A Study on the Impact of Poly(3-hexylthiophene) Chain Length and Other Applied Side-Chains on the NO2 Sensing Properties of Conducting Graft Copolymers

The detection and concentration measurements of low concentrations of nitrogen dioxide (NO2) are important because of its negative effects on human health and its application in many fields of industry and safety systems. In our approach, conducting graft copolymers based on the poly(3-hexylthiophene) (P3HT) conducting polymer and other side-chains, polyethylene glycol (PEG) and dodec-1-en, grafted on a poly(methylhydrosiloxane) backbone, were investigated. The grafts containing PEG (PEGSil) and dodec-1-en (DodecSil) in two variants, namely, fractions with shorter (hexane fraction -H) and longer (chloroform fraction -CH) side-chains of P3HT, were tested as receptor structures in NO2 gas sensors. Their responses to NO2, within the concentration range of 1–20 ppm, were investigated in an nitrogen atmosphere at different operating temperatures—room temperature (RT) = 25 °C, 50 °C, and 100 °C. The results indicated that both of the copolymers with PEG side-chains had higher responses to NO2 than the materials with dodec-1-en side-chains. Furthermore, the results indicated that, in both cases, H fractions were more sensitive than CH fractions. The highest response to 1 ppm of NO2, from the investigated graft copolymers, had PEGSil H, which indicated a response of 1330% at RT and 1980% at 100 °C. The calculated lower-limit of the detection of this material is lower than 300 ppb of NO2 at 100 °C. This research indicated that graft copolymers of P3HT had great potential for low temperature NO2 sensing, and that the proper choice of other side-chains in graft copolymers can improve their gas sensing properties.


Introduction
In the contemporary era of civilization, one of the major problems facing society is air pollution. One of the most common forms of pollution is nitrogen dioxide (NO 2 ), which is associated with adverse effects on human health. This is because, at high concentrations, it can cause inflammation of the airways [1]. NO 2 also contributes to the formation of secondary particulate aerosols and tropospheric ozone (O 3 ) in the atmosphere-both are significant air pollutants due to their adverse effects on human health [2]. Thus, NO 2 monitoring is important in the automotive and the energetic industries [3,4]. The nitro compounds are also components of vapors of explosive materials, such as trinitrotoluene, Research Department Explosive, and nitroglycerine [5][6][7].
ft-dodec-1-en]), with different P3HT chain lengths, were tested as gas receptor thin films in resistance NO2 sensors. This paper presents material synthesis and characterization, a sensors fabrication process description, and a study of the NO2 sensing properties of the obtained devices. Our research shows that the proper choice of CP chain length and other side-chains in graft copolymers can improve their gas sensing properties. The investigated materials show higher responses to single ppms of NO2 than they do to other P3HT-based materials that are functionalized by other methods. Thus, we conclude that the grafting method appears to be a prospective way of obtaining tailor-made gas sensing materials.

Materials and Methods
The synthesis of the vinyl terminated regioregular poly(3-hexylthiophene) (vin-rrP3HT), via the Grignard Metathesis Method (GRIM) method, and the synthesis of the vinyl terminated rrP3HT, was conducted based on the procedure described in the literature in [29] (all of the materials were from Sigma Aldrich, Saint Louis, MO, USA). The polymers of two average Mn, namely, a 10,000-chloroform fraction (CH) and a 4000-hexane fraction (H)-as determined by the gel permeation chromatography (GPC) calibrated on the polystyrene standard-were prepared. The calculated average of the P3HT repeat units were 60 and 24 for the CH and H fractions, respectively. The investigated graft polymers were obtained using the method described in the patent application [30]. The synthesis was based on the grafting of the vinyl terminated P3HT with a different chain length and PEG, or dodec-1-en onto poly(methylhydrosiloxane) (PMHS) chains (all of the materials were from Sigma Aldrich, Saint Louis, MO, USA). The scheme of the synthesis of the obtained P3HT graft copolymers is presented in Figure 1. The chemical structures of the investigated materials were confirmed by 1 H Nuclear Molecular Resonance ( 1 H-NMR) and Fourier-transform infrared spectroscopy with the attenuated total reflectance (FTIR-Atr). The IR spectroscopy was carried out on a Perkin-Elmer Spectrum-Two (Waltham, MA, USA) spectrometer with a Universal Attenuated Total Reflectance accessory UATR (Single Reflection Diamond) module.
An 1 H-NMR analysis of products was performed for solutions in CDCl3 on a Varian Unity Inova (Palo Alto, CA, USA) spectrometer with a resonance frequency of 300 MHz, using tetramethylsilane (TMS) as the internal standard.
The obtained graft copolymers and polymers, which are the receptor materials, were deposited on the interdigital transducers (IDT) using spin-coating method. The IDTs with gold electrodes on the Si/SiO2 substrates were described in detail in our previous works [11,18]. All of the polymers were dissolved in chloroform (CHCl3) (POCH, Gliwice, Poland) in the proportion of 2.5 mg of polymer to 1 mL of solvent. The solutions were dropped on the rotating transducers in amounts of The chemical structures of the investigated materials were confirmed by 1 H Nuclear Molecular Resonance ( 1 H-NMR) and Fourier-transform infrared spectroscopy with the attenuated total reflectance (FTIR-Atr). The IR spectroscopy was carried out on a Perkin-Elmer Spectrum-Two (Waltham, MA, USA) spectrometer with a Universal Attenuated Total Reflectance accessory UATR (Single Reflection Diamond) module.
An 1 H-NMR analysis of products was performed for solutions in CDCl 3 on a Varian Unity Inova (Palo Alto, CA, USA) spectrometer with a resonance frequency of 300 MHz, using tetramethylsilane (TMS) as the internal standard.
The obtained graft copolymers and polymers, which are the receptor materials, were deposited on the interdigital transducers (IDT) using spin-coating method. The IDTs with gold electrodes on the Si/SiO 2 substrates were described in detail in our previous works [11,18]. All of the polymers were dissolved in chloroform (CHCl 3 ) (POCH, Gliwice, Poland) in the proportion of 2.5 mg of polymer to 1 mL of solvent. The solutions were dropped on the rotating transducers in amounts of 25 µL. During all of the coating processes, the spin rate of the IDTs were kept on a constant level of 500 rpm. Further details about the spin-coating process can be found in our previous work [18], where different graft copolymers and pure P3HT that were obtained by the same technological process, were tested as NO 2 receptor materials.
The obtained sensing films were characterized using the atomic force microscopy (AFM) technique using the NTGRA Prima system (NT-MDT, Moscow, Russia) with a semi-contact mode and using HA-HR probes (NT-MDT) with a 260 kHz work frequency. The typical curative radius of the tip of the used probes is less than 10 nm, the tip length is ≤1 µm, and the cantilever length is 123 µm. All of the measurements were performed using a scanning frequency of 0.5 Hz (scan speed for 5 × 5 µm areas was approximately 5 µm/s). The AFM data were processed and analyzed using the dedicated software Nova 1.1.0.1824 (NT-MDT, Moscow, Russia).
The sensors were placed on a thick film heater on an alumina substrate and were electrically connected with chip feedthroughs using an ultrasonic wire bonding method with 25 µm of gold wire. The temperature was controlled using a SR94 controller (Shimaden, Tokyo, Japan) with a Pt100 temperature sensor. Further details concerning the sensors' electrical connections, heating, and temperature control can be found in our previous work [11]. The four sensors (each with a different graft copolymer) were tested simultaneously using a testing chamber and mass flow controllers that were based on a measurement system that was also presented in detail in our previous work [11]. The sensors' resistance was measured using a multi-switch unit 34970A (Agilent, Santa Clara, CA, USA) within a 10 MΩ range.
The sensors' reaction to the small concentrations of NO 2 within the range of 1 ppm to 20 ppm, in different operating temperatures (RT = 25 • C, 50 • C, and 100 • C), were tested. The gas mixture was prepared from a calibration mixture (100 ppm NO 2 in N 2 ) and a carrier gas (pure N 2 ), using gas a dosing system that was based on the mass flow controllers. The research was conducted in a nitrogen atmosphere in order to describe the interactions of the polymeric materials with NO 2, without the interference of oxygen and other reactive gases. During all of the experiments, the relative humidity (RH) of the gas mixture was kept at a constant level of 6 ± 1%.
The sensors' responses were calculated according to the following formula: where R a is a base resistance in a pure carrier gas and R g is a resistance in a target gas.

Materials Characterization
Both FTIR and 1 H-NMR were used in order to identify and characterize the four copolymers. The assignments of the chemical shifts are presented in Table 1. The NMR has allowed for the compositional and structural determination of each copolymer. The 1 H-NMR spectra of the copolymer DodecSil CH and PEGSil CH, presented in Figure A1 (Appendix A), show the representative spectra for all of the copolymers that were used used (both with P3HT chain lengths).  In Figure A2, the FTIR-Atr representative spectra of graft copolymers are presented. Detailed analyses of the IR signals are presented in Table 2. The spectra of the grafted copolymers, not only shows all of the characteristic peaks of the siloxane groups of Si-O-Si at 1090 cm −1 and at 1020 cm −1 and those of P3HT, but also shows the absorption bands of poly(ethylene glycol), such as CH 2 and CH 3 , bending at 1467.8 cm −1 and at 1342.4 cm −1 C-O-C. This indicates that the P3HT, poly(ethylene glycol), and dodecyl were successfully grafted onto the polysiloxane backbone. Table 2. Assignments of the main Fourier-transform infrared spectroscopy with attenuated total reflectance (FTIR-Atr) features of DodecSil CH and PEGSil CH. The morphology of all of the prepared polymer sensing films was investigated using the AFM method. These measurements showed that all of the films were homogenous and smooth, with root mean squares (RMS) smaller than 2 nm (Table 3). Since all of the results were similar, the representative AFM height of the images (of 5 × 5 µm areas between the IDT electrodes) has been presented in Figure 2 (the complementary phase images are presented in Figure A3). The thickness of all of the films were measured using the AFM method and found to be approximately 20 ± 5 nm. The base resistance of the sensing structures measured at RT in pure N 2 are presented in Table 3. In Figure A2, the FTIR-Atr representative spectra of graft copolymers are presented. Detailed analyses of the IR signals are presented in Table 2. The spectra of the grafted copolymers, not only shows all of the characteristic peaks of the siloxane groups of Si-O-Si at 1090 cm −1 and at 1020 cm −1 and those of P3HT, but also shows the absorption bands of poly(ethylene glycol), such as CH2 and CH3, bending at 1467.8 cm −1 and at 1342.4 cm −1 C-O-C. This indicates that the P3HT, poly(ethylene glycol), and dodecyl were successfully grafted onto the polysiloxane backbone. Table 2. Assignments of the main Fourier-transform infrared spectroscopy with attenuated total reflectance (FTIR-Atr) features of DodecSil CH and PEGSil CH. The morphology of all of the prepared polymer sensing films was investigated using the AFM method. These measurements showed that all of the films were homogenous and smooth, with root mean squares (RMS) smaller than 2 nm (Table 3). Since all of the results were similar, the representative AFM height of the images (of 5 × 5 μm areas between the IDT electrodes) has been presented in Figure 2 (the complementary phase images are presented in Figure A3). The thickness of all of the films were measured using the AFM method and found to be approximately 20 ± 5 nm. The base resistance of the sensing structures measured at RT in pure N2 are presented in Table 3.

Gas Sensing Mesurements
The conductivity of the studied graft copolymers is provided by the presence of the CP (P3HT) chains. The application of the polymethylsiloxane backbone is responsible for the positive film forming process and adhesive properties. Additionally, siloxane has good gas permeability values and, as a result it does not significantly interfere with the gas diffusion processes in sensing films. The other side-chains, PEG or dodec-1-en, were applied to improve the mechanical properties of the investigated graft copolymers. PEG is a well-known polyelectrolyte, which increases gas and charge transfer in the sensing films. PEG has positive NO x adsorption properties [31]. Thus, it is added to our grafts in order to improve the material sensing properties, while dodec-1-en is only the internal plasticizer and should not show additional interaction with the analyte. Thus, the addition of this comonomer will improve the mechanical properties of the sensing material.
In all cases, NO 2 causes a decrease of the sensors' resistance (this is a response increase) as it is a typical p-type semiconductor reaction to the oxidizing gas action. The sensing mechanism is based on the P3HT doping process, where NO 2 is an electron acceptor dopant [32,33]. The reaction of P3HT with NO 2 leads to an increase in the holes of concentration, because of the formation of polarons followed by bipolaron formation [22,34].
For low concentrations of NO 2 (lower than 10 ppm) at RT, polarons are generated, causing the resistance to drop, seen in Figure 3a. However, all of the investigated materials stay in their semiconducting states. After some time at higher NO 2 concentrations, the polaron concentration becomes sufficient for bipolaron generation to occur. This can be seen in Figure 3a, for 10 and 20 ppm peaks, where a very fast resistance drop (response increase) to small values (even hundreds of ohms) is observed. Consequently, polymers switched to conducting states. Furthermore, the differences between the time and concentration of where bipolaron generation starts shows that CP chain length has an influence on this process [35,36]. P3HT H has a shorter effective conjugation length and prefers to generate stable polarons, in contrast to longer P3HT CH chain, where the generation of bipolarons is more probable. This leads us to conclude that, for NO 2 sensing applications, the semiconducting state of P3HT is required for proper NO 2 concentration measurement, and its conducting state is an upper limit for the proper operation sensors. However, it has to be stressed that the adsorption processes are reversible in all cases, after the NO 2 removal sensor resistance returns to its initial value. Unfortunately, at RT, the regeneration process is slow (Figure 3a) and an operating temperature elevation is required to improve sensor dynamics.
At higher temperatures (50 • C and 100 • C), a switch to the conducting state was not observed in measured the concentration range, as seen in Figure 3b,c. Thermal activation causes faster NO 2 desorption/polarons recombination. Consequently the bipolarons were not generated, the sensor operated faster, and regeneration was more efficient.
The significant differences in sensor responses were observed between graft copolymers with PEG and those with dodec-1-en side-chains. According to our prediction, PEG improved sensor responses due to the abovementioned interaction with NO 2 . The sensor responses to 5 ppm of NO 2 at 50 • C, shown in Figure 4a, and to 1 ppm of NO 2 at different temperatures, shown in Figure 4b, are collected and shown in Figure 4 for an easy comparison. It is evident that, in all cases, PEGSils' responses were higher than those of DodecSils. Furthermore, in all cases, H fractions exhibited higher responses to NO 2 than CH fractions did. The highest responses were observed at the temperature of  The sensors' scalability at 100 °C was confirmed using the calibration curves presented in Figure 5. In this case, the gas sensors demonstrate the logarithmic dependence between the response and the concentration. In order to show the sensors' potential for low concentration detection, the calibration curves are presented in a semi-logarithmic scale with logarithmic approximations. Figure 5 shows that the proposed sensors have great potential for sub-ppm and ppb levels of NO2 concentration measurements.  The sensors' scalability at 100 °C was confirmed using the calibration curves presented in Figure 5. In this case, the gas sensors demonstrate the logarithmic dependence between the response and the concentration. In order to show the sensors' potential for low concentration detection, the calibration curves are presented in a semi-logarithmic scale with logarithmic approximations. Figure 5 shows that the proposed sensors have great potential for sub-ppm and ppb levels of NO2 concentration measurements. The sensors' scalability at 100 • C was confirmed using the calibration curves presented in Figure 5. In this case, the gas sensors demonstrate the logarithmic dependence between the response and the concentration. In order to show the sensors' potential for low concentration detection, the calibration curves are presented in a semi-logarithmic scale with logarithmic approximations. Figure 5 shows that the proposed sensors have great potential for sub-ppm and ppb levels of NO 2 concentration measurements.  Table 4 presents a comparison of the responses to NO2 of PEGSil H and other P3HT-based receptor materials that were recently investigated in the literature. This summary shows that the graft copolymers investigated in our current and previous work [18] show higher responses to NO2 than clean P3HT and P3HT that are functionalized using other methods. In this case, the responses to 1 ppm of NO2, at all measured temperatures were significantly higher than the responses of the other compared materials' responses to 4 ppm or 5 ppm of NO2 (Table 4). This confirms that the investigated materials have a greater potential for the measuring of NO2 in sub-ppm and ppb concentration levels than other P3HT-based materials. Thus, we can conclude that our strategy for improving the NO2 sensing properties of CPs, using the grafting method, is a promising advance in the research of tailor-made functionalized receptor materials. The obtained results show that the responses of the proposed graft copolymers are comparable with MOX based sensors and can be considered as an alternative to these sensors for NO2 sensing. The advantages of these materials include a relatively low operating temperature and the numerous technologies available for the application on transducers, (such as spin-coating, printing, drop-coating, and dip-coating methods). However, long-term stability and selectivity of such materials still require further investigation. * Response calculated as (Ra − Rg)/Ra or (Ia − Ig)/Ia.  Table 4 presents a comparison of the responses to NO 2 of PEGSil H and other P3HT-based receptor materials that were recently investigated in the literature. This summary shows that the graft copolymers investigated in our current and previous work [18] show higher responses to NO 2 than clean P3HT and P3HT that are functionalized using other methods. In this case, the responses to 1 ppm of NO 2, at all measured temperatures were significantly higher than the responses of the other compared materials' responses to 4 ppm or 5 ppm of NO 2 (Table 4). This confirms that the investigated materials have a greater potential for the measuring of NO 2 in sub-ppm and ppb concentration levels than other P3HT-based materials. Thus, we can conclude that our strategy for improving the NO 2 sensing properties of CPs, using the grafting method, is a promising advance in the research of tailor-made functionalized receptor materials. The obtained results show that the responses of the proposed graft copolymers are comparable with MOX based sensors and can be considered as an alternative to these sensors for NO 2 sensing. The advantages of these materials include a relatively low operating temperature and the numerous technologies available for the application on transducers, (such as spin-coating, printing, drop-coating, and dip-coating methods). However, long-term stability and selectivity of such materials still require further investigation. P3HT/ZnO NS-NR composite 4 ppm 60% * RT [38] RGO-P3HT composite 4 ppm 40% * RT [39] P3HT:ZnO (ratio6:2) hybrid 5 ppm 20% * RT [40] P3HT-SnO 2 composite 5 ppm 500% 100 • C [41] Poly-(bisdodecylquaterthiophene) 5 ppm 300% * RT [26] poly-(bisdodecylthioquaterthiophene) 5 ppm 450% * RT [26] P3HT:ZnO-nanowire composite 4 ppm 30% * RT [42] * Response calculated as (R a − R g )/R a or (I a − I g )/I a .

Conclusions
In summary, this work presents the results of research on new graft copolymers that contain the semiconductor rrP3HT chain in their structures. The aim of the work was to examine the sensor's response to NO 2, in order to check whether the addition of rrP3HT to the polymethylhydrosiloxane chain, and the additional grafting of various types of comonomers, would affect the rrP3HT sensor properties. Our approach shows that the proper selection of the graft copolymers, and the length of CP chains, improve the sensing properties of the polymer gas receptor material. The investigated materials show higher responses to single ppms of NO 2 than other P3HT-based materials (cleaned or functionalized by other methods). These responses are comparable to many MOX-based sensors and can be considered as an alternative to these sensors for NO 2 sensing. The advantages of the proposed grafted copolymers-based sensors include their ability to operate at a relatively low temperature and the numerous technologies available for the application of such materials on transducers. The obtained sensing structures show the potential to measure low concentrations at sub-ppm and ppb level, making them useful for many applications. Based on these results found in our research, we presented a novel alternative method of obtaining the tailor-made gas sensing materials, based on CPs.

Conclusions
In summary, this work presents the results of research on new graft copolymers that contain the semiconductor rrP3HT chain in their structures. The aim of the work was to examine the sensor's response to NO2, in order to check whether the addition of rrP3HT to the polymethylhydrosiloxane chain, and the additional grafting of various types of comonomers, would affect the rrP3HT sensor properties. Our approach shows that the proper selection of the graft copolymers, and the length of CP chains, improve the sensing properties of the polymer gas receptor material. The investigated materials show higher responses to single ppms of NO2 than other P3HT-based materials (cleaned or functionalized by other methods). These responses are comparable to many MOX-based sensors and can be considered as an alternative to these sensors for NO2 sensing. The advantages of the proposed grafted copolymers-based sensors include their ability to operate at a relatively low temperature and the numerous technologies available for the application of such materials on transducers. The obtained sensing structures show the potential to measure low concentrations at sub-ppm and ppb level, making them useful for many applications. Based on these results found in our research, we presented a novel alternative method of obtaining the tailor-made gas sensing materials, based on CPs.
Acknowledgments: Publication supported as a part of the Rector's grants in the area of scientific research and development works. Silesian University of Technology, grants numbers: 05/040/RGJ18/0022 and 04/010/RGJ18/0077. The present work was partially sponsored by the Polish National Science Centre "NCN" within the grant 2016/23/B/ST5/03103 and Silesian University of Technology, Faculty of Electrical Engineering within the grant BKM/567/Re4/2017.

Author Contributions: Marcin
Procek designed and carried out the gas sensing experiments, designed the experiment stand, conducted sensor preparation technology processes, participated in material characterization, analyzed the data and participated in preparing the paper. Agnieszka Stolarczyk designed and dealt with the materials syntheses, participated in material characterization, analyzed the data, described sensing mechanisms and participated in preparing the paper. Kinga Kepska participated in materials syntheses and characterization.

Conflicts of Interest:
The authors declare no conflict of interest.   Marked by letters of the hydrogen atoms group assigned to particular observed signal groups visible in the 1 H-NMR spectrum. Polymers' macroparticles schemes are shown in the spectrograms. Figure A2. Fourier-transform infrared spectroscopy with the attenuated total reflectance (FTIR-Atr) spectra of (a) PEGSil CH; (b) DodecSil CH. Figure A2. Fourier-transform infrared spectroscopy with the attenuated total reflectance (FTIR-Atr) spectra of (a) PEGSil CH; (b) DodecSil CH.