Thermoelectric (TE) materials convert thermal
to electrical energy, and vice versa, making them ideally suited to power
generation and refrigeration applications. TE devices are attractive because
of their highly reliable, silent, and vibration-free operation; lack of
compressed gases, chemicals, or other consumables; and complete scalability.
However, because of their relatively high cost and low efficiency, TE devices
have been restricted to applications for which high reliability, portability,
or small size are important, such as generators in satellites and space
probes, and automotive seat coolers. The primary goal of researchers in
the field has been to increase the efficiency of TE materials, expressed
by the figure of merit:
ZT=S2σT/κ
where S is the Seebeck coefficient, σ is the electrical conductivity,
T is the absolute temperature, and κ is the thermal conductivity. Substantial
ZT enhancements are predicted for nanostructured materials due to both
quantum confinement effects and increased phonon scattering at interfaces. Our
work is an attempt to realize these enhancements through the synthesis
of nanowire arrays of known TE materials (bismuth chalcogenides, bismuth
antimony, and CoSb
3) by electrochemical deposition.
Electrochemical Deposition of Nanowires
Widespread use of TE energy
conversion requires that the materials and devices can be synthesized/fabricated
inexpensively on large scales without sacrificing control of the composition
and spatial arrangement. Electrodeposition is a well-known technique that
is used industrially on large scales, and has a number of variables that
can be tuned for composition and morphology control. We are interested
in the electrodeposition of thermoelectric nanowires using porous aluminum
oxide (alumina) as a template. In this process, the electrically conducting
TE material is electrodeposited from solution within the empty channels
of an electrically insulating alumina template. This yields dense arrays
of parallel, cylindrical, high aspect ratio nanowires.
Porous Anodic Alumina
Porous alumina templates can be fabricated with a wide range of pore
diameters, pore spacings, and template thicknesses. We start by anodizing
polished aluminum foil in an acidic bath. This oxidizes the aluminum and
initiates the formation of porous alumina (Al2O3).
Anodizing aluminum in the right acids spontaneously yields cylindrical
nanochannels that, under the right conditions, are hexagonally ordered. The
applied voltage and duration of aluminum anodization determines the final
alumina template thickness. Our home-made porous alumina templates are typically
50-75 µm thick with 20-40 nm diameter pores; nanowires grown in templates
with these specifications will have aspect ratios well over 1000. Once the
aluminum is anodized, a metal electrode is sputtered on one face of the template;
this metal layer will serve as the cathode during the electrodeposition process.
Nanowire and Array Characterization
As we incorporate our nanowire arrays into prototype TE devices,
we need to be able to correlate the electrochemical synthesis with the
thermal and electrical properties of the nanowires. Characterization techniques
used for the films and the nanowire arrays are x-ray diffraction (crystallinity
and orientation), scanning electron microscopy (morphology and pore filling
ratio), and energy dispersive spectrometry (composition). A thorough characterization
of the microstructure is critical for understanding how the synthesis,
properties and performance in TE devices are related. Transmission electron
microscopy (TEM) couples spectrometry for chemical composition with near-simultaneous
imaging and diffraction for microstructure characterization. The ability
to determine chemical composition at the nanoscale provides a check of
overall homogeneity and dopant distribution. X-ray diffraction (XRD) tells
us how the electrochemical synthesis imparts a preferred orientation into
the polycrystalline-nanowire array, but TEM is required for orientation
information from individual nanowires. Grain size and grain boundary orientation
are also important because phonons and electrons may be scattered with
different efficiencies depending on their mean free paths and orientation
mismatch between grains. We are currently optimizing methods for gaining
a semi-quantitative understanding of nanowire microstructure in hopes of
being able to rapidly predict how the synthesis of a given nanowire array
will determine its thermal and electrical properties.
• Collaborators
