Natural enzyme is produced by living cells, and participates in almost all life activities, which has a strong specific catalytic activity and a well application prospect in many fields^[1,2]. However, the natural enzyme content in animals is extremely low, and quite sensitive to physical and chemical factors such as reaction pH, temperature, etc., which limits the application in various field^[3,4,5]. Therefore, it is of great significance to develop a mimic enzyme with high catalytic efficiency and stable reaction, and make the catalytic mechanism clear. So far, the reported materials with enzymatic activity are mainly including: 1) noble metal nanomaterials such as platinum, gold, and palladium, which have been reported to exhibit various enzyme-like activities such as peroxidase, superoxide dismutase, and catalase^[6,7,8]; 2) metal oxides and metal sulfide nanomaterials, for example, Fe₃O₄ nanomaterials showing peroxidase activity^[9], and Co₃O₄ displaying catalase activity for colorimetric detection SO₃^2-[10]; 3) non-metallic nanomaterials, for example, carbon nanotubes and graphene oxide, having peroxidase-like activity^[11,12,13]. Among them, Pt and Pt-based bimetallic or multi-metallic nanoparticles have a variety of enzyme activities and have been applied in many fields. For example, Kong et al.^[14] have synthesized Pt NPs in situ using DNA as a template, and used the catalase activity to detect nucleic acids. At present, in the aspect of colorimetric detection, there are many applications of catalase- like activity of Pt and Pt-based nanomaterials. Therefore, it is meaningful to study their oxidase-like activity.

Ascorbic acid is an essential nutrient for the human body. It has functions of regulating human immunity, preventing gum bleeding and promoting certain amino acid metabolism^[15,16]. However, the human body cannot self-synthesize ascorbic acid and must be taken from the outside world such as food and medicine, and excessive intake or lack of intake can cause certain diseases^[17]. The main sources of ascorbic acid required by the human body are vegetables and fruits^[18]. Therefore, it is important for us to accurately measure the ascorbic acid content in food or drugs using a simple method. At present, the main methods for determining ascorbic acid content include gas chromatography, fluorescence, electrochemical analysis and electrophoresis^[19,20,21]. All of them have the disadvantages of complicated process, time consuming and high cost. Therefore, it is very important to develop an efficient, economical and easy to operate method. Among all detection methods, colorimetric biosensing has a unique advantage because of high sensitivity and low cost^[22,23]. It is based on the relationship between the color of the solution and the absorbance. The signal of detection events is transformed into the color of the reactant. So it has the characteristics of visualization, makes it possible for field analysis^[24,25].

Therefore, we adoped a facile modified polyol process, using polyvinylpyrrolidone (PVP) as a surface blocking agent, a certain concentration of hydrochloric acid as an oxidizing etchant, and prepared Pt-Au dendritic nanoparticles by adjusting precursor ratio and reaction temperature. The oxidase-like activity was tested using TMB as a chromogenic substrate. Furthermore, we have established a new method for colorimetric detection of ascorbic acid, in which the presence of AA can cause a change in the color of the reaction system and is visible.

Chloroplatinic acid hexahydrate (H₂PtCl₆·6H₂O, chloroauric acid trihydrate (HAuCl₄∙3H₂O) were obtained from Sigma-Aldrich. And ethylene glycol (C₂H₆O₂), 3,3′,5,5′-tetramethylbenzidine (TMB), PVP were purchased from Alfa Aesar. Hydrochloric acid, sulfuric acid, ethanol, n-hexane, and acetic acid- sodium acetate buffer solution (0.2 mol/L, pH 4.0) were purchased from Aladdin, China. The water used was deionized water (18 MΩ·cm). All the chemicals were used without further purification.

Absorbance spectra of the reaction solution was measured using an ultraviolet-visible spectrophotometer (Lamda950), and the cuvette path length was 1 cm. A transmission electron microscope (TEM, Tecnai G2 F20) and a scanning electron microscope (SEM, FEI-Magellan) were used to observe the sample morphology. The infrared spectrum was obtained by a Fourier infrared spectrometer (NICOLET Is10). X-ray diffraction (XRD) characterization was performed by a D8 ADVANCE diffractometer.

In a 50 mL three-necked flask, 2.5 mL of ethylene glycol was refluxed at 180 ℃ for 5 min with oil bath. After which, 0.34 mL of HCl (0wt%-37wt%) was added. Then, 0.94 mL of a mixed solution of H₂PtCl₆·6H₂O (0.0625 mol/L) and HAuCl₄∙3H₂O (0.0625 mol/L) (both in molor ratio of 9 : 1-5 : 1) and 4 mL of PVP were added into the above solution for 10 times within 5 min. During the reaction, the reactants were continuously stirred and mixed uniformly. After 20 min, the flask was taken out, naturally cooled to room temperature. Finally, the sample was collected by centrifugation.

60 μL Pt-Au DNPs (0.75 mg/mL), 140 μL of TMB, were added to 2.8 mL of acetic acid-acetate buffer solution (0.2 mol/L, pH 4.0) at 35 ℃. The kinetic measurements were performed in a time course mode using a spectrophotometer to measure the absorbance spectra at the first 3 min of the reaction system at 652 nm every 15 s. Lineweaver-Burk Graphs was used to calculate Michaelis-Menten constants (K_m):

1/v=K_m/V_max(1/[S]+1/K_m)

where [S] is the substrate concentration, and V_max is the maximum velocity of the reaction. K_m: Michaelis constant, V_max: maximum reaction rate.

40 μL TMB (5 mmol/L) and 60 μL Pt-Au DNPs solution (0.75 mg/mL) were added to 2.9 mL acetic acid- sodium acetate buffer solution (0.2 mol/L, pH 4.0), then incubated the solution at 35 ℃ with water bath. UV- visible absorption spectra were used to trace the reaction.

40 μL TMB(5 mmol/L) and 60 μL Pt-Au DNPs solution (0.75 mg/mL) were added to 2.8 mL acetic acid- sodium acetate buffer solution (0.2 mol/L, pH 4.0). After that, the mixture was incubated at 35 ℃ for 10 min. Than 50 μL AA solution were added into it. Their UV- visible absorption spectra were measured at 200-800 nm.

Fig. 1(a, b) show SEM and TEM images of the synthesized Pt-Au DNPs, with diameter of about 35 nm. It can be clearly seen that the nanoparticles have a distinct dendritic structure, which has the effect of increasing the specific surface area of the prepared nanomaterial. At the same time, the agglomeration of nanoparticles is a little serious, probably due to the large specific surface area (Fig. 1(a)). The HRTEM further indicates that the lattice spacing of the parallel stripes is 0.231 nm (Fig. 1(c)), which corresponds to the (111) crystal plane and between the lattice constant of Pt (0.228 nm) and Au (0.235 nm)^[25], indicating that alloy structure may be formed. EDS elemental mapping images show that Pt and Au in the nanoparticles are homogeneously distributed which supports that the Pt-Au DNPs are bimetallic alloy (Fig. (d-1, d-2)) (EDS elemental analysis showed in Fig. S1). The XRD pattern also confirmed the formation of the AuPt alloy (Fig. S2). ICP-OES analysis demonstrated that the nanoparticles with an Au/Pt molar ratio of 4.66/1, which corresponded to the proportion of the added precursor.

The reaction time is critical to the morphology of as-synthesized Pt-Au DNPs. To explore the formation process, the nanoparticles prepared for different reaction-time were characterized. As show in Fig. 2, the nanoparticles which had dendritic structure were observed after the reaction for 5 min (Fig. 2(a)). When the reaction time was 10 min, the dendritic particles grew bigger (Fig. 2(b)). Until the reaction was extended to 20 min (Fig. 2(c)), Pt-Au DNPs were obtained with good dispersibility, high dendritic structure and uniform size. The yielded nanoparticles became smaller and the dendritic structure gradually disappeared when the reaction time was 40 min (Fig. 2(d)), which may result from the fact that the prolonged heat treatment damaged the instable nanoparticle structure. This indicates that reasonable control of reaction time is helpful to obtain materials with good morphology and excellent properties.

It is known from the existing literature that HCl is an important additive for the formation of dendritic structure, and O₂/Cl^- is a commonly used oxidizing etchant and has the function of regulating the morphology of nanoparticles^[26]. Currently, there are many reports that O₂/Cl^- can perform the better characteristics of oxidative etching in acidic environment than those in alkaline and neutral environment, and hydrochloric acid with high concentration is beneficial to the formation of Pt nanoparticle with dendritic structure^[26,27,28]. As shown in Fig. 3 (a, b), the particles agglomeration was severe and did not form when the concentration of HCl was 0wt% or 18.5wt%; and these nanoparticles with dendritic structure and uniform size were formed when the concentration of HCl 37wt% (Fig. 3(c)). At the same time, it was found that the progress of the reaction became slower and the yield became lower (Fig. S3) as the concentration of HCl increased, indicating that the high concentration of HCl was very effective for oxidative etching.

PVP is also a commonly used morphology control agent for the preparation of noble metal nanoparticles, which can be adsorbed on the crystal surface to regulate anisotropic growth. At the same time, PVP acts as a stabilizer and a blocking agent, and also could protect the nanoparticles from agglomeration^[29,30,31]. When the PVP concentration was 0.0187 mol/L (Fig. 4(a)), only some little particulate matter could be seen at the edge of the material, and a uniform stable structure could not be formed, which may indicate that the amount of PVP is insufficient to coat a specific crystal face (previous HRTEM analysis showed that the main exposed crystal plane of Pt-Au DNPs was (111) plane), and some studies suggest that PVP is selective for crystal plane coating, which makes it unable to uniformly anisotropic grow^[32]. When PVP concentration was 0.0375 mol/L the grain structure became regular and uniform in size, as shown in Fig. 4(b). And when the PVP concentration was increased to 0.0750 mol/L, it’s found that the higher concentration had little effect on its morphology. However, the amount of organic matter adsorbed on the surface of the nanoparticles increased as shown in Fig. 4(c), and the transmission electron microscope showed that the surface of the nanoparticles was covered with a film. Infrared test result indicated that the main component of the film was PVP (Fig. S4).

Using TMB as a substrate, the oxidase-like activity of Pt-Au DNPs was assessed by testing the degree of oxidation of TMB without adding H₂O₂. As shown in Fig. 5, Pt-Au DNPs catalyzed the oxidation of TMB to produce a typical blue product with a strong absorption peak at 652 nm, which is consistent with the literature ^[33,34]. TMB was almost completely not oxidized when no catalyst was added. The catalytic performance of Pt-Au DNPs was slightly decreased after surface deoxidation treatment of Pt-Au DNPs (200 ℃, holding for 2 h under Ar atmosphere), although the reduction was much weaker than the decrease of catalyst performance after the reaction solution was purged with nitrogen (Fig. 5), indicating that the source of oxygen in the catalytic oxidation of TMB included both adsorbed and dissolved oxygen, and the effect of dissolved oxygen on catalytic performance might be greater. Meanwhile, this experiment showed that when the reaction conditions were changed, such as contents of hydrochloric acid and PVP, the catalytic performance was basically consistent with its morphological change. The more obvious the nano-dendritic structure, the better the catalytic performance. However, there were exceptions, when the PVP concentration was 0.0750 mol/L, the morphology of the nanoparticles did not change obviously but catalytic activity was very low (Fig. S5), implying that a large amount of PVP might occupy the surface active site of the catalyst, which reduced the performance. The catalytic activity of Au-Pt alloy NPs increased with the increase of Pt / Au ratio (Fig. S6), consistent with the literature^[35].

Different experimental conditions had a great influence on the oxidase-like activity of Pt-Au DNPs. Therefore, the optimal reaction conditions were explored by changing the concentration of Pt-Au DNPs, temperature, pH, and reaction time. As shown in Fig. 6(b, d), the absorption intensity of the reaction system reached the maximum at 652 nm when the temperature was lower at 35 ℃ and the pH of the reaction solution was 4.0, indicating that the activity of Pt-Au DNPs was strongest under this reaction conditions. The above results could be confirmed by the color depth of ox-TMB solution (inset in Fig. 6(c, d)). Similarly, the TMB oxidation reaction under different concentrations of catalyst (Fig. 6(a, c)) showed that the absorption intensity increased with the increase of Pt-Au DNPs concentration and reaction time. When the concentration of Pt-Au DNPs reached 15 μg/mL and the reaction time reached about 8 min, the increasing rate of absorption intensity gradually slowed down. Therefore, to facilitate experimentation and save catalyst usage, the following experiments were carried out using the reaction conditions of pH 4.0, temperature 35 ℃, 15 μg/mL Pt-Au DNPs, and 10 min of reaction time.

To further understand oxidase activity of Pt-Au DNPs, the time course mode was used to explore initial velocity of the reaction and kinetic parameters. The Michaelis-Menten curve was established by changing TMB concentration and tracking the absorbance of the reaction solution at 652 nm for the first 3 min of the reaction (Fig. 7(a)). At the same time, the K_m of Pt-Au DNPs for TMB oxidation was calculated to be 0.22 mmol/L, and the V_max was 282 nmol/(L·s) according to the Lineweaver- Burk plot (Fig. 7(b)). For natural enzymes, the smaller the K_m, the higher the affinity of the enzyme to the substrate. At the same time, as shown in Table S2, the K_m obtained in this experiment was compared with those reported in other literatures, and the value was found to be small, indicating that the affinity of Pt-Au DNPs with TMB is high. Comparing the measured K_m with the products obtained by changing the reaction parameters, it’s found that the lower K_m value was accompanied by a better morphology of the catalyst (obvious and dispersioned dendritic structure). But when preparing the product by doubling PVP amount, the K_m value was rather high (Fig. S7), which might link to the PVP adsorption on the surface of the nanoparticles occupying the active site.

To better understand the catalytic process, different concentrations of AA were added before the reaction, and then the intensity of absorbance peak at 652 nm was measured as a function of time. Fig. 8(a) showed that AA delayed the reaction, and the higher the AA concentration, the stronger the retardation was, but there were not substantially effect on the plot slope. It indicated that AA was an effective antioxidant, and TMB started to oxidize when AA was consumed. As shown in Fig. 8(b), the absorption intensity decreased instantaneously when AA was added, indicating that AA could effectively reduce the already oxidized TMB, and over time, the reduced TMB was oxidized again. However, as the number of AA additions increased, the TMB oxidation rate became slower. This implied that the continuous reaction would reduce the catalytic performance of Pt-Au DNPs, which might be related to catalyst being poisoned. Meanwhile, the performance of Pt-Au DNPs was severely reduced by continuous reaction (Fig. S8(a)), but centrifugal washing could obviously restore its performance. Fig. S8(b) showed that after prolonging the reaction time to re-oxidize TMB to the same level as the first test, the limit of detection (LOD) (87 nmol/L) obtained by retesting AA was not much different from the first test (78 nmol/L), indicating the possibility of recycling. At the same time, Pt-Au DNPs could directly oxidize AA (Fig. S9). Based on the above research, the scheme and design principle of testing AA are shown in Scheme 1, containing two steps: TMB is oxidized to ox-TMB, and ox-TMB is reduced by AA.

As described above, Pt-Au DNPs have oxidase-like activity and can catalyze the oxidation of TMB to form a blue product (ox-TMB). By using AA to inhibit the activity of Pt-Au DNPs and reduce ox-TMB, a new colorimetric assay was formed. Based on this detection method described in above Materials and Methods, a standard curve for detecting AA was established. As shown in Fig. 9(a), the absorbance value at 652 nm gradually decreased with increasing AA concentration, indicating that the content of ox-TMB and the addition of AA content satisfied a certain relationship. The analytical curve showed a linear range of 1-15 μmol/L (Fig. 9(a)). After calculation, the LOD was 78 nmol/L. It indicated that our method was more sensitive than those previously reported results (Table S3). Fig. S10 showed that the product performance was best when the precursor ratio is 5 : 1. To test the selectivity of this method, several ions (HCO₃^-, Na⁺, Br^-, K⁺, Al³⁺, NO₃^-, Cu²⁺, Cl^-, I^-, F^-, SO₄^2-, Zn²⁺, CH₃COO^-), glucose, lysine, and so on were used. However, the results showed that these substances had little effect on the reduction of oxidase-like activity of ox-TMB and Pt-Au DNPs, which could be neglected (Fig. 9(b)). After the method was put into detecting AA in real sample (Table S1), the recoveries of the sample range were from 103.68% to 103.71%. Therefore, the method is considered to be an effective tool for determining AA content.

In summary, Pt-Au DNPs is synthesized through a modified polyol process, in which PVP and HCl are served as shape-control agent and the oxidase-like activity of nanoparticles can be adjusted by controlling the amount of the additions. Kinetic studies show a typical Michaelis-Menten kinetics, indicating that Pt-Au DNPs has good affinity for TMB. And AA can effectively inhibit the oxidation of TMB, while reducing the already ox-TMB, and AA itself can also be oxidized by Pt-Au DNPs. In particular, Pt-Au DNPs has a certain reusable potential. Based on above reaction process, a simple colorimetric method is established for detecting AA with high sensitivity and selectivity. This study provides some ideas for its application in food and drug testing, optical sensing and other aspects.

Supporting materials related to this article can be found at https://doi.org/10.15541/jim20200005

Pt-Au Dendritic Nanoparticles with High Oxidase-like Activity for Detection of Ascorbic Acid

CHENG Qin^1,2, YANG Yong², YANG Lili²

1. Shanghai Applied Radiation Institute, Shanghai University, Shanghai 201800, China; 2. State Key Laboratory of High Performance Ceramics and Superfine Microstructure, Shanghai Institute of Ceramics, Chinese Academy of Sciences, Shanghai 200050, China

参考文献:

[1]CHAI D H, MA Z, YAN H, et al. Synergistic effect of sandwich polyoxometalates and copper-imidazole complexes for enhancing the peroxidase-like activity. RSC Advances, 2015, 5: 78771-78779.

[1]ASATI A, SANTRA S, KAITTANIS C, et al. Oxidase-like activity of polymer-coated cerium oxide nanoparticles. Angewandte Chemie-International Edition, 2009, 48(13): 2308-2312.

[2]WANG G L, XU X F, CAO L H, et al. Mercury(II)-stimulated oxidase mimetic activity of silver nanoparticles as a sensitive and selective mercury(II) sensor. RSC Advances, 2014, 4(12): 5867-5872.

[3]YADAV P K, SINGH V K, CHANDRA S, et al. Green synthesis of fluorescent carbon quantum dots from azadirachta indica leaves and their peroxidase-mimetic activity for the detection of H₂O₂ and ascorbic acid in common fresh fruits. ACS Biomaterials Science & Engineering, 2019, 5(2): 623-632.

[4]DARABDHARA G, SHARMA B, DAS M A, et al. Cu-Ag bimetallic nanoparticles on reduced graphene oxide nanosheets as peroxidase mimic for glucose and ascorbic acid detection. Sensors and Actuators B-Chemical, 2017, 238: 842-851.

[5]WANG N, LI B, QIAO F, et al. Humic acid-assisted synthesis of stable copper nanoparticles as a peroxidase mimetic and their application in glucose detection. Journal of Materials Chemistry B, 2015, 3: 7718-7723.

[6]Yu C J, CHEN T H, JIANG J Y, et al. Lysozyme-directed synthesis of platinum nanoclusters as a mimic oxidase. Nanoscale, 2014, 6(16): 9618-9624.

[7]AN Q, SUN C, LI D, et al. Peroxidase-like activity of Fe₃O₄@carbon nanoparticles enhances ascorbic acid-induced oxidative stress and selective damage to PC-3 prostate cancer cells. ACS Applied Materials & Interfaces, 2013, 5(24): 13248-13257.

[8]WU T, MA Z, LI P, et al. Colorimetric detection of ascorbic acid and alkaline phosphatase activity based on the novel oxidase mimetic of Fe-Co bimetallic alloy encapsulated porous carbon nanocages. Talanta, 2019, 202: 354-361.

[9]MI C, WANG T, ZENG P, et al. Determination of ascorbic acid via luminescence quenching of LaF₃:Ce,Tb nanoparticles synthesized through a microwave-assisted solvothermal method. Analytical Methods, 2013, 5(6): 1463-1468.

[10]WANG X, HAN Q, CAI S, et al. Excellent peroxidase mimicking property of CuO/Pt nanocomposites and their application as an ascorbic acid sensor. Analyst, 2017, 142: 2500-2506.

[11]GAO C, ZHU H, CHEN J, et al. Facile synthesis of enzyme functional metal-organic framework for colorimetric detecting H₂O₂ and ascorbic acid. Chinese Chemical Letters, 2017, 28: 1006-1012.