916 IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS, VOL. 3, NO.
3, JUNE 1997
Advances in Selective Wet Oxidation
of AlGaAs Alloys
Kent D. Choquette, Member, IEEE, Kent M. Geib, Carol I. H. Ashby, Ray D. Twesten,
Olga Blum, Hong Q. Hou, Member, IEEE, David M. Follstaedt, B. Eugene Hammons,
Dave Mathes, and Robert Hull
(Invited Paper)
Abstract— We review the chemistry, microstructure, and
processing of buried oxides converted from AlGaAs layers
using wet oxidation. Hydrogen is shown to have a central
role in the oxidation reaction as the oxidizing agent and to
reduce the intermediate product As2 O3 to As. The stable oxide
is amorphous (Alx Ga10x )2 O3 which has no defects along the
oxide/semiconductor interfaces but can exhibit strain at the oxide
terminus due to volume shrinkage. The influence of gas flow,
gas composition, temperature, Al-content, and layer thickness on
the oxidation rate are characterized to establish a reproducible
process. Linear oxidation rates with Arrhenius activation
energies which strongly depend upon AlAs mole fraction are
found. The latter produces strong oxidation selectivity between
AlGaAs layers with slightly differing Al-content. Oxidation
selectivity to thickness is also shown for layer thickness <60 nm.
Differences between the properties of buried oxides converted
from AlGaAs and AlAs layers and the impact on selectively
oxidized vertical cavity laser lifetime are reported.
Index Terms—Materials processing, materials science and tech-
nology, optoelectronic device, semiconductor device fabrication,
semiconductor lasers.
I. INTRODUCTION
T HE FORMATION of a robust native oxide on silicon is
the foundation of Si integrated circuit technology which
has spawned multibillion dollar markets. The attributes of SiO
which set it apart from the native oxides of other semicon-
ductors include its high density, low interface state density,
mechanical stability, and insulating properties [1]. Obtaining
an analogous native oxide on III–V compound semiconductors
such as GaAs has been pursued for decades but has proven
elusive. Among other issues, these native oxides tend to be
chemically and mechanically unstable and nonuniform [2], [3].
However, the wet oxidation of Al-containing semiconductors
such as AlGaAs at elevated temperatures was discovered
by researchers at the University of Illinois to form to a
phase of Al O which is mechanically stable and has a low
Manuscript received June 19, 1997; revised June 26, 1997. This work
was supported by the U.S. Department of Energy under Contract DE-AC04-
94AL85000.
K. D. Choquette, K. M. Geib, C. I. H. Ashby, R. D. Twesten, O.
Blum, H. Q. Hou, D. M. Follstaedt, and B. E. Hammons are with the
Center for Compound Semiconductor Science & Technology, Sandia National
Laboratories, Albuquerque, NM 87185 USA.
D. Mathes and R. Hull is with the Department of Materials Science,
University of Virginia, Charlottesville, VA 22903 USA.
Publisher Item Identifier S 1077-260X(97)07196-7.
1077–260X/97$10.00  1997 IEEE
Authorized licensed use limited to: Tampere University. Downloaded on January 04,2021 at 09:27:57 UTC from IEEE Xplore. Restrictions apply.
refractive index [4]. Although of limited impact to date on
microelectronics, this oxide has had significant impact on
optoelectronic devices, in particular edge-emitting lasers and
vertical-cavity surface-emitting lasers (VCSEL’s).
Wet oxidation of AlAs was first reported in 1979, where the
native (Gibbsite phase) oxide was formed at the low oxidation
temperature used (100 C) as evident by the greater thickness
of the oxide compared to the original AlAs layer [5]. Ten years
later during studies of atmospheric degradation of AlAs [6],
it was discovered that wet oxidation of Al-containing semi-
conductors (AlGaAs, AlInAs, AlInGaP, etc.) at temperatures
above 300 C produces a mechanically robust oxide with
a low refractive index and reduced thickness relative to the
unconverted semiconductor layer [4], [7]. The utility of this
oxide has been demonstrated in a variety of edge emitting
laser structures [8]–[10] where the oxide layers provide both
index-guiding [11] and buried current apertures [12].
Dramatic performance advances have been achieved in
selectively oxidized VCSEL’s which incorporate buried ox-
ide apertures for efficient electrical and optical confinement.
Oxide-confined VCSEL’s have used oxide apertures under a
dielectric distributed Bragg reflector to produce a significant
reduction of the threshold current [13]. Monolithic oxide-
confined VCSEL’s exploit the oxidation selectivity of the
low refractive index (high Al-content) layers within a semi-
conductor distributed Bragg reflector to form one or more
buried oxide aperture(s) [14], [15]. These VCSEL’s have
demonstrated the lowest threshold current [16] and voltage
[14] as well as record high power conversion efficiency [17],
[18]. The improved electrical confinement afforded by buried
oxide apertures is evident in the lower threshold voltage of
selectively oxidized VCSEL’s as compared to other device
structures [19]. Furthermore, because the refractive index is
reduced from 3.0 for the original AlGaAs layer to 1.6 for
the oxidized layer [11], index-guiding optical confinement
is obtained as evident from the low threshold current [13],
[16] and the VCSEL emission characteristics [20], [21]. The
low refractive index of the oxide has also been exploited
to fabricate VCSEL’s with high index contrast mirrors com-
posed of GaAs–Al O layers [22]. Other applications of the
AlGaAs oxide have included birefringent optical waveguides
[23], photonic lattices [24], buried microlenses [25], and
metal–insulator–semiconductor transistors [26], [27].
CHOQUETTE et al.: ADVANCES IN SELECTIVE WET OXIDATION OF AlGaAs ALLOYS 917

Fig. 1. Apparatus for reproducible wet oxidation of AlGaAs alloys that

employs mass flow gas controllers, a constant temperature water bubbler,
and a three zone furnace.

Understanding the chemistry, microstructure, and process-

ing of buried oxides converted from AlGaAs is essential
for the development of a robust fabrication technology for (a)
optoelectronic and microelectronic devices. We review the
present understanding of selective oxidation of AlGaAs, par-
ticularly as applied to the fabrication of selectively oxidized
VCSEL’s. In Section II we describe our oxidation system,
the sample structures, and experimental methods used to
characterize buried oxides. The oxidation chemistry and oxide
microstructure are discussed in Section III. In Section IV we
examine the oxidation rate dependencies, including the effects
of gas flow, steam partial pressure, temperature, Al-content,
and layer thickness. In addition, the differences between the
properties of buried oxides converted from AlGaAs and AlAs
layers, including the implications on VCSEL lifetime, are
discussed. Conclusions and future work are finally presented
in Section V.

II. EXPERIMENT
The wet oxidation of AlGaAs is a relatively simple
procedure: expose high Al-content layers to water vapor
transported in an inert gas within an elevated temperature
(350 C–500 C) environment [7]. However, establishment of
a stable and reproducible oxidation process requires careful
control of several process parameters. In Fig. 1, we sketch our
oxidation system which incorporates mass flow controllers,
a stable water vapor source, and a three-zone 4-in-diameter
tube furnace. The steam is supplied by bubbling N or other
inert gas through a 2-l flask of deionized water, which in turn
is immersed within a constant-temperature bath maintained at
90 C. The gas passes through the bubbler and into a three
zone furnace through tubes heated to avoid condensation. As
will be shown in Section IV, the stringent control of gas flow,
bubbler water temperature, and furnace temperature enables a
stable and reproducible oxidation process.
The oxidation samples are composed of GaAs–AlGaAs
multilayers [typically VCSEL epitaxial material as sketched in
Fig. 2(a)] [14], [15], which are grown by metal–organic vapor
phase epitaxy (MOVPE). MOVPE with in situ monitoring has
the advantages of easy accessibility of the complete AlGaAs
alloy range and stringent compositional control and uniformity
[28]. Mesas of various geometries and sizes are defined and
etched by reactive ion etching to expose the AlGaAs layers
at the mesa sidewall for lateral oxidation. The samples are
typically etched immediately before oxidation, although delays
between etching and oxidation of AlGaAs do not influence the
(b)
Fig. 2. (a) Cross section sketch of selectively oxidized VCSEL samples
composed of GaAs–AlGaAs multilayers and buried oxide layers. The highest
Al-containing layers on each side of the optical cavity extend further into the
mesa than the other oxidized layers, forming an aperture at the mesa center.
(b) Mesa top view under infrared illumination showing extent of buried lateral
oxidation, where the central darker region is the unoxidized aperture.
oxidation rates. The oxidation furnace is allowed to equilibrate
with carrier gas and steam flowing before each oxidation
run. The oxidation samples are placed on a large preheated
platen which is inserted into the furnace. The composition of
the epitaxial layers vary such that one or more layers will
oxidize laterally within the mesa to a greater extent than the
surrounding layers as depicted in Fig. 2(a) [14]. The refractive
index difference between the oxidized and unoxidized region
is apparent under visible or infrared illumination as shown in
Fig. 2(b). The lateral extent of the oxide is thus measured from
the edge of the square mesas, averaging in both directions to
minimize error.
The oxides are also examined by Raman scattering spec-
troscopy and cross section transmission electron microscopy
(TEM). Raman spectra are measured using an excitation
wavelength of 514.5 nm with a 500-mW probe beam line
focused to apply less than 85 W/cm on the sample. Two
methods are used to prepare TEM samples. Some samples
are thinned by a combination of mechanical polishing and
Ar–ion milling with the layers held in compression during

Authorized licensed use limited to: Tampere University. Downloaded on January 04,2021 at 09:27:57 UTC from IEEE Xplore. Restrictions apply.
918 IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS, VOL. 3, NO. 3, JUNE 1997
Fig. 3. Raman spectra of partially oxidized Al0:98 Ga0:02 As layer at (a) the
beginning and (b) the end of the oxidation process showing AlGaAs, As2 O3 ,
and As peaks.
polishing to prevent delamination of the oxidized layers from
the surrounding semiconductor. Cross section TEM samples of
selectively oxidized VCSEL’s are also prepared using focused
ion beam etching [29] in order to image specific locations such
as the oxide terminus. TEM images were formed using 200
keV electrons, where care is taken to minimize beam-induced
oxide crystallization and densification during electron beam
exposure [30], [31].
III. OXIDE CHEMISTRY AND MICROSTRUCTURE
To elucidate the oxidation chemistry, planar oxide films
have been characterized by Raman spectroscopy [32],
[33]. Raman spectra obtained from a planar oxidized
Al Ga As layer at two different times during oxidation
are shown in Fig. 3. The prominent feature at 220 cm
is the superposition of several peaks arising from crystalline
As (198 and 257 cm ) and amorphous As (227 cm ).
The broad peak centered at 475 cm is identified as an
intermediate reaction product, amorphous As O . Notice that
the AlAs-like phonon peak intensity at 400 cm in Fig. 3(b)
decreases with oxidation time as the AlGaAs is consumed.
However, there is a relatively constant ratio between the signal
intensities for As and As O in Fig. 3(a) and (b), showing the
formation and loss of each of these products reaches a steady
state during oxidation.
To form Al O the oxidation state of Al is made more
positive by oxidation, and thus the oxidation state of another
constituent in the reaction must become more negative, i.e.
it must be reduced. In dry oxidation, this is accomplished
by the change from zero-valent O or O atoms to O .
However, for a wet oxidation process that uses steam in an
inert carrier gas, the oxygen in water is already in the 2
oxidation state. Consequently, the agent for removing electrons
(the oxidizing agent) from AlAs to produce Al O is H
from water. Moreover as discussed below, the resulting zero-
Authorized licensed use limited to: Tampere University. Downloaded on January 04,2021 at 09:27:57 UTC from IEEE Xplore. Restrictions apply.
valent H can produce volatile As species by reduction of the
intermediate As O .
The following reactions [33] are both consistent with the
observed products during wet oxidation of AlAs and have
thermodynamically favorable Gibbs free energies at the tem-
peratures employed [34]:
2 AlAs 6 H O(g) Al O As O 6H
473 kJ/mol (1)
As O 3H 2 As 3 H O(g)
131 kJ/mol (2)
As O 6H 2 As 3 H O(g)
1226 kJ/mol (3)
Note that replacing AlAs with GaAs in (1) produces
10 kJ/mol. This indicates that exchanging Ga for Al in
Al Ga As will make (1) less favorable and thereby retard
wet oxidation as the Ga content increases [see Fig. 8(a)].
Conversion of As O to As produces the relatively constant
As O Raman intensity observed in Fig. 3. Since elemental
As is detected in the oxidized film [32], [35], the relatively
constant As Raman intensity in Fig. 3 reflects a balance
between formation by (2) and (3) and loss of a portion of
the As. Although arsine formation using H is unfavorable
77 kJ/mol), reaction of As with atomic hydrogen
is favorable 471 kJ/mol). The hydrogen produced
by Al oxidation will not initially be H , but should be weakly
bound to the semiconductor, allowing transformation from As
to AsH to occur. Either As or AsH could serve as the volatile
species for removal of As from the oxidized film.
Two other experiments indicate the central role that H plays
in the wet oxidation reaction [33]. First, the greater ability of
atomic hydrogen rather than H to convert As O to As has
been demonstrated in annealing studies of partially oxidized
AlGaAs films. Although annealing partially oxidized samples
in forming gas H –Ar produces little change in the Raman
spectra (see Fig. 3), an anneal of the same sample for 2 h at
400 C in atomic hydrogen (generated using a hot W filament)
produces a 10% reduction in the As O signal intensity and
a concomitant 10% increase in the As signal. Secondly, a
reaction mechanism that relies on the participation of H should
be susceptible to a competing reaction that will consume it.
For example, molecular oxygen would react with and remove
H and thus the presence of O should suppress (2) and (3).
We show in Section IV that oxidation using O as the carrier
gas inhibits the wet oxidation of AlGaAs [see Fig. 6(b)]. This
is consistent with a rapid consumption of atomic hydrogen
to form water 1148 kJ/mole O before it can
reduce the As O to As.
The microstructure of the oxide obtained from TEM analysis
is shown in Fig. 4, and oxide apertures within VCSEL’s are
shown in Fig. 5. The microstructure of oxidized Al Ga As
for 0.8 is found to be independent of both Al
content and thickness [31]. Electron diffraction patterns from
these oxides indicate that the oxide is an amorphous solid
solution of Al Ga O . Previous studies have identified
the oxide as - Al Ga O [36], [37]. The discrepancy
CHOQUETTE et al.: ADVANCES IN SELECTIVE WET OXIDATION OF AlGaAs ALLOYS 919

(a)

(a)

(b)

(b)
Fig. 4. Cross section transmission electron microscopy bright field images
of (a) oxidized layer converted from 100–nm-thick AlAs surrounded by layers
of GaAs; and (b) oxide/semiconductor interface in high resolution.

arises from the transformation of the amorphous oxide to
polycrystalline phase either by annealing after oxidation
[37] or by electron beam induced crystallization during TEM
examination [31]. The bright field image in Fig. 4(a) reveals
only granular amorphous contrast within the oxide, with
a 17-nm-thick interface zone at the oxide/semiconductor
interface. The different amorphous contrast of this thin zone
may arise from the presence of As O , corresponding to a
region rich in the intermediate product near the reaction front.
No extended defects are found along the oxide/semiconductor
interface in Figs. 4 or 5 [15]. As seen in the high-resolution
image of Fig. 4(b), a smooth transition over several lattice
constants is apparent between the oxide and semiconductor
[31], [38]. Note that the formation of interface voids at the
oxide/semiconductor interface due to oxide densification has
been observed under persistent electron beam exposure [30],
[31]. Interface voids formed during TEM examination are
apparent in the images of oxide apertures within VCSEL’s
shown in Fig. 5.
The thickness of an oxide formed at temperatures above
300 C tends to be reduced compared to the original semicon-
ductor layer [4], [7]. For example, the volume per Al atom in
AlAs is (3.57 Å , while in -Al O it is (2.85 Å , which
theoretically corresponds to a 20% linear contraction of -
Al O layers. Experimentally, the linear shrinkage is measured
(c)
Fig. 5. Cross section transmission electron microscopy images of VC-
SEL oxide aperture terminus (see arrows) formed from: (a) 84-nm-thick
Al0:98 Ga0:02 As with no defects apparent along the oxide/semiconductor
interface or at the oxide terminus[ = (311)]; (b) 84-nm-thick AlAs showing
evidence of strain at the oxide terminus [ = (400)]; and (c) 14-nm-thick
Al0:98 Ga0:02 As embedded within a 160-nm-thick Al0:92 Ga0:08 As layer
resulting in a tapered profile at each oxide terminus.
to be 12 to 13% in oxide layers formed from AlAs [38],
[39]; by comparison, shrinkage for oxidized Al Ga As
layers is measured to be 6.7% [37]. Although the Ga–O bond
length is greater than the Al–O bond, it does not fully account
for the reduced shrinkage of oxidized AlGaAs layers. The
reduced thickness of the oxide can result in strain fields at
the oxide terminus as manifest in the TEM images of the
AlAs oxide apertures within the VCSEL shown in Fig. 5(b).
In contrast, oxide apertures formed from Al Ga As do
not exhibit evidence of strain as depicted in Fig. 5(a) [40].
Further differences between the properties of oxides formed
from AlAs versus AlGaAs are discussed below.
IV. OXIDATION PROCESSING
A. Oxidation Rate Dependencies
To develop a manufacturable AlGaAs wet oxidation process,
the influence of process parameters such as gas flow, gas

Authorized licensed use limited to: Tampere University. Downloaded on January 04,2021 at 09:27:57 UTC from IEEE Xplore. Restrictions apply.
920 IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS, VOL. 3, NO. 3, JUNE 1997
(a)
(b)
Fig. 6. (a) Oxidation rate of buried 84-nm-thick Al0:98 Ga0:02 As layers as
a function of N2 and steam flow through furnace at 440  C. Note sufficient
flow insures a vapor saturated regime. (b) Lateral oxidation extent of buried
84-nm-thick Al0:98 Ga0:02 As layers using various carrier gases.
composition, temperature, Al composition, and layer thickness
on the rate of oxidation have been characterized [41]. Under-
standing the dependencies of the oxidation rate will dictate the
degree of control of these parameters required in an oxidation
system, as well as the compositional control and uniformity
required in the epitaxial structures, to achieve a reproducible
fabrication process.
In Fig. 6(a), we show the oxidation rates obtained from 84-
nm-thick Al Ga As layers as a function of N gas flow
bubbled through deionized water at two bath temperatures
[41]. For sufficient gas flow in Fig. 6(a), the system is in
a water vapor saturated regime, which insures the oxidation
reaction is not reactant limited. Specifically for our 4-in
diameter furnace, 3 l/min of N is sufficient to maintain a
constant oxidation rate. Moreover, the oxidation rate increase
with increasing bath temperature apparent in Fig. 6(a) follows
Authorized licensed use limited to: Tampere University. Downloaded on January 04,2021 at 09:27:57 UTC from IEEE Xplore. Restrictions apply.
from the increase of the steam partial pressure. Estimates of
the water partial pressure indicate that the oxidation reaction is
linear in water concentration. Note that the water temperature
in the bubbler (at constant bath temperature) is found to
vary with changing gas flow, reinforcing the importance of
maintaining a constant gas flow [41]. In Fig. 6(b), we show
a comparison of wet oxidation using various carrier gases
bubbled through water. The oxidation rates vary only slightly
with N , Ar, or forming gas (3% H N as the carrier gas.
In contrast, O bubbled through water completely suppresses
the AlGaAs oxidation, consistent with the consumption of
hydrogen as discussed in Section III.
The oxidation temperature dependence is presented in Fig. 7
[42]. The extent of the lateral oxidation of an Al Ga As
layer is plotted in Fig. 7(a). The lateral oxidation has a
temperature-dependent linear oxidation rate without an in-
duction time preceding the onset of oxidation [15]. Linear
oxidation rates have also been reported by other researchers
[5], [43], [44]. Using a model of silicon oxidation [45] the
oxidation thickness, ox , achieved in a time is
(4)
where is related to the diffusion constant of the reactants
moving through the oxide and is related to the oxidation
reaction rate constant and/or the rate of reactant supply at the
oxide boundary. In the limit of short oxidation times and/or
thin oxide thickness, (4) yields linear growth:
(5)
The linear oxidation rates apparent in Fig. 7(a) indicate
that the lateral oxidation of AlGaAs, over the range of
350 C–500 C and for lengths 50 m, is reaction rate
limited rather than determined by the diffusion of water vapor
through the oxide to the reaction front. For greater oxidation
lengths or higher temperatures, parabolic growth rates may
arise [46], [47]. We observe a lack of an induction time before
oxidation, even for samples that are pre-etched and exposed
to the atmosphere prior to oxidation. For example, the open
circles in Fig. 7(a) at 440 C, correspond to samples that
were etched 70 days before oxidation. Hence, controlled and
reproducible oxidation of AlGaAs can be attained for device
fabrication.
The oxidation rates in Fig. 7(b) depend exponentially on
temperature and obey an Arrhenius relationship [42]. Thus,
stable temperature control is necessary to insure reproducible
oxidation. Plotted in Fig. 7(c) are the oxidation activation
energies derived from the data in Fig. 7(b) for various AlGaAs
alloys. The extremely strong compositional dependence of
the oxidation rates results from the compositional dependence
of the activation energies. This is specifically illustrated in
Fig. 8(a) where the oxidation rate of Al Ga As for
varying from 1 to 0.84, changes by more than two orders of
magnitude [14]. Thus, a high degree of oxidation selectivity
between AlGaAs layers can be obtained with only a small
change in Al concentration. It is this oxidation selectivity
which can be exploited for fabrication of buried oxide layers
within a device, such as a VCSEL: with minute changes of Ga
CHOQUETTE et al.: ADVANCES IN SELECTIVE WET OXIDATION OF AlGaAs ALLOYS 921

(a) (b)

(c)
Fig. 7. (a) Lateral oxidation extent of buried 84-nm-thick Al0:98 Ga0:02 As layers versus temperature. The open circles at 440  C are samples that were
=
etched and exposed to the atmosphere 70 days before oxidation. (b) Alx Ga10x As oxidation rate of 84-nm-thick layers versus temperature for x 1.0, 0.98,
and 0.92. (c) Arrhenius activation energy for the oxidation reaction of Alx Ga10x As versus Al composition.

concentration we can select out a specific or multiple AlGaAs
layer(s) to form buried oxide layers for electrical and optical
confinement [see Fig. 2(a)]. However, the data in Fig. 8(a) also
indicates that stringent compositional control and uniformity
is critical for a manufacturable fabrication technology. Using
the oxidation system shown in Fig. 1, we have measured the
variation of the Al Ga As oxidation rate to be 2%
across a quarter of a 3-in wafer [41]. It is interesting to
note that from Fig 8(a), we estimate that a 2% oxidation
nonuniformity of Al Ga As corresponds to an AlAs mole
fraction variation of 0.001.
The thickness of the oxidized layer will influence the
oxidation rate [48], [49]. Fig. 8(b) shows the oxidation rate
of AlAs layers (surrounded by GaAs layers) with thicknesses
varying from 10–250 nm. For thickness 60 nm, a relatively
constant lateral oxidation rate is observed; for thinner layers
the oxidation rate dramatically decreases. For example, the
oxidation rate of a 10-nm-thick AlAs layer is reduced by
two orders of magnitude compared to a 100-nm-thick layer
in Fig. 8(b). Comparing Fig. 8(a) and (b), it is obvious that
the oxidation rate dependence on thickness can compensate
for the compositional dependence: thin layers of AlAs may
oxidize slower than thick layers of AlGaAs. Note that linear
oxidation rates are still observed for oxidation of thin layers,
indicating that decreased reactant transport through a thin layer
is not responsible for the reduced oxidation rate. Variations of
the surface energy at the oxide terminus [49] or changes of
strain with thickness will influence the oxidation rate.
Finally, the composition of the surrounding layers will affect
the oxidation rate and shape of the oxide terminus through
diffusion and/or supply of reactants [50], [51]. For example,
the thickness of a GaAs barrier between Al Ga As and
Al Ga As has been shown to influence the oxidation
rate of the latter by mediating the diffusion of reactants [50].

Authorized licensed use limited to: Tampere University. Downloaded on January 04,2021 at 09:27:57 UTC from IEEE Xplore. Restrictions apply.
922 IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS, VOL. 3, NO. 3, JUNE 1997

Furthermore, a rapidly oxidizing layer can supply reactants

to its surrounding layers such that vertical oxidation occurs
leading to an enhanced effective lateral oxidation rate as well
as a tapered profile at the oxidation terminus [51]. Fig. 5(c)
shows two tapered oxide apertures each composed of a 14-nm-
thick Al Ga As layer embedded within a 160-nm-thick
Al Ga As layer. The lateral oxidation rate of the thicker
Al Ga As layer is enhanced by a factor of 2 over the
rate shown in Fig. 8(a). Therefore to determine the oxidation
rate and profile of buried layers of AlGaAs, composition and
thickness of the intended oxide layer and the surrounding
layers must all be taken into account.

B. Oxide Composition Effects

Wet oxidation of AlAs has been commonly used for many
device demonstrations. The use of the binary AlAs rather than
AlGaAs alloys as an oxidation layer obviously relaxes the (a)
compositional control required during growth. However, the
properties of the oxides formed from AlAs versus AlGaAs
are found to differ in terms of oxidation isotropy, mechanical
stability, and induced strain [40]. In the following we compare
the properties of AlAs and AlGaAs oxides employed within
selectively oxidized VCSEL’s.
Top views of light-emitting VCSEL’s using different
Al Ga As layers in the oxide aperture layers are shown
in Fig. 9 [40]. The illuminated region in each mesa center
corresponds to the unoxidized portion of the current aperture
which defines the laser cavity [see Fig. 2(a)]. Independent of
composition, the current aperture resulting from a square mesa,
regardless of its crystallographic orientation, tends to also be
square as shown in Fig. 9(a), implying isotropic oxidation
[14]. However, for high Al-content layers 0.94) a slight
crystallographic oxidation dependence can be observed from
circular mesas. For example, Fig. 9(b) shows an approximately
square aperture results from a circular mesa when oxidizing
AlAs. The sides of the square oxide aperture which is formed (b)
Fig. 8. (a) Alx Ga10x As oxidation rate at 420  C versus composition for
are aligned along the crystal axes. Hence, the oxidation
along the crystal axes is slightly faster than along 100-nm-thick layers surrounded by 100-nm-thick GaAs layers; (b) AlAs
the axes. This crystallographic oxidation anisotropy is oxidation rate at 400  C versus AlAs layer thickness.
consistent with the lower surface reactivity of planes
as compared to in GaAs [52]. Notice for 0.92 cycling is particularly insidious during other post-oxidation
in Al Ga As, each Al atom on average will have one Ga fabrication steps which require temperatures of 100 C (e.g.,
atom as its second nearest neighbor, implying a different bond photoresist baking, polyimide curing, etc.). The difference in
configuration for the Al atoms in the ternary versus the binary mechanical stability apparent in Fig. 10 is consistent with
compound. The crystallographic reactivity of AlAs will thus the degree of strain observed in Fig. 5(a) and (b) for oxide
be modified as compared to AlGaAs. Isotropic oxidation is apertures formed from Al Ga As and AlAs, respectively.
verified in Fig. 9(c) which reveals a circular aperture from a Therefore, the mechanical instability of mesas containing AlAs
circular mesa using an oxide layer composition of 0.92. oxide layers likely arise from the strain observed at the AlAs
The mechanical stability of mesas which contain buried oxide terminus. The addition of a small amount of Ga to the
oxide layers are also found to depend upon the original layer oxidation layer and/or the use of thin AlAs layers (e.g., 20
composition [40]. Shown in Fig. 10 is a comparison of VCSEL nm) is found to mitigate this mechanical instability.
mesas which have been oxidized and then subjected to rapid Finally, the composition of the oxidized layer has been
thermal annealing to 350 C for 30 s. The mesas contain- shown to impact the reliability of selectively oxidized VC-
ing Al Ga As in the oxide layer shown in Fig. 10(a) SEL’s [40]. Preliminary reports using oxidized AlAs lay-
are unaffected by the anneal, while the mesas in Fig. 10(b) ers in VCSEL’s described device lifetimes of only several
which contain AlAs oxide apertures delaminate along the minutes [53]. More encouraging VCSEL reliability results
oxide/semiconductor interfaces. This sensitivity to thermal have been obtained by oxidizing Al Ga As layers [54].

Authorized licensed use limited to: Tampere University. Downloaded on January 04,2021 at 09:27:57 UTC from IEEE Xplore. Restrictions apply.
CHOQUETTE et al.: ADVANCES IN SELECTIVE WET OXIDATION OF AlGaAs ALLOYS
(a)
(b)
(c)
Fig. 9. Top view of a mesas showing illuminated oxide aperture with (a) an
Al0:98 Ga0:02 As oxidized layer in square mesa; (b) an AlAs current aperture
in a circular mesa; (c) an Al0:92 Ga0:08 As current aperture in a circular mesa.
In (a)–(c) the VCSEL current aperture is the light region in center surrounded
by a metal contact with a width of 15 m.
In Fig. 11(a), we show the aging characteristics of oxide-
confined VCSEL’s with various sizes that employ oxidized
Al Ga As current apertures. At a ambient temperature
of 80 C, the current required to maintain 1 mW output
power remains relatively unchanged for nearly 10 h, which
is comparable to the reliability of commercial ion implanted
VCSEL’s [55]. In Fig. 11(b), we show the aging characteristics
from VCSEL’s fabricated from two different wafers which
employ oxidized AlAs apertures [40]. Under more gentle aging
conditions (constant current operation at room temperature),
we nevertheless observe monotonic degradation of the output
power within tens of hours of operation. Note the strain field
Authorized licensed use limited to: Tampere University. Downloaded on January 04,2021 at 09:27:57 UTC from IEEE Xplore. Restrictions apply.
923
(a)
(b)
Fig. 10. Top view of VCSEL mesas after oxidation and rapid thermal
annealing to 350  C for 30 s. The mesas contain Alx Ga10x As oxide apertures
with: (a) x = 0.98, and (b) x = 1.0.
at the oxide terminus seen in Fig. 5(b) is within 100 nm of
the quantum-well active region, making it a potential source
of defects that could propagate into the VCSEL active region.
We conclude the addition of a small concentration of Ga into
the oxidized layer greatly enhances the laser lifetime, although
further study of the reliability and degradation mechanisms of
selectively oxidized VCSEL’s is warranted.
V. CONCLUSION
A basic understanding of the chemistry, microstructure,
and processing of the wet oxidation of AlGaAs alloys will
enable a robust fabrication technology. We have shown that
hydrogen has a central role in the oxidation reaction, serving
as the oxidizing species to form Al O and As O as well
as reduction of As O to As. This explains the orders-of-
magnitude faster oxidation of high Al-content semiconductors
with water in an oxygen-free carrier gas than with pure
oxygen. The oxide converted from AlGaAs is found to be
an amorphous solid solution of Al Ga O . For buried
oxide layers, no intrinsic defects or voids are detected at
the oxide/semiconductor interface, although a decrease in the
layer volume upon oxidation can produce strain fields around
the oxide terminus. Reproducible and linear oxidation rates
can be achieved by control of the gas flow, steam partial
924 IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS, VOL. 3, NO. 3, JUNE 1997
(a)
(b)
Fig. 11. Aging characteristics of various sizes of selectively oxidized VC-
SEL’s. (a) Current density required to achieve 1 mW output at 80  C
for VCSEL’s containing Al0:98 Ga0:02 As oxide apertures. (b) Output power
obtained at 25  C under constant current operation for VCSEL’s containing
AlAs oxide apertures.
pressure, and oxidation temperature in the oxidation apparatus.
The strong compositional dependence of the oxidation rate
is a consequence of the compositionally dependent Arrhenius
activation energies for AlGaAs oxidation, and is consistent
with the variation with composition of the Gibbs energy of
formation. In addition to layer composition, the oxidation rate
is dependent upon the layer thickness and the composition of
the surrounding layers. Finally, the properties of buried oxides
formed from AlGaAs alloys, rather than AlAs, are found to be
superior in terms of oxidation isotropy, mechanical stability
to thermal cycling, and VCSEL reliability. The latter two
properties apparently are due to the reduced strain observed
in VCSEL’s containing oxidized apertures of AlGaAs instead
of AlAs.
Authorized licensed use limited to: Tampere University. Downloaded on January 04,2021 at 09:27:57 UTC from IEEE Xplore. Restrictions apply.
Further work will be required to develop selective wet
oxidation of AlGaAs into a manufacturable technology. Al-
though we have developed a stable fabrication process, in
order to scale up to full wafer manufacture, issues pertaining
to reproducibility and uniformity obviously remain. Moreover,
devices which employ buried oxides such as VCSEL’s com-
monly require careful calibrations of each wafer to determine
the oxidation extent for device fabrication. Thus developing a
means to predefine the length of the oxide layer within a given
device structure would be advantageous. Finally, analogous
to process simulators available for Si oxidation, predictive
models of the wet oxidation of AlGaAs will be valuable. As
we have shown, simulating the oxidation rate and/or oxidation
profile requires a detailed understanding of the dependence
of the oxidation rate on the structure and composition of the
intended oxide layer and its surrounding layers. In spite of
these remaining challenges, the motivation to develop a wet
oxidation technology is clearly driven by the high performance
demonstrated from devices which incorporate these oxides,
such as selectively oxidized VCSEL’s. Therefore, selective
wet oxidation of AlGaAs alloys may play an important role in
the manufacture of future optoelectronic and microelectronic
devices.
ACKNOWLEDGMENT
The authors thank R. P. Schneider, Jr., K. L. Lear, M.
Haggerot Crawford, J. P. Sullivan, T. J. Drummond, and H. C.
Chui and acknowledge the technical assistance of J. Nevers,
S. Limary, J. Banas, and D. Moran. Sandia is a multiprogram
laboratory operated by Sandia Corporation, a Lockheed Martin
Company, for the U.S. Department of Energy.
REFERENCES
[1] E. Katz, “Oxidation,” in VLSI Technology, S. M. Sze, Ed. New York:
McGraw-Hill, 1983, ch. 4.
[2] C. W. Wilmsen, “Oxide/III-V compound semiconductor interfaces,” in
Physics and Chemistry of III–V Compound Semiconductor Interfaces, C.
W. Wilmsen, Ed. New York: Plenum, 1985, ch. 7.
[3] J. M. Dallesasse, N. El-Zein, N. Holonyak, Jr., K. C. Hsieh, R. D. Burn-
ham, and R. D. Dupuis, “Environmental degradation of Alx Ga10x As-
GaAs quantum well heterostructures,” J. Appl. Phys., vol. 68, pp.
2235–2238, 1990.
[4] J. M. Dallesasse, N. Holonyak, Jr., A. R. Sugg, T. A. Richard, and N. El-
Zein, “Hydrolyzation oxidation of Alx Ga10x As-AlAs-GaAs quantum
well heterostructures and superlattices,” Appl. Phys. Lett., vol. 57, pp.
2844–2846, 1990.
[5] W. T. Tsang, “Self-terminating thermal oxidation of AlAs epilayers
grown on GaAs by molecular beam epitaxy,” Appl. Phys. Lett., vol.
33, pp. 426–429, 1978.
[6] J. M. Dallesasse, P. Gavrilovic, N. Holonyak, Jr., R. W. Kaliski, D. W.
Nam, E. J. Vesely, and R. D. Burnham, “Stability of AlAs in AlGaAs-
AlAs-GaAs quantum well heterostructures,” Appl. Phys. Lett., vol. 56,
pp. 2436–2438, 1990.
[7] N. Holonyak, Jr, and J. M. Dallesasse, U.S. Patent 5 262 360, 1993.
[8] J. M. Dallesasse and N. Holonyak, Jr., “Native-oxide stripe-geometry
Alx Ga10x As-GaAs quantum well heterostructure lasers,” Appl. Phys.
Lett., vol. 58, pp. 394–396, 1991.
[9] S. A. Maranowski, F. A. Kish, S. J. Caracci, N. Holonyak, Jr., J.
M. Dallesasse, D. P. Bour, and D. W. Treat, “Native oxide defined
In0:5 (Alx Ga10x As)0:5 P quantum well heterostructure window lasers
(660 nm),” Appl. Phys. Lett., vol. 61, pp. 1688–1690, 1992.
[10] Y. Cheng, P. D. Dapkus, M. H. MacDougal, G. M. Yang, “Lasing char-
acteristics of high-performance narrow stripe InGaAs-GaAs quantum
well lasers confined by AlAs native oxide,” IEEE Photon. Technol. Lett.,
vol. 8, pp. 176–178, 1996.
CHOQUETTE et al.: ADVANCES IN SELECTIVE WET OXIDATION OF AlGaAs ALLOYS
[11] F. A. Kish, S. J. Caracci, N. Holonyak, Jr., J. M. Dallesasse, K. C. Hsieh,
M. J. Ries, S. C. Smith, and R. D. Burnham, “Planar native oxide index
guided AlGaAs-GaAs quantum well heterostructure lasers,” Appl. Phys.
Lett., vol. 59, pp. 1755–1757, 1991.
[12] S. A. Maranowski, A. R. Sugg, E. I. Chen, and N. Holonyak, Jr.,
“Native oxide top and bottom confined narrow stripe p-n AlGaAs-GaAs-
InGaAs quantum well heterostructure laser,” Appl. Phys. Lett., vol. 63,
pp. 1660–1662, 1993.
[13] D. L. Huffaker, D. G. Deppe, K. Kumar, and T. J. Rogers, “Native-oxide
defined ring contact for low threshold vertical-cavity lasers,” Appl. Phys.
Lett., vol. 65, pp. 97–99, 1994.
[14] K. D. Choquette, R. P. Schneider, Jr., K. L. Lear, and K. M. Geib,
“Low threshold voltage vertical-cavity lasers fabricated by selective
oxidation,” Electron. Lett., vol. 30, pp. 2043–2044, 1994.
[15] K. D. Choquette, K. L. Lear, R. P., Schneider, Jr., K. M. Geib, J. J.
Figiel, and R. Hull “Wavelength Insensitive Performance of Robust
Selectively Oxidized Vertical-Cavity Lasers,” Photon. Technol. Lett.,
vol. 7, pp. 1237–1239, 1995.
[16] G. M. Yang, M. H. MacDougal, and P. D. Dapkus, “Ultralow threshold
current vertical-cavity surface emitting lasers obtained with selective
oxidation,” Electron. Lett., vol. 31, pp. 886–888, 1995.
[17] K. L. Lear, K. D. Choquette, R. P. Schneider, Jr., S. P. Kilcoyne, and
K. M. Geib, “Selectively oxidized vertical-cavity surface emitting lasers
with 50% power conversion efficiency,” Electron. Lett., vol. 31, pp.
208–209, 1995.
[18] B. Weigl, M. Grabherr, G. Reiner, and K. J. Ebeling, “High efficiency
selectively oxidized MBE grown vertical-cavity surface emitting lasers,”
Electron. Lett., vol. 32, pp. 557–558, 1996.
[19] K. D. Choquette, K. L. Lear, R. P. Schneider, Jr., and K. M. Geib,
“Cavity characteristics of selectively oxidized vertical-cavity lasers,”
Appl. Phys. Lett., vol. 66, pp. 3413–3415, 1995.
[20] D. L. Huffaker, J. Shin, and D. G. Deppe, “Lasing characteristics of low
threshold microcavity lasers using half-wave spacer layers and lateral
index confinement,” Appl. Phys. Lett., vol. 66, pp. 1723–1725, 1995.
[21] K. L. Lear, K. D. Choquette, R. P. Schneider, Jr., and S. P. Kil-
coyne, “Modal analysis of a small surface emitting laser with a selec-
tively oxidized waveguide,” Appl. Phys. Lett., vol. 66, pp. 2616–2618,
1995.
[22] M. H. MacDougal, P. D. Dapkus, V. Pudikov, H. Zhao, and G. M. Yang,
“Ultralow threshold current vertical-cavity surface emitting lasers with
AlAs oxide-GaAs distributed Bragg reflectors,” IEEE Photon. Technol.
Lett., vol. 7, pp. 229–231, 1995.
[23] A. Flore, V. Berger, E. Rosencher, N. Laurent, S. Theilmann, N.
Vodjdani, and J. Nagle, “Huge birefringence in selectively oxidized
GaAs/AlAs optical waveguides,” Appl. Phys. Lett., vol. 68, pp.
1320–1322, 1996.
[24] M. J. Ries, E. I. Chen, and N. Holonyak, Jr., “Photopumped laser
operation of a planar disorder and native oxide defined AlAs-GaAs
photonic lattice,” Appl. Phys. Lett., vol. 68, pp. 2035–2037, 1996.
[25] O. Blum, K. L. Lear, H. Q. Hou, and M. E. Warren, “Buried refractive
microlenses formed by selective oxidation of AlGaAs,” Electron. Lett.,
vol. 32, pp. 1406–1408, 1996.
[26] E. I. Chen, N. Holonyak, Jr., and S. A. Maranowski, “Alx Ga10x As
metal-oxide semiconductor field effect transistors formed by lateral wa-
ter vapor oxidation of AlAs,” Appl. Phys. Lett., vol. 66, pp. 2688–2690,
1995.
[27] P. A. Grudowski, R. V. Chelakara, and R. D. Dupuis, “An
InAlAs/InGaAs metal-oxide-semiconductor field effect transistor using
the native oxide of InAlAs as the gate insulation layer,” Appl. Phys.
Lett., vol. 69, pp. 388–390, 1996.
[28] H. Q. Hou, H. C. Chui, K. D. Choquette, B. E. Hammons, W. G.
Breiland, and K. M. Geib, “Highly uniform and reproducible vertical
cavity surface emitting lasers grown by metalorganic vapor phase
epitaxy with in situ reflectometry,” IEEE Photon. Technol. Lett., vol.
8, p. 1285–1287, 1996.
[29] R. Hull, D. Bahnck, F. A. Stevie, L. A. Koszi, and S. N. G. Chu,
“Microscopic studies of semiconductor lasers utilizing a combination
of transmission electron microscopy, electroluminescence imaging and
focused ion beam sputtering,” Appl. Phys. Lett., vol. 62, pp. 3408–3410,
1993.
[30] F. A. Kish, S. J. Caracci, N. Holonyak, Jr., K. C. Hsieh, J. E. Baker,
S. A. Maranowski, A. R. Sugg, J. M. Dallesasse, R. M. Fletcher,
C. P. Kuo, T. D. Osentowski, and M. G. Craford, “Properties and
use of In0:5 (Alx Ga10x )0:5 As and Alx Ga10x As native oxides in
heterostructure lasers,” J. Electron. Mater., vol. 21, pp. 1133–1139,
1992.
[31] R. D. Twesten, D. M. Follstaedt, and K. D. Choquette, “Microstructure
and interface properties of laterally oxidized Alx Ga10x As,” in Vertical-
Authorized licensed use limited to: Tampere University. Downloaded on January 04,2021 at 09:27:57 UTC from IEEE Xplore. Restrictions apply.
925
Cavity Surface Emitting Lasers, K. D. Choquette and D. G. Deppe, Eds.,
Proc. SPIE, vol. 3003, pp. 55–61, 1997.
[32] C. I. H. Ashby , J. P. Sullivan, P. P. Newcomer, N. A. Missert, H.
Q. Hou, B.E. Hammons, M. J. Hafich, and A.G. Baca, “Wet oxidation
of Alx Ga10x As: temporal evolution of composition and microstructure
and the implications for metal–insulator–semiconductor applications,”
Appl. Phys. Lett., vol. 70, pp. 2443–2445, 1997.
[33] C. I. H. Ashby, J. P. Sullivan, K. D. Choquette, K. M. Geib, and H.
Q. Hou, “Wet oxidation of AlGaAs: The role of hydrogen,” J. Appl.
Phys., 1997.
[34] O. Kubaschewski, C. B. Alcock and P. J. Spencer, Materials Thermo-
chemistry. London, U.K.; Pergamon, 1993.
[35] Z. Liliental-Weber, M. Li, G. S. Li, C. Chang-Hasnain, and E. R. Weber,
“Structure of III–V oxides,” in Proc. Microscopy and Microanalysis, G.
W. Bailey, J. M. Corbett, R. V. W. Dimlich, J. R. Michael, and N.
J. Zaluec, Eds. San Francisco, CA: San Francisco Press, 1996, pp.
942–943.
[36] S. Guha, F. Agahi, B. Pezeshki, J. A. Kash, D. W. Kisker, and N A.
Bojarczuk, “Microstructure of AlGaAs-oxide heterolayers formed by
wet oxidation,” Appl. Phys. Lett., vol. 68, pp. 906–909, 1996.
[37] R. D. Twesten, D. M. Follstaedt, K. D. Choquette, and R. P. Schneider,
Jr., “Microstructure of laterally oxidized Alx Ga10x As layers in vertical-
cavity lasers,” Appl. Phys. Lett., vol. 69, pp. 19–21, 1996.
[38] T. Takamori, K. Takemasa, and T. Kamijoh, “Interface structure of se-
lectively oxidized AlAs/GaAs,” Appl. Phys. Lett., vol. 69, pp. 659–661,
1996.
[39] M. H. MacDougal, H. Zhao, P. D. Dapkus, M. Ziari, and W. H.
Steier, “Wide bandwidth distributed Bragg reflectors using oxide/GaAs
multilayers,” Electron. Lett., vol. 30, pp. 1147–1149, 1994.
[40] K. D. Choquette, K. M. Geib, H. C. Chui, B. E. Hammons, H. Q.
Hou, T. J. Drummond, and R. Hull, “Selective oxidation of buried
AlGaAs versus AlAs layers,” Appl. Phys. Lett., vol. 69, pp. 1385–1387,
1996.
[41] K. M. Geib, K. D. Choquette, H. Q. Hou, and B. E. Hammons,
“Fabrication issues of oxide-confined VCSEL’s,” in Vertical-Cavity
Surface Emitting Lasers, K. D. Choquette and D. Deppe, Eds., Proc.
SPIE, 1997, vol. 3003, pp. 69–74.
[42] K. D. Choquette, K. M. Geib, H. C. Chui, H. Q. Hou, and R. Hull,
“Selective oxidation of buried AlGaAs for fabrication of vertical-cavity
lasers,” Materials Res. Soc. Proc., vol. 421, pp. 53–61, 1996.
[43] R. S. Burton and T. E. Schlesinger, “Wet thermal oxidation of
Alx Ga10x As compounds,” J. Appl. Phys., vol. 76, pp. 5503–5507,
1994.
[44] H. Nickel, “A detailed experimental study of the wet oxidation kinetics
of Alx Ga10x As layers,” J. Appl. Phys., vol. 78, pp. 5201–5203, 1995.
[45] B. E. Deal and A. S. Grove, “General relationship for the thermal
oxidation of silicon,” J. Appl. Phys., vol. 36, p. 3770, 1965.
[46] F. A. Kish, S. A. Maranowski, G. E. Hofler, N. Holonyak, Jr., S. J.
Caracci, J. M. Dallesasse, and K. C. Hsieh, “Dependence on doping
type (p/n) of the water vapor oxidation of high gap Alx Ga10x As,”
Appl. Phys. Lett., vol. 60, pp. 3165–3167, 1992.
[47] M. Ochiai, G. E. Giudice, H. Temkin, J. W. Scott, and T. M. Cockerill,
“Kinetics of thermal oxidation of AlAs in water vapor,” Appl. Phys.
Lett., vol. 68, pp. 1898–1890, 1996.
[48] J. H. Kim, D. H. Lim, K. S. Kim, G. M. Yang, K. Y. Lim, and H.
J. Lee, “Lateral wet oxidation of Alx Ga10x As-GaAs depending on its
structure,” Appl. Phys. Lett., vol. 69, pp. 3357–3359, 1996.
[49] R. L. Naone and L. A. Coldren, “Surface energy model for the thickness
dependence of the lateral oxidation of AlAs,” J. Appl. Phys., 1997.
[50] O. Blum, C. I. H. Ashby, and H. Q. Hou, “Barrier layer thickness control
of selective wet oxidation of AlGaAs for embedded optical elements,”
Appl. Phys. Lett., vol. 70, pp. 2870–2872, 1997.
[51] R. L. Naone, E. R. Hegbloom, B. J. Thibeault, and L. A. Coldren,
“Oxidation of AlGaAs layers for tapered apertures in vertical cavity
lasers,” Electron. Lett., vol. 33, pp. 300–301, 1997.
[52] W. Ranke, Y. R. Xing, and G. D. Shen, “Orientation dependence of
oxygen absorption on a cylindrical GaAs crystal,” J. Vac. Sci. Technol.,
vol. 21, pp. 426–428, 1982.
[53] D. L Huffaker, J. Shin, and D. G. Deppe, “Low threshold half-wave
vertical cavity lasers,” Electron. Lett., vol. 30, pp. 1946–1947, 1994.
[54] K. L. Lear, S. P. Kilcoyne, R. P. Schneider, Jr., and J. A. Nevers,
“Life testing oxide confined VCSELs: Too good to last?” in Fabrication,
Testing, and Reliability of Semiconductor Lasers, M. Fallahi and S. C.
Wang, Eds., Proc. SPIE, 1996, vol. 2683, pp. 114–122.
[55] J. K. Guenter, R. A. Hawthorne, D. N. Granville, M. K. Hibbs-
Brenner, and R. A. Morgan, “Reliability of proton-implanted VCSEL’s
for data communications,” in Fabrication, Testing, and Reliability of
Semiconductor Lasers, M. Fallahi and S. C. Wang, Eds., Proc. SPIE,
1996, vol. 2683, pp. 102–103.
926 IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS, VOL. 3, NO. 3, JUNE 1997
Kent D. Choquette (M’97) received the B.S. degrees in engineering physics
and applied mathematics from the University of Colorado, Boulder, in 1984,
the M.S. and Ph.D. degrees in materials science from the University of
Wisconsin, Madison, in 1985 and 1990, respectively.
After a Post-Doctoral appointment at AT&T Bell Laboratories in Murray
Hill, NJ, where he developed vacuum integrated fabrication processes for op-
toelectronic devices, he joined Sandia National Laboratories in Albuquerque,
NM, in 1992. As a Senior Member of Technical Staff at Sandia, he has been
engaged in research into novel fabrication and the physics of vertical-cavity
surface emitting lasers and other optoelectronic devices. He has written over
90 technical publications and two book chapters, and has presented several
invited talks on VCSEL’s.
Dr. Choquette has served on technical committees for LEOS and OSA
conferences, and is a member of the IEEE Lasers and Electro-Optics Society
and the Optical Society of America.
Kent M. Geib received the M.S. degree in electrical engineering from
Colorado State University, Ft. Collins, CO.
He has been a member of the VCSEL development team at Sandia
National Laboratories in Albuquerque, New Mexico for the past three years.
He worked at Colorado State University as a Research Associate for 15
years implementing surface analytical spectroscopies to study the chemical
compositions and the development of the interfaces of III–V compound
semiconductors with their oxides and/or metals. His current research focus
is on the development of advanced processes for the fabrication of novel
optoelectronic devices.
Mr. Geib is a member of the American Vacuum Society and Sigma Xi.
Carol I. H. Ashby received the B.S. degree from the University of Idaho,
Moscow, and the Ph.D. degree in inorganic chemistry from the University of
Illinois, Urbana.
In 1979, she joined Sandia National Laboratories, Albuquerque, NM, where
she has performed research in a variety of areas including hydrogen-first
wall reactions for fusion reactors, carrier-controlled laser etching of com-
pound semiconductors, novel etching processes for lithium niobate, improved
etchants for high-Tc superconductors, compound semiconductor electronic
passivation, and wet oxidation of compound semiconductors. She is the author
of six book chapters and numerous scientific publications.
Ray D. Twesten received the B.S. degree in physics from Wayne State
University and the M.S. and Ph.D. degrees in physics from the University
of Illinois, Urbana-Champaign, where he used in situ electron microscopy to
study the formation of interfaces and dynamic processes on UHV surfaces.
He is a Post-Doctoral Research Associate at Sandia National Laboratories
where he uses transmission electron microscopy to investigate the properties
of semiconducting materials. He recently joined the staff of the Center for
Microanalysis of Materials at the University of Illinois.
Olga Blum received the B.S. degree from the University of Illinois,
Champaign-Urbana in 1987, and the M.S. and Ph.D. degrees from the
University of California, Berkeley, in 1990 and 1992, all in electrical
engineering.
In 1993, she joined Sandia National Laboratories in Albuquerque, NM.
Her research topics include microlens fabrication and integration with
VCSEL’s and oxidation of AlAsSb AlGaAs. She is currently working on
long wavelength VCSEL’s and integration of VCSEL’s with MEM’s.
Dr. Blum has served as a chair of the New Focus Award Committee. She
was awarded Newport Corporation fellowship while she was at UC Berkeley.
She is a member of the Optical Society of America.
Authorized licensed use limited to: Tampere University. Downloaded on January 04,2021 at 09:27:57 UTC from IEEE Xplore. Restrictions apply.
Hong Q. Hou (S’92–M’93) received the Ph.D. degree in electrical engineering
from the University of California, San Diego, in 1993.
His thesis research centered on 1.3-mm optical modulators based on
InAsP–InP strained quantum structures grown by gas-source molecular beam
epitaxy. Currently, he is a senior member of technical staff at Sandia National
Laboratories in Albuquerque, NM. His principal interests include the design
and epitaxial growth of a variety of optoelectronic and electronic devices by
metal–organic vapor phase epitaxy, especially vertical-cavity surface emitting
lasers. His past professional experience include a Research Assistant at the
Institute of Physics, Chinese Academy of Science, Beijing (1985–1988), a
Visiting Scholar at the Institute of Physical and Chemical Research, RIKEN,
Japan (1988–1990), and a Post-Doctoral Member of technical staff at AT&T
Bell Laboratories, Holmdel, NJ (1993–1995) He has published over 80 papers
in technical journals.
Dr. Hou is a member of the American Physical Society, the Optical Society
of America, and the Electrochemical Society (ECS).
David M. Follstaedt received the Ph.D. degree from the University of Illinois,
Urbana-Champaign, and the B.S. degree in physics from Texas Technological
University, Lubbock, TX.
He. is a Senior Member of Technical Staff at Sandia National Laboratories.
He uses transmission electron microscopy to evaluate microstructures of new
materials and relate them to other properties. His research in III–V semicon-
ductor alloys has included composition modulation, chemical ordering, defects
in GaN, and oxidation for device processing. Other interests include cavities
formed in semiconductors by He implantation, and forming and evaluating
high-strength layers on metals by ion implantation.
B. Eugene Hammons was born in Kansas City, KS, in October 1944.
He received the A.S. degree in electrical engineering technology from the
Missouri Institute of Technology, Kansas City, MO, in 1964.
He began work at Sandia National Laboratories in the Physics of Solids
Research Division in 1964. He is presently a member of technical staff in
the Center for Compound Semiconductor Science and Technology at Sandia,
and has been involved in thin film semiconductor growth for the past several
years. His present research interests include the metal–organic vapor phase
epitaxial (MOVPE) growth of new semiconductor devices and materials.
Dave Mathes received the B.S. degree in physics from Rochester Institute of
Technology, Rochester, NY, in 1995. He is currently working toward the Ph.D.
degree in the Materials Science Department of The University of Virginia,
Charlottesville.
His research consists of using transmission electron microscopy and focused
ion beam sputtring techniques to investigate the structural, chemical, and
mechanical properties of III-V oxides grown from AlGaAs and InAlP.
Robert Hull received the B.A. degree in physics and D.Phil degree in
materials science from Oxford University, Oxford, U.K.
He worked for two years at Hewlett Packard Laboratories and nine years
at Bell Laboratories prior to joining the University of Virginia as a tenured
Associate Professor in the Materials Science and Engineering Department in
1994. His research concentrates upon mechanisms of epitaxial growth, thin-
film microstructure, and electronic materials. He is currently president of the
Materials Research Society.