Ulrich Fekl
Research Subjects
Ulrich Fekl has a total of 80 co-authors. Here are the top co-authors by number of shared publications:
Alan J. Lough
Department of Chemistry University of Toronto 80 St. George St. Toronto Ontario Canada M5S 3H6
19 shared publicationsDaniel J. Harrison
Department of Chemical and Physical Sciences, University of Toronto Mississauga, 3359 Mississauga Rd, Mississauga, Ontario, L5L 1C6, Canada
8 shared publications- David Armstrong
University of Toronto, Mississauga Campus, 3359 Mississauga Road, Mississauga, Ontario, Canada L5L 1C6.
6 shared publications Neilson Nguyen
Department of Chemical and Physical Sciences, University of Toronto Mississauga, 3359 Mississauga Rd, Mississauga, Ontario, L5L 1C6, Canada
6 shared publications- Alan J. Lough
University of Toronto
5 shared publications Antonio G. De Crisci
University of Toronto, 3359 Mississauga Road N, Mississauga, Ontario L5L 1C6, Canada, Fax: +1-905-828-5425
5 shared publications
Latest Publications
- 12 March 2026
- Published by American Chemical Society (ACS) in Journal of Chemical Information and Modeling
- Vol. 66 (6) , 3103-3116
Parametrized Quantum Circuit Learning for Quantum Chemical Applications
In the field of quantum machine learning (QML), parametrized quantum circuits (PQCs)─constructed using a combination of fixed and tunable quantum gates─provide a promising hybrid framework for tackling complex machine learning problems. Despite numerous proposed applications, there remains limited exploration of data sets relevant to quantum chemistry. In this study, we investigate the potential benefits and limitations of PQCs on two chemically meaningful data sets: (1) the BSE49 data set, containing bond separation energies for 49 different classes of chemical bonds, and (2) a data set of water conformers, where coupled-cluster singles and doubles (CCSD) wave functions are predicted from lower-level electronic structure methods using the data-driven coupled-cluster (DDCC) approach. We construct a comprehensive set of 168 PQCs by combining 14 data encoding strategies with 12 variational ansätze, and evaluate their performance on circuits with 5 and 16 qubits. Our initial analysis examines the impact of circuit structure on model performance using state-vector simulations. We then explore how circuit depth and training set size influence model performance. Finally, we assess the performance of the best-performing PQCs on current quantum hardware, using both noisy simulations ("fake" backends) and real quantum devices. Our findings underscore the challenges of applying PQCs to chemically relevant problems that are straightforward for classical machine learning methods but remain nontrivial for quantum approaches.
- 12 March 2026
- Published by American Chemical Society (ACS) in Journal of Chemical Information and Modeling
- Vol. 66 (6) , 3103-3116
Parametrized Quantum Circuit Learning for Quantum Chemical Applications
In the field of quantum machine learning (QML), parametrized quantum circuits (PQCs)─constructed using a combination of fixed and tunable quantum gates─provide a promising hybrid framework for tackling complex machine learning problems. Despite numerous proposed applications, there remains limited exploration of data sets relevant to quantum chemistry. In this study, we investigate the potential benefits and limitations of PQCs on two chemically meaningful data sets: (1) the BSE49 data set, containing bond separation energies for 49 different classes of chemical bonds, and (2) a data set of water conformers, where coupled-cluster singles and doubles (CCSD) wave functions are predicted from lower-level electronic structure methods using the data-driven coupled-cluster (DDCC) approach. We construct a comprehensive set of 168 PQCs by combining 14 data encoding strategies with 12 variational ansätze, and evaluate their performance on circuits with 5 and 16 qubits. Our initial analysis examines the impact of circuit structure on model performance using state-vector simulations. We then explore how circuit depth and training set size influence model performance. Finally, we assess the performance of the best-performing PQCs on current quantum hardware, using both noisy simulations ("fake" backends) and real quantum devices. Our findings underscore the challenges of applying PQCs to chemically relevant problems that are straightforward for classical machine learning methods but remain nontrivial for quantum approaches.
- 17 December 2025
- Published by American Chemical Society (ACS) in Inorganic Chemistry
- Vol. 65 (1) , 776-790
Ligand Substitution in Ni(CO) 4 ─Computational Discovery of a Long-Suspected “Dark Pathway” and a Detailed Analysis of the Intrinsic Reaction Coordinate
The reaction mechanism for carbon monoxide self-exchange in nickel tetracarbonyl and the substitution of CO by various incoming ligands was investigated in depth using DFT calculations. Early experimental work from the 1960s suggested that ligand substitution in Ni(CO)4 proceeds via a unimolecular dissociative (D) mechanism. Although a bimolecular interchange (I-type) pathway had been proposed, definitive evidence for such a mechanism is still lacking. Our calculations now provide convincing support for an interchange route: all transition states (TSs) for the one-step interchange process were located and validated by vibrational analysis. The ligands studied─hydrogen cyanide, pyridine, trimethylamine, trimethylphosphine, triphenylphosphane, and trimethylarsine─show reaction barriers below 20 kcal/mol, largely independent of the functional used (B3LYP, B3PW91, or ωB97XD with 6-311+G(d,p)). Despite the bimolecular nature of the reaction and the diversity of donor atoms, the activation barriers proved surprisingly insensitive to ligand identity. To uncover more subtle ligand-dependent effects, additional B3LYP/6-311+G(d,p) calculations tracked topological and conceptual DFT indices along the full-reaction coordinates; for pyridine, NICS aromaticity indices were also followed. These analyses reveal nuanced differences among the ligands and suggest that the interchange pathway, though hidden under typical conditions, could become dominant with appropriate ligands and high concentrations.
- 17 December 2025
- Published by American Chemical Society (ACS) in Inorganic Chemistry
- Vol. 65 (1) , 776-790
Ligand Substitution in Ni(CO) 4 ─Computational Discovery of a Long-Suspected “Dark Pathway” and a Detailed Analysis of the Intrinsic Reaction Coordinate
The reaction mechanism for carbon monoxide self-exchange in nickel tetracarbonyl and the substitution of CO by various incoming ligands was investigated in depth using DFT calculations. Early experimental work from the 1960s suggested that ligand substitution in Ni(CO)4 proceeds via a unimolecular dissociative (D) mechanism. Although a bimolecular interchange (I-type) pathway had been proposed, definitive evidence for such a mechanism is still lacking. Our calculations now provide convincing support for an interchange route: all transition states (TSs) for the one-step interchange process were located and validated by vibrational analysis. The ligands studied─hydrogen cyanide, pyridine, trimethylamine, trimethylphosphine, triphenylphosphane, and trimethylarsine─show reaction barriers below 20 kcal/mol, largely independent of the functional used (B3LYP, B3PW91, or ωB97XD with 6-311+G(d,p)). Despite the bimolecular nature of the reaction and the diversity of donor atoms, the activation barriers proved surprisingly insensitive to ligand identity. To uncover more subtle ligand-dependent effects, additional B3LYP/6-311+G(d,p) calculations tracked topological and conceptual DFT indices along the full-reaction coordinates; for pyridine, NICS aromaticity indices were also followed. These analyses reveal nuanced differences among the ligands and suggest that the interchange pathway, though hidden under typical conditions, could become dominant with appropriate ligands and high concentrations.
- Open Access
- 18 September 2024
- Published by Royal Society of Chemistry (RSC) in Physical Chemistry Chemical Physics
- Vol. 26 (39) , 25412-25417
Heteroatom-vacancy centres in molecular nanodiamonds: a computational study of organic molecules possessing triplet ground states through σ-overlap
Can doped nanodiamonds with a central vacancy be small-molecule-sized and still retain a desirable triplet ground state? Computational results are encouraging for doped and hollow decamantane derivatives of C3v symmetry.
- 18 September 2024
- Published by Royal Society of Chemistry (RSC) in Physical Chemistry Chemical Physics
- Vol. 26 (39) , 25412-25417
Open AccessHeteroatom-vacancy centres in molecular nanodiamonds: a computational study of organic molecules possessing triplet ground states through σ-overlap
Can doped nanodiamonds with a central vacancy be small-molecule-sized and still retain a desirable triplet ground state? Computational results are encouraging for doped and hollow decamantane derivatives of C3v symmetry.
- Open Access
- 8 September 2024
- Published by American Chemical Society (ACS) in ChemRxiv
Comment on "Structure of a novel form of carbon: dehydropolycondensed adamantane?" by J. S. Rigden, K. J. Koivusaari, R. J. Newport, D. A. Green, G. Bushnell-Wye and J. Tomkinson, J. Mater. Chem. 1996, 6, 449.
This comment addresses the identification of the crystalline phase observed after Wurtz-Fittig coupling (using sodium) and heat treatment of 1,3,5,7-tetrabromoadamantane in J. S. Rigden, K. J. Koivusaari, Robert J. Newport, D. A. Green, G. Bushnell-Wye and J. Tomkinson, J. Mater. Chem. 1996, 6, 449, where a preparation intended to yield "dehydropolycondensed adamantane" (V. I. Kasatochkin, Y. P Kudryavtsev, V. M. Elizen, O. I. Egorova, A. M. Sladkov and V. V. Korshak, Dokl. Akad. Nauk. USSR 1976, 231, 1358) was re-investigated. Using powder X-ray diffraction (P-XRD), the crystalline phase in question was assigned by Ridgen et al. in 1996 to be a carbon-based "graphitic" material. We show here that this assignment is incorrect. A direct comparison of the P-XRD pattern reported in 1996 with the P-XRD pattern generated from the sodium bromide structure reveals that the crystalline phase observed is in fact sodium bromide (NaBr).
- 8 September 2024
- Published by American Chemical Society (ACS) in ChemRxiv
Open AccessComment on "Structure of a novel form of carbon: dehydropolycondensed adamantane?" by J. S. Rigden, K. J. Koivusaari, R. J. Newport, D. A. Green, G. Bushnell-Wye and J. Tomkinson, J. Mater. Chem. 1996, 6, 449.
This comment addresses the identification of the crystalline phase observed after Wurtz-Fittig coupling (using sodium) and heat treatment of 1,3,5,7-tetrabromoadamantane in J. S. Rigden, K. J. Koivusaari, Robert J. Newport, D. A. Green, G. Bushnell-Wye and J. Tomkinson, J. Mater. Chem. 1996, 6, 449, where a preparation intended to yield "dehydropolycondensed adamantane" (V. I. Kasatochkin, Y. P Kudryavtsev, V. M. Elizen, O. I. Egorova, A. M. Sladkov and V. V. Korshak, Dokl. Akad. Nauk. USSR 1976, 231, 1358) was re-investigated. Using powder X-ray diffraction (P-XRD), the crystalline phase in question was assigned by Ridgen et al. in 1996 to be a carbon-based "graphitic" material. We show here that this assignment is incorrect. A direct comparison of the P-XRD pattern reported in 1996 with the P-XRD pattern generated from the sodium bromide structure reveals that the crystalline phase observed is in fact sodium bromide (NaBr).
- Open Access
- 26 January 2024
- Published by Wiley in Chemistry – A European Journal
- Vol. 30 (14) , e202302998
Organic Polar Crystals, Second Harmonic Generation, and Piezoelectric Effects from Heteroadamantanes in the Space Group R3m
Polar crystalline materials, a subset of the non‐centrosymmetric materials, are highly sought after. Their symmetry properties make them pyroelectric and also piezoelectric and capable of second‐harmonic generation (SHG). For SHG and piezoelectric applications, metal oxides are commonly used. The advantages of oxides are durability and hardness – downsides are the need for high‐temperature synthesis/processing and often the need to include toxic metals. Organic polar crystals, on the other hand, can avoid toxic metals and can be amenable to solution‐state processing. While the vast majority of polar organic molecules crystallize in non‐polar space groups, we found that both 7‐chloro‐1,3,5‐triazaadamantane, for short Cl‐TAA, and also the related Br‐TAA (but not I‐TAA) form polar crystals in the space group R3m, easily obtained from dichloromethane solution. Measurements confirm piezoelectric and SHG properties for Cl‐TAA and Br‐TAA. When the two species are crystallized together, solid solutions form, suggesting that properties of future materials can be tuned continuously.
- 26 January 2024
- Published by Wiley in Chemistry – A European Journal
- Vol. 30 (14) , e202302998
Open AccessOrganic Polar Crystals, Second Harmonic Generation, and Piezoelectric Effects from Heteroadamantanes in the Space Group R3m
Polar crystalline materials, a subset of the non‐centrosymmetric materials, are highly sought after. Their symmetry properties make them pyroelectric and also piezoelectric and capable of second‐harmonic generation (SHG). For SHG and piezoelectric applications, metal oxides are commonly used. The advantages of oxides are durability and hardness – downsides are the need for high‐temperature synthesis/processing and often the need to include toxic metals. Organic polar crystals, on the other hand, can avoid toxic metals and can be amenable to solution‐state processing. While the vast majority of polar organic molecules crystallize in non‐polar space groups, we found that both 7‐chloro‐1,3,5‐triazaadamantane, for short Cl‐TAA, and also the related Br‐TAA (but not I‐TAA) form polar crystals in the space group R3m, easily obtained from dichloromethane solution. Measurements confirm piezoelectric and SHG properties for Cl‐TAA and Br‐TAA. When the two species are crystallized together, solid solutions form, suggesting that properties of future materials can be tuned continuously.
- 30 December 2023
- Published by Royal Society of Chemistry (RSC) in Physical Chemistry Chemical Physics
- Vol. 26 (5) , 4071-4082
Applications of noisy quantum computing and quantum error mitigation to “adamantaneland”: a benchmarking study for quantum chemistry
Modeling isomers of C10H16 and carbocationic rearrangements of C10H15+ using the variational quantum eigensolver algorithm and quantum state tomography correction.
- 30 December 2023
- Published by Royal Society of Chemistry (RSC) in Physical Chemistry Chemical Physics
- Vol. 26 (5) , 4071-4082
Applications of noisy quantum computing and quantum error mitigation to “adamantaneland”: a benchmarking study for quantum chemistry
Modeling isomers of C10H16 and carbocationic rearrangements of C10H15+ using the variational quantum eigensolver algorithm and quantum state tomography correction.
- 9 September 2021
- Published by American Chemical Society (ACS) in Journal of Chemical Education
- Vol. 98 (10) , 3189-3206
Covalent d-Block Organometallics: Teaching Lewis Structures and sd/sd2Hybridization Gives Students Additional Explanations and Powerful Predictive Tools
- 9 September 2021
- Published by American Chemical Society (ACS) in Journal of Chemical Education
- Vol. 98 (10) , 3189-3206
Covalent d-Block Organometallics: Teaching Lewis Structures and sd/sd2Hybridization Gives Students Additional Explanations and Powerful Predictive Tools
- Open Access
- 17 March 2021
- Published by American Chemical Society (ACS) in ChemRxiv
Covalent d-Block Organometallics: Teaching Lewis Structures and sd/sd2 Hybridization Gives Students Additional Explanations and Powerful Predictive Tools
Despite tremendous efforts by instructors and textbook authors, students find it difficult to develop useful chemical intuitions about preferred structures, structural trends, and properties of even the most common d-block element organometallic species, that is d6, d8, and d10 systems. A full molecular orbital analysis of a transition metal species is not always feasible or desirable, and crystal field theory, while generally useful, is often too simplistic and limited. It would be helpful to give students of organometallic chemistry an additional toolkit that helps them to understand d-block compounds, in particular highly covalent ones. It is well known in the research literature in organometallic chemistry that hybridization arguments involving s and d orbitals (such as sd and sd2 hybridization for d8 and d6 systems, respectively) provides useful insight. However, this knowledge is much underused in undergraduate teaching and not taught in undergraduate textbooks. The purpose of this article is to make descriptions of bonding that are based on s,d-hybridized orbitals more accessible in a way that is directly useful for undergraduate teaching. Geometries of unusual low-coordinate structures can be successfully predicted. An in-depth physical explanation for the trans-influence, the weakening of a bond due to a strong bond trans to it, is provided. A clear explanation is given for why the cis isomer normally more stable than the trans isomer in square-planar d8 complexes of the type MR2L2 (R = alkyl/aryl, L = relatively weakly bonded neutral ligand). Similarly, the relative stability of fac versus mer isomers in octahedral d6 complexes of the type MR3L3 is explained. Relevant to catalysis, the method explains why strongly donating ligands do not always facilitate oxidative addition and why 12-electron and 14-electron Pd(0) species are thermodynamically much more accessible than one might expect. The method capitalizes on 1st year knowledge such as the ability to write Lewis structures and to use hybridization arguments. It also ties into the upper-year experience, including graduate school, where covalent d-block complexes may be encountered in research and where the hybridization schemes described here naturally emerge from using the NBO formalism. It is discussed where the method might fit into the inorganic curriculum.
- 17 March 2021
- Published by American Chemical Society (ACS) in ChemRxiv
Open AccessCovalent d-Block Organometallics: Teaching Lewis Structures and sd/sd2 Hybridization Gives Students Additional Explanations and Powerful Predictive Tools
Despite tremendous efforts by instructors and textbook authors, students find it difficult to develop useful chemical intuitions about preferred structures, structural trends, and properties of even the most common d-block element organometallic species, that is d6, d8, and d10 systems. A full molecular orbital analysis of a transition metal species is not always feasible or desirable, and crystal field theory, while generally useful, is often too simplistic and limited. It would be helpful to give students of organometallic chemistry an additional toolkit that helps them to understand d-block compounds, in particular highly covalent ones. It is well known in the research literature in organometallic chemistry that hybridization arguments involving s and d orbitals (such as sd and sd2 hybridization for d8 and d6 systems, respectively) provides useful insight. However, this knowledge is much underused in undergraduate teaching and not taught in undergraduate textbooks. The purpose of this article is to make descriptions of bonding that are based on s,d-hybridized orbitals more accessible in a way that is directly useful for undergraduate teaching. Geometries of unusual low-coordinate structures can be successfully predicted. An in-depth physical explanation for the trans-influence, the weakening of a bond due to a strong bond trans to it, is provided. A clear explanation is given for why the cis isomer normally more stable than the trans isomer in square-planar d8 complexes of the type MR2L2 (R = alkyl/aryl, L = relatively weakly bonded neutral ligand). Similarly, the relative stability of fac versus mer isomers in octahedral d6 complexes of the type MR3L3 is explained. Relevant to catalysis, the method explains why strongly donating ligands do not always facilitate oxidative addition and why 12-electron and 14-electron Pd(0) species are thermodynamically much more accessible than one might expect. The method capitalizes on 1st year knowledge such as the ability to write Lewis structures and to use hybridization arguments. It also ties into the upper-year experience, including graduate school, where covalent d-block complexes may be encountered in research and where the hybridization schemes described here naturally emerge from using the NBO formalism. It is discussed where the method might fit into the inorganic curriculum.
- 22 February 2021
- Published by Wiley in International Journal of Quantum Chemistry
- Vol. 121 (11) , e26638
Group‐IV‐based selective C-H bond activation of a diamondoid—A density functional theory study
Selective synthetic modifications on diamondoids (well‐defined molecular nano‐diamonds) are difficult. Group IV organometallic complexes are promising systems for CH bond activation. We study the CH bond activation of the hard‐to‐activate CH2 positions of the adamantyl group (the smallest diamondoid) using density functional theory. As a platform for activation, we tether the adamantyl group to the cyclopentadienyl in a substituted bis‐cyclopentadienyl group IV metal diphenyl complex. The mechanism proposed in the experimental paper reporting the activation of C(Me)2‐Ph or t‐Bu groups using Zr(IV) involves an η2‐benzyne complex intermediate. Our computational work confirms the two‐step mechanism proposed for activation of C(Me)2‐Ph or t‐Bu using Zr(IV) and further extends it to Ti(IV) and Hf(IV). The first step forms the benzyne complex. In the second step, the cyclopentadienyl‐bound group is activated through a mechanism that closely resembles a σ‐bond metathesis when the benzyne complex is described as a metallacyclopropene We demonstrate through computations that the two‐step reaction can be successfully applied to the adamantyl group. CH bond activation of adamantyl is found to be thermodynamically and kinetically feasible. The important question of regioselectivity is addressed. We predict that α‐CH bond activation will be achievable with Ti (thermodynamic control), and selective γ‐CH bond activation with Zr (kinetic or thermodynamic control).
- 22 February 2021
- Published by Wiley in International Journal of Quantum Chemistry
- Vol. 121 (11) , e26638
Group‐IV‐based selective C-H bond activation of a diamondoid—A density functional theory study
Selective synthetic modifications on diamondoids (well‐defined molecular nano‐diamonds) are difficult. Group IV organometallic complexes are promising systems for CH bond activation. We study the CH bond activation of the hard‐to‐activate CH2 positions of the adamantyl group (the smallest diamondoid) using density functional theory. As a platform for activation, we tether the adamantyl group to the cyclopentadienyl in a substituted bis‐cyclopentadienyl group IV metal diphenyl complex. The mechanism proposed in the experimental paper reporting the activation of C(Me)2‐Ph or t‐Bu groups using Zr(IV) involves an η2‐benzyne complex intermediate. Our computational work confirms the two‐step mechanism proposed for activation of C(Me)2‐Ph or t‐Bu using Zr(IV) and further extends it to Ti(IV) and Hf(IV). The first step forms the benzyne complex. In the second step, the cyclopentadienyl‐bound group is activated through a mechanism that closely resembles a σ‐bond metathesis when the benzyne complex is described as a metallacyclopropene We demonstrate through computations that the two‐step reaction can be successfully applied to the adamantyl group. CH bond activation of adamantyl is found to be thermodynamically and kinetically feasible. The important question of regioselectivity is addressed. We predict that α‐CH bond activation will be achievable with Ti (thermodynamic control), and selective γ‐CH bond activation with Zr (kinetic or thermodynamic control).
- 1 February 2021
- Published by Canadian Science Publishing in Canadian Journal of Chemistry
- Vol. 99 (2) , 154-160
Platinum(IV) complexes of a bulky and exceptionally donating alkyl: 2-adamantyl
The 2-adamantyl (2-Ad) group is a particularly challenging alkyl to install on a transition metal, with the only reported 2-Ad complexes of platinum being those of Pt(II). The question of whether 2-Ad complexes with platinum in the oxidation state +IV can be made and whether they are stable is answered here. The Pt(IV) compound trans-[(bpy)PtMe2(2-Ad)I] (bpy = 2,2′-bipyridine) was synthesized via oxidative addition of iodomethane to [(COD)PtMe(2-Ad)] (COD = 1,5-cyclooctadiene) in the presence of bpy. Iodide abstraction with silver triflate (triflate anion, OTf− = CF3SO3−) produced trans-[(bpy)PtMe2(2-Ad)(OTf)]. Both complexes were unambiguously characterized by 1D NMR (1H, 13C) spectroscopy and X-ray crystallography. Both show evidence for the large steric bulk of the 2-Ad group (through the conformation of the complex adopted) and also for its exceptionally pronounced trans-influence (very long bonds to iodide/triflate trans to adamantyl). Both of the new platinum(IV) adamantyl complexes were extremely stable, even to prolonged heating.
- 1 February 2021
- Published by Canadian Science Publishing in Canadian Journal of Chemistry
- Vol. 99 (2) , 154-160
Platinum(IV) complexes of a bulky and exceptionally donating alkyl: 2-adamantyl
The 2-adamantyl (2-Ad) group is a particularly challenging alkyl to install on a transition metal, with the only reported 2-Ad complexes of platinum being those of Pt(II). The question of whether 2-Ad complexes with platinum in the oxidation state +IV can be made and whether they are stable is answered here. The Pt(IV) compound trans-[(bpy)PtMe2(2-Ad)I] (bpy = 2,2′-bipyridine) was synthesized via oxidative addition of iodomethane to [(COD)PtMe(2-Ad)] (COD = 1,5-cyclooctadiene) in the presence of bpy. Iodide abstraction with silver triflate (triflate anion, OTf− = CF3SO3−) produced trans-[(bpy)PtMe2(2-Ad)(OTf)]. Both complexes were unambiguously characterized by 1D NMR (1H, 13C) spectroscopy and X-ray crystallography. Both show evidence for the large steric bulk of the 2-Ad group (through the conformation of the complex adopted) and also for its exceptionally pronounced trans-influence (very long bonds to iodide/triflate trans to adamantyl). Both of the new platinum(IV) adamantyl complexes were extremely stable, even to prolonged heating.