(456e) First-Principles Theoretical Analysis of Doping in II-VI Compound Semiconductor Nanocrystals

Singh, T., University of Massachusetts - Amherst
Maroudas, D., University of Massachusetts, Amherst
Mountziaris, T. J., University of Massachusetts

Nanocrystalline structures of compound semiconductors, such as the II-VI compounds ZnS, CdSe, and ZnSe, exhibit size-dependent optoelectronic properties and form the basis for a new generation of nanoelectronic and photovoltaic devices, as well as biological labels. Doping in semiconductor nanocrystals allows for precise control of their optical and electronic properties. However, in spite of its feasibility, doping of semiconductor nanocrystals has been an extremely difficult task. Self-purification of nanocrystals makes the incorporation of dopants into nanocrystals even more difficult. Typically, Mn-doped ZnSe nanocrystals have been grown using hot-injection organometallic synthesis. It has been observed that increase in anion:cation ratios leads to higher doping efficiencies. This has been attributed to adsorption of Mn dopant atoms into ZnSe nanocrystals through their surface facets, in particular ZnSe(001)-(2x1) surface facets. In this context, we aim at obtaining a fundamental and quantitative understanding of dopant adsorption and diffusion on II-VI semiconductor nanocrystal surfaces, which can help elucidate the mechanisms of dopant incorporation into growing nanocrystals.

In this presentation, we report results on dopant adsorption and diffusion on surface facets of ZnSe and ZnSe1-xSx nanocrystals based on first-principles density functional theory (DFT) calculations within the generalized gradient approximation. In our DFT calculations, we have employed slab supercells, plane-wave basis sets, and the projector-augmented wave method. We have also implemented the climbing-image nudged elastic band method to construct fully optimized dopant diffusion pathways. We have made indirect comparisons between our DFT analysis and experimental findings on doping efficiencies.

We have computed the surface energies of the three high-symmetry surfaces [(001), (110), and (111)] of ZnSe and examined the corresponding surface reconstructions and their stability as a function of the anion (Se) chemical potential. Based on these findings, we have constructed the equilibrium crystal shape (ECS) of ZnSe nanocrystals as a function of the Se chemical potential and found that anion-rich reconstructed surface facets are present in the ECS of a ZnSe nanocrystal. We have computed the binding energies of Mn dopant atoms onto all of the ZnSe nanocrystal facets and found that they depend strongly on the surface morphology and nanocrystal shape. We conclude that all anion-rich surfaces contribute toward dopant adsorption onto ZnSe nanocrystal surface facets.

In addition, our DFT calculations indicate that the binding energy for Mn adsorption onto various sites of all surface facets increases with increasing dopant surface concentration. This low binding energy at low dopant surface concentration explains the doping difficulties during nanocrystal growth. Furthermore, we have analyzed several dopant migration pathways for Mn diffusion on the ZnSe(001)-(2x1) surface. The calculated activation barriers for migration along the Se dimer rows range from 0.17 eV to 0.43 eV, depending upon the dopant surface concentration. Due to the low activation barriers, dopant atoms can migrate fast along the Se dimer rows without substantial surface relaxation. However, migration of the Mn atom across the Se dimer rows from the dimer site (Mn adsorbed onto the Se dimer) to the trough site (Mn adsorbed in the trough between Se dimers) is governed by a high-barrier pathway. Finally, the above analysis of dopant adsorption and diffusion on nanocrystal facets has been extended to ZnSe1-xSx nanocrystals in order to examine the effects of nanocrystal alloying on the doping efficiency.