Effects of oxygen gas pressure on properties of iron oxide films grown by pulsed laser deposition

Effects of oxygen gas pressure on properties of iron oxide films grown by pulsed laser deposition

Journal of Alloys and Compounds 552 (2013) 1–5 Contents lists available at SciVerse ScienceDirect Journal of Alloys and Compounds journal homepage: ...

790KB Sizes 0 Downloads 39 Views

Journal of Alloys and Compounds 552 (2013) 1–5

Contents lists available at SciVerse ScienceDirect

Journal of Alloys and Compounds journal homepage: www.elsevier.com/locate/jalcom

Effects of oxygen gas pressure on properties of iron oxide films grown by pulsed laser deposition Qixin Guo a,c,⇑, Wangzhou Shi a, Feng Liu a, Makoto Arita b, Yoshifumi Ikoma b, Katsuhiko Saito c, Tooru Tanaka c, Mitsuhiro Nishio c a b c

Department of Physics, Shanghai Normal University, Shanghai 200234, China Department of Materials Science and Engineering, Kyushu University, Fukuoka 819-0395, Japan Synchrotron Light Application Center, Department of Electrical and Electronic Engineering, Saga University, Saga 840-8502, Japan

a r t i c l e

i n f o

Article history: Received 4 September 2012 Received in revised form 12 October 2012 Accepted 18 October 2012 Available online 27 October 2012 Keywords: Magnetite films Pulsed laser deposition Oxygen gas pressure X-ray absorption fine structure Magnetic properties

a b s t r a c t Iron oxide films were grown on sapphire substrates by pulsed laser deposition at oxygen gas pressures between 1  105 and 1  101 Pa with a substrate temperature of 600 °C. Atomic force microscope, X-ray diffraction, Raman spectroscopy, X-ray absorption fine structure, and vibrational sample magnetometer analysis revealed that surface morphology and crystal structure of the iron oxide films strongly depend on the oxygen gas pressure during the growth and the optimum oxygen gas pressure range is very narrow around 1  103 Pa for obtaining single phase magnetite films with high crystal quality. Ó 2012 Elsevier B.V. All rights reserved.

1. Introduction Magnetite has attracted enormous attention because of its half metallic ferromagnetic nature, high Curie temperature, and the presence of a metal–insulator transition [1–3]. These properties make it a promising material for applications in spin electronic devices such as magnetic tunnel junctions for magnetic random access memory [4]. Many growth methods such as electron beam deposition [4], molecular beam epitaxy [5–7], and sputtering [8–11], have been used to fabricate magnetite films. The magnetite film on sapphire system is an important candidate for heteroepitaxial magnetic tunneling junction structures which exhibit giant tunnel magnetroresistance [12]. Pulsed laser deposition (PLD) is a promising technique for preparing thin films due to its advantages which include stoichiometric transfer, growth from an energetic beam, reactive deposition, and inherent simplicity for the growth of multilayered structures [13]. Especially, for growing nanometer order thin films, PLD is one of the most advanced techniques because the growth can be controlled precisely. In PLD process for growing iron oxide films, oxygen gas pressure is one of the most important parameters affecting on thin film properties. Parames et

⇑ Corresponding author at: Synchrotron Light Application Center, Department of Electrical and Electronic Engineering, Saga University, Saga 840-8502, Japan. Tel.: +81 952 28 8500; fax: +81 952 28 8651. E-mail address: [email protected] (Q. Guo). 0925-8388/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jallcom.2012.10.088

al. [14] tried to grow magnetite thin films on sapphire substrates by PLD in reactive atmospheres of oxygen and argon, at working pressure of 8  102 Pa. They have investigated the microstructure and magnetic properties of the obtained magnetite films and have reported the saturation magnetization of the magnetite film is about 315 emu/cm3, which is much smaller than the value for bulk magnetite. In this article, we present on the effects of oxygen pressure on the properties of the iron oxide films in PLD because a deeper knowledge on the growth of the magnetite thin films with high quality is essential for device applications. We have succeeded in fabrication of magnetite film with saturation magnetic moment close to the value of bulk magnetite. 2. Experimental Iron oxide films were grown on c-plane sapphire substrates by pulsed laser deposition in atmosphere of oxygen. The sapphire substrates were cleaned ultrasonically in organic solvents, chemically etched in a hot H3PO4:H2SO4 (1:3) solution, rinsed in deionized water, and blown dry in nitrogen gas before they were introduced into the growth chamber. Pulsed KrF excimer laser at a wavelength of 248 nm with a repetition rate of 2 Hz and a energy density of 2 J/cm2 at the target surface was used to ablate magnetite target (99.9%). The substrate to target distance was 50 mm. High purity oxygen gas were introduced through mass flow controllers after the growth chamber was evacuated below 1  105 Pa using a turbomolecular pump. The target was rotated during the growth to avoid crater formation. The substrate temperature was kept at 600 °C while the oxygen gas pressure was controlled at a given value between 1  105 and 1  101 Pa. After the growth, thickness of the iron oxide films was determined by a surface step profile analyzer. The thickness of these films varies from 44.5 to 80.4 nm. Since the thickness of the film

Q. Guo et al. / Journal of Alloys and Compounds 552 (2013) 1–5

102

1.4 1.2 1.0 0.8

Roughness (nm)

Growth Rate (nm/min.)

2

10

0.6

10-6 10-5 10-4 10-3 10-2 10-1 Pressure (Pa) Fig. 1. Dependence of the growth rate of the iron oxide films on oxygen gas pressure.

grown at a given oxygen gas pressure increased linearly with increasing the growth time, we determined growth rate simply dividing film thickness by growth time. The crystallographic structure of the iron oxide films was analyzed by X-ray diffraction (XRD) using Ka emission line of copper. Raman measurements were performed on a micro-Raman system equipped with a classic charge-coupled device detector. The Raman scattering was recorded in the backscattering geometry of the zðx; Þz configuration using an Ar ion laser at 488 nm. The surface morphology and roughness were studied by atomic force microscope (AFM) on 10  10 lm2 areas under ambient conditions. X-ray absorption fine structure (XAFS) experiments were carried out at beam line BL11 of Saga Light Source with a Si (1 1 1) double crystal monochromator using synchrotron radiation. The XAFS spectra were collected by recording the conversion electron yield at room temperature. Vibrational sample magnetometer (VSM) was used to investigate the magnetic properties of the iron oxide films.

1

10-6 10-5 10-4 10-3 10-2 10-1

Pressure (Pa) Fig. 3. Dependence of the surface roughness of the iron oxide on oxygen gas pressure.

3. Results and discussion Fig. 1 shows the dependence of the growth rate of the iron oxide films on oxygen gas pressure. The growth rate remains almost constant when the pressure is increased to 1  103 Pa, then decreases with further increasing pressure. The largest growth rate at the pressure of 1  103 Pa is observed to be 1.34 nm/min, suggesting that the thickness of the grown films can be controlled precisely in our PLD growth system. Fig. 2 presents AFM images of the iron oxide films grown at various oxygen gas pressures. The root mean square (RMS) surface

Fig. 2. AFM images of the iron oxide films grown at different oxygen gas pressures: (a) 1  105, (b) 1  103, (c) 1  102 and (d) 1  102 Pa.

3

(410) Fe3O4

(006)

(111) FeO

-Al2O3

Q. Guo et al. / Journal of Alloys and Compounds 552 (2013) 1–5

(a)

Intensity (arb. units)

(b)

(c)

(c)

(006) Fe2O3 30

(333) Fe3O4

(b)

(006) Fe2O3

Intensity (arb. units)

(222) Fe3O4

(a)

(d)

(d) 40

50 2 (deg.)

60

200

400 600 800 Raman Shift (cm-1)

1000

Fig. 4. XRD patterns of the iron oxide films grown at different oxygen gas pressures: (a) 1  105, (b) 1  103, (c) 1  102 and (d) 1  101 Pa.

Fig. 5. Raman spectra of the iron oxide films grown at different oxygen gas pressures: (a) 1  105, (b) 1  103, (c) 1  102 and (d) 1  101 Pa.

roughness of the iron oxide films obtained from Fig. 2 is shown in Fig. 3 as a function of the oxygen gas pressure. The surface roughness of the film grown at the pressure of 1  105 Pa is about 54.4 nm. With increasing oxygen gas pressure, however, the surface roughness of the films dramatically decreases to the order of a few nanometers. Fig. 4 shows the h  2h XRD patterns of the iron oxide films as a function of oxygen gas pressures. For the sample grown at low oxygen gas pressure (1  105 Pa), the peaks related to (1 1 1) wüstite (FeO) and (4 1 0) magnetite (Fe3O4) planes are observed in Fig. 4(a), suggesting both wüstite and magnetite phases exist in the obtained films. When the oxygen gas pressure is increased to 1  103 Pa, only two diffraction peaks related to the (2 2 2) and (3 3 3) reflections from magnetite are clearly observed as shown in Fig. 4(b). As the gas pressure is further increased, however, only (0 0 6) hematite (Fe2O3) plane are observed (Fig. 4(c) and (d)) while the peaks related to the (2 2 2) and (3 3 3) reflections from magnetite phase completely disappear. It is well known that Raman spectroscopy is a powerful technique for characterizing iron oxide phases because it is highly sensitive to the chemical composition of crystalline materials [15]. Thus, we performed Raman spectroscopy measurements in order to further confirm the crystal structure of the iron oxide films. Fig. 5 shows the Raman spectra obtained at room temperature for the iron oxide films grown at different oxygen gas pressures. For comparison, the observed Raman peaks from the iron oxides

films grown at different oxygen gas pressures from Fig. 5 and phonon frequencies of magnetite, hematite, and wüstite reported in the literature [15] are summarized in Table 1. The Raman peak observed at 662 cm1 from the sample grown at the oxygen gas pressure 1  105 Pa (Fig. 5(a)) is related with magnetite. With increasing the oxygen gas pressure to 1  103 Pa, the intensity of this Raman peak increases and other Raman peaks related with magnetite exhibit at 309 and 533 cm1 (Fig. 5(b)). As the oxygen gas pressure was further increased, however, the Raman peaks related with magnetite phase disappear and the peaks related with hematite phase emerge at 225, 298, 409, and 609 cm1 as shown in Fig. 5(c) and (d). We note that the peak appeared at 418 cm1 is ascribed to the sapphire substrate. The results are consistent with the XRD analysis, suggesting that the crystal structure of the iron oxide films strongly depends on the oxygen gas pressure during the growth. Next, we studied the local structure around the Fe atoms of the iron oxide films by XAFS. Fig. 6 shows the Fe K edge XAFS spectra acquired from the iron oxide films grown at different oxygen gas pressures. All spectra exhibit similar sharp rise with a main structure B accompanied by a pre-edge peak A. The pre-edge peak A is ascribed to the Fe 1s–3d electric dipole-forbidden transition. This weak-intensity peak is known to be a consequence of the Fe 3d– 4p orbital mixing [16]. From Fig. 6, it is clear that both the pre-edge peak and the main peak shift toward higher energy with the increase of oxygen gas pressure during the growth of the films. It

4

Q. Guo et al. / Journal of Alloys and Compounds 552 (2013) 1–5

Table 1 Observed Raman peaks from the iron oxide films grown at different oxygen gas pressures and phonon frequencies of magnetite, hematite, and wüstite reported in literatures [15]. Observed Raman peaks (cm1) 1  10

5

Reported phonon frequencies (cm1) 3

Pa

1  10

– – – – – – – 662

Pa

– – 309 – 533 – – 662

1  10

2

1

Pa

1  10

225 298 – 409 – 609 – –

225 298 – 409 – 609 – –

B (a) A

Intensity (arb. units)

(b)

(c)

(d)

7110

7140 Energy (eV)

7170

Fig. 6. Fe K edge XAFS spectra of the iron oxide films grown at different oxygen gas pressures: (a) 1  105, (b) 1  103, (c) 1  102 and (d) 1  101 Pa.

has been reported that the pre-edge feature is sensitive to Fe redox states in minerals and silicate melts [17]. Wilke et al. [17] have measured Fe K-edge XAFS spectra of a series of crystalline Fe2+- and Fe3+-bearing model compounds in order to correlate characteristics of the pre-edge feature with oxidation state and local coordination environment of Fe atoms. They found that the position of centroid of the pre-edge peak is very useful for determining Fe oxidation state. It is well known that all Fe3+ ions are located in a slightly distorted octahedral environment in the hexagonal corundum structure of hematite while one-third of the iron ions are located on tetrahedral sites (all Fe3+; filled bonds) and

Pa

Magnetite

Hematite

Wüstite

– – 306 – – – – 668

226 229 – 411 – 612 – –

– – – – – – 652 –

two-third are located on octahedral sites (one-half being Fe2+ and the other half Fe3+) in the cubic inverse spinel structure of magnetite [18]. Thus, we believe that the shift observed in Fig. 6 is due to the variation of Fe oxidation state from wüstite (Fe2+) and magnetite (Fe2+/Fe3+) (1  105 Pa) to magnetite (Fe2+/Fe3+) (1  103 Pa) and then hematite (Fe3+) (1  102, 1  101 Pa) with the increase of the oxygen gas pressure during the growth, which is consistent with the observed XRD and Raman data. Magnetic hysteresis loops of the iron oxide films grown at different oxygen gas pressures are shown in Fig. 7. The external magnetic field was applied parallel to the plane of the sapphire substrate. From Fig. 7, we find that the saturation magnetic moment of the single phase magnetite thin film grown at the oxygen gas pressure of 1  103 Pa is 492 ± 10 emu/cm3, which is in good agreement with the value reported for bulk magnetite [6]. Since the physical properties such as magnetic behavior of magnetite is strongly affected due to the presence of antiphase boundaries, which are growth defects resulting from the nucleation of islands which corresponding to a stacking fault in the iron cation sublattice [7], the observed bulk-like magnetic properties from our sample suggests that the magnetite film grown at the gas pressure of 1  103 Pa is of high crystal quality without antiphase boundaries. With increasing or decreasing the gas pressure from 1  103 Pa, the magnetic moment of the films dramatically decreases as shown in Fig. 7(a), (c), and (d) due to the properties of hematite and wüstite phases in the films. Here, it is worth to note that Parames et al. have used same PLD technique to grow magnetite thin films on sapphire substrates in reactive atmospheres of oxygen and argon at oxygen pressures of 4.8  102 Pa and 6.4  102 Pa [14]. The best saturation magnetic moment of their magnetite films has been reported to be 315 emu/cm3 which is much smaller than our value of 492 ± 10 emu/cm3 obtained in this work. Based on the experimental results, we conclude that the crystal structure of the iron oxide films strongly depends on the oxygen gas pressure during the growth and the optimum oxygen gas pressure range is very narrow around 1  103 Pa for obtaining single phase magnetite films with high crystal quality. This phenomenon can be explained as follows: growth of the oxide film is limited by the arrival rate of the iron and oxygen atoms, desorption rate and reaction rate of these atoms at the substrate in the pulsed laser deposition of iron oxide film using magnetite target with oxygen gas. At low gas pressure (1  105 Pa), there is no sufficient oxygen atoms at the substrate for forming magnetite phase. The rough surface is formed because both wüstite and magnetite are grown on the sapphire substrate with random orientation. With increasing the oxygen gas pressure to 1  103 Pa, the oxygen atoms at the substrate increases to the optimal number for forming magnetite films. Surface roughness of the film is considerably reduced since magnetite phase having a cubic system preferentially grows on the sapphire (0 0 1) surface along [1 1 1] direction. When the oxygen gas pressure is further increased, however, the oxygen atoms are surplus to form magnetite phase, which lead to the growth of hematite

5

Q. Guo et al. / Journal of Alloys and Compounds 552 (2013) 1–5

500

(a) M (emu/cc)

M (emu/cc)

500

0

-8000 -4000

0 4000 H (Oe)

-8000 -4000

8000

500

(c)

M (emu/cc)

M (emu/cc)

0

-500

-500

500

(b)

0

0 4000 H (Oe)

8000

(d)

0

-500

-500 -8000 -4000

0 4000 H (Oe)

-8000 -4000

8000

0 4000 H (Oe)

8000

Fig. 7. Magnetic hysteresis loops of the iron oxide films grown at different oxygen gas pressures: (a) 1  105, (b) 1  103, (c) 1  102 and (d) 1  101 Pa.

phase with the same crystal structure of sapphire at relatively smooth surface. 4. Conclusions We have grown iron oxide films on sapphire substrates by pulsed laser deposition at oxygen gas pressures between 1  105 and 1  101 Pa with a substrate temperature of 600 °C. By AFM, XRD, Raman spectroscopy, XAFS, and VSM analysis, we revealed that surface morphology and crystal structure of the iron oxide films strongly depend on the oxygen gas pressure during the growth and the optimum oxygen gas pressure range is very narrow around 1  103 Pa for obtaining single phase magnetite films with high crystal quality.

References [1] [2] [3] [4] [5] [6] [7]

[8] [9] [10] [11]

Acknowledgements

[12] [13]

The authors are grateful to the staff of the Saga Light Source for their useful advice. Thanks also go to Y. Mitsuishi, K. Nakamura, and T. Konomi for their help in the sample preparation and measurement. This work was partially supported by the Scientific Research (No. 23560369) and the Partnership Project for Fundamental Technology Researches of the Ministry of Education, Culture, Sports, Science and Technology, Japan.

[14] [15] [16] [17] [18]

Z. Zhang, S. Satpathy, Phys. Rev. B 44 (1991) 13319. Y. Yanase, K. Siratori, J. Phys. Soc. Jpn. 53 (1984) 312. E.J.W. Verwey, Nature 144 (1939) 327. S. Jain, A.O. Adeyeye, C.B. Boothroyd, J. Appl. Phys. 97 (2005) 093713. S. Arora, H. Wu, R. Choudhary, I. Shvets, O. Mryasov, H. Yao, W. Ching, Phys. Rev. B 77 (2008) 134443. M. Ferhat, K. Yoh, Appl. Phys. Lett. 90 (2007) 112501. J. Moussy, S. Gota, A. Bataille, M. Guittet, M. Gautier-Soyer, F. Delille, B. Dieny, F. Ott, T. Doan, P. Warin, P. Bayle-Guillemaud, C. Gatel, E. Snoeck, Phys. Rev. B 70 (2004) 174448. H. Liu, E. Jian, H. Bai, R. Zheng, H. Wei, X. Zhang, Appl. Phys. Lett. 83 (2003) 3531. J. Hong, S. Lee, Y. Jung, J. Lee, K. Yoon, K. Kim, C. Kim, C. Lee, Appl. Phys. Lett. 83 (2003) 1590. Y. Peng, C. Park, D. Laughlin, J. Appl. Phys. 93 (2003) 7957. S. Soeya, J. Hayakawa, H. Takahashi, K. Ito, C. Yamamoto, A. Kida, H. Asano, M. Matsui, Appl. Phys. Lett. 80 (2002) 823. F. Schedin, E. Hill, G. Laan, G. Thornton, J. Appl. Phys. 96 (2004) 1165. D.J. Lowndes, D.B. Geohegan, A.A. Puretzky, D.P. Norton, C.M. Rouleau, Science 273 (1996) 898. M.L. Parames, Z. Viskadourakis, M.S. Rogalski, J. Mariano, N. Popovici, J. Giapintzakis, O. Conde, Appl. Surf. Sci. 253 (2007) 8201. M. Lubbe, A. Gigler, R. Stark, W. Moritz, Surf. Sci. 604 (2010) 679. T. Westre, P. Kennepohl, J. DeWitt, B. Hedman, K. Hodgson, E. Solomon, J. Am. Chem. Soc. 119 (1977) 6297. M. Wilke, F. Farges, P. Petit, G. Brown Jr., F. Martin, Am. Mineral. 86 (2001) 714. Th. Schedel-Niedrig, W. Weiss, R. Schlogl, Phys. Rev. B 52 (1995) 17449.