Electrochemical System for Field Control of Hg2+ Concentration in Wastewater Samples

The paper presents the validation of an electrochemical procedure for on-site Hg2+ ions determination in wastewater samples using a modified carbon screen-printed electrode (SPE) with a complexing polymeric film based on poly(2,2′-(ethane-1,2-diylbis((2-(azulen-2-ylamino)-2-oxoethyl)azanediyl))diacetic acid) (polyL). Using metal ions accumulation in an open circuit followed by anodic stripping voltammetry, the SPE-polyL electrode presents a linear range in the range of 20 µg/L to 150 µg/L, with a limit of detection (LOD) = 6 µg/L, limit of quantification (LOQ) = 20 µg/L, and an average measurement uncertainty of 26% of mercury ions. The results obtained in situ and in the laboratory using the SPE-polyL modified electrode were compared with those obtained by the atomic absorption spectrometry coupled with the cold vapor generation standardized method, with the average values indicating excellent recovery yields.


Introduction
The toxicity of heavy metals imposes increasing concerns because of the effects they produce on human health and the ecological system. Mercury is one of the most toxic metals and at the same time one of the most widespread pollutants in nature, coming from natural and anthropogenic sources [1]. It is an element that does not biodegrade in the environment and is well known for its accumulation in the aquatic ecosystem. This element from the atmosphere contaminates waters (rivers, lakes, oceans) and soils. In the aquatic environment, inorganic mercury is transformed into organic mercury (methylmercury-the most toxic form). Populations that have a diet based especially on seafood or that consumes certain species of fish (king mackerel, shark, tuna, etc.) also have an increased risk of methylmercury poisoning.
Mercury (Hg 2+ ) and its compounds are extremely toxic to the environment. Due to its accumulation in food, it is extremely harmful even at very low amounts. In addition, due to its strong absorption into biological tissues, it exhibits a very slow rate of elimination. Through this bioaccumulation process, mercury easily passes from the aquatic system into the food chain. Exposure and consumption of this element, even at very low concentrations, can cause neurological and movement disorders, kidney failure, cancer, liver dysfunction, and disorders of the endocrine systems [2]. According to the Environmental Protection Agency (EPA), the maximum allowed limit for Hg 2+ is less than 10 nM [3]; therefore, monitoring environmental contamination with Hg 2+ is of great interest.
Recently, nanoparticles have been used to remediate soils and waters contaminated with metals. These nanoparticles are used due to their large surface area, high adsorption capacity, fast diffusion speed, and significant chemical reactivity [33][34][35]. In order to prevent environmental pollution with mercury, strategies based on nanomaterials have been developed using graphene oxide, carbon nanotubes, or carboxymethyl cellulose or chitosan as adsorbent material to remove mercury from polluted water [36].
Electroanalytical methods are an alternative solution to the traditional methods of mercury detection in water samples. Despite the fact that numerous studies have been published based on the use of disposable SPEs for monitoring metal contaminants in the environment [37,38], there are only a few studies that employed SPEs to detect mercury levels in the environment [39]. For instance, Joe Wang was the first to employ a SPE gold-based electrode for potentiometric stripping measurements of trace mercury levels, achieving a detection limit of 0.5 µg/L [40]. In order to detect mercury, Bernalte et al. in 2011 employed a commercially available carbon-based SPE electrode enhanced with gold nanoparticles, obtaining a LOD of 1.1 µg/L for testing on samples of water and dust using square-wave anodic stripping voltammetry [41][42][43]. In addition, electrodeposited gold on a SPE was reported by Mandil et al. for mercury detection, achieving a LOD of 1.5 µg/L [44].
This study aimed to validate the electrochemical method for the determination of mercury from wastewater samples using a modified SPE with complexing polymeric film based on polyL (the structure of the monomer is presented in Figure 1.). In addition, the present work intends to make a step further into the development of an electrochemical portable device, with low running costs and easy-to-operate system based on a modified SPE for mercury detection. of native enzymes, which is why they were applied in the detection of mercury based on color variation [32]. Recently, nanoparticles have been used to remediate soils and waters contaminated with metals. These nanoparticles are used due to their large surface area, high adsorption capacity, fast diffusion speed, and significant chemical reactivity [33][34][35]. In order to prevent environmental pollution with mercury, strategies based on nanomaterials have been developed using graphene oxide, carbon nanotubes, or carboxymethyl cellulose or chitosan as adsorbent material to remove mercury from polluted water [36].
Electroanalytical methods are an alternative solution to the traditional methods of mercury detection in water samples. Despite the fact that numerous studies have been published based on the use of disposable SPEs for monitoring metal contaminants in the environment [37,38], there are only a few studies that employed SPEs to detect mercury levels in the environment [39]. For instance, Joe Wang was the first to employ a SPE goldbased electrode for potentiometric stripping measurements of trace mercury levels, achieving a detection limit of 0.5 µg/L [40]. In order to detect mercury, Bernalte et al. in 2011 employed a commercially available carbon-based SPE electrode enhanced with gold nanoparticles, obtaining a LOD of 1.1 µg/L for testing on samples of water and dust using square-wave anodic stripping voltammetry [41][42][43]. In addition, electrodeposited gold on a SPE was reported by Mandil et al. for mercury detection, achieving a LOD of 1.5 µg/L [44].
This study aimed to validate the electrochemical method for the determination of mercury from wastewater samples using a modified SPE with complexing polymeric film based on polyL (the structure of the monomer is presented in Figure 1.). In addition, the present work intends to make a step further into the development of an electrochemical portable device, with low running costs and easy-to-operate system based on a modified SPE for mercury detection.

Equipment
The SPEs were purchased from BVT Technologies (Strazek, Czech Republic). The characteristics of these electrodes are: carbon working electrode (2 mm diameter); auxiliary carbon electrode; Ag/AgCl reference electrode. The dimensions of these electrodes were mass: 0.5 g; length: 25.40 mm; width: 7.26 mm; thickness: 0.63 mm.
PalmSens4 portable potentiostat connected to a laptop and equipped with PSTrace software, capable of processing all the information provided by the potentiostat from the electrochemical cell (EC) and rendering it in the form of graphics on a computer screen. The connection between the EC and the potentiostat was made with the help of a connector from Metrohm DropSens (reference CAC4MMH).
A Thermo M6 spectrometer equipped with cold vapors and a Hg cavity cathode lamp was used to perform the atomic absorption spectrometry with cold vapor generation (AAS-CV) standardized.
A Millipore Direct-Q water purification system from Merck was used a source of purified water.

Electrochemical Procedures for Hg 2+ Analysis
The SPE modified electrode was prepared in supporting electrolyte (0.1 M) obtained from acetonitrile and TBAP containing 1.5 mM L using the previously described procedure [45]. An example of successive cyclic voltammograms of L on SPE is presented in Figure S1 (see Supplementary Information), which emphasize an accumulation of polymeric material on the electrode surface during electropolymerization. After its transfer into free monomer electrolyte ( Figure   This four-stage procedure was applied for the validation of the proposed method for Hg 2+ detection ( Figure S3 from Supplementary Information), using at this level only standard solutions from mercuric acetate.
To validate the electrochemical method, the following performance parameters of the method were verified: linearity and working range, LOD and LOQ, accuracy, repeatability, intermediate precision, recovery, robustness, and measurement uncertainty.

Alternative Method for Control of Hg 2+ in Wastewater Samples
An alternative method used for quality control of the Hg 2+ ions in wastewater samples, in order to validate the electrochemical procedures applied in the study, was the EN ISO 12846:2012 standardized method [46]. The technique of this standardized method is AAS-CV.

Wastewater Sampling
Seventeen wastewater samples were collected from an industrial site in order to determine mercury. Duplicate samples were preserved with KBrO 3 -KBr in hydrochloric acid and transported to the analytical laboratory for mercury determination with the AAS-CV technique. In addition, electrochemical procedures were applied in the laboratory using PalmSens4 portable potentiostat equipment. On-site, the electrochemical procedures were performed and the concentration of Hg 2+ ions was determined using the same device.

Linearity and Working Range
To establish the linearity of the method, the calibration curve was performed on six points in the concentration range from 20 to 150 µg/L Hg 2+ (20 µg/L; 50 µg/L; 60 µg/L; 80 µg/L; 100 µg/L; 150 µg/L). For each concentration level, three replicates were analyzed; statistical parameters were calculated using the average of these three replicates. Good correlation coefficient (R = 0.9996) and determination coefficient (R 2 = 0.9993) values were obtained, with the values being close to 1.
The linearity of the regression curve calculated with Equation (1) indicated a value of 99.07%, which falls within the range of ±1% [47]. The value of the method coefficient of variation (1.23%) was less than 2% according to Horwitz function, situating this approach as a good electrochemical method [48].
(1) Figure 3 presents the calibration curve that was used to establish the performance parameters of the method.

Accuracy
To determine the accuracy of the method, ten individual replicates were performed at five concentration levels (20 µg/L Hg 2+ , 50 µg/L Hg 2+ , 80 µg/L Hg 2+ , 100 µg/L Hg 2+ , 150 µg/L Hg 2+ ). The replicates were carried out by a single analyst,in the same day using the same equipment and the same working conditions and method. The obtained results are presented in Table 1.

Accuracy
To determine the accuracy of the method, ten individual replicates were performed at five concentration levels (20 µg/L Hg 2+ , 50 µg/L Hg 2+ , 80 µg/L Hg 2+ , 100 µg/L Hg 2+ , 150 µg/L Hg 2+ ). The replicates were carried out by a single analyst, in the same day using the same equipment and the same working conditions and method. The obtained results are presented in Table 1.

Repeatability
The repeatability of the method was carried out on ten individual replicates having the concentrations 20 µg/L Hg 2+ , 50 µg/L Hg 2+ , 80 µg/L Hg 2+ , 100 µg/L Hg 2+ , and 150 µg/L Hg 2+ . The replicates were realized by a single analyst on the same day using the same equipment and the same work method. The obtained results (standard deviation of the repeatability, repeatability, and the repeatability variation coefficient) are presented in Table 2. The repeatability and repeatability variation coefficient are presented in the following equations: *Repeatability (r) = 2.8 × S r, **RSDr = (S r × 100)/X average (5)

Intermediate Precision
To determine the intermediate precision (R), twelve replicates were analyzed at three concentration levels (20 µg/L, 50 µg/L, and 100 µg/L Hg 2+ ). The tests were carried out by two analysts in three days using the same equipment and the same working condition and method. The obtained results (average concentration, standard deviation of the intermediate precision (S R ), intermediate precision, and standard deviation of the intermediate precision RSD R ) are presented in Table 3.

Interference
The selectivity of the polyL material was checked and discussed in our previous paper [45]. In addition, we can say that, as can be seen from Figure S4 (see Supplementary Information), the mercury ions stripping peak current remained at 90% from the initial height even when five mass equivalents of interference ions I (I: (Zn(II), Cd(II), Pb(II), Ni(II), Co(III), and Cu(II)) were added.

Recovery
The recovery test indicates, according to the literature, "the fraction of the analyte added to the test sample (fortified or spiked) prior to analysis which is measured by the method" [48].
Recovery percentages were calculated after analyzing six replicates of standard solutions of 50 µg/L and 100 µg/L. Same standard solutions was added to real samples without Hg 2+ (spike matrix sample) and six replicates with same standard solutions added to the blanks (spike blank) with the following equation: Recovery test data indicated recovery percentage situated in the range of 98.2% ± 5% (100 µg/L) to 110% ± 7% (50 µg/L).

Robustness
The robustness tests were performed using the Youden and Steiner partial factorial model [49,50]; a model in which three factors can be varied and analyzed with only four experiments. This efficient method can be used for "in-house" validation of the methods. Depending on the procedure used, a high value (+) and a low value (−) are established for each factor, varying by a maximum of 10% the normal value of the factor. The partial factorial model for three factors is presented in Table 4. Table 4. The partial factorial model for testing the robustness of the analytical method.

Experiment
Factor The absolute effect (bias) of each factor from A to C can be evaluated with the following formula: where: ∑ Y A+ = sum of Y results, where factor A has the high value (+), Y 1 + Y 3 , n = 2; ∑ Y A− = sum of Y results, where the factor A has the low value (−), Y 2 + Y 4 , n = 2. The effect of a factor can be considered significant if it exceeds the value of 1.4 multiplied by S R , where S R is the standard deviation of the intermediate precision value, according to Equation (10): This formula could be used in the case of the new method's "in-house" validation. S R could be replaced with other value, such as standard deviation of the original procedure taken from the last control test.
The relative effects of each factor can be estimated using the following formula: The factors selected for testing the robustness of the electrochemical method for the determination of Hg 2+ using DPV technique with the SPE-polyL modified electrodes were the parameters that influence the retention of Hg 2+ ions on the surface of SPE-polyL electrodes during DPV procedures: pH (pH = 3), reaction time (20 min), acetate buffer concentration (0.05 M).
Each experiment was performed in six replicates so that the Y i value of each experiment represented their average value. The tested concentration was 20 µg/L. Two robustness tests were performed: one in which all three parameters were modified by 10% compared to the normal value, and the other in which two parameters were modified by 10% (reaction time and acetate buffer concentration), and one was modified by 3.5% (pH value). Table 5 presents the values of the factors for each experiment.  Table 6 presents the values of Hg 2+ concentration (µg/L) for both tests; values obtained when the pH, reaction time, and acetate buffer concentration were modified according to the data for A, B, and C factors (Table 5).  Table 7 presents the data obtained in test 1 to verify the robustness of the electrochemical method. The ΣY A+ represents the sum of average values from T1-Y 1 and T1- . Values for absolute effect and relative effect for each factor were calculated using Equations (9) and (11). The comparison value used, according to Equation (11), is 10.5 µg/L, with S R for 20 µg/L being 7.52 µg/L (Table 3). Table 8 presents the data obtained in test 2 to verify the robustness of the method. The values of the factors and their influence on the method were calculated in the same way as in test 1.

Uncertainty Budget
In the estimation of the measurement uncertainty, data from the quality certificates of the reagents, the analytical balance, graduated flasks, and pipettes used in the entire procedures, as well as data from the precision and recovery tests, were used. Table 9 presents the values of uncertainty at three different concentrations.

Electrochemical Analysis of Hg 2+ in Wastewater Samples
To verify the functionality of the electrochemical procedure with SPE modified with polyL films, mercury concentration from wastewater samples was determined in the laboratory and on-site using the portable potentiostat PalmSens4 connected to a laptop and equipped with PSTrace software. The obtained results were compared with those from the AAS-CV standardized method. Table 10 presents the values of Hg 2+ concentrations in seventeen wastewater samples.

Discussion
Usually, electrochemical studies report only few descriptive statistics and metrics, such as the standard deviation of the results, and thus complex validations of the proposed methods are not carried out, both in the case of electrochemical methods used to determine metals and pharmaceutical compounds.
In our previous studies, we introduced statistics and metrics other than standard deviation in the experimental tests, such as repeatability, intermediate precision, and measurement uncertainty [51,52].
In the present study, full validation test for a new method was applied in order to offer a solution for Hg 2+ ion monitoring in wastewater to the analytical laboratories that do not have expensive equipment and specialized staff for quality control of water.
Thus, from the linearity test it was observed that the variation coefficient of the method is 1.23%, a very good value according to the literature [47]. Regarding repeatability (r) and intermediate precision (R) of the analytical methods, the literature data indicate that at the level from LOQ to 2LOQ, repeatability must be less than 25% and intermediate precision around 30% [47]. Obtained values reported in Tables 2 and 3 indicated that the r and R values fell within the accepted range. For concentration from 2LOQ to 10 LOQ, r must be lower than 15% and R must be lower than 20% [47]. Repeatability and intermediate precision expressed in terms of percentage for the concentrations between 50 µg/L and 150 µg/L (Tables 2 and 3) were situated below the admissible values.
Regarding recovery percentage, the literature indicates a recovery range between 80% and 110% for the concentration of 100 µg/L [47]: values in which the recovery tests fall.
From the data obtained in the robustness tests, it can be seen that the pH and the acetate buffer concentration do not significantly influence the electrochemical determination of Hg 2+ using the DPV technique with the SPE-polyL modified electrodes. The absolute effects of factors A (pH) and C (acetate buffer concentration) in both tests had lower values than the comparison value of 10.5 µg/L (2.85 µg/L in test 1, 4.84 µg/L in test 2). In terms of relative effect, variation with ± 10% of factors A and C values indicates an influence from 15.4% (test 1) to 31.2% (test 2). The pH value situated in the range 2.7 to 3.3 pH units and acetate buffer concentration from 0.045 M to 0.05 M does not influence the electrochemical procedure.
Instead, in both tests, the absolute effect of factor B (reaction time) had a higher value than the comparison one (14.0 µg/L in test 1, 13.5 µg/L in test 2). In addition, a higher value of relative effect was obtained: 107% (test 1) to 210% (test 2).
The reaction time between the SPE-polyL modified electrode and the solution with Hg 2+ ions is crucial for the method. The 10% variation of the reaction time indicates a major change; the values obtained in both tests (Tables 7 and 8) show that the change of this factor determines a lack of the method robustness.
The "in-house" validated method for the determination of Hg using modified SPE with complexing polymeric film based on polyL leads to precise and accurate results, with a measurement uncertainty of 26% at 50 µg/L Hg 2+ , which represents the maximum allowed limit for wastewater originating from treatment plants and discharged into natural receptors. For these reasons, the method is suitable for controlling and monitoring on-site of Hg 2+ ions in wastewater samples.
The data presented in Table 10 regarding Hg 2+ concentration in wastewater samples show comparable results both in the field and in the laboratory using the electrochemical procedure with SPE-polyL modified electrodes. In addition, the electrochemical results (EC lab, EC on-site) were in the same range with those obtained from the AAS-CV technique.

Conclusions
The validated method allows the determination of the Hg 2+ concentration directly at the pollution source using a method with low complexity and affordable equipment.
The proposed procedure for the determination of Hg 2+ using the SPE-polyL modified electrode has a wide linear range, low LOD and LOQ values, suitable repeatability, intermediate precision, and uncertainty values. According to the robustness tests, pH and acetate buffer concentration does not influence the method in the range studied (+10% variation), while reaction time has a strong effect in the procedure. Hence, the last parameter should be carefully considered to obtain adequate results. The results obtained in situ and in the laboratory were compared with those obtained using the standardized method, with the average values indicating very good recovery yields and the Hg 2+ concentrations being both in the linear range of the calibration curve and below the quantification limit. The electrochemical procedure for the determination of Hg 2+ using the SPE-polyL electrode is suitable for field monitoring of wastewater.

Patents
Invention patent application OSIM (Romanian State Office for Patents and Brands) no. A00464/29.07.2022. A method for obtaining a screen-printed electrode modified with polymeric films and an electrochemical procedure for field determination of mercury concentration in wastewater.

Supplementary Materials:
The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/s23031084/s1, Figure S1: Succesive cyclic voltamograms recorded in 1.5 mM L in 0.1M TBAP, acetonitrile solution at carbon SPE (0.1 V/s); Figure S2: Succesive cyclic voltamograms recorded in 0.1 M TBAP, acetonitrile solution at SPE-polyL modified electrode (0.1 V/s); Figure S3: DPV stripping curves recorded on SPE electrode modified with polyL before and after mercury complexation; the accumulation was performed in acetate buffer at pH 3.0 containing 10 -7 M Hg(II) for a time of 20 min; and the reduction was performed at −1.3 V for 15 s; Figure S4