Abstract

We present the detailed study of activation energy E_a according to the thickness of the magnetic layer (t_Co=0.7,0.8 and 1nm). The study was carried out at room temperature by means of polar magneto-optical Kerr effect magnetometry (PMOKE) using a He–Ne laser (λ=633nm). We found that the activation field μ₀H_a, the coercive field μ₀H_C and the average activation energy E_a are weak for the sample with thickness t_Co = 1 nm.

1. Introduction

The miniaturization of the devices using ferromagnetic materials made grow the interest for these materials during these last decades for the researchers. Research on these materials is directed either towards the comprehension of very fundamental mechanisms or towards an important prospects for applications such as ultra-high density storage ^{1, 2}. Indeed, the writing of the elementary bits of information is traditionally done by application of an impulsion of magnetic field. Thus the magnetization reversal dynamic plays a fundamental role in the creation of these elementary bits of information. Energy necessary to create a first reversed magnetic field is called activation energy. A perfect control of the parameters controlling the activation energy would make it possible to control the electric current necessary for the creation of the elementary bits of information. Some work has been devoted to the energy of activation ^{3, 4, 5} but these works did not discuss the effect the thickness of the magnetic layer on energy of activation.

The aim of this paper is to show that the thickness of the magnetic layer can influence the value of the activation energy.

2. Material and Methods

Si(100) substrate is beforehand cleaned by ultrasounds in an acetone bath. After the cleaning, this substrates is thermally oxidized in a furnace at 1200°C during 2 hours. This time is sufficient for the formation of an oxide layer on the silicon surface substrate.

Au/Co/Au films were prepared by electron beam evaporation in an ultrahigh vacuum chamber, with a base pressure about of and approximately during deposition on SiO₂, at room temperature.

A first 25 nm thick Au film is deposited on the substrate at a deposition rate of , as calibrated with a quartz microbalance, followed by annealing at during to reduce the surface roughness.

The Au film is (111) textured, as shown by X-ray diffraction (Figure 1(a)). Figure 1(b) shows the 2D AFM image of the Au buffer layer after annealing. The surface roughness (root mean square: rms) was measured to be about 0.2 nm. Using the surface corrugation obtained from 2D AFM, we estimate a lateral grain size of 40-60 nm. Cobalt layers with thicknesses () equal to 1,0.8 and 0.7nm are then deposited on the Au/SiO₂ at a deposition rate of 0.2nm/min. Finally, a second Au layer with a thickness about of 5nm is deposited on top of the cobalt layers.

The (111) texture of the Au buffer layer suggests, in each case, a possible epitaxial growth of the cobalt layer with the Hexagonal Close-Packed (0001) structure ^{6, 7, 8}.

Magnetic hysteresis loops, at a field sweep rate of , were recorded at room temperature (RT) by polar magneto-optical Kerr effect magnetometry (PMOKE) using a He–Ne laser (). On the hysteresis loops we measured the coercive fields and the nucleation fields . Table 1 shows magnetic quasi statistic parameters deduced from the hysteresis loops of the three samples.

The energy needed to reverse magnetization can be expressed in the following way ^{3, 4}:

(1)

where is an activation energy at zero field i.e. thermal energy required to initiate the magnetization reversal in the absence of the field, is the saturation magnetization and is the Barkhausen volume (the magnetization volume that reverses during a single activation event). In this context, the time so that a sample is demagnetized, under the applied field , should follow the Arrhenius-Néel law:

(2)

We recorded the reduced magnetization reversal curves in time. From magnetization reversal curves vs we deduced vs . The fit of vs by using equation (2) allowed us to determine and . On Figure 2 are represented vs and their fitting.

The experimental dots of Figure 2 show that evolves under the Arrhenius law. The adjustments of these experimental dots by Eq. (2) gives the values of , and .

3. Results and Discussions

On Figure 2, we notice in the three cases that there is an agreement between the adjustment curve and the experimental dots. In Table 2 are summarized the values of , and . For the three samples is , what mean that this parameter does not vary too much according to the thickness of the magnetic layer. The value of is the same magnitude order we found on (Pt/Co)₃ multilayers ⁵.

The values of , in Table 2, are in general weak compared to that we found on our (Pt/Co)₃ multilayers ⁵. in magnetic layer having 0.8 nm thickness is twice higher than those of the magnetic layers having thicknesses = 1 nm and 0.7 nm. The highest value of the activation energy found in this sample (nm) let’s think that this sample would have more defects than the two others. This sample has also the highest value of . If we emit the assumption that the magnetization saturation is almost of the same magnitude order for the three samples ⁶ then the greatest value of would be in the magnetic layer of 0.8 nm thickness. This lets think that reversed initial volumes would be larger than those of the two other samples. Thus, magnetization reversal in this sample would be done by a mode different from that of the two other samples. In the samples of thickness = 1 nm and 0.7 nm, the weakness of the values of activation energy lets think about the magnetization reversal dynamics dominated by the magnetic domain wall motion. These deductions are in agreement with our previous works ⁵ where we showed that the activation volume and the activation energy are highs when magnetization reverse mainly by several nucleate centers due to the inhomogeneities.

Knowing and for each sample, it is possible to estimate these times of demagnetization in zero field but under the temperature effect only at 300 K. In fact, Eq. (2) at zero field become:

(3)

We found respectively = , and for 0.7 nm, 0.8 nm and 1 nm.

for where is the activation field that cancels the energy barrier. Under these conditions the activation energy is given by:

(4)

By taking into account the values of found, in Table 2, Eq. (4) gives respectively = 10.93 mT, = 13.03 mT and = 7.43 mT for the samples with 0.7 nm, 0.8 nm and 1 nm. In the three cases the value of is lower than that of what shows clearly that the magnetization reversal is well initiated before the sample is demagnetized. The lowest value of is found for the sample with = 1 nm. The lowest values of and found for this sample shows that the reversal of its magnetization would not require enough of electrical energy.

4. Conclusion

We studied ultrathin cobalt films with thickness 0.7, 0.8 and 1 nm. We extracted for these three samples the average activation energy in zero field. does not vary linearly according to the thickness of the magnetic layer. We found that the activation field , the coercive field and the average activation energy are weak for the sample with thickness = 1 nm. This result shown that with this thickness one can reverse magnetization with a low electrical energy.

Acknowledgments

This work is the result of a collaboration between the LMOP laboratory (Tunisia) and the University of Kara, Togo. We acknowledge the help of LMOP in Tunisia and that of «équipe Micro et Nano-Magnétisme de l’Institut Néel à Grenoble, France».

References