Unveiling the potential of heterogeneous catalysts for molecular solar thermal systems

Hot Paper Unveiling the Potential of Heterogeneous Catalysts for Molecular Solar Thermal Systems Alberto Gimenez-Gomez,[a] Benjamin Rollins,[b] Andrew Steele,[b] Helen Hölzel,[c, d] Nicolò Baggi,[e] Kasper Moth-Poulsen,[c, d, e, f] Ignacio Funes-Ardoiz,*[a] and Diego Sampedro*[a] Solar energy utilization has gained considerable attention due to its abundance and renewability. However, its intermittent nature presents a challenge in harnessing its full potential. The development of energy storing compounds capable of capturing and releasing solar energy on demand has emerged as a potential solution. These compounds undergo a photochemical transformation that results in a high-energy metastable photoisomer, which stores solar energy in the form of chemical bonds and can release it as heat when required. Such systems are referred to as MOlecular Solar Thermal (MOST)-systems. Although the photoisomerization of MOST systems has been vastly studied, its back-conversion, particularly using heterogeneous catalysts, is still underexplored and the development of effective catalysts for releasing stored energy is crucial. Herein we compare the performance of 27 heterogeneous catalysts releasing the stored energy in an efficient Norbornadiene/ Quadricyclane (NBD/QC) MOST system. We report the first benchmarking of heterogeneous catalysts for a MOST system using a robust comparison method of the catalysts’ activity and monitoring the conversion using UV-Visible (UV-Vis) spectroscopy. Our findings provide insights into the development of effective catalysts for MOST systems. We anticipate that our assay will reveal the necessity of further investigation on heterogeneous catalysis. Introduction In recent years, the need for sustainable energy solutions has become increasingly urgent due to the high demands of our modern world.[1] Energy-storing compounds based on solarinduced photoisomerization reactions have emerged as a promising avenue for addressing this challenge. These compounds undergo a photochemical transformation, resulting in the generation of a high-energy metastable photoisomer that stores solar energy in the form of chemical bonds. This energy can be released as heat when needed, making these systems ideal for providing energy on demand. Known as MOlecular Solar Thermal (MOST) systems,[2–5] they hold significant potential for revolutionizing the way we generate and store energy.[22–25] To be used as MOST systems, molecules must possess several key features (Figure 1a). First, they should absorb solar light efficiently, thanks to a good match between the solar spectrum and their absorption spectrum. Secondly, they should provide a high energy storage density, balancing storage energy and molecular weight. Finally, the back-conversion barrier from the high-energy photoisomer to the parent molecule should be large enough to store energy without reverting spontaneously, but easy to overcome when using a catalyst. In recent developments within the field, there have been endeavors to employ heterogeneous catalysts as triggers for the heat release in NBD/QC based MOST systems,[6–15] particularly cobalt,[7,16,22] and gold-based ones.[3,18–20] Additional desirable features for its integration into a device include a closed-cycle system that does not involve hazardous chemical elements in its daily performance. Experimental Section Protocol for Batch Catalytic Testing All catalysts tested were provided by Johnson Matthey (JM) and we selected the one of the best performing NBD/QC molecule-pair[17] [a] A. Gimenez-Gomez, Dr. I. Funes-Ardoiz, Prof. Dr. D. Sampedro Department of Chemistry Instituto de Investigación Química de la Universidad de La Rioja (IQUR) Universidad de La Rioja Madre de Dios 53, 26006 Logroño (Spain) E-mail: [email protected] [email protected] [b] B. Rollins, Dr. A. Steele Johnson Matthey Technology Centre Blounts Court Road, Sonning Common, RG49NH Reading, (UK) [c] Dr. H. Hölzel, Prof. Dr. K. Moth-Poulsen Department of Chemistry and Chemical Engineering Chalmers University of Technology Kemivägen 4, 41296 Gothenburg (Sweden) [d] Dr. H. Hölzel, Prof. Dr. K. Moth-Poulsen Department of Chemical Engineering Universitat Politècnica de Catalunya, EEBE Eduard Maristany 10–14, 08019 Barcelona (Spain) [e] Dr. N. Baggi, Prof. Dr. K. Moth-Poulsen The Institute of Materials Science of Barcelona, ICMAB-CSIC Bellaterra, 08193 ■■2x city■■Barcelona (Spain) [f] Prof. Dr. K. Moth-Poulsen Catalan Institution for Research & Advanced Studies, ICREA Pg. Lluís Companys 23, 08010 Barcelona (Spain) Supporting information for this article is available on the WWW under https://doi.org/10.1002/chem.202303230 © 2023 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH. This is an open access article under the terms of the Creative Commons Attribution Non-Commercial NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. Wiley VCH Freitag, 22.12.2023 2401 / 329912 [S. 88/91] 1 Chem. Eur. J. 2024,30, e202303230 (1 of 4) © 2023 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH Chemistry—A European Journal www.chemeurj.org Research Article doi.org/10.1002/chem.202303230 to develop our protocol (see Figure 1(b) for the molecule studied and Table S1 and Table S2 from the Supporting Information for the detailed list of catalysts). Then, we developed a simple protocol to compare different heterogeneous catalysts robustly and rapidly using UV-Vis spectroscopy to characterize the conversion, thanks to the very small overlap between NBD and QC absorption. To start the catalyst testing, NBD1 is first solubilized in toluene and then subjected to full irradiation towards QC1 using a LuzchemTM photoreactor that utilizes 14×8 W lamps with alternated UV-A and UV-B wavelength emission (see Figure 2). The duration of irradiation varies depending on the concentration and volume of the solution, but it typically ranges around 30 min for 10 mL of NBD1 at 104M concentration. To confirm the conversion of NBD1 to QC1, UV-Vis spectroscopy measurements are utilized (λNBD1=326 nm, λQC1= 290 nm). A threshold below 5 % of the intensity of NBD1 absorption band determines when the NBD1 has been fully irradiated. A calibration curve was prepared to assess this value and the range of linearity of the curve spans from 1×105M to 1×104M (see Figure 2). The value of absorption intensity of the NBD1 at 340 nm is used to calculate the concentration, to minimize the overlap between NBD1 and QC1 absorption bands, which is significant at the absorption maximum at 326 nm. After complete irradiation of the sample, catalysis was performed to investigate the activity of several catalysts in the back-conversion process. To assess the catalyst efficiency, a weight ratio of 10% w/w catalyst with respect to NBD1 is used, which typically involves adding the NBD1 solution to a vial containing 1 mg of catalyst per 10 mg of NBD1 in 10 mL of toluene, resulting in an NBD1 concentration of 4.5×103M. Due to the concentration exceeding the detection range of the UV-Vis spectrometer, dilution is required prior to measurement, typically at a ratio of 1:100, resulting in an NBD1 concentration of 4.5×105M before irradiation. In addition, solution is filtrated through a 0.22 μm PTFE filter to avoid dispersion artifacts in the measurement (See Figure S35 in the Supporting Information). Photometric quantification using UV-Vis has the disadvantage that does not allow to identify trace amounts of byproducts. An NMR experiment was set up (see Figure S34) to thoroughly assess the full conversion of NBD1 into QC1 and the catalyzed back-reaction, while also discarding the presence of any byproduct. After recording the UV-Vis spectrum, the concentration of NBD1 is calculated using a previously established calibration curve. This enables a straightforward comparison of catalyst efficiency. To recover the NBD1 from the catalytic media, a crystallization process was followed. The process involved the following steps: i) removing the catalyst from the solution by filtration using a 0.22 μm PTFE filter; ii) thoroughly drying the solution to eliminate any traces of toluene; iii) diluting the oily residue in the minimum amount of ethyl acetate necessary to dissolve the 10 mg of NBD1; iv) subsequently adding three times more hexane. It is worth noting that smaller volumes yield better results due to the higher final concentration achieved; v) heating the solution using a heat gun until complete solubility is achieved; and vi) allowing it to crystallize overnight in a freezer. The quantity of solvent used may vary, but a reasonable approximation is around 1 mL of ethyl acetate and 3 mL of hexane per experiment (10 mL of toluene for every 10 mg of NBD1). Typically, the various experiments were recovered simultaneously to increase concentration and facilitate crystallization. The comparative method developed in this study employs minimal amounts of catalyst and generates a low waste of NBD1 and solvents, particularly when considering the recrystallization process. Compared to traditional NMR experiments, the use of UV-Vis spectroscopy provides a costand time-effective alternative, which is easily replicable and yields robust results across a variety of catalysts, including homogeneous ones. Results and Discussion A closer look at the 10 best-performing catalysts at shorter times showed that only 5 Pt-based catalysts convert more than 50% of QC to NBD and that two Pt catalysts over carbon support outperform the rest of the pool reaching at least 90% conversion around the 4 h of catalysis (Figure 3). Also, platinum-based catalysts represent 9 of the 10 best-performing catalysts, indicating the privileged role of this metal in the backconversion reaction. We then analyzed the back-conversion at longer times (3 days) to explore the different factors that may affect the catalyst activity (Figure 4). We began by analyzing similar catalysts with different particle sizes, which can be extremely important when moving MOST systems to flow regimes. Figure 1. a) Schematic view of a MOST system. b) 2-Cyano-3-(4-methoxyphenyl)-norbornadiene (NBD1) isomerization into 2-Cyano-3-(4-methoxyphenyl)-quadricyclane (QC1). c) Output of this work. Wiley VCH Freitag, 22.12.2023 2401 / 329912 [S. 89/91] 1 Chem. Eur. J. 2024,30, e202303230 (2 of 4) © 2023 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH Chemistry—A European Journal Research Article doi.org/10.1002/chem.202303230 15213765, 2024, 1, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/chem.202303230 by Readcube (Labtiva Inc.), Wiley Online Library on [22/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Most of the catalysts were in a powder form, although a series of them was acquired in relatively large particle size (Pt and Pd 200 μm-particles, Pd-2 mm chunks, and Pt-spheres under 1 mm). All these catalysts are dispersed over carbonbased support. Although conversion was partial, due to the limited surface area of the catalyst, palladium achieved higher conversion than platinum throughout the tests. These results suggest that smaller particle size provides a higher backconversion when the metal-core is matched, which can be correlated to the higher surface area in contact with the solution at the same time. This leads to the next categorization of the catalysts: the metal loading. Two catalysts were acquired with the same support and same particle size (powder around 50 μm) with Figure 2. Schematic view of the catalyst comparison protocol. First, NBD1 is irradiated under UV-A/B light for 30 min.; then the catalyst is added at a 10% w/w to the sample. For every data point over time, an aliquot is diluted and filtered before the measurement. UV-Vis spectra of QC1 catalysis into NBD1 at different irradiation times are recorded. Measurements at 340 nm are interpolated into the NBD1 calibration curve and the percentage of back-conversion is calculated. Inset graph corresponds to NBD1 calibration curve. Figure 3. Back-conversion activity of the 10 best-performing catalysts in the first 4 h of catalysis. Figure 4. Back-conversion achieved by every catalyst evaluated after a 3-day protocol. Platinum-based catalysts outperform the rest of the pool, representing 9 of the top 10 catalysts. Further details on the catalysts in Table S1 and Table S2 in the Supporting Information. Color code: Pt (grey), Pd (blue), Ru (brown). Wiley VCH Freitag, 22.12.2023 2401 / 329912 [S. 90/91] 1 Chem. Eur. J. 2024,30, e202303230 (3 of 4) © 2023 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH Chemistry—A European Journal Research Article doi.org/10.1002/chem.202303230 15213765, 2024, 1, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/chem.202303230 by Readcube (Labtiva Inc.), Wiley Online Library on [22/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License different platinum loadings (5 and 10% in weight). The latter one could achieve complete conversion under the 24 h mark, while the 5% loaded catalyst reached 83.2% of the conversion at the end of the 3-day experiment. Although this is a clear result towards a high loading-high activity conclusion, other factors might influence in a non-linear way the amount of metal used. Besides the economic cost increase, a different type of support might not get benefited from larger amounts of metal. Indeed, the benchmark only has a 3% weight platinum loading and performs 6× faster (around 240 min). There are other factors that affect the catalyst activity that were not considered in this work. For example, variations in the adsorption and desorption of products and byproducts on the catalyst’s surface may lead to differing availability of the active surface, and thus the catalytic activity.[26] Conclusions We have reported a fast and efficient protocol for heterogeneous catalyst testing in the back-conversion reaction of a QC to NBD isomerization pair. This will be key in the development of realistic MOST closed-cycle systems in the future, as the back-conversion reaction is much less explored than photochemical isomerization. Also, this protocol can be easily adapted to related reactions or catalysts when there is no competing absorption between both isomers in the MOST system. Similarly, after conducting a thorough examination of a set of commercial catalysts and analyzing key features that impact the rate of the back-conversion reaction, the privileged role of platinum as the metal core and the importance of a relatively small particle size catalyst with carbon-based support was demonstrated, providing catalyst Pt/C #14-1 as the best performing catalyst for this transformation among the 27 catalysts tested. This highlights the necessity of a multi-variant analysis comparison that can consider the influence of different parameters simultaneously to design the optimal catalyst for this MOST system. Acknowledgements We thank the European Union’s H2020 research and innovation program under grant agreement N:951801 (MOST H2020-EICFETPROACT-2019-951801). Some results are part of the project TED2021-131896B-I00 financed by MCIN/AEI/10.13039/ 501100011033 and the European Union “Next Generation EU”/ PRTR. I. F.-A. thanks the MICINN of Spain for financial support (IJC2020-045125-I). As part of the FETPROACT MOST project, all catalysts tested were provided by Johnson Matthey (JM). Conflict of Interests The authors declare no conflict of interest. Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request. Keywords: energy ·heat release ·heterogeneous catalysis · MOST ·UV-Vis [1] BP, bp Statistical Review of World Energy 2021, (71st edition),2022, 1. [2] Z. Wang, P. Erhart, T. Li, Z.-Y. Zhang, D. Sampedro, Z. Hu, H. A. Wegner, O. Brummel, J. Libuda, M. B. Nielsen, K. Moth-Poulsen, Joule. 2021,5, 3116–3136. [3] E. Franz, C. Stumm, F. Waidhas, M. Bertram, M. Jevric, J. OrregoHernández, H. Hölzel, K. Moth-Poulsen, O. Brummel, J. Libuda, ACS Catal. 2022,12, 13418–13425. [4] A. Lennartson, K. Moth-Poulsen, in Molecular Devices for Solar Energy Conversion and Storage, Vol. 1 (Eds: H. Tian, G. Boschloo, A. Hagfeldt) Springer Nature, Singapore. 2018, pp. 327–352. [5] A. Gimenez-Gomez, L. Magson, B. Peñin, N. Sanosa, J. Soilán, R. Losantos, D. Sampedro, Photochemistry 2022,2, 694–716. [6] M. Quant, A. Lennartson, A. Dreos, M. Kuisma, P. Erhart, K. Börjesson, K. Moth-Poulsen, Chem. Eur. J. 2016,22, 13265–13274. [7] M. Jevric, A. U. Petersen, M. Mansø, S. Kumar Singh, Z. Wang, A. Dreos, C. Sumby, M. B. Nielsen, K. Börjesson, P. Erhart, K. Moth-Poulsen, Chem. Eur. J. 2018,24, 12767–12772. [8] Z. Wang, R. Losantos, D. Sampedro, M. A. Morikawa, K. Börjesson, N. Kimizuka, K. Moth-Poulsen, J. Mater. Chem. A. 2019,7, 15042–15047. [9] J. L. Elholm, A. E. Hillers-Bendtsen, H. Hölzel, K. Moth-Poulsen, K. V. Mikkelsen, Phys. Chem. Chem. Phys. 2022,24, 28956–28964. [10] M. J. Kuisma, A. M. Lundin, K. Moth-Poulsen, P. Hyldgaard, P. Erhart, J. Phys. Chem. C. 2016,120, 3635–3645. [11] F.-Y. Meng, I.-H. Chen, J.-Y. Shen, K.-H. Chang, T.-C. Chou, Y.-A. Chen, Y.- T. Chen, C.-L. Chen, P.-T. Chou, Nat. Commun. 2022,13, 797. [12] Z. Wang, Z. Wu, Z. Hu, J. Orrego-Hernández, E. Mu, Z.-Y. Zhang, M. Jevric, Y. Liu, X. Fu, F. Wang, T. Li, K. Moth-Poulsen, Cell Rep. 2022,3, 100789. [13] S. Miki, Y. Asako, M. Morimoto, T. Ohno, Z. Yoshida, T. Maruyama, M. Fukuoka, T. Takada, Bull. Chem. Soc. Jpn. 1988,61, 973–981. [14] K. Maruyama, H. Tamiaki, S. Kawabata, J. Chem. Soc.-Perkin Trans. 1986, 2, 543–549. [15] K. Maruyama, H. Tamiaki, J. Org. Chem. 1986,51, 602–606. [16] Z. Wang, H. Hölzel, K. Moth-Poulsen, Chem. Soc. Rev. 2022,51, 7313– 7326. [17] R. Eschenbacher, T. Xu, E. Franz, R. Low, T. Moje, L. Fromm, A. Gorling, O. Brummel, R. Herges, J. Libuda, Nano Energy. 2022,95, 107007. [18] P. Lorenz, T. Luchs, A. Hirsch, Chem. Eur. J. 2021,27, 4993–5002. [19] F. Hemauer, U. Bauer, L. Fromm, C. Weiss, A. Leng, P. Bachmann, F. Dull, J. Steinhauer, V. Schwaab, R. Grzonka, A. Hirsch, A. Gorling, H. P. Steinruck, C. Papp, ChemPhysChem. 2022,23. [20] E. Franz, J. Jung, A. Kunz, H. A. Wegner, O. Brummel, D. Mollenhauer, J. Libuda, J. Phys. Chem. Lett. 2023,14, 1470–1477. [21] T. Luchs, P. Lorenz, A. Hirsch, ChemPhotoChem. 2020,4, 52–58. [22] Z. Wang, A. Roffey, R. Losantos, A. Lennartson, M. Jevric, A. U. Petersen, M. Quant, A. Dreos, X. Wen, D. Sampedro, K. Börjesson, K. Moth-Poulsen, Energy Environ. Sci. 2019,12, 187–193. [23] A. Dreos, K. Börjesson, Z. Wang, A. Roffey, Z. Norwood, D. Kushnir, K. Moth-Poulsen, Energy Environ. Sci. 2017,10, 728–734. [24] J. Orrego-Hernández, H. Hölzel, Z. Wang, M. Quant, K. Moth-Poulsen, in Molecular Photoswitches: Chemistry, Properties, and Applications, Vol. 2. (Ed: Z. L. Pianowski), Wiley-VCH, Weinheim. 2022 pp. 351–378. [25] J. Orrego-Hernández, H. Hölzel, M. Quant, Z. Wang, K. Moth-Poulsen, Eur. J. Org. Chem. 2021,38, 5337–5342. [26] R. S. Shamsiev, J. Mol. Model. 2023,29, 342. Manuscript received: October 2, 2023 Accepted manuscript online: November 10, 2023 Version of record online: November 29, 2023 Wiley VCH Freitag, 22.12.2023 2401 / 329912 [S. 91/91] 1 Chem. Eur. J. 2024,30, e202303230 (4 of 4) © 2023 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH Chemistry—A European Journal Research Article doi.org/10.1002/chem.202303230 15213765, 2024, 1, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/chem.202303230 by Readcube (Labtiva Inc.), Wiley Online Library on [22/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License