In situ study of surface reactions of atomic layer deposited Lax Al2-x O3 films on atomically clean In0.2 Ga0.8 As

Article abstract of DOI:10.1063/1.3009303

The surface reactions of Lax Al2-x O3 ultrathin films deposited on atomically clean In0.2 Ga0.8 As by atomic layer deposition are studied by in situ high resolution x-ray photoelectron spectroscopy. Using 1:2 alternating cycles of La2 O3 and Al2 O3 results in a La:Al concentration ratio of 1:10. We found that the Lax Al2-x O3 InGaAs interface consisted of interfacial Ga-suboxides and As-As bonds but no As- or In-oxides were detected. This suggests an interface formed by Ga-O-Al and Ga-O-La bonds from the precursor reaction.

Full text of DOI:10.1063/1.3009303

In situ study of surface reactions of atomic layer deposited La x Al 2 − x O 3 films on
atomically clean In 0.2 Ga 0.8 As
F. S. Aguirre-Tostado, M. Milojevic, B. Lee, J. Kim, and R. M. Wallace
Citation: Applied Physics Letters 93, 172907 (2008); doi: 10.1063/1.3009303
View online: http://dx.doi.org/10.1063/1.3009303
View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/93/17?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
Improved thermal stability and electrical properties of atomic layer deposited HfO2/AlN high-k gate dielectric
stacks on GaAs
J. Vac. Sci. Technol. A 33, 01A136 (2015); 10.1116/1.4903367
Effective passivation of In0.2Ga0.8As by HfO2 surpassing Al2O3 via in-situ atomic layer deposition
Appl. Phys. Lett. 101, 172104 (2012); 10.1063/1.4762833
Reconstruction dependent reactivity of As-decapped In0.53Ga0.47As(001) surfaces and its influence on the
electrical quality of the interface with Al2O3 grown by atomic layer deposition
Appl. Phys. Lett. 99, 193505 (2011); 10.1063/1.3659688
Half-cycle atomic layer deposition reaction studies of Al 2 O 3 on In 0.2 Ga 0.8 As (100) surfaces
Appl. Phys. Lett. 93, 202902 (2008); 10.1063/1.3033404
Synchrotron radiation photoemission spectroscopic study of band offsets and interface self-cleaning by
atomic layer deposited Hf O 2 on In 0.53 Ga 0.47 As and In 0.52 Al 0.48 As
Appl. Phys. Lett. 93, 182103 (2008); 10.1063/1.3020298
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
130.18.123.11 On: Sat, 20 Dec 2014 14:33:30

APPLIED PHYSICS LETTERS 93, 172907 ͑2008͒
In situ study of surface reactions of atomic layer deposited La Al O
3
x
2−x
films on atomically clean In Ga As
0
.2
0.8
a͒
b͒
F. S. Aguirre-Tostado, M. Milojevic, B. Lee, J. Kim, and R. M. Wallace
Department of Materials Science and Engineering, University of Texas at Dallas,
Richardson, Texas 75083, USA
͑
Received 13 July 2008; accepted 8 October 2008; published online 31 October 2008͒
The surface reactions of La Al O ultrathin films deposited on atomically clean In Ga As by
x
2−x
3
0.2
0.8
atomic layer deposition are studied by in situ high resolution x-ray photoelectron spectroscopy.
Using 1:2 alternating cycles of La O and Al O results in a La:Al concentration ratio of 1:10. We
2
3
2
3
found that the La Al O /InGaAs interface consisted of interfacial Ga-suboxides and As–As bonds
x
2−x
3
but no As- or In-oxides were detected. This suggests an interface formed by Ga–O–Al and Ga–O–La
bonds from the precursor reaction. © 2008 American Institute of Physics.
͓
DOI: 10.1063/1.3009303͔
The search for a suitable high-␬ dielectric on III-V high-
mobility substrates for surface channel metal-oxide-
semiconductor ͑MOS͒ applications has led to resurgence in
interest of interface studies. For scaled structures beyond
clean InGaAs surface reconstructed ͑2ϫ4͒. After the surface
cleaning, either 1.4 or 10 nm La Al O films were depos-
x
2−x
3
1
5
ited in situ using an integrated SUNALE™ ALD reactor.
For comparison, a 1 nm thick Al O film was deposited un-
2
3
2
0 nm gate lengths, the dielectric constant for Al O
der identical ALD conditions on an InGaAs sample sepa-
rately. The La Al O ALD deposition was carried out at a
2
3
͑
␬ϳ9͒ can be enhanced by the addition of other high-␬ ox-
x
2−x
3
ides ͑HfO or La O ͒ while preserving its useful thermal
substrate temperature of 200 °C to minimize the thermal de-
sorption of volatile As and In in a base pressure of flowing
N of 10 mbar. The La Al O film was deposited by re-
2
2
3
1
,2
stability. Recent reports have investigated molecular beam
3
deposited LaAlO and atomic layer deposition ͑ALD͒ of
3
2
x
2−x
3
4
HfAlO on InGaAs, due to the higher electron mobility and
peating 1 cycle of La O ͑i.e., La precursor+water͒ and
2 3
3
mitigation of Fermi level pinning problems associated with
2 cycles of Al O ͑i.e., Al precursor+water+Al precursor
2 3
5
–7
GaAs surface states impacting MOS devices.
+water͒ for a total of five times. The integrated ALD reactor,
Differences in the initial surface leads to different reac-
tion channels and therefore to different interface bond
transfer chambers, and XPS system enable the interrogation
1
6
of the surface without spurious contamination. The XPS
8
,9
arrangements.
The nucleation of ALD grown Al O
data were obtained using an Al K␣ monochromatic x-ray
2
3
1
can depend strongly on the availability of surface–OH
groups rather than on the deposition conditions. Lim
source ͑0.25 eV line width͒ and a hemispherical analyzer
1
0
͑pass energy=15 eV͒ equipped with seven Channeltron®
1
7
et al. reported the deposition of LaAlO ͑LAO͒ films on Si
detectors. The take-off angle from the substrate surface was
45°, with an analyzer acceptance angle of 16°. The deconvo-
lution of XPS spectra was performed self-consistently using
the software AANALYZER ͑Ref. 18͒ with fixed values for pa-
rameters such as Lorentzian/Gaussian ratios, spin-orbit split-
ting, and branching ratios, which are either known or deter-
3
using
a
La
amidinatei
precursor
͑tris͑N-NЈ-
11
diisopropylacetamidinato͒La ͓͑ Pr -amd͒ -La͔͒ and water.
2
3
They reported a low C contamination ͑ϳ1%͒ and no self-
decomposition below 350 °C.
In this letter we report on the surface chemical
reactions of La Al O and Al O deposited on atomically
1
4
mined directly from the atomically clean InGaAs substrate.
x
2−x
3
2
3
clean InGaAs substrates by ALD. A recently developed
Figure 1 shows the Al 2p and O 1s XPS spectral regions
1
2
La-amidinate
precursor
͑tris͑N-NЈ-diisopropyl-
for a 1.4 nm thick La Al O film grown using alternating
x
2−x
3
i
formamidinato͒La ͓͑ Pr -fmd͒ -La͔͒ together with water is
employed for La O deposition and trimethyl aluminum
cycles of La O and Al O on atomically clean In Ga As
2
3
2
3
2
3
0.2
0.8
as well as the comparison spectra for the 1 nm thick pure
2
3
1
9
͑
TMA͒ with water for Al O . In situ ALD and x-ray photo-
Al O . The binding energy of the Al 2p ͑74.9 eV͒ is con-
2 3
20,21
2
3
electron spectroscopy ͑XPS͒ analyses are used to study the
interface bonding arrangement and the relevant reaction
paths are discussed as well.
sistent with O–Al–O bonding environment.
The O 1s
shows two peaks for the pure Al O films ͑at 531.6 and
2
3
533.6 eV͒ and three peaks for the La Al O film ͑531.6,
x
2−x
3
2
The 1ϫ1 cm substrate consisted of a 13.5 nm
533.0, and 530.0 eV͒. The first peak is as attributed to oxy-
1
3
20,21
In Ga As layer grown by molecular beam epitaxy on a
gen in the Al–O–Al bonds,
and La–O–H bonds.
and the second to Al–O–H
0
.2
0.8
2
2,23
semi-insulating GaAs͑001͒ wafer with an intermediate
The third peak, appearing only in
5
35 nm thick GaAs buffer layer. The InGaAs and GaAs
the La Al O film, is chemically shifted by −1.6 eV with
x 2−x 3
1
7
−3
buffer layers were doped with Si ͑n-type, 1ϫ10 cm ͒. The
InGaAs native oxides were removed using an in situ atomic
H treatment ͑AHT͒ at a substrate temperature of 450 °C.
This surface preparation method provides an atomically
respect to the Al–O–Al peak and is identified as oxygen
forming Al–O–La bonds. The chemical shift is likely caused
1
4
24
by an additional charge transfer from the La to the O atom
as the Al and La atoms possess different electronegativities
2
5
͑
1.61 and 1.10, respectively͒. No La–O–La bonding
26
a͒
͑528.8 eV͒ is detected. The O 1s in Fig. 1 shows no sig-
Electronic mail: jiyoung.kim@utdallas.edu.
Electronic mail: rmwallace@utdallas.edu.
b͒
nificant difference in the observed amount of hydroxides be-
0
003-6951/2008/93͑17͒/172907/3/$23.00
93, 172907-1
30.18.123.11 On: Sat, 20 Dec 2014 14:33:30
© 2008 American Institute of Physics
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
1

172907-2
Aguirre-Tostado et al.
Appl. Phys. Lett. 93, 172907 ͑2008͒
FIG. 2. ͑Color online͒ As 2p3/2, Ga 2p3/2, and In 3d5/2 showing the surface
FIG. 1. ͑Color online͒ Normalized Al 2p and O 1s XPS peaks for 1 nm of
LaAlO3 vs 1 nm of Al O . In both cases Al 2p shows a single chemical
reaction at the initial growth of LaAlO vs Al O . The formation of As–As
3
2
3
2
3
and Ga-suboxides occurs in a similar manner for the two films. The forma-
tion of In oxide is not observed.
component around 74.9 eV corresponding to fully oxidized Al. O 1s shows
a main peak around 531.7 eV corresponding to La–O–Al and Al–O–Al bond
arrangements for LaAlO and Al O , respectively. The small feature at 2 eV
3
2
3
above corresponds to hydroxyl groups.
Ga O suggesting the formation of a Ga-suboxide most
2
3
likely in the following bond environment: substrate-Ga–
2
8
tween the two samples. From the XPS analysis, the Al–OH
amount is ϳ10% of the total oxygen in the film and is at-
tributed to residual water incorporated during the ALD
growth at 200 °C.
The total C content detected in the films ͑not shown͒
measured by the XPS C 1s feature to be ϳ4% for the 10 nm
LAO films. For the 1.4 nm LAO film, C–O, C–O–H, and
C–C bondings were detected, while the thicker 10 nm LAO
film indicated the presence of C–O and/or C–O–H bonding
without any observation of carbonate peaks. By comparison,
O–Al ͑or La͒. Similarly, the As 2p3/2 shows an additional
component at 0.63 eV above the position of the As-bulk peak
2
7
͑1322.4 eV͒ and is associated with As–As bonding. The
area ratio of the nonbulk/bulk Ga and As components are
0.31 and 0.49, respectively. Assuming that these chemical
species are confined at the interface forming a layer, the
above amounts correspond to a full monolayer for each
case. Given the amounts of reacted Ga and As at the oxide-
substrate interface and chemical bonding detected, we de-
duce that the reacted interface is chemically abrupt.
Figure 3 shows the Al 2p, O 1s, and La 3d peaks com-
paring 1.4 nm versus 10 nm thick La Al O films. Oxi-
dized Al and La species are evident in both films, with the
expected XPS satellite structure for La. The relative atomic
concentrations were calculated using calibrated atomic sen-
sitivity factors obtained using XPS and Rutherford back-
2
9
1
4
XPS measurement of the Al O film indicates the presence
2
3
3
0
of C–C bonding with a C concentration ϳ4%. We note that
these levels of C are somewhat higher than previously re-
ported secondary ion mass spectroscopy results for
x
2−x
3
3
1
i
11
La͑ PrAMD͒ on Si at 300–330 °C, and may be attributed
3
to the lower deposition temperature employed here for the
InGaAs substrates.
scattering spectroscopy from sputter deposited LaAlO thick
3
Figure 2 shows the XPS spectra of the As 2p3/2,
films. From the calculated atomic concentrations the anion to
cation atomic ratio is 3:2. The Al:La atomic ratio is 11.5
which gives an x=0.16 for La Al O .
In conclusion, we have shown that the ALD growth of
La0.16Al1.84O and pure Al O on atomic hydrogen-treated
InGaAs results in similar interface bonding arrangement
while producing the equivalent of one monolayer of As–As
bonding and Ga–O–Al͑La͒ bridging between the substrate
and the film. The employment of an atomically clean InGaAs
substrate and ALD deposition of La Al O using TMA/
Ga 2p3/2, and In 3d
5/2
for the clean InGaAs surface
͑
bottom͒, 1 nm Al O /InGaAs ͑middle͒, and 1.4 nm
2
3
x
2−x
3
La Al O /InGaAs ͑top͒. The In, Ga, and As peaks in the
x
2−x
3
clean substrate show a single chemical component that is
free of detectable surface oxides and As–As bonds as a result
of the AHT. After the deposition of 1 nm of Al O or
3
2
3
1
4
2
3
La Al O , the In 3d region still shows a single chemical
x
2−x
3
5/2
component ͑444.2 eV͒, indicating that the In-deficient
1
4
surface results in no In substrate reaction with the overly-
ing oxide.
x
2−x
3
i
water and ͑ Pr -fmd͒ La/water chemistry results in the for-
2
3
In contrast, the Ga 2p region shows an additional chemi-
cal component at 0.4 eV above the Ga-bulk component and
mation of a single monolayer interfacial layer.
2
7,28
is associated with the formation of Ga-suboxide.
The
The authors thank J. Suydam, D. Shenai, and M. Rous-
chemical shift found here is substantially lower than in
seau of Rohm and Haas LLC for the La precursor employed
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
30.18.123.11 On: Sat, 20 Dec 2014 14:33:30
1

172907-3
Aguirre-Tostado et al.
Appl. Phys. Lett. 93, 172907 ͑2008͒
electron. Eng. 84, 2154 ͑2007͒.
7
8
Y. Xuan, P. D. Ye, and T. Shen, Appl. Phys. Lett. 91, 232107 ͑2007͒.
C. L. Hinkle, A. M. Sonnet, E. M. Vogel, S. McDonnell, G. J. Hughes, M.
Milojevic, B. Lee, F. S. Aguirre-Tostado, K. J. Choi, H. C. Kim, J. Kim,
and R. M. Wallace, Appl. Phys. Lett. 92, 071901 ͑2008͒.
F. S. Aguirre-Tostado, M. Milojevic, K. J. Choi, H. C. Kim, C. L. Hinkle,
E. M. Vogel, J. Kim, T. Yang, Y. Xuan, P. D. Ye, and R. M. Wallace, Appl.
Phys. Lett. 93, 061907 ͑2008͒.
9
0
1
R. L. Puurunen, J. Appl. Phys. 97, 121301 ͑2005͒.
B. S. Lim, A. Rahtu, P. de Rouffignac, and R. G. Gordon, Appl. Phys.
Lett. 84, 3957 ͑2004͒.
The La precursor is provided by Rohm and Haas Electronic Materials
LLC.
The InGaAs/GaAs epitaxial structure was grown by Intelliepi.
F. S. Aguirre-Tostado, M. Milojevic, C. L. Hinkle, E. M. Vogel, R. M.
Wallace, S. McDonnell, and G. J. Hughes, Appl. Phys. Lett. 92, 171906
11
1
2
1
1
3
4
5͑
2008͒.
1
1
1
1
See: http://www.picosun.com/.
6
7
8
R. M. Wallace, ECS Trans. 16, 255 ͑2008͒.
Omicron Nanotechnology: http://www.omicron.de.
AANALYZER is a software for XPS peak deconvolution. See: http://
www.qro.cinvestav.mx/~aanalyzer.
1
2
9
The Al O was deposited in situ separately on In0.2Ga0.8As using ten con-
2
3
secutive cycles of TMA and de-ionized water pulses to form an ϳ1 nm
thick film.
0
O. Renault, L. G. Gosset, D. Rouchon, and A. Ermolieff, J. Vac. Sci.
Technol. A 20, 1867 ͑2002͒.
FIG. 3. ͑Color online͒ Al 2p, O 1s, and La 3d comparison of 1 vs 10 nm
LaAlO films. There are no detectable differences in the LaAlO films com-
21
22
3
3
D. Leinen, A. Fernandez, J. P. Espinos, J. P. Holgado, and A. R. Gonzalez-
Elipe, Appl. Surf. Sci. 68, 453 ͑1993͒.
position or chemical bonding. The atomic composition calculated from XPS
is La0.03Al0.37O0.60
.
M. R. Alexander, G. E. Thompson, and G. Beamson, Surf. Interface Anal.
2
9, 468 ͑2000͒.
2
2
3
4
A. Nylund and I. Olefjord, Surf. Interface Anal. 21, 283 ͑1994͒.
P. S. Bagus, F. Illas, G. Pacchioni, and F. Parmigiani, J. Electron Spec-
trosc. Relat. Phenom. 100, 215 ͑1999͒.
in this work. This work was supported in part by the FCRP
on Materials, Structures, and Devices and by “System IC
2010” project funded by MKE in Korea.
25
CRC Handbook of Chemistry and Physics, edited by D. R. Lide 84th ed.
CRC, Boca Raton, FL, 2004͒, pp. 9–74.
1
26͑
P. Sivasubramani, M. J. Kim, B. E. Gnade, R. M. Wallace, L. F. Edge, D.
T. L. Barr, J. Phys. Chem. 82, 1801 ͑1978͒.
2
7
G. Schlom, H. S. Craft, and J.-P. Maria, Appl. Phys. Lett. 86, 201901
S. Arabasz, E. Bergignat, G. Hollinger, and J. Szuber, Appl. Surf. Sci.
252, 7659 ͑2006͒.
S. Ingrey, W. M. Lau, and N. S. McIntyre, J. Vac. Sci. Technol. A 4, 984
͑1986͒.
The intensity ratio of a monolayer is calculated using the effective attenu-
ation lengths obtained from the EAL-NIST database and the following
formula: Ifilm/Isubstrate=1=1 exp͑d/␭ cos ␪͒, where d is the interplanar dis-
tance, ␭ is the EAL and ␪ is the take-off angle measured from the surface
plane.
The atomic sensitivity factors found for La 3d5/2, Al 2p, and O 1s are 8.11,
0.094, and 0.7, respectively.
D. F. Mullica, H. O. Perkins, C. K. C. Lok, and V. Young, J. Electron
Spectrosc. Relat. Phenom. 61, 337 ͑1993͒.
2͑
2005͒.
28
29
L. F. Edge, D. G. Schlom, P. Sivasubramani, R. M. Wallace, B. Holländer,
and J. Schubert, Appl. Phys. Lett. 88, 112907 ͑2006͒.
3
N. Goel, P. Majhi, W. Tsai, M. Warusawithana, D. G. Schlom, M. B.
Santos, J. S. Harris, and Y. Nishi, Appl. Phys. Lett. 91, 093509 ͑2007͒.
T. Yang, Y. Xuan, D. Zemlyanov, T. Shen, Y. Q. Wu, J. M. Woodall, P. D.
Ye, F. S. Aguirre-Tostado, M. Milojevic, S. McDonnell, and R. M.
Wallace, Appl. Phys. Lett. 91, 142122 ͑2007͒.
4
5
6
30
L. J. Brillson, M. L. Slade, R. E. Viturro, M. K. Kelly, N. Tache, G.
Margaritondo, J. M. Woodall, P. D. Kirchner, G. D. Pettit, and S. L.
Wright, Appl. Phys. Lett. 48, 1458 ͑1986͒.
3
1
G. Brammertz, M. Heyns, M. Meuris, M. Caymax, and D. Jiang, Micro-
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
30.18.123.11 On: Sat, 20 Dec 2014 14:33:30
1