- Email: [email protected]
The effect of target rotation rate on structural and morphological properties of thin garnet films fabricated by pulsed laser deposition
Optics & Laser Technology 43 (2011) 609–612
Contents lists available at ScienceDirect
Optics & Laser Technology journal homepage: www.elsevier.com/locate/optlastec
The effect of target rotation rate on structural and morphological properties of thin garnet films fabricated by pulsed laser deposition M.M. Tehranchi a,b, S.M. Hamidi a,n, A. Hasanpour c, M. Mozaffari d, J. Amighian d a
Laser and Plasma Research Institute, G.C., Shahid Beheshti University, Evin, Tehran, Iran Physics Department, G.C., Shahid Beheshti University, Evin, Tehran, Iran c Department of Physics, Payame nour University, Tehran, Iran d Department of Physics, Isfahan University, Isfahan, Iran b
a r t i c l e in fo
abstract
Article history: Received 26 May 2010 Received in revised form 16 August 2010 Accepted 31 August 2010 Available online 28 September 2010
The effect of target rotation rate on the structural and morphological properties of pulsed laser deposition grown Bi:YIG garnets is investigated. The rotation rate dependence of the surface morphology and magnetic properties of the thin films were studied using atomic force microscopy combined with a magneto-optical measurement setup. The results show that decrease in the target rotation rate can also increase the roughness, the index of refraction, and the surface skewness and can decrease Faraday rotation by an order of magnitude. & 2010 Elsevier Ltd. All rights reserved.
Keywords: Garnet thin films Pulsed laser deposition Faraday rotation
1. Introduction Bi-substituted garnet thin films have attracted much attention for magneto-optical applications such as magnetic-field sensors, optical switchers and magneto-optical isolators [1–3]. One of the best candidate techniques to fabricate garnet thin films is pulsed laser deposition (PLD) [4]. There are several factors associated with the PLD technique for fabrication of these films that have to be controlled carefully to optimize their properties. These parameters are the substrate temperature, heat treatment [5] and the frequency of substrate rotating [6]. In the past, the effects of these parameters were investigated, but the effects of target rotating frequency on the optical and magneto-optical properties of garnet films were not considered. The surface behavior of thin films is an important factor in determining the performance of optical devices and plays an important role in all optical and magneto-optical applications of thin films. In this respect, we studied the magnetic and structural characteristics of Bi:YIG films prepared by PLD at different substrate temperatures and target rotating frequencies. 2. Experimental procedure 2.1. Target preparation A stoichiometric target with the nominal composition of BiY2Fe5O12 was fabricated from Y2O3, Fe2O3 and Bi2O3 powders n
Corresponding author. E-mail address: [email protected] (S.M. Hamidi).
0030-3992/$ - see front matter & 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.optlastec.2010.08.009
by the mechanical alloying method and then solid state sintering at 850 1C for 10 h. The phase formations of the as-milled and annealed Garnet nano-powders were investigated by an X-ray diffractometer.
2.2. Thin film preparation Magnetic garnet thin films have been deposited onto gadolinium gallium garnet (Gd3Ga5O12) substrates using the third harmonic output from an Nd:YAG laser (355 nm, 6 ns pulse duration, 10 Hz repetition rate and laser fluence of 3 J/cm2), which was focused on a rotating target. The resultant plasma cloud of material was condensed onto the substrate, which was positioned directly in front of the target at a distance of around 4 cm. The substrate was held at room temperature and at 600 1C for various samples. All depositions were carried out in a pure oxygen partial pressure of 60 mbar. To investigate the effect of target rotation rate, we deposited thin films of 100 nm thickness at three different rotation rates such as 72 (1.2 Hz), 150 (2.5 Hz) and 186 (3.1 Hz) round per minute (rpm). Film composition and thickness were inferred from Rutherford backscattering spectroscopy (RBS) spectra. X-ray diffraction (XRD) scans were made to investigate the structural properties of the thin films. The magnetic hysteresis (M–H) loops of all the samples were recorded in parallel configuration (in-plane) using the magneto-optical Faraday rotation (FR) under AC magnetic field and polar magneto-optical Kerr rotation (MOKE) measurement. Optical microscopy and atomic force microscopy (AFM) were used to examine the surface quality of these films. Optical properties
610
M.M. Tehranchi et al. / Optics & Laser Technology 43 (2011) 609–612
were measured using a UV-shimadzu spectroscope in the wavelength range 400–900 nm.
3. Results and discussion 3.1. Substrate temperature effect The presence of garnet phase formations on the target was investigated by X-ray diffraction, with results shown in Fig. 1. To investigate the effect of substrate temperature, we deposited two Bi:YIG thin films on GGG substrate at room temperature (named as S11) and at 600 1C (S22). Directly after deposition, the S11 film was annealed at 700 1C for 2 h with rapid thermal annealing (RTA) and the S22 film was annealed in the deposition chamber under 200 mbar of oxygen pressure for 10 min. The XRD pattern of two samples of Bi:YIG thin films is shown in Fig. 2. The wide angle scan shown in Fig. 2 reveals only the (444) and (888) reflections of the film and substrate that are sharper in the S22 sample with substrate temperature of 600 1C. From the insets of Fig. 2(b), it can be seen that the coherence length for X-rays in the film is rather long (at least 100 nm) because the film peaks are narrow [1]. The lattice constant of the film is 12.43 A˚ for the (444) direction and 12.37 A˚ for the (888) direction, which is very similar to garnet bulk lattice constant; this means that the lattice mismatch between the film and substrate is low in the growth direction. The full width at half maximum (FWHM) of the (444) film direction is 0.131 and 0.171 for the (888) direction; see the inset of Fig. 2(b) that indicates the good growth direction of (111).
250
420
Intensity
200 150 422
100 50 15
640
400 211
221
20
25
642 521
321
30
511
35 40 45 2θ degree
444
50
800
55
60
65
Fig. 1. XRD of Bi:YIG nano-powders used as target in the experiment.
The magnetic behavior of thin films, investigated with FR measurement, indicated that sample S22 provided better-quality film with higher value of magnetization and FR than S11. As mentioned above, the best structural and magnetic properties of thin films were achieved at layer with substrate temperature of 600 1C. Thus we fixed our deposition parameters and investigated the effect of target rotation rate on the properties of laser deposited thin films.
3.2. Rotation rate effect To investigate the target rotation rate, we deposited three samples with three different frequencies (named as S1, S2 and S3). The optical and mechanical properties of Bi:YIG samples were evaluated by measuring the levels of surface roughness and optical constants (n and k). The measurement results for three types of Bi:YIG films are summarized in Table 1. It can be seen that the surface roughness increases with decrease in rotation frequency, and this is confirmed in Fig. 3, which shows threedimensional (3D) images of samples at scan rate of 1 Hz and their fast Fourier transform (FFT). The usual application of FFT filtering is to eliminate periodic noise from uniform pattern images, such as atomic level resolution structures [7]. This is useful for the studies of surface quality with its symmetric behavior. As shown in Fig. 3(a), the FFT image is symmetric, bright and mostly confined to the region close to the origin. This indicates that the lower frequency components are of the highest intensity. This fact can be seen in Fig. 3(b) and (c), as depicted in Table 1, with asymmetric behavior and larger amount of roughness. The maximum roughness, average roughness, root mean square, root mean square value of the surface departures within the sampling area and surface skewness, which is the measure of asymmetry of surface deviations about the mean plane, all increased with decrease in rotation frequency (Fig. 4). In fact, the surface skewness can be used to describe the shape of the topography height distribution of these samples. For a Gaussian surface, which has a symmetrical shape for the surface height distribution, the skewness is zero. For an asymmetric distribution of surface heights, the skewness may be negative if the distribution has a longer tail at the lower side of the mean plane or positive if the distribution has a longer tail at the upper side of the mean plane. This parameter can give some indications of the existence of spiky features. The increase in entropy with decrease in target rotation rate indicates that the nucleation and growth of pores are taking place spontaneously. Because of the formation of porous and rough coatings, the root mean square and average roughness values increased with decrease in target rotation rate. On the other hand, the real and imaginary parts of refractive index of the samples were calculated from the transmittance and absorption spectra using the Kramers–Kroning relation [8]. Table 1 Optical properties and roughness of Bi:YIG thin films with different rotating frequencies.
Fig. 2. XRD of Bi:YIG thin films at different substrate temperature: (a) room temperature and annealed at 700 1C and (b) with substrate temperature fixed at 600 1C.
Samples
S1
S2
S3
Rotating rate Maximum roughness (Sy) Average Average roughness (Sa) Root mean square (Sq) Surface skewness (Ssk) Entropy n K
186 rpm 74.66 nm 39.2 nm 5.09 nm 7.68 nm 0.1124 8.88 1.97 0.23
150 rpm 231.40 nm 93.53 nm 16.26 nm 23.91 nm 0.8695 10.55 2.43 0.17
72 rpm 284.56 nm 106.38 nm 19.23 nm 29.48 nm 1.2361 10.77 2.49 0.05
M.M. Tehranchi et al. / Optics & Laser Technology 43 (2011) 609–612
611
Fig. 3. Contact AFM image of Bi:YIG samples that are prepared at target rotating frequency: (a) 3 Hz, (b) 2.5 Hz and (c) 1.7 Hz and their fast Fourier transform.
Fig. 4. Change in surface skewness (circle) and entropy (square) as a function of rotating frequency.
The values at 635 nm are summarized in the two end rows of Table 1. Hansen and Krumme [9] presented an empirical investigation of the refractive index as a function of wavelength and bismuth substitution x in BixY3 xFe5O12. They showed that Bi:YIG exhibits a higher refractive index than pure YIG. The refractive index increases with increase in bismuth content, as well as with the presence of other elements. For example BixY3 xFe5O12 films with x¼0, 0.1 and 1:43 showed refractive indices of 2.3, 2.34 and 2.56, respectively. The refractive index of our sample with x¼1 at 635 nm must be equal to 2.4 that is on the order of our measured data. As listed in Table 1, the refractive index increased with decrease in rotating frequency, which is confirmed with entropy of samples (Fig. 5). Finally, we investigated the magnetic behavior of these thin films. For this purpose we measured the FR hysteresis loop at 635 nm (Fig. 6). Our result showed that the magnetic hysteresis loop of samples changes as follows: the ratio Mr/Ms is maximum in S2 and this coefficient decreased in S3 and S1. In addition, the magnetic coercivity field (Hc) in these samples increased with decrease in target rotation rate. Based on the relatively high coercivity field in S3, one may deduce that the magnetic moments
612
M.M. Tehranchi et al. / Optics & Laser Technology 43 (2011) 609–612
Fig. 5. Change in refractive index (circle) and roughness (square) as a function of rotating frequency.
in Table 1, the refractive indices of sample S2 are closer to real quantity of garnet thin films, which is the main reason for the best answer of this sample. These results were confirmed by the aid of MOKE measurement. In addition we show that the amount of magnetic anisotropy changed in these samples because of the change in target rotation rate. The sign and magnitude of the growth-induced anisotropy constant are determined by the same atomic parameters and interactions that determine those of the cubic anisotropy but in addition depend on substrate orientation, film composition and growth conditions such as melt composition and rate of rotation. Magnetic anisotropy can be varied either by non-magnetic ions substituted on iron sites and dodecahedral sites or by magnetically anisotropic transition metal ions [9]. In all these cases an ordering mechanism of the substituted ions takes place and is controlled by the special conditions during the growth process. The effect of a non-cubic distribution of ions has been investigated theoretically and in principle explains the occurrence of the non-cubic anisotropy. The substitution of transition metal ions on iron sites in most cases leads to much larger anisotropy than that of non-magnetic ions, with the exception of the substitution of Bi3 + ions on dodecahedral sites, which yields comparably large anisotropy values [10].
4. Conclusion In this paper, we present a study of Bi:YIG thin films deposited using the PLD technique at various substrate temperatures and frequencies of target rotation. The results show that substrate temperature plays an important role in improving the properties of garnet thin films. The speed of target rotation can affect the roughness and entropy of thin films and consequently the refractive indices and magneto-optical response of thin films. The highest value of magnetization that has been observed was in a sample deposited at 600 1C with a rotation rate of 150 rpm. References
Fig.6. Measured Faraday rotation hysteresis loop as a function of applied magnetic field.
have a tendency to be in the plane of the film. In fact, the existence of wide and deep faults on the surface of S3 causes inplane demagnetization field, which as a result partly balances the effect of the perpendicular demagnetization. The value of surface skewness larger than 1 in sample S3 indicates the existence of spiky features on the sample surface. This leads to an increase in pinning of magnetic moment and consequently decrease in the FR response. Furthermore, as listed
[1] Kahl S, Khartsev SI, Grishin AM, Kawano K, Kong G, Chakalov RA, Abell JS. J Appl Phys 2002;91(12):9556. [2] Jalili- Roudsari AA, Denyshenkov VP, Khartsev SI, Grishin AM, Adachi N, Okuda T. IEEE Trans Mag 2001;37:2454. [3] Zhao W. Sensors Actuators A 2001;89:250. [4] Laulajainen M, Paturi P, Raittila J, Huhtinen H, Abrahamsen AB, Andersen NH, Laiho R. J Magn Magn Mater 2004;279:218. [5] Kumar N, Prasad S, Misra DS, Venkataramani N, Bohra M, Krishnan R. J Magn Magn Mater 2008;320:2233. [6] Rojas R, Krafft C, Nistor I, Zhang D, Mayergoyz ID. J Appl Phys 2004;95(11):6885. [7] User’s guide to autoprobe CP and M5, Part III, PSI ProScan Software Version 1.5; 1998 [Chapter 2]. [8] M. Mozaffari, A. Hasanpour, J. Amighian, S. M. Hamidi, M. M. Tehranchi and M. Ghanaatshoar. Proceedings of the 1st international symposium on functional materials, Hilton Kuala Lumpur, Malaysia, 2005. [9] Hansen P, Krumme J-P. Thin Solid Films 1984;114:69. [10] Hamidi SM, Tehranchi MM. J Magn Magn Mater 2010, submitted for publication.
Contents lists available at ScienceDirect
Optics & Laser Technology journal homepage: www.elsevier.com/locate/optlastec
The effect of target rotation rate on structural and morphological properties of thin garnet films fabricated by pulsed laser deposition M.M. Tehranchi a,b, S.M. Hamidi a,n, A. Hasanpour c, M. Mozaffari d, J. Amighian d a
Laser and Plasma Research Institute, G.C., Shahid Beheshti University, Evin, Tehran, Iran Physics Department, G.C., Shahid Beheshti University, Evin, Tehran, Iran c Department of Physics, Payame nour University, Tehran, Iran d Department of Physics, Isfahan University, Isfahan, Iran b
a r t i c l e in fo
abstract
Article history: Received 26 May 2010 Received in revised form 16 August 2010 Accepted 31 August 2010 Available online 28 September 2010
The effect of target rotation rate on the structural and morphological properties of pulsed laser deposition grown Bi:YIG garnets is investigated. The rotation rate dependence of the surface morphology and magnetic properties of the thin films were studied using atomic force microscopy combined with a magneto-optical measurement setup. The results show that decrease in the target rotation rate can also increase the roughness, the index of refraction, and the surface skewness and can decrease Faraday rotation by an order of magnitude. & 2010 Elsevier Ltd. All rights reserved.
Keywords: Garnet thin films Pulsed laser deposition Faraday rotation
1. Introduction Bi-substituted garnet thin films have attracted much attention for magneto-optical applications such as magnetic-field sensors, optical switchers and magneto-optical isolators [1–3]. One of the best candidate techniques to fabricate garnet thin films is pulsed laser deposition (PLD) [4]. There are several factors associated with the PLD technique for fabrication of these films that have to be controlled carefully to optimize their properties. These parameters are the substrate temperature, heat treatment [5] and the frequency of substrate rotating [6]. In the past, the effects of these parameters were investigated, but the effects of target rotating frequency on the optical and magneto-optical properties of garnet films were not considered. The surface behavior of thin films is an important factor in determining the performance of optical devices and plays an important role in all optical and magneto-optical applications of thin films. In this respect, we studied the magnetic and structural characteristics of Bi:YIG films prepared by PLD at different substrate temperatures and target rotating frequencies. 2. Experimental procedure 2.1. Target preparation A stoichiometric target with the nominal composition of BiY2Fe5O12 was fabricated from Y2O3, Fe2O3 and Bi2O3 powders n
Corresponding author. E-mail address: [email protected] (S.M. Hamidi).
0030-3992/$ - see front matter & 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.optlastec.2010.08.009
by the mechanical alloying method and then solid state sintering at 850 1C for 10 h. The phase formations of the as-milled and annealed Garnet nano-powders were investigated by an X-ray diffractometer.
2.2. Thin film preparation Magnetic garnet thin films have been deposited onto gadolinium gallium garnet (Gd3Ga5O12) substrates using the third harmonic output from an Nd:YAG laser (355 nm, 6 ns pulse duration, 10 Hz repetition rate and laser fluence of 3 J/cm2), which was focused on a rotating target. The resultant plasma cloud of material was condensed onto the substrate, which was positioned directly in front of the target at a distance of around 4 cm. The substrate was held at room temperature and at 600 1C for various samples. All depositions were carried out in a pure oxygen partial pressure of 60 mbar. To investigate the effect of target rotation rate, we deposited thin films of 100 nm thickness at three different rotation rates such as 72 (1.2 Hz), 150 (2.5 Hz) and 186 (3.1 Hz) round per minute (rpm). Film composition and thickness were inferred from Rutherford backscattering spectroscopy (RBS) spectra. X-ray diffraction (XRD) scans were made to investigate the structural properties of the thin films. The magnetic hysteresis (M–H) loops of all the samples were recorded in parallel configuration (in-plane) using the magneto-optical Faraday rotation (FR) under AC magnetic field and polar magneto-optical Kerr rotation (MOKE) measurement. Optical microscopy and atomic force microscopy (AFM) were used to examine the surface quality of these films. Optical properties
610
M.M. Tehranchi et al. / Optics & Laser Technology 43 (2011) 609–612
were measured using a UV-shimadzu spectroscope in the wavelength range 400–900 nm.
3. Results and discussion 3.1. Substrate temperature effect The presence of garnet phase formations on the target was investigated by X-ray diffraction, with results shown in Fig. 1. To investigate the effect of substrate temperature, we deposited two Bi:YIG thin films on GGG substrate at room temperature (named as S11) and at 600 1C (S22). Directly after deposition, the S11 film was annealed at 700 1C for 2 h with rapid thermal annealing (RTA) and the S22 film was annealed in the deposition chamber under 200 mbar of oxygen pressure for 10 min. The XRD pattern of two samples of Bi:YIG thin films is shown in Fig. 2. The wide angle scan shown in Fig. 2 reveals only the (444) and (888) reflections of the film and substrate that are sharper in the S22 sample with substrate temperature of 600 1C. From the insets of Fig. 2(b), it can be seen that the coherence length for X-rays in the film is rather long (at least 100 nm) because the film peaks are narrow [1]. The lattice constant of the film is 12.43 A˚ for the (444) direction and 12.37 A˚ for the (888) direction, which is very similar to garnet bulk lattice constant; this means that the lattice mismatch between the film and substrate is low in the growth direction. The full width at half maximum (FWHM) of the (444) film direction is 0.131 and 0.171 for the (888) direction; see the inset of Fig. 2(b) that indicates the good growth direction of (111).
250
420
Intensity
200 150 422
100 50 15
640
400 211
221
20
25
642 521
321
30
511
35 40 45 2θ degree
444
50
800
55
60
65
Fig. 1. XRD of Bi:YIG nano-powders used as target in the experiment.
The magnetic behavior of thin films, investigated with FR measurement, indicated that sample S22 provided better-quality film with higher value of magnetization and FR than S11. As mentioned above, the best structural and magnetic properties of thin films were achieved at layer with substrate temperature of 600 1C. Thus we fixed our deposition parameters and investigated the effect of target rotation rate on the properties of laser deposited thin films.
3.2. Rotation rate effect To investigate the target rotation rate, we deposited three samples with three different frequencies (named as S1, S2 and S3). The optical and mechanical properties of Bi:YIG samples were evaluated by measuring the levels of surface roughness and optical constants (n and k). The measurement results for three types of Bi:YIG films are summarized in Table 1. It can be seen that the surface roughness increases with decrease in rotation frequency, and this is confirmed in Fig. 3, which shows threedimensional (3D) images of samples at scan rate of 1 Hz and their fast Fourier transform (FFT). The usual application of FFT filtering is to eliminate periodic noise from uniform pattern images, such as atomic level resolution structures [7]. This is useful for the studies of surface quality with its symmetric behavior. As shown in Fig. 3(a), the FFT image is symmetric, bright and mostly confined to the region close to the origin. This indicates that the lower frequency components are of the highest intensity. This fact can be seen in Fig. 3(b) and (c), as depicted in Table 1, with asymmetric behavior and larger amount of roughness. The maximum roughness, average roughness, root mean square, root mean square value of the surface departures within the sampling area and surface skewness, which is the measure of asymmetry of surface deviations about the mean plane, all increased with decrease in rotation frequency (Fig. 4). In fact, the surface skewness can be used to describe the shape of the topography height distribution of these samples. For a Gaussian surface, which has a symmetrical shape for the surface height distribution, the skewness is zero. For an asymmetric distribution of surface heights, the skewness may be negative if the distribution has a longer tail at the lower side of the mean plane or positive if the distribution has a longer tail at the upper side of the mean plane. This parameter can give some indications of the existence of spiky features. The increase in entropy with decrease in target rotation rate indicates that the nucleation and growth of pores are taking place spontaneously. Because of the formation of porous and rough coatings, the root mean square and average roughness values increased with decrease in target rotation rate. On the other hand, the real and imaginary parts of refractive index of the samples were calculated from the transmittance and absorption spectra using the Kramers–Kroning relation [8]. Table 1 Optical properties and roughness of Bi:YIG thin films with different rotating frequencies.
Fig. 2. XRD of Bi:YIG thin films at different substrate temperature: (a) room temperature and annealed at 700 1C and (b) with substrate temperature fixed at 600 1C.
Samples
S1
S2
S3
Rotating rate Maximum roughness (Sy) Average Average roughness (Sa) Root mean square (Sq) Surface skewness (Ssk) Entropy n K
186 rpm 74.66 nm 39.2 nm 5.09 nm 7.68 nm 0.1124 8.88 1.97 0.23
150 rpm 231.40 nm 93.53 nm 16.26 nm 23.91 nm 0.8695 10.55 2.43 0.17
72 rpm 284.56 nm 106.38 nm 19.23 nm 29.48 nm 1.2361 10.77 2.49 0.05
M.M. Tehranchi et al. / Optics & Laser Technology 43 (2011) 609–612
611
Fig. 3. Contact AFM image of Bi:YIG samples that are prepared at target rotating frequency: (a) 3 Hz, (b) 2.5 Hz and (c) 1.7 Hz and their fast Fourier transform.
Fig. 4. Change in surface skewness (circle) and entropy (square) as a function of rotating frequency.
The values at 635 nm are summarized in the two end rows of Table 1. Hansen and Krumme [9] presented an empirical investigation of the refractive index as a function of wavelength and bismuth substitution x in BixY3 xFe5O12. They showed that Bi:YIG exhibits a higher refractive index than pure YIG. The refractive index increases with increase in bismuth content, as well as with the presence of other elements. For example BixY3 xFe5O12 films with x¼0, 0.1 and 1:43 showed refractive indices of 2.3, 2.34 and 2.56, respectively. The refractive index of our sample with x¼1 at 635 nm must be equal to 2.4 that is on the order of our measured data. As listed in Table 1, the refractive index increased with decrease in rotating frequency, which is confirmed with entropy of samples (Fig. 5). Finally, we investigated the magnetic behavior of these thin films. For this purpose we measured the FR hysteresis loop at 635 nm (Fig. 6). Our result showed that the magnetic hysteresis loop of samples changes as follows: the ratio Mr/Ms is maximum in S2 and this coefficient decreased in S3 and S1. In addition, the magnetic coercivity field (Hc) in these samples increased with decrease in target rotation rate. Based on the relatively high coercivity field in S3, one may deduce that the magnetic moments
612
M.M. Tehranchi et al. / Optics & Laser Technology 43 (2011) 609–612
Fig. 5. Change in refractive index (circle) and roughness (square) as a function of rotating frequency.
in Table 1, the refractive indices of sample S2 are closer to real quantity of garnet thin films, which is the main reason for the best answer of this sample. These results were confirmed by the aid of MOKE measurement. In addition we show that the amount of magnetic anisotropy changed in these samples because of the change in target rotation rate. The sign and magnitude of the growth-induced anisotropy constant are determined by the same atomic parameters and interactions that determine those of the cubic anisotropy but in addition depend on substrate orientation, film composition and growth conditions such as melt composition and rate of rotation. Magnetic anisotropy can be varied either by non-magnetic ions substituted on iron sites and dodecahedral sites or by magnetically anisotropic transition metal ions [9]. In all these cases an ordering mechanism of the substituted ions takes place and is controlled by the special conditions during the growth process. The effect of a non-cubic distribution of ions has been investigated theoretically and in principle explains the occurrence of the non-cubic anisotropy. The substitution of transition metal ions on iron sites in most cases leads to much larger anisotropy than that of non-magnetic ions, with the exception of the substitution of Bi3 + ions on dodecahedral sites, which yields comparably large anisotropy values [10].
4. Conclusion In this paper, we present a study of Bi:YIG thin films deposited using the PLD technique at various substrate temperatures and frequencies of target rotation. The results show that substrate temperature plays an important role in improving the properties of garnet thin films. The speed of target rotation can affect the roughness and entropy of thin films and consequently the refractive indices and magneto-optical response of thin films. The highest value of magnetization that has been observed was in a sample deposited at 600 1C with a rotation rate of 150 rpm. References
Fig.6. Measured Faraday rotation hysteresis loop as a function of applied magnetic field.
have a tendency to be in the plane of the film. In fact, the existence of wide and deep faults on the surface of S3 causes inplane demagnetization field, which as a result partly balances the effect of the perpendicular demagnetization. The value of surface skewness larger than 1 in sample S3 indicates the existence of spiky features on the sample surface. This leads to an increase in pinning of magnetic moment and consequently decrease in the FR response. Furthermore, as listed
[1] Kahl S, Khartsev SI, Grishin AM, Kawano K, Kong G, Chakalov RA, Abell JS. J Appl Phys 2002;91(12):9556. [2] Jalili- Roudsari AA, Denyshenkov VP, Khartsev SI, Grishin AM, Adachi N, Okuda T. IEEE Trans Mag 2001;37:2454. [3] Zhao W. Sensors Actuators A 2001;89:250. [4] Laulajainen M, Paturi P, Raittila J, Huhtinen H, Abrahamsen AB, Andersen NH, Laiho R. J Magn Magn Mater 2004;279:218. [5] Kumar N, Prasad S, Misra DS, Venkataramani N, Bohra M, Krishnan R. J Magn Magn Mater 2008;320:2233. [6] Rojas R, Krafft C, Nistor I, Zhang D, Mayergoyz ID. J Appl Phys 2004;95(11):6885. [7] User’s guide to autoprobe CP and M5, Part III, PSI ProScan Software Version 1.5; 1998 [Chapter 2]. [8] M. Mozaffari, A. Hasanpour, J. Amighian, S. M. Hamidi, M. M. Tehranchi and M. Ghanaatshoar. Proceedings of the 1st international symposium on functional materials, Hilton Kuala Lumpur, Malaysia, 2005. [9] Hansen P, Krumme J-P. Thin Solid Films 1984;114:69. [10] Hamidi SM, Tehranchi MM. J Magn Magn Mater 2010, submitted for publication.