Introduction

Heavy metal ions (Hg2+, Pb2+, Cd2+ and As3+) are hazardous substances in the environmental, pharmaceutical and biomedical analysis [1, 2]. Among them, the accumulation of Hg2+ in the human body which damages the central nervous system, kidney, liver and some important cell functions is inactivated which leads to a wide variety of disease. Heavy metal ions are used in industries for manufacturing batteries, thermometers, cathodic tubes, mercury vapour lamps and other electronic productions [3, 4]. The waste from the above industries is source of metal contamination, and hence, it will be beneficial if selective method is available for determination of mercury in the environment.

Schiff bases are usually made by the reaction of an amine and aldehyde group, and sometimes it possesses the fluorescence properties with biological activity [5]. The Schiff base compounds are very reactive towards metal ions and are considered as excellent chelating agents. 1,2,4,5-tetrasalicylideneaminobenzene (TSAB) is a ligand prepared by condensation of salicylaldehyde and 3, 3′-diaminobenzidine. This ligand is a symmetrical molecule and has chiral properties.

The advantage of multi-walled carbon nanotubes (MWCNTs) has received a great attention due to its good stability on the electrode, chemical properties and wide potential range [6,7,8]. In electroanalytical methods, the presence of nanomaterials on the surface of the electrode has enhanced the sensitivity for determination of analytes in stripping analysis [9,10,11,12]. Numerous spectroscopy methods were used for trace analysis of mercury such as atomic absorption spectroscopy, inductively coupled plasma–atomic emission spectroscopy (ICP-AES) and inductively coupled plasma–mass spectrometry (ICP-MS) [13,14,15,16,17]. Unfortunately, these spectroscopy methods are more expensive, high operating cost, high labour and analytical cost [18, 19]. On the other hand, electroanalytical techniques are the cheapest and highly sensitive methods for determination of mercury. Therefore, electroanalytical techniques including stripping voltammetry are relatively widespread. The major advantages of anodic stripping voltammetry includes simplicity of the equipment, inexpensiveness, less electrical power consumption, more sensitivity and low detection limit. One of the disadvantages with the anodic stripping voltammetry for the metal ion determination with mercury electrode is the toxicity associated with mercury to the human system [20,21,22] and mercury-free electrodes are the need of the hour.

In this study, we have demonstrated the use of MWCNT/octadentate TSAB ligand to modify the graphite electrode for the electrochemical determination of Hg2+. The synthesised TSAB is an octadentate Schiff base ligand and is shown in Scheme 1. The complex formation reaction and voltammetry behaviour of Hg2+ at TSAB/MWCNT modified electrode are reported for the first time. The TSAB/MWCNT modified electrode has enhanced the sensitivity of determination of Hg2+ in seawater and lake water. The stripping voltammetric results of real samples were compared with atomic absorption spectrometry (AAS) and the details are discussed in this paper.

Scheme 1
scheme 1

a) Synthesized of 1,2,4,5-tetrasalicylideneaminobenzene (TSAB) Ligand, b) preparation of MWCNT/TSAB Ligand modified electrode, c) Mechanism of Hg2+ in TSAB/MWCNT modified electrode using anodic stripping voltammetry

Full size image

Experimental

Chemical

MWCNT (size 110 nm and purity ≥ 90%) and graphite electrode (3 mm Dia) was purchased from Sigma-Aldrich. Salicylaldehyde and 1,2,4,5-tetraaminobenzene tetrahydrochloride, mercury acetate and sodium acetate were purchased from SRL, India. Other chemicals were used for analytical grade.

Apparatus

Scanning electron microscopy (SEM) was performed with Hitachi instrument (S-3400 N). Voltammetric determinations were carried out with a CHI 660 B electrochemical workstation (CHI instruments, USA). A conventional three-electrode system, consisting of a TSAB/MWCNT working electrode, a saturated calomel reference electrode (SCE) and a platinum wire counter electrode, was employed.

Synthesis of TSAB ligand

The ligand was prepared by the reported procedure [23] with a small reconstruction. Sodium methoxide was added to a stirred suspension of the 1,2,4,5-tetraaminobenzene tetrahydrochloride (0.93 g, 3.3 mmol) in 50 mL dry methanol under nitrogen atmosphere until reaction was complete, which resulted in a yellow solution. Salicylaldehyde (1.61 g, 13.2 mmol) was then immediately added dropwise to the warm solution. The mixture was refluxed overnight and the orange solid obtained was filtered, washed several times with methanol and ether and air-dried yield (90%, mp 270–290 °C).

Fabrication of modified electrode

Paraffin wax-impregnated graphite electrodes (PIGE) have been used for electrode modification. The spectroscopic grade graphite rods have been impregnated with paraffin wax procedure developed by Scholz et al. [24]. Briefly, the graphite rods with the diameter of 3 mm and 4 cm length were put into the paraffin under a vacuum and kept until no more bubbles evolve from the rod. After that, the graphite rods were stripped and the excess paraffin from the surface of electrodes was cleaned. The one end of the electrode surface was polished with emery and smooth paper to get a mirror-like finish and rinsed with distilled water. The impregnation of graphite with paraffin helps to fill all the pores and in order to prevent the high background currents and spoiling of the electrode ingression of electrolyte or analyte solution [25]. Thus, we have chosen PIGE for the base electrode for modification. First, 0.1 mg MWCNT dispersion was prepared by dissolving 0.1 mg of MWCNT material in 1 mL of ethanol and sonicated for 15 min. Then, 10 μL of MWCNT suspension was coated on the bare electrode surface and allowed to dry for 10 min. Afterwards, the TSAB ligand was dissolved in 1 mL of acetone and 10 μL of the 1 mmol ligand was drop cast on the MWCNT electrode and dried in air for 5 min.

Voltammetric determination for Hg2+ ion

The voltammetric determination for Hg2+ procedure consisted of preconcentration, deposition potential, stripping and regeneration steps. In preconcentration steps, the modified electrode was dipped in a 0.1-M NaNO3 solution which contained 30 nM of Hg2+ stirred for 180 s such that the ligand on the electrode surface forms complexes. Afterwards, the TSAB/MWCNT electrode was then removed and transferred into a fresh background electrolyte of 0.1 M NaNO3. A negative potential of − 0.1 V was applied for 60 s so that the metal ion was reduced on the electrode surface. The metal ion on the electrode surface was stripped by scanning the potential from − 0.1 to 0.7 V. Then, the electrode was regenerated by immersing the electrode in 0.01 M EDTA solution (pH 4.5) as Hg2+ was removed from the electrode as water soluble Hg2+-EDTA complex and the electrode was washed with distilled water.

Results and discussion

Characterisation of MWCNT/TSAB ligand

The synthesised ligand was characterised by Fourier transform infrared (FT-IR) spectroscopy and nuclear magnetic resonance (NMR) spectroscopy. The modified electrode was characterised by SEM with energy dispersive x-ray analysis (EDAX) analysis, electrochemical impedance spectroscopy (EIS) spectroscopy and square wave anodic stripping voltammetry (SWASV). Fig. S1 shows the FT-IR spectrum of the ligand. The strong absorption band appears at 3315 cm−1 which could be assigned to be presence of OH groups. The presence of C=N and C–O groups were observed at 1611 and 1210 cm−1, respectively. H1-NMR spectra data of TSAB ligand are shown in Fig. S2. A sharp singlet peak is observed at 8.016 ppm were assigned to CH=N proton and OH proton of four phenolic group is observed at 4.82 ppm. The aromatic protons show peaks at 6.553-7.993 ppm. All the spectroscopic results are in good agreement with the reported values [23].

Electrochemical characterisation of TSAB/MWCNT

The surface bound feature of the bare electrode, MWCNT and TSAB/MWCNT ligand modified electrodes in 1 mM [Fe(CN)6]3−/4− solution was investigated by cyclic voltammetry (CV) and EIS as shown in Fig. 1.

Fig. 1
figure 1

Cyclic voltammogram of bare electrode (a), MWCNT electrode (b) and (c) TSAB/MWCNT (A), Impedance plots (B) of bare electrode (a), MWCNT electrode (b), TSAB/MWCNT ligand modified electrode (b) with 1 mM[Fe (CN)6]3-/4- in 0.1 M acetate buffer. Inset: Randles equivalent circuit

Full size image

Fig. 1(A) show the CV of bare electrode, MWCNT and TSAB/MWCNT modified electrodes in 1 mM of [Fe (CN) 6]3−/4−. The figure shows that the TSAB/MWCNT electrode exhibited higher anodic peak (Ip) and lower cathodic peak (Epc) compared to the bare electrode and MWCNT electrode. The electroactive surface area was estimated using Randles-Sevcik Eq. (1).

$$ \mathrm{Ip}=2.69\times {10}^5{AD}^{\raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{$2$}\right.}{n}^{\raisebox{1ex}{$3$}\!\left/ \!\raisebox{-1ex}{$2$}\right.}{v}^{\raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{$2$}\right.}C $$
(1)

where A is the area of the electrode (cm2), D is the diffusion coefficient of the molecule in solution (cm2/s), C is the concentration of the molecule in the bulk solution (mol/cm3) and ν is the scan rate (V/S) for 1 mM of [Fe(CN)6]3−/4− in 0.1 M acetate buffer solution (ABS). The electroactive surface area of the bare electrode, MWCNT electrode and TSAB/MWCNT modified electrode demonstrated in the present work was calculated to be 0.012, 0.052 and 0.082 cm2, respectively.

In Fig. 1(B), its shows the Nyquist plots of the electrochemical impedance study for the bare (PIGE) electrode (curve a), MWCNT electrode (curve b) and TSAB/MWCNT modified electrode (curve c) in 1 mM [Fe(CN)6]3−/4− solution. The modified Randle’s equivalent circuit is given as an inset in Fig. 1(B). The resistance in solution (Rs) for the bare electrode was 52.36 Ω, for the MWCNT electrode was 38.16 Ω and for the TSAB/MWCNT electrode was 23.4 Ω. The charge transfer resistance (Rct) values observed were 7490 Ω for the bare electrode, 917.2 Ω for the MWCNT electrode and 94.83 Ω for the TSAB/MWCNT electrode.

The conductivity of the MWCNT electrode and TSAB/MWCNT electrode is given in Eq. (2)

$$ \sigma =\frac{l}{A}\times \frac{1}{R} $$
(2)

where l is the length of the electrode, A is the surface area of the electrode and R is the conductance of the MWCNT electrode and MWCNT/TSAB ligand electrode. By applying Eq. (2), it shows that the electrode conductance of the bare electrode is 17.7 × 10−4 S cm−1, of the MWCNT electrode is 14.5 × 10−4 S cm−1 and of the TSAB/MWCNT ligand electrode is 140 × 10−4 S cm−1. From the equation, it was conformed that the TSAB/MWCNT electrode conductivity is higher than that of the MWCNT electrode.

The phase degree was calculated by using Eq. (3)

$$ \phi ={\tan}^{-1}\left[\frac{1}{1+2R\mathrm{s}/R\mathrm{ct}}\right] $$
(3)

From the equation, the phase degree (ϕ) for the bare electrode was 44.4°, for the MWCNT electrode was 33.8° and for the TSAB/MWCNT was estimated to be 28.0°. According to Eq. (2), the phase angle (ϕ) is directly proportional to (Rct), respectively, so that the phase angles decrease at the TSAB/MWCNT ligand modified electrode and it has enhanced the electron transfer kinetics at the electrode interface. Therefore, this TSAB/MWCNT ligand modified electrode was suitable for stripping analysis.

SEM with EDAX image for different electrodes

The surface morphology of the bare electrode, MWCNT electrode, TSAB electrode, MWCNT/TSAB ligand modified electrodes and MWCNT/[TSAB-Hg2+] metal complex was studied by SEM with EDAX analysis, and the image obtained is shown in Fig. 2. The bare electrode has a smooth surface as shown in Fig. 2(A) and the EDAX image showed that the carbon element peak present in Fig. 2(B) and TSAB/MWCNT ligand is like the rod structure in Fig. 2(E), and the EDAX image showed the presence of carbon and oxygen peak in Fig. 2(F). The MWCNT modified electrode structure shows flexible surface (Fig. 2(C)) and the EDAX image for the MWCNT electrode showed an increase in carbon element peak (Fig. 2(D)) compared with the bare electrode. The preconcentration of Hg2+ on the modified electrode (metal complex) shows the needle-shape structure (Fig. 2(G)), and in the EDAX image, the results showed that the presence of Hg2+ peak (Fig. 2(H)) on the surface of MWCNT/TSAB ligand electrode is observed.

Fig. 2
figure 2

SEM with EDAX image for bare electrode (A and B), MWCNT modified (C and D), MWCNT/TSAB (E and F) and MWCNT/TSAB-Hg2+ (G and H)

Full size image

Voltammetric behaviour of TSAB/MWCNT modified electrode

Electrochemical detection of Hg2+ that is based on the TSAB/MWCNT ligand modified electrode using CV and square wave anodic stripping voltammetry was studied as shown in Fig. 3. In CV, the TSAB/MWCNT ligand modified electrode is immersed in 0.1 M NaNO3 containing 0.5 μM of Hg2+ ion for 180 s (Fig. 3(A)). After preconcentration, the modified electrode is washed with double distilled water and then transferred into freshly NaNO3 solution (pH 5.0). From the CV of the modified electrode (curve b), a well-defined anodic peak (0.34 mV) and cathodic peak (0.38 mV) were observed. In the absence of Hg2+ ion (curve a), no peak shows. The CV result implies that the metal ion was preconcentrated at the TSAB/MWCNT electrode surface. Therefore, the modified electrode is used for the determination of Hg2+ ion using anodic stripping voltammetry.

Fig. 3
figure 3

(A) Cyclic voltammograms of modified electrode in 0.1 M NaNO3 solution containing 0.5 μM Hg2+ without (a) and with (b) 180 s preconcentration at the scan rate of 50 mV s−1. Reduction potential − 1.2 V for 60 s. (B) Stripping current of 30 nM for Hg2+ ion at PIGE (a), MWCNT (b) and MWCNT/TSAB modified electrode (c) in 0.1 M NaNO3. Preconcentration time 180 s, deposition potential − 0.1 V, amplitude 50 mV

Full size image

In SWASV, preconcentration was performed by immersing the electrode in 0.1 M NaNO3 solution containing 30 nM of Hg2+ ion under stirred conditions. The electrode was transferred into fresh electrolyte and reduced to − 0.1 V for 60 s. The anodic stripping voltammetry was carried out from − 0.1 to 0.7 V. The result obtained with the stripping analysis is shown in Fig. 3(B). From the figure, it can be observed that the modified electrode shows a high current enhancement compared to the bare electrode (MWCNT without ligand). The nature of ligand containing imine and hydroxyl groups is in coordinate with Hg2+ in solution. Therefore, the TSAB/MWCNT ligand modified electrode has high sensitivity current for determination of Hg2+ using anodic stripping voltammetry (ASV). The stripping mechanism of Hg2+ using the TSAB/MWCNT ligand modified electrode is explained by the following steps:

  1. 1)

    Preconcentration step for surface MWCNT electrode (0.1 M NaNO3, for 180 s pH − 5.0)

$$ {\left(\mathrm{TSAB}/\mathrm{MWCNT}\right)}_{\mathrm{surface}}+{{\mathrm{Hg}}^{2+}}_{\mathrm{aqueous}}={\left({\mathrm{Hg}}^{2+}-\mathrm{TSAB}/\mathrm{MWCNT}\right)}_{\mathrm{surface}}\left(\mathrm{Complex}\ \mathrm{formation}\right) $$
  1. 2)

    Reduction step (0.1 M NaNO3, pH − 5.0, − 0.1 V)

$$ {\left({\mathrm{Hg}}^{2+}-\mathrm{TSAB}/\mathrm{MWCNT}\right)}_{\mathrm{surface}}+2\mathrm{e}-\kern1em ={\left({\mathrm{Hg}}^0-\mathrm{TSAB}/\mathrm{MWCNT}\right)}_{\mathrm{surface}}\left[\mathrm{Reduction}\ \mathrm{step}\right] $$
  1. 3)

    Stripping step (0.1 M NaNO3, pH − 5.0, − 0.1 to 0.7 V)

$$ {\left({\mathrm{H}\mathrm{g}}^0-\mathrm{TSAB}/\mathrm{MWCNT}\right)}_{\mathrm{surface}}+{2}^{\mathrm{H}+\kern0.5em }=\kern0.75em {\left(\mathrm{TSAB}/\mathrm{MWCNT}\right)}_{\mathrm{surface}}+\mathrm{Hg}2{+}_{\mathrm{aqueous}}\left[\mathrm{Stripping}\ \mathrm{step}\right] $$

Optimisation parameters

Figure 4 shows the electrochemical detection of Hg2+ ion using anodic stripping voltammetry. Using the TSAB/MWCNT ligand modified electrode, various parameters such as preconcentration time, effect of pH, different supporting electrolyte and stripping media were investigated.

Fig. 4
figure 4

(A) The effect of supporting electrolytes: 0.1 M solutions of (a) KNO3, (b) NH4NO3, (c) acetate buffer and (d) NaNO3. (B) Effect of pH and (C) preconcentration time for Hg2+ ions containing 50 nM at modified electrode

Full size image

In electrochemical determination, the choice suitable for different supporting electrolytes of a stripping medium on the current response was investigated using KNO3, NH4NO3, acetate buffer and NaNO3. Figure 4(A) shows the stripping voltammogram for Hg2+ ion which was preconcentrated from these different media. The maximum current response was found in 0.1-M NaNO3 solution as the most suitable medium for preconcentration of Hg2+ ion. Hence, in subsequent experiments, the 0.1-M NaNO3 solution was used for preconcentration of Hg2+ ion.

The effect of pH that was evaluated for Hg2+ in the pH range of 3.0–6.0 in the 0.1-M NaNO3 medium was studied and the results obtained are given in Fig. 4(B). The preconcentration of Hg2+ increases from pH 3.0–5.0 and then decreases from 5.0 to 6.0, showing a maximum pH 5.0 in 0.1 M NaNO3 solution. The protonation of the TSAB ligand occurs at low pH which has resulted in low preconcentration amount as complexation does not occur with protonated TSAB. On the other hand, metal hydroxide forms at higher pH, reducing the amount of Hg2+ present in the solution leading to lower amount of Hg2+ preconcentration. A pH 5.0 is most suitable for the preconcentration of Hg2+ of the electrode surface. Hence, subsequent experiments have been carried out at pH 5.0 in 0.1 M of NaNO3 medium.

The influence of preconcentration on the stripping peak current at different time intervals from 60 to 420 s was studied in Fig. 4(C). The amount of Hg2+ preconcentrated at the TSAB/MWCNT ligand modified electrode surface was found to increase with the increase in the preconcentration time. Therefore, the stripping peak current of Hg2+ also increased. Thus, preconcentration time of 180 s was fixed in order to shorten the analysis time.

Analytical performance

Calibration curve

In SWASV, the TSAB/MWCNT ligand modified electrode was the response between the stripping peak (Ip) and concentration (C) of Hg2+ that were investigated. In Fig. 5, after the accumulation of Hg2+ for 180 s, the Ip that is proportional to the C ranging from 2.4 to 180 nM was observed; the regression equation is expressed as Ip (μA) = (0.019 ± 0.0003) x + (0.027 ± 0.03) [Hg2+] with a correlation coefficient of R2 = 0.996, and the lowest detectable concentration of Hg2+ is achieved at 0.8 nM after 180 s accumulation. A calibration graph has been drawn by plotting the Ip vs. C of Hg2+ taken and is shown as an inset in Fig. 5. Further, linear range and the detection limits obtained with the TSAB/MWCNT ligand modified electrode have shown better sensitivity and linear range with previously reported modified electrodes (Table 1). From the table, the present modified electrode showed a better linear working range and low detectable Hg2+.

Fig. 5
figure 5

Anodic stripping voltammograms of different concentrations of Hg2+ ion in the range from 2.4 to 180 nM on TSAB/MWCNT modified electrode in 0.1 M NaNO3 solutions. Inset: calibration plot

Full size image
Table 1 Comparison of different electrodes for determination of Hg2+
Full size table

Reproducibility and stability

To examine the reproducibility, stability of the TSAB/MWCNT electrode for electrochemical sensors of Hg2+ ion was investigated. The reproducibility of electrodes was performed by four electrodes prepared with the determination of standard solution of 50 nM Hg2+ ion. The RSD for the response between electrodes was 3.0% for Hg2+, respectively. The stability of stripping voltammetry for the same modified electrode was used for 3 weeks, and the maximum deviation obtained was 3.5%, respectively. The stability of the modified electrode is shown in Fig. 6. The results indicate that the multi-walled carbon nanotube/TSAB ligand electrode has good sensitivity, reproducibility and long-time stability.

Fig. 6
figure 6

Stability for modified electrode containing 50 nM of Hg2+ in 0.1 M NaNO3 solution

Full size image

Anti-interference studies

Under optimal conditions, the interference studies were carried out by the addition of some metal ions which contain the 50 nM of Hg2+ ion. The results that show in the presence of 50 nM concentration of Pb2+, Cd2+, Zn2+, Sn2+, Ni2+,Co2+, Ag+ and 30 nM of Cu2+ and As3+ have shown insignificant effect on the stripping peak current for 50 nM Hg2+ as shown in Fig. 7. The prepared electrode shows excellent reproducibility and stability and the result of the interference was also investigated.

Fig. 7
figure 7

Effects of various metal ions on the analytical signal of 50 nM Hg2+ in the presence of 50 nM concentration of Pb2+, Cd2+, Zn2+, Sn2+, Ni2+, Co2+, Ag+, Cu2+ and As3+. Other experimental conditions are the same as in Fig. 4

Full size image

Real sample analysis

The use of the MWCNT/TSAB ligand modified electrode for the determination of Hg2+ in well water and lake water. The lake water sample was collected from Kodaikanal lake and well water from poonamallee from Tamil nadu, India. The results are shown in Table 2, and the recovery test had been carried out by spiking in different concentrations of Hg2+ to the water samples which was analysis of the prepared electrode. The result of well water (Fig. 8a) and lake water (Fig. 8b) are shown in Fig. 8. From Table 2, it was confirmed that the determination of Hg2+ at TSAB/MWCNT ligand modified electrode was not affected by sample matrix, and the results of well water and lake water corroborated with those obtained by atomic absorption spectrometry (AAS), with the R.S.D values lower than 2.3 and 3.0% for Hg2+ which showed a good recovery in the presence of Hg2+ for analysis. In all the samples, good recoveries (between 100.4 and 104%) had been obtained (Table 2). Therefore, the prepared electrode was extremely showed good recovery for determination of Hg2+ in different water samples.

Table 2 Determination of Hg2+ in real samples
Full size table
Fig. 8
figure 8

Anodic stripping voltammetry of well water (A) and lake water containing (B) 50 and 100 nM of Hg2+ in 0.1 M NaNO3 solution

Full size image

Conclusions

Fabrication of MWCNT and synthesised TSAB ligand modified electrode was prepared. Compared with a MWCNT electrode, the TSAB/MWCNT ligand shows an extremely improves in stripping current peak for the Hg2+. The developed method is highly selective and shows good sensitivity and wide linearity ranges for Hg2+. The electrode modified with the MWCNT/TSAB ligand exhibited high sensitivity and selectivity for the determination of Hg2+ and the LOD was 0.8 nM at S/N = 3 was achieved. Due to the combination of the good stability of MWCNT and the strong chelating ability of Hydroxyl groups to Hg2+. The performances of the prepared method were favourably utilised for determining Hg2+ in well water and lake water with good recovery ranging from 100.4% to 104%. Additionally, the sensor showed remarkable stability and repeatability.