Proceedings of ISES Solar World Congress 2007: Solar Energy and Human Settlement
552
require a spectrally selective surface that is capable of
absorbing as much as possible of the solar radiation
incident on it, while it re-emits very little in the near and far
infra-red wavelength ranges. There exists already a range
of solar collectors that use solar selective coatings [5]
mainly for domestic water heating and space heating
applications. The global use of these devices has been
very low due to the high cost of production of the selective
surface which is the most important component in solar
collectors. Solar selective coatings have been manufactured
by sputtering and thermal evaporation techniques [2], both
of which have high and prohibitive initial investment cost
of equipment and consumable materials. There is therefore
a need to investigate new inexpensive production
techniques for solar selective surfaces that match or excel
the present commercial coatings in performance and
durability characteristics. Lately, there has been a trend to
move towards the use of sol-gel techniques in the
production of solar selective coatings.[3, 6] The sol-gel
technique brings with it much better control of the
selectivity characteristics of coatings due to easy control of
production parameters such as composition, particle size,
particle size distribution, coating thickness, morphology
and so on. Traditionally, sol-gel techniques employ
alkoxides as precursor materials of oxides in which solar
absorbing particles are embedded. In this work, an even
less expensive and more environmentally friendly
technique of using acetates instead of alkoxides is
investigated. The investigation is conducted on carbon
nanoparticles dispersed in ZnO and NiO layers. Carbon is
used as the absorbing particles, while ZnO and NiO have
been chosen for their appropriate near and far infra-red
reflectance characteristics. A further added advantage of
carbon is that it is relatively easy to form carbon
nanoparticles in situ in the oxide matrix using a carbon
precursor compound.
2. SAMPLE PREPARATION
Rolled aluminium was used as substrates in the form of 55
mm squares that minimised edge-effects of coatings and at
the same time fitted well into a 60 mm diameter tube
furnace. The substrates were treated in dilute phosphoric
acid bath held at 60 °C for 20 minutes to remove an oxide
layer that normally forms on aluminium in air. The
substrates were rinsed thoroughly in distilled water and
then dried by blowing with nitrogen gas.
The preparation technique for the solar selective coatings
was as follows: Separate solutions of zinc acetate and
nickel acetate dissolved in 50 ml absolute ethanol were
prepared. The zinc and nickel acetates were used as ZnO
and NiO precursor materials, respectively. Diethanolamine
(DEA) was added to the oxide precursor solutions to act as
a chelating agent. A solution of α-D-glucopyranoside,
used as a carbon precursor, was intermixed with each of the
oxide precursor solutions and stirred with a magnetic stirrer.
The stirring was necessary to make a homogeneous
solution. After a period of stirring a few drops of
polyethylene glycol (PEG) were added and the stirring
continued until a sol was formed. PEG was employed as a
structure directing template.
The sol was spin coated at 4000 rotations per minute on the
aluminium substrates. The spin coated samples were
heat-treated in a nitrogen gas environment in a tube furnace.
The heat treatment sintered the oxide matrix and at the
same time reduced the α-D-glucopyranoside to carbon.
The furnace temperature was set at 550 °C for one hour.
A furnace temperature ramp rate of 5 °C per minute was
used at the start and end of the heat treatment stage to avoid
cracks in the coatings.
3. OPTICAL MEASUREMENTS AND STRUCTURAL
DETERMINATION
Normal spectral reflectance measurements were made on
the samples. A Lambda 900 spectrophotometer was used
for measuring reflectance in the 0.3 to 2.5 μm wavelength
range and a Bruker Tensor 27 spectrophotometer was used
for reflectance measurements in the 2.5 to 20 μm
wavelength range. The spectral reflectance measurements
were used to calculate the total solar absorptance, α and the
total hemispherical thermal emittance, ε according to
equations (1) and (2), respectively (2, 6):
()(1 ())
()
Rd
Id
λλλ
α
λλ
−
=
∫
∫
(1)