A two-stage deep freezing chemisorption cycle driven by low-temperature heat source

Abstract

A two-stage chemisorption cycle suitable for deep-freezing application driven by low- temperature heat source was proposed. Through two-stage desorption processes, the two-stage cycle can break through the limitations of the heating temperature and ambient cooling temperature. The two-stage cycle based on CaCl₂/BaCl₂-NH₃ working pair can utilize the heat source with a temperature of above 75°C, and simultaneously realize deep-freezing all the year round. Experimental results and performance prediction show that the adsorption quantity of calcium, theoretical coefficient of performance (COP) and optimized specific cooling power (SCP) of the CaCl₂/BaCl₂-NH₃ chemisorption system are 0.489 kg/kg (salt), 0.24 and 120.7 W/kg, when the heating temperature, ambient cooling temperature, pseudo-evaporating temperature and mass ratio of reacting salt and expanded graphite are 85, 30, -20, and 4∶1, respectively.

Keywords

adsorption / freezing / desorption / heat source

Introduction

As an environmentally benign and energy saving technology, adsorption refrigeration has recently attracted much attention. Adsorption refrigeration systems can be driven by low-grade heat with a temperature lower than 100°C which is abundant in many industries, and can also be supplied from solar energy [1,2].

For air conditioning conditions, Saha [3,4], Kasiwagi [3,4] and Akisawa [4] proposed a muti-stage cycle driven by waste heat at a temperature of 50°C, and it mainly adopted silica gel-water as working pair. However, such type of cycle has never been utilized in chemisorption due to its difference from physical adsorption. Physical adsorption performances are influenced by two parameters, the temperature and pressure. The operating principle of the physical adsorption cycle is illustrated in Fig. 1. When the condensing pressure is set as p_c, which corresponds to environmental temperature T_c, equilibrium desorption temperature varies from T_g1 to T_g3 for a single stage cycle. The maximum equilibrium desorption temperature (T_g2) of a two-stage physical adsorption cycle is much lower than that (T_gc) of a single stage physical adsorption cycle. By contrast, in physical adsorption processes, the chemisorption processes are only influenced by one parameter. Line 1-2, shown in Fig. 1, is the equilibrium chemisorption line. It shows that when the equilibrium pressure is set as p_c, the equilibrium reaction temperature is limited to T_g3, and it won’t be changed unless the equilibrium pressure varies.

For application in ice-making conditions driven by low temperature heat source, several adsorption working pairs have been researched, such as activated carbon-methanol, activated carbon-ammonia and mental chlorides-ammonia. Pons and Guilleminot [5] developed a prototype utilizing activated carbon-methanol as working pair, which could produce approximately six kilograms of ice per square meter of solar panel for solar radiation of approximately 20 MJ/d. Critoph [6,7] and Tamainot-Telto [6] mentioned a solar vaccine refrigerator which could maintain a cold box at a temperature of 0.1°C during daytime. Erhard et al. [8] studied a solar powered adsorption ice-making unit utilizing SrCl₂-NH₃ as working pair. This system operated about 2000 times. Its coefficient of performance (COP) was between 0.05 and 0.08, whereas its specific cooling power (SCP) per kilogram of adsorbent was approximately 10 W/kg.

The evaporating temperatures of existing adsorption ice-making systems mentioned above are all higher than -15°C. Moreover, most of these systems have to be powered by a heat source with a temperature higher than 85°C. At present, no adsorption working pairs are suitable for deep-freezing application (refrigeration temperature lower than -15°C), when the heating temperature is below 90°C and ambient temperature higher than 30°C [9]. To realize deep-freezing applications driven by low temperature heat source, a new type of cascaded thermochemical system utilizing BaCl₂-NH₃ as working pair was developed by Le Pierres et al. [10]. This system comprised two reactors, two condensers, and two evaporators (one evaporator was inside one of the reactors). A refrigeration temperature of -33°C could be obtained when the ambient temperature was 25°C. However, the drawback of such a system lies in its complex structure. Two reactors were required in the cascading ice-making systems, and the refrigeration effect of one reactor (reactor I) was thermally linked to the other reactor (reactor II). The theoretical coefficient of performance (COP_i) for one single system in the cascading systems was lower than 0.5, which means that reactor I should be two times larger than reactor II, since reactor I provided cooling power for reactor II in order to get lower refrigeration temperature. Such a structure will lead to a COP theoretically lower than 0.17 [10].

For deep-freezing condition application driven by a low temperature heat source, a two-stage adsorption refrigeration cycle is proposed in this paper in order to realize a cooling output lower than -15°C when heating temperature is between 70°C and 90°C. Although two different metal chlorides working in different temperature ranges are necessary, the structure of this type of two-stage adsorption system is relatively simple because only one condenser and one evaporator are required.

Adsorption working pairs comparison

Adsorption working pairs for ice-making condition are mainly activated carbon-methanol, activated carbon-ammonia, and mental chloride-ammonia.

The adsorption and desorption performances of activated carbon-methanol and activated carbon-ammonia working pairs were tested, and the main problem of these two working pairs is low concentration for ice-making condition [9]. The maximum concentration is only 0.02 kg/kg while the evaporating temperature, ambient temperature and heating temperature are -20°C, 25°C and 100°C, respectively. Besides, that value comes close to zero when the heating temperature is lower than 90°C.

Metal chloride-ammonia working pairs suitable for utilization of low temperature heat source are mainly BaCl₂-NH₃, SrCl₂-NH₃ and CaCl₂-NH₃. In deep-freezing application driven by low temperature heat source, the single-stage chemisorption cycle is mainly limited to unattainable equilibrium adsorption temperature in adsorption process or equilibrium desorption temperature in desorption process. The equilibrium adsorption temperature is as low as 12°C when pseudo-evaporating temperature comes to -25°C for the BaCl₂-NH₃(8-0) single-stage chemisorption cycle, as shown in Fig. 2 [11]. As to the CaCl₂-NH₃(8-4), SrCl₂-NH₃(7-1) and CaCl₂-NH₃(4-2) single-stage chemisorption cycles, equilibrium desorption temperatures are 90°C, 95°C and 102°C, respectively, when ambient cooling temperature is 35°C.

None physical adsorption or chemisorption working pairs are suitable for deep-freezing application driven by low-temperature heat source. New ways to break through limitations of ambient cooling temperature and heating temperature should be found to meet demands of deep-freezing condition.

Design of two-stage chemisorption cycle

A novel two-stage chemisorption cycle is proposed to realize deep-freezing when the heating temperature is lower than 90°C and ambient cooling temperature as high as 35°C. As demonstrated in Fig. 3, the two-stage chemisorption cycle needs two kinds of metal chlorides working in different temperature range, such as barium chloride and calcium chloride. The calcium chloride and barium chloride serve as high temperature salt and low temperature salt, respectively, since the equilibrium reacting temperature of calcium chloride is higher than that of barium chloride under the same equilibrium reacting pressure.

The two-stage chemisorption cycle comprises two-stage desorption processes and adsorption refrigeration process. During the first-stage desorption process, the high temperature adsorber is heated by heat source Q_H1, and the low temperature adsorber cooled by cooling source Q_L2 at ambient temperature. The refrigerant is desorbed from the high temperature adsorber to the low temperature adsorber, accompanied by the regeneration process of high temperature salt. During the second-stage desorption process, the low temperature adsorber is heated by heat source Q_H2 while the condenser is still cooled by cooling source Q_L1 at ambient temperature. The refrigerant is desorbed from the low temperature adsorber and condensed in the condenser with the regeneration process of low temperature salt. The adsorption refrigeration process proceeds at the same time as the second-stage desorption process. The high temperature adsorber is cooled by cooling source Q_L1 at ambient temperature and adsorbs the refrigerant from the evaporator. The evaporation of liquid refrigerant produces cooling effect.

The operating principles of the two-stage chemisorption cycle based on CaCl₂/BaCl₂-NH₃ working pair is presented in Fig. 4 [11]. In the two-stage chemisorption cycle, the maximum desorption temperature is 75°C when cooling temperature is set at 35°C and equilibrium adsorption temperature during adsorption refrigeration process is 38°C when pseudo-evaporating temperature comes to -25°C. All of these imply that the two-stage chemisorption cycle can realize deep-freezing application when the heating temperature is higher than 75°C and cooling temperature is lower than 38°C. Thus, such a two-stage cycle is theoretically suitable for utilization of solar energy and industrial waste heat with a temperature above 75°C and can realize deep-freezing application all the year round.

Adsorption performance test

Experimental test unit and results

The performance testing unit mainly comprises two adsorbers, an evaporator/condenser, a pressure transmitter and four platinum resistance thermometers, as exhibited in Fig. 5. The pseudo-evaporating temperature is controlled by the ethylene glycol circulation in the jacket. An ambient temperature -37°C can be realized by the cooling ethylene glycol with a volumetric concentration set at 50%. The adsorption quantities are calculated by the change of pressure difference between the two ends of liquid ammonia column and vapour ammonia column, as expressed in Eq. (1). To ensure a relatively high measurement accuracy, the vapor tube of the pressure transmitter is heated with nickel-chromium wires to prevent condensation of ammonia by certain heat input. The photograph of the test unit is displayed in Fig. 6.

(1)\mathrm{\Delta }x=\frac{1}{m_{salt}}·\mathrm{\Delta }\left[\left(1-\frac{{\nu }^{\text{'}}\left(T_e\right)}{{\nu }^{\text{'}\text{'}}\left(T_e\right)}\right)·\frac{A_c}{\mathrm{g}}·\mathrm{\Delta }p+\frac{V}{{\nu }^{\text{'}\text{'}}\left(T_e\right)}\right].

In which, Δxis cycle adsorption quantity, kg/kg (salt); {\nu }^{\text{'}} and {\nu }^{\text{'}\text{'}} are specific volume of saturated vapor ammonia and liquid ammonia, respectively, m³/kg; A_c is the effective area of cross section of ammonia in the evaporator/condenser, m²; V is the volume of the evaporator/condenser, m³; Δp is the pressure difference between the two ends of liquid ammonia column and vapor ammonia column,, Pa; m_salt is the mass of the reacting salt, kg; and g is the gravity acceleration, m/s².

The expanded graphite in the adsorbents can prevent agglomeration during the reacting process to improve the heat and mass transfer performances. It can be manufactured by heating natural expandable graphite in an oven with an ambient temperature of 700°C for 12 to 15 min. The characteristic parameters of the adsorbents are listed in Table 1. The molar ratio of calcium chloride and barium chloride is 8∶6, which corresponds to their theoretical adsorption density.

As previously mentioned, the two-stage chemisorption cycle comprises two-stage desorption processes and adsorption refrigeration process. Only when the first-stage desorption process were finished can the adsorption process of calcium chloride begin. In the experiments, the temperature ranges of heating, cooling and pseudo-evaporating temperatures were 70°C to 90°C, 25°C to 35°C and -25°C to+5°C, respectively, and the temperature gradient was taken as 5°C. As to a certain compound condition, the heating temperatures were the same in the two stage desorption processes, and so were the cooling temperatures the three reacting processes. The adsorption quantities of calcium chloride in the adsorption process are summarized in Table 2.

The adsorption quantity increases when the heating temperature and pseudo-evaporating temperature increase and cooling temperature decreases. The maximum value is 0.87 kg/kg (salt) (approximately 94.5% of the theoretical value), when the heating, cooling and pseudo-evaporating temperatures are 90°C, 25°C and 5°C, respectively. Especially, the adsorption quantity of calcium chloride are more than 0.3 kg/kg (salt) even when the heating, cooling and pseudo-evaporating temperatures are 70°C, 30°C and -20°C, respectively.

The desorption quantities of barium chloride shown in Table 3 are only related to the heating temperature and cooling temperature, since the pseudo-evaporating temperature was set at 5°C. The data at different pseudo-evaporating temperatures can also be calculated by conservation of mass of reacting ammonia, as expressed in Eq. (2).

(2)m_{ca}·\mathrm{\Delta }{x}_{ca}=m_{ba}·\mathrm{\Delta }{x}_{ba}.

Experimental error analysis

The adsorbent mass is measured by the balance (BS2202S) with a measuring error of±0.01 g. The pressure difference between the vapor end and liquid end of the evaporator/condenser is tested by the smart pressure transmitter (Fig. 5), whose testing error is 0.2%. According to Eq. (3), the relative errors of adsorption/desorption quantities of calcium chloride and barium chloride are 3.6% and 1.3%, respectively.

(3)\left|{\displaystyle \frac{\mathrm{d}\left(\mathrm{\Delta }x\right)}{\mathrm{\Delta }x}}\right|\le \left|{\displaystyle \frac{\mathrm{d}{m}_{\mathrm{s}\mathrm{a}\mathrm{l}\mathrm{t}}}{m_{\mathrm{s}\mathrm{a}\mathrm{l}\mathrm{t}}}}\right|+\left|{\displaystyle \frac{\mathrm{d}\left(\left(1-{\nu }^{\prime }\left(T_{\mathrm{e}}\right)/{\nu }^{\prime \prime }\left(T_{\mathrm{e}}\right)\right)·\left(A_{\mathrm{c}}/g\right)·\mathrm{\Delta }\left(\mathrm{\Delta }p\right)\right)}{\mathrm{\Delta }\left(\left(1-{\nu }^{\prime }\left(T_{\mathrm{e}}\right)/{\nu }^{\prime \prime }\left(T_{\mathrm{e}}\right)\right)·\left(A_{\mathrm{c}}/g\right)·\mathrm{\Delta }p+V/{\nu }^{\prime \prime }\left(T_{\mathrm{e}}\right)\right)}}\right|.

Performance prediction of two-stage chemisorption system

In the two-stage chemisorption system, two heat inputs are needed to achieve cooling output. The theoretical COP_i considering sensible heat of refrigerant can be calculated by Eqs. (4)–(6). The COP of the two-stage chemisorption system based on CaCl₂/BaCl₂-NH₃ working pair is approximately 0.23 to 0.26 when the ranges of cooling temperature and pseudo-evaporating temperature are 25°C to 35°C and -25°C to 5°C, respectively, as shown in Fig. 7.

(4)Q_{vap}=m_{ca}\cdot \mathrm{\Delta }{x}_{ca}\cdot \left[\mathrm{\Delta }{H}_{vap}-C_p\cdot \left(T_c-T_e\right)\right],

(5)Q_{heat}=m_{ca}\mathrm{\Delta }{x}_{ca}\mathrm{\Delta }{H}_{r-ca}+m_{ba}\mathrm{\Delta }{x}_{ba}\mathrm{\Delta }{H}_{r-ba},

(6)COP=\frac{\mathrm{\Delta }{H}_{vap}-{\mathit{\text{C}}}_p\cdot \left(T_c{-T}_e\right)}{\mathrm{\Delta }{H}_{r-ca}+\mathrm{\Delta }{H}_{r-ba}}.

The SCP has a great impact on the commercialization process of the adsorption system. As to the CaCl₂/BaCl₂-NH₃ two-stage chemisorption system, the mass ratio of barium chloride and calcium chloride is derived as 1.405 from the theoretical molar ratio of the two mental chlorides. The mass of calcium chloride and barium can be calculated if the cooling power of the system is specified at a certain value. The optimization strategy of SCP takes the real adsorption/desorption quantities into consideration, and the results with different ratios of salt and expanded graphite are given in Fig. 8.

The heating temperature, cooling temperature and cycle period are 85°C, 30°C and 30 min, respectively. The optimized SCP is 110.9 to 146.8 W/kg when pseudo-evaporating temperature increases from -25°C to 5°C and the mass ratio of metal chloride and expanded graphite is 4∶1. Especially, optimized SCP of the two-stage chemisorption system increases approximately by 22% when pseudo-evaporating temperature is -20°C.

Conclusions

A two-stage chemisorption cycle suitable for deep freezing application driven by low temperature heat source was proposed, and performance prediction of the CaCl₂/BaCl₂-NH₃ two-stage chemisorption system based on experimental results of adsorption/desorption quantities was made. The main conclusions are as follows.

1) In the two-stage chemisorption cycle, the regeneration processes of high temperature salt and low temperature salt are finished by the two stage desorption processes. Compared to the single-stage chemisorption cycle, the two-stage cycle can break through the limitations of heating temperature and ambient cooling temperature. Dynamic characteristic analysis shows that the two-stage chemisorption cycle based on the CaCl₂/BaCl₂-NH₃ working pair is suitable for utilization of solar energy and industrial waste heat with a temperature of above 75°C and can realize deep-freezing application all the year round.

2) Experimental study shows that the two-stage chemisorption cycle can utilize the heat source with a temperature of above 70°C, and most of the adsorption quantities of calcium chloride under compound operating conditions are above 0.3 kg/kg (salt). The COP_i of the two-stage chemisorption system is 0.23 to 0.26, considering the sensible heat of liquid refrigerant. The optimized SCP is above 110 W/kg, when the heating temperature, ambient cooling temperature, pseudo-evaporating temperature, and mass ratio of salt and expanded graphite are 85°C, 30°C, -25°C to 5°C and 4∶1.

References

[1] Wang D C, Xia Z Z, Wu J Y, Wang R Z, Zhai H, Dou W D. Study of a novel silica gel-water adsorption chiller. Part I. Design and performance prediction.International Journal of Refrigeration, 2005, 28(7): 1073-1083

[2] Anyanwu E E. Review of solid adsorption solar refrigerator I: an overview of the refrigeration cycle.Energy Conversion and Management, 2003, 44(2): 301-312

[3] Saha B B, Kashiwagi T. Experimental investigation of an advanced adsorption refrigeration cycle.ASHRAE Trans, 1997, 103(2): 50-58

[4] Saha B B, Akisawa A, Kashiwagi T. Solar/waste heat driven two-stage adsorption chiller: the prototype.Renewable Energy, 2001, 23(1): 93-101

[5] Pons M, Guilleminot J J. Design of an experimental solar powered solid-adsorption ice maker.ASME-Journal of Solar Energy Engineering,1986, 108(4): 332-337

[6] Critoph R E, Tamainot-Telto Z. Solar sorption refrigerator.Renewable Energy, 1997, 12(4): 409–417

[7] Critoph R E. Laboratory testing of an ammonia carbon solar refrigerator.In:ISES, Solar World Congress, Budapest, Hungary,1993

[8] Erhard A, Spindler K, Hahne E. Test and simulation of a solar powered solid sorption cooling machine.International Journal of Refrigeration, 1998, 21(2): 133-141

[9] Wang L W, Wang R Z, Wu J Y, Wang K, Wang S G. Adsorption ice makers for fishing boats driven by the exhaust heat from diesel engine: choice of adsorption pair.Energy Conversion and Management,2004, 45(13,14): 2043-2057

[10] Le Pierres N, Mazet N, Stitou D. Experimental results of a solar powered cooling system at low temperature.International Journal of Refrigeration, 2007, 30(6): 1050-1058

[11] Neveu P, Castaing J. Solid-gas chemical heat pumps: field of application and performance of the internal heat of reaction recovery process.Heat Recovery Systems and CHP, 1993, 13(3): 233-251

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag Berlin Heidelberg