Environmental Engineering Reference
In-Depth Information
Fig. 4.38 Photograph of the epoxy-bonded plate with glued spacers
The plates were made in a special Te
on mould. The powder was mixed with
epoxy resin and after vacuuming, the plates were cold pressed inside the Te
fl
on
mould. The composite plate stayed in the mould for 24 h, when it was removed and
subsequently cured under ambient conditions. The thickness of the plates was
approximately 0.4 mm. However, in order to study the impact of the epoxy and the
plate composition in detail, different epoxy-bonded magnetocaloric plates were
made and analysed regarding their magnetocaloric, thermal and cycling properties
and further directly compared to the samples prepared by the TDR technique (as
explained above). The samples had a volume fraction of magnetocaloric powder
between 45 and 60 %. As was expected, the sintered plates have somewhat better
magnetocaloric and especially thermal properties compared to the epoxy-bonded
plates. The main reason for that is the impact of the epoxy, since it has a relatively
high speci
fl
c heat and a low thermal conductivity. The sample with the lowest
amount of epoxy exhibited the highest adiabatic temperature change (but about
50 % smaller than the sintered samples). In the next step the AMR was made using
epoxy-bonded magnetocaloric plates (see Fig. 4.38 ) and applied in the magnetic
refrigerator. The AMR consists of plates with a thickness of 0.4 mm and a spacing
of 0.1 mm and was produced by gluing, which is better than the current state of the
art, which involves the TDR method (but its magnetocaloric properties are poorer).
A very similar idea was also applied by Skokov et al. [ 145 ] with La(Fe,Si) 13
powder. The authors used 5 wt% of silver epoxy and powder between 50 and
300
m. The samples were then pressed into thin plates with up to 0.3 GPa. The
results for the optimized epoxy-bonded samples (powder of 100
ʼ
m and compacted
pressure of 0.1 GPa) show a very promising magnetocaloric effect (in some cases an
even higher adiabatic temperature change compared to the bulk samples). They also
fabricated an AMR based on that technique with plate thickness and a channel size
of 0.6 mm (but this was not applied to the magnetic refrigerator).
It is not yet clear which fabrication method or even which magnetocaloric
material is going to be successfully applied in any future (commercial) magnetic
refrigerator. As many times pointed out in this chapter, the AMR geometry (besides
the magnetocaloric effect) plays one of the most important roles in a highly ef
ʼ
cient
magnetic regenerator and therefore suitable fabrication methods must be applied.
However, its time
and cost ef
ciency are required as well.
Search WWH ::




Custom Search