Yesinowski James P, Purdy Andrew P, Wu Huaqiang, Spencer Michael G, Hunting Janet, DiSalvo Francis J
Chemistry Division, Naval Research Laboratory, Washington, DC 20375, USA.
J Am Chem Soc. 2006 Apr 19;128(15):4952-3. doi: 10.1021/ja0604865.
A strategy is demonstrated for identifying unambiguously and characterizing quantitatively the effects of distributions of conduction electron concentrations arising from intentional or unintentional dopants in semiconductors by magic-angle spinning (MAS) NMR. The 71Ga MAS NMR spectra of a number of chemically synthesized GaN samples with no intentional doping show inhomogeneously broadened absorptions to high frequency of the main peak. These broad signals are shown, from spin-lattice relaxation time measurements as a function of shift position in a single sample, to be due to Knight shifts arising from degenerate conduction electrons. For a GaN sample with Ge as an intentional dopant at the 0.13% (wt) level, the spectrum is dramatically broadened and shifted to high frequency by up to several hundred parts per million. Analysis of the inhomogeneously broadened line shape yields a quantitative probability density function for electron carrier concentration in the bulk sample that reflects significant compositional heterogeneity due to a variety of possible sources.
本文展示了一种通过魔角旋转(MAS)核磁共振(NMR)明确识别并定量表征半导体中有意或无意掺杂剂引起的传导电子浓度分布效应的策略。一些未有意掺杂的化学合成氮化镓(GaN)样品的71Ga MAS NMR谱显示,主峰高频处存在不均匀展宽的吸收峰。通过在单个样品中测量自旋 - 晶格弛豫时间作为位移位置的函数,表明这些宽信号是由简并传导电子引起的奈特位移所致。对于在0.13%(重量)水平有意掺杂锗的GaN样品,光谱显著展宽并向高频移动高达百万分之几百。对不均匀展宽线形的分析得出了体样品中电子载流子浓度的定量概率密度函数,该函数反映了由于多种可能来源导致的显著成分异质性。