In situ synthesis of long tubular water-dispersible polyaniline with core/shell gold and silver@graphene oxide nanoparticles for gas sensor application

Abstract

Gold nanoparticles (Au NPs) with graphene oxide (GO) shell (Au@GO), silver nanoparticles (Ag NPs) with GO shell (Ag@GO), and gold silver nanoparticles (AuAgNPs) with GO shell (AuAg@GO) were synthesized employing a cationic surfactant. The prepared core@shell structures were used for in situ synthesis of long tubular polyaniline structures employing cetyl trimethyl ammonium bromide (CTAB) as a soft template. This process led to a notable enhancement in the tubular nanostructure of PANI, extending its length beyond 10 μm, in the case of using core/shell Au@GO, Ag@GO, and AuAg@GO structures. To evaluate their applicability and compatibility, the dispersibility of these nanocomposites was assessed in three distinct solvents: water, dimethyl sulfoxide (DMSO), and N-Methyl-2-pyrrolidone (NMP). Subsequently, the dedoping of PANI within the prepared nanocomposites was scrutinized using UV–Visible (UV–Vis) spectroscopy, which revealed a reduction in the I₇₅₀/I₃₁₅ ratio from 1.00 to 0.66 when subjected to water and NMP solvents, respectively. Notably, the dedoping of the AuAg@GO/PANI nanocomposite was predominantly observed in NMP, attributable to the presence of hydrogen bonding interactions and the basic properties of NMP. In terms of ionic conductivity, it was observed that the prepared nanocomposite exhibited its highest conductivity in a water-based medium, registering at 1982 μs. Furthermore, the AuAg@GO/PANI nanocomposite exhibited superior sensing capabilities in comparison to PANI-based gas sensor devices, particularly when exposed to acetone, CO₂, NO₂, and H₂S. Remarkably, at room temperature (25 °C), the AuAg@GO/PANI nanocomposite displayed rapid response and recovery times, with values of 279 s, 431 s, 335 s, and 509 s for 1 ppm concentrations of CO₂, NO₂, H₂S, and acetone, respectively. The sensitivity of these sensors towards acetone, CO₂, NO₂, and H₂S, was quantified by analyzing the slope of the response versus the target gas concentration, revealing the AuAg@GO/PANI nanocomposite to exhibit the highest sensitivity, particularly towards NO₂.

1. Introduction

Monitoring gases, especially toxic ones, is of paramount importance across various domains such as agriculture, automotive, security, aviation, healthcare, environmental monitoring, and industry [1,2]. To address this need, a wide range of sensors based on both organic conductive polymers and inorganic metal oxides has been extensively explored [3]. Among these, chemical gas sensors utilizing conducting polymers like polypyrrole (PPy), polyaniline (PANI), poly (3,4-ethylene-dioxythiophene) (PEDOT), and polythiophene (P3HT) have garnered attention due to their cost-effectiveness, lightweight nature, flexibility, facile synthesis processes, and the ability to operate at room temperature [4,5]. Notably, polyaniline (PANI) stands out as one of the most promising conductive materials, capable of transitioning between various oxidation states, including non-conductive, fully oxidized, and reduced forms, such as pernigraniline, and the conducting emeraldine salt, which can be achieved through slight doping [6]. PANI's simplicity of synthesis, inherent self-doping characteristics, and high solubility in alkaline solutions render it particularly attractive for industrial applications and sensor development [6]. However, despite its merits, PANI has faced challenges related to high response times, low sensitivity, inadequate stability, susceptibility to humidity, and incomplete reversibility of sensor responses [7]. In response to these limitations, researchers have explored various strategies and materials, including PANI nanostructures and PANI nanocomposites.

Studies have demonstrated the synthesis of PANI nanostructures at the nanoscale through both soft [[8], [9], [10], [11], [12]] and hard templates [[13], [14], [15], [16]]. In the hard template method, diverse types of hard templates, such as porous templates [15] and zeolite channel [16], have been introduced by researchers. However, employing hard templates necessitates complex post-treatment procedures for PANI extraction, which can potentially damage the PANI structure. Therefore, soft templates, including micelles [11], vesicles [8], PANI oligomers [9,10], carbon nanotubes (CNT) [17], biological molecules like vitamin C [18], reactive oxide templates [19], and emulsifiers [20], have gained preference due to their controllable properties. Research findings indicate that the utilization of PANI in nanoparticle, nanofiber, nanowire, and nanotube forms has been proposed to enhance sensor response times by increasing the surface-to-volume ratio [21,22]. Furthermore, PANI nanocomposites, combining the advantages of both inorganic materials and PANI, have exhibited remarkable responses to gases such as hydrogen sulfide (H₂S), ammonia (NH₃), hydrogen (H₂), nitrogen dioxide (NO₂), hydrochloric acid (HCl), sulfur dioxide (SO₂), carbon monoxide (CO), carbon dioxide (CO₂), and other volatile organic compounds (VOCs) [[23], [24], [25]]. Essentially, the incorporation of metallic nanoparticles (NPs) into PANI has enhanced gas sensor performance through two mechanisms. First, n-type semi-conductive nanoparticles introduced into PANI can create depletion regions with reduced conductivity at PANI/nanoparticle interfaces, thereby widening the depletion region can enhance sensor response [26]. Second, the application of inorganic nanostructures like gold (Au), silver (Ag), palladium (Pd), and copper (Cu) facilitates gas adsorption on the nanostructure surface due to their catalytic properties [27]. For instance, Kinyanjui et al. introduced a novel approach for PANI-gold nanoparticle (Au NPs) synthesis, where they demonstrated that HAuCl₄ could serve as both a metal salt and an oxidizing agent. In this process, the synthesized gold nanoparticles exhibited uniform dispersion within the PANI matrix. However, due to the amorphous structure of PANI, the surface area of the resulting PANI-gold nanoparticles was limited [28]. Iole Venditti et al. employed an Au/PANI nanocomposite for the detection of H₂O and NH₃, achieving a 1.9% ppm⁻¹ improvement in response compared to PANI alone [29]. Researchers have also explored graphene oxide (GO) combined with PANI, observing conductivity and capacitance dependencies on the PANI/GO nanocomposite microstructure when using varying ratios of cetyl trimethyl ammonium bromide (CTAB)-stabilized GO [30]. Graphene oxide, known for its 2D structure and advantageous properties such as ease of fabrication, cost-effectiveness, rapid processability, facile functionalization through oxidation, and a large surface area (2.6 × 103 m²/g), holds significant promise for gas sensor applications [31,32]. Furthermore, researches used activated charcoal/PANI composite for selective sensing of methanol vapor [33]. Polyaniline montmorillonite reduce graphene oxide polymer nanocomposite material was also used for detection of hydrogen cyanide gas [34]. Montmorillonite-reduced graphene oxide combination enhances the material's structural and electrical properties. In gas sensing applications, MRGO demonstrates improved sensitivity and selectivity due to its increased surface area and the synergistic effects between graphene and montmorillonite. Researchers also used Ag-decorated ZnO nanorods/PANI nanocomposite for detecting Ammonia gas [35]. A gas sensing response of 59% for 120 ppm NH₃ gas was achieved with a 10 % loading of ZnO, and further enhancement was demonstrated by embedding Ag-decorated ZnO nanorods in the PANi matrix, resulting in a highly selective and sensitive sensor with a chemiresistive response of 70% at 120 ppm NH₃ and a recovery time of less than 120 s at room temperature.

In this article, we introduce a novel architecture featuring long tubular PANI nanocomposites, using an AuAg@GO core/shell structure and CTAB as the soft cationic surfactant. Our study explores the impact of CTAB concentration on the tubular structure of PANI in the presence of the AuAg@GO nanocomposite. We also investigate the effects of different solvents (water, dimethyl sulfoxide (DMSO), N-Methyl-2-pyrrolidone (NMP)) on conductivity, solubility, and de-doping of the nanocomposites. To the best of our knowledge, no prior scientific reports have detailed the in-situ synthesis of AuAg@GO and PANI with a controllable polymer shape, utilizing CTAB as a soft template. Furthermore, the synthesized AuAg@GO/PANI nanocomposite is assessed as a gas sensor for the detection of acetone, CO₂, NO₂, and H₂S, revealing significant reductions in response time and recovery time. The proposed PANI nanocomposite architecture holds promise for applications in sensors, catalysis, energy storage, and optoelectronic devices.

2. Experimental section

2.1. Materials

Ammonium persulfate (APS), silver nitrate (AgNO₃), sodium borohydride (NaBH₄) and cetyl trimethyl ammonium bromide (CTAB) were purchased from Sigma-Aldrich and were used as received. Chloroauric acid (HAuCl₄), trisodium citrate (TSC), aniline, and graphite powder were acquired from Merck and used as received without any purification. All the solvents were also obtained from Merck and deionized water was used for the synthesis throughout the experiment.

2.2. Synthesis of gold and silver nanoparticles

Silver nanoparticles were synthesized using chemical reduction of AgNO₃ with NaBH₄ as reducing and stabilizing agent [36][Fig. S1]. Gold nanoparticles were synthesized using turkevich method [37][Fig. S2]. The detailed process of synthesizing silver NPs and gold NPs are illustrated in the supporting information.

2.3. Synthesis of core/shell Au@GO and Ag@GO nanostructure

CTAB powder was dissolved in deionized (DI) water (15 mM) and stirred for 6 h, so that no bubbles were left. 4 mL of Au NPs solution (10⁻¹³ M) was added to 4 mL of prepared CTAB solution and stirred for 30 min until micelles were formed on the outer layer of nanoparticles. Therefore, negative zeta potential of Au NPs due to TSC changed to positive because of CTAB micellar structure (not shown here). Filtered graphene oxide (2 mL, 1 mg/mL) was added dropwise to the solution (Fig. S3). The detailed process of synthesizing GO is illustrated in the supporting information. Core/shell structure of Au@GO was prepared in this way and stored at 8 °C for further experiments (Fig. S4) [38]. The same procedure was applied for the preparation of Ag@GO nanostructure (Fig. S5).

2.4. Synthesis of long tubular Au@GO/PANI, Ag@GO/PANI and AuAg@GO/PANI nanocomposites

To study the impact of soft template concentration on PANI nanostructure, CTAB was added, without any core/shell nanocomposite, in portions of 5, 10, 15, 20 and 25 mL. In this procedure, CTAB solutions were added to 0.2 mL of aniline and HCl (10 mL, 1 N) and stirred for 15 min. APS (0.6 g) was dissolved in 10 mL HCl (1 N) and stirred for 5 min. APS solution was added dropwise to the prepared monomer/CTAB/HCl solution at 0 °C and was stirred for 24 h. In accordance with the PANI nanostructure, 10 mL CTAB was chosen as the optimum amount for the rest of experiments.

For the synthesis of Au@GO/PANI, Ag@GO/PANI, and AuAg@GO/PANI nanocomposites the same procedure was applied. In the first step of polymerization, however, 10 mL CTAB was also added to the NPs@GO solutions. 10 mL CTAB and 10, 20, and 30 mL of core/shell solutions of Au@GO and Ag@GO were added separately to the monomer/HCl mixture followed by dropwise addition of the APS solution for the synthesis of Au@GO/PANI and Ag@GO/PANI nanocomposites. AuAg@GO/PANI nanocomposite was prepared by adding 10, 15, and 20 mL of each of the Au@GO/PANI and Ag@GO/PANI solutions to the monomer/HCl mixture followed by dropwise addition of the APS solution. These were named Au₁₀@GO/PANI, Au₂₀@GO/PANI and Au₃₀@GO/PANI for Au nanocomposites, Ag₁₀@GO/PANI, Ag₂₀@GO/PANI and Ag₃₀@GO/PANI for Ag nanocomposites and Au₁₀Ag₁₀@GO/PANI, Au₁₅Ag₁₅@GO/PANI and Au₂₀Ag₂₀@GO/PANI for Au and Ag nanocomposites. Table 1 illustrates compositions of synthesized nanocomposites. The mixtures were stirred overnight at 0 °C the color changed from yellow to light red then purple, blue and green, consecutively. This procedure is illustrated in Scheme 1.

Table 1.

Compositions of synthesized nanocomposites.

Sample name	CTAB (mL) (15 mM)	Au@GO solution (10−13 M)	Ag@GO solution (10−13 M)	Aniline solution in HCl (mL)
Au10@GO/PANI	10	10	–	10
Au20@GO/PANI	10	20	–	10
Au30@GO/PANI	10	30	–	10
Ag10@GO/PANI	10	–	10	10
Ag20@GO/PANI	10	–	20	10
Ag30@GO/PANI	10	–	30	10
Au10Ag10@GO/PANI	10	10	10	10
Au15Ag15@GO/PANI	10	15	15	10
Au20Ag20@GO/PANI	10	20	20	10

Scheme 1.

Synthesis procedure of metal NPs@GO/PANI nanocomposite.

To compare structural impact of the soft template, the same procedure was adopted for PANI synthesis, this time without introducing any CTAB and core/shell solutions.

2.5. Characterization methods

In this experiment, Fourier transform infrared spectroscopy (FTIR), UV–Vis spectroscopy (UV–Vis), X-ray diffraction (XRD) analysis were performed to characterize prepared nanocomposites. FTIR analysis was carried out using Bomem (Hartmann & Braun) with accelerating voltage of 230 V. XRD patterns were acquired from EQUINOX3000 Intel (Cu Kα, λ = 0.15418 nm, 40 kV, 30 mA). UV–Vis was performed using Specord 210 Analytik jena (220–240 V). Field emission scanning electron microscopy (FESEM), energy-dispersive X-ray spectroscopy and elemental map were performed using KYKY-EM3200 with the resolution of 6.0 nm and accelerating voltage of 30 kV to study Au@GO/PANI and Ag@GO/PANI nanostructures. Ionic conductivity was measured using 4510 Jenway conductivity meter in solution. Thermogravimetric analysis (TGA) was performed using Mettler Toledo instrument in the range of 25–800 °C. Transmission electron microscopy (TEM) was performed using Philips FM208S 100Kv instrument.

3. Result and discussion

3.1. Au@GO/PANI, Ag@GO/PANI and AuAg@GO/PANI characterization

Nine samples of Au₁₀@GO/PANI, Au₁₅@GO/PANI, Au₂₀@GO/PANI, Ag₁₀@GO/PANI, Ag₁₅@GO/PANI, Ag₂₀@GO/PANI, Au₁₀Ag₁₀@GO/PANI, Au₁₅Ag₁₅@GO/PANI, and Au₂₀Ag₂₀@GO/PANI underwent analysis using a four-point probe, with the resulting data revealing that the nanocomposites Au₂₀@GO/PANI, Ag₁₀@GO/PANI, and Au₁₅Ag₁₅@GO/PANI exhibited the highest levels of conductivity (Table S1). Consequently, these particular samples were chosen for subsequent procedures.

The molecular structures of PANI, Au₂₀@GO/PANI, Ag₁₀@GO/PANI, and Au₁₅Ag₁₅@GO/PANI were characterized through Fourier transform infrared spectroscopy (FTIR), as depicted in Fig. 1. Notably, the FTIR analysis revealed the presence of specific vibrational modes within the polymers. The C C stretching vibrations of the quinoid ring and the C C stretching vibrations of benzenoid rings of PANI appeared at approximately 1587 cm⁻¹ and 1498 cm⁻¹, respectively. These characteristic peaks provide evidence of amine and imine nitrogen units within the polymer's backbone. Furthermore, the observation of in-plane and out-of-plane C–H bending in the aromatics of the 1,4-aromatic ring at 1132 cm⁻¹ and 824 cm⁻¹, respectively, indicated a linear PANI structure [39]. Additional peaks at 1294 cm⁻¹, corresponding to C–N stretching of secondary amines in the polymer, and at 1109 cm⁻¹, indicative of delocalized charges in the polymer backbone resulting from the formation of –NH⁺ between benzenoid and quinoid rings [40], were also discerned. Vibrations in the range of 3000–3500 cm⁻¹ were attributed to N–H stretching, and further differentiation of N–H secondary amine types was evident at 2978 cm⁻¹ and 2843 cm⁻¹. CTAB, a surfactant, displayed distinctive peaks at 2850 cm⁻¹ and 2918 cm⁻¹, arising from C–H stretching vibrations in methyl and methylene groups, overlapping with N–H stretching vibrations in PANI.

Fig. 1.

FTIR spectra of PANI, Au₂₀@GO/PANI, Ag₁₀@GO/PANI and Au₁₅Ag₁₅@GO/PANI nanocomposites.

GO, known for its oxygen-functionalized groups, exhibited notable peaks at around 3450, 1700, and 1100 cm⁻¹, corresponding to O–H, C O, and C–O groups, respectively. Hydroxyl peaks were evident at 3400 cm⁻¹, with an additional peak at 1070 cm⁻¹ attributed to epoxy groups. In core/shell nanocomposites, a minor peak at approximately 1700 cm⁻¹ was observed, associated with carbonyl groups. The comparatively low intensity of this carbonyl peak, in comparison to the PANI characterization peaks, can be attributed to the low concentration of GO sheets in the nanocomposite.

Notably, the FTIR analysis of Au₂₀@GO/PANI, Ag₁₀@GO/PANI, and Au₁₅Ag₁₅@GO/PANI nanocomposites unveiled shifts in the PANI peaks towards lower wavenumbers, along with a reduction in peak intensity. These FTIR shifts were notably amplified with the introduction of Au and Ag nanoparticles, signifying significant alterations in the PANI structure resulting from interactions between the nanoparticles and the nitrogen units in the PANI backbone.

3.2. X-ray diffraction (XRD) analysis

X-ray diffraction (XRD) patterns of PANI, Au₂₀@GO/PANI, Ag₁₀@GO/PANI, and Au₁₅Ag₁₅@GO/PANI nanocomposites are illustrated in Fig. 2. The diffraction peaks observed at 15°, 20°, and 25° can be confidently assigned to the (011), (020), and (200) crystal planes of PANI [31], respectively, providing critical structural information about the polymer.

Fig. 2.

XRD patterns of PANI, Au₂₀@GO/PANI, Ag₁₀@GO/PANI and Au₁₅Ag₁₅@GO/PANI nanocomposites.

In the Au₂₀@GO/PANI sample, distinct peaks characteristic of gold nanoparticles (Au NPs) are visible at 2Ɵ angles of 38° (111), 44° (200), 64° (220), and 78° (311) (corresponding to JCPDS card No. 01–1172). Likewise, in the case of the Ag₁₀@GO/PANI sample, silver nanoparticles (Ag NPs) exhibit their own distinctive peaks at 2Ɵ values of 33° (111), 42° (200), 65° (220), and 77° (311) (in accordance with JCPDS card NO. 04–0783) [41]. These specific peaks affirm the presence and crystalline nature of both Au and Ag nanoparticles within the nanocomposites. Notably, the formation of these characteristic peaks in the XRD patterns signifies the successful integration of Au and Ag nanoparticles during in-situ PANI synthesis.

Furthermore, an additional peak at approximately 2Ɵ = 10° is evident, indicating an interlayer distance of 0.88 nm within the graphene oxide (GO) sheets, as calculated using Bragg's equation. This finding underscores the structural arrangement of GO within the nanocomposites and its role in the overall composite structure.

It is worth noting that cetyltrimethylammonium bromide (CTAB), a surfactant, displays characteristic peaks at 17°, 20°, and 23°, which, notably, overlap with some of the peaks associated with PANI [42]. The coexistence of these peaks is highlighted in Fig. 2, underscoring the presence and influence of CTAB within the nanocomposite structure, which can have implications for the composite's properties and performance.

The crystallinity and structure of four nanocomposite samples, namely Au₂₀@GO/PANI, Ag₁₀@GO/PANI, PANI, and Au₁₅Ag₁₅@GO/PANI, were investigated through XRD (X-ray diffraction) analysis. In this study, the Full Width at Half Maximum (FWHM) values obtained from the XRD tests were crucial indicators of the nanocomposites' structural characteristics. For Au₂₀@GO/PANI, the FWHM value of 5.1 suggests a moderate crystallinity, indicating a well-defined crystal structure with some degree of disorder. The presence of gold nanoparticles (Au NPs) on graphene oxide (GO) and polyaniline (PANI) may contribute to the observed crystalline features. Similarly, Ag₁₀@GO/PANI exhibits a FWHM value of 5.3, indicating a comparable crystallinity to Au₂₀@GO/PANI. The incorporation of silver nanoparticles (Ag NPs) into the nanocomposite, along with GO and PANI, contributes to the overall crystalline structure. PANI, with a tubular structure when used without nanoparticles, exhibits a FWHM value of 5.5. The slightly higher FWHM suggests a moderate crystallinity. On the other hand, Au₁₅Ag₁₅@GO/PANI exhibits a FWHM value of 4.3, indicating a slightly narrower peak compared to PANI. This suggests a higher crystallinity for Au₁₅Ag₁₅@GO/PANI, possibly influenced by the combination of gold and silver nanoparticles along with GO and PANI. In summary, the XRD results suggest that the nanocomposites Au₂₀@GO/PANI and Ag₁₀@GO/PANI have similar crystallinities, while PANI exhibits moderate crystallinity. The Au₁₅Ag₁₅@GO/PANI nanocomposite shows a slightly higher crystallinity compared to PANI, possibly due to the combination of different metal nanoparticles. These findings provide valuable insights into the structural characteristics of the nanocomposites, paving the way for a deeper understanding of their properties and potential applications.

3.3. Elemental and structural analysis of prepared nanocomposites

The utilization of Field Emission Scanning Electron Microscopy (FESEM) images has been instrumental in highlighting the significant impact of various volumes of cetyltrimethylammonium bromide (CTAB) (at a concentration of 15 mM) on the structural configuration of polyaniline (PANI), as depicted in Fig. 3. The introduction of CTAB, which acts as a soft template, has brought about notable structural transformations in PANI, which can be favorably compared to the unaltered PANI structure presented in Fig. 3. In the absence of CTAB, PANI exhibits an amorphous morphology devoid of any tubular structure. However, with the incorporation of CTAB, PANI undergoes a remarkable transformation, adopting a more organized and structured configuration. Notably, at lower CTAB volumes, the dominance of the tubular structure is less pronounced, a consequence of the diminished CTAB concentration. Optimal results, in terms of achieving a well-defined tubular PANI structure, are observed when CTAB is added in the range of 10–20 mL (at 15 mM concentration), as described in Fig. 3.

Fig. 3.

SEM images of PANI with adding different volumes a) 5 mL, b) 10 mL, c) 15 mL, d) 20 mL, e) 25 mL of CTAB and f) no CTAB, during PANI polymerization.

The structural configurations of the synthesized nanocomposites are illustrated in Fig. 3a-c, while Fig. 3f represents the PANI structure with no CTAB. Fig. 3 illustrates SEM images of PANI with adding different volumes of 5 mL (a), 10 mL (b), 15 mL (c), 20 mL (d), 25 mL (e) of CTAB and f) no CTAB, during PANI polymerization. The underlying principle governing the effect of CTAB on PANI transformation is elucidated in Scheme 2. CTAB, a surfactant, is characterized by a positively charged polar head and a nonpolar tail. In an aqueous environment, CTAB molecules form micellar structures. During the synthesis of PANI, these CTAB micelles aggregate into a rod-like arrangement, while aniline monomers are attracted to the amine groups of CTAB, initiating the polymerization process. As the polymerization proceeds, the PANI encapsulates the rod-shaped CTAB micelles. Subsequent washing steps with water, acetone, and HCl solution lead to the removal of CTAB molecules, leaving behind a tubular PANI structure. This tubular PANI exhibits enhanced conductivity and superior dispersibility in water, distinguishing it from the amorphous PANI.

Scheme 2.

Tubular polyaniline synthesis using CTAB as a soft template.

With the optimal concentration of CTAB for tubular PANI synthesis established, solutions containing Au₂₀@GO, Ag₂₀@GO, and Au₁₅Ag₁₅@GO core/shell nanoparticles (NPs) were introduced with 10 mL of CTAB to investigate their influence on the PANI structure. The addition of these prepared nanocomposites has had a pronounced positive effect on the development of tubular PANI structures. In the Au₂₀@GO/PANI sample, certain amorphous regions were initially evident, but these disappeared upon the introduction of Ag NPs (as depicted in Fig. 4). Fig. 4 illustrates FESEM images of Au@GO/PANI (a), Ag@GO/PANI (b), AuAg@GO/PANI nanocomposites (c), close look of Au@GO/PANI (a'), Ag@GO/PANI (b'), AuAg@GO/PANI nanostructure (c'), EDX patterns of Au@GO/PANI (a''), Ag@GO/PANI (b''), and AuAg@GO/PANI (c''). This intriguing phenomenon can be attributed to the strong interaction between the PANI backbone and the π-π bonds in graphene oxide [43,44]. When core/shell nanoparticles are introduced to aniline during the synthesis process, they engage with the polymer backbone, as elucidated in Scheme 3. Particles that are not absorbed by the PANI backbone are effectively eliminated through repeated washing steps. Elemental mapping has corroborated the presence of gold and silver nanoparticles, revealing a uniform dispersion of these NPs within the samples.

Fig. 4.

FESEM images of a) Au₂₀@GO/PANI b) Ag₁₀@GO/PANI, c) Au₁₅Ag₁₅@GO/PANI nanocomposites, close look of a') Au₂₀@GO/PANI, b') Ag₁₀@GO/PANI, c') Au₁₅Ag₁₅@GO/PANI nanostructure, EDX patterns of a'') Au₂₀@GO/PANI, b'') Ag₁₀@GO/PANI and c'') Au₁₅Ag₁₅@GO/PANI.

Scheme 3.

Schematic polymerization mechanism of emeraldine salt of polyaniline in the presence of Au@GO nanoparticles using APS.

Additionally, we investigated the presence of oxygen atoms to assess the dispersion of graphene oxide (GO), revealing that oxygen atoms were uniformly distributed near gold and silver nanoparticles. This distribution is depicted in Fig. 5, which illustrates elemental map images of: a) silver (Ag), b) gold (Au), c) oxygen (O) atoms, d) carbon (C), e) silver and gold atoms (Ag and Au), and f) silver, gold, oxygen, and carbon atoms (Ag, Au, O, and C) in the Au₁₅Ag₁₅@GO/PANI nanocomposite. Energy-Dispersive X-ray (EDX) analysis further confirmed the presence of gold, silver, and oxygen atoms, substantiating the integration of Au@GO and Ag@GO structures within the PANI nanocomposites.

Fig. 5.

Illustrates elemental map images of a) Ag, b) Au, c) oxygen (O) atoms, d) carbon (C), e) Ag and Au atoms, and f) Ag, Au, O, and C atoms in Au₁₅Ag₁₅@GO/PANI nanocomposite.

3.4. Thermogravimetric analysis

Thermogravimetric analysis (TGA) was conducted on both pristine PANI and Au₁₅Ag₁₅@GO/PANI samples, as depicted in Fig. 6a. Existing literature has established that PANI experiences three notable weight loss stages, each occurring at distinct temperature ranges: approximately 100 °C, 300 °C, and 500 °C. These stages correspond to the removal of moisture, the exclusion of doping ions from the polymer structure, and the eventual degradation of the polymer backbone, respectively [45,46].

Fig. 6.

a) TGA curves of PANI and Au₁₅Ag₁₅@GO/PANI nanocomposite, b) TEM image of tubular Au₁₅Ag₁₅@GO/PANI nanostructure.

In the initial temperature range, between 30 °C and 100 °C, the analysis aimed to eliminate any moisture present in both the PANI and nanocomposite samples. The results indicated that Au₁₅Ag₁₅@GO/PANI retained a higher quantity of water molecules. This propensity is attributed to the hydrophilic nature imposed upon the nanocomposite by the presence of Au@GO, Ag@GO, and CTAB components.

Within the temperature range of 400 °C–500 °C, a notable mass loss occurred, signifying the stage of doping in the Au₁₅Ag₁₅@GO/PANI sample. This weight loss is indicative of the expulsion of doping ions as the sample undergoes thermal decomposition.

Subsequently, as the temperature was raised further, approximately between 500 °C and 550 °C, the polymer backbone underwent decomposition, leading to a more pronounced weight loss in the Au₁₅Ag₁₅@GO/PANI nanocomposite. This phenomenon suggests that the bulk structure of PANI exhibited a higher degree of thermal stability in comparison to the nanotube structure of Au₁₅Ag₁₅@GO/PANI. A Transmission Electron Microscopy (TEM) image in Fig. 6b depicts the nanotube structure of Au₁₅Ag₁₅@GO/PANI. This structural analysis provides essential insights into the intricate nanocomposite architectures and their potential applications. The nanotube structure, with its enhanced heat transfer properties, exhibited a more accelerated decomposition under the higher temperature conditions. This insight provides valuable information regarding the thermal behavior and stability of these materials, critical for various potential applications.

3.5. Impact of solvents on PANI solubility and conductivity

The UV–Visible (UV–Vis) spectra of PANI nanotubes incorporated with Au₂₀@GO/PANI, Ag₁₀@GO/PANI, and Au₁₅Ag₁₅@GO/PANI, in three distinct solvents—NMP, water, and DMSO—are presented in Fig. 7a–c, respectively. PANI exhibits two prominent absorption peaks at 315 nm and 455 nm. The absorption at 315 nm corresponds to the π-π* transition within benzenoid rings, while the peak at 455 nm arises from polaron-π* transitions. Additionally, a broader peak within the 700–800 nm range corresponds to π-polaron transitions [8,47,48]. These polaronic transition peaks serve as an indicator of the presence of charge carriers in PANI. The ratio of the polaronic band (750 nm) to the π-π* (315 nm) transition band, denoted as I₇₅₀/I₃₁₅, provides insights into the average oxidation state of PANI. The results, compiled in Table 2 for ease of comparison, reveal that the doping state of the three samples is influenced by the choice of solvents. In NMP, a significant reduction in the I₇₅₀/I₃₁₅ ratio suggests dedoping of PANI, attributed to the formation of hydrogen bonds (C O⋯N–H) between the C O groups present in NMP and the amine groups of PANI [49]. The trend observed for doping preservation in the various solvents is as follows: water > DMSO > NMP. This order suggests that water has the least adverse effect on the doping status of PANI.

Fig. 7.

UV–Vis absorbance and solubility of PANI, Au₂₀@GO/PANI, Ag₁₀@GO/PANI and Au₁₅Ag₁₅@GO/PANI samples in a, a') NMP, b, b') water and c, c') DMSO solvents respectively.

Table 2.

Intensity ratio of I₇₅₀/I₃₁₅ of Au₂₀@GO/PANI, Ag₁₀@GO/PANI, Au₁₅Ag₁₅@GO/PANI samples.

	I750/I315
Sample name	Water	DMSO	NMP
Au20@GO/PANI	1.01	0.52	0.51
Ag10@GO/PANI	0.84	0.47	0.40
Au15Ag15@GO/PANI	1.00	0.75	0.66
PANI	0.31	0.37	0.59

Both gold nanoparticles (Au NPs) and graphene oxide (GO) exhibit unique absorption characteristics. Au NPs display plasmonic absorption at approximately 520 nm, while GO has characteristic peaks at around 250 nm and 300 nm, corresponding to the π-π and π-π* transitions within hexagonal carbonic and oxygen-containing groups, respectively. The relatively strong absorbance of benzenoid and imine groups in PANI, in comparison to the absorption of Au NPs and GO, makes the plasmon absorption less visible.

The solubility of PANI nanocomposites in water, NMP, and DMSO is visually represented in Fig. 7a'-c'. Specifically, 4 mg/mL of Au₂₀@GO/PANI, Ag₁₀@GO/PANI, and Au₁₅Ag₁₅@GO/PANI were dispersed in each of the solvents, followed by stirring for 3 h at room temperature. In NMP and DMSO, the PANI dispersion remained stable for several weeks, indicating good solubility. However, in water, the dispersion settled after 3 h. Initially, all mixtures exhibited a green color, but after 2 days, the mixture of Au₁₅Ag₁₅@GO/PANI in NMP transitioned to a blue color. This color change is indicative of the dedoping of emeraldine salt to emeraldine base, a phenomenon consistent with the UV–Vis results. Conversely, the color of the nanocomposite remained green in DMSO even after 1 week, highlighting its sustained doping state. These solubility and dedoping characteristics hold significant implications for the practical use and processing of these materials.

The emeraldine salt form of PANI is characterized by the presence of one molecule of hydrochloric acid per every two aniline units. This specific composition imparts ionic conductivity to PANI. The acidic hydrogen within the emeraldine salt interacts with the imino groups of PANI, which exhibit a high degree of conjugation. Furthermore, this interaction between polarons and anions, a component of the acid, serves to neutralize positive charges [50,51]. This combination of factors contributes to PANI's ionic conductivity.

The ionic conductivity of the prepared Au₂₀@GO/PANI, Ag₁₀@GO/PANI, and Au₁₅Ag₁₅@GO/PANI samples was assessed in various solvents, and the results demonstrated that the PANI nanocomposites exhibited the highest ionic conductivity in the following order: water > DMSO > NMP. The comprehensive dataset is summarized in Table 3. The comparatively lower ionic conductivity observed in NMP can be attributed to the higher degree of dedoping experienced by PANI, a finding that aligns with the UV–Vis results. This conductivity data provides essential insights into the electrochemical properties and performance of these nanocomposites in different solvent environments, which is vital information for various potential applications.

Table 3.

Ionic conductivity of Au₂₀@GO/PANI, Ag₁₀@GO/PANI, and Au₁₅Ag₁₅@GO/PANI samples in water, DMSO, NMP solvents.

	Conductivity in solvents (μs)
Sample name	Water	DMSO	NMP
Au20@GO/PANI	1667	108	42
Au15Ag15@GO/PANI	1982	91	49
Ag10@GO/PANI	997	140	13

3.6. Gas sensor analysis

To evaluate gas sensing performance, the most promising metal nanocomposites, selected based on their superior conductivity and nanotube structure, were meticulously deposited (using a 50 μL drop-casting method) onto comb-like gold electrodes, as depicted in Scheme 4. The selected nanocomposites, namely Au₂₀@GO/PANI, Ag₁₀@GO/PANI, and Au₁₅Ag₁₅@GO/PANI, were dissolved in water and then meticulously applied to the gold electrodes.

Scheme 4.

Schematic preparation of gas sensor based on prepared metal nanocomposites.

Subsequently, three different gases—nitrogen dioxide (NO₂), carbon dioxide (CO₂), and hydrogen sulfide (H₂S)—were employed to assess the gas sensing efficiency and retention time of the meticulously prepared nanocomposites. The responsiveness of these sensor systems was quantitatively defined through the utilization of Equation (1):

$$\raisebox{1ex}{$\Delta R$}\!\left/ \!\raisebox{-1ex}{$R_0$}\right.\%=\frac{R_g-R_0}{R_0}*100$$ (1)

here, R_g represents the resistance of the sensor when exposed to the target gas, and R₀ is the resistance of the sensor when exposed to air.

The nanotube structure of the PANI film, owing to its significantly larger surface area when compared to shapeless PANI, enhances its interactions with gases [52]. Moreover, the metal nanoparticles play a catalytic role, as evidenced by the observed reduction in sensor resistance upon exposure to NO₂. It is essential to note that the conductivity of PANI is not solely determined by the degree of protonation but is also influenced by the degree of oxidation. Reports in the literature have documented a decline in the resistance of PANI upon contact with NO₂, in line with our understanding of the oxidative nature of NO₂ gas. When an oxidizing agent, such as NO₂, comes into contact with the π-electrons of PANI, electrons are transferred from PANI to NO₂. As a result, the positively charged P-type PANI experiences an increased density of holes relative to electrons. This augmented charge carrier density contributes to higher conductivity in PANI. The presence of metal nanoparticles, with their catalytic properties, can further facilitate charge transfer and enhance response efficiency. The transient resistance profiles of gas sensors utilizing Au₂₀@GO/PANI, Ag₁₀@GO/PANI, and Au₁₅Ag₁₅@GO/PANI in contact with NO₂ are depicted in Fig. 8a. The response and recovery times of PANI and Au₁₅Ag₁₅@GO/PANI sensors towards NO₂ are illustrated in Fig. 9a and a', respectively. Notably, the response time for PANI is 883 s, whereas Au₁₅Ag₁₅@GO/PANI exhibits a significantly reduced response time of 431 s. The recovery time for PANI is 445 s, while Au₁₅Ag₁₅@GO/PANI showcases an even swifter recovery time of 296 s.

Fig. 8.

Resistance of Au₂₀@GO/PANI, Ag₁₀@GO/PANI and Au₁₅Ag₁₅@GO/PANI nanocomposites gas sensors in contact with a) NO₂, b) H₂S, c) CO₂ gases, and d) acetone.

Fig. 9.

Response and recovery times of the Au₁₅Ag₁₅@GO/PANI nanocomposite gas sensors to 1 ppm a) NO₂, b) H₂S, c) CO₂, and d) acetone and PANI gas sensor to 1 ppm a') NO₂, b') H₂S, c') CO_2, and d') acetone.

Furthermore, H₂S was effectively detected by the sensors composed of Au₂₀@GO/PANI, Ag₁₀@GO/PANI, and Au₁₅Ag₁₅@GO/PANI nanocomposites, as illustrated in Fig. 8b. Prior research has indicated that H₂S interacts with gold nanoparticles, forming AuS, which results in the generation of H+ ions (as shown in Equation (2)) and, consequently, a reduction in the resistance of Au₂₀@GO/PANI and Au₁₅Ag₁₅@GO/PANI nanocomposites [53,54]. Silver nanoparticles also interact with H₂S (as described in Equation (3)), although to a lesser extent than the interaction between Au and H₂S. The chemical equations involved are presented below:

These reactions elucidate the mechanistic foundations of H₂S detection and provide insights into the sensor's sensitivity and selectivity.

$$H_{2}S+Au\to AuS+2H^{+}$$ (2)

$$H_{2}S+Ag\to AgS+2H^{+}$$ (3)

The response and recovery times of PANI and Au₁₅Ag₁₅@GO/PANI sensors towards H₂S are illustrated in Fig. 9b and b', respectively. Notably, the response time for PANI is 929 s, whereas Au₁₅Ag₁₅@GO/PANI exhibits a significantly reduced response time of 335 s. The recovery time for PANI is 807 s, while Au₁₅Ag₁₅@GO/PANI showcases an even swifter recovery time of 277 s. These results unequivocally demonstrate that the inclusion of Au and Ag nanoparticles enhances the sensing capabilities of the nanocomposite when compared to pristine PANI.

Furthermore, the sensors based on Au₂₀@GO/PANI, Ag₁₀@GO/PANI, and Au₁₅Ag₁₅@GO/PANI were evaluated for their response to carbon dioxide (CO₂), revealing superior sensing performance compared to PANI-based sensors. The underlying sensing mechanism for PANI and PANI nanocomposites in the context of CO₂ detection hinges on two key mechanisms. First, primary amines are capable of interacting with CO₂ to form carbamates. To effectively trap CO₂, the involvement of two primary amines is required, as illustrated in Equation (4):

$${CO}_2+2RN{H}_2\leftrightarrow {RNHCOO}^{-}+{RNH}_3^{+}$$ (4)

This mechanistic insight underscores the sensitivity of the nanocomposite sensors to CO₂, elucidating the improved sensing efficiency of Au₂₀@GO/PANI, Ag₁₀@GO/PANI, and Au₁₅Ag₁₅@GO/PANI in comparison to their PANI-based counterparts.

In this intricate process, a pair of amine electrons forms a bond with the carbon atom of CO2, leading to the creation of a zwitterion. Subsequently, another amine group dissociates a proton from the zwitterion, resulting in the formation of carbamate [55,56]. This dual interaction signifies a reduction in the number of free amine groups and, in turn, a decrease in the charge density of protons. Consequently, the net effect is an increase in electrical resistance.

Alternatively, the second scenario unfolds with the formation of a p-n junction between the Au and Ag nanoparticles (NPs) and PANI. In this instance, a depletion layer is established at the interface of the metal NPs and PANI. When CO₂ is introduced into the chamber, PANI nanocomposites adsorb CO₂ molecules, liberating electrons within the junction. This phenomenon augments the thickness of the depletion layer, subsequently elevating the electrical resistance [54]. Fig. 8c represents the transient resistance profiles of gas sensors employing nanocomposites in response to CO₂. The response and recovery times for Au₁₅Ag₁₅@GO/PANI and PANI toward CO₂ are represented in Fig. 9c and c', respectively, and notably, Au₁₅Ag₁₅@GO/PANI achieves a swifter response time of 279 s and a more rapid recovery time of 366 s, compared to the corresponding times of 592 s and 789 s for PANI alone. These findings explicitly underscore the enhanced response characteristics of the nanocomposite in contrast to PANI-based sensors, with both mechanisms contributing to their improved sensing efficiency. Acetone was also monitored and Au₂₀@GO/PANI, Ag₁₀@GO/PANI and Au₁₅Ag₁₅@GO/PANI based sensors showed better sensing in comparison with PANI-based sensor as shown in Fig. 8d. The transient resistance variation of the Sensors based on Au₂₀@GO/PANI, Ag₁₀@GO/PANI, and Au₁₅Ag₁₅@GO/PANI to different concentrations of acetone vapor are shown in Fig. 7d. Previous results investigated that some analytic gases, such as ketones, may change the crystallinity of conducting polymers [57]. The strong interaction between small ketones like acetone and nitrogen atoms of polyaniline led to the expansion of polyaniline chains to more stretching conformation [58]. As a result, polymer crystallinity increases, and electrical resistance declines. Fig. 9d and d' compare response and recovery times of PANI and Au₁₅Ag₁₅@GO/PANI sensors toward acetone which are 867 s, 672 s, and 193 s, 509 s, respectively.

For enhanced comparability, all the sensor responses for the prepared sensors have been combined in Fig. 10. Fig. 10 illustrates selectivity of the PANI, Au₂₀@GO/PANI, Ag₁₀@GO/PANI and Au₁₅Ag₁₅@GO/PANI nanocomposite gas sensors toward 4 ppm of NO₂, H₂S, acetone, and CO₂ gases. Standard deviation (SD) values of the set of sensors are calculated using the following equation (5):

$$\text{SD}=\surd \frac{\sum {\left(Xi-X\right)}^2}{N}$$ (5)

Where X_i is each individual baseline measurement, X is the mean of the baseline measurements, and N is the number of baseline measurements. According to the calculations, the Au₂₀@GO/PANI has SD value of 0.3, 0.1, 0.2, and 0.2 in contact with NO₂, CO₂, H₂S, and acetone, respectively. Ag₁₀@GO/PANI has SD value of 0.3, 0.05, 0.2, and 0.2 in contact with NO₂, CO₂, H₂S, and acetone, respectively. Au₁₅Ag₁₅@GO/PANI nanocomposite gas sensor has SD value of 0.3, 0.1, 0.2, and 0.1 in contact with NO₂, CO₂, H₂S, and acetone, respectively. PANI gas sensor has SD value of 0.3, 0.07, 0.05, and 0.1 in contact with NO₂, CO₂, H₂S, and acetone, respectively.

Fig. 10.

Selectivity of the PANI, Au₂₀@GO/PANI, Ag₁₀@GO/PANI and Au₁₅Ag₁₅@GO/PANI nanocomposite gas sensors toward 4 ppm acetone, NO₂, H₂S, and CO₂ gases.

For further analysis, detection limit (DL) of the sensors were measured using the following equation (6):

DL = K × SD (6)

In the abovementioned equation (6) K is a constant (K = 3). To obtain DL of the sensors, 4 blank samples were tested using each sensor. SD values of the sensors using blank sample were determined. The blank sample is a sample without analyte. DL of the sensors were calculated using SD values of the sensors. DL of the Au₂₀@GO/PANI gas sensor, Ag₁₀@GO/PANI gas sensor, and Au₁₅Ag₁₅@GO/PANI nanocomposite gas sensor are 0.6, 0.45, and 0.3, respectively.

Table 4 illustrates a comparison table showing response and recovery times of nanocomposites toward CO₂, NO₂, and H₂S gases studied by other researchers. As shown in Table 4, the present sensor based on Au₁₅Ag₁₅@GO/PANI shows detection range of 0.6–4, 1–4, 0.6–4 (ppm) for detecting NO₂, CO₂, and H₂S, respectively. Response time and recovery times of the sensor based on Au₁₅Ag₁₅@GO/PANI are compatible with other researches in this field.

Table 4.

Comparison table showing response and recovery times of nanocomposites toward CO₂, NO₂, acetone, and H₂S gases studied by other researchers.

Gas	Sample	Detection range (ppm)	Response time	Recovery time	Operating temperature	Reference
NO2	Au NP-MWCNTs	0.1–1	>600	-	RT.	[59]
	Au NP-rGO	0.5–5	132	386	50°C	[60]
	Au15Ag15@GO/PANI	0.6–4	431	296	RT.	present paper
CO2	PANI/chloroalumium phthalocyanine	0.05–7.20	250	200	27°C	[61]
	Amino-functionalized Gr/PANI nanofiber	20–2000	10	20	RT.	[62]
	Au15Ag15@GO/PANI	1–4	279	366	RT.	present paper
H2S	PANI/CuCl2	2.5–100	1000	-	20°C	[63]
	PANI/Au	0.1–100 (ppb)	<120	<300	RT.	[53]
	Au15Ag15@GO/PANI	0.6–4	335	277	RT.	present paper
Acetone	ZnO/graphene quantum dots (GQDs)/PANI	500	15	27	25°C	[57]
	Au15Ag15@GO/PANI	0.6–4	193	509	RT.	This paper

Fig. 11 presents the calibration curves depicting the responses of the set of sensors when in contact with various gases such as NO₂, CO₂, acetone, and H₂S. These curves serve as a visual representation of how the sensors' responses vary with different concentrations of each gas.

Fig. 11.

Calibration curves of the Au₂₀@GO/PANI, Ag₁₀@GO/PANI and Au₁₅Ag₁₅@GO/PANI nanocomposite gas sensors' response in contact with a) NO₂, b) H₂S, c) CO₂, and d) acetone.

In addition to the curves, the accompanying equations of the linear trend lines are displayed on the graph. These trend lines help quantify the relationship between the sensor responses and the concentration of the gases. The equations are instrumental in determining the sensitivity of each sensor. Sensitivity is derived from the slope of the linear trend line in the calibration curve.

To calculate sensitivity, the following general formula can be employed:

Sensitivity = Change in Sensor Response/Change in Gas Concentration

The calculated sensitivity values for each sensor are then utilized to assess their performance in detecting different gases. The Au₂₀@GO/PANI gas sensor has sensitivity of 2.4085, 0.8303, 0.6221, and 1.1965 toward NO₂, CO₂, acetone, and H₂S, respectively. The Ag₁₀@GO/PANI gas sensor has sensitivity of 1.9194, 0.4117, 0.515, and 1.4341 toward NO₂, CO₂, acetone, and H₂S, respectively. The Au₁₅Ag₁₅@GO/PANI gas sensor has sensitivity of 4.2649, 0.9377, 1.4902, and 2.828 toward NO₂, CO₂, acetone, and H₂S, respectively. These sensitivity values offer insights into the sensors' ability to detect and respond to varying concentrations of specific gases. Higher sensitivity values generally indicate a more responsive sensor, making them crucial parameters for evaluating the efficacy of gas sensors in practical applications. As results show, Au₁₅Ag₁₅@GO/PANI gas sensor shows the highest sensitivity toward gases in comparison to the two other gas sensors. The results also indicates higher sensitivity of Au₁₅Ag₁₅@GO/PANI gas sensor toward NO₂.

4. Conclusion

We have successfully developed an efficient in-situ method for preparing Au@GO/PANI nanostructures, with CTAB acting as a soft template to precisely control the tubular structure of PANI. We conducted thorough analyses of the nanocomposite's dispersibility in various solvents, noting the highest ionic conductivity of 1982 μs in a nanocomposite/water dispersion, which significantly decreased to 49 μs in N-Methyl-2-pyrrolidone (NMP), confirming the dedoping effect of NMP. Importantly, the AuAg@GO/PANI nanocomposite exhibited exceptional sensing capabilities, surpassing PANI-based gas sensor devices in detecting CO₂, NO₂, and H₂S. Notably, at room temperature, the AuAg@GO/PANI nanocomposite displayed rapid response times of 279, 431, and 335 s, along with efficient recovery times of 366, 296, and 277 s when exposed to 1 ppm concentrations of CO₂, NO₂, and H₂S, respectively. In contrast, PANI exhibited significantly slower response and recovery times for the same gases, highlighting the enhancing role of core@shell nanoparticles in gas sensor performance. Additionally, the AuAg@GO/PANI nanocomposite offers an environmentally friendly and cost-effective solution-processing method for fabricating efficient gas sensors. This research underscores the potential of the proposed architecture in various applications, including sensors, optoelectronics, and energy conversion."

CRediT authorship contribution statement

Zahra babaei: Writing – original draft, Methodology, Conceptualization. Bahareh Rezaei: Writing – original draft, Data curation. Ehsan Gholami: Data curation. Faramarz Afshar Taromi: Writing – review & editing, Supervision. Amir Hossein Haghighi: Writing – review & editing, Supervision.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper (article entitled “In situ synthesis of long tubular water-dispersible polyaniline with core/shell gold and silver@graphene oxide nanoparticles for gas sensor application”).

Appendix A. Supplementary data

The following is the Supplementary data to this article.