Reviewer #1
Reviewer #1 - 1st comment: Page 4, line 39. The manuscript writes that the high viscous encapsulants compromises efficiency, but there is no data to support. And will the small pressure applied to a ~100 nm film on a rigid substrate really influences the PCE?
In brief, we compared the spreadability and damage extent under pressure. Our analysis shows that low-viscosity encapsulants can spread completely under a minimal pressure of 0.03 kPa, equivalent to the weight of the glass substrate. However, high-viscosity encapsulants require more than two orders of magnitude higher pressure to achieve the same spread, clearly demonstrating the potential for damage to the thin (~100 nm) film on a rigid substrate. These findings are detailed in the response to Reviewer #2's 4th comment.
We hope this clarifies the issue and demonstrates the comprehensive analysis we conducted regarding the impact of encapsulant viscosity and applied pressure on device efficiency.
Reviewer #1 - 2nd comment: The manuscript should provide some cross-sectional SEM of the dead zone such as P3 gap, this is the real difference between "Laminar" and "Complete".
In all cases, the adhesion between the encapsulant and the electrode was stronger than the adhesion between the electrode and the BHJ layer. Thus, the delaminated encapsulant retained the electrode, acting as a template. By examining the transmittance differences in transmission mode OM, we can determine if the encapsulant filled specific areas based on the transmittance differences between the encapsulant and the electrode. The Complete encapsulation filled the P1 and P2 scribed areas entirely, including the deepest grooves created by the P2 laser etching process, where energy is most concentrated. In contrast, the Laminar encapsulation did not adequately fill the P2 area. Therefore, the revised Figure S* clearly demonstrates that the Laminar encapsulant did not fill the dead zone, whereas the Complete encapsulant did. - Figure R*: High magnitude cross sectional SEM
- **Figure S***: T-OM
Figure S*. OM images of the dead zone areas (P1, P2, P3) in the delaminated G2G encapsulated devices. (A) Cross-sectional schematic of encapsulant nanoimprint and selective transmission within dead zone areas. Transmission mode OM image of delaminated (B) laminar and (C) complete encapsulants.
Reviewer #1 - 3rd comment: Please check the caption of Figure 1, there seems no description of figure C and D, and the caption of Figure 1B seems wrong, should the green part be glass and the red part be epoxy?
Response: Reviewer 1 pointed out an issue with the caption of Figure 1, noting that there is no description for Figures 1C and 1D, and that the caption for Figure 1B seems incorrect. The green part should be glass and the red part should be epoxy. We apologize for this oversight. The errors have been corrected in revised Figure 1.
Figure 1. Glass-to-Glass Encapsulation for ST-OSMs.
(A) Schematic illustration of the glass-to-glass encapsulation strategy for the three types of encapsulations compared with an unprotected module.
(B-D) UHR FE-SEM cross-sectional images and EDS mapping of the three types of encapsulations. The color maps indicate the characteristic elements of glass (yellow), voids (green), film (red), and epoxy (orange). See also Figure S1.
Reviewer #1 - 4th comment: In Figure 2B, why was the acceptor not significant degraded even in the unprotected device? Can the author provide some explanation?
Reviewer #1 - 5th comment: Why the series connection region is so weak? If the reason is the oxidation of the thin Ag electrode, the weak part is not only the SCR region, but the whole electrode.
In our study, the difference in stability between the active and encapsulated electrode and the inactive electrode is thoroughly discussed in response to Reviewer #3's 8th comment. In brief, the SCR is more susceptible to degradation due to the structural differences in the series connections, which may not be as uniformly protected as the rest of the electrode in the laminar encapsulation.
We hope this explanation clarifies the concerns and demonstrates our comprehensive approach to addressing the encapsulation challenges and their impact on device stability and performance.
Reviewer #2 - 1st comment: ST-OSMs typically show a trade-off between PCE and AVT, which, from there, the figure of merit (or light utilization efficiency, LUE) can be justified. This paper shows a LUE value of 1.94%, which is comparatively very low compared to some of the reported papers with LUE of more than 2.5% LUE for ST-OSMs (>10 cm2 area), E.g., Adv. Mater. 34 (2022) 2110569 and Nano Energy, 121 (2024) 10921. Instead of showing the graph of "PCE vs area", it is also important to show the "LUE vs area". This additional data could potentially enhance the merit of this work and inspire further research in the field.
The relatively low transmittance is primarily due to the limited transmittance of the MoO3/Ag/MoO3 top transparent electrode used in this study. As mentioned in the comprehensive optical analysis in Figure S* in response to Reviewer 3’s 2nd comment, this electrode has lower transmittance compared to other sputtering-based electrodes like ITO or solution-processed electrodes such as AgNWs and PEDOT. Specifically, the PEDOT top electrode exhibited nearly 100% quantum utilization efficiency (QUE; EQE(λ) + T(λ)) for the entire visible spectrum, as studied in our previous paper. However, as stated in the introduction, forming a top transparent electrode uniformly without damage over a large area remains a significant challenge and was unavoidable. This consideration was unnecessary because all previous research was conducted on small areas which is highlighted in the black ellipse in Figure S*.
Current concerns regarding 'stability' and 'scalability' are slowing down the progress in organic photovoltaic (OPV) research. Our research team has achieved a near world-record efficiency of 13.9% over a 200 cm² area for opaque OPVs, as shown in Figure R2. Additionally, as discussed in Reviewer 3's 2nd comment, we have fabricated a 20 cm² OSM with a thinner BHJ layer to enhance LUE, achieving 2.65% LUE (PCE: 8.3%, AVT: 32.0%).
We aim to present excellent references for the practical application of OPVs, particularly focusing on ST-OSMs. Added the above discussion of transmittance and low LUE to the paragraph pointing out the limitations of the study. We appreciate your understanding of this aspect. - Revised manuscript: A comprehensive literature review, presented in Figures 3C and 3D and tabulated in Table S1 and S2, situates our findings within the state-of-the-art efficiency and stability of ST-OSMs with areas larger than 10 cm2. To the best of our knowledge, the achieved T85 lifetime and PCE of 10.37% are unprecedented benchmarks for large-area ST-OSMs. However, the large-area ST-OSM in this work exhibits relatively lower LUE due to the reflection of the OMO electrode used as the top transparent electrode (Figure S). We have not yet found a transparent electrode for ST-OSM that offers higher transmittance, uniformity, and does not damage the underlying interface.* Additionally, this study was conducted in a lab-scale environment, which lacks the controlled humidity, dust, or automation conditions typically found in industrial settings. These factors could influence both the reproducibility and scalability of the results. Thus, fab-scale production may require adaptations to address electrode materials and procedural variances encountered in larger-scale operations. This adaptation is crucial to ensure that our findings can be effectively applied in real-world settings, forming a solid foundation for the commercialization of ST-OSMs. - Revised Figure R → Revision에서만 포함될 피규어.*
- **Figure S***
Figure S*. LUE versus area comparing previous ST-OSM study and this study. Open circles represent LUE values calculated using the PCE for the active area, while solid circles represent LUE values calculated using the PCE for the module area. The ellipses highlight the distribution of data points for comparison.
Reviewer #2 - 2nd comment: The increase in resistance within the SCRs under different encapsulation methods is highlighted as a significant issue. However, the author does not explain why the laminar-sealing method provides poor protection within these regions. Could the authors provide more insights into the different structural weaknesses of the laminar-sealed method compared with the complete-sealed method (epoxy encapsulant)?
In brief, the SCR is more susceptible to degradation due to its exposure to environmental factors and the intrinsic weaknesses in the series connections. The laminar-sealing method may not provide as uniform protection as the complete-sealed method with epoxy encapsulant. Specifically, laminar encapsulation tends to leave micro-gaps and inconsistencies in coverage, making these regions more vulnerable to oxidation and moisture ingress. In contrast, the complete-sealed method with epoxy forms a more continuous and impermeable barrier, effectively protecting the entire electrode, including the SCRs.
We hope this explanation clarifies the concerns and demonstrates our comprehensive approach to addressing the encapsulation challenges and their impact on device stability and performance.
Reviewer #2 - 3rd comment: It is not so obvious why NOA-60 outperforms other types of epoxy in terms of device performance, see Fig S3. Is it merely due to the low viscosity of NOA-60? To what extent can the low viscosity value be generalized to improve device performance after G2G (glass-to-glass) encapsulation? The material, NOA, is well known commercial one and have different code names. What are the effects of other NOAs? Fundamentally, it is necessary to supplement the physicochemical influence analysis of capping materials on devices.
Our hypothesis was to determine the most appropriate encapsulant for large-area organic solar cells by comparing different viscosities within a widely used product range in the mature solar cell market. Due to the sensitivity of OSMs to pressure, UV exposure, and heat, encapsulants that involve these processes are unsuitable.
In this response, NOA was selected as an encapsulant for its short curing time. Initially, the curing process was evaluated by ensuring the surface was free of fingerprints, indicating sufficient polymerization. To further evaluate the protective ability of the encapsulants, we conducted tensile shear adhesive strength experiments with a single lap configuration as shown in Figure S*. (https://doi.org/10.1016/j.ijadhadh.2012.03.007) This figure illustrates the degree of radical polymerization of the epoxy by curing duration.
As the UV curing time increases in the early stages, the encapsulant becomes strong enough to withstand linearly higher stress, indicating that the epoxy polymer becomes denser and thus better protects the underlying layers. We found that at least 30 seconds of curing time is necessary within a short curing period to ensure adequate protection, exceeding the strength of glass. Although we also prepared samples with a 60-second curing time for comparison, which might have greater stretchability, further comparisons were not pursued since the objective was to minimize UV curing time. Consequently, due to its spreadability and fast curing properties, we concluded that NOA is suitable as a complete encapsulant for organic solar cells. We have thoroughly supplemented this content in the manuscript.
Given this rationale, we did not conduct additional experiments with other similar NOA series products, as the results would likely be predictable. However, we acknowledge the importance of comprehensive physicochemical influence analysis among various encapsulants. For more detailed information, please refer to our response to Reviewer #3's fifth comment, where we address this analysis in detail. - Revised Experimental Section: Tensile shear adhesive strength evaluation were conducted using glass substrates and encapsulation materials. A 25.4 mm × 25.4 mm piece of encapsulant was applied to the cleaned glass substrate (101.6 mm × 25.4 mm with a thickness of 1.1T) for single lap joint configuration.[ref] Then, the specimens were systematically subjected to UV curing for various durations (1 ~ 60 s). The tensile shear adhesive strength of each specimen was measured using a Universal Testing Machine (UTM), applying a tensile load at a constant crosshead speed until failure occurred. - Revised Manuscript: [레퍼 추가 절대 기억해] - Revised Figure S*
Figure S*. Tensile shear adhesive strength of the encapsulant as a function of UV curing duration. As the UV curing time increases, the encapsulant exhibits higher tensile strength and strain capacity, indicating enhanced polymerization and density of the epoxy.
Reviewer #2 - 4th comment: The device with 3035b and 3124L shows a significant drop in almost all parameters, Jsc, FF, and Voc. How is it possible if the encapsulant is located on the bottom side of the illumination light, i.e., no parasitic absorption from the encapsulant? It can also be inferred from the I-V curve that the series resistance is increasing significantly using 3035b; what would be the reason behind this phenomenon?
The low-viscosity NOA-60 spread across the entire surface with approximately 25 Pa pressure from the weight of the 1.1T glass. In contrast, 3035b and 3124L require pressures of 30 kPa and 50 kPa, respectively, using a hot press machine at 23 °C to completely spread due to their high viscosity. To further illustrate the significant effect of the much higher pressure required for these encapsulants, we have included photographs and EL mapping images of 20 cm² devices after pressure application. Note that the EL mapping image of the fresh 20 cm² device is located in Figure S*.
As shown in the EL map in Figure S*, the areas where pressure was applied show direct damage. This confirms that excessive pressure during encapsulation can severely impact the series resistance of OSMs, as described in the original manuscript. The applied pressure caused the areas to lose their optoelectrical features, significantly degrading the performance of the solar cells. Unlike inorganic semiconductors (e.g., CIGS), OSMs are typically only a few hundred nanometers thick and are sensitive to pressure. - Revised Experimental Section: Encapsulation of ST-OSMs: For G2G encapsulation, cover glass substrates (15 × 16.5 cm) were cleaned, as described above. NOA-60 epoxy encapsulant was poured onto the module’s front side, spread evenly under the weight of the cover glass. “To ensure the complete spreading of the highly viscous ThreeBond 3035b and 3124L epoxy encapsulants on the module surface, pressures of 30 kPa and 50 kPa, respectively, were applied using a hot press machine at room temperature.” After spreading, epoxy encapsulants were cured with UV light at an intensity of 3 J/cm² for 30 s. - Revised Figure S*:
**Figure S***: The photograph of ThreeBond 3035b encapsulated 200 cm² module, stored for 18 months in ambient conditions.
**Figure S***: The photograph of ThreeBond 3035b encapsulated 200 cm² module, stored for 18 months in ambient conditions.
- **Revised Figure S***: small area module → [3035b, 3124L] x [Low, Moderate, High pressures] 6종 + NOA 1종, 최소 총 7종에 대한 Photography (spreadability) 및 EL Mapping 결과. [월요일 수행]
Reviewer #2 - 5th comment: Is the AVT of 18.78% measured only on BHJ film or the complete device with its corresponding electrode, capping material or other also?
Reviewer #2 - 6th comment: The author also needs to provide the details of the substrate used for the large area device, including how many submodules or sub-cells are comprised to support the module. A photograph of a large area (or geometry) of the ST-OPV device is encouraged.
Reviewer #2 - 7th comment: The scalability section mentions the occurrence of local hotspots in large-area modules, leading to a reduction in fill factor (FF). Could the authors provide a more thorough explanation of the distribution and causes of these hotspots? This would significantly enhance the reader's understanding of the issue. I suggest referring to the following paper: https://doi.org/10.1063/1.3073857