Review of recent advances toward transcritical C[O.sub.2] cycle technology.
In the early twentieth century, carbon dioxide (C[O.sub.2]) was extensively used as the refrigerant for vapor-compression refrigeration systems, especially in marine systems, due to its nonflammability and acceptable toxicity. However, C[O.sub.2] has unfavorable characteristics of lower critical temperature and higher operating pressure compared to ammonia ([NH.sub.3]) and sulfur dioxide ([SO.sub.2]). After the advent of chlorofluorocarbons (CFCs) in 1928, the use and interest in carbon dioxide as a refrigerant significantly diminished until the revival of interest in natural working fluids in the early 1990s due to the environmental concerns of global warming. Table 1 shows a comparison of the direct global warming potentials among some fluorocarbon-based refrigerants and C[O.sub.2].
A paper by Lorentzen and Pettersen (1992) appears to be the first open literature about revisiting the use of C[O.sub.2] as a viable alternative refrigerant to deal with global warming and ozone depletion. Since then, the transcritical cycle technology using C[O.sub.2] as the refrigerant has been the subject of increased research activities and is considered a possible replacement for vapor-compression cycle technologies that use fluorocarbon-based refrigerants in certain applications.
One of the prominent applications is automotive air conditioning. By now, most of the automobile manufacturers have prototype systems of C[O.sub.2] air conditioners. Several new and innovative designs for heat exchangers, compressors, and valves have emerged from studies in this area. This application has recently received increased attention based on the European Parliament's July 2006 vote to phase out the refrigerant HFC-134a, which is currently used in automotive air conditioning. This vote marks the first international action to ban a hydrofluorocarbon (HFC) refrigerant due to its high global warming potential. HFC refrigerants were introduced in response to the Montreal Protocol to phase out chlorine-containing refrigerants to protect the Earth's ozone layer. Another application for C[O.sub.2] is environmental control units (ECUs), which are packaged air-to-air air conditioners that are used in the cooling of mission critical electronics and personnel. The US Army currently maintains roughly 22,000 ECUs of varying capacity, either in service, in storage, or on order, that use HCFC-22 as their refrigerant. These need to be replaced by 2010, as this is the current phase-out date for HCFC-22 in new equipment due to the refrigerant's ozone depletion potential. A third application that shows great promise for transcritical C[O.sub.2] systems is heat pump water heaters. In fact, the first commercial product has already been introduced on the Japanese market. Finally, transcritical C[O.sub.2] technology has been proposed as the cooling system in glass door coolers and vending machines based on the interest by a leading soft drink manufacturer to use "green" cooling technologies.
This paper presents a description of the C[O.sub.2] systems that are proposed for each of these applications and the performance results obtained with these systems. Both theoretical modeling results and experimental data will be presented. Whenever feasible, a comparison to the existing fluorocarbon-based technology will be given.
BASIC TRANSCRITICAL C[O.sub.2] CYCLE
The critical point of C[O.sub.2] is 30.85[degrees]C and 73.53 bars. Thus, many refrigeration and air-conditioning applications span the critical point in that the heat absorption temperature is below and the heat rejection temperature is above the critical temperature. This implies a transcritical cycle in which the evaporator operates as a familiar vapor-liquid two-phase device but the condenser is replaced by a supercritical heat rejection device called a gas cooler. Figure 1 shows the cycle state points indicated in a p-h diagram. The transcritical C[O.sub.2] cycles were drawn based on the following assumptions: gas cooler outlet temperature of 30[degrees]C, evaporation temperature of 0[degrees]C, superheat of 0 K, and isentropic compression and isenthalpic expansion processes.
[FIGURE 1 OMITTED]
The cycle shown in Figure 1 illustrates the basic transcritical cycle. Assuming a constant gas cooler exit temperature, the cycle can be operated at different gas cooler pressures by using the expansion valve to adjust the back (high-side) pressure. Varying the gas cooler pressure has two impacts on the cycle performance. As the gas cooler pressure increases, the enthalpy difference across the evaporator (cooling capacity) increases. At the same time, the enthalpy difference across the compressor (compressor work input) increases. Since these impacts have opposing effects on cycle performance, the efficiency of the transcritical C[O.sub.2] cycle, typically indicated by its coefficient of performance (COP), can be maximized by adjusting the gas cooler pressure. Figure 2 presents the COP as a function of gas cooler pressure for several evaporation temperatures and gas cooler exit temperatures. Figure 2 was generated by conducting a thermodynamic cycle analysis in EES (Klein 2004) using the assumptions on isentropic compression and isenthalpic expansion processes and a superheat of 0 K.
Figure 2 indicates that both the gas cooler exit temperature and the evaporation temperature have significant impacts on the maximum COP. It can also be seen from Figure 2 that the discharge pressure should be in the order of 90-100 bars or higher depending on the application. This means that when the evaporation temperature is, for instance, 0[degrees]C, the discharge temperature in single-stage compression with dry saturated suction vapor will be about 70[degrees]C-80[degrees]C. This temperature can be adjusted by varying the discharge pressure and the suction vapor state using a suction-to-liquid line heat exchanger or, possibly, some liquid injection. A suction-to-liquid line heat exchanger, as used quite commonly in refrigeration and air-conditioning equipment, may also result in a 5%-10% increase in cycle efficiency (Lorentzen and Pettersen 1993; Robinson and Groll 1998a).
[FIGURE 2 OMITTED]
TRANSCRITICAL C[O.sub.2] CYCLE MODIFICATIONS
The basic transcritical C[O.sub.2] cycle is thermodynamically less efficient compared to the conventional vapor-compression cycle if one assumes that the evaporation temperatures are the same and that the gas cooler exit temperature is equal to the condensing temperature. However, several studies in the literature indicate that these assumptions are incorrect. Due to the superior heat transfer characteristics of C[O.sub.2] compared to halocarbon refrigerants, the gas cooler exit temperature of a transcritical C[O.sub.2] system is significantly closer to the heat sink temperature than the condensing temperature of conventional vapor-compression systems for similar heat exchanger designs (Robinson and Groll 2000). While the differences in the evaporator approach temperature are not as large as the ones on the heat rejection side, studies also indicate that the evaporation temperature of a C[O.sub.2] system is closer to the heat source temperature than the evaporation temperature of halocarbon systems for similar heat exchanger designs (Robinson and Groll 2000). Even considering better heat exchanger approach temperatures, the basic transcritical C[O.sub.2] cycle may still be less efficient than the conventional vapor-compression cycle; thus, several advanced cycle designs have been considered that show promise well beyond the basic cycle.
The compression power consumption may be reduced by utilizing technologies such as multiple-effect compression (e.g., a single compressor with economizer) (Voorhees 1905) or two-stage compression (Hall 1889) that were already available for C[O.sub.2] vapor-compression cycles at the beginning of the twentieth century. In addition, these technologies can be applied to increase the overall temperature lift between the lowest heat source temperature and the highest heat sink temperature, resulting in a major expansion of possible applications. A two-stage cycle with intercooling is shown in Figure 3. The intercooler and gas cooler can be implemented in one heat exchanger. The transcritical two-stage cycle with intercooling provides operating characteristics that are of great importance to the cycle efficiency. Figure 4 presents two different two-stage transcritical C[O.sub.2] cycles with intercooling in a pressure-enthalpy diagram. Both cycles have an evaporation temperature of 0[degrees]C, a gas cooler exit temperature of 35[degrees]C, a gas cooler pressure of 10 MPa, and isentropic compression and isenthalpic expansion processes. The cycles differ in their choice of intermediate pressure, i.e., 7 versus 8 MPa. It can be seen from Figure 4 that the choice of intermediate pressure has a significant impact on the work input of the two compressors.
[FIGURE 3 OMITTED]
It can also be seen from Figure 4 that intercooling is only possible when the intermediate pressure is high enough so that the first-stage compressor discharge temperature is higher than the outdoor, i.e., gas cooler, exit temperature. For lower intermediate pressures, the two-stage compression process equals a single-stage compression process and no benefit in cycle efficiency can be achieved.
Figure 5 presents the COP of the two-stage transcritical C[O.sub.2] cycle with intercooling as a function of intermediate pressure for the same operating conditions as used in Figure 4. The COP reaches a maximum at a relatively high intermediate pressure. This is a substantial difference from subcritical two-stage cycles with intercooling, where the intermediate pressure is typically chosen as the square root of the high-side times the low-side pressure. To explain the COP behavior as a function of intermediate pressure, Figure 4 needs to be considered. It can be seen from Figure 4 that as the intermediate pressure increases above the critical pressure, the second-stage compression process moves quickly into the supercritical region, where the slopes of the isentropic lines are steeper than in the conventional superheated region. In fact, the inlet state of the second-stage compression process moves to lower enthalpies than the critical enthalpy due to a slope of the isotherms of nearly zero in the supercritical region. As a result, the enthalpy increase through the second-stage compressor decreases sharply as the intermediate pressure rises across the critical pressure. Just before the slope of the isotherms changes again toward lower enthalpies, the total compression work reaches a minimum and, thus, the COP reaches a maximum. At maximum COP, the enthalpy difference across the intercooler (at intermediate pressure) is significantly larger than the enthalpy difference across the gas cooler (at high-side pressure) and, thus, the intercooler is the main heat rejection heat exchanger of the overall cycle. Depending on operating conditions, the two-stage transcritical C[O.sub.2] cycle with intercooling provides a 12% to 24% increase in efficiency compared to the efficiency of the basic transcritical cycle (Baek et al. 2005).
[FIGURE 4 OMITTED]
[FIGURE 5 OMITTED]
Work Recovery Expansion Machine
Further improvement of the basic transcritical cycle can be made by decreasing the relatively high throttling losses of the transcritical C[O.sub.2] cycle. This can be amended by a number of well-known methods (Lorentzen 1983). One favorable solution is to recover the expansion work directly by using an expansion machine, since the properties of C[O.sub.2] make this feasible. With a conventional refrigerant such as R-134a, most of the theoretical expansion work comes from the flash gas, and the p-V diagram for the expansion process becomes very narrow with a low mean pressure. For C[O.sub.2] the situation is quite different, with most of the work in the liquid phase, a high mean pressure, and a small volume requirement (Lorentzen 1994). An expansion work recovery machine that is included in the basic transcritical C[O.sub.2] cycle may result in a 14%-17% increase in cycle efficiency (Robinson and Groll 1998a).
Two prototype scroll expanders were built and tested (Huff et al. 2003). Scroll-type expanders were chosen over piston-type expanders because scroll expanders do not need active valve control. The two prototype scroll expanders were operated at high and low pressures of 8 and 4.5 MPa, respectively, and the speed of the scroll expanders was controlled by the generator load. It was reported that there was optimal expander speed for each scroll expander where the maximum mechanical power output and the maximum isentropic efficiency were achieved. For the second prototype, the mechanical power output increased from approximately 350 to 750 W when the expander speed changed from 1400 to 1800 rpm. The isentropic efficiency increased from 20% at an expander speed of 1400 rpm to the maximum of 42% at 1800 rpm. The authors concluded that, for better performance of the expander, the scrolls needed to be optimized with careful considerations of leakage passages and contact forces between the two scrolls.
A multi-vane type expander with a chamber volume of 64 [mm.sup.3] was prototyped and tested by Fukuta et al. (2003). A maximum isentropic efficiency of 43% was obtained at an inlet pressure of 9.1 MPa, an outlet pressure of 4.1 MPa, an inlet temperature of 40[degrees]C, and a rotor speed of 2000 rpm. The authors implied that better internal flow leakage control would result in a performance increase of the expander.
An even better option to increase cycle efficiency is to integrate the expansion work recovery machine in a transcritical two-stage C[O.sub.2] cycle with intercooling. Since the optimum COP of the transcritical two-stage cycle with intercooling is reached at a relatively high intermediate pressure, the work input of the second-stage compressor is much smaller compared to the work input of the first-stage compressor. As it turns out, the work input required to drive the second-stage compressor is approximately equal to the work output that is generated by an expansion machine. Baek et al. (2005) studied how the system COP changes when the work-producing device directly drives the second-stage compressor. The analysis was carried out at an evaporation temperature of 0[degrees]C, a superheat of 0 K, gas cooler outlet temperatures of 35[degrees]C and 40[degrees]C, a high-side pressure of 10 MPa, and isentropic expansion and compression processes as well as isentropic expansion and compression process efficiencies of 50% and 70%. The authors found that, depending on intermediate pressure, an increase of up to 42% in cycle efficiency compared to the efficiency of the basic transcritical cycle can be achieved.
Nickl et al. (2005) presented field test results of an integrated three-stage expander-compressor that directly drives the second-stage compressor in a C[O.sub.2] refrigeration plant. The second-stage compressor had inlet and outlet pressures of approximately 8.2 and 10 MPa, respectively. The authors reported that the system COP of the three-stage expander-compressor cycle improved by at least 40% compared to the one of a cycle with isenthalpic expansion process. It is interesting to note that the increase in COP of up to 42% as theoretically predicted by Baek et al. (2005) agrees well with the experimentally measured results of Nickl et al. (2005) at approximately the same operating pressure of the second-stage compression process.
A C[O.sub.2] swing piston expander was built and tested (Haiqing et al. 2006). Due to the complexity and cost of using electromagnetic valves, the inlet valve was controlled by a camshaft mechanism. The prototype expander was coupled with a motor and a generator through one shaft. The motor started the expander. Once the expander was running, the motor was powered off and the electric power from the generator was measured by a power meter. The shaft speed of the expander was controlled by varying the load of the generator. The expander was tested at an inlet pressure of 10 MPa, an inlet temperature of 35[degrees]C, and an outlet pressure of 4 MPa. It was shown that the isentropic efficiency increased, topped off, and gradually decreased as the generator load changed from 300 to 1000 W. The maximum isentropic efficiency of approximately 31% was observed at a generator load of 600 W. The shaft speed of the expander decreased from approximately 1600 to 800 rpm as the generator load increased from 300 to 1000 W. It was also reported that the expander was running at steady-state during these tests.
Vortex Tube Expansion Device
Another cycle modification to reduce the expansion process losses that has been investigated in the literature considers the replacement of the expansion valve by a vortex tube (Li et al. 2000).
The energy separation effect of a vortex tube is often called the Ranque-Hilsch effect in honor of Georges J. Ranque, who first discovered the phenomenon in 1931, and Rudolf Hilsch, who carried out the first fundamental investigation of the effect and improved Ranque's vortex tube design. To enable the energy separation effect, a compressed high-pressure gas enters the vortex tube tangentially through the inlet nozzle and develops an approximately axis-symmetric vortex flow. A cold gas stream at low pressure leaves the tube through a central orifice right next to the entrance nozzle, while at the far end of the tube a hot gas stream at low pressure is exhausted near the wall. The detailed mechanism behind the Ranque-Hilsch effect of a vortex tube is still unclear, although it is generally attributed to the exchange of mechanical work caused by the velocity difference between the gas near the wall and the gas in the center. The Ranque-Hilsch effect has been proven by experiments with single-phase gases, such as air, as the working fluids.
Theoretically, it is possible to assume that a vortex tube should be able to operate as the expansion valve of the transcritical C[O.sub.2] cycle as schematically indicated in Figure 6. In this case, the high-pressure C[O.sub.2] exiting the gas cooler enters the vortex tube. Due to the Ranque-Hilsch effect, saturated liquid C[O.sub.2] at evaporation pressure (state 4) leaves the cold end of the vortex tube, while superheated vapor C[O.sub.2] at evaporation pressure (state 6) exits the hot end of the vortex tube. The superheated vapor enters an auxiliary heat exchanger and dissipates heat to the environment. The cooled vapor stream is then mixed with the saturated liquid again before entering the evaporator to extract heat from the low-temperature heat source. It must be noted, though, that thus far no experimentally functional vortex tube in which a liquid-vapor mixture is separated has been demonstrated. If such a device could be developed, then a vortex tube that results in 100% separation of superheated vapor from saturated liquid could provide up to a 37% increase in cycle efficiency, and a vortex tube that results in 50% separation could provide approximately a 20% increase in cycle efficiency compared to the efficiency of the basic transcritical cycle.
Ejector Expansion Device
An experimentally validated approach to recover some of the expansion process losses of the transcritical C[O.sub.2] cycle is based on replacing the expansion valve by an ejector (Akagi et al. 2005; Li and Groll 2005). The ejector-expansion refrigeration cycle was first proposed by Kornhauser (1990). An updated schematic of the cycle is shown in Figure 7. The state points of the cycle presented in Figure 7 can be shown in a C[O.sub.2] pressure-enthalpy diagram, as indicated in Figure 8.
The ejector working processes can be explained as follows. The motive stream leaves the gas cooler at gas cooler pressure and enters the ejector. The motive stream expands in the motive nozzle from the high pressure to the receiving chamber pressure. The enthalpy of the motive stream reduces and the velocity increases. At the same time, the suction stream leaves the evaporator at evaporator pressure and enters the ejector in the receiving chamber, where its velocity increases. The suction stream mixes with the motive stream in the mixing section and they become one stream at an intermediate mixing pressure and velocity. This stream further increases its pressure in the diffuser by converting its kinetic energy into internal energy. The stream leaves the ejector at a pressure slightly higher than the evaporator pressure and enters a separator to separate saturated liquid from the saturated vapor. The saturated liquid expands to the evaporator pressure and enters the evaporator, while the saturated vapor enters the compressor. The cycle efficiency increases because the compressor operates at a reduced pressure ratio compared to the cycle without an ejector.
[FIGURE 6 OMITTED]
In order to provide adequate control of the cycle at various operating conditions, a motive stream control mechanism as invented by Gay (1931) and revisited by Takeuchi et al. (2002) for C[O.sub.2] air-conditioning applications or another flow control mechanism, such as a vapor bypass to regulate the quality at the evaporator inlet, is necessary.
[FIGURE 7 OMITTED]
[FIGURE 8 OMITTED]
Li and Groll (2005) performed theoretical analyses for the transcritical C[O.sub.2] refrigeration cycle with an ejector-expansion device. The analyses were carried out with the following assumptions: a gas cooler pressure of 10 MPa, a gas cooler refrigerant outlet temperature of 40[degrees]C, an evaporation temperature of 5[degrees]C, a superheat of 5 K, an entrainment ratio of 0.55 for the ejector, and a pressure difference of 0.03 MPa between the evaporator and the receiving chamber. In addition, the isentropic efficiencies of the motive stream and suction stream nozzles were assumed to be 90% each, and the diffuser pressure recovery coefficient was assumed at 0.75. The theoretical analyses showed that the transcritical ejector-expansion cycle could provide an approximately 16% increase in cycle efficiency compared to the efficiency of the basic transcritical cycle. A similar performance improvement had been reported by Ozaki et al. (2004). In contrast to the vortex tube, ejector-expansion devices have been successfully implemented in experimental systems (Elbel and Hrnjak 2004) and even commercial products (Akagi et al. 2005).
Incorporating Thermoelectric Devices
In 1834, Jean Peltier discovered the inverse of the Seebeck effect, now known as the Peltier effect. He found that a voltage applied to a thermocouple causes a temperature difference between the junctions. This results in a small heat pump, later referred to as a thermoelectric cooler (TEC). A Peltier cooler/heater or thermoelectric heat pump is a solid-state active heat pump that transfers heat from one side of the device to the other. The direction of heat transfer is controlled by the polarity of the current; reversing the polarity will change the direction of transfer and, thus, the sign of the heat absorbed/rejected. Using this technology, it is possible to subcool the refrigerant that is leaving a gas cooler or condenser below the ambient air temperature.
Winkler et al. (2006) proposed the novel concept of employing thermoelectric devices at strategic locations within a condenser or evaporator in order to compensate for air or refrigerant flow maldistributions. The authors showed that a thermoelectric device, if used as a dedicated subcooler at the gas cooler outlet of a transcritical C[O.sub.2] system, increases the cooling capacity by 20% and the system COP by 16%, even taking the external power consumption into account. A further increase in system efficiency could be achieved if additional thermoelectric devices are used during the desuperheating of the hot C[O.sub.2] gas leaving the compressor, when the temperature difference between the C[O.sub.2] and the environment is large, to generate the electricity that is needed to drive the thermoelectric devices used in the subcooler. The authors also studied the effect of using a thermoelectric cooling device at the condenser outlet of conventional refrigeration cycles for R-134a, R-410A, and R-404A. The cycles for these refrigerants were simulated at the evaporating temperature of 4.44[degrees]C and the condensing temperature of 46.1[degrees]C for air-conditioning applications. Then, the COP improvements for each cycle were plotted as a function of the temperature lift in the thermoelectric cooling device and compared to the COP improvement for a C[O.sub.2] cycle with thermoelectric cooling at the outlet of a gas cooler. It should be noted that, for the R-134a, R-404A, and R-410A cycles, the power consumption of the thermoelectric device was not included in the system COP. The COP improvement comparison showed that even though the extra power consumption by the thermodynamic device was considered for the C[O.sub.2] cycle, the C[O.sub.2] cycle showed much higher improvement in COP compared to the three conventional cycles. For example, the COP improvement for the C[O.sub.2] cycle was at least three times higher than those of the three conventional refrigeration cycles at the temperature lift of 4 K.
Summary of Cycle Modifications
Figure 9 presents a comparison of the relative performance of the various transcritical C[O.sub.2] cycles presented in the previous chapters. The performance of the basic transcritical cycle is set at 100%. It can be seen from Figure 9 that the highest improvement of the basic single-stage cycle can be achieved by replacing the expansion device with a work recovery expansion machine. However, a work recovery expansion machine is also considered the most costly improvement compared to an internal heat exchanger or ejector, which has no moving parts. It can also be seen from Figure 9 that the highest overall cycle performance can be achieved by using a two-stage cycle with intercooling and a work recovery expansion machine to drive the second-stage compressor.
[FIGURE 9 OMITTED]
C[O.sub.2] COMPONENT DESIGN CONSIDERATIONS
The compression process of the transcritical cycle using C[O.sub.2] involves a compression ratio, which is greatly reduced compared to that of other refrigerants, resulting in the potential of much better compressor efficiencies. On the other hand, the absolute pressures are much higher than are encountered in traditional refrigeration technology. R-134a and R-22 exhibit high-end pressures of up to 20 bars, based on the application. Depending on the cycle configuration, the high-end pressure in a transcritical cycle using C[O.sub.2] might be as high as 140 bars. This high pressure is tolerable since it can be confined inside small-diameter tubes, which form the spine of the heat exchanger. Due to the fact that the volumetric refrigeration capacity of C[O.sub.2] is five to eight times higher than that of R-134a or R-22, the volumetric flow rate is much smaller for the same cooling capacity. Thus, the pipe diameters of the refrigeration piping and, therefore, the weight and occupied volume of the heat exchangers can be decreased by 30% compared to conventional halocarbon refrigerant heat exchangers (Robinson and Groll 1998b) while achieving the same heat transfer rates. Because of the small area over which the pressure acts in a small-diameter tube, this geometry is ideal for high-pressure systems. In particular, microchannel (or minichannel) heat exchangers have been investigated as the design of choice for C[O.sub.2] heat exchangers (Yin et al. 2001). Microchannel heat exchangers use several multi-port extruded (MPE) aluminum tube bands that are stacked with accordion-style fins between two tubes, as shown in Figure 10. Each MPE tube has many parallel microchannels. Due to the nature of the manufacturing process, the high pressures are not expected to be a problem in the heat exchanger components. The compressor for the transcritical C[O.sub.2] cycle requires a thicker shell to contain the high pressure compared to conventional refrigeration compressors. However, since the volumetric capacity of the fluid is large, the compression chamber size is smaller than a current refrigeration compressor for the same cooling capacity. Moreover, if a two-stage C[O.sub.2] compressor design with internal intermediate pressure is used, the shell thickness can be reduced. Thus, it is possible to have a C[O.sub.2] compressor that is as light as a conventional refrigeration compressor (Yamasaki et al. 2004).
[FIGURE 10 OMITTED]
Due to the high pressure differential, C[O.sub.2] compressors should be carefully designed to minimize leakage losses in the compression chamber and friction losses in bearings. To reduce flow leakage, Suss and Veje (2004) used piston rings for their prototype reciprocating compressor and Maeyama et al. (2006) reduced a radial clearance by enlarging the rolling piston for their rotary compressor. It is also important to minimize reexpansion loss due to the high gas density of C[O.sub.2]. Since reexpansion loss is directly related to the discharge port area, the port diameter and depth need to be optimized to reduce the loss. Too small a port diameter could result in high overpressure loss, but too large a port diameter could result in high back-flow loss. Higuchi et al. (2006) reduced the discharge port height of a swing compressor to reduce the reexpansion loss.
Proper oil selection and oil return to the compressor are also very important to ensure the reliability of C[O.sub.2] compressors. Polyolester (POE) and polyalkylene glycol (PAG) oils seem to be the choice of lubricants. In comparison to PAG oil, POE oil has relatively better miscibility and solubility but lower dynamic viscosity (Hauk and Weidner 2000). Due to the low solubility of PAG oil, it can be trapped in the refrigerant circuit. Thus, when PAG oil is used, proper oil return to the compressor should be considered.
TRANSCRITICAL C[O.sub.2] CYCLE APPLICATIONS
All told, the transcritical cycle using C[O.sub.2] may be a feasible alternative to conventional vapor-compression technology for certain applications that take advantage of the distinct C[O.sub.2] characteristics. However, issues regarding energy efficiency and high pressure have to be addressed. Five of the more promising applications for transcritical C[O.sub.2] technology will be evaluated in more detail in the following sections. These applications are automotive air conditioners, residential air-conditioning and heat pump systems, military standard environmental control units, heat pump water heaters, and cooling systems for glass door coolers and vending machines.
Automotive Air Conditioners
Lorentzen and Pettersen (1993) discussed an experimental study in which they compared the performance of a standard commercial R-12 automotive air-conditioning system and a laboratory prototype C[O.sub.2] system with a similar design cooling capacity. The R-12 system had a sliding vane rotary compressor with a displacement of 150 [cm.sup.3], while the C[O.sub.2] system had a wobble-plate axial piston compressor with a displacement of 26 [cm.sup.3]. The experimental results showed that the prototype C[O.sub.2] system delivered equal or slightly higher system efficiencies than the R-12 system. It was demonstrated that a suction line heat exchanger in the prototype C[O.sub.2] system resulted in improvement in the system efficiency compared to the prototype C[O.sub.2] system without a suction line heat exchanger, especially at high ambient temperature.
In another paper, Pettersen (1997) noted that in comparison to a highly efficient R-134a mobile air-conditioning system, a laboratory prototype C[O.sub.2] system performed slightly better at a low ambient temperature below approximately 25[degrees]C, but above this ambient temperature the C[O.sub.2] system performed lower than the R-134a system.
McEnaney et al. (1998) reported a similar trend in system efficiency in relation to the ambient temperature when their prototype C[O.sub.2] mobile air-conditioning system was compared to the R-134a air-conditioning system of a small passenger vehicle. They observed that the C[O.sub.2] system performance was higher than that of the R-134a system by 40% or more at the ambient temperature below 40[degrees]C.
Preissner et al. (2000) compared the performance of automotive air-conditioning systems using C[O.sub.2] and R-134a with heat exchangers of approximately equal size. Unlike the previously discussed studies, suction line heat exchangers were used in both systems. The evaporating conditions were controlled with variable size expansion devices. Measurements were carried out over a typical range of outdoor temperatures (25[degrees]C to 45[degrees]C) for idling and driving conditions of 1000 and 1800 rpm. Depending on the compressor speed and the ambient temperatures, the capacity of the C[O.sub.2] system ranged from 13% lower to 20% higher when compared to the R-134a system, whereas the COP was 11% to 23% lower. It was noted that the inclusion of a suction line heat exchanger in the R-134a system resulted in a better system efficiency in comparison to the same R-134a system without a suction line heat exchanger. The R-134a system was equipped with a compressor and heat exchangers that are currently applied in vehicles and represent the latest technology, whereas the C[O.sub.2] components were still considered prototypes under development. It was suggested that further improvements are needed with respect to the C[O.sub.2] components that were available to the authors so that C[O.sub.2] systems are competitive with R-134a systems.
Hirao et al. (2000) conducted a study of an automobile air-conditioning system that uses the transcritical C[O.sub.2] cycle and incorporated a high-pressure control system. The high-pressure control system made it possible to improve the efficiency of the compressor and, thus, to maximize the system COP. The C[O.sub.2] system was made up of an experimentally manufactured gas cooler that consisted of multi-flow air paths, a microchannel evaporator, and a scroll type compressor. During the experiments, the gas cooler outlet temperature was kept at a constant value of 37[degrees]C and the compressor speed was varied. The results showed that the COP of the C[O.sub.2] system was almost equal to the COP of an existing HFC-134a system at the rating point of 2400 rpm. It was verified that the high pressure could be controlled to the appropriate value where the COP reaches a maximum corresponding to the given outside temperature. It was also confirmed that a COP that is equal to or higher than the COP of an HFC-134a can be realized for some operating conditions.
Giannavola et al. (2000) presented the results of experiments conducted on a prototype R-744 (C[O.sub.2]) system operating in a heat pump mode with the heat rejection path lying entirely within the supercritical region. The prototype system was sized for a compact car, but the experiments were conducted to provide baseline data for scale-up to sport utility vehicle size. The data presented in the paper were for a limited range of operation and revealed the potential for significant heating capacity at relatively low ambient temperatures. The following operational characteristics were observed:
* Low ambient temperatures degraded the steady-state capacity by 22% to 26% depending on the indoor temperature.
* The maximum capacity and efficiency occurred at the most extreme operating conditions. This coincides with the time when they are needed the most for warm-up conditions.
* The engine heat can be used exclusively to reduce emissions during start-up instead of being diverted immediately for cabin comfort.
These conclusions were qualitative because they were observed from data taken with a system that used heat exchangers not designed for reversible operation. The authors suggested that further experiments with optimized heat exchangers and control strategies are needed to establish the quantitative basis for more definitive conclusions.
Brown et al. (2002a) carried out a comparative simulation study of R-134a and C[O.sub.2] systems by using the simulation models CYCLE-11.UA and CYCLE-11.UA-C[O.sub.2]. The heat exchanger configurations of the R-134a system were derived from those of the C[O.sub.2] system studied by McEnaney et al. (1999). The air-side configurations were the same for both systems. However, the refrigerant circuits of the R-134a heat exchangers were redesigned to have a refrigerant-side pressure drop comparable to that of a conventional R-134a system. The C[O.sub.2] compressor had the same swept volume as the C[O.sub.2] compressor studied by McEnaney et al. (1999). The R-134a compressor was sized to deliver the same cooling capacity as the C[O.sub.2] system at the compressor speed of 1000 rpm and an ambient temperature of 43.3[degrees]C. Simulation was carried out at condenser/gas cooler air inlet temperatures of 32.2[degrees]C, 43.3[degrees]C, and 48.9[degrees]C and an evaporator air inlet temperature of 26.7[degrees]C. Two compressor speeds, 1000 and 3000 rpm, were selected for low--and high-speed operation. At the compressor speed of 1000 rpm, both systems delivered comparable cooling capacity systems, but at the compressor speed of 3000 rpm, the C[O.sub.2] system had a lower cooling capacity by 3.6% and 7.2% at the gas cooler/condenser air inlet temperatures of 43.3[degrees]C and 48.9[degrees]C, respectively. At 1000 rpm, the C[O.sub.2] compressor consumed approximately 20%, 29%, and 36% more power as the gas cooler/condenser air inlet temperature increased from 32.2[degrees]C to 43.3[degrees]C and 48.9[degrees]C, respectively. At 3000 rpm, a similar trend was observed but the difference in compressor power was larger. Thus, the COP of the R-134a system was approximately 21% to 60% better than the COP of the C[O.sub.2] system for all operating conditions. The performance improvement increased with the compressor speed and the gas cooler/condenser air inlet temperature. At a given compressor speed, the system efficiency always decreased with the gas cooler/condenser air inlet temperature. It was noticeable to see that this trend occurred even though the C[O.sub.2] compressor was given a higher isentropic efficiency at all simulation conditions to credit its lower compression ratio. Thus, the authors concluded that even though C[O.sub.2] has favorable transport properties, the C[O.sub.2] system equipped with a suction line heat exchanger to reduce expansion losses and a compressor that had a higher isentropic efficiency than an R-134a compressor did not overcome the thermodynamic disadvantages of the transcritical cycle. The authors observed that the higher entropy production at the gas cooler compared to the R-134a condenser was the main cause for the poor performance of the C[O.sub.2] system.
Direct global warming potential (GWP) of refrigerants alone cannot capture the true impact of the refrigeration systems on global warming because it disregards the indirect global warming impact due to the energy consumption of the systems. Two other indices are available to capture the indirect global warming impacts due to the energy use of the system throughout its lifetime: the total equivalent warming impact (TEWI) and the life-cycle climate performance (LCCP). Sumantran et al. (1999) published TEWI comparisons of R-134a and C[O.sub.2] systems for small-and mid-sized cars. It was shown that the TEWI of the C[O.sub.2] systems was lower in a mild climate (Frankfurt, Germany) but higher in warm or hot climates (Tokyo, Japan, and Miami, FL, USA). Thus, the use of C[O.sub.2] automotive air-conditioning systems on a global basis may not achieve the common goal of the reduction of greenhouse gas emissions.
Recently, three leading chemical companies announced three new refrigerants that meet the European Union's restrictions on the use of refrigerants with a GWP of higher than 150 for automotive air-conditioning applications starting from 2011. It is said that all new refrigerants are to be nonflammable and low in toxicity with zero ozone depletion potential and low global warming potential (GWP of 10 and 40 for Fluid H and DP-1, respectively). Initial tests show that Fluid H and DP-1 have
a similar thermodynamic cycle performance to R-134a and are compatible with R-134a automotive air-conditioning technology (Spatz 2006; Minor 2006).
Residential Air-Conditioning and Heat Pump Systems
Bullock (1997) investigated the feasibility of C[O.sub.2] as a refrigerant in residential unitary systems by comparing the system efficiencies of a mid-efficient R-22 unit and a C[O.sub.2] unit. Both units had equal cooling capacities and the same compressor isentropic efficiencies. He noted that the C[O.sub.2] unit could deliver comparable system efficiency to the R-22 unit only if significant improvements were made in all components of the C[O.sub.2] unit, especially in hermetic C[O.sub.2] compressor efficiency and heat exchanger designs.
Aarlien and Frivik (1998) investigated the performance of a prototype C[O.sub.2] system for residential heat pump applications and compared it to the performance of a conventional R-22 system in cooling and heating modes. Microchannel heat exchangers were used for the indoor and outdoor units in each system. All heat exchangers had the same core dimensions. A hermetic type compressor was used for the baseline R-22 system, whereas an open type compressor was used for the prototype C[O.sub.2] unit. Due to the different types of compressors, the total compressor power consumption for the C[O.sub.2] system was determined by measuring the shaft power and factoring in the assumed motor efficiency of 0.9. For the R-22 compressor, the total power consumption was directly measured, which included the motor power. The authors reported that in cooling mode, the COPs of the C[O.sub.2] system were 0.5%, 3.1%, and 14.5% lower than those of the R-22 system at three different cooling test conditions. In heating mode, however, the COP values of the C[O.sub.2] system were 3.3%, 14.1%, and 8.0% higher than those of the R-22 system at three different heating test conditions. Although the motor efficiency of the C[O.sub.2] compressor was taken into account in calculating the COP, an extra cooling capacity required to cool down the motor was not considered.
Brown et al. (2002b) evaluated the feasibility of using C[O.sub.2] as an R-22 substitute for residential air-conditioning applications. The R-22 and C[O.sub.2] vapor-compression systems were simulated using the modeling tools CYCLE-11.UA and CYCLE-11.UA-C[O.sub.2], respectively. Each system was configured to have state-of-the-art microchannel heat exchangers as evaporator and condenser/gas cooler. The R-22 heat exchangers were designed based on the C[O.sub.2] microchannel heat exchangers, but the refrigerant circuits were modified to have similar pressure drops. The C[O.sub.2] compressor was modeled to be more efficient than the R-22 compressor. A suction line heat exchanger was used for only the C[O.sub.2] system. The simulation result showed that the COP of the C[O.sub.2] system was approximately 40% or more lower than that of the R-22 system and that the compressor power consumption of the C[O.sub.2] system was approximately 38% or more higher than that of the R-22 system. In cooling capacity, the C[O.sub.2] and R-22 systems delivered comparable cooling capacity at the gas cooler/condenser air inlet temperature below 35[degrees]C. Above that temperature, the C[O.sub.2] cooling capacity was lower than the R-22 cooling capacity, and the difference increased with the gas cooler/condenser air inlet temperature. The authors concluded that better transport properties of C[O.sub.2] and a better isentropic efficiency of the C[O.sub.2] compressor did not overcome the inherent thermodynamic disadvantages of the transcritical cycle in comparison to the R-22 vapor-compression cycle. Based on the fact that the efficiencies of R-410A systems are comparable to those of R-22 systems for the same applications, the authors implied that the C[O.sub.2] system would perform lower than R-410A systems as well.
Cho et al. (2005) compared the cooling performance of a C[O.sub.2] system to conventional R-22, R-410A, and R-407C systems. Their prototype C[O.sub.2] system consisted of an inverter-driven scroll compressor, a finned-tube gas cooler, an electronic expansion valve, a finned-tube evaporator, and an accumulator, but it had no suction line heat exchanger. The performance of the C[O.sub.2] system was measured at the standard rating conditions, while the performances of the conventional R-22, R-410A, and R-407C systems were obtained from the open literature (Ha 1999; Choi and Kim 2004; Park et al. 2004). At the system charge of 10% less than optimal charge, the cooling capacity of the C[O.sub.2] system was reduced by 11%, whereas the cooling capacities of the R-22, R-410A, and R-407C systems were reduced by 0.7%, 1.1%, and 3%, respectively. At that same undercharge, the COP of the C[O.sub.2] system decreased by 10%, whereas the cooling capacities of the R-22, R-410A, and R-407C systems were reduced by 2%, approximately 3%, and 5.5%, respectively. Based on these findings, it was concluded that the C[O.sub.2] system requires extremely careful refrigerant charge control to deliver the best performance.
A report by Little (2002) presented a comparison of the LCCP of four 10.5 kW (3 ton) residential air-conditioning systems with the refrigerants R-22, R-407C, R-410A, and C[O.sub.2]. The C[O.sub.2] system had the lowest LCCP, 27500 kg C[O.sub.2], followed by the R-410A system with 29300 kg C[O.sub.2], the R-407C system with 29400 kg C[O.sub.2], and the R-22 system with 29850 kg C[O.sub.2]. It should be noted that the analysis assumed equal system efficiency (12 SEER) for all systems. As a result, the indirect warming associated with the system power consumption accounts for 27466 kg C[O.sub.2] for all systems, which is equal to 99%, 94%, 93%, and 92% of the LCCP. The assumption of equal system efficiency has to be questioned. According to Brown et al.'s (2002b) study, the COP of the R-22 system was at least 41% higher than that of the C[O.sub.2] system when both systems delivered the same cooling capacity of 10.5 kW (30 tons). In other words, the C[O.sub.2] system consumed approximately 38% more than the R-22 system. If this is the case, the C[O.sub.2] system would have the highest LCCP.
Military Standard Environmental Control Units
Military standard environmental control units (MIL-STD ECUs) are used by the US Army to provide an air-conditioned environment for their equipment and personnel in the field. MIL-STD ECUs are similar in style to packaged unitary systems but have unique requirements that make them different from commercial air conditioners (Calkins 2000). They are designed to meet both high stresses and environmental extremes. They must meet high vibration and impact requirements resulting from military off-road and rail transport requirements. MIL-STD ECUs are performance rated at a 48.8[degrees]C condenser ambient and 32.2[degrees]C, 50% relative humidity evaporator ambient. Currently, they do not operate as heat pumps due to the poor performance of R-22 at low ambient temperatures. Because military systems are mobile, the primary design concerns for ECUs are weight and size. For example, a 10.55 kW ECU is contained in a volume of approximately 1 [m.sup.3] and weights approximately 200 kg. The volume and weight concerns constrain the system to have smaller coils than commercially available packaged systems. Heavier fan assemblies are required to handle the increased pressure drop that results from the smaller coils and to provide additional airflow. In addition, ECUs are designed to minimize the maximum power draw. As a result, they are not optimized for efficiency over the range of operating conditions.
Currently, the US Army is seeking an alternative refrigerant to replace R-22 in their ECU applications by 2010. C[O.sub.2] is one of the options considered because it has minimal direct environmental impact compared to other refrigerants and because some of its characteristics make it an especially attractive refrigerant to the ECU application (Manzione 1998).
A breadboard transcritical C[O.sub.2] ECU was designed and built by Cutler et al. (2000) to investigate the performance potential of C[O.sub.2] in both air-conditioning and heat pumping modes. The unit used a reciprocating semi-hermetic prototype compressor. The evaporator and gas cooler were both designed using microchannel technology. The refrigerant follows a cross-counterflow path with respect to the airflow direction in order to maximize effectiveness. Experimental results of cooling capacity and COP of the prototype C[O.sub.2] ECU were not published.
Li and Groll (2004) presented a theoretical study that focused on the performance evaluation of a 10.55 kW C[O.sub.2] ECU. The modeling results were generated using the system simulation model ACCO2 (Ortiz et al. 2003). ACCO2 models air-to-air heat pump/air-conditioning systems using C[O.sub.2] as the refrigerant. The predictions of ACCO2 were validated using experimental results obtained with a breadboard C[O.sub.2] system. ACCO2 was then used to simulate the performance of a basic transcritical C[O.sub.2] ECU with a microchannel evaporator and a gas cooler. Both heat exchangers had similar dimensions of width, height, and depth as those of the R-22 ECUs currently used by the US Army. The compressor model was based on the measured performance of a prototype semi-hermetic C[O.sub.2] compressor (Hubacher et al. 2002). Figures 11 and 12 present the results for cooling capacity and cooling COP as a function of the outdoor temperature, respectively. It can be seen from Figures 11 and 12 that the C[O.sub.2] ECU can meet and exceed the cooling capacity and cooling COP as stipulated by the US Army, which are a cooling capacity of 12 kW and a COP of 0.9 at an outdoor temperature of 48.9[degrees]C, an indoor temperature of 32.2[degrees]C, and an indoor relative humidity of 50%.
Heat Pump Water Heaters
It has been shown that tap water heating is one of the most promising applications for the transcritical C[O.sub.2] cycle. The temperature glide of C[O.sub.2] during supercritical heat rejection results in a very good temperature adaptation when heating tap water, which inherits a large temperature glide itself. This, together with the efficient compression and good heat transfer characteristics of C[O.sub.2], makes it possible to design very efficient heat pumping systems. In laboratory settings, heating COPs of up to 4.3 were obtained for a prototype system when heating tap water from 9[degrees]C to 60[degrees]C at an evaporation temperature of 0[degrees]C (Neksa et al. 2000). These results led to a seasonal performance factor of about 4 for an Oslo, Norway, climate using ambient air as the heat source.
Saikawa et al. (2000) started the development of a C[O.sub.2] heat pump water heater for residential use in 1999. A prototype system was designed and constructed; the authors presented the details of the prototype system and the experimental performance. Based on the prototype system, five field test units were developed and installed in five houses in Japan. The units consisted of a heat pump unit and a hot water storage tank with a capacity of 200 L. The size of the heat pump unit was 81 x 31 x 64 cm and the size of the hot water tank was 95 x 40 x 130 cm. The first announcement of the commercial availability of the heat pump water heater was published by Japan Air Conditioning, Heating and Refrigeration News in 2000 (JARN 2000). The lowest ambient air temperature experienced by the field test units was -20[degrees]C. From the test results it was confirmed that the field test unit could produce 90[degrees]C hot water at -20[degrees]C ambient air temperature. The average yearly COP for Tokyo, Japan, operating conditions was estimated from these studies to be 3.0. By now, the technical development of the commercial system has been completed, although small improvements are ongoing, such as the incorporation of an ejector device in the overall cycle (Akagi et al. 2005). Several thousands of these units have been sold on the Japanese market. From their experience, the authors concluded that C[O.sub.2] heat pump water heaters are highly energy efficient and can produce high-temperature hot water at very low ambient air temperatures. Hashimoto (2006) graphically showed that the heat pump water heater with a COP of 3 would consume 30% less primary energy than the combustion-based water heater with a COP of 0.78. The author assumed that the efficiency of power generation and transmission was 0.37, which means that only 37% of the primary energy would be actually delivered to the heat pump water heater. Based on these assumptions and operating conditions, the combustion-based water heater would consume primary energy of 100 W to deliver a heating capacity of 78 W. In comparison, the heat pump water heater would need 70 W of primary energy to achieve the same heating capacity and, thus, would be more energy efficient.
[FIGURE 11 OMITTED]
[FIGURE 12 OMITTED]
Mukaiyama et al. (2000) developed a residential heat pump water heater using C[O.sub.2] as the working fluid. The C[O.sub.2] circuit of the prototype system was composed of a rotary two-stage compressor, a gas cooler, an expansion valve, and an evaporator. The components were specially designed for use with C[O.sub.2] and installed in the outdoor unit of an air conditioner. The nominal heat output of the unit was about 4 kW and the hot water temperature was about 90[degrees]C. Based on the experimental data, the TEWI values were calculated for a Tokyo, Japan, climate. According to the result presented by the authors, the C[O.sub.2] heat pump water heater reduced the TEWI by about 40% compared to a conventional hot water boiler.
Zakeri et al. (2000) reported that the first commercial pilot plant of a heat pump water heater with C[O.sub.2] as the working fluid was built and installed in Norway. The C[O.sub.2] heat pump system uses heat from the condenser of an ammonia refrigeration plant as the heat source. In the cascade heat exchanger, ammonia condenses (either partial or full condensation) and C[O.sub.2] evaporates. The heat pump system produces hot water with a temperature of 70[degrees]C to 80[degrees]C. The operating system for the installed C[O.sub.2] heat pump system is very stable, with little variation in the heat source temperature. At the design condition of an evaporation temperature of 20[degrees]C, the heat pump system has a heating capacity of about 22 kW when heating water from 10[degrees]C to 75[degrees]C. According to the preliminary results, at an evaporation temperature of 14.3[degrees]C, a heating COP of 5.77 is achieved and tap water is heated from 6.7[degrees]C to 66[degrees]C. At a lower evaporation temperature of 11.8[degrees]C, the heating COP was reduced by 5.8% and tap water was heated from 7.1[degrees]C to 67[degrees]C.
Cooling Systems for Glass Door Coolers and Vending Machines
Based on the interest of a leading soft drink manufacturer to use "green" cooling technologies, several manufacturers have begun investigating transcritical C[O.sub.2] systems as the cooling system in glass door coolers and vending machines. Veje and Suss (2004) presented the first test results of such a cooling system with a cooling capacity of 800 W. The authors measured the steady-state performance of the C[O.sub.2] system and compared it to the performance of a conventional R-134a system. Using a thermal expansion valve instead of a capillary tube, the authors reported up to 18% in energy reductions for the glass door cooler and up to 37% in energy reduction for the vending machine. These promising results led to increased research activities with respect to this application.
Transcritical cycle technology using C[O.sub.2] as the refrigerant offers the potential for efficient refrigeration and air conditioning in certain applications, without the ozone depletion and global warming problems associated with conventional refrigerants. The practical feasibility of the basic cycle and the additional features that are needed to enhance the performance of the basic cycle have been investigated to demonstrate this potential. However, introduction of transcritical C[O.sub.2] systems in the marketplace has been very slow. This is mainly due to manufacturers' significant upfront investments, which are part of a change to a drastically new technology, and the associated first costs for the customer. Research is still ongoing with respect to overall system performance, heat exchange component innovations, and compressor development, to name a few topics. However, it is the authors' opinion that this research has slowed down in recent years due to the limited induction of the technology in the market. To some extent, the demonstration of the suitability of transcritical C[O.sub.2] technology as a viable substitute for current vapor-compression technology with HFC refrigerants is still missing and can only happen through technical innovations. In the end, trade-offs between operating and capital costs and evolving legislation in the global economy will govern the widespread implementation of transcritical C[O.sub.2] cycle technology.
Aarlien, R., and P.E. Frivik. 1998. Comparison of practical performance between C[O.sub.2] and R-22 reversible heat pumps for residential use. Proceedings of the IIR-Gustav Lorentzen Conference on Natural Working Fluids, Oslo, Norway, pp. 388-96.
Akagi, S., J.-F. Wang, and E. Hihara. 2005. Characteristic of two-phase ejector in carbon dioxide operated refrigeration cycle. Proc. of the 8th IEA Heat Pump Conf., Las Vegas, NV, May 31-June 2.
Baek, J.S., E.A. Groll, and P.B. Lawless. 2005. Theoretical performance of transcritical carbon dioxide cycle with two-stage compression and intercooling. Proc. IMechE, vol. 219, Part E: J. Process Mechanical Engineering, Special Issues Paper, pp. 187-95.
Brown, J.S., S.F. Yana-Motta, and P.A. Domanski. 2002a. Comparative analysis of automotive air conditioning systems operating with C[O.sub.2] and R134a. International Journal of Refrigeration 25:19-32.
Brown, J.S., Y. Kim, and P.A. Domanski. 2002b. Evaluation of carbon dioxide as an R-22 substitute for residential air conditioning. ASHRAE Transactions 108(2):954-64.
Bullock, C.E. 1997. Theoretical performance of carbon dioxide in subcritical and transcritical cycles. Refrigerant for the 21st century, ASHRAE/NIST Refrigerants Conference, Gaithersburg, MD, pp. 20-26.
Calkins, F. 2000. R-22 baseline testing of the MIL-STD 36,000 Btu/hr ECU. Proc. of the 4th IIR-Gustav Lorentzen Conference on Natural Working Fluids at Purdue, West Lafayette, IN, July, pp. 83-89.
Cho, H., C. Ryu, Y. Kim, and H.Y. Kim. 2005. Effects of refrigerant charge amount on the performance of a transcritical C[O.sub.2] heat pump. International Journal of Refrigeration 28:1266-73.
Choi, J., and Y. Kim. 2004. Influence of the expansion device on the performance of a heat pump using R407C under a range of charging conditions. International Journal of Refrigeration 27:378-84.
Cutler, B., Y. Hwang, L. Bogdanic, and R. Radermacher. 2000. Development of a transcritical carbon dioxide environmental control unit. Proc. of the 4th IIR-Gustav Lorentzen Conference on Natural Working Fluids, Purdue University, West Lafayette, IN, July 25-28, pp. 91-98.
Elbel, S.W., and P.S. Hrnjak. 2004. Effects of internal heat exchanger on performance of transcritical C[O.sub.2] systems with ejector. Proceedings of 10th International Refrigeration and Air Conditioning Conference at Purdue, Purdue University, West Lafayette, IN, paper no. R166.
Gay, N.H. 1931. Refrigeration system, US patent no. 1,836,318.
Giannavola, M.S., R. Murphy, J.M. Yin, M.H. Kim, C.W. Bullard, and P.S. Hrnjak. 2000. Experimental investigation of an automotive heat pump prototype for military, SUV and compact cars. Proc. of the 4th IIR-Gustav Lorentzen Conference on Natural Working Fluids, Purdue University, West Lafayette, IN, July 25-28, pp. 115-22.
Fukuta, M., T. Yanagisawa, and R. Radermacher. Performance prediction of vane type expander for C[O.sub.2] cycle. Proceedings of the 21st IIR International Congress on Refrigeration, Washington, DC.
Ha, J. 1999. Experimental study on the characteristics of flow rate distribution and performance of a multi-heat pump system. Master's thesis, Korea University, Seoul, Korea.
Hall, J.E. 1889. First two-stage C[O.sub.2] compressor, Great Britain. As cited in A History of Refrigeration Throughout the World, R. Thevenot, trans. by J.C. Fidler, Paris, France: International Institute of Refrigeration, 1979.
Haiquing, G., M. Yitai, and L. Minxia. 2006. Some design features of C[O.sub.2] swing piston expander. Applied Thermal Engineering 26:237-43.
Hashimoto, K. 2006. Technology and market development of C[O.sub.2] heat pump water heaters (ECO CUTE) in Japan. IEA Heat Pump Centre Newsletter 24(3):12-16.
Hauk, A., and E. Weidner. 2000. Thermodynamic and fluid-dynamic properties of carbon dioxide with different lubricants in cooling circuits for automobile application. Industrial Engineering Chemistry Research 39(12):4646-51.
Higuchi, M., H. Mori, H. Taniwa, K. Ida, and Y. Nabetai. 2006. Development of the high efficiency and low noise compressor for C[O.sub.2] heat pump water heaters. Proceedings of the Eighteenth International Compressor Engineering Conference at Purdue, Purdue University, West Lafayette, IN, C100.
Hirao, T., H. Mizukami, M. Takeuchi, M. Taniguchi, and A. Yoshioka. 2000. Development of air conditioning system using C[O.sub.2] for automobiles. Proc. of the 4th IIR-Gustav Lorentzen Conference on Natural Working Fluids, Purdue University, West Lafayette, IN, July 25-28, pp. 193-200.
Hubacher, B., E.A. Groll, and C. Hoffinger. 2002. Performance measurements of a semi-hermetic carbon dioxide compressor. Proc. of the Int'l Refrig. and Air Conditioning Conf. at Purdue, Purdue University, West Lafayette, IN, July 16-19, pp. 477-86.
Huff, H.-J., R. Radermacher, and M. Preissner. 2003. Experimental investigation of a scroll expander in a carbon dioxide air-conditioning system. Proceedings of the 21st IIR International Congress of Refrigeration, Washington DC, paper ICR0485.
JARN. 2000. TEPCO/Denso jointly develop C[O.sub.2] heat pump water heater. Japan Air Conditioning, Heating and Refrigeration News 32(4):13.
Klein, S.A. 2004. Engineering Equation Solver (EES). Madison, WI: F-Chart Software.
Kornhauser, A.A. 1990. The use of an ejector as a refrigerant expander. Proc. of the 1990 USNC/IIR-Purdue Refrigeration Conf., Purdue University, West Lafayette, IN, pp. 10-19.
Li, D., and E.A. Groll. 2004. Theoretical performance evaluation of a carbon dioxide based environmental control unit (ECU) with microchannel heat exchangers. Proc. of the 6th IIR-Gustav Lorentzen Conference on Natural Working Fluids, Glasgow, Scotland, Aug. 30-Sept. 1.
Li, D., and E.A. Groll. 2005. Transcritical C[O.sub.2] refrigeration cycle with ejector-expansion device. Int'l J. Refrigeration 28(5):766-73.
Li, D., J.S. Baek, E.A. Groll, and P.B. Lawless. 2000. Thermodynamic analysis of vortex tube and work output expansion devices for the transcritical carbon dioxide cycle. Proc. of the 4th IIR-Gustav Lorentzen Conference on Natural Working Fluids, Purdue University, West Lafayette, IN, July 25-28, pp. 433-40.
Little, A.D. 2002. Global comparative analysis of HFC and alternative technologies for refrigeration, air conditioning, foam, solvent, aerosol propellant, and fire protection applications. Final report to the Alliance for Responsible Atmospheric Policy, March 21.
Lorentzen, G. 1983. Throttling, the internal haemorrhage of the refrigeration process. London: The Institute of Refrigeration, December, pp. 39-47.
Lorentzen, G. 1994. C[O.sub.2] refrigerant in large heat pumps. IEA Heat Pump Centre Newsletter 12(1):35-37.
Lorentzen, G., and J. Pettersen. 1992. Development of a transcritical carbon dioxide environmental control unit. Proceedings of the IIR International Symposium on Refrigeration, Energy, and Environment, Trondheim, Norway, pp. 147-63.
Lorentzen, G., and J. Pettersen. 1993. A new, efficient and environmentally benign system for car air-conditioning. Int'l J. Refrigeration 16(1):4-12.
Maeyama, H., T. Yokoyama, and H. Nakao. 2006. Development of the compressor for C[O.sub.2] heat pumps with the single rotary mechanism. Proceedings of the Eighteenth International Compressor Engineering Conference at Purdue, Purdue University, West Lafayette, IN, C056.
Manzione, J.A. 1998. Development of carbon dioxide environmental control unit for the US Army. Proc. of the IIR-Gustav Lorentzen Conference on Natural Working Fluids, Oslo, Norway, June, pp. 297-302.
McEnaney, R.P., D.E. Boewe, J.M. Yin, Y.C. Park, C.W. Bullard, and P.S. Hrnjak. 1998. Experimental comparison of mobile A/C systems when operated with transcritical C[O.sub.2] versus conventional R134A. Proceedings of the Seventh International Refrigeration Conference at Purdue, Purdue University, West Lafayette, IN, pp. 145-50.
McEnaney, R.P., Y.C. Park, J.M. Yin, and P.S. Hrnjak. 1999. Performance of the prototype of a transcritical R744 mobile A/C system. SAE International Congress and Exposition, Detroit, Michigan, paper no. 1999-01-0872.
Minor, B.H. 2006. DuPont next generation refrigerant MAC global industry. SAE Alternative Refrigerant Systems Symposium 2006, June 27-29, Phoenix, AZ.
Mukaiyama, H., O. Kuwabara, K. Izaki, S. Ishigaki, and T. Susai. 2000. Experimental results and evaluation of residential C[O.sub.2] heat pump water heaters. Proc. of the 4th IIR-Gustav Lorentzen Conference on Natural Working Fluids, Purdue University, West Lafayette, IN, July 25-28, pp. 67-74.
Neksa, P., H. Rekstad, G.R. Zakeri, P.A. Schiefloe, and M.C. Svensson. 2000. Heat pump water heaters with C[O.sub.2] as the working fluid. Short Course on Fundamentals of the Transcritical C[O.sub.2] Cycle Technology, Purdue University, West Lafayette, IN, July 24.
Nickl, J., G. Will, H. Quack, and W.E. Kraus. 2005. Integration of a three-stage expander into a C[O.sub.2] refrigeration system. International Journal of Refrigeration 28:1219-24.
Ortiz, T.M., D. Li, and E.A. Groll. 2003. Evaluation of the performance potential of C[O.sub.2] as a refrigerant in air-to-air air conditioners and heat pumps: System modeling and analysis. Final Report of ARTI Project 610-10030, Herrick Labs 2003-20, Report No. 1275-2.
Ozaki, Y., H. Takeuchi, and T. Hirata. 2004. Regeneration of expansion energy by ejector in C[O.sub.2] cycle. Proceedings of the 6th IIR-Gustav Lorentzen Conference on Natural Working Fluids at Glasgow, Glasgow, UK.
Park, T., M. Lee, J. Jung, and G. Jang. 2004. Comparison of performance variation between R22 and R410A refrigerant systems. Korea Journal of Air-conditioning and Refrigeration 15(3):166-76.
Pettersen, J. 1997. Experimental results of carbon dioxide in compression systems. Proc. of Refrigerant for the 21st Century, ASHRAE/NIST Refrigerants Conference, Gaithersburg, Maryland, pp. 27-37.
Preissner, M., B. Cutler, S. Singanamalla, Y. Hwang, and R. Radermacher. 2000. Comparison of automotive air-conditioning systems operating with C[O.sub.2] and R-134a. Proc. of the 4th IIR-Gustav Lorentzen Conference on Natural Working Fluids, Purdue University, West Lafayette, IN, July 25-28, pp. 185-92.
Robinson, D.M., and E.A. Groll. 1998a. Efficiencies of transcritical C[O.sub.2] cycles with and without an expansion turbine. Int'l J. Refrigeration 21(7):577-89.
Robinson, D.M., and E.A. Groll. 1998b. Heat transfer analysis of air-to-C[O.sub.2] two-phase heat absorption and supercritical heat rejection. HVAC & R Research 4(4):327-45.
Robinson, D.M., and E.A. Groll. 2000. Theoretical performance comparison of C[O.sub.2] transcritical cycle technology versus HCFC-22 technology for a military packaged air conditioner application. HVAC & R Research 6(4):325-48.
Saikawa, M., K. Hashimoto, T. Kobayakawa, K. Kusakari, M. Ito, and H. Sakakibara. 2000. Development of prototype of C[O.sub.2] heat pump water heater. Proc. of the 4th IIR-Gustav Lorentzen Conference on Natural Working Fluids, Purdue University, West Lafayette, IN, July 25-28, pp. 51-58.
Spatz, M.W. 2006. Update on an ultra-low GWP for mobile air conditioning applications. SAE Alternative Refrigerant Systems Symposium 2006, June 27-29, Phoenix, AZ.
Sumantran, V., B. Khalighi, K. Saka, and S. Fischer. 1999. An assessment of alternative refrigerants for automotive applications based on environmental impact. SAE Alternative Refrigerant Systems Symposium 1999, June 28-July 1, Scottsdale, AZ.
Suss, J., and C. Veji. 2004. Development and performance measurements of a small compressor for transcritical C[O.sub.2] applications. Proceedings of the Seventeenth International Compressor Engineering Conference at Purdue, Purdue University, West Lafayette, IN, C112.
Takeuchi, H., Y. Kume, and H. Oshitani. 2002. Ejector cycle system, US patent no. 6,438,993 B2.
Veji, C., and J. Suss. 2004. The transcritical C[O.sub.2] cycle in light commercial refrigeration applications. Proc. of the 6th IIR-Gustav Lorentzen Conference on Natural Working Fluids, Glasgow, Scotland, Aug. 30-Sept. 1.
Voorhees, G.T. 1905. Dual effect compression system, USA. As cited in A History of Refrigeration Throughout the World, R. Thevenot, trans. J.C. Fidler, Paris, France: International Institute of Refrigeration, 1979.
Winkler, J., V. Aute, B. Yang, and R. Radermacher. 2006. Potential benefits of thermoelectric elements used with air-cooled heat exchangers. Proc. of the Int'l Refrigeration and Air Conditioning Conference at Purdue, Purdue University, West Lafayette, IN, July 17-20, paper R091.
Yamasaki, H., M. Yamanaka, K. Matsumoto, and G. Shimada. 2004. Introduction of transcritical refrigeration cycle utilizing C[O.sub.2] as working fluid. Proceedings of the Seventeenth International Compressor Engineering Conference at Purdue, Purdue University, West Lafayette, IN, C090.
Yin, J., C. Bullard, and P. Hrnjak. 2001. R-744 gas cooler model development and validation. Int'l J. Refrigeration 24:652-59.
Zakeri, G.R., P. Neksa, H. Rekstad, K. Lang-Ree, and T. Olson. 2000. Results and experiences with the first commercial pilot plant C[O.sub.2] heat pump water heater. Proc. of the 4th IIR-Gustav Lorentzen Conference on Natural Working Fluids, Purdue University, West Lafayette, IN, July 25-28, pp. 59-66.
Eckhard A. Groll, DrEng
Jun-Hyeung Kim, PhD
Associate Member ASHRAE
Received January 16, 2007; accepted March 7, 2007
Eckhard A. Groll is a professor of mechanical engineering and director of Global Initiatives, Co-operative Education and Professional Experiences, and Jun-Hyeung (Jay) Kim is a post-doctoral research associate, Purdue University, School of Mechanical Engineering, Ray W. Herrick Laboratories, West Lafayette, IN.
Table 1. Comparison of Direct Global Warming Potentials of Some Refrigerants (a) Global warming potential in relation to C[O.sub.2] for 100 years of potential time horizon R-12 R-22 R-134a R404A R407C R410A C[O.sub.2] 10600 1700 1300 3784 1653 1975 1 a. IPCC. 2001. Climate Change, 2001: The Scientific Basis, Intergovernmental Panel on Climate Change. Cambridge, UK: Cambridge University Press.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||REVIEW ARTICLE|
|Author:||Groll, Eckhard A.; Kim, Jun-Hyeung|
|Publication:||HVAC & R Research|
|Date:||May 1, 2007|
|Previous Article:||Admission (economizer) port optimization of a Voorhees modified reciprocating C[O.sub.2] compressor.|
|Next Article:||International Journal of HVAC & R Research.|