Thin-Film Solar Cells for Solar Thermal Cooling, Heating, and Energy Storage Systems: Materials, Manufacturing, and Emerging Applications

Abstract

Thin-film solar cells (TFSCs) remain a cornerstone of the global transition toward renewable energy, characterized by consistent reductions in manufacturing costs and steady gains in power conversion efficiency. In addition to electricity generation, TFSCs play an important role in advanced solar thermal cooling, heating, and energy storage systems, where their tunable optical absorption, low thermal mass, and flexibility enable integration with photovoltaic–thermal (PV/T) collectors, thermally driven cooling cycles, and hybrid thermal–electrical storage architectures. This paper provides a comprehensive review of prominent TFSC technologies, including copper indium gallium selenide (CIGS), cadmium telluride (CdTe/CdS), amorphous silicon (a-Si), copper zinc tin sulfide (CZTS), organic photovoltaics (OPVs), and metal halide perovskite solar cells (PSCs), with a focus on their material structures, performance specifications, and current efficiency benchmarks. Compared to state-of-the-art reviews, this article distinguishes itself by addressing next-generation innovations, cross-domain solar thermal–photovoltaic applications, and economic analysis. Specifically, the integration of machine learning and simulation-based material dynamics is examined to accelerate material discovery, process optimization, and the characterization of novel TFPV components relevant to coupled thermal–electrical energy systems. Furthermore, the study explores how additive manufacturing is transforming the industry through the development of high-efficiency electrodes, electrohydrodynamic atomization for thin-film deposition, and the fabrication of flexible solar arrays suitable for thermally integrated and building-scale energy systems, including space applications. By integrating advancements in module efficiency, scalable manufacturing approaches, and techno-economic analysis, this paper positions TFSCs as sustainable, resource-abundant technologies essential for next-generation solar thermal cooling, heating, and energy storage infrastructures.

1. Introduction

The global transition toward sustainable and low-carbon energy systems has intensified the demand for photovoltaic technologies that are efficient, scalable, and economically viable [1,2]. While crystalline silicon photovoltaics continue to dominate commercial markets, their rigid form factor, high material usage, and energy-intensive manufacturing processes impose constraints on further cost reduction and system integration [1,3]. As a result, alternative photovoltaic technologies capable of complementing conventional silicon-based systems have become increasingly important [1]. Among these, thin-film solar cells (TFSCs) have emerged as a promising class of photovoltaic devices due to their reduced material consumption, mechanical flexibility, and compatibility with large-area manufacturing [1,4]. These characteristics not only support scalable electricity generation but also facilitate integration with solar thermal heating, cooling, and energy storage systems, particularly within photovoltaic–thermal and hybrid energy architectures.

TFSCs are characterized by semiconductor absorber layers that are typically only a few micrometers thick and are deposited on substrates such as glass, metal, or flexible polymers [1,5]. This structural configuration enables lightweight devices, lower embodied energy, and the possibility of scalable deposition techniques, including roll-to-roll and solution-based processing [1,6]. These advantages make thin-film photovoltaics particularly attractive for applications where conventional rigid silicon modules are impractical, including building-integrated photovoltaics, portable and wearable electronics, transportation systems, aerospace platforms, and off-grid energy generation [7,8,9].

Sustained research and industrial development over the past several decades have led to significant improvements in the performance and reliability of thin-film photovoltaic technologies [1,2]. Mature platforms such as CdTe and copper indium gallium selenide have achieved laboratory-scale efficiencies exceeding twenty-two percent, with commercial modules approaching the performance of crystalline silicon [2,10,11,12]. Amorphous silicon has evolved from early low-efficiency devices into a critical material for silicon heterojunction and tandem architectures [3,13,14]. In parallel, emerging thin-film materials, including kesterite-based absorbers and halide perovskites, have expanded the technological landscape by offering pathways toward high-efficiency, earth-abundant, and next-generation tandem solar cells [15,16,17].

Despite these advances, several fundamental and practical challenges continue to limit the widespread adoption of TFSCs [1,18]. Performance losses associated with defect-mediated recombination, open-circuit voltage deficits, material instability, and interface degradation remain critical issues [5,19,20]. In addition, concerns related to material toxicity, elemental scarcity, and long-term durability must be addressed to ensure sustainable deployment at scale [1,10,21]. These challenges are inherently multi-dimensional and cannot be resolved through materials development or device engineering alone [1].

In response, recent research has increasingly adopted data-driven and multi-scale approaches to accelerate the design and optimization of thin-film photovoltaic systems [22,23,24]. Machine learning techniques are now widely used to model nonlinear relationships between fabrication parameters, material properties, and device performance, enabling performance prediction, inverse design, and automated process optimization [22,23,24]. At the same time, additive manufacturing and printed electronics have introduced new opportunities for scalable and low-waste fabrication of TFSCs on flexible and unconventional substrates [6,9,25]. These manufacturing innovations are particularly relevant for organic and perovskite-based photovoltaics, where solution processability enables high-throughput production and novel form factors [6,26].

Complementing experimental and data-driven methods, molecular dynamics simulations and atomistic modeling have provided valuable insight into the microscopic mechanisms governing thin-film solar cell behavior [19,27,28]. Such simulations enable detailed investigation of grain boundary effects, defect formation, interfacial stability, thermal transport, and degradation pathways that are difficult to probe experimentally [19,27]. When integrated with device-level modeling and experimental validation, these atomistic approaches offer a powerful framework for understanding performance limits and improving long-term stability [20,28].

Beyond technical performance, economic considerations play a decisive role in determining the viability of thin-film photovoltaic technologies [1,29,30]. Manufacturing cost structures, life-cycle cost analysis, and levelized cost of electricity are strongly influenced by material selection, fabrication methods, and operational lifetime [29,31,32]. For printed and organic thin-film solar modules in particular, achieving competitive costs requires simultaneous advances in efficiency, durability, and scalable manufacturing processes [6,32].

This review provides a comprehensive overview of TFSC technologies within this evolving research landscape and thermal applications [1]. The paper first introduces the fundamental performance parameters, construction, and operating principles of thin-film photovoltaic devices [1]. It then presents a detailed discussion of major thin-film technologies, including CdTe, amorphous silicon, copper zinc tin sulfide-based cells, copper indium gallium selenide, and perovskite solar cells, highlighting their material properties, device architectures, performance evolution, and remaining challenges [1,5,16,17,18]. The review further examines emerging research directions, including machine learning-assisted design, additive manufacturing and printed photovoltaics, molecular dynamics simulations, and cost-driven analysis of thin-film modules [22,23,25,29]. By integrating materials science, data-driven modeling, manufacturing innovation, and economic assessment, this work aims to provide a unified perspective on the current state and future potential of TFSCs for sustainable energy systems [1,2].

2. Fundamentals of Thin-Film Solar Cells

2.1. Construction

TFSCs are constructed using a multilayer architecture in which each layer performs a distinct optical and electrical function, as schematically illustrated in Figure 1. The device is typically built on a rigid or flexible substrate, such as glass, metal foil, or polymer, which provides mechanical support and determines thermal compatibility and deployment flexibility. A transparent conductive oxide (TCO) layer, commonly indium tin oxide (ITO) or zinc oxide (ZnO), serves as the front electrode by combining high optical transparency with low electrical resistivity to enable efficient photon transmission and lateral charge collection. The photoactive absorber layer constitutes the functional core of TFSCs, where incident photons generate electron–hole pairs that are separated across a p–n junction or heterojunction interface; representative absorber materials include amorphous silicon (a-Si), cadmium telluride (CdTe), copper indium gallium selenide (CIGS), copper zinc tin sulfide (CZTS), organic semiconductors, and metal halide perovskites. A buffer or junction layer is often incorporated to optimize band alignment and suppress recombination losses. The back contact, typically composed of highly conductive metals such as molybdenum, silver, or aluminum, completes the electrical circuit and must exhibit minimal resistive losses. Additional functional layers, including anti-reflection coatings, passivation layers, and encapsulation barriers, may be integrated to enhance optical absorption, improve carrier lifetime, and ensure environmental stability. This multilayer structure enables the effective integration of TFSCs with photovoltaic–thermal collectors, thermally activated cooling technologies, hybrid thermal–electrical energy storage systems, and a wide range of other adaptable applications [33,34].

2.2. Working Principle

TFSCs operate based on a p–n semiconductor junction mechanism, as illustrated in Figure 2. When photons with energy equal to or greater than the absorber bandgap are absorbed in the photoactive layer, electrons are excited from the valence band to the conduction band, generating electron–hole pairs. The built-in electric field across the depletion region drives electrons toward the n-type region and holes toward the p-type region, enabling charge separation and suppressing recombination. The collected electrons flow through the external circuit to deliver electrical power and recombine with holes upon returning to the p-type side. Under illumination, the photocurrent flows opposite to the diode forward-bias direction, while a small thermally generated dark current flows from the p-type to the n-type region and remains negligible under normal operating conditions. From a materials science perspective, TFSC performance is governed by the absorber bandgap, semiconductor properties, junction quality, defect density, carrier mobility, and interfacial recombination. Efficiency can be improved through bandgap engineering for optimal spectral matching, defect passivation, advanced junction and interface design, and low-resistivity transport layers to minimize series resistance and carrier losses. Notably, because the absorber bandgap can be engineered to selectively harvest specific spectral bands, the fraction of incident energy not converted to electricity is available as recoverable heat, a characteristic that underpins the dual electrical–thermal functionality exploited in PV/T and hybrid energy storage configurations.

2.3. Performance Parameters

The performance of TFSCs is fundamentally evaluated through current–voltage (I–V) characteristics, from which key electrical parameters are extracted. These parameters provide quantitative insight into charge generation, recombination losses, resistive effects, and carrier extraction. Because thin-film photovoltaics rely on comparatively thin absorber layers and are often limited by defect states and interface quality, parameter-level analysis is essential for diagnosing loss mechanisms and guiding device optimization.

The illuminated I–V curve describes the relationship between output current and voltage under operating conditions. As shown in Figure 3, the curve spans from the short-circuit condition, where the current is maximum, and voltage is zero, to the open-circuit condition, where the voltage is maximum and current is zero. The curvature and slopes of the I–V response encode internal losses such as series resistance, shunt leakage, and recombination. In thin-film devices, these deviations are commonly associated with grain boundaries, interface traps, and non-uniform absorber layers, making I–V analysis a primary diagnostic tool.

From the $I–V$ curve, the first key quantity is the short-circuit current ($I_{sc}$), which represents the maximum photocurrent generated when the cell terminals are shorted. Figure 4 shows the $I_{sc}$ curve, which primarily reflects photon absorption and carrier collection. For thin-film technologies, $I_{sc}$ is strongly influenced by absorber thickness, optical confinement, and defect-assisted recombination. Since thin absorbers reduce material usage but can limit absorption, improving $I_{sc}$ through light-trapping and passivation is crucial for sustaining high efficiency.

In contrast, the open-circuit voltage ($V_{oc}$) captures the voltage limit of the device when no external current flows. $V_{oc}$ reflects quasi-Fermi level splitting and is therefore highly sensitive to recombination pathways. In thin-film photovoltaics, $V_{oc}$ is frequently constrained by bulk defects and interface states, making junction quality, band alignment, and defect passivation central to improvement. The $V_{OC}$ can be expressed as

$$V_{oc}={\displaystyle \frac{nkT}{q}}ln\left({\displaystyle \frac{I_L}{I_o}}+1\right),$$

(1)

where n is the diode ideality factor, k is Boltzmann’s constant, T is temperature, q is the elementary charge, $I_L$ is the light-generated current, and $I_o$ is the saturation current.

While $I_{sc}$ and $V_{oc}$ define the extremes of the I–V curve, practical operation occurs at the maximum power point (MPP), where the product of current and voltage is maximized. The MPP therefore determines the usable power output and energy yield of a module [35]. In thin-film systems, environmental variation and degradation can shift the MPP over time, so maintaining a stable operating point is important for field reliability and predictable performance.

Finally, the fill factor ($FF$) summarizes how closely the I–V curve approaches an ideal rectangular shape, and is defined as

$$FF={\displaystyle \frac{V_{MP}\times I_{MP}}{V_{OC}\times I_{SC}}}.$$

(2)

The FF captures the combined impact of resistive losses and recombination. In thin-film devices, the FF is particularly sensitive to contact design, series resistance at interfaces, shunt pathways, and layer uniformity. Because the power conversion efficiency scales with $I_{sc}\times V_{oc}\times FF$, even modest FF improvements can translate into meaningful efficiency gains.

Accordingly, the overall power conversion efficiency $\eta$ of a solar cell is given by

$$\eta ={\displaystyle \frac{V_{OC}\times I_{SC}\times FF}{P_{in}}},$$

(3)

where $P_{in}$ is the incident optical power. Collectively, $I_{sc}$, $V_{oc}$, MPP, and FF provide a compact but powerful framework for diagnosing loss mechanisms and directing materials and device-level improvements in thin-film solar cells.

3. Major Thin-Film Solar Cell Types

3.1. Cadmium Telluride (CdTe) and Selenium-Alloyed (CdSeTe) Photovoltaics

Cadmium telluride (CdTe) represents the most prominent thin-film photovoltaic technology currently in production, distinguished by its durable absorber material, chemical stability, and highly versatile manufacturing methods [36]. While early devices were centered on nominally pure CdTe absorbers, state-of-the-art high-efficiency devices now employ selenium-alloyed CdSeTe/CdTe architectures, in which selenium incorporation near the front region narrows the effective bandgap to approximately 1.4 eV and enhances long-wavelength absorption and photocurrent [37,38]. These advances, together with improved front-interface design and Group-V doping strategies, have increased the certified research-cell efficiency to 23.1% [39]. The absorber bandgap range of modern CdSeTe/CdTe devices remains close to the single-junction optimum and corresponds to a detailed-balance efficiency ceiling above 30%, indicating substantial room for further improvement in the $V_{OC}$ and fill factor [38,40], while commercial large-area modules typically achieve around 19% [36].

A schematic of an ultra-thin CdSeTe/CdTe device is shown in Figure 5, with an FTO/TO front contact, a CdSexTe1-x region near the junction, a CdTe layer, and a back contact. The same schematic also compares CdSe/CdTe thickness ratios of 75/725, 150/650, 270/530, and 400/400 nm within a total absorber thickness of 0.8 μm, illustrating how selenium content is varied in the device stack [37].

Sol–gel spin-coated Mg-doped ZnO thin films have been reported to show Mg-dependent changes in crystallinity/surface morphology and optical transparency (bandgap-related optical response), supporting MgZnO as a tunable wide-bandgap window/ETL material in such device stacks [41]. To facilitate efficient hole collection, a copper telluride ($CuTe$) hole transport layer (HTL) is utilized in conjunction with a gold cathode. Maintaining process temperatures within 550 °C is critical for preserving the integrity of these materials during fabrication [42].

As shown in the energy-band diagram in Figure 6, the operational mechanism involves photogenerated electrons moving from the p-type to the n-type junction toward the anode, and subsequently traveling through an external circuit to the cathode. Here, they recombine with holes provided by the HTL.

The evolution of CdTe technology has been characterized by consistent improvements in laboratory and module-level performance. Recent advancements between 2021 and 2025 have focused on selenium alloying ($CdSeTe$) to enhance the short-circuit current ($J_{sc}$) by narrowing the bandgap at the front interface [43]. Furthermore, the transition toward Group-V doping (e.g., arsenic or antimony) has significantly stabilized the $V_{OC}$ compared to traditional copper-based methods. Table 1 outlines the key performance parameters and milestones through 2025.

Beyond standalone electricity generation, CdTe/CdSeTe photovoltaics are also relevant to solar thermal cooling, heating, and energy storage applications because their relatively low temperature coefficient, good performance under high-temperature and diffuse-light conditions, and compatibility with lightweight superstrate, façade, and glazing configurations make them attractive for building-integrated photovoltaic–thermal (BIPV/T) and hybrid solar energy systems [44,45]. In such configurations, heat extracted from the back side of the module can be recovered for domestic hot water, space heating, or low-temperature process heat, while the associated reduction in cell operating temperature helps preserve electrical output [46]. The same coupled thermal–electrical behavior is relevant to thermally assisted cooling architectures, including PV/T systems coupled with heat pumps, dehumidification units, or hybrid cooling loops, and to thermal energy storage configurations that use water tanks or phase-change materials to shift useful heat beyond peak solar hours [47,48]. Accordingly, the significance of CdTe/CdSeTe devices in this review extends beyond high-efficiency power conversion alone to multifunctional solar platforms in which electrical generation, heat recovery, cooling support, and thermal buffering are integrated within the same system footprint [44].

The historical trajectory and technological maturation of cadmium telluride (CdTe) thin-film photovoltaics are summarized in Table 1. The timeline shows a sustained and well-documented rise in conversion efficiency, advancing from early ∼6% prototypes in the 1970s to 23.1% class devices by 2024, followed by continued progress and consolidation of performance and manufacturability during 2017–2026. The early gains were enabled by process innovations that established reproducible, high-quality junction formation, particularly the widespread adoption of close-space sublimation (CSS) for dense absorber growth and chemical bath deposition (CBD) for controlled buffer layers. As the technology matured, device designs evolved from basic heterojunction concepts toward highly engineered superstrate architectures incorporating optimized transparent conducting oxides (TCOs), band-aligned window layers, and rigorous post-deposition treatments, such as $CdC{l}_2$ activation, to passivate defects, promote grain growth, and suppress interfacial recombination, thereby improving the $V_{oc}$ and fill factor.

A critical processing step in CdTe and CdSeTe device fabrication is CdCl2 activation. This post-deposition treatment is central to achieving high-performance CdTe-based photovoltaics because it promotes grain growth and recrystallization, drives chlorine redistribution and segregation at grain boundaries and heterointerfaces, and suppresses defect-assisted non-radiative recombination, thereby improving junction quality and carrier collection. These microstructural and electronic changes are strongly associated with increases in minority-carrier lifetime, $V_{OC}$, fill factor, and overall power conversion efficiency. Moreover, recent work on CdTe and CdSeTe/CdTe absorbers continues to identify chloride activation as a key step in absorber evolution and interface optimization [49,50].

More recent milestones emphasize that CdTe performance improvements are increasingly driven by precise defect and doping control, interface engineering, and stability-oriented device design rather than by incremental process tuning alone. In particular, modern CdSeTe/CdTe stacks and Group-V doping strategies have been used to enhance p-type conductivity and mitigate non-radiative recombination pathways, supporting higher $V_{oc}$ and strengthening the prospects for improved long-term operational stability. In parallel, the post-2017 period also reflects a broader shift toward sustainability and bankability considerations, including durability evidence from long-term outdoor exposure, reduced reliance on copper-related defect chemistry through alternative doping approaches, and continued advances in module-scale implementation. Overall, Table 1 captures how CdTe has progressed from laboratory demonstrations to a mature thin-film platform, with 2017–2026 developments highlighting the transition from efficiency-driven optimization to integrated progress in performance, stability, and manufacturing practicality.

Table 1. Sequential developments and historical milestones of CdTe/CdSeTe solar cells.

Year	Applied Methodology/Innovation	$\mathit{\eta }$ (%)	Developed By/Significance
1972	First CdTe/CdS cell via graded gap thin film; 3-step VPD for p-CdTe and vacuum evaporation for n-CdS.	6.0 (B)	Bonnet and Rabenhorst [51]
1976	Ceramic thin film (20 $\mathsf{\mu }$m) via screen printing; achieved 0.2 $\mathsf{\Omega }$-cm resistivity.	8.1 (P)	Nakayama et al. (MEI) [52]
1982	Thin poly-CdS/CdTe via CSS on SLG/In2O3; oxygen induction for shallow junction ($V_{oc}=750$ mV).	10.5 (P)	Tyan and Perez (Eastman Kodak) [53]
1982	ITO/CdTe and CdS/CdTe via E-beam evaporation; Br-in-methanol etching and H-heating surface treatments.	7.5–10.5 (P)	Werthen et al. (Stanford Univ.) [54]
1982	Prep. of p-CdTe on n-CdTe via VTD; optimized shallow $p^{+}n{n}^{+}$ form ($V_{oc}=850$ mV).	11.0 (P)	Cohen et al. (CNRS) [55]
1983–1984	Cu-addition in C paste to reduce series resistance; achieved high $J_{sc}$ (22 mA/cm2) via screen printing.	12.8 (P)	Kuribayashi and Matsumoto (MEI) [56,57]
1990	Glass/ITO/CdS/CdTe/Cu–Au device; reduced bandgap and enhanced $J_{sc}$ (28 mA/cm2).	13.1 (P)	Morris et al. (Queensland Univ.) [58]
1991–1992	CBD prep. of CdS using thiourea; optimized CSS p-CdTe temperature and composition.	13.4–14.6 (P)	Chu et al. (Univ. of South Florida) [59,60]
1993	CBD n-CdS buffer and CSS p-CdTe absorber optimization; $V_{oc}$ reached 843 mV and FF of 0.745.	15.8 (P)	Britt and Ferekides (USF) [61]
1997	MOCVD for ultra-thin CdS (50 nm) on ITO; CdTe deposited via CSS on glass substrate.	16.0 (P)	Ohyama et al. (MBI) [53]
2001	Polycrystalline CTO/ZTO/CdS/CdTe formation; achieved record $J_{sc}$ (26 mA/cm2) and FF (0.773).	16.5 (P)	Wu et al. (NREL) [62]
2006–2016	Analysis of near-ideal bandgap (1.45 eV); VTD/CSS refinement leading to commercial records.	17.3–22.1 (P)	Morales/First Solar [63]
2017–2020	Long-term field reliability and accelerated aging validation.	—	NREL/First Solar [64]
2022	Group-V (As) doping in CdSeTe/CdTe for reduced recombination.	—	Danielson et al. [65]
2023	Cu-free CdSeTe engineering via Group-V doping.	22.3 (P)	Mallick et al. [66]
2024	New NREL-certified record for small-area CdTe research cells.	23.1 (P)	First Solar/NREL [67]
2025	Global cross-check of independently confirmed efficiency records and updates across PV technologies (including thin-film categories) for contextual benchmarking.	—	Green et al.; Solar Cell Efficiency Tables (Version 66) [39]

Note: B: substrate; P: superstrate; MEI: Matsushita Electric Industrial; CNRS: French National Center for Scientific Research; MBI: Matsushita Battery Industrial; NPI: National Polytechnic Institute.

Table 1. Sequential developments and historical milestones of CdTe/CdSeTe solar cells.

Year	Applied Methodology/Innovation	$\mathit{\eta }$ (%)	Developed By/Significance
1972	First CdTe/CdS cell via graded gap thin film; 3-step VPD for p-CdTe and vacuum evaporation for n-CdS.	6.0 (B)	Bonnet and Rabenhorst [51]
1976	Ceramic thin film (20 $\mathsf{\mu }$m) via screen printing; achieved 0.2 $\mathsf{\Omega }$-cm resistivity.	8.1 (P)	Nakayama et al. (MEI) [52]
1982	Thin poly-CdS/CdTe via CSS on SLG/In2O3; oxygen induction for shallow junction ($V_{oc}=750$ mV).	10.5 (P)	Tyan and Perez (Eastman Kodak) [53]
1982	ITO/CdTe and CdS/CdTe via E-beam evaporation; Br-in-methanol etching and H-heating surface treatments.	7.5–10.5 (P)	Werthen et al. (Stanford Univ.) [54]
1982	Prep. of p-CdTe on n-CdTe via VTD; optimized shallow $p^{+}n{n}^{+}$ form ($V_{oc}=850$ mV).	11.0 (P)	Cohen et al. (CNRS) [55]
1983–1984	Cu-addition in C paste to reduce series resistance; achieved high $J_{sc}$ (22 mA/cm2) via screen printing.	12.8 (P)	Kuribayashi and Matsumoto (MEI) [56,57]
1990	Glass/ITO/CdS/CdTe/Cu–Au device; reduced bandgap and enhanced $J_{sc}$ (28 mA/cm2).	13.1 (P)	Morris et al. (Queensland Univ.) [58]
1991–1992	CBD prep. of CdS using thiourea; optimized CSS p-CdTe temperature and composition.	13.4–14.6 (P)	Chu et al. (Univ. of South Florida) [59,60]
1993	CBD n-CdS buffer and CSS p-CdTe absorber optimization; $V_{oc}$ reached 843 mV and FF of 0.745.	15.8 (P)	Britt and Ferekides (USF) [61]
1997	MOCVD for ultra-thin CdS (50 nm) on ITO; CdTe deposited via CSS on glass substrate.	16.0 (P)	Ohyama et al. (MBI) [53]
2001	Polycrystalline CTO/ZTO/CdS/CdTe formation; achieved record $J_{sc}$ (26 mA/cm2) and FF (0.773).	16.5 (P)	Wu et al. (NREL) [62]
2006–2016	Analysis of near-ideal bandgap (1.45 eV); VTD/CSS refinement leading to commercial records.	17.3–22.1 (P)	Morales/First Solar [63]
2017–2020	Long-term field reliability and accelerated aging validation.	—	NREL/First Solar [64]
2022	Group-V (As) doping in CdSeTe/CdTe for reduced recombination.	—	Danielson et al. [65]
2023	Cu-free CdSeTe engineering via Group-V doping.	22.3 (P)	Mallick et al. [66]
2024	New NREL-certified record for small-area CdTe research cells.	23.1 (P)	First Solar/NREL [67]
2025	Global cross-check of independently confirmed efficiency records and updates across PV technologies (including thin-film categories) for contextual benchmarking.	—	Green et al.; Solar Cell Efficiency Tables (Version 66) [39]

Note: B: substrate; P: superstrate; MEI: Matsushita Electric Industrial; CNRS: French National Center for Scientific Research; MBI: Matsushita Battery Industrial; NPI: National Polytechnic Institute.

The historical trajectory and technological maturation of CdTe-based thin-film photovoltaics are summarized in Table 1. The timeline shows a sustained rise in conversion efficiency from early CdS/CdTe heterojunction devices to present-day CdSeTe/CdTe architectures, which exceed 23% efficiency. Early progress was enabled by process innovations such as close-space sublimation and chemical bath deposition, whereas more recent improvements have been driven by selenium alloying, oxide front-interface engineering, and Group-V doping strategies that improve carrier lifetime, hole density, and junction quality [38,49].

Despite its commercial dominance, several technical hurdles prevent CdSeTe/CdTe from reaching its theoretical efficiency limit of ∼30%. A significant challenge remains the “voltage deficit”, where $V_{oc}$ is limited by non-radiative recombination at the interfaces and low carrier concentration [36,68]. Furthermore, the traditional use of copper ($Cu$) for p-type doping, while beneficial for conductivity, introduces long-term stability concerns due to the rapid diffusion of $Cu$ atoms, which leads to device degradation [69]. Additionally, the formation of a low-resistance ohmic back contact is complicated by the high work function of p-type CdTe, often resulting in an undesirable Schottky barrier.

Table 2. Performance parameters of CdTe/CdSeTe thin-film cells.

Parameter	Symbol	Value	Context/Comparison
Open-circuit voltage	$V_{OC}$	917 mV	Major breakthrough. Previous generations were limited to ∼850–880 mV. The rise is attributed to improved hole density ($p>{10}^{16}$) via Arsenic doping.
Short-circuit current	$J_{SC}$	32.0 mA/cm2	Approaching the theoretical limit for 1.4 eV bandgap. Achieved via CdSeTe grading.
Fill factor	$FF$	∼79.5%	Still lagging behind silicon (∼84%) and GaAs (∼86%). Limited by series resistance and contact non-idealities.
Efficiency	$\eta$	23.1%	Verified by NREL. The highest efficiency for any polycrystalline thin-film technology, closing in on CIGS.
Cell area	A	0.45 cm2	Standard aperture area for research champion cells.
Dopant	–	Arsenic (As)	Replacement of copper (Cu) improves long-term stability and voltage.

provides a performance summary of research-scale polycrystalline thin-film solar cells, highlighting a verified efficiency of 23.1%. This performance level is supported by a $V_{OC}$ of 917 mV, achieved through the use of arsenic (As) doping to improve hole density. Despite these advancements in voltage and current, the fill factor (FF) of 79.5% remains lower than that of silicon or GaAs technologies due to existing series resistance and contact non-idealities.

3.2. Amorphous Silicon (a-Si)

Amorphous silicon (a-Si), specifically in its hydrogenated form (a-Si:H), is a well-established thin-film photovoltaic (PV) technology that distinguishes itself from crystalline silicon (c-Si) through its disordered atomic structure. This lack of long-range order results in a direct-like bandgap typically ranging from $1.7$ to $1.8$ eV, which is significantly wider than that of c-Si [70,71]. Consequently, a-Si:H exhibits a high absorption coefficient across the visible spectrum, allowing for the capture of substantial incident light with active layers less than 1 $\mathsf{\mu }$m thick—approximately 1/100th the thickness of a standard silicon wafer [72,73].

The fundamental device architecture utilized is the p-i-n (or n-i-p) junction, as depicted in Figure 7. Unlike crystalline cells, where carrier transport is diffusion-driven, the low carrier mobility and short diffusion lengths in amorphous silicon necessitate a drift-driven mechanism. This is achieved by sandwiching a thick intrinsic (i-type) layer between thin, heavily doped p-type and n-type layers, thereby establishing a strong internal electric field that sweeps photogenerated carriers toward the electrodes [70,74]. Beyond standalone thin-film modules, a-Si:H has become an integral component of silicon heterojunction (HJT) technology, where ultra-thin layers of intrinsic and doped a-Si:H are used to passivate the surfaces of n-type crystalline silicon wafers [3,14]. This combination reduces the surface recombination velocity to nearly zero, facilitating a $V_{oc}$ exceeding 740 mV and pushing commercial module efficiencies toward $25\%$ [2,14].

The performance of a-Si:H cells is defined by their “stabilized” efficiency, which accounts for the degradation occurring during the initial hundreds of hours of light exposure. While single-junction thin films achieve stabilized efficiencies around $10\%$, their integration into multi-junction (tandem) and hybrid heterojunction architectures has led to record-breaking results [2,74].

The historical progression and technological evolution of hydrogenated amorphous silicon ($\alpha$-Si:H) photovoltaics are summarized in

Table 3. Sequential development and historical milestones of a-Si solar cells.

Year	Applied Methodology/Innovation	$\mathit{\eta }$ (%)	Developed By/Significance
1976–77	First a-Si:H p-i-n cell developed via discharge doping; Schottky barrier formation with 1.1 eV height for Pt cells.	2.4 (S); 4.0 (S)	Carlson and Wronski (RCA Labs) [71,75]
1978	M-I-S junction with Schottky barrier and insulating layer to equilibrate the low work function of nickel.	4.8 (S)	Wilson and McGill (Heriot-Watt Univ.) [76]
1980	Fabricated 1.19 cm2 p-i-n structure to decrease loss characterizations.	6.1 (S)	Carlson (RCA Labs) [77]
1981–82	a-SiC:H via plasma decomposition; a-SiC:H/a-Si:H heterojunction with $V_{oc}=887$ mV and $J_{sc}=15.21$ mA/cm2.	7.14 (S); 8.04 (S)	Tawada et al. (Osaka Univ.) [78,79]
1986	Insertion of thin-film at the p-i interface to improve short-wavelength response and FF (0.771).	8.43 (S)	Arya et al. (Solarex Corp.) [80]
1986	Glass substrate without ARC; achieved high $V_{oc}$ (12,670 mV) and $J_{sc}$ (78.47 mA/cm2).	9.63 (S)	Yamazaki et al. (SEL) [81]
1992	Double-junction with dual-bandgap device utilizing appropriate deposition methods.	11.0 (M)	Guha et al. (USSC) [82]
1996–97	a-Si:H/a-Si:Ge dual-junction; triple-junction spectrum-splitting and TCO/p-n tunnel-junction improvements.	11.8 (M); 13.0 (M)	Yang et al. (USSC) [83,84]
2013	Triple-junction device developed using a-Si:H/$\mu$c-Si:H/$\mu$c-Si:H formation.	13.4 (M)	Kim et al. (LGEARI) [85]
2015–16	Diode/triode PECVD for rate-controlled a-Si:H; textured substrates and minimized light-induced degradation (4%).	10.22 (S); 14.04 (M)	Matsui and Sai (AIST) [86,87,88]
2017	HJT integration: a-Si used as passivation in silicon heterojunction (SHJ) cells.	26.7 (HJT)	Kaneka [13]—Critical role in c-Si records.
2021	Flexible/ultra-thin: development of a-Si cells on flexible substrates for wearables.	∼10 (flex)	Various researchers [89]—new form factors for IoT.
2021	Record for M6-size p-type HJT cells (a-Si:H passivation/contact layers).	26.07 (HJT)	LONGi; certified by ISFH
2022	Benchmark for full-size HJT (a-Si:H passivation/contact layers).	26.81 (HJT)	LONGi; verified by ISFH, utilizing electrically optimized nanocrystalline-silicon contacts [90,91]
2024	Laboratory HJT back contact (HBC) reaching theoretical ceilings.	27.30 (HBC)	LONGi; certified by ISFH, surpassing the 27% threshold via laser-patterning and silver-free metallization [92]
2025	Resource-efficient processing: record-low silver consumption (1.4 mg/Wp).	—	Fraunhofer ISE; achieved via copper-based metallization on the rear side to drive material sustainability [93]
2025	Peer-reviewed validation of 27%-class cells and 25.4% modules.	27.0 (cell); 25.4 (module)	Nature Communications; links lab and module [92]
2025	Development of transparent a-Si1−xCx:H for BIPV applications.	—	Renewable Energy; transparent absorbers [94]
2026	Industry claim of HJT movement using microcrystalline silicon.	26.92 (HJT)	Maxwell Tech; ISFH-certified (half-cell) [95]

. The timeline highlights a transition from early proof-of-concept p-i-n and Schottky barrier structures in the late 1970s to complex multi-junction architectures that address the material’s inherent stability challenges, such as the Staebler–Wronski effect. Significant milestones include the introduction of wide-bandgap $\alpha$-SiC:H window layers in the 1980s and the development of triple-junction configurations using $\mu$c-Si:H in the 2010s to achieve higher stabilized efficiencies. Most recently, the role of $\alpha$-Si has evolved from a standalone absorber to a critical passivation layer in silicon heterojunction (HJT) cells and a primary candidate for flexible, ultra-thin power sources in the growing Internet of Things (IoT) ecosystem.

Table 4. Performance characteristics of triple-junction amorphous and microcrystalline silicon solar cells.

Parameter	Value	Unit	Description/Context
Cell structure	a-Si:H/$\mu$c-Si:H/$\mu$c-Si:H	–	Triple-junction thin-film stack on substrate
Open-circuit voltage ($V_{oc}$)	1.922	V	Exceptionally high voltage due to series connection of 3 junctions.
Short-circuit current ($J_{sc}$)	9.94	mA/cm2	Current constrained by the lowest-current sub-cell (current matching).
Fill factor (FF)	73.5	%	Indicates high quality of interfaces and tunnel junctions.
Conversion efficiency ($\eta$)	14.0	%	Stabilized efficiency (after 1000 h light soaking). Initial efficiency was >15%.
Active area	1.05	cm2	Designated illumination area.
Degradation ratio	∼4%	–	Significantly lower than single-junction a-Si (typically 10–30%).

details the technical parameters of a triple-junction thin-film silicon solar cell that utilizes a stacked architecture of amorphous (a-Si:H) and microcrystalline ($\mu$c-Si:H) silicon. This configuration yields a high $V_{OC}$ of 1.922 V by connecting three junctions in series, though the short-circuit current is limited to 9.94 mA/cm² to maintain current matching between sub-cells. A key highlight is the 14.0% stabilized efficiency, which demonstrates superior resistance to light-induced degradation compared to single-junction cells, showing a degradation ratio of only 4% after extensive light soaking.

A key limitation of hydrogenated amorphous silicon (a-Si:H) absorbers is the Staebler–Wronski effect (SWE), in which prolonged illumination induces metastable defect states—primarily silicon dangling bonds within the amorphous network [70,96]. The resulting increase in defect density enhances Shockley–Read–Hall recombination, shortens effective carrier lifetime, and typically causes an initial efficiency drop of about 15–$30\%$ before reaching a quasi-stabilized state [72,74].

Band offsets and bending at the a-Si:H(i)/c-Si interface establish the selective contact behavior required for efficient carrier extraction; however, SWE-induced gap states in a-Si:H (and related interface states) introduce additional recombination pathways that reduce collection efficiency and device performance. Common mitigation approaches include reducing the effective a-Si:H thickness to strengthen the internal field and shorten transport paths, employing hydrogen dilution (and optimized passivation) during deposition to improve bonding stability, and using more stable mixed-phase absorbers or contact layers (e.g., incorporating nanocrystalline silicon, $nc$-Si:H) to suppress light-induced degradation [73,97].

3.3. Copper Zinc Tin Sulfide (CZTS)

Copper zinc tin sulfide (Cu₂ZnSnS₄, CZTS) and its selenized analogs (Cu₂ZnSn(S,Se)₄, CZTSSe) are widely regarded as leading earth-abundant and low-toxicity alternatives to conventional thin-film photovoltaic materials. Composed exclusively of abundant elements and free from scarce indium or toxic cadmium, kesterite absorbers offer a promising route toward sustainable, large-scale photovoltaic deployment. The bandgap of CZTS-based compounds is tunable from approximately 1.0 eV for the pure selenide to about 1.5 eV for the pure sulfide, corresponding to a theoretical efficiency limit exceeding 30% under the Shockley–Queisser framework.

As of 2025, the certified world-record efficiency for laboratory-scale CZTSSe solar cells is 14.6% (with a best reported laboratory value of 14.9%), achieved by the Institute of Physics at the Chinese Academy of Sciences [98]. In parallel, pure sulfide CZTS devices have reached a certified efficiency of 13.2%, reported by the University of New South Wales, through the implementation of hydrogen-assisted defect passivation [16]. These advances reflect steady progress in addressing the intrinsic limitations of kesterite absorbers, particularly those associated with defect-related recombination and interface losses.

Table 5. Chronological evolution of efficiency of CZTS-based solar cells.

Year	Efficiency ($\mathit{\eta }$)	Organization	Material	Key Innovation/Milestone	Ref.
1997	0.66%	Nagaoka (Katagiri)	CZTS	First functional device; sputtering + sulfurization.	[15]
2008	6.70%	Nagaoka (Katagiri)	CZTS	Introduction of soaking processes.	[99]
2010	9.60%	IBM	CZTSSe	Solution processing (hydrazine) breakthrough.	[100]
2021	13.0%	NREL/U. Delaware	Ag-CZTSSe	Ag alloying breaks the 8-year stagnation.	[101]
2023	13.8%	IOP/CAS	CZTSSe	Certified record. Se partial-pressure tuning.	[102]
2024	14.6% (cert)	CAS/NREL	Ag-CZTSSe	Gradient bandgaps and multi-element alloying.	[103]
2025	14.9% (lab)	CAS	CZTSSe	Suppressing element inhomogeneity.	[103]
2025	13.2% (cert)	UNSW	CZTS	Sulfide record. Hydrogen passivation.	[103]

summarizes the historical evolution of CZTS-based solar cell efficiencies and the corresponding technological milestones, from the first observation of photovoltaic behavior in 1988 to present-day record devices approaching 15% efficiency. The timeline illustrates a shift from early sulfurization and sputtering-based fabrication routes toward more sophisticated solution processing, alloying strategies, and interface engineering. It further documents the resolution of a prolonged efficiency stagnation period through targeted compositional engineering, including Ag alloying and Ge incorporation.

The state of CZTS solar cell technology in 2025 reflects clear progress beyond the prolonged efficiency stagnation observed between 2015 and 2020. Recent gains have been primarily driven by advances in defect engineering, absorber compositional control, and interface optimization. Unlike superstrate thin-film technologies such as CdTe, CZTS devices employ a substrate-type configuration in which the metallic back contact is deposited first, followed by sequential growth of the functional layers and a transparent front contact. The modern CZTS device stack comprises several key layers.

Substrate and back contact: Soda-lime glass (SLG) is commonly used as the substrate and serves as a sodium (Na) source. During high-temperature annealing, Na diffuses into the absorber, contributing to grain boundary passivation and enhanced p-type conductivity. A bilayer molybdenum (Mo) back contact with a thickness of approximately 500–800 nm is deposited by sputtering. A thin MoS₂ interfacial layer forms during sulfurization and, when kept below ∼200 nm, provides quasi-ohmic contact; excessive thickness increases series resistance and reduces the fill factor.

Graded absorber layer: High-performance devices employ a compositionally graded Cu₂ZnSn(S,Se)₄ absorber with a thickness of 1.5–2.5 $\mathsf{\mu }$m. The absorber is typically Cu-poor (Cu/(Zn + Sn) ≈ 0.75–0.85) and Zn-rich (Zn/Sn ≈ 1.1–1.2) to suppress deep defects and secondary phases. Partial substitution of Cu with Ag reduces Cu–Zn disorder and band tailing, while Ge incorporation promotes grain growth and improves back-contact band alignment. Additional bandgap grading via S/Se compositional variation assists carrier separation and collection.

Buffer layer: Cadmium sulfide (CdS), deposited by chemical bath deposition (50–80 nm), remains the reference buffer material. However, Cd-free alternatives, such as Zn(O,S), ZnSnO, and (Zn,Mg)O, are increasingly used because their tunable band alignment reduces interface recombination. These buffers typically provide a small positive conduction-band offset, lowering recombination velocities to below ${10}^3$ cm/s.

Window layer and front contact: A thin intrinsic ZnO (i-ZnO) layer suppresses shunting, followed by a transparent conductive oxide such as Al-doped ZnO (AZO) or indium tin oxide (ITO). This bilayer front contact combines high optical transparency with adequate lateral conductivity.

Hydrogen passivation: Post-deposition annealing in hydrogen-containing atmospheres enables hydrogen diffusion into the kesterite lattice, passivating dangling bonds and deep-level defects. Minority-carrier lifetimes increase from the sub-nanosecond regime to values exceeding 10 ns, enabling recent record efficiencies in pure-sulfide CZTS devices.

The certified 2025 CZTSSe record device exhibits an efficiency of 14.6%, with an $V_{OC}$ of ≈576 mV, a short-circuit current density of 35–36 mA/cm², and a fill factor of approximately 72–74%. The pure-sulfide record (13.2%) shows higher $V_{OC}$ values of 723–800 mV but lower current density. Despite this progress, a large $V_{OC}$ deficit, exceeding 600 mV, remains the dominant limitation for CZTS technology.

Under illumination, photogenerated electrons in p-type CZTS are transported across the heterojunction into the n-type buffer and collected at the front contact, while holes are extracted through the Mo back contact. Performance losses are primarily associated with Cu–Zn disorder, interface recombination at the CZTS/buffer junction, and deep Sn-related trap states that promote Shockley–Read–Hall recombination.

Table 6. Comparison of verified performance parameters for kesterite solar cells from CAS and UNSW (updated April 2026).

Parameter	CAS [103]	UNSW [104]	CAS (2026 Record [105])
Material	Pure sulfide (CZTS)	Pure sulfide (CZTS)	Sulfo-selenide (CZTSSe)
Efficiency ($\eta$)	13.2% (certified)	11.4% (certified)	15.45%(certified)
Voltage ($V_{oc}$)	831 mV	723 mV	602 mV
Current ($J_{sc}$)	22.8 mA/cm2	24.7 mA/cm2	34.1 mA/cm2
Fill factor (FF)	69.5%	64.0%	75.3%
Key innovation	Gradient bandgap	Hydrogen passivation	$L{i}_{2}Sn{S}_3$ interphase

compares representative high-performance kesterite devices reported by UNSW and CAS, illustrating trade-offs between high-bandgap pure-sulfide and lower-bandgap sulfo-selenide alloy systems.

3.4. Copper Indium Gallium Selenide (CIGS)

Copper indium gallium selenide (CIGS) photovoltaics have progressed from early laboratory-scale demonstrations to one of the most mature and high-performing thin-film photovoltaic technologies. Continuous advances in absorber engineering, particularly through alkali post-deposition treatment (PDT) and controlled alloying strategies, have enabled certified single-junction efficiencies exceeding 23.6%, placing CIGS among the highest-efficiency thin-film technologies and in direct competition with crystalline silicon [12]. In addition to high conversion efficiency, CIGS exhibits exceptional structural versatility. Flexible devices fabricated on polyimide or metallic foils have demonstrated certified efficiencies above 22.2%, enabling lightweight, conformable, and building-integrated photovoltaic (BIPV) applications [106,107]. Although the use of relatively scarce elements such as indium presents cost and supply considerations, CIGS remains highly attractive due to its radiation tolerance, long-term operational stability, and robust performance under diverse environmental conditions.

Figure 8 shows the substrate-type heterojunction configuration of a canonical CIGS device. It follows a employing either rigid soda-lime glass (SLG) or flexible polymer/metal substrates. A typical layer sequence consists of a substrate, a molybdenum (Mo) back contact, a p-type CIGS absorber, an n-type CdS buffer layer, and a transparent conductive oxide (TCO) front contact. The Mo back contact provides high electrical conductivity and thermal stability while promoting the formation of a thin interfacial ${\mathrm{MoSe}}_2$ layer that ensures favorable ohmic contact with the absorber [18,20].

The p-type Cu(In_1−xGa_x)Se₂ absorber, with a typical thickness of 1.5–2.5 $\mathsf{\mu }$m, constitutes the primary photoactive region. This polycrystalline chalcopyrite material offers a tunable bandgap ranging from approximately 1.0 to 1.7 eV through Ga compositional grading, allowing optimization of the trade-off between $V_{OC}$ voltage and photocurrent [27,108]. Proper bandgap engineering and defect passivation have enabled absorber layers that routinely support device efficiencies exceeding 15%.

In CIGS architecture, proper bandgap engineering is practically implemented through precise compositional grading, specifically by tuning the gallium-to-indium ratio, Ga/(Ga + In), throughout the depth of the absorber layer during the deposition process [109]. This strategy allows for the creation of a “double-graded” or “notch” bandgap profile. By increasing the gallium concentration toward the back contact, a quasi-electric back-surface field is established. This field effectively repels photo-generated electrons, thereby suppressing rear-interface recombination. Simultaneously, increasing the gallium content toward the front buffer-layer interface optimizes the $V_{OC}$ without significantly compromising the absorption of the solar spectrum in the bulk material, facilitating highly efficient carrier transport [110]. Furthermore, while CIGS absorbers inherently contain point defects and grain boundary dislocations, strategic defect passivation mitigates their detrimental impacts. The recent literature highlights the critical role of alkali post-deposition treatment (PDT) utilizing heavy alkali metal fluorides, such as KF, RbF, or CsF, as highly effective passivating materials. During PDT, these alkali elements segregate at the grain boundaries and the CIGS surface. The primary passivation mechanism involves neutralizing deep-level donor-like defects, such as indium-on-copper antisites (InCu), and mitigating copper vacancies (VCu). This chemical interaction often induces the formation of wide-bandgap secondary phases (e.g., KInSe2 or RbInSe2) at the interface, shifting the surface valence band downward. Consequently, trap-assisted Shockley–Read–Hall (SRH) recombination at the buffer–absorber interface is significantly suppressed, yielding substantial enhancements in both the $V_{OC}$ and the fill factor of the devices [111].

The operational performance of CIGS solar cells is highly sensitive to thermal variations, exhibiting both reversible electrical changes and irreversible thermomechanical stresses. Under thermal heating, the device experiences a reversible decline in power conversion efficiency. This loss is primarily driven by a substantial decrease in the $V_{OC}$. At elevated temperatures, the intrinsic carrier concentration increases, which elevates the reverse saturation current (dark current) and accelerates non-radiative recombination rates. Concurrently, thermal expansion induces a slight narrowing of the material’s bandgap, which marginally increases the short-circuit current but is insufficient to compensate for the severe voltage degradation. Nevertheless, CIGS thin-film modules typically exhibit a more favorable temperature coefficient for maximum power compared to traditional crystalline silicon, rendering them relatively robust in high-temperature operating environments [112].

Beyond transient operational fluctuations, continuous thermal stress and cyclical heating and cooling (thermal cycling) can induce permanent thermomechanical degradation. The CIGS device architecture is a stack of distinct thin films, including the glass or polyimide substrate, molybdenum back contact, CIGS absorber, CdS buffer, and ZnO-based transparent conducting oxide window layer. Each of these materials possesses a significantly different coefficient of thermal expansion (CTE). Repeated thermal cycling exacerbates mechanical shear stresses at these heterointerfaces. Over time, this mismatch can lead to microcracking, increased series resistance, or catastrophic delamination. Furthermore, prolonged exposure to elevated temperatures can activate unwanted diffusion processes, such as the migration of copper within the absorber or the interdiffusion of elements across the buffer interface, which introduces deep-level trap states and permanently degrades device longevity [113,114].

To form the heterojunction, an n-type cadmium sulfide (CdS) buffer layer is commonly deposited by chemical bath deposition (CBD). In contemporary device architectures, Cd-free buffer stacks such as Zn(O,S,OH)_x/Zn_0.8Mg_0.2O are increasingly adopted to mitigate toxicity concerns and enhance short-wavelength response [11].

In addition to Zn-based buffer layers, alternative oxide- and sulfide-based buffer materials, including metal-oxide and metal-sulfide compounds, have been investigated to further improve device performance and environmental compatibility. For example, Hossain [115] demonstrated the fabrication and characterization of CIGS solar cells employing indium sulfide (In₂S₃) buffer layers deposited using physical vapor deposition (PVD). The use of In₂S₃ as a buffer layer offers several advantages, including reduced toxicity compared to cadmium-based materials, improved optical transparency, and favorable conduction-band alignment with the CIGS absorber. Such oxide- and sulfide-based buffer technologies have attracted increasing attention due to their potential to enhance carrier transport, minimize interface recombination, and support the development of environmentally sustainable thin-film photovoltaic devices.

Following the formation of the buffer and absorber interfaces, the front contact typically comprises an intrinsic ZnO layer followed by aluminum-doped ZnO (AZO), which together provide high optical transparency and lateral conductivity. Industrial fabrication employs scalable processes, including co-evaporation, sputtering, electrodeposition, and rapid thermal treatments.

Performance improvements in CIGS technology have been closely linked to alkali PDT using potassium (K), rubidium (Rb), and cesium (Cs), which effectively passivate grain boundaries and interface defects, leading to enhanced carrier lifetimes and diode quality [20,107]. Solar Frontier reported a certified efficiency of 23.35% for a Cd-free CIGS device in 2019 [11]. More recently, in 2024, a world-record efficiency of 23.64% was achieved for a $(Ag,Cu)(In,Ga)S{e}_2$ (ACIGS) solar cell through high-concentration silver alloying and a “hockey-stick” gallium compositional profile, yielding open-circuit voltages above 760 mV and short-circuit current densities approaching 40 mA/cm² [12,116].

The historical evolution and key technological milestones of CIGS photovoltaics are summarized in

Table 7. Sequential development and historical milestones of CIGS solar cells.

Year	Applied Methodology/Innovation	$\mathit{\eta }$ (%)	Developed By/Significance
1976–77	First CIGS cells (p-CuInSe2/n-CdS) via tungsten–halogen illumination; vacuum deposition of ternary/binary materials.	4.5 (R), 5.7 (R)	Kazmerski et al. (Univ. of Maine) [117]
1980–82	Polycrystalline CdS/CuInSe2 via continuous evaporation; sputtering on cheap substrates; mixed ZnxCd1−xS for $V_{oc}$ enhancement.	5.7 (R), 7.5 (R), 9.5 (R), 10.6 (R)	Mickelsen and Chen (Boeing Aerospace) [118,119,120,121]
1985	Introduced ZnO with thin (Cu,Zn)S or CdS; ZnO ARC enhanced $J_{sc}$ by 25%.	11.2 (R)	Potter et al. (ARCO Solar) [122]
1990	Polycrystalline ZnO/CdZnS/CuInGaSe2 via PVD, CVD, and reactive sputtering.	12.5 (R)	Devaney et al. (Boeing) [123]
1993	CuInGaSe2 on flexible Mo/Ti/Al via E-beam evaporation and H2Se gas reaction.	8.3 (F)	Başol et al. (ISET) [124]
1994–95	Co-evaporation of In/Ga/Se; increased bandgap and $V_{oc}$ (654 mV).	15.9 (R), 17.1 (R)	Gabor and Tuttle et al. (NREL) [125,126]
1996	Polymeric substrate for low-temp CdS/ZnO; grid design modifications with i-ZnO.	9.3 (F), 17.7 (R)	Başol (ISET) [127]/Tuttle (NREL) [128]
1999	Cd-free cell via PVD and direct ZnO deposition; improved ZnO/CdS/Cu(In,Ga)Se2/Mo.	15 (R), 18.8 (R)	Contreras et al. (NREL) [129]
2003	Enhanced ZnO/CdS/CuInGaSe2 via advanced characterization.	19.2 (R)	Ramanathan et al. (NREL) [130]
2005	Improvement via reduced diode saturation current and space-charge recombination.	19.5 (R)	Contreras et al. (NREL) [131]
2008	Reduced recombination via Ga-poor (In-rich) film termination.	19.9 (R)	Repins et al. (NREL) [132]
2010	Max efficiency for flexible Mo substrate (3-stage co-evaporation); 20.3% for rigid cells.	14.6 (F), 20.3 (R)	Niki (AIST) [133]/Jackson (ZSW) [134]
2011–12	Reduced SS substrate temperature; enhanced CIGS growth performance.	17.1 (F), 17.7 (F)	Reinhard and Pianezzi (EMPA) [5,135]
2013	Highest efficiency for polymer foil substrate.	20.4 (F)	EMPA and FhG-ISE [20]
2013–14	Static co-evaporation with Zn(O,S) buffer; K-doped film to increase Ga.	20.4 (R), 20.8 (R)	Powalla and Jackson (ZSW) [136,137]
2014	High deposition rates and improved absorber co-evaporation; record 21.7%.	21 (R), 21.7 (R)	Herrmann (Solibro) [138]/ Jackson (ZSW) [139]
2015–16	Improved CIS absorber layer; alkali PDT (Rb, Cs) for improved diode quality.	22.3 (R), 22.6 (R)	Solar Frontier [140]/Jackson (ZSW) [141]
2017	Cs-alkali treatment and absorber modification for wider bandgap.	22.9 (R)	Wu et al. (AIST) [142]
2019	Roll-to-cell process on SS foil; record Cd-free double-buffer Cu(In,Ga)(Se,S)2.	20.56 (F), 23.35 (R)	Bayman (NREL) [61]/ Nakamura (AIST) [11]
2022	Certified flexible CIGS record on polymer film; improved device stack and processing on lightweight substrate for high-efficiency flexible PV.	22.2 (F)	EMPA; independently certified measurement (flexible CIGS record) [143,144]
2024	World-record single-junction CIGS via high-concentration Ag alloying and steep back-contact Ga grading in (Ag,Cu)(In,Ga)Se2 (ACIGS), enabling improved $V_{oc}$ and reduced recombination losses.	23.64 (R)	Keller et al. (Nature Energy); certified efficiency for ACIGS [12]
2024	Narrow-bandgap CIGSe engineered for tandem applications; certified record efficiency with record-low $V_{oc}$ deficit, supporting high-efficiency multi-junction integration strategies.	20.26 (R)	Zhang et al. (Nature Communications); certified narrow-bandgap CIGSe for tandem relevance [145]
2025	Updated global benchmarks consolidating independently confirmed efficiencies for CIGS and other PV technologies.	—	Green et al., Solar Cell Efficiency Tables (Version 66) [39]
2025	Official efficiency cross-check resource reflecting latest confirmed records across PV technology classes (including CIGS).	—	NREL Best Research-Cell Efficiency Chart (PDF) [146]

Note: R: rigid substrate; F: flexible substrate. ISET: International Solar Electric Technology; NREL: National Renewable Energy Laboratory; ZSW: The Center for Solar Energy and Hydrogen Research; EMPA: Swiss Federal Laboratories for Materials Science and Technology.

. The progression from early p-n heterojunction devices to modern high-efficiency architectures highlights the critical roles of bandgap grading, interface engineering, and alkali PDT. The table further distinguishes advances on rigid substrates from those achieved on flexible platforms, underscoring the broad applicability of CIGS technology across both conventional and emerging photovoltaic markets.

Table 8. Efficiency records and device specifications for CIGS photovoltaics.

Device Category	Institution	Efficiency ($\mathit{\eta }$)	Area (cm2)	Source
World record (rigid)	Uppsala Univ.	23.64%	1.00 (aperture)	[12]
Flexible record	Empa	22.2%	∼0.5	[144]
Module record	Avancis	20.3%	527 (aperture)	[147]

summarizes the performance metrics for three key categories of copper indium gallium selenide (CIGS) photovoltaic technology. A laboratory-scale rigid cell developed by Uppsala University holds a record efficiency of 23.64% on a 1.00 cm² aperture area. Empa has demonstrated high-performance flexible CIGS technology, reaching an efficiency of 22.2% on a smaller scale. In contrast, the module-scale device from Avancis maintains an efficiency of 20.3% over a significantly larger aperture of 527 cm². This progression highlights the slight efficiency trade-off required when transitioning from optimized laboratory cells to commercial-scale modules.

It is important to distinguish between laboratory-scale, module-scale, and real-world system efficiencies when evaluating photovoltaic performance. Laboratory efficiencies typically represent optimized small-area cells under controlled conditions and therefore indicate the theoretical performance limits of a given technology. In contrast, module-level efficiencies reflect larger-area devices that include interconnection losses, material non-uniformities, and packaging effects. Real-world system efficiencies are often lower due to environmental factors such as temperature variations, shading, dust accumulation, and degradation over time. For most thin-film photovoltaic systems, operational efficiencies in field conditions are typically several percentage points lower than laboratory-record values, emphasizing the importance of scalable device engineering and durability-focused design strategies. Furthermore, system-level efficiencies at real-world implementation scales are influenced by additional balance-of-system losses, including inverter inefficiencies, wiring resistance, and installation-related performance variations.

Despite these advancements, the industry faces persistent technical and economic challenges. A primary technical hurdle is the $V_{oc}$ deficit observed in wide-bandgap CIGS compositions; as the gallium content increases to raise the bandgap for tandem applications, the efficiency often plateaus due to non-radiative recombination at deep-level defects and hetero-interfaces [108,148]. Economically, due to the rarity of constituent materials like indium, CIGS cells are typically costlier than other thin-film PVs. The reliance on indium poses long-term sustainability concerns, while the toxicity of traditional cadmium-based buffers continues to invite regulatory scrutiny [18].

3.5. Organic

Organic photovoltaics (OPVs) have undergone a radical transformation over the past three decades, evolving from early inefficient bilayer structures into highly tuned, morphologically complex bulk heterojunctions (BHJs). Unlike inorganic thin films that generate free charge carriers directly upon photon absorption, OPVs generate tightly bound electron–hole pairs (excitons) characterized by high binding energies (0.3–0.5 eV). To facilitate charge dissociation, the active layer must comprise a carefully engineered interpenetrating network of electron-donating and electron-accepting organic semiconductors.

Historically, OPV research was dominated by fullerene-based acceptors, predominantly PCBM (${\mathrm{PC}}_{61}\mathrm{BM}$ and ${\mathrm{PC}}_{71}\mathrm{BM}$). While fullerenes provided excellent electron mobility and favorable phase separation when blended with donor polymers like P3HT or PTB7, they imposed severe theoretical limits on device performance. Fullerenes suffer from weak optical absorption in the visible and near-infrared (NIR) spectra, limited energy-level tunability, and a high driving force requirement for charge separation, which inherently limited the $V_{oc}$ [149]. Furthermore, fullerene derivatives are prone to thermal dimerization and severe morphological degradation under continuous heat and light, severely curtailing their utility in solar thermal applications.

The most significant paradigm shift in contemporary OPV research and the primary driver of recent efficiency surges has been the transition to non-fullerene acceptors (NFAs). The introduction of A-D-A (acceptor–donor–acceptor) type small molecules, most notably the ITIC family and the subsequent Y-series (e.g., Y6), shattered previous performance ceilings [150]. NFAs exhibit robust and tunable optical absorption extending well into the NIR spectrum (up to 900–1000 nm), highly adjustable highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) levels, and non-isotropic crystalline packing that enhances charge transport while minimizing non-radiative recombination losses [149].

Table 9. Sequential development and historical milestones of organic solar cells.

Year	Applied Methodology/Innovation	$\mathit{\eta }$ (%)	Developed By/Significance
1986	First bilayer heterojunction organic solar cell (copper phthalocyanine and perylene derivative).	∼1.0	Tang (Eastman Kodak) [151]
1995	Introduction of the bulk heterojunction (BHJ) concept, blending MEH-PPV with $C_{60}$ derivatives.	2.9	Yu et al./Heeger Group [152]
2005	P3HT:PCBM blends optimized via thermal annealing, establishing the standard fullerene baseline.	∼5.0	Ma et al. [153]
2015	Introduction of ITIC, demonstrating the viability of high-performance non-fullerene acceptors.	6.8	Lin et al. [154]
2019	Development of the Y6 NFA, enabling highly efficient charge generation with very low voltage loss.	15.7	Yuan et al. [150]
2021	Optimization of PM6:Y6 derivatives (e.g., L8-BO) and ternary blend architectures.	18.2	Jin et al. [155]
2024+	Refined multi-fibril network morphologies and tandem OPV structures.	>19.4	Current NREL benchmarks [146]

outlines the historical milestones of OPV development, demonstrating the rapid acceleration in efficiency driven by NFA integration, pushing the technology from single-digit efficiencies into direct competition with established thin-film technologies.

Driven by NFA innovations, the power conversion efficiency (PCE) of laboratory-scale, single-junction OPVs has officially surpassed $19.4\%$, with tandem configurations exceeding $20\%$ [146]. However, a critical distinction must be made between spin-coated laboratory cells operating at miniature areas (<0.1 ${\mathrm{cm}}^2$) and scalable, mass-manufactured modules. The grand challenge of OPV commercialization lies in translating these high efficiencies through continuous, solution-based additive manufacturing techniques such as slot-die coating, roll-to-roll (R2R) processing, and electrohydrodynamic atomization (EHDA).

When scaled to module sizes (>20 ${\mathrm{cm}}^2$), state-of-the-art printed NFA modules consistently achieve stabilized efficiencies between $12\%$ and $14.46\%$ [156]. The slight drop from laboratory records is attributed to increased sheet resistance in transparent conductive electrodes and inherent coating inhomogeneities over large areas.

Despite this efficiency gap, printed OPVs offer unparalleled form-factor advantages for weight-sensitive and conformal integrations. For instance, Saravanapavanantham et al. recently demonstrated a highly scalable, printed OPV module deposited onto an ultra-thin polyethylene terephthalate (PET) substrate [157]. While the module efficiency was optimized for mechanical resilience rather than peak performance (yielding $5.2\%$ to $7.1\%$ depending on the electrode), the resulting fabric-integrated PV system achieved an extraordinary power-to-weight ratio of $370\mathrm{W}/\mathrm{kg}$. This metric represents a power density approximately eighteen times greater than that of conventional rigid crystalline silicon cells, highlighting the specific niche of OPVs in advanced architectural, portable, and aerospace systems.

A major focus of modern renewable infrastructure is the hybridization of photovoltaics with thermal management and energy storage. While standard silicon modules suffer severe efficiency degradation at elevated temperatures, OPVs possess unique thermodynamic, thermo-mechanical, and morphological properties that fundamentally alter their behavior in photovoltaic–thermal (PV/T) and building-integrated (BIPV) systems.

Organic thin films have an inherently low thermal mass. The active layers are typically 100–300 nm thick, and when printed on polymer substrates (e.g., PET or PI), the entire module possesses a specific heat capacity and thermal inertia drastically lower than glass-encapsulated inorganic cells. In a coupled PV/T fluid-loop system, this low thermal mass prevents the solar cell from acting as a thermal bottleneck. The OPV layer reaches thermal equilibrium with the heat-extraction fluid almost instantaneously, allowing for highly responsive thermal regulation and minimizing thermal lag during intermittent cloud cover or fluctuating solar irradiance.

Conventional crystalline silicon exhibits a negative temperature coefficient for efficiency, typically degrading by −0.4%/°C to −0.5%/°C as temperatures rise above 25 °C, primarily due to increased intrinsic carrier concentration dropping the $V_{oc}$. Conversely, charge transport in OPVs relies on thermally assisted inter-site hopping mechanisms. Consequently, as the temperature of an OPV module increases, the charge carrier mobility often increases. While the $V_{oc}$ in OPVs still declines with heat, the short-circuit current ($J_{sc}$) and fill factor (FF) frequently improve, leading to a net temperature coefficient that is near zero, or in some specialized NFA blends, slightly positive up to roughly 60–70 °C [158]. This thermo-optic property makes OPVs uniquely advantageous for integration into BIPV/T façades, where the module is expected to operate continuously at elevated temperatures.

Despite favorable temperature coefficients, the absolute thermal limit of OPVs is governed by morphological stability. Under prolonged heat (>80 °C), organic blends are susceptible to spinodal decomposition, where the finely mixed donor and acceptor phases over-aggregate into macro-domains, drastically reducing the interfacial area required for exciton dissociation. Traditional fullerenes fail rapidly under these conditions. However, advanced NFA systems—particularly those utilizing halogen-rich end groups or high glass-transition ($T_g$) polymer donors—demonstrate exceptional morphological “locking” mechanisms. Recent studies incorporating UV–ozone-treated ${\mathrm{MoO}}_x$ hole-transporting layers or cross-linkable additives have successfully suppressed high-temperature phase segregation, extending the $T_{80}$ lifetime (the time required to degrade to $80\%$ of initial efficiency) to thousands of hours under continuous 80 °C thermal stress [158,158].

The mechanical flexibility and low-temperature (<150 °C) solution-processing requirements of OPVs enable seamless, monolithic integration with thin-film energy storage devices. Unlike CdTe or CIGS, which require high-temperature vacuum deposition that would melt polymer separators, OPVs can be printed directly alongside or atop flexible solid-state batteries (e.g., printed Li-S or Zinc-ion batteries) and planar supercapacitors. This additive manufacturing synergy allows for the roll-to-roll fabrication of “photo-rechargeable” power sheets that natively smooth solar intermittency without external wiring or rigid housing, a critical enabler for smart-building sensors and wearable autonomous electronics.

In organic photovoltaics (OPVs), their thermo-mechanical properties offer advantages for building-integrated photovoltaic–thermal (BIPV/T) facades and thermal energy storage systems. OPVs feature a low thermal mass, allowing them to reach thermal equilibrium with heat-extraction fluids rapidly. This prevents the solar module from acting as a thermal bottleneck, facilitating responsive heat transfer for building heating or hybrid cooling loops, while the modules themselves demonstrate operational stability at elevated temperatures [159]. To manage the morphological stability of active layers under thermal stress, established PV/T structural designs integrate phase-change materials (PCMs) as a cooling mechanism. These PCMs serve as passive thermal energy storage buffers: they absorb excess heat through latent melting during peak solar irradiation and release it as the ambient temperature drops [160,161]. Applying this architecture to OPVs helps protect non-fullerene acceptor (NFA) networks from thermal degradation and shifts useful thermal energy beyond peak daylight hours for continuous building climate control.

Thin-film solar cells have the potential to turn traditionally unconventional surfaces into energy sources. The realization of the potential of this technology will require solar cells that generate greater amounts of electricity, with greater usage flexibility, at a cheaper price [6,162]. Saravanapavanantham et al. [157] proposed an organic photovoltaic module (OPV) that aims to achieve these two targets. They fabricated a thin-film organic solar cell with a scalable and economical manufacturing process that can be utilized in weight-sensitive applications, such as drones and human skin. The research goal was to fabricate an OPV on a flexible substrate, and to use a scalable and cost-effective printing process. To validate the effectiveness of these two goals, they chose both a traditional glass substrate and a flexible heat-sensitized polyethylene terephthalate (PET) surface. For both of the substrates, they chose the traditional evaporated top electrode fabrication process, and a scalable solution-based printing process. Finally, they compared the efficiency as free-standing films and after transfer lamination onto fabric films. The final fabric–PV system had a thickness of about $50\mathsf{\mu }$m, with a density of 105 g/m² and output of 730 W/kg when freestanding, and 370 W/kg when deployed on fabric. It generated about eighteen times more power per unit weight than traditional crystalline-based cells.

Figure 9C shows the manufacturing process of a typical printed organic PV solar cell. For the glass substrate fabrication process, ITO glass was selected. And for the flexible substrate, heat-sensitized PET material was used. A class 1000 cleanroom was used to ensure good surface quality. After slitting and trimming, a fluoropolymer release layer was bladecoated, and then a thin film of parylene was added through CVD onto the glass or PET substrates. The slot-die (flood coat) process was used to add the ITO–silver–ITO bottom-electrode layer. An infrared laser created P1 scribes spaced at 13 mm to structure the modules, and a dielectric layer was slot-die-coated over the scribed areas to prevent shorts and to guide the placement of subsequent layers. This was followed by slot-die coating of the $Sn{O}_2$ ETL, $PV2000$:$PCBM$ organic photovoltaic layer, and $Mo{O}_3$ or $PEDOT$:$PSS$ HTL. It should be noted that both perovskites and organic photovoltaics (OPVs) are flexible photovoltaic materials that allow slot-die coating, spray coating, and ink-jet printing [163,164,165]. However, organic bulk heterojunction material was chosen as the photoactive material due to its non-toxicity. A composite of PEDOT and silver nanowires (AgNW) was slot-die-coated as the conductive layer, resulting in a high transmission (85%) and low sheet resistance (15–17 $\mathsf{\Omega }$/m²). This layer minimized roughness and achieved isotropic conductivity, essential for device stability.

The glass-based device was used as a free-standing solar cell, whereas flexible PET-based devices can be employed either as free-standing films or laminated onto textile substrates. A durable Dyneema composite fabric was used as the cell carrier. UV-curable or pressure-sensitive adhesives were employed to transfer-laminate the solar cells onto the fabric. The lamination layer mitigates stress-induced damage during bending [166] and compression [167]. Similar lamination-based fabrication strategies are now widely adopted for flexible and wearable electronic devices [168,169].

Table 10. Performance comparison of solar cells fabricated on glass substrates using evaporated and printed top electrodes under uncoated (glass) and parylene-coated encapsulation conditions; adapted from [157].

Metric	Evaporated + Uncoated	Evaporated + Parylene	Printed + Uncoated	Printed + Parylene
$V_{OC}$ [V]	0.76	0.75	0.75	0.76
$J_{\mathrm{SC}}$ [mA cm−2]	14.4	11.9	14.7	11.1
FF [%]	64.7	62.5	58.3	61.6
PCE [%]	7.1	5.6	6.5	5.2

shows that glass-based devices with evaporated top electrodes under uncoated conditions achieve the highest power conversion efficiency (PCE) of 7.1%. When parylene encapsulation is applied, the PCE decreases to 5.6%. Devices employing scalable printed top electrodes reach a PCE of 6.5% in the uncoated configuration and 5.2% after parylene coating. Overall, the results demonstrate that flexible and printed-electrode solar cells fabricated using scalable processes can achieve efficiencies comparable to those of conventional glass-substrate devices with evaporated electrodes.

3.6. Perovskite

Given the theoretical power conversion limitation of traditional crystalline silicon photovoltaics, metal halide perovskites have emerged as a promising class of semiconductor materials. Figure 10 shows the cell structure of a typical perovskite atom. These materials are defined by their ${\mathrm{ABX}}_3$ crystal structure, which mirrors the arrangement found in calcium titanate (${\mathrm{CaTiO}}_3$). In these structures, the A site is typically occupied by an organic cation such as methylammonium (${\mathrm{CH}}_3{\mathrm{NH}}_3^{+}$; ${\mathrm{MA}}^{+}$) or formamidinium ($\mathrm{CH}{\left({\mathrm{NH}}_2\right)}_2^{+}$; ${\mathrm{FA}}^{+}$), or an inorganic cation like cesium (${\mathrm{Cs}}^{+}$). The B site is usually a divalent metal cation, most commonly lead (${\mathrm{Pb}}^{2+}$) or tin (${\mathrm{Sn}}^{2+}$), and the X site is a monovalent halide anion such as iodide (${\mathrm{I}}^{-}$), bromide (${\mathrm{Br}}^{-}$), or chloride (${\mathrm{Cl}}^{-}$) [170]. The rapid development of perovskite solar cells (PSCs) is evidenced by the progress in certified power conversion efficiency (PCE), which has increased from 3.8% in 2009 to a current single-junction record of 27.3% in early 2026 [39]. Furthermore, perovskite/silicon tandem configurations have attained 34.85% efficiency, exceeding the Shockley–Queisser limit of single-junction silicon [171].

Table 11. Certified perovskite efficiency records as of 2025–2026.

Technology Category	Configuration	Certified PCE (%)	Institutional Leader	Year
Single-junction perovskite	Small area	27.3 [39]	SooChow Univ./UNSW	2025
Perovskite–silicon tandem	Monolithic 2T	35.0 [171]	LONGi Solar	2025
Perovskite–silicon tandem	Large area	33.0 [171]	LONGi Solar	2025
Perovskite–perovskite tandem	All-perovskite (small area)	30.1 [171]	Nanjing University	2024
Perovskite–CIGS tandem	Flexible monolithic	27.3 [173]	Nanjing University	2025
Triple junction (P-P-Si)	Monolithic 3-junction	27.06 [171]	University of Sydney	2025

shows the efficiency gains in PSCs, which are fundamentally linked to the unique optoelectronic properties of the perovskite lattice. Unlike the rigid covalent bonding found in silicon, the ionic nature of the perovskite crystal results in a soft lattice that exhibits high defect tolerance [170]. This structural flexibility allows the material to maintain performance levels despite the presence of point defects that would typically act as recombination centers in other semiconductors. Perovskites also possess high absorption coefficients, enabling thin films of less than a micrometer to capture a significant portion of incident sunlight, which reduces material consumption. A critical advantage of the perovskite system is its bandgap tunability. By altering the chemical composition of the A, B, and X components, researchers can shift the absorption window from approximately 1.15 eV to over 3.0 eV [170]. This tunability facilitates the development of tandem solar cells. In a monolithic two-terminal (2T) device, a wide-bandgap perovskite top cell captures high-energy photons, while an underlying narrow-bandgap silicon or CIGS bottom cell absorbs the infrared portion of the spectrum, minimizing thermalization losses and maximizing spectral utilization [173].

To address challenges related to stability and charge extraction, several engineering strategies have been developed, focusing on interfaces, dopants, and thermal resilience. The interface between the perovskite absorber and charge transport layers is often the primary site for charge recombination. Interface engineering using self-assembled monolayers (SAMs), such as Me-4PACz, has become essential for inverted (p-i-n) structures [174]. Recent advancements include the use of asymmetric self-assembled molecules, such as HTL201, which provide improved coverage on textured silicon surfaces and reduce recombination at buried interfaces [174]. Additionally, p-type antimony-doped tin oxides (ATO) have been utilized to replace NiO as the interlayer; the chemical stability of ATO prevents undesired redox reactions at the interface while maintaining superior electrical conductivity [170]. Passivation strategies further mitigate defects. Techniques such as a targeted two-step immersion-cleaning process allow for the selective passivation of defects while removing residual agents that might otherwise introduce secondary instabilities [175]. Furthermore, the design of spatial isomers, such as tris(p-chlorophenyl)benzene (TCBP), enables precise coordination with iodide vacancies, which has resulted in certified efficiencies of 25.02% and enhanced operational longevity [175].

Dopant engineering has evolved from traditional hygroscopic materials to more stable alternatives. Metallocenium salts, such as ferrocenium dopants, have been utilized to enhance hole extraction by up to 45% while improving moisture resistance compared to conventional LiTFSI-based dopants [176]. Other innovations include dual-doping methods using zinc and formate additives to synergistically improve optical properties and thermal stability while reducing defect densities [170]. Doping with magnesium (Mg) in ZnO electron transport layers has also been shown to improve optical tunability and reduce defect-related trap densities [170].

Thermal engineering is critical for applications in extreme environments. A dual-molecular reinforcement strategy has been reported to address thermal expansion mismatches between the perovskite and its substrate. This involves incorporating alpha-lipoic acid to create a polymerized network at grain boundaries and using sulfonium-based derivatives (DMSLA) to anchor the perovskite to the electrode [177]. This “anchored net” allows the layers to adapt to temperature swings between −80 °C and 80 °C, with devices retaining 84% of their efficiency after rigorous thermal cycling [177].

Due to the toxicity of lead, research into lead-free alternatives continues. ${\mathrm{Cs}}_2{\mathrm{AgBiBr}}_6$ has shown potential due to its material stability and long carrier lifetime, though its efficiency remains lower than lead-based counterparts [178]. Experimental records for hydrogenated ${\mathrm{Cs}}_2{\mathrm{AgBiBr}}_6$ have reached 6.37%, while numerical simulations for all-perovskite lead-free tandems using antimony-doped ${\mathrm{Cs}}_2{\mathrm{AgBiBr}}_6$ and tin-based absorbers suggest theoretical potentials as high as 28.22% [178].

Beyond standalone power generation, the unique optical properties of metal halide perovskites make them compelling for hybrid photovoltaic–thermal (PV/T) heating and cooling systems. Because the perovskite bandgap is highly tunable, researchers have developed semi-transparent modules that selectively absorb high-energy photons for electricity while transmitting the low-energy, sub-bandgap infrared spectrum directly to an underlying solar thermal collector. This spectral-splitting approach allows the system to simultaneously generate electricity and heat. Furthermore, perovskite PVs inherently possess much smaller temperature coefficients than traditional silicon—reaching approximately −0.08 rel %/°C [179]. This means they suffer less reversible efficiency loss when heated. Consequently, provided that irreversible thermal degradation is managed through robust encapsulation or compositional engineering, their exceptional high-temperature electrical retention makes them suited for PV/T operations, where the module must operate at elevated temperatures to supply effective heat recovery, thermally assisted cooling, or localized space heating.

Perovskite solar cells are expected to become a leading technology for future space power systems, particularly for large-area, high-voltage arrays and small satellites operating under low-light and low-temperature conditions. However, they face parallel but distinct challenges for terrestrial and space deployment, and future research must address both environments simultaneously. On Earth, the primary barriers remain long-term operational stability, moisture and oxygen sensitivity, scalable manufacturing, and bankability, all of which must be solved to enable large-scale commercialization and tandem integration with silicon. In space, priorities shift toward radiation tolerance, thermal cycling durability, vacuum compatibility, and the development of space-qualified substrates and device architectures optimized for end-of-life performance rather than peak efficiency. The space environment itself provides ideal fabrication conditions, eliminating moisture degradation and enabling passive thermal annealing. A 1 MW array could be manufactured from only 12 kg of precursor material [26], dramatically reducing launch cost and enabling on-demand power generation for lunar bases, satellites, and deep-space missions.

3.7. Comparison

Table 12. Qualitative comparison of five thin-film photovoltaic technologies.

Comparison Criterion	a-Si:H	CdTe	CIGS	Perovskite	Organic (OPVs)
Material availability	Abundant (Si-based)	Constrained by tellurium availability	Constrained by indium and gallium availability	Abundant; uses low-cost, earth-abundant synthetic precursors [26,170]	Abundant (carbon-based feedstocks); avoids elemental scarcity [156]
Toxicity/regulatory concern	Generally low	Cadmium-related regulatory concern (requires controlled handling and recycling)	Generally low (material criticality is a larger concern than toxicity)	Lead-related toxicity and environmental concerns; requires stringent encapsulation and recycling [178]	Generally low; transitioning to non-halogenated, green solvents to reduce compliance overhead [156]
Flexible-substrate compatibility	High (can be deposited on flexible substrates)	Typically low (commonly rigid glass modules)	High (commercially demonstrated on flexible substrates)	High; compatible with low-temperature solution processing and printing [26]	Excellent; highly scalable via low-temperature solution- processing (e.g., R2R printing) on ultra-thin polymers [156,157]
Stability/durability	Lower stabilized efficiency due to light-induced degradation (Staebler–Wronski effect)	High durability and mature field performance	High durability with strong field performance	Currently lower than inorganic thin films; sensitive to moisture, oxygen, and heat [176,177]	Improving; advanced NFA systems demonstrate 10-year stability thresholds with strong morphological “locking” [158,180]
Typical deployment/ application	Small-power and lightweight applications; some BIPV use	Utility-scale power plants; large-area deployment	Rooftops, portable power, and BIPV	Emerging in tandems (with Si), BIPV, and indoor light harvesting (IoT) [171,173]	BIPV/T façades, wearable autonomous electronics, portable power, and smart-building sensors [157,158]
Temperature performance	Good high-temperature performance relative to crystalline silicon	Good high-temperature performance relative to crystalline silicon	Good high-temperature performance relative to crystalline silicon	Good performance relative to crystalline silicon; suitable for low-light and space conditions [26,177]	Excellent; near-zero to slightly positive temperature coefficient up to 70 °C, with ultra-low thermal mass [158]

summarizes the key qualitative differences between five established thin-film PV technologies. CdTe is widely deployed at the utility scale, while a-Si:H and CIGS offer distinct advantages in lightweight and flexible applications. Emerging thin-film options such as OPVs and metal halide perovskites are increasingly being studied; however, their stability and lead-management challenges remain important considerations for commercialization.

4. Advanced Technologies for Solar Thermal Management and Energy Storage

4.1. Thermal Cooling

Thermal management is a fundamental challenge in the application of solar cells, as the conversion efficiency and operational reliability of photovoltaic (PV) devices are strongly influenced by their operating temperature. During solar energy conversion, only a fraction of the incident irradiance is transformed into electrical power, while the remainder is dissipated as heat within the semiconductor layers, encapsulation materials, and supporting structures. This heat accumulation elevates the junction temperature, leading to reduced $V_{OC}$, increased carrier recombination, and accelerated material degradation. For conventional crystalline silicon solar cells, efficiency losses of approximately $0.4–0.65\%$ per degree Celsius have been widely reported, underscoring the necessity of effective thermal cooling strategies in practical PV applications [181,182].

Thermal cooling techniques have therefore evolved as an integral component of modern photovoltaic system design. These techniques aim not only to enhance instantaneous electrical performance but also to extend device lifetime, improve operational stability, and enable reliable deployment in high-irradiance and building-integrated environments.

Table 13. Comparison of primary photovoltaic thermal management strategies [181,182,183,184].

Strategy	Mechanism	Energy Req.	Primary Advantage	Main Drawback
Active	Forced convection	High	High heat-flux removal capability	Parasitic power consumption and added maintenance
Passive	Natural convection	None	High reliability and low cost	Limited cooling capacity under low wind/low $\Delta T$
PV/T	Heat recovery (coupled)	Low–Med	Higher overall (electrical + thermal) efficiency	Added system complexity and integration constraints
PCM	Latent heat storage	None	Near-isothermal temperature regulation	Low thermal conductivity and finite storage capacity
Radiative	Sky/space emission	None	Potential for sub-ambient cooling	Strong dependence on climate, humidity, and sky conditions

provides a comparative overview of the primary thermal management strategies discussed in this section.

Active thermal cooling methods employ externally powered components to enhance heat extraction from solar cells through forced convection, conduction, or thermoelectric effects. Common active approaches include forced-air cooling using fans, liquid cooling via water or nanofluid circulation, and thermoelectric cooling modules [183,184]. While heat pipes are fundamentally passive, they are frequently integrated into active systems to facilitate rapid transport to externally cooled heat exchangers. By increasing the heat-transfer coefficient and promoting rapid thermal dissipation, active cooling systems can achieve substantial reductions in PV operating temperature, particularly under high-solar-flux conditions.

Liquid-based cooling systems generally offer superior thermal performance compared to air-based systems due to the higher heat capacity and thermal conductivity of liquids. These systems have been shown to significantly improve temperature uniformity across PV modules, thereby mitigating localized hot spots that accelerate degradation. However, the advantages of active cooling are often offset by parasitic energy consumption, increased system complexity, and maintenance requirements. Consequently, their adoption is typically only justified in applications where high power density, thermal energy recovery, or hybrid energy generation is required.

4.1.1. Passive Thermal Cooling Strategies

Passive cooling strategies rely on natural heat-transfer mechanisms and do not require external energy input, making them attractive for large-scale and building-integrated photovoltaic systems. These approaches include natural convection, conductive heat spreading using fins or heat sinks, phase-change material (PCM) integration, and radiative cooling surfaces [182]. Passive systems are characterized by their simplicity, robustness, and low operational cost, although their cooling effectiveness is inherently constrained by environmental conditions, such as ambient temperature and wind speed.

Heat sinks and extended surfaces enhance convective heat transfer by increasing the effective surface area available for heat dissipation. Their performance depends strongly on geometric design, material selection, and airflow conditions. PCMs, on the other hand, provide thermal buffering by absorbing excess heat during phase transition, thereby maintaining the solar cell near an optimal operating temperature during peak irradiance periods. While PCMs effectively suppress temperature fluctuations, their inherently low thermal conductivity can limit sustained heat removal, necessitating careful system design or hybridization with conductive enhancers.

4.1.2. Photovoltaic–Thermal (PV/T) Cooling Systems

Photovoltaic–thermal (PV/T) systems integrate thermal collectors with PV modules to simultaneously generate electricity and recover useful thermal energy. By extracting heat from the rear surface of PV panels, PV/T systems reduce cell temperature while supplying thermal energy for domestic hot water, space heating, or coupling with heat pumps [185]. These systems are commonly classified into air-based, liquid-based, and heat-pipe-based configurations, each offering distinct trade-offs between efficiency, complexity, and cost.

Liquid-based PV/T collectors generally achieve higher thermal efficiency and better temperature regulation than air-based systems, making them suitable for applications requiring stable thermal output. Heat-pipe-assisted PV/T systems further enhance thermal transport and decouple heat collection from fluid circulation, improving reliability under variable operating conditions. PV/T technology is particularly attractive for building-integrated photovoltaic (BIPV) applications, where thermal management directly influences both electrical performance and indoor thermal comfort.

4.1.3. Phase-Change-Material-Based Cooling

Phase-change-material-based cooling has emerged as a promising passive approach for regulating solar cell temperature under transient thermal loads. PCMs absorb latent heat during melting, thereby limiting temperature rise during periods of high solar irradiance. This quasi-isothermal behavior improves electrical stability and reduces thermal stress on PV components. However, the practical effectiveness of PCM-based cooling is often constrained by slow heat-transfer rates and eventual thermal saturation under prolonged exposure.

To overcome these limitations, hybrid PCM systems incorporating metallic fins, copper channels, or heat pipes have been proposed. These enhancements significantly improve heat conduction within the PCM and facilitate more effective heat extraction [186]. Such hybrid PCM configurations are particularly relevant for BIPV systems, where space constraints, long-duration thermal loads, and architectural integration demand compact and reliable thermal management solutions.

4.1.4. Radiative Cooling and Spectrally Selective Surfaces

Radiative cooling represents a passive thermal management strategy that enables solar cells to dissipate heat directly to outer space by emitting infrared radiation through the atmospheric transparency window ($8–13\mathsf{\mu }\mathrm{m}$). Recent advances in photonic materials and surface engineering have enabled the development of spectrally selective coatings that exhibit high mid-infrared emissivity while maintaining strong solar transmittance in the visible spectrum [187,188].

Radiative cooling structures, including multilayer dielectric stacks, metamaterials, and micro- or nano-patterned surfaces, have demonstrated reductions in PV operating temperatures of several degrees Celsius under real outdoor conditions. These temperature reductions translate into measurable gains in electrical efficiency and improved operational stability without additional energy consumption. The integration of radiative cooling materials into PV modules and BIPV systems is therefore considered a promising pathway for sustainable, long-term thermal management, particularly in hot and arid climates.

4.1.5. Hybrid and Emerging Thermal Cooling Concepts

Hybrid thermal cooling strategies combine multiple cooling mechanisms to overcome the limitations of individual approaches and maximize overall system performance. Examples include PV/T systems coupled with PCM thermal buffering, radiative cooling coatings applied to actively cooled modules, and multifunctional BIPV designs that integrate structural, thermal, and electrical functionalities [182,184]. Emerging research increasingly emphasizes system-level optimization, material innovation, and lifecycle performance assessment to ensure scalability and economic viability.

In summary, the maturation of advanced thermal management strategies is essential for enabling high-performance, durable, and sustainable solar cell applications. Future research directions are expected to focus on multifunctional materials, adaptive thermal control, and seamless integration of cooling technologies into next-generation photovoltaic and building energy systems [182,185].

4.2. Thermal Heating

This subsection discusses how thin-film solar cells can contribute to thermal heating, mainly for buildings and low-temperature industrial uses. This discussion therefore moves from general heating concepts to a comparative assessment of the thin-film solar cell families considered in this review, namely, cadmium telluride (CdTe), copper indium gallium selenide (CIGS), copper zinc tin sulfide (CZTS), amorphous silicon (a-Si), organic photovoltaics (OPVs), and perovskite solar cells (PSCs), with emphasis on their relevance to hybrid thermal-energy systems. Particular attention is given to conversion performance, temperature sensitivity, durability under thermal stress, scalability, and compatibility with building-integrated photovoltaic/thermal (BIPV/T) configurations.

4.2.1. Fundamentals of Solar-Based Thermal Heating

Solar-based thermal heating captures sunlight to deliver useful heat for domestic hot water, space heating, and process pre-heating. Since heating demand represents a major share of building energy use, recovering low- and medium-temperature heat can noticeably reduce overall consumption and fossil-fuel dependence [46].

Two main approaches are relevant. Direct solar thermal collectors (e.g., flat-plate and evacuated-tube systems) convert solar radiation mainly into heat. By contrast, PV-assisted heating uses photovoltaic electricity to operate pumps, controls, circulation loops, or heat pumps. PV/T collectors combine both functions by generating electricity while simultaneously extracting heat from the PV layer. This is particularly useful because PV electrical output generally decreases as operating temperature rises; heat extraction therefore improves electrical stability while also providing usable thermal energy [46,189].

Thin-film technologies are attractive for heating-linked integration because they can cover large building surfaces with low structural burden. Recent BIPV studies also emphasize lightweight form factors, flexible installation, façade compatibility, and comparatively good performance under diffuse or low-light conditions, which are beneficial for urban or partially shaded sites [190].

4.2.2. Technology Comparison for Heating-Oriented Integration

For thermal-heating applications, not all thin-film technologies offer the same balance of performance, durability, and deployment readiness. Current BIPV-oriented comparisons indicate that CdTe and CIGS are the most commercially mature thin-film options, with module efficiencies reported around 18.6% and 19.2%, respectively, while a-Si remains lower in efficiency but attractive for semi-transparent and diffuse-light applications [190]. Temperature sensitivity is also important in heating-linked systems because module temperature often rises during operation. Reported temperature coefficients suggest that a-Si and CdTe are generally less temperature-sensitive than CIGS, while perovskites show very strong thermal-performance potential but still face unresolved stability and encapsulation challenges under practical outdoor operation [190,191].

A second distinction concerns scalability and durability. CdTe and CIGS already have the strongest near-term balance between efficiency, manufacturability, and building-scale integration [190]. Amorphous silicon remains attractive where low weight, semi-transparency, and façade use are more important than peak power density [190]. CZTS is promising because of its low-toxicity and earth-abundant constituent elements, but its practical performance remains below the leading commercial thin-film families [190]. OPVs offer excellent mechanical flexibility and low structural burden, but heating-linked deployment still requires better thermal durability and morphology control. In one representative OSC study, device parameters improved up to an optimum annealing temperature near 90 °C and deteriorated at higher temperatures, illustrating the sensitivity of organic systems to thermal history [192]. PSCs offer the highest long-term upside for hybrid heating envelopes because of their high efficiency potential and favorable temperature behavior, but their real-world implementation still depends on improved moisture resistance, encapsulation, and long-term operational stability [190,191].

Table 14. Performance–stability–scalability comparison of thin-film solar cell families for solar thermal heating integration.

Technology	Indicative Performance	Heating-Oriented Strengths	Main Stability/Scalability Limitation	Near-Term Suitability
CdTe	Cell/module efficiencies around 21.1%/18.6%; relatively low temperature coefficient [190,191]	Strong hot-climate suitability, mature manufacturing, good match for roof/façade BIPV/BIPV-T deployment [190,191]	Cadmium-related environmental handling and end-of-life management still require careful control [190]	Strong
CIGS	Cell/module efficiencies around 21.7%/19.2%; commercially mature and flexible [190]	High efficiency with good architectural adaptability, especially for lightweight and flexible BIPV surfaces [190]	Higher temperature sensitivity than CdTe and comparatively more complex manufacturing routes [190,191]	Strong
a-Si	Cell/module efficiencies around 10.2%/9.1%; lower power density but mild temperature response [190,191]	Useful for semi-transparent, diffuse-light, and lightweight envelope applications; attractive for façades and glazing-related integration [190]	Lower efficiency and area-normalized output than CdTe/CIGS; performance penalties when high power density is needed [190]	Moderate/niche
CZTS	Promising but still lower practical efficiency than leading commercial thin-film families [190]	Earth-abundant, lower-toxicity material platform; interesting from sustainability and future hybrid-building perspectives [190]	Limited commercial maturity and weaker current performance envelope for large-scale deployment [190]	Emerging
OPVs	Rapidly improving but still thermally more sensitive than the leading commercial thin-film families in heating-linked contexts [192]	Very lightweight, mechanically flexible, and suitable for conformal or ultra-low-load surfaces [192]	Performance is sensitive to thermal history, morphology evolution, and long-term durability under elevated-temperature operation [192]	Emerging/niche
PSCs	Very high efficiency potential; cell/module values around 27.3%/22.8% reported in recent BIPV comparisons [190]	Excellent long-term potential for high-efficiency hybrid envelopes, including semi-transparent and tandem concepts [190,191]	Moisture/UV/thermal stability, encapsulation, and large-scale reliability remain the key bottlenecks [190,191]	High potential, medium term

Note: Efficiency, commercial maturity, and BIPV deployment discussion are synthesized mainly from [190]. Temperature-sensitivity comparison is supported by [191]. OPV thermal-history and annealing-dependent behavior are supported by [192].

summarizes this material-to-system comparison from the viewpoint of solar thermal heating integration.

4.2.3. Thin-Film PV/T Systems for Heating Applications

PV/T collectors are widely used for water and space-heating support. A typical design places an air or water heat-extraction layer behind the PV module and transfers heat to a storage tank or to a hydronic or warm-air loop [46,189]. Thin-film modules can ease roof, wall, and façade mounting, including building-integrated PV/T (BIPV/T) layouts [190].

The key system benefit is improved total energy utilization from the same area. Instead of losing PV heat to the ambient environment, PV/T systems recover part of it as useful heat while moderating PV temperature [189]. Recent work also shows that combined performance depends strongly on design choices such as thermal contact, heat-transfer path, exchanger material, emissivity, and operating conditions, so careful engineering is essential [189,193].

Hybrid configurations are especially attractive for heating applications. A BIPV/T unit can preheat air for an air-source heat pump, improving cold-weather operation and lowering purchased heating electricity [193]. In one recent BIPV/T–BISAH–ASHP case study, coupling the solar air-heating array to the heat pump reduced space-heating electricity consumption by 6.5% in a net-zero house configuration [193]. Thermal buffering with PCMs is another practical route. A recent PVT–PCM study reported that adding water cooling and organic PCM reduced PV cell temperature from 59 °C to 36 °C and improved electrical efficiency from 6.1% to 9.5%, while maintaining useful thermal recovery [47].

Heat-exchanger design also matters. A recent hybrid PVT analysis comparing stainless steel, aluminum, and copper pipes found the highest overall efficiency with copper, showing that heat-extraction materials should be selected not only for conductivity but also for manufacturability, cost, and long-term reliability [189].

Figure 11 illustrates a practical thin-film BIPV/T heating layout that combines roof/façade integration with storage and a heat-pump-assisted supply for domestic hot water and space heating.

Figure 11. System-level schematic of a thin-film BIPV/T heating configuration integrated with thermal storage and an air-source heat pump, conceptually synthesized from representative PV/T, BIPV/T, and heat-pump-coupled heating studies [46,189,190,193]. Numbered callouts correspond to Table 15.

Figure 11. System-level schematic of a thin-film BIPV/T heating configuration integrated with thermal storage and an air-source heat pump, conceptually synthesized from representative PV/T, BIPV/T, and heat-pump-coupled heating studies [46,189,190,193]. Numbered callouts correspond to Table 15.

Table 15. Numbered callouts used in Figure 11 for the thin-film BIPV/T building heating configuration, synthesized from representative BIPV/T, PV/T, and solar-heating integration concepts in [46,189,190,193].

ID	Component	Role in the Heating System
1	Thin-film BIPV/T roof	Primary solar aperture on the roof; provides electricity and recoverable heat for building heating services.
2	Thin-film BIPV/T façade	Additional solar aperture on vertical surfaces; supports energy yield in urban/shaded settings and extends collection area.
3	Thermal absorber/backplate	Collects and transfers heat from the PV back side into the working fluid to reduce PV temperature and recover useful heat.
4	Fluid loop piping	Transfers thermal energy between BIPV/T collectors, storage, and heating subsystems with minimal distribution losses.
5	Circulation pump	Drives the working fluid through the collector loop; enables controlled heat extraction based on demand and operating set-points.
6	Hot water storage tank	Stores collected heat for later use; smooths demand–supply mismatch and improves system-level utilization.
7	Air-source heat pump (outdoor unit)	Upgrades low-grade heat (and/or uses electricity) to meet heating demand efficiently, especially in cold conditions.
8	Space heating loop (radiator/floor)	Delivers thermal energy for indoor comfort heating; can be hydronic radiators or underfloor heating circuits.
9	Domestic hot water outlet	Supplies hot water demand for residential or service uses; typically coupled with tank stratification and mixing control.
10	Inverter + electrical loads path	Converts PV DC to AC for building loads; can also supply pumps/controls and support heat pump operation.

Table 15. Numbered callouts used in Figure 11 for the thin-film BIPV/T building heating configuration, synthesized from representative BIPV/T, PV/T, and solar-heating integration concepts in [46,189,190,193].

ID	Component	Role in the Heating System
1	Thin-film BIPV/T roof	Primary solar aperture on the roof; provides electricity and recoverable heat for building heating services.
2	Thin-film BIPV/T façade	Additional solar aperture on vertical surfaces; supports energy yield in urban/shaded settings and extends collection area.
3	Thermal absorber/backplate	Collects and transfers heat from the PV back side into the working fluid to reduce PV temperature and recover useful heat.
4	Fluid loop piping	Transfers thermal energy between BIPV/T collectors, storage, and heating subsystems with minimal distribution losses.
5	Circulation pump	Drives the working fluid through the collector loop; enables controlled heat extraction based on demand and operating set-points.
6	Hot water storage tank	Stores collected heat for later use; smooths demand–supply mismatch and improves system-level utilization.
7	Air-source heat pump (outdoor unit)	Upgrades low-grade heat (and/or uses electricity) to meet heating demand efficiently, especially in cold conditions.
8	Space heating loop (radiator/floor)	Delivers thermal energy for indoor comfort heating; can be hydronic radiators or underfloor heating circuits.
9	Domestic hot water outlet	Supplies hot water demand for residential or service uses; typically coupled with tank stratification and mixing control.
10	Inverter + electrical loads path	Converts PV DC to AC for building loads; can also supply pumps/controls and support heat pump operation.

4.2.4. Building and Industrial Heating Applications

In buildings, thin-film PV/T and BIPV/T systems mainly support domestic hot water and space heating. Roof- or façade-integrated collectors can feed a storage tank or provide preheated air or water that lowers the conventional heating load [46,189]. In cold climates, coupling BIPV/T with heat pumps is particularly relevant because even modest air preheating can improve heat-pump stability and seasonal performance [193].

For low-temperature industrial processes, PV/T can assist tasks such as make-up water preheating, washing or cleaning lines, and moderate drying steps. In these cases, the thermal output reduces fossil-fuel use during daylight hours, while the electrical output can operate pumps, sensors, and basic controls [46,189]. In retrofit contexts, thin-film building integration can also improve energy performance by offsetting purchased electricity and lowering heating demand without imposing major structural penalties [190].

Agriculture and greenhouse heating is another relevant target area. Active solar thermal greenhouse systems already use air heating, basal heating, and backwall heating, often combined with short- or long-term heat storage [194]. Reviews of PCM-assisted greenhouse systems further show that thermal storage can raise greenhouse temperatures during critical nighttime periods and reduce daily temperature fluctuations, which is directly relevant for stable low-temperature solar heating applications [195]. In this sense, thin-film-assisted solar heating is not restricted to buildings alone; it also fits controlled-environment agriculture, where lightweight surfaces and distributed collection areas are useful.

4.2.5. Limitations and Optimization Strategies

Real performance is mainly shaped by heat losses, the control strategy, and material durability. PV/T collectors lose heat through convection, conduction, and radiation, and additional distribution losses appear when ducts or pipes are long or poorly insulated [189]. These losses become more visible under windy weather, non-ideal collector orientation, or when storage is located far from the collector field.

Long-term exposure also matters. Outdoor operation involves repeated temperature cycling and solar radiation exposure that can accelerate performance decay and increase maintenance needs, especially in building-linked heating systems [191]. A second challenge is the electrical–thermal trade-off: extracting more heat can improve thermal delivery and reduce PV temperature, but it can also require higher flow rates, larger exchangers, and extra pumping power. Therefore, system optimization usually focuses on thermal contact quality, flow control, realistic operating set-points, and the matching of collector output to heating demand [189,193].

Durability of thermal components must be considered as well. Selective absorber coatings are widely used in solar thermal hardware, yet aging mechanisms such as oxidation, diffusion, and thermally induced coating degradation can reduce long-term stability if material choice and service conditions are not properly managed [196]. In addition, some thin-film devices are themselves sensitive to temperature history. Organic solar cell studies show that thermal treatment can significantly change morphology, recombination pathways, and device parameters, reinforcing the need for controlled thermal operating windows in integrated systems [192].

Overall, the most robust results come from system-level design rather than from PV material selection alone. In practice, better outcomes are usually achieved by combining suitable thin-film technology choice with PCM-assisted storage, compact heat-transfer layouts, durable absorber materials, and demand-matched controls that limit unnecessary thermal losses [47,193,195].

4.3. Energy Storage

Energy storage is essential for overcoming the intermittency challenges associated with solar energy, particularly in systems utilizing thin-film photovoltaic (PV) technologies. Solar energy generation fluctuates throughout the day due to weather conditions and time of day, which makes it difficult to match energy production with demand. Energy storage systems (ESSs) play a crucial role in ensuring that energy can be stored during periods of excess generation and used during low-generation times, such as at night or during cloudy weather. This subsection discusses the importance of energy storage in thin-film PV applications, including electrical and thermal storage systems, as well as the role of hybrid storage solutions and smart energy management.

4.3.1. Importance of Energy Storage in Solar Systems

Solar energy, while abundant, is highly intermittent. The mismatch between solar energy generation and consumption patterns presents a significant challenge for integrating solar energy into the power grid. This challenge is particularly relevant for thin-film solar technologies, which are often used in distributed and hybrid systems. Thin-film solar cells are lightweight, flexible, and suitable for applications such as building-integrated photovoltaics (BIPV), portable power systems, and off-grid installations. However, to fully harness the potential of thin-film PV, it is necessary to address the intermittency of solar energy through energy storage.

Energy storage in solar systems can provide grid stability, improve reliability, and ensure that energy is available when it is needed most. In the context of thin-film solar systems, energy storage is particularly beneficial for off-grid applications, where there is no access to a centralized power grid. Here, batteries and other storage technologies can store excess energy produced during the day for use at night or during periods of low solar irradiance. Additionally, building-integrated systems can benefit from energy storage by enabling the use of solar energy even when the sun is not shining [197,198].

4.3.2. Electrical Energy Storage Integration

The integration of electrical energy storage with thin-film PV systems has become a focal point for improving system performance and energy reliability. Batteries, particularly lithium-ion and sodium-ion technologies, have proven to be effective for coupling with thin-film solar systems. Lithium-ion batteries are widely used in applications ranging from consumer electronics to electric vehicles, offering high energy density, long cycle life, and relatively low maintenance. Sodium-ion batteries, on the other hand, are gaining attention due to their lower cost and more abundant raw materials.

In thin-film PV applications, lightweight and modular battery systems provide significant advantages. For example, in portable power systems and wearable electronics, where weight and flexibility are key considerations, thin-film PV systems coupled with lightweight batteries are ideal. These systems are also compatible with smart energy management systems, which optimize energy storage and usage through real-time monitoring and control. This integration enables intelligent dispatch of stored energy, ensuring that energy is used efficiently when required. In addition, the modular nature of these systems allows for scalable deployments that can be tailored to meet the needs of specific applications [199,200].

4.3.3. Thermal Energy Storage Systems

While electrical energy storage is critical, thermal energy storage offers complementary benefits, especially in thin-film PV and photovoltaic–thermal (PV/T) systems. Thermal energy storage can store excess heat generated during peak sunlight hours, which can then be used during periods of low sunlight or at night. There are several approaches to thermal storage, including sensible heat storage, latent heat storage using phase-change materials (PCMs), and thermochemical storage.

Sensible heat storage works by raising the temperature of a material, such as water or molten salts, which can then release the stored heat as it cools down. Latent heat storage, on the other hand, utilizes phase-change materials that absorb or release heat as they transition between solid and liquid phases. This type of storage is highly efficient because it can store more heat per unit mass compared to sensible heat storage. Thermochemical storage involves chemical reactions that can absorb or release heat, providing high energy density and potentially long-term storage.

In PV/T systems, thermal energy storage can enhance the overall efficiency of solar energy systems by capturing and storing the excess heat generated by the PV cells. This stored heat can be used for heating water, space, or even for generating electricity in combined heat and power systems. The integration of thermal storage with thin-film PV systems can significantly improve their performance, especially in off-grid or hybrid solar applications [201,202].

4.3.4. Hybrid Storage and Smart Energy Management

The future of solar energy systems lies in hybrid storage solutions that combine both electrical and thermal storage capabilities. These hybrid systems can offer enhanced flexibility and efficiency by providing multiple modes of energy storage that cater to the varying needs of different applications. For instance, an integrated system that combines electrical storage with thermal storage can store both electrical energy and excess heat, optimizing the overall system performance.

Smart energy management systems play a crucial role in the efficient operation of hybrid storage systems. These systems use control algorithms to predict energy demand and supply, manage the charging and discharging cycles of batteries, and optimize the usage of stored energy. The integration of energy storage systems with smart grids and building automation systems further enhances the capabilities of solar PV systems. By incorporating artificial intelligence (AI) and predictive control, these systems can anticipate energy demand and adjust the energy storage and usage accordingly, ensuring optimal performance.

As illustrated in Figure 12, a representative hybrid architecture couples thin-film PV generation with a smart energy management hub that coordinates power routing and storage dispatch. The PV array supplies DC electricity that is conditioned (DC–DC conversion) before entering the controller, which then performs load balancing by allocating power among the building load, the lithium-ion battery pack (electrical storage), and grid interaction through the power-grid interface. In parallel, the system includes a phase-change material (PCM) thermal tank for thermal storage; excess energy can be directed to the thermal subsystem to store heat for later use, providing longer-duration buffering compared to the battery. The figure also highlights how AI optimization and predictive control signals inform the hub, enabling proactive scheduling of battery charge/discharge and thermal charging based on anticipated demand, thereby improving self-consumption and reducing reliance on the grid [73,203].

Hybrid storage solutions, along with smart energy management, are expected to play a significant role in future integrated energy networks. These systems can help achieve higher efficiency, reduced energy costs, and greater system reliability, making them an essential component of the next generation of solar power systems [73,203].

5. Cost Economics of Thin-Film Solar Modules

5.1. Module Manufacturing Costs and Cost Structure

A primary requirement for any photovoltaic technology is to provide abundant, affordable, and sustainable electricity, which necessitates a deep understanding of its manufacturing cost structure. The manufacturing costs of thin-film solar cells vary significantly depending on material choices, deposition methods, and production scale, though they generally benefit from lower semiconductor material usage compared to traditional crystalline silicon technologies.

Cadmium telluride (CdTe) currently represents the most cost-competitive thin-film technology, with a minimum sustainable price (MSP) benchmarked around $0.28/W [204]. Its economic advantage stems from a highly streamlined, vertically integrated manufacturing process with moderate-to-low capital expenditure (CAPEX) intensity. In contrast, copper indium gallium selenide (CIGS) modules face higher manufacturing costs, with an MSP of approximately $0.48/W [204]. The cost structure of CIGS is constrained by the reliance on scarce and price-volatile elements such as indium and gallium, as well as the need for complex vacuum-based deposition and sulfurization/selenization processes that increase capital, maintenance, and operational expenses [205,206].

Amorphous silicon (a-Si) modules have an estimated manufacturing cost on the order of $0.7/W, reflecting lower stabilized efficiencies and light-induced degradation effects that limit economic performance relative to CdTe and CIGS technologies [205,207]. The cost structure of an optimized a-Si module is distributed such that materials account for roughly 42% of the total cost, while equipment depreciation contributes a significant 28% [208]. To bypass the material scarcity and toxicity issues of CIGS and CdTe, copper zinc tin sulfide/selenide (CZTS/CZTSSe) relies on earth-abundant materials, achieving projected manufacturing costs between $41 and $52 per square meter [209]. The compatibility of CZTSSe with low-temperature, solution-based deposition techniques offers substantial reductions in both CAPEX and operational expenditures (OPEX) [209].

Metal halide perovskite solar cells share this low-temperature, solution-processing advantage, though early-stage manufacturing costs currently sit around $0.57/W due to low mass-production yields [210]. For a 100 MW perovskite production line, the cost structure is remarkably material-heavy: materials account for nearly 70% of total costs, while CAPEX and OPEX represent only 15% each [210]. Similar to other emerging thin films, the active perovskite layer accounts for a mere 6% of the materials budget, whereas the conductive glass substrate (e.g., FTO or ITO) dominates at 30% to 35% of material costs [210].

A similarly material-dominated cost structure is observed in organic photovoltaics (OPVs). A comprehensive engineering cost estimation for mass-manufactured OPV modules on flexible plastic substrates was presented by Mulligan et al. Their analysis, based on a P3HT:PCBM active-layer system, projected a total manufacturing cost of $7.85/m² with an uncertainty of ±$2.35/m². The cost structure is noteworthy because it is dominated by material inputs: materials account for approximately 71% of the total module cost, operating costs contribute 25%, and capital investment in equipment represents only 4%. The single largest component is the PET substrate and encapsulation barrier, which together account for $3.50/m² (63% of the total materials budget).

5.2. Life-Cycle Costing and Levelized Cost of Electricity (LCOE)

Beyond the initial manufacturing expense, the long-term economic feasibility of thin-film photovoltaics is best assessed through life-cycle costing, which determines the levelized cost of electricity (LCOE), the average cost per kilowatt-hour ($/kWh) produced over the system’s lifetime. Mature cadmium telluride (CdTe) modules achieve a highly competitive LCOE of $65/MWh, driven by gigawatt-scale manufacturing and 25- to 30-year bankable service lifetimes [205]. Copper indium gallium selenide (CIGS) faces a slightly higher LCOE, in the high $50s/MWh range, limited by material supply constraints and elevated project financing costs [205]. Among emerging technologies, metal halide perovskites show a projected LCOE between $0.03 and $0.06/kWh [30], though current early-stage estimates sit at $0.22/kWh due to low pilot yields and unresolved stability issues [210]. Meanwhile, copper zinc tin sulfide/selenide (CZTSSe) must still reach approximately 15% module efficiency to match the marginal LCOE of CIGS [209], and the LCOE of amorphous silicon (a-Si) remains constrained by light-induced degradation effects over its operational lifetime [205]. Early LCOE projections for OPVs were hampered by short operational lifetimes; a system lasting only 3 to 5 years yielded uncompetitive LCOEs ranging from $\$0.26$ to $\$0.69/\mathrm{kWh}$ due to the inability to amortize the initial balance-of-system (BOS) and installation costs [211].

5.3. Comparative Economic Analysis

When benchmarked against one another, the cost structures of thin-film technologies show divergent paths to competitiveness. The leading mature thin-film technology, cadmium telluride (CdTe), sets the economic benchmark, with a minimum sustainable manufacturing price of approximately $0.28/Wp [204]. In comparison, copper indium gallium selenide (CIGS) modules face a higher manufacturing cost, near $0.48/Wp, while amorphous silicon (a-Si) costs remain elevated, around $0.73/Wp, largely confining the latter to niche architectural applications [204,207]. For emerging technologies, cost projections based on pre-industrial pilot lines are highly dependent on the final stabilized module efficiency and active area. Current metal halide perovskite manufacturing costs sit at approximately $0.57/Wp [210], while CZTSSe modules offer projected production costs of $52/m², which could translate to highly competitive per-watt pricing if 15% module efficiency is achieved [209]. Cost projections for OPV modules range from 1.35/Wp to 4.09/Wp (approximately $1.88/Wp to $5.69/Wp).

A key advantage for emerging solution-processed thin films (such as perovskites, CZTSSe, and OPVs) is the significantly lower capital investment required. An OPV production facility, for instance, was estimated to require an investment of approximately 0.5 million per MWp of annual capacity, equivalent to around $0.695 million/MWp. Similarly, perovskite roll-to-roll manufacturing facilities possess an estimated CAPEX as low as $0.15 to $0.17/Wp [204]. This is substantially lower than the capital intensity for mature vacuum-based thin-film technologies like CIGS, which can reach $200 to $350 million per GW [205], making mature technologies up to 6.5 times higher in capital intensity. This lower barrier to entry could enable more decentralized, smaller-scale production scenarios.

5.4. Advances in Manufacturing to Reduce Costs

Innovations in manufacturing processes are critical to realizing the cost projections across all thin-film technologies. For mature technologies like cadmium telluride (CdTe), high-speed vapor-transport deposition (VTD) has streamlined production, enabling the conversion of a glass substrate to a finished module in less than three hours [207]. Similarly, copper indium gallium selenide (CIGS) and amorphous silicon (a-Si) have benefited from advancements in high-throughput roll-to-roll (R2R) processing on flexible substrates and optimized sputtering or plasma-enhanced chemical vapor deposition (PECVD) techniques, which substantially lower capital expenditures [205].

Emerging materials such as metal halide perovskites and CZTSSe leverage low-temperature, solution-based deposition methods, including nanocrystal inks, spray pyrolysis, and slot-die coating to bypass expensive vacuum equipment [209]. Slot-die coating, in particular, offers material utilization efficiencies up to 95%, making it a highly cost-effective and scalable alternative to wasteful spin-coating for perovskite module fabrication [212].

Ink-jet printing has also emerged as a promising additive, non-contact deposition technique that dramatically improves material utilization. As an additive method, it deposits material only where needed, resulting in near-zero material waste compared to the significant waste (assuming 30–50%) common in traditional subtractive methods, such as screen printing. Ink-jet printing can simplify processing by eliminating the need for physical masks and enabling rapid, software-driven design changes. The scalability of printing-based deposition for large-area thin-film modules has been demonstrated through roll-to-roll-fabricated organic photovoltaic modules and scalable coated perovskite modules, confirming its suitability for industrial manufacturing [32,212].

5.5. Economic Barriers and Path to Competitiveness

The most significant barrier to achieving a competitive levelized cost of electricity (LCOE) for OPV technology remains its operational lifetime. Most studies report $T_{80}$ lifetimes of only 2–5 years under accelerated testing, which is insufficient for widespread utility adoption. As noted in sensitivity analyses, extending this lifetime to 10 years is a major technical milestone for economic viability. Encouragingly, materials innovation (e.g., non-fullerene acceptors) has shown the potential for improved stability.

Similarly, metal halide perovskites suffer from moisture and UV-induced degradation that restrict module warranties and elevate project financing costs, making the demonstration of a 25-year lifespan a strict prerequisite for outperforming mature silicon modules [205,210]. For amorphous silicon (a-Si), the primary barrier is the Staebler–Wronski light-induced degradation effect, which caps stabilized commercial efficiencies at 6–9% and drives its LCOE above market benchmarks [205]. Copper zinc tin sulfide/selenide (CZTSSe) faces a large $V_{OC}$ deficit, driven by antisite defects and band tailing, preventing it from exceeding ∼15% efficiency despite its ultra-low material costs [206,209]. Conversely, mature thin films like CIGS and CdTe face material availability constraints (indium, gallium, and tellurium) and cadmium toxicity regulations, while CIGS also contends with high capital intensity [207,209].

To achieve grid parity, OPV technology must reach several key targets. Module efficiencies need to consistently exceed 10% in large-area formats, and long-term operational stability must improve substantially. For perovskites, scaling up production yields above 90% and reaching 25% module efficiency could drop manufacturing costs below $0.32/W, yielding a highly competitive LCOE of $0.06/kWh [30,210]. Meanwhile, CdTe and CIGS continue to advance toward higher efficiencies through absorber and interface engineering approaches [36,205]. Ultimately, integrating advanced thin films into multi-junction tandem architectures—such as perovskite/silicon or perovskite/CIGS—remains a promising pathway to exceed single-junction limits and reduce long-term electricity costs [207,213].

5.6. Sustainability and Material Availability

Material availability and regulatory considerations are central to long-term thin-film photovoltaic scalability. The toxicity of cadmium in CdTe and the scarcity of critical raw materials such as indium and gallium in CIGS pose sustainability and supply-chain challenges [205,206]. CdTe partly offsets these concerns through established recycling and favorable energy payback performance, with an energy payback time (EPBT) of 0.8–1.5 years reported for thin-film implementations in building-integrated contexts [190,207].

To bypass material scarcity and toxicity, thin-film technologies such as amorphous silicon (a-Si) and copper zinc tin sulfide/selenide (CZTSSe) rely on earth-abundant, environmentally benign elements, supporting a sustainable manufacturing profile [206,209]. Metal halide perovskite solar cells also offer favorable material and process characteristics; nevertheless, commercial sustainability depends on mitigating lead (Pb) toxicity and leaching risks through robust encapsulation and cradle-to-cradle recycling frameworks [205,214]. Overall, sustainability and material availability considerations will increasingly influence the long-term cost competitiveness of thin-film photovoltaics as deployment scales.

To address concerns related to material toxicity, elemental scarcity, and long-term stability, several realistic strategies are being developed for sustainable large-scale commercialization of thin-film photovoltaic technologies. These include the use of earth-abundant materials such as CZTSSe and amorphous silicon to reduce dependence on scarce elements like indium and gallium. Recycling and closed-loop recovery strategies for cadmium and tellurium-based modules are also being implemented to minimize environmental risks and improve resource efficiency. For perovskite devices, improved encapsulation and moisture-resistant designs are being developed to enhance stability and prevent material leakage. Additionally, scalable manufacturing techniques such as roll-to-roll processing and tandem device architectures are being explored to support cost-effective large-area production.

6. Future Directions

6.1. Advanced Architecture and Deployment Strategies for Thin-Film Photovoltaics

A primary focus across thin-film solar technologies is the development of tandem and multi-junction architectures to optimize spectral harvesting. Perovskite cells are frequently paired with other thin films, such as in perovskite/CdTe stacks and monolithic perovskite/CIGS tandems. The latter utilizes CIGS as a narrow-bandgap bottom subcell to capture infrared light, yielding $V_{oc}$ values near 1.765 V and certified efficiencies reaching 24.6 percent [148]. Similarly, pure-sulfide CZTS, with its 1.5 eV bandgap, is being investigated as a suitable top cell for monolithic CZTS/Si tandems. This approach relies on transparent back contacts (like ITO or FTO) to target theoretical efficiencies exceeding 30 percent while maintaining an earth-abundant and low-carbon composition. Amorphous silicon (a-Si:H) also plays a functional role in perovskite/silicon tandem cells as a passivation and carrier-selective layer, facilitating efficiencies beyond 33 percent [14,97]. Additionally, multi-junction cells pairing a-Si:H with lower-bandgap materials like $\mu c$-Si:H or a-SiGe:H are used to mitigate the efficiency impacts of the Staebler–Wronski effect [72,74].

Beyond tandem structures, physical versatility drives the integration of thin films into urban environments, particularly in building-integrated photovoltaics (BIPV). Both a-Si:H and CIGS offer aesthetic adaptability and structural advantages. CIGS modules on flexible substrates can weigh under 2 kg/m², enabling the green retrofitting of aging residential buildings where traditional silicon modules exceed structural load limits [7]. Concurrently, a-Si:H provides semi-transparency and consistent performance under diffuse or indoor light conditions [72,73]. For utility-scale applications, current CdTe research includes the development of bifacial modules designed to capture albedo light. Advancing these technologies for widespread deployment requires addressing material stability, manufacturing scalability, and resource scarcity. In CdTe, efforts are directed toward implementing Group-V dopants to achieve stable hole densities exceeding ${10}^{16}$ cm⁻³ [215], alongside improved tellurium recycling protocols to offset material scarcity [42,43]. For perovskites, research focuses on robust material compositions, metal oxide transport layers, moisture management, and scalable vapor-deposition or roll-to-roll manufacturing. Solving these material challenges would enable perovskites to function as a dual-use platform for both terrestrial decarbonization and extraterrestrial applications. Future aerospace deployments aim to utilize in-space fabrication strategies, such as 3D printing (as illustrated in Figure 13), to provide lightweight power for satellites, lunar infrastructure, and deep-space missions.

6.2. Techno-Economic Outlook

Future research on thin-film photovoltaic technologies should prioritize integrated techno-economic assessments that couple module manufacturing cost, operational lifetime, degradation behavior, and end-of-life management within unified life-cycle costing and levelized cost of electricity (LCOE) frameworks. Life-cycle studies and techno-economic analyses have quantified important cost and impact contributors, such as the influence of manufacturing electricity use and absorber-layer contributions on per-kWh outcomes [31], and the non-negligible role of end-of-life recycling costs and recovered-value offsets for thin-film modules [216]. However, harmonized models that consistently span material sourcing, module fabrication, system integration, operation, and recycling remain limited [36]. Accordingly, establishing standardized assumptions for lifetime, degradation rate, manufacturing yield, and production scale is essential, because stability- and lifetime-dependent parameters can dominate LCOE outcomes, particularly for emerging thin-film platforms [29,217].

From a cost-economics standpoint, commercially mature CdTe and CIGS technologies currently provide the most reliable benchmarks for thin-film competitiveness due to established manufacturing pathways and favorable life-cycle performance at current efficiencies [31,36,206]. Earth-abundant kesterite technologies (CZTS, CZTSSe) remain compelling for long-term sustainable cost reduction because their material systems avoid the scarcity constraints associated with indium and tellurium [208,218], but their near-term competitiveness is still limited by efficiency and voltage-deficit challenges that must be addressed to translate material advantages into lower LCOE [36,218]. Metal halide perovskite solar cells and perovskite-based tandems offer strong prospects for disruptive cost reduction enabled by low-temperature processing and scalable manufacturing concepts [206], yet published LCOE analyses and tandem replacement modeling show that their economic viability remains contingent on achieving long operational lifetimes and low degradation rates [29,217].

Accordingly, future cost-focused research should emphasize: (i) lifetime extension and stability engineering, with LCOE-linked validation; (ii) scalable high-yield manufacturing, with quantified process-energy and yield impacts; (iii) recycling-integrated cost modeling that captures net end-of-life costs and recovered value; and (iv) large-area module performance validation to ensure that device-level advances translate into bankable module- and system-level economics [31,36,206,216]. Progress in these areas is expected to yield larger LCOE reductions than incremental efficiency gains alone, thereby shaping the relative competitiveness of thin-film photovoltaic technologies in next-generation solar energy systems [29,36].

6.3. Application of Machine Learning in Thin-Film Solar Cell Research

Future progress in thin-film solar cell technologies will increasingly rely on the integration of machine learning (ML) as a core component of material discovery, device optimization, and manufacturing control rather than as a standalone analytical tool. Given the strong coupling between composition, microstructure, processing conditions, and device physics, ML is well positioned to enable accelerated exploration of high-dimensional design spaces and to guide decision making beyond the limits of conventional trial-and-error experimentation [22,23].

One important future direction is the development of ML-driven frameworks for fabrication parameter optimization and compositional tuning across earth-abundant and emerging absorber systems such as CZTSSe and Sb₂(S_xSe_1−x)₃. Data-driven models capable of identifying dominant variables and interaction effects can support targeted defect mitigation strategies and efficiency enhancement, thereby shortening experimental cycles and improving reproducibility [24,219,220]. Extending such approaches toward multi-objective optimization that simultaneously considers efficiency, stability, and manufacturability represents a critical next step.

Beyond forward performance prediction, inverse design enabled by ML is expected to play a growing role in tailoring thin-film optical and electronic structures toward predefined performance targets. Neural network-based inverse models offer a pathway to rapidly generate material stacks and multilayer configurations with customized optical absorption or reflection characteristics, substantially reducing human-driven iteration [221]. Expanding inverse design methodologies to include stability and degradation constraints would further enhance their practical impact.

Another major opportunity lies in the convergence of ML with autonomous and scalable fabrication platforms. Closed-loop experimental systems, in which ML algorithms iteratively propose and evaluate processing conditions, provide a foundation for self-optimizing thin-film deposition workflows and rapid convergence toward high-performance devices [180]. In parallel, ML-enabled process monitoring during scalable manufacturing can facilitate early detection of defects, thickness non-uniformities, and process drift, thereby improving yield and reproducibility [222].

To illustrate the practical relevance of machine learning for TFSC optimization, several real-life examples can be highlighted. Zhang et al. [180] demonstrated a machine learning-guided robotic platform for optimizing perovskite thin-film parameter spaces, enabling accelerated exploration of deposition conditions and rapid convergence toward high-performance perovskite solar cells. This represents a clear case in which ML was applied directly to fabrication optimization rather than only post hoc data analysis. In addition, Laufer et al. [222] reported deep learning-based augmented process monitoring for scalable perovskite thin-film fabrication, showing how ML can support early detection of process deviations, thickness non-uniformities, and manufacturing defects during large-area production. Together, these studies demonstrate that machine learning is already being used in practice to improve device performance, process control, and reproducibility in thin-film photovoltaic manufacturing.

ML-based surrogate models for optical and spectral characterization are also expected to become increasingly important for high-throughput experimentation. Rapid extraction of film thickness and optical constants from spectroscopy data using ML can replace computationally expensive fitting routines and enable real-time feedback during deposition [223]. Such capabilities are essential for bridging laboratory-scale optimization with industrial-scale production.

Finally, widespread adoption of ML in thin-film photovoltaics will require continued progress in data availability, standardization, and physical interpretability. Hybrid and physics-informed ML frameworks that embed known physical constraints into learning architectures offer a promising pathway to improve generalization across material systems and enhance model reliability [224,225]. Collectively, these developments indicate that ML will evolve into a central enabling technology for next-generation thin-film solar cells, supporting faster innovation cycles and more robust pathways from material discovery to large-scale manufacturing.

6.4. Additive Manufacturing in Thin-Film Solar Cell Fabrication

Additive manufacturing (AM), or 3D printing, is transforming the photovoltaic sector by facilitating the realization of complex geometries and advanced material architectures that were previously unattainable through traditional subtractive methods. This paradigm shift enables the fabrication of specialized components such as optimized housings for silicon reference cells and high-performance counter electrodes for dye-sensitized solar cells (DSSCs), as well as the potential for fully integrated, printed solar panel systems. For instance, the use of stainless steel–bronze composites in 3D-printed housings has been shown to match or exceed the thermal regulation capabilities of standard aluminum components, thereby ensuring higher precision and reliability in standard solar cell performance characterization [226].

To further illustrate the practical relevance of additive manufacturing for TFSC optimization, several real-life examples may be noted. Akin et al. [227] demonstrated fully additively manufactured (using filament-based material extrusion and cold spray techniques) counter electrodes for dye-sensitized solar cells (DSSCs), reporting photo-conversion efficiencies approximately 2.5 times higher than those of conventional glass-based counterparts due to improved electrical conductivity and favorable microstructural connectivity. James and Contractor [228] further showed that fused deposition modeling 3D printing can be used to implement nature-inspired fractal flexible counter electrodes for DSSCs, emphasizing the role of geometry control in scalable electrode design and device-relevant charge-transfer behavior. At the interconnect level, Jaramillo-Mora et al. [229] reported that laser-assisted AM reinforced the metal–plastic bonding within the structure and enhanced performance of conductive traces with low electrical resistance. These applications of AM are relevant to current collection and series-resistance management in flexible and large-area thin-film photovoltaic modules.

In the context of organic photovoltaics, electrohydrodynamic atomization (EHDA) offers a high-resolution printing alternative for the fabrication of P3HT:PCBM active layers. By precisely controlling parameters such as flow rate and applied voltage to achieve a stable cone-jet mode, researchers can tailor the film’s topography and morphology. Studies indicate that the surface roughness is highly sensitive to processing conditions, typically ranging from $1.938$ to $3.910\mathrm{nm}$, with optimal smoothing occurring at annealing temperatures of 150 °C. Furthermore, higher annealing temperatures have been shown to reduce film mass ($3.235$ to $23.471\mathrm{mg}$) and thickness ($597.5\mathrm{nm}$ to $1.60\mathsf{\mu }\mathrm{m}$) while simultaneously reducing the distance between PCBM clusters by up to $82\%$. This morphological refinement results in a robust visible absorption spectrum peaking at $500\mathrm{nm}$, with absorption intensity scaling proportionally with the number of deposition passes [25].

Beyond terrestrial applications, AM-driven thin-film solar arrays are increasingly critical for space exploration. Their high specific power, mechanical flexibility, and low mass make them ideal for deployment on small spacecraft, inflatable habitats, and even extravehicular spacesuits. To address the labor-intensive nature of traditional assembly, the print-assisted photovoltaic panel assembly (PAPPA) process has been introduced to automate the production of flexible arrays. Rigorous testing of PAPPA-fabricated coupons has demonstrated remarkable durability; the arrays maintain performance stability through folding cycles and even exhibit slight power increases following thermal cycling, proving their resilience in extreme orbital environments [9].

Finally, the development of second-generation thin-film cells continues to address historical limitations such as material scarcity and performance degradation under diffuse lighting. A notable innovation in this area is the implementation of nature-inspired fractal electrodes in amorphous silicon (a-Si) solar cells. By utilizing fractal geometries, researchers have achieved energy densities up to 30 times higher than those of traditional planar electrodes [228]. These bio-inspired designs, coupled with selective electrochemical co-deposition techniques for characterizing sheet resistance and material distribution, represent a significant advancement in the pursuit of high-efficiency, flexible, and sustainable photovoltaic systems.

6.5. Molecular Dynamics Analysis for Advancing Materials and Performance

Molecular dynamics (MD) simulations are emerging as a powerful tool to probe the atomic-scale phenomena that govern thin-film solar cell performance. Recent studies in polycrystalline CdTe, employing atomistic simulations with machine-learned force fields, show that grain boundaries strongly hinder thermal transport. Specifically, a $\Sigma 9$ grain boundary can reduce the lattice thermal conductivity by over $90\%$ compared to a single-crystal film due to enhanced phonon scattering at the defective interface [230]. Such thermal insights are crucial as heat management affects both the efficiency and longevity of thin-film modules.

MD techniques have also been extended to organic photovoltaic (OPV) materials. Simulations of conjugated polymer thin films under strain indicate that mechanical deformation can alter polymer conformation and improve charge transport pathways. For example, stretching a P3HT (poly(3-hexylthiophene)) film by $150\%$ was found to increase the polymer’s conjugation length from ∼2.7 nm to ∼4.7 nm, promoting greater electronic delocalization [231]. These findings shed light on the molecular mechanisms behind the remarkable flexibility and stretchability of next-generation solar cells.

At critical interfaces, MD studies are clarifying how interfacial engineering can mitigate degradation. In silicon heterojunction cells, oxygen doping of the amorphous Si passivation layer was shown to anchor hydrogen and reduce its diffusion toward the crystalline silicon; this helped preserve roughly twice as much post-illumination efficiency gain over 30 days compared to an undoped interface [28].

Together, these advances underscore a broader trend of leveraging atomistic simulation to guide improvements in thin-film photovoltaics. By capturing defect dynamics, ion transport, grain growth, and interfacial phenomena in silico, MD simulations provide insight into failure mechanisms and performance limits that are difficult to observe experimentally. Going forward, this approach points to a promising research direction: integrating high-accuracy MD (increasingly powered by machine learning) with experiments and device-level modeling. Such multi-scale studies will enable researchers to predict how microscopic changes, like improved grain boundary passivation or molecular additives that enhance interfacial adhesion, translate into macroscopic device stability.

7. Conclusions

Thin-film solar cells have emerged as a vital class of photovoltaic technologies for advancing sustainable energy systems, offering distinctive advantages in material efficiency, mechanical flexibility, and manufacturing scalability. Through decades of sustained research, thin-film photovoltaics have evolved from low-efficiency laboratory devices into commercially competitive technologies capable of addressing diverse application requirements beyond those of conventional crystalline silicon solar cells. This review has examined the fundamental principles, device architectures, and performance metrics that define thin-film solar cells, providing a foundation for understanding their operational behavior and efficiency limits in practical applications such as thermal heating, cooling, and energy storage systems.

Major thin-film technologies, including cadmium telluride (CdTe), cadmium selenium telluride (CdSeTe), amorphous silicon (a-Si), copper zinc tin sulfide (CZTS), copper indium gallium selenide (CIGS), organic photovoltaics (OPVs), and perovskites, exhibit application-specific advantages in integrated energy systems. CdTe, CdSeTe, and CIGS, with high absorption and thermal stability, are well suited for photovoltaic thermal heating applications. Amorphous silicon and OPV systems support low-temperature and building-integrated heating and cooling due to their tunable bandgap, light weight, and flexibility. Perovskites and CZTS offer strong potential for cooling and energy storage integration through their tunable optoelectronic properties and compatibility with tandem and hybrid photovoltaic storage systems. Mature platforms such as CdTe, CdSeTe, and CIGS enable reliable heating integration, while emerging materials such as perovskites and kesterites are promising for multifunctional cooling and storage applications. Future work should investigate the coupled optoelectronic and thermal behavior of these systems in integrated heating, cooling, and energy storage configurations, with an emphasis on interface stability, defect evolution under thermal cycling, and performance optimization in multifunctional photovoltaic thermal and storage platforms.

Beyond materials and device engineering, this work highlights the future directions and growing importance of interdisciplinary approaches in overcoming persistent challenges related to efficiency losses, stability, scalability, and cost. Machine learning has become an increasingly powerful tool for modeling complex fabrication spaces, enabling performance prediction, inverse design, process optimization, and automated defect detection. Additive manufacturing and printed electronics have further expanded the design and deployment space of thin-film solar cells by enabling low-waste, high-throughput fabrication on flexible and unconventional substrates. In parallel, molecular dynamics simulations and atomistic modeling have provided critical insight into grain boundary behavior, interfacial stability, thermal transport, and degradation mechanisms, supporting a deeper understanding of performance limitations and long-term reliability.

Economic considerations remain central to the widespread adoption of thin-film photovoltaic technologies. Manufacturing cost structures, life-cycle cost analysis, and levelized cost of electricity are strongly coupled to material availability, fabrication methods, and operational lifetime. Continued progress in scalable manufacturing, coupled with improvements in device stability and efficiency, is essential for achieving cost competitiveness with established energy technologies, particularly for printed and organic photovoltaic modules.

Overall, thin-film solar cells represent a versatile and rapidly advancing photovoltaic platform with the potential to play a critical role in the global transition toward sustainable energy. Future progress will depend on the continued integration of materials science, data-driven modeling, advanced manufacturing, and economic analysis. By fostering closer collaboration across these disciplines, thin-film photovoltaic technologies can move beyond incremental improvements and contribute meaningfully to resilient, efficient, and accessible energy systems worldwide.