Optimized green synthesis of silver nanoparticles from guarana seed skin extract with antibacterial potential

2.1 Materials

Silver nitrate (Sigma Aldrich, purity > 99%), sodium hydroxide (Sigma Aldrich, purity > 99%), hydrochloric acid (Sigma Aldrich, purity > 99%), and distilled water were used throughout the synthesis reactions. For antibacterial characterization, Mueller–Hinton agar and broth were utilized along with Staphylococcus aureus (CCCD-S009) and Escherichia coli (CCCD-CC01).

2.2 Taguchi design

In this experiment, nine runs were conducted, selecting four factors – pH, temperature, guarana seed skin extract concentration, and silver nitrate concentration – each with three levels, as detailed in Table 1. The particle size of AgNPs was chosen as the response variable.

A Taguchi orthogonal array L9 (34) was used with an S/N ratio optimized for the “smaller is better” criterion, aiming to achieve the smallest possible AgNP size, as shown in Table 2.

2.3 Statistical analysis

A 3D surface plot was generated using Statistica software to visualize the interactions between factors. Additionally, an analysis of variance (ANOVA) was performed at a 90% confidence level (p > 0.1) to assess the statistical significance of each factor’s effect on the response variable (AgNP size). The ANOVA helped identify the key parameters influencing the AgNP size. This analysis also included calculating the signal-to-noise (S/N) ratio for each experimental run, following the “smaller is better” criterion, using Minitab 17 software (Minitab Inc., Pennsylvania), which helped minimize variability while optimizing for the smallest AgNP size.

2.4 Preparation of guarana seed skin extract

Guarana seed skins (Paullinia cupana) were collected in Maués, a region near Manaus (Amazonas, Brazil). The seed skins were separated from the seeds, washed three times with distilled water to remove impurities, and sun-dried for 12 h. The dried material was then stored in a humidity-free environment. To reduce the particle size, the guarana seed skins were ground using dry ball milling for 1 h. Based on parameters predicted from the Taguchi design, different guarana seed skin extracts (3%, 5%, and 7% w/v) were prepared by adding guarana seed skin to 100 mL of water preheated to 60°C, using the infusion method for 20 min. The aqueous solution was filtered with Whatman No. 1 filter paper, and the extracts were stored for subsequent analyses.

2.5 Synthesis of AgNPs

The synthesis of AgNPs was conducted based on nine experimental runs, as described by the Taguchi design (Table 1). Initially, guarana seed skin extracts were prepared in varying concentrations (3%, 5%, and 7% w/v) and added dropwise under vigorous agitation to aqueous silver nitrate solutions (1, 5, and 10 mM) in a 1:1 ratio (guarana seed extract concentration to silver nitrate solution concentration) on a magnetic stirrer. The temperature (40°C, 50°C, or 60°C) and pH (5, 7, and 11) were adjusted according to the experimental setup for each run.

The reduction reaction for AgNP formation was maintained for 24 h. To prevent the photoactivation of silver nitrate, the reaction vessels were wrapped in aluminum foil. After synthesis, the AgNPs were centrifuged at 6,000 rpm for 20 min. The resulting precipitate was washed three times with deionized water and subsequently lyophilized. The dried material was then stored for further analysis.

2.6 Chemical analysis of guarana seed skin

Quantitative chemical analyses of 20 g of guarana seed skin were conducted using X-ray fluorescence (PW 2400, Philips), equipped with a 3 kW tube and rhodium target. Loss on ignition was determined by burning the samples for 1 h at 1,000°C.

For further analysis, liquid chromatography coupled with mass spectrometry (LC/MS) was performed using an LCQ Fleet system (Thermo Scientific) with electrospray ionization in positive mode at an energy of 20–35 eV. The mass-to-charge (m/z) range analyzed was 100–600 Da. Extracts were prepared in methanol (HPLC grade) at a concentration of 10 ppm, and data were processed using Xcalibur 2.0.7 software.

2.7 Characterization of AgNPs

The concentration of AgNPs was measured using inductively coupled plasma optical emission spectrometry (ICP-OES, ICPE-9820, Shimadzu) in triplicate at 25 ± 1°C. AgNP samples were diluted to a 1:20 mass ratio in water, and the colloidal suspension was kept in the dark to prevent oxidation.

The biochemical properties of AgNPs were assessed by ultraviolet-visible spectroscopy (UV-Vis, UV-1800, Shimadzu) in the range of 190–700 cm⁻¹, identifying the presence of AgNPs, extract compounds, and their dispersity. Particle size distribution and hydrodynamic size were analyzed through dynamic light scattering (DLS, Zetasizer Nanosizer, Malvern), with samples dispersed in distilled water and sonicated for 1 h before analysis.

Fourier-transform infrared spectroscopy (FTIR, Cary 630 Agilent) was used to identify compounds in the guarana extract responsible for reducing and stabilizing AgNPs. The guarana seed skin extract was analyzed in diffuse reflectance mode from 500 to 400 cm⁻¹ with an 8 cm⁻¹ resolution at 25 ± 1°C for comparison.

The crystallinity and crystallite size of lyophilized AgNPs were evaluated by X-ray diffractometry (XRD, D2 Phaser, Bruker). Diffractograms were compared to Joint Committee on Powder Diffraction Standards (JCPDS) patterns using X’pert HighScore Plus software. The average crystalline size was calculated using the Debye–Scherrer equation (Eq. 1)

where D is the average crystalline size, k is a geometric factor (0.9), λ is the X-ray wavelength (0.15418 nm), and β is the full width at half-maximum of the XRD peak at the diffraction angle θ.

Scanning electron microscopy (SEM, JSM IT500HR, Jeol) was used to examine the morphology of the biosynthesized AgNPs, providing details on internal structure. Lyophilized AgNPs were coated with platinum for SEM analysis. Transmission electron microscopy (TEM, 1400 Flash, Jeol) was conducted on the supernatant from the third wash; samples were dried for 8 h and mounted on a copper grid with formvar paper.

Additionally, X-ray fluorescence (XRF, model PW 2400, Philips) was employed to determine oxide composition in guarana seed skin. Approximately 20 g of seed skin was analyzed using a 3 kW tube with a rhodium target.

2.8 Antibacterial activity assay

The antibacterial activity of AgNPs was tested against Gram-positive Staphylococcus aureus (CCCD-S009) and Gram-negative Escherichia coli (CCCD-E005) following the Clinical Laboratory Standards Institute guidelines [28]. Bacterial suspensions were prepared in 0.85% saline solution to reach a 0.5 McFarland standard. Using a sterile swab, these suspensions were seeded onto Müeller–Hinton agar plates. Wells were created on the agar surface, and 100 µL of the AgNP solution was added to each well. Negative controls included the guarana seed skin extract without nanoparticles, while levofloxacin (250 µg·mL⁻¹) served as a positive control for both bacteria. The plates were sealed and incubated at 37°C for 24 h. Following incubation, inhibition zones were observed and measured using a caliper. Each experiment was repeated three times, and the average zone of inhibition was calculated.

The minimum inhibitory concentration (MIC) was determined using a 96-well microtiter plate by the broth microdilution method based on the reduction of resazurin (7-hydroxy-3H-phenoxazin-3-one 10-oxide). In this method, the color change of resazurin from blue (indicating absence of growth) to pink (indicating bacterial growth) allows for MIC determination. Commercially sourced strains (Cefar Diagnóstica Ltd) were used in this assay. For each bacterium, 100 µL of microbial inoculum at 5 × 10⁵ CFU·mL⁻¹ was mixed with 100 µL of AgNPs in triplicate within a sterile 96-well microplate. Positive controls included levofloxacin (250 μg·mL⁻¹), while negative controls used only the microbial inoculum and guarana extract. Sterility control wells contained 100 µL of sterile Mueller–Hinton medium.

Following initial incubation at 37°C for 24 h, 30 µL of resazurin was added to each well. Plates were incubated again at 37°C for 2 h to observe color changes. In this assay, a pink color indicates bacterial growth (no antimicrobial effect), while a blue color indicates inhibition of bacterial growth.

3.1 Optimization of AgNP synthesis

According to the Taguchi design, nine experiments were conducted in random order under the conditions outlined in Tables 1 and 2. Previous studies have demonstrated that smaller AgNPs exhibit enhanced antibacterial activity [29,30,31]. Therefore, the average size of AgNPs was analyzed as the response variable, with results summarized in Table 3.

Main effects plots were used to explore the relationships between the factors (pH, reaction temperature, guarana seed skin extract concentration, and silver nitrate concentration) and the response variable (average AgNP size), as illustrated in Figure 1. The interpretation of the S/N ratio plot was observed to be opposite to the main effects plot. The optimal level for each factor was determined by selecting the highest S/N ratio, which corresponds to increased robustness and improved process performance.

Figure 1

Influence of single factors on the average size of AgNPs and respective S/N ratio: (a) concentration of silver nitrate (C _AgNO3), (b) temperature of the reaction, (c) pH, and (d) concentration of guarana seed skins extract (C _g).

A decrease in silver nitrate concentration from 13.6 mM (Level 1) to 12.1 mM (Level 2) resulted in a smaller average AgNP size, likely due to an imbalance between the reactant and reducing agent. However, as the AgNO₃ concentration increased from 5 mM (Level 2) to 10 mM (Level 3), the average AgNP size increased from 12.1 to 40.1 nm. This increase may be attributed to a higher availability of Ag⁺ ions at elevated AgNO₃ concentrations, which accelerates the bioreduction reaction. Similar findings were reported by Tuzun et al. [32], where AgNPs synthesized with the Anthemis Tricolor Boiss aqueous extract showed hydrodynamic sizes of approximately 442 nm at 1 mM silver nitrate and 515 nm at 3 mM.

The concentration of plant extract is a significant factor in AgNP synthesis [25,33]. Higher extract concentrations can lead to the formation of larger particle clusters, likely due to increased extract concentration from 1% (Level 1) to 10% (Level 3). In previous research, the green synthesis of AgNPs using Plantago major L. leaf ethanolic extract also yielded larger nanoparticles with increased extract concentrations [34]. Zulfiqar et al. [33] suggested that higher plant extract amounts can accelerate reduction rates, potentially causing uncontrolled nucleation and the formation of larger AgNPs.

AgNP size has been shown to be temperature-dependent, as the synthesis kinetics are influenced by the reaction temperature [7,35]. In this study, AgNP size decreased from 27.4 nm at 40°C (Level 1) to 12.4 nm at 50°C (Level 2), likely due to insufficient precursor availability at lower temperatures. Soto et al. [36] reported similar findings, observing that higher synthesis temperatures (80°C) produced smaller AgNPs compared to lower temperatures (20°C). Rapid reduction rates at higher temperatures are thought to favor homogeneous nucleation, yielding smaller AgNPs [37]. In contrast, when the temperature increased from 50°C (Level 2) to 60°C (Level 3), AgNP size increased from 12.4 to 26.2 nm. This may be due to increased molecular kinetic energy at higher temperatures, leading to faster consumption of silver ions and larger particle sizes [38].

pH is also a crucial factor influencing the shape, morphology, and size of AgNPs [3,33]. In this study, as pH increased from 5 (Level 1) to 11 (Level 3), the average AgNP size decreased from 34.8 to 7.23 nm. Under acidic conditions, the nucleation process is slower, and stabilization during AgNP synthesis is less effective, resulting in larger nanoparticles [36]. In contrast, alkaline conditions favor rapid nucleation, likely due to the availability of functional groups in guarana seed skin, such as alkaloids (e.g., caffeine, theophylline, and theobromine), which enhance stabilization.

3.2 ANOVA

ANOVA is a robust statistical tool within Taguchi design that evaluates the effect of each parameter on overall variability in the results. The sequential sums of squares (Seq SS) represent the variation for each component of the model, while adjusted mean squares (Adj MS) indicate the amount of variation a term explains, assuming all other terms are present in the model regardless of entry order. The contribution percentage in the ANOVA table quantifies each factor’s impact on total Seq SS. The p-value provides a probability measure against the null hypothesis, with lower values indicating stronger evidence. Degrees of freedom (DF) represent the data’s informational content, determined by the sample’s number of observations; here, DF = 4. The ANOVA results for factors influencing the average AgNP size are summarized in Table 4.

When comparing factor contributions, pH was found to have the highest impact at 34.02%, followed closely by silver nitrate concentration at 33.61%. Conversely, guarana seed skin extract concentration and temperature showed minimal effects, contributing only 7.61% and 0.06%, respectively. The results also indicated a notable margin of error (24.70%) attributed to noise in particle size measurements. The Pareto chart in Figure 2 highlights two significant factors – pH and silver nitrate concentration – while the concentration and temperature of guarana seed skin extract exhibited limited impact. This may result from the selected levels of guarana seed skin extract and temperature being too close, causing variations insufficient to significantly affect AgNP size.

Figure 2

Pareto chart of the standardized effects (response in size [nm] ∝ = 0.1) for pH, concentration of silver nitrate (C _AgNO3), concentration of guarana seed skins extract (C _g), and temperature.

A three-dimensional (3D) surface plot (Figure 3) illustrates the combined influence of silver nitrate concentration (CAgNO₃), temperature, pH, and guarana seed skin extract concentration (Cg) on the size.

Figure 3

3D response surface plots showing the effects of temperature, pH, C _g, and C _AgNO3 on the average size of AgNPs. Four pairs of parameters were investigated: (a) Temperature and pH, (b) C _g and pH, (c) C _AgNO3 and C _g, and (d) C _AgNO3 and pH.

Based on the S/N ratio and ANOVA results, the optimized conditions were identified as follows: AgNO₃ concentration at 0.005 M (Level 2), pH at 11 (Level 3), reaction temperature of 50°C (Level 2), and guarana seed skin extract concentration of 1% (m/v) (Level 1). This condition was predicted to produce the smallest AgNP size, and thus, the corresponding sample was selected for chemical characterization and antibacterial testing.

3.3 Chemical composition of guarana seed skin extract

The LC/MS fingerprint of guarana seed skin extract was obtained in positive mode within a mass range of 100–600 m/z (Figure 4), allowing for the investigation of its chemical profile. This analysis identified caffeine as a possible major compound in the aqueous extract. Barreto et al. [39] reported optimized extraction conditions for bioactive compounds in guarana seeds, detecting high caffeine levels of 56.8 mg·g⁻¹ using ultrasound-assisted extraction.

Figure 4

LC/MS fingerprints of guarana seed skin extract.

In addition to the prominent peak at m/z 195, several other ions (e.g., m/z 170, 317, and 365) with high relative intensity were observed in the guarana seed skin extract, suggesting a complex matrix of alkaloid compounds. To further clarify the composition, auxiliary techniques such as HPLC and nuclear magnetic resonance (NMR) are recommended to precisely identify the compounds present.

The LC/MS spectrum of the ion at m/z 195 displayed a base peak at m/z 138, along with less intense ions at m/z 163, 151, and 110 (Figure 5), consistent with the structure of caffeine [40,41]. The spectrum of protonated caffeine at m/z 195 revealed an initial loss of 57 Da (O–CNCH₃) due to retro-Diels–Alder rearrangement, resulting in a base peak at m/z 163, with subsequent losses of 32 Da (CH₃CH═O), 44 Da (CH₃CH–O), and CO (m/z 138–110). This study is the first to report caffeine as part of the chemical composition in guarana seed skin extract.

Figure 5

Proposed products from the fragmentation of caffeine.

The spectrum of protonated caffeine at íon m/z 195 displayed the initial loss of 57 Da (O–CNCH₃) by retro-Diels–Alder rearrangement, and a base peak at m/z 163 showed a neutral loss of 32 Da (CH₃CH═O). Moreover, a loss at m/z 151 of 44 Da (CH₃CH–O) and, lastly, the loss of carbon monoxide (CO) (m/z 138 → 110) could be identified. Thus, this is the first report of the caffeine present in the chemical composition of the seed skin guarana extract.

XRF was conducted to identify and quantify oxides present in guarana seed skin, a method that provides insights into the metal content contributing to the extract’s composition. The main oxides identified are summarized in Table 5, with potassium oxide (K₂O), calcium oxide (CaO), and silicon dioxide (SiO₂) being the most abundant at 0.918%, 0.487%, and 0.478%, respectively. A high loss on ignition of 97.417% was observed, likely due to the combustion of organic matter in the seed skin.

These results are consistent with findings by Adolfo et al. [42], who detected Fe and Ni in guarana, with contents of 1.004 and 0.022 μg·g⁻¹, respectively. Variations in oxide content across samples may be influenced by factors such as the region of guarana collection, environmental stressors, and biological variables.

Understanding the chemical composition of guarana seed skin is crucial for identifying compounds that interact with AgNPs in further analyses and for hypothesizing the potential mechanisms between seed skin compounds and biogenic AgNPs.

3.4 Characterization of selected AgNPs

AgNPs were successfully synthesized using guarana seed skin extract as an efficient reducing and capping agent. UV-Vis spectroscopy (Figure 6) confirmed the formation of AgNPs in the colloidal solution, as evidenced by an absorption band in the 400–500 nm range, which intensified with longer reaction times.

Figure 6

Absorption spectra of AgNPs observed in 1 h, 24 h, and 30 days of reaction.

In the UV-vis absorption spectra, the guarana seed skin extract shows two distinct absorption bands. A band at 190–250 nm is associated with the cinnamoyl system, while a strong band observed at 250–290 nm is likely attributed to caffeine [43,44]. Previous studies reported the presence of caffeine in coffee samples analyzed by UV-vis spectroscopy, showing absorption bands at λ_max 275 and 330 nm [45]. A reduction in the intensity of the guarana seed skin extract absorption peak was also observed, likely due to its involvement in the reduction of silver ions to AgNPs.

In the first hour of the reaction (Figure 6), no absorption band characteristic of AgNPs (400–500 nm) is visible. After 24 h, an SPR (surface plasmonic resonance) absorption peak is present between 350 and 500 nm, indicating the formation of stabilized AgNPs in the colloidal dispersion. After 30 days (Figure 7), a slight increase in SPR peak intensity is visible, suggesting both the presence and good stability of AgNPs, with no aggregation observed. The initial absorption peak at 438 nm remains visible after 30 days, and a red shift in the peak signifies an increase in the average AgNP size [1].

Figure 7

AgNPs SPR peak intensity.

The appearance of the SPR band and the color change to yellowish-brown (Figure 8) in the solution result from the cumulative oscillation of the conducting metal surface electrons in resonance, known as SPR, indicated by a λ_max of 438 nm, which is typical for AgNPs [3,46,47]. After a few hours, the solution’s color remained unchanged, indicating a decrease in the reduction process. After 30 days, no significant color change was observed, suggesting no further reaction.

Figure 8

Color change of guarana seed skin extract and AgNPs in colloidal solution in 1 h, 24 h, and 30 days.

Morphological information of AgNPs, such as particle size distribution, shape, and aggregation, can also be inferred using UV-vis techniques [33,48]. AgNPs produced using guarana seed skin extract were isotropic and spherical in shape, indicated by a single SPR peak, as confirmed by TEM. The broadening of the UV-vis spectrum suggests a polydisperse nature. After 24 h, the absorbance reaches its highest value, indicating a larger quantity of AgNPs. This suggests that flavonoids, polyphenols, and alkaloids present in the guarana seed skin extract likely acted as reducing and stabilizing agents. Similar results were reported in the green synthesis of AgNPs using Bauhinia variegata extract [49]. In that work, a broad absorption band at 415 nm, characteristic of SPR in AgNPs, was observed. The formation of the SPR peak is influenced by particle size, shape, and the properties of the formed particles [8,11]. Bauhinia variegata extract served as a stabilizing and reducing agent in converting silver ions to AgNPs.

XRD was also performed to identify the crystalline nature and crystal size of AgNPs, as shown in Figure 9. When compared with the standard JCPDS (064706) for silver, the XRD pattern confirmed the presence of AgNPs. The spectrum shows peaks at 2θ values of 38.30°, 44.44°, 64.47°, and 77.65°, corresponding to the (111), (200), (220), and (311) crystallographic planes of the face-centered cubic (fcc) structure. Additional peaks at 2θ values of ∼26° and ∼33° were observed, suggesting the presence of cristobalite (SiO₂), possibly originating from the Paullinia cupana extract on the AgNP surface [50]. The Debye–Scherrer equation was used to calculate the crystal size of AgNPs based on the (111) plane, resulting in a size of approximately 9.34 nm. Broader peaks indicate the presence of smaller crystallites [51]. The crystal size observed was smaller than the average AgNP size measured by DLS and TEM, which is due to differences in measurement methods. XRD calculates the crystal size of crystalline regions only, while TEM provides the overall particle size, and DLS measures the hydrodynamic size, including solvation layers or stabilizing agents on the nanoparticle surface [52,53]. XRD patterns confirm the potential of guarana seed skin extract as a powerful reducing and stabilizing agent for the synthesis of crystalline AgNPs. These results are consistent with those of Soliman et al., who synthesized AgNPs using Bauhinia variegata extract, obtaining four primary peaks at 2θ values of 38°, 44°, 64°, and 77°, corresponding to the (111), (200), (220), and (311) planes of silver [49].

Figure 9

XRD spectrum of the biosynthesized AgNPs.

FTIR was employed to identify the functional groups on the AgNP surface (Figure 10 and Table 6). Similarities and marginal shifts in peak positions between both spectra indicate the presence of residual guarana seed skin extract involved in capping, reducing, and stabilizing AgNPs. Specifically, the FTIR spectrum of guarana seed skin extract presented a broad band at 3,214 cm⁻¹, corresponding to –OH functional groups in alcohols and phenolic compounds. The band at 2,912 cm⁻¹ corresponded to C–H stretching in methyl and methylene groups from lipids. The 2,092 cm⁻¹ band corresponds to amine salts and NH₄⁺ stretching of amines, while the 1,682 cm⁻¹ band corresponds to C═O stretching in proteins and/or esters. N–H bending of aromatic amines at 1,619 cm⁻¹ was also observed. Bands at 1,216, 1,173, and 1,129 cm⁻¹ are attributed to CH and CH₂ stretching, typical of alkaloids and flavonoids in the extract. Similarly, the FTIR spectrum of AgNPs displayed 13 main peaks: 3,264, 2,899, 2,067, 1,720, 1,628, 1,517, 1,378, 1,367, 1,125, 1,058, 973, 836, and 736 cm⁻¹. The strong absorption band at 3,264 cm⁻¹ (OH stretching) indicates functional groups from guarana seed skin extract surrounding the AgNPs. The slight shifts in some bands between the spectra of guarana seed skin extract and AgNPs likely result from their interactions, suggesting that biomolecules from the guarana seed skin extract play a crucial role in the reduction and stabilization of AgNPs.

Figure 10

FTIR spectra of the guarana seed skin extract and the as-prepared AgNPs.

The surface properties and stability of AgNPs (Figure 11) were evaluated using zeta potential measurements. Typically, zeta potential values lower than −30 mV or higher than +30 mV indicate good suspension stability due to strong electrostatic repulsion between particles, which prevents aggregation [54]. The average zeta potential of AgNPs was measured at −11.2 mV, with an average particle size of 63.9 ± 1.64  nm and a particle size distribution of 0.448. This negative zeta potential suggests that the AgNP surface charge, likely originating from functional groups such as OH−, COO−, and CO− in the guarana seed skin extract, contributes to stabilization, as observed in FTIR analysis. According to DLVO theory, nanoparticle stability in suspension results from the balance between attractive van der Waals forces and repulsive electrostatic interactions [55]. The negative charge on the AgNP surface helps prevent aggregation through electrostatic or steric repulsion. Similar results were reported by Ejaz et al., who synthesized AgNPs using Thymus vulgaris extract, achieving a zeta potential of −7.46 ± 4.93 mV, indicating a stable negative surface charge [47]. Consistent findings were reported by Jeon et al., who used Black Mulberry leaf extract, obtaining AgNPs with a zeta potential of −56.6 ± 0.56 mV and a hydrodynamic size of 170.17 ± 12.65 nm [56]. The negative zeta potential indicated uniformly dispersed AgNPs with good colloidal stability, with larger hydrodynamic sizes than those measured by TEM, likely due to the presence of an organic layer on the AgNP surface.

Figure 11

Biosynthesized AgNPs from guarana seed skin extract: (a) Zeta potential at pH of 6 and particle size distribution. (b) Hydrodynamic size distribution.

The concentration of silver in the biosynthesized AgNPs was measured by ICP-OES, yielding a concentration of 180 ± 0.15 ppm after 24 h. Under optimal conditions requiring a 5 mM solution of AgNO₃, this concentration corresponds to a silver reduction efficiency of 33.36%. Factors such as the complexation of Ag⁺ ions with biomolecules in guarana seed skin extract may limit reduction. Additionally, the availability of reducing and stabilizing agents, or silver nitrate, may be insufficient in green synthesis processes. Parameters like pH, reaction temperature, and plant extract concentration also play critical roles in green synthesis of AgNPs [3,35,36]. For instance, Taleb Safa and Koohestani reported AgNP concentrations of 2,750 ppm using green tea extract as a reducing agent, highlighting variability across synthesis conditions [57].

SEM and TEM analyses (Figure 12) were conducted to examine the morphology, size, and size distribution of AgNPs.

Figure 12

SEM images of AgNPs obtained by green synthesis using guarana seed skin extract at different magnification levels: (a) ×5,000, (b) ×20,000, (c) ×100,000, and (d) ×200,000.

In SEM images captured at high magnifications, quasi-spherical AgNPs are evident, consistent with observations in TEM. Similar findings were reported by Ghasemi et al. [58], who synthesized AgNPs using leaf extract of Rubus discolor as a stabilizing and reducing agent. Their SEM images displayed spherical AgNPs with polydispersity and an average particle size of 38 nm.

The TEM images (Figure 13) further illustrate the shape and size of the AgNPs synthesized with guarana seed skin extract. A quasi-spherical shape is observed, with a uniform particle diameter of approximately 26.11 nm. This TEM-measured particle diameter is smaller than that obtained from DLS measurements, as expected. DLS measures the hydrodynamic size, which includes ions and stabilizing agents adsorbed on the AgNP surface, whereas TEM analysis involves dry particles. Some agglomerated particles can be observed in the micrographs, but they were excluded from the size measurement. The presence of potential capping agents, such as alkaloids, proteins, and flavonoids from the guarana seed skin extract, surrounding the AgNPs is also visible in the TEM images, indicating their role in colloidal stabilization.

Figure 13

TEM images of AgNPs obtained by green synthesis using guarana seed skin extract. (a) AgNP micrograph at the scale of 200 nm. Inset: Histogram of AgNP size distribution. (b) Presence of guarana seed skin extract.

These findings align with those of Ahmed et al. [59], who developed an eco-friendly and cost-effective method for AgNP synthesis using Calendula arvensis extract as a reducing and stabilizing agent. Their optimized synthesis, based on a Taguchi design incorporating factors like silver nitrate concentration and extract temperature, yielded stable colloidal AgNPs with irregular morphologies and an average diameter of approximately 45 nm, achieving maximum SPR wavelength under optimal conditions.

3.5 Proposed mechanism of AgNP formation

The proposed chemical mechanism (Figure 14) for the reduction of silver ions to metallic AgNPs involves the major components found in guarana seed skin extract, as identified by EM/SM analysis.

Figure 14

A possible chemical mechanism for the silver ion to the metallic AgNP of seed skin guarana extract via caffeine as a reducing agent.

Studies indicate that caffeine (1,3,7-trimethylxanthine), a primary compound in guarana seed skin extract, can act as an electron donor under alkaline conditions (pH 11) due to the constraints of carbon valency and its chelating action, leading it to convert into a quinone form [60]. In this process, silver ions are reduced by electrons donated from caffeine molecules, which subsequently aggregate to form clusters and eventually colloid-stable AgNPs in an aqueous medium (Figure 15).

Figure 15

Stabilization of AgNPs performed by caffeine molecule present in the guarana seed skin extract.

The observed decrease in the intensity of the −OH vibration in FTIR analysis suggests that hydroxyl groups play a role in the reduction of silver ions to AgNPs, further supporting caffeine’s involvement in both reducing and stabilizing the silver ions. Other compounds within the guarana seed skin extract, such as quercetin, polyphenols, terpenoids, quinic acid, and vanillic acid, may also contribute as reducing and stabilizing agents [29,61,62].

It has been demonstrated that the effectiveness of silver ion reduction in the formation of AgNPs depends on the biomolecules present in plant extracts. Typically, macromolecules and phytochemicals such as catechins, flavonoids, ascorbic acid, alkaloids, proteins, and volatile acids serve as the primary reducing and stabilizing agents for AgNPs [46,47,56,58,63]. However, when plant extracts are utilized as sources of stabilizing and reducing agents, complex reactions can occur, and the exact mechanism remains a topic for further investigation and discussion.

3.6 Antibacterial activity of AgNPs

The antibacterial activity of AgNPs was evaluated against Gram-positive S. aureus and Gram-negative E. coli. Figure 16 illustrates the antibacterial effectiveness of the biosynthesized AgNPs, guarana seed skin extract, and the positive control (Levofloxacin) using the agar well diffusion method. The results were assessed based on the mean diameter of the zone of inhibition (ZOI), as summarized in Table 7. The minimum inhibitory concentrations (MICs) for S. aureus and E. coli were determined to be 50 μg/mL (Figure 17). In both cases, AgNPs demonstrated a promising dose-dependent antibacterial effect.

Figure 16

Antibacterial activity in terms of zone of inhibition of the biosynthesized AgNPs mediated by guarana seed skin extract against (a) E. coli and (b) S. aureus.

The exact antibacterial mechanism of AgNPs remains incompletely understood, with multiple potential actions proposed in the literature [1,8,11]. One possible mechanism involves the release of silver ions from AgNPs, which can disrupt bacterial cell walls. Studies suggest that AgNPs can destabilize the outer membrane of E. coli, facilitating interaction with the underlying peptidoglycan layer [64]. Additionally, AgNPs may release silver ions that bind to thiol groups in bacterial proteins, leading to protein inactivation and impaired cellular function [65].

The antibacterial activity of AgNPs may also be influenced by their size and morphology. TEM images reveal that the AgNPs synthesized in this study are small (around 26 ± 10 nm) and quasi-spherical, which may enhance their interaction with bacterial cells. Based on literature reports and the observed antibacterial activity in this study, it can be concluded that AgNPs synthesized from guarana seed skin extract represent a promising bio-resource for antibacterial applications.

Figure 17

Determination of MIC of AgNPs in 24 h against S. aureus and E. coli.