Using APL format - Wayne State University Physics and Astronomy

6 downloads 582 Views 51KB Size Report
bubbler, automated valves, a pressure controller, mass flow controllers, and precursor flow ... Maine, 5737 Jenness Hall, Orono, Maine 04469; electronic mail:.
APPLIED PHYSICS LETTERS

VOLUME 76, NUMBER 25

19 JUNE 2000

Highly spin-polarized chromium dioxide thin films prepared by chemical vapor deposition from chromyl chloride W. J. DeSisto,a) P. R. Broussard, T. F. Ambrose, B. E. Nadgorny, and M. S. Osofsky Naval Research Laboratory, Washington, DC 20375-5347

共Received 4 January 2000; accepted for publication 25 April 2000兲 Highly spin-polarized chromium dioxide (CrO2) thin films were deposited on 共100兲 TiO2 substrates by chemical vapor deposition using chromyl chloride as a precursor. The spin polarization, as measured by the point contact Andreev reflection technique, was 81⫾3%. X-ray diffraction ␪/2␪ scans indicated the films grew completely 共100兲 oriented, in registry with the 共100兲 oriented TiO2 substrate. X-ray diffraction ␾ scans on the CrO2 共110兲 reflection indicated the expected twofold symmetry, with no evidence of misaligned material. The resistivity at room temperature was 240 ␮⍀ cm and decreased to 10 ␮⍀ cm at 5 K, consistent with metallic behavior. The films were ferromagnetic with a Curie temperature of 395 K and a coercivity of ⬃100 Oe at 298 K. The use of chromyl chloride as a precursor resulted in efficient and controlled CrO2 film growth. © 2000 American Institute of Physics. 关S0003-6951共00兲04125-5兴

Spin polarized transport effects in materials has become an important and rapidly developing area of basic research and technology. It is expected that this new field, known as magnetoelectronics, will spur the development of new devices which cannot be realized with existing semiconductorbased electronics.1 A central component of these devices are ferromagnetic materials that provide current with a high degree of spin polarization 共ideally 100%兲. Chromium dioxide (CrO2) is one such material that has been predicted to be 100% spin polarized at the Fermi level based upon band structure calculations.2 Recent point contact experiments have indicated the spin polarization in CrO2 approaches 100%.3 Ultrathin layers of highly spin-polarized CrO2 have potential applications in giant magnetoresistance 共GMR兲 devices. Therefore it is important to develop efficient and controlled methods for preparing CrO2 films. Chromium dioxide (CrO2) is a ferromagnet with a Curie temperature of ⬃395 K that crystallizes with the rutile structure 共tetragonal, P42 /mnm兲. The fabrication of single-phase thin films of CrO2 is often difficult, requiring epitaxial growth on appropriate substrates. Furthermore, chromium forms many oxides including CrO3, Cr2O5, CrO2, and Cr2O3, with Cr2O3 the most stable. CrO2 has been shown to irreversibly reduce to Cr2O3 at temperatures between 250 and 460 °C4–6 placing clear temperature constraints on the growth process. Despite these constraints, there have been some attempts to prepare CrO2 thin films by chemical vapor deposition 共CVD兲. Examples include chemical vapor transport 共CVT兲 of CrO2Cl2 in a sealed tube at 3 atmospheres pressure,7 photodecomposition of CrO2Cl2, 8 and of Cr共CO兲6. 9 It was not clear whether the films grown by CVT were single phase CrO2. In the photodecomposition experiments, the CrO2 films either contained Cr2O3 or were amorphous. More recently, thin film growth efforts have involved a兲

Present address: Department of Chemical Engineering, University of Maine, 5737 Jenness Hall, Orono, Maine 04469; electronic mail: [email protected]

chemical vapor deposition using CrO3 as a precursor based on the Ishibashi process10 or a high-pressure bomb.5 Because of the properties of the CrO3 precursor, a solid that sublimes at ⬃260 °C and also partially decomposes, conventional CVD precursor handling equipment was not used. State-ofthe-art precursor handling equipment includes a precursor bubbler, automated valves, a pressure controller, mass flow controllers, and precursor flow sensors to deliver precise quantities of precursor into the reactor. Instead, a two-zone furnace was used: one zone to sublime the precursor and a second zone to completely decompose the precursor on the substrate. Several groups have prepared CrO2 thin films on a variety of substrates using this process.11–13 In this letter, an alternative, efficient, and controlled CVD process for preparing high quality epitaxial CrO2 共100兲 films on TiO2 共100兲 substrates is described. This process uses chromyl chloride (CrO2Cl2) as a precursor. This precursor is a liquid with a room temperature vapor pressure of 24 Torr, and is therefore, compatible with conventional CVD precursor handling equipment. The CrO2 films were prepared in either a pure argon or oxygen atmosphere. To compare these films with previously published work, the results of structural, magnetic, transport, and spin polarization measurements on the CrO2 films are presented. CrO2Cl2 共Aldrich, 99%兲 was transferred to a bubbler and held at 0 °C for all experiments. The carrier gases, either high-purity oxygen or argon 共Air Products, 99.99%兲 were used without further purification. The gas flow into the reactor, a tube furnace heated to 400 °C, was between 20 and 40 sccm. Substrates of 共100兲 TiO2 were cleaned in trichloroethane, acetone, methanol, isopropanol, and a 20% HF solution. The film thickness was measured with x-ray fluorescence and the film microstructure was examined in a Digital Instruments atomic force microscope 共AFM兲. The CrO2 films were 2000 Å thick, having a shiny and black appearance. The growth rate was 33 Å min⫺1, independent of the carrier gas (O2 or Ar兲, and reproducible. AFM measurements indicated the films had a root-mean-square 共rms兲 roughness be-

0003-6951/2000/76(25)/3789/3/$17.00 3789 © 2000 American Institute of Physics Downloaded 22 Jan 2006 to 141.217.4.72. Redistribution subject to AIP license or copyright, see http://apl.aip.org/apl/copyright.jsp

3790

Appl. Phys. Lett., Vol. 76, No. 25, 19 June 2000

DeSisto et al.

FIG. 2. Resistivity vs temperature of a 2000-Å-thick CrO2 film on (100)TiO2.

FIG. 1. 共a兲 X-ray diffraction of a 2000-Å-thick CrO2 film indicating 共100兲 film orientation (CrO2 peaks labeled, 䊏 indicates substrate peaks兲, and 共b兲 phi scan on the CrO2共110兲 line showing expected twofold symmetry around the 共200兲 direction.

tween 35 and 60 Å over a 5 ␮m2 scan area. The films grew with a granular microstructure with individual grain sizes from 0.5 to 2 ␮m. The crystal structure of the films was studied using a Philips MRD x-ray diffraction 共XRD兲 system, with a fourcrystal Ge 220 monochromator on the incident beam, and a Scintag x-ray diffraction system, both with Cu K ␣ 1 radiation (␭⫽1.5405 Å). ␪/2␪ scans were taken both along and at various angles to the growth direction, as well as rocking curves and ␾ scans 共where ␪/2␪ is set for a particular reflection at an angle ␺ to the film normal, and the film is rotated about the film normal兲 to look for misoriented grains. XRD ␪/2␪ scans 关Fig. 1共a兲兴 indicated the films grew completely 共100兲 oriented, in registry with the 共100兲 oriented TiO2 substrate. In addition, there was no evidence of impurities in the film, including Cr2O3. The rocking curves for the CrO2 共200兲 reflection in the films is of order 0.1°. The ␾ scan on the CrO2 共110兲 reflection shows the expected twofold symmetry, with no evidence of misaligned material 关Fig. 1共b兲兴. The measured lattice constants for the material are a⫽4.395 Å, b⫽4.443 Å, and c⫽2.916 Å, with an uncertainty of 0.001 Å. This shows that although the films are under compression along the growth direction 共⬃0.5%兲, the in-plane lattice constant b is expanded by ⬃0.5%, while c is the same as the bulk value. Figure 2 shows the electrical resistivity, ␳, for a 2000 Å CrO2 film. The resistivity at room temperature is 240 ␮⍀ cm and decreases to 10 ␮⍀ cm at 5 K, consistent with metallic behavior. This data compares favorably to Li et al.13 for measurements taken along the c axis, and is consistent with the epitaxial quality of the CrO2 films. A high resistivity ratio ( ␳ 298 K / ␳ 5 K ⫽24) further exemplifies the high quality of these epitaxial CrO2 thin films. The magnetic properties of the CrO2 films, as measured in a superconducting quantum interference device 共SQUID兲

magnetometer, are shown in Fig. 3. The hysteresis measurements were made with the magnetic field oriented in the plane of the film along a substrate edge. No distinction was made between particular substrate edges. A large magnetic field (H⬎4 kOe) was needed to saturate the magnetization while only a relatively small coercivity (HC⬍100 Oe) was observed 关Fig. 3共a兲兴. The saturation magnetization was determined to be 670 and 365 emu/cm3 at 5 and 298 K, respectively. These magnetization values are slightly lower than previously published values.14 However, this discrepancy is probably due to the error in the determination of the total volume of CrO2 material, since a nonuniform amount of CrO2 grew on the back 共unpolished side兲 of the substrate. Figure 3共b兲 shows the classic temperature dependence of the magnetization using an applied field of 500 Oe. The Curie temperature of the film 共395 K兲 agrees well with the bulk value. The spin polarization of CrO2 was measured by the point contact Andreev reflection 共PCAR兲 technique, which is described in detail elsewhere.3,15 At least 10 different junctions, established by gently pressing a superconducting 共Pb兲 tip into the CrO2 film, with contact resistance between 1 and 100 ⍀ were measured for each sample. Several different samples

FIG. 3. 共a兲 Hysteresis curve of a CrO2 film measured at 298 K, and 共b兲 temperature dependent magnetization obtained in a 500 Oe field. Downloaded 22 Jan 2006 to 141.217.4.72. Redistribution subject to AIP license or copyright, see http://apl.aip.org/apl/copyright.jsp

DeSisto et al.

Appl. Phys. Lett., Vol. 76, No. 25, 19 June 2000

3791

ties of fabricating GMR and/or tunnel junction devices based on CrO2, and thus opens up new opportunities in magnetoelectronics. The authors acknowledge and thank Dr. A. Berry for many helpful discussions and J. Trotter for assistance with some of the measurements. T.A. and B.N. were supported by NRC and ASEE fellowships, respectively. This work was supported by the Office of Naval Research. G. A. Prinz, Science 282, 1660 共1998兲. K. J. Schwarz, J. Phys. F: Met. Phys. 16, L211 共1986兲. 3 R. J. Soulen, Jr., J. M. Byers, M. S. Osofsky, B. Nadgorny, T. Ambrose, S. F. Cheng, P. R. Broussard, C. T. Tanaka, J. Nowak, J. S. Moodera, A. Barry, and J. M. D. Coey, Science 282, 85 共1998兲. 4 K. P. Ka¨mper, W. Schmidt, G. Gu¨ntherodt, R. J. Gambino, and R. Ruf, Phys. Rev. Lett. 59, 2788 共1987兲. 5 L. Ranno, A. Barry, and J. M. D. Coey, J. Appl. Phys. 81, 5774 共1997兲. 6 K. Ko¨hler, M. Maciejewski, H. Schneider, and A. Baiker, J. Catal. 157, 301 共1995兲; M. Maciejewski, K. Ko¨hler, H. Schneider, and A. Baiker, J. Solid State Chem. 119, 13 共1995兲. 7 L. Ben-Dor and Y. Shimony, J. Cryst. Growth 24Õ25, 175 共1974兲. 8 C. Arnone, M. Rothschild, J. G. Black, and D. J. Erlich, Appl. Phys. Lett. 48, 1018 共1986兲. 9 F. K. Perkins, C. Hwang, M. Onellion, Y-G. Kim, and P. Dowben, Thin Solid Films 198, 317 共1991兲. 10 S. Ishibashi, T. Namikawa, and M. Satou, Mater. Res. Bull. 14, 51 共1979兲. 11 K. Suzuki and P. M. Tedrow, Solid State Commun. 107, 583 共1998兲. 12 K. Suzuki and P. M. Tedrow, Phys. Rev. B 58, 11597 共1998兲. 13 X. W. Li, A. Gupta, T. R. McGuire, P. R. Duncombe, and G. Xiao, J. Appl. Phys. 85, 5585 共1999兲. 14 B. L. Chamberland, Crit. Rev. Solid State Mater. Sci. 7, 1 共1977兲. 15 S. K. Upadhyay, A. Palanisami, R. N. Louie, and R. A. Buhrman, Phys. Rev. Lett. 81, 3247 共1998兲. 16 G. E. Blonder, M. Tinkham, and T. M. Klapwijk, Phys. Rev. B 25, 4515 共1982兲. 17 B. Nadgorny, R. J. Soulen Jr., M. Osofsky, K. Swider, P. Lubitz, A. Gupta, X. W. Li, and G. Xiao 共unpublished兲. 1 2

FIG. 4. Normalized conductance as a function of bias voltage for a CrO2 thin film: solid circles–experimental data; solid line–the modified BTK fit at T⫽1.7 K, ⌬(T)⫽1.2 meV; fitting parameters: P⫽80%, Z⫽1.3.

were studied. All samples were highly spin polarized. Figure 4 shows typical conductance data for a CrO2 film as a function of bias voltage, V, normalized at VⰇ⌬(T)/e 关 ⌬(T) –superconducting gap at temperature T, e–electron charge兴. Experimental curves for each junction were fitted separately by the modified3 BTK16 theory with only two fitting parameters, spin polarization, P and barrier strength Z.16 The spin polarization for each sample was obtained by averaging over the resulting values of P. The spin polarization for a CrO2 thin film was found to be 81⫾3%, which is comparable to the values obtained for the CrO2 films fabricated by other techniques.3,17 The Z range was 0.6–1.3. In summary, an efficient and controllable method for depositing high quality epitaxial CrO2 thin films was developed using a standard CVD technique with CrO2Cl2 as a precursor. The films were metallic, smooth, epitaxial, and highly spin polarized. This technique enhances the possibili-

Downloaded 22 Jan 2006 to 141.217.4.72. Redistribution subject to AIP license or copyright, see http://apl.aip.org/apl/copyright.jsp