Th& ColidFilms, 189 (1990) 73 79 ,METALLURGICALAND PROTECTIVELAYERS
ZrB 2 C O A T E D O N C O P P E R P L A T E BY C H E M I C A L V A P O U R D E P O S I T I O N , A N D ITS C O R R O S I O N A N D O X I D A T I O N S T A B I L I T I E S SEIJI MOTOJIMA,KIMIEFUNAHASHIAND KAZUYUKIKUROSAWA* Department of Applied Chemistry, Faculty of Engineering, Gifu University, Gifu 501-11 (Japan) (Received October 16, 1989;accepted February 13, 1990)
A copper plate was coated with Z r B 2 layers from a gas mixture of Z r C 1 4 + B C I 3 + H 2 + A r at temperatures in the range 700-900°C. The lowest deposition temperature was very low, about 690 °C, which is 100-200 °C lower than that on other substrates. The corrosion stabilities of the copper plate to HC1 and HNO3 solutions were improved outstandingly by coating with the Z r B 2 layers.
l. INTRODUCTION Zirconium diboride (ZrB2; melting point, 3038 °C) is a representative refractory compound having excellent characteristics such as extreme hardness, affording good wear and abrasion resistivities, high chemical stability against corrosion or attack by molten salts and metals, and good oxidation stability 1'2. Zirconium diboride also has the lowest bulk electrical resistivity (7 ~tf~cm) among various metal borides 3, high thermoionic emission, excellent diffusion barrier performance 4' 5, etc. Accordingly, ZrB2 is a potential candidate for applications such as protective coatings against wear, abrasion and corrosion, for electrical contacts or diffusion barriers in semiconductor device fabrication, and for use as a photothermal solar absorber 2. ZrB 2 layers have been prepared by a chemical vapour deposition ( C V D ) 2'6'7, sputtering s'9 etc. Among these processes, the CVD process is the most important for obtaining dense and adherent ZrB2 coatings on metal substrates. We have reported on the preparation of TiB 2, TaB and TaB 2 layers on a plate of copper and copper alloys, and their corrosion and abrasion resistivities against acid solutions, sea water and whirled sea sands have been determined 10-14. In this work, we obtained a ZrB 2 layer on a copper plate by CVD using a gas mixture o f Z r C l 4 + BCI 3 + H2 + Ar at temperatures in the range 700-900 °C, and the corrosion stabilities to acid solutions and the oxidation stability of the ZrB2-coated copper plate were examined. *Present address: Industrial Research Institute of Aichi PrefectureGovernment, Nishishin-kawa, Hitotsugi-cho, Kariya-shi,Aichi-ken448, Japan. 0040-6090/90/$3.50
© ElsevierSequoia/Printed in The Netherlands
s. MOTOJIMA, K. FUNAHASHI, K. KUROSAWA
2. EXPERIMENTAL PROCEDURES
The apparatus used in this work is shown schematically in Fig. 1. A substrate (copper plate, 0.2mm x 10mm × 10mm) was hung from a tungsten hook in the central part of a vertical reaction tube (quartz, 26mm inner diameter) which was heated from the outside by a nichrome element. The copper plate was abraded by emery papers sequentially up to 1200 grade, degreased ultrasonically in acetone, and d(ied. Zirconium tetrachloride was prepared in situ by the chlorination of zirconium sponge at about 700 ~C, and boron trichloride was prepared by the chlorination of boron carbide (B4C) a t 800 "C, argon being used as a carrier gas. Total gas flow rates and combined flow rates ofZrC14 + BCI 3 were fixed at 5.7 m l - is and 0.20 m l - i s respectively. The ZrBz-coated copper plate (hereafter called the coated plate) was immersed in acid solutions at a given temperature, and the weight loss caused by corrosion was measured at intervals. The coated plate used as a sample for the corrosion or oxidation test was obtained at a temperature of 700 °C and a source gas flow ratio (BC13:ZrCI4, hereafter called the gas flow ratio (B:Zr)) of 1.5, unless otherwise described.
fE Fig. 1. Schematic diagram of the apparatus: A, reaction tube (quartz, 26mm inner diameter); B, substrate (copper plate, 0.2 mm x 10 mm x 10 mm); C, zirconium sponge; D, boron carbide (B4C); E, H2; F, C12 +Ar; G, argon or H2; H, chromel alumel thermocouple; I, gas outlet.
3. RESULTS AND DISCUSSION 3.1. Deposition parameters The effect of the reaction temperature on the mass gain of the copper plate after
CVD Z r B 2 ON C u
30 min of reaction time is shown in Fig. 2. An appreciable mass gain of the copper plate indicating deposition of the ZrB 2 layer was observed at a temperature of 700 °C, and the mass gain increased steeply with increasing reaction temperature, followed by retardation at temperatures above 800 °C. The lowest deposition temperature at which the characteristic colours of the copper plate disappeared completely, giving way to the tarnished silver colours of ZrB 2 layers, was about 690 °C, which is 100-200 °C lower than that on other substrates. The activation energy for the deposition of the ZrB 2 layers in the temperature range 750-900 °C was estimated to be 13.6 kcal mo1-1. This lowering effect of the deposition temperature of the ZrB 2 layers may be attributable to the strong promotion effect of the nucleation of ZrB 2 crystallites and/or to some catalytic activity for surface reaction on the copper plate at the initial stage of deposition. The same lowering effect of the deposition temperature was observed for TiB2 layers on the copper and copper alloys 1o-14. The effect of the gas flow ratio B:Zr on the mass gain of the copper plate after 30 min of reaction time is shown in Fig. 3. The maximum mass gain was obtained at a gas flow ratio B:Zr of about 1.5, which is lower than the stoichiometry (B:Zr = 2), irrespective of the reaction temperature, and the mass gain decreased steeply above or below this ratio. The ZrCl4-rich condition may be attributed to the different reactivity to hydrogen (the reduction of BC13 proceeds faster than that of ZrC14). A similar tendency has also been observed during the CVD of TiB2.
'~ 3 tD 4J
700 750 8 0' 0 8 '5 0 ' Reaction Temperature
2 4 13(31s/ZrC14
Fig. 2. Effect of reaction temperature on the mass gain of the copper plate for gas flow ratios BCI3:ZrCI 4 of 1.5 (©) and 2.0 (O) (total gas flow rate, 5.7 m l - 1 s; combined flow rate of ZrC14 + BCI 3, 0.20 m l - 1 s; reaction time, 30 min). Fig. 3. Effect o f g a s flow ratio BCIa:ZrCI4 on the mass gain of the copper plate for reaction temperatures of 850 ° C (IS]), 800 ° C (/x) and 750 ° C (©) (reaction time, 30 min; total gas flow rate, 5.7 m l - 1 s; combined flow rate of ZrC1, + BCI3, 0.20 m l - 1 s).
The mass gain of the copper plate was examined for the reaction temperature range 700-900 °C and reaction times of 0-120 min. The mass gain increased linearly with increasing reaction time irrespective of reaction temperature.
s. MOTOJIMA, K. FUNAHASHI, K. KUROSAWA
3.2. Identification and morphology X-ray diffraction analysis of the deposited layers obtained at a reaction temperature of 700 °C revealed only a single phase o f Z r B 2 (hexagonal). The surface appearances of the ZrB 2 layers obtained with a gas flow ratio B:Zr of 1.5 and for the temperature range 700 800 °C are shown in Fig. 4. Polycrystallites having apparent crystal facets were deposited at a reaction temperature as low as 700 °C, and the growth of the polycrystallites was accelerated outstandingly with increasing reaction temperature, whereby they formed plate-like crystals, which grew perpendicularly to the substrate surface at a temperature of 800°C (Fig. 4(c)). The temperature at which the growth of polycrystallites is accelerated is about 50-100 "~C lower than that for TiB 2 or TAB2, and a surface morphology similar to that shown in Figs. 4(b) and 4(c) was observed at a temperature of 850 °C for the TiB 2 layers deposited on a phosphor bronze plate 13. No apparent effect of the gas flow ratio B:Zr on the surface morphology was observed at the ratios B:Zr in the range 1 2.
(c) Fig. 4. Surface appearance of the ZrB 2 layers, for reaction temperatures of(a) 700" C, (b) 750 "~Cand (c) 800'C (reaction time 30 rain; gas flow ratio BCI3:ZrCI 4, 1.5).
Spontaneous crack formation on the ZrB 2 layer of 5 ~tm thickness or peeling of the layer from the copper plate was not observed at all after one month of exposure to the atmosphere. Furthermore, peeling of the ZrB 2 layer from the copper plate was not observed at all under a scratch test, although the formation of cracks and depressions on the layer by the action of a weighted scratch indenter was observed. That is, adherence of the ZrB 2 layer with the copper plate was excellent.
CVD ZrB 2 ON Cu
3.3. Oxidation and corrosion stabilities The stability to oxidation of the coated plate was estimated from the mass gain in an air atmosphere. The effect of the thickness of the ZrB 2 layer on the mass gain for I h of oxidation time is shown in Fig. 5. The mass gain decreased steeply with increasing thickness of the ZrB 2 layer and attained a constant value at a thickness abo~ : 2 p.r,1 for an oxidation temperature of 700 °C and above 8 p.m for 800 °C. The coated plate was immersed in 12 N HCI solutions maintained at 60 °C, and the mass loss was measured. The mass loss of the coated plate decreased linearly with increasing immersion time, and the rate of mass loss was much smaller than that of the bare copper plate. The effect of the thickness of the ZrB 2 layer on the mass loss is shown in Fig. 6. The mass loss of the coated plate decreased steeply with increasing thickness of'the ZrB 2 layer and attained a constant value at a thickness of about 4 ~tm, irrespective of the immersion time. However, the value obtained after 6 h of immersion was about 3 times of that for 3 h, because the ZrB 2 layer itself dissolves very slightly in 12 N HC1 solution. (~m)
4 6 Thickness (~m)
i 1 51~/
Fig. 5. Effect of the thickness o f the ZrB 2 layer on the mass gain of the ZrB2-coated copper plate exposed in air at elevated temperatures of 800 ° C (@), and 700 ° C (O) (reaction temperature, 700 ° C; gas flow ratio BC13:ZrCla, 1.5; oxidation time, I h). Fig. 6. Effect of the thickness of the ZrB 2 layer on the mass loss of the ZrB2-coated copper plate immersed in 12 N HC1 solutions for 3 h (O) and 6 h (@) (reaction temperature, 700 °C; gas flow ratio BC13: ZrCI4, 1.5; temperature o f 12 N HCI, 60 ° C ).
The coated plate was immersed in 3.2 N or concentrated H N O 3 solutions at room temperature, and the mass loss was measured. The effect of the immersion time in 3.2 N H N O 3 solutions on the mass loss of the coated plate obtained at a reaction temperature of 750 °C is shown in Fig. 7, with a bare copper plate as a reference sample. The bare copper plate dissolved very rapidly in 3 . 2 N H N O a solution. On the contrary, mass loss of the coated plate was scarcely observed for up to 2-3 h of immersion; at longer times increasing mass loss took place. The effect of the thickness of the ZrB 2 layer on the mass loss of the coated plate immersed in 3.2 N H N O 3 solution for 6 h is shown in Fig. 8. The mass loss of the coated plate obtained at a reaction temperature of 700 °C decreased steeply with increasing
S. MOTOJIMA, K. FUNAHASHI, K. KUROSAWA
Immersion Time (hrs) 2 4 6 8 10
Thickness 2 4
Fig. 7. Effect ofthe immersion time ofthe ZrB2-coated copper plate in 3.2 N HNO3 solutions of the mass loss, for ZrB 2 layer thicknesses of 1.7 l~m (A), 2.9 l~m (I-3), 4.6 l~m (O) and 7.7 p.m (©) and for a bare copper plate (- - - ) (reaction temperature, 750 ° C; gas flow ratio BC]3:ZrC] ~, 1.5; temperature of 3.2 N H N O 3, room temperature).
Fig. 8. Effect o f t h i c k n e s s o f t h e Z r B 2 1 a y e r o n t h e m a s s l o s s o f t h e Z r B a - c o a t e d c o p p e r p l a t e i m m e r s e d i n 3.2N H N O 3 solutions for reaction temperatures of 700'~C (©) and 750°C (Q) (gas flow ratio BC13:ZrCI4, 1.5; temperature of 3.2 N HNO3, room temperature; immersion time in 3.2 N HNOa, 6 h).
thickness of the ZrB 2 layer and reached zero at a thickness of about 51~m. The mass loss of the coated plated obtained at 750 °C was larger than that for 700 °C; this was probably caused by the grain growth, which resulted in the formation of openings or crevices in the ZrB 2 layer as shown in Fig. 4(b). The effect of the immersion time of the coated plate in concentrated HNO3 solutions maintained at room temperature on the mass loss is shown in Fig. 9, in
Immersion Time 30 60 90
o _o,,a~,O_O_o_o 2
 i I
Fig. 9. E•ect•ftheimmersi•ntim.e•fthe•rB2•c•atedc•pp•rp•ateinc•ncentratedHN•3s••uti•ns•n the mass loss, for ZrB2 layer thicknesses of 7.6 gm (D), 12.6 gm (O) and 13.9 gm (©) and for a bare copper plate (. . . . ) (reaction temperature, 750 ° C; gas flow ratio BC13: ZrCI4, 1.5, temperature of concentrated H N O 3, room temperature).
CVD Z r B 2 ON C u
which the c o a t e d plate was o b t a i n e d at a t e m p e r a t u r e o f 750 °C. It can be seen t h a t the m a s s o f the c o a t e d plate with a thick Z r B 2 layer decreased very g r a d u a l l y u p to a b o u t 40 m i n o f i m m e r s i o n time for 7.6 lam thick coats, to a b o u t 70 m i n for 12.6 ~tm c o a t s a n d to a b o u t 100min for 13.9~tm coats; the m a s s loss then accelerated, p r o b a b l y as a result o f dissolution o f the ZrB2 layer itself. 4. CONCLUSIONS A c o p p e r plate was c o a t e d with Z r B 2 layers by C V D using a gas m i x t u r e o f ZrCl4, BC13, H 2 a n d a r g o n at t e m p e r a t u r e s in the range 700-900 °C. G r a i n g r o w t h o f the Z r B 2 layer was accelerated at t e m p e r a t u r e s a b o v e 750°C, which is c o m p a r a b l e with the result for TiB 2, to f o r m plate-like crystals on the surface. T h e m a x i m u m d e p o s i t i o n rate o f the Z r B 2 layer was o b t a i n e d at a gas flow ratio BC13: Z r C l 4 o f 1.5, a n d the rate decreased steeply a b o v e o r below this value. T h e lowest d e p o s i t i o n t e m p e r a t u r e was very low, a b o u t 690 °C, which is 100-200 °C lower t h a n that o n o t h e r substrates. The c o r r o s i o n stabilities o f the c o p p e r plate to 12 N HCI, 3.2 N H N O 3 a n d c o n c e n t r a t e d H N O a solutions were o u t s t a n d i n g l y i m p r o v e d by c o a t i n g with the Z r B 2 layers. REFERENCES
1 W.C. Tripp and H. C. Grahan, J. Electrochem. Soc., 118 (1971) 1195. 2
E. RandichandD. D. Allred, ThinSolidFilms, 83(1981)393.
3 M.-A. Nicolet, Thin Solid Films, 52(1978)415. 4 J.R. Shappiro, J. J. Finnegan, R. A. Lux and D. C. Fox, Thin Solid Films, 119 (1984) 23. 5 J. R. Shappiro, J. J. Finnegan, R. A. Lux, D. C. Fox, J. Kwiatkowski, H. Katt¢lus and M.-A. Nicolet, J. Vac. Sci. Technol. A,3 (1985) 2255. 6 I.E. Campbell, C. F. Powell, D. H. Nowichi and B. W. Gonser, J. Electrochem. Sot., 96 (1949) 318. 7 J.J. Gebhardt and R. F. Cree, J. Am. Ceram. Soe., 48 (1965) 262. 8 J.R. Shappiro and J. J. Finnegan, Thin Solid Films, 107 (1983) 81. 9 U.K. Chakrabarti, H. Bartz, W. C. Dautremont-Smith, J. W. Lee and T. Y. Kometani, J. Vac. Sei. Teehnol. A,5 (1987) 196. 10 S. Motojima, M. Yamada and K. Sugiyama, J. Nucl. Mater., 105 (1982) 335. 11 S. Motojima and H. Kosaki, J. Mater. Sei. Lett., 4 (1985) 1350. 12 S. Motojima and K. Kobayashi, J. Less-Common Met., 114 (1985) 375. 13 S. Motojima and H. Hotta, J. Less-Common Met., 141 (1988) 327. 14 S. Motojima and R. Azuma, J. Mater. Sci. Lett., 23 (t988) 4375.