Journal of Alloys and Compounds448(2008)
73–76
The magnetic entropy change in CoMnSb
alloys with different crystal sizes
Shandong Li a,b,∗
a Department of Physics,Fujian Normal University,Fuzhou350007,China
b National Laboratory of Solid State Microstructure and Department of Physics,
Nanjing University,Nanjing210093,China
Received30November2006;received in revid form11March2007;accepted12March2007
Available online16March2007
Abstract
The magnetocaloric effect(MCE)in CoMnSb has been investigated by comparing two samples with d
ifferent crystal size.One sample is the ingot with crystal size of120nm,referred as sample A.The other with average crystal size of30nm has been fabricated by rapid solidification method,referred as sample B.It has been found that crystal size dramatically affects the magnetic properties and MCE for CoMnSb alloy.For example,in comparison with sample A,sample B exhibits a lower magnetization and Curie temperature,but an enhanced refrigerant capacity and broader working temperature range.The facts indicate that sample B is superior to the ingot in practical application.The above results are explained in terms of the effect of crystal size and atomic disorder on the intrinsic magnetic properties and magnetic entropy change.
©2007Elvier B.V.All rights rerved.
Keywords:Nanostructured materials;Magnetocaloric;Transition metals alloys and compounds
1.Introduction
In recent years,the magnetic refrigerants have drawn an increasing attention becau they are more protective towards our living environment than the conventional vapor-cycle refrig-erant[1,2].In comparison with gas refrigerators,magnetic refrigerators have a number of advantages,such as high effi-ciency,small volume and ecological cleanliness.The magnetic refrigeration makes u of the cycl
星期一到星期日英文
es of magnetization and demagnetization of a magnetic material,so that the develop-ment of new materials with a giant MCE is strongly desired. The rearch for materials with large magnetocaloric effect is being continued since the discovery of MCE in iron by Warburg about100years ago[1].Some magnetic materials with afirst-order or cond-order transition have attracted much attention, since they have large MCE[2–6].
∗Correspondence address:Department of Physics,Fujian Normal University, Fuzhou350007,China.Tel.:+8659183486160;fax:+86-591-83465313.
E-mail address:
It is known that above15K,Ericsson cycle is ud in magnetic
refrigeration in order to remove the effect of the lattice entropy
[7].Thermodynamic analysis shows that efficient operation of
an ideal Ericsson cycle requires a constant-induced magnetic
entropy change as a function of temperature over the required
operating range[8].If a magnetic working material has a large
magnetic entropy change(| S M|)peak at the transition tem-perature,but falls off rapidly on either side,it is not suitable
for u in devices utilizing the Ericsson cycle[3].Therefore,
it is significant to explore a magnetocaloric material with high
MCE and wide operating temperature span and/or to widen the
operating temperature span for the high MCE material by u
of some novel methods.It was reported that nanoparticles fab-
ricated by rapid solidification or chemical method,may have
relatively wider working temperature span[9].
In our previous work,the MCE of CoMnSb alloy has been
reported as an exploration for uful MCE materials[10].Due
to the large Mn magnetic moments,this kind of rare-earth-free
alloy may be a potential candidate of large MCE materi-
als.In order to extend the operating temperature span and to
enhance the refrigerant capacity of CoMnSb alloy,the nanocrys-
talline CoMnSb alloy has been fabricated by rapid solidification
0925-8388/$–e front matter©2007Elvier B.V.All rights rerved. doi:10.1016/j.jallcom.2007.03.052
74S.Li/Journal of Alloys and Compounds448(2008)73–76 method.In this paper,the effect of crystal size and atomic dis-
order on the MCE of CoMnSb alloy have been investigated by
comparing the MCE of two samples with different crystal sizes.
2.Experimental procedure
Two types of CoMnSb alloys with different crystal sizes have been prepared
by an induction-melting and a melt-spinning method,respectively.The high
purity metals of Co,Mn and Sb were melted in an induction melting furnace
for three times under Ar atmosphere.Then,the ingot was aled in a quartz
tube under vacuum atmosphere(less than3×10−3Pa).The aled ingot was
annealed at873K for30h for eliminating inner stress.The annealed sample
with large crystal size was referred to as sample A,while the other sample with
small crystal size,fabricated by melt-spun part of the ingot at a circumference
speed of30m/s in vacuum,was referred to as sample B.
The magnetic properties of the samples were measured by using vibrating
sample magnetometer(VSM)and superconducting quantum interference device
(SQUID)magnetometer.The microstructure of the materials was characterized
by an X-ray diffractometer(XRD)with Cu K␣radiation.
3.Results and discussion
Fig.1shows the XRD curves for the samples A and B.The
indexes of CoMnSb facets were signed in Fig.1.As illustrated,
both samples are compod of a single pha of CoMnSb.It can
also be en that the diffraction peaks of sample A are great
sharper than tho of sample B,indicating that the crystal size
of sample A is great coarr than that of sample B.The crystal
sizes of samples A and B,calculated by Scherrer equation,are
about120and30nm,respectively.
The temperature dependence of magnetization for both sam-
因为你英文
ples was measured by VSM in the magneticfield of0.2T.Fig.2
shows the M–T relationship curves for samples A and B.It can be
en that:(1)the transition temperature of sample B is slightly
lower than that of sample A.The Curie temperatures are471
金山 翻译and468K for samples A and B,respectively,and(2)the mag-
netization of sample B is slightly lower than that of sample A at
temperature range less than T C.
Fig.3shows a ries of magnetization isotherms measured at
different temperatures in the vicinity of Curie temperature,T C,
with the maximum appliedfield of0.9T.The magnetic entropy
压岁钱英文change,| S M|,was determined as a function of temperature
and
Fig.1.The XRD traces for samples A and
B.
Fig.2.The temperature dependence of magnetization for samples A and B.
magneticfield from isothermal magnetization curves by u of
Maxwell equation:
S M=
H
2
H1
∂M(H,T)
∂T
H
d H(1)
Fig.4shows the plots of| S M|versus temperature of sam-ples A and B for the magneticfield changing from0to0.9T,
respectively.Although,the maximum value of| S M|for sample
B(1.32J/kg K)is smaller than that for sample A(2.06J/kg K),
a broader peak for sample B is obrved in the| S M|–T curves,
indicating that sample B may be superior to the bulk material
for practical application in Ericsson cycle.
In practice,how much heat can be transferred between the cold and hot sinks in one ideal refrigeration cycle is characterized
by the refrigerant capacity[11].The refrigerant capacity,q,is
defined as
q=
T
hot
T cold
S(T,P, H)P, H d T
(2)
Fig.3.The magnetization isotherms measured at different temperatures near T C
for samples A and B.
S.Li/Journal of Alloys and Compounds448(2008)73–76
75
Fig.4.The plots of| S perature for samples A and B with H=0.9T. where T cold and T hot are the temperature of the cold and hot
sinks,respectively.Therefore,when two different materials are
ud in the same refrigeration device,the material with higher
refrigerant capacity is expected to perform better,since it will
support transport of greater amounts of heat in a real cycle,
provided all parameters of a magnetic refrigerator remain the
same.From Fig.4,it can be en that the optimum operating
temperature range of sample B is wider than that of sample A.In
order to accurately evaluate the refrigerant capacity for materials
with different peak site and operating temperature span,we take
the temperature range between the full-width at half maximum
as the calculating temperature span in Eq.(2).The temperature
spans for samples A and B are466–475and452.3–477.4K,
respectively.For the magneticfield changing from0to0.9T,the
refrigerant capacities for samples A and B are15.3and27.3J/kg,
respectively,according to Eq.(2).In addition,even taking the
same temperature span of433–483K,the refrigerant capacity
of sample B(42.83J/kg)is slightly larger than that of sample
A(41.28J/kg).Conquently,sample B is superior to sample A
tmoin practical utilization.Comparing with sample A,the smooth
peak of magnetic entropy change and larger refrigerant capacity
of sample B indicate that sample B is a preferential choice rather
than sample A.
It is believed that magnetic properties of CoMnSb alloy with
crystal size of120nm can be considered as the bulk ones.
The saturation magnetization(M s)of sample A was measured
by SQUID at3T and2K.If,approximately,taking the M s
of94.2533emu/g at2K as the real saturation magnetization
M s(0)of CoMnSb alloy,the calculated saturation magnetiza-
tion is3.978B/f.u.for sample A.This value is very clo to
the theoretical and experimental result of M s∼4.0B/f.u.for CoMnSb alloy[12].While the crystal size is decreasing to small
30nm),the magnetic exchange interaction and mag-
netic anisotropy deviate from the bulk material,giving ri to a
slight reduction of the magnetization and T C[13,14].This phe-
你在眺望着谁
nomenon was widely obrved in nanocrystalline ferromagnetic
systems[15,16].Moreover,the effect of crystal size distribution
on the inner magnetic he saturation magnetiza-
tion,T C)for nanocrystallite materials is great larger than that for
bulk one[17].With raising temperature,the nanocrystallite fer-romagnetic material transforms to the paramagnetic state prior
to the bulk one.As a result,a relatively lower transition tem-
perature in M–T curve for nanocrystalline materials than that
for bulk ones is expected.In addition to that,the distribution
of crystal size for the nano-ferromagnetic materials also gives
ri to afluctuation of T C accordingly.Therefore,the broaden-
ing of| S M|peak and enhancement of refrigerant capacity in sample B in comparison with sample A can be,at least partially,
attributed to the broad T C distribution induced by small size and
its distribution.
Atomic disorder generally occurs in half-Heusler alloys,such
as CoMnSb and NiMnSb[18,19].In our previous work[20],the
atomic disorder of CoMnSb alloys was reduced by annealing the
现在进行时态sample at1323K for50h.A superstructure with low atomic dis-
order was formed for the sample annealed at high temperature.
The broadening of operating temperature span and the enhance-
ment of refrigerant capacity can be attributed to the formation
of the superstructure.Ref.[20]implies that the atomic disorder
in CoMnSb alloy deteriorates the MCE.In this study,sample
A was annealed at873K for30h for the aims of eliminating
inner stress and reducing the atomic disorder.Sample B was not
annealed for avoiding the grain growth.It is well known that the
atomic disorder is stronger for the sample prepared by quench-
ing than the ingot.This can also be demonstrated by the XRD
results.The diffraction peaks of sample B are slightly shifted
to left side in comparison with tho of sample A,suggesting
a relatively larger atomic disorder in quenched sample B than
in sample A.Therefore,the measured refrigerant capacity is
lower than the real value due to the stronger atomic disorder in
sample B.In other words,the effect of crystal size on MCE is
partially reduced by atomic disorder for the quenched sample.
The improvement of MCE in sample B is dominated by crystal
size effect.
4.Conclusion
The magnetic properties and magnetocaloric effect of
CoMnSb alloys with different crystal sizes have been investi-
gated.Comparing to the sample with crystal size larger than
100nm,the refrigerant capacity is enhanced and the operating
temperature span is extended for the sample with crystal size as
small as veral tens of nanometer.The results suggest that
CoMnSb alloy with small crystal size is superior to the larger
one in practical application.
Acknowledgements
This work wasfinancially supported by National Science
Foundation of China(NSFC)for Young Scientists(Grant No.:
10504010),Key Project of Fujian Provincial Department of Sci-
ence&Technology(2006H0018)and Science Foundation of
Fujian Province of China(2006J0152and2005J023). References
[1]E.Warburg,Ann.Phys.(Leipzig)13(1881)141.
[2]V.K.Pecharsky,K.A.Gschneidner Jr.,Phys.Rev.Lett.78(1997)4494.
76S.Li/Journal of Alloys and Compounds448(2008)73–76
[3]B.J.Korte,V.K.Pecharsky,K.A.Gschneidner Jr.,J.Appl.Phys.10(1998)
5677.
[4]F.W.Wang,X.X.Zhang,F.X.Hu,Appl.Phys.Lett.77(2000)1360.
[5]O.Tegus,E.B¨u rck,K.H.J.Buschow,F.R.de Boer,Nature415(2002)
150.
[6]S.D.Li,M.M.Liu,Z.R.Yuan,L.Y.L¨u,Z.C.Zhang,Y.B.Lin,Y.W.Du,J.
Alloys Compd.427(2007)15.
[7]T.Hashimoto,T.Kuzuhara,M.Sahashi,K.Inomata,A.Tomokiyo,H.
Yayama,J.Appl.Phys.62(1987)3873.
[8]A.Smaili,R.Chahine,J.Appl.Phys.81(1997)824.
[9]D.H.Wang,S.L.Tang,H.D.Liu,S.D.Li,J.R.Zhang,Y.W.Du,Jpn.J.
Appl.Phys.40(2001)6815.
[10]S.D.Li,M.M.Liu,Z.G.Huang,F.Xu,W.Q.Zou,F.M.Zhang,Y.W.Du,泸江日语网
确认英文J.Appl.Phys.99(2006)063901.
[11]V.P.Pecharsky,K.A.Gschneidner Jr.,J.Appl.Phys.90(2001)4614.[12]V.Knofontov,G.Melnyk,M.Wojcik,S.Wurmehl,K.Kroth,S.Reiman,
oprah winfreyP.Blaha,C.Felr,Phys.Rev.B74(2006)134426.
[13]R.H.Kodama,S.H.Makhlouf,A.E.Berkowitz,Phys.Rev.Lett.79(1997)
1393.
[14]J.M.D.Coey,Phys.Rev.Lett.27(1971)1140.
[15]T.Sato,T.Iijima,M.Seki,N.Inagaki,J.Magn.Magn.Mater.65(1987)
252.
[16]J.F.L¨o ffler,J.P.Meier,B.Doudin,J.P.Anrmet,W.Wagner,Phys.Rev.
B57(1998)2915.
[17]R.H.Kodama,J.Magn.Magn.Mater.200(1999)359.
[18]C.PalmstrØm,MRS Bull.(October)(2003)725.
[19]K.Kaczmarska,J.Pierre,J.Tobola,R.V.Skolozdra,Phys.Rev.B60(1999)
373.
[20]S.D.Li,Z.R.Yuan,L.Y.L¨u,M.M.Liu,Z.G.Huang,F.M.Zhang,Y.W.Du,
Mater.Sci.Eng.A428(2006)332.
