Aniline, formic acid (FA), ammonium persulfate (APS), ammonia and sulfuric acid (H₂SO₄) were purchased from Sinopharm Chemical Reagent Co. Ltd. (Beijing, China). Polyamide 6 (PA6, Mw = 21,000 g/mol) was obtained from ZIG ZHENG Industrial Co. Ltd (Taibei, Taiwan). All chemicals and reagents were used as received, except for aniline, which was distilled before use.

PANI/PA6 composite nanofibers were prepared by electrospinning and in situ polymerization. PA6 nanofibers were firstly prepared by electrospinning PA6/FA solutions of 20% PA6 concentration. Then, aniline and APS were dissolved separately in aqueous solutions, and the H⁺ concentrations of the aqueous solutions were adjusted by H₂SO₄ to 1.5 mol/L. The mole ratio of aniline to APS was 1:1. The electrospun PA6 nanofibers, were then immersed into the aniline/H₂SO₄ solution for 30 min. Successive polymerization was finally initiated by dropping the acid aqueous solution of APS into the above diffusion bath. PANI was synthesized on the surface of PA6 nanofibers and doped with H₂SO₄ at 0–5 °C for 5 h. After the reaction, the samples were taken out, washed with deionized water, and dried in vacuum at 65 °C for 12 h. The steps for synthesis of PANI/PA6 composite nanofibers are illustrated in Figure 1.

TiO₂-PANI/PA6 composite nanofibers were obtained by depositing TiO₂ onto PANI/PA6 substrate at room temperature for different times via RF magnetron sputtering. High purity titanium discs (99.99%) of 50 mm diameter was used as sputtering targets. High purity argon (99.999%) and oxygen (99.999%) were used as the sputtering and reactive gases, respectively. A diffusion pump was used to get the desired 9.8 × 10⁻⁴ Pa base pressure. The argon and oxygen flow rates were controlled separately by mass flow meters. The distance between target and substrates was kept at 60 mm. Before each sputtering-deposition step, the target was pre-sputtered in argon for 10 min to clean the target surface. The sputtering conditions are listed in Table 1.

The structure and surface morphology of PANI/PA6 and TiO₂-PANI/PA6 composite nanofibers were observed with a field emission scanning electron microscope (FESEM, S-4800, Hitachi, Tokyo, Japan) with a golden coating. Fourier transform infrared (FTIR) spectrum of all the samples was obtained with a resolution of 4 cm⁻¹ in the range of 400–4,000 cm⁻¹ by using a NEXUS 470 spectrometer (Nicolet, Madison, WI, USA).

A sensing system for ammonia sensing test was fabricated by the following processes. A home-made Au electrode with a gap of 0.5 mm between two Au stripes was firstly prepared by depositing Au on phenolic resin, and then PANI/PA6 and TiO₂-PANI/PA6 composite nanofibers were pasted onto the open area between the two electrodes as shown in Figure 2.

To test the ammonia sensitivity, the sensing electrode was placed in a lab-made sensing system as shown in Figure 3. The sensing set-up consisted of an airtight test chamber with 4,500 mL volume, a heater pad and a fan. Two minutes after the system reached a steady-state, a certain amount of ammonia was injected with a microsyringe through the intake valve, and with the help of a heater pad, the ammonia was heated to evaporation. The resistance changes of PANI/PA6 and TiO₂-PANI/PA6 composite nanofibers sensors were monitored and recorded automatically by an Agilent electrometer and a computer. A constant voltage of 5.0 V was used as the DC power supply. All the tests were conducted at room temperature (25 ± 1 °C) with a relative humidity of 65 ± 1%.

During the measurements, the actual ammonia volumes injected were 0.67, 1.35, 2.02, 2.69 and 3.37 μL, corresponding to the ammonia vapor with the concentration of 50, 100, 150, 200 and 250 ppm, respectively. After the ammonia was introduced to the chamber, the resistance of the sensors was recorded for 250 s, then the test chamber was flushed with dry air consecutively for another 250 s to make sure that a relatively steady state had achieved before next cyclic test. The sensitivity is defined as (R_i – R₀)/R₀, where R_i and R₀ are the resistance of sensors in ammonia and in air, respectively. The response and recovery time are defined as the time of 90% total resistance change. Each result was the average value of fivetimed tests.

It is well known that structure and morphology can have a significant effect on the sensing properties of materials. The SEM images of PA6 nanofibers, PANI/PA6 composite nanofibers and TiO₂-PANI/PA6 composite nanofibers sputtered for different times are shown in Figure 4. It can be seen that the surface morphology of PA6nanofibersappeared smooth, while PANI/PA6 composite nanofibers had a rough surface and more uniform diameter because of the PANI coating, as indicated in Figure 4(a,b). The SEM images clearly revealed that the TiO₂-PANI/PA6 composite nanofibers had very rough surface with porous structures, as presented in Figure 4(c–e). This could be attributed to the impact of high-energy particles during sputtering deposition of TiO₂. It is obvious that such porous structure exhibited higher specific surface area than PANI/PA6, which facilitated the diffusion of ammonia vapor in sensing materials. However, TiO₂-PANI/PA6 with 90 min deposition time showed a distorted surface structure as the excessive sputtering time could damage the integrity of the PANI coating.

The FTIR spectra of PANI/PA6 and TiO₂-PANI/PA6 composite nanofibers are shown in Figure 5. For the PANI/PA6 composite nanofibers, the peaks around 1,476.15 cm⁻¹ and 1,559.92 cm⁻¹ were assigned to C=C stretching vibrations of the benzenoid and quinoid rings, respectively. The peaks at 1,298.85 cm⁻¹ and 798.54 cm⁻¹ resulted from the C–N stretching vibration of the secondary aromatic amine and the C–H bending vibration, respectively. The characteristic peak of Q=NH⁺–B was also observed at around 1,118.85 cm⁻¹. All these peaks were identical to those of PANI. On the other hand, the C=O stretching vibration peak of amide in PA6 was also observed at 1,636.84 cm⁻¹, while the amide N–H stretching vibration peak overlapped with the C=C stretching peak of the PANI quinoid rings, thus the peak at 1,559.92 cm⁻¹ was shown to be broader. All the characteristic peaks of PANI/PA6 could be observed in the spectrum of TiO₂-PANI/PA6 composite nanofibers, and the characteristic bands around 616.44 cm⁻¹ and 573.27 cm⁻¹ was attributed to Ti–O bending vibration of TiO₂. It also can be observed that incorporation of TiO₂ nanoparticles leads peaks of PANI/PA6 to shift slightly to lower wave number, indicating that some interaction existed between PANI and TiO₂.