Assessing the Feasibility of Integrating a Thermal Separational Method with PV Recycling Technologies

Abstract

The growing volume of end-of-life photovoltaic (PV) panels, projected to reach 60–78 million tons by 2050, poses significant environmental challenges. With landfilling being the most cost-effective but unsustainable disposal method, developing eco-friendly processes to recover valuable materials is essential. One potential solution for recovering raw materials from PV panels is thermal treatment. Therefore, in this study, PV modules were heat-treated at a low heating rate, and their components were manually separated with an average efficiency of 90%. The recovered silicon wafers and tempered glass sheets were utilized to fabricate new PV panels using lamination technology. The applied heating parameters enabled the cells to be removed from the PV panels without structural damage. However, the results of electroluminescence tests showed that thermal treatment significantly damages the p-n junctions, rendering direct reuse in new panels unfeasible. The thermal separation methods outlined in this study offer valuable opportunities for industries employing various PV-panel-recycling technologies. These methods lay the groundwork for environmentally responsible management and recovery of materials from end-of-life solar panels, advancing sustainable recycling practices.

1. Introduction

Photovoltaic (PV) modules typically last about 25 years, and this lifespan can be extended to 30 using improved materials or technologies. The lifespan-reducing factors include extreme weather, natural disasters, fires, poor maintenance, structural defects, and damage during transport or installation. At the end-of-life stage, photovoltaic (PV) panels may become hazardous waste, with projections estimating that there will be 60–78 million tons of silicon-based PV panel e-waste by 2050 [1,2,3].

Recycling PV panels helps reduce environmental impacts by recovering materials like silicon and silver for reuse in new panels and electronic devices. Mechanical, thermal, and chemical methods are used worldwide, and grinding or shredding methods are often incorporated to recover tempered glass and metals while discarding or reusing residual materials [4]. First-generation mono- and polycrystalline silicon panels, comprising tempered glass (75%), polymers (10%), aluminum (8%), silicon cells (5%), and trace metals, dominate due to their efficiency and cost. Improper disposal, such as in landfills, releases harmful substances like lead and tin, contaminating ecosystems. Recovering rare, valuable metals like silver and copper is vital to managing the growing amount of waste from PV panels, which will surge over the next two decades. Ongoing research explores physical, thermal, and chemical recycling methods, each with unique advantages and drawbacks [5,6,7,8].

Physical methods for PV panel recycling include cutting, grinding, and milling [9]. Studies have demonstrated the use of techniques such as removing binding materials and components with hot knives or shears, followed by milling. Frames, junction boxes, and cables are first manually removed, and raw materials like silicon and nickel can be recovered via smelting. Grinding with jaw crushers or dual-rotor mills separates materials like glass and polymers into distinct fractions, aiding recycling [10,11,12,13]. Glass recovery rates of up to 100% have been achieved, though metal recovery is less efficient compared to chemical methods [14].

In the thermal method, high temperatures are employed to remove EVA adhesives and backsheets from PV panels, enabling the recovery of valuable materials but also producing toxic gases during combustion. To address challenges regarding EVA’s adhesive properties, Doni et al. (2012) introduced a delamination step to physically separate silicon wafers from glass before thermal treatment. Mechanical removal of the glass often leaves adhesive residues on the silicon cells, complicating their recovery; however, thermal treatment effectively eliminates EVA and other organic materials [15]. Wang et al. (2012) developed a combined thermal–chemical recycling process, starting with heating the panels to 330 °C to remove the Tedlar backsheet, followed by heating to 400 °C to remove the EVA adhesive, enabling the recovery of glass, silicon wafers, and metals. Silicon and copper were later extracted using wet-chemical methods [16]. Farrell et al. (2019) observed that the gas produced during EVA pyrolysis has a heating value comparable to that of biodiesel, presenting an additional energy source alongside material recovery [17]. Dobra et al. (2022) assessed the environmental impact of thermal processing, focusing on harmful emissions from Tedlar combustion, and found that pre-removal of Tedlar significantly reduced fluorine-based pollutants while also shortening the separation time for other components by 45% [18]. Jongsung et al. (2016) investigated the extraction of valuable raw materials from PV panels through thermal treatment in an oxygen atmosphere. The peak combustion temperature and heating rate were systematically adjusted during the process. Their results demonstrated that at a peak temperature of 450 °C and a heating rate of 15 °C per minute, the EVA foil in the laminates could be eliminated entirely [19]. This study provided a foundational framework for the research presented in this paper. J. Weinfurtner’s 1996 patent detailed a method involving the thermal treatment of panels at 500 °C in an oxygen atmosphere, followed by cooling via spraying water and separating glass and other materials [20]. These methods highlight the challenges of EVA adhesive removal, the potential energy value of pyrolyzed polymers, and the environmental risks associated with Tedlar^® combustion, which necessitate its pre-removal. Continuous improvements in thermal and combined recycling methods are advancing efficiency and sustainability in solar panel material recovery [21,22].

The wet-chemical delamination of EVA involves dissolving the adhesive layer to enhance its plasticity, forming a gel-like structure that reduces adhesion and enables the separation of PV panel components. Miller-Chou et al. (2003) highlighted the risks associated with volumetric changes during chemical treatment, which can cause silicon wafer fractures [23]. Organic solvents such as trichloroethylene and toluene are commonly used, with Doi et al. (2001) achieving intact cell recovery after dissolving EVA in trichloroethylene at 80 °C for 10 days [24]. Kim and Lee (2012) found toluene could completely dissolve EVA, though cell damage and hindrances occurred with other solvents [25]. Ortho-dichlorobenzene was identified as being effective without damaging silicon cells, though dissolution remains time-intensive and requires the use of volatile solvents [26]. Shin et al. (2013) employed inorganic methods, recovering silicon cells with efficiencies comparable to new ones via precipitation reactions [27]. Lovato et al. (2021) accelerated dissolution with supercritical CO₂, achieving 100% recovery of glass and metals [28]. Tembo et al. (2021) used hexane to recover 92% of cells in 24 h [29], and Pang et al. (2021) proposed using microwave-assisted technology to enhance organic solvent-based separation, with trichloroethylene being the most effective solvent [30]. Despite advancements, wet-chemical methods often involve long processing times and the use of hazardous solvents, presenting challenges for industrial application.

Mechanical, chemical, and thermal methods for recycling solar panels are subject to various challenges, including the emission of toxic gases and chemicals. Combustion and pyrolysis release hazardous substances, while mechanical processes require a substantial amount of energy. Chemical methods contribute additional emissions from reagents and organic solvents, whose handling and regeneration present further difficulties. Laboratory studies by Tammaro et al. (2015) and Bogacka et al. (2020) on thermal processing showed that 0.589 kg of solar panel waste consumed 4.9 kWh over 75 min, producing solid, liquid, and gaseous byproducts, with an indirect emission amount equivalent to 1.7 kg CO₂ per kilogram of silicon [31,32]. Thermal methods also require the use of flue gas purification to address pollutants like lead and cadmium, which can be mitigated with filtration techniques [33]. In addition to the metals evaporating from the ribbons, the thermal degradation of EVA (ethylene-vinyl acetate) foil in an oxygen atmosphere theoretically generates 2.85 g of CO₂, 0.01 g of CO, and 0.30 g of VOC (volatile organic carbon) per 1 g of ethylene-vinyl acetate copolymer [34,35]. During the combustion of PVF (polyvinyl fluoride) and PVDF (polyvinylidene fluoride), which serve as the backsheet materials of panels, halogen-containing flue gases are released [36], further increasing the environmental impact of the emitted waste gases. Therefore, proper treatment of flue gases generated during thermal processes is highly recommended. Numerous techniques have been developed for this purpose, including incineration, adsorption, absorption, condensation, photocatalytic oxidation, plasma catalysis, membrane separation, and biological methods such as biofiltration [37]. However, chemical treatment processes can produce toxic byproducts such as nitrogen oxide compounds, fluorides, and silicon compounds, posing challenges for waste management [38]. Wet-chemical processing methods, such as nitric acid treatment and the use of organic solvents, have significantly greater environmental impacts compared to mechanical and thermal methods. These factors highlight the environmental trade-offs of existing solar-panel-recycling methods. While thermal and chemical approaches are effective, their emissions and energy demands necessitate further optimization to minimize their environmental footprints [39].

The economic feasibility of a given recycling technology is the most critical parameter from an implementation perspective, alongside assessing the environmental impact and efficiency of recycling pathways. According to the reviewed studies, recycled silicon could attain a market value of approximately 350 USD/ton if local supply chains effectively integrate it as a raw material with adequate purity and yield. Overcoming technological barriers is essential to reduce costs and make recycling commercially viable, especially for high-purity silicon. Thermal processing could become the cheapest method if the value of silicon exceeds 350 USD/ton. Improvements in silicon quality and yield are crucial to increasing revenue, as is optimizing operational and capital expenditures in recycling processes. Effective policy interventions and technological innovations are vital to promoting solar panel recycling in place of environmentally harmful disposal practices [40].

Based on the information outlined above, it is crucial to select a photovoltaic (PV) panel recycling pathway that is as simple as possible, requires minimal equipment, incurs low maintenance costs, avoids the use of environmentally harmful materials, and enables the recovery of raw materials in near-pristine condition and significant quantities. Ideally, the recovered materials should be directly reusable, significantly enhancing the economic attractiveness of the process for potential investors. However, as detailed in this study, several factors can hinder the economic feasibility of recycling.

This study focuses on using the thermal decomposition method for processing PV panels, particularly targeting the recovery of intact silicon cells with minimal energy consumption. This method involves removing polymer binding layers, such as thermoplastic polyolefin (POE), to access valuable materials like silicon wafers and metals. In our experiments, we investigated the effects of treatment with organic solvents and thermal decomposition techniques. Although thermal treatment degrades the p-n junction, making the cells unsuitable for new PV panel production without further processing, this work does not address p-n junction regeneration to avoid additional costs. Instead, it focuses on the simplest recycling process for reusing cells in basic applications. This study highlights the limitations of thermal recycling, particularly the degradation of the p-n junction during high-temperature treatment. It emphasizes the key parameters and steps required to recover intact silicon wafers, serving as a foundational approach to thermal PV recycling. The experimental work demonstrated the removal of POE through heating in an oxygen atmosphere, followed by cleaning and reuse of silicon wafers and tempered glass. The recovered materials were used to assemble PV panels, and their electrical performance was measured. While this method represents a minimal recycling approach, it provides valuable insights into optimizing thermal processing for silicon recovery. This study underscores the need for further advancements to overcome technological and economic barriers in thermal PV panel recycling.

2. Materials and Methods

2.1. Preparation of Solar Panels

The solar panels used in this study were manufactured through a laminating process at an industrial partner’s facility. Due to confidentiality agreements, specific product names and lamination parameters are not disclosed in this article. The primary components of the panels are the silicon wafers, which were standard monocrystalline, anti-glare surfaceless wafers sourced from China. These wafers measure 156 mm × 156 mm, are n-type-phosphorus-doped, and deliver a power output of 5 W at 0.5 V and 2 A under 1000 W/m² of illumination. To minimize electrical losses, the wafers were segmented into smaller pieces using a laser cutter. The laser process only etches each wafer’s surface to avoid damaging the depletion zone, with the final cutting being completed manually.

The segmented pieces were sorted by polarity and soldered in series on an automated production line to form electrical strings. Copper ribbons, coated with a lead-based solder (approximately 60 wt.% tin and 40 wt.% lead), were used as the connecting material. They were sourced from China and externally fluxed. The strings were then arranged on thermally tempered soda-lime glass, also sourced from China, which served as the primary structural layer. Before assembly, the glass was cleaned with isopropyl alcohol. A binding and insulating polyethylene-octene elastomer (POE) foil was applied to the glass. Polytetrafluoroethylene (PTFE) strips were added to shield the glass during soldering. It is worth mentioning that EVA adhesive is increasingly being replaced by POE (polyolefin elastomer) foil, as the acetic acid released during the aging of EVA has a corrosive effect on the surfaces of the ribbons [41].

The strings were fixed in place by partially heating the wafers to adhere the binding foil, and Schottky diodes were soldered onto ribbons to isolate individual strings. Additional layers, including glass foil for stability, binding foil, and a back layer made of black and white isotactic polypropylene reinforced with an aluminum layer (sourced from India), were applied. Following assembly, a preliminary inspection and voltage test were performed, and the panels were placed in an oil-vacuum laminator. The panels underwent heating at approximately 160 °C under vacuum for 600 s before cooling. Excess material was trimmed after cooling. The structure of the final laminate and the respective weights and thicknesses of its layers are outlined in

Table 1. General size and mass of each layer of the solar panels used in this study.

Layer		Thickness	Weight
		[mm]	[g]
1.	Tempered soda-lime glass	3.2 ± 0.2	558.3 ± 3.1
2.	POE foil	0.2 ± 0.05	29.6 ± 1.3
3.	Wafer	0.3 ± 0.02	20.3 ± 0.5
4.	Ribbon	0.2 ± 0.01	7.4 ± 0.4
5.	POE foil	0.2 ± 0.05	29.6 ± 1.3
6.	Backsheet	0.2 ± 0.05	32.6 ± 1.8
Total		4.3 ± 0.38	677.9 ± 8.4

. The deviation values for thickness have been provided in accordance with the manufacturer’s datasheets.

2.2. Testing Methods

2.2.1. Wet-Chemical and Thermal Treatment of Solar Panels

The experimental work was based on already published studies, which investigated thermal and wet-chemical decomposition pathways in order to remove EVA/POE layers from solar panels [19].

To appropriately determine the processing method suitable for the panels, the solar modules were first ground using a Fritsch Pulverisette 13 model disc mill (Fritsch GmbH, Idar-Oberstein, Germany). The fractions resulting from grinding were subsequently treated with organic solvents (toluene ≥ 99.5% from Sigma-Aldrich GmbH (Merck Life Science Kft., a subsidiary of Merck KGaA, Darmstadt, Germany) and xylene ≥ 99.5% from Sigma-Aldrich GmbH) and subjected to heat treatment in a Nabertherm Top 60-type furnace (Nabertherm GmbH, Lilienthal, Germany). The efficiency of plastic removal was monitored through mass measurements using a Sartorius Secura1102-1S model balance (Sartorius GmbH, Göttingen, Germany).

For large, factory-new solar panels, only thermal decomposition was performed. After the thermal treatment, valuable materials such as intact silicon cells and glass sheets were manually separated from the metal-based ribbons remaining after combustion. The separated cells and glass sheets underwent ultrasonic cleaning: the first three washing cycles were performed using distilled water and were followed by three additional cycles with ethanol (≥99.5%, from Sigma-Aldrich GmbH).

2.2.2. Manufacturing of Solar Panels from Recycled Materials

The harvested raw materials sourced from the thermal process were then reused to create new solar modules based on the manufacturing process previously described. Two types of recycled panels were produced. In the first type, all components (POE foil, backsheet, ribbons, and tempered glass) were new except for the recycled silicon cells. In the second type, only the glass was recycled, with the other components (cells, POE, backsheet, and ribbons) derived from unused materials. The quality of the recovered materials was assessed using various material characterization methods, while the electrical performance of the panels was evaluated through multiple tests.

2.2.3. Electroluminescence (EL) Test

The EL test was employed to assess panel degradation and determine the active surface area of the wafers. This method is based on the phenomenon where solar cells emit light in the infrared spectrum when a forward voltage is applied to their terminals (electroluminescence). An infrared-sensitive camera (Nikon DSLR camera with infrared-filter, Nikon GmbH, Tokyo, Japan) records this emitted light, producing an image that highlights the active regions of the wafer as bright areas, while inactive or non-conducting parts remain dark. The EL-tester was developed as part of a previous research project, with all subunits and software being developed by the solar manufacturing company. The image-processing system used for the measurements captured images with a resolution of 2464 × 1632 pixels, with an image DPI value of 96 pixels/inch.

2.2.4. Flash Power Test

Solar panel power output can be evaluated using various methods, among which there is the flash power test, a rapid approach. In this test, a predetermined burst of light is directed onto the panel’s surface, while its electrical output is measured against a variable dummy load connected to a power meter. The maximum power output is calculated from these electrical measurements. The experimental setup includes a custom-built station equipped with an Ecoprogetti Ecosun Plus LED panel (Ecoprogetti GmbH, Padua, Italy) as the light source, positioned above a glass surface where the panel is placed face down. Electrical connections are made to an ITECH IT8713 programmable DC electronic load device (ITECH Electronics Co., Ltd., New Taipei City, Taiwan) for accurate readings. It offers a voltage measurement accuracy of ±(0.025% + 0.025% FS), with a resolution of 0.1 mV in the 0–18 V range and 1 mV in the 0–80 V range. For current measurement, it provides an accuracy of ±(0.05% + 0.05% FS), with a resolution of 0.01 mA in the 0–4 A range and 0.1 mA in the 0–40 A range. Additionally, the power measurement accuracy is ±(0.1% + 0.1% FS), with a resolution of 10 mW. The measurements were performed at room temperature, where the irradiance was 1000 W/m².

2.2.5. Microstructural Examination Using Scanning Electron Microscopy (SEM)

Microstructural analysis was performed using a scanning electron microscope (FEI/ThermoFisher Apreo S, Thermo Fisher Scientific Brno s.r.o, Brno, Czech Republic) under high-vacuum conditions at an accelerating voltage of 20.0 kV. Before imaging was conducted, the samples were cleaned in an ultrasonic bath using ethanol and acetone. Back-scattered electron images were captured, and the chemical composition was determined using an EDAX/Ametek Octane Elect Plus energy-dispersive X-ray spectroscopy system (EDAX Ametek Octane Elect Plus, Ametek GmbH, Wiesbaden, Germany).

3. Results

3.1. Preliminary Experiments

As mentioned earlier, the two primary methods for processing photovoltaic (PV) panels involve dissolving the EVA or POE layers using organic solvents or applying thermal treatment. Based on experimental parameters reported in the literature [19], preliminary investigations were conducted to examine the degradation behavior of the PV panels under solvent-based and high-temperature conditions.

For these experiments, strips measuring 10 cm in length and 3 cm in width were cut from the PV panels and ground using a disc mill. The disc spacing was set to 1 mm, and the rotation speed was maintained at 400 rpm. Following the grinding process, two fractions were separated using a sieve with a mesh size of 2 mm. The fraction containing particles larger than 2 mm primarily consisted of a polymer material with embedded silicon cells and metal ribbons. In contrast, the fraction smaller than 2 mm consisted mainly of tempered glass fragments.

To determine the plastic content of each fraction, they were heated. The treatment was carried out at a combustion temperature of 450 °C, with a heating rate of 1.65 °C/min and a holding time of 1 h. The results of these investigations are summarized in

Table 2. Loss on ignition of individual fractions.

		Mainly Glassy Fraction	Shreds with High Plastic Content
Initial weight	[g]	10.07 g	2.08 g
Weight after heat treatment	[g]	10.06 g	0.53 g
Loss on ignition	[%]	0.09%	74.65%

.

As shown by the results, the fraction larger than 2 mm exhibited an ignition loss of approximately 75%, indicating that this phase primarily contains the plastic-based POE adhesive, which is targeted for removal during treatment. The fraction smaller than 2 mm contained only minimal amounts of plastic particles. This suggests that by carefully optimizing grinding parameters, it is possible to separate the tempered glass from the other components of the PV panel in a single step.

For the solvent-based investigations, two commonly used solvents [24], toluene and xylene, were selected. The aim of the experiments was to evaluate the efficiency of these solvents in removing the POE adhesive. Tests were conducted at both low (25 °C) and high (110 °C) temperatures, at which the previously identified high-plastic-content fraction was used. Equal amounts of the fraction were exposed to the solvent for 8 h under magnetic stirring. The efficiency of dissolution was monitored through mass measurements, with the results summarized in

Table 3. Mass changes of the shreds with high plastic content induced by solvent treatment.

		Toluene		Xylene
Temperature	[°C]	25.0	110.0	25.0	110.0
Initial weight	[g]	2.0	2.1	2.0	2.0
Weight after dissolution	[g]	1.7	1.3	1.8	1.5
Dissolution loss	[%]	15.7	38.7	14.3	27.5

.

The results indicate that the highest dissolution efficiency was achieved using toluene at a treatment temperature of 110 °C. However, this mass change was still only half of the material loss achieved through thermal treatment. Furthermore, it is important to note that the industrial application of organic solvents poses significant environmental challenges. Additionally, heating, evaporating, and regenerating the solvents incur further costs.

Based on the insights gained from the preliminary experiments, intact PV panels were processed using thermal treatment methods.

3.2. Thermal Degradation of Strings

To study the thermal behavior of the cells in photovoltaic (PV) panels, a string consisting of 13 cells was subjected to heat treatment. The string comprised cells with silver conductive lines, ribbons, and solder, as shown in Figure 1. The thermal process was conducted at a peak firing temperature of 450 °C, with a heating rate of 1.65 °C per minute and a holding time of 1 h, and a complete string was treated to monitor changes occurring in the subunits. The microstructure of the same cells was analyzed before and after the treatment using SEM.

During the heating process, the string disassembled due to the softening of the solder, which was expected given that the melting point of solder is a similar temperature (based on the manufacturer’s data sheet). Resistance measurements conducted before and after the treatment revealed minimal changes, showing slight inconsistencies in both decreasing and increasing directions. However, the measured forward voltages significantly decreased, with reductions exceeding 0.4 V in some cases, potentially indicating damage to the p-n junction. Nevertheless, this could not be conclusively verified using the available measurement equipment (a digital multimeter). The damage to the p-n junction was clarified in subsequent stages of the experimental work through electroluminescence and flash testing. The initial experiment demonstrates that it is possible to reassemble a new string from the recovered cells following heat treatment, but only if the p-n junctions remain intact. Scanning electron microscopy (SEM) analysis of the heat-treated string revealed that the silver conductive lines detached from the wafer surfaces that were originally in contact with the solder (Figure 2).

This phenomenon was not observed in other areas where no solder was present, and the silver conductive lines in these regions remained intact, with no changes to their microstructures (Figure 3).

Based on the results obtained during the thermal processing of the strings and polymers, the entire panel was heat-treated using the same parameters (450 °C, 1.65 °C/min, and 1 h holding time). During the combustion of complete panels, it was noted that the cells remained intact and did not fracture, contrary to the findings reported by Jongsung et al. in their study [19]. In their publication, they recommended the use of counterweights during heat treatment to prevent cell breakage caused by gases released during combustion. However, their study utilized larger cells and a heating rate an order of magnitude higher than in this research. For smaller cells, it was found that applying a low heating rate effectively prevents mechanical degradation during heat treatment in an oxygen-rich atmosphere, as indicated by the intact cells shown in Figure 4.

Following combustion, the recovery of raw materials was performed manually, involving sorting (a glass sheet, wafers, and metals) and the removal of ash, predominantly originating from the backsheet.

The backsheet’s aluminum foil layer, situated between polypropylene layers, provides protection against mechanical damage and moisture. To ensure adequate material availability, three panels were thermally processed. Wafers were cleaned using ultrasonic cleaning with ethanol, requiring at least three washing cycles with distilled water and ethanol, using fresh solvents each time. The post-treatment wafers displayed missing silver conductive lines, as shown in Figure 5. Approximately 10% of the wafers were lost due to breakage, though 3% of these were partially damaged but reusable, retaining intact conductive lines. Overall, ~86% of panel materials were recoverable after cleaning; these can be observed in

Table 4. Recovery rates of individual materials after thermal treatment (see Table 1 for the approximate initial masses).

Layer		Treatment 1	Treatment 2	Treatment 3	Av. Recovery
		[g]	[g]	[g]	[%]
1.	Tempered soda-lime glass	558.35	558.12	557.88	99.98 ± 0.42
2.	Polymer foil	2.92	2.21	2.54	2.78 ± 0.39
3.	Wafer	17.92	18.78	18.18	90.17 ± 2.16
4.	Ribbon	7.31	7.27	7.22	98.39 ± 0.05
Total		586.50	586.38	585.82	86.48 ± 0.05

. The recovery efficiencies were high for specific components: glass (~100%), silicon cells (88–92%), and copper-based conductive lines (95–98%). These findings underscore the feasibility of material recovery with minimal losses during thermal processing. Once a sufficient number of solar cells had been processed, the next step was to produce solar cells using recycled raw materials.

The extraction rates obtained during the experiment were expected to yield similar results during the thermal treatment of first-generation panels, as the construction of those panels is comparable to the panel described in this study (glass–foil–cell–foil–backsheet). However, thermal treatment may present challenges in the case of glass–glass laminates, where the backsheet does not burn off, leaving insufficient space for the gases generated during the process to properly escape [43].

3.3. Silver-Based Adhesive

The first step in utilizing the recovered cells is to manufacture new strings, which are essential for assembling a complete photovoltaic panel. A silver-based adhesive was introduced to address issues in solar panel repair, leveraging technology from “Shingler-type” solar panels. This adhesive enables wafer-to-wafer bonding, replacing copper-based conductive lines, and ensures electrical conduction. The adhesive, a silver-filled bisphenol resin, cures quickly at room temperature and forms robust, conductive bonds, as verified through microscopy and X-ray analysis. Handling difficulties arose due to the adhesive’s viscosity, necessitating dilution experiments using organic solvents, as the original resin was intended to be applied using a silk-screen printer. As the use of such an instrument was not a feasible step for us at the time of the experiment, dilution was required in order to aid in the application of the resin. Ethanol provided the best consistency and was adopted for further applications. Using syringes, the diluted adhesive was applied precisely to repair missing conductive lines, forming wafer-to-wafer connections. This method reduced string size but increased mechanical stability, with the active surface area reduced due to overlapping. A minimum overlap of 2 mm per cell was determined necessary for proper adhesion. New wafers were placed at the string ends, allowing direct attachment of ribbons via soldering with fluxed and silver-coated strands. The final outcome was the production of Shingler-style strings from recycled solar cells, offering a viable method for solar panel reassembly.

3.4. Production and Testing of Laminates with Recycled Components

After successfully producing string units of adequate quantity and size for one complete laminate each, the experiments continued at a PV panel production line, where two types of panels were manufactured. In the first type, original subunits were used for all components except the solar cells, which were replaced with recycled strings. The second type incorporated a recovered glass sheet, while all other materials were brand new. The recycled glass was cleaned using detergent, isopropyl alcohol, and distilled water, leaving it suitable for use despite two minor iridescent surface defects.

Panel manufacturing was conducted using the production technology detailed in earlier sections (Section 2.1). Assembly was conducted using a Bosch pallet system. Initially, a tempered glass sheet was placed in the pallet, followed by a layer of laminating (POE) foil. Two Teflon/PTFE strips were added along the longer edges of the glass and foil to protect them during the soldering phase. Next, six solar cells, arranged according to their polarity, were laid out to form a cell matrix. During matrix assembly, conductive ribbons were soldered to the ends of the strings, with one ribbon incorporating diodes. The ends of the two outer ribbons served as connections. In this case, the production of PV panels utilizing recycled cells involved the “shingler” technology, while panels incorporating recycled glass were manufactured using the standard soldering technology, as described previously (Figure 6).

Subsequently, a polypropylene sheet was placed on the back of the cell matrix, followed by another POE foil layer and a second isotactic polypropylene backsheet. Electrical contacts were formed by attaching thick, bent ribbons using a stencil, and connections were routed through designated openings (Figure 7). Assembly concluded with optical inspection and voltage testing before proceeding to laminate production.

Laminating was performed in a dual-chamber, oil-heated laminator under single-step conditions at 150 °C and 5 mbar pressure for 8–12 s, followed by slow decompression. During this process, molten foil material appeared as excess along panel edges, which required removal. Solar panels made from different recycled materials are shown in Figure 8.

3.5. EL Imaging and Performance of Panels Made from Recycled Cells

The panel made from recycled cells was analyzed using EL imaging, but the cells showed no (Figure 9a) or minimal (Figure 9b) infrared emission. This confirmed that the diodes were completely degraded during thermal processing, making them unsuitable for reuse in solar panels. To test adhesive performance, a positive control (Figure 9c) was created using new cells, showing proper EL imaging and validating the silver-based adhesive method.

As shown in Figure 10, both the entire panel and individual strings produced black EL images, indicating complete cell failure. In contrast, the panels manufactured using recycled glass performed adequately, with some soldering defects noted. Nevertheless, these defects can be overlooked in the context of glass recycling.

Despite some terminal voltage, performance testing showed that the recycled panel could not produce usable power. These findings confirm that the current thermal processing method is ineffective for solar panel recycling. The degraded diodes and lack of usable energy output highlight the limitations of this approach.

The electrical parameters of the panels made from recycled materials are summarized in

Table 5. Electrical parameters of photovoltaic panels manufactured with recycled cells and glass (F.F. denotes the fill factor).

	F.F.	Temp.	Isc	Voc	Pm	Ipm	Vpm	Rs	Rsh
	[%]	[°C]	[A]	[V]	[W]	[A]	[V]	[mOhm]	[Ohm]
re-wafer	0.00	27.10	0.00	0.00	0.00	0.00	0.00	0.00	0.00
re-glass	75.43	27.10	0.26	51.22	10.18	0.23	43.72	19,688.58	974.93

. These results clearly demonstrate that cells recovered through thermal treatment are not suitable for producing functional photovoltaic panels without further processing.

Successful reuse would require alternative recovery methods or advanced post-treatment of wafers. Significant adjustments are necessary to improve the feasibility of solar panel recycling through thermal processing. This method, in its current state, cannot support the production of functional solar panels from recycled materials.

One potential method for regenerating solar cells is etching. During this process, the silver electrodes, the anti-reflection coating (ARC), the emitter layer, and the p-n junction were removed using a mixture of nitric acid (HNO₃, 60%) and hydrofluoric acid (HF, 49%) in a 4:1 ratio. In the second step of the etching process, the silver busbars were removed from the rear side using a potassium hydroxide (KOH, 30%) solution. Another etching method involves the removal of the SiN_x-based ARC layer and aluminum from the rear side in the first step using phosphoric acid (H₃PO₄). Subsequently, the silver, emitter layer, and p-n junction were removed with the previously mentioned HF and HNO₃ mixture [44].

Jongsung et al. also developed an alternative method where silver was first removed using HNO₃ solution (60%), followed by the removal of the ARC, emitter, and p-n layers through grinding. After grinding, the cells were further treated with a KOH solution. These methods effectively remove the degraded layers formed during thermal treatment. However, after the regeneration steps, it is necessary to re-establish the p-n junction, emitter layer, ARC layer, and silver electrodes. Despite this complicated procedure, this technology is more favorable than the complete remanufacturing of wafers [19].

3.6. Economic Analysis of the Thermal Treatment

The costs associated with various recycling and processing technologies typically constitute collection costs, processing costs (operational costs, labor costs, and depreciation), and neutralization costs (landfill fees and hazardous waste treatment costs). On the revenue side, income is generated from the sale of recovered materials during processing, primarily from glass, aluminum frames, ribbons (copper, tin, and lead), silver, silicon, and any energy potentially generated during the process.

According to analyses published in previous studies, the amortization cost of thermal recycling ranges between USD 315 and 473/ton, the operational cost ranges between USD 373 and 533/ton, and maintenance costs (personnel and overhead) range between USD 37 and 135/ton. Compared to simple shredding and mechanical processing, thermal treatment is the most expensive among the currently available industrial-scale techniques. Wet-chemical processing, on the other hand, has only been validated at the laboratory level, making its associated costs difficult to predict.

For shredding-based processing, a significant portion of the revenue comes from selling aluminum frames and ribbons. Additionally, the revenue from mechanical processing is significantly increased by the sale of metallurgical-grade silicon (USD 59–94/ton of processed solar panels), while in the case of thermal processing, the revenue is further enhanced by the sale of solar-cell-grade silicon (USD 194–775/ton of processed solar panels). Solar-cell-grade silicon refers to material that, after recycling, can be directly used to manufacture new solar panels without additional preparation. The second-largest revenue source during processing is the sale of recovered silver, the value of which depends on the purity of the recovered material (approximately USD 446–702/ton of processed solar panels). Based on these analyses, the expected cost of thermal processing is around USD 1034–1420/ton, while the revenue generated by this technology is projected to be approximately USD 742–1339/ton. However, in regions lacking silicon-processing facilities or PV-manufacturing industries, the value of the recovered materials is lower, making mechanical recycling a more practical solution.

These analyses, however, do not account for the possibility that the p-n junction in the cells may be damaged during thermal treatment, requiring additional post-processing. They also do not consider the potential revenue from syngas obtained during the pyrolysis of PV panels, which could be used for energy sales or reducing maintenance costs [40,45,46].

As outlined in the introduction, five different alternative solar-panel-processing technologies are available, each with distinct environmental and economic implications: landfill disposal, shredding, mechanical processing, thermal processing, and wet-chemical processing. Landfill disposal has the advantage of being the most economically favorable option; however, it does not allow for the recovery of materials from used solar panels. Shredding technology typically generates profit only through the sale of tempered glass, resulting in a low environmental impact and minimal installation costs for the processing plant. However, its main drawback, similar to landfill disposal, is that valuable materials remain locked in adhesives and cannot be recovered.

The main advantage of mechanical processing is that, apart from airborne dust, it does not release other environmentally harmful substances. However, its economic feasibility is limited. Wet-chemical methods, on the other hand, have the benefit of avoiding thermal impacts on the cells, enabling the recovery of silicon-based wafers in reusable form. The primary disadvantage consists of the numerous hazardous and carcinogenic chemicals required to dissolve EVA foil, making this approach undesirable from a green chemistry perspective.

Thermal processing can release harmful substances into the air, necessitating proper treatment, and involves relatively high initial investment costs compared to other technologies. Additionally, further post-treatment of the recovered cells is required to integrate them back into solar panels. Nonetheless, its primary advantage is its relatively simple, single-step separation of panel components. Depending on the specific technology, syngas can also be extracted during the process, making thermal processing economically attractive. Thus, thermal treatment can be used not only to recover materials but also for energy production [40,43,45].

4. Discussion

The initial step in processing the photovoltaic (PV) panels involved grinding them using a disc mill, followed by the size-based separation of the resulting fractions. The results showed that tempered glass could be effectively removed from the panels’ surfaces, leaving only minimal residual POE on the glass surfaces. This finding is particularly valuable for processing techniques that rely on the sale of tempered glass as a primary revenue source. Furthermore, the coarse fraction left after grinding, containing over 75 wt.% POE, has potential for syngas production. Under inert heating conditions, the reaction produces a high-calorific-value product [17] along with a solid-phase byproduct rich in valuable metals, such as silicon and copper. Supposedly, the electrical energy generated from the syngas can be reintegrated into the system, reducing operational costs.

Thermal treatment experiments showed that cell degradation within panels can be prevented by using a low heating rate (1.65 °C/min). This is essential for continuous-furnace technologies, enabling shorter cycle times and improving the economic feasibility of material recovery.

In contrast, organic solvents were less effective in removing POE. This method not only contaminates the solvent but also makes the dissolved polymer unsuitable for processes like syngas production. Therefore, thermal treatment is clearly the preferred approach for the initial stage of PV panel processing.

After thermal treatment, the subunits within the PV panels can be easily separated manually without the need for advanced techniques. Cleaning these subunits is also straightforward. The method described in this study enables the recovery of approximately 86.5 wt.% of the PV panel material, with recovery rates for valuable materials (glass, cells, and ribbons) exceeding 90%. However, thermal degradation of the p-n junctions in silicon wafers limits their potential for reuse.

Recovered cells can be used to produce new strings through the Shingler method, using adhesives with silver-based bisphenol resin. These strings are compatible with standard photovoltaic-panel-manufacturing processes and do not require technological modifications. However, electrical tests indicate that these strings are not suitable for direct reuse in functional solar panels without further processing.

In summary, the thermal treatment method presented in this study allows for the recovery of tempered glass, silicon wafers, and copper-containing ribbons from photovoltaic (PV) panels without causing significant physical damage. The recovered glass can be directly reused in new solar panel production, increasing revenue from processed materials. However, silicon cells require post-processing due to thermal damage to the p-n junction. Despite this limitation, the thermal processing method offers significant technological advantages. The subunits of the panel can be manually separated, eliminating the need for complex or costly separation techniques. Although the tin and lead coatings on the ribbons are degraded during thermal treatment, they can be easily separated and isolated for further metallurgical processing.

This technology highlights several areas where the costs of PV panel recycling can be reduced and identifies aspects where improvements are necessary to make the process both economically viable and environmentally sustainable.