A Flexible and Polymer‐Based Chemiresistive CO2 Gas Sensor at Room Temperature

CO2 sensing is important in many applications ranging from air‐quality monitoring to food packaging. In this study, an amine‐functionalized copolymer, poly(N‐[3‐(dimethylamino)propyl]‐methacrylamide‐co‐2‐N‐morpholinoethyl methacrylate) (p(D‐co‐M)) is synthesized, offering moderate basicity suitable for a wide CO2 detection range. Taking advantage of this characteristic of p(D‐co‐M), this polymer is used for designing a chemiresistive, low‐cost, flexible, and reversible CO2 sensor. The p(D‐co‐M)‐based sensors show a noticeable decrease in their direct current resistance and alternating current impedance upon exposure to a wide range of CO2 concentration (1–100%) at room temperature with a response and a recovery time of 6 and 14 min, respectively. Additionally, the p(D‐co‐M)‐based sensors demonstrate a favorable selectivity to CO2 in the presence of interfering gases including methanol, ethanol, toluene, and acetone. Surface potential measurements show an increase of +6.34 V upon exposure to humidity and CO2, indicating the protonation of the polymer's amine sites, facilitating the detection of CO2 in the wet environment. This sensor is efficient for detecting CO2 concentration released during fermentation of kimchi as a food model.


Introduction
Carbon dioxide (CO 2 ) detection plays a crucial role in different fields such as airquality monitoring, [1] fire detection, [2] green-house gas monitoring, [3] health, [4] marine and environmental studies, [5] and food supply chain. [6] Conventionally, different methods including fluorescence, [7] gas chromatography, [8] infrared spectrometry, [9] photoacoustic spectroscopy, [10] and severinghaus type electrode [11] have been used for CO 2 measurement. Although traditional CO 2 detection methods are highly selective and sensitive, they usually suffer from relatively high cost, bulkiness, high power consumption, and vulnerability to electromagnetic interference. [10,12] In addition, these techniques require materials (e.g., metal oxide, glass, silicone) that are not compatible with growing interests in new applications such as food packaging. [13] For these reasons, the demand for flexible, compact, low-cost, low-power, and lightweight CO 2 sensors has increased, recently. Metal oxide gas sensors are the predominant type of gas sensors that are currently on the market. [14] These sensors are based on ceramic materials, including ZnO, [15] SnO 2 , [16] TiO 2 , [17] In 2 O 3 , [18] CuO, [19] CdO, [20] and WO 3 , [21] making them rigid and bulky. In addition, most of them operate at high temperatures (300-700 °C), which is not desirable due to the demand for high voltage (10-100 V) and subsequently short lifetime. [22] Therefore, low-power, flexible, and light CO 2 sensors were developed based on carbon nanotubes (CNTs), [23] aluminium oxide, [24] graphene, [25] and polyethyleneimine (PEI). [26] These materials have been employed by researchers to fabricate CO 2 sensors operating at low temperatures (25-100 °C) and consuming milliwatts of power with a detection range of 10-4000 ppm. [22] However, the sensing feasibility can be enhanced by increasing the detection range to higher concentrations, matching the CO 2 concentration found in food packaging. [6a] In most flexible sensors, a sensing component that is responsive to the analyte is deposited on a flexible substrate. Therefore, selecting the substrate and the sensing material are two key factors for the development of a flexible gas sensor. Paper, polyethylene naphthalate, poly(ethylene terephthalate) (PET), polyimide, poly(dimethylsiloxane), and textile (e.g. cotton and fabric) are the most commonly used substrates for flexible CO 2 sensing is important in many applications ranging from air-quality monitoring to food packaging. In this study, an amine-functionalized copolymer, poly(N-[3-(dimethylamino)propyl]-methacrylamide-co-2-N-morpholinoethyl methacrylate) (p(D-co-M)) is synthesized, offering moderate basicity suitable for a wide CO 2 detection range. Taking advantage of this characteristic of p(D-co-M), this polymer is used for designing a chemiresistive, low-cost, flexible, and reversible CO 2 sensor. The p(D-co-M)-based sensors show a noticeable decrease in their direct current resistance and alternating current impedance upon exposure to a wide range of CO 2 concentration (1-100%) at room temperature with a response and a recovery time of 6 and 14 min, respectively. Additionally, the p(D-co-M)-based sensors demonstrate a favorable selectivity to CO 2 in the presence of interfering gases including methanol, ethanol, toluene, and acetone. Surface potential measurements show an increase of +6.34 V upon exposure to humidity and CO 2 , indicating the protonation of the polymer's amine sites, facilitating the detection of CO 2 in the wet environment. This sensor is efficient for detecting CO 2 concentration released during fermentation of kimchi as a food model.
sensors. [22,27] Paper has been utilized extensively in many flexible electronics since it is one of the cheapest and most readily available substrate materials. [28] Its recyclability, lightweight, and eco-friendly characteristics are the other advantages of the paper substrate. [27,28] Despite its advantages, the paper substrate loses its mechanical integrity and stability in wet conditions. [28] Consequently, the high humidity in the surrounding environment of paper-based sensors leads to a weak affinity between the paper substrate and the sensing material, resulting in their poor performance. [28] PET is another common flexible substrate that has been used in gas sensors. [27,28] It is low-cost, odorless, non-toxic, highly flexible, transparent, and suitable for direct use in food packaging. [27,28] Unlike paper, it offers mechanical strength, resistance to some solvents (insulating properties), and discrete temperature tolerance. [28,29] CO 2 -responsive polymers have received a great deal of attention for developing sensors because of their high flexibility and good sensitivity at room temperature. [27,30] In contact with CO 2 , one or more of the physical attributes of these polymers (e.g., electrical resistance, volume, transparency) changes accordingly. For sensing applications, the variation in physical properties of CO 2 -responsive polymers can be used to measure CO 2 concentration. In the presence of CO 2 , the functional groups in the polymer chains change from neutral to charged state (or vice versa) (Equation (1)). [31] Amidine, [32] guanidines, [33] and amines [34] are the most widely reported functional groups present in the chemical structure of the common CO 2 -responsive polymers. [35] Those containing amine groups-especially tertiary amines-have gained significant interest due to their simple synthesis compared with those possessing amidine and guanidine groups. [35] Compared with amidine and guanidine groups, tertiary amines are a moderate base (pK aH = 6 -7), resulting in good switchability between neutral and charged states (easy to protonate and deprotonate). [34] Although it is known that a higher pK aH results in a higher degree of protonation (DOP), it makes the reversible reaction more difficult. [35] For example, amidine is a weak base (pK aH = 5.4) with a low DOP, [32a] and in contrast, guanidine is a super base (pK aH = 13.5) which is challenging to deprotonate. [33a] Last but not least, primary (Equation (2)) and secondary amines (Equation (3)) have NH bonds that react with CO 2 in water, forming carbamate salt. [36] The carbamate formation is not preferred for sensing applications as it is not reversed unless at high temperatures, which demands extra heat, cost, and time. [ Acrylamide and isooctylacrylate, poly (γ-aminopropylethoxy/ propylethoxysiloxane), polypyrrole, poly(anthranilic acid), ethylenediamine, poly(propylenimine), polystyrene-bound ethylene diamine, poly(N,N-dimethylaminoethyl methacrylate), poly(N,N-diethylaminoethyl methacrylate), and PEI are some examples of amine-functionalized polymers which have been employed for CO 2 detection. [35,38] Sun et al. developed PEI and PEI/starch-based sensors, which were sensitive to the CO 2 concentration within the range of 1000 to 5000 ppm at room temperature. [26] Nevertheless, these sensors are not flexible and have a long response time of 12 to 19 min and a long recovery time of more than 20 min which are problematic for some applications. [26] As another example, Siefker et al. developed a resonant mass sensor based on PEI for the detection of CO 2 . [1a] Although this sensor was cheap, required low-power, highly selective, and sensitive to CO 2 , it was not flexible which is an undesirable property for some applications such as food packaging. [1a] Our Previous study explored the performance of a CO 2 sensor based on poly(N- [3-(dimethylamino)propyl] methacrylamide) (pDMAPMAm) through the proton hopping mechanism. [39] Despite its capability for CO 2 detection at room temperature, the sensor suffered from the non-linear and irreversible response to CO 2 at different concentrations, due to the saturation of amine sites, which hindered its potential as a reversible sensor for accurate CO 2 measurements. [39] The drawback of saturation of amine sites can be resolved by adjusting the basicity of pDMAPMAm. Therefore, adding a less basic monomer such as 2-N-morpholinoethyl methacrylate (MEMA) (pK aH of homopolymer = 4.9) [31,40] would provide more free amine sites (especially at lower pH). MEMA is commercially available as an inexpensive monomer for CO 2 detection. The advantage of using MEMA to adjust the basicity of a superabsorbent polymer hydrogel for enhanced CO 2 capturing has been recently shown by Jansen-van Vuuren et al. [41] This work aims to develop a polymer-based CO 2 sensor to overcome the drawbacks of the existing CO 2 sensing devices, including inflexibility, high operating temperature, and limited detection range. Herein, we synthesized a novel chemiresistor material composed of DMAPMAm and MEMA monomers by free radical polymerization. The introduced chemiresistor was used as a CO 2 -responsive component for the fabrication of a CO 2 sensor. The electrical responses of the fabricated CO 2 sensing device were studied at the range of 1-100% using direct current (DC) resistance and alternating current (AC) impedance measurements. In addition, the reversibility, response/recovery time, and selectivity of the fabricated sensor were explored. Ultimately, the successful application of the fabricated sensor for in situ CO 2 monitoring in a model food packaging was demonstrated.

Synthesis and Characterization of p(D-co-M) Copolymer
The chemical structure of the p(D-co-M) copolymer is shown in Figure 1a. The results of proton nuclear magnetic resonance ( 1 H-NMR) analysis acquired with CDCl 3 in Figure 1b confirmed the chemical bonding between DMAPMAm and MEMA monomers and the formation of p(D-co-M) by conducting free radical polymerization at 60 °C overnight in ethanol as a solvent. The characteristic peaks of pDMAPMAm were detected at a) 0.99-1.10 ppm, d) 1.66 ppm, b) 1.95 ppm, f) 2.27, e) 2.40 ppm, and c) 3.20 ppm. [42] The characteristic peaks of pMEMA were detected at g) 0.9-1.10, h) 1.75 ppm, j) 2.52, k) 2.64 ppm, l) 3.73 ppm, and i) 4.10 ppm. [40,43] Both pDMAPMAm and pMEMA signals can be observed in the p(D-co-M) spectra, which confirms that the polymerization has been successful. The p(D-co-M) composition ratio was roughly estimated using the ratio between the methyl protons from the DMAPMAm segment at f) 2.27 ppm and the methylene protons from the MEMA segment at i) 4.10 ppm. The ratio of DMAPMAm to MEMA segments in p(D-co-M) was 5. The number of repeating units of DMAPMAm and MEMA was determined using this ratio in combination with the Mn from size exclusion chromatography (SEC), as shown in Table 1.
To further confirm the successful polymerization of p(D-co-M), Fourier-transform infrared spectroscopy (FTIR) was carried out. Figure 1c shows the FTIR spectra of pDMAPMAm, p(D-co-M), and pMEMA. The characteristic absorption bands at 1113 cm −1 and 1723 cm −1 were attributed to COC and CO, respectively, in the structure of the pMEMA polymer. [30a] Similarly, the characteristic peaks of pDMAPMAm polymer such as NH (amide II), CO (amide I), and NH, appeared at 1525 cm −1 , 1629 cm −1 , and 3330 cm −1 , respectively. [30b,42] The observation of the distinctive peaks from DMAPMAm and MEMA in the FTIR spectra of p(D-co-M) confirms the existence of both monomers within the copolymer structure.  Since the basicity of the polymer has a significant effect on its response to CO 2 , [37a] it is important to know the pK aH values of the synthesized copolymer. Therefore, titration with 1 m hydrochloric acid (HCl) was performed on dilute aqueous solutions of polymer (1 wt% in deionized water) to measure the pK aH values of p(D-co-M). As shown in Figure 1d, the pK aH values of p(D-co-M) copolymer were 7.3 and 4.4 which were attributed to the DMAPMAm and MEMA chains, respectively. Of note, these pK aH values were within the range obtained by titration of each homopolymer separately, 7.9 for pDMAPMAm and 4.1 for pMEMA. These observations confirmed the existence of both amine side groups with different     pK aH values within the p(D-co-M) structure which is advantageous because of providing more free amine sites over a wide range of pH.

CO 2 Sensing Mechanism Model
As shown above, our newly synthesized p(D-co-M) polymer possessed a broad pK aH range (7.3 and 4.4) and suitable basic properties which are essential for efficient CO 2 responsitivity. Therefore, we assessed the feasibility of using p(D-co-M) for designing a real-time CO 2 sensor. The fabricated p(D-co-M) sensor and the sensing mechanism model are sketched in Figure 2. The interaction between the polymer chains and CO 2 is based on acid-base chemistry shown in Equation (1). In the presence of water, the amine sites of the polymer react with the carbonic acid generated by CO 2 , forming bicarbonate salt. [37a] This reaction is reversible when CO 2 is removed from the system. Therefore, the CO 2 -responsive polymer can alternate between the neutral and charged states depending on the presence or absence of CO 2 . [37a] This switchable behavior in the charge state of amine side groups is translated to corresponding changes in the electrical resistance of the polymer. It is noteworthy to mention that the pH of a solid or incompletely dissolved polymer is unknown. [41] However, here, we hypothesized that the carbonic acid generation alters the pH of the polymer to more acidic values. As shown in Figure 1d, p(D-co-M) has two different pK aH values, therefore, its molecular composition depends on the pH of the system. The possible protonation states of the copolymer in various pH values are presented in Figure 2. Before exposure to CO 2 , the polymer is basic and both amine sites are neutral. However, by introducing CO 2 to the humidified system, the pH decreases due to carbonic acid production. Once pH drops below 7. 3  protonated, forming -N(CH 2 ) 2 + . With further reduction in pH, both amine sites are largely protonated.

Fabrication and Evaluation of the p(D-co-M)-Based CO 2 Sensors
In order to design a chemiresistive CO 2 sensor utilizing the CO 2 -responsive p(D-co-M) polymer, it is necessary to determine the optimum polymer concentration to accurately and rapidly measure the concentration of CO 2 with high sensitivity within the range of 1 and 100%. To this end, various concentrations of p(D-co-M) including 1, 10, 40, and 70 wt% solutions were deposited between the carbon black (CB) electrodes which were printed on PET substrates (Figure 2). The sensors were then exposed to 3% CO 2 and their DC electrical resistance was measured as the sensor response. The results in Figure 3 present the sensors' response and their half-response time. Clearly, higher ink concentrations led to weaker responses to CO 2 . The strongest response (highest sensitivity) was found to be 67% which is attributed to the 1 wt% ink. In terms of response time, 1 wt% ink also revealed the shortest half-response time of 2.6 min. Therefore, 1 wt% ink was selected as the optimum concentration for the fabrication of p(D-co-M) sensors hereafter.
It is anticipated that CO 2 -responsive polymers that possess tertiary amine groups cannot interact with CO 2 in their dried form, and the presence of water is pivotal for polymer protonation. Since the DOP and the resistance depend on the water content and the humidity level, it is important to measure the effect of this parameter on the CO 2 response for p(D-co-M) polymers. Therefore, the DC resistance of the fabricated p(D-co-M) sensor was measured as a function of relative humidity (RH) as shown in Figure 4. By increasing the humidity, the relative resistance of the p(D-co-M) sensor decreased due to the absorption of water, which facilitated proton conduction. [30d] The largest reduction in resistance occurred at lower humidity levels, which reached 0.05 at 70%. For the higher humidity levels, the resistance decreased, albeit not as much as in the first section, until it plateaued for values above 76%. Thus, to eliminate the impact of humidity on the response of the sensor, all the experiments were conducted at RH ≥ 90%. Considering that the RH in the environment inside food packaging is ≈100%, working in high humidity levels is one of the key advantages of the p(D-co-M) sensors.
In order to investigate the electrical response of the p(D-co-M) sensor to CO 2 , it was exposed to CO 2 at concentrations ranging from 1 to 100%, and its DC resistance and AC impedance were monitored. The results in Figure 5a demonstrate that the relative resistance of the p(D-co-M) sensor significantly dropped from 1 to 0.12 by increasing the CO 2 concentration inside the sensing chamber from 1 to 20%, which was attributed to the protonation of the amine groups. A reduction in the relative resistance was still observed for concentrations above 20%, before reaching 0.06 at 100% CO 2 . Figure 5b shows the impedance profile of the sensor at 1 kHz. A decrease in the relative impedance of the sensor from 1 to 0.6 upon exposure to higher CO 2 concentrations was consistent with the observed decrease in the relative resistance. The results in Figure 5 also underline that p(D-co-M) is a suitable material for designing CO 2 sensors, which are compatible with both AC and DC measurements. The systems based on AC can enable the fabrication of wireless sensors while those based on DC lead to simpler electronics.
To further investigate the surface electrical properties and surface charge of p(D-co-M) in response to CO 2 , the surface potential was measured by Kelvin probe force microscopy (KPFM) (Figure 6a,b, and Figure S1, Supporting Information). The contact potential difference between the tip and the polymer surface was −6.3 V and 0.04 V before and after exposure to CO 2, respectively. Therefore, the surface potential was increased by + 6.34 V as a result of exposure to CO 2 . The reason for this observation is the protonation of p(D-co-M) tertiary amine sites, which accumulates positive charges on the polymer's surface and increases its potential. [34b,37a] To determine the surface morphology and roughness of the p(D-co-M) film before and after exposure to CO 2 , atomic force microscopy (AFM) measurement was used. The results in Figure 6c-f demonstrate that the surface roughness of p(D-co-M) film decreased from 1327 to 684 nm after exposure to humidity and CO 2 .
The reversibility of the response of the p(D-co-M) sensor to CO 2 was monitored for six cycles of alternate exposure to humidified CO 2 and humidified N 2 at room temperature. As shown in Figure 7a, the relative resistance of the sensor rapidly declined from 1 to 0.06 in 6 min upon exposure to CO 2 , while it leveled off after removing CO 2 from the chamber. By switching Adv. Mater. Technol. 2023, 8,   the gas to N 2 , the resistance increased to its initial value within 14 min. The longer recovery time of 14 min compared with the response time of 6 min indicates that the CO 2 desorption is slower than its adsorption on tertiary amine sites. The successive CO 2 and N 2 bubbling for six cycles showed the reproducibility of the p(D-co-M) sensor (Figure 7b).
To investigate the effect of interfering gases on the response of the p(D-co-M) to CO 2 , the sensor was exposed to a broad range of functional chemical groups (i.e., alcohol, ketone, aromatic hydrocarbon, and amine) typically found in food packaging environments. Therefore, the selectivity tests were performed with a variety of confounding species including ethanol, methanol, acetone, toluene, and ammonia vapors. It is worth noting that the presence of ethanol, methanol, acetone, and toluene in food packaging environments is attributed to the use of these solvents in multilayer packaging materials or printing and laminating processes, which can change the taste or cause unexpected health problems in humans. [44] The selectivity testing was carried out with two different experimental setups.
In the first method, 50 µL of each selected volatile organic compound was injected into the system (5 min between each injection) at a constant background of 100% CO 2 (Figure 8a). In the second method, the sensor was first exposed to the individual interfering gases and then to CO 2 with a constant background of that interferer gas (Figure 8b-f). Testing under these two conditions allows for determining any effects on the response of the sensor when both CO 2 and interferer are present. The sensor revealed a distinguishable response to CO 2 in the presence of all the interferers tested. However, there is an apparent cross-sensitivity for ammonia, as can be seen in Figure 8a,f. The reason for this observation is the high solubility of ammonia in water which results in its high reactivity. [30e] Based on similar phenomena, Firat et al. developed a paper-based electrical gas sensor that was tested with different water-soluble gases. [30e] Their sensor showed a low sensitivity to CO 2 in contrast to a very high sensitivity to ammonia because of their different solubilities in water. They reported that the solubility of ammonia in water was ≈2000 times greater than CO 2 .  An experiment was set up to assess the feasibility of using the designed sensor for the detection of food spoilage or CO 2 release during the fermentation of live bacteria. To this end, a model system was used in which we measured the CO 2 released during the fermentation of kimchi. Kimchi contains live bacteria and produces CO 2 at room temperature due to continuous lactic acid fermentation. [45] Therefore, it is a good candidate to test the developed CO 2 sensor. To test the performance of our sensor in this real-life scenario, the p(D-co-M) sensor was mounted in a food container (1 L) with 200 g kimchi. A commercial reference CO 2 sensor and a RH-temperature sensor were also added to the setup to monitor the CO 2 concentration, humidity, and temperature during the experiment (Figure 9a). As can be seen in Figure 9b, by fermentation of kimchi, the concentration of CO 2 inside the container increased over time, and consequently, the relative resistance of the sensor fell gradually. Of note, the humidity and temperature changes inside the sensing chamber were negligible during the test (Figure 9c).
The key parameters of the p(D-co-M) sensor and other chemiresistive sensors reported in the literature are listed and compared in Table 2. Compared with the listed studies, the CO 2 sensor developed in this work was easy to fabricate and had a wide range of detection (1-100%). As shown in Table 2, most of the reported sensors were inflexible and worked at high temperatures. However, the sensor developed in this study is flexible and operates at room temperature. Furthermore, the p(D-co-M) sensor was reversible with the response and recovery times of 6 and 14 min, respectively. Altogether, the fabricated sensor offered superior features that overcome the challenges that exist in current CO 2 sensing devices. Therefore it is suitable for many applications including fermentation and brewing, [46] beverage industries, [47] and food packaging. [13]

Conclusions
A chemiresistive sensor was developed based on a novel CO 2responsive polymer, p(D-co-M), via free radical polymerization at 60 °C. The p(D-co-M) sensors exhibited high sensitivity to a wide range of CO 2 concentrations (1-100%) at room temperature with a response and a recovery time of 6 and 14 min, respectively. Both DC resistance and AC impedance decreased upon exposure to CO 2, which is attributed to the protonation of amine groups in the p(D-co-M) chains due to the interaction between the polymer and carbonic acid produced by CO 2 in a humidified environment. Surface potential measurements showed an increase of +6.34 V upon exposure to humidity and CO 2 , indicating the protonation of the polymer's amine sites. The fabricated sensor also rendered high reproducibility and selective response to CO 2 in the presence of other interferers. Finally, the proposed sensing device presented a remarkable response when exposed to CO 2 released by kimchi as a food model, representing its successful application for in situ food monitoring.
Compared to other amine-functionalized polymers used for CO 2 detection, this polymer exhibited a broad pK aH range with moderate basicity which is beneficial for CO 2 sensing and reversibility of sensors. The p(D-co-M)-based sensors developed here offer superior properties for the fabrication of flexible, low-cost, highly sensitive, reversible, and selective CO 2 sensors in comparison with current commercial CO 2 sensors. In addition, the newly developed sensors can operate at elevated humidity conditions and in the presence of other gases for CO 2 sensing. Overall, the advantages and superiority of the proposed polymer may promote the fabricated sensors as a strong candidate for CO 2 sensing compatible with both DC and AC measurements.  PEI-functionalized CNTs [48] 200-1000 600 600 Room temperature Yes Yes PEDOT/PSS, graphene [49] 400-4200 NA NA 50 and 60 No Yes ZnO nanoflakes [50] 200-1025 9-17 9-17 250 Yes Yes PEI-PANI [51] 50-5000 440 ± 313 601 ± 292 Room temperature Yes No SPES/e-MWCNTs [52] 500-5000 480-840 1620 Room temperature Yes No PES/e-MWCNTs [52] 500-5000 600-960 1620 Room temperature Yes No PEI/PEG [53] 200-250 000 251 300 Room temperature Yes No Y-HEC [54] Figure 10. Gas sensing measurement setup. CO 2 and N 2 gas flow rates into the chamber were controlled by installing two mass flow controllers. The sensing chamber includes the p(D-co-M) sensor, water, and a reference CO 2 sensor with an upper limit of 100% to record the CO 2 concentration. The measurement unit includes a precise source meter and a potentiostat for continuously recording the DC resistance and AC impedance of the sensor.
Synthesis of p(D-co-M): DMAPMAm (10.87 mL, 0.06 mol) and MEMA (11.44 mL, 0.06 mol) with a molar ratio of 1:1 were mixed with AIBN (2.8 mL, 0.00055 mol) and ethanol (10 mL, 0.17 mol) as the initiator and solvent, respectively. The prepared mixtures were purged by N 2 (45 mL min −1 ) for 20 min to eliminate any trapped oxygen. After degassing, they were incubated at 60 °C overnight to complete the free radical polymerization under sealed conditions. To evaporate the remaining ethanol, the synthesized copolymers were transferred to glass Petri dishes and were then kept at 60 °C for 2 h before being stored for further usage at room temperature.
Polymer Characterization: 1 H-NMR spectroscopy was performed on a Bruker 800 MHz spectrometer using CDCl 3 as the solvent and 5 mm NMR tubes at room temperature. SEC measurement was performed on a UFLC Shimadzu Prominence SEC system with two Phenogel columns (5 µm, 104 Å, and 105 Å) running in DMAc using BHT/LiBr at 0.05 wt% as eluent at a flow rate of 1 mL min −1 at 50 °C. FTIR was carried out on a Thermo Scientific Nicolet 6700 spectrometer fitted with an ATR accessory (diamond crystal) at an angle of incidence of 90°. The data were collected at a resolution of 4 cm −1 over the range of 450−4000 cm −1 from the average of 32 scans.
Sensor Fabrication: To fabricate the p(D-co-M)-based sensors, a configuration of two CB electrodes (1 mm apart) was first printed on the PET substrates using a 3D-Bioplotter (EnvisionTEC, Germany). The CB ink was extruded through a 250 µm nozzle, and the printing parameters, including temperature, pressure, and speed of printing, were optimized at 21 °C, 3.2 bar, and 10 mm s −1 . After printing, the CB electrodes were dried at 37 °C for 6 h for further usage. Then, 50 µL of the copolymer solution (1 wt% in ethanol) was drop-casted along with the CB electrodes and was dried for the sensing measurements.
Sensor Evaluation: To evaluate the response of the fabricated sensors to CO 2 , a gas-sensing measurement setup was designed (Figure 10). The prepared sensors were mounted in a measurement chamber with inlet and outlet ports for the target gas (either N 2 or CO 2 or a mixture of them) to flow inside the chamber. All measurements were carried out in a humified environment (RH ≥ 90%). For this purpose, the measurement chamber containing water was maintained inside the oven at 37 °C overnight before each experiment. In addition, the gas flow was always humidified by passing through the water. CO 2 and N 2 gas flow rates into the chamber were controlled by installing two mass flow controllers (FMA-2600A Upstream Valve, OMEGA, USA). A reference CO 2 sensor (ExplorIR-W-100, CO 2 METER, USA) with an upper limit of 100% was used to record the CO 2 concentration inside the measurement chamber during the sensing experiments.
To investigate the CO 2 sensing of p(D-co-M) copolymers, the fabricated sensors were exposed to different CO 2 concentrations (1-100%) under high humidity levels (RH ≥ 90%) at room temperature. The DC resistance was continuously measured with a precise source meter (KEITHLY 2450, Textronix, USA). The response of the sensor was expressed as R/R 0 (relative resistance), where R 0 is the initial resistance and R is the resistance of the sensors during the experiment. The AC impedance was evaluated using a potentiostat (ZIVE SP1, WonATech, South Korea). The absolute impedance of the sensors (Z) was measured at a constant frequency of 1 kHz, and the response of the fabricated sensor was expressed as Z/Z 0 (relative impedance) where Z 0 is the initial and Z is the impedance of the sensors during the experiment. AFM and KPFM using a Pt-Ir coated tip were carried out on a Bruker Dimension ICON SPM to explore the surface morphology, roughness, and potential of the p(D-co-M) film before and after exposure to CO 2 .

Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.