DOI: 10.1021/cg901156z
Effect of Supersaturation on the Growth of Zinc Oxide Nanostructured Films by Electrochemical Deposition
2010, Vol. 10 1189–1193
B. N. Illy,† B. Ingham,‡ and M. P. Ryan*,† †
Department of Materials, Imperial College London, Exhibition Road, London SW7 2AZ, United Kingdom, and ‡Industrial Research Limited, P.O. Box 31-310, Lower Hutt, New Zealand Received September 21, 2009; Revised Manuscript Received January 7, 2010
ABSTRACT: The changes in crystal growth habit of electrodeposited zinc oxide with zinc nitrate concentration are explained by changes in the levels of saturation at the electrode. Three growth regimes are found between 0.5 and 50 mM. For concentrations less than 2 mM, the growth is one-dimensional. Nanorods grow by screw dislocations from the outside inward, no coalescence is observed, and their surface shows pyramid-like features. For concentrations above 20 mM, the growth is twodimensional. Large levels of supersaturation favor the nucleation on the low indexes faces and large sheets are observed. In the intermediate regime of growth, the growth is pseudo three-dimensional. Nanorods with a conical ends grow initially before coalescing and forming dense films.
Introduction Zinc oxide is a wide band gap semiconductor that has emerged as a promising material in many areas such as solar cells,1 field emission displays,2 sensors,3 and varistors.4 Recently, cathodic electrodeposition of ZnO has been developed for the fabrication of nanostructured films because it is a simple and effective method compared with other chemical and physical methods such as chemical bath deposition and vapor phase growth. The method is based on the control of the production of hydroxide at the electrode by reduction of precursor species such as oxygen5 or zinc nitrate:6 NO3 - þ H2 O þ 2e - f NO2 - þ 2OH -
ð1Þ
1=2O2 þ H2 O þ 2e - f 2OH -
ð2Þ
This leads to the precipitation of ZnO on the electrode according to the reaction: Zn2þ þ 2OH - f ZnO þ H2 O
ð3Þ
The morphology of the film can be controlled by varying the deposition parameters: temperature,7 applied potential,6 deposition time,8 chemical composition of the deposition bath,9 and substrate.10 The Zn(II) concentration is a key parameter in controlling the morphology and properties of the films; for example, at low concentrations nanorods have been synthesized with interesting lasing properties,11 whereas at high concentrations nanoplates that exhibit excellent field emission properties have been obtained.12 It has been reported that films deposited at both low and high concentrations are of good crystallographic quality, forming nanorods or nanoplates that are single crystals in both cases.9,13 The Lincot and Pauporte groups have performed pioneering work to understand how the concentration influences the kinetics of deposition.14-16 However, the change in morphology due to changes in concentration is not completely understood. Elias and co-workers13 have carried out a systematic study of the influence of Zn(II) at low concentrations. They explained the
formation of one-dimensional (1-D) nanostructures by a limited diffusion of Zn ions which preferentially react at the top of the rods and thus lead to a growth along the c-axis. This explanation has been refuted by Belghiti and co-workers,16 who argue that such a limitation should lead to the formation of conical structures. They explained the formation of wirelike structures at low zinc concentration by a sharp increase of the pH at the beginning of the deposition leading to a change of the growing unit and a preferential growth on the top metastable face of the rods, whereas the nonpolar side faces are not reactive enough for further growth. This model explains well the change in aspect ratio observed when the zinc precursor is changed but predicts the formation of conical ended rods at high concentration and flat ended rods at low concentration, whereas the opposite is actually observed experimentally. Moreover, no study has been carried out yet to explain the change of morphology observed in a wide concentration range. To understand the change in morphology, it is necessary to take into account both thermodynamic equilibrium and growth kinetics. We have recently developed methodologies based on synchrotron-based X-ray absorption spectroscopy and diffraction to directly measure the growth and crystallography of ZnO nanostructures in solution.8,17 Together with electron microscopy characterization, these provide excellent tools to understand the influence of concentration on the nanostructures. We propose that the supersaturation of the solution is the key element in controlling the morphology. Experimental Section
*To whom correspondence should be addressed. E-mail: m.p.ryan@ imperial.ac.uk.
The electrolyte was composed of Zn(NO3)2 (Riedel DeHa€en, 98%) at a concentration ranging from 0.5 to 50 mM and 0.1 M of KCl (BDH, 99%) as a support electrolyte. The applied potential was -0.75 V vs Ag/AgCl (note that nitrate ions are not reduced at this potential and so do not contribute to the cathodic reaction). The temperature of the solution during the deposition was 65 ( 2 °C and oxygen gas was bubbled 20 min prior to, and during, the deposition. Electrodeposition experiments were carried out in a standard 3 electrode setup. A 20 nm thick sputtered gold film on Mylar was used as the working electrode, which doubles as the X-ray window of the cell; the films were used as-sputtered: they are very smooth and
r 2010 American Chemical Society
Published on Web 02/02/2010
pubs.acs.org/crystal
1190
Crystal Growth & Design, Vol. 10, No. 3, 2010
Illy et al.
Figure 1. Effect of the Zn(II) concentration on the morphology of the films. FESEM images of films deposited with an initial concentration of (a) 0.5 mM, (b) 2 mM, (c) 5 mM, (d) 10 mM, (e) 20 mM, and (f) 50 mM. The top inset in panels (b-d) presents the FESEM cross section of the films. The deposition time was 30 min. (g) Film deposited with 1 mM for 60 min; the inset presents a high magnification of the top surface of the rods. (h) Films deposited with 5 mM for 30 s. (i) Film deposited with 50 mM for 30 s. exhibit a (111) texture (not shown). An Ag/AgCl reference electrode (þ0.205 V vs NHE) and a Pt wire were used as the reference electrode and counter electrode, respectively. The electrochemistry was controlled with a Radiometer Voltalab PGZ402 potentiostat. In situ X-ray absorption spectroscopy experiments were conducted on beamline X10C at the National Synchrotron Light Source, Brookhaven National Laboratory, New York. The experimental setup is described in detail in ref 18. In situ Synchrotron X-ray diffraction experiments were carried out using beamline 11-3 at the Stanford Synchrotron Radiation Laboratory. The X-ray wavelength was 0.9736 A˚ and beam size was 0.15 mm (h) 0.05 mm (v). The samples were measured in a grazing incidence geometry. The principle of the method can be found in ref 17. A Leo Gemini 1525 field-emission scanning electron microscope (FESEM) equipped with an Oxford instruments energy dispersive X-ray spectroscopy (EDS) for microanalysis was used for surface topography and quantitative compositional analysis.
Results Microscopy. Figure 1 presents the FESEM images of films deposited with different Zn(NO3)2 concentrations. For low concentrations (less than 2 mM), the film is composed of isolated rods with a hexagonal shape (Figure 1a,b). Some rods exhibit triangular shapes because of twinning defects. For intermediate concentrations (between 5 and 20 mM), the film is a dense layer (Figure 1c-e) formed of clearly coalesced rods. For high concentrations (greater than 20 mM), the film is composed of large sheets (Figure 1f). The sheets are a few micrometers long and only a few nanometers thick. The
concentration regions less than 2 mM, between 2 and 20 mM, and above 20 mM will be called “low”, “medium”, and “high” concentrations respectively throughout this paper. We will first present the differences observed for the different concentration regions in our experiments before discussing individually the mechanism of growth in each region. Figure 1g presents FESEM images of samples prepared with a low zinc nitrate concentration after deposition times of 60 min. ZnO nanorods are orientated perpendicular to the substrate, and no coalescence is visible. Pyramid-like structures are clearly visible on the top of the rods (see inset Figure 1g). If the concentration is increased to 5 mM, the top face of the rod is flat and the substrate is densely covered after a few minutes, even though the films are still composed of rods (see cross section in Figure 1c,d). Figure 1h shows a FESEM image of the film after 15 s of deposition. The nuclei have a rough conical shape. Figure 2 shows the time dependence of the rod diameter for 1 and 5 mM Zn(NO3)2, which exhibits a dramatic change in the growth behavior in going from low to medium concentrations. The size was calculated from measurements of at least 200 features on three separate samples for each set of conditions. An increase of the diameter with time is clearly visible in the medium concentration region, whereas the diameter is constant for low concentrations. In the medium concentration region, if the concentration is further increased above 5 mM, the growth direction is progressively shifted away from the direction
Article
Crystal Growth & Design, Vol. 10, No. 3, 2010
1191
Figure 2. Evolution of the rod diameter with time as a function of Zn(II) concentration: (a) 1 mM (triangle) and (b) 5 mM (square) of Zn(NO3)2 The rod diameters have been measured from the FESEM images on three samples deposited under identical conditions. Figure 4. In-situ grazing incidence X-ray diffraction scans at the end of a 30 min deposition at (a) low Zn(II) concentration (1 mM), (b) medium concentration (5 mM), and (c) high concentration (50 mM). The peaks corresponding to ZnO, Au, and Zn5(OH)8Cl2 have been marked the squares, triangles, and circles, respectively.
Figure 3. Distribution of the nanorod diameters as a function of the Zn(II) concentration. (a) 0.5 mM, (b) 1 mM, (c) 2 mM, (d) 5 mM, and (e) 10 mM. The rod diameters have been measured from the FESEM images on two samples deposited in the same conditions. Gaussian fits of the distributions have been indicated by lines. The deposition time is 30 min.
perpendicular to the substrate (Figure 1e,f). Figure 3 presents the size distribution of the diameter of the nanorods after 30 min of deposition. For low concentrations, the distribution is very narrow. When the concentration is increased, the mean rod diameter increases and the dispersion becomes broader. Thus, the rod diameter distribution can be controlled simply by varying the concentration independently of other deposition parameters. EDS was carried out on the films: the film deposited from an electrolyte containing 5 mM or less of Zn(NO3)2 are entirely composed of Zn and O; for higher concentrations some Cl- was detected in the film. X-ray Diffraction. X-ray diffraction was performed in situ during the deposition.17 Within the uncertainties of the method, the only change observed with time was an increase in the amount of deposited material. Figure 4 presents X-ray diffraction (XRD) scans performed in situ at the end of the depositions with the solution removed. For 1 mM and 5 mM, only peaks from ZnO (having a wurtzite structure) and the substrate are observed. For a higher concentration of Zn(NO3)2, two extra peaks appear which can be identified as belonging to Zn5(OH)8Cl2.14,19 It has been shown that this phase can be removed by annealing at 350 °C for 1 h and that
the resulting films produce intense green luminescence at room temperature under femtosecond pulse ultraviolet light excitation.19 The formation of this zinc hydroxylchlorite phase is dependent on the relative solubility of zinc oxide and zinc hydroxylchlorite species.14 This is in good agreement with the EDS measurements showing that the films were composed of the elements Zn, O, and Cl. At more negative potentials, the intensity of the ZnO peaks is smaller than Zn5(OH)8Cl2 and additional peaks corresponding to other zinc hydroxylchlorite phases appear (XRD not shown). It should be noted that increasing the deposition potential (i.e., to less negative values) is a simple way to produce purer ZnO materials and reduce the need for annealing.20 Electrochemical Measurements. Figure 5a shows the variation with time of the absolute value of the current density as a function of the concentration. The current comes from the reduction of the nitrate and oxygen ions in solution (eqs 1 and 2). The three regions are clearly visible in Figure 5a. The curves obtained for low concentrations are very noisy and remain quite constant with time. When the zinc nitrate concentration is medium to high, a peak, followed by a plateau, appears. This peak has been previously linked to the nucleation process.21 During the nucleation, the true active area increases due to the formation of nuclei and the deposition current increases progressively. When the nanostructures coalesce, the active surface decreases. The peak in the current transient curve should be related to the maximum of the active surface area before coalescence. The peak gets higher and narrower when the concentration is further increased. The increase in the height of the peak in the medium and high concentration is related to the change in morphology. Figure 1i presents FESEM images of the films after 30 s of deposition at high concentrations, showing the formation of nanosheets. The nanosheets have a very high aspect ratio, and thus the active surface for the reduction reaction is higher than the rods and therefore the peak value is also higher. The fact that the peak is narrower indicates that the nucleation rate increases with an increase of the Zinc nitrate concentration. No peak is visible at low
1192
Crystal Growth & Design, Vol. 10, No. 3, 2010
Illy et al.
Discussion There are two fundamental models of crystal growth mechanism when random nuclei are generated on existing flat surfaces: the Kossel-Stranski-Volmer (KSV) mechanism22,23 (two-dimensional (2D) nucleation growth) and the BesselCabrera-Frank (BCF) mechanism24 (spiral growth). In the KSV mechanism, 2-D nucleation occurs preferentially near the edges, kinks, and steps of the substrate. When all the high index facets have disappeared, the crystal will continue to grow by 2D nucleation of new material depositing on the low index facets.24 In the BCF mechanism, crystal faces grow via the outward displacement of a growth spiral originating from screw dislocations in the central region of the face.25 The former occurs only at relatively high supersaturation levels, whereas the latter can happen at low supersaturation levels. The solubility product Ks of ZnO can be written as K s ¼ ½Zn2þ ½OH - 2
Figure 5. Electrochemical measurements as a function of Zn(II) concentration: (a) Variations of the current density with time. (b) Variations of the nominal thickness (tn) with time. (c) Variations of the current efficiency with time.
concentrations simply because there is no coalescence of the nanostructures. Figure 5b presents the variation of the volume of Zn(II) per unit area of surface (“nominal thickness”, tn) with time obtained from the X-ray absorption measurements.8 It is a direct measure of the rate of the ZnO deposition reaction (eq 3). For low concentrations the growth is quasilinear. For medium concentrations, the growth curves are composed of two linear parts, as reported earlier.8 When the concentration is increased, the change in slope of the curves is less distinct. Figure 5c presents the variation of the current efficiency of the reaction with time. It is the ratio of the actual amount of ZnO deposited to the amount of material predicted by Faraday’s law for the charge passed. The differences between the three cases can be clearly seen. For low concentrations, the current efficiency is very low, less than 10%. A sharp increase is observed during the first few minutes of deposition. This indicates that the reaction is more efficient on ZnO than on gold. For medium concentrations, the curves have a similar shape to the low concentration curves, but the current efficiency is about 40%. When the concentration is high, the shape of the curve is completely modified. No peak is observed, and the current efficiency increases sharply before reaching a plateau at 70%.
ð4Þ
The saturation of the solution is strongly dependent on the initial concentration of zinc nitrate as it is a precursor of both zinc and, for lower applied potentials, hydroxide ions. Low Supersaturation ([Zn(II)] < 2 mM). The supersaturation of the solution is low and the growth follows a BCF mechanism. The pyramidal-like structures at the top of the rods are the steps of screw dislocations. The diameter of the rods is set during the nucleation and stays constant during the growth because the growth proceeds from the outside inward. Similar types of growth have been reported for chemical bath deposition for low level of supersaturation.25 No change in the morphology and growth kinetics have been observed by varying the applied potential, which promote the formation of hydroxide. The growth is probably limited by the diffusion of Zn(II). The current efficiency is therefore low because the production of hydroxide is faster than the Zn(II) diffusion. High Supersaturation ([Zn(II)] > 20 mM). In this domain, the supersaturation of the solution is very high. 2-D nucleation of low index faces is more favorable and the growth follows a KSV mechanism. Large plates are formed. The growth rate is dramatically increased (Figure 5b). The comparison of SEM images (Figure 1f,i) shows that the number and diameter of the nanosheets increase dramatically with time, whereas the thickness does not change significantly. The growth perpendicularly to the sheets is blocked by absorption of Cl-. This happens preferentially on the polar faces and blocks the growth in that direction as previously reported.9 The fact that the sheets grow perpendicularly to the substrate may be explained by Coulomb repulsion between the Cl- terminated (002) face and the negatively polarized substrate.19 The continuous nucleation of new plates happen on the (002) faces previously formed and large stacks of plates are observed. The formation of Zn5(OH)8Cl2 is observed because for high levels of supersaturation its solubility is close to that of ZnO.14 The current efficiency first increases because the reaction of hydroxide formation is more efficient on ZnO than gold.26 It is lower than 100% due to the formation of other phases and the diffusion of hydroxide toward the bulk solution. Medium supersaturation (2 mM < [Zn(II)] < 20 mM). In this intermediate region, the concentration of the oxygen and zinc precursor are close to stoichiometry and a transition between the two mechanisms above is observed. During the nucleation, the pH at the vicinity of the electrode is low and
Article
Crystal Growth & Design, Vol. 10, No. 3, 2010
so is the supersaturation. Conic structures are formed following a BCF mechanism. The increase of the pH at the electrode with time leads to an increase of the supersaturation and a transition to a KSV mechanism. The low index faces disappear by the addition of material on the side of the rods and the formation of flat (00l) surfaces with a hexagonal symmetry is observed. An increase of the current efficiency is observed compared to lower concentration because of the increase of the concentration of zinc nitrate in solution and thus a decrease of the diffusion limitation. Conclusion Three concentration regions have been identified where the deposition behavior of ZnO nanorods differs notably. For low Zn(II) concentrations (less than 2 mM), the nanorods grow along screw dislocations: pyramidal structures are observed on top of the rods, and over time the rods maintain a constant diameter and do not coalesce. For medium Zn(II) concentrations (2 mM to 20 mM), the nanorods grow in a 3-D fashion and ultimately coalesce. The growth rate behavior clearly shows an initial nucleation process followed by 3-D growth. For high Zn(II) concentrations (above 20 mM), large nanosheets are formed and Cl- ions absorb to the surface to form zinc hydroxylchloride phases. By controlling the supersaturation in solution, highly tailored morphologies can be produced. Acknowledgment. The authors wish to acknowledge S. C. Hendy (Industrial Research Limited, New Zealand) for fruitful discussions. XAS data reported in this paper were collected at beamline X10C at the National Synchrotron Light Source. Use of the National Synchrotron Light Source, Brookhaven National Laboratory, was supported by the U.S. Department of Energy, Office of Basic Energy Sciences. XRD data were collected at beam line 11-3 at the Stanford Synchrotron Radiation Lightsource, a national user facility operated by Stanford University on behalf of the U.S. Department of Energy, Office of Basic Energy Sciences. The authors wish to thank Hugh Isaacs and Kenneth Sutter (Materials Science at BNL), and Larry Fareria and Mike Sansone (NSLS at BNL) for their technical support. Funding was provided in part by the New Zealand Foundation for Research, Science and Technology under contract CO8X0409.
References (1) Takahashi, K.; Nishi, T.; Suzaka, S.; Sigeyama, Y.; Yamaguchi, T.; Nakamura, J.; Murata, K. Performance improvement by inserting an electrodeposited ZnO into ITO/organic solid interface in organic solid-state solar cells. Chem. Lett. 2005, 34 (6), 768–769. (2) Wang, X. Influence of S incorporation on the luminescence property of ZnO nanowires by electrochemical deposition. Phys Lett. A 2008, 372 (16), 2900–2903. (3) Kenanakis, G.; Vernardou, D.; Koudoumas, E.; Kiriakidis, G.; Katsarakis, N. Ozone sensing properties of ZnO nanostructures grown by the aqueous chemical growth technique. Sensor Actuator B Chem. 2007, 124 (1), 187–191. (4) Pillai, S. C.; Kelly, J. M.; McCormack, D. E.; O’Brien, P.; Ramesh, R. The effect of processing conditions on varistors prepared from nanocrystalline ZnO. J. Mater. Chem. 2003, 13 (10), 2586–2590.
1193
(5) Peulon, S.; Lincot, D. Cathodic electrodeposition from aqueous solution of dense or open-structured zinc oxide films. Adv. Mater. 1996, 8 (2), 166–170. (6) Izaki, M.; Omi, T. Transparent zinc oxide films prepared by electrochemical reaction. Appl. Phys. Lett. 1996, 68 (17), 2439–2440. (7) Goux, A.; Pauporte, T.; Chivot, J.; Lincot, D. Temperature effects on ZnO electrodeposition. Electrochim. Acta 2005, 50 (11), 2239– 2248. (8) Ingham, B.; Illy, B. N.; Ryan, M. P. In situ synchrotron studies of ZnO nanostructures during electrochemical deposition. Curr. Appl. Phys. 2008, 8 (3-4), 455–458. (9) Xu, L.; Guo, Y.; Liao, Q.; Zhang, J.; Xu, D. Morphological control of ZnO nanostructures by electrodeposition. J. Phys. Chem. B 2005, 109 (28), 13519–13522. (10) Pauporte, T.; Cortes, R.; Froment, M.; Beaumont, B.; Lincot, D. Electrocrystallization of epitaxial zinc oxide onto gallium nitride. Chem. Mater. 2002, 14 (11), 4702–4708. (11) Pauporte, T.; Lincot, D.; Viana, B.; Pelle, F. Toward laser emission of epitaxial nanorod arrays of ZnO grown by electrodeposition. Appl. Phys. Lett. 2006, 89 (23), 233112. (12) Pradhan, D.; Kumar, M.; Ando, Y.; Leung, K. T., Efficient field emission from vertically grown planar ZnO nanowalls on an ITOglass substrate. Nanotechnology 2008, 19, (3). (13) Elias, J.; Tena-Zaera, R.; Levy-Clement, C. Electrochemical deposition of ZnO nanowire arrays with tailored dimensions. J. Electroanal. Chem. 2008, 621 (2), 171–177. (14) Peulon, S.; Lincot, D. Mechanistic study of cathodic electrodeposition of zinc oxide and zinc hydroxychloride films from oxygenated aqueous zinc chloride solutions. J. Electrochem. Soc. 1998, 145 (3), 864–874. (15) Pauporte, T.; Lincot, D. Hydrogen peroxide oxygen precursor for zinc oxide electrodeposition II--Mechanistic aspects. J. Electroanal. Chem. 2001, 517 (1-2), 54–62. (16) El Belghiti, H.; P., T.; Lincot, D. Mechanistic study of ZnO nanorod array electrodeposition. Phys. Status Solidi 2008, 205 (10), 2360–2364. (17) Ingham, B.; Illy, B. N.; Toney, M. F.; Howdyshell, M. L.; Ryan, M. P. In situ Synchrotron X-ray diffraction experiments on electrochemically deposited ZnO nanostructures. J. Phys. Chem. C 2008, 112 (38), 14863–14866. (18) Kerkar, M.; Robinson, J.; Forty, A. J. In In Situ Structural Studies of the Passive Film on Iron and Iron/Chromium Alloys Using X-Ray Absorption Spectroscopy; General Discussion on Structure of Surfaces and Interfaces as Studied Using Synchrotron Radiation, Manchester, England, Apr 04-06, 1990; Royal Society Chemistry: Manchester, England, 1990; pp 31-40. (19) Wang, F.; Liu, R.; Pan, A.; Cao, L.; Cheng, K.; Xue, B.; Wang, G.; Meng, Q.; Li, J.; Li, Q.; Wang, Y.; Wang, T.; Zou, B. The optical properties of ZnO sheets electrodeposited on ITO glass. Mater. Lett. 2007, 61 (10), 2000–2003. (20) Illy, B. Electrodeposition of zinc oxide nanostructured films. Ph.D. Thesis, Imperial College London, London, 2009. (21) Pletcher, D.; S. E. G., Southampton electrochemistry group, Instrumental Methods in Electrochemistry; Horwood Publishing: U.K., 2001; p 442. (22) Kossel, W., Zur Theorie des Kristallwachstums, Nachr.; Ges. Wiss. G€ottingen, 1927. (23) Stranski, Zur Theorie des Kristallwachstums, Nachr. Z. Phys. Chem 1928, A136, 259-278. (24) Burton, W. K.; Cabrera, N.; Frank, F. C. The growth of crystals and the equilibrium structure of their surfaces. Phil. Trans. R. Soc. Lond. Math. Phys. Sci. 1951, 243 (866), 299–358. (25) Govender, K.; Boyle, D. S.; Kenway, P. B.; O’Brien, P. Understanding the factors that govern the deposition and morphology of thin films of ZnO from aqueous solution. J. Mater. Chem. 2004, 14 (16), 2575–2591. (26) Goux, A.; Pauporte, T.; Lincot, D. Oxygen reduction reaction on electrodeposited zinc oxide electrodes in KCl solution at 70 °C. Electrochim. Acta 2006, 51 (15), 3168–3172.
