ZnO Nanostructures Application in Electrochemistry : Influence of Morphology

,


INTRODUCTION
ZnO is an n-type semiconductor material with wide bandgap (3.37 eV) and high exciton binding energy (60 meV).The outstanding popularity of ZnO and their nanostructures application are constantly increasing.The ZnO multifunctionality is interesting for various applications, such as catalysts, 1 photovoltaics, 2,3 sensors, 4−9 mixed oxide varistors, 10 rubber, 11 and concrete additives, 12,13 serving to large-scale industry needs.ZnO nanostructures can be manufactured by using miscellaneous synthesis methods that, according to the processes used, can be divided into chemical synthesis methodshydrothermal synthesis, 14−16 combustion, 17−20 electrochemical deposition, 21 chemical vapor deposition, 22−24 spray pyrolysis, 25 and so onand physical methodsarc discharge method, 26,27 radio-frequency magnetron sputtering, 11,28 molecular beam epitaxy, 29 pulsed laser deposition, 30 and so on.
Many works are devoted to the physical and electrochemical property dependence on ZnO structure.The control of the size and the shape of ZnO nanostructures is essential for many advanced applications. 1,3,10,11,43Therefore, there are many studies done to investigate the size and shape influence on physical properties (photoluminescence, transmittance, absorbance, etc. 14,44 ) and electrochemical properties (photocatalytic activity, peak separation, active surface, potential window measurements, etc. 1−3,27,43,45−48 ).Until now the dependence of the ZnO nanostructure forms on the electrochemical performance of this material is not evident.
Morphological and structural properties of the material strongly affect its properties and, therefore, define electrochemical applications.The structure of the electrode surface contributes to the cyclic voltammetry characteristics and is essential for several important parameters such as the active electrode area calculated by the Randles−Sevcik equation and peak separation to determine the number of electrons transferred.Therefore, a cyclic voltammetry study using potassium ferricyanide, as an indicator, is a convenient tool for surface characterization of electrodes, where the charge transfer at the interface between electrolyte and electrode must occur either by a barrier or through defects.Because ΔE p is a function of electron transfer rate, hence the lower the ΔE p , the greater the electron transfer rate, thus minimizing the background current. 49−52 Therefore, the aim of this work is to investigate the influence of ZnO nanostructure morphology (size, shape, and porosity) on its electrochemical properties.Thus, we used different morphologies of ZnO nanostructures for comparison: (i) ZnO tetrapods (ZnO-Ts) which we synthesized by the combustion method 18 and separated into two fractions by size; (ii) commercially available ZnO nanorods and nanoparticles.ZnO-Ts of different sizes were made under the same conditions (temperature, synthesis time, etc.) and afterward were separated by centrifugation into two fractions by size, in such a way ruling out the influence of synthesis conditions on their properties.To estimate the structure and morphology influence on electrochemical properties, SEM, TEM, BET, XRD analysis, and UV−vis spectroscopy were used; electrochemical properties were measured by using cyclic voltammetry, employing Au electrodes with a thin film of ZnO nanomaterial.Compared to other ZnO nanostructures, large ZnO-Ts demonstrated the lowest peak separation and the highest active surface values, which can be attributed to their pore size and shape.

MATERIALS AND METHODS
2.1.ZnO Nanostructure Preparation.ZnO-Ts were synthesized by the combustion method as described in previous work. 18Briefly, micrometer-sized Zn particles were introduced from the top ending of the quartz tube into a vertical furnace under an ambient air atmosphere.During the combustion process occurring at 1173 K, ZnO-Ts were formed, and the obtained powder was collected from the bottom ending of the quartz tube on a filter.As-obtained ZnO-Ts were dispersed in absolute ethanol and separated by centrifugation using two rotation speeds (1000 and 3000 rpm) into two different fractions of the supernatant of the liquid, and these fractions were used in this work.The size of the particles that precipitate due to centrifugal force is proportional to the particle size; therefore, larger diameter ZnO-Ts (LT, Table 1) were obtained in the supernatant solutions at 1000 rpm compared to smaller ZnO-Ts after 3000 rpm (ST).
ZnO nanoparticles from Sigma-Aldrich (product no.544906), particle size <100 nm, are produced by a vapor method using electric arcs to vaporize precursor materials, which are then carefully condensed to produce nanoparticles.In this paper nanoparticles are labeled as NP (Table 1).
ZnO nanoparticles from Sigma-Aldrich (product no.544906), particle size <100 nm, are produced by a vapor method using electric arcs to vaporize precursor materials, which are then carefully condensed to produce nanoparticles.In this paper nanoparticles are labeled as NP (Table 1).
Commercially available ZnO (Kadox-25, New Jersey Zinc Co.) obtained by combustion of zinc metal, exhibiting wurtzite crystal nanocrystals with the prevalence of hexagonal facets, 53 was used for comparison.ZnO Kadox nanocrystals show a moderate aspect ratio, and a predominant nanorod-like structure can be assumed.All ZnO nanostructures were not modified, and commercial ZnO nanostructures were used as received.
The working electrode was prepared on a glass substrate by sputtering of 100 nm thick, 10 mm long, and 5 mm wide gold strip through the mask.A layer of ZnO nanostructures was deposited on a glass substrate heated to 353 K by airbrush spraying of 1 mg/mL solution of ZnO nanostructures in ethanol after mild sonication.Silver paint was used for the low resistance contact with potentiostat clips on the part, which was not immersed into the liquid.
Resistivity measurements of ZnO nanostructures were made by using ZnO nanostructure solutions in ethanol after mild sonication and deposited by airbrush spraying between two electrodes with a gap of 20 μm on a glass substrate heated to 353 K; the layer thickness was kept at 15 μm for all ZnO nanomaterials.Measurements were performed at room temperature (298 K) with ambient air (30% RH); the resistivity was calculated from the I−V curve linear region −5 to −10 V as R = dV/dI. 54.2.Electrochemical Stability Window (Cyclic Voltammetry and Potential Window Measurement).The electrochemical measurements were performed by a computercontrolled potentiostat (Metrohm, Autolab).A standard threeelectrode cell (volume 100 mL) was used.The potential window measurement was made in 20 mL of 0.1 M KCl electrolyte solution.Measurement of ΔE p was made with 100 μM K 4 [Fe(CN) 6 ] dissolved in 0.1 M KCl.During the electrochemical measurements an Ag/AgCl reference electrode (RE) of 2 mm diameter (World Precision Instruments, DRIRFE-2) and a Pt counter electrode (CE) of 0.5 mm diameter (Bio-Logic Science Instruments, A-002222) were used.The working electrode was a glass electrode covered with various ZnO nanostructures on a gold strip.For all types of ZnO nanostructures the coating thickness was maintained at 5 μm.The simplified electrochemical measurements setup and SEM images of various ZnO nanostructures used in this study are shown in Figure 1.
The potential window was evaluated by using cyclic voltammetry measurements, which were made by using 0.1 M KCl solution as electrolyte.Potassium hexacyanoferrate(II), K 4 [Fe(CN) 6 ], in the amount of 100 μM (99.5% purity) was used for electron transfer measurements.Only freshly prepared solutions were used for the measurement and were not deaerated during the experimental runs.The Journal of Physical Chemistry C 2.3.Structural Characterization.A PANalytical X'Pert PRO XRD diffractometer was used for the X-ray powder diffraction (XRD) measurements.Technical specification: Ni filtered Cu radiation in a standard Bragg−Brentano geometry.The International Centre for Diffraction Data (ICDD) database was used to compare XRD patters and to identify the phase.The Debye−Scherrer equation was used to calculate crystallite sizes; the constant k 1 value of 0.94 was used in this work. 55Zeiss Evo50 SEM instrument connected with an Oxford energy dispersive X-ray detector was used.In addition, a JEOL 3010-UHR TEM instrument with 300 kV operating voltage was used to investigate the crystalline structure of the samples.Technical specifications: 300 kV, 2K × 2K pixels Gatan US1000 CCD camera.
To determine the Brunauer−Emmett−Teller (BET) surface area and the porosity properties, characterization of the samples' surface area and N 2 adsorption−desorption measure-  The Journal of Physical Chemistry C ments were performed at 77 K on a Micromeritics ASAP 2020 instrument.
A double-beam UV−vis spectrophotometer (Varian Cary UV 300 Bio) was exploited for optical absorbance measurements in the wavelength range 190−800 nm.All samples were dispersed in absolute ethanol, and spectra were estimated in the transmission mode.
Veusz, Edraw Max 9.2, and Adobe Photoshop CC 2019 image acquisition tools and image processing software packages were used.Shape software was used for the external morphology (faces) representation of crystals.

RESULTS AND DISCUSSION
3.1.Morphology and Structure.We synthesized ZnO tetrapods exploiting the combustion method, 17,18 which has several advantages, such as low cost as well as continuous and high yield synthesis. 18,19ZnO-Ts subsequently were separated by size into two fractions by using a centrifuge: large ZnO tetrapods were collected at 1000 rpm and small tetrapods at 3000 rpm.Therefore, since the ZnO tetrapod structure was not affected by synthesis conditions, it gives the opportunity to investigate how the morphology (i.e., dimensions) influences the electrochemical properties.
To compare various morphologies, commercially available ZnO nanostructures were used.High aspect ratio commercial ZnO nanostructures, i.e., nanorods, were manufactured by using the chemical vapor deposition method.Other commercial ZnO nanomaterials with low aspect ratio, i.e., nanoparticles, were synthesized by using the physical arc discharge method.Moreover, ZnO NRs were synthesized by the chemical route (chemical vapor deposition), whereas ZnO NPs were synthesized by the physical route, condensing nanoparticles after arc discharge.This gives a possibility to evaluate electrochemical properties of two distinct ZnO nanostructure synthesis methods.
From SEM images of the electrode containing the different ZnO nanostructures illustrated in Figures 1, a particular orientation of the tetrapod structures with respect to the other nanostructures (nanorods and nanoparticles) can be observed.This characteristic will be discussed later when morphology and structure as obtained by TEM images will be shown.
Figure 2a−d shows TEM and HRTEM images of STs, LTs, NRs, and NPs.ZnO-Ts prepared by the earlier developed combustion method have a typical tetrapod structure consisting of four connected nanorods (legs) as imaged in the low-resolution TEM images (top-left insets of Figure 2a,b).ZnO tetrapods collected after 3000 rpm (STs) have smaller leg diameters (5 nm) and shorter length (50 nm) compared to those collected after 1000 rpm (LTs), which exhibit mean diameters and leg lengths of 20 and 150 nm, respectively.The atop leg region of a ST is HRTEM imaged in Figure 2a, where well-defined interference fringes corresponding to ZnO (101̅ 0) planes, as obtained from the fast-Fourier transform (FFT) images (bottom-left inset of Figure 2a), show two bright spots which are associated with 0.28 Å regularly spaced planes.The bottom region of a leg (ST) is HRTEM imaged in Figure 2b.Well-defined interference fringes corresponding to ZnO (101̅ 0) planes, as obtained from the FFT images (bottomleft inset of Figure 2b), show two bright spots which are associated with 0.28 Å regularly spaced planes.HRTEM shows the highly crystalline structure of ZnO-Ts; the most exposed surfaces are laterally exposed prismatic (nonpolar) facets of ZnO-T legs.
ZnO NRs calcined at 400 °C display a hexagonal prismatic habit (top-left inset of Figure 2c) with a typical preferential growth direction.A more precise investigation reveals that the ZnO crystals are elongated along the prismatic (101̅ 0) direction and are highly crystalline, as testified by the 0.28 Å spaced well-defined interference fringes, whose corresponding SAED pattern is imaged in the bottom-left inset of Figure 2c.Furthermore, such ZnO nanocrystals have apparently rounded terminations made of small facets together with defects (terraces and steps). 56t is not surprising that ZnO-Ts and NRs expose the [0001] preferential growth direction due to the fact that their synthesis involves high-temperature treatments, under which {101̅ 0} and {112̅ 0} prismatic facets are more stable, 56 while {0001} facets, formed of only one type of ion (polar surfaces), may reconstruct with the formation of microsteps. 57In the case of the ZnO NPs the situation is completely different (Figure 2d).Nanoparticles are larger (50−80 nm), exposing a more complex habit, which is composed of a variety of surfaces due to their shapes (top-left inset of Figure 2d); as also shown in the ZnO constructed crystal (Figure 2c), interference fringes with 0.28 Å space correspond to (101̅ 0) planes.We can conclude that because of their morphology, ZnO NPs exhibit a larger amount of polar terminations compared to ZnO elongated structures.
It is important to emphasize that tetrapods by their nature are made of legs branching radially from the joint region, in a way that if a tetrapod is on a flat surface, one leg is always arranged vertically, while the others are oriented downward.This characteristic can be a determining factor for electron transport lowering the number of needed junctions in percolation path of tetrapod based layer compared to a more complex path of other geometries due to jump from one nanoparticle to another.A HRTEM image of a connection region of tetrapods is shown in Figure 3.
UV−vis spectroscopy was used to study the optical properties of the ZnO nanostructures.UV−vis spectra of various ZnO nanostructures dispersed in absolute ethanol are shown in Figure 4a.Because of the direct electron transitions from the valence band to the conduction band, a sharp absorption in the 350−600 nm interval is seen.The spectra reveal a characteristic absorption peak of ZnO at the wavelength of ∼370 nm, which can be assigned to the intrinsic band gap absorption of ZnO due to the electron transitions from the valence band to the conduction band (O 2p → Zn 3d ). 58,59Comprehensively, moving from the pink curve (ST) (Figure 4a, inset) to the dark violet curve (LT), and then up to The Journal of Physical Chemistry C the light blue (NR) and dark green curves (NP), a shift to the higher wavelength is observed.The peak shift for smaller ZnO nanostructures can be associated with the quantum size effects.It is worthy to note that NRs and NPs behave as 3D systems because of their roughly remarkable sizes, while TPs approach properties of low-dimensional systems (i.e., leg diameter).Decreasing the ZnO size leads to smaller excitation frequencies.A blue-shift is observed in the extinction behavior, and the absorption edge is positioned at smaller wavelengths. 60t can be seen that with the increase of the particle size and shape, the slope of the shoulders becomes steeper (Figure 4b, LT and ST curves).This can be associated with the increase of the visible part scattering effects for larger ZnO nanostructures. 17 2 -adsorption/desorption isotherms at 77 K and the associated pore size distributions (PSDs), determined by applying the DFT (density functional theory) method, of the ZnO nanostructures after the electrode preparation are shown in Figure 4.All isotherms are of type IV with hysteresis loops, indicating the mesoporous character.The contribution of small micropores is negligible.This observation is confirmed by PSDs reported in Figure 4b.In more detail, a very broad distribution of pores in the 20−400 Å range is shown for all samples, with larger contribution of the smaller mesopores (20−60 Å).Interestingly, this prominent distribution is remarkably intense for the ZnO ST and roughly negligible for the ZnO NR sample.The weak hysteresis loop observed for the adsorption/desorption isotherms (Figure 4b) and the wide distributions of the pore sizes (Figure 4c) can be explained with a porosity coming from the aggregation state of the ZnO tetrapods, nanorods, or nanoparticles in the electrode. 61In fact, while for short tetrapods (ST), nanorods (NR), and nanoparticles (NP) the empty spaces could be more efficiently filled, the packing of the long tetrapods (LT) is prevented by their long legs.
The BET surface area values and pore sizes of the ZnO nanostructures, calculated from the N 2 -adsorption isotherms, are summarized in Figure 4d.Tetrapod-shaped ZnO nanostructures had the highest surface area compared to ZnO NRs and NPs.Also it is known that BET analysis evaluates only the open pores.Because of this, the tetrapods may accumulate and form agglomerates with the closed type pores (intercluster pores). 62We can assume that all samples exhibited a broader and extended mesoporosity.Because of mesoporosity properties developing a high specific surface area, molecular species can be adsorbed; therefore, finally pore filling occurred through capillary condensation governed by various factors largely of kinetic origin. 62t can be concluded that even though the surface area of all ZnO nanostructure is in the range 8−22 m 2 /g (Figure 4b,c), the best results are obtained with ZnO LTs and STs, probably due to higher porosity they created in the film assembling.The results obtained from the Debye−Scherrer equation show a significant increase in FWHM values from 0.136 for nanoparticles to 0.37 for small ZnO-Ts.The FWHM is sensitive to the variation in microstructure; therefore, it affects the size of the crystals.The increasing FWHM values can be due to the decrease of crystal sizes: from ∼63 to 23 nm and from ∼85 to 61 nm.The obtained results for different faces are reported in Figure 5.An increase in stacking faults and structural disorder widens the XRD peaks. 63.2.Electrochemical Properties.It is known that electrochemical deposition parameters (potential, temperature, time, surface coverage, and compactness) affect ΔE p values.For example, Zawawi et al. 52 determined that the anodic peak current increased and the peak potential shifted to a more positive potential with the decrease in temperature, and the ZnO electrode prepared at 333 K was quasi-reversible with a peak separation of 0.42 V. 52 It was also determined that the ZnO-modified ITO (indium tin oxide covered glass) prepared at −0.9 V deposition potential showed the smallest ΔE p value compared to electrode prepared at −1.0 and −1.1 V.However, to the best of our knowledge, there is no systematic analysis of the ZnO nanomaterial shape and morphology influence on the electrochemical peak separation value (ΔE p ).
Therefore, we compared the influence of various ZnO nanostructure morphologies on electrochemical properties, such as potential window, peak-to-peak separation, and calculated active surface.
3.2.1.Electrochemical Reactions of ZnO Nanostructures under Different Potentials.The potential window, showing the electrochemical stability of solvents and electrolytes, is an important measure for determining the oxidation and reduction potential. 64The electrochemical characteristics of potential window for the ZnO nanostructures were estimated from the cyclic voltammograms in a 0.1 M KCl solution (Figure 6).The scan range of −0.2 to 0.7 V (vs Ag/AgCl) and the scan rate of 25 mV s −1 were used.Even though the The Journal of Physical Chemistry C background current slightly increased, no considerable decrease of potential was observed for different ZnO morphologies; the potential window was around 1.2 V for all ZnO nanostructure types.The potential window was determined by the threshold current of 0.2−0.3μA.
3.2.2.Electrochemical Reactions of Ferricyanide on the ZnO Electrode.It is known that the inner-sphere surfacesensitive ferricyanide ([Fe(CN) 6 ] 3− ) redox couple is dependent on the electrochemical reactivity and surface morphology of the working electrode.For example, in the reversible system the electrochemical peak separation (ΔE p ) can be used to determine the number of electrons transferred; accordingly, a fast one-electron process exhibits ΔE p = 59 mV. 65,66s a result, to investigate electronic carrier transport in various ZnO nanostructures, the reversible redox couple of ferricyanide/ferrocyanide was used.The cyclic voltammetry of the redox reaction including [Fe(CN) 6 ] 3−/4− (100 μM) in a 0.1 M KCl solution (Figure 7) was employed to evaluate the potential window of ZnO electrodes.
The electron transfer rate is faster than the rate of ferricyanide diffusion from the solution to the electrode surface due to the concentration gradient, which forms due to ferricyanide reduction to ferrocyanide.The theoretical value of one-electron transfer in the ideal case would be ΔE pT = 0.059 V at 298 K. 66 Therefore, the peak current I p (measured as the distance from the baseline to the peak) is dependent on the active area of the electrode as is calculated by using the Randles−Sevcik equation, 67,68 where the constant k 2 = 2.2 × 10 5 and D is the diffusion coefficient, D = 7.2 × 10 −6 cm 2 s −1 for [Fe(CN) 6 ] 3−/4− . 69,70t can be noted that anodic and cathodic peaks have a symmetric shape for ST, LT, and NR curves which indicates the reversible redox reaction of ferricyanide occurred in these systems.However, the electrode covered with NPs has an asymmetric shape which indicates an irreversible redox reaction, which can be associated with the lower surface area and pore size.
9][50][51][52]65,66 The electron transfer rate can be evaluated by using ΔE p , which corresponds to the potential difference between the oxidation and reduction peaks in the cyclic voltammograms. To meure the oxidation peak, the potential window was changed by altering the voltage interval.By the increase of the voltage to 1.1 V the oxidation peak changes from 0.4 to 0.7 V (Figure 7).The spreading −0.4 to 1.0 V indicates Au dissolution.71,72 To determine the accurate ΔE p value, the −0.4 to 1.0 V interval for the potential window measuring was further used.
Figures 8a and 8b show cyclic voltammograms of [Fe-(CN) 6 ] 3−/4− for the different morphologies of ZnO nanomaterials and the bare Au electrode.The ΔE p decreased from 230.7 mV (for NP) to 61.7 mV (for LT).This decrease indicates an enhancement in electrochemical activity because of the changing ZnO surface area and surface morphology (Figures 8c and 8d).Larger than 0.059 V, ΔE p indicate that ferricyanide molecules were not adsorbed but were free to diffuse toward the electrode surface, 73 as it can be noticed in Figure 8a,c.
3.2.3.Mechanisms Influencing Electrochemical Performance of Various ZnO Nanomaterials.The electrochemical properties of ZnO nanostructures are influenced by two simultaneous mechanisms: (i) the electrochemical reactions of ZnO nanostructures under different potentials and (ii) the electrochemical reactions of ferricyanide on the ZnO electrode.The electrochemical reaction of ferricyanide using electrodes with various ZnO morphologies is considered as a diffusioncontrolled process similar to the single electron transfer quasireversible electrochemical process. 73The best results are reached by using the electrode covered with LT which has the biggest active surface (0.095 cm 2 ) and its practical ΔE p value of 61.7 mV, which is close to the theoretical 59 mV (Figure 7d).Consequently, a higher active area is connected to a higher electron transfer rate.Low or close to the theoretical ΔE p value indicates a kinetically favorable route which minimizes background current and hence indicates electrode suitability for possible applications, e.g., biosensing. 49,50,52he high aspect ratio of the tetrapods might also lead to a higher electron conduction through the layer.ZnO nanostructure resistivity measurements showed resistivity of STs and LTs in the order of 10 2 Ω•m (Figure 9), which is consistent with earlier results, 74 whereas other ZnO nanostructures demonstrated an order of magnitude higher resistivity.The low resistivity of ZnO tetrapods could be associated with the one-dimensional morphology of their legs, which could  The Journal of Physical Chemistry C provide a direct and stable pathway for rapid electron transport. 75he electrochemical performance of ZnO nanomaterials may be affected by electron transfer at the surface active sites and by the conduction through the layer from the active site to the electrode.The quality of the crystal influences the electron transfer rate at the surface, since surface defects can act as active sites for electron transfer. 76However, the direct electron conduction from the active sites to the electrode is important, as the resistance of the material can increase the peak separation due to voltage drop in the material. 77,78Because ZnO nanostructures have the same crystal structure, their resistivity is mostly determined by the barriers at the particle− particle junctions, formed at the electron percolation path through the ZnO nanostructure layer from the surface to the Au electrode, as it is shown schematically in Figure 10.The specific ZnO tetrapod geometry of elongated legs spreading radially in four different directions gives the advantage of fewer barriers on the percolation path ensuring lower voltage drop, whereas a lower aspect ratio ZnO morphology (NP and NR)   The Journal of Physical Chemistry C percolation path encounters more junctions deteriorating the electron conduction, which is consistent with the resistivity measurements.
Different ZnO nanostructure morphology influences the layer packing during deposition, influencing pore sizes and forms.Consequently, reactant diffusion in ZnO nanostructures electrodes affects the electron transfer rate.ZnO tetrapods (LTs and STs) were characterized by the smallest pores, and it can be assumed that there is a lack of intercluster pores, which might indicate peculiar pores with openings only at one end at two ends (through pores), which could cause the [Fe-(CN) 6 ] 3−/4− molecule easier to adsorb and react on the surface. 62onsequently, the unique structure of tetrapods shows better electrochemical properties compared to other ZnO nanostructure counterparts due to geometry of their onedimensional legs arranged in four directions, leading to more efficient electron transfer from the active site to the electrode; at the same time the ZnO tetrapod layer packing during deposition results in both high porosity and active area, ensuring low peak separation.The effect is more pronounced for longer leg tetrapods (LTs), leading to peak separation ΔE p , approaching the theoretical value.

CONCLUSIONS
We examined the influence of the ZnO nanostructure morphology on the electrochemical properties using cyclic voltammetry and potential window measurements.Compared to ZnO nanorods and nanoparticles, ZnO tetrapods demonstrated better electrochemical properties.Interestingly, the larger tetrapods (LT diameter of 120 nm and ST diameter of 40 nm) showed the highest active surface value (0.095 cm 2 ) and a ΔE p of 61.7 mV, which is close to the theoretical ZnO value (59 mV).UV−vis analysis showed a blue-shift in connection with the decrease of ZnO nanostructures size, which can be associated with the quantum size effect.Moreover, with the increase of the ZnO nanostructure size and shape, the slope of the shoulders became steeper since the scattering effects dominate.The one-dimensional legs arrangement of ZnO tetrapods compared to other ZnO nanostructure counterparts leads to superior electrochemical properties due to more efficient electron conduction; at the same time both porosity and active area are kept high as a result of prevented packing.The effect is more pronounced for longer leg tetrapods (LTs), leading to peak separation ΔE p , approaching the theoretical value.These results are important for the future ZnO nanostructure applications in electrochemistry, where the influence of morphology on the properties should be considered.

Figure 2 .
Figure 2. TEM and HRTEM of ZnO tetrapods after centrifugation at (a) 1000 rpm (LTs) and (b) 3000 rpm (STs).(c) ZnO nanorods and (d) ZnO nanoparticles.Representation of the (101̅ 0) nonpolar surfaces is shown in (a−c), while the simulated surface of the ZnO nanoparticles is illustrated in (d).In the top-right insets, representations of the HRTEM selected regions of tetrapods are depicted.In the bottom-left insets, fast-Fourier transform (FFT) images of the selected regions in (a−d) are illustrated.

Figure 3 .
Figure 3. HRTEM images illustrating four leg joint regions in (a) STs and (b) LTs.Leg preferential directions are also illustrated.

Figure 6 .
Figure 6.Cyclic voltammograms for potential windows measurement of various ZnO nanostructures and bare gold electrode (inset).

Figure 8 .
Figure 8.(a) Cyclic voltammograms of [Fe(CN) 6 ] 3−/4− measured for electrodes, coated with ZnO nanostructures.(b) Cyclic voltammogram of [Fe(CN) 6 ] 3−/4− measured for glass electrode covered with bare gold without ZnO materials.ΔE p values are denoted by bars at the top.(c) Active area and (d) ΔE p values for different ZnO morphology.The theoretical ΔE pT = 59 mV value is marked on the axis.

Figure 10 .
Figure 10.Electron transport in ZnO nanostructures: schematics of percolation path (red) through the ZnO nanostructure layer with the marked barriers (blue crosses) at the particle−particle junctions.