phys. stat. sol. (a) 196, No. 1, 161–164 (2003) / DOI 10.1002/pssa.200306376

© 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 0031-8965/03/19603-0161 $ 17.50+.50/0

Kerr magnetometer based on a differential amplifier

J. Wrona*, 1, T. Stobiecki1, R. Rak1, M. Czapkiewicz1, F. Stobiecki2, L. Uba3, J. Korecki4, T. Ślęzak4, J. Wilgocka-Ślęzak4, and M. Rots5 1 Department of Electronics, University of Mining and Metallurgy, Krakow, Poland 2 Institute of Molecular Physics, Polish Academy of Science, Poznan, Poland 3 Institute of Physics, University of Białystok, Poland 4 Department of Solid State, University of Mining and Metallurgy, Krakow, Poland 5 Instituut voor Kern-en Stralingsfysica, Leuven, Belgium

Received 1 July 2002, accepted 15 October 2002 Published online 19 March 2003

PACS 33.55.Ad, 73.21.Ac

The Kerr magnetometer with differential amplifier which in contrast to lock-in detection technique, is especially recommended for rapid measurements up to 1.5 kHz is described. The calibration of the Kerr rotation angle was performed on the standard sputtered films (with thickness from 2 nm to 50 nm) and Fe-wedge sample, prepared by MBE-technique, in the range of thickness from 1ML to 50 ML of Fe. The test measurements on spin valve structure were also performed.

Introduction Multilayers in the form of artificial superlattices in the range of thickness of several monoatomic layers consisting of different magnetic materials (e.g. soft or hard ferromagnets, antiferro-magnets, semiconductors, insulator) form nanosystems which can be applied as spin electronics devices. Such elements require a sensitive magnetometer for measurements of the hysteresis loops. In the present paper we show the high sensitivity and low cost solution for the Kerr magnetometer based on differential amplifier which in contrast to lock-in detection technique is especially recommended for rapid measure-ments up to 1.5 kHz. Experimental setup The arrangement for the Kerr rotation measurements is shown in Fig. 1. The He–Ne laser (Z) with 5 mW (λ = 633 nm) and a Glan–Thompson prism polarizer (P) with an extinction ratio 1 × 10–6 were used to obtain linearly polarized light perpendicular (s-polarization) or parallel (p-polarization) to the plane of incidence. The beam reflected from the surface of the sample (S) placed in the centre of the air – Helmholtz coils (H) (or special construction electromagnet [1]) then passes through Wollaston prism (W) (with an extinction ratio 1 × 10–6) which splits the beam into two mutually orthogonal linearly polarized beams. These beams are focused by a lens (L) on the photodiodes D1 and D2 of the differential amplifier (A). The amplifier delivers both the common and the differential signals which are proportional to the sum and difference of the light intensities of the beams, respectively. These signals are measured by a high resolution (16 bit) and rapid (200 kHz) AD converter in PCI 6035 – com-puter card. The computer controls the voltage of a programmable bipolar current source (KEPCO BOP 36-12M) which is used to drive the current through the Helmholtz coils (or electromagnet) to provide a

* Corresponding author: e-mail: wrona@agh.edu.pl, Phone: +48 12 617 30 44, Fax: +48 12 617 35 50,

162 J. Wrona et al.: Kerr magnetometer based on a differential amplifier

PCPCI-6035

BN

C-2

090

KEPCO

AI

01

20

AO

Htm-11n

RS-232

Z

PW

L

H

S

35̊

D1

D2

DifferentialSignal

Common Signal

Hp

AUD

UC

RW

Fig. 1 Schematic layout of Kerr magnetometer arrangement.

sweeping magnetic field up to 1.5 kHz. The magnetic field is accurately calibrated before measurements of the hysteresis loop using a teslameter (Htm-11n). The MOKE system is programmed by LabView. Calibration The differential to common signal ratio (which is proportional to the Kerr rotation angle) was accurately calibrated as a function of the polarization-plane rotation angle. Additionally the calibra-tion of the Kerr rotation angle was performed on standard sputtered Fe films with a thickness from 2 nm to 50 nm of Fe (Fig. 2) using a Kerr spectrometer (in Institute of Physics at the University of Białystok)

based on the polarization modulation technique. To calculate the Kerr rotation of Fe for s and p polarized light as a function of the film thickness the phenomenological matrix formalism was employed, based on Maxwell theory and developed by Zak et al. [2]. The data for the optical and magneto-optical constants

polarization p

Thickness Fe [nm]

0 10 20 30 40 50

Θm

ax [m

in]

0

1

2

3

4

polarization s

Thickness Fe [nm]0 10 20 30 40 50

Θm

ax [m

in]

0

1

2

3

4

Our deviceBialystok Univ.Calculation

Fig. 2 Kerr rotation angle of Fe sputtered films measured in saturation field (0.5 T) vs. thickness of Fe.

phys. stat. sol. (a) 196, No. 1 (2003) 163

Thickness Fe [ML]0 10 20 30 40 50

Θm

ax [m

in]

-0.4

-0.2

0.0

0.2

0.4

polarization ppolarization s

µ0H [mT]-300 -200 -100 0 100 200 300

Θm

ax [m

in]

-0.04

-0.02

0.00

0.02

0.042 ML1 ML

.

Fig. 3 Kerr rotation angle of MBE prepared Fe-wedge-Fe sample vs. thickness. The arrows indicate the transi-tion from in-plane to perpendicular anisotropy.

of bulk Fe were assumed: n = 2.87–i3.36 and Voight constant Q = 0.037–i0.0066 from [2] and refractive index of glass substrate 1.5151 for wavelength of laser 633 nm. Above the thickness of 25 nm the ex-perimental and calculated data agree very well but below this, the difference between them increases with decreasing thickness of Fe due to nonstoichiometric Fe oxide about 2 nm thick [1, 4]. The ultrathin Fe wedge film was prepared by MBE technique on a Au(001) 170 Å thick buffer layer grown on MgO single crystal substrate. The thickness of Fe was in the range of 1 ML and 50 ML (1 ML of Fe = 1.45 Å). The sample was capped by 30 Å of Au to prevent oxidation. The longitudinal Kerr effect measurement proved in-plane magnetic anisotropy for Fe thickness between 50 and 5 ML, showing decrease of Kerr rotation with decreasing Fe thickness (Fig. 3). Below critical thickness (5 ML) drastic increase of Kerr rotation was observed due to the change of the easy magnetization axis to the perpendicular direction combined with the polar contribution to the measured signal [3]. Discussed measurements on Fe wedge sample proof that 2 ML Fe is still ferromagnetic at room temperature (Fig. 4). The resolution our MOKE magnetometer is estimated on 0.001 min.

µ0H [mT]-25 -20 -15 -10 -5 0 5

J [T

]

-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

Rot

atio

n [m

in]

-0.3

-0.2

-0.1

0.0

0.1

0.2

0.3

KerrR-VSMCalculation

Fig. 4 Kerr rotation hysteresis loops of MBE prepared Fe wedge sample.

Fig. 5 Comparison of the spin valve hysteresis loops: magnetization (R-VSM), Kerr angle rotation and calculated.

164 J. Wrona et al.: Kerr magnetometer based on a differential amplifier

Test measurements on spintronic elements On selected spin valve structures and an array of mag-netic dots measurements of the hysteresis loops were performed [1]. A spin valve Ta(52 Å)/Co (44 Å)/Cu(22.09 Å)/Co(44 Å)/FeMn(100 Å)/Ta(52 Å) structure sputtered on Si(100) is demonstrated here as an example (Fig. 5). The signal of the magnetooptical hysteresis loop, in contrast to magnetiza-tion, due to intensity absorption effect, is higher for pinned layer (lying on the top of multilayer system), but the switching fields are almost the same. The calculation of magnetization hysteresis loop agree very well with measurement of R-VSM (Resonance Vibrating Sample Magnetometer). The fitting parameters are the following: interlayer exchange energy J1 = 7.9 × 10–6 J/m2, exchange biased energy EEB = 113 × 10–6 J/m2 and uniaxial anisotropy of free and pinned layer Ku1 = 2.05 × 103 J/m3 and Ku2 = 2.4 × 103 J/m3, respectively. Conclusion We conclude that the designed magnetometer is highly sensitive (with resolution of 0.001 min) and recommended for rapid measurements. Construction is simple both from the electronic as well as the mechanical point of view and its cost is very low.

Acknowledgements This work was partially supported by the State Committee for Scientific Research (grant No. PZB/KBN/044/P03/2001 and 11.120.68).

References

[1] J. Wrona, doctor thesis, (University of Mining and Metallurgy Krakow, 2002). [2] J. Zak, E. R. Moog, C. Liu, and S. D. Bader, J. Magn. Mat. 89, (1990) 107. [3] C. Liu and S. D. Bader, J. Magn. Mat. 93, (1991) 307. [4] K. Postava, J. F. Bobo, M. D. Ortega, B. Raquet, H. Jaffres, E. Snoeck, M. Goiran, A. R. Fert, J. P. Redoules,

J. Pištora, and J. C. Ousset, J. Magn. Magn. Mat. 163, 8 (1996).