Molecular solar thermal energy storage (MOST) systems based on photochromic molecules that undergo photoisomerization to high‐energy isomers are attractive for storage of solar energy in a closed‐energy cycle. One challenge is to control the discharge time of the high‐energy isomer. Here we show that incorporation of a strong acceptor substituent in the seven‐membered ring of the dihydroazulene/vinylheptafulvene (DHA/VHF) couple increases the half‐life of the energy‐releasing VHF‐to‐DHA back‐reaction from hours to more than a day in a polar solvent. For some derivatives, the absorption maximum of the photo‐active DHA is also significantly redshifted, thereby better matching the solar spectrum. Synthetic protocols and kinetics studies are presented together with a computational study of the energy densities of the systems and excitation spectra. The computations show that the increased lifetime of the high‐energy isomer is counter‐balanced by a lower energy storage capacity in vacuo than for the parent system, but a slightly higher energy density than for the parent system in a polar solvent.

Cu is currently the most effective monometallic catalyst for producing valuable multicarbon-based (C 2+ ) products, such as ethylene and ethanol, from the CO 2 reduction reaction (CO 2 RR). One approach to optimize the activity and selectivity of the metal Cu catalyst is to functionalize the Cu electrode with a molecular modifier. We investigate from a data standpoint whether any reported functionalized Cu catalyst improves the intrinsic activity and/or multicarbon product selectivity compared to the performance of bare Cu foil and the best single crystal Cu facets. Our analysis shows that the reported increases in activity are due to increased surface roughness and disappear once normalized with respect to electrochemical surface area. The intrinsic activity generally falls below that of the bare Cu foil reference, both for total and product-specific current, which we attribute to nonselective blocking of active sites by the modifier on the surface. Instead, an analysis of various polymer diffusion coefficients indicates that the modifier allows for easier diffusion of CO 2 compared to H 2 O to the surface, leading to greater selectivity for CO 2 RR and C 2+ products. As such, our analysis finds no catalyst for CO 2 RR that intrinsically outperforms bare Cu.

Copper offers unique capability as catalyst for multicarbon compounds production in the electrochemical carbon dioxide reduction reaction. In lieu of conventional catalysis alloying with other elements, copper can be modified with organic molecules to regulate product distribution. Here, we systematically study to which extent the carbon dioxide reduction is affected by film thickness and porosity. On a polycrystalline copper electrode, immobilization of porous bipyridine-based films of varying thicknesses is shown to result in almost an order of magnitude enhancement of the intrinsic current density pertaining to ethylene formation while multicarbon products selectivity increases from 9.7 to 61.9%. In contrast, the total current density remains mostly unaffected by the modification once it is normalized with respect to the electrochemical active surface area. Supported by a microkinetic model, we propose that porous and thick films increase both local carbon monoxide partial pressure and the carbon monoxide surface coverage by retaining in situ generated carbon monoxide. This reroutes the reaction pathway toward multicarbon products by enhancing carbon–carbon coupling. Our study highlights the significance of customizing the molecular film structure to improve the selectivity of copper catalysts for carbon dioxide reduction reaction.

The catalytic reduction of CO 2 /CO is key to reducing the carbon footprint and producing the chemical building blocks needed for society. In this work, we performed a theoretical investigation of the differences and similarities of the CO 2 /CO catalytic reduction reactions in gas, aqueous solution, and aprotic solution. We demonstrate that the binding energy serves as a good descriptor for the gaseous and aqueous phases and allows catalysts to be categorized by reduction products. The CO* vs O* and CO* vs H* binding energies for these phases give a convenient mapping of catalysts regarding their main product for the CO 2 /CO reduction reactions. However, for the aprotic phase, descriptors alone are insufficient for the mapping. We show that a microkinetic model (including the CO* and H* binding energies) allows spanning and interpreting the reaction space for the aprotic phase.

The dihydroazulene/vinylheptafulvene (DHA/VHF) photocouple is a promising candidate for molecular solar heat batteries, storing and releasing energy in a closed cycle. Much work has been done on improving the energy storage capacity and the half-life of the high-energy isomer via substituent functionalization, but similarly important is keeping these improved properties in common polar solvents, along with being soluble in these, which is tied to the dipole properties. However, the number of possible derivatives makes an overview of this combinatorial space impossible both for experimental work and traditional computational chemistry. Due to the time-consuming nature of running many thousands of computations, we look to machine learning, which bears the advantage that once a model has been trained, it can be used to rapidly estimate approximate values for the given system. Applying a convolutional neural network, we show that it is possible to reach good agreement with traditional computations on a scale that allows us to rapidly screen tens of thousands of the DHA/VHF photocouple, eliminating bad candidates and allowing computational resources to be directed toward meaningful compounds.