L Esaki R Tsu Communication Superlattice and Negative Differential Conductivity in Semiconductors Abstract: We consider a one-dimensional periodic potential, r “superlattice,” in monocrystalli ne semiconductors formed y a periodic variation o f all oy c omposi tion or of impurity density introduced during epitaxial growth. f the period of a superlattice, of the order o f 100A, is shorter than the electron mean free pat h, a seri es of narro w allowed and forbi dden bands s expec ted due o the subdivision of the Brill ouin zone nt o a series of minizone s. f the scattering time f electrons meets threshold condition, the combined effect f the energy and momentum beyond an inflection point in the -k relati on; this results in negative differentia l conductance in the direction of the superlatti ce. The study f superlattices and observations of quantum mechanical effects on a new physical scale may provide a valuable area o f investigation in the fieId of semiconductors. Introduction We consider theoretically a one-dimensional periodic po- tential, or “superlattice,” in monocrystalline semiconduc- tors. This superlattice potential would be obtained by a periodic variation of alloy composition or of impurity density introduced during epitaxial growth. This technique would enable one to vary arbit rarily t he amplitude and periodicity of the superlattice potential over a range of values, although one period probably could not be made much shorter than lOOA (about 20 times as long as the lattice constant of the host crystal). If this distance, which is comparable to the junction width in a tunnel diode,’ is shorter han he electron mean ree path , ne may expect to observe stro ng energy dispersion effects in the proposed structure. These effects would allow observation of familiar quan tum mechanical properties in a new domain of physical scale, due o very narrow allowed and for- bidden energy bands associated with a series of minizones in the Brillouin zone, not seen in the host crystal. It should be poss ible to obta in a novel class of man-made semicon- ductor mater ials, a t least as far as electr onic properties are concerned, and one expec ts the prope rties t o depend not only on the band parameters of the host crystal, but also on the char acteristics of the superlattice. We have analyzed the dynamics of conduction electrons in a superlatti ce structure which, we think, is realizable with techniques described here. Although the one-dimen- Yorktown Heights, New York 10598. North Carolina under Co ntract DAHC04-69-GO069. The authors are located at the IBM Thomas J. Watson Research Center, * Research sponsored i n part by the Army Research Off ice, Durham, sional lattice per se is an eleme ntary subj ect, the results contain important implications for the direction of experi- mental effort. We have found that, in the direction of the superlattice (perpendic ular t o the superlattice pla nes), the narrow wave-vec tor zon es and the narr ow ener gy bands make it possible for electrons to be excited beyond the energy corresponding to an E-vs.-k inflection point with moderate electric fields. The resulting negative conduc- tance could lead t o new ultra-hig h-speed device s.? These device s would have virtually no frequency limi tation except when the energy quantum for the fre quen cy involved is a significant fraction of the width of the narr ow nergy band. Since the potentials envisioned are small compared with band gap energies of the host semiconductors, a nd since the properties depend on a sustained periodic variation, the structure should be viewed as a perturbed bulk crystal rather than as a series of junctions. Materials The achievement of well-defined superlattice structure with a period of, say, lOOA will require considerable effort even with the use of the most advanced epitaxial thin-film technologies. Th e materials should be well-known semi- conductors and their alloys; for examples, Ge, Si, Ge-Si alloys, 111-V compounds and their alloys, 11-VI com- semiconductors for a negative mass amplifier. wherein transverse effective t H. Kromer proposed using the heavy hole band in Ge, Si and other masses were said to become negative for excited electro ns actually ho les) [Phys. Reu. 109, 1856 195811. Application of the effect, however, has not turne d out to be practical. 61 SUPERLATTICE NEGATIVE DIFFERENTIAL CONDUCTIVITY
5
Embed
Superlattice and Negative Differential Conductivity in Semiconductors
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
8/9/2019 Superlattice and Negative Differential Conductivity in Semiconductors