Spin Seebeck effect has been investigated in Pt/ε-Fe2O3 bilayers. The ε-Fe2O3 thin layer with 40 − 70 nm thickness were deposited by a spin-coating method on Y:ZrO2(100) substrates. The prepared layers are highly oriented with the easy magnetic a-axis parallel to the film surface. The magnetic hysteresis loops measured at room temperature with magnetic field parallel to the layer exhibit coercive fields up to 11.6 kOe, which is so far the highest value measured for ε-Fe2O3 thin layer samples. The shape of the spin Seebeck hysteresis loops is similar to the shape of magnetization for single phase layers with coercive field around 10 kOe. In some prepared layers a small amount of secondary soft ferrimagnetic phase is revealed by a constricted shape of magnetization loops, in contrast to spin Seebeck loops, where no constriction is observed. A difference in encountered in the case of layers with a small amount (1 − 2 volume%) of secondary soft ferrimagnetic phase, which is revealed by a constricted shape of magnetization loops, in contrast to spin Seebeck loops, where no constriction is observed.

Our study targets some of the long-standing questions concerning the somewhat mysterious properties of chalcopyrite CuFeS 2 . We show that defect chemistry in connection with charge transfer within the structure is responsible for the unusual electronic and magnetic properties of CuFeS 2 . Specifically, our model addresses weak ferromagnetism and the high mobility of carriers on the background of a rigid antiferromagnetic structure. We show that defect structure can, counterintuitively, boost the mobility of free carriers due to defect-modified charge transfer. Further, the defect-modified charge transfer induces the weak ferromagnetism both in the Cu-and Fe-sublattice. This new view opens up space for further investigations and applications of charge transfer compounds. 45 d-band dispersion and treat the Fe ions as impurities within an 46anion-hosting band structure. 18,19,23 47 The exact valence state of copper ions remains in debate, 48 although it is clear that number of d electrons is slightly lower 49 than 10, which would correspond to the Cu +1 state. 22,29,30 The 50 electrical transport complexity is documented by a rather wide 51 range of reported band-gap values derived from various 52 experiments (i.e., 0.33 eV, 31 0.53 eV, 3 and 2.6 eV 5 ). 53 Furthermore, the electron mobility spans a rather wide range 54 from 10 −3 to 10 2 cm 2 V −1 s −1 depending on the temperature 55 and doping and Fermi level (E F ). Thus, the nature of the 56 electrical transport can be described within various models 57 starting with hopping transport in the low-temperature region 58 and, inevitably, ending with band transport at higher 59 temperatures. Some peculiar interactions have been included 60 to account for the "mysterious behaviour" 23 of chalcopyrite 61 including negative charge transfer (CT) energy. The 62 contribution of CT-driven hybridization between the Fe3d 63 and S3p orbitals has also been discussed previously. 19 64 Furthermore, band calculations can shed more light on the 65 delocalized Fe3d electrons, 26,32,33 thus attributing markedly to 66 a consistent picture as for the spectroscopic properties and 67 AFM ordering. However, the magnitude of the free-carrier

No abstract

Rod-like particles of silica-coated maghemite: synthesis via akaganeite, characterization and biological properties,