Skip to main content
NCBI home page
RSC Advances logoLink to RSC Advances
. 2026 Aug 3;16(39):44470–44478. doi: 10.1039/d6ra03667h

Metal-enhanced fluorescent sensor system for bisphenol a determination using AgNP-modified molecularly imprinted polymer chips

Daria Yarynka a,✉, Roman Nikolaiev a, Oleksandr Brovko b, Tetyana Sergeyeva a
PMCID: PMC13430516  PMID: 42549099

Abstract

A metal-enhanced fluorescence sensor system was developed for the detection of bisphenol A (BPA) using molecularly imprinted polymer (MIP) films with in situ synthesized silver nanoparticles (AgNPs). The sensor design combines the high selectivity of ethylene glycol methacrylate phosphate (EGMP)-containing MIP films and the fluorescence amplification capabilities of AgNPs via metal-enhanced fluorescence (MEF) phenomenon. Analytical characteristics were evaluated, including detection limit, linear dynamic range and selectivity. The proposed MEF sensor demonstrated a detection limit of 0.1 nM for BPA and a linear dynamic range of 0.1 nM–25 µM. The thin AgNP-MIP films immobilized on glass slides exhibited significantly higher selectivity toward BPA compared to its structurally related analogues (2-, 3-, 4-nitrophenols, phenol, o- and p-cresols, resorcinol, catechol). The MEF sensor system was successfully applied to detect BPA in spiked wastewater samples, demonstrating potential for robust and sensitive environmental monitoring.

A novel metal-enhanced fluorescence (MEF) sensor system for BPA detection was developed, achieving a low LOD of 0.1 nM. Nanostructured EGMP-based AgNP-MIP chips were successfully used as highly efficient recognition elements.graphic file with name d6ra03667h-ga.webp

1. Introduction

Bisphenol A (BPA) is a common industrial chemical compound that has been widely used in the production of polycarbonate plastics and epoxy resins since the mid-20th century.1 As a consequence of its massive global production and chemical stability, BPA has become a widespread environmental contaminant by the beginning of the 21st century. When bisphenol A enters the human body, it can mimic the activity of estrogen.2 This characteristic classifies BPA as an endocrine-disrupting chemical (EDC),3 triggering significant concern about its serious impact on human health and aquatic ecosystems. Bisphenol A has been associated with various health problems, including reproductive disorders, developmental abnormalities, and neurobehavioral issues.4,5 Notably, even low-dose exposure to BPA induces significant physiological responses, which complicates risk assessment and the implementation of stringent regulatory measures.6

Moreover, the regulation of BPA remains crucial for public health. This chemical compound has been detected in food packaging,7 water sources,8 and consumer products,9 underscoring the need for reliable, rapid, and highly sensitive analytical approaches for its monitoring.

Traditional detection techniques, such as high-performance liquid chromatography (HPLC),10 mass spectrometry (MS),11 gas chromatography (GC),12 and capillary electrophoresis (CE),13 offer high accuracy and specificity. However, they typically require expensive instrumentation, extensive sample preparation, and highly qualified personnel, which limits their applicability for on-site or routine analysis. Additionally, conventional methods for BPA analysis include immunochromatographic assays and electrochemical sensors.14–16 Although these approaches have achieved excellent analytical performance, their practical deployment is often constrained by operational complexity, limited portability, or susceptibility to matrix effects.

Among the various emerging technologies, electrochemical sensors based on iron metal–organic frameworks (Fe-MOFs) and molecularly imprinted polymers (MIPs) have demonstrated remarkable sensitivity, with detection limits as low as 0.1 µM and 0.04 µM, respectively.17,18 Furthermore, molecularly imprinted electrochemical sensors (MIES) offer a cost-effective and portable approach for BPA detection, with detection limits comparable to conventional techniques.19 Other innovative approaches, including sensors using recombinant proteins, peptides, and nanomaterial-based platforms, have achieved remarkable detection performance. For instance, a sensor using a designer probe (LacI-BPA) on reduced graphene oxide electrodes achieved a wide linear range of 100 fM to 10 nM.20 Similarly, a peptide-modified gold electrode showed excellent performance with a broad dynamic detection range of 1–5000 nM.21 The incorporation of nanomaterials, metal–organic frameworks, and ionic liquids has further improved sensor performance.22 Nevertheless, these systems often involve multi-step fabrication, limited long-term stability, or insufficient robustness for real-sample analysis, highlighting the need for simpler yet reliable sensing architectures.

To overcome these limitations, research has increasingly focused on developing optical biosensors, especially fluorescence-based platforms.23 These systems offer simplicity, high sensitivity, and potential for miniaturization and on-site detection. Over the past few years, various optical strategies for BPA detection have been explored, including quantum dots,24 upconversion nanoparticles,25 and metal–organic framework26 with detection limits in the nanomolar range.

However, conventional fluorescence sensing of BPA often suffers from background autofluorescence of environmental matrices and low quantum yields of the BPA, necessitating efficient signal amplification strategies. Metal-enhanced fluorescence (MEF) has emerged as a promising solution to overcome these constraints by utilizing noble metal nanostructures to manipulate and amplify emission intensity.27 It is known that silver nanoparticles (AgNPs), in particular, can exhibit strong localized surface plasmon resonance (LSPR) effects that significantly boost fluorescence emission when fluorophores are in close proximity.29 Crucially, recent studies in metal-manipulated fluorescence (MMF) demonstrate that signal amplification does not strictly require a far-field spectral overlap between the nanoparticle's primary dipole LSPR band and the fluorophore's transitions.30 Moreover, the overall enhancement is heavily driven by a complex of near-field phenomena, including intense localized electric fields (“hot-spots”) and modifications of radiative decay rates on the nanometer scale.30,31 Recent studies at the single-nanoparticle level have confirmed that the number of local hot spots and the strict control over the metal-fluorophore spacing are the dominant factors governing the enhancement, rather than far-field resonance matching.30 These suggest alternative pathways for signal amplification depending strictly on the spatial distribution of the components, thereby expanding the applicability of MEF-based sensing systems. Such a mechanism is highly relevant for nanostructured composite architectures, such as molecularly imprinted polymers, where the precise positioning of the target molecule within the nanoparticle's enhanced near-field zone can prevent non-radiative quenching and determine the final fluorescence enhancement.30,31

When combined with synthetic recognition elements such as molecularly imprinted polymers (MIPs), MEF-based sensors offer a powerful strategy for selective and sensitive detection.32,33 MIPs are robust, cost-effective alternatives to biological receptors, capable of forming specific recognition sites complementary in shape and functionality to target analytes.34,35 Their chemical stability and reusability make MIP-based sensors particularly suitable for environmental monitoring and field applications where biological receptors would be unstable.36 Molecularly imprinted plasmonic sensors utilized for environmental pollutants include detection of pesticides (atrazine, cyanazine, and simazine),37 pharmaceuticals,38 and other contaminants in water samples.36

Despite these advances, the integration of MIPs with plasmonic nanostructures for BPA detection has been limited. The inherent compatibility of MIPs with nanomaterials makes them ideal candidates for integration into advanced sensor platforms. In our previous research, we demonstrated the successful application of silver nanoparticles for the development of highly sensitive metal-enhanced fluorescence sensors based on molecularly imprinted polymers (MIPs) with embedded silver nanoparticles for the detection of foodborne mycotoxins (aflatoxin B1 and zearalenone).32,33 However, directly translating this strategy to BPA introduces an additional challenges, as BPA requires a specialized polymer matrix capable of managing both specific biomimetic binding and mediating the fragile metal-fluorophore interactions near the UV-visible boundary without inducing fluorescence quenching. To close this research gap, this work introduces a novel chip-based sensing platform that explicitly utilizes the in situ fabrication of AgNPs within a pre-formed ethylene glycol methacrylate phosphate (EGMP)-based MIP tailored specifically for BPA. The in situ reduction mechanism ensures that the synthesized AgNPs are spatially confined within the MIP structure, enabling plasmonic coupling while minimizing quenching and attenuation effects in complex wastewater matrices. Herein, we systematically investigate the influence of silver nitrate precursor concentration on the AgNPs formation and the resulting fluorescence amplification. The developed MIP-MEF chip platform demonstrates high selectivity against structurally similar phenolic compounds, establishes a superior stability profile, and achieves a nanomolar detection limit in heavily-contaminated wastewater samples, offering a robust and easily miniaturizable platform for the rapid monitoring of endocrine-disrupting compounds.

2. Results and discussion

To develop a highly sensitive MEF sensor system, BPA-selective MIPs were synthesized using ethylene glycol methacrylate phosphate (EGMP) as a functional monomer at a monomer-to-template ratio of 1 : 1, previously optimized in our earlier work.39 Building on this, the first stage of the current study investigated the influence of silver nitrate concentration on the metal-enhanced fluorescence phenomenon. Specifically, EGMP-based MIP films were synthesized by adding varying amounts of silver nitrate to monomer mixtures. As a result, the fluorescent sensor response toward BPA was significantly affected by the formation of AgNP within the polymer matrix (Fig. 1 and S1).

Fig. 1. Fluorescence responses of AgNP-containing MIP thin-film sensors synthesized with varying silver nitrate concentrations, in response to the addition of 25 µM BPA. The measurements were carried out in 10 mM sodium phosphate buffer (pH 6.0) containing 10 mM NaCl and 10% acetonitrile. *The values represent the difference in signal between the MIP and NIP chips. The signal labelled “0” denotes the differential response of chips prepared without silver nanoparticles.

Fig. 1

Open in a new tab

The fluorescence intensity toward BPA exhibited a pronounced maximum at 1 mM silver nitrate, with a signal more than 3 times higher than that of the AgNO3-free MIP films (Fig. S2). At both lower and higher AgNO3 concentrations, the fluorescence response decreased substantially, indicating that optimal fluorescence enhancement is achieved only within a narrow range of silver nitrate concentrations. At low AgNO3 levels, the number of formed AgNPs was likely too low to induce significant enhancement effects. In contrast, high silver nitrate (88 mM) concentrations may have promoted particle aggregation,40 thereby reducing enhancement efficiency. Crucially, the concentration of 1 mM AgNO3 allowed achievement of the optimal distance between the fluorophore and the metal surface required to maximize the local electromagnetic field enhancement while successfully preventing fluorescence quenching. Based on these results, 1 mM silver nitrate was selected as an optimal concentration for further sensor chips fabrication.

To confirm the successful formation and optical properties of silver nanoparticles (AgNPs) within the polymer matrix, transmission electron microscopy (TEM) and UV-vis absorption spectroscopy were employed. The results of this dual characterization are presented in Fig. 2.

Fig. 2. Structural and optical characterization of the AgNPs formed in the structure of MIP thin films synthesized with EGMP as a functional monomer: (a) TEM microphotograph showing AgNPs formed in situ within the MIP thin film; (b) UV-vis absorption spectra of the glass chip (1), bare MIP film immobilized on the glass chip (2), and AgNP-MIP film immobilized on the glass chip (3).

Fig. 2

Open in a new tab

The TEM imaging (Fig. 2a and S3) showed that the AgNPs synthesized in situ during the MIP polymerization were primarily spherical, with diameters ranging from 10 to 30 nm. Statistical analysis of the TEM micrographs (Fig. S4) indicated a relatively narrow particle size distribution, with the synthesized silver nanoparticles exhibiting an average diameter of 15.9 ± 0.5 nm. The silver nanoparticles were uniformly distributed throughout the polymer structure, suggesting that the chosen concentration of AgNO3 (1 mM) promoted controlled nucleation and growth while effectively minimizing aggregation.40

These structural findings are further supported by the UV-vis absorption analysis (Fig. 2b). To isolate the optical contribution of the silver nanoparticles, spectra were recorded for three types of surfaces: bare glass substrates, MIP films without added silver nitrate, and the final AgNP-MIP composite film. While the glass substrate and the bare MIP films exhibited no significant absorption in the visible range, the AgNP-MIP chips displayed a distinct and well-defined surface plasmon resonance (SPR) peak centered at 420 nm. The presence of this characteristic absorption maximum confirms the successful reduction of silver ions into metallic AgNPs and their stable incorporation into the MIP matrix. Furthermore, the symmetry of the SPR peak at 420 nm correlates with the spherical morphology and narrow size distribution observed in the TEM images. Such morphology is consistent with conditions that provide optimal fluorescent enhancement,32,33 further validating the choice of 1 mM of silver nitrate for MEF sensor fabrication. The use of optimized silver nitrate concentration resulted in a uniform distribution and stable size of formed AgNPs within the polymer matrix, which in turn enhanced the sensitivity of the sensor system.

It is worth noting that typically, metal-enhanced fluorescence requires a significant spectral overlap between the nanoparticle LSPR band and the fluorophore.28 In our system, the silver nanoparticles peak at 420 nm, while bisphenol A excites at 278 nm and emits at 310 nm, showing no direct overlap (Fig. S2). Despite this, the AgNP-MIP platform provides a strong signal amplification. The AgNP-MIP chip demonstrated a 3.2-fold increase in fluorescence intensity at 310 nm compared to the MIP film without AgNP. Crucially, the spectral profile remained unchanged and no new bands appeared. This confirms that the enhancement comes strictly from the target molecules, ruling out any background luminescence or optical artifacts.

This effective enhancement can be explained by two main factors. First, although the core SPR maximum of our AgNPs is at 420 nm, its shorter-wavelength tail extends into the UV region, overlapping with the excitation band of BPA. This allows the 15 nm nanoparticles to act as local optical antennas that concentrate the excitation light and enhance the excitation efficiency.31 Second, recent studies show that fluorescence enhancement depends more on local “hot spots” and the metal-fluorophore distance than on far-field resonance matching, with the maximum enhancement typically occurring at a critical distance of approximately 5 nm.30,31 Here, the nanostructured polymer matrix plays a key role. The MIP cavities act as a precise spatial framework that holds the captured BPA molecules at an optimized distance from the silver surface. This close arrangement allows the system to maximize the local near-field enhancement while preventing non-radiative quenching, leading to a highly sensitive analytical response.31 Additional fluorescence lifetime and spectral-overlap analyses further supported this AgNP-mediated fluorescence enhancement mechanism (Section S5, Fig. S5). The AgNP-containing MIP system showed altered fluorescence decay, an increase in apparent energy-transfer efficiency from ∼0.70 to 0.93, a 1.4-fold rise in the normalised spectral overlap integral, and estimated donor – AgNP coupling distances of approximately 4.3–4.8 and 3.2–3.6 nm, consistent with efficient plasmon-assisted fluorescence without dominant non-radiative quenching.

Following the morphological characterization of the AgNPs synthesized within the MIP films' structure, the analytical performance of the developed sensor system toward BPA was evaluated. The calibration curve was constructed by plotting the fluorescence sensor response as a function of the decimal logarithm of BPA concentration over the range of 0.1 nM–25 µM (Fig. 3). The significant difference between AgNP-MIP and the corresponding AgNP-NIP confirms the successful formation of BPA-specific binding sites during the imprinting process. The resulting calibration plot exhibited excellent linearity (R2 > 0.9663 for AgNP-MIP) within this interval, confirming the quantitative capability of the proposed MEF sensor system.

Fig. 3. Calibration curves of AgNP-containing MIP-film-based MEF sensor systems for BPA detection. Fluorescence responses of MIP and NIP thin polymer films, immobilized on glass slides and synthesized using EGMP as the functional monomer with the addition of 1 mM silver nitrate. Measurements were carried out in 10 mM sodium phosphate buffer (pH 6.0) containing 10 mM NaCl and 10% acetonitrile.

Fig. 3

Open in a new tab

The limit of detection (LOD) and the limit of quantification (LOQ) were determined using the standard IUPAC and ICH statistical criteria based on the rigorous formulas LOD = 3.3 × σ/S and LOQ = 10 × σ/S, respectively.41,42 Here, σ represents the standard deviation of the blank analytical signal calculated from ten independent replicate measurements (n = 10) of the blank sensor chips, and S is the slope of the linear calibration plot. Under these conditions, the LOD was determined to be 0.1 nM, and the LOQ was found to be 1 nM. The linear dynamic range of the MEF sensor was from 0.1 nM to 25 µM. In our previous study,39 a colorimetric MIP-based sensor for BPA detection exhibited a linear range from 5 to 250 µM, with a LOD of 5 µM. By contrast, the present metal-enhanced fluorescence sensor platform reduces the LOD to 0.1 nM – an improvement of approximately 50 000-fold.

This significant enhancement effect results from the synergistic molecular imprinting, enabling selective recognition of BPA and metal-enhanced fluorescence from embedded AgNPs, which amplifies the fluorescence signal of the target analyte. To critically evaluate the analytical performance of the developed AgNP-modified MIP chips and position them within the field of optical sensors, a comprehensive comparison with other recently reported optical platforms for bisphenol A was performed, as summarized in Table 1. Moreover, the achieved value of LOD is notably lower than those reported for many previously developed optical sensors for BPA detection, where typical LODs range from 32 nM to 1 µM (e.g.43–46).

Table 1. Comparison of the developed sensor with other reported fluorescence-based sensors for bisphenol A detection.

Sensor platform/materials Detection strategy Linear range LOD Tested matrices Reference
BPA-aptamer system Fluorescence 0–1300 µM 32 nM Tap water 43
Au-Fe alloy nanoparticles/horseradish peroxidase Colorimetry 1.10–70.1 µM 0.58 µM Polycarbonate plastic samples 44
Laccase enzyme/ABTS/polyvinyl alcohol hydrogel beads Colorimetry 1.0–10.0 µM 0.5 µM Model phenolic solutions 45
Magnetic particles and quantum dots/immunochromatographic test strip Fluorescence/lateral flow immunoassay 0.3–100 µg mL−1 0.3 ng mL−1 Drinking and natural water 24
UCNPs/tetramethylrhodamine Fluorescence (FRET) 0.1–25 ng mL−1 0.05 ng mL−1 Tap water, river water, disposable paper cup water 25
Zinc-based porphyrin metal–organic framework Fluorescence quenching 0.050–20 nM 0.902 nM Polycarbonate food packaging films 26
MIP chips Colorimetry 5–25 µM 5 µM Wastewater 39
AgNP-MIP chips Fluorescence (MEF) 0.1 nM–25 µM 0.1 nM Wastewater This work
Open in a new tab

The selectivity of the developed sensor system was further assessed by comparing its fluorescence response to BPA with that obtained for several structurally related compounds, including similar phenolic analogues (Fig. 4). Each analyte was tested at the same concentration under identical experimental conditions. The results clearly demonstrated that the fluorescence response toward BPA was significantly higher than for any of the analogues tested, confirming the minimal cross-reactivity of the MEF sensor with other phenolic compounds and indicating high molecular recognition capability of the proposed AgNP-MIP chips.

Fig. 4. Selectivity profile of AgNP-containing MIP-film-based MEF sensor system. Differential responses represent the signal difference between AgNP-MIP and corresponding NIP films after the addition of each analyte at a concentration of 10 µM. Measurements in 10 mM sodium phosphate buffer (pH 6.0) with 10 mM NaCl and 10% acetonitrile.

Fig. 4

Open in a new tab

Beyond their analytical performance, the proposed sensor chips offer exceptional mechanical robustness and ease of handling, ensuring reliable operation in various environments.

As shown in Fig. 5, the chips demonstrated remarkable long-term stability, retaining over 95% their initial response even after 20 months of storage under ambient condition.

Fig. 5. Long-term storage stability profile of the AgNP-containing MIP-film-based MEF sensor system. The fluorescence responses were recorded for a series of sensor chips upon exposure to 1 µM BPA over a 20 month period.

Fig. 5

Open in a new tab

To evaluate the method's resistance to non-phenolic matrix interferences, the sensor was applied to the analysis of real wastewater samples, which inherently contain complex backgrounds of surfactants, inorganic ions, and organic matter. Finally, the MEF AgNP-containing MIP-film sensor chips were applied to analyze BPA in spiked wastewater samples. The reproducibility of the proposed sensor was evaluated by repeated measurements across different chips and days, showing stable and consistent performance (Table 2).

Table 2. The results of BPA analysis in spiked wastewater samples (n = 8).

Sample Added, µM Found, µM Recovery (R, %) % RSD
Filtrate of the city dump (lake Zelene, v. Kotsyubyns'ke, Kyiv region) 3 2.83 ± 0.26 94.3 9.1
Filtrate of the city dump (lake Blakytne, Horenka, Kyiv region) 1 1.04 ± 0.07 104.0 6.6
Filtrate of the city dump (river Syrets, Kyiv) 0.5 0.51 ± 0.09 102.4 16.7
River Lybid, Kyiv 0.1 0.09 ± 0.02 93.2 16.4
Lake Vyrlytsia, Kyiv 0.025 0.027 ± 0.003 108.9 12.2
River Dnipro, Kyiv 0.001 0.0009 ± 0.0002 92.6 18.5
Open in a new tab

As shown in Table 2, the applicability of the developed AgNP-containing MIP-film-based MEF sensor system for real sample analysis was evaluated using wastewater samples spiked with BPA at concentrations of 3, 1, 0.5, 0.1, 0.025, and 0.001 µM. Under the optimized measurement conditions, the calculated recovery rates were 94.3%, 104%, 102.4%, 93.2%, 108.9%, and 92.6%, respectively, with corresponding relative standard deviations (RSDs) of 9.1%, 6.6%, 16.7%, 16.4%, 12.2%, and 18.5%.

The recoveries obtained were within or close to the generally acceptable range of 80–120%,42 indicating the sensor's suitability for trace-level BPA quantification in complex matrices. Notably, the tested concentrations, including the lowest level of 0.001 µM (approx. 0.23 µg L−1), are significantly lower than current regulatory standards, such as the EU Drinking Water Directive (2020/2184) limit of 2.5 µg L−1.47 Slightly elevated recovery and higher RSD (above 10%) at the lowest tested concentrations (0.025 and 0.001 µM) reflect signal variability near the detection limit, which arises from a combination of instrumental noise near the detection boundary and matrix interferences from real wastewater samples. Wastewater represents an exceptionally complex matrix containing a vast array of non-phenolic interferences, including suspended solids, surfactants, inorganic salts, and diverse dissolved organic matter.48 To mitigate these effects in routine field analyses, future protocols could incorporate optimized sample pre-treatment steps, such as fine membrane filtration or solid-phase extraction, to minimize particulate interference and stabilize the background optical signal. Nevertheless, the acceptable recovery rates obtained in these experiments confirm that the developed AgNP-MIP chips retain their efficiency even when exposed to a heavily contaminated non-phenolic matrix, such as the tested wastewater. Importantly, while most state-of-the-art optical sensors summarized in Table 1 were validated using relatively clean media, such as tap, drinking, or purified water, the developed AgNP-MIP chip platform demonstrates improved tolerance to matrix effects, maintaining reliable analytical performance and low detection limit even when challenged with environmental wastewater samples.

As demonstrated by the comparative data in Table 1, while modern optical platforms relying on aptamers, enzymatic systems (such as horseradish peroxidase), or biological antibodies offer high sensitivity, their practical deployment in environmental monitoring is frequently limited by the fragile nature of these bioreceptors. Furthermore, strategies utilizing colloidal substrates, quantum dots, or upconversion nanoparticles (UCNPs) often face challenges regarding matrix interferences and poor structural stability in complex wastewater samples. The proposed AgNP-MIP chip sensor platform successfully addresses these common limitations. The synergistic integration of silver nanoparticles within the nanostructured MIP ensures robust signal amplification and high selectivity, offering a clear methodological advance for practical wastewater analysis.

3. Experimental

3.1. Preparation of molecularly imprinted polymer films and synthesis of AgNP-embedded MIP sensor chips

MIP films selective for BPA were synthesized via in situ photopolymerization as described.32,33 The pre-polymerization mixture consisted of EGMP as the functional monomer, BPA as the template molecule, TGDMA/OUA as the cross-linker, PEG 20.000 as the porogen, DMF as solvent, and 2,2′-dimethoxy-2-phenylacetophenone as the photoinitiator (Table 3). Glass slides were pretreated with γ-methacryloxypropyltrimethoxysilane for covalent immobilization of polymer films. The mixture was polymerized under UV light (λ = 365 nm, 3.4 W m−2) for 30 minutes between two glass slides held together with clips.

Table 3. Compositions of the MIP and NIP films' monomer mixtures for the preparation of sensor chips.

Monomer/oligomer MIP NIP
BPA, mg 10 —
EGMP, mg 9.2 9.2
TGDMA/OUA, mg 190.8 190.8
DMF, µL 100 100
PEG, mg 30 30
2,2′-Dimethoxy-2-phenylacetophenone, mg 0.1 0.1
Open in a new tab

To incorporate silver nanoparticles, various concentrations of AgNO3 (0.1–88 mM) were added to the pre-polymerization mixture. The formation of AgNPs occurred during polymerization relies on a synchronized, one-step photo-induced process, enabling uniform dispersion of AgNPs within the MIP matrix. Under continuous ultraviolet (UV) irradiation, the UV light acts simultaneously as a polymerization initiator and a clean reducing agent. The formation mechanism of silver nanoparticles embedded within the polymeric matrix follows our previously reported protocol.32,33

After polymerization, MIP films were subjected to Soxhlet extraction in ethanol for 8 hours, followed by heated distilled water (80 °C) for another 8 hours to remove the template, PEG 20.000 and unreacted compounds.32 Non-imprinted polymer (NIP) films were prepared using the same composition but excluding BPA (see Table 3). After polymerization and extraction, the AgNP-containing MIP films were dried at room temperature and used as sensor chips.

The thickness of the MIP films immobilized on glass surfaces was determined to be 60 µm, as measured with an Adoric electronic digital caliper (0–300 µm range; Ningbo, China).32

3.2. Characterization of AgNPs

Transmission electron microscopy (TEM) was employed to characterize the morphology, size and distribution of silver nanoparticles embedded within the structure of the MIP films. For imaging, MIP samples were prepared on copper grids using the same photopolymerization protocol and procedure as described.32,33 The monomer mixture (see Table 3), containing 1 mM silver nitrate, was used for MIP sample preparation. The resulting MIP film with in situ synthesized AgNPs on the grid was subjected to TEM analysis. Microstructural imaging was performed using a JEM-1230 transmission electron microscope (JEOL, Japan) operated at an accelerating voltage of 50–120 kV.

The particle size distribution and statistical analysis of the TEM images were conducted using ImageJ software.49

The optical properties and surface plasmon resonance (SPR) of the synthesized nanocomposites were characterized using UV-vis absorption spectroscopy. Absorption spectra were recorded in the wavelength range of 320–600 nm using a Specord 210 plus spectrophotometer. All spectra were recorded at room temperature. A clean glass slide was used as a reference to account for background transitions.

3.3. Fluorescence measurement procedure

Sensor chips were incubated in 10 mL of BPA-containing sample (10 mM sodium phosphate, pH 6.0, with 10% acetonitrile and 10 mM sodium chloride).39 Fluorescence was measured using a spectrofluorimeter Jasco FP-8200 (Japan), with excitation at 278 nm and emission detected with a maximum of BPA fluorescence at 310 nm. Values of the fluorescent responses recorded directly from the film surfaces were plotted as a function of BPA concentration in the samples. The differential sensor response was defined as the difference in fluorescence intensity between the MIP and the corresponding NIP films.

3.4. Analysis of wastewater samples

Collected wastewater samples were spiked with varying BPA concentrations (0.001, 0.025, 0.1, 0.5, 1, and 3 µM) for analysis with the proposed MEF sensor system. For sample preparation, 1 mL of each environmental water sample was mixed with 9 mL of a working BPA solution at the corresponding concentration level. The prepared spiked samples were analyzed using the MEF sensor system, following the fluorescence measurement procedure described previously (see 3.3 Section). Recovery and related standard deviation rates were calculated to assess the applicability of the sensor in real matrices.

4. Conclusions

In this study, a metal-enhanced fluorescence (MEF) sensor system based on molecularly imprinted polymer (MIP) films with in situ embedded silver nanoparticles was developed for the selective and sensitive detection of bisphenol A (BPA). The effect of silver nitrate concentrations added to monomeric mixtures, as well as the morphology of synthesized AgNPs, on the fluorescence enhancement phenomenon was investigated. Optimization of silver nitrate concentration in EGMP-based MIP films enabled maximal fluorescence enhancement with an improvement of approximately 50 000-fold, yielding a competitive limit of detection (0.1 nM) and a wide linear dynamic range (0.1 nM–25 µM). The sensor demonstrated high selectivity toward BPA over structurally related phenolic compounds (2-, 3-, 4-nitrophenols, phenol, o- and p-cresols, resorcinol, catechol). Application of the proposed MEF sensor system to spiked wastewater samples at concentrations ranging from 3 µM to 0.001 µM resulted in recovery rates of 92.6–108.9% with relative standard deviations of 6.6–18.5%, confirming the method's capability for trace-level BPA detection in real environmental matrices. Compared to previously reported MIP-based colorimetric sensor, the proposed MEF-based sensing platform offers a substantial improvement in detection limit (LOD 5 µM vs. 0.1 nM) while preserving robust selectivity and reproducibility. Beyond its sensitivity, the system offers a key practical advantage: the immobilization of MIP films onto rigid glass surfaces provides superior mechanical stability and ease of handling compared to conventional free-standing MIPs. Combined with a remarkable shelf-life stability of 20 months under ambient conditions, these results highlight the potential of combining plasmonic nanostructures with synthetic recognition elements for practical, portable, and cost-effective environmental monitoring of endocrine-disrupting compounds such as BPA.

Author contributions

D. Yarynka: formal analysis and investigation, data curation, visualization, writing – original draft preparation. R. Nikolaiev: investigation, writing – editing. O. Brovko: conceptualization, investigation. T. Sergeyeva: conceptualization, methodology, writing – reviewing and editing, supervision.

Conflicts of interest

The authors declare that they have no competing interests.

Supplementary Material

RA-016-D6RA03667H-s001

Acknowledgments

This work was financially supported by the National Academy of Sciences of Ukraine. Daria Yarynka, Roman Nikolaiev and Tetyana Sergeyeva received financial support from Simmons Foundation (IMBG Simons Foundation Grant for Ukrainian institutions No. SFI-PD-Ukraine-00017453). The authors gratefully acknowledge the Armed Forces of Ukraine and all defenders of Ukraine, whose resilience and sacrifice made this work possible during the full-scale war.

Data availability

Data will be available from the corresponding author Daria Yarynka upon reasonable request.

Supplementary information (SI) is available. See DOI: https://doi.org/10.1039/d6ra03667h.

References

  1. Xing J. Zhang S. Zhang M. Hou J. Comp. Biochem. Physiol., Part C: Toxicol. Pharmacol. 2022;254:109275. doi: 10.1016/j.cbpc.2022.109275. [DOI] [PubMed] [Google Scholar]
  2. Kania M. Adamowska A. Śniatała A. Bartkowiak H. Grubski D. Ziarnik K. Nadolny F. Jabłoński J. J. Educ. Health Sport. 2024;75:56271. doi: 10.12775/JEHS.2024.75.56271. [DOI] [Google Scholar]
  3. Kawa I. A. Fatima Q. Mir S. A. Jeelani H. Manzoor S. Rashid F. Diabetes Metab. Syndr. 2021;15:803–811. doi: 10.1016/j.dsx.2021.03.031. [DOI] [PubMed] [Google Scholar]
  4. Talpade J. Shrman K. Sharma R. Gutham V. Singh R. P. Meena N. S. Entomol J. Zool. Stud. 2018;6:394–397. [Google Scholar]
  5. Milczarek-Banach J. Miśkiewicz P. J. Med. Sci. 2020;89:e441. doi: 10.20883/medical.e441. [DOI] [Google Scholar]
  6. Kassotis C. D. Vandenberg L. N. Demeneix B. A. Porta M. Slama R. Trasande L. Lancet Diabetes Endocrinol. 2020;8:719–730. doi: 10.1016/S2213-8587(20)30128-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Sadrabad E. K. Hashemi S. A. Nadjarzadeh A. Askari E. Mohajeri F. A. Ramroudi F. Food Sci. Nutr. 2023;11:3718–3728. doi: 10.1002/fsn3.3398. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. dos Santos Costa M. J. Araújo J. V. S. Moura J. K. L. da Silva Moreno L. H. Pereira P. A. da Silva Santos R. Moura C. V. R. Water, Air, Soil Pollut. 2022;233:362. doi: 10.1007/s11270-022-05834-z. [DOI] [Google Scholar]
  9. Kubwabo C. Kosarac I. Stewart B. Gauthier B. R. Lalonde K. Lalonde P. J. Food Addit. Contam. 2009;26:928–937. doi: 10.1080/02652030802706725. [DOI] [PubMed] [Google Scholar]
  10. Pop C. E. Miu B. A. Németh D. Wolff R. Mihăilescu D. F. Avramescu S. M. Mernea M. Discover Appl. Sci. 2024;6:171. doi: 10.1007/s42452-023-05617-z. [DOI] [Google Scholar]
  11. Zhao Y. Boukherroub R. Liu L. Li H. Zhao R. S. Wei Q. Yu X. Chen X. J. Hazard. Mater. 2023;459:132336. doi: 10.1016/j.jhazmat.2023.132336. [DOI] [PubMed] [Google Scholar]
  12. Martín-Pozo L., Martín-Bueno J., Moscoso-Ruiz I. and Zafra-Gómez A., Emerging Contaminants in the Environment, 2022, pp. 465–493, 10.1016/B978-0-323-85160-2.00008-1 [DOI] [Google Scholar]
  13. Yahaya N. Huang Z. A. Yan B. Chen D. D. Food Chem. 2022;372:131220. doi: 10.1016/j.foodchem.2021.131220. [DOI] [PubMed] [Google Scholar]
  14. Raysyan A. Schneider R. J. Biosensors. 2021;11:231. doi: 10.3390/bios11070231. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Tajik S. Beitollahi H. Nejad F. G. Zhang K. Le Q. V. Jang H. W. Kim S. Y. Shokouhimehr M. Sensors. 2020;20:3364. doi: 10.3390/s20123364. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Zhang Y. Lei Y. Lu H. Shi S. Wang P. Ali Z. Li J. Food Chem. 2021;346:128895. doi: 10.1016/j.foodchem.2020.128895. [DOI] [PubMed] [Google Scholar]
  17. Maridevaru M. C. Dube A. Kaimal R. Al Souwaileh A. Kannadasan S. Anandan S. Analyst. 2024;149:3325–3334. doi: 10.1039/D4AN00499J. [DOI] [PubMed] [Google Scholar]
  18. Tan F. Cong L. Li X. Zhao Q. Zhao H. Quan X. Chen J. Sens. Actuators, B. 2016;233:599–606. doi: 10.1016/j.snb.2016.04.146. [DOI] [Google Scholar]
  19. Pan Y. Wu M. Mi S. Shi P. Zhao N. Zhu Y. Karimi-Maleh H. Ye C. Lin C.-T. Fu L. Sensors. 2023;23:8656. doi: 10.3390/s23208656. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Kim K. S. Jang J. R. Choe W. S. Yoo P. J. Biosens. Bioelectron. 2015;71:214–221. doi: 10.1016/j.bios.2015.04.042. [DOI] [PubMed] [Google Scholar]
  21. Yang Y. Yu J. Yin J. Shao B. Zhang J. J. Agric. Food Chem. 2014;62:11130–11137. doi: 10.1021/jf5037933. [DOI] [PubMed] [Google Scholar]
  22. Dodevska T. Acta Chim. Slov. 2025;72:217–237. doi: 10.17344/acsi.2024.8918. [DOI] [PubMed] [Google Scholar]
  23. Nath P. Mahtaba K. R. Ray A. Sensors. 2023;23:5053. doi: 10.3390/s23115053. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Taranova N. A. Bulanaya A. A. Zherdev A. V. Dzantiev B. B. Sensors. 2025;25:7328. doi: 10.3390/s25237328. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Li Q. Bai J. Ren S. Wang J. Gao Y. Li S. Zhang X. Zhang Q. Gao Z. Anal. Bioanal. Chem. 2019;411:171–179. doi: 10.1007/s00216-018-1425-8. [DOI] [PubMed] [Google Scholar]
  26. Pang Y. Cao Y. Han J. Xia Y. He Z. Sun L. Liang J. Food Control. 2022;132:108551. doi: 10.1016/j.foodcont.2021.108551. [DOI] [Google Scholar]
  27. Singh K. Pombeiro A. J. Garazade I. M. Sun Q. Mohan B. Microchem. J. 2024;197:109888. doi: 10.1016/j.microc.2023.109888. [DOI] [Google Scholar]
  28. Nandhini G. U. Minakshi M. Sivasubramanian R. Dharmalingam G. Nanomaterials. 2026;16:298. doi: 10.3390/nano16050298. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Raj A. and Thomas R. K., Optical and Molecular Physics, 2021, pp. 43–69 [Google Scholar]
  30. Blake-Hedges J. M. Greenspan S. H. Kean J. A. McCarron M. A. Mendonca M. L. Wustholz K. L. Chem. Phys. Lett. 2015;635:328–333. doi: 10.1016/j.cplett.2015.06.083. [DOI] [Google Scholar]
  31. Li J. F. Li C. Y. Aroca R. F. Chem. Soc. Rev. 2017;46:3962–3979. doi: 10.1039/C7CS00169J. [DOI] [PubMed] [Google Scholar]
  32. Sergeyeva T. Yarynka D. Lytvyn V. Demydov P. Lopatynskyi A. Stepanenko Y. Brovko O. Pinchuk A. Chegel V. Analyst. 2022;147:1135–1143. doi: 10.1039/D1AN02173G. [DOI] [PubMed] [Google Scholar]
  33. Yarynka D. Chegel V. Piletska E. Piletsky S. Dubey L. Dubey I. Nikolaiev R. Brovko O. Sergeyeva T. Analyst. 2023;148:2633–2643. doi: 10.1039/D2AN01991D. [DOI] [PubMed] [Google Scholar]
  34. Saylan Y. Kılıç S. Denizli A. Processes. 2024;12:177. doi: 10.3390/pr12010177. [DOI] [Google Scholar]
  35. Lazarević-Pašti T. Tasić T. Milanković V. Potkonjak N. I. Chemosensors. 2023;11:35. doi: 10.3390/chemosensors11010035. [DOI] [Google Scholar]
  36. Rebelo P. A. C. Seguro I. Nouws H. P. A. Delerue-Matos C. Pacheco J. G. Chemosensors. 2023;11:318. doi: 10.3390/chemosensors11060318. [DOI] [Google Scholar]
  37. Saylan Y. Akgönüllü S. Çimen D. Derazshamshir A. Bereli N. Yılmaz F. Denizli A. Sens. Actuators, B. 2017;241:446–454. doi: 10.1016/j.snb.2016.10.017. [DOI] [Google Scholar]
  38. Ayankojo A. G. Reut J. Öpik A. Furchner A. Syritski V. Biosens. Bioelectron. 2018;118:102–107. doi: 10.1016/j.bios.2018.07.042. [DOI] [PubMed] [Google Scholar]
  39. Yarynka D. Honcharenko A. Gorbach L. Piletska E. Piletsky S. Brovko O. Sergeyeva T. Anal. Bioanal. Chem. 2024;416:7121–7129. doi: 10.1007/s00216-024-05616-y. [DOI] [PubMed] [Google Scholar]
  40. Sobczak-Kupiec A. Malina D. Wzorek Z. Zimowska M. Micro Nano Lett. 2011;6:656–660. doi: 10.1049/mnl.2011.0152. [DOI] [Google Scholar]
  41. Analytical Methods Committee, Analyst, 1987, 112, 199–204.. doi: 10.1039/AN9871200199 [DOI] [Google Scholar]
  42. ICH Harmonised Tripartite Guideline, Validation of analytical procedures: text and methodology, Q2(R1), 2005
  43. Wei Y. Zhou Y. Wei Y. Dong D. Wang L. Anal. Methods. 2021;13:1816–1822. doi: 10.1039/D1AY00180A. [DOI] [PubMed] [Google Scholar]
  44. Liang X. Wang H. Wang H. Pei G. Anal. Methods. 2015;7:3952–3957. doi: 10.1039/C5AY00090D. [DOI] [Google Scholar]
  45. Zofair S. F. Khan M. A. Salman K. A. Younus H. Catal. Lett. 2024;155(32) doi: 10.1007/s10562-024-04877-7. [DOI] [Google Scholar]
  46. Du W. Liu J. Li H. Deng C. Feng Q. Tan Y. Yang S. Wu Z. Xiao F. Anal. Chem. 2023;95:4220–4226. doi: 10.1021/acs.analchem.2c05662. [DOI] [PubMed] [Google Scholar]
  47. Directive (EU) 2020/2184 of the European Parliament and of the Council of 16 December 2020 on the quality of water intended for human consumption (recast). https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A32020L2184
  48. Dsikowitzky L. and Schwarzbauer J., in Pollutant Diseases, Remediation and Recycling, ed. E. Lichtfouse, J. Schwarzbauer and D. Robert, Springer, Cham, 2013, vol. 4, pp. 65–123. doi: 10.1007/978-3-319-02387-8_2 [DOI] [Google Scholar]
  49. Zhang S. Wang C. Methods Protoc. 2023;6(4):63. doi: 10.3390/mps6040063. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

RA-016-D6RA03667H-s001
RA-016-D6RA03667H-s001.pdf (643.2KB, pdf)

Data Availability Statement

Data will be available from the corresponding author Daria Yarynka upon reasonable request.

Supplementary information (SI) is available. See DOI: https://doi.org/10.1039/d6ra03667h.

Articles from RSC Advances are provided here courtesy of Royal Society of Chemistry

RESOURCES