2000 Simcoe Street North, Oshawa, Ontario, Canada, L1H 7K4. Fuku-shi, Fukui, , Japan

Transcription

1 Applied Mechanics and Materials Online: ISSN: , Vol. 42, pp doi:1.428/ 213 Trans Tech Publications, Switzerland Determination of Specific Heat of EutecticIndium Bismuth- Tin Liquid Metal Alloys as a Test Material for Liquid Metal - Cooled Applications Adam Lipchitz 1,a, Glenn Harvel 1,b, Takeyoshi Sunagawa 2,c 1 Faculty of Energy Systems and Nuclear Sciences, University of Ontario Institute of Technology, 2 Simcoe Street North, Oshawa, Ontario, Canada, L1H 7K4 2 Department of Applied Nuclear Technology, Fukui University of Technology, Gakuen, Fuku-shi, Fukui, , Japan a adam.lipchitz@uoit.ca, b glenn.harvel@uoit.ca, c sunagawa@fukui-ut.ac.jp Keywords:Liquid Metals, Liquid Metal Cooled Nuclear Reactors, Specific Heat, Eutectic Alloys Abstract. Currently, Russia, India, China, France, South Korea, and Japan are actively pursuing liquid metal cooled applications such as liquid cooled metal nuclear reactor concepts. The liquid metal coolants being considered for these designs are sodium, lead and lead-bismuth eutectic; these designs utilize reactive and toxic materials at temperatures up to 173 K for nuclear power plant operations and other similar applications. To simulate these systems with the actual coolant material requires a high level of safety systems. Use of these materials in university experimental laboratory settings is difficult due to the safety hazards and that lead (Pb) is a designated substance requiring special permission to use. Therefore, a less toxic and less reactive liquid metal that can be used to simulate liquid metal cooled flows will allow for a greater number of investigations and experimentation of liquid metal flow with regards to the field of thermal hydraulics. Good candidates for a liquid metal experimental fluid are alloys from the indium-bismuth-tin system such as Field s metal, which by weight percent is 51% indium, 32.5% bismuth and 16.5% tin and possesses a melting temperature of 333 K. However, the thermodynamic properties of Field s metal and similar alloys in their liquid state are not well described in literature. This work experimentally measures the specific heat of the eutectic alloys of theindium-bismuth-tin tertiary system using a differential scanning calorimeter technique and analyzes the results to determine if the thermodynamic properties of the system have sufficient scaling for experimental modeling applications. The results verify the melting temperatures of the alloys and establish a relationship between temperature and specific heat. Introduction Liquid metals have been used to cool advanced power generation systems such as nuclear power plants and submarines since the 197s[1,2]. Typically, the liquid metals are either a pure elemental metal, such as lithium, lead or sodium or a eutectic alloy, for examplelead-bismuth or sodium-potassium[3,4,5]. Currently, the international Generation IV Forum has selected liquid metal cooled reactors as one of six concepts to explore for future nuclear power generation designs and technologies as these systems can achieve thermal efficiencies on the order of 4%. The forum has selected three liquid metals as the possible coolant for this type of reactor; sodium, lead and lead-bismuth[1]. Therefore, it is essential that the thermal-hydraulics of liquid metals and especially liquid metal eutectic alloys be investigated and understood using modern techniques and methods. However, in Canada; sodium, due to its chemical reactivity and lead (Pb) due to its toxicity, are difficult materials to study inuniversity laboratory settings without special considerations to health and safety[6,7].despite this problem, it is still necessary to perform experiments at the University level for education, training, and research on liquid metals to improve the overall understanding of their use in a nuclear reactor environment. Therefore, if a liquid metal could be developed as an experimental model then experimentation in this field could be increased significantly. For water cooled reactors, Freons were commonly used as a test fluid to simulate the fluid behaviour. Thus, the same methods can be applied here to liquid All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans Tech Publications, (ID: , Pennsylvania State University, University Park, USA-6/3/16,18:38:5)

2 186 Recent Trends in Materials and Mechanical Engineering II metals. In order to achieve this objective, the thermo-physical properties of a liquid metal experimental fluid need to be characterized[8,9,1]. Unfortunately, the fluid properties of many eutectic alloys have not been well documented in the liquid region in the public literature[11,12,13,14]. This workutilizes a method to measure the specific heat of eutectic liquid metal alloys in the indium-bismuth-tin system and compare the values to that of a lead-bismuth eutectic. The compositions of the various alloys used in this study are described in Table 1 along with their critical (melting) temperatures. Table 1: Composition by weight percent of eutectic liquid metal alloys [11,13] Indium-Bismuth-Tin Eutectic Alloys Alloy Name (Weight Percent) Field s Metal (51 In /32.5 Bi/ 16.5 Sn) Local Eutectic Reactions for In-Bi-Sn System Indalloy 27 (54 Bi /29.7 In/ 16.3 Sn) Indalloy 174 (57 Bi/26 In/17 Sn) Other Eutectic Alloys Bismuth-Tin (52 Bi/48 Sn) Lead Bismuth (45.5 Pb/55.5 Bi) Critical Temperature/ Reaction 333 [K] Eutectic 354 [K] 352 [K] 415 [K] Eutectic 397 [K] Eutectic The study investigates the eutectic alloys of the In-Bi-Snternary system;where there are three different eutectic alloys. The primary eutectic alloy Field s metal and two local eutectic alloys whose compositions are presented in Table 1. Field s metal is considered to be the primary eutectic reaction occurring in the ternary system as it has the lowest critical temperature 333 K. Therefore, Field s metalalso has the highest potential for the easiest use in a laboratory setting of the three eutectic alloys. Field s metal also has several other advantageous properties; it is non-reactive with air and water, responds to a magnetic field and is easily fabricated[8]. However, little is publicly available about its thermo-physical properties in a liquid state or the other In-Bi-Sn alloys[12,15] even though its material thermodynamic properties were recently determined and optimized[16]. This study examines the specific heat of all three In-Bi-Sn alloys. The property of specific heat (C p ) is chosen as it is a fundamental parameter to thermal hydraulics and fluid dynamics research.specific heat is dependent on the material, its state of matter and the temperature. Specific heat in this work is measured using a differential scanning calorimeter, which has been demonstrated to effectively measure the specific heat of liquid metals[17,18,19]. Experimental Design The In-Bi-Sn alloys tested and investigated are the same as those presented in Tables 1 and 2; Field s metal, Indalloy-27, and Indalloy 174. The alloys were fabricated at University of Ontario Institute of Technology. The alloys are created in a carbon-rich environment and after formation are filtered for oxides based on a method developed by Sunagawa et al. [2]. A reference sample of Field s metal was acquired from Fukui University of Technology created by Dr. Sungawais also tested and compared to the UOIT fabricated sample of Field s metal.the metal alloys specific heat is measured using a TA instruments Q2 v.23.4 differential scanning calorimeter (DSC). DSCs work by heating two materials, the sample and the reference, at a constant specified rate. The calorimeter records the time-dependent temperature of each material using a thermocouple. As the temperatures of two materials heat capacity are different the rate at which they heat and the corresponding temperatures will also be different. This difference is innate to materials themselves and can be used to calculate the specific heat. Figure 1 is a schematic of the type differential scanning calorimeter employed.

3 Applied Mechanics and Materials Vol Figure 1: Schematic of a TA Q2 series differential scanning calorimeter [19] There are several methods to calculate specific heat from a DSC. The method used in this work is called the cell constant method[19]. The cell constant method uses the following equation: C =. (1) where, E is the cell constant, H is the heat flow, m is the mass and H r is the heating rate. The systems cell constant is determined using a built-in function of the DSC software. A pure sample of Indium is required to run until it reaches its melting point as this is a known reference sample.next the background heat rate is determined by placing two empty pans into the DSC. The heat flow is calculated by subtracting the background from the recorded data. The TA instruments Q2 v.23.4 differential scanning calorimeter was used with nitrogen purge gas and platinum reference pans. The samples are placed inside a reference pan and sealed.each sample is heated at a rate of 5 C/min up to 15 C except in the case of Sn-Bi, where it was heated to 18 C due to its higher melting temperature. Results and Discussions The melting temperatures and specific heat of all three alloys were calculated and are presented in this section. Figure 2 is a description of the heat flow curves presented in Figures 3-5. The heat flow changes as energy is added or removed from the system. For metals a significant change in the positive direction in heat flow is from the solidification or endothermic reaction of the liquid. A significant change in the negative direction is an exothermic reaction. In the following figures the exothermic reaction is the melting of the metal alloy. For the specific heat measurement the reaction of interested is heating section of the curve where temperature is rising. Figure 2: Description of heat flow vs. temperature curve for a differential scanning calorimeter with exothermic and endothermic reactions

4 188 Recent Trends in Materials and Mechanical Engineering II The melting temperature occurs at the point where the tangent of the inflection point intersects the base of the curve. In order to determine the melting temperature the point can be determined from an examination of heat flow curve. A tie line is drawn from the beginning of the reaction (across the valley) to the end of the reaction. From here a line is drawn from the inflection point until it intersects with the tie line. The point where the lines intersect is the eutectic melting temperature. The method is demonstrated in Figure 3 with Field s metal. Field s metal has a double peak shape in its eutectic reaction. Therefore, two points of interest are determined. The first point is the intersection of the tie line and inflection point, therefore the eutectic melting point. The alloy then goes through a brief transition period before the melting reaction is concluded. A line is drawn that is tangent to the larger curve. The second point is where this line intersects with the tie line, which is the melting temperature. Both samples of Field s metal produced very similar heat curves with a double valley shape during its melting reaction. The first smaller valley occurs at approximately 59.5 C, while the second valley occurs at 61 C. The double valley suggests a transition period where the melting reaction does not emit a linear amount of energy but rather a smaller initial release of energy and then a larger secondary release. Figures 4 and 5 are the entire curves for the three fabricated alloys and the reference sample of Field s metal Heat Flow (mw) Temperature ( C) Figure 3: Determination of melting temperature for a liquid metal based upon the experimentalcalorimetric data.the melting point is determined at the intersection of the tangent to the slope and the baseline Heat Flow (mw) Heat Flow (mw) Temperature ( C) a) b) -1 Temperature ( C) Figure 4: Heat flow versus temperature for a) fabricated sample of Field's metal b) reference sample for the determination of melting temperature and specific heat calculations

5 Applied Mechanics and Materials Vol Heat Flow (mw) Heat Flow (mw) a) Temperature ( C) -2 b) Temperature ( C) Figure 5: Heat flow versus temperature for a) Indalloy-174 and b) Indalloy-127for the determination of melting temperature and specific heat calculations The other two alloys, Indalloy-174 and Indalloy-27, have a smooth single valley with a single point where the melting reaction occurs. This shape allows for an easier determination of the melting temperature with a single value of 79 C and 8.5 C for Indalloy-174 and Indalloy-27 respectively. Table 2 lists the melting temperatures and the percent difference between the reference values and experimental data. The calculated melting temperatures are all within 1 C of the reference values, which are in the range of acceptable accuracy for the application of ± 2 C[19]. Table 2: Measured melting temperatures of eutectic In-Bi-Sn alloys using a differential scanning calorimeter Alloy Recorded Melting Temperature [K] Reference Melting Temperature [K][13] Percent Difference [%] Field s Metal 332.5, Indalloy Indalloy The specific heat is calculated using the cell constant method, explained in Section 2, from the heat curve experimental data. Figure 7 displays the specific heat and temperature data. The peaks occur at the melting temperatures of the alloys and have the same shape of heat curve exothermic reaction expect in the positive direction. The area of interest is the specific heat of the alloys in their liquid state, which occurs to the right of the peak. Immediately, after the peak there is significant variation of up to 25% from the median value. Therefore, the median value is used to calculate the approximated specific heat in that region. Table 3: Uncertainty of Specific Heat Calculation Parameter Uncertainty Indium [E] ± 5 [%] Mass ±.5 [mg] Temperature ± 1 [K] Heat Flow ±5E-8[mW] Furthermore, it is important to note that the area between the curve and the tie-line or the area within the exothermic reaction is representative of the heat generated in the experiment. Therefore, this area can be used to calculate the heat of fusion (enthalpy of melting). However, this analysis was not performed as it is not needed for the creation of the thermal hydraulic model. Table 4 lists the calculated specific heat at 15 C or 423 K. Field s metal has the highest heat capacity of the three alloys. At 423 KLead-Bismuth Eutectic (LBE) has a specific heat in the range of J/kg-K. Therefore, Indalloy-27 has a specific heat most similar to Lead-Bismuth out of the alloys investigated. The result is not unexpected as Indalloy-27 has the highest bismuth content and therefore should be closest physically to LBE. Field s metal however is comprised of nearly two-thirds a combination of indium and tin which have a significantly higher specific heat than lead and bismuth.

6 19 Recent Trends in Materials and Mechanical Engineering II The experimental results are used to create an equation with the capability to approximate the specific heat for the purpose of inclusion into a numerical model. The correlations are only an approximation of the specific heat within a specified temperature. The calculated specific heats based on the correlations are presented in Figure 7. The correlations are presented in Table 4. The regions immediately before and after the spike in specific heat that occurs during melting have significant variance and have been corrected for the determination of the correlations. In the future a more detailed investigation of specific heat in this region may yield a more accurate correlation if it is necessary to have accurate knowledge near phase transitions. However, the correlations are considered sufficient for the approximation of specific heat in numerical models with a7% variance from the experimental data. Table 4: Experimentally determined specific heat of eutectic liquid metal alloys and reference values for other liquid metals Alloy Cp [J/kg-K] At 423[K] Field s Metal 25 Indalloy Indalloy Sn-Bi 263 Other Liquid Metals at T = 573 [K][11][21] Indium 247 Tin 244 Lead 145 Bismuth 14 Lead-Bismuth 15 Sodium 134 Lithium 428 Table 5: Experimentally derived correlations for the approximation of specific heat for a numerical model Alloy Correlation Temperature Range [ C] Field s Metal Fabrication x 1-3 T + 3 x 1-6 T Field s Metal Reference x 1-4 T + 2 x 1-6 T Indalloy x 1-3 T + 3 x 1-6 T Indalloy x 1-3 T + 2 x 1-6 T Cp(J/g-K) Bi-29.7 In Sn 57 Bi-26 In-17 Sn 58 Bi-42 Sn Temperature (K) Figure 6: Specific heat versus temperature for In-Bi-Sn eutectic alloys and eutectic Sn-Bi with error of ±25% in the liquid region.

7 Applied Mechanics and Materials Vol Specific Heat (J/g K) Field's Metal - Fabircated Field's Metal - Reference.5 Indalloy 27 Indalloy Temperature ( K) Figure 7: Calculated heat capacity of In-Bi-Sn eutectic alloys based upon experimentally determined correlations with a variance of ±7% from the experimental data Concluding Remarks The following concluding remarks were obtained from the completion of this study; The specific heat of the three eutectic alloys of interest in the In-Bi-Sn ternary system was experimentally determined. Field s metal is observed to have a higher specific heat than the other two alloys as expected due to the higher indium and tin content. Indalloy 27 has a specific heat similar to lead-bismuth eutectic. The melting temperatures of the In-Bi-Sn eutectic alloys have been confirmed and verified by this work. The temperatures are in agreement with the published data from the Indium Corporation as oppose to some critical temperatures of theses alloys published in literature. The heat rate curves also determine there is a transition occurring during melting for Field s metal that is undetermined and warrants further investigation. Correlations for the relationship between specific heat and temperaturehave been developed to approximate the specific heat of the In-Bi-Sn eutectic alloys. The correlations are considered suitable for anumerical model of a system utilizing these alloys as a liquid coolant. However, the correlations require further investigation and verification if they need to be used beyond the measured range of data or if more accuracy near the phase transition boundary is required. The data collected in this work can now be used as inputs into a thermal-hydraulic model to approximate the change in specific heat with temperature and establish the lower bound with respect to temperature for the model. Acknowledgments The authors wish to acknowledge and thank TheophileImbert at UOIT and on loan from Universite Joseph Fourier for his technical assistance and support.we also acknowledge Dr. Frank McCluskey, Dr. Matthew Kaye, Dr. GhausRivzi and Dr. UsmanSaeedof the University of Ontario Institute of Technology Materials Science Research Group for access to the equipment and support regarding the differential scanning calorimeter and for valuable discussions. This work is supported by NSERC of Canada Discovery Grants and Ontario Graduate Scholarships.

8 192 Recent Trends in Materials and Mechanical Engineering II References [1] (21) Gen IV International Forum. [Online]. [2] M. Matsuura, M. Hatori, and M. Ikeda, "Design and modification of steam generator safety system for FBR MONJU," Nuclear Engineering and Design, vol. 237, pp , 27. [3] N. Tanatugu, Y. Fuji-E, and T. Suita, "Frictional pressure drop for NaK-N2 two-phase flow in rectangular cross section channel of large aspect ratio," Journal of Nuclear Science and Technology, vol. 1, no. 4, pp , [4] J.M. Tourneir and M. El-Genk, "Reactor lithium heat pipes for HP-STMCs Space Reactor Power System," in Space Technology and Advances International Forum, 24. [5] International Atomic Energy Agency, "Status Report: Super-Safe, Small and Simple Reactor (4S)," IAEA, Vienna, 76, 211. [6] K. Uchimura, G.D. Harvel, T. Matsumato, M. Kanzaki, and J.S. Chang, "An image processing approach for two-phase interfaces visualized by a real-time neutron radiography technique," Flow Measurement and Instrumentation, vol. 9, pp , [7] T. Hibiki et al., "Study of flow characteristics in gas-molten metal mixture pool," Nuclear Engineering and Design, vol. 196, pp , 2. [8] A. Lipchitz, L. Laurent, and Harvel G.D., "Suitability of Eutectic Field's Metal for use in an Electromagnetohydrodynamically-enhanced Experimental Two-Phase Flow Loop," in 2th International Conference on Nuclear Engineering, Anaheim, USA., 212. [9] P. Roy, R.L. Orr, and R. Hultgren, "The thermodynamics of bismuth-lead alloys," Journal of Physical Chemistry, vol. 64, no. 8, pp , 196. [1] A. Terlain, "Thermodynamic relationships and heavy metal interaction with other coolants," in Handbook on lead-bismuth eutectic alloy and lead proerties, material compatibility, thermal-hydraulics and technologies., 27, ch. 3, pp [11] V. Sobolev and G. Benamati, "Thermophysical and electrical properties," in Handbook on lead-bismuth eutectic alloy and lead properties, material compatibility, thermal-hydraulics and technologies.: OECD, 27, ch. 2, pp [12] K.A. Shaikh, S. Li, and C. Liu, "Development of a latchable microvalve employing a low melting temperature metal alloy," Journal of Micoelectromechanical Systems, vol. 17, no. 5, pp , 28. [13] Indium Corporation, "Indalloy Specialty Alloys Properties Chart," 211. [14] R. Hultgren, Selected Values of the Thermodynamic Properties of Binary Alloys. Metals Park, Ohio, USA: ASM, [15] W.D. Callister Jr., Fundamentals of Materials Science and Engineering, 2nd ed. Hoboken, USA: John Wiley & sons, 25. [16] V.T. Witusiewicz, U. Hect, B. Bottger, and S. Rex, "Thermodynamic re-optimisation of the Bi-In-Sn system based on new experimental data," Journal of Alloys and Compounds, vol. 428, pp , 27. [17] A. Schilling and O. Jeandupeux, "High-accuracy differential thermal analysis: A tool for calorimetric investigations on small high-temperature-superconductor specimens," American Physical Society Physical Review B, vol. 52, no. 13, pp , 1995.

9 Applied Mechanics and Materials Vol [18] A. Maitre, J.M. Fiorani, and M. Vilasi, "Thermodynamic study of phase equilibria in the Pb-Bi-Hg system," Journal of Phase Equilibria, vol. 23, no. 4, pp , 22. [19] TA Instruments, "Differential Scanning Calorimetry,". [2] T. Sunagawa, A. Lipchitz, and G.D. Harvel, "Study of eutectic low melting alloy using In, Bi and Sn," Fukui University of Technology Journal, 213. [21] M.J. Richardson, "Specific heat capacities," Kaye & Laby, National Physical Laboratory, 213.

10 Recent Trends in Materials and Mechanical Engineering II 1.428/ Determination of Specific Heat of Eutectic Indium Bismuth-Tin Liquid Metal Alloys as a Test Material for Liquid Metal - Cooled Applications 1.428/