2540
Appl. Phys. Lett., Vol. 76, No. 18, 1 May 2000
Lee et al.
Si, the W layer was removed using a boiling H O solution,
2
2
before XPS analysis. Before sputtering the surface of the
sample, a strong peak of Si in the SiO peak was detected,
2
probably due to the oxide formed during the H O treatment.
2
2
However, a peak indicating the silicon nitride ͑Si–N͒ bond-
ing status was clearly seen after sputtering the sample for 5 s,
and a peak of pure silicon appeared with further sputtering.
This result implies that the segregated nitrogen has a Si–N
bonding status.
The thermal stability of the W N /poly-Si structure
6
7
33
treated by RTA at 1000 °C under nitrogen atmosphere was
evaluated at the annealing temperatures of 850, 1000, and
1
050 °C in a hydrogen environment for 2 min. It was found
that the RTA-treated W N /poly-Si structure was stable up
6
7
33
to 1000 °C for 2 min annealing, as shown in Fig. 6, due to
the nitrogen-segregated layer acting as a barrier between W
and poly-Si.
FIG. 6. X-ray diffraction spectra of the rapidly thermal annealed
W N /poly-Si structure after hydrogen annealing at temperatures of ͑a͒
67
33
8
50, ͑b͒ 1000, and ͑c͒ 1050 °C for 2 min.
In summary, the structure of W N /poly-Si converted
6
7
33
into ␣-phase W/nitrogen-segregated layer/poly-Si layers by
RTA at 1000 °C for 1 min. The ␣-phase W layer was used as
a low-resistive electrode, and the nitrogen-segregated layer at
the interface of W/poly-Si acted as an in situ formed diffu-
sion barrier, which successfully suppressed the silicidation
between W and poly-Si. The nitrogen-segregated layer was
an amorphous structure and revealed a Si–N bonding status.
observe any peaks on WSi with the samples tested, even
after high-temperature annealing at 1000 °C ͑Fig. 2͒. To in-
x
vestigate the causes of the lack of formation of WSi , the
x
behavior of nitrogen was investigated using SIMS analysis.
Ϫ3
Equilibrium solubility of nitrogen in W is about 10 at. %
at 1000 °C. This indicates that excessive nitrogen in the
W N films may be diffused out and also redistributed after
6
7
33
RTA at 1000 °C.
Figure 3 shows the SIMS depth profiles of tungsten,
oxygen, silicon, and nitrogen before and after RTA at
1
W.-K. Yeh, Y.-C. Shiau, and M.-C. Chen, J. Electrochem. Soc. 144, 214
͑
1997͒.
2
3
4
J. B. Lasky, J. S. Nakos, O. J. Cain, and P. J. Geiss, IEEE Trans. Electron
Devices ED-38, 262 ͑1991͒.
J. S. Byun, B. H. Lee, J.-S. Park, D. K. Sohn, S. J. Choi, and J. J. Kim, J.
Electrochem. Soc. 145, 3228 ͑1998͒.
H. Wakabayashi, T. Andoh, K. Sato, K. Yoshida, H. Miyamoto, T.
Mogami, and T. Kunio, Tech. Dig. Int. Electron Devices Meet., 447
͑1996͒.
M. T. Takagi, K. Miyashita, H. Koyama, K. Nakajima, K. Miyano, Y.
Akasaka, Y. Hiura, S. Inaba, A. Azuma, H. Koike, H. Yoshimura, K.
Suguro, and H. Ishiuchi, Tech. Dig. Int. Electron Devices Meet., 455
͑1996͒.
Y. Akasaka, S. Suehiro, K. Nakajima, T. Nakasugi, K. Miyano, K. Kasai,
H. Oyamatsu, M. Kinugawa, M. Takayanagi, M. Takagi, K, Agawa, F.
Matsuoka, M. Kakumu, and K. Suguro, Proceedings of the 12th Interna-
tional VLSI Multilevel Interconnection Conference ͑VMIC’95͒, Santa
Clara, CA ͑1995͒, p. 168.
W. A. Metz, N. J. Szluk, G. Miller, and H. O. Hayworth, IEEE Electron
Device Lett. EDL-6, 372 ͑1985͒.
C. W. Lee, Y. T. Kim, and J. Y. Lee, Appl. Phys. Lett. 64, 619 ͑1994͒.
S.-L. Zhang, R. Palmans, J. Keinonen, C. S. Petersson, and K. Maex,
Appl. Phys. Lett. 67, 2998 ͑1995͒.
1000 °C for 1 min. The intensities of nitrogen were normal-
ized by the intensity of tungsten and silicon on the right axis.
In the case of the as-deposited film, the concentration of
nitrogen was uniformly distributed in depth, but after RTA at
1000 °C, most of the nitrogen atoms were dissipated and par-
5
6
tially segregated at the W/poly-Si interface, as in Fig. 3͑b͒.
To evaluate the characteristics at the interface of the WNx
and poly-Si, TEM analysis was conducted with the sample
annealed at 1000 °C. As shown in Fig. 4͑a͒, the interface of
W and poly-Si was extremely uniform and there was no
layer of WSi . The nitrogen-segregated layer is clearly seen
x
in the high-resolution TEM photograph in Fig. 4͑b͒. The
thickness of the segregated layer was about 2 nm and the
structure was amorphous. From the analysis results of SIMS
and TEM, it may be concluded that the segregated layer acts
as a barrier protecting any interaction between W and poly-
Si, even at such a high RTA temperature of 1000 °C.
In order to identify the chemical bonding status of the
segregated layer, XPS analysis was performed ͑Fig. 5͒. To
avoid the ‘‘interface effect’’ at the interface of W and poly-
7
8
9
10
Y. Hiura, A. Azuma, K. Nakajima, Y. Akasaka, K. Miyano, H. Nitta, A.
Honjo, K. Tsuchida, Y. Toyoshima, K. Suguro, and Y. Kohyama, Tech.
Dig. Int. Electron Devices Meet., 389 ͑1998͒.
K. Affolter, H. Kattelus, and M.-A. Nicolet, Mater. Res. Soc. Symp. Proc.
47, 167 ͑1985͒.
11
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:
42.104.240.194 On: Tue, 16 Dec 2014 16:29:59
1