A high-pressure isostatic lamination technique to fabricate versatile carbon electrode-based perovskite solar cells
一种高压等静压层压技术,用于制造基于碳电极的多功能钙钛矿太阳能电池
Perovskite solar cells (PSCs) with evaporated gold (Au) electrodes have shown great efficiencies, but the maturity of the technology demands low-cost and scalable alternatives to progress towards commercialisation. Carbon electrode-based PSCs (C-PSCs) represent a promising alternative, however, optimising the interface between the hole transport layer (HTL) and the carbon electrode without damaging the underlying functional layers is a persistent challenge. Here, we describe a lamination technique using an isostatic press that can apply exceedingly high pressure to physically form an HTL/carbon interface on par with vacuum-evaporated electrodes, without damaging the device. Research-scale C-PSCs with a power conversion efficiency (PCE) of up to 20.8% are demonstrated along with large-area C-PSCs with PCEs of 19.8% and 16.9% for cell areas of 0.95 cm2 and 5.5 cm2 , respectively. The unencapsulated C-PSCs significantly outperform the Au-electrode devices in accelerated operational stability testing (ISOS-L-1), retaining 84% of the initial PCE after 1000 h. Additionally, this versatile technique is also used to fabricate flexible, roll-to-roll printed C-PSCs with efficiencies of up to 15.8%.
带有蒸发金 (Au) 电极的钙钛矿太阳能电池 (PSC) 已显示出很好的效果,但该技术的成熟需要低成本和可扩展的替代方案才能迈向商业化。基于碳电极的 PSC (C-PSC) 代表了一种很有前途的替代方案,然而,在不损坏底层功能层的情况下优化空穴传输层 (HTL) 和碳电极之间的界面是一项持续的挑战。在这里,我们描述了一种使用等静压机的层压技术,该技术可以施加极高的压力,以物理方式形成与真空蒸发电极相当的 HTL/碳界面,而不会损坏器件。证明了功率转换有效效率和效率 (PCE) 高达 20.8% 的研究规模 C-PSC 以及 0.95 cm 2 和 5.5 cm 2 细胞面积的 PCE 分别为 19.8% 和 16.9% 的大面积 C-PSC 。未封装的 C-PSC在加速操作稳定性测试 (ISOS-L-1) 中明显优于 Au 电极器件,在 1000 小时后保留了 84% 的初始 PCE。此外,这种多功能技术还用于制造效率高达 15.8% 的灵活、卷对卷印刷的C-PSC。
Perovskite solar cells (PSCs) have been developed rapidly in the past decade, with their record power conversion efficiency (PCE) now exceeding 26%1 . While gold (Au) serves as the preferred back contact electrode for these highly efficient PSCs, its material cost and energy-intensive thermal vacuum evaporation hinder the low-cost and high-throughput device fabrication required for commercial production2,3 . Carbon-based electrodes have proven to be a promising alternative, particularly due to their suitable work function (5.0 eV) and electrical conductivity for charge transfer4,5 . These solution-processable electrodes also offer improved stability and can be adapted to scalable roll-to-roll (R2R) deposition6–10. Replacing the evaporated metal gold electrode with a solution-processable alternative also offers significant cost savings in the manufacture of PSC products. Various cost models have been developed throughout the literature, each indicating that the costs associated with a vacuum-deposited electrode (including the material cost of gold and the equipment purchase and running costs of a high-throughput evaporator) are the highest cost component for the whole PSC stack11,12. For example, recently, it was predicted that the cost reduction of transitioning from evaporated gold to screen-printed carbon and silver will reduce the cost from ~150 USD m−2 to less than 100 USD m−2 for fully printed R2R C-PSCs13. However, carbon electrode-based PSCs (C-PSCs) often suffer from performance loss due to the sensitivity of the underlying functional layers to adverse solvents and high-temperature processing, along with poor interface contact between the electrode and the underlying device stack。
钙钛矿太阳能电池 (PSC) 在过去十年中发展迅速,其创纪录的功率转换效率 (PCE) 现已超过 26%1。虽然金 (Au) 是这些高效 PSC 的首选背接触电极,但其材料成本和能源密集型热真空蒸发阻碍了商业生产所需的低成本和高通量器件制造2,3。碳基电极已被证明是一种很有前途的替代品,特别是因为它们具有合适的功函数 (5.0 eV) 和电荷转移的导电性4,5。这些可溶液加工的电极还具有更高的稳定性,可适应可扩展的卷对卷 (R2R) 沉积6–10。用可溶液加工的替代品代替蒸发额定金属金电极,还可以显著节省 PSC 产品的制造成本。文献中已经开发了各种成本模型,每个模型都表明与真空沉积电极相关的成本(包括金的材料成本和高通量蒸发器的设备购买和运行成本)是整个 PSC 堆栈的最高成本组成部分11,12。例如,最近预测,从蒸发金过渡到丝网印刷碳和银的成本降低将使完全印刷的 R2R C-PSC 的成本从 ~150 USD m-2 降低到不到 100 USD m-213。然而,由于底层功能层对不利溶剂和高温加工的敏感性,以及电极与底层器件堆栈之间的界面接触不良,基于碳电极的 PSC (C-PSC) 经常遭受性能损失。
While carbon pastes can be directly deposited onto the underlying device stack, lamination techniques such as pneumatic hot pressing have been employed to address concerns related to solvent leaching and hightemperature processing16–18. Recent reports on laminated C-PSCs have showcased competitive efficiencies compared to control devices having evaporated Au electrodes, with the best research-scale C-PSCs now surpassing 20% PCE1. Furthermore, laminated carbon electrodes have been used to fabricate perovskite solar modules, showcasing efficiencies of up to 16.01% (10 cm2 active area) using a playdough-like carbon electrode23. However, the widely used pneumatic plate-to-plate press lamination method poses several limitations. Firstly, the pressure that can be applied by these systems is greatly limited by the risk of damaging the cell due to uneven pressure distributions, with reported pressures for C-PSC lamination typically below 1 MPa, but ranging up to 30 MPa24,25. As such, the carbon surface is often specifically modified to achieve the appropriate adhesion20. Secondly, this process is limited by its scalability, requiring one-by-one device fabrication to achieve high
虽然碳浆可以直接沉积在底层器件堆栈上,但已采用气动热压等层压技术来解决与溶剂浸出和高温加工相关的问题16-18。最近关于层压 C-PSC 的报告显示,与具有蒸发 Au 电极的对照设备相比,其效率具有竞争力,目前最好的研究规模的 C-PSC 已超过 20% PCE1。此外,叠层碳电极已被用于制造钙钛矿太阳能组件,使用橡皮泥状碳电极的效率高达 16.01%(10 cm2 有效面积)23。然而,广泛使用的气动板对板压合方法存在一些局限性。首先,由于压力分布不均匀,这些系统可以施加的压力受到损坏电池的风险的极大限制,据报道,C-PSC 层压的压力通常低于 1 MPa,但最高可达 30 MPa24,25。因此,碳表面通常经过专门改性,以实现适当的附着力20。其次,该工艺受其可扩展性的限制,需要逐个制造器件才能实现高
efficiency. Additionally, achieving uniform pressure distribution across large surfaces, while also preventing localised stress concentrations, has hindered the development of large-area C-PSCs with comparable performance to research-scale cells.
效率。此外,在大表面上实现均匀的压力分布,同时也防止了局部应力集中,这阻碍了大面积 C-PSC 的开发,其性能可与研究规模的细胞相媲美。
Isostatic pressing provides a suitable solution to overcome these challenges as the pressure being applied to the sample within the isostatic chamber is transmitted undiminished in all directions. Therefore, isostatic pressing permits the application of exceedingly high pressure without the risk of damaging the device, while also mitigating the risk of localised stress concentrations on the sample. Additionally, this technique is suitable for large-area cells and modules as the pressure is applied uniformly across the entire surface area. Cold isostatic pressing (CIP) has been reported for the fabrication of PSCs on only several occasions, with reported efficiencies of up to 11.6%27,28. This has also been demonstrated as a scalable manufacturing technique, as Dexit et al. investigated its feasibility for the scalable fabrication of solid-state battery components, including an extensive techno-economic analysis of the process.
等静压为克服这些挑战提供了合适的解决方案,因为施加在等静压室内样品上的压力在所有方向上都不减弱。因此,等静压允许施加极高的压力,而不会损坏设备,同时还降低了样品上局部应力集中的风险。此外,该技术适用于大面积电池和模块,因为压力均匀地施加在整个表面积上。据报道,冷等静压 (CIP) 仅用于制造 PSC 的次数,据报道效率高达 11.6%27,28。这也已被证明是一种可扩展的制造技术,因为 Dexit 等人研究了其可扩展制造固态电池组件的可行性,包括对该工艺的广泛技术经济分析。
In this work, CIP is used for the fabrication of highly efficient C-PSCs, by laminating a bilayer electrode comprising a coated carbon and coated silver (Ag) film onto the HTL of various PSC device stacks. The CIP lamination technique is shown to apply very high mechanical pressure (up to 380 MPa) to form a seamless physical connection between the HTL and carbon film, without damaging the device, without heat and without the need for additional morphology modifications to the carbon film. Using this technique, we report research-scale C-PSCs with PCEs of over 20%, matching the performance of the control cells with evaporated Au electrodes. Furthermore, the advantages of high and uniform pressure are showcased as we report record efficiencies for large-area C-PSCs of up to 19.8% and 16.9% for cell areas of 0.95 cm2 and 5.5 cm2 , respectively. The C-PSCs exhibit remarkable stability, greatly outperforming the control devices after 1000 h of accelerated operational testing (ISOS-L-1). Finally, the broad applicability of this method is demonstrated by laminating the coated electrodes onto flexible, R2R-processed PSCs exhibiting PCEs as high as 15.8% and those with custom screen-printed electrode designs. All the C-PSCs in this work are fabricated entirely using low-temperature processes (≤150 °C) to align with high-throughput R2R fabrication processes and to minimise overall energy expenditure.
在这项工作中,CIP 用于将包含涂层碳和涂层银 (Ag) 膜的双层电极层压到各种 PSC 器件堆栈的 HTL 上,从而制造高效的 C-PSC。CIP 层压技术被证明可以施加非常高的机械压力(高达 380 MPa),以在 HTL 和碳膜之间形成无缝的物理连接,而不会损坏设备,无需加热,也不需要对碳膜进行额外的形态修改。使用这种技术,我们报告了 PCE 超过 20% 的研究规模的 C-PSC,与具有蒸发 Au 电极的对照细胞的性能相匹配。此外,我们报告了大面积 C-PSC 的创纪录效率,分别为 0.95 cm2 和 5.5 cm2,高和均匀压力的优势分别高达 19.8% 和 16.9%。C-PSCs 表现出显着的稳定性,在加速操作测试 (ISOS-L-1) 1000 小时后性能大大优于控制装置。最后,通过将涂层电极层压到柔性的 R2R 加工的 PSC 上,证明了该方法的广泛适用性,这些 PSC 的 PCE 高达 15.8%,并且具有定制的丝网印刷电极设计。这项工作中的所有 C-PSC 均完全使用低温工艺 (≤150 °C) 制造,以与高通量 R2R 制造工艺保持一致,并最大限度地减少整体能源消耗。
Results and discussion
结果与讨论
The CIP lamination technique CIP involves submerging a sample into a chamber of ambient-temperature fluid, either liquid or gas, which is then isostatically pressurised. For this lamination process, two separate device components were required: the PSC device stack fabricated up to the HTL, and the coated bilayer electrode, as shown in Fig. 1a. In this work, both rigid and flexible R2R PSCs were used. The coated bilayer electrode was prepared as described in our previous work16. Briefly, a commercial silver (Ag) paste was bar coated onto a nonstick “release layer” (silicone-coated PET film) and dried on a hotplate at 135 °C for 2 min, resulting in a dry film thickness of ~20 µm. A commercial carbon-based paste was then bar coated onto the dry Ag film, followed by another brief drying stage on a hotplate at 135 °C, resulting in an optimised dry film thickness of ~60 µm. Both the Ag paste and carbon-based paste were used without modification and there were no additional steps to modify the morphology of the resulting carbon film.
CIP 层压技术 CIP 涉及将样品浸入环境温度流体(液体或气体)室中,然后进行等静压。对于这种层压工艺,需要两个单独的器件组件:制造到 HTL 的 PSC 器件堆栈,以及涂层双层电极,如图 1a 所示。在这项工作中,同时使用了刚性和柔性 R2R PSC。按照我们之前的工作16 中的描述制备了涂层双层电极。简而言之,将市售银 (Ag) 浆料涂布在不粘“离型层”(硅涂层 PET 薄膜)上,并在 135 °C 的热板上干燥 2 分钟,得到 ~20 μm 的干膜厚度。然后将商业碳基浆料棒涂在干 Ag 薄膜上,然后在 135 °C 的加热板上进行另一个短暂的干燥阶段,从而获得 ~60 μm 的最佳干膜厚度。Ag 浆料和碳基浆料均未经改性使用,并且没有额外的步骤来改变所得碳膜的形态。
The flexible bilayer electrode on the release layer was cut to size, placed lightly onto the underlying PSC stack fabricated up to the HTL, and sealedin a vacuum-seal bag as shown in Fig. 1b. The samples were then submerged into the water-filled CIP chamber. The sealed bag held the loose electrode in position before the application of pressure and effectively protected the devices from the water. The chamber was then pressurised, applying evenly distributed pressure to the submerged samples, and held at the set pressure for 30 s to form a strong bond between the carbon layer and the HTL, as illustrated in Fig. 1c. The sample was then removed from the sealed bag and the release layer was peeled away from the electrode, exposing the coated Ag film for electrical connection. The final device structure is shown in Fig. 1d, incorporating tin(IV) oxide (SnO2) as the electron transport layer (ETL) and 2,2′,7,7′-tetrakis-(N,N-di-4-methoxyphenylamino)−9,9′-spirobifluorene (Spiro-OMeTAD) as the HTL.
将释放层上的柔性双层电极切割成一定尺寸,轻轻地放在制造到 HTL 的底层 PSC 堆栈上,并密封在真空密封袋中,如图 1b 所示。然后将样品浸入充满水的 CIP 室中。密封袋在施加压力之前将松散的电极固定到位,并有效地保护设备免受水的影响。然后对腔室加压,对浸没样品施加均匀分布的压力,并在设定压力下保持 30 秒,以在碳层和 HTL 之间形成牢固的键,如图 1c 所示。然后从密封袋中取出样品,从电极上剥去离型层,露出涂层的 Ag 薄膜以进行电气连接。最终的器件结构如图 1d 所示,采用氧化锡 (SnO2) 作为电子传输层 (ETL),2,2',7,7'-四基-(N,N-di-4-甲氧基苯氨基)−9,9'-螺芴 (Spiro-OMeTAD) 作为 HTL。
The entire isostatic lamination process of pressurising, holding for 30 s, and depressurising took 3 min, and with the lab-scale instrument used in this work (chamber diameter 10 cm, depth 55 cm), tens of devices were typically pressed simultaneously. Photos of the CIP apparatus showing the operational scale are provided in Fig. S1. The resulting C-PSCs after isostatic pressure are shown in Fig. 1e.
加压、保持 30 s 和减压的整个等静压层压过程需要 3 min,并且使用这项工作中使用的实验室规模仪器(腔室直径 10 cm,深度 55 cm),通常同时压制数十个设备。显示操作规模的 CIP 设备的照片如图 S1 所示。等静压后得到的 C-PSC 如图 1e 所示。
Conductivity of the coated carbon/Ag electrode
涂层碳/Ag 电极的电导率
The choice of electrode materials and fabrication processes plays a pivotal role in achieving efficient C-PSCs. In the bilayer electrode arrangement, the carbon film serves several roles: to create a strong bond to the underlying HTL due to its high porosity and roughness, to facilitate vertical charge transfer through its thickness, and to act as a relatively thick protective buffer layer to prevent ion migration and interaction between the adjacent Ag electrode film and the underlying perovskite. The planar carbon film used in this work has a sheet resistance of roughly 8.7 Ω □−1 at an approximate dry-film thickness of 60 µm. However, relying solely on the carbon electrode is not suitable for fabricating large-area C-PSCs due to the substantial increase in the series resistance within the cell, which would result in a large drop in the fill factor (FF). Therefore, a highly conductive top contact is required to fabricate efficient large-area C-PSCs. In this approach, a conductive silver paste is employed, a choice driven by its ease of printability (and hence scalability) and high electrical conductivity. The Ag dry film, with a thickness of ~20 µm, results in a sheet resistance of roughly 70 mΩ □−1 . When combined, the resulting bilayer carbon/Ag electrode has a sheet resistance of roughly 26 mΩ □−1 as measured from the carbon side of the electrode.
电极材料和制造工艺的选择在实现高效的 C-PSC 方面起着关键作用。在双层电极布置中,碳膜起着多种作用:由于其高孔隙率和粗糙度,与底层 HTL 形成牢固的结合,促进其通过其厚度进行垂直电荷转移,并充当相对较厚的保护缓冲层,以防止离子迁移和相邻 Ag 电极膜与底层钙钛矿之间的相互作用。这项工作中使用的平面碳膜在大约 60 μm 的干膜厚度下具有大约 8.7 Ω □−1 的片状电阻。然而,由于电池内的串联电阻大幅增加,仅依靠碳电极并不适合制造大面积 C-PSC,这会导致填充因子 (FF) 大幅下降。因此,需要高导电性的顶部触点来制造高效的大面积 C-PSC。在这种方法中,采用了导电银浆,这种选择是由其易于印刷性(以及因此的可扩展性)和高导电性驱动的。厚度为 ~20 μm 的 Ag 干膜产生大约 70 mΩ □−1 的薄层电阻。当组合时,所得的双层碳/Ag 电极具有大约 26 mΩ □−1 的薄层电阻,从电极的碳侧测量。
When the CIP pressure was applied, the thickness, morphology and conductivity of the coated electrode films were altered. Before lamination, the thickness of the coated films was ~20 µm and ~60 µm for the coated Ag and carbon layers, respectively. Following the application of pressure, both films were compressed, with the Ag film reducing to ~16 µm and the carbon film reducing to ~40 µm, as shown in Fig. S2. The application of this extreme pressure also influences the morphology of the carbon film. The SEM images in Fig. S3 and the 3D profiles in Fig. S4 show the compaction of the carbon film with increasing pressure, particularly as the graphite flakes present in the carbon film are pushed into the carbon film and tend to orient parallel to the surface of the underlying substrate. In comparison, the SEM image of the pristine carbon film shows graphite flakes oriented randomly and protruding from the film. This is clearly shown in Fig. S4 as the surface area roughness decreases from 1.227 µm to 0.442 µm on the application of 240 MPa lamination pressure (Table S1).
当施加 CIP 压力时,涂层电极膜的厚度、形态和电导率会发生变化。层压前,涂布的 Ag 层和碳层的涂布薄膜厚度分别为 ~20 μm 和 ~60 μm。施加压力后,两层薄膜都被压缩,Ag 膜减少到 ~16 μm,碳膜减少到 ~40 μm,如图 S2 所示。这种极端压力的应用也会影响碳膜的形态。图 1 中的 SEM 图像。S3 和图 3 中的 3D 剖面图S4 显示碳膜随着压力的增加而压实,特别是当碳膜中存在的石墨片被推入碳膜并趋于平行于底层基材的表面时。相比之下,原始碳膜的 SEM 图像显示石墨片随机定向并从薄膜中突出。这在图 1 中清楚地显示出来。S4 在施加 240 MPa 层压压力时,表面积粗糙度从 1.227 μm 降低到 0.442 μm(表 S1)。
The compaction of the electrode film has a direct impact on its electrical conductivity. Isostatic pressing of the electrode at 240 MPa resulted in a 30% reduction in sheet resistance, down to ~18 mΩ □−1 . This reduction can be attributed to the pressure-induced compaction of particles within the film, subsequently improving the interparticle connectivity and conductivity. In comparison, the sheet resistance of an 80 nm evaporated Au film was measured to be ~575 mΩ □−1 . Consequently, the superior conductivity of the coated carbon/Ag electrode, combined with the ease of fabrication and lower cost, is more suitable for the scalable production of PSCs, providing a viable alternative to traditional evaporated Au films. The results of electrode conductivity measurements are provided in Table S2.
电极膜的压实直接影响其电导率。在 240 MPa 下对电极进行等静压导致薄层电阻降低 30%,降至 ~18 mΩ □−1 。这种减少可归因于薄膜内颗粒的压力诱导压实,从而改善了颗粒间的连通性和导电性。相比之下,经测得 80 nm 蒸发的 Au 薄膜的薄层电阻为 ~575 mΩ □−1 。因此,涂层碳/Ag 电极的卓越延展性,加上易于制造和较低的成本,更适合 PSC 的放大生产,为传统蒸发 Au 薄膜提供了可行的替代方案。电极电导率测量的结果见表 S2。
Influence of lamination pressure on the carbon/HTL interface
层压压力对碳/HTL 界面的影响
The interface between the HTL and the electrode plays a crucial role in charge extraction, ultimately affecting the device’s performance. To examine the influence of isostatic pressure on this interface, the CIP pressure was varied from 35 MPa to the maximum allowable pressure of 380 MPa. The 35 MPa pressure variant was insufficient to properly adhere the coated carbon/Ag electrode to the underlying HTL, resulting in the electrode easily delaminating from the device when the release layer was peeled away (Supplementary Movie 1). The cross-sectional Scanning Electron Microscopy (SEM) image of the 35 MPa sample is shown in Fig. 2a and indicates significant regions of electrode delamination from the underlying device. At 100 MPa the adhesion improved, and the non-stick release layer was removed without causing any visible delamination of the electrode.
HTL 和电极之间的界面在电荷提取中起着至关重要的作用,最终影响器件的性能。为了检查等静压对该界面的影响,CIP 压力从 35 MPa 变化到最大允许压力 380 MPa。35 MPa 压力变体不足以将涂层碳/Ag 电极正确粘附到下面的 HTL 上,导致当释放层被剥离时,电极很容易从器件上分层(补充视频 1)。35 MPa 样品的横截面扫描电子显微镜 (SEM) 图像如图 2a 所示,并显示了底层器件的电极分层的重要区域。在 100 MPa 时,粘附力得到改善,不粘脱模层被去除,而不会导致电极出现任何可见的分层。
However, further destructive analyses showed that the electrode could be easily peeled away from the underlying HTL due to weak bonding at the interface (Supplementary Movie 2), evident by the small cavities seen in the cross-sectional SEM image (Fig. 2b). Pressures of 170 MPa, 240 MPa and 380 MPa resulted in strong adhesion between the carbon electrode and the underlying Spiro-OMeTAD HTL with an almost seamless connection at the interface as shown in Fig. 2c. As per Fig. 2d–g, no visible differences were noted for PSC precursor stacks up to the HTL (no electrode) across all pressures. The quality of the interface between the HTL and carbon electrode directly corresponded with the C-PSC photovoltaic performance, presented in Fig. 2h, showing that the best performance was achieved at pressures of 170 MPa and above which was directly related to an improvement in the FF (Fig. S5). The open circuit voltage (Voc) and short circuit current (Jsc) were only marginally impacted by the pressure above 35 MPa (Fig. S5).
然而,进一步的破坏性分析表明,由于界面处的粘合较弱,电极很容易从下面的 HTL 上剥离(补充视频 2),横截面 SEM 图像中看到的小腔就证明了这一点(图 2b)。170 MPa、240 MPa 和 380 MPa 的压力导致碳电极和下面的 Spiro-OMeTAD HTL 之间具有很强的粘附力,界面处几乎无缝连接,如图 2c 所示。根据图 2d-g,在所有压力下,直到 HTL(无电极)的 PSC 前驱体堆栈均未观察到明显差异。HTL 和碳电阻之间的界面质量与 C-PSC 光伏性能直接对应,如图 2h 所示,表明在 170 MPa 及以上的压力下实现了最佳性能,这与 FF 的改善直接相关(图 S5)。开路电压 (Voc) 和短路电流 (Jsc) 仅受 35 MPa 以上压力的轻微影响(图 S5)。
Interestingly, the maximum pressure of 380 MPa resulted in a slight decrease in the PCE, which can be attributed to a drop in FF (Fig. S5). Upon further investigation, it was found that this was due to the occurrence of shunts and micro-shorts due to extreme local pressure, resulting in a drop in shunt resistance. Thermal maps of the devices with an applied voltage bias (1 V forward bias) revealed a clear trend that these short circuits were occurring around the edge of the laminated electrode. This is shown in Fig. S6, where shunts are observed around the edge of the laminated electrode. Although the CIP applies uniform pressure across the device’s surface, the thickness disparity between the electrode and the surface of the cell (roughly 100 µm in total, including the release layer) coupled with the thickness of the vacuum-seal bag used, prevents the bag from conforming precisely to the contour of the electrode. Consequently, under these extreme pressure conditions, pressure becomes concentrated at the edges of the electrode. Optimisation of the electrode design and using a thinner vacuum-sealing bag would mitigate this phenomenon.
有趣的是,380 MPa 的最大压力导致 PCE 略有下降,这可以归因于 FF 的下降(图 S5)。经过进一步调查,发现这是由于极端的局部压力导致分流和微短路的发生,导致分流电阻下降。施加电压偏置(1 V 正向偏置)的器件的热图揭示了一个明显的趋势,即这些短路发生在叠层电极的边缘周围。如图 1 所示。S6,在层压电极的边缘周围观察到分流。尽管 CIP 在设备表面施加均匀的压力,但电极和电池表面之间的厚度差异(总共约 100 μm,包括释放层)加上所用真空密封袋的厚度,使袋子无法精确贴合电极的轮廓。因此,在这些极端压力条件下,压力集中在电极的边缘。优化电极设计并使用更薄的真空密封袋将缓解这种现象。
To further investigate the effect of lamination pressure, the CIP hold time was increased from 30 s to 10 min for 35 MPa and 240 MPa samples. The longer hold time at 35 MPa (35* sample set in Fig. 2h) significantly improved the C-PSC performance, closely matching the results for the 100 MPa sample. However, the performance was still ~20% lower than samples processed at pressures of 170 MPa and above. Conversely, the samples processed at 240 MPa for 10 min (240* sample set in Fig. 2h) showed no further improvement in PCE compared to the 30 s samples and even exhibited a slight drop in performance. Due to throughput considerations, lengthy hold times are undesirable for scaled production, making the 240 MPa, 30 s parameter set the most appropriate for the susequent experiments.
为了进一步研究层压压力的影响,将 35 MPa 和 240 MPa 样品的 CIP 保持时间从 30 s 增加到 10 min。在 35 MPa 下的较长保持时间(图 2h 中的 35* 样品组)显著提高了 C-PSC 性能,与 100 MPa 样品的结果非常接近。然而,性能仍比在 170 MPa 及以上压力下处理的样品低 ~20%。相反,在 240 MPa 下处理 10 分钟的样品(图 2h 中的 240* 样品组)与 30 s 样品相比,PCE 没有进一步改善,甚至表现出性能略有下降。由于通量限制,长时间的保持时间对于规模化生产是不可取的,这使得 240 MPa、30 s 参数设置最适合后续实验。
Electrical Impedance Spectroscopy (EIS) was also employed to understand the interfacial properties of the C-PSCs. Figure 2i compares the EIS results from samples prepared at 35 MPa (the best working sample from this set), 100 MPa, 170 MPa, 380 MPa, and a control Au electrode sample. The resistive and capacitive elements of the results were quantified by applying a Nyquist equivalent circuit, as shown by the inset in Fig. 2i. Constant phase elements (CPE) accounted for the nonideality of the lowfrequency capacitive element (CLF). The EIS response magnitude decreased with increasing CIP pressure, directly correlating with the parameters obtained from the Nyquist fit (see Table S3). Notably, samples processed at 35 MPa and 100 MPa resulted in a significantly higher low-frequency resistance (RLF) than those processed at 170 MPa, 380 MPa and the Au control sample, indicating inadequate interface contact between the HTL and carbon electrode.
电阻抗谱 (EIS) 也被用来了解 C-PSC 的界面特性。图 2i 比较了在 35 MPa(该组中最好的工作样品)、100 MPa、170 MPa、380 MPa 和对照 Au 电极样品制备的样品的 EIS 结果。通过应用奈奎斯特等效电路对结果的电阻和电容元件进行量化,如图 2i 中的插图所示。恒定相位元件 (CPE) 解释了低频电容元件 (CLF) 的非理想性。EIS 响应幅度随着 CIP 压力的增加而降低,与从 Nyquist 拟合获得的参数直接相关(见表 S3)。值得注意的是,在 35 MPa 和 100 MPa 下处理的样品导致比在 170 MPa、380 MPa 和 Au 对照样品下处理的样品具有明显更高的低频电阻 (RLF),这表明 HTL 和碳电极之间的界面接触不足。
In terms of quantifying the bonding of the carbon electrode to the underlying Spiro-OMeTAD HTL, a tape peel-off test was conducted to provide a preliminary assessment of this bonding strength. It’s important to note that this test will only reveal the weakest interface within the entire device stack. Results indicated that the weakest interface was between the perovskite and the Spiro-OMeTAD layer, with the exposed perovskite layer being observed upon removing the tape. This finding is consistent with previous studies of the mechanical integrity of PSCs subjected to double cantilever beam experiments34. Consequently, it can be inferred that the
在量化碳电极与底层 Spiro-OMeTAD HTL 的粘合方面,进行了胶带剥离测试,以初步评估这种粘合强度。请务必注意,此测试只会揭示整个设备堆栈中最弱的接口。结果表明,最弱的界面是在钙钛矿和 Spiro-OMeTAD 层之间,在去除胶带时观察到暴露的钙钛矿层。这一发现与之前对双悬臂梁实验下 PSC 机械完整性的研究一致34。因此,可以推断出
bonding between the carbon electrode and the Spiro-OMeTAD layer must be stronger than between the HTL and the perovskite layer.
碳电极和 Spiro-OMeTAD 层之间的键合必须比 HTL 和钙钛矿层之间的键合更强。
Optimised device performance
优化的设备性能
The current density-voltage (J-V) curves for the best research-scale C-PSC (masked area of 0.16 cm2 ) in forward and reverse scan directions are presented in Fig. 3a. The champion C-PSC exhibited a reverse-scan PCE of 20.8% (Voc = 1.10 V,Jsc = 23.2 mA cm−2 , FF = 81.3). Minimal hysteresis was observed, with a forward-scan PCE of 20.4% (Voc = 1.10 V, Jsc = 23.2 mA cm−2 , FF = 80.9) and a stabilised PCE of 20.3% after 100 s at maximum power point (MPP) tracking. The reproducibility of the CIP lamination technique is demonstrated by the box plots in Fig. 3b, comparing the PCE of cells with laminated carbon-only electrodes, laminated carbon/ Ag bilayer electrodes and evaporated Au electrodes. While the devices with evaporated Au electrodes display a slightly narrower distribution and higher average PCE, the high-quality physical interface between the HTL and carbon formed using the high-pressure CIP lamination method enables the reliable fabrication of efficient C-PSCs, consistently matching the perfor mance of their costly evaporated Au counterparts. Therefore, this physical lamination technique provides a suitable third alternative to atomically deposited electrodes (thermal vacuum evaporation) and molecularly deposited electrodes (direct solution deposition). The J–V curve of the best device with an evaporated Au electrode compared to the laminated carbon and laminated bilayer carbon/Ag electrodes is provided in Fig. S7. Furthermore, the batch data for devices shown in Fig. 3b is also provided in Table S5, showing the specific Voc, Jsc, FF and PCE for cells with different electrode configurations.
图 3a 显示了最佳研究规模 C-PSC(掩蔽面积为 0.16 cm2 )在正向和反向扫描方向上的电流密度-电压 (J-V) 曲线。冠军 C-PSC 的反向扫描 PCE 为 20.8% (Voc = 1.10 V,Jsc = 23.2 mA cm-2 ,FF = 81.3)。观察到最小的滞后,前向扫描 PCE 为 20.4% (Voc = 1.10 V,Jsc = 23.2 mA cm-2 ,FF = 80.9),在最大功率点 (MPP) 跟踪 100 秒后稳定 PCE 为 20.3%。图 3b 中的箱线图证明了 CIP 层压技术的可重复性,该图比较了具有层压纯碳电极、层压碳/Ag 双层电极和蒸发 Au 电极的电池的 PCE。虽然具有蒸发 Au 电极的器件显示出略窄的分布和较高的平均 PCE,但使用高压 CIP 层压方法形成的 HTL 和碳之间的高质量物理界面能够可靠地制造高效的 C-PSC,始终与其昂贵的蒸发 Au 对应物的性能相匹配。因此,这种物理层压技术为原子沉积电极(热真空蒸发)和分子沉积电极(直接溶液沉积)提供了合适的第三种替代方案。图 S7 提供了与叠层碳和叠层双层碳/Ag 电极相比,具有蒸发 Au 电极的最佳器件的 J-V 曲线。此外,表 S5 中还提供了图 3b 所示器件的批次数据,显示了具有不同电极配置的电池的特定 Voc、Jsc、FF 和 PCE。
The increase in Jsc for evaporated Au cells is attributed to double path light absorption made possible by the highly reflective Au electrode, which is not possible for the non-reflective carbon electrodes35,36. This is further validated in Fig. 3c, comparing the EQE response of the two cell types, wherein the cell with an evaporated Au electrode exhibits greater light absorption at longer wavelengths (650 nm–820 nm), aligning with the onset of spectral reflectance of gold and enabling double path absorption37. The improved FF of the devices featuring the laminated carbon/Ag bilayer electrode can be attributed to the higher conductivity of the coated Ag film compared to a thin evaporated Au film, as previously described, and shown in Table S2. However, this could also be due to establishing greater surface area contact between the laminated electrode and underlying Spiro-OMeTAD, formed by applying very high pressure.
蒸发 Au 电池的 Jsc 增加归因于高反射 Au 电极实现的双路径光吸收,而非反射碳电极则无法实现35,36。图 3c 进一步验证了这一点,比较了两种电池类型的 EQE 响应,其中具有蒸发 Au 电极的电池在较长波长 (650 nm–820 nm) 下表现出更大的光吸收,与金的光谱反射率一致,并实现了双路径吸收37。如前所述,具有层压碳/Ag 双层电极的器件的改进 FF 可归因于与蒸发的薄 Au 膜相比,涂层 Ag 膜的导电性更高,如表 S2 所示。然而,这也可能是由于在层压电极和底层 Spiro-OMeTAD 之间建立了更大的表面积接触,这是通过施加非常高的压力形成的。
Stability of C-PSCs compared to evaporated Au Unencapsulated PSCs with laminated carbon/Ag electrodes and evaporated Au electrodes were testedfor stability in different environmental conditions. As expected, both PSC configurations were stablefor more than 1500 h in an inert environment (storage in a N2 purge box, Fig. S8). Differences in the device stability were observed once the devices were placed in an ambient atmosphere. Fig. S8 also shows the ISOS-D-1 intermittent test results for PSCs stored in the ambient laboratory environment (dark cupboard, 25 °C, 40 – 80% relative humidity (RH)) for over 1000 h, indicating that the devices with a laminated carbon/Ag electrode were slightly more stable in the presence of ambient moisture and oxygen compared to the devices with an evaporated Au electrode38. More notably, the superior stability of the C-PSCs is showcased in Fig. 3d, giving the results of an ISOS-L-1 test where the unencapsulated devices were subjected to continuous MPP tracking (1 sun illumination, 30 °C, 11% RH). Each of the cells had an initial PCE of ~19%.As depicted in Fig. 3d, the C-PSC suffersfrom an initial“burn-in”loss within the first 24 h of testing, attributed to an initial drop in the Voc (Fig. S9). Following this initial“burn-in” period, the device exhibits great stability, retaining more than 84% of the initial PCE after 1000 h of MPP tracking. Conversely, the control cell with an evaporated Au electrode degraded to under 20% of its initial PCE after just 200 h of testing in the same conditions. Aligning with the PSC stability consensus statement, the PCE was normalised to reflect this ageing scenario, extrapolating the Tburn-in to T0 based on the stabilised degradation slope, the T80 lifetime is calculated to be 1250 h, two orders of magnitude greater than that for the evaporated Au cell (Fig. S9)38. These results highlight one of the major advantages of carbon-based electrodes, mitigating device degradation due to the ingress of moisture and oxygen to the sensitive perovskite layer while also preventing elemental migration from evaporated metal electrode.
测试了 C-PSC 与带有层压碳/Ag 电极和蒸发 Au 电极的蒸发 Au 未封装 PSC 相比在不同环境条件下的稳定性。正如预期的那样,两种 PSC 配置在惰性环境中稳定超过 1500 小时(储存在 N2 吹扫箱中,图 S8)。一旦将设备放置在环境大气中,就会观察到设备稳定性的差异。无花果。S8 还显示了在周围实验室环境(暗橱、25 °C、40 – 80% 相对湿度 (RH))中储存超过 1000 小时的 PSC 的 ISOS-D-1 间歇性测试结果,表明与具有蒸发 Au 电极的设备相比,具有层压碳/Ag 电极的设备在环境水分和氧气存在下略更稳定38。更值得注意的是,C-PSC 的卓越稳定性在 Fig. 3d 中得到了展示,给出了 ISOS-L-1 测试的结果,其中未封装的器件经受连续 MPP 跟踪(1 次太阳照射,30 °C,11% RH)。每个细胞的初始 PCE 为 ~19%。如 Fig. 3d 所示,C-PSC 在测试后的前 24 小时内出现最初的“老化”损失,这归因于 Voc 的初始下降(图 S9)。在这个初始“老化”期之后,该器件表现出极好的稳定性,在 MPP 跟踪 1000 小时后保留超过 84% 的初始 PCE。相反,在相同条件下仅测试 200 小时后,带有蒸发 Au 电极的对照电池降解至其初始 PCE 的 20% 以下。 与 PSC 稳定性共识声明一致,将 PCE 标准化以反映这种老化情况,根据稳定的降解斜率将 Tburn-in 外推到 T0,计算出 T80 寿命为 1250 小时,比蒸发的 Au 电池的寿命大两个数量级(图 S9)38。这些结果突出了碳基电极的主要优势之一,即减轻由于水分和氧气进入敏感的钙钛矿层而导致的器件退化,同时还可以防止蒸发的金属电极的元素迁移。
Large-area C-PSCs
大面积 C-PSC
A critical benefit of isostatic pressing over other lamination techniques is the ability to apply uniform pressure across large areas. To showcase this advantage, we progressed to fabricating large-area C-PSCs. The J-V curve of the champion large-area C-PSC (0.95 cm2 active area) is shown in Fig. 3e, demonstrating a reverse-scan PCE of 19.8% (Voc = 1.12 V, Jsc = 22.7 mA cm−2 , FF = 78.0), highlighting significant progress in C-PSC efficiency for a cell active area of ~1 cm2 (Fig. 3g and Table S4). This device exhibited a stabilised PCE of just under 19.5% (Fig. 3f) and was also used to operate a small fan on a sunny winter day in Melbourne as shown in Supplementary Movie 3. Progressing further, we fabricated even larger C-PSCs with a 5.5 cm2 active area at the cell level (as opposed to an interconnected module) that demonstrated a champion reverse-scan PCE of 16.9% (Voc = 1.11 V, Jsc = 22.3 mA cm−2 , FF = 68.2), and a stabilised PCE of 15.7%.
与其他层压技术相比,等静压的一个关键优势是能够在大面积上施加均匀的压力。为了展示这一优势,我们开始制造大面积 C-PSC。冠军大面积 C-PSC(0.95 cm2 有效面积)的 J-V 曲线如图 3e 所示,显示反向扫描 PCE 为 19.8%(Voc = 1.12 V,Jsc = 22.7 mA cm-2,FF = 78.0),突出了 ~1 cm2 电池有效面积的 C-PSC 效率的显着进步(图 3g 和表 S4)。该设备的稳定 PCE 略低于 19.5%(图 3f),并且还用于在墨尔本阳光明媚的冬日操作小型风扇,如补充电影 3 所示。进一步发展,我们制造了更大的 C-PSC,在电池水平上具有 5.5 cm2 的活性面积(与互连模块相反),其冠军反向扫描 PCE 为 16.9%(Voc = 1.11 V,Jsc = 22.3 mA cm-2,FF = 68.2),稳定 PCE 为 15.7%。
A vital aspect of this result is attributed not only to the high and uniform pressure applied by the CIP but also to the highly conductive coated Ag film acting as the lateral charge highway following extraction by the carbon layer. While devices with a carbon-only laminated electrode showed comparable performance to the laminated carbon/Ag bilayer electrode for research-scale cells, a significant performance loss was observed for largearea carbon-only laminated devices (Fig. S10). Notably, a substantial drop in FF was observed without the highly conductive coated Ag film. As such, the bilayer electrode design used in this work provides a critical solution to achieving high-performing largearea C-PSCs. The achievement of record
这一结果的一个重要方面不仅归功于 CIP 施加的高压和均匀压力,还归功于高导电性涂层 Ag 薄膜在被碳层萃取后充当侧向电荷高速公路。虽然对于研究规模的电池,具有纯碳层压电极的器件显示出与层压碳/Ag 双层电极相当的性能,但对于大面积的纯碳层压器件,观察到明显的性能损失(图 S10)。值得注意的是,在没有高导电涂层 Ag 薄膜的情况下,观察到 FF 的大幅下降。因此,本研究中使用的双层电极设计为实现高性能大面积 C-PSC 提供了关键解决方案。创纪录的成就
breaking efficiencies for large-area C-PSCs underscores the advantages of this approach in pushing the boundaries of C-PSC performance.
大面积 C-PSC 的断裂效率强调了这种方法在突破 C-PSC 性能界限方面的优势。
The results presented in Fig. 3 strongly demonstrate that the CIP lamination technique can establish a robust interface between the HTL and the electrode, on par with thermally evaporated Au, directly matching the photovoltaic performance whilst also exhibiting greatly improved stability in the presence of extrinsic forces such as light, moisture and oxygen.
图 3 中呈现的结果有力地表明,CIP 层压技术可以在 HTL 和电极之间建立一个坚固的界面,与热蒸发的 Au 相当,直接与光伏性能相匹配,同时在光、湿气和氧气等外在力存在下也表现出大大提高的稳定性。
Flexible, R2R and custom-designed C-PSCs
灵活的 R2R 和定制设计的 C-PSC
To further showcase the versatility of this alternative lamination technique, flexible C-PSCs were also fabricated. These devices employed a similar n-i-p structure: PET-TCE/SnO2/FA0.4MA0.6PbI3/Spiro-OMeTAD (TCE = transparent conductive electrode), however, each of the layers up to the HTL were deposited either by R2R reverse-gravure coating (SnO2) or R2R slotdie coating (perovskite and Spiro-OMeTAD). More details on the materials and device fabrication are given in the experimental procedures. Using the CIP lamination method, a roll of flexible PSCs was sealed and pressed simultaneously in the chamber, as shown in Fig. 4a, b, posing a significant advantage compared to other lamination techniques which would require one-by-one processing. The flexible, R2R-fabricated C-PSCs demonstrated a maximum PCE of 15.8% (Voc = 1.04 V, Jsc = 19.1 mA cm−2 , FF = 79.2), as shown in Fig. 4c, marking one of the highest reports for a R2R-fabricated PSC with a vacuum-free electrode.
为了进一步展示这种替代层压技术的多功能性,还制造了柔性 C-PSC。这些器件采用类似的 n-i-p 结构:PET-TCE/SnO2/FA0.4MA0.6PbI3/Spiro-OMeTAD(TCE = 透明导电电极),但是,直到 HTL 的每一层都是通过 R2R 反向凹版涂层 (SnO2) 或 R2R 狭缝模涂层(钙钛矿和 Spiro-OMeTAD)沉积的。实验程序中给出了有关材料和器件制造的更多细节。使用 CIP 层压方法,一卷柔性 PSC 在腔室中同时密封和压制,如图 4a、b 所示,与其他需要逐一加工的层压技术相比,具有显着优势。柔性的 R2R 制造的 C-PSC 表现出 15.8% 的最大 PCE(Voc = 1.04 V,Jsc = 19.1 mA cm-2,FF = 79.2),如图 4c 所示,标志着具有无真空电极的 R2R 制造的 PSC 的最高报告之一。
Finally, custom-designed electrodes were also fabricated by screen printing before being laminated onto the underlying PSC using the isostatic press. In this respect, the CIP lamination technique offers a clear advantage over other methods such as uniaxial pressing which would fail to apply uniform pressure distribution across the entire electrode geometry. An image of the custom-designed electrode laminated onto a PSC is shown in Fig. 4d, with a photocurrent map of the device provided in Fig. 4e showing the uniform interface contact between the carbon electrode and HTL.
最后,还通过丝网印刷制造了定制设计的电极,然后使用等静压机层压到下面的 PSC 上。在这方面,CIP 层压技术比其他方法(如单轴压制)具有明显的优势,因为单轴压制无法在整个电极几何形状上施加均匀的压力分布。图 4d 显示了层压到 PSC 上的定制设计电极的图像,图 4e 中提供的器件的光电流图显示了碳电极和 HTL 之间的均匀界面接触。
Conclusion
结论
In conclusion, we found that using an isostatic press is an effective technique for the vacuum-free lamination of back electrodes for PSCs. Using the isostatic press, extremely high pressure could be applied without localised stress to laminate pre-coated electrodes from a carrier film to the PSC device stack to form an effective physical and electrical contact at the interface between the electrode and HTL, on par with vacuum evaporation deposition. Moreover, the approach intrinsically eliminated the risk of degradation caused by solvent leaching and/or high-temperature processing that are commonly observed in solution-based deposition. The effectiveness of the technique was proven by demonstrating record PCEs for C-PSCs at various scales including small cells with up to 20.8% PCE, matching the performance of the same-scale control devices with evaporated Au electrodes. The extreme pressure was combined with the benefits of a coated bilayer electrode consisting of carbon for soft interface contact and silver for high conductivity. Therefore, performance loss was minimised for large-area C-PSCs, with record efficiencies of 19.8% and 16.9% for cell areas of 0.95 cm2 and 5.5 cm2 , respectively. The C-PSCs demonstrated excellent operational stability compared to the evaporated Au cells, retaining over 84% of their initial PCE after more than 1000 h of unencapsulated MPP tracking in lowhumidity air. Finally, we demonstrated the versatility of the CIP lamination technique for producing flexible and R2R printable PSCs with up to 15.8% PCE. This alternative lamination technique is readily scalable and suitable to produce both rigid and flexible C-PSCs, with the potential to simultaneously laminate printed electrodes onto thousands of cells with diverse sizes and electrode designs in minutes. These findings underscore the critical role of interface contact in enhancing C-PSC performance, and the results pave the way for developing low-cost, efficient, and reliable perovskite solar cells.
总之,我们发现使用等静压机是 PSC 背电极无真空层压的有效技术。使用等静压机,可以施加极高的压力,而不会产生局部应力,将预涂电极从载体膜层压到 PSC 器件堆栈上,从而在电极和 HTL 之间的界面处形成有效的物理和电气接触, 与真空蒸发沉积相当。此外,该方法从本质上消除了溶剂浸出和/或高温处理引起的降解风险,这些风险在基于溶液的沉积中很常见。该技术的有效性通过证明各种规模的 C-PSC 创纪录的 PCE 来证明,包括具有高达 20.8% PCE 的小细胞,与具有蒸发 Au 电极的相同规模控制设备的性能相匹配。极压与涂层双层电极的优点相结合,该电极由用于软界面接触的碳和用于高导电性的银组成。因此,大面积 C-PSC 的性能损失最小,0.95 cm2 和 5.5 cm2 的细胞面积的效率分别为 19.8% 和 16.9%,创纪录。与蒸发的 Au 电池相比,C-PSC 表现出优异的操作稳定性,在低湿度空气中经过 1000 多个小时的未封装 MPP 跟踪后,仍保留了超过 84% 的初始 PCE。最后,我们展示了 CIP 层压技术的多功能性,用于生产 PCE 高达 2% 的柔性和 R15.8R 可打印 PSC。 这种替代层压技术易于扩展,适用于生产刚性和柔性 C-PSC,有可能在几分钟内将打印的电极同时层压到数千个具有不同尺寸和电极设计的电池上。这些发现强调了界面接触在增强 C-PSC 性能中的关键作用,结果为开发低成本、高效和可靠的钙钛矿太阳能电池铺平了道路。
Methods
方法
Glass-based PSC device fabrication
玻璃基 PSC 器件制造
The ITO-coated glass substrates (15 Ω □−1 , Yingkou Shangneng Photoelectric Material Co. Ltd.), were cleaned sequentially using detergent (Deconex solution 5% v/v), deionised water, acetone (Chemsupply), and 2-propanol (Sigma-Aldrich) in an ultrasonic bath for 10 min each, after which the substrates were dried under a nitrogen flow and then ultraviolet (UV) ozone treated for 20 min. For research-scale (0.16 cm2 active area) and large area (~1 cm2 active area) devices, the glass substrates were 25 mm × 25 mm. For the larger area devices (~5 cm2 active area), the glass substrates were 50 mm × 50 mm.
使用清洁剂(Deconex 溶液 5% v/v)、去离子水、丙酮 (Chemsupply) 和 2-丙醇 (Sigma-Aldrich) 在超声波浴中依次清洁 ITO 涂层玻璃基板(15 Ω □−1,营口尚能光电材料有限公司),每个在超声波浴中清洗 10 分钟,然后在氮气流下干燥基板,然后紫外线 (UV) 臭氧处理 20 分钟。对于研究规模(0.16 cm2 有效面积)和大面积(~1 cm2 有效面积)器件,玻璃基板为 25 mm × 25 mm。对于较大面积的器件(~5 cm2 有效面积),玻璃基板为 50 mm × 50 mm。
SnO2 ETL. The SnO2 nanoparticle electron transport layer (ETL) was synthesised as described previously39. Briefly, 0.68 g of tin(IV) chloride pentahydrate (SnCl4·5H2O, 98%, Sigma-Aldrich) was dissolved in 20 mL of an EtOH/H2O mixture (50% v/v). 0.1 mL of ethylenediamine (EDA, 99%, Sigma-Aldrich) was added to the solution at room temperature
SnO2 ETL。 如前所述合成 SnO2 纳米粒子电子传递层 (ETL)39。简而言之,将 0.68 g 氯化锡 (IV) 五水合物(SnCl4·5H2O,98%,Sigma-Aldrich)溶于 20 mL EtOH/H2O 混合物 (50% v/v) 中。在室温下向溶液中加入 0.1 mL 乙二胺(EDA,99%,Sigma-Aldrich)
under stirring, after which stirring was continued for 2 h until the solution became clear. The solution was then filtered (PTFE, pore size 0.45 µm) before transferring to a microwave reactor (Monowave 400, AntonPaar) and processing at 100 °C for 30 min. The suspension was then centrifuged at 7000 rpm for 5 min to remove the supernatant. The resulting precipitate was washed with deionised water and then with ethanol, with the washing assisted by ultra-sonication for 10 min. The final precipitate of SnO2 paste was re-dispersed using 2.2 g of the wet precipitate (solid loading ~4 w/w%) in 10 mL of an aqueous solution of 0.12 M KOH.
在搅拌下,继续搅拌 2 小时,直到溶液澄清。然后过滤溶液(PTFE,孔径 0.45 μm),然后转移到微波反应器(Monowave 400,AntonPaar)并在 100 °C 下处理 30 分钟。然后将悬浮液以 7000 rpm 离心 5 分钟以去除上清液。用去离子水洗涤所得沉淀物,然后用乙醇洗涤,在超声处理辅助下洗涤 10 分钟。使用 2.2 g 湿沉淀物(固体负载量 ~4 w/w%)在 10 mL 0.12 M KOH 的水溶液中重新分散 SnO2 糊状物的最终沉淀物。
Perovskite. The FA0.88Cs0.12PbI3 perovskite solution was prepared by dissolving 1 M lead iodide (PbI2, Sigma-Aldrich) with 5% M excess, 0.88 M formamidinium iodide (CH(NH2)2I, FAI, Greatcell Solar), 0.12 M caesium iodide (CsI, Sigma-Aldrich) and 23 M% methylammonium chloride (MACl, Greatcell Solar) with respect to the PbI2 concentration in 0.5 mL of dimethylformamide (DMF, Sigma-Aldrich) and 0.096 mL of N-methyl-2-pyrrolidone (NMP, Sigma-Aldrich). The perovskite solution (80 µL) was spin coated onto the SnO2 layer at 5000 rpm for 75 s inside the glovebox. The films were then immediately annealed at 70 °C for 1 min inside the glovebox before they were taken out of the glovebox and annealed on a hotplate for 10 min at 150 °C, in the ambient lab environment (20 – 25 °C and 30–50% RH).
钙钛矿。FA0.88Cs0.12PbI3钙钛矿溶液的制备方法是将 1 M 碘化铅(PbI2,Sigma-Aldrich)与 5% M 过量的 M、0.88 M 甲酰胺碘化铵(CH(NH2)2I,FAI,Greatcell Solar)、0.12 M 碘化铯(CsI,Sigma-Aldrich)和 23 M% 甲基氯化铵(MACl,Greatcell Solar)溶解在 0.5 mL 二甲基甲酰胺(DMF,Sigma-Aldrich)和 0.096 mL N-甲基-2-吡咯烷酮(NMP, Sigma-Aldrich)。将钙钛矿溶液 (80 μL) 以 5000 rpm 的速度旋涂到手套箱内的 SnO 2 层上 75 秒。然后,立即将薄膜在手套箱内于 70 °C 退火 1 分钟,然后从手套箱中取出,并在周围实验室环境(20 – 25 °C 和 30–50% RH)中在 150 °C 的热板上退火 10 分钟。
9 mg n-hexylammonium bromide (HABr, Greatcell Solar) in 5 mL of 2-propanol was spin coated at 5000 rpm for 30 s on top of the perovskite film to form a passivation layer before drying on a hot plate at 100 °C for 5 min.
将 9 mg 正己基溴化铵(HABr,Greatcell Solar)溶于 5 mL 2-丙醇中,以 5000 rpm 的速度旋涂在钙钛矿薄膜上 30 秒,形成钝化层,然后在 100 °C 的热板上干燥 5 分钟。
Spiro-OMeTAD HTL. The hole transport layer (HTL) solution contained 73 mg of 2,2′,7,7′-tetrakis-(N,N-di-4-methoxyphenylamino)−9,9′-spirobifluorene (Spiro-OMeTAD, Luminescence Technologies Corp.) in 1 mL of chlorobenzene (CB, Sigma-Aldrich) to which 18 μL of a stock solution containing 520 mg lithium bis(trifluoromethanesulfonyl)imide (LiTFSI, Sigma-Aldrich) in 1 mL acetonitrile, 30 μL 4-tert-butylpyridine (tBP, Sigma-Aldrich) and 10 μL of tris-(2-(1H-pyrazol-1-yl)−4-tertbutylpyridine)-cobalt(III)tris(bis(trifluoromethylsulfonyl)imide)) (FK209, Luminescence Technologies Corp.) stock solution (300 mg FK209 in 1 mL acetonitrile) were added. Then, the HTL solution (70 µL) was spin-coated at 3000 rpm for 30 s onto the perovskite layer. Finally, for evaporated gold devices, an 80 nm gold electrode was evaporated using an Angstrom Engineering thermal evaporator under vacuum base pressure of <10−6 to form pixels having an active area of around 0.2 cm2 .
Spiro-OMeTAD HTL. 空穴传输层 (HTL) 溶液在 1 mL 氯苯(CB,Sigma-Aldrich)中含有 73 mg 2,2′,7,7′-四氢-(N,N-二-4-甲氧基苯氨基)−9,9′-螺双芴(Spiro-OMeTAD,发光技术公司)和 1 mL 氯苯(CB,Sigma-Aldrich)中含有 18 μL 储备液,其中含有 520 mg 双(三氟甲磺酰基)酰亚胺锂(LiTFSI,Sigma-Aldrich)和 10 μL 三(2-(1H-吡唑-1-基)−4-叔)的储备液加入丁基吡啶)-钴(III)三(双(三氟甲基磺酰基)酰亚胺))(FK209,发光技术公司)储备液(300 mg FK209,溶于1 mL乙腈中)。然后,将 HTL 溶液 (70 μL) 以 3000 rpm 的速度旋涂到钙钛矿层上 30 秒。最后,对于蒸发的金器件,使用 Angstrom Engineering 热蒸发器在 <10-6 的真空基础压力下蒸发 80 nm 金电极,形成有效面积约为 0.2 cm2 的像素。
R2R fabrication of flexible PSCs up to the HTL
R2R 制造高达 HTL 的柔性 PSC
The FA0.4MA0.6PbI3 perovskite film was fabricated using a two-step deposition process. The first solution (PbI2/FAI) was prepared by dissolving 0.8 M PbI2 (Sigma-Aldrich) and 0.4 M FAI (Greatcell Solar) in 1 mL anhydrous DMF (Sigma-Aldrich). The solution was stirred at 70 °C for more than 1 h and cooled before transferring to the slot-die head. The second solution was separately prepared by mixing 0.3 M methylammonium iodide (CH3NH3I, MAI, Greatcell Solar) in 5 mL anhydrous 2-propanol (IPA, Sigma-Aldrich).
FA0.4MA0.6PbI3 钙钛矿薄膜采用两步沉积工艺制成。通过将 0.8 M PbI 2 (Sigma-Aldrich) 和 0.4 M FAI (Greatcell Solar) 溶解在 1 mL 无水 DMF (Sigma-Aldrich) 中来制备第一种溶液 (PbI2/FAI)。将溶液在 70 °C 下搅拌 1 小时以上并冷却,然后转移到狭缝模头。通过将 0.3 M 甲基ammo碘化铵(CH3NH3I,MAI,Greatcell Solar)与 5 mL 无水 2-丙醇(IPA,Sigma-Aldrich)混合,分别制备第二种溶液。
The fabrication of flexible PSCs was undertaken in the ambient laboratory environment (25 °C, 30–50% RH) using a 25 mm wide transparent conductive electrode (TCE) polyethylene terephthalate (PET) substrate (125 µm thickness) having an 8 Ω □−1 sheet resistance from Solutia (OPV8). The SnO2 was coated using the reverse-gravure coating method (16 rpm roller speed, 0.3 m min−1 web speed, 11.5 mm coating width), followed by passing over an in-line hot plate at 135 °C for about 5 s. The PET/TCE/SnO2 film then underwent R2R infra-red treatment for 8 min. The first perovskite precursor solution (PbI2+ FAI) was then slot-die coated onto the film (20 µL min−1 flow rate, 0.3 m min−1 web speed, 11.5 mm coating width). A N2 gas-knife blower was installed behind the solution head and nitrogen was blown onto the continuously moving wet film. The second perovskite precursor solution (MAI) was then similarly deposited via slot-die coating (65 µL min−1 flow rate, 0.3 m min−1 web speed, 11.5 mm coating width). Using this two-step deposition process, the intermediate precursor film (PbI2 + FAI) undergoes an almost instant transformation to the photoactive perovskite once coated with the MAI solution in IPA. The film was immediately passed over a hot plate at 135 °C for about 5 s. The PET/TCE/SnO2/Perovskite film was then rewound and the Spiro-OMeTAD solution was deposited via slot-die coating (20 µL min−1 flow rate, 0.3 m min−1 line speed, 7 mm coating width). Typically, these PSC stacks were made at lengths of 3 – 8 m. The rolls were then cut prior to attachment of the carbon/Ag bilayer electrode to lengths of around 200 mm, although lengths of over 1 m were also trialled by attaching multiple strips of the coated bilayer electrode.
柔性 PSC 的制造是在周围实验室环境(25 °C,30-50% RH)中使用 25 mm 宽的透明导电电极 (TCE) 聚对苯二甲酸乙二醇酯 (PET) 基材(125 μm 厚),具有 8 Ω □−1 薄层电阻的首诺 (OPV8)。使用反向凹版涂布方法(16 rpm 辊速,0.3 m min-1 卷筒纸速度,11.5 mm 涂布宽度)涂布 SnO2,然后在 135 °C 下通过在线热板约 5 s。PET/TCE/SnO2 薄膜随后进行 R2R 红外处理 8 min。然后将第一种钙钛矿前驱体溶液 (PbI2+ FAI) 狭缝模头涂布到薄膜上(20 μL min-1 流速,0.3 m min-1 卷筒纸速度,11.5 mm 涂层宽度)。在溶液头后面安装一个 N2 气刀鼓风机,将氮气吹到连续移动的湿膜上。然后,第二种钙钛矿前驱体溶液 (MAI) 以类似方式通过狭缝模头涂层(65 μL min-1 流速,0.3 m min-1 卷筒纸速度,11.5 mm 涂层宽度)沉积。使用这种两步沉积工艺,一旦在 IPA 中涂覆 MAI 溶液,中间前驱体薄膜 (PbI2 + FAI) 几乎立即转变为光活性钙钛矿。薄膜立即在 135 °C 的热板上传递约 5 秒。然后重卷 PET/TCE/SnO2/钙钛矿薄膜,并通过狭缝模头涂层(20 μL min-1 流速,0.3 m min-1 线速度,7 mm 涂层宽度)沉积 Spiro-OMeTAD 溶液。通常,这些 PSC 膜组的长度为 3 – 8 m。然后在将碳/银双层电极连接到大约 200 mm 的长度之前切割卷材,尽管也通过连接多个涂层双层电极条来试验超过 1 m 的长度。
Coated carbon/Ag bilayer electrode fabrication
涂层碳/Ag 双层电极制造
The coated electrodes were fabricated with a commercial carbon paste from Dycotec Materials (DM-CAP-4701S) and a conductive Ag paste from DuPont (PV416 conductor paste). The carbon paste used in this work is composed of a blended carbon mixture (<50% wt.), blended adhesion resin (<20% wt.) and an aromatic solvent systemfor the remaining weight. Due to proprietary rights, the exact composition of this paste is unknown. The pastes were sequentially manually bar-coated onto the non-stick side of a silica-coated PET substrate (class 55, The Griff Network) of 60 µm thickness and 120 mm width using a glass rod. The coated films were typically around 90 mm wide and 200 mm long. The coated films were then sequentially dried on a hot plate at 135 °C for 2 min and 5 min for the Ag and carbon paste, respectively. The dry-film thickness of the Ag was ~20 µm and the carbon dry-film thickness was ~50 µm. The electrode film was cut to size depending on the relevant device area. For research scale devices (0.16 cm2 active area), the electrode film was cut to a size of ~4 mm × 25 mm. For the large area devices (~1 cm2 active area) the electrode film was cut to a size of
涂层电极是用 Dycotec Materials 的商用碳浆 (DM-CAP-4701S) 和杜邦的导电 Ag 浆料(PV416 导电浆料)制造的。本工作中使用的碳浆由混合碳混合物 (<50% wt.)、混合粘合树脂 (<20% wt.) 和剩余重量的芳香族溶剂系统组成。由于所有权,这种浆料的确切成分尚不清楚。使用玻璃棒将浆料依次手动棒涂到厚度为 60 μm 且宽度为 120 mm 的二氧化硅涂层 PET 基材(第 55 类,Griff Network)的不粘面上。涂层薄膜通常约为 90 毫米宽和 200 毫米长。然后将涂布的薄膜依次在 135 °C 的热板上干燥 2 分钟和 5 分钟,用于 Ag 和碳浆。Ag 的干膜厚度为 ~20 μm,碳干膜厚度为 ~50 μm。根据相关设备区域将电极膜切割成合适的尺寸。对于研究规模的设备(0.16 cm2 有效面积),将电极膜切割成 ~4 mm × 25 mm 的尺寸。对于大面积器件(~1 cm2 有效面积),将电极膜切割成
~6 mm × 17 mm. For the larger area devices (~5 cm2 active area) the electrode film was cut to a size of ~10 mm × 50 mm. For the flexible R2Rfabricated devices, the electrode film was cut to a size of ~4 mm wide and lengths of up to 200 mm.
~6 毫米× 17 毫米。对于较大面积的器件(~5 cm2 有效面积),将电极膜切割成 ~10 mm × 50 mm 的尺寸。对于柔性 R2R制造的器件,电极膜被切割成 ~4 mm 宽的尺寸,最长可达 200 mm。
CIP electrode lamination
CIP 电极层压
For both the rigid and flexible PSCs, the coated electrode was placed lightly on top of the Spiro-OMeTAD HTL and then the devices were vacuum-sealed in polymer bags. The devices were submerged into the CIP chamber (Quintus Technologies LLC., LCIP42260) and pressurised to the set pressure and held at the set pressure for 30 s. For a 240 MPa sample, the total time to pressurise, hold, and release was 3 min. The flexibility of the bilayer electrode configuration on the release layer aids the C-PSC fabrication process and mitigates the risk of localised mechanical stress as the electrode can conform to the shape of the underlying PSC precursor stack. All PSCs were stored in a dry box (<10% RH) for 72 h before characterisation.
对于刚性和柔性 PSC,将涂层电极轻轻放在 Spiro-OMeTAD HTL 的顶部,然后将器件真空密封在聚合物袋中。将设备浸入 CIP 室 (Quintus Technologies LLC., LCIP42260) 中,加压至设定压力并在设定压力下保持 30 秒。对于 240 MPa 样品,加压、保持和释放的总时间为 3 min。离型层上双层电极配置的灵活性有助于 C-PSC 制造过程,并降低局部机械应力的风险,因为电极可以符合底层 PSC 前驱体堆栈的形状。表征前,所有 PSCs 在干燥箱 (<10% RH) 中储存 72 h。
Device characterisation
设备特性
Photovoltaic performance measurements were undertaken in an ambient laboratory atmosphere (25 °C, 30-70% RH) using a class AAA solar simulator (Oriel, Xenon-lamp light source), with a 16 channel Bio-Logic VMP3 potentiostat as defined in the work by Surmiak et al. 40. The solar simulator was calibrated to 1-sun (1000 W m−2 ) AM 1.5 G illumination using a certified reference cell (Enlitech with KG-2 filter, certified by Enlitech in accordance with IEC 60904-1:2006, last calibration in August 2021). J-V measurements were carried out in the forward and reverse scan directions over the voltage range −0.2 V to 1.2 V at a 20 mV s−1 scan rate and a 50 ms settling time after 10 s of light soaking. The spectral mismatch factor was calculated between reference cell and testing cells to be 0.983. For researchscale cells, a shadow mask was used to define a cell active area of 0.16 cm2 . For large-area cells, the cell area was determined by scanning the cell (with a reference ruler) with an office scanner and mapping the total electrode area using ImageJ software. For the R2R C-PSCs, a shadow mask was used to define an active area of 0.076 cm2 .
光伏性能测量是在实验室环境(25 °C,30-70% RH)中使用 AAA 级太阳模拟器(Oriel,氙灯光源)和 16 通道 Bio-Logic VMP3 恒电位仪进行的,如 Surmiak 等人 40 的工作中所定义。使用经过认证的参考电池(带有 KG-2 滤光片的 Enlitech,由 Enlitech 根据 IEC 60904-1:2006 认证,最后一次校准于 2021 年 8 月),将太阳模拟器校准为 1 太阳 (1000 W m−2 ) AM 1.5 G 照明。在 −0.2 V 至 1.2 V 电压范围内以 20 mV s−1 的扫描速率和 50 ms 的建立时间在光浸 10 s 后在 50 ms 的建立时间内在正向和反向扫描方向上进行 J-V 测量。计算出参比池和测试池之间的光谱失配因子为 0.983。对于研究规模的细胞,使用阴影掩模来定义 0.16 cm2 的细胞有效面积。对于大面积细胞,通过使用办公室扫描仪扫描细胞(用参考尺)并使用 ImageJ 软件绘制总电极面积来确定细胞面积。对于 R2R C-PSC,使用阴影掩模来定义 0.076 cm2 的活性区域。
The long-termMPP tracking stability testwas performed under a white LED light source calibrated to 1-sun illumination intensity with the reference cell mentioned previously. The unencapsulated cells were held under MPP tracking conditions, whilst also measuring the Voc, Jsc and a reversescan J-V curve every 10 min. The temperature of the cells was maintained at roughly 30 °C and 11% RH by a continuous low N2 flow across the samples.
长期 MPP 跟踪稳定性测试是在校准为 1 太阳照明强度的白色 LED 光源下进行的,并使用前面提到的参考池。未封装的细胞保持在 MPP 跟踪条件下,同时每 10 分钟测量一次 Voc、Jsc 和反向扫描 J-V 曲线。通过样品中持续的低 N2 流,将细胞的温度保持在大约 30 °C 和 11% RH。
External quantum efficiency (EQE) measurements were performed using an incident photon-to-current conversion efficiency (IPCE) measurement apparatus from Peccell Technologies, Inc (PEC-S20) and calibrated with a silicon reference photodiode.
使用 Peccell Technologies, Inc (PEC-S20) 的入射光子到电流转换效率 (IPCE) 测量设备进行外部量子效率 (EQE) 测量,并使用硅参考光电二极管进行校准。
Impedance spectroscopy (IS) was carried out from 1 MHz to 0.1 Hz in dark conditions under 0 V DC voltage bias and 10 mV applied AC voltage using an electrochemical workstation (RST5200, Zhengzhou Shiruisi Instrument Co., Ltd.).
使用电化学工作站(RST5200,郑州世瑞思仪器有限公司)在黑暗条件下,在 0 V 直流电压偏置和 10 mV 外加交流电压下,从 1 MHz 到 0.1 Hz 进行阻抗谱 (IS)。
Sheet resistance measurements were undertaken using a 4-point probe (Jandel RM3000) with an applied current of 99.999 mA for the highly conductive Ag film and the C/Ag bilayer film and 10 mA for the coated C-alone film.
使用 4 点探针 (Jandel RM3000) 进行薄层电阻测量,高导电性 Ag 薄膜和 C/Ag 双层薄膜的施加电流为 99.999 mA,涂层 C 单独薄膜的施加电流为 10 mA。
SEM images were taken with a Zeiss Merlin field-emission SEM (FESEM) operated at an accelerating voltage of 3 kV. Images were acquired using an In-lens detector and a working distance of 7 mm. Prior to imaging, a 5 nm layer of Ir was sputter-coated onto the sample.
SEM 图像是使用在 3 kV 加速电压下运行的 Zeiss Merlin 场发射 SEM (FESEM) 拍摄的。使用镜头内检测器和 7 mm 的工作距离采集图像。在成像之前,将 5 nm 的 Ir 层溅射镀膜到样品上。