IUVSTA 15th International Vacuum Congress (IVC-15), AVS 48th International Symposium (AVS-48), 11th International Conference on Solid Surfaces (ICSS-11)
    Dielectrics Tuesday Sessions
       Session DI-TuM

Paper DI-TuM6
Kinetic Study on ZrO@sub 2@-CVD using Zirconium-Tetra-tertiary-Butoxide

Tuesday, October 30, 2001, 10:00 am, Room 130

Session: High K Dielectrics II
Presenter: T. Kawamoto, University of Tokyo, Japan
Authors: T. Kawamoto, University of Tokyo, Japan
Y. Shimogaki, University of Tokyo, Japan
Correspondent: Click to Email
ZrO@sub 2@ films have lately attracted much attention as high-k dielectrics for high performance ULSI gate applications. We carried out kinetic study on ZrO@sub 2@-CVD by using hot-wall tubular reactor to understand the basic reaction mechanisms. Zirconium-Tetra-tertiary-Butoxide (ZTB) was used as the Zr source and injected into the reactor by bubbling with He. We investigated the fractured surface of the sample by FE-SEM to obtain the growth-rate and step coverage of deposited ZrO@sub 2@ films on Si substrate with microtrenches. The deposition profile at 663K showed the increase in the up-stream part, and then the decrease was observed in the down-stream part. This implies that this reactive process has the special path, precursors decomposed into intermediate species by the gas-phase reaction, and the intermediate species deposits on growing surface. Furthermore, we examined depositions into the trench substrates with different aspect ratio at several positions in the reactor, and could found out that the step coverage became poorer, as samples were laid further from inlet. As a result of carrying out fitting these experimentally observed step coverage profiles with a single species model, the predicted composition profiles did not match well with the all observed profiles. This result indicates that some growth species contribute to film growth. Then, by fitting with two species model, we determined reactive sticking coefficients of two growth species; @eta@@sub 1@=0.0095 and @eta@@sub 2@=0.20, respectively. And we found out that the less active growth species (@eta@@sub 1@=0.0095) existed dominantly in the up-stream part, on the other hand, the high active growth species (@eta@@sub 2@=0.20) did the same role in the down-stream part. The dependence of the step coverage quality on the distance from inlet could be explained by the relative ratio of these species. The film properties may have different characteristics by this precursor ratio.