406
I. de Moraes and N. M. Dempsey
over the last 20 years. This is because, despite extensive searches for ternary and
quaternary phases containing RE and TM elements, no new magnetic phases having
intrinsic properties better than those of Nd 2 Fe 14 B have been discovered.
The maximum operating temperature of NdFeB-based magnets in hybrid electric
vehicles and wind turbines reaches as high as 180°C. The coercivity of RE-TM
magnets decreases with increasing temperature, and to maintain sufficient coercivity
at elevated temperature, magnet manufacturers increased the anisotropy of the main
hard magnetic phase (R 2 Fe 14 B) through a partial substitution of Nd, which is a
light rare earth (LRE) element, by a heavy rare earth (HRE) such as Dy or Tb.
One drawback of substituting HRE for LRE is that the remanent magnetisation of
the magnet is reduced, because while the moments on LRE atoms align parallel
to the moments on the Fe atoms in the R 2 Fe 14 B phase, those of HRE align antiparallel. Another major concern with HRE substitution is that HREs are much less
abundant, and thus much more expensive, than LREs. The growing demand for
certain LREs and HREs for use in magnets and in other applications (e.g. lighting
and visual displays, catalysers…) coupled with limited mining and separation of
REs, essentially concentrated in China, led to the so-called RE-crisis. The cost and
supply risks associated with REs have been a driving factor in defining the main
directions in permanent magnet research today (Fig. 17.3). The rest of this chapter
deals with one specific research direction, namely the development of hard—soft
magnetic nanocomposites.
17.1.2 Hard—Soft Magnetic Nanocomposites
Following an experimental report by Coehoorn on the attainment of M r /M s values of
over 0.5 and coercivity of 0.3T in isotropic Fe 3 B–Fe–Nd 2 Fe 14 B magnets containing
just 15 vol.% hard magnetic phase [3], Kneller and Hawig proposed an approach to
significantly increase the energy product of permanent magnets. Their idea was to
produce a nanostructured composite material that combines a hard magnetic phase
exchange coupled to a high magnetisation soft phase [4], as illustrated in Fig. 17.4.
They showed that the length scale of the soft phase should be roughly of the order
of the domain wall width of the hard phase, i.e. a few nanometres. The application
of an external field to an exchange coupled hard-soft nanocomposite will produce a
twisted spin state in the soft material, and upon removal of the external field, the twist
will spring back to the original state of full remanence, giving rise to the term “spring
magnet” (Fig. 17.4—right). The concept was further developed by Skomski and Coey,
who predicted that it may be possible to achieve an energy product in excess of a
megajoule per cubic metre in an optimised textured Sm 2 Fe 17 N 3 /Fe 65 Co 35 hard/soft
nanocomposite [5]. Skomski recently revisited the idea of predicting the ideal nanostructure of aligned hard/soft nanocomposites and found that soft-in-hard structures
should have higher coercivity than hard-in-soft structures and that ideal nanostructures should be made of soft cubes or long soft rods [6]. The intrinsic magnetic
properties of candidate hard and soft magnetic materials are listed in Table 17.1.
I. de Moraes and N. M. Dempsey
over the last 20 years. This is because, despite extensive searches for ternary and
quaternary phases containing RE and TM elements, no new magnetic phases having
intrinsic properties better than those of Nd 2 Fe 14 B have been discovered.
The maximum operating temperature of NdFeB-based magnets in hybrid electric
vehicles and wind turbines reaches as high as 180°C. The coercivity of RE-TM
magnets decreases with increasing temperature, and to maintain sufficient coercivity
at elevated temperature, magnet manufacturers increased the anisotropy of the main
hard magnetic phase (R 2 Fe 14 B) through a partial substitution of Nd, which is a
light rare earth (LRE) element, by a heavy rare earth (HRE) such as Dy or Tb.
One drawback of substituting HRE for LRE is that the remanent magnetisation of
the magnet is reduced, because while the moments on LRE atoms align parallel
to the moments on the Fe atoms in the R 2 Fe 14 B phase, those of HRE align antiparallel. Another major concern with HRE substitution is that HREs are much less
abundant, and thus much more expensive, than LREs. The growing demand for
certain LREs and HREs for use in magnets and in other applications (e.g. lighting
and visual displays, catalysers…) coupled with limited mining and separation of
REs, essentially concentrated in China, led to the so-called RE-crisis. The cost and
supply risks associated with REs have been a driving factor in defining the main
directions in permanent magnet research today (Fig. 17.3). The rest of this chapter
deals with one specific research direction, namely the development of hard—soft
magnetic nanocomposites.
17.1.2 Hard—Soft Magnetic Nanocomposites
Following an experimental report by Coehoorn on the attainment of M r /M s values of
over 0.5 and coercivity of 0.3T in isotropic Fe 3 B–Fe–Nd 2 Fe 14 B magnets containing
just 15 vol.% hard magnetic phase [3], Kneller and Hawig proposed an approach to
significantly increase the energy product of permanent magnets. Their idea was to
produce a nanostructured composite material that combines a hard magnetic phase
exchange coupled to a high magnetisation soft phase [4], as illustrated in Fig. 17.4.
They showed that the length scale of the soft phase should be roughly of the order
of the domain wall width of the hard phase, i.e. a few nanometres. The application
of an external field to an exchange coupled hard-soft nanocomposite will produce a
twisted spin state in the soft material, and upon removal of the external field, the twist
will spring back to the original state of full remanence, giving rise to the term “spring
magnet” (Fig. 17.4—right). The concept was further developed by Skomski and Coey,
who predicted that it may be possible to achieve an energy product in excess of a
megajoule per cubic metre in an optimised textured Sm 2 Fe 17 N 3 /Fe 65 Co 35 hard/soft
nanocomposite [5]. Skomski recently revisited the idea of predicting the ideal nanostructure of aligned hard/soft nanocomposites and found that soft-in-hard structures
should have higher coercivity than hard-in-soft structures and that ideal nanostructures should be made of soft cubes or long soft rods [6]. The intrinsic magnetic
properties of candidate hard and soft magnetic materials are listed in Table 17.1.