A Sensitive Epinephrine Sensor Based on Photochemically Synthesized Gold Nanoparticles

Abstract

In this study, gold nanoparticles (AuNPs) and AuNPs-graphene oxide (AuNPs@GO) nanostructures were synthesized in aqueous media using an in-situ photochemical method with bis-acyl phosphine oxide (BAPO) photoinitiator as a photoreducing agent in the presence of HAuCl₄. The parameters for synthesis were arranged to obtain stable and reproducible dispersions with desirable chemical and optical properties. Both AuNPs and AuNPs@GO were employed as sensing platforms for the detection of epinephrine in two concentration ranges: micromolar (µM) and nanomolar (nM). Field emission scanning electron microscopy (FE-SEM), Dynamic Light Scattering (DLS), UV-Vis absorption, fluorescence emission, and Fourier Transform Infrared (FT-IR) spectroscopy techniques were used to investigate the morphological, optical, and chemical properties of the nanostructures as well as their sensing ability towards epinephrine. Fluorescence spectroscopy played a crucial role in demonstrating the high sensitivity and effectiveness of these systems, especially in the low concentration (nM) range, confirming their strong potential as fluorescence-based sensors. By constructing calibration curves on best linear subranges, limit of detection (LOD) and limit of quantification (LOQ) were calculated with two different approaches, SE_intercept and S_y/x. Among all the investigated nanostructures, AuNPs@GO exhibited the highest sensitivity towards epinephrine. The efficiency and reproducibility of the in-situ photochemical AuNPs synthesis approach highlight its applicability in small-molecule detection and particularly in analytical and bio-sensing applications.

1. Introduction

Nanotechnology is a multidisciplinary field that enables scientists to develop significant novel materials with superior performance at the nanoscale for various applications. Among the wide range of nanomaterials, metallic nanoparticles, especially gold nanoparticles (AuNPs), have attracted significant attention due to their unique chemical, optical, and electrical properties. Because of their strong surface plasmon resonance (SPR), high surface area-to-volume ratio, and biocompatibility, AuNPs are versatile tools for applications in various fields, such as catalysis, drug delivery, biomedicine, coating technology, polymer science, and sensing applications [1,2,3,4,5].

However, the method used for the synthesis of AuNPs plays a significant role in the adjustment of the desired properties and functionalities of the nanoparticles. Conventional methods such as chemical synthesis often require harsh conditions, toxic reagents, or multiple steps. Subsequent physical methods, on the other hand, require more energy and lead to increased steps during the synthesis [1,2]. In contrast, in-situ photochemical synthesis offers an environmentally friendly and controllable alternative [6,7,8]. In this method, the reduction of Au ions to AuNPs is achieved via reduction by a photoinitiator. The initiation capabilities and formed free radicals have a strong effect on the synthesis of AuNPs. Hence, a suitable photoinitiator and optimization parameters for the synthesis of nanoparticles have great importance [9,10,11].

Incorporating nanoparticles with graphene oxide (GO) leads to further enhancements of nanoparticle performance in sensor applications [12]. GO has excellent mechanical properties, good electrical conductivity, and abundant oxygen-containing functional groups, and can be dispersed in aqueous media [12,13,14,15]. These properties are ideal when incorporating GO with metallic nanoparticles for sensor applications. As a result of these properties, AuNPs@GO exhibits beneficial properties in sensor applications, such as less aggregation, improved dispersion in aqueous media, and better interaction with analytes. In sensor applications, these synergistic effects can improve selectivity and sensitivity [12,13,14,15,16,17].

Epinephrine is an important hormone and neurotransmitter that has a vital role in the human body as part of the catecholamine hormone group. It is responsible for the response to stress and is also involved in various physiological processes, such as cardiovascular and metabolic regulation. Abnormal levels of epinephrine in the human body are associated with a range of medical conditions, such as hypertension, heart failure, and some neurodegenerative diseases. Therefore, sensitive and accurate detection of epinephrine is important for clinical diagnostics [18,19,20].

Due to the importance of epinephrine and the versatility of AuNPs, there has been a remarkable contribution in the literature for epinephrine detection studies based on AuNPs [21,22,23,24,25,26,27]. However, to the best of our knowledge, comprehensive studies utilizing especially in-situ photochemically synthesized AuNPs and AuNPs@GO for small-molecule detection at both µM and nM ranges are limited.

For this reason, this study addresses a green and efficient synthesis method of AuNPs and, at the same time, demonstrates high-performance sensing of epinephrine at low concentration ranges. In detail, AuNPs and AuNPs@GO-based sensing platforms were prepared for the detection of epinephrine at micromolar (µM) and nanomolar (nM) concentration ranges. Both AuNPs and AuNPs@GO were synthesized via a facile and environmentally friendly in-situ photochemical method using BAPO as a photoreducing agent. The morphological, optical, and chemical properties of the synthesized nanostructures were characterized using field-emission scanning electron microscopy (FE-SEM), Dynamic Light Scattering (DLS), UV-Vis absorption spectroscopy, fluorescence emission spectroscopy, and Fourier-transform infrared (FT-IR) spectroscopy in conjunction with epinephrine detection studies. Sensing performances of nanostructures were evaluated based on analytical parameters, such as linear regression, on the best linear subranges, calculation of the limit of detection (LOD), and limit of quantification (LOQ) using two different approaches, namely SE_intercept and S_y/x. Among the tested nanostructures, it was found that AuNPs@GO demonstrates the highest sensitivity and lowest detection limits for epinephrine. These results show the potential of in-situ photochemically synthesized AuNPs and AuNPs@GO sensing platforms for small-molecule detection in analytical and bio-sensing applications.

2. Materials and Methods

2.1. Materials

Tetrachloroauric(III) acid trihydrate (HAuCl₄, ≥99.9% trace metals basis, Sigma-Aldrich, St. Louis, MO, USA), phenylbis(2,4,6-trimethylbenzoyl)phosphine oxide (BAPO, Sigma-Aldrich), DL-epinephrine hydrochloride (>98.0% (T), HPLC, TCI Co. Tokyo, Japan), graphene oxide powder (15–20 sheets, 4–10% edge-oxidized, average 15–20 layers, Sigma-Aldrich), ethanol absolute (ACS, ISO Reagent, Ph. Eur., ISOLAB GmbH, Schweitenkirchen, Germany), and deionized water were used as received.

2.2. Instruments

UV-Vis absorption spectra were recorded using a Varian Cary 50 UV-Vis spectrophotometer (Agilent, Santa Clara, CA, USA). Fluorescence emissions were recorded using a Jobin Yvon-Horiba Fluoromax-P spectrophotometer (Jobin Yvon-Horiba, Kyoto, Japan). Fourier Transform Infrared (FT-IR) spectra were recorded using a Thermo-Scientific Nicolet 6700 FT-IR Spectrophotometer (Waltham, MA, USA). Hamamatsu Lightningcure LC-8 (Shizuoka, Japan) was used for irradiation. Particle size distribution was determined via dynamic light scattering (DLS) with a Malvern Nano ZS zetasizer (Worcestershire, UK). FE-SEM images were taken using Thermo Fisher Quattro ESEM (Waltham, MA, USA).

2.3. Photochemical Preparation of AuNPs and AuNPs@GO

In-situ photochemical preparation of AuNPs was achieved using a 3 mL aqueous solution consisting of 3.33 × 10⁻⁴ M HAuCl₄·H₂O as precursor and 3.33 × 10⁻⁵ M BAPO as photoinitiator/photoreducing agent. The solution was irradiated for 600 s at 10 cm distance with a Hamamatsu Lightningcure LC-8 unit. SPR band formation was recorded by a UV-Vis absorption spectrophotometer at time intervals to confirm the formation of AuNPs.

For the synthesis of AuNPs@GO, the same procedure above was followed, using an aqueous GO suspension (0.5 mg/10 mL). The GO suspension was prepared by dispersing GO powder in deionized water and sonicating for 30 min to obtain uniform dispersion. After combining GO dispersion with HAuCl₄ and BAPO, the solution was allowed to stand for 15 min to promote interaction between GO and Au ions, which leads to the in-situ formation of AuNPs on the GO surface.

2.4. Preparation of Epinephrine and Detection Solutions

Freshly prepared epinephrine stock solutions were used in all experiments to maintain stability. For detection studies, 2 mL of AuNPs or AuNPs@GO solutions were used. Cumulative additions of epinephrine in the range of 2–1000 µL either from 10⁻⁴ M or 10⁻⁶ M epinephrine stock solutions were made, resulting in final epinephrine concentrations ranging from 0.1 to 33.3 µM and 1 to 333 nM, respectively.

The pH values of all prepared solutions were measured before and after epinephrine addition. No significant change to acidic or basic conditions was observed. pH remained within the range of 6.0 to 6.5 throughout all experiments consistently.

2.5. UV-Vis and Fluorescence Spectroscopy Studies

UV-Vis spectroscopy was used to monitor the synthesis of AuNPs and AuNPs@GO under UV irradiation and to investigate their interaction with epinephrine. Spectra were recorded in the range of 200 to 800 nm. Characteristic SPR peaks of AuNPs at ~540 nm and AuNPs@GO at ~530 nm were monitored.

Additionally, fluorescence spectroscopy was used to monitor potential emission changes due to epinephrine interaction with AuNPs and AuNPs@GO nanostructures. An appropriate excitation wavelength of λ_exc = 275 nm was used.

After each cumulative epinephrine addition to the AuNPs and AuNPs@GO solutions, absorbance and fluorescence intensity changes were recorded. Control experiments were conducted with equivalent volumes of deionized water to take into account the dilution effects. All measurements were carried out in triplicate or more, and the average values were used in the calculations.

2.6. LOD and LOQ Calculations

LOD and LOQ were calculated manually using two statistical methods (SE_intercept and S_y/x). A custom Python script (Python 3.13.3) utilizing the NumPy (2.2.5) and Matplotlib (3.10.1) libraries was used to process the raw absorbance data.

Small differences in initial absorbance values may occur between measurements, even when using the same nanoparticle solutions. For this reason, the raw absorbance data from both the analyte and blank experiments were corrected for their individual starting baselines (baseline, t = 0, no addition) as shown in Equations (1) and (2).

$$\Delta A_{\mathrm{Analyte}}=A_{\mathrm{Analyte}}-A_{\mathrm{Start}, \mathrm{Analyte}}$$

(1)

$$\Delta A_{\mathrm{Blank}}=A_{\mathrm{Blank}}-A_{\mathrm{Start}, \mathrm{Blank}}$$

(2)

where $\Delta A_{\mathrm{Analyte}}$ and $\Delta A_{\mathrm{Blank}}$ are the measured changes after each addition, and $A_{\mathrm{Start}, \mathrm{Analyte}}$ and $A_{\mathrm{Start}, \mathrm{Blank}}$ are the initial (baseline, t = 0) absorbance values.

The net corrected signal ($\Delta A_{\mathrm{Corrected}}$) was then calculated by subtracting the blank absorbance change from the analyte absorbance change, which isolates the true analyte-induced absorbance change by removing the blank response.

$$\Delta A_{\mathrm{Corrected}}=\Delta A_{\mathrm{Analyte}}-\Delta A_{\mathrm{Blank}}$$

(3)

For each addition (corresponding to a specific volume and a resulting concentration), the change in analyte absorbance ($\Delta A_{\mathrm{Analyte}}$), blank absorbance ($\Delta A_{\mathrm{Blank}}$), and net corrected absorbance values were calculated ($\Delta A_{\mathrm{Corrected}}$).

To account for dilution effects from cumulative additions, the final analyte concentration $C_{\mathrm{final}}$ was calculated using Equation (4):

$$C_{\mathrm{final}}={\displaystyle \frac{C_{\mathrm{stock}}\times V_{\mathrm{added}}}{V_{\mathrm{initial}}+V_{\mathrm{added}}}}$$

(4)

where $C_{\mathrm{stock}}$ is the stock analyte concentration, $V_{\mathrm{added}}$ is the cumulative volume added, and $V_{\mathrm{initial}}$ is the initial volume of the nanoparticle solution.

The paired values of $\Delta A_{\mathrm{Corrected}}$ and $C_{\mathrm{final}}$ for each addition were used for calibration, LOD and LOQ calculations, residual analysis, and plotting.

Calibration modeling was performed using least-squares regression to fit a linear model (Equation (5)), where the slope reflects system sensitivity and the intercept adjusts for baseline offset:

$$\Delta A_{\mathrm{Corrected}}=\mathrm{slope}\times C_{\mathrm{final}}+\mathrm{intercept}$$

(5)

The fit quality was assessed via residual analysis (Equation (6)), and the coefficient of determination R² was used to evaluate model performance (Equation (7))

$$r_i=\Delta A_{\mathrm{Corrected},i}-\Delta A_{\mathrm{Fitted},i}$$

(6)

$$R^2=1-{\displaystyle \frac{\sum r_i^2}{\sum {\left(\Delta A_{\mathrm{Corrected},i}-\overline{\Delta A_{\mathrm{Corrected}}}\right)}^2}}$$

(7)

LOD and LOQ were first calculated using the standard error of the intercept (SE_intercept) and the slope of the calibration curve as shown in Equations (8) and (9). Here, LOD represents the minimum detectable concentration, while LOQ indicates the minimum quantifiable concentration with acceptable precision:

$$\mathrm{LOD}={\displaystyle \frac{3\times {\mathrm{SE}}_{\mathrm{intercept}}}{\mathrm{slope}}}$$

(8)

$$\mathrm{LOQ}={\displaystyle \frac{10\times {\mathrm{SE}}_{\mathrm{intercept}}}{\mathrm{slope}}}$$

(9)

Additionally, LOD and LOQ were calculated using the S_y/x method. S_y/x is the standard error of regression, which quantifies the scatter of data points around the fitted line. It was calculated as follows (Equation (10))

$$S_{y/x}=\sqrt{{\displaystyle \frac{\sum {\left(y_i-\widehat{y_i}\right)}^2}{n-2}}}$$

(10)

where $y_i$ is the observed absorbance, $\widehat{y_i}$ is the predicted value, and n is the number of data points. In other words, S_y/x is the square root of the sum of squared residuals divided by the degrees of freedom. LOD and LOQ using this method were calculated according to Equations (11) and (12).

$$LOD={\displaystyle \frac{3.3\times S_{y/x}}{slope}}$$

(11)

$$LOQ={\displaystyle \frac{10\times S_{y/x}}{slope}}$$

(12)

All LOD and LOQ calculations were strictly performed on raw (non-normalized) absorbance data to ensure the detection limits reflect the true analytical signal, free from any post-processing adjustments.

3. Results and Discussion

3.1. Photochemical Preparation of AuNPs and AuNPs@GO in Aqueous Media

The photochemical synthesis of AuNPs and AuNP@GO was carried out in aqueous media using BAPO as the photoreducing agent, which is dissolved in a small amount of ethanol, and initiated by irradiation with a medium-pressure mercury lamp. The proposed mechanism for the in-situ photochemical formation of AuNPs in the presence of BAPO, as a Type I photoinitiator, is presented in Figure 1. Upon irradiation of the BAPO photoinitiator, two radicals are generated according to the well-established initiation mechanism of BAPO. Both the benzoyl and phosphinoyl radicals are not only efficient in initiating photopolymerization of monomers but also play a crucial role in the presence of HAuCl₄ for the photoreduction of ionic gold species (Au³⁺) to metallic gold nanoparticles (Au⁰), as depicted in Figure 1. During the photochemical preparation, strong SPR band of AuNPs around 540 nm and AuNPs@GO around 530 nm were observed and as shown in Figure 2 and Figure 3 colorless solutions turned to reddish dark purple color after certain irradiation times.

AuNPs and AuNPs@GO were synthesized with the aim of enabling highly sensitive detection of the key neurotransmitter epinephrine and facilitating the development of an efficient biosensor. To evaluate the effect of GO on sensor performance, AuNPs@GO was prepared due to the additional benefits of GO, such as its large surface area, excellent dispersibility in water, and abundance of oxygen-containing functional groups, which facilitate strong interactions with both AuNPs and target molecules like epinephrine. The improved stability, sensitivity, and electron transfer are expected to be obtained from the hybrid structure of AuNPs@GO, which would lead to more efficient biosensors.

The photochemical synthesis of AuNPs was monitored by following the SPR band values at the characteristic wavelengths. To confirm the adequacy of the irradiation time, additional experiments were performed with illumination periods longer than 600 s. However, no significant changes were observed in the UV-Vis spectra beyond this point. Therefore, solutions irradiated for 600 s were used in all subsequent studies.

The SPR maximum of AuNPs was found at 540 nm (Figure 2). A slight blue shift in the SPR peak was observed when GO was incorporated into the system and was detected at 530 nm (Figure 3), which possibly indicates the interactions between the AuNP surface and the oxygen-containing functional groups on GO, which can influence the surface chemistry and stabilization of the nanoparticles. These maximum SPR bands, at 540 nm for AuNPs and 530 nm for AuNPs@GO, were used as the reference wavelengths for absorbance measurements in the detection studies.

3.2. Sensing Performance of AuNPs for Epinephrine Detection

SPR band changes of AuNPs with the addition of epinephrine in the range of 0.1–33.3 µM concentration are given in Figure 4. An increasing epinephrine concentration followed a consistent decrease in the SPR band, while the peak position remained unchanged. The decrease of the SPR band is possibly due to interactions between epinephrine and AuNPs.

As shown in Figure 5, a more distinct decrease in absorbance was observed compared to the results obtained in the µM range when the epinephrine concentration was reduced to the nanomolar range (1–333 nM).

The obtained results confirmed that efficient surface interactions at lower epinephrine concentrations are possibly due to favorable binding dynamics. These results proved that epinephrine was effectively adsorbed onto the AuNP surface even in the nanomolar range, leading to measurable changes in the SPR signal.

The FT-IR spectrum of (i) epinephrine, (ii) AuNPs, and (iii) AuNPs after the addition of epinephrine is given in Figure 6. Significant absorption peaks can be identified for epinephrine (curve i) at 3300 cm⁻¹ (–OH and –NH stretching), 1600 cm⁻¹, 1485 cm⁻¹, and 1340 cm⁻¹, 1200–1000 cm⁻¹ (C–O stretching), which are attributed to the stretching frequencies of phenolic –OH or aliphatic –OH. It was observed that epinephrine binds to the AuNPs surface by weak coordination bonds involving its phenolic –OH and/or amine groups. As shown in Figure 6, spectral shifts of the functional groups were observed in the FT-IR analysis, which supported this interaction. When epinephrine was added to the AuNPs solution, the broad absorption band near 3300 cm⁻¹ disappeared, which corresponds to the O–H and N–H stretching vibrations of phenolic and amine groups. For this reason, these functional groups are thought to be directly involved in interactions with the AuNP surface, possibly through weak coordination or hydrogen bonding. The disappearance of epinephrine’s characteristic vibrational bands in the AuNPs + Epinephrine FT-IR spectrum suggests strong surface adsorption of epinephrine. These interactions may lead to IR signal suppression due to molecular immobilization, electronic coupling, and surface-induced damping of vibrational transitions [23].

Fluorescence emission spectra of AuNPs before and after adding a certain amount (0.1–33.3 µM) of epinephrine solution are displayed in Figure 7. The weak fluorescence emission wavelength for epinephrine was found to be at 320 nm, and no significant change in intensity was observed until a threshold of 20 µM (500 µL addition) was reached, indicating a strong interaction between epinephrine and the AuNPs surface. Metal-enhanced fluorescence (MEF) effects can be responsible for this enhancement and the suppression of non-radiative decay pathways due to the adsorption of epinephrine via its phenolic and amine groups. Additionally, partial oxidation of epinephrine on the AuNPs surface may contribute to forming new emissive species.

Best linear subrange regression analyses for AuNPs at 540 nm, corresponding to two distinct epinephrine concentration ranges: (a) µM and (b) nM, are presented in Figure 8. In both cases, linearity was well maintained, as reflected by high R² values of 0.98984 (µM) and 0.9626 (nM), indicating that consistent and reproducible spectral responses were obtained. The calculated analytical parameters are summarized in

Table 1. Calculated parameters for epinephrine detection using AuNPs with SE_intercept and S_y/x methods.

Name	Parameter	Method
		SE of Intercept	SE of Regression
AuNPs + Epinephrine (µM)	LOD	1.77 µM	3.05 µM
	LOQ	5.90 µM	9.23 µM
	R2	0.98984
	Slope (Sensitivity)	0.0052
	SEintercept or Sy/x	0.00307	0.00480
	Best linear subrange	1.48 to 23.08 µM
AuNPs + Epinephrine (nM)	LOD	56.03 nM	74.21 nM
	LOQ	186.76 nM	224.89 nM
	R2	0.9626
	Slope (Sensitivity)	1.76 × 10−4
	SEintercept or Sy/x	0.00328	0.00395
	Best linear subrange	47.62 to 333.33 nM

. LOD values were found to be 1.77 µM (SE_intercept) and 3.05 µM (S_y/x) in the µM range. In the nM range, significantly lower LOD values of 56.03 nM and 74.21 nM were determined, respectively. Similar results for the LOQ values were obtained. LOQ values were calculated as 5.90 µM and 9.23 µM in the µM range, and 186.76 nM and 224.89 nM in the nM range. These results indicate that the AuNPs are capable of detecting epinephrine at nM levels without additional needs such as surface modifications or amplification techniques. Best linear subranges were determined as 1.48–23.08 µM and 47.62–333.33 nM for the µM and nM ranges, respectively, showing that flexible and quantifiable detection was achieved across both mid- and low-concentration levels.

The observed visual color changes are given in Figure 9a, representing before and after epinephrine addition to the AuNPs solution. A deep reddish dark purple color was observed for AuNPs, while a slight fading of color after the addition of 33.3 µM epinephrine was observed, and more fading in color was seen with the addition of 333 nM epinephrine. FE-SEM images (Figure 9b–d) were acquired using a backscatter detector at 100,000× magnification. Although some clustering was present, individual nanoparticles were distinguishable. Particle sizes were measured in the 46–62 nm range for AuNPs, 24–55 nm after 33.3 µM epinephrine addition, and 32–56 nm after 333 nM epinephrine addition, indicating the successful synthesis of nanoparticles. Size measurements of the AuNPs on FE-SEM images are given in Supplementary Materials in Figures S5–S7.

3.3. Sensing Performance of AuNPs@GO for Epinephrine Detection

SPR band of AuNPs@GO at 530 nm after the addition of epinephrine solution in the 0.1–33.3 µM range is given in Figure 10. A decrease in SPR band intensity was observed by increasing the epinephrine concentration, indicating consistent surface interaction of AuNPs@GO and epinephrine.

The SPR bands of AuNPs@GO at 530 nm following the addition of epinephrine in the 1–333 nM range are given in Figure 11. A notable decrease in absorbance was observed, which confirms that the interaction between epinephrine and the nanostructure surface remains detectable even at low concentrations.

In the case of AuNPs@GO, a stronger interaction with epinephrine is expected due to additional π–π stacking and hydrogen bonding between epinephrine and the oxygen-containing functional groups on the GO surface. This expected result was proven with FT-IR spectra as shown in Figure 12 for (i) epinephrine, (ii) AuNPs@GO, and (iii) AuNPs@GO after the addition of epinephrine. The disappearance of the 3300 cm⁻¹ band was observed upon the addition of epinephrine to AuNPs@GO, similar to AuNPs. In fact, the stronger interaction in the AuNPs@GO system possibly results from the abundance of oxygen-containing functional groups on GO, which promotes both hydrogen bonding and π–π interactions with the aromatic ring of epinephrine. Restricting the vibrational freedom of O–H and N–H bonds reduces IR activity. The adsorption of epinephrine onto AuNPs and AuNPs@GO may be attributed to the overall decrease in spectral intensity. This adsorption can induce partial aggregation or electronic coupling effects, resulting in a reduction of the concentration of well-dispersed, optically active nanoparticles, which in turn leads to a decrease in the absorbance intensity of the entire spectrum. These spectral changes provide further evidence of effective epinephrine binding to the hybrid nanostructure.

Fluorescence studies were conducted using epinephrine concentrations in both the micromolar (µM) and nanomolar (nM) ranges for the AuNPs@GO sensor system. Figure 13 and Figure 14 recorded the fluorescence response of the AuNPs@GO sensor system, where epinephrine concentration ranges from µM to nM.

After adding the epinephrine solution, a similar behavior was observed in the fluorescence emission spectra of AuNPs@GO compared to the results obtained for AuNPs. The AuNPs@GO system exhibited a more pronounced fluorescence intensity increase upon the addition of epinephrine (Figure 13) than the AuNPs. However, the sharp increase in emission intensity within the µM concentration range occurred at a lower concentration, approximately 13.04 µM.

In this case, synergistic interactions between epinephrine and the oxygen-containing functional groups on GO via hydrogen bonding and π–π stacking could be considered in addition to coordination with AuNPs. These interactions lead to increased rigidity and reduced non-radiative relaxation of the epinephrine molecule, resulting in amplified fluorescence intensity. The hybrid structure of AuNPs@GO promotes more efficient adsorption and stabilization of epinephrine, which leads to an enhancement of the overall fluorescence response. When the concentration was in the nM range, the 320 nm emission band slightly increased, while a new emission peak at around 400 nm was observed (Figure 14). However, this new peak was not observed at the micromolar range (Figure 13). This is possibly due to specific interactions between epinephrine and the AuNPs@GO surface, such as π–π stacking and hydrogen bonding, which create a new emissive environment. This surface-mediated fluorescence enhancement is most effective when binding sites are available (i.e., at low concentrations), and diminishes at higher concentrations due to surface saturation and self-quenching effects.

Best linear subrange regression analyses for AuNPs@GO at 530 nm, corresponding to epinephrine additions in the µM and nM ranges, are presented in Figure 15. In contrast to AuNPs, improved detection performance in the nM range was demonstrated by the AuNPs@GO, as evidenced by the lowest LOD values achieved in this study: 26.55 nM (SE_intercept) and 49.09 nM (S_y/x). Calculated analytical parameters are given in

Table 2. Calculated parameters for epinephrine detection using AuNPs@GO based on SE_intercept and S_y/x methods.

Name	Parameter	Method
		SE of Intercept	SE of Regression
AuNPs@GO + Epinephrine (µM)	LOD	2.80 µM	3.96 µM
	LOQ	9.34 µM	11.99 µM
	R2	0.95591
	Slope (Sensitivity)	0.00799
	SEintercept or Sy/x	0.00746	0.00959
	Best linear subrange	1.48 to 13.04 µM
AuNPs@GO + Epinephrine (nM)	LOD	26.55 nM	49.09 nM
	LOQ	88.50 nM	148.77 nM
	R2	0.98189
	Slope (Sensitivity)	1.89 × 10−4
	SEintercept or Sy/x	0.00167	0.00281
	Best linear subrange	14.78 to 285.71 nM

. Even a moderate linearity was observed in the µM range (R² = 0.95591), a stronger correlation was observed in the nM range (R² = 0.98189), which indicates that enhanced response at lower analyte concentrations was promoted by the AuNPs@GO hybrid structure. In this case, a similar trend was observed for LOQ values, with 88.50 nM and 148.77 nM calculated for the nM range, and in comparison, higher values: 9.34 µM and 11.99 µM calculated for the µM range. These results indicate that enhanced detection capabilities at low concentrations were achieved through the incorporation of GO into the system. The best linear subranges were identified as 1.48–13.04 µM and 14.78–285.71 nM. Taken together, the data suggest that a more effective sensing platform for trace-level detection of epinephrine was provided by the AuNPs@GO nanostructures when compared to AuNPs, particularly in the nM range, where high sensitivity is critical for potential biosensing applications.

When two systems are compared, the AuNPs@GO nanostructures exhibited the most favorable analytical performance, especially in the nM concentration range. These results show the beneficial role of GO in enhancing sensitivity towards epinephrine. In order to assure the robustness of statistical calculations, all parameters were calculated with two different methods, namely SE_intercept and S_y/x. This way, consistent and cross-validated detection metrics were calculated. Additionally, residuals vs. epinephrine concentration were plotted for all regression models and are provided in the Supplementary Materials (Figures S1–S4).

The color changes observed in the AuNPs@GO sensing system before and after the addition of epinephrine are shown in Figure 16a. A reddish deep purple color with a grayish hue was observed before the addition of epinephrine. Following the addition of both 33.3 µM and 333 nM epinephrine, the color shifted to lighter purple tones, appearing more faded. Figure 16b–d shows FE-SEM images acquired using a backscatter detector at magnifications of 50,000× and 100,000×. AuNPs were seen to be clustered on the GO surface but remained individually distinguishable. Particle size distributions were measured in the range of 43–135 nm for AuNPs@GO, 91–154 nm after the addition of 33.3 µM epinephrine, and 28–167 nm following 333 nM epinephrine addition, indicating the successful synthesis of nanostructures. Size measurements of the AuNPs@GO on FE-SEM images are given in Supplementary Materials in Figures S8–S10.

Dynamic light scattering (DLS) measurements were performed to determine the hydrodynamic sizes of the nanoparticles. A hydrodynamic diameter of 476.1 nm was measured for AuNPs@GO. Following the addition of 333 nM epinephrine, the size decreased to 119 nm (see

Table 3. DLS size distributions of AuNPs and AuNPs@GO detection solutions before and after total Epinephrine (333 nM) addition.

Name	Size (nm)
AuNPs	407.7
AuNPs + 333 nM Epinephrine	95.3
AuNPs@GO	476.1
AuNPs@GO + 333 nM Epinephrine	119

). A similar size reduction was observed for AuNPs, which decreased from 407.7 nm to 95.3 nm after Epinephrine addition. These results suggest strong interactions between epinephrine molecules and nanostructures, possibly leading to more compact and stabilized assemblies in solutions. Although some agglomeration was observed by FE-SEM images, DLS values reflect the solvated state and dynamic aggregation profile of the nanoparticles in dispersion. The difference between the two techniques was attributed to their main differences: FE-SEM is capable of providing direct measurements of the AuNPs size in a dry environment with high spatial resolution. On the other hand, DLS reflects the AuNPs’ behavior in aqueous medium, in which solvation effects and interparticle interactions are included. Therefore, FE-SEM imaging was found to give more accurate size information of AuNPs, while DLS was beneficial to understand the dispersion stability and colloidal properties of the AuNPs in aqueous media.

A proposed interaction mechanism between epinephrine and AuNPs is illustrated in Figure 17. The interaction is believed to occur primarily through the catechol moiety, where adjacent hydroxyl groups are known to bind to the AuNPs surface via coordination or hydrogen bonding. Upon binding, changes in the local surface environment are induced, leading to observable shifts in color and UV-Vis spectra. While minor contributions from other functional groups of epinephrine cannot be excluded, the catechol unit is considered the dominant site of interaction.

4. Conclusions

In conclusion, AuNPs and AuNPs@GO were successfully synthesized using a well-optimized in-situ photochemical method in aqueous media. This environmentally friendly approach was carried out using bis-acylphosphine oxide (BAPO) as the photoreducing agent and produced stable, reproducible colloidal dispersions. Both nanostructures were used for epinephrine detection in the concentration ranges of µM and nM. In the nM range, a better performance was obtained with AuNPs@GO compared to the AuNPs. This finding was supported by comprehensive analyses with spectroscopic methods. For AuNPs@GO, lower LOD and LOQ values were calculated, and still, the fluorescence intensity was detectable. This result was attributed to the incorporation of GO into the system, which likely promoted stronger interactions between nanostructures and epinephrine. The increase in sensitivity results from synergistic effects between AuNPs and GO due to the high surface area of GO, which provides more active sites for AuNPs and gives better adsorption and immobilization of molecules (e.g., epinephrine, dopamine). All analytical parameters were calculated with two different methods, namely SE_intercept and S_y/x. This approach led to statistically more reliable results. Overall, the results showed that the AuNPs@GO system offers a simple, effective, and environmentally friendly strategy for detecting low concentrations of small molecules like epinephrine and may be extended to other sensing applications in the future.