Alumina atomic layer deposition nanocoatings on primary diamond ...

Report 2 Downloads 62 Views
Available online at www.sciencedirect.com

Diamond & Related Materials 17 (2008) 185 – 189 www.elsevier.com/locate/diamond

Alumina atomic layer deposition nanocoatings on primary diamond particles using a fluidized bed reactor Xinhua Liang a , Guo-Dong Zhan b , David M. King a , Jarod A. McCormick a , John Zhang b , Steven M. George a , Alan W. Weimer a,⁎ a

Department of Chemical and Biological Engineering, University of Colorado, Boulder, Colorado 80309, USA b Smith International, Inc., Houston, Texas 77205, USA Received 28 June 2007; received in revised form 11 November 2007; accepted 3 December 2007 Available online 8 December 2007

Abstract Ultra-thin alumina films are successfully deposited on primary micron-sized diamond particles in a scalable fluidized bed reactor. The studies of fluidization at reduced pressure show that micron-sized diamond particles can be fluidized with the assistance of vibration. Alumina films are grown at 177 °C by atomic layer deposition (ALD) using sequential exposures of Al(CH3)3 and H2O. The deposited alumina films are characterized by X-ray photoelectron spectroscopy, transmission and scanning electron microscopy, inductively coupled plasma-atomic emission spectroscopy, and surface area. The results indicate that the alumina films are conformally coated on the primary diamond particle surface, and the growth rate of alumina is 0.12 nm per coating cycle. © 2007 Elsevier B.V. All rights reserved. Keywords: Alumina film; Atomic layer deposition (ALD); Fluidized bed reactor

1. Introduction Due to its extreme hardness, diamond has long been used in cutting and abrasive applications [1–3]. The advent of highpressure/high-temperature (HP/HT) synthesis methods [4–7] led to the discovery of polycrystalline diamond grit and the manufacture of polycrystalline diamond compact (PDC) materials [8]. PDC cutters are well known and widely used as the cutting element in drill bits. Such cutters generally comprise a PDC table formed on a hard metal tungsten carbide (WC) substrate by a HP/HT sintering process. Cobalt is normally present in the diamond bonding in the PDC press [9–11]. At the elevated temperature and pressure where diamond-to-diamond bonding occurs, the cobalt migrates into the diamond grit and helps to catalyze the bonding process. However, chemical adsorbates such as oxygen- and nitrogen-containing functional groups on the surface of the diamond grit are found to be detrimental to the formation of diamond-to-diamond bonding [12,13]. The ⁎ Corresponding author. E-mail address: [email protected] (A.W. Weimer). 0925-9635/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.diamond.2007.12.003

occurrence of oxidation and graphitization of PDC cutters during the service also limits their applications. Therefore, there is a need to change the surface properties of diamond grit by particle coatings while maintaining the bulk properties. Coatings can also be used to functionalize, protect, strengthen or modify the properties of the particle. Despite the importance of coating particles, very few methods exist to deposit conformal ultra-thin films on particles. Traditional chemical vapor deposition (CVD) processes are incapable of coating such an ultra-thin conformal film on diamond. Their high operating temperature requirements and tendency to produce non-uniform and granular films via undesired gas-phase reactions also make them non-ideal [14]. Atomic layer deposition (ALD) is a nanocoating process that is an ideal method for such an application. The ALD processing method is similar to CVD, but allows conformal atomic level control over the deposition process [15– 23]. This control during ALD is achieved by introducing the reactants individually, separated by purge steps, in a sequential manner and carrying out self-limiting surface reactions that occur during each step on the substrate surface. In contrast, both reactants are present simultaneously during CVD. Consequently,

186

X.H. Liang et al. / Diamond & Related Materials 17 (2008) 185–189

the CVD reaction occurs continuously and there is no automatic control over the film thickness. Due to its high chemical and thermal stability, alumina (Al2O3) is selected as a coating material for the present study. Alumina ALD has been carried out by splitting the reaction of trimethylaluminum (TMA) and H2O into two self-limiting halfreactions [15–17]:   ðAÞ AlOH⁎ þ AlðCH3 Þ3 → AlOAlðCH3 Þ2 ⁎ þ CH4

ð1Þ

ðBÞ AlðCH3 Þ⁎ þ H2 O→AlOH⁎ þ CH4

ð2Þ

where ⁎ designates the surface species. The TMA and H2O halfreactions are performed in an ABAB…binary sequence to grow Al2O3. The growth rate of the films deposited after sequential exposures of TMA and H2O is nominally 0.11–0.13 nm per coating cycle at the reaction temperature of 177 °C [19,20]. A convenient method for dispersing and applying the coating on the diamond particles is gas fluidization in a fluidized bed reactor, which has the capability of good mixing, large gas-solid contact area and high efficiency of mass and heat transfer. With appropriate expansion of a bed of particles, the particles behave much as a fluid and the precursors can be introduced into the bed for reaction with the surface of the particles. However, Geldart Type C particles, such as micron-sized diamond particles, are difficult to fluidize due to strong interparticle forces. Channels form and particles fluidize as aggregates [24– 26]. To date, several studies concerning the fluidization of aggregates of fine particles have been carried out with externally assisted fluidization involving the use of additional forces generated by mechanical vibration [25–27], sound activation [28,29], or a magnetic field [30] to enhance the dynamics of the fluidization in the bed. Mechanical vibration has proven to be an effective means for improving the fluidization of cohesive particles [25–27]. Vibration can break up the large aggregates generated by cohesive interparticle forces, thereby achieving stable fluidization. The Al2O3 ALD coating on diamond film substrates for electronic device applications has been demonstrated before [31]. The Al2O3 ALD nanocoating process on diamond particles in larger quantities has not been demonstrated, although Ti-ALD [32] and Si-ALD [33] on diamond particles has been carried out. This paper demonstrates that ultra-thin Al2O3 nanocoatings can be successfully deposited on primary diamond particles by ALD using a scalable fluidized bed reactor. The technology has a variety of industrial applications. 2. Experimental The fluidized bed system consists of a reactor column, a vibration generation system, a gas flow control system, and a data acquisition and control system with LabView®, which has been described in detail previously [21,22]. The reactor itself is composed of a stainless steel column, surrounded by a clamshelltype furnace. A porous stainless steel plate with a pore size of 0.5 μm serves as the gas distributor. Mechanical vibration is

generated by two synchronized vibro-motors mounted on the two sides of a steel-made base that rests on four large springs. A vibration frequency of 20 Hz and amplitude of 3 mm are used. Nitrogen gas is used as fluidization gas and purge gas. The flow of nitrogen gas is maintained using a mass flow controller from MKS instruments. The flow rate of precursors, TMA obtained from Sigma Aldrich and deionized H2O, is adjusted using needle valves to ensure that a precursor pressure is high enough for particle fluidization. The pressure of the precursor vapors is kept constant during all the reactions. All valves used to provide the transient dosing are automatically controlled by LabView® and pressure measurements are recorded to monitor the progress of each dosing cycle. The diamond particles used in this study have the particle size of ~ 2–22 µm. The surface area of the particles is 0.39 m2/g, and the bulk density is 3.52 g/cm3. For a typical run, about 50 g of diamond particles are loaded into the reactor. Precursors are fed separately through the distributor of the reactor using the driving force of their vapor pressures. The reaction temperature is 177 °C. Before the reaction, the particles are outgassed at 177 °C for about 2 h. Each precursor is fed for enough time to make sure that all active sites are saturated. N2 is then fed as the purge gas to help remove precursor from the system. The system is pumped down to 50 mTorr pressure prior to the dose of the next precursor. In this manner, there is no overlap of two precursors. The ALD sequential surface reactions are easily carried out and undesirable chemical vapor deposition (CVD) is avoided. The composition of Al2O3 films is characterized using a PHI 5600 Physical Electronics X-ray photoelectron spectroscopy (XPS) system with a high-energy resolution analyzer. The conformality of the Al2O3 nanocoatings on the diamond particles is confirmed by a Philips CM 10 transmission electron microscope (TEM). The aluminum concentration on diamond particles is determined using an Applied Research Laboratories ICP-AES 3410+ inductively coupled plasma atomic emission spectroscope (ICP-AES). The morphologies of particles before and after coating are observed using a JSM-6480LV scanning electron microscope (SEM). The specific surface area is measured using a Quantachrome Autosorb-1. 3. Results and discussions 3.1. Fluidization studies at reduced pressure Fine particles are usually difficult to fluidize due to the highly cohesive nature of these particles, resulting from Van der Waals adhesion forces [34]. The fluidization quality progressively decreases when the absolute pressure decreases below about 150 Torr. In this study, fluidization behavior of diamond particles is investigated at 177 °C and reduced pressure (about 4 Torr) using mechanical vibration to help overcome interparticle forces. The fluidization experiments are carried out with an initial pressure of about 50 mTorr. To examine fluidization at low pressures, the pressure drop across the fluidized bed is recorded for a range of high purity N2 gas flow rates. To obtain a baseline pressure profile, pressure drop values are obtained

X.H. Liang et al. / Diamond & Related Materials 17 (2008) 185–189

187

Fig. 1. Pressure drop across the fluidized bed versus superficial gas velocity for diamond particles at 177 °C and reduced pressure.

Fig. 3. Transmission electron micrograph of alumina coated diamond particle after 100 cycles.

without particles in the reactor. These values are then subtracted from the pressure drop values obtained for the reactor with particles. This provides the pressure drop resulting from the particle bed alone. The pressure drop across the fluidized bed of particles reaches a constant value at the minimum fluidization velocity. At this point, all of the diamond particles are being fluidized. The fluidization behavior of diamond particles in the fluidized bed reactor is shown in Fig. 1. Pressure drop across the bed increases and reaches a plateau region when the superficial gas velocity increases above 0.37 cm/s, suggesting that the bed is fully fluidized. Mechanical vibration improves the fluidization quality by generating a pressure fluctuation that is transferred to the bed via a gas gap [35]. This helps to partly overcome some interparticle forces and reduce both the average size and the segregation of agglomerates in the bed, thus improving the fluidization quality of cohesive particles.

XPS measurements are carried out on uncoated and Al2O3 coated diamond particles after 100 cycles. The analysis is

performed using an aluminum source, pass energy of 187.85 eV and an energy step of 0.8 eV. In Fig. 2, the spectrum for the uncoated diamond particles shows a strong C 1s photoelectron peak at 284.7 eV. In contrast, the carbon spectrum for coated particles reveals a much weaker photoelectron intensity at 284.7 eV. This is expected for the diamond particles encapsulated by conformal Al2O3 coatings. However, the carbon XPS signal cannot be completely attenuated since some of it corresponds to surface carbon. Photoelectrons from the Al2O3 coated particles are observed at 118.7 eV (Al, 2s), 73.9 eV (Al, 2p) and 530.7 eV (O, 1s), which correspond to Al–O bonds of Al2O3. The XPS results verify the composition of the deposited Al2O3 films on the diamond particles. Previous studies confirmed that Al2O3 ALD films grown at 177 °C had an amorphous structure [23]. A transmission electron micrograph of the Al2O3 coated diamond particles provides visual confirmation that Al2O3 films are conformally coated on the particle surface. A representative TEM image of coated particles after 100 coating cycles is shown in Fig. 3. The TEM image reveals that the Al2O3 film is very conformal and smooth. The film thickness is ~ 12 nm, indicating that the growth rate of Al2O3 on diamond at 177 °C is

Fig. 2. X-ray photoelectron spectroscopy spectra of (a) uncoated and (b) alumina coated diamond particles after 100 cycles.

Fig. 4. Aluminum concentration on diamond particles versus the number of coating cycles.

3.2. Alumina ALD on diamond particle surfaces

188

X.H. Liang et al. / Diamond & Related Materials 17 (2008) 185–189

Fig. 5. Alumina film thickness versus the number of coating cycles.

0.12 nm per cycle. This growth rate is similar to those growth rates for Al2O3 films deposited on other substrates at 177 °C [19,20]. The concentration of Al2O3 on the coated particles is characterized by ICP-AES. Fig. 4 shows that the concentration of aluminum is almost directly proportional to the number of coating cycles, which indicates a constant growth rate of Al2O3 films and a linear dependence between the film thickness and the number of growth cycles. Therefore, deposition can begin during the first cycle. Recent FTIR studies of the diamond

Fig. 7. Surface area of diamond particles versus the number of coating cycles.

surface indicate the presence of hydroxyl groups on diamond surface [33]. These may result from adsorbed oxygen containing functional groups which can initiate the Al2O3 ALD growth during the first coating cycle. The thickness of Al2O3 films on particle surface can be estimated using the measured aluminum concentration on the particle surface, an Al2O3 density of 3.5 g/cm3, and the surface area of the uncoated diamond particles (0.39 m2/g). The Al2O3 film thickness has been calculated (see Appendix A) using the ICP-AES data and are summarized in Fig. 5. From these results, it is seen that the film grows linearly with the number of coating cycles at a growth rate of 0.12 nm Al2O3 per cycle. These results agree with the result observed from TEM imaging. SEM imaging (Fig. 6) and surface area analysis (BET, Fig. 7) before and after particle ALD coating indicate that primary diamond particles are coated without becoming aggregated. For SEM carried out at 15 kV, there does not appear to be any difference between the particles before and after coating. Likewise, the surface area of the coated particles after 50 cycles was 0.34 m2/g, which is very close to that of uncoated diamond particles. This indicates that primary particles are coated as opposed to bridging multiple particles together as aggregates. 4. Conclusions

Fig. 6. Scanning electron micrographs of (a) uncoated and (b) alumina coated diamond particles after 100 cycles.

Ultra-thin Al2O3 films are deposited on primary micronsized diamond particles by atomic layer deposition (ALD). The deposition is carried out in a scalable fluidized bed reactor at 177 °C and a reduced pressure using mechanical vibration to overcome interparticle forces between primary particles. The composition of alumina films on the diamond particle surface is confirmed by X-ray photoelectron spectroscopy. The conformality of the deposited films is observed by TEM; the growth rate of Al2O3 at 177 °C is 0.12 nm per cycle. The results of inductively coupled plasma-atomic emission spectroscopy indicate that no nucleation period is needed for alumina ALD on diamond particle surface, and deposition can begin during the first coating cycle. The alumina film growth rate based on ICP-AES data corresponds well with the growth rate observed from TEM imaging. The results of SEM and surface area of the uncoated and nanocoated diamond particles show that no

X.H. Liang et al. / Diamond & Related Materials 17 (2008) 185–189

agglomeration is observed during the coating process. Alumina can be deposited on diamond particles in a scalable fluidized bed reactor for a variety of industrial applications. Acknowledgements The authors thank Fred Luiszer and Thomas Giddings at the University of Colorado for assistance with the ICP-AES and TEM analysis. Appendix A VAl2 O3 ;m3 =g ¼

½ppm Al mol Al mol Al2 O3  0:5  27 g Al mol Al 106 

tfilm;nm ¼

ð3Þ

102 g Al2 O3 1  qAl2 O3 ;g=m3 mol Al2 O3

VAl2 O3 ;m3 =g 109 nm  m Surface aream2 =g

ð4Þ

References [1] N.E. Garner, Journal of Petroleum Technology 19 (1967) 937. [2] A. Ortega, D.A. Glowka, Society of Petroleum Engineers Journal 24 (1984) 121. [3] K. Saijo, M. Yagi, K. Shibuki, S. Takatsu, Surface & Coatings Technology 47 (1991) 646. [4] E.A. Ekitnov, A.S. Ivanov, A.F. Pal', A.N. Ryabinkin, A.O. Serov, A.N. Starostin, V.E. Fortov, R.A. Sadykov, N.N. Mel'nik, A. Presh, Doklady Physics 50 (2005) 351. [5] G.S. Bobrovnitchii, E.S. Osipov, M. Filgueira, International Journal of Refractory Metals & Hard Materials 21 (2003) 251. [6] M.G. de Azevedo, A. Potemkin, A.L.D. Skury, R.N.D. Faria, Diamond and Related Materials 10 (2001) 1607. [7] Y.S. Ko, T. Tsurumi, O. Fukunaga, T. Yano, Journal of Materials Science 36 (2001) 469. [8] T.J. Shin, J.O. Oh, K.H. Oh, D.N. Lee, Diamond and Related Materials 13 (2004) 488. [9] F.T.C. Lima, G.S. Bobrovnitchii, M. Filgueira, Advanced Powder Technology Iv 498–499 (2005) 225.

189

[10] S. Spriano, Q. Chen, L. Settineri, S. Bugliosi, Wear 259 (2005) 1190. [11] C. Campillo, S. Ilias, C.F.M. Borges, M. Moisan, L. Martinu, New Diamond and Frontier Carbon Technology 11 (2001) 147. [12] O. Semchinova, D. Uffmann, H. Neff, E.P. Smirnov, Journal of the European Ceramic Society 16 (1996) 753. [13] O.K. Semchinova, J. Graul, H. Neff, G. Holzhuter, E.P. Smirnov, V.Y. Davydov, Diamond and Related Materials 8 (1999) 2140. [14] K. Greene, D. Egan, S. Kelly, S. Nailer, S. Melody, Industrial Diamond Review 66 (2006) 57. [15] J.R. Wank, S.M. George, A.W. Weimer, Powder Technology 142 (2004) 59. [16] L.F. Hakim, S.M. George, A.W. Weimer, Nanotechnology 16 (2005) S375. [17] L.F. Hakim, C.L. Vaughn, H.J. Dunsheath, C.S. Carney, X.H. Liang, P. Li, A.W. Weimer, Nanotechnology 18 (2007) art. No. 345603. [18] D.M. King, J.A. Spencer, X.H. Liang, L.F. Hakim, A.W. Weimer, Surface & Coatings Technology 201 (2007) 9163. [19] J.W. Elam, Z.A. Sechrist, S.M. George, Thin Solid Films 414 (2002) 43. [20] L.F. Hakim, J. Blackson, S.M. George, A.W. Weimer, Chemical Vapor Deposition 11 (2005) 420. [21] X.H. Liang, L.F. Hakim, G.D. Zhan, J.A. McCormick, S.M. George, A.W. Weimer, J.A. Spencer, K.J. Buechler, J. Blackson, C.J. Wood, J.R. Dorgan, Journal of the American Ceramic Society 90 (2007) 57. [22] X.H. Liang, S.M. George, A.W. Weimer, N.H. Li, J.H. Blackson, J.D. Harris, P. Li, Chemistry of Materials 19 (2007) 5388. [23] L.F. Hakim, J.A. McCormick, G.D. Zhan, A.W. Weimer, P. Li, S.M. George, Journal of the American Ceramic Society 89 (2006) 3070. [24] D. Geldart, Powder Technology 7 (1973) 285. [25] L.F. Hakim, J.L. Portman, M.D. Casper, A.W. Weimer, Powder Technology 160 (2005) 149. [26] J.R. Wank, S.M. George, A.W. Weimer, Powder Technology 121 (2001) 195. [27] C.B. Xu, J. Zhu, Powder Technology 161 (2006) 135. [28] Q.J. Guo, H. Liu, W.Z. Shen, X.H. Yan, R.G. Jia, Chemical Engineering Journal 119 (2006) 1. [29] Q.J. Guo, M.H. Wang, Y. Li, C.H. Yang, Chemical Engineering & Technology 28 (2005) 1117. [30] Q. Yu, R.N. Dave, C. Zhu, J.A. Quevedo, R. Pfeffer, AIChE Journal 51 (2005) 1971. [31] N. Kawakami, Y. Yokota, T. Tachibana, K. Hayashi, K. Kobashi, Diamond and Related Materials 14 (2005) 2015. [32] J.B. Zang, J. Lu, Y.H. Wang, X.H. Qi, Y.G. Yuan, Advances in Grinding and Abrasive Technology Xiii 304–305 (2006) 48. [33] J.B. Zang, J. Lu, X.H. Qi, Y.H. Wang, Diamond and Related Materials 15 (2006) 1434. [34] J. Visser, Powder Technology 58 (1989) 1. [35] A.W. Siebert, D. Highgate, M. Newborough, Applied Thermal Engineering 19 (1999) 37.