CuO-ZIF-8 modified electrode surface as a new electrochemical sensing platform for detection of free chlorine in aqueous solution

Abstract

This work has applied metal–organic frameworks (MOFs) with high adsorbability and catalytic activity to develop electrochemical sensors to determine free chlorine (free-Cl) concentrations in aqueous media. A zeolitic imidazolate frameworks, Zn(Hmim)₂ (ZIF-8) has been synthesized and incorporated with CuO nanosheets to decorate a glassy carbon electrode (GCE) and provide a new sensor for free-Cl determination. The as-prepared ZIF-8 and CuO-ZIF-8 composites have been characterized by FESEM, EDX, XRD, and FT-IR analyses. Cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) utilized to characterize the CuO-ZIF-8/GC modified electrode electrochemically, demonstrated the ability of the sensor to measure free-Cl concentration. Using differential pulse voltammetry (DPV) and under the optimal conditions, the prepared CuO-ZIF-8/GC modified electrode showed a linear response in the 0.25–60 ppm range with a 12 ppb detection limit (LOD) for free-Cl concentration. Finally, the fabricated sensor was applied to analyze free-Cl from actual swimming pool water samples with promising 97.5 to 103.0% recoveries.

Introduction

Disinfection with chlorine is used daily to treat drinking water, waste and swimming pool waters and effluents from chemical industries^1–4. When chlorine gas bubbles or chlorine-based compounds such as sodium hypochlorite (NaOCl), are added to water, they are both hydrolyzed to form hypochlorite ion and hypochlorous acid (ClO^– and HClO, respectively)^5,6. The sum of the dissolved chlorine gas, ClO^– and HClO in water is defined as free chlorine (free-Cl)⁷. The HClO/ClO^- ratio is sensitive to pH since the drinking water pH is in the 6.5–9.5 range. Depending on the pH value of drinking water, each of the HOCl or OCl^- species can be dominant⁸.

During water disinfection, added chlorine should be limited to residual chlorine in water within a specified range. Low chlorine concentrations lead to ineffective disinfection, while high concentrations may form trihalomethane, which can affect the respiratory tract, immune system, cardiovascular system, central nervous system and other health problems in living beings^9–13. Therefore, the recommended chlorine concentration in drinking water by WHO is below five ppm¹⁴. Hence, the residual chlorine concentration in water should be continually detected and monitored. The standard N, N-diethyl-p-phenylenediamine (DPD) colorimetric technique is typically used to measure the free chlorine concentration in water and wastewater¹⁵. Different methods, such as chromatography¹⁶, spectrophotometry¹⁷, chemiluminescence¹⁸, colorimetric¹⁹ and electrochemical methods^6,20 have already been employed to find residual chlorine in water. Among them, electrochemical methods, especially those employing electrochemical sensors, are more attractive considering their high selectivity and sensitivity, low price, no need for additional reagents and simple procedures in continuous monitoring²¹. The electrochemical measurement of the free-Cl concentration in solutions occurs through the reduction of either HOCl or OCl^–, depending on the media pH²².

The common challenge that the electrochemical method faces for determining free-Cl is overlapping the dissolved oxygen reduction potential with that of free-Cl⁶. In addition, biofouling, mineralization and instability in long-period applications are the main disadvantages of the electrochemical sensors⁷. Several electrochemical sensors have been developed to determine free chlorine in the past decade. For instance, a gold-based electrode was used by Campo et al. as the working electrode with an applied potential of + 0.15 V. They were able to overcome the overlapping problem of the reduction potentials of the dissolved oxygen and free-Cl. However, the electrode materials were expensive³. Murata et al. reported the development of a boron-doped diamond electrode to determine the concentration of free-Cl with a + 1.1 V working potential. BDD electrodes can be operated continually with no maintenance. However, the operating potential was high²³. A Prussian Blue electrode modified with surfactant has been used by Salazar et al. to detect free-Cl with a − 0.2 V working potential at pH = 5.5²¹. In addition, free-Cl concentration was measured by a polydopamine@electrochemically reduced graphene oxide-modified electrode with acceptable sensitivity while the operating potential was + 0.15 V⁶. Tomei et al. have reported another electrochemical sensor by a CB-modified printed electrode to determine the free-Cl concentration at a low applied potential of − 0.1 V, which could overcome the fouling problem²⁰. An electrochemical self-cleaned silicon sensor chip has been designed by Yin et al. to determine the chlorine concentration in tap water. A LOD of 0.056 ppm was shown by the designed sensor²⁴. A conductive silver membrane has been recently prepared for free-Cl reduction using polyethyleneimine (PEI) and polydopamine (PDA). The results showed a LOD and detection range of 0.07 ppm and 0.1–20 ppm, respectively. However, the effect of common species has not been investigated²⁵.

Metal–organic frameworks (MOFs), porous compounds with high catalytic behavior, have been used to design a novel electrochemical sensor to detect free-Cl. They are classical porous compounds widely employed in different fields of chemical separation, gas separation and storage, catalysis, sensing and membrane construction, sensing^26–32. Easy synthetic methods, high catalytic activity, tunable chemical functionalization and reversible physical adsorption with target analytes make MOFs excellent candidates in chemical sensors^33–36. In addition to their application as probes in luminescent and colorimetric sensors to detect environmental contaminants, they have shown potential ability as electrode modifiers because of their large pore volumes, high surface area, selective adsorbability, high catalytic activity and facility of chemical functionalization^37–41. The use of MOFs in electrochemical sensors is restricted by their low electrical conductivity and relative mechanical stability in aqueous media⁴². To overcome this problem, MOFs can be incorporated with other functional compounds^43,44. In this regard, metal ions and oxides as electroactive species can enhance electrocatalytic characteristics and selectivity of MOF composites. Moreover, ligands containing conjugated systems, including pyridines, aromatic acids and imidazoles or benzimidazolescan improve the transfer of electrons into the framework⁴⁵. Hence, the application of MOFs in electrochemistry relies on the design of MOF composites with exciting architectures. In this respect, researches has been performed on applying metal oxides in functionalized MOF composites in recent years to improve their electrocatalytic properties^46–48. Among the various MOFs used to modify electrodes in electrochemical sensors, ZIF-8, which is a subset of the zeolitic imidazolate frameworks (ZIFs) series, has demonstrated potential for a diversity of applications due to its thermal/chemical stability, easy crystallization and uniform pore size (3.4 Å)^49–52. Multiple nitro groups in ZIF-8 framework and other features make it a promising adsorbent for hypochlorite ions in aqueous media. Hence, in the current study, the CuO-ZIF-8 composite has been fabricated successfully and applied as the electrocatalyst for free chlorine reduction in water. Decorating ZIF-8 by CuO could improve the electrocatalytic properties of ZIF-8 and enhance its performance in the electroreduction of free-Cl.

Experimental

Reagents and materials

Chemicals in this work were analytical grades and used directly as received. Double distilled water (DDW) was applied to prepare aqueous solutions. Copper sulfate pentahydrate (CuSO₄·5H₂O), zinc nitrate hexahydrate (Zn(NO₃)₂·6H₂O), polyvinyl pyrrolidone (PVP, Mw = 58,000), 2-methylimidazole (MeIM), sodium hydroxide (NaOH), ethanol and methanol were supplied by Merck Chemical Co. Sodium hypochlorite (4.00–5.00%) was purchased from Dr. Mojallali Chemical Co. and applied to prepare free chlorine (hypochlorite (v/v)) and its concentration was spectrophotometrically determined before use⁵³. To prepare phosphate buffer solutions (PBS) with different pH values, monosodium hydrogen phosphate (NaH₂PO₄) and disodium hydrogen phosphate (Na₂HPO₄), which were obtained from Sigma-Aldrich Chemical Co., were mixed in different proper ratios and adjusted using a Metrohm 691pH/ion meter.

Synthesis of CuO-ZIF-8

CuO-ZIF-8 was synthesized using a literature procedure with some modifications⁵⁴. To prepare CuO-ZIF-8, firstly CuO nanosheets were synthesized by the following method. 334.6 mg of CuSO₄·5H₂O was dissolved in DDW (100 mL), followed by the slow addition of 5 M NaOH (1.5 mL). The obtained solution was then continuously stirred for two hours at ambient temperature. Last, the CuO nanosheet was separated by centrifugation and after washing several times with DDW, dried at 60 °C in vacuo overnight.

In a typical synthesis of CuO-ZIF-8, 0.2 mmol of the as-synthesized CuO nanosheets was first dispersed in methanol (25 mL). 1 mmol of Zn(NO₃)₂·6H₂O and 60 mg of polyvinyl pyrrolidone was added to this solution and pretreated by ultrasonic dispersion for 20 min (solution A). Next, solution B containing 800 mg of 2-methylimidazole in 25 mL of methanol was prepared and added to solution A under stirring. The final solution was stirred for one hour at ambient temperature and kept for 24 h under static conditions. The mixture was separated by centrifugation, washed thoroughly with methanol, and set in a vacuum oven to dry (overnight at 60 °C). In the next step, the dry precipitate was mixed with 25 mL of ethanol, and obtained mixture was stirred at 50 °C for 12 h to separate the residual ions and surfactant. After being centrifuged and washed with ethanol, the final product was dried at 100 °C for ten hours in vacuo. ZIF-8 nanocrystals were fabricated following the same method except adding CuO nanosheets.

Modified electrode preparation

Before any experiment, the surface of the GCE was polished with slurry alumina powder (0.05 and 0.3 mm) on the cotton cloth and then, ultrasonically cleaned in ethanol and DDW, respectively to ensure the removal of any pollution. In the next step, 5 mg of the prepared CuO-ZIF-8 and ZIF-8 powder were ultrasonically dispersed in 0.5 mL of DDW separately to form a uniform solution. 0.5 µL of each dispersion was then added dropwise to the GCE surface and dried using an infrared lamp at a distance of 15 cm to avoid damaging the surface of GCE. Finally, ZIF-8/GCE and CuO-ZIF-8/GCE were obtained for the following electrochemical studies.

Materials characterization

All the electrochemical investigations were performed by a potentiostat/galvanostat (Ivium V21508, Vertex) with a cell of three electrodes in an aqueous medium as the electrolyte. A glassy carbon electrode (GCE, 2 mm diameter), a platinum wire and an Ag/AgCl reference electrode (3.0 mol L^–1 KCl) (Azar electrode Co., Urmia, Iran) served as the working, reference, and counter electrodes, respectively. FESEM images were obtained by MIRA3 (Czech Republic) coupled with EDX. In addition, the chemical characterization of the as-prepared ZIF-8 and CuO-ZIF-8 composites was studied by FTIR spectroscopy (Nicolet 800, USA). The XRD characterization was analyzed using a PHILIPS PW3710 instrument (Netherlands) using Ni-filtered Cu Kα.

Results and discussion

Morphology and structure of ZIF-8 and CuO-ZIF-8

The FESEM images of CuO, ZIF-8, and CuO-ZIF-8 are shown in Fig. 1a–d. The CuO nanosheets exhibited a foliaceous structure with a 100–500 nm average diameter (see Fig. 1a). As depicted in Fig. 1b, ZIF-8 showed characteristic hexahedral morphology with an average particle size in the 100–200 nm range. The corresponding FESEM images of CuO-ZIF-8 composites are shown in Fig. 1c,d. As observed, CuO nanosheets have been inserted into the ZIF-8 surface and attached, resulted in a rough surface morphology with lacking edge-sharpness nanostructure. Noticeably, no aggregation of free CuO was observed in the images.

Figure 1.

FESEM images of (a) CuO, (b) ZIF-8 and (c,d) CuO-ZIF-8, (e) EDS spectrum of CuO-ZIF-8 and (f–k) EDS mapping images of CuO-ZIF-8.

EDS mapping technique was applied to explore the dispersion of CuO-ZIF-8 NPs. As observed, Fig. 1f–k shows that Cu, Zn, C, N, and O elements have been uniformly distributed within CuO-ZIF-8 nanocomposite. As shown, the Cu content was the least, which aligns with the FESEM images and the CuO attachment on the ZIF-8 nanoparticle surfaces. Furthermore, the chemical characterization of CuO-ZIF-8 NPs was performed by EDS analysis (Fig. 1e). The EDS results revealed the presence of Zn, Cu, O, N, and C elements in the CuO-ZIF-8 nanocomposite, confirming the formation of this nanocomposite. The XRD patterns of ZIF-8 and CuO-ZIF-8 are presented in Fig. 2a. All reflection peaks of the XRD pattern matched well with the crystal structure of ZIF-8 presented in the other reports^55,56. The prominent peaks at 2θ = 7.41°, 10.45°, 12.77°, 14.77°, 16.53°, 18.07°, 22.19°, 24.53°, 26.73°, 29.71°, 30.61°, and 32.43° correspond to the (011), (002), (112), (022), (013), (222), (114), (233), (134), (044), (244), and (235) related to ZIF-8 crystal planes, respectively. Similar diffraction peaks were observed in both XRD patterns of ZIF-8 and CuO-ZIF-8 composite, proving the stability of the ZIF-8 structure even after mixing with CuO. Moreover, no characteristic peaks corresponding to the bulky free CuO particles were observed in the XRD images of the CuO-ZIF-8 composite, indicating the dispersion of CuO on ZIF-8.

Figure 2.

(a) XRD patterns and (b) FT-R spectra of ZIF-8 and CuO-ZIF-8, BET isotherms and BJH (inset) of (c) ZIF-8, and (d) CuO-ZIF-8.

The FTIR spectra of the ZIF-8 and CuO-ZIF-8 composite is shown in Fig. 2b. The vibration band around 413 cm^–1 is due to Zn-N stretching vibration⁵⁷. The signal at 1143 cm^–1 corresponds to the aromatic C–N stretching mode. The intense peak in the 1310–1513 cm^–1 range is ascribed to the stretching of the entire ring. The peak at 991 cm^–1 is due to the ring in-plane bending mode. However, the peaks at around 691 and 750 cm^–1 are due to the out-of-plane bending of the 2-methylimidazole ring⁵⁸. The vibration mode of C=N appeared at 1584 cm^–1 and the bands at 3135 and 2929 cm^–1 correspond to the stretching of the aromatic and aliphatic C–H in the imidazole, respectively⁵⁷. Furthermore, the peaks related to the Cu–O stretching vibration can be observed at 618 and 510 cm^–159. In addition, the peak at 3428 cm^–1 is associated with the O–H stretching of H₂O, which could be adsorbed physically.

The surface area and pore structures of ZIF-8 and CuO-ZIF-8 composite were characterized by nitrogen adsorption–desorption isotherms and the results were exhibited in Fig. 2c,d. Based on the IUPAC classification, both the samples displayed Ι type whit hysteresis of type IV. The BET surface for ZIF-8 and CuO-ZIF-8 were obtained to be 893 and 1708.4 m²/g, respectively, indicating the big surface area of the samples specially CuO-ZIF-8 composite, which can lead to high sensing ability. From BJH plot of samples (insets of Fig. 2c,d), they are microporous and partly mesoporous.

Electrochemical characterization of CuO-ZIF-8/GCE

The CV data of the bare GC, ZIF-8/GC, and CuO-ZIF-8/GC modified electrodes in the solution of 5 mM [Fe(CN)6]^3–/4– (Fig. 3a) were used to study the electrochemical effects of the various modifiers on the electrode surface. For the bare GCE, the Epc and Epa corresponding to the reduction and oxidation peak currents are observed at + 0.16 and + 0.28 V, respectively. The intensity of the Fe⁺²/Fe⁺³ probe current has increased upon modifying the GC electrode with ZIF-8 MOF. This indicates that a faster charge transfer occurs between the solution and the electrode surface in the presence of ZIF-8 MOF. The CuO-ZIF-8/GCE showed a higher current. The high current response values of CuO-ZIF-8/GCE, in comparison with ZIF-8/GCE, resulted from the presence of CuO in the structure of MOF and thus, the increased conductivity of ZIF-8.

Figure 3.

(a) CV and (b) EIS plots of different modified electrodes obtained at a GCE, ZIF-8/GCE and CuO-ZIF-8/GCE in a 0.1 M KCl solution containing 5.0 mM K₃[Fe(CN)₆]^3-/4- at a scan rate of 0.05 V s^-1 (c) Relevant simulated equivalent circuit for fitted EIS plot of CuO-ZIF-8/GCE with the resultant data.

Furthermore, EIS technique was applied to assess the surface-modified electrode probe. For this purpose, a solution of 5 mM [Fe(CN)₆]^3–/4– in 0.1 M KCl, with a 100 kHz-0.01 Hz frequency range was used to obtain the Nyquist diagrams of bare GCE, ZIF-8/GCE, and CuO-ZIF-8/GCE. The Nyquist diagrams displayed linear and semi-circular sections at low and high frequencies, corresponding to the diffusion and electron transfer-limited processes, respectively. The diameter of the semicircle in the Nyquist plot indicates the resistance of charge transfer (R_ct), reflects the kinetics of the electron transfer on the modified electrode surface and affects the electrode sensitivity⁶⁰. As observed in Fig. 3b, the impregnation of the GCE surface with ZIF-8 decreased the charge transfer resistance (R_ct). The R_ct of the bare GCE and ZIF-8/GCE were 7217 and 4182 Ω, respectively. By stabilizing the CuO-ZIF-8 on the GC electrode surface, the diameter of semicircles and charge transfer resistance decreased to lower frequencies (R_ct 2927 Ω). These results indicated the faster charge transfer on CuO-ZIF-8/GCE compared to ZIF-8/GCE and bare GCE, confirming the role of CuO as an electron mediator, which enhances the electrocatalytic activity of CuO-ZIF-8 composite. The equivalent circuit compatible with the Nyquist diagram recorded in the presence of Fe⁺²/Fe⁺³ was depicted in Fig. 3c. To obtain a satisfactory impedance simulation, it is necessary to replace the capacitor C with a constant phase element (Q) in the equivalent circuit. The most widely accepted explanation for the presence of constant phase element behavior and depressed semicircles on solid electrodes is microscopic roughness, causing an inhomogeneous distribution in the solution resistance as well as in the double-layer capacitance. R₂ and Q₂ are related to Fe⁺²/Fe⁺³ redox couple.

Electrochemical behavior of free-Cl at CuO-ZIF-8/GCE

The electrochemical properties of free-Cl at the modified electrodes (bare GCE, ZIF-8/GCE, and CuO-ZIF-8/GCE) were studied in 0.2 M PB solution and pH 6.0 (Fig. 4a). Also, for comparison, the CVs of the mentioned electrode in the absence of free-Cl were shown in the inset curve of Fig. 4a. As observed, in the presence of 4 ppm free-Cl, there are no clear cathodic or anodic peaks on the CV curves of bare GC. The ZIF-8/GC modified electrode showed an increase in the background current to the negative area, which means the ZIF-8/GCE could sense the free-Cl. However, a small cathodic peak were observed. The CuO-ZIF-8/GC modified electrode showed a clear cathodic peak at − 0.1 V with increased free-Cl reduction peak current. This demonstrated the better electrocatalytic activity and performance of the CuO-ZIF-8 composite in free-Cl detection compared to bare GCE and ZIF-8/GCE due to the extensive surface area and outstanding adsorption properties of ZIF-8 besides good conductivity of CuO. Since the HClO/ClO^- ratio is pH-dependent, HClO is the main species at low pH (pH < 7.5)⁸. Thus, pH 6.0 was used in this experiment to ensure a higher concentration of HClO since the presence of dissolved oxygen can affect the hypochlorite reduction reaction⁶¹. On the other hand, ZIF-8 is chemically unstable under acidic conditions⁶². Hence, pH was adjusted close to the neutral condition. Thus, depending on pH value in the experiment, the observed reduction peak can be related to the reduction of HClO, according to follows equation^6,32,61:

$$\text{HClO}+{\text{2e}}^{-}\to {\text{Cl}}^{-}+{\text{OH}}^{-},$$ 1

Figure 4.

(a) CVs of bare GCE, ZIF-8/GCE and CuO-ZIF-8/GCE in the presence of 4 ppm of free-Cl, Inset: CVs of bare GCE, ZIF-8/GCE and CuO-ZIF-8/GCE in the absence of free-Cl (b) CVs of CuO-ZIF-8/GCE at different concentration of free-Cl in 0.2 M PBS (pH 6.0), scan rate: 0.05 V s^–1. Inset: extensive CV of CuO-ZIF-8/GCE in the presence of 2 ppm of free-Cl.

Figure 4b displays the CVs of CuO-ZIF-8/GCE in the presence and absence of different free-Cl concentrations (0.0–2.0 ppm). The CVs show no specific cathodic peaks corresponding to chlorine without free-Cl. After adding 0.5 ppm of free-Cl, a weak cathodic peak appears around − 0.1 V. The observed current of the cathodic peak systematically enhances with the increase of free-Cl concentration. This confirms that the CuO-ZIF-8/GCE provides excellent sensitivity to free-Cl. Furthermore, the extensive CV of the CuO-ZIF-8/GCE in the presence of 2.0 ppm free-Cl from − 1 to 0.6 V is observed in the inset curve of Fig. 4b. As observed, the reduction peak of the dissolved oxygen is wholly separated from that of the free-Cl. A reduction peak related to O₂ is observed in all the plots in Fig. 4b and the peak corresponding to 2.0 ppm free-Cl is shown as an example.

The chronoamperometry technique was applied to investigate the electrochemical kinetic behavior of the CuO-ZIF-8/GC electrode towards the free-Cl reduction. Figure 5a shows the current curve versus time for CuO-ZIF-8/GCE at a − 0.1 V constant potential (vs. Ag/AgCl) in different ranges of free-Cl concentration from 0 to 150 ppm. The results revealed the enhancement of the steady-state current as free-Cl concentration increased. By plotting the calibration curve for the data obtained from Fig. 5a, a linear relationship with coefficient of 0.985 was obtained. This result confirmed the ability of the designed sensor to measure the concentration of free-Cl in a wide range. Also, the free chlorine diffusion coefficient (D) was estimated using the data obtained from chronoamperograms (slope of the I plots versus t^−1/2 that shown in Fig. 5(b)) and based on the Cottrell Eq. (2)⁶³.

$$\text{I}={\text{nFACD}}^{1/2}{\text{(}\pi \text{t)}}^{-1/2}.$$ 2

Figure 5.

(a) Chronoamperograms of CuO-ZIF-8/GCE in the presence of various free-Cl concentrations (0, 2, 5, 10, 20, 60, 100 and 150 ppm) in a 0.2 mol L^–1 PBS recorded at – 0.1 V, (b) plot of current vs. t^−1/2 based on data from chronoamperograms.

That n, F, A, and C are the number of the transferred electrons, Faraday constant, active area of the electrode (obtained via Randel’s equation using the data from Fig. 3a), and free-Cl concentration, respectively. According to reaction (1), the number of electrons transferred is 2 for HClO and a value of 0.0289 cm² for A was obtained. The D value for free-Cl was 4.41 × 10^–6 cm² s^–1.

DPV technique was conducted to find free-Cl concentration on the CuO-ZIF-8/GC modified electrode, given its better resolution and higher sensitivity for quantitative analysis than the CV technique. Figure 6a shows that plotting free-Cl concentration vs. the corresponding peak currents in the 0.25–60 ppm range provided a straight line. The calibration plot presented the linear concentration ranges from 0.25 to 60 ppm with R² = 0.9932 (Fig. 6b). The estimated LOD for free-Cl at an S/N ratio of 3 was 12 ppb. The results were comparable to the analytical data reported by other electrochemical sensors for free-Cl measuring (see Table 1).

Figure 6.

(a) DPV response for CuO-ZIF-8/GCE (at 0.025 Vs^–1) in a 0.2 mol L^–1 PBS (pH 6.0) containing various concentration of 0.25, 0.5, 0.75, 1.25, 1.5, 2.5, 3, 4, 10, 15, 20, 25, 45 and 60 ppm of free-Cl. (b) calibration curves obtained for peak current vs. free-Cl concentration.

Table 1.

Comparison of the sensing ability for free-Cl with different electrochemical sensors.

Electrode	Reaction	pH	Applied potential (V)	Limit of detection (ppm)	Linear range (ppm)	Stability (days)	Refs.
Graphite electrode	Oxidation	8	1.4	1.0	1.0–300.0	N/A	64
Au film microelectrode	Reduction	5.5	0.35	0.02	0.02–5	> 10	65
Polymelamine-SPCE	EC	7	0.125	0.4	0.4–521	> 7	66
Boron-doped diamond electrode	Oxidation	8	1.4	0.016	0.1–100	> 3 months	23
CuO-NPs/MWCNT/epoxy	Reduction	5.5	− 0.35	0.0006	0.056–450	> 7	60
SPCE/PB-BZT	Reduction	5.2	− 0.2	0.0825	0.008–3	> 3 months	67
PDA@ERGO/GCE	EC	7.2	0.15	0.0032	0.736–16	> 28	6
CB-SPE	Reduction	5	− 0.1	0.01	0.05–200	N/A	20
PPMM@PDA/PEI@Ag	Reduction	N/A	− 0.3	0.07	0.1–20	N/A	25
silicon-BDD/Pt based electrode	Reduction	8.5	− 0.35	0.056	5–200	> 50 days	24
CuO-ZIF-8/GCE	Reduction	6	− 0.1	0.012	0.25–60	> 1 month	This work

N/A not available.

Figure 7 exhibits the Nyquist diagrams of CuOZIF-8/GC electrode registered at reduction peak potential for different amounts of free-Cl. The increase in free-Cl concentrations reduced the diameters of semicircles and appeared at the low frequency limit, indicating the decreased charge transfer resistance at higher amounts of chlorine.

Figure 7.

EIS plots of the CuO-ZIF-8/GC electrode with varying concentrations of free-Cl at pH 6.0.

Interference, reproducibility, and Stability studies of CuO-ZIF-8/GCE

The ability of the CuO-ZIF-8/GC electrode to determine the concentration of free-Cl in the presence of such interfering ions as SO₄^2–, SO₃^2–, NO₃^–, NO₂^–, HCO₃^–, and CO₃^2– was investigated. Selected ions are typically present in water or may be existed after the purification process. Different species were prepared individually, and each was added to 1.5 ppm of free-Cl supported by the electrolyte. After adding 50 µL of the ions with the final concentration of 2.5 ppm to the buffer solution, the peak current of free-Cl slowly decreased in the presence of SO₄^2–, SO₃^2–, and NO₂^– due to a competitive adsorption of these ions to the CuO-ZIF-8 modified active sites whereas the interference responses for HCO₃^–, CO₃^2– and NO₃^– were negative. The peak current of free-Cl was slowly increased (as observed in Fig. 8). Furthermore, the reproducibility of the CuO-ZIF-8/GC modified electrode was evaluated by comparing the differential pulse voltammetry measurements of 8 ppm free-Cl using the three various electrodes. Based on the obtained results, the relative standard deviation (RSD) was 3.9%.

Figure 8.

Effects of various interference ions on the free-Cl signals in a 0.2 mol L^–1 PBS containing a concentration of 1.5 ppm of free-Cl (pH 6.0).

Additionally, the repeatability of the CuO-ZIF-8/GCE was investigated by differential pulse voltammetry with an RSD of 3.31%. This RSD value exhibits satisfactory reproducibility and repeatability for the modified electrode. Finally, after storing the CuO-ZIF-8/GC modified electrode for 30 days, the electrode response reduced to 93% of its initial current, indicating the excellent stability of CuO-ZIF-8/GCE.

Free chlorine measurement in real sample

The potential application of the prepared CuO-ZIF-8/GC electrode in real situation was investigated to determine the free-Cl concentration in actual swimming pool water samples by standard addition method. Various chorine standard solutions were prepared in PBS, and the current response was measured after adding 10 µL spikes of swimming pool water. As shown in Table 2, the free-Cl concentrations were properly evaluated at 12.7 and 7.9 in the actual samples with acceptable recovery of 103.2% and 97.5%, respectively.

Table 2.

Measurement of free-Cl concentration in the samples of swimming pool water using the CuO-ZIF-8/GC sensor.

Sample	Added (ppm)	Found (ppm)	Recovery (%)	RSD (%) (n = 3)
Swimming pool-1	12.3	12.7	103.2	4.2
Swimming pool-2	8.1	7.9	97.5	2.8

Conclusions

An efficient CuO-ZIF-8/GC modified electrode-based electrochemical sensor has been designed to determine the free-Cl concentration in aqueous media. Compared to bare GCE, the ZIF-8/GC modified electrode showed better sensing ability considering its high surface area, selective adsorbability and catalytic activity. Introducing CuO nanosheets into the ZIF-8 structure promoted the electrocatalytic reduction of free-Cl. Modified electrodes presented superior analytical properties in measuring free-Cl concentration, such as a LOD of 12 ppb, broad concentration ranges of 0.25 to 60 ppm, acceptable repeatability, reproducibility, and good stability. Moreover, this sensor could be satisfactorily applied to determine the free-Cl concentration in swimming pool water samples directly.

Author contributions

Somayeh Jafari, Seied Mahdi Pourmortazavi, Ali Ehsanib, Somayeh Mirsadeghi wrote the main manuscript text and prepared all figures. All authors reviewed the manuscript.

Data availability

The authors declare that all data generated or analysed during this study are included in this published article.

Competing interests

The authors declare no competing interests.