A Simple and Fast Colorimetric Assay for the Detection of Chloride Ions Based on Etching of Silver Triangular Nanoplates

Abstract

It is of essential importance to precisely detect chloride ions (Cl^–) for environmental monitoring and protection. However, major challenges still remain for reliable, simple, and rapid detection of Cl^–. Herein, we present a simple and rapid colorimetric method for detecting Cl^– by utilizing unmodified silver triangular nanoplates (AgNPrs) as the colorimetric agent. The detection principle is based on the Cl^–-induced etching of AgNPrs into round nanodisks, resulting in a blue shift in the absorption peak and an observable color transition from blue to yellow. A good linear relationship between Cl^– concentration and the peak blue shift of absorption spectra is achieved in the range of 140–220 μM, with a naked-eye detection limit of 96 μM. Due to its superior sensitivity and selectivity, the proposed colorimetric sensor is successfully applied for discriminating Cl^– from real water samples. Besides, the colorimetric assay is a valid method for investigating binding sites and the change of Gibbs free energy (ΔG) during the Cl^–-induced etching process of AgNPrs.

Introduction

Chloride ions (Cl^–) are common anions in natural water bodies. However, excessive Cl^– can negatively impact water quality. In drinking water, high levels of Cl^– may cause a salty taste, affecting its palatability. More importantly, elevated Cl^– concentrations indicate pollution in the water body. The discharge of domestic sewage and industrial wastewater, such as wastewater from printing and dyeing and electroplating, significantly increases Cl^– levels in water. Through the detection of Cl^–, it is possible to assess whether the water has been contaminated by Cl^–-containing wastewater, allowing for timely action to protect water resources. Therefore, detecting Cl^– in real-world water samples is of great significance.

So far, various strategies have been employed for the detection of Cl^–, including fluorescent, , ion chromatography, , electrochemistry, , and spectrophotometry. , Although these methods provide high sensitivity measurements, they require expensive equipment, tedious procedures, long analysis time, and professional training. Therefore, a reliable, simple, rapid, and accurate approach for the detection of Cl^– in real water samples is urgently required.

Colorimetric sensing has garnered significant attention because it is low-cost and easy to operate with the naked eye. , Noble metal nanoparticles have been widely employed as colorimetric agents due to the strong absorption from the visible to near-infrared region. Among them, silver triangular nanoplates (AgNPrs) display unique localized surface plasmon resonance properties with an ultrahigh extinction coefficient, theoretically enabling broader applications in colorimetric analysis. Their optical properties are tunable throughout the visible and near-IR regions of the spectrum by adjusting the aspect ratios and corner sharpness through oxidation etching reactions, which can be observed by the naked eye with no need for an additional instrument. Therefore, AgNPrs stand out as a remarkable chromogenic agent for the visual detection. Recent studies have demonstrated that colorimetric assays based on the etching of AgNPrs have been widely applied in chemsensing and biosensing. − For instance, a nonenzymatic colorimetric biosensor was developed for single-nucleotide variation identification based on the etching of AgNPrs. The colorimetric biosensor based on the etching of AgNPrs offers notable advantages in terms of its portability, cost efficiency, and operational simplicity. However, AgNPrs are commonly functionalized with biomolecules to act as colorimetric agents, a process that entails intricate experimental procedures. Additionally, the comparatively longer detection time hinders their applicability in rapid on-site analysis.

In this study, we designed a sensitive, simple, rapid, and accurate colorimetric sensor for the determination of Cl^– by utilizing unmodified AgNPrs as a colorimetric agent (Scheme ). After incubation with Cl^–, AgNPrs were etched from triangular prisms to round nanodisks. This etching process results in a blue shift in the absorption spectra peak and a corresponding color transition from blue to yellow. Based on this working principle, the colorimetric method shows a good response toward Cl^– detection. The proposed colorimetric sensor exhibited excellent selectivity toward Cl^– and was successfully utilized for the quantification of Cl^– in real water samples, which offers a new alternative strategy for Cl^– detection. Compared to other colorimetric assays, this proposed method using unmodified AgNPrs directly as the colorimetric agent, without the requirement of surface modification, simplifies the detection process. Moreover, the detection time has been shortened to 10 min, making it more suitable for on-site rapid analysis. Besides, it demonstrated that the binding sites and the change in ΔG during the etching of AgNPrs by Cl^–.

1. Mechanism of the Colorimetric Detection for Cl^–

Results and Discussion

Preparation and Characterization of AgNPrs

AgNPrs were synthesized according to the literature. AgNPrs function as colorimetric agents without the need for surface modification. The colorimetric assay utilizing unmodified AgNPrs offers several advantages including rapid detection and ease of operation. The as-prepared AgNPrs were characterized via extinction spectra and transmission electron microscopy (TEM). Extinction spectra of AgNPrs exhibited a sharp absorption maximum at approximately 695 nm (Figure S1a), which is ascribed to the dominant in-plane dipole plasmon resonance of AgNPrs. The colloidal solution of AgNPrs appeared blue due to strong absorption in the near-infrared region (inset of Figure S1a). As shown in Figure S1b in the TEM image, the AgNPrs exhibited sharp tips. Together, the above results confirmed the successful preparation of AgNPrs.

Principle of Cl^– Detection

Scheme schematically depicts the principle of the colorimetric assays. Previous literature studies have confirmed that the etching reaction is based on the interaction between the sharp tips of AgNPrs and Cl^–. , The standard reduction potential of Ag⁺/Ag⁰ is +0.78 V. In the presence of Cl^–, the standard reduction potential of Ag⁰ is decreased (E⁰ AgCl/Ag⁰ = +0.22 V). It indicates that the presence of Cl^– will reduce the reducing ability of Ag, promoting the oxidation of Ag to Ag⁺ and its rapid combination with Cl^– to form the AgCl precipitate. The vertex regions (facet [110]) of AgNPrs have a high crystal surface energy and are not completely protected by the protective agent. Thus, Cl^–, acting as the etching agent, can selectively etch the [110] facets of AgNPrs, making the triangle gradually turn into a planar disk-like.

To prove the principle for the analysis of Cl^–, the Cl^–-induced color change was systematically investigated. Consistent with expectations, when Cl^– were introduced to the AgNPrs solution, a blue shift of the absorption spectral peak occurred (Figure a), and the blue color of AgNPrs colloids was converted to the yellow color (Figure b). These results confirmed the reliability of the detection mechanism. In this work, we quantified the Cl^– by measuring the color change response before and after the etching of AgNPrs.

1.

Proof of concept investigation of the principle of the colorimetric assay. Absorption spectra (a) and corresponding photographs (b) of AgNPrs before and after addition of Cl^– (300 μM). TEM images of AgNPrs before (c) and after (d) treatment with Cl^– (300 μM). (e) Absorption spectra of the AgNPrs under different conditions: AgNPrs (black line); AgNPrs incubated with 0.12 mg/mL HRP (red line); and AgNPrs incubated with 0.12 mg/mL HRP and 300 μM Cl^– (blue line). (f) Corresponding photographs of AgNPrs at different conditions: AgNPrs incubated with 0.12 mg/mL HRP (left); AgNPrs incubated with 0.12 mg/mL HRP and 300 μM Cl^– (right).

To validate the morphological changes of AgNPrs, TEM images were acquired before and after the addition of Cl^–. As presented in Figure c,d, the morphology of AgNPrs transformed from a triangle to a circular disk after the addition of Cl^– (300 μM). The above results clearly indicated the etching of AgNPrs triggered by the target Cl^–.

To further validate the variations in the absorption spectral peak driven by the target Cl^–, horseradish peroxidase (HRP) was used as a termination agent to inhibit Cl^–-mediated etching, thereby obtaining specific morphologies. Figure e demonstrates that the absorption peak undergoes a slight red shift due to the change in the reflectivity index of the AgNPrs surface microenvironment, confirming the successful functionalization of AgNPrs with HRP. As AgNPrs were coated with HRP via unspecific physical adsorption on the silver surfaces, Cl^– failed to adhere to the surface of the AgNPrs, effectively freezing their morphologies. Consistent with expectations, the introduction of Cl^– into the AgNPrs solution did not induce a blue shift of the absorption spectral peak (Figure e). At the same time, the color of AgNPrs solution remained consistently blue throughout the observation period (Figure f). These results indicated that the variations in the absorption spectral peak were directly driven by the target Cl^–.

Optimization of Conditions

To attain an optimal detection performance, the reaction conditions were optimized. A wavelength shift of the in-plane dipole surface plasmon band of the AgNPrs was employed to determine the conditions, including the size of the AgNPrs, the etching reaction time, the pH of the AgNPrs solution, and the concentration of the AgNPrs solution in this system.

The absorption spectral peak is highly sensitive to the local electromagnetic distribution in the AgNPrs, which was regulated by the edge length and the rounded radius. Thus, the size of AgNPrs must be selected carefully (with the same initial wavelength peak) to achieve optimal detection performance. Toward this goal, AgNPrs with different initial absorption peaks were incubated with a fixed concentration of Cl^–. Their absorption spectra were recorded before and after the etching reaction (Figure a). Under a fixed Cl^– concentration, the wavelength shift was found to reach a maximum and then declined as the initial wavelength peak value increased. Thus, AgNPrs with the initial wavelength peak at 695 nm were chosen as the optimal conditions for the rest of the experiments. Subsequently, the influence of etching reaction time on the etching process was explored. To this end, the absorption spectra of the AgNPrs solution were continuously recorded at 10 min intervals after the addition of Cl^–. Figure S2 depicts time-resolved absorption spectra of the AgNPrs solution upon the addition of Cl^–. With increasing incubation time, the peak blue shift of absorption spectra increased gradually due to the etching of AgNPrs by Cl^– (Figure b). The blue shift of the absorption spectral peak reached the maximum value and then remained constant after 10 min, suggesting that the etching reaction had almost finished after incubating for 10 min. In the control, without the addition of Cl^– solution, no significant blue-shift fluctuation in the absorption spectra was observed (Figure S5), indicating the stability of the AgNPrs during the process of absorption spectra collection. For a better quantification, 10 min was chosen as the optimal incubation time. Then, the influence of media pH on the detection of Cl^– was explored. For this purpose, the pH of the AgNPrs solution is adjusted by using a buffer solution. After incubating with a fixed concentration of Cl^–, absorption spectra of AgNPrs at different pH values were recorded (Figure S3). It was found that the peak shift value of AgNPrs solution reached the maximum at pH 6.8 (Figure c). This is attributable to the formation of AgOH precipitates under highly basic conditions, which thereby inhibited the etching reaction. Under acidic conditions, H⁺ accelerated the oxidative dissolution of AgNPrs. Therefore, pH 6.8 was selected as the optimal pH value. Finally, the influence of AgNPrs concentration on the detection of Cl^– was explored. To this end, the concentration of the AgNPrs solution can be adjusted through centrifugal concentration or the addition of ultrapure water. After treatment with a fixed concentration of Cl^–, absorption spectra of AgNPrs at different concentrations were recorded (Figure S4). At the original concentration, a significant change in the peak shift of the AgNPrs solution was observed (Figure d). Therefore, the initial concentration was selected as the optimal concentration of AgNPrs.

2.

Optimization of the experimental conditions. (a) Distribution curve corresponding to the wavelength shift values of different sized AgNPrs treated with Cl^– (300 μM) for 10 min. (b) Plots of peak shift vs incubation time. (c) Peak shift of the colorimetric probe incubated with 300 μM Cl^– at different pH values. (d) Peak shift of the colorimetric probe at different concentrations incubated with 300 μM Cl^–. The error bars are the standard deviation of the mean from three measurements.

Detection of Cl^–

Under the optimized conditions, the sensitivity of the colorimetric detection was assessed based on the peak blue shift of absorption spectra upon adding different concentrations of Cl^– (0 to 300 μM). Representative images recording the Cl^– concentration-dependent etching process are shown in Figure a. As the concentration of Cl^– increased, the color of the AgNPrs colloids gradually changed from blue to mauve and ultimately to yellow. Figure b depicts typical absorption spectra under different Cl^– concentrations, where a blue shift of the absorption spectral peak was observed with increasing Cl^– concentration. Figure c shows the plot of the wavelength shift (Δλ) versus Cl^– concentration ranging from 0 to 300 μM. Notably, the peak blue shift of absorption spectra increased linearly with the Cl^– concentration in the range of 140–220 μM. The linear regression equation (Δλ = 1.45163C_(Cl−) – 147.22394) for Cl^– detection was obtained (Figure d). The corresponding correlation R ² was calculated as 0.9912. The detection limit (LOD) for Cl^– was calculated as 96 μM at a signal-to-noise ratio of 3. Thus, our proposed method provides a simple and rapid colorimetric approach for the determination of Cl^–. The lowest concentration for quantitative detection of this method was 140 μM, which was lower than the allowable maximum residue limit of Cl^– (7.05 mM) in drinkable water as per the World Health Organization (WHO). Compared with previously reported Cl^– detection methods, our proposed strategy exhibits significantly higher detection sensitivity. ,,−

3.

Cl^– detection performances of the colorimetric probe. Photographs (a) and corresponding absorption spectra (b) of the colorimetric probe at Cl^– concentrations of 0, 20, 50, 100, 120, 140, 160, 180, 200, 220, 240, 260, 280, and 300 μM. (c) Plot of the wavelength shift versus Cl^– concentration in the range of 0–300 μM. (d) The linearity of the wavelength shift versus the Cl^– concentration in the range of 100–220 μM. (e) Peak shift of the colorimetric probe incubated with 300 μM Cl^– and a single anionic ion (300 μM for H₂PO₄ ^–, HPO₄ ^2–, HCO₃ ^–, OH^–, CO₃ ^2–, HF₂ ^–, F^–, SO₄ ^–, NO₃ ^–, AgNO₃+I^–, AgNO₃+Br^–, and Mix). A photograph of the respective reaction solutions is shown in the inset. The error bars represent the standard deviation of the three measurements.

Stable and batch-to-batch reproducible signal output is a fundamental prerequisite for Cl^– detection. To evaluate the stability of the colorimetric assay, continuous absorption spectra measurements were conducted on the same batch of the colloidal solution of AgNPrs for 180 min (Figure a). The time-dependent absorption spectra remain nearly constant, confirming the excellent stability of the colorimetric assay (Figure b). To demonstrate the roles of trisodium citrate and polyvinylpyrrolidone (PVP) in the stabilization of silver nanoparticles, trisodium citrate and PVP were removed by centrifugation. After centrifugation to remove trisodium citrate and PVP, AgNPrs were resuspended in ultrapure water, and a blue shift of the absorption spectral peak was observed (Figure S6). This behavior results from the enormous sensitivity of the AgNPrs to suffer tip truncation or partial dissolution after removing trisodium citrate and PVP. To evaluate the batch-to-batch reproducibility of the colorimetric assay, five batches of the colorimetric assay under the same conditions were investigated. As shown in Figure c, all sensors exhibited a similar performance, with an interbatch RSD of 1.79% in the detection of Cl^–. This result suggests that the colorimetric assay exhibited good batch-to-batch reproducibility.

4.

Stable and batch-to-batch reproducibility of the colorimetric assay. (a) Time-resolved absorption spectra of the AgNPrs. (b) Plots of peak shift vs time. (c) Peak shift of five different batches of colorimetric probes for Cl^– sensors. The error bars are the standard deviation of the mean from the three measurements.

To verify the selectivity of this strategy, several kinds of interfering substance ions were selected for a control test, including H₂PO₄ ^–, HPO₄ ^2–, HCO₃ ^–, OH^–, CO₃ ^2–, HF₂ ^–, F^–, SO₄ ^–, NO₃ ^–, I^–, Br^–, and Mix. As expected, Cl^– triggered a noticeable color change (the inset of Figure e) and a marked enhancement of peak shift in absorption spectra (Figure S9). This can be rationally ascribed to the specific chemical interaction between Cl^– and the [110] facets of AgNPrs. Other interfering ions (H₂PO₄ ^–, HPO₄ ^2–, HCO₃ ^–, OH^–, CO₃ ^2–, HF₂ ^–, F^–, SO₄ ^–, and NO₃ ^–) induced a slight blue shift in the absorption peak (Figure e). Nevertheless, we found that Br^– or I^– induced a blue shift in absorption spectra (Figure S7a,b). To minimize such interference by Br^– and I^–, which may coexist with Cl^– in real samples, we employed AgNO₃ as a masking agent. This is because AgBr or AgI provides a greater solubility product constant for the formation of stable precipitation. AgNO₃ can preferentially combine with Br^– and I^–. As shown in Figures S7c,d and e, after the addition of AgNO₃ to form stable precipitation, absorption spectra of the AgNPrs after exposure to these solutions display negligible changes, thus illustrating that AgNO₃ shows a good masking effect. Meanwhile, the masking agent AgNO₃ solution can effectively mask Br^– and I^– without affecting the response signal of Cl^– to the sensor (Figure S8). Together, these results clearly indicated the excellent selectivity of this proposed method.

After proving the excellent selectivity capability, we investigated the capabilities of the probes for Cl^– in real-life water samples. To avoid the interference of large-molecule proteins in the detection of Cl^– in real-life water samples, ultrafiltration membrane separation experiments are operated to remove large-molecule proteins. The real-life water was pretreated by spiking with different concentrations of Cl^–, and absorption spectra of AgNPrs were recorded. The experimental results were analyzed with the Mohr method and our proposed method (Table ). The results show that the concentrations of Cl^– in tap and groundwater samples were close to the results analyzed with the Mohr method. Compared with the results of the Mohr method, this sensing system could accurately detect Cl^– in tap and groundwater samples, with relative errors ranging from 0.355 to 1.838%. Taken together, these preliminary results indicated the potential use of this protocol for practical applications.

1. Detection of Cl^– in Tap and Ground Water.

sample	number of samples	Cl– (μM) (proposed method)	Cl– (μM) (Mohr method)	relative error (%)	average relative error (%)
tap water	sample 1	639.1	634.1	0.789	0.443
	sample 2	647.3	648.5	0.185
	sample 3	622.5	620.3	0.355
groundwater	sample 1	752.9	747.3	0.749	1.24
	sample 2	766.7	775.5	1.134
	sample 3	747.4	761.4	1.838

Hill Plot and Gibbs Free Energy

The platform of the etching of AgNPrs could be utilized to explore the binding sites in the etching of AgNPrs triggered by the target Cl^–. To ascertain the number of binding sites during the etching of AgNPrs, a Hill plot analysis was performed. Toward this goal, the binding sites in the etching of AgNPrs triggered by the target Cl^– were investigated as a function of Cl^– concentration by using the peak blue shift of absorption spectra. Typical absorption spectra of the AgNPrs at different Cl^– concentrations are presented in Figure c. As illustrated in Figure a, the concentration-dependent peak blue shift of absorption spectra exhibited a sigmoidal curve, which was fitted with the Hill equation (eq S6) in Origin Software (see Supporting Information Section 6 for detailed derivation). The Hill coefficient (n) was calculated to be 4.32, indicating that Cl^– binding to the vertex areas of AgNPrs was a positively cooperative process. Specifically, the etching reaction in the vertex areas (facet [110]) of AgNPrs occurs following the binding of approximately four Cl^– ions via physical adsorption or weak coordination interaction, rather than directly forming bonds through ionic bonds. In addition, the dissociation constant (K _d) was calculated as 158.95 μM. The above results indicated the thermodynamic driving mechanism of the etching reaction, thus providing experimental evidence for the interaction between Cl^– and AgNPrs.

5.

Gibbs free energy of the etching of AgNPrs. (a) Hill plot for the concentration-dependent wavelength shift of the colorimetric probe exposed to different concentrations of Cl^– (0, 20, 50, 100, 120, 140, 160, 180, 200, 220, 240, 260, 280, and 300 μM). (b) ΔG versus the Cl^– concentration of 0, 20, 50, 100, 120, 140, 160, 180, 200, 220, 240, 260, 280, and 300 μM.

The platform of the etching of AgNPrs could also be used to investigate the relation of Gibbs free energy change (ΔG) and Cl^– concentrations in the etching of AgNPrs triggered by the target Cl^–. The standard Gibbs free energy change (ΔG ⁰) related to the etching of AgNPrs triggered by the target Cl^– was calculated at 25 °C using the following equation.

$$\mathrm{\Delta }{G}^{\mathrm{°}}=\mathit{RT}\ln K_{\mathrm{d}}$$ 1

ΔG ⁰ was calculated to be 12.98 kJ/mol. The relationship between ΔG and ΔG ⁰ was expressed by the following equation.

$$\mathrm{\Delta }G=\mathrm{\Delta }{G}^{\mathrm{°}}+\mathit{RT}\ln Q$$ 2

In addition, the relationship between the reaction quotient (Q) and the concentration of Cl^– [Cl^–] was expressed by the following equation (see Supporting Information Section 7 for detailed derivation).

$$\mathrm{Q}=\frac{K_{\mathrm{s}\mathrm{p}}}{[{\mathrm{Cl}}^{-}]}$$ 3

Combining eqs and , the direct relation of ΔG to the etching of AgNPrs triggered by the target Cl^– was calculated at 25 °C using the following equation (see Supporting Information Section 7 for detailed derivation).

$$\mathrm{\Delta }G=\mathrm{\Delta }{G}^{\mathrm{°}}+\mathit{RT}\ln \frac{K_{\mathrm{sp}}}{[{\mathrm{Cl}}^{-}]}$$ 4

The direct relationship between ΔG and the concentration of Cl^– could be fitted to eq , as shown in Figure b. It was found that the ΔG value increased linearly with increasing Cl^– concentration. The concentration of Cl^– ions indirectly influenced the ΔG of a chemical reaction under nonstandard conditions by affecting the reaction quotient (Q). The thermodynamic relationship between ΔG and concentration provides a fundamental explanation for how the sensing reaction occurs and varies with concentration, serving as the essential basis for sensors to achieve precise, sensitive, and quantitative detection.

Conclusions

In conclusion, we demonstrated a novel, simple, and rapid colorimetric detection of Cl^– by using unmodified AgNPrs as the probe. AgNPrs were etched from triangle to round by Cl^–, leading to a noticeable color change from blue to yellow. On this basis, Cl^– could be readily quantified by the naked eye without the need for an additional equipment. The proposed method exhibited ultrasensitive Cl^– detection capability in solution. Moreover, this kind of sensor exhibits excellent specificity toward Cl^–. Thus, in the real water samples assay, excellent relative errors between 0.355 and 1.838% were obtained. In addition to the common advantages of colorimetric assays, such as simplicity, high sensitivity, and low cost, this proposed method using unmodified AgNPrs offered further benefits, including rapid detection and ease of operation. Besides, it provided a valid platform for investigating the binding sites and the change in ΔG during the etching of the nanoprisms by Cl^–.

Experimental Section

Synthesis of AgNPrs

AgNPrs were prepared according to a previous report. More details can be found in Supporting Information Section 2.

Absorption Spectra of AgNPrs and Cl^– Detection

UV–vis spectrophotometry was used to record the absorption spectrum of the AgNPrs solution. For the analysis of Cl^–, absorption spectra of AgNPrs solution were collected before and after introduction of the target solutions with different Cl^– concentrations. More details can be found in Supporting Information Section 3.

Glossary

Abbreviations

Cl^–

chloride ions

AgNPrs

silver triangular nanoplates

TEM

transmission electron microscopy

HRP

horseradish peroxidase

Δλ

wavelength shift

LOD

detection limit

WHO

World Health Organization

n

Hill coefficient

K _d

dissociation constant

ΔG ⁰

standard Gibbs free energy change

ΔG

Gibbs free energy change

Q

reaction quotient

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.5c12694.

• Materials and instrumentation, synthesis of the AgNPrs, detection of Cl^–, detection of Cl^– in tap and groundwater, derivation of the Hill equation, calculation of Gibbs free energy, and supporting figures (PDF)

L.X.L., N.F., and Y.F.L. designed the project. N.F. conceived and designed the research. X.D.H. performed the experiments, generated the data, took photographs, prepared all the graphs, and wrote the initial draft of the manuscript. N.F. and Y.F.L. analyzed the data and interpreted the results. A.R.W. critiqued and revised the manuscript.

The authors declare no competing financial interest.