3.1 Characterisation of CuCo-PTC MOF
CuCo-PTC MOF was prepared by the coordination polymerisation between Co, Cu, and H3PTC in DMF and ethanol. The shape and dimension of the product were obtained by scanning electron microscopy (SEM) and transmission electron microscopy (TEM). Fig. 2a and 2b show the prepared CuCo-PTC MOF, which presents a rough surface, and stratified and disordered accumulation. The chemical composition was determined by energy dispersive X-ray spectroscopy (EDS), which showed that the CuCo-PTC MOF contains Cu, Co, O, and C elements, and that these elements have a uniform distribution in samples (Fig. 3). We also tested for organic ligands H3PTC and CuCo-PTC MOF using Fourier transform infrared spectroscopy, as shown in Fig. S1a. The most important functional group of H3PTC is carbonyl at 1700 cm-1, which produces a red shift in the crystallisation process. The disappearance of the signal at 1700 cm-1 in CuCo-PTC MOF indicates that the carboxyl group of the organic ligand is fully protonated and bound to the organometallic ion and that the carbonyl absorption peak of CuCo-PTC MOF shifted to 1568 cm-1. The X-ray diffraction testing for the organic ligand H3PTC and CuCo-PTC MOF material showed that the removal of reaction solvent influences the changes in pore structure, which finally leads to the amorphisation of MOF. Therefore CuCo-PTC MOF material does not show a well-defined diffraction peak in its XRD pattern (Fig. S1b), like the organic ligand.
We also investigated the surface element composition and chemical state of CuCo-PTC MOF material. The X-ray photoelectron spectroscopy (XPS) measurement showed that Co atoms are in a metallic state corresponding to the Co2+ reduction, with the binding energy of 778.68 eV for Co 2p1/2. The increase in the binding energy of Co 2p1/2 in CuCo-PTC MOF compared to standard cobalt metal (BE = 778.2 eV) is due to N–ligand interaction (Fig. 4a) [13]. The nitrogen adsorption experiments to determine the structure of the holes of CuCo-PTC MOF, shown in Fig. 4b, revealed that CuCo-PTC MOF had a specific surface area of 16.5125g·cm3. Finally, we conducted thermogravimetric analysis (TG) in the air at a slope of 10 ℃·min-1 to test the thermal stability of CuCo-PTC MOF. The curves showed three important weightlessness steps. As shown in Fig. 4c, the TG curve indicated that the thermal stability temperature was 450 ℃ owing to the decomposition of the tricarboxylate linker. The decomposition of the skeleton and the formation of mixed-valence Co3O4 happened at 0–90 ℃. Thermogravimetric analysis of CuCo-PTC MOF showed that it had high thermal stability at a specific temperature.
3.2 Design and detection mechanism of colourimetric probes
The use of colourimetric probes to detect NO (Fig. 2) showed that, when NO is not present, all OPD in the solution oxidizes to OPDox, and the hydroxyl radical is produced by CuCo-PTC MOF catalysis of H2O2. The fluorescence emission by hydroxyl radical changes the colour of the solution from colourless to yellow. When NO exists, the OPD in solution first reacts with NO to form N-(2-hydrazinophenyl) methylamine. When CuCo-PTC MOF and H2O2 were added to the reaction solution, the concentration of oxidizable OPD decreases with increasing NO concentration. Therefore, this sensing strategy is feasible for NO detection in terms of method design.
3.3 Feasibility of NO sensing
Fluorescence emission spectroscopy was performed on the solution containing 0.1 mol·L-1 of PBS, 10 mg·L-1 of CuCo-PTC MOF, 26 μmol·L-1 of OPD and 12 mol·L-1 of hydrogen peroxide. As Fig. 5 shows, when NO (6 μmol·L-1) was added to the detection system, the fluorescence quenched at 568 nm at the excitation wavelength of 520 nm. These results show that fluorescence colourimetry can be used for NO detection.
3.4 Composition optimisation
The key to obtaining a good signification signal is the optimum composition of the reaction solution because the detection of NO by fluorescence colourimetry involves the oxidation of OPD by hydrogen peroxide and related chemical reactions. First, we optimize the concentration of the PBS. As Fig. S2 shows, we investigated the concentration range of PBS between 0.001 mol·L-1 and 0.1 mol·L-1 in the NO detection system. For the PBS concentration of 0.1 mol·L-1, the Uv-vis absorption of the gold nanoparticles solution reached the maximum, meaning that the optimum concentration of PBS is 0.1 mol·L-1.
Next, we researched the influence of the concentration of the catalyzer CuCo-PTC MOF on the intensity of fluorescence signal in the detection system. As shown in Fig. S3, the fluorescence emission of the detection system reaches the maximum when the concentration of the CuCo-PTC MOF is 8 mg·L-1.
We subsequently optimized the concentration of OPD, which acts as the fluorescent emitter in the detection system, as shown in Fig. S4. The fluorescence emission of the detection system reached the maximum when the concentration of OPD was 24 μmol·L-1. We then optimized the concentration of H2O2, as shown in Fig. S5, the best concentration of H2O2 was 16 mol·L-1. Finally, we also found that in the presence of CuCo-PTC MOF catalyst, the detection system could be oxidized by H2O2 within 5 min (Fig. S6).
3.5 Sensitivity
The fluorescent emission intensity of OPDox can be controlled by changing reaction conditions. The fluorescence intensity is proportional to the concentration of OPDox, and with the oxidation of OPD, the fluorescence intensity increases. The variation in composition can be directly measured because the signal change can be quantified. NO and OPD react to form N-(2-hydrazinophenyl) methylamine, which can indirectly change the concentration of OPDox. Therefore, it can achieve the purpose of detecting NO (Technological process 1). Under the optimum reaction conditions, the concentrations of the components were 0.1 mol·L-1 of PBS, 10 mg·L-1 of CuCo-PTC MOF, 26 μmol·L-1 of OPD, and 12 mol·L-1 of hydrogen peroxide. The fluorescence titration curve showed that with the increase in NO concentration, the fluorescence intensity at 568 nm decreased linearly (Fig. 6) until the concentration of NO reached 9 μmol·L-1 (Fmax = 568 nm) and the fluorescence intensity of the reaction solution reached the minimum value. The linear equation can be fitted by Y = −114.969X + 675.182 (R2 = 0.996). The limit of detection (LOD) between NO and the reaction system is 0.15 μmol·L-1, and the optical sensor shows high sensitivity to NO.
3.6 Selectivity
We studied the selectivity of the detection system for NO detection, including some of the normal interferences, such as NO2-, NO3-, K+, Ca2+, and glucose, recording the fluorescence intensity of reaction solution. As shown in Fig. 7, the fluorescence intensity of OPDox in the solution had no weakening when there were only 3 μmol·L-1 interferences in the reaction solution. When 3 μmol·L-1 of NO coexisted with 3 μmol·L-1 interferences in the reaction solution, the fluorescence intensity of OPDox in the solution weakened. The results showed no apparent interferences for NO detection, which confirmed that the optical sensor had relatively high selectivity.
3.7 Actual sample testing
Cell lysate was prepared for evaluating the NO detection capability of FMC detection strategy in actual samples. The fluorescence emission intensity of cell lysate samples at different concentrations of NO was collected to evaluate the NO recovery rate of the FMC detection strategy. Table 1 lists all the results. There was no NO in unmarked cell lysate. Then, we had the standard addition recovery experiment in the proper range for evaluating this method's detection potential for NO in the actual environment. We added 0.3, 1.0, 3.0, and 6.0 μmol·L-1 NO to the samples and repeated the experiment three times. As shown in Table 1, the interferences of positive samples (cell lysates) did not influence NO detection. The recovery rate was between 98.5% and 103.6%. The relative standard deviation was between 0.4% and 1.8%. The results showed that this sensing strategy has great application potential for actual samples in NO detection.