Optimizing the synthesis conditions of silver nanoparticles using corn starch and their catalytic reduction of 4-nitrophenol

Regular corn starch containing approximately 74% amylopectin and 26% amylose, silver nitrate $\left( {\rm AgN}{{{\rm O}}_{3}} \right)$ (>$99.9\%$), sodium hydroxide (NaOH), sodium borohydride (${\rm NaB}{{{\rm H}}_{4}}$) and 4-nitrophenol (4-NP) were purchased from Sigma–Aldrich (St. Louis, USA). The glassware was washed with ${\rm HN}{{{\rm O}}_{3}}/{\rm HCl}$ (3:1) mixture and subsequently with highly purified distilled water $\left(18\,{\rm M}\Omega {\rm m} \right)$.

Weighted amount (1 g) of CS powder was dissolved in 100 ml of distilled water to prepare 1% (wt%) solution in a hot water ($>50{{\,}^{\circ }}{\rm C}$) with stirring at certain intervals to ensure complete dissolution of CS. The different concentrations of ${\rm AgN}{{{\rm O}}_{3}}$ (0.1, 1, 10 and 100 mmol l⁻¹) solution were prepared. The efficiency and the synthesis of SNPs in the biopolymer matrix were investigated under various experimental conditions by varying the ${\rm AgN}{{{\rm O}}_{3}}$ concentration, changing the temperature of the reaction and changing the reaction time for the reactants. These factors possibly play an important role in the fabrication of NPs and affect the size, shape and particle distribution.

The absorption spectra of the prepared samples were measured by UV-visible (UV-Vis) using a spectrophotometer (Thermo Scientific Evolution 201 UV-vis, USA) that scanned in a wavelength range of $200-900\,{\rm nm}$. Transmission electron microscopy (TEM) images were taken using JEOL TEM-210 model, which operates at 120 kV to study the size, morphology, and shape of the synthesised SNPs. Elemental composition analysis of SNP samples was carried out using an SEM (JOEL Model JED-2300, USA) supplied with an energy dispersive spectrometer detector. The x-ray diffraction (XRD) measurement was performed on x-ray diffractometer (Panalytical X'Pert Pro 3050/60) working at 30 kV and 100 mA and spectrum was recorded by Cu Kα radiation $\left(\lambda =1.5406\,{{\rm }\mathring{\rm A} } \right)$ in the range $2\theta ={{20}^{\circ }}$ to ${{80}^{\circ }}$. Fourier transform infrared (FTIR) transmittance spectra of silver/starch nanocomposite powder were recorded at room temperature, using FT-IR spectrometer (Thermo Fisher Nicolet iS10, USA) in the range of $4000-400\,{\rm c}{{{\rm m}}^{-1}}$ and using KBr pellets. Zeta potential measurements were carried out using Malvern instruments.

The catalytic activity of CS-SNPs was investigated by examining the reduction of 4-NP in 4-aminophenol (4-AP) in the presence of ice cold freshly prepared ${\rm NaB}{{{\rm H}}_{4}}$ which acts as the reducing agent [12, 14, 30]. To estimate (assess) the catalytic reduction of 4-NP, 2 ml of 4-NP $\left(0.06\times {{10}^{-3}}\,{\rm M} \right)$ was added to $0.5\,{\rm ml}$ of ${\rm NaB}{{{\rm H}}_{4}}$ ($0.06\,{\rm M}$) in a quartz cuvette of 1 cm path length. This was followed by the addition of $300\,\mu {\rm l}$ of CS-SNP different silver/starch nanocomposite solution samples and mixed well quickly. The reaction was monitored by UV-vis spectrophotometer by recording the absorption spectra. All experiments were carried out for various silver/starch nanocomposite samples under ambient conditions at room temperature $\left(25{{\,}^{\circ }}{\rm C} \right)$.

3.1.1. Effect of silver nitrate concentration and temperature on the synthesis of silver nanoparticles.

For the synthesis of SNPs, 25 ml of the previously prepared CS colloidal solution ($1\%$) was mixed with 5 ml of ${\rm AgN}{{{\rm O}}_{3}}$ (0.1, 1, 10 and 100 mmol l⁻¹) in a basic pH medium for certain reaction time. Figure 1(a) shows the ultraviolet-visible (UV-vis) spectra of the absorbance intensity against various ${\rm AgN}{{{\rm O}}_{3}}$ concentrations. At 0.1 mmol l⁻¹, there was no observable localised surface plasmon resonance (LSPR) peak, which is a characteristic peak that occurs when the electrons in the nanoparticles interact with the electromagnetic radiation. In general, the noble metal nanoparticles such as silver and gold afford strong extinction and scattering spectra due to such property [31].

Zoom In Zoom Out Reset image size

According to the Mie theory, the position of the UV-vis spectrum absorption peak directly depends on the size of the synthesised nanoparticles as follows:

$$\begin{align} \newcommand{\e}{{\rm e}} \displaystyle I={{I}_{0}}\left(\frac{1+{\rm co}{{{\rm s}}^{2}}\theta }{2{{R}^{2}}} \right){{\left(\frac{2\pi }{\lambda } \right)}^{4}}\left(\frac{{{n}^{2}}-1}{{{n}^{2}}+1} \right){{\left(\frac{d}{2} \right)}^{6}},\nonumber \end{align} \tag{ 1 }$$

where $R$ is the distance between the particles and the observer, $d$ is the diameter of the particles, $\theta$ is the scattering angle and n is the refractive index of the particles [31–33].

The absorbance intensity increased with the increasing ${\rm AgN}{{{\rm O}}_{3}}$ concentrations (1, 10, and 100 mmol l⁻¹). The LSPR peak at the maximum wavelength (${{\lambda }_{{\rm max}}}$) for 1 mmol l⁻¹ and 10 mmol l⁻¹ was 414 and 405 nm, respectively, whereas a slight red shift of the LSPR peak (${{\lambda }_{{\rm max}}}=418\,{\rm nm}$) was detected at 100 mmol l⁻¹, possibly due to the formation of the large particles that may lead to the agglomeration and/or aggregation of the particles due to the supersaturation of the ${\rm AgN}{{{\rm O}}_{3}}$ concentration [34].

Figure 1(b) shows the UV-vis absorption spectra of SNPs capped by CS at different temperatures (50, 60, 70, 80 and $90{{\,}^{\circ }}{\rm C}$) and higher pH for a certain interval. The absorption intensity increased with the increasing reaction temperature. Furthermore, the colour of the solutions gradually changed with a fixed LSPR peak at ${{\lambda }_{{\rm max}}}=412\,{\rm nm}$. However, at $90{{\,}^{\circ }}{\rm C}$, there was a blue shift of the LSPR peak (${{\lambda }_{{\rm max}}}=405\,{\rm nm}$), indicating the formation of small nanoparticles [23, 35, 36]. Also, the colour change of the synthesised SNP samples at different temperatures was observed in the inset in figure 1(a)(i)–(v)). Hence, the gradual change of the colour, indicates the formation of the SNPs due to the reduction of ${\rm A}{{{\rm g}}^{+}}$ ions. Since, it is known that polysaccharides such as CS could be hydrolyzed by an amylase enzyme or by heat treatment at high temperature of $80{{\,}^{\circ }}{\rm C}$ or higher [31]. Since the increase of the reaction temperature from $50{{\,}^{\circ }}{\rm C}$ to $90{{\,}^{\circ }}{\rm C}$, that will enhance the amylose crystalline helical structure content to leach from the amorphous amylopectin of CS molecules [27, 37]. Whereas, this reaction could not complete without the heat supply. Also, the reaction will be optimized at the very high alkaline basic medium (more than pH 9) [38, 39].

3.1.2. Effect of the pH of the reaction medium on the synthesis of silver nanoparticles.

To demonstrate the role of different pH on the synthesis of SNPs, 25 ml of corn starch solution was mixed with 5 ml of ${\rm AgN}{{{\rm O}}_{3}}$ (optimum suitable concentration), which was added drop wise at a hot plate at $90{{\,}^{\circ }}{\rm C}$ with continuous stirring. Simultaneously, the pH of the mixture was adjusted to 3, 5, 7, 9 and 11 with a few drops of a freshly prepared 0.1 N HCl or 0.1 N NaOH solution.

According to the UV-vis absorption spectrum shown in figure 2(a), there was no change in the solution colour at low pH (3–7) compared to the blank sample (pure CS), indicating that SNPs were not formed. However, by increasing the pH (9–11), both the colour and the LSPR peak of the mixtures changed (figure 2(a), inset (i)–(v)) from faint yellow and ${{\lambda }_{{\rm max}}}=410\,{\rm nm}$ (pH 9) to deep yellow colour and ${{\lambda }_{{\rm max}}}=405\,{\rm nm}$ (pH 11), indicating that the SNPs were formed (figure 2(b)). It was reported by $U$. Uthumporn et al [31] that, the gelatinization process of corn CS was achieved at $86.2{{\,}^{\circ }}{\rm C}$, therefore the heating of the CS at $90{{\,}^{\circ }}{\rm C}$ could enhance and improve the hydrolysis of CS amylose content to maltose molecules. Then, after the addition of NaOH which dissociates into sodium ions (${\rm N}{{{\rm a}}^{+}}$) and hydroxyl ions (${\rm O}{{{\rm H}}^{-}}$), these ${\rm O}{{{\rm H}}^{-}}$ ions oxidise the CS molecules and generate heat. This will be followed by the dissociation of oxidized CS molecules into glucose units, whereas this reaction is enhanced in the presence of both high temperature (more than $80{{\,}^{\circ }}{\rm C}$) and very high alkaline basic reaction medium (i.e. pH 11). The reduction of Ag⁺ ions was done by the glucose units which contain a carbonyl group that named aldehyde, that is responsible for the reduction of ${\rm A}{{{\rm g}}^{+}}$ ions to SNPs (${\rm A}{{{\rm g}}^{0}}$) [38]. This is consistent with the mechanism proposed by Shervani and Yamamto [25], where Ag₂O molecules dissociate into silver ions (${\rm A}{{{\rm g}}^{+}}$), which subsequently get reduced to SNPs (${\rm A}{{{\rm g}}^{0}}$) in the presences of the electron pairs produced from CS fragment oxidation. In addition, the synthesised NPs could be stabilised by the helical amylopectin residue of CS molecules as shown in the proposed reaction mechanism in figure 3 forming Ag/CS nanocomposite. Hence, the CS acts not only as a reducing agent in an alkaline medium, but also as a capping (stabilising) agent.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

3.1.3. Effect of reaction time on the synthesis of silver nanoparticles.

The effect of the reaction time on the synthesis of SNPs, the reaction was carried out at different reaction time durations (15, 30, 45, 60 and 90 min). Then, the samples from the reaction medium were withdrawn at different time intervals, followed by recording the UV-vis absorption spectra of the fabricated SNPs. Figure 4(a) shows a gradual increase in the absorption intensity at different time intervals; the LSPR peak was observed at ${{\lambda }_{{\rm max}}}=405\,{\rm nm}$. Furthermore, the calculation of the full width at half maximum (FWHM) of the recording data indicates that the FWHM gradually decreases with the increasing reaction time. In addition, the FWHM value at the lowest reaction time interval (15 min) was 76 nm, whereas it was 110 nm at the highest reaction time interval (i.e. 90 min) [figure 4(b)]. This could be correlated to the dependence of SNP synthesis on the reaction time interval [40]. However, using NaOH to elevate the pH of the medium to 11 accelerates the reduction of silver salt under different conditions and promotes the capping of the synthesised SNPs by CS fragments. Therefore, the reaction is carried out under different physical conditions to obtain smaller SNPs capped by a biopolymeric CS matrix [41, 42].

Zoom In Zoom Out Reset image size

High resolution (HR) transmission electron microscopy (TEM) measurements showed that the synthesised SNPs capped by corn starch were bigger in size ($9-90\,{\rm nm}$; average particle diameter:$9.71\,{\rm nm}$) at pH 7 (figure 5), while at high pH (9 and 11), the SNPs were smaller in size (average particle size diameter: 8.9 and 7.68 nm, respectively). The spherical particles and the particle size distribution were relatively homogeneous for alkaline pH (pH 11), which is inconsistent with the UV-vis spectra [42]. Moreover, the selected area electron diffraction (SAED) image of the sample shows the concentric circles of diffraction pattern indicating that the obtained SNPs is highly crystalline [figure 5(i)]. In addition to, the HR-TEM images ascribes the space lattice of the metallic SNPs with 0.19 and 0.23 nm as shown in figure 5(j). Therefore, the TEM images confirm that the synthesised SNPs are metallic in nature.

Zoom In Zoom Out Reset image size

X-ray diffraction (XRD) pattern depicted three distinct, intense peaks, in the diffraction spectrum of 2θ, ranging from $20{{\,}^{\circ }}$ to $80{{\,}^{\circ }}$. XRD pattern of SNPs stabilised by CS (10 mmol l⁻¹) was characterised by three distinct, intense peaks at $38.30{{\,}^{\circ }}$, $43.55{{\,}^{\circ }}$ and $77.50{{\,}^{\circ }}$, representing the characteristic diffraction peaks of SNPs that were indexed to (111), (220) and (311) crystalline planes of the face centred cubic crystalline structure of metallic silver, respectively [figure 6(a)]. These results are inconsistent with an XRD spectrum of pure crystalline silver structure that has been published by the joint committee on powder diffraction standard (JCPDS file No. 00-004-0783) [43]. Furthermore, the average particle size can be estimated using the Scherrer formula:

$$\begin{align} \newcommand{\e}{{\rm e}} \displaystyle D=\frac{k\lambda }{\beta {\rm cos}\theta },\nonumber \end{align} \tag{ 2 }$$

where $D$ is the particle diameter (nm), $k$ is a constant (a shape factor) that equals to 0.9, $\lambda$ is the wavelength of the x-ray diffraction source (0.1541 nm), $\beta$ is the full width at half maximum and $\theta$ is the half diffraction angle (in degrees). Therefore, the average particle size diameter was calculated from the most predominant intense diffraction peak at $2\theta =38.30{{\,}^{\circ }}$ [44]. Thus, the calculated average particle size diameter was 7.31 nm, which is very close to that obtained from the TEM histogram distribution [45].

Zoom In Zoom Out Reset image size

Figure 6(b) shows a characteristic peak around 3.0 keV, which corresponds to the binding energy of ${\rm A}{{{\rm g}}_{L}}$, and other peaks near $1.0\,{\rm keV}$, which correlate to the presence of C, O and Na impurities. The EDX spectrum confirms the presence of SNPs [46, 47].

The structural changes due to the interaction between the starch molecules and synthesised SNPs were investigated by using Fourier transformation infrared (FT-IR) spectra. Figure 7(a) shows the FT-IR absorption spectra of native starch and SNP/starch nanocomposite. A strong absorption band at $3445\,{\rm c}{{{\rm m}}^{{{\rm {\text -}}1}}}$ was attributed to the O-H stretching of starch. In addition, a sharp band of asymmetric stretching of the C-H band was assigned at $2925\,{\rm c}{{{\rm m}}^{-{\rm 1}}}$. Moreover, a strong band observed at $1645\,{\rm c}{{{\rm m}}^{-{\rm 1}}}$ was attributed to the water adsorbed in the amorphous amylose region of starch [48]. The characteristic peaks at 1085 and $1025\,{\rm c}{{{\rm m}}^{-1}}$ were due to the C-O-H bending of starch molecules and anhydroglucose ring O-C stretch, respectively [49]. The shift at $3445$ and $1645\,{\rm c}{{{\rm m}}^{-{\rm 1}}}$ bands to a lower wavelength at $3404$ and $1596\,{\rm c}{{{\rm m}}^{-{\rm 1}}}$ in the presence of SNPs was possibly due to the interaction of SNPs with the OH groups of SNP/CS nanocomposites. Furthermore, the other two characteristic bands at 1085 and $1025\,{\rm c}{{{\rm m}}^{-{\rm 1}}}$ shifted to 1076 and $1020\,{\rm c}{{{\rm m}}^{-{\rm 1}}}$, respectively, due to the coordination between the oxygen atom of the starch OH polar groups and Ag ions [47].

Zoom In Zoom Out Reset image size

Figure 7(b) represents the ZP measurements of SNPs stabilised by corn starch. The ZP of the SNPs was $-23\,{\rm mV}$. The negative ZP of the fabricated SNPs is due to the OH group of the capping agent (CS). The current result is inconsistent with that reported in the literature, where the ZP of the measured sample solution was above $+20\,{\rm mV}$ or below $-20\,{\rm mV}$, which were considered stable [47, 50, 51].

To investigate the catalytic reduction of 4-NP to 4-AP in the presence of excessive ${\rm NaB}{{{\rm H}}_{4}}$, the reaction progress was monitored by a UV-vis spectrophotometer. 4-NP (2 ml; $0.06\times {{10}^{-3}}\,{\rm M}$) with a yellow colour showed absorption peaks at $227\,{\rm nm}$ and $317\,{\rm nm}$ [figure 8(a)] due to $\pi \to {{\pi }^{*}}$ (from the ring of phenol) and $n\to {{\pi }^{*}}$ (from a lone pair of electron in the oxygen and nitrogen atoms), respectively [52]. Hence, after the addition of ${\rm NaB}{{{\rm H}}_{4}}$, the colour of the reaction mixture changed to yellow green and the absorption peak was shifted to 400 nm due to the formation of 4-NP ions as shown in figure 8(a). This reaction could stay unchanged for several hours due to the presence of  −NO₂ group of 4-NP aromatic compounds, which is inert to the reduction of ${\rm NaB}{{{\rm H}}_{4}}$.

Zoom In Zoom Out Reset image size

Moreover, the mutually repelling negative ions 4-NP and ${\rm BH}_{4}^{-}$ suggest that this reaction cannot proceed at all [5]. This is a probe catalytic reaction model, which is thermodynamically preferred (${{E}_{0}}$ of 4-NP/4-AP is $-0.76\,{\rm V}$). The large potential barrier between the donor ${\rm BH}_{4}^{-}$ and acceptor [4-NP] molecules decreases the feasibility of this reaction. To overcome this kinetic potential barrier, the addition of metallic NPs simplifies the catalytic reduction of 4-NP to 4-AP. The removal of 4-NP, which is one of the most carcinogenic, toxic compounds, is the primary target. The intermediate product 4-AP has medical and industrial applications such as paracetamol and dye production. The addition of 0.3 ml of fabricated SNPs, at optimized synthesis conditions (i.e. 1 mmol l⁻¹ ${\rm AgN}{{{\rm O}}_{3}}$ in the presence of CS (10 g l⁻¹) for 15 min at $90{{\,}^{\circ }}{\rm C}$), decreased the absorption characteristic peak at 400 nm, fading with time depending on the size of the nanocatalyst at different pH (7, 9 and 11), as shown in figures 8(b)–(d); a new growing peak was assigned at $293\,{\rm nm}$. The rate of catalytic reduction was accelerated by elevating the reaction medium pH. The reaction time was approximately 16 min for silver/starch nanocomposite that was synthesised at neutral and alkaline medium with pH 7, whereas it was 9 min and 8 min for the preparation medium of the reaction at pH 9 and 11, respectively. This may be due to the small size of the synthesised SNPs, confirmed by HR-TEM images, representing a higher surface-volume ratio that eases the catalytic reduction by facilitating the transfer of electrons from the donor ion to the acceptor ion as shown in figure 9 [53]. In order to evaluate quantitatively, the catalytic activities of the fabricated silver nanocatalyst, the following pseudo first-order rate kinetic reaction was used as the model reaction:

$$\begin{align} \newcommand{\e}{{\rm e}} \displaystyle \frac{[A]}{[{{A}_{0}}]}=-Kt,\nonumber \end{align} \tag{ 3 }$$

where $K$ is the pseudo first-order rate constant, $t$ is the reaction time, $A$ is the absorbance at a certain time and ${{A}_{0}}$ is the absorbance at time $t=0$.

Zoom In Zoom Out Reset image size

The absorbance can be estimated from the absorbance peak at 400 nm. Figure 10 shows the graphical representation of ${\rm ln}[A]$ against the reaction time ($t$), where the rate constant ($K$) can be estimated from the mentioned relationship. Hence, the rate constant for the silver/starch nanocomposite prepared at pH 7 was $9.333\times {{10}^{-4}}\,{{{\rm s}}^{-1}}$, whereas it was $7.167\times {{10}^{-4}}\,{{{\rm s}}^{-1}}$ and $5.333\times {{10}^{-4}}\,{{{\rm s}}^{-1}}$ for SNPs synthesised at pH 9 and 11, respectively. This result confirms that the smaller the used NPs, the faster the 4-NP catalytic reduction rate (table 1).

Table 1. Catalytic activity of SNPs for the reduction of 4-NP.

Reaction medium pH	Reaction time (min)	Ka (s)	Correlation coefficient $\left({{R}^{2}} \right)$
7	16	$9.333\times {{10}^{-4}}$	0.9465
9	9	$7.167\times {{10}^{-4}}$	0.9253
11	8	$5.333\times {{10}^{-4}}$	0.9574

Zoom In Zoom Out Reset image size