A Molecularly Imprinted CdSe@SiO₂ Core–Shell Nanohybrid: A Ratiometric Fluorescent Probe for Ultra‑Sensitive 4‑Nitrophenol Detection

Background

Nitrophenols are widely used in herbicides, pesticides, dyes, and pharmaceuticals, making them among the most common environmental contaminants [1]. 4‑Nitrophenol (4‑NP) is particularly hazardous, classified by the EPA as a priority pollutant with a permissible drinking‑water limit of 60 ng mL⁻¹ [3]. Accordingly, reliable detection methods are essential. Traditional analytical techniques—chromatography, electrochemical, chemiluminescent, and fluorescence assays—offer varying trade‑offs between sensitivity, cost, and operational complexity [4–11]. Fluorescence detection is attractive for its simplicity and speed; however, single‑emission probes suffer from intensity fluctuations caused by excitation variations, probe concentration changes, and interference from quenchers such as heavy metals or reactive oxygen species [15–18]. Ratiometric fluorescence, which measures the intensity ratio of two emission bands, mitigates these issues through self‑calibration, yielding more reliable quantification [19]. While several ratiometric probes have shown improved sensitivity for pollutants like Hg²⁺ and H₂S [20–22], few combine ratiometric readouts with molecular imprinting to achieve both high sensitivity and selectivity for trace analytes [25,26].

To address this gap, we constructed a dual‑emission sensor that merges a ratiometric FRET system with a MIP tailored to 4‑NP. CdSe quantum dots embedded in a silica shell (CdSe@SiO₂) serve as a reference fluorophore, while organosilane‑functionalized CDs act as the FRET donor. The silica shell protects the toxic Cd and Se cores and provides an inert, transparent matrix for FRET. The MIP layer, formed via sol–gel polymerization in the presence of 4‑NP, creates selective recognition sites that enhance binding and quenching efficiency. This architecture delivers both high sensitivity (due to FRET) and high selectivity (due to MIP).

Methods

Materials

Tetraethoxysilane (TEOS), Triton X‑100, and petroleum ether were purchased from Tianjin Kemiou Chemical Reagent Co., Ltd. (Tianjin, China). Cyclohexane, 4‑NP, hexyl alcohol, ammonium hydroxide (25 wt %), absolute ethyl alcohol, methylbenzene, and isopropyl alcohol were sourced from Guangfu Chemical Reagent Co., Ltd. (Tianjin, China). 3‑Aminopropyltrimethoxysilane (APTMS) and anhydrous citric acid were obtained from Aladdin Chemical Reagent Co., Ltd. (Shanghai, China). Carboxyl‑modified CdSe/ZnS quantum dots were supplied by Wuhan Jiayuan Quantum Dot Technological Development Co., Ltd. (Wuhan, China). All reagents were analytical grade and used as received. Deionised water (18.2 MΩ cm) was prepared with a Sartorius Arium® Pro VF system.

Synthesis of CdSe@SiO₂

A reverse microemulsion was formed by mixing 7.7 mL cyclohexane, 1.77 mL Triton X‑100, 1.8 mL n‑hexanol, and 400 µL of the CdSe QD solution (8 µM). After the microemulsion stabilized, 50 µL TEOS and 200 µL 25 wt % ammonium hydroxide were added. The system was stirred at 25 °C for 24 h, then broken with 36 mL isopropyl alcohol. The resulting precipitate was repeatedly washed with ethanol by centrifugation until the supernatant showed no fluorescence. The purified particles were finally dispersed in toluene under ultrasonication.

Synthesis of Organosilane‑Functionalized CDs

APTMS (10 mL) was heated to 185 °C, followed by rapid addition of 0.5 g anhydrous citric acid under vigorous stirring. After 1 min at 185 °C, the mixture cooled to 25 °C. The dark‑green product was purified by extracting with petroleum ether (×5, 1:1 volume ratio). The lower phase yielded approximately 2 mL of organosilane‑functionalized CDs.

Coating of CdSe@SiO₂ with CDs and MIP Layer

10 µL of organosilane‑functionalized CDs were added to 25 mL toluene containing 5 mg CdSe@SiO₂. The mixture was heated at 113 °C under reflux for 12 h with stirring, then centrifuged and resuspended in 2 mL ethanol. 0.2 mg 4‑NP was introduced and reacted for 2 h at 25 °C. TEOS (25 µL) and ammonium hydroxide (25 µL) were then added, and the system was stirred for another 5 h at 25 °C. After three precipitation/centrifugation cycles and ethanol washing, the product—CdSe@SiO₂/CDs/MIP—was dispersed in ethanol for use. Non‑imprinted controls (NIP) were prepared identically but without 4‑NP.

Adsorption Capacity Determination

For each sample, 0.1 mg 4‑NP was added to 1 mL (1.5 mg mL⁻¹) of the nanohybrids and stirred for 120 min. After centrifugation (12 000 rpm, 15 min), the supernatant was analysed by UV–vis at 400 nm. The adsorption capacity (Q) was calculated via Q=(C_0-C_t)V/W, where C_0 and C_t are the initial and final concentrations, V is the solution volume, and W is the mass of nanohybrids.

Characterisation

High‑resolution TEM (JEOL JEM‑2100, 200 kV) was used to visualise morphology. FTIR spectra were recorded on a Nicolet Magna IR‑560 (20 scans, 4 cm⁻¹ resolution). Fluorescence measurements were performed on a Cary Eclipse spectrophotometer (1 cm cells). UV–vis spectra were acquired with a TU‑1810 spectrophotometer (1 cm cells).

Fluorescent Detection of 4‑NP

1 mL of the CdSe@SiO₂/CDs/MIP in ethanol (1.5 mg mL⁻¹) was mixed with 2 mL ethanol and the desired amount of 4‑NP. After 10 min of incubation at room temperature (pH ≈ 7.0), fluorescence spectra were recorded at 350 nm excitation (10 nm slits). The 10‑min incubation was chosen based on precedent for ratiometric probes [33,34].

Results and Discussion

Initial spectral analysis confirmed that CdSe QDs (emission 460 nm) and CDs (emission 615 nm) do not overlap, enabling a clear ratiometric readout. Importantly, the emission band of CDs overlaps the absorption of 4‑NP, allowing FRET‑mediated quenching at 455 nm while the CdSe reference remains unchanged.

The synthetic route (Scheme 1) involved (i) silica coating of CdSe QDs to form CdSe@SiO₂, (ii) attachment of organosilane‑functionalized CDs, and (iii) encapsulation with a TEOS‑based MIP layer. This architecture preserves the photoluminescence of the QDs, secures the CDs within the matrix, and generates 4‑NP‑specific cavities.

Schematic representation of the process employed for preparation of the molecularly imprinted polymer‑coated dual‑emission CdSe@SiO₂/CD nanohybrids

Morphology and Optical Properties

HRTEM images (Fig. 1) show uniform spherical particles with average diameters of 46.7 ± 2.5 nm (CdSe@SiO₂), 53.6 ± 2.7 nm (CdSe@SiO₂/CD), and 66.4 ± 2.0 nm (CdSe@SiO₂/CD/MIP). The increase in size reflects sequential layering of CDs and the MIP shell. The MIP thickness (~12.8 nm) is optimal for FRET proximity (<10 nm) [43,44]. FTIR spectra (Fig. 1g) confirm the presence of Si–O–Si, C=O, C–H, and NH₂ vibrations, and the characteristic NO₂ peaks after imprinting, verifying successful MIP formation.

Characterisation of the prepared hybrid nanoparticles. (a,b) CdSe@SiO₂; (c,d) CdSe@SiO₂/CD; (e,f) CdSe@SiO₂/CD/MIP. (g) FTIR spectra of CdSe@SiO₂, CdSe@SiO₂/CD, CdSe@SiO₂/CD/MIP, and 4‑NP.

Fluorescence Stability and Sensitivity

Fluorescence spectra before and after template removal (Fig. 2a) show that the CD emission at 615 nm remains intact, while the 455 nm band recovers upon 4‑NP removal, indicating reversible quenching. The probe maintains >95 % of its 455 nm intensity over 120 min, confirming excellent photostability (Fig. 2b). The adsorption capacity of the MIP nanohybrid (9.1 mg g⁻¹) is markedly higher than that of the NIP (1.58 mg g⁻¹), reflecting the selective cavity design.

Fluorescence of the prepared hybrid nanoparticles. (a) Emission before/after 4‑NP removal. (b) Photostability over time. (c) Adsorption capacities of MIP vs NIP. (d) Fluorescence quenching curves for MIP and NIP.

Selectivity Assessment

To evaluate specificity, the sensor was challenged with phenol, 2‑NP, and hydroquinone. The quenching efficiencies for these analogs were substantially lower than for 4‑NP (Fig. 3), attributable to mismatched absorption overlap and cavity mismatch. 2‑NP, despite structural similarity, exhibited only one‑third the quenching, confirming the MIP’s selective recognition.

The selectivity of the prepared ratiometric fluorescent probe. Fluorescence responses to 4‑NP, phenol, 2‑NP, and hydroquinone at 4.8 µg mL⁻¹.

Quantitative Detection of 4‑NP

Incremental addition of 4‑NP (0.051–13.7 µg mL⁻¹) produced a pronounced decrease in the 455 nm band while the 615 nm reference remained constant (Fig. 4a). Plotting the intensity ratio F_b/F_r against 4‑NP concentration yields a linear relationship (R² = 0.985) over the working range, with a calculated detection limit of 0.026 µg mL⁻¹ (3σ/k). This LOD is well below the EPA guideline of 60 ng mL⁻¹, demonstrating the sensor’s suitability for real‑world monitoring. Compared with literature methods (Table 1), the proposed nanohybrid matches or surpasses current electrochemical and fluorescence techniques in both sensitivity and linearity.

The detection of 4‑NP. (a) Fluorescence spectra at increasing 4‑NP concentrations (350 nm excitation). (b) Effect of 4‑NP on the F_r/F_b ratio for the MIP‑coated nanohybrids. Concentrations: 0.051, 2.1, 4.8, 6.7, 8.9, 11.52, 13.7 µg mL⁻¹.

Conclusions

We have engineered a molecularly imprinted CdSe@SiO₂ core–shell nanohybrid that functions as a ratiometric fluorescent probe for ultra‑trace detection of 4‑NP. The dual‑emission system leverages FRET‑mediated quenching of CDs by 4‑NP while maintaining a stable CdSe reference signal. The MIP layer affords high binding affinity and selectivity, yielding a linear response from 0.051 to 13.7 µg mL⁻¹ and a detection limit of 0.026 µg mL⁻¹—well below regulatory limits. This work demonstrates that combining molecular imprinting with ratiometric fluorescence offers a powerful strategy for sensitive, selective, and reliable monitoring of hazardous environmental pollutants.