Dr. Alexander Croy

Dr. Croy's research mainly focuses on computational efforts, especially machine learning techniques and simulations for the development and description of nanomaterials. His research interests also include electronic olfaction and the associated interdisciplinary challenges. He is affiliated with the Chair of Physical and Theoretical Chemistry as Akademischer Rat ("assistant professor/lecturer"). 

  • Perceptronics: Combination of sensor technology and machine learning.
    Image: A. Bierling
    Olfactorial PerceptronicsExternal link

    The perceptronics research group works as an interdisciplinary team on topics of perceptive electronics (short:perceptronics) for olfaction.

Alexander Croy, Dr
Senior Researcher and Lecturer (akad. Rat)
Professorship of Theoretical Chemistry
Room 105
Lessingstraße 4
07743 Jena Google Maps site planExternal link

26 Publikationen filtern

Die Publikationen filtern

Highlighted authors are members of the University of Jena.

  1. Semiconductor Bloch equations in Wannier gauge with well-behaved dephasing

    Year of publicationStatusReview pendingPublished in:Computer physics communications : an international journal and program library for computational physics and physical chemistry M. Thümmler, T. Lettau, A. Croy, U. Peschel, S. Gräfe
    The semiconductor Bloch equations (SBEs) with a dephasing operator for the microscopic polarizations are a well established approach to simulate high-harmonic spectra in solids. We discuss the impact of the dephasing operator on the stability of the numerical integration of the SBEs in the Wannier gauge. It is shown that the commonly used phenomenological approach to apply dephasing is ill-defined in the presence of band crossings and leads to artifacts in the carrier distribution. They are caused by rapid changes of the dephasing operator matrix elements in the Wannier gauge, which render the convergence of the simulation in the stationary basis infeasible. In the comoving basis, also called Houston basis, these rapid changes can be resolved, but only at the cost of a largely increased computation time. As a remedy, we propose a modification of the dephasing operator with reduced magnitude in energetically close subspaces. This approach removes the artifacts in the carrier distribution and significantly speeds up the calculations, while affecting the high-harmonic spectrum only marginally. To foster further development, we provide our parallelized source code.
    University Bibliography Jena:
    fsu_mods_00029532External link

  2. High-Performance Phototransistor Based on a 2D Polybenzimidazole Polymer

    Year of publicationPublished in:Advanced Materials A. Prasoon, P. Dacha, H. Zhang, E. Unsal, M. Hambsch, A. Croy, S. Fu, N. Ngan Nguyen, K. Liu, H. Qi, S. Chung, M. Jeong, L. Gao, U. Kaiser, K. Cho, H. Wang, R. Dong, G. Cuniberti, M. Bonn, S. Mannsfeld, X. Feng
    University Bibliography Jena:
    fsu_mods_00025285External link

  4. MORE-Q, a dataset for molecular olfactorial receptor engineering by quantum mechanics

    Year of publicationPublished in:Scientific data L. Chen, L. Medrano Sandonas, P. Traber, A. Dianat, N. Tverdokhleb, M. Hurevich, S. Yitzchaik, R. Gutierrez, A. Croy, G. Cuniberti
    We introduce the MORE-Q dataset, a quantum-mechanical (QM) dataset encompassing the structural and electronic data of non-covalent molecular sensors formed by combining 18 mucin-derived olfactorial receptors with 102 body odor volatilome (BOV) molecules. To have a better understanding of their intra- and inter-molecular interactions, we have performed accurate QM calculations in different stages of the sensor design and, accordingly, MORE-Q splits into three subsets: i) MORE-Q-G1: QM data of 18 receptors and 102 BOV molecules, ii) MORE-Q-G2: QM data of 23,838 BOV-receptor configurations, and iii) MORE-Q-G3: QM data of 1,836 BOV-receptor-graphene systems. Each subset involves geometries optimized using GFN2-xTB with D4 dispersion correction and up to 39 physicochemical properties, including global and local properties as well as binding features, all computed at the tightly converged PBE+D3 level of theory. By addressing BOV-receptor-graphene systems from a QM perspective, MORE-Q can serve as a benchmark dataset for state-of-the-art machine learning methods developed to predict binding features. This, in turn, can provide valuable insights for developing the next-generation mucin-derived olfactory receptor sensing devices.
    University Bibliography Jena:
    fsu_mods_00022155External link

  5. Monosaccharide-Derived Enantioselectivity in SWCNT Chemoresistive VOC Sensing

    Year of publicationPublished in:Chemistry: a European Journal A. Shitrit, Y. Sukhran, N. Tverdokhleb, L. Chen, A. Dianat, R. Gutierrez, S. Körbel, A. Croy, G. Cuniberti, M. Hurevich, S. Yitzchaik
    Semiconducting single-walled carbon nanotubes (sc-SWCNTs) are of great potential for vapor sensing. However, sc-SWCNTs lack recognition features for discriminating between sparsely functionalized moieties, molecules with similar structural features, and enantiomer pairs. This becomes a major setback in discriminating between volatile organic compounds (VOCs). Here, we used two galactosides decorated with aromatic groups as a recognition layer in chemoresistive sc-SWCNT sensors to produce chiral preference toward six terpenoid enantiomers. The multichirality and multifunctionality of a monosaccharide scaffold were exploited to maximize the limited interacting features associated with VOCs. The developed system establishes a robust and tunable platform for enantioselective gas sensing.
    University Bibliography Jena:
    fsu_mods_00028980External link

  6. A dataset of laymen olfactory perception for 74 mono-molecular odors

    Year of publicationPublished in:Scientific data A. Bierling, A. Croy, T. Jesgarzewsky, M. Rommel, G. Cuniberti, T. Hummel, I. Croy
    The molecular structure of an odor determines whether and how it is perceived by humans. However, the principles of how odorant chemistry links to perceptual patterns remain largely unknown and are primarily studied using odor rating datasets from highly trained olfactory experts, such as perfumers. This limits our knowledge of typical odor perception and its variability over individuals. We provide a dataset featuring free descriptions, evaluative ratings, and qualitative labels for 74 chemically diverse mono-molecular odorants, rated by a large sample of young adults. A total of 1,227 participants described and rated the odors, and completed questionnaires covering their demographic background, personality traits, and the role of olfaction in their daily lives. The dataset offers a valuable foundation for research aimed at understanding the fundamentals of olfactory perception.
    University Bibliography Jena:
    fsu_mods_00022143External link

  7. A standardized lexicon of body odor words crafted from 17 countries

    Year of publicationPublished in:Scientific data A. Bierling, A. Croy, F. Bilem, L. Bloy, F. Ho, A. Jimenez, P. Kyjaková, M. Mastinu, N. Power Guerra, U. Sailer, A. Schirmer, E. Silva, V. Surakka, L. Takau, E. Thunell, K. Verma, B. Żyżelewicz, A. Majid, I. Croy
    Body odors offer a unique window into the physiological and psychological profile of the emitter. This information, broadcast in nonverbal communication, significantly shapes social interactions. However, effectively digitizing body odors requires a precise framework for perceptual operationalization. Previous research has used a very limited number of verbal terms, such as pleasant, intense, or attractive, which fails to adequately capture qualitative differences. To address this gap, we elicited body odor descriptions from 2,607 participants across 17 countries and 13 languages. All these descriptions are presented here in one dataset, together with a condensed list of 25 body odor words (BOW). Those terms reliably differentiated between body states, and were validated in a separate study with a different group of 155 perceivers. The dataset, available as a web application, provides a novel operationalization of body odor impressions, which is a precondition for studying olfaction in human nonverbal communication, for perception-based digitization of body odors and for comparative studies.
    University Bibliography Jena:
    fsu_mods_00022154External link

  8. Charge carrier mobilities in γ-graphynes: a computational approach

    Year of publicationStatusReview pendingPublished in:Nanoscale / Royal Society of Chemistry ; National Centre for Nanoscience and Technology (NCNST), Beijing E. Unsal, A. Pecchia, A. Croy, G. Cuniberti
    Graphynes, a class of two-dimensional carbon allotropes, exhibit exceptional electronic properties, similar to graphene, but with intrinsic band gaps, making them promising for semiconducting applications. The incorporation of acetylene linkages allows for systematic modulation of their properties. However, the theoretical characterization of graphynes remains computationally demanding, particularly for electron-phonon coupling (EPC) analyses. Here, we employ the density functional tight binding method within the DFTBEPHY framework, providing an efficient and accurate approach for computing EPC and transport properties. We investigate the structural, mechanical, electronic, and transport properties of graphynes, comparing transport calculations using the constant relaxation-time approximation and the self-energy relaxation-time approximation (SERTA) alongside analytical models based on parabolic- and Kane-band approximations. For graphyne, the SERTA relaxation time is 0.63 (1.69) ps for holes (electrons). In graphdiyne, the relaxation time is 0.04 (0.14) ps for holes (electrons). While the hole mobilities in graphyne are on the order of 103 cm2 V-1 s-1, the electron mobilities reach up to 104 cm2 V-1 s-1. In graphdiyne, the mobility values for both types of charge carriers are on the order of 102 cm2 V-1 s-1. The phonon-limited mobilities at room temperature in graphyne fall between those of graphene and MoS2, while in graphdiyne, they are comparable to those of MoS2.
    University Bibliography Jena:
    fsu_mods_00028741External link

  9. From Local Atomic Environments to Molecular Information Entropy

    Year of publicationPublished in:ACS Omega A. Croy
    The similarity of local atomic environments is an important concept in many machine learning techniques, which find applications in computational chemistry and material science. Here, we present and discuss a connection between the information entropy and the similarity matrix of a molecule. The resulting entropy can be used as a measure of the complexity of a molecule. Exemplarily, we introduce and evaluate two specific choices for defining the similarity: one is based on a SMILES representation of local substructures, and the other is based on the SOAP kernel. By tuning the sensitivity of the latter, we can achieve good agreement between the respective entropies. Finally, we consider the entropy of two molecules in a mixture. The gain of entropy due to the mixing can be used as a similarity measure of the molecules. We compare this measure to the average and best-match kernel. The results indicate a connection between the different approaches and demonstrate the usefulness and broad applicability of the similarity-based entropy approach.
    University Bibliography Jena:
    fsu_mods_00012828External link

  11. Tailoring phosphine ligands for improved C-H activation: insights from Δ-machine learning

    Year of publicationPublished in:Digital Discovery: a journal for new thinking on machine learning, robotics and AI T. Huang, R. Geitner, A. Croy, S. Gräfe
    Transition metal complexes have played crucial roles in various homogeneous catalytic processes due to their exceptional versatility. This adaptability stems not only from the central metal ions but also from the vast array of choices of the ligand spheres, which form an enormously large chemical space. For example, Rh complexes, with a well-designed ligand sphere, are known to be efficient in catalyzing the C-H activation process in alkanes. To investigate the structure-property relation of the Rh complex and identify the optimal ligand that minimizes the calculated reaction energy ΔE of an alkane C-H activation, we have applied a Δ-machine learning method trained on various features to study 1743 pairs of reactants (Rh(PLP)(Cl)(CO)) and intermediates (Rh(PLP)(Cl)(CO)(H)(propyl)). Our findings demonstrate that the models exhibit robust predictive performance when trained on features derived from electron density (R² = 0.816), and SOAPs (R² = 0.819), a set of position-based descriptors. Leveraging the model trained on xTB-SOAPs that only depend on the xTB-equilibrium structures, we propose an efficient and accurate screening procedure to explore the extensive chemical space of bisphosphine ligands. By applying this screening procedure, we identify ten newly selected reactant-intermediate pairs with an average ΔE of 33.2 kJ mol−¹, remarkably lower than the average ΔE of the original data set of 68.0 kJ mol−¹. This underscores the efficacy of our screening procedure in pinpointing structures with significantly lower energy levels.
    University Bibliography Jena:
    fsu_mods_00013567External link

  12. Site-selective chemical reactions by on-water surface sequential assembly

    Year of publicationPublished in:Nature Communications A. Prasoon, X. Yu, M. Hambsch, D. Bodesheim, K. Liu, A. Zacarias, N. Nguyen, T. Seki, A. Dianat, A. Croy, G. Cuniberti, P. Fontaine, Y. Nagata, S. Mannsfeld, R. Dong, M. Bonn, X. Feng
    Controlling site-selectivity and reactivity in chemical reactions continues to be a key challenge in modern synthetic chemistry. Here, we demonstrate the discovery of site-selective chemical reactions on the water surface via a sequential assembly approach. A negatively charged surfactant monolayer on the water surface guides the electrostatically driven, epitaxial, and aligned assembly of reagent amino-substituted porphyrin molecules, resulting in a well-defined J-aggregated structure. This constrained geometry of the porphyrin molecules prompts the subsequent directional alignment of the perylenetetracarboxylic dianhydride reagent, enabling the selective formation of a one-sided imide bond between porphyrin and reagent. Surface-specific in-situ spectroscopies reveal the underlying mechanism of the dynamic interface that promotes multilayer growth of the site-selective imide product. The site-selective reaction on the water surface is further demonstrated by three reversible and irreversible chemical reactions, such as imide-, imine-, and 1, 3-diazole (imidazole)- bonds involving porphyrin molecules. This unique sequential assembly approach enables site-selective chemical reactions that can bring on-water surface synthesis to the forefront of modern organic chemistry.
    University Bibliography Jena:
    fsu_mods_00009561External link
Pagination Page 1