Open Journal of Microphysics, 2011, 1, 19-22
doi:10.4236/ojm.2011.12003 Published Online August 2011 (
Copyright © 2011 SciRes. OJM
Structural Stabilizing Effect of Zn Substitution on
MnAl and Its Magnetic Properties
H. X. Wang1, P. Z. Si1,*, W. Jiang1, J. G. Lee2, C. J. Choi2, J. J. Liu3,
Q. Wu1, M. Zhong1, H. L. Ge1
1School of Materials Science and Engineering, China Jiliang University,
Hangzhou, China
2Korea Institute of Materials Science, Changwon, Gyeongnam, R. Korea
3Faculty of Materials Science and Chemical Engineering,
Ningbo University, Ningbo, China
May 17, 2011; revised June 28, 2011; accepted July 12, 2011
The effect of Zn substitution on the structure and magnetic properties of τ-MnAl has been investigated sys-
tematically. It is found that Zn substitution can stabilize the structure of τ-phase and thus a significant
amount of τ-phase can be produced. Zn increases the coercivity and saturation magnetization of the τ-MnAl
but reduces the Currie temperature. However, excess Zn is detrimental to the magnetic parameters. The op-
timum magnetic performance was found in samples with Zn substitution to 2.9% Mn atoms and 3.5% Al
atoms, respectively.
Keywords: MnAl, Magnetic Property, Zn Substitution
1. Introduction
The development of rare earth free permanent magnets is
becoming more and more important with increasing cost
and decreasing reserve of rare earth resources. Ferro-
magnetic τ-MnAl, as a low cost rare-earth-free magnet,
has received continuous attention since its discovery in
1958 for its superior magnetic performance in compari-
son with magnetically hard ferrites and Alnicos [1-3].
The τ-MnAl has a Currie temperature of 661 K, a man-
ganese moment of 1.94 µB, and an enhanced magnetic
anisotropy. The structure of ferromagnetic τ-MnAl is
tetragonal, which can be interpreted to arise from a non-
magnetic cubic structure by two subsequent steps,
namely an electronic distortion due to spin polarization
followed by a structural distortion into the tetragonal
system [4]. Since the τ-MnAl is metastable, it is difficult
to obtain pure phase of τ-MnAl, which usually is pro-
duced by a rapid quench of the high temperature ε-phase
followed by isothermal annealing at temperatures be-
tween 400˚C and 700˚C for a short while, or by cooling
the τ-phase at a rate of 10˚C/min [2]. Long time anneal-
ing would result in decomposition of the τ-phase to the
equilibrium γ-phase and β-phase. In order to stabilize the
τ-MnAl, the doping effect of several elements, including
Ti, Cu, Ni, C, B, etc., has also been studied [2,5]. It was
found that the addition of carbon to MnAl can stabilize
the τ-phase [6,7]. Reports on the partial substitution of
Mn in MnAl by Ni or Co could also be found [8,9]. The
effect of introducing other elements into MnAl system
has not been studied. The purpose of this work is to in-
vestigate the effect of doping Zn on the structure and
magnetic properties of the Mn-Al system.
2. Experiments
High purity (>99.9%) Mn, Al and Zn in nominal compo-
sition of Mn54Al46-δZnδ (δ = 0, 1.6, 3.3, 4.7, 5), Mn53.5
Zn0.5Al46, Mn53.4Zn1.6Al46, Mn51.8Zn3.2Al46, and Mn49Zn5
Al46 were melted by using induction meting in an argon
atmosphere. Then the melt was quenched in water. After
that the quenched ingots were annealed at 420˚C for 1 h
in vacuum, respectively. The structure of the samples
was determined by using a powder x-ray diffractometer
(XRD) with Cu Kα radiation while the magnetic proper-
ties were measured by using a Vibrating Sample Magne-
tometer (VSM) in fields up to 1.5 T. Thermal analysis on
the samples was carried out in argon atmosphere with a
20 H. X. WANG ET AL.
temperature sweep rate of 20˚C/min.
3. Results and Discussion
The x-ray powder diffraction patterns of the Mn54Al46-δ
Znδ (δ = 0, 1.6, 3.3, 5) water quenched samples and their
counterparts after 420˚C and one hour vacuum heat-
treatment are shown in Figure 1, respectively. The XRD
patterns for the as-quenched Mn54Al46, as shown in Fig-
ure 1(a), could be mainly indexed with orthorhombic
ε-MnAl phase, which is stable at temperatures above
870˚C and is maintained to room temperature during
quench. The weak broadened peak in the vicinity of 43˚
indicates the presence of γ and/or β-phase as minor phase
in the sample. Figure 1(b) shows that the heat-treatment
to Mn54Al46 induces the transformation of ε-phase to a
more stable Al-rich γ-phase and Mn-rich β-phase. The
diffraction peaks for meta-stable τ-phase could also be
found but are very weak. Strong diffraction peaks appear
in both water-quenched and heat-treated Mn54Al44.4Zn1.6,
as shown in Figures 1(c) and (d), indicating stabilizing
effect of Zn on the structure of τ-phase. Trace amount of
ε-phase, β-phase and γ-phase are detected in the as-
quenched Mn54Al44.4Zn1.6. After heat-treatment, the ε-
phase disappears while the diffraction intensity of β- and
γ-phase made almost no change, indicating a phase
transformation of ε-phase to τ-phase during heat-treat-
ment. For Mn54Al42.7Zn3.3, the τ-phase presents in the
as-quenched samples but disappears after heat-treatment,
as seen in Figures 1 (e) and (f). These phenomena indi-
cate that the τ-phase formed in Mn54Al44.4Zn1.6 is more
stable than that formed in Mn54Al42.7Zn3.3. With increas-
ing Zn content, the τ-phase, β-phase and γ-phase coexists
in both water-quenched and heat-treated Mn54Al41Zn5
samples as seen in Figures 1(g) and (h).
It should be noted that for sample with δ = 1.6, τ-phase
is the major phase while for other samples β-phase is the
major one. This result indicates that substitution of a
certain amount of Zn to Al is beneficial for the formation
of τ-phase. We speculate that atomic size may play an
important rule in this process. It is known that the atomic
size of Zn is slightly smaller than that of Al. Since the
minimum internal energy in MnAl occurs very close to
c/a = 1 (c and a are the lattice parameters), thus the
tetragonal τ-phase is reported to be meta-stable [4].
When a small number of Al atoms were substituted by
smaller Zn, a local lattice distortion that make tetragonal
τ-phase more stable might occur to maintain internal
energy minimum. However, for samples with increasing
substitution δ 3.3, β-phase rather than ε or τ-phase
formed as major phase in water quenched samples. Since
β-phase is very stable and thus the heat-treatment has
little effect on the structure of these samples.
Figure 2 shows the effect of Zn substitution to Al on
the magnetic properties of Mn54Al46-δZnδ (δ = 0, 1.6, 3.3,
5). For heat-treated samples, the coercivity and the satu-
ration magnetization increase first and then decrease with
increasing Zn substitution to Al. A maximum coercivity
of 0.157 T was observed in heat-treated Mn54Al44.4Zn1.6.
The 0.157 T coercivity in our Mn-Al-Zn system is lower
than the 0.34 T coercivity as reported in the Mn-Al-C
samples [7]. Since the parameter of coercivity is very
sensitive to microstructures, we tend to believe that cer-
tain partial substitution of Zn to Al might result in the
formation of localized lattice defects that hinder mag-
netization reversal. The increasing saturation magnetiza-
tion with Zn substitution was ascribed to the structural
stabilizing effect of Zn on ferromagnetic τ-MnAl and its
presence as major phase in the samples. Excess Zn
would result in the formation of more β-phase and
γ-phase that are detrimental to the magnetic performance.
In comparison with heat-treated samples, most wa-
ter-quenched samples exhibit a lower coercivity except
Mn54Al42.7Zn3.3, as seen in Figure 2. However, for sam-
ples with δ 3.3, the effect of heat-treatment on the coer-
Figure 1. XRD patterns for (a) Mn54Al46, (b) Heat-treated
Mn54Al46, (c) Mn54Al44.4Zn1.6, (d) Heat-treated Mn54Al44.4
Zn1.6, (e) Mn54Al42.7Zn3.3, (f) Heat-treated Mn54Al42.7Zn3.3, (g)
Mn54Al41Zn5, and (h) Heat-treated Mn54Al41Zn5. The heat-
treatment for sample (b), (d), (f), and (h) was carried out in
vacuum under 420˚C for one hour.
Figure 2. The effect of Zn substitution to Al on the coerciv-
ity and saturation magnetization of Mn54Al46-δZnδ (δ = 0,
1.6, 3.3, 5).
Copyright © 2011 SciRes. OJM
civity is decreasing due to the presence of stable β-phase
and γ-phase in the samples. Heat-treatment has a strong
effect on the magnetic properties of Mn54Al44.4Zn1.6.
Figure 3 shows the XRD patterns of the water-
quenched and the heat-treated ingots of Mn53.5Zn0.5Al46,
Mn53.4Zn1.6Al46, Mn51.8Zn3.2Al46, and Mn49Zn5Al46. It is
interesting that τ-phase formed in all water-quenched
samples, indicating a strong structural stabilizing effect
of Zn substitution to Mn on τ-MnAl. However, the
β-phase and γ-phase were also formed in all samples
during water-quench. For Mn53.5Zn0.5Al46, as seen in
Figures 3(a) and (b), by comparing the relative diffrac-
tion intensity of τ-phase and β-/γ-phase, we noticed that
the amount of τ-phase increased after heat-treatment.
Figures 3(c) and (d) shows that the heat-treatment has
little effect on the structure of Mn53.4Zn1.6Al46. For
Mn51.8Zn3.2Al46, ε-phase was detected in water-quenched
sample while it disappears after heat-treatment, as shown
in Figures 3(e) and (f). The intensity of τ-phase peaks
were enhanced after heat-treatment, indicating a ε τ
transformation in this process. Figures 3(g) and (h) shows
that both water-quenched and heat-treated Mn49Zn5Al46
were composed of τ-phase, β-phase and γ-phase, while
heat-treatment had almost no effect on the structure.
Figure 3 shows that no pure τ-phase could be obtained in
Zn substituting Mn process, for comparison Figures 1(c)
and (d) shows that pure τ-phase could be obtained in Zn
substituting Al process.
The effect of Zn substitution to Mn on the magnetic
properties of Mn54Al46, Mn53.5Zn0.5Al46, Mn53.4Zn1.6Al46,
Mn51.8Zn3.2Al46 and Mn49Zn5Al46 are shown in Figure 4.
Both the saturation magnetization and the coercivity of
the heat-treated samples increase first and then decrease
with increasing Zn content. The maximum coercivity
was observed in Mn53.5Zn0.5Al46 while the maximum
saturation magnetization was observed in Mn53.4Zn1.6Al46.
The enhancement of the magnetic performance for sam-
ples with lower Zn substitution indicates that Zn is bene-
ficial for the formation of τ-phase. However, Figure 4
also shows that higher Zn substitution is detrimental to
the magnetic properties, suggesting a limited effect of Zn
addition. It is interesting to note that heat-treatment in-
creases the coercivity when Zn substitution is low and
decreases the coercivity when Zn content is high.
Figure 5 plots the M-T and ΔT-T curves for the sam-
ples. The Currie temperature of Mn54Al42.7Zn3.3, Mn54
Al44.4Zn1.6, Mn53.5Zn0.5Al46 determined by using dM/dT
method is in the vicinity of 371K-373 K, which is lower
than the 388 K reported for τ-MnAl[3]. The result indi-
cates that Zn substitution reduces the Currie temperature
of τ-MnAl. The endothermal peak observed in Figure
5(d) was ascribed to the ferromagnetic-paramagnetic
transition of τ-MnAl in Mn54Al42.7Zn3.3.
Figure 3. XRD patterns for (a) Mn53.5Zn0.5Al46, (b) Heat-
treated Mn53.5Zn0.5Al46, (c) Mn53.4Zn1.6Al46, (d) Heat-treated
Mn53.4Zn1.6Al46, (e) Mn51.8Zn3.2Al46, (f) Heat-treated Mn51.8
Zn3.2Al46, (g) Mn49Zn5Al46, and (h) Heat-treated Mn49Zn5Al46.
The heat-treatment for sample (b), (d), (f), and (h) was car-
ried out in vacuum under 420˚C for one hour.
Figure 4. The effect of Zn substitution to Mn on the mag-
netic properties of Mn54Al46, Mn53.5Zn0.5Al46, Mn53.4Zn1.6
Al46, Mn51.8Zn3.2Al46 and Mn49Zn 5Al46.
Figure 5. Temperature dependence of magnetization of
heat-treated samples (a) Mn54Al42.7Zn3.3, (b) Mn54Al44.4Zn1.6,
(c) Mn53.5Zn0.5Al46 in an applied field of 0.05 T and a sweep
rate of 5˚C/min. Line (d) plots the T-ΔT data from thermal
analysis on Mn54Al42.7Zn3.3.
4. Conclusions
Zn substitution can stabilize the structure of τ-phase. Zn
increases the coercivity and saturation magnetization of
the τ-MnAl but reduces the Currie temperature. It should
be noted that excess Zn is detrimental to the magnetic
Copyright © 2011 SciRes. OJM
Copyright © 2011 SciRes. OJM
parameters. The optimum magnetic performance was
found in samples with Zn substitution to 2.9% Mn atoms
and 3.5% Al atoms, respectively.
5. Acknowledgements
This work was supported by the Natural Science Foun-
dation of China (Nos. 10874159, 11074227, 50801039),
Zhejiang Provincial Natural Science Foundation of
China (No. R6110362), and Fundamental R&D Program
for Core Technology of Materials funded by the Ministry
of Knowledge Economy, R. Korea.
6. References
[1] H. Kono, “On the Ferromagnetic Phase in Manga-
nese-Aluminum System,” Journal of the Physical Society
of Japan, Vol. 13, No. 12, 1958, pp. 1444-1451.
[2] Q. Zeng, I. Baker, J. B. Cui and Z. C. Yan, “Structural
and Magnetic Properties of Nanostructured Mn-Al-C
Magnetic Materials,” Journal of Magnetism and Mag-
netic Materials, Vol. 308, No. 2, 2007, pp. 214-226.
[3] J. H. Park, Y. K. Hong, S. Bae, J. J. Lee, J. Jalli, G. S.
Abo, N. Neveu, S. G. Kim, C. J. Choi and J. G. Lee,
“Saturation Magnetization and Crystalline Anisotropy
Calculations for MnAl Permanent Magnet,” Journal of
Applied Physics, Vol. 107, No. 9, 2010, pp. 09A731.
[4] Y. Kurtulus and R. Dronskowski, “Electronic Structure,
Chemical Bonding, and Spin Polarization in Ferromag-
netic MnAl,” Journal of Solid State Chemistry, Vol. 176,
No. 2, 2003, pp. 390-399.
[5] Y. Sakka, M. Nakamura and K. Koshimoto, “Rapid
Quenching and Properties of Hard Magnetic Materials in
MnAl-X (X=Ti, Cu, Ni, C, B) Systems,” Journal of Ma-
terials Science, Vol. 24, No. 12, 1989, PP. 4331-4338
[6] W. H. Dreizler and A. Menth, “Transformation Kinetics
of the Ferromagnetic Alloy Mn-Al-C,” IEEE Transac-
tions on Magnetics, Vol. 16, No. 3, 1980, PP. 534-536.
[7] D. C. Crew, P. G. McCormick and R. Street, “MnAl and
MnAlC Permanent Magnets Produced by Mechanical
Alloying,” Scripta Metallurgica et Materialia, Vol. 32,
No. 3, 1995, PP. 315-318.
[8] M. Mat, A. Mor and N. Koh, “Crystal Structure and
Magnetic Properties of Mn-Al-Ni Ferromagnetic Films,”
IEEE Translation Journal on Magnetics in Japan, Vol. 6,
No. 2, 1991, pp. 134-140
[9] R. Kainuma, M. Ise, K. Ishikawa, I. Ohnuma and K.
Ishida, “Phase Equilibria and Stability of the B2 Phase in
the Ni–Mn–Al and Co–Mn–Al systems,” Journal of Al-
loys and Compounds, Vol. 269, 1998, pp. 173-180.