16 Magnetic Fluids for Thermoelectricity
389
preparation of ionic nanofluids made of interacting nanoparticles is faced with a
formidable challenge on its own. Ferrofluids, in particular, contain freely moving
magnetic nanoparticles that tend to form chains and aggregates through attractive
magnetic forces, unless screened properly by other repulsive forces (electrostatic,
steric, etc.) which require a fine tuning of the coating ions of individual nanoparticles
as well as the surrounding ionic condition (e.g. electrolyte ions, counter- and co-ions).
The ferrofluids used in the following example studies are composed of maghemite
(γ-Fe 2 O 3 ) nanoparticles with average particle diameters in the range of 6–8 nm. They
are either ionically stabilized ([30, 32, 33]) or with polymers ([34]) and dispersed
in polar liquids (water and organic solvent). The thermoelectric measurements were
performed in the thermogalvanic cell condition, i.e. a small amount of redox couple
molecules were always present.
A schematic image of the thermoelectric measurement cell is shown below. In
order to properly measure the “basic” liquid TE property, following precautions
should be taken.
• Cell body: The thermal conductance of the material should be smaller or comparable to that of the liquid sample. The material or the design must be able to
accommodate the liquid dilation at high temperature. The material must be inert
(no chemical reactions with the fluid can occur) and impermeable.
• Electrodes: The material must be electrochemically stable, possess high thermal
conductivity (to ensure good heat transfer from the heater/cooler to the liquid).
• Faraday cage, electrometer: For highly resistive samples, it is desirable to shield
the cell from the environmental EM waves, and use a high input-impedance
voltmeters (10
11
or higher).
• Measurements: In order to avoid introducing convective motions of the fluid and/or
nanoparticles, the cell should be heated from the top (Fig. 16.2).
Due to the thermodiffusion effect, especially that of magnetic nanoparticles, the
thermoelectric voltage (thus the Seebeck coefficient) of nanofluids evolves over a
long period of time, from several minutes to several days, depending on the viscosity
of the liquid, the cell geometry, the hydrodynamic size of the nanoparticles and
their concentration. Such a slow process can become experimentally cumbersome;
however, it is precisely this time dependency of Se that provides us the means to distinguish between different thermoelectric phenomena taking place inside the complex,
magnetic nanofluids.
16.2.2 Experimental Determination of Eastman Entropy
of Transfer in Ferrofluids
In 2015, Huang et al. have published the first experimental study on the Seebeck
coefficient (S e
ini ) in ferrofluids, combined with the corresponding Soret coefficient
S T [31]. The ferrofluids used here consist of charge-stabilized nanoparticles dispersed
389
preparation of ionic nanofluids made of interacting nanoparticles is faced with a
formidable challenge on its own. Ferrofluids, in particular, contain freely moving
magnetic nanoparticles that tend to form chains and aggregates through attractive
magnetic forces, unless screened properly by other repulsive forces (electrostatic,
steric, etc.) which require a fine tuning of the coating ions of individual nanoparticles
as well as the surrounding ionic condition (e.g. electrolyte ions, counter- and co-ions).
The ferrofluids used in the following example studies are composed of maghemite
(γ-Fe 2 O 3 ) nanoparticles with average particle diameters in the range of 6–8 nm. They
are either ionically stabilized ([30, 32, 33]) or with polymers ([34]) and dispersed
in polar liquids (water and organic solvent). The thermoelectric measurements were
performed in the thermogalvanic cell condition, i.e. a small amount of redox couple
molecules were always present.
A schematic image of the thermoelectric measurement cell is shown below. In
order to properly measure the “basic” liquid TE property, following precautions
should be taken.
• Cell body: The thermal conductance of the material should be smaller or comparable to that of the liquid sample. The material or the design must be able to
accommodate the liquid dilation at high temperature. The material must be inert
(no chemical reactions with the fluid can occur) and impermeable.
• Electrodes: The material must be electrochemically stable, possess high thermal
conductivity (to ensure good heat transfer from the heater/cooler to the liquid).
• Faraday cage, electrometer: For highly resistive samples, it is desirable to shield
the cell from the environmental EM waves, and use a high input-impedance
voltmeters (10
11
or higher).
• Measurements: In order to avoid introducing convective motions of the fluid and/or
nanoparticles, the cell should be heated from the top (Fig. 16.2).
Due to the thermodiffusion effect, especially that of magnetic nanoparticles, the
thermoelectric voltage (thus the Seebeck coefficient) of nanofluids evolves over a
long period of time, from several minutes to several days, depending on the viscosity
of the liquid, the cell geometry, the hydrodynamic size of the nanoparticles and
their concentration. Such a slow process can become experimentally cumbersome;
however, it is precisely this time dependency of Se that provides us the means to distinguish between different thermoelectric phenomena taking place inside the complex,
magnetic nanofluids.
16.2.2 Experimental Determination of Eastman Entropy
of Transfer in Ferrofluids
In 2015, Huang et al. have published the first experimental study on the Seebeck
coefficient (S e
ini ) in ferrofluids, combined with the corresponding Soret coefficient
S T [31]. The ferrofluids used here consist of charge-stabilized nanoparticles dispersed