Does Good Wetting Sufficient for Free Infiltration of Porous Performs?

N. Frage(a)*, S. Barzilai(a,b), H. Nagar(a) and N. Froumin(a),

a Department of Materials Engineering, Ben-Gurion University, P.O.Box 653, Beer-Sheva 84105, Israel

b Department of Materials, NRC-Negev, P.O.Box 9001, Beer-Sheva 84190, Israel

ABSTRACT

Correlation between wetting behavior and infiltration of porous performs was investigated. Wetting and infiltrating experiments were conducted for In-1at%Ti melt on Er2O3, TiB2, Al2O3 and CaF2 substrates. Good wetting and spontaneous infiltration were observed for Er2O3, TiB2 substrates, while no infiltration, in spite of good wetting, was observed for Al2O3 and CaF2 substrates. These results were attributed to the presence of the gaseous phase, which appears as a result of the substrate evaporation (CaF2 substrate) or a chemical interaction between the substrate and Ti dissolved in the melt (Al2O3 substrate) It was suggested and experimentally confirmed that under condition of constrained gas evacuation from the porous media a thin solid layer is formed at the melt surface and prevents liquid penetration into the porous performs.

Keywords: wetting, infiltration, volatile compound, thermodynamic calculations, Al2O3, fluoride, CaF2.

* Corresponding author. Tel:+97286461468; fax:+97286479441; e-mail

1. INTRODUCTION

Pressure less infiltration is a promising technique for fabrication a metal-ceramic composites. Good wetting between a liquid metal and a ceramic phase is necessary for successful spontaneous penetration of a melt into a porous ceramic perform [1-5]. Nevertheless, for some systems, which display sufficient wetting, the free infiltration does not take place. The sufficient wetting in these systems is achieved by addition of active elements, which reacts with the substrate and a new interfacial layer is formed. In this case, the reaction products may fill open pores of a preform and mechanically prevent metal penetration [6]. Altering of the melt composition due to depletion of the active element, during penetration may also be a reason for the contact angle increasing and stopping spontaneous infiltration.

Recently, we have observed that in the CaF2/(In-1%atTi) system, where a sufficient wetting (contact angle of about 15deg) is achieved as a result of a preferential Ti adsorption at the interface and no formation of a new condensed phase takes place, spontaneous infiltration does not occur. The specific feature of this system is related to the high vapor pressure of CaF2 at temperatures higher then 1273 K. The question is: how does the gaseous phase (substrate vapor or gas releases during metal-ceramic interaction) affect the spontaneous infiltration? This question is discussed in the present study.

2. EXPERIMENTAL

Four types of substrates, which display different nature of interaction with In-Ti alloy, were chosen for the analysis. Two oxide substrates: Al2O3 (the interaction leads to the formation of gaseous aluminum sub-oxide Al2O) and Er2O3 (only condensed phases are expected as a result of the interaction). Two non-oxide substrates: TiB2, (no gaseous phase is expected as a result of the interaction) and CaF2 (substrate with high vapor pressure). Dense ceramic substrates for the sessile drop experiments and porous preforms (~30% of open porosity) for infiltration were prepared by powder metallurgy technique. The samples for wetting experiments were polished down to the 1 mm diamond paste level and ultrasonically cleaned in acetone and ethanol. In-1%atTi alloy was prepared in-situ during heating. The wetting experiments were performed under a dynamic vacuum (4·10-3 Pa). The contact angles were measured directly from the magnified profile images of the drops. The composition of the metal/ceramic interfaces was studied using SEM (Jeol GSM 5600), equipped with an EDS analyzer. The interfacial region beneath the In-Ti drop was analyzed by X-ray Photoelectron Spectroscopy (XPS), accompanied with Ar-sputtering depth profile using the ESCALAB 250 system. Further experimental details were given in previous studies [7-9].

3. RESULTS

3.1 Wetting Experiments

The spreading kinetics of In and In-1at%Ti drops on the CaF2, Er2O3, Al2O3 and TiB2 substrates are presented in Fig. 1. The experimental results indicate the pure In does not wet the substrates (θ≈120-135°) and the contact angle remains almost constant during 30 min of contact (Fig. 1a). The addition of 1at% Ti to the melt improved wetting and the contact angles decrease toward ~5-30° (Fig. 1b). No new phase formation at the interface in the CaF2/In-Ti and TiB2/In-Ti systems was detected. Extremely thin Ti-rich interfacial layers were detected in the Al2O3/In-Ti and Er2O3/In-Ti systems.

Fig. 1: Contact angle for In (a) and for (In-1at%Ti) (b) on Al2O3, Er2O3, TiB2 and CaF2 substrate at 1173K.

3.2 Infiltration

SEM micrographs (Fig. 2) indicate that In-1at%Ti alloy spontaneously penetrates into the porous Er2O3 and TiB2 preforms, while no infiltration into the porous Al2O3 and CaF2 preforms takes place, even though, the drops remained on the porous bodies display relatively low contact angle.

Fig. 2: SEM micrographs of the In-1at%Ti atop different porous performs, after 30 min of contact at 1173K.

4. DISCUSSION

Low contact angle and absence of a thick layer of the reaction product, which may fill the open pores, have to provide the conditions for spontaneous infiltration [2,3]. However, no spontaneous infiltration takes place in the CaF2/(In-1at%Ti) and Al2O3/(In-1at%Ti) systems. These experimental observations can not be explained by the mechanisms mentioned above. We suggest that a reason of this feature is related to the presence of a gaseous phase in the systems, which appears as a result of the substrate evaporation (CaF2 substrate) or a chemical interaction at the metal/ceramic interface (Al2O3 substrate). This gaseous phase may be easily evacuated form the flat interface of the substrate exposed to vacuum, while in a porous media, where evacuation of the gas is constrained, its condensation at the surface of the melt may take place and a condensed film, even extremely thin, may prevent liquid penetration.

The equilibrium vapor pressure of CaF2 for a flat solid/gas interface at 1173K is equal to 10-10atm [10]. It has to be pointed out that the vapor pressure may be higher within capillaries due to their surface curvature and surface curvature of CaF2 particles in the sintered porous preform [11]. The massive evaporation of the CaF2 substrate was observed experimentally in [9]. In order to confirm that the condensation of CaF2 on the melt surface occurs under condition of a constrained gas evacuation additional experiments were performed. In-1at%Ti alloy was placed onto fully dense CaF2 substrate and covered by small CaF2 crucible, which suppresses CaF2 gas evacuation. It was established that after 30 min of contact at 1173K the drop did not spread and the contact angle was 120° (Fig. 3a). The same drop was heated again under vacuum without cover and its spreading took place (Fig.3b). An XPS depth profile analysis was preformed for the surface of the drops. Relatively thick CaF2 (more than 100nm) layer was detected on the surface of the covered drop (Fig. 4a), while no presence of F and Ca was detected on the drop exposed to vacuum (Fig. 4b).

Fig. 3: The In-1at%Ti after wetting experiment on CaF2 substrate at 1173K for 30 min. (a) drop was covered by CaF2 crucible (b) drop was not covered.

Fig. 4: XPS depth profile for the surface of the covered drop (a) and the drop exposed to vacuum (b).

In the Al2O3/(In-1at%Ti) system the presence of the gaseous phase is attributed to the chemical reaction at the triple line between Al2O3 and Ti dissolved in the melt. This interaction provides the formation of Al2O according to Eq. 1.

(1)

The equilibrium partial pressure of Al2O for a given titanium activity was calculated as 10-11atm at 1173K [14]. For the substrate exposed to vacuum, Al2O may be easily evacuated from the area in the vicinity of the triple line. In this case, reaction 1 continuously occurs and enables the drop spreading. For the constrained gas evacuation the Al2O oxidation by oxygen presented in the experimental chamber (reaction 2) have to be taken into account.

(2)

Using thermodynamic data [10] the values of the equilibrium constant K and the partial pressures product were calculated. At 1173K the value of is equal to 10-45atm2 [10]. Thus, even extremely small amount of oxygen in the experimental chamber (10-34atm of O2) is sufficient for thin Al2O3 layer formation. This layer prevents the melt penetration into the porous Al2O3.

The experiment with covered drop was also conducted for the Al2O3/(In-1at%Ti) system. After 30 min of contact at 1173K, the covered drop again did not spread and the contact angle was ~110° (Fig. 5a). According to the Auger analysis (Fig. 6) small amount of aluminum is detected at the surface of the covered drop. The same drop after heating under vacuum without cover displays low contact angle (Fig. 5b). No Al content was detected on the surface of the drop exposed to vacuum.

Fig. 5: The In-1at%Ti drops after wetting experiment on Al2O3 substrate at 1173K for 30 min. (a) the drop was covered by Al2O3 crucible, (b) the drop was not covered.

Fig. 6: Auger spectra of the covered drop surface.

The same wetting experiments were also conducted for the Er2O3 and TiB2 substrates. For both systems the contact angles of the covered and uncovered drops were similar.

5. CONCLUSIONS

Wetting and infiltrating experiments were conducted for In-1at%Ti melt on Er2O3, TiB2, Al2O3 and CaF2 substrates. Good wetting and spontaneous infiltrating were observed for Er2O3, TiB2 substrates, while no infiltration, in spite of good wetting, was observed for Al2O3 and CaF2 substrates. The results of the experiments with covered drop allow to conclude that the reason of this feature is related to the presence of the gaseous phase in the systems, which appears as a result of the substrate evaporation (CaF2 substrate) or chemical interaction at the metal/ceramic interface (Al2O3 substrate). This gaseous phase is easily evacuated form the interface of the substrate exposed to vacuum and low contact angle is achieved, while in a porous media, where evacuation of the gas is constrained, the formation of thin solid films on the melt surface takes place, which prevent liquid penetration into porous performs.

ACKNOWLEDGEMENTS

The authors wish to thank Mr. Etaay Meydany and to Mr. Amir Hagag for their expert technical assistance. This work was supported by the grant N0138-05 from the Israeli Council of High Education and the Israeli Atomic Energy Commission.

REFERENCES

  1. Metals Handbook, vol. 7, "Powder Metal Technologies and Applications, Section: Shaping and Consolidation Technologies", ASM International.
  2. F. Delannay, L. Froyen, A Deruyttere , J. of Mater. Sci. 22 (1986) 1.
  3. N. Eustathopoulos, M.G. Nicholas and B. Drevet, Wettability at High Temperatures, Pergamon Materials Series, 1999.
  4. K.P. Trumble, Acta Mater. 46 (1998) 2363.
  5. G. Kaptay, T. Barczy, J. Mater. Sci. 40 (2005) 2531.
  6. S. Tariolle, F. Thévenot, M. Aizenstein, M.P. Dariel, N. Frumin, N. Frage, Journal of Solid State Chemistry 177 (2004) 400.
  7. S. Barzilai, M. Aizenshtein, M. Lomberg, N. Froumin, N. Frage, Solid State Sci. 9 (2007) 338.

8.  S. Barzilai, M. Aizenshtein, M. Lomberg, N. Froumin, N. Frage, J. Alloys. Compd. 452 (2008) 154.

  1. . M. Aizenshtein, S. Barzilai, N. Froumin, N. Frage, J. Mat. Sci. 43 (2008) 1259.

10.  Thermodynamic Data-Base SSUB3, version 3.1 (2001). Produced by Scientific Group Thermo-data Europe

  1. Y.M. Chiang, D. Birnie, W. D. Kingery, Physical ceramics-principles for ceramic science and engineering-Ch. 5, 1ed., John Wiley and sons Inc., (1997).

1-171