X-ray Photoelectron Spectroscopic Studies of Cu-AI alloy/SiO 2 Interfaces
- PDF / 1,049,253 Bytes
- 6 Pages / 417.6 x 639 pts Page_size
- 82 Downloads / 151 Views
generation of interfacial layers through copper alloying. Small amounts of alloying additions to copper have been shown to suppress copper diffusion into SiO 2, without significantly affecting the resistivity of copper.' Previous studies have shown that Al is one of the best choices for an alloying additive to copper.'2 The addition of I at.% Al to Cu has been shown to improve copper's oxidation resistance by diffusion of Al to the surface. Here, the films remained3 microscopically smooth and that there was no detectable diffusion of Cu into the Si0 2 substrate. The Cu-Al alloy/SiO2 interfacial reaction is a crucial issue to be investigated due to stress, diffusion, adhesion and aging effects. Therefore, the clarification of the characteristics of this interface would help increase the understanding of the mechanism of interface formation for this class of materials. In this work, we investigate the interfacial reaction of Cu-Al alloy/Si02 interfaces by XPS analysis of a buried interface. The chemical states of A!/SiO 2 interfaces of sputter and e-beam deposited films are examined and compared with those of Cu-AI alloy/SiO2 interfaces. The interface formation during BTS is also examined by XPS. This paper reports the results of our investigation of the interfacial reaction of Cu-Al alloy/SiO 2 interfaces. Our XPS data shows that the interfacial layer is composed of Al-O and Cu-O-Al bonds. The interface formation initiates at room temperature and develops during further thermal anneal and/or BTS treatment.
347 Mat. Res. Soc. Symp. Proc. Vol. 564 © 1999 Materials Research Society
EXPERIMENT P-typed, silicon wafers were used for all experiments. After RCA cleaning, a high quality thermal oxide was grown in dry oxygen to a thickness about 1OOOA. Thin films of pure aluminum were either sputter deposited at 2kW or deposited using an electron-beam evaporation technique. Thin films of Cu with various Al contents were sputter deposited either by cosputtering Cu and Al or by using a Cu alloy target. The deposition conditions for Cu-5at.%Al were as follows. The sputtering powers on Cu and Al targets were 2 kW and 0.24 kW respectively. The base pressure in the vacuum chamber for all depositions was in the range of 10-7 Torr. The argon pressure during sputtering was 5 mTorr. The sputtering power on the Cu-lat.%A1 alloy target was 1.5 kW. These blanket films were directly examined by XPS. The MOS capacitors were formed by sputtering through a shadow mask. Pure Cu was sputter deposited on Cu-Al alloy films in sequence without breaking the vacuum and at a sputtering power of 2.5 kW. The sample geometry was 7000 A Cu/600 A Cu-5at.%Al/1000 A SiO 2 and
7000 A Cu/3000 A Cu-lat.%Al/1000 A SiO 2. MOS capacitors were investigated at varying times using BTS with an electric field of 1.5 MV/cm at 250'C on a thermal chuck while N 2 was flowing into the chamber. A computer-controlled testing system consisting of a HP 4280A C-V meter and HP 4140B I-V meter performed the C-V and I-V measurements respectively. High frequency C-V characteris
Data Loading...
