Solid-state cooling, part I: vapor cycles and Peltier devices.
Peltier devices, commonly called thermoelectrics, have been available commercially for nearly 50 years, with active research ongoing on improvements. The other solid-state cooling technologies are still in the research stage and have not resulted in any mainstream commercial products.
The magnetocaloric effect is a solid-state effect in which certain solid materials respond to a change in applied magnetic field with a temperature change. However, the temperature change is so small that regenerative cycles with a liquid heat transport fluid as an integral part of the cycle are needed for any of the normal building air-conditioning or food refrigeration applications, so it is really a hybrid solid/fluid cycle.
The advantages and potential advantages of solid-state cooling devices include the absence of working fluids having environmental issues (e.g., high ODP and/or GWP), small volume and weight, and the ability to make small cooling capacity devices without any scale-down loss of efficiency.
Because solid-state cooling devices are not subject to the thermodynamic losses that occur in the vapor compression cycle, potentially, they could provide cooling efficiencies above the efficiencies at which vapor cycles have plateaued in recent years. The latter point is the rationale for much of the research being pursued in solid-state cooling.
While it is true that vapor compression cycle cooling equipment operates at efficiencies less than the ideal Carnot cycle due to inherent losses, solid-state cooling devices have losses of their own.
The key challenge to implementing solid-state cooling is to find materials and configurations that minimize their inherent losses and can exceed the efficiency levels that are routinely obtained with conventional vapor compression cycle equipment and to do so at a comparable cost to conventional vapor compression cycle equipment.
When considering the discrepancy between the actual and the ideal efficiency of any cooling cycle, the high and low working fluid temperatures (or the high and low solid material surface temperatures) will be considered, because all cooling equipment ultimately needs to transfer heat from the air or liquid being cooled to the cooling cycle and from the cooling cycle to the air or liquid heat sink.
Starting with the vapor compression cycle, there are four main losses that account for most of the discrepancy between actual and ideal efficiency for a given set of low and high (evaporating and condensing) temperatures, as indicated on the vapor cycle P-h diagram (Figure 1):
* Less than ideal compressor efficiency;
* Rejection of the compressor discharge superheat portion of the waste heat above the condensing temperature;
* Constant enthalpy expansion, dissipating potentially recoverable work; and
* Pressure losses in the coils and interconnecting tubing.
We will take as an example a vapor cycle air conditioner operating at the standard AHRI rating temperatures for refrigerant compressors for air conditioning (130[degrees]F [54[degrees]C] condensing, 45[degrees]F [7[degrees]C] evaporating, 20[degrees]F [-7[degrees]C] suction vapor superheat, 15[degrees]F [-9[degrees]C] liquid subcooling), with a compressor with the best coefficient of performance (COP) that is commercially available Btu/W-h EER or COP of 3.22, with R-410A, motor-compressor isentropic efficiency of 74%).
Allowing for 1[degrees]F (0.6[degrees]C) of saturated temperature loss due to pressure drops on the high and low sides (for average evaporating and condensing temperatures of 46[degrees]F [7.8[degrees]C] and 129[degrees]F [53.9[degrees]C], respectively), the Carnot COP of a Carnot cycle (TC /(TH-TC)) with TC and TH of 46[degrees]F [7.8[degrees]C] and 129[degrees]F [53.9[degrees]C], respectively, is 6.10. Adjusting for the suction vapor superheat and liquid subcooling gives an ideal COP of 6.2, so the vapor cycle COP is 52% of ideal. Real solid-state cooling devices also will be subject to deviations from ideal.
[FIGURE 1-2 OMITTED]
The basics of Peltier devices are illustrated in Figure 2. The typical construction of a thermoelectric module is an array of N- and P-type bismuth telluride pellets on hot and cold side ceramic substrates. Electric current alternately flows through the P and N pellets, always flowing in the same direction through each type of pellet. The Peltier effect causes heat to flow from the cold side to the hot side, in the same direction of current flow through the P-type pellets and in the opposite direction of current flow through the N-type pellets.
Because thermoelectric materials have a finite electric resistivity and a finite thermal conductivity, these devices are subject to two primary losses, electric resistance loss and thermal conduction, that increase the power input and reduce the net flow of heat from cold to hot. Electric resistance voltage loss raises the voltage that must be applied to drive current through the thermoelectric module, increasing the required power input. At the same time, the electric resistance heat dissipated in the module reduces the net heat flow from the cold to the hot side.
Due to the finite thermal conductivity of the thermoelectric material, heat is conducted from the hot to the cold side, again, reducing the net flow of heat from the cold to the hot side. The thermoelectric figure of merit, ZT, of a thermoelectric material is a dimensionless term that balances the strength of the thermoelectric effect with the losses due to thermal conductivity and electric resistivity:
ZT = [S.sup.2]T/pk,
where S is the Seebeck coefficient for the material, T is the absolute temperature, p is the electrical resistivity of the material and k is the thermal conductivity of the material. Higher ZT values correlate directly with higher COP.
In conventional, commercially available devices, the electric resistance and thermal conduction losses are significant enough to limit ZT at room temperature to 1.0, with the maximum COP being well below the COP of vapor cycle systems. (1,2,3,4,5,6) Despite the low COP, commercially available thermoelectric devices offer the advantage of being compact and suitable for low-capacity cooling requirements such as small picnic coolers and electronic chip-cooling applications. For COPs that are competitive with mainstream vapor cycle air-conditioning and refrigeration equipment, considerably higher ZT values, of 3 and higher, are needed.
A considerable amount of ongoing research is directed toward reducing both of these losses relative to the Peltier heat pumping capacity. Typically, the goals of reducing thermal conductivity while increasing electrical conductivity (the inverse of electric resistivity) are not compatible. While this holds for the semiconductor class of materials that provide the best thermoelectric performance, semiconductors have unique characteristics that can be exploited.
Much research has focused on creating quantum wells or quantum dots that provide greater electron mobility for the electrons that are actually transporting heat (reducing electric resistivity), while decreasing the thermal conductivity at the same time. The bulk of the reported research has explored the use of thin films, using various semiconductor fabrication techniques, or nanocomposites to construct these quantum features.
A variety of material systems have been identified that can be fabricated in this manner to achieve higher ZT. Examples range from bismuth telluride (1,6) to combinations of bismuth and tellurium and other elements such as antimony and selenium (3,6) to lead-selenium-tellurium (4) to polyphenyl ether molecule chains (5) to cobaltates on a silicon substrate. (7)
As indicated previously, thermoelectric devices with COPs better than conventional air-conditioning and refrigeration equipment have not been commercialized. Research results reported in recent years include small-scale measurements of room temperature ZT values approaching the nominal goal of 3 needed to begin to produce efficiency-competitive cooling equipment.
Assuming that viable, higher efficiency thermoelectric cooling systems emerge for mainstream cooling applications, the potential for energy savings is large. The total primary energy consumed annually for air conditioning and refrigeration in residential and commercial buildings in the U.S. (in 2006) is approximately 6.2 quads. (8) Significant additional energy is consumed for mobile air conditioning. The ultimate savings will depend on the magnitude of efficiency improvement over vapor cycle and the extent of market penetration, neither of which can be predicted now.
Conventional thermoelectric modules have an economic advantage over the vapor cycle where small cooling capacities are needed, generally if less than 50 W of cooling capacity is needed. For these small capacities, one or several thermoelectric modules can provide the required cooling capacity, with significantly less space and weight and at lower cost than a vapor cycle cooling system. While the cooling COP is considerably lower with currently available thermoelectric modules, at small capacity the energy use is often not a major issue.
Among the common air-conditioning and refrigeration products, the cooling capacity needed for residential refrigerators, ranging from 100 W to 400 W, is a modest increase from the small capacity requirements that can be met by a few thermoelectric modules. To be commercially viable in this context, ZT values above 3 and lower costs will be needed. For the larger capacity requirements for residential and commercial air conditioning and commercial refrigeration, ZT values above 3 will be needed and the cost will need to be even lower.
(1.) Bottner, H., G. Chen, and R. Venkatasubramanian. 2006. "Aspects of thin-film superlattice thermoelectric materials, devices, and applications." MRS Bulletin 31(3):211 - 217. http://tinyurl.com/superlattice.
(2.) Yang, R., G. Chen. "Nanostructured thermoelectric materials: from superlatttices to nanocomposites." Materials Integration 18(9):31 - 36. http://tinyurl.com/nanostructured.
(3.) Bottner, H. 2005. "Micropelt miniaturized thermoelectric devices: small size, high cooling power densities, short response times." 2005 ICT Conference. http://tinyurl.com/micropelt.
(4.) Harman, T. C., et al. 2002. "Quantum dot superlatttice thermoelectric materials and devices." Science 297(9):2229 - 2232.
(5.) 2010. "Quantum effects help reduce waste heat." Materials Today. http://tinyurl.com/materials-today.
(6.) Venkatasubramanian, R., et al. 2001. "Thin-film thermoelectric devices with high room-temperature figures of merit." Nature 413:597 - 602.
(7.) 2005. "Scientists grow thermoelectric cobaltate thin films on silicon." Science Daily. http://tinyurl.com/cobaltate.
(8.) Energy Efficiency and Renewable Energy. 2010. 2009 Buildings Energy Data Book. U.S. Department of Energy. http://buildingsdatabook. eere.energy.gov/.
By John Dieckmann, Member ASHRAE; Alissa Cooperman; and James Brodrick, Ph.D., Member ASHRAE
John Dieckmann is a director and Alissa Cooperman is a technologist in the Mechanical Systems Group of TIAX, Cambridge, Mass. James Brodrick, Ph.D., is a project manager with the Building Technologies Program, U.S. Department of Energy, Washington, DC
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||EMERGING TECHNOLOGIES|
|Author:||Dieckmann, John; Cooperman, Alissa; Brodrick, James|
|Date:||Mar 1, 2011|
|Previous Article:||Fed budget & ASHRAE.|
|Next Article:||Material selection to O&M: materials-related IAQ.|