Open Access
Add to BibSonomyAbstract
A Faraday Effect sensor with Nd2Fe14B biasing magnetic film was described. Ta/Nd2Fe14B/Ta films were grown by magnetron sputtering method. The magnetic domain in the sensor with the Nd2Fe14B biasing magnetic film can persist its distribution. The average linearity error of Faraday Effect sensor with biasing magnetic film decreased from 1.42% to 0.125% compared with non-biasing magnetic film, and the measurement range increased from 820 Oe to 900 Oe.
©2012 Optical Society of America
1. Introduction
Faraday Effect sensors had been extensively researched and developed during more than 40 years [1]. Most of this effort had been focused on magneto-optical current sensor (MOCS), as it had several advantages over conventional current transformer, such as effective isolation from high potentials, immunity against electromagnetic interferences, high dynamic range, no saturation effects, high bandwidth, compact and lightweight design, et al. [2]. There was an increasing demand for high accuracy MOCS in electric power industry [3–5].
In these sensors, garnet film was used as Faraday element. A large error due to the small domain wall coercive force in the small magnetic field region was assumed and studied [6,7]. Ga-substituted BiRIG films were prepared for high sensitivity [8], converged beam optical system of sensor can also get high linearity [9]. Permanent thin film with perpendicular anisotropy had many potential applications in the field of magnetic microelectro-mechanical system, as well as for biasing [10], and playing a fundamental role in the progressive miniaturization of devices. Bulk permanent magnets had been used to linearize and improve the sensitivity of reflective Faraday Effect sensor as biasing magnet [11]. Biasing magnetic field was also used in magnetoresistors to improve their sensitivity and linearize the transducing effect. Thermomagnetic patterned NdFeB magnetic film can be used to fabricate microflux sources by single pulsed laser irradiation method [12]. The measurement range and linearity error of MOCS and the magnetic domain dynamics of garnet were very important for scientific research and application.
In order to improve the performance of MOCS, one method based on producing a Faraday rotation angle of the polarization of the optical beam propagating through the garnet, by means of Nd2Fe14B biasing magnetic film was demonstrated. The films had been grown on garnet substrate by magnetron sputtering method. This resulted in clear advantages such as smaller sensor size, simple design, lightweight, et al.
2. Experiments
The garnet film (Granopt Co., Ltd) with a thickness of 0.39 mm and 3 mm × 3 mm square was used as substrate. Ta(50 nm)/Nd2Fe14B(5 µm)/Ta(50 nm) films (shown in Fig. 1 ) were deposited onto garnet substrate through a stainless steel mask in a magnetron sputtering system with base pressure <1 × 10−5 Pa. Ta films were used as both buffer and capping layer to prevent diffusion into the garnet film and oxidation. During deposition, the substrate was heated to 400 °C, after deposition, the sample was annealed at temperature 550 °C for 30 min in a vacuum of <1 × 10−4 Pa to crystallize Nd2Fe14B phase. The crystalline structure was conducted by a Rigaku D/Max-3C X-ray diffraction (XRD) with Cu Kα radiation. Magnetic properties were measured with the external magnetic field either perpendicular or parallel to the garnet film plane, using Versa-lab (50K, 3T, Quantum Design) at room temperature. Then the sample was magnetized in magnetizer YD-20, and the maximum magnetic field was more than 30000 Oe. The magnetic direction was out of plane.
The garnet films with multi-domain structure and the evolution of domain structures under externally magnetic field (Hext) were observed by complementary metal oxide semiconductor image sensors in polarized light microscope (POM) and transmission illumination. The objectives and eyepieces were in magnifications of 5x and 10x.The garnet with perpendicular magnetization was placed with its easy magnetization axis parallel to the incident beam, was shown in Fig. 1(b).
The experimental set-up was shown schematically in Fig. 2 . The Faraday rotation angle was measured using Faraday Effect sensors with garnet and garnet/Ta/Nd2Fe14B/Ta as a sensor head. A 1550 nm optical fiber laser and InGaAs Quadrants detector were used as light source and detector, respectively. The analyzer was at an azimuth angle of 45° with respect to the polarizer. The Hext from −940 Oe to 940 Oe was applied in the thickness direction and the light was propagating along the same direction.
The Faraday rotation angle θ can be got from the following expression
where V is the Verdet constant of garnet, B is the magnetic field and L is the thickness of garnet. Linearity error (%) was calculated byHin is the applied magnetic field, Hout is the sensor output, n is the number of measurements.
3. Results and discussion
Figure 3 showed XRD patterns of Nd2Fe14B biasing magnetic films deposited at 400 °C and annealed at 550 °C for 30 min in a vacuum of <1 × 10−4 Pa. It can be noticed that the prominent peaks were (004) (2θ≈30.6°), (006) (2θ≈44.8°) and (008) (2θ≈60°) reflections of Nd2Fe14B for the biasing magnetic film, indicating that c axis of the films was perpendicular to garnet substrate.
Fig. 3 XRD patterns for Garnet/Ta/Nd2Fe14B/Ta thin films deposited at 400 °C and annealed at 550 °C for 30 min.
Figure 4 showed the typical M-H loops measured in an applied field of 25KOe. The coercivity of biasing magnetic film were 1.8T (perpendicular to the film plane) and 1.4T (in the film plane), respectively. It can be noticed that, in spite of the demagnetizing field, the coercivity of Nd2Fe14B film in the perpendicular direction was much higher than its in the plane direction, indicating that there was a perpendicular magnetic anisotropy.
Fig. 4 Hysteresis loops at room temperature of Nd2Fe14B biasing magnetic film The loop shown in solid black boxes is for applied field perpendicular to the film plane and the in red circle is for field in the film plane.
Ta(50 nm)/Nd2Fe14B(5 μm)/Ta(50 nm) films were grown on garnet film, as shown in Fig. 1. This was grown by magnetron sputtering method and the Ta and Nd2Fe14B alloy were targets. The films surface was very smooth and clean. A clear step on garnet implied that the stainless steel masks prevented the Ta/Nd2Fe14B/Ta diffusion while sputtering. The diffraction effects can be ignored and the laser beam can be considered that it went through in parallel and straight line, as the gap in Nd2Fe14B film was 200μm far more than the laser wavelength.
The micro-structure, evolution and dynamics characterizations of garnet magnetic domains as a function of the Hext, observed by transmissive POM, were shown in Fig. 5(a) –5(c). The magnetic domain (Fig. 5(a)) remained pronounced random strips and labyrinths structures under zero magnetic field, as the domain self-energy and interaction energy were much greater than the thermal fluctuation energy [13-14]. With the Hext increased, like thermal fluctuation, the magnetic domains decreased obviously and showed cells and mixed states of stripes, shown in Fig. 5(b). However the magnetic domain cannot recover to the initial state when the Hext decreased and equaled to zero (as shown in Fig. 5(c)), as the domain self-energies and interaction energy were greater than the thermal fluctuation energy and there was no other external field.
Fig. 5 POM image of magnetic domain in garnet at different Hext Hext = 0 Oe; (b)Hext = 400 Oe; (c)Hext = 0 Oe.
The dynamics characteristics of garnet with Nd2Fe14B biasing magnetic film had been studied as a function of the Hext, as shown in Fig. 6(a) –6(c). At the value of Hext equaled to 0 Oe, the magnetic domains in garnet film were represented by regularly distribution symbols, shown in Fig. 6(a). The magnetic domain distribution lied in a certain direction and suggested an affect by Nd2Fe14B biasing magnetic films. The magnetic domains decreased and showed cells when the Hext increased, as shown in Fig. 6(b), as the Hext and Nd2Fe14B biasing magnetic film had combined effect on garnet. The magnetic domains in garnet can transform back into the initial state (Fig. 6(c)) and be the same direction distribution. The magnetic domain distribution of garnet persisted in the sensor with Nd2Fe14B biasing magnetic field. And further studies concerning conformal structures are needed, as the evolution of domain structures under Hext was very important.
Fig. 6 POM image of magnetic domain in garnet with biasing magnetic film at different Hext Hext = 0 Oe; (b)Hext = 400 Oe; (c)Hext = 0 Oe.
On other hand, garnet film has large Faraday Effect and excellent transparency in the near infrared light. The Faraday rotation angles of garnet and with Nd2Fe14B biasing magnetic film were measured by Faraday rotation measurement experimental setup at room temperature. The Results were displayed in Fig. 7 , The Faraday angle curve (black line) of garnet with biasing magnetic film was moved down and right (when the Hext>0), and move up and left (when the Hext<0) compared with the curve with only garnet(red line). The measurement range increased from 820 Oe to 900 Oe.
The slope of a curve at the Hext means the slope of the tangent and it indicates the rate of change at a particular instant. The slope of Faraday rotation angle curve in the sensor with Nd2Fe14B biasing magnetic film was about 0.064, change slowly, when Hext<8Oe, while the slope of Faraday rotation angle curve with non-biasing magnetic film was about 0.68, change enormously, when Hext<8Oe. So, the Faraday rotation angles of two types of sensor as a function of Hext changed differently.
The linearity error can be got from Eq. (2). The average linearity error of garnet was about 1.42%, caused by the magnetic domain wall coercive force of garnet film, while the average linearity error of garnet with Nd2Fe14B biasing magnetic film was 0.125%. It can be seen that, when the Nd2Fe14B biasing magnetic film, Bbias was applied in the garnet film to generate an additional rotation of φ, The contrast between the peak to peak modulation in the biasing film and non-biasing film can be given by
where Iφ(θF) was the output intensity of light beam in the sensor with Nd2Fe14B biasing magnetic film, was the output intensity of light beam in the sensor with non-biasing magnetic film. For small rotations, from Eq. (3), the contrast grew gradually as the sin 2φ ≤ 1, and the biasing magnetic film played an important in improving the linearity of sensor. So, the linearity and measurement range of Faraday Effect sensor with Nd2Fe14B biasing magnetic film was better and wide than the non-biased sensor.4. Summary
In conclusion, we have developed a Faraday Effect sensor by using Ta/Nd2Fe14B/Ta biasing magnetic films. The XRD and VSM results indicated that the Nd2Fe14B biasing magnetic film had perpendicular magnetic anisotropy. For the Faraday effect sensor, the magnetic domain of garnet persist in the sensor with Nd2Fe14B biasing magnetic film, and was proposed to improve the linearity and measurement range. And this is very important for the sensor manufacturing, smaller sensor size, simple design, lightweight.
Acknowledgments
This work were supported by Chinese Academy of sciences for Key Topics in Innovation Engineering (Grant No KGCX-2-YW-150-4), Suzhou Special funds for major science and technology enterprises technological innovation (Grant No SG201022).
References and links
1. S. Saito, Y. Fujii, K. Yokoyama, J. Hamasaki, and Y. Ohno, “The laser current transformer for EHV power transmission lines,” IEEE J. Quantum Electron. 2(8), 255–259 (1966). [CrossRef]
2. P. Ripka, “Electric current sensors: a review,” Meas. Sci. Technol. 21(11), 112001 (2010). [CrossRef]
3. B. Yi, B. C. B. Chu, and K. S. Chiang, “Magneto-optical electric-current sensor with enhanced sensitivity,” Meas. Sci. Technol. 13(61–N), 63 (2002).
4. Y. N. Ning, Z. P. Wang, A. W. Palmer, K. T. V. Grattan, and D. A. Jackson, “Recent progress in optical current sensing techniques,” Rev. Sci. Instrum. 66(5), 3097–3011 (1995). [CrossRef]
5. J. G. Bai, G.-Q. Lu, and T. Lin, “Magneto-optical current sensing for applications in integrated power electronics modules,” Sens. Actuators A Phys. 109(1-2), 9–16 (2003). [CrossRef]
6. N. Itoh, Y. Yoshikawa, H. Minemoto, and S. Ishizuka, “Optical magnetic field probe sensor with high accuracy using iron garnet films,” in proceeding of the eleventh International Conference on Optical Fiber Sensors,” Advanced Sensing Photonics (Japan Society of Applied Physics, Sapporo, 1996), pp. 638–641.
7. O. Kamada, “Magneto-optical properties of (BiGdY) iron garnets for optical magnetic field sensors,” J. Appl. Phys. 79(8), 5976–5978 (1996). [CrossRef]
8. O. Kamada, H. Minemoto, and N. Itoh, “Magneto-optical properties of (BiGaY)3Fe5O12 for optical magnetic field sensors,” J. Appl. Phys. 75(10), 6801–6803 (1994). [CrossRef]
9. N. Itoh, H. Minemoto, D. Ishiko, and S. Ishizuka, “Optical magnetic field sensors with high linearity using Bi-substituted rare earth iron garnets,” IEEE Trans. Magn. 31(6), 3191–3193 (1995). [CrossRef]
10. V. Neu, A. Anane, S. Wirth, P. Xiong, S. A. Shaheen, and F. J. Cadieu, “Design optimization for a SmCo-biased colossal magnetoresistive thin film device,” J. Appl. Phys. 87(9), 5350 (2000). [CrossRef]
11. H. Guerrero, R. Perez del Real, R. Fernandez de Caleya, and G. Rosa, “Magnetic field biasing in Faraday effect sensors,” Appl. Phys. Lett. 74(24), 3702–3704 (1999). [CrossRef]
12. F. Dumas-Bouchiat, L. F. Zanini, M. Kustov, N. M. Dempsey, R. Grechishkin, K. Hasselbach, J. C. Orlianges, C. Champeaux, A. Catherinot, and D. Givord, “Thermomagnetically patterned micromagnets,” Appl. Phys. Lett. 96(10), 102511 (2010). [CrossRef]
13. M. Ohkoshi, “Formation and stability of small-size bubbles in garnet films,” J. Appl. Phys. 92(1), 370–373 (2002). [CrossRef]
14. M. Mino and H. Yamazaki, “Magnetic domain structure in thin film under alternate magnetic field,” J. Magn. Magn. Mater. 272–276, E509–E510 (2004). [CrossRef]
