1 Introduction

The properties of nanomaterials do not depend only on their size but also on their morphology (shape). In fact, the morphology plays a crucial role in tuning various properties and activities of nanomaterials [1,2,3,4]. Recently, we defined the morphology effects of nanomaterials as “a variation in the physical and/or chemical properties of a nanostructure or its activities, taking place due to changes in its morphology” [2]. Lucas et al. [5] found that the elastic modulus of ZnO nanobelt depending strongly on the aspect ratio, which was attributed to the existence of stacking faults in the structure of nanobelts growing along specific directions. Yang et al. [3] suggested that tuning the optical properties of nano-sized ZnO can be achieved by varying their morphology. Vinod et al. [6] and Lv et al. [7] ascribed the red shift in band gap energy to the flower morphology, microstructure of ZnO, and/or Burstein–Moss effect. Yuan et al. [8] reported that the ZnO morphology had a substantial effect on the electrochemical properties of the anodes. The biological activity of the nanomaterials was also observed to change with their morphology. Pang et al. [9] synthesized different morphologies of Cu2O structures. They found that when Cu2O changing from cubic morphology to octahedral morphology, the antibacterial activity dramatically changed. The results were attributed to the fact that Cu2O crystals with different shapes have different surface planes and different surface planes would have different desorption and adsorption capabilities towards the bacteria, leading to different antibacterial activities. Pal et al. [10] observed that compared to rod- and sphere-shape nano-Ag, triangular-shape Ag nanoparticles exhibited stronger antibacterial property. Tong et al. [4] found that nano-TiO2 morphology plays a critical role in controlling the phototoxicity to bacteria. Under simulated solar irradiation, TiO2 nanotubes and nanosheets were observed less phototoxic than their rod- and sphere- shape counterparts. Hamilton et al. [11] found that the toxicity to both mice and alveolar macrophages induced by longer TiO2 nanobelts was higher than their shorter, spherical counterparts.

Many attempts have been carried out to understand the mechanism behind the morphology effect. For example, Li et al. [12] synthesized different morphologies of ZnO; rods, rings, disks and screw caps, by hydrothermal method. They found that the more polar planes, the more enhanced catalytic activity. They attributed this positive relationship to the fact that VO defects can be easily generated in the polar planes. Therefore, ZnO sample with more polar planes contains a higher concentration of VO and thereby a higher catalytic activity. Very recently, we studied the effects of morphology of ZnO nanostructures on their optical and photocatalytic properties [2]. In this report, two different morphologies of ZnO nanoflowers were synthesized using hydrothermal method. We observed that the optical and photocatalytic properties of the prepared ZnO nanoflowers changed with their morphology. The mechanism behind the morphology effect was attributed to a change of native defect concentration with morphology. However, in this study, two different additives were used to obtain two different ZnO nanoflowers. Therefore, it might be thought that using different additives during the synthesis of those morphologies might play a role in changing their properties.

In the present work, we investigate the correlation among morphology, VO and properties of ZnO nanoflowers. Three different morphologies of ZnO nanoflowers have been synthesized via hydrothermal method in the presence of the same additive namely diethylamine (DEA). The morphology was tuned by adjusting the concentration of DEA. The prepared ZnO flowers were found having considerable VO in their structure and the concentration of VO was found to be changed with the morphology. The optical, photocatalytic, and antibacterial properties were also observed changing with the morphology. Our study delineates the relationships among the morphology, VO and the properties of ZnO nanoflowers.

2 Experimental section

2.1 Materials

Zn (NO3)2·6H2O (Alfa Aesar, 99%), NaOH (Alfa Aeaser, 98%), ethanol (Sigma, 99%) and DEA (Sigma Aldrich, 99%) were used as received. Deionized water (DIW), with a resistivity of 18.2 MΩ cm, was used in the experiment.

2.2 Synthesis procedure

Zn (NO3)2·6H2O and NaOH were dissolved in DIW in 1:10 ratio and mixed with 10 ml ethanol to form 50 ml. Then, 1.5, 1 and 0.5 ml of DEA were added to 15 ml of the above solution separately. Then the resultant solutions were sonicated for 40 min, before being transferred to Teflon-lined autoclaves. The hydrothermal processing was conducted for 1 h at 180 °C. Afterwards, obtained precipitates were thoroughly washed with DIW and ethanol and collected by centrifugation. The obtained products were dried at 80 °C for 12 h in a hot air oven.

2.3 Analytical method

The samples were analyzed by X-ray diffraction (Rigaku smart Lab-II, Cu Kα radiation, λ = 1.5414 Å), scanning electron microscopy (SEM, Hitachi S-3400N), and X-ray photoelectron spectroscopy (XPS, ESCALAB 250Xi, Thermo Scientific). The room temperature photoluminescence (PL) spectra of the samples ZF1, ZF2, and ZF3 were investigated utilizing a Hitachi F-2700, spectrophotometer with an excitation wavelength of 325 nm. Ultraviolet–Visible diffuse reflectance spectroscopy (DRS) was recorded on a Hitachi U-3010.

2.4 Estimation of photocatalytic activity

Photocatalytic activities of ZnO nanoflowers samples were estimated by analyzing the photodegradation of phenol under sunlight irradiation. 0.1 g of ZnO nanoflowers was dispersed in separate aqueous phenol solution (100 ml, 10 mg/l). Prior to reaction the suspension was stirred for 30 min in dark to attain an adsorption–desorption equilibrium. Aliquots of samples were collected at certain time intervals, centrifuged to remove the photocatalyst and analyzed by a Beckman Coulter (DU 730) UV–Vis spectrophotometry.

2.5 Electrodes preparation

The ZnO nanoflowers electrodes were fabricated using an electrophoretic deposition method. 10 mg of ZnO powder and 5 mg of iodine were dispersed in 20 ml of acetone. The electrophoretic deposition process was carried out by applying a voltage of 15 V between two parallel FTO glass substrates with a 1 cm distance for 5 min. Finally, the FTO/ZnO nanoflowers electrodes were heated to 450 °C for 30 min under a nitrogen atmosphere.

2.6 Electrochemical studies

To measure the photocurrents and electrochemical impedance spectroscopy (EIS), an electrochemical analyzer (CHI-660E, USA) was used. A standard three-electrode cell was used with FTO/ZnO nanoflowers electrodes as working electrodes, a platinum wire and a standard calomel electrode (SCE) as a counter and reference electrodes, respectively. Simulated sunlight irradiation was obtained from a 400 W Xe lamp (100 mW/cm2). The electrolyte solution was a 0.1 M Na2SO4 aqueous solution. The photoresponses of the photocatalysts as light on and off were measured at 0.0 V.

2.7 Disc diffusion method

Disc diffusion method for antibacterial susceptibility testing was performed according to Lakshmeesha et al. [13] with minor modification to assess the existence of antibacterial activities of nanoparticles. Fifteen milliliters of the molten Muller Hinton agar (40 °C) was poured into sterile petridishes. Overnight grown test bacterial cultures were attuned to 0.5 McFarland standard and used to lawn on Muller Hinton agar plates by using a sterile swab. Sterial discs impregnated with a ZnO nanoflowers (20 mg/ml) concentrations were placed on the Mueller Hinton agar surface, ceftriaxone disc (30 µg) was used as positive control. The plates were kept in incubator at 35 °C and read at 24 h. Each experiment was repeated in triplicate. Zone diameters in mm were measured at the point where bacterial growth decreased abruptly.

2.8 Statistical analysis

The Antibacterial assay was carried out in triplicate and the data were expressed as mean ± Standard error (SE) (n = 3). Statistical analysis was done using two-way ANOVA using GraphPad Prism version 3.00 for Windows (GraphPad Software, San Diego, Calif.). The difference was considered significant when P < 0.05.

3 Results and discussion

Figure 1 depicts the XRD patterns of as-prepared ZnO nanoflowers. All the XRD diffraction peaks are well matched with the typical hexagonal wurtzite phase of ZnO crystals with space group P63mc (JCPDS: 36-1451). No other peaks corresponding to other crystalline impurities were observed in the XRD patterns indicating the purity of the prepared ZnO nanoflowers.

Fig. 1
figure 1

XRD patterns of the samples (a) ZF1, (b) ZF2, and (c) ZF3

Full size image

To study the morphology of prepared ZnO nanoflowers, SEM analysis was employed. Figure 2 depicts the SEM images of the ZnO nanoflowers synthesized via hydrothermal route at 180 °C for 1 h using different concentrations of DEA, (1.5 ml: sample ZF1, 1 ml: sample ZF2 and 0.5 ml: sample ZF3). Using 1.5 ml of DEA resulted in the formation of ZnO nanoflowers consisting of a large number of nanorods with small diameters (Fig. 2a, d). Using 1 ml of DEA resulted in decreasing the number of nanorods and increasing their diameters (Fig. 2b, e). While 0.5 ml of DEA resulted in the formation of the smallest number of nanorods with the largest diameters (Fig. 2c, f). That is to say that increasing the concentration of DEA led to an increase in the number of nanorods and a decrease in their diameters.

Fig. 2
figure 2

SEM images of the samples a and d ZF1, b and e ZF2, and c and f ZF3

Full size image

The above results denote obviously that concentration of additives, in hydrothermal synthesis of ZnO, is potential in inducing the anisotropic growth of the crystal faces resulting in tuning the morphology of the products. Indeed, the variation of DEA concentration resulted in a variation in the degree of supersaturation. The variation in the degree of supersaturation led to a change in the nucleation and crystal growth rates and thereby changed the morphology of ZnO nanoflowers [2, 14, 15]. Careful observation on XRD patterns demonstrates that intensity ratio of (002) to (101) peaks I[002/101] of the samples ZF1 is higher than that of the samples ZF3. This increase in the value of I[002/101] indicates a higher growth of ZnO in [0002] direction in ZF1 than ZF3, therefore the flowers in the samples ZF1 have petals with smaller diameters than those of the sample ZF3.

3.1 Growth mechanism of ZnO nanoflowers

As discussed above, with decreasing the concentration of DEA, the diameters of nanorods (nanoflowers petals) increased while their number decreased. Based on the heterogeneous growth mechanism [2], the formation mechanism of obtained ZnO nanoflowers can be explained as follows. It is widely believed that the crystals growth mechanism in solution occurs by the formation of nuclei (nucleation process) and complexes like [Zn(OH)4]2− called growth units (crystal growth process). In the later process, the growth units are incorporated into the crystal lattice at interfaces [16]. When the concentration of nuclei is much less than that of growth units, the growth units can easily attach to each nucleus surface at various sites and dehydrate forming single crystals. In the following process, these crystals grow preferentially along c-axis to form branches (petals) resulting in the growth of ZnO nanoflowers [16, 17].

We assumed that when the amount of DEA was 1.5 ml (sample ZF1), a large quantity of growth units along with a small quantity of ZnO nuclei were formed. Therefore, a large number of growth units attached to each nucleus surface at different sites. Subsequently, ZnO nanoflowers with plenty of branches were formed. When the amount of DEA was reduced to 1 ml, the number of growth units decreased compared to that of nuclei leading to a decrease in the number of growth units attaching to each ZnO nucleus. Hence, ZnO nanoflowers with smaller number of branches were formed (sample ZF2). When the amount of DEA was reduced to 0.5 ml (sample ZF3), the growth units concentration further decreased compared to that of ZnO nuclei leading to a further decrease in the number of growth units attaching to each ZnO nucleus. Therefore, ZnO nanoflowers in the sample ZF3 have the smallest number of branches (Scheme 1). Thus, the reduction in the number of petals of ZnO nanoflowers from sample ZF3 to sample ZF1 resulted because of a decrease in the concentration ratio of growth units to ZnO nuclei.

Scheme 1
scheme 1

Schematic diagram of growth mechanism of the different morphologies of ZnO nanoflowers

Full size image

3.2 The effects of ZnO nanoflowers morphology on their properties

3.2.1 Optical properties

To study the effect of ZnO nanoflowers morphology on the optical properties, DRS spectra of the three samples ZF1, ZF2 and ZF3 were measured and presented in Fig. 3a. The estimated band gaps from Tauc plots (Fig. 3b) were 3.01, 3.04, and 3.1 eV for the samples ZF1, ZF2, and ZF3, respectively. It is clear that the band gaps of the three samples exhibited redshift compared to bulk band gap of ZnO (3.37 eV) [18] and the value of redshift changed with respect to morphology.

Fig. 3
figure 3

a DRS spectra, b Tauc plots, and c PL spectra of the samples (I) ZF1, (II) ZF2 and (III) ZF3

Full size image

The authors speculated that the reason behind this redshift and variation in its value with the morphology might be due to the existence of native defects in ZnO nanoflowers structure. To confirm the existence of native defects in ZnO nanoflowers structure, PL study was carried out.

Generally, the higher amount of native defects, the lower recombination rate of photoinduced electron–hole, the lower PL intensity [19,20,21,22]. PL spectra of the three samples are shown in Fig. 3c. The defect bands at 450, 468, and 522 nm are minimum in ZF1, being stronger in ZF2 and is strongest in ZF3 implying the presence of native defects in ZnO nanoflowers with a concentration decreases from the sample ZF1 to ZF3. It is worth to notice that the defect concentration varies with morphology of ZnO nanoflowers. However, determining native defects type from PL spectrum is highly controversial [23,24,25]. Therefore, an accurate analysis is required to determine their type.

3.2.2 XPS analysis

XPS is a potential tool to investigate the surface defects. To determine the type of native defects confirmed by PL study, XPS analysis was performed. Figure 4 shows O1s XPS spectra for the different morphologies of ZnO nanoflowers. Obviously, the O1s spectra of ZF1, ZF2, and ZF3 are asymmetric, indicating the presence of at least two oxygen species in near-surface region. Therefore, typical O1s peaks of ZnO nanoflowers samples were fitted into two peaks. (1) The first one is located at lower binding energies (O1) (~ 530.05–530.12 eV) and is attributed to the O2− ion in the ZnO wurtzite structure (2) the second one is located at higher binding energies (O2) (~ 531.06–531.58 eV) and is attributed to oxygen-deficient regions within the ZnO matrix [26,27,28]. It is believed that the changes in the O2/O1 area ratio could be related to the variations in the concentration of the VO [22, 26, 29]. The (O2/O1) area ratio is highest in ZF1 being weaker in ZF2 and is minimum in ZF3 (Fig. 4) implying the existence of VO in ZnO nanoflowers with a concentration decreases from the sample ZF1 to ZF3.

Fig. 4
figure 4

O1s XPS spectra of the samples a ZF1, b ZF2 and c ZF3

Full size image

3.2.3 Photocatalysis studies

To investigate the morphology effect on the photocatalytic activity of ZnO nanoflowers, phenol degradation over ZF1, ZF2, and ZF3 under sunlight irradiation was carried out. Figure 5a displays the photodegradation curves of phenol over the various morphologies of ZnO nanoflowers. The photocatalysis follows the pseudo first order kinetics law for various morphologies as displayed in Fig. 5b. The first-order rate constants (k) were calculated by taking the slopes of − ln(C/C0) versus t, where C0 and C are the concentrations of phenol at the irradiation times 0 and t min, respectively [30, 31]. Noticeably, the order of photocatalytic activity for the reduction of phenol is ZF1 > ZF2 > ZF3 which is similar the order of VO concentration. Thus, the highest VO, the highest photocatalytic activity. The mechanism behind that will be discussed later in detail.

Fig. 5
figure 5

a The degradation curves and b the degradation kinetics of phenol over the samples (I) ZF1, (II) ZF2, and (III) ZF3 under sunlight, c the photocurrent density response with light on/off cycles under simulated sunlight irradiation and d Nyquist plots (dotted lines in dark and solid lines under simulated sunlight irradiation) of the samples (I) ZF1, (II) ZF2, and (III) ZF3 [0.1 M Na2SO4]

Full size image

3.2.4 Photocurrent study

It is believed that the photocatalytic reaction is an electrochemical process and therefore, the photocurrent is considered as being equivalent to the photocatalytic performance. Moreover, the photocurrent magnitude directly reflected the number of the photoinduced electrons and holes but with less regards to the degrading substrate difference (FTO) [32]. To further analyze the effect of morphology on the recombination probability of photo-induced carriers in the samples and to verify the sunlight photocatalytic activity resulted from VO, the photocurrent experiment was carried out for the samples ZF1, ZF2 and ZF3 electrodes, under simulated sunlight. Figure 5c shows the photocurrent density measured for various morphologies of ZnO nanoflowers as a function of time without voltage via three on–off cycles of light-on and light-off. As expected, the photocurrent density followed the same trend of photocatalytic activity for the reduction of phenol (ZF1 > ZF2 > ZF3) indicating that the amount of VO influenced the number of photogenerated carriers, and the photoactivity was affected, accordingly [32]. This can be ascribed to that the sample with higher VO exhibits lower electron hole recombination and higher visible light absorption. Besides, the photocurrents measurements further confirmed the band gap values of ZF1, ZF2, and ZF3; VO play a role of band gap narrowing instead of just forming the trap centers or active centers [33, 34].

3.2.5 EIS

To interpret the charge transfer resistance as well as the separation efficiency of the photogenerated electron–hole pairs, EIS was done. Figure 5d depicts the typical Nyquist plots of the samples ZF1, ZF2, and ZF3 scanned from 105 to 1 Hz both under simulated sunlight and in the dark. A smaller arc radius indicates more efficient transportation and separation of the charge carriers [35]. As displayed in Fig. 5d, among the three samples, ZF1 has the smallest arc radius followed by ZF2 and ZF3 in order under simulated sunlight and in dark. The fact helps interpret the change of photocatalytic property with morphology of ZnO nanoflowers. Thus, the variation of VO concentration with respect to morphology led to a variation in the efficiency of charge transportation and separation and hence the photocatalytic activity [36].

3.2.6 Antibacterial activity study

Figure 6a, b and Table 1 show antibacterial activity of ZnO nanoflowers against E. coli as Gram-negative and S. aureus as Gram-positive bacteria. From Fig. 6a, it is evident that zone of inhibition increases in the order of ZF3 < ZF2 < ZF1. The increase in zone clearly indicates the antibacterial activity is dependent on morphology of the ZnO nanoflowers, which is in direct relationship to VO. The details of this relationship is discussed in the next section.

Fig. 6
figure 6

a Antibacterial activity of ZnO nanoflowers on (A) E. coli and (B) S. aures: (1-FZ1, 2-FZ2 and 3-FZ3 20 mg/ml respectively) positive control (CTR 30 µg—ceftriaxone disc) and b morphology dependent inhibition of E. coli and S. aureus by ZnO nanoflowers ((I)ZF1,(II) ZF2 & (III) ZF3) 30 mg, data are shown as mean ± standard error (n = 3)

Full size image
Table 1 Effect of ZnO nanoflowers (ZF1, ZF2 & ZF3) 30 mg on cell viability of E. coli and S. aureus
Full size table

The results are expressed as the mean ± standard error, (n = 3). Treated groups showed statistically significant differences from the control group by the two-way ANOVA (P < 0.05).

3.3 The mechanism of morphology effect

As discussed above, the optical, photocatalytic, and antibacterial properties were observed changing with the morphology of ZnO nanoflowers. XPS and PL analyses confirmed the existence of VO in ZnO nanoflowers with concentration varies with their morphology. We believe that, VO concentration is the key reason behind the morphology effect in tuning the optical, photocatalytic, and antibacterial properties of the prepared ZnO nanoflowers.

3.3.1 Optical properties

VO play as self-dopants forming acceptor states in the band gap of ZnO near to the valence band. When the concentration of the VO is high enough, VO states overlap with the valence band pushing it upwards towards the conduction band. That result in the band gap narrowing [32, 33, 37, 38]. The increase in the VO concentration leads to an increase in the number of energy states above the valence band and hence increases the band gap narrowing. Thus, in our case, various morphologies of ZnO nanoflowers display different band gap values because they have different concentration of VO (Scheme 2).

Scheme 2
scheme 2

Schematic diagram of the morphology effect mechanism

Full size image

3.3.2 Photocatalytic activity

Tuning the photocatalytic activity of ZnO nanoflowers by tuning their morphology is also attributed to the variation of VO concentration with morphology. Generally, tuning the photocatalytic activity can be achieved by tuning four important parameters: light absorption, charge recombination, charge separation, and reactive oxygen species (ROS) generation [39]. In our case, (i) the change of VO concentration with morphology led to a change in the band gap values as well as the visible light absorption (Fig. 3a, b) and hence photocatalytic activity changed accordingly. (ii) As discussed above, PL, XPS, and photocurrent studies revealed the correlation among VO concentration, ZnO nanoflowers morphology, charge recombination, and photocatalytic activity. In fact, during the photocatalytic reaction, VO play as electron acceptors and temporary traps of photoinduced electrons reducing the surface recombination of photoinduced electrons and holes [22, 40]. Thus, the higher VO concentration, the lower electron–hole recombination rate, the lower PL intensity, and the higher photocatalytic activity [22, 41, 42]. (iii) In addition, EIS study correlated the charge separation efficiency to VO concentration as well as ZnO nanoflowers morphology. (iv) Furthermore, VO can facilitate capturing photoinduced electrons by the adsorbed O2, simultaneously generating superoxide radical anions [2, 22, 43, 44]. Thus, a variation in ZnO nanoflowers morphology is accompanied with a variation in VO concentration which is associated with a variation in light absorption, charge recombination, charge separation, and ROS generation leading to a variation in photocatalytic activity (Scheme 2).

3.3.3 Antibacterial activity

The mechanism behind variation in antibacterial activity with morphology can be explained as follows. ZnO nanoflowers can generate ROS, which results in antioxidative suppression leading to cell death. That is, ZnO as a semiconductor can generate ROS. Electrons in valence band can be promoted to the conduction band, leaving behind holes. The electron and hole will migrate to ZnO nanoflowers surface to react with O2 and −OH ions, respectively. This leads to the formation of superoxide (•O2−) and hydroxyl (•OH−) radicals [45]. Various ROS will be produced triggering redox-cycling cascades in the cells, causing permanent oxidative damage. It was proposed that there is a positive relationship between ROS generation and VO [46]. Moreover, it was reported that antibacterial activity can be tuned by tuning the VO concentration and the higher VO concentration, the higher antibacterial activity [47]. In this way, the change in the antibacterial activity with morphology results from the change of VO concentration with respect to morphology.

Thus, the change in the morphology of ZnO nanoflowers was associated with a variation in the concentration of VO which played the key role in tuning the optical, photocatalytic and antibacterial properties.

4 Conclusions

In this study, different morphologies of ZnO nanoflowers have been synthesized via hydrothermal method by varying the DEA concentration. The DEA concentration played a significant role in controlling the concentration ratio of growth units to nuclei leading to tuning the morphology of ZnO nanoflowers. Tuning their morphology led to tuning their optical, photocatalytic, and antibacterial properties. The tuning effect of morphology was attributed to Vo concentration which played as self-dopants, hence, a change in their concentration led to a change in the band gap (optical properties). Moreover, VO played a vital role in tuning the concentration of ROS and thereby tuning the photocatalytic and antibacterial activities. The findings of this study provide a fundamental understanding of the mechanism behind morphology effect in tuning ZnO nanoflowers properties.