[PDF]
http://dx.doi.org/10.3952/lithjphys.44508
Open access article / Atviros prieigos straipsnis
Lith. J. Phys. 44, 375–379 (2004)
MOCVD KINETICS OF STRONTIUM
BISMUTH TANTALATE THIN FILM GROWTH
M. Šilinskas, M. Lisker, B. Kalkofen, S. Matichyn, and E. Burte
Institute of Micro- and Sensor Systems, Otto von Guericke
University, Universitätsplatz 2, 39106 Magdeburg, Germany
Received 26 May 2004
Dedicated to the 100th anniversary of Professor K. Baršauskas
The thin films of BiOx, SrxTayOz,
and strontium bismuth tantalate (SBT) were deposited by
metalorganic chemical vapour deposition (MOCVD) on 150 mm silicon
(100) wafers. Some of the wafers were pre-deposited with Pt
electrodes. The substrate temperature and the deposition pressure
were varied from 300 to 600 °C and from 0.35 to 7 mbar,
respectively. Bi(/\//)3 (triallyl bismuth) and
Sr[Ta(OEt)5(OC2H4OMe)]2
(strontium bis[tantalum(pentaethoxy)(2-methoxyethoxide)] were used
as a Bi precursor and as a Sr–Ta precursor, respectively. A liquid
delivery system was used to supply and to evaporate the precursor
into the reactor. X-ray photoelectron spectroscopy (XPS) and
ellipsometry were carried out to characterize the film properties.
The growth rate of the MOCVD of BiOx and SrxTayOz
was compared to the growth rate of SBT to obtain information about
mutual interaction between precursors. The deposition rate of
bismuth oxide thin films was low (∼10 nm/h at 0.35 mbar) and did
virtually not depend on the temperature. On the contrary, the
growth rate of strontium tantalate films depended strongly on the
temperature. The deposition rate of the SBT films was similar to
the bismuth oxide film deposition, which slightly increased with
increasing substrate temperature. However, the deposition rate of
SBT was always lower than the deposition rate of the single
precursors. The growth rate significantly depends on pressure. The
decrease of the deposition pressure in the reactor chamber reduces
the deposition rate of BiOx, SrxTayOz,
and SBT, but on the other hand, it improves the uniformity of the
film thickness. The XPS measurements showed a deficit of bismuth
in the SBT films even though the concentration of the Bi precursor
was several times higher. The XPS depth-profiling by Ar+
ion sputtering indicated different bond characteristics of Ti, Sr,
and Bi before and after ion beam bombardment.
Keywords: metalorganic chemical vapour deposition, SBT, XPS
PACS: 81.15.Gh, 77.84.–s, 79.60.–i
STRONCIO BISMUTO TANTALATO DANGŲ
AUGINIMO KINETIKA, NAUDOJANT METALOORGANINĮ CHEMINĮ GARŲ
NUSODINIMO METODĄ
M. Šilinskas, M. Lisker, B. Kalkofen, S. Matichyn, E. Burte
Otto von Guericke universitetas, Magdeburgas, Vokietija
BiOx, SrxTayOz
ir stroncio bismuto tantalato (SBT) dangos buvo sudaromos and 150
mm silicio (100) padėklo, naudojant metaloorganinį cheminį garų
nusodinimo metodą iš Bi(/\//)3 (trialilbismuto) ir iš
Sr[Ta(OEt)5(OC2H4OMe)]2
(stroncio bis[tantalo (pentaetoksi)(2-metoksietoksido)]). Dangoms
apibūdinti naudojome spektroskopinę elipsometriją ir Röntgen’o
spindulių fotoelektronų spektroskopiją (RFS). Norint teisingai
parinkti SBT auginimo sąlygas, iš pradžių tyrėme bismuto oksido
bei stroncio tantalato auginimo kinetiką. Paaiškėjo, kad bismuto
oksidas auga lėtai (∼10 nm/h, esant 0,35 mbar slėgiui), o jo
auginimo sparta beveik nepriklauso nuo temperatūros. Priešingai,
stroncio titanato auginimo sparta labai priklauso nuo
temperatūros. SBT auginimo sparta buvo mažesnė nei BiOx,
SrxTayOz, ir
ji buvo būdingesnė BiOx. Dangų auginimo sparta
taip pat labai priklausė ir nuo slėgio kameroje: didėjant slėgiui,
didėjo nusodinimo sparta, bet kartu didėjo ir plėvelės storio
netolygumai. Naudojant RFS, parodyta, kad dangose trūksta bismuto,
net jei Bi(/\//)3 koncentracija du kartus didesnė už
Sr[Ta(OEt)5(OC2H4OMe)]2
koncentraciją.
References / Nuorodos
[1] J. Baliga, Semicond. Intern. 23, 83 (2000),
http://dx.doi.org/10.5414/ALP23083
[2] L. Mendicino, V. Vartanian, B. Goolsby, and P. Brown, Semicond.
Intern. 25, 61 (2002),
[3] T. Li, Y. Zu, S.B. Desu, C.-H. Peng, and M. Nagata, Appl. Phys.
Lett. 68, 616 (1996).,
http://dx.doi.org/10.1063/1.116486
[4] S. Zafar, V. Kaustik, P. Laberge, P. Chu, R. Hance, P. White, D.
Taylor, B. Melnick, and S. Gillespie, J. Appl. Phys. 82,
4469 (1997),
http://dx.doi.org/10.1063/1.366179
[5] K. Amanuma, T. Hase, and Y. Miyasaka, Appl. Phys. Lett. 66,
221 (1995),
http://dx.doi.org/10.1063/1.113140
[6] D. Wang, T. Yu, A. Hu, D. Wu, A. Li, Z. Liu, and N. Ming, Appl.
Phys. Lett. 79, 2237 (2001),
http://dx.doi.org/10.1063/1.1406979
[7] N. Nukaga, M. Mitsuya, and H. Funakubo, Jpn. J. Appl. Phys. 39,
5496 (2000),
http://dx.doi.org/10.1143/JJAP.39.5496
[8] R. Dat, J. Lee, O. Auciello, and A. Kingon, Appl. Phys. Lett. 67,
572 (1995),
http://dx.doi.org/10.1063/1.115173
[9] N. Nukaga, M. Saito, K. Okuwada, and H. Fanakubo, Jpn. J. Appl.
Phys. 41, 710 (2002),
http://dx.doi.org/10.1143/JJAP.41.L710
[10] Y. Park, K. Min, K. Cho, S. Heo, C. Lim, M. Lee, T. Lee, H.
Kim, and Y. Kim, J. Vac. Sci. Technol. A 19, 916 (2001),
http://dx.doi.org/10.1116/1.1354601