1、578 ASHRAE TransactionsABSTRACT Microencapsulated phase change material (MPCM) and carbon-nanotube based nanofluids have both been investi-gated as heat transfer coefficient enhancers separately. In this paper, we investigate the potential manifold benefit of using a blend of both as a new heat tran
2、sfer fluid. The effect of percent-age of MPCM that undergoes phase change and the compo-sition of the new blend of heat transfer fluid have been investigated. A computer simulation code reveals that the best composition for the MPCM-nanofluid blend depends on the actual percentage of phase change th
3、at takes place in the process. Better heat transfer rates can be achieved when high quantity of MPCM undergoes phase change.INTRODUCTIONIncreasing intensity and performance of heat transfer in heat exchanger equipment is one of the most pressing issues in the world today. Information technology appl
4、ications as well as data centers have seen a growth in processing speed which has resulted in higher dissipation of heat. In HVAC applica-tions, increasing the heat transfer rate can lead to reduction of the required flow rate of heat transfer fluids which conse-quently saves energy through decrease
5、d pumping power.The introduction of Microencapsulated Phase Change Material (MPCM) in form of slurries as heat transfer fluids has been well accepted among researchers. MPCM slurries consist of a base fluid such as water that contains a relatively low mass fraction of microcapsule filled with phase
6、change material such as octadecane or paraffin wax. The microcap-sules vary in size between 1 to 10 m to avoid breakage during continuous pumping. High effective heat capacity is expected when the phase change undergoes phase change while is flow-ing under constant heat flux or temperature condition
7、s.Several experimental as well as numerical studies have been published recently (Ravi et al. 2009, Alkan et al. 2009, Alvarado et al. 2007, Chen et al. 2008, Goel et al. 1994, Bai and Lu 2003, Yamagishi et al. 1999 and Mulligan et al. 1996). Improvement in heat transfer coefficient as much as 3 tim
8、es has been reported. Use of carbon nanotubes (CNT) dispersed in water has also been under intense investigation recently (Amrollahi et al. 2008, Vadasz 2006, Patel et al. 2008, Hsu et al. 2008, Garg et al. 2009). Most of the literature reports heat transfer enhance-ment with nanofluids as working f
9、luid, though there is no total agreement among researchers on the magnitude of enhance-ment. In this paper we propose combining MPCM and CNT to a base working fluid medium in order to quantify the benefits of the anticipated collective effect of additional heat capacity and thermal conductivity enha
10、ncement. ANALYSISTransport PropertiesAnalysis of any heat transfer fluid in a thermal systems needs well defined thermal properties. Density ( ), specific heat (cp), thermal conductivity (k) and dynamic viscosity () are the four thermophysical properties that are essential in any convection heat tra
11、nsfer problem. With regards to density, there is minimal discrepancy in the literature. Pak and Cho (1998), compared the density measurement results for Al2O3and TiO2in distilled water with those estimated by the simple mixing theory at 298K (536.4R) and up to a volume concen-tration of 5%, and foun
12、d that the maximum deviation is 0.6%. Density measurements by Vajjha et al. (2009) showed that for System Analysis of MPCM Slurry Enhanced with Carbon Nanotubes as Heat Transfer FluidHessam Taherian, PhD Jorge L. Alvarado, PhD, PEMember ASHRAE Associate Member ASHRAEHessam Taherian is a research ass
13、ociate and Jorge L. Alvarado is an assistant professor in the Department of Engineering Technology and Industrial Distribution of Texas A&M University, College Station, TX.AB-10-0212010, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashrae.org). Published in AS
14、HRAE Transactions (2010, Vol. 116, Part 2). For personal use only. Additional reproduction, distribution, or transmission in either print or digital form is not permitted without ASHRAEs prior written permission.2010 ASHRAE 579nanofluids, the simple mixing equation can be used with mini-mal deviatio
15、n.Equation 1 is postulated based on the conservation of mass principle and by modifying the mixing model for two components (MPCM and CNT) added to the base fluid.(1)The density of the MPCM can be predicted accurately using Equation 2 (Chen et al., 2008). Furthermore, Roy and Avanic (2001) suggested
16、 that the suspension density would not change by more than 1-2% for phase change material concentrations of 10-20% as long as the specific gravity of the phase change material was greater than 0.8.(2)With regards to the specific heat of the MPCM slurries, Alisetti and Roy (1999) through numerical an
17、alysis, proposed and investigated several functions for the specific heat during phase change process. They found that the exact shape of the function during phase change was not crucial to the analysis of laminar convective heat transfer inside a circular tube as the difference among various models
18、 was less than 4%. Equation 3 has been proposed by Mulligan et al. (1996) based on energy balance to calculate the specific heat of MPCM slurries.(3)The specific heat of the PCM capsules will be evaluated using Equation 4 (Chen et al. 2008).(4)Buongiorno (2006) recently proposed the use of Equation
19、5 for the specific heat of nanofluids whereas, Pak and Cho (1998) suggest using the simpler Equation 6.(5)(6)However, Zhou and Ni (2008) found that the result of their experiments fits Equation 5 well, and had considerable devi-ation when using Equation 6. For the mixture of MPCM and nanofluid, an e
20、nergy balance equation leads to:(7)The thermal conductivity is the most difficult of the prop-erties to model. For MPCM slurries most researchers use Maxwells (1954) relation as given in Equation 8.(8)For nanofluids, in addition to Maxwells relation there are other models available in the literature
21、. Equation (9) has been proposed by Hamilton and Crosser (1962):(9)where n=3/ and is called the sphericity and defined as the ratio of the surface areas of a sphere with the volume equal to that of the nanoparticle.To estimate the combined effect of nanofluid and MPCM slurry, available models for bi
22、nary mixtures of liquids may be used. Maloka (2007) reviewed several complicated models and proposed a new model. However, according to the study made by Focke (2008), all models deviate by less than 10% from the much simpler “linear blending rule” presented in Equation 10.(10)For the dynamic viscos
23、ity of suspensions, Einsteins theory has been presented by Drew (1998) in the form of Equation 11.(11)Brinkman (1952) and Thomas (1965) proposed Equa-tions 12 and 13, respectively for the effective viscosity of Newtonian suspensions. Both Equations are valid for 0.25.(12)(13)Equations 11-13 can be s
24、imilarly used for nanofluids and MPCM slurries as long as the assumption of Newtonian behavior holds. In order to estimate the effective viscosity of the mixture of CNT fluid and MPCM slurry, the Refutas method (Maples 1993) which has originally been used for blends of hydocarbons may be used for bl
25、ends of MPCM slurry and nanofluid. The method is based on calculating Viscosity Blending Index (VBI) of each component of the blend:(14)then calculating VBN of the blend using Equation (15):(15)the kinematic viscosity of the blend can be calculated using Equation (16):(16)wfnpnpMPCMMPCMbf1 np MPCM()
26、+=MPCM1 y+()cssyc+-=Cp,effCp,bf1 MPCM()MPCMCp,MPCMT-+=Cp,MPCMCp,cyCp,s+()csycs+()MPCM-=Cp,nfpCp,np1 ()bfCp,bf+nf-=Cp,nfCp,np+ 1 ()Cp,bf=Cp,eff1eff- (npnpCp,npMPCMMPCMCp,MPCM+=bf1 np MPCM()Cp,bfMPCMMPCM T)+kwfkbf2kbfkMPCM2 kMPCMkbf()+2kbfkMPCM kMPCMkbf+-=knfkbfknpn 1()kbfn 1()kbfknp()+knpn 1()kbf kbf
27、knp()+-= keffxAkAxBkB+=effbf12.5+()=effbf11 ()2- 5=effbf12.5 10.052+=VBI 14.534 ln ln v 0.8+() 10.975+=VBIBlendxAVBIAxBVBIB+=v exp expVBIBlend10.97514.534-0.8=2010, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashrae.org). Published in ASHRAE Transactions (201
28、0, Vol. 116, Part 2). For personal use only. Additional reproduction, distribution, or transmission in either print or digital form is not permitted without ASHRAEs prior written permission.580 ASHRAE TransactionsFinally, the dynamic viscosity of the mixture can be calculated with prior knowledge of
29、 the mixtures mass density. The pressure drop can be estimated using the Darcy-Weisbach equation for turbulent flow in pipes.Heat Transfer and Pressure Drop of MPCM-CNT Heat Transfer FluidIn order to show the enhancing effect of the new heat transfer fluid, a concentric-tube exchanger has been consi
30、d-ered as the heat transfer device. Table 1 shows the heat exchanger parameters as well as the heat transfer fluid prop-erties. The results have shown that for 1kW (3412 Btu/hr) only 0.05 L/s (0.79 GPM) of heat transfer fluid flow is required.Due to the nature of the mixing process, dimensional para
31、meters of multi-walled carbon nanotubes (MWCNT) can only be given within a range rather than a definite fixed value, so median values were considered for computational analysis. The specific heat value for carbon nanotubes was taken from Hepplestone et al. (2006).The heat exchanger depicted in Figur
32、e 1 shows schematic diagram of a concentric-tube heat exchanger.Which was considered for the analysis. Water flows as the cooling liquid in the outer tube of the heat exchanger. As indicated in Table 1, the volume fractions of CNT in the nano-fluid and MPCM in the slurry are considered constant and
33、equal to 1% and 10%, respectively. The design temperatures have been selected in a way that the average inlet fluid temper-ature is around 27C (80.6F) which is the melting point of the phase change material (i.e. octadecane). At first, a desired temperature drop for water (cooling fluid) was specifi
34、ed. Since both outlet temperature of the heat transfer fluid and heat transfer area are unknown, the simula-tion starts with an initial guess for the heat transfer fluid outlet temperature. Knowing the average temperature of both fluids, their thermophysical properties can be calculated using the eq
35、uations shown above. The value for Cp,effdepends on the temperature drop (or difference) in the heat transfer fluid as seen in Equation 7. Then Nusselt numbers for water and the heat transfer fluid are determined using Dittus-Boelter corre-lation (Equation 17) for turbulent flow (Winterton 1998).(17
36、)In Equation 17, m is equal to 0.3 for the fluid that is being cooled and 0.4 for the fluid that is being heated. The overall heat transfer coefficient U is then calculated using Equation 17 and the corresponding heat transfer coefficients (U = 1/ho+ 1/hi). The required heat transfer area is determi
37、ned using Equa-tion 18 for counter-flow configuration.(18)Using energy balance, the value of the heat transfer fluid outlet temperature can be corrected until convergence is reached. The algorithm briefly described above was incorpo-rated into a computer code which was developed using EES software (
38、Klein 2008) to solve for the heat exchanger design problem to determine required surface area.ResultsIn Figure 2, is the percentage of MPCM that has under-gone phase change in the process. Figure 2 also shows that when high percentages of phase change material undergoes phase change, an increasing m
39、ass fraction of nanofluid in the blend will cause Cp,effto drop sharply from 20.1 (4.8) to 4.3kJ/Figure 1 Schematic diagram of a concentric-tube heat exchanger.Nu 0.023Re0.8Prm=Table 1. Concentric-Tube Heat Exchanger Analysis ParametersInner-Tube Inside Diameter 0.0134 (0.528) m (in)Inner-Tube Wall
40、Thickness0.0012 (0.047) m (in)Outer-Tube Inside Diameter0.02 (0.787) m (in)Water-Side Temperature Raise2 (3.6) K (R)Carbon Nanotube Volume Fraction1 %MPCM Volume Fraction10 %Fluid Flow Rate 0.05 (0.79) L/s (GPM)Water Flow Rate 0.12 (1.90) L/s (GPM)Fluid Inlet Temperature31 (87.8) C (F)AQULMTD-=2010,
41、 American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashrae.org). Published in ASHRAE Transactions (2010, Vol. 116, Part 2). For personal use only. Additional reproduction, distribution, or transmission in either print or digital form is not permitted without ASHRAEs
42、 prior written permission.2010 ASHRAE 581kgK (1.03 Btu/lbF). At low values, the effective specific heat is not affected considerably by change in the blend composition. If there is no phase change in the fluid, Cp,effincreases slightly with increasing the amount of CNT.Figure 3 shows that the effect
43、ive thermal conductivity of the blend increases by increasing the amount of nanofluid in the blend regardless of the percentage of the MPCM that changes phase. This is expected since higher amount of CNT should result in higher thermal conductivity.The overall heat transfer coefficient of the heat e
44、xchanger almost increases linearly with increasing the amount of nano-fluid for cases where a low fraction of phase change happens as shown in Figure 4. As the fraction of material that changes phase increases to a value over 75%, the linearity of the curve is disturbed and the curve exhibits a mini
45、mum value. The Uvalue first decreases and then increases again with increasing amount of nanofluid. The U value is at its highest when 100% phase change occurs and no nanofluid is present in the blend. This can be explained by the dominance of T in the effective specific heat model which reaches a m
46、inimum value at 100% phase change.Figure 5 demonstrates the effect of parameters on the temperature drop of the heat transfer fluid passing through the heat exchanger. As expected, the lowest temperature change happens in the case when 100% phase change takes place, and there is no nanofluid in the
47、blend. This is due to high value of the effective specific heat of the fluid as described above. The change with the fraction of nanofluid in the blend is linear in all cases.One should expect that the required surface area of the heat exchanger to be lowest when phase change effect is at its highes
48、t. This is shown in Figure 6. However, if phase change is less than 50% then it is beneficial to have a high amount of nanofluid in the blend. At low phase change rates, the required Figure 2 Effective specific heat of the fluid as a function of blending percentage.Figure 3 Effective thermal conductivity of the blend.Figure 4 Overall heat transfer coefficient
