The EU Energy Performance of Buildings Directive (Directive 2010/31/EU) poses a major challenge, as it promotes the transformation of existing buildings into nearly zero-energy buildings (NZEB). In this work, we present the case of study of a lecture hall building, owned by the University of Valladolid (Spain), that is currently being refurbished into a NZEB by integration of renewable energy sources (RES), also in line with the requirements from Directive 2009/28/EC. As part of its major renovation, not only Trombe walls and geothermal energy are to be incorporated but also a building-integrated solar photovoltaic (BIPV) system to address the electricity needs and reduce the building’s energy use and GHGs in a cost-effective manner. The environmental profile of this BIPV system has been investigated using life cycle impact assessment (LCIA), assessing the net emissions of CO2 and the damages caused in a comparative context with conventional electricity-generation pathways. In spite of the small power installed in this first stage (designed to cover only an annual energy consumption of about 13,000 kWh, around 6% of the total demand), it can be concluded that significant environmental benefits are gained using this system.
A NZEB is defined in Article 2(2) of the Directive 2010/31/EU as “a building that has a very high energy performance, as determined in accordance with Annex I. The nearly zero or very low amount of energy required should be covered to a very significant extent by energy from renewable sources, including energy from renewable sources produced on-site or nearby”. This directive, adopted in 2010 and translated into RD 235/2013 in Spain, has led EU countries to significantly step up their efforts to take advantage of the opportunities presented by nearly zero-energy buildings, according to COM(2013)483 report. Further, the Commission recently published a Recommendation (EU) 2016/1318 on guidelines for the promotion of NZEBs and best practices to ensure that, by 2020, all new buildings are NZEBs. This emphasis can be mainly ascribed to the importance of the building sector for energy efficiency improvements and its contribution to energy security and the 2030 framework for climate and energy policy.
In parallel to requirements for new buildings, Article 9(2) also requires EU member states to put in place support policies to stimulate the refurbishment of existing building stocks—old and inefficient and frequently renovated at a slow pace—towards nearly zero-energy levels. Nonetheless, in contrast to new public-owned buildings (for which the NZEB concept will apply by 2019), in this case, there are no target dates or an obligation to set minimum energy performance requirements [
The University of Valladolid (UVa), a public university in the autonomous region of Castile and Leon (Spain), has made significant efforts in the past years to improve the energy efficiency of its buildings. The best example has been the LUCIA building, which has a zero CO2 balance (i.e., it is a real zero energy building, ZEB) and uses 100% RES, leading to a 98-point (Platinum) rating in the LEED (Leadership in Energy and Environmental Design) certification, ranking as the most efficient building in Europe and the second most efficient in the world (after Pixel building, in Australia) in 2015. Furthermore, it was chosen as a highlighted case in March 2016 by Build Up (the European portal for energy efficiency in buildings) as a showcase pioneer energy autonomous public building.
The UVa has now undertaken the transformation of several other existing buildings by implementing passive and bioclimatic design approaches to reduce energy demand in maintenance as well as only using on-site renewable energies (biomass, photovoltaic, and geothermal). One of these buildings is the lecture hall, in which on-site geothermal pipes are to be installed for both heating and cooling, in combination with passive heating systems (Trombe walls). Moreover, PV panels have also been integrated in the façade to cover the electricity needs. It is worth noting that, in terms of RES, PV panels are the most common option, with nearly 70% of the NZEB examples using them [
Analyses of photovoltaic applications in ZEBs and NZEBs have received significant attention in the past two years (e.g., [
Taking into consideration that PV technology is associated with some environmental concerns (the production of PV cells is accompanied by a high rate of emissions during manufacturing, consequently causing a high impact on the environment, and the PV industry utilizes a variety of chemicals, many of which are relatively toxic to the human health and the environment, as noted by Menoufi et al
The lecture hall building (Figure
General view of the lecture hall building before the installation of the PV panels.
In a first stage of the transformation of the building, PV panels were to be installed as shades in its façade (Figure
General view of the lecture hall façade after the installation of the PV panels (a) and detail of the PV panels used as shades (b).
Monthly average global and diffuse horizontal irradiance experimental values [
Horizontal irradiance (kWh/m2/month) | Temperature (°C) | |||||
---|---|---|---|---|---|---|
Experimental [ |
PVsyst | |||||
Global | Direct | Global | Direct | Diffuse | ||
January | 57.6 | 31.2 | 56.5 | 26.9 | 30.5 | 2.0 |
February | 90.3 | 54.3 | 77.8 | 45.1 | 32.7 | 3.50 |
March | 134.1 | 84.6 | 127.7 | 76.1 | 51.1 | 7.30 |
April | 165.9 | 105 | 152.1 | 91.3 | 60.8 | 9.80 |
May | 194.4 | 123 | 185.7 | 115.1 | 70.6 | 14.6 |
June | 226.5 | 160.8 | 207.9 | 145.5 | 62.4 | 19.9 |
July | 232.5 | 174.3 | 213.0 | 161.9 | 51.1 | 22.7 |
August | 203.7 | 147.6 | 185.4 | 137.2 | 48.2 | 22.3 |
September | 157.8 | 110.7 | 139.8 | 99.3 | 40.5 | 18.2 |
October | 100.8 | 62.7 | 92.7 | 57.5 | 35.2 | 12.6 |
November | 65.4 | 36.3 | 59.7 | 31.0 | 28.7 | 6.6 |
December | 49.2 | 25.5 | 48.1 | 23.6 | 24.5 | 3.3 |
Average | 139.9 | 93.0 | 128.9 | 84.2 | 44.7 | 11.9 |
Total | 1678.2 | 1116 | 1546 | 1010 | 536 |
In the design of the BIPV system, the optimum components’ sizing and the orientation and inclination of the solar panels were evaluated and validated using PVsyst v.6 (PVsyst SA, Satigny, Switzerland) software (in line with other studies [
Examples of the shading analyses conducted with PVsyst: shading without panels (a), with the panels only above the top row of windows (b), and with the chosen solution—in which the panels are distributed above both rows of windows—(c) on 21st of June at different times (from top to bottom: 10 am, noon, 2 pm, and 4 pm).
It should also be clarified that, provided that the degradation of the modules reduces efficiency over the lifetime, a linear degradation declining to 80% of the initial efficiency at the end of a 30-year lifetime (i.e., 0.7% per year or 10% on average during the entire lifetime [
The reasoning behind the LCA is that the electricity produced by the PV installation will replace the same amount of electricity in the existing grid. If the environmental impact of the PV installation per kWh is smaller than the one for the grid electricity production, there will be a net saving of environmental damage.
LCA is not a performance indicator but rather a research method used for the quantitative assessment of material used, energy flows, and environmental impacts of products. It has been widely applied in the building industry, because it cannot only provide a more comprehensive and reasonable analysis on the energy and environment impact for the whole life cycle but also be used to determine top design priorities and quantitatively inform sustainable design decision making for various buildings [
As regards the specific LCA analysis of PV systems, the most popular approach in the literature has been the Energy Payback Time (EPBT), which may not be sufficient, as it does not provide a comprehensive environmental performance prospective [
The selected attributional LCA (ALCA) approach provides a compilation of all relevant inputs and outputs, as well as an evaluation of the feasible environmental impacts. Life cycle assessments based on the standard specification ISO 14040:2006, as the one presented herein, consist of four interdependent elements: (i) the goal and scope definition, (ii) the life cycle inventory (LCI) analysis, (iii) the life cycle impact assessment, and (iv) the interpretation of the results. Due to the associated complex environmental modeling, the LCIA step is the most critical and data intensive one, but it can be dealt with through different LCA software programs. In this case, SimaPro 8 (Pre Sustainability, Amersfoort, The Netherlands) software was chosen. The most recent International Energy Agency (IEA) PVPS Task 12 guidelines [
The functional unit for the LCA is defined as 1 m2 of module area (the most usual for quantifying environmental impacts of buildings or for quantifying energy gains on roofs), but 1 kWh of electricity fed into the grid will also be used in the discussion for the comparison of the PV system as a replacement of the set of energy resources used in the Spanish power grid mix.
Regarding the system boundaries of the study, a “cradle to gate” approach was chosen, considering the whole life cycle of the BIPV system, including all expenses to produce required energies, materials, and auxiliaries (raw material acquisition, materials processing, and manufacturing phases); the transportation of produced and used modules; all inputs involved in the installation of the PV system; its maintenance during the utilization phase; and its treatment/disposal/recycling at the end of the system’s life. The end-of-life phase has been modelled according to Latunussa et al
In relation to the geographical scope, production was considered for various countries representing the actual cell and module production sites. All datasets on used materials and energies were based on country representative datasets. The use phase was assumed to be in Valladolid (Spain), as noted above.
The life cycle inventory was carried out using with Ecoinvent 3.1 and ELCD (European reference Life Cycle Database) databases (updated in June 2016, based on 2015 LCI data).
In respect to the life cycle impact assessment, as noted above, two damage-oriented LCIA methodologies were considered: the EI-99 methodology, which is taken as the reference methodology, and EPS 2000 methodology. LCIA methodologies differ in some parameters: the approach of modeling the environmental impact, the impact categories, the endpoint and damage categories considered, characterization, normalization, and weighting factors. Applying various methodologies assists in having a more comprehensive image of a system’s environmental performance and its relative effects on the different environmental areas of protection. According to the structure of the version used of the EPS 2000 methodology, only characterization and weighting were considered, disregarding the normalization step, whereas in the EI-99 methodology characterization, normalization and weighting were considered. As in the work by Menoufi et al
In connection with the carbon footprint, the global warming potential (GWP) of air emissions associated to the life cycle of the BIPV system was assessed with the IPCC 2013 method, which lists the climate change factor with a timeframe of 20 and 100 years and is based on the Intergovernmental Panel on Climate Change 2013 report [
A system consisting of customized semitransparent glass-Tedlar® modules, supplied by Onyx Solar (Ávila, Spain), was chosen. Four strings of 25 photovoltaic glasses each, adding up to 100 connected modules, were connected to an IGPLUS 150 V-3 (Fronius, Pettenbach, Austria) inverter. Out of the 25 photovoltaic glasses in each string, 18 were large panels (1.430 × 850 mm) with 28 6
The BIPV system accounts for a total surface of 106 m2 and has a peak power of 9.7 kWp, with an estimated annual power generation of ca. 13 MWh. Panels were installed with a 31° tilt and a 14° S azimuth (Figure
The impact assessment results per 1 m2 of module area from EI-99, which is the reference methodology used in this research, are summarized in Table
Impact assessment results using the EI-99 methodology per 1 m2 of module area (itemized by impact category).
Safeguard subject | Impact category | Damage (Pt) | |||||
---|---|---|---|---|---|---|---|
Manufacturing of PV panels | Installation† | Maintenance | Dismantling | Recycling | Total | ||
Human health | Carcinogens | 22.868 | 0.767 | 9.088 | 0.743 | 0.016 | 33.482 |
Respiratory organics | 0.006 | 0.023 | 0.004 | 0.001 | 0.000 | 0.035 | |
Respiratory inorganics | 7.843 | 22.244 | 4.563 | 2.293 | 0.116 | 37.059 | |
Climate change | 0.675 | 3.086 | 0.530 | 0.337 | 0.017 | 4.645 | |
Ionizing radiation | 0.004 | 0.023 | 0.005 | 0.004 | 0.000 | 0.036 | |
Ozone layer depletion | 0.000 | 0.003 | 0.000 | 0.000 | 0.000 | 0.003 | |
|
|||||||
Ecosystem quality | Ecotoxicity | 2.209 | 0.809 | 0.547 | 0.212 | 0.003 | 3.780 |
Acidification/eutrophication | 0.158 | 0.380 | 0.175 | 0.041 | 0.002 | 0.755 | |
Land use | 0.093 | 0.393 | 0.056 | 0.024 | 0.001 | 0.567 | |
|
|||||||
Resources | Minerals | 1.347 | 0.542 | 0.261 | 0.152 | 0.001 | 2.304 |
Fossil fuels | 1.369 | 7.268 | 2.230 | 0.669 | 0.042 | 11.577 | |
|
|||||||
Total | 36.572 | 35.537 | 17.460 | 4.474 | 0.199 | 94.242 |
†Includes not only the transport and installation of the PV panels but also the manufacturing of all the other components of the BIPV system.
Table
Impact assessment results using the EPS 2000 methodology per 1 m2 of module area (itemized by impact category).
Safeguard subject | Impact category | Damage (Pt) | |||||
---|---|---|---|---|---|---|---|
Manufacturing of PV panels | Installation† | Maintenance | Dismantling | Recycling | Total | ||
Human health | Life expectancy | 80.980 | 23.586 | 12.366 | 7.489 | 0.404 | 124.825 |
Severe morbidity | 15.205 | 3.125 | 2.227 | 1.311 | 0.072 | 21.940 | |
Morbidity | 2.323 | 0.569 | 0.406 | 0.253 | 0.013 | 3.565 | |
Severe nuisance | 3.051 | 5.340 | 0.817 | 0.114 | 0.002 | 9.324 | |
Nuisance | 1.150 | 0.601 | 0.261 | 0.123 | 0.006 | 2.141 | |
|
|||||||
Ecosystem production capacity | Crop growth capacity | 0.124 | 0.036 | 0.053 | 0.013 | 0.001 | 0.227 |
Wood growth capacity | −0.552 | −0.134 | −0.129 | −0.062 | −0.003 | −0.880 | |
Fish and meat production | −0.033 | −0.008 | −0.014 | −0.003 | 0.000 | −0.058 | |
Soil acidification | 0.026 | 0.014 | 0.007 | 0.003 | 0.000 | 0.050 | |
Prod. cap. irrigation water | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 | |
Prod. cap. drinking water | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 | |
|
|||||||
Abiotic stock resource | Depletion of reserves | 802.403 | 590.150 | 248.804 | 57.317 | 0.824 | 1699.498 |
|
|||||||
Biodiversity | Species extinction | 0.581 | 0.157 | 0.122 | 0.080 | 0.003 | 0.943 |
|
|||||||
Total | 905.258 | 623.436 | 264.921 | 66.639 | 1.321 | 1861.575 |
†Includes not only the transport and installation of the PV panels but also the manufacturing of all the other components of the BIPV system.
On the subject of the impact categories, the total impact score obtained with the EI-99 methodology is dominated by two impact categories associated with the damage to human health, namely, carcinogens (35.5% of the total impact) and respiratory inorganics (39.3%), followed by fossil fuels (12.3%). This is attributed to the carcinogenic substances used and the emissions induced (which directly affect the respiratory health system) during the processing of the PV system constituting components. The depletion of the fossil fuel resources indicates the surplus energy that will be needed by the future generation in order to extract fossil fuels and use it to manufacture the corresponding components.
If instead we refer to the EPS 2000 analysis, one impact category prevails: the depletion of reserves (91.3%), followed by life expectancy (6.7%), and severe morbidity (1.2%). The high score of the depletion of reserve impact category is explained by the high amount of fossil fuels extracted and used for the manufacturing of the used materials, while the life expectancy impact category reflects the expected shortening of average individual lifetimes (years of lost life) due to the impact of the corresponding manufacturing processes.
The differences between the results from the two methodologies should be ascribed to the fact that EPS 2000 method puts a great emphasis on the manufacturing phase, while EI-99 takes into consideration the entire life cycle. As mentioned above, the EI-99 methodology is considered as the reference, and EPS 2000 methodology is used in order to check and compare the coherency of the environmental performance results from another methodology perspective.
Upon comparison of the BIPV system as replacement of the set of energy resources used in the power grid mix (Figure
Environmental impacts ascribed to electricity consumption from Spanish power grid and from on-site PV panels, according to EI-99 methodology.
With regard to the carbon footprint analysis, the climate change factor with a timeframe of 100 years was assessed using the IPCC 2013 methodology. The results indicate that the manufacturing phase would account for ca. 66.8% of the total emission, followed by the installation (14.3%), the maintenance (11.2%), and the end-of-life phase (7.6% in the dismantling and 0.3% in the recycling). Throughout its 30-year lifetime, the system would lead to emissions of 52,716 kg CO2eq. Since the total power generation of the BIPV system over its 30-year lifetime would add up to 350,033 kWh (taking into consideration the decrease in efficiency due to degradation of the panels, as noted above), the emission factor would be 0.15 kg CO2eq/kWh. This value is in good agreement with those reported for a BICPV system for different cities (Barcelona, Seville, Paris, Marseille, London, and Aberdeen) [
For the same electricity consumption, using the Spanish power grid mix would lead to emissions of 0.456 kg CO2eq/kWh (an average of 0.257 kg CO2/kWh over the 2013–2017 period + the carbon dioxide equivalents of all non-CO2 gases (namely, CH4, N2O, SF6, HFCs, and CFCs)). Thus, the carbon footprint would be reduced by 67%, thanks to the implementation of the BIPV system.
With reference to the obtained average life cycle greenhouse gas (GHG) emissions for the system (150 g CO2eq/kWh), it is almost twice the average value for mono-Si PV technology reported by Nugent and Sovacool [
Apropos of the LCE analysis conducted with PVsyst, it yielded a value of emission savings of 3.54 tons CO2/year, which implies that it only considered the mix LCE (ca. 0.27 kg CO2/kWh) and neglected the carbon footprint associated with all the PV system life cycle phases except for the use phase.
In estimating potential environmental impacts, LCA, by its very nature, associates with uncertainties [
Impact of BIPV system lifetime assumption on the emission factor, considering a reduction in the module efficiency of 0.7% per year.
System lifetime (years) | Power generation over lifetime (kWh) | Total GHG emissions (kg CO2) | Emission factor (kg CO2/kWh) | Δ emission factor† (%) |
---|---|---|---|---|
25 | 296,694 | 51,729 | 0.178 | 15.77 |
26 | 307,513 | 51,926 | 0.171 | 12.12 |
27 | 318,255 | 52,124 | 0.166 | 8.75 |
28 | 328,922 | 52,321 | 0.160 | 5.62 |
29 | 339,515 | 52,519 | 0.155 | 2.92 |
30 | 350,033 | 52,716 | 0.151 | — |
31 | 360,478 | 52,914 | 0.146 | −2.53 |
32 | 370,850 | 53,111 | 0.142 | −4.91 |
33 | 381,149 | 53,309 | 0.138 | −7.13 |
34 | 391,376 | 53,506 | 0.135 | −9.22 |
35 | 401,531 | 53,704 | 0.131 | −11.19 |
†Taking a 30-year system lifetime as a reference.
Lamnatou et al
The degradation rate can also lead to uncertainty in the LCA. Table
Impact of the module degradation rate on the emission factor, for a 30-year system lifetime.
Reduction in module efficiency per year (%) | Power generation over lifetime (kWh) | Emission factor (kg CO2/kWh) | Δ emission factor† (%) |
---|---|---|---|
0.3 | 370,484 | 0.142 | 5.52 |
0.4 | 365,228 | 0.144 | 4.16 |
0.5 | 360,069 | 0.146 | 2.79 |
0.6 | 355,005 | 0.148 | 1.40 |
0.7 | 350,033 | 0.151 | — |
0.8 | 345,152 | 0.153 | −1.41 |
0.9 | 340,361 | 0.155 | −2.84 |
1 | 335,656 | 0.157 | −4.28 |
1.1 | 331,038 | 0.159 | −5.74 |
†Taking a 0.7% reduction in module efficiency per year as a reference.
In the matter of the failure rate of modules, a default 8% was chosen for the initial analysis. As shown in Table
Impact of the module failure rate on the emission factor and on EI-99 impact estimation.
Failure rate (%) | Total GHG emissions (kg CO2) | Emission factor (kg CO2/kWh) | Δ emission factor† (%) | EI-99 impact for entire system (kPt) | Δ EI-99 impact† (%) |
---|---|---|---|---|---|
13 | 54,722 | 0.156 | 3.81 | 9.781 | 2.09 |
12 | 54,278 | 0.155 | 2.96 | 9.811 | 1.79 |
11 | 53,643 | 0.153 | 1.76 | 9.862 | 1.28 |
10 | 53,195 | 0.152 | 0.91 | 9.909 | 0.81 |
9 | 53,164 | 0.152 | 0.85 | 9.920 | 0.70 |
8 | 52,716 | 0.151 | — | 9.990 | — |
7 | 52,084 | 0.149 | −1.20 | 10.041 | −0.51 |
6 | 52,053 | 0.149 | −1.26 | 10.052 | −0.62 |
5 | 51,605 | 0.147 | −2.11 | 10.103 | −1.13 |
4 | 51,158 | 0.146 | −2.96 | 10.169 | −1.79 |
3 | 50,942 | 0.146 | −3.37 | 10.221 | −2.32 |
†Taking an 8% failure rate as a reference.
The influence of spectrum and latitude on the annual optical performance would also have an impact on the overall performance of the system and is another source of uncertainty. It should be stressed that the scarcity of case studies on LCA of BI solar thermal, BIPV, BIPVT, and BICPV systems—showcased by Lamnatou et al
The environmental impact of conventional residential photovoltaic systems in regions with low solar irradiation (900–1000 kWh/m2/year, a typical value for Northern Europe and Canada) was investigated by Laleman et al
Amongst the studies on LCA of mono-Si BIPV systems in different climatic conditions, we may refer, for example, to those by Seng et al
Concerning more recent studies on BICPV systems (which show about 13% lower impact per kWp than the BIPV systems without concentration due to the fact that there is less PV cell material [
Finally, another important aspect in terms of the uncertainty of the LCA would be related to the shading effect of the BIPV system. Given that the façade of the lecture hall is oriented to the south, there were significant heat and glare problems in the classrooms that led to the lowering of blinds and the use of artificial light during the day. The installation of the BIPV shading panels above the windows, projecting out for a meter from the façade, has allowed full visibility from the inside while blocking off direct sunlight, significantly reducing aforementioned unwanted heat and excessive glare in the classrooms. Since the use of artificial light will thus be reduced, the electricity consumption of the building from now onwards should actually be lower than the values indicated in Section
As part of a framework strategy for a resilient energy union with a forward-looking climate change policy, the EU has placed emphasis both on the construction of new NZEB buildings and on the transformation of existing buildings into NZEBs. The design of a PV system integrated into one of these later buildings (a lecture hall of the University of Valladolid, Spain), validated with PVsyst tool, is presented. The LCA of the BIPV system over a 30-year span was conducted using two damage-oriented LCIA methodologies (EPS2000 and EI-99), which agreed in the significant impact associated with the wiring system, close to that of the PV manufacturing. The assessment of the BIPV system environmental impact as a replacement of the energy resources used in the Spanish power grid mix indicated that a significant impact reduction (53%) would be achieved through the installation of the BIPV system. The results from the IPCC 2013 methodology further supported this claim, since the PV system led to a 67% lower carbon footprint.
The authors declare that there is no conflict of interest regarding the publication of this article.
The authors would like to gratefully acknowledge the financial support by Consejería de Educación de la Junta de Castilla y León through Project no. VA029U16.