C.M. Comrie

Pt redistribution during Ni(Pt) silicide formation

We report on a real-time Rutherford backscattering spectrometry study of the erratic redistribution of Pt during Ni silicide formation in a solid phase reaction. The inhomogeneous Pt redistribution in Ni(Pt)Si films is a consequence of the low solubility of Pt in Ni2Si compared to NiSi and the limited mobility of Pt in NiSi. Pt further acts as a diffusion barrier and resides in the Ni2Si grain boundaries, significantly slowing down the Ni2Si and NiSi growth kinetics. Moreover, the observed incorporation of a large amount of Pt in the NiSi seeds indicates that Pt plays a major role in selecting the crystallographic orientation of these seeds and thus in the texture of the resulting Ni1−xPtxSi film.

Dominant diffusing species during cobalt silicide formation

The dominant moving species during cobalt monosilicide and cobalt disilicide formation has been examined using a thin tantalum layer as a metal marker. The marker data obtained following the formation of CoSi from Co2Si showed that monosilicide growth was essentially due Si diffusion only. When used to study CoSi2 formation, the data indicated that silicon was also the dominant moving species during disilicide formation, although a noninsignificant amount of cobalt diffusion was also observed to take place.

The influence of Pt redistribution on Ni1−xPtxSi growth properties

We have studied the influence of Pt on the growth of Ni silicide thin films by examining the Pt redistribution during silicide growth. Three different initial Pt configurations were investigated, i.e., a Pt alloy (Ni+Pt/⟨Si⟩), a Pt capping layer (Pt/Ni/⟨Si⟩) and a Pt interlayer (Ni/Pt/⟨Si⟩), all containing 7 at. % Pt relative to the Ni content. The Pt redistribution was probed using in situ real-time Rutherford backscattering spectrometry (RBS) whereas the phase sequence was monitored during the solid phase reaction (SPR) using in situ real-time x-ray diffraction. We found that the capping layer and alloy exhibit a SPR comparable to the pure Ni/⟨Si⟩ system, whereas Pt added as an interlayer has a much more drastic influence on the Ni silicide phase sequence. Nevertheless, for all initial sample configurations, Pt redistributes in an erratic way. This phenomenon can be assigned to the low solubility of Pt in Ni2Si compared to NiSi and the high mobility of Pt in Ni2Si compared to pure Ni. Real-time RBS furt...

Marker and radioactive silicon tracer studies of PtSi formation

Marker and radioactive 31Si experiments have been performed to investigate atomic diffusion during PtSi formation. The marker work used a thin metallic layer (Ti, Co, Ni) as a marker. Analysis of the marker displacement indicated growth dominated by silicon diffusion (∼90%). The interpretation of data from the radioactive tracer experiments is less clear cut. However, when examined in conjunction with the marker results, it would appear that either PtSi growth took place by silicon substitutional diffusion or by a mixed interstitial mechanism (i.e., a mixture of interstitial and interstitialcy diffusion). Arguments are presented to suggest that silicon vacancy diffusion during silicide growth is the most likely mechanism. This interpretation is found to be generally consistent with other recently published work on PtSi formation.

Stability of NiSi2 and CoSi2 in contact with their free metal

Rutherford backscattering has been used to study metal/disilicide thin‐film interactions for Ni and Co. Upon heating, the metal reacted with disilicide to produce the phase M2Si in both cases. On further heating the M2Si itself reacted with the disilicide to form MSi. In the case of Co it was found that after all the metal had been converted to CoSi in this way, the reaction stopped. However, with Ni the disilicide substrate continued to dissociate into NiSi and Si even after all the original Ni had reacted to form NiSi. The stability of NiSi2 under various conditions was investigated and it appears that twin requirements of a crystalline silicon substrate on which the excess Si can regrow and nucleation sites in the form of NiSi are necessary in order to induce dissociation.

Diffusion of silicon in Pd2Si during silicide formation

Inert and radioactive markers have been used to study the mechanism of diffusion during Pd2Si formation. With the aid of Ti as an inert marker it has been shown that silicon is the dominant diffusing species during polycrystalline Pd2Si formation. When radioactive silicon is used as a marker it is found that the radioactive silicon is uniformly distributed throughout the Pd2Si after silicide formation. The self‐diffusion of silicon in Pd2Si was investigated and found to be much lower than that necessary to produce a uniform radioactive silicon distribution, had silicon diffused by a grain boundary or pure interstitial mechanism. It is therefore proposed that silicon diffuses by a vacancy mechanism during silicide formation.

Nucleation and diffusion during growth of ternary Co1−xNixSi2 thin films studied by complementary techniques in real time

The growth kinetics of ternary Co1−xNixSi2 thin films was studied in real time. The “Kissinger” method was applied to the results of ramped sheet resistance measurements to extract the apparent activation energy for the growth process. By simultaneously acquiring sheet resistance, x-ray diffraction and laser light scattering data on one hand and combining resistance measurements and Rutherford backscattering spectrometry on the other hand, we could distinguish between the initial, nucleation controlled thin film growth, and the subsequent diffusion controlled growth. The apparent activation energy for the initial growth decreases with increasing Ni concentration as a result of a lower nucleation barrier for the ternary disilicide. The markedly different microstructure of the ternary Co1−xNixSi2 films with respect to pure CoSi2 layers lies at the origin of a lower activation energy for the diffusion controlled growth of the ternary films. Despite the low activation energy, these films grow at a much slower...

Study of Pt/Ge interaction in a lateral diffusion couple by microbeam Rutherford backscattering spectrometry

Abstract A scanning electron microscope and nuclear microprobe have been used to investigate the interaction between germanium and platinum in a lateral diffusion couple. When an island of germanium on a platinum film was annealed at 500°C three distinct phases were observed to form, two inside and one outside the original island. Micro-Rutherford backscattering spectrometry carried out with the nuclear microprobe suggested that these phases were Pt2Ge, Pt3Ge2 and PtGe (with the latter forming outside the original island). Micro-Rutherford backscattering spectrometry also showed that after reaction the platinum was not uniformly distributed throughout the film in the island region, nor was it uniformly distributed in the reaction zone. Information regarding the thickness of the reacted films, and of the distribution of the various components in the film, are readily accessible via micro-Rutherford backscattering spectrometry, thus demonstrating the power of this technique for lateral diffusion studies.

Simultaneous real-time x-ray diffraction spectroscopy, Rutherford backscattering spectrometry, and sheet resistance measurements to study thin film growth kinetics by Kissinger plots

When the Kissinger method is used to investigate thin film growth kinetics, activation energies obtained are often significantly higher than those of Arrhenius plots based on isothermal studies. The reason for the higher activation energies is related to the sensitivity of the Kissinger analysis to nucleation effects. In fact, this often undesirable effect opens the possibility of studying nucleation barriers in a semiquantitative way. Furthermore, we show that these nucleation effects can be filtered out by a more careful application of the Kissinger method, and activation energies that are consistent with Arrhenius plots are then obtained.

On the growth kinetics of Ni(Pt) silicide thin films

We report on the effect of Pt on the growth kinetics of δ-Ni2Si and Ni1−xPtxSi thin films formed by solid phase reaction of a Ni(Pt) alloyed thin film on Si(100). The study was performed by real-time Rutherford backscattering spectrometry examining the silicide growth rates for initial Pt concentrations of 0, 1, 3, 7, and 10 at. % relative to the Ni content. Pt was found to exert a drastic effect on the growth kinetics of both phases. δ-Ni2Si growth is slowed down tremendously, which results in the simultaneous growth of this phase with Ni1−xPtxSi. Activation energies extracted for the Ni1−xPtxSi growth process exhibit an increase from Ea = 1.35 ± 0.06 eV for binary NiSi to Ea = 2.7 ± 0.2 eV for Ni1−xPtxSi with an initial Pt concentration of 3 at. %. Further increasing the Pt content to 10 at. % merely increases the activation energy for Ni1−xPtxSi growth to Ea = 3.1 ± 0.5 eV.