Are the Materials Properties of Indiumnitride Dominated by Defects?
- PDF / 437,159 Bytes
- 8 Pages / 612 x 792 pts (letter) Page_size
- 68 Downloads / 200 Views
0994-F02-01
Are the Materials Properties of Indiumnitride Dominated by Defects? Petra Specht, William Hong, and Eicke R. Weber Dept. of Mat.Sci.&Eng., UC Berkeley, and Mat.Sci.Div., LBNL, Berkeley, CA, 94720
ABSTRACT Indium nitride (InN) is a promising yet technologically challenging material with a high defect density and unusual material properties. Its high electron mobility may be utilized in high power electronic devices, and its high absorbance and low energy optical response make it a promising candidate for multi-junction, high-efficient solar cell technology. Studies of absorption and photoluminescence optical response of epitaxial InN resulted in a large correction of the fundamental bandgap from the originally proposed 1.9 eV to now below 0.7 eV. Yet, it is still debated if the commonly measured optical transitions below the original high bandgap values are actually caused by a large concentration of defects, on the order of 1020/cm3, rather than reflecting a low fundamental bandgap. Many applications of this material, e.g. in high-efficient solar cell technology, are primarily dependent on the successful production of a contacted p-n junction, which has not yet been achieved. This contribution addresses the controversy in the bandgap discussion of InN. Valence electron energy loss spectroscopy (VEELS) of InN allows spatially resolved analysis of the density of states in the transmission electron microscope (TEM). Standard optical characterization is compared with TEM results. INTRODUCTION The fundamental bandgap of (Al,Ga,In)ñ(N,P,As,Sb) compounds is reduced when a group-III element or a group-V element is replaced with one of a higher atomic mass Z. In 2001, when a correction of the bandgap of InN from the original 1.9 eV to ~0.7 eV was proposed, InN was named as the only exception to this rule (Figure 1). In order to explain the exceptional status of only one III-V compound, various theories were developed which describe this assumed uniqueness of InN [1], [2]. In most publications which support the low bandgap model of InN [3], [4] it is not mentioned that ìadjustingî the bandgap also requires changing the original bandgap of most InGaN alloys (Figure 2). It is plausible that the increase in crystal quality of epitaxial InN with the availability of molecular beam epitaxy (MBE) materials could have resulted in more accurate measurements of its fundamental bandgap. However, this argument does not apply to InGaN alloys with indium content of 40% or less, which were already available in high crystal quality in 1995 when Nakamura et al. published high bandgap results for InGaN [5]. Between 1995 and 2001 the optical bandgap in InGaN alloys constantly shifted towards lower energy values (see for example [6], [7], [8]) until finally a minimum was reached (Figure 2, intermediate energy values are omitted for clarity). Correspondingly, the bowing parameter of the InGaN system was calculated to an unusually high value if 1.9 eV was assumed as the fundamental bandgap of InN (for example see [8]: 3.68 eV and [9]
Data Loading...
