A simple and template free synthesis of branched ZnO nanoarchitectures for sensor applications

Abstract

Strong electrophilic natured acetaldehyde present in various food and beverages damages genetic material and induces diseases like atherosclerosis. Detection and quantification of such a carcinogen poses a major challenge. In this context, a novel room temperature acetaldehyde sensor made up of hierarchical ZnO nanostructures and prepared by a simple and template-free method has been reported. ZnO nanostructures were grown on glass substrates by a chemical spray pyrolysis technique at the substrate temperature of 523 K. Different nanostructures, namely tiny nanoplatelets, branched nanorods and thicker nanoplatelets, were formed by an annealing process. The crystal structures, morphologies and optical absorbances of the hierarchical ZnO nanostructures were investigated by X-ray diffraction (XRD), field emission scanning electron microscopy (FE-SEM) and UV-vis spectrophotometry, respectively. The branched nanorods showed an excellent sensing response towards 20 to 500 ppm of acetaldehyde vapour. The role of high density junctions of the branched ZnO architecture in enhancing the vapour sensing performance has been highlighted. The observed selectivity, range of detection and stability of the branched ZnO nanorods have proven their potential as a sensing element for the detection of acetaldehyde.

---

1. Introduction

The American Conference of Governmental Industrial Hygienists (ACGIH) and the Environment Agency in Japan reported that the maximum permitted concentration of acetaldehyde is only about 10 and 50 ppm respectively.¹ Japan, USA and the Netherlands listed acetaldehyde as a hazardous air pollutant.² Consumers may be exposed to acetaldehyde in several ways because of its presence in cheese, cooked beef, chicken, oak and tobacco leaves, and it is also commonly found in alcoholic beverages. Due to its strong electrophilic nature, it is believed to react with DNA to induce biological changes such as mutagenesis and carcinogenesis.³ It also reacts with low density lipoproteins (LDH) and causes diseases such as atherosclerosis and acute alcoholic liver disease.⁴ Moreover, acetaldehyde can be used as a biomarker for wine, beer, yoghurt and meat quality assessment applications.⁵ Hence, there is a crucial demand to explore sensors which are capable of monitoring acetaldehyde in real time environments. Detection of acetaldehyde can be performed in several ways such as colorimetric, chemiresistive, gas spectrometric, spectrofluorometric, chromatographic, chemiluminescence, liquid chromatography and enzymatic methods.^6,7

Due to high performance, easy operation and low cost, metal oxide based chemical sensors have been exploited for numerous applications in various fields like automotive, industrial, aerospace, medical, domestic, security and food industries.^8,9 Several metal oxides such as TiO₂, Al₂O₃, SnO₂, MoO₃, ZnO, V₂O₅ and WO₃ have been employed to detect hazardous gases in room and elevated operating temperatures.^10,11 Among them, ZnO is a commonly used sensing material due to its unique catalytic, optical and electrical properties.^12–14 ZnO nanostructures with tailored shapes and sizes are often considered to improve the performance of gas sensors. Several methods have been employed to tune the sensitivities and selectivities of these materials, such as element doping, functionalization with noble metals and heterostructure formation.^15–17 To date, various morphologies of ZnO, including quantum dots, nanowires, nanotubes, nanoneedles, nanobelts, nanosheets, nanowalls, etc., have been extensively employed to fabricate gas sensors.^18–22

ZnO with different nanostructured morphologies has been preferred for sensing applications because of its very high surface to volume ratio, grain dimension comparable to the space charge region and superior stability.^23,24 Recent studies revealed that branched/networked types of special morphologies are often associated with enhanced gas sensing properties. Complex surface morphologies have received greater research interest due to their novel properties such as enhanced electrical conductivity, thermal stability, catalytic properties and low cost.^25,26 Mostly, seeded growth approaches have been reported for the preparation of branched 1D nanostructures. This multistep preparation is a major obstacle which limits their practical applications. In addition, the preparation methods require costly equipment and highly skilled experts.^27–29 ZnO thin films have been deposited by several methods, but a spray pyrolysis technique is preferred owing to its advantages such as simplicity, large productivity, cost effectiveness and being environmentally safe since water is used as the solvent and no vacuum is required.^30–32

Branched 1D nanostructures have been efficiently used in a wide variety of applications such as solar cells, field emitters, photo detectors, supercapacitors, transparent EMI shielding, fuel cells, etc.³³ However, limited reports are available on gas sensor applications. Branched ZnO heterostructures were used to detect ethanol, n-butanol, NO₂, etc., at elevated operating temperatures.^34–38 Most gas sensors operate at higher temperatures, which is not beneficial for practical applications in biological or explosive environments. To our knowledge no work has been reported on the room temperature detection of acetaldehyde using ZnO nanostructures. Hence, in this work, a simple and template-free synthesis of branched ZnO architectures has been devised and their acetaldehyde sensing abilities at room temperature were investigated.

2. Experimental section

2.1 Film deposition

The hierarchical ZnO nanostructures derived from a simple chemical spray pyrolysis technique (HOLMARC, HO-TH-04, India) were deposited on glass substrates at the temperature of 523 K.³⁹ Various forms of nanostructures were obtained by annealing the as-deposited samples at 523, 623 and 723 K. As a first step, 0.1 mol of anhydrous zinc chloride (ZnCl₂, Merck, Purity 99%) was dissolved in 50 mL of deionized water (Millipore, USA) and subjected to constant stirring for 1 h. Glass substrates (Blue Star, Mumbai) were ultrasonically cleaned with acetone, ethanol and deionized water. The substrates were then positioned on the substrate heater of the spray system. Prior to spraying, the prepared precursor solution was loaded into the dispenser. The precursor solution was sprayed by an atomizer on glass substrates maintained at a constant temperature of 523 K. 50 mL of precursor solution was sprayed over the substrate with spray rate and area of 2 mL min⁻¹ and 15 cm min⁻¹, respectively. The substrate to nozzle distance and carrier gas pressure were fixed to 15 cm and 2 mbar, respectively.

2.2 Characterization

The structural and morphological properties were studied using an X-ray diffractometer (D8 Focus, Bruker, Germany) and a field emission scanning electron microscope (JEOL, 6701F, Japan), respectively. The optical properties of the thin films were studied using a UV-vis spectrophotometer (Perkin Elmer, Lambda 25, USA) in the wavelength range of 200 to 800 nm with a scan rate of 50 nm min⁻¹. The water contact angle over the film surface was measured using a contact angle (CA) goniometer (ramé-hart, Model 250, USA). Electrical and acetaldehyde sensing properties were examined using an electrometer (Keithley, 6517A, USA).

2.3 Sensor fabrication and measurements

The room temperature (303 K and 55% RH) sensing abilities of the ZnO nanoarchitectures were examined in a sealed testing chamber of 5 L capacity endowed with gas inlet and outlet valves. The Ohmic contacts were established on the surface of the sensing element of area 10 mm × 10 mm using zero resistance copper wire and highly conducting silver paste.^40,41 The two electrodes were separated by a 5 mm distance. Details of the complete sensing set up and the precise geometry of the electrical contacts can be found in our previous work.⁴¹ The resistances of the sensing elements were continuously monitored by a two probe method using a high resistance electrometer (Keithley 6517A). The desired concentration of acetaldehyde was injected using a microliter syringe through a septum provision in the sealed chamber. The concentration of the target vapour⁴² in the testing chamber was fixed using eqn (1).where, δ is the density of acetaldehyde, V_r is the volume of acetaldehyde injected, R is the universal gas constant, T is the absolute temperature, M is the molecular weight, P_b is the pressure inside the chamber and V_b is the volume of the chamber. The sensor response (S) was calculated using the relation R_a/R_g, where R_a and R_g are the resistances of the sensing elements in air and in the target gas. Since the as-deposited films showed an unstable base resistance in an air atmosphere, only annealed films were considered for the sensing studies. All the sensing measurements were recorded at room temperature.

3. Results and discussion

3.1 Morphological studies

Fig. 1 shows the formation of ZnO thin films at the substrate temperature of 523 K as well as at the annealing temperatures of 523, 623 and 723 K. This scheme clearly depicts the shape transformation of as-deposited ZnO nanoplatelets (Fig. 2a) into diverse nanostructures such as tiny platelets (Fig. 2b), branched nanorods (Fig. 2c) and thicker nanoplatelets (Fig. 2d) as a function of the annealing temperature. The as-deposited ZnO nanoplatelets were randomly connected to each other and the thickness of these nanoplatelets were found to be in the range 110–130 nm. Some irregular particles were also observed along with the nanoplatelets and these could be attributed to the incomplete decomposition of chloride salt.³⁹ Hence, the annealing process was carried out in order to prepare ZnO nanostructures without precursor residues. While annealing the as-deposited nanoplatelets at 523 K, tiny nanoplatelets were formed. After increasing the annealing temperature to 623 K, the surface morphology showed the co-existence of 2D nanoplatelets and 1D nanorods. In this structure, the roots of the nanorods were formed at the surface of the ZnO nanoplatelets and grown outward with lengths ranging from 300 to 330 nm, filling the gaps between the nanoplatelets. At 723 K, the nanoplatelets became denser and thicker with shorter nanorods than those formed at 623 K.

Fig. 1 Synthesis strategy to hierarchical ZnO nanostructures.

Fig. 2 FESEM images of (a) larger nanoplatelets, (b) tiny nanoplatelets, (c) branched nanorods and (d) thicker nanoplatelets.

3.2 Formation mechanism

Based on the observed results, a growth mechanism for the ZnO nanoarchitectures has been proposed. Generally, structure directing or capping agents such as sodium dodecyl sulfate, cetyltrimethylammonium bromide, thiourea and thiocarbamide have been used along with precursor salts (zinc acetate, zinc nitrate and zinc chloride) to obtain various ZnO nanostructures.^43–45 Spherical morphologies were most commonly achieved when using zinc acetate or nitrate precursors without any structure directing agents. Zinc chloride however has an inherent property of forming hexagonal rod/platelet like structures without any capping agents. Previously, Smith et al. have extensively investigated this unique property and concluded that the byproduct formed during the pyrolysis process itself acts as a structure directing agent. At the pyrolysis region, the byproduct, namely HCl, acts as a capping or structure directing agent and controls the growth of the nuclei, resulting in various morphologies.^39,46–48 The possible reaction mechanism for the formation of ZnO from zinc chloride as the precursor is given in eqn (2).

ZnO formation originates from the reaction between Zn²⁺ and OH⁻ ions. As the precursor solution is sprayed onto the pre-heated substrate, ZnO forms from the decomposition of zinc hydroxide. Firstly, thinner nanoplatelets are grown on glass substrates at the substrate temperature of 523 K with the assistance of the byproducts formed during decomposition. The as-deposited film resembles the simonkolleite (Zn₅(OH)₈Cl₂·2H₂O) crystal structure; it was then subjected to annealing to obtain a ZnO-only film. In the early stages of the annealing process (523 K), thinner nanoplatelets broke up into smaller units and acted as nucleation centers for the growth of nanorods. This growth occurred heterogeneously on both sides of the tiny nanoplatelets (ESI Fig. S1†). The nanorod growth rate was however limited due to a lack of thermal energy. Consequently, at an intermediate temperature (623 K) all of the tiny nanoplatelets were converted into smaller grains. These smaller grains acted as seeds for the growth of ZnO nanorods in a lateral direction. At a higher temperature (723 K), due to a coalescence process, smaller grains combined together to form larger grains. Hence, the growth of larger nanorods originated from fusing individual nanorods together.

3.3 Structural studies

XRD measurements (Fig. 3) were carried out to determine the crystal plane orientations and crystallite sizes of the as-deposited and annealed ZnO nanoplatelets. Patterns of the as-deposited ZnO thin films showed relatively low intensity ZnO peaks and several other peaks corresponding to ZnO, Zn and ZnOH. This might be due to the incomplete decomposition of the precursor salt whose crystal structure resembles that of simonkolleite (Zn₅(OH)₈Cl₂·2H₂O) (JCPDS 07-0155).⁴⁹ The annealed films showed a dominant (002) crystal plane orientation which indicated that the films were preferentially oriented along the c axis, well matched with JCPDS card no: 36-1451. The average crystallite sizes of the ZnO thin films were estimated with reference to the (002) plane using Scherrer’s formula,²¹ and they were found to be 33, 36 and 38 nm for tiny nanoplatelets, branched nanorods and thicker nanorods, respectively.

Fig. 3 XRD patterns of the hierarchical ZnO nanostructures.

3.4 Optical studies

Fig. 4 shows the optical absorbance spectra of the films in the wavelength range of 350 to 800 nm. The major differences in the absorbance values of the three different nanoarchitectures can be primarily attributed to the scattering of incident light at the grain boundaries. In the case of the tiny nanoplatelets, the presence of a higher surface coverage with lesser voids both captures and traps the incident light to a maximum extent. The absorption was further increased by a multiple reflection effect when light interacts with the tiny nanoplatelets and branched nanorods, which in-turn extends the spatial and temporal light absorption.⁵⁰ However, in the case of the thinner and thicker nanoplatelets the lower surface coverage and large number of open voids facilitate less absorption of incident light. In other words, the surface filling factor and branched architectures play an important role in determining the light absorption through multiple scattering.

Fig. 4 Optical absorbance spectra of the hierarchical ZnO nanostructures.

3.5 Wettability studies

The contact angle (CA) between water and the film surface determines the wettability nature of the surface, i.e. whether it is hydrophilic (<90°) or hydrophobic (>90°).⁵¹ Fig. 5 shows the water wettability properties of the as-deposited and annealed ZnO thin films. The as-deposited film exhibited a CA of about 80°. Except for this case, all other annealed ZnO thin films showed hydrophobic natures. The obtained results are consistent with the morphological studies. The as-deposited film surface showed hexagonal shaped ZnO platelets with smooth surfaces (relatively low roughness), which afford the hydrophilic nature. Once the film was subjected to annealing treatment, the change in morphology, namely to tiny nanoplatelets, branched nanorods and thicker nanoplatelets, resulted in increased surface roughness. Films with higher surface roughness trap more air pockets beneath the water drop and hence the contact angle increases.⁵² The film annealed at 623 K showed a quite different morphology and displayed a high CA of about 134°. Thus, these results prove that annealing can effectively change the wettability nature of the thin films.

Fig. 5 Wettability nature of the various ZnO nanostructures.

3.6 Sensing studies

3.6.1 Selectivity. Selectivity of the sensing element can be defined as the capability of the sensor to give a response to specific gases over others. In fact, a highly selective sensor can be used to detect a specific gas/vapour when it is exposed to a multicomponent gas/vapour environment. Therefore, the responses of the ZnO nanostructures were tested in the presence of 100 ppm of various vapours, namely ammonia, acetone, ethanol, methanol, formaldehyde, toluene and acetaldehyde. Fig. 6a and b summarize the sensing responses of the three sensing elements toward the test vapours. Surprisingly, all three different nanostructures showed a high response towards acetaldehyde vapour. Meanwhile, their responses to other vapours did not exceed 13. The branched nanorods showed a remarkable response which was around 16 and 4 times higher than those of tiny and thicker nanoplatelets, respectively. Noticeably, these sensor elements showed a considerable response to formaldehyde. The selectivity towards acetaldehyde might be due to its lesser dissociation energy (364 kJ mol⁻¹) compared with other vapours, namely ammonia (435 kJ mol⁻¹), acetone (393 kJ mol⁻¹), ethanol (436 kJ mol⁻¹), formaldehyde (364 kJ mol⁻¹) and toluene (368 kJ mol⁻¹). Even though the dissociation energies of acetaldehyde and formaldehyde are the same, the number of electrons released during the sorption process made acetaldehyde more sensitive to detection than formaldehyde.³⁹

Fig. 6 (a) Selectivity of the sensing elements and (b) enlarged view of the low sensing response region.

3.6.2 Transient and response studies. Herein, we studied the room temperature sensing responses of the three different morphologies toward various concentrations of acetaldehyde. The transient response–recovery characteristics are shown in Fig. 7a. A fall in resistance within a few seconds is observed when injecting the reducing type target vapour inside the chamber, indicating the n-type semiconductor behaviour of the sensing element.⁵³ All the films recovered to their baseline values once the target vapour was exhausted. It is also obvious that as the concentration of the target vapour increased, the response amplitude also increased. Some fluctuations in the readings were observed and these might be due to environmental turbulence around the sensor. The lowest detection limit varied from 50 ppm, 20 ppm and 5 ppm for tiny nanoplatelets, branched nanorods and thicker nanoplatelets, respectively.

Fig. 7 (a) Transient resistance responses and (b) response trends for the various ZnO nanostructures.

On the basis of the data collected upon sequential exposure to acetaldehyde at different concentrations, the response curves are shown in Fig. 7b. The response values for 100 ppm of acetaldehyde were 36, 587 and 140 for tiny nanoplatelets, branched nanorods and thicker nanoplatelets, respectively. The obtained sensing response for the branched nanorods was about 16 and 4 times higher than those of the tiny and thicker nanoplatelets, respectively. For the tiny nanoplatelets, response values of 4.1, 36, 72, 125.8 and 128.5 were observed for 50, 100, 250, 500 and 1000 ppm of acetaldehyde, respectively. The response becomes more or less saturated at over 500 ppm in the case of the tiny nanoplatelets. The sensing responses of the branched nanorods were 2.8, 3.6, 66, 92, 587, 3083 and 7872 for 20, 30, 50, 75, 100, 250 and 500 ppm of acetaldehyde respectively. Interestingly, a change of 4 orders of magnitude (4.7 × 10⁶ Ω for 500 ppm) was observed with respect to the baseline resistance (3.7 × 10¹⁰ Ω). It is apparent that the branched nanorod sensors showed a wide detection range of 20 ppm to 1000 ppm. Even at 5 ppm, the thicker nanoplatelets showed a response of 4.3, but the sensor was saturated at 100 ppm of acetaldehyde. In contrast, no response was recorded at lower concentrations for the tiny nanoplatelets and branched nanorods.

3.6.3 Response/recovery times. Response and recovery times are very crucial parameters for a gas sensor. The response and recovery times were calculated from time vs. resistance plots, and are defined as the time taken to attain 90% and 10% of target gas resistance from the baseline resistance, respectively.⁴¹ ESI Fig. S2† shows the typical response recovery times of the three different sensing elements. As the concentration was increased, the response time was reduced to a few seconds due to the simultaneous interaction of a higher number of target molecules with the sensing element. Also, the recovery time of the sensing element constantly increased with increasing concentration due to slow desorption of target gas molecules from the surface. Response/recovery times of 116/44 s (tiny nanoplatelets), 112/54 s (branched nanorods) and 97/186 s (thicker nanoplatelets) were observed for 100 ppm of acetaldehyde. Furthermore, branched nanorods showed a very fast recovery of 7 s for 20 ppm of acetaldehyde. Also, for the branched nanorods, the reproducibility was checked five times towards 100 ppm of acetaldehyde, and a fair reproducibility behaviour was observed. The reason behind the excellent selectivities, high sensitivities and quick response and recovery times for the hierarchical nanostructures is explained in the following section.

3.6.4 Sensing mechanism. ZnO is one of the highly versatile sensing materials used to detect various gases. It is well known that the surface morphology plays a significant role in determining the sensor response of a semiconductor material. Gas sensors are surface-controlled types, where the dimensions, surface states and quantities of adsorbed oxygen molecules influence the sensing performance significantly. Moreover, the electrical properties of one-dimensional (1D) ZnO nanostructured arrays are exceedingly sensitive to adsorbed species, because of their high surface-to-volume ratio, which allows the surface atoms to have more opportunities to participate in surface reactions. Compared with 1D & 2D nanostructures, 3D nanostructures facilitate enhancement of the properties of the sensing material.^54–57 In particular, branched/networked types of morphologies are often associated with enhanced gas sensing properties.^34–38 The gas sensing mechanisms of most metal oxide semiconductor sensors can be recognized as involving changes in their electrical conductivities in the presence and absence of target gases.

When the hierarchical nanostructures have been continuously exposed to the ambient atmosphere, ambient oxygen is adsorbed on the ZnO surface, thereby resulting in the formation of a depletion layer and ultimately increasing the resistance. According to the commonly accepted Barsan and Weimar conduction model,⁵⁸ the extent of domination of oxygen species adsorption depends on temperature. Hence, at room temperature O₂⁻ adsorption is dominant and can be written as in eqn 3.

O_{2(atmosphere)} + e⁻_{(ZnO surface)} → O⁻_{2(ZnO surface)} (3)

Since acetaldehyde is a reducing gas, it gets oxidized into CO₂ and H₂O while interacting with adsorbed oxygen on the ZnO surface. Hence, the trapped electrons are put back into the conduction band (i.e. thinning of the space charge region, thus decreasing the potential barrier) of the sensing element, leading to an increase in conductivity. The proposed acetaldehyde sensing mechanism is given in eqn (4).

2CH₃CHO + 5O⁻_{2(ZnO surface)} → 4CO₂↑ + 4H₂O↑ + 5e⁻_{(ZnO surface)} (4)

In order to validate the proposed sensing mechanism, a simple lime water test was carried out, as described in detail in our previous work.³⁹

In summary, the branched nanorod morphology showed the maximum response. Fig. 8a–d show enlarged views of the branched nanorod morphology. It is clearly observed that branched nanorods root from the backbones of tiny grains and grow to around 300 nm in length (Fig. 8a). The branched nanorod sensor provides more pathways to electron exchange during gas interaction, owing to stems made up of tiny grains ranging from 40 to 60 nm (Fig. 8b) and networked nanorods around 300 nm in length (Fig. 8c). In addition, the tips of the nanorods show a hexagonal morphology (Fig. 8d). The large amount of open space between the nanorods is also highly beneficial for the diffusion of gases, which may have enhanced the sensing performance. At higher concentrations, high sensing responses of 3083 and 7872 were observed for 250 and 500 ppm. These may be owing to the switching from a highly non-activated state into an activated state for the entire hierarchical branched nanowires at higher concentrations. Generally, the crystallinity of a material is often associated with its sensing performance. When the crystallinity increases, the intrinsic resistance of the material decreases.^59,60 The obtained resistance values for the three different nanoarchitectures (tiny nanoplatelets, branched nanorods and thicker nanoplatelets) are 1.8 × 10¹¹, 3.7 × 10¹⁰ and 1.3 × 10¹¹ Ω. Among them, the branched nanorods, with a higher crystallinity, showed a lower resistance, indicating the significantly improved electron transport between the stems and rods, and consequently resulting in a high sensing performance.

Fig. 8 High magnification FESEM images of (a) branched nanorods, (b) stems of branched nanorods, (c) a complete view of the nanorods, (d) the tips of the hexagonal nanorods and (e) thicker nanoplatelets.

Though the tiny and thicker nanoplatelets possessed more open spaces than the branched nanorods, the presence of both the large active area (rods & stems) and open space made the branched nanorods more sensitive than the other structures. The calculated responses were found to be 36 and 139.7 for 100 ppm of acetaldehyde for the tiny and thicker nanoplatelets, respectively. To understand the enhanced response of the thicker nanoplatelets, the morphology was further examined and an enlarged view is shown in Fig. 8e. The image elucidates nanorods with larger diameters laterally grown from the stems, which may be the reason for the better response of the thicker nanoplatelets in comparison to the tiny ones.

To demonstrate the sensing mechanism, the schematics of the three different morphologies are depicted in Fig. 9. J_Total is referred to as the number of junctions in the film morphology. The tiny nanoplatelets showed only one kind of junction, namely intergranular contacts in the stems (J₁). The thicker nanoplatelets showed an additional type of junction between the intergranular contacts and the nanorods (J₂). In contrast, the branched nanorods posses four different kinds of junctions, namely intergranular contacts in the stems (J₁), intergranular-nanorod (J₂), nanorod-nanorod (J₃) and intergranular contacts in the nanorods (J₄). The surface depletion regions of individual nanorods and high potential barriers created in the junctions facilitate a 2-fold superior sensitivity compared with the other nanostructures. In the literature, Park et al. systematically studied the densities of the junctions in networked SnO₂ nanowires and their sensing properties. They reported that the sensing performance was outstanding for films with morphologies of high density junctions.⁶¹ Furthermore, the effect of humidity on the sensing responses of the ZnO nanoarchitectures towards 100 ppm of acetaldehyde were studied and are shown in Fig. 10. At lower humidity levels, due to a lesser number of adsorbed OH ions on the ZnO surface, the availability of more active sites for gas–solid interactions resulted in a better response. At higher humidity levels, as expected, the response was reduced due to the inhibition of active sites on the ZnO surface by the OH ions.^40,42 The co-efficients of variation for the sensing responses of the tiny nanoplatelets, branched nanorods and thicker nanoplatelets were found to be 44%, 6% and 14%, respectively, with reference to various humidity levels. These results can be correlated with the measured contact angle values. The obtained contact angles are in the following order: branched nanorods (CA = 134°) ≤ thicker nanoplatelets (CA = 119°) ≤ tiny nanoplatelets (CA = 91°). Since nanostructures with large contact angles are more hydrophobic, they do not allow for OH ions to be adsorbed onto their surface. Hence, the film with the branched nanorod architecture showed a lower co-efficient of variation for its sensing response at various humidity levels.

Fig. 9 Schematic diagram showing the densities of junctions in the hierarchical nanostructures. Potential barrier built up in (J1) intergranular contacts in stems, (J2) between inter-granular contacts and nanorods, (J3) nanorod-nanorod and (J4) intergranular contacts in nanorods.

Fig. 10 Response towards 100 ppm of acetaldehyde with respect to various humidity levels.

With respect to the previous reports on ZnO acetaldehyde sensors, Calestani et al. achieved a response of 47.5 for 50 ppm using ZnO tetrapods at the operating temperature of 400 °C.²⁴ Giberti et al. studied acetaldehyde detection using ZnO powders prepared by a sol–gel technique and reported the sensor performance, but this was limited to a 10 ppm concentration at the operating temperature of 450 °C.⁶² Similarly, Rai et al. prepared ZnO nanorods using a microwave assisted hydrothermal method and the highest response achieved was 5.30 for 250 ppm at 400 °C.⁶³ Recently, Zhang et al. discussed the acetaldehyde sensing performance of ZnO nanosheets and reported a response of ∼80 for 1 ppm at 220 °C.⁶⁴ All of these reports study the performances of the ZnO sensors at elevated operating temperatures. In contrast, in the present work we highlight the advantages of room temperature sensing, a wide detection range, a simple fabrication procedure, and cost effective synthesis route to ZnO sensing elements. Therefore, ZnO nanostructures with multiple junctions have been identified as promising candidates to detect acetaldehyde at room temperature.

4. Conclusion

In conclusion, through annealing treatment of spray deposited ZnO thin films, various hierarchical nanostructures, namely tiny nanoplatelets, branched nanorods and thicker nanoplatelets, were formed. The first implementation of branched ZnO nanorods as a room temperature acetaldehyde sensor has been successfully accomplished. The branched nanorod structure provided a response of 587 for 100 ppm of acetaldehyde with response and recovery times of 112 and 54 s, respectively. Combined morphologies, such as nanorods rooting from nanoplatelets, were able to generate more active centers for interactions with gas molecules and hence increased the sensing response. Under similar test conditions, tiny and thicker nanoplatelets showed low responses with narrow ranges of detection. These results demonstrated that the hierarchical ZnO nanostructures can be used to fabricate a low power, cost effective acetaldehyde sensor with a better figure of merit.

Acknowledgements

The authors wish to express their sincere thanks to the Department of Science & Technology, New Delhi, India for financial support (Project ID: INT/SWD/VINN/P-04/2011 & SR/FST/ETI-284/2011(C)). They also wish to acknowledge SASTRA University, Thanjavur for extending infrastructure support to carry out this work.

References

1. K. Mitsubayashi, H. Amagai, H. Watanabe and Y. Nakayama, Sens. Actuators, B, 2003, 95, 303–308.
2. E. Przyk, A. Świtaj-Zawadka, P. Konieczka, J. Szczygelska-Tao, J. F. Biernat and J. Namieśnik, Anal. Chim. Acta, 2003, 488, 89–96.
3. M. E. Ghica, R. Pauliukaite, N. Marchand, E. Devic and C. M. A. Brett, Anal. Chim. Acta, 2007, 591, 80–86.
4. T. Miyake and T. Shibamoto, J. Chromatogr. B: Biomed. Sci. Appl., 1998, 719, 213–216.
5. A. Avramescu, T. Noguer, M. Avramescu and J.-L. Marty, Anal. Chim. Acta, 2002, 458, 203–213.
6. A. a. Mohamed, A. T. Mubarak, Z. M. H. Marestani and K. F. Fawy, Talanta, 2008, 74, 578–585.
7. W. Ma and W. R. Klemm, Alcohol, 1997, 14, 469–472.
8. N. Ramgir, N. Datta, M. Kaur, S. Kailasaganapathi, A. K. Debnath, D. K. Aswal and S. K. Gupta, Colloids Surf., A, 2013, 439, 101–116.
9. Y.-B. Hahn, R. Ahmad and N. Tripathy, Chem. Commun., 2012, 48, 10369–10385.
10. X. Liu, S. Cheng, H. Liu, S. Hu, D. Zhang and H. Ning, Sensors, 2012, 12, 9635–9665.
11. R. Boppella, P. Manjula, S. Arunkumar and S. V. Manorama, Chem. Sens., 2014, 4, 1–22.
12. A. Wei, L. Pan and W. Huang, Mater. Sci. Eng., B, 2011, 176, 1409–1421.
13. J. Kennedy, P. P. Murmu, E. Manikandan and S. Y. Lee, J. Alloys Compd., 2014, 616, 614–617.
14. F. Fang, J. Futter, A. Markwitz and J. V. Kennedy, Mater. Sci. Forum, 2012, 700, 150–153.
15. C. Zhao, G. Zhang, W. Han, J. Fu, Y. He, Z. Zhang and E. Xie, CrystEngComm, 2013, 15, 6491.
16. S. Wang, Z. Li, P. Wang, C. Xiao, R. Zhao, B. Xiao, T. Yang and M. Zhang, CrystEngComm, 2014, 16, 5716.
17. F. Fang, J. Futter, A. Markwitz and J. Kennedy, Nanotechnology, 2009, 20, 245502.
18. J. Huang and Q. Wan, Sensors, 2009, 9, 9903–9924.
19. J. H. Lee, Sens. Actuators, B, 2009, 140, 319–336.
20. G.-C. Yi, C. Wang and W. Il Park, Semicond. Sci. Technol., 2005, 20, S22–S34.
21. K. Sivalingam, P. Shankar, G. K. Mani and J. B. B. Rayappan, Mater. Lett., 2014, 134, 47–50.
22. E. Gunasekaran, P. Shankar, G. K. Mani and J. B. B. Rayappan, Eur. Phys. J.: Appl. Phys., 2014, 67, 20301.
23. A. Gurlo, Nanoscale, 2011, 3, 154–165.
24. D. Calestani, R. Mosca, M. Zanichelli, M. Villani and A. Zappettini, J. Mater. Chem., 2011, 21, 15532.
25. Z. Lou, F. Li, J. Deng, L. Wang and T. Zhang, ACS Appl. Mater. Interfaces, 2013, 5, 12310–12316.
26. Y. Liu, Y. Jiao, Z. Zhang, F. Qu, A. Umar and X. Wu, ACS Appl. Mater. Interfaces, 2014, 6, 2174–2184.
27. S. Sarkar and D. Basak, Sens. Actuators, B, 2013, 176, 374–378.
28. S. An, S. Park, H. Ko, C. Jin, W. I. Lee and C. Lee, Thin Solid Films, 2013, 547, 241–245.
29. J. Liu, W. Wu, S. Bai and Y. Qin, ACS Appl. Mater. Interfaces, 2011, 3, 4197–4200.
30. P. S. Patil, Mater. Chem. Phys., 1999, 59, 185–198.
31. J. B. Mooney and S. B. Radding, Annu. Rev. Mater. Sci., 1982, 12, 81–101.
32. D. Perednis and L. J. Gauckler, J. Electroceram., 2005, 14, 103–111.
33. J. Y. Lao, J. G. Wen and Z. F. Ren, Nano Lett., 2002, 2, 1287–1291.
34. H.-U. Lee, K. Ahn, S.-J. Lee, J.-P. Kim, H.-G. Kim, S.-Y. Jeong and C.-R. Cho, Appl. Phys. Lett., 2011, 98, 193114.
35. Y. Zhang, J. Xu, Q. Xiang, H. Li, Q. Pan and P. Xu, J. Phys. Chem. C, 2009, 113, 3430–3435.
36. J. Zhang, S. Wang, M. Xu, Y. Wang, B. Zhu, S. Zhang, W. Huang and S. Wu, Cryst. Growth Des., 2009, 9, 3532–3537.
37. M.-W. Ahn, K.-S. Park, J.-H. Heo, D.-W. Kim, K. J. Choi and J.-G. Park, Sens. Actuators, B, 2009, 138, 168–173.
38. S.-W. Choi, A. Katoch, G.-J. Sun and S. S. Kim, Sens. Actuators, B, 2013, 181, 787–794.
39. G. K. Mani and J. B. B. Rayappan, Sens. Actuators, B, 2014, 198, 125–133.
40. G. K. Mani and J. B. B. Rayappan, Sens. Actuators, B, 2013, 183, 459–466.
41. G. K. Mani and J. B. B. Rayappan, Appl. Surf. Sci., 2014, 311, 405–412.
42. R. Pandeeswari and B. G. Jeyaprakash, Sens. Actuators, B, 2014, 195, 206–214.
43. M. Duta, D. Perniu and a. Duta, Appl. Surf. Sci., 2014, 306, 80–88.
44. T. Dedova, O. Volobujeva, J. Klauson, A. Mere and M. Krunks, Nanoscale Res. Lett., 2007, 2, 391–396.
45. E. Sonmez, T. Karacali, A. E. Ekinci and M. Ertugrul, IEEE Trans. Nanotechnol., 2011, 10, 532–536.
46. A. Smith, Thin Solid Films, 2000, 376, 47–55.
47. Á. Smith and R. Rodriguez-clemente, Thin Solid Films, 1999, 345, 192–196.
48. A. Smith, J. Laurent, D. S. Smith, J. Bonnet and R. Rodriguez, Thin Solid Films, 1995, 266, 20–30.
49. J. Sithole, B. D. Ngom, S. Khamlich, E. Manikanadan, N. Manyala, M. L. Saboungi, D. Knoessen, R. Nemutudi and M. Maaza, Appl. Surf. Sci., 2012, 258, 7839–7843.
50. X. Sun, Q. Li, J. Jiang and Y. Mao, Nanoscale, 2014, 6, 8769–8780.
51. Y. Yuan and T. R. Lee, Surface Science Techniques, Springer Berlin Heidelberg, Berlin, Heidelberg, 2013, vol. 51.
52. Y. Li, M. Zheng, L. Ma, M. Zhong and W. Shen, Inorg. Chem., 2008, 47, 3140–3143.
53. R. L. Vander Wal, G. W. Hunter, J. C. Xu, M. J. Kulis, G. M. Berger and T. M. Ticich, Sens. Actuators, B, 2009, 138, 113–119.
54. M. R. Alenezi, S. J. Henley, N. G. Emerson and S. R. P. Silva, Nanoscale, 2014, 6, 235–247.
55. A. Kolmakov and M. Moskovits, Annu. Rev. Mater. Res., 2004, 34, 151–180.
56. C. Soldano, E. Comini, C. Baratto, M. Ferroni, G. Faglia and G. Sberveglieri, J. Am. Ceram. Soc., 2012, 95, 831–850.
57. E. Comini, C. Baratto, G. Faglia, M. Ferroni, a. Vomiero and G. Sberveglieri, Prog. Mater. Sci., 2009, 54, 1–67.
58. N. Barsan and U. Weimar, J. Electroceram., 2001, 7, 143–167.
59. H. Xu, X. Liu, D. Cui, M. Li and M. Jiang, Sens. Actuators, B, 2006, 114, 301–307.
60. Y. Lü, W. Zhan, Y. He, Y. Wang, X. Kong, Q. Kuang, Z. Xie and L. Zheng, ACS Appl. Mater. Interfaces, 2014, 6, 4186–4195.
61. J. Park, S. Choi and S. Kim, J. Phys. Chem. C, 2011, 115, 12774–12781.
62. A. Giberti, M. C. C. Carotta, B. Fabbri, S. Gherardi, V. Guidi and C. Malagù, Sens. Actuators, B, 2012, 174, 402–405.
63. P. Rai, H.-M. Song, Y.-S. Kim, M.-K. Song, P.-R. Oh, J.-M. Yoon and Y.-T. Yu, Mater. Lett., 2012, 68, 90–93.
64. S.-L. Zhang, J.-O. Lim, J.-S. Huh, J.-S. Noh and W. Lee, Curr. Appl. Phys., 2013, 13, S156–S161.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra09652e

---