People involved with project:
Landi, B. J.; Raffaelle, R. P.; Gennett, T.(RIT); Krainsky, I. L.; Landis, G.; Bailey, S. G.(NASA Glenn RC)
Contractor Christine Chevalier/Analex (HiTemp RF Transistor Modeling)

Objective:
To produce a highly efficient thermionic generator that is suitable for waste heat recovery in aeronautical applications. We will overcome the inefficiency commonly associated with these devices through the use of nanostructured emitters in a composite thermionic vacuum diode arrangement. The materials employed will be thermally and structurally compatible with conventional aircraft such that they will be suitable for integration within a conventional engine nacelle. The conversion of waste heat from the engine to usable electricity will be directed towards supplementing and/or replacing some of the current electrical requirement placed on turbine auxiliary power units, as well as reduce the cooling requirements on the next-generation high operating temperature aircraft engines. A factor of 4 improvement in specific power over conventional APUs, as well as a reduction in overall aircraft noise, fuel consumption, and cost are targeted.

Background:
Nearly 200 years ago it was demonstrated that air in the neighborhood of a hot-body would conduct electricity. In 1899, J.J. Thomson showed that the discharge from a carbon filament in a vacuum tube was carried by negatively charged electrons. In 1903, the Nobel laureate Owen W. Richardson showed that a metallic anode would emit a limited number of electrons that would increase very rapidly with temperature. The maximum current i at any temperature T was governed by the law

(1)
where k is Boltzmann’s constant, and A and w are specific constants of the material.1 Simply stated, this means electrons can escape the surface of a hot material when their kinetic energy corresponding to their velocity normal to the surface exceeds the work function of the material. (It is interesting to note that one of the first materials used to demonstrate this law was carbon). It was not long before it was realized that this phenomena would allow heat to be changed into electricity or direct thermoelectric conversion. Similarly, it was also demonstrated that this phenomenon could be used to extract heat from a hot body as well under the application of an applied current.
Thermoelectric (TE) performance is measured in terms of a dimensionless figure of merit zT. This figure of merit is expressed as

(2)
where S is the Seebeck coefficient, is the electrical conductivity, and ? is the thermal conductivity of the junction.2 Recently it has been proposed that tremendous advantages in the zT values are possible through quantum confinement in nanostructured materials. It has been shown that zT increases exponentially with decreasing nanoscale diameter. Practical nanostructured thermoelectric converters with zT > 5 have been designed.3-4 Thermionic devices are predicted to have efficiencies of greater that 20% at 400 oC, low operational costs, and higher reliability that IC motor-generator sets.

A thermionic energy converter or “thermal diode” consists of a cathode, vacuum gap, and anode. Heating the cathode causes the electrons to “boil off”, transverse the gap, and be absorbed into the colder anode where they can be connected to an external load. In an ordinary parallel plate configuration using metallic electrodes the limitations included high manufacturing costs and temperatures above 1100 oC in order to have reasonable efficiencies. This has limited their application to nuclear power converters in space probes and satellites. However, recent advances in thermionics, including the use of nanotechnology, have opened the door to new applications.

An added benefit of the proposed technology development is that it may also provide a viable means of removing unwanted heat from aircraft without the need of compressor gasses and plumbing required in conventional mechanical cooling systems. As a thermoelectric cooler, a 1 cm square device could theoretically provide 5000 W of cooling with a higher coefficient of performance (COP) than a compressed gas system and a less that 10% of the size and weight.5 Mahan predicted that thermionic device efficiencies as high as 80% if the Carnot value are possible, as compared to high-performance compressors at only 50%.6 It was recently reported by in Aviation Week that Cool Chips PLC, Gilbraltar, had produced a prototype thermionic device which demonstrated a 10 A quantum mechanical electron tunneling current over a wafer area of approximately 9 sq. cm.

Approach:
High purity single wall carbon nanotubes are produced via laser vaporization in a high temperature furnace under argon using a 755 nm alexandrite laser and a doped graphite target. The raw “soot” is purified by a nitric acid refluxing followed by an annealing in oxygen. We have used this method to produce single wall carbon nanotubes which are over 99% pure by weight using this method (see Figure 1).11

Figure 1. High resolution TEM images of (a) “Raw” laser vaporization soot; (b) single-wall nanotube bundles after reflux

Figure 1. (c) high-purity (>99% by weight) single wall carbon nanotubes; (d) nanotube “tape.”

Recently, a sol-gel technique was developed which produced an inorganic-nanotube composite film which would be suitable for a thermionic device. This technique produced TiO2 films with homogeneously embedded carbon nanotubes. 12 This methodology is quite attractive because it permits the ease of doping profiles and allows for the possibility of depositing materials of varying thickness from 10 nm up to 1 micron. Sol-gel procedures for producing TiO2 are well established and have been described in the literature.13-14 The nanotube/TiO2 composites will be made by first dispersing the nanotubes in methanol, followed by the direct addition to the sol-gel solution before complete reaction of the TiO2 precursor. This procedure has been shown to produce an excellent dispersion, whose homogeneity increases over time. The sol-gel solutions will be cast directly onto metallized ?-alumina substrates with through-vias and heat-treated. Photoresist masking will be used to deposit first titanumium and then indium sidewalls. The devices are then assembled by making a sandwich of two of these electrodes and heating them under high vacuum to provide the metal-sealed vacuum between the anode and cathode. The separation is controlled by the difference in the metal side-wall and the nanotube composite thickness (see Figure 2).

Figure 2. Prototype Nanotube Thermoionic Power Converter All the nanotube material is being tested for thermionic emission, and work function is measured in high vacuum chamber.

Literature:

1. G.N. Hatsopoulos and E.P. Gyftopoulos, Thermionic Energy Conversion (MIT Press, Cambridge, 1973-79), Vols. I and II.
2. H.J. Goldsmit, Electronic Refrigeration (Pion, London, 1986).
3. L.D. Hicks, T.C. Harman, X. Sun, and M.S. Dresselhaus, Phys. Rev. B 53, R10493 (1963).
4. T.C. Harman, D.L. Spears, and M.J. Manfra, J. Electron. Mater. 25, 1121 (1996).
5. Aviation Week and Space Technology, Jan 14., P. 425 (2002).
6. Aviation Week and Space Technology, June 24., P. 425 (2002).
7. G.D. Mahan, J. appl. Phys., 76, 4362 (1994).
8. R.W. Sigel, et.al., Scripta Mater 44, 2061 (2001).
9. Peigney, et.al., Chem. Phys. Lett 352, 20 (2002).
10. G.L. Hwang and K.C Hwang, J. of Mater. Chem. 11, 1722 (2001).
11. A.C. Dillon, et. al., Advanced Materials, 11, 1354-1358(1999).
12. P.Vincent, A. Brioude, C. Journet, S. Rabaste, S.T. Purcell, J. Le Brusq, J. C. Plenet, J. Non-Crystalline Solids, in press.
13. L. Klein, Sol-gel Technology for Thin Films, Fibers Preforms, Electronics and Specialty Shapes, Noyes Publication (1988).
14. A. Bahtat, M. Bonazaoui, M. Bahtat, C. Garapon, B. Jacquier, J. Mugnier, J. Non-Cryst Sols. 16, 202 (1996).
15. Elich, J. M.; Landi, B. J.; Raffaelle, R. P.; Gennett, T.; Krainsky, I. L.; Landis, G.; Bailey, S. G. "Single Wall Carbon Nanotube Thermionic Emitters," Proceedings of the First Int. Energy Conversion Engineering Conference: Portsmouth, VA, p. 233 – 236 August 2003.