Enhanced Quality, Growth Kinetics, and Photocatalysis of ZnO

Publication Date (Web): August 5, 2015 ... They can be grown by low-cost solution methods on Al (covered) ... Citation data is made available by parti...
0 downloads 0 Views 4MB Size
Page 1 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

Enhanced Quality, Growth Kinetics and Photocatalysis of ZnO Nanowalls Prepared by Chemical Bath Deposition Kingsley O. Iwu1, Vincenzina Strano1, Alessandro Di Mauro1, Giuliana Impellizzeri1, Salvo Mirabella1* 1

MATIS IMM-CNR and Dipartimento di Fisica e Astronomia, Università di Catania, via S.

Sofia 64, 95123 Catania, ITALY

1 ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 26

ABSTRACT: ZnO nanowalls (NWLs) represent a non-toxic, abundant and porous material, with promising applications in sensing and photocatalysis. They can be grown by low-cost solution methods on Al (covered) substrates - Al(OH)4- generated in situ is assumed to be responsible for engendering the NWL morphology. Here, we grew ZnO NWLs by chemical bath deposition (at 70-95°C). The roles of pH, concentration of Al(OH)4-, and growth time on the thickness and quality of NWL film were experimentally investigated and the growth kinetics was explained in terms of a self-screening model. Increasing the chemical bath pH from 5.7 to 7.4 led to a 40% thicker film and more NWLs per unit area of the substrate - due to increased concentration of Al(OH)4- - but these were accompanied by the presence of embedded micro/nano particles. We propose the use of anodized Al as a way to enhance the growth rate and density of the NWLs with no detrimental effect on film quality. Compared with non anodized Al, NWL film grown on anodized Al (at the lower pH) showed a higher growth rate, an excellent film quality, and a higher photocatalytic activity in the degradation of toxic methyl orange.

Keywords: ZnO; nanowalls; film thickness and quality; anodic aluminium oxide; photocatalysis

2 ACS Paragon Plus Environment

Page 3 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

1. Introduction

Porous materials are important for applications in areas such as biomedicine, microelectronics and photonics. ZnO is an important, abundant and non-toxic wide bandgap material with application in light emitting diodes [1], photocatalysis [2], and UV- [3,4], near infrared [5], bio [6, 7], pH [8], and gas [9] sensors. Among the various ZnO nanostructures, the so-called “nanowalls” (NWLs) [10] have attracted increasing interest because of their huge surface-tovolume ratio and extremely thin wall thicknesses [2, 9, 11, 12]. ZnO NWLs or nanoplatelets (crystalline, 10-200 nm thick) grow vertically on substrates, with an intertwined, honeycomb-like pattern and c-axes parallel to the substrates. [12] ZnO NWLs can be prepared by physical [9, 13] or chemical [14] vapour deposition techniques, with operating temperatures higher than 800 °C. Alternatively, they can be synthesized at 90 – 100 °C in aqueous Zinc solution, during growth period of 30 min to several hours [2, 11, 12, 15]. Unlike the vacuum techniques, the low temperature approach is suitable for ZnO NWL growth on flexible (plastic) substrates, which can lead to much lower fabrication cost. This has recently been demonstrated in the fabrication of a fully flexible pH sensor with an extended gate thin film transistor configuration, which showed a near-ideal Nernstian response (59 mV/pH, with an ideality factor close to 1) [8]. Aluminium or aluminium covered substrate is important for obtaining ZnO NWLs in the low temperature chemical bath deposition (CBD) technique, with hexamethylenetetramine (HMTA) often employed. The latter generally fixes the pH of the solution at around 6 [16], yielding hydroxyl ions which react with Al to form Al(OH)4-. It is now well established that the binding of Al(OH)4- to the Zn2+ terminated surfaces is essential for blocking ZnO growth along the 3 ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 26

(0001) direction and promoting lateral growth [2, 11, 12]. With Al wafer and equimolar solution of Zinc nitrate hexahydrate (ZN) and HMTA, Ye et al. demonstrated that by increasing pH from 6 to 12 (with ammonium hydroxide, AH), thinner nanowalls could be obtained. They attributed this to an increasing concentration and adsorption of Al(OH)4- on the (0001) surfaces of ZnO NWLs [2]. NWLs of Zn-Al carbonate hydroxide hydrate, a layered compound, and ZnO nanorods were obtained at pH 10 with zinc acetate, AH, and e-beam evaporated Al film [17]. The concentration of Al(OH)4- influenced the particular product obtained – at high concentration of the Al complex, made possible by using a 1 µm Al film, only the Zn-Al carbonate hydroxide hydrate NWLs were obtained, while a 100 nm Al film gave the layered compound first and later, ZnO nanorods on top. The authors attributed the formation of ZnO nanorod to the exhaustion of Al(OH)4- needed to ensure the NWL morphology. It therefore follows that the growth rate of ZnO NWLs can be controlled by manipulating the rate of Al(OH)4- generation. The latter can be enhanced by increasing pH, but this may not be suitable for Al or some pH sensitive substrates, whose fast corrosion in a highly basic solution can affect the quality of NWLs grown on them. A fast (10 min) sonochemical route for ZnO NWL growth on Al and alumina substrates (with HMTA and ZN) was reported by Nayak et al.[18]. However, there is no information on whether the growth dynamics and film quality differ between the two substrates. Moreover, prolonged sonication caused the NWLs to peel off. Here we report on how increasing pH, with or without the presence of HMTA, can affect the growth rate and quality of ZnO NWL film. Prior anodization of Al, which is one way to increase its oxide layer [19], was used as a means of enhancing the rate of generation of Al(OH)4-. Its effects on not only growth rate, but also film quality, growth temperature and photocatalytic

4 ACS Paragon Plus Environment

Page 5 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

activity, are presented. NW film quality is defined in terms of homogeneity in film thickness across the substrate and the presence or absence of embedded ZnO (micro/nano) particles. 2. Experimental

Zn(NO3)2 6H2O ( ≥ 99%), HMTA ( ≥ 99.5%) and ZnO colloids (40 wt %, ethanol solution) were obtained from Sigma Aldrich. Deionised water (18.2 MΩ•cm) was used as solvent for ZnO NWL growth. CBD of ZnO NWLs was typically carried out with 25 mM each of Zinc nitrate and HMTA, with or without AH. Except otherwise indicated, the growth was performed at 95 °C for 300 s on Al or anodised Al film (100 nm thick, sputtered on silica over silicon substrates). To seed the substrate, 20 µl of 1 or 0.1 wt % ZnO colloid (diluted with isopropanol, particle size < 40 nm) was spin coated on 1 cm2 of substrate at 7000 rpm for 30 s, followed by heating at 120 °C for 20 min. The plain Al film were sonicated in soapy water, water, ethanol and acetone before use. Anodization of Al was carried out at 5 V (vs SCE, VersaSTAT 4 Potentiostat) in 0.4 M sulphuric acid for 2 min, yielding a porous oxide layer [19]. A Gemini 152 field emission SEM Carl Zeiss SUPRA 25 scanning electron microscope (FEG-SEM) was used to image the morphology (plan view) and thickness (cross view) of ZnO NWL films. In the case of nonuniformity in thickness, the thicker part of each sample was used for determining film thickness. Rutherford Backscattering Spectrometry (RBS, 2.0 MeV He+ beam atnormal incidence) with normal or glancing detection modes (165° or 105° backscattering angle, respectively) was performed with a 3.5 MV HVEE Singletron accelerator system. RBS spectra were simulated through SimNRA software [20].

5 ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 26

Photoluminescence (PL) analyses were performed by pumping with the 325 nm line of a HeCd laser. The pump power was 1.5 mW and the laser beam was chopped through an acousto-optic modulator at a frequency of 55 Hz. The PL signal was analyzed by a single grating monochromator and detected with a Hamamatsu visible photomultiplier. Spectra were recorded with a lock-in amplifier using the acousto-optic frequency as a reference. All the measurements were performed at room temperature. We tested the photocatalytic activity of ZnO NWLs by the degradation of methyl orange (MO), a toxic azo-dye commonly used in the textile industry. ZnO NWL samples were irradiated with a UV light source (8 W power; at 365 nm, 20 nm of full width at half maximum, irradiance of 1.1 mW cm-2) for 60 min in order to remove any organic compound on the sample surface [21]. Then the samples (1 cm2 each) were immersed in an aqueous MO solution (2 ml, starting MO concentration of 1.0x10-5 M). To disentangle the adsorption of MO on ZnO from its photoinduced degradation, the samples were left immersed in the solution under dark condition for 20 hours, after which the MO concentration remained constant (about 50% of the starting value). Thereafter, the samples were irradiated with the UV light and at fixed intervals, the absorbance of the solution was measured between 350 and 650 nm (Perkin-Elmer Lambda 35 UV/VIS spectrometer). The peak at 464 nm was used in evaluating the concentration of MO, based on the Lambert-Beer law [22, 23]. The decomposition of the MO solution in the absence of any catalyst was checked as a reference.

6 ACS Paragon Plus Environment

Page 7 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

3. Results and discussion

Table 1: Summary of different conditions for ZnO NWL syntheses and the resulting film thicknesses. sample ZNW1 ZNW2 ZNW3

Solution content ZN + HMTA ZN + AH ZN + HMTA + AH

Solution pH 5.7 7.4 7.4

thickness [nm] 850 600 1200

To investigate the effect of the growth solution composition, ZnO NWLs were prepared using HMTA (ZNW1), AH (ZNW2) or both (ZNW3), as summarized in Table 1. When only HMTA was used, measured pH was 5.7, while the addition of 16 mM AH – with or without HMTA – increased the pH to 7.4. Fig. 1 reports the plan views of the three samples. In comparison with other samples, ZNW1 presents the lowest density of ZnO NWLs, while both ZNW2 and ZNW3 have a more interlaced structure. This can be attributed to an increased nucleation of NWLs as a result of a higher concentration of Al(OH)4- at high pH. There is however a significant number of embedded particles in ZNW2 (Fig. 1b), as well as an inconsistency in the density of NWLs across a wide area of the substrate as presented in Fig. S1 in the supplementary information. Fig. 1b therefore represents the best NWL density for ZNW2. Further increase in AH concentration to 32 mM gave the same film thickness as ZNW2 but with even worse quality (not shown).

7 ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 26

Figure 1. Plan view SEM micrographs of ZnO NWL films: (a) ZNW1 (b) ZNW2 (c) ZNW3. Scale bar is the same for all the images.

Uniformity in the density of the NWLs improves significantly in ZNW3, though it still contains embedded particles (see also Fig S1). This presence of embedded particles is likely due to a fast

8 ACS Paragon Plus Environment

Page 9 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

precipitation of ZnO in solution at higher pH. As shown in Table 1, the thicknesses of NWL films are 850, 600, and 1200 nm for ZNW1, ZNW2 and ZNW3, respectively. This indicates that increasing pH – as a way of increasing the concentration of Al(OH)4- – does not necessarily lead to increase in film thickness on its own. The combination of HMTA and higher pH (ZNW3) gives a faster growth (Fig. 1c), implying that this combination is essential for both higher nucleation and growth rates. Fig. 2 reports the RBS analyses performed on ZNW1, ZNW2, and Al film. The thickness of ZNW3 hindered the acquisition of a reliable RBS spectrum. The peaks at 1.6, 1.1 and 0.75 MeV are related to He+ ions backscattered by Zn, Al and O atoms, respectively [24]. Al film shows a sharp peak at 1.1 MeV (O signal at 0.71 MeV comes from the silica below). ZNW1 and ZNW2 have a similar Zn peak at 1.6 MeV, while only ZNW1 shows a strong Al peak at 1.05 MeV (shifted to lower energy because of the NWL film on top). Only a very small signal at 1.05 MeV is observed for ZNW2. This indicates a fast corrosion of the Al film during ZNW2 growth at higher pH and may be responsible for the smaller thickness and poor quality of its ZnO film, as this fast corrosion might have delayed or impeded the heterogeneous NWL nucleation. The reason this deleterious effect was not seen in ZNW3 (prepared at the same pH) can be related to the presence of HMTA. One possible explanation is that HMTA helped in ensuring a high Zn2+ concentration by reducing the precipitation of ZnO in the chemical bath (for example, through chelating Zn2+ [25]). This then would enhance NWL nucleation and growth by making more Zn2+ available. It is also possible that steric hindrance caused by interaction of HMTA with alumina (Al3+) inhibited its fast dissolution.

9 ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 26

Figure 2. RBS spectra of ZNW1, ZNW2 and Al film. The inset shows the experimental setup and the sample structure (porous ZnO film on Al/silica/silicon).

The area of Zn peak (AZn) is related to the amount of Zn in the NWL film, which is 0.82x1017 and 1.05x1017 Zn at./cm2 in ZNW1 and ZNW2, respectively. In both cases, the O signal is in agreement with a Zn:O of 1:1. Based on these data and growth time (300 s), one can estimate the deposition rate of ZnO molecules in this NWL structures, which is about 3x1014 ZnO molecules/(cm2×s). It is worthy of note that ZnO nanorods, grown with the same synthetic condition as ZNW1, show a vertical growth rate of 12 nm/min [16], which translates to around 4x1014 ZnO molecules/(cm2×s). This demonstrates that the precipitation kinetics of ZnO molecules is not greatly affected by the presence of Al(OH)4-, whose main function is related to the shape of the solid ZnO. ZNW2 contains a larger amount of ZnO compared to ZNW1, evidencing a more dense structure. To evaluate the filling factor (FF, fraction of volume occupied by ZnO) in these two samples, an equivalent thickness (teq) can be derived from the 10 ACS Paragon Plus Environment

Page 11 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

RBS spectra, as

, where ρZn (2.05x1022 Zn at./cm3) is the density of Zn atoms in ZnO.

FF is the ratio of teq to the measured thickness. We obtained teq and FF of 40 nm and 4.7% for ZNW1, and 51 nm and 8.5% for ZNW2. It is therefore clear that ZnO NWL film is a highly porous material, with pore size in the range of 50-200 nm. The impact of Al(OH)4- on ZnO NWL film thickness and quality was further investigated by carrying synthesis on substrates seeded with ZnO colloids. ZNW3_0.1% and ZNW3_1% were grown using the synthetic conditions of ZNW3 on Al film, spin coated with 0.1 and 1 wt% ZnO colloidal solutions, respectively. ZNW3_0.1% yielded a film thickness (Fig. 3c) and porosity (Fig. 3a) similar to ZNW3 - its seed layer was too thin to be viewed by SEM in cross section but was clearly seen in plan view (not shown). With a 70 nm ZnO seed layer, ZNW3_1% shows a reduced ZnO NWL film thickness (Fig. 3d) of 600 nm, with a lower porosity (Fig. 3b) than even ZNW1. This implies that with the thicker seed layer there was a significant reduction in the concentration of Al(OH)4- , due to obstruction to penetration of the synthetic solution to the Al substrate and/or movement of Al(OH)4- to the surface of the seed layer. In any case, the Al films were not completely corroded in these samples.

11 ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 26

Figure 3. SEM micrographs of plan (a) and cross-section (c) of ZNW3_0.1% and plan (b) and cross section (d) of ZNW3_1% . The scale bar is the same for each kind.

Experimental data presented up to now support the view that: i) the Al(OH)4- complex is essential for the growth of NWLs; ii) the production rate of this complex largely affects the thickness and porosity of

NWL film; iii) a thicker film can be obtained with a higher

concentration of Al(OH)4-, which is itself achieved by increasing pH. However, the latter is detrimental to film quality. We therefore propose here a preliminary, controlled oxidation of Al film through anodization, as a way of enhancing the rate of generation of the complex at lower pH. We assume that dissolution of anodic aluminium oxide should be faster than that of pure Al. A comparison of the Al film before and after this anodization process is presented in Fig. 4. The

12 ACS Paragon Plus Environment

Page 13 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

RBS spectra (symbols) were obtained in glancing detection mode to enhance the depth resolution at the surface. In this configuration, Al and O peaks are at 1.35 and 1.05 MeV. Spectral simulation (lines) were performed with SimNRA software. According to the sample structure reported in the table in Fig. 4(b), Al film contains a very thin native Al oxide layer at the surface, while the oxide layer on anodized Al is thicker and fairly stoichiometric. SEM micrographs (insets in Fig. 4) reveal that the surface of pure Al is somehow rough, comprising big and small particles, while that of the anodized film is smoother, exhibiting the usual porous anodic aluminium oxide structure (pore size ~ 10 nm).

Figure 4. (a) RBS spectra (symbols) of Al and anodized Al films. The insets show SEM micrographs in plan view of Al film (top) and anodized Al film (bottom). The scale bar is the same for all the images. (b) Table reporting the sample structures derived from the simulation (lines) of RBS spectra. 13 ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 26

ZnO NWLs (ZNW1_anod) was grown on anodised aluminium under the same conditions as ZNW1. It can be seen from Fig. 5a that ZNW1_anod film is quite uniformly thick across a wide area of the substrate, unlike ZNW1 which shows a variation in film thickness (Fig. 5b). This lack of uniformity is also seen in ZNW3 (see also Fig. S1), leading us to conclude that it does not depend on the concentration of the aluminium complex. It can be related to the surface roughness of the non anodised Al film, indicating a morphology-dependent varying rate of Al(OH)4generation across the substrate surface. Anodization therefore helps to smoothen the surface of the substrate, ensuring uniform rate of the complex generation. It should be noted that the dark spots in Fig. 5b are not necessarily void of NWLs, though they may be less populated. The insets of Fig. 5 indicate that the thickness of ZNW1_anod is 1300 nm thick, compared to 850 nm for ZNW1, representing an increase of 53 %. This is similar to the difference between ZNW1 and ZNW3, implying that the concentration of Al(OH)4- can be increased at a lower pH value when Al is replaced with anodic aluminium oxide. Therefore, ZNW1_anod possesses a larger surface area than ZNW1.

14 ACS Paragon Plus Environment

Page 15 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

Figure 5. Low magnification SEM plan image of (a) ZNW1_anod, and (b) ZNW1. The insets are their cross sectional views. Scale bar in (a) is the same for (b).

Photoluminescence (PL) spectra were obtained for ZNW1, ZNW1_anod and ZNW3 (Fig. S2), showing in all cases a peak centred at 560 nm, attributed to radiative recombination from band gap defect states. This indicates that increasing the concentration of Al(OH)4- in order to impact the growth rate and quality of ZnO NWL film does not affect the energy level of the surface defects. As presented in Fig. S1, there is little or no evidence of embedded particles in ZNW1_anod (giant particles from solution can be seen from time to time on the surface of all samples though). Anodization therefore helps in not only realising a uniformly thick ZnO NWL film, but

15 ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 26

is also useful in increasing the film thickness (or decreasing growth time) without recourse to higher (basic) pH which could compromise film quality and substrate integrity. Various ZNW1_anod samples were grown for different duration in order to gain insight into their growth kinetics. It can be seen from Fig. 6a-e that after 10 s of growth the substrate is only sparsely covered by NWLs, with the coverage increasing after 30s. At 90s the substrate becomes uniformly and densely covered, which was also evident by visual observation. There appears to be little or no change in porosity from 90 to 300 s. The sample grown for 900 s appears smoother than the ones grown for 90 s and 300 s, with average thickness of a single NWL foil around 20 nm (Fig. S3). As shown in Fig. 6(f), there is an explosive growth within 90s, with the film thickness reaching 800 nm, while the growth rate decreases at higher duration. The presence of ZnO crystal structure was confirmed with high resolution X-ray diffraction analyses in an earlier paper where NWLs were grown on non anodized Al, for 60 min at 90 °C [8]. In that case XRD pattern revealed a peculiar splitting of the (002) peak into two components (corresponding to d-spacing of 0.261 nm and 0.265 nm) indicating the occurrence of a non uniform strain applying to the zincite planes [8]. Similar XRD pattern was found for ZNW1 and ZNW1_anod grown at 300 and 900 s, with the (002) peak split into two components at lower diffraction angles (Fig. S4). In addition, the films grown for a longer time exhibits more intense peaks, probably due to the thicker film or to progressive crystallization. The peaks for the samples grown for 300 s are generally weak, so XRD analysis on samples grown for lesser duration was considered unnecessary.

16 ACS Paragon Plus Environment

Page 17 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

Figure 6. SEM plan views of ZNW1_anod grown for (a) 10 s, (b) 30 s, (c) 90 s, (d) 300 s, and (e) 900 s; the scale bar is the same for all. (f) NWL film thickness vs. growth time for ZNW1_anod.

The growth kinetics of ZnO NWLs indicates a process highly affected by the concentration of Al(OH)4-. Taking into account also the results of the seeded substrate (Fig. 3), we can assume that initially, a maximum flux of the Al(OH)4- complex is attained just over the surface of the substrate, ensuring a growth rate as high as 15-20 nm/s at the early stage. With increasing NWL film thickness, the flux of Al(OH)4- coming from the substrate decreases, because the growing NWLs screen the Al film from the solution and, in addition, can obstruct the diffusion of Al complex to the top of the film. The diffusion of the Al complex can be very fast compared to its generation, thus we assume that the flux of the Al(OH)4- complex is essentially limited by its 17 ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 26

generation. The following two hypotheses are the basis for a self-screening model developed to explain the growth kinetics. We assume that: i) the flux (φ) of Al(OH)4- generated at the surface of the substrate decreases exponentially with film thickness (s); ii) the growth rate of the film is proportional to φ. These bring to fore the following equations:

(1)

(2) where a, b and φ0 (flux of Al(OH)4- at beginning) are the only parameters. The analytical solution for s(t), given that at t=0 s=0, is s (t ) = α × ln(1 + βt ) .

where α =

(3)

1 , and β = abφ0 . Fig. 7 reports the time evolution of the NWL thickness for a

ZNW1_anod (half-close circles) and ZNW1 (open circles) grown at 95°C, and for ZNW1 (open squares) grown at 70 °C. The error bars are as large as the sizes of the symbols.

18 ACS Paragon Plus Environment

Page 19 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

Figure 7. NWL film thickness versus growth time for ZNW1_anod (half filled symbols) and ZNW1 (open symbols) grown at 95 °C (circles) and 70 °C (squares). The lines represent the model fits to the data based on the reported equation and parameters (α and β).

It is obvious that NWL films grown on anodized Al show the faster growth rate. The best fit procedure applied on the ZNW1_anod at 95°C gives the following: α = 0.53 µm and β = 0.043 s1

. The first parameter accounts for the self-screening effect while the second one contains the

initial flux of Al complex (φ0) and the product a×b. Assuming that the self-screening parameter (a) is not affected by the Al film anodization, ZNW1 grown at 95°C was fitted by fixing α = 0.53 µm and obtaining β = 0.017 s-1. This indicates that anodization increased the initial flux of Al

complex, as expected. To fit the data of ZNW1 at 70°C we fixed α = 0.53 µm, and we obtained a best fit with β = 0.0024 s-1. This is expected given the lower synthesis temperature. 19 ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 26

Finally, the film thickness of ZNW1_anod at 70 °C (300 s growth time, half-closed square) is as high as that of ZNW1 grown at 95 °C for the same duration (Fig. 7). In addition, the former possesses a good quality – uniformity in NWL thickness and absence of embedded particle – as shown in Fig. S5. The photocatalytic activity of ZnO NWLs grown at 95 °C, with or without prior anodization, was tested by the degradation of methyl orange solution. Given the sample area and the RBS estimation of AZn, the mass of ZnO was about 17 µg for ZNW1. Figure 8(a) shows the time evolution of the residual concentration (C) of MO clearly indicating a photocatalytic activity of the synthesized ZnO NWLs [C was normalized to the initial concentration (C0), which was obtained after adsorption test in the dark]. No response was obtained for two reference samples (Al and anodized Al films). ZNW1_anod (black and white circles) shows a higher photoactivity than ZNW1 (white circles). According to the Langmuir-Hinshelwood model [26], the photodegradation reaction rate (k) of water contaminants can be estimated by the following equation:

where t is the irradiation time. Therefore, it is possible, by a simple fitting procedure, to extract the MO photo-degradation . The rate for ZNW1_anod and ZNW1 are 3.2×10-6 and 2.3×10-6 s-1, respectively. If normalized to the mass of ZnO, the photocatalytic activity of ZNW1 becomes 0.13 (s-1 g-1), which is in the same order of magnitude as Ref. [2] (k of 1.5×10-3 s-1 for 2 mg ZnO). In fact, we used a less performing configuration (ZnO film on a substrate rather than ZnO powder).

20 ACS Paragon Plus Environment

Page 21 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

Figure 8. Apparent photodegradation test of Methyl Orange (MO). (a) Normalized MO concentration during irradiation time for ZNW1_anod (half filled symbols) or ZNW1 (open symbols), and for reference samples (Al, anodized Al, and MO solution). (b) Photo-degradation rates for the discoloration process.

We show in Fig. 8 (b) the photo-degradation rate for the discoloration process, normalized to the value obtained for MO decomposition without any catalyst (“MO” case in the abscissa axis). ZNW1 and ZNW1_anod show a clear photo-degradation activity, with the latter being 40% higher than the former. This difference can be explained by the larger surface area of

21 ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 26

ZNW1_anod. As observed by SEM plan images (not shown), the discoloration test did not change the NWLs structure.

4. Conclusions

In conclusion, we studied and modelled the growth of ZnO nanowalls grown by chemical bath deposition (70-95°C) on Al film. The growth rate of the film was enhanced by increasing pH from 5.7 to 7.4, which translated to increasing the concentration of Al(OH)4-. However this was accompanied by a poor film quality. The enhancement of growth rate and a good film quality could however be achieved at the lower pH by employing a prior anodization of the Al film, Furthermore, with anodised Al, growth temperature can be decreased to 70 °C with no negative effect on NWL film quality, and with a film thickness similar to that obtained at 95 °C with non anodised Al. We developed a self-screening model to describe the dependence of ZnO NWL film thicknesses on the flux of Al(OH)4-. ZnO NWLs grown on anodised Al showed a 40 % higher photocatalytic activity than corresponding NWLs grown on pure Al, attributed to a larger surface area of the former.

Author Information *Corresponding Author Salvo Mirabella, [email protected]

22 ACS Paragon Plus Environment

Page 23 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

Author Contributions KOI conceived the study, performed the samples synthesis, contributed to data interpretation, model, and manuscript drafting. VS contributed to sample synthesis and SEM analysis. ADM and GI performed the MO degradation tests. SM conceived the study, contributed to SEM and RBS analyses, model development, and manuscript revision. All authors have given approval to the final version of the manuscript. Funding Sources This work has been partially sponsored by the projects PLAST_ICs (PON02_00355_3416798 PON 2007-2013) and FP7 European Project WATER (Grant Agreement 316082).

NOTES The Authors declare no competing financial interest.

Acknowledgement The authors thank G. Franzò (MATIS CNR-IMM) for PL analyses, G. Fortunato and L. Maiolo (CNR-IMM) for scientific discussion and supply of substrates, and G. Pantè (MATIS CNRIMM) for expert technical assistance.

Supporting Information. Supporting Information Available: PL and large area SEM images of ZnO NWLs with or without Al anodization process. This material is available free of charge via the Internet at http://pubs.acs.org.

23 ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 26

REFERENCES [1] [2] [3] [4] [5] [6] [7]

[8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20]

[21] [22]

Alvi, N.H.; Usman, Ali S.M.; Hussain, S.; Nur, O., Willander, M. Scr. Mater., 2011, 64, 697-700. Ye, C.; Bando, Y.; Shen, G.; Golberg, D. J. Phys. Chem. B, 2006, 110, 15146-15151. Hatch, S.M.; Briscoe, J.; Sapelkin, A.; Gillin, W. P.; Gilchrist, J. B.; Ryan, M. P.; Heutz, S.; Dunn S. J. Appl. Phys., 2013, 113, 204501. Hassan, J.J.; Mahdi, M. A.; Kasim, S. J.; Naser, M. A.;, Abu Hassan, H.; Hassan, Z.; Appl. Phys. Lett., 2012, 101, 261108. Pescaglini, A.; Martín, A.; Cammi, D.; Juska, G.; Ronning, C.; Pelucchi, E.; Iacopino, D.; Nano Lett., 2014. 14, 6202-6209 Kim, J.Y.; Jo, S.-Y.; Sun, G.-J.; Katoch, A.; Choi, S.-W.; Kim, S. S. Sens. Actuators, B 2014, 192, 216-220. Fulati, A.; Usman, S. M. A.; Asif, M. H.; Hassan Alvi, N. Ul; Willander, M.; Brännmark, C.; Strålfors, P.; Börjesson, S. I.; Elinder, F.; Danielsson, B. Sens. Actuators, B 2010, 150, 673–680 Maiolo, L.; Mirabella, S.; Maita, F.; Alberti, A.; Minotti, A.; Strano, V.; Pecora, A., Shacham-Diamand, Y.; Fortunato, G. Appl. Phys. Lett., 2014, 105, 093501. Wang, X.; Ding, Y.; Li, Z.; Song, J.; Lin Wang, Z. J. Phys. Chem. C, 2009, 113, 17911794. Ng, H. T.; Li, J.; Smith, M. K.; Nguyen, P.; Cassell, A.; Han, J.; Meyyappan, M. Science 2003, 300, 1249 Gao, S.Y.; Li, H.D.; Yuan, J.J.; Li, Y.A.; Yang, X.X.; Liu, J.W. Appl. Surf. Sci., 2010, 256, 2781-2785. Cheng, J.P.; Zhang, X.B.; Luo, Z.Q. Surf. Coat. Technol., 2008, 202, 4681-4686. Lao, J.Y.; Huang, J.Y.; Wang, D.Z.; Ren, Z.F.; Steeves, D.; Kimball, B.; Porter, W. Appl. Phys. A: Mater. Sci. Process., 2004, 78, 539-542. Maeng, J.; Jo, G.; Choe, M.; Park, W.; Kwon, M.-K.; Park, S.-J.; Lee, T. Thin Solid Films, 2009, 518, 865-869. Yao, L.; Zheng, M.; Li, C.; Ma, L.; Shen, W. Nanoscale Res. Lett. 2012, 7, 1-8. Strano, V.; Urso, R.G.; Scuderi, M.; Iwu, K. O.; Simone, F.; Ciliberto, E.; Spinella, C.; Mirabella, S. J. Phys. Chem. C, 2014, 118, 28189-28195. Koh, Y.W.; Lin, M.; Tan, C.K.; Foo, Y. L.; Loh, K. P. J. Phys. Chem. B, 2004, 108, 1141911425. Nayak, A.; Katzenmeyer, A.M.; Gosho, Y.; Tekin, B.; Islam, M. S. Appl. Phys. A: Mater. Sci. Process., 2012, 107, 661-667. Bwana, N. N. J. Nanopart. Res., 2008, 10, 313-319. Mayer, M. SIMNRA a Simulation Program for the Analysis of NRA, RBS and ERDA, Proceedings of the 15th International Conference on the Application of Accelerators in Research and Industry, J. L. Duggan and I.L. Morgan (eds.). AIP Conf. Proc. 1999, 475, 541–544.www.simnra.com Wang, R.; Hashimoto, K.; Fujishima, A.; Chikuni, M.; Kojima, E.; Kitamura, A.; Shimohigoshi, M.; Watanabe, T. Nature 1997, 388, 431-432 IUPAC. Compendium of Chemical Terminology, 2nd ed. (the "Gold Book"). Compiled by A.D. McNaught and A. Wilkinson, Blackwell Scientific Publications, Oxford, UK

24 ACS Paragon Plus Environment

Page 25 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

[23] [24] [25] [26]

(1997). (b) XML on-line corrected version: http://goldbook.iupac.org (2006- ) created by M. Nic, J. Jirat, and B. Kosata; updates compiled by A.D. Jenkins. Scuderi, V.; Impellizzeri, G.; Romano, L.; Scuderi M.; Nicotra, G.; Bergum, K.; Irrera, A.; Svensson, B. G.; Privitera, V. Nanoscale Res Lett, 2014, 9, 458 Sberna, P.; Scapellato, G. G.; Piluso, N.; Boninelli, S.; Miritello, M.; Crupi, I.; Bruno, E.; Privitera, V.; Simone, F.; Mirabella, S. Appl. Phys. Lett., 2013, 103, 221902 Pachauri, V.; Kern, K.; Balasubramanian, K. Chem. Phys. Lett. 2010, 498, 317–322 Chong, M. N.; Jin, B.; Chow, C. W. K.; Saint, C. Water Res. 2010, 44, 2997-3027

25 ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 26

FOR TABLE OF CONTENTS USE ONLY TITLE: Enhanced Quality, Growth Kinetics and Photocatalysis of ZnO Nanowalls Prepared by Chemical Bath Deposition

AUTHORS: Kingsley O. Iwu1, Vincenzina Strano1, Alessandro Di Mauro1, Giuliana Impellizzeri1, Salvo Mirabella1*

Brief Synopsis (60 words) ZnO nanowalls grow vertically on Al covered substrates. We synthesize them by chemical bath deposition enhancing the generation of Al(OH)4- (driving the NWL growth mode) by preliminary oxidation of the Al film. The NWL growth kinetics is investigated and summarized in a selfscreening model. Excellent film quality and good photocatalytic activity in the degradation of toxic methyl orange are shown.

26 ACS Paragon Plus Environment