- Email: [email protected]
Thermal analysis of slurry ice production system using direct contact heat transfer of carbon dioxide and water mixture
Available online at www.sciencedirect.com
International Communications in Heat and Mass Transfer 35 (2008) 756 – 761 www.elsevier.com/locate/ichmt
Thermal analysis of slurry ice production system using direct contact heat transfer of carbon dioxide and water mixture☆ S. Thongwik a , N. Vorayos a , T. Kiatsiriroat a,⁎, A. Nuntaphan b a
b
Department of Mechanical Engineering, Faculty of Engineering, Chiang Mai University, Chiang Mai, 50200, Thailand Thermal Technology Research Laboratory, Mae Moh Training Center, Electricity Generating Authority of Thailand, Mae Moh, Lampang 52220, Thailand Available online 27 March 2008
Abstract This research studies the heat transfer characteristic during ice formation of a direct contact heat transfer between carbon dioxide and water mixture. This research is divided into two parts. For the first part, the low temperature carbon dioxide, between − 15 and − 60 °C, is injected into water initially at 28 °C and exchanges heat directly. The flow rate of carbon dioxide is varied between 0.003 and 0.017 kg/s while the volume of water is between 1 and 3 L. From the experiment, it is found that the effectiveness of the direct contact heat transfer between the carbon dioxide and the water is closed to 100%. Moreover, the lumped model is found to be used for predicting the temperature of water and the mass of ice formation quite well. However, for this technique, the blockage of ice around the injector always occurs. To avoid this problem, slurry ice is considered instead of pure ice. In the second part of this work, the thermal behavior of the slurry ice by direct contact heat transfer technique between carbon dioxide and water mixture is investigated. Compressor oil and Tween 60 are mixed with water for making slurry ice and it is found that the suitable composition is water/oil/Tween 60 at 100/6/1 by volume. Again, the lumped model can be used for predicting the temperature of water mixture and the mass of slurry ice quite well. © 2008 Elsevier Ltd. All rights reserved. Keywords: Direct contact heat transfer; Ice storage; Ice slurry; Carbon dioxide; Lumped model
1. Introduction At present, air-conditioning system of commercial building consumes approximately 60% of electrical demand and it brings to get high electrical cost. To overcome this problem, an ice thermal energy storage has been considered. Since, the electrical cost of Thailand during the nighttime is cheaper than that of during the daytime, therefore, ice thermal energy storage system of which the ice is produced in the night and used for airconditioning during the daytime seems to be appropriate. Chaichana et al. [1] reported that an ice thermal energy storage can reduce the electrical cost approximately 55%. Moreover, Braun [2] and Carey et al. [3] reported that the coefficient of performance (COP) of an ice thermal energy storage system is higher than that of a conventional air-conditioning system. ☆
Communicated by W. J. Minkowycz. ⁎ Corresponding author. E-mail address: [email protected] (T. Kiatsiriroat).
0735-1933/$ - see front matter © 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.icheatmasstransfer.2008.02.007
Ice-on-coil system is a commercial ice thermal energy storage. An evaporator coil of this system is submerged in the water and ice is formed around its evaporator tube. Since the thermal conductivity of ice is very low, therefore, the performance of the evaporator is reduced with the increasing of ice thickness. To avoid this problem, direct contact heat transfer technique is proposed. The concept for producing ice is to inject cold refrigerant into water and exchange heat directly. Nuntaphan [4] reported that the effectiveness of a direct contact heat transfer between R12 and water is close to 100%. Moreover, Kiatsiriroat et al. [5] investigated the performance of a refrigeration cycle using a direct contact evaporator. In this study, R12 was selected as the refrigerant. They found that the COP of this system was around 3.4–3.6 and the ice is looked like slurry. Kiatsiriroat et al. [6] proposed the heat transfer correlation for the direct contact heat transfer between R12water and R22-water in the form of dimensionless parameters which were Stanton number, Stephan number and dimensionless pressure.
S. Thongwik et al. / International Communications in Heat and Mass Transfer 35 (2008) 756–761
757
Nomenclature Cp h L S M : m: Q T t Δt Δtice UA UV V
Specific heat capacity (J/kg K) Enthalpy (J/kg) Latent heat process Sensible heat process Mass (kg) Mass flow rate (kg/s) Heat transfer rate (Watt) Temperature (°C) Time (s) Time duration (s) Time duration for producing ice (s) Overall heat loss coefficient and area (W/K) Volumetric heat transfer coefficient (W/m3 K) Volume of water and water mixture (m3)
Greek symbols ρ Density (kg/m3) λ Latent heat of freezing (J/kg) Subscripts a Ambient CO2 Carbon dioxide i Inlet ice Ice L Latent heat period max Maximum o Outlet S Sensible heat period V Volume w Water
In this work, carbon dioxide, one type of a natural fluid, is used as the dispersed fluid in the direct contact heat transfer for ice production. Normally, a blockage of ice around an injector of the cold refrigerant always occurs. One interesting method to avoid this problem is to produce slurry ice. Nuntaphan [4] reported that when small amount of compressor oil is mixed with water, slurry ice is formed. However, there is no report about the suitable composition. This study is divided into two parts. For the first part, the formation of ice derived from the injection of carbon dioxide into pure water is investigated. While the second part of this work, the water is mixed with a compressor oil and a surfactant (Tween 60) for producing slurry ice and the suitable composition is investigated.
Fig. 1. Schematic sketch of the experimental set-up.
water is stored in a cylindrical test tube of 0.0762 m in diameter and 2 m in height. The volume of water is between 1 and 3 L. The temperature of water, the inlet temperature of carbon dioxide and the ambient air are measured by a set of K-type thermocouples having ± 0.1 °C accuracy and recorded every 10 s. Notice that the carbon dioxide is injected into the water till most of water becomes ice. The mass of ice in a storage tank is measured by a high precision weighing machine having ± 0.1 g accuracy. The second part of this work is to produce a slurry ice. The procedure is the same as the previous part. However, the compressor oil and Tween 60 [Polyoxyethylene sorbitan monostearate, C24H46O6U(C2H4O)n] are mixed with water. The compositions of the compressor oil and the surfactant
2. Experimental set-up Fig. 1 shows the schematic sketch of the experimental set-up. Low temperature carbon dioxide from a storage tank is injected into water in a storage tank and exchanges heat directly. For the first part of this work, the initial water temperature is around 28 °C while the inlet temperature of carbon dioxide is varied between − 15 °C and − 60 °C. The mass flow rate of carbon dioxide is varied between 0.003 and 0.017 kg/s. The
Fig. 2. Comparison of water temperature decreasing from the experiments and the lumped model.
758
S. Thongwik et al. / International Communications in Heat and Mass Transfer 35 (2008) 756–761
Tween 60 are varied between 0 and 20% and 0 and 6% by volume, respectively. 3. Mathematical model From the previous studies [4–6], it is found that when a refrigerant is injected into water, a high turbulence of water is found and the temperature of water in the whole volume is rather uniform. Therefore, the lumped model could be used for predicting the water temperature and the amount of ice formation. In case of the direct contact between carbon dioxide and water, the energy balance can be written as d : ðMw hw Þ ¼ mCO2 hCO2 ;i hCO2 ;o ðUAÞðTa Tw Þ: ð1Þ dt The left hand side of the above equation is the enthalpy change of water with time while the first and the second terms of the right hand side are the rate of enthalpy change of carbon dioxide and the rate of heat loss to the environment, respectively. During the sensible heat period, Eq. (1) is modified and the temperature of water in the storage tank could be : m CO2 hCO2 ;i hCO2 ;o ðUAÞðTa Tw Þ TwtþDt ¼ Twt þ Dt: Mw Cpw ð2Þ Moreover, during ice formation period, the water temperature is constant at its freezing point and the mass of ice during a time interval could be predicted from : m CO2 hCO2 ;i hCO2 ;o ðUAÞðTa Tw Þ Mice ¼ Dtice : ð3Þ kw 4. Results and discussion 4.1. Direct contact heat transfer of carbon dioxide and pure water Fig. 2 shows the effect of the mass flow rate and the inlet temperature of carbon dioxide and the volume of water on the decrease of water temperature in the storage tank. Notice that the
Fig. 4. The comparison of mass of ice from the experiment and the model.
volume of water is 1–3 L, the mass flow rate and the inlet temperature of carbon dioxide is varied between 0.003 and 0.017 kg/s and −35 °C and −50 °C, respectively. The initial temperature of the water is approximately 28 °C. The undoubtedly result shows that the decrease of water volume is directly proportional to the mass flow rate of carbon dioxide and inversely proportional to the volume of water and the inlet temperature of carbon dioxide. Moreover, the lumped model can predict the temperature of water in the storage tank quite well. Fig. 3 shows the effectiveness of a direct contact heat transfer between water and carbon dioxide. Notice that the effectiveness can be calculated from : Q e ¼ : CO2 ; ð4Þ Q max : : ð5Þ Q CO2 ¼ mCp CO2 TCO2 ;o TCO2 ;i ; : : Q max ¼ mCp CO2 Tw TCO2 ;i :
ð6Þ
From the figure, it is found that the effectiveness from this work agrees well with the results of Nuntaphan [4] which is close to 100%. However, for the water volume of 1 L, the effectiveness is slightly lower than those of 2 L and 3 L. This result comes from the shorter retention time of heat exchanging. In case of latent heat period, Fig. 4 shows the comparison of mass of ice from the experiment with the results from the model. It is found that the lumped model can predict the mass of ice approximately 77.3% within ± 10% variation. Notice that the ice from the direct contact heat transfer process has high porosity. Fig. 5 shows the ice formation at various conditions. In some conditions, there is a blockage of ice around the injector and it obstructs the flow of carbon dioxide. 4.2. Direct contact heat transfer of carbon dioxide and water– oil mixture
Fig. 3. Effectiveness of the direct contact heat transfer between water and carbon dioxide.
In this part, 11 samples of water–oil mixture are tested by exchange heat with low temperature carbon dioxide via direct
S. Thongwik et al. / International Communications in Heat and Mass Transfer 35 (2008) 756–761
759
Fig. 5. Ice formation at various states.
contact heat transfer technique. The surfactant, Tween 60, is used to merge the water and the compressor oil. The testing method is the same as the previous chapter and the compositions of water–oil mixture are given in Table 1. It is found that at the compositions of numbers 8–11, all of water–oil mixture becomes slurry. When the composition of compressor oil is lower than 6%, such as cases 1–7, a blockage of ice is formed around the injector. At these conditions, the compressor oil composition is rather low, therefore, part of non-merging water becomes plain ice and the blockage of ice around the injector occurs. In our experiment, the composition of water/compressor oil/Tween 60 at 100/6/1 is selected as a water mixture for producing slurry ice. Fig. 6 shows the slurry ice at this composition. In this part the lumped model is also used for predicting the water–oil mixture temperature and the mass of slurry ice during a time interval. Notice that, from our test, it is found that the specific heat and the enthalpy of freezing of water–oil mixture are close to that of pure water (lower than 1% difference). Therefore, the properties of water are used for calculating the temperature and mass of ice slurry. Fig. 7 shows the temperature of water–oil mixture and it is found that the lumped model can predict the temperature quite well. Fig. 8 shows the comparison of mass of slurry ice from the lumped model compared with that of the experiment and it is
found that the model can predict 100% of the experimental data within ± 10% variation. 4.3. Heat transfer analysis of direct contact heat exchange This research also studies the effect of parameters on the volumetric heat transfer coefficient (UV) of a direct contact heat transfer technique. In this work UV can be calculated from : Q CO2 ; UV ¼ V DT
ð7Þ
DT ¼ Tw TCO2 ;i :
ð8Þ
Where V is volume of water. During the sensible heat process, the volumetric heat transfer coefficient of this study is compared with the previous works and the results are shown in Table 2. Table 2 shows the volumetric heat transfer coefficient of the present study and is close to those of Goodwin et al. [8] and Kiatsiriroat et al. [6] for non-flow or stagnant storage medium. It is less than those of Blair [7] and Sidemen and Get [9] more than 10 times for injection of dispersed-phase agent into flowing water. In the case of the result of Goodwin et al. [8], the
760
S. Thongwik et al. / International Communications in Heat and Mass Transfer 35 (2008) 756–761
Table 1 Water mixture composition and form of ice No. Water/oil/Tween60 Color (by volume)
Solubility Blockage of ice Form of around injector ice
1
100/1/2
Good
Yes
2
100/3/0.5
Good
Yes
3
100/3/1
Good
Yes
4
100/3/2
Fair
Yes
5
100/3/4
Fair
Yes
6
100/3/6
Good
Yes
7
100/4/1
Fair
Yes
8
100/6/1
Fair
No
Slurry + plain ice Slurry + plain ice Slurry + plain ice Slurry + plain ice Slurry + plain ice Slurry + plain ice Slurry + plain ice Slurry
9
100/8/1
Fair
No
Slurry
10
100/9/1
Fair
No
Slurry
11
100/20/1
Fair
No
Slurry
Clear yellow Cloudy white Cloudy white Cloudy white Cloudy white Clear pink Cloudy white Cloudy white Cloudy white Cloudy white Cloudy white
volumetric heat transfer coefficient is still low (3–8 kW/m3 K) even though the water is continuously flowing. This result may come from the very high mass of water and mass flow rate at the storage tank then the dispersed-phase agent could not merge throughout the water in the whole tank. In this work, the models for predicting the volumetric heat transfer coefficient are also developed as for sensible heat: : 0:050828 UV ;S ¼ 729:31Mw0:99955 m 1:0117 ; CO2 DT
ð9Þ
storage medium temperature drops from its initial temperature at room temperature about 28 °C to its freezing point at 0 °C and the latent heat period occurs when the water temperature is at 0 °C till ice is forming. The values of the volumetric heat transfer coefficient from the correlations are compared with those from the experimental data for both sensible heat and latent heat periods as shown in Figs. 9 and 10. The model can predict all of the experimental values within ± 10%. Increasing of mass flow rate of carbon dioxide will get high turbulence in the storage medium thus high volumetric heat transfer coefficient is obtained. The heat transfer coefficient is less with the increasing of mass of water due to the decreasing of turbulence in the storage medium. It is found that temperature difference between water and carbon dioxide, has a slight effect on the volumetric heat transfer coefficient compared to those of previous parameters. 5. Conclusion
for latent heat: : 0:066906 UV ;L ¼ 704:52ðMw Mice Þ0:98204 m 1:0186 : CO2 DT
Fig. 7. The temperature of water mixture at various conditions.
ð10Þ
Both equations could be applied for pure water and water–oil mixture. The sensible heat period will be considered when the
From this research, it could be concluded as – The decreasing of water temperature is directly proportional to the mass flow rate of carbon dioxide and inversely
Table 2 Comparison of working conditions of direct contact heat transfer systems
Solution phase change
Fig. 6. Slurry ice from the composition of water mixture of water/oil/Tween 60 at 100/6/1.
Dispersed phase UV (kW/m3 K) Mass flow rate of dispersed phase (kg/s) Mass flow rate of water (kg/min) Diameter of storage tank (m)
Present study
Blair [7]
Goodwin et al. [8]
Sideman and Gat [9]
Kiatsiriroat et al. [6]
Water, water mixture CO2 1–25 0.003– 0.017
Water
Water
Water
Water
R113 122 0.023
Pentane 3–8 1.90
Pentane 180 0.022
R12, R22 2–16, 2–52 0.02–0.08
–
0.101
17.1
0.042
–
0.07
0.095
0.61
0.07
0.30
S. Thongwik et al. / International Communications in Heat and Mass Transfer 35 (2008) 756–761
– – – –
Fig. 8. The comparison of mass of ice from the experiment and the model.
761
proportional to the volume of water and the inlet temperature of carbon dioxide. The effectiveness of a direct contact heat transfer between carbon dioxide and water is close to 100%. The suitable composition of water–oil mixture, water/oil/ Tween, for producing slurry ice is at 100/6/1 by volume. The lumped model can predict the water temperature and the amount of ice in case of sensible heat and latent heat periods quite well. The volumetric heat transfer coefficient is inversely proportional to the mass of water and water mixture. On the other hand, the mass flow rate of carbon dioxide and temperature difference between inlet and outlet the storage tank are directly proportional.
Acknowledgements The authors would like to thank Thermal Technology Research Laboratory, Mae Moh Training Center, Electricity Generating Authority of Thailand for facilitating the testing equipment and the Commission on Higher Education, Thailand for funding this research study. References
Fig. 9. Comparison of volumetric heat transfer coefficient from the correlation and the experimental data during sensible heat period.
Fig. 10. Comparison of volumetric heat transfer coefficient from the correlation and the experimental data during latent heat period.
[1] C. Chaichana, W.S.C. William, A. Lu, An ice thermal storage computer model, Applied Thermal Engineering 21 (2001) 1769–1778. [2] J.E. Braun, Reducing energy cost and peak electric demand though optimal control of building thermal storage, ASHRAE Transaction 96 (2) (1990). [3] C.W. Carey, J.W. Mitchell, W.A. Beckman, The control of ice storage system, ASHRAE Transaction (1995) 32–39. [4] A. Nuntaphan, Performance analysis of a refrigeration cycle using a direct contact evaporator, M.Eng. Thesis, Department of Mechanical Engineering, Chiang Mai University, Thailand, (1998). [5] T. Kiatsiriroat, P. Siriplubpla, A. Nuntaphan, Performance analysis of a refrigeration cycle using a direct contact evaporator, Int. J. Energy Research 22 (1998) 1179–1190. [6] T. Kiatsiriroat, N. Vorayos, S. Vithayasai, A. Nuntaphan, N. Vorayos, Heat transfer prediction for a direct contact ice thermal energy storage, Energy Conversion & Management 44 (2003) 497–508. [7] C.K. Blair, Heat transfer characteristics of a three-phase volume boiler direct contact heat exchanger, M.S. Thesis, University of Utha, Salt Lake City, (1976). [8] P. Goodwin, M. Cobin, R. Boehm, Evaluation of the flooding limits and column, ASME National Heat Transfer Conference, Denver, 85-HT-49, 1985. [9] S. Sideman, Y. Gat, Direct contact heat transfer with change of phase: spray column study of three phase heat exchanger, AICHE Journal 95 (1) (1996) 1206–1213.
International Communications in Heat and Mass Transfer 35 (2008) 756 – 761 www.elsevier.com/locate/ichmt
Thermal analysis of slurry ice production system using direct contact heat transfer of carbon dioxide and water mixture☆ S. Thongwik a , N. Vorayos a , T. Kiatsiriroat a,⁎, A. Nuntaphan b a
b
Department of Mechanical Engineering, Faculty of Engineering, Chiang Mai University, Chiang Mai, 50200, Thailand Thermal Technology Research Laboratory, Mae Moh Training Center, Electricity Generating Authority of Thailand, Mae Moh, Lampang 52220, Thailand Available online 27 March 2008
Abstract This research studies the heat transfer characteristic during ice formation of a direct contact heat transfer between carbon dioxide and water mixture. This research is divided into two parts. For the first part, the low temperature carbon dioxide, between − 15 and − 60 °C, is injected into water initially at 28 °C and exchanges heat directly. The flow rate of carbon dioxide is varied between 0.003 and 0.017 kg/s while the volume of water is between 1 and 3 L. From the experiment, it is found that the effectiveness of the direct contact heat transfer between the carbon dioxide and the water is closed to 100%. Moreover, the lumped model is found to be used for predicting the temperature of water and the mass of ice formation quite well. However, for this technique, the blockage of ice around the injector always occurs. To avoid this problem, slurry ice is considered instead of pure ice. In the second part of this work, the thermal behavior of the slurry ice by direct contact heat transfer technique between carbon dioxide and water mixture is investigated. Compressor oil and Tween 60 are mixed with water for making slurry ice and it is found that the suitable composition is water/oil/Tween 60 at 100/6/1 by volume. Again, the lumped model can be used for predicting the temperature of water mixture and the mass of slurry ice quite well. © 2008 Elsevier Ltd. All rights reserved. Keywords: Direct contact heat transfer; Ice storage; Ice slurry; Carbon dioxide; Lumped model
1. Introduction At present, air-conditioning system of commercial building consumes approximately 60% of electrical demand and it brings to get high electrical cost. To overcome this problem, an ice thermal energy storage has been considered. Since, the electrical cost of Thailand during the nighttime is cheaper than that of during the daytime, therefore, ice thermal energy storage system of which the ice is produced in the night and used for airconditioning during the daytime seems to be appropriate. Chaichana et al. [1] reported that an ice thermal energy storage can reduce the electrical cost approximately 55%. Moreover, Braun [2] and Carey et al. [3] reported that the coefficient of performance (COP) of an ice thermal energy storage system is higher than that of a conventional air-conditioning system. ☆
Communicated by W. J. Minkowycz. ⁎ Corresponding author. E-mail address: [email protected] (T. Kiatsiriroat).
0735-1933/$ - see front matter © 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.icheatmasstransfer.2008.02.007
Ice-on-coil system is a commercial ice thermal energy storage. An evaporator coil of this system is submerged in the water and ice is formed around its evaporator tube. Since the thermal conductivity of ice is very low, therefore, the performance of the evaporator is reduced with the increasing of ice thickness. To avoid this problem, direct contact heat transfer technique is proposed. The concept for producing ice is to inject cold refrigerant into water and exchange heat directly. Nuntaphan [4] reported that the effectiveness of a direct contact heat transfer between R12 and water is close to 100%. Moreover, Kiatsiriroat et al. [5] investigated the performance of a refrigeration cycle using a direct contact evaporator. In this study, R12 was selected as the refrigerant. They found that the COP of this system was around 3.4–3.6 and the ice is looked like slurry. Kiatsiriroat et al. [6] proposed the heat transfer correlation for the direct contact heat transfer between R12water and R22-water in the form of dimensionless parameters which were Stanton number, Stephan number and dimensionless pressure.
S. Thongwik et al. / International Communications in Heat and Mass Transfer 35 (2008) 756–761
757
Nomenclature Cp h L S M : m: Q T t Δt Δtice UA UV V
Specific heat capacity (J/kg K) Enthalpy (J/kg) Latent heat process Sensible heat process Mass (kg) Mass flow rate (kg/s) Heat transfer rate (Watt) Temperature (°C) Time (s) Time duration (s) Time duration for producing ice (s) Overall heat loss coefficient and area (W/K) Volumetric heat transfer coefficient (W/m3 K) Volume of water and water mixture (m3)
Greek symbols ρ Density (kg/m3) λ Latent heat of freezing (J/kg) Subscripts a Ambient CO2 Carbon dioxide i Inlet ice Ice L Latent heat period max Maximum o Outlet S Sensible heat period V Volume w Water
In this work, carbon dioxide, one type of a natural fluid, is used as the dispersed fluid in the direct contact heat transfer for ice production. Normally, a blockage of ice around an injector of the cold refrigerant always occurs. One interesting method to avoid this problem is to produce slurry ice. Nuntaphan [4] reported that when small amount of compressor oil is mixed with water, slurry ice is formed. However, there is no report about the suitable composition. This study is divided into two parts. For the first part, the formation of ice derived from the injection of carbon dioxide into pure water is investigated. While the second part of this work, the water is mixed with a compressor oil and a surfactant (Tween 60) for producing slurry ice and the suitable composition is investigated.
Fig. 1. Schematic sketch of the experimental set-up.
water is stored in a cylindrical test tube of 0.0762 m in diameter and 2 m in height. The volume of water is between 1 and 3 L. The temperature of water, the inlet temperature of carbon dioxide and the ambient air are measured by a set of K-type thermocouples having ± 0.1 °C accuracy and recorded every 10 s. Notice that the carbon dioxide is injected into the water till most of water becomes ice. The mass of ice in a storage tank is measured by a high precision weighing machine having ± 0.1 g accuracy. The second part of this work is to produce a slurry ice. The procedure is the same as the previous part. However, the compressor oil and Tween 60 [Polyoxyethylene sorbitan monostearate, C24H46O6U(C2H4O)n] are mixed with water. The compositions of the compressor oil and the surfactant
2. Experimental set-up Fig. 1 shows the schematic sketch of the experimental set-up. Low temperature carbon dioxide from a storage tank is injected into water in a storage tank and exchanges heat directly. For the first part of this work, the initial water temperature is around 28 °C while the inlet temperature of carbon dioxide is varied between − 15 °C and − 60 °C. The mass flow rate of carbon dioxide is varied between 0.003 and 0.017 kg/s. The
Fig. 2. Comparison of water temperature decreasing from the experiments and the lumped model.
758
S. Thongwik et al. / International Communications in Heat and Mass Transfer 35 (2008) 756–761
Tween 60 are varied between 0 and 20% and 0 and 6% by volume, respectively. 3. Mathematical model From the previous studies [4–6], it is found that when a refrigerant is injected into water, a high turbulence of water is found and the temperature of water in the whole volume is rather uniform. Therefore, the lumped model could be used for predicting the water temperature and the amount of ice formation. In case of the direct contact between carbon dioxide and water, the energy balance can be written as d : ðMw hw Þ ¼ mCO2 hCO2 ;i hCO2 ;o ðUAÞðTa Tw Þ: ð1Þ dt The left hand side of the above equation is the enthalpy change of water with time while the first and the second terms of the right hand side are the rate of enthalpy change of carbon dioxide and the rate of heat loss to the environment, respectively. During the sensible heat period, Eq. (1) is modified and the temperature of water in the storage tank could be : m CO2 hCO2 ;i hCO2 ;o ðUAÞðTa Tw Þ TwtþDt ¼ Twt þ Dt: Mw Cpw ð2Þ Moreover, during ice formation period, the water temperature is constant at its freezing point and the mass of ice during a time interval could be predicted from : m CO2 hCO2 ;i hCO2 ;o ðUAÞðTa Tw Þ Mice ¼ Dtice : ð3Þ kw 4. Results and discussion 4.1. Direct contact heat transfer of carbon dioxide and pure water Fig. 2 shows the effect of the mass flow rate and the inlet temperature of carbon dioxide and the volume of water on the decrease of water temperature in the storage tank. Notice that the
Fig. 4. The comparison of mass of ice from the experiment and the model.
volume of water is 1–3 L, the mass flow rate and the inlet temperature of carbon dioxide is varied between 0.003 and 0.017 kg/s and −35 °C and −50 °C, respectively. The initial temperature of the water is approximately 28 °C. The undoubtedly result shows that the decrease of water volume is directly proportional to the mass flow rate of carbon dioxide and inversely proportional to the volume of water and the inlet temperature of carbon dioxide. Moreover, the lumped model can predict the temperature of water in the storage tank quite well. Fig. 3 shows the effectiveness of a direct contact heat transfer between water and carbon dioxide. Notice that the effectiveness can be calculated from : Q e ¼ : CO2 ; ð4Þ Q max : : ð5Þ Q CO2 ¼ mCp CO2 TCO2 ;o TCO2 ;i ; : : Q max ¼ mCp CO2 Tw TCO2 ;i :
ð6Þ
From the figure, it is found that the effectiveness from this work agrees well with the results of Nuntaphan [4] which is close to 100%. However, for the water volume of 1 L, the effectiveness is slightly lower than those of 2 L and 3 L. This result comes from the shorter retention time of heat exchanging. In case of latent heat period, Fig. 4 shows the comparison of mass of ice from the experiment with the results from the model. It is found that the lumped model can predict the mass of ice approximately 77.3% within ± 10% variation. Notice that the ice from the direct contact heat transfer process has high porosity. Fig. 5 shows the ice formation at various conditions. In some conditions, there is a blockage of ice around the injector and it obstructs the flow of carbon dioxide. 4.2. Direct contact heat transfer of carbon dioxide and water– oil mixture
Fig. 3. Effectiveness of the direct contact heat transfer between water and carbon dioxide.
In this part, 11 samples of water–oil mixture are tested by exchange heat with low temperature carbon dioxide via direct
S. Thongwik et al. / International Communications in Heat and Mass Transfer 35 (2008) 756–761
759
Fig. 5. Ice formation at various states.
contact heat transfer technique. The surfactant, Tween 60, is used to merge the water and the compressor oil. The testing method is the same as the previous chapter and the compositions of water–oil mixture are given in Table 1. It is found that at the compositions of numbers 8–11, all of water–oil mixture becomes slurry. When the composition of compressor oil is lower than 6%, such as cases 1–7, a blockage of ice is formed around the injector. At these conditions, the compressor oil composition is rather low, therefore, part of non-merging water becomes plain ice and the blockage of ice around the injector occurs. In our experiment, the composition of water/compressor oil/Tween 60 at 100/6/1 is selected as a water mixture for producing slurry ice. Fig. 6 shows the slurry ice at this composition. In this part the lumped model is also used for predicting the water–oil mixture temperature and the mass of slurry ice during a time interval. Notice that, from our test, it is found that the specific heat and the enthalpy of freezing of water–oil mixture are close to that of pure water (lower than 1% difference). Therefore, the properties of water are used for calculating the temperature and mass of ice slurry. Fig. 7 shows the temperature of water–oil mixture and it is found that the lumped model can predict the temperature quite well. Fig. 8 shows the comparison of mass of slurry ice from the lumped model compared with that of the experiment and it is
found that the model can predict 100% of the experimental data within ± 10% variation. 4.3. Heat transfer analysis of direct contact heat exchange This research also studies the effect of parameters on the volumetric heat transfer coefficient (UV) of a direct contact heat transfer technique. In this work UV can be calculated from : Q CO2 ; UV ¼ V DT
ð7Þ
DT ¼ Tw TCO2 ;i :
ð8Þ
Where V is volume of water. During the sensible heat process, the volumetric heat transfer coefficient of this study is compared with the previous works and the results are shown in Table 2. Table 2 shows the volumetric heat transfer coefficient of the present study and is close to those of Goodwin et al. [8] and Kiatsiriroat et al. [6] for non-flow or stagnant storage medium. It is less than those of Blair [7] and Sidemen and Get [9] more than 10 times for injection of dispersed-phase agent into flowing water. In the case of the result of Goodwin et al. [8], the
760
S. Thongwik et al. / International Communications in Heat and Mass Transfer 35 (2008) 756–761
Table 1 Water mixture composition and form of ice No. Water/oil/Tween60 Color (by volume)
Solubility Blockage of ice Form of around injector ice
1
100/1/2
Good
Yes
2
100/3/0.5
Good
Yes
3
100/3/1
Good
Yes
4
100/3/2
Fair
Yes
5
100/3/4
Fair
Yes
6
100/3/6
Good
Yes
7
100/4/1
Fair
Yes
8
100/6/1
Fair
No
Slurry + plain ice Slurry + plain ice Slurry + plain ice Slurry + plain ice Slurry + plain ice Slurry + plain ice Slurry + plain ice Slurry
9
100/8/1
Fair
No
Slurry
10
100/9/1
Fair
No
Slurry
11
100/20/1
Fair
No
Slurry
Clear yellow Cloudy white Cloudy white Cloudy white Cloudy white Clear pink Cloudy white Cloudy white Cloudy white Cloudy white Cloudy white
volumetric heat transfer coefficient is still low (3–8 kW/m3 K) even though the water is continuously flowing. This result may come from the very high mass of water and mass flow rate at the storage tank then the dispersed-phase agent could not merge throughout the water in the whole tank. In this work, the models for predicting the volumetric heat transfer coefficient are also developed as for sensible heat: : 0:050828 UV ;S ¼ 729:31Mw0:99955 m 1:0117 ; CO2 DT
ð9Þ
storage medium temperature drops from its initial temperature at room temperature about 28 °C to its freezing point at 0 °C and the latent heat period occurs when the water temperature is at 0 °C till ice is forming. The values of the volumetric heat transfer coefficient from the correlations are compared with those from the experimental data for both sensible heat and latent heat periods as shown in Figs. 9 and 10. The model can predict all of the experimental values within ± 10%. Increasing of mass flow rate of carbon dioxide will get high turbulence in the storage medium thus high volumetric heat transfer coefficient is obtained. The heat transfer coefficient is less with the increasing of mass of water due to the decreasing of turbulence in the storage medium. It is found that temperature difference between water and carbon dioxide, has a slight effect on the volumetric heat transfer coefficient compared to those of previous parameters. 5. Conclusion
for latent heat: : 0:066906 UV ;L ¼ 704:52ðMw Mice Þ0:98204 m 1:0186 : CO2 DT
Fig. 7. The temperature of water mixture at various conditions.
ð10Þ
Both equations could be applied for pure water and water–oil mixture. The sensible heat period will be considered when the
From this research, it could be concluded as – The decreasing of water temperature is directly proportional to the mass flow rate of carbon dioxide and inversely
Table 2 Comparison of working conditions of direct contact heat transfer systems
Solution phase change
Fig. 6. Slurry ice from the composition of water mixture of water/oil/Tween 60 at 100/6/1.
Dispersed phase UV (kW/m3 K) Mass flow rate of dispersed phase (kg/s) Mass flow rate of water (kg/min) Diameter of storage tank (m)
Present study
Blair [7]
Goodwin et al. [8]
Sideman and Gat [9]
Kiatsiriroat et al. [6]
Water, water mixture CO2 1–25 0.003– 0.017
Water
Water
Water
Water
R113 122 0.023
Pentane 3–8 1.90
Pentane 180 0.022
R12, R22 2–16, 2–52 0.02–0.08
–
0.101
17.1
0.042
–
0.07
0.095
0.61
0.07
0.30
S. Thongwik et al. / International Communications in Heat and Mass Transfer 35 (2008) 756–761
– – – –
Fig. 8. The comparison of mass of ice from the experiment and the model.
761
proportional to the volume of water and the inlet temperature of carbon dioxide. The effectiveness of a direct contact heat transfer between carbon dioxide and water is close to 100%. The suitable composition of water–oil mixture, water/oil/ Tween, for producing slurry ice is at 100/6/1 by volume. The lumped model can predict the water temperature and the amount of ice in case of sensible heat and latent heat periods quite well. The volumetric heat transfer coefficient is inversely proportional to the mass of water and water mixture. On the other hand, the mass flow rate of carbon dioxide and temperature difference between inlet and outlet the storage tank are directly proportional.
Acknowledgements The authors would like to thank Thermal Technology Research Laboratory, Mae Moh Training Center, Electricity Generating Authority of Thailand for facilitating the testing equipment and the Commission on Higher Education, Thailand for funding this research study. References
Fig. 9. Comparison of volumetric heat transfer coefficient from the correlation and the experimental data during sensible heat period.
Fig. 10. Comparison of volumetric heat transfer coefficient from the correlation and the experimental data during latent heat period.
[1] C. Chaichana, W.S.C. William, A. Lu, An ice thermal storage computer model, Applied Thermal Engineering 21 (2001) 1769–1778. [2] J.E. Braun, Reducing energy cost and peak electric demand though optimal control of building thermal storage, ASHRAE Transaction 96 (2) (1990). [3] C.W. Carey, J.W. Mitchell, W.A. Beckman, The control of ice storage system, ASHRAE Transaction (1995) 32–39. [4] A. Nuntaphan, Performance analysis of a refrigeration cycle using a direct contact evaporator, M.Eng. Thesis, Department of Mechanical Engineering, Chiang Mai University, Thailand, (1998). [5] T. Kiatsiriroat, P. Siriplubpla, A. Nuntaphan, Performance analysis of a refrigeration cycle using a direct contact evaporator, Int. J. Energy Research 22 (1998) 1179–1190. [6] T. Kiatsiriroat, N. Vorayos, S. Vithayasai, A. Nuntaphan, N. Vorayos, Heat transfer prediction for a direct contact ice thermal energy storage, Energy Conversion & Management 44 (2003) 497–508. [7] C.K. Blair, Heat transfer characteristics of a three-phase volume boiler direct contact heat exchanger, M.S. Thesis, University of Utha, Salt Lake City, (1976). [8] P. Goodwin, M. Cobin, R. Boehm, Evaluation of the flooding limits and column, ASME National Heat Transfer Conference, Denver, 85-HT-49, 1985. [9] S. Sideman, Y. Gat, Direct contact heat transfer with change of phase: spray column study of three phase heat exchanger, AICHE Journal 95 (1) (1996) 1206–1213.