Design of nitrobenzene vapor sensor based on PCN-224

It is vital to develop a quick and reliable approach for nitrobenzene in gas phase detection. In this work, the quartz crystal microbalance (QCM) platform was used to take advantage of the porphyrin-based metal-organic frameworks (MOFs) of PCN-224 as a novel sensing material for the sensitive and selective detection of nitrobenzene vapor. Besides, a quick response time of only 16 s was achieved. Meanwhile, the frequency shift in gas sensor could be proportionally quenched in correlation with the concentration of nitrobenzene vapor. Additionally, this PCN-224-based sensor demonstrated great stability to nitrobenzene vapor. This foreshadowed the huge potential of nitrobenzene detection in gas phase for practical applications in public security.


Introduction
The production of azo colors, insecticides, herbicides, munitions, and anilines all depend on the use of nitrobenzene, a highly poisonous and carcinogenic nitrocellulose aromatic chemical [1,2]. Toxicological studies have demonstrated that nitrobenzene at very low concentrations poses a high risk to the environment and human health, and has been certified as a 2B carcinogen by the United States Environmental Protection Agency [3]. Detection of nitrobenzene vapor has earned paramount significance because of its extreme volatility, toxicity, and high vapor pressure [4]. Notably, special attention should be paid to that nitrobenzene is volatile In this work, PCN-224 was successfully prepared. A simple drop coating method was used to attach PCN-224 to the surface of QCM. This is a novel QCM sensor and its gassensing performance was studied using nitrobenzene vapor as the target gas. The sensing results showed that PCN-224 based QCM sensor exhibited significant sensing properties for nitrobenzene vapor at room temperature, including high sensitivity and selectivity, rapid response/recovery speed, and high stability.

Chemicals and materials
The chemicals and materials for synthesizing the PCN-224 are displayed in the Supporting Information.

Synthesis of PCN-224
The process of synthesizing the PCN-224 are displayed in the Supporting Information.

Characterization, fabrication and test methods of the QCM Sensor
The characterization instruments and methods of PCN-224 are displayed in Supporting Information.
The fabrication and test methods of the PCN-224 based QCM sensor are obtained from a previous report and are described in detail in the Supporting Information. Figure S1 in the Supporting Information shows a schematic of the testing system. Figure 1 presents the scanning electron microscope (SEM) image of the PCN-224 sample and corresponding grain size distribution. The scale bar of Fig. 1a is 500 nm. It is observed that the PCN-224 crystals exhibited a cubic shape. Besides, the PCN-224 crystals possessed uniform and regular geometry, which should be attributed to the forming of well-crystallized PCN-224 during the synthesis process. As shown in Fig. 1b, the grain size of PCN-224 crystal is distributed in the range of 200-700 nm, which indicates that there are some differences in the stirring and nucleation rates in different regions during the synthesis process.

Materials characterization
The high resolution transmission electron microscope (HRTEM) image showed the PCN-224 crystals were solid (Fig. 2a). We further confirmed the elemental components of PCN-224 with TEM-mapping of a selected crystal in a HAADF-STEM image. TEM-mapping confirms the homogeneous distribution of elements including C, N, O, and Zr (Fig. 2b).
The Fourier transform infrared (FTIR) spectra of the PCN-224 is depicted in Fig. 3b. The asymmetric vibrational absorption near 1662 cm − 1 should be ascribed to C = O groups from 5,10,15,20-tetrakis (4-carboxyphenyl) porphyrin [18]. The peak near 1541 cm − 1 is favored by the C = C stretching vibration. The wide peak located at 3418 cm − 1 is assigned to the O-H vibrations. The peak near 971 cm − 1 is attributed to the N-H bond adsorbs vibration [19]. The vibrational bands around 1408 and 1605 cm − 1 are assigned  [20].
The structure information of PCN-224, including BET surface area, pore volume, and pore size, was investigated based on experimental data of nitrogen (N 2 ) adsorptiondesorption isotherms. Figure 4 shows the N 2 adsorption isotherm of PCN-224 exhibiting a reversible type-I adsorption behavior, indicating the characteristic of microporous property. As shown in the table in Fig. 4, the average pore size of PCN-224 is 2.8864 nm. The BET surface area of PCN-224 is 2532.3 m 2 g − 1 . The total pore volume of PCN-224 is 1.0309 mL g − 1 .

Gas Sensing Properties
The PCN-224 was adopted to fabricate gas sensor. Generally, the gas sensors based on QCM platform do not need heating to work better. Therefore, we only studied the frequency shift at several common ambient temperatures, including 293 K, 298 K, 303 K, and 308 K. The working temperature column chart to 50 ppm nitrobenzene vapor is given in Fig. 5a. This column chart shows that the PCN-224-QCM has obvious high frequency shift values to 50 ppm nitrobenzene vapor. The column chart also illustrates that the frequency shift value is negatively correlated with the ambient temperature, which should be attributed to the endothermic reaction of adsorption. It is obvious that the change of ambient temperature will change the frequency shift of PCN-224-QCM to nitrobenzene vapor, which may weaken the accuracy in practical application. But we can use some post-processing methods, such as error compensation method, to improve the practicability of PCN-224-QCM in the future.  that counted from fresh air flowing in to achieving the 90% recovery frequency shift change is about 26 s, which displays a fast response and recovery speed. Short-term repeatability of the gas sensors is related to the signal reliability. Therefore, the cyclic repeatability test for three loops is shown in Fig. 7b. The frequency shift curve of PCN-224-QCM sensor can recover to the baseline well for each loop, there are no obvious deviation during the three cyclic tests. This result proves the favorable short-term repeatability of PCN-224-QCM sensor.
To research the relationship between the frequency shift value and the gas concentration, the gradient gas concentration test was given in Fig. 7c. With the concentration of nitrobenzene vapor changing from 5 to 100 ppm, the frequency shift value of the PCN-224-QCM sensor gradually increases. To observe the relationship more clearly, the dotline plot transformed from Fig. 7c is shown in Fig. 7d. The stage from 5 to 100 ppm is obvious linear relationship. The linear fitting equation is y = -1.05 x − 11.50, of which y is the frequency shift value, and x is the concentration of nitrobenzene vapor. Besides, the goodness of fit value is 0.9928 closed to 1, demonstrating the great reliability of the above equation.
The potential nitrobenzene sensing mechanism should be attributed to porphyrin structure of PCN-224. First, the Selectivity is one of the most important properties of gas sensors. There are six kinds of gases tested to evaluate the selectivity, as shown in Fig. 5b. The PCN-224-QCM with PCN-224 modification performed strong tendency to nitrobenzene vapor sensing. Consequently, the radar map transformed from Fig. 5b is summarized in Fig. 6a. It is obvious that PCN-224-QCM has the selectivity to nitrobenzene vapor. If we want to measure the anti-interference capability of the above sensor, the comparison between the best and second sensing gases should be provided. Define R best as the frequency shift of sensor to the best, and R second as the frequency shift of sensor to the second, then the calculation method is as follows [21].
Anti-interference ability: R best /R second . As shown in Fig. 6b, the anti-interference capability of the PCN-224-QCM is 3.98. This result proves the anti-interference capacity of PCN-224-QCM sensor.
With the excellent response and anti-interference ability to nitrobenzene vapor, we further evaluated PCN-224-QCM sensor to the nitrobenzene vapor. The frequency shift curve of PCN-224-QCM sensor to 5 ppm nitrobenzene vapor at the ambient temperature (298 K) and default humidity (50% RH) is given in Fig. 7a. With the nitrobenzene vapor flowing into the testing chamber, the response time of achieve 90% frequency shift is about 16 s, and the recovery time which is called as decay ratio. As shown in Fig. 8b, all decay ratios are lower than 10%, which means that the PCN-224-QCM sensor can still maintain more than 90% response capability in a wide concentration range of nitrobenzene vapor after five weeks.
Finally, we list the comparison between nitrobenzene vapor sensors in reported work and this work. It is obviously that other reported nitrobenzene vapor sensors have not mentioned relevant investigation on selectivity, while the PCN-224-QCM reported in this paper has the ability to resist the interference of five common gases. Besides, PCN-224-QCM has a wide detection range for nitrobenzene vapor (5-100 ppm). The above data demonstrate the excellent performance of the nitrobenzene vapor sensor we hollow ring structure of porphyrin and the benzene ring are very easy to occur π-π adsorption, which is one of the main causes of gas sensing mechanism [22]. Secondly, the N atom in porphyrin and nitro group of nitrobenzene tend to be adsorbed by hydrogen bond interaction [16].
Long-term stability testing is a necessary method to measure the long-term reliability of sensors. The frequency shift values of the PCN-224-QCM sensor for five weeks are given (Fig. 8a). After the first term five weeks to age, the sensor met the stable stage, and the tendency to decrease is quite slight during this term, which proves the good longterm stability of the PCN-224-QCM sensor. In addition, we calculated the ratio of the frequency shift after five weeks to the initial frequency shift under various concentrations,

Supporting information
Chemicals and Materials; Synthesis of CN-224; characterization; fabrication and test methods of the QCM sensor; schematic of the gas testing system.

Declarations
Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. designed in Table 1. It is obviously that other reported nitrobenzene vapor sensors have not mentioned relevant investigation on selectivity, while the PCN-224-QCM reported in this paper has the ability to resist the interference of five common gases. Besides, PCN-224-QCM has a wide detection range for nitrobenzene vapor (5-100 ppm). The above data demonstrate the excellent performance of the nitrobenzene vapor sensor we designed.

Conclusion
In summary, porphyrin-based PCN-224 has been firstly developed for the rapid and specific detection of nitrobenzene in gas phase based on the elaborated QCM sensing platform. Significant sensing characteristics of this sensor included excellent sensitivity, selectivity, and quick