The Room Temperature Highly Sensitive Ammonia Gas Sensor Based on Polyaniline and Nitrogen-Doped Graphene Quantum Dot-Coated Hollow Indium Oxide Nanofiber Composite

Hollow indium trioxide (In2O3) nanofibers fabricated via an effectively combined method of electrospinning and high-temperature calcination were coated with nitrogen-doped graphene quantum dots (N-GQDs) prepared by a hydrothermal process through electrostatic interaction. The N-GQD-coated hollow In2O3 nanofibers served as a core for the synthesis of polyaniline (PANI)/N-GQD/hollow In2O3 nanofiber ternary composites using in situ chemical oxidative polymerization. The chemical structure and morphology of the fabricated ternary composites were characterized using Fourier transform infrared, field-emission scanning electron microscopy, and transmission electron microscopy. The gas-sensing performances of the ternary composites were estimated by a homemade dynamic test system which was supplied with a real-time resistance acquisition platform at room temperature. The response value of the PANI/N-GQD/hollow In2O3 nanofiber sensor with a loading of 20 wt% N-GQD-coated hollow In2O3 nanofiber and an exposure of 1 ppm NH3 was 15.2, which was approximately more than 4.4 times higher than that of the PANI sensor. This ternary composite sensor was proved to be very sensitive in the detection of NH3 at a range of concentration between 0.6 ppm and 2.0 ppm at room temperature, which is crucial in the detection of hepatic or kidney disease in human breath. The PANI/N-GQD/hollow In2O3 nanofiber sensor also revealed higher selectivity and repeatability when exposed to 1.0 and 2.0 ppm NH3 at room temperature. Because of the excellent selectivity and repeatability in the detection of 1.0 and 2.0 ppm NH3 at room temperature achieved in this study, it is considered that the PANI/N-GQD/hollow In2O3 nanofiber composite sensor will be a favored gas-sensing material applied on human breath for the detection of hepatic or kidney disease.


Introduction
Human breath comprises various amounts of gas compounds in a range of concentration from a few ppt to thousands of ppm and a lot of humidity [1][2][3]. A composition change significantly depends on many issues, for example, age, gender, and health condition. A breakdown product of proteins, ammonia (NH 3 ), normally converted into urea via the liver and evacuated by the kidneys, reveals specific relevance to hepatic or kidney disease [3][4][5][6][7], and when either of these two organs collapses, the concentration of NH 3 increases from many hundred ppb when healthy up to several ppm. In accordance with a previous report, a concentration of NH 3 less than 1.1 ppm is considered to be healthy, whereas greater than 1.6 ppm is regarded as unhealthy [8]. Since the concentrations of the borderline between healthy and unhealthy are unsure, an intermediate concentration range is identified for clarity. Hence, NH 3 gas sensors are utilized to monitor the level of gas and sustain it within a specific limit.

Preparation of Graphene Nanoribbon and Indium Trioxide
The N-GQDs, prepared by a hydrothermal process using urea as a nitrogen source, were synthesized according to previous reports [30,34]. First, 0.18 g urea and 0.21 g CA were mixed under stirring at room temperature for 30 min in a 10 mL beaker. Then, the mixture was transferred to a poly(tetrafluoroethylene) reactor and heated at 160 • C for 4 h. The reactants were further treated by adding ethanol to the solution and centrifuging at 5000 rpm for 2 h to obtain the N-GQD specimens. Finally, the product was washed with distilled water (DI-water) and subsequently purified by a dialysis bag for 24 h.
Indium nitrate hydrate was used as an indium source to prepare the hollow In 2 O 3 nanofibers. In a typical fabrication process, 3.5 g PVP and 1.1 g In(NO 3 ) 3 were mixed in 12 mL ethyl alcohol and 10.6 mL DMF, and the solution was stirred to dissolve PVP and In(NO 3 ) 3 completely for 10 h. Then, the solution was filled into a 20 mL syringe including a 0.5 mm diameter metallic needle for electrospinning. A 20 kV high-voltage power was applied to the metallic needle tip, set at a feeding rate of 0.3 mL/h, and the distance between the needle and the collector was about 15 cm. After the 24 h electrospinning process, the obtained composite nanofibers were thermally annealed to fabricate hollow In 2 O 3 nanofibers [29,32]. The thermal annealing process using calcination was operated at 800 • C for 3 h with a heating rate of 5 • C/min.

Synthesis of Polyaniline/Nitrogen-Doped Graphene Quantum Dot/Hollow Indium Trioxide Nanofiber Composites
The polyaniline (PANI)/N-GQD/hollow indium trioxide nanofiber composites were prepared using in situ chemical oxidative polymerization. Scheme 1a shows a schematic illustration of this preparation. In a typical preparation process, a certain weight ratio of N-GQD powders was dispersed in DI-water, mixed with hollow In 2 O 3 nanofibers through electrostatic interaction, and sonicated for 1 h. The reactants were further treated by adding ethanol to the solution and centrifuging at 5000 rpm for 2 h to obtain the N-GQD-coated hollow In 2 O 3 nanofiber specimens. Finally, the product was washed with DI-water and subsequently purified by a dialysis bag for 24 h. The obtained N-GQD/hollow In 2 O 3 nanofiber was added to 50 mL HCl solution and sonicated for 2 h. After that, the aniline monomer was added to the dispersed solution of N-GQD/hollow In 2 O 3 nanofiber and stirred for 1 h. Subsequently, ammonium persulfate dissolved in 20 mL HCl solution was progressively added to the mixed solution of aniline monomer/N-GQD/hollow In 2 O 3 nanofiber and reacted for 3 h at 0 • C. The obtained PANI/N-GQD/hollow In 2 O 3 nanofiber composite was filtered, rinsed several times with methanol and DI-water, and vacuum dried at 60 • C for 24 h. For comparison, the PANI/hollow In 2 O 3 nanofiber composites were synthesized using the same method as for the PANI/N-GQD/hollow In 2 O 3 nanofiber composites. Scheme 1. The schematic illustration of (a) preparation of hollow In2O3 nanofibers, N-GQDs, and PANI/N-GQD/hollow In2O3 nanofiber composites and (b) preparation of the PANI/N-GQD/hollow In2O3 nanofiber sensor.

Analytical Procedures
The molecular structures of the fabricated PANI/N-GQD/hollow In2O3 nanofiber composites were determined by Fourier transform infrared (FTIR), Raman, and wideangle X-ray diffraction (WAXD). FTIR spectra were determined with a PerkinElmer Spectrum One spectrometer (Waltham, Massachusetts, USA) in a range of 400-4000 cm −1 . Raman spectra were obtained on a Renishaw system 1000 (Renishaw Inc., West Dundee, IL, USA) using an argon ion laser performing at 514.5 nm with a CCD detector. WAXD measurement experiments were operated using an X-ray diffractometer (Bruker D8, BRUKER AXS, Inc., Madison, WI, USA) equipped with a Ni-filtered Cu Kα radiation. The WAXD experiments were recorded in the range of 2θ = 1.5°-40° with an increment of 1°/min. The morphology of all samples was classified using field-emission scanning electron microscopy (FESEM) and transmission electron microscopy (TEM). The measurement of FESEM was carried out using a JEOL JSM-6700F field-emission instrument (JEOL Ltd., Tokyo, Japan). All specimens were coated with gold to avoid charging. The measurement of TEM was carried out using JEOL JEM-2010 (JEOL Ltd., Tokyo, Japan). Samples for TEM measurement were prepared by a Reichert Ultracut ultramicrotome. The particle size distribution was obtained using a Zetasizer Nano ZS instrument (Malvern Instruments Ltd., Worcestershire, UK). The specific surface area was determined using N2 sorption isotherms by the BET and BJH methods using a gas sorption analyzer (Quantachrome AutoSorb IQ, Montgomeryville, PA, USA).
Scheme 1b shows a schematic illustration of the sensor preparation. The gas-sensing properties of the sensors were evaluated by a homemade dynamic test system equipped with a real-time resistance acquisition platform at 25 °C and a relative humidity of 40 ± 5%. The gas concentrations of NH3 samples and interfering gas samples, including CH3OH, C2H5OH, and C3H6O, were investigated by altering the mixing ratio of nitrogen and test samples. The sensor response is given using the equation of R = Rg/Ra, where Rg and Ra are the sensor resistances in test gases and air, respectively [30]. The slope of the response-concentration fitting curve is designated as the sensitivity of the sensors. The response time and recovery time were calculated as the time taken for the sensor to attain 90% of total resistance change from its initial resistance.

Synthesis, Structure, and Morphology of PANI/N-GQD/Hollow In2O3 Nanofiber Composites
A characteristic TEM micrograph of N-GQDs is shown in Figure 1a. The average diameter of the N-GQDs, determined using a Nano ZS instrument, was about 6 nm ( Figure 1b). Raman spectra were used to obtain the absorption bands of the N-GQDs and

Analytical Procedures
The molecular structures of the fabricated PANI/N-GQD/hollow In 2 O 3 nanofiber composites were determined by Fourier transform infrared (FTIR), Raman, and wide-angle X-ray diffraction (WAXD). FTIR spectra were determined with a PerkinElmer Spectrum One spectrometer (Waltham, Massachusetts, USA) in a range of 400-4000 cm −1 . Raman spectra were obtained on a Renishaw system 1000 (Renishaw Inc., West Dundee, IL, USA) using an argon ion laser performing at 514.5 nm with a CCD detector. WAXD measurement experiments were operated using an X-ray diffractometer (Bruker D8, BRUKER AXS, Inc., Madison, WI, USA) equipped with a Ni-filtered Cu Kα radiation. The WAXD experiments were recorded in the range of 2θ = 1.5 • -40 • with an increment of 1 • /min. The morphology of all samples was classified using field-emission scanning electron microscopy (FESEM) and transmission electron microscopy (TEM). The measurement of FESEM was carried out using a JEOL JSM-6700F field-emission instrument (JEOL Ltd., Tokyo, Japan). All specimens were coated with gold to avoid charging. The measurement of TEM was carried out using JEOL JEM-2010 (JEOL Ltd., Tokyo, Japan). Samples for TEM measurement were prepared by a Reichert Ultracut ultramicrotome. The particle size distribution was obtained using a Zetasizer Nano ZS instrument (Malvern Instruments Ltd., Worcestershire, UK). The specific surface area was determined using N 2 sorption isotherms by the BET and BJH methods using a gas sorption analyzer (Quantachrome AutoSorb IQ, Montgomeryville, PA, USA).
Scheme 1b shows a schematic illustration of the sensor preparation. The gas-sensing properties of the sensors were evaluated by a homemade dynamic test system equipped with a real-time resistance acquisition platform at 25 • C and a relative humidity of 40 ± 5%. The gas concentrations of NH 3 samples and interfering gas samples, including CH 3 OH, C 2 H 5 OH, and C 3 H 6 O, were investigated by altering the mixing ratio of nitrogen and test samples. The sensor response is given using the equation of R = R g /R a , where R g and R a are the sensor resistances in test gases and air, respectively [30]. The slope of the response-concentration fitting curve is designated as the sensitivity of the sensors. The response time and recovery time were calculated as the time taken for the sensor to attain 90% of total resistance change from its initial resistance.

Results
Synthesis, Structure, and Morphology of PANI/N-GQD/Hollow In 2 O 3 Nanofiber Composites A characteristic TEM micrograph of N-GQDs is shown in Figure 1a. The average diameter of the N-GQDs, determined using a Nano ZS instrument, was about 6 nm ( Figure 1b). Raman spectra were used to obtain the absorption bands of the N-GQDs and  Figure 1c. In this figure, two strong absorption bands at 1581 cm −1 (G mode) and at 1338 cm −1 (D mode) are observed [34]. are shown in Figure 1c. In this figure, two strong absorption bands at 1581 cm −1 (G mode) and at 1338 cm −1 (D mode) are observed [34]. The morphology and structure of the fabricated In2O3 nanofibers were determined using WAXD, FESEM, and TEM. Figure 2a exhibits a typical FESEM micrograph of an electrospun In (NO3)3/PVP nanofiber, which displays a continuous fibrous morphology with smooth and uniform surfaces. The average diameter is about 540 nm. After calcination in air, the diameter of the In2O3 nanofibers decreased drastically to 165 nm, and the surfaces became coarse, as presented in Figure 2b. In addition, the microstructure of hollow In2O3 nanofibers was further identified by TEM, as illustrated in Figure 2c. The crystalline structure of the hollow In2O3 nanofibers was determined using a WAXD technique. The WAXD diffraction pattern of the hollow In2O3 nanofibers, as presented in Figure 2d, exhibits five strong diffraction peaks at 2θ = 21.7°, 30.6°, 35.4°, 51.2°, and 60.7°, assigned to the (211), (222), (400), (440), and (622) crystalline planes of In2O3, respectively. This finding suggests that the crystalline structure of the hollow In2O3 nanofibers is a cubic phase [18]. The morphology and structure of the fabricated In 2 O 3 nanofibers were determined using WAXD, FESEM, and TEM. Figure 2a exhibits a typical FESEM micrograph of an electrospun In (NO 3 ) 3 /PVP nanofiber, which displays a continuous fibrous morphology with smooth and uniform surfaces. The average diameter is about 540 nm. After calcination in air, the diameter of the In 2 O 3 nanofibers decreased drastically to 165 nm, and the surfaces became coarse, as presented in Figure 2b. In addition, the microstructure of hollow In 2 O 3 nanofibers was further identified by TEM, as illustrated in Figure 2c. The crystalline structure of the hollow In 2 O 3 nanofibers was determined using a WAXD technique. The WAXD diffraction pattern of the hollow In 2 O 3 nanofibers, as presented in Figure 2d, exhibits five strong diffraction peaks at 2θ = 21.7 • , 30.6 • , 35.4 • , 51.2 • , and 60.7 • , assigned to the (211), (222), (400), (440), and (622) crystalline planes of In 2 O 3 , respectively. This finding suggests that the crystalline structure of the hollow In 2 O 3 nanofibers is a cubic phase [18].
The chemical structure and morphology of the fabricated polyaniline (PANI)-coated hollow In 2 O 3 nanofibers synthesized using in situ chemical oxidative polymerization were identified using the FTIR, FESEM, and TEM methods. Figure 3 shows the FTIR spectra of PANI-coated composites with various weight ratios of hollow In 2 O 3 nanofiber. The FTIR data of neat PANI and hollow In 2 O 3 nanofiber were also presented in this figure.
The absorption peaks of hollow In 2 O 3 nanofiber occurring at 538, 567, and 600 cm −1 were attributed to the In-O-In stretching vibration [35]. The characteristic peak of PANI observed at 800 cm −1 was attributed to the C-N •+ stretching vibration, and the absorption band at 1240 cm −1 was ascribed to a C-H out-of-plane bending vibration of the 1,4-disubstituted aromatic rings [36]. The absorption peaks at 1112 and 1294 cm −1 were assigned to the C=N and C-N stretching vibrations, respectively [36]. The FTIR spectra of PANI coated with significant amounts of hollow In 2 O 3 nanofiber were approximately identical to those of pure PANI, indicating that the surface of the hollow In 2 O 3 nanofibers was coated with PANI to form PANI/hollow In 2 O 3 nanofiber composites. It was also found that the  Figure 4 shows the FESEM and TEM images of PANI/hollow In 2 O 3 nanofiber composites with various loading ratios of hollow In 2 O 3 nanofibers. By adding the hollow In 2 O 3 nanofiber to PANI, the diameters of the PANI/hollow In 2 O 3 nanofiber composites decreased as the weight ratio of hollow In 2 O 3 nanofiber increased to 20 wt%. Notably, in Figure 4a, the diameters of the PANI/hollow In 2 O 3 nanofiber composites decreased from 180-200 nm to 140-160 nm for 20 wt% hollow In 2 O 3 nanofiber. The decreasing diameter in the PANI/hollow In 2 O 3 nanofiber composites can be attributed to the increasing weight ratio of hollow In 2 O 3 nanofiber, which reduces the ratio of aniline monomer/hollow In 2 O 3 nanofiber and possibly gives more active sites for the absorption and polymerization of aniline monomers on the surface of the hollow In 2 O 3 nanofiber, thereby decreasing the coating thickness of PANI layers. The data of the coating thickness of the PANI layer for the PANI/hollow In 2 O 3 nanofiber composites, calculated using TEM images, is listed in Table 1.
As the content of hollow In 2 O 3 nanofiber increases, the diameter of PANI/hollow In 2 O 3 nanofiber composites slightly increases, which can be ascribed to the aggregation effect. Figure 5 shows the specific surface area of the PANI/hollow In 2 O 3 nanofiber composites with various loading ratios of hollow In 2 O 3 nanofiber. The data of the specific surface area for the PANI/hollow In 2 O 3 nanofiber composites are also recorded in Table 1. It is clear that the specific surface area increases with the increase in weight ratio of hollow In 2 O 3 nanofiber to 20 wt%. This observation is in good agreement with the FESEM/TEM images. The chemical structure and morphology of the fabricated polyaniline (PANI)-coated hollow In2O3 nanofibers synthesized using in situ chemical oxidative polymerization were identified using the FTIR, FESEM, and TEM methods. Figure 3 shows the FTIR spectra o   Figure 4 shows the FESEM and TEM images of PANI/hollow In2O3 nanofiber composites with various loading ratios of hollow In2O3 nanofibers. By adding the hollow In2O3 nanofiber to PANI, the diameters of the PANI/hollow In2O3 nanofiber composites decreased as the weight ratio of hollow In2O3 nanofiber increased to 20 wt%. Notably, in Figure 4a, the diameters of the PANI/hollow In2O3 nanofiber composites decreased from 180-200 nm to 140-160 nm for 20 wt% hollow In2O3 nanofiber. The decreasing diameter in the PANI/hollow In2O3 nanofiber composites can be attributed to the increasing weight ratio of hollow In2O3 nanofiber, which reduces the ratio of aniline monomer/hollow In2O3 nanofiber and possibly gives more active sites for the absorption and polymerization of aniline monomers on the surface of the hollow In2O3 nanofiber, thereby decreasing the coating thickness of PANI layers. The data of the coating thickness of the PANI layer for the PANI/hollow In2O3 nanofiber composites, calculated using TEM images, is listed in Table 1. As the content of hollow In2O3 nanofiber increases, the diameter of PANI/hollow In2O3 nanofiber composites slightly increases, which can be ascribed to the aggregation effect. Figure 5 shows the specific surface area of the PANI/hollow In2O3 nanofiber composites with various loading ratios of hollow In2O3 nanofiber. The data of the specific surface area for the PANI/hollow In2O3 nanofiber composites are also recorded in Table 1. It is clear that the specific surface area increases with the increase in weight ratio of hollow In2O3 nanofiber to 20 wt%. This observation is in good agreement with the FESEM/TEM images.  The response and recovery attributes of neat PANI and various hollow In 2 O 3 nanofiber ratios of PANI/hollow In 2 O 3 nanofiber sensors were comparatively investigated to estimate the effect of the weight ratio of hollow In 2 O 3 nanofiber on the NH 3 -sensing performance of the PANI/hollow In 2 O 3 nanofiber sensor. The dynamic response-recovery curves of the composite sensors, with exposure of 1 ppm NH 3 at room temperature, are presented in Figure 6. It was observed that the PANI/hollow In 2 O 3 nanofiber sensor reacted with an improvement in resistance when exposed to NH 3 and the resistance dropped down to the original state after the NH 3 was substituted with dry air. This result reveals a distinctive performance of the PANI/hollow In 2 O 3 nanofiber composite film and an excellent reversibility of the PANI/hollow In 2 O 3 nanofiber sensor. Particularly, the PANI/hollow In 2 O 3 nanofiber sensor exhibited much higher response values than the pure PANI, indicating that the hollow In 2 O 3 nanofiber plays a promising role in NH 3 -sensing measurements. It was also noticed that the PANI/hollow In 2 O 3 nanofiber sensor with 20 wt% hollow In 2 O 3 nanofiber loading achieved a maximum response which was far superior to other composite specimens. In order to estimate the effect of different weight ratios of hollow In 2 O 3 nanofiber on the NH 3 -sensing performance of the PANI/hollow In 2 O 3 nanofiber sensor, the plot of the specific surface area, the coating thickness of the PANI layer, and the response values for the PANI/hollow In 2 O 3 nanofiber composites versus the adding ratios of hollow In 2 O 3 nanofiber is shown in Figure 7. It is clear that the 20 wt% hollow In 2 O 3 nanofiber loaded PANI/hollow In 2 O 3 nanofiber composite possesses the highest specific surface area, the smallest coating thickness of the PANI layer, and the highest response values of the PANI/hollow In 2 O 3 nanofiber composites. Therefore, 20 wt% hollow In 2 O 3 nanofiber loading was used as the optimal loading ratio to fabricate the PANI/N-GQD/hollow In 2 O 3 nanofiber composites. Figure 8a illustrates the response of PANI, 20 wt% PANI/hollow In 2 O 3 nanofiber, and 20 wt% N-GQD-coated hollow In 2 O 3 PANI/N-GQD/hollow In 2 O 3 nanofiber. The response value of pure PANI was about 3.6, but with the loading of 20 wt% hollow In 2 O 3 nanofiber into PANI the response value was extensively enhanced to 11.2. The response value of the PANI/N-GQD/hollow In 2 O 3 nanofiber sensor was 15.2 with the loading of coated N-GQD on the surface of the hollow In 2 O 3 nanofiber. The improvement of the sensing performance was ascribed to the existence of a p-n junction between the p-type PANI and n-type N-GQD-coated hollow In 2 O 3 nanofiber [18,36]. The sensing reversibility and repeatability of the PANI/N-GQD/hollow In 2 O 3 nanofiber sensor to 1.0 ppm NH 3 are shown in Figure 8b. The response value of the PANI/N-GQD/hollow In 2 O 3 nanofiber gas sensor came back to the initial response value after exposure to 1.0 ppm NH 3 . This characteristic behavior of response and recovery to 1.0 ppm NH 3 confirmed excellent reproducibility in the process of five continues cycles. This result suggests that the PANI/N-GQD/hollow In 2 O 3 nanofiber sensor has good repeatability.
Polymers 2021, 13, x FOR PEER REVIEW Figure 5. The BET curves of (a) neat PANI polymer matrix, (b) 2.5 wt% PANI/hollo nanofiber, (c) 5 wt% PANI/hollow In2O3 nanofiber, (d) 10 wt% PANI/hollow In2O3 nanofibe wt% PANI/hollow In2O3 nanofiber, (f) 20 wt% PANI/hollow In2O3 nanofiber, and (g) PANI/hollow In2O3 nanofiber composites. value of the PANI/N-GQD/hollow In2O3 nanofiber gas sensor came back to response value after exposure to 1.0 ppm NH3. This characteristic behavior o and recovery to 1.0 ppm NH3 confirmed excellent reproducibility in the pro continues cycles. This result suggests that the PANI/N-GQD/hollow In2O3 sensor has good repeatability. The PANI/N-GQD/hollow In2O3 nanofiber sensor was applied to detect NH3 in a concentration ranging from 0.6 ppm to 2.0 ppm at room temperature and to estimate the NH3-sensing performance of the composite sensor used in the analysis of human breath for hepatic or kidney disease. The NH3-sensing performance of the PANI/N-GQD/hollow In2O3 nanofiber sensor compared with the PANI and PANI/hollow In2O3 nanofiber sensors is shown in Figure 9a. These results indicate that the response of each sensor PANI/N-GQD/hollow In2O3 nanofiber sensor exhibited better sensing performance to NH3 at room temperature than previously described devices. Consequently, commingled p-type PANI and n-type N-GQD-coated hollow In2O3 nanofibers could be applied as an effectual approach for improving the NH3-sensing response of sensors.  The PANI/N-GQD/hollow In 2 O 3 nanofiber sensor was applied to detect NH 3 in a concentration ranging from 0.6 ppm to 2.0 ppm at room temperature and to estimate the NH 3 -sensing performance of the composite sensor used in the analysis of human breath for hepatic or kidney disease. The NH 3 -sensing performance of the PANI/N-GQD/hollow In 2 O 3 nanofiber sensor compared with the PANI and PANI/hollow In 2 O 3 nanofiber sensors is shown in Figure 9a. These results indicate that the response of each sensor immediately increased as the exposure to NH 3 increased and then dropped significantly to an initial state after exposure to dry air. When the concentration of the analyte increased, the response of each sensor increased significantly. These results suggest that all responses of the three sensors had almost the same trend but the response values of the three sensors at the same concentration were exceptionally different. The response values of the PANI/N-GQD/hollow In 2 O 3 nanofiber sensors were about 9.4, 11.6, 15.  Figure 10 shows the responses of the three sensors as a function of various concentrations of NH 3 . These curves reveal that the correlation between the response values and the NH 3 concentration is nearly linear. The corresponding functions were determined as y = 2.04x + 2.55, y = 15.23x − 3.88, and y = 23.95x − 10.41 for the PANI, PANI/hollow In 2 O 3 nanofiber, and PANI/N-GQD/hollow In 2 O 3 nanofiber sensors, respectively. The correlation coefficients, R 2 , were approximately equal to 0.994. The slopes of the corresponding lines, designated as the sensitivities of the sensors, reveal that the sensitivity of the PANI/N-GQD/hollow In 2 O 3 nanofiber sensor was superior to those of the PANI and PANI/hollow In 2 O 3 nanofiber sensors. These findings suggest that the PANI/N-GQD/hollow In 2 O 3 nanofiber sensor has a greater ability to detect NH 3 , recommending it for suitable use as a favored substance for NH 3 gas detection. A comparison of the sensing performance of the PANI/N-GQD/hollow In 2 O 3 nanofiber sensor and previously described devices was recorded in Table 2. It is evident that the PANI/N-GQD/hollow In 2 O 3 nanofiber sensor exhibited better sensing performance to NH 3 at room temperature than previously described devices. Consequently, commingled p-type PANI and n-type N-GQD-coated hollow In 2 O 3 nanofibers could be applied as an effectual approach for improving the NH 3 -sensing response of sensors. Figure 10 presents the synergistic effect of the PANI/N-GQD/hollow In 2 O 3 nanofiber ternary material, from which it is apparent that the combination of N-GQD and hollow In 2 O 3 nanofibers enhances the response of PANI to NH 3 gas. PANI-coated N-GQD/hollow In 2 O 3 nanofibers can generate new chemical bonds on the oxygen-containing defects of N-GQD and hollow In 2 O 3 nanofiber surfaces. The extensive special surface area of N-GQDs and hollow In 2 O 3 nanofibers further improves the contact sites with PANI, which can provide a considerable number of adsorption sites for NH 3 gas. This observable fact has previously been proved in Figures 4, 5 and 7. Consequently, the synergistic arrangement of N-GQD, hollow In 2 O 3 nanofiber, and PANI creates exceptional sensing properties superior to those of pure PANI and PANI/hollow In 2 O 3 nanofiber sensors. The characteristic of p-n heterojunctions generated between the p-type PANI and n-type N-GQD-coated hollow In 2 O 3 nanofibers could be attributed as another NH 3 -sensing mechanism of the PANI/N-GQD/hollow In 2 O 3 nanofiber film. The positions of the conduction band (E c ) and valence band (E v ) in In 2 O 3 , the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) for PANI, and the Fermi levels (E f ) regarding the vacuum level are presented in the energy band gap structure diagram in Figure 10. The band gaps for PANI and In 2 O 3 are 2.96 and 3.85 eV, respectively [15,37]. As the occurrence of the heterojunction between the p-type PANI and n-type N-GQD-coated hollow In 2 O 3 nanofiber is present, a self-established electronic field of depletion layer is generated. When the sensor is subjected to NH 3 gas, the In 2 O 3 electrons and PANI holes shift in opposite directions until the new Fermi level achieves equilibrium. During this procedure, electron transfer from the n-type N-GQD-coated hollow In 2 O 3 nanofiber to the p-type PANI is delayed, attributable to the potential barrier, which enlarges the depletion layer thickness and resistance of the sensor [37]. The p-n heterojunction performs as an indicator amplifier and makes it easier to effectively detect tiny quantities of NH 3 [38].  electrons and PANI holes shift in opposite directions until the new Fermi level achieves equilibrium. During this procedure, electron transfer from the n-type N-GQD-coated hollow In2O3 nanofiber to the p-type PANI is delayed, attributable to the potential barrier, which enlarges the depletion layer thickness and resistance of the sensor [37]. The p-n heterojunction performs as an indicator amplifier and makes it easier to effectively detect tiny quantities of NH3 [38]. For practical applications, the repeatability, reversibility, and selectivity of synthesized gas sensors are critical. Actually, gas sensors are normally exposed to numerous analytes, and it is assumed that the target analyte is detected accurately without being influenced by other species. Figure 11 exhibits the selectivity of the PANI/N-GQD/hollow In2O3 nanofiber sensor toward ammonia (NH3), methanol (CH3OH), ethanol (C2H5OH), and acetone (CH3COCH3) with a concentration of 1.0 and 2.0 ppm. The selectivity test shown in Figure 11 is the first test in sensor characterization. It was noticeably illustrated that the PANI/N-GQD/hollow In2O3 nanofiber sensor had a highlevel response property to ammonia but displayed approximately no response to other gases. The mechanism of NH3 selectivity may be related to the surface absorption of NH3 on the interface of the PANI/N-GQD/hollow In2O3 nanofiber sensor. The de-doping  For practical applications, the repeatability, reversibility, and selectivity of synthesized gas sensors are critical. Actually, gas sensors are normally exposed to numerous analytes, and it is assumed that the target analyte is detected accurately without being influenced by other species. Figure 11 exhibits the selectivity of the PANI/N-GQD/hollow In 2 O 3 nanofiber sensor toward ammonia (NH 3 ), methanol (CH 3 OH), ethanol (C 2 H 5 OH), and acetone (CH 3 COCH 3 ) with a concentration of 1.0 and 2.0 ppm. The selectivity test shown in Figure 11 is the first test in sensor characterization. It was noticeably illustrated that the PANI/N-GQD/hollow In 2 O 3 nanofiber sensor had a high-level response property to ammonia but displayed approximately no response to other gases. The mechanism of NH 3 selectivity may be related to the surface absorption of NH 3 on the interface of the PANI/N-GQD/hollow In 2 O 3 nanofiber sensor. The de-doping response between NH 3 and PANI plays an important role in improving NH 3 -sensing performance, leading to a selective response to NH 3 [18,34]. Subsequently, it can be established that the PANI/N-GQD/hollow In 2 O 3 nanofiber sensor exhibited selectivity toward NH 3 at room temperature as against other gases.
Polymers 2021, 13, x FOR PEER REVIEW 16 of Figure 11. The selectivity of the PANI/GQD/hollow In2O3 nanofiber film sensor tested at the concentrations of 1 and 2 ppm toward NH3, methanol, ethanol, and acetone.

Conclusions
We have shown outstanding gas-sensing performances of PANI/N-GQD/hollow In 2 O 3 nanofiber ternary composites effectively synthesized by in situ chemical oxidation polymerization. The response value of the 20 wt% N-GQD-coated hollow In 2 O 3 nanofiber added to the PANI/N-GQD/hollow In 2 O 3 nanofiber sensor was 15.1 at an exposure of 1 ppm NH 3 , which was approximately 4.4 and 1.4 times higher than those of the PANI and PANI/hollow In 2 O 3 nanofiber sensors, respectively. The PANI/N-GQD/hollow In 2 O 3 nanofiber sensor showed the highest response at room temperature in detecting NH 3 concentrations ranging from 0.6 ppm to 2.0 ppm, as compared to those of the PANI and PANI/hollow In 2 O 3 nanofiber sensors; this is critical for detecting hepatic or kidney disease in human breath. The PANI/N-GQD/hollow In 2 O 3 nanofiber sensor also showed higher repeatability and selectivity on exposure to 1.0 and 2.0 ppm NH 3 at room temperature. Owing to the exceptional repeatability and selectivity in the detection of 1.0 and 2.0 ppm NH 3 at room temperature reported in this study, it is considered that the PANI/N-GQD/hollow In 2 O 3 nanofiber composite sensor will be a favored gas-sensing material for the detection of hepatic or kidney disease in human breath.