MAGNETIC AND ELECTRICAL PROPERTIES OF ELECTRON BEAM GUN DEPOSITED [Mn/Al] MULTILAYERED FILMS

2231 | P a g e M a y 0 7 , 2 0 1 5 MAGNETIC AND ELECTRICAL PROPERTIES OF ELECTRON BEAM GUN DEPOSITED [Mn/Al] MULTILAYERED FILMS R.Ramanna, T.Sankarappa*, T. Sujatha, J.S. Ashwajeet and P.J. Sadashivaiah a Department of Physics, Gulbarga University, Gulbarga, Karnataka, India b Shridevi Institute of Engineering and technology, Tumkur, Karnataka, India *Author for correspondence: sankarappa@rediffmail.com ABSTRACT


INTRODUCTION
Magnetic and electrical transport studies on magnetic multilayers have been of great interest in the recent years.Multilayers consisting of ferromagnetic metals and nonmagnetic metals were investigated for structure, magnetic and electrical studies [1][2][3][4][5][6].The role of nonmagnetic spacer in tuning the interlayer exchange coupling (IEC) between two neighboring magnetic layers has been first reported by Grunberg [7.] Subsequently, giant magnetoresistance (GMR) and oscillations in IEC with varying width of the nonmagnetic spacer were discovered [8][9][10].In some multilayers, chromium (Cr) spacer was used and investigated interface roughness [11,12], magnetic and electronic structures [13].An anomalous behavior of low temperature resistivity has been observed in the films, Co/M/Co (M = Cr or Cr/Ag or Ag/Cr) [14].The effect of Cu interlayer on grain size was studied in sputtered Fe/Cu multilayers [15].Structural and magnetic properties were probed in Fe/Cu multilayers for varied Fe layer thickness [16].Interlayer diffusion of nonmagnetic metals C, Cr, Pt [17,18], Ag [19], Al and Cu into the magnetic layers has been studied [20].Some Mn-based alloys and compounds exhibited ferromagnetism though they did not contain ferromagnetic elements.The systems, Mn/Al [21], Mn/C/Si [22,23] and Mn/SiO2/Si [24] showed ferromagnetism at room temperature.In these systems, it was learnt that Mn lattices become ferromagnetic and that the ferromagnetic regions were located near the Mn/Al interfaces [25].It is known that the interlayer coupling depends on the thickness of spacer layer.Here we report on structural, magnetic and electrical properties of multilayered films, [Mn(60nm)/Al(20nm)]n (where n= 1,2,9 represent number of repeats) labeled as MAM1, MAM2 and MAM3.For the first time a detailed studies of structural, magnetic and low temperature electrical properties of [Mn/Al] multilayers are reported.

EXPERIMENTAL
The [Mn/Al] n; n= 1, 2, 9 films were deposited in a diffusion pump evacuated coating unit using electron beam gun evaporation method at a temperature of 473K.X-ray diffraction (GIXRD) studies were carried out in Brucker-D8 advance diffractometer with Cu-Kα radiation of 1.5406 Å wavelength.Surface morphology has been investigated by Atomic force microscope (AFM).Average surface roughness of the films were determined on a scan area of 1 μm x 1 μm using Nanoscope software.Magnetic hysteresis studies were carried out in an MPMS SQUIDvibrating sample magnetometer (VSM).The resistivity measurements were done by following a four point method in the temperature range from 5 K to 300K in an Oxford Instrument make setup.

Grazing incidence X-ray diffraction (GIXRD) studies:
The XRD spectra for 2θ between 20° and 60° for MAM1 film is shown in Fig. 1(a) and the Gaussian fit to the peak is depicted in Fig. 1(b).The spectra exhibit single sharp peak at around 43° and that could be due to a plane of Mn crystals indicating (411) reflection.This result is comparable with that reported for Mn-Al multilayers [21].No other peaks are seen in the spectrum up to 60°.The average grain sizes have been estimated using Scherrer's formula [26], D = (0.9λ)/(B cosθB), where D is the grain size, B the angular width in terms of 2θ, θB the Bragg angle and λ the wavelength of the radiation.The interplanar spacing of the films were calculated using the relation, d = λ/(2 sin θB) [Table .1].The average grain sizes obtained were of about 21nm and interplanar spacing of 2.1Ǻ for all the three films.These observations show that the present films are nanocrystalline in nature [26].Crystalline to quasicrystalline phase transition was observed at 873k in [Mn/Al] multilayers due to interfacial diffusion [27].No such transition can be expected in the present films as the experiments are conducted much below 873k.

AFM
AFM images in contact mode with a scan area of 1 μm x 1 μm have been recorded.The AFM images in 2D and 3D for the present films are shown in Fig. 2. The height verses distance profiles were sketched [28].This analysis provided information on average grain size, D (pair of blue dots) and average surface roughness, h [pair of green dots in Fig. (2a)] of the films.It can be seen that D and h increases with increase of t [Table 2].Increase of grain sizes on the surface with increase of t is quite obvious on the 3D images Surface roughness also increased with increase in t.The grain sizes obtained from XRD and AFM are quite different.It can be due to the fact that AFM images provide sizes of grains only on the surface only.Where as XRD gives information about average size of the grains in the entire bulk of the sample.Therefore, the grain sizes obtained by these two techniques are different.

Magnetization
The magnetization, M, as a function of applied field, H, was measured at two different temperatures of 150K and 200K for the fields applied parallel to the surface of the films.The recorded hysteresis (M-H) loops for the present films are shown in Fig. 3.All the three films exhibited ferromagnetism at these two temperatures.Coercive field, Hc, saturation magnetization, Ms, remanent magnetization, Mr, were determined from the hysteresis loops and they are tabulated in Table .3.Hc is found to be increasing with increase of temperature for MAM1 film and it is decreasing as a function of temperature for MAM3.It means that film MAM1 become magnetically hardened with increasing temperature and MAM3 become softened with temperature.
Saturation magnetization Ms, remenant magnetization Mr and antiferromagnetic coupling decreased with increasing temperature in the present films [Table .3].The size of the saturation magnetization measured in the present film is very small and that is expected for when the films are grown on the glass substrates [21].The increase of saturation magnetization with decrease in temperature observed in these films agree with the results of other multilayers [29].Total magnetization has been found to be dependent on the Mn/Al layer thickness [25].In the present films Mn/Al layers thickness was kept constant and the number of repeats was varied.Due to non-availability of the instrument magnetization of MAM2 film could not be measured.The finite values of (1-S) obtained for these films confirm the existence of antiferromagnetic type of interaction between Mn layers through similar to what has been observed in [26,30].The strength of antiferromagnetic coupling decreases with increasing temperature [10].M a y 07, 2 0 1 5

Resistivity Studies
The measured variations of ρ with T for all the three films are displayed in Fig. 4(a-c).All the three films exhibited semiconducting to metal transition (MST) [31].The MST transition shown by the present films is similar to Ni/Al/Ni sandwich films and Zno doped systems [32,33].Transition temperature, Tc measured are recorded in Table .4. Tc decreases with increasing number of repeats of bilayers in the films.This means that MAM1 has got longer tail in the semiconducting region than the metallic region.Where as other two films exhibited longer tails in the metallic region than the semiconducting region.

Resistivity variation in MAM1 film
The conductivity data of MAM1 film in the semiconducting region has been analysed using Mott's small polaron hopping and variable range hopping models due to Mott's and Greaves.The plots of ln(σT) versus (1/T) for MAM1 film as per Mott's SPH model [34] were made and is shown in Fig. 5(a).The least square linear line was fit to the data for temperatures above θD, where θD is the Debye's temperature below which the data deviates from linearity.From the slope, the activation energy, Ea, was determined to be 12.16 meV.
For the data below θD, the variable-range hopping (VRH) models due to Mott [35] and Greave [36] have been applied.The plot of ln(σ) versus T -1/4 as per Mott's (VRH) model has been made and shown in Fig. 5 (b).Similarly, the plot of ln(σT 1/2 ) versus T -1/4 has been sketched and shown in Fig. 5 (c).Linear lines were fit to the data at high temperatures in both the plots and density of states at Fermi level N(EF), were estimated using slopes and intercepts.The value of α required for N(EF) calculation was taken to be 10nm -1 as mentioned in reference [37,38].The N(EF) values so obtained were 1.389X10 28 ev -1 cm -3 and 7.587X10 25 ev -1 cm -3 due to Mott's (VRH) and Greaves (VRH) models respectively.The data in the metallic region for MAM1 film was fit to the linear expression, ρ= A+BT and the coefficients of A and B were determined (Table 5).

Resistivity variation in MAM2 and MAM3 films
Semiconducting data of MAM2 film has been analysed using Mott's small polaron hopping model similar to that performed for MAM1.The activation energy, Ea, was determined to be 2.08 meV.Activation energies determined for MAM1 and MAM2 films are much smaller than that measured in Cu-Ni multilayers [39] and they were predicted to be due to grain boundary scattering.
The resistivity data in the semiconducting region for MAM3 was very limited and therefore, has not been considered for analysis.A careful observation of variation of ρ with T revealed that there exists three different power laws for the measured temperature range as shown in Fig. 6 (a-c) for MAM3 film.Hence, the following expressions were fit to the data for different temperature ranges.
Where, ρ(0) is the residual resistivity which is taken to be equal to the measured value at 5K in all the films.By regressional analysis, the coefficients a1 & a2, exponents k, m & n and constants A & B were extracted.The fit parameters thus obtained are tabulated in Table 5.All the three films produced exponent n to be near unity.For pure nonmagnetic metals, the linearity between resistivity and temperature is expected at high temperature [40].For MAM2 and MAM3 films, in the temperature range, 45K ≤ T ≤ 100K, the exponent, m is found to be 1.96 and 2.53 respectively and that is in agreement with the results on a magnetic layer [2].In this range of temperature, electronphonon s-d scattering dominates over electron-magnon scattering which is expected to be dominant below 20K.For the temperature, T ≤ 45K, the exponent, k is found to be 3.34, which is greater than 2 observed for bulk Fe [41].

CONCLUSIONS
(i) The Multilayered films, [Mn(60nm)/Al(20nm)]n; n = 1, 2 and 9 were deposited on to glass substrates by electron beam gun method at 473 K under high vacuum conditions.The structure, grain sizes, surface roughness were probed by grazing incidence X-ray diffraction (GIXRD and atomic force microscope (AFM).
(ii) At two different temperatures of 150K and 200K, magnetic hysteresis loops were recorded in a MPMS SQUIDvibrating sample magnetometer (VSM).At these two temperatures, coercive field saturation magnetization, rem anent magnetization and antiferromagnetic coupling were determined and discussed.
(iii) Electrical resistivity as a function of temperature in the range from 5K to 300K has been measured by four point method for all the films.All the three films exhibited semiconducting to metal transition.The data corresponds to metallic region has been analysed and relation between resistivity and temperature were established.Semiconducting data of has been analysed polaron hopping models.
It is for the first time that antiferromagnetic interaction between Mn layers through Al layer and MST have been measured in [Mn/Al] multilayered films.

Fig. 1 :
Fig.1: (a) GIXRD spectra for MAM1 film and (b) spectra around the peak position.The solid line is a Gaussian fit to the peak.

Fig. 7 .
Fig.7.Plots of resistivity, ρ versus temperature, T for MAM3 film for (a) 10K to 45K (b) 45K to 100K and (c) 100K to 300K.The continuous curve passing through the data points are fits to the data.