9 Magnetism of Individual Nanoparticles Probed by X-Ray …
231
These experiments directly demonstrate that the enhanced magnetic energy barriers in the magnetically blocked iron nanoparticles are not related with the free surface
of the pure, metallic iron nanoparticles. The findings also confirm that a Néel-type surface anisotropy is not sufficient to explain the enhanced magnetic anisotropy observed
in the magnetically blocked iron nanoparticles even when taking non-collinear surface spin configurations into account, as discussed in [5, 38, 55]. Moreover, the oxidation experiments confirm that enhanced magnetic energy barriers can be observed
in partially oxidized iron nanoparticles and that interactions of the metallic core with
the oxide layer are not necessary as proposed in [23], and, as seen in the in situ
oxidation experiments, also not sufficient to establish magnetically blocked states in
initially superparamagnetic nanoparticles. The in situ oxidation investigations show
that the origin for the large magnetic anisotropy barriers is indeed very local and
resides either deep in the particle volume or at the interface with the substrate, which
is most likely not affected by the oxidation.
9.3.3 Iron Nanoparticles Deposited on Different
Substrates—The Role of the Interface
In the previous sections it was shown that gas-phase grown iron nanoparticles with
sizes ranging from 8 to 20 nm deposited on silicon substrates can exist in both a
superparamagnetic state, as expected from bulk and surface contributions to the total
magnetic anisotropy, and in a metastable state, with uniaxial anisotropy and significantly enhanced magnetic energy barriers when compared to the bulk properties
[38]. It was further shown that the enhanced magnetic stability of the magnetically
blocked nanoparticles is not related to their free surface [53]. In this section, we
discuss the role of the interface to the substrate for the magnetic anisotropy of the
iron nanopartices. The interface between different magnetic and non-magnetic materials is well known to modify magnetic phenomena in thin films and multilayers and
can be used to control magnetism at the nanoscale [56]. Interface effects on the
magnetic anisotropy of supported clusters and nanoparticles were also frequently
reported, see for instance [57, 58]. Ensemble measurements performed under UHV
conditions revealed, for instance, that iron nanoparticles deposited on single crystalline copper surfaces can have strongly enhanced magnetic anisotropy energies
[28], which can even give rise to magnetically blocked states at room temperature
[45], while deposition on W(110) leads only to a superparamagnetic state at room
temperature with the magnetic fluctuations occuring mainly in the surface plane [59].
However, the sample preparation conditions varied in the different experiments and,
therefore, it remained unclear whether the observed effects are indeed due to the
interface or reflect intrinsic properties of the nanoparticles.
This issue was addressed in an XPEEM study, where iron nanoparticles with
sizes in the range from 8 to 20 nm were deposited on three single crystalline substrates, namely Cu(001), NiO(110), and W(110) in addition to the Si substrates [60].
231
These experiments directly demonstrate that the enhanced magnetic energy barriers in the magnetically blocked iron nanoparticles are not related with the free surface
of the pure, metallic iron nanoparticles. The findings also confirm that a Néel-type surface anisotropy is not sufficient to explain the enhanced magnetic anisotropy observed
in the magnetically blocked iron nanoparticles even when taking non-collinear surface spin configurations into account, as discussed in [5, 38, 55]. Moreover, the oxidation experiments confirm that enhanced magnetic energy barriers can be observed
in partially oxidized iron nanoparticles and that interactions of the metallic core with
the oxide layer are not necessary as proposed in [23], and, as seen in the in situ
oxidation experiments, also not sufficient to establish magnetically blocked states in
initially superparamagnetic nanoparticles. The in situ oxidation investigations show
that the origin for the large magnetic anisotropy barriers is indeed very local and
resides either deep in the particle volume or at the interface with the substrate, which
is most likely not affected by the oxidation.
9.3.3 Iron Nanoparticles Deposited on Different
Substrates—The Role of the Interface
In the previous sections it was shown that gas-phase grown iron nanoparticles with
sizes ranging from 8 to 20 nm deposited on silicon substrates can exist in both a
superparamagnetic state, as expected from bulk and surface contributions to the total
magnetic anisotropy, and in a metastable state, with uniaxial anisotropy and significantly enhanced magnetic energy barriers when compared to the bulk properties
[38]. It was further shown that the enhanced magnetic stability of the magnetically
blocked nanoparticles is not related to their free surface [53]. In this section, we
discuss the role of the interface to the substrate for the magnetic anisotropy of the
iron nanopartices. The interface between different magnetic and non-magnetic materials is well known to modify magnetic phenomena in thin films and multilayers and
can be used to control magnetism at the nanoscale [56]. Interface effects on the
magnetic anisotropy of supported clusters and nanoparticles were also frequently
reported, see for instance [57, 58]. Ensemble measurements performed under UHV
conditions revealed, for instance, that iron nanoparticles deposited on single crystalline copper surfaces can have strongly enhanced magnetic anisotropy energies
[28], which can even give rise to magnetically blocked states at room temperature
[45], while deposition on W(110) leads only to a superparamagnetic state at room
temperature with the magnetic fluctuations occuring mainly in the surface plane [59].
However, the sample preparation conditions varied in the different experiments and,
therefore, it remained unclear whether the observed effects are indeed due to the
interface or reflect intrinsic properties of the nanoparticles.
This issue was addressed in an XPEEM study, where iron nanoparticles with
sizes in the range from 8 to 20 nm were deposited on three single crystalline substrates, namely Cu(001), NiO(110), and W(110) in addition to the Si substrates [60].