Towards Reduced Solar PV Industry Waste: A Feasibility Assessment of Novel Ownership Models

Since 2010, solar Photovoltaic (PV) has been the single fastest growing power generation technology worldwide. However, given that the useful lifetime of a PV installation currently stands at 25 years and that current industry End-of-Life (EOL) management techniques, focus primarily on recycling and disposal, it has been estimated that by 2050, there will exist 78 million tonnes of hazardous solar PV waste.

One potential solution that could aid in mitigating this impending environmental crisis, is determining whether or not the lifetime of commercial and residential solar PV installations can be elongated from the industry standard of 25 years to 50 years. Two novel solar PV ownership models, “The Cascading Tiered Commercial Ownership model” (CTCO) and “The Elongated Residential Ownership model” (ERO) have been created by projecting the technical outputs and economic Net Present Values (NPV) of a 60 kwp commercial and 4.8 kWp residential installation operated over a 50 as opposed to 25 year period.

As expected, the Business as Usual (BAU) model which required that the commercial residential installations be decommissioned and replaced at 25 years, produced more energy over a 50 year period than both lifetime elongation models. However, the cascading tiered ownership model and the elongated residential ownership model had an NPV that was higher than the BAU model. Feed-in-Tariff (FIT) analysis identified that a rate of more than €0.25 per kWh would be required for the BAU model to be favoured, while the rate of module degradation favoured the elongated ownership model for all rates under 2% per year. Alterations to the FIT at 25 years assuming preference for environmentally sustainable business models, led to a greater disparity in results in favour of the novel ownership models. Irradiation levels only favoured the BAU ownership model when in excess of 1750 kWh/m2. Altogether, the projected technical output of both hypothesised ownership models suggests that elongation of PV system lifetime is economically advantageous and should be considered as a viable alternative to other models in both commercial and residential market segments.

Energy demand is increasing throughout the world due to unprecedented population growth and economic development [1]. While non-renewable energy sources have the capacity to meet present demands, the finite aspect and adverse environmental effects associated with them is driving global strategy towards adoption of sustainable energy generation sources [2]. The solar PV industry has long been identified as having the potential to capture a significant proportion of total global energy generation, in both commercial and residential market segments [3,4]. In 2019, solar PV generation surpassed bioenergy as the third-largest renewable electricity technology after onshore wind and hydropower, accounting for 2.7% of total global electricity supply [5]. In addition, global solar PV annual new installations surpassed wind in 2016, with 78gw of new installed capacity relative to 55gw. Although solar PV offers many advantages as a form of energy generation with respect to both renewable and non-renewable alternatives, there are questions emerging regarding the environmental integrity of the industry at the End of Life (EOL) stage [6]. When the useful life of a solar module is over, it becomes a form of hazardous waste due to the wide variety of integrated components present within the system [7]. As the PV industry is expected to continue to develop and grow rapidly in the coming years, EOL PV waste will naturally follow [6].

To date little progress has been made in relation to how the industry may process this impending waste stream [8]. Additionally, due to the relatively low supply of EOL PV waste generated heretofore, there has been little incentive to explore EOL management techniques beyond those on the waste management hierarchy such as disposal and recycling [9]. However, the dynamic nature of PV technologies makes EOL management within the industry extremely difficult, both technically and economically [10]. Therefore, in alignment with the WMH, it follows that an effective waste management technique would be to reduce the amount of waste being produced, as well as postponing the costly dismantling of components [11]. One such method of reduction of PV EOL waste could involve extending the lifetime of PV systems beyond the industry-standard of 25 years, which would involve alterations to current industry business and ownership models that don’t foresee a useful life for PV systems beyond this timeframe.

Although the current business models of PV solar are relatively well established, they emphasise immediate deployment of technology and single ownership models [12] rather than sustainable industry growth for the future. Typically, both commercial and residential investors view solar PV as an asset they will own and operate for a set 25 years until the performance warranty expires. However, this 25 year timeframe is based on PV applications from the 1970s and does not consider the rapid technological development of the industry. PV modules are now considerably more powerful than the current field-study versions, it is likely that their useful lifetimes may be far more than 25 years. Other researchers [13] stated: “a modern 300w monocrystalline panel performing at 80% in 25 years’ time is still more efficient that a brand new 100 w module that we currently base our projected waste estimations on”.

One method of extending product life is to diverge from a linear economic model towards a reuse or circular economic model [14] The PV industry has followed the linear economic model, raw material extraction, production, use and disposal, and typically has one owner over the course of its lifetime, for both commercial and residential applications. The reuse economy model considers the waste management hierarchy and aims to integrate more sustainable use of raw materials into the model, while allowing for some non-recyclable waste to be disposed of. A circular economy is an idealistic economic model that aims to eliminate waste while encouraging the continual use of all resources. However, irrespective of economic model chosen, the industry will inevitably generate significant quantities of waste, thus undermining the renewable status of this technology [15].

A business model with the potential to reduce PV industry waste is the cascading commercial ownership model. The premise of the model relies on the extension of the lifetime of the PV system beyond 25 years using tiered ownership to significantly reduce industry waste. The term ‘tiered ownership’ refers directly to the investor level, i.e. a tier one organisation makes the initial purchase. The asset is then sold prematurely, e.g. at expiry of the inverter warranties at 10 years, whereby investment risk is mitigated and payback time is shortened. A tier two organisation or investment then takes ownership of the asset at a heavily reduced price compared with a new installation, with a guaranteed efficiency performance over 90%. The tier two organisation agrees to a greater amount of risk, such as Balance of System (BoS) degradation, however, this enables a unique entry point into the market. This cascading tiered ownership continues until the risk is too high for a new investor or efficiency decreases below a useful percentage that donation to an institution is the only logical step, whereby each organisation involved would gain value in terms of Corporate Social Responsibility (CSR) reputational enhancement, or alternatively, decommissioning of the system. Under this synthesised model, the lifetime of a PV system could extend beyond 50 years. From a commercial market point of view, there is also potential to reduce and mitigate risk for individual organisations in a quantitative way, which is integral to any investment opportunity [16]. Another commercial model, with potential to greatly reduce the amount of PV industry waste is the elongated ownership model. The elongated residential ownership model assumes a continuation of the useful lifetime of a residential PV system, whereby the owner would retain the system beyond the widely accepted 25 year system lifetime [17-19]. The main objective of this study is to evaluate the feasibility of extending the lifetime of solar PV systems using a synthesised cascading tiered commercial ownership model and an elongated residential ownership model with a view towards reduction of industry EOL waste.

Annual AC output in kWh for all models

For years 1-25 and 26-50 of both BAU commercial and residential models and for years 1-25 of the CTCO and ERO models, only, Eq. 2 was used to estimate the respective PV installations’ annual AC output in kWh. Eq. 2 is the Standard Assessment Procedure equation (Eq. 1) as defined by SEAI, 2020 and BRE 2016, to which an annual performance degradation rate degree has been applied. The annual performance degradation rate is calculated as an annual percentage of the PV retailer’s 25 year performance degradation warranty.

For years 26-50 of both the CTCO and ERO models, Eq. 3 was used to estimate the respective PV installations’ annual AC output in kWh. Eq. 3 assumes that at year 26 the initial PV installation will have a total annual power output that is 80% that of the initial installation (0.8_b). It also assumes that this annual power output will be further degraded for years 26-50 by the annual degradation rate degree.

$Annual AC Output \left(kWh\right)=0.8 x kWp x S x Zpv\text{ (1)}$

where:

• 0.8 = the assumed performance ratio of the system [20] (BRE 2016)
• kWp = installed peak power
• S = annual solar radiation
• Zpv = over shading factor (a value of 1 was used in this study to indicate no shading as per an optimally installed system)

$Annual AC output (kW{h}_{Yr})=(\left(0.8 x kWp x Zpv\right) x (1-degr. x Yr))\text{ (2)}$

where:

• degr. = Yearly % of the PV retailers total performance degradation warranty over 25 years.
• Yr = Year (1-25) in which the PV installation’s annual AC output is being calculated

For years 26-50, 25 is subtracted from each year so that the equations 1-25 year range applies.

$Annual AC output (kW{h}_{Yr})=\left(0.8 x kWp x Zpv\right) x ({0.8}_b-degr. x \left(Yr -25\right))\text{ (3)}$

where:

• 0.8b = total annual power output at year 26 as a percentage of the total annual power output at year 0.
• degr. = % annual performance degradation rate for years 26-50.
• Yr = Year (26-50) in which the PV installation’s annual AC output is being calculated

Annual revenue calculated from annual AC output in kWh

Annual revenue generated by selling the annual electricity produced (kWh_Yr) to the grid at a specific feed in tariff

$Annual Revenue \left(€\right)=Annual AC output (kW{h}_{Yr})\times FIT (\frac{€}{kWh}) \text{ (4)}$

where:

FIT = Feed in tariff in € per kWh

Total system lifetime cost of BAU and CTCO models

$Total system lifetime cost \left(€\right)=Capital cost \left(2\right)+O \& M cost+Decommissioning cost (2)\text{ (5)}$

$Total system lifetime cost \left(€\right)=Capital cost+ O \& M cost-Sell on Price\text{ (6)}$

where:

Sell on price = to the price that is achieved when one tier organisation sells the PV installation to another.

$Total system lifetime cost \left(€\right)=Capital cost+ O \& M cost+Decommissioning cost\text{ (7)}$

where:

The cost of decommissioning was estimated using the cost of disconnection per panel (€25.50) and the cost of hardware disconnection per panel (€34) and multiplying the result by the number of panels in the system [21] (Solar Quotes, 2020).

BAU and CTCO models’ Net Present Values (NPV)

Net Present Value (NPV) is the difference between the present value of cash inflows and outflows over a defined periods (t) for capital projects. A positive NPV indicates a profit over the lifetime of the project, while a negative NPV signifies a loss.

$NPV= {\displaystyle \sum }_{t=0}^N\frac{Revenu{e}_t-Cos{t}_t}{{(1+d)}^t}\text{ (8)}$

where:

• N = number of years of economic analysis
• t = year variable
• Revenue = revenue generated by system in year t
• Cost = cost of system in year t
• d = discount rate (assumed to be not applicable in this study)

The objective of this study was to analyse the economic and technical feasibility of elongating solar PV system lifetimes from the current BAU commercial and residential industry standard practice to the proposed CTCO and ERO models, by applying novel approaches to ownership.

Assessing the applicability of the Cascading Tiered Commercial Ownership (CTCO) model as an alternative to current commercial models

The primary differences between the CTCO model and that of the BAU commercial model is that the useful lifetime of the installation is extended from 25 years to 50 and that the number of owners throughout the lifetime of the system is greater than one (Tiers). The ‘cascading’ and ‘tiered’ aspects of the proposed CTCO model propose new strategies by which ownership and as such risk can be transferred for the duration of the installation. However, before these ownership models can be considered, the technical and economic feasibility of elongating the installation’s lifetime needs to be assessed first.

CTCO and BAU commercial models-system description

The proposed CTCO and the BAU commercial ownership models both consider a 50 year period in which electricity generated at a theoretical site in a non-specific location, with no shading, receiving an average annual irradiance level of 1250 kWh/m2 [22], is being sold to the main grid via a set FIT price. Although, capital costs attributed to the installation, decommissioning and replacement of the inverter at 10 year intervals as per industry norm are identical, the BAU model’s 25 year lifespan means that two as opposed to one, decommissioning and installation events are required over the 50 year project lifespan.

Each CTCO installation which assumes a total nominal power output of 60kWp, comprises of 200 roof-mounted Hanwha Q Cells Q Peak G4.1 300 mono crystalline solar panels, each of which have a nominal power output of 300 Wp and an efficiency of 18%. The inverter used was the Fronius Symo 15.0-3-M, has an efficiency of 98.1%, a maximum AC nominal output (p_ac,r) of 15 kW and an expected lifespan of 10 years as per the industry norm. The rest of the BoS components are of a generic brand and battery storage usage was not considered for this study.

CTCO and BAU commercial models-technical analysis

A degradation rate (.degr) of 0.08% was applied to Eq.2 to calculate both the CTCO and BAU models respective total power outputs from year 1 to 25. This degradation rate was selected as it represents a standard PV retailer performance warranty that stands at 20% total performance degradation over 25 years [15,23,24]. Although, Q cells’ performance warranty specifies a 16.4% total performance degradation over 25 years, the less favourable industry standard PV retailer degradation rate warranty was selected, as in real world settings, warranties are between the retailer and the customer and not the supplier.

As the BAU commercial model requires that the PV installation be replaced twice over the 50 year period, total annual power output is calculated by applying Eq. 2 twice for years 1-25 and 26-50, respectively. However, as the CTCO model simply extend the lifespan of the initial installation from years 26-50, Eq. 3 in which a degradation rate (degr.) of 1% was applied. As the degradation rate beyond 25 years has not been studied extensively a degr. value of 1% was chosen as it was the average of the 0.5%, 1% and 1.5% degradation rates that could applied as proposed by [25-27] respectively.

As expected, the BAU model which applied Eq. 2, for years 1-25 and 26-50, produced a total power output (2,688,000 kWh) that significantly outperformed the CTCO model (2,349,000 kWh) which applied Eq. 2 for years 1 -25 and Eq. 3 for years 26-50.

CTCO and BAU commercial models-economic analysis

The total 50 year revenue that could be generated by selling the BAU and CTCO models respective annual power output to the grid was calculated from Eq. 4 using an assumed set FIT price that is 60% (0.1296 EUR/kWh) of the average EU Member State energy retailer price for main grid electricity (0.216 EUR/kWh) [28]. As annual power outputs for both models were identical for years 1-25, the higher revenue generated by the BAU model (€348,364.80) when compared to that of the CTCO model (€304,430.40), resulted from the difference in power outputs that occurred between years 26-50 only. Although, the total revenue that could be generated by the BAU model over a 50 year period was found to be superior to that of the CTCO model, lifetime costs attributed to each model needed to be assessed. Capital costs per PV installation (€64,253.6) were calculated from Eq. 5 using industry and literature based cost assumptions for PV modules, inverters installation and the PV systems Balance of System (BoS) components. Additionally, for each PV installation, there will be one decommissioning cost and continuous Operational and Maintenance (O&M) costs such as inverter replacement at a 10 year interval. The BAU scenarios when applied over a 50 year period will require two PV installations, two decommissioning’s and a continuous O&M cost (Eq. 6). In the BAU scenario, as the asset is typically owned by one owner/organisation throughout the entire system lifetime, total system lifetime costs for the 50 year period which stand at €193,571.2 (Table 1) are incurred by one owner/organisation.

Although, both commercial ownership models were considered stand-alone capital projects, i.e. a large initial investment by an organisation that earns an expected net economic benefit over time through savings and income that occur due to the project, which typically ‘pays off’ the initial capital investment, the primary difference between the CTCO and current commercial models is that the number of owners throughout the lifetime of the system is greater than one. The ‘cascading’ and ‘tiered’ aspects of the model refer to the method of ownership transfer, whereby the PV system (the asset) is initially funded by a tier One organisation seeking a large Return on Investment (ROI) from a high capital cost, with minimal risk. The asset is subsequently sold to a Tier Two organisation when risk increases, i.e. when BoS warranties expire at 10 years, at a reduced cost. This acts as a market entry point that would otherwise be unavailable for the Tier Two organisation due to the high initial capital cost of solar PV projects. The Tier Two organisation then invests in O & M, such as BoS replacement costs, to mitigate this risk. This tiered ownership continues in a cascading fashion until the technical performance of the system no longer justifies the economic risk. At some point along the cascading ownership transfer line, the higher tiered organisations may decide that donation of the PV system to an institution or a charity could have a greater benefit to the organisation in terms of CSR enhancement, rather than attempting to sell the depreciating asset. The timing of tiered organisation ownership transfer, installation-, decommissioning-, O&M- and BoS replacement-costs for both the CTCO and BAU models are detailed in the supplemental table S1.

As the CTCO model proposes to transfer ownership of the asset at 10 year intervals (Tiers), Total system lifetime costs are calculated as a summation of each tiered ownership level’s total costs across the 50 year lifetime of the project. As only the final tiered organisation that is in possession of the PV installation at year 50, is responsible for decommissioning costs, Eq. 6, which calculates a tiered organisation’s total costs as the difference between the capital costs incurred (exclusive of decommissioning) and the price at which the PV installation has been sold onto a subsequent tiered organisation, is applied to all other tiered organisations. Eq. 7 which is applied to final tier organisation in possession of the PV installation at year 50, is Eq. 6 to which decommissioning costs are added but that no sell-on value is subtracted. For all tiered organisation to which actual costs were estimated using Eq. 6, the sell-on value was calculated by assuming that the value of the PV installation depreciated by 5% per year over a 10 year period (50% of initial capital cost), and that the asset continued to depreciate at the same rate with respect to capital price. The sell-on value of 50% was used as a pessimistic depreciation rate considering the PV modules still perform at 90% by guarantee. The two CTCO options presented pertain to; 1.) a Tier 4 organisation maintaining ownership of the PV installation from year 30 until the end of the project in year 50; and 2.) a scenario whereby at year 40 a Tier 4 organisation sells the asset to a Tier 5 organisation/institution both produced the same total life time cost value of €117,417.60. The requirement that the costs of two installations and two decommissioning events be incurred by the BAU model for the 50 year project duration, meant that its total lifetime costs of €194,214.4 were significantly higher than that of both CTCO model options.

CTCO and BAU Commercial models - economic viability as a function of NPV

The lifetime cost and lifetime revenue for both the BAU and both CTCO model options are presented in (Figure 1). Although, the BAU model revenue was found to be superior to the cascading model revenue, it had significantly higher lifetime costs. To this end, Net Present Value (NPV) which is the most commonly used method to analyse commercial capital projects [29] (NREL, 2011) was applied using Eq. 8. The NPV for both the BAU which had a single tier organisation ownership and option 1 and 2 of the CTCO tiered ownerships that were calculated from Eq. 8 to which it was assumed that no discount applied are presented in (Table 2). The total NPV for the BAU and both CTCO models was €154,793.60 and €187,012.80, respectively. However, it can be seen that the NPV was significantly lower for some year grouping/tiers for both models. Years 20-30 and 40-50 of the BAU model have a low NPV of €15,827.46 and €10,710.61, respectively, while years 40-50 for option 2 of the CTCO model exhibit a low NPV of €11,008.85. Additionally, as the NPV presented for years 31-50 for option 1 of the CTCO model is a total figure for this 20 year period (€52,009), it can either be interpreted as €0 NPV for years 31-40 or €26,004.85 for both 31-40 and 41-50, periods. For all models, the low NPV is due largely to the significant cost of decommissioning that was not the burden of previous year groupings/tiers within the model. As the BAU model incurs two decommissioning events during the project lifetime, there are two periods in which a low NPV value is observed, however, in all instances the NPVs are still profitable.

Figure 1: Lifetime cost evolution (€) and lifetime revenue evolution (€) for both commercial ownership models.

The cascading commercial solar PV ownership model was shown to be both technically and economically feasible. Although producing less energy than the BAU commercial model over the lifetime, the cascading commercial ownership model produced a significant amount of energy beyond 25 years of operation. The FIT revenue generated by this system output, combined with the lower total lifetime system cost, led to a favourable NPV of €181,061.21 compared to the BAU commercial ownership model (€161,615.36). Additionally, the payback time for each tier within the cascading ownership model was shorter than that of the BAU model, indicating favourable investment conditions of the proposed model. The elongated residential solar PV ownership model was both technically and economically feasible. The elongated ownership model generated less electricity throughout the lifetime of the system than the BAU residential model. However, the same proportion of the energy demand of the house was met as the BAU model until year 45, and surplus solar energy was still sold to the main grid until the same time, thus generating FIT revenue for the owner. The elongation of the system lifetime led to a favourable NPV of €8,335.40 in comparison with that of the BAU residential model (€5,786.80). The scenario analyses conducted in this experiment highlighted some of the conditions in which the feasibility of the hypothesised ownership models could remain favourable. The rate of FIT received for solar energy exportation to the main grid would have to be more than €0.25 kWh for the entire lifetime of the project to favour the BAU models. The rate of module degradation would have to be greater than 2% year for the technical output of the systems to reduce to the extent that the BAU models were more economically feasible. Any decrease in FIT rate beyond 20% applied to the BAU model after 25 years of operation would lead to greater total revenue generation for the hypothesised models. Variation in irradiation levels was also shown to have a significant impact on the feasibility of such systems, with values greater than 1750 kWh/m2 favouring the BAU models slightly in terms of NPV. A key point to also be addressed is in regard to the module manufacturing guarantees which define the useful life of a PV system especially if novel ownership models – such as those proposed here – are to be integrated into future PV projects.

1. Kannan N, Vakeesan D. Solar energy for future world: A review. Renewable and Sustainable Energy Reviews. 2016;62:1092-1105. doi: 10.1016/j.rser.2016.05.022.
2. Schou P. Polluting non-renewable resources and growth. Environmental and Resource Economics. 2000;16(2):211-227. doi: 10.1023/A:1008359225189.
3. Hoffmann W. PV solar electricity industry: Market growth and perspective. Solar Energy Materials and Solar Cells. 2006;90(18):3285-3311. doi: 10.1016/j.solmat.2005.09.022.
4. Crago CL, Koegler E. Drivers of growth in commercial-scale solar PV capacity. Energy Policy. 2018;120:481-491. doi: 10.1016/j.enpol.2018.05.047.
5. IEA. 2020. https://www.iea.org/
6. Shin J, Park j, Park N. A method to recycle silicon wafer from end-of-life photovoltaic module and solar panels by using recycled silicon wafers. Solar Energy Materials and Solar Cells. 2017;162:1-6. doi: 10.1016/j.solmat.2016.12.038.
7. Huang P, Negro SO, Hekkert MP, Bi K. How China became a leader in solar PV: An innovation system analysis. Renewable and Sustainable Energy Reviews. 2016;64:777-789. doi: 10.1016/j.rser.2016.06.061.
8. IRENA. End-of-Life Management: Solar Photovoltaic Panels. 2017. https://tinyurl.com/v7x9e974
9. Tao J, Yu S. Review on feasible recycling pathways and technologies of solar photovoltaic modules. Solar Energy Materials and Solar Cell. 2015;141:108-124. doi: 10.1016/j.solmat.2015.05.005.
10. Chowdhury MS, Rahman KS, Chowdhury T, Nuthammachot N, Techato K, Akhtaruzzaman M, Tiong SK, Sopian K, Amin N. An overview of solar photovoltaic panels end-of-life material recycling. Energy Strategy Reviews. 2020;27:100431. doi: 10.1016/j.esr.2019.100431.
11. Dias PR, Benevit MG, Veit HM.  Photovoltaic solar panels of crystalline silicon: Characterization and separation. Waste Management & Research. 2016;34(3):235-245. doi: 10.1177/0734242X15622812.
12. Abreu J, Wingartz N, Hard Ny. New trends in solar: A comparative study assessing the attitudes towards the adoption of rooftop PV. Energy Policy. 2019;128(C):347-363. doi: 10.1016/j.enpol.2018.12.038.
13. Mokhtar M, Wagdy R, Abouelatta M, IA. Design of PV modules including a layer between solar cells and glass cover to increase pv module lifetime. International Journal of Computer Applications. 2018;182(19):18-23. doi: 10.5120/ijca2018917950.
14. Baker-Brown D. The re-use atlas: A designer's guide towards the circular economy. London: RIBA Publishing; 2017. https://tinyurl.com/2p96ecuj
15. Salim K, Stewart RA, Sahin O, Dudley M. Drivers barriers and enablers to end-of-life management of solar photovoltaic and battery energy storage systems: A systematic literature review. Journal of Cleaner Production. 2019;211:537-554. doi: 10.1016/j.jclepro.2018.11.229.
16. Tomosk S, Haysom JE, Wright D. Quantifying economic risk in photovoltaic power projects. Renewable Energy. 2017;109:422-433. doi: 10.1016/j.renene.2017.03.031.
17. Karneyeva Y, Wüstenhagen R. Solar feed-in tariffs in a post-grid parity world: The role of risk, investor diversity and business models. Energy Policy. 2017;106:445-456. doi: 10.1016/j.enpol.2017.04.005.
18. Osterwalder A, Pigneur Y, Clark T, Smith A. Business model generation: A handbook for visionaries, game changers, and challengers. 1st ed. Hoboken, New Jersey: Wiley; 2010. p.288. https://tinyurl.com/mrfeutnv
19. Sica D, Malandrino O, Supino S, Testa M, Lucchetti MC. Management of end-of-life photovoltaic panels as a step towards a circular economy. Renewable and Sustainable Energy Reviews. 2018;82:2934-2945. doi: 10.1016/j.rser.2017.10.039.
20. Building Research Establishment Ltd (BRE), National calculation methodologies for energy rating of dwellings (SAP/RdSAP), on behalf of the Department of Energy and Climate Change (DECC) 2016.
21. Solar Quotes. 2020. https://tinyurl.com/mr4xaxmx
22. Solar GIS. 2020. https://tinyurl.com/2evvajtf
23. Jordan DC, Silverman TJ, Sekulic B, Kurtz SR. National Renewable Energy Lab, GCO. PV degradation curves: Non-linearities and failure modes. Progress in photovoltaics. 2017;25(7):583-591. doi: 10.1002/pip.2835.
24. Quansah DA, Adaramola MS. Ageing and degradation in solar photovoltaic modules installed in northern Ghana. Solar energy. 2018;173:834-847. doi: 10.1016/j.solener.2018.08.021.
25. Quintana MA,  King DL, Hosking FM, Kratochvil JA, Johnson RW, Hansen BR, Dhere NG,  Pandit MB. Diagnostic analysis of silicon photovoltaic modules after 20-year field exposure. IEEE. 2000;1420-1423. doi: 10.1109/PVSC.2000.916159.
26. Sánchez-Friera P, Piliougine M, Peláez J, Carretero J, Sidrach de Cardona M. Analysis of degradation mechanisms of crystalline silicon PV modules after 12 years of operation in Southern Europe. Progress in photovoltaics. 2011;19(6):658-666. doi: 10.1002/pip.1083.
27. Vignola F, Peterson J, Kessler R, Mavromatakis F, Dooraghi M, Sengupta M. Use of pyranometers to estimate PV module degradation rates in the field. IEEE. 2016;116-1021. doi: 10.1109/PVSC.2016.7749764.
28. Eurostat. 2020. https://tinyurl.com/2w82a59a
29. Wade A, Stolz P, Frischknecht R, Heath G, Sinha P, National Renewable Energy Lab, GCO. The Product Environmental Footprint (PEF) of photovoltaic modules-lessons learned from the environmental footprint pilot phase on the way to a single market for green products in the European Union. Progress in Photovoltaics: Research and Applications. 2018;26(8):553-564. doi: 10.1002/pip.2956.