- Email: [email protected]
Preparation and investigation on density and surface tension of quaternary bromides for concentrating solar power
Solar Energy Materials & Solar Cells 157 (2016) 709–715
Contents lists available at ScienceDirect
Solar Energy Materials & Solar Cells journal homepage: www.elsevier.com/locate/solmat
Preparation and investigation on density and surface tension of quaternary bromides for concentrating solar power Yaxuan Xiong a,n, Jianfeng Shi a, Yuting Wu b, Chongfang Ma b, Hongbing Chen a a
Beijing Key Laboratory of Heating, Gas Supply, Ventilation and Air Conditioning, Beijing University of Civil Engineering and Architecture, Beijing 100044, China b Key Laboratory of Enhanced Heat Transfer and Energy Conservation of Ministry of Education & Key Laboratory of Heat Transfer and Energy Conversion of Beijing Municipality, Beijing University of Technology, Beijing 100124, China
art ic l e i nf o
a b s t r a c t
Article history: Received 11 April 2016 Received in revised form 26 June 2016 Accepted 12 July 2016 Available online 8 August 2016
In order to improve the unsteady high temperature heat transfer performance of receivers in solar thermal power stations the receivers should be optimized further. High temperature could help to optimize heat transfer of the receiver. Based the selection of heat pipes and bromides this work prepared six samples of quaternary bromides with different composition ratios of KBr, LiBr, NaBr, and CaBr2 and their density and surface tension were measured and analyzed. Experimental results show that density of quaternary bromides shows very good linearity with salt temperature, and density of quaternary bromides is higher than that of Solar salt at a same temperature. Also, quaternary bromide NO.2 has maximum density, and quaternary bromide NO.3 has minimum thermal expansion coefficient while quaternary bromide NO.6 has the minimum density and maximum thermal expansion coefficient at the same temperature. In addition, quaternary bromide NO.1 has a maximum surface tension and fitting curve slope at a same temperature among the six samples of quaternary bromides. Their density and surface tension were fitted with linear fit with their temperature respectively. Furthermore, that the composition ratio has evident influence on the density and surface tension of quaternary bromides has been concluded. & 2016 Elsevier B.V. All rights reserved.
Keywords: Concentrating solar power Heat pipes Quaternary bromides Density Surface tension
1. Introduction A concentrating solar power (CSP) system is usually a largescale way of solar energy to generate electricity, where a key problem is the coupled unsteady heat transfer from solar radiation to high temperature heat transfer fluid (HTF). Heat pipes with Alkali metals as HTF are of efficient heat transfer elements, which have been employed in CSP to improve the heat transfer [1–4]. However, Alkali metals heat pipes in case of fracture may cause fire or even serious explosion, which may bring about big economic losses and heavy casualties. Molten inorganic salts(Molten salts) as efficient HTFs have also been widely used to enhance heat transfer in the concentrating power tower receiver and the thermal energy storage (TES) system in CSP system [5], which is of low vapor pressure, wide operating temperature range, low cost and to be inert in comparison to Alkali metals. Thus employing molten salts as the working fluid of heat pipes would promote the heat transfer stability and safety of receivers in CSP system. Molten salts that have been evaluated experimentally and n
Corresponding author. E-mail address: [email protected] (Y. Xiong).
http://dx.doi.org/10.1016/j.solmat.2016.07.016 0927-0248/& 2016 Elsevier B.V. All rights reserved.
analytically for CSP include mainly nitrates, Chlorides and carbonates. Currently, Multi-component salt went to the center of efficient heat transfer material development due to the mismatch of melting point, working temperature and temperature range, decomposition temperature of single salt. Density and surface tension are two fundamental thermo-physical properties of HTF as the working fluid of heat pipes, which have direct influence on the comprehensive heat transfer performance-the larger the density and surface tension are, the higher the heat transfer power is and the lower the heat transfer resistance of heat pipes is. Several nitrates, carbonates and chlorides have been developed and their density and surface tension have been studied extensively by authors [6–18]. However, these salts have either high melting point or low decomposition temperature or density, which is unsuitable for heat pipes. It is found that molten bromide mixtures can satisfy high working temperature, density and surface tension. Decomposition temperature of quaternary bromides with different component ratios of NaBr, KBr, CaBr2 and LiBr is proved to be up to 900 °C while their melting point is as low as 300 °C [19]. It is difficult to find a suitable molten salt serving as the working fluid of high temperature heat pipes. Density and surface tension are two fundamental thermo-physical properties of molten salt as the
710
Y. Xiong et al. / Solar Energy Materials & Solar Cells 157 (2016) 709–715
working fluid of heat pipes. Data of density and surface tension of quaternary bromide mixtures are rare and have ever been measured specially. In the present work density and surface tension of quaternary bromides with different component ratios of NaBr, KBr, CaBr2 and LiBr were measured. The Purpose of this work intends to provide data for selection of molten salt for heat pipes in concentrating solar power system. This work provides basic data of density and surface tension for molten salt selection used for heat pipes in concentrating solar power systems. The results of density, thermal expansion coefficient and surface tension of quaternary bromides would be presented and discussed in the following.
2. Experimental 2.1. Preparations of quaternary bromides Materials used in the presented experiments are sodium bromide (AR), potassium bromide (AR), lithium bromide (AR) and calcium bromide (AR), which are all from the manufacturer: Tianjin Zhiyuan ChemicalReagen Co. Ltd. All bromides were treated by dehydration before the preparations. 6 quaternary bromide mixtures were prepared based on the different mass proportions of NaBr-KBr-LiBr-CaBr, as shown in Table 1. The mass of mixed bromides were weighed by using a precision balance with a precision of 0.1 mg produced by Mettler Toledo Instrument Company. The bromides were mixed well by a high-speed disintegrator. The prepared samples were stored in a drying box kept at 200 °C for experiments.
Fig. 1. Schematic diagram of the density probe.
2.2. Density measurement Density is an important thermo-physical property of molten salts and has great significance in practical and theoretical research. In order to eliminate the influence resulted from the surface tension the density of the molten quaternary bromides is measured by the improved Archimedean principle [20] in the present work. The crucible is made of high pure alumina with purity of 99.9% and the density probe is made of 316 L stainless steel in present experiments which have extremely low reactivity with bromide salts with the protection of inert gas - Helium. An accurate density of molten salts prepared can be obtained according Eq. (1) as follows:
ρ=
M1 − M2 Vρ
(1)
Where M1 is the weight of the probe when the surface of the melt is at position 1, M2 is the weight of the probe at position 2, while Vρ is the volume of the probe between position 1 and 2. Vρ is measured in advance with the same method by using densityknown molten salts at different temperature. Schematic diagrams of the density probe and the experimental apparatus are show in Figs. 1 and 2. Fig. 2. Schematic diagram of the experimental apparatus. Table 1 mass ratio of quaternary bromides. NO.
1 2 3 4 5 6
Mass fraction, % NaBr
KBr
CaBr2
LiBr
5 5 8 11 14 17
25 27 31 35 39 43
15 18 16 14 12 10
55 50 45 40 35 30
An electric balance with an accuracy of 0.1 mg is used to measure the weight of the probe. The relative error is measurement of density is estimated to be with 70.1%。. 2.3. Surface tension measurement The Pull-Off method originating from the du Noüy Ring method is employed to measure the surface tension of the molten salts, which is based on the measurement of maximum equilibrium force to pull a probe out of the molten salts. The surface tension
Y. Xiong et al. / Solar Energy Materials & Solar Cells 157 (2016) 709–715
711
probe used here is made of 316 L stainless steel which has a very low reactivity with the protection of inert gas - Helium and good wetting property with the measured quaternary bromides to ensure measurements accurately. The surface tension can be expressed by Eq. (2) as
σ=
C ( M2 − M1)g 4π R
(2)
where s is the surface tension, R is the average radius of the bottom ring of the probe, M1 is the static weight of the probe dipped into the molten salts surface, M2 is the maximum equilibrium force; g is the gravity acceleration, and C is a constant. The radius R at room temperature can be measured in advance, and then R can be modified in high temperature by the following formula as
R = R T0 3 1 + 3β( T − T0)
(3)
where RT0 is the radius of the probe at room temperature, β is the linear thermal expansion coefficient of 316 L stainless steel, which is 1.9 10 5 K 1. The constant C can be obtained by measuring a surface-tension-known molten salt at various temperatures. A same type of electric balance is used to measure the weight of the probe at different equilibrium. The front view of the surface-tension probe is given in Fig. 3 and the Schematic diagram of surface-tension rig is shown as Fig. 2. During the experiment the surface-tension probe is pulled up with 0.01 m/s to achieve force equilibrium. When experiments are carried out the experimental apparatus was placed in an air conditioning room to keep the temperature.
3. Results and discussion In order to eliminate the effects of internal natural convection of molten salts the experiments in this work were carried out during the cooling process of the molten quaternary bromides. The
Fig. 4. Density data of molten LiNO3 and molten Solar Salt in comparison with available data.
cooling speed is less than 2 K/min to eliminate the influence of the inner flow of the molten salt interior. 3.1. Calibration of test rig Density and Surface tension are two important thermo-physical parameters which affects the heat transfer capability of molten salts. In order to calibrate the test rig the famous Solar Salt (NaNO3:60%, KNO3:40%) and analytical pure LiNO3 are chose. The temperature dependence of the density and surface tension of molten LiNO3 and molten Solar Salt are shown in Figs. 4 and 5 respectively. Comparisons between density data from this work and that from references show that the maximum bias in the temperature range in this work is estimated to be less than 0.5% while the maximum bias between the surface tension data and that from References is less than 1.0%. 3.2. Density measurement and discussion Density of the six quaternary bromides was measured employing the above experimental apparatus and the density probe. The density data of different quaternary bromides are shown in Figs. 6 to 12 respectively.
Fig. 3. schematic diagram of the surface-tension probe.
Fig. 5. Surface tension data of Solar salt and NaNO3 in comparison with the available data.
712
Y. Xiong et al. / Solar Energy Materials & Solar Cells 157 (2016) 709–715
Fig. 6. Density curve of quaternary bromides NO.1.
Fig. 8. Density curve of quaternary bromides NO.3.
Fig. 7. Density curve of quaternary bromides NO.2.
2.60
measured value fitting curve
2.59 2.58 3
Density/(g/m )
Fig. 6 shows that the density of sample NO.1 falls into 2.614 and 2.529 g/cm3 and decreases linearly with the increase of temperature from 704 K to 823 K. Density of sample NO.2 falls into 2.658 and 2.585 g/cm3 and decreases linearly with the increase of temperature from 678 K to 792 K as seen from Fig. 7. Density data of both quaternary bromides show very good linearity with salt temperature, which is consistent in trend to that of Solar Salt. For sample NO.3, as shown in Fig. 8, the density data fall into 2.640 and 2.560 g/cm3, and decreases linearly with the increase of temperature from 666 K to 797 K. The density data for sample NO.4 fall into 2.624 and 2.552 g/cm3 and decreases linearly with the increase of temperature from 660 K to 767 K as shown in Fig. 9. The fitting curves are straight lines as those in Figs. 6 and 7. Seeing from Figs. 10 and 11, density of sample NO.5 descends from 2.598 to 2.530 g/cm3 with the rise of temperature from 676 K to 788 K, while density of sample NO. 6 descend from 2.538 to 2.443 g/ cm3 with the rise of temperature from 733 K to 851 K linearly. Taking Figs. 6–11 into consideration it can be seen that that there is a good linearity between density and temperature for each quaternary bromides. Also we know that sample NO.2 has maximum density, which is the best working fluid among the six quaternary bromides for heat pipes from the view of density, while sample NO.6 has the minimum density at the same temperature; Density of sample NO.2 is greater than that of sample NO.3, density of sample NO.3 is greater than that of sample NO.1 and density of sample NO.1 is greater than that of sample NO.4,
Fig. 9. Density curve of quaternary bromides NO.4.
2.57 2.56 2.55 2.54 2.53 2.52 660
680
700
720
740
760
780
800
Temperarture/ Fig. 10. Density curve of quaternary bromides NO.5.
while density of sample NO.4 is greater than that of sample NO.5. For thermal storage higher density would reduce the volume of the TES tank. So sample NO.2 among the six samples should be the best choice, and then is sample NO.3 or sample NO.1. Fitting correlations of Density data of the 6 quaternary bromides are listed in Table 2.
Y. Xiong et al. / Solar Energy Materials & Solar Cells 157 (2016) 709–715
713
Fig. 11. Density curve of quaternary bromides NO.6.
Table 2 Correlated formulas of density of quaternary bromides. No.
σ = a − b × 10−4T
1 2 3 4 5 6
a
b
3.1273 3.0869 3.0382 3.0921 2.9964 3.0892
7.3171 6.3456 5.9607 7.0698 5.8842 7.5492
R2
Temperature range/°C
0.986 0.993 0.957 0.986 0.958 0.987
704 823 678 792 666 797 660 767 676 788 733 851
Fig. 12. Temperature dependence of thermal expansion coefficient of quaternary bromides.
From the density data and fitting correlations the thermal expansion coefficient of the 6 quaternary bromides can be obtained by employing its definition as follows:
β=−
1 dρ ρ dT
(9)
The thermal expansion coefficient of quaternary bromides depending on temperature is shown in Fig. 12. From Fig. 12 it can be seen that the thermal expansion coefficient of the quaternary bromides increases linearly approximately with the rise of temperature. Sample NO.6 has the maximum thermal expansion coefficient at the same temperature, which is greater than that of sample NO.1; thermal expansion coefficient of sample NO.1 is greater than that of sample NO.4, thermal expansion coefficient of sample NO.4 is greater than that of sample NO.5, and thermal expansion coefficient of sample NO.4 is greater than that of sample NO.2, while sample NO.3 has the minimum thermal expansion coefficient. So sample NO.3 needs least reserved expansion volume in a TES tank either for either sensible thermal storage or for PCM thermal storage. 3.3. Surface tension measurement and discussion All samples are melting after being heated to 620 K. When the temperature was above 770 K, there was some volatile molten salt adhering to the probe. This made it difficult to calculate M1 and M2 accurately, which may result in relatively large experimental errors. Therefore, the measuring temperature range was set within a range of 633–773 K. Measured surface tensions of the six quaternary bromides are shown in Figs. 13–18 respectively, and Table 3 lists the fitting correlations of experiment data. Fig. 13 shows the surface tension of sample NO.1 which ranges linearly from 0.1240 to 0.1152 N m 1 with the rise of temperature
Fig. 13. Measured surface tension values and fitting curve for quaternary bromides NO.1.
from 703 K to 763 K. Fig, and there is a good agreement between the measured surface tension and the fitting curve for sample NO.1. For sample NO.2, Fig. 14 indicates that the surface tension falls within 0.11728–0.11450 N m 1 from 673 K to 758 K, while the surface tension diminishes linearly with the increases of salt temperature as with sample NO.1. From Figs. 15 and 16 we know The surface tension of sample NO.3 is within 0.11753–0.11260 N m 1 in 648–733 K while for sample NO.4 The surface tension ranges linearly within 0.11803– 0.11375 N m 1 in 633–722 K. The tendency for surface tensions to decreases linearly with the increase of salt temperature is also observed for sample NO.3 and sample NO.4. For Figs. 17 and 18 there are similar trends for the changing of surface tensions with the increase of salt temperature for sample NO.5 and sample NO.6. The surface tension of sample NO.5 is within 0.11551–0.11273 N m 1 in 663–737 K while for sample NO.4 the surface tension ranges linearly within 0.11812– 0.11249 N m 1 in 695–773 K.
714
Y. Xiong et al. / Solar Energy Materials & Solar Cells 157 (2016) 709–715
Fig. 14. Measured surface tension values and fitting curve for quaternary bromides NO.2.
Fig. 15. Measured surface tension values and fitting curve for quaternary bromides NO.3.
Fig. 17. Measured surface tension values and fitting curve for quaternary bromides NO.5.
Fig. 18. Measured surface tension values and fitting curve for quaternary bromides NO.6. Table 3 Correlated formulas of surface tension of quaternary bromides. No.
1 2 3 4 5 6
Fig. 16. Measured surface tension values and fitting curve for quaternary bromides NO.4.
s¼ a b 10 5T a
b
0.2276 0.1397 0.1550 0.1485 0.1406 0.1681
14.71 3.3232 5.57888 4.7981 3.7755 7.2009
R2
Temperature range/K
0.9912 0.996 0.999 0.992 0.999 0.999
703 763 673 758 678 733 633 722 663 737 695 773
Detailed surface tension curves fitted in Figs. 13–17 for sample NO.1 to NO. 6 are listed in Table 3. Based on Figs. 13–18 a conclusion can be achieved that all the fitting curves of surface tension of the six quaternary bromides have similar trends that surface tension decreases linearly with the increase of temperature, which is consistent with the results of literatures on other liquids [7]. Moreover, it shows that the fitting curve of surface tension of sample NO.1 has a maximum slope and
Y. Xiong et al. / Solar Energy Materials & Solar Cells 157 (2016) 709–715
a maximum surface tension at the same temperature, which is the best working fluid among the six quaternary bromides for heat pipes from the view of surface tension, which benefits to the heat transfer of molten salts, among all the six quaternary bromides. Surface tension of sample NO.6 changes slower than that of sample NO.1 but faster than NO.3, while Surface tension of sample NO.3 changes slower than that of sample NO.6 but faster than NO.4. Surface tension of sample NO.5 changes slower than that of sample NO.4 but faster than NO.2. From Table 1 and Figs. 13–18 we also find that the composition of quaternary bromides has significant effects on surface tension: the slopes of the fitting curves of the samples first increase and then decrease as the mass fractions of NaBr and KBr increase and those of LiBr and CaBr2 decrease. The less the surface tension is, the weaker the heat transfer ability is.
4. Conclusions Mixed molten salts are considered as promising medium for both heat transfer and thermal storage in solar thermal power plants. In the present work six novel quaternary bromides are prepared by mixing NaBr, KBr, LiBr and CaBr in different proportions. And then the improved Archimedean principle and the PullOff method were employed to measure the density and surface tension of six quaternary bromides respectively, and the thermal expansion coefficient for each quaternary bromide were calculated and fitted. For density, there is a good linearity between density and temperature for each quaternary bromide. Sample NO.2 has maximum density, which is greater than that of sample NO.3; Density of sample NO.3 is greater than that of sample NO.1, density of sample NO.1 is greater than that of sample NO.4, and density of sample NO.4 is greater than that of sample NO.5, while sample NO.6 has the minimum density. For heat transfer of heat pipes sample NO.2 should be the best choice, and then is sample NO.3 and sample NO.1. Thermal expansion coefficients of the six quaternary bromides increase linearly approximately with the rise of temperature. But there are evident difference in the values and fitted curves. Sample NO.6 has the maximum thermal expansion coefficient while sample NO.3 has the minimum thermal expansion coefficient at a same temperature. The sample NO.3 would reduces the reserved expansion volume in a TES tank for either sensible thermal storage. For surface tension, the six quaternary bromides have similar trends that the surface tension descends linearly with the rise of temperature. Fitting curve of surface tension of sample NO.1 has a maximum slope and the maximum surface tension at the same temperature, which benefits to the heat transfer of molten salts, among all the six quaternary bromides. Surface tension of sample NO.6 changes slower than that of sample NO.6 but faster than NO.1, while Surface tension of sample NO.3 changes slower than that of sample NO.6 but faster than NO.4 and Surface tension of sample NO.5 changes slower than that of sample NO.4 but faster than NO.2. The composition of quaternary bromides also has significant effects on surface tension: the slopes of the fitting curves of the samples first increase and then decrease as the mass fractions of NaBr and KBr increase and those of LiBr and CaBr2 decrease.
715
Funding sources This work was supported by the National Nature Science Foundation of China under Project [grant numbers 51206004]; and the Key Project of Beijing Natural Science Foundation under Project [grant number 3151001].
References [1] S. Almsater, W. Saman, F. Bruno, Performance enhancement of high temperature latent heat thermal storage systems using heat pipes with and without fins for concentrating solar thermal power plants, Renew. Energy 89 (2016) 36–50, http://dx.doi.org/10.1016/j.renene.2015.11.068. [2] S. Tiari, S. Qiu, M. Mahdavi, Discharging process of a finned heat pipe–assisted thermal energy storage system with high temperature phase change material, Energy Convers. Manag. 118 (2016) 426–437, http://dx.doi.org/10.1016/j. enconman.2016.04.025. [3] Z. Liao, A. Faghri, Thermal analysis of a heat pipe solar central receiver for concentrated solar power tower, Appl. Therm. Eng. 102 (2016) 952–960, http: //dx.doi.org/10.1016/j.applthermaleng.2016.04.043. [4] Y. Cao, Separate-type heat pipe solar receivers for concentrating solar power, Front. Heat. Pipes (2015), http://dx.doi.org/10.5098/fhp.6.1. [5] L.X. Sang, M. Cai, N. Ren, Y.T. Wu, C. Burda, Improving the thermal properties of ternary carbonates for concentrating solar power through simple chemical modifications by adding sodium hydroxide and nitrate, Sol. Energy Mater. Sol. Cells 124 (2014) 61–66, http://dx.doi.org/10.1016/j.solmat.2014.01.025. [6] G.J. Janz, U. Krebs, H.F. Siegenthaler, R.P.T. Tomkins, Molten salts: volume 3 nitrates, nitrites, and mixtures: electrical conductance, density, viscosity, and surface tension data, J. Phys. Chem. Ref. Data 1 (1972) 581–672, http://dx.doi. org/10.1063/1.3253103. [7] H. Sasaki, E. Tokizaki, K. Terashima, S. Kimura, Density measurement of molten silicon by an improved Archimedian method, J. Cryst. Growth 139 (1994) 225–230, http://dx.doi.org/10.1016/0022-0248(94)90170-8. [8] Q. Peng, J. Ding, X.L. Wei, J.P. Yang, X.P. Yang, The preparation and properties of multi-component molten salts, Appl. Energy 87 (2010) 2812–2817, http://dx. doi.org/10.1016/j.apenergy.2009.06.022. [9] J.G. Cordaro, N.C. Rubin, R.W. Bradshaw, Multicomponent molten salt mixtures based on nitrate/nitrite anions, J. Sol. Energy Eng. 133 (2011) 011–014, http: //dx.doi.org/10.1115/1.4003418. [10] J.W. Raade, D. Padowitz, Development of molten salt heat transfer fluid with low melting point and high thermal stability, J. Sol. Energy Eng. 133 (2011) 1–6, http://dx.doi.org/10.1115/1.4004243. [11] T. Bauer, N. Pfleger, N. Breidenbach, M. Eck, D. Laing, et al., Material aspects of solar salt for sensible heat storage, Appl. Energy 111 (2013) 1114–1119, http: //dx.doi.org/10.1016/j.apenergy.2013.04.072. [12] T. Wang, D. Mantha, R.G. Reddy, Novel low melting point quaternary eutectic system for solar thermal energy storage, Appl. Energy 102 (2013) 1422–1429, http://dx.doi.org/10.1016/j.apenergy.2012.09.001. [13] Y. Marcus, Surface tension and cohesive energy density of molten salts, Thermochim. Acta 571 (2013) 77–81, http://dx.doi.org/10.1016/j. tca.2013.08.018. [14] C.J. Li, P. Li, K. Wang, E.E. Molina, Survey of properties of key single and mixture halide salts for potential application as high temperature heat transfer fluids for concentrated solar thermal power systems, AIMS Energy 2 (2014) 133–157, http://dx.doi.org/10.3934/energy.2014.2.133. [15] N. Ren, Y.T. Wu, C.F. Ma, L.X. Sang, Preparation and thermal properties of quaternary mixed nitrate with low melting point, Sol. Energy Mater. Sol. Cells 127 (2014) 6–13, http://dx.doi.org/10.1016/j.solmat.2014.03.056. [16] F. Aqra, Surface tension of molten metal halide salts, J. Mol. Liq. 200 (2014) 120–121, http://dx.doi.org/10.1016/j.molliq.2014.09.052. [17] L.X. Sang, M. Cai, Y.B. Zhao, N. Ren, Y.T. Wu, et al., Mixed metal carbonates/ hydroxides for concentrating solar power analyzed with DSC and XRD, Sol. Energy Mater. Sol. Cells 140 (2015) 167–173, http://dx.doi.org/10.1016/j. solmat.2015.04.006. [18] X.H. An, J.H. Cheng, P. Zhang, Z.F. Tang, J.Q. Wang, Determination and evaluation of the thermophysical properties of an alkali carbonate eutectic molten salt, Faraday Discuss. (2016), http://dx.doi.org/10.1039/C5FD00236B. [19] G.J. Janz, R.P.T. Tomkinns, C.B. Allen, J.R. Downey, S.K. Singer, Molten salts: volume 4, part 3, bromides and mixtures, iodides and mixtures - electrical conductance, density, viscosity, and surface tension data, J. Phys. Chem. Ref. Data 6 (1977) 411–587, http://dx.doi.org/10.1063/1.555552. [20] H. Sasaki, E. Tokizaki, K. Terashima, S. Kimura, Density measurement of molten silicon by an improved Archimedean method, Journal of Crystal Growth, 139, 1994, pp. 225–230, 10.1016/0022–0248(94)90170-8. 10.1016/0022–0248(94) 90170-90178.
Contents lists available at ScienceDirect
Solar Energy Materials & Solar Cells journal homepage: www.elsevier.com/locate/solmat
Preparation and investigation on density and surface tension of quaternary bromides for concentrating solar power Yaxuan Xiong a,n, Jianfeng Shi a, Yuting Wu b, Chongfang Ma b, Hongbing Chen a a
Beijing Key Laboratory of Heating, Gas Supply, Ventilation and Air Conditioning, Beijing University of Civil Engineering and Architecture, Beijing 100044, China b Key Laboratory of Enhanced Heat Transfer and Energy Conservation of Ministry of Education & Key Laboratory of Heat Transfer and Energy Conversion of Beijing Municipality, Beijing University of Technology, Beijing 100124, China
art ic l e i nf o
a b s t r a c t
Article history: Received 11 April 2016 Received in revised form 26 June 2016 Accepted 12 July 2016 Available online 8 August 2016
In order to improve the unsteady high temperature heat transfer performance of receivers in solar thermal power stations the receivers should be optimized further. High temperature could help to optimize heat transfer of the receiver. Based the selection of heat pipes and bromides this work prepared six samples of quaternary bromides with different composition ratios of KBr, LiBr, NaBr, and CaBr2 and their density and surface tension were measured and analyzed. Experimental results show that density of quaternary bromides shows very good linearity with salt temperature, and density of quaternary bromides is higher than that of Solar salt at a same temperature. Also, quaternary bromide NO.2 has maximum density, and quaternary bromide NO.3 has minimum thermal expansion coefficient while quaternary bromide NO.6 has the minimum density and maximum thermal expansion coefficient at the same temperature. In addition, quaternary bromide NO.1 has a maximum surface tension and fitting curve slope at a same temperature among the six samples of quaternary bromides. Their density and surface tension were fitted with linear fit with their temperature respectively. Furthermore, that the composition ratio has evident influence on the density and surface tension of quaternary bromides has been concluded. & 2016 Elsevier B.V. All rights reserved.
Keywords: Concentrating solar power Heat pipes Quaternary bromides Density Surface tension
1. Introduction A concentrating solar power (CSP) system is usually a largescale way of solar energy to generate electricity, where a key problem is the coupled unsteady heat transfer from solar radiation to high temperature heat transfer fluid (HTF). Heat pipes with Alkali metals as HTF are of efficient heat transfer elements, which have been employed in CSP to improve the heat transfer [1–4]. However, Alkali metals heat pipes in case of fracture may cause fire or even serious explosion, which may bring about big economic losses and heavy casualties. Molten inorganic salts(Molten salts) as efficient HTFs have also been widely used to enhance heat transfer in the concentrating power tower receiver and the thermal energy storage (TES) system in CSP system [5], which is of low vapor pressure, wide operating temperature range, low cost and to be inert in comparison to Alkali metals. Thus employing molten salts as the working fluid of heat pipes would promote the heat transfer stability and safety of receivers in CSP system. Molten salts that have been evaluated experimentally and n
Corresponding author. E-mail address: [email protected] (Y. Xiong).
http://dx.doi.org/10.1016/j.solmat.2016.07.016 0927-0248/& 2016 Elsevier B.V. All rights reserved.
analytically for CSP include mainly nitrates, Chlorides and carbonates. Currently, Multi-component salt went to the center of efficient heat transfer material development due to the mismatch of melting point, working temperature and temperature range, decomposition temperature of single salt. Density and surface tension are two fundamental thermo-physical properties of HTF as the working fluid of heat pipes, which have direct influence on the comprehensive heat transfer performance-the larger the density and surface tension are, the higher the heat transfer power is and the lower the heat transfer resistance of heat pipes is. Several nitrates, carbonates and chlorides have been developed and their density and surface tension have been studied extensively by authors [6–18]. However, these salts have either high melting point or low decomposition temperature or density, which is unsuitable for heat pipes. It is found that molten bromide mixtures can satisfy high working temperature, density and surface tension. Decomposition temperature of quaternary bromides with different component ratios of NaBr, KBr, CaBr2 and LiBr is proved to be up to 900 °C while their melting point is as low as 300 °C [19]. It is difficult to find a suitable molten salt serving as the working fluid of high temperature heat pipes. Density and surface tension are two fundamental thermo-physical properties of molten salt as the
710
Y. Xiong et al. / Solar Energy Materials & Solar Cells 157 (2016) 709–715
working fluid of heat pipes. Data of density and surface tension of quaternary bromide mixtures are rare and have ever been measured specially. In the present work density and surface tension of quaternary bromides with different component ratios of NaBr, KBr, CaBr2 and LiBr were measured. The Purpose of this work intends to provide data for selection of molten salt for heat pipes in concentrating solar power system. This work provides basic data of density and surface tension for molten salt selection used for heat pipes in concentrating solar power systems. The results of density, thermal expansion coefficient and surface tension of quaternary bromides would be presented and discussed in the following.
2. Experimental 2.1. Preparations of quaternary bromides Materials used in the presented experiments are sodium bromide (AR), potassium bromide (AR), lithium bromide (AR) and calcium bromide (AR), which are all from the manufacturer: Tianjin Zhiyuan ChemicalReagen Co. Ltd. All bromides were treated by dehydration before the preparations. 6 quaternary bromide mixtures were prepared based on the different mass proportions of NaBr-KBr-LiBr-CaBr, as shown in Table 1. The mass of mixed bromides were weighed by using a precision balance with a precision of 0.1 mg produced by Mettler Toledo Instrument Company. The bromides were mixed well by a high-speed disintegrator. The prepared samples were stored in a drying box kept at 200 °C for experiments.
Fig. 1. Schematic diagram of the density probe.
2.2. Density measurement Density is an important thermo-physical property of molten salts and has great significance in practical and theoretical research. In order to eliminate the influence resulted from the surface tension the density of the molten quaternary bromides is measured by the improved Archimedean principle [20] in the present work. The crucible is made of high pure alumina with purity of 99.9% and the density probe is made of 316 L stainless steel in present experiments which have extremely low reactivity with bromide salts with the protection of inert gas - Helium. An accurate density of molten salts prepared can be obtained according Eq. (1) as follows:
ρ=
M1 − M2 Vρ
(1)
Where M1 is the weight of the probe when the surface of the melt is at position 1, M2 is the weight of the probe at position 2, while Vρ is the volume of the probe between position 1 and 2. Vρ is measured in advance with the same method by using densityknown molten salts at different temperature. Schematic diagrams of the density probe and the experimental apparatus are show in Figs. 1 and 2. Fig. 2. Schematic diagram of the experimental apparatus. Table 1 mass ratio of quaternary bromides. NO.
1 2 3 4 5 6
Mass fraction, % NaBr
KBr
CaBr2
LiBr
5 5 8 11 14 17
25 27 31 35 39 43
15 18 16 14 12 10
55 50 45 40 35 30
An electric balance with an accuracy of 0.1 mg is used to measure the weight of the probe. The relative error is measurement of density is estimated to be with 70.1%。. 2.3. Surface tension measurement The Pull-Off method originating from the du Noüy Ring method is employed to measure the surface tension of the molten salts, which is based on the measurement of maximum equilibrium force to pull a probe out of the molten salts. The surface tension
Y. Xiong et al. / Solar Energy Materials & Solar Cells 157 (2016) 709–715
711
probe used here is made of 316 L stainless steel which has a very low reactivity with the protection of inert gas - Helium and good wetting property with the measured quaternary bromides to ensure measurements accurately. The surface tension can be expressed by Eq. (2) as
σ=
C ( M2 − M1)g 4π R
(2)
where s is the surface tension, R is the average radius of the bottom ring of the probe, M1 is the static weight of the probe dipped into the molten salts surface, M2 is the maximum equilibrium force; g is the gravity acceleration, and C is a constant. The radius R at room temperature can be measured in advance, and then R can be modified in high temperature by the following formula as
R = R T0 3 1 + 3β( T − T0)
(3)
where RT0 is the radius of the probe at room temperature, β is the linear thermal expansion coefficient of 316 L stainless steel, which is 1.9 10 5 K 1. The constant C can be obtained by measuring a surface-tension-known molten salt at various temperatures. A same type of electric balance is used to measure the weight of the probe at different equilibrium. The front view of the surface-tension probe is given in Fig. 3 and the Schematic diagram of surface-tension rig is shown as Fig. 2. During the experiment the surface-tension probe is pulled up with 0.01 m/s to achieve force equilibrium. When experiments are carried out the experimental apparatus was placed in an air conditioning room to keep the temperature.
3. Results and discussion In order to eliminate the effects of internal natural convection of molten salts the experiments in this work were carried out during the cooling process of the molten quaternary bromides. The
Fig. 4. Density data of molten LiNO3 and molten Solar Salt in comparison with available data.
cooling speed is less than 2 K/min to eliminate the influence of the inner flow of the molten salt interior. 3.1. Calibration of test rig Density and Surface tension are two important thermo-physical parameters which affects the heat transfer capability of molten salts. In order to calibrate the test rig the famous Solar Salt (NaNO3:60%, KNO3:40%) and analytical pure LiNO3 are chose. The temperature dependence of the density and surface tension of molten LiNO3 and molten Solar Salt are shown in Figs. 4 and 5 respectively. Comparisons between density data from this work and that from references show that the maximum bias in the temperature range in this work is estimated to be less than 0.5% while the maximum bias between the surface tension data and that from References is less than 1.0%. 3.2. Density measurement and discussion Density of the six quaternary bromides was measured employing the above experimental apparatus and the density probe. The density data of different quaternary bromides are shown in Figs. 6 to 12 respectively.
Fig. 3. schematic diagram of the surface-tension probe.
Fig. 5. Surface tension data of Solar salt and NaNO3 in comparison with the available data.
712
Y. Xiong et al. / Solar Energy Materials & Solar Cells 157 (2016) 709–715
Fig. 6. Density curve of quaternary bromides NO.1.
Fig. 8. Density curve of quaternary bromides NO.3.
Fig. 7. Density curve of quaternary bromides NO.2.
2.60
measured value fitting curve
2.59 2.58 3
Density/(g/m )
Fig. 6 shows that the density of sample NO.1 falls into 2.614 and 2.529 g/cm3 and decreases linearly with the increase of temperature from 704 K to 823 K. Density of sample NO.2 falls into 2.658 and 2.585 g/cm3 and decreases linearly with the increase of temperature from 678 K to 792 K as seen from Fig. 7. Density data of both quaternary bromides show very good linearity with salt temperature, which is consistent in trend to that of Solar Salt. For sample NO.3, as shown in Fig. 8, the density data fall into 2.640 and 2.560 g/cm3, and decreases linearly with the increase of temperature from 666 K to 797 K. The density data for sample NO.4 fall into 2.624 and 2.552 g/cm3 and decreases linearly with the increase of temperature from 660 K to 767 K as shown in Fig. 9. The fitting curves are straight lines as those in Figs. 6 and 7. Seeing from Figs. 10 and 11, density of sample NO.5 descends from 2.598 to 2.530 g/cm3 with the rise of temperature from 676 K to 788 K, while density of sample NO. 6 descend from 2.538 to 2.443 g/ cm3 with the rise of temperature from 733 K to 851 K linearly. Taking Figs. 6–11 into consideration it can be seen that that there is a good linearity between density and temperature for each quaternary bromides. Also we know that sample NO.2 has maximum density, which is the best working fluid among the six quaternary bromides for heat pipes from the view of density, while sample NO.6 has the minimum density at the same temperature; Density of sample NO.2 is greater than that of sample NO.3, density of sample NO.3 is greater than that of sample NO.1 and density of sample NO.1 is greater than that of sample NO.4,
Fig. 9. Density curve of quaternary bromides NO.4.
2.57 2.56 2.55 2.54 2.53 2.52 660
680
700
720
740
760
780
800
Temperarture/ Fig. 10. Density curve of quaternary bromides NO.5.
while density of sample NO.4 is greater than that of sample NO.5. For thermal storage higher density would reduce the volume of the TES tank. So sample NO.2 among the six samples should be the best choice, and then is sample NO.3 or sample NO.1. Fitting correlations of Density data of the 6 quaternary bromides are listed in Table 2.
Y. Xiong et al. / Solar Energy Materials & Solar Cells 157 (2016) 709–715
713
Fig. 11. Density curve of quaternary bromides NO.6.
Table 2 Correlated formulas of density of quaternary bromides. No.
σ = a − b × 10−4T
1 2 3 4 5 6
a
b
3.1273 3.0869 3.0382 3.0921 2.9964 3.0892
7.3171 6.3456 5.9607 7.0698 5.8842 7.5492
R2
Temperature range/°C
0.986 0.993 0.957 0.986 0.958 0.987
704 823 678 792 666 797 660 767 676 788 733 851
Fig. 12. Temperature dependence of thermal expansion coefficient of quaternary bromides.
From the density data and fitting correlations the thermal expansion coefficient of the 6 quaternary bromides can be obtained by employing its definition as follows:
β=−
1 dρ ρ dT
(9)
The thermal expansion coefficient of quaternary bromides depending on temperature is shown in Fig. 12. From Fig. 12 it can be seen that the thermal expansion coefficient of the quaternary bromides increases linearly approximately with the rise of temperature. Sample NO.6 has the maximum thermal expansion coefficient at the same temperature, which is greater than that of sample NO.1; thermal expansion coefficient of sample NO.1 is greater than that of sample NO.4, thermal expansion coefficient of sample NO.4 is greater than that of sample NO.5, and thermal expansion coefficient of sample NO.4 is greater than that of sample NO.2, while sample NO.3 has the minimum thermal expansion coefficient. So sample NO.3 needs least reserved expansion volume in a TES tank either for either sensible thermal storage or for PCM thermal storage. 3.3. Surface tension measurement and discussion All samples are melting after being heated to 620 K. When the temperature was above 770 K, there was some volatile molten salt adhering to the probe. This made it difficult to calculate M1 and M2 accurately, which may result in relatively large experimental errors. Therefore, the measuring temperature range was set within a range of 633–773 K. Measured surface tensions of the six quaternary bromides are shown in Figs. 13–18 respectively, and Table 3 lists the fitting correlations of experiment data. Fig. 13 shows the surface tension of sample NO.1 which ranges linearly from 0.1240 to 0.1152 N m 1 with the rise of temperature
Fig. 13. Measured surface tension values and fitting curve for quaternary bromides NO.1.
from 703 K to 763 K. Fig, and there is a good agreement between the measured surface tension and the fitting curve for sample NO.1. For sample NO.2, Fig. 14 indicates that the surface tension falls within 0.11728–0.11450 N m 1 from 673 K to 758 K, while the surface tension diminishes linearly with the increases of salt temperature as with sample NO.1. From Figs. 15 and 16 we know The surface tension of sample NO.3 is within 0.11753–0.11260 N m 1 in 648–733 K while for sample NO.4 The surface tension ranges linearly within 0.11803– 0.11375 N m 1 in 633–722 K. The tendency for surface tensions to decreases linearly with the increase of salt temperature is also observed for sample NO.3 and sample NO.4. For Figs. 17 and 18 there are similar trends for the changing of surface tensions with the increase of salt temperature for sample NO.5 and sample NO.6. The surface tension of sample NO.5 is within 0.11551–0.11273 N m 1 in 663–737 K while for sample NO.4 the surface tension ranges linearly within 0.11812– 0.11249 N m 1 in 695–773 K.
714
Y. Xiong et al. / Solar Energy Materials & Solar Cells 157 (2016) 709–715
Fig. 14. Measured surface tension values and fitting curve for quaternary bromides NO.2.
Fig. 15. Measured surface tension values and fitting curve for quaternary bromides NO.3.
Fig. 17. Measured surface tension values and fitting curve for quaternary bromides NO.5.
Fig. 18. Measured surface tension values and fitting curve for quaternary bromides NO.6. Table 3 Correlated formulas of surface tension of quaternary bromides. No.
1 2 3 4 5 6
Fig. 16. Measured surface tension values and fitting curve for quaternary bromides NO.4.
s¼ a b 10 5T a
b
0.2276 0.1397 0.1550 0.1485 0.1406 0.1681
14.71 3.3232 5.57888 4.7981 3.7755 7.2009
R2
Temperature range/K
0.9912 0.996 0.999 0.992 0.999 0.999
703 763 673 758 678 733 633 722 663 737 695 773
Detailed surface tension curves fitted in Figs. 13–17 for sample NO.1 to NO. 6 are listed in Table 3. Based on Figs. 13–18 a conclusion can be achieved that all the fitting curves of surface tension of the six quaternary bromides have similar trends that surface tension decreases linearly with the increase of temperature, which is consistent with the results of literatures on other liquids [7]. Moreover, it shows that the fitting curve of surface tension of sample NO.1 has a maximum slope and
Y. Xiong et al. / Solar Energy Materials & Solar Cells 157 (2016) 709–715
a maximum surface tension at the same temperature, which is the best working fluid among the six quaternary bromides for heat pipes from the view of surface tension, which benefits to the heat transfer of molten salts, among all the six quaternary bromides. Surface tension of sample NO.6 changes slower than that of sample NO.1 but faster than NO.3, while Surface tension of sample NO.3 changes slower than that of sample NO.6 but faster than NO.4. Surface tension of sample NO.5 changes slower than that of sample NO.4 but faster than NO.2. From Table 1 and Figs. 13–18 we also find that the composition of quaternary bromides has significant effects on surface tension: the slopes of the fitting curves of the samples first increase and then decrease as the mass fractions of NaBr and KBr increase and those of LiBr and CaBr2 decrease. The less the surface tension is, the weaker the heat transfer ability is.
4. Conclusions Mixed molten salts are considered as promising medium for both heat transfer and thermal storage in solar thermal power plants. In the present work six novel quaternary bromides are prepared by mixing NaBr, KBr, LiBr and CaBr in different proportions. And then the improved Archimedean principle and the PullOff method were employed to measure the density and surface tension of six quaternary bromides respectively, and the thermal expansion coefficient for each quaternary bromide were calculated and fitted. For density, there is a good linearity between density and temperature for each quaternary bromide. Sample NO.2 has maximum density, which is greater than that of sample NO.3; Density of sample NO.3 is greater than that of sample NO.1, density of sample NO.1 is greater than that of sample NO.4, and density of sample NO.4 is greater than that of sample NO.5, while sample NO.6 has the minimum density. For heat transfer of heat pipes sample NO.2 should be the best choice, and then is sample NO.3 and sample NO.1. Thermal expansion coefficients of the six quaternary bromides increase linearly approximately with the rise of temperature. But there are evident difference in the values and fitted curves. Sample NO.6 has the maximum thermal expansion coefficient while sample NO.3 has the minimum thermal expansion coefficient at a same temperature. The sample NO.3 would reduces the reserved expansion volume in a TES tank for either sensible thermal storage. For surface tension, the six quaternary bromides have similar trends that the surface tension descends linearly with the rise of temperature. Fitting curve of surface tension of sample NO.1 has a maximum slope and the maximum surface tension at the same temperature, which benefits to the heat transfer of molten salts, among all the six quaternary bromides. Surface tension of sample NO.6 changes slower than that of sample NO.6 but faster than NO.1, while Surface tension of sample NO.3 changes slower than that of sample NO.6 but faster than NO.4 and Surface tension of sample NO.5 changes slower than that of sample NO.4 but faster than NO.2. The composition of quaternary bromides also has significant effects on surface tension: the slopes of the fitting curves of the samples first increase and then decrease as the mass fractions of NaBr and KBr increase and those of LiBr and CaBr2 decrease.
715
Funding sources This work was supported by the National Nature Science Foundation of China under Project [grant numbers 51206004]; and the Key Project of Beijing Natural Science Foundation under Project [grant number 3151001].
References [1] S. Almsater, W. Saman, F. Bruno, Performance enhancement of high temperature latent heat thermal storage systems using heat pipes with and without fins for concentrating solar thermal power plants, Renew. Energy 89 (2016) 36–50, http://dx.doi.org/10.1016/j.renene.2015.11.068. [2] S. Tiari, S. Qiu, M. Mahdavi, Discharging process of a finned heat pipe–assisted thermal energy storage system with high temperature phase change material, Energy Convers. Manag. 118 (2016) 426–437, http://dx.doi.org/10.1016/j. enconman.2016.04.025. [3] Z. Liao, A. Faghri, Thermal analysis of a heat pipe solar central receiver for concentrated solar power tower, Appl. Therm. Eng. 102 (2016) 952–960, http: //dx.doi.org/10.1016/j.applthermaleng.2016.04.043. [4] Y. Cao, Separate-type heat pipe solar receivers for concentrating solar power, Front. Heat. Pipes (2015), http://dx.doi.org/10.5098/fhp.6.1. [5] L.X. Sang, M. Cai, N. Ren, Y.T. Wu, C. Burda, Improving the thermal properties of ternary carbonates for concentrating solar power through simple chemical modifications by adding sodium hydroxide and nitrate, Sol. Energy Mater. Sol. Cells 124 (2014) 61–66, http://dx.doi.org/10.1016/j.solmat.2014.01.025. [6] G.J. Janz, U. Krebs, H.F. Siegenthaler, R.P.T. Tomkins, Molten salts: volume 3 nitrates, nitrites, and mixtures: electrical conductance, density, viscosity, and surface tension data, J. Phys. Chem. Ref. Data 1 (1972) 581–672, http://dx.doi. org/10.1063/1.3253103. [7] H. Sasaki, E. Tokizaki, K. Terashima, S. Kimura, Density measurement of molten silicon by an improved Archimedian method, J. Cryst. Growth 139 (1994) 225–230, http://dx.doi.org/10.1016/0022-0248(94)90170-8. [8] Q. Peng, J. Ding, X.L. Wei, J.P. Yang, X.P. Yang, The preparation and properties of multi-component molten salts, Appl. Energy 87 (2010) 2812–2817, http://dx. doi.org/10.1016/j.apenergy.2009.06.022. [9] J.G. Cordaro, N.C. Rubin, R.W. Bradshaw, Multicomponent molten salt mixtures based on nitrate/nitrite anions, J. Sol. Energy Eng. 133 (2011) 011–014, http: //dx.doi.org/10.1115/1.4003418. [10] J.W. Raade, D. Padowitz, Development of molten salt heat transfer fluid with low melting point and high thermal stability, J. Sol. Energy Eng. 133 (2011) 1–6, http://dx.doi.org/10.1115/1.4004243. [11] T. Bauer, N. Pfleger, N. Breidenbach, M. Eck, D. Laing, et al., Material aspects of solar salt for sensible heat storage, Appl. Energy 111 (2013) 1114–1119, http: //dx.doi.org/10.1016/j.apenergy.2013.04.072. [12] T. Wang, D. Mantha, R.G. Reddy, Novel low melting point quaternary eutectic system for solar thermal energy storage, Appl. Energy 102 (2013) 1422–1429, http://dx.doi.org/10.1016/j.apenergy.2012.09.001. [13] Y. Marcus, Surface tension and cohesive energy density of molten salts, Thermochim. Acta 571 (2013) 77–81, http://dx.doi.org/10.1016/j. tca.2013.08.018. [14] C.J. Li, P. Li, K. Wang, E.E. Molina, Survey of properties of key single and mixture halide salts for potential application as high temperature heat transfer fluids for concentrated solar thermal power systems, AIMS Energy 2 (2014) 133–157, http://dx.doi.org/10.3934/energy.2014.2.133. [15] N. Ren, Y.T. Wu, C.F. Ma, L.X. Sang, Preparation and thermal properties of quaternary mixed nitrate with low melting point, Sol. Energy Mater. Sol. Cells 127 (2014) 6–13, http://dx.doi.org/10.1016/j.solmat.2014.03.056. [16] F. Aqra, Surface tension of molten metal halide salts, J. Mol. Liq. 200 (2014) 120–121, http://dx.doi.org/10.1016/j.molliq.2014.09.052. [17] L.X. Sang, M. Cai, Y.B. Zhao, N. Ren, Y.T. Wu, et al., Mixed metal carbonates/ hydroxides for concentrating solar power analyzed with DSC and XRD, Sol. Energy Mater. Sol. Cells 140 (2015) 167–173, http://dx.doi.org/10.1016/j. solmat.2015.04.006. [18] X.H. An, J.H. Cheng, P. Zhang, Z.F. Tang, J.Q. Wang, Determination and evaluation of the thermophysical properties of an alkali carbonate eutectic molten salt, Faraday Discuss. (2016), http://dx.doi.org/10.1039/C5FD00236B. [19] G.J. Janz, R.P.T. Tomkinns, C.B. Allen, J.R. Downey, S.K. Singer, Molten salts: volume 4, part 3, bromides and mixtures, iodides and mixtures - electrical conductance, density, viscosity, and surface tension data, J. Phys. Chem. Ref. Data 6 (1977) 411–587, http://dx.doi.org/10.1063/1.555552. [20] H. Sasaki, E. Tokizaki, K. Terashima, S. Kimura, Density measurement of molten silicon by an improved Archimedean method, Journal of Crystal Growth, 139, 1994, pp. 225–230, 10.1016/0022–0248(94)90170-8. 10.1016/0022–0248(94) 90170-90178.