TMS Logo

About the 8th Biennial Workshop on OMVPE: Technical Program, Monday Morning Sessions (April 14)



8TH BIENNIAL WORKSHOP ON ORGANOMETALLIC VAPOR PHASE EPITAXY
April 13-17, 1997 · Dana Point, California

The following papers will be presented at the 8th Biennial Workshop on OMVPE, on Monday morning, April 14th, 1997. The calendar of events describes the entire technical program.

SESSION CHAIR:
Ron Moon, Hewlett Packard Laboratories, Palo Alto, CA 94304-1392
Previous Session     Next Session

SESSION II: IN SITU MONITORING I

10:40 am

In Situ Spectroscopic Reflectance and Virtual Interface Analysis of MOCVD-Grown AlxGa1-xAs Alloys: W.G. Breiland, H.Q. Hou, B.E. Hammons, Sandia National Laboratories, Albuquerque, NM 87185-0601 and S.A. Chalmers, Pacific Lightwave, 10655 Roselle St., Suite G, San Diego, CA 92121

Much of the time and expense of MOCVD growth is spent in performing numerous calibration runs to fine-tune the recipe for an actual device structure growth. We have eliminated most of this effort by using a single-wavelength, in situ reflectance monitor coupled with a virtual interface analysis method for use as a pre-growth calibration tool. We have verified to three-digit precision that the growth rate of an AlxGa1-xAs alloy is the sum of the growth rates of the individual AlAs and GaAs contributions. Applications of the pre-growth technique on a commercial EMCORE MOCVD reactor have allowed us to grow AlxGa1-xAs with precise composition control over the 0<x<1 range after only a single one-hour calibration run. Recently, we have extended the in situ monitor capabilities to record a complete 512-point spectrum (700nm to 1000nm) every second during a growth run. Virtual interface analysis has been used to extract high temperature optical constants of AlxGa1-xAs alloys. Based on ex-situ composition determinations. The dispersion curves may be used to extract thickness and aluminum composition from in situ spectral reflectance data. This work was supported by the United States Department of Energy under Contract DE-AC04-94AL85000. Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United States Department of Energy.

11:00 am

Growth and In Situ Characterization of AlGaN/GaN on Sapphire: Kevin Killeen, M.J. Ludowise, W.H. Perez, S.D. Lester, Y-L. Chang, D. Lefforge and J.N. Miller, Hewlett Packard Laboratories, Palo Alto, CA 94304

Heterostructures of GaN/AlGaN are recognized as important components of nitride optoelectronic and electronic devices. In this paper we report on the deposition of AlxGa(1-x)N (O £ x £ 1) films on sapphire substrates in a multiwafer, vertical stagnation flow reactor using conventional reactants (TMGa, TMAl, NH3). The growth is monitored using normal incidence, in situ reflectometry which spatially samples the epilayer thickness and composition across the wafers positioned on the (rotating) susceptor. We have measured the AlGaN growth rates and compositions as a function of TMAl mole fraction for growth temperatures from 900 to 1050°C and growth pressures from 76 to 700 Torr. Optical constants at 633 nm have been determined at growth temperatures for AlGaN over the entire composition range. The AlGaN layers were also characterized by ex-situ photoluminescence (300°K), SIMS and X-ray diffraction.

11:20 am

Observation of Alloy Ordering in Real Time using In Situ X-ray Scattering: D.W. Kisker*, P.H. Fuoss**, G.B. Stephenson*, S. Brennan***, and S. Kurtz****, IBM Research Division*, Yorktown Heights, NY 10598; AT&T Bell Laboratories**, Murray Hill, NJ 07974; SSRL***, Stanford, CA 94309; NREL****, Golden, CO

The ordering of III-V semiconductor alloys such as InxGal1-xAs and InxGa1-xP has important implications for the resulting properties of devices which are fabricated from these materials. However, the detailed mechanism for the formation of these ordered materials is not yet understood. It has been proposed to be related to the presence of a reconstruction during growth, but, it has been difficult to verify this relationship directly. Furthermore, the typically proposed reconstruction, (2x4), has never been observed on the related GaAs (001) surface during OMVPE. Therefore, there seems to be an inconsistency between the proposed mechanisms and experimentally observed results. In this work, we have been able to directly monitor the formation of ordered InxGalxAs lattice matched to InP during OMVPE growth using in-situ x- ray scattering. Our results include the dependence on growth parameters such as temperature and growth rate. In addition, we have directly determined for the first time the surface reconstruction which is present on the InxGal xAs layer during OMVPE growth of this alloy. We will discuss the similarities and differences between the nature of the surface of InxGa1-xAs and the binary, GaAs.

11:40 am

Surface Ordering of GaInP: J.F. Geisz, W. McMahon, A.E. Kibbler, D.J. Friedman, and J.M. Olson, National Renewable Energy Laboratory, 1617 Cole Blvd., Golden, CO 80401

Alloy ordering of GaO.5In O.5P grown by organometallic vapor phase epitaxy (OMVPE) has long been known to be affected by growth conditions and substrate misorientation. A popular model for this ordering requires a 2x4 reconstruction of the (100) growth terrace, but few techniques have been applied to the study of these surfaces due to the relatively high pressures present during OMVPE growth. We have used reflectance difference spectroscopy (RDS) as an in-situ tool to observe the surface condition (i.e., bond anisotropy) of GaInP during growth on (100) substrates misoriented 6 toward (111)B and correlated these spectra with low energy electron diffraction (LEED) and scanning tunneling microscope (STM) images of these surfaces measured at room temperature in a UHV analytical chamber attached directly to the OMVPE system. We find that surfaces previously identified as "2x4like" from in situ optical probes (surface photoadsorption and RDS) are actually 2x1 reconstructed. This implies that a 2x4 reconstructed surface is not a necessary condition for surface ordering of GaInP, contrary to the above terrace model. Furthermore, the LEED spots are split, indicating the formation of evenly spaced monolayer steps, consistent with step-driven, ordering models. We have also examined GaInP surfaces on a variety of substrate misorientations prepared under various conditions. The implications of these observations will be discussed.

12:00 pm

Process Development on a Vertical Reactor with Stagnation Flow Geometry: Steve Hummel, Pamela Langhoff and Kevin Killeen, Hewlett Packard Laboratories, 3500 Deer Creek Rd., Palo Alto, CA 94303

A vertical reactor with stagnation flow geometry has been implemented for the growth of AlGaAs and AlGaInP materials. It has been demonstrated both numerically and experimentally that the stagnation flow geometry provides an expanded operational parameter space, as well as desirable uniformity, stability and scalability behavior. Such a geometry also provides a short and normal optical path to the wafer surface, permitting the advantages of insitu reflectance monitoring to be easily applied. This presentation will summarize the influence of chamber design on epitaxial growth behavior as observed during optimization of a 1x2" vertical reactor. The large operational parameter space in concert with in-situ reflectance monitoring led to the rapid process development of 780 nm laser devices. Results to date include AlGaAs-based DBRs with thickness uniformity better than 0.33%, controllable and reproducible carbon doping using CBr4, and 3 QW, 780 nm broadarea lasers with threshold current densities as low as 448 A/cm2. The discussion will include the effects of growth parameters on growth rate and uniformity of thickness and alloy composition.


The information on this page is maintained by TMS Customer Service Center (csc@tms.org).

Search TMS Meetings Page About TMS TMS OnLine