SOLUTION PROCESSED HIGH PERFORMANCE TRANSPARENT ORGANIC PHOTOVOLTAIC CELLS

Informacje na temat projektu

SOLPROCEL

Identyfikator umowy o grant: 604506

Projekt został zamknięty

Data rozpoczęcia 1 Listopada 2013

Data zakończenia 31 Października 2016

Finansowanie w ramach

• Specific Programme "Cooperation": Nanosciences, Nanotechnologies, Materials and new Production Technologies

Koszt całkowity € 3 771 798,00

Wkład UE € 2 860 434,00

Koordynowany przez

FUNDACIO INSTITUT DE CIENCIES FOTONIQUES
Spain

Final Report Summary - SOLPROCEL (SOLUTION PROCESSED HIGH PERFORMANCE TRANSPARENT ORGANIC PHOTOVOLTAIC CELLS)

Executive Summary:

The work towards the main goal of SOLPROCEL which was to reach a high performance semi-transparent photovoltaic technology that could be fully solution processed was implemented as planned during the 36 month duration of the project. Within WP2 we considered the development of polymers with a wide band absorption suitable for an up-scale industrial production. This worked was performed in a close collaboration between FhG-IAP and SP. In the last year of SOLPROCEL, SP considered the industrial fabrication of donor polymers. Production and cost of such polymers were estimated by SP. Within WP2 FhG-IAP and SP developed cross-linkers to stabilize the nano-morphology of the active layer blend, which resulted in the opening of a completely new product line for SP. Within WP2 ICFO and FhG-IAP made progress towards increasing the power conversion efficiency, and by implementing a new optical cavity configuration SOLPROCEL demonstrated power conversion efficiencies above 11%. Within WP3 SOLPROCEL implemented a fully solution processed semi-transparent solar cell using the ZnO and WO3 nanoparticles provided by Nanograde, the Ag nanowires from ras, and high performance blends such as PC71BM:PCE-10. By the end of SOLPROCEL we were able to fabricate fully solution processed cells with a power conversion efficiency above 4% and a transparency above 15%. For large scale production of such nanoparticles Nanograde built a pilot plant reactor with the capability to provide approximately 10 times larger quantities than at the start of the project. As for the Ag nanowires, ras developed a product showing comparable or even superior performance than vacuum deposited transparent electrodes (e.g. ITO) that can be produced in amounts sufficient to serve the growing OPV market. Within the same WP3 FAU developed a photonic crystal technology based on the Nanograde nanoparticle layers readily applicable to large scale devices. Within WP4 progress made by ICFO continued towards the fabrication of devices with a stable performance under normal operation conditions. Using high performance polymer blends with initial efficiencies higher than 9%, SOLPROCEL reached under 1 sun AM1.5G illumination (ISOS-L-1) lifetimes (T80) larger than 14000 h. This corresponds to a 7 year lifetime for outdoor testing under optimal orientation and 10 year lifetime for vertical applications.. We also showed that the photonic crystals were very stable under high temperature (85°C) cycles. Within WP5 SOLPROCEL fabricated 10x10 cm2 modules with a 4.5% efficiency and a transparency larger than 15%. Such modules were benchmarked by COMSA against similar technologies from two different OPV companies. In WP6 COMSA carried out lamination by testing different lamination materials, procedures, as well as different cell architectures. In WP7 several activities for the dissemination of SOLPROCEL results were carried out. A very successful workshop on organic photovoltaics was organized in Barcelona by COMSA.

Project Context and Objectives:
Summary description of project context and objectives

To achieve SOLPROCEL’s goal of obtaining a high performance fully solution processed semi-transparent photovoltaic technology, we divided the work in a total of seven workpackages. Research and development activities were grouped in WPs from 2 to 5, while in WP6 we conducted a benchmarking against other existing semi-transparent PV solutions and in WP7 we considered the optimal path for facilitating an effective industrialization and commercialization. The main achievements against the initial objectives set within each one of the WPs are summarized below.

WP2
The main goal in here was to increasing the performance of currently available OPV cells by reaching higher efficiencies and higher transparencies. This required the combination of the synthesis of new materials, device architecture optical design, and the implementation of different device fabrication procedures. At Fraunhofer IAP we developed the synthesis of innovative absorber terpolymers in a small lab scale. A singularity of such terpolymers is the combination of three different monomers units (donor, acceptor and donor-acceptor) which might lead to tune the absorption out to the IR-range up to 900 nm. These polymers were synthesized by Stille coupling reaction. The IR absorption of this new absorber terpolymer was proved to be displaced in the IR range and may lead to an efficiency of 9-10%. Investigations in normal and inverted OPV cells gave average efficiencies of around 5.5%. The synthesis and up-scaled synthesis of diboronic dithophene as donor monomer and dibromofluoro-benzthiadiazole as acceptor was achieved properly by SPECIFIC POLYMERS and batches of about 7g were obtained in both cases. The products developed in the project here and most especially the donor and acceptor monomers are now available in SPECIFIC POLYMERS catalog. Within WP2, another important aspect is the long term stability of the active cell blend. Photo- or thermo-crosslinkable agents based on organic bisazides were developed by SPECIFIC POLYMERS in order to stabilize the blend, create a polymer network and thus avoid the macrophase segregation of the blend component after heating. In the case of P3HT/PCBM and the blend based on the new absorber terpolymer and PCBM it was demonstrated that the crosslinking lead to thermal stabilization of the morphology. Surprisingly, unlike the results on P3HT:PCBM and PCE-10:PCBM layers, for the new terpolymer blend we did not observe any micro-scale crystallization of the PCBM component even after 22 hours annealing at 150°. At ICFO we worked towards increasing the power conversion efficiency by implementing a new optical cavity configuration that we used to demonstrate power conversion efficiencies above 11% using high performance commercial blends. This latter approach is fully compatible to optimize light harvesting in the new materials developed by Fraunhofer IAP and SP.

WP3
The aim of WP3 was to develop all the non-active material parts of an organic photovoltaic device based on solution processable formulations. The materials developed in this WP can be split into three classes, which were developed separately: Conductive electrode materials, based on silver nanowires, semiconducting selective charge transport layers, based on inorganic metal oxide nanoparticles and optical layers based on organic-inorganic hybrid systems.
The silver nanowires were developed by RAS based on a liquid synthesis process. Starting off with simple recipe to produce Ag-NW of rather low quality, in the course of the project, RAS improved the production process, developed further purifications, scaled up the production and implemented a state-of-the art quality assurance system. The final developed product now shows comparable or even superior performance than vacuum deposited transparent electrodes (e.g. ITO) and can be produced in amounts sufficient to serve the growing OPV market.

Printable inorganic buffer layers were imagined to replace current vacuum deposited EBL and HBL materials such as evaporated Ca or MoO3. By using its patented flame spray pyrolysis particle synthesis method and the broad know-how of suspension formulation, Nanograde developed several EBL (WO3 MoO3) and HBL (ZnO, AZO, IGZO) formulations that showed very high performance in printed solar cells. Due to an unfortunate incompatibility with the finally applied absorber system, Nanograde’s EBL inks did not deliver the desired performance in the end and were replaced with, still printable, PEDOT:PSS formulations.

Furthermore, based on the same particle synthesis method, Nanograde developed printable layers with high transparency and very low or very high refractive index. These materials are based on polymer-nanoparticle hybrid systems. In parallel SPECIFIC POLYMERS developed fully organic low-RI monomers and cross-linked thin layer materials that could in principle be used in the same way. Due to slightly superior performance Nanograde’s materials were preferred in the final product. By combining several layers of high- and low-RI material, photonic crystals with a performance comparable to vacuum deposited systems could be printed.

Finally, the materials developed in WP3 were combined to an all-solution-processed semitransparent organic solar cell. The cells the FAU was able to produce with the materials provided by RAS, Nanograde and SP showed very promising results, even if the originally set goal of a PCE>9% at >30% transparency could not be fully reached. It was however clearly shown, that incorporating a photonic crystal in to the semi-transparent PV system was beneficial for the performance at high transparencies.

WP4
The main goal was to identify failure mechanisms in semi-transparent OPVs, and propose cell configurations that would lead to stable devices. During the first year of SOLPROCEL we established that materials used to fabricate the devices were highly stable under “no-stress” conditions. It was also shown that this changed dramatically when devices were exposed to an aggressive environment including oxygen, humidity, temperature changes or intense light. We concluded that the only viable option to limit the effect of such agents was to encapsulate the devices. We showed that a proper encapsulation constitutes en effective barrier to oxygen and moisture. However, for obvious reasons, such encapsulation could not be used as an effective barrier to high temperature/temperature oscillations or to block light. During the second year we approached such issues separately. On the one hand, we introduced the cross-linker developed in WP2 into the PCE-10:PC71BM blend and tested its effect under several conditions. On the other hand, we established several kinds of optical filtering and implemented a pre-treatment procedure to the polymer blend to prevent degradation of the device under the solar simulator illumination. During the last year we defined the final configuration to reach long cell lifetimes. Degradations studies on semi-transparent cells indicated that when opaque cells are converted into semi-transparent ones no new degradation mechanisms appear. For encapsulated cells, in both cases, it was shown that the major source of degradation is the illumination by the UV light. However, we demonstrated that the detrimental effect of UV light can be eliminated by including the proper UV filtering in addition to a new procedure which we implemented to stabilize the blend nano-morphology. For the opaque devices we demonstrated lifetimes beyond 7 years assuming a direct orientation of the cell in relation to the sun position, and above 10 years for a vertical orientation of the cell corresponding to the standard orientation for a PV window when integrated in a building. Regarding the photonic crystals to be incorporated on the semi-transparent cells, we saw an excellent stability under different degradation conditions.

WP5
The objective of WP5 was to use materials and processes developed in WP2, WP3 and WP4 to fabricate large-area, fully printed semitransparent solar modules and photonic crystals on a size of ~ 10cm x 10cm and finally combine them into a single device. For module fabrication, we chose to structure all the layers of the printed module with a femtosecond laser after coating, to create a series connection between the single cells. This approach has the advantage of a high geometric fill factor and a rather homogeneous appearance due to the small width of the laser lines. With respect to photonic crystals, Nanograde fabricated upscale batches with a volume of a few liters, FAU evaluated these batches and used them to produce photonic crystals with the doctor blading method, on a substrate size of 10cm x 10cm on both glass and flexible substrates. These PCs were characterized with respect to the homogeneity of their optical properties. Concerning the combination of solar modules and photonic crystals, we investigated the suitability of different lamination materials, we studied influence of the incidence angle of the illuminating light, which is of high importance for the successful integration of the up-scaled solar module in BIPV applications such as windows. Since the up-scaling facilities at FAU only allowed ambient conditions, we chose absorber polymers which showed good compatibility with solution-processing in air, and combined them with the buffer layers provided by Nanograde and the nanowires (NWs) produced by ras. As a final result, we fabricated semitransparent modules with an efficiency of 4.5% and a transparency of 16%, which exhibited negligible losses compared to the single-cell reference. All solution-processed layers were either doctor-bladed or slot-die coated, which means that they are fully compatible with roll-to-roll production for future upscaling. The fabricated photonic crystals exhibited good optical quality and homogeneity and also good agreement with the simulated curves, indicating smooth interfaces and a performance similar to evaporated mirrors. The combination of semitransparent module and photonic crystal resulted in a current increase of 19% for the thinner, more transparent solar cell and a current increase of 6.7% for the thicker, less transparent solar cell. In addition, the study of the incidence angle revealed that the increase due to the photonic crystal is approximately constant for incidence angles up to 60°, indicating the excellent suitability of the photonic crystal approach for window applications.

WP6
The main objective of WP6 was the integration of SOLPROCEL modules produced in WP5 into building-façade elements and the benchmarking of 3 OPV BIPV technologies to evaluate their transparency, performance and stability in building integrated environment conditions. The first task of WP6 focused on the definition and validation of materials and processes to perform the optimal lamination of SOLPROCEL modules. Several lamination tests were carried out with different temperatures to determine the best lamination temperature profile to achieve the better homogeneity and melting of the lamination material, and the right adherence and mechanical properties to be installed as architectural elements in building façades. As the first tests showed severe degradation of the cells after lamination, a second round of tests was conducted in which the cells were protected with a pre-encapsulating glass fixated with DELO-KATIOBOND prior to lamination to prevent them from being damaged. The validated lamination material used was EVA with a temperature profile of 85ºC in 60 minutes. For the final set-up of OPV modules integrated in building-façade elements, the requirements for BIPV applications were taken into account. These include design strategies (module size and shape, glass characteristics, color, uniformity, frames, junction box, wiring and connectors), their functionality (stability, sun protection, insulation) and the safety conditions (materials cohesion, electrical safety). In particular, for the SOLPROCEL semi-transparent OPV modules, the fabrication of the up-scaled modules, as well as the following procedures for their integration into building elements were defined while considering both the initial (105mm x 105mm modules) and the future approach (fully printed 1m x 1m modules). Finally, SOLPROCEL modules were benchmarked alongside two commercial OPV technologies. The results were satisfactory in terms of transparency, as the SOLPROCEL modules showed superior results than the others, and acceptable in terms of efficiency, even though new lines of research should be conducted within the materials and manufacturing process to ensure a higher stability which is critical for BIPV applications.
WP7
The objectives of WP7 were the dissemination of the SOLPROCEL project development and results and the design of a plan for an optimal exploitation of the technology developed in SOLPROCEL, considering the manufacturing process, the aggregated cost analysis and the Life Cycle Analysis of SOLPROCEL modules, as well as the standardization procedure of OPV modules for BIPV applications.
The dissemination of the SOLPROCEL project was conducted in a series of conferences, newsletters and publications, as well as to relevant platforms and associations. On October 7th 2016, COMSA CORP organized the OPV Workshop: A new technology to market, which had a scientific approach and also a market-oriented approach in relation to BIPV. The SOLPROCEL project was presented during the event as well as MUJULIMA and ARTESUN projects which also devloped OPV technologies. The publication of the results of SOLPROCEL project was conducted through multiple articles in scientific publications, invited talks in conference at worldwide level, poster presentations and exhibitions, both national and international. The results of the benchmarking of the SOLPROCEL semi-transparent OPV modules were exposed during University lectures as well as in events and publications related to the sector. Regarding the exploitation of the modules, a large-scale manufacturing process was designed aiming at a future production scenario of 1 million m2 of 1m x 1m modules with 5% efficiency and a 90% active area. The overall manufacturing cost of the OPV modules in the aforementioned scenario would add up to 0.44€/Wp (0.45€/Wp including estimated costs for potentially up-scaled WP2 materials). According to the expected impact, by 2030 the SOLPROCEL modules should have an efficiency of 15%, which would decrease their cost to 0.16 €/Wp, lower than the expected cost of c-silicon (0.25 €/Wp). The Life Cycle Analysis performed on the SOLPROCEL semi-transparent OPV modules showed very satisfactory Energy Payback Time and GHG emission values in comparison to the conventional silicon technology. Finally, the definition of the Roadmap for the Standardization procedure for OPV modules identified two Standardization Technical Committees at national and international levels, and a new work item proposal started in the Technical Working Group of “Wafers, cells and modules” (CLC/TC82/WG1) of the “Solar PV energy systems” (AEN/CTN206/82) Standardization Technical Committee.

Project Results:
Main S&T results

WP2: High performance active materials and configurations
Task 2.1: Synthesis of the optimal active layer (M1-M24, FhG-IAP)
One of the main objectives in WP2 was to produce high performance photovoltaic organic polymers. To do so, within Task 2.1 Fraunhofer IAP developed the synthesis of innovative absorber terpolymers in a small lab scale in the lab while SPECIFIC POLYMERS was in charge of the design of an up-scaled production procedure for the most promising polymer candidates Singularity of such terpolymers (compared to copolymers commonly used) is the combination of three different monomers units (donor, acceptor and donor-acceptor) which might lead to tune the absorption out to the IR-range up to 900 nm. These polymers were synthesized by Stille coupling reaction using donor, acceptor and donor-acceptor monomers. Syntheses of all three monomers were successfully achieved as well as the synthesis of the corresponding terpolymers. Two different ways for the polymerization were tried: the polymerisation in normal flask (KE29/3) and the microwave assistant polymerization (KE29/2). The introduction of the additional donor acceptor unit dithieno pyrazine with two phenoxyoctyl side chains into the polymer backbone should lead to polymers with a better solubility like it was described in our patent application [2]. The introduction of an oxygen atom into the cyclopentane ring leads to the dithienopyrane and shift the absorption later on in the polymer more in IR-range [1]. We optimized further the conditions of the conventional polymerization process. The change of the solvent from toluene to tretraline led to polymers with higher molecular weight (KE29/3-2). The IR absorption of this new absorber terpolymer was proved to be displaced in the IR range and might lead to an efficiency of 9-10%. The synthetic route was further optimized concerning the applied solvent and the catalyzer systems for the polymerization process, to reach polymers with high molecular weights and small polydispersity. Investigations in normal and inverted OPV cells gave average efficiencies of around 5.5%.

Figure 2.1. Terpolymer 29/2
Task 2.2: Production of new absorber polymers (M13-M36, SP)

Nevertheless, the studies performed by Fraunhofer IAP to obtain this absorber terpolymer revealed that a part of the synthetic routes used are much too difficult to consider for an industrial upscale production. As a consequence, SOLPROCEL consortium decided on January 2015 that SPECIFIC POLYMERS, in close collaboration with Fraunhofer IAP, would work on an alternative synthesis route to produce the absorber polymer. That decision was based on three main reasons: monomers synthesis difficulties (number of steps, not industrial scale methods), toxicity of the synthetic routes (Stille coupling reaction, organostannic compounds) and production cost. These reasons were described more precisely in M12 annual report. All these considerations pushed FRAUNHOFER and SPECIFIC POLYMERS to find an alternative synthesis route to produce the absorber polymer. Alternative absorber polymer targeted for the up-scale production study was a copolymer based on 2 monomers. The donor monomer targeted was a diboronic dithophene monomer (A’) and acceptor monomer was the dibromofluorobenzthiadiazole monomer (C’). Finally, the synthesis of the absorber copolymer would be achieved by Suzuki coupling polymerization between the monomers (A’) and (C’).The synthesis and up-scaled synthesis of both monomers were achieved properly by SPECIFIC POLYMERS and batches of about 7g were obtained in both cases. A special attention was given to the purification of these monomers since it appeared that a purity level of 99.9% was mandatory to obtain the desired high molecular weight absorber polymer. The purity of monomers (C’) and (A’) obtained by SPECIFIC POLYMERS were 99.9% and 99.2% respectively. The highest purity of monomer (C’) could have been confirmed by LCMS analysis whereas the one of monomer (A’) was only determined by 1H NMR titration. Both monomers were used to study the Suzuki polymerization reaction of this monomer couple. The optimization of the reaction condition and the use of the monomers exhibiting the highest degree of purity allowed obtaining an absorber polymer characterized by a mean molecular weight of 4250 g/mol. Such molecular weight was significantly lower than those obtained by Fraunhofer. In addition, the obtained product also contained low molecular weight oligomer chains and residual unreacted monomers and no treatments were found to recover the pure polymers. More than a treatment issue, the relatively low molecular weight and the presence of impurity indicated that a further purification step had to be determined and implemented to reach a purity of 99.9% and thus higher molecular weight. Consequently, SPECIFIC POLYMERS was unable to produce a 10g production of the absorber polymer within the scope of SOLPROCEL Project. Nevertheless, the work performed during SOLPROCEL project regarding the synthesis and the up-scaled study of donor/acceptor monomers and absorber polymers allowed SPECIFIC POLYMERS to acquire new skills and competences. The synthesis involved in SOLPROCEL project implied totally new chemistries for the company. This part of the project was very challenging but, within 3 years, the SOLPROCEL project allowed SPECIFIC POLYMERS to acquire new synthesis methods (Suzuki polymerization) and new purifications methods (Flash chromatography). Most of the monomers and polymers usually produced by the company can be prepared at 50-100g scale but only with a purity level of 95-98%. In this project, it was figured out that monomers purity of 99.9% was required which pushed SPECIFIC POLYMERS researchers to implement new purification methods and an intensive work was done during the project to acquire skills in the column chromatography technique. Such technique is now used in the company for another project and is offering new opportunities in application sectors like optoelectronics but also biomedicine. The products developed in the project here and most especially the donor and acceptor monomers are now available in SPECIFIC POLYMERS catalog.
Task 2.3: Cross-linking reaction for the stabilization of the blend systems (M7-M30, FhG-IAP, SP)
Within WP2, another important aspect considered in Task 2.3 was the long term stability of the active cell blend. A Photo- or thermo-crosslinkable agent was developed by SPECIFIC POLYMERS in order to stabilize the blend, create a polymer network and thus avoid the macrophase segregation of the blend component after heating. Regarding this singular point, SPECIFIC POLYMERS and Fraunhofer-IAP decided to focus on bi-functional crosslinking agent bearing azide. SPECIFIC POLYMERS (SP) developed the synthesis of various diazido crosslinking agents in order to improve the long term stability of the active polymer layer blend. All crosslinking agents are bi-functional azide compounds that are able to react with double bonds or labile hydrogen present in the polymer blend. SP synthesized 10 different diazide structures. The chemical structures of crosslinking agents synthesized by SPECIFIC POLYMERS are presented in Table 2.1. All different crosslinking agents characterized by different structures and various molecular weights were obtained and their structures were confirmed by 1H NMR and IFTR analysis. The Fraunhofer IAP studied the crosslinking reactions between the diazides and different active blend systems. In the case of P3HT:PCBM and the blend based on our new developed absorber terpolymer KE40 and PCBM, it could be evidenced that the crosslinking lead to thermal stabilization of the morphology. But upon long-term (20 h) heating, the device efficiency dropped by ca. 2/3 for the KE40-PC71BM blend layers. For PCE-10:PC71BM layers however, the solar cell characteristics appeared to be adversely affected by the presence of the crosslinker even before crosslinking. This might be due to the fragile morphology of the PCE-10:PC71BM BHJ. Upon further heating the power conversion efficiency dropped to around 1% in both crosslinked and non-crosslinked samples.
Table2. 1. Crosslinking agents synthesized by SP.

In order to complete the stability data on KE29:PC71BM solar cells, the thermal ageing studies of devices were extended with non-cross-linked active layer. Surprisingly, and unlike the results on P3HT:PCBM and PCE-10:PCBM layers, a micro-scale crystallization of the PCBM component was not observed even after a 22 hours annealing at 150°. The efficiency of KE29-based devices was maintained after the heating process. Accordingly, no phase segregation was observed for these layers. Owing to the thermal stability, the retained device efficiency of KE29:PC71BM devices was higher than the efficiency of PCE-10:PC71BM devices for heating durations longer than 22 h at 150°C (Fig.2.2). The polymeric crosslinker SP-SOLPROCEL-115, a poly(styrene-co-styrene azide), was used to prepare solvent stable layers with high PC61BM or PC71BM content as electron transport layer in an inverted OPV ell architecture. Operating devices with PCE-10: PCBM as active blend were reliably produced and had a moderate efficiency of ca. 5.6%. Note that the PEI layer was essential in these devices for proper device characteristics.

Figure 2.2 Power conversion efficiency (PCE) and microscopy images of PCE-10:PC71BM and KE29:PC71BM OPV cells after varied durations of heating of the active layer at 150°C.
Task 2.4: Fabrication of opaque single junction and tandem cells (M7-M36, ICFO, FhG-IAP)
Figure 2.3. a) Scheme of Two-resonance tapping in an optical cavity, b) Electric field intensity depending on MCL thickness.
Within task 2.4 we implemented an optical optimization of the single junction devices based on the effect described as a two-resonance tapping in an optical cavity. Such cavity is based on a combination of a metallic and dielectric cavity layer, MCL and DCL, respectively, using an active layer (AL) based on PCE10:PC71BM blend, as shown in Figure 2.3. a. The high reflective mirror (HRM) in Fig 2.3.a is typically a 100 nm Ag layer. A simulation of the electrical field enhancement that can be achieved when fixing the DCL at an optimum thickness and optimizing for the MCL is shown in Figure 2.3.b. A broadband electrical field enhancement can be achieved by such structure, thus outperforming the ITO based cells with a short circuit current that may be larger by more than 2mA/cm2. Efficiencies reached the 11.13% in these devices.


Table 2.2. Comparison of PV parameters for the ITO reference (top), and cavity device (bottom).



We also designed and fabricated 4-terminal architectures, with the aim of achieving higher efficiencies. Fig 2.4a shows the structure using PCE10:PCBM in both subcells. The devices are separated by a dielectric layer, and the inner semi-transparent electrodes are based on thin Au films. The subcells are not subjected to current-matching in this type of configuration, thus the total photocurrent generated by the structure is equal to the sum of the individual subcells photocurrent. Figure 3b shows the calculated sum of photocurrents depending on the characteristics of a non-absorbing intermediate dielectric layer. Optimal cases for this dielectric can lead to photocurrents as high as 18.4 mA/cm2, surpassing the best single device photocurrent with ITO, which is 16.6 mA/cm2 as seen in Table 2.3.

b)
Figure 2.4. a) Scheme of 4-terminal device with PCE10:PCBM and Au inner electrodes, b) Calculated sum of photocurrents of the subcells as a function of the refractive index and thickness of a non-absorbing dielectric layer.
In an experimentally fabricated device, we obtained a photocurrent of 17.35 mA/cm2, which represents a 4% improvement over the single device with ITO, as seen in Table 2.3. The Voc and FF of the structure must still be optimized to improve the efficiency of the device, compared to the single.
Table 2.3. Comparison of experimental photocurrent by the 4-terminal device with gold and the single device with ITO.
Subcell Exp. 4-T gold
Jsc (mA/cm2) Exp. Single
Jsc (mA/cm2)
Top cell 10.24 16.6
Bottom cell 7.11
Sum of Jsc 17.35 16.6


WP3: Solution processed non-active OPV nano-materials
Task 3.1: Development of solution-processed buffer and/or recombination layers (M1-M24, Nanograde, FAU, ICFO)
The aim of Task 3.1 was to develop fully solution processable and easily printable semiconducting materials for electron- (HBL) and hole-transport layers in organic solar cells. These materials were to replace the state of the art materials (HBL: sol-gel ZnO, EBL: evaporated MoOx or printed PEDOT:PSS) which all show a variety of limitations such as high-temperature curing, vacuum-deposition process or very high price.
Nanograde together with FAU and ICFO was able to develop several HBL inks (ZnO or Al:ZnO with varying work function to adjust to energy levels of absorber polymers) and EBL inks (WOx or MoOx) that could be adjusted to any desired printing process (doctor-blade, slot-dye, ink-jet, etc.) by changing the solvent mixture and thus viscosity and drying behavior.
The HBL inks were quickly proven to be very suitable for the planned application (see figure, 1 left) and shown great results in PCE measurements. Nanograde’s “N-10” (ZnO in 2-propanol) became the standard HBL formulation for many OPV research groups around the world. Even though the EBL inks have not shown the desired performance and had in the course of the project been replaced by commercial PEDOT:PSS, Nanograde has still pushed their commercialization and has received good feedback from R&D teams outside the SOLPROCEL project.

Figure 1: Exemplary results generated with Nanograde’s ZnO HBL formulations showing very high performance (left) and image of bottled 1L products, developed in this project and subsequently commercialized on R&D scale.
Summarizing, Task 3.1 was very successful regarding HBL formulations for the SOLPROCEL project but only moderately successful regarding EBL formulations, even though the developed formulations have proven to perform adequately with certain absorber polymers and Solar Cell architectures outside of the SOLPROCEL project.
Task 3.2: Development of novel solution-processed photonic layers (M1-M24, Nanograde, SP, ICFO, FAU)
In task 3.2 a fully printable one-dimensional photonic crystal for the light harvesting enhancement while allowing semi-transparency of the OPV cells was to be developed. While Specific Polymers pursued a fully organic approach, while Nanograde worked on organic-inorganic hybrid layers.
Specific Polymers was able to develop and produce UV-curable monomers with resulting refractive indices of n>1.7 as well as n<1.4. These materials were optically immaculate but have not been used in the photonic crystal due to wettability problems during the coating steps.
Nanograde’s approach allowed for easily printable layers with good wettability and refractive indices of n>1.8 as well as n<1.3. These layer could easily be stacked on top of each other without redissolution by either simply fully drying the layers or subsequently UV-curing them.

Figure 2: Image of photonic crystals with varying double layer amounts and thus varying transmission/reflection characteristics.
At FAU the printing process of Nanograde’s inks was developed and improved to a level where the reflection and transmission characteristics of the photonic crystals could be reproducibly tuned to certain wavelengths. By changing the thickness of each layer within the stack, the reflectance maximum of the PC can be shifted to the absorption maximum wavelength of the applied absorber polymer within the solar cell. By changing the amount of high-n and low-n layers the absolute amount of transmitted and reflected light can be adjusted to a value where the highest PCE gain can be expected together with the lowest necessary overall transparency reduction.
In summary, Task 3.2 has been very successful and the SOLPROCEL consortium has gained the ability to produce photonic crystals with simple coating methods that match or even outperform state of the art vacuum-deposited PCs.
Task 3.3: Optimization of silver nanowire synthesis, upscale and cost efficient production (M1-M30, ras, FAU )
The objectives of the work in Task 3.3 were the successful scale-up of the synthesis and purification of silver nanowires, the development of methods for quality assurance and a cost estimation for the production. The main results are the installation of a pilot plant reactors for the synthesis and for the purification, to guarantee an optimized nanowire synthesis and an optimal product quality, respectively. Upscale a cost-efficient production of silver nanowires was realized by use of the pilot plant reactor. For further optimization, a quality assurance was established.
Optimization of silver nanowire synthesis by pilot plant reactor for synthesis
To optimize the silver nanowire synthesis a pilot plan reactor was set up. The maximum filling volume of the reactor comprises 20 l and allows a temperature control of ±0.5 °C within a temperature range of 25-150 °C. Stirring in different speed can be realized with a rotating stirrer with a propeller agitator. With respect to the cost efficiency of the silver nanowire production the scheduled quantities of the synthesis were varied. The first test runs were done with quantities of 10 kg; afterwards the quantity was increased to 16 kg and 13 kg. Numerous test runs were carried out successfully with respect reproducibility and silver nanowire quality.
Pilot plant reactor for purification
To remove the continuous phase (EG) and other impurities after synthesis, RAS developed a special phase separation process. A special solvent mixture is added to initiate a phase separation due to solubility-effects: The result of the phase separation is a pasty concentrate with silver nanowires with a concentration of around 4 wt% silver. This process was successfully established and scaled up to enable a purification in the same quantities compared to the pilot plant scale synthesis.
Quality assurance (QA)
To guarantee a high and stable quality of the synthesized nanowires, methods for quality assurance were developed and established. The quality assurance can be subdivided into three main steps. In QA step 1 the quality of the raw solution is analyzed. Important facts are the homogeneity of the raw product such as size and diameter of the silver nanowires that are analyzed by SEM images. After the purification of the raw product QA step 2 is carried out. Therefore, the yield of the synthesis is determined and the content of side products, which could not be removed by the purification step are analyzed by SEM. QA step 1 and 2 are classified as passed, if the nanowires show a mean length of 10 - 30 µm with a mean diameter of around 50 nm at a yield of min. 85 % and a low to medium content of small particles.
For the last QA step a formulation with defined silver nanowire content is prepared and coatings are prepared according to a constant procedure. Using a 4- point meter the surface resistance is measured and certain values have to be reached to pass QA step 3.
Using the pilot plant reactor and the optimized synthesis for silver nanowires the last batches (#1355-#1365), scheduled qualities of 13kg and 16kh, passed the QA step 1-3.
For further information on the storage behavior and altering of the formulation long time stability was analyzed. It could be found that the 1-K system is not stable over a period of 24h. The 2-K system showed good performance over a period of several days up to weeks. The main result of the long-time stability studies is: the surface resistance of coatings freshly prepared with altered concentrate was comparable to the surface resistance achieved with coatings freshly prepared from not altered concentrate.

Task 3.4 Fabrication of semi-transparent OPV devices including Ag nanowires, nano-particle inks, and solution-processed PC (M13-M30, FAU, ICFO).
FAU successfully fabricated solution-processed semitransparent solar cells and combined them with solution-processed photonic crystals. The resulting devices show an efficiency increase which is consistent with the prediction of optical simulations. Due to the easy tunability of the reflection maximum of the photonic crystal by changing the layer thicknesses, it can be adapted to the specific absorption spectrum of a particular absorber. To further understand the usefulness of the photonic crystal approach in terms of efficiency vs. transparency, we performed a theoretical investigation for which cases the combination with a photonic crystal is beneficial compared to simply increasing the layer thickness. It was concluded that the photonic crystal is always advantageous below a certain transparency (i.e. above a certain efficiency), and has the largest effect for absorbers with small band gaps, i.e. broad absorption spectra. In real devices for which the short-circuit current often drops with increasing thickness, the positive effect of the photonic crystal is even more pronounced. This observation was confirmed using devices with three different absorber materials.
References:
(1) C. Bronnbauer, J. Hornich, N. Gasparini, F. Guo, B. Hartmeier, N. A. Luechinger, C. Pflaum, C. J. Brabec, K. Forberich, “Printable Dielectric Mirrors with Easily Adjustable and Well-Defined Reflection Maxima for Semitransparent Organic Solar Cells, 2015, Adv. Optical Mater., 3, 1424-1430.
(2) C. Bronnbauer, N. Gasparini, C. J. Brabec, and K. Forberich, “Guideline for Efficiency Enhancement in Semi-Transparent Thin-Film Organic Photovoltaics with Dielectric Mirrors,” Adv. Opt. Mater., vol. 4, pp. 1098–1105, 2016.


WP4: OPV degradation mechanisms and device stability
Task 4.1: Identification of failure mechanisms, configurations and components in opaque OPV devices (M7 –M24, ICFO, FAU, FhG-IAP, Nanograde)
During the first year in Task 4.1 we established that materials used to fabricate the solar devices were highly stable under “no-stress” conditions. It was also shown that this changed dramatically once the fabricated devices were exposed to an aggressive environment including oxygen, humidity, temperature changes or UV light. We concluded that the only viable option to limit the effect of such agents was to encapsulate the devices. We showed that a proper encapsulation constitutes an effective barrier to oxygen and moisture. However, for obvious reasons, such encapsulation cannot be used as an effective barrier to high temperature/temperature oscillations or to block UV light.

We continued this work with the aim to obtain stable devices under any conditions and connect it with the activities in task 4.3 by studying the stability of encapsulated cells under the illumination from 1 sun mimicking realistic outdoor conditions. In the majority of the tests performed within this task and task 4.3 we considered the following cell architecture: ITO(Transparent electrode)/ZnO(ETL)/PCE10:PC71BM(Blend)/MoOx(HTL)/Ag(back electrode). From this architecture we concluded that the photoactive and electron transporting layers are the two layers that play the most relevant role in the overall stability of encapsulated devices under illumination: A 20 nm zinc oxide (ZnO) electron transporting layer (ETL) and a 95 ± 5 nm BHJ photoactive layer composed of the donor PTB7-Th and the acceptor [6,6]-phenyl-C71-butyric acid methyl ester (PC71BM, Figure 4.1.1B) can be seen in the cross-sectional scanning electron microscopy (SEM) image of the inverted cell architecture shown in Figure 4.1.1A.
Figure 4.1.1. A) SEM device cross-section, B) Blend material used in the stability experiments

We found that efficiency evolution curves for encapsulated cells under illumination could be accurately fitted using the superposition of two linearly independent exponential functions
"η (t) =A1·" "e" ^"-t/τ1" " + A2·" "e" ^"-t/τ2 " " "
where τ1 and τ2 are the time constants for the fast and slow exponential decay components, respectively. The fitting function was normalized in such a way that the sum of the A1 and A2 constant amplitudes would be equal to unity at t = 0. In a series of photo-stability tests we performed on encapsulated cells employing various UV cut-off filters (L-37, GG400 and GG455) we observed that photo-induced degradation could be greatly suppressed employing UV filters with cut off wavelength longer than 370 nm (See Figure 4.1.2) indicating that degradation is strongly accelerated with UV-photons with a cut-off wavelength in the 330-370 nm range. Hence, burn-in loss should be caused by deep UV-light induced degradation of the photo-active layer.
Figure 4.1.4. Amplitudes A1 of fast decay component for reference and Non-ZnO devices in terms of various wavelength cut-off filters (Without UV filter, L-370,GG400 and GG455) (square), and the normalized absorption of photoactive layer (90 nm) on soda-lime glass (circle).

Task 4.2: Identification of failure mechanisms, configurations and components in semi-transparent OPV devices (M13 – M30, FAU, Nanograde, ras)
In the implementation of this task we considered separately the stability of the Ag NW thin films and that of the photonic crystals. We performed accelerated stability on the Ag NW and considered basic stability issues for the fabricated photonic crystals. The conclusion for the Ag NW was that: Electrical properties of Ag NWs degrade fastest under light and air, Ag NWs do not degrade in the dark, and that Light/air and damp heat cause different degradation routes. For the photonic crystals we showed an excellent stability under different degradation conditions. The change in the spectral position of the transmittance minimum could be solved by encapsulation of the samples (which is a required step for organic solar cells as indicated in Task 4.1) and/or by pre-treatment of the samples. Alternatively, if the shift is known, it can be accounted for during fabrication, so that the photonic crystals exhibit the desired spectrum after degradation.

Task 4.3: Semi-transparent OPV device stability tested mimicking realistic outdoor conditions (M19–M36, ICFO,FAU)
Within Task 4.3 we implemented several kinds of optical filtering and a pre-treatment procedure to prevent degradation of the solar device under the solar simulator illumination and/or temperature changes. We performed accelerated tests and measured a T80 close to 15,000 hours. The key to achieve stable cells against light-stress was to obtain stable nano-scale morphologies for the blends. We used a PTB7-Th: PC71BM blend prepared from an optimized PCBM weight ratio and 1, 2-Dichlorobenzene (DCB) solvent. After fabrication optimizations, such cells could achieve similar device performance as standard cells with average PCEs of 9.3%. Figure 4.3.1a and figure 4.3.1b show the normalized PCE evolutions for both standard cells and morphology stabilized cells under full sun and UV-filtered (GG400) illumination, respectively. A huge improvement was achieved using morphology stabilized blends and the lifetime was increased by two orders of magnitude, reaching up to 400 hours even when no UV filter was employed. When a UV filter (GG400) was applied, a double-exponential extrapolated lifetime beyond 7 years was obtained assuming that 5.5 hours of light soaking in the lab are equal to one day of sun light soagking.

Figure 4.3.1. long-term lifetime tests for both standard cells and optimized cells under continuous AM 1.5G one sun illumination (a) and UV-filtered (GG400) illumination (b).

The architecture of the inverted semi-transparent cells used was similar to the stable opaque device. The semi-transparent device introduces a difference in the top electrode, which instead of being a 100 nm thick silver layer, was formed by a MoO3/Ag/MoO3 (MAM) structure using thin 10 nm thick silver in between the two MoO3 layers, as shown in Figure 4.3.2.

Figure 4.3.2. Device structure of semi-transparent cells.

Semi-transparent solar cells were fabricated by spin-casting 0.3M ZnO sol-gel solution on the pre-cleaned patterned ITO glass substrates (Lumtec, 15 Ω/sq ), then annealed on hotplate at 180 oC for 10 min to form 20 nm thick electron transporting layer. The blends were prepared by dissolving PTB7-Th: PC71BM in mixture solvent of o-dichlorobenzene (DCB) and DIO (3% Vol) with a total concentration of 36 mg mL−1 at 60 o C overnight and then spin-coated on the ZnO layer with controlled thickness of 95-100 nm. The resulting photoactive films were dried in vacuum (< 5× 10− 6 mbar) for at least one hour. Finally, MoO3 (10 nm, 0.5 Å/s) and Ag (10 nm, 6 Å/s) were sequentially deposited on the top of photoactive layer through a shadow mask by thermal evaporation in vacuum of < 5×10−6 mbar. Finally, the anti-reflection coating layer of 100 nm of MoO3 (1 Å /s) was deposited on top of thin Ag layer. The active area was 0.09 cm2. Solar cells were encapsulated with pre-cleaned glass slides using a UV-curable epoxy (DELO, LP655,) in N2-filled glovebox before measuring in air.

Measurement of stability in encapsulated organic semi-transparent devices under solar illumination. Encapsulated semi-transparent cells were exposed to solar irradiance using a Solar Simulator ABET 2000. Out of eight devices, seven were let in open circuit conditions, while one was connected to a 750 Ohms resistor. Under such specified conditions for stability test and IV curve measurements, the cells suffered from a minor degradation. In no case we observed any additional and abrupt degradation different than the behavior of opaque cells.

Measurement of stability in encapsulated organic semi-transparent devices under high humidity conditions. Encapsulated semi-transparent solar cells were exposed to conditions of high humidity (RH 90%) at 21ºC. A climatic chamber Vötsch was used for the tests. Samples were measured before the test, after 2 hours and then after 24 hour. The PV parameters of the cells remained stable during the period of test. Encapsulation with the UV-curable epoxy and the glass can adequately protect the semi-transparent solar cells from humidity conditions as high as RH 90%.

WP5: Up-scaling to OPV modules for commercially competitive applications
Task 5.1: Large area printing of a photonic crystal structure (M13-M36, FAU, Nanograde, ICFO)
In this task, up-scale batches of high refractive index and low refractive index inks from Nanograde were used to coat large-areas (10cm x 10cm on glass substrates and 20cm x 20cm on plastic substrates). Due to the excellent wetting and coating properties of the inks, we could achieve good optical quality and homogeneity of the PCs on both glass and PET substrates. Figure 5.1 shows the visual appearance of the layers as well as the influence of the thickness gradient in the coating direction. This gradient could be improved by changing the coating method from doctor blading to slot die coating. Some remaining point defects can be attributed to agglomerate formation in the ink, which could be improved by ink fabrication in a real production line. It has to be noted that those defects only affect the aesthetic quality of the layers and not the function of the PCs in combination with a semitransparent solar cell.

Figure 5.1: Fully printed upscaled PCs on PET (a) and glass substrates (b). c) Influence of the thickness gradient on the transmission spectra of the printed PC.
Just like the PCs fabricated on small substrates, the measured transmittance and reflectance spectra show good agreement with optical simulations performed with the transfer matrix method, signifying smooth interfaces and a performance similar to evaporated mirrors. The transmittance in the short-wavelength region was additionally improved by using a commercial low-refractive index ink which allowed the formation of denser layers. Optical simulations further confirm that porosity and disorder are responsible for relatively low transmittance in the short wavelength region and high haze.

Task 5.2 Large-area printing of OPV (M13-M36, FAU, ICFO, ras, SP, FhG-IAP, nanograde)

To produce large-area modules, the most important step is to create a series connection between the single cells which constitute the module. To this end, we chose laser structuring with a femto-second laser as it allows a high geometric fill factor and a comparatively homogeneous appearance of the modules. Since the up-scaling facilities at FAU only allow processing in ambient conditions, we focused on materials which are compatible with solution processing in air.

After establishing the architecture and especially the laser parameters for the different structuring steps, we used the up-scaled materials from Nanograde and ras, namely ZnO as buffer layer and nanowires as a semitransparent top electrodes, to fabricate semitransparent modules with an efficiency of 4.5% (with respect to the active area) and a transparency of 16% (~25% for state-of-the-art nanowires). The electrical and optical parameters of these modules are displayed in Figure 5.2. These modules show negligible losses compared to the single-cell reference, and the materials developed by ras and Nanograde perform at the same level as state-of-the-art materials. Modules fabricated with ras nanowires exhibit a slightly inhomogeneous appearance which can be attributed to bad wetting, which will be optimized by ras after the end of the project. All solution-processed layers are either doctor-bladed or slot-die coated, which means that they are fully compatible with roll-to-roll production for future upscaling.


Figure 5.2: Electrical and optical performance of up-scaled semitransparent modules with ras nanowires as top electrode. a) IV curve and performance data of the single cell reference with the thicker cell corresponding to the layer thicknesses which were chose for the module, b) JV data for one of the modules consisting of 12 single cells, c) summary of device parameters for the single cell and the module, d) digital image of the module (substrate size: 10cm x 10cm, active area 8cm x 8cm), e) measured total and diffuse transmission of the module with ras nanowires and reference nanowires as top electrodes.
Task 5.3 Combination of solution-processed OPV with solution-processed PC (M19-M36, FAU)
To investigate the suitability of different lamination methods, we performed degradation studies of various encapsulation materials. Due to the observed degradation, the printed PCs were simply attached to the printed modules at the edges. Figure 5.3 shows the visual appearance of the large-area modules that were combined with large-area photonic crystals for the benchmarking studies.

Figure 5.3: Digital images of the combination of large-area solar modules with large-area photonic crystals. a) solar module, b) two different photonic crystals with reflection maxima at 515nm (pink, left) and 465nm(yellow, right). c) combination of solar module with pink PC, d) combination of solar module with yellow PC.
These images also indicate how the combination of solution-processed module with solution-processed PC also allows to modify the color appearance, creating additional opportunities for window integration. With respect to efficiency, EQE measurements show a Jsc increase of 19% for a semitransparent cell with a transparency of 52% and a Jsc increase of 6.7% for a cell with a transparency of 24%.
Finally, to further assess the suitability of the photonic crystal approach for window integration, we studied the angle-dependence of Jsc which is generated by a solar cell combined with a PC. EQE measurements revealed that the improvement which is caused by the PC is largely independent of the angle of incidence up to 60°, meaning that the concept is highly suitable for windows where the angle of the sun varies during the course of the day.
References:
C. Bronnbauer, A. Riecke, M. Adler, J. Hornich, G. Schunk, C. J. Brabec, K. Forberich, „Printing of large-scale, flexible, long-term stable dielectric mirrors with suppressed side interferences“, submitted


WP6: End-users benchmarking
Task 6.1 Lamination of semi-transparent OPV cells (M10 – M21, COMSA EMTE, ICFO)
The main objective of this task is the definition, demonstration and validation of materials and industrial lamination process for semi-transparent SOLPROCEL OPV cells to be integrated in Façade Building Elements. Two laminating approaches and methodologies were performed. The first one of them was carried out in July (M21) using ENLIGHT as an encapsulant material and the second one in October (M24) using both ENLIGHT and EVA in different lamination tests.
LAMINATION TESTS IN JULY 2015
FAU prepared 8 substrates with a PCE10 absorber layer with 6 cells on each substrate and 5 substrates with a DPP absorber layer with also 8 cells on each substrate. The DPP cells had previously been pre-encapsulated. ICFO prepared 5 PCE10 devices with 8 OPV cells in each device. Several lamination profiles for the encapsulant material were tested with different peak temperatures and lamination times in order to minimize the degradation of the OPV cells. Along with the cells, glass plate testers were used in every lamination process to verify which lamination profile guaranteed better homogeneity and melting of the encapsulant as well as good adherence. An adherence Pummel test was performed to the glass plates after the lamination.
Both PCE devices from FAU and ICFO showed severe degradation after the first lamination tests. However, the DPP devices from FAU retained their average efficiency of 1.4%. The results proved that the encapsulation procedure is, in principle, suitable for SOLPROCEL solar cells, but that the ENLIGHT material severely damages the solar cell if it is applied without additional protection, as pre-encapsulation is. The results of the Pummel Test performed on the glass plates concluded that the minimum profile temperature profile validated for the lamination with ENLIGHT encapsulant is 110ºC – 30 minutes.
LAMINATION TESTS IN OCTOBER 2015
A different lamination approach was taken to avoid the severe degradation of the OPV cells, that is, to protect the OPV cells previous to the lamination with a pre-encapsulating glass fixated to the active layer with a UV cured resin without applying vacuum or temperature. With this approach, the EVA encapsulant could also be tested as it requires lower lamination temperature profiles and the pre-encapsulation would protect the cells from the damaging dye products generated from EVA.
For this lamination tests FAU provided 10 devices with 6 cells on each device, which means a total of 60 cells, with PCE10 absorber layer, Ag nanowires developed in previous WPs and pre-encapsulation protection. ICFO provided a total of 11 devices with 8 cells in each device, which adds up to a total of 88 cells, with PCE10 absorber layer and pre-encapsulation protection. The pre-encapsulation material chosen was the epoxy resin DELO-KATIOBOND, which is solvent-free, UV-curing and has a high barrier function against water vapor. The lamination temperature profile for ENLIGHT was the one determined in the previous tests (110ºC – 30 minutes). For EVA, the profile was 85ºC – 60 minutes.
The semitransparent PCE10 cells provided by FAU were measured after fabrication, after pre-encapsulation, after lamination and again after another period of ten days during which they were stored in inert atmosphere. After lamination, we may observe a drop in efficiency from approx. 1.4% to 0.7% which is mainly due to a drop in VOC from 0.78V to 0.52V. No substantial degradation was observed after the storage period after degradation. We therefore attribute this degradation to the temperature treatment during the lamination process. Since there was a long interruption between the initial fabrication of the cells and the final lamination for technical reasons, this drop in VOC might also be caused by intrinsic degradation of the cells. Semi-transparent PCE10 cells from ICFO were also tested in the laboratory, before the lamination, just after, and six, thirteen and twenty-six days after the lamination. After the lamination we may observe a drop from approximately 4.81% average efficiency of the tested cells laminated with ENLIGHT to an average 4.10% efficiency. When analyzing the efficiency drop from the EVA-laminated devices, we observe a drop from 5.02% to 4.41% after the lamination. Results obtained from six to twenty-six days after the lamination show minimum degradation of the efficiency of the cells for both, ENLIGHT and EVA laminated cells. Finally, because of the lower lamination temperature (85ºC – 60 minutes), EVA was chosen as the encapsulant.
Task 6.2 Preparation of semi-transparent OPV modules for benchmarking (M19 – M27, COMSA EMTE)
The main objective of T6.2 is the preparation of the OPV modules developed in WP5, as well as identifying the optimal procedure to integrate it into the building elements.
The first step towards the definition of the integration requirements of the SOLPROCEL OPV modules in building elements is the fabrication of up-scaled modules. The project fabricated entirely printed flexible solar modules with size of 105 x 105 mm² by slot-die coating. Due to the challenges of laser structuring a module with a bottom nanowire electrode, these modules were printed on a DMD electrode. The architecture is based on PV4610:PCBM active layer, a AgNWs semi-transparent top electrode and ZnO nanoparticles. These modules consist of 19 single cells. In contrast to most of the work reported in the literature, these modules were fabricated in inverted architecture which is expected to result in an improved long-term stability. As seen in T6.1 for the optimal development of future steps towards the final set-up of the OPV modules, these are pre-encapsulated with a transparent barrier foil. The encapsulation glue was the UV-curing Delo Katiobond.
In order to define the process and requirements for the integration of SOLPROCEL OPV modules in building elements it is important to differentiate between two approaches:
Initial approach: the modules that have been developed in SOLPROCEL with dimensions of 105x105 mm, 10V, 0.02A per modules and a power of 0.2Wp.
Future approach: the fabrication of larger scale SOLPROCEL modules in order to have a fully printed 1m x 1m module surface.

Task 6.3 Benchmarking semi-transparent OPV for façades (M25–M36, COMSA EMTE, ICFO)
The objective of T6.3 has been the performance of a comparative assessment of transparency, efficiency and stability between the technology developed in SOLPROCEL project in WP5 (Technology A) and two commercial OPV modules (Technologies B and C). The benchmarking was performed in a real building environment using an outdoor experimental set-up where the OPV modules were integrated in a south-oriented façade. The modules were characterized both in terms of electrical performance and spectral transmittance. The maximum electric power output was continuously monitored with MPPT developed specifically for the assessed modules. Moreover, daily measurements of the I-V curve and five spectral transmittance measurements were taken throughout the benchmarking period. A second batch of technology A was produced after the first did not show a satisfactory performance in terms of efficiency and FF.
Table Summary of the benchmarking results
Tech. τ -_i'400-800 (%) τ -_f'400-800 (%) Ef i (%) Ef f (%) Efficiency variation (%)
A 17.4 18.1 1.49 1.30 -12.75
A, batch 1 19.9 19.8 1.13 0.99 -12.40
A1 18.8 18.0 1.33 1.21 -9.02
A, batch 2 13.8 15.5 2.02 1.76 -12.87
B 14.22 13.53 3.59 3.01 -16.16
B2&B3 14.10 13.55 3.82 3.40 -11.00
C 11.2 10.6 2.36 2.44 +3.40


The results obtained indicate that transmittances of technology A in the visible range are the highest in average. The lowest transmittances are those of technology C, with values of about 11%. In between, technology B transmits 14.2% at the beginning and 13.5% at the end of the monitoring. In terms of transmittance variations, all the modules report a small difference from the beginning to the end, except for A batch 2 which increases 2 percentage units. The reason for this increase is associated with the fact that the modules were not light soaked when tests were started.

As for efficiencies, technology B outperforms the other two technologies achieving an initial efficiency of 3.59% and a final efficiency of 3.01%. After B, technology C reports an almost constant efficiency around 2.4%. Close to this value, the following highest efficiency is obtained by the second batch of technology A with initial and final values of 2.02% and 1.76%. The first batch of technology A reaches the lowest efficiency values, near 1%. For technology C, it can be noted that efficiency remains almost constant over the time; however for technologies A and B the relative efficiency variation takes values of 12.6% and 16.9% respectively. It should be noted that for technology B, the module B1 performs comparatively worse from the beginning than B2 and B3, which present a relative decrease in efficiency of near 11%. Moreover, it is worth mentioning that in the case of the first batch of A, module A1 outperforms A2 and A3 and suffers a lower efficiency reduction of 9% (being the lowest of all the modules except C) presenting higher efficiencies (1.33% at the beginning, 1.21% at the end). For this reason and as it has been indicated in several parts of the document, the representability of modules B1, A2 and A3 is not considered to be the best and specific comments about were included accordingly.
In summary, SOLPROCEL modules show a significantly better transparency compared to the other commercial technologies tested, which was a predominant point in the project’s main objectives in order to implement this technology in building semitransparent façades. Even though their efficiency is slightly lower than the others’, the values obtained paired with the better transparency results confirm the potential competitiveness in the market of the SOLPROCEL developed modules in these two aspects. However, only technology C shows stability during the benchmarking period.
In order to achieve a higher stability for the integration in buildings of SOLPROCEL modules, it is concluded that new lines of research should be conducted within the materials and manufacturing process of technology A. A possible alternative path to increase stability could be the development of a manufacturing processed similar to that of technology C, which takes place under a controlled atmosphere instead of open air.
WP7: Dissemination and exploitation
Task 7.1 Positioning SOLPROCEL at relevant Platforms, Associations and Conferences (M13-M36, COMSA CORP, ICFO, FAU)
The presentation of the SOLPROCEL project was possible in a series of conferences, newsletters, publications, etc. From the developed activities related to this task, it is important to highlight the following:
II ForoSolar “Retos y oportunidades del nuevo modelo energético” (COMSA CORP)
Publication of information about SOLPROCEL in COMSA CORP’s corporate press/media site. (COMSA CORP)
Publication of information about SOLPROCEL in an article in the press (El Periódico)
Presentation by Dr. Jordi Martorell to disseminate SOLPROCEL in different target groups such as Col·legi d’Enginyers and Foment (ICFO)
Building blocks of the future: 1st Czech-German Workshop on nanotechnology entrepreneurship (ras)
“Produktion der Zukunft: Mit Nanotechnologie effizienter wirtschaften” (ras)
Aachen Dresden International Textile Conference (ras)
SPIE Optics + Photonics 2015 (FAU)
Poster Presentation done by Dr. Alain Graillot in EUROSENSORS XXX 2016 in Budapest presenting the development of new low refractive index materials and active layer crosslinkers of interest for opto-electronic devices.

Task 7.2 Workshop organization (M19-M36, ICFO, FAU, COMSA CORP)
On October 7th 2016 COMSA CORP organized the OPV Workshop: “A New Technology Market”, which had a scientific approach and also a market-oriented approach in relation to BIPV. On the previous and following days of the workshop, communication actions were carried out to increase its visibility. During the workshop, proactive actions were performed on social networks:
Special report on El Periódico de Cataluña
Live coverage on Twitter
Dissemination on Press Room and Corporate Portal of COMSA CORP
Interview with Albert Cot for a special article in Solar News magazine
Elaboration of video-interviews of the speakers with reduced format and subsequent dissemination through social networks.
Task 7.3 Dissemination via the publication of results (M13-M36, ICFO, FAU)
The publication of the results of the project has been conducted through the following activities:
13th European Conference on Molecular Electronics (Fraunhofer and SP)
Poster presentation at Mathero Summer School (FAU)
Publication in “Advanced Optical Materials 3”, pages 1424-1430 (FAU)
Technical article of SOLPROCEL project in the magazine Sun&Wind Energy (COMSA CORP)
Advertisement in the magazine Sun&Wind Energy (COMSA CORP)
News about the evolution of the project on the Press Room and Corporate Portal and dissemination via corporate social networks (COMSA CORP)
As well as publication in scientific publication in peer review articles detailed in Template A1.
Task 7.4 Dissemination of SOLPROCEL’s technology benchmarking (M31-M36, COMSA CORP, ICFO)
The dissemination of the benchmarking results will be done in the courses of Energy efficiency and renewable energies and Control, Certifications and Auditing of the Degree in Agricultural and Food Engineering and the Masters Degree in Industrial Engineering, respectively. Moreover, the results will be presented in the following events and publications:
INSPIRES research center seminar
Article included in an internal publication of COMSA Group Conocimiento y Obra
International Conference on Building Integrated Renewable Energy Systems (BIRES 2017)
Task 7.5 Exploitation of semi-transparent OPV cells (M25-M36, COMSA CORP)
For the OPV modules large scale production the slot die technique has been chosen as the optimal method for the application of each layer. However, for the patterning lines which interconnect top and bottom electrodes and active layer, a laser system will be used. This patterning process is higher in precision and throughput and is also mask-free.
The flow-chart of the manufacturing of the modules is the following:


The estimated manufacturing cost of the SOLPROCEL modules (1m x 1m, 90% active area) adds up to 0,44 €/Wp for large scale manufacturing (1 million m2 per year), which will be a competitive price in the market.
The LCA of the production of pre-laminated OPV modules has been performed using project data for the core-processes (photonic crystal fabrication and pre-lamination of OP module with photonic crystal) and specific literature data for the fore-ground process (solar module fabrication). Main conclusions of LCA are the following:
The results show a very promising Energy Pay-back Time (EPBT), less than 4 months, for the SOLPROCEL OPV pre-laminated modules.
ITO-silver and silver-silver OPV cells have very similar environmental performance according to the present calculated indicators. The Primary Energy value is 255 MJ/m2 and the GWP is 14.1 kg CO2 eq/m2.
The most contributing steps to both impact categories are S1 (ITO or silver electrode processing), S4 (PEDOT:PSS deposition) and S7 (photonic crystal fabrication), followed close by S5 (silver electrode deposition).
The most contributing chemicals to both impact categories are: PEDOT:PSS, ITO, PET and silver.

In order to fulfil the standardization requirements for the entry of the OPV modules into the market, a standardization process has been carried out by COMSA Corp. in collaboration with AENOR.
In relation to the outcomes of the project, two approaches for the OPV modules have been considered for the identification of potentially interested standardization committees (at National, European and International level):
As translucent construction elements with the purpose of being considered in a similar way as any other construction component in order to allow sunlight getting into the building, like a window or a glazed curtain walling (CONSTRUCTION SECTOR)
As electric generator devices with solar radiation as the source of energy (PHOTOVOLTAIC SECTOR.
In the Spanish scenario two potentially interested groups were identified for each of the mentioned approaches: AEN/CTN 85 “CLOSURE OF FRAMES IN BUILDING AND RELATED PRODUCTS” and AEN/CTN 206/SC 82 “SOLAR PHOTOVOLTAIC ENERGY SYSTEMS”. While no response was received by the group related to the construction sector, the photovoltaic-related standardization committee did provide feedback for the standardization of the SOLPROCEL project. It also mentioned two working groups (both European and international) in which the scope of the SOLPROCEL project could fit: CLC/TC82/WG1 “Wafers, cells and modules” and IEC/TC82/WG2 “Modules, non-concentrating”.
According to the information provided by CLC/TC82/WG1:
There are no current activities at the CENELEC (CLC) level on organic PV.
At the IEC/TC82/WG2 there is a project team which is collecting the state-of-the-art regarding the characterization of slow responding devices, including OPV, DSSC and perovskite cells. This information will be included in a Technical Report, which is a type of standardization document and can be used as a basis for future International Standard developments.
The technology specific parts of the standard series IEC 61215-1 (Terrestrial photovoltaic (PV) modules – Design qualification and type approval) will be circulated among the national committees as a Final Draft International Standard. Taking into account the SOLPROCEL project’s scope, the most interesting part is IEC 61215-1-5: Special requirements for testing of flexible (non-glass superstrate) photovoltaic (PV) modules.
There is a work in progress at the technical committee IEC TC113 “Nanotechnology for electrotechnical products and systems” where the ongoing project is the technical specification EC/TS 62876-2-1 Ed 1.0: Nanotechnology – Reability assessment – Part 2.1: Nano-enabled photovoltaic – Stability test.
With regard to starting new work items within a standardization group of the results of SOLPROCEL project indicate that there is a gap in the standards at either the CENELEC or IEC level, then it is necessary to make a proposal for a new work item.
It was finally concluded that CLC/TC82/WG1 is the proper group to propose a new standardization work item. The following actions to be performed once the standardization technical committee has been identified are the development of a proposal in the proper draft international standard format and the submission of the New Work Item Proposal to IEC/TC82/WG2 “Modules, non-concentrating” through the appropriated forms.

Potential Impact:
Impact and Exploitation of Project:

Specific Polymers
SPECIFIC POLYMERS is a SME providing Research and Development services to industrial and academic customers in a very wide range of application fields. Thanks to the SOLPROCEL project, SP has been involved in a polymer chemistry field, the organometallic cross-coupling polymerization, in which the company was not involved so far. Thus, it has been possible to acquire competences and skills in (i) organic chemistry allowing the synthesis of donor and acceptor monomers and (ii) cross-coupling polymerization that enabled SP to prepare absorber polymers. Additional research work is needed on the cross-coupling polymerization reactions. In addition, due to the required level of purity of monomers targeted in this project, SP also developed new purification methods that were not used before (purification chromatography). It is a major step forward for the company, in this field of research but also for all other application fields SP is involved in. Beyond the field of absorber polymers, SP also developed low-RI polymeric thin layer based on poly(HFPO) and difluoro acrylate crosslinkers. Both these monomers or cross-linked things have already attracted some customers in the field of optics. Interest in such materials pushed SP to developed a wider range of poly(HFPO) based products. Last but not least, the cross-linkers developed in the scope of SOLPORCEL project are of high interest since they can be applied in lots of application fields. Indeed, it was proved during the projects that the azide moieties can react with polymeric chains bearing double bonds. Thus, such crosslinking additives seem to be a crosslinking pathway of great interest in many fields and SP is working on projects using similar azido-crosslinking agent in Rubbers. To conclude, all research and development progress made during the project offer the possibility for SPECIFIC POLYMERS to be involved in new application fields and to propose innovative solutions to its customers.

Nanograde
In the scope of SOLPROCEL Nanograde has developed the basis for a multitude of coating formulations that can not solely be used in the field of organic photovoltaics but also in a variety of related printed electronics applications such as perovskite solar cells, OLEDs or Displays. Furthermore some formulations developed in the project can be used for applications in completely different fields such as security printing or microoptics.
Nanograde has basically developed two families of coatings, the electronically active materials used as buffer layer materials in WP3 and the optically active materials used for the creation of the photonic crystal in WP3. Both material families are based on metal oxide nanoparticles that are produced by Nanograde’s patented flame spray pyrolysis process and subsequently dispersed into solvents and mixed with certain additives. The details of each product family and its potential exploitation is explained in more detail below
Electronic Materials (EBL & HBL)
Nanograde’s buffer layer materials (selective electron- and hole conductive layers) are based on dispersions of semi-conducting metal oxide nanoparticles such as zinc oxide or tungsten oxide. Based on the findings within the SOLPROCEL project Nanograde has been able to develop a broad portfolio of different types of EBL and HBL inks, all of them freely adjustable to necessary coating conditions. These materials have in the last three years been developed and distributed as R&D samples to research institutes and R&D departments of large electronics manufacturers alike all around the globe.
As with semi-conducting layers a huge number of prerequisites must be met for an effective final product, for each customer and each application the formulation must be adjusted or remade from scratch. A few examples of the most promising materials will further illustrate this complexity:
• Nanograde N-10: Zinc oxide. HBL for OPV and EIL for OLED
• Nanograde N-20X: Al-doped zinc oxide. HBL for OPV and perovskite PV, EIL for OLED
• Nanograde N-21X: Work function modified Al-doped zinc oxide. Improved light soaking stability. HBL for OPV and perovskite PV, EIL for OLED
• Nanograde P-10: Tungsten trioxide. EBL for OPV and HIL for OLED
All of the above mentioned formulations are currently under extensive testing in several electronics manufacturing companies all around the world and are potential candidates for large-volume production of printed electronic devices. Even though the results achieved with these materials in the SOLPROCEL project have shown to be only moderate, their exploitation may prove to be very important for Nanograde. With production volumes quickly expected to surpass 1Mio m2, the necessary amounts of formulation could be several tons per year.
Optical Materials (Low-n & High-n)
The optically active materials developed by Nanograde for the SOLPROCEL projects are based on transparent metal oxide nanoparticles with either very high or very low refractive index. Combined with the right type and amount of binders, layers with any RI between 1.2 and 2 can be achieved. Such layers can, if printed very thin and stacked on top of each other be used for the creation of Bragg mirrors (photonic crystals) or as thicker coated single layers be used as index matching layers in a variety of optoelectronic devices or as security feature in expensive printware such as banknotes. A few examples are:
• Anti-reflective coatings for windows, lenses or displays
• Photonic crystals for semi-transparent solar cells
• Security feature in banknotes
• Index matching layer in OLED stacks
• Index matching layer in Display assemblies
• IR reflector in greenhouse foils or glasses
Nanograde is currently involved in several co-operations with various industrial partners where these materials are tested for their suitability. As these materials can potentially be applied to such a broad variety of applications, their sales potential is considered very high.

ras
Within the framework of SOLPROCEL RAS has amongst others worked on the topics
• improve quality of silver nanowires
• formulations with silver nanowires
• scale-up of both the silver nanowire synthesis and purification in a pilot plant scale
These topics are from the point of view of RAS very important for the planned exploitation of the results, which is described more detailed in the following.
Within the framework of the SOPROCEL project RAS has successfully improved the quality of the silver nanowires and formulations thereof. In particular, RAS has worked on the reduction of the diameter, the avoidance of impurities and side products during the synthesis and the variation of length of silver nanowires. The achieved high quality of silver nanowires form the basis of the future exploitation. The better the quality of the silver nanowires, the better is the performance in a device, for example a solar cell. Therefore, the improvement with respect to quality increases the competitiveness of silver nanowires manufactured by RAS.
Furthermore, RAS has successfully developed methods to purify silver nanowires. In principle RAS has developed three different qualities with respect to purification:
• raw product of silver nanowires
• concentrate of silver nanowires obtained with a solvent extraction step
• highly purified silver nanowire obtained with several washing procedures
During the project, it turned out that for the application in solar cells highly purified silver nanowires are mandatory. However, the other qualities can indeed be used in other electronic applications. RAS is currently verifying whether the raw product can be used for example in antistatic coatings and applications, where the requirements are less strict. The concentrate of silver nanowires can be used in different transparent and conductive coatings. With the achieved quality for highly purified silver nanowires RAS can deliver raw materials for liquid processed transparent electrodes for use in the electronics sector, especially as transparent electrodes for solar cells. The future pricing for the three qualities will be different as well. Consequently, RAS considers to provide different qualities of silver nanowires in the future, depending on the customer application and requirements.
Another important development during the project was the successful scale-up of both the silver nanowire synthesis and purification in a pilot plant scale. These experiences form the basis of a further scale-up that RAS has currently started. Potential customers are mostly located in the fields of electronics and technical textiles.
Consequently, RAS is at this stage holding talks to different potential customers and planning marketing actions. For example RAS will present the silver nanowire technology at the European Coatings Show in spring next year. The first lighthouse projects and applications for commercial use are foreseen between 2017 and 2018. In the following table the exploitable foreground and its planned commercialization is summarized.
Type of Exploitable Foreground
Descriptionof exploitable foreground Confidential
Click on YES/NO Foreseen embargo date
dd/mm/yyyy Exploitable product(s) or measure(s) Sector(s) of application Timetable, commercial or any other use Patents or other IPR exploitation (licences) Owner & Other Beneficiary(s) involved
Improved quality of silver nanowires no electrode for solar cell electronics 2018 Beneficiary RAS AG
increased quantity of silver nanowires no electrode for solar cell 1. electronics
2. technical textiles 2017 Beneficiary RAS AG


COMSA
The aim of SOLPROCEL project is the development of an OPV technology suitable for its implementation in building glass façades so that building large glass surfaces can turn into cost-effective photovoltaic panels with homogeneous transparency, allowing the passage of visible light, and large energy generation. According to the topic FP7-NMP-2013-SMALL-7, the expected outcomes regarding the impact are (i) achieving an efficiency of the OPV module of at least 15% in a relevant environment and a considerable improvement in the modules’ lifetime by 2030, (ii) improving the efficiency of material use and OPV production processes, (iii) reaching a more favorable cost/efficiency ratio compared to inorganic PV and (iv) contributing to the implementations of the SET plan, particularly to the Materials Roadmap Enabling Low Carbon Energy Technologies. Nevertheless, because of their structure and organic materials, stability has been the critical point in all OPV technologies that have emerged to this day. As building elements, their stability should be of at least 10 years.
According to the results of the benchmarking that has been conducted to test the modules developed in SOLPROCEL alongside with two other commercial technologies, the former achieve a higher transparency, even though their efficiency values are slightly lower than the commercial technologies. New lines of research should be considered within the manufacturing of SOLPROCEL modules in order to achieve a higher stability that is absolutely necessary for their integration in buildings. A possible alternative path to increase stability could be the development of a manufacturing process similar to that of technology C, which takes place under a controlled atmosphere instead of open air.
A potential market size of 1.8 million square meter of eligible surface for modules with the adequate performance and stability is presumed to be a realistic scenario in 2030. This market size would represent between 1.4 and 1.8 billion Euros in sales, which is encouraging for future investments. In the aforementioned 2030 scenario, when considering the target efficiency of 15% and an active area of 90%, this eligible surface would be able to achieve a production of around 230 MW per year. Taking the modules cost at 21.59€/m2, as indicated in D7.2 the price per Wp of the modules would be 0.16€/Wp, significantly lower than the presumed cost of inorganic PV modules, which is estimated to be at around 0.25€/Wp by 2030.
In terms of environmental impact, the larger presence of energy self-generation systems in the share of renewables that would be achieved in the previous scenario would also imply a very large decrease in CO2 emissions related to the total energy mix in the grid of between 8,000 and 12,000 tCO2/year.

List of Websites:
http://www.solprocel.eu/