Langmuir 2008, 24, 5359-5365
5359
Chemisorption of Phosphoric Acid and Surface Characterization of As Passivated AlN Powder Against Hydrolysis Ibram Ganesh, Susana M. Olhero, Aurora B. Araújo, Maria R. Correia, Govindan Sundararajan, and José M. F. Ferreira* Centre for AdVanced Ceramics, International AdVanced Research Centre for Powder Metallurgy and New Materials (ARCI), Hyderabad - 500 005, A.P., India, Department of Ceramics and Glass Engineering, CICECO, UniVersity of AVeiro, -3810193 AVeiro, Portugal, Department of Physics, UniVersity of AVeiro, AVeiro, P-3810193, Portugal ReceiVed January 9, 2008. ReVised Manuscript ReceiVed February 16, 2008 By simply refluxing a commercial AlN powder in a mixture solution of ethanol, H3PO4, and Al(H2PO4)3 for 24 h at 80 °C, the powder was successfully passivated against hydrolysis. The phosphate layer formed on the surface of AlN powder was found to be quite stable toward protecting the powder from hydrolysis. The efficacy of the coating was established by suspending the treated and the untreated powders in water for 72 h and subsequently characterizing them by X-ray diffraction (XRD), Fourier transform infrared (FT-IR), X-ray photoelectron spectroscopy (XPS), transmission electron microscopy (TEM), and Raman analysis. The good dispersing behavior of the treated AlN powder in water was confirmed by the low viscosity of an AlN suspension containing 50 vol % solids demonstrating the viability of replacing organic solvents by water in colloidal processing of AlN-based ceramics.
Introduction Polycrystalline AlN ceramics have received considerable attention on account of their attractive properties such as high thermal conductivity, low thermal expansion, high electrical resistivity, and excellent resistant to attack by molten metals.1–5 At moderate temperatures (∼200 °C), AlN thermal conductivity exceeds that of copper. Owing to these important properties, AlN-based materials are employed in various applications including heat engines, crucibles for molten aluminum and gallium arsenide, airborne ballistic armor materials, and substrates and packaging for high power or high-density assemblies of microelectronic components, chip carriers and heat sinks.1–5 Although AlN can be consolidated by hot isostatic pressing, pressureless sintering at temperatures of 1600–2000 °C is increasingly used.3–5 Conventional dry-pressing of powder mixtures of AlN and small quantities of sintering aids, such as SiO2, Y2O3, Al2O3, CaO, and so forth, is the most currently used consolidation method for AlN-based products, followed by sintering at >1750°C to achieve full density.6–9 Expensive and extensive machining is often required when the final shape of the products consolidated by dry-pressing is not simple. Complexshaped products have been most suitably consolidated by nonaqueous colloidal processing routes such as injection molding, slip casting, tape casting, and so forth.1–9 However, recent economic and environmental concerns related to the use of organic-based media have stimulated an increasing interest toward * To whom correspondence should be addressed. Tel.: 351-234-370242. Fax: 351-234-370204. E-mail:
[email protected]. (1) Sheppard, L. M. Am. Ceram. Soc. Bull. 1990, 69, 1801. (2) Knudsen, A. K. Am. Ceram. Soc. Bull. 1995, 74, 97. (3) Virkar, A. V.; Jackson, T. B.; Cutler, R. A. J. Am. Ceram. Soc. 1989, 72, 2031. (4) Collange, A.; Grosseau, P.; Guilhot, B.; Disson, J. P.; Joubert, P J. Eur. Ceram. Soc. 1997, 17, 1897. (5) Lindqvist, K.; Carlstro¨m, E.; Persson, M.; Carlsson, R. J. Am. Ceram. Soc. 1989, 72(1), 99. (6) Jarrige, J.; Bouzouita, K.; Doradoux, C.; Billy, M. J. Eur. Ceram. Soc. 1993, 12, 279. (7) Lee, R. J. Am. Ceram. Soc. 1991, 74(9), 2242. (8) Liao, H.; Coyle, T. W. J. Am. Ceram. Soc. 1995, 78(5), 1291. (9) Du, S.; Gao, L.; Li, F.; Liu, Z.; Jei-Oh, Y. J. Mater. Sci. 1996, 31, 3679.
processing AlN powders from aqueous suspensions. Unfortunately, unprotected AlN powders undergo an easy hydrolysis process when in contact with water molecules (liquid or vapor) and gradually transform into γ-AlOOH, Al(OH)3, or γ-Al2O3 as revealed by thermal analysis,10,11 X-ray photoelectron spectroscopy (XPS),12,13 diffuse-reflectance Fourier transform infrared (FT-IR) spectroscopy14 and auger electron spectroscopy.15 So far, a great variety of methods have been devised to passivate AlN powder against hydrolysis.16–22 Some successful methods involved the use of hydrophobic substances as coating agents, which hinder the preparing of high solids loading (>50 vol.%) slurries usually required for aqueous processing of complexshaped parts.16 In a recent study,17 sebacic acid (a dicarboxylic acid) was employed to protect AlN powder from hydrolysis and also to provide hydrophilic -COOH groups on the powder surface to facilitate its dispersion in water. In another study, Fukumoto et al.,18 investigated the effect of temperature and surface area on the hydrolysis behavior of AlN powder and observed faster hydrolysis with increasing specific surface area and temperature. In another study, Krnel and Kosmac19,20 studied the hydrolysis behavior of AlN at pH 1 and pH 3 by using several inorganic acids including HF, HCl, HNO3, H2SO4, H3PO4, and Al(H2PO4)3. (10) Saito, N.; Ishizaki, C.; Ishizaki, K. J. Ceram. Soc. Jpn., Int. E 1994, 102, 299. (11) Nicolaescu, I. V.; Tardos, G.; Riman, R. E. J. Am. Ceram. Soc. 1994, 77(9), 2265. (12) Metselaar, R.; Reenis, R.; Chen, M.; Gorter, H.; Hintzen, H. T. J. Eur. Ceram. Soc. 1995, 15, 1079. (13) Liao, H. M.; Sodhi, R. N. S.; Coyle, T. W. J. Vac. Sci. Technol., A 1993, 11(5), 2681. (14) Highfield, J. G.; Bowen, P. Anal. Chem. 1989, 61, 2399. (15) Saito, N.; Ishizaki, K. J. Am. Ceram. Soc. 1996, 79(5), 1213. (16) Groat, E. A.; Mroz, T. J. Am. Ceram. Soc. Bull. 1994, 73(11), 75–78. (17) Shimizu, Y.; Hatano, J.; Hyodo, T.; Egashira, M. J. Am. Ceram. Soc. 2000, 83, 2793. (18) Fukumoto, S.; Hookabe, T.; Tsubankino, H. J. Mater. Sci. 2000, 35(11), 2743. (19) Krnel, K.; Kosmac, T. J. Am. Ceram. Soc. 2000, 83(6), 1375. (20) Krnel, K.; Kosmac, T. J. Eur. Ceram. Soc. 2001, 21, 2075. (21) Uenishi, F. M.; Hashizume, K. N. Y.; Yokote, T. U.S. Patent 4,923,689, May 8, 1990. (22) Morisada, Y.; Sakurai, T.; Miyamoto, Y. Int. J. Appl. Ceram. Technol. 2004, 1(4), 374.
10.1021/la800075b CCC: $40.75 2008 American Chemical Society Published on Web 04/10/2008
5360 Langmuir, Vol. 24, No. 10, 2008
They noticed the stability of AlN against hydrolysis at pH 1 regardless of the acid used; however, at pH 3, the stability was found to depend on the nature of the acid involved. They also found that Al(H2PO4)3 provided a very effective protective coating against hydrolysis.19 However, the use of acid solutions at pH A-AlN > A-AlN-72 h. The amount of N detected in the A-AlN-72 h powder is negligible. This is due to the occurrence of extensive hydrolysis and to the fact that the soft X-rays (1–3 keV) used in the XPS analysis do not penetrate more than a 30 Å depth from the surface of the sample. Because of the high thickness of the aluminum hydroxide layer formed on the surface of AlN particles, the soft X-rays could not reach the core of AlN particles, whereas hard X-rays used in the XRD study (Figure 2) could detect some remaining AlN crystals. Figure 6 and Table 1 show XPS peaks and BE values of Al 2p core levels belonging to four AlN powders. No appreciable chemical shifts could be seen in the BE values of Al for all analyzed powders, and the values match very well with those reported in the literature.12,13,29 The absence of noticeable chemical shifts in the BE of Al atoms is not surprising since all of them possess a +3 oxidation state. The small differences in the BE values reported in Table 1 are within the allowed range and could be due to the minor changes in the experimental conditions. The concentration of Al detected in different powders is as follows: A-AlN-72 h > T-AlN-72 h > T-AlN > A-AlN. As a result of the formation of Al(OH)3 upon hydrolysis of AlN, the surface concentrations of Al and O increase at the expenses of nitrogen, which escapes as NH3 gas. Very interestingly, the A-AlN powder exhibited the lowest Al concentration and highest N concentration among the four powders. This indicates that the A-AlN powder has relatively low oxygen concentration, in good agreement with the technical data sheet from the supplier. Figure 7 shows the P 2p photoelectron peaks of four different AlN powders and the BE values recorded (Table 1) are according to the literature reports.12,13,27,29 In the case of P also, no chemical shift is seen in BE values because of the availability of only a +3 oxidation state for the P atom. As expected, T-AlN and T-AlN-72 h reveal higher P concentrations than the other two powders, confirming the adsorption of a phosphate layer onto the surface of treated AlN particles. Surprisingly, even the A-AlN powder exhibits a small amount of P that can be regarded as an impurity. The atomic percentage ratios of IN/(IO + IN + IAl + IP), IP/(IO + IN + IAl + IP), and IO/IN as determined by XPS reported in Table 1 show almost the same values for the A-AlN, T-AlN, and T-AlN-72 h powders, confirming the formation of a stable phosphate layer onto the surface of treated AlN powder. For the A-AlN-72 h powder, the oxygen concentration increased and nitrogen concentration decreased several times, confirming the
IN/(IO + IN + IAl + IP)
IO/(IO + IN + IAl + IP)
IP/(IO + IN + IAl + IP)
IO/IN ((3%)
7.69
60.43
16.24
7.85
7.59
60.28
15.46
7.94
5.41
59.80
15.83
11.05
1.143
69.90
2.29
61.15
considerable extent of hydrolysis after 72 h of contact with water, leading to the formation of Al(OH)3 and NH3. The effects of the phosphate coating treatment on the surface morphology were analyzed by TEM for the four powders studied. Figure 8a,b shows the TEM micrographs of the T-AlN-72 h and A-AlN-72 h powders, respectively. Different magnifications were used to account for the coating layer thickness and the extent of hydrolysis. Very thin phosphate layers of about 25 nm could be measured for the T-AlN and T-AlN-72 h powders. This last powder also showed very fine (5–10 nm) particles of aluminum hydroxide due to incipient hydrolysis (Figure 8a). Contrarily, the A-AlN-72 h powder (Figure 8b) revealed a thick (100–300 nm) transparent layer of aluminum hydroxide around each particle, and a dark AlN core due to incomplete hydrolysis, as detected by XRD (Figure 2).
Figure 8. TEM micrographs of the T-AlN-72 h (a) and A-AlN-72 h (b) powders. The insets correspond to DDPs obtained from SAED analysis.
5364 Langmuir, Vol. 24, No. 10, 2008
Figure 9. Raman spectra of the various AlN powders investigated.
Figure 10. Viscosity versus shear rate of an aqueous AlN slurry containing 50 vol % solids.
The rings and spots shown in the digital diffraction patterns (DDP) of the selected-area electron diffraction (SAED) analysis (Figure 8a,b) account for the existence of periodic contrasts, which correspond to different sets of atomic planes of the crystalline structure. The presence of continuous concentric rings in DDPs of powders indicates that both powders consist of polycrystallites, and the grains in the in-plane directions are in a high degree of random orientation.30 However, the distances between the concentric rings in DDPs confirmed that the crystallites in the T-AlN-72 h powder are solely of AlN, whereas in the case of A-AlN-72 h they consist exclusively of Al(OH)3.31 Thus these SAED observations are well inline with the XRD, XPS, and TEM results. The Raman spectra of the four AlN powders recorded at room temperature are shown in Figure 9. Some studies have indicated that it is possible to obtain the Raman spectrum of layers as thin as 35 Å.32–38 AlN, like other members of the III-nitride family, is a tetrahedrally coordinated binary compound.32 III-nitrides (30) Lopez-Cartes, C.; Perez-Omil, J. A.; Pintado, J. M.; Calvino, J. J.; Kang, Z. C.; Eyring, L. Ultramicroscopy 1999, 80, 19. (31) Di Lello, B. C.; Moura, F. J.; Solorzano, I. G. Mater. Sci. Eng. C 2001, 15, 67. (32) Strite, S.; Morkoc, H. J. Vac. Sci. Technol., B 1992, 10, 1237. (33) Tinkham, M. Group Theory and Quantum Mechanics; McGraw Hill: New York, 1964. (34) Guang-Yan, Z., Guo-Xiang, L., and Yu-Fang, W. Lattice Vibration Spectroscopy; Higher Education Press: Beijing, 1992; pp 236–242. (35) Carlone, C.; Lakin, K. M.; Shanks, H. R. J. Appl. Phys. 1984, 55, 4010.
Ganesh et al.
crystallize in the wurtzite and zincblende structures. For the wurtzite structure, the space group is C6ν4 (P63mc).32 According to factor group analysis at the Γ point, at the center of the Brillouin Zone, the optical phonons are reducible into the following representation:33–35 Γ opt ) 1A1 + 1E1 + 2E2 + 2B1, where the A1 (polarized in the z direction) and E1 (polarized in the (x, y) plane) modes are both Raman active and infrared-active; the two E2 (l, low; h, high) modes are Raman active only, and the two B1 modes are silent modes. Because of the long-range macroscopic electrostatic field, the A1 and E1 modes split into A1L and E1T (L, longitudinal; T, transverse) for parallel propagation to the c-axis and A1T and E1L for perpendicular propagation.36 In AlN, the E1(TO) and A1(TO) Raman frequencies are grouped together in the ∼60 cm-1 frequency range, the E1(LO) and A1(LO) frequencies are in the ∼20 cm-1 range, and the LO-TO group splitting is ∼220 cm-1. According to the different scattering geometry configurations (selection rules), the observable modes for the different backscattering configurations can be obtained. Liu et al.36 applied the selection rules and determined that the frequencies of the A1(TO), A1(LO), E1(TO), E1(LO), E2l, and E2h modes are, respectively, 612, 898, 665, 910, 243, and 654 cm-1 in Wurtzite AlN. These results are close to those measured by Bergman,37 but different from those reported by Carlone et al.35 and Bergman et al.37 Different results obtained by several researchers could be due to the mass difference of AlN samples. Also, the different substrates for the AlN films affect the Raman spectrum.35 According to Brafman et al.,38 in polycrystalline AlN without preferred orientation, all the optical phonons will appear during the Raman scattering. As shown in Figure 9, A-AlN and T-AlN powders reveal frequencies of E21, A1(TO), E2h, E1(TO), A1(LO), and E1(LO) optical phonons at 252, 614, 658, 672, 894, and 912 cm-1, respectively. These measured results are the same as those obtained by McNeil et al.,39 but slightly different from those reported in references 34-40.34–40 The different fabrication methods used for AlN and different sample conditions for these various tests cause the Raman spectra to be a little different.35 Very interestingly, T-AlN-72 h powder also exhibited frequency bands attributed only to AlN crystals. The A-AlN-72 h powder revealed additional bands that could be attributed to the formed aluminum hydroxide on the surface of AlN powder due to AlN hydrolysis along with bands due to AlN. The traditional Raman spectrum of the bayerite, Al(OH)3, with wave numbers at 3651, 3543, and 3427 cm-1 (with the shoulder at 3440 cm-1) is also seen. In addition, a very weak signal was detected at approximately 3470 cm-1.32–38 Since, in this study, the analysis was performed only between 200 and 900 cm-1, it is difficult to confirm the presence of bayerite on the basis of hydroxyl groups. Raman bands for diaspore, Al(OH)3, are reported at 497, 447, and 327 cm-1 and attributed to different types of stretching and bending modes of Al-O-Al bonds, and those for corundum are seen at 580, 563, 449, 430, 417, 379, and 252 cm-1.41 However, the frequencies observed for Al(OH)3 in the case of A-AlN-72 h does not match those of diaspore or corundum. Therefore, these frequency bands could be attributed to aluminum trihydroxide, Al(OH)3, as detected by XRD study. As far as the stability of treated AlN powder is concerned, the Raman results (36) Liu, M. S.; Nugent, K. W.; Prawer, S.; Bursill, L. A.; Peng, J. L.; Tong, Y. Z. Int. J. Mod. Phys. B 1998, 12, 1963. (37) Bergman, L.; Mitra, D.; Cengiz, D.; Robert, F.; Christman, J. A.; Alexson, D.; Nemanich, R. J. J. Appl.Phys. 1999, 85, 3535. (38) Brafman, O.; Lengyel, G.; Mitra, S. S.; Gielisse, P. J.; Plendl, J. N.; Mansur, L. C. Solid State Commun. 1968, 6, 523. (39) McNeil, L. E.; Grimsditch, F. R. H. J. Am. Ceram. Soc. 1993, 76, 1132. (40) Wang, H.; Jiecai, H.; Yongtin, Z.; Shanyi, D. J. Chin. Ceram. Soc. 2000, 28, 26.
AlN Powder PassiVated against Hydrolysis
also support very well the results obtained from XPS and TEM studies. The T-AlN powder exhibited a zeta potential (ζ) value of -58 mV in water under the influence of 1 wt % Duramax 3005 (at approximately pH 9). Normally, ζ values greater than (50 mV allow the preparation of stable and high solids loading slurries.1-5 In the present case, aqueous AlN suspensions containing 50 vol % solids could be successfully prepared. Figure 10 shows a typical viscosity versus shear rate curve, confirming the good flowing properties. A shear thinning behavior is observed in the low shear rate range, followed by a near Newtonian behavior for shear rates >50 s-1, with the viscosity reaching a plateau of about 0.1 Pa · s. These rheological properties reveal the good suitability of the highly concentrated AlN suspensions for colloidal processing.
Conclusions The following conclusions can be drawn from the present study: (41) Shoval, S.; Michaelian, K. H.; Boudeulle, M.; Panczer, G.; Lapides, I.; Yariv, S. J. Therm. Anal. Calorim. 2002, 60, 205.
Langmuir, Vol. 24, No. 10, 2008 5365
(1) Ortho-phosphoric acid and aluminum dihydrogen phosphate could be successfully used to protect AlN powder against hydrolysis using a simple experimental procedure; (2) The phosphoric-acid-treated AlN powder was found to be stable in water for periods as long as 72 h; (3) Low-viscosity aqueous slurries with solids loading as high as 50 vol % could be prepared using the phosphoric acid treated AlN powder, which exhibited suitable rheological properties for colloidal processing; (4) FT-IR, XPS, and TEM techniques confirmed the presence of the protecting phosphate layer on the surface of phosphoricacid-treated AlN powder and its stability against hydrolysis. Acknowledgment. I.G. thanks SERC-DST (Government of India) for the awarded BOYSCAST fellowship (SR/BY/E-04/ 06). S.M.O. wishes to thank the Foundation for Science and Technology (FCT) of Portugal for the financial support under Grant SFRH/BPD/27013/2006. A.B.A. thanks CICECO for the financial support. LA800075B
