- Email: [email protected]
Experimental investigation on supercooling, thermal conductivity and stability of nanofluid based composite phase change material
Journal of Energy Storage 17 (2018) 47–55
Contents lists available at ScienceDirect
Journal of Energy Storage journal homepage: www.elsevier.com/locate/est
Experimental investigation on supercooling, thermal conductivity and stability of nanofluid based composite phase change material Munyalo Jotham Muthoka, Zhang Xuelai* , Xu Xioafeng Institute of Thermal Storage Technology, Merchant Marine College, Shanghai Maritime University, Shanghai, 201306, China
A R T I C L E I N F O
Article history: Received 22 September 2017 Received in revised form 28 November 2017 Accepted 9 February 2018 Available online 24 February 2018 Keywords: Phase change material Thermal conductivity Supercooling Aggregation Barium chloride dehydrate
A B S T R A C T
With increasing concerns over global warming, there is compelling need to apply energy technologies to improve energy efficiencies. One of the new technologies is the use of phase change materials (PCMs) to store energy and release it on demand. However, most of these materials undergo supercooling, aggregation and have low thermal conductivity that inhibits effective heat transfer. In this paper supercooling, stability, energy storage,thermal conductivity and latent heat of fusion of water based nanofluid of barium chloride dehydrate (BaCl22H2O) were experimentally studied by adding a mass fraction of 0.2 w.t%–1 wt.% magnesium oxide (MgO) and 0.2 wt.%–1 wt.% multi-walled carbon nanotubes (MWCNTs). Results show that by adding separately mass fraction of 1.% of MgO and MWCNTs reduce the supercooling degree of barium chloride dehydrate by 85% and 92% respectively. At the same mass fraction, thermal conductivity also improves by 6% on addition of MWCNTs and 17% on addition of MgO. However, the enthalpies reduce by 7% at 1 wt.% MgO and by 12.3% at 1 wt% MWCNTs. It was found that MgO exhibited relatively higher thermal conductivity and a lower reduction in latent heats at a mass fraction of 1 wt.%. Surfactant was also found to prevent aggregation at low temperatures. © 2018 Elsevier Ltd. All rights reserved.
1. Introduction Cool thermal energy storage (CTES) technology, which is storing cool thermal energy in order to bridge the time gap between the energy availability and energy use is being considered as one of the best method for thermal energy management [1,2]. The CTES is useful in central air-conditioning systems in large buildings [3,4], supermarkets refrigeration, high powered electronic cooling applications, marine applications, pharmaceutical field and various industrial cooling process applications where the cooling requirement is highly intermittent. The efficient operation of CTES system reduces the additional energy consumption during its cyclic charging and discharging operations. Therefore, PCMs have been applied to increase thermal energy storage capacity of different systems [5]. The use of PCMs provides higher heat storage capacity and a constant temperature behavior during charging and discharging compared to sensible heat storage. High energy storage density is a desirable property of any storage system which guarantees smaller size for the storage system. In the application of PCMs, the most used phase change is solid–liquid
* Corresponding author. Tel.: +86 13127992577; fax: +86 02138282925. E-mail addresses: [email protected] (J.M. Munyalo), [email protected] (X. Zhang). https://doi.org/10.1016/j.est.2018.02.006 2352-152X/© 2018 Elsevier Ltd. All rights reserved.
phase change as the volume is easily controlled. The substances used can be organic such as paraffin and fatty acids or inorganic such as aqueous salts solutions. Water based inorganic salts are widely used as a PCMs because of low cost, readily available and environmental friendly [6]. However, the most serious problem encountered when using these PCMs is the supercooling phenomenon during solidification [7,8], aggregation at low temperature [9,10] and a low thermal conductivity [11]. A lower degree of supercooling results in a better coefficient of performance of the refrigeration system [12] and a high thermal conductivity reduces the solidification and melting time of PCMs. In recent years, many researchers have studied ways of reducing supercooling and at the same time increase thermal conductivity [13,14]. One of the ways is to use nanoparticles to make nanofluid which can be used as a PCMs. Researchers have pointed out that nanostructures can reduce the supercooling degree and increase thermal conductivity [13,15,16]. Qinbo et al. [17] carried out thermal conductivity, supercooling degree, latent heat, specific heat, and rheological behaviors of nano PCMs by suspending a small amount of TiO2 nanoparticles in saturated barium chloride aqueous solution; his experimental results showed that with volume fraction of 1.130%, the thermal conductivities of nanofluids PCMs were enhanced by 12.76% at 5 C and the supercooling degree was reduced by 84.92% but the latent heat and specific heat were slightly decreased. Yudong et al. [15] carried out an experimental study
48
J.M. Munyalo et al. / Journal of Energy Storage 17 (2018) 47–55
temperature thermostat bath has a temperature range of 65 C to 100 C; its precision is 0.05 C. For the purpose of this experiment, the low temperature thermostat bath’s temperature was set at 20 C for freezing and 20 C for melting. The T-type thermocouples which were used to take temperature measurements at various points have a precision of 0.5 C and thermal response period of 0.4 s.
on the supercooling degree and nucleation behavior of nanofluids phase change material which were prepared by adding small fraction of graphene oxide nanosheets in deionized water without any dispersants. The results showed that the supercooling degree was reduced by 69.1%, and the nucleation started in advance, reducing the time by 90.7%. Liu et al. [18] carried out a research on nucleation rates and supercooling degree of deionized water and graphene oxide nanofluids using classical nucleation theory. He found that whereas the supercooling degree of deionized water was approximately 31.5 K, the supercooling degrees of four different concentrations of graphene oxide nanofluids were 7.98, 7.93, 3.05, and 3.03 K concluding that the supercooling degrees were reduced by more than 74% with the corresponding increase in volume concentration. Stability of nanofluid has also been carried out by many researchers [19,20] where the focus was on stability with respect to pH, zeta potential, particle size distribution, and its effect on viscosity and thermal conductivity. Most of these previous researchers’ mainly focused on the nanofluid PCMs used in the air conditioning cool storage, where the phase change temperatures are above 0 C. However, some enterprises, such as ice and chemical plant, need the cool storage temperature below 0 C. At the moment there are few researches and applications for nanofluid PCMs in sub-zero temperature cool storage available [5]. In this paper, considering the application in water ice making plant where the temperature is between 1 C to 0 C. While the phasechange temperature of 24 wt.% BaCl22H2O solution is about 7.5 C and considering heat losses together with overcoming supercooling problem of water in some cases [21–23], the 7.5 C temperature difference is appropriate. Compared with other phase change materials that have found use in sub-zero temperature application such NaCl–H2O and CaCl–H2O, barium chloride dehydrate is found to have the highest latent heat of fusion. So it was screened out as a suitable candidate for PCMs. The main focus in this paper was to experimentally study supercooling, stability, latent heat of fusion and thermal conductivity of barium chloride dehydrates under the effect of different cooling temperatures, freeze cycling and nano particles concentration of MgO and MWCNTs. Nanofluids were prepared using a two-step method. N, N-dimethyl formamide was used as surfactant to stabilize the dispersion of MWCNTs nanoparticles in aqueous barium chloride dehydrate. SDS surfactant was used for MgO nanofluid to enhance its stability.
In order to evaluate how different factors influence supercooling degree of water based barium chloride dehydrate, different cooling temperatures and nanoparticle concentrations were used. The supercooling characteristic test was carried out using constant temperature bath (DC 6115, Shanghai Hengping Equipment Company). To avoid the contamination of barium chloride dehydrate solution which may interfere with the supercooling degree and thermal conductivity measurements, different beakers, glass rod and tweezers were placed into the ultrasonic machine to clean them by means of ultrasonic vibrations. Polyurethane sealing covers were used to cover the tops of the beakers to avoid the content from being contaminated by external environment. The thermocouples wires pass through the polyurethane sealing covers and stretch into the solution with thermocouple probes being placed at the mid position of the beaker where temperature is taken. An electronic balance (FA2004) which was used to measure the weights of experimental materials has a precision of 0.1 mg. Thermal conductivities of nanofluids were measured by the Hot disk (TPS 500, Thermal Constant Analyser,TCA, Sweden) where KS-1 probe sensor was used. Calibration of instrument with DI water was performed before starting off the measurements of thermal conductivities of the nanofluids. Thermal conductivity of DI water at 20 C was found to be 0.602 W/mK. The latent heat of the base fluids and the nanofluids were measured with a differential scanning calorimeter (DSC 200, NETZSCH, German) with an accuracy of 1.0%. The DSC instrument was first calibrated using a standard high-purity indium specimen prior to use. For each sample at least five measurements were carried out and average value with a standard deviation of less than 1% was given as a result. To test stability of nanofluid, a volume of nanofluid sample was kept in glass beaker and photographs were taken at regular time interval.
2. Material and methods
2.3. Preparation of nanofluid
2.1. Experimental setup
A BaCl22H2O (produced by Sinopharm Chemical Reagent Co, Ltd purity 99.5%) without additives was prepared by dissolving it in deionized water to make an aqueous solution of 24 wt.% BaCl22H2O. The solution was evenly stirred in order to ensure all the barium chloride dehydrate dissolves. The nanofluid was prepared by a two-step method. The nanoparticle has a bigger specific surface area and surface energy because of its small size. It is apt to reunite in the process of preparation, reprocessing and the application therefore obtaining a well-distributed and stabilized nanofluids is a critical step. Five samples of 0.2, 0.4, 0.6, 0.8 and 1 wt.% MgO and 1 wt.% of SDS as surfactant (Supplied by Aladdin Industrial Corporation, Shanghai, China) were prepared for experiment. First the solution was stirred in a magnetic stirrer for 30 min and a then ultrasonicated for 45 min at temperature of 50 C, with a working time of 5 s and stop time of 10 s. The same mass fraction was used to prepare MWCNTs (Supplied by Aladdin Industrial Corporation,Shanghai,China) nanofluid. 1 wt.% of N,Ndimethyl formamide, having a molecular formular C3H7NO (Supplied by Aladdin Industrial Corporation,Shanghai,China) was first put into each five samples then stirred in a magnetic stirrer (HJ-6A,West Jintan Zhengrong Experiment Equipment Company,
The experiment was set up as shown in Fig. 1. It consists of agilent data logger (34972A), computer, low-temperature thermostatic bath (DC-6515, supplied by Shanghai HengPing Instrument Factory), thermocouples and polyurethane sealing cover. The low-
Fig. 1. Experimental setup. 1-computer; 2-agilent data logger; 3-thermocouple; 4-polyurethane sealing cover; 5-beaker; 6-nanofluid PCM; 7-bracket; 8-low- temperature thermostat bath.
2.2. Experimental technique
J.M. Munyalo et al. / Journal of Energy Storage 17 (2018) 47–55
49
Fig. 2. Magnetic stirring of Nano fluid PCM.
China) before adding the MWCNTs and stirring further for 30 min. The solutions were then transferred to ultrasonic vibrator for 45 min with the same setting as for MgO nanofluid. Samples of prepared MWCNTs are shown in Fig. 2 and the properties of the nanoparticles are shown in Table 1. 3. Results and discussion 3.1. Supercooling As shown in Fig. 3(a), when 24 wt.% BaCl22H2O solutions having a melting point of 7.5 C was exposed to different cooling temperatures, the highest supercooling degrees occurred at temperatures of 15 C and 20 C with 6 C and 6.5 C respectively. At 35 C, no supercooling that was witnessed with the other intermediate temperatures having very minimal supercooling degree of less than 0.5 C. This generally shows that supercooling degree of aqueous barium chloride dehydrate is affected by the cooling temperature. The higher the cooling temperature is, the higher the supercooling degree and the lower the cooling temperature is, the lower supercooling degree. This is attributed to the fact that nucleus forms very fast when there is high heat removal but when there is low heat removal the nucleus forms very slowly hence low nucleation. By using low temperatures to charge low temperature PCM, the supercooling degree can be eliminated completely. It is obvious that when charging in low temperature, charging time is greatly reduced as it takes shorter time to charge in low temperature than in high temperature though one has to strike a balance between cost and benefit as low temperature comes at extra cost. As shown in Fig. 3(b),when aqueous barium chloride dehydrate underwent four freezing and melting cycles at a cooling temperature of 20 C, supercooling degree was reduced from 6 C at the first cycle to about 1 C at the fourth cycle. The reason for reduction in supercooling degree could have been that during subsequent freezing some parts of the solution could have been already supercooled hence prompting quicker formation of nucleus. This is in good agreement with findings in
the literature [25]. When 0.2–1% mass fraction of MgO was added to barium chloride dehydrate as shown in Fig. 5(a) supercooling degree was observed to reduce by 85%, when the same quantity of MWCNT was added to 24 wt.% barium chloride dehydrate solution, supercooling degree was reduced by 92% as shown in Fig. 5(b). Fig. 6 shows supercooling for different mass fraction for MWCNTs and 24% barium chloride dehydrate. When nanoparticles are added into the base fluid, the supercooling degree is reduced. This phenomenon can be clearly explained by the mechanism of heterogeneous nucleation. The energy barrier for a cluster to be overcome before it grows irreversibly as a crystal can be defined Eq. (1) [26] as
DG ¼
16pg 3sl 1 : 23cosucos3 u 3 Dg 2 4
ð1Þ
Where DG* is gibbs free energy, gSL is the interface solid–liquid free energy, Dg is the free energy difference per particle between liquid and crystalloid. The factor 16/3 is related to the shape of the nucleus, which is assumed to be spherical and u is the contact angle. From Eq. (1) it can be clearly seen that, the reduction of u reduces DG*, if the u is equal to zero degree, the effect of nucleation is the highest. The relation between u and surface energy is defined in Eq. (2) as cosu ¼
dLB dSB dLS
ð2Þ
Where dLB is the specific surface free energy between nanoparticle and fluid; dSB is the specific surface free energy between crystal nucleus and nanoparticle; dLS is the specific surface free energy between crystal nucleus and fluid. The relationship between the forces is shown in Fig. 4. If dSB is less, the cos u tends to one, and the DG* also tends to zero. The crystal nucleus can be formed nearly without supercooling degree. In accordance with the principle of similar structure and corresponding dimension, the crystal face structure of nanoparticles and base fluid is similar and the surface free energy dSB will be less.For barium chloride dehydrates solutions, the nucleation is mainly heterogeneous and the critical
Table 1 Physical parameters of experimental materials [24]. Material
Specific heat (kJ kg1 K1)
Latent heat of fusion(kJ kg1)
Purity (%)
Density(g cm3)
Size of nanoparticle
Thermal conductivity (Wm1 K1)
Deionized water BaCl22H2O MgO MWCNTs
4.182 – 0.88 0.63
330 – – –
100 99.5 95.5 99.5
0.998 3.098 3.58 0.35
– – Diameter 40 nm Inner dia. 5–10 nm Outer dia. 10–20 nm Length. 10–20 mm
0.602 – 48 1500
50
J.M. Munyalo et al. / Journal of Energy Storage 17 (2018) 47–55
Fig. 3. Super cooling of 24 wt% barium chloride dehydrate (a) different cooling temperatures;(b)freezing cycles.
size of barium chloride dehydrate molecule clusters determines whether or not the cluster can grow to a solid. The internal surface of sample holder provides a surface for heterogeneous nucleation highly depending on the wet angle barium chloride dehydrate. Addition of nanoparticles provides extra surfaces for heterogeneous nucleation. Besides, the size of the nanoparticles is larger than that of barium chloride dehydrate and could be larger than the critical size for solid crystals to grow, which makes it more easily for the size of the barium chloride dehydrate crystals on the nanoparticles to exceed the critical size. Therefore, the nanoparticles provide nucleus on which barium chloride dehydrate solutions molecule accumulate to form solid crystals, and thus reduce the supercooling. Also, the development of crystalloid depends on its heat transfer, which will release abundant latent heat when it is crystallizing, if the heat cannot be removed, crystallization will be hold-up. The nanofluids have the higher thermal conductivity than the base liquid, therefore, the speed of growth for crystalloid will accelerate and the rate of freezing enhanced greatly. Although some authors have reported that the relationship between the supercooling degree and the loading of the MWCNTs is not clear [27], in this paper there was decrease in supercooling degree as the concentration of MWCNTs increased. The increase in thermal conductivity increases the release of latent heat of fusion hence reducing supercooling degree [28]. 3.2. Thermal conductivity First, thermal conductivity of deionized water was measured at 20 C as 0.602 W/mK. This was done to check the accuracy of the hot disk machine. Then the thermal conductivity of aqueous 24 wt. % barium dehydrate was measured to be 0.6146 W/mK. The thermal conductivity of MgO at a concentration of 0.2–1% had a range of 0.7215 W/mK–0.7913 W/mK while that of MWCNTs with a
Fig. 4. Forces acting on a crystal nucleus.
similar concentrations had a range of between 0.6279 W/mK– 0.6780 W/mK. The variation of thermal conductivity and percentage increase with respect to mass fraction of MgO and MWCNTs are shown in Fig. 7. It is clear that thermal conductivity of MWCNTs barium chloride dehydrate composite was found to be lower than that of MgO barium chloride dehydrate. Despite high thermal conductivity of MWCNTs the reported increase in thermal conductivity is unexpected compared to a composite that contains MgO. In fact Mulat et al. [29] reported a decrease in thermal conductivity of MWCNTs/PCM composite compared to the pure PCM. Wu et al. [27] also reported thermal conducticvity increase of less than 10% with a mass fraction of upto 2.8 wt.% of MWCNT/PCM composite. At very low mass fractions, the multi walled carbon nano tube continuous network has not been fully formed and therefore the thermal conductivity enhancement is cancelled by the high interface resistance. After increasing mass fraction to a certain level, the network is now formed and enhancement in thermal conductivity is achieved. After achieving this mass fraction, thermal conductivity increases linearly with the mass fraction of MWCNTs. For the MWCNTs to have a meaningful increase in thermal conductivity mass fraction should be above 1 wt.% of the PCM composite but a high increase has a negative effect on the PCM latent heat. Another reason of lower thermal conductivity could be the use of a surfactant and also due to the poor phonon coupling at the PCM-filler interfaces and oxidation of MWCNTs in the preparation process [30]. 3.3. Latent heat The latent heat of 24 wt.% aqueous barium chloride dehydrates was measured as 269 kJ/kg. After addition of nanoparticles of mass fraction of between 0.2% 1%, the latent heat of nano composite PCM decreased by 7% upon adding1 wt.% MgO and by 12.3% on addition of 1 wt.% MWCNTs.The decrease in the latent heat of different samples could be caused due to the interaction between barium chloride dehydrate and the nanoparticles. If nanopartcicle and PCM interaction potential is greater than PCM molecules, the latent heat enhancement would increase. The MWCNTs and MgO do not undergo phase change during the freezing temperature of composite PCM which is about 7.5 C therefore, the enthalpies of the nanocomposite PCM are expected to be lower than those of pure PCM. The decrease in latent heat for MWCNTs PCM composite is higher than MgO PCM; the reason could be attributed to the use of surfactant and oxidation during preparation. Another reason probably may be the realignment of
J.M. Munyalo et al. / Journal of Energy Storage 17 (2018) 47–55
51
Fig. 5. Different mass fraction of additive and 24% barium chloride dehydrate (a) MgO; (b) MWCNTs.
Fig. 6. Comparison for different mass fraction for MWCNT and 24% barium chloride dehydrate (a) 0.2 (b) 0.4 (c) 0.6 (d) 0.8 (e) 1.
Fig. 7. Different mass fraction of MWCNTs and MgO and 24% barium chloride dehydrate (a) Effect of nanoparticle on thermal conductivity (b) percentage increase in thermal conductivity.
52
J.M. Munyalo et al. / Journal of Energy Storage 17 (2018) 47–55
Fig. 8. Effects of mass fraction of nanopaticles on latent heat.
molecules of the matrix PCM in the presence of MWCNTs [30]. The variation of latent heat with respect to nanoparticle mass fraction is shown in Fig. 8. Fig. 9 can be used for optimization purpose whereby depending on the requirements, one can trade off between enhanced thermal conductivity and reduction in latent heat of fusion. 3.4. Cool storage and discharge capacity Fig. 10(a) and (b) shows the cooling curve of 24 wt.% Barium Chloride dehydtrate aqueous solution and MWCNT and MgO nanofluids PCMs respectively, with mass fraction of 1 wt.% under the cool thermal energy storage condition. The cooling curves for both nanofluid PCM show a similar trend. In Fig. 10(a), as an example, the curve ABCDFH is the freezing curve of 24 wt.% Barium chloride dehydrate aqueous solutions while the curve ABEG is that of MWCNT nanofluid PCM.As it can be seen, the process of thermal cool storage include the following three/four steps: section AB is sensible heat thermal cool storage of the liquid phase; section BC is supercooling of 24 wt.% BaCl2–2H2O, which is very much reduced in both MWCNT and MgO nanofluid PCMs due to the faster formation of ice crystal on MWCNT and MgO nanoparticles. It can be seen that suspension of nanoparticles in the nanofluids PCMs not only function to enhance heat transfer, but also as a nucleating agent hence reducing supercooling of the nano PCM. Section BE
(for MWCNT–BaCl2–2H2O) and DF (for 24 wt.% BaCl2–2H2O) is latent heat cool thermal energy storage process, where the two samples possess almost equal phase change temperature for both MWCNT and MgO nanofluid PCM; section EG (for MWCNT–BaCl2– 2H2O) and FH (for 24w.% BaCl2–2H2O) is sensible heat cool thermal energy storage of the solid ice phase. It can be observed from Fig. 10 that by adding a mass fraction of nanoparticles in 24 wt .% BaCl2– 2H2O solutions, the end points of latent heat cool thermal energy storage is shifted from F to E for both MWCNT and MgO nanofluid. As shown in Fig. 10(a) and (b), the time of ice storage decreases from 35 mins to 23 mins and from 37minsto 25 mins for MWCNT and MgO nanofluid PCM respectively. This can be attributed to faster nucleation and probably the enhancement of heat transfer due to the high thermal conductivity of the nanofluids PCMs. Fig. 11 illustrates the melting processes of the three samples above. The curve ABCDEF is the general melting curve of 24 wt.% BaCl2–2H2O, MWCNT- BaCl2–2H2O and MgO- BaCl2–2H2O. It can be seen clearly that the melting processes include the following three sections: section AB is the sensible heat of the solid ice, section BD is the latent heat of melting, section DE (for nanofluid) and DF (for 24 wt% BaCl2–2H2O) is the sensible heat of the liquid phase. It is evident in Fig. 11 that the end point of melting process is shifted from F to E by adding a mass fraction of 1 wt.% nanoparticles in 24 wt% BaCl2– 2H2O solution, and the total melting time decreases from 90 min to
Fig. 9. Effects of mass fraction of nanopaticles on latent heat.
J.M. Munyalo et al. / Journal of Energy Storage 17 (2018) 47–55
53
Fig. 10. Cooling curves of BaCl2–2H2O (a) with MWCNT(b) with MgO.
70 mins. The faster melting rate of nanofluid is attributed to a higher thermal conductivity that increases thermal diffusivity. Thermal diffusivity is given by
a¼
l rc p
ð3Þ
Where a is thermal diffusivity, l is thermal conductivity, r is the density and cp is specific heat capacity. It is evident from Eq. (3), that the increase in thermal conductivity results in increase in thermal diffusivity hence the nanofluid responds faster to changes in temperature than pure aqueous barium chloride dehydrates. During melting, the bath temperature was heated together with samples to reflect real life situation unlike in cooling where the freezing temperature is set in advance. As shown in Fig. 10, the cooling dynamics are almost the same due to higher cooling rate unlike in melting process. 3.5. Stability of nanofluids There are many ways of measuring nanofluid stability, among them sedimentation technique is the simplest techniques [31]. A volume of nanofluid sample is kept in glass beaker and photographs are taken at regular time intervals. The first image is taken
immediately after the preparation of sample and it continued until the nanoparticles settled completely in the container. The samples of MgO–Barium chloride dehydrate nanofluid were prepared as mentioned previously with and without the addition of surfactants. The volume fraction of 0.2 and 1 wt.% the nanoparticles were chosen. The relation between sedimentation velocity and size of the particles in stationery state is given by Stokes law as V¼
2R2 rp rl g 9m
ð4Þ
where V is the sedimentation velocity of the particles; R is the radius of the particles; m is the viscosity of the liquid medium; rp and rL are the densities of the particles and liquid medium, respectively; g is the acceleration due to gravity. Stability of the nanofluid can be enhancement by reducing the sedimentation velocity of the nanoparticles. According to Stokes law, the sedimentation velocity can be reduced by using nanoparticles with smaller diameters. However, the smaller the nanoparticles diameter is, the higher the surface energy will be and this will increase possibility of agglomeration. Therefore, there exists a challenge in preparing the nanofluid where a balance has to be struck between decreasing the sedimentation velocity and at the same time suppressing
Fig. 11. Melting curves of MWCNT-BaCl2–2H2O, MgO-BaCl2–2H2O, and BaCl2–2H2O.
54
J.M. Munyalo et al. / Journal of Energy Storage 17 (2018) 47–55
Fig. 12. Nanofluid without surfactant after 24 h (a) 0.2 wt% MgO (b) 1 wt% MgO.
Fig. 13. Nanofluid with surfactant held at 6 C after 24 h (a) 0.2 wt% MgO (b) 1 wt% MgO.
agglomeration. One way of suppressing agglomeration without affecting the sedimentation velocity is the addition of surfactants to the nanofluids. As shown in Fig. 12, the stability of MgO and 24 wt.% barium chloride dehydrate without surfactant showed poor stability after only 24 h. As shown in Fig. 13, when 1 wt.% of SDS added and the solution kept at low constant water bath for 24 h, it showed no separation. The MWCNT nanofluid was covalent fuctionalized by use of N,N-dimethyl formamide and after being kept at a low temperature bath, it showed no separation after 24 h (Fig. 14). 4. Conclusions In this paper, the supercooling, stability and thermal properties of a new nanofluid composite phase change material with two kinds of nanoparticles was developed and investigated. The following conclusion can be drawn;
1. The new phase change material latent heat of fusion and the supercooling degree decrease due to the addition of both MgO and MWCNTs. The latent heat decreases by 12.3% on addition of 1 wt.% MWCNTs due to realignment of molecules of the matrix PCM in the presence of MWCNTs. 2. The thermal conductivity enhancement ratios of the nano PCM increase with the loading of nanofillers however, thermal conductivity enhancement in MWCNT nanofluid was lower than expected. The MgO nanofillers have the greatest thermal conductivity enhancement of up to 17% at mass fraction of 1 wt.%. 3. Stability of the nanofluid at low temperature is also increased by use of surfactant as it was found there was no aggregation. The new nano composite phase change material can be used in some industries such as water ice making plant which require the cooling temperature to be below zero.
J.M. Munyalo et al. / Journal of Energy Storage 17 (2018) 47–55
55
Fig. 14. Nanofluid with surfactant held at 6 C after 24 h (a) 0.2 wt.% MWCNT (b) 1 wt.% MWCNT.
Acknowledgement This is study benefits from financial support of the Key Program of Shanghai Science and Technology Commission (Grant No. 16040501600). References [1] V. Kumaresan, P. Chandrasekaran, N. Maitreyee, A.K. Maini, R. Velraj, Role of PCM based nanofluids for energy efficient cool thermal storage system, Int. J. Refrig. 36 (2013) 1641–1647. [2] R.K. Sharma, P. Ganesan, V.V. Tyagi, H.S.C. Metselaar, S.C. Sandaran, Developments in organic solid–liquid phase change materials and their applications in thermal energy storage, Energy Convers. Manage. 95 (2015) 193–228. [3] G. Li, D. Liu, Y. Xie, Study on thermal properties of TBAB–THF hydrate mixture for cold storage by DSC, J. Therm. Anal. Calorim. 102 (2010) 819–826. [4] L. Gang, H. Yunho, R. Reinhard, Review of cold storage materials for air conditioning application, Int. J. Refrig. 35 (2012) 2053–2127. [5] E. Oró, A. Gracia, A. Castell, M.M. Farid, L.F. Cabeza, Review on phase change materials (PCMs) for cold thermal energy storage applications, Appl. Energy 99 (2012) 513–533. [6] J. Lisi, P. Lan, C. Ying, M. Songping, L. Xing, Improving the supercooling degree of titanium dioxide nanofluids with sodium dodecylsulfate, Appl. Energy 124 (2014) 248–255. [7] Refat Al-Shannaq, Jamal Kurdi, Shaheen Al-Muhtaseb, Michelle Dickinson, Mohammed Farid, Supercooling elimination of phase change materials (PCMs) microcapsules, Energy 87 (2015) 654–662. [8] B. Sandnes, J. Rekstad, Supercooling salt hydrates: stored enthalpy as a function of temperature, Sol. Energy 80 (2006) 616–625. [9] T.X. Li, D.L. Wu, F. He, R.Z. Wang, Experimental investigation on copper foam/ hydrated salt composite phase change material for thermal energy storage, Int. J. Heat Mass Transfer 115 (2017) 148–157. [10] Y. Fan, C. Yingying, L. Xingbo, X. Jiale, L. Chiahsun, L. Qi, T. Peng, D. Tao, Dispersion stability of thermal nanofluids, Prog. Nat. Sci.: Mater. Int. 27 (2017) 531–542. [11] Florian Kleiner, Konrad Posern, Andrea Osburg, Thermal conductivity of selected salt hydrates for thermochemical solar heat storage applications measured by the light flash method, Appl. Therm. Eng. 113 (2017) 1189–1193. [12] S.L. Che, P.P. Wang, T.S. Lee, An experimental investigation of nucleation probability of supercooled water inside cylindrical capsules, Exp. Therm. Fluid Sci. 18 (1998) 299–306. [13] A. Safari, R. Saidur, F.A. Sulaiman, X. Yan, J. Dong, A review on supercooling of Phase Change Materials in thermal energy storage systems, Renew. Sustain. Energy Rev. 70 (2017) 905–919. [14] Mohamed M. Tawfik, Experimental studies of nanofluid thermal conductivity enhancement and applications: a review, Renew. Sustain. Energy Rev. 75 (2017) 1239–1253. [15] L. Yudong, L. Xin, H. Pengfei, H. Guanghua, Study on the supercooling degree and nucleation behavior of water-based graphene oxide nanofluids PCM, Int. J. Refrig. 50 (2015) 80–86.
[16] W.L. Cui, L.L. YuanY.P. Sun, X.L. Cao, X.J. Yang, Experimental studies on the supercooling and melting/freezing characteristics of nano- copper/sodium acetate trihydrate composite phase change materials, Renew. Energy 99 (2016) 1029–1037. [17] H. Qinbo, W. Shuangfeng, T. Mingwei, L. Yudong, Experimental study on thermophysical properties of nanofluids as phase-change material (PCM) in low temperature, Energy Convers. Manage. 64 (2012) 199–205. [18] Y.D. Liu, J.Q. Wang, C.J. Su, S.C. Geng, Y.K. Gao, Q.G. Peng, Nucleation rate and supercooling degree of water-based graphene oxide nanofluids, Appl. Therm. Eng. 115 (2017) 1226–1236. [19] M.A. Khairul, S. Kalpit, D. Elham, A. Reza, M. Behdad, Effects of surfactant on stability and thermo-physical properties of metal oxide nanofluids, Int. J. Heat Mass Transfer 98 (2016) 778–787. [20] L. Xiaoke, Z. Changjun, L. Xinyu, L. Wenliang, Stability and enhanced thermal conductivity of ethylene glycol-based SiC nanofluids, Int. J. Heat Mass Transfer 89 (2015) 613–619. [21] L. Yudong, L. Xin, H. Pengfei, H. Guanghua, Study on the supercooling degree and nucleation behavior of water-based graphene oxide nanofluids PCM, Int. Refrig. 50 (2015) 80–86. [22] B.C. Jean-Pierre, D. Thomas, C.L. Jean, Ice slurry production using supercooling phenomenon, Int. Refrig. 331 (2010) 196–204. [23] P. Chandrasekaran, M. Cheralathan, R. Velraj, Effect of fill volume on solidification characteristics of DI (deionized)water in a spherical capsule: an experimental study, Energy 90 (2015) 508–515. [24] S. Omid, A. Mohammad, Effects of temperature and particles concentration on the dynamic viscosity of MgO-MWCNT/ethylene glycol hybrid Nano fluid: experimental study, Phys. E 84 (2016) 564–570. [25] S. Anant, D. Buddhi, R.L. Sawhney, Thermal cycling test of few selected inorganic and organic phase change materials, Renew. Energy 33 (2008) 2606– 2614. [26] I. Yuji, S. Masahiro, K. Masaaki, I. Kohei, Theoretical analysis of supercooled states of water generated below the freezing point in a PEFC, Int. J. Heat Mass Transfer 74 (2014) 215–227. [27] S.Y. Wu, X. Tong, C.D. Nie, D.Q. Peng, S.G. Gong, Z.Q. Wang, The effects of various carbon nanofillers on the thermal properties of paraffin for energy storage applications, J. Therm. Anal. Calorim. 124 (2016) 181–188. [28] V. Kumaresa, R. Velraj, N. Maitreyee, A.K. Maini, Phase change material-based nanofluids for heat transfer enhancement in latent heat thermal energy storage system, Int. J. Green Nanotechnol. 4 (2012) 541–546. [29] P. Murat, K. Sengül, N.T. Ümit, Y. Kerim, Thermal property investigation of multi walled carbon nanotubes (MWCNTs)Embedded phase change materials (PCMs), Therm. Thermomech. Phenom. Electron. Syst. 15 (2016) 639–644. [30] F. Xin, W.F. Li, D. Qing, W. Xiao, L.Y. Xiao, F.H. Jian, T.Y. Zi, H.C. Guan, C.H. Ya, Ke FC.Increased thermal conductivity of eicosane-based composite phase change materials in the presence of graphene nanoplatelets, Energy Fuels 27 (2013) 4041–4047. [31] W. Baojie, Z. Changjun, L. Xiaoke, Experimental investigation on stability and thermal conductivity of diathermic oil based TiO2 nanofluids, Int. J. Heat Mass Transfer 104 (2017) 537–543.
Contents lists available at ScienceDirect
Journal of Energy Storage journal homepage: www.elsevier.com/locate/est
Experimental investigation on supercooling, thermal conductivity and stability of nanofluid based composite phase change material Munyalo Jotham Muthoka, Zhang Xuelai* , Xu Xioafeng Institute of Thermal Storage Technology, Merchant Marine College, Shanghai Maritime University, Shanghai, 201306, China
A R T I C L E I N F O
Article history: Received 22 September 2017 Received in revised form 28 November 2017 Accepted 9 February 2018 Available online 24 February 2018 Keywords: Phase change material Thermal conductivity Supercooling Aggregation Barium chloride dehydrate
A B S T R A C T
With increasing concerns over global warming, there is compelling need to apply energy technologies to improve energy efficiencies. One of the new technologies is the use of phase change materials (PCMs) to store energy and release it on demand. However, most of these materials undergo supercooling, aggregation and have low thermal conductivity that inhibits effective heat transfer. In this paper supercooling, stability, energy storage,thermal conductivity and latent heat of fusion of water based nanofluid of barium chloride dehydrate (BaCl22H2O) were experimentally studied by adding a mass fraction of 0.2 w.t%–1 wt.% magnesium oxide (MgO) and 0.2 wt.%–1 wt.% multi-walled carbon nanotubes (MWCNTs). Results show that by adding separately mass fraction of 1.% of MgO and MWCNTs reduce the supercooling degree of barium chloride dehydrate by 85% and 92% respectively. At the same mass fraction, thermal conductivity also improves by 6% on addition of MWCNTs and 17% on addition of MgO. However, the enthalpies reduce by 7% at 1 wt.% MgO and by 12.3% at 1 wt% MWCNTs. It was found that MgO exhibited relatively higher thermal conductivity and a lower reduction in latent heats at a mass fraction of 1 wt.%. Surfactant was also found to prevent aggregation at low temperatures. © 2018 Elsevier Ltd. All rights reserved.
1. Introduction Cool thermal energy storage (CTES) technology, which is storing cool thermal energy in order to bridge the time gap between the energy availability and energy use is being considered as one of the best method for thermal energy management [1,2]. The CTES is useful in central air-conditioning systems in large buildings [3,4], supermarkets refrigeration, high powered electronic cooling applications, marine applications, pharmaceutical field and various industrial cooling process applications where the cooling requirement is highly intermittent. The efficient operation of CTES system reduces the additional energy consumption during its cyclic charging and discharging operations. Therefore, PCMs have been applied to increase thermal energy storage capacity of different systems [5]. The use of PCMs provides higher heat storage capacity and a constant temperature behavior during charging and discharging compared to sensible heat storage. High energy storage density is a desirable property of any storage system which guarantees smaller size for the storage system. In the application of PCMs, the most used phase change is solid–liquid
* Corresponding author. Tel.: +86 13127992577; fax: +86 02138282925. E-mail addresses: [email protected] (J.M. Munyalo), [email protected] (X. Zhang). https://doi.org/10.1016/j.est.2018.02.006 2352-152X/© 2018 Elsevier Ltd. All rights reserved.
phase change as the volume is easily controlled. The substances used can be organic such as paraffin and fatty acids or inorganic such as aqueous salts solutions. Water based inorganic salts are widely used as a PCMs because of low cost, readily available and environmental friendly [6]. However, the most serious problem encountered when using these PCMs is the supercooling phenomenon during solidification [7,8], aggregation at low temperature [9,10] and a low thermal conductivity [11]. A lower degree of supercooling results in a better coefficient of performance of the refrigeration system [12] and a high thermal conductivity reduces the solidification and melting time of PCMs. In recent years, many researchers have studied ways of reducing supercooling and at the same time increase thermal conductivity [13,14]. One of the ways is to use nanoparticles to make nanofluid which can be used as a PCMs. Researchers have pointed out that nanostructures can reduce the supercooling degree and increase thermal conductivity [13,15,16]. Qinbo et al. [17] carried out thermal conductivity, supercooling degree, latent heat, specific heat, and rheological behaviors of nano PCMs by suspending a small amount of TiO2 nanoparticles in saturated barium chloride aqueous solution; his experimental results showed that with volume fraction of 1.130%, the thermal conductivities of nanofluids PCMs were enhanced by 12.76% at 5 C and the supercooling degree was reduced by 84.92% but the latent heat and specific heat were slightly decreased. Yudong et al. [15] carried out an experimental study
48
J.M. Munyalo et al. / Journal of Energy Storage 17 (2018) 47–55
temperature thermostat bath has a temperature range of 65 C to 100 C; its precision is 0.05 C. For the purpose of this experiment, the low temperature thermostat bath’s temperature was set at 20 C for freezing and 20 C for melting. The T-type thermocouples which were used to take temperature measurements at various points have a precision of 0.5 C and thermal response period of 0.4 s.
on the supercooling degree and nucleation behavior of nanofluids phase change material which were prepared by adding small fraction of graphene oxide nanosheets in deionized water without any dispersants. The results showed that the supercooling degree was reduced by 69.1%, and the nucleation started in advance, reducing the time by 90.7%. Liu et al. [18] carried out a research on nucleation rates and supercooling degree of deionized water and graphene oxide nanofluids using classical nucleation theory. He found that whereas the supercooling degree of deionized water was approximately 31.5 K, the supercooling degrees of four different concentrations of graphene oxide nanofluids were 7.98, 7.93, 3.05, and 3.03 K concluding that the supercooling degrees were reduced by more than 74% with the corresponding increase in volume concentration. Stability of nanofluid has also been carried out by many researchers [19,20] where the focus was on stability with respect to pH, zeta potential, particle size distribution, and its effect on viscosity and thermal conductivity. Most of these previous researchers’ mainly focused on the nanofluid PCMs used in the air conditioning cool storage, where the phase change temperatures are above 0 C. However, some enterprises, such as ice and chemical plant, need the cool storage temperature below 0 C. At the moment there are few researches and applications for nanofluid PCMs in sub-zero temperature cool storage available [5]. In this paper, considering the application in water ice making plant where the temperature is between 1 C to 0 C. While the phasechange temperature of 24 wt.% BaCl22H2O solution is about 7.5 C and considering heat losses together with overcoming supercooling problem of water in some cases [21–23], the 7.5 C temperature difference is appropriate. Compared with other phase change materials that have found use in sub-zero temperature application such NaCl–H2O and CaCl–H2O, barium chloride dehydrate is found to have the highest latent heat of fusion. So it was screened out as a suitable candidate for PCMs. The main focus in this paper was to experimentally study supercooling, stability, latent heat of fusion and thermal conductivity of barium chloride dehydrates under the effect of different cooling temperatures, freeze cycling and nano particles concentration of MgO and MWCNTs. Nanofluids were prepared using a two-step method. N, N-dimethyl formamide was used as surfactant to stabilize the dispersion of MWCNTs nanoparticles in aqueous barium chloride dehydrate. SDS surfactant was used for MgO nanofluid to enhance its stability.
In order to evaluate how different factors influence supercooling degree of water based barium chloride dehydrate, different cooling temperatures and nanoparticle concentrations were used. The supercooling characteristic test was carried out using constant temperature bath (DC 6115, Shanghai Hengping Equipment Company). To avoid the contamination of barium chloride dehydrate solution which may interfere with the supercooling degree and thermal conductivity measurements, different beakers, glass rod and tweezers were placed into the ultrasonic machine to clean them by means of ultrasonic vibrations. Polyurethane sealing covers were used to cover the tops of the beakers to avoid the content from being contaminated by external environment. The thermocouples wires pass through the polyurethane sealing covers and stretch into the solution with thermocouple probes being placed at the mid position of the beaker where temperature is taken. An electronic balance (FA2004) which was used to measure the weights of experimental materials has a precision of 0.1 mg. Thermal conductivities of nanofluids were measured by the Hot disk (TPS 500, Thermal Constant Analyser,TCA, Sweden) where KS-1 probe sensor was used. Calibration of instrument with DI water was performed before starting off the measurements of thermal conductivities of the nanofluids. Thermal conductivity of DI water at 20 C was found to be 0.602 W/mK. The latent heat of the base fluids and the nanofluids were measured with a differential scanning calorimeter (DSC 200, NETZSCH, German) with an accuracy of 1.0%. The DSC instrument was first calibrated using a standard high-purity indium specimen prior to use. For each sample at least five measurements were carried out and average value with a standard deviation of less than 1% was given as a result. To test stability of nanofluid, a volume of nanofluid sample was kept in glass beaker and photographs were taken at regular time interval.
2. Material and methods
2.3. Preparation of nanofluid
2.1. Experimental setup
A BaCl22H2O (produced by Sinopharm Chemical Reagent Co, Ltd purity 99.5%) without additives was prepared by dissolving it in deionized water to make an aqueous solution of 24 wt.% BaCl22H2O. The solution was evenly stirred in order to ensure all the barium chloride dehydrate dissolves. The nanofluid was prepared by a two-step method. The nanoparticle has a bigger specific surface area and surface energy because of its small size. It is apt to reunite in the process of preparation, reprocessing and the application therefore obtaining a well-distributed and stabilized nanofluids is a critical step. Five samples of 0.2, 0.4, 0.6, 0.8 and 1 wt.% MgO and 1 wt.% of SDS as surfactant (Supplied by Aladdin Industrial Corporation, Shanghai, China) were prepared for experiment. First the solution was stirred in a magnetic stirrer for 30 min and a then ultrasonicated for 45 min at temperature of 50 C, with a working time of 5 s and stop time of 10 s. The same mass fraction was used to prepare MWCNTs (Supplied by Aladdin Industrial Corporation,Shanghai,China) nanofluid. 1 wt.% of N,Ndimethyl formamide, having a molecular formular C3H7NO (Supplied by Aladdin Industrial Corporation,Shanghai,China) was first put into each five samples then stirred in a magnetic stirrer (HJ-6A,West Jintan Zhengrong Experiment Equipment Company,
The experiment was set up as shown in Fig. 1. It consists of agilent data logger (34972A), computer, low-temperature thermostatic bath (DC-6515, supplied by Shanghai HengPing Instrument Factory), thermocouples and polyurethane sealing cover. The low-
Fig. 1. Experimental setup. 1-computer; 2-agilent data logger; 3-thermocouple; 4-polyurethane sealing cover; 5-beaker; 6-nanofluid PCM; 7-bracket; 8-low- temperature thermostat bath.
2.2. Experimental technique
J.M. Munyalo et al. / Journal of Energy Storage 17 (2018) 47–55
49
Fig. 2. Magnetic stirring of Nano fluid PCM.
China) before adding the MWCNTs and stirring further for 30 min. The solutions were then transferred to ultrasonic vibrator for 45 min with the same setting as for MgO nanofluid. Samples of prepared MWCNTs are shown in Fig. 2 and the properties of the nanoparticles are shown in Table 1. 3. Results and discussion 3.1. Supercooling As shown in Fig. 3(a), when 24 wt.% BaCl22H2O solutions having a melting point of 7.5 C was exposed to different cooling temperatures, the highest supercooling degrees occurred at temperatures of 15 C and 20 C with 6 C and 6.5 C respectively. At 35 C, no supercooling that was witnessed with the other intermediate temperatures having very minimal supercooling degree of less than 0.5 C. This generally shows that supercooling degree of aqueous barium chloride dehydrate is affected by the cooling temperature. The higher the cooling temperature is, the higher the supercooling degree and the lower the cooling temperature is, the lower supercooling degree. This is attributed to the fact that nucleus forms very fast when there is high heat removal but when there is low heat removal the nucleus forms very slowly hence low nucleation. By using low temperatures to charge low temperature PCM, the supercooling degree can be eliminated completely. It is obvious that when charging in low temperature, charging time is greatly reduced as it takes shorter time to charge in low temperature than in high temperature though one has to strike a balance between cost and benefit as low temperature comes at extra cost. As shown in Fig. 3(b),when aqueous barium chloride dehydrate underwent four freezing and melting cycles at a cooling temperature of 20 C, supercooling degree was reduced from 6 C at the first cycle to about 1 C at the fourth cycle. The reason for reduction in supercooling degree could have been that during subsequent freezing some parts of the solution could have been already supercooled hence prompting quicker formation of nucleus. This is in good agreement with findings in
the literature [25]. When 0.2–1% mass fraction of MgO was added to barium chloride dehydrate as shown in Fig. 5(a) supercooling degree was observed to reduce by 85%, when the same quantity of MWCNT was added to 24 wt.% barium chloride dehydrate solution, supercooling degree was reduced by 92% as shown in Fig. 5(b). Fig. 6 shows supercooling for different mass fraction for MWCNTs and 24% barium chloride dehydrate. When nanoparticles are added into the base fluid, the supercooling degree is reduced. This phenomenon can be clearly explained by the mechanism of heterogeneous nucleation. The energy barrier for a cluster to be overcome before it grows irreversibly as a crystal can be defined Eq. (1) [26] as
DG ¼
16pg 3sl 1 : 23cosucos3 u 3 Dg 2 4
ð1Þ
Where DG* is gibbs free energy, gSL is the interface solid–liquid free energy, Dg is the free energy difference per particle between liquid and crystalloid. The factor 16/3 is related to the shape of the nucleus, which is assumed to be spherical and u is the contact angle. From Eq. (1) it can be clearly seen that, the reduction of u reduces DG*, if the u is equal to zero degree, the effect of nucleation is the highest. The relation between u and surface energy is defined in Eq. (2) as cosu ¼
dLB dSB dLS
ð2Þ
Where dLB is the specific surface free energy between nanoparticle and fluid; dSB is the specific surface free energy between crystal nucleus and nanoparticle; dLS is the specific surface free energy between crystal nucleus and fluid. The relationship between the forces is shown in Fig. 4. If dSB is less, the cos u tends to one, and the DG* also tends to zero. The crystal nucleus can be formed nearly without supercooling degree. In accordance with the principle of similar structure and corresponding dimension, the crystal face structure of nanoparticles and base fluid is similar and the surface free energy dSB will be less.For barium chloride dehydrates solutions, the nucleation is mainly heterogeneous and the critical
Table 1 Physical parameters of experimental materials [24]. Material
Specific heat (kJ kg1 K1)
Latent heat of fusion(kJ kg1)
Purity (%)
Density(g cm3)
Size of nanoparticle
Thermal conductivity (Wm1 K1)
Deionized water BaCl22H2O MgO MWCNTs
4.182 – 0.88 0.63
330 – – –
100 99.5 95.5 99.5
0.998 3.098 3.58 0.35
– – Diameter 40 nm Inner dia. 5–10 nm Outer dia. 10–20 nm Length. 10–20 mm
0.602 – 48 1500
50
J.M. Munyalo et al. / Journal of Energy Storage 17 (2018) 47–55
Fig. 3. Super cooling of 24 wt% barium chloride dehydrate (a) different cooling temperatures;(b)freezing cycles.
size of barium chloride dehydrate molecule clusters determines whether or not the cluster can grow to a solid. The internal surface of sample holder provides a surface for heterogeneous nucleation highly depending on the wet angle barium chloride dehydrate. Addition of nanoparticles provides extra surfaces for heterogeneous nucleation. Besides, the size of the nanoparticles is larger than that of barium chloride dehydrate and could be larger than the critical size for solid crystals to grow, which makes it more easily for the size of the barium chloride dehydrate crystals on the nanoparticles to exceed the critical size. Therefore, the nanoparticles provide nucleus on which barium chloride dehydrate solutions molecule accumulate to form solid crystals, and thus reduce the supercooling. Also, the development of crystalloid depends on its heat transfer, which will release abundant latent heat when it is crystallizing, if the heat cannot be removed, crystallization will be hold-up. The nanofluids have the higher thermal conductivity than the base liquid, therefore, the speed of growth for crystalloid will accelerate and the rate of freezing enhanced greatly. Although some authors have reported that the relationship between the supercooling degree and the loading of the MWCNTs is not clear [27], in this paper there was decrease in supercooling degree as the concentration of MWCNTs increased. The increase in thermal conductivity increases the release of latent heat of fusion hence reducing supercooling degree [28]. 3.2. Thermal conductivity First, thermal conductivity of deionized water was measured at 20 C as 0.602 W/mK. This was done to check the accuracy of the hot disk machine. Then the thermal conductivity of aqueous 24 wt. % barium dehydrate was measured to be 0.6146 W/mK. The thermal conductivity of MgO at a concentration of 0.2–1% had a range of 0.7215 W/mK–0.7913 W/mK while that of MWCNTs with a
Fig. 4. Forces acting on a crystal nucleus.
similar concentrations had a range of between 0.6279 W/mK– 0.6780 W/mK. The variation of thermal conductivity and percentage increase with respect to mass fraction of MgO and MWCNTs are shown in Fig. 7. It is clear that thermal conductivity of MWCNTs barium chloride dehydrate composite was found to be lower than that of MgO barium chloride dehydrate. Despite high thermal conductivity of MWCNTs the reported increase in thermal conductivity is unexpected compared to a composite that contains MgO. In fact Mulat et al. [29] reported a decrease in thermal conductivity of MWCNTs/PCM composite compared to the pure PCM. Wu et al. [27] also reported thermal conducticvity increase of less than 10% with a mass fraction of upto 2.8 wt.% of MWCNT/PCM composite. At very low mass fractions, the multi walled carbon nano tube continuous network has not been fully formed and therefore the thermal conductivity enhancement is cancelled by the high interface resistance. After increasing mass fraction to a certain level, the network is now formed and enhancement in thermal conductivity is achieved. After achieving this mass fraction, thermal conductivity increases linearly with the mass fraction of MWCNTs. For the MWCNTs to have a meaningful increase in thermal conductivity mass fraction should be above 1 wt.% of the PCM composite but a high increase has a negative effect on the PCM latent heat. Another reason of lower thermal conductivity could be the use of a surfactant and also due to the poor phonon coupling at the PCM-filler interfaces and oxidation of MWCNTs in the preparation process [30]. 3.3. Latent heat The latent heat of 24 wt.% aqueous barium chloride dehydrates was measured as 269 kJ/kg. After addition of nanoparticles of mass fraction of between 0.2% 1%, the latent heat of nano composite PCM decreased by 7% upon adding1 wt.% MgO and by 12.3% on addition of 1 wt.% MWCNTs.The decrease in the latent heat of different samples could be caused due to the interaction between barium chloride dehydrate and the nanoparticles. If nanopartcicle and PCM interaction potential is greater than PCM molecules, the latent heat enhancement would increase. The MWCNTs and MgO do not undergo phase change during the freezing temperature of composite PCM which is about 7.5 C therefore, the enthalpies of the nanocomposite PCM are expected to be lower than those of pure PCM. The decrease in latent heat for MWCNTs PCM composite is higher than MgO PCM; the reason could be attributed to the use of surfactant and oxidation during preparation. Another reason probably may be the realignment of
J.M. Munyalo et al. / Journal of Energy Storage 17 (2018) 47–55
51
Fig. 5. Different mass fraction of additive and 24% barium chloride dehydrate (a) MgO; (b) MWCNTs.
Fig. 6. Comparison for different mass fraction for MWCNT and 24% barium chloride dehydrate (a) 0.2 (b) 0.4 (c) 0.6 (d) 0.8 (e) 1.
Fig. 7. Different mass fraction of MWCNTs and MgO and 24% barium chloride dehydrate (a) Effect of nanoparticle on thermal conductivity (b) percentage increase in thermal conductivity.
52
J.M. Munyalo et al. / Journal of Energy Storage 17 (2018) 47–55
Fig. 8. Effects of mass fraction of nanopaticles on latent heat.
molecules of the matrix PCM in the presence of MWCNTs [30]. The variation of latent heat with respect to nanoparticle mass fraction is shown in Fig. 8. Fig. 9 can be used for optimization purpose whereby depending on the requirements, one can trade off between enhanced thermal conductivity and reduction in latent heat of fusion. 3.4. Cool storage and discharge capacity Fig. 10(a) and (b) shows the cooling curve of 24 wt.% Barium Chloride dehydtrate aqueous solution and MWCNT and MgO nanofluids PCMs respectively, with mass fraction of 1 wt.% under the cool thermal energy storage condition. The cooling curves for both nanofluid PCM show a similar trend. In Fig. 10(a), as an example, the curve ABCDFH is the freezing curve of 24 wt.% Barium chloride dehydrate aqueous solutions while the curve ABEG is that of MWCNT nanofluid PCM.As it can be seen, the process of thermal cool storage include the following three/four steps: section AB is sensible heat thermal cool storage of the liquid phase; section BC is supercooling of 24 wt.% BaCl2–2H2O, which is very much reduced in both MWCNT and MgO nanofluid PCMs due to the faster formation of ice crystal on MWCNT and MgO nanoparticles. It can be seen that suspension of nanoparticles in the nanofluids PCMs not only function to enhance heat transfer, but also as a nucleating agent hence reducing supercooling of the nano PCM. Section BE
(for MWCNT–BaCl2–2H2O) and DF (for 24 wt.% BaCl2–2H2O) is latent heat cool thermal energy storage process, where the two samples possess almost equal phase change temperature for both MWCNT and MgO nanofluid PCM; section EG (for MWCNT–BaCl2– 2H2O) and FH (for 24w.% BaCl2–2H2O) is sensible heat cool thermal energy storage of the solid ice phase. It can be observed from Fig. 10 that by adding a mass fraction of nanoparticles in 24 wt .% BaCl2– 2H2O solutions, the end points of latent heat cool thermal energy storage is shifted from F to E for both MWCNT and MgO nanofluid. As shown in Fig. 10(a) and (b), the time of ice storage decreases from 35 mins to 23 mins and from 37minsto 25 mins for MWCNT and MgO nanofluid PCM respectively. This can be attributed to faster nucleation and probably the enhancement of heat transfer due to the high thermal conductivity of the nanofluids PCMs. Fig. 11 illustrates the melting processes of the three samples above. The curve ABCDEF is the general melting curve of 24 wt.% BaCl2–2H2O, MWCNT- BaCl2–2H2O and MgO- BaCl2–2H2O. It can be seen clearly that the melting processes include the following three sections: section AB is the sensible heat of the solid ice, section BD is the latent heat of melting, section DE (for nanofluid) and DF (for 24 wt% BaCl2–2H2O) is the sensible heat of the liquid phase. It is evident in Fig. 11 that the end point of melting process is shifted from F to E by adding a mass fraction of 1 wt.% nanoparticles in 24 wt% BaCl2– 2H2O solution, and the total melting time decreases from 90 min to
Fig. 9. Effects of mass fraction of nanopaticles on latent heat.
J.M. Munyalo et al. / Journal of Energy Storage 17 (2018) 47–55
53
Fig. 10. Cooling curves of BaCl2–2H2O (a) with MWCNT(b) with MgO.
70 mins. The faster melting rate of nanofluid is attributed to a higher thermal conductivity that increases thermal diffusivity. Thermal diffusivity is given by
a¼
l rc p
ð3Þ
Where a is thermal diffusivity, l is thermal conductivity, r is the density and cp is specific heat capacity. It is evident from Eq. (3), that the increase in thermal conductivity results in increase in thermal diffusivity hence the nanofluid responds faster to changes in temperature than pure aqueous barium chloride dehydrates. During melting, the bath temperature was heated together with samples to reflect real life situation unlike in cooling where the freezing temperature is set in advance. As shown in Fig. 10, the cooling dynamics are almost the same due to higher cooling rate unlike in melting process. 3.5. Stability of nanofluids There are many ways of measuring nanofluid stability, among them sedimentation technique is the simplest techniques [31]. A volume of nanofluid sample is kept in glass beaker and photographs are taken at regular time intervals. The first image is taken
immediately after the preparation of sample and it continued until the nanoparticles settled completely in the container. The samples of MgO–Barium chloride dehydrate nanofluid were prepared as mentioned previously with and without the addition of surfactants. The volume fraction of 0.2 and 1 wt.% the nanoparticles were chosen. The relation between sedimentation velocity and size of the particles in stationery state is given by Stokes law as V¼
2R2 rp rl g 9m
ð4Þ
where V is the sedimentation velocity of the particles; R is the radius of the particles; m is the viscosity of the liquid medium; rp and rL are the densities of the particles and liquid medium, respectively; g is the acceleration due to gravity. Stability of the nanofluid can be enhancement by reducing the sedimentation velocity of the nanoparticles. According to Stokes law, the sedimentation velocity can be reduced by using nanoparticles with smaller diameters. However, the smaller the nanoparticles diameter is, the higher the surface energy will be and this will increase possibility of agglomeration. Therefore, there exists a challenge in preparing the nanofluid where a balance has to be struck between decreasing the sedimentation velocity and at the same time suppressing
Fig. 11. Melting curves of MWCNT-BaCl2–2H2O, MgO-BaCl2–2H2O, and BaCl2–2H2O.
54
J.M. Munyalo et al. / Journal of Energy Storage 17 (2018) 47–55
Fig. 12. Nanofluid without surfactant after 24 h (a) 0.2 wt% MgO (b) 1 wt% MgO.
Fig. 13. Nanofluid with surfactant held at 6 C after 24 h (a) 0.2 wt% MgO (b) 1 wt% MgO.
agglomeration. One way of suppressing agglomeration without affecting the sedimentation velocity is the addition of surfactants to the nanofluids. As shown in Fig. 12, the stability of MgO and 24 wt.% barium chloride dehydrate without surfactant showed poor stability after only 24 h. As shown in Fig. 13, when 1 wt.% of SDS added and the solution kept at low constant water bath for 24 h, it showed no separation. The MWCNT nanofluid was covalent fuctionalized by use of N,N-dimethyl formamide and after being kept at a low temperature bath, it showed no separation after 24 h (Fig. 14). 4. Conclusions In this paper, the supercooling, stability and thermal properties of a new nanofluid composite phase change material with two kinds of nanoparticles was developed and investigated. The following conclusion can be drawn;
1. The new phase change material latent heat of fusion and the supercooling degree decrease due to the addition of both MgO and MWCNTs. The latent heat decreases by 12.3% on addition of 1 wt.% MWCNTs due to realignment of molecules of the matrix PCM in the presence of MWCNTs. 2. The thermal conductivity enhancement ratios of the nano PCM increase with the loading of nanofillers however, thermal conductivity enhancement in MWCNT nanofluid was lower than expected. The MgO nanofillers have the greatest thermal conductivity enhancement of up to 17% at mass fraction of 1 wt.%. 3. Stability of the nanofluid at low temperature is also increased by use of surfactant as it was found there was no aggregation. The new nano composite phase change material can be used in some industries such as water ice making plant which require the cooling temperature to be below zero.
J.M. Munyalo et al. / Journal of Energy Storage 17 (2018) 47–55
55
Fig. 14. Nanofluid with surfactant held at 6 C after 24 h (a) 0.2 wt.% MWCNT (b) 1 wt.% MWCNT.
Acknowledgement This is study benefits from financial support of the Key Program of Shanghai Science and Technology Commission (Grant No. 16040501600). References [1] V. Kumaresan, P. Chandrasekaran, N. Maitreyee, A.K. Maini, R. Velraj, Role of PCM based nanofluids for energy efficient cool thermal storage system, Int. J. Refrig. 36 (2013) 1641–1647. [2] R.K. Sharma, P. Ganesan, V.V. Tyagi, H.S.C. Metselaar, S.C. Sandaran, Developments in organic solid–liquid phase change materials and their applications in thermal energy storage, Energy Convers. Manage. 95 (2015) 193–228. [3] G. Li, D. Liu, Y. Xie, Study on thermal properties of TBAB–THF hydrate mixture for cold storage by DSC, J. Therm. Anal. Calorim. 102 (2010) 819–826. [4] L. Gang, H. Yunho, R. Reinhard, Review of cold storage materials for air conditioning application, Int. J. Refrig. 35 (2012) 2053–2127. [5] E. Oró, A. Gracia, A. Castell, M.M. Farid, L.F. Cabeza, Review on phase change materials (PCMs) for cold thermal energy storage applications, Appl. Energy 99 (2012) 513–533. [6] J. Lisi, P. Lan, C. Ying, M. Songping, L. Xing, Improving the supercooling degree of titanium dioxide nanofluids with sodium dodecylsulfate, Appl. Energy 124 (2014) 248–255. [7] Refat Al-Shannaq, Jamal Kurdi, Shaheen Al-Muhtaseb, Michelle Dickinson, Mohammed Farid, Supercooling elimination of phase change materials (PCMs) microcapsules, Energy 87 (2015) 654–662. [8] B. Sandnes, J. Rekstad, Supercooling salt hydrates: stored enthalpy as a function of temperature, Sol. Energy 80 (2006) 616–625. [9] T.X. Li, D.L. Wu, F. He, R.Z. Wang, Experimental investigation on copper foam/ hydrated salt composite phase change material for thermal energy storage, Int. J. Heat Mass Transfer 115 (2017) 148–157. [10] Y. Fan, C. Yingying, L. Xingbo, X. Jiale, L. Chiahsun, L. Qi, T. Peng, D. Tao, Dispersion stability of thermal nanofluids, Prog. Nat. Sci.: Mater. Int. 27 (2017) 531–542. [11] Florian Kleiner, Konrad Posern, Andrea Osburg, Thermal conductivity of selected salt hydrates for thermochemical solar heat storage applications measured by the light flash method, Appl. Therm. Eng. 113 (2017) 1189–1193. [12] S.L. Che, P.P. Wang, T.S. Lee, An experimental investigation of nucleation probability of supercooled water inside cylindrical capsules, Exp. Therm. Fluid Sci. 18 (1998) 299–306. [13] A. Safari, R. Saidur, F.A. Sulaiman, X. Yan, J. Dong, A review on supercooling of Phase Change Materials in thermal energy storage systems, Renew. Sustain. Energy Rev. 70 (2017) 905–919. [14] Mohamed M. Tawfik, Experimental studies of nanofluid thermal conductivity enhancement and applications: a review, Renew. Sustain. Energy Rev. 75 (2017) 1239–1253. [15] L. Yudong, L. Xin, H. Pengfei, H. Guanghua, Study on the supercooling degree and nucleation behavior of water-based graphene oxide nanofluids PCM, Int. J. Refrig. 50 (2015) 80–86.
[16] W.L. Cui, L.L. YuanY.P. Sun, X.L. Cao, X.J. Yang, Experimental studies on the supercooling and melting/freezing characteristics of nano- copper/sodium acetate trihydrate composite phase change materials, Renew. Energy 99 (2016) 1029–1037. [17] H. Qinbo, W. Shuangfeng, T. Mingwei, L. Yudong, Experimental study on thermophysical properties of nanofluids as phase-change material (PCM) in low temperature, Energy Convers. Manage. 64 (2012) 199–205. [18] Y.D. Liu, J.Q. Wang, C.J. Su, S.C. Geng, Y.K. Gao, Q.G. Peng, Nucleation rate and supercooling degree of water-based graphene oxide nanofluids, Appl. Therm. Eng. 115 (2017) 1226–1236. [19] M.A. Khairul, S. Kalpit, D. Elham, A. Reza, M. Behdad, Effects of surfactant on stability and thermo-physical properties of metal oxide nanofluids, Int. J. Heat Mass Transfer 98 (2016) 778–787. [20] L. Xiaoke, Z. Changjun, L. Xinyu, L. Wenliang, Stability and enhanced thermal conductivity of ethylene glycol-based SiC nanofluids, Int. J. Heat Mass Transfer 89 (2015) 613–619. [21] L. Yudong, L. Xin, H. Pengfei, H. Guanghua, Study on the supercooling degree and nucleation behavior of water-based graphene oxide nanofluids PCM, Int. Refrig. 50 (2015) 80–86. [22] B.C. Jean-Pierre, D. Thomas, C.L. Jean, Ice slurry production using supercooling phenomenon, Int. Refrig. 331 (2010) 196–204. [23] P. Chandrasekaran, M. Cheralathan, R. Velraj, Effect of fill volume on solidification characteristics of DI (deionized)water in a spherical capsule: an experimental study, Energy 90 (2015) 508–515. [24] S. Omid, A. Mohammad, Effects of temperature and particles concentration on the dynamic viscosity of MgO-MWCNT/ethylene glycol hybrid Nano fluid: experimental study, Phys. E 84 (2016) 564–570. [25] S. Anant, D. Buddhi, R.L. Sawhney, Thermal cycling test of few selected inorganic and organic phase change materials, Renew. Energy 33 (2008) 2606– 2614. [26] I. Yuji, S. Masahiro, K. Masaaki, I. Kohei, Theoretical analysis of supercooled states of water generated below the freezing point in a PEFC, Int. J. Heat Mass Transfer 74 (2014) 215–227. [27] S.Y. Wu, X. Tong, C.D. Nie, D.Q. Peng, S.G. Gong, Z.Q. Wang, The effects of various carbon nanofillers on the thermal properties of paraffin for energy storage applications, J. Therm. Anal. Calorim. 124 (2016) 181–188. [28] V. Kumaresa, R. Velraj, N. Maitreyee, A.K. Maini, Phase change material-based nanofluids for heat transfer enhancement in latent heat thermal energy storage system, Int. J. Green Nanotechnol. 4 (2012) 541–546. [29] P. Murat, K. Sengül, N.T. Ümit, Y. Kerim, Thermal property investigation of multi walled carbon nanotubes (MWCNTs)Embedded phase change materials (PCMs), Therm. Thermomech. Phenom. Electron. Syst. 15 (2016) 639–644. [30] F. Xin, W.F. Li, D. Qing, W. Xiao, L.Y. Xiao, F.H. Jian, T.Y. Zi, H.C. Guan, C.H. Ya, Ke FC.Increased thermal conductivity of eicosane-based composite phase change materials in the presence of graphene nanoplatelets, Energy Fuels 27 (2013) 4041–4047. [31] W. Baojie, Z. Changjun, L. Xiaoke, Experimental investigation on stability and thermal conductivity of diathermic oil based TiO2 nanofluids, Int. J. Heat Mass Transfer 104 (2017) 537–543.