Intermetallics 15 (2007) 805e809www.elsevier.com/locate/intermet

Deposition of Cr-modified silicide coatingson NbeSi system intermetallics

Chen Chen, Chungen Zhou, Shengkai Gong, Shusuo Li, Yue Zhang, Huibin Xu*

School of Materials Science and Engineering, Beijing University of Aeronautics and Astronautics, Beijing 100083, PR ChinaAvailable online 8 December 2006

Abstract

In order to improve the oxidation resistance of NbeSi system intermetallics, silicide was deposited on the substrate by molten salt or by packcementation, and Cr was deposited by pack cementation. Phase and microstructure were observed by scanning electro microscope (SEM), X-raydiffraction (XRD) and energy diffusion spectrum (EDS). It was found that phases of NbSi2, CrSi2, Cr2Nb were formed. The result showed thatthe high temperature oxidation resistance of the NbeSi system intermetallics will be improved by applying the Cr doped Si coatings. Maybe itwas attributed by the formation of SiO2 and Cr2O3 which could prevent the penetration of oxygen into the inner coatings and the substrate.� 2006 Elsevier Ltd. All rights reserved.

Keywords: B. Diffusion; B. Oxidation; C. Coatings, intermetallic and otherwise; C. Vapour deposition; D. Microstructure

1. Introduction

NbeSi system intermetallics have great potential as hightemperature structural materials because of their low density,high melting points and high strength at high temperatures.NbeSi system intermetallics must be the better choice thanmost of the alloys. However, the oxidation resistance ofNbeSi system intermetallics is extremely poor at high temper-atures [1e3]. In order to use this kind of intermetallics in air athigh temperatures, they are modified by elements such as Hf,Cr and so on. Even though, the oxidation resistance was not sosatisfied as hoped. Then there must be coatings on them.

To achieve better oxidation resistance, silicon was filteredonto the substrate by molten salt, by pack cementation andby other methods. So there is NbSi2 coating at the surface[4]. NbSi2 is a kind of intermetallics. When it was heated inthe open air at high temperatures, the protective coating SiO2

was formed. SiO2 is a glassy substance and it will stop the ox-ygen coming further into the substrate. And the oxidation

* Corresponding author. Tel.: þ86 10 8231 7117; fax: þ86 10 8233 8200.

E-mail address: xuhb@buaa.edu.cn (H. Xu).

0966-9795/$ - see front matter � 2006 Elsevier Ltd. All rights reserved.

doi:10.1016/j.intermet.2006.10.033

resistance was really enhanced, proved by the oxidation exper-iment. Unfortunately, SiO2 layers were consumed heavilywhen heated in the open air for a long time. The reasons ofwhich may be, first of all, SiO2 layers were volatilized in thevery high temperature region; second, silicon was diffusedinto the substrate when heated for a long time. So chromiumwas doped into the NbSi2 coatings by pack cementation [5].It was suspected that after the deposition of Cr and Si, bothNbSi2 and Cr2Nb will be made. When it was tested in the ox-idation experiment SiO2 and Cr2O3 will be formed. Both ofthem will protect the diffusion of the oxygen into the innercoating and the substrate.

2. Experimental procedures

The alloys included high purity niobium, titanium, silicon,chromium, alumina, and hafnium and were prepared by arcmelting. The specimens with dimension of F 13 mm� 2 mmwere spark-cut from ingots, then ground and cleaned ultrason-ically in acetone. The nominal composition of the alloy is24Tie16Sie6Cre6Ale2Hf.

806 C. Chen et al. / Intermetallics 15 (2007) 805e809

Fig. 1. Cross-sectional SEM image of the niobium alloy with Cr, Si co-deposition coating at 1100 �C heated for (a) 8 h and (b) 16 h under the argon field.

The halide pack cementation was used to deposit the chro-mium onto the substrate and NbSi2 coating. The componentsfor the pack cementation process consisting of Cr or Si powder,halide activator (AlF3 or NH4Cl), and inert filter (Al2O3 pow-der) were weighed, mixed and ground smaller than 75 mm.The process is listed in Table 1. The samples and the mixtureswere all put into the Al2O3 crucibles and heated under the ar-gon shield [6].

After diffusion, the samples were cleaned ultrasonicallyand the packing inert masses were removed gently from the

surface of the samples. The samples were examined by scan-ning electron microscopy (SEM), X-ray diffraction (XRD) andenergy dispersive spectrum (EDS).

3. Results and discussion

3.1. Cr, Si co-deposition

Fig. 1 shows the cross-sectional SEM images of the Cr, Sico-deposition coating formed on NbeSi system intermetallics

807C. Chen et al. / Intermetallics 15 (2007) 805e809

heated for 8 h and 16 h at 1100 �C. And Fig. 2 shows theatomic concentration curve of the corresponding coating.The plot where the component was obtained is shown in theimage. It is seen that the surface of the coatings was smooth.From the image and the concentration profile it was obviousthat in the coating the concentration of Si and Cr was veryhigh, and in the substrate it decreased rapidly. At the sametime, Nb also increased apparently. So it was proved that co-depositing Cr and Si as coating on the NbeSi system interme-tallics were made. Obviously Si and Cr were added into thealloys, and were enriched and concentrated at the surface ofthe alloys. It was also seen in the X-ray diffraction patternin Fig. 3.

The Cr, Si co-deposited coatings formed on NbeSi systemintermetallics was predominantly by the inward diffusion ofCr and Si. This conclusion is made on the basis that the coat-ings were free from pack inclusions, which were known to becontained in the coatings that formed by the outward growthmechanism. It was seen apparently from the cross-sectionalSEM image. The inward growth mechanism of the coatingswas in accordance with the results of Van Loo and Rieck[7,8].

There are two processes when Cr and Si were co-deposited,one was heated for 8 h and the other for 16 h. From the cross-sectional image it was seen that with the increase in insulationtime the thickness of the coating also increased. And in theconcentration profile there is no difference between the con-centration of Si and Cr in the two processes. So it is provedthat the doped atomics Cr can protect the diffusion of siliconinto the substrate when heated for a long time.

3.2. Chromizing followed by siliconing

3.2.1. Both Cr and Si deposited by pack cementationIn this process Cr and Si were deposited separately and by

pack cementation. At first, Cr was deposited for 4 h at1100 �C, after that, Si was deposited on the chromized speci-mens at 1050 �C for 8 h.

Fig. 4 shows the cross-sectional image and can be seen thatthe coating was much thicker than Cr, Si co-deposition pro-cess. When Cr, Si was co-deposited NH4Cl was definitely

Table 1

The mixture ratio of the pack cementation powders

Cr, Si

co-deposition

Cr deposition

by pack

cementation

Si deposition

by molten

salt

Si deposition

by pack

cementation

Si(g) 10 / 10 15

Cr (g) 20 25 / /

NH4Cl (g) 10 10 / 2

NaF (g) / / 13 /

Al2O3 (g) 60 65 83

NaCl (g) / / 30 /

KCl (g) / / 35 /

Na2SiF6 (g) / / 12 /

Temperature (�C) 1100 1100 1000 1050

Heating time (h) Variance 4 8 8

the most suitable for halide activity. But the halide activityof NaF was much effective than NH4Cl when Si was depositedby pack cementation.

Fig. 5 shows the concentration profile of the coating on ni-obium alloy through pack cementation chromizing followedby siliconing. In this process, Cr was deposited at first thenSi was deposited. During deposition of silicon, Cr was not fur-ther diffused into the substrate. One of the reasons of this phe-nomenon was the huge size of the Cr atom. It was shown in theX-ray diffraction pattern in Fig. 6 that in this process, thephase of CrSi2 was formed. The peak of the phase of CrSi2was very sharp and high. The new phase CrSi2 could notprevent from Si diffusing into the substrate. So the coatingbecomes much thicker than the co-deposition process.

It is also seen in the X-ray diffraction pattern that bothCrSi2 and Cr2Nb phases were apparent. Because of thesephases, the coating is considered to have good oxidation re-sistance performance. When they were heated in the openair SiO2 and Cr2O3 will be formed. They will prevent thediffusion of the oxygen into the inner coating and thesubstrate.

This coating was predominantly by the inward diffusion ofCr and Si. It was seen apparently in the cross-sectional SEM

Fig. 2. Concentration profile of the Cr, Si co-deposition coating heated at

1100 �C for (a) 8 h and (b) for 16 h under the argon field.

808 C. Chen et al. / Intermetallics 15 (2007) 805e809

image that the coatings were free from pack inclusions, whichwere known to be contained in the coatings that were formedby the outward growth mechanism. And in the X-ray diffrac-tion pattern no peak about the pack inclusions was found.

Fig. 3. X-ray diffraction pattern of the Cr, Si co-deposition coating heated at

1100 �C for (a) 8 h and (b) heated for 16 h under the argon field.

Fig. 4. Cross-sectional SEM image of the coating on niobium alloy through

pack cementation chromizing followed by siliconing.

Actually in this kind of process, the quality of the coatingwas better than co-deposition process, but the oxidation resis-tance would be tested in the future experiment.

3.2.2. Pack cementation chromizing followedby siliconing by molten salt

The difference between these processes from the last onewas the method of depositing Si onto the chromized speci-mens. The method of Si deposition in the last process wasby pack cementation and in this process it was by moltensalt. The deposition was carried out at 1000 �C for 8 h withoutthe argon field. And the process of Cr deposition was the sameas the last one.

Fig. 7 showed the cross-sectional SEM image and it was re-viewed that the thickness of the coating was close to the onewhen Si coating was prepared by pack cementation. Because

Fig. 5. Concentration profile of the coating on niobium alloy through pack ce-

mentation chromizing followed by siliconing.

Fig. 6. X-ray diffraction pattern of the coating on niobium alloy through pack

cementation chromizing followed by siliconing.

809C. Chen et al. / Intermetallics 15 (2007) 805e809

in these two process chromium was deposited onto the sub-strate by pack cementation for the same hours. It was obvi-ously in the X-ray diffraction pattern that the phase CrSi2was formed.

It was seen obviously from the X-ray diffraction pattern thatphases of NbSi2, Cr2Nb and CrSi2 were apparent. These phaseswere considered to be a good oxidation resistant composition.Because when they were heated in the oxidation experiment,SiO2 and Cr2O3 were formed. They will prevent the diffusionof the oxygen into the inner coating and the substrate.

The growth mechanism of this coating was also predomi-nantly by the inward diffusion of Cr and Si. It was seen appar-ently in the cross-sectional SEM image and concentrationprofile in Fig. 8 that the coatings were free from pack inclu-sions, which were known to be contained in the coatingsthat were formed by the outward growth mechanism. And inthe X-ray diffraction pattern no peak about the pack inclusionswas found (Figs. 3e9).

Fig. 7. Cross-sectional SEM image of the coating on niobium alloy through

pack cementation chromizing followed by siliconing by molten salt.

Fig. 8. Concentration profile of the coating on niobium alloy through pack ce-

mentation chromizing followed by siliconing by molten salt.

4. Conclusions

1. The process of preparing Cr, Si coatings was successful.The coating, which was made by both pack cementationand by molten salt, was dense and thick. In the cross-sectional SEM images it was seen obviously that thegrowth mechanism of the coatings, which were formedon NbeSi system intermetallics with Cr and Si was pre-dominantly by the inward growth mechanism. This con-clusion was made on the basis that the coatings werefree from pack inclusions, which are known to be con-tained in the coatings that formed by an outward growthmechanism.

2. The composition of the coatings was NbSi2, Cr2Nb andCrSi2. It was seen apparently in the X-ray diffraction pat-tern. These phases were considered to have a good oxida-tion resistant composition. When they were heated in theopen air, carrying in the oxidation experiment, SiO2 andCr2O3 were going to be formed, thus they will protectthe diffusion of the oxygen into the inner coating andthe substrate. And then the purpose of oxidation resistancewill be achieved.

References

[1] Bewlay BP, Jackson MR, Lipsitt HA. Metall Trans A 1996;27A:380.

[2] Ma CL, Kasama A, Tan Y, Tanaka H, Tanaka R, Mishima Y, et al. Report

of the 123rd committee on heat-resisting materials and alloys, vol. 40,

No. 3. Japan Society for the Promotion of Science; 1999. p. 349e60.

[3] Yanmabe-Mitarai Y, Ro Y, Nakazawa S, Harada H. J Jpn Inst Met

2000;64:1068e75.

[4] Suzuki O Ryosuke, Ishikawa Masayori, Ono Katsutoshi. J Alloys Compd

2002;336:280e5.

[5] Fukumoto Michihisa, Matsummura Yuzi, Hayashi Shigenari,

Narita Toshio. Mater Trans 2003;44(4):731e5.

[6] Zhou Chungen, Xu Huibin, Gong Shengkai, Yang Ying, Kim Kyoo Young.

Surf Coat Technol 2000;132:117e23.

[7] Van Loo FJJ, Rieck GD. Acta Metall 1973;21:61.

[8] Van Loo FJJ, Rieck GD. Acta Metall 1973;21:73.

Fig. 9. X-ray diffraction pattern of the coating on niobium alloy through pack

cementation chromizing followed by siliconing by molten salt.