We develop a practical first-principles methodology to determine nonradiative carrier capture coefficients at defects in semiconductors. We consider transitions that occur via multiphonon emission. Parameters in the theory, including electron-phonon coupling matrix elements, are computed consistently using state-of-the-art electronic structure techniques based on hybrid density functional theory. These provide a significantly improved description of bulk band structures, as well as defect geometries and wavefunctions. In order to properly describe carrier capture processes at charged centers, we put forward an approach to treat the effect of long-range Coulomb interactions on scattering states in the framework of supercell calculations. We also discuss the choice of initial conditions for a perturbative treatment of carrier capture. As a benchmark, we apply our theory to several hole-capturing centers in GaN and ZnO, materials of high technological importance in which the role of defects is being actively investigated. Calculated hole capture coefficients are in good agreement with experimental data. We discuss the insights gained into the physics of defects in wide-band-gap semiconductors, such as the strength of electron-phonon coupling and the role of different phonon modes.

We study the spin and orbital dynamics of single nitrogen-vacancy (NV) centers in diamond between room temperature and 700 K. We find that the ability to optically address and coherently control single spins above room temperature is limited by nonradiative processes that quench the NV center's fluorescence-based spin readout between 550 and 700 K. Combined with electronic structure calculations, our measurements indicate that the energy difference between the 3 E and 1 A 1 electronic states is approximately 0.8 eV. We also demonstrate that the inhomogeneous spin lifetime (T Ã 2 ) is temperature independent up to at least 625 K, suggesting that single NV centers could be applied as nanoscale thermometers over a broad temperature range. DOI: 10.1103/PhysRevX.2.031001 Subject Areas: Quantum Information, Semiconductor Physics, SpintronicsThe negatively charged nitrogen-vacancy (NV) center spin in diamond stands out among individually addressable qubit systems because it can be initialized, coherently controlled, and read out at room temperature [1]. The defect's robust spin coherence [2] and optical addressability via spin-dependent orbital transitions [3] have enabled applications ranging from quantum information processing [4][5][6][7] to nanoscale-magnetic and electric-field sensing [8][9][10]. While it has been shown that NV center spins can be optically polarized up to at least 500 K [11,12], little is known about what processes limit the spin's optical addressability and coherence at higher temperatures. Understanding these processes is important to hightemperature field-sensing applications and will aid the search for new defect-based spin qubits analogous to the NV center [13,14] by identifying the aspects of its orbital structure responsible for its high-temperature operation.The NV center's optical-spin polarization and opticalspin readout result from a spin-selective intersystem crossing (ISC). Although optical transitions between the spin-triplet ground ( 3 A 2 ) and excited ( 3 E) states [1.945 eV zero-phonon line (ZPL)] are typically spin conserving, the 3 E state can also relax to the 3 A 2 state via an indirect pathway that involves a nonradiative, triplet to singlet ISC and subsequent transitions through at least one additional singlet [ Fig. 1(a)]. The 3 E ISC is much stronger for the m s ¼ AE1 3 E sublevels than for the m s ¼ 0 sublevel, which facilitates spin readout through the resulting spindependent photoluminescence (PL) and initializes the spin into the m s ¼ 0 3 A 2 sublevel with high probability (P m s ¼0 $ 0:8) through repeated optical excitation [15]. Despite the singlet pathway's critical role in preparing and interrogating the spin, open questions remain regarding the number and energies of the singlets involved. Recent experiments have established that it consists of at least two singlets of 1 A 1 and 1 E symmetry separated by 1.19 eV [16][17][18], and have shown that the 1 E state lifetime is strongly temperature dependent below 300 K [17,19]. The details of the 3 E ISC remain unc...

In this work we present theoretical calculations and analysis of the vibronic structure of the spin-triplet optical transition in diamond nitrogen-vacancy (NV) centres. The electronic structure of the defect is described using accurate firstprinciples methods based on hybrid functionals. We devise a computational methodology to determine the coupling between electrons and phonons during an optical transition in the dilute limit. As a result, our approach yields a smooth spectral function of electron-phonon coupling and includes both quasi-localized and bulk phonons on equal footings. The luminescence lineshape is determined via the generating function approach. We obtain a highly accurate description of the luminescence band, including all key parameters such as the Huang-Rhys factor, the Debye-Waller factor, and the frequency of the dominant phonon mode. More importantly, our work provides insight into the vibrational structure of NV centres, in particular the role of local modes and vibrational resonances. In particular, we find that the pronounced mode at 65 meV is a vibrational resonance, and we quantify localization properties of this mode. These excellent results for the benchmark diamond (NV) centre provide confidence that the procedure can be applied to other defects, including alternative systems that are being considered for applications in quantum information processing.

Hexagonal BN (h-BN) is attracting a lot of attention for two-dimensional electronics and as a host for single-photon emitters. We study the properties of native defects and impurities in h-BN using density functional theory with a hybrid functional. Native vacancy and antisite defects have high formation energies, and are unlikely to form under thermodynamic equilibrium for typical growth conditions. Self-interstitials can have low formation energies when the Fermi level is near the band edges, and may form as charge compensating centers; however, their low migration barriers render them highly mobile, and they are unlikely to be present as isolated defects. The defect chemistry of h-BN is most likely dominated by defects involving carbon, oxygen, and hydrogen impurities. Substitutional carbon and oxygen, as well as interstitial hydrogen and boron vacancy-hydrogen complexes, are low-energy defects in h-BN. Based on our results, we can rule out several proposed sources for defect-related luminescence in h-BN. In particular, we find that the frequently observed 4.1 eV emission cannot be associated with recombination at CN, as has been commonly assumed. We suggest alternative assignments for the origins of this emission, with CB as a candidate. We also discuss possible defect origins for the recently observed single-photon emission in h-BN, identifying interstitials or their complexes as plausible centers.