Adsorption and photocatalytic activity of biosynthesised ZnO nanoparticles using Aloe Vera leaf extract

In this article, we demonstrates the growth of phase pure ZnO nanostructures from Aloe-Vera leaf extract and degradation of an organic dye-Malachite Green (MG)- from aqueous medium using the same as catalyst. Adsorption mechanisms were evaluated using Lagergren’s pseudo-first-order, pseudo-second-order and intraparticle diffusion kinetic models. X-Ray diffraction data showed that the synthesised ZnO is crystalline with hexagonal wurtzite phase. Average crystallite size and lattice strain was estimated from Scherrer equation and Williamson-Hall analysis with the help of Rietveld refinement data. Crystallite size obtained from Scherrer method is 12.62 nm while that from Williamson-Hall analysis is 19.27 nm. Uniform growth of ZnO nano-sheets were confirmed by FE-SEM analysis. Optical characterisation was carried by UV-Visible spectroscopy and the band gap ZnO nanoparticles was found to be 3.19 eV. Zn-O stretching vibrations were recorded at 550 cm−1 using FTIR spectrophotometer. Results showed that biosynthesised ZnO nanosheets are particularly effective for the degradation of MG dye.


Introduction
ZnO is a well-known n type II-VI semiconductor having hexagonal wurtzite crystal structure with a broad direct energy bandgap ranging from 3.1 to 3.3 eV and high exciton binding energy. Owing to the unique characteristics, ZnO nanostructures are widely used in various fields like optoelectronic device manufacturing [1], photovoltaics [2], biosensors [3], solar cell [4] and energy storage devices [5]. US Drug and Food Administration recognized ZnO as a safe companion for environment and food materials due to its adsorption capacity of UV light and transparent nature in visible light [6]. ZnO nanoparticles (NP)s are highly active against microbes and organic dyes [7,8]. ZnO has several advantages such as low cost, non-toxic, chemically stable and high availability in nature [8].
TiO 2 is the widely used photocatalyst owing to its suitable electronic and optical properties, low toxicity and higher physical and chemical stability. Similar to TiO 2, ZnO possess favourable exceptional optical, electrical and mechanical properties. Another important factor is the production cost, which is upto 75% lower when compared with TiO 2 and Al 2 O 3 nanoparticles [9]. As per the literature, ZnO is reported to exhibit higher absorption efficiency over a larger fraction of solar spectrum compared to TiO 2 [10]. Larger bandgap of ZnO enables the utilization of ultraviolet wavelengths. Also in comparison with TiO 2 , ZnO has higher photocatalytic efficiency and has good environmental stability.
Several methods are in practice to synthesis ZnO nanoparticles including hydrothermal [11], sol gel [12], combustion [13], precipitation [14], electro deposition [15], spray pyrolysis [16] and biosynthesis [17]. Each method has its own contributions in tuning the shape and size of ZnO NPs. Biosynthesis, otherwise known as green synthesis is an innovative method to prepare NPs. In biosynthesis, instead of chemicals, natural elements like plant parts, fungi, bacteria and enzymes act as stabilizing and reducing agents [18]. Green synthesis methods are low cost, simple and environmental friendly due to its ease in availability of reducing agents and production of non-toxic materials [19]. Nowadays synthesis of ZnO using some medicinal plants like Azadirachta indica [20], Calatropis procera [21], Polygonum Chinense [6], Mimosa Pudica [22], Moringa oleifera [23], Aloe barbadensis miller [24], Solanum nigrum [25] are being reported.
Availability of clean and affordable drinking water is a globally challenging issue. A lot of procedures named osmosis, ultrafiltration, adsorption and flocculation are in practice to remove these impurities. Not all them are much effective but are very costly time consuming process. Toxic by-products are also produced in some cases. Photocatalysis is regarded as an excellent method for treating industrial dye effluents and hydrogen evolution [26]. ZnO [27], TiO 2, [28] g-C 3 N 4 [29,30] etc are some of the widely used photocatalyst for the dye removal. Absence of toxic by-products makes photocatalysis more attractive. Here, we synthesized ZnO nanoparticles from Aloe Vera extract and dye removal efficiency of ZnO NPs against Malachite green dye in the presence and absence of sunlight is studied. Also, this study focusses on the adsorption kinetic study which provides the mechanism behind adsorption and predicts the rate of adsorption.
Kinetics was studied using pseudo first order (PFO), pseudo second order (PSO) and intra-particle diffusion (IPD) model.

Materials
Aloe Vera fresh leaves collected from home garden were used for the synthesis. Zinc Nitrate hexahydrate (Zn (NO 3 ) 2 .6H2O) (98.977% purity) and Sodium hydroxide (NaOH) (98%) pellets was purchased from Emplura and Emparta chemicals respectively. Double distilled water was used in the entire experiment for washing and preparing solutions. In order to investigate adsorption and photo catalytic study, Malachite Green (MG) dye was purchased from Nice Chemicals. Figure 1 shows the synthesis routes of ZnO nanoparticles from Aloe Vera leaf extract. Aloe Vera leaves were washed in double distilled water for removing mud and other impurities. Washed leaves were then transferred to a clean beaker and 250 ml distilled water was added to it. Beaker was then placed on a heating mantle and boiled for about 45 min till the colour changed into pale yellow. As-prepared solution was separated into extract using Whatsmann filter paper. The filtered leaf extract was transferred into clean container and stored in refrigerator for further experiments and studies. 30 ml of 1 M Zinc Nitrate solution was made using double distilled water. 30 ml of leaf extract was then added to the solution and stirred for 2 min pH of the solution was maintained at 10 by adding appropriate quantity of NaOH (5 M) solution. The whole solution was stirred for about an hour. After the reaction time, resultant solution was filtered using filter paper. Product was then washed by ethanol and water respectively to remove impurities. Resultant powder was dried in an oven at 80 C.

Characterization
Structural, morphological, optical and vibrational properties of the synthesised ZnO nanostructures were studied using respective characterisation techniques. The structural analysis was carried out by powder x-ray diffraction technique using Rigaku Miniflex 600 with CuKα (λ = 1.5406 Å) radiation in the range of 20°to 80°. Morphology of the synthesised sample was investigated by Field Emission Scanning Electron Microscope (FESEM). UV Visible absorption spectrum of ZnO nanoparticles were recorded in the range of 200-900 nm using JascoV-760 spectrophotometer and Photoluminescence emission spectrum was obtained from Horiba Fluoromax Spectrofluorometer. Fourier Transform Infrared (FTIR) measurements in the range of 500-4000 cm −1 were recorded using Perkin Elmer Spectrometer.

Adsorption studies
Adsorption experiments were carried out in a 500 ml beaker where 10 ppm of MG dye solution was prepared initially. 10 mg ZnO nanoparticles were then added to the dye solution. The beaker was well wrapped with aluminium foil in order to avoid photolysis reaction and is kept for stirring. A known amount of sample was withdrawn from the solution at predetermined intervals to monitor the absorbance using UV-Visible spectrophotometer.

Photocatalytic activity of ZnO nanoparticles against malachite green dye
Biosynthesized ZnO nanoparticles were introduced as catalyst to study photocatalytic removal of malachite green under solar irradiation. Accurately weighed Malachite green powder (2 mg) was dissolved in 200 ml double distilled water to prepare 10 ppm dye solution and is kept under constant stirring. 10 mg of the synthesised ZnO was then added to the dye solution (sample -A). First 30 min of stirring process was done in dark to achieve adsorption equilibrium [31,32]. After attaining equilibrium, the solution was exposed to sunlight under constant stirring. Similarly, to study the effect of Aloe Vera leaf extract, if any, on the degradation of MG dye, a 10 ppm dye solution with 10 ml Aloe Vera leaf extract in the absence of ZnO nanoparticles was exposed to sunlight under constant stirring (sample -B). Absorption spectrum for both the samples was recorded using UV-Visible Spectrometer at regular intervals of 30 min

Crystallite size calculation and microstrain analysis
Physical and mechanical properties of crystalline nanoparticles strongly depends on its crystallite size [37]. Crystallite size of the synthesised ZnO can be calculated from Debye-Scherrer equation [38] and Williamson-Hall method [39].

a) Scherrer Method
Crystallite size of the nanoparticles can be calculated from the full width at half maximum of the diffracted peaks. Total broadening of the diffraction peak is due to the sample and instrument dependent effects. For this reason, a standard silicon sample with small microstrain and large particle size is used to findout the instrumental broadening [38,39]. Let b expt be the measured width and b stand be the width due to standard sample i.e., instrumental width, then actual broadening β can be written as Average crystallite size was calculated using the Scherrer equation where 'D' is average crystallite size, 'k' is called Scherrer constant, 'λ' is the wavelength of the incident Cu-Kα radiation (1.5406 Å). Peak broadening at lower angle is more suitable for the calculation of crystallite size [41]. Crystallite size calculated using Scherrer equation is found to be 12.62 nm.

b) Williamson-Hall method
In Scherrer method, the significance of microstrain is neglected even though its effect is present in the diffraction pattern. Lattice strain is one of the main factor that induces peak broadening. Reason for strain includes the sinter or contact stress, grain boundary, faults in stacking and coherent stress [39,42]. Stokes and Wilson derived the contribution of microstrain to the line broadening as They proposed a method-Williamson Hall (WH) method-for including the crystallite size and microstrain contributions to the line broadening analysis which relied on reflection order. Thus WH plot is considered as one of the simple and effective methods for counting the crystallite size and microstrain in the line broadening studies of XRD profile. Total broadening is thus given by where ′β′ is the sum of the broadening due to crystallite size (b D ) and that due to microstrain (b e ). Therefore On rearranging, the individual contributions to the broadening of reflections [43] can be expressed as ¢ 4 sin provides the microstrain e ¢ ¢ and the mean crystallite size can be obtained from the intercept [44]. The straight line plot indicates the homogeneous distribution of particle size and microstrain [45] Average crystallite size calculated from W-H plot is 19.27 nm. The crystallite size estimated from Scherrer equation and W-H plot showed negligible variation. This variation may be due to the difference in averaging the particle size distribution [38,46]. Also, the variation in crystallite size points out that the strain also should be taken into consideration while calculating the crystallite size from XRD profile [47].
Microstrain can induce a greater broadening in the diffraction peak [44]. Peak broadening due to crystallite size is used in the Scherrer equation while effect of microstrain is considered in the Williamson-Hall method.
• Estimation of micro strain Plot of βcosθ versus 4 sinθ is analysed to study the strain in ZnO lattice. Value of microstrain is measured from the slope of the line in figure 3.
We obtained a negative slope for the ZnO nanoparticles. Negative slope reveals the compressive nature of residual stress. In the case of nano ranged particles, negative plot of W-H plot indicates the presence of compressive strain while positive slope represents the tensile strain [37,48,49]. Crystallite size and strain obtained from Scherrer equation and WH plot is shown is table 1.

Structural analysis: rietveld refinement
Rietveld method is used to refine the XRD data. In rietveld refinement, we are fitting calculated profile (containing all structural and instrumental parameters) to experimental data [50,51]. In refinement the rietveld algorithm optimizes the model function such that the weighted sum of squared difference between observed and computed intensity values are minimized. Quality of refinement is judged by the reliability indices given by weighted profile factor (R wp ), expected weighted profile factor (R exp ) and the goodness of fit (χ 2 ). R p is explained as a function of the integrated intensities of the peaks. This index gives a measure of quality of the refined crystalline structure.
R wp shows the square root of the weighted difference between observed and computed intensity values. It should be analysed to verify the convergence of refinement.
[ ] is the uncertainity estimate for y o i , . y o i , is the observed intensity values. R exp is an estimation of the best possible R profile (R wp ) that could be obtained based on the statistical noise of the experimental diffraction pattern. i.e., is the expected statistical value for R wp . The goodness of fit (c 2 ) is calculated by The ideal value for c 2 is 1. R p is explained as a function of the integrated intensities of the peaks. This index gives a measure of quality of the refined crystalline structure.  parameters (a, b, c), unit cell volume (V), the

FE-SEM analysis
Morphology of the synthesised ZnO nanostructure is characterised by FE-SEM images (shown in figure 4). FE-SEM micrographs clearly revealed the formation of randomly distributed ZnO nanosheets with well-defined particle boundary. The well-proportioned ZnO nanosheets have thickness around 15-40 nm. Similar nanosheet structures has been already reported by Hazmi et al [53] and Bazta et al [54].

UV-vis absorption spectrum
A comparative optical absorbance spectrum of the bio-synthesised ZnO nanosheets and aloe vera leaf extract taken by UV -visible spectrometer in the range of 300 nm-900 nm is shown in figure 5(a). Presence of strong excitonic absorption peak at 359 nm of ZnO nanosheets indicates the electronic transition from O 2p to Zn 3d, corresponding to the band-to-band transition of the ZnO energy band structure [54,55]. A keen observation on the spectrum points out a negligible, low intensity visible region emission (around 550 nm) which can be attributed to the defect induced emission due to the existence of Zn or O vacancies [56]. Existence of excitonic peak in the ultraviolet region along with an absorption in the visible region clearly depicts the UV as well as visible light activity of ZnO nanosheets. High transparency in the visible region and low transparency in the UV regions make it a good material for photovoltaic applications [57,58]. Absence of other peaks in the spectrum indicate the good optical property of synthesised ZnO. Aloe vera leaf extract showed a bump around 300 nm which is likely to have originated from the polyphenolic compounds [59].
Optical bandgap energy of the synthesised ZnO nanosheets was analysed from Davis and Mott expression which relates incident photon energy (hν) with absorption coefficient (α) through the Tauc plot equation [60] a n = n - where A is a constant independent of energy or band tailing parameter [61], h is Planck's constant, n is the frequency of incident photon (in Hz), E g is the optical bandgap energy in eV and n is power factor. Depending upon the nature of transition, n may have values or 1 2 , 3 2 , 2 3 for direct allowed, direct forbidden, indirect  allowed or indirect forbidden transitions respectively. In this case, value of n is ½. Optical bandgap energy is calculated by plotting a graph between a n h 2 ( ) and photon energy ( n h ). Extrapolating linear portion of the curve to a n h 2 ( ) = 0 gives the optical energy bandgap. Measured value of E g for the biosynthesised ZnO NPs is 3.19 eV ( figure 5(b)) and it matches well with the literature [62][63][64].

Photoluminescence spectrum
Room temperature Photoluminescence (RTPL) spectrum of ZnO is designated with two emission bands; the intense sharp UV emission at around 380 nm and a broad visible emission ranging from 400 to 700 nm [65]. UV emission-the near band edge emission (NBE)is regarded as the key characteristics emission of ZnO [66]. Another band located in the visible spectral region is generally known as deep level emission (DLE) [67]. NBE emission peak is typically attributed to excitonic recombination [68,69] while the emergence of DLE is usually connected with intrinsic or extrinsic defects. RTPL spectra of bio-synthesised ZnO NPs are recorded using fluoromax spectrophotometer. Figure 6 describes the emission spectrum of biosynthesied ZnO NPs with excitation wavelength 358 nm at room temperature. Sample exhibits a notable UV emission peak at 388 nm and an optically active defect states in the blue and green spectral regions. Two shoulders are visible at 446 nm and 464 nm (blue emission) and a broad peak at around 555 nm (green emission).
Emission peak observed at 388 nm can be assigned to the band edge transition or exciton combination of the wide-band-gap ZnO [66,70] and is in good agreement with UV-vis measurement. It is found that the intense characteristic band edge emission is generally observed in the range of 384 nm-391 nm [71]. In principle there are a number of defect states within the bandgap of ZnO [72]. Zn interstitials and oxygen vacancies are known to  be the predominant ionic defect types [73]. Blue emission of ZnO nanoparticles are proposed with transitions involving Zn interstitial defect states. Emission lines at 446 nm and 466 nm have been reported in the literature [74,75]. Emission at 446 nm is attributed to the transition between shallow donor (oxygen vacancy) to the valence band [74]. Oxygen and Zn vacancies or interstitials and their complexes are considered as the cause for 466 nm emission peak [74]. Green emission is the most commonly observed deep level emission in ZnO nanostructures. Broad peak at 555 nm can be related to the singly ionized oxygen vacancies [65,76]. Different hypothesis including singly ionized oxygen vacancies, antisite oxygen [71], doubly charged oxygen vacancy [77] and surface defects [78] have been proposed for the origin of green emission. Although these assumptions are observed in literature, origin of these defects especially the green emission have been controversial for quite a long time [79].

FTIR analysis
FTIR analysis gives an insight into the vibrational fingerprints of ZnO nanoparticles. Figure 7 shows the FTIR spectrum of synthesised ZnO nanoparticles. Most commonly the vibration peaks of metal oxides are seen below 1000 cm −1 region. The peak at 550 cm −1 clearly shows the presence of Zn-O stretching vibration [80]. The other peaks in the range 500-4000 cm −1 represents the vibration of some other molecules attached to the surface of Zinc Oxide nanoparticles. The -C-H bending vibration and C=C stretching vibration of alkane group are observed at 1381 cm −1 and 1641 cm −1 respectively [65]. Vibrations due to the biomolecules attached to the Zinc Oxide Nanoparticles are recorded at 835 cm −1 and 686 cm −1 which attributes to C-H stretching vibrations present in the alkanes and bending vibrations present in aromatic rings respectively [23]. Peak at 3420 cm −1 corresponds to OH stretching vibrations and it shows the absorption of CO 2 and molecular water on the surface of nanoparticles [26].

XPS analysis
Inorder to study the surface chemistry and oxidation states of ZnO nanoparticles, x-ray Photo electron spectroscopy was utilized. C 1s peak at 284 eV was utilized to analyse the binding energies of Zn and O. XPS wide scan spectrum for ZnO nanoparticles is shown in figure (8) and it indicates the presence of core elements Zn and O. No other peaks are observed revealing the purity of synthesised sample. Core level spectra of Zn2p ( figure 8(b)) displays a simple spin-orbit doublets at 1022.1 eV and 1045.2 eV corresponding to Zn 2p 3/2 and 2p 1/2 electronic states indicating the 2 + state of Zn in the sample. Similar peaks have been reported by [81]. Core level spectra of O 1s state in the ZnO sample is illustrated in figure 8(c). Spectra clearly shows the presence of a peak at binding energy 531.6 eV. According to the reports of Dupin et al [82] this peak can be assigned to Ospecies.

Photocatalytic study of ZnO NPs against malachite green
ZnO acts as good photocatalyst and degrades the dangerous organic dyes in aqueous medium under solar irradiation. The process of photocatalysis involves different steps and a schematic representation of photocatalysis mechanism is shown in figure 9. ZnO-Malachite Green mixture exposed to sunlight generates   with O 2 to produce superoxide radical anion of oxygen [84].
OH· and -O 2 · then reacts with and decomposes the dye into non-toxic bi-products.

Dye e reduction products
cb +  Dye OH degradation products · As the time progresses, blue colour of the sample-A gradually vanishes and finally become colourless while no notable changes was observed for sample-B. Readings are taken in a time interval of 30 min and the obtained spectrum is shown in figures 10(a) and (c). From the spectrum, it is clear that as the reaction goes on, the characteristic peak of Malachite green (617 nm) decreases exponentially and vanishes after 4 h in the case of sample-A while no decrease in peak was noted even after 4 h for sample-B. This clearly indicates that the decrease in dye concentration is purely due to the presence of ZnO photocatalyst and not due to Aloe Vera leaf extract.
The percentage of dye degradation of sample (a) is calculated by the equation, is the initial concentration and 'C' is the concentration of dye at time t.

Adsorption studies
The amount of MG dye adsorbed q t (mgg −1 ) by the adsorbent at time t, was calculated using the equation C 0 and C e (mg l -1 ) are the concentrations of MG dye at initial and any time t respectively, V is the volume of solution in litres, m is the mass of adsorbent in grams. Figure 12 shows the plot of percentage of dye degradation with time.
It is observed that 76.73% of initial dye concentration was removed after 180 min showing a good adsorption capacity of ZnO nanoparticles over the removal of MG dye in aqueous medium. Adsorption and photocatalysis results shows that ZnO nanostructures could be used as an efficient catalyst for MG dye removal both in the absence and presence of light irradiation.

Kinetic study
To gain an idea on the unseen mechanism behind the observed adsorption phenomena, adsorption kinetics is investigated. Adsorption kinetics predict the rate of dye removal as well as draw a picture on the dynamics of sorption reaction [87]. Kinetics behind the pollutant removal are divided into two categories: (i) kinetic models which describe the interaction between pollutant and reaction sites on the adsorbent surface (Pseudo-first order, Psuedo-second order and Langmuir kinetic model) (ii) Diffusion models. Former kinetic model do not take diffusion into account. Among the various models in practice, Lagergren pseudo first order (PFO) and pseudo second order (PSO) rate equations are commonly applied to examine the adsorption mechanism. Determining best-fit kinetic model is the most common way to predict the suitable adsorption kinetic expression. PFO, PSO and Intra particle diffusion models are discussed in this study.
• Pseudo first order rate equation can be expressed as where q e and q t are the amount of solute adsorbed per unit mass of adsorbent at equilibrium and any time t (min) respectively, k 1 (min −1 ) is the rate constant of PFO reaction.
• Pseudo second order rate equation is given by where k 2 (mg g −1 min −1 ) is the rate constant of PSO reaction. Experimental kinetic results were modelled according to the above mentioned reactions. Kinetic parameters and regression coefficient obtained from the simulations using Origin 9 software is listed in the table 3. The validity of kinetic model is tested by the magnitude of regression coefficient (r 2 ). The r 2 value for PFO model is 0.98984 and that for PSO is 0.98807. Obtained results suggest that Lagergren pseudo first order model is suitable for modelling the MG adsorption onto ZnO. Figure 13 illustrates the obtained PFO model.

• Intraparticle Diffusion (IPD) Model
Intraparticle diffusion plays an important role in the adsorption process. Several diffusion processes are known to affect the adsorption. To evaluate the roll of intraparticle diffusion (IPD), the kinetic data was fitted into intraparticle diffusion model given by Weber and Morris [88], where q t is the amount of dye adsorbed at time t(mgg −1 ) and k id is the IPD rate constant (mgg −1 min −1/2 ), c is the boundary layer thickness. Slope of the plot between q t and t 0.5 gives the value of k id . If the line passes through the origin, it can be concluded that intraparticle diffusion is the controlling factor and if it do not pass through the origin, intraparticle diffusion as well as some degree of other boundary layer might also control the rate of adsorption [89].  Obtained plot ( figure 14) showed multilinearity characterisations showcasing three step adsorption process; initial portion represents the fast transfer of MG dye molecules on to the ZnO surface by physical/chemical force, the subsequent portion is attributed to the IPD or the rate determining step and the third region represents the saturation/equilibrium [38]. Also, the plot do not pass through the origin indicating that intraparticle diffusion is not the sole reason for adsorption in this case. This result suggest the joint functioning of intraparticle with surface sorption for the intake [90] of dye using ZnO nanoparticle. IPD constants calculated from the plots are given in table 4.

Conclusion
In this study, we successfully synthesised biogenic ZnO nanostructures from Aloe vera leaf extract. XRD analysis confirmed the formation of hexagonal wurtzite ZnO. Crystallite size estimated from Scherrer equation (12.62 nm) and WH plot (19.27 nm) showed small variation which may be due to the difference in averaging the particle size. Negative slope indicated the presence of microstrain in ZnO lattice. Structural refinement by retvield method obtained cell parameters and atomic positions which are in good agreement with previous reports. FESEM micrographs showed the formation of randomly distributed ZnO nanosheets. Optical bandgap energy of the synthesised ZnO measured from Tauc plot is 3.19 eV. In the PL spectrum, sample exhibited a UV emission (at 388 nm), blue emission (at 446 and 464 nm) and a broad green emission (at around 555 nm ). FTIR analysis gave the vibrational outlook of ZnO nanosheets. Under solar irradiation, ZnO was found to remove around 85% of the initial MG dye concentration within 4 h. On the other hand, 76.23% of the initial dye concentration was removed after 3 h by adsorption mechanism. Adsorption kinetics was described by pseudo first order model with r 2 factor 0.98984 giving best fit to the experimental data. Intraparticle diffusion was not the sole reason for adsorption of MG dye since the line of plot does not pass through the origin. This work demonstrated that the ZnO nanostructures could be exploited as a feasible and eco-friendly photocatalyst for the treatment of dye-polluted waste water.