Comparison of two different isolation schemes for n-type InP by helium implantation

Abstract

N/sup +/ InP layers on semi-insulating InP were electrically isolated by helium implantation using two different schemes. An energy of either 55 keV or 600 keV was used to place the peak of the damage either in the n/sup +/ doped layer or deep inside the substrate, respectively. Helium isolation implants were performed at different doses ranging from 5/spl times/10/sup 11//cm/sup 2/ to 1/spl times/10/sup 16//cm/sup 2/. A gradual increase in the sheet resistivity from its pre-implanted value of 432 /spl Omega///spl square/ to a maximum value of 1.2/spl times/10/sup 7/ /spl Omega///spl square/ and 3/spl times/10/sup 6/ /spl Omega///spl square/ was obtained for 55 keV and 600 keV He/sup +/ implants respectively. Maximum sheet resistivity is obtained at a dose of 2/spl times/10/sup 14//cm/sup 2/ called the threshold dose for both isolation schemes. The effect of He+ dose on the semi-insulating (SI) InP substrate is also reported. The substrate becomes less resistive with increase in He/sup +/ dose. At a dose of 1/spl times/10/sup 15//cm/sup 2/, the data suggests that the sheet resistivity value for the 600 keV implants is due to the semi-insulating substrate itself and not from the n-type InP layer. This means that the maximum sheet resistivity measured for the 600 keV implants is not the optimum value. The evolution of sheet resistivity as a function of post-implant annealing temperature is also examined for a helium dose of 2/spl times/10/sup 14//cm/sup 2/ for both isolation schemes. Both the 55 keV and 600 keV implants show more or less the same trend of annealing behaviour. A broad thermally stable region with a sheet resistivity value in the order of /spl sim/10/sup 7/ /spl Omega///spl square/ persists even after post-implant annealing temperatures of /spl sim/500/spl deg/C for both isolation schemes.