Growth kinetic study of ionic liquid mediated synthesis of gold nanoparticles using Elaeis guineensis (oil palm) kernels extract under microwave irradiation

2.1 Materials

Hydrogen tetrachloroaurate (HAuCl₄·3H₂O, 99.99%) and IL[EMIM][OAc], (1-ethyl-3-methyl-imidazolium-acetate), >97% were purchased from Sigma-Aldrich Chemical Co. (St. Louis, MO). Oil palm kernels were collected from Felcra Berhad Nasaruddin oil palm plantation located at Bota, Perak, Malaysia. All other reagents used in the experiments were analytical grade and were used without any further purification.

2.2 Methods

3.1 UV–vis spectroscopy analysis

UV–vis photospectrometer has been frequently used to monitor the synthesis, sizing and stability of AuNPs in aqueous form (Nasrollahzadeh et al., 2015b). When light photons strike with surface of AuNPs, these become excited and exhibit a strong absorption band in the visible region. This takes place when the frequency of the electromagnetic field is resonant with the coherent electron motion what is known as “surface plasmon resonance (SPR) absorption. Preliminary assessment for presence of AuNPs can be observed from pinkish color appeared in the reaction mixture. Initial color of aqueous extract of OPK was initially translucent white and turned into pale yellow after addition of chloroauric acid. On heating, colloidal solution exhibits intense color (ranging from yellowish to pink to ruby red) due to its SPR. Using CH, time for reduction of trivalent gold ions to gold particles was changed dramatically with increase in temperature. At room temperature, time for color change was more than 6 h which was significantly reduced to 17 min on increase in temperature from 30 °C to 100 °C as shown in Fig. 1 inset. UV–vis analysis was performed for colloidal samples which showed an increase in absorbance with increase in temperature as shown in Fig. 1. It was also observed that increase in temperature up to 80 °C resulted in shifting of SPR peaks toward lower wavelength region i.e. from 554 nm to 544 nm which endorsed the formation of small particles (Bhuvanasree et al., 2013). Surface plasmon band is actually created due to oscillation of metal electrons in the presence of visible light that change their position due to change of particles size (Joseph & Mathew, 2015). At high temperature, reactants are consumed rapidly to yield smaller particles which appeared in a blue shifting of plasmon band (Ibrahim, 2015). However, increase in temperature toward 100 °C resulted in slight red shift of SPR peaks which may be due to agglomeration process and larger particle size formation at very high temperature (Gan et al., 2012). It was also resulted in sizable elevation of SPR peaks from horizontal axis at longer wavelength region (600–700 nm) as shown in Fig. S1. Similar pattern was observed with AuNPs synthesized using Dracocephalum kotschyi leaf extract where increase in temperature was resulted in blue shifting of SPR peaks. However, crossing a certain temperature limit, the position of SPR peaks was red shifted which is due to formation of larger particles size (Dorosti & Jamshidi, 2016).

Close

Fig. 1

UV–vis spectrometer: absorbance and λ_max at different temperatures, Inset; time for color change at different temperature.

Export to PPT

Reaction was also carried out using MW irradiation to find out its impact on rate of reduction reaction and synthesis of AuNPs. Using MW irradiation, color of reaction mixture was changed within 1 min. The time for color change using MW irradiation was significantly less than using CH i.e. 17 min - 6 h. MW induced AuNPs synthesis in ILs results in enhanced heating rate with fast nucleation to complete the synthesis reaction in less time (Richter et al., 2013). Therefore, MW synthesis was appeared in much faster rate of reaction in comparison with CH synthesis, not only at room temperature but even at 100 °C. The SPR peaks absorbance intensities were increased with increase of irradiation time as shown in Fig. 2. High temperature rise of solution within short period of time promoted greater number of nuclei that provided high concentration of particles in solution. This phenomenon was also reflected by blue shifting of λ_max which indicated the formation of smaller particles as shown in Fig. 2 inset. The positions of SPR peaks using MW irradiation were found in the region of 524–529 nm which were significantly toward lower wavelength region when compared with CH where these peaks were appeared in the region of 544–554 nm. After MW irradiation of 3 min, a minor difference in absorbance value was observed that indicated the completion of gold ions to AuNPs in the reaction mixture. Sharp SPR bands, high absorbance with lower wavelength using MW irradiation in comparison with CH is an index for synthesis of small AuNPs with less rate of agglomeration (Nadagouda et al., 2011).

Close

Fig. 2

UV–vis spectrometer; comparison of AuNPs using CH and MW irradiation, Inset; effect of MW heating time on absorbance.

Export to PPT

3.2 TEM image analysis

TEM image analysis was carried out for colloidal AuNPs synthesized using CH prepared at 80 °C. The main reasons to select this temperature i.e. 80 °C as a representative temperature are attributed to its sharp SPR peak and smaller wavelength value which are symptoms of formation of uniform and small particles (Bhuvanasree et al., 2013; Rai et al., 2010). TEM images suggested the presence of predominant spherical shape of AuNPs along with some triangular and multiple twinned structure as shown in Fig. 3a and b. Statistical data for more than 200 particles confirmed an average diameter of particles as 12.18 ± 3.14 nm which was found slightly larger when compare with 8.72 ± 2.21 nm achieved through MW synthesis (Irfan et al., 2019). MW irradiation reduced the average diameter of AuNPs as illustrated in Fig. 3c and d that was in agreement with the blue shifting of λ_max observed through UV–vis photospectrometer (Fig. 2 inset). TEM histogram revealed a particles size distribution ranging from 4.65 nm to 19.47 nm synthesized using CH (Fig. 3e). However, using MW irradiation, AuNPs were relatively in narrow range of 4.07 nm to 16.35 nm as demonstrated in Fig. 3f. Rapid heating and augment MW absorption tendency of IL at the initial stage lead to super saturation of gold atom in the solution that provided short time for nucleation which resulted in formation of small AuNPs (Nadagouda et al., 2011; Seol et al., 2011). The rapid nucleation in a short period of time also played a vital role to get the monodisperse nanoparticles (Zayed & Eisa, 2014). This resulted in fast reduction of metallic ions into metallic species simultaneously followed by conversion and growth to stable particles. The significant blue shifting of SPR bands, high absorbance and a rapid reaction between kernel extract and chloroauric acid to synthesize AuNPs show the symptoms of supremacy of this heating mode over CH.

Close

Fig. 3

TEM image and histogram of AuNPs synthesized using CH (a, b and e) and MW irradiation (c, d and f).

Export to PPT

3.3 XPS and EDAX analysis

To evaluate the surface chemistry of AuNPs, XPS analysis of AuNPs was performed using OPK extract which showed a strong presence of C, N, O and Au peaks as shown in Fig. 4a. XPS analysis for AuNPs sample showed a strong observation of Au spectrum. Fig. 4a inset illustrates the Au spectrum peaks identified for 4f_5/2 and 4f_7/2 at 88.08 eV and 84.38 eV, respectively. The difference in B.E for these characteristics peaks was 3.7 eV that is in agreement with the reported value (Kannan et al., 2014). Intensity for peak at 4f_7/2 was observed much higher indicating the dominant presence of metallic gold (Au⁰) in the mixture. It endorsed that gold ions present in the reaction mixture has been reduced and converted into AuNPs. Bulk Au(0) atoms are known to have Au 4f_7/2 binding energy (BE) of 84.0 eV (Yong & Hahn, 2013). However, a small positive shift of 0.38 eV observed at Au 4f_7/2 indicated the strong interaction of anions with the metallic gold at their interferences (Maji et al., 2014). It was in accordance with the positive shift of BE reported in [6,6]-phenyl-C61 butyric acid methyl ester (PCBM) functionalized AuNPs (Yong & Hahn, 2013). Fig. 4b shows the EDAX spectrum which also confirmed the presence of Au. EDAX spectrum also displayed peaks for C, O and N which might be appeared due to organic moieties and IL present in the mixture of AuNPs. Peak for Si is due to the grid used for making a thin film of AuNPs for further analysis.

Close

Fig. 4a

XPS survey spectra of AuNPs; Inset: spectrum for Au 4f band.

Export to PPT

Close

Fig. 4b

EDAX spectra of AuNPs, Inset EDAX map.

Export to PPT

3.4 FTIR analysis

FTIR spectroscopy analysis was performed to observe the spectrum of AuNPs to identify the vibrational frequencies in response to different bio-functional groups of OPK extract those participate in the conversion reaction of gold ions to AuNPs. FTIR spectrum of OPK extract before and after reaction with gold precursor was performed and major bands were observed at 3453.1 cm⁻¹, 1661.2 cm⁻¹, 1125.6 cm⁻¹ and 659.3 cm⁻¹ as shown in Fig. 5. Band observed at 3453.1 cm⁻¹ indicated the O—H stretching vibration of alcohol and phenolic compounds present in the kernels extract. Phenolic groups and flavonoids compounds are also reported as active bio-groups present in oil palm mills effluent (Gan et al., 2012). These have been consider as reducing agent involved in reduction of gold ions and have the ability to work as natural stabilizers (Hatamifard et al., 2016; Irfan et al., 2017). Bands present at 1661.2 cm⁻¹ and 1125.6 cm⁻¹ are raised due to presence of amide C⚌O stretching and C—N stretching of aliphatic amine or phenolic groups respectively. There was a clear change in intensity of band in the wavenumber of 659.3 cm⁻¹ which corresponds to the aromatic C—H bending and showed the possible involvement of aromatic C—H groups in the synthesis process. Change in band positions at 3453.1, 1661.2 and 1125.6 cm⁻¹ showed significant presence of phenols, flavonoids and protein molecules that play dominant role in reduction and stabilization of AuNPs (Irfan et al., 2017; Nasrollahzadeh et al., 2014).

Close

Fig. 5

FTIR analysis for OPK extract before and after reaction (extract + AuNPs).

Export to PPT

3.5 Growth behavior of AuNPs using OPK extract

The SPR band is influenced by nature of metal present in the solution, its size and shape and surrounding medium (Ghosh & Pal, 2007). The position of SPR peaks is also changed with interparticle distance. Zhong et al. explained that a well isolated colloidal solution could be exhibited by formation of only single SPR peaks (Zhong et al., 2004). However, when two particles approach to each other to form an aggregate, it provides two absorbance maxima. First peak for formation of single particles is located at lower wavelength region and attributed to quadrupole plasmon excitation in couple sphere. The other peak identified toward longer wavelength region (600–700 nm) could be due to formation of aggregates and attributed as dipole plasmon resonance of AuNPs. The maximum shift in peaks occurs when interparticle distance approaches to zero give rise to maximum electrodynamics interactions. On observing the changes in the SPR peak positions of synthesized AuNPs through periodic UV–vis analysis (Irfan et al., 2019), three distinct phases could be identified to interpret UV–vis data for OPK extract as demonstrated in Fig. 6. In the first phase, primary peak for AuNPs was observed at 527 nm. This peak consistently grew up to 22 days due to formation of more nanoparticles with minor blue shifting toward 524 nm as shown in Fig. 6 phase 1. Hereafter, absorbance value become constant up to 30 days probably due to existence of a kinetic equilibrium between new nanoparticles and aggregated of already formed nanoparticles that shifted the SPR band toward higher wavelength (phase II). In phase II, no significant difference in λ_max and absorbance values of AuNPs mixture was observed. However, slightly upward movement of secondary peak at longer wavelength region (600–700 nm) was observed that might be due to some formation of aggregates (Fig. 6 Phase II). In Phase III, absorbance value started to decrease at 524 nm while peak positions in the longer wavelength region were still increasing upward (Fig. 6 Phase III). This change in peak position at longer wavelength region is attributed to dipole-plasmon excitation in coupled spheres showing increase of particles size. Initially, when rate of reaction was high, SPR band at 524 nm was increased continuously showing formation of new nanoparticles. These particles were probably capped due to the presence of bio-compounds and ILs to resist the agglomeration. With passage of time, primary peak formed at 524 nm continuously moved down showing cessation of formation of new particles and shifted toward longer wavelength region exhibiting an increase in particles size.

Close

Fig. 6

UV–vis spectra; different phases demonstrating rate of particle formation and aggregation of AuNPs as a function of time, Phase I: rising of primary peaks showing formation of new particles, Phase II: kinetic equilibrium attained between formation of new particles and aggregates of earliest particles, Phase III, shifting of secondary peaks at longer wavelength region indicating increase in particles size.

Export to PPT

TEM image analysis was performed periodically in our previous study which showed an increase in particles size from 8.72 ± 2.21 nm to 9.39 ± 2.12, 10.21 ± 2.04, 11.30 ± 1.91 and 12.67 ± 1.77 after day 30, day 60, day 90 and day 120 (Irfan et al., 2019). Particle size was further increased to 13.76 ± 1.92 and 14.82 ± 1.70 nm after day 150 and day 180 as shown in Fig. 7. Small nanoparticles initially formed by nucleation experience strong Van der Waals force of attraction and tend to aggregate due to presence of AuCl₄^- adsorbed on these primary nanoparticles. However, strong electrostatic and electrosteric repulsion forces caused by bio-compounds and ILs ions impart colloidal stability to AuNPs (Seol et al., 2011). This phenomenon is confirmed by TEM images which showed dispersed AuNPs without any aggregation even after six months of their preparation. The uniform dispersion of particles were remained maintain in the solution and TEM analysis even after six months showed the isolation nature of AuNPs. The slight upward movement of SPR peaks at longer wavelength could be due to change of di-electric properties of surrounding layer along with increase in particles size.

Close

Fig. 7

TEM analysis for AuNPs after (a) day 150, (b) days 180.

Export to PPT

3.6 Mechanism for growth kinetics of AuNPs

Conventional growth of particles can be divided into different stages. First stage starts with rapid formation of nuclei that immediately coalesce to form big particles in the next phase. Afterward diffusion growth of particles continues slowly through further reduction of gold precursor and coalescence with other particles (Polte et al., 2010; Polte et al., 2012). During bio-synthesis of nanoparticles, the particles size increase slowly due to massive availability of bio stabilizers present in the leaves extract. The bio-stabilizers available in the plant extract starts to shield the particles surface to control the growth of particles. In case of IL mediated synthesis of AuNPs, stability can be achieved by couple action of both bio-compounds and ILs. In addition to bio-compounds, anions present in the reaction mixture cap the surface of AuNPs while alkyl chain directed outward to give rise to steric repulsion force which is in accordance of Derjaugin-Landau-Verwey-Overbeek (DLVO) theory (Janiak, 2013). An insignificant change in particles’ size indicated by TEM analysis suggests about capping ability of bio stabilizers and IL which hinders the growth of particle for a long period of time. Protective layer offsets the Van der Waals attractive forces giving rise to electrostatic and electrosteric repulsions between adsorbed ions and counter ions impeding the aggregation of nanoparticles. Intrinsic high charge occupied by IL produced strong electrostatic forces and long alkyl chain directed outward may causes electrosteric repulsion resulting in the strong repulsive forces and causing stabilization of metal particles.

According to TEM analysis, particle size was increased incessantly although pattern was not absolute linear. Our first approach was to use Lifshitz-Slyozov-Wagner (LSW) model to explain the particle growth that explains increase of particles’ size as a function of time. Coarsening is actually dependent on the solubility of solid phase on surface of particle according to Gibbs-Thomson relation. Most of the colloidal systems are lyophobic in nature having less affinity for disperse medium that results in the rapid precipitation of particles. Gibb’s free energy reduces when particles aggregate together to form bulk material associated to their higher stability. If the particle growth is driven dominantly by diffusion growth, the particle growth takes the form of a kinetic equation also known as OR growth pattern. According to OR growth, the average diameter of nanoparticles has cube root dependence on the time and follows the equation d³ − d₀³ = 8Kt, where d and d₀ are the mean diameter at initial and after time t, respectively (Viswanatha et al., 2007). The rate constant K can be estimated by K =  $\frac{8\gamma {{\mathit{DV}}^{\mathit{2}}}_m{C}_{\infty }}{9\mathit{RT}}$ , where C_∞ is bulk solubility of gold particles, γ is surface energy of gold 0.72–1.26 J/m² (Holec et al., 2014) and V_m is the molar volume taken as 10.2 cm³/mole. D represents the Diffusion constant (Chaudhuri & Paria, 2011) and is given as $D=\frac{{\mathit{TK}}_B}{6\pi \eta a}$ where T shows the temperature of colloidal in Kelvin, K_B is Boltzmann constant, η is dynamic viscosity of solvent and a is hydrodynamic size of AuNPs. To investigate the Ostwald growth pattern, a graph was plotted between cubes of radius versus time as shown in Fig. 8. It showed a continuous increase in particles size over time, however, suggested a partial deviation from OR growth at initial stage. The time dependence r³ showed limited deviation from linearity particularly at earlier days but follows the linear relation reasonably well with time signifying the diffusion-limited growth also termed as OR. Nevertheless, it suspects the involvement of any other growth pattern which was resulted in fractional deviation particularly at initial phase. During early period, it could be considered that strong surface adsorption by anions hindered the growth of particles and decelerates the OR of nanoparticles that was resulted in small change in particle size but surface attachment mechanism was still uncertain (Dare et al., 2015; Zhang et al., 2010). On observing the particles by high resolution TEM image analysis, few quasi particles were observed which were linked each other to give a paired structure as shown in Fig. 9. The presence of these types of particles endorsed the involvement of another growth pattern named as OA. This type of growth pattern directs the formation of nanoparticles occupying irregular shapes which are not projected in precipitation-based growth. In accordance to OA, primary particles attached to a central particle to give an irregular spherical particle as were witnessed by TEM analysis. The attachment of primary particles can be affected by change of surface properties imparted by organic moieties and IL present on the surface of the particles. It is reported that during the early stage of growth when particle size is smaller there are more chances for OA growth and its time period can be changed with increase of surface adsorption (Huang et al., 2003b; Penn, 2004). When two small particles collide with each other to form secondary particles, interpretation of OA kinetic model can be illustrated as; $A_1+A_1\stackrel{k_1}{\to }B$ , where A₁ shows the primary particles and B is the secondary particles as a result of coalescence of primary particles while k₁ is reaction constant. Based on these assumptions, OA kinetic model can be deduced as (Zhang et al., 2010) $$d=\frac{d_0(\sqrt[3]{2}{k}_{1}t+1)}{(k_{1}t+1)}$$ where d₀ is initial mean diameter and d is average diameter at time t.

Close

Fig. 8

Cube radius of AuNPs as a function of time.

Export to PPT

Close

Fig. 9

(left side) HRTEM images of 4 different particles (right side) HRTEM images with lines to indicate the interfaces position.

Export to PPT

Fig. 10 represents a comparison between the mean particles’ size as a function of time based on experimental data derived through TEM analysis and theoretical data calculated using OR model. It was observed that theoretical curve predicted by using OR model seems to be well fitted with the experimental data. However, experimental curve was observed fractional below than OR curve. The discrepancy for linearity at early stage could be due to involvement of many factors such as presence of capping agent i.e. imidazolium acetate as well as other bio stabilizers present in OPK extract that slows down the particle dissolution and decreases OR ripening (Zhang et al., 2010). It could also be due to involvement of some other growth mechanism, deviations in some of the model’s approximation or partial variation in the particles sizes obtained experimentally. It was also noticed that both of these growths curves were relatively distant from each other at the early stage but bow to connect each other with the passage of time. This pattern actually endorsed the involvement of OA growth mechanism at initial stages when rate of coalescence is considered to be very high (Ribeiro et al., 2005). Theoretical particles size was also calculated to examine the involvement of OA growth pattern. Using TEM size analysis, rate of reaction was determined using first order equation i.e. ln d/d₀ = kt as shown in Fig. 11a. Derived data by OA kinetic model also predicts an increase in particle size as illustrated in Fig. 11b. However on comparing this theoretical curve with experimental curve, it showed the involvement of this growth mechanism at initial stage as shown in Fig. 10. During UV–vis analysis, it was also observed an increase in SPR absorption peak intensities up to 22 days which indicated the formation of more new particles in the solution. The kinetic energy of these particles with smaller radius is expected to be high which leads toward higher rate of collision. Therefore, these particles could adopt OA growth at expense of their higher coalescence particularly at early stage. Similar type of growth pattern was also reported in ZnS nanocrystal where initial growth was mainly contributed by OA but followed by OR at longer time (Huang et al., 2003b). Therefore, it might be suggested that OA growth got slow with time and consequently, it was replaced by OR ripening which resulted in more linear behavior of d³ with time. It could be due to the strong surface adsorption by anions which initially hinders the OR growth duration (Dare et al., 2015; Zhang et al., 2010). However, with the passage of time, capping species replaced with water molecules and resulted in higher growth rate with a major contribution by OR.

Close

Fig. 10

Comparison of actual particles size measured by TEM analysis versus theoretical particles size measured by OR and OA model.

Export to PPT

Close

Fig. 11

(a) Rate constant using first order equation for increase in particle size, (b) Particle growth predicted by OA model.

Export to PPT