- Email: [email protected]
Thermophysical properties of Ca(NO3)2-NaNO3-KNO3 mixtures for heat transfer and thermal storage
Solar Energy 146 (2017) 172–179
Contents lists available at ScienceDirect
Solar Energy journal homepage: www.elsevier.com/locate/solener
Thermophysical properties of Ca(NO3)2-NaNO3-KNO3 mixtures for heat transfer and thermal storage Y.Y. Chen, C.Y. Zhao ⇑ Institute of Engineering Thermophysics, Shanghai Jiao Tong University, Shanghai 200240, China
a r t i c l e
i n f o
Article history: Received 23 July 2016 Received in revised form 18 January 2017 Accepted 20 February 2017
Keywords: Ternary molten salt Thermophysical property Phase change material Heat transfer fluid Thermal storage
a b s t r a c t In this study calcium nitrate, sodium nitrate, and potassium nitrate were mixed to form cheap ternary molten salts based on different weight ratios. These molten salts can be used as both sensible heat storage materials and latent heat storage materials. In addition, they can be directly used as heat transfer fluids due to their low freezing temperatures. The results indicated that the mixture (Ca(NO3)2:NaNO3:KNO3 = 32:24:44 wt%) had the best performance for latent heat storage with its enthalpy of 67 J/g and melting point of about 80 °C. The specific heat capacity (1.7 J/(g °C) for the solid phase and 1.2 J/(g °C) for the liquid phase), viscosity (next to zero at 200 °C), thermal conductivity (about 1–3 W/(m K)), thermal decomposition, and cycle stability of the molten salts were measured by DSC, Malvern Kinexus Ultra+, a transient plane thermal conductivity meter, STA, and XRD, respectively. The thermophysical properties including the low manufacturing cost showed that the molten salts have great potential applications in thermal storage systems. Ó 2017 Elsevier Ltd. All rights reserved.
1. Introduction Because of the disparity between energy supply and demand, thermal storage plays a crucial role in renewable energy and industrial waste heat recovery. Heat storage technology and heat transfer fluids are very important components in thermal storage. Heat storage technology, can be classified as latent heat storage, sensible heat storage, thermochemical heat storage, and adsorptive heat storage (Sharma et al., 2009; Zhao and Tian, 2013). Among the four kinds of technology, latent heat storage has received much attention in recent decades because it has merits such as large heat capacity, and no significant temperature variation in the process of phase transition (Zuo et al., 2005). Heat transfer fluid (HTF), is an important component of thermal-energy storage (TES) systems. The main energy storage system is based on a sensible energy storage system, which is equipped with two tanks: a ‘‘cold” tank to collect solar heat and a ‘‘hot” tank to store thermal energy (Kearney et al., 2002; Kearney, 2004). If the storage system in these tanks is direct, the heat transfer fluid can also be used as a storage medium (Javier Ruiz-Cabarias et al., 2017). For both latent heat storage and heat transfer technology, the materials have a predominant effect on their use. In the case of latent heat storage, phase-change material selection is very crucial (Yan et al., 2015) and preparation and characterization of ⇑ Corresponding author. E-mail address: [email protected] (C.Y. Zhao). http://dx.doi.org/10.1016/j.solener.2017.02.033 0038-092X/Ó 2017 Elsevier Ltd. All rights reserved.
phase-change materials have been investigated by many researchers. Paul et al. (2015)studied a eutectic mixture of galactitol and mannitol. When the molar ratio of galactitol and mannitol is 3:7, they have a melting point of 153 °C and a high latent heat of fusion (292 J/g). Warzoha et al. (2015) implemented random HGNF(herringbone style graphite nanofibers) into paraffin (IGI 1230A) and quantified the effect of HGNF on the paraffin in both solid and liquid phases. Lee et al. (2014) investigated the thermal characterization of EG (expanded graphite) and the composition of EG and erythritol. They found that the latent heat and thermal conductivity improved with an increase of the EG interlayer distance under the assistance of the thermal equilibrium technique. Zhao et al. (2015) studied binary nitrate salt mixtures consisting of NaNO3 and Ca(NO3)2 with different molar ratios. They found that when the molar ratio is 3:7 (Ca(NO3)2:NaNO3), the sample has the best heat storage performance and this ratio is considered as the best eutectic composition. Roget et al. (1980) investigated a binary molten salt (LiNO3-NaNO3) and a ternary molten salt (LiNO3-KNO3NaNO3). They measured their physical properties such as thermal cycle stability, corrosion, and enthalpy. Judith and Calvet (2013) investigated a molten salt mixture which combined KNO3, NaNO3, and Ca(NO3)2 and the best composition was considered to be 36 wt % Ca(NO3)2, 16 wt% NaNO3, and 48 wt% KNO3. In t most of the commercial solar thermal power plants, synthetic oil is used as the heat transfer fluid. But the high price of synthetic oil poses a huge limitation on its utilization. In order to remedy this, recent heat transfer fluids use inorganic molten salts. From the 1980s, a molten
Y.Y. Chen, C.Y. Zhao / Solar Energy 146 (2017) 172–179
nitrate salt mixture which is 60% NaNO3 and 40% KNO3(Solar Salt) has been used as an attractive candidate for heat transfer fluid in the Concentrating Solar Power (CSP) pilot plant (Kearney, 2004; Mar and Kramer, 1981; Herrmann et al., 2004) because of its great thermal stability, the lowest cost, and highest melting point. The second commercial salt is Hitec Salt, which is a mixture of 53 wt % KNO3, 40 wt% NaNO2, and 7 wt% NaNO3 with a melting point of 142 °C (Gil, 2010). The third salt mixture is comprised of 28 wt% NaNO3, 52 wt% KNO3, and 20 wt% LiNO3, and this mixture’s thermal property was studied by Olivares and Edwards (2013) and Fernandez et al. (2014). The working temperature range of this mixture is 130–600 °C, and its viscosity is 0.03 Pa S at 300 °C (Olivares et al., 2012), which is good for fluid flow properties. But the high price of LiNO3 ($4.32/kg) poses a huge limitation on its development. The fourth salt mixture is Carbonate salt (Vignarooban et al., 2015) with 32.1 wt% Li2CO3, 33.4 wt% Na2CO3, and 34.5 wt% K2CO3, which was proposed to replace the nitrate salt to increase the thermal stability temperature limit. The melting point of this mixture is 400 °C and the decomposition temperature is 800–850 °C. Among the mentioned materials which are used for both latent heat storage and heat transfer fluid, nitrate salts have wide use in actual practice due to their merits such as high heat storage capacity (enthalpy), favorable thermal stability, satisfactory compatibility with containers, low corrosion (Javier Ruiz-Cabarias et al., 2017), low vapor pressure, low viscosity in most cases, and chemical inertness (Zhao and Wu, 2011). Reviewing relevant references (Paul et al., 2015; Warzoha et al., 2015; Lee et al., 2014; Zhao et al., 2015; Roget et al., 1980; Judith and Calvet, 2013; Mar and Kramer, 1981; Herrmann et al., 2004; Gil, 2010; Olivares and Edwards, 2013; Fernandez et al., 2014; Olivares et al., 2012; Vignarooban et al., 2015; Zhao and Wu, 2011; Li et al., 2016; Fernandez and Perez, 2016; Prieto et al., 2016; Brandon and Davidson, 2017; Myers and Yogi Goswami, 2016; Zhou and Eames, 2016), Ca (NO3)2, NaNO3, and KNO3 are eligible as both phase change materials and heat transfer fluid due to their insignificant corrosion, no toxicity, and low price. Although several reports mentioned Ca (NO3)2, NaNO3 or KNO3 as an additive in the mixtures, the study remains incomplete. For example, Judith and Calvet (2013) used Ca(NO3)2-NaNO3-KNO3 as phase-change materials, but the data was incomplete (only 6 groups) and the thermal conductivity was not presented in his study. Roget et al. (1980) prepared a ternary molten salt (LiNO3-KNO3-NaNO3), but the price of LiNO3 is much higher than that of Ca(NO3)2, which is not economical in solar plants. Moreover, Zhao et al. (2015) used Ca(NO3)2 as an additive in NaNO3, but its’ melting point (220 °C) isn’t suitable for low or medium temperatures. The present study prepared a ternary molten salt consisting of KNO3, NaNO3, and Ca(NO3)2 based on different weight ratios. Experiments showed that this mixture has a melting point of 80 °C and the maximum stable temperature is 600 °C. However, this mixture focuses mainly on low and medium temperatures (less than 200 °C). Hence, other thermophysical properties such as thermal conductivity and viscosity were measured in the temperature range of 50–200 °C. Further study indicated a correlation between the heating rate and heat enthalpy, and the economic impact id discussed at the end of this study. Overall, the experimental results showed that these ternary molten salts are suitable candidates as thermal storage materials and heat transfer fluids.
173
by Sinopharm Chemical Reagent Co. Ltd. The method of preparing the formulations was to mix the anhydrous form of the single components at room temperature. In order to produce the anhydrous form of calcium nitrate, the tetrahydrate form was heated in two stages: (i) at 150 °C for 10 h and (ii) at 350 °C for 20 h. Then the anhydrous salts were put in a muffle furnace at 150 °C to ensure drying. In order to avoid water absorption, especially for the extremely anhydrous calcium nitrate, the handling and mixing of the salts, as well as putting samples in the testing pan, were performed in a box under dry nitrogen. To weigh the samples based on 32 sets of data, an analytical balance with a resolution of 0.1 mg was used. 2.2. Simultaneous Thermal Analyzer Fusion heat, cycle stability, and the endothermic peak’s property under different heat rates were measured by a Simultaneous Thermal Analyzer (STA 8000, Perkin Elmer) which provides realtime analysis of sample weight change and heat flux. The weight and temperature accuracy of the STA were 0.0001 mg and 0.01 °C, respectively. The temperature accuracy of the oven was 0.1 °C. To measure the fusion heat, the 32 groups of samples were heated from 50 °C to 250 °C at a rate of 10 °C/min under N2 purging of 20 L/min. The samples were heated 33 times in three stages for the cycle stability test: (i) from 50 °C to 250 °C at a rate of 10 °C/ min, (ii) at 250 °C for 1 min, and (iii) from 250 °C to 50 °C at the rate of 10 °C/min. In order to determine the endothermic peak’s property at different heating rates, the sample was measured at different heating rates of 5 °C/min, 10 °C/min, 15 °C/min, and 20 °C/min. 2.3. Differential scanning calorimetry Differential scanning calorimetry (DSC 8000, Perkin Elmer) was used to determine the specific heat capacity of the mixtures. The temperature calibration was performed with benzophenone and caffeine. The heat flow calibration was carried out by indium. After the aforementioned calibration, the samples were heated from 50 °C to 80 °C and from 200 °C to 250 °C, respectively, in order to measure the samples’ solid and liquid specific heat capacity. 2.4. X-ray diffraction X-ray diffraction (XRD) measurements were carried out to investigate the remnants after thermal decomposition. The sample was heated by STA from 200 °C to 950 °C at a rate of 10 °C/min, and then, XRD was used to investigate the remnants at ambient temperature. 2.5. Transient plate method The transient plate method was performed to measure the sample’s thermal conductivity. This methods measure the transient thermal conductivity and the instrument’s uncertainty was about 10%. 2.6. Malvern Kinexus Ultra+
2. Material preparation and experimental methods 2.1. Material preparation The molten salt was prepared from Ca(NO3)24H2O(purity > 99%), NaNO3(purity > 99%), and KNO3(purity > 99%) manufactured
Malvern Kinexus Ultra+ was used to test the sample’s viscosity and the apparatus’s uncertainty was about 10%. In order to avoid the water absorption, the experiments were performed under a flow of either dry nitrogen or dry synthetic air. The viscosity was measured with a 40 mm diameter stainless steel parallel plate.
174
Y.Y. Chen, C.Y. Zhao / Solar Energy 146 (2017) 172–179
3. Results and discussion 3.1. Fusion heat Fusion heat and melting point are important parameters for molten salts. Some nitrate salts’ enthalpy and melting points are shown in Fig. 1 and Table 1. Of 32 test groups, only 3 groups showed a significant fusion heat. The STA thermograms obtained for these three Ca(NO3)2-NaNO3-KNO3 mixtures are shown in Fig. 2 and Table 2. In the figure, the observed enthalpy changed with the weight ratios of Ca(NO3)2, NaNO3, and KNO3 and this phenomenon can be explained by the mismatch between the fixed and experimental weight ratios. Specifically, when the three nitrate salts are mixed according to the weight proportions, they can only form a eutectic salt at a fixed proportion which is not known in advance. If the weight proportion is not the right value, there will be surplus Ca(NO3)2, NaNO3, KNO3, or a combination of them. When the mixtures are heated, the eutectic composition melts first, and then the surplus salts melt (Zhao et al., 2015). The different data will result in various masses of the ternary eutectic composition, and the heat of fusion will subsequently be different (as shown in Table 2). In Table 2, the Tmi1, Tmi2, and Tmi3 (melting point) are similar since the melting point only depends on the materials of the mixture. Furthermore, Fig. 2 and Table 2 also indicate an invariant temperature value (120 °C), which is the peak temperature. Tg in Fig. 2 and Table 2 represents the temperature of glass transition and its value is about 50 °C. In Fig. 2, there is a small observed enthalpy between Tg and Tm. This phenomenon can be attributed to the crystallization reaction which is prior to the melting of the mixture. This subtle energy is used to transform the state from a metastable to stable crystal structure. After this process, this ternary salt begins to quickly melt. Actually, when heating the materials with a vitreous state, they always require a part of the energy (subtle energy as shown in Fig. 2) to transform their state from a metastable to thermodynamically stable crystal structure (Judith and Calvet, 2013). 3.2. Cycle stability Among the three groups of samples, the first (Ca:Na: K = 34:20:46 wt%) and third samples (Ca:Na:K = 32:24:44 wt%) had the highest fusion heat, which was 51 J/g and 67 J/g,
Table 1 Enthalpy of the three nitrate salts. Material
Enthalpy
Ca(NO3)2 NaNO3 KNO3
145 J/g 177 J/g 88 J/g
Fig. 2. STA curves of the mixing salts.
respectively. Hence, these two samples were used to test the cycle stability. The endothermic and exothermic curves obtained from STA are shown in Fig. 3(a), (b) and (c) and Fig. 4(a) and (b), respectively. Table 3 gives the enthalpy of each peak. As shown in Fig. 3(a), (b) and (c), endothermic peak 2 of the first sample and endothermic peak 3 of the third sample had the great reproducibility. Additionally, exothermic peak 1 of the first sample and exothermic peaks 1 and 2 of the third sample also had favorable reproducibility during the test. The weight loss of the first and third samples was 0.2% and 0.25%, respectively. The above properties of the two samples indicate that they have satisfactory chemical stability ranging from 50 °C to 250 °C. For the third sample, its initial point of the exothermic peak was Tsi = 180 °C and the final point was Tsf = 100 °C. Its enthalpy of the exothermic peak (peak 1 and 2) was 11 J/g which accounts for approximately 15% of the enthalpy of the endothermic peak. For the first sample, the initial point of its exothermic peak was Tsi = 160 °C and the final point was Tsf = 100 °C. The enthalpy of the exothermic peak (peak 1) was 2 J/g which accounts for 4% of the enthalpy of the endothermic peak. The possible reason for the difference between the enthalpy of the exothermic peak and that of the endothermic peak in the two samples is the rate of nucleating. In the cooling process, the rate of the cooling is faster than the rate of nucleating. When the cooling process ends, the two samples will not be thoroughly nucleated, which will result in the smaller enthalpy of the exothermic peak compared to that of endothermic peak. Therefore, the appropriate rate of cooling is very critical. 3.3. Specific heat capacity
Fig. 1. Ternary phase diagram of Ca(NO3)2-NaNO3-KNO3 mixtures.
Although latent heat is used to store energy in latent heat storage, latent and sensible heat storage are closely related. On one hand, before the materials reach the temperature of phase change, they use the sensible heat to store energy. On the other hand, due to the extremely low thermal conductivity of the phase change materials, the temperature difference in the inner area of the materials is huge, which will lead to the fact that when some parts begin
175
Y.Y. Chen, C.Y. Zhao / Solar Energy 146 (2017) 172–179 Table 2 Samples’ fusion heat and glass transition temperature. Number
Initial point
Ca:Na:K = 32:20:48 (wt%) Ca:Na:K = 32:24:44 (wt%) Ca:Na:K = 34:20:46 (wt%)
Tmi1 = 80 °C Tmi2 = 84 °C Tmi3 = 86 °C
Final point Tmf1 = 180 °C Tmf2 = 190 °C Tmf3 = 200 °C
Heat of fusion 45 J/g 67 J/g 51 J/g
Density 3
1455 kg/m 1460 kg/m3 1482 kg/m3
Heat of fusion
Temperature of the peak
Temperature of glass transition
65 kJ/L 98 kJ/L 76 kJ/L
Tpeak,1 = 120 °C Tpeak,2 = 120 °C Tpeak,3 = 120 °C
Tg,1 = 50 °C Tg,2 = 50 °C Tg,3 = 50 °C
(a) First sample Ca:Na:K=34:20:46 (wt%)
(a) First sample Ca:Na:K=34:20:46 (wt%)
(b) Third sample Ca:Na:K=32:24:44 (wt%)
(b) Third sample Ca:Na:K=32:24:44 (wt%)
Fig. 4. Samples’ cycle curves (the X axis is time).
Table 3 Enthalpy of each peak. Sample
Enthaply of Peak, 1
Enthaply of Peak, 2
Enthaply of Peak, 3
Ca:Na:K = 34:20:46 (wt%) Ca:Na:K = 32:24:44 (wt%)
2 J/g 1.4 J/g
51 J/g 9.6 J/g
0 67 J/g
phase transformation, the others still have not reached the temperature of transition. Hence, in actual applications, the specific heat is very crucial. The following expression (Li et al., 2016) shows the heat storage capacity of phase change systems.
(c) Comparison between the two samples Fig. 3. Samples’ cycle curves (the X axis is temperature).
Z Q¼
Tm
Ti
Z mCp dT þ mam Dhm þ
Tf
mCp dT Tm
ð1Þ
176
Y.Y. Chen, C.Y. Zhao / Solar Energy 146 (2017) 172–179
Based on DSC, the specific heat capacity of the third sample (because it has the highest fusion heat) was measured and the result is shown in Fig. 5(a) and (b). Based on these curves, the equation of the specific heat capacity was calculated and can be expressed as follows: For the solid phase, the curve can be expressed by a quadratic function as:
Csp ¼ A1 þ B1 X þ C1 X2
ðA1 ¼ 3:70785; B1 ¼ 0:06525;
5
C1 ¼ 5:10854e Þ
ð2Þ
For the liquid phase, the curve can be expressed by a quadratic function as:
ClP ¼ C1 þ D1 X þ E1 X2
ðC1 ¼ 0:26884; D1 ¼ 0:01016;
E1 ¼ 1:6466e5 Þ
ð3Þ
In Fig. 5(a), the curve quickly rises when the temperature exceeds 68 °C, indicating that the heat capacity rises with an increase in temperature, while in the range of 50–68 °C, the curve is almost stable regardless of the variation of temperature. The possible reason is that in this temperature range, the sample is metastable vitreous. In Fig. 5(b), the liquid curve quickly rises with an increase in temperature, which indicates that the specific heat capacity of the liquid phase is very sensitive to temperature variation.
Fig. 6. Thermal conductivity of the third sample (Ca:Na:K = 32:24:44 wt%).
3.4. Thermal conductivity Thermal conductivity was determined by the transient plate method and the result is shown in Fig. 6 and Table 4. Based on Fig. 6 and Table 4, the thermal conductivity rises when the temperature reaches 120 °C, 150 °C, and 180 °C. This difference can be attributed to the state change. In other words, when the temperature is 120 °C and 150 °C, the sample is molten and with an increase of temperature, more parts of the sample transform from the molten state to the liquid state, so the thermal conductivity rises with an increase of temperature. When the temperature reaches 180 °C, the entire sample melts and all of its parts have become liquid, so the thermal conductivity is higher than that at 120 °C and 150 °C. 3.5. Endothermic peak
(a) Curve of solid phase (50-80
The endothermic peak’s property at different heating rates was investigated by STA and the result is presented in Fig. 7(a) and (b). As shown in Fig. 7(a) and (b), the endothermic peak becomes wider with an increase of heating rate (Roget et al., 1980). Since the endothermic peak becomes wider with an increase of heating rate, it will be insignificant or even disappear in the experiments. This phenomenon can be attributed to the mismatch between the heating rate and the materials’ phase transition speed. In other words, the materials need some time to transform their phases, and a faster heating rate will result in inadequate melting of the materials. Table 5 shows the temperature (Tmi represents the initial temperature, Tmf reflects the final temperature). 3.6. Thermal decomposition Thermal decomposition was investigated by STA and XRD. The result is presented in Figs. 8 and 9(a) and (b). In Fig. 8, the sample decomposed at 650 °C and 750 °C, respectively. Below 600 °C, there
Table 4 Thermal conductivity K = 32:24:44 (wt%).
(b) Curve of liquid phase (200-250 Fig. 5. Specific heat capacity curves of the third sample (Ca:Na:K = 32:24:44 wt%).
of
the
third
sample
Ca:Na:
Temperature (°C)
Thermal conductivity (W/m K)
120 150 180
1.090 1.929 3.194
177
Y.Y. Chen, C.Y. Zhao / Solar Energy 146 (2017) 172–179
Fig. 8. Decomposition of the third sample Ca:Na:K = 32:24:44 (wt%).
(a) First sample Ca:Na:K=34:20:46 (wt%)
(a) Decomposing at 650 (b) Third sample Ca:Na:K=32:24:44 (wt%) Fig. 7. Samples’ endothermic peak performance at different heating rates.
Table 5 Samples’ melting temperature range at different heating rates.
Tmi,1-Tmf,1 Tmi,2-Tmf,2 Tmi,3-Tmf,3 Tmi,4-Tmf,4
Ca:Na:K = 34:20:46 (wt%)
Ca:Na:K = 32:24:44 (wt%)
75–170 °C 80–180 °C 95–210 °C 105–235 °C
75–175 °C 85–195 °C 100–220 °C 105–235 °C
was no decomposition. This phenomenon shows that the mixture is very stable below 600 °C. In Fig. 9(a) and (b), when the temperature ranges from 650 °C to 750 °C, the first decomposition happens, with the main products being KNO3, CaO, and NaNO3. When the temperature exceeds 750 °C, the main remnants are Ca (OH)2, KNO3, CaO, and NaNO3. This indicates that Ca(NO3)2 decomposes to CaO and Ca(OH)2, but KNO3 and NaNO3 are still stable. 3.7. Viscosity The viscosity of the first sample (Ca:Na:K = 34:20:46 wt%) and third sample (Ca:Na:K = 32:24:44 wt%) are shown in Fig. 10 (a) and (b). In the figure, the viscosity of the two samples decreased with an increase in temperature, and when the temperature
(b) Decomposing at 750 Fig. 9. XRD of remnants.
ranged from 115 to 200 °C, the viscosity was very much lower than the viscosity in the temperature range of 100–115 °C. This phenomenon can be attributed to the state change. In other words, when the temperature is lower than 115 °C, the state of the two samples is molten, then with an increase in temperature, the state transforms from molten to liquid, so the viscosity reduces to a very
178
Y.Y. Chen, C.Y. Zhao / Solar Energy 146 (2017) 172–179 Table 6 Recent price of nitrates. Nitrate
Price ($/Metric Ton)
Ca(NO3)24H2O KNO3 NaNO3 LiNO3
300 930 330 7980
4. Economic analysis
(a) Viscosity of the molten salts mixtures 100-200
Since the price of the materials will have a direct influence on their utilization, they must have the lowest price to decrease the capital investment as much as possible. The prices of some nitrates are presented in Table 6. Based on Table 6, the price of the third sample (Ca:Na:K = 32:24:44 wt%) is $626 per Metric Ton, and the price of the first sample(Ca:Na:K = 34:20:46 wt%) is $640 per Metric Ton. Therefore, the third sample should be chosen as both the phase-change material and heat transfer fluid. 5. Conclusions
(b) Viscosity of the molten salts mixtures (115-200 ) Fig. 10. Viscosity of the molten salts mixtures.
small value which is almost zero. The viscosity of the two samples in the temperature range of 115–200 °C is shown in Fig. 10(b). In this figure, the viscosity of the third sample (Ca:Na: K = 32:24:44 wt%) was a little greater than that of the eleventh sample (Ca:Na:K = 34:20:46 wt%) in the temperature range of 115–200 °C, but the difference was very small. Hence the viscosity of the two samples can be considered similar. However, when the temperature ranges from 100 to 115 °C (as shown in Fig. 10(a)), the two samples are molten and the viscosity of the first sample was very much greater than that of the third sample. This difference can be explained by an increase in the amount of calcium nitrate which is contained in the molten salts mixture. Molten salts containing calcium nitrate tends to form a glassy state, and this can increase the viscosity (Judith and Calvet, 2013; Vignarooban et al., 2015). Hence, the viscosity of the first sample was greater than that of the third sample when both the samples are molten because the first sample contains more calcium nitrate than the third sample. However, when the temperature rises, the samples change their state from molten to liquid, and the amount of calcium nitrate contained in the molten salts mixtures may have little effect on the viscosity(as shown in Fig. 10(b)) and other elements in the molten salts mixtures may have more influence. This needs further investigation. Overall, both the third and first sample can be expected to be chosen as the heat transfer fluids at a normal operating temperature for a thermal solar plant.
This study made ternary salts by using Ca(NO3)2, NaNO3, and KNO3, which can be used as both phase-change materials and heat transfer fluids. Among the ternary salts, the mixture of Ca(NO3)2, NaNO3, and KNO3 with a weight ratio of 32:24:44 (wt%) is considered to be the optimum proportion for its best cycle stability, specific heat capacity, viscosity, thermal conductivity, and economy cost. Its melting point is 80 °C and fusion heat is 67 J/g. The STA test showed that the thermal cycle stability is great in the temperature range of 50–250 °C and the XRD test showed that the upper limit temperature is 600 °C. The thermal conductivity of the optimal mixture (Ca:Na:K = 32:24:44 wt%) is about 1–3 W/ (m K) and the viscosity is very small at the normal operating temperature. Furthermore, the cost of the optimal mixture (Ca:Na: K = 32:24:44 wt%) is about $626 per Metric Ton, which is acceptable in the industry. In summary, the mixture (Ca:Na: K = 32:24:44 wt%) can be applied as both a phase-change material and heat transfer fluid in solar plants due to its significant fusion heat, low viscosity, cheap price, and excellent thermal stability. Acknowledgement This work is supported by the National Key Basic Research Program of China (973 Project: 2013CB228303). References Brandon, J., Davidson, Jane H., 2017. Demonstration of prototype molten salt solar gasification reactor. Sol. Energy 145, 224–230. Fernandez, A.G., Perez, F.J., 2016. Improvement of the corrosion properties in ternary molten nitrate salts for direct energy storage in CSP plants. Sol. Energy 134, 468–478. Fernandez, A.G., Ushak, S., Galleguillos, H., Perez, F.J., 2014. Development of new molten salts with LiNO3 and Ca(NO3)2 for energy storage in CSP plants. Appl. Energy 119, 131–140. Gil, A., 2010. State of the art on high temperature thermal energy storage for power generation. Part 1-concepets, materials and modelization. Renew. Sustain. Energy Rev. 14 (1), 31–35. Herrmann, U., Kelly, B., Price, H., 2004. Two-tank molten salt storage for parabolic trough solar power plants. Energy 29 (5–6), 883–893. Ruiz-Cabarias, F. Javier, Jove, Aleix, Prieto, Cristina, et al., 2017. Materials selection of steam-phase change material (PCM) heat exchange for thermal energy storage systems in direct steam generation facilities. Sol. Energy Mater. Sol. Cells 159, 526–535. Judith, C., Calvet, N., 2013. Ca(NO3)2-NaNO3-KNO3 Molten salt mixtures for direct thermal energy storage systems in parabolic trough plants. J. Sol. Energy Eng. ASME 5. 021016-1-7. Kearney, D., 2004. Engineering aspects of a molten salt heat transfer fluid in a trough solar field. Energy 29 (5–6), 861–870. Kearney, D., Kelly, B., Mahoney, R., 2002. Assessment of a molten heat transfer fluid in a parabolic trough solar field. Sol. Energy Eng. 125 (2), 170–176.
Y.Y. Chen, C.Y. Zhao / Solar Energy 146 (2017) 172–179 Lee, S.Y., Shin, H.K., Park, M., Rhee, K.Y., Park, S.J., 2014. Thermal characterization of erythritol/expanded graphite composites for high thermal storage capacity. Carbon 68, 67–72. Li, Wei, Hu, Chen, et al., 2016. The electrochemical synthesis of LiNbO2 in molten salts and its application for lithium ion batteries with high rate capacity. Electrochim. Acta 189, 231–236. Mar, R.W., Kramer, C.M., 1981. Pressure-temperature-composition relationships for heated draw salt systems. Sol. Energy Mater. 5 (1), 71–79. Myers Jr., Philip D., Yogi Goswami, D., 2016. Thermal energy storage using chloride salts and their eutectics. Appl. Therm. Eng. 109, 889–900. Olivares, R., Edwards, W., 2013. LiNO3-NaNO3-KNO3 salt for thermal energy storage: thermal stability evaluation in different atmosphere. Thermochim. Acta 560, 34–42. Olivares, R., Chen, C., Wright, S., 2012. The thermal stability of molten lithiumsodium-potassium carbonate and the influence of additives on the melting point. J. Sol. Energy Eng.-Trans. ASME 134, 041002. Paul, A., Shi, L., Bielawski, C.W., 2015. A eutectic mixture of galactitol and mannitol as a phase change material for latent heat storage. Energy Convers. Manage. 103, 139–146. Prieto, Cristina, Miro, Laia, et al., 2016. Temperature distribution and heat losses in molten salts tanks for CSP plants. Sol. Energy 135, 518–526. Roget, F., Favotto, C., Rogez, J., 1980. Study of the KNO3-LiNO3 and KNO3-NaNO3LiNO3 eutectics as phase change materials for thermal storage in a low temperature solar power plant. Sol. Energy 6 (24), 581–587.
179
Sharma, A., Tyagi, V.V., Chen, C.R., Buddi, D., 2009. Review on thermal energy storage with phase change and applications. Renew. Sustain. Energy Rev. 13, 318–345. Vignarooban, K., Xu, X.H., Aravy, A., et al., 2015. Heat transfer fluids for concentrating solar power systems – a review. Appl. Energy 146, 383–396. Warzoha, R.J., Weigand, R.M., Fleischer, A.S., 2015. Temperature-dependent thermal properties of a paraffin phase change material embedded with herringbone style graphite nanofibers. Appl. Energy 137, 716–725. Yan, T., Wang, R.Z., Li, T.X., Wang, L.W., 2015. A review of promising candidate reactions for chemical heat storage. Renew. Sustain. Energy Rev. 43, 13–31. Zhao, C.Y., Tian, Y., 2013. A review of solar collectors and thermal energy storage in solar thermal applications. Appl. Energy 104, 538–553. Zhao, C.Y., Wu, Z.G., 2011. Thermal property characterization of a low melting – temperature ternary nitrate salt mixture for thermal energy storage systems. Sol. Energy Mater. Sol. Cells 95, 3341–3346. Zhao, C.Y., Ji, Y.N., Xu, Z.G., 2015. Investigation of the Ca(NO3)2-NaNO3 mixture for latent heat storage. Sol. Energy Mater. Sol. Cells 140, 281–288. Zhou, Dan, Eames, Philip, 2016. Thermal characterization of binary sodium/lithium nitrate salts for latent heat storage at medium temperatures. Sol. Energy Mater. Sol. Cells 157, 1019–1025. Zuo, Z.Y., Ding, J., Xiao, X.X., 2005. Progress in the research of phase change materials with medium temperature. Mod. Chem. Ind. 25 (12), 15–19.
Contents lists available at ScienceDirect
Solar Energy journal homepage: www.elsevier.com/locate/solener
Thermophysical properties of Ca(NO3)2-NaNO3-KNO3 mixtures for heat transfer and thermal storage Y.Y. Chen, C.Y. Zhao ⇑ Institute of Engineering Thermophysics, Shanghai Jiao Tong University, Shanghai 200240, China
a r t i c l e
i n f o
Article history: Received 23 July 2016 Received in revised form 18 January 2017 Accepted 20 February 2017
Keywords: Ternary molten salt Thermophysical property Phase change material Heat transfer fluid Thermal storage
a b s t r a c t In this study calcium nitrate, sodium nitrate, and potassium nitrate were mixed to form cheap ternary molten salts based on different weight ratios. These molten salts can be used as both sensible heat storage materials and latent heat storage materials. In addition, they can be directly used as heat transfer fluids due to their low freezing temperatures. The results indicated that the mixture (Ca(NO3)2:NaNO3:KNO3 = 32:24:44 wt%) had the best performance for latent heat storage with its enthalpy of 67 J/g and melting point of about 80 °C. The specific heat capacity (1.7 J/(g °C) for the solid phase and 1.2 J/(g °C) for the liquid phase), viscosity (next to zero at 200 °C), thermal conductivity (about 1–3 W/(m K)), thermal decomposition, and cycle stability of the molten salts were measured by DSC, Malvern Kinexus Ultra+, a transient plane thermal conductivity meter, STA, and XRD, respectively. The thermophysical properties including the low manufacturing cost showed that the molten salts have great potential applications in thermal storage systems. Ó 2017 Elsevier Ltd. All rights reserved.
1. Introduction Because of the disparity between energy supply and demand, thermal storage plays a crucial role in renewable energy and industrial waste heat recovery. Heat storage technology and heat transfer fluids are very important components in thermal storage. Heat storage technology, can be classified as latent heat storage, sensible heat storage, thermochemical heat storage, and adsorptive heat storage (Sharma et al., 2009; Zhao and Tian, 2013). Among the four kinds of technology, latent heat storage has received much attention in recent decades because it has merits such as large heat capacity, and no significant temperature variation in the process of phase transition (Zuo et al., 2005). Heat transfer fluid (HTF), is an important component of thermal-energy storage (TES) systems. The main energy storage system is based on a sensible energy storage system, which is equipped with two tanks: a ‘‘cold” tank to collect solar heat and a ‘‘hot” tank to store thermal energy (Kearney et al., 2002; Kearney, 2004). If the storage system in these tanks is direct, the heat transfer fluid can also be used as a storage medium (Javier Ruiz-Cabarias et al., 2017). For both latent heat storage and heat transfer technology, the materials have a predominant effect on their use. In the case of latent heat storage, phase-change material selection is very crucial (Yan et al., 2015) and preparation and characterization of ⇑ Corresponding author. E-mail address: [email protected] (C.Y. Zhao). http://dx.doi.org/10.1016/j.solener.2017.02.033 0038-092X/Ó 2017 Elsevier Ltd. All rights reserved.
phase-change materials have been investigated by many researchers. Paul et al. (2015)studied a eutectic mixture of galactitol and mannitol. When the molar ratio of galactitol and mannitol is 3:7, they have a melting point of 153 °C and a high latent heat of fusion (292 J/g). Warzoha et al. (2015) implemented random HGNF(herringbone style graphite nanofibers) into paraffin (IGI 1230A) and quantified the effect of HGNF on the paraffin in both solid and liquid phases. Lee et al. (2014) investigated the thermal characterization of EG (expanded graphite) and the composition of EG and erythritol. They found that the latent heat and thermal conductivity improved with an increase of the EG interlayer distance under the assistance of the thermal equilibrium technique. Zhao et al. (2015) studied binary nitrate salt mixtures consisting of NaNO3 and Ca(NO3)2 with different molar ratios. They found that when the molar ratio is 3:7 (Ca(NO3)2:NaNO3), the sample has the best heat storage performance and this ratio is considered as the best eutectic composition. Roget et al. (1980) investigated a binary molten salt (LiNO3-NaNO3) and a ternary molten salt (LiNO3-KNO3NaNO3). They measured their physical properties such as thermal cycle stability, corrosion, and enthalpy. Judith and Calvet (2013) investigated a molten salt mixture which combined KNO3, NaNO3, and Ca(NO3)2 and the best composition was considered to be 36 wt % Ca(NO3)2, 16 wt% NaNO3, and 48 wt% KNO3. In t most of the commercial solar thermal power plants, synthetic oil is used as the heat transfer fluid. But the high price of synthetic oil poses a huge limitation on its utilization. In order to remedy this, recent heat transfer fluids use inorganic molten salts. From the 1980s, a molten
Y.Y. Chen, C.Y. Zhao / Solar Energy 146 (2017) 172–179
nitrate salt mixture which is 60% NaNO3 and 40% KNO3(Solar Salt) has been used as an attractive candidate for heat transfer fluid in the Concentrating Solar Power (CSP) pilot plant (Kearney, 2004; Mar and Kramer, 1981; Herrmann et al., 2004) because of its great thermal stability, the lowest cost, and highest melting point. The second commercial salt is Hitec Salt, which is a mixture of 53 wt % KNO3, 40 wt% NaNO2, and 7 wt% NaNO3 with a melting point of 142 °C (Gil, 2010). The third salt mixture is comprised of 28 wt% NaNO3, 52 wt% KNO3, and 20 wt% LiNO3, and this mixture’s thermal property was studied by Olivares and Edwards (2013) and Fernandez et al. (2014). The working temperature range of this mixture is 130–600 °C, and its viscosity is 0.03 Pa S at 300 °C (Olivares et al., 2012), which is good for fluid flow properties. But the high price of LiNO3 ($4.32/kg) poses a huge limitation on its development. The fourth salt mixture is Carbonate salt (Vignarooban et al., 2015) with 32.1 wt% Li2CO3, 33.4 wt% Na2CO3, and 34.5 wt% K2CO3, which was proposed to replace the nitrate salt to increase the thermal stability temperature limit. The melting point of this mixture is 400 °C and the decomposition temperature is 800–850 °C. Among the mentioned materials which are used for both latent heat storage and heat transfer fluid, nitrate salts have wide use in actual practice due to their merits such as high heat storage capacity (enthalpy), favorable thermal stability, satisfactory compatibility with containers, low corrosion (Javier Ruiz-Cabarias et al., 2017), low vapor pressure, low viscosity in most cases, and chemical inertness (Zhao and Wu, 2011). Reviewing relevant references (Paul et al., 2015; Warzoha et al., 2015; Lee et al., 2014; Zhao et al., 2015; Roget et al., 1980; Judith and Calvet, 2013; Mar and Kramer, 1981; Herrmann et al., 2004; Gil, 2010; Olivares and Edwards, 2013; Fernandez et al., 2014; Olivares et al., 2012; Vignarooban et al., 2015; Zhao and Wu, 2011; Li et al., 2016; Fernandez and Perez, 2016; Prieto et al., 2016; Brandon and Davidson, 2017; Myers and Yogi Goswami, 2016; Zhou and Eames, 2016), Ca (NO3)2, NaNO3, and KNO3 are eligible as both phase change materials and heat transfer fluid due to their insignificant corrosion, no toxicity, and low price. Although several reports mentioned Ca (NO3)2, NaNO3 or KNO3 as an additive in the mixtures, the study remains incomplete. For example, Judith and Calvet (2013) used Ca(NO3)2-NaNO3-KNO3 as phase-change materials, but the data was incomplete (only 6 groups) and the thermal conductivity was not presented in his study. Roget et al. (1980) prepared a ternary molten salt (LiNO3-KNO3-NaNO3), but the price of LiNO3 is much higher than that of Ca(NO3)2, which is not economical in solar plants. Moreover, Zhao et al. (2015) used Ca(NO3)2 as an additive in NaNO3, but its’ melting point (220 °C) isn’t suitable for low or medium temperatures. The present study prepared a ternary molten salt consisting of KNO3, NaNO3, and Ca(NO3)2 based on different weight ratios. Experiments showed that this mixture has a melting point of 80 °C and the maximum stable temperature is 600 °C. However, this mixture focuses mainly on low and medium temperatures (less than 200 °C). Hence, other thermophysical properties such as thermal conductivity and viscosity were measured in the temperature range of 50–200 °C. Further study indicated a correlation between the heating rate and heat enthalpy, and the economic impact id discussed at the end of this study. Overall, the experimental results showed that these ternary molten salts are suitable candidates as thermal storage materials and heat transfer fluids.
173
by Sinopharm Chemical Reagent Co. Ltd. The method of preparing the formulations was to mix the anhydrous form of the single components at room temperature. In order to produce the anhydrous form of calcium nitrate, the tetrahydrate form was heated in two stages: (i) at 150 °C for 10 h and (ii) at 350 °C for 20 h. Then the anhydrous salts were put in a muffle furnace at 150 °C to ensure drying. In order to avoid water absorption, especially for the extremely anhydrous calcium nitrate, the handling and mixing of the salts, as well as putting samples in the testing pan, were performed in a box under dry nitrogen. To weigh the samples based on 32 sets of data, an analytical balance with a resolution of 0.1 mg was used. 2.2. Simultaneous Thermal Analyzer Fusion heat, cycle stability, and the endothermic peak’s property under different heat rates were measured by a Simultaneous Thermal Analyzer (STA 8000, Perkin Elmer) which provides realtime analysis of sample weight change and heat flux. The weight and temperature accuracy of the STA were 0.0001 mg and 0.01 °C, respectively. The temperature accuracy of the oven was 0.1 °C. To measure the fusion heat, the 32 groups of samples were heated from 50 °C to 250 °C at a rate of 10 °C/min under N2 purging of 20 L/min. The samples were heated 33 times in three stages for the cycle stability test: (i) from 50 °C to 250 °C at a rate of 10 °C/ min, (ii) at 250 °C for 1 min, and (iii) from 250 °C to 50 °C at the rate of 10 °C/min. In order to determine the endothermic peak’s property at different heating rates, the sample was measured at different heating rates of 5 °C/min, 10 °C/min, 15 °C/min, and 20 °C/min. 2.3. Differential scanning calorimetry Differential scanning calorimetry (DSC 8000, Perkin Elmer) was used to determine the specific heat capacity of the mixtures. The temperature calibration was performed with benzophenone and caffeine. The heat flow calibration was carried out by indium. After the aforementioned calibration, the samples were heated from 50 °C to 80 °C and from 200 °C to 250 °C, respectively, in order to measure the samples’ solid and liquid specific heat capacity. 2.4. X-ray diffraction X-ray diffraction (XRD) measurements were carried out to investigate the remnants after thermal decomposition. The sample was heated by STA from 200 °C to 950 °C at a rate of 10 °C/min, and then, XRD was used to investigate the remnants at ambient temperature. 2.5. Transient plate method The transient plate method was performed to measure the sample’s thermal conductivity. This methods measure the transient thermal conductivity and the instrument’s uncertainty was about 10%. 2.6. Malvern Kinexus Ultra+
2. Material preparation and experimental methods 2.1. Material preparation The molten salt was prepared from Ca(NO3)24H2O(purity > 99%), NaNO3(purity > 99%), and KNO3(purity > 99%) manufactured
Malvern Kinexus Ultra+ was used to test the sample’s viscosity and the apparatus’s uncertainty was about 10%. In order to avoid the water absorption, the experiments were performed under a flow of either dry nitrogen or dry synthetic air. The viscosity was measured with a 40 mm diameter stainless steel parallel plate.
174
Y.Y. Chen, C.Y. Zhao / Solar Energy 146 (2017) 172–179
3. Results and discussion 3.1. Fusion heat Fusion heat and melting point are important parameters for molten salts. Some nitrate salts’ enthalpy and melting points are shown in Fig. 1 and Table 1. Of 32 test groups, only 3 groups showed a significant fusion heat. The STA thermograms obtained for these three Ca(NO3)2-NaNO3-KNO3 mixtures are shown in Fig. 2 and Table 2. In the figure, the observed enthalpy changed with the weight ratios of Ca(NO3)2, NaNO3, and KNO3 and this phenomenon can be explained by the mismatch between the fixed and experimental weight ratios. Specifically, when the three nitrate salts are mixed according to the weight proportions, they can only form a eutectic salt at a fixed proportion which is not known in advance. If the weight proportion is not the right value, there will be surplus Ca(NO3)2, NaNO3, KNO3, or a combination of them. When the mixtures are heated, the eutectic composition melts first, and then the surplus salts melt (Zhao et al., 2015). The different data will result in various masses of the ternary eutectic composition, and the heat of fusion will subsequently be different (as shown in Table 2). In Table 2, the Tmi1, Tmi2, and Tmi3 (melting point) are similar since the melting point only depends on the materials of the mixture. Furthermore, Fig. 2 and Table 2 also indicate an invariant temperature value (120 °C), which is the peak temperature. Tg in Fig. 2 and Table 2 represents the temperature of glass transition and its value is about 50 °C. In Fig. 2, there is a small observed enthalpy between Tg and Tm. This phenomenon can be attributed to the crystallization reaction which is prior to the melting of the mixture. This subtle energy is used to transform the state from a metastable to stable crystal structure. After this process, this ternary salt begins to quickly melt. Actually, when heating the materials with a vitreous state, they always require a part of the energy (subtle energy as shown in Fig. 2) to transform their state from a metastable to thermodynamically stable crystal structure (Judith and Calvet, 2013). 3.2. Cycle stability Among the three groups of samples, the first (Ca:Na: K = 34:20:46 wt%) and third samples (Ca:Na:K = 32:24:44 wt%) had the highest fusion heat, which was 51 J/g and 67 J/g,
Table 1 Enthalpy of the three nitrate salts. Material
Enthalpy
Ca(NO3)2 NaNO3 KNO3
145 J/g 177 J/g 88 J/g
Fig. 2. STA curves of the mixing salts.
respectively. Hence, these two samples were used to test the cycle stability. The endothermic and exothermic curves obtained from STA are shown in Fig. 3(a), (b) and (c) and Fig. 4(a) and (b), respectively. Table 3 gives the enthalpy of each peak. As shown in Fig. 3(a), (b) and (c), endothermic peak 2 of the first sample and endothermic peak 3 of the third sample had the great reproducibility. Additionally, exothermic peak 1 of the first sample and exothermic peaks 1 and 2 of the third sample also had favorable reproducibility during the test. The weight loss of the first and third samples was 0.2% and 0.25%, respectively. The above properties of the two samples indicate that they have satisfactory chemical stability ranging from 50 °C to 250 °C. For the third sample, its initial point of the exothermic peak was Tsi = 180 °C and the final point was Tsf = 100 °C. Its enthalpy of the exothermic peak (peak 1 and 2) was 11 J/g which accounts for approximately 15% of the enthalpy of the endothermic peak. For the first sample, the initial point of its exothermic peak was Tsi = 160 °C and the final point was Tsf = 100 °C. The enthalpy of the exothermic peak (peak 1) was 2 J/g which accounts for 4% of the enthalpy of the endothermic peak. The possible reason for the difference between the enthalpy of the exothermic peak and that of the endothermic peak in the two samples is the rate of nucleating. In the cooling process, the rate of the cooling is faster than the rate of nucleating. When the cooling process ends, the two samples will not be thoroughly nucleated, which will result in the smaller enthalpy of the exothermic peak compared to that of endothermic peak. Therefore, the appropriate rate of cooling is very critical. 3.3. Specific heat capacity
Fig. 1. Ternary phase diagram of Ca(NO3)2-NaNO3-KNO3 mixtures.
Although latent heat is used to store energy in latent heat storage, latent and sensible heat storage are closely related. On one hand, before the materials reach the temperature of phase change, they use the sensible heat to store energy. On the other hand, due to the extremely low thermal conductivity of the phase change materials, the temperature difference in the inner area of the materials is huge, which will lead to the fact that when some parts begin
175
Y.Y. Chen, C.Y. Zhao / Solar Energy 146 (2017) 172–179 Table 2 Samples’ fusion heat and glass transition temperature. Number
Initial point
Ca:Na:K = 32:20:48 (wt%) Ca:Na:K = 32:24:44 (wt%) Ca:Na:K = 34:20:46 (wt%)
Tmi1 = 80 °C Tmi2 = 84 °C Tmi3 = 86 °C
Final point Tmf1 = 180 °C Tmf2 = 190 °C Tmf3 = 200 °C
Heat of fusion 45 J/g 67 J/g 51 J/g
Density 3
1455 kg/m 1460 kg/m3 1482 kg/m3
Heat of fusion
Temperature of the peak
Temperature of glass transition
65 kJ/L 98 kJ/L 76 kJ/L
Tpeak,1 = 120 °C Tpeak,2 = 120 °C Tpeak,3 = 120 °C
Tg,1 = 50 °C Tg,2 = 50 °C Tg,3 = 50 °C
(a) First sample Ca:Na:K=34:20:46 (wt%)
(a) First sample Ca:Na:K=34:20:46 (wt%)
(b) Third sample Ca:Na:K=32:24:44 (wt%)
(b) Third sample Ca:Na:K=32:24:44 (wt%)
Fig. 4. Samples’ cycle curves (the X axis is time).
Table 3 Enthalpy of each peak. Sample
Enthaply of Peak, 1
Enthaply of Peak, 2
Enthaply of Peak, 3
Ca:Na:K = 34:20:46 (wt%) Ca:Na:K = 32:24:44 (wt%)
2 J/g 1.4 J/g
51 J/g 9.6 J/g
0 67 J/g
phase transformation, the others still have not reached the temperature of transition. Hence, in actual applications, the specific heat is very crucial. The following expression (Li et al., 2016) shows the heat storage capacity of phase change systems.
(c) Comparison between the two samples Fig. 3. Samples’ cycle curves (the X axis is temperature).
Z Q¼
Tm
Ti
Z mCp dT þ mam Dhm þ
Tf
mCp dT Tm
ð1Þ
176
Y.Y. Chen, C.Y. Zhao / Solar Energy 146 (2017) 172–179
Based on DSC, the specific heat capacity of the third sample (because it has the highest fusion heat) was measured and the result is shown in Fig. 5(a) and (b). Based on these curves, the equation of the specific heat capacity was calculated and can be expressed as follows: For the solid phase, the curve can be expressed by a quadratic function as:
Csp ¼ A1 þ B1 X þ C1 X2
ðA1 ¼ 3:70785; B1 ¼ 0:06525;
5
C1 ¼ 5:10854e Þ
ð2Þ
For the liquid phase, the curve can be expressed by a quadratic function as:
ClP ¼ C1 þ D1 X þ E1 X2
ðC1 ¼ 0:26884; D1 ¼ 0:01016;
E1 ¼ 1:6466e5 Þ
ð3Þ
In Fig. 5(a), the curve quickly rises when the temperature exceeds 68 °C, indicating that the heat capacity rises with an increase in temperature, while in the range of 50–68 °C, the curve is almost stable regardless of the variation of temperature. The possible reason is that in this temperature range, the sample is metastable vitreous. In Fig. 5(b), the liquid curve quickly rises with an increase in temperature, which indicates that the specific heat capacity of the liquid phase is very sensitive to temperature variation.
Fig. 6. Thermal conductivity of the third sample (Ca:Na:K = 32:24:44 wt%).
3.4. Thermal conductivity Thermal conductivity was determined by the transient plate method and the result is shown in Fig. 6 and Table 4. Based on Fig. 6 and Table 4, the thermal conductivity rises when the temperature reaches 120 °C, 150 °C, and 180 °C. This difference can be attributed to the state change. In other words, when the temperature is 120 °C and 150 °C, the sample is molten and with an increase of temperature, more parts of the sample transform from the molten state to the liquid state, so the thermal conductivity rises with an increase of temperature. When the temperature reaches 180 °C, the entire sample melts and all of its parts have become liquid, so the thermal conductivity is higher than that at 120 °C and 150 °C. 3.5. Endothermic peak
(a) Curve of solid phase (50-80
The endothermic peak’s property at different heating rates was investigated by STA and the result is presented in Fig. 7(a) and (b). As shown in Fig. 7(a) and (b), the endothermic peak becomes wider with an increase of heating rate (Roget et al., 1980). Since the endothermic peak becomes wider with an increase of heating rate, it will be insignificant or even disappear in the experiments. This phenomenon can be attributed to the mismatch between the heating rate and the materials’ phase transition speed. In other words, the materials need some time to transform their phases, and a faster heating rate will result in inadequate melting of the materials. Table 5 shows the temperature (Tmi represents the initial temperature, Tmf reflects the final temperature). 3.6. Thermal decomposition Thermal decomposition was investigated by STA and XRD. The result is presented in Figs. 8 and 9(a) and (b). In Fig. 8, the sample decomposed at 650 °C and 750 °C, respectively. Below 600 °C, there
Table 4 Thermal conductivity K = 32:24:44 (wt%).
(b) Curve of liquid phase (200-250 Fig. 5. Specific heat capacity curves of the third sample (Ca:Na:K = 32:24:44 wt%).
of
the
third
sample
Ca:Na:
Temperature (°C)
Thermal conductivity (W/m K)
120 150 180
1.090 1.929 3.194
177
Y.Y. Chen, C.Y. Zhao / Solar Energy 146 (2017) 172–179
Fig. 8. Decomposition of the third sample Ca:Na:K = 32:24:44 (wt%).
(a) First sample Ca:Na:K=34:20:46 (wt%)
(a) Decomposing at 650 (b) Third sample Ca:Na:K=32:24:44 (wt%) Fig. 7. Samples’ endothermic peak performance at different heating rates.
Table 5 Samples’ melting temperature range at different heating rates.
Tmi,1-Tmf,1 Tmi,2-Tmf,2 Tmi,3-Tmf,3 Tmi,4-Tmf,4
Ca:Na:K = 34:20:46 (wt%)
Ca:Na:K = 32:24:44 (wt%)
75–170 °C 80–180 °C 95–210 °C 105–235 °C
75–175 °C 85–195 °C 100–220 °C 105–235 °C
was no decomposition. This phenomenon shows that the mixture is very stable below 600 °C. In Fig. 9(a) and (b), when the temperature ranges from 650 °C to 750 °C, the first decomposition happens, with the main products being KNO3, CaO, and NaNO3. When the temperature exceeds 750 °C, the main remnants are Ca (OH)2, KNO3, CaO, and NaNO3. This indicates that Ca(NO3)2 decomposes to CaO and Ca(OH)2, but KNO3 and NaNO3 are still stable. 3.7. Viscosity The viscosity of the first sample (Ca:Na:K = 34:20:46 wt%) and third sample (Ca:Na:K = 32:24:44 wt%) are shown in Fig. 10 (a) and (b). In the figure, the viscosity of the two samples decreased with an increase in temperature, and when the temperature
(b) Decomposing at 750 Fig. 9. XRD of remnants.
ranged from 115 to 200 °C, the viscosity was very much lower than the viscosity in the temperature range of 100–115 °C. This phenomenon can be attributed to the state change. In other words, when the temperature is lower than 115 °C, the state of the two samples is molten, then with an increase in temperature, the state transforms from molten to liquid, so the viscosity reduces to a very
178
Y.Y. Chen, C.Y. Zhao / Solar Energy 146 (2017) 172–179 Table 6 Recent price of nitrates. Nitrate
Price ($/Metric Ton)
Ca(NO3)24H2O KNO3 NaNO3 LiNO3
300 930 330 7980
4. Economic analysis
(a) Viscosity of the molten salts mixtures 100-200
Since the price of the materials will have a direct influence on their utilization, they must have the lowest price to decrease the capital investment as much as possible. The prices of some nitrates are presented in Table 6. Based on Table 6, the price of the third sample (Ca:Na:K = 32:24:44 wt%) is $626 per Metric Ton, and the price of the first sample(Ca:Na:K = 34:20:46 wt%) is $640 per Metric Ton. Therefore, the third sample should be chosen as both the phase-change material and heat transfer fluid. 5. Conclusions
(b) Viscosity of the molten salts mixtures (115-200 ) Fig. 10. Viscosity of the molten salts mixtures.
small value which is almost zero. The viscosity of the two samples in the temperature range of 115–200 °C is shown in Fig. 10(b). In this figure, the viscosity of the third sample (Ca:Na: K = 32:24:44 wt%) was a little greater than that of the eleventh sample (Ca:Na:K = 34:20:46 wt%) in the temperature range of 115–200 °C, but the difference was very small. Hence the viscosity of the two samples can be considered similar. However, when the temperature ranges from 100 to 115 °C (as shown in Fig. 10(a)), the two samples are molten and the viscosity of the first sample was very much greater than that of the third sample. This difference can be explained by an increase in the amount of calcium nitrate which is contained in the molten salts mixture. Molten salts containing calcium nitrate tends to form a glassy state, and this can increase the viscosity (Judith and Calvet, 2013; Vignarooban et al., 2015). Hence, the viscosity of the first sample was greater than that of the third sample when both the samples are molten because the first sample contains more calcium nitrate than the third sample. However, when the temperature rises, the samples change their state from molten to liquid, and the amount of calcium nitrate contained in the molten salts mixtures may have little effect on the viscosity(as shown in Fig. 10(b)) and other elements in the molten salts mixtures may have more influence. This needs further investigation. Overall, both the third and first sample can be expected to be chosen as the heat transfer fluids at a normal operating temperature for a thermal solar plant.
This study made ternary salts by using Ca(NO3)2, NaNO3, and KNO3, which can be used as both phase-change materials and heat transfer fluids. Among the ternary salts, the mixture of Ca(NO3)2, NaNO3, and KNO3 with a weight ratio of 32:24:44 (wt%) is considered to be the optimum proportion for its best cycle stability, specific heat capacity, viscosity, thermal conductivity, and economy cost. Its melting point is 80 °C and fusion heat is 67 J/g. The STA test showed that the thermal cycle stability is great in the temperature range of 50–250 °C and the XRD test showed that the upper limit temperature is 600 °C. The thermal conductivity of the optimal mixture (Ca:Na:K = 32:24:44 wt%) is about 1–3 W/ (m K) and the viscosity is very small at the normal operating temperature. Furthermore, the cost of the optimal mixture (Ca:Na: K = 32:24:44 wt%) is about $626 per Metric Ton, which is acceptable in the industry. In summary, the mixture (Ca:Na: K = 32:24:44 wt%) can be applied as both a phase-change material and heat transfer fluid in solar plants due to its significant fusion heat, low viscosity, cheap price, and excellent thermal stability. Acknowledgement This work is supported by the National Key Basic Research Program of China (973 Project: 2013CB228303). References Brandon, J., Davidson, Jane H., 2017. Demonstration of prototype molten salt solar gasification reactor. Sol. Energy 145, 224–230. Fernandez, A.G., Perez, F.J., 2016. Improvement of the corrosion properties in ternary molten nitrate salts for direct energy storage in CSP plants. Sol. Energy 134, 468–478. Fernandez, A.G., Ushak, S., Galleguillos, H., Perez, F.J., 2014. Development of new molten salts with LiNO3 and Ca(NO3)2 for energy storage in CSP plants. Appl. Energy 119, 131–140. Gil, A., 2010. State of the art on high temperature thermal energy storage for power generation. Part 1-concepets, materials and modelization. Renew. Sustain. Energy Rev. 14 (1), 31–35. Herrmann, U., Kelly, B., Price, H., 2004. Two-tank molten salt storage for parabolic trough solar power plants. Energy 29 (5–6), 883–893. Ruiz-Cabarias, F. Javier, Jove, Aleix, Prieto, Cristina, et al., 2017. Materials selection of steam-phase change material (PCM) heat exchange for thermal energy storage systems in direct steam generation facilities. Sol. Energy Mater. Sol. Cells 159, 526–535. Judith, C., Calvet, N., 2013. Ca(NO3)2-NaNO3-KNO3 Molten salt mixtures for direct thermal energy storage systems in parabolic trough plants. J. Sol. Energy Eng. ASME 5. 021016-1-7. Kearney, D., 2004. Engineering aspects of a molten salt heat transfer fluid in a trough solar field. Energy 29 (5–6), 861–870. Kearney, D., Kelly, B., Mahoney, R., 2002. Assessment of a molten heat transfer fluid in a parabolic trough solar field. Sol. Energy Eng. 125 (2), 170–176.
Y.Y. Chen, C.Y. Zhao / Solar Energy 146 (2017) 172–179 Lee, S.Y., Shin, H.K., Park, M., Rhee, K.Y., Park, S.J., 2014. Thermal characterization of erythritol/expanded graphite composites for high thermal storage capacity. Carbon 68, 67–72. Li, Wei, Hu, Chen, et al., 2016. The electrochemical synthesis of LiNbO2 in molten salts and its application for lithium ion batteries with high rate capacity. Electrochim. Acta 189, 231–236. Mar, R.W., Kramer, C.M., 1981. Pressure-temperature-composition relationships for heated draw salt systems. Sol. Energy Mater. 5 (1), 71–79. Myers Jr., Philip D., Yogi Goswami, D., 2016. Thermal energy storage using chloride salts and their eutectics. Appl. Therm. Eng. 109, 889–900. Olivares, R., Edwards, W., 2013. LiNO3-NaNO3-KNO3 salt for thermal energy storage: thermal stability evaluation in different atmosphere. Thermochim. Acta 560, 34–42. Olivares, R., Chen, C., Wright, S., 2012. The thermal stability of molten lithiumsodium-potassium carbonate and the influence of additives on the melting point. J. Sol. Energy Eng.-Trans. ASME 134, 041002. Paul, A., Shi, L., Bielawski, C.W., 2015. A eutectic mixture of galactitol and mannitol as a phase change material for latent heat storage. Energy Convers. Manage. 103, 139–146. Prieto, Cristina, Miro, Laia, et al., 2016. Temperature distribution and heat losses in molten salts tanks for CSP plants. Sol. Energy 135, 518–526. Roget, F., Favotto, C., Rogez, J., 1980. Study of the KNO3-LiNO3 and KNO3-NaNO3LiNO3 eutectics as phase change materials for thermal storage in a low temperature solar power plant. Sol. Energy 6 (24), 581–587.
179
Sharma, A., Tyagi, V.V., Chen, C.R., Buddi, D., 2009. Review on thermal energy storage with phase change and applications. Renew. Sustain. Energy Rev. 13, 318–345. Vignarooban, K., Xu, X.H., Aravy, A., et al., 2015. Heat transfer fluids for concentrating solar power systems – a review. Appl. Energy 146, 383–396. Warzoha, R.J., Weigand, R.M., Fleischer, A.S., 2015. Temperature-dependent thermal properties of a paraffin phase change material embedded with herringbone style graphite nanofibers. Appl. Energy 137, 716–725. Yan, T., Wang, R.Z., Li, T.X., Wang, L.W., 2015. A review of promising candidate reactions for chemical heat storage. Renew. Sustain. Energy Rev. 43, 13–31. Zhao, C.Y., Tian, Y., 2013. A review of solar collectors and thermal energy storage in solar thermal applications. Appl. Energy 104, 538–553. Zhao, C.Y., Wu, Z.G., 2011. Thermal property characterization of a low melting – temperature ternary nitrate salt mixture for thermal energy storage systems. Sol. Energy Mater. Sol. Cells 95, 3341–3346. Zhao, C.Y., Ji, Y.N., Xu, Z.G., 2015. Investigation of the Ca(NO3)2-NaNO3 mixture for latent heat storage. Sol. Energy Mater. Sol. Cells 140, 281–288. Zhou, Dan, Eames, Philip, 2016. Thermal characterization of binary sodium/lithium nitrate salts for latent heat storage at medium temperatures. Sol. Energy Mater. Sol. Cells 157, 1019–1025. Zuo, Z.Y., Ding, J., Xiao, X.X., 2005. Progress in the research of phase change materials with medium temperature. Mod. Chem. Ind. 25 (12), 15–19.