Using TD-DFT to rationalize the optical spectra of organic compounds

Denis Jacquemin

CEISAM, UMR CNRS 6230, 2, rue de la Houssinière, Université de Nantes, France

Video Recording

Video is available only for registered users.

Abstract

During this lecture, I will illustrate the successes and failures of Time-Dependent Density Functional Theory (TD-DFT) in simulating the properties of electronically excited-states, with a specific interest on conjugated organic structures of intertest for dye chemistry. This will start by general explanations on how to perform and analyse TD-DFT calculations for non-experienced users [1-3]. Various advices will be given regarding the selection of basis sets, exchange-correlation functionals and models for modelling solvent effets. The pros and cons of the method will be illustrated. Next, the typical errors of TD-DFT for both 0-0 energies and band shapes will be discussed [4,5]. Single-reference methods representing alternatives to TD-DFT will be briefly discussed in this context, e.g., ADC(2), CC2 and BSE/GW [6]. Finally, I will present applications dedicated to some challenging applications, such as, e.g., cyanine derivatives [7] and excited-state intramolecular proton transfer [8].

Molecular Magnetic Compounds: Calculating Magnetic Couplings Using the DFT Method

Henrique C. S. Junior

Universidade Federal Fluminense (UFF) – Rio de Janeiro, Brazil

Video Recording

Video is available only for registered users.

Abstract

Molecular Magnetic Compounds can have several different features, from single molecules with long range couplings to complex polymeric chains with strong interactions. The field of Molecular Magnetism occupy itself in understanding how these systems interact and in chemically modulating magnetic couplings by selecting adequate ligands and paramagnetic centers. The task to devise how the infinity of magnetic systems couple with each other in a crystal structure is possible with a methodological First-priciples Bottom-up (FPBU) approach and the use of Computational Methods like the Density Functional Theory (DFT), allowing for fast and accurate results. In this presentation we show, using examples of weak and strong interactions, how to apply the FPBU approach to choose reasonable magnetic systems as relevant candidates for magnetic coupling studies, functionals and basis sets leading to better results and how to use the broken-symmetry approach to obtain magnetic couplings (J).

Benchmarking reaction energies, barrier heights, and noncovalent interaction energies

Gershom (Jan M.L.) Martin

Department of Organic Chemistry
Weizmann Institute of Science
7610001 Reḥovot, Israel
http://compchem.me

Video Recording

Video is available only for registered users.

Abstract

While density functional theory has made great strides, even the best exchange-correlation functionals are about one order of magnitude less accurate than can be achieved using modern wavefunction ab initio techniques. The latter have a well-defined road map for refinement in accuracy; however, their steep computational cost scaling with system size limits their use to relatively small molecules. While some applications (e.g., atmospheric chemistry, fine thermochemistry) demand such levels of accuracy, a perhaps more important application is the creation of benchmark datasets for the parametrization and validation of density functional, reactive force fields, and other lower-cost methods.
Using the case of total atomization energies, we will discuss the breakdown of molecular binding energies into their constituent components, as well as the optimal convergence strategy for each. By such “layered approximations” as implemented in the Weizmann-n series of thermochemistry protocols [1,2,3] (and its ‘competitor’, the HEAT approach [4]), CPU times and memory requirements can be drastically reduced versus brute-force approaches. The introduction of explicitly correlated coupled cluster theory brings still larger molecules within reach, as long as non dynamical correlation effects are not too important. (See [5] for a discussion of static correlation diagnostics.)

We will illustrate some of the concepts using the W4-11 atomization energy benchmark [6], the DBH24 barrier heights benchmark [7], the HFREQ27 vibrational frequencies benchmark [8], and several recent benchmarks for noncovalent interactions such as the S66x8 set of biomolecule dimer potentials, [9], conformational energies of the proteinogenic amino acids, [10], and water clusters [11].

Computational medicinal chemistry

Dr. Robert J. Doerksen

Associate Dean, Graduate School
Associate Professor of Medicinal Chemistry, Department of BioMolecular Sciences
Research Associate Professor, Research Institute of Pharmaceutical Sciences
University of Mississippi, University, MS, USA

Video Recording

Abstract

A wide variety of computational chemistry methods are useful in the search for new drugs. These approaches are collectively termed computational medicinal chemistry. A typical small molecule drug (molecular weight < 500 Da) needs to interact with or react with a protein target to achieve its useful pharmacological effect. Its path through the human body can also include changing protonation state, crossing lipid barriers, being carried by proteins, and undergoing metabolic transformations. A series of computational methods can be used to study the progress of a drug through the body in the various stages of pharmacokinetics and pharmacodynamics. Three-dimensional representations of both the drug and of what it interacts with are often helpful. For this, conformational search and methods to calculate and rank the relative energies of conformations are necessary. Many electronic structure properties of the drug molecule can be calculated, which can be used to characterize the molecule and predict its behavior. Protein modeling is also important to carry out, including effective use of experimental structural information. The conformations of drug and target can then be used in molecular docking which in turn can serve as a key step in virtual screening to find, from a database of known structures, drug hits with never-before-reported useful pharmacological activity at targets of interest. This presentation will include examples of best-practice application of these methods, such as for identifying selective protein kinase inhibitors or cannabinoid receptor ligands.

Computational enzymology

James W. Gauld, Professor

Department of Chemistry and Biochemistry,
University of Windsor,
Windsor,
Ontario, N9B 3P4
Canada

Abstract

Elucidating the properties and chemistry of enzymes has long been of significant importance. This is due in part to the fact that they are central to many physiological processes that occur in cells. In particular, they are critical for ensuring that metabolically important reactions that occur within cells and organisms occur with life-sustaining rates, efficiency, and accuracy. Furthermore, they often achieve this under relatively mild conditions. Thus, in addition to the fundamental knowledge to be gained, they also present tremendous potential health and industrial benefits.
Indeed, it has been estimated that in the US more than 90% of chemical and pharmaceutical manufacturing requires catalysts.¹ Meanwhile, due to their critical physiological roles, enzymes are often the desired target of therapeutic drugs. Recently, the World Health Organization declared "antibiotic resistance one of the biggest threats to global health, food security, and development".² Rational design is a powerful tool for developing new drugs to combat this present and growing threat. For those that target enzymes this requires detailed knowledge of the latter's active site structures, properties, and mechanisms. Unfortunately, this knowledge is often at best limited.
^3,4Computational enzymology, in its broadest sense, is the use of computers to study the properties and mechanisms of enzymes. However, one of its major goals is to elucidate the catalytic mechanism of an enzyme or enzymes, the role of their key active site residues, and/or surrounding protein and solvent environment. Nowadays there are a range of computational methods available to the researcher that can be brought to bear on such challenges including molecular dynamics, quantum mechanical (QM)-chemical cluster, quantum mechanical/molecular mechanic (QM/MM). Increasingly, it is common to complementarily apply several of these methods.
In this lecture we will discuss several key aspects of computational enzymology including chemical model construction, commonly applied computational methods and their complementary application and challenges. These will be illustrated using examples from the literature as well as our own research in the Gauld group.

Quantum-chemical calculation of spectroscopic parameters

Cristina Puzzarini

Department of Chemistry “Giacomo Ciamician”, University of Bologna
https://site.unibo.it/rotational-computational-spectroscopy

Video Recording

Video is available only for registered users.

Abstract

Implementation of very accurate ab initio methods on one hand and improvements in computer facilities on the other hand allow the determination of structural, molecular, and spectroscopic properties of small- to medium-size molecules to a very high accuracy. The predictive capabilities have such an accuracy that theoretical calculations can guide, support and even challenge experimental determinations. To perform accurate quantum-chemical calculations of spectroscopic parameters, post-HF methods, such as the coupled cluster ones, should be employed in conjunction with extrapolative and additive techniques in order to account for basis set and wave function truncation errors as well as to include important corrections, such as those related to core correlation and vibrational effects.
Concerning rotational spectra, the starting point is the computation of an accurate equilibrium geometry for the evaluation of the corresponding rotational constants.  Next, quadratic and cubic force constants allow the determination of vibrational corrections to rotational constants together with the quartic and sextic centrifugal-distortion constants. Finally, accurate dipole moment components and, when needed, quadrupole-coupling constants complete the list of the needed quantities. The most effective strategy relies on coupled-cluster (CC) evaluations (at the CC singles and doubles augmented by a perturbative treatment of triple excitations, CCSD(T)) of equilibrium geometries, properties, and, possibly, quadratic force constants in a normal-mode representation (i.e., harmonic frequencies), also considering extrapolation to the complete basis set (CBS) limit and core− valence correlation (CV) contributions. These results can be complemented by density functional theory (DFT) evaluations of the anharmonic contributions (employing hybrid or double hybrid functionals, like B3LYP and B2PLYP, in conjunction with medium-sized basis sets). For larger molecules, benchmark studies suggest that the computationally expensive (and slowly converging) evaluation of harmonic frequencies at the CCSD(T)/CBS level augmented by CV corrections can be replaced by effective B2PLYP computations without any dramatic reduction of the accuracy for both rotational and vibrational spectroscopy investigations.
Concerning vibrational spectroscopy, the simulation of fully anharmonic spectra including fundamental, overtones, and combination bands  requires, in addition to the quantities discussed above for the rotational spectra, quartic (at least semi-diagonal) force constants together with second and third derivatives of the electric dipoles (for IR spectra). Once again, hybrid and double hybrid functionals perform very well in the computation of anharmonic contributions, provided that, for the evaluation of the electric dipoles, diff use functions are properly included in the basis set. Subsequently, the vibrational problem can be solved by either variational or perturbative approaches. For semi-rigid molecules, second-order vibrational perturbation theory (VPT2) is particularly effective, provided that nearly resonant contributions are treated by means of a variational approach, thus leading to the so-called generalized VPT2 model (GVPT2).