Cationic carbon dot–silver nanohybrids with enhanced stability and redox control

Abstract

Silver nanoparticles (AgNPs) show promise for biomedical applications, but their functional stability remains a major challenge. Here, the photochemical synthesis of hybrid AgNPs using carbon dots (CDs) as both reducing and capping agents (CDs–AgNPs) is described, producing highly stable nanostructures under physiological conditions. Comparisons with traditionally synthesized AgNPs, using Irgacure-2959 photoreduction and sodium borohydride methods, indicate that CDs–AgNPs have better colloidal stability across a range of pH levels (4.5–8), ionic strengths (up to 500 mM in NaCl), and oxidative conditions, especially against peroxyl radicals for at least 2 h, whereas other AgNP formulations destabilize within minutes. Spectroscopic analysis confirmed the presence of CDs on the AgNP surface, forming a protective shell rich in amino groups. This shell prevents nanoparticle clumping and oxidation, while also enhancing antioxidant capacity compared to gallic acid and BHT standards, outperforming other standard AgNPs. In vitro cytotoxicity tests on cancerous and non-cancerous cell lines showed toxicity toward MCF cell lines at the highest concentration used, with CDs–AgNPs showing increased activity likely due to a pro-oxidant cellular mechanism. These findings demonstrate a green and reproducible method for creating carbon–silver nanohybrids with improved stability and redox control. This hybrid structure offers a platform to reduce instability and potential side effects associated with metallic nanomaterials, supporting their advancement in therapeutic applications.

Supplementary Information

The online version contains supplementary material available at 10.1186/s11671-026-04779-0.

Introduction

Silver nanoparticles (AgNPs) are widely used in the biomedical field due to their remarkable physicochemical properties, including conductivity, heat transfer, and optical properties [1, 2]. In addition, their biological functions are widely described, including antimicrobial and prophylactic activities, as well as moderate/low toxicity [3, 4], which has prompted the use of AgNPs in novel therapies such as radiotherapy [5] and antibiofilm treatments [6–8], among others. However, AgNPs can still cause nanotoxicity in mammalian cells by inducing oxidative stress, inflammation, protein denaturation, and genotoxicity [9–11], ultimately leading to a double-edged sword dilemma when considering the use of metal nanoparticles for biomedical purposes. Regarding this drawback, several studies have reported that a major part of these consequences arises from the poor stability of AgNPs under biological conditions [12]. Therefore, a gap remains in improving the stability of AgNPs formulations for biomedical applications.

The stability of AgNPs and their properties are closely linked to the chosen synthesis pathways, with both top-down and bottom-up routes being extensively explored in the literature [13]. Yet, greater attention has been given to the latter due to the ease with which factors such as growth, shape, and capping agents can be manipulated, among others [14]. For instance, the Turkevich method, the most widely used, was initially regarded as a protocol for producing gold nanoparticles by reducing gold ions with sodium citrate at elevated temperatures [15]. Of course, this method has been modified in several studies; for example, the use of sodium borohydride (NaBH₄) as a reductive agent was one such modification. These optimizations have enabled the synthesis of various metal nanoparticles, including AgNPs, across a broad range of sizes [16–18], and have been extensively employed for different biomedical applications of AgNPs [19]. Another method that has garnered interest is the photochemical pathway, in which reduction is performed using labile molecules that are prone to forming free radicals upon light irradiation [20], which has the advantage of controlling morphology in the development of highly monodispersed, small-sized metal nanoparticles [21]. Still, developing stable AgNPs through this methodology requires the use of capping agents to mitigate the instability of these usually small-sized AgNPs (< 10 nm) [22]. In this regard, the AgNPs synthesis requires a reducing agent to reduce silver ions and a capping agent to prevent nanoparticle aggregation and further agglomeration, which can be fulfilled by a single reagent acting as both reducing and capping agent, or by different molecules or substrates that serve those roles. As expected, their selection influences the size and morphology of AgNPs, as well as their final physical, chemical, and biological properties [23, 24]. One of the most commonly used capping agents is citrate, which stabilizes colloidal AgNPs by providing a negative surface charge and ensuring uniform size and morphology via electrostatic interactions, without affecting the initial nucleation process in AgNP synthesis [25]. Furthermore, citrate-AgNPs in biological applications demonstrate an increased cell uptake compared to other capping agents, such as PEG, without significantly altering their toxicity [26].

Although not extensively explored, carbon dots (CDs) are nanoparticles smaller than 10 nm that have also been established as potential reduction and capping agents for AgNPs synthesis, particularly in catalysis [27]. Previous research has established that the charged surface of CDs can act as a reducing agent in the synthesis of silver nanoparticles (AgNPs) using various methods, including the Turkevich-like method [28, 29]. Particularly in terms of superficial charge, cationic CDs have attracted interest because they have been described as having potential therapeutic effects, such as serving as carriers for negatively charged bioactive molecules, inhibiting tumor cell growth, exhibiting antibacterial activity, and facilitating permeation of cell membranes, which is attributed to their improved cell-nanomaterial interactions [30–32]. While considering their photoluminescent properties, they can act as a photo-initiator agent through photoinduced electron transfer [33, 34]. In this regard, CDs´ optical properties, particularly their photoluminescence, are related to their structure, which consists of carbonaceous sp² and sp³ domains. These regions exhibit multiple electron transitions on the surface, suggesting their ability to capture excitons and induce the oxidation of CD’s surface when the nanoparticle is excited by UV light [35, 36]. In particular, the use of CDs for AgNPs preparation supports the development of a hybrid composite (CDs-AgNPs) that enhances various properties in metal components [37]. For instance, hybrid composites of carbon dots and silver nanoparticles demonstrate high catalytic activity in degrading organic dyes [38]. Meanwhile, the use of nanohybrid composites coated with an NIR dye (CyOH) enhanced the antitumor effect by increasing singlet oxygen levels compared with CyOH alone [39].

Therefore, previous research supports the use of carbon nanomaterials as reducing or capping agents in the development of AgNPs, providing essential activity as a protective coating layer that reduces aggregation and surface oxidation [40]. This makes them a perfect candidate for establishing new AgNPs formulations for next-generation biomaterials. However, other relevant questions remain unexplored. For instance, in biological environments, the presence of reactive oxygen species (ROS) represents extreme conditions for the stability of metal nanoparticles, which commonly leads to toxicity in the cellular environment [12]. Moreover, the impact on cytotoxicity and its performance against other synthesis pathways requires further study to propose this nanocomposite as a next-gen AgNPs [41]. Thus, this work proposes a green CDs-AgNPs formulation approach prepared by a photochemical method, employing cationic carbon dots as reducing and capping agents, and postulates it as a nanoparticle with enhanced stability against biological stresses. Therefore, their stability is tested under common biological stresses, including ionic strength, pH, and free radicals. Also, their antioxidant activity is evaluated. In addition, its cytocompatibility has been established and compared with that of other AgNPs formulations obtained by the Turkevich-like method (T-AgNPs) and photochemical procedures (P-AgNPs), and their physical, chemical, and biological properties have been discussed.

Experimental

Materials

Silver nitrate (AgNO₃), Poly (ethylene imine) (PEI, Mw: 20,000 g/mol), Chitosan (CS, cod. 448,877), 2-hydroxy-1-[4-(2-hydroxyethoxy) phenyl]- 2- methyl-1-propanone (I-2959), Trisodium citrate dihydrate, Sodium borohydride (NaBH₄), Sodium chloride (≥ 99.5%), Phosphate Buffered Saline (P4417), Tris, Sodium acetate, acetic acid, chloride acid, sodium hydroxide, 2,2’-azobis(2-methylpropionamidine) dihydrochloride (AAPH), 2,2′-azinobi (3-ethylbenzothiazoline-6-sulphonic acid) (ABTS), 2,2′-azo-bis (2-amidino propane) (ABAP), 2,4,6-tri(2-pyridyl)-s-triazine (TPTZ), 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox), ferric chloride (FeCl₃), and 2,2-Diphenyl-1-picrylhydrazyl (DPPH) were purchased and used as received from Sigma-Aldrich. All solutions were prepared using Milli-Q water from an Adrona CB1901 system.

Synthesis of carbon dots and AgNPs

This study examined three different synthesis methods to develop AgNPs and subsequently evaluated their stability and cytotoxicity under different conditions. Each procedure is presented below.

Synthesis of carbon dots (CDs)

Carbon dots were synthesized according to the method described by Santos and colleagues [42]. Briefly, a mixture of 10 mL of 2% m/v Chitosan (CS) (dissolved in 0.5 M acetic acid) and 50 mg of PEI dissolved in 3 mL of Milli-Q water was irradiated in a microwave for 8 min. After irradiation, the CDs were placed in a poly (tetrafluoroethylene) (PTFE) container, positioned in a hydrothermal reactor, and sealed. The system was maintained at 90 °C for 4 h. The resultant solution was filtered through a 0.22 μm filter, yielding a clear and transparent solution. Subsequently, the solution was lyophilized and stored in the dark at room temperature.

Synthesis of AgNPs using CDs as reducing and stabilizing agents (CDs-AgNPs)

To develop the CDs-AgNPs, different concentrations of CDs solutions were evaluated (1, 2.5, 5, 10, 15, and 20 mg of lyophilized CDs in 13.2 mL of Milli-Q water), which were used for subsequent AgNPs synthesis. Then, 300 μL of AgNO₃ (10 mM) was added to the CDs, followed by 1.5 mL of dehydrated trisodium citrate (10 mM). The solution was purged with N₂ for 30 min. Then, using a UV LED (365 nm), the solution was irradiated for an additional 30 min while continuously stirring at 400 rpm. Finally, a translucent yellow-like color was obtained.

Synthesis of AgNPs by the Turkevich method (T-AgNPs)

The T-AgNPs were prepared using a protocol similar to that previously reported by Wuithschick and collaborators for the synthesis of AuNPs [18], with minor modifications to the AgNPs synthesis. Shortly, a 10 mL solution of AgNO₃ (0.2 mM) was stirred and heated at 400 rpm and 90 ºC, respectively, until thermal equilibrium was reached. Then, 1 mL of trisodium citrate dihydrate (10 mM) at 90 °C was incorporated into the solution, with continued stirring for 12 min. Afterward, 5 μL of NaBH₄ (5.285 mM) was added, and the solution was stirred at 1200 rpm for 1 min. Subsequently, the stirring rate decreased back to 400 rpm for 1 h. Finally, the T-AgNPs suspension appeared as a translucent, yellow-colored solution.

Synthesis of AgNPs by photochemical reaction (P-AgNPs)

The P-AgNPs were prepared according to the protocol reported by Urzúa and colleagues [43]. In brief, a solution containing AgNO₃ (0.2 mM), I-2959 (0.2 mM), and sodium citrate (1 mM) was prepared and subsequently purged with N₂ for 30 min to remove the oxygen. The solution was then irradiated with a UV LED (365 nm) at room temperature for 30 min. The final P-AgNPs suspension was translucent yellow.

Characterization of AgNPs

Surface plasmon resonance (SPR), or plasmon band, was continuously evaluated as a rapid yet unequivocal indicator of metal nanoparticle stability. For this, the UV–Vis spectra of all AgNPs formulations under all tested conditions were recorded using a UV–Vis spectrophotometer (V-750, Jasco). All tests were performed in triplicate.

Size and morphology of the AgNPs were determined using transmission electron microscopy (TEM) on a Talos FF200C G2 (Thermo Scientific) at 120 kV. Gold-carbon grids (300 mesh) were used, and 30 μL of each diluted AgNP sample at a 1/10 concentration was applied to them. Additionally, ζ-potential values were measured for all samples using a Zetasizer (Nano-ZS90) after incubation for 1 h at 37 ºC. Each measurement was conducted in triplicate.

The molecular structure of each AgNPs formulation prepared was evaluated using a Fourier Transform Infrared Spectrophotometer (FT-IR) with an attenuated total reflectance (ATR) accessory (Spectrum Two FTIR, Perkin Elmer). The measurements were conducted by performing 64 scans for each sample. Due to stability considerations, T-AgNPs and P-AgNPs were measured as aqueous dispersions, whereas CDs-AgNPs were measured as a powder post-lyophilization using a BK-FD10 freeze-dryer (Biobase), as their stability allowed this process without altering their properties.

X-ray powder diffraction (XRD) measurements were performed using a RIGAKU Multipurpose SmartLab model in the 2θ region of 2°–80° with a Cu lamp (λ = 1.5418 Å). The sampling interval was 0.02°.

Ionic stability of AgNPs

Ionic stability was assessed by analyzing the plasmon band for each AgNPs formulation using a UV–Vis spectrophotometer (V-750, Jasco). To do this, the AgNPs solutions were diluted in a 1:1 ratio with NaCl solutions at different concentrations and measured every 10 min for up to 60 min. The NaCl final concentrations used were 10, 100, and 500 mM.

Stability of AgNPs at different pH

The stability of different AgNPs formulations was tested at different pH values. To achieve this, AgNPs were diluted in a buffer solution at a 1:4 ratio (AgNP:buffer), and changes in their stability were monitored by observing variations in the plasmon band using a UV–Vis spectrophotometer (V-750, Jasco). The buffer solutions used (10 mM) included acetate buffer (pH 4.48), phosphate-buffered saline (PBS; pH 7.38), and Tris buffer (pH 8). All experiments were conducted at room temperature and in triplicate.

Stability of AgNPs in the presence of peroxyl radicals

The stability of AgNPs formulations in the presence of radicals was monitored through their respective plasmon peaks using UV–Vis spectroscopy. Briefly, 500 µL of the AgNPs suspension was diluted with 500 µL of Milli-Q water, incubated at 37 °C for 30 min, and its UV–Vis spectrum was recorded as the AgNPs control. Afterwards, AAPH radical was introduced into the nanoparticle’s suspension at a final concentration of 10 mM. The AgNPs suspension exposed to AAPH was incubated, and the UV–Vis spectra were recorded at different time intervals over a 2-h period. Stability was assessed considering the λ max displacement and the full width at half maximum (FWHM) of the plasmon.

Antioxidant activity

Hydrogen-atom transfer (HAT) and single-electron transfer (SET) methods were employed to assess the antioxidant capacity of various AgNPs. The total reactive antioxidant power (TRAP) assay was conducted using the HAT method. In contrast, SET methods involved evaluating both the ferric reducing antioxidant power (FRAP) and DPPH· radical scavenging assays. Detailed descriptions of each procedure are provided below.

Total reactive antioxidant power (TRAP) assay

The method followed the protocol described by Romay et al. [44]. One volume of a 10 mM solution of ABAP (2,2′-azo-bis (2-amidino propane)) was mixed with the same volume of a 150 μM solution of ABTS (2,2′-azinobi (3-ethylbenzothiazoline-6-sulphonic acid)) using PBS 100 mM at a pH of 7.4 (TRAP solution). The mixture was incubated at 45 °C for 30 min and then cooled to room temperature for use. The AgNPs sample solution was mixed with the TRAP solution, and the absorbance was measured after 50 s at 734 nm against the ABTS solution as a blank. The absorbance values were interpolated in a Trolox standard curve, and the TRAP values were expressed in mM Trolox equivalent antioxidant capacity (mM TEAC). All measurements were performed in triplicate.

Ferric reducing antioxidant power (FRAP) assay

Ferric reducing power was measured as described by Dudonné et al. [45]. Freshly TPTZ reactive solution was prepared (10 volumes of 300 mM acetate buffer, pH 3.6, with 1.0 volume of 10 mM TPTZ (2,4,6-tri(2-pyridyl)-s-triazine) in 40 mM hydrochloric acid, and 1.0 volume of 20 mM ferric chloride). FRAP reagent (3.0 mL) was mixed with deionized water (300 μL) and the tested AgNPs sample (100 μL). The mixture was incubated for 30 min at 37 °C in a water bath, and the absorbance was measured at 593 nm using water as the blank solution. The obtained absorbance values were interpolated in a Trolox calibration curve, and the FRAP values were expressed in mM Trolox equivalent antioxidant capacity (mM TEAC). All measurements were performed in triplicate.

Radical scavenging assays using (DPPH·)

The DPPH assay was performed as described by Brand–Williams et al. [46]. The AgNPs sample (0.1 mL) was mixed with a freshly prepared 50 µM (DPPH·) solution (2.9 mL), and ethanol (0.1 mL) was used as the control. The samples and control solutions were incubated for 15 min at room temperature, and the absorbance (A) was measured at 517 nm. The following Eq. (1) calculates the inhibition (%):

1

Based on the inhibition values (%) obtained, the IC₅₀ value was determined using the dose–response equation.

In vitro evaluation

In this study, the employed cell lines were obtained from the American Type Culture Collection (Rockville, MD, USA): HT-29 (human colon cancer; ATCC NO. HTB-38), PC-3 (human prostate cancer; ATCC NO. CRL-1435), MCF-7 (human breast cancer; ATCC NO. HTB-22), and MCF-10A (human mammary gland; ATCC NO. CRL-10317). All cell lines were grown in Dulbecco’s Modified Eagle Medium/Nutrient Mixture F-12 (DMEM-F12) containing 10% FCS, 100 U/mL penicillin, 100 μg mL⁻¹ streptomycin, and 1 mM glutamine. Cells were seeded into 96-well microplates at 100 μL per well, with a plating density of 3 × 10³ cells/well. After a 24 h incubation at 37 °C (under a 5% humidified carbon dioxide atmosphere to allow cell attachment), cells were treated with different concentrations of AgNPs (CDs-AgNPs, T-AgNPs, and P-AgNPs) and incubated for 72 h under the same conditions. The AgNPs stock formulations were prepared in Milli-Q water.

Cytotoxicity

The Sulforhodamine B (SRB) assay was used according to the method of Skehan et al. [47]. Cell density was determined using a microplate reader (wavelength 540 nm). Untreated cells served as the negative control. In addition, all samples were tested at concentrations ranging from 30 to 0.3 nM AgNPs, using Milli-Q water as the carrier solvent. All measurements were replicated three times. Finally, SigmaPlot software calculated the viability percentage.

Statistical analysis

All data were reported as mean values ± standard deviation (SD). Due to the nonparametric nature of the data, a Kruskal–Wallis ANOVA was conducted at the 95% confidence level using STATISTICA 7.0.

Results

CDs-AgNPs synthesis

CDs were synthesized using a previously reported method described by Santos et al. [42], which employed chitosan and poly(ethylene imine) as carbon sources, yielding cationic carbon dots. Consequently, the synthesis of CDs-AgNPs was achieved through a photochemical method, as depicted in Fig. 1. The rationale for proposing this method stems from the photoexcited states that CDs can attain, making them suitable as photoinitiator agents that function as electron donors under UV irradiation (365 nm). Upon irradiation, a nucleation process of silver atoms on the CDs' surface via electrostatic interaction is favored [48]. Nonetheless, the resulting nanocomposite (CDs and AgNO₃ only) quickly aggregated and precipitated, losing its plasmon signal after a few hours, indicating poor stability as a nanocomposite. Therefore, adding a capping agent was necessary to improve nanoparticle stability. In this regard, sodium citrate was chosen as the extra capping agent, as its stabilizing capacity for AgNPs has been widely documented in the literature [49]. During nanoparticle synthesis, the citrate’s carboxylate group (COO⁻) electrostatically interacts with the positively charged silver ions, promoting the formation of a negatively charged layer on the surface of the AgNPs. This negative layer enhances colloidal stability by repulsive electrostatic forces, preventing agglomeration between the nanoparticles [50, 51]. As a result, the obtained CDs-AgNP showed improved colloidal stability for at least three months. This proposed photochemical approach differs from those using common chemical reduction methods, such as a Turkevich-like process [29], or other well-known techniques, such as green synthesis [52], because it enables the consistent production of more uniform nanoparticles with better control over their size and shape. However, it is important to recognize that each synthesis method has its own advantages and disadvantages. In this context, a limitation of the photochemical route in this study is that it requires a co-stabilizing agent (citrate) that improves the colloidal stability of the final nanocomposite. To continue, once prepared, the CDs-AgNPs were characterized physicochemically, and their stability against ionic strength, pH, and free radicals was assessed and compared with those of the other two classical AgNPs synthesis methods: the Turkevich (T-AgNPs) and photochemical (P-AgNPs) procedures. The three AgNP formulations were prepared and evaluated using the same precursor concentration (AgNO₃) and trisodium citrate as capping agents to determine the CDs effect for the redox control in the silver nanocomposite CDs-AgNPs.

Fig. 1.

Schematic representation of silver nanoparticle preparation using carbon dots as a photoinitiator (CDs-AgNPs). Pictures on the right show the sample at various preparation stages

Characterization

To develop the CDs-AgNPs, different concentrations of CDs were evaluated as photo-initiators to achieve the most suitable formulation, established in terms of the plasmon band (Fig. 2a). Consequently, the resulting AgNPs were analyzed by UV–Vis spectroscopy and compared to those obtained through photochemical and Turkevich synthesis methods (Fig. 2b). The nanomaterials exhibit a characteristic absorbance peak around 400 nm associated with the surface plasmon resonance (SPR) band of AgNPs [53]. After UV irradiation, the maximum absorbance and sustained peak were observed for the CDs-AgNPs formulation prepared with 1 mg/mL of CDs. The irradiation time was set to 30 min to ensure maximum conversion of CDs-AgNPs (Figure S1). Regarding the CDs' fluorescence emission, they exhibited two excitation peaks at 275 and 370 nm, which, upon irradiation, produced broad, high-intensity emission bands (Figure S2). However, after serving as reducing and capping agents in the synthesis of the AgNPs, these bands are diminished. This phenomenon can be attributed to their incorporation onto the AgNPs, which quenches the luminescence of the CDs. Similarly, AgNPs obtained via photochemical and Turkevich protocols showed SPR peaks around 400 nm (Fig. 2b). However, the AgNPs containing CDs exhibited an enhanced plasmon band compared to those obtained by reducing silver nitrate using the photochemical and Turkevich protocols. Previous work by Beiraghi and Najibi-Gehraz (2017) demonstrated the potential role of CDs in enhancing the SPR effect on AgNPs when evaluating their capacity for the specific detection of copper ions, corroborating these results in developing the nanohybrid structure through AgNO₃ reduction in the presence of CDs [54]. Under these considerations, further experiments were conducted using the CDs-AgNPs formulation at a concentration of 1 mg/mL.

Fig. 2.

UV–Vis spectra of different AgNPs. a CDs-AgNPs at different concentrations of CDs after 30 min of UV irradiation. b Plasmon band of AgNPs synthesized using three different methods: CDs-AgNPs, T-AgNPs, and P-AgNPs

The average size, distribution, and morphology of all of the tested AgNPs formulations (i.e., with CDs, photochemical, and Turkevich-like) were assessed by TEM (Figure S3). The size distribution plots in Fig. 3 show that CDs-AgNPs have an average size of 6.07 (± 1.72) nm, representing the smallest size among the other formulations. Meanwhile, P-AgNPs show an average size of 7.16 (± 1.88) nm, and Turkevich-like (T-AgNPs) displayed the largest size and heterogeneous distribution at 9.26 (± 3.05) nm. Regarding their morphology, all three formulations have a quasi-spherical shape. Equivalent results have been previously reported for the chemical and photochemical synthesis with sizes below 10 nm [55–57]. On the other hand, hybrid structures generated with CDs were reported by Jahanbakhshi and Habibi (2016), who synthesized CDs-AgNPs via UV irradiation at 240 nm. In this work, the nanomaterial exhibited a spherical morphology with sizes ranging from 4 to 8 nm, demonstrating the ability of CDs to function as photo-initiators, which enables the formation of controlled, small-sized nanohybrid structures [58].

Fig. 3.

TEM results of different tested AgNPs formulations. Distribution plots were created by measuring the diameter of 100 individual nanoparticles from TEM images of each sample. The red line represents the normal distribution, illustrating the size distribution. Representative TEM images of each formulation are shown as insets, scale bars = 200 nm. a CDs-AgNPs; b Turkevich-AgNPs; and c Photochemical-AgNPs

The ζ potential was measured to evaluate the surface charge of the AgNPs formulation. CDs-AgNPs exhibited a positive surface charge of 7.5 (± 0.26) mV. Meanwhile, P-AgNPs and T-AgNPs presented negative surface charges of −16.83 (± 1.45) mV and −13.36 (± 1.95) mV, respectively. The observed differences in the measured surface charge can be attributed to their capping agents; in the case of CDs-AgNPs, CDs are primarily composed of cationic components (PEI and CS), indicated by the presence of amine groups. Conversely, in the P- and T-AgNPs formulations, trisodium citrate is the sole capping agent, containing a carboxyl group that promotes the negative charge. This assertion is confirmed by the FT-IR results (Fig. 4a), where it is observed that for CDs-AgNPs, the peaks at 3350 cm⁻¹ correspond to the stretching vibration of N–H and O–H bonds, around 1550 cm⁻¹ correspond to the N–H bond, and the intense band around 1360 cm⁻¹ describes the bending of C-H and COO– groups of CS [59]. In contrast, P-AgNPs and T-AgNPs presented a peak near 3400 cm⁻¹, which can be associated with the O–H stretching vibrations from water. This is mainly because these two formulations were evaluated in an ultrapure water suspension; under drying conditions, these AgNPs are unstable and quickly oxidize. Nevertheless, all three samples exhibit a similar peak at 1600 cm⁻¹, which is associated with the anti-symmetric COO- stretching vibration of the trisodium citrate [60].

Fig. 4.

Characterization of the tested AgNPs formulations. a FT-IR spectra of the tested samples, including the employed CDs as precursors. The CDs and CDs-AgNPs were evaluated in powder form, whereas T-AgNPs and P-AgNPs were assessed in Milli-Q water due to their instability in the solid state. b XRD diffractogram of the CDs-AgNPs. The different measured planes are indicated within the figure

The crystalline structure of CDs-AgNPs was evaluated by XRD (Fig. 4b). The diffractogram showed strong peaks around 38°, 44°, 64.5°, and 77.5°, corresponding to (111), (200), (220), and (311) planes, respectively, suggesting the formation of AgNPs. Meanwhile, an amorphous phase is observed between 5 and 20º, which can be associated with the CDs. These results correlate with previous literature, which has verified the formation of nanohybrid structures using metallic Ag [61, 62].

Ionic and pH stability

A relevant parameter for any potential use of metal nanoparticles in the biomedical context is their stability with respect to ionic strength and pH [63, 64]. Initially, the changes in the AgNPs' plasmon bands were evaluated over time as NaCl concentration increased. This test provides useful insights because ions in solution generally shield the electrostatic repulsion between metal nanoparticles, thereby decreasing their colloidal stability, which can lead to aggregation and further precipitation at higher salt concentrations [65]. Now, when evaluating the plasmon band of the CDs-AgNPs after 1 h of exposure in a 10 mM NaCl solution, no significant variations were observed (Fig. 5a). In contrast, T-AgNPs (Fig. 5b) showed an increase in the plasmon band intensity, followed by a notable decrease after 1 h. Furthermore, P-AgNPs exhibited a higher plasmon intensity at 1 h compared to the control, with a narrowed plasmon peak (Fig. 5c). At higher salt concentrations, after 1 h, CDs-AgNPs maintained their initial plasmon peak at 100 mM NaCl (Fig. 5d); however, T-AgNPs and P-AgNPs quickly destabilized (within the first minute) at this 100 mM NaCl (Fig. 5e and f, respectively). This suggests a poor ionic stability of the AgNPs even against a slightly saline environment, compared to the physiological condition of 154 mM (ref). Evermore, the CDs-AgNPs showed a highly saline robustness, with a plasmon λ > 400 nm after 1 h of exposure to 500 mM NaCl (with a decrease in plasmon intensity). While the T-AgNPs and P-AgNPs showed the same drastic destabilization (Figure S4). These results indicate increased ionic stability and resilience in a saline environment, promoted by the stabilization of CDs in the CDs-AgNPs. On the other hand, the stability of AgNPs at different pHs is another crucial factor, as this property is valuable for designing nanomaterials for biological applications, given the diversity of pH environments within the human body [66, 67], where pH changes can cause changes in surface charge, which in turn, can cause possible chemical and structural modifications [68]. Furthermore, it has been shown that the rate of silver ion release from AgNPs depends on pH [69]. Figure 5g–i shows the main results obtained for changes in the plasmon band of the synthesized AgNPs. As for CDs-AgNPs, this nanomaterial demonstrates robustness and remains stable under all pH conditions tested. In the case of P-AgNPs, the nanomaterial shows colloidal stability at pH 4.45 and pH 8. In contrast, T-AgNPs showed lower stability, demonstrating dependence across the entire pH range evaluated. As mentioned above, changes in the plasmon band of AgNPs in response to pH changes have been described, with similar results obtained by Zhou and collaborators (2016). The authors established that variations in the plasmon band, especially when the pH is reduced, may be associated with an increase in dissolved silver [70]. Fernando and Zhou (2019) reached a similar conclusion, linking changes in the plasmon band to the release of silver and nanoparticle aggregation [71]. These results highlight that a carbon structure can protect the AgNPs from bioenvironmental conditions, unlike other synthesis methods, in which the plasmon of AgNPs undergoes significant changes over time.

Fig. 5.

UV–Vis spectra of AgNPs under different stresses. Top: after 1 h exposure to 10 mM NaCl solution, a CDs-AgNPs, b T-AgNPs, c P-AgNPs. Middle: after 1 h exposure to 100 mM NaCl solution, d CDs-AgNPs, e T-AgNPs, f P-AgNPs. Bottom: after 1 h exposure to different pH conditions, g acetate buffer pH 4.5, h PBS pH 7.4; and i Tris buffer pH 8.0

AgNPs stability towards free radicals’ action

Among the most extreme biological conditions that nanomaterials designed for biomedical applications encounter are their exposure to highly reactive species, whether derived from oxygen (ROS) or nitrogen (RNS). The action of free radicals is one of the most common factors that promote instabilities of metal nanoparticles at the biological level [72]. Therefore, the proposed formulations were evaluated against the peroxyl radical derived from AAPH, a commonly studied free radical model that stems from an azo compound and thermally decomposes into the free radical [73]. Figure 6 shows the changes in the stability of the different AgNP formulations, as indicated by the plasmon band at 400 nm. Additional measurement times are shown in Figure S5. For CDs-AgNPs, no significant variations were observed in the UV–Vis spectrum after 2 h (Fig. 6a), which is further demonstrated by analyzing the full width at half maximum (FWHM) and the λ_max shift. After 2 h, T-AgNPs showed a reduction in the intensity of the plasmon band, a λ_max shift from 400 to 370 nm, and an increase in the FWHM due to the formation of a second band around 480 nm (Fig. 6b). Finally, the P-AgNPs formulation shows the lowest stability under peroxyl radical exposure (Fig. 6c), with a reduction in plasmon intensity, a shift in λ_max, and an increase in FWHM, similar to those observed for T-AgNPs. It is important to note that, for both T-AgNPs and P-AgNPs, the plasmon band shifts almost immediately. To further confirm whether these observations were related to changes in size and surface charge, the hydrodynamic size and ζ potential were measured. Under all tested conditions, the free radical caused the aggregation of AgNPs and the formation of metal oxides. For both nanoparticles, P-AgNPs and T-AgNPs, a rapid increase in hydrodynamic size was observed: from 14.3 to 221 nm (+ 1555.5%) and from 15.3 to 274 nm (+ 1800.8%), respectively. In addition, CDs-AgNPs showed an increase in hydrodynamic size from 27.8 nm to 86 nm after radical exposure (+ 312.1%). On the other hand, changes in the ζ potential were also observed, as the presence of the radical neutralized the surface charge and caused the agglomeration of the nanoparticles, especially in P-AgNPs (from −16.83 to −4.95 mV) and T-AgNPs (from −13.36 to −7.54 mV). In the case of CDs-AgNPs, an increase in the ζ value was observed, possibly due to neutralization of citrate by radical-mediated interactions. This, together with the increase in hydrodynamic size resulting from nanoparticle aggregation, is further supported by the carbon-layer stabilization of the cationic carbon dots. As a result, the positive surface charge increased from 7.5 to 14.4 mV. This initial damage, which affects the coating agent rather than the metal content in the core, is consistent with the findings of Ahumada and colleagues (2018). They observed rapid aggregation of AgNPs led by AAPH exposure. However, HSA-coated AgNPs remain relatively stable under radical exposure [74]. Overall, these results demonstrate that the proposed CDs-AgNPs exhibit superior stability against variations in ionic strength, pH changes, and free radicals. This shows the inherent ability of CDs to scavenge ROS, acting as a protective layer for silver nanoparticles.

Fig. 6.

Changes in the plasmon band followed by UV–Vis spectra of AgNPs exposed to peroxyl radicals at 37 °C. a CDs-AgNPs, b T-AgNPs, c P-AgNPs, d Full width at half maximum (FWHM); and e λ max evaluated for each plasmon band. All the results were obtained in triplicate

Several studies have highlighted the potential of AgNPs to modulate ROS levels within the intracellular environment. Depending on the circumstances, they can either exacerbate ROS levels, leading to oxidative damage, or exhibit ROS scavenging properties, thereby reducing ROS levels [75]. The results presented here regarding the stability of nanohybrid structures are consistent with previous research. The developed polymer shell of the CDs acts as an electron donor, protecting the AgNPs from instabilities caused by radical exposure. For instance, Nakal-Chidiac and coworkers (2020) developed a silver nanocluster via a photochemical method that involved irradiating I-2959 with UV light in the presence of chitosan, which served as both a ligand and stabilizer. This nanomaterial exhibited enhanced resistance to oxidation by hydrogen peroxide (H₂O₂) and to varying pH conditions during incubation, which is attributable to the dense multiamino polysaccharide coating [76]. Concurrent studies, such as those by Badi and coworkers (2021), have shown that the obtained nanohybrid structures (FeNCDs) exhibit peroxidase activity due to a synergistic effect between the CDs and the metal structure (Fe) [77]. Furthermore, Li and colleagues (2022) evaluated N-doped CDs, demonstrating behavior similar to that of nanozymes through peroxide activity with the superoxide radical (O2⁻), which promotes the stabilization of the nanomaterial in a state of ROS imbalance within a biological context [78]. Therefore, these findings provide insights into a new synthesis pathway for designing novel nanomaterials, due to the enhanced resilience of AgNPs for biomedical applications. The introduction of carbon dots into nanocomposites can enhance ROS scavenging activity and prolong the lifetime of AgNPs [79, 80].

Antioxidant activity

Based on the established stability against free radicals, their antioxidant activity was further evaluated using three commonly explored methods: DPPH assay to determine stable free radicals scavenging activity, ferric-reducing antioxidant power (FRAP) to assess the level of metal ions reduction, and total reactive antioxidant power (TRAP) as a competitive method [81]. The evaluation of the antioxidant activity of AgNPs is crucial to prevent and alleviate the damage they can cause to biological systems, especially when ROS production increases. This capability can improve the resistance of nanomaterials and the biological environment to oxidative stress, which has been reported to be altered in several pathophysiological processes [82, 83]. To determine the antioxidant activity of the formulated AgNPs, the NANoPoLC algorithm was used to estimate nanoparticle concentration (Eq. 2), which requires only knowledge of the metal salt concentration, polydispersity value, size, and morphology [84]. The calculated results showed concentrations related to the synthesis process of 0.374 nM for CDs-AgNPs, 0.308 nM for T-AgNPs, and 0.31 nM for P-AgNPs.

2

In almost all cases analyzed, AgNPs formulations demonstrated superior antioxidant activity compared to the control antioxidants (gallic acid and BHT), as shown in Table 1. For CDs-AgNPs, concentrations of 0.576 mM for TRAP and 0.015 mM for FRAP were determined, with an IC₅₀ of 2.14 nM in the DPPH assay. In contrast, T-AgNPs exhibited concentrations of 0.3 mM, 0.01 mM, and 0.74 nM for TRAP, FRAP, and DPPH, respectively. Meanwhile, P-AgNPs showed TRAP and FRAP values of 0.32 mM and 0.0095 mM, respectively, but did not demonstrate DPPH radical scavenging activity. At first glance, these results suggest that all formulations could be excellent antioxidants; however, it is important to note that these effects are subject to certain experimental limitations. For example, when comparing organic molecules with a metal surface, specific sites (functional groups) that act as electron donors are identified [85]. In contrast, AgNPs act as electron reservoirs, leading to the first significant difference in comparison [86]. In addition, reaction time is a vital factor to consider. Along with the use of standardized methods to assess antioxidant activity, the results are consistent with previously demonstrated trends in the evaluation of AgNPs' stability against peroxyl radicals [87, 88]. Therefore, it is essential to note that further research is needed to better understand and clarify the mechanism underlying this superior stability, especially in more complex biological systems. However, based on the AAPH and antioxidants results, it is evident that the CDs-AgNPs demonstrated superior stability when compared to the other AgNPs formulations, which can be attributed to the presence of cationic CDs, as previous reports have indicated an electron/hydrogen transfer mechanism that enhances their ROS scavenging capacity [89, 90]. In this context, previous studies using the DPPH assay have reported nanohybrid structures of AgNPs combined with carbon nanomaterials, demonstrating improved antioxidant activity compared to AgNPs without the carbon coatings [91, 92]. It is interesting to note that the antioxidant capability of the nanohybrid structure may be related to the N-doped carbon dots developed by PEI-CS. For instance, Chen and coworkers (2020) developed CDs via the hydrothermal method using CS as a carbon source, demonstrating increased hydroxyl scavenger activity by transferring hydrogen atoms from numerous NH₂ groups [93]. Additionally, the FRAP assay, which measures the reduction of Fe⁺³ to Fe⁺², has previously been evaluated for CDs prepared using sodium citrate and 4-aminophenyl boronic acid synthesized by the hydrothermal method. In this work by Ahlawat and coworkers (2022), it was concluded that these CDs possess antioxidant properties, restoring cellular homeostasis and exerting a preventive effect on the transformation of amyloid proteins [94]. Similar properties of CDs have been evaluated using the ABTS method, which showed stronger antioxidant activity than the TRAP and DPPH assays, using mushrooms to produce CDs via the hydrothermal method [95].

Table 1.

Total antioxidant capacity per nanoparticle

	TRAP (TEAC mM)	FRAP (TEAC mM)	DPPH (IC50 nM)
CDs-AgNPs	0.5760 ± 0.0372b	0.0150 ± 0.0004	2.1418 ± 0.0469a
T-AgNPs	0.3060 ± 0.0188a	0.0104 ± 0.0004	0.7436 ± 0.0466b
P-AgNPs	0.3268 ± 0.0247a	0.0095 ± 0.0002	n.d
Gallic acid	1.1280 ± 0.0128c	1.7229 ± 0.0183a	n.d
BHT	1.0688 ± 0.0222d	1.5202 ± 0.0769b	2.7252 ± 0.0112c

n.d.: Not detectable, ^a−d Super index letters show significant differences p < 0.05

The results showed mean values ± standard deviation (n = 5) for the different nanoparticles and antioxidant references

TRAP, ABTS· radical scavenger; FRAP, Ferric ion reductase capacity; and DPPH, DPPH· radical scavenger

Cytotoxicity

The potential cytotoxicity of several AgNPs was evaluated in different cell lines, including HT-29, PC-3, and MCF-7 as cancer cell line models, and MCF-10 as a non-cancer cell line, using a Sulforhodamine B (SRB) assay (Fig. 7). These results revealed differences in cytotoxicity, influenced by the synthesis methods. In response to AgNPs exposure, MCF-7 human breast cancer cells demonstrate susceptibility to the highest concentration of AgNPs among the three synthesis methods, compared to the other evaluated cell lines. Furthermore, the CDs-AgNPs showed greater potential to enhance cytotoxicity in the breast cancer cell line MCF-7 (Fig. 7a). However, this nanomaterial also showed decreased citocompatibility in the non-tumorigenic epithelial cell line MCF-10, which was not observed with T-AgNPs (Fig. 7b) or P-AgNPs (Fig. 7c). On the other hand, in HT-29 and PC-3 cancer cell lines, none of the AgNP formulations showed a significant cytotoxic effect. The specific susceptibility of the MCF-7 cell line to AgNP exposure, produced using various reducing agents, has been previously documented. For instance, Vivek and coworkers (2012) reported an increase in apoptotic cell induction in MCF-7 cells compared to normal breast cells (HBL-100) after exposure to AgNPs synthesized with Annona squamosa leaf extract as a reducing agent [96]. Furthermore, Khatoon and collaborators (2023) obtained AgNPs that showed a greater cytotoxic effect using NaBH₄ as a reducing agent [97]. Additionally, according to Ghosal and colleagues (2020), after evaluating the application of CDs-AgNPs nanocomplexes in a MCF-7 cell line, they observed a synergistic cytotoxic effect by increasing intracellular ROS, which led to cell apoptosis [98]. However, they did not test such cytotoxic effect in a non-tumoral cell line. Regarding the susceptibility of MCF-10 to CDs-AgNPs exposure, the development of CDs involves the use of CS and PEI as carbon sources. Previous research by Santos and collaborators (2023) demonstrated that application of PEI in the formulation produced highly toxic CDs at the highest concentration of 500 μg/mL, as evaluated in NIH/3T3 fibroblasts, which could explain the cytotoxic effect of the nanohybrid material formulation [42]. At this stage, further research is needed to uncover the mechanism that distinguishes the responses of these cells from those of others evaluated. However, the observed results in cytotoxicity and antioxidant activity may be linked to a pro-oxidant mechanism. In this regard, CDs-AgNPs demonstrated greater stability against peroxyl radicals and better antioxidant capacity than other silver nanoparticles. While these properties can enhance the resistance of nanomaterials and the biological environment to oxidative stress, an excess of these molecules can have the opposite effect, becoming pro-oxidant and inducing reductive stress [99]. It has been reported that several molecules typically considered important antioxidants, when present at high concentrations, can act as pro-oxidants, increasing ROS levels, oxidative damage, and cytotoxicity [100–103].

Fig. 7.

Cytotoxicity assay of HT-29, MCF-7, MCF-10, and PC3 cell lines after exposure to different concentrations of AgNPs. a CDs-AgNPs, b T-AgNPs, and c P-AgNPs. ^a−d Super index letters show significant differences p < 0.05

Conclusion

In this work, AgNPs were successfully synthesized using CDs as reducing and capping agents via a photochemical pathway, providing a green, rapid, and reproducible method for producing these metal nanoparticles. Furthermore, to understand and compare the stability properties of the CDs-AgNPs, the results were contrasted with two common synthesis methods that used the same starting reagents, differing only in the presence of CDs. In this regard, the AgNPs analyzed in this study were similar in size (< 10 nm) and quasi-spherical in shape. Additionally, spectroscopic analyses and in vitro studies showed that the proposed CD-AgNPs formulation exhibited superior stability under various conditions, including high ionic stability up to 500 mM NaCl, pH stability between 4.5 and 8, and resistance to free radical destabilization for up to 2 h, while traditional synthesis methods destabilized within the first minute of exposure to these conditions. Similar results were also observed when evaluating their antioxidant activity. These findings clearly show that the CDs serve as a protective layer, closely associated with the polymeric surface of this type of nanomaterial. This protective layer not only reduces the agglomeration and oxidation rates of AgNPs upon exposure to biologically relevant free radicals but also enhances antioxidant activity due to the high abundance of amino groups. Overall, the observed scavenging activity indicates that the nanohybrid structure of CDs-AgNPs could act as a modulator in biological redox processes. In contrast, all AgNPs formulations exhibited similar cytotoxic profiles against both cancerous and non-cancerous cell lines, with some exceptions. Although all AgNPs showed cytotoxicity in MCF-7 cells, the cytotoxic effect of CDs-AgNPs was greater. However, a similar effect was also seen in MCF-10, a non-tumorous cell line, at concentrations above 15 nM, unlike T-AgNPs and P-AgNPs, which demonstrated lower or negligible cytotoxicity. While the mechanism behind the link between the enhanced stability, redox properties, and observed cytotoxicity was not part of the study, it is hypothesized that CDs-AgNPs act as pro-oxidants at higher concentrations. In this regard, the current study has limitations that should be weighed and considered to guide future research. For example, a detailed evaluation of the nanocomposite structure is necessary to determine the type of composite formed during AgNPs synthesis. Additionally, regarding cellular activity, it is worth investigating whether the nanoparticle acts as an antioxidant and how it interacts with different cell lines. Although more research is needed to understand the biological interactions of CDs-AgNPs, this study offers valuable insights into designing AgNPs for biomedical applications. It confirms that creating hybrid nanostructures with cationic carbon dots can extend the lifespan of metal-based nanomaterials in biological environments, reduce unwanted side effects caused by instability, and potentially provide a therapeutic effect under biological conditions.

Author contributions

NS performed the comprehensive investigation (experimental section) and helped with the formal analysis and manuscript writing. JM contributed to investigating the stability of the nanocomposites and assisted with manuscript revisions. MSA contributed to the DLS, ζ potential, lifetime measurements, results analysis, and manuscript writing. CJ and JV contributed to performing the antioxidant and cytotoxic *in vitro* assays, as well as the results analysis and manuscript writing. MA contributed to conceptualization, formal analysis, methodology, resources, supervision, and overall manuscript writing.

Funding

This research was funded by Agencia Nacional de Investigación y Desarrollo ANID-FONDECYT Regular 1241229 and ANID-Vinculacion Internacional-FOVI240066 (MA).

Data availability

Data will be made available on request.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent to publication

Not applicable.

Competing interest

The authors declare no competing interests.