Introduction

Ascorbic acid (AA) is an important biological cofactor and antioxidant, which participate in various biochemical and physiological processes [1]. Abnormal levels of AA are associated with scurvy, cancer, cardiovascular disease, and some mental illness [2,3,4,5]. It is of great importance to develop convenient and rapid methods for AA detection with high sensitivity in a vast range of areas such as drugs, foods, and clinical diagnosis. Recently, various methods have been exploited for AA determination, including high-performance liquid chromatography (HPLC), capillary electrophoresis, electrochemistry, chemiluminescence, and fluorescence [6,7,8,9,10,11,12]. Colorimetric assay has emerged as an important sensing platform due to its rapidity, simplicity, low-cost and easy detection by naked eyes. Nowadays, enzyme-like nanomaterials-based sensors have drawn great interest as colorimetric tools [13].

Nanozyme is a class of nanomaterials that possess intrinsic enzyme-like properties, which can be applied as an enzyme substitute to catalyze the biochemical reactions of the substrates under physiological conditions [14]. Owning to the advantages of low cost, high stability, and ease of production, nanozyme has excellent prospects in molecular detection and some other areas. The peroxidase and oxidase-like nanomaterials can efficiently catalyze the oxidation of colorless substrates to produce colored products, which can be used as colorimetric sensors [15, 16]. Different from peroxidase-mimics, oxidase-mimicking nanozymes can directly use dissolved oxygen molecules as the oxidant [17]. Herein, the oxidase-mimics are much more attractive in biosensing, which can effectively eliminate the interferences of the extra addition of H2O2. Most of the reported oxidase-mimics concerned in metal oxide nanoparticles and noble metal nanoparticles [15], such as Au [18], Pt[19], CeO2 [20], NiO [21], MnO2 [22], Mn3O4 [23], and V2O5 nanozymes [24]. However, the catalytic activities of these metal nanomaterials are severely deteriorated by the spontaneous aggregation and instability. Improving the catalytic activity and stability of the nanoparticle-based nanozyme is still challenging.

The growth of nanoparticles on specific support can efficiently prevent aggregation and increase active sites, which is an effective strategy to enhance the catalytic activity [25,26,27]. Graphene nanosheets, as an excellent two-dimensional (2D) carbon support, is regarded as a promising candidate due to the characteristics of high mechanical strength, large surface area, high chemical stability, and good biocompatibility [28,29,30]. However, the agglomeration and restacking of graphene sheets can lead the decrease of active sites. Porous graphene, which possesses abundant pores on the surface, could efficiently avoid agglomeration and increase the surface area [31,32,33,34]. We have proposed several strategies to deposit different nanoparticles (including TiO2 nanoparticles, Fe/Ni bimetallic nanoparticles, Zn-BTC MOF, and so on) on the surface of porous graphene, which exhibited good potential applications in organic pollutant degradation [35], rare earth adsorption [36, 37], separation [38], and molecule detection [39].

In this work, we successfully combined porous reduced graphene oxide nanosheets (p-rGO) with Mn3O4 nanoparticles (Mn3O4@p-rGO) through an in-situ fabrication strategy. Mn3O4 nanoparticles were synthesized by reusing the absorbed Mn2+ on the surface of GO as the precursor, and the p-rGO with porous structure was obtained by directly calcinating in open air. The decoration of Mn3O4 nanoparticles on the surface of p-rGO nanosheets effectively enhanced the oxidase-like activity owing to the larger surface area, more active sites, and accelerated electron transfer efficiency (as shown in Scheme 1). Furthermore, a simple and fast colorimetric sensor was constructed for AA detection with satisfactory sensitivity and selectivity. It’s proved that Mn3O4@p-rGO is a promising oxidase alternative with potential applications in molecular sensing and disease diagnosis.

Scheme 1
scheme 1

Schematic illustration the mechanism the boosted oxidase-like activity of Mn3O4@p-rGO

Full size image

Experimental section

Preparation of Mn 3 O 4 @p-rGO nanocomposites

Aqueous graphite oxide dispersion was prepared via the modified Hummers method [40]. Specifically, with stirring in an ice bath, graphite flakes (0.5 g), and NaNO3 (0.5 g) was added to 25 mL concentrated H2SO4. KMnO4 (3 g) was then slowly added with the temperature of the suspension lower than 20 °C throughout the process. After 2 h, the mixture was transferred to a 40 °C water bath for an additional hour. Subsequently, 40 mL of H2O was added dropwise to increase the temperature. The reaction was performed at 90 °C for 0.5 h and followed by further addition of 5 mL H2O2. Then, the mixture was centrifuged to remove the soluble ions, and 5 M KOH solution was added to neutralize the products. Let stand for 1 h, the black precipitates were washed with water for three times and dried in vacuum at 60 °C. At last, the samples were rapidly transferred to a muffle furnace at 500 °C for 10 min, and the final product was collected and denoted as Mn3O4@p-rGO nanocomposite.

Oxidase-mimic activity assay

The oxidase-mimic activity was evaluated by using TMB as a chromogenic substrate. Typically, 20 μL of TMB (10 mM), 30 μL of Mn3O4@p-rGO nanocomposites (100 mg/L), and 40 μL of acetate buffer (100 mM, pH 4.0) were mixed with 110 mL DI water, and the absorbance spectrum was recorded after incubated for 10 min. The kinetic assays were performed by changing the TMB concentrations (0.025, 0.05, 0.075, 0.1, 0.15, 0.2, 0.3, 0.4, 0.5, 0.6, and 0.7 mM) in the reaction mixtures. All reaction solutions were incubated at room temperature (25 ℃) unless otherwise stated, and the absorbance at 652 nm was recorded at 1-s interval over 5 min. All data were collected with three independent measurements.

Detection of AA

For colorimetric detection of AA, 30 μL of Mn3O4@p-rGO nanocomposites solution (100 mg/L), 40 μL of acetate buffer (100 mM, pH 4.0), and various concentrations of AA (final concentrations: 0–100 μM) were mixed in a 0.5-mL tube. Then, 20 μL of TMB (10 mM) solution was added, and further diluted with ultrapure water to a final volume of 200 μL. These mixtures were incubated for 10 min, and then the UV–vis absorbance spectrum was recorded. Similarly, the interference experiment was performed by replacing AA with some other agents. All data were collected with three independent measurements.

Detection AA in fruit juice

Fresh fruits were purchased from market, and the edible part of these fruits were isolated. Then, the fruit juices were obtained with an electric juicer, the juice were centrifuged at 3000 rpm for 10 min to remove the fruit residues. The supernatants were collected and diluted tenfold with 100 mM acetate buffer (pH 4.0) and stored at 4 ℃. Under the detection procedures, 20 μL of the diluted juices were added into the reaction systems, and further detected as mentioned above. Totally, all fruit juices were diluted 100-fold to fit the sensing response range. The AA concentrations in these samples were also detected by HPLC method and 2, 4-dinitrophenylhydrazine colorimetry method according to the Chinese standard GB 5009.86–2016. All data were collected with three independent measurements.

Results and discussion

Characterizations of Mn 3 O 4 @p-rGO nanocomposites

As illustrated in Fig. 1A, a facile in-situ fabrication strategy was applied for the synthesis of the Mn3O4@p-rGO nanocomposites. Firstly, Mn2+-GO suspension was obtained by the modified Hummers method. Under this process, KMnO4 was both used as a strong oxidant for graphite oxidation and as the manganese source. Mn2+ ions were absorbed on the surface of GO nanosheets, on account of the strong coordination interaction of the oxygen containing groups on the GO nanosheets. After being neutralized with KOH and exposed in air, the absorbed Mn2+ was in situ converted into Mn3O4 nanoparticles [41]. After calcination under high temperature in open air for a short time, the GO nanosheets were oxidized to form porous structure as confirmed by the characterizations of the morphology and structure of the nanocomposites as follows.

Fig. 1
figure 1

A Schematic illustration of fabrication process of Mn3O4@p-rGO. B TEM image of Mn3O4@p-rGO nanocomposites. C XRD pattern and D XPS spectrum of Mn3O4@p-rGO nanocomposites

Full size image

As displayed in Fig. 1B, the porous structure of p-rGO nanosheets can be easily identified from the TEM image, and the Mn3O4 nanoparticles uniformly distributed on the surface of p-rGO nanosheets. The p-rGO provided a great support for the formation of the small Mn3O4 nanoparticles and prevent them from aggregation. On the other hand, Mn3O4 nanoparticles also protect the p-rGO nanosheets from burning up at high temperature in open air. The typical (101) crystal plane of Mn3O4 nanoparticles can be observed from the crystal plane spacing of 0.49 nm [26]. As shown in Fig. 1C, the typical diffraction peaks in the XRD pattern of the nanocomposites match well with the standard hausmannite structure of Mn3O4 (JCPDS no. 24–0734), suggesting the high purity of the Mn3O4 nanoparticles [26, 42]. Besides, the broad peak at around 24° implied the existence of the (002) plane of the porous graphene nanosheets. In the Raman spectra of the nanocomposites (Fig. S1), the intensity ratio between the D and G bonds (ID/IG) was 1.05, implying the enhanced electron conductivity [43]. Moreover, the prominent peak at 647 cm−1 associated with the phonon lines of Mn3O4 [42, 43]. Thus, the successful composition of p-rGO nanosheets and Mn3O4 nanoparticles is confirmed.

Furthermore, XPS was performed to study the electronic structure and surface composition of the Mn3O4@p-rGO nanocomposites. The XPS survey spectrum (Fig. 1D) reveals the presence of Mn, C, and O elements in the nanocomposites. As shown in the high-resolution Mn 2p spectrum (Fig. S2A), there are two distinct peaks at 641.8 (Mn 2p3/2) and 653.6 eV (Mn 2p1/2) with a spin-energy separation of 11.8 eV, which agrees well with the earlier reported Mn3O4 [26, 44]. The C 1 s spectrum (Fig. S2B) can be deconvoluted into four components at 284.8, 286.6, 288.2, and 289.1 eV, which correspond to the C–C/C = C, C-O, C = O, and O-C = O bonds, respectively [42]. Most of the oxygen atoms in the surface of porous graphene may residue from GO. As shown in Fig. S2C, the O 1 s spectrum reveals the presence of Mn–O-Mn (530.2 eV), Mn-OH (531.5 eV), and C-O/C = O (532.8 eV), which are consistent well with the previous results [42]. Overall, the Mn3O4 nanoparticles were successfully deposited on the surface of p-rGO.

Enzyme mimicking activity

The typical chromogenic substrates (TMB, ABTS, and OPD) were selected to test the oxidase-like performance of Mn3O4@p-rGO nanocomposites. As displayed in Fig. 2A and Fig. S3, all of these substrates were oxidized to generate the chromatic products with the characteristic UV–vis absorption spectra under the presence of Mn3O4@p-rGO. Besides, the obvious color change can be observed, suggesting the versatile applicability of Mn3O4@p-rGO as an oxidase mimic. To further evaluate the oxidase-mimic activity, TMB was selected as the model substrate, and the absorbance intensity at 652 nm (A652 nm) was monitored to assess the catalytic performance. As shown in Fig. 2B, the catalytic kinetics of different catalysts were investigated. Notably, the oxidase activity of the catalysts follows the order: Mn3O4@p-rGO > Mn3O4@GO > Mn3O4 > p-rGO. Interestingly, the catalytic activity of Mn3O4@p-rGO nanocomposites is higher than the sum of the individual catalytic activity of Mn3O4 and p-rGO, exhibiting a good synergistic effect. Thus, the Mn3O4 nanoparticles decorated on p-rGO exhibited the highest oxidase activity, which was selected as the best oxidase-mimic excellent candidate. As previously reported, both GO and p-rGO were applied as the supports for boosting the catalytic activities of nanoparticles, owing to the increased surface area and catalytic sites [30, 45, 46]. Herein, the catalytic activity of Mn3O4@p-rGO was superior to that of Mn3O4@GO, which may mainly attribute to the better electron transfer ability of p-rGO.

Fig. 2
figure 2

A UV−vis absorption spectra of different reaction systems (inset: photographs of different reaction systems); B time-dependent absorbance changes under the presence of different catalysts. All data were represented as means ± standard deviation of three independent measurements

Full size image

Furthermore, the effects of catalyst concentration, reaction pH, temperature, and storage time on the catalytic activity of the nanozyme were evaluated. As shown in Fig. S4A and S4B, 15 mg/L of Mn3O4@p-rGO and pH = 4.0 were optimized as the best reaction conditions. The catalytic activity of Mn3O4@p-rGO peaked at 45 ℃ over the range of 15–65 ℃ (Fig. S4C). To facilitate the practical application, the following experiments were carried out at 25 ℃. Besides, the catalytic activity of the nanozyme exhibited no apparent change even after store for four months (Fig. S4D), suggesting its outstanding stability.

To evaluate the oxidase-mimic activity of Mn3O4@p-rGO, the steady-state kinetic parameters were further analyzed. By monitoring the catalytic reaction velocity (V) with the addition of different concentrations of TMB, the obtained plot follows the typical Michaelis–Menten curve (Fig. 3A). The Michaelis constant (Km) and maximum initial velocity (Vmax) were calculated to be 0.1135 mM and 164.5 × 10−8 M s−1, respectively. Compared with most reported oxidase-mimics (including Mn3O4 nanomaterials with different morphologies and some other nanoparticle-decorated graphene nanocomposites), Mn3O4@p-rGO had a comparable Km value and higher Vmax (Table S1), suggesting that Mn3O4@p-rGO has a high affinity toward TMB and excellent oxidase-mimic activity.

Fig. 3
figure 3

A Steady-state kinetic assay of Mn3O4@p-rGO on TMB oxidation. B Catalytic performance of Mn3O4@p-rGO in N2/O2-saturated conditions. C The effects of different radical scavengers on the catalytic activity of Mn3O4@p-rGO. D EPR spectra for detection of 1O2 before and after reaction for 5 min. All data were represented as means ± standard deviation of three independent measurements

Full size image

Oxidase-mimic nanozymes can catalyze the substrates oxidation in the presence of dissolved O2 as the electron acceptor. Hence, the catalytic activity of Mn3O4@p-rGO in N2/O2-saturated solution was studied. As shown in Fig. 3B, the catalytic activity increased obviously after the solution was purged with O2, while it was inhibited under a N2-saturated solution. Thus, the dissolved O2 participated in the oxidation of TMB.

To further figure out the possible mechanism, different kinds of radical scavengers were used to verify the possible reactive oxygen species (ROS, including hydroxyl radical (·OH), singlet oxygen atom (1O2), and superoxide radical (O2−)) involving the catalytic process. As shown in Fig. 3C, TBA (·OH scavenger) and p-benzoquinone (O2− scavenger) exhibited weak effects on the catalytic activity even at high concentrations, suggesting that there was almost no ·OH and O2− generated in the reaction system. Obviously, the catalytic activity was sharply reduced with the addition of NaN3 (1O2 scavenger), indicating the generation of plenty of 1O2 in the solution. Furthermore, the existence of these reactive oxygen species was investigated by ESR technology, DMPO and TEMP were used as the spin-trapping reagents for the capture of O2− or ·OH, and 1O2, respectively. As shown in Fig. S5, the weak EPR signals attributed to O2− and ·OH confirmed the absence of O2·− or ·OH in the reaction solution. In contrast, the strong triplet EPR peaks of TEMPOL verified the existence abundant of 1O2 free radicals (Fig. 3D).

Based on these results, the possible catalytic mechanism for TMB oxidation can be illustrated as follows: the dissolved O2 efficiently adsorbed on the mesoporous structure of p-rGO, and then it was immediately activated to generate 1O2 by the increasing active sites of Mn3O4 nanoparticles. Subsequently, the generated 1O2 with strong oxidizing ability immediately oxidized TMB to the blue ox-TMB.

Colorimetric detection of AA

AA is a kind of typical antioxidant, which widely exist in fruit. With the addition of AA in the Mn3O4@p-rGO-TMB reaction system, the generated 1O2 can be directly eliminated by AA, which greatly inhibits the chromogenic reaction of TMB. Meanwhile, the blue ox-TMB can also be transformed into colorless TMB in the existence of AA. Based on this principle and the excellent oxidase enzyme mimic activity of Mn3O4@p-rGO nanocomposites, a fast, sensitive assay could be established. As displayed in Fig. 4 A, with the increase of AA concentration (0–100 μM), the absorbance of ox-TMB gradually decreased, and the color of the solution fading dark blue to colorless. The absorption differences at 652 nm (ΔA652) were further calculated, which exhibited good linearity to the AA concentrations in the range of 0.5 μM to 80 μM (inset of Fig. 4B). The limit of detection (LOD) for AA was calculated to be 0.278 μM according to the 3σ/m criterion (where σ is the standard deviation of the blank and m is the slope of the calibration plot). The proposed Mn3O4@p-rGO-TMB sensing platform exhibited higher sensitivity and wider sensing range than most of the reported probes (Table S2). Additionally, compared with the peroxidase-mimic probes, the proposed sensing platform is much easier to operate and facilitates the application in complex conditions.

Fig. 4
figure 4

A Adsorption spectra of the Mn3O4@p-rGO-TMB reaction system with different concentrations of AA (0, 0.5, 1, 2, 3, 5, 7.5, 10, 15, 20, 30,40, 50, 60, 70, 80, 90, and 100 μM). B Absorbance difference (ΔA652 = A − A0,) under the presence of different amount of AA; inset: the linear calibration plot of ΔA652 against the AA concentration (0.5–80 μM), and the visual picture corresponding to the AA concentrations. C Interference of different agents for detection of AA, the concentration of these interfere agents were 250 μM, and AA was 100 μM; inset: corresponding photographs of the reaction solutions. All data were represented as means ± standard deviation of three independent measurements

Full size image

To further assess the selectivity of the Mn3O4@p-rGO-TMB detection system toward AA, the potential effects of different interferential substances (such as citric acid, benzoic acid, adipic acid, uric acid, and some amino acids) were measured. As shown in Fig. 4C, the presence of these interferential substances has no influence on the detection of AA, and no obvious color change can be observed even in a much higher concentration.

The practical usability of Mn3O4@p-rGO for AA detection was carried out in some different fresh fruits, such as pitaya, grape, and navel orange. As displayed in Table S3, the contents of AA in these fruits juice were determined, and the standard addition method was applied by adding different amounts of AA in the solution. The recoveries of the samples range from 97.8 to 104.2%, indicating the feasibility and reliability of the proposed method for AA determination in practical samples. Furthermore, HPLC method and 2,4-dinitrophenylhydrazine colorimetry method were also applied to detect the concentrations of the fruit juices. As listed in Table S4, the detected concentrations by our method were consistent with these reference methods, indicating the as-prepared method presented acceptable precision and accuracy.

Conclusions

The Mn3O4@p-rGO nanocomposite was successfully prepared through an in-situ fabrication strategy. Owning to the synergistic effect between Mn3O4 nanoparticles and p-rGO nanosheets, the hybrid exhibited enhanced oxidase-like activity compared with Mn3O4 nanoparticles or p-rGO nanosheets alone. A simple and fast colorimetric sensor of Mn3O4@p-rGO-TMB detection system was constructed for AA detection with excellent sensitivity and selectivity. Furthermore, practical application for AA detection in real samples also demonstrated its good feasibility and reliability. This work indicates the decoration of nanoparticles on specific support is an efficient strategy to enhance their enzyme-like activities, which exhibit promising potential in food safety, pharmacy, biosensing, and disease diagnosis.