J. Electroanal. Chem., 111 (1980) 61--70 © Elsevier Sequoia S.A., Lausanne - - Printed in The Netherlands 61 INITIAL STAGES OF ANODIC O X I D A T I O N OF P O L Y C R Y S T A L L I N E COPPER E L E C T R O D E S IN A L K A L I N E SOLUTION J O H N M.M. D R O O G , C O R R I E A. A L D E R L I E S T E N , P E T E R T. A L D E R L I E S T E N and G O S S E A. B O O T S M A Van't Hoff Laboratory, State University of Utrecht, Padualaan 8, Utrecht (The Netherlands) (Received 12th O c t o b e r 1979; in revised f o r m 23rd J a n u a r y 1980) ABSTRACT The first stages o f the anodic o x i d a t i o n o f polycrystalline c o p p e r electrodes in N a O H solutions were studied by potential sweep v o l t a m m e t r y and ellipsometry. F o r m a t i o n o f bulk C u 2 0 was f o u n d to be preceded by e l e c t r o s o r p t i o n o f o x y g e n species, that occurs in t w o successive stages, each represented by a current peak, corresponding to a different s u b m o n o layer state with a different adsorption energy. This surface o x i d e was f o r m e d via r a n d o m electrodeposition. The width o f the first current peak indicates the presence o f lateral attractive interactions in the c h e m i s o r b e d layer. The surface layer did n o t s h o w any ageing effect. (I) I N T R O D U C T I O N Although the anodic oxidation o f copper in alkaline solutions has been extensively studied b y means of galvanostatic [1--7 ] and p o t e n t i o d y n a m i c techniques [7--20], the initial oxidation processes have received little attention. A study o f the first oxidation processes on the bare metal surface should help us to obtain a better understanding of the copper anode. Ambrose et al. [11] reported cyclic voltammograms o f copper in 1 mol dm -3 KOH. A voltammogram similar to the one t h e y published is shown in Fig. 1, for copper in 1 tool dm -3 NaOH. They were the first to observe b y careful examination at high sensitivity, and especially at higher sweep rates, the presence of a very small peak (peak 1 in Fig. 1) with a potential negative to the peak due to formation o f Cu20 (peak 2). Peak 3 is generally associated with formation of Cu(II) species [11,13,19,20]. Peaks 4 and 5 correspond to the Cu(II)/Cu(III) redox couple [11,19]. Peak 6 is assigned to reduction of Cu(II) to Cu(I) [11,13] or to reduction of Cu(I) to Cu [19], and peak 7 represents complex electroreduction to Cu [11,13,19]. Peak 1 was tentatively assigned b y Ambrose et al. to the formation of soluble species according to Cu + 2 O H - -~ Cu(OH)~ + e (1) Recently, Fletcher et al. [20] have reported cyclic voltammetry measurements on polycrystalline copper electrodes in 0.1 mol dm -3 LiOH. They concluded that the first anodic process is dissolution o f Cu(I). At somewhat higher poten- 62 20 ~/mA 3 10 0 10 E/V 1'~ - 1~0 oL5 0 05 Fig. 1. Triangular s w e e p v o l t a m m o g r a m for a c o p p e r e l e c t r o d e in 1 mol d m -3 NaOH a f t e r several o x i d a t i o n and r e d u c t i o n cycles, 22°C, d E / d t = 90 m V s -] , E vs. " A r g e n t h a l " elect r o d e (Ag/AgCl, KC1 (3M)//, E = 207 m V vs N H E ) A p p a r e n t surface area 0 28 c m 2. tiais, their voltammograms show a peak, which was attributed to the formation of a monolayer of Cu20, followed by a potential region where a thick Cu20 layer is formed. Miller [8], on the other hand, concluded from experiments with rotating ring<lisc electrodes that for Cu20 no pre-film free dissolution could be detected. Therefore, the initial processes on copper anodes deserve further attention. In this paper we report on combined electrochemical and ellipsometric measurements of carefully polished polycrystalline electrodes in NaOH solution. The ellipsometric technique was expected to give information on the possible formation o f surface layers. Our results indicate t h a t the first stage in the anodising of copper electrodes (peak 1 in Fig. 1) is the electrosorption o f oxygen species. This deposition occurs in two successive stages represented by two current peaks corresponding to different submonolayer states. (II) E X P E R I M E N T A L The working electrodes were copper discs (dia. 6 m m , thickness 4--5 mm) cut from a polycrystalline copper rod (Materials Research, 99.999% pure) and sealed into acrylic resin (Technovite 4071) in such a way t h a t only the top surface o f the metal was exposed to the solution. The electrodes were m o u n t e d on glass specimen holders, which were adjustable to allow proper alignment of the specimens to be studied by ellipsometry. The electrode surfaces were lightly rubbed with carborundum paper and polished with diamond pastes containing diamond particles as small as 0.25 pm, and then with alumina powders containing particles as small as 0.05 pm. Next the electrodes were, in some cases, chemically etched in 20% nitric acid. After polishing the electrodes were cleaned ultrasonically in twice<listilled water. Before measurements were taken, the copper surfaces were cathodically 63 reduced at --1200 mV to remove any oxide films t h a t might have been present. Sometimes it was necessary to activate the electrodes electrochemically to remove adsorbed impurities from the surface (see Section III.l.1). An Ingold "Argenthal" electrode [Ag/AgCl, KC1 (3M)] ; E = 207 mV vs. NHE), separated from the main compartment, served as reference electrode. All potentials quoted in this paper are given with reference to this electrode. The wavelength o f the ellipsometer light was 632.8 nm and the angle o f incidence was 65 + 0.5 ° . Further experimental details were the same as in ref. 21. (III) R E S U L T S (III.1) Cyclic voltammetry measurements (111.1.1) Effect of potential sweep limit Figure 2 shows a typical voltammogram recorded with a sweep rate o f 20 mV s -1 and increasing sweep widths for the potential region o f peak 1. Provided t h a t the electrodes have been carefully cleaned, the general shape o f the curves is the same for mechanically polished electrodes, electrodes etched in 20% nitric acid, and for electrodes t h a t have been roughened by oxidation-reduction cycles over the potential region from --1.6 V to 0.8 V. With roughened electrodes the size o f the voltammograms increased by a factor of about 4. '/BA 20 10 -10 E/V I I -08 I I i -06 Fig. 2. P o t e n t i o d y n a m i c c h a r g i n g curves f o r c o p p e r u p t o various p o t e n t i a l s ; 1 tool d m -3 NaOH, 22°C, dE~dr = 20 m V s -1. A p p a r e n t surface area 0 . 2 8 c m 2. 64 Figure 3 shows t h a t the voltammogram changes when the electrode is left in contact with the solution for several hours at a potential E = --0.8 V. Current densities are smaller, and the sharp peaks disappear. The original voltammogram can be found again after "electrochemical activation" of the electrode. The activation consists of applying some short oxidation pulses (0.1--0.2 s at E = 1000 mV), and of subsequently reducing the products formed at potentials Where hydrogen evolution is just beginning. The voltammograms in Fig. 2 clearly show two anodic peaks at potentials negative to the peak that corresponds to bulk formation of Cu20. The first is at --680 mV and has its cathodic counterpart at --690 mV. The second is at --630 mV and has its counterpart at - 6 4 0 mV. The peak at - 6 8 0 mV is very sharp and its half-width is estimated to be no more than 15 mV. Except in the case with the highest reversal potential, the current drops almost immediately to cathodic values after reversal of the potential sweep. It should be noted t h a t the anodic curves almost coincide. The charges under the anodic part of the curves are n o t markedly different from those in the corresponding cathodic parts. They are represented in Fig. 4. Comparison of the integrated charge-potential profile with the potentiodynamic i--E curve shows that the Q--E line exhibits little structure in relation to that indicated differentially by the i--E curve itself. Figure 5 gives a typical example of the changes in the ellipsometric parameters A and ~ as a function of the electrode potential during a triangular sweep. The plotted parameters 5A and ~ are defined by ~A = ~ -- A and 5~ = ~ -~, where ~ and ~ represent the parameters for the bare copper substrate at ] I I i i 10 60 4C 2C -10 ,E/ V i -1 i 08 J I -06 J -07 , E/V -06 , -0 Fig. 3. P o t e n t i o d y n a m i c charging curves for c o p p e r in 1 m o l d m -3 NaOH, 22°C, d E / d r = 20 m V s -1 : (a) clean e l e c t r o d e surface; (b) e l e c t r o d e surface left in c o n t a c t w i t h t h e solut i o n for a b o u t 2 h at a p o t e n t i a l E = --0.8 V. Fig. 4. Charges vs. t h e p o t e n t i a l o f scan reversal. Cu in 1 mol d m -3 NaOH, 22°C, d E / d r 20 m V s -1 . = 65 i ~- i i i 5A/deg 02 04 06 01 02 E/V I 1 i -O8 i i I -06 Fig. 5. Change in A and ~ during potential scanning up to --540 mV. Cu in 1 mol dm -3 NaOH, 22°C, d E ~ d r = 20 mV s -1. --1000 mV. The values of :~ and ~ were measured for 15 different electrodes. From these we calculated the optical constants o f the pure copper, which proved to be n = 0.39 + 0.13 and k = 3.80 + 0.22. We f o u n d t h a t the ellipsometer light had no influence on the cyclic voltammograms. Figure 5 shows a slow increase in b o t h 8A and 5~ in the region --1000 mV to --700 mV. At --700 mV there is a m u c h sharper increase in 5A and 5~. There is a very little hysteresis between the positive and the negative sweep. (111.1.2) Effect o f sweep rate Voltammograms were measured at sweep rates s varying from 10 mV s -1 to 2 V s -1 . For the two anodic and the two cathodic peaks in the voltammograms, plots of In ip vs. I n s give straight lines with slopes between 0.90 and 0.94. The peak potentials hardly shift (10--20 mV) with increasing sweep rate. In Fig. 6 the anodic charges Qa are plotted for voltammograms measured at different sweep rates. The potential of sweep reversal was --580 mV. Maximum values of ~iA during a potential sweep are plotted in the same figure. (111.2) Cyclic voltammetry with potential sweep arrest at the upper potential Figure 7 shows voltammograms at a sweep rate of 20 mV s -1, where the sweep was arrested at various potentials E u of sweep reversal for 0, 15, 30 and 60 s. The charges under the cathodic current peaks indicate whether or n o t extra oxidation products have been formed during the sweep arrest. From Figs. 7a and 7b it is clear t h a t during the first 15 s of waiting time a little more oxidation has taken place, because the cathodic currents are slightly higher. The curves 66 i i , i ,~l ~ i I , ,,i 02 2c s / m Y s~1 - - L I 10 I I I I 50 Fig. 6 Qa a n d ~ A vs. LIL I00 I I i I 500 sweep rate i i I I 1000 s w h e r e t = 3 0 s c o i c i d e e x a c t l y w i t h t h o s e w h e r e t = 1 5 s. A l s o in Fig. 7 c t h e f u r t h e r oxidation can be seen to have been small. Here a very slight difference can be seen between the curve with a waiting time of 15 s and that with a waiting t i m e o f 3 0 s. In t h e v o l t a m m o g r a m s w i t h an arrest p o t e n t i a l o f - - 5 4 0 m V t h e charge under the cathodic current keeps increasing considerably with waiting I0 '&A 10 '/~A C -1C -IC IC ~/~A IF f/. A 0 -I0 E/V _0~8 L J -06 -08 -06 Fig. 7. Voltammograms with potential-sweep arrest of t seconds at the upper potential. Cu in 1 tool dm -3 NaOH, 22°C, d E ~ d r = 20 mV s -1; (a, b, c) t = 0, 15 and 30 s; (d) t = 0, 15, 30 and 60 s. E u = --660 mV (a);--600 mV ( b ) ; - - 5 6 0 mV (c); --540 mV (d). 67 time. Part of the reduction current is n o w shifted to potentials more negative than --700 mV. In this case the ellipsometric measurements also show a continuing increase in 5A during the sweep arrest. The peak potentials o f the cathodic waves do n o t shift with waiting time in any of the voltammograms in Fig. 7. (III.3) Potential step measurements Current responses to single potential step perturbations with starting potential --0.8 V and upper potential Eu in the region --700 mV to --500 mV are shown in Fig. 8. The potential step was followed b y a negative potential sweep, with a sweep rate of 20 mV s -1. The current--time curve contains information a b o u t the kinetics o f the oxidation process. The charge under the cathodic reduction curve gives the a m o u n t o f oxidation products, and the potentials of the cathodic current peaks give some indication o f the character of the anodically formed species. /-0 ,/~A 30 20 IC t tA ' 2~0 1'o ® 1c .10i , -20 ® E/V -08 -06 -0z, Fig. 8. P o t e n t i a l s t e p m e a s u r e m e n t s (starting p o t e n t i a l - - 0 . 8 V) f o l l o w e d b y p o t e n t i o d y n a m i c r e d u c t i o n ( d E / d t = 20 m V s -1 ). (1) E u = - - 6 9 5 m V ; (2) E u = - - 6 5 5 m V ; (3) E u = - - 5 9 0 m V ; (4) E u = - - 5 5 5 m V ; (5) E u = - - 5 4 0 m V . 68 The current--time curves are all monotonically falling curves, with the exception of the curve at E u = --540 mV where after about 10 s the current starts rising again. In the reduction curves it is seen t h a t the a m o u n t o f charge increases markedly between Eu = --555 mV and Eu = --540 mV. (IV) DISCUSSION The reversible Cu/Cu20 potential at a pH = 14 is k n o w n to be --563 mV vs. the "Argenthal" reference electrode [22]. In Fig. 2 two anodic peaks are seen in the voltammogram at potentials lower than this reversible potential for bulk oxide formation. The ellipsometric curves in Fig. 5 show t h a t there is a clear change at a potential o f --700 mV, i.e. at the beginning of the first peak in the voltammogram. In this region no detectable surface roughening occurs; as from the first sweep the current--voltage curves remain the same, as do the ellipsometric parameters ~ and 5. It is thus clear that the anodic peaks in Fig. 2 represent electrosorption of oxygen species and that the cathodic peaks represent the corresponding desorption process. With potential sweeps up to --550 mV with a sweep rate of 20 mV s -1, the deposition process is accompanied by an a m o u n t of charge of 40 pC. If for our electrodes the roughness factor is taken as 1 and an equal distribution of the three lowest index planes is assumed, then a complete monolayer of Cu(I) atoms corresponds to 70 pC. The voltammograms in Fig. 2 show that electrosorption of oxygen species occurs in two successive stages each represented by a peak corresponding to a different submonolayer state, with a different adsorption energy. The problem here is whether or n o t the multiple states of oxygen chemisorption arise simply because o f a p r i o r i heterogeneity of the surface, e.g. exposure of various crystal planes in which the adsorption sites have different energies. The fact that the current peak at --680 mV is very sharp indicates the existence of areas on the electrode which have only one type o f adsorption site, with attractive lateral interactions between the adsorbed species [29,30]. For example, a p r i o r i heterogeneity has been f o u n d in the underpotential deposition of metal atoms on foreign metal substrates [23] and, very recently, in hydrogen chemisorption on platinum electrodes [24]. Gas phase studies have shown that oxygen chemisorption on copper single crystals is a very plane-specific reaction, and that the C u ( l l 0 ) surface is much more reactive to oxygen than the (100) surface, which in turn is more reactive than the (111) surface [25--28]. However, the existence of a p r i o r i heterogeneity can be proved only by investigating this electrochemical reaction at carefully prepared copper single crystal surfaces. Figure 3 shows t h a t when the electrode is left in contact with the electrolyte for a long time, the electrode becomes contaminated. This means that a number o f adsorption sites are blocked and are no longer active in the chemisorption of oxygen species. Apparently these adsorbed impurities can be oxidised anodically, and the reaction products can be removed at potentials where hydrogen evolution is just beginning. The scan arrest measurements do n o t show any shift o f the cathodic current peak, so an "ageing" process such as t h a t reported for oxygen chemisorption on platinum and gold electrodes [31--33] was not found. 69 At the highest potential (Fig. 7d) the oxidation becomes more and more irreversible in the sense t h a t a more cathodic potential is required to reduce the oxide film. At potentials lower than --560 mV, the oxidation--reduction behaviour of copper is rather reversible, compared to noble metal electrodes where hysteresis occurs between oxide formation and reduction of the surface oxide film [34--35]. From the identical dependence of Qa and 5A on the sweep rate s (Fig. 6) we conclude t h a t 5 A is proportional to the charge Qa and t h a t there is no evidence for dissolution of copper species. This is in contrast with measurements on silver electrodes in NaOH [21], where considerable dissolution o f silver species was shown to occur. Current--time curves in Fig. 8 are monotonically falling curves. This means t h a t there is r a n d o m electrodeposition, because otherwise a peak, characteristic for a nucleation and growth process, would have been f o u n d in the i--t profile. The same conclusion follows from the voltammograms in Fig. 2 where, upon sweep reversal, there is never a continuing increase of current. The initial part of the i--t curve at --500 mV is not essentially different from the initial parts o f the other curves. This means t h a t at potentials for f o r m a t i o n of bulk Cu20 a surface film is first formed, and this is followed by in-depth oxidation. (V) CONCLUSION The first stage in the anodic oxidation o f polycrystalline copper electrodes in NaOH solutions is the electrosorption of oxygen species. The process occurs in two successive stages, represented by two current peaks, corresponding to different submonolayer states. ACKNOWLEDGMENTS The authors wish to t h a n k Dr. J.H. Sluyters and Dr. M. Sluyters-Rehbach for valuable discussions. REFERENCES 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 E. Mflller, Z. E l e k t r o c h e m . , 1 3 ( 1 9 0 7 ) 1 3 3 . W. F e i t k n e c h t a n d H . W . L e n e l , Helv. C h i m . A c t a , 2 7 ( 1 9 4 4 ) 7 7 5 . A. H i e k l i n g a n d D. T a y l o r , T r a n s . F a r a d a y S o c . , 4 4 ( 1 9 4 8 ) 2 6 2 . J.S. Halhday, Trans. Faraday Soc., 50 (1954) 171. R.W. O h s e , Z. P h y s . C h e m . N . F . , 21 ( 1 9 5 9 ) 4 0 6 . M . J . D l g n a m a n d D . B . G i b b s , C a n . J. C h e m . , 4 8 ( 1 9 7 0 ) 1 2 4 2 . H . P . L e c k i e , J. E l e c t r o c h e m . S o c . , 1 1 7 ( 1 9 7 0 ) 1 4 7 8 . B. Miller, J. E l e c t r o c h e m . S o c . , 1 1 6 ( 1 9 6 9 ) 1 6 7 5 . N.A. Hampson, J.B. Lee and K.J. MacDonald, J. Electroanal. Chem., 32 (1971) 165. P . L . B o n o r a , G.P. P o n z a n o a n d M. B a s s o h , A n n . C h i m . , 6 2 ( 1 9 7 2 ) 6 3 6 . J. A m b r o s e , R . G . B a r r a d a s a n d D.W. S h o e s m i t h , J. E l e c t r o a n a l . C h e m . , 4 7 ( 1 9 7 3 ) 4 7 . J. A m b r o s e , R . G . B a r r a d a s a n d D.W. S h o e s m l t h , J. E l e c t r o a n a l . C h e m . , 4 7 ( 1 9 7 3 ) 6 5 . D . D . M a c d o n a l d , J. E l e c t r o c h e m . S o c . , 1 2 1 ( 1 9 7 4 ) 6 5 1 . D.W. S h o e s m l t h , T.E. R u m m e r y , D. O w e n a n d W. L e e , J. E l e c t r o c h e m . S o c . , 1 2 3 ( 1 9 7 6 ) 7 9 0 . D.W. S h o e s m i t h , T.E. R u m m e r y , D. O w e n a n d W. L e e , E l e c t r o c h i m . A c t a , 2 2 ( 1 9 7 7 ) 1 4 0 3 . D.W. S h o e s m i t h a n d W. Lee, E l e c t r o c h i m . A c t a , 2 2 ( 1 9 7 7 ) 1 4 1 1 . V . P . S e v e r d e n k o , V . A . L a b u n o v , E.M. K o s a r e w t s h , L . V . K o g i e t o v a n d V . M . P a r k u m , T h i n S o l i d Films, 41 (1977) 243. 70 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 V. A s h w o r t h a n d D. Fa~rhurst. J. E l e c t r o c h e m . Soc., 1 2 4 ( 1 9 7 7 ) 5 0 6 . A.M. Castro L u n a de M e d i n a , S.L. M a r c h i a n o a n d A.J. A r v / a , J. A p p . E l e c t r o c h e m . , 8 ( 1 9 7 8 ) 1 2 1 . S. F l e t c h e r , R . G . B a r r a d a s a n d J.D. P o r t e r , J. E l e c t r o c h e m . Soc., 1 2 5 ( 1 9 7 8 ) 1 9 6 0 . J.M.M. D r o o g , P.T. A l d e r l i e s t e n a n d G.A. B o o t s m a , J. E l e c t r o a n a l . C h e m . , 99 ( 1 9 7 9 ) 173. N. de Z o u b o v , C. V a n l e u g e n h a g h e a n d M. P o u r b a i x m M. P o u r b a a x ( E d , ) , A t l a s o f E l e c t r o c h e m x c a l E q u t h b r i a in A q u e o u s S o l u t i o n s , P e r g a m o n Press, L o n d o n , 1 9 6 6 , p. 3 8 4 . D M. K o l b m H. G e r i s c h e r a n d C.W. T o b i a s (Eds.), A d v a n c e s in E l e c t r o c h e m i s t r y and E l e c t r o c h e m i c a l E n g i n e e n n g , Vol. 11, Wiley, N e w Y o r k , 1 9 7 8 , Ch. 2. K. Y a m a m o t o , D.M. K o l b , R. K o t z a n d G. L e h m p f u h l , J. E l e c t r o a n a l . C h e m . , 96 ( 1 9 7 9 ) 2 3 8 . F.H.P.M. H a b r a k e n , E.Ph. K l e f f e r a n d G.A. B o o t s m a , Surf. Sci., 83 ( 1 9 7 9 ) 4 5 . F.H.P.M. H a b r a k e n a n d G . A . B o o t s m a , Surf. Scl., 87 ( 1 9 7 9 ) 3 3 3 . F.H.P.M. H a b r a k e n , G.A. B o o t s m a , P. H o f m a n n , S. H a c h l c h a a n d A M. B r a d s h a w , Surf. Sca , 88 (1979) 285. F.H.P.M. H a b r a k e n , C.M.A.M. Mesters a n d G . A . B o o t s m a , Surf. Sci., to b e p u b l i s h e d . E. Gilead1 a n d B.E. C o n w a y m B.E. C o n w a y a n d J . O ' M . Bockris (Eds.), M o d e r n A s p e c t o f E l e c t r o c h e m i S t r y , Vol. 3, B u t t e r w o r t h s , L o n d o n , 1 9 6 4 , Ch. 5. A S a d k o w s k i , J. E l e c t r o a n a l . C h e m . , 97 ( 1 9 7 9 ) 2 8 3 . H A n g e r s t e m - K o z l o w s k a , B.E. C o n w a y a n d W B.A. S h a r p , J. E l e c t r o a n a l C h e m . , 43 ( 1 9 7 3 ) 9. N R. De T a c c o m , A.J. C a l a n d r a a n d A.J. A r v i a , J E l e c t r o a n a l . C h e m . , 51 ( 1 9 7 4 ) 25. C M, F e r r o , A J. C a l a n d r a a n d A.J. A r v f a , J. E l e c t r o a n a l . C h e m . , 59 ( 1 9 7 5 ) 2 3 9 . R. W o o d s m A.J. Bard ( E d . ) E l e c t r o a n a l y t l c a l C h e m x s t r y , Vol. 9, Marcel D e k k e r , N e w Y o r k , 1 9 7 6 G. B~langer a n d A.K. V ~ h m A K. V1]h ( E d ) O x i d e s a n d O x i d e Films, Vol. 5, Marcel D e k k e r , N e w Y o r k , 1 9 7 7 , Ch. 1.