Invited: Jeffrey Neaton
Excited States & Spectroscopy of Organic Semiconductors from First Principles
Organic semiconductors are a highly tunable, diverse class of cheap-to-process materials promising for next-generation optoelectronics, for example solar cells. Further development of new organic materials requires new intuition that links molecular-scale morphology to underlying excited-state properties and phenomena. Here, I will discuss the use of first-principles density functional theory and many-body perturbation theory - within the GW approximation and the Bethe-Salpeter equation approach - for computing and understanding spectroscopic properties of selected organic semiconductor crystals, including acenes from benzene to hexacene; PTCDA; perfloropentecene; and TIPS-pentacene. For both gas-phase and crystals, our quantitative calculations agree well with transport gaps extracted from photoemission and inverse photoemission data, and with measured polarization-dependent optical absorption spectra. Introducing a new analysis, we elucidate the nature of low-lying solid-state singlet and triplet excitons, which have significant binding energies and charge-transfer character in these systems, and assess the effects of dynamic disorder associated with room-temperature structural fluctuations. In addition, we rationalize trends in predicted singlet and triplet excitation energies in the context of recently proposed mechanisms for singlet fission. Insights into new materials, as well as implications for recent spectroscopic experiments, are discussed.
Invited: Hongjun Xiang
Computational Design of Energy Materials
Co-authors: Suhuai Wei, Xingao Gong
Diamond silicon (Si) is the leading material in the current solar cell market. However, diamond Si is an indirect band gap semiconductor with a large energy difference (2.3 eV) between the direct gap and the indirect gap, which makes it an inefficient absorber of light. We recently proposed to alloy Si with its mutated III−V or II−VI semiconductors such as AlP or MgS to improve the optical properties: the C1c1-Si3AlP phase has a larger fundamental band gap and a smaller direct band gap and much higher absorption in the visible light region than Si1. On the other hand, we develop a novel inverse band structure design approach based on the evolution algorithm to predict the metastable Si phases with better optical properties than diamond Si. Using our new method, we predict a cubic Si20 phase with a quasi-direct gap of 1.55 eV, which is a promising candidate for making thin-film solar cells2. Furthermore, we have developed a multi-objective evolution algorithm to optimize several desirable properties simultaneously. Employing the Inverse-Design of Materials by Multi-Objective Differential Evolution (IM2ODE) package, we predict two low-energy low-band-gap TiO2 phases for high-efficiency solar energy conversion3.
 J.-H. Yang, Y. Zhai, H. Liu, H. J. Xiang*, X. G. Gong*, and S.-H. Wei, J. Am. Chem. Soc. 134, 12653 (2012).
 H. J. Xiang*, Bing Huang, Erjun Kan, Su-Huai Wei, and X. G. Gong, Phys. Rev. Lett. 110, 118702 (2013).
 H. Z. Chen, Y. Zhang, X. G. Gong*, and H. J. Xiang*, J. Phys. Chem. C. (2014).
Invited: Noa Marom
Describing the Electronic Structure of Organic Semiconductors by GW Methods
Organic photovoltaics are attractive for large area, low cost applications and for flexible modules. However, their relatively low efficiency leaves much to be desired. Insight from computation may help improve device performance by designing new organic semiconductors and ordered nanostructured interfaces. It is important to obtain an accurate description of the electronic structure of organic semiconductors, including the fundamental gaps and absolute positions of the donor HOMO and the acceptor LUMO at the interface. This requires going beyond ground state density functional theory (DFT). For this purpose, many-body perturbation theory is often used within the GW approximation, where G is the one particle Green function and W is the dynamically screened Coulomb interaction. Typically, GW calculations are performed as a non-self-consistent perturbative correction to DFT eigenvalues, known as G0W0. The predictive power of G0W0 is limited by a strong dependence of the results on the DFT starting point. Self-interaction errors (SIE), spurious charge transfer, and incorrect ordering and hybridization of molecular orbitals may propagate from the DFT level to G0W0. These issues may be addressed by judiciously choosing a hybrid DFT starting point or by going beyond G0W0 to a higher level of self-consistency. Here, this is demonstrated for prototypical organic semiconductors, such as phthalocyanines and anhydrides.
Invited: Stefan K. Estreicher
Heat Flow in Semiconductor Nanostructures Containing Defects: A First-Principles Study
Co-authors: Byungkyun Kang
The lattice contribution to heat transport in materials containing surfaces, boundaries, dislocations, impurities, and other defects is commonly described in terms of "phonon scattering": defects are static scattering centers for bulk phonons. But the dynamics of defects should not be ignored, especially in nanostructures. If the spatially-localized vibrational modes of the defects are included in the calculations, phonon scattering is seen to involve phonon trapping at defects for substantial periods of time, followed by the decay of the vibrational excitation into combination of bulk phonons. Phonon trapping reduces the thermal conductivity. The interplay between delocalized (bulk) and localized (defect-related) vibrational modes could be controlled as (1) the lifetime of trapped phonons varies with the nature of defect, and (2) the decay of the excitations involves phonons of lower frequency and this can be used for heat control at heterostructures. In this talk, we will illustrate these processes in Si nanowires. The defects considered are the surface, mono-atomic delta layers of Ge or C compared to random distributions of the same impurities. We will show that if the surface is H-saturated, the localized surface modes (Si-H wag modes) couple resonantly to each other much faster than they decay into the bulk, leading to distinct bulk and surface thermal conductivities. Thermal equilibrium in the nanowire is reached only after the bulk (faster) and surface (slower) contributions have propagated.