EDP Sciences logo
Submit your paper
Search
Menu
All issues Volume 17 (2026) EPJ Photovolt., 17 (2026) 6 Full HTML
Issue
EPJ Photovolt.
Volume 17, 2026
Special Issue on ‘EU PVSEC 2025: State of the Art and Developments in Photovoltaics', edited by Robert Kenny and Carlos del Cañizo
Article Number 6
Number of page(s) 13
DOI https://doi.org/10.1051/epjpv/2025030
Published online 03 February 2026

© C. Grand et al., Published by EDP Sciences, 2026

Licence Creative CommonsThis is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

1 Introduction

Photovoltaic (PV) modules and systems are indubitable contributors to energy generation from renewable sources. According to IEA’s PVPS report T1-43:2024 “Trends in Photovoltaic Applications 2024”, at the start of 2024 over 1.6 TW of PV systems were operational, generating more than 2,135 TWh of electricity. This output accounted for 8.3% of the global electricity demand and PV systems were estimated to reduce greenhouse gas emissions by approximately 920 million tons of CO2 [1]. When considering sustainability of photovoltaic modules through the methodology of a life cycle assessment (LCA), innovations focused on increased energy efficiency and reliability to reduce the levelized cost of electricity from solar panels also have benefits for the LCA of this technology, most directly impacting the in-use phase [2]. The design of such efficient and reliable modules requires a multi-material system with a layered structure and complex interfaces as detailed in Figure 1, where polymers are the base material for encapsulant layers and may be used in the backsheet layer.

The encapsulant layer is critical to limiting losses and failure mechanisms by protecting the cell and metallization from physical and environmental stressors and maintaining electrical insulation [3]. Furthermore, the development of new, cutting-edge modules also requires innovative approaches, materials, and formulations in the encapsulant layers to enable implementation of more efficient cell technologies along with increasing module reliability [4,5], hence also impacting positively the in-use phase of the module life cycle. Furthermore, when considering the other stages beyond the in-use phase in the full LCA, and the End-of-Life (EoL) phase in particular, innovative technologies have been developed to remove the encapsulant layers for module separation and material recovery [6–10]. With the Waste of Electrical and Electronic Equipment (WEEE) directive stating that all producers supplying photovoltaic solar panels to the European market must finance the costs of collecting and recycling EoL photovoltaic solar panels in Europe [11,12], further focus is directed towards repair, collection, and recovery of modules, and hence on the contributions from encapsulants on circularity.

As mentioned, encapsulant layers in modules rely on polymers: ethylene-vinyl acetate (EVA) and polyolefin elastomers (POE, considered to be copolymers of ethylene and α-olefin) are two commonly used polymer types in this market, with encapsulant films containing POE continuing to grow, either as a monolayer or as a multilayer in combination with EVA (referred to as EPE films) [13]. The generic structures of EVA and POE polymers are shown in Figure 2. It is important to note that while EVA or POE polymers are the majority components, encapsulant films are formulated systems and include additives, one of which is typically a peroxide to yield a crosslinked film in the module with the thermomechanical performance required to meet module qualification standards [14,15]. When reviewing the use phase of modules, thoughtful encapsulant material selection depending on the cell type and module construction may minimize failures, hence maximizing reliability and lifetimes [4,16]. When thinking about the EoL of modules with the most common encapsulants currently, the crosslinked EVA or crosslinked POE materials recovered (either from manufacturing rework or from modules after use in the field) are expected to have widely different properties compared to the original, non-crosslinked EVA or POE polymer, hence creating a challenge for closed-loop polymer recycling for modules. In particular, the crosslinked, thermoset materials are expected to have distinct rheological behavior compared to the original, thermoplastic polymer, thus limiting their use as-recovered [17]. This change in rheology points to the need for further recycling assessments of the recovered crosslinked polymers from PV modules.

In addition to encapsulants, backsheet layers also rely on polymers. The materials used in backsheets tend to vary from those found in the encapsulant layer, due to the different performance requirements, and the backsheet itself is often a multilayer, multi-material system [18]. As with encapsulant materials, technology of the backsheets is evolving and materials used may include polyethylene terephthalate (PET), fluoropolymers, such as polyvinylidene fluoride (PVDF), and polyolefins, such as polyethylene (PE) or polypropylene (PP) [19,20].

Given that polymers are the base for encapsulant and backsheet layers, end-of-life recycling technology of polymers from other applications can be leveraged to benefit circularity of PV modules. Over 450 million metric tons of plastics are produced annually on a global scale, but only around 12% are being collected for recycling [21–23]. Based on data from 2019, 40% of the global plastic comes from packaging applications; the remaining 60% are from non-packaging applications [24,25]. For example, recycling of rigid bottles from packaging applications is well established, where polyethylene terephthalate (PET) and high-density polyethylene (HDPE) represent the majority of recycled plastics. By contrast, and as noted for photovoltaic modules, polymers from non-packaging applications are more difficult to collect and recycle, as the overall system including polymers is typically meant to enable long lifetimes and high reliability, where different types of polymers, combinations of polymers with other materials such as glass and metals (among others), or the use of different additives can complicate these formulations. Given the multicomponent design of the modules, ideally, a kind of dismantling or disassembly of solar modules into their individual components to obtain interesting secondary raw materials, such as individual polymer fractions could be considered [26,27]. Several companies offer recycling technologies for photovoltaic solar panels [28], with upstream processes to remove the frames and connecting cables with the corresponding boxes as the first step implemented as state of the art by several companies [29,30]. Subsequently, the separation of the module layers to recover desired fractions can be completed using water jet separation. In this process, high pressure is used to remove the material layer-by-layer down to the PV glass, and the individual polymer flakes can then be sorted [31,32]. Another example of separation of the polymer-rich layers by physical and chemical processes, such as swelling, is provided in the literature [33]. The resulting output streams include aluminum, glass, polymers, silicon powder, silver and other metals, and while many of these output streams can be reused or recycled (e.g., glass, aluminum, copper, and silver), the polymer stream is often directed to other applications, such as the cement industry or used in road construction [34–36]. Although these applications could also be considered for reuse of the polymers coming from the photovoltaic solar panels, this still follows a linear approach and is not circular.

Complementary recycling routes exist for polymers: mechanical and advanced (a.k.a. chemical) recycling. In mechanical recycling, the polymer is maintained: the collected polymer-rich fraction is shredded, sorted, washed, and extruded and pelletized. However, not all polymers can be mechanically recycled. For that reason, chemical recycling, which breaks down the input material to smaller units, has been gaining attention. It should be noted that the collection, shredding, sorting, and washing steps utilized with mechanical recycling are also included with chemical recycling. Three technologies are currently recognized for chemical recycling: depolymerization, pyrolysis, and gasification. Pyrolysis is a thermal process that converts the solid input material into different products; these output products include liquid hydrocarbons (pyrolysis oil), gas, and solid char/residue. The principal output stream is pyrolysis oil, which after upgrading to remove undesired impurities/components, can be used as a feedstock for steam crackers to produce olefinic monomers, which can ‘Close the Loop’ through the production of recycled polymers. Pyrolysis is a targeted recycling route for polyolefinic materials, such as EVA and POE, to favor closed-loop circularity for materials found in encapsulant layers by producing polymers with electrical and rheological properties required by photovoltaic applications. This approach favors closed-loop recycling of polymers from encapsulant layers to achieve circularity of photovoltaic modules, with limited impact on module reliability during use. As this article focuses on pyrolysis, depolymerization and gasification will not be discussed. However, readers are referred to some of the many review articles detailing the different chemical recycling technologies [37–39].

Currently, energy recovery, via pyrolysis for example [32], appears to be the main EoL treatment for encapsulants from PV modules. To complement this path of energy recovery, which can contribute to emissions [40], this work focuses on characterization of recovered encapsulants from PV modules as an initial feasibility analysis of closed-loop recycling of encapsulant materials from photovoltaic modules via mechanical separation for recovery of polymer-rich layers, followed by chemical recycling to produce feedstocks for EVA or POE production. As polymer backsheet layers could end up in the same polymer-rich stream from end-of-life modules, the assessment of recovered backsheet materials complements that of the recovered encapsulant materials, to outline some of the challenges that may arise from contaminants or combined polymer streams. One example is the potential presence of fluoropolymers in polymer backsheets, which limits the compatibility of mixed polymer streams with recycling processes [41], and with pyrolysis processes in particular due to HF gas formation [42].

This work complements prior research assessing the polymer stream from photovoltaic solar panels, where thermogravimetric analysis and pyrolysis coupled to gas chromatography with mass spectrometric detection were used to evaluate the polymer stream based on an EVA-rich encapsulant layer and a backsheet layer from photovoltaic solar panel recycling [43]. Various olefinic compounds and oxygenated molecules were identified; however, no quantification was provided, nor was an evaluation on potential inorganic contamination performed. Such information is critical to understand the type and levels of undesired impurities that could be present in a pyrolysis oil obtained from the polymer stream from photovoltaic solar panels. The Deutsche Gesellschaft für Kreislaufwirtschaft und Rohstoffe mbH (DKR) is the German system for classifying polymer waste and is commonly used in Europe, and while there is no open-source specification for pyrolysis feedstocks, several articles have been published on the lab- and pilot-scale pyrolysis of DKR-350 [44,45]. Therefore, the open DKR-350 specification for mixed plastics was used as a guideline in this report [46]. Presented herein is a first pass assessment and comprehensive evaluation of the polymer stream from photovoltaic solar panels for potential as an input feedstock for pyrolysis.

thumbnail Fig. 1

Schematic description of the layers and materials making up a generic photovoltaic module.
thumbnail Fig. 2

Structure of (a) ethylene-vinyl acetate (EVA) and (b) polyolefin elastomers (POE) polymers.

2 Material and methods

2.1 Materials

Modules from 3 manufacturers were recovered either after manufacturing (production scrap) or after use in the field and disassembled as described in the following section. The modules were either based on crystalline silicon (cSi) or thin-film cell technology. The separation of the panels yielded 5 polymer-rich samples as listed in Table 1: two from the backsheet layer labeled Backsheet 1 and Backsheet 2, and three from the encapsulant layer labeled Encapsulant 1, Encapsulant 2, and Encapsulant 3. Backsheet 1, Encapsulant 2, and Encapsulant 3 come from module production at the time of dismantling, meaning that they were not subjected to stressors as would an EoL panel, such as electrical bias, ultraviolet exposure, humidity, or temperature. Encapsulant 2 was laminated to the crystalline silicon cell, whereas Encapsulant 3 was recovered prior to lamination to the cell. It is interesting to note the variation in physical aspect of the recovered fractions compared to the input materials, as illustrated by the encapsulant layer. The input encapsulant film is designed to be colorless as seen in Figure 3b and transparent when integrated in the module, while some recovered samples, such as Encapsulant 2 and Encapsulant 3, show varying degrees of gray-to-black color. Encapsulant 3, which has not been in contact with other components of the module, exhibits a light grey color. It is hypothesized that this material could be based on technology similar to smart wire contacting technology (SWCT) and include both polymer and some metal content by design [47]. Furthermore, Encapsulant 2 has been laminated to the cSi cell and its darker color compared to Encapsulant 3 could be the result of contamination from materials present in the module at the interface with the encapsulant layer, for example from the silicon cell, and metallization and interconnect, as well as the strong mechanical stresses encountered during the separation process.

thumbnail Fig. 3

(a) Pictures of the recovered polymer-rich samples, (b) Picture of a typical embossed encapsulant film prior to lamination in a module.
Table 1

Description of the recovered samples for analysis.

Table 2

Comparison of theoretical and estimated C:H:O ratio.

thumbnail Fig. 4

Elemental analysis (CHN-O) results for the two backsheet samples and three encapsulant samples. Note: oxygen values were estimated by subtracting the carbon, hydrogen, and nitrogen values from 100%.

2.2 Module separation and material recovery

Two key technologies were used to obtain polymer samples from end-of-life modules for the initial tests. For the backsheet samples, taking the “Backsheet 2” sample from Module 2 as an example, a complete PV module without frames and J-boxes was treated using high-pressure waterjet technology (according to the apparatus and method in patent DE102021109591B4) [48]. The backsheet material could be easily extracted from the blasting material and made available for the tests. The backsheet of this old module was very easy to remove and fell off in larger flakes than usual. We suspected that contact with the substrate was no longer very strong. The module was a monofacial glass/backsheet crystalline silicon module.

This was contrasted by a so-called thin-film module, Module 3, which had also been in the field for a long time. These modules are usually double-glazed modules, which require a different technology to open the sandwich. Typically, the photoactive layer is located directly on one of the two glass plates. This layer can be manipulated using flash discharges, allowing the entire glass plate to be removed using vacuum suction cups after complete exposure of the thin-film module. Subsequently, the semiconductor and conductor residues remaining on the polymer fraction were removed using a hydrometallurgical process (according to technology published in EP000002961855B1) [49], and the clean polymer-rich material was removed manually and cut for further testing.

Encapsulant 2 and Encapsulant 3 were recovered during module production and underwent mechanical treatment for further separation.

2.3 Cryogenic milling

All samples were cryo-milled prior to analytical evaluation. A Retsch CryoMill with a 50 mL stainless steel grinding jar and autofill liquid nitrogen vessel was utilized (Verder Scientific, Belgium). Liquid nitrogen was used as the cryogen (Air Liquide, Belgium). An amount of material was added to the 50 mL stainless steel griding jar, making sure to not overfill the grinding jar, and the material was cryo-milled for 5 min or until a fine powder was obtained.

2.4 Fourier transform infra-red (FTIR) spectroscopy

Fourier Transform Infra-red (FTIR) spectroscopy was performed on the cryo-milled material. A Nicolet iS50 FTIR spectrometer was used in attenuated total reflection (ATR) mode to collect spectra on the samples (Thermo Fisher, Netherlands). A total of 16 scans at 4 cm−1 were co-added to obtain the IR spectra. OMNIC software (Version 9.13.1256) provided with the instrument was used for data collection and processing.

2.5 Elemental analysis (CHN-O)

The elemental composition of the samples was determined using a Vario EL Cube CHN-O elemental analyzer from Elementar (Beun De Ronde, Netherlands). The EL cube operating software provided with the instrument was used for instrument control and data processing. Approximately 5 mg of the cryo-milled material was weighed into the sample cup and analyzed by combustion followed by separation using gas-selective columns to trap the gases until released by heating prior to detection by thermal conductivity. Triplicate sample preparations were prepared and analyzed. It should be noted that the oxygen content was estimated by subtracting the carbon, hydrogen, and nitrogen values from 100%.

2.6 X-ray fluorescence (XRF)

The inorganic composition of the samples was determined using a Zetium 4 kW X-Ray Fluorescence (XRF) analyzer (Malvern Panalytical, Netherlands). The samples were analyzed in a P1 cup with a 4 μm prolene film. Omnian software provided with the instrument was used for instrument control and data processing. A relative error of around 20% is given for the XRF results.

2.7 Thermogravimetric analysis (TGA)

The weight loss distribution for the samples was determined using a Mettler Toledo TGA/DSC 3+. The purge gas was either nitrogen or air at a flow rate of 30 mL/min. For the first heating run nitrogen gas was used, and the oven program was as follows: 30 °C (2 min) – 20 °C/min – 900 °C (10 min). The oven was allowed to cool down to and equilibrate at 200 °C, the purge gas was switched to air, and then the second heating run was performed using the same ramp of 20 °C/min to 900 °C with a 10 min isothermal hold. Instrument control and data processing were performed using the STARe software provided with the instrument.

3 Results and discussion

3.1 Material characterization

All samples were cryo-milled prior to analysis by a variety of analytical techniques; it should be noted that only cryo-milled material was analyzed to obtain the overall composition of the samples, which is more relevant when considering a material as a potential feedstock for pyrolysis recycling. FTIR was used to identify the polymer type present, elemental analysis was used to determine the amount of carbon, hydrogen and nitrogen (oxygen was estimated), XRF was utilized for semi-quantitative determination of the inorganic elements present, and TGA was used to estimate the amount of polymer content present. While the sample set is limited and does not allow for a systematic comparison, these samples provide a first view into the range of composition that could be expected both from the input materials based on module design and from output material impacted by the separation method selected (here, mechanical separation).

From the FTIR analysis, the polymer type was determined, and Table 2 gives an overview of the identified polymers. Identification was performed using reference spectra contained in the spectral library and comparing to the spectra obtained from the samples. Clear differences in polymer type were seen between the backsheet and encapsulant samples; additionally, differences were also observed between the encapsulant samples. Both Backsheet 1 and Backsheet 2 contained PET. Encapsulant 1 included polyvinyl butyral (PVB) and EVA, while Encapsulant samples 2 and 3 consisted of ethylene-rich polymers, akin to what can be expected from POE-based encapsulants. The differences between backsheet and encapsulant materials were expected based on the distinct property requirements of the two layers and the information on materials used by manufacturers of these different films.

The elemental analysis (CHN-O), inorganic analysis (XRF), and weight loss distributions (TGA) are complementary to one another. From the elemental analysis, it was clear to see that the two backsheet samples contained the lowest amount of carbon (approximately 46 wt.% and 59 wt.% for Backsheet 1 and 2, respectively) and highest amount of oxygen, estimated at around 50 wt.% and 36 wt.%, respectively (Fig. 4). As stated above, the two backsheet samples are composed of PET, which has a formula of (C10H8O4)n. Encapsulant 1 has the next highest level of carbon, at about 63 wt.%. This sample comprises PVB and EVA, which have the following formulas: (C8H14O2)n and (C2H4)n(C4H6O2)m, respectively. Encapsulants 2 and 3 contained the highest carbon content at around 79 wt.% and 83 wt.%, respectively. These samples both contained ethylene-rich polymers thought to be polyolefin elastomers as common encapsulant materials discussed in the introduction, with a typical example being ethylene/1-octene copolymer, which has a formula of (C2H4)n(C8H16)m. Nitrogen was detected at less than 0.5 wt.% for the two backsheet samples and was below the detection limit of 0.1 wt.% for the three encapsulant samples.

While it is expected that the carbon content corresponds solely to the polymer(s) present in the samples, the oxygen content can originate from not only the polymer (in the case of oxygen-containing polymers such as PET, PVB, or EVA), but also from the inorganic components present. From the inorganic data presented in Fig. 5, it is clear to see that Backsheet 1 contains the highest total level of inorganic elements at around 22 wt.%. The largest contributor to Backsheet 1 is aluminum, which is present at roughly 18 wt.%. The aluminum present in this backsheet sample could be by design, as aluminum layers in the backsheet provide a moisture barrier, with this type of backsheet available from multiple suppliers [50]. Backsheet 2 contains around 1.1 wt.% total inorganics with titanium making up the majority at 1 wt.%. The presence of titanium in both backsheet samples is thought to stem from titanium dioxide incorporated by design into backsheets for performance requirements [19]. Both backsheet samples also show some levels of silicon. The origin of the silicon content in Backsheet 1, and to a lower extent in Backsheet 2, is unclear. One potential source of silicon in backsheets is silica used as a reinforcing filler. Another possibility is that a silicon oxide coating is used between the layers; however, the level of silicon detected is too high for that scenario to be likely. Encapsulant 1 has negligible inorganic content (<0.1 wt.%), while Encapsulants 2 and 3 have approximately 7 wt.% and 2.7 wt.% total inorganic content. Encapsulant 1 was recovered from a module based on thin-film cell technology after in-field use, whereas Encapsulants 2 and 3 were recovered from production. The different parameters between Encapsulant 1 and Encapsulant 2 and 3 (module design, cell type, encapsulant type, stress experienced by the module prior to separation, and separation method) do not allow for a clear explanation of the variation in total inorganic content between Encapsulant 1 and Encapsulant 2 and 3. For Encapsulant 2, silicon is the major contributor at about 5.5 wt.%, while bismuth at about 1.5 wt.% is the major contributor for Encapsulant 3. As noted previously, Encapsulant 3 had a light grey color, which is explained by the presence of bismuth and tin. In the case of Encapsulant 2, which displayed a darker color compared to Encapsulant 3, the inorganic content is thought to mainly come from contamination from the neighboring interfaces with the silicon cell, as well as metal-rich electrical connections (for example, silver likely coming from gridlines), in addition to bismuth and tin already present in the film before lamination. In particular, when comparing Encapsulant 2, which has been laminated to the cSi cell, and Encapsulant 3, which has not been in contact with the cSi cell, around 5 wt% of silicon is measured in Encapsulant 2, indicating that the combination of lamination with the mechanical separation process used in this study leads to a certain level of metallic components in the polymer fraction. This is hypothesized to be due to ion migration into the encapsulant layer or more likely from cohesive failure within the cSi cell, within the silicon or solder material, during the mechanical separation step.

As the C:H:O ratio of polymers is well known, the theoretical and estimated C:H:O ratios are included in Table 2. The theoretical ratio is based on the polymers identified by FTIR and the formulas given above. Estimating the C:H:O ratios was challenging, as C-H-N-O can also be incorporated into the inorganic compounds present. To estimate the C:H:O ratios, the total inorganic content obtained from XRF was subtracted from the total oxygen values estimated from the elemental analysis. As can be seen in Table 2, the theoretical and estimated C:H:O ratios only line up for the three encapsulant samples. The deviations between the theoretical and estimated C:H:O ratios are likely due to a multitude of factors. First, it is not accurate to assume that all inorganic content will contain oxygen. Metal (M0) or inorganic compounds containing C-H-N-O can be present. The estimated ratios assume that the inorganic content determined by XRF has an equivalent oxygen content. XRF only provides data on the inorganic element and not the molecular formula of the compound present in the formulation. Therefore, significant errors can be introduced into the estimated values; for example, oxygen is known to be present in many inorganic compounds (e.g., SiO2, TiO2), but the XRF data cannot define if Si is present as a metal (Si0), as an inorganic filler (SiO2), or as a polymer (PDMS: CH3[Si(CH3)2O]nSi(CH3)3). Second, oxidation of the polymers during outdoor exposure may also increase the oxygen content of the polymer [51]. The two backsheet samples, which showed deviations between the theoretical and estimated C:H:O ratios, contained the highest levels of oxygen. Both samples were shown to contain PET, which is an oxygen-containing polymer, and these samples also contain detectable levels of inorganic elements, which can also contribute to the total oxygen content (e.g., oxides), incorporating additional error into the estimated C:H:O ratios. Furthermore, oxidation of PET present in polymer backsheets may further contribute to the difference between theoretical and estimated C:H:O ratios [52,53]. For the three encapsulant samples, the estimated and theoretical C:H:O values aligned. Encapsulant 1 showed the highest oxygen level and was shown to contain EVA and PVB, both oxygen-containing polymers. Encapsulant 1 also showed negligible inorganic content, indicating that the total oxygen content observed originated from the polymers. Encapsulants 2 and 3 showed the lowest total oxygen content. These samples consisted of ethylene-rich polymer that does not contain oxygen, indicating that any oxygen present would be non-polymeric or the result of polymer oxidation.

TGA was used to estimate the weight loss information for the samples. Experiments were performed such that in addition to the polymer content, char and residue amounts could also be estimated. By heating under air, the amount of carbonaceous char present in the samples was determined. This differs from the residue, which should consist of the inorganic components. The weight loss information was used to estimate the amount of polymer content, which showed significant weight loss in the 300–550 °C temperature range. More specifically, the two backsheet samples exhibited weight loss in the 400-500 °C range, while Encapsulant 1 showed weight loss between 300-450 °C and Encapsulants 2 and 3 between 450-500 °C. The char content was determined under air, where weight loss was observed in the 500 – 650 °C range. Finally, the residue was the amount remaining after the second heating took place. Overall, these results are complementary to the elemental and inorganic analyses. As can be seen in Figures 6 and 7, Backsheet 1 contains the lowest polymer content (∼ 61 wt.%) and highest residue (∼ 37 wt.%). This sample also showed the lowest carbon content and highest inorganic content (described above). Backsheet 2 yielded about 83 wt.% polymer content and had significant char (∼12.5 wt.%) and residue (∼5 wt.%) levels. The level of char detected in Backsheet 2 seems high; however, this is likely attributed to the fact that some of the inorganics present in Backsheet 2 are in the oxide state (as determined from comparison of the theoretical and estimated C:H:O ratios) and the fact that the molecular formulas of the inorganic compounds present are unknown. These combined factors could result in the higher-than-expected amount of char in Backsheet 2. Encapsulant 1 showed an extremely high polymer content of ∼98 wt.% and relatively low char and residue values (<1 wt.% and ∼ 1 wt.%, respectively). Additionally, from the TGA curve, it is clear to see the different decomposition temperature of the PVB and EVA polymers. Encapsulant 2, which had an inorganic content around 7 wt.%, contained about 91 wt.% polymer and ∼ 9 wt.% residue. Lastly, Encapsulant 3 also contained an extremely high polymer content of ∼ 98 wt.% and a residue value of ∼ 2 wt.%; this matches well with the high carbon content (∼83 wt.%) and relatively low inorganic content (∼2.7 wt.%) observed for this sample. The low residue amount observed for Encapsulant 3 while still containing ∼2.7 wt.% inorganic content is contradictory to the results obtained for Encapsulant 1, where low char and residue were observed in combination with non-detectable inorganic content. This can possibly be explained by the types of inorganic elements identified in Encapsulant 3, where Bi and Sn were the predominant elements detected. Volatilization can occur during TGA causing inorganic elements to not be detected in the residue. As the molecular formula of the inorganic components is unknown, it is possible that volatilization occurred, resulting in the low residue value (∼2 wt.%) observed for Encapsulant 3.

thumbnail Fig. 5

Inorganic composition (XRF) results for the two backsheet samples and three encapsulant samples.
thumbnail 6

Weight loss data (TGA) results for the two backsheet samples and three encapsulant samples.
thumbnail Fig. 7

TGA curves for the two backsheet samples [A) Backsheet 1, and B) Backsheet 2] and the three encapsulant samples [C) Encapsulant 1, D) Encapsulant 2, and E) Encapsulant 3].

3.2 Fit with chemical recycling processes

Chemical recycling via pyrolysis is the focus of this manuscript, where the pyrolysis oil is the desired output stream in order to ‘Close the Loop.’ To determine the fit of these materials with pyrolysis, the data presented above were considered and compared to the DKR-350 guidelines. The authors would like to re-emphasize the fact that there is no open-source specification for pyrolysis feedstocks, and the DKR-350 guidelines are not extensive. One point of consideration is the polymer content, as a higher polymer content is expected to offer a higher yield of pyrolysis oil. The three encapsulant samples showed higher polymer content (all above 90 wt.%) compared to the two backsheet samples. Additionally, the polymer content obtained from the TGA data for the three encapsulant samples is in line with the guideline of maximum 10 wt.% total amount of impurities taken from DKR-350. Another point of consideration that further eliminates the two backsheet samples for consideration is the presence and amount of PET. According to DKR-350, a maximum guideline of 4 wt.% PET is allowed, and as PET was the only plastic polymer identified in the two backsheet samples, the level of PET present was above the 4 wt.% guideline.

The high polymer content observed for the three encapsulant samples corresponds well with the data obtained from elemental analysis (CHN-O), where the three encapsulant samples showed higher carbon content compared to the two backsheet samples. Additionally, the amount of residue obtained from the TGA analysis also matched well with the inorganic content determined by XRF. Backsheet 1 and Encapsulant 2 showed the highest levels of total inorganic content (∼ 22 wt.% and ∼7 wt.%, respectively) and correspondingly, these two samples also showed the highest levels of residue (∼37 wt.% and ∼ 9 wt.%, respectively). With respect to pyrolysis, the inorganic components are expected to end up in the solid char/residue output stream; therefore, materials with higher amounts of inorganic content are less desirable when trying to optimize the oil output.

Based on these results, of the five samples evaluated (two backsheet samples and three encapsulant samples), Encapsulant 1 and Encapsulant 3 show the most promise as a potential feedstock for pyrolysis. The next analysis of these encapsulant samples to assess their fitness for use in a chemical recycling process should be done under pyrolysis conditions to produce a pyrolysis oil, where lab-scale pyrolysis is a first step. The pyrolysis oil produced from such lab-scale experiments must be evaluated, providing completeness to the work presented herein. Data from the potential feedstock and corresponding pyrolysis oil are required if encapsulants from PV solar panel applications are to be considered as potential feedstock for pyrolysis, thus enabling a closed-loop recycling process for polymer-rich materials.

Beyond the fit of individual samples with a pyrolysis recycling process, the characterization of these five samples reveals a diverse range of both organic and inorganic compositions. Consequently, effective sorting of the resulting polymer mixture may be essential to maximize the value of the different polymer types, such as isolating polyolefins from encapsulant layers and PET from backsheet layers, for further processing. Various techniques are employed to sort polymers within mixed plastic waste streams, with the selection depending on the sorting facility’s capabilities and the intended end products. Spectroscopic technologies are commonly used, wherein plastics are conveyed under light sources and identified based on their distinct spectral signatures. Near-Infrared (NIR) spectroscopy is frequently applied to differentiate polymer types like PET, PE, PP, PVC, and PS; however online in a sorting facility, only limited polymers can be sorted simultaneously using this technique. Furthermore, its accuracy may be compromised by multilayer configurations, surface contamination, or dark-colored materials. Additional optical methods, such as visible (VIS) spectroscopy for color separation, X-ray fluorescence (XRF) for detecting heavier elements (e.g., chlorine and bromine), and advanced techniques like Raman spectroscopy, can also be utilized typically offline and albeit sometimes at higher costs. Beyond optical approaches, physical separation methods play a significant role; these rely on properties such as density and surface charge. Sink-float separation leverages water to differentiate plastics by density, while magnetic density separation uses both magnetic fluid and field to create density gradients. The triboelectric effect can be harnessed to separate binary polymer mixtures by exploiting differences in surface charge—for instance, rubbing together polyethylene (PE) and polypropylene (PP) and then exposing them to an electric field causes PP, which tends to acquire a positive charge relative to PE ([+] PP – PET – PS – PE – PVC [–]), to move toward the negative electrode [38,54].

Considering the polymers present in EoL PV modules, current sorting technologies are capable of separating mixed streams containing backsheets and encapsulants, yet certain challenges persist. Polymers such as PET, EVA, and PE can generally be identified and separated through established spectroscopic methods like NIR. However, online NIR is limited in the number of polymers that can be sorted, and this is compounded by the detection of thin layers of fluorinated polymers in backsheets, such as PVF or PVDF, remains problematic; these layers are often either too thin or present in insufficient quantities for reliable identification by NIR, which would likely detect only the dominant polymer, such as PET. Furthermore, while metallic layers within these materials can be identified using XRF, this technique is not available online for sorting facilities and it cannot detect elements with atomic numbers below 10, such as fluorine. Given these limitations, further research is needed to evaluate alternative spectroscopic methods and/or assess physical separation methods for isolating fluorinated polymers, especially when considering the need for online techniques that can be implemented in sorting facilities.

Pyrolysis is a promising chemical recycling technology for handling heterogeneous and contaminated end-of-life (EoL) plastic streams, especially when mechanical recycling is not possible. However, pyrolysis feasibility is challenged by the variable composition of the resulting feedstock, which often contains a mix of undesired polymers (such as PET identified in this work or fluorinated polymers that may be present), various organic and inorganic contaminants, and fluctuates depending on the input waste stream. These impurities can cause operational issues in pyrolysis plants and affect downstream processes like upgrading and steam cracking. To mitigate these challenges, measures should be implemented at the collecting and sorting stages, through the selection of the most suitable pyrolysis technology, and by adding pyrolysis oil treatment (upgrading) steps. Despite its potential, pyrolysis remains a relatively small-scale process, and its economic viability requires in-depth assessment through a detailed business model [38,55].

4 Conclusion

Recycling of photovoltaic modules presents the opportunity to recover polymer-rich fractions as feedstock for chemical recycling via pyrolysis, as an alternative to using the recovered polymer materials for energy recovery. To assess the potential of the recovered polymers as input material for chemical recycling, comprehensive characterization of the chemical composition, including carbon content and contaminant type and level, was conducted using FTIR, elemental analysis, XRF, and TGA on an albeit limited set of samples. Mechanical separation of three modules spanning different designs, cell technologies, and history yielded five polymer-rich samples, two from the backsheet layers and three from encapsulant layers. Characterization of these five polymer-rich layers from photovoltaic recycling streams already indicates a wide range of polymer type (PET, PVB, EVA and ethylene-rich polymer) and content, carbon content, and contamination depending on the source of the sample, material type, and history of the module, pointing to the complexity of polymer recycling from PV applications. Especially relevant, samples Encapsulant 1 and Encapsulant 3 show promise as potential feedstock for pyrolysis given their carbon content of 63 wt.% and 83 wt.% respectively, balanced with inorganic content below 3 wt.%. Encapsulant 2 is also a candidate for further work as it contained 79 wt.% carbon from 91 wt.% polymer and around 9 wt.% residue, hence fitting within the DKR-350 guideline. Further lab-scale pyrolysis could be conducted on these samples to generate a pyrolysis oil for analysis. In contrast, backsheet samples, dominated by PET and higher inorganic content, are less suitable due to exceeding impurity thresholds of the DKR-350 guidelines and potential process incompatibilities.

In summary, recycling polymer materials from PV modules can further enable closed-loop circularity thanks to polyolefin production from alternative feedstocks and reincorporation into polyolefin encapsulant, while maintaining the targeted performance and reliability for PV applications. While these results suggest potential for recovered polymers for chemical recycling to generate hydrocarbon feedstocks, they also point to the need for sorting polymer fractions recovered from photovoltaic modules or additional cleaning processes prior to considering chemical recycling technologies for recycling. Future work should expand sample diversity, integrate lab-scale then process-scale pyrolysis, and refine sorting and purification methods that could be required prior to pyrolysis to enable closed-loop circularity of polymers in PV modules.

Acknowledgments

The authors gratefully acknowledge Dr. Jeff Munro and Dr. Yuyan Li for their revisions of the manuscript, as well as Dow for support of part of the research disclosed in this article.

Funding

Part of the work has received funding from the European Union’s Horizon Europe research and innovation programme under Grant Agreement No. 101122298 “QUASAR”.

Conflicts of interest

The authors have nothing to disclose.

Data availability statement

Data associated with this article cannot be disclosed due to legal reasons.

Author contribution statement

Conceptualization, Caroline Grand; Methodology, Melissa N. Dunkle, Adriana Ferreira, Wolfram Palitzsch; Formal Analysis, Melissa N. Dunkle, Adriana Ferreira, Caroline Grand; Data Curation, Melissa N. Dunkle; Writing – Original Draft Preparation, Melissa N. Dunkle, Caroline Grand, Wolfram Palitzsch; Writing – Review & Editing, Melissa N. Dunkle, Caroline Grand, Wolfram Palitzsch, Adriana Ferreira, Guy Beaucarne.

References

  1. M. de l’Epine, I. Kaizuka, Trends Photovolt. Appl. 2024. https://doi.org/10.69766/jnew6916 [Google Scholar]
  2. H. Mirletz, S. Ovaitt, S. Sridhar, T.M. Barnes, Prioritizing circular economy strategies for sustainable PV deployment at the TW scale, EPJ Photovolt. 15, 18 (2024). https://doi.org/10.1051/epjpv/2024015 [CrossRef] [EDP Sciences] [Google Scholar]
  3. S. Chunduri, The role of encapsulation films & backsheets in solar module durability & efficiency, Online. Duesseldorf : TaiyangNews UG (haftungsbeschraenkt), 2025. [Accessed 28 July 2025]. Available from: https://taiyangnews.info/technology/the-role-of-encapsulation-films-backsheets-in-solar-module-durability-efficiency [Google Scholar]
  4. N.Tz. Dintcheva, E. Morici, C. Colletti, Encapsulant materials and their adoption in photovoltaic modules: a brief review, Sustainability 15, 9453 (2023). https://doi.org/10.3390/su15129453 [Google Scholar]
  5. S. Chunduri, Transition of encapsulation materials & structures from EVA to alternative technologies, Online. Duesseldorf : TaiyangNews UG (haftungsbeschraenkt), 2025. [Accessed 28 July 2025]. Available from: https://taiyangnews.info/technology/transition-of-encapsulation-materials-structures-from-eva-to-alternative-technologies [Google Scholar]
  6. A.K. Schnatmann, F. Schoden, E. Schwenzfeier-Hellkamp, Sustainable PV module design—review of state-of-the-art encapsulation methods, Sustainability. 14, 9971 (2022). https://doi.org/10.3390/su14169971 [Google Scholar]
  7. J. Lu, C. Wang, J. Zhu, Y. Wu, J. Ruan, Delamination of components for recovery of waste crystalline photovoltaic modules by three-step treatments of separating fluorinated coating, heating and ultrasonication, Chem. Eng. J. 506, 160335 (2025). https://doi.org/10.1016/j.cej.2025. 160335 [Google Scholar]
  8. S.L. Shaw, M.L. Rencheck, G.A. Siegfried, C. Libby, A circular economy roadmap for solar photovoltaics, Sol. Energy 276, 112580 (2024). https://doi.org/10.1016/j.solener.2024.112580 [Google Scholar]
  9. G. Ansanelli, G. Fiorentino, M. Tammaro, A. Zucaro, A life cycle assessment of a recovery process from end-of-life photovoltaic panels, Appl. Energy. 290, 116727 (2021). https://doi.org/10.1016/j.apenergy.2021.116727 [Google Scholar]
  10. M.M. Lunardi, J.P. Alvarez-Gaitan, J.I. Bilbao, R. Corkish, Comparative life cycle assessment of end-of-life silicon solar photovoltaic modules, Appl. Sci. 8, 1396 (2018). https://doi.org/10.3390/app8081396 [Google Scholar]
  11. Directive - 2012/19 - EN - EUR-Lex. Directive 2012/19/EU of the European Parliament and of the Council of 4 July 2012 on waste electrical and electronic equipment (WEEE) (recast) Text with EEA relevance. Online. 24 July 2012. [Accessed 23 July 2025]. Available from: https://eur-lex.europa.eu/eli/dir/2012/19/oj [Google Scholar]
  12. Directive - EU - 2024/884 - EN - EUR-Lex. Directive (EU) 2024/884 of the European Parliament and of the Council of 13 March 2024 amending Directive 2012/19/EU on waste electrical and electronic equipment (WEEE) (Text with EEA relevance). Online. 19 March 2024. [Accessed 23 July 2025]. Available from: https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=celex%3A32024L0884 [Google Scholar]
  13. J. Lin, Encapsulant technology and supply-demand dynamics. Online. 27 September 2023. [Accessed 28 July 2025]. Available from: https://www.infolink-group.com/energy-article/solar-topic-encapsulant-technology-supply-demand-dynamics [Google Scholar]
  14. M. Wendt, P. Wessel, R. Gottschalg, A. Mordvinkin, R. Heidrich, Crosslinking behavior of ethylene‐vinyl acetate copolymer encapsulants in dependence of the additive composition, Prog. Photovolt.: Res. Appl. 33, 1322 (2025). https://doi.org/10.1002/pip.3849 [Google Scholar]
  15. M.L. Pliquet, T. Béjat, M. Serasset, A. Descormes, E. Mofakhami, E. Voroshazi, Enhancing photovoltaic modules encapsulation: optimizing lamination processes for polyolefin elastomers (POE) through crosslinking behavior analysis, Sol. Energy Mater. Sol. Cells 267, 112725 (2024). https://doi.org/10.1016/j.solmat.2024.112725 [CrossRef] [Google Scholar]
  16. P.M. Thorat, S.P. Waghmare, A. Sinha, A. Kumar, G.S. Tamizhmani, Reliability analysis of field-aged glass/glass PV modules: influence of different encapsulant types, in 2020 47th IEEE Photovoltaic Specialists Conference (PVSC) (2020), Vol. 00, p. 1816. https://doi.org/10.1109/pvsc45281.2020.9300551 [Google Scholar]
  17. S.-H. Schulze, A. Apel, D. Dassler, C. Ehrich, Cure state assessment of EVA-copolymers for PV-applications comparing dynamic-mechanical, dielectric and calorimetric properties, Sol. Energy Mater. Sol. Cells 143, 411 (2015). https://doi.org/10.1016/j.solmat.2015.07.024 [Google Scholar]
  18. M. Owen-Bellini, An overview of backsheet materials for photovoltaic modules, in DuraMAT May 2020 Webinar. Online. 1 May 2020. [Accessed 28 July 2025]. Available from: https://www.duramat.org/docs/duramatlibraries/assets/webinars/duramat_webinar_may2020.pdf?sfvrsn=7934f6b5_1 [Google Scholar]
  19. K.J. Geretschläger, G.M. Wallner, J. Fischer, Structure and basic properties of photovoltaic module backsheet films, Sol. Energy Mater. Sol. Cells 144, 451 (2016). https://doi.org/10.1016/j.solmat.2015.09.060 [Google Scholar]
  20. S. Chunduri, Different backsheet structures in PV modules (TaiyangNews UG (haftungsbeschraenkt), Duesseldorf, 2025). Online [Accessed 28 July 2025]. Available from: https://taiyangnews.info/technology/different-backsheet-structures-in-pv-modules [Google Scholar]
  21. Plastics use by polymer - projections. Online. 21 March 2024. [Accessed 28 July 2025]. Available from: https://data-explorer.oecd.org/vis?tenant=archive&df[ds]=DisseminateArchiveDMZ&df[id]=DF_PLASTIC_USE_V2_2&df[ag]=OECD&lom=LASTNPERIODS&lo=5&to[TIME_PERIOD]=false&vw=ov [Google Scholar]
  22. Plastic waste collected for recycling. Online. 26 March 2024. [Accessed 28 July 2025]. Available from: https://data-explorer.oecd.org/vis?tenant=archive&df[ds]=DisseminateArchiveDMZ&df[id]=DF_PLASTIC_WASTE_6&df[ag]=OECD&lom=LASTNPERIODS&lo=5&to[TIME_PERIOD]=false [Google Scholar]
  23. W. Gao, T. Hundertmark, G. Pais, A. Ryba, J. Wallach, Addressing the challenges of plastic waste. Online. 2 September 2022. [Accessed 28 July 2025]. Available from: https://www.mckinsey.com/industries/chemicals/our-insights/addressing-the-challenges-of-plastic-waste-circularity-and-leakage [Google Scholar]
  24. OECD. Global plastics outlook: economic drivers, environmental impacts and policy options (OECD Publishing, Paris, 2022). Online [Accessed 28 July 2025]. Available from: https://doi.org/10.1787/de747aef-en [Google Scholar]
  25. M. Castell-Rudenhausen, A. Almasi, I. Vanderreydt, S. Colgan, T. Nielsen, Non-packaging plastics in Europe (European Environment Agency, 2022). Online [Accessed 29 April 2025]. ETC/CE Report 2022/7. Available from: https://www.eionet.europa.eu/etcs/etc-ce/products/etc-ce-products/etc-ce-report-2022-7-non-packaging-plastics-in-europe [Google Scholar]
  26. T.B. Anwar, K.M. Hanson, K. Lam, C.J. Bardeen, Using nanosecond laser pulses to debond the glass-EVA layer from silicon photovoltaic modules, Waste Manag. 187, 275 (2024). https://doi.org/10.1016/j.wasman.2024.07.013 [Google Scholar]
  27. J. Ko, K. Kim, J.W. Sohn, H. Jang, H.-S. Lee, D. Kim, Y. Kang, Review on separation processes of end-of-life silicon photovoltaic modules, Energies 16, 4327 (2023). https://doi.org/10.3390/en16114327 [CrossRef] [Google Scholar]
  28. K. Wambach, C. Libby, S. Shaw, Report IEA-PVPS T12-28:2024: Advances in Photovoltaic Module Recycling, 2024 [Google Scholar]
  29. NPC incorporated | Global leader of solar module manufacturing equipment. Online. [Accessed 28 July 2025]. Available from: https://www.npcgroup.net/eng/ [Google Scholar]
  30. Innovation radar > Innovation > Disassembling of frame, junction box and electrical components from End-of-Life photovoltaic panels. Online. 10 January 2023. [Accessed 28 July 2025]. Available from: https://innovation-radar.ec.europa.eu/innovation/49552 [Google Scholar]
  31. EU-funded research project: recycling of solar panels. Online. 15 April 2025. [Accessed 29 October 2025]. Available from: https://audiovisual.ec.europa.eu/en/media/video/I-271159 [Google Scholar]
  32. C. Farrell, A.I. Osman, X. Zhang, A. Murphy, R. Doherty, K. Morgan, D.W. Rooney, J. Harrison, R. Coulter, D. Shen, Assessment of the energy recovery potential of waste photovoltaic (PV) modules, Sci. Rep. 9, 267 (2019). https://doi.org/10.1038/s41598-019-41762-5 [Google Scholar]
  33. M. Królikowski, M. Fotek, P. Żach, M. Michałowski, Development of a recycling process and characterization of EVA, PVDF, and PET polymers from end-of-life PV modules, Materials 17, 821 (2024). https://doi.org/10.3390/ma17040821 [Google Scholar]
  34. A.B. de Melo, T.N.M. Mendonça, Cement blocks with EVA waste for extensive modular green roof: contribution of the components for thermal insulation, Rev. IBRACON Estrut. Mater. 10, 92 (2017). https://doi.org/10.1590/s1983-41952017000100006 [Google Scholar]
  35. K. Yan, G. Li, Q. Li, A. Diab, L. You, M. Wang, Compound modification of asphalt mixture using ethylene-vinyl acetate copolymer and amorphous poly alpha olefin, Construct. Build. Mater. 341, 127705 (2022). https://doi.org/10.1016/j.conbuildmat.2022.127705 [Google Scholar]
  36. F. Xu, Y. Zhao, K. Li, Using waste plastics as asphalt modifier: a review, Materials 15, 110 (2021). https://doi.org/10.3390/ma15010110 [Google Scholar]
  37. J. Jiang, K. Shi, X. Zhang, K. Yu, H. Zhang, J. He, Y. Ju, J. Liu, From plastic waste to wealth using chemical recycling: a review, J. Environ. Chem. Eng. 10, 106867 (2022). https://doi.org/10.1016/j.jece.2021.106867 [Google Scholar]
  38. K. Ragaert, L. Delva, K. Van Geem, Mechanical and chemical recycling of solid plastic waste, Waste Manag. 69, 24 (2017). https://doi.org/10.1016/j.wasman.2017.07.044 [Google Scholar]
  39. M. Solis, S. Silveira, Technologies for chemical recycling of household plastics – a technical review and TRL assessment, Waste Manag. 105, 128 (2020). https://doi.org/10.1016/j.wasman.2020.01.038 [Google Scholar]
  40. H.-H. Fan, C. Salas-Redondo, A. Chalaux, Life cycle assessment of an innovative high-value-recovery crystalline silicon photovoltaic recycling process − the environmental impacts and benefits, EPJ Photovolt. 15, 35 (2024). https://doi.org/10.1051/epjpv/2024034 [Google Scholar]
  41. P. Padhamnath, Recent progress in the recovery and recycling of polymers from end-of-life silicon PV modules, Sustainability 17, 4583 (2025). https://doi.org/10.3390/su17104583 [Google Scholar]
  42. K.K. Patel, S. Mallick, Pyrolysis gas analysis of ethylene-vinyl acetate (EVA) and poly-vinylidene fluoride (PVDF) polymers encapsulated in c-Si photovoltaic solar module to facilitate recycling, Circ. Econ. Sustain. 4, 1835 (2024). https://doi.org/10.1007/s43615-024-00361-x [Google Scholar]
  43. X. Chen, J. Liu, Z. Lin, J. Lu, L. Li, Y. Chen, F. Evrendilek, C. Yang, H. Yuan, Y. He, Reutilizing waste photovoltaic plastics by pyrolysis: dynamics, thermodynamics, mechanisms, products, and optimization, Fuel 384, 134005 (2025). https://doi.org/10.1016/j.fuel.2024.134005 [Google Scholar]
  44. H.C. Genuino, M. Pilar Ruiz, H.J. Heeres, S.R.A. Kersten, Pyrolysis of mixed plastic waste (DKR-350): effect of washing pre-treatment and fate of chlorine, Fuel Process. Technol. 233, 107304 (2022). https://doi.org/10.1016/j.fuproc.2022.107304 [Google Scholar]
  45. H.C. Genuino, M.C.P. van Eijk, S.R.A. Kersten, M. Pilar Ruiz, Pyrolysis of real packaging plastic waste streams in a fluidized-bed pilot plant, Energy Fuels 38, 2188 (2024). https://doi.org/10.1021/acs.energyfuels.3c04114 [Google Scholar]
  46. Deutschland, Duales System. Product Specification 04/2009 Fraction-No. 350 (Mixed Plastics). 1 April 2009. Online : Der Gruner Punkt-Duales System Deutschland GmbH. [Accessed 28 July 2025]. Available from: https://www.nedvang.nl/wp-content/uploads/2019/03/Mix-kunststoffen-DKR-350.pdf [Google Scholar]
  47. A. Faes, M. Despeisse, J. Levrat, J. Champliaud, N. Badel, M. Kiaee, T. Söderström, Y. Yu, R. Grischke, M. Gragert, J. Ufheil, P. Papet, B. Strahm, G. Cattaneo, J. Cattin, Y. Baumgartner, A. Hessler-Wyser, C. Ballif, SmartWire solar cell interconnection technology, in Proc. 29th Eur. Photovoltaic Sol. Energy Conf. (2014), p. 2555 [Google Scholar]
  48. PALITZSCH, Wolfram. Process and device for separating multilayer composite materials. DE102021109591B4. 2023 [Google Scholar]
  49. PALITZSCH, Wolfram. Hydrometallurgical process for recovery of metals and/or semimetals from waste materials containing compound semiconductor materials and/or back contact materials and/or transparent electrically conducting oxides (TCOs). EP2961855B1. 2020 [Google Scholar]
  50. Thin film backsheet with aluminum for solar panels - dyMat Apye | Coveme. Online. [Accessed 28 July 2025]. Available from: https://www.coveme.com/dymat-a-pye-l-backsheet-274/ [Google Scholar]
  51. C. Buerhop, T. Pickel, O. Stroyuk, J. Hauch, I.M. Peters, An insight into a combined effect of backsheet and EVA encapsulant on field degradation of PV modules, Energy Sci. Eng. 11, 4168 (2023). https://doi.org/10.1002/ese3.1572 [CrossRef] [Google Scholar]
  52. F. Zeng, Y. He, S. Xu, Q. Liu, L. Xi, Y. Huang, M. Kong, Y. Huang, Y. Niu, Y. Lv, G. Li, Accelerated degradation of PET-based photovoltaic backsheets under UV and acetic acid exposure, Polymer 312, 127671 (2024). https://doi.org/10.1016/j.polymer.2024.127671 [Google Scholar]
  53. F.J. Horne, J.J. Liggat, W.A. Macdonald, S.W. Sankey, Photo‐oxidation of poly(ethylene terephthalate) films intended for photovoltaic backsheet, J. Appl. Polym. Sci. 137, 48623 (2020). https://doi.org/10.1002/app.48623 [Google Scholar]
  54. S. Serranti, G. Bonifazi, Use of recycled plastics in eco-efficient concrete (2019) p. 9. https://doi.org/10.1016/b978-0-08-102676-2.00002-5 [Google Scholar]
  55. V.K. Soni, G. Singh, B.K. Vijayan, A. Chopra, G.S. Kapur, S.S.V. Ramakumar, Thermochemical recycling of waste plastics by pyrolysis: a review, Energy Fuels 35, 12763 (2021). https://doi.org/10.1021/acs.energyfuels. 1c01292 [Google Scholar]

Cite this article as: Caroline Grand, Melissa N. Dunkle, Wolfram Palitzsch, Adriana Ferreira, Guy Beaucarne, Characterization of Mechanically Recovered Photovoltaic Encapsulants and Backsheets as Potential Feedstocks for Pyrolysis Chemical Recycling, EPJ Photovoltaics 17, 6 (2026), https://doi.org/10.1051/epjpv/2025030

All Tables

Table 1

Description of the recovered samples for analysis.

In the text
Table 2

Comparison of theoretical and estimated C:H:O ratio.

In the text

All Figures

thumbnail Fig. 1

Schematic description of the layers and materials making up a generic photovoltaic module.
In the text
thumbnail Fig. 2

Structure of (a) ethylene-vinyl acetate (EVA) and (b) polyolefin elastomers (POE) polymers.
In the text
thumbnail Fig. 3

(a) Pictures of the recovered polymer-rich samples, (b) Picture of a typical embossed encapsulant film prior to lamination in a module.
In the text
thumbnail Fig. 4

Elemental analysis (CHN-O) results for the two backsheet samples and three encapsulant samples. Note: oxygen values were estimated by subtracting the carbon, hydrogen, and nitrogen values from 100%.
In the text
thumbnail Fig. 5

Inorganic composition (XRF) results for the two backsheet samples and three encapsulant samples.
In the text
thumbnail 6

Weight loss data (TGA) results for the two backsheet samples and three encapsulant samples.
In the text
thumbnail Fig. 7

TGA curves for the two backsheet samples [A) Backsheet 1, and B) Backsheet 2] and the three encapsulant samples [C) Encapsulant 1, D) Encapsulant 2, and E) Encapsulant 3].
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.