Effects of Rapid Thermal Annealing on Indium Antimonide Bismide | AIChE

Effects of Rapid Thermal Annealing on Indium Antimonide Bismide

Mid-wave and long-wave infrared (MWIR and LWIR respectively) photodetection is of vital importance to a variety of applications including astronomical, medical, and military imaging [1]. Current photodetectors operating in this region utilize HgCdTe, a material which contains toxic mercury and often experiences low uniformities that can impede device performance [2]. InSbBi has emerged as a promising alternative to HgCdTe for MWIR and LWIR photodetectors due to the significant redshift in absorption which results from incorporating small amounts of Bi into InSb [1]. Still, InSbBi remains a relatively underexplored material.

This study looks at the effect of rapid thermal annealing (RTA) on InSbBi grown at low substrate temperatures via molecular beam epitaxy (MBE). Annealing is a common post-growth method used on highly mismatched alloys (HMMAs) to improve material quality, specifically for those grown at low substrate temperatures [3]. Early predictions expect InSbBi to be an HMMA. It is also involved in device formation. Many other dilute bismides have shown increases in photoluminescent (PL) intensity post-annealing [3][4][5]. On the other hand, the effect of annealing on the absorption spectra of dilute bismides has been variable. [3][4][5]. In this study, PL measurements were utilized to analyze changes in PL intensity and shifts in the absorption spectrum. Furthermore, as MBE-grown InSbBi is produced at low substrate temperatures, Bi droplets which can impede device performance tend to form. In this study, atomic force microscopy (AFM) is utilized to analyze the effect of annealing on the morphology of Bi droplets.

Annealing InSbBi between 100°C to 250°C was found to significantly decrease PL intensity, and annealing between 375°C to 475°C caused PL to no longer be observable. Furthermore, there were no significant shifts in absorption spectra. AFM measurements were conflicting, though it is possible that Bi droplet size slightly decreased with annealing. Still, droplets were clearly present following ever annealing temperature.

As annealing is a necessary step in device formation, it is valuable to know this step will result in a degradation in optical quality. Future processes for creating InSbBi devices should utilize annealing temperatures of 100°C or lower to prevent PL degradation. Furthermore, alternative methods for improving PL intensity which could compensate can be devised. Additionally, as annealing does not have a significant effect on Bi droplet size, alternatives such as wet and dry etching, should be explored to increase the potential of functional InSbBi-based devices.