Design of high-performance probes with enhanced sensitivity is crucial in detecting biomarkers for the early diagnosis of degenerative diseases [1-3]. Based on the unique structural and functional advantages of nanomaterials [4-6], several meticulously designed probes have been proven effective in various sensing scenarios [7-10]. However, many of these probes necessitate sophisticated fabrication techniques and have high manufacturing costs. Therefore, there is still a significant demand for universal strategies that can easily and systematically upgrade conventional probes to meet the requirements of cutting-edge, high-demanding detections.

Previously, we successfully developed a unique fluorescence platform by incorporating gold nanoclusters (AuNCs) into a nanogel of chitosan derivative, glycol chitosan (GC) [11]. This GC assembly served a dual purpose: it not only confined the Au clusters to enhance the fluorescent intensity of the probe but also utilized the positively charged amino groups of GC to attract sulfide ions through electrostatic forces, thereby synergistically improving the sensitivity of the detection. However, the preliminary idea of Enrichment-enhanced Detection Strategy was yet to be verified independently.

Dopamine (DA) is a crucial neurotransmitter and chemical messenger in central and peripheral nervous systems, which plays key roles in human metabolism, immune, memory and cognition [12-15]. The DA fluctuation is closely connected to severe physical and neurological diseases including Alzheimer, Parkinson, schizophrenia [16-18]. The diagnosis of these diseases requires high-precision measurement of DA. Various techniques have been developed for the sensitive detection of DA. Among them, fluorescent sensing is considered as one of the most promising detection strategies owing to its high sensitivity, good selectivity and real-time merits [19-22]. Carbon dots (CDs), as a new type of carbon-based material, have gained significant attention in recent years due to their unique optical properties, low cytotoxicity, and easy fabrication process [23,24]. Their potential applications range from bioimaging to biosensing [25]. For example, Soto et al. reported a N-doped carbon dots for highly sensitive and selective detection of DA in human fluids at nanomolar concentrations [26]. In another study, Liang et al. developed a carbon dots-gold nanoclusters hybrid for the ratiometric fluorescence detection of DA [27]. Li et al. fabricated a series of graphene quantum dots to explore the influence of element doping on fluorescence sensing of DA [28]. However, in the case of DA-sensing, the complicated analytical environment, extremely low DA level still recalls for more smart and effective strategies to systemically upgrade the probes for DA sensing.

In this study, we used the CDs encapsulated in DA (CDs@GC) detection system as a representative model to rule-out the enhancement of fluorescent intensity and validate the feasibility of the Enrichment-enhanced Detection strategy (EED strategy) independently. As shown in Scheme 1, the CDs were prepared by one-step hydrothermal method. By mixing the CDs with GC, the self-assembled CDs@GC probe was easily obtained through noncovalent interactions such as electrostatic interactions and hydrogen bonding. Studies have shown that DA could form DA quinone by ambient O₂ in alkaline condition [29,30], and quenched the fluorescence of nanoprobes. The positively charged amino groups on the surface of GC could enable noncovalent interaction with DA-DA quinone through electrostatic interactions. Despite having nearly identical fluorescent profiles, the assembled CDs@GC probe exhibited a 12-fold increase in sensitivity compared to the bare CDs probes, thus confirming the feasibility of the EED strategy. We believe that this EED strategy presents a promising approach to optimizing conventional probes in high-demand detection scenarios.

In a typical procedure, CDs were synthesized via the hydrothermal technique using citric acid and thiourea as precursors (Fig. 1A). The obtained brownish solution exhibited good aqueous solubility and showed strong fluorescence under UV irradiation. The CDs@GC were easily obtained via noncovalent interactions including electrostatic interaction and hydrogen bond. The purified CDs and CDs@GC were characterized using various techniques to determine their structural and morphological composition. Fig. 1B shows the transmission electron microscopy (TEM) images of the CDs and exhibited a uniform dispersion without apparent aggregation. The high-resolution TEM images of CDs showed a lattice spacing of 0.15 nm, which is close to the (100) facet of graphite carbon (Fig. 1C). Dynamic light scattering (DLS) analysis were also conducted and the average particle size of CDs were about 6 nm (Fig. 1D). After assembled with GC, a well-organized self-assembled structure were obtained (Figs. 1E and F) and the average size of CDs@GC was increased to about 21 nm (Fig. 1G). Moreover, Fourier transform infrared (FTIR) and X-ray photoelectron spectroscopy (XPS) experiments were conducted to investigate the functional groups of the CDs@GC. In the FTIR analysis, the stretching vibrations of O-H, N-H, C-H, and C=O were clearly observed (Fig. S1 in Supporting information). Further, Fig. S2A (Supporting information) shows that the CDs@GC comprised carbon, nitrogen, oxygen, and sulfur. In Fig. S2B (Supporting information), the spectrum of N 1s demonstrates the presence of pyridinic N at 399.5 eV, pyrollic N at 400.0 eV, and graphitic N at 401.5 eV. The high-resolution spectrum of C 1s (Fig. S2C in Supporting information) reveals peaks at 284.4 eV, 285.0 eV and 287.4 eV, which are corresponded to the C=C/C–C, C-N and N-C=O bonds. The O 1s XPS spectrum in Fig. S2D (Supporting information) shows two peaks at 531.1 eV and 532.6 eV, which are attributed to C–OH and C=O. Excitation-dependent PL behavior of CDs and CDs@GC were investigated, which is common in fluorescent carbon materials due to the complex surface state affecting the band gap of CDs (Fig. S3A and S3B). Both of CDs and CDs@GC displayed satisfied quantum yield (QY) of 21.7% and 23.5% (using quinine sulfate as a reference, Fig. S4 and Table S1 in Supporting information).

Furthermore, the PL stability of CDs and CDs@GC to the effects of pH and ionic strength were investigated. According to Figs. S3C and D (Supporting information), the PL intensity remained stable when the pH adjusted from 6 to 12. Meanwhile, there was negligible influence on the fluorescence intensity of CDs or CDs@GC even the salinity was up to 3 mol/L (Figs. S3E and F in Supporting information). In addition, the stability of the as-prepared CDs and CDs@GC was evaluated (Fig. S5 in Supporting information). The fluorescence of CDs and CDs@GC remained almost unchanged after 4 h of continuous irradiation under UV light, which reveals the ideal photostability of the probes. Besides, only about 5% fluorescence decreased at the end of 8-week period when stored at 4 ℃, indicating the satisfactory long-term stability of both CDs and CDs@GC. These indicated that CDs and CDs@GC exhibited almost identical fluorescent profiles including excitation-dependent PL behavior and PL stability in despite of the difference in morphology.

Before performing the sensitivity assay, detection parameters (pH, temperature and incubation time) were optimized for the system. As shown in Fig. S6A (Supporting information), the value of ΔI/I₀ increased within the pH range from 6.0 to 11.0. It has been reported that DA could produce polydopamine in strong alkaline solutions, which is not conducive to sensitive detection of dopamine. Therefore, moderate alkali condition (pH 9.0) was selected for the sensing system. On the other hand, temperature affects the reaction rate, which is of great significance to the detection process. As shown in Fig. S6B (Supporting information), 50 ℃ is the optical reaction temperature. The incubation time of the reaction system was also investigated to determine the optimal response time of the fluorescence sensor. According to Fig. S6, the value of ΔI/I₀ increased dramatically within 1 h; while after 1 h, slight increases could be observed. To achieve a satisfied response, 1 h of pre-incubation time was used for further experiments.

Under optical conditions, PL spectra of CDs@GC as well as bare CDs with 0.5–100 µmol/L DA was studied and showed in Fig. 2. The PL intensity of CDs and CDs@GC at 460 nm gradually decreased with the increase of DA concentrations (Figs. 2A2 and B2), and was quenched along with the increasing concentration of DA. For CDs, the linear equation is calculated I = 320.03 − 2.19c with a correlation coefficient of 0.9984 (Fig. 2A3). While for CDs@GC, the linear regression equation is I₁ = 338.28 − 16.48c₁ for low concentration (≤10 µmol/L) and I₂ = 191.70 − 1.61c₂ for higher concentration (≥10 µmol/L), with a correlation coefficient of 0.9806 and 0.9935, respectively (Fig. 2B3). The limit of detection (LOD = 3σ/K) for DA was 44.6 nmol/L for CDs and 3.7 nmol/L for CDs@GC. Impressively, though with similar fluorescent property, the CDs@GC achieved 12-fold improvement in the sensitivity than the bare CDs. Thus, it was the change of interaction between probe and DA rather than the change of fluorescent profile that contribute to the enhancement of DA sensing.

The dependence of the intensity of luminescence with increasing concentrations of DA followed the Stern-Volmer formalism: I₀/I = 1 + K_SV[DA], where I₀ is the luminescence signal of CDs in the absence of DA, and I is the luminescence signal with an increasing concentration of DA. K_SV is the Stern-Volmer constant, which indicates the quenching efficiency between the nanoprobes and DA. The calculated K_SV were found to be 0.00909 and 0.08632 for CDs and CDs@GC, respectively, which confirmed the assembled GC indeed promoted sensitive detection of DA.

The selectivity of the probes toward DA were also studied. Under identical conditions, only DA displayed a significant luminescence quenching, while other species (Na⁺, K⁺, Ca²⁺, Mg²⁺, GSH, Ala, Gly, Phe, Leu, glucose, tyramine, β-phenethylamine and catechol) produced negligible fluorescence changes even at the concentration of 10-fold higher than that of DA (Fig. S7 in Supporting information). Comparing with some other fluorescent nanoprobes, CDs@GC also has excellent sensitivity (Table S2 in Supporting information).

Dopamine could be easily oxidized to quinone by ambient O₂ in alkaline conditions (Fig. 3A) [29,30]. Herein, to investigate the sensing mechanism, the UV–vis absorption spectra of DA and DA quinone have been measured. As shown in Fig. 3B, the emission spectrum of CDs@GC (yellow region) shows obvious spectral overlap with the absorption band of generated DA quinone, which directly cause the fluorescence quenching of the probe. The zeta potential of the sensing system was also detected. As shown in Fig. 3C, the CDs are negatively charged and the zeta potential is about -16.43 mV. Whereas the CDs@GC are positively charged with a zeta potential of 2.56 mV, suggesting the nanocomplex has been successfully formed. When adding 50 µmol/L DA to the solution, the zeta potential of CDs@GC decreased to -12.21 mV. The anion-cation pair of DA⁻ and CDs@GC-NH₂⁺ could be formed via electrostatic interaction, which makes DA possess a special affinity to the CDs@GC. Meanwhile, studies have demonstrated that quinones could dynamically react with chitosan's amines via Michael addition or Schiff base reaction [31-33] and form a DA “reservoir” to further raise the regional DA concentration. Thus, the DA could be electrostatically attracted and dynamically bonded to the CDs@GC probe, which could synergistically contribute to the good sensitivity and selectivity.

Complexes formation induced fluorescence quenching is general categorized to static quenching. Herein, the time-resolved fluorescence spectra of the CDs@GC with/without DA were measured to verify the quenching mechanism. The fluorescence decay curves are shown in Fig. 3D. In the absence and presence of DA, the fluorescence lifetimes of the CDs@GC were almost the same, implying that the quenching mechanism obeys static quenching rather than dynamic quenching. The overlap between absorbance spectrum of the acceptor (DA quinone species) and excitation spectrum of the donor (CDs@GC), and the constant lifetime value demonstrated the occurrence of inner filter effect (IFE) process in the system. Furthermore, three DA homologues (tyramine, β-phenethylamine and catechol) were chosen to explore the fluorescence quenching behavior of CDs@GC. As shown in Fig. 3E, both tyramine and β-phenethylamine hardly show any fluorescence quenching phenomenon, while the catechol and DA both exhibit fluorescence quenching to some extent. Therefore, we deduce that the catechol group is the crucial structure for the quenching of CDs@GC, which is consistent with the above results.

In recent years, the portability and practicality of smartphones have been favored in various detection fields [34-36]. The related applications on smartphones can transform color signals into digital dataset for analysis to achieve the quantitative detection of analytes. To verify the possibility of the sensing platform, firstly we tested the sensing system on filter paper. As shown in Fig. 4A, with the concentration of DA increases, the fluorescence intensity of the filter paper significantly decreased. Considering that single color change (dark blue to light blue) could not be easily recognized, we added a common fluorescent dye, riboflavin into the system. With the increase of DA concentration, the fluorescence of the system showed a change from blue to yellow, demonstrating the feasibility for the construction of the sensing platform (Fig. 4B). Next, the portable sensing platform was created by the combination of a smartphone, a mini laser, a sample curve and a dark chamber. The RGB values corresponding to the photos are obtained through a color recognizer APP (Color Picker) for further statistical analysis (Fig. 4C). After the reaction was completed, a series of photos with colors ranging from blue to yellow green were taken with a smartphone under a 365 nm ultraviolet mini-laser. According to Fig. 4D, there is a good linear relationship between the RGB ratio (B/G) and the DA concentration. The low concentration linear regression equation is y₁ = 1.8047 - 0.0764x₁ and high concentration linear regression equation is y₂ = 1.1086 - 0.0088x₂, with a correlation coefficient of 0.9914 and 0.9978, respectively. In addition, the corresponding CIE chromaticity coordinates shift from the blue area to the yellow area, thus realizing the visual determination of DA (Fig. 4E). The above results indicate that the probe has a more sensitive response in the low concentration range. More importantly, compared with the standard curve measured in solution state, the ratio of the slopes of the two segments is almost equal, indicating the reliability of digital signal conversion for dopamine detection (Table S3 in Supporting information).

To test the practicability of the CDs@GC nanosensor, the concentration of DA in human sweat samples and urine samples was detected using standard addition method, and the results are shown in Tables S4 and S5 (Supporting information). Under optimized conditions, the mean recoveries of the spiked samples were in the range of 90%–115% with the relatively low relative standard deviation (RSD) under 1.0% at spiked concentrations of 10, 15, 20, 25, 30 µmol/L in human sweat or urine samples. These results indicate that the CDs@GC nanosensor based platform is effective and reliable for the detection of DA in complicated samples and could further confirm the feasibility of this EED theory.

In summary, we developed an innovative “Enrichment-enhanced Detection” strategy for the enhanced detection of dopamine using a CDs@GC nanosensor. The GC of the nanocomplex can effectively increase the local concentration of the analyte to significantly improve the sensitivity and selectivity of the sensing system. The LOD of the nanosensor was determined as 3.7 nmol/L, demonstrating 12-fold enhancement in the sensitivity compared to bare CDs probe with similar fluorescent profile. Subsequently, we applied this strategy to the design of a high-performance portable platform for the visual detection of dopamine (DA) to further explore the feasibility of EED strategy in real case. The success of this CDs@GC-DA detecting system exhibited the potential of this novel strategy as a universal optimizing pathway for the development of high-performance probes.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Xilin Bai: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Validation, Visualization, Writing – original draft, Writing – review & editing. Wei Deng: Investigation. Jingjuan Wang: Writing – original draft. Ming Zhou: Supervision, Writing – original draft, Writing – review & editing.

The author gratefully acknowledges the financial support from the National Natural Science Foundation of China (No. 21904007), the Fundamental Research Funds for the Central Universities (China, No. 2412022QD008), the Jilin Provincial Department of Education (China), the Key Laboratory of Nanobiosensing and Nanobioanalysis at Universities of Jilin Province (China), and the Analysis and Testing Center of Northeast Normal University (China).

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.109959.