Environmental Engineering Reference
In-Depth Information
total mass of gadolinium was 142.7 g. However, this AMR had plates positioned
with the normal to the main surface being perpendicular to the magnetic
eld.
Further experiments were performed on three AMRs that consisted of packed-
bed gadolinium particles. One AMR comprised small Gd cylinders with a diameter
of 2.5 mm, a length of 4 mm and a total mass of 120.5 g. The other AMR was
lled
with 135 g of spheres, having a diameter in the range 0.35
0.5 mm. In the third
AMR, 93 g of Gd powder was used. The size of the powder particles was also in the
range 0.35
-
0.5 mm.
Initially, no-load temperature-span tests were conducted on parallel plate AMRs.
The best no-load temperature span of 19.8 K was achieved using an AMR with
parallel plates that had a smaller spacing. In the AMR with parallel plates that have a
wider spacing, the no-load temperature span was 16 K. The no-load temperature span
for the AMR with plates positioned with the normal to the main surface being
perpendicular to the magnetic
-
eld was 14 K due to the higher demagnetization effect.
The best performance of the packed-bed AMRs was obtained with spheres
(15.5 K). In the powder-based AMR 7 K of no-load temperature span was obtained.
The worst performance was demonstrated by the AMR consisting of cylinders
(temperature span of only 4 K). The poor performance of this AMR was mainly due
to a too large diameter and length of magnetocaloric cylinders, resulting in an
inef
cient heat transfer.
Tests on the cooling power were only performed for the best of the evaluated
AMRs (parallel plates with a smaller spacing). In this case, the zero-temperature
span-speci
c cooling power was 39.7 W kg 1 . At 13.5 K of temperature span the
device produced 11.3 W kg 1 of speci
c cooling power.
In the next experimental study from 2014, other magnetocaloric materials were
tested in the same experimental device as in Tu
32 ] and compared to
the best performing Gd parallel plate AMR [ 32 ]. Three different AMRs were built,
with two, four and seven layers of the La
š
ek et al. [ 30
-
Si material [ 33 ]. The Curie
temperatures of two-layered AMR were 291.2 and 296.8 K, respectively. The Curie
temperatures of the four-layered AMR were 291.2, 296.8, 303 and 308 K,
respectively. The seven-layered AMR had additional materials to the AMR with
four layers. Two materials were added at the cold end, having their Curie tem-
peratures at 280.8 and 283.8 K, and additional material was added to the hot end
with a Curie temperature of 312 K. The plate thickness of all seven magnetocaloric
materials was 0.5 mm, with a spacing of 0.2 mm. The total mass of each AMR was
144 g, regardless of the number of layers. The highest no-load temperature span
was similar for the seven-layered, as well as for the four-layered, AMR, and it was
approximately 20 K. The two-layered AMR
Fe
Co
-
-
-
s performance was slightly lower, with
a no-load temperature span of about 16 K. The speci
'
c cooling power of the seven-
layered and four-layered AMR was approximately 15 W kg 1 at a temperature span
of 8 K. The two-layered AMR preformed approximately 11 W kg 1 of speci
c
cooling power at 5 K. Furthermore, the in
uence of the hot-side (heat sink) tem-
perature was also investigated. In this manner, the highest no-load temperature span
of the device was achieved by Gd parallel plate AMR. It was approximately 23.5 K
at the heat sink temperature of about 300 K.
fl
Search WWH ::




Custom Search