Experimental study of an automobile exhaust heat-driven adsorption air-conditioning laboratory prototype by using palm activated carbon-methanol.INTRODUCTION
Air conditioning in an automobile is basically the production of desired indoor air conditions, independent of the outdoor conditions. However, air-conditioning technology is required to evolve due to the new environmental regulations, notably Montreal Protocol in 1987 and European Commission Regulation 2037/2000. These regulations were put in place in response to concerns about the depletion of the ozone layer and an increase in global warming, which began a phase-out of first chlorofluorocarbons (CFCs) then hydro-chlorofluorocarbons (HCFCs). As a result, this trend has led to a strong demand for a new air-conditioning technology. Among the proposed air-cooling technologies, the adsorption air-cooling system has shown potential. The advantages of this system are that it's long lasting, has low-cost maintenance, uses non-polluting refrigerants, is environmentally friendly, and can be powered by heat or solar waste (Dieng and Wang 2001). Unfortunately, no working prototype has been practically run in present automobiles due to various restrictions, including sizing and cooling capacity limitations. In general, the adsorption cycle can be categorized into two main cycles: the intermittent cycle and the continuous cycle. The intermittent cycle seems unsuitable for automobile application because it cannot provide continuous cooling as needed. Thus, a continuous cycle was adopted in this research to continuously produce a cooling effect via two adsorbers (containing four total adsorbent beds) that operate intermittently.
A preliminary study was performed by Suzuki (1993) to elucidate the technological limits associated with the application of adsorption cooling systems to automobiles. But while the study showed some potential for utilizing this method in vehicle air conditioning, the author performed a simulations study only and no experimental work was carried out to verify his claim. Aceves (1996) had carried out an experimental analysis of the applicability of an adsorption system for electric vehicle air conditioning. The coefficient of performance (COP) of the system (with zeolite and water as a working pair) was approximately 0.28. His studies indicated that conventional compression air conditioners were superior to adsorption systems due to their higher COPs and more compact size. The drawback of using zeolite and water as a working pair is that a very low operating pressure is needed. Meanwhile, Sato et al. (1997) have presented a multiple-stage adsorption air-conditioning system for vehicles. Although the efficiency of the multiple-stage adsorption system was improved, the size of the system also increased and its control system became more complex. Zhang (2000) has described an experimental intermittent adsorption cooling system driven by the waste heat of a diesel engine. Zeolite 13X-water is used as the working pair and a finned double-tube heat exchanger is used as the adsorber. The COP and specific cooling power (SCP) of the system is 0.38 and 25.7 W/kg, respectively. Wang et al. (2001) have studied an adsorption air conditioning for a bus driven by waste heat from exhausted gases. The working pair for this system is activated carbon and ammonia with the cooling power of 2.58 kW and a COP of 0.16. The activated carbon is pressurized to a density of about 900 kg/[m.sup.3] in to fit additional adsorbent into the adsorber. The total weight of the two adsorbers is about 248 kg and occupied about 1.0 [m.sup.2].
Lu et al. (2004) presented experimental studies on the practical performance of an adsorption air-conditioning system powered by exhausted heat from a diesel locomotive. The system was incorporated with one adsorbent bed and utilizes zeolite and water as a working pair to provide chilled water for conditioning the air in the driver's cab of the locomotive. Their experimental results showed that the adsorption system is technically feasible and can be applied for space air conditioning. Under typical running conditions, the average refrigeration power ranging from 3.0 to 4.2 kW has been obtained. However, this system may not be suitable for automobile application due to its size and high regenerative temperature. Inoue et al. (2006) have described an air-conditioner that uses the cooling water of an internal combustion engine, which includes a compressive refrigerator and an adsorption type refrigerator. The compressive refrigerator is used to control the temperature of the air to be blown into a passenger compartment of the vehicle. The adsorbent generates adsorption heat when the adsorbent adsorbs the adsorbate, and desorbs the adsorbate when the adsorbent is heated by coolant water from the internal combustion engine. On the other hand, Henning and Mittelbach (2006) have disclosed an adsorption heat pump for air conditioning a passenger car. Their system is based on a quasi-continuous operation of adsorption heat pump with the used of cold and heat accumulators.
In our research work, a laboratory prototype of exhaust heat-driven adsorption air-conditioning system was designed, built, and tested in laboratory to study the replacement of a conventional vapor compression air-conditioning system in automobile. The prototype consists of a novel adsorbers system design, coupled with bioresource material of high adsorptive capacities for present invention application.
The current prototype consists of an exhaust heat-driven adsorption air-conditioning system, which is comprised of adsorbers, flow control modules, an evaporator, a condenser, an expansion valve, blowers, and an engine. These adsorbers generally have the same function as the mechanical compressor in a conventional vapor-compression system. However, these adsorbers contain adsorptive material and adsorbate as working pairs, instead of using CFCs or HCFCs, to generate the cooling effect required by using heat from the exhaust gas. As a result, these adsorbers are also called "thermal compressors." These adsorbers are first linked to the condenser via the flow control module, which consist of a few check valves. The condenser is then connected through an expansion valve to the evaporator, which in turn is connected back to the adsorbers. The associated adsorbers, condenser, expansion valve, and evaporator are integrated in a closed-loop operation. The engine, blower, and exhaust passage are integrated in an open-loop operation.
The current prototype exhibits several advantages compared to the conventional compression-based air-conditioning system commonly used in contemporary automobiles, such as:
1. Fuel consumption and unwanted gas emission (such as nitrogen oxides and carbon dioxide) would be reduced as overall engine load required is decreased.
2. It is more environmentally friendly as methanol (which leads to neither ozone depletion nor global warming) was used as a working fluid compared with conventional refrigerants.
3. A low regeneration temperature (less than 300 [degrees]F [150.0 [degrees]C]) can be used to power the system.
4. It operates using fewer moving parts, necessitates lower maintenance costs, and has a simple system structure.
Adsorbent and Adsorbate Pair
In our research work, the biomass of granular palm-derived activated carbon (Figure 1) and methanol was used as a working pair. Palm-derived activated carbon (Table 1) was selected because it is locally available, offers a high adsorptive capacity, and is also low cost. Methanol, which has low boiling point of about 148[degrees]F (64.5[degrees]C), is used since a low regeneration temperature is required for the operation of the adsorption system. Besides, methanol has a high latent heat of vaporization of ~ 473 Btu/lb (1100 kJ/kg), which is essential for increasing the cooling capacity and reducing the quantity of adsorbate used.
[FIGURE 1 OMITTED]
Table 1. Properties of the Palm-Derived Activated Carbon Property Value Particular shape Granular (size < 0.12 in. or 3.0 mm) Density 26.90 lb/[ft.sup.3] or 0.431 g/[cm.sup.3] Heat of adsorption 774 Btu/lb or 1800 kJ/kg Iodine number 1180 Total pore volume 13.8-16.6 [in.sup.3]/lb or 0.5-0.6 [cm.sup.3]/g Surface area 4,882,669-5,370,936 [ft.sup.2]/lb or 1000-1100 [m.sup.2]/g Moisture (% max) Below 5%
Development of the Prototype
The most crucial and complex part in this prototype is the adsorbers. CATIA 3D, a graphic software was used to design the adsorber. The adsorbers were designed in such as way (illustrated in Figure 2) as to maximize the quantity of activated carbon and to improve the heat transfer between the adsorbent and exhaust gas. Two identical adsorbers were constructed, each adsorber consisting of two adsorbent beds Each adsorbent bed was packed with approximately 1.8 lb (0.8 kg) of granular palm-activated carbon in a stainless steel net. The dimensions of the adsorbers are 1.31 ft (40 cm) in length, 0.66 ft (20 cm) in width, and 0.33 ft (10 cm) in height. Six radial stainless steel fins symmetrically distributed in the adsorbent bed are employed to intensify heat conduction. A four-stroke petrol 5HP (3.7 kW) engine was used to supply the heat source required during the regeneration process. The heat from the exhaust gas can reach over 300[degrees]F (150.0[degrees]C), which was more than enough to operate the prototype. The condenser used is a type of air-finned-tube aluminum heat exchangers that is attached with a 12V d.c. fan to increase the heat rejection rate. Meanwhile, a hanging type of air-finned-tube aluminum heat exchanger, which consists of a cooling coil of 3412 Btu/h (1.0 kW) and two blowers powered by a 12V d.c. motor with a motor speed controller were integrated to the prototype. The detail design of the complete system is given in Abdullah and Leo (2008).
[FIGURE 2 OMITTED]
Figure 3 shows the placement and integration of the components in the prototype, which consists of two adsorbers, a blower, an evaporator attached to two blowers, a condenser attached to a fan, an expansion valve, four check valves, three three-way valves, an engine, and several pipe connectors. Before the prototype is ready to be tested, it is evacuated and charged with 24.4 [in.sup.3] (400 mL) of methanol. The quantity of methanol charged was lower compared to the adsorption capacity of activated carbon in order to prevent the activated carbon from becoming saturated, which could reduce the system performance.
[FIGURE 3 OMITTED]
Operation of the Prototype
This prototype generally works in two main phases; desorption (regeneration) phase and adsorption phase. Figure 4 shows the schematic diagram of the entire system during the first half-cycle of operation. As shown in this figure, the system begins the process when Adsorber 1 is heated by the exhaust gas released from the engine. At the same time, Adsorber 2 is cooled by the blower. After a few minutes of heating, the adsorber temperature can be raised up to 248[degrees]F (120.0[degrees]C). The three-way valves (Valve 1 and Valve 2) are used to divert air from the blower and divert exhaust gas to the adsorbers, respectively, while another three-way valve (Valve 3) is used to bypass the exhaust heat to prevent the adsorbers from overheating. During the heating process, methanol is desorbed and then pressurized by the adsorber. Due to the high-pressure difference, the check valve (Valve 4) that connected Adsorber 1 to the condenser is automatically opened and Valve 5 is closed. The high temperature and pressure methanol vapors are then transmitted to the condenser. When the methanol vapors touch the cool internal surface of the condenser, the vapors are condensed to form a lower temperature high-pressure liquid. A check valve (Valve 8) was placed near the inlet of the condenser to avoid reversing the process and to prevent methanol liquid from accumulating inside the tube. To increase the heat rejection rate, a ten-blade fan was mounted at the back of the condenser.
[FIGURE 4 OMITTED]
As Adsorber 1 is continuously heated, more methanol is desorbed from the adsorbent. Consequently, the pressure increased and forces the methanol liquid to travel via an expansion valve (Valve 9). A filter was placed at the inlet of the valve to prevent any dust or impurity from blocking it, which can cause malfunction of the system. When the high-pressure methanol liquid enters the evaporator, it vaporizes spontaneously due to lower pressure inside the evaporator. When the methanol vaporizes, it absorbs large amounts of heat from cooling space. These vapors are then adsorbed by Adsorber 2, low in both pressure and temperature, where Valve 6 is closed while Valve 7 is opened. In this half cycle of operation, Adsorber 1 became discharge side while Adsorber 2 acted as suction side to generate cooling effect in the evaporator.
During the second half cycle, exhaust gas was diverted to heat Adsorber 2 (desorption phase) while Adsorber 1 (adsorption phase) was cooled by the blowing air. In this half cycle, Adsorber 1 acted as suction side (Valve 4 is closed and Valve 6 is opened) while Adsorber 2 became discharge side (Valve 5 is opened and Valve 7 is closed). The temperature of Adsorber 2 increased and caused the methanol to desorb from the activated carbon. The methanol vapors then traveled to the condenser and condensed. On the other hand, the temperature of Adsorber 1 decreased by the air blown from the blower. The same processes as the previous half cycle are repeated but now the methanol vapors from the evaporator is adsorbed by Adsorber 1. As a result, a continuously cooling effect was achieved by merely providing means of heating and cooling of the adsorbers intermittently.
The prototype was tested inside an open laboratory, where the exhaust gas was allowed to flow out to the surrounding. A test chamber, made by using Perspex, was built and installed at the evaporator outlet to reduce the effect of the engine heat and flowing air on the experiment results. During the test run, K Type thermocouples were used to measure temperature variation of the evaporator, condenser and the engine. Two thermocouples were attached to the evaporator, one thermocouple located at the back and another one at the front of the evaporator coil, to measure the temperature variation of the inlet air and outlet air through the cooling coil. Thermocouples were also placed near the inlet and outlet of the condenser to determine the average condensation temperature of the methanol vapors, while another thermocouple was located at the exhaust outlet for measuring exhaust temperature. Besides, thermocouples were used to measure the adsorbers temperature at different locations, where average readings were taken. In addition, the pressures between the adsorber-condenser and evaporator-adsorber were measured by using two compound vacuum gauges (range: -1 to +3 bars). A humidity meter was used to measure the variation of humidity inside the laboratory before and during the experiments.
The performance of the adsorption cooling system is commonly evaluated using two performance factors: the coefficient of performance (COP) and specific cooling power (SCP). In general, COP is the amount of cooling produced by an adsorption cooling system per unit heat supplied (Anyanwu 2004; Zhang 2000) as shown below:
COP = [[Q.sub.ev]/[Q.sub.de]] (1)
[Q.sub.ev] is the quantity of heat transferred through the evaporator
[Q.sub.de] is the quantity of heat adsorbed by the adsorber during the desorption phase
The SCP, on the other hand, is defined as the ratio between the cooling production and the cycle time per unit of adsorbent weight, as given below:
SCP = [[Q.sub.ev]/([t.sub.eye][m.sub.ac])] (2)
[t.sub.cyc] is the cycle time
[m.sub.ac] is mass of the activated carbon
Since SCP relates to both the mass of adsorbent and the cooling power, it reflects the size of the system. For a nominal cooling load, higher SCP values indicate the compactness of the system.
RESULTS AND DISCUSSION
Temperature Profiles of the Prototype
A series of experiments have been carried out to determine the best operating temperature for this prototype. Figure 5 presents a simple T-S diagram of the prototype and its design operating temperatures. With these operating conditions, the system was operated to show the heat distributions profile of the adsorbers (as shown in Figure 6) by using a thermography camera. In this figure, the Adsorber 2 was in the desorption phase while Adsorber 1 was in adsorption phase. During the desorption phase, the adsorber was heated by the exhaust heat to an average temperature of 257 [degrees]F (125[degrees]C). At the same time, the other adsorber was cooled by the blowing air and adsorb methanol vapor from the evaporator.
[FIGURE 5 OMITTED]
[FIGURE 6 OMITTED]
Variation of Temperature with Various Types of Pressure Regulating Devices
Two common types of pressure regulating devices were tested to determine the most suitable valve that could provide the lowest cooling temperature in the shortest time, the thermal expansion valve and orifice tubes (orifice size: 0.012 in. [0.3 mm], 0.016 in. [0.4 mm], 0.020 in. [0.5 mm], and 0.024 in. [0.6 mm]). The initial ambient temperature during all the experiments was controlled around 84[degrees]F (29.0[degrees]C) with an initial relative humidity around 85% that was decreased to about 60% during operation. The data gathered was presented in Figure 7, which showed the trend of the cooling coil temperature over time for the first half-cycle during startup operation. Based on this figure, the cooling generated in the cooling coil by using a thermal expansion valve was the least successful compared to orifice tubes. It also showed that the size of the orifice influenced the cooling generated and also the time taken to start cooling. Among the four types of orifice tubes, the 0.012 in. (0.3 mm) orifice tube produced the highest cooling effect but took the longest time whereas 0.024 in (0.6 mm) orifice tube produced the lowest cooling effect in shortest time. In order to increase the system's efficiency, the cooling temperature must be lowest and the time taken to achieve this cooling must be shortest. As a result, the 0.020 in. (0.5 mm) orifice tube was the best compromise between desired cooling effect and low cycle time.
[FIGURE 7 OMITTED]
Variation of Temperature during Operations
A series of experiments have been conducted to determine the cycle time of the system, where it was set to be around 20 minutes per cycle. This means every 10 minutes of operation, the adsorber in desorption phase was shifted to adsorption phase and vice-versa. Figures 8a and 8b showed the temperature variation of the cooling coil and the chilled air for five sets of data collected under the same operational condition. Based on these experimental results, the average temperature of the cooling coil was around 53[degrees]F (11.4[degrees]C) with the temperature range between 49 [degrees]F and 58[degrees]F (9.5[degrees]C and 14.7[degrees]C), while the temperature range of the chilled air fluctuated between 69 [degrees]F and 77[degrees]F (20.7[degrees]C and 25.2[degrees]C) with an average temperature of 73[degrees]F (22.6[degrees]C). During the transition period of desorption and adsorption phases, the cooling coil temperature was increased due to lack of methanol flow into the evaporator. However, the cooling coil temperature started to decrease again when the methanol vapors released from Adsorber 2 were condensed in the condenser and reached the evaporator. Simultaneously, Adsorber 1 adsorbs the methanol vapors from the evaporator. Every 10 minutes of operation, the exhaust gas was diverted to heat Adsorber 1 while Adsorber 2 was cooled by the blower. The same processes were repeated and a continuously cooling effect was produced by using two adsorbers operated intermittently.
[FIGURE 8 OMITTED]
Performance of the Prototype
From the experimental results, the cycle COP of the prototype is approximately 0.19 and the SCP is around 614 Btu/h-lb (396.6 [Wkg.sup.-1] ). As the quantity of palm-activated carbon used was approximately 3.5 lb (1.6 kg) in each adsorber, the total cooling power that can be achieved is only 635 W. Table 2 shows the comparison of COP, SCP, and cooling power for the current research with some related previous research. The COP obtained herein is average compared to the work performed by other researchers. However, the SCP found is quite high ~ 400 W per kg of activated carbon used for the present study.
Table 2. Comparison of COP, SCP and Cooling Power Investigators Heat Working Pairs COP SCP Cooling Source (W/kg) Power (kW) Present study Exhaust Activated 0.19 396.6 0.635 heat carbon-methanol Aceves (1996) Electric Zeolite/water 0.28 - - heater Zhang (2000) Exhaust Zeolite 0.38 25.7 - heat 13X-water Wang et al. Exhaust Activated 0.16 - 2.58 (2001) heat carbon- ammonia Lu et al. Exhaust Zeolite/water 0.18-0.21 - 3.0-4.2 (2004) heat
The performance of the current prototype is lower compared to the conventional vapor-compression system (COP generally more than 1) due to a low heat transfer rate to the activated carbon during the desorption phase and from the activated carbon during the adsorption phase. It is to be noted that a vapor compression system is driven by electrical power; whereas the system in this study is driven by exhaust heat as the conversion efficiency from heat to power is actually accounted for.
A working prototype of exhaust heat-driven adsorption air-conditioning system using palm-derived activated carbon-methanol has been built, commissioned and laboratory tested. The experimental results showed the employment of adsorption technologies in automobile air-conditioner are feasible and promising; however, there is a need to further enhance the efficiency and the associated control system for effective on-the-road application.
The present study is supported by the Universiti Malaysia Sarawak's Special Project Grant No. DI/04/2007/(04). The first author would like to thank the Ministry of Science, Technology, and Innovation of Malaysia for the MOSTI fellowship award. Both authors would like to thank all staff for their continuous encouragements throughout this project.
Aceves, S.M. 1996. Analytical comparison of adsorption and vapor compression air conditioners for electric vehicle applications. Energy Resources Technology, Transactions of the ASME 118(1):16-21.
Abdullah, M.O., and S.L. Leo. 2008. Heat driven adsorption air-conditioning system for automobile. Malaysian Patent Application PI20081641, submitted May 6, 2008.
Anyanwu, E.E. 2004. Review of solid adsorption solar refrigeration II: An overview of the principles and theory. Energy Conversion and Management 45(7):1279-95.
Dieng, A.O., and R.Z. Wang. 2001. Literature review on solar adsorption technologies for ice-making and air conditioning purposes and recent developments in solar technology. Renewable and Sustainable Energy Review 5(4):313-42.
Henning, H.M., and W. Mittelbach. 2006. Passenger car air-conditioning systems with adsorption heat pumps. U.S. Patent WO 2006/048244, published November 5, 2006.
Inoue, S., S. Ooishi, H. Satoh, and H. Mieda. 2006. Air-conditioner for vehicle. Japan Patent JP2005212735.
Lu, Y.Z., R.Z. Wang, S. Jianzhou, Y.X. Xu, and J.Y. Wu. 2004. Practical experiments on an adsorption air-conditioner powered by exhausted heat from a diesel locomotive. Applied Thermal Engineering 24:1051-9.
Sato, H., S. Honda, H. Tanaka, and T. Terao. 1997. Adsorptive type refrigeration apparatus. U.S. Patent 5619866.
Suzuki, M. 1993. Application of adsorption cooling systems to automobiles. Heat Recovery Systems & CHP 13(4):335-40.
Wang, R.Z., W. Wang, and T.F. Qu. 2001. Research and development on waste heat driven adsorption bus air-conditioning system. Final Report SJTU-UTRC Joint Research Program, Shanghai Jiao Tong University (SJTU) and United Technologies Research Center (UTRC), China.
Zhang, L.Z. 2000. Design and testing of an automobile waste heat adsorption cooling system. Applied Thermal Engineering 20(1):103-14.
Received June 30, 2009; accepted October 4, 2009
Leo Sing Lim is a research specialist at the Energy Research Group Laboratory and Mohammad Omar Abdullah is a Senior Lecturer at the Department of Chemical Engineering and Energy Sustainability at the Universiti Malaysia Sarawak (UNIMAS), Sarawak, Malaysia.
Leo Sing Lim, PhD
Mohammad Omar Abdullah, PhD, CEng