Controllable Synthesis of a Porous PEI-Functionalized Co3O4/rGO Nanocomposite as an Electrochemical Sensor for Simultaneous as Well as Individual Detection of Heavy Metal Ions.

The present study focuses on the strategy of employing an electrochemical sensor with a porous polyethylenimine (PEI)-functionalized Co3O4/reduced graphene oxide (rGO) nanocomposite (NCP) to detect heavy metal ions (HMIs: Cd2+, Pb2+, Cu2+, and Hg2+). The porous PEI-functionalized Co3O4/rGO NCP (rGO·Co3O4·PEI) was prepared via a hydrothermal method. The synthesized NCP was based on a conducting polymer PEI, rGO, nanoribbons of Co3O4, and highly dispersed Co3O4 nanoparticles (NPs), which have shown excellent performance in the detection of HMIs. The as-prepared PEI-functionalized rGO·Co3O4·PEI NCP-modified electrode was used for the sensing/detection of HMIs by means of both square wave anodic stripping voltammetry (SWV) and differential normal pulse voltammetry (DNPV) methods for the first time. Both methods were employed for the simultaneous detection of HMIs, whereas SWV was employed for the individual analysis as well. The limits of detection (LOD; 3σ method) for Cd2+, Pb2+, Cu2+, and Hg2+ determined using the rGO·Co3O4·PEI NCP-modified electrode were 0.285, 1.132, 1.194, and 1.293 nM for SWV, respectively. Similarly, LODs of Cd2+, Pb2+, Cu2+, and Hg2+ were 1.069, 0.285, 2.398, and 1.115 nM, respectively, by DNPV during simultaneous analysis, whereas they were 0.484, 0.878, 0.462, and 0.477 nM, respectively, by SWV in individual analysis.

Introduction

Over the past several decades, heavy metal ions (HMIs) have been attracting great attention as they are extremely injurious to the biosphere, and even their minute aggregates are of much threat to human health.[1−3] In order to prevent this, the most sensitive and rapid removal method of trace HMIs is the need of today. Until now, technologically advanced approaches such as electrochemical exploration, anodic stripping voltammetry (ASV), and differential pulse ASV (DPASV) methods have been employed due to their fast response, convenience, and low cost.[4−6]

So far, a stronger electrochemical sensing interaction in an aqueous solution for Pb2+ and Hg2+ has also been practiced on the modified carbon electrode (GCE) for DPASV by using an amine functionalized graphene oxide..[7] Similarly, DPASV has also been performed for the simultaneous recognition of Cd2+, Pb2+, Cu2+, and Hg2+[8] by using gold nanoparticles (NPs) on the surface of GCE. Here, differential normal pulse voltammetry (DNPV) was used only for the simultaneous sensing/detection of HMIs because the simultaneous detection of HMIs using the DNPV technique results in strong intermetallic interaction; therefore, this method was not used in individual analysis.

Recently, for highly sensitive and selective recognition of HMIs, nanomaterials with various functionalities and accumulating abilities to adjust electrochemical electrodes for specific target HMIs have been reported.[9−15] Among them, Co3O4 is an important semiconductor oxide that has scientific application in energy storage and redox catalysis.[16] However, the concrete use of Co3O4 has been unfulfilled yet, mostly owing to particle aggregation, expansion–contraction in volume, poor cycling stability, and capacity loss.[17,18]

To enhance the electronic conductivity and to avoid aggregation, Co3O4 NPs have been used with carbon-based materials, such as amorphous carbon,[19] carbon nanofibers,[20] carbon nanotubes,[21,22] and graphene/reduced graphene oxide (rGO).[23−25] In particular, it has been verified that graphene/rGO not only improves the electronic conductivity but also prevents the aggregation of Co3O4 NPs. Thus, the design of the Co3O4/graphene composite is the most competent selection for progress of the electrochemical activity of Co3O4.[26,27]

As reported earlier,[28] the described N-doped PC-Co3O4 nanocomposite (NCP) that possesses numerous N functional groups deliver additional binding sites and improved electrochemical performance. The N-doped PC-Co3O4 NCPs have efficiently relieved the aggregation of Co3O4 and exhibited excellent cycle performance capacity. Similarly, graphene-based Co3O4 composites have obtainable stability, sustainability, high specific capacity, and excellent electrochemical performance.[29−31]

The most important characteristic of Co3O4 nanocrystals is their unsaturated surface atoms bound with supplementary atoms. These atoms have strong chemical activities to efficiently adsorb metal ions. Generally, Co3+ ions on the exterior are unimpregnated and bonded with several dangling bonds, which can produce catalytic active sites. Also, it is reported that the surface with extra dangling bonds possesses additional active sites and therefore has strong chemical activities.[32−34] This means that the direct nucleation, growth, and attachment of Co3O4 nanocrystals with a suitable support are responsible features in determining the catalytic behavior. Therefore, the exposed crystal facets of nanocrystals are believed to play a significant role in adjusting the overpotential for the reaction in different contexts. The present study focuses on the detection of HMIs and the effect of increased potential on voltammetry methods.[35] For Co3O4 nanocrystals, the facet-dependent electrochemical activities toward HMIs have also been reported. For example, the (111) facet of Co3O4 nanoplates is considered to be better than the (001) facet of Co3O4 nanocubes in electrochemical sensing.[36] Few studies have investigated HMI incorporation on active facets of Co3O4 nanoribbons to explore their stripping behavior by electrochemical methods.

In this work, we have tried to combine a conductive polymer polyethylenimine (PEI) with Co3O4 and rGO [from expandable graphite (EG)] to formulate an electrochemical platform for individual as well as simultaneous analysis of HMIs (Cd2+, Pb2+, Cu2+, and Hg2+) in solution by applying stripping voltammetry (SWV) and DNPV methods, respectively. During the formation of rGO·Co3O4·PEI NCP, GO from EG was simultaneously reduced to rGO along with the precipitation of dense Co3O4[26] with good electron transport property. Meanwhile, PEI was fabricated hydrothermally with less active site Co2+ to enhance the interaction between Co3O4 and rGO. Moreover, the conductive polymer PEI possesses a lot of N functional groups that provide more binding sites to detect HMIs. Most importantly, we designed a thin Co3O4 nanoribbon structure with active planes of (110) and (220)[37,38] to enhance the activity as well as the stability of Co3O4. The design strategy for the formation of a porous rGO·Co3O4·PEI NCP is given in Scheme . This porous rGO·Co3O4·PEI NCP with thin Co3O4 nanoribbons was found to be a favorable material for the electrochemical detection of HMIs.

Scheme 1

Design and Synthesis of rGO·Co3O4, rGO·PEI, and Porous rGO·Co3O4·PEI NCPs

Results and Discussion

Characterization of the Porous rGO·Co3O4·PEI NCP

Structure and Morphology of the rGO·Co3O4·PEI NCP

The methods for the preparation of rGO·Co3O4·PEI-1, rGO·Co3O4·PEI-2 NCP, and rGO·PEI NCP are summarized in the Supporting Information. The morphology of rGO·PEI, rGO·Co3O4·PEI-1, rGO·Co3O4·PEI-2, and rGO·Co3O4·PEI NCP was analyzed by transmission electron microscopy (TEM) (Figures , 2 and S1–S6), scanning electron microscopy (SEM), and energy-dispersive X-ray spectroscopy (EDS).

Figure 1

TEM and SEM images of the porous rGO·Co3O4·PEI NCP. (a,d) TEM images of rGO·Co3O4·PEI, (b,c,i) SEM images, and (e–h) EDS maps corresponding to (i).

Figure 2

TEM and HRTEM images of the porous rGO·Co3O4·PEI NCP. (a–c) TEM images of rGO·Co3O4·PEI, (d,e) HRTEM images of rGO·Co3O4·PEI, and (f) selected-area electron diffraction (SAED) pattern of rGO·Co3O4·PEI.

TEM images in Figure a illustrate a porous rGO·Co3O4·PEI NCP structure with both nanoribbons and Co3O4 NPs covered by the conducting layer of PEI. Both NPs and nanoribbons can be easily seen in Figure a,d. The dense Co3O4 NPs ranging from 18 to 47 nm covering the rGO layer can be seen in Figure S3. As shown in Figure c,e–h, the EDS mapping elucidates the occurrence of all elements C, O, Co, and N in the rGO·Co3O4·PEI NCP. The well-bonded N atom of the conducting polymer PEI in porous rGO·Co3O4·PEI NCP provides more active sites due to the heterojunction among rGO and Co3O4. Thus, it is expected to be used as an electrochemical sensor.

For rGO·Co3O4·PEI NCP, the graphene layers acting as a substrate, Co3O4, and the conducting layer of PEI in the porous rGO·Co3O4·PEI NCP are clearly seen in Figures , 2, S3, and S4. The Co3O4 NPs and thin nanoribbons are confirmed by TEM and high-resolution TEM (HRTEM) on the surface of rGO and can be clearly seen in Figures a–e and S3e,f. The pores in the rGO·Co3O4·PEI NCP can be clearly seen in Figure S3c,d. The EDS spectrum (Figure S3g) further confirms the formation of the porous rGO·Co3O4·PEI NCP with weight (%) for Co, O, C, and N of 68, 27, 3, and 1%, respectively. This indicates that rGO·Co3O4 is covered by a thin conducting layer of PEI, as presented in Figures S1–S4.

The lattice fringe spacings of ∼0.29 and 0.46–0.54 nm (Figures c–e, 3g,i, S6d–f, S3, and S4) correspond to (220) and (110) planes of the Co3O4 nanoribbon and 0.24 and 0.32 nm for (111) and (311) planes of cubic Co3O4, respectively. The thin nanoribbon Co3O4 of 6.5–11 nm width (Figure b,d) and cubic Co3O4 crystals of 24–56 nm size can be clearly indicated in Figures a,b, S3, and S4.

Figure 3

Co3O4 nanoribbon/NP crystal plane structure and HRTEM images of rGO·Co3O4·PEI. (a–c) Surface atomic configuration of Co3O4 in (220), (111), and (110) planes, (d–f) surface atomic textures in the (220), (111), and (110) planes, (g) HRTEM image illustrating the (220) and (110) planes of the Co3O4 nanoribbon/NP on rGO·Co3O4·PEI, (h) SAED pattern of the Co3O4 (111) plane on rGO·Co3O4·PEI, and (i) crystal growth along the (110) plane of the Co3O4 nanoribbon on rGO·Co3O4·PEI.

The TEM images of the porous rGO·Co3O4·PEI-1 NCP illustrate that when the Co(NO3)2 solution concentration was turned down to half, less nanoribbon/nanorod structures were obtained compared to rGO·Co3O4·PEI (Figure S4). TEM images also show a porous structure (Figure S4f). As shown in Figure S6d,f, the Co3O4 nanoribbon or nanorod shows a (220) plane. Therefore, it can be deduced that the Co3O4 nanoribbon primarily grows along the {110} or {220} direction and favorably displays the (110) or (220) planes (Figures and S6). Co3O4 has a spinel structure with Co3+ attached with six oxygen atoms in an octahedral relationship (Figure S6b) and Co2+ in a tetrahedral coordination (Figure S6c). The previous Co3+ is considered as the active site, while the last is nearly inactive.[38,40] The compact planes are (220), (111), and (110) with their surface atomic arrangements as presented in Figure a–c. It has been proved that the (111) plane comprises Co2+ cations, whereas the (220) and (110) planes are mainly imperturbable Co3+ cations. Actually, the surface differential diffraction research has proved that Co3+ cations exist merely on the (110)/(220) plane.[41−43] Further information about the nanoribbon Co3O4 flat top, atomic plane, and side plane can be found in Co3O4 nanorod formation.[44,45] The planar density for the (111) plane is 0.185 × 1020 atoms nm–2 (Figure S7), and for the (110) and (220) planes, it is 0.113 × 1020 and 0.113 × 1020 atoms nm–2, respectively.

The TEM results illustrated in Figures , 3, S3, and S4 are in accordance with X-ray diffraction (XRD) spectra. Figure f displays the SAED configuration of the porous rGO·Co3O4·PEI NCP. The diffraction rings of (220) are in agreement with the Co3O4 nanoribbon, and the (111), (311), (400), (422), (511), and (440) planes correspond to Co3O4 NPs, demonstrating the presence of Co3O4 crystals (JCPDS. no 42-1467)[46−49] in the rGO·Co3O4·PEI NCP.

Characterization of the rGO·Co3O4·PEI NCP

The crystal structures of rGO, Co3O4, and rGO·Co3O4·PEI were characterized by XRD (Figure a). The characteristic curving peaks appear at 26.59 and 44.86° attributed to the (002) and (101) planes of rGO and the peaks at 19.6, 31.24, 36.63, 44.86, 59.14, and 65.17° confirm the (111), (220), (311), (400), (511), and (440) crystalline planes of Co3O4, respectively (JCPDS no. 42-1467).[49−51] The broad curve in the XRD pattern at 23.59° is recognized as the construction of layers in the rGO·Co3O4·PEI NCP, which is in accordance with HRTEM and TEM images (Figures and 2).[52,53]

Figure 4

XRD pattern of the rGO·Co3O4·PEI NCP (a), BET and pore size distribution (b,c), FTIR spectra (d,e), and TGA curve (f).

N2 adsorption/desorption was performed to study the pore size distribution and surface areas of the porous rGO·Co3O4·PEI NCP (Figure b,c). The porous rGO·Co3O4·PEI NCPs possess a BET surface area of 261.15 cm3 g–1 with a total pore volume of 0.612 cm3 g–1. As illustrated in Figure c, the average pore size of the porous rGO·Co3O4·PEI NCP was 9.4 nm, in close resemblance with the reports of t N-doped macroporous and mesoporous graphene-based metal oxides and N-doped PC NCPs.[54,55]

The Fourier-transform infrared (FTIR) spectra of rGO and rGO·Co3O4·PEI NCP (Figure d,e) were recorded to additionally explore the property of the porous rGO·Co3O4·PEI NCP. The widespread absorption peaks at 3000–3700 cm–1 are ascribed to the −NH and −OH groups in the rGO·Co3O4·PEI NCP. Similarly, the peak at 1625 cm–1 is assigned to the −OH bending vibration of absorbed water molecules, and the involvement of aromatic C=C (1218 and 1225 cm–1) further confirms the successful reduction of GO.[56] The peaks at 567 and 661 cm–1 are allotted to Co–O vibrations. The FTIR study additionally confirms that Co3O4 NPs are effectively deposited on the surfaces of rGO.[57−61]

Figures f and S8 show the thermogravimetric analysis (TGA) curve of rGO·Co3O4·PEI NCP. The weight loss observed below 112 °C is generally recognized to be due to dehydration or drying out of the NCP. The additional weight loss phase from 112 to 407 °C is due to the decomposition of the rGO·Co3O4·PEI NCP, and the temperature range after almost 500 °C might be ascribed to Co3O4. On the basis of the TGA curve, the loss in the weight ratio of rGO·Co3O4·PEI NCPs of about 40 wt % is almost similar to the result reported[62] for the formation of the Co3O4–carbon composite.

Electrochemical Characteristics of rGO and rGO·Co3O4·PEI NCP

Cyclic Voltammetry, Electrochemical Impedance Spectroscopy, and Mott–Schottky Measurements of the rGO·Co3O4·PEI NCP

The cyclic voltammetry (CV), electrochemical impedance spectroscopy (EIS), and Mott–Schottky (MS) measurements of the rGO·Co3O4·PEI-modified electrodes were carried out in a solution of 5 mM Fe(CN)63–/4–, 1 M KCl using the Fe(CN)63–/4– redox couple, and —1 M NaAc–Hac was used to adjust the pH to 5.0. The results are shown in Figure .

Figure 5

CV curves (a), EIS curves (b,c), and MS plots (d) of rGO·Co3O4·PEI NCP (5 mM Fe(CN)63–/4–, 1 M KCl, 1 M NaAc–HAc, pH 5 ± 02, 0.1 mV s–1).

As shown in Figure S11a, both the anodic (0.27 V) and cathodic (0.18 V) peaks of GO are less compressed than those of rGO, rGO·PEI, and rGO·Co3O4-modified electrodes. The electrochemical gaps are adjusted to the spots of the redox couples. It can be sensed that at the anodic current curve of the rGO·Co3O4-modified electrode, there are two redox couples from 0.2 to 0.6 V/s. They are sited at 0.27/0.56 V/s and 0.63/0.45 V/s, respectively. As presented in Figure a, both the anodic (0.41 V) and cathodic (0.02 V) peak of GCE are less thick/broad than those of the rGO·Co3O4·PEI-modified electrode. The peaks around 0.29 (I) and 0.57 V (III) in the first cycle are ascribed to the decline of Co3O4 to Co along with the construction of a solid electrolyte edge as well as the development of a solid electrolyte connection reaction on the exterior of the rGO·Co3O4·PEI NCP (same as that reported[63,64] for the Co3O4 nanowall in Li-ion batteries). At the anodic current curve, two redox couples are detected from 0.2 to 0.6 V/s. They are positioned at 0.22/0.47 V and 0.57/0.647 V, respectively, arising due to the reversible conversion attributed to Co3O4 and CoOOH(II/III) and similar reversible conversions between CoOOH and CoO2(III/IV).[57,62,65] The peaks at 0.72 (V) and 0.29 V (I) in the first cycle are ascribed to the depletion of Co3O4 to Co, along with the development of a solid electrolyte edge on the exterior of rGO·Co3O4·PEI, with 0.65μA current greater than GCE, as shown in Figure a.[65]

The porous PEI-functionalized Co3O4/rGO NCPs were characterized by X-ray photoelectron spectroscopy (XPS) spectrum as shown in Figure S13. The strong peaks at 287.9, 400, 538.2, and 783.2 eV analogous to the distinct peaks of C 1s, N 1s, O 1s, and Co 2p indicate the presence of C, N, O, and Co elements, respectively, in the sample. The Co 2p1/2 and Co 2p3/2 spectra can be deconvoluted into four peaks at 794.8 and 793.2 eV for Co 2p1/2, 782.7 and 781.4 eV for Co 2p3/2, showing the characteristic of the Co3O4 phase (Figure S13b). Similarly, for co-doping of graphene with Co3O4 and PEI O 1s and N 1s spectrum linkage and interaction with graphene and Co3O4 shown in Figure S13c,d confirm rGO·Co3O4·PEI NCP.

The EIS experiments were performed between −0.2 and 0.6 V at room temperature, as illustrated in Figure b,c. Figure b shows Nyquist plots of GCE and rGO·Co3O4·PEI (Figure c). It might be perceived that the rGO·Co3O4·PEI electrode has the smallest diameter (825 Ω), compared to GO (2868 Ω), GCE (1271 Ω), and rGO·Co3O4 (1849 Ω) (Figure S11b), signifying that the rGO·Co3O4·PEI-modified electrode possesses a smaller Rct value than GCE (Table S1). The results show that the electrochemical reaction on the surface of anode/electrolyte for rGO·Co3O4·PEI is more prominent than that on the GCE electrode, which is almost the same as reported by ref (57) for the Co3O4/rGO-0.50 electrode. Meanwhile, the film resistance and resistance of charge transfer (RS and Rct) of the rGO·Co3O4·PEI NCP are lower than those of GCE (Table S1). Further suitable results are displayed in Table S1, indicating that the RS (135.6 Ω) and Rct (419.7 Ω) of the rGO·Co3O4·PEI NCP-modified electrode are lesser in comparison with those of the pure GCE electrode (193 and 1070 Ω, respectively).

The rGO·Co3O4·PEI NCP exhibited a negative (adverse) slope for the MS plot and hence exhibit a p-type semi-conducting performance, as illustrated in Figure d and Table S2; the carrier density (Nds) of rGO·Co3O4·PEI NCP was calculated as 1.56 × 1018 (with a quiet time of 2 s, an amplitude of 0.005 V, and an increased potential of 0.05 V).

SWV and DNPV Voltammetry Analyses Intended for the Sensing/Detection of Cd2+, Pb2+, Cu2+, and Hg2+ by Means of the rGO·Co3O4·PEI-Modified Electrode

SWV and DNPV electrochemical analyses were used to detect the four HMIs (Cd2+, Pb2+, Cu2+, and Hg2+). Figure a,b illustrates the SWV and DNPV peaks from 1 to 10 μM concentration of Cd2+, Pb2+, Cu2+, and Hg2+ aqueous solutions with an increased potential of 7 mV for SWV and 1 mV for DNPV at pH 5 ± 02. For the stripping analysis of HMIs using both SWV and DNPV, the obtained results in the form of voltammetric peaks are shown in Figure a,b, and the rGO·Co3O4·PEI NCP was subject to the simultaneous investigation of HMIs. The vertices at almost −0.79, −0.58, −0.22, and 0.11 V are assigned to Cd2+, Pb2+, Cu2+, and Hg2+ on SWV, respectively (Figure a); similarly, the peaks at almost −0.801, −0.573, −0.162, and 0.114 V are allocated to Cd2+, Pb2+, Cu2+, and Hg2+, respectively, on DNPV analysis. The peaks at 0.02 V for SWV and −0.35 V for DNPV between Cu2+ and Hg2+ (Figure a,b) are assigned to the occurrence of inter-metallic interactions.[66,67]

Figure 6

SWV and DNPV results of rGO·Co3O4·PEI NCP, (a) SWV curves and (b) DNPV curves from 1 to 10 μM with an increased potential of 0.007 V (7 mV) for SWV and an increased potential of 0.001 V (1 mV) for DNPV, (c,d) increased potential of 0.001–0.007 V for SWV at 10 and 7 μM, respectively (pH, 5 ± 02).

Figure c shows the increased potential from 0.001 to 0.007 V for the SWV analysis at 10 μM concentrations of four HMIs (Cd2+, Pb2+, Cu2+, and Hg2+). Similarly, Figure d illustrates the increased potential of 0.001–0.004 V for SWV at 7 μM concentration. The peaks at an increased potential of 0.007 V (7 mV) are recommended to be the best for the 1–10 μM concentration of Cd2+, Pb2+, Cu2+, and Hg2+ for both individual (Figure S9) and simultaneous analyses of SWV (Figure c,d). Figure S9 illustrates the individual analysis of the rGO·Co3O4·PEI NCP for SWV. Figure S9a illustrates it for Cu2+ (1 μM) with an increased potential from 0.001 to 0.007 V (1 mV) and Figure S9b illustrates the same for Cu2+ (3 μM) with an increased potential from 0.007 (7 mV) to 0.049 V, demonstrating that 7 mV potential is the best for the individual measurement of Cd2+, Pb2+, Cu2+, and Hg2+. The sensing performance of rGO, rGO·PEI, and rGO·Co3O4 NCPs was satisfactory but not strong enough as that of the rGO·Co3O4·PEI NCP (Figure S12).

To check the increased potential effect on voltammetric peaks, the potential was varied from 1 to 7 mV; when the potential was increased from 1 to 5 mV, the intensity of voltammetry peaks increased a little. However, the voltammetry spectrum becomes better than before at an increased potential of 7 mV. Therefore, an increased potential of 7 mV is recommended to be the best for the simultaneous sensing/detection of Cd2+, Pb2+, Cu2+, and Hg2+ on SWV, as shown in Figure c,d.

Increased Potential Effect on the Voltammetry Spectrum

Figure shows the increased potential effect on SWV peaks for both SWV and DNPV analyses. The increased potential effect from 0.001 to 0.005 V for DNPV at 10 and 5 μM for simultaneous analysis is shown in Figure a,b. When the potential was increased from 1 to 5 mV, the intermetallic interaction increased, and almost three new voltammetry peaks at −0.47, −0.23, and 0.001 V were observed, which is shown in pink with other parameters in Figure a,b. Figure c,d illustrates the effect of the increased potential of 4–7 mV for SWV at 5 μM concentration and 1 and 3 mV for SWV at 3 μM concentration. The vertices at −0.799, −0.599, −0.23, and 0.11 V are attributed to Cd2+, Pb2+, Cu2+, and Hg2+, respectively; the intensity of voltammetric peaks increases with the increase in potential from 1 to 7 mV (Figure c,d). However, two stronger intermetallic interactions between Pb2+–Cu2+[68] and Cu2+–Hg2+[69] at −0.49 and 0.011 V were also observed for SWV measurements. Similarly, the less intense peaks arise at 0.02 V for SWV and −0.35 V for DNPV between Cu2+ and Hg2+ (Figure a,b), which are due to the occurrence of inter-metallic interactions.[69] Due to stronger intermetallic interactions, the DNPV voltammetry method was not proceeded further. However, at an increased potential of 1 mV for simultaneous DNPV analysis, the limit of detection (LOD) and sensitivity were found to be good (Table S4).

Figure 7

Voltammetric curves (a,b) for the increased potential of 0.001–0.007 V for DNPV at 10 μM and 5 μM concentrations, respectively, for simultaneous analysis, (c) for the increased potential of 0.004–0.007 V for SWV at 5 μM concentration, and (d) for the increased potential of 0.001 V (1 mV) and 0.003 V (3 mV) carried out on SWV at 3 μM concentration.

Stripping Analysis of Pb2+, Cd2+, Hg2+, and Cu2+

Individual Depriving Performance Analysis for Cd2+, Cu2+, Hg2+, and Pb2+ Using SWV

The SWV response of the rGO·Co3O4·PEI-modified electrode (GCE), linearization equations, and calibration curves for the individual voltammetry examination of Cd2+, Pb2+, Cu2+, and Hg2+ from 1 to 10 μM concentration are presented in Figure a–d and Table S5.

Figure 8

SWV response of the rGO·Co3O4·PEI NCP-modified electrode and calibration curves for the individual investigation of (a) Cu2+, (b) Hg2+, (c) Pb2+, and (d) Cd2+ in the 1–10 μM concentration range, while other parameters were the same as in Figure in acetate buffer (0.1 M).

The respective calibration curves of rGO·Co3O4·PEI for the simultaneous sensing/detection of Cd2+, Pb2+, Cu2+, and Hg2+ corresponding to Figure a are shown in Figure S10 and Table S3. For Cd2+, Pb2+, Cu2+, and Hg2+, the inevitable LODs were 0.285, 1.13, 1.19, and 1.29 nM for SWV (Table S3); similarly, they were 1.07, 0.285, 2.40, and 1.12 nM for DNPV (Table S4) for simultaneous detection, whereas they were 0.484, 0.878, 0.462, and 0.477 nM for SWV (Table S5) for individual analysis of Cd2+, Pb2+, Cu2+, and Hg2+. The linearization equation and the calculation of the corresponding coefficients for concurrent determination of Cd2+, Pb2+, Cu2+, and Hg2+ are enumerated in Tables S6 and S7.

As shown in Figure S10, the corresponding calibration curves for Cd2+, Pb2+, Cu2+, and Hg2+ for simultaneous analysis by SWV were built from 1 up to 10 μM. Similarly, the corresponding calibration curves for the four HMIs (inset in Figure and Table S6) represent the linearization equations with comparable correlation coefficients of 0.99, 0.97, 0.99, and 0.98 for individual analysis by SWV when the rGO·Co3O4·PEI NCP was employed. The LODs for simultaneous sensing/detection of Cd2+, Pb2+, Cu2+, and Hg2+ with the use of the rGO·Co3O4·PEImodified electrode were estimated to be 0.285, 1.13, 1.19, and 1.29 nM for SWV (Table S3); similarly, they were 1.07, 0.285, 2.40, and 1.12 nM for DNPV (Tables S4 and S7) during simultaneous detection (3σ method), respectively. The fallouts acquired by the use of the rGO·Co3O4·PEI NCP for LOD in both DNPV and SWV voltammetry analyses were found to be inferior to the values recommended by the World Health Organization (WHO) (see Table S8).

Sensitivity and LOD Calculation of the rGO·Co3O4·PEI NCP

For individual analysis with the SWV voltammetric method, the sensitivities of the rGO·Co3O4·PEI-modified electrode calculated for Cd2+, Pb2+, Cu2+, and Hg2+ are 3.50, 9.60, 13.9, and 14.6 μA/μM (Table S9). The lowest detectable (LOD) concentrations of Cd2+, Pb2+, Cu2+, and Hg2+ for rGO·Co3O4·PEI reach 0.484, 0.878, 0.462, and 0.477 nM, respectively. For simultaneous analysis by SWV, the sensitivities of the rGO·Co3O4·PEI-modified electrode calculated for Cd2+, Pb2+, Cu2+, and Hg2+ are 1.30, 2.70, 4.50, and 6.30 μA/μM, respectively, while the lowest measurable concentrations of Cd2+, Pb2+, Cu2+, and Hg2+ by rGO·Co3O4·PEI are 0.285, 1.130, 1.190 and 1.29 nM, respectively. Similarly, for simultaneous analysis by DNPV, the sensitivities of the rGO·Co3O4·PEI-modified electrode calculated for Cd2+, Pb2+, Cu2+, and Hg2+ are 1.20, 1.20, 2.90, and 2.60 μA/μM, respectively; while the lowest measurable concentrations of Cd2+, Pb2+, Cu2+, and Hg2+ detected by rGO·Co3O4·PEI are calculated as 1.07, 0.285, 2.40, and 1.12 nM, respectively.

A comparison of the rGO·Co3O4·PEI NCP with previous work for the sensing/detection of Cd2+, Pb2+, Cu2+, and Hg2+ is shown in Table S10. It can be illustrated from the present work that the porous rGO·Co3O4·PEI-modified electrode with a PEI thin cover on the exterior of the rGO·Co3O4 and Co3O4 nanoribbon structure has improved the sensitivity for both individual analysis and simultaneous analysis of Cd2+, Pb2+, Cu2+, and Hg2+ in 1–10 μM concentration. The HRTEM images reveal that the Co3O4 nanoribbons mainly have (110) and (220) planes, preferring the occurrence of active Co3+ species on the surface. This might interact with HMIs directly, and the Co2+ plane will interact with the PEI thin film. The mechanism for the detection of HMIs by using rGO·Co3O4·PEI is shown in Scheme . For the porous rGO·Co3O4·PEI NCP, the conducting polymer PEI on the surface of Co3O4 nanoribbons and rGO (from EG) possessing a lot of N functional groups can provide more binding sites to bond with the objective HMIs and hence improve the electrochemical performance. Also, the Co3O4 nanoribbons with active (110) and (220) planes possess exclusive properties such as high electrical conductivity and chemical sensitivity and the ability to adsorb HMIs. The Co3O4 nanoribbons and NPs well dispersed on thin-layer rGO nanosheets can effectively buffer the disturbance initiated by structural and volume variations of Co3O4 nanoribbons and NPs, which might efficiently relieve or prevent Co3O4 NPs from aggregation. Thus, the design of the rGO·Co3O4·PEI NCP is one of the impressive ways to carry out the electrochemical routine of Co3O4.

Scheme 2

Mechanism of Detection of HMIs by Using the rGO·Co3O4·PEI NCP; (a) Interaction of PEI with Co2+ on rGO·Co3O4·PEI, (b) Co3O4 NP Interaction with HMIs on rGO·Co3O4·PEI, and (c) Detection of HMIs and the Interaction between PEI, rGO, and Co3O4 with HMIs

Furthermore, the porous structure of the rGO·Co3O4·PEI NCP can allow the solution of HMIs to easily accumulate the target metal ions on the electrode surface, thereby showing excellent sensing performance. This new nanomaterial will be helpful in both laboratory and industrial scale applications.

Conclusions

In the present work, we tried to formulate an electrochemical method by combining the conducting polymers PEI and Co3O4 nanoribbons/NPs with rGO for the analysis of four selected (Cd2+, Pb2+, Cu2+, and Hg2+) HMIs in a solution by SWV and DNPV for the first time. For the SWV method, by simultaneous analysis (3σ method) LODs of Cd2+, Pb2+, Cu2+, and Hg2+ using the rGO·Co3O4·PEI NCP-modified electrode are 0.285, 1.132, 1.194 and 1.293 nM, respectively; similarly, LODs are 1.069, 0.285, 2.398, and 1.115 nM for DNPV, , respectively whereas they are 0.484, 0.878, 0.462, and 0.477 for SWV by individual analysis of Cd2+, Pb2+, Cu2+, and Hg2+, respectively.

The enhanced electrochemical performance can be ascribed to three factors: (1) rGO could be used to prevent the aggregation of Co3O4 nanoribbons/NPs, resulting in fast migration of electrons; (2) PEI might prevent the aggregation of Co3O4 nanoribbons/NPs and provide more binding sites to bond with the target HMIs as they possess a lot of N functional groups; (3) the Co3O4 nanoribbons with active (110) and (220) planes retain the chemical sensitivity and adsorb HMIs. The approach developed in the present work provides a novel route for low-cost and comprehensive porous rGO·Co3O4·PEI with highly promising applications in heavy metal detection.

Experimental Section

Chemicals

All the chemicals used in this study were of analytical grade and used without further purification. Cobalt nitrate hexahydrate [Co(NO3)2·6H2O, 99%] and sodium hydroxide (NaOH, 96%) were provided by Xilong Chemical Co., Ltd. Hydrochloric acid (HCl, 37%) was obtained from Beijing Chemical Works. Branched PEI (MW = 600, 99%) was bought from Alfa Aesar Chemical Co. Ltd. (Tianjin, China). Lead nitrate [Pb(NO3)2], cadmium nitrate tetrahydrate [Cd(NO3)2·4H2O], copper nitrate trihydrate [Cu(NO3)2·3H2O], mercury nitrate [Hg(NO3)2], acetic acid, and sodium acetate were provided by Tianjin Chemical Works. EG of 300 mesh used in the preparation of rGO·PEI and rGO·Co3O4·PEI NCP was obtained from Qingdao Tianyuan Chemical Co., Ltd. Concentrated sulfuric acid (H2SO4, 98%) was purchased from Yaohua Chemical Reagent Co. Ltd. (Tianjin, China).

Synthesis of the rGO·Co3O4·PEI NCP

The thin-sheet-like structure of EG was attained by heating expandable graphite in a microwave oven. Graphite oxide was prepared by the Hummers method.[39] Then, 20 mL of PEI (2 g/L) and 80 mL of EG solution (0.03 g/L) were mixed with deionized (DI) water (100 mL) under magnetic stirring, the pH was adjusted (9.0 ± 0.2), and the solution was allowed to stand for 1 h; then it was sonicated for 0.5 h, and 20 mL of Co(NO3)2 solution (2 g/L) was added dropwise to the solution. During the mixing process, the pH of the solution was kept at 9.0 ± 0.2. The flow rate of air was maintained as 50 mL/min for 20 min. Then, the pH of the solution (12.0 ± 0.2) was balanced by the gradual addition of NaOH solution and the air flow was maintained for 2 h. After the completion of aeration, the solution was further sonicated for 0.5 h followed by pH adjustment (12 ± 0.2). Finally, the mixed precursor solution was placed for 24 h at room temperature (RT), the precipitate was filtered and washed with DI water; the pH was adjusted to about 7. The obtained precipitates were dispersed in DI water and transferred to a 100 mL Teflon-lined stainless steel autoclave. The hydrothermal method was done at 190 °C for 5 h. The obtained rGO·Co3O4·PEI NCP was named rGO·Co3O4·PEI. When the Co(NO3)2 solution concentration was 30 and 40 mL, the obtained NCPs were named rGO·Co3O4·PEI-1 and rGO·Co3O4·PEI-2, respectively.

Characterizations

The morphological and surface compositions for NCPs were studied by SEM (HITACHI S-4800), EDS, and TEM (JEOL-2100) analyses. X-ray powder diffraction (XRD; D/Max-IIIB-40 kV, Japan, Cu Kα radiation, λ = 1.5406 Å) analysis was used to define the crystalline structure and the composition of the assembled composites. FTIR measurements (PerkinElmer spectrometer, KBr pellet technique) and Raman spectroscopy (Jobin Yvon HR 800 micro-Raman spectrometer) were also performed. The surface composition study was done by XPS(AXIS ULPRA DLD Shimadzu Corporation). The porous PEI-functionalized Co3O4/rGO NCPs were characterized by XPS using an Mg Kα X-ray source (hν = 1.254 MeV) and a 120 mm arched electron energy analyzer, with an energy resolution of ∼0.8 eV. Thermogravimetric (TG) quantification was achieved with a Shimadzu TGA-50.

For electrochemical experiments, a traditional three-electrode system electrochemical workstation (Beijing Company CHI660D) was used. The Ag/AgCl electrode (no 218 ± 3 potential) purchased from the Shanghai Branch of the Inst. Eq. Company and platinum (Pt J110, with the snap width below 1 mm) made at Tianjin Ida Heng Sheng Tech. Dev. Co., Ltd. were used as the reference and auxiliary electrodes, respectively. SWV and DNPV electrochemical analyses were conducted to detect the four HMIs (Cd2+, Pb2+, Cu2+, and Hg2+). Both SWV and DNPV were performed with a three-electrode cell using a CHI660D electrochemical workstation.

The GCE coated with rGO·Co3O4·PEI was used as the working electrode for EIS, CV, MS measurements, and SWV experiments. The CV responses of rGO·Co3O4·PEI-modified electrodes were studied in Fe(CN)63–/4– (5 mM) solution comprising KCl (1 M) using the Fe(CN)63–/4– redox couple, and the pH was attuned to 5 ± 02 with acetate buffer solution (CH3COOH/CH3COONa). The EIS experiments were carried out between −0.2 and 0.6 V at RT with Ag/AgCl as the reference electrode and a Pt electrode as the counter electrode. EIS experiments were completed with an initial potential of 0.294 V in the frequency range of 1 Hz to 0.1 μHz.shikeying2008@163.com