The k p Method

This foreword recalls how the author first encountered the k·p method in a semiconductor physics course at Harvard in fall 1956, taught by thesis adviser William Paul. Paul described it as a semiempirical theory tool that had proved useful for reading the cyclotron resonance experiments reported by Dresselhaus, Kip and Kittel. Shockley had already outlined the perturbation technique in a 1950 Physical Review paper, though it is now nearly forgotten. In 1958, Harvey Brooks, who had returned to Harvard as dean of the Division of Engineering and Applied Physics, lectured on how k·p could predict and fit nonparabolicities of band extrema in semiconductors. He had recently discussed with Evan Kane at General Electric's Schenectady labs Kane's work on the nonparabolicity of InSb. The author found the talk a compelling use of quantum mechanics on practical problems. During the author's thesis work, optical measurements of semiconductor effective masses as they varied with temperature and carrier concentration called for theoretical explanation. With help from Paul and Brooks, the author saw how k·p could interpret that data simply.