1 Introduction

The search for a sustainable renewable energy (SRE) source has been a major topic for the 21 st century to replace depleting and greenhouse gases emitting fossil fuels. Among various SRE’s, hydrogen (H2) has been identified as the best candidate to be used as an alternative to the greenhouse gas emission bound fossil fuel-based energy systems [1,2,3,4,5]. This is because of its high energy density, eco-friendliness and readily available nature [6, 7]. H2 gas can be generated from different approaches which include steam reforming of natural gases or fossil fuels, gasification of coal, and water splitting [6, 7]. Water splitting by either electrocatalysis or photocatalysis is favoured as compared to other H2 production methods due to its zero greenhouse gas emissions, which is of great significance to the environment [7]. Photocatalysis water splitting requires an effective semiconductor to harvest the solar light with a band gap closer to the one of hydrogen (1.23 eV) whereas electrochemical water splitting needs an electrocatalyst which can reduce the overpotential [8, 9]. In terms of electrochemical water splitting, the most used electrocatalysts are the scarce and expensive Pt-based [10, 11]. To commercialize electrochemical water splitting, there is a strong need to replace the Pt-based electrocatalysts with cheap and earth-abundant/metal-free electrocatalysts [12, 13]. Earth’s abundant metal-based electrocatalysts, which include metal oxides, carbines and clusters, metal chalcogenides, MXene, metal organic frameworks, metal-free carbon-based materials (graphene, carbon nanotubes and activated carbons) and intrinsic conducting polymers (ICPs) have been reported as an alternative to the Pt-based electrocatalysts [14,15,16,17]. However, materials such as metal dichalcogenides, oxides, carbides suffer from poor conductivity and possess limited electrochemical active sites which hampers their usage in HER [16]. On the other hand, MXene sheets tend to aggregate because of weak Van der Waal forces resulting in inferior HER activities [17].

Among these above-mentioned materials, ICPs such as polypyrrole (Ppy), polythiophene (PT), poly(3,4-ethylenedioxythiophene) (PEDOT) and polyaniline (PANI) have been widely used as electrode materials for electrochemical applications (batteries, supercapacitors and sensors) due to their low cost, ease of synthesis, fast charge/discharge kinetics, high charge density, and environmental stability [18, 19]. Chemical oxidative and electrochemical polymerization methods are the most common methods to produce the ICPs [20, 21]. Amongst these two methods, chemical polymerization is the most preferred as it is simple to undertake and allows large scale production unlike the rapid electrochemical process which does not allow production of large quantities, requires specialized equipment and skilled personnel, high initial setup costs and it is highly dependent on the nature and condition of the electrode [20, 21]. From the mentioned ICPs, PANI has been studied extensively as an electrocatalyst support (electron transport channel) for metal electrocatalysts for various applications including HER because of their π-conjugate structure, proton conductivity and unique redox properties [22]. In addition, the inclusion of nitrogen atoms in its structure aids in the connection of metal ions to the polymer chain which leads to an increase in electron transmission in the structure [22]. Furthermore, the nitrogen element of PANI can readily trap H+ from hydrated ions to form protonated amino groups. This makes it possible for H+ to have a higher positive charge density on the electrode surface, which improves catalytic performance and accelerate the transformation of H+ to H2 for HER resulting in fast kinetics and low overpotential [23].

To our best knowledge, there are few reports on the use of polyaniline and its derivatives as metal-free HER electrocatalysts. This is because of its poor processibility, potent electrostatic interaction between the polymer chain and aromatic structure, mechanical instability resulting in active material loss (PANI peeling off from the solid electrode surface) and that ICPs in general are almost non-conductive in the potential window of HER [24]. This is illustrated by limited studies for the application of polymeric materials as metal-free electrocatalysts. For example, El-Deeb and co-coworkers [25] fabricated GC/PANI from glassy carbon (GC) electrode with polyaniline (PANI) for HER applications in Et3NHCl/[Bu4N][BF4]-CH3CN solution. The prepared GC/PANI exhibited outstanding HER properties and excellent stability. In another study, Aydin and Koleli reported the hydrogen evolution of PANI, Ppy, and Ppy/PANI electrocatalysts deposited on Pt surface [26]. The HER results followed the Volmer-Tafel process in which the Tafel step is the rate-limiting step. Herein, we present the first time preparation of PANI homopolymer and copolymers such as poly(aniline-co-3-aminobenzoic acid) (P(ANI-co-ABA), poly(aniline-co-triphenylaniline) (P(ANI-co-TPA) and poly(aniline-co-3-nitroaniline) (P(ANI-co-3NI) as an electrocatalysts for HER in acidic medium. The PANI homopolymer and copolymers were fabricated using the chemical polymerization route and their structural confirmations were evaluated using various analytical techniques. From the prepared PANI homopolymer and copolymers, it was deduced that P(ANI-co-3NI) exhibited excellent HER performance in acidic medium with a Tafel slope of 47.9 mV.dec−1 and charge transfer resistance of 16.71 Ω. In addition, P(ANI-co-3NI) possessed the highest electrochemical surface area (ECSA) of 28.43 cm2.

2 Materials and methods

2.1 Material

Aniline (C6H5NH2), 3-nitroaniline (H2NC6H4NO2), 3-aminobenzoic acid (H2NC6H4CO2H) iron (III) chloride (FeCl3) and 2,4,6-triphenylaniline (C24H19N) were purchased from Sigma Aldrich, South Africa. Ammonium persulfate ((NH4)2S2O8, APS) was purchased from Riedel-en-Haen. Ethanol (C2H5OH, EtOH) and hydrochloric acid (HCl) were purchased at Rochelle chemicals, South Africa. All reagents were used as received except aniline which was double distilled before use.

2.2 Synthesis of PANI and its copolymers

Polyaniline (PANI) homopolymer and copolymers (Scheme 1) were prepared by chemical oxidation polymerization of aniline [27]. Exactly, 1 ml of double distilled aniline monomer was dissolved in a solution of 10 ml HCl/100 ml deionized water in a 250 ml round – bottom flask. For copolymer synthesis, the exact amount (1:1 ratio) of 3-aminobenzoic acid (ABA), 2,4,6-triphenylaniline (TPA) and 3-nitroaniline (3NI) were added together with aniline during synthesis. The solution was stirred for 30 min at 50 ̊C. Thereafter, 2.40 g of ammonium persulfate (APS), (NH4)2S2O8 and 1.88 g of iron (lll) chloride (FeCl3) were added respectively to the solution. The mixture was stirred for another 3 h at 50 ̊C and the content of the reaction was placed in the oven at 50 ̊C overnight to evaporate the solvents. The remaining content was washed with ethanol and dried at 50 ̊C. To differentiate between the synthesized polymer and copolymers, the products were named as follows: polyaniline homopolymer- PANI and copolymers as P(ANI-co-ABA), P(ANI-co-TPA) and P(ANI-co-3NI) for copolymerization of aniline with ABA, TPA and 3NI, respectively.

Scheme 1
scheme 1

Chemical polymerization of (a) Polyaniline (PANI), (b) Poly(aniline-co-3-aminobenzoic acid) (P(ANI-co-ABA), (c) Poly(aniline-co-2,4,6-triphenylaniline) (P(ANI-co-TPA) and (d) Poly(aniline-co-3-nitroaniline) (P(ANI-co-3NI)

Full size image

2.3 Modification of the working electrode

For the modification of the working electrode, glassy carbon electrode (GCE) with a diameter of 3 mm (area = 0.071 cm2) was used as a working electrode substrate. Prior to the modification, the GCE working electrode was polished with alumina oxide (Al2O3) powder (0.10, 0.05, and 0.003 μm), washed with distilled water and ultrasonicated for 30 min and air dried. To prepare the electrocatalyst ink, 5 mg of the synthesized materials were dissolved in a solution consisting of 800 µl of deionized water, 100 µl of isopropanol and 100 µl of Nafion solution. The mixture was then ultrasonicated for an hour. For the modification, 10 µl of the prepared ink was drop-casted onto the surface of the pre-cleaned GCE working electrode (catalytic loading = 0.704 mg.cm−2) and allowed to dry at room temperature. The modified electrodes were kept safe until use.

2.4 Structural characterization

The IR spectra (for deducing the functional groups of the prepared materials) were acquired from the Spectrum II Perkin Elmer with platinum crystal on the attenuated total reflectance from the range 400–4000 cm−1 at a resolution of 4 cm−1. The X-ray diffraction (D8 Advance, Bruker Germany) equipped with Co radiation with a wavelength of 1.78897 Å coupled with a position-sensitive detector (LYNXEYE) in Bragg Brentano geometry using a power setting of 35 kV and 40 mA was used to acquire the diffraction patterns. The morphologies of the prepared polymers were collected by Auriga Field-Emission Scanning Electron Microscopy (Zeiss) with Energy Dispersive X-ray spectroscopy (EDS) for elemental composition. Thermal analysis was done from Simultaneous Thermal Analyzer (STA) Perkin Elmer 6000 from 30 to 700 °C at a scan speed of 10 °C.min−1 under nitrogen environment. The mass of the polymer materials added to the crucible was approximately 30 mg. The specific surface area and the pore distribution of prepared materials under different synthesis conditions were measured with TriStar II 3020 surface area measurement instrument (Micromeritics Inc. USA) at 77 K. The samples were degassed at 100 °C for 10 h in nitrogen gas. The surface area and pore sizes were determined following the BET and BJH methods, respectively. The ultra-violet spectra were measured from the SP-UV 500 UV/Vis spectrophotometer. The solutions (14 ppm) of the polymers were prepared in DMF solution.

2.5 Electrochemical studies

The electrochemical and HER studies were investigated in the three-electrode system of the Biologic SP150 electrochemical workstation using bare and electrocatalyst modified glassy carbon (GCE), Silver/Silver chloride (Ag/AgCl) and the platinum wire as a working electrode (3 mm diameter, 0.071 cm2 area), reference electrode and counter electrode, respectively. The Ag/AgCl reference electrode potential was calibrated to Reversible Hydrogen Electrode (RHE) using Eq. 1 [28]:

$$\text{ERHE} = \text{EAg}\backslash \text{AgCl} + 0.059 pH + E^\circ\text{Ag}\backslash\text{AgCl}, E^\circ\text{Ag}\backslash\text{AgCl} = 0.1976 V$$
(1)

Determination of electrochemical properties of the materials were performed using cyclic voltammetry (CV) at the scan rate of 0.02–0.10 V.s−1 from the potential of −1.0–1.0 V in 0.50 M H2SO4 (pH = 0) electrolytic solution. The surface coverages (Γ), diffusion coefficients (D) and electrochemical surface areas (ECSA) of the electrocatalysts were estimated using the Eq. 2 [29], 3 [30] and 4 [31], respectively:

$$\:{i}_{p}=\frac{{n}^{2}{F}^{2}A{\Gamma\:}v}{4RT}$$
(2)
$$\:{I}_{p}=\left(2.686\:\times\:\:{10}^{5}\right){n}^{\frac{3}{2}}AC{D}^{\frac{1}{2}}{v}^{\frac{1}{2}}$$
(3)
$$\:ECSA=\frac{{C}_{dI}\times\:A}{{C}_{s}}$$
(4)

where ip, n, F, A, v, R, T, C Cdl and Cs are the peak current, number of electron transfer (n = 2), Faradays constant (F = 96500 C.s), area of the GCE working electrode (A = 0.071 cm2), scan rate, gas constant (R = 8.314 J.mol−1.K−1), temperature (T = 298.15 K), concentration of the electroactive species on the electrode, specific capacitance of GCE to be 0.035 mF.cm−2 (standard for acidic solution) obtained from the literature [31], respectively.

Linear sweep voltammetry (LSV) scans of the GCE modified working electrodes were measured between the applied potential − 1.0 V–0.5 V at a low scan rate of 5 mV.s−1 in 0.5 M H2SO4. The LSV data was fitted into Tafel analysis to determine the Tafel slope, b using the relation in Eq. 5 [28]:

$$\:\eta\:=blog\:\left(j\right)+a$$
(5)

where \(\:\eta\:\) is the overpotential, j defines the current density and a is a constant.

Electrochemical impedance spectroscopy (EIS) Nyquist and Bode plots were determined using EIS between the frequency 100 kHz to 700 mHz at 1.5 V vs. RHE to deduce the chemical and interfacial processes on the surface of electrode [32]. The lifetime (\(\:\tau\:\)) of an electron at a maximum peak frequency (fp) (Hz or s−1) of the electrocatalysts were estimated using the following relations (Eq. 6 [33]).

$$\:\tau\:=\frac{1}{2\pi\:{f}_{p}}$$
(6)

Chronoamperometry (CA) investigations were done for 15 h at 0.36 V. From the CA data, the number of active sites (n) were estimated from the integral area of CA and charge (Q) following the relation (Eq. (7)) [34]:

$$\:n=\frac{Q}{2F}$$
(7)

The turnover frequency (TOF) which estimate the intrinsic property of the electrocatalysts and is known as the total number of H2 molecules produced/evolved in an electrocatalytical active site per second was determined using the Eq. 8 [34]:

$$\:TOF=\frac{I}{2Fn}$$
(8)

3 Results and discussions

3.1 Structural characterizations

The Fourier transform (FTIR) spectra of prepared PANI homopolymer and copolymers (P(ANI-co-ABA), P(ANI-co-TPA) and P(ANI-co-3NI)) are displayed in Fig. 1(a). The spectra of PANI and copolymers showed a characteristic peak at the regions of 1486–1490 and 1564–1625 cm−1 which are attributed to the C = C vibrations of the benzenoid ring and quinoid ring, respectively [35,36,37]. As compared to PANI homopolymer, the intensities of these above-mentioned peaks of copolymers are reduced, and similar observations were reported by Ding et al. [38]. In addition, the peak at around 1287 cm−1 in all synthesized materials corresponds to the N-H bending and the one at 1236 cm−1 is due to the asymmetric constituent of the C-C stretching modes [30, 36, 38]. In copolymers (P(ANI-co-ABA), P(ANI-co-TPA) and P(ANI-co-3NI)), these peaks shifted slightly to higher wavenumbers. The in-phase and out-of-phase C-H bonding modes of the fabricated polymeric materials resonate at around 1034 and 832 cm−1, respectively [30, 37]. In addition, the FTIR spectrum of P(ANI-co-3NI) displayed peaks at around 1510 and 1349 cm−1 which are because of the asymmetric and symmetric stretching modes of the NO2 group of the 3-nitroaniline [39]. Based on the above observation and comparison, the closer resemblance of the FTIR of PANI and copolymers tells that the polymer backbone of the materials are similar and the reduction in intensities of the peaks shows that the fabrication of the copolymers was successful.

Fig. 1
figure 1

The (a) FTIR spectra, and (b) XRD diffraction patterns of PANI and copolymers (P(ANI-co-ABA), P(ANI-co-TPA) and P(ANI-co-3NI)

Full size image

The XRD diffraction patterns of the polyaniline homopolymer and copolymers are shown in Fig. 1(b) below to probe its crystallinity. PANI homopolymer showed two broad diffraction peaks at 2θ = 22 and 29o due to the 100 and 110 phases with the JMPDS card number (00-065-0826) which can be attributed to the amorphous characteristic’s nature of PANI and this is in consistent with our previous works [30, 37, 40, 41]. The XRD of P(ANI-co-TPA) and P(ANI-co-3NI) copolymers showed the peaks which correspond to that of PANI. This confirms the successful copolymerization process and the polymer backbone is not hindered. Interestingly, the XRD of P(ANI-co-ABA) showed a crystalline nature, and this can be seen by the sharp diffraction peaks obtained. The highly ordered diffraction peaks are in agreement with the work by Gao and co-workers [42] and the poly(3-aminobenzoic acid) homopolymer reported by Ozturk et al. [43]. The crystalline nature is attributed to the orthorombic nature of P(ANI-co-ABA) and it is assumed that the copolymer assumes a zigzag polymer chain configuration [42]. In addition, the sharpness of the peaks is attribute to the degree of orientation of the polymer backbone [44]. The Debye-Scherrer equation (Eq. 9) was used to evaluate the crystallite sizes (D) of the as-synthesized materials [44, 45]:

$$\:D=\frac{K\lambda\:}{\beta\:\text{cos}\theta\:}$$
(9)

where, K is the Scherrer constant, β represents the full width at half maximum (FWHM), λ is the wavelength of Co radiation, and θ is the angle between the incident and reflected rays. The D obtained (Table 1) for the as-prepared materials were 7.18 nm for PANI, 140 nm for P(ANI-co-ABA), 6.45 nm for P(ANI-co-TPA) and 21.33 nm for P(ANI-co-3NI). The obtained crystallite sizes were in good agreement with the reported ones in the literature for polymeric materials [44,45,46]. The values of the crystallite sizes obtained were used to calculate the dislocation density (δ) which can be related to the degree of crystallinity of the materials using the relation (Eq. 10) and the values obtained are shown in Table 1 [47]:

$$\:\delta\:=\frac{1}{{D}^{2}}$$
(10)

The dislocation density of PANI is 194 × 1014 lines.m−2, whereas that of P(ANI-co-ABA) is 0.510 × 1014 lines.m−2, 240 × 1014 lines.m−2 for P(ANI-co-TPA) and the one for P(ANI-co-3NI) is 22.0 × 1014 lines.m−2. The dislocation density results are consistent with the reported ones and they show the high degree of crystallinity in the polymeric materials [44]. The interchain separation length (R) which can be associated with the polymeric layer conductivity and also represents the hopping distance of electrons from one chain to another, was determined using the relation (Eq. 11) [44, 45]:

$$\:R=\frac{5\lambda\:}{8\text{sin}\theta\:}$$
(11)

The interchain separation lengths of the polymeric materials are shown in Table 1 and the R value of P(ANI-co-3NI) was found to be 4.07 Å which is the lowest as compared to other materials. It was reported that the conductivity of the polymers increases with the decrease in the interchain separation length [45]. Thus, the conductivity of P(ANI-co-3NI was higher than that other materials. The lattice strain (Ɛ) was calculated with the following relation (Eq. 12) [44, 48]:

$$\:\epsilon\:=\frac{\beta\:cos\theta\:}{4}$$
(12)

The obtained lattice strains (Table 1) were 5.60 × 10−3, 29.9 × 10−3, 6.52 × 10−3 and 1.97 × 10−3 for PANI, P(ANI-co-ABA), P(ANI-co-TPA) and P(ANI-co-3NI), respectively. The lattice strain of P(ANI-co-3NI) was found to be the lowest and this signifies the broadening effect which can be attributed to the crystalline size [44, 45].

Table 1 Structural (XRD and BET) and physicochemical (UV-Vis) properties of the as-synthesized materials
Full size table

The synthesized polymeric materials’ surface area and pore structure features were ascertained using the Brunauer-Emmett-Teller (BET) nitrogen adsorption-desorption isotherms and the Barret-Joyner-Halenda (BJH) pore size distributions, and the produced materials’ apparent hysteresis loops which are displayed in Fig. 2(a). The IUPAC states that these isotherms are typical type IV, indicating the presence of mesoporous structures. Table 1 provides a summary of the materials’ BET surface areas and pore volumes. According to Table 1’s BET data, P(ANI-co-3NI) has a specific surface area of 40.864 m2.g−1, and the BJH technique yielded a pore volume of approximately 1.075 cm3.g−1 whereas PANI homopolymer has a BET surface area of 19.89 m2.g−1 and the pore volume of approximately 1.808 cm3.g−1. In terms of other copolymers, P(ANI-co-ABA) and P(ANI-co-TPA) exhibited the surface area of 10.063 and 32.223 m2.g−1, respectively and the pore volumes of 0.00574 and 0.398 cm3.g−1, respectively. The higher surface area of P(ANI-co-3NI) signifies the increased in the number of active sites which can be beneficial for HER. The thermal stability investigation was done using thermogravimetric analysis (TGA). Figure 2(b) shows the comparison of the weight loss between PANI and copolymers upon heating at the rate of 10 °C per minute in a nitrogen condition with the temperature range of 25–700 °C. PANI and P(ANI-co-3NI) displayed three steps of weight loss in the TGA thermogram, whereas P(ANI-co-TPA) and P(ANI-co-ABA) showed two steps. The first steps, at around 100 °C observed in the thermograms of PANI, P(ANI-co-ABA), P(ANI-co-TPA) and P(ANI-co-3NI), are from the loss of water molecules and release of HCl [43, 49, 50]. The steady degradation at around 120 °C on the thermogram of P(ANI-co-TPA) is due to the loss of oligomer and this tell that the thermal stability of the said copolymer is very low. For P(ANI-co-ABA), the total carbonization was observed at around 325 °C. The second step in the degradation of PANI and P(ANI-co-3NI) was seen at around 325 °C which is attributed to the melting and volatilization of the polymeric backbone in the latter and former materials [51]. The final step, which is due to the full carbonization of the polymeric chains at higher temperatures, is observed at temperatures higher than 650 °C [49, 50]. By analysing the thermograms of PANI and P(ANI-co-3NI), it can be deduced that the structural make-up and orientation of the two materials resemble each other, and this is in agreement with the FTIR.

Fig. 2
figure 2

The (a) BET surface areas and (b) TGA thermograms of PANI and copolymers (P(ANI-co-ABA), P(ANI-co-TPA) and P(ANI-co-3NI)

Full size image

3.2 Scanning electron microscope

Fig. 3 shows the SEM images of the as-prepared PANI, P(ANI-co-ABA), P(ANI-co-TPA) and P(ANI-co-3NI) materials. The microscopic images of PANI (Fig. 3(a)), P(ANI-co-ABA) (Fig. 3(b)), P(ANI-co-TPA) (Fig. 3(c)) and P(ANI-co-3NI) (Fig. 3(d)) demonstrate the agglomerated, irregular granular particles in their amorphous nature. This is in agreement with the XRD patterns of the prepared materials, which were amorphous for PANI and other copolymers except for P(ANI-co-ABA) which was crystalline. The EDS analysis was done in order to deduce the elemental composition of the prepared materials, and the results are shown in Fig. 4(a-d) and the atomic percentages in the within the polymer materials synthesized are given in Table 2. The EDS analyses of the prepared materials display the presence of nitrogen (N), oxygen (O), sulphur (S), chlorine (Cl) and iron (Fe) with different atomic percentages and observations were reported by Djara et al. [51]. The presence of N and O are within the polymer backbone because of the type of monomers used during the oxidative chemical polymerization. The existence of S, Cl and Fe are from the bi-oxidants (APS and FeCl3) and the acid dopant (HCl) used in the synthesis. It is reported that during polymerization process, the Cl from the HCl can form the counterion of protonic doping and Fe act as a Lewis acid to form a Lewis adduct with the N atom on the polymer backbone and this can improve the conductivity of the polyaniline [52].

Fig. 3
figure 3

SEM micrographs of (a) PANI, (b) P(ANI-co-ABA), (c) P(ANI-co-TPA) and (d) P(ANI-co-3NI)

Full size image
Fig. 4
figure 4

EDS elemental composition of (a) PANI, (b) P(ANI-co-ABA), (c) P(ANI-co-TPA) and (d) P(ANI-co-3NI)

Full size image
Table 2 Atomic percentages of prepared materials from the EDS analysis
Full size table

3.3 Optical studies

The fabricated PANI homopolymer and copolymers were subjected to physicochemical characterization by using UV/Vis spectroscopy and the results are shown in Fig. 5. The first absorption peak (Fig. 5 (a)) observed between 300–360 nm is attributed to π-π* transitions in the benzenoid ring, which are associated with the conjugation between the rings in the polymeric chains [38, 51, 53]. With respect to PANI homopolymer, the peak on P(ANI-co-TPA) showed a hypochromic effect (a decrease in the absorption intensity) and this can be due to the steric repulsion which resulted from the conjugation of the bulk groups attached to the monomer, 2,4,6-triphenylaniline. For P(ANI-co-ABA), the peak showed the hypsochromic effect and appeared around at 300 nm. The hyperchromic was observed when it comes to the absorption spectrum of P(ANI-co-3NI) which can be attributed to the NO2 group, which is electron withdrawing, enhancing the excitation of electrons. Figure 5 (b) and 4 (c) present the indirect and direct Tauc plot variations to determine the band gap and the values are shown in Table 1. The indirect band gaps were found to be 2.04, 2.58, 2.30 and 1.80 eV for PANI, P(ANI-co-ABA), P(ANI-co-TPA) and P(ANI-co-3NI), respectively. In addition, PANI, P(ANI-co-ABA), P(ANI-co-TPA) and P(ANI-co-3NI), exhibited a direct band gaps of 2.44, 2.87, 2.60 and 1.73, respectively. Based on these deductions, it can be concluded that the electric conductivity of P(ANI-co-3NI) is higher than that of PANI and other copolymers. Niyitanga and Jeong reported that the smaller band gap is essential for HER electrocatalysts as it suggest that the material possesses higher electrical conductivity [54]. Based on that assentation, it can be concluded that the HER properties of P(ANI-co-3NI) are enhanced as compared to other polymeric materials as it exhibited smaller optical band gap.

Fig. 5
figure 5

(a) The comparison absorption spectra (b) and (c) are direct and indirect Tauc plots of PANI, P(ANI-co-3NI), P(ANI-co-ABA) and P(ANI-co-TPA) in 14.00 ppm of DMF

Full size image

3.4 Electrochemical characterization

The oxidation-reduction (redox) and electrochemical studies of the prepared materials were investigated by employing cyclic voltammetry at different scan rates in 0.50 M H2SO4 electrolytic medium on a GCE working electrode substrate. Comparison CV voltammograms of the prepared materials are displayed in Fig. 6(a) and the individual materials are shown in Fig. 6(b-f). From Fig. 6(a) and (b), it can be observed that the CV of bare GCE shows no apparent cathodic peak current, with an anodic peak current appearing at around − 0.50 V. The non-appearance of the redox peaks is due to the inert characteristic of bare GCE. After modifications with the electrocatalysts, there was an enhancement in the current of the modified GCE. The improvement in the electrochemical behaviour is due to the presence of electroactive species adsorbed on the surface of GCE. It is noteworthy to illustrate that the CV comparison of P(ANI-co-3NI) showed the highest current response as compared to PANI, P(ANI-co-ABA and P(ANI-co-TPA). The increase in the current response in the P(ANI-co-3NI) can be supported by the optical studies which showed high electrical conductivity on the mentioned copolymer as compared to PANI homopolymer and other copolymers. To fully comprehend the electrochemical behaviour of the bare GCE and as-synthesized polymeric electrocatalysts, individual CV voltammograms were plotted and shown in Fig. 6(b-f). As mentioned above, the cyclic voltammogram of bare GCE (Fig. 6(b)) showed no cathodic peak current with the only peak appearing at around − 0.50 V, which may be attributed to unknown species adsorbed on the surface. The CV of PANI Fig. 6(c)) showed no distinct reversible cathodic and anodic peak currents which are normally observed in the cyclic voltammogram of PANI. Similar observations were seen on the cyclic voltammogram of P(ANI-co-ABA) Fig. 6(d)). The voltammogram of P(ANI-co-TPA) (Fig. 6(e)) showed appearance of small double cathodic peaks at around 0.2627 and 0.7092 V and corresponding anodic peaks at around 0.1256 and 0.5617 V. The peak current ratios (ip.a./ipc) for the obtained peaks were − 0.9060 and − 0.3517, respectively. The obtained CV of P(ANI-co-TPA) was totally different from the CV of 2,4,6-triphenylaniline reported by Lund [55]. The difference in the CV is a clear indication that there was a structural change when comparing, which shows that there was copolymerization between aniline and 2,4,6-triphenylnitrobenzene. For P(ANI-co-3NI) (Fig. 6(f)), the cyclic voltammogram showed improved electrochemical response as compared to other materials. The first redox peaks on the CV of P(ANI-co-3NI) at cathodic peak current at 0.3525 V and the corresponding anodic peak at 0.1287 V with the ip.a./ipc of −0.7703 are attributed to the generation of the monomer radical cations, and the second redox peaks at cathodic peak current at 0.6907 V and the corresponding anodic peak at 0.5177 V with the ip.a./ipc of 0.4398 are due to the coupling of radical cations to generate copolymer [56]. The highest occupied molecular orbital energy (EHOMO) and the lowest unoccupied molecular orbital energy (ELUMO) were estimated to deduce the electronic properties of PANI and its copolymers. The EHOMO and ELUMO were determined using Eqs. 13 and 14, where Eox and Ered represent the potential for oxidation and reduction, respectively [57].

$$\:{E}_{HOMO}=-\left[{E}_{ox}+4.75\right]eV$$
(13)
$$\:{E}_{LUMO}=-\left[{-(E}_{red}+4.75)\right]eV$$
(14)

The calculated EHOMO were − 5.317, −5.350, −5.013 and − 5.095 eV and ELUMO were − 4.611, −4.608, −4.624 and − 4.621 eV for PANI, P(ANI-co-ABA), P(ANI-co-TPA and P(ANI-co-3NI), respectively (Table 3). The high values of EHOMO and the low values of ELUMO on the electrocatalysts suggest that the electrocatalysts have electron-donating ability and electron-accepting ability, respectively [58]. There are various reactivity indices such as global hardness (η), which tells about the chemical reactivity; electronegativity (χ) which is the power of a material to attract electrons; electrophilicity index (ω) which is defined as the ability of the material to accept electrons; chemical potential (µ) which is the electron escaping tendencies at equilibrium, as well as global softness (σ) which can be used to determine the electronic properties of the electrocatalysts [46, 59,60,61]. The reactivity indices were estimated experimentally from the EHOMO and ELUMO using the Koopmans theorem [62, 63] and the obtained parameters are tabulated in Table 3. The global hardness of the P(ANI-co-3NI) was less than that of other materials, and this indicates that the reactivity of said copolymer was greater than that of other materials. This is supported by the low optical and electrochemical band gaps. The electronegativities of the as-synthesized materials were found to be in similar ranges, and this tells us that upon copolymerization, the ability of the materials to pull electrons towards themselves was not compromised. Furthermore, from the estimation of chemical potential and electrophilicity indices, the materials showed greater electron accepting and donating ability as observed in the EHOMO and ELUMO. From these obtained chemical reactivity parameters, it can be concluded that the electrochemical properties of the PANI were enhanced upon copolymer formations.

Fig. 6
figure 6

(a) Comparison cyclic voltammogram of GCE and prepared materials and CV of (b) GCE, (c) PANI, (d) P(ANI-co-ABA), (e) P(ANI-co-TPA and (f) P(ANI-co-3NI) in 0.5 M H2SO4 at 0.10 V.s−1 scan rate

Full size image
Table 3 Electrochemical parameters of the prepared materials
Full size table
$$\:{E}_{g}={\varvec{E}}_{\varvec{H}\varvec{O}\varvec{M}\varvec{O}}-{\varvec{E}}_{\varvec{L}\varvec{U}\varvec{M}\varvec{O}};\:\varvec{\chi\:}=-\raisebox{1ex}{$1$}\!\left/\:\!\raisebox{-1ex}{$2$}\right.\left[{\varvec{E}}_{\varvec{H}\varvec{O}\varvec{M}\varvec{O}}+{\varvec{E}}_{\varvec{L}\varvec{U}\varvec{M}\varvec{O}}\right]\:;$$
$$\:\:\varvec{\mu\:}=-\varvec{\chi\:}=\raisebox{1ex}{$1$}\!\left/\:\!\raisebox{-1ex}{$2$}\right.\left[{\varvec{E}}_{\varvec{H}\varvec{O}\varvec{M}\varvec{O}}+{\varvec{E}}_{\varvec{L}\varvec{U}\varvec{M}\varvec{O}}\right]\:;\:\varvec{\eta\:}=\raisebox{1ex}{$1$}\!\left/\:\!\raisebox{-1ex}{$2$}\right.\left[{\varvec{E}}_{\varvec{H}\varvec{O}\varvec{M}\varvec{O}}-{\varvec{E}}_{\varvec{L}\varvec{U}\varvec{M}\varvec{O}}\right]\:;\:$$
$$\:\varvec{\omega\:}=\raisebox{1ex}{${\varvec{\mu\:}}^{2}$}\!\left/\:\!\raisebox{-1ex}{$2\varvec{\eta\:}$}\right.\:;\:\varvec{\Delta\:}{\varvec{N}}_{\varvec{M}\varvec{A}\varvec{X}}=\raisebox{1ex}{$-\varvec{\chi\:}$}\!\left/\:\!\raisebox{-1ex}{$\varvec{\eta\:}$}\right.$$

The influence of scan rate on the current response in electrochemistry was conducted at a scan rate of 0.02–0.10 V.s−1 in 0.50 M H2SO4 and the results are displayed in Fig. 7. Figure 7(a-d) illustrates that the increase in the scan rate results in a rise in both cathodic and anodic peak currents, which indicative of adsorption occurring on the electrode surface. The plot of log scan rate against log current was plotted to deduce if there is either a diffusion-controlled or adsorption-controlled process on the electrode/electrolyte interface, and it was reported that the slopes near 0.50 show that the process is diffusion-controlled while the slopes near 1.00 describe the adsorption-controlled process [64]. The slopes obtained in these studies ranges between 0.670 and 0.861 (Fig. 8(a)). PANI, P(ANI-co-ABA), P(ANI-co-TPA) and P(ANI-co-3NI) have the slopes of 0.670, 0.674, 0.861 and 0.773, respectively and this shows that these materials possesses both diffusion- and adsorption-controlled processes [64, 65]. The surface density of the electrochemical reactive species on the GCE electrode surface was determined by calculating the surface coverage (Г) following Eq. 2 [29]. From the slope of the plot of current versus scan rate displayed in Fig. 8(b), the surface density/concentration of P(ANI-co-3NI) was found to be 7.452 × 10−10 mol.cm−2 which is higher than the one of PANI (1.567 × 10−10 mol.cm−2), P(ANI-co-ABA) (3.368 × 10−10 mol.cm−2), P(ANI-co-TPA) (0.8034 × 10−10 mol.cm−2) and 2.685 × 10−14 mol.cm−2 for bare GCE. To determine the diffusion coefficient (D) of the as-synthesized materials, the plot of cathodic peak currents as a function of the square root of scan rate (Fig. 8(c)) was plotted and the Randles-Sevcik (Eq. 3) [30] was used. The determined D were found to be 4. 804 × 10−8 cm2.s−1 (PANI), 21.05 × 10−8 cm2.s−1 (P(ANI-co-ABA)), 2.450 × 10−8 cm2.s−1 (P(ANI-co-TPA)) and 108.8 × 10−8 cm2.s−1 (P(ANI-co-3NI)). The increase in diffusion coefficient suggests that the diffusional transport was increased upon copolymerization with 3-nitroaniline. The electrochemical surface areas (ECSA) of the prepared materials were estimated from the capacitance double layer (Cdl) (Fig. 8(d)). It was reported that the increase in the Cdl of the material suggests that more active sites were exposed [31]. From the study, Cdl of the GCE, PANI, P(ANI-co-ABA), P(ANI-co-TPA) and P(ANI-co-3NI) were found to be 0.160, 1.159, 2.368, 0.8034 and 4.914 mF.cm−2, respectively. The estimated ECSA determined using Eq. 4 were found to be 2.351 cm2 (PANI), 4.804 cm2 (P(ANI-co-ABA)), 1.630 cm2 (P(ANI-co-TPA)) and 9.968 cm2 (P(ANI-co-3NI)). The increase in ECSA of the P(ANI-co-3NI) suggests that more active sites were exposed, and this is a significant parameter in HER [66].

Fig. 7
figure 7

Multiscan cyclic voltammograms of (a) PANI, (b) P(ANI-co-ABA), (c) P(ANI-co-TPA) and (d) P(ANI-co-3NI)

Full size image
Fig. 8
figure 8

(a) Log of current as a function of log of scan rate, (b) current versus scan rate, (c) Randles-sevcik plot and (d) capacitance double layer determination plot of the as-synthesized materials

Full size image

3.5 Electrocatalytic HER studies

After investigating the electrochemical properties of the as-synthesized polymeric materials, their HER performance was evaluated using LSV, EIS, and chromoamperometry (CA). The obtained LSV polarization curves are shown in Fig. 9(a). The LSV for pure GCE showed a poor electrocatalytic response as compared to modified GCE electrodes. This can be attributed to the low ECSA, which resulted in poor intrinsic properties of the pure GCE. For the modified GCE, the modified electrodes exhibited similar onset potentials in the range of 0.30–0.31 V (vs. RHE). It is noteworthy to state that the current density response of the prepared electrocatalysts is less than the benchmark 10 mA.cm−2 [67]. The less than benchmark response may be attributed to limited ECSA as well as the absence of metallic impregnation of the polymeric materials. However, P(ANI-co-3NI) was able to achieve the high current density of 6.00 mA.cm−2 at the potential of 750 mV (vs. RHE) as compared to PANI (1.50 mA.cm−2), P(ANI-co-ABA) (0.50 mA.cm−2), and P(ANI-co-TPA) (2.20 mA.cm−2) at the same potential. The high obtained exchange current density of P(ANI-co-3NI) may be attributed to the outstanding and exceptional electrochemical responses obtained in the CV results as well as ECSA, surface coverage, and diffusion constant. The obtained current densities are higher as compared to the ones reported by Valiollahi and co-workers on the electrocatalytic responses of PEDOT [68]. From this, it can be deduced that copolymerization of polyaniline with its derivatives can be used as an efficient way to increase HER performance. To account for the HER reaction mechanisms, the LSV data was fitted into Tafel analysis according to Eq. 5 [28], and the Tafel plots are displayed in Fig. 9(b). The obtained Tafel slopes (b) were 389.5, 64.6, 57.3, 33.1, and 47.9 mV.dec−1 for GCE, PANI, P(ANI-co-ABA), P(ANI-co-TPA), and P(ANI-co-3NI), respectively. The obtained Tafel slopes PANI, P(ANI-co-ABA) and P(ANI-co-3NI), suggest that the rate-determining step is the Volmer-Heyrovsky step, whereas for P(ANI-co-TPA), the Volmer-Tafel mechanism is the rate-determining process. As the current density of the excellent electrocatalyst (P(ANI-co-3NI) did not reach the ideal one of 10 mA.cm−2, to compare with the reported ones, the overpotential to reach a current density of 1.0 mA.cm−2 was used as reported in previous studies of polyaniline-based materials. The comparison studies are shown in Table 4. As compared to the reported work, the P(ANI-co-3NI) showed inferior HER properties as it required an overpotential of 318 mV to reach a current density of 1.0 mA.cm−2. The inferior HER properties can be due to absence of metallic components as most electrocatalysts compared have Ni. However, the Tafel slope obtained for P(ANI-co-3NI) was less as compared to the reported work. Based on these observations, incorporation of metallic component or composites formation (binary, ternary or quaternary) can be used to enhance the HER parameter of the polyaniline homopolymer and copolymers. The HER mechanism taking place in P(ANI-co-3NI) which showed superior HER activity is given on Scheme 2 in acidic medium. The superior HER activity is influenced by the number of nitrogen atoms in the polymer backbone which can adsorb the protons and facilitate HER. Mechanistically, the proton from the electrolyte (H2SO4) adsorb onto the surface of P(ANI-co-3NI) to generate Hads@P(ANI-co-3NI)/GCE. This is followed by the adsorbed hydrogen proton reacting with the protons from the electrolyte to generate molecular H2.

Fig. 9
figure 9

(a) LSV polarization curves and (b) Tafel plots for GCE, PANI, P(ANI-co-ABA), P(ANI-co-TPA) and P(ANI-co-3NI)

Full size image
Table 4 Comparison of P(ANI-co-3NI) with reported Ni-based PANI materials in acidic medium
Full size table
Scheme 2
scheme 2

HER mechanism on P(ANI-co-3NI)

Full size image

The EIS was conducted to gain a better understanding and context of the HER kinetics properties of the as-prepared electrocatalysts. The EIS Nyquist plots of the materials are shown in Fig. 10(a) and the expanded version in Fig. 10(b) with the insert equivalent circuit used. The EIS results of P(ANI-co-3NI) showed a smaller semi-circle as compared to other materials, and this reveals the fast-charge transfer process. Table 5 shows the values obtained from the EIS analyses of the electrocatalysts. The Rct of the P(ANI-co-3NI) was found to be 16.17 Ω as compared to the one of PANI (25.81 Ω), P(ANI-co-ABA) (57.43 Ω) and P(ANI-co-TPA) (24.17 Ω) which further suggest the fast kinetics and increase in electrode conductivities. The charge transfer resistance, Rct were used to calculate the exchange current density (jo) where \(\:{j}_{o}=\raisebox{1ex}{${i}_{o}$}\!\left/\:\!\raisebox{-1ex}{$A$}\right.\) and the apparent heterogeneous electron transfer rate constant (ko) following the Eqs. (15) and (16) [73, 74]:

$$\:{i}_{o}=\frac{RT}{nF{R}_{ct}}$$
(15)
$$\:{k}^{o}=\frac{{i}_{o}}{nFAC}=\frac{{j}_{o}}{nFC}$$
(16)

Where the other terms have been already cited. The jo and ko reveal the intrinsic rate of transfer of electrons between the electrode and the electrolyte. The calculated jo values were found to be 7.01 mA.cm2 (PANI), 3.16 mA.cm2 (P(ANI-co-ABA)), 7.48 mA.cm2 (P(ANI-co-TPA)) and 10.83 mA.cm2 (P(ANI-co−3NI)). The estimated ko values were found to be 7.26 × 10−6 cm.s−1 (PANI), 3.27 × 10−6 cm.s−1 (P(ANI-co-ABA)), 7.76 × 10−6 cm.s−1 (P(ANI-co-TPA)) and 11.21 × 10−6 cm.s−1 (P(ANI-co−3NI)). The most significant values of jo and ko related to P(ANI-co−3NI) indicate faster electron transfer process [73, 74]. The Bode plots (Fig. 10(c-f)) of the materials were used to deduce the lifetime of the electron in the system. The lifetime of the electrons (Eq. 6) [33] were 1.83, 2.54, 1.91, 1.89, and 2.18 ms for GCE, PANI, P(ANI-co-ABA), P(ANI-co-TPA), and P(ANI-co-3NI), respectively. The electron lifetime for all the electrocatalysts was greater than that of pure GCE, with PANI showing a longer lifetime. P(ANI-co-3NI) had the lifetime closer to the one of PANI and this suggest that the electronic properties of the materials was not compromised greatly.

Fig. 10
figure 10

(a) and (b) Nyquist plots of as-prepared materials as well as Bode plots of (c) PANI, (d) P(ANI-co-ABA), (e) P(ANI-co-TPA) and (f) P(ANI-co-3NI)

Full size image
Table 5 The EIS and bode parameters of the prepared electrocatalyst
Full size table

To probe the long-term stability, the CA for PANI and P(ANI-co-3NI) was conducted for 2 h, and the results are displayed in Fig. 11(a). The P(ANI-co-3NI) material showed a good long-term stability and does not show a significant drop but it shows the accumulation and release of hydrogen gas bubbles on the surface of the electrode signified by the rough and irregular shape [75]. To estimate the amount of hydrogen on the surface of the electrode, TOF was determined electrochemically using Eq. 8 [34] from the number of active sites (n) as shown in Eq. 7 [34]. The estimated number of active sites were 5.46 × 10−8 mol for PANI and 1.59 × 10−7 mol for P(ANI-co-3NI). The higher number of active sites in P(ANI-co-3NI) suggests that the intrinsic property of P(ANI-co-3NI) is greater than that of PANI homopolymer. The obtained TOF (Fig. 10(b)) were 0.25 mmol H2.s−1 for PANI and 3 mmol H2.s−1 for P(ANI-co-3NI). The higher TOF value for P(ANI-co-3NI) suggests that the copolymerization route can be used as an effective way to improve the intrinsic properties of electrocatalysts. In this regard, more work must be done on using the copolymers as an electron transport channel enhancer for other metallic and non-metallic electrocatalysts.

Fig. 11
figure 11

(a) Long-term stability and (b) TOF polarization plot of PANI and P(ANI-co-3NI)

Full size image

4 Conclusions

In this work, we demonstrated the fabrication of polyaniline homopolymer and copolymer through a chemical polymerization method. The structural characterization revealed that the copolymerization route does not disrupt polymer formation and generates polymeric materials with enhanced properties such as optical properties. The electrochemical characterization showed that the P(ANI-co-3NI) exhibited exceptional chemical reactivity indices, which makes it a great candidate for HER. P(ANI-co-3NI) possessed outstanding HER properties achieving a current density of 6 mA.cm−2 at the potential of 750 mV vs. RHE, high conductivity, electron lifetime as well as fast charge transfer process as compared to other materials. Furthermore, the TOF, which is the measure of intrinsic property of the electrocatalyst, was improved. To further improve the HER properties of polyaniline and copolymers, we believe that the copolymers can be used as an electron transport channel for secondary abundant materials such as metal oxides, sulphides, and other carbonaceous materials.