Tunable Non-Enzymatic Glucose Electrochemical Sensing Based on the Ni/Co Bimetallic MOFs

Constructing high-performance glucose sensors is of great significance for the prevention and diagnosis of diabetes, and the key is to develop new sensitive materials. In this paper, a series of Ni2Co1-L MOFs (L = H2BPDC: 4,4′-biphenyldicarboxylic acid; H2NDC: 2,6-naphthalenedicarboxylic acid; H2BDC: 1,4-benzenedicarboxylic acid) were synthesized by a room temperature stirring method. The effects of metal centers and ligands on the structure, compositions, electrochemical properties of the obtained Ni2Co1-L MOFs were characterized, indicating the successful preparation of layered MOFs with different sizes, stacking degrees, electrochemical active areas, numbers of exposed active sites, and glucose catalytic activity. Among them, Ni2Co1-BDC exhibits a relatively thin and homogeneous plate-like morphology, and the Ni2Co1-BDC modified glassy carbon electrode (Ni2Co1-BDC/GCE) has the highest electrochemical performance. Furthermore, the mechanism of the enhanced glucose oxidation signal was investigated. It was shown that glucose has a higher electron transfer capacity and a larger apparent catalytic rate constant on the Ni2Co1-BDC/GCE surface. Therefore, tunable non-enzymatic glucose electrochemical sensing was carried out by regulating the metal centers and ligands. As a result, a high-sensitivity enzyme-free glucose sensing platform was successfully constructed based on the Ni2Co1-BDC/GCE, which has a wide linear range of 0.5–2899.5 μM, a low detection limit of 0.29 μM (S/N = 3), and a high sensitivity of 3925.3 μA mM−1 cm−2. Much more importantly, it was also successfully applied to the determination of glucose in human serum with satisfactory results, demonstrating its potential for glucose detection in real samples.


Introduction
Glucose is one of the essential nutrients for human metabolism, but diabetes can be triggered when blood sugar levels are higher than normal. As we all know, diabetes is one of the most common chronic diseases in our daily life has and is a serious threat to people's health [1,2]. Therefore, there is a significant medical value and commercial prospect to developing a stable and accurate glucose sensor for rapid real-time monitoring of blood glucose levels. To develop cheap and practical sensors, researchers have tried many new materials, according to the different catalytic mechanisms of these materials. The detection methods can be divided into colorimetry, fluorescence spectrometry, electrochemical method, etc. [3][4][5]. In contrast, the electrochemical method has the advantages of being convenient, rapid, sensitive, and economical and is the most widely used method for the determination of glucose content [6,7]. At present, most of the small blood glucose monitors on the market are enzyme-based blood glucose sensors, which have the advantages of high sensitivity and selectivity. However, there are also some insurmountable defects, such as the high cost and harsh survival conditions of biomolecular enzymes, and they are also susceptible to the influence of the external environment, such as temperature and pH. Therefore, it is essential to prepare enzyme-free glucose sensors that have a simple in

Characterization of Ni 2 Co 1 -L MOFs
Additionally, the crystallinity and phase purity of the synthesized Ni 2 Co 1 -L MOFs were examined using XRD, and the results are shown in Figure 1A-C. With the exception of some deviations in the relative intensities, their XRD patterns were in good agreement with the simulated patterns, which indicated the formation of high purity [34][35][36]. By comparing the FTIR spectra in Figure 1D, it can be seen that some absorption bands are the same in Ni 2 Co 1 -BPDC, Ni 2 Co 1 -NDC, and Ni 2 Co 1 -BDC MOFs. These indicated that they share common organic functional groups. The bending vibrations of the aromatic ring are about 678, 742, and 803 cm −1 [37]. Furthermore, they all have two characteristic peaks around 1375 and 1583 cm −1 , which correspond to the asymmetric stretching vibration peak (as) and the symmetric (ss) stretching vibration peak of the carboxyl group. The presence of a broad absorption band at 3394 cm −1 and the characteristic peak at 3598 cm −1 are due to the -OH vibration of water molecules present in the material structures [38]. All of these confirmed the successful introduction of ligands into the synthesized MOFs. In particular, the XRD diffraction peaks and FTIR patterns of Ni 2 Co 1 -BDC were basically in agreement with those Molecules 2023, 28, 5649 3 of 13 of the monometallic Ni-BDC and Co-BDC ( Figure S1A,B in Supporting Information), which indicated that the introduction of Ni, Co bimetals did not affect the crystalline structure of MOFs. TGA analysis was applied to investigate the thermal stability of Ni 2 Co 1 -L MOFs in the range of 25-700 • C. The results are shown in Figure S1C. The main weight loss of Ni 2 Co 1 -L MOFs in the range of 25-170 • C was attributed to water and residual solvents adsorbed within the pores [39]. The second weight loss around 200-280 • C was attributed to coordinated water molecules, and the weight loss around 400 • C was assignable to decomposition of the carboxylate ligand [40]. As a comparison, Ni 2 Co 1 -BDC had less percent weight loss, which resulted in better stability.
presence of a broad absorption band at 3394 cm −1 and the characteristic peak at 3598 cm −1 are due to the -OH vibration of water molecules present in the material structures [38]. All of these confirmed the successful introduction of ligands into the synthesized MOFs. In particular, the XRD diffraction peaks and FTIR patterns of Ni2Co1-BDC were basically in agreement with those of the monometallic Ni-BDC and Co-BDC ( Figure S1A,B in Supporting Information), which indicated that the introduction of Ni, Co bimetals did not affect the crystalline structure of MOFs. TGA analysis was applied to investigate the thermal stability of Ni2Co1-L MOFs in the range of 25-700 °C. The results are shown in Figure S1C. The main weight loss of Ni2Co1-L MOFs in the range of 25-170 °C was attributed to water and residual solvents adsorbed within the pores [39]. The second weight loss around 200-280 °C was attributed to coordinated water molecules, and the weight loss around 400 °C was assignable to decomposition of the carboxylate ligand [40]. As a comparison, Ni2Co1-BDC had less percent weight loss, which resulted in better stability. Then, the morphological characteristics of the synthetic materials were investigated using the SEM technique, and the corresponding images are shown in Figure 2. It can be seen from Figure 2A that the Ni2Co1-BPDC is a layered stacking structure consisting of nanosheets with smaller sizes. Ni2Co1-NDC ( Figure 2B) is a bulky structure. On the contrary, Ni2Co1-BDC ( Figure 2C) is a relatively thin and homogeneous plate-like morphology, which possesses a 2D structure at the micron size. Then, the elemental distribution and content were characterized by EDS. From Figure 2D and Table S1 (Supporting Information), it can be seen that C, O, Ni, and Co are uniformly distributed on the surface of Ni2Co1-BDC. Then, the morphological characteristics of the synthetic materials were investigated using the SEM technique, and the corresponding images are shown in Figure 2. It can be seen from Figure 2A that the Ni 2 Co 1 -BPDC is a layered stacking structure consisting of nanosheets with smaller sizes. Ni 2 Co 1 -NDC ( Figure 2B) is a bulky structure. On the contrary, Ni 2 Co 1 -BDC ( Figure 2C) is a relatively thin and homogeneous plate-like morphology, which possesses a 2D structure at the micron size. Then, the elemental distribution and content were characterized by EDS. From Figure 2D and Table S1 (Supporting Information), it can be seen that C, O, Ni, and Co are uniformly distributed on the surface of Ni 2 Co 1 -BDC.

Electrochemical Performance of Ni2Co1-L MOFs
First of all, the electrochemical active areas of different Ni2Co1-L MOF nanosheets were compared by calculating the electric double-layer capacitance (Cdl) based on the CV tests in a non-Faradaic region (e.g., from 0.10 to 0.20 V) [41]. The results are shown in Figure 3A-C. According to the slope of the linear relationship between the capacitive currents difference (Δj = janodic − jcathodic) at 0.15 V and the scan rates ( Figure 3D), the Cdl of the Ni2Co1-BPDC/GCE, Ni2Co1-NDC/GCE, and Ni2Co1-BDC/GCE was obtained to be 3.20, 4.65, and 5.35 mF cm −2 , respectively. It is known that the Cdl is proportional to the electrochemically active area. The Ni2Co1-BDC/GCE had the largest active area, which may be derived from its 2D lamellar morphology.

Electrochemical Performance of Ni 2 Co 1 -L MOFs
First of all, the electrochemical active areas of different Ni 2 Co 1 -L MOF nanosheets were compared by calculating the electric double-layer capacitance (C dl ) based on the CV tests in a non-Faradaic region (e.g., from 0.10 to 0.20 V) [41]. The results are shown in Figure 3A-C. According to the slope of the linear relationship between the capacitive currents difference (∆j = j anodic − j cathodic ) at 0.15 V and the scan rates ( Figure 3D), the C dl of the Ni 2 Co 1 -BPDC/GCE, Ni 2 Co 1 -NDC/GCE, and Ni 2 Co 1 -BDC/GCE was obtained to be 3.20, 4.65, and 5.35 mF cm −2 , respectively. It is known that the C dl is proportional to the electrochemically active area. The Ni 2 Co 1 -BDC/GCE had the largest active area, which may be derived from its 2D lamellar morphology.

Electrochemical Performance of Ni2Co1-L MOFs
First of all, the electrochemical active areas of different Ni2Co1-L MOF nanosheets were compared by calculating the electric double-layer capacitance (Cdl) based on the CV tests in a non-Faradaic region (e.g., from 0.10 to 0.20 V) [41]. The results are shown in Figure 3A-C. According to the slope of the linear relationship between the capacitive currents difference (Δj = janodic − jcathodic) at 0.15 V and the scan rates ( Figure 3D), the Cdl of the Ni2Co1-BPDC/GCE, Ni2Co1-NDC/GCE, and Ni2Co1-BDC/GCE was obtained to be 3.20, 4.65, and 5.35 mF cm −2 , respectively. It is known that the Cdl is proportional to the electrochemically active area. The Ni2Co1-BDC/GCE had the largest active area, which may be derived from its 2D lamellar morphology.  It is well known that one of the key factors affecting the catalytic performance of electrode material is the number of active sites, which can be measured in the electrochemically active region. Here, the electrocatalytic activity of Ni 2 Co 1 -L MOFs was compared by recording the CV curves in 0.1 M NaOH at a scan rate of 100 mV/s, and the results are shown in Figure 4A. It is noteworthy that the GCE and Ni 2 Co 1 -BPDC/GCE show no obvious redox peak, suggesting poor electrochemical activity. The Ni 2 Co 1 -NDC/GCE shows a pair of asymmetric redox peaks at around 0.42 and 0.24 V, and in the CV curves of the Ni 2 Co 1 -BDC/GCE more obvious redox peaks appear at around 0.46 and 0.34 V. These redox peaks most likely resulted from the redox of their metal centers Ni 2+ and Co 2+ . Additionally, the Ni 2 Co 1 -BDC/GCE has a higher peak current, suggesting the largest number of exposed active sites. Then, glucose was chosen as the target molecule and the catalytic activity of these materials on glucose was investigated. Figure 4B shows the CV curves of different GCEs after the addition of 0.1 mM glucose. No obvious redox peaks were observed on the surfaces of the GCE and Ni 2 Co 1 -BPDC/GCE. For the Ni 2 Co 1 -NDC/GCE and Ni 2 Co 1 -BDC/GCE, both of them show an oxidation peak with enhanced signal compared with Figure 4A without glucose. It was found that the Ni 2 Co 1 -BDC/GCE had the most obvious phenomenon, which indicated that it had the strongest catalytic activity for the oxidation of glucose. This can be attributed to the unique disk-like structure of Ni 2 Co 1 -BDC, as it exposes more active sites easily, like the results we mentioned above. The electrode reaction mechanism of the redox process can be expressed by the following equation [42]: Molecules 2023, 28, x FOR PEER REVIEW 5 of 13 It is well known that one of the key factors affecting the catalytic performance of electrode material is the number of active sites, which can be measured in the electrochemically active region. Here, the electrocatalytic activity of Ni2Co1-L MOFs was compared by recording the CV curves in 0.1 M NaOH at a scan rate of 100 mV/s, and the results are shown in Figure 4A. It is noteworthy that the GCE and Ni2Co1-BPDC/GCE show no obvious redox peak, suggesting poor electrochemical activity. The Ni2Co1-NDC/GCE shows a pair of asymmetric redox peaks at around 0.42 and 0.24 V, and in the CV curves of the Ni2Co1-BDC/GCE more obvious redox peaks appear at around 0.46 and 0.34 V. These redox peaks most likely resulted from the redox of their metal centers Ni 2+ and Co 2+ . Additionally, the Ni2Co1-BDC/GCE has a higher peak current, suggesting the largest number of exposed active sites. Then, glucose was chosen as the target molecule and the catalytic activity of these materials on glucose was investigated. Figure 4B shows the CV curves of different GCEs after the addition of 0.1 mM glucose. No obvious redox peaks were observed on the surfaces of the GCE and Ni2Co1-BPDC/GCE. For the Ni2Co1-NDC/GCE and Ni2Co1-BDC/GCE, both of them show an oxidation peak with enhanced signal compared with Figure 4A without glucose. It was found that the Ni2Co1-BDC/GCE had the most obvious phenomenon, which indicated that it had the strongest catalytic activity for the oxidation of glucose. This can be attributed to the unique disk-like structure of Ni2Co1-BDC, as it exposes more active sites easily, like the results we mentioned above. The electrode reaction mechanism of the redox process can be expressed by the following equation [42]: Co 3+ -H2BDC + OH − + glucose → Co 2+ -H2BDC glucolactone + H2O + e − (4)

Electrochemical Oxidation of Glucose on Ni2Co1-L/GCEs
To further demonstrate the electrocatalytic activity of the series Ni2Co1-L MOFs for glucose oxidation, the i-t response of glucose on the surface of different GCEs was investigated. As shown in Figure 5A, i-t curves were recorded at 0.5 V for different GCEs after successive addition of 0.1 mM glucose in 0.1 M NaOH solution. No current step signal of glucose was observed on the bare GCE, indicating poor electrochemical activity. The current step signal of the Ni2Co1-BPDC/GCE had a slight increase, but it is difficult to maintain a smooth state. In contrast, a remarkable increase was noticed in both the

Electrochemical Oxidation of Glucose on Ni 2 Co 1 -L/GCEs
To further demonstrate the electrocatalytic activity of the series Ni 2 Co 1 -L MOFs for glucose oxidation, the i-t response of glucose on the surface of different GCEs was investigated. As shown in Figure 5A, i-t curves were recorded at 0.5 V for different GCEs after successive addition of 0.1 mM glucose in 0.1 M NaOH solution. No current step signal of glucose was observed on the bare GCE, indicating poor electrochemical activity. The current step signal of the Ni 2 Co 1 -BPDC/GCE had a slight increase, but it is difficult to maintain a smooth state. In contrast, a remarkable increase was noticed in both the Ni 2 Co 1 -NDC/GCE and Ni 2 Co 1 -BDC/GCE, and the current step signal of the Ni 2 Co 1 -BDC/GCE had the largest value.
Ni2Co1-NDC/GCE and Ni2Co1-BDC/GCE, and the current step signal of the Ni2Co1-BDC/GCE had the largest value.
To explore the reason for the signal enhancement effects for the oxidation of glucose, electrochemical impedance spectroscopy (EIS) measurements were employed to investigate the electron transfer resistance of different Ni2Co1-L/GCEs during glucose oxidation. In the Nyquist plot, the semi-circle diameter represents the charge transfer resistance (Rct) for the electrode surface active species [43]. Herein, based on Figure 5B, the fitted values of Rct for the oxidation of glucose on the GCE, Ni2Co1-BPDC/GCE, Ni2Co1-NDC/GCE, and Ni2Co1-BDC/GCE were 85.4, 43.3, 27.2, and 12.0 KΩ, respectively. The greatly decreased Rct values revealed that the Ni2Co1-BDC/GCE improved the electron transfer ability of glucose, consequently resulting in higher oxidation signals. Subsequently, the chronoamperometry experiment was applied to study the electrochemical kinetics property, which can explain the signal enhancement mechanism To explore the reason for the signal enhancement effects for the oxidation of glucose, electrochemical impedance spectroscopy (EIS) measurements were employed to investigate the electron transfer resistance of different Ni 2 Co 1 -L/GCEs during glucose oxidation. In the Nyquist plot, the semi-circle diameter represents the charge transfer resistance (R ct ) for the electrode surface active species [43]. Herein, based on Figure 5B, the fitted values of R ct for the oxidation of glucose on the GCE, Ni 2 Co 1 -BPDC/GCE, Ni 2 Co 1 -NDC/GCE, and Ni 2 Co 1 -BDC/GCE were 85.4, 43.3, 27.2, and 12.0 KΩ, respectively. The greatly decreased R ct values revealed that the Ni 2 Co 1 -BDC/GCE improved the electron transfer ability of glucose, consequently resulting in higher oxidation signals.
Subsequently, the chronoamperometry experiment was applied to study the electrochemical kinetics property, which can explain the signal enhancement mechanism more deeply. Figure 5C-F shows the charge (Q)-time (t) curves for the GCE, Ni 2 Co 1 -BPDC/GCE, Ni 2 Co 1 -NDC/GCE, and Ni 2 Co 1 -BDC/GCE in 0.1 M NaOH in the absence (a) and presence of 1 mM glucose (b). The apparent catalytic rate constant (K cat ) was determined according to the equation [44].
I cat /I L = (πk cat Ct) 1/2 (5) where I cat represents the catalytic current of analytes on the Ni 2 Co 1 -L/GCEs; I L is the limiting current in the absence of analytes; C and t represent the analyte's concentration and time elapse, respectively. The linear regression equation between I cat /I L with t 1/2 is given in the inset of Figure 5C-F. The k cat for the GCE, Ni 2 Co 1 -BPDC/GCE, Ni 2 Co 1 -NDC/GCE, and Ni 2 Co 1 -BDC/GCE was calculated as 1.43 × 10 2 , 1.77 × 10 2 , 3.01 × 10 2 , and 3.69 × 10 3 M −1 ·s −1 . In summary, the Ni 2 Co 1 -BDC/GCE had the strongest electron transfer capacity and the highest electrocatalytic rate constant for glucose oxidation and consequently exhibited the largest signal response. The possible reasons are summarized as follows: First, the Ni, Co bimetallic provided active metal oxidation sites for glucose. Second, the 2D lamellar structure was more beneficial to the contact of active sites and accelerated electron transfer and mass transfer. In summary, the Ni 2 Co 1 -BDC/GCE had good electrochemical properties, which can be applied to the construction of a glucose electrochemical sensing platform.
To further investigate the reaction kinetic characteristics of the Ni 2 Co 1 -BDC/GCE for glucose oxidation, the CV curves of the electrode at different sweep rates (50-300 mV/s) were tested in the NaOH solution containing 0.1 mM glucose ( Figure 6A). It can be seen from Figure 6B that both anodic and cathodic peak currents (I p ) increased in proportion to the square root of the scan rates. These results indicated that the electrochemical kinetics were diffusion-controlled. Moreover, the anodic peak potentials (E pa ) shifted more positively and the cathodic potentials (E pc ) shifted more negatively with increases in scan rates, which suggested the quasi-reversible electrochemical reaction process of the Ni 2 Co 1 -BDC on the electrode surface in the majority of activation sites [45]. Figure 6C presents the CVs of the Ni 2 Co 1 -BDC/GCE, including various concentrations of glucose. It was found that the peak current density of anodic oxidation increased with the augment of glucose concentration from 0 to 2 mM, which indicated that the Ni 2 Co 1 -BDC/GCE had good catalytic performance for glucose oxidation.
Molecules 2023, 28, x FOR PEER REVIEW 7 of 13 more deeply. Figure 5C-F shows the charge (Q)-time (t) curves for the GCE, Ni2Co1-BPDC/GCE, Ni2Co1-NDC/GCE, and Ni2Co1-BDC/GCE in 0.1 M NaOH in the absence (a) and presence of 1 mM glucose (b). The apparent catalytic rate constant (Kcat) was determined according to the equation [44].
In summary, the Ni2Co1-BDC/GCE had the strongest electron transfer capacity and the highest electrocatalytic rate constant for glucose oxidation and consequently exhibited the largest signal response. The possible reasons are summarized as follows: First, the Ni, Co bimetallic provided active metal oxidation sites for glucose. Second, the 2D lamellar structure was more beneficial to the contact of active sites and accelerated electron transfer and mass transfer. In summary, the Ni2Co1-BDC/GCE had good electrochemical properties, which can be applied to the construction of a glucose electrochemical sensing platform.
To further investigate the reaction kinetic characteristics of the Ni2Co1-BDC/GCE for glucose oxidation, the CV curves of the electrode at different sweep rates (50-300 mV/s) were tested in the NaOH solution containing 0.1 mM glucose ( Figure 6A). It can be seen from Figure 6B that both anodic and cathodic peak currents (Ip) increased in proportion to the square root of the scan rates. These results indicated that the electrochemical kinetics were diffusion-controlled. Moreover, the anodic peak potentials (Epa) shifted more positively and the cathodic potentials (Epc) shifted more negatively with increases in scan rates, which suggested the quasi-reversible electrochemical reaction process of the Ni2Co1-BDC on the electrode surface in the majority of activation sites [45]. Figure 6C presents the CVs of the Ni2Co1-BDC/GCE, including various concentrations of glucose. It was found that the peak current density of anodic oxidation increased with the augment of glucose concentration from 0 to 2 mM, which indicated that the Ni2Co1-BDC/GCE had good catalytic performance for glucose oxidation.

Non-Enzymatic Glucose Sensing Based on the Ni2Co1-BDC/GCE
The Ni2Co1-BDC/GCE was used to construct the enzyme-free glucose sensing because of its excellent electrochemical properties. To achieve the optimum sensing performance, the material preparation conditions, such as the effects of the bimetallic type and ratio (Ni, Ni:Co = 1:1, 2:1, 1:2, Co), the type and concentration of the applied base source (NaOH, KOH, TEA, TEAH) on the glucose oxidation signal were optimized. As shown in Figure S2A-C, the strongest current step signal was obtained with the

Non-Enzymatic Glucose Sensing Based on the Ni 2 Co 1 -BDC/GCE
The Ni 2 Co 1 -BDC/GCE was used to construct the enzyme-free glucose sensing because of its excellent electrochemical properties. To achieve the optimum sensing performance, the material preparation conditions, such as the effects of the bimetallic type and ratio (Ni, Ni:Co = 1:1, 2:1, 1:2, Co), the type and concentration of the applied base source (NaOH, KOH, TEA, TEAH) on the glucose oxidation signal were optimized. As shown in Figure  S2A-C, the strongest current step signal was obtained with the Ni 2 Co 1 -BDC/GCE (3 mmol KOH). Then, the test conditions (potential, modification amount, dispersing solvent) were also optimized. As shown in Figures 7A and S2D-F, the strongest current step signal was obtained under the test conditions of the test voltage of 0.50 V, dispersing solvent of DMF, and modification amount of 4 µL (5 µL Nafion).
The Ni2Co1-BDC/GCE constructed in this study had a higher sensitivity, a lower detection limit, and a wider linearity range than the other transition metal electrochemical sensors constructed with enzyme-free glucose previously reported in the literature, as summarized in Table 1. Therefore, the Ni2Co1-BDC/GCE was found to have broad application prospects in the construction of glucose sensors.
The interference of some possible co-existing biomolecules with the detection of glucose signals at Ni2Co1-BDC/GCE electrodes was investigated. The concentration of glucose in human serum is considered to be about 10 times higher than the concentration of the interfering biomolecule in the anti-interference test [47]. Therefore, the anti-interference test of the Ni2Co1-BDC/GCE was performed by successive addition of 100 µM glucose, 10 µM ascorbic acid (AA), 10 µM dopamine (DA), 10 µM uric acid (UA), and 1 mM KCl in 0.1 M NaOH. As can be seen from Figure 7D, the Ni2Co1-BDC/GCE had a very obvious current response to glucose, but the current response to these interferences was almost negligible, indicating that the sensor had good anti-interference ability. The relationship between glucose concentration and its oxidation signal was investigated under optimal conditions. Figure 7B shows the amperometric responses obtained for the Ni 2 Co 1 -BDC/GCE with successive additions of different concentrations of glucose in 0.1 M NaOH solution under the above optimal conditions. An excellent linear relationship was observed between the glucose concentration and its response signal in the range of 0.5 to 2899.5 µM ( Figure 7C). The linear regression equation was described as I pa (µA) = 274.8 C (mM) + 6.900 with R 2 = 0.997. The sensitivity of the Ni 2 Co 1 -BDC/GCE was 3925.3 µA mM −1 cm −2 , and the detection limit was 0.29 µM (S/N = 3). The current response was gradually saturated with the gradual increase in glucose concentration, probably because the electrode surface was partially covered by adsorbed reaction intermediates, thus not providing enough electroactive sites for the oxidation process of glucose [46]. The Ni 2 Co 1 -BDC/GCE constructed in this study had a higher sensitivity, a lower detection limit, and a wider linearity range than the other transition metal electrochemical sensors constructed with enzyme-free glucose previously reported in the literature, as summarized in Table 1. Therefore, the Ni 2 Co 1 -BDC/GCE was found to have broad application prospects in the construction of glucose sensors.
The interference of some possible co-existing biomolecules with the detection of glucose signals at Ni 2 Co 1 -BDC/GCE electrodes was investigated. The concentration of glucose in human serum is considered to be about 10 times higher than the concentration of the interfering biomolecule in the anti-interference test [47]. Therefore, the anti-interference test of the Ni 2 Co 1 -BDC/GCE was performed by successive addition of 100 µM glucose, 10 µM ascorbic acid (AA), 10 µM dopamine (DA), 10 µM uric acid (UA), and 1 mM KCl in 0.1 M NaOH. As can be seen from Figure 7D, the Ni 2 Co 1 -BDC/GCE had a very obvious current response to glucose, but the current response to these interferences was almost negligible, indicating that the sensor had good anti-interference ability. Moreover, the reproducibility of the proposed Ni 2 Co 1 -BDC/GCE-based sensors was further evaluated. The relative standard deviation (RSD) of five independently manufactured Ni 2 Co 1 -BDC/GCE measurements of 100 µM glucose was calculated as 3.06%. Meanwhile, the RSD of five consecutive glucose additions with one Ni 2 Co 1 -BDC/GCE was only 2.14%. The lower RSD values justified the good reproducibility of the fabricated sensors. Several Ni 2 Co 1 -BDC/GCEs were stored in air at room temperature, and the current responses for 100 µM glucose were measured each day. The current response retained 93.37% of its original value after a week, as illustrated in Figure S2G, indicating good stability of the as-prepared electrochemical sensor.
To study the practical value of the Ni 2 Co 1 -BDC/GCE, the i-t method was used to detect human serum samples, and the accuracy and reliability of the electrode were evaluated. The standard glucose solution and different amounts of serum were added sequentially to 10 mL of 0.1 M NaOH at a voltage of 0.50 V ( Figure S2H, Supporting Information). Then, blood glucose data measured from different human serum samples were compared with hospital tests, which showed satisfying results (Table 2). It can be seen that the blood glucose concentrations obtained from the Ni 2 Co 1 -BDC/GCE were well consistent with the hospital test results, which demonstrated the reliable practical value of the developed glucose sensor.

Instruments
Electrochemical performance experiments were carried out using a CHI 660E electrochemical workstation with a traditional three-electrode system. The saturated calomel electrode (SCE, Rosemead, CA, USA, in saturated KCl solution), platinum wire electrode, and the modified glassy carbon electrode (GCE) served as the reference electrode, counter electrode and the working electrode, respectively. The morphological characterization of the Ni 2 Co 1 -L MOFs was performed with a scanning electron microscope (SEM, Hitachi SU-8000) operated at an accelerating voltage of 10 kV. The composition and crystal structure of the Ni 2 Co 1 -L MOFs were checked by X-ray diffraction (XRD, Rigaku RINT 2500×) with Cu-Ka radiation. The Fourier transform infrared (FTIR) spectra of the samples were collected using the Avatar 360 Nicolet instrument.

Synthesis of Ni 2 Co 1 -L MOFs
In a typical preparation procedure, 0.67 mmol (0.1667 g) Ni(CH 3 COO) 2 ·4H 2 O and 0.33 mmol (0.0822 g) Co(CH 3 COO) 2 ·4H 2 O (Ni:Co = 2:1) were dissolved in 5 mL of DMF to form solution A. Meanwhile, 5 mL of ultrapure water was used to dissolve 3 mmol (0.1680 g) of KOH, and then 1 mmol of organic ligand (L = H 2 BPDC, 0.2423 g; H 2 NDC, 0.2162 g; H 2 BDC, 0.1663 g) was added. After the above solution was clear, 5 mL of DMF was added to form solution B. Solution A was poured into solution B and reacted for 1 h under rapid stirring conditions. Finally, the mixed solution was centrifuged at 6000 rpm for 10 min. The obtained precipitate was centrifuged and washed by DMF and ethanol (1:1) several times and dried at 60 • C. Finally, Ni 2 Co 1 -BPDC, Ni 2 Co 1 -NDC, and Ni 2 Co 1 -BDC MOFs were obtained.

Preparation of the Ni 2 Co 1 -L MOFs Modified GCE
For the modification, 5 mg/mL of the material suspension was prepared by ultrasonic dispersion of 2.5 mg of the Ni 2 Co 1 -L MOFs powder in 495 µL of DMF, followed by the addition of 5 µL of Nafion. Meanwhile, the surface of the GCE (diameter: 3.0 mm) was polished using 0.05 µm alumina slurry and then washed with ethanol and ultrapure water in an ultrasonic bath. After that, 4 µL of Ni 2 Co 1 -L dispersion was pipetted onto the surface of the GCE, which was dried with an infrared baking lamp. Then, the Ni 2 Co 1 -BPDC/GCE, Ni 2 Co 1 -NDC/GCE, Ni 2 Co 1 -BDC/GCE were obtained and used as the working electrodes.

Conclusions
In summary, a series of Ni/Co bimetallic MOFs (Ni 2 Co 1 -BPDC, Ni 2 Co 1 -NDC, Ni 2 Co 1 -BDC) were successfully prepared under simple room temperature conditions. The effects of metal centers and ligand types on the morphology and electrochemical sensing properties of the Ni 2 Co 1 -L MOFs were demonstrated by different characterization approaches. It was demonstrated that Ni 2 Co 1 -BDC had an ultrathin and a homogeneous disk-like structure with a large electrochemical active area, which can expose abundant active sites. Moreover, an excellent electrochemical sensing capability for glucose detection was demonstrated by the prepared electrochemical sensor, which obtained a wider linear range, a lower detection limit, and a more stable interference immunity. Last, satisfactory results were also found for the detection of human serum samples, and the utility and accuracy were further demonstrated.