We have a YouTube channel where we have putting short talks about what we do. Take a look at some of these below!  Each one of these was a lightning talk at a recent conference in the field, a cool new format in our virtual world. 

VCTC 2020

Emily talks about the organic synthesis of borylnitrenes:

We performed a systematic computational investigation into borylnitrenes, which can undergo different rearrangements with varying substituents, to help advance experimental design. Density function theory (DFT) calculations were performed to find the energy gaps between singlet and triplet states. In addition, we found that the singlet states were either found to form a ring following carbon insertion or rearrange to an iminoborane. The outcome depends on the substituents. There were not generally observed geometry changes for the triplet states. This is useful because it will help to start screening a database of compounds to find those which are both easy to activate and undergo appropriate rearrangements.

Tina talks about using the transition structure factor to learn about finite size effects in metals:

In periodic solids, errors in the energies are typically thought to converge to the thermodynamic limit (TDL), or bulk, with an  scaling, where  is the number of electrons in the system. However, in our analysis of the coupled cluster doubles correlation energy in the high density of the uniform electron gas, where coupled cluster is taken to be exact, we found the error scales as  Through analysis of the part of the coupled cluster wavefunction known as the transition structure factor, whose convergence to the TDL mirrors that of the errors, we show where and how each convergence rate occurs and how they can be applied to solids in general.   

Laura talks about her research on machine learning algorithms for Coupled Cluster Correlation Energy:

We use machine learning to analyze the correlation energy from coupled cluster in uniform electron gas calculations. The aim is to take a relatively small system, calculate its coupled cluster wavefunction, and predict the thermodynamic limit correlation energy. Preliminary data show promise and our next step is to transfer this to real solids for use in materials design.

ESW 2020

Tina talks about the convergence rate of the finite size effects in the thermodynamic limit of connectivity-twist-averaged coupled cluster calculations: 

We performed a numerical analysis on the thermodynamic limit (TDL) extrapolation power laws for the coupled cluster doubles correlation energy in the uniform electron gas using a new cost- reducing twist averaging method we recently developed.[1] The thermodynamic limit energies were found for a range of densities. The high-density limit, where exact TDL values are known, was then used to determine the convergence rate which showed a different convergence than the accepted 1/N rate.

Hayley Talks about combining i-FCIQMC with projection based embedding:

We present proof of concept results for the fully quantum embedded full configuration interaction quantum Monte method.  In this fully quantum method, we embed the intiator version of full configuration interaction in density functional theory, called i-FCIQMC-in-DFT. Here we show that that i-FCIQMC-in-DFT has comparable accuracy to coupled cluster singles and doubles with perturbative triples embedded in density functional theory, CCSD(T)-in-DFT, by calculating dissociation energies of diatomic systems physisorbed on benzene. In addition, by generating a dissociation curve of hydrogen fluoride on benzene we show improvement over CCSD(T)-in-DFT in the bond breaking region. 

Laura talks about frozen core coupled cluster calculations:

We use finite coupled cluster calculations of the uniform electron gas to study the effect of orbital freezing on the correlation energy in solids. Taking a finite electron gas and freezing out core electrons starting from the orbitals with the lowest kinetic energy, we show that the fraction of energy lost is approximately linear in the number of frozen orbitals. This could be of interest to those practitioners applying coupled cluster theory to solids.