Download Control of crystalline texture in polycrystalline alumina

Control of crystalline texture in polycrystalline alumina
ceramics by electrophoretic deposition in a strong
magnetic field
T. Uchikoshi,a) T.S. Suzuki, H. Okuyama, and Y. Sakka
National Institute for Materials Science (NIMS), Tsukuba, Ibaraki 305-0047, Japan
(Received 2 September 2003; accepted 9 February 2004)
Highly crystalline-textured pure dense alumina ceramics were fabricated from spherical
alumina powder without any seed particles and sintering additives by electrophoretic
deposition (EPD) in a strong magnetic field of 10 T. The crystalline texture was
confirmed by x-ray diffraction (XRD) for alumina ceramics deposited at 10 T followed
by sintering at 1873 K. The angle between the directions of the magnetic and electric
fields (␸B-E) was altered to control the dominant crystal faces of the ␣-alumina
monoliths. The average orientation angles estimated from the XRD diagram of the
samples prepared at ␸B-E ⳱ 0°, 45°, and 90° were 16.52°, 45.15°, and 84.90°,
respectively. Alumina/alumina laminar composites with different crystalline-oriented
layers were also fabricated by alternately changing the ␸B-E layer by layer during EPD
in a 10 T magnetic field. It was demonstrated that by using this technique, it is
possible to control the crystalline orientation by changing the angle of E versus B
during the EPD.
I. INTRODUCTION
Recently, there have been a number of studies concerning the fabrication of textured ceramics, as they have
anisotropic mechanical, thermal, and electrical properties, which are similar to single crystals.1 Textured ceramics have been produced by a variety of techniques;
such as tape casting,2 hot forging or deformation,3–5 eutectic solidification,6 and templated or seeded grain
growth.7–9 Ceramic platelets, fibers, or whiskers are used
as seed particles along with some additives to promote
the anisotropic grain growth during sintering.9 Although
it is important to make a distinction between the microstructural texture and crystalline texture,10 they have often been confused.
There has been increased interest in fabricating crystalline-textured materials using the influence of an external magnetic field against the magnetic anisotropy of the
materials. Many materials in asymmetric (noncubic)
crystalline structures have anisotropic magnetic susceptibilities, ⌬␹ ⳱ ␹|| −␹⊥, associated with their crystal structures, where ␹|| and ␹⊥ are the susceptibilities parallel
and perpendicular to the magnetic principal axis, respectively. When a single crystal of these materials is placed
in a magnetic field, the crystal is rotated and the
a)
Address all correspondence to this author.
e-mail: [email protected]
DOI: 10.1557/JMR.2004.0198
J. Mater. Res., Vol. 19, No. 5, May 2004
crystallographic axis of high ␹ is aligned in the direction
of the magnetic field. The driving force of the magnetic
alignment is the energy of the crystal anisotropy and is
given as11
⌬E = ⌬␹VB2 Ⲑ 2␮0
,
(1)
where V is the volume of the material, B is the applied
magnetic field, and ␮0 is the permeability in a vacuum.
This alignment occurs when the energy of anisotropy is
higher than the energy of thermal motion, that is
⌬E ⬎ kT
,
(2)
where k is Boltzmann’s constant.
Generally, the magnetic susceptibilities of feeble magnetic materials (|␹| ⳱ 10−3∼10−6) are quite low in comparison with those of ferromagnetic materials (|␹| ⳱
102∼104), and the ⌬E of feeble magnetic materials is
much lower than kT in a conventional magnetic field
generated by a permanent magnet (B ⳱ ∼0.1 T). Therefore, the influence of a magnetic field on feeble magnetic
materials has not yet been taken into consideration. The
recent development of superconducting magnet technologies has provided academic laboratories with magnetic fields as high as approximately 10 T. Under such
strong magnetic fields, the magnetization force acting on
feeble magnetic materials is not negligible.
Recently, it has been reported that crystalline-textured
ceramics such as alumina,12 titania,13-15 zinc oxide,15,16
© 2004 Materials Research Society
1487
T. Uchikoshi et al.: Control of crystalline texture in polycrystalline alumina ceramics by electrophoretic deposition in a strong magnetic field
and SiC whisker-dispersed alumina17 have been fabricated by slip casting under 10 T. To extend the use of a
strong magnetic field for the colloidal processing of ceramics, we have proposed the use of both magnetic and
electric fields, that is, electrophoretic deposition (EPD)
in a strong magnetic field.18–21 EPD is a colloidal process
wherein ceramic bodies are directly shaped from a stable
colloid suspension by a dc electric field.22–24 A schematic illustration of the concept is shown in Fig. 1. Ceramic particles dispersed in a solvent are rotated due to
their magnetic anisotropy and then deposited on a substrate. It is essential for each particle to be single crystalline and be deflocculated for their rotation in a solvent.18 We have reported that crystalline-textured alumina18–21 and titania25,26 are fabricated by EPD in a
strong magnetic field.
This paper reports the significant advantages of this
facile and precise technique to synthesize crystallinetextured alumina monoliths and alumina/alumina laminate composites by controlling the directions between the
magnetic and electric fields during EPD in a superconducting magnetic environment.
II. EXPERIMENTAL
Spherical, single crystalline ␣-alumina particles
(Sumitomo Chem. Co., Ltd., Tokyo, Japan, AKP-50, average particle size of 0.2 ␮m, high purity of >99.99%)
were dispersed at pH 4 in distilled water by ultrasound,
and then a deflocculated aqueous suspension with a
10 vol% solid content was prepared. The zeta-potential
of alumina measured by the laser Doppler velocimetry
method was +40 mV at pH 4. The suspension was placed
in a superconducting magnet (Japan Magnet Technology,
Kobe, Japan, JMTD-10T100 with a room temperature
bore of 100 mm), and then a strong 10 T magnetic field
was applied to the suspension to rotate each particle. A
pair of electrodes, with an area of 25 × 35 mm2 and
20 mm spacing and held on a phenol resin support, was
put in the suspension, and then an electric current was
applied. The center of the magnetic field was at the center of the two electrodes. The magnetic field was maintained in the suspension during the EPD at a constant
voltage of 30 V at room temperature. A palladium sheet
was used as the cathodic substrate to absorb hydrogen
produced by electrolysis of the solvent.27 A schematic
illustration of the apparatus is shown in Fig. 2. The direction of the electric field relative to the magnetic field
(the angle between the vectors E and B) was altered
(0°, 45°, 90°) to control the dominant crystal faces. The
sintering was conducted at a fixed temperature of 1873 K
for 2 h in air out of the magnetic field. The density of the
sintered compacts was measured by Archimedes’ method
using kerosene. The degree of crystalline orientation of
the specimen was characterized by x-ray diffraction
(XRD) analysis. The microstructure was observed using
a violet laser color 3D probe microscope (Keyence,
Osaka, Japan, VK-9500).
III. RESULTS AND DISCUSSION
The use of a palladium substrate effectively suppressed the bubble formation at the cathode, and dense,
bubble-free deposits were obtained. The green density of
the deposits was approximately 60% of the theoretical
density (TD) regardless of the deposition in and out of
the magnetic field. The sintered density of the deposits at
1873 K for 2 h was approximately 97% of TD.
The degree of crystalline orientation of the asdeposited specimen characterized by XRD analysis was
highly improved with the grain growth during sintering.20 Figure 3 shows the XRD patterns of the ␣-alumina
deposited at 10 T, followed by sintering at 1873 K for
2 h. The angle of E versus B was fixed at 0° (␸B-E ⳱ 0°)
during the deposition. The XRD analysis was carried out
for the cross-sectional planes that were parallel and perpendicular to the substrate. Hereafter, these planes were
designated as the TOP and SIDE planes, respectively. To
characterize the XRD peaks, the interplanar angles ␾hkl
between the planes (hkl) and the basal plane (00l) were
calculated for a hexagonal unit cell of ␣-alumina (a ⳱
FIG. 1. Schematic diagram of the concept of the electrophoretic deposition in a strong magnetic field: (a) deflocculated suspension, (b) alignment
of the particles in a strong magnetic field, and (c) electrophoretic deposition in the magnetic field.
1488
J. Mater. Res., Vol. 19, No. 5, May 2004
T. Uchikoshi et al.: Control of crystalline texture in polycrystalline alumina ceramics by electrophoretic deposition in a strong magnetic field
FIG. 2. Schematic illustration of the EPD apparatus in a superconducting magnet.
FIG. 3. XRD patterns of the ␣-alumina deposited at 10 T followed by
sintering at 1873 K. The angle of E versus B was fixed at 0° (␸B-E ⳱
0°) during the deposition.
FIG. 4. XRD patterns of the ␣-alumina deposited at 0 T
(external to the magnetic field) followed by sintering at 1873 K.
0.4758 nm, c ⳱ 1.2991 nm).28 The standard XRD data of
␣-alumina from ICDD29 with the ␾hkl have been described in a previous paper.18 The diffraction peaks of
the planes at low interplanar angles such as (006)(␾006 ⳱
0°), (0012)(␾0012 ⳱ 0°), and (1010)(␾1010 ⳱ 17.5°) are
characteristic of the TOP. In contrast, the diffraction
peaks of the planes at high interplanar angles such as
(110)(␾110 ⳱ 90°), (030)(␾030 ⳱ 90°), (220)(␾220 ⳱
90°), and (211)(␾211 ⳱ 83.16°) are characteristic of the
SIDE. The XRD data clearly show the crystallite orientation of the ␣-alumina prepared in the strong magnetic
field of 10 T. It is also shown that the c axis is easily
aligned along the magnetic field. Similar XRD results
were observed when the angle of E versus B was fixed at
180° (␸B-E ⳱ 180°).
Figure 4 shows the XRD patterns of the ␣-alumina
deposited at 0 T (external to the magnetic field) followed
by sintering at 1873 K for 2 h. No difference is observed
between the XRD patterns of the TOP and the SIDE. It is
obvious that the specimen prepared with no magnetic
field has a randomly oriented polycrystalline structure.
Figure 5 shows the variation in the XRD patterns of
J. Mater. Res., Vol. 19, No. 5, May 2004
1489
T. Uchikoshi et al.: Control of crystalline texture in polycrystalline alumina ceramics by electrophoretic deposition in a strong magnetic field
fixes the orientation of each particle; the c axis of ␣-alumina is aligned parallel to B in the suspension. When an
electrical field is then applied to the oriented particles,
they move along with the electric field lines retaining
their orientation relative to the magnetic field lines and
then deposit on the substrate. The degree of crystalline
orientation was estimated from the intensity of the XRD
peaks using the following equation30:
FIG. 5. Changes in the XRD patterns of the TOP planes with the angle
between the directions of B and E. The interplanar angles ␾hkl between
the planes (hkl) and the basal plane (00l) are also noted in the figure.
the TOP planes with the angle between the directions of
B and E (␸B-E). The ␾hkl of the appeared peaks are also
shown in the figure. When E is parallel to B (␸B-E ⳱ 0°),
the diffraction peaks of the planes at low interplanar
angles (␾hkl is close to 0°) are dominant. When ␸B-E is
changed to 45°, the dominant diffraction peaks change to
the planes of the intermediate interplanar angles (␾hkl is
close to 45°). Finally, when E is perpendicular to B (␸B-E
⳱ 90°), the dominant diffraction peaks change to the
planes of high interplanar angles (␾hkl is close to 90°).
This result is explained as follows. The magnetic field
␾ = 兺共Ihkl ⳯ ␾hkl兲 Ⲑ 兺Ihkl ,
(3)
where ␾ is the average orientation angle of the crystals of
the TOP, and Ihkl is the intensity of the (hkl) reflection.
The average orientation angles of the samples prepared at
␸B-E ⳱ 0°, 45°, and 90° were 16.52°, 45.15°, and 84.90°,
respectively. This result clearly shows that the dominant
crystal faces on the TOP surface can be controlled by
varying the angle of E versus B. When a charged particle
moves across a magnetic field line (E × B ⫽ 0), the
particle should be affected by the Lorentz force. However, its effect on the alignment of the particles seems to
be not that strong.
Figure 6 shows a schematic illustration and the crosssectional microstructure of a crystalline-textured alumina/alumina laminate composite prepared by alternately changing ␸B-E ⳱ ±45° layer by layer during EPD
in a magnetic field of 10 T. The crystalline orientation is
changed at the interface of the two layers. Figure 7 shows
the cross-sectional microstructure of a laminar composite
prepared by alternately changing ␸B-E ⳱ 0° and 45°
layer by layer. Laminar composites with oriented and
randomly oriented layers have been fabricated by the
alternate EPD of the suspensions placed in and out of a
superconducting magnet.21
FIG. 6. Microstructures of the cross-sectional plane of a crystalline-textured alumina/alumina laminate composite prepared by alternately
changing ␸B-E ⳱ ±45° during EPD in a magnetic field of 10 T.
1490
J. Mater. Res., Vol. 19, No. 5, May 2004
T. Uchikoshi et al.: Control of crystalline texture in polycrystalline alumina ceramics by electrophoretic deposition in a strong magnetic field
the Ministry of Education, Culture, Sports, Science and
Technology, Japan.
REFERENCES
FIG. 7. Microstructures of the cross-sectional plane of a crystallinetextured alumina/alumina laminate composite prepared by alternately
changing ␸B-E ⳱ 0° and 45° during EPD in a magnetic field of 10 T.
Electrophoretic deposition in a strong magnetic field is
an excellent method to fabricate crystalline textured ceramic thick bodies. This method can also be applied to
prepare crystalline-oriented, or specified crystal face,
thin films for functional applications.
IV. CONCLUSIONS
The magnetic field fixes the orientation of each particle; the c axis of ␣-alumina is aligned parallel to B in
the suspension. When an electrical field is then applied to
the oriented particles, they move along with the electric
field lines while retaining their orientation against the
magnetic field lines, and then deposit on the substrate.
By varying the angle between the vectors E and B, the
crystalline orientation in the bulk and dominant crystal
faces at the surface can be controlled.
ACKNOWLEDGMENTS
The authors wish to thank Mr. Koji Kuramoto of the
Keyence Co. for his help with the microscopic observations and Prof. Patrick S. Nicholson of McMaster University and Dr. Partho Sarkar of the Alberta Research
Council for their valuable comments. This research was
financially supported by the Hosokawa Powder Technology Foundation and the Budget for Nuclear Research of
1. E. Suvaci and G.L. Messing, J. Am. Ceram. Soc. 83, 2041 (2000).
2. K. Hirao, M. Ohashi, M.E. Brito, and S. Kanzaki, J. Am. Ceram.
Soc. 78, 1687 (1995).
3. T. Takenaka and K. Sakata, Jpn. J. Appl. Phys. 19, 31 (1980).
4. Y. Ma and K.J. Bowman, J. Am. Ceram. Soc. 74, 2941 (1991).
5. Y. Yoshizawa, M. Toriyama, and S. Kanzaki, J. Am. Ceram. Soc.
84, 1392 (2001).
6. V.S. Stubican and R.C. Bradt, Annu. Rev. Mater. Sci. 11, 267
(1981).
7. S.H. Hong and G.L. Messing, J. Am. Ceram. Soc. 82, 867 (1999).
8. E. Suvaci, K-S. Oh, and G.L. Messing, Acta Mater. 49, 2075
(2001).
9. M.M. Seabaugh, I.H. Kerscht, and G.L. Messing, J. Am. Ceram.
Soc. 80, 1181 (1997).
10. D. Brandon, D. Chen, and H. Chan, Mater. Sci. Eng. A 195, 189
(1995).
11. P. De Rango, M. Less, P. Lejay, A. Sulpice, R. Tournier, M. Ingold,
P. Germi, and M. Pernet, Nature 349, 770 (1991).
12. T.S. Suzuki, Y. Sakka, and K. Kitazawa, Adv. Eng. Mater. 3, 490
(2001).
13. T.S. Suzuki and Y. Sakka, Jpn. J. Appl. Phys. 41, L1272 (2002).
14. Y. Sakka, T.S. Suzuki, N. Tanabe, S. Asai, and K. Kitazawa, Jpn.
J. Appl. Phys. 41, L1416 (2002).
15. T.S. Suzuki and Y. Sakka, Key Eng. Mater. 248, 191 (2003).
16. T.S. Suzuki and Y. Sakka, Chem. Lett. 12, 1204 (2002).
17. T.S. Suzuki, Y. Sakka, and K. Kitazawa, J. Ceram. Soc. Jpn. 109,
886 (2001).
18. T. Uchikoshi, T.S. Suzuki, H. Okuyama, and Y. Sakka, J. Mater.
Res. 18, 254 (2003).
19. T. Uchikoshi, T.S. Suzuki, H. Okuyama, and Y. Sakka, Electrophoretic Deposition: Fundamentals and Applications, ECS Procs.
2002-21, 9 (2002).
20. T. Uchikoshi, T.S. Suzuki, H. Okuyama, and Y. Sakka, J. Mater.
Sci. 39, 861 (2004).
21. T. Uchikoshi, T.S. Suzuki, H. Okuyama, and Y. Sakka, J. Eur.
Ceram. Soc. 24, 225 (2004).
22. P. Sarkar and P.S. Nicholson, J. Am. Ceram. Soc. 79, 1987 (1996).
23. I. Zhitomirsky, Adv. Colloid & Interface Sci. 97, 279 (2002).
24. A.R. Boccaccini and I. Zhitomirsky, Curr. Opin. Solid State
Mater. Sci. 6, 251 (2002).
25. T. Uchikoshi, T.S. Suzuki, H. Okuyama, and Y. Sakka, Trans.
Mater. Res. Soc. Jpn. (in press).
26. T. Uchikoshi, T.S. Suzuki, F. Tang, H. Okuyama, and Y. Sakka,
Ceram. Int. (in press)
27. T. Uchikoshi, K. Ozawa, B.D. Hatton, and Y. Sakka, J. Mater.
Res. 16, 321 (2001).
28. B.D. Cullity, Elements of X-ray Diffraction (Addison Wesley,
London, 1959), p. 460.
29. JCPDS No. 10-173. International Center for Diffraction Data:
Newton Square, PA.
30. M. Tahashi, M. Ishihara, K. Sassa, and S. Asai, Mater. Trans. 44,
285 (2003).
J. Mater. Res., Vol. 19, No. 5, May 2004
1491