Iodide-Mediated Etching of Gold Nanostar for the Multicolor Visual Detection of Hydrogen Peroxide

A multicolor visual method for the detection of hydrogen peroxide (H2O2) was reported based on the iodide-mediated surface etching of gold nanostar (AuNS). First, AuNS was prepared by a seed-mediated method in a HEPES buffer. AuNS shows two different LSPR absorbance bands at 736 nm and 550 nm, respectively. Multicolor was generated by iodide-mediated surface etching of AuNS in the presence of H2O2. Under the optimized conditions, the absorption peak Δλ had a good linear relationship with the concentration of H2O2 with a linear range from 0.67~66.67 μmol L−1, and the detection limit is 0.44 μmol L−1. It can be used to detect residual H2O2 in tap water samples. This method offered a promising visual method for point-of-care testing of H2O2-related biomarkers.


Introduction
Hydrogen peroxide (H 2 O 2 ) is widely used in fields such as the textile, chemical, environmental protection, medical, and food industries due to its ecological friendliness, strong oxidation ability, and ability to kill bacteria and bleach [1][2][3][4]. However, the direct addition of H 2 O 2 in food production is prohibited by most countries, as residual H 2 O 2 and its derived active substances can cause oxidative damage to human health [5,6]. Food-grade H 2 O 2 residue is considered a "recognized safe substance" (JECFA, 2004), and according to the regulations of the Food and Agriculture Organization of the United States, the maximum concentration of H 2 O 2 in dairy products is limited to 0.05%. In the Chinese national standard (GB53009.226-2016), the H 2 O 2 content in food is strictly limited to less than 3 mg kg −1 . Excessive H 2 O 2 may cause Alzheimer's, Parkinson's, amyotrophic lateral sclerosis, diabetes, traumatic brain injury, ischemia/reperfusion injury, learning and memory impairment, and other diseases [7][8][9][10]. Therefore, developing a simple and effective method to monitor H 2 O 2 concentration is of great significance.
At present, the commonly used technologies for H 2 O 2 detection include the colorimetric method [11][12][13], electrochemical method [14,15], fluorescence method [16,17], liquid chromatography [18,19], SERS [20,21], and others. Colorimetric detection has advantages in terms of simplicity, cost-effectiveness, and sensitivity in visual monitoring [22,23]. Most colorimetric methods used for H 2 O 2 detection are based on the color reaction of substrate 3, 3, 5, 5,-tetramethylbenzidine (TMB) [22,24,25] and o-phenylenediamine (OPD) [26][27][28] oxidized by H 2 O 2 under the action of a catalyst to quantify H 2 O 2 . However, these organic compounds are toxic and are easily oxidized by O 2 . Therefore, developing a stable H 2 O 2 detection method without chromogenic substrates is meaningful. Precious metal nanostructures have many excellent properties, such as good light scattering, absorption, and photothermal conversion capabilities [11,29]. Au NS is an uncommon type of gold nanostructure. In 2006, Nehl et al. [30] systematically studied gold nanostructures and multi-branched nanoparticles with sharp tips and unique local surface plasmon resonance of gold nanostructure. In 2006, Nehl et al. [30] systematically studied gold nanostructures and multi-branched nanoparticles with sharp tips and unique local surface plasmon resonance (LSPR) characteristics. Au NS is a promising choice for biosensor application due to its sharp corners, plasma resonance, and lightning rod effect [31,32]. Research has shown that the LSPR characteristics of Au NS are related to their size, shape, and composition. Chemical etching can change the length of the sharp angle of Au NS and the surface plasmon resonance of Au NS, which means that the chemical etching of the sharp angle of Au NS will shift the LSPR absorption peak of Au NS and produce a blue color change [12,13,33]. The etching process can be adjusted by iodide to quickly and sensitively generate the blue shift of the surface plasmon resonance spectrum [34].
Considering the important role of H2O2, this work intended to construct a multicolor sensor based on the reaction between H2O2, Au NS, and I − (Scheme 1A). Firstly, H2O2 reacts with I − to generate I2 with the help of horseradish peroxidase (HRP). The remaining I − etches the sharp corners of Au NS, which changes the absorption peak of LSPR shifts and the color of the solution. With the increasing concentration of I − , the degree of blue shift (Δλ) increases. The color of the mixture changes from blue to purple and finally turns pink, and Δλ can be used for the quantification of H2O2 (Scheme 1B). Scheme 1. Schematic illustration of iodide-mediated etching of Au NS for H2O2 detection (A) and color changes with increasing H2O2 (B).

Apparatus
The transmission electron microscope (FE-TEM) images of Au NS were collected by JEOL200 kV field emission transmission electron microscope (JEM-2100 F, JEOL, Tokyo, Japan). We used KQ5200DE ultrasonic cleaner (Kunshan Ultrasonic Instrument Co., Ltd., Kunshan, China) to perform ultrasonic treatment on the sample and the UV-Vis spectrum was recorded by using a Tecan Spark microplate reader (Tecan, Switzerland).

Synthesis of Au NS [34]
Golden Seed: Add a trisodium citrate solution (4 mL, 1%) to a rapidly stirred boiling HAuCl 4 aqueous solution (50 mL, 0.5 mmol L −1 ) and boil for 30 min. The solution color changes from purple-red to wine-red. Then, cool the wine-red solution to room temperature to obtain the gold seed solution, and store it at 4°C for later use.

Detection of H 2 O 2
After incubation at 37°C for 40 min, 50 µL HRP (40 ng mL −1 ), 100 µL KI and 100 µL H 2 O 2 of different concentrations were added into 250 µL Tris-HCl (10 mmol L −1 , pH = 3.0) buffer solution. Take 50 µL and add it to 250 µL Au NS. After the reaction at 37°C for 25 min, the ultraviolet absorption spectra of each pore were recorded by a Tecan microplate reader. The calibration curve of ∆λ and H 2 O 2 concentration was constructed by drawing the wavelength shift change values (∆λ) of different concentrations of H 2 O 2 , where wavelength change value ∆λ = λ 0 − λ (λ 0 is the peak value of Au NS, λ is the peak value measured after the etching of Au NS).

Detection of H 2 O 2 in Tap Water
A standard addition method was used to evaluate the feasibility of iodide-mediated Au NS etching for detecting H 2 O 2 in tap water. Follow the above steps (drawing a standard curve) to determine the concentration of H 2 O 2 in tap water by recording the change in wavelength offset (∆λ).

Characterization of AuNS
AuNS are synthesized by a seed-mediated method. First, small gold nanoparticles are synthesized, and then gold seeds are added to the growth solution containing HAuCl 4 , a reducing agent, and surfactant. The newly reduced Au • grows on the surface of the seeds to form Au NS [18,35,36]. The UV-Vis spectrum measured using a Tecan microplate reader is shown in Figure 1A. The absorbance intensities of the Au NS solution at different wavelengths were recorded as the optical density (OD). It can be seen that the Au NS has two SRP bands located at 736 nm and 550 nm, respectively. λ 0 represents the maximum absorption peak of Au NS at 736 nm. Figure 1B shows that Au NS has multiple sharp edges. Compared to smooth surfaces, the multiple sharp edges of Au NS have special sensitivity to the etching of I − , and iodide binds to the surface during the etching process [36]. After I − etched Au NS, the sharp corners become shorter, and Au NS are etched into spherical nanostructures ( Figure 1C). Due to the fact that gold atoms are "soft" Lewis acids, and I − is a classic "soft" Lewis donor [34], iodide has a high affinity for gold. Oxidized Au ions and iodide can chelate at the interface of Au NS [37], resulting in uneven or even larger Au NS sizes after etching.

Feasibility of Iodide-Mediated Etching of AuNS for Detecting H2O2
To investigate the feasibility of using iodide-mediated surface etching of detecting H2O2, we first validated the UV-Vis spectra of different mixtures. As Figure 2A,B, when only Au NS exists, the absorption peak was measured at 73 ure 2B, curve e). After adding a final concentration of I − as 80 µmol L −1 , the peak visible absorption peak changed from 736 nm to 588 nm, resulting in a signifi shift of the UV peak absorption peak ( Figure 2B, curve d). The addition of H2O separately does not affect the etching of AuNS by I − (curves b and c in Figure  when both exist simultaneously does the absorption peak of the mixture shift to as in Figure 2B). This is because the addition of H2O2 or HRP alone into the reacti does not affect the concentration of I − , and all I − is used for etching AuNS. After tion of both, HRP can catalyze the oxidation of iodide to elemental iodine in the of H2O2 [34]. It led to a decrease in the remaining concentration of iodide. A decr ing degree of Au NS was also observed which resulted in a red shift of the absorp of Au NS. In addition, we validated the relationship between the concentration the degree of etching. When the concentration of I − increases from 0 mmol L −1 to L −1 , the etching degree gradually increases and reaches the plateau ( Figure 2D corresponding absorption peak shifts from 736 nm to 556 nm. At the same time of the solution changes from dark blue to purple to pink (Inset of Figure 2D). 2C, it can be seen that as the concentration of I − increases from 0 mmol L −1 to 0

Feasibility of Iodide-Mediated Etching of AuNS for Detecting H 2 O 2
To investigate the feasibility of using iodide-mediated surface etching of AuNS for detecting H 2 O 2 , we first validated the UV-Vis spectra of different mixtures. As shown in Figure 2A,B, when only Au NS exists, the absorption peak was measured at 736 nm ( Figure 2B, curve e). After adding a final concentration of I − as 80 µmol L −1 , the peak of the UV visible absorption peak changed from 736 nm to 588 nm, resulting in a significant blue shift of the UV peak absorption peak ( Figure 2B, curve d). The addition of H 2 O 2 and HRP separately does not affect the etching of AuNS by I − (curves b and c in Figure 2B). Only when both exist simultaneously does the absorption peak of the mixture shift to red (curve as in Figure 2B). This is because the addition of H 2 O 2 or HRP alone into the reaction system does not affect the concentration of I − , and all I − is used for etching AuNS. After the addition of both, HRP can catalyze the oxidation of iodide to elemental iodine in the presence of H 2 O 2 [34]. It led to a decrease in the remaining concentration of iodide. A decrease etching degree of Au NS was also observed which resulted in a red shift of the absorption peak of Au NS. In addition, we validated the relationship between the concentration of I − and the degree of etching. When the concentration of I − increases from 0 mmol L −1 to 1.2 mmol L −1 , the etching degree gradually increases and reaches the plateau ( Figure 2D), and the corresponding absorption peak shifts from 736 nm to 556 nm. At the same time, the color of the solution changes from dark blue to purple to pink (Inset of Figure 2D). In Figure 2C, it can be seen that as the concentration of I − increases from 0 mmol L −1 to 0.08 mmol L −1 , the etching degree gradually increases. When the concentration of I − increases from 0.05 mmol L −1 to 0.06 mmol L −1 , the maximum absorption peak of the Au NS solution shows a large blue shift (∆λ). However, the increase in ∆λ becomes little from 0.06 mmol L −1 to 0.08 mmol L −1 . Therefore, an I − concentration of 0.06 mmol L −1 was selected for the sensitive detection of H 2 O 2 . The above experimental results indicate that this method based on the iodide-mediated etching of AuNS for detecting H 2 O 2 has good feasibility.

Optimization of Experimental Conditions
To obtain a more sensitive response, several parameters in the experimental system were optimized, such as the time for I − etching AuNS, the incubation time for the reaction between HRP and H2O2, and the reaction temperature ( Figure 3). As shown in Figure 3A, the reaction time for I − -induced etching of AuNS was optimized. After adding a concentration of 0.06 mmol L −1 , the plateau was reached with a reaction time of 25 min, indicating that the etching of AuNS by I − can reach stability after 25 min. Figure 3B shows the optimization of incubation time for HRP and H2O2. At this point, the concentration of H2O2 is 33.3 µmol L −1 . As the time of HRP catalyzing H2O2 prolongs, the etching degree gradually decreases and reaches stability. This is because the amount of I − consumed by H2O2 under HRP catalysis increases, resulting in a decrease in the remaining I − , leading to a decrease in the degree of Au NS etching and a shift in the UV absorption peak. Finally, we optimized the reaction temperature. The experimental results indicate that an increase in temperature contributes to the etching of the reaction system ( Figure 3C), but a higher temperature will reduce the catalytic efficiency of HRP. Therefore, we chose a reaction time of 25 min for I − -induced etching of AuNS, an incubation time of 40 min for the reaction between HRP and H2O2, and a reaction temperature of 37 ℃.

Optimization of Experimental Conditions
To obtain a more sensitive response, several parameters in the experimental system were optimized, such as the time for I − etching AuNS, the incubation time for the reaction between HRP and H 2 O 2 , and the reaction temperature ( Figure 3). As shown in Figure 3A, the reaction time for I − -induced etching of AuNS was optimized. After adding a concentration of 0.06 mmol L −1 , the plateau was reached with a reaction time of 25 min, indicating that the etching of AuNS by I − can reach stability after 25 min. Figure 3B shows the optimization of incubation time for HRP and H 2 O 2 . At this point, the concentration of H 2 O 2 is 33.3 µmol L −1 . As the time of HRP catalyzing H 2 O 2 prolongs, the etching degree gradually decreases and reaches stability. This is because the amount of I − consumed by H 2 O 2 under HRP catalysis increases, resulting in a decrease in the remaining I − , leading to a decrease in the degree of Au NS etching and a shift in the UV absorption peak. Finally, we optimized the reaction temperature. The experimental results indicate that an increase in temperature contributes to the etching of the reaction system ( Figure 3C), but a higher temperature will reduce the catalytic efficiency of HRP. Therefore, we chose a reaction time of 25 min for I − -induced etching of AuNS, an incubation time of 40 min for the reaction between HRP and H 2 O 2 , and a reaction temperature of 37°C. Biosensors 2023, 13, x FOR PEER REVIEW 6 of 10

Performance Analysis of Iodide-Mediated AuNS Etching for Detecting H2O2
Under optimal experimental conditions, the linear response range and detection limit of H2O2 detected by the constructed sensor platform were investigated by monitoring the offset value of the absorption peak (Δλ). As shown in Figure 4A (Table S1), this method is rapid and sensitive, and shows a multicolor change which has great potential application in the field of point-of-care testing.

Performance Analysis of Iodide-Mediated AuNS Etching for Detecting H 2 O 2
Under optimal experimental conditions, the linear response range and detection limit of H 2 O 2 detected by the constructed sensor platform were investigated by monitoring the offset value of the absorption peak (∆λ). As shown in Figure 4A (Table S1), this method is rapid and sensitive, and shows a multicolor change which has great potential application in the field of point-of-care testing.

Performance Analysis of Iodide-Mediated AuNS Etching for Detecting H2O2
Under optimal experimental conditions, the linear response range and detection limit of H2O2 detected by the constructed sensor platform were investigated by monitoring the offset value of the absorption peak (Δλ). As shown in Figure 4A,B, with the increase in H2O2 concentration in the range of 0.67-66.67 µmol L −1 , Δλ gradually increases and reaches a plateau. In the range of 0.67-33.33 µmol L −1 , Δλ has a strong linear relationship with the concentration of H2O2 ( Figure 4C). The linear equation is Δλ = −2.84CH2O2 + 123.69 (R 2 = 0.9908, n = 3) and the LOD value is 0.44 µmol L −1 (n = 20, 3σ0/K). Compared with other reported H2O2 sensors (Table S1), this method is rapid and sensitive, and shows a multicolor change which has great potential application in the field of point-of-care testing.

The Selectivity and Interference Study of Iodide-Mediated Etching of AuNS for Detecting H 2 O 2
Tap water often contains a variety of salts, so it is necessary to explore the interference of different ions in the detection of hydrogen peroxide in tap water. To evaluate the selectivity and interference study of this method for H 2 O 2 detection in tap water, six different interfering substances (K + , Na + , Cl − , NO 3 − , SO 4 2− , CO 3 2− ,) were selected for the study. Figure 5A shows that under the same experimental conditions, except for H 2 O 2 , even if the concentration of other interfering substances is 50 times higher than the concentration of H 2 O 2 , the ∆λ shows no significant decrease. The above results indicate that this method has high selectivity for the detection of H 2 O 2 . Subsequently, we selected the above-mentioned interfering substances for the interference study. As shown in Figure 5B, the ∆λ value of H 2 O 2 detection with the addition of interfering substances did not change.
The results indicate that this method has good stability for detecting H 2 O 2 .

The selectivity and Interference Study of Iodide-Mediated Etching of AuNS for Detecting H2O2
Tap water often contains a variety of salts, so it is necessary to explore the interference of different ions in the detection of hydrogen peroxide in tap water. To evaluate the selectivity and interference study of this method for H2O2 detection in tap water, six different interfering substances (K + , Na + , Cl − , NO3 − , SO4 2− , CO3 2− ,) were selected for the study. Figure  5A shows that under the same experimental conditions, except for H2O2, even if the concentration of other interfering substances is 50 times higher than the concentration of H2O2, the Δλ shows no significant decrease. The above results indicate that this method has high selectivity for the detection of H2O2. Subsequently, we selected the above-mentioned interfering substances for the interference study. As shown in Figure 5B, the Δλ value of H2O2 detection with the addition of interfering substances did not change. The results indicate that this method has good stability for detecting H2O2.

Detection of H2O2 in Real Samples
To evaluate the potential applicability of this method for detecting H2O2 in real samples, tap water was selected as the actual sample, and the accuracy of the method was verified by the spiked recovery method. The specific experimental steps are as follows: Take 30% H2O2 and dilute it with tap water to different concentrations. Add 50 µL HRP (40 ng mL −1 ) and 100 µL NaI to Tris-HCl (10 mmol L −1 , pH = 3.0) buffer solution. After incubating with 100 µL different concentrations of H2O2 at 37 °C for 40 min, 50 µL of the above solution was added to 250 µL Au NS, and the reaction was carried out at 37 °C for 25 min. UV measurement was performed after the reaction was completed. Unless otherwise specified, all detection experiments were independently repeated three times. Statistical analysis was conducted on the data obtained from at least three independent experiments. The data of all detection methods are represented as mean ± standard deviation (SD). The recovery rate of this method is between 92.8-101.6%, and the relative standard deviation (RSD) is between 3.4-4.4% (Table 1). These results indicated the superior accuracy and potential application of this method based on the iodide-mediated etching of AuNS for detecting H2O2.

Detection of H 2 O 2 in Real Samples
To evaluate the potential applicability of this method for detecting H 2 O 2 in real samples, tap water was selected as the actual sample, and the accuracy of the method was verified by the spiked recovery method. The specific experimental steps are as follows: Take 30% H 2 O 2 and dilute it with tap water to different concentrations. Add 50 µL HRP (40 ng mL −1 ) and 100 µL NaI to Tris-HCl (10 mmol L −1 , pH = 3.0) buffer solution. After incubating with 100 µL different concentrations of H 2 O 2 at 37 • C for 40 min, 50 µL of the above solution was added to 250 µL Au NS, and the reaction was carried out at 37 • C for 25 min. UV measurement was performed after the reaction was completed. Unless otherwise specified, all detection experiments were independently repeated three times. Statistical analysis was conducted on the data obtained from at least three independent experiments. The data of all detection methods are represented as mean ± standard deviation (SD). The recovery rate of this method is between 92.8-101.6%, and the relative standard deviation (RSD) is between 3.4-4.4% (Table 1). These results indicated the superior accuracy and potential application of this method based on the iodide-mediated etching of AuNS for detecting H 2 O 2 .

Conclusions
In this work, AuNS was utilized to construct a multicolor sensor for detecting H 2 O 2 based on a rapid and sensitive surface etching of Au NS. AuNS are etched into spherical nanostructures in the presence of HRP, I − , and H 2 O 2 . The etching process also produces color changes with different concentrations of H 2 O 2 which can be used for visual detection. This method exhibits satisfactory sensitivity and selectivity. Under the conditions of an incubation time of 40 min, a reaction time of 25 min, a temperature of 37 • C, and I − concentration of 0.06 mmol L −1 , the blue shift value of the absorption peak ∆λ shows a good linear relationship with the concentration of H 2 O 2 , with a linear range of 0.667~66.67 µmol L −1 , and a detection limit of 0.44 µmol L −1 . Because of the usage of the biological enzyme of HRP in this work, it is important to control the temperature during the analysis. This shortcoming might be solved by using a nanozyme to replace HRP. Considering the significant role of H 2 O 2 in food chemistry and biochemistry, this work offered a new multicolor visual method for point-of-care testing and on-site analysis.