Full text of DOI:10.1063/1.114960

Synthesis of aluminum oxide thin films: Use of aluminum trisdipivaloylmethanate as a
new low pressure metal organic chemical vapor deposition precursor
E. Ciliberto, I. Fragalà, R. Rizza, G. Spoto, and G. C. Allen
Citation: Applied Physics Letters 67, 1624 (1995); doi: 10.1063/1.114960
View online: http://dx.doi.org/10.1063/1.114960
View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/67/11?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
Metal-organic chemical vapor deposition of aluminum oxide thin films via pyrolysis of dimethylaluminum
isopropoxide
J. Vac. Sci. Technol. A 28, 238 (2010); 10.1116/1.3294718
Metal-organic chemical vapor deposition of high-dielectric-constant praseodymium oxide films using a
cyclopentadienyl precursor
Appl. Phys. Lett. 96, 012105 (2010); 10.1063/1.3275706
In situ metal-organic chemical vapor deposition atomic-layer deposition of aluminum oxide on GaAs using
trimethyaluminum and isopropanol precursors
Appl. Phys. Lett. 93, 031902 (2008); 10.1063/1.2960574
Properties of lanthanum oxide thin films deposited by cyclic chemical vapor deposition using tris(isopropyl-
cyclopentadienyl)lanthanum precursor
J. Appl. Phys. 100, 024111 (2006); 10.1063/1.2218465
Fabrication of aluminum oxide thin films by a lowpressure metalorganic chemical vapor deposition technique
Appl. Phys. Lett. 62, 681 (1993); 10.1063/1.108838
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:
130.239.20.174 On: Mon, 24 Nov 2014 20:06:57

Synthesis of aluminum oxide thin films: Use of aluminum
tris-dipivaloylmethanate as a new low pressure metal organic chemical
vapor deposition precursor
E. Ciliberto,a) I. Fragala, R. Rizza, and G. Spoto
`
`
Dipartimento di Scienze Chimiche, Universita di Catania, Via le A. Doria 6, Catania, Italy
G. C. Allen
Interface Analysis Centre, 121 St. Michael Hill, Bristol, United Kingdom
͑Received 21 March 1995; accepted for publication 30 June 1995͒
Amorphous aluminum oxide thin films have been produced with low pressure metal organic
chemical vapor deposition technique using the aluminum tris-dipivaloylmethanate volatile
precursor. Different carrier gases were used for the depositions. The surfaces of the films were
analyzed using x-ray photoelectron spectroscopy and secondary ion mass spectrometry. A very low
content of carbon was verified when oxygen and water vapor saturated argon were used as carrier
gases. A higher hydration percentage of the deposited material was verified when water vapor was
present during the deposition process. © 1995 American Institute of Physics.
The deposition of aluminum oxide films has been of
considerable importance in both pure and applicative scien-
tific research. The interest is connected with the applications
of these films in various technological fields. The high di-
electric constant, the low permeability to alkali ions, the high
radiation resistance, and the high thermal conductivity prop-
erties of aluminum oxide are of interest in microelectronics
and very large scale integration ͑VLSI͒ circuit production.1–4
Al2O3 films are also widely used as dielectric layers in the
fabrication of capacitive humidity sensors5 and are interest-
ing candidates for planar optic waveguides6 because of their
low refractive index ͑Ͻ1.7͒. The high hardness and the high
corrosion resistance also make Al2O3 films attractive candi-
dates for passivation of metal surfaces.7,8
deposition using the tris͑2,2,6,6-tetramethylheptane-3,5-
dionate͒ aluminum complex Al͑dpm͒ as a new volatile
͓
͔
3
precursor in a low pressure MOCVD reactor. This complex
was chosen because of the simpler synthetic procedure, its
higher stability in air, the better volatility and thermal stabil-
ity compared to that of Al͑acac͒3.21 The low acidity of the
related Hdpm acid relative to the Hacac species22 is more-
over, of advantage. The deposited films have been analyzed
by surface and bulk techniques to evaluate the oxide stoichi-
ometry and variations along the vertical profile, the chemical
oxidation state of atom components, the carbon content, the
crystal phase, the coating homogeneity and, finally, the thick-
ness of the films obtained.
Dipivaloylmethane ͑2,2,6,6-tetramethylheptane-3,5-di-
one, Adrich Chemical Co. Ltd.͒ and the aluminum nitrate
nonahydrate ͑Al͑NO3͒3•9H2O, Fluka͒ were used without fur-
ther purification for the preparation of Al͑dpm͒3. The syn-
thetic route successfully adopted for various ␤-diketone che-
Until now, coating techniques of aluminum oxide thin
films have spurred a large interest in connection with a vari-
ety for experimental methods.1,9–13 Among them, low pres-
sure metalorganic chemical vapor deposition ͑LP-MOCVD͒
has been applied successfully. This technique has found an
increasing application in a wide variety of research fields.14
In fact, the ability to kinetically control the deposition pro-
cess and to pattern highly uniform coatings on complex-
shaped objects allied to the facility for large-scale produc-
tion, renders this technique is a very powerful tool. Some
problems, however, still remain open to question. Among
them, the substrate reactivity and possible carbon codeposi-
tion are of particular importance. In both cases, these aspects
involve reactivity and the thermal stability of the volatile
precursors. Trimethylaluminum Al͑CH ͒ ,15 aluminum
lates was used.23 Al͑dpm͒ was then purified by re-
3
crystallization from ethanol and by vacuum sublimation ͑mp
265–266 °C͒.
The aluminum oxide film deposition was carried out in a
horizontal quartz hot-well CVD reactor. The flow rate of
several carrier gases ͑Ar, O2, Ar/H2O) was kept at 100
ml minϪ1. The total pressure was 10 Torr. A source tempera-
ture of 150 °C was used for the precursor. The reactor depo-
sition zone including the susceptor was maintained at
600 °C. Glass slides and Si ͑100͒ wafers were used as sub-
strates, after cleaning with a thorough degreasing procedure
using ultrasonic scrubbing and then drying in a nitrogen flux.
Silicon wafers were also chemically etched by dipping in a
diluted HF solution to remove the native oxide and used
immediately or stored under dried nitrogen atmosphere. The
specimens were mounted on a porous alumina susceptor, par-
allel to the gas streams.
͓
͔
3 3
chloride
AlCl ,16
aluminum
triisopropoxide
͓
͔
3
͓Al͑i-OC3H7͒3͔,17 aluminum acetylacetonate Al͑acac͒ ͔,18
͓
3
aluminum 2-ethylhexanoate Al͑C H CO ͒ 19, and alumi-
͓
͔
2 3
7
15
20
num borohydride Al͑BH ͒ ͔ have been used so far for
͓
4 3
Al2O3 film growth. Both the deposition rate as well as the
quality of these films have been found correlated with the
nature of the carrier and reaction gases.
In this letter, we report on aluminum oxide thin-film
An x-ray diffractometer ͑Philips PW1130͒ equipped with
a monochromatized Cu K␣ source was used for XRD mea-
surements. The spectra did not show any peaks, thus indicat-
a͒
Electronic mail: cilibert@ictuniv.unict.it
1624
Appl. Phys. Lett. 67 (11), 11 September 1995
0003-6951/95/67(11)/1624/3/$6.00
© 1995 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:
130.239.20.174 On: Mon, 24 Nov 2014 20:06:57

FIG. 2. Deconvoluted O 1s spectrum of aluminum oxide film deposited
using Ar/H2O.
FIG. 1. ͑a͒ XPS survey spectrum from an as-deposited aluminum oxide film
produced using Ar carrier gas. ͑b͒ XPS survey spectrum from the same
sample after sputtering with 3 kV Arϩ ions for 30 s.
30 s Arϩ sputter for the amorphous Al2O3 film deposited
under Ar/H2O flow. Curve fitting procedure shows two dif-
ferent peaks attributed to O2Ϫ ions ͑530.3 eV, FWHM 1.83
eV͒ and to OHϪ ͑531.6 eV, FWHM 1.83 eV͒,25 respectively.
ing that the material constituting the coating films was amor-
phous.
The ratio IOHϪ /IO was 0.17 in the case of the film depos-
X-ray photoelectron spectroscopy ͑XPS͒ and secondary
ion mass spectrometry ͑SIMS͒ analyses were performed to
investigate the composition of the films. The XPS spectra
were acquired using an Al K␣ monochromatized source
͑PHI 5600 Monochromator System͒. An electron flood gun
was used to compensate the charging effects that occurred.
The thickness of the films was estimated by XPS depth pro-
file using a 3 keV Arϩ ion gun. The sputter rate ͑145 Å
minϪ1, Arϩ, 3 keV, 25 mA filament current, 0.8 nA ion cur-
rent͒ was estimated by reference to the thickness of Al2O3
films which were grown on silicon in the same conditions,
and then, calibrated by Rutherford Backscattering Spectrom-
etry ͑2 meV, Heϩ, Van de Graaff accelerator͒.
tot
ited under Ar/H2O. On the contrary 0.02 and 0.04 ratios were
observed when O2 and Ar gases, respectively, were used.
These observations provide an indication of a higher hydra-
tion percentage in the films grown in the presence of water
vapor.
The quantitative Al/Otot atom ratios observed for the
sputtered films were consistent with the observed IOHϪ /IO
tot
trends. For the Ar/H2O grown film, an Al/Otot ratio of 0.41
was observed, whereas ratios of 0.51 and 0.55 were observed
for the films grown using, respectively, Ar and O2 carrier
gases as compared with theoretical values of 0.33 for
Al͑OH͒ and 0.66 for Al2O3.
3
As expected,24 different deposition rates were obtained
using different carrier gases. An average deposition rate of
25 Å minϪ1 was observed using Ar carrier gas. The rate in-
creased to 40 Å minϪ1 with O2. The highest rate ͑70 Å
minϪ1͒ was observed when water vapor-saturated Ar gas was
used.
The modified Auger parameters ͑about 1038.6 eV, Fig.
3͒, evaluated with respect to the position of the O KVV and
Figure 1 shows the wide scan XPS spectra of a represen-
tative as-deposited amorphous Al2O3 film ͑Ar carrier gas͒
before and after 30 s Arϩ sputtering. Before sputtering ͓Fig.
1͑a͔͒, peaks associated to Al 2p ͑73.7 eV͒, Al 2s ͑118.6 eV͒,
O 1s ͑530.5 eV͒, and O KVV ͑978.4 eV͒ become evident
͑C 1s reference BEϭ284.5 eV͒. Spectral features due to C 1s
ionization show a high atomic percentage of carbon impuri-
ties on the surface ͑9.8%͒. After 30 s Arϩ sputter cycle, the
carbon surface contamination was reduced to 1.1% and, thus,
no impurities other than Ar, are evident ͓Fig. 1͑b͔͒. Identical
procedures adopted for films deposited under either O2 or
Ar/H2O as carrier gases unambiguously indicate that the car-
bon content is, in every case, lower than 1%. In every case,
Al2O3 stoichiometry was found to be reproducible both lat-
erally and vertically.
Figure 2 shows the narrow scan of the O 1s region after
FIG. 3. Modified Auger parameter plot for some deposited films.
Appl. Phys. Lett., Vol. 67, No. 11, 11 September 1995
Ciliberto et al.
1625
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:
130.239.20.174 On: Mon, 24 Nov 2014 20:06:57

film grown under pure Ar, and provide further evidence of
surface hydration as observed with XPS measurements.
From these results we can affirm that the Al͑dpm͒ pre-
3
cursor allows us to obtain films of amorphous Al2O3 by LP-
MOCVD with a low content of carbon impurities. The evi-
dence that carbon impurities are mainly confined to the
surface of all the analyzed films shows that the deposition
process occurs without carbon contamination.
Preliminary mass spectrometry data seem to indicate
that the volatile acid species Hdpm, the hydrolysis product
of Al͑dpm͒3, is produced during the deposition process.
This observation might explain the lower carbon content ob-
served here compared to other films prepared using different
metalorganic precursors.
The observed deposition rates for the three cases exam-
ined suggest that water plays an important role in speeding
up the film growth mechanisms. At the same time, these
films showed a more evident hydration percentage, which
has implications for certain characteristics like density, re-
fractive index, and dielectric properties.
FIG. 4. Valence band region ͑a͒ for the as-deposited aluminum oxide film,
͑b͒ after sputtering with 3 kV Arϩ ions for 30 s.
1 K. P. Pande, V. K. R. Nair, and D. Gutierrez, J. Appl. Phys. 54, 5436
͑1983͒.
2 M. Ishida, I. Katakabe, T. Nakamura, and O. Ohtake, Appl. Phys. Lett. 52,
1326 ͑1988͒.
the O 1s peaks, also agree with data already reported by
Wagner et al. for Al2O3.26
3 K. Sawada, M. Ishida, I. Katakabe, T. Nakamura, and N. Ohtake, Appl.
Phys. Lett. 52, 1672 ͑1988͒.
4 L. G. Meiners, Thin Solid Films 113, 85 ͑1984͒.
5 V. K. Khanna and R. K. Nahar, Appl. Surf. Sci. 28, 247 ͑1987͒.
6 M. K. Smit, G. A. Acket, and C. J. van der Laan, Thin Solid Films 138,
171 ͑1986͒.
Moreover, no significant changes dependent upon the
deposition operating conditions were observed in the valence
band region ͑4–13 eV͒. A FWHM of 6.87 eV became appar-
ent in the as-deposited amorphous Al2O3 film ͓Fig. 4͑a͔͒. The
band shows a shape similar to that reported for the
␣-Al2O3.27 Two main features are, thus, observed. The higher
BE feature must be associated to a density of states having a
dominant Al–O bonding nature. The remainder is due to
states having O 2p nonbonding contribution. Arϩ sputtering
͓Fig. 4͑b͔͒ caused pronounced intensity ratio variations be-
tween the two features probably associated with the removal
of oxygenated species contaminating the film surface. Thus,
the final shape of the VB region might be considered as
originating from the deposited film surface, and the clear
evidence of a structural shape could be related to the short
range order ͑octahedral environment of aluminum͒ in the
amorphous film surface.
No meaningful differences were observed on the nega-
tive SIMS spectra ͑VG TOF-SIMS, Gaϩ, 20 kV, 5 ␮A ion
current͒ of films grown both under Ar and O2. They showed
peaks corresponding to OϪ ͑16 amu͒, OHϪ ͑17 amu͒, OϪ2 ͑32
amu͒, AlOϪ ͑43 amu͒, and AlOϪ2 ͑59 amu͒ species. In the
films grown under Ar/H2O other peaks corresponding to
AlOϪ3 ͑75 amu͒ and AlO2HϪ ͑60 amu͒ fragments, respec-
tively, were observed. These observations are entirely con-
sistent with the IOHϪ /IOϪ ratios of 0.24, observed in the case
7 V. A. C. Haanappel, H. D. van Corbach, T. Fransen, and P. J. Gellings,
Thin Solid Films 230, 138 ͑1993͒.
8 H. D. van Corbach, V. A. C. Haanappel, T. Fransen, and P. J. Gellings,
Thin Solid Films 239, 31 ͑1994͒.
9 R. Solanki, W. H. Ritchie, and G. J. Collins, Appl. Phys. Lett. 43, 454
͑1983͒.
10 O. B. Ajaki, M. S. Akanni, J. N. Lambi, H. D. Burrows, O. Osasona, and
B. Podor, Thin Solid Films 138, 91 ͑1986͒.
11 G. S. Hogashi and C. G. Fleming, Appl. Phys. Lett. 55, 1963 ͑1989͒.
12 C. A. T. Salama, J. Electrochem. Soc. 117, 913 ͑1970͒.
13
¨
W. Kubler, Thin Solid Films 199, 257 ͑1991͒.
14 J. T. Spencer, Prog. Inorg. Chem. 41 ͑1994͒.
15 L. H. Hall and W. C. Robinette, J. Electrochem. Soc. 118, 1624 ͑1971͒.
16 R. Colmet, R. Naslain, P. Hagenmuller, and C. Bernard, J. Electrochem.
Soc. 129, 1367 ͑1982͒.
17 J. A. Aboaf, J. Electrochem. Soc. 114, 948 ͑1967͒.
18 J. S. Kim, H. A. Marzouk, P. J. Reucroft, J. D. Robertson, and C. E.
Harmin, Jr., Appl. Phys. Lett. 62, 681 ͑1993͒.
19 T. Maruyama and T. Nakai, Appl. Phys. Lett. 58, 2079 ͑1991͒.
20 J. A. Glass, Jr., S. S. Kher, and J. T. Spencer, Chem. Mater. 4, 530 ͑1992͒.
21 I. Yoshida, H. Kobayashi, and K. Ueno, Bull. Chem. Soc. Jpn. 47, 2203
͑1974͒.
22 N. S. Al-Niaimi and H. A. Hamid, J. Inorg. Nucl. Chem. 39, 849 ͑1977͒.
23 E. W. Berg and N. M. Herrera, Anal. Chim. Acta 60, 117 ͑1972͒.
24 J. Saraie, K. Ono, and S. Takeuchi, J. Electrochem. Soc. 136, 3139 ͑1989͒.
25 T. Tsuchida and H. Takahashi, J. Mater. Res. 9, 2919 ͑1994͒.
26 C. D. Wagner, D. A. Zatko, and R. H. Raymond, Anal. Chem. 52, 1445
͑1980͒.
27 V. E. Henrich and P. A. Cox, The Surface Science of Metal Oxide ͑Cam-
bridge University Press, Cambridge, 1994͒, p. 139.
of films grown under Ar/H2O and of 0.02 evaluated for the
Published without author corrections
1626
Appl. Phys. Lett., Vol. 67, No. 11, 11 September 1995
Ciliberto et al.
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:
130.239.20.174 On: Mon, 24 Nov 2014 20:06:57