Designing CoS1.035 Nanoparticles Anchored on N-Doped Carbon Dodecahedron as Dual-Enzyme Mimics for the Colorimetric Detection of H2O2 and Glutathione.

In recent years, the exploration of the nanozyme, an artificial enzyme with the structure and function of natural enzymes, has become a hot topic in this field. Although significant progress has been made, it is still a huge challenge to design nanozymes with multiple enzyme-like catalytic activities. In this work, we have successfully fabricated a colorimetric sensing platform to mimic peroxidase-like and oxidase-like activities by the CoS1.035 nanoparticles decorated N-doped carbon framework porous dodecahedrons (abbreviated to CoS1.035/N-C PDHs). And the catalytic mechanism of CoS1.035/N-C PDHs toward the peroxidase-like and oxidase-like activities is systematically explored. The results display that CoS1.035/N-C PDHs can catalyze the oxidation of the colorless substrate 3,3,'5,5'-tetramethylbenzidine (TMB) into blue oxidized TMB (ox-TMB) by disintegrating H2O2 or the physically/chemically absorbed O2 into different ROS species (·OH or O2 ·-) in the presence or absence of H2O2. Therefore, on the basis of the dual-enzyme mimic activities of CoS1.035/N-C PDHs, the bifunctional colorimetric sensing platform is established for H2O2 detection with a wide linear range of 0.5-120 μM and glutathione detection with a linear range of 1-60 μM, respectively. This work provides an efficient platform for dual-enzyme mimics, expanding the application prospect of Co-based chalcogenides as enzyme mimics in biosensing, medical diagnosis, and environment monitoring.

Introduction

Most natural enzymes, as powerful biocatalysts, are a class of biological macromolecules (a few are RNAs) with high selectivity and catalytic activity, which play a substantial role in many biological processes. However, owing to the difficulty of preparation, high cost, poor stability, and other inherent defects, their application is seriously hindered.[1,2] In the last 10 years, as novel artificial enzymes, nanozymes have been developed to substitute natural enzymes.[3] They are defined as nanomaterials with enzyme-like properties (which can simulate the structure and function of enzymes).[4,5] Since magnetic Fe3O4 nanoparticles were first proven to show intrinsic peroxidase-like properties in 2007,[6] a surge of inorganic nanomaterials has been found to possess enzyme-like properties, such as noble metals,[7] metal oxides[8] and sulfides,[9−11] carbon materials,[12] and metal organic frameworks and their derivatives.[13,14] Due to the diverse structure, easy surface modification, adjustable activity, large surface area, simple preparation, low cost, and high stability (tolerance to strong acid and alkali and resistance to high temperature), nanozymes have broad application prospects in the fields of chemical sensing, biological analysis, medical diagnosis, and environmental conservation.[15−18] However, compared with natural enzymes, nanozymes have no substrate specificity, and their catalytic activity is still low.[19] Moreover, generally, most nanozymes only display one characteristic enzyme-like catalytic activity, which restrains their further application in the fields of analysis, sensing, and others. Thus, it is a huge challenge to design highly efficient nanozymes with multiple enzymatic mimetic properties, which exhibit promising applications for colorimetric sensing.

Recently, transition metal chalcogenides (TMCs) were reported to own intrinsic enzyme-like catalytic properties.[20,21] For example, Xia et al.[22] reported that the MoS2 quantum dot possesses excellent peroxidase-like activity through precisely controlling the existence form of the MoS2 quantum dot with the help of Fe3+ ions. Based on the finding, they established a reliable colorimetric detection platform for pyrophosphate. Besides, He et al.[23] prepared a CuS concave polyhedral superstructure through a green solvothermal reaction, which exhibits intrinsic peroxidase-like activity, as it can catalyze TMB and OPD oxidation quickly in the presence of hydrogen peroxide. Likewise, Wang et al.[24] synthesized the hollow CuS nanocube, which also displays outstanding peroxidase-like property. And they found that the excellent catalysis property is attributed to the existence of copper, endowing the hollow CuS nanocube with the characteristics of Fenton’s reagents. On the basis of the research of Wang et al., Song et al.[9] designed FeS2 nanoparticles as peroxidase mimicase owing to their suitable band gap and extraordinary absorption coefficient. And then, a sensitive colorimetric method for H2O2 and glutathione (GSH) detection was developed. Considering the similar electron structure of Fe, Cu, and Co, we reasonably speculate that the cobaltous sulfides might also hold a great potential to be a mimic enzyme. Although some progress has been made in this field,[25] the exploration of cobaltous sulfides as dual-enzyme mimics and the subsequent application in colorimetric sensing are still in their infancy. Therefore, it is of great significance to develop a bifunctional nanozyme with excellent enzyme-like activity and stability.

Based on the above-mentioned condition, in this work, we designed CoS1.035/N-C porous dodecahedrons (PDHs) as bifunctional mimic enzymes, in which CoS1.035 nanoparticles are uniformly dispersed on a porous N-doped carbon dodecahedron framework. The CoS1.035/N-C PDH nanozyme owns multiple advantages, such as the fact that the dodecahedron carbon framework offers abundant channels for electrons, facilitating electron transfer from the catalyst to reactant, and the porous property brings about a huge surface area to promote the fast contact of substrates with the catalytically active sites. Besides, N atoms dope into the carbon framework, which alter the surface charge property of the CoS1.035/N-C PDH nanozyme catalyst. In addition, the CoS1.035 nanoparticles are anchored on the N-doped carbon dodecahedron framework, which could prevent the active nanoparticles from being directly exposed to the reaction solution. All of the above endow the CoS1.035/N-C PDH nanozyme catalyst with glorious dual-enzyme mimic activities and stability. As a result, CoS1.035/N-C PDHs exhibit satisfactory peroxidase-like and oxidase-like activities, which could catalyze the oxidation of the colorless substrate 3,3,′5,5′-tetramethylbenzidine (TMB) into blue oxidized TMB (ox-TMB) in either the presence or absence of H2O2. Consequently, combined with the inoxidizability of glutathione (GSH), a bifunctional colorimetric sensing platform for the visual detection of H2O2 and GSH was designed. The proposed method shows great potential application in clinical diagnosis, environmental protection, and the food industry.

Results and Discussion

Synthesis and Characterization of CoM/N-C (M = O, S, Se, Te) PDHs

As shown in Scheme , first, the precursor ZIF-67 nanomaterial was prepared by self-assembly of cobalt nitrate and dimethyl imidazole at room temperature, and then the Co/N-C intermediate was obtained by annealing the as-prepared ZIF-67 dodecahedron under a N2 flow atmosphere. During this process, the nitrogenous organic ligands were carbonized to form porous N-doped carbon framework dodecahedrons, while the Co2+ ions were reduced to Co nanoparticles by the derived carbon. Finally, the CoM/N-C porous dodecahedrons (marked as CoM/N-C PDHs) with the morphology of the precursor were fabricated by introduction of the chalcogen element (M = O, S, Se, Te) through high heat treatment.

Scheme 1

Schematic Presentation of the Synthesis Procedure of CoM/N-C PDHs (M = O, S, Se, Te)

To explore the structural feature and stability of the synthesized ZIF-67 precursor, field emission scanning electron microscope (FE-SEM), X-ray diffraction (XRD), and thermogravimetry (TG) analysis were carried out. As shown in Figures S1a and S3a,e, the as-obtained ZIF-67 precursor exhibits the morphology of a typical rhombic dodecahedron with an average size of 751.9 ± 7.2 nm (Figure S1b). And the XRD pattern (Figure S1c) shows that the ZIF-67 dodecahedron has good crystallinity even without high heat treatment, suggesting the high thermal stability. The high thermal stability makes the ZIF-67 dodecahedron an ideal carbon-based precursor to maintain its initial porous morphology and avoid excessive vaporization during the annealing process. Thermogravimetry analysis in Figure S1d displays that the weight loss (4.67%) of the ZIF-67 dodecahedron below 550 °C is attributed to the removal of solvents and dehydration. When the temperature is elevated from 550 to 600 °C, the nitrogenous organic ligand is totally decomposed to form a N-doped carbon framework causing a significant weight loss (31.54%). In this process, a large number of pores are produced due to the decomposition of organic linkers. Meanwhile, Co2+ ions are reduced to Co nanoparticles by the derived carbon, inducing the transformation from the ZIF-67 precursor into the Co/N-C intermediate. When the temperature is above 600 °C, there is a degree of weight loss probably due to the structural collapse of the N-doped carbon framework. These results show that when the annealed temperature is above 550 °C, the pure Co/N-C intermediate can be successfully obtained. Therefore, three Co/N-C intermediates under different annealed temperatures (550, 600, and 700 °C) were prepared. Figure S2 reveals that the XRD diffraction peaks of the three samples are consistent with the cubic Co phase (JCPDS no. 89-4307) without a sign of the precursor, indicating that the ZIF-67 precursor transforms into the Co/N-C intermediate successfully. As shown in Figure S3b,f, the Co/N-C intermediate (550 °C) perfectly inherits the dodecahedron morphology of the ZIF-67 precursor, but the active Co nanoparticles have not been formed obviously. When carbonized at 600 °C, the Co/N-C (600 °C) dodecahedron with a slightly sunken surface can be obtained, and Co nanoparticles with a small size are evenly distributed on the N-doped carbon framework (Figure S3c,g). At 700 °C, the N-doped carbon framework of the Co/N-C (700 °C) dodecahedron collapses slightly, and Co nanoparticles increase in size due to agglomeration (Figure S3d,h). It can be seen that the annealing temperature can affect the collapsed degree of the N-doped carbon framework and the size of Co nanoparticles, which may further affect the catalytic activity of the subsequently synthesized CoM/N-C (M = O, S, Se, Te) PDHs. Therefore, we compared the catalytic performance of the above three intermediates, as shown in Figure S4. It is not difficult to find that the as-prepared Co/N-C (600 °C) exhibits the highest catalytic activity among the three intermediates, which may be attributed to the smaller size and uniform distribution of the active centers on the N-doped carbon framework. After comprehensive consideration, the Co/N-C (600 °C) intermediate was selected for subsequent research.

To better regulate the catalytic performance, CoM/N-C (M = O, S, Se, Te) PDHs with a concave surface were synthesized and characterized. Figure shows the SEM images of CoM/N-C (M = O, S, Se, Te) PDHs. It can be seen that except for Co3O4/N-C PDHs, other samples, such as CoM/N-C (M = S, Se, Te) PDHs, keep the concave dodecahedron morphology. The Co3O4/N-C nanostructure is composed of small particles as building blocks. Its morphology collapses completely probably due to the destruction of the carbon framework during calcination in the air. In addition, the sizes of CoM/N-C (M = O, S, Se, Te) PDHs are 373.5 ± 4.2 nm (Co3O4/N-C), 696.3 ± 13.1 nm (CoS1.035/N-C), 724.3 ± 5.5 nm (Co0.85Se/N-C), and 659.4 ± 3.6 nm (CoTe/N-C), respectively, which are smaller than those of ZIF-67. The results show that the diameter of CoM/N-C (M = O, S, Se, Te) PDHs decreases after heat treatment. Figure S5 exhibits the XRD patterns of the obtained CoM/N-C (M = O, S, Se, Te) PDHs. The diffraction peaks of all samples match well with the corresponding standard cards, indicating the acquisition of pure phases. Besides, element mapping analysis was used to measure the percentages of each element in the CoM/N-C (M = O, S, Se, Te) PDHs and intermediate Co/N-C PDHs, as displayed in Table S1, which indirectly proves the successful synthesis of the above materials. More convincing evidence is offered by the inductive coupled plasma atomic emission spectrometer (ICP-AES) in Table S2, in which the Co/M mole ratios are well consistent with the results in XRD. Figure S6 displays the uniform distribution of C, N, Co, and the chalcogen element (M = O, S, Se, Te) over the entire architecture of porous dodecahedrons. The above observations prove the successful synthesis of four CoM/N-C (M = O, S, Se, Te) PDHs. Then, we compared their catalytic performance as exhibited in Figure S7. Obviously, among these nanomaterials, CoS1.035/N-C PDHs exhibit the best peroxidase-like activity. As is well known, the performance of catalysts is mainly determined by their composition, structure, and morphology. As shown in Figure S6b–e, four CoM/N-C (M = O, S, Se, Te) materials have similar morphology with porous dodecahedrons (PDHs). The difference is that the structure of Co3O4/N-C PDHs has completely collapsed owing to the destruction of the carbon framework during calcination in the air. This reason may be attributed to the suboptimal performance of Co3O4/N-C PDHs. Besides, an ideal peroxidase- and oxidase-like catalyst should serve as the electron acceptor, which could receive the electrons from reactants. In three CoM/N-C (M = S, Se, Te) PDHs, tellurides have typical metallic properties; thus, its ability to gain electrons is limited. And sulfur is more electronegative than selenium; consequently, CoS1.035/N-C PDHs own higher peroxidase- and oxidase-like activities. To explore the intrinsic factor of the enhanced activity, more elaborate characterizations toward CoS1.035/N-C PDHs were carried out.

Figure 1

The SEM images and size distributions of CoM/N-C PDHs (M = O, S, Se, Te). (a, e) Co3O4/N-C; (b, f) CoS1.035/N-C; (c, g) Co0.85Se/N-C; and (d, h) CoTe/N-C.

Figure shows the scanning electron microscope (SEM) images, transmission electron microscope (TEM) images, and high-resolution transmitting electron microscope (HRTEM) images of CoS1.035/N-C PDHs. In Figure a,c, the large-scale CoS1.035/N-C PDHs with a uniform size can be observed. They inherit the porous dodecahedron morphology of intermediate Co/N-C PDHs perfectly, suggesting the excellent structural stability. And a large number of CoS1.035 nanoparticles are evenly distributed on the N-doped carbon dodecahedron framework, as shown in Figure b,d. No significant agglomeration has been discovered. Further observation (Figure e) reveals that the size of CoS1.035 nanoparticles falls in the range of 15–30 nm, as indicated by the red circles. Notably, some light and shade contrasts (as indicated by the yellow arrows) can be discerned in the enlarged TEM image (Figure e), implying the presence of abundant pores. Thus, the nitrogen adsorption–desorption isotherm was investigated to prove the porous structure. As exhibited in Figure S8, CoS1.035/N-C PDHs display a combination of type-I and type-IV isotherm characteristics with an additional H4-type broad hysteresis loop at p/p0 = 0.5–0.99, revealing the existence of a mesoporous structure in CoS1.035/N-C PDHs. The pore-size distribution of CoS1.035/N-C PDHs mainly focuses on 3.89 nm. Such porous structure brings about a high surface area (230.692 m2 g–1). This intriguing structure is believed to be advantageous for the exposure of active sites and mass transport of the substrates. Besides, the HRTEM image in Figure f reveals a set of regular lattice fringes with an interplanar crystal distance of 0.195 nm corresponding to the (102) lattice plane of CoS1.035, which is consistent with the XRD results.

Figure 2

(a, b) The SEM images, (c, d) TEM images, and (g) HRTEM images of CoS1.035/N-C PDHs.

To further investigate the surface chemical components and chemical states of the CoS1.035/N-C PDHs, X-ray photoelectron spectroscopy (XPS) measurement was performed. The binding energies of all spectra are calibrated strictly using the standard C 1s peak (284.8 eV). As shown in Figure a, the XPS survey spectrum of CoS1.035/N-C PDHs exhibits the presence of C, N, O, S, and Co elements, wherein the O element is derived from the surface oxidation (531.1 eV), dissociative oxygen (532.0 eV), and adsorbed oxygen (533.4 eV), which is proven by the O 1s XPS spectrum (Figure b). And Figure c–f exhibits, respectively, XPS spectra of the C 1s, N 1s, S 2p, and Co 2p region for the samples. The asymmetrical C 1s peak in Figure c is composed of three independent peaks, which could be assigned to COO– (288.5 eV),C=N (285.8 eV), and C=C (284.8 eV), respectively.[26] For the N 1s region (Figure d), the original peak is deconvoluted into four peaks, that is, graphitic N (400.7 eV), pyrrolic N (399.5 eV), Co–N (399.1 eV), and pyridinic N (398.7 eV).[27] The graphitic N atoms are inserted into the carbon layer and bond to three adjacent carbon atoms, which may have a p-doping effect on the N-doped carbon, while the pyrrolic N and pyridinic N atoms increase the density of the electronic states of carbon and open the band gap,[27] altering the surface charge property. Consequently, it facilitates the electron transfer from the catalyst to reactant, contributing to the prominent catalytic activity. In the S 2p region, as shown in Figure e, the binding energies of two satellites, S 2p1/2, S 2p3/2, and S-Co species are observed at 170.1, 168.9, 165.5, 164.0, and 162.1 eV, respectively.[28,29] For the Co 2p region, as illustrated in Figure f, it could be resolved into seven peaks, involving satellites (at 803.1 and 786.7 eV), Co 2p1/2 (at 797.7 and 793.6 eV), Co 2p3/2 (at 783.7 and 781.5 eV), and the Co–S bond (at 778.5 eV). The peaks located at 781.5 and 778.5 eV are attributed to the Co–N and Co–S species, respectively. These observations confirm the existence of Co2+/Co3+ ions on the surface of CoS1.035/N-C PDHs.[28,30] The above results show that the CoS1.035/N-C PDHs were successfully prepared, which are in accord with the XRD and HRTEM analysis. Additionally, Raman measurement is an efficient tool to detect the structural properties of carbon-based materials. The Raman spectrum in Figure S9 reveals two characteristic peaks of the D band (1347 cm–1) and G band (1588 cm–1) of carbon materials. The D band reflects the disordered carbon or defective graphitic structures resulting from edge doping of N atoms in the carbon framework. The G band represents the existence of the graphitic structure.[31] The ID/IG value of 1.17 implies the presence of a large number of defects in the N-doped carbon framework. These defects are in favor of accelerating the interfacial electron transfer between the catalyst and substrates, which could partially account for the excellent catalytic activity of CoS1.035/N-C PDHs.

Figure 3

(a) The survey scan of CoS1.035/N-C. The high-resolution XPS spectra of (b) O 1s, (c) C 1s, (d) N 1s, (e) S 2p, and (f) Co 2p in the CoS1.035/N-C PDHs, respectively.

Exploring the Peroxidase-Like Activity of CoS1.035/N-C PDHs

Considering the structural advantages, such as the hollow dodecahedron morphology with a huge surface area, stable N-doped carbon framework, and ultrafine CoS1.035 nanoparticles, the CoS1.035/N-C PDH material is deemed as a promising enzyme mimic catalyst. Therefore, the catalytic activities of the CoS1.035/N-C PDHs were investigated by performing the peroxidase-mediated oxidation reactions of 3,3,′5,5′-tetramethylbenzidine (TMB) in the presence of H2O2. As demonstrated in Figure a, CoS1.035/N-C PDHs could catalyze the oxidation of TMB by H2O2 to generate the typical blue color and an absorption peak at 652 nm, which could be applied as a colorimetric probe to oxTMB. In contrast, the individual CoS1.035/N-C + H2O2 or TMB + H2O2 system does not display the obvious colorimetric response and absorbance, implying the outstanding peroxidase-like activity of CoS1.035/N-C PDHs. More distinct phenomenon exhibits in the time-course profile (Figure S10a). The intensity of the absorption peak gradually rises and the color of the test solution becomes darker as time goes on. Besides, the responses of absorbance are particularly rapid accompanied with the concentration increases of TMB and H2O2, as shown in Figure S10b,c, proving the fast reaction rate.

Figure 4

(a) The UV–vis absorption spectra of different test systems in the absence and presence of 20 mM H2O2, 15 μg/mL CoS1.035/N-C PDHs, and 50 μg/mL TMB. (b) The UV–vis absorption spectra of 15 μg/mL CoS1.035/N-C PDHs with 20 mM H2O2 in the different substrates (50 μg/mL). (c) The zeta potential measurements of Co/N-C and CoM/N-C PDHs (M = O, S, Se, Te). The error bars represent the standard deviation values of three measurements. (d) The UV–vis absorption spectra of 15 μg/mL CoS1.035/N-C PDHs, 20 mM H2O2, and 50 μg/mL TMB in the absence or presence PTA.

To prove the specificity of CoS1.035/N-C PDHs toward substrates, reaction systems with different chromogenic substrates, such as TMB, OPD, and ABTs, were also employed for comparison. By visual inspection and spectral measurements, the chromogenic substrates TMB and OPD could be catalyzed by the CoS1.035/N-C PDHs to produce blue oxTMB (λ = 652 nm) and yellow oxOPD (λ = 450 nm) with specific absorption peaks, respectively, as shown in Figure b. However, there is no obvious change for the ABTS substrate. The experimental results show that CoS1.035/N-C PDHs can catalyze the oxidation of positively charged TMB and uncharged OPD but cannot catalyze the oxidation of negatively charged ABTS, which may be related to the surface charge of CoS1.035/N-C PDHs. To verify this hypothesis, the surface zeta potential (ζ) was analyzed by a dynamic light scattering instrument. As shown in Figure c, CoS1.035/N-C PDHs are negatively charged; thus, they could bond with TMB and OPD by the electrostatic interaction, indicating that the peroxidase-like activity of CoS1.035/N-C PDHs demonstrates a certain extent of specificity toward chromogenic substrates. Moreover, compared with Co/N-C PDHs (−14.9 mV), Co3O4/N-C PDHs (−11.7 mV), Co0.85Se/N-C PDHs (−13.9 mV), and CoTe/N-C PDHs (−11.6 mV), CoS1.035/N-C PDHs have the strongest negative charge (−17.3 mV), which explains the optimal peroxidase-like activity among the five catalysts, as shown in Figure S7. As reported in the literature, the most artificial peroxidases are able to catalyze H2O2 by generating reactive oxygen species (ROS) like hydroxyl radicals (·OH).[32,33] Therefore, we investigated the ROS in this reaction system by adding the specific radical quencher of hydroxyl radicals (·OH). Phthalic acid (PTA) is a radical scavenger that could react with ·OH to impede the catalytic oxidation of TMB.[34,35] Obviously, there are a blue color and strong absorbance at 652 nm in the test system of CoS1.035/N-C + TMB + H2O2 (Figure d). However, when a certain amount of PTA is added into the above reaction solution, the blue color and absorbance at 652 nm decay distinctly, confirming that PTA could eliminate the ·OH radicals generated when CoS1.035/N-C PDHs exhibited their peroxidase-like activity. Interestingly, a slight oxidation of TMB existed even when the ·OH radicals were eliminated completely by PTA. That is probably because CoS1.035/N-C PDHs own a certain extent of oxidase-like activity, which could catalyze the direct oxidation of TMB by the dissolved oxygen. These observations indicate the generation of ·OH in the reaction process that is closely related to the peroxidase-like catalytic mechanism of CoS1.035/N-C PDHs, similar to other nanozymes.[36,37]

Explore the Oxidase-Like Activity of CoS1.035/N-C PDHs

Based on the above results, we found that the CoS1.035/N-C PDH catalyst also reveals an oxidase-like activity, which could oxidize TMB to produce a blue color with a weak absorbance at 652 nm even in the absence of H2O2 (Figure a). Notably, the absorbance extremely increases when a certain amount of H2O2 is added into the reaction system, implying that the peroxidase-like activity of the CoS1.035/N-C PDH catalyst is higher than its oxidase-like activity. In general, the oxidation of TMB without H2O2 is driven by dissolved oxygen. Hence, to verify the action of dissolved O2, the contrast experiments by bubbling inert gas into the system of N2, air, and O2 for 30 min were carried out. As displayed in Figure b, compared to the air-saturated reaction system, the absorption intensity of oxTMB at 652 nm catalyzed by CoS1.035/N-C PDHs exhibits an obvious promotion in the O2-saturated condition and a significant decrease in the N2-filled system, illustrating that the catalytic oxidation of TMB demands the participation of dissolved O2. The dissolved O2 as the electron acceptor could combine with electrons from TMB to produce ROS (O2·–), which is the real oxidant. In Figure c, the oxidase-like activity of CoS1.035/N-C PDHs is slightly impeded when the p-benzoquinone (PBQ), a scavenger of the superoxide radical (O2·–), is introduced into the reaction system without H2O2, suggesting that small doses of O2·– are produced in the oxidase-like reaction system.

Figure 5

(a) The UV–vis absorption spectra of various reaction systems in an acetate buffer solution (pH = 5.0). (b) The UV–vis absorption spectra of reaction systems in air-saturated, O2-saturated, and N2-saturated systems. The reaction system consists of TMB (50 μg/mL), CoS1.035/N-C PDHs (15 μg/mL), H2O2 (10 mM) in HAc-NaAc buffer solution (pH = 5.0). (c) The UV–vis absorption spectra of 15 μg/mL CoS1.035/N-C PDHs and 50 μg/mL TMB in the absence or presence PBQ. The inset is the corresponding photograph.

Catalytic Mechanism of CoS1.035/N-C PDHs

The electron paramagnetic resonance (EPR) spectrum in Figure a offers more direct evidences. The 5,5-dimethyl-1-pyridine N-oxide (DMPO) is selected as the spin capture reagent, which could react with oxygen-centered free radicals, such as ·OH and O2·–, to generate the more stable free radical adducts.[38] Consequently, the production of ·OH and O2·– was also monitored by incorporating DMPO. As depicted in Figure a, the characteristic spectrum of the DMPO/OH adduct with an intensity of 1:2:2:1 can be detected.[39] Likewise, Figure a displays the signal of the DMPO/O2·– adduct with six characteristic peaks.[40] Based on the above results, a possible mechanism for the generation of ·OH and O2·– by the CoS1.035/N-C PDH catalyst with peroxidase- and oxidase-like activities is provided in Figure b. In the presence of H2O2, the oxidation of TMB is on account of the electron transfer from the nonbonding orbital (NBO) of TMB to the lowest unoccupied molecular orbital (LUMO) of H2O2. In this process, the CoS1.035/N-C PDH catalyst receives the lone pair electron from NBO of TMB and then transfers it into the LUMO of H2O2, resulting in the breakage of O–O bonds and production of ·OH. Meanwhile, the TMB is catalytically oxidized into oxTMB due to the loss of electrons. In the absence of H2O2, the electrons from TMB rapidly transfer to the CoS1.035/N-C PDH catalyst, and then the dissolved O2 absorbed on the surface of the CoS1.035/N-C PDH catalyst serves as the electron acceptor, which receives the electrons to form the ROS of O2·–. In parallel, the TMB is catalytically oxidized into oxTMB with the responses of color and absorbance. In summary, the CoS1.035/N-C PDH catalyst could disintegrate H2O2 and the physically/chemically absorbed O2 into different ROS species (·OH and O2·–), consequently illustrating the prominent dual-enzyme mimic activities.

Figure 6

(a) The typical EPR spectrum: ·OH and O2·– originated from the CoS1.035/N-C PDHs +H2O2+ DMPO system. (b) The proposed reaction mechanism of dual-enzyme mimic activities for CoS1.035/N-C PDHs.

Analysis of Steady-State Kinetic of CoS1.035/N-C PDHs

To explore the enzyme-like catalytic properties of CoS1.035/N-C PDHs, TMB and H2O2 were used as substrates to study the steady-state kinetics of the system by changing one and keeping the other unchanged in the appropriate concentration range. Figure a,c shows the steady-state kinetic studies of H2O2 (concentration range: 0.1–40 mM) and TMB (concentration range: 0.02–1.4 mM), respectively. Figure b,d shows the Michaelis–Menten curves of H2O2 and TMB, respectively, which are obtained by fitting the Michaelis–Menten function. Among them, the Michaelis constants Km and Vm are obtained from the following equation:wherein Vm is the maximum reaction rate, V0 is the initial rate, [S] refers to the substrate (H2O2 or TMB) concentration, and Km is the substrate concentration when the reaction rate reaches half of the maximum reaction rate. The values of Vm and Km reflect the catalytic activity and affinity of the catalyst with the substrate. The higher the Vm is, the better is the catalytic activity of the catalyst. The smaller the Km value is, the stronger is the affinity between the catalyst and substrate. According to Figure and Figure S11, the Km and Vm values of CoS1.035/N-C PDHs are listed in Table S3. The Km value of CoS1.035/N-C PDHs with substrate H2O2 and TMB is 0.0348 and 0.124 mM, respectively, and the Vm value is 4.645 × 10–8 and 6.642 × 10–7. As expected, the Km value of CoS1.035/N-C PDHs is lower than that of Co/N-C, Co3O4/N-C, Co0.85Se/N-C, and CoTe/N-C, while the Vm value of CoS1.035/N-C PDHs is higher than their corresponding values (see Table S3). From these experimental results, it is concluded that CoS1.035/N-C PDHs have remarkable affinity and catalytic activity. And their catalytic activity is comparable to that of horseradish peroxidase (HRP), as exhibited in Table S3. Consequently, they can be used to replace natural enzymes. Such a high catalytic activity of CoS1.035/N-C PDHs might be attributed to the following merits: (1) the high surface area and abundant pores facilitate the rapid diffusion of reaction molecules to the catalytic active sites, and (2) N-doping could increase the density of the electronic states of carbon, accelerating the interfacial electron transfer between the catalyst and substrates, which promotes H2O2/dissolved O2 to produce ROS with strong oxidation ability. Consequently, CoS1.035/N-C PDHs reveal remarkable catalytic performances.

Figure 7

Steady-state kinetic study using the Michaelis–Menten model (a, b) for CoS1.035/N-C by varying the concentration of H2O2 with a fixed amount of TMB and (c, d) varying the concentration of TMB with a fixed amount of H2O2. The error bars represent the standard deviation values of three measurements.

Optimizing the Reaction Conditions of CoS1.035/N-C PDHs

To further master the bifunctional enzyme-like catalytic properties of CoS1.035/N-C PDHs, the reaction conditions affecting the activity of CoS1.035/N-C PDHs were optimized. Several significant parameters, such as pH, concentration, temperature, and reaction time, were researched to acquire the optimal catalytic activity. As is well known, the pH value has a huge effect on the catalytic activity of natural enzymes. Therefore, we first explored the dependence of pH value on peroxidase-like and oxidase-like activities. From Figure a and Figure S12a, it can be concluded that the optimal peroxidase-like activity of CoS1.035/N-C PDHs focuses in the pH range of 4.2–5.4. And the satisfactory pH range for oxidase-like activity is determined to be 4.6–5.2, as demonstrated by the more intense colorimetric responses in reaction solutions (inset in Figure S12b). As a result, both peroxidase- and oxidase-like activities could achieve the optimum state at a pH value of 5.0, which is close to the condition (pH = 4.5) of natural HRP. Besides, the catalytic activities display a change tendency of increase first and then decrease as the reaction temperature rises (Figure b and Figure S12c,d). And the peroxidase-like reaction system exhibits an activity–temperature relationship similar to that of the oxidase-like reaction system. As such, 25 °C is chosen for the best catalytic performance. According to Figure c and Figure S12e,f, the absorbance at 652 nm increases with time and then remains constant after 10 min. In addition, Figure d–f and Figure S12g–l show that the absorbance is concentration-dependent. To maximize the enzyme-like catalytic activities, the optimal concentration of CoS1.035/N-C PDHs, TMB, and H2O2 is also identified as 15 μg·mL–1, 50 μg·mL–1, and 10 mM, respectively. Consequently, in the subsequent experiments, the optimized conditions are as follows: CoS1.035/N-C PDHs = 15 μg·mL–1, TMB = 50 μg·mL–1, H2O2 = 10 mM, pH = 5.0, T = 25 °C, and t = 10 min.

Figure 8

(a) The effect of solution pH on the catalytic activity of CoS1.035/N-C. (b) The effect of temperature on the catalytic activity of CoS1.035/N-C. (c) The effect of time on the catalytic activity of CoS1.035/N-C in the absence and presence of H2O2. (d) The effect of CoS1.035/N-C concentration on the catalytic activity of CoS1.035/N-C. (e) The effect of TMB concentration on the catalytic activity of CoS1.035/N-C. (f) The effect of H2O2 concentration on the catalytic activity of CoS1.035/N-C. The error bars represent the standard deviation values of three measurements.

Considering the influence of leaching ions on the catalytic activity, the CoS1.035/N-C PDHs catalyst was immersed in the above reaction solution under experimental conditions for 25 days, and then the supernatant was collected by centrifugation. Following that, the supernatant was used to catalyze TMB in the presence and absence of H2O2. Figure S13a demonstrates that the peroxidase- and oxidase-like catalytic activities are not dependent on the leaching ions; that is to say, the dual-enzyme mimic catalytic activities of CoS1.035/N-C PDHs are derived from the nanomaterial itself. What’s more, the stability and reproducibility of CoS1.035/N-C PDHs were tested by recycling and reusable experiments. After five cycles, the catalytic activity of CoS1.035/N-C PDHs still remained at 70.3 and 59.8% of their initial activity, revealing the favorable reproducibility (Figure S13b). Beyond that, the catalytic activities are almost unchanged after preserving in an aqueous solution for 25 days (Figure S13c). The excellent stability is superior to many cobalt-based mimic enzymes, as shown in Table S4. Moreover, no obvious change was found in the XRD data (Figure S14a) and SEM image (Figure S14b) after the stability test, further indicating that CoS1.035/N-C PDHs own outstanding long-term stability.

Determination and Colorimetric Assay of H2O2

H2O2 is a biomolecule widely used in food, chemistry, environmental conservation, and medical treatment fields. In chemical production, H2O2 can be applied to the preparation of fine chemicals. The addition of H2O2 in the food process can kill food spoilage bacteria and play a preservative role;[41] in the field of environmental remediation, it is used for environmental pollutant control and wastewater treatment;[42] in medical treatment, H2O2 is used for wound treatment, disinfection, and sterilization.[43] In addition, as one of the representative reactive oxygen species, the content of H2O2 in the body is also closely related to many diseases, such as Alzheimer’s disease,[44] Parkinson’s disease,[45] chronic obstructive pulmonary disease (COPO),[46] and cancer.[47] Therefore, it is very critical to construct a fast and sensitive real-time analysis platform of H2O2 for environmental protection, food, and disease diagnosis. Due to the advantages of being visible to the naked eye, fast response, low cost, and simple operation, the colorimetric method for the determination of H2O2 has gradually captured people’s attention.

Therefore, based on the optimal reaction conditions, a H2O2 sensing platform was engineered. Figure a is the UV–vis absorption spectrum. When there is no H2O2, the absorbance value of the system is the lowest and the blue color of the solution is the lightest (embedded photo (a) in Figure a). With the increase of H2O2 content, the absorbance value becomes larger and the visual blue becomes darker (embedded graphs (b–j) in Figure a). Besides, as illustrated in Figure b, in a certain range, the concentration of H2O2 has a linear relationship with the absorbance. The linear range of H2O2 concentration is 0.5–20 μM (R1 = 0.9979) and 20–120 μM (R2 = 0.9978), respectively. The detection limit is 50 nM, which is comparable to or better than that of previous reports (see Table S4). These results show that CoS1.035/N-C PDHs with peroxidase-like activity have a wide linear range (0.5–120 μM) and quite low detection limit (50 nM) as a H2O2 sensor and can be used for the colorimetric detection of H2O2 just by color comparison.

Figure 9

(a) The characteristic spectra of CoS1.035/N-C (15 μg/mL) + TMB (50 μg/mL) + HAc-NaAc buffer solution (pH = 5) + different doses of H2O2 (from 0 to 120 μM) system at 652 nm. Inset: a color contrast photo of H2O2 detection from 0 to 120 mM. (b) A concentration–response curve for the sensitive determination of H2O2 under the optimized conditions. Inset: a liner calibration plot for the detection of H2O2. The error bars represent the standard deviation of three measurements. (c) Anti-interference of the TMB turn to be blue under the H2O2 and CoS1.035/N-C with peroxidase mimics. From left to right: blank, Ca2+, Na+, Mg2+, Al3+, Zn2+, Co2+, Ni2+, Fe3+, Ba2+, Li+, and Cr3+ (the concentration of the substance in the system: TMB, 50 μg/mL; HAc-NaAc buffer solution, pH 5.0; CoS1.035/N-C, 15 μg/mL; H2O2, 10 mM; and interference ions, 50 mM).

To verify the reliability of the existing method, several common interferences in lakes and tap water were measured, as shown in Figure c, namely, Ca2+, Na+, Mg2+, Al3+, Zn2+, Co2+, Ni2+, Fe3+, Ba2+, Li+, and Cr3+, which indicated that the proposed method had good anti-interference ability. Additionally, to further confirm the potential application of this method in H2O2 detection, the H2O2 concentration in labeled real samples was detected. The recoveries were investigated by adding a known amount of the standard H2O2 solution to the sample. As shown in Table S5, the average recoveries of H2O2 at three spiked levels range from 82.0 to 110.6%. The satisfactory recovery shows that the method has a potential application prospect in the determination of H2O2 concentration in lakes and tap water.

Determination and Colorimetric Assay of GSH

Glutathione (GSH) is the most abundant tripeptide thiol in eukaryotic cells, and an abnormal GSH level is able to reflect many types of diseases, such as liver injury, cancer, and acquired immune deficiency syndrome.[9] Therefore, it is necessary to establish a simple and reliable detection method of GSH level for disease diagnosis. Considering that GSH is an antioxidant, it can inhibit the oxidation of TMB. The nanozymes with peroxidase- and oxidase-like activity can be applied to detect such antioxidants.[48,49] So a biosensor is constructed with the CoS1.035/N-C PDH nanozyme to realize the visual detection of GSH. As witnessed in Figure a, with the increase of GSH concentration, the absorbance at 652 nm gradually declines accompanied with color fading of the reaction system. Figure b exhibits the calibration curve for detecting GSH under the optimal reaction conditions. The values of ΔA652 (ΔA652 = Ablank – AGSH) are proportional to GSH concentrations in the linear range of 1–60 μM with a low detection limit of 42 nM (S/N = 3). The detection limit based on this method is superior to most limits reported in the literature to date (see Table S6). More importantly, the CoS1.035/N-C PDH nanozyme also displays a prominent selectivity and anti-interference ability for detecting GSH. As shown in Figure c,d, various types of inorganic metal ions and biological molecules, for instance, K+, Na+, Mg2+, Al3+, Ca2+, Ni2+, Co2+, Zn2+, Ba2+, tryptophan, glycine, lysine, aspartic acid, glutamic acid, serine, methionine, arginine, valine, and threonine, are adopted as latent interferences. We found that even if the concentration of the latent interferences is 80 times higher than that of GSH, there is no significant effect on GSH detection. The results indicate the great potential of CoS1.035/N-C PDHs to detect the GSH.

Figure 10

(a) The UV–vis absorption spectrum changes of the mixing solution consisting of TMB (50 μg/mL), CoS1.035/N-C catalyst (15 μg/mL), and H2O2 (10 mM) in the absence or presence of varied concentrations of GSH. (b) The linear calibration plot for GSH detection. (c) Anti-interference performance of the GSH detection. From left to right: blank, Na+, K+, Mg2+, Al3+, Ca2+, Ni2+, Co2+, Zn2+, and Ba2+. The concentration of the substance in the system: TMB, 50 μg/mL; HAc-NaAc buffer solution, pH 5.0; CoS1.035/N-C, 15 μg/mL; H2O2, 10 mM; GSH, 60 μM; and interference ions, 5 mM. (d) Selectivity of this detection platform for GSH assay. From left to right: GSH, tryptophan, glycine, lysine, aspartic acid, glutamic acid, serine, methionine, arginine, valine, and threonine (the concentration of the substance in the system: TMB, 50 μg/mL; HAc-NaAc buffer solution, pH 5.0; CoS1.035/N-C, 15 μg/mL; H2O2, 10 mM; GSH, 60 μM; and other amino acids, 60 μM). The error bars represent the standard deviation values of three measurements.

Conclusions

In summary, we have successfully designed the CoS1.035/N-C PDHs as bifunctional enzyme mimics with superior peroxidase-like and oxidase-like activities using a facile wet chemical method followed by high heat treatment. It can rapidly convert colorless TMB into blue ox-TMB in the presence and absence of H2O2. And the catalytic mechanism of CoS1.035/N-C PDHs toward the dual-enzyme mimic activities was revealed by investigating the reactive oxygen species. The satisfactory dual-enzyme mimic activities of CoS1.035/N-C PDHs are ascribed to the structural advantages, such as the abundant pores, huge exposure surface area, N atom doping, and stable carbon framework. Therefore, the bifunctional platform for the colorimetric detection of H2O2 and GSH has been developed, which exhibits a wide linear range and a low detection limit. The present research opens the way for the development of Co-based chalcogenides as dual-enzyme mimics in the fields of biochemical sensing, biomedicine, and environmental monitoring.

Experimental Section

Chemicals and Materials

Cobalt nitrate hexahydrate (Co(NO3)2·6H2O, 99.99%), 2-methylimidazole (C4H6N2, 98%), and sulfur sublimed (S, 99.99%) were provided by Chengdu Kelong Chemical Co. Selenium (Se, 99.99%) was obtained from SAAN Chemical Technology Co. Tellurium (Te ,99.99%) was purchased from Shanghai Macklin Biochemical Co., Ltd. Ethyl alcohol (C2H5OH, 99.5%, AR), Ltd.Ltd. 3,3′,5,5′-Tetramethylbenzidine dihydrochloride (TMB, 98%) was obtained from Aladdin. Hydrogen peroxide solution (H2O2, 30%) was obtained from Chuandong Chemical. In these experiments, all raw materials were not further purified, and deionized (DI) water (18.2 MΩ·cm) was used throughout the experiments.

Apparatus

The nanomaterials were characterized by different techniques. Field-emission transmission electron microscopy (FETEM, FEI, Tecnai G2 F20, 200 kV) and scanning electron microscopy (FESEM, Hitachi, SU820, 3.0 kV) were used to analyze the micromorphology and surface structure of the as-obtained products. Elemental mapping data were collected by SEM. The crystalline structure and phase purity of the samples were analyzed by X-ray diffraction (XRD) on an XRD-6100 X-ray diffractometer (2 kW, NF, Cu tube) at room temperature. X-ray photoelectron spectroscopy (XPS) data were obtained with an ESCALAB 250 X-ray photoelectron spectrometer (Thermo, USA). Binding energies (BEs) were calibrated by setting the measured BE of C 1s to 284.8 eV. EPR data were collected by an A300-10/12 electron paramagnetic resonance spectrometer (Bruker, Germany). The absorption spectra were measured using a UV-2550 UV–vis spectrophotometer (Shimadzu, Kyoto, Japan).The zeta potential data were obtained by dynamic light scattering (Brookhaven, USA). A dissolved oxygen meter (Beijing, Chiana) was adapted to analyze the content of dissolved oxygen in the system. A high-speed TGL-16M centrifuge (Xiangyi, China) was used in the purification of samples.

Synthesis of ZIF-67 and Its Derivatives

Preparation of ZIF-67

The ZIF-67 dodecahedrons were synthesized by a familiar room-temperature precipitation method.[1] First, 300 mg of Co(NO3)2·6H2O was dissolved in 150 mL of CH3OH to form solution A, and 5 g of 2-methylimidazole (C4H6N2) was dissolved in 150 mL of CH3OH to gain solution B. Then, solution B was rapidly added into solution A under magnetic stirring. Subsequently, the synthetic mixture was stirred for 24 h at room temperature. The final product was centrifuged and washed several times with CH3OH. Finally, after drying at 60 °C for 12 h, the ZIF-67 dodecahedrons were obtained.

Synthesis of Co/N-C PDHs

The Co/N-C PDHs with the morphology of ZIF-67 were prepared by pyrolyzing the as-obtained ZIF-67 dodecahedrons in a N2 atmosphere. First, 100 mg of ZIF-67 dodecahedrons was spread in a graphite boat. Then, the graphite boat was placed in a tube furnace; after calcining the graphite boat at 600 °C for 1.5 h under a N2 atmosphere, the black powder of Co/N-C PDHs was obtained.

Synthesis of Co3O4/N-C PDHs

The as-prepared Co/N-C PDHs were first spread in the graphite boat, and then the graphite boat was placed in a tube furnace to calcine for 1.5 h at 600 °C. Afterward, the CoO/N-C was obtained.

Synthesis of CoM/N-C (M = S, Se, Te) PDHs

First, 50 mg of the as-prepared Co/N-C PDHs and 30 mg of sulfur sublimed/selenium powder/tellurium powder were placed downstream and upstream of the graphite boat, respectively. Then, the graphite boat was transferred to a tube furnace to calcine for 1.5 h at 600 °C at a heating rate of 0.5 °C/min under a N2 atmosphere. Finally, after natural cooling, the CoM/N-C was obtained.

Evaluation of the Enzyme-Like Activity of Co1.035S/N-C PDHs

With TMB as the chromogenic substrate, the peroxidase-mimicking catalytic capacity of Co1.035S/N-C PDHs was evaluated. In the presence of H2O2, Co1.035S/N-C PDHs can rapidly catalyze the oxidation of colorless TMB into blue oxidized TMB. First, the steady-state kinetic study of Co1.035S/N-C PDHs was carried out. Then, with the evolution of reaction time, the absorbance of the available reaction system at 652 nm was recorded by a UV-2550 spectrophotometer. Experimental conditions were as follows: Co1.035S/N-C PDHs, 15 μg·mL–1; TMB, 50 μg·mL–1; H2O2, 10 mM; HAc-NaAc buffer solution, pH = 5; and reaction time, 10 min. After 10 min of reaction, the absorbance of the system reached a stable value. Hence, this result reveals that the optimized reaction time is 10 min. Following that, under the optimized time, other factors affecting the peroxidase-like capacity of Co1.035S/N-C PDHs such as pH and substrate concentration (including TMB and H2O2) were elaborately investigated. In addition, the affinity between Co1.035S/N-C PDHs and the substrate was investigated by calculating Km. Keeping the content of catalysts constant and altering the concentration of substrate, the reaction kinetics of the system were recorded by a UV-2550 spectraphotometer. Finally, the Lineweaver–Burk plot was applied to fit the obtained curve and analyze data.