WP 5

Magnetic characterization


M 5.1. Identification of the types of magnetism in fluorinated samples: anti/ferro/super/magnetism, 2D magnetism

Graphene nanoribbons with zigzag and chiral edges have been suggested to give rise to magnetic ordering. Our method allows one to create a nanoribbon is to obtain spin-polarized regions on graphene by the selective conversion of sp2 carbon state to sp3 one. The residual pi-system regions (also called nanoroads) can be considered as reasonable alternatives to their purely carbonic counterparts. IFGC (intercalated fluorinated graphite compounds) share many common features with graphene: the layer–layer interaction is significantly weakened, and essential graphene physics is preserved. Recently it was confirmed that fluorine atoms randomly attached to graphene basal plane behave as magnetic atoms (develop spin-half paramagnetism – A. Geim). We have shown that the pi-network appearing in the strips between the regular fluorine patterns on the basal plane of the IFGC share similarities with magnetic organic polymers and is responsible for non-Curie paramagnetism, short-range and in certain cases long-range magnetic ordering.

 M 5.2 Implementation of successful tailoring approach in production of graphite-based nanomagnetic structures

A scientific highlight within this WP is the first muon spin resonance measurements on graphene (P2, P7). Due to the relatively large muon penetration depth, gram-scale samples are required, so the first μSR experiment in graphene was performed on chemically-produced material. Unlike other types of nanostructured carbon, here an evident muon spin precession was observed. Although it is usually the fingerprint of magnetic order, here it was demonstrated to originate from muon-hydrogen nuclear dipolar interactions. The μ-H distance, fitted from the observed spin evolution, is dm-H » 1.75 Å which matches the inter-proton distance in a CH2 group. This proves the formation of CHMu (analogous to CH2). This CHMu proved to be stable up to 1250 K where the signal still persists.

Three possible trapping sites are present in graphene: in-plane vacancies, zig-zag and armchair edges. The different dipolar fields acting on the muon in these defects allow to distinguish among these different sites. These results rule out the formation of defect induced ferromagnetic or antiferromagnetic order in chemically synthesized graphene and highlight the role of stacking.

For a quantitative analysis of the magnetic behavior of magnetism dependent on defect concentration, the concentration of defects has to be estimated. In particular the amount of in-plane vacancies on graphene surfaces can be estimated by SQUID magnetometry, which shows that all the samples display a Curie T dependence of magnetization (apart from a sample dependent ferromagnetic contribution consistent with the amount of metal impurities in the precursors). Since the ideal graphene has a diamagnetic behavior, the paramagnetic contribution observed in the temperature dependence of magnetization provide a quantitative estimation of defects concentration. A qualitative estimation can also be obtained by Raman spectroscopy, since the ratio of the D and G band intensities (Id/Ig) provides an indication of the concentration of defects present in graphene. By comparing the two measurements, we found a qualitative agreement between the two techniques for all tested samples, and the one for which the muon spin resonance (µSR) histogram is reported in figure 2 showed 1300(150) ppm paramagnetic defects concentration.

In order to provide an estimate of the hyperfine coupling assuming the presence of long-range magnetic order, we perform first-principles calculations on a model system with muonium bound to carbon atoms in one sublattice of pristine graphene. Such a distribution of chemisorption sites results in ferromagnetic correlation. Electronic states introduced by the covalently bound species in graphene are quasi-localized in nature, that is, their degree of localization shows a logarithmic dependence on the total concentration of defects contributing to the zero-energy band. The calculated hyperfine couplings show a pronounced logarithmic dependence. This estimate shows that local fields of the order of 100 G can be expected for systems with large (x > 0.01) concentration of defects, giving rise to magnetic order. Defect concentrations x = 10−5 − 10−4 correspond to local fields of the order of 1 G. These results allow us to conclude that long-range magnetic order in disordered graphene samples can be reliably measured by means of the µSR technique.

An increase in the impurity concentration in a semiconductor enhances the interaction between impurity centers and leads to the second order phase transition from the insulator to the metal state (MIT). In this case, conductivity at zero temperature can be observed. The problem of such phase transitions is widely discussed in the literature. Apart from the conductivity in the MIT transition region, the magnetic properties of the semiconductor also change. The Curie paramagnetism of noninteracting impurity electrons, which is typical of the insulator state, is transformed into the Pauli paramagnetism of the metal state. Therefore, analysis of the magnetic properties of a semiconductor in the MIT phase transition region involves the estimation of the magnetic susceptibility of the semiconductor. For this purpose, the electron paramagnetic resonance (EPR) method is suitable in principle. However, some problems associated with a considerable change in the conductivity of the sample appear in this case. We have developed the method for determining the magnetic susceptibility for poor conductors using Ge : As as an example. The method consists of three parts. 1. The use of EPR resonance with two antinodes of the magnetic field with the experimental sample at one antinode and the standard sample at the other. 2. Measurement of the temperature dependence of the resistivity of the sample (poor conductor) to take into account the nonuniform distribution of the microwave field in it. 3. Double integration of the measured positive part of the derivative of the Dyson resonance line. The procedure gives the value of spin density with an error not exceeding 15%.

Impurity paramagnetic susceptibility of n-Ge has been measured in the vicinity of the insulator– metal phase transition by two techniques: ESR and SQUID. A comparison of these methods demonstrated that their results qualitatively coincide at low temperatures, but differ at high temperatures. The advantage of the ESR technique is in its substantially higher relative precision, but SQUID makes it possible to obtain on principle absolute values of the magnetic susceptibility. It was found that, in the present-day state of development, the SQUID method has a substantially poorer precision, compared with ESR.

Microscopic studies of magnetic samples revealed an intriguing feature of the nanopatterned sample bombarded by medium energy ions. Atomic force microscopy (AFM) shows that the damages are not random but form intricate patterns and motifs appearing on the HOPG (highly oriented pyrolytic graphite) surface which are apparently related to the intrinsic texture of graphite. For all radiation doses, the surface features display long tracks with the hillocks grown on them. These tracks sometimes cross at the angles close to 60 or 90°, suggesting that the surface patterns reflect the intrinsic mosaicity of graphite. The surface of the pristine (non-bombarded) samples is flat, so the surface features reveal hidden grain boundaries (buried underneath) propagating along the c axis of the graphite crystal.

The localization of defects at the grain boundaries suggests that mechanism of magnetism for pristine and irradiated graphite has a common origin. Intrinsic ferromagnetism of graphite is related to the presence of grain boundaries which can be considered as 2D periodic networks of point defects measurements on irradiated samples.

The MFM signals reveal as the dark and bright areas, but they contain apart from magnetic also electrostatic contributions and also artefacts connected with sharp changes in topography on the defects. In order to ensure that the MFM image contrast is caused by the HOPG magnetization and not due to topographical artifacts, the tip magnetization was reversed by 180° and a MFM scan was obtained from the same area. The non-magnetic contributions are reliably withdrawn by making the measurements at opposite tip magnetizations. Somewhat unexpectedly, only linear defects change their magnetic response from dark to bright whereas round-shape defects looked bright at both tip orientations. This observation supports the model that magnetic areas are located at the grain boundaries of HOPG.

M 5.3 Description of magnetic behaviour of transmutation doped IV group elements near the metal-insulator transition,

DMS (diluted magnetic semiconductor) ferromagnetism with Si or Ge as the host semiconductor is obviously attractive because of their greater compatibility with existing silicon based technology. In Si Mn impurities favor interstitial position which significantly complicates the synthesis of a uniform DMS system. While strong hybridization of Mn 3d states with s,p states in Si occurs if Mn enter substitutional (MnSi) positions as acceptors. Binary compounds of 3d metals with Si are weak itinerant magnets of helicoidal type with Curie temperature <50K (no hysteresis loop). A small deviation of the stoihiometry results in drastic changes of magnetic properties. Tc reaches more than 300 K. Carriers are spin polarized. We explain experimental results within the model of exchange between local magnetic moments through the spin fluctuations. Properties of these structures strongly depend on substrate.

We developed a theory which describes the tunnel coupling between a continuum of states in the quantum well (QW) and an impurity bound state located outside of the QW. We utilized the well known Fano approach for calculation of the matrix elements for the direct interband optical transitions in the QW. For such transitions the tunnel coupling of the 2D QW states with the impurity states leads to the drop of the luminescence spectral density at the frequency corresponding to the configuration resonance. This modification of the spectra leads to an integral circular polarization of the light emitted from the QW provided the bound hole state is split in the projection of the hole angular momentum. The key advantage of the approach used in the present study is that the unknown eigenfunctions of the system are expressed through those of the uncoupled states. Any effects on the localized state can be translated into effects for the whole coupled system. For this reason it is capable of describing other effects expected in such systems like anisotropy of the holes g-factor in the QW induced by the paramagnetic impurity or the indirect exchange interaction between the bound states provided by the 2D free carriers in the QW. The presented theory treats the configuration interaction between a continuum of states in the QW and a paramagnetic impurity located outside of the QW.