Ultrasonic Assisted Synthesis of Size-Controlled Cu-Metal-Organic Framework Decorated Graphene Oxide Composite: Sustainable Electrocatalyst for the Trace-Level Determination of Nitrite in Environmental Water Samples.

Excess levels of nitrite ion in drinking water interact with amine functionalized compounds to form carcinogenic nitrosamines, which cause stomach cancer. Thus, it is indispensable to develop a simple protocol to detect nitrite. In this paper, a Cu-metal-organic framework (Cu-MOF) with graphene oxide (GO) composite was synthesized by ultrasonication followed by solvothermal method and then fabricated on a glassy carbon (GC) electrode for the sensitive and selective determination of nitrite contamination. The SEM image of the synthesized Cu-MOF showed colloidosome-like structure with an average size of 8 μm. Interestingly, the Cu-MOF-GO composite synthesized by ultrasonic irradiation followed by solvothermal process produce controlled size of 3 μm colloidosome-like structure. This was attributed to the formation of an exfoliated sheet-like structure of GO by ultrasonication in addition to the obvious influence of GO providing the oxygen functional groups as a nucleation node for size-controlled growth. On the other hand, the composite prepared without ultrasonication exhibited 6.6 μm size agglomerated colloidosome-like structures, indicating the crucial role of ultrasonication for the formation of size-controlled composites. XPS results confirmed the presence of Cu(II) in the as-synthesized Cu-MOF-GO based on the binding energies at 935.5 eV for Cu 2p3/2 and 955.4 eV for Cu 2p1/2. The electrochemical impedance studies in [Fe(CN)6]3-/4- redox couple at the composite fabricated electrode exhibited more facile electron transfer than that with Cu-MOF and GO modified electrodes, which helped to utilize Cu-MOF-GO for trace level determination of nitrite in environmental effluent samples. The Cu-MOF-GO fabricated electrode offered a superior sensitive platform for nitrite determination than the Cu-MOF and GO modified electrodes demonstrating oxidation at less positive potential with enhanced oxidation current. The present sensor detects nitrite in the concentration range of 1 × 10-8 to 1 × 10-4 M with the lowest limit of detection (LOD) of 1.47 nM (S/N = 3). Finally, the present Cu-MOF-GO electrode was successfully exploited for nitrite ion determination in lake and dye contaminated water samples.

Introduction

Nitrite has been well-recognized as a critical “hypoxic buffer”, and it is potentially regulating the hypoxic vasodilatation, mitochondrial respiration, modulation of ischemia–reperfusion tissue injury, and obstruction.[1] According to WHO, 3 mg/L nitrite is the maximum allowable level in drinking water.[2] Therefore, drinking water containing more than the permissible limit of nitrite leads to several complications, which include spontaneous abortions, intrauterine growth restriction, and premature birth.[3,4] The excess of nitrite remarkably reduces the blood capacity to transport oxygen from irreversible conversion of oxyhemoglobin to methemoglobin and subsequently leads to methemoglobinemia (blue baby syndrome).[1,3,4] In addition, nitrite interacts with compounds containing amine functional groups to form carcinogenic nitrosamines, which cause stomach cancer,[5−7] and hence, it is very much essential to determine nitrite concentrations in ultratrace level. Gas chromatography,[8] fluorescence spectroscopy, HPLC,[9] electrochemical methods,[10−16] and Raman spectroscopy[17] have been frequently used to detect nitrite. Electrochemical methods are ideal over other methods due to their superior sensitivity, excellent selectivity, and easy operation. Though nitrite can be identified by its electro-oxidation as well as reduction, the former is highly favored in regular applications since easily reducible nitrate and molecular oxygen interfere in the nitrite reduction.[18] Nitrite is electroactive, but it cannot be detected by bare glassy carbon (GC) electrodes. This is due to the adsorption of the oxidation products and intermediates on the electrode surface.[12,16−20] Moreover, the requirement of high overpotential for its oxidation at a bare electrode encourages researchers to modify the electrode with different materials.[11−18]

Metal–organic frameworks (MOFs) are highly crystalline materials. Inorganic units (metal ions or metal clusters) and organic ligands bond through covalent or coordinate bonds to form MOFs.[21−25] Recently, MOFs have attained much consideration because of their remarkable surface area, high thermal stability, and large pore volume.[23−28] Research on MOFs has attracted much attention due to their diverse and tunable porosities.[22] The selection of metal ions, linkers, solvents, and synthetic procedures tunes the MOF with controlled size and shape.[22] MOFs have been widely applied in catalysis, energy and gas storage, drug delivery, adsorption, and sensors.[21−31] However, their solubility, chemical stability, and electrical conductivity significantly hamper the full potential of MOFs. MOFs synthesized by the solvothermal method with diverse morphology cannot meet the requirements of a good electrocatalyst. In order to overcome these problems, MOF composites with metal nanoparticles and carbon nanomaterials have been prepared.[21,23,31] Among the different carbon-based materials, graphene oxide (GO) has attracted increasing attention in recent years.[21,32] GO, a 2D amphiphilic material, contains many hydroxyl, epoxy, and carboxylic acid containing functional groups on its conjugated planes and along the sheet edges.[32] The coexistence of aromatic sp2 features and oxygen containing functionalities allow GO to participate in a wide range of interactions. Due to the solubility and open sheet-like structure of GO, the basal plane and also the edges of the GO sheet can be functionalized.[33] The decoration of the GO sheet with various functional groups on both sides can offer a novel polyaromatic platform in its performance in chemistry as well as a hybrid 2D-nanobar building block, which can perform in supramolecular assembly.[21−23,33] A composite material formed with GO could have enhanced optical, electrical, thermal, and mechanical properties.[21,33−36] Few reports on MOF–carbon nanomaterial composite based sensors for nitrite are available in the literature.[11,12,14] Saraf et al. fabricated a Cu-MOF/rGO electrode for nitrite sensing in a wide concentration range with a limit of detection (LOD) of 33 nM.[11] Kung et al. achieved a LOD of 2.1 μM for nitrite using zirconium-based MOF-525 as electrode material.[12] Wang et al. reported a nitrite sensor based on an α-Fe2O3 polyhedral electrocatalyst derived from CNT–MOF with 0.15 μM LOD and 334 μA mM–1 cm–2 sensitivity.[14] Although the aforementioned reports showed the determination of nitrite with good sensitivity and selectivity, the preparation of the composite material is difficult and the electrodes prepared by carbon paste are relatively unstable with poor reproducibility. Thus, it is crucial to develop a simple method to prepare a composite material for the stable determination of nitrite. The ultrasonication approach for synthesizing nanomaterials is considered as a green and environmentally benign approach and has received much attention recently.[37,38]

This study reports the synthesis of a Cu-MOF–GO composite by ultrasonication followed by solvothermal treatment. The composite prepared without ultrasonication of GO leads to the formation 6.6 μm colloidosome-like Cu-MOF in contrast to 8 μm Cu-MOF in the absence of GO. Interestingly, the size of the colloidosome-like structure was reduced from 6.6 to 3 μm when the GO was ultrasonicated. The decreased in particle size is due to the interaction of oxygen functional groups on GO with Cu-MOF followed by molecular collision induced by ultrasonication. The composite material was then fabricated with a GC electrode and effectively exploited for nitrite determination with high sensitivity and selectivity. It exhibited excellent electrocatalytic activity by showing nitrite oxidation with 100 mV lower overpotential and 2-fold oxidation current enhancement in comparison to the bare electrode. The detection of a wide range of nitrite concentration from 10 nM to 0.1 mM was achieved by amperometry with 1.47 nM (S/N = 3) LOD.

Results and Discussion

Structural Analysis of Cu-MOF–GO Composite

The formation of Cu-MOF and its composite with GO was initially confirmed by FT-IR spectroscopy, and the spectra of azelaic acid (AZA), Cu-MOF, GO, and Cu-MOF–GO are shown in Figure . The obtained vibrational frequencies and their assignments are given in Table S1.

Figure 1

FT-IR spectra of (a) AZA, (b) Cu-MOF, (c) GO, and (d) Cu-MOF−GO.

AZA exhibits peaks at 677, 724, 1086, 1420, 1703, 2936, and 3440 cm–1 (Figure , curve a). The peak appearing at 677 cm–1 corresponds to the −C–H bending vibration.[39,40] The peak appearing at 724 cm–1 was ascribed to the −CH bending vibration in the outside plane. The sharp peak at 1086 cm–1 was due to the −C–O– stretching vibration.[40] The broad peak appearing at 1420 cm–1 corresponds to the −OH bending vibration. The intense peak at 1703 cm–1 was due to the stretching vibration of −C=O.[39] The peak appearing at 2936 cm–1 was ascribed to the −CH stretching vibration. The broad peak appearing at 3440 cm–1 corresponds to the −OH stretching vibration.[39−41] The FT-IR spectrum of Cu-MOF shows peaks at 614, 725, 1101, 1436, 1586, 2847, 2929, and 3448 cm–1 (Figure , curve b). The peak at 614 cm–1 corresponds to the Cu–O stretching frequency,[42] whereas the peak appearing at 725 cm–1 was due to the −CH bending vibration.[40,41] The peak appearing at 1101 cm–1 corresponds to the stretching vibration of −C–O. This suggests that the −C–O functional group was involved in the formation of the network structure. The −OH bending vibration at 1420 cm–1 was shifted (curve a) to 1436 cm–1 (curve b) due to the interaction between the carboxylic functional group and the metal center. Further, the formation of Cu-MOF was confirmed by the remarkable peak shifting of the −C=O functional group at 1586 cm–1 (curve b) from 1703 cm–1 (curve a). The observed peaks at 2847 and 2929 cm–1 were attributed to the stretching vibration of the aliphatic C–H group.[39,40] The −OH stretching frequency was shifted from 3440 to 3448 cm–1, and also the peak becomes sharp compared to AZA. GO exhibited broad peaks in its spectrum at 1090, 1387, 1634, and 3459 cm–1 of the −C–O– stretching, −OH bending, −C=O stretching, and −OH stretching vibrations, respectively (Figure , curve c).[43] Finally, the composite material shows peaks at 626, 716, 1107, 1422, 1574, 2848, 2922, and 3432 cm–1 in the FT-IR spectrum (Figure , curve d). The obtained bands were similar to those of Cu-MOF with slight shift, indicating the interaction between GO and Cu-MOF. The shifting of a peak from 1574 to 1586 cm–1 for Cu-MOF suggests the interaction of the metal ion with the carboxylic functional group of GO. Further, the interaction was also understood from the shifting of the hydroxyl functional group from 3448 to 3432 cm–1. The obtained FT-IR results clearly reveal the successful formation of Cu-MOF–GO composite material.

XRD studies were carried out to determine the crystallinity of the synthesized Cu-MOF and its composite with GO and are shown in Figure . The XRD pattern of Cu-MOF shows peaks at 10.97°, 12.76°, 13.53°, 14.78°, 18.59°, 22.58°, 26.26°, 29.74°, 36.57°, and 42.27° corresponding to (2 2 0), (2 2 2), (4 0 0), (4 2 0), (4 4 0), (5 5 1), (7 3 1), (7 5 1), (7 7 3), and (8 8 2) planes (curve a). The obtained peaks were well consistent with the similarly reported patterns of Cu-MOF, suggesting the successful formation of Cu-MOF.[12,17] On the other hand, GO exhibits a single sharp 2θ peak at 12.15° due to (0 0 2) reflection planes of the graphitic carbon surface[43] (curve b). When Cu-MOF and GO are reacted together to form a Cu-MOF–GO composite, it exhibits an XRD pattern quite similar to that of Cu-MOF, which indicates the existence of well-defined MOF units in composite (curve c). However, it is interesting to notice that the relative intensity of Cu-MOF–GO is less than that of Cu-MOF, indicating that the more ordered crystalline structure of Cu-MOF was affected by the presence of GO. It is likely that GO can function as an efficient nucleation site for the growth of Cu-MOF, and also the presence of a sharp peak at 2θ = 12.15° confirmed that the composites have GO in their assembly, not reduced GO (r-GO).

Figure 2

XRD patterns of (a) Cu-MOF, (b) GO, and (c) Cu-MOF–GO.

Further, the composition of Cu-MOF with GO was confirmed by Raman spectroscopy (Figure S1). The Raman peaks appearing at 276 and 496 cm–1 were attributed to Cu–O.[44,45] The intense peaks exhibited at 746 and 827 cm–1 were due to the −CH bending vibrations of both ligand and GO. The less intense peaks at 1225 and 1423 cm–1 were responsible for the −C=C stretching vibration of the phenyl ring on the GO surface.[44] The peaks at 1456 and 1546 cm–1 were due to the stretching of the carboxylic functional group, which were assigned to symmetric and asymmetric −COO–, respectively.[44,45] However, the D and G bands of GO were overlapped with Cu-MOF. These results clearly suggest the successful formation of Cu-MOF–GO composite.

XPS is the most important analytical tool to reveal the presence of different oxidation states of elements in the synthesized compound. The survey spectrum of the synthesized Cu-MOF–GO exhibits three different binding energies at 284.6, 531.2, and 965.2 eV corresponding to carbon, oxygen, and copper (Figure A), and the individual regions were deconvoluted to study their binding in the composite matter. The less intense peak appearing at 566.4 eV was attributed to the surface oxygen from hydroxyl, epoxy, and carboxylic functional groups. Figure B shows the deconvoluted spectrum of the C 1s region, which exhibits two major peaks at 285.5 and 288.7 eV of C–C and −COO–, respectively.[40,46] The deconvoluted O 1s region displays the binding energy at 532.2 eV due to the graphitic lattice oxygen (Figure C),[40,46] and the deconvoluted spectrum of copper shows two major peaks at 935.5 and 955.4 eV attributed to the Cu 2p3/2 and Cu 2p1/2, respectively, along with two satellite peaks at 945.2 and 963.15 eV, and the obtained satellite peaks confirmed that Cu is present as Cu2+ in the composite (Figure D).[46]

Figure 3

XPS of Cu-MOF–GO on GC substrate. (A) Survey and (B–D) deconvoluted spectra from (B) C 1s, (C) O 1s, and (D) Cu 2p regions.

Morphological Analysis of Cu-MOF–GO Composite

Effect of Ultrasonication

The SEM images of the synthesized Cu-MOF, GO, and Cu-MOF–GO at different magnifications are shown in Figure . The morphology of the Cu-MOF looks like a colloidosome structure, and the MOFs uniformly covered the whole surface with a size of 8 μm (Figure a). In the case of GO, densely packed graphene sheets on the GC substrate with a layered structure were observed (Figure b). Interestingly, after the addition of GO into Cu-MOF, the resulting composite still maintained the colloidosome-like structure with controlled size of 3 μm (Figure c). The drastic decrease in particle size of Cu-MOF is mainly due to the influence of oxygen containing functional groups in GO and ultrasonic waves. It has been already reported that GO can serve as a template for stabilizing as well as reducing the size of nanomaterials.[47] On the other hand, it has also been established that the ultrasonic treatment of GO could decrease the GO precipitates by increasing the dispersibility and completely exfoliating the multilayered GO into monolayer GO sheets with active functional sites, and long-term ultrasonication leads to further fragmentation in GO sheets.[48] The exfoliated GO sheets with active functional sites controlled the size and sustained the shape of the MOF, and hence, the size was decreased after the formation of composite. Generally, the synthesized MOF particles are randomly distributed, and therefore it is difficult to control the size of MOFs on the GO surface. Moreover, it is also difficult to coordinate the metal ion and ligand simultaneously with the GO, which often results in limited utilization of the GO sheets in the composite formation.[39] To prove the role of ultrasonication in the size-controlled formation of Cu-MOF–GO, the composite material was synthesized without ultrasonication. The SEM image of the composite prepared without ultrasonication also exhibited colloidosome-like structure, but the average size was found to be 6.6 μm (Figure S2). When compared to Cu-MOF, the size of the composite was decreased from 8 to 6.6 μm, which indicates that GO influenced the growth of Cu-MOF.

Figure 4

SEM images obtained at various magnifications for (a) Cu-MOF, (b) GO, and (c) Cu-MOF–GO.

On the other hand, the size of the composite was drastically decreased from 6.6 to 3.0 μm after ultrasonication of GO, indicating the limited utilization of GO sheets in the formed composite material without ultrasonication. Moreover, the ultrasonication treatment produced the dense and uniform size of Cu-MOF–GO composites. It also lead to the exfoliation of GO flakes into GO sheets with the provision of active functional sites, which can interact easily with metal ion to form a coordination bond to form a size-controlled composite material. In addition, sufficient interaction between exfoliated GO sheets and Cu ions can provide the dense and homogeneous formation of Cu-MOF–GO by the incorporation of Cu-MOF with the GO layer. Recently, Qiu et al. reported the synthesis of Al-MOF of aggregative amorphous particles with 1.50 μm size and observed a drastic decrease in size from 1.5 μm to 350 nm after the introduction of GO.[39] Jahan et al. demonstrated the bifunctional properties of GO and proved that the oxygen and carboxylic acid functional groups on either side of the sheet act as structure directing agents in the growth of microstructures.[49]

Effect of GO Concentration

Since the introduction of GO is highly influential in the size- and shape-controlled synthesis of Cu-MOF–GO, the weight ratio of GO in the composite was tuned, and the resulting size and the corresponding morphology were analyzed by SEM. Different weight ratios of GO from 0.1 to 0.4 mg/mL were incorporated with Cu-MOF to form the composites and were investigated (Figure ).

Figure 5

SEM images obtained at various magnifications for Cu-MOF composited with (a) 0.1, (b) 0.2, (c) 0.3 and (d) 0.4 mg/mL of GO.

In the case of 0.1 mg/mL GO in Cu-MOF, the size of the colloidosome Cu-MOF was found to be 6 μm (Figure a). Further increasing the concentration of GO to 0.2 and 0.3 mg/mL decreased the size of the Cu-MOF to 5 and 3 μm, respectively (Figure b,c). However, the colloidosome-like Cu-MOF structure was agglomerated with enlarged size after increasing the GO concentration to 0.4 mg/mL with Cu-MOF (Figure d). The obtained results suggest that the concentration of GO also plays a crucial role in controlling the size of Cu-MOF.

The elemental composition of the Cu-MOF–GO composite material is further characterized by EDX analysis. The EDX spectrum of the composite displays peaks at 0.27 keV for carbon and 0.53 keV for oxygen and two peaks at 0.91 and 8.04 keV for Cu of Kα and Lα shells, respectively (Figure S3). The EDX mapping analysis exhibited the homogeneous distribution of carbon (red), oxygen (green), and copper (blue) in the Cu-MOF–GO composite (Figure ). The above results indicate that the synthesized composite contains Cu-MOF and GO.

Figure 6

Elemental mapping of Cu-MOF–GO composite material.

Characterization of Cu-MOF–GO Fabricated GC Substrate

ATR-FT-IR spectral studies are highly useful in the structural analysis of the fabricated substrates. The ATR-FT-IR of Cu-MOF–GO composite modified GC substrate is shown in Figure S4. The composite modified substrate exhibited similar stretching frequencies as those shown by Cu-MOF–GO (Figure d). It shows the O–Cu–O stretching at 618 cm–1, the −CH stretching and bending vibrations at 2401 and 734 cm–1, respectively, the −OH stretching and bending vibrations at 3512 and 1434 cm–1, respectively, and the stretching of −C=C at 1608 cm–1.[41−43] The obtained ATR-FT-IR data suggest that modification of the Cu-MOF–GO composite on GC substrate does not affect the functional groups significantly.

Electrochemical Behavior of Cu-MOF–GO Fabricated GC Electrode

The as-synthesized Cu-MOF–GO composite was fabricated on a GC electrode, and its electrochemical behavior is analyzed by cyclic voltammetry (CV) in 0.2 M phosphate buffer (PB) solution. The CV of the composite fabricated electrode displays the oxidation peak at 0.04 V and a reduction peak at −0.25 V due to Cu2+/Cu+ (Figure S5).[50] The obtained redox response of Cu confirms the successful modification of Cu-MOF–GO composite on the GC electrode.

Further, the electron transfer rate of the modified electrode between the electrode/electrolyte interfaces can be investigated by electrical impedance spectral studies. The Nyquist planes of bare GC and GO, Cu-MOF, and Cu-MOF–GO fabricated GC electrodes in 1 mM [Fe(CN)6]3–/4–/0.1 M KCl mixture at 0.01 to 100 000 Hz scanning frequencies are displayed in Figure , and the results are simulated with a Randles equivalent circuit model (inset, Figure ).[40] The semicircle of the Nyquist plane can be highly useful in the determination of charge transfer resistance (RCT) of the electrode.[51] It was found that bare GC and GO, Cu-MOF, and Cu-MOF–GO fabricated electrodes exhibited the respective RCT values of 32.32, 26.58, 16.83, and 8.62 kΩ. Obviously, it can be found that the composite modified electrode exhibited a much lower RCT value than the GC and other (GO, Cu-MOF) fabricated electrodes, indicating the higher electron transfer kinetics. Further, the electron transfer rate constant (ket) at the electrode/electrolyte interface for different fabricated electrodes was evaluated (eq ).[40,52]where R, T, F, A, C°, and n have their usual significance.[40] The bare electrode exhibited ket of 1.02 × 10–4 cm2 s–1, whereas GO, Cu-MOF, and Cu-MOF–GO composite modified GC electrodes exhibited 1.25 × 10–4, 1.97 × 10–4, and 3.85× 10–4 cm2 s–1, respectively. The higher ket value attained from the composite fabricated electrode suggests the facile and faster electron transfer reaction at this electrode than that at the GO and Cu-MOF fabricated electrodes.

Figure 7

Nyquist planes of bare GC and GO, Cu-MOF, and Cu-MOF–GO modified GC electrodes in 1 mM [Fe(CN)6]3–/4– in 0.1 M KCl at 0.01 to 100000 Hz scanning frequencies. Inset, equivalent circuit for fitting.

Electrochemically active surface area (EASA) was estimated for bare and GO, Cu-MOF, Cu-MOF–GO composite modified GC electrodes using the Anson equation (eq ).[53,54]where the slope of the Anson plot (a) (Q vs t1/2) is obtained from chronocoulometric studies, n, F, and C have their usual significance in which n = 1 and C = 1 mM, and diffusion coefficient (D) (6.7 × 10–6 cm2 s–1). The EASA is calculated from the above inputs for different electrodes and found to be 0.062, 0.078, 0.094, and 0.136 cm2, respectively, for bare and GO, Cu-MOF, and Cu-MOF–GO composite fabricated GC electrodes. Thus, the composite fabricated electrode exhibited 2.2-fold higher EASA than the unmodified electrode. The size of the Cu-MOF was decreased after it formed a composite with GO, which in turn enhanced the EASA. This is evidenced from SEM analysis.

Electrochemical Oxidation of Nitrite

After studying the electrochemical characteristics of the modified electrodes, they were exploited for nitrite determination in 0.2 M PB solution (pH 7.2). The CVs of 0.5 mM nitrite at bare and GO, Cu-MOF, and Cu-MOF–GO modified GC electrodes at 50 mV/s are shown in Figure . The bare GC electrode shows the oxidation of nitrite at 0.92 V (curve a), whereas the GO modified electrode exhibited a broad oxidation at 0.87 V (curve b) with enhanced oxidation current. When the electrode is fabricated with Cu-MOF, it exhibits nitrite oxidation at 0.90 V with enhanced current (curve c). On the other hand, the nitrite oxidation is highly catalyzed at the Cu-MOF–GO composite fabricated electrode. It exhibits nitrite oxidation potential at 0.78 V (curve d), which is 140 mV less overpotential and 3.2-fold enhanced current compared with the bare electrode. Moreover, the oxidation potential of nitrite was highly stable at the Cu-MOF–GO/GC electrode even after 8 cycles (curve d, solid and dotted lines). The enhanced oxidation current of nitrite was due to higher electroactive surface area in addition to the synergetic effect of GO and Cu-MOF.

Figure 8

CVs for 0.5 mM nitrite at (a) bare GC and (b) GO, (c) Cu-MOF, and (d) Cu-MOF–GO fabricated electrodes in PB solution (pH 7.2) at 50 mV/s. (e) Cu-MOF–GO/GC electrode in the absence of nitrite. (solid line, 1st cycle, and dotted line, after 8 cycles)

Further, the nitrite oxidation is studied at different Cu-MOF–GO loaded GC electrodes, and the corresponding CVs are shown in Figure S6. The 1 mg/mL Cu-MOF–GO fabricated electrode shows nitrite oxidation at 0.79 V (curve a) while upon increase of the loading to 2 mg/mL, the oxidation peak was shifted to more positive potential of 0.86 V with increased current (curve b). Further increasing the loading level to 3 mg/mL resulted in the nitrite oxidation appearing at 0.87 V (curve c). A well-defined sharp oxidation peak at 0.78 V with an enhanced oxidation current was observed for 4 mg/mL loading level (curve d). Further increasing the loading level of Cu-MOF–GO to 5 mg/mL resulted in decrease of the oxidation current with shift in the oxidation toward more positive potential (curve e). This decreasing current response of nitrite was mainly due to the poor dispersion of Cu-MOF–GO. The obtained results clearly suggest that nitrite oxidation is highly dependant on the loading level of Cu-MOF–GO.

The kinetics of electrochemical nitrite oxidation was studied by utilizing the modified electrode in 0.5 mM nitrite at different sweep rates, and the results are shown in Figure S7. When the sweep rates varied from 10 to 100 mV/s, the oxidation current increased. The plot of oxidation peak current against the square root of sweep rate was linear with R2 = 0.9988 (Figure S7, inset) implying that the oxidation of nitrite is diffusion-controlled process. The possible nitrite oxidation mechanism at the Cu-MOF–GO fabricated electrode is shown in eqs –5.[54−56]

Sensitive Determination of Nitrite by DPV and Amperometry

Since the Cu-MOF–GO composite at 4 mg/mL exhibited excellent electrochemical response for nitrite oxidation, it was further utilized for the quantitative determination of nitrite. Figure exhibits the differential pulse voltammograms (DPVs) of nitrite at the Cu-MOF–GO modified GC electrode. The composite fabricated electrode exhibits nitrite oxidation at 0.65 V for the initial addition of 5 μM nitrite (Figure , curve b). Further addition of each 5 μM nitrite increases its oxidation current without affecting the oxidation potential, indicating that stable determination of nitrite was possible at the composite fabricated electrode. A linear relationship was observed upon plotting nitrite oxidation current against the added nitrite concentration with 0.9987 correlation coefficient (Figure , inset).

Figure 9

DPVs of each 5 μM nitrite addition at GC/Cu-MOF–GO electrode (a, 5 μM; b−i, 10−45 μM). Inset, correlation plot of oxidation current vs nitrite concentration.

The amperometric studies were further carried out to investigate the sensitivity, linear range of detection, and LOD of the present fabricated electrode for nitrite. Figure S8 displays the amperometric response for nitrite at the Cu-MOF–GO modified GC electrode in a continuously stirred 0.2 M PB solution at an applied potential of +1.0 V. It shows the response current for each 2 μM nitrite addition with 50 s sample interval, and the steady state was reached within 3 s. The plot of nitrite oxidation current against its concentration is linear with R2 = 0.9994 (Figure S7, inset). Further, determination of nitrite in a wide range of concentration from 10 nM to 0.1 mM was also studied by amperometry (Figure ). The response current was increased linearly with the increasing nitrite concentration from 10 nM to 0.1 mM at the Cu-MOF–GO modified GC electrode (Figure ). The plot of nitrite oxidation current against its concentration is linear with R2 = 0.9927 (Figure , inset).

Figure 10

Amperometric curves of nitrite at Cu-MOF–GO/GC electrode in PB solution at constant applied potential, +1.0 V. Each addition is responsible for (a) 0.01, (b) 0.1, (c) 0.5, (d) 1, (e) 2, (f) 10, (g) 20, (h) 35, and (i) 50 μM and (j) 0.10 mM nitrite at 50 s intervals. Inset, plot of current vs concentration of nitrite.

The present fabricated electrode exhibited the lowest LOD of 1.47 nM (S/N = 3) with an outstanding sensitivity of 1522.5 μA mM–1 cm–2 in the excellent linear range for concentration detection of 10 nM to 0.1 mM against nitrite determination when compared to the recently reported MOF-based electrochemical sensors (Table ).[11,12,14,57−60] Very importantly, the present electrode fabrication is very simple, and the fabricated sensor is highly stable against nitrite determination.

Table 1

Comparison of the Present Cu-MOF–GO Fabricated Sensor with the Reported Sensors Towards Electrochemical Sensing of Nitrite

electrode material	electrolytes (pH)	oxidation potential (V)	techniques	LOD (nM)	ref
Cu-MOF/rGOa/GCEb	0.1 M PBS	0.76	AMP	33	(11)
MOF-525/FTOc	KCl	0.85	AMP	2100	(12)
α-Fe2O3–CNTsd/GCE	7.0	0.78	AMP	150	(14)
GO–MnNH2TPPe/GCE	4.0	0.76	AMP	1100	(57)
Ag–rGO/GCE	7.4	0.82	DPV	12	(58)
Cu/MWCNTsf/GCE	7.0	0.93	AMP	1800	(59)
PANIg/GO/GCE	5.0	0.83	AMP	500	(60)
Cu-MOF–GO/GCE	7.2	0.66	AMP	1.47	this work

Reduced graphene oxide.

Glassy carbon electrode.

Fluorine doped tin oxide substrate.

Carbon nanotubes.

Metalloporphyrin.

Multiwalled CNTs.

Polyaniline.

Anti-interference Ability of the Cu-MOF–GO Electrode

Selectivity is one of the key issues in real-time electrochemical sensors, since it decides the practicability of the sensor. Thus, the anti-interference ability of the Cu-MOF–GO composite fabricated sensor toward nitrite determination in the presence of various possible physiological interferents and co-interfering agents is examined by amperometry. Figure shows the amperometric response of Cu-MOF–GO/GC electrode toward nitrite determination in the presence of several co-interfering agents like Na+, K+, Mg2+, Cl–, F–, SO42–, and NO32– (curves b–h) and physiological interferences including urea and oxalate (curves i and j). The amperogram exhibits the response current for 5 μM nitrite addition initially (curve a) and no significant change in current for further additions of 500 μM of the aforementioned interfering agents. These results suggest the excellent specificity of the fabricated sensor toward nitrite determination in the presence of 100-fold higher concentrations of other agents.

Figure 11

Amperometric current response for the successive additions of (a) 5 μM nitrite, (b–e) 500 μM each Na+, K+, Mg2+, and Cl–, and (f–j) F–, SO42–, NO32–, urea, and oxalate at Cu-MOF–GO/GC electrode in PB solution at the applied potential +1.0 V.

Stability and Reproducibility of the Nitrite Sensor

To evaluate the stability of Cu-MOF–GO modified GC electrode, the DPVs for 10 μM nitrite were monitored in PB solution frequently at 10 min intervals. It was observed that the nitrite oxidation current remains the same with a relative standard deviation (RSD) of 1.5% for 10 repetitive measurements. This result indicates that the fabricated sensor is highly stable. To find out the reproducibility, four different electrodes were fabricated with the Cu-MOF–GO composite and their DPV response currents against nitrite oxidation were investigated in 10 repetitive experiments. The obtained RSD of 1.8% confirmed that the present fabricated sensor is highly reproducible in nitrite determination.

Analysis of Nitrite in Environmental Samples

The practical utility of the present sensor was tested for nitrite in lake and industrial effluent water samples. They were collected from Kodaikanal and Erode regions, Tamil Nadu, India. Figure A shows the DPVs of lake water samples in PB solution, and it does not exhibit any oxidation peak within the studied potential window (curve b), indicating the absence of nitrite in the lake water sample. When 5 and 10 μM nitrite were spiked into lake water sample (curves c and d), an oxidation peak was obtained at 0.66 V due to the oxidation of nitrite. Figure B shows the DPVs of the industrial effluent water in 0.2 M PB solution. It shows an oxidation peak at 0.66 V (curve b), which may be due to nitrite oxidation. To ascertain that the observed oxidation peak is due to nitrite, known concentrations of commercial nitrite (5 and 10 μM) were added to the industrial effluent sample (curves c and d). The enhancement of the oxidation current at 0.66 V without shifting the potential suggested that the oxidation peak was due to nitrite. The recovery results for nitrite addition in lake and industrial effluent samples are summarized in Table S2. The proposed method showed ∼99% recovery for spiked nitrite in lake and industrial effluent water samples, suggesting that the present modified electrode could be applied in real sample analysis.

Figure 12

DPVs obtained for (a) absence and (b) presence of (A) lake water and (B) industrial effluent water and (c, d) after the addition of 5 and 10 μM nitrite to lake water and industrial effluent water samples containing in 0.2 M PB solution (pH 7.2) at Cu-MOF–GO modified electrode.

Experimental Section

Chemicals

Azelaic acid (AZA) from Tokyo Chemical Industry Co., Ltd., copper(II) nitrate trihydrate, N,N-dimethylformamide (DMF), sodium dihydrogen phosphate, and disodium hydrogen phosphate from Merck, India, and GO from GO Advanced Solution Sdn Bhd, Malaysia, were purchased. The chemicals utilized in this study are analar grade.

Cu-MOF–GO Synthesis and Its Fabrication for Electrochemical Sensors

The Cu-MOF–GO composite was synthesized by ultrasonication followed by the solvothermal approach. Briefly, Cu(NO3)2 (0.3 g) and different weight ratios of GO (0.1–0.4 mg/mL) were mixed well in 70 mL of DMF and ultrasonicated for 2 h followed by the successive addition of AZA (0.29 g) into the mixture. The reaction mixture was then ultrasonicated for a further 20 min, poured into a 100 mL autoclave and heated to 120 °C in an oven for 24 h, and then cooled at room temperature. The final product of solid Cu-MOF–GO was then filtered and allowed to solvent exchange with ethanol numerous times and finally dried (Scheme ). Cu-MOF was synthesized to compare the electrochemical performance of Cu-MOF–GO composite.

Scheme 1

Illustration of the Preparation of Cu-MOF–GO Composite

The electrochemical sensor for nitrite ion was constructed by fabricating the as-synthesized Cu-MOF–GO composite on the well-polished GC electrode surface. Briefly, 4 mg/mL composite material was dispersed in water and ultrasonicated for 20 min to form a homogeneous dispersion. The above dispersed composite (7 μL) was then fabricated on the surface of a GC electrode by drop-casting and dried in air, and then 7 μL of 0.5% Nafion in ethanol was cast on the composite modified electrode and dried in air. The resultant Cu-MOF–GO fabricated GC electrode was utilized for electrochemical studies.

Characterization of Cu-MOF–GO and Its Fabrication

The crystallographic structure, chemical bonding, and functional groups of the Cu-MOF–GO composite were analyzed with Fourier transform-infrared (FT-IR) (JASCO FT-IR 460 plus) and XRD (PANalytical X’pert[3]) techniques. XPS (PHI 5000 Versa Probe II, FEI Inc.) analysis was carried out to study the electronic state and the bonding of GO with Cu-MOF. The obtained XPS data were deconvolutated using the software XPSPEAK 4.1. Raman spectroscopy was analyzed with a He–Cd laser IK5651R-G from Kimmon Electric Ltd., Japan. For DPV measurements, the following parameters were employed: pulse width = 0.06 s, amplitude = 0.05 V, sample period = 0.02 s, and pulse period = 0.2 s. The topography of the composite was analyzed by SEM (VEGA3 TESCAN) binding with EDX probe analyzer (Bruker Nano, Germany). All the electrochemical investigations (CHI, 643B workstation, Austin) were analyzed at 27 °C under N2 atmosphere in a 3-electrode system (GC working electrode; Ag/AgCl reference electrode; Pt wire counter electrode).

Conclusions

The present work demonstrated the fabrication of a colloidosome-like Cu-MOF–GO composite electrode and its sensitive and selective nitrite determination by DPV and amperometry. The composite was synthesized from azelaic acid, Cu salt, and GO in DMF by ultrasonic treatment followed by solvothermal method at 120 °C for 24 h. The SEM images of Cu-MOF showed colloidosome like structure with size of 8 μm whereas the Cu-MOF–GO composite exhibited similar morphology with controlled size of 3 μm. This was attributed to the formation of exfoliated GO sheets by ultrasonication in addition to the influence of GO as a nucleation node for size-controlled growth. Under optimized conditions, the Cu-MOF–GO modified GC electrode was utilized as a sensitive, selective, and stable electrochemical scaffold for the determination of nitrite. The lowest LOD of 1.47 nM (S/N = 3) with a sensitivity of 1522.5 μA mM–1 cm–2 and excellent linear range of concentration detection of 10 nM to 0.1 mM toward nitrite determination was achieved using the present sensor when compared to the reported MOF-based electrochemical sensors. Finally, the present sensor was used to determine nitrite in industrial effluent samples and lake water samples.