One of the strengths in the UCL Chemistry Department is structure prediction, so I was excited to collaborate with Scott Woodley on the low energy structures formed for clusters of In2O3. Scott recently co-authored a nice review of global optimisation techniques in Nature Materials.

In our work, one of the key results was that the energy landscape found for a classical potential model that we developed and density functional theory were very similar. The project also allowed me to try out quasi-particle calculations (GW theory) within the FHI-AIMS code to investigate the spectroscopic properties.

I have collected the low energy structures that we found for (In2O3)n, from 1 ≤ n ≤ 10, which are available here for those who are curious of the nanochemistry of indium oxide.

I attended a stimulating TYC soirée this week on issues relating to exchange and correlation in Density Functional Theory. Nicola Marzari painted the picture of Narcissus, where the reflection (delocalisation) of an electron arises from the infamous self-interaction error of DFT, or in other words, the absence of electron exchange to exactly cancel the Coulomb interaction of an electron with itself.

The interesting point for me was the distinction between the long and short-range self-interaction error, which manifests in different types of chemical problems, e.g. short-range hybridisation of metal d - anion p orbitals versus long-range charge transfer reactions for mixed oxidation states. Each regime can be corrected differently, depending on the level of approximation: SIC; DFT+U; Hybrid Functionals; Enforced Koopman's Condition, etc. In reality, no correction method solves all issues, but there is now no excuse for blindly applying uncorrected local/semi-local functions to problems that they have been demonstrated to be inappropriate for.

The second talk of the night was from Ali Alavi on Quantum Monte Carlo. I saw the same talk in Shanghai, so on second viewing, it became more digestible. Conceptually, his approach is simple: start with any set of single-particle wavefunctions (e.g. from Hartree-Fock or Density Functional Theory) and sample the multi-determinant space in a stochastic fashion to obtain a good approximation to the many-body wavefunction, at a fraction of the cost of an explicit calculation. The trick is in the algorithm, which Ali appears to have finely crafted over the last few years, combining the best parts of evolution and genocide for his “walkers”. The approach has now been extended to periodic model problems, and has the potential to mature into a new benchmark technique for solid-state calculations.

One difference (of semantics) to note is that while Marzari would argue that the principal error of DFT is in the "exchange" term, by definition in the latter approach the same error falls under electron "correlation" - perhaps, evidence of exchange-correlation duality?

As the field of dilute magnetic semiconductors continues to thrive in the slums of caveat emptor calculations, it appears that intermediate band gap solar cells are following a similar fate.

Building upon much earlier work in the field, in 1997 Luque and Marti calculated that introducing an intermediate band gap state (IB) into a semiconductor could offer a means to achieve solar conversion efficiencies of up to 63.1%, well beyond the single-junction limit. This could be achieved in many ways, but one of the most straightforward is through impurity doping.

The principal assumption of the high efficiency is that there exists a half-filled, localised defect band 0.7 eV above the valence band (as it is half-filled, valence -> IB and IB -> conduction band optical transitions are independent). Generally, a localised half-filled band will be subject to a Peierls distortion, so it is not an easy task. A perfect case for electronic structure theory to lead the way...

One of the early disasters was doping of the III-V materials GaAs and GaP, e.g. an excellent IB was predicted when Ti is substituted on the *anion* site (over a series of papers!). Much of the subsequent work focused on transition metal doping and alloying of oxides, II-VI and III-V compounds; however, it is well demonstrated that standard local/semi-local density functionals cannot deal with transition metals, especially when you are interested in quantitative band positions and optical absorption properties. In particular, it is difficult to see how a partially occupied 3d band could be stable with respect to a Jahn-Teller distortion. The same issue applies to recent work on Ti and Cr doping of the Ga site in CuGaS2, where the results will be highly sensitive to the treatment of the 3d states and to the form of ionic charge compensation.

The main danger here is that experimentalists read these papers, get their hopes up, and ultimately lose respect for the predictive role of material simulation.