“The noise is the signal” said more than a decade ago Rolf Landauer, founder of theory of quantum electron transport in nanostructures MAGNETRANS group is unique in Spain and one of few in Europe with expertise in experimental studies of low frequency noise (1/f noise and shot noise (SN)) in magnetic nanostructures. Shot noise provides information on electron correlations on nanoscale which is inaccessible from electron transport data. Normalized shot noise (Fano factor) which is1 for uncorrelated or direct tunneling, is expected to be suppressed below 1 for correlated tunneling processes and to exceed 1 for cotunelling. Our recent studies (in collaboration with group of Dr. Moodera, MIT, Boston, USA and prof. Y. Tserkovnyak, UCLA, USA) of shot noise in Co(80Å)/Al2O3(14Å)/Py (100Å) magnetic tunnel junctions (MTJs) without doping and with Cr δ-doping of the insulating barrier (Al2O3/Cr(δÅ)/Al2O3) have shown that SN is reduced for the intermediate resistances with F ≈ 0.65–0.8. Sub-Poissonian shot noise indicates correlated electron tunneling, most probably through the localized states formed by the defects inside the barrier. On the other hand, the SN is found to be enhanced for the antiparallel alignment of the ferromagnetic electrodes in the MTJs with tunneling weakly affected by spin-flip scattering. We proposed a simple model, which considers trap-assisted sequential tunneling, that qualitatively explains the main experimental results.

<

More reading: R. Guerrero, F.Aliev,  Y. Tserkovnyak, T. S. Santos, and J. S. Moodera,  Phys. Rev. Lett. 97, 266602 (2006).

In the magnetic tunneling phenomena the main efforts up to now have been concentrated on the increase of tunneling magnetoresistance(TMR) value. We have grown fully epitaxial Fe/MgO/Fe MTJs and Fe(100)/Fe–C/MgO(100)/ Fe(100)on MgO(100) substrate (in collaboration with group of Dr. Tiusan, University Poincare, Nancy, France) and measured at low temperature (10 K) high voltage bias dynamic conductivity (up to 2.7 V) and shot noise (up to 1 V) magnetic tunnel junctions as a function of the magnetic state. Our junctions show large tunnel magnetoresistance (185% at 300 K and 330% at 4 K) and multiple sign inversions in the magnetoresistance for bias polarity when the electrons scan the electronic structure of the bottom Fe–C interface. The shot noise shows a Poissonian behavior (Fano factor is close to 1), indicating direct tunneling. These experiments demonstrate a pure spin-dependent direct tunneling mechanism and validate the high structural quality of the MgO barrier.

Knowledge of low frequency noise on electron devices is important both from fundamental and applied points of view. Although microscopic theory is still absent for so called 1/f noise dominating at very low frequencies, in the first approximation it may be understood as due to different relaxation times of the defects unavoidably present inside the barrier or as a result of fluctuations of the magnetization.
Recently we have realised first experimental study of electron transport and low frequency noise in magnetic tunnel junctions with a thin layer of silicon introduced inside the insulating barrier (in collaboration with Dr. Moodera, USA, Profs. Barnás and Dugaev, Poland). Here we investigated the role of a layer of impurities of different thickness in the conductance, on the TMR and on the voltage noise. Both electron transport and noise measurements indicate a breakdown of Coulomb blockade for Si layers thickness exceeding one monolayer. The conductance results have been explained using the Larkin-Matveev approach, supporting a crossover from a diluted impurities regime to a Coulomb blockade regime in the higher thickness silicon layer.

One of the most promising future applications of MTJs is as ultra-sensitive detectors of magnetic field in biology, for example detection of brain activity.
This possibility has motivated us to investigate possibility of interface engineering of fully epitaxial MTJs in order to improve dramatically signal to noise ratio (work done in collaboration with University Poincare, Nancy, France)

Our recent studies revealed record low 1/ f noise at room temperature in fully epitaxial Fe(100)/MgO(100)/Fe(100) magnetic tunnel junctions. We have reported on room temperature 1/ f noise in fully epitaxial Fe (45nm) /MgO (2.6nm) /Fe (10nm) magnetic tunnel junctions (MTJs) with and without carbon doping of the Fe/MgO bottom interface. We have found that the normalized noise (Hooge factor) asymmetry between parallel and antiparallel states may strongly depend on the applied bias and its polarity. Both types of MTJs exhibit record low Hooge factors being at least one order of magnitude smaller than previously reported.

More reading: R. Guerrero, F. G. Aliev, R. Villar, T. Santos, and J. Moodera, phys. stat. sol. (a) 205, 1040, (2008)

More reading: R. Guerrero, D. Herranz, F. G. Aliev, F. Greullet, C. Tiusan, M. Hehn, and F. Montaigne, Appl. Phys. Lett. 91, 132504 (2007)

More reading: F. G. Aliev, R. Guerrero, D. Herranz, R. Villar, F. Greullet, C. Tiusan and M. Hehn, Appl. Phys. Lett. 91, 232504 (2007)

---

One of future direction in spintronics is related to possibility of switching the magnetic state of the device not only by external field but also by current. This second approach is scalable with reduced dimensions of the nanostructures.
Motivated by recent theoretical predictions of much lower critical current needed to move DWs in antiferromagnets than in ferromagnets, and having in mind complicated DW structure of antiferromagnets, such as Fe/Cr multilayers, we have proposed to investigate current-driven magnetization dynamics in synthetic [Fe/Cr]10 multilayer antiferromagnets by using low-frequency voltage noise measurements.

Indeed, we observed suppression of the noise above a critical current density of about 2´10^5 A/cm². Theoretical estimations suggest that this effect may be attributed to current induced motion of domain walls in the antiferromagnet. The observed critical current density is about one order of magnitude smaller than for ferromagnetic systems. Our results are relevant for possible applications of antiferromagnetic metal spintronics in, e.g., magnetic memory storage technology.

(a) Dependence of the Hooge factor (α) on the magnetic field for MML1 at 77 K with different applied current densities. (b) Hooge factor plotted on a logarithmic scale vs magnetic field for MML1 at 77 K with two different applied current densities. The normalized noise vs field measurements are done with a current density of j=3.3´10^5 A/cm² and the results are shown both for increasing and decreasing magnetic field. The experiments have been carried out with dc current parallel to the external magnetic field and to the magnetic easy (001) axis

More reading: D. Herranz, R. Guerrero, R. Villar, F.G.Aliev, A. C. Swaving, R. A. Duine, C. van Haesendonck and I. Vavra, Phys. Rev. B79, 134423 (2009)

---

MAGNETRANS group has recently open a new research line related to experimental study of magnetization dynamics above GHz range (currently 8.5 GHz and between 300K and 1.5K, soon up to 40 GHz) in different magnetic and superconducting nanostructures including magnetic dots, spintronic devices and other nanostructures such as arrays of nanowires.

Study of the magnetization dynamics at room temperature in periodic arrays of 50 nm thick FeNi (Py) circular magnetic dots of 500 nm radius and different center to center distance (1200 and 2500 nm) had been performed using a broadband magnetometer based on Vector Network Analyzer which works between 300 kHz and 8.5 GHz. A comparison between the dynamic response, ferromagnetic resonance (FMR) and its line-width, with static magnetic characteristics such as magnetization curves had been presented and lead to more profound vision about magnetic vortex state. The FMR peak appears just above the nucleation field and is perfectly described by Kittel formula taking into account the demagnetizing factor of an individual magnetic dot. In addition to FMR we observed a spin wave resonance below the uniform mode (mode B at Figure), which could be attributed to spin waves in confined systems. The FMR line-width shows a significant broadening close to the field region corresponding to nucleation of magnetic vortex.

More reading: J.F.Sierra, A.A.Awad, G.N.Kakazei, F.J.Palomares and F.G.Aliev ”Broadband magnetic response of periodic arrays of FeNi dots” IEEE Transactions on Magnetics,  44, pp.3063-3066  (2008).

---

We have studied linear spin dynamics in the vortex state of Permalloy cylindrical dots subjected to in-plane bias magnetic field. We demonstrate experimentally by a broadband ferromagnetic resonance technique and by simulations that breaking the cylindrical symmetry of the magnetic vortex gradually changes and suppresses the azimuthal spin eigenmodes below the vortex nucleation field and leads further to the appearance of new eigenmodes. We had found that the spin-wave eigenmodes (n,m) classification based on the number of nodes in radial and azimuthal directions could be applicable to the vortex shifted by magnetic field only below the vortex nucleation field. An important conclusion that could be noticed easily from the Figure is that the microwave excitation in different directions with respect to the bias field involves excitation of different spin wave modes. The parallel microwave field pumping is shown to be a unique tool to observe spin excitation modes localized near the strongly shifted vortex core for the bias field between the vortex nucleation and annihilation fields. Meanwhile, the perpendicular field pumping, which excites the spin waves throughout the entire dot, reveals a crossover between two dynamic vortex regimes near the nucleation field). Our results are important for understanding the dynamics in different vortex systems in confined biased conditions, e.g., in arrays of vortex nano-oscillators excited by a spin-polarized current, natural vortex systems (typhoons, cyclones, vortex galaxies).

More reading: F.G.Aliev, J.F.Sierra, A.A.Awad, et al., PRB, accepted to PRB

---

Ferromagnetic resonance and static magnetic properties of CoFe/Al2O3/CoFe/Py and CoFe/Al2O3 /CoFeB/Py magnetic tunnel junctions and of 25 nm thick single-layer Permalloy (Py) films have been studied as a function of temperature down to 2 K. The temperature dependence of the ferromagnetic resonance excited in the Py layers in magnetic tunnel junctions shows “knee-like” enhancement of the resonance frequency accompanied by an anomaly in the magnetization near 60 K(the figure shows the resonance frequency as a function of temperature for different samples). We attribute the anomalous static and dynamic magnetic response at low temperatures to interface stress induced magnetic reorientation transition at the Py interface which could be influenced by dipolar soft-hard layer coupling through the Al2O3 barrier.

More reading: J. F. Sierra, V. V. Pryadun, F. G. Aliev, S. E. Russek, M. García-Hernández, E. Snoeck, and V. V. Metlushko, Temperature dependent dynamic and static magnetic response in magnetic tunnel junctions with Permalloy layers Appl. Phys. Lett., v.93, p. 172510, (2008)

---

Recently we observed unexpected resonant response in [Fe/Cr]10 multilayers epitaxially grown on MgO(100) substrates which exists only when both ac current and dc magnetic field are simultaneously applied. The magnitude of the resonances is determined by the multilayer magnetization proving their intrinsic character. The reduction of interface epitaxy leads to non-linear dependence of the magnitude of resonances on the alternating current density. We suggest that the existence of the interface transition zone could facilitate the subatomic vibrations in thin metallic films and multilayers grown on bulk insulating substrates.

    

More reading: F.G.Aliev, V.V.Pryadun, E. Snoeck, “Unexpected resonant response in [Fe(001)/Cr(001)]10 /MgO(001) multilayers in magnetic field”  Phys. Rev. Lett., 102, 035503 (2009)

---

The possibility of ferroelectric memory in very thin films is closely associated with stability of single domain ferroelectric state in films with real metallic electrodes.  It has been argued that due to finite screening length in real metals the single domain state is at best metastable in thin ferroelectric films. Therefore, to reveal the possibilities of the ferroelectric memories one has to determine not the minimal film thickness compatible with ferroelectricity as many authors do but to determine the thickness corresponding to a sufficiently long escape time from the metastable state permitting realization of the memory of a desired duration. The problem of calculation of this escape time is very complicated but a possible path to realize the calculation has been indicated.

More reading: A.M. Bratkovsky and A.P.Levanyuk “Continuous theory of ferroelectric states in ultrathin films with real metallic electrodes” Journal of Computational and Theoretical Nanoscience 6, 1 (2009).