A Caltech Library Service

The effects of oxygen on the formation of Ni, Pd, and Pt silicides


Scott, David Martin (1982) The effects of oxygen on the formation of Ni, Pd, and Pt silicides. Dissertation (Ph.D.), California Institute of Technology.


NOTE: Text or symbols not renderable in plain ASCII are indicated by [...]. Abstract is included in .pdf document.

A systematic study of the effects of implanted oxygen on the formation of Ni, Pt and Pd silicides has been carried out using 4He+ backscattering spectrometry (BS), [...] nuclear reaction analysis (NRA) and x-ray photoelectron spectroscopy (XPS) for the analysis. A detailed presentation of the NRA technique is given as this technique was central to this study and is not as generally known as are BS and XPS (Chapter II). The depth resolution of this technique is found to be ~ 150 to 200 Å in Ni. The oxygen sensitivity is essentially limited by measurement time (~ 10 hrs) to ~ 10(19) to 10(20) 0/cm2.

The effect of oxygen on the formation of Ni2Si is shown to vary with the initial location of the oxygen (Chapter III). Ni is the dominant diffusing species in Ni2Si formation. Oxygen initially located in the Ni film is found to build up during annealing at the Ni-Ni2Si interface until a diffusion barrier to the Ni is formed. The XPS data shows this barrier to be SiO2. Oxygen picked up during annealing from the ambient also plays a role in the buildup of this barrier. Once Ni2Si growth is halted, the second phase NiSi begins to nucleate and grow. The result is the simultaneous presence of Ni, Ni2Si and NiSi in the implanted samples. The threshold dose [...] necessary for barrier formation is ~ 1.2 x 10(16) 0/cm2. This is equivalent to ~ 26 Å of stoichiometric fused quartz if present as a layer. For the oxygen initially present in the Si, the oxygen is incorporated into the Ni2Si layer without an interfacial accumulation taking place. The relative reduction in oxygen density parallels that of the Si density as Si forms Ni2Si.

We model these observations in terms of the asymmetries that are present in this system with regard to the moving species in Ni2Si formation and with regard to the chemical reactivity of oxygen with Ni and Si.

The effects of implanted oxygen on NiSi formation are studied for the case of the oxygen initially present in the Ni2Si film on a Si <100> substrate. Upon annealing, NiSi grows with square root of time in both implanted and unimplanted samples. This disagrees with the linear rate previously reported. The growth is slightly slower for implanted samples. The slowing is uncorrelated with the amount of oxygen, suggesting that a structural change due to ion implantation is the cause. During NiSi formation, oxygen is incorporated into the NiSi film without interfacial accumulation, but the oxygen distribution is seen to move towards the surface. This motion is explained in terms of a simple model based on the chemical affinity of oxygen to Si and Ni and the fact that Ni is the moving species in NiSi growth. The shift in the oxygen peak position during NiSi formation enables the implanted oxygen to act as a diffusion marker. This confirms that Ni is the diffusing species in NiSi formation. The use of implanted oxygen as a diffusion marker in thin film studies is briefly explored.

We have also studied the effect of impurity oxygen initially present in a Pt film on Pt2Si formation (Chapter IV). We found that the redistribution of the oxygen during annealing, subsequent barrier formation and threshold oxygen dose were all identical to that of the corresponding Ni case. This result is shown to be consistent with the asymmetries present in the chemistry of oxygen relative to Pt and Si and in the initial location of the oxygen relative to the moving species (Pt). The case of oxygen initially located in the Si was not investigated as this case most likely is also identical to the corresponding Ni case.

For the impurity oxygen initially present in a Pd film on a Si <100> substrate (Chapter V), we found that the oxygen is incorporated into the Pd2Si without a diffusion barrier being formed. The implanted oxygen has no effect on the growth of Pt2Si. We find oxygen to be mobile in the Pd at an annealing temperature of only 250°C. Upon annealing the oxygen diffuses to the Pd-Pt2Si interface, to react and form SiO2 there. Simultaneously Si diffuses through the Pd2Si layer and forms additional silicide at the Pd-Pd2Si interface thereby incorporating the SiO2 into the Pd2Si. Thus barrier formation does not occur. We also show that our results identify Si as the dominant diffusing species during Pd2Si formation rather than both Pd and Si as previously reported.

Generalizing from the result of our study of the effects of impurity oxygen on the formation of the silicides of Ni, Pt and Pd we present a conceptual framework for impurity effects in metal silicide formation (Chapter VI). This model relies on the asymmetries present with regard to the initial location of the impurity relative to the moving species and with regard to the chemical affinities of the impurity relative to the reacting species. The results for impurity N in the Ni-Si system are briefly compared with predictions of the model and shown to agree.

Further work suggested by this model is summarized next (Chapter VII). The cases covered are impurity N in the Ni-Si, Pt-Si and Pd-Si systems along with impurity C in the Ni-Si and Pt-Si systems. Also covered are the preliminary results of contact restivity measurements after barrier formation has occurred for the cases of impurity N and O in the Ni-Si system. The results of this additional work are shown to be consistent with the model.

Some parts of this thesis have already been published under the following titles:

1. "The Effect of oxygen on the Growth Kinetics of Nickel Silicides," D. M. Scott, P. J. Grunthaner, B. Y. Tsaur, M-A. Nicolet and J. W. Mayer, in Proceedings of the Symposium on Thin Film Interfaces and Interactions, J. E. E. Baglin and J. M. Poate, eds., (The Electrochemical Soc. Princeton, Vol. 80-2, 148 (1980)

2. "Modification of Nickel Silicide Formation by Oxygen Implantation," D. M. Scott and M-A. Nicolet, Nuct. Inst. Meth. 182/183, 655 (1981)

3. "Implanted Oxygen in NiSi Formation," D. M. Scott and M-A. Nicolet, Phys. Stat. Sol. (a), 66, 773 (1981)

4. "Oxygen Impurity Effects at Metal/Silicide Interfaces: Formation of Silicon Oxide and Suboxides in the Ni/Si System," P. J. Grunthaner, F. J. Grunthaner, D. M. Scott, M-A. Nicolet, and J. W. Mayer, J. Vac. Sci. Tech. 19, 641 (1981)

5. "Alteration of Ni Silicide Formation by N Implantation," L. Wielunski, D. M. Scott, M-A. Nicolet and H. von Seefeld, Appl. Phys. Lett., 38, 106 (1981)

6. "Retardation and Suppression of Nickel Silicide Formation by N+ Implantation," D. M. Scott, L. Wielunski, H. von Seefeld and M-A. Nicolet, Nucl. Inst. Meth., 182/183 661 (1981)

Item Type:Thesis (Dissertation (Ph.D.))
Degree Grantor:California Institute of Technology
Division:Engineering and Applied Science
Major Option:Applied Physics
Thesis Availability:Restricted to Caltech community only
Research Advisor(s):
  • Nicolet, Marc-Aurele (advisor)
  • Mayer, James W. (co-advisor)
Thesis Committee:
  • Unknown, Unknown
Defense Date:3 March 1982
Record Number:CaltechETD:etd-09062006-140245
Persistent URL:
Default Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:3361
Deposited By: Imported from ETD-db
Deposited On:25 Sep 2006
Last Modified:26 Dec 2012 02:59

Thesis Files

[img] PDF (Scott_dm_1982.pdf) - Final Version
Restricted to Caltech community only
See Usage Policy.


Repository Staff Only: item control page