Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.
Advertisement
Scientific Reports volume 16, Article number: 4986 (2026)
1220
2
Metrics details
Bifacial photovoltaic (PV) modules have been receiving increasing attention because of the harvesting light from both front and back sides. End-of-Life (EoL) PV modules output grow annually, which are rich in recyclable resources such as silicon and metals. A critical prerequisite for recovery is the separation of the laminate. This study presented a novel and rapid separation strategy by laser (1200 W power, 2000 Hz frequency, 5% duty cycle), achieving complete separation of the silicon cells from the Ethylene Vinyl Acetate (EVA) interlayer. Characterizations analysis showed that the surface morphology of anti-reflection coatings (ARCs) on the silicon cells was damaged. In addition, the content of silicon nitride, main component of ARCs, was decreased. Furthermore, the analysis of EVA verified the cleavages C-O bond, while the cross-linked structure of EVA was not disrupted. Thus, complete separation was driven by ARCs disappearance and the bond cleavage. Life cycle assessment demonstrated that this approach was more environmentally friendly than thermal and chemical methods. It also overcame the drawbacks of product mixing and difficult sorting associated with mechanical methods, enabling simple, rapid separation and easy large-scale automation. This strategy dramatically reduces the energy consumption and provides sustainable pathway for recycling EoL bifacial PV modules.
The global photovoltaic (PV) industry has experienced rapid growth in recent years. The International Energy Agency predicts that by 2030, the volume of waste PV modules will reach 1.7-8 million tons, and surge to 60–78 million tons by 20501. Given the presence of the heavy metal lead in End-of-Life (EoL) PV modules, the European Union has classified them as hazardous waste, which mandates proper recycling and disposal2,3,4. Meanwhile, PV modules fall under the category of electronic waste, often referred to as “urban mines”. They contain not only high-value metals such as tin, lead, copper, silver, and aluminum but also valuable materials including tempered glass and silicon (Si)5,6,7. In turn, the recycling of EoL PV modules can alleviate shortages of primary mineral resources while mitigating the high energy consumption and pollution associated with the traditional metallurgical industry8.
EoL PV modules are categorized into monofacial and bifacial types. Unlike monofacial PV modules, bifacial modules replace the backsheet with glass. Additionally, their silicon cells are equipped with silver wires and anti-reflection coatings (ARCs) on both sides (Fig. 1). By harvesting light from both front and back sides, bifacial PV modules achieve higher PV output9,10,11. From the backside to the frontside, they comprise an external junction box, aluminum frame, and an internal laminated structure—sequentially including glass, upper EVA (Ethylene Vinyl Acetate) layer, silicon cells layer, lower EVA layer, and glass. While bifacial PV modules manufacturing are more costly than monofacial modules as a result of additional materials (e.g., double glass) and processes (e.g., post-screen printing contacts), their levelized cost of electricity still holds a competitive advantage10,12,13. The International Technology Roadmap for photovoltaic predicts that the market share of bifacial PV modules will hit 85% by 203214. However, the research community has conducted relatively few studies on the recycling of bifacial PV modules. To date, the focus of most PV module recycling studies remains on backsheet-based modules.
For the recycling of PV modules, junction boxes and backsheets are typically removed via mechanical methods initially15,16,17. Subsequently, the remaining PV laminate is separated, and the resulting materials are recovered. In the current research, three primary methods are widely employed for laminate dissociation: thermal treatment, chemical processes, and physical methods16,18,19,20. Thermal treatment typically entails high-temperature heating to incinerate or pyrolyze EVA, thereby achieving dissociation. However, direct thermal treatment induces release of toxic gases, thereby requiring additional gas treatment21. Chemical processes typically involve immersing EVA in chemical solvents (e.g., toluene, xylene, trichloroethylene, etc.) to either dissolve or swell it, or utilizing supercritical fluid technology. EVA consists of non-polar ethylene segments and polar vinyl acetate segments. This unique structure enables EVA to be dissolved by certain non-polar and polar solvents22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37. Physical methods involve first crushing the laminate using equipment such as crushers or high-pressure pulse systems. Material recovery subsequently occurs either by utilizing compositional differences among particle sizes or by separation techniques such as electrostatic separation and flotation38,39,40,41,42,43,44,45,46. Each method has inherent limitations: thermal treatment requires large-scale equipment, and suffers from high energy consumption plus the need for waste gas treatment; chemical processes are plagued by low treatment efficiency, costly solvents, and large volumes of waste liquid generation; physical methods often produce mixed products and low separation efficiency, creating challenges for subsequent resource purification and recovery15,18,47,48,49,50.
As an emerging separation technology proposed in recent years, laser-based separation offers distinct advantages for interlayer separation of PV laminates, including low pollution, high efficiency, and strong selectivity. Li et al. irradiated the silver-aluminum coating of silicon cells using a pulsed laser, successfully stripping and recovering EVA without altering its properties51. Their study noted that the silver-aluminum coating on the silicon cells backside absorb laser energy, causing localized temperature increases at the interface and weakening the bonding strength of EVA. By contrast, Anwar et al. used a pulsed laser to directly transmit through glass and EVA to irradiate silicon cells52. They attributed the complete separation of the EVA-silicon cells interface to chemical changes or vaporization of the ultra-thin EVA layer at the interface under instantaneous high temperatures. This prevented re-adhesion to the silicon cells and thus separated the bonding surface from the inside. Notably, both studies only achieved separation of specific layers within the PV laminates and lacked a clear elaboration of the separation mechanism.
This study proposed a continuous laser-based method for separation of bifacial PV laminates. This method enabled separate recovery of silicon cells from bifacial PV laminates, with selective separation at the silicon cells-EVA and no residual EVA on the silicon cells surface. The study focused on three key objectives: (i) Determining the optimal laser parameters for separation. (ii) Investigating changes in EVA during laser processing via Fourier-transform infrared spectroscopy (FTIR), Raman spectroscopy, and Thermogravimetric analysis (TGA). (iii) Characterizing the surface morphology of silicon cells using scanning electron microscopy/energy-dispersive X-ray spectroscopy (SEM/EDS), and X-ray photoelectron spectroscopy (XPS). This work highlights three core scientific innovations: selective removal of the ARCs, establishment of the mechanistic connection between ARCs degradation behavior and EVA debonding efficiency, and demonstration of continuous laser’s superiority in PV recycling.
Since PV modules are designed to convert light energy into electricity, their design inherently prioritizes high light transmittance for glass and encapsulant materials. Specifically, PV glass requires a light transmittance of over 90%, while EVA encapsulants exhibit minimal absorption of light within the 380–2200 nm wavelength range This provides a robust foundation for laser-based separation51,53. The laser penetrates the glass and EVA to irradiate the silicon cells, with nearly all its energy absorbed by the cells. Given the absence of external electrical devices or energy storage systems, the generated electrical energy cannot be consumed or stored. Thus, almost all laser energy is converted into heat. To prevent EVA pyrolysis (and subsequent waste gas generation) due to temperature increases during laser processing, it is critical to determine EVA’s pyrolysis temperature. TGA shows that EVA in the bifacial modules studied exhibited mass loss starting at 243 ℃ (Fig. S11). Consequently, during laser processing, the temperature of PV modules must be maintained below this threshold. Moreover, as a thermoplastic material, EVA is typically integrated into PV modules via hot-pressing during manufacturing. It begins to soften with reduced viscosity at 60 ℃, loses viscosity entirely around 150 ℃, and regains viscosity upon cooling to room temperature54,55.
To investigate the effect of temperature on PV module separation, a PV strip was placed on a heating stage with the temperature set to 190 ℃. The PV strips took 14 s to reach 190 ℃ on the heating stage, whereas laser treatment required 38 s to achieve the same temperature. Both heating methods were conducted under continuous temperature monitoring, with heating terminated immediately upon reaching 190 ℃. However, the temperature of the PV strips continued to rise within a few seconds after the laser irradiation ceases. Therefore, laser scanning was stopped when the monitored temperature reached 176 ℃, resulting in a maximum temperature peak of 189 ℃. Such stringent control over temporal and temperature parameters effectively excluded the influence of the total thermal budget on the experimental results (Fig. 2).
The separation behavior at this temperature was shown in Fig. 3a: both the glass-EVA and EVA-silicon cells interfaces separated, but EVA remained adhered to the silicon cells surface. Silicon cells were irregularly distributed on the upper and lower EVA layers. This was because the temperature exerted the same effect on the upper and lower contact surfaces between silicon cells and EVA. These observations indicated that temperature can reduce EVA viscosity, thereby facilitating separation, but separation driven solely by temperature lacks selectivity.
A 600 W laser was then used to scan the PV modules with a 5% duty cycle. This parameter set was selected due to its moderate temperature rise rate. Scanning was halted once the temperature reached 190 ℃, after which the glass, together with EVA, was peeled away (Fig. 3a). Notably, no EVA residue was observed on the silicon cells surface, and most of the cells separated from the upper EVA layer. The selective separation of the silicon cells-upper EVA interface was achieved .
The experiment was repeated using the same parameters: after the PV strips cooled, scanning was performed again, and this process was repeated ten times. During scanning, the deep blue color of the silicon cells surface gradually faded to gray (Fig. 3b). This phenomenon indicated that the ARCs were damaged under laser irradiation. So we reached a preliminary conclusion. Once the ARCs was damaged, EVA loses its attachment sites, leading to weakened adhesion. However, since ARCs damage was confined to the laser-scanned side, the ARCs on the opposite side of the silicon cells remained intact. This leaves EVA on the non-scanned side still tightly bonded to the cells, resulting in most of the silicon cells remaining adhered to the lower EVA layer.
To verify this preliminary conclusion, five PV strips with varying degrees of ARCs damage were prepared, and peeling force tests were performed on them (Fig. S10). The relationship between their gray discoloration area (corresponding to ARCs damage) and the peeling force is presented in Table 1. It can be clearly observed that as the area of the damaged ARC region increases, the required pulling force decreases.
This observation presented a critical opportunity for laser-based selective separation. The subsequent section will optimize laser parameters to identify conditions yielding the optimal selective separation effect.
First, PV strips were scanned at a maximum power of 1200 W with a 100% duty cycle. At a 100% duty cycle, the laser operates continuously, so no frequency parameter applies. Thick smoke and open flames emerged during scanning, forcing the process to stop within 2 s. The scanned PV strip was shown in Fig. 4a: the silicon cells surface turned completely gray, attributed to ARCs damage. After lifting the glass, most silicon cells were found attached to the lower EVA layer. The upper EVA-silicon cells interface had separated. A yellow oily substance was observed on the silicon cells surface, identified as EVA pyrolysis products56,57. Black deposits on EVA were presumed to result from EVA carbonization at high temperatures. Excessive laser power caused the PV strip to absorb large amounts of energy, leading to uncontrolled, rapid temperature rise.
Bifacial PV module structure and silicon cells structure diagram.
Schematic diagram of separating photovoltaic laminate with laser.
(a) Thermal treatment vs. laser treatment effect diagram. (b) The anti-reflective coatings gradually disappeared as the number of scans increases.
(a) Photovoltaic strips after laser irradiation at 1200 W with a duty cycle of 100%. (b) Photovoltaic strips after laser irradiation under conditions of 500 W and a duty cycle of 100%. (c) Photovoltaic strips after laser irradiation under conditions of 200 W and a duty cycle of 100%. (d) Photovoltaic strips after laser irradiation under conditions of 500 W, 40% duty cycle and 200 Hz. (e) The photovoltaic strips after laser irradiation under the conditions of 1200 W, duty cycle of 5%, and frequency of 2000 Hz.
To control temperature and eliminate smoke/flames, the power was reduced to 500 W. Scanning at 500 W and 100% duty cycle avoided open flames. While the heating rate was controllable, a faint odor was still detected during laser irradiation. The scanned PV strip (Fig. 4b) showed most of the silicon cells surface had turned gray. After peeling, black deposits were still present, but no yellow oily substance formed. The vast majority of silicon cells remained attached to the lower EVA layer, with good selective separation.
To further prevent exhaust gas pollution from EVA denaturation, power was reduced to 200 W. PV strips were scanned until the temperature reached 190 ℃, and no exhaust gas was detected by VOC detection equipment. The PV strip (Fig. 4c) showed no color change on the silicon cells surface. After peeling, silicon cells adhered to both upper and lower EVA layers. This separation was consistent with the heating stage effect described above. These results confirmed that ARCs damage was power-dependent: higher power leads to more effective ARCs degradation.
Next, the effect of frequency on separation was investigated. PV strips were scanned at 500 W, 40% duty cycle, and 200 Hz. The effective power output per unit time at 40% duty cycle was equivalent to that of a 200 W laser operating at 100% duty cycle. After scanning to 190 ℃, the PV strip (Fig. 4d) exhibited blue-white stripes. This occurred because the low frequency prevented the laser’s effective irradiation coverage from exceeding its displacement per unit time (at a 30 cm/s scanning speed). Specifically, under the conditions of 2000 Hz and a 5% duty cycle, the laser-off time per cycle was calculated as 475 µs. During this period, the PV strips traveled a distance of 0.1425 mm at a scanning speed of 30 cm/s. Since the beam spot width (0.4 mm) was larger than this displacement, enhanced uniformity of anti-reflective coatings (ARCs) degradation was achieved. In contrast, at 200 Hz (with the same 5% duty cycle), the traveled distance (1.425 mm) exceeded the beam spot width, leading to the formation of blue-white ablation stripes. After peeling, some silicon cells remained attached to the upper EVA layer, failing to maximize the laser’s separation benefits.
Two solutions exist for this issue: reducing scanning speed or increasing frequency. However, reducing scanning speed would decrease processing efficiency, it was an undesirable outcome, so the frequency was set to its maximum value of 2000 Hz. Reducing the duty cycle effectively regulated power output per unit time, preventing rapid temperature rose while avoiding exhaust gas generation and unsymmetrical ARCs damage. At 1200 W and 2000 Hz, the duty cycle was gradually reduced to 5%, at which point no exhaust gas was detected by VOC equipment.
Scanning the PV strip at 1200 W, 2000 Hz and 5% duty cycle to 190 ℃ yielded the result shown in Fig. 4e: the ARCs was uniformly damaged. Blue streaks were caused by laser scattering from glass cracks, which led to less ARCs damage. The upper EVA-silicon cells interface was completely separated, with nearly all silicon cells attached to the lower EVA layer. This was attributed to that the interfacial adhesion of the upper EVA layer on the laser-irradiated surface was significantly weaker than that on the non-irradiated side. This difference in interfacial adhesion may originate from the surface modifications of the silicon cells. Irradiating the opposite glass side with the laser allowed easy scraping of the silicon cells and PV ribbons (Fig. 4e).
To investigate the separation mechanism, characterization tests were conducted on EVA and silicon cells, starting with silicon cells analysis. Figure 5a presented the surface morphology and elemental composition of the pristine silicon cells. At 50 μm magnification, the silicon nitride on the surface exhibited a distinct rhombic structure. Figure 5b showed the surface morphology and elemental composition of Sample 2. Its central region retained the same morphology as the pristine silicon cells (Sample 1), while its peripheral regions displayed morphological changes induced by laser treatment. Partial destruction of silicon nitride resulted in a significant decrease in nitrogen content and a corresponding increase in silicon content. Figure 5c illustrated the surface morphology and elemental composition of Sample 3. Thorough laser treatment led to the complete disappearance of silicon nitride, with no rhombic structure observable. Notably, the same laser-induced morphological features were observed in Fig. 5c as in Fig. 5b. Additionally, Fig. 5c also exhibited porous structures induced by the high temperature generated by the laser, which was consistent with the findings of Coyne et al58. The spherical agglomerated structure, meanwhile, aligned with results reported in Ulmeanu et al. and Kumar et al59. The significant increase in oxygen content was attributed to the formation of silicon oxide, while the continuous decrease in nitrogen content further confirmed the progressive destruction of silicon nitride.
XPS analysis was performed on the three samples, with their full survey spectra shown in Fig. 6a-c. In the peak fitting analysis, the background type was set to Tougaard, and the FWHM constraints were specified as (0.2, 5). The full spectra of Sample 1 and Sample 2 were nearly identical, whereas Sample 3 exhibited a strong peak at 532.7 eV—assigned to silicon oxide—consistent with the EDS results. Following laser treatment (Sample 2), the nitrogen content (defined as the atomic percentage of nitrogen relative to the total atomic content of nitrogen, silicon, and oxygen) was lower than that of the pristine silicon cells (Sample 1). After complete laser-induced removal of the ARCs (Sample 3), the nitrogen proportion relative to nitrogen, oxygen and silicon further decreased. This aligned with the elemental proportion trends observed via EDS.
The high-resolution Si2p spectrum of Sample 1 (Fig. 6d) showed two distinct peaks: one at 101.9 eV (assigned to N-Si-N) and another at 102.6 eV (assigned to Oₓ-Si-Nγ), which aligned with the findings of Chen et al. and the Handbook of X-ray Photoelectron Spectroscopy61,62. For Sample 2 (Fig. 6e), the high-resolution Si2p spectrum revealed a significant decrease in the intensity of the N-Si-N peak (101.9 eV), accompanied by the emergence of a new peak at 103.1 eV (assigned to silicon oxide). This indicated partial destruction of silicon nitride. In the case of Sample 3 (Fig. 6f), the silicon nitride peaks were weakened to near-undetectable levels, while the silicon oxide peak was highly prominent. Additionally, a peak at 99.3 eV (assigned to elemental Si) appears, which was attributed to unoxidized silicon exposed following the destruction of the silicon nitride layer. The progressive attenuation and near-disappearance of ARCs-specific silicon nitride peaks (N-Si-N and Ox-Si-Ny) with increasing laser treatment confirmed structural breakdown of the ARCs. The emergence of an elemental silicon peak (99.3 eV) was observed in the fully treated sample, which arose from unoxidized silicon exposed only after the ARC layer was removed.
Collectively, the aforementioned SEM/EDS and XPS analyses confirmed that the ARCs on the laser-irradiated surface of the silicon cells was damaged. In the manufacturing process of PV modules, glass and silicon cells are bonded using EVA as the adhesive via vacuum hot pressing63. More specifically, the EVA is adhered directly to the ARCs of the silicon cells. Consequently, the damage to the ARCs on the irradiated side resulted in the loss of EVA’s original adhesive sites, ultimately leading to complete adhesion of all silicon cells to the lower EVA layer. Notably, Samples 1 and 2 exhibit significantly higher carbon content, indicating the potential presence of organic components on the silicon cells surfaces. For Sample 3, which underwent additional laser processing, these organic components were pyrolyzed, leading to a reduction in carbon content. These organic species are inferred to derive from EVA pyrolysis, highlighting the need for further analysis of EVA.
To investigate whether EVA undergoes property changed during laser processing and to determine their role in facilitating separation. FTIR spectroscopy, Raman spectroscopy, and TGA were performed on pristine EVA and laser processed EVA. Laser energy is selectively absorbed by the silicon cells and confined to the EVA-silicon cells interface (irradiated area). This process induces only structural and chemical modifications within a limited interfacial EVA layer (≈ 10–50 μm), including the breaking of C-O bond and mild melting of the EVA matrix. Thus, all the aforementioned characterizations were targeted specifically at this thin interfacial EVA layer.
FTIR results were shown in Fig. 7a, the laser processed EVA exhibited the disappearance of the peak at 1029 cm− 1 (corresponding to the C-O bond) indicating a deacetylation reaction, which aligned with the first step of EVA pyrolysis57,64. This phenomenon occurred as follows: during laser irradiation, the silicon cells surface absorbs energy, causing a rapid temperature rise. This led to the temperature at the silicon cells-EVA interface exceeding EVA’s thermal stability threshold. Notably, given that the mass of silicon cells was much smaller than that of glass and EVA, glass and EVA heated up slowly during heat transfer. Thus, while the bulk temperature measured during the experiment did not exceed 190 ℃, the local temperature at the silicon cells-EVA interface had already reached the threshold required for EVA’s initial pyrolysis. During the laser processing, it was observed that the surface temperature of the silicon cells rose to 289 ℃ and stabilized, indicating that heat absorption and heat dissipation were in equilibrium. This temperature satisfies the conditions for the deacetylation reaction but does not meet the criteria required for the main-chain pyrolysis of EVA. This effectively accounts for the high carbon content observed in XPS analysis, which derives from acetic acid generated during the initial pyrolysis of EVA. Furthermore, the elimination of acetic acid contributed to the reduction in interfacial adhesion. Retention of EVA’s other characteristic peaks confirms the integrity of its main chain.
Raman spectroscopy results (Fig. 7b) indicated that the characteristic peaks of the two EVA samples (pristine EVA and laser processed EVA) were generally consistent. The main difference resided in peak intensity: both 1065 cm− 1 and 1447 cm− 1 correspond to the characteristic peaks of the C-O bond, with an intensity of 161065. In contrast, for laser processed EVA, the intensities of these two C-O bond peaks decreased to 1256. This observation further confirmed a reduction in C-O bonds, supporting the occurrence of the deacetylation reaction (consistent with FTIR findings). Notably, the peaks in the 2800–3000 cm− 1 range (corresponding to C-H stretching vibrations in EVA’s main chain) were perfectly matched across the two samples. This indicated that EVA’s cross-linked structure remained intact67.
TGA showed that in the first stage (below 400 ℃), the weight loss curves of the two EVAs overlapped initially. At 360.15 ℃, the weight loss rate of pristine EVA exceeded that of laser processed EVA. This was attributed to the partial thermal decomposition of the laser processed EVA. Specifically, the EVA in contact with the silicon cells had already undergone preliminary deacetylation during laser processing64,68. Since this contact-area EVA accounted for only a small fraction of the total EVA mass, the weight loss rates of the two samples remained consistent before 360.15 ℃. Beyond this temperature, pristine EVA still required extensive acetic acid removal, while the laser processed EVA was nearly finished with acetic acid release. This led to the temperature-dependent weight loss percentage of pristine EVA being exceeded by the laser processed sample—a trend reflected in the first intersection of the two curves in Fig. 7c.
In the second stage (main chain pyrolysis, above 400 ℃64,68), the curve trend mirrored that of the first stage: the weight loss rates overlapped in the early phase but diverged in the later phase. This followed the same mechanism: for samples of equal mass, the laser processed EVA had a higher proportion of intact main chain (since partial low-molecular-weight components had already been removed during laser processing). During main chain pyrolysis, this higher main chain proportion caused the mass percentage of the laser processed EVA to change more rapidly with temperature compared to pristine EVA. This difference was visualized in Fig. 7d.
The above analyses collectively confirm that the selective separation of EVA and silicon cells is attributed to the combined effects of ARCs ablation and alterations in EVA properties. Specifically, ARCs ablation eliminated EVA’s adhesion sites on the silicon cells surface. Concurrently, EVA property modification led to the loss of its viscosity, while C–O bond cleavage in the thin interfacial EVA layer weakened its interfacial adhesion.
The SimaPro 9.4.0.1 software (https://simapro.com) was used to perform an LCA, comparing the environmental impacts of the method developed in this study with those proposed by two other researchers for PV laminate separation. Pang et al. employed a chemical method using trichloroethylene (scenario 3)69. Gao et al. adopted a combined physical-thermal approach (scenario 2)70. This study used a laser layering strategy (scenario 1). The LCA was conducted on all three processes at the laboratory scale, assessing the separation of silicon cells from 100 g PV laminations. The system boundary was defined to include the process from obtaining photovoltaic lamination to separating silicon cells. Upstream processes, such as the transportation of PV modules and the initial stripping of Al frame and junction box from the modules, were excluded from the analysis. The materials and energy consumption involved in the production of the equipment themselves, such as laser devices and milling machines, were also excluded from the analysis. In contrast, the electricity and chemicals consumed for the treatment of PV strips, as well as the products obtained post-treatment, were included in the analysis.
Based on the Ecoinvent 3 database, the input and output flows of the processes involved in this study as well as detailed energy flow data were calculated. The analysis was calculated according to their energy flow. Detailed energy flow data are shown in Table S1-3. LCA evaluated 18 environmental impact categories, of which 12 were selected for presentation, including but not limited to climate change, ozone depletion, terrestrial acidification, freshwater eutrophication, human toxicity, photochemical oxidant formation, particulate matter formation, terrestrial ecotoxicity, ionising radiation, urban land occupation and fossil depletion (Table S4). These twelve impact categories were selected due to their strong correlation with the electricity consumption and chemical usage involved in the three scenarios examined in this study.
This study’s method was more environmentally friendly than the other two. By avoiding chemical reagents and thermal steps, CO2 emissions and fossil fuel consumption were significantly reduced. In addition, the laser-based method in this study offered higher efficiency than the other two: if the laser width was increased to half the width of the PV laminate, only 40 s would be required to process a 160 × 100 cm PV laminate. It also outperformed the other two methods in terms of other environmental impacts, as was illustrated in Fig. 8. However, the LCA process was solely based on small-scale treatment processes at the laboratory scale. For industrial-scale production, the reduction in environmental costs resulting from the scale effect should also be taken into account.
SEM/EDS and elemental contents of (a) Sample 1: pristine Si cells. (b) Sample 2: laser processed Si cells. (c) Sample 3: Si cells with the anti-reflection coatings completely removed.
(a) The full XPS spectra of (a) pristine Si cells, (b) laser processed Si cells and (c) Si cells with the anti-reflection coatings completely removed. The Si2p spectra of (d) pristine Si cells, (e) laser processed Si cells and (f) Si cells with the anti-reflection coatings completely removed.
(a) FTIR and (b) Raman analysis of the pristine EVA and laser processed EVA. (c) TGA and (d) Deriv. Weight of the pristine EVA and laser processed EVA.
Environmental impacts of the three scenarios.
The laser penetrates the glass and EVA layers to irradiate the silicon cells. This achieved separation of the silicon cells-EVA interface by destroying the ARCs and inducing the denaturation of thin interfacial EVA directly contacting with the silicon cells. Temperature elevation effectively reduces EVA viscosity, aiding separation, but cannot alone enable selective separation.
Collectively, characterization results support the separation mechanism: XPS and electron microscopy analyses confirm ARCs degradation and silicon oxide formation on the silicon cells surface. Higher laser power enhancing ARCs destruction and higher frequencies promoting uniform ARCs damage. FTIR and Raman spectroscopy reveal that EVA undergoes deacetylation at the instantaneous high temperature of the silicon cells-EVA interface, causing viscosity loss, it is an additional key driver of separation. This is corroborated by thermogravimetric analysis. Laser-based separation enables efficient silicon cells recovery from bifacial PV modules, with the equipment easily adaptable to industrialization and automation. LCA studies have confirmed that this method is more environmentally friendly than thermal and chemical approaches.
This study offers novel insights into laser-assisted PV laminate separation and validates a practical approach for silicon cells recovery from PV modules. Future research should address industrialized challenges in processing full-size, large-area PV modules, such as, integration with existing recycling infrastructure, and real-time thermal monitoring for process optimization.
The scheme proposed in this study incurs a certain degree of increase in processing costs compared with the traditional mechanical crushing scheme, owing to the introduction of laser equipment. This cost arises from the large area of PV modules to be processed in actual production, which may involve the simultaneous operation of multiple laser devices, leading to additional expenses related to equipment procurement and energy consumption. Furthermore, the laser-based scheme is not applicable to certain types of PV modules that contain regions without silicon cells (Fig. S12).
The laser automated scanning system is equipped with a Hongben HB-C1500 continuous laser, with the laser’s central wavelength being 1080 ± 10 nm. The laser output head is mounted on a servo motor-controlled guide rail and programmed via a computer to operate in the horizontal plane along a predefined trajectory at a set speed. The overall setup is shown in Fig. S1. Commercially retired bifacial PV modules used in the experiments are illustrated in Fig. S2. The heating stage is a self-constructed pure resistance heating platform, regulated via a transformer to enable precise temperature control (Fig. S3). The temperature of the entire PV strips is measured using an infrared (IR) sensor thermometer, while the temperature of the silicon cells is monitored via thermocouples (Fig. S4). For volatile organic compound (VOC) detection, the MiniRAE 3000 + portable handheld monitor (RAE Systems/Honeywell) with a sensor resolution of 0.1 ppm is employed (Fig. S5). Peeling force measurements are conducted using a DK-F200S tensile tester manufactured by Deka (Fig. S6).
The adjustable parameters of the laser include output power, frequency, and duty cycle. The output frequency (f) refers to the number of laser pulses emitted per unit time, as denoted by count n in Fig. S7a. The interval between two consecutive laser emissions is defined as the period (T), with each period lasting 1/n seconds. The duty cycle is defined as the proportion of time within one cycle during which the laser emits energy. For instance, a 50% duty cycle means the laser emits energy for half the duration of one cycle; a 70% duty cycle indicates energy emission for 70% of the cycle duration, as illustrated in Fig. S7b.
First, the junction box and frame were removed. Bifacial PV laminate was cut into 8 × 5 cm strips (length × width) for subsequent use, as shown in Fig. S8. These strip-shaped bifacial PV strips were heated to 190 ℃ using a heating stage; silicon cells were extracted and labeled as Sample (1) New PV strips were then subjected to laser treatment, with parameters set as follows: power ranging from 100 to 1200 W, duty cycle from 100% to 5%, and frequency from 200 to 2000 Hz. The laser scanning speed was set to 30 cm/s, while the rectangular beam spot had a length of 5 cm and a width of 0.4 mm. During scanning, an IR sensor electronic thermometer continuously monitored the temperature. Scanning was halted once the temperature of surface glass reached 190 ℃, followed by cooling. During scanning, thermocouple probe was inserted into the side of the PV strips and tightly attached to the surface of the silicon cells to monitor their temperature. After multiple scans, the ARCs partially degraded, allowing the light-exposed side of the silicon cells to be easily manually separated from EVA. Through repeated trials, the optimal laser parameters were determined as 1200 W power, 2000 Hz frequency, and 5% duty cycle. The opposite side of the silicon cells were re-irradiated under these parameters, enabling easy scraping of the cells. These cells was collected for analysis and labeled as Sample (2) The complete experimental process is illustrated in Fig. 2. Separated silicon cells were further processed by direct surface irradiation using the optimal laser parameters until the ARCs was completely removed. These cells were collected for characterization and labeled as Sample (3) All three samples were shown in Fig. S9. Experiments under each parameter set were performed in triplicate (n = 3), and the results obtained from the three PV strips under the same parameter set were consistent with each other.
TGA was conducted on both pristine EVA from PV modules and laser processed EVA using a thermal analyzer (TA Instruments, Q500, US). The temperature was ramped from room temperature to 550 °C at a heating rate of 10 °C/min, with the entire experiment performed under an air atmosphere at a flow rate of 50 mL/min; sample mass was 5 mg. The objectives were twofold: first, to determine the pyrolysis temperature of EVA, thereby preventing waste gas generation due to excessive temperatures during laser treatment; second, to compare the thermogravimetric curves of pristine and laser processed EVA to investigate changes in their properties.
FTIR spectroscopy (Thermo Fisher, Scientific Nicolet iS20, US) was performed on pristine EVA and laser processed EVA. Measurements were conducted in attenuated total reflection (ATR) mode over a wavenumber range of 600–4000 cm− 1. The aims were to investigate the effects of laser treatment on EVA.
A Raman spectrometer (HORIBA Scientific, LabRAM HR Evolution, Japan) was used to analyze pristine EVA and laser processed EVA. The excitation laser wavelength for the measurement was 785 nm, and the measurement wavenumber range was 50–4000 cm− 1. The aims were to investigate the effects of laser treatment on EVA.
Cold Field Emission Scanning Electron Microscope (S-4800 SEM, Hitachi, Japan). Electron Gun – Cold cathode field emission type. Accelerating voltage 15 kV, working distance 4 mm − 1.0 nm and accelerating voltage 1 kV, working distance 1.5–2.0 nm. EDS analysis at a point or over user-defined regions (Point&ID), between any two points (LineScan, TruLine, and QuantLine), EDS element mapping (LayerMap, AutoLayer and TruMap), and EDS phase mapping (AutoPhaseMap). The objective was to analyze the surface morphological changes and variations in elemental content of silicon cells following different levels of laser treatment.
X-ray Photoelectron Spectroscopy (Thermo ESCALAB 250XI XPS, Thermo Kalpha, US). The monochromated X-ray beam can be focused to spot sizes ranging from 900 μm to 200 μm. XPS is based on the photoelectric effect that can identify the elements that exist within a material (elemental composition) or are covering its surface, as well as their chemical state, and the overall electronic structure and density of the electronic states in the material. The elements analyzed are carbon and Si, aiming to examine the material changes on the surface of silicon cells.
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.
IRENA, IEA-PVPS. End of life management: Solar photovoltaic panels. https://www.irena.org/publications/2016/Jun/End-of-life-management-Solar-Photovoltaic-Panels (2016).
Zapf-Gottwick, R. et al. Leaching hazardous substances out of photovoltaic modules. Int. J. Adv. Appl. Phys. Res. 2(7) (2015).
Nover, J. et al. Long-term leaching of photovoltaic modules. Jpn. J. Appl. Phys. 56 (8S2), 08MD02 (2017).
Article Google Scholar
Nover, J., Zapf-Gottwick, R., Feifel, C., Koch, M. & Werner, J. H. Leaching via weak spots in photovoltaic modules. Energies 14 (3), 692 (2021).
Article CAS Google Scholar
Kwok, K. H., Savaget, P., Fukushige, S. & Halog, A. The necessity for end-of-life photovoltaic technology waste management policy: A systematic review. J. Clean. Prod. 461, 142497 (2024).
Article Google Scholar
Nain, P. & Anctil, A. End-of-Life Solar Photovoltaic Waste Management: A Comparison as Per European Union and United States Regulatory Approaches. Vol. 21. 200212 (Resources, Conservation & Recycling Advances, 2024).
Nowakowski, P. Urban Mining of e-Waste: Conversion of Waste to Wealth. Management of Electronic Waste: Resource Recovery, Technology and Regulation. 152–172 (2024).
Deng, B. et al. Urban mining by flash joule heating. Nat. Commun. 12 (1), 5794 (2021).
Article ADS CAS PubMed PubMed Central Google Scholar
Mahim, T. M., Rahim, A. H. M. A. & Rahman, M. M. Review of mono-and bifacial photovoltaic technologies: A comparative study. IEEE J. Photovolt. 14 (3), 375–396 (2024).
Article Google Scholar
Sun, X., Khan, M. R., Deline, C. & Alam, M. A. Optimization and performance of bifacial solar modules: A global perspective. Appl. Energy. 212, 1601–1610 (2018).
Article ADS Google Scholar
Yin, H. P. et al. Optical enhanced effects on the electrical performance and energy yield of bifacial PV modules. Sol. Energy. 217, 245–252 (2021).
Article ADS Google Scholar
Kumbaroğlu, G. S., Çamlibel, M. E. & Avcı, C. Techno-economic comparison of bifacial vs monofacial solar panels. Eng. Struct. Technol. 13 (1), 7–18 (2021).
Google Scholar
Fertig, F. et al. Economic feasibility of bifacial silicon solar cells. Prog. Photovoltaics Res. Appl. 24 (6), 800–817 (2016).
Article Google Scholar
Equipment, V. P. International Technology Roadmap for Photovoltaic (ITRPV). Results 2020. Vol. 12. 1–74 (2021).
Wang, X., Tian, X., Chen, X., Ren, L. & Geng, C. A review of end-of-life crystalline silicon solar photovoltaic panel recycling technology. Sol. Energy Mater. Sol. Cells. 248, 111976 (2022).
Article CAS Google Scholar
Wang, G., Liao, Q. & Xu, H. Anticipating future photovoltaic waste generation in china: navigating challenges and exploring prospective recycling solutions. Environ. Impact Assess. Rev. 106, 107516 (2024).
Article Google Scholar
Jadhav, N. B., Gajare, O., Zele, S., Gogate, N. & Joshi, A. Current status and challenges in silver recovery from end-of-life crystalline silicon solar photovoltaic panels. Sol. Energy. 283, 113027 (2024).
Article Google Scholar
Su, P., He, Y., Feng, Y., Wan, Q. & Li, T. Advancements in end-of-life crystalline silicon photovoltaic module recycling: current state and future prospects. Sol. Energy Mater. Sol. Cells. 277, 113109 (2024).
Article CAS Google Scholar
Sanathi, R., Banerjee, S. & Bhowmik, S. A technical review of crystalline silicon photovoltaic module recycling. Sol. Energy. 281, 112869 (2024).
Article CAS Google Scholar
Singh, R. & Mondal, P. Insights into the recycling of discarded solar panels: Challenges and future outlook. Sustain. Mater. Technol. e01481 (2025).
Huang, W. H., Shin, W. J., Wang, L., Sun, W. C. & Tao, M. Strategy and technology to recycle wafer-silicon solar modules. Sol. Energy. 144, 22–31 (2017).
Article ADS CAS Google Scholar
Tembo, P. M., Heninger, M. & Subramanian, V. An investigation of the recovery of silicon photovoltaic cells by application of an organic solvent method. ECS J. Solid State Sci. Technol. 10 (2), 025001 (2021).
Article ADS CAS Google Scholar
Sah, D., Saini, P. & Kumar, S. Recovery and analysis of polymeric layers from waste solar modules by chemical route. Sol. Energy. 244, 31–39 (2022).
Article ADS Google Scholar
Keerthivasan, T., Madhesh, R., Srinivasan, M. & Ramasamy, P. Photovoltaic recycling: enhancing silicon wafer recovery process from damaged solar panels. J. Mater. Sci.: Mater. Electron. 35 (12), 880 (2024).
CAS Google Scholar
Prasad, D. S., Sanjana, B., Kiran, D. S., Kumar, P. S. & Ratheesh, R. Process optimization studies of essential parameters in the organic solvent method for the recycling of waste crystalline silicon photovoltaic modules. Sol. Energy Mater. Sol. Cells. 245, 111850 (2022).
Article CAS Google Scholar
Kim, Y. & Lee, J. Dissolution of ethylene vinyl acetate in crystalline silicon PV modules using ultrasonic irradiation and organic solvent. Sol. Energy Mater. Sol. Cells. 98, 317–322 (2012).
Article CAS Google Scholar
Doi, T. et al. Experimental study on PV module recycling with organic solvent method. Sol. Energy Mater. Sol. Cells. 67 (1–4), 397–403 (2001).
Article CAS Google Scholar
Wang, C., Lu, J., Qin, B., Zhu, J. & Ruan, J. Decapsulating waste photovoltaic laminated modules by the combination treatment of thermal field and the solvent of the N-methyl-2-pyrrolidone. Waste Manage. 191, 182–190 (2025).
Article CAS Google Scholar
Lu, J., Wang, C., Zhu, J., Wu, Y. & Ruan, J. 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).
Article CAS Google Scholar
Lee, J., Duffy, N., Petesic, J., Witheridge, T. & Allen, J. Comparative assessment of solvent chemical delamination of end-of-life solar panels. Waste Manage. 190, 122–130 (2024).
Article CAS Google Scholar
Min, R. et al. Effective decapsulation method for photovoltaic modules: Limonene-induced EVA controlled swelling under sonication and debonding mechanism analysis. J. Clean. Prod. 450, 141917 (2024).
Article CAS Google Scholar
Yu, Y. et al. Green recycling of end-of-life photovoltaic modules via deep-eutectic solvents–Part B. Chem. Eng. J. 512, 162345 (2025).
Article CAS Google Scholar
Min, R. et al. A novel method for layer separation of photovoltaic modules by using green reagent EGDA. Sol. Energy. 253, 117–126 (2023).
Article ADS CAS Google Scholar
Xu, G. et al. Recover value materials from waste photovoltaic modules as secondary resource: layer separation by eco-friendly reagent DMC combined pyrolysis. Sol. Energy Mater. Sol. Cells. 279, 113282 (2025).
Article CAS Google Scholar
Li, K. et al. Recycling of solar cells from photovoltaic modules via an environmentally friendly and controllable swelling process by using dibasic ester. Clean Technol. Environ. Policy. 25 (7), 2203–2212 (2023).
Article CAS Google Scholar
Briand, A. et al. Deformation-induced delamination of photovoltaic modules by foaming ethylene-vinyl acetate with supercritical CO2. J. CO2 Utilization. 59, 101933 (2022).
Article CAS Google Scholar
Birtürk, A. & Celiktas, M. S. Subcritical water delamination: A promising path to efficient recycling of critical minerals. J. Clean. Prod. 469, 143147 (2024).
Article Google Scholar
Granata, G., Pagnanelli, F., Moscardini, E., Havlik, T. & Toro, L. J. Recycling of photovoltaic panels by physical operations. Sol. Energy Mater. Sol. Cells. 123, 239–248 (2014).
Article CAS Google Scholar
Pagnanelli, F. et al. Physical and chemical treatment of end of life panels: An integrated automatic approach viable for different photovoltaic technologies. Waste Manage. 59, 422–431 (2017).
Article CAS Google Scholar
Tokoro, C., Nishi, M. & Tsunazawa, Y. Selective grinding of glass to remove resin for silicon-based photovoltaic panel recycling. Adv. Powder Technol. 32 (3), 841–849 (2021).
Article CAS Google Scholar
Song, B. P. et al. Recycling experimental investigation on end of life photovoltaic panels by application of high voltage fragmentation. Waste Manage. 101, 180–187 (2020).
Article CAS Google Scholar
Akimoto, Y., Iizuka, A. & Shibata, E. High-voltage pulse crushing and physical separation of polycrystalline silicon photovoltaic panels. Miner. Eng. 125, 1–9 (2018).
Article CAS Google Scholar
Zhao, P. et al. A novel and efficient method for resources recycling in waste photovoltaic panels: high voltage pulse crushing. J. Clean. Prod. 257, 120442 (2020).
Article CAS Google Scholar
Dias, P. R. et al. High yield, low cost, environmentally friendly process to recycle silicon solar panels: Technical, economic and environmental feasibility assessment. Renew. Sustain. Energy Rev. 169, 112900 (2022).
Article CAS Google Scholar
de Souza, R. A. & Veit, H. M. Study of electrostatic separation to concentrate silver, aluminum, and silicon from solar panel scraps. Circular Econ. 2 (1), 100027 (2023).
Article Google Scholar
Fiandra, V., Sannino, L. & Andreozzi, C. Photovoltaic waste as source of valuable materials: A new recovery mechanical approach. J. Clean. Prod. 385, 135702 (2023).
Article CAS Google Scholar
Preet, S. & Smith, S. T. A comprehensive review on the recycling technology of silicon based photovoltaic solar panels: Challenges and future outlook. J. Clean. Prod. 448, 141661 (2024).
Article CAS Google Scholar
Maghraby, Y. R., Ibrahim, A. H., Tayel, A., Azzazy, H. M. E. S. & Shoeib, T. Towards sustainability via recycling solar photovoltaic panels, A review. Sol. Energy. 285, 113085 (2025).
Article CAS Google Scholar
Wang, J., Feng, Y. & He, Y. The research progress on recycling and resource utilization of waste crystalline silicon photovoltaic modules. Sol. Energy Mater. Sol. Cells. 270, 112804 (2024).
Article CAS Google Scholar
Trivedi, H. K., Meshram, A. & Gupta, R. Recycling of photovoltaic modules for recovery and repurposing of materials. J. Environ. Chem. Eng. 11 (2), 109501 (2023).
Article CAS Google Scholar
Li, X. et al. Back EVA recycling from c-Si photovoltaic module without damaging solar cell via laser irradiation followed by mechanical peeling. Waste Manage. 137, 312–318 (2022).
Article CAS Google Scholar
Anwar, T. B., Hanson, K. M., Lam, K. & Bardeen, C. J. Using nanosecond laser pulses to debond the glass-EVA layer from silicon photovoltaic modules. Waste Manage. 187, 275–284 (2024).
Article Google Scholar
Deubener, J., Helsch, G., Moiseev, A. & Bornhöft, H. Glasses for solar energy conversion systems. J. Eur. Ceram. Soc. 29 (7), 1203–1210 (2009).
Article CAS Google Scholar
Song, H. J., Lee, D., Kim, C. & Na, J. H. Improved performance of bifacial photovoltaic modules with low-temperature processed textured rear reflector. Appl. Sci. 14 (19), 8718 (2024).
Article CAS Google Scholar
Zhu, J., Montiel-Chicharro, D., Betts, T. & Gottschalg, R. Development of Adhesive and Cohesive Failures in EVA-Backsheet Structures in Environmental Testing (2016).
Pern, F. J. & Czanderna, A. W. Characterization of ethylene vinyl acetate (EVA) encapsulant: Effects of thermal processing and weathering degradation on its discoloration. Sol. Energy Mater. Sol. Cells. 25 (1–2), 3–23 (1992).
Article CAS Google Scholar
Allen, N. S., Edge, M., Rodriguez, M., Liauw, C. M. & Fontan, E. Aspects of the thermal oxidation, yellowing and stabilisation of ethylene vinyl acetate copolymer. Polym. Degrad. Stab. 71 (1), 1–14 (2000).
Article Google Scholar
Coyne, E. et al. STEM (scanning transmission electron microscopy) analysis of femtosecond laser pulse induced damage to bulk silicon. Appl. Phys. A. 81 (2), 371–378 (2005).
Article ADS CAS Google Scholar
Kumar, R., Mavi, H. S. & Shukla, A. K. Macro and microsurface morphology reconstructions during laser-induced etching of silicon. Micron 39 (3), 287–293 (2008).
Article CAS PubMed Google Scholar
Ulmeanu, M., Jipa, F., Radu, C., Enculescu, M. & Zamfirescu, M. Large scale microstructuring on silicon surface in air and liquid by femtosecond laser pulses. Appl. Surf. Sci. 258 (23), 9314–9317 (2012).
Article ADS CAS Google Scholar
Chen, F. et al. Short process recovery of silver and purification mechanism of crystalline silicon deep etching from end-of-life photovoltaic cells. Chem. Eng. J. 510, 161651 (2025).
Article CAS Google Scholar
Chastain, J. & King, R. C. Handbook of X-ray Photoelectron Spectroscopy (Perkin-Elmer Corporation, 1992).
Pern, F. J. Pv module encapsulation–materials, process, and reliability. In 16th Workshop on Crystalline Silicon Solar Cells and Modules: Materials and Processes. 111 (2006).
Yang, J. et al. A fluorine-restrained pyrolysis process for sustainable photovoltaic modules recycling: Two-stage decomposition of EVA and fluorine-containing backsheets. Resour. Conserv. Recycl. 225, 108611 (2026).
Article CAS Google Scholar
Peike, C., Kaltenbach, T., Weiß, K. A. & Koehl, M. Non-destructive degradation analysis of encapsulants in PV modules by Raman spectroscopy. Sol. Energy Mater. Sol. Cells. 95 (7), 1686–1693 (2011).
Article CAS Google Scholar
Ren, Y. et al. Two-dimensional Fourier transform Raman correlation spectroscopy study of composition-induced structural changes in a series of ethylene/vinyl acetate copolymers. J. Phys. Chem. B. 103 (31), 6475–6483 (1999).
Article CAS Google Scholar
Hirschl, C. et al. In-line determination of the degree of crosslinking of ethylene vinyl acetate in PV modules by Raman spectroscopy. Sol. Energy Mater. Sol. Cells. 152, 10–20 (2016).
Article CAS Google Scholar
Hoffendahl, C. et al. Decomposition mechanism of fire retarded ethylene vinyl acetate elastomer (EVA) containing aluminum trihydroxide and melamine. Polym. Degrad. Stab. 113, 168–179 (2015).
Article CAS Google Scholar
Pang, S. et al. Enhanced separation of different layers in photovoltaic panel by microwave field. Sol. Energy Mater. Sol. Cells. 230, 111213 (2021).
Article CAS Google Scholar
Gao, S. et al. Recycling of silicon solar panels through a salt-etching approach. Nat. Sustain. 7 (7), 920–930 (2024).
Article Google Scholar
Download references
This work was financially support from the National Natural Science Foundation of China (no.52270130) and the Science and Technology Committee Foundation of Shanghai (23DZ1201503).
School of Resources and Environmental Engineering, Shanghai Polytechnic University, Shanghai, 201209, China
Chenglong Zhang, Zhengzhong Zhao, Ruixue Wang & Xiaonuan Wang
School of Environmental Science and Engineering, Tongji University, Shanghai, 200092, China
Youcai Zhao
PubMed Google Scholar
PubMed Google Scholar
PubMed Google Scholar
PubMed Google Scholar
PubMed Google Scholar
Xiaonuan Wang: Writing – review & editing, visualization; Zhengzhong Zhao: Methodology, investigation, writing – Original Draft; Ruixue Wang: Project administration, formal analysis; Youcai Zhao: resources; Chenglong Zhang: Conceptualization, supervision, funding acquisition;
Correspondence to Xiaonuan Wang.
The authors declare no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Below is the link to the electronic supplementary material.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
Reprints and permissions
Zhang, C., Zhao, Z., Wang, R. et al. Separate silicon cells from end-of-life bifacial glass photovoltaic modules using continuous lasers. Sci Rep 16, 4986 (2026). https://doi.org/10.1038/s41598-026-35277-z
Download citation
Received:
Accepted:
Published:
Version of record:
DOI: https://doi.org/10.1038/s41598-026-35277-z
Anyone you share the following link with will be able to read this content:
Sorry, a shareable link is not currently available for this article.
Provided by the Springer Nature SharedIt content-sharing initiative
Advertisement
Scientific Reports (Sci Rep)
ISSN 2045-2322 (online)
© 2026 Springer Nature Limited
Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.
