Chemical Space of Molecular Nanomotors: Optimizing Photochemical Properties for One- and Two-photon Applications
Alexander Mielke, Alexander Scrimgeour, Enrico Tapavicza · 2025
Read Chemical Space of Molecular Nanomotors: Optimizing Photochemical Properties for One- and Two-photon Applications by Alexander Mielke, Alexander Scrimgeour, Enrico Tapavicza (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-20328.txt.
Introduction
Alexander Mielke, Alexander Scrimgeour, and Enrico Tapavicza search chemical space for better light-driven molecular motors, in a 2025 paper. Tiny motors turned by light promise uses in materials and medicine. Past work improved single candidate molecules one at a time. Their study instead screens a large family of designs at once.
The data set held just over two thousand motors with varied push-pull groups. Top candidates absorbed two-photon light far more strongly than known motors. That trait matters for medical uses that need deep-penetrating red light. A new score checks that the excited state keeps the character needed for motion. Three machine learning models were tested for predicting the traits. Their accuracy was excellent, the authors report. Learned models could replace costly quantum calculations in such searches.
Catalog record
| CREATOR | Alexander Mielke, Alexander Scrimgeour, Enrico Tapavicza |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 52 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-20328
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Chemical Space of Molecular Nanomotors: Optimizing Photochemical Properties for One- and Two-photon Applications Alexander Mielke, Alexander Scrimgeour, Enrico Tapavicza Abstract Light-driven molecular nanomotors hold promise for applications in material science and biomedicine. Significant efforts have focused on improving their efficiency, often targeting single candidate molecules. Here, we present a systematic data-driven approach to design nanomotors with high isomerization quantum yields for one- and two-photon applications, the latter being critical for biomedical applications requiring near-infrared light. We analyze the excited state properties of a dataset of 2016 nanomotors substituted with electron-donating and electron-withdrawing (push-pull) groups. Among the the top candidates, we achieved an increase in two-photon absorption strengths of up to two orders of magnitude compared to existing nanomotors. To ensure that the π − π ∗ \pi-\pi^{*} -character of the excited state is preserved, which is necessary to achieve the required photoisomerization, we introduce a photoreactivity score, that gauges the excited state character based on the transition. Furthermore, we benchmark three machine learning (ML) models—Kernel Ridge Regression, XGBoost, and a Neural Network—using physical and connectivity-based molecular descriptors. The excellent accuracy of our ML predictions holds promise to replace computationally costly quantum chemistry calculations in chemical space explorations. keywords American Chemical Society, L a T e X † † affiliation: Institute of Chemistry and Pharmacy, University of Regensburg, Universitaetsstrasse 31, 93041 Regensburg, Germany † † affiliation: Contributed equally to this work † † affiliation: Institute of Chemistry and Pharmacy, University of Regensburg, Universitaetsstrasse 31, 93041 Regensburg, Germany † † affiliation: Contributed equally to this work † † affiliation: Institute of Chemistry and Pharmacy, University of Regensburg, Universitaetsstrasse 31, 93041 Regensburg, Germany † † affiliation: Department of Chemistry and Biochemistry, California State University, Long Beach, 1250 Bellflower Boulevard, Long Beach, California 90840-9507, United States † † email: [email protected] † † abbreviations: IR,NMR,UV † † suppinfo: 1 Introduction In the last two decades, the development of light-driven molecular nanomotors (MNMs) has led to a variety of possible applications, such as mesoscale photoresponsive materials Deng et al. 2024 and applications in biology García-López et al. 2017 and medicine García-López et al. 2019 . Similar to photoswitches, which have been extensively proposed for medical applications Velema et al. 2014 ; Grathwol et al. 2019 , light-driven MNMs also have the advantage to be controllable with high spatial and temporal precision. A large number of the second generation MNMs are based on the overcrowded alkene architecture Koumura et al. 1999 ; Koumura et al. 2002 ; Pooler et al. 2021 (Figure 1 ). To achieve unidirectional rotation, photochemical EZ-isomerization ( h\nu ) alternates with thermal helix inversion (THI, \Delta ), as shown in Figure 1 . Figure 1: Mechanism of unidirectional rotation of an overcrowded alkene MNM, here exemplified for the generic isomer pair (3-D,5-A/3-D,8-A)-X . Photochemical steps are denoted by h\nu , thermal steps are denoted by \Delta . D and A are unspecified donor and acceptor substituents, respectively. The chiral center is marked with an asterisk. A major objective in designing MNMs is to increase the frequency of rotation, which entails both increasing EZ-photoisomerization quantum yield and THI rates. While increasing the THI rates has been the subject of several studies, Koumura et al. 2002 ; Koumura et al. 2000 here we focus on maximizing the photochemical EZ-isomerization quantum yield. The photochemical isomerization quantum yield depends on two key quantities: Thompson and Tapavicza 2018 ; Tapavicza et al. 2018 i) the amount of photons absorbed by the MNM, and ii) the probability of successful EZ-photoisomerization once the molecule is excited. In case of a typical one-photon absorption, the oscillator strength f describes the ability to absorb photons. A prime objective in designing efficient MNMs is therefore increasing the oscillator strength. However, applications in biological tissue often require near infrared (NIR) excitation due to its better penetration depth and lower phototoxiticity. Ash et al. 2017 Recently large progress has been made in developing photoresponsive materials with one-photon excitation energies in the NIR. Hartinger et al. 2025 The typical energy gap of overcrowded alkenes, however, lies in the UV/Vis region. Therefore, excitation with NIR radiation can only be achieved by a two-photon process, leading to a doubling of the corresponding one-photon excitation wavelength (or halfing of the energy gap). Although two-photon processes pose challenges in practical applications due to their low sensitivity, a variety of molecular sensitizers for photodynamic therapy for two-photon absorption (TPA) have been developed Bolze et al. 2017 . Recently several prototype MNMs have been harnessed to function by two-photon excitations. Liu et al. 2019 ; Guinart et al. 2024 However, due to the low two-photon absorption strength (TPAS), this process still requires high radiation intensity, which can cause unwanted tissue damage in biological applications. Alabugin 2019 Therefore, increasing TPAS is a major route to pave the way for two-photon applications in photodynamic therapy. Once the molecule is promoted to the excited state, it still needs to undergo the desired photochemical transformation, to achieve the desired mechanical motion. In the case of EZ-isomerization, the probability of isomerization is related to the orbital character of the electronic transition: a \pi-\pi^{*} character typically leads to the desired double-bond EZ-isomerization Lucia-Tamudo et al. 2025 . Thus, increasing absorption strength does not necessarily lead to increased quantum yield. One still needs to ensure that the state reached by excitation has the required \pi-\pi^{\ast} orbital character. Lately, a large effort has been undertaken in applying computational tools to study the mechanism of MNMs and to increase their efficiency Stauch and Dreuw 2016 ; Nikiforov et al. 2016 ; Oruganti et al. 2015 ; Nikiforov et al. 2016 ; Filatov et al. 2022 ; Filatov et al. 2024 ; Wen et al. 2023 ; Lucia-Tamudo et al. 2025 . In an attempt to initiate a data-driven approach Vela et al. 2019 ; Tapavicza et al. 2021 to identify promising candidates for nanomotor applications, we systematically explore the chemical space of substituted MNMs. Since one of our objectives is to apply MNM to TPA, we study derivatives of a Feringa-type Augulis et al. 2009 overcrowded alkene MNM substituted with electron donor-acceptor pairs (push-pull systems) (Figure 2 ), which have been shown to effectively increase the TPAS Albota et al. 1998 ; Pawlicki et al. 2009 . Figure 2: (A) Base structure of the overcrowded alkene MNM with atom numbering. Positions R 1 -R 8 are used as possible sites for substituents. For each candidate molecule, one position is substituted with one electron donor group (B) and another position is substituted with one electron acceptor group (C). We restrict our study to molecules with one donor-acceptor pair only. Hammett parameters ( \sigma_{p} ), indicating electron-donating or electron-withdrawing strengths are given Hansch et al. 1991 . To this end, we generated a comprehensive molecular dataset of 2016 molecules and analyzed the impact of substituent groups on photochemical excited-state properties (Figure 2 ). For all generated structures, excited state quantum mechanical calculations were performed using time-dependent density functional theory (TDDFT) Runge and Gross 1984 . The effects of the donor and acceptor groups and occupied positions on the excitation energies \omega , oscillator strengths f , and TPAS \sigma values were analyzed. An additional analysis of the transition densities was also performed to