Fabrication of porous NiMn2O4 nanosheet arrays on nickel foam as an advanced sensor material for non-enzymatic glucose detection

In this work, porous NiMn2O4 nanosheet arrays on nickel foam (NiMn2O4 NSs@NF) was successfully fabricated by a simple hydrothermal step followed by a heat treatment. Porous NiMn2O4 NSs@NF is directly used as a sensor electrode for electrochemical detecting glucose. The NiMn2O4 nanosheet arrays are uniformly grown and packed on nickel foam to forming sensor electrode. The porous NiMn2O4 NSs@NF electrode not only provides the abundant accessible active sites and the effective ion-transport pathways, but also offers the efficient electron transport pathways for the electrochemical catalytic reaction by the high conductive nickel foam. This synergy effect endows porous NiMn2O4 NSs@NF with excellent electrochemical behaviors for glucose detection. The electrochemical measurements are used to investigate the performances of glucose detection. Porous NiMn2O4 NSs@NF for detecting glucose exhibits the high sensitivity of 12.2 mA mM−1 cm−2 at the window concentrations of 0.99–67.30 μM (correlation coefficient = 0.9982) and 12.3 mA mM−1 cm−2 at the window concentrations of 0.115–0.661 mM (correlation coefficient = 0.9908). In addition, porous NiMn2O4 NSs@NF also exhibits a fast response of 2 s and a low LOD of 0.24 µM. The combination of porous NiMn2O4 nanosheet arrays and nickel foam is a meaningful strategy to fabricate high performance non-enzymatic glucose sensor. These excellent properties reveal its potential application in the clinical detection of glucose.

. Figure 2b shows the crystal structure of the spinel NiMn 2 O 4 sample. Nickel ions and manganese are adopted the cubic structure with mixed valence states for the spinel NiMn 2 O 4 structure; nickel and manganese occupy randomly but totally proportionally (Ni:Mn = 1:2) in the interstices of oxygen stacking tetrahedron and octahedron 43,44 . These cations occupy the cubic lattice composing by close-packed oxygen anions (O 2− ) 45 . For the spinel NiMn 2 O 4 materials, the low crystalline can improve their electrochemical performances due to numerous of loosely packed atoms being available for redox reaction 46 .
The morphology and structure of porous NiMn 2 O 4 NSs@NF sensor electrode are characterized by SEM. Different magnification SEM images of porous NiMn 2 O 4 NSs@NF are shown in Fig. 3. Low magnification SEM image shows a panoramic view (Fig. 3a) nanosheets is about 20 nm on average. More detailed structures are further shown in Figs. S1-S4. There is space between the smaller NiMn 2 O 4 nanosheets forming 3D spatial structure, which can facilitate the electrolyte ions diffusion and provide larger surface area for electrocatalytic reactions. The unique 3D core-shell structure greatly reduces distance for the diffusion/transport of electrolyte ions, which can be attributed to the opening structure and excellent performances of NiMn 2 O 4 nanosheets. Figure 4 shows TEM and HRTEM images of porous NiMn 2 O 4 nanosheet arrays (porous NiMn 2 O 4 nanosheets are scratched from nickel foam consisted of stacking nanosheets). Figure 4a displays a low magnification TEM image. An overall contour of NiMn 2 O 4 nanosheet is observed in the low magnification TEM image. From this low magnification TEM image, the length and width of this porous NiMn 2 O 4 nanosheet can be clearly seen about 600 nm and 400 nm, respectively. Figure 4b shows medium magnification TEM image. Medium magnification TEM image further shows the unique core-shell structure of the NiMn 2 O 4 nanosheet arrays. Figure 4c shows high magnification TEM image. This unique core-shell structure can be clearly observed in high magnification TEM image. These smaller NiMn 2 O 4 nanosheets, called as "shell" nanosheets, are distributed uniformly on the surface of the "core" NiMn 2 O 4 nanosheet. Some "shell" NiMn 2 O 4 nanosheets are marked by the yellow dash line in Fig. 4c. The thickness of the "shell" NiMn 2 O 4 nanosheets is about 10 nm. This core-shell structure can effectively provide lager surface area and markedly shorten the ion diffusion distance 47 . Figures Fig. 4f. The SAED patterns are composed of several light diffraction circles. Four major diffraction circles can be observed on the SAED patterns. These major diffraction circles clearly match with the (444), (622), (400) and (311) planes of porous NiMn 2 O 4 nanosheet, respectively, representing the existence of porous NiMn 2 O 4 nanosheet and its polycrystalline structure 48 .
To further analyze the elemental composition and oxidation state, porous NiMn 2 O 4 NSs@NF sensor electrode is characterized by XPS and the results are analyzed with based on Gaussian-Lorentzian fitting method. Figure 5a displays the full XPS survey spectra of porous NiMn 2 O 4 NSs@NF electrode, which mainly contains the elements of Ni, Mn and O. Figure 5b shows the Ni spectrum of porous NiMn 2 O 4 NSs@NF electrode. Two peaks with the binding energies at 853.9 and 855.4 eV correspond to the Ni 2p 3/2 49,50 . The peak at 872.6 eV corresponds to the Ni 2p 1/2 51 . Two peaks with the binding energies located at 860.9 and 879.3 eV as shown in Fig. 5b are the satellite (Sat.) peaks of the Ni 2p 3/2 and Ni 2p 1/2, respectively 52 . Figure 5c shows Mn spectrum of porous NiMn 2 O 4 NSs@NF electrode. Two spin-orbit peaks in Mn spectrum are deconvolved into four peaks. Two deconvolved peaks are observed at 641.0 and 642.5 eV, which correspond to Mn 2p 3/2. Two deconvolved peaks are observed at 654.0 and 652.6 eV, corresponding to Mn 2p 1/2, which is consistent with the previous reported literature 53 . Two deconvolved peaks are observed at 641.0 and 652.6 eV, which correspond to the correlative peaks of Mn 2+ ; two deconvolved peaks are observed at 642.5 and 654.0 eV, which correspond to the correlative peaks of Mn 3+ binding energy 54 . Figure 5d shows the O spectrum of porous NiMn 2 O 4 NSs@NF electrode. The resolved peak at binding energy of 529.4 eV is indexed to typical metal oxygen bonds (M-O-M) or the lattice oxygen [55][56][57] . The peak for O 1 s at 530.6 eV is attributed to metal-O-H from metal surface hydroxyl groups 33,58 . The peak at 531.6 eV is attributed to a larger number of defect sites with a low oxygen coordination normally observed in materials with small particles 55 . The energy dispersive spectroscopy (EDS) mappings of porous NiMn 2 O 4 NSs@NF sensor electrode are shown in Fig. S8. These EDS mappings indicate that Ni, Mn and O elements are uniformly distributed on porous NiMn 2 O 4 nanosheet arrays, which in agreement with XRD and XPS characterizations.
The surface area and porosity are two important factors, which can critically influence the sensing performances for detection of glucose. The surface area and porosity of the porous NiMn 2 O 4 NSs@NF sensor electrode is further analyzed by Brunauer-Emment-Teller (BET) nitrogen isothermal adsorption and desorption test. Figure S9 shows a typical BET nitrogen adsorption and desorption isotherms of NiMn 2 O 4 NSs@NF electrode. Nitrogen adsorption and desorption isotherms are plotted as quantity volume (V m ) on the y-axis and relative pressure (P/P 0 ) on the x-axis based to BET experimental data. According to the BET test, the BET specific surface area of porous NiMn 2 O 4 NSs@NF electrode is calculated to be 77.5 m 2 g −1 . This large specific area can effectively increase the utilization of NiMn 2 O 4 as an electrochemically active material in the process of glucose electrochemical detection. The adsorption/desorption isotherms also show a hysteresis, which can be classified as a type IV isotherm according to the profile of the hysteresis loop in a range of 0.5 < P/P 0 < 1.0 59 . Inset in Fig. S9 shows the corresponding pore-size distribution with calculated by the Barrette Joynere Halenda (BJH) method based on BET experimental data. The pore size distribution image shows a wide pore-size distribution characteristic, which can be attributed to the porous nano-structure of NiMn 2 O 4 NSs@NF electrode. The pore size distribution image presents that the average pore size of NiMn 2 O 4 NSs@NF electrode is about 9.6 nm. Porous structure of NiMn 2 O 4 (2019) 9:18121 | https://doi.org/10.1038/s41598-019-54746-2 www.nature.com/scientificreports www.nature.com/scientificreports/ NSs@NF electrode with large surface area provides the highway for transportation of electrons and ions between the electrolytes and electrode material, which is in favour of non-enzymatic glucose detection.
Porous NiMn 2 O 4 NSs@NF electrode is used directly as a sensor electrode to test its electrocatalytic activity toward glucose detection in 0.5 M NaOH electrolyte. Figure 6a presents CV curves of porous NiMn 2 O 4 NSs@ NF electrode at sweep rate ranging from 10 to 100 mV s −1 . Each CV curve displays a pair of redox peak. These peaks in CV curves can be attributed to the electrochemical redox reactions/electrocatalytic oxidation reactions of porous NiMn 2 O 4 NSs@NF electrode. The electrocatalytic oxidation reaction for glucose detection are shown as follows formulae (1-4) 60-62 :   As shown in CV curve at sweep rate of 10 mV s −1 , a pairs of redox peaks at +0.23/+0.43 V can be observed. With the increase of scan rates, the positive potential shift of the anodic peaks and negative potential shift of cathodic peaks are also observed. The separation of peak to peak (ΔEp) increases linearly with the increasing scan rates. These phenomena may be attributed to the increase of overpotential 63 . Figure 6b shows the corresponding fitting curves of the response currents vs. the square root of sweep rates. The corresponding fitting curves show the linear dependencies on the sweep rates, indicating that the electron transfer process of electrode is the reversible and diffusion-controlled electrochemical redox process 64    Binding energy (eV) Intensity (a.u.) EIS is employed to test the electrochemical impedance property of porous NiMn 2 O 4 NSs@NF sensor electrode. The electrochemical impedance test is employed in the frequency range from 10 5 Hz to 10 −2 Hz at 5 mV in a three-electrode cell with 0.5 M NaOH electrolyte solution. Figure S10 shows Nyquist plot of porous NiMn 2 O 4 NSs@NF electrode. Nyquist plot shows a semicircle at higher frequencies and a long positive-slope line at the lower frequencies. This diameter of semicircle corresponds to the charge transfer resistance, indicating the electron transfer kinetics of the charge transfer process at the working electrode/electrolyte interface 65 . From the inset in Fig. S10, the small diameter of the semicircle reveals the good electric conductivity of porous NiMn 2 O 4 NSs@NF electrode. At the lower frequencies, this positive-slope line corresponds to the Warburg impedance (Z w ). From Nyquist plot, the slope line with an inclination angle approaching 60° reveals the good diffusion kinetics between the electrode surface and electrolyte 64,66 . The low electrochemical impedance indicates that porous NiMn 2 O 4 NSs@NF electrode can provide an efficient electron transfer pathway and fast current response for glucose detection.
The amperometric tests are performed to test the electrochemical response property of porous NiMn 2 O 4 NSs@NF sensor electrode. Under optimal conditions, the amperometric responses are tested at a potential of +0.45 V in 0.5 M NaOH electrolyte solution. Figure 7 shows typical amperometric response curves for various concentrations of glucose. Figure 7a shows the amperometric responses of porous NiMn 2 O 4 NSs@NF electrode with consecutive step changes of the glucose concentration at a potential of 0.45 V (the concentration range of glucose in electrolyte bath is 0.99-67.30 μM). As can be seen from Fig. 7a, porous NiMn 2 O 4 NSs@NF electrode shows the good amperometric response at the glucose concentration range from 0.99 μM to 67.30 μM. Figure 7b shows the corresponding fitting curves of the amperometric responses vs. glucose concentrations (the concentration range is 0.99-67.30 μM). The amperometric response increases linearly with the increase of glucose concentration. The linear fitting regression equation is expressed as y (mA cm −2 ) = 0.01224x + 0.3228 (R 2 = 0.9982). Porous NiMn 2 O 4 NSs@NF electrode delivers a sensitivity of 12.2 mA mM −1 cm −2 at the window concentrations of 0.99-67.30 μM. In addition, the limit of detection (LOD) of glucose detection is calculated to be using the following equation 67 . LOD = 3 SD/S, where, S is the slope of the calibration curve (0.01224 mA μM −1 cm −2 ) and SD is the standard deviation of blank (9.9 × 10 −4 mA cm −2 ). The detection limit is calculated to be 0.24 µM.  Figure 7f shows the corresponding fitting curves (the linear concentration range is 0.925-5.310 mM).
The amperometric response time is crucial parameter for the electrochemical sensor in non-enzymatic glucose detection. The response time of porous NiMn 2 O 4 NSs@NF electrode is obtained by amperometric measurements in the different glucose concentration in 0.5 M NaOH electrolyte at 0.45 V. Figure 8 presents the response time of porous NiMn 2 O 4 NSs@NF electrode. With the addition of glucose to electrolyte solution, the glucose oxidation current increases rapidly and then reaches to the steady state. The time begin from the current increase until the current signal to the stead state value is defined as the response time of the sensor. Figure 8a shows an observed response of the sensor is 3 s at a gluconic concentration of 5.964 μM. Figure 8b shows an observed response of the sensor is 2 s at a gluconic concentration of 0.115 mM, which is considered a quick response time. The quick amperometric response time is attributed to good sensibility, excellent electronic conductivity and efficient catalytic ability selectivity of porous NiMn 2 O 4 NSs@NF electrode. The comparison for the sensing performances of porous NiMn 2 O 4 NSs@NF electrode and other transition metal oxide materials is listed as shown in Table S1. This comparison table shows the excellent sensing performances of porous NiMn 2 O 4 NSs@NF electrode compared to the reported sensor.
Anti-interference property of porous NiMn 2 O 4 NSs@NF sensor electrode is crucial for non-enzymatic electrochemical detection of glucose. It is well-known that saccharides have similar electrochemical reaction behaviors or the interferential behaviors and chloride can lead to the catalyst poisoning to glucose detection. Thus, these compounds cannot be ignored. We investigate the amperometric responses from the other saccharides or chloride such as CA, urea, AA, NaCl in a 0.5 M NaOH electrolyte solution. With the addition of 1.0 mM glucose to the 0.5 M NaOH electrolyte, a distinct response current at 400 s can be observed from Fig. 8c. When upon addition www.nature.com/scientificreports www.nature.com/scientificreports/ of other interference compounds such as CA (0.1 mM), urea (0.1 mM), AA (0.1 mM), NaCl (0.1 mM), the current responses cannot be observed or the response current are acceptable and negligible compared to the response current of glucose molecules. With the additions of 1.0 mM glucose, two distinct amperometric responses at 200 s to 900 s can be observed toward glucose detection. The low current responses for other saccharides or chloride indicate that porous NiMn 2 O 4 NSs@NF have the good selectivity for the electrochemical determination of glucose. Considering that the glucose level is at least 30~50 times higher than those of interfering species in human serum, these interference species produce negligible current responses compared with glucose molecules in a 0.5 M NaOH electrolyte solution 63 . Therefore, these result reveals that porous NiMn 2 O 4 NSs@NF electrode will be well used toward the detection of glucose in practice.
The long-term stability of porous NiMn 2 O 4 NSs@NF sensor electrode is examined after 30 days. The stability of porous NiMn 2 O 4 NSs@NF electrode is measured by CV sweep at 20 mV s −1 in 0.5 M NaOH with 1 mM glucose. Figure 8d shows CV curves of porous NiMn 2 O 4 NSs@NF electrode after 1 day and 30 days. From CV curves, we can observe that no distinct decrease for the peak current after 30-days storage. CV curves almost remains the same shape. In addition, the current response (64.5 mA cm −2 ) maintains 95.1% of the primitive response (67.8 mA cm −2 ) after one month storage. These results indicate the excellent electrochemical stability of porous NiMn 2 O 4 NSs@NF sensor electrode. ≥99.0%), urea (CO(NH 2 ) 2 ; ≥99.0%), glucose (C 6 H 12 O 6 ·H 2 O; α D : +52.5~+53.0°), ascorbic acid (C 6 H 8 O 6 , AA; ≥99.7%) and citric acid (C 6 H 8 O 7 , CA; ≥99.5%) were purchased and obtained from Tianjin Guangfu Technology Development Co. Ltd.. Hydrochloric acid (HCl, 36.0~38.0%) was obtained and purchased from Jinzhou Ancient City Chemical Reagents Factory. Sodium chloride (NaCl; ≥99.5%) and sodium hydroxide (NaOH; ≥96.0%) were obtained and purchased from Tianli Chemical Reagent Co. Ltd.. Ammonium fluoride (NH 4 F; ≥98.0%) was obtained and purchased from Tianjin Fuchen Chemical Reagents Factory. Nickel foam was purchased and obtained from Taiyuan Liyuan Lithium Technology Co. Ltd., more detailed technical parameters of nickel foam were shown in Table S2. De-ionized water (18.3 MΩ cm at 25 °C) was purified and obtained by Milli-Q water system to prepare all solutions. In our work, all chemical reagents and materials were also used without further purification unless otherwise described. fabrication of porous niMn 2 o 4 nSs@nf electrode. The typical procedure, porous NiMn 2 O 4 NSs@NF were fabricated via a simple hydrothermal method followed by a heat treatment. Briefly, 2 mmol Ni(NO 3 ) 2 ·6H 2 O (0.582 g), 12 mmol NH 4 F (0.444 g) and 30 mmol CO(NH 2 ) 2 (1.800 g) were dissolved into 40 mL deionized water under continuous electromagnetic stirring. A nickel foam (length × width × thickness = 20 mm × 10 mm × 1 mm) was treated and purified with 3 M HCl for 15 min for the remove of the oxide layer on nickel foam. Then, the acid-treated nickel foam was cleaned sequentially by copious amounts of de-ionized water. Then, this pre-treated nickel foam was immersed into above mixed solution. 4 mmol KMnO 4 (0.632 g) was dissolved into 40 mL de-ionized water and then was added into the previous solution. After stirring for 30 min, the mixed solution was transferred into a 100 ml-Teflon-lined stainless steel hydrothermal reactor and followed by heating at 110 °C for 8 h. After cooling to the ambient temperature, the precursor was collected and washed with de-ionized water and ethanol thoroughly to remove residual ions. The precursor was dried at 60 °C for 2 h. Finally, the precursor was converted to porous NiMn 2 O 4 NSs@NF under 350 °C for 2 h and then naturally cooled to the ambient temperature.
instruments and characterizations. X-ray diffraction (XRD) measurement was carried out using the Rigaku RAD-3C diffractometer instrument (Cu Kα, λ = 1.5405 Å, 35 kV, 20 mA, 2-Theta angles: 10°-70°). The morphology and structure were investigated by scanning electron microscopy (SEM, JEOL S-4800) under the condition of 3.0 kV operating voltage. Transmission electron microscopy (TEM, JEOL JEM-2100F microscopy) with an energy dispersive X-ray spectroscope (EDS) was also carried out to investigate the element distributions of porous NiMn 2 O 4 NSs@NF under the condition of 200 kV accelerating voltage. X-ray photoelectron spectroscopy (XPS, ESCALB-MKII250) was performed to analyze the elemental compositions and its valence of porous NiMn 2 O 4 NSs@NF under a monochromatic 150 W Al Kα source radiation. Nitrogen adsorption/desorption