Controlling nanoparticle aggregation via gel point shift in the in situ photochemical synthesis of plasmonic epoxy-based nanocomposites

Abstract

Plasmonic metal/polymer nanocomposites are hybrid materials that integrate the intrinsic properties of metal nanoparticles (NPs) with the mechanical strength and thermal stability offered by the polymer matrix. These materials can be fabricated using in situ photochemical synthesis, in which both the metal NPs and the polymer matrix are generated simultaneously in a one-pot process. While this approach is rapid and simple, the ability to control the organization of particles within the polymer matrix remains a challenge. This is a key factor for the preparation of plasmonic materials with well-tuned properties. In this study, we demonstrate that the gel point of a crosslinking epoxy matrix can serve as an effective tool to control the aggregation process of in situ generated silver NPs. The epoxy matrix was formulated as a copolymerization reaction of a diepoxy monomer (diglycidyl ether of bisphenol A, DGEBA) with a monoepoxy monomer (phenyl glycidyl ether, PGE). By varying the DGEBA/PGE ratio, we were able to precisely shift the gel point of the matrix, during the simultaneous formation of silver NPs. This behavior was monitored using dynamic rheology. SAXS analysis, combined with UV-visible spectroscopy and transmission electron microscopy (TEM), allowed us to demonstrate that shifting the gel point toward higher conversions leads to a dimerization process of primary silver NPs, resulting in a significant increase in the plasmonic response of the material. Based on the experimental evidence presented, the dimerization mechanism is discussed.

---

Introduction

Noble metal nanoparticles (NPs), such as gold and silver, exhibit unique optical properties determined by their surface plasmons (collective oscillation of the conduction band electrons), which depend on factors such as shape, size, organization, and the refractive index of the medium in which they are dispersed.^1–3 This tunability of plasmonic properties enables the optimization of their optical response for specific applications in fields such as sensing, imaging, and energy harvesting, among others.^1,4,5 When metal NPs are incorporated into a polymer matrix to create a hybrid nanocomposite, the overall performance of the material is significantly improved due to the properties of both components. These materials combine the mechanical robustness and thermal stability provided by the matrix with the intrinsic properties of metal NPs. Furthermore, it is possible to regulate specific properties of the material through the influence that the matrix can exert on the morphology and organization of the NPs.⁶ This synergy between both components serves as a valuable tool for creating innovative designs and functionalities for a wide range of applications.

Metal/polymer nanocomposites can be obtained using either ex situ or in situ methods. The ex situ approach involves an initial step where well-defined metal NPs are fabricated, followed by a subsequent step in which these preformed particles are incorporated into a polymer matrix.⁷ Alternatively, the NPs can be mixed with reactive monomers, which are then polymerized to form the matrix.⁸ Frequently, an intermediate synthetic step is necessary in which the surface of the particles is functionalized with appropriate molecules or polymer chains to help them maintain colloidal stability during their integration into the matrix.^9,10

Within these ex situ methods, various strategies have been reported to optimize or modulate the plasmonic properties of metal/polymer nanocomposites. For instance, techniques such as nanoprinting,¹¹ electron beam lithography^12,13 and ion beam lithography¹⁴ have been employed to generate metal NPs with controlled size and distribution. However, these technologies are too expensive for scaling up to large surface areas. Alternatively, colloidal synthesis^15–17 is a valuable method for preparing metal NPs with well-defined size and shape, although the ability to control the arrangement of these particles during their incorporation into the matrix is somewhat limited. To overcome this challenge, metal–organic frameworks (MOFs),¹⁸ carbon nanotubes (CNTs)¹⁹ and block copolymers²⁰ have been utilized as templates to guide the distribution and orientation of metal NPs in polymer matrices. Despite the promising results from laboratory-scale synthesis of these nanocomposites, scaling up the processes for industrial applications remains a challenging task.

On the other hand, in situ preparation methods offer a faster and simpler approach to creating metal/polymer nanocomposites, avoiding multi-step procedures and complex synthetic protocols. These methods involve the direct formation of metal NPs within the polymer matrix via chemical or photochemical reduction of a metal salt precursor previously dissolved in the matrix.^21,22 For example, this concept has been employed to prepare several metal/polymer nanocomposites using high-temperature chemical reduction to transform gold or silver ions adsorbed onto polymers, including chitosan,²³ poly(methyl methacrylate),^24,25 acrylic²⁶ and polystyrene.^27,28 Although these examples have shown that the size and shape of the resulting NPs can be controlled by adjusting parameters such as temperature and metal precursor concentration, high-temperature conditions often lead to NPs with high polydispersity,²⁹ which can limit the use of these plasmonic nanocomposites in various applications. For this reason, in situ photochemical methodologies performed at room temperature offer a more convenient approach. Moreover, light-mediated synthesis provides spatiotemporal control and allows for easy modulation of reaction kinetics by adjusting the irradiation intensity.^21,30

Yagci et al.³¹ first reported on the fabrication of silver/epoxy nanocomposites through simultaneous processes of visible-light-induced electron transfer and cationic polymerization. This approach involves the excitation of a photoinitiator upon irradiation with visible light, leading to the generation of free radicals, which reduce a metal salt resulting in the formation of silver NPs. Concurrently, a strong Brønsted acid is produced, which initiates the cationic ring-opening polymerization of epoxy monomers. In this approach, silver NPs and epoxy matrix are generated concomitantly in a one-pot process. The same concept was also applied for the preparation of gold/epoxy^32,33 and silver/methacrylate^34,35 nanocomposites.

In previous work, we investigated the particle formation mechanism during the in situ photochemical synthesis of silver/epoxy plasmonic materials, which were fabricated using this method of simultaneous metal reduction and matrix polymerization.³⁶ Through time-resolved synchrotron SAXS experiments combined with dynamic rheology measurements, we demonstrated that particle nucleation begins at very early stages of irradiation, followed by a growth process dominated by Ostwald ripening and dynamic coalescence mechanisms. This particle growth regime, driven by diffusion and coalescence processes, is arrested when polymerization of the epoxy matrix reaches the gel point. Therefore, our findings led us to hypothesize that the gel point could be used as a valuable synthetic tool to control the aggregation process of metal NPs within the epoxy matrix. In the present work, we tested this hypothesis.

To achieve this, silver NPs were generated in situ during the copolymerization of a diepoxy monomer (diglycidyl ether of bisphenol A, DGEBA) with a monoepoxy monomer (phenyl glycidyl ether, PGE). By varying the DGEBA/PGE ratio, we were able to precisely shift the gel point of the epoxy matrix, as demonstrated through dynamic rheology. SAXS analysis, combined with UV-visible spectroscopy and transmission electron microscopy (TEM), allowed us to show that under the irradiation conditions used, the shift of the gel point toward higher conversions leads to a dimerization process of primary particles, resulting in a significant increase in the plasmonic response of the material. These findings are highly valuable for developing simple and rapid synthetic protocols that yield plasmonic nanocomposites with well-tuned properties.

Experimental section

Materials

For this study, a diglycidyl ether of bisphenol A (DGEBA, Sigma-Aldrich DER 332) epoxy monomer with an epoxy equivalent weight of 174.3 g per equiv., was used. Phenyl glycidyl ether (PGE), silver hexafluoroantimonate (AgSbF₆), and poly(ethylene glycol) (PEG, M_n = 200 g mol⁻¹) were supplied by Sigma-Aldrich. Boron trifluoride monoethylamine complex (BF₃·MEA, Sigma-Aldrich) and camphorquinone (CQ, Sigma-Aldrich) were used as thermal initiator and photoinitiator, respectively, for the cationic ring-opening polymerization of epoxy monomers. Fig. 1 shows the chemical structures of these materials, which were used as received.

Fig. 1 Chemical structures of the materials used in this study: (a) phenyl glycidyl ether, (b) camphorquinone, (c) BF₃·MEA, and (d) DGEBA.

Sample preparation

Silver/epoxy nanocomposites were prepared by in situ photochemical synthesis following a previously reported method.³⁶ First, a silver salt solution was obtained by dissolving AgSbF₆ (50 mg) in PEG (200 μL), at room temperature. On the other hand, different mixtures of DGEBA and PGE monomers were prepared, with the PGE content varying from 0 (neat DGEBA) to 75 wt%. Both components are easily miscible at room temperature and the resulting blends were named DGEBA/PGE (100 − x) : x, being x the mass percentage of PGE. Then, an appropriate amount of the silver solution was added to each DGEBA/PGE blend to obtain samples with 2 wt% AgSbF₆. Finally, these samples were photoactivated by incorporating 2 wt% CQ, and then irradiated at room temperature using a blue LED source (OptoTech, Germany) emitting in the wavelength range between 410–530 nm, with an irradiation intensity of 15 mW cm⁻².

Characterization techniques

Fourier transform infrared spectroscopy (FTIR). To monitor the conversion of epoxy groups during irradiation time, a Nicolet 6700 Thermo Scientific IR device was used. Measurements were performed in the near-IR range (4000–7000 cm⁻¹), from 32 co-added scans with 4 cm⁻¹ resolution. The sample was sandwiched between two glass slides separated by a 50 μm Teflon spacer, and then irradiated at room temperature during regular time intervals. Immediately after each irradiation, a spectrum was collected. The same IR device was also used to evaluate the conversion of epoxy groups in DGEBA/PGE blends polymerized at 140 °C, with BF₃·MEA as thermal initiator (at a ratio of 0.048 mol BF₃·MEA/mol of epoxy groups). In this case, the sample was placed in a heated transmission cell (HT-32, Spectra-Tech) equipped with a programmable temperature controller (Omega, Spectra-Tech, ΔT = ±1 °C). The height of the epoxy peak located at 4530 cm⁻¹ was used to calculate the degree of conversion, taking the peak at 4620 cm⁻¹ as a reference.^37,38

UV-Visible spectroscopy. An Agilent 8453 diode-array spectrophotometer was used to measure UV-vis spectra in the range between 200–800 nm. The sample was sandwiched between two quartz slides separated by a 50 μm Teflon spacer, and then irradiated at room temperature during regular time intervals. Spectra were collected immediately after each irradiation, using a clean quartz slide as a blank. The formation of Ag nanoparticles was assessed by following the evolution of the plasmon band at 410 nm.

Rheometry. An Anton Paar rheometer (Physica MCR-301) equipped with a CTD 600 thermochamber was used to monitor changes in storage (G′) and loss (G″) moduli during polymerization of DGEBA/PGE blends at 140 °C, with BF₃·MEA as thermal cationic initiator. Measurements were performed using a parallel-plate configuration (25 mm diameter and 1 mm gap), in an oscillatory mode at 1 Hz frequency and 1% amplitude. The gel point was assigned as the crossover point between G′ and G″.³⁹

Small-angle X-ray scattering (SAXS). A XEUSS 1.0 HR (XENOCS, Grenoble) apparatus equipped with a Pilatus 100 K detector (Dectris, Switzerland) and a GENIX 3D micro-focus X-ray tube (λ = 0.15419 nm) was used to record SAXS curves at room temperature. The sample-detector distance was 1354 mm. Measurements were performed with an acquisition time of 3 min to guarantee good quality SAXS patterns. Circular integration of the 2D data was performed through standard procedures. 1D-SAXS curves were analyzed with the SASfit software.^36,40,41

Transmission electron microscopy (TEM). An LKB ultramicrotome equipped with a diamond knife was used to section the silver/epoxy nanocomposites at room temperature. Sectioned samples of 60 nm thickness were obtained and collected on copper grids for observation. TEM images were acquired with a JEOL 100CX transmission electron microscope at an acceleration voltage of 80 kV.

Results and discussion

Recently, we investigated the mechanism of particle formation in metal/epoxy nanocomposites prepared by in situ photochemical synthesis.³⁶ In this method, silver NPs are generated during polymerization of the epoxy matrix in a one-pot process. As depicted in Scheme S1, visible-light irradiation produces the excitation of CQ, which then reacts with a hydrogen donor (which can be the monomer itself or another reducing agent) generating a ketyl free radical. This radical reduces the metallic salt forming silver NPs and a strong Brønsted acid, which in turn initiates the cationic ring-opening polymerization of epoxy monomers. It was shown that silver NPs grow rapidly during the pre-gel period of the polymerization reaction through diffusion and coalescence processes. However, these processes are arrested by diffusional constraints at the gel point, where the particle mean size reaches a maximum value. These previous findings motivated us to explore the possibility of using the gelation of the epoxy matrix as a versatile synthetic tool to control the particle growth process.

With this idea in mind, we first analyzed different DGEBA/PGE blends to analyze how the gel point is affected by the diepoxy/monoepoxy ratio in the matrix. The gel point was determined by rheometry of DGEBA/PGE samples polymerized at 140 °C with BF₃·MEA as thermal cationic initiator. Fig. 2 and S1 show the evolution of the storage (G′) and loss (G″) moduli as a function of reaction time for samples with different DGEBA : PGE weight ratios. From the crossover point between G′ and G″, it can be noted that the gel point shifts towards longer reaction times as the DGEBA content in the mixture decreases. This is an expected behavior, since DGEBA provides the crosslinking points for network formation.

Fig. 2 Evolution of the storage (G′) and loss (G″) moduli during the cationic ring-opening polymerization at 140 °C of blends with different DGEBA/PGE weight ratios: (a) 100 : 0; (b) 40 : 60; and (c) 30 : 70.

Near-infrared spectroscopy was used to determine the overall conversion of epoxy groups at the gel point time for each DGEBA/PGE blend analyzed (see Fig. S2 and S3). For a proper comparison, the reaction conditions used in both devices (rheometer and IR spectrometer) were nearly identical. As shown in Fig. 3, the gel point conversion increases very slowly at low PGE concentrations but rises abruptly above 50 wt% PGE. Based on these results, we selected formulations containing between 50 and 75 wt% PGE, for which gelation is significantly delayed. Note that the behavior shown in Fig. 3 (measured at 140 °C) is assumed to be comparable to that at room temperature, at which silver/epoxy nanocomposites are prepared. This assumption is based on the fact that the same epoxy polymerization mechanism is used at both temperatures, i.e., cationic ring-opening polymerization, which is relatively little affected by temperature in the range studied.⁴² Furthermore, the obtained values are in good agreement with those reported by Vidil et al.⁴³ for the cationic copolymerization of DGEBA with PGE initiated by tetrafluoroborate at 40 °C. Using Fourier transform mechanical spectroscopy, the authors found that gelation occurs at conversion values of 0.07, 0.32 and 0.47 for blends with 0, 65 and 75 wt% PGE, respectively.⁴³

Fig. 3 Evolution of the gel point conversion as a function of PGE concentration for DGEBA/PGE blends polymerized at 140 °C, with BF₃·MEA as cationic initiator. The blue line was drawn to guide the eye.

Then, DGEBA/PGE solutions containing AgSbF₆ and CQ were irradiated at room temperature with a blue LED source. During irradiation, the overall conversion of epoxy groups was monitored by near-IR spectroscopy. The results obtained for the 100 : 0, 50 : 50 and 25 : 75 DGEBA/PGE samples are shown in Fig. 4. As can be seen, the three formulations exhibited a comparable kinetic behavior at early stages of the curing process, indicating that the initial polymerization rate is not significantly affected by the diepoxy/monoepoxy ratio. These results denote that the chemical reactivity of the epoxy groups provided by DGEBA and PGE is practically the same. However, after approximately 50 min of irradiation, the 100 : 0 sample deviated and did not reach full conversion due to vitrification effects. Clearly, the addition of PGE produced a decrease in the glass transition temperature (T_g) of the network. This behavior is expected, as a decrease in crosslinking density results in greater chain flexibility and an increase in the number of accessible conformations, which ultimately leads to a reduction in T_g. To further support this interpretation, differential scanning calorimetry (DSC) measurements were conducted, revealing that T_g decreases from 103 °C for the 100 : 0 sample to 26 °C and 3 °C for the 50 : 50 and 25 : 75 samples, respectively (Fig. S4).

Fig. 4 Overall conversion of epoxy groups versus irradiation time for samples containing AgSbF₆, CQ and different DGEBA/PGE weight ratios. Irradiations were carried out at room temperature with a blue LED source (15 mW cm⁻², in the range 410–530 nm).

Fig. 5a displays UV-vis absorption spectra recorded during irradiation at room temperature for the 100 : 0 DGEBA/PGE sample containing AgSbF₆ and CQ. As can be seen, a plasmon absorption band centered at 410 nm developed with time due to the formation of silver NPs. The growth rate of this band became increasingly slower after 900 s irradiation (about 0.05 conversion), which coincides, within experimental error, with the gel point for this formulation. Under the same irradiation conditions, the 25 : 75 sample exhibited a much sharper increase of the plasmon band, as shown in Fig. 5b. In fact, the absorption intensity increased until reaching the saturation levels of the UV-vis detector in the spectrophotometer after 1500 s of irradiation (∼0.2 conversion), while the sample was still in the pre-gel period (see Fig. 3). These curves, which are indicated in red color in Fig. 5b, reveal a substantial enhancement in the plasmonic properties of this material. In addition, a small red-shift of the absorption maximum was observed with irradiation time, consistent with a gradual increase in particle size, as illustrated in Fig. S5.

Fig. 5 Evolution of the plasmon band of silver NPs during irradiation at room temperature of samples containing AgSbF₆, CQ and different DGEBA/PGE weight ratios: (a) 100 : 0 and (b) 25 : 75. In (c), the maximum absorbance of the plasmon band is plotted as a function of irradiation time for the formulations shown in (a) and (b). Irradiations were carried out with a blue LED source (15 mW cm⁻², in the range 410–530 nm). The spectra shown in red color in (b) correspond to irradiation times at which the plasmon band reached saturation levels of the UV-vis detector in the spectrophotometer.

For comparison purposes, the maximum absorbance of the plasmon band was plotted as a function of irradiation time for both formulations (Fig. 5c). Note that all three graphs in Fig. 5 have the same scale on the absorbance-axis. As can be seen, the plasmon band not only reached higher absorbance values but also increased at a much faster rate in the 25 : 75 sample compared to the 100 : 0 sample. Although the monoepoxy monomer was incorporated to shift the gel point, it must be taken into account that the addition of PGE also produces a decrease in the viscosity of the reaction medium during the pre-gel period, which favors the diffusion and coalescence of particles. This could explain why the plasmon band evolved at different rates in both samples.

The silver/epoxy nanocomposites obtained after irradiation at room temperature were analyzed by SAXS. This technique provides statistically reliable information on the shape and size of individual particles, as well as the type of interaction between neighboring particles. Fig. 6a shows the scattering intensity curves in a double logarithmic diagram for samples with different DGEBA/PGE ratios. The 100 : 0 sample exhibits a plateau in the low-q range (Guinier regime) followed by a Gaussian decay at higher q values, without the presence of spatial correlation peaks. This profile is characteristic of a set of uncorrelated quasi-spherical nanoparticles embedded in a homogeneous matrix, which is consistent with our previous results.³⁶ As can be seen, the SAXS curves corresponding to different DGEBA/PGE ratios exhibited marked differences in the low-q range. In this region, the slope of the scattering curves becomes progressively more negative with the PGE content in the formulation, which is consistent with an aggregation process of primary particles that evolves as the gel point is delayed.

Fig. 6 SAXS analysis of silver/epoxy nanocomposites formulated with different DGEBA/PGE ratios. (a) Scattering intensity (I) as a function of the scattering vector (q), and (b) Pair-Distance Distribution Function (PDDF).

To gain insight into the particle aggregation process, SAXS data were used to calculate the Pair-Distance Distribution Function (PDDF) for each formulation, using the SasView 3.1.2 software package.⁴⁴ Further details of the PDDF calculation procedure are provided in the SI. The PDDF represents the histogram of correlation distances (r) between two points in a particle. For example, for a spherical particle, the most probable correlation distance (maximum in the PDDF(r) curve) is the radius of the particle; while for any finite particle, the value of r_max where PDDF(r) = 0 represents the size of the particle.

As shown in Fig. 6b, the PDDF(r) curve for the 100 : 0 sample presents a unimodal distribution, with a peak located at 5.8 nm and a r_max value of 19.8 nm. This information, in combination with that of Fig. 6a, allows us to infer the presence of a distribution of spherical primary NPs with a mean radius of 5.8 nm and a maximum particle diameter of about 20 nm (suggesting also particle size distribution). For the 50 : 50 sample, the PDDF(r) curve clearly shows two particle populations, one of them corresponding to primary NPs with a mean radius of 8.4 nm, and the other corresponding to aggregates virtually formed by two primary particles. Fig. S6 shows the deconvolution of the 50 : 50 PPDF(r) curve, where the distribution function corresponding to aggregates presents a maximum at 19.8 nm, i.e. approximately twice that of the primary particles. For 40 : 60, 30 : 70 and 25 : 75, the aggregate population has a larger contribution than that of primary NPs. Note that r_max in these cases (∼40 nm) is twice that of primary NPs in the 100 : 0 sample. These results suggest that the addition of PGE promotes a controlled aggregation process, where primary particles tend to aggregate in pairs. This aggregation process evolves with the PGE content, as the gel point is delayed.

Model SAXS curves were fitted to the SAXS data shown in Fig. 6a using the SASfit software.⁴¹ Data corresponding to the 100 : 0 sample were modeled assuming a form factor of spherical objects with a log–normal size distribution. Regarding the samples containing PGE, two contributions to the form factor were considered: one corresponding to spherical primary particles and the other to aggregated structures. A log–normal size distribution was assumed for each of these components. The best-fitting parameters for each sample are presented in the SI (Table S1). An excellent fitting of the SAXS models to the data was achieved with these parameters, as shown in Fig. 7.

Fig. 7 Fitting of the experimental SAXS data shown in Fig. 6a. Data were modeled using the SASfit software package.⁴¹ The solid lines are the fitting curves, and the symbols represent the experimental data. SAXS curves were displaced vertically for visual clarity.

The silver/epoxy nanocomposites obtained after irradiation at room temperature were also analyzed by transmission electron microscopy (TEM). Fig. 8a and b shows TEM images from ultrathin sections of the 100 : 0 and 25 : 75 samples, respectively. As can be seen, a dispersion of spherical silver NPs homogeneously distributed in the epoxy matrix was obtained for the 100 : 0 sample (Fig. 8a). These NPs have a maximum diameter of 18.4 nm, as assessed by TEM image analysis using the Image-Pro Plus software. On the other hand, the 25 : 75 sample (Fig. 8b) presents a dispersion of dimeric structures with different degrees of coalescence. The inset in Fig. 8b shows a dimer of primary silver NPs at higher magnification. These structures are consistent with our previous predictions from SAXS analysis. Fig. 8c and d shows histograms of particle size distributions obtained by TEM image analysis for the 100 : 0 and 25 : 75 samples, respectively. Superimposed on these histograms are the particle size distributions obtained by fitting the SAXS curves for these samples, as shown in Fig. 7. As can be seen, there is good agreement between the SAXS results and the TEM image analysis. Additional TEM images and the corresponding particle size histogram for an intermediate composition (50 : 50) are provided in the SI (Fig. S7), further supporting the influence of blend composition on nanoparticle formation.

Fig. 8 Top: TEM images from ultrathin sections of samples with different DGEBA/PGE weight ratios: (a) 100 : 0 and (b) 25 : 75. In both images, the scale bar indicates a length of 50 nm. The inset in (b) shows a dimer of primary silver NPs at higher magnification. Bottom: Histograms of particle size distributions obtained by TEM image analysis for: (c) 100 : 0 and (d) 25 : 75. In these plots, the blue lines represent the particle size distributions derived from fitting the SAXS curves shown in Fig. 7.

Previous studies have shown that the random aggregation of metal nanoparticles follows step-growth kinetics.^45,46 In this process, two primary NPs aggregate to form a dimer, which can then combine with other primary particle to create a trimer or with another dimer to form a tetramer, and so on. Therefore, dimerization is only a transient intermediate step of the overall aggregation process, which ultimately leads to the formation of bulk aggregates. Our ability to finely tune the gel point by adjusting the DGEBA/PGE ratio allowed us to effectively quench the particle aggregation process when dimeric structures became the predominant aggregates. A typical feature of step-growth kinetics is that the aggregation number increases slowly with time, but accelerates markedly in the later stages of the process. Therefore, the possibility of freezing the aggregation process at relatively short timescales provides a valuable synthetic tool to capture structures with a low aggregation number, such as dimers in this work.

As previously explained, the formation of silver NPs resulted in the development of a plasmonic band centered at 410 nm (Fig. 5). This plasmonic effect is due to collective oscillations of conduction electrons on the surface of the NPs when they are irradiated with visible light. These oscillations generate a strong electromagnetic field near the surface of the particles. Now, when two primary NPs come close to form a dimer, the electromagnetic field strength in the junction between the particles increases dramatically due to a coupling effect between the particles. This plasmon coupling manifests itself as a marked increase in light absorption, as seen in Fig. 5b.

Plasmonic dimers are of extreme interest in a vast range of applications, such as surface-enhanced Raman spectroscopy (SERS), photothermal therapy, material processing, sensing, and heat-assisted nanochemistry.^46–48 Therefore, the method presented in this work offers a simple and flexible synthetic approach for the in situ one-pot preparation of plasmonic nanocomposites with significant technological relevance. The method can be naturally applied in photo-curing 3D printing technologies.

Conclusions

We studied the effect of the gel point of an epoxy matrix on the arrangement and distribution of in situ generated silver NPs. The epoxy matrix was formulated as a copolymerization reaction of DGEBA and PGE. By varying the diepoxy/monoepoxy ratio, we were able to precisely shift the gel point of the matrix during the simultaneous formation of silver NPs. We demonstrated that shifting the gel point toward higher conversions leads to a dimerization process of primary silver NPs, resulting in a significant increase in the plasmonic response of the material. It is well known that dimerization is only a transient intermediate step in the overall aggregation process, which ultimately leads to the formation of bulk aggregates. Our ability to precisely tune the gel point allowed us to effectively quench the particle aggregation process when dimeric structures became the predominant aggregates. These results demonstrate that the gel point can be used as a valuable synthetic tool to control the aggregation process of silver NPs within the epoxy matrix and, in this way, fabricate plasmonic nanocomposites with well-tuned properties.

Author contributions

Agustina B. Leonardi: investigation; formal analysis; writing – review & editing. Nancy M. Cativa: investigation; formal analysis; writing – review & editing. Gustavo F. Arenas: investigation; formal analysis; methodology; funding acquisition; writing – review & editing. Marcelo Ceolín: investigation; formal analysis; methodology; writing – review & editing. Ignacio E. dell'Erba: conceptualization; investigation; supervision; writing – review & editing. Walter F. Schroeder: conceptualization; funding acquisition; supervision; writing – original draft.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information: proposed mechanism for the in situ photochemical synthesis of silver/epoxy nanocomposites; evolution of the storage (G′) and loss (G″) moduli during polymerization at 140 °C of blends with different DGEBA/PGE weight ratios; FTIR spectra measured to determine the overall conversion of epoxy groups at the gel point for the different DGEBA/PGE weight ratios; second heating DSC scans of silver/epoxy nanocomposites prepared with different DGEBA/PGE ratios; UV–Vis spectra recorded during irradiation at room temperature for the sample formulated with a 25 : 75 DGEBA/PGE ratio; deconvolution of the Pair-Distance Distribution Function (PPDF) corresponding to the 50 : 50 sample; TEM images and the corresponding particle size histogram for the 50 : 50 sample; detailed description of the PDDF calculation procedure; best-fitting parameters from SAXS data and description of the models used. See DOI: https://doi.org/10.1039/d5nr03338a.

Acknowledgements

The financial support of the following institutions is gratefully acknowledged: National Research Council (CONICET, Argentina), National Agency for the Promotion of Research, Technological Development and Innovation (Agencia I+D+i, Argentina), and University of Mar del Plata.

References

1. Y. Guo, Z. Xu, A. G. Curto, Y. J. Zeng and D. Van Thourhout, Plasmonic semiconductors: materials, tunability and applications, Prog. Mater. Sci., 2023, 138, 101158.
2. M. P. Mcoyi, K. T. Mpofu, M. Sekhwama and P. Mthunzi-Kufa, Developments in localized surface plasmon resonance, Plasmonics, 2025, 20, 5481–5520.
3. Y. Hang, A. Wang and N. Wu, Plasmonic silver and gold nanoparticles: shape- and structure-modulated plasmonic functionality for point-of-caring sensing, bio-imaging and medical therapy, Chem. Soc. Rev., 2024, 53, 2932–2971.
4. W. Rao, D. Wang, T. Kups, E. Baradács, B. Parditka, Z. Erdélyi and P. Schaaf, Nanoporous gold nanoparticles and Au/Al₂O₃ hybrid nanoparticles with large tunability of plasmonic properties, ACS Appl. Mater. Interfaces, 2017, 9, 6273–6281.
5. Z. Gao, S. Shao, W. Gao, D. Tang, D. Tang, S. Zou, M. J. Kim and X. Xia, Morphology-invariant metallic nanoparticles with tunable plasmonic properties, ACS Nano, 2021, 15, 2428–2438.
6. P. Xu, X. Han, B. Zhang, Y. Dua and H.-L. Wang, Multifunctional polymer-metal nanocomposites via direct chemical reduction by conjugated polymers, Chem. Soc. Rev., 2014, 43, 1349–1360.
7. J. Li, H. Liu, J. Guo, Z. Hu, Z. Wang, B. Wang, L. Liu, Y. Huang and Z. Guo, Flexible, conductive, porous, fibrillar polymer-gold nanocomposites with enhanced electromagnetic interference shielding and mechanical properties, J. Mater. Chem. C, 2017, 5, 1095–1105.
8. I. E. dell'Erba, C. E. Hoppe and R. J. J. Williams, Synthesis of silver nanoparticles coated with OH-functionalized organic groups: dispersion and covalent bonding in epoxy networks, Langmuir, 2010, 26, 2042–2049.
9. K. Wieszczycka, K. Staszak, M. J. Woźniak-Budych, J. Litowczenko, B. M. Maciejewska and S. Jurga, Surface functionalization - The way for advanced applications of smart materials, Coord. Chem. Rev., 2021, 436, 213846.
10. S. V. Kaymaz, H. M. Nobar, H. Sarıgül, C. Soylukan, L. Akyüz and M. Yüce, Nanomaterial surface modification toolkit: principles, components, recipes, and applications, Adv. Colloid Interface Sci., 2023, 322, 103035.
11. B. Liu, S. Liu, V. Devaraj, J. Ai, Y. Han, J. Feng, Y. Yin and Y. Zhang, Metal 3D nanoprinting with coupled fields, Nat. Commun., 2023, 14, 4920.
12. M. H. V. Werts, M. Lambert, J. P. Bourgoin and M. Brust, Nanometer scale patterning of langmuir-blodgett films of gold nanoparticles by electron beam lithography, Nano Lett., 2002, 2, 43–47.
13. M. Tayyab, L. Zizhe, S. Rauf, Z. Xu, R. U. R. Sagar, F. Faiz, Z. Tayyab, R. U. Rehman, M. Imran, A. Waheed, R. Javed, A. Surulinathan, Z. Zafar, X.-Z. Fu and J.-L. Luo, Advanced fabrication techniques for polymer–metal nanocomposite films: state-of-the-art innovations in energy and electronic applications, Chem. Sci., 2025, 16, 3362–3407.
14. M. Horák, K. Bukvišová, V. Švarc, J. Jaskowiec, V. Křápek and T. Šikola, Comparative study of plasmonic antennas fabricated by electron beam and focused ion beam lithography, Sci. Rep., 2018, 8, 9640.
15. P. Yang, J. Zheng, Y. Xu, Q. Zhang and L. Jiang, Colloidal synthesis and applications of plasmonic metal nanoparticles, Adv. Mater., 2016, 28, 10508–10517.
16. M. L. Personick, in One Hundred Years of Colloid Symposia: Looking Back and Looking Forward, ed. R. Nagarajan, American Chemical Society, Washington, 2023, vol. 1457, ch. 11, pp. 247–283, ACS Symposium Series.
17. H. You, S. Yang, B. Ding and H. Yang, Synthesis of colloidal metal and metal alloy nanoparticles for electrochemical energy applications, Chem. Soc. Rev., 2013, 42, 2880–2904.
18. Q. Yang, W. Liu, B. Wang, W. Zhang, X. Zeng, C. Zhang, Y. Qin, X. Sun, T. Wu, J. Liu, F. Huo and J. Lu, Regulating the spatial distribution of metal nanoparticles within metal-organic frameworks to enhance catalytic efficiency, Nat. Commun., 2017, 8, 14429.
19. V. Pérez-Luna, S. Oros-Ruiz, E. Pérez and M. Quintana, Single-walled carbon nanotubes as one-dimensional template for CdS nanoparticles growth, Phys. Status Solidi B, 2017, 254, 1700197.
20. F. Aubrit, F. Testard, A. Paquirissamy, F. Gobeaux, X. Wang, F. Nallet, P. Fontaine, V. Ponsinet and P. Guenouna, Ligand-free synthesis of gold nanoparticles incorporated within cylindrical block copolymer films, J. Mater. Chem. C, 2018, 6, 8194–8204.
21. K. Kashihara, Y. Uto and T. Nakajima, Rapid in situ synthesis of polymer-metal nanocomposite films in several seconds using a CO₂ laser, Sci. Rep., 2018, 8, 14719.
22. D. Yu, X. Sun, X. Chen, W. Ma, J. Sun, C. Zhou, Z. Xie and S. Zhoub, In situ hydrosilane reduction and preparation of gold nanoparticle-gel glass composites with nonlinear optical properties, J. Mater. Chem. C, 2018, 6, 5624–5629.
23. S. Kumar-Krishnan, E. Prokhorov, M. Hernández-Iturriaga, J. D. Mota-Morales, M. Vázquez-Lepe, Y. Kovalenko, I. C. Sánchez and G. Luna-Bárcenas, Chitosan/silver nanocomposites: synergistic antibacterial action of silver nanoparticles and silver ions, Eur. Polym. J., 2015, 67, 242–251.
24. M. Alsawafta, S. Badilescu, A. Paneri, V.-V. Truong and M. Packirisamy, Gold-poly(methyl methacrylate) nanocomposite films for plasmonic biosensing applications, Polymers, 2011, 3, 1833–1848.
25. A. Akhavan, N. Sheikh and R. Beteshobabrud, Polymethylmethacrylate/silver nanocomposite prepared by γ-ray, J. Nucl. Sci. Technol., 2010, 50, 80–84.
26. W. Di Cianni, M. de la Mata, F. J. Delgado, J. Hernández-Saz, M. Herrera, S. I. Molina, M. Giocondo and A. Sanz de León, Polymer nanocomposites for plasmonics: in situ synthesis of gold nanoparticles after additive manufacturing, Polym. Test., 2023, 117, 107869.
27. X. Ni, Z. Wu, X. Gu, D. Wang, C. Yang, P. Sun and Y. Li, In situ growth of clean Pd nanoparticles on polystyrene microspheres assisted by functional reduced graphene oxide and their excellent catalytic properties, Langmuir, 2017, 33, 8157–8164.
28. N. K. Ray, H. Patel, R. Muñoz-Espí, K. Landfester and V. Gundabala, Synthesis of silver-coated polystyrene latex through in situ metal reduction, MRS Commun., 2022, 12, 952–958.
29. F. W. Zeng, D. Zhang and J. B. Spicer, Palladium nanoparticle formation processes in fluoropolymers by thermal decomposition of organometallic precursors, Phys. Chem. Chem. Phys., 2018, 20, 24389–24398.
30. Y. Liu, A. Naseri, T. Li, A. Ostovan, E. Asadian, R. Jia, L. Shi, L. Huang and A. Z. Moshfegh, Shape controlled photochemical synthesis of noble-metal nanocrystals based on reduced graphene oxide, ACS Appl. Mater. Interfaces, 2022, 14, 16527–16537.
31. Y. Yagci, M. Sangermano and G. Rizza, A visible light photochemical route to silver-epoxy nanocomposites by simultaneous polymerization-reduction approach, Polymer, 2008, 49, 5195–5198.
32. R. N. Schmarsow, I. E. dell'Erba, M. S. Villaola, C. E. Hoppe, I. A. Zucchi and W. F. Schroeder, Effect of light intensity on the aggregation behavior of primary particles during in situ photochemical synthesis of gold/polymer nanocomposites, Langmuir, 2020, 36, 13759–13768.
33. Y.-J. Chen, W.-H. Chang, C.-Y. Li, Y.-C. Chiu, C.-C. Huang and C. H. Lin, Direct synthesis of monolayer gold nanoparticles on epoxy based photoresist by photoreduction and application to surface-enhanced raman sensing, Mater. Des., 2021, 197, 109211.
34. S. V. Asmussen and C. I. Vallo, Synthesis of silver nanoparticles in surfactant-free light-cured methacrylate resins, Colloids Surf., A, 2015, 466, 115–124.
35. M. N. Siddiqui, H. H. Redhwi, E. Vakalopoulou, I. Tsagkalias, M. D. Ioannidou and D. S. Achilias, Synthesis, characterization and reaction kinetics of PMMA/silver nanocomposites prepared via in situ radical polymerization, Eur. Polym. J., 2015, 72, 256–269.
36. I. E. dell'Erba, F. D. Martínez, C. E. Hoppe, G. E. Elicabe, M. Ceolín, I. A. Zucchi and W. F. Schroeder, Mechanism of particle formation in silver/epoxy nanocomposites obtained through a visible-light-assisted in situ synthesis, Langmuir, 2017, 33, 10248–10258.
37. J. Mijović and S. Andjelić, A study of reaction kinetics by near-infrared spectroscopy. 1. Comprehensive analysis of a model epoxy/amine system, Macromolecules, 1995, 28, 2787–2796.
38. N. Poisson, G. Lachenal and H. Sautereau, Near- and mid-infrared spectroscopy studies of an epoxy reactive system, Vib. Spectrosc., 1996, 12, 237–247.
39. M. L. Auad, S. R. Nutt, P. M. Stefani and M. I. Aranguren, Rheological study of the curing kinetics of epoxy-phenol novolac resin, J. Appl. Polym. Sci., 2006, 102, 4430–4439.
40. A. F. Craievich, in Handbook of Sol-Gel Science and Technology, ed. L. Klein, M. Aparicio and A. Jitianu, Springer, Cham, 2nd edn, 2016, ch. 39, pp. 1185–1230.
41. I. Breßler, J. Kohlbrecher and A. F. Thünemann, SASfit: A tool for small-angle scattering data analysis using a library of analytical expressions, J. Appl. Crystallogr., 2015, 48, 1587–1598.
42. G. Odian, Principles of Polymerization, Wiley-Interscience, Hoboken, 4th edn, 2004.
43. T. Vidil, M. Cloître and F. Tournilhac, Control of gelation and network properties of cationically copolymerized mono- and diglycidyl ethers, Macromolecules, 2018, 51, 5121–5137.
44. This work was benefited from the use of the SasView application, originally developed under NSF award DMR-0520547.
45. F. Wang, V. N. Richards, S. P. Shields and W. E. Buhro, Kinetics and mechanisms of aggregative nanocrystal growth, Chem. Mater., 2014, 26, 5–21.
46. M. Zhang, Y. Xu, X. Peng, H. Chen and H. Wang, Controllable synthesis of gold nanoparticle dimers via site-selective growth, Chem. Commun., 2022, 58, 7932–7935.
47. N. Zohar, L. Chuntonov and G. Haran, The simplest plasmonic molecules: metal nanoparticle dimers and trimmers, J. Photochem. Photobiol., C, 2014, 21, 26–39.
48. L. Tian, C. Wang, H. Zhao, F. Sun, H. Dong, K. Feng, P. Wang, G. He and G. Li, Rational approach to plasmonic dimers with controlled gap distance, symmetry, and capability of precisely hosting guest molecules in hotspot regions, J. Am. Chem. Soc., 2021, 143, 8631–8638.

---