Gold Nanoparticles-Thiol-Functionalized Reduced Graphene Oxide Coated Electrochemical Sensor System for Selective Detection of Mercury Ion

Reduced graphene oxide (rGO) was purchased from Carbon Solution, USA. Dodecanethiol was purchased from Alfa-Aeser. Mercury (II) nitrate purchased from Acros Organics. Gold chloride trihydrate (HAuCl₄.3H₂O), ascorbic acid, sodium borohydride, zinc acetate trihydrate, copper sulfate pentahydrate, cobalt acetate, manganese acetate, lead acetate and nickel nitrate were purchased from Sigma-Aldrich, India. Silver nitrate (AgNO₃) was received from SRL, India.

All the solutions were prepared by using milli-Q water. The 0.1 M PBS (pH = 7.0) was prepared using potassium monohydrogen phosphate (K₂HPO₄) and potassium dihydrogen phosphate (KH₂PO₄) which were used as an electrolyte solution. All the metal salt solutions were prepared at a required concentration by dissolving the corresponding salts in the aqueous solution.

The electrochemical measurements were performed with Biologic science instrument (SP-300), Electrochemical workstation from CH Instruments, USA (Model: CHI-760E) and Klyte 1.2 (Potentiostat, India). GCE (3 mm in diameter) was used as the working electrode, Ag/AgCl (1 M KCl) as the reference electrode and platinum wire as the counter electrode in a three-electrode system. The surface morphology of Au-NPs and rGO-SH were analyzed by field-emission scanning electron microscope (FE-SEM-EDS, FEI-Quanta-200FEG) and FEI-Tecnai F20 transmission electron microscope at accelerating voltage of 200 kV. The absorption spectrum of Au-NPs was characterized by 2000c nanodrop spectrophotometer, Nanodrop technologies, USA.

The rGO powder (2 mg) was dispersed in 15 mL dodecanethiol via the probe sonication for 1 hr with the amplitude of 40% with the pulse of 5 sec On and 2 sec OFF. During the ultrasonication process, an ice-bath was used to control the temperature. As obtained rGO/SH dispersion was highly stable up to 4 weeks without any settling effect at room temperature which may due to dodecanethiol having a long aliphatic carbon chain group which strongly attached with rGO sheets (Scheme 1, Step-1).

Scheme 1. (Step 1) Preparation of rGO/SH dispersion. (Step 2) Fabrication of electrochemical sensor to detect Hg²⁺ ions by DPV.

10 ml aqueous solution containing 1 mM HAuCl₄.3H₂O and 0.1 M cetyltributylammonium bromide (CTAB) were prepared. Further, 0.2 mM AgNO₃ (10 mL) and 2 mM ascorbic acid added until it turns the color from orange to colorless. Finally, 10–100 μL of 1 mM NaBH₄ (freshly prepared) added until pinkish violet color appears.³¹

The surface of GCE was cleaned with α-alumina powder (0.05 μm) on a polishing cloth after that washed with double-distilled water until it gives a mirror-like surface. Further, the GCE was activated by potential cycling from −0.5 to 0.5 at a scan rate of 50 mV/s in 0.1 M H₂SO₄ for 10 cycles. Subsequently, the electrode was tested by recording cyclic voltammograms (CVs) in 5 mM K₄[Fe(CN)₆] and K₃[Fe(CN)₆) in 0.1 M KCl solution.^32–34

10 μL rGO/SH dispersion was placed on the surface of GCE by drop casting method. After complete drying at 40°C in a hot air-oven, 10 μL of AuNPs solution was coated on the GCE/rGO-SH surface and then it dried again in hot air oven at 40°C for 10 min (Scheme 1, Step 2). Later, the GCE/rGO-SH/Au-NPs was gently washed using distilled water to remove any loosely bounded materials from the electrode surface. After that the GCE/rGO-SH/Au-NPs is dipped in different concentration of Hg²⁺ solution (Note: Hg(NO₃)₂ is sparingly soluble in water, so to obtain a clear solution, 3–4 drops of HNO₃ was added) for 20 min and then it was scanned between −0.2 and 0.3 V by cyclic voltammetry in 0.1 M PBS (during this potential scan, adsorbed Hg²⁺ on the electrode surface get reduced to Hg). Later, the electrochemical oxidation of Hg present on the GCE/rGO-SH/Au-NPs was studied by differential pulse voltammetry (DPV). Differential pulse voltammograms (DPVs) were recorded by potential scanning from 0.0 to 0.3 V at a scan rate of 50 mV/s in 0.1 M PBS. The deposited Hg get oxidized to Hg²⁺ ion and giving the corresponding oxidation peak at 0.172 V (Scheme 1, Step 2).³⁵

Fig. 1 shows the photographic images of dodecanethiol (image i) and rGO dispersed in dodecanethiol, (image ii) resulting the black color solution as obtained by probe-sonication. Schematic diagram of the functionalization of rGO-SH with Au-NPs for the detection of Hg ion, is shown in Scheme 1. As it shown in Fig. 1, image (ii), rGO sheets were well dispersed (as it confirmed by SEM and TEM analysis) in dodecanethiol solution up to four weeks without any aggregation.

Zoom In Zoom Out Reset image size

Next, the formation of Au-NPs was characterized by UV-Vis absorption spectrum. A strong absorption peak appeared at 520 nm indicated the formation of Au-NPs (Fig. 2a). Furthermore, the interaction between Au-NPs and Hg²⁺ ions was studied by UV-Vis spectroscopy. UV-Vis spectra were recorded before and after addition of 5 μM concentration of Hg²⁺ into Au-NPs solution (Fig. 2b). The addition of Hg²⁺ exhibited a peak shift of 10 nm (blue curve). It revealed that Au-NPs had interaction with the Hg²⁺, so the absorption spectrum was down (blue) shifted. It was also reported that Au-NPs could form amalgam with Hg²⁺ ions.³⁶

Zoom In Zoom Out Reset image size

FT-IR spectra were recorded for dodecanethiol and –SH functionalized rGO powder (Fig. 2c, red and blue curves). Fig. 2c, red curve shows the absorption peak at (peak i) 745 cm⁻¹ and (peak v) 2900 cm⁻¹ correspond to stretching and bending of –CH₃, (iv) peak at 2800 cm⁻¹ is for –CH₂ stretching and the peak position³⁷ (peak iii) at 1600 cm⁻¹ corresponds to carbonyl group. Those peaks were observed in dodecanethiol and rGO-SH except (ii) the peak at 1200 cm⁻¹ corresponds to thioketals C=S formed between carbonyl group in rGO and -SH of dodecanethiol. Since this peak was not appeared in dodecanethiol, so it confirmed the covalent interaction between the rGO and dodecanethiol.³⁸ An another peak (vi) was observed in rGO-SH due to rGO material which corresponds to –OH group (Fig. 2c).^33,34

The samples for FE-SEM were prepared by drying overnight in the hot air oven and collected the powder samples. TEM sample was prepared by coating on the copper grid and then it was dried at the room temperature. Fig. 3 shows the surface morphologies of (image a) rGO-SH as characterized by the FE-SEM. Further, TEM image which revealed sheets like structures of rGO-SH film, as it confirmed the successful exfoliation of rGO sheets in dodecanethiol (image b).³⁹ Dodecanethiol did increase the rGO sheets interlayer distance after functionalization with basal plane of rGO containing the carbonyl group. It might form thioketals which made the material to be more functional. Next, the Au-NPs were also characterized by TEM. The average particle sizes of Au-NPs were in the range of ∼10 nm (Fig. 3c).

Zoom In Zoom Out Reset image size

The solution of 5 μM Hg²⁺ was prepared and characterized by DPV technique at the (a) bare-GCE, (b) GCE/rGO/–SH and (c) GCE/rGO-SH/Au-NPs (Fig. 4). All these modified electrodes were dipped in the standard solution of 5 μM Hg²⁺ (5 mL) for 20 min after that it was dried in hot air oven for 5 min. Further the electrodes were immersed in the distilled water to remove if any loosely bounded soluble materials present on the electrodes surface. After that all the modified electrodes were used as working electrodes to record DPVs in the potential range from 0.0 to 0.3 V at a scan rate of 100 mV/s using PBS as an electrolyte (pH = 7) (Fig. 4). The oxidation peak was not observed for Hg at 0.172 V on bare GCE (curve i) and GCE/rGO-SH (curve ii). It corroborated that these electrodes were not suitable to attract Hg²⁺ ions toward the electrodes surface. Interestingly, before immersion of GCE/rGO-SH/Au-NPs in Hg²⁺ ion, it did not show any oxidation peak. But, after the immersion of GCE/rGO/-SH/Au-NPs in Hg²⁺ ion, the GCE/rGO/-SH/Au-NPs only showed a strong oxidation peak at 0.172 V (curve iv), it confirmed that this electrode material is suitable for sensing of mercury ions with the combination of rGO-SH/Au-NPs film because of the strong Au-NPs affinity toward the Hg²⁺ ion (Fig. 4).

Zoom In Zoom Out Reset image size

We also described the schematic presentation of electrode active material synthesis and detecting Hg²⁺ ion in Fig. 5. The rGO-SH film may acted as a nucleophile as it had the sulfur group which easily interacted with Au-NPs.⁴⁰ The interaction between rGO-SH/Au-NPs film and Hg²⁺ ion might have occurred through the amalgamation type reactions. During the potential cycling between 0.3 and −0.2 V, the assembled Hg²⁺ ions get reduced to Hg⁰. Later, when DPV was recorded (anodic scan), the attached Hg⁰ was get oxidized back to Hg²⁺ ion at 0.172 V.³⁵

Zoom In Zoom Out Reset image size

Figs. 6a and 6b shows CVs of the of (a) GCE/rGO/Au-NPs and (b) GCE/rGO-SH/Au-NPs recorded in 0.1 M KCl solution in the potential range from −0.2 to 0.3 V at a scan rate of 50 mV/s. The stability of the modified electrodes was analyzed by cyclic voltammetry (CV) by comparing the materials of rGO dispersed in dimethylformamide (DMF) and in dodecanethiol. Both dispersions were prepared after probe sonication independently for 1 hr. Two individual GCE's were modified with rGO/DMF and rGO/SH dispersions using 10 μL of each dispersion on the electrode surface. After that, 10 μL of Au-NPs dispersion was placed on the both electrode's surface and dried (in hot air oven for 5 min) it as a second layer. Finally, the CVs were recorded by applying potential between −0.2 and 0.3 V at a scan rate of 50 mV/s for 10 cycles in KCl solution. The GCE/rGO/SH film exhibited enhanced anodic peak currents due to higher loading of Au-NPs compared to the GCE/rGO/DMF film (Figs. 6a and 6b). The reason might be due to presence of –SH group on the rGO sheets; it had strong affinity toward the Ag-NPs. This result confirmed that rGO-SH functionalized film can be utilized for fabrication of Hg²⁺ sensor.

Zoom In Zoom Out Reset image size

We also studied the self-assembly behavior of Au-NPs on the rGO-SH film. Fig. 7 shows the time variation studies upon self-assembly of Au-NPs onto the rGO-SH film modified GCE. The oxidation peak increases with the immersion time in Au-NPs dispersion. After the immersion of GCE/rGO-SH in Au-NPs dispersion, the CVs were recorded in the potential range from −0.2 to 0.3 at a scan rate of 50 mV/s. The oxidation peak currents were increased with the immersion time due to the continuous adsorption of Au-NPs on the rGO-SH film modified electrode (Fig. 7). The self-assembling nature of nanoparticles was increased up to 3 hr. However, after 4 hr of immersion time, the oxidation peak current was again started to decrease which indicated the saturation effect of Au-NPs self-assembly on the modified electrode's surface. This study has also proved the existence of strong interaction between rGO-SH film and Au-NPs (Fig. 7).

Zoom In Zoom Out Reset image size

The different concentrations of mercury ion solutions were prepared by diluting the stock solution of 1 mM Hg²⁺ solution. The prepared various Hg²⁺ concentration solutions were analyzed by DPV technique using the GCE/rGO-SH/Au-NPs. The GCE/rGO-SH/Au-NPs was immersed in different concentration of Hg²⁺ ion (5 mL) solution using the same procedure as given in Hg sensor fabrication section. The Hg oxidation current was gradually increased from the lower concentration to higher which indicated the high sensitivity of the electrode (Fig. 8a). Fig. 8b shows the calibration curve plotted against Hg²⁺ ions concentration vs. oxidation peak current in the range from 1–10 μM in 0.1 M PBS (Each data point is the mean value of three measurements recorded using three different GCE/rGO-SH/Au-NPs modified electrodes, Fig. 8b). The linear graph was well fitted with the correlation coefficient (R²) value of 0.972 (Fig. 8b). The detection limit of Hg²⁺ was found to be 0.2 μM (S/N = 3).²¹ The reproducibility of the GCE/rGO-SH/Au-NPs sensor was also analyzed by detecting the concentration of 5.0 μM Hg²⁺ in 0.1 M PBS. The obtained data indicated that the modified electrode possesses an acceptable reproducibility with a relative standard deviation (RSD) of 5.52% for five parallel experiments with different electrodes.

Zoom In Zoom Out Reset image size

The GCE/rGO-SH/AuNPs based sensor was studied with different metal ions (Co²⁺, Ni²⁺, Cu²⁺, Mn²⁺, Zn²⁺, Al³⁺, Pb²⁺, and Hg²⁺). The standard solutions of all metal ions (5 μM) were prepared and the analysis procedure was repeated as given in Hg sensor fabrication section. As shown in Fig. 9, the Hg oxidation was only exhibited the higher peak current at 0.172 V. All other metal ions did not show significant oxidation current compared to Hg. From this interference metal ion analysis, it was confirmed that the rGO-SH-AuNPs film is very selectively interacted with Hg²⁺ ion compared to other metal ions which may be due to the unique combination of rGO-SH and AuNPs (Fig. 9).

Zoom In Zoom Out Reset image size

The as-prepared GCE/rGO-SH/Au-NPs was utilized as a sensor to measure the known concentration of Hg²⁺ spiked into the real water samples. The tap water samples (collected from our laboratory) (1 mL) were spiked with 1 and 5 μM concentration of Hg²⁺ ions and made up to 10 mL solution using distilled water. The DPV was recorded as given in Electrochemical detection of mercury ion section. The Hg recovery was estimated as it was in the range of 94.0–97.5%. We envisage that this GCE/rGO-SH/Au-NPs can be applied to detect Hg²⁺ ions selectively in the environmental samples (Table I).