English

• Physiological and pathological changes in organisms frequently lead to alterations in blood components, making detecting these changes of crucial clinical importance [1-3]. For example, imbalanced Glu levels can lead to a variety of diseases, such as hyperglycemia and hypoglycemia [4-6]. Sustained high Glu caused by various reasons can lead to diabetes, which is the most important disease related to Glu concentration. In addition, excessive Chol levels in the blood can cause hypercholesterolemia, which can lead to so-called "diseases of wealth" such as coronary atherosclerotic heart disease [7, 8]. Therefore, it is extremely important to accurately detect the concentrations of Glu and Chol in blood for the evaluation of human health. Clinically, Glu and Chol can be detected by their enzymatic products H₂O₂ [9, 10]. So far, various techniques such as electrochemistry [11], fluorescence [12] and chemiluminescence [13] have been applied for the detection of H₂O₂. Among them, fluorescence measurement is widely used for its high sensitivity, stability and accuracy [14]. Compared with single emission signal, dual emission ratio fluorescent probe is less affected by non-analyte-related factors, which can improve the accuracy and reliability of experimental results [15]. Therefore, it is of great significance to construct a new ratio dual-emission fluorescent probe for detecting H₂O₂.

  Metal-organic frameworks (MOFs) are nano-materials coordinated by organic ligands and metal ions, which were widely used in catalysis, environment and biosensing because of their large surface area, porosity and rich active sites [16]. However, the detection of Glu and Chol by the ratio fluorescent probe based on MOFs is not common. Among MOFs materials, NH₂−MIL-53(Al) not only has strong fluorescence, but also shows excellent stability in water and organic solvents. Therefore, it is feasible to design a ratio fluorescent probe based on NH₂−MIL-53(Al) to detect Glu and Chol. The technology of simultaneous detection of Glu and Chol from a single whole blood sample is still insufficient and plasma should be separated in advance, because the existence of blood cells will interfere with the detection results. Besides, plasma also contains other substances necessary for maintaining health, such as protein and other organic compounds, which are usually molecular targets for clinical diagnosis [17]. Therefore, the preparation of pure plasma samples can prevent the contamination of target biomarkers by ntracellular components released by hemolysis, as well as minimize optical interference caused by a high concentration of red blood cells (RBCs). Generally, the ideal plasma separation technology for point of care testing (POCT) should meet the following requirements: rapid separation, high purity, high protein recovery rate and low hemolysis. Due to the lack of laboratory infrastructure and the limitation of separation technology, it is still a challenge to separate a large amount of plasma from whole blood for on-site diagnosis. In actual clinical diagnosis, plasma separation is achieved by centrifugation, but these instruments are expensive, bulky, not portable and consume a lot of blood, which limits their application in POCT and resource-poor areas [18-20]. To overcome these shortcomings, many microfluidic technologies have been proposed for plasma separation, including inertial force separation [21], acoustic wave separation [22], gravitational sedimentation [23], microfiltration [24, 25]. However, many of these methods typically require substantial quantities of whole blood and are time-consuming, which will affect the timeliness, reliability and accuracy of the analysis results [26]. According to the "size sieving" principle, membrane-based plasma separation has attracted wide attention because of its portability, low cost and low hemolysis level [27]. More importantly, plasma separation and detection are primarily conductedseparately, rendering them impractical for portable applications. Therefore, the POCT diagnosis platform needs to develop a simple, cheap and fast equipment for plasma separation and detection.

  Here, we designed a plasma separator based on membrane and a dual-emission fluorescence sensing system based on NH₂–MIL-53(Al)/OPD for direct analysis of Glu and Chol in blood. The separation membrane is supported by the micro-column array and placed vertically, so that blood cells settle in the direction parallel to the membrane as much as possible, and plasma flows out in the direction perpendicular to the membrane, instead of being directly deposited on the membrane as in the existing design, which greatly reduces the blockage of blood cells on the membrane, improves the separation efficiency and avoids excessive hemolysis. NH₂–MIL-53(Al)/OPD sensing systems show dual emission response to H₂O₂, a catalytic product of Glu and Chol. Due to the FRET, the response of fluorescence intensity ratio (F_{574 nm}/F_{434 nm} or F_{554 nm}/F_{434 nm}) increases with the increase of H₂O₂ concentration, accompanied by obvious color change. The visual detection of Glu and Chol can be realized by capturing RGB [red (R), green (G) and blue (B)] values with smart phones. This integrated device based on smart phone has been successfully used to determine Glu and Chol in real blood samples with high precision, which can provide a general platform for sensing the biocatalytic process of H₂O₂ production. The whole process of serum separation and target detection is shown Scheme 1.

  Scheme 1

  

  Scheme 1.  Schematic diagram of integrated device for plasma separation and visual detection of Glu and Chol based on NH₂−MIL-53(Al) and o-phenylenediamine (OPD) combined with smart phone.

  DownLoad: Full-Size Img PowerPoint

  Aluminum chloride hexahydrate (AlCl₃·6H₂O), 2-aminoterephthalic acid (NH₂–H₂BDC), glucose oxidase (GOx), hydrogen peroxide (H₂O₂, 30%), horseradish peroxidase (HRP), glucose, sucrose, starch, aminoacids, various ions, and o-phenylenediamine (OPD) were obtained from Aladdin Reagent Co. (Shanghai, China). Cholesterol (Chol), and cholesterol oxidase (ChOx) were purchased from Macklin Biochemical Co., Ltd. (Shanghai, China). Blood separation membrane (asymmetric polysulfone membrane (Vivid GR, Pall Life Sciences, East Hills, NY)).

  XRD measurements were carried out on a D/max-2400 X-ray powder diffractometer (Japan, Rigaku), using CuKα radiation (40 kV, 30 mA) source with a resolution of 0.02° and a scanning speed of 0.5° min^-1. Laser cutter (Shen Zhen Da Hong Laser Equipment Co., Ltd.) was used to cut the membrane and double-sided tape. SEM images were recorded on an ULTRA Plus (Germany Zeiss), operated under high vacuum conditions to visualize the morphology and size distribution of the particles. Micro-precision CNC platform (Suzhou Hengwei Precision Technology Co., Ltd.) was used for processing the plasma separator.

  The integrated device consists of a 3D printed sample box with length, width and height of 9.5, 7.5 and 7.5 cm, respectively, a 365 nm ultraviolet LED lamp, a smart phone, arduino uno, micropump (P/N T6–1IC-03–1EEP, Parker Hannifen) and a plasma separator. Black resin was used as 3D printing material to reduce the reflection of ultraviolet light. The plasma separator was made with machine tool engraving technology. As shown in Fig. 1a, the plasma separator consists of two poly(methyl methacrylate) (PMMA) plates with a length, width and height of 16 mm, 5.4 mm and 16 mm, respectively. A 14 mm high × 12 mm wide × 2 mm deep groove was machined in the PMMA board with a precision milling machine. Then, a micro-pillar array with a length of 500 µm, a width of 500 µm, a depth of 400 µm and a pitch of 500 µm was precisely machined on the surface of the groove to be used as a plasma separation membrane support. At its bottom, there is an outlet with a diameter of 1.6 mm for connecting the blood separation device with the detection device. The separation membrane was cut to a length of 14 mm and a width of 12 mm using a laser machine. The double-sided adhesive tape (Deli Group Co., Ltd.) was cut into a rectangular frame which outer side and the inner side coincide with the separation membrane and the edge of the micro-column array. The blood separation membrane and the edge of the groove except the micro-column array were connected together by a double-sided rectangular frame. Due to the support of microcolumns to the membrane, the gaps between microcolumns and the blood separation membrane will form a closed cavity. Then, the 2 p.m. grooves were combined again with double-sided adhesive tape to form the plasma separator (Fig. 1b). The detection chip was made of 2 p.m. plates (The length, width and thickness are 6 cm, 4 cm and 2 mm, respectively) by thermocompression bonding. Four connecting holes with a diameter of 1.6 mm and three detecting holes with a diameter of 8 mm were drilled on the upper plate, respectively. Machining a channel with a width of 0.3 mm and a depth of 0.2 mm on the bottom plate (Fig. 1c). Finally, the separation device and micropump were connected with the detection chip by using a catheter with a diameter of 1.6 mm to form an integrated device (Fig. 1d).

  Figure 1

  

  Figure 1.  Schematic diagram of preparation of plasma separator. (a) Exploded view of plasma separator. (b) Plasma separator. (c) Thermocompression bonding of detection chip. (d) Integrated separation and detection device.

  DownLoad: Full-Size Img PowerPoint

  NH₂−MIL-53(Al) was prepared according to the reported method [28]. Briefly, 0.0724 g AlCl₃·6H₂O and 0.053 g 2-aminoterephthalic acid were dissolved in 15 mL deionized water respectively, then the two solutions were mixed and ultrasonicated for 30 min and transferred to a 50 mL high-pressure reaction kettle and reacted at 140 ℃ for 8 h. After cooling at room temperature, the light yellow product was collected by centrifugation, washed with DMF and deionized water for three times, and dried at 80 ℃ for 24 h.

  For the detection of H₂O₂, 80 µL of H₂O₂ standard solution with different concentrations was added to 320 µL of probe solution, which contained 40 µL of NH₂−MIL-53(Al) solution (0.2 mg/mL), 80 µL of OPD (40 mmol/L), 195 µL of PBS buffer (50 mmol/L, pH 6.0) and 5 µL of HRP (0.004 mg/L). Then the composite system was incubated in the dark for 15 min and the fluorescence intensity was measured under the excitation of 339 nm.

  The detection of Glu and Chol was similar to that of H₂O₂. Briefly, 80 µL of Glu/Chol standard solution with different concentrations was added to 320 µL of corresponding probe solution. The components of the probe solution are consistent with those of H₂O₂ except that the amount of PBS was changed to 185 µL, and 10 µL of GOx (0.6 mg/mL)/ChOx (0.04 mg/mL) were added. Then the probe solution containing Glu and Chol was incubated in the dark for 45 min and 25 min, respectively. Finally, the fluorescence intensity was measured under excitation of 339 nm.

  Whole blood was collected from SD male rat weighing about 200 g (Jiangsu Jicui Yikang Biotechnology Co., Ltd.) using a blood collection tube containing anticoagulant K₂EDTA to prevent blood cell aggregation. Blood samples diluted 10 times with PBS were loaded into the separator. Then 80 µL of probes containing GOx and ChOx were added to the corresponding detection holes respectively. Start the power supply and adjust the output power of the driving pump through the potentiometer of Arduino uno. Under the driving force provided by the micropump, the serum will fill the detection cell with a volume of 100 µL and reacted with the probe. With the help of smart phone, RGB values could be read by using App (Live Color Selector), and the linear relationship between G/B value and analyte concentration was established. Experiments involving animals were approved by the Animal Ethics Committee at the East China Normal University (Approved number: R20230603).

  Compared with traditional centrifugation, membrane provides a fast, convenient and effective method for separating plasma from whole blood. The membrane of most plasma separation devices was usually placed horizontally at the bottom, so blood cells will be directly deposited on the membrane. However, this method will make the membrane quickly blocked by blood cells, which will eventually affect the separation time and recovery rate of plasma [29]. In order to overcome this shortcoming, we vertically place the plasma separation membrane in the separator. Before separation, blood cells were firstly deposited at the bottom of the separator by gravity in the vertical direction. This arrangement made the blood cells settle in the direction parallel to the membrane as much as possible, instead of directly depositing on the membrane as in the existing design, which greatly reduces the blockage of the membrane caused by blood cells, improves the separation efficiency and avoids excessive hemolysis. First, 0.6 mL of fresh whole blood was added into the plasma separator and allowed to stand for 5 min (Figs. 2a and b). As the density of blood cells is higher than that of plasma, standing still can make blood cells deposit at the bottom of the chamber as much as possible under the action of gravity. At this time, the fluid at the upper part of the separator is much clearer than that at the bottom (Fig. 2c). When pressure was applied, the transmembrane pressure difference will force plasma to ooze in the direction perpendicular to the membrane, while blood cells will be intercepted by the membrane. The micro-column array in the separator can provide support for the membrane and prevent the membrane from deformation and rupture during separation (Figs. 2d and e). Because high shear stress can damage blood cells and lead to hemolysis [30], we evaluated the possible hemolysis caused by the membrane separation process. The absorbance of each plasma sample (hemolyzed water, blood diluted by PBS, centrifugation and membrane-based separation) in the range of 450–700 nm (Fig. 3a). There is no significant difference between the level of free hemoglobin and that obtained by centrifugation, which indicatevs that the plasma separation method based on membrane will not cause much hemolysis.

  Figure 2

  

  Figure 2.  Steps of plasma separation process: (a) fill the blood sample into the plasma separation device with a pipette; (b) gravity sedimentation of blood cells; (c) after 5 min, blood cells settled at the bottom; (d) the plasma is collected by the pressure provided by the pump and sent to the detection orifice; (e) plasma flow direction.

  DownLoad: Full-Size Img PowerPoint

  Figure 3

  

  Figure 3.  (a) Hemolysis analysis results. (b) SEM image of NH₂−MIL-53(Al). (c) XRD Patterns of NH₂−MIL-53(Al). (d) Fluorescence spectral responses of NH₂−MIL-53(Al), NH₂−MIL-53(Al)+OPD+HRP+H₂O₂, and HRP+OPD+H₂O.

  DownLoad: Full-Size Img PowerPoint

  SEM was performed to observe the morphology and structure of the sample. The SEM image of NH₂−MIL-53(Al) shows its unique three-dimensional gear shape (Fig. 3b), and the XRD pattern matches well with the previous reports (Fig. 3c). The diffraction peaks at 8.8°, 11.7°, 15.2°, 17.6° and 26.5° could be assigned as (101), (200), (011), (202), (020) lattice planes of NH₂−MIL-53(Al) [31], which indicating that it was successfully synthesized. The existence of amino group significantly improves its dispersibility in aqueous solution, allowing the entire sensing process to be conducted in a relatively uniform system. In addition, NH₂−MIL-53(Al) nanoparticles showed good fluorescence stability and stable sensitivity to the target within one week (Figs. S1 and S2 in Supporting information), which provided a stable environment for the fluorescence detection of bioactive molecules in blood.

  The changes in fluorescence optical characteristics of the system before and after the reaction were investigated to better understand the sensing process. The produced NH₂−MIL-53(Al) nanomaterial showed a significant emission peak at 434 nm when excited at 339 nm. Under the catalysis of HRP, OPD can be rapidly oxidized by H₂O₂ to oxOPD. Simultaneously, the oxidation products assembled on NH₂−MIL-53(Al) through hydrogen bonding and π-π interaction, which effectively quenches the fluorescence of NH₂−MIL-53(Al) and produces new emission peaks (Fig. 3d). The UV–vis absorption spectrum of the mixed system (OPD+HRO+H₂O₂) effectively overlapped with the fluorescence emission spectrum of NH₂−MIL-53(Al) (Fig. 4a), proving that the quenching phenomenon might be caused by the inner-filter effect (IFE) or fluorescence resonance energy transfer (FRET). Time-resolved fluorescence spectra were carried out to further study the quenching mechanism. As shown in Fig. 4b, with the addition of OPD and H₂O₂, the fluorescence lifetime of NH₂−MIL-53(Al) decreases. Therefore, the quenching mechanism of NH₂−MIL-53(Al) might be attributed to the FRET. These results are different from IFE, which typically demonstrates the same fluorescence lifetime [32]. The response of the fluorescence intensity ratio (F_{574 nm}/F_{434 nm} or F_{554 nm}/F_{434 nm}) increased due to FRET effect, accompanied by a distinct color change from weak to strong as the target concentration increases.

  Figure 4

  

  Figure 4.  (a) UV–vis absorption spectrum of NH₂−MIL-53(Al), OPD, NH₂−MIL-53(Al)+OPD and OPD+HRP+H₂O₂ and the fluorescence emission spectrum of NH₂−MIL-53(Al). (b) Fluorescence decay curves of NH₂−MIL-53(Al) in the absence and presence of H₂O₂ and OPD.

  DownLoad: Full-Size Img PowerPoint

  To further explore the potential applications of the research, the key lies in achieving the best sensing performance. Therefore, a systematic optimization of various parameters affecting the experiment was conducted, including temperature, pH value, and incubation time. As shown in Fig. S3 (Supporting information), the ratio of fluorescence intensity (F_{574 nm}/F_{434 nm}) increases as the temperature rises, reaching its peak value at 35 ℃. PBS buffer was used to optimize the pH value of the reaction. The fluorescence intensity ratio reached the maximum at pH 6.0. This can be attributed to the fact that weak acidic conditions are beneficial to the oxidation of OPD by H₂O₂ catalyzed by HRP, so pH 6.0 was chosen as the best reaction condition. Finally, the response dynamics of the sensing system to the target H₂O₂ was studied. The results shown that the reaction between H₂O₂ and NH₂−MIL-53(Al), OPD tends to be equilibrium after 14 min. For Glu and Chol, the optimum reaction time was 45 min and 25 min respectively, and the best temperature and pH value were consistent with H₂O₂. This disparity arises from the varying catalytic efficiencies of GOx and Chox.

  Based on the optimal experimental conditions mentioned above, the fluorescence response of the NH₂−MIL-53(Al) and OPD-based ratiometric fluorescence method for detecting H₂O₂ was investigated. The fluorescence intensity at 434 nm steadily dropped as the H₂O₂ concentration increased, while it gradually increased at 574 nm (Fig. 5a). Therefore, the fluorescence intensity ratio (F_{574 nm}/F_{434 nm}) also increases with the increase of H₂O₂ concentration, and has a linear correlation (F_{574 nm}/F_{434 nm} = 0.0102 C_H − 0.0080, R = 0.997) (Fig. 5b). The detection limit (LOD) was calculated to be 0.12 µmol/L based on a signal-to-noise (S/N) ratio of 3.

  Figure 5

  

  Figure 5.  (a) The emission spectrum of the system in the presence of different concentrations of H₂O₂. (b) The linear curve between the fluorescence intensity ratio and H₂O₂.

  DownLoad: Full-Size Img PowerPoint

  The sensitive response of NH₂−MIL-53(Al)/HRP/OPD to H₂O₂ provided a general platform for biomedical applications, capable of detecting any reaction involving H₂O₂ production. Many metabolites in human body can be oxidized to produce H₂O₂, such as Glu, Chol, lactic acid, choline, uric acid, sarcosine and xanthine [33]. Here, as a conceptual demonstration, we choose Glu and Chol as detection targets to evaluate the practical application performance of this method. As expected, with the increase of Glu/Chol concentration, the change of fluorescence intensity is consistent with the change of H₂O₂ (Figs. 6a and b). Figs. 6c and d show the linear correlation between the fluorescence intensity ratio and the concentrations of Glu and Chol, with the linear range of 10–200 µmol/L (F_{574 nm}/F_{434 nm} = 0.0054 C_Glu + 0.0237, R = 0.994, F_{554 nm}/F_{434 nm} = 0.0189 C_Chol − 0.1733, R = 0.995). Based on 3σ blank/slope, the detection limits of Glu and Chol are 0.64 µmol/L and 1.4 µmol/L, respectively. In comparison to the developed monitor approach [34-37], this method is more portable and can directly detect Glu and Chol in blood without centrifugation (Table 1).

  Figure 6

  

  Figure 6.  The changes of fluorescence emission spectra with the concentrations of Glu (a) and Chol (b). Calibration curve between fluorescence intensity ratio and concentrations of Glu (c) and Chol (d).

  DownLoad: Full-Size Img PowerPoint

  Table 1

  

  Table 1.  Comparison with other analysis methods.

  DownLoad: CSV

  Probe	Glu			Chol		Ref.
  	Linear range (µmol/L)	LOD (µmol/L)		Linear range (µmol/L)	LOD (µmol/L)
  AuNCs/ChOx@ZIF-8/PEI	0.1–2.4	0.073		/	/	[34]
  Bio@Ag NPs	10–400	53.39		0.5–40	8.16	[35]
  GSH−CuNCs-Pb2+-Zr4+	0.8–50	0.37		6–80	2.7	[36]
  AgNC-GOx/Ag+-FP	50–5000	50		/	/	[37]
  NH2−MIL-53(Al) /OPD	10–200	10		10–200	10	This work

  The anti-interference performance plays a decisive role in the practical application of the sensor. To assess the specificity of this method to the target in practical application, some common potential interfering substances such as l-arginine (Arg), l-cysteine (Cys), sucrose, fructose, Na⁺, K⁺, Ca²⁺, Fe²⁺, Zn²⁺ were measured. As shown in Figs. 7a and b, only Glu and Chol showed strong response compared with the interferents, which indicated that the method had strong specific response to the target analyte.

  Figure 7

  

  Figure 7.  Selectivity of sensing system to Glu (a) and Chol (b). The concentration of Glu and Chol was 0.1 mmol/L. The concentration of interfering substance was 1 mmol/L.

  DownLoad: Full-Size Img PowerPoint

  Through the variation and superposition of three RGB colors, various colors can be obtained. This color standard is one of the most extensively used color systems, encompassing nearly all colors visible to the naked eye [38, 39]. By converting the color changes into RGB signals through App (Live Color Selector), we improve the detection accuracy. Under UV-LED light, as the concentration of Glu and Chol increased, the G value of the reaction system gradually increased, while the B value gradually decreased. Fig. 8 shows the linear curve between the G/B value and the concentrations of Glu and Chol (G/B = 27.47 C_Glu + 0.624, R = 0.9924 and G/B = 0.052 C_Chol + 0.533, R = 0.9959). The portable biosensor based on smart phone has a linear range of 10–200 µmol/L and the detection limit of 10 µmol/L, which fulfills the requirements for testing real samples. Furthermore, this sensor can concurrently detect multiple samples without the need for complex and costly analytical instruments, which provides a new idea for detecting additional markers in human serum.

  Figure 8

  

  Figure 8.  Colorimetric responses of sensing system toward the Glu (a) and Chol (b) with various concentrations from 0 µmol/L to 200 µmol/L under UV light. Linear relationshipbetween the G/B value and the concentrations of Glu (c) and Chol (d).

  DownLoad: Full-Size Img PowerPoint

  To validate its feasibility and practicality in biological samples, the proposed integrated method was also used to detect Glu and Chol in three actual mouse blood samples. There was no obvious difference between the contents of Glu and Chol detected by this method and those obtained by commercial kits, which demonstrates that this sensing system has potential feasibility in practical application. Furthermore, recovery experiments were performed to assess the impact of the potential matrix in the sample on the analysis results. Equal volumes of Glu and Chol standard solutions (0, 20 and 50 µmol/L) were added to serum samples diluted 400 times and 100 times, respectively. The recovery rate was between 78.36% and 112%, and the relative standard deviation (RSD) was < 4%, indicating that this method can quantitatively analyze Glu and Chol in real samples. All results were illustrated in Table S1 (Supporting information). In addition, the changes of Glu and Chol concentrations in blood of mice before and after feeding were analyzed by this method. As depicted in Fig. S4 (Supporting information), the concentration of Glu noticeably decreased after a 24-h fast. Following a 3-h feeding period, the blood Glu concentration increased and reached a stable level. Due to the minor trauma inflicted on mice during this process, it becomes difficult for the Glu content to return to its normal levels within a short period of time. On the contrary, the Chol content did not change significantly within a few hours before and after feeding. Lu et al. also proved through animal experiments that compared with mice fed with conventional feed, HMGCR protein in liver and total Chol in blood and liver of mice fed with high-sugar and high-fat food for a long time increased significantly [40]. All results show that the device can be used for accurate quantification and monitoring of Glu and Chol. The photo of the device is shown in Fig. S5 (Supporting information).

  In conclusion, an integrated device for separating plasma from whole blood samples and accurately detecting Glu and Chol is designed. In the process of plasma separation, due to the vertical design of the separation membrane and the gravity deposition of blood cells, the separation effect of the membrane is improved. Based on NH₂−MIL-53(Al)/OPD fluorescent dual-emission probe and smart phone, the visual detection of Glu and Chol was realized. The proposed method has good reproducibility and reliability which provided a universal portable detection platform for the biocatalytic process of H₂O₂ production without centrifugation of blood.

  CRediT authorship contribution statement

  Fangbing Wang: Writing – review & editing, Writing – original draft, Methodology, Investigation. Qiankun Zeng: Methodology. Jing Ren: Resources. Min Zhang: Writing – review & editing, Supervision, Conceptualization. Guoyue Shi: Supervision.

  Acknowledgments

  This work was supported by the National Natural Science Foundation of China (Nos. 22274053 and 22274051), the Shanghai Municipal Science and Technology Major Project ("Beyond Limits manufacture"), and Natural Science Foundation of Chongqing, China (No. CSTB2023NSCQ-MSX0339).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110494.

  
  
• 1. [1]

     C.O. Adetunji, O.S. Michael, S. Rathee, et al., Mater. Today Adv. 13 (2021) 100198.

  2. [2]

     S. O'Neill, L. O'Driscoll, Obes. Rev. 16 (2015) 1–12. doi: 10.1111/obr.12229

  3. [3]

     J. Khunger, M. Malhotra, N. Kumar, et al., Blood 134 (2019) 4959. doi: 10.1182/blood-2019-125048

  4. [4]

     J. Gall, E. Frisdal, R. Bittar, et al., Atherosclerosis 252 (2016) e205.

  5. [5]

     H. Takase, K. Hayashi, F. Kin, et al., Hypertens. Res. 46 (2023) 236–243. doi: 10.1038/s41440-022-01035-7

  6. [6]

     L. Haverals, K.V. Dessel, A. Verrijken, et al., Nutr. Diabetes 9 (2019) 16.

  7. [7]

     D.P. Yang, W.W. Guo, Z.F. Cai, et al., Sens. Actuators B: Chem. 260 (2018) 642–649.

  8. [8]

     J. Gao, W.Z. Huang, Z.P. Chen, C.Q. Yi, L.L. Jiang, Sens. Actuators B: Chem. 287 (2019) 102–110.

  9. [9]

     C.B. Ma, Y. Zhang, Q. Liu, Y. Du, E.K. Wang, Anal. Chem. 92 (2020) 5319–5328. doi: 10.1021/acs.analchem.9b05858

  10. [10]

      C.Y. Hong, X.X. Zhang, C.Y. Wu, et al., ACS Appl. Mater. Interfaces 12 (2020) 54426–54432. doi: 10.1021/acsami.0c15900

  11. [11]

      S. Singh, M. Singh, K. Mitra, et al., Electrochim. Acta 258 (2017) 1435–1444.

  12. [12]

      Z.L. Yuan, S.H. Chen, B. Liu, J. Mater. Sci. 52 (2017) 164–172. doi: 10.1007/s10853-016-0318-5

  13. [13]

      Z.J. Liu, L. Wang, P.F. Liu, et al., Food Chem. 357 (2021) 129753.

  14. [14]

      W.X. Lv, X. Wan, J.H. Wu, et al., Chin. Chem. Lett. 30 (2019) 1635–1638.

  15. [15]

      R.G. Gui, H. Jin, X.N. Bu, et al., Coord. Chem. Rev. 383 (2019) 82–103.

  16. [16]

      L. Guo, Y. Liu, R.M. Kong, et al., Anal. Chem. 91 (2019) 2453–12460.

  17. [17]

      M. Kersaudy-Kerhoas, E. Sollier, Lab Chip 13 (2013) 3323–3346. doi: 10.1039/c3lc50432h

  18. [18]

      B. Kim, S. Oh, D.W. You, S.Y. Choi, Anal. Chem. 89 (2017) 1439–1444. doi: 10.1021/acs.analchem.6b04587

  19. [19]

      M.F. Ulum, L. Maylina, D. Noviana, D.H.B. Wicaksono, Lab Chip 16 (2016) 1492–1504.

  20. [20]

      C.C. Liu, S.C. Liao, J.Z. Song, et al., Lab Chip 16 (2016) 553–560.

  21. [21]

      S.S. Zhou, T. Hu, F. Zhang, et al., Anal. Chem. 92 (2020) 1574–1581. doi: 10.1021/acs.analchem.9b04852

  22. [22]

      A. Tajudin, K. Petersson, A. Lenshof, et al., Lab Chip 13 (2013) 1790–1796. doi: 10.1039/c3lc41269e

  23. [23]

      X.B. Zhang, Z.Q. Wu, K. Wang, et al., Anal. Chem. 84 (2012) 3780–3786. doi: 10.1021/ac3003616

  24. [24]

      J. Hauser, G. Lenk, J. Hansson, et al., Anal. Chem. 90 (2018) 13393–13399. doi: 10.1021/acs.analchem.8b03175

  25. [25]

      J. Hauser, G. Lenk, S. Ullah, et al., Anal. Chem. 91 (2019) 7125–7130. doi: 10.1021/acs.analchem.9b00204

  26. [26]

      W.S. Mielczarek, E.A. Obaje, T.T. Bachmann, M. Kersaudy-Kerhoas, Lab Chip 16 (2016) 3441–3448.

  27. [27]

      C.C. Lien, P.J. Chen, A. Venault, et al., J. Membr. Sci. 584 (2019) 148–160.

  28. [28]

      H.N. Zhao, Z.P. Xing, S.Y. Su, et al., Appl. Catal. B: Environ. 291 (2021) 120106.

  29. [29]

      K.R. Baillargeon, L.P. Murray, R.N. Deraney, C.R. Mace, Anal. Chem. 92 (2020) 16245–16252. doi: 10.1021/acs.analchem.0c04127

  30. [30]

      J.H. Son, S.H. Lee, S. Hong, et al., Lab Chip 14 (2014) 2287–2292.

  31. [31]

      Y.B. Guan, M., Xia, X.H. Wang, W.C. Cao, A. Marchettib, Inorganica Chim Acta 484 (2019) 180–184.

  32. [32]

      Y.H. Yuan, J.Z. Jiang, S.P. Liu, et al., Sens. Actuators B: Chem. 242 (2017) 545–553.

  33. [33]

      Y. Zhao, H. Liu, Y.N. Jiang, et al., Anal. Chem. 90 (2018) 14578–14585. doi: 10.1021/acs.analchem.8b04691

  34. [34]

      J.L. Ding, W.Q. Zhang, F.Y. Xue, et al., Microchim. Acta 189 (2022) 203.

  35. [35]

      H.V. Xu, Y. Zhao, Y.N. Tan, ACS Appl. Mater. Interfaces 11 (2019) 27233–27242. doi: 10.1021/acsami.9b08708

  36. [36]

      F. Qu, Q.Q. Yang, B.J. Wang, J.M. You, Talanta 207 (2020) 120289.

  37. [37]

      S.S. Jiang, Y.F. Zhang, Y.C. Yang, et al., ACS Appl. Mater. Interfaces 11 (2019) 10554–10558. doi: 10.1021/acsami.9b00010

  38. [38]

      W.R. Zhang, Y.M. Zhang, F.L. Xie, et al., Adv. Mater. 32 (2020) 2003121.

  39. [39]

      K.K. Zheng, Y. Tong, S.H. Zhang, et al., Adv. Funct. Mater. 31 (2021) 2102599.

  40. [40]

      X.Y. Lu, X.J. Shi, A. Hu, et al., Nature 588 (2020) 479–484. doi: 10.1038/s41586-020-2928-y

• 

  Scheme 1  Schematic diagram of integrated device for plasma separation and visual detection of Glu and Chol based on NH₂−MIL-53(Al) and o-phenylenediamine (OPD) combined with smart phone.

  下载: 全尺寸图片 幻灯片
  

  Figure 1  Schematic diagram of preparation of plasma separator. (a) Exploded view of plasma separator. (b) Plasma separator. (c) Thermocompression bonding of detection chip. (d) Integrated separation and detection device.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Steps of plasma separation process: (a) fill the blood sample into the plasma separation device with a pipette; (b) gravity sedimentation of blood cells; (c) after 5 min, blood cells settled at the bottom; (d) the plasma is collected by the pressure provided by the pump and sent to the detection orifice; (e) plasma flow direction.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  (a) Hemolysis analysis results. (b) SEM image of NH₂−MIL-53(Al). (c) XRD Patterns of NH₂−MIL-53(Al). (d) Fluorescence spectral responses of NH₂−MIL-53(Al), NH₂−MIL-53(Al)+OPD+HRP+H₂O₂, and HRP+OPD+H₂O.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) UV–vis absorption spectrum of NH₂−MIL-53(Al), OPD, NH₂−MIL-53(Al)+OPD and OPD+HRP+H₂O₂ and the fluorescence emission spectrum of NH₂−MIL-53(Al). (b) Fluorescence decay curves of NH₂−MIL-53(Al) in the absence and presence of H₂O₂ and OPD.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  (a) The emission spectrum of the system in the presence of different concentrations of H₂O₂. (b) The linear curve between the fluorescence intensity ratio and H₂O₂.

  下载: 全尺寸图片 幻灯片
  

  Figure 6  The changes of fluorescence emission spectra with the concentrations of Glu (a) and Chol (b). Calibration curve between fluorescence intensity ratio and concentrations of Glu (c) and Chol (d).

  下载: 全尺寸图片 幻灯片
  

  Figure 7  Selectivity of sensing system to Glu (a) and Chol (b). The concentration of Glu and Chol was 0.1 mmol/L. The concentration of interfering substance was 1 mmol/L.

  下载: 全尺寸图片 幻灯片
  

  Figure 8  Colorimetric responses of sensing system toward the Glu (a) and Chol (b) with various concentrations from 0 µmol/L to 200 µmol/L under UV light. Linear relationshipbetween the G/B value and the concentrations of Glu (c) and Chol (d).

  下载: 全尺寸图片 幻灯片

  Table 1.  Comparison with other analysis methods.

  Probe	Glu			Chol		Ref.
  	Linear range (µmol/L)	LOD (µmol/L)		Linear range (µmol/L)	LOD (µmol/L)
  AuNCs/ChOx@ZIF-8/PEI	0.1–2.4	0.073		/	/	[34]
  Bio@Ag NPs	10–400	53.39		0.5–40	8.16	[35]
  GSH−CuNCs-Pb2+-Zr4+	0.8–50	0.37		6–80	2.7	[36]
  AgNC-GOx/Ag+-FP	50–5000	50		/	/	[37]
  NH2−MIL-53(Al) /OPD	10–200	10		10–200	10	This work

  下载: 导出CSV
•