
generated during production and the potentially avoidable emissions,
this results in a negative emissions balance – more emissions are avoided

Fig. 5. Comparison of embodied CO2 eq.-emissions from BIPV façade production with German PV module production chain analysed with static and prospective
approach for 2023.

Fig. 6. Comparison of the static and dynamic accumulated CO2 eq.-emissions of the BIPV façade and the reference façade over life cycle.

6
C. Popp et al. Energy & Buildings 330 (2025) 115336

Fig. 7. Absolute (abs.) and relative (rel.) embodied CO2 eq.-emissions from production of the BIPV façade for the different scenarios of production chain and
installation time.

than emitted (Fig. 8 for German production chain). their embodied CO2 eq.–emissions from production during use phase
Comparing the whole life cycle of the different scenarios shows that and reached negative emissions. This provides benefits for the whole
PV module production in Germany achieves a better overall balance building. Even in 2033 – when the electricity mix contains less emissions
over the life cycle than mixed and entire Chinese module production and the potential benefits gained by energy export are lower – and even
chain due to the lower emissions embedded in production (Fig. 9). for the Chinese production – with higher embedded emissions – the BIPV
Looking at the temporal influence, it becomes clear that the installation façade reaches a negative emission balance and becomes an emission
in 2023 achieves significantly higher emission reductions over the life sink. Under the study’s boundary conditions and the chosen projections,
cycle than in 2033. However, all BIPV façade systems achieve negative BIPV is worthwhile – even in 2033!
emission values in both the global balance and the building balance. As a The conventional reference façade in direct comparison only con­
result, negative emission balances are achieved and all cases can be tributes embedded emissions to the building. It has to be taken into
considered as emission sinks. Reduction factors comparing the life cycle account that the reduction in emissions of the fibre cement façade is
CO2 eq.-emissions of the BIPV façade to the conventional façade result in primarily based on the improvement in the electricity mix and higher
values between − 5.85 (DE 2023) to − 0.43 (CN 2033) from global secondary fractions in the aluminium of the substructure. No further
perspective and − 3.42 (DE 2023) to − 0.18 (CN 2033) from building material-related reductions in cement and in the material consumption
perspective. While reduction factors of between − 5.85 and − 3.08 are itself were recorded in this study. Accordingly, fibre cement façades may
achieved in 2023, the benefits in 2033 are in the range of − 1.15 to also offer further potential reductions that were not recorded. Further­
− 0.18. more, as the fibre cement façade contains concrete and therefore high
embodied emissions the reference façade represents a system with high
4. Conclusion environmental impact. Further façade systems should be analysed for
better comparison and comprehensive conclusions.
The results showed, that all analysed BIPV façade scenarios reduced Nevertheless, the transformation of the electricity mix also reduces

Fig. 8. Balance of CO2 eq.-emissions over life cycle and impact on the building of the BIPV and fibre cement façade for the German production chain and installation
in 2023 (model 3) resp. 2033 (model 4). Environmental impacts and benefits in module B6 are allocated to the building according to the self-consumption rate.

7
C. Popp et al. Energy & Buildings 330 (2025) 115336

Fig. 9. Development of greenhouse gas emissions of the analysed scenarios over the life cycle from building and global perspective.

the potentially achievable benefits of the BIPV system, which is why the updated projection reports for the grid mix from 2024 (UBA and IEA
emissions associated with production are becoming more relevant. In Energy Outlook) prospect faster transition towards a renewable energy
order to minimise emissions, preferably European (BI)PV products supply for Germany and China than the reports used in this study. In this
respectively products with a high ratio of renewable energies in the context, embodied emissions are likely to be reduced more than assumed
entire production chain should therefore be used. According to the used in this study. At the same time, however, the benefits achievable through
projections, the transformation process is taking place at different grid electricity compensation will also be reduced. This is a correlation
speeds in China and Europe. Therefore, the margin of embodied carbon that requires further analyses in order to draw universal, forward-
of PV module production between Germany and China will grow in the looking conclusions on the life cycle greenhouse gas balance of BIPV
next years. Disregarding the current world marked, the silicon and PV systems.
supply chain and their production capacities as well as the economic Furthermore, the projection reports provide different scenarios to
situation in PV industry, it would be beneficial for climate change cover both conservative and progressive future developments. In this
mitigation to produce silicon, wafer, PV cells and PV modules with low study the more conservative scenario for Germany and the more pro­
embodied CO2 eq.–emissions in Germany resp. the European Union and gressive scenario for China were chosen. In addition, the electricity mix
install them in China to improve the Chinese grid-mix, as a lower of individual manufacturers in the PV production chain may differ from
emission intensity is available and further projected in the German that of the domestic electricity grid.
electricity mix. It becomes clear that no ultimate statement can be reached, as it can
With the improvement of the electricity mix, the embedded emis­ be questioned and clarified again with the next projection report and
sions for production, construction phase, replacement and end of life are there is a wide range of influencing parameters and boundary condi­
also reduced significantly, which was quantified by the prospective LCA tions. In order to provide a better understanding of the future-oriented
in this study. As this study focuses on the projection of the electricity life cycle greenhouse gas balance of BIPV systems, a range of results
mix, the improvements of production technologies and the de­ covering the spectrum between conservative and progressive scenarios
velopments in secondary shares of metals, even higher potentials of should be determined. This also shows the need for variant studies and
BIPV façade systems would be determined by an entire prospective sensitivity analyses. These will analyse the impact and sensitivity of
modelling of all material flows of the whole process chain and by individual assumptions, influencing factors and scenarios. This will help
adapting the background data. to assess the robustness of this analysis and provide more objective
The static approach delivers conservative results. By using a pro­ results.
spective approach with reliable and reviewed projections and data a Furthermore, the availability and choice of reliable data is crucial for
more precise picture of the future processes and emissions can be the quality and validity of the findings. For example, high reductions are
reached. This helps to show potentials for the reduction of emissions identified for the inverters with the prospective approach, although the
over the whole life cycle, gives a better assessment of the expected future reduction in power density was chosen conservatively compared to the
situation and supports planning decisions. Chinese technology roadmap [22]. A more precise breakdown of the
However, the regularly published projection reports, there is the projected mass reduction of the individual components would be help­
problem of the limited durability or timeliness of the results of the ful, as these contribute to the overall emissions of the inverters with
prospective analyses, as new projections will in turn provide new input strongly varying intensity. In general more up to date LCI data on the PV
data. Newly published projection reports therefore also offer the op­ modules, the electrical installation and the inverter would be helpful for
portunity to question, extend and update the analyses. For example, the reliable projections.

8
C. Popp et al. Energy & Buildings 330 (2025) 115336

This work was supported by the Federal Ministry for Economic Af­ J. Clean. Prod. 113 (2016) 532–540, [Link]
jclepro.2015.11.052.
fairs and Climate Action on the basis of a resolution of the German
[12] A.A.W. Karl, E. Maslesa, M. Birkved, Environmental performance assessment of the
Bundestag [grant number KK5140504GM1]. use stage of buildings using dynamic high-resolution energy consumption and data
on grid composition, Build. Environ. 147 (2019) 97–107, [Link]
CRediT authorship contribution statement [Link].2018.09.042.
[13] C. Roux, P. Schalbart, E. Assoumou, B. Peuportier, Integrating climate change and
energy mix scenarios in LCA of buildings and districts, Applied Energy, Bd. 184, S.
Christian Popp: Writing – review & editing, Writing – original draft, 619–629, Dez. 2016, doi: 10.1016/[Link].2016.10.043.
Visualization, Validation, Software, Project administration, Methodol­ [14] M. Beraldi, [Position paper] Dynamic life cycle assessment explained, lab
recherche environnement. Zugegriffen: 4. Dezember 2023. [Online]. Verfügbar
ogy, Investigation, Funding acquisition, Formal analysis, Data curation, unter: [Link]
Conceptualization. Clara Uhlig: Writing – original draft, Software, cycle-assessment-explained/.
Formal analysis, Data curation. Michael Engelmann: Writing – review [15] B.A. Wender, Illustrating Anticipatory Life Cycle Assessment for Emerging
Photovoltaic Technologies, Environ. Sci. Technol., Bd. 48, Nr. 18, S. 10531–10538,
& editing, Supervision, Resources. Sep. 2014, doi: 10.1021/es5016923.
[16] R. Frischknecht, P. Stolz, G. Health, M. Raugei, P. Sinha, M. de Wild-Scholten,
Declaration of competing interest Methodology Guidelines on Life Cycle Assessment of Photovoltaic, 4th Edition., Bd.
Report IEA-PVPS T12-18:2020. International Energy Agency - Photovoltaic Power
systems Programme, 2020.
The authors declare that they have no known competing financial [17] R. Frischknecht, P. Stolz, L. Krebs, M. de Wild-Scholten, P. Sinha, Life cycle
interests or personal relationships that could have appeared to influence inventories and life cycle assessments of photovoltaic systems 2020. in PVPS Task
12 PV Sustainanility. 2020. doi: 10.2172/1561526.
the work reported in this paper.
[18] N. Thonemann, A. Schulte, D. Maga, How to conduct prospective life cycle
assessment for emerging technologies? a systematic review and methodological
Appendix A. Supplementary data guidance, Sustainability, Bd. 12, Nr. 3, Art. Nr. 3, Jan. 2020, doi: 10.3390/
su12031192.
[19] A. Liebich, J. Müller, D. Münter, C. Wingenbach, R. Vogt, LCAst – Prospektive
Supplementary data to this article can be found online at [Link] Ökobilanzen auf Basis der ecoinvent-Datenbank, ifeu gGmbH: Institut für Energie-
org/10.1016/[Link].2025.115336. und Umweltforschung. Zugegriffen: 30. März 2024. [Online]. Verfügbar unter:
[Link]
ecoinvent-datenbank.
Data availability [20] R. Frischknecht, LCI modelling approaches applied on recycling of materials in
view of environmental sustainability, risk perception and eco-efficiency, Int. J. Life
Data will be made available on request. Cycle Assess 15 (7) (2010) 666–671, [Link]
[21] IRTPV and VDMA, International Technology Roadmap for Photovoltaic (ITRPV) -
2022 Results, 14th Edition. 2023.
References [22] China Photovoltaic Industry AssociationCPIA, China PV Industry Development
Roadmap 2022-2023, China Photovoltaic Industry Association, Feb. 2023.
[1] R. Besseau, An open-source parameterized life cycle model to assess the Zugegriffen: 28. März 2024. [Online]. Verfügbar unter: [Link]
environmental performance of silicon-based photovoltaic systems, Prog. Photovolt. chinese-roadmap-for-pv-2022-2023-chinese-version.
Res. Appl. 31 (9) (2023) 908–920, [Link] [23] IAI, Public Access, International Aluminium Institute. Zugegriffen: 2. Februar
[2] S. Herceg, I. Kaaya, J. Ascencio-Vásquez, M. Fischer, K.-A. Weiß, L. Schebek, The 2024. [Online]. Verfügbar unter: [Link]
influence of different degradation characteristics on the greenhouse gas emissions public-access/.
of silicon photovoltaics: a threefold analysis, Sustainability, Bd. 14, Nr. 10, S. 5843, [24] B. Bookhagen, Dialogplattform Recyclingrohstoffe - Steckbrief Eisen und Stahl,
Mai 2022, doi: 10.3390/su14105843. Bundesanstalt Für Geowissenschaften Und Rohstoffe (2023), [Link]
[3] A. A. Khan, Global Warming Potential of Photovoltaics with State-of-The Art 10.25928/QYAW-1B33.
Silicon Solar Cells: Influence of Electricity Mix, Installation Location and Lifetime, [25] B. Bookhagen, Dialogplattform Recyclingrohstoffe - Steckbrief Kupfer,
2. Mai 2023, Rochester, NY: 4435731. doi: 10.2139/ssrn.4435731. Bundesanstalt Für Geowissenschaften Und Rohstoffe (2023), [Link]
[4] C. Reichel, A. Müller, L. Friedrich, S. Herceg, M. Mittag, D.H. Neuhaus, CO2 10.25928/ACQR-5Y24.
emissions of silicon photovoltaic modules – impact of module design and [26] R.O. Harthan, H. Förster, K. Borowski, H. Böttcher, S. Braungardt, V. Bürger,
production location, gehalten auf der WCPEC-8, 8th World Conference on Projektionsbericht 2023 für Deutschland. Umweltbundesamt, 2023. Zugegriffen: 6.
Photovoltaic Energy Conversion, Milan, Italy, Sep. 2022. [Online]. Verfügbar September 2023. [Online]. Verfügbar unter: [Link]
unter: [Link] publikationen/projektionsbericht-2023-fuer-deutschland.
publications/conference-paper/wcpec-8/Reichel_5DV234.pdf. [27] ÖKOBAUDAT. 2022. Zugegriffen: 23. Januar 2024. [Online]. Verfügbar unter:
[5] S. Herceg, A.-K. Briem, M. Fischer, P. Brailovsky, T. Dannenberg, M. Held, A [Link]
comparative life cycle assessment of pv modules - influence of database and uuid=fb094f56-89a3-438e-b354-
background system, 2021, Zugegriffen: 24. Juli 2023. [Online]. Verfügbar unter: da2eb2c15631&version=20.23.050&stock=OBD_2023_I&lang=de.
[Link] [28] ÖKOBAUDAT. Zugegriffen: 23. Januar 2024. [Online]. Verfügbar unter: https://
[6] B. Steubing, G. Wernet, J. Reinhard, C. Bauer, E. Moreno Ruiz, The ecoinvent [Link]/[Link]/datasetdetail/[Link]?uuid=aeefc9b4-
database version 3 (part II): analyzing LCA results and comparison to version 2, f414-499a-8f9d-5a5bf07e2421&version=20.23.050&stock=OBD_2023_
The International Journal of Life Cycle Assessment, Bd. 21, S. 1–13, Sep. 2016, doi: I&lang=de.
10.1007/s11367-016-1109-6. [29] U.R. Fritsche, H.-W. Greß, Der nichterneuerbare kumulierte Energieverbrauch und
[7] R. Arvidsson, B. A. Sandén, M. Svanström, Prospective, Anticipatory and Ex-Ante – THG- Emissionen des deutschen Strom- mix im Jahr 2022 sowie Ausblicke auf
What’s the Difference? Sorting Out Concepts for Time-Related LCA, gehalten auf 2030 und 2050, IINAS, Darmstadt, Okt. 2023. [Online]. Verfügbar unter: https://
der SETAC Europe 33rd Annual Meeting, Dublin, Ireland, Mai 2023. [Online]. [Link]/app/uploads/2023/10/IINAS_2023_KEV_THG_Strom-2022_2030-2050.
Verfügbar unter: [Link] pdf.
[8] M. Pehnt, Dynamic life cycle assessment (LCA) of renewable energy technologies, [30] IEA, World Energy Outlook 2018, International Energy Agency, 2018. [Online].
Renew. Energy 31 (1) (2006) 55–71, [Link] Verfügbar unter: [Link]
renene.2005.03.002. 9d93-d81b938217cb/World_Energy_Outlook_2018.pdf.
[9] A. Levasseur, P. Lesage, M. Margni, L. Deschênes, R. Samson, Considering time in [31] R. Itten, R. Frischknecht, M. Stucki, Life Cycle Inventories of Electricity Mixes and
LCA: dynamic LCA and its application to global warming impact assessments, Grid, treeze Ltd., Uster, Schweiz, Version 1.3, Juli 2014. [Online]. Verfügbar unter:
Environ. Sci. Technol. 44 (8) (2010) 3169–3174, [Link] [Link]
es9030003. [32] S. Klauß, Entwicklung einer Datenbank mit Modellgebäuden für energiebezogene
[10] M. Kim, Identifying methods to reduce emission intensity of centralised Untersuchungen, insbesondere der Wirtschaftlichkeit, Zentrum für
Photovoltaic deployment for net zero by 2050: Life cycle assessment case study of a Umweltbewusstes Bauen e.V., Kassel. [Online]. Verfügbar unter: [Link]
30 MW PV plant, Prog. Photovolt. Res. Appl. 31 (2023), [Link] [Link]/BBSR/DE/forschung/programme/zb/Auftragsforschung/
pip.3747. 5EnergieKlimaBauen/2010/DatenbankModellgebaeude/DL_Endbericht.pdf;
[11] C. Roux, P. Schalbart, B. Peuportier, Accounting for temporal variation of jsessionid=8F3599B9E57463E332C9E4734322D864.live21324?
electricity production and consumption in the LCA of an energy-efficient house, __blob=publicationFile&v=1.