From Waste to Value: A Comprehensive Review of Perovskite Solar Cell Recycling Technologies

Abstract

The rapid progress of perovskite solar cells (PSCs) has established them as a groundbreaking technology for sustainable energy. However, the sustainability of their lifecycle is still hindered by challenges related to material toxicity and end-of-life management. This review comprehensively assesses emerging recycling technologies, with a particular focus on their effectiveness in recovering perovskite compounds, transparent conductive oxides, and metallic contacts. Mechanical separation, solvent-based dissolution, thermal decomposition, and hybrid methods are compared in terms of recovery rates, purity levels, energy consumption, and scalability. Current challenges, such as the generation of secondary waste, the instability of recovered perovskites, and economic barriers, are critically analyzed alongside emerging solutions, including the use of non-toxic solvents, vacuum-assisted recovery, and the integration of closed-loop manufacturing. By evaluating lifecycle impacts and cost–benefit trade-offs, this work outlines pathways for transforming PSC waste into high-value secondary resources, thereby promoting both environmental sustainability and industrial competitiveness.

1. Introduction

As global energy demand continues to surge, the imperative for developing and harnessing clean and sustainable energy sources has grown ever more pressing [1,2,3,4,5,6,7]. Solar energy, as a renewable resource, has garnered widespread attention. In particular, perovskite solar cells (PSCs) have emerged as one of the most promising technologies [8,9,10,11,12,13,14], achieving remarkable breakthroughs in power conversion efficiency (PCE) [15]. This significant progress can be attributed to their exceptional optoelectronic properties, cost-effective fabrication processes [16], and tunable material compositions. However, PSCs still confront several challenges, such as the diversity and heterogeneity of their constituent materials, environmental pollution stemming from lead contamination [17,18,19,20,21,22], and the inadequate recovery and recycling of key materials [20]. These issues not only impede the path to commercialization but also have profound negative impacts on the environment and resource utilization. Therefore, it is of utmost urgency for the community to pay greater attention to the interplay between environmental sustainability, materials recycling, and cost reduction, which is vital for enhancing the overall sustainability of PSCs and facilitating their large-scale deployment [18,23,24,25].

To ensure the long-term sustainable development of PSCs, establishing efficient closed-loop recycling processes and achieving a circular economy for materials have emerged as critical research priorities [26,27,28]. However, the recycling and utilization of PSCs still confront several challenges. Firstly, the leakage of toxic metals such as lead, along with the inevitable use of hazardous solvents during material synthesis and device recycling, may pose risks to the environment and human health [18,19,20]. Secondly, existing recycling processes are immature and inefficient, and it is difficult to guarantee the quality and purity of recovered materials, thereby limiting their effective integration into circular economy pathways [29,30,31]. Thirdly, the lack of systematic life-cycle assessment (LCA) and techno-economic analysis (TEA) means that the feasibility, cost-effectiveness, and environmental benefits of recycling and reuse remain inadequately validated [32,33]. Importantly, PSC recycling should not be merely regarded as waste treatment; rather, it serves as a crucial link connecting environmental protection, resource utilization, and industrial upgrading [24,25]. This perspective not only addresses toxic pollution and resource wastage during industrialization but also provides essential technical support for the large-scale deployment of PSCs. Therefore, it is imperative for both academia and industry to focus on the environmental life cycle performance of PSCs and to develop green solvents and innovative recycling strategies that can accelerate the commercial implementation of photovoltaic (PV) technologies [34,35,36,37,38].

Among the approaches for managing end-of-life (EoL) PSC modules, traditional strategies primarily consist of refurbishment and reuse, secure landfilling, and high-temperature thermal treatment. However, in comparison with these conventional methods, recycling and circular economy strategies are more in line with the core requirements of PSCs for sustainable development and commercialization [14,17,18,20]. Currently, the key recycling routes encompass aqueous recycling, layer-by-layer extraction and regeneration, physical separation combined with chemical purification, and the design of detachable cells or modules [27,29,30,31,35,36,37]. These methods offer a diverse range of options for reducing the life cycle environmental impact of PSCs, improving resource efficiency, and supporting sustainable commercialization. Specifically, aqueous recycling enables the recovery and remanufacturing of entire EoL modules, often achieving high material purity and favorable refabrication performance [30,31]. Additionally, layer-by-layer extraction and regeneration allow for the separation, purification, and reuse of key functional materials, thereby realizing the circular use of high-value components [35,36,37,39]. Furthermore, physical separation coupled with chemical purification or ion exchange routes enables the effective recovery of multicomponent materials through mechanical dismantling. Finally, detachable cell designs adopt modular architectures that enhance the dismantling efficiency of retired devices and reduce recycling costs [36,40,41,42]. Nevertheless, despite recent progress, most studies still focus on single-component recovery and lack systematic, modular recycling schemes for complete PSC modules. There remain persistent challenges in terms of material purity, impurity control, and restoration of device efficiency and stability. Moreover, from the perspectives of TEA, LCA, and industrial deployment, current research has not yet achieved closed-loop validation and lacks comprehensive data on module recycling, material reprocessing, device remanufacturing, and long-term operation [18,26,32,33,37,43,44,45,46,47]. Therefore, to promote the sustainable development of PSCs, a comprehensive review of PSC recycling and cyclic utilization strategies is urgently required (Figure 1) [27,48,49,50,51,52].

Figure 1. Overview of sustainable recycling and cyclic utilization strategies for PSCs.

Herein, this review conducts a systematic and comprehensive summary of the recycling and reuse strategies for the various functional layers in PSCs. Initially, we delve into the recycling technologies specifically tailored for the perovskite absorber layer. Subsequently, we provide an overview of the recovery methods applicable to charge transport layers. Next, we conduct an in-depth analysis of the recycling routes for electrodes and glass substrates. Then, through specific case studies, we focus on the synergistic interplay between TEA and LCA, utilizing the LCA perspective to shed light on the environmental benefits of PV technologies. Afterwards, we evaluate solvent use and management, with an emphasis on integrated strategies centered around green solvent systems and solvent recovery efficiency. Finally, we offer a forward-looking outlook on sustainable circular technologies for perovskite PV modules and propose innovative solutions to propel the green recycling and commercialization of PSCs.

2. Recycling and Cyclic Utilization of Perovskite Absorber Layer

The metal halide perovskite absorber layer serves as the photoactive core of PSCs, directly governing light absorption and charge generation [53]. This thin film typically follows the ABX₃ structure, where “A” represents organic or inorganic monovalent cations, “B” is a divalent metal cation, and “X” denotes halide anions. The specific combination of these elements critically determines the optoelectronic properties, film quality, and, ultimately, the PCE and stability of PSCs. Figure 2 summarizes the common material systems used in PSCs [17].

Figure 2. Commonly used materials for each component of PSCs.

However, the pursuit of high efficiency in PSCs is intrinsically linked to a significant environmental challenge: the pervasive use of lead. A 1 GW perovskite PV installation could contain approximately 3.5 tons of lead [54], which, due to the ionic crystal nature of perovskites, can readily leach into the environment, posing risks to soil and water contamination, as well as neurotoxicity [55,56,57,58]. Figure 3 illustrates the in vivo distribution and multi-organ toxicity of lead [59,60]. This issue has raised the fundamental question: how can we reconcile the high performance of PSCs with environmental responsibility? Early strategies, such as the development of lead-free perovskites or physical encapsulation, have encountered limitations in efficiency and device stability [10,61,62,63]. Consequently, the focus has shifted toward direct lead recycling from EoL devices, a strategy that aligns with circular economy principles and offers the dual benefits of mitigating toxicity and recovering valuable materials [64,65,66,67]. The following sections critically evaluate the evolution of absorber-layer recycling technologies, emphasizing their methodological advances, limitations, and trajectories toward scalable circularity.

Figure 3. In vivo distribution and multi–organ toxicity of lead. (a) Schematic overview of absorption, distribution, and excretion of Pb compounds in the human body. Reproduced with permission from Ref. [59]. Copyright 2016, Springer Nature. (b) The toxic bio–hazards of Pb on human organs and systems. Reproduced with permission from Ref. [60]. Copyright 2021, Springer Nature.

2.1. Adsorption–Desorption Method

The adsorption–desorption method has emerged as a highly adaptable approach for lead recovery, using engineered adsorbents to selectively capture Pb²⁺ while often enabling concurrent iodide recovery. Ren et al. [31] demonstrated nearly complete Pb immobilization (~100%) and >95% iodide recovery using commercial zeolite (Figure 4a), yielding recycled PbI₂ that enabled PSCs with a PCE of 21.0%—comparable to commercial precursors.

Figure 4. (a) Schematic illustration of zeolite−enabled cyclic recovery of both iodide and lead from degraded PSCs: (I) peeling off the metal electrodes using adhesive tape; (II) removal of the organic HTL by immersion in ethyl acetate; (III) decomposition of the perovskite layer via brief immersion in water; and (IV) removal of PbI₂ by immersion in water. Reproduction with permission from Ref. [31]. Copyright 2021, American Chemical Society. (b) Schematic illustration of Pb²⁺ absorption mechanisms: ion exchange in HAP and dissolution followed by recrystallization in WH. (c) Energy−dispersive X−ray spectroscopy (EDS) results for HPy proving the persistence of Ca²⁺ in Pb−HAP. (d) EDS results for Pb−WH proving the absence of Ca²⁺ and Mg²⁺ cations. Reproduction with permission from Ref. [29]. Copyright 2022, Wiley-VCH.

However, the performance of adsorption systems is governed by the purity of the recovered PbI₂, which many early adsorbents failed to achieve. Hydroxyapatite (HAP), for example, introduces Ca²⁺ impurities that impair subsequent perovskite film quality. Hong et al. [29] addressed this limitation by designing biomimetic whitlockite (WH) nanoparticles, which operate via a dissolution–recrystallization mechanism rather than ion exchange, achieving a record Pb²⁺ adsorption capacity of 2339 mg g⁻¹ and producing high-purity PbI₂ suitable for high-efficiency PSCs (Figure 4b–d). Notably, Xie et al. [68] developed a reactant-recycling strategy to extract Pb from used lead–acid batteries, followed by wet-chemical conversion into high-crystallinity PbI₂ with 99.9% purity.

More recent advances have shifted from post-leakage remediation to device-integrated capture. As shown in Figure 5a, Luo et al. [69] embedded a bioinspired cage-trap (BCT) material within PSCs, enabling in situ Pb sequestration and subsequent closed-loop recovery. This approach achieved a 96.07% Pb recycling rate and yielded reconstructed PSCs with a PCE of 22.08%. Additionally, the BCT-encapsulated PSCs exhibited markedly enhanced long-term stability, maintaining over 92% of initial PCE after 1000 h under 50% relative humidity. Such embedded strategies mark an important transition toward proactive, system-level recycling architectures.

2.2. Electrochemical Deposition Method

In contrast to the adsorption–desorption method, electrochemical deposition techniques focus on the recovery of high-purity lead metal, often through molten-salt electrolysis. Wang et al. [70] employed a molten-salt electrolysis method at 450 °C, achieving a lead recovery rate of 97.65% while avoiding organic solvents. Although this method is effective in producing high-purity lead, its significant energy requirements and operational complexity present substantial barriers to large-scale implementation, particularly in terms of cost and energy efficiency.

An innovative alternative is the photoelectrochemical (PEC) system, which integrates real-time lead detection with remediation and recovery. Dang et al. [71] developed a PEC system that couples photo-oxidation with ion intercalation, enabling the effective immobilization and removal of lead ions. This method offers the potential for intelligent recycling platforms that simultaneously monitor environmental conditions and recover resources, advancing the concept of smart recycling. While promising, PEC systems are still in their early stages, and challenges remain related to scalability, energy efficiency, and long-term operational stability. The field must navigate a trade-off between high-purity electrochemical recovery and lower-energy PEC approaches, each with distinct economic and environmental implications.

2.3. Solvent Extraction Method

Solvent extraction remains widely used due to its operational simplicity and ability to process entire device stacks. However, solvent toxicity and waste generation necessitate greener alternatives. Schmidt et al. [72] demonstrated a solvent-free route by exploiting the temperature-dependent solubility of PbI₂ in hot water, enabling high-purity PbI₂ recovery and producing residues sufficiently Pb-depleted to qualify as non-hazardous waste (Figure 5b).

To further simplify the process, Hu et al. [73] achieved a Pb removal rate of 99.84% via aqueous leaching combined with ultrasound-assisted desorption. However, a critical gap in their study is the lack of validation of device performance using the recovered materials, which remains a common limitation in the literature. Beyond simple dissolution, Feng et al. [74] proposed a butylamine-coordination route that facilitates direct recrystallization of MAPbI₃, enabling a genuine closed-loop perovskite synthesis cycle. At the module level, Wang et al. [75] developed a comprehensive one-key-reset disassembly strategy based on methylamine solutions, enabling rapid separation of all layers. Devices reconstructed from recycled components maintained an average PCE of 20.08% after two cycles (only 2.8% degradation), demonstrating that full-device circularity is technically achievable, though process duration and solvent handling remain key bottlenecks.

Figure 5. (a) Schematic of sustainable Pb management in perovskite solar modules, including I Pb precipitation, II Pb adsorption, III Pb desorption and IV Pb recycle. Reproduction with permission from Ref. [69]. Copyright 2023, Springer Nature. (b) Scheme of the facile recycling process developed. The process includes mechanical pretreatment (cutting or shredding) (1), hot aqueous extraction (2), solid–liquid separation via filtration (3), PbI₂ precipitation by cooling (4) and recovery (5), PbI₂ can be reused in new perovskites, and the solid waste (W) is depleted of Pb. Reproduction with permission from Ref. [72]. Copyright 2023, Elsevier. (c) Device architectures and J–V curves of fresh and refabricated PSCs. Reproduction with permission from Ref. [76]. Copyright 2024, Elsevier. (d) Recycling procedure for PSCs: (I) removal of the Au electrode using adhesive tape; (II) removal of the HTM by immersion in chlorobenzene; (III) transformation of the perovskite into MAI and PbI₂ followed by extraction of MAI into water; (IV,V) removal of PbI₂ and TiO₂ using DMF; (VI) preparation of a new TiO₂ film; (VII) formation of the perovskite film on recycled FTO using recycled PbI₂; (VIII) preparation of the HTM layer; and (IX) evaporation of the Au top electrode. Reproduction with permission from Ref. [30]. Copyright 2016, American Chemical Society.

Figure 5. (a) Schematic of sustainable Pb management in perovskite solar modules, including I Pb precipitation, II Pb adsorption, III Pb desorption and IV Pb recycle. Reproduction with permission from Ref. [69]. Copyright 2023, Springer Nature. (b) Scheme of the facile recycling process developed. The process includes mechanical pretreatment (cutting or shredding) (1), hot aqueous extraction (2), solid–liquid separation via filtration (3), PbI₂ precipitation by cooling (4) and recovery (5), PbI₂ can be reused in new perovskites, and the solid waste (W) is depleted of Pb. Reproduction with permission from Ref. [72]. Copyright 2023, Elsevier. (c) Device architectures and J–V curves of fresh and refabricated PSCs. Reproduction with permission from Ref. [76]. Copyright 2024, Elsevier. (d) Recycling procedure for PSCs: (I) removal of the Au electrode using adhesive tape; (II) removal of the HTM by immersion in chlorobenzene; (III) transformation of the perovskite into MAI and PbI₂ followed by extraction of MAI into water; (IV,V) removal of PbI₂ and TiO₂ using DMF; (VI) preparation of a new TiO₂ film; (VII) formation of the perovskite film on recycled FTO using recycled PbI₂; (VIII) preparation of the HTM layer; and (IX) evaporation of the Au top electrode. Reproduction with permission from Ref. [30]. Copyright 2016, American Chemical Society.

2.4. Other Innovative Methods

A number of emerging strategies address niche recycling needs or specific degradation pathways. Chhillar et al. [67] demonstrated in situ regeneration of degraded PbI₂ films using MAI solution, with performance strongly dependent on the original film fabrication route. Similarly, Deng et al. [76] conducted systematic solvent screening and identified DMF as an optimal Pb solubilizer; using KI as a precipitant, they recovered PbI₂ while simultaneously introducing passivating K⁺ into regenerated films, enabling PSCs with a champion PCE of 22.78% (Figure 5c) and >90% PCE retention under high humidity for 1200 h. Beyond solution-based regeneration, thermal decomposition–assisted in situ recycling has also emerged as a distinctive pathway for absorber-layer recovery. Xu et al. [77] developed a controlled thermal decomposition strategy that converts CH₃NH₃PbI₃ films directly back into high-crystallinity PbI₂ scaffolds without introducing secondary pollutants.

Furthermore, layer-by-layer dissolution processes have been foundational in recovering multiple components, such as FTO substrates and PbI₂. Binek et al. [30] developed a process enabling the efficient recovery of these materials, proving that recycling can be economically and environmentally beneficial (Figure 5d). These methods not only support lead recovery but also establish the viability of recovering other valuable materials, such as conductive substrates, in a way that minimizes waste and reduces manufacturing costs. Meanwhile, to address the gap in research concerning the recycling of unused perovskite powders that have degraded under poor storage conditions, Ha et al. [78] introduced a customized approach termed the Perovskite Powder Recycling Method (PPRM). When applied to PSCs, PPRM enabled the reconstruction of degraded powders to achieve a performance comparable to or surpassing that of the original materials, yielding an impressive PCE of >18.9% in a large-area device with an active area of 2.21 cm².

As recycling technologies have diversified, their sustainability evaluation has expanded beyond laboratory-scale metrics such as recovery yield, precursor purity, and PCE retention to include TEA and LCA [79]. This reflects a shift from proof-of-concept validation to system-level appraisal of cost, energy demand, and environmental impact. For example, Oh et al. [80] employed combined TEA-LCA to benchmark three hydrometallurgical routes for PSCs—HAP/Fe adsorption, WAC-resin adsorption, and water extraction—and revealed a clear trade-off: WAC-resin adsorption was economically optimal, with a treatment cost of 54 USD m⁻² due to adsorbent reusability and low DMF consumption, whereas water extraction minimized global warming potential (GWP −0.04 kg CO₂-eq) by using water as the sole solvent. Such studies provide a quantitative basis for comparing recycling pathways and demonstrate that economic and environmental optima may diverge.

Yet the central challenge remains scaling: translating laboratory protocols into industrially viable, low-cost, and energy-efficient recycling workflows [19,81,82,83]. Future progress will depend on the development of trigger-responsive “smart” materials, device architectures designed for disassembly, and system-level life-cycle integration, ensuring that perovskite PVs achieve not only record efficiencies but also long-term sustainability aligned with global circular economy objectives.

3. Recycling and Cyclic Utilization of Charge Transport Layers

Having established the recycling paradigms for the photoactive perovskite absorber, the pursuit of a fully circular PSC economy necessitates the integration of charge transport layers (CTLs) into the resource recovery framework. CTLs, also referred to as charge-selective layers, comprise the hole transport layer (HTL) and electron transport layer (ETL), which selectively extract and transport photogenerated charges to establish the internal electric field of PSCs. Common hole transport materials (HTMs) include Spiro-OMeTAD, NiOₓ, PTAA, PEDOT: PSS, and self-assembled monolayers (SAMs), while frequently used electron transport materials (ETMs) encompass fullerenes, ZnO, PCBM, and SnO₂. However, compared with the substantial efforts devoted to recycling the light-absorbing layer, systematic studies on the separate recovery and recycling of CTLs remain scarce [84].

A marked research imbalance exists: most studies on CTL recycling focus on n-i-p (conventional) architectures, whereas p-i-n (inverted) structures remain underexplored. This disparity arises from intrinsic architectural and material factors. In n-i-p devices, a discrete and soluble small-molecule HTL such as Spiro-OMeTAD atop the perovskite facilitates the sequential layer-by-layer dissolution that underpins prevalent recycling methods [35,36,85]. In contrast, CTL recovery in p-i-n structures presents distinct and more complex challenges. The p-type CTL (e.g., PEDOT: PSS or NiOₓ) is typically deposited first onto the transparent conductive oxide (TCO) substrate, forming an ultrathin, strongly bonded substrate that is highly vulnerable to damage during the removal of the overlying perovskite and n-type CTL. Moreover, the organic CTLs common in inverted architectures are often highly sensitive to the solvents required to dissolve the perovskite, complicating intact recovery. Consequently, current p-i-n recycling efforts prioritize the recovery of lead and the valuable TCO substrate, often treating the underlying HTL as a sacrificial layer or attempting its co-recovery with the substrate [86]. Nonetheless, insights from conventional recycling can inform strategies for inverted structures. As research interest shifts toward p-i-n devices due to their advancing efficiency, stability, and suitability for tandem cells and large-area modules, dedicated efforts targeting CTL recycling in such architectures are expected to increase, ultimately addressing this critical knowledge gap.

This limited progress arises from several interrelated factors. First, the economic incentive is weak: compared with the perovskite absorber containing high-value lead and the expensive TCO substrates, many CTL materials have relatively low unit costs, reducing the economic attractiveness of their recycling. Second, the technical challenges are more complex. CTLs are typically present as ultra-thin films in intimate contact with adjacent layers, making high-purity, low-damage separation technically daunting. Third, CTL materials are highly diverse, ranging from organic small molecules and polymers to inorganic metal oxides, with markedly different chemical and physical properties that hinder the development of universal recycling strategies. Finally, research priorities have naturally focused on addressing the most pressing environmental toxicity and critical material scarcity issues (namely lead and indium). Collectively, these factors have relegated CTL recycling to a secondary priority, creating a critical gap in the lifecycle management of PSCs that must be addressed to avert future waste streams of functional materials. Nonetheless, achieving closed-loop recycling and sustainable development of complete PSCs devices necessarily requires effective resource management of CTLs as well. In this context, this chapter reviews recent progress and critically examines the current state of recycling for both HTMs and ETMs.

3.1. Recycling of Hole Transport Layers

Research on HTL recycling has predominantly focused on Spiro-OMeTAD, with most strategies relying on selective, layer-by-layer dissolution to recover both the organic HTL and other high-value components. Song et al. [35] proposed a facile closed-loop recycling strategy to recover and reuse the hole transport material and other retrievable components from obsolete PSCs by sequentially dissolving the Spiro-OMeTAD and perovskite layers with chlorobenzene (CB) and DMF, respectively (Figure 6a). PSCs refabricated from the recovered materials delivered an enhanced PCE of up to 23.41% together with an open-circuit voltage (Voc) of 1.17 V, outperforming control devices based on fresh materials (20.77%, 1.11 V). This work demonstrates that, when appropriately purified, recycled HTM and absorber precursors can not only preserve but even improve device performance, thereby providing a foundational, albeit simplified, proof-of-concept for material recovery. However, the economic and environmental costs of the chlorinated and aprotic solvents employed were not yet addressed. Similarly, Le Khac et al. [86] successfully recovered Spiro-OMeTAD from inverted PSCs via a one-step chemical treatment. They utilized aqua regia to dissolve the various layers and employed CB solvent for selective dissolution to extract Spiro-OMeTAD. The purity of the recovered material was confirmed via spectroscopic analysis, though the performance of devices fabricated with recycled components requires systematic evaluation.

Figure 6. (a) Conceptual schematic of the closed-loop recycling process for EoL PSCs. Reproduction with permission from Ref. [35]. Copyright 2025, Wiley-Blackwell. (b) Schematic diagram of closed-loop recycling process for perovskite solar devices. Reproduction with permission from Ref. [36]. Copyright 2024, Royal Society of Chemistry. (c) Schematic of the recycling process using a water-based solution. Three main additives (NaOAc, NaI, and H₃PO₂) are incorporated to address perovskite solubility, phase purity, and solution stability issues in the aqueous solution. Reproduction with permission from Ref. [24]. Copyright 2025, Springer Nature. (d) Schematic illustration of the recovery processes for three critical components of PSCs, namely SnO₂-coated ITO substrates, PbI₂, and Spiro-OMeTAD. The recycling and restoration processes are indicated by green arrows, while the refabrication processes are indicated by gray arrows. Reproduction with permission from Ref. [85]. Copyright 2025, Royal Society of Chemistry.

Building on this concept, Wu et al. [36] developed a layer-by-layer solvent extraction process for closed-loop recycling that quantitatively illustrates the recoverability of different components. Their protocol achieved high recovery rates: nearly 100% for ITO/SnO₂ substrates, 87% for MAPbI₃, and 66% for Spiro-OMeTAD (Figure 6b). The HTL was dissolved in CB and purified via column chromatography; although residual additives remained, devices employing recycled Spiro-OMeTAD reached a PCE of 17.3%, comparable to the 17.5% obtained with fresh HTL. The perovskite layer was dissolved in γ-butyrolactone (GBL) and recrystallized, while the reclaimed substrates required an additional SnO₂ coating to restore performance. A full device fabricated from all recycled components achieved a champion PCE of 17.1%, closely matching the 17.7% PCE of fresh devices. TEA and LCA demonstrated significant economic and environmental benefits. At the same time, the incomplete HTL recovery and additional SnO₂ deposition indicate that further process optimization is needed to minimize material losses and processing steps.

To alleviate solvent-related environmental burdens and improve overall sustainability, subsequent work has focused on green solvent systems. Valentina Larini et al. [85] developed a green-solvent-based recycling strategy for PSCs, achieving high recovery rates for critical components: nearly 100% for ITO/SnO₂ substrates, 99.1% for PbI₂, and 89% for Spiro-OMeTAD. The process involved sequential dissolution using ethyl acetate (EtOAc) to remove Spiro-OMeTAD and gold, followed by dissolution of the perovskite layer in deionized water and precipitation of PbI₂ with ethanol; the reclaimed ITO/SnO₂ substrates were then cleaned and treated with UV–ozone for reuse (Figure 6d). Devices fabricated from fully recycled materials retained 98.4% of their initial PCE, with an average PCE of 18.9%. Performance analysis indicated that a significant increase in fill factor (FF) effectively compensated for minor reductions in Voc and short-circuit current density (Jsc). LCA further demonstrated the environmental superiority of this approach over landfill disposal, with a 28% reduction in carbon footprint and improvements across other impact categories. Notably, the process remained advantageous through three recycling iterations, and a closed-loop system was established by distilling and reusing the solvents. Taken together, this work not only establishes a greener protocol but also underscores a critical paradigm shift: environmental compatibility in recycling can be synergistically achieved with high device performance, challenging the perceived trade-off between sustainability and efficiency.

Moving beyond solvent substitution, a more holistic approach aims to integrate HTL recycling into a comprehensive system that recovers all valuable device components. Xiao et al. [24] recently reported a low-cost, green, holistic recycling strategy for PV, utilizing an efficient aqueous solution enhanced by sodium acetate (NaOAc), sodium iodide (NaI), and hypophosphorous acid (H₃PO₂) to address key challenges in perovskite dissolution, phase purity, and solution stability (Figure 6c). This approach achieves near-complete recovery of all valuable components—including the perovskite absorber, Spiro-OMeTAD, metal electrodes, and conductive substrates—thereby enabling a closed-loop “cradle-to-cradle” material cycle. Devices fabricated after five rounds of degradation–recycling exhibited a PCE of 21.8%, with stability comparable to fresh devices under accelerated aging conditions (85 °C thermal stress and continuous light soaking). Combined LCA and TEA confirmed substantial environmental benefits, including a 96.6% reduction in resource depletion and a 68.8% decrease in human toxicity (cancer effects), along with a reduced levelized cost of electricity (LCOE). From a circular-economy perspective, this work suggests that HTL recycling is most effective when embedded in system-level strategies that simultaneously recover the absorber, CTLs, electrodes, and substrates, rather than being treated as an isolated unit operation. This work thus transcends incremental improvements and helps redefine the ultimate goal of PSC recycling: not as a series of disconnected steps, but as an integrated, holistic material metabolism process.

3.2. Recycling of Electron Transport Layers

While the recovery of organic HTLs demonstrates the feasibility of molecular-level recycling, the strategies for ETLs confront a distinct set of material and economic realities. In contrast to the HTL, which is typically dissolved and recovered as an independent organic layer [87], recycling strategies for the ETL follow distinct technical pathways. Rather than focusing on extracting and purifying the ETL material itself, most approaches aim to clean and regenerate the attached transparent conductive substrate, thus enabling reuse of the integrated “substrate/ETL” structure [88,89,90]. This strategy is adopted because common inorganic ETL materials are firmly bonded to the substrate and possess relatively low intrinsic value, whereas the TCO substrate supporting them is among the most costly components in the device. Consequently, ETL recycling is closely linked to substrate reuse and to preserving the structural and electronic integrity of the TCO/ETL stack over multiple cycles.

Direct recycling and reuse of TCO/ETL substrates have been demonstrated for both TiO₂- and SnO₂-based architectures. Zhu et al. [91] developed an efficient recycling and reuse method for FTO/TiO₂ substrates in carbon-based all-inorganic CsPbIBr₂ solar cells. The process involved immersing degraded devices in DMF to dissolve the CsPbIBr₂ and simultaneously detach the top carbon electrode, thereby exposing the intact FTO/TiO₂ substrate. The substrate was then cleaned via ultrasonication and sequential solvent rinsing to obtain a purified recycled substrate suitable for device refabrication. Ling et al. [92] further demonstrated that ETL design can be tailored for both rapid deposition and robust recyclability. They reported a rapid chemical bath deposition method for preparing Sb-doped SnO₂ ETLs using a concentrated Sn precursor stabilized with ethanol and incorporating Sb doping, which shortened the deposition time from several hours to just 5 min, showing strong potential for high-throughput fabrication. More importantly, the SnO₂ ETL substrates could be effectively recycled after removing the upper layers of degraded PSCs. Even after three consecutive rounds of recycling and device re-fabrication, the refurbished PSCs retained about 98% of their initial PCE, highlighting the robustness and reusability of the SnO₂ ETMs and underscoring the importance of durable ETL/substrate stacks for cyclic utilization.

In advancing ETL recycling, improving recovery efficiency must be accompanied by the development of environmentally friendly recycling processes. Larini et al. [93] made significant strides in this regard. Recognizing that the solvent DMF, commonly used for perovskite active-layer removal, is highly toxic and environmentally hazardous—posing a major obstacle to the greening and scalability of recycling technologies—the researchers adopted dimethyl sulfoxide (DMSO), a green solvent classified as non-hazardous, for the recovery of SnO₂/ITO substrates. New solar cells fabricated using the recycled substrates achieved a champion PCE of 22.6%, matching that of devices using pristine substrates. By demonstrating that high-performance ETL/substrate recycling can be realized using a low-toxicity solvent, this work provides an important step toward aligning ETL recycling with green chemistry principles and large-scale industrial implementation. Together with the high-value electrode and TCO recycling routes discussed in the next section, ETL/substrate recycling thus forms a key element of device-level circularity in PSCs.

4. Recycling and Cyclic Utilization of Electrodes

From a full life-cycle perspective, the recycling and cyclic utilization of electrode components in PSCs combine two critical attributes: high value and high environmental sensitivity. Precious metals and TCOs contribute disproportionately to both material cost and embodied energy, while their production relies on energy- and emission-intensive operations such as mining, high-temperature metallurgy, and vacuum deposition. If these components are simply landfilled or incinerated at EoL, critical metal resources are squandered and heavy-metal release may pose non-trivial ecological and health risks. Accordingly, in constructing a sustainable development pathway for perovskite PVs, electrode recycling must be treated not as a peripheral EoL option, but as a central design lever [94]. This chapter focuses on two major classes of electrode-related materials—metallic electrodes and TCO/glass substrates—and traces how their recycling has evolved from simple component recovery toward integrated, design-for-recycling concepts increasingly supported by TEA and LCA.

4.1. Recycling of Metal Materials

The back electrode of PSCs generally employs highly conductive metal films to provide ohmic contact and reflect unabsorbed light. In laboratory-scale devices, Au, Ag, Cu, Al and, in some cases, Ni are commonly used [95]. Among them, Au offers excellent efficiency and stability owing to its favorable work function alignment and chemical inertness, but its high price renders it a major cost driver: the Au electrode alone can account for nearly 20% of PSC cost [96], and NREL analyses indicate that large-scale use of Au or Ag would make modules economically uncompetitive [97]. This economic pressure has been the primary driver for developing metal-electrode recycling strategies.

A baseline route to metal recovery relies on selective dissolution of the perovskite and charge-transport layers to liberate intact metal films. Kim et al. [98] established a widely cited protocol using polar aprotic solvents (DMF, GBL, DMSO), where a brief (~30 s) immersion dissolves CH₃NH₃PbI₃ and spiro-MeOTAD, enabling recovery of the Au electrode and mesoporous TiO₂-coated TCO glass (Figure 7a). The reclaimed substrates, after rinsing and annealing at 500 °C, were reused to fabricate new devices that maintained ~15% PCE over 10 recycling cycles with negligible losses in Voc, Jsc or FF (Figure 7b). The recovered Au contained only 0.41% Pb, meeting industrial refining standards. Similarly, layer-by-layer dissolution has yielded Au without detectable contamination by foreign metals [27]. These studies establish that high-purity metal recovery is technically straightforward; however, they also typify a “recover-and-refine” model that still requires energy-intensive purification and re-deposition and depends on toxic solvents, thereby limiting net environmental benefit when full process chains are considered.

Figure 7. (a) Schematic illustration of the detailed process for recycling PSCs via selective dissolution. During immersion in a polar aprotic solvent, the deposited metal electrode peels away from the device, leaving the clear ETL-coated substrate behind, while the HTL and perovskite layers dissolve. (b) Normalized trends in Jsc, Voc, FF, and efficiency of MALI-based PSCs fabricated on substrates recycled for 1st, 2nd, 3rd, 4th, 5th, and 10th times, compared with those of the MALI-PSC fabricated on a newly prepared substrate. Reproduction with permission from Ref. [98]. Copyright 2016, Springer Nature. (c) J–V curves of evaporated and nanoporous Au/PSCs; the evaporated and nanoporous Au electrodes are reused through two recycling processes. Reproduction with permission from Ref. [96]. Copyright 2020, Wiley-VCH.

To move beyond this essentially linear paradigm, Yang et al. [96] proposed a design-for-recycling strategy focused on electrode architecture. Instead of conventional evaporated Au, they developed a freestanding nanoporous Au (nano-Au) film that can be mechanically delaminated and directly reused. In their process, degraded devices and a fresh membrane support are co-immersed in acetone; dissolution of the perovskite and HTL releases the nano-Au film, which floats and is captured by the membrane for reuse (Figure 7c). The high PCE is primarily attributed to the nano-Au electrode’s high specific surface area (133.0 m² g⁻¹) and its crack-inhibiting porous architecture, which ensures intimate contact with the HTL and preserves electrical continuity during repeated recycling. In contrast, the evaporated Au electrode, with a much lower specific surface area of only 1.7 m² g⁻¹, is prone to folding and tearing during recycling, leading to degraded interfacial contact and increased recombination resistance. PSCs incorporating the recycled nano-Au electrode achieved 16.5% PCE—markedly higher than devices using recycled evaporated Au—and retained high performance over 12 recycling cycles, with only slight degradation mainly attributed to reduced active-area coverage and residual chemicals affecting Jsc and FF (Figure 8a–d). This approach substantially reduces Au consumption, energy use and process complexity by preserving the electrode as a functional component rather than degrading it to bulk metal.

Figure 8. (a) Schematic flow of the fabrication and restoration processes for the nanoporous Au electrode in PSCs. (b) Photographs of the nano-Au film recycled six times in PSCs. (c) Changes in PV parameters in the nano-Au/PSCs during 12 recycling iterations. (d) Noble Au consumption in evaporated and nanoporous Au films as the number of recycling iterations increases. Reproduction with permission from Ref. [96]. Copyright 2020, Wiley-VCH.

Viewed together, current studies on metal-electrode recycling reveal a clear trajectory: from solvent-based separation aimed at recovering metal mass to architected electrodes designed for facile disassembly and direct reuse. At the same time, they expose unresolved challenges. First, most strategies still rely on noble metals, so resource constraints and price volatility are not fundamentally alleviated. Second, the mechanical robustness and contact reliability of freestanding or nano-structured electrodes under real operating and handling conditions remain to be demonstrated at module scale. Third, the solvent and energy footprints of the overall recycling chain must be quantified rigorously, especially when polar aprotic solvents are involved. Future work should therefore extend design-for-recycling principles to earth-abundant metals (Cu, Ni, Al), explore solvent-free or green-solvent release methods, and validate recycling schemes under realistic field-aged module conditions, ensuring that electrode circularity becomes a robust pillar of PSCs sustainability rather than an elegant but isolated laboratory demonstration.

At the end of this section, a dedicated discussion on the recycling and circular utilization of carbon-electrode PSCs (C-PSCs) is included, along with a brief perspective. Most laboratory-scale PSCs still employ Ag electrodes; however, the ultrathin Ag layer is difficult to extract, and current hydrometallurgical or electrochemical recovery routes may incur costs exceeding the value of the reclaimed Ag [18,99]. Although Cu and Al reduce manufacturing costs, their inferior chemical stability often necessitates additional encapsulation, which can complicate EoL recycling [100]. Thus, for metal-based PSCs, future work should prioritize optimizing metal-separation techniques to enhance recycling feasibility. For large-area manufacturing compatible with industrial scale-up, replacing noble-metal electrodes with carbon is emerging as a promising strategy to reduce cost and improve process compatibility [101]. Specifically, the total cost of C-PSCs can be reduced from $86.49 (n–i–p) and $81.31 (p–i–n) to $41.16, underscoring a decisive cost advantage [102]. Carbon electrodes are typically deposited via energy-efficient slot-die coating, which is amenable to roll-to-roll processing, in contrast to the energy-intensive physical vapor deposition required for Au/Ag electrodes [103,104]. Consequently, carbon electrodes exhibit remarkable economic viability and industrial potential, considering both their material cost and manufacturing energy consumption. Although systematic data on electrode recycling for C-PSCs remain scarce, it is reasonable to anticipate that recycling may adopt a layer-delamination approach. Given the chemical stability of carbon and its low reactivity with common solvents, separating the carbon layer could be more straightforward than delaminating metal films, and the recovered carbon electrode might be directly reusable after cleaning.

4.2. Recycling of TCO/Glass Substrates

The front TCO/glass substrate is arguably the most critical electrode-related component from both economic and environmental standpoints. ITO, FTO and indium–zinc oxide (IZO) coatings on glass typically account for 30–50% of PSC material cost and more than half of the module-level GWP [33,105,106]. Their bulky, high-value nature and energy-intensive production make them prime targets for cyclic utilization. At the same time, the perovskite absorber introduces Pb-based toxicity concerns, creating a strong rationale to couple TCO/glass recovery with closed-loop Pb management.

Early studies established the intrinsic robustness and reusability of TCO/glass substrates bearing metal-oxide scaffolds. Kadro et al. [27] demonstrated that sequential selective dissolution of organic and perovskite layers in regular n–i–p devices allowed complete recovery of FTO/glass with its mesoporous TiO₂ layer. PSCs remanufactured on these recycled FTO/mp-TiO₂ substrates exhibited PCEs comparable to those of freshly fabricated devices, providing a compelling proof-of-concept for substrate recyclability. Huang et al. [94] then optimized a low-temperature solution process—combining organic-solvent washing, ultrasonic cleaning and UV–ozone treatment—to regenerate glass/FTO/TiO₂ substrates from degraded devices (Figure 9a). After two recycling cycles, planar and mesoporous PSCs reconstructed on recycled substrates retained peak PCEs of 11.87% and 11.03%, respectively. Extending to carbon-based all-inorganic devices, Zhu et al. [91] separated FTO/TiO₂ substrates from degraded CsPbIBr₂ PSCs by DMF soaking followed by ultrasonication (Figure 9b); structural characterization showed that the recycled substrates had clean surfaces and crystal structures essentially identical to pristine ones. Collectively, these works show that oxide/TCO stacks can withstand multiple recycling cycles without catastrophic performance loss, provided that cleaning protocols are properly tuned.

Figure 9. (a) Schematic illustration of the process for fabricating efficient PSCs by recycling glass/FTO/TiO₂ substrates from degraded devices. Reproduction with permission from Ref. [94]. Copyright 2017, American Chemical Society. (b) Illustration of the main steps to recycle the FTO/TiO₂ substrate from a degraded carbon-based, all-inorganic CsPbIBr₂ PSC for re-fabricating a new one, along with XRD patterns of the same samples and a bare FTO glass sheet. Reproduction with permission from Ref. [91]. Copyright 2020, American Chemical Society. (c) Roadmap for the recycling of perovskite solar modules. Reproduction with permission from Ref. [54]. Copyright 2021, Springer Nature. (d) Proposed remanufacturing route for carbon-based perovskite PV modules, based on a thermally assisted mechanochemical method to separate the back-glass, encapsulants, degraded perovskite, and carbon layers. This allows the reuse of the remaining layers and significantly reduces the GWP of a perovskite PV module. Reproduction with permission from Ref. [26]. Copyright 2024, American Chemical Society.

A major conceptual step has been the integration of substrate recycling with closed-loop Pb recovery. Chen et al. [54] developed a holistic module-level process in which thermal delamination non-destructively separated glass–glass encapsulated modules, enabling direct recovery of intact TCO/glass and back cover glass. The perovskite absorber was treated with a carboxylic-acid cation-exchange resin to adsorb Pb²⁺, followed by acid washing and precipitation to regenerate high-purity PbI₂ (Figure 9c). The overall Pb recovery efficiency reached 99.2%; solar cells fabricated with the recycled PbI₂ achieved a median PCE of 20.4%, closely matching the 21.0% efficiency of devices based on fresh high-purity PbI₂. Deng et al. [107] adopted a related dissolution–reprecipitation strategy: degraded perovskite films were dissolved in a DMF:DMSO (9:1) mixture and supplemented with MAI to form a regenerated precursor, while the same solvent system was used to clean and recycle the ITO/glass substrate. Carbon-based PSCs prepared from fresh and recycled materials reached champion PCEs of 16.63% and 15.30%, respectively. Zhang et al. [108] further demonstrated a dissolution–precipitation protocol tailored to carbon-based devices, achieving efficient Pb recovery while enabling direct reuse of FTO/c-TiO₂/m-TiO₂ substrates. These studies collectively demonstrate that TCO/glass substrate reuse and Pb closed-loop cycling are technically compatible and can be realized without prohibitive performance penalties.

System-level analyses have clarified why TCO/glass recycling is so impactful and how process design should respond. Bogachuk et al. [26] used LCA to show that the glass substrate—the core of the TCO/glass assembly—dominates the GWP of commercial perovskite modules. Conventional “bulk glass recycling” via crushing and remelting offers only modest benefit due to high energy demand. To address this, they proposed a thermally assisted mechanochemical remanufacturing route that dismantles glass–glass modules, removes degraded perovskite and electrodes, and directly reuses the glass and metal-oxide layers (Figure 9d), reducing the GWP of remanufactured devices by 24–33%. Complementary TEA by Xiao et al. [109] revealed that FTO glass contributes 58–73% of total PSC material cost, severely limiting economic viability if discarded at end of life. A key bottleneck is the strong, often irreversible adhesion of conventional metal-oxide ETMs (TiO₂, SnO₂) to TCO surfaces. Xiao et al. addressed this by introducing soluble PbSO₄ nanoparticles with surface metallicity as an ETL. Devices using PbSO₄ ETMs achieved high efficiencies (24.1%, 0.1 cm²) and certified module performance (17.9%, 204.9 cm²), while the PbSO₄ layer could be selectively dissolved in ethanolamine, allowing non-destructive recovery of pristine FTO glass. The material cost of PbSO₄ was only ~4% that of standard ETLs. Within a cradle-to-cradle LCA framework, this design-for-recycling strategy reduced resource depletion by 96.6% and human toxicity (cancer effects) by 68.8% relative to landfill disposal [24] (Figure 10a,b), illustrating how electrode-adjacent design choices can propagate into substantial system-level gains.

Figure 10. (a) System boundary of LCA considering the proposed recycling strategy as the EoL scenario. (b) Comparison of full-spectrum midpoint impact categories between recycling and landfill according to the Environmental Footprint (EF) v3.0 method (values normalized to the landfill scenario for better comparison). CED, cumulative energy demand; MR, material resources: metals/minerals; POF, photochemical oxidant formation: human health; IR, ionizing radiation: human health; HTN, human toxicity: non-cancer effects; HTC, human toxicity: cancer effects; CC, climate change. Reproduction with permission from Ref. [24]. Copyright 2025, Springer Nature.

As these technological concepts mature, solvent management and process chemistry have emerged as critical levers for aligning TCO/glass recycling with green-chemistry principles and the broader solvent-circularity strategies discussed in Section 5. Preeti et al. [110] demonstrated that patterned ITO/glass substrates can be recovered using a single alkali-hydroxide solvent, simplifying waste streams and process control. Larini et al. [93] replaced DMF with the greener solvent DMSO in recovering SnO₂-coated ITO substrates for n–i–p devices, achieving a champion PCE of 22.6% on recycled substrates indistinguishable from those on pristine ones. Amrein et al. [111] developed an environmentally benign process targeting simultaneous recovery of indium and silver via nanofiltration. Augustine [112] and Gallegos et al. [113] further showed that ITO-coated glass can be recycled under mild alkaline conditions, reinforcing the feasibility of low-toxicity, low-temperature dissolution protocols. These works collectively signal a shift from “recycling at any cost” toward recycling that is itself sustainable, where solvent choice, waste minimization and process simplicity are treated as first-order design criteria.

Finally, economic analyses underscore why TCO/glass recycling is not optional but structurally necessary. As summarized in Table 1, material costs account for ~70% of total PSC recycling cost, with FTO glass alone representing ~30% of the module material cost, and capital-intensive equipment contributing an additional ~15% [114]. Under current industrial conditions, discarding TCO/glass substrates at EoL is incompatible with competitive cost structures. Looking ahead, the trajectory of TCO/glass recycling will be determined not solely by recovery yields, but by the coupled optimization of device durability, dismantling logistics, solvent and waste management, and scalable recycling infrastructure. Closing this lab-to-fab gap will require coordinated efforts across materials science, device engineering, process systems engineering and environmental assessment, ensuring that TCO/glass substrates evolve from disposable carriers into core assets within a genuinely circular perovskite PV economy.

Table 1. Economic analysis of PSC recycling process.

Cost Components	Proportion of Total Cost	Cost Contribution	Representative Materials/Components	Cost Factors and Analysis
Material Cost	~70%	0.39 $ W−1	FTO glass, ITO, C60, perovskite absorber, NiOₓ, SnOₓ, Cu, encapsulation materials	Dominant cost contributor; mainly determined by the prices of transparent conductive substrates and charge transport materials; strongly affected by module efficiency and yield
└FTO Glass	~30% (of mat. cost)	10.2 $ m−2	FTO-coated glass	Most expensive single material; limited cost reduction potential; development of low-cost TCO substrates is required
└ITO (rear electrode)	11.5% (of mat. cost)	3.38 $ m−2	ITO sputtering target	Widely used in optoelectronics; price insensitive to PV manufacturing scale
└C60 (ETL)	~7.4% (of mat. cost)	2.16 $ m−2	Fullerene C60	High material price due to synthesis limitations; replacement by inorganic ETLs (e.g., SnO2) is desirable
└Perovskite absorber layer	~6% (of mat. cost)	1.89 $ m−2	Lead halide perovskite	Intrinsically low material cost; not a major cost driver
└Other materials	~14.7% (of mat. cost)	-	Al frame, back glass, POE, junction box, sealants	Comparable to conventional PV modules
Capital cost (equipment depreciation)	~15%	0.092 $ W−1	Vacuum deposition systems, coating equipment, laser scribing tools	High upfront investment; per-unit cost decreases with scale-up; limited impact compared to materials cost
Other Costs	~14.7%	0.088 $ W−1	Electricity, labor, land rent, maintenance	Electricity dominates (~60%) due to extensive use of vacuum equipment

Notes: Data in this table are taken from Ref. [114] and are calculated assuming a production capacity of 100 MW yr⁻¹, a module efficiency of 15%, and a manufacturing yield of 50%, representative of the current industrialization status. Material costs are first calculated on an area basis ($ m⁻²) and then converted into power-normalized costs ($ W⁻¹) based on module efficiency and yield. └ refers to the classification in Material Cost.

Although the aforementioned cases focus on rigid glass–TCO substrates, the key question is whether these recycling strategies can be extended to polymer-based TCO substrates used in flexible devices, thereby addressing a critical gap. Polymer-supported TCOs (e.g., ITO on PET or PEN) are increasingly important for lightweight PSCs in emerging applications such as aerospace and the Internet of Things (IoT) [115,116,117]. In principle, approaches including low-temperature solvent cleaning, selective dissolution, and layer-by-layer delamination are transferable. However, flexible substrates pose distinct challenges owing to limited thermal stability and solvent sensitivity; thermal debonding or aggressive chemical treatments must be substantially gentler to avoid polymer deformation or dissolution. Encouragingly, the shift toward milder, low-temperature recycling routes should benefit polymer-based perovskite devices. Design-for-recycling may further require readily separable interlayers or soluble contact materials to enable layer-by-layer disassembly. With appropriate interface engineering and encapsulation design, the inherently lower adhesion of certain layers on polymers may even be advantageous for mechanical peeling and reuse. Overall, lightweight demands in aerospace and IoT underscore the urgency of this research frontier. We predict that future recycling solutions for flexible PSCs will be co-developed with device architectures, prioritizing gentle, dismantle-friendly designs that facilitate material recycling and ensure the sustainability of next-generation lightweight perovskite photovoltaics.

5. Management of Solvents

Previous sections addressed the recycling and circular utilization of solid components in PSCs, including absorber layers, CTLs, electrodes, and substrates. However, TEA and environmental assessments indicate that the viability of these routes is strongly method-dependent and governed by coupled trade-offs among cost, resource efficiency, and environmental footprint. Table 2 summarizes the reuse and recycling technologies introduced in this review. As summarized in Table 1 and Table 2, adsorption–desorption can deliver high Pb recovery (>95%) [29] but relies on comparatively costly adsorbents, whereas solvent extraction is operationally simple yet shifts the burden to toxicity control and downstream waste treatment [72,73]. LCA and TEA further suggest that aqueous-based recycling can reduce GWP [24], but may introduce energy-intensive unit operations during scale-up. Integrated strategies can improve environmental performance at the system level, but typically demand higher upfront capital and tighter integration of unit operations. Collectively, these comparisons clarify that “best” recycling is not defined by recovery alone; it emerges from balancing recovery performance against solvent and energy demand, infrastructure requirements, and EoL liabilities. Future work should therefore prioritize low-energy, low-toxicity integrated flowsheets—for example, coupling solvent recovery [38] with lead-capture modules [118]—to reduce both operating cost and environmental burden. With continued progress in greener solvent systems and automation, PSC recycling can move toward closed-loop operation and ultimately support “design for recycling.”

Despite this emphasis on solid-component recovery, the large volumes of organic solvents used throughout solution processing and recycling are often treated as ancillary consumables rather than part of the “material system,” and are consequently omitted from many recycling schemes. This omission is consequential: solvents often dominate mass throughput, mobilize and accumulate toxic species (notably Pb²⁺), and contribute disproportionately to life-cycle burdens through production, emissions, and end-of-life handling [119]. Accordingly, solvent management warrants dedicated discussion as a central—rather than peripheral—lever for improving PSC sustainability. In the following sections, we first delineate the principal environmental and life-cycle burdens associated with current solvent use and end-of-life handling, and then critically evaluate process-level strategies for solvent recovery, purification, and cyclic utilization that can reduce impacts across the PSC value chain.

5.1. Impact of Solvents in Recycling

Large quantities of polar organic solvents are indispensable at every stage of PSCs solution processing and device fabrication, and their consumption typically exceeds that of solid materials [120]. Because these solvents are not incorporated into the final device stack, they have often been excluded from the “material system” in conventional PV research. Within the LCA framework, however, they emerge as “hidden contributors” to the sustainability profile of perovskite PVs. Solvents such as DMF, DMSO, N-methyl-2-pyrrolidone (NMP), and CB have therefore become critical bottlenecks for industrialization due to their toxicity, volatility, and demanding waste management requirements. Elevating solvent management to a central topic shifts the sustainability discussion from the materials level to the process level and is essential for achieving a genuinely closed-loop value chain [121].

DMF is almost indispensable in PSC processing because of its excellent PbI₂ solubility and film-forming capability, yet its toxicity is a major concern. The European Chemicals Agency (ECHA) classifies DMF as a Substance of Very High Concern (SVHC) for its reproductive toxicity and environmental hazards, and EU regulations limit its residual content in PV products to below 0.3%. Readily absorbed through the skin or by inhalation, DMF exposure can induce liver damage, dermatitis, and neurological disorders. In comparison, DMSO is less toxic and generally outperforms DMF in metrics such as ecotoxicity and occupational safety. Nevertheless, many routinely used solvents remain harmful. Antisolvents such as chlorobenzene and toluene are highly volatile and toxic: chlorobenzene is persistent and bioaccumulative, posing long-term risks to aquatic ecosystems and potential neurotoxicity and carcinogenicity in humans, while toluene is a recognized neurotoxin associated with chronic neurological damage and adverse reproductive and developmental outcomes. Moreover, polar aprotic solvents such as NMP, N,N-dimethylacetamide (DMAc), and γ-butyrolactone (GBL) are also listed in ECHA toxicity inventories, indicating notable ecological and occupational risks [34]. Taken together, these examples show that simply substituting DMF with other conventional high-boiling aprotic solvents affords only limited gains in sustainability and does not fundamentally resolve solvent-related hazards.

At the production scale, the magnitude and fate of solvent use further underscore the need for circular management. In pilot-scale PSCs manufacturing, approximately 312 metric tonnes of organic solvents such as DMF are consumed per 1 GW of capacity. Most of these solvents volatilize during coating and film formation and are captured by wet scrubbing, generating large volumes of dilute waste liquids. Common disposal methods—on-site incineration and deep-well injection—not only squander potentially valuable resources but also introduce additional environmental burdens. Incineration requires extra energy input and emits CO₂, NOₓ, and other pollutants, whereas deep-well injection buries toxic organics with long-term risks of soil and groundwater contamination under limited monitoring [122,123]. LCA studies indicate that such destruction-based strategies account for a substantial fraction of the solvent life-cycle environmental footprint, especially for high-risk solvents like DMF, and mere disposal does not fully eliminate the associated hazards. With tightening environmental regulations and rising prices of organic solvents, the prevailing linear model of “single use + disposal” has become unsustainable. Global waste organic solvent generation is estimated to reach 4.15 million metric tonnes by 2025, with an annual growth rate of ~3.5% [124]. These trends highlight the urgent need to redesign solvent use in PSCs manufacturing from a linear paradigm toward circular, recovery-oriented process schemes.

Table 2. PSC recycling strategies and performance reported in the literature.

Architectures	Recycling Methods	Recycled Materials	Times Recycled	Recovery Rate [%]	PCE Comparison [%]	Refs.
glass/ITO/SnO2/Perovskite/Spiro-OMeTAD/Ag	Systematic screening to identify optimal recycling parameters	PbI2, ITO, Ag	4	Pb in PbI2: 96.63 Ag Purity: 99.15	Fresh: 20.76 Recycled: 22.78	[76]
glass/ITO/Poly-TPD/MeO-2PACz/Perovskite/LiF/C60/SnO2/ITO/Ag	Physico-chemical recycling method utilizing hot water	PbI2	2	Pb: 92.6–100	-	[72]
glass/ITO/SnO2/FAPbI3/Spiro-OMeTAD/Au	Bioinspired “cage traps”	PbI2	-	Pb Purity: 96.07	Fresh: 22.37 Recycled: 22.08	[69]
glass/ITO/SnO2/perovskite/spiro-OMeTAD or PCBM/Au	WH nanoparticles as Pb adsorbent	PbI2	-	Pb Absorption: 2339 mg g−1	Commercial: 19.31 Recycled: 19.00	[29]
glass/FTO/TiO2/SnO2/perovskite/Spiro-OMeTAD/Ag	Ion-exchange method	PbI2	-	Pb Stabilization: ~100 Iodide Recovery: >95	Fresh: 21.58 Recycled: 21.58	[31]
glass/FTO/PbSO4/m-TiO2/perovskite/Spiro-OMeTAD/Au	Recyclable ETL based on PbSO4 nanoparticles	FTO Substrate	-	-	PbSO4: 24.1 TiO2: 23.7 Module (204.9 cm2): 17.9	[109]
glass/ITO/PTAA/Cs0.1FA0.9PbI3/C60/BCP/Cu	Weak-acid cation-exchange resin to adsorb Pb	PbI2, ITO Substrate	-	Pb: 99.2	Commercial: 21.0 Recycled: 20.4 ITO/glass: Comparable	[54]
glass/ITO/SnO2/MAPbI3/spiro-OMeTAD/Au	Layer-by-Layer Solvent Extraction	All Major Components	-	Total: 99.97	Fresh: 17.5 Recycled: 17.3–17.6	[36]
glass/FTO/TiO2/CH3NH3PbI3/spiro-OMeTAD/Au	Sequential Solvent Extraction at Room Temperature	FTO Substrate, Au, PbI2	2	-	Fresh: 16.1 1st Recycled: 16.0 2nd Recycled: 15.0	[27]
glass/ITO/PTAA/FA0.97MA0.03PbI3/PCBM/BCP/Ag	Multistage air-gap membrane distillation system	DMF solvent	-	-	commercial DMF (99.5 wt %): 20.07 recovered DMF (94.2 wt %): 19.97	[38]
glass/FTO/TiO2/MAPbI3/spiro-OMeTAD/Au	Two-step selective dissolution process	DMF solvent, CB	-	-	Fresh: 25.12 Recycled: 25.02	[34]
glass/ITO/4PADCB/perovskite/C60/SnO2/Ag	Resin adsorption–desorption-precipitation	PbI2 precursor	5	Pb: 97.1	Commercial: 18.39 Recycled: 17.72	[125]
glass/FTO/CBD SnO2/perovskite/spiro-OMeTAD/Au	Ultrasonic cleaning in DMF followed by water rinsing and drying	SnO2	3	98	From 22.6 to 22.1	[92]
glass/ITO/SnO2/FAPbI3/Spiro-OMeTAD/Au	Using DMSO as green solvent	SnO2-coated ITO substrate	-	-	Same: 22.6	[93]
glass/ITO/SnO2/perovskite/spiro-OMeTAD/Ag	Sequential dissolution using CB and DMF	spiro-OMeTAD, Ag, PbI2, ITO/SnO2, DMF	4	Ag purity: 99.17	Fresh: 20.77 Recycled: 23.41	[35]
glass/ITO/SnO2/FAPbI3/Spiro-OMeTAD/Au	Multi-step process using green solvents	Spiro-OMeTAD, ITO substrate	3	Overall: 81 Spiro-OMeTAD: 89	1st recycle: 98.4% of fresh 3rd recycle: 84.2% of fresh	[85]
Perovskite PV	Holistic recycling: aqueous-based for perovskite	All Major Components	5	Perovskite: 99.0 ± 0.4	Fresh: 22.1 ± 0.9 Recycled: 21.9 ± 1.1	[24]
glass/ITO/SnO2/perovskite/spiro-OMeTAD/Au	One-key-reset method using bleacher solution	All Major Components	2	-	Fresh: 20.65 1st Recycled: 20.30 2nd Recycled: 20.08	[75]

Table 2. PSC recycling strategies and performance reported in the literature.

Architectures	Recycling Methods	Recycled Materials	Times Recycled	Recovery Rate [%]	PCE Comparison [%]	Refs.
glass/ITO/SnO2/Perovskite/Spiro-OMeTAD/Ag	Systematic screening to identify optimal recycling parameters	PbI2, ITO, Ag	4	Pb in PbI2: 96.63 Ag Purity: 99.15	Fresh: 20.76 Recycled: 22.78	[76]
glass/ITO/Poly-TPD/MeO-2PACz/Perovskite/LiF/C60/SnO2/ITO/Ag	Physico-chemical recycling method utilizing hot water	PbI2	2	Pb: 92.6–100	-	[72]
glass/ITO/SnO2/FAPbI3/Spiro-OMeTAD/Au	Bioinspired “cage traps”	PbI2	-	Pb Purity: 96.07	Fresh: 22.37 Recycled: 22.08	[69]
glass/ITO/SnO2/perovskite/spiro-OMeTAD or PCBM/Au	WH nanoparticles as Pb adsorbent	PbI2	-	Pb Absorption: 2339 mg g−1	Commercial: 19.31 Recycled: 19.00	[29]
glass/FTO/TiO2/SnO2/perovskite/Spiro-OMeTAD/Ag	Ion-exchange method	PbI2	-	Pb Stabilization: ~100 Iodide Recovery: >95	Fresh: 21.58 Recycled: 21.58	[31]
glass/FTO/PbSO4/m-TiO2/perovskite/Spiro-OMeTAD/Au	Recyclable ETL based on PbSO4 nanoparticles	FTO Substrate	-	-	PbSO4: 24.1 TiO2: 23.7 Module (204.9 cm2): 17.9	[109]
glass/ITO/PTAA/Cs0.1FA0.9PbI3/C60/BCP/Cu	Weak-acid cation-exchange resin to adsorb Pb	PbI2, ITO Substrate	-	Pb: 99.2	Commercial: 21.0 Recycled: 20.4 ITO/glass: Comparable	[54]
glass/ITO/SnO2/MAPbI3/spiro-OMeTAD/Au	Layer-by-Layer Solvent Extraction	All Major Components	-	Total: 99.97	Fresh: 17.5 Recycled: 17.3–17.6	[36]
glass/FTO/TiO2/CH3NH3PbI3/spiro-OMeTAD/Au	Sequential Solvent Extraction at Room Temperature	FTO Substrate, Au, PbI2	2	-	Fresh: 16.1 1st Recycled: 16.0 2nd Recycled: 15.0	[27]
glass/ITO/PTAA/FA0.97MA0.03PbI3/PCBM/BCP/Ag	Multistage air-gap membrane distillation system	DMF solvent	-	-	commercial DMF (99.5 wt %): 20.07 recovered DMF (94.2 wt %): 19.97	[38]
glass/FTO/TiO2/MAPbI3/spiro-OMeTAD/Au	Two-step selective dissolution process	DMF solvent, CB	-	-	Fresh: 25.12 Recycled: 25.02	[34]
glass/ITO/4PADCB/perovskite/C60/SnO2/Ag	Resin adsorption–desorption-precipitation	PbI2 precursor	5	Pb: 97.1	Commercial: 18.39 Recycled: 17.72	[125]
glass/FTO/CBD SnO2/perovskite/spiro-OMeTAD/Au	Ultrasonic cleaning in DMF followed by water rinsing and drying	SnO2	3	98	From 22.6 to 22.1	[92]
glass/ITO/SnO2/FAPbI3/Spiro-OMeTAD/Au	Using DMSO as green solvent	SnO2-coated ITO substrate	-	-	Same: 22.6	[93]
glass/ITO/SnO2/perovskite/spiro-OMeTAD/Ag	Sequential dissolution using CB and DMF	spiro-OMeTAD, Ag, PbI2, ITO/SnO2, DMF	4	Ag purity: 99.17	Fresh: 20.77 Recycled: 23.41	[35]
glass/ITO/SnO2/FAPbI3/Spiro-OMeTAD/Au	Multi-step process using green solvents	Spiro-OMeTAD, ITO substrate	3	Overall: 81 Spiro-OMeTAD: 89	1st recycle: 98.4% of fresh 3rd recycle: 84.2% of fresh	[85]
Perovskite PV	Holistic recycling: aqueous-based for perovskite	All Major Components	5	Perovskite: 99.0 ± 0.4	Fresh: 22.1 ± 0.9 Recycled: 21.9 ± 1.1	[24]
glass/ITO/SnO2/perovskite/spiro-OMeTAD/Au	One-key-reset method using bleacher solution	All Major Components	2	-	Fresh: 20.65 1st Recycled: 20.30 2nd Recycled: 20.08	[75]

5.2. Solvent Recovery and Circular Processes

In light of these challenges, moving PSCs manufacturing from linear solvent use to closed-loop management is no longer optional but imperative. Achieving solvent circularity, however, requires integrated strategies for recovery, purification, and reuse that preserve device performance while addressing hazardous waste streams.

At the device-recycling level, Kim et al. [34] proposed an integrated closed-loop strategy to mitigate the environmental risks associated with toxic solvents used during PSC recycling. They first showed that conventional one-step dissolution routes (Figure 11a), which rely heavily on strongly polar solvents such as DMF, typically involve liquids with high toxicity scores (Figure 11b), and most highly polar candidates are classified as hazardous. In addition, residual Spiro-OMeTAD in the reclaimed solvent led to severe performance degradation of refabricated devices, revealing a key compatibility issue often overlooked in simple solvent reuse scenarios. To address these challenges, Kim et al. developed a two-step selective dissolution process: CB was first employed to selectively dissolve and recover the Spiro-OMeTAD layer, and DMF was subsequently used to dissolve the perovskite absorber. Hematite nanoparticles were then introduced to adsorb toxic Pb²⁺ from the DMF (Figure 11d). PSCs refabricated using the recycled solvents and recovered materials achieved a PCE of 25.02%, comparable to pristine devices (25.12%), and exhibited excellent operational stability. This work demonstrates that careful sequence design and impurity management are crucial for enabling high-performance device refabrication while simultaneously recovering both functional materials and solvents. At the same time, the continued reliance on CB and DMF indicates that such closed-loop schemes should ideally be complemented by the development of safer solvent systems.

Figure 11. (a) Typical PSC recycling process that generates toxic solvents. (b) Hazardous scores based on the ACS GCI-PR guideline versus the polarity of commonly used solvents in the PSC recycling process, where a more reddish color indicates higher toxicity. (c) Recycling of CB and DMF in a two-step selective dissolution process. Reproduction with permission from Ref. [34]. Copyright 2023, American Chemical Society. (d) UV aging test results of encapsulated perovskite mini-modules under continuous AM 1.5 G solar irradiation for over 800 h. Reproduction with permission from Ref. [38]. Copyright 2025, American Association for the Advancement of Science.

At the process-engineering scale, Zheng et al. [38] developed a multistage air-gap membrane distillation (MAMD) system to efficiently recover DMF from dilute waste solutions using industrial waste heat. Their MAMD system achieved a DMF enrichment factor of up to 314, increasing the concentration from 0.3 to 94.2 wt %, and maintained stable operation for 60 h. The recovered DMF (94.2 wt %) was subsequently employed for perovskite minimodule fabrication, yielding a certified stabilized power output of 19.97% (Figure 11c). The narrow device-to-device efficiency spread, state-of-the-art PCE, and small hysteresis collectively demonstrate that high-purity recovered DMF can satisfy the stringent process window required for industrial PSC fabrication (Figure 12a,b). Compared with dissolution–refabrication strategies that integrate solvent recovery within the device recycling workflow, the MAMD approach targets bulk solvent streams and leverages low-grade heat, offering a promising pathway for factory-level solvent circularity. Future work will need to address membrane fouling, long-term durability, and integration with upstream scrubber systems to fully realize its industrial potential.

Figure 12. (a) Recovery of DMF organic solvent using a MAMD system. Schematic illustration of the MAMD system for DMF organic solvent recycling, highlighting its applicability in industrial production processes, including solar cell manufacturing, semiconductor fabrication, pharmaceutical production, and fine chemical synthesis. (b) Layout of a MAMD system in cross-flow mode, with the effluent from the previous stage flowing into the subsequent stage following the direction indicated by the solid red arrow, and the distillate from each stage following the direction indicated by the dashed blue arrow. Reproduction with permission from Ref. [38]. Copyright 2025, American Association for the Advancement of Science.

From an LCA perspective, Rodriguez-Garcia et al. [105] unequivocally demonstrated that solvent management is a decisive factor in achieving circularity in PSC recycling. They conducted a comprehensive LCA of 13 techniques for recovering TCO-coated glass, identifying distillation as the optimal method for solvent recovery. The study emphasized that the multiple reuse of solvents can enable various recycling pathways to offer greater environmental benefits compared to the production of new materials. Viewed alongside the experimental demonstrations by Kim et al. and Zheng et al., this LCA evidence underscores that solvent recovery and reuse are not marginal optimizations, but central design criteria for environmentally meaningful PSC recycling processes.

Solvent recovery and management in PSC manufacturing require not only reclaiming the bulk solvents DMF and DMSO, but also removing dissolved Pb species that otherwise prevent safe solvent recirculation and compliant handling. Park et al. [118] addressed this solvent-purification bottleneck by deploying a magnetic Fe-incorporated hydroxyapatite (HAP/Fe) adsorbent. Fe incorporation strengthens the negative surface potential via charge delocalization (Figure 13a), thereby enhancing Pb²⁺ uptake in non-aqueous media and enabling rapid magnetic separation. After treatment, the recovered polar solvent contained <15 ppb Pb, and the captured lead was reprecipitated as PbI₂ with 99.97% recovery, linking solvent reuse with precursor regeneration (Figure 13e). Complementarily, Zou et al. [125] implemented a closed-loop solvent management scheme using an iminodiacetic-acid chelating resin (CH-90Na), which selectively extracts Pb from waste organic solvents with a maximum capacity of 208.0 mg g⁻¹. This process lowered the Pb concentration in the treated solvent to 0.02 ppm (below 5 ppm). Subsequent acid desorption and iodide precipitation produced high-purity PbI₂ (~99.99%) with 97.1% overall recovery. The resin remained effective over multiple adsorption–desorption–regeneration cycles, ensuring the solvent could be repeatedly purified (Figure 13b–d). Taken together, these studies indicate that integrating targeted Pb capture into solvent recovery can transform Pb-contaminated waste solvents into reusable process streams while simultaneously regenerating PbI₂ feedstock. This approach facilitates closed-loop solvent management and reduces lead-related environmental, health, and safety risks in PSC production.

Figure 13. (a) Illustration of the use of a HAP/Fe composite for treating a Pb-containing pollutant solution and the PbI₂ recovery process after Pb removal/separation. Reproduction with permission from Ref. [118]. Copyright 2020, Springer Nature. (b) Experimental flowchart of the resin-based lead recovery process from waste organic solvents. (c) Comparison of the full spectra of recovered PbI₂ and commercial PbI₂. (d) Relationship between the number of resin recycling cycles and its adsorption capacity and adsorption efficiency for Pb(II) from waste organic solvents. Reproduction with permission from Ref. [125]. Copyright 2025, Elsevier. (e) Concentration of Pb before and after adsorption using HAP/Fe composites with different Fe weight percentages; 20 mL of a 2 mM PbI₂/DMF solution was used as the Pb stock solution. Reproduction with permission from Ref. [118]. Copyright 2020, Springer Nature.

6. Conclusions and Outlook

In summary, this review conducts a systematic investigation into recycling technologies that facilitate the transformation of PSCs from waste materials into high-value secondary resources. A wide variety of methods, ranging from physical separation to chemical dissolution and hybrid approaches, have achieved varying degrees of success in recovering key components such as lead, organic materials, and transparent conductive substrates. Despite these significant advancements, challenges like high processing costs, environmental safety concerns, and scalability limitations persist. Looking forward, future research should prioritize optimizing recycling processes to improve efficiency and reduce environmental impact. Innovations in material design, such as the development of lead-free or less toxic perovskite materials, hold the potential to simplify the recycling process. Additionally, establishing standardized recycling protocols and fostering collaborative partnerships between industry and academia will be of utmost importance for large-scale implementation. The following are suggestions for the future focus areas and research directions in this field:

(1)

Scalable and automated recycling processes. Current recycling methods for PSCs remain largely laboratory-scale, with limited industrial applicability. Future advancements will prioritize the development of automated, high-throughput recycling systems capable of handling large-scale PSC waste from manufacturing defects, EoL modules, and BIPV installations. For instance, the integration of AI-driven sorting technologies could enable real-time identification and separation of PSC components, reducing manual labor and improving recovery efficiency. Additionally, roll-to-roll recycling lines could be adapted for PSCs, leveraging their thin-film structure to achieve cost-effective material recovery. This shift toward automation will align with the industry’s push for GW-scale production, ensuring recycling infrastructure keeps pace with rapid capacity expansions [126,127,128].

(2)

Stability-enhanced recycling for long-term reliability. As PSCs approach commercialization, their long-term stability under real-world conditions becomes critical for recycling viability. Future research will focus on designing recyclable PSCs with inherent stability, such as all-inorganic perovskites or 2D/3D heterostructures that resist decomposition. Additionally, self-healing materials could extend PSC lifespans, reducing the frequency of recycling interventions. For EoL modules, non-destructive testing will enable selective recycling, prioritizing components with residual functionality for refurbishment rather than material recovery. These advancements will ensure recycled PSCs meet 25-year performance warranties, a benchmark set by silicon rivals [74,86,129].

(3)

Closed-loop material recovery and reuse. The next frontier in PSC recycling lies in closed-loop systems that minimize virgin material consumption. Key targets include recovering high-purity Pb from degraded perovskite layers. Innovations such as electrochemical extraction and solvent-based purification have demonstrated Pb recovery rates exceeding 95%, with the potential to reintroduce recycled Pb into new PSC production. Similarly, TCO glass can be reclaimed through laser scribing and chemical etching, preserving its optical and electrical properties for reuse. Furthermore, organic additives could be recovered via distillation or chromatography, reducing reliance on expensive synthetic pathways [130,131].

Although the field of PSC recycling is still in its infancy, the strides made thus far are promising. Through the collaborative and sustained efforts of researchers, policymakers, and industry participants, the dream of turning waste PSCs into valuable resources is on the horizon, thereby laying a solid foundation for a more sustainable and robust solar energy future.