A priori, the magnetoresistance (MR) of a spin valve is positive, which simply means that the resistance of the AP state is higher than the resistance of the P state. However, many groups reported negative MR, that is, the resistance of the AP state is smaller than the resistance of the P state (see the figure above). This MR sign inversion is possible because at one of the interfaces the majority carriers become the minority carriers and vice versa. In the parallel state the electrodes are aligned in the same direction, but when carriers reach the second electrode they encounter high scattering. Therefore, the state of the resistance is high. An analogous situation is encountered in the AP state where the magnetization of the electrodes is antiparallel but the carriers encounter low scattering when reaching the spin detector.
One of the possibilities to achieve negative MR is by antiferromagnetic (AF) coupling since an AF layer will invert the majority and minority spins. For instance, negative MR has been measured in MTJs where Ciudad and coworkers demonstrated that the inversion in their devices was due to a chemical reaction caused by the LiF. An CoF2 antiferromagnetic layer was produced being the responsible of the MR sign inversion.
Molecules are also capable to invert the sign of the magnetoresistance without reacting chemically with the electrodes but by orbital overlap or hybridization. In their case, the spin filtering is due to the fact that the localized molecular levels broaden when put in contact with metals, and the broadening is different for spin-up and spindown branches when the metal is ferromagnetic (see Figure above). This situation generates preferential channels near the Fermi level, allowing passing more spin-up or spin-down carriers. Furthermore, the first molecular layer makes an additional selection of the spin polarization, changing the spin polarization coming from FM.
Thus, the spin filtering has to be studied individually for the interface given a couple of materials. The direct demonstration of spin filtering was given by Atodiresei et al. at the Fe / H2Pc interface. Nevertheless, the model was firstly proposed by Barraud et al.in 2010. Later, this ability of a metal/molecule interface to invert the sign of the magnetoresistance was named spinterface by Sanvito.
REFERENCES
[1] C. Barraud, P. Seneor, R. Mattana, S. Fusil, K. Bouzehouane, C. Deranlot, P. Graziosi, L. Hueso, I. Bergenti, V. Dediu, et al. “Unravelling the role of the interface for spin injection into organic semiconductors”. Nature Physics 6.8 (2010), pp. 615–620.
[2] S. Sanvito. “Molecular spintronics: The rise of spinterface science”. Nature Physics 6.8 (2010), pp. 562–564.
[3] D.Ciudad, M. Gobbi, C. J. Kinane, M. Eich, J. S. Moodera, and L. E. Hueso. “Sign control of magnetoresistance through chemically engineered interfaces”. Advanced Materials 26.45 (2014), pp. 7561–7567.
[2] S. Sanvito. “Molecular spintronics: The rise of spinterface science”. Nature Physics 6.8 (2010), pp. 562–564.
[3] D.Ciudad, M. Gobbi, C. J. Kinane, M. Eich, J. S. Moodera, and L. E. Hueso. “Sign control of magnetoresistance through chemically engineered interfaces”. Advanced Materials 26.45 (2014), pp. 7561–7567.
