High-Temperature Interaction of Liquid Gd with Y2O3
- 463 Downloads
The sessile drop method combined with contact heating procedure was applied for the investigation of high-temperature interaction between liquid Gd and Y2O3 substrate. Real-time behavior of Gd sample in flowing inert gas (Ar) atmosphere upon heating to and at temperature of 1362 °C was recorded using high-speed high-resolution CCD camera. The results evidenced that molten Gd wets Y2O3 substrate (the contact angle θ < 90°) immediately after melting of metal sample observed at T = 1324 °C (Tm = 1312 °C). During the first 3 min of the sessile drop test, the contact angle dropped from θ = 52° to θ = 24° and then stabilized at the final value of θ f * = 33°. The solidified Gd/Y2O3 couple was subjected to structural characterization using optical microscopy, scanning electron microscopy coupled with x-ray energy-dispersive spectroscopy. The results evidenced that the wettability in the Gd/Y2O3 system has a reactive nature and the leading mechanism of the interaction between liquid Gd and Y2O3 is the dissolution of the ceramic in the liquid metal responsible for the formation of a deep crater in the substrate under the drop. Therefore, the final contact angle θf*, estimated from the side-view drop image, should be considered as an apparent value, compared to the more reliable value of θf = 70° measured on the cross section of the solidified couple.
KeywordsGd interfaces reactivity sessile drop wettability Y2O3
The development of metallic materials technology, especially alloys with highly reactive components, is inextricably linked to the search for new, refractory ceramic materials for crucible-assisted melting and casting processes. Such ceramics must be characterized by a set of features, e.g., good mechanical strength, high erosion and corrosion resistance to molten metals, and low levels of impurities, which could significantly affect the metal properties upon entering into the melt. The same requirements apply for materials of containers (e.g., crucibles, holders, and capillaries) used for measurements of thermophysical properties of liquid and semisolid alloys; those reliable values are particularly important in modeling and computer simulation of liquid-assisted processes. Examples include widely used calorimetric, dilatometric, and thermogravimetric tests as well as measurements of electrical and thermal conductivity, surface tension, and density of liquid metals and alloys. The characteristic feature of container-assisted methods is a relatively long time of contact of the melt with the container material.
In the available literature, there is no information on a chemically stable material that could be used for high-temperature contact with gadolinium, although quite different refractories have been reported to be used in the works where Gd-rich alloys or Gd-containing compounds were synthesized or examined at high temperatures using container-assisted methods (Ref 1-9).
Waring and Schneider deposited the Gd2O3-TiO2 compounds by heating them in air at 1000 °C and placed samples on platinum plates (Ref 1), while Klimm et al. used platinum crucibles for DTA studies of LiREF4 materials (RE = Eu, Gd, Tb, Dy, Ho) at temperatures up to 1300 °C (Ref 2). Literature has also documented the use of crucibles made of aluminum oxide (Al2O3) for DTA studies of gadolinium-containing materials. They were used by both Wang et al. for the ZrO2-REO1.5 system (RE = La, Nd, Sm, Gd, Dy, Yb) at temperatures up to 1700 °C (Ref 3) and S. Schmitz et al. for the Gd-Ti and Gd-Ti-Al-Cu alloys at 1300 °C (Ref 4). None of these works, however, clearly indicates the ceramic materials suitable for the melting or testing of molten gadolinium-rich alloys.
However, the preliminary study of interaction of liquid Gd with Y2O3, ZrO2-3 wt.%Y2O3 (3YSZ), and TiO2 substrates reported by Sobczak et al. (Ref 6) showed that unexpectedly among these oxides, Y2O3 exhibits the highest reactivity in contact with liquid Gd. Moreover, the study (Ref 6) confirmed the first findings on yttria-based ceramics Y2O3, ZrO2-3 wt.%Y2O3 (3YSZ), and ZrO2-5 wt.%Y2O3 (5YSZ) reported by Kaban et al. (Ref 7) and showing that the reactivity of yttria-stabilized zirconia with Gd-based alloys containing Ti increases with the increase in the amount of Y2O3. Therefore, both Y2O3 and 5YSZ were excluded as refractories suitable for measurements of the thermophysical properties of Gd-Ti alloys by substrate- or container-assisted technique in work (Ref 7).
Recent study by Turalska et al. (Ref 8) on the high-temperature behavior and the reactivity of Gd/3YSZ couple in flowing Ar showed that liquid gadolinium wets 3YSZ with the final contact angle of 67° at 1362 and 1412 °C. Good wettability was explained by the formation of a continuous layer of the wettable reaction product Gd2Zr2O7, growing at the drop/substrate interface as well as on the substrate surface beyond the drop. The same authors examined the wettability of dense TiO2 substrates with liquid Gd (Ref 9) using the identical testing conditions as those applied in the research described in Ref 8). The results of wetting tests of the Gd/TiO2 couple show that liquid Gd does not wet TiO2 substrate and that it forms high contact angles of θ ~ 100° at both 1362 °C and 1412 °C. It was also observed that in the Gd/3YSZ as well as in the Gd/TiO2 couples, the interaction of Gd with the ceramic leads to the formation of an interfacial layer and is accompanied by the dissolution of the substrate in the liquid drop. Important difference is in the structure of reactively formed interfacial layer. In the Gd/3YSZ couple, it represents one single-phase Gd2Zr2O7 while in the Gd/TiO2 couple, it is composed of two sublayers with different structure and chemical composition: Gd2TiO5 in the drop side sublayer and Gd2Ti2O7 in the substrate-side sublayer.
The materials used were pure Gd of 99.99% purity (Sigma-Aldrich) and polycrystalline Y2O3 substrate made by high-pressure, high-temperature synthesis. The Y2O3 substrate was polished to a roughness of Ra ≈ 120 nm and then ultrasonically cleaned in isopropanol (C3H8O alcohol) for 5 min. Directly before the high-temperature test, the substrate was preheated in air at 1700 °C for 2 h.
The sessile drop method was applied for investigation of high-temperature wetting behavior and reactivity between liquid Gd and Y2O3 substrate using an experimental set up described in details in Ref 11. During the high-temperature test, the Gd/Y2O3 couple was contact-heated to the test temperature of Texp = 1362 °C with a heating rate of 12 °C/min under flowing inert gas (Ar, 99.9992%) under a pressure of 850-900 hPa. After isothermal heating at Texp for 5 min, the couple was cooled down to the room temperature at a rate of 24 °C/min.
The behavior of the Gd sample on ceramic substrate was registered using an MC1310 high-speed, high-resolution digital camera. In the first stage of the experiment, including heating until sample melting and isothermal heating of the couple at 1362 °C, the couple behavior was recorded at a rate of 10 frames per second (fps). The next stage, which involved cooling to the solidification temperature, was recorded at a rate of 1 fps. The collected images were used for estimation of the contact angle values θ (θl—left-side angle; θr—right-side angle and θ—average value of θl and θr) with a help of the ASTRA2 software developed by ICMATE-CNR, Genoa, Italy (Ref 12, 13) as well as for making a real-time movie of the high-temperature test (SUPPLEMENT #1).
Structural and chemistry characterization of solidified couple was made both on its top surface and on cross section using Keyence VHX-700F optical microscope (OM) and Hitachi TM3000 scanning electron microscope (SEM) equipped with energy-dispersive x-ray spectroscopy (EDS) analyzer.
Contact angle values calculated and measured for the Gd/Y2O3 system
Contact angle θ, °
θ 0 *
θ f *
Isothermal heating of the system at the test temperature Texp = 1362 °C for 5 min (Fig. 1) does not affect the contact angle and the persistent averaged value of 33° suggests that the Gd/Y2O3 system has achieved its thermodynamic equilibrium before reaching the test temperature. This fact combined with the low value of the final contact angle recorded in this study θf* = 33° is typical for systems with high reactivity.
However, the value of Gibbs energy for the above reaction (ΔGr), calculated in this study with the HSC software and the database contained therein (Ref 19) for a temperature of 1362 °C, showed that ΔGr is positive and amounts to 14 kcal/mole, i.e., the formation of the Gd2O3 phase according to the redox reaction (1) is rather unlikely.
Considering the fact that the Gd2O3-Y2O3 quasibinary phase diagram is well established (e.g., Ref 17) and there are no any reports on ternary oxides in the Gd2O3-Y2O3 system, the results of structural examinations are of a great importance for the explanation of the mechanism of high-temperature interaction taking place in the Gd/Y2O3 system.
Both optical microscopy (Fig. 6) and scanning electron microscopy (Fig. 7 and 8) showed the occurrence of a deep crater in the Y2O3 substrate under the solidified drop and the presence of new phases in initially pure Gd sample. Taking into account that the drop matrix became rich in oxygen and yttrium, particularly near the substrate, and all precipitates in the drop also contain high amounts of yttrium and oxygen, we may conclude that the formation of a deep crater in the substrate is associated with a strong dissolution of yttria in molten Gd while the new phases were precipitated from Y- and O-saturated melt Gd(Y,O) during its solidification. Structure examination also shows that the heterogeneous nucleation of new phases takes place both at the drop/substrate interface and on the surface of the drop. Since the needle-like precipitates in the 2PL area are rich in Y and they contain Gd and O we suggest that they present Y2O3-based phase, designated as (Y,Gd)2O3 and formed after wetting tests during cooling of the Gd(Y,O) melt saturated with Y and O at high temperature. Therefore, the phenomenon responsible for high-temperature interaction and phase transformations in the Gd/Y2O3 couple is related to dissolution–reprecipitation mechanism but not with direct interfacial reactions. Moreover, we suggest that the Gd-rich discontinuous oxide layer containing Y, well distinguished at the drop/substrate interface (Fig. 8) as well as individual precipitates at the drop surface of similar chemical composition, correspond to the (Gd,Y)2O3 phase. Although this phase is based on Gd2O3, it was not reactively formed and represents the residual fragments of primary (native) gadolinium oxide film, in which the Y2O3 substrate was dissolved during the high-temperature test.
From the analysis of available binary phase diagrams, it has to be emphasized that the Gd/Y2O3 system is unique since both Gd and Y as well as their oxides Gd2O3 and Y2O3 show unlimited solubility in both liquid and solid states. That is why high-temperature interaction in the Gd/Y2O3 sessile drop couple is dominated by a strong dissolution of Y2O3 both in the liquid metal and in primary gadolinium oxide covering the Gd sample. Subsequently, it results in the formation of a deep crater in the substrate and new phases that correspond to solid solutions, i.e., Gd(Y), (Y,Gd)2O3 and (Gd,Y)2O3. It is worth to mention that at this stage of the research, it is impossible to distinguish a clear difference between two solid solutions, (Gd,Y)2O3 and (Y,Gd)2O3, due to unlimited solubility between Y2O3 and Gd2O3.
A strong dissolution of the substrate in liquid metal is the main reason why the final contact angle measured in this study θf* = 33° should be considered as apparent value because the real position of the drop/substrate interface is hidden due to the formation a deep crater in the substrate. Moreover, the position of the triple line is additionally affected by a strong liquid metal penetration inside the substrate.
Upon cooling, oxygen solubility in the liquid Gd decreases rapidly. Therefore, oxygen present in the liquid is rejected either by heterogeneous nucleation of the (Gd,Y)2O3 crystals both at the drop surface and at the drop/substrate interface or by their homogeneous nucleation at the residual fragments of primary gadolinium oxide film with dissolved yttrium (Fig. 10d). Consequently, the area around the (Gd,Y)2O3 precipitates becomes depleted in oxygen, while the area in the drop near the substrate, especially in its crater, becomes oversaturated in yttrium. Thus, in the next step of cooling, it makes possible the heterogeneous nucleation and growth of the (Y,Gd)2O3 phase in the form of needles surrounded by almost pure Gd (Fig. 10d).
The phenomenon of significant substrate dissolution in liquid metal, often observed in metal/metal systems, has also been reported in some metal/non-oxide ceramic systems, e.g., in Au-Ni/ZrB2 (Ref 20), Ni/SiC (Ref 21), and Ni/TiB2 (Ref 22). The results of this study demonstrate that the dissolutive wetting can be a leading mechanism of interaction of liquid metals not only with metallic materials or metallic-like compounds such as metal borides, carbides, and nitrides, but also with oxides.
Experimental study of the high-temperature interaction between liquid gadolinium and yttrium oxide substrate performed under an inert atmosphere upon heating to and at a temperature 1362 °C shows a good wettability in the Gd/Y2O3 system with a final value of the contact angle θ f * = 33°. However, this value should be considered as apparent one because the real position of the drop/substrate interface is hidden due to the formation of a deep crater in the substrate, while the position of the triple line is affected by a strong liquid metal penetration inside the substrate. The contact angle measured at the triple point on the cross section of solidified sessile drop couple was θf= 70°, and it is considered as to the more reliable value that reflects the wetting properties of the Gd/Y2O3 couple at 1355-1362 °C. Whatever the real contact angle is, the results of this study leave no doubt about a good wetting in the Gd/Y2O3 system, in which dissolutive wetting mechanism, typical for metal/metal systems, is responsible for significant structural changes in the Gd/Y2O3 couple during the high-temperature interaction. This is related to the specific situation, where the unlimited solubility occurs both between pure Gd and Y as well as between their oxides. Therefore, the use of yttria either as a bulk material or a protective coating in contact with metallic melts rich in Gd, particularly for measurements their thermophysical properties by container- or substrate-assisted methods, should be done with care and efforts to decrease the contact area and the contact time.
The above findings, formulated from experimental work, show that when information on the phase equilibria in the metal/refractory systems is missing, for the selection of a proper refractory resistant to aggressive attract of a molten metal, the analysis of the Ellingham diagram of the stability of compounds is insufficient and it must be assisted with detailed experimental studies of high-temperature interaction between the metal and the selected refractory.
This study was done in the frame of collaboration program between the German Academic Exchange Service DAAD and the Ministry of Science and Higher Education of Poland. Financial supports from the National Science Centre of Poland (program HARMONIA, Project No. 193624) and the German Academic Exchange Service DAAD (Project No. DAAD-56269397) are acknowledged. The authors acknowledge M.Sc. Janina Radzikowska and Dr. Rafał Nowak for technical assistance in structural characterization.
Supplementary material 1 (AVI 1452 kb)
- 2.D. Klimm, I.A. dos Santos, I.M. Ranieri, and S.L. Baldochi, Phase equilibria and crystal growth for LiREF4 scheelite crystals, in Materials Research Society symposia proceedings, 2009, 1111–D01-07Google Scholar
- 5.P. Turalska, N. Sobczak, A. Polkowska, G. Bruzda, A. Kudyba, I. Kaban, N. Mattern, and J. Eckert, Wettability and Reactivity of Liquid Gd in Contact with Al2O3 Ceramics, Surf. Eng., 2017, 22(4), p 41–48Google Scholar
- 6.N. Sobczak, P. Turalska, M. Homa, A. Siewiorek, G. Bruzda, R. Nowak, A. Kudyba, I. Kaban, N. Mattern, and J. Eckert, High temperature interaction between liquid Gd-containing alloys and selected oxides, in Proceedings of the 72nd World Foundry Congress, Nagoya, 2016Google Scholar
- 9.P. Turalska, M. Homa, R. Nowak, G. Bruzda, N. Sobczak, I. Kaban, N. Mattern, and J. Eckert, Wettability, Reactivity and Interfaces in the Gd/TiO2 System, Trans. Foundry Res. Inst., 2017, LVII(4), p 303–308Google Scholar
- 10.T.B. Reed, Free Energy Formation of Binary Compounds, MIT Press, Cambridge, 1971Google Scholar
- 13.ASTRA Reference Book, Oct. 2007, IENRI, ReportGoogle Scholar
- 14.http://periodictable.com/Elements/064/data.html, Accessed 19 Dec 2018
- 17.MTDATA—Phase Diagram Software from the National Physical Laboratory Calculated Gd2O3-Y2O3 phase diagram. http://resource.npl.co.uk/mtdata/gdyo.htm. Accessed 22 Feb 2018
- 18.J.W. McMurray, Thermodynamic Modeling of Uranium and Oxygen Containing Ternary Systems with Gadolinium, Lanthanum, and Thorium, PhD Thesis, University of Tennessee. http://trace.tennessee.edu/utk_graddiss/3152/, 2014. Accessed 18 May 2017
- 19.HSC Chemistry® 6.0—Software Copyright© by Outotec Research Oy, Finland, ISBN 13:978-952-9507-12-2Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.