Steven C. Erwin

Doping semiconductor nanocrystals

Doping—the intentional introduction of impurities into a material—is fundamental to controlling the properties of bulk semiconductors. This has stimulated similar efforts to dope semiconductor nanocrystals. Despite some successes, many of these efforts have failed, for reasons that remain unclear. For example, Mn can be incorporated into nanocrystals of CdS and ZnSe (refs 7–9), but not into CdSe (ref. 12)—despite comparable bulk solubilities of near 50 per cent. These difficulties, which have hindered development of new nanocrystalline materials, are often attributed to ‘self-purification’, an allegedly intrinsic mechanism whereby impurities are expelled. Here we show instead that the underlying mechanism that controls doping is the initial adsorption of impurities on the nanocrystal surface during growth. We find that adsorption—and therefore doping efficiency—is determined by three main factors: surface morphology, nanocrystal shape, and surfactants in the growth solution. Calculated Mn adsorption energies and equilibrium shapes for several nanocrystals lead to specific doping predictions. These are confirmed by measuring how the Mn concentration in ZnSe varies with nanocrystal size and shape. Finally, we use our predictions to incorporate Mn into previously undopable CdSe nanocrystals. This success establishes that earlier difficulties with doping are not intrinsic, and suggests that a variety of doped nanocrystals—for applications from solar cells to spintronics—can be anticipated.

The structure of silicon surfaces from (001) to (111)

Abstract We describe the structure of silicon surfaces oriented between (001) and (111) as determined by scanning tunneling microscopy (STM) and first-principles, total-energy calculations. In addition to reviewing and reproducing the structures reported for the few surfaces previously studied, we describe a number of additional surfaces in order to provide a complete overview of the (001)-to-(111) surface morphology. As the sample orientation is titled from (001) to (111) (ϑ=0 to 54.7°), the surface morphology varies as follows: (1) Si(001) to Si(114) = (001)-like surfaces composed of dimers separated by steps (both rebonded and nonrebonded); (2) Si(114) to Si(113) =mesoscale sawtooth facets composed of the stable (114)−2 × 1 and (113)−3 × 2 planes; (3) Si(113) to Si(5 5 12) =mesofacets composed of (113)−3 × 2 and (5 5 12)-like planes; (4) Si(5 5 12) to ∼Si(223) =nanoscale sawtooth facets composed of (5 5 12)-like and unit-cell-wide (111)−7 × 7 planes; and (5) ∼Si(223) to Si(111)=(111)−7 × 7 terraces separated primarily by single- and triple-layer steps. The change in the surface morphology is accompanied by a change in the composition of surface structural units, progressing from (001)-like structures (e.g. dimers, rebonded steps, and tetramers) to (111)-like structures (π-bonded chains, adatoms and dimer-chain walls). The resultant morphology is a delicate balance between the reduction of dangling bond density achieved by the formation of these structural units, and the resulting surface stress associated with their unusual bond angles and bond lengths.

Theory of the {open_quotes}Honeycomb Chain-Channel{close_quotes} Reconstruction of {ital M}/Si( 111) -(3{times}1)

First-principles electronic-structure methods are used to study a structural model for Ag/Si(111) -(3{times}1) , recently proposed on the basis of transmission electron diffraction data. The fully relaxed geometry for this model is far more energetically favorable than any previously proposed, partly due to the unusual formation of a Si double bond in the surface layer. The calculated electronic properties of this model are in complete agreement with data from angle-resolved photoemission and scanning tunneling microscopy. {copyright} {ital 1998} {ital The American Physical Society}

Theoretical Fermi-Surface Properties and Superconducting Parameters for K3C60

Quantitative theories of superconductivity in alkali-doped C60 require an accurate and detailed description of the Fermi surface. First-principles calculations of Fermi-surface properties and electronic parameters for K3C60, the prototype fulleride-superconductor, are reported. The Fermi surface has two sheets; the first is free-electron-like, and the second is multiply-connected, forming two interlocked symmetry-equivalent pieces that never touch. The calculated (clean limit) London penetration depth is Λ = 1600 �. Comparing the Fermi velocity with the experimental coherence length leads to a superconducting pairing strength λ ∼ 5, indicating very strong coupling. Partial nesting in the second Fermi-surface sheet may favor coupling to short-wavelength q,0,0 optic modes.

Epitaxial ferromagnetic Mn5Ge3 on Ge(111)

Ferromagnetic Mn5Ge3 thin films were grown on Ge(111) with solid-phase epitaxy. The epitaxial relationship between the alloy film and substrate is Mn5Ge3(001)//Ge(111) with [100]Mn5Ge3//[110]Ge. The alloy films exhibit metallic conductivity and strong ferromagnetism up to the Curie temperature, TC=296 K. These epitaxial alloy films are promising candidates for germanium-based spintronics.

Tailoring ferromagnetic chalcopyrites

If magnetic semiconductors are ever to find wide application in real spintronic devices, their magnetic and electronic properties will require tailoring in much the same way that bandgaps are engineered in conventional semiconductors. Unfortunately, no systematic understanding yet exists of how, or even whether, properties such as Curie temperatures and bandgaps are related in magnetic semiconductors. Here we explore theoretically these and other relationships within 64 members of a single materials class, the Mn-doped II-IV-V2 chalcopyrites (where II, IV and V represent elements from groups II, IV and V, respectively); three of these compounds are already known experimentally to be ferromagnetic semiconductors. Our first-principles results reveal a variation of magnetic properties across different materials that cannot be explained by either of the two dominant models of ferromagnetism in semiconductors. On the basis of our results for structural, electronic and magnetic properties, we identify a small number of new stable chalcopyrites with excellent prospects for ferromagnetism.

Self-compensation in manganese-doped ferromagnetic semiconductors

We present a theory of interstitial Mn in Mn-doped ferromagnetic semiconductors. Using density-functional theory, we show that under the nonequilibrium conditions of growth, interstitial Mn is easily formed near the surface by a simple low-energy adsorption pathway. In GaAs, isolated interstitial Mn is an electron donor, each compensating two substitutional Mn acceptors. Within an impurity-band model, partial compensation promotes ferromagnetic order below the metal-insulator transition, with the highest Curie temperature occurring for 0.5 holes per substitutional Mn.

A Stable High-Index Surface of Silicon: Si(5 5 12)

A stable high-index surface of silicon, Si(5 5 12), is described. This surface forms a 2 x 1 reconstruction with one of the largest unit cells ever observed, 7.7 angstroms by 53.5 angstroms. Scanning tunneling microscopy (STM) reveals that the 68 surface atoms per 2 x 1 unit cell are reconstructed only on a local scale. A complete structural model for the surface is proposed, incorporating a variety of features known to exist on other stable silicon surfaces. Simulated STM images based on this model have been computed by first-principles electronic-structure methods and show excellent agreement with experiment.

Spin Injection and Detection in Silicon

Spin injection and detection in silicon is a difficult problem, in part because the weak spin-orbit coupling and indirect gap preclude using standard optical techniques. Two ways to overcome this difficulty are proposed, both based on spin-polarized transport across a heterojunction. Using a realistic transport model incorporating the relevant spin dynamics of both electrons and holes, it is argued that symmetry properties of the charge current can be exploited to detect electrical spin injection in silicon using currently available techniques.

Intrinsic magnetism at silicon surfaces

It has been a long-standing goal to create magnetism in a non-magnetic material by manipulating its structure at the nanoscale. Many structural defects have unpaired spins; an ordered arrangement of these can create a magnetically ordered state. In this article we predict theoretically that stepped silicon surfaces stabilized by adsorbed gold achieve this state by self-assembly, creating chains of polarized electron spins with atomically precise structural order. The spins are localized at silicon step edges having the form of graphitic ribbons. The predicted magnetic state is supported by recent experimental observations, such as the coexistence of double- and triple-period distortions and the absence of edge states in photoemission. Ordered arrays of surface spins can be accessed by probes with single-spin sensitivity, such as spin-polarized scanning tunnelling microscopy. The integration of structural and magnetic order is crucial for technologies involving spin-based computation and storage at the atomic level.