The XRD spectra from samples treated at the temperature interval of 400 to 1000°C for one hour, show a progressive reduction in intensities and number of titanium peaks and substrate phases, and on the other hand, a progressive increase in intensities and number of the titanium carbide (TiC) reflections. This phase has no preferential orientation. We also notice the apparition of iron, titanium and chromium oxides at high temperatures (Fig. 3).

 Since titanium has higher efficiency to carbon than both Cr and Fe, the formation of TiC is due to the reaction between carbon from substrate and titanium thin film. It results a consumption of the titanium layer and the increase of TiC peaks in number and intensity with annealing temperature. The apparition of iron, titanium and chromium oxides, Fe2O3, TiO2, Cr2O3, is probably caused by insufficient applied conditions during the thermal treatments (insufficient vacuum) particularly at higher temperatures, were the formation of these oxides is favored. This may indicate that these films have an open structure with voids along the grain boundaries. The same structure has been observed for TiN by Al-jaroudi [5].

 3. 2. Auger electron spectroscopy

 AES depth profiles of 100C6/Ti sample before annealing are shown in figure 4. Because of the natural oxidation of titanium film during their storage at room temperature before AES analysis, one observes a high content of carbon and oxygen in the superficial region of the samples. After elimination of the contaminated layer, the titanium film seems have a relatively high purity and a constant concentration up to the interface. This justifies the homogeneity of thin films achieved by the magnetron sputtering method. The slight increase of the oxygen content at the interface is caused either by the adsorption of the oxygen on the substrate surface before the Ti deposition or by its incorporation into the Ti layer during the deposition process. Gheriani [6] showed the effect of substrate surface ion bombardment etching on reaction between thin films and substrates.

 Figure 5 shows AES profiles of 100C6/Ti samples heat treated at 700°C during 60 minutes. It reveals that the titanium concentration decreases significantly and that of carbon increases. This fact leads to the reaction of these elements and tothe formation and growth of TiC carbide. It can be also noted the diffusion of Ti towards the steel substrate, and carbon and iron towards the thin coating layer.

 In the rich region in Ti, up to 60 minutes of ion etching with argon ions, we observe that the titanium concentration decrease progressively with a low speed. However, in the region from 60 to 85 minutes of sputtering, we note a speed decrease in titanium concentration. We can say that the reaction starts at the substrate- titanium film interface, and then develops towards the superficial layers of the sample, as observed in Gheriani [7] work.

 

3. 4. Micro-hardness measurement

 Micro hardness tests were performed by Vickers method under small load (40g). Figure 8 represents the micro-hardness variation with annealing temperature for substrate and coated samples. One notices for the curve corresponding to the substrate, an increase of the micro-hardness until a maximum value of 600 Kg/mm2 at 700°C, then a relatively soft decrease. However, the micro-hardness values of coated sample are more important and reach their maximum of 3200 Kg/mm2 at 900°C, then begin to decrease. The observed increase of the micro-hardness with annealing temperature is due to the formation and the growth of titanium carbide. However, the reduction of the micro hardness after the maximal value is probably caused by the effect of soft ferrite formation beneath of the TiC layer, since carbon diffuses from region lying near the substrate surface. So, when hard TiC layer is laying on soft ferrite, during the micro hardness measurements (even when using small charge P = 40 g) ferrite is deformed and it results in reading of smaller values of micro hardness. Besides, the formation of iron, titanium and chromium oxides can also contribute to the decrease of the micro hardness.

4. Conclusion

 Titanium carbide coatings are obtained from a film of titanium deposited by magnetron sputtering method on steel substrates (1%wt of carbon content) and annealed in vacuum at different temperatures. TiC is forming by high diffusion rate of carbon from substrate towards titanium thin film in the studied temperature range. The formation of titanium carbide is accompanied by an important increase of the micro-hardness.

 Acknowledgments

 The authors thank Prof. Marie Paule Delplancke-Ogletree and M. Lazlo Szabo from Industrial Chemistry Department, ULB University (Bruxelles-Belgium) for their help in XRD and SEM analysis. Special thanks are due to Prof. P. M. Ossi, Department of Nuclear Engineering Polytechnic of Milano, 20133 Milano Italy, for useful discussions relating to this work.

        References

 [1] Teghil R.; Applied Surface Science, Vol. 4, N°173, pp. 233-241 (2001).

 [2] Bhatt H. D.; Thin Solid Films, Vol. 8, N° 342, pp. 214-220 (1999).

 [3] Mitsuo A.; Surface and Coatings Technology, Vol. 6, N° 98, pp. 98-103 (1998).

 [4] Méçabih S.; Physica A, Vol. A, N° 285, pp. 392-396 (2000).

 [5] Al-Jaroudi M. Y.; Thin Solid Films Vol. 40, N° 195, pp. 265-277 (1991).

 [6] Gheriani R.; Sciences & Technologie, Univ. Constantine Vol. 1, N°9, pp. 19-22 (1998).

 [7] Gheriani R.; Surface and Coatings Technology, Vol. 180, N° 181, pp. 49-52 (2004).