February 2, 2012
Computation of intermolecular interaction energies via Kohn-Sham density functional theory (KS-DFT) within the local density (LDA) or generalized gradient (GGA) approximations is hindered by their inaccurate inclusion of medium and long range dispersion interactions. Computation of inter- and intra-macrobiomolecule interaction energies, in particular, requires a fairly accurate yet not overly expensive methodology. Dr. Christos Deligkaris, as part of his PhD research, and Prof. Jorge H. Rodriguez implemented a method to compute intermolecular interaction energies by including an empirical correction for dispersion which is explicitly valid over a range of intermolecular distances. The method was designed to predict interaction energies with an accuracy consistent with distance-dependent (DD) reference energies obtained from coupled cluster, CCSD(T), ab-initio calculations. The resulting methodology, which we named B3LYP-DD, yields interaction energies with an accuracy generally better than 1 kcal mol-1 for different types of noncovalent complexes, over a range of intermolecular distances and interaction strengths, relative to the expensive CCSD(T)/CBS standard. For a training set of dispersion interacting complexes, B3LYP-DD interaction energies in combination with diffuse functions display absolute errors equal to or smaller than 0.68 kcal mol-1. The details of the B3LYP-DD methodology are given in ``Physical Chemistry Chemical Physics" published by the Royal Society of Chemistry (RSC). [Christos Deligkaris and Jorge H. Rodriguez, PCCP (2012), DOI: 10.1039/C2CP23673G] The method potentially allows accurate computation of energies and geometric structures of fairly large biomolecular systems such as protein-ligand interactions.
November 21, 2011
Prof. Jorge H. Rodriguez and his team use methods of computational quantum mechanics to investigate the biochemical function and structure of metal containing enzymes. Metalloenzymes catalyze many important biochemical reactions. Rodriguez and his team realize that metal-mediated biochemical reactions are related to the interaction between valence electrons of the reactant species. Such valence electrons are microscopic particles properly described by quantum mechanics. Taking advantage of powerful parallel-processing supercomputers, Rodriguez and his group solve the fundamental equations of quantum mechanics to elucidate physico-chemical properties and reactivities of metal centers in enzymes. The quantum effects are incorporated via Kohn-Sham spin density functional theory (SDFT) and, when appropriate, its relativistic extensions.
Rodriguez and his group pioneer the emerging field of "Quantum Biochemistry". As an example, Rodriguez and his team have predicted the electronic and geometric structures of a key peroxo reaction intermediate in the catalytic cycle of the enzyme methane monooxygenase hydroxylase (MMOH). This enzyme catalyzes an important reaction, namely the conversion of methane (CH4) to methanol (CH3OH). During such catalytic process, the di-iron center of MMOH must pass through various key stages known as reaction intermediates. Often, there are serious difficulties that prevent obtaining structures of reaction intermediates from experiment. However, it may be possible to obtain other types of experimental information based on spectroscopic techniques such as 57Fe Mössbauer spectroscopy. By using algorithms developed in his group, able to accurately predict 57Fe Mössbauer spectral parameters, Rodriguez and his team have been able to predict the geometric structure of a key reaction intermediate of MMOH for which there is no experimental crystallographic structure. In an article published in Dalton Transactions [Dalton Trans., 2012, DOI: 10.1039/C1DT11656H], Rodriguez and his former graduate student Dr. Teepanis Chachiyo present the predicted structural and electronic details of a key peroxo intermediate of MMOH. Rodriguez and Chachiyo started this project several years ago. The main structural results on the MMOH intermediate and its consistency with 57Fe Mössbauer spectral parameters were reported in Chachiyo's PhD thesis in 2005. Later, to further support their findings, the coauthors computed optical (UV-vis) parameters of the same intermediate and Raman parameters of other structuraly-related intermediates since, currently, there are no reliable Raman parameters for MMOH peroxo. In their Dalton publication, the coauthors show that the predicted structure is consistent with several experimental parameters. In particular, the predicted structure is fully consistent with experimental 57Fe Mössbauer spectral parameters. Mössbauer isomer shifts and quadrupole splittings computed with the algorithm developed at Purdue are, within experimental error, in excellent agreement with experiment.
July 28, 2009
Our research group has implemented an accurate computational methodology for predicting the effects of spin-orbit coupling (SOC), a relativistic effect, on physico-chemical properties of metallo-proteins and (bio)inorganic complexes. In particular, a phenomenon known in the field of molecular magnetism as "zero field splitting" (ZFS), which removes the ground state spin-degeneracy of a system, has been predicted with very good accuracy. Aquino and Rodriguez published a paper [J. Phys. Chem. A, Articles ASAP, July 2009] describing a methodology, based on the combination of spin density functional theory and perturbation theory (SDFT-PT), which is capable of predicting the ZFS of transition metal-containing complexes of relevance in metallo-enzyme biochemistry, inorganic/bioinorganic chemistry, and molecular magnet-based nanotechnology. Their latest paper follows a previous paper on the subject [J. Chem. Phys., Vol. 123, 204902, 2005].
May 17, 2006
Professor Jorge Rodriguez has been interviewed for a special "career development" article sponsored by Science Magazine. In a May 17, 2006 special supplement titled "Chemistry: First Principles-Calculations" the author Mike May features an interview with Prof. Rodriguez and three other experts about new career opportunities for scientists working on first-principles and quantum chemical computational methods. The full article can be found at:
September 8, 2005
Spin is a quantum property of electrons. Chemists and physicists have long known that many iron-containing molecules often have a "net spin" associated with their outermost electrons. Most often, iron complexes that are excited with light will make a transition from a low energy electronic configuration to a higher energy configuration but will eventually return to their lowest energy state. Chemists and physicists call this lowest energy state the "molecular ground state". The Rodriguez group has been using quantum physics, quantum chemistry, and supercomputers to study a remarkable class of iron complexes which, upon excitation with light, do not return to their spin ground state.
Our research group has developed techniques of computational quantum mechanics to study iron-containing "spin-photoswitches". Contrary to common iron complexes, spin photoswitches can be excited with light but, after making a transition to a higher spin and higher energy state, these do not return to their original ground state. Instead, these remarkable molecular crystals remain in a metastable high spin state as long as their temperature is below some 50 Kelvin. This phenomenom, which has been previously discovered by experimentalists, is called it light-induced excited-state spin trapping "LIESST".
Rodriguez used time-dependent density functional theory in conjunction with supercomputers to elucidate the initial excitations that take place at the onset of the LIESST process [J. Chem. Phys., Vol. 123, 094709, 2005]. In addition, Chachiyo and Rodriguez published a paper [J. Chem. Phys., Vol. 123, 094711, 2005] that describes a computational algorithm to follow the structural rearrangements and changes in spin state during the LIESST process. The computational studies show that there is a sequence of spin states which the photoswitches can go through, forming a pathway which takes them from the ground state to the metastable high spin state. This is a novel description of the mechanism by which iron spin photoswitches work. These theoretical results will likely aid to the development of better quality molecular photoswitches which, by remaining trapped in a metastable spin state at high temperatures, may be used in molecular-level memory storage of liquid display technological applications. Molecular-level memory storage is seen as a powerful futuristic way to store huge amounts of information in very small (nanoscale) regions of space. The work done by the Rodriguez group is an example of how computational quantum physics and quantum inorganic chemistry can contribute to the development of novel nanoscale technological devices.
The research was supported in part by a NSF CAREER award CHE-0349189 (JHR).
FIGURE CAPTION: The iron complex shown in the figure is a spin photoswitch that, upon excitation with green light, can change its spin from S=0 to S=2. The Rodriguez group has used supercomputers, quantum physics, and quantum chemistry to discover novel mechanisms about the spin-photoswitching process.
January 15, 2004
The National Science Foundation (NSF) has granted a CAREER award to Jorge Rodríguez, Assistant Professor of Physics at Purdue. This award intends to foster Prof. Rodríguez's theoretical and computational research on biomolecular and mesoscopic physics and, in particular, on quantum models of the electronic structure and biological function of magnetically ordered metalloproteins. This award also intends to contribute to Prof. Rodríguez's various educational efforts. According to NSF, the CAREER is its "most prestigious award for new faculty members. The CAREER program recognizes and supports the early career-development activities of those teacher-scholars who are most likely to become the academic leaders of the 21st century." The award carries with it funding for five consecutive years.
September 14, 2002
First principle calculations based on density functional theory(DFT), in conjunction with high performance supercomputers, were carried out to obtain these results. Our computational method is potentially a powerful and novel alternative to susceptibility or other experimental methods which have been traditionally used to determine magnetic parameters. We have reported these results in an extensive publication [J. Chem. Phys. 116, 6253-6270, 2002].