Introduction

Conducting polymers, due to the special physical and chemical properties, are becoming more and more attractive in a diversity of fields with the rapid development of electric industries and information technologies [1]. Conducting polymers include polyacetylene (PA) [2, 3], polyaniline (PANI) [4,5,6], polymelamine (PMel) [7, 8], polypyrrole (PPy) [9, 10], and polythiophene (PT) [9, 11]. Among these conducting polymers, PPy, as one of the most popular candidates, finds widespread applications ranging across solar cell [10], electrochromic device [12], photodiode [13], supercapacitor [14, 15], and electrochemical sensor [16,17,18] due to the excellent features of high conductivity, high stability, good bio-compatibility, facile fabrication, and low cost [17, 19]. Morphologically, various forms of PPy nanomaterials have been fabricated, including cauliflower nanospheres [10], nanowire [20, 21], and nanotubes [22]. Among these, three-dimensional nanotube has a high specific surface area and accessible active sites for ion transportation. Moreover, their inner and outer diameters as well as the aspect ratio of the tubes are tunable for different applications. This fabrication process and the form are controllable during synthesis typically by an electro-spinning technique.

Nano-structured PPy has often been used as a conducting matrix or support for deposition of metal nanoparticles because of its porous structure, excellent conductivity, and high specific surface area [17, 23]. Meanwhile, the metal agglomeration can be abated due to the interaction of pyrrolic N and metal [24, 25]. PPy-Pt [10, 16], PPy-Ag [21, 26, 27], and PPy-Au [28, 29] have been successfully fabricated for adsorption, catalysis, energy storage, and sensing applications. Base metal copper nanoparticles possess unique catalytic activity in a variety of fields including chemical synthesis, sensor, and energy storage [30,31,32]. In addition, it has excellent advantages of good conductivity, strong redox property, low cost, and natural abundance, compared with the noble metals. The combination of Cu and PPy may achieve the synergetic effect for catalysis application. However, in the PPy/metal composite preparation, PPy was prepared mostly by electropolymerization [25, 33] or bulk synthesis [34, 35], which leads to dense and thick film, difficult for the even deposition of the highly dispersed metal particles. Three-dimensional PPy nanotube is supposed to provide huge surface area and a large quantity of active sites for the effective interaction of PPy and Cu, preventing the agglomeration of Cu. However, to our knowledge, the Cu nanoparticles decorated PPy nanotube composite has not been reported to date.

In this article, we present a PPy nanotube fabrication method using Py polymerization on an electrospun PS fiber followed by solvent etching. The morphology of PPy nanotubes can be tuned by Py concentration. By convenient one-step electrolysis, Cu nanoparticles are deposited on the PPy tubes. The influence of PPy structure on Cu deposition and the catalytic performance to nitrite reduction are investigated. The well-defined three-dimensional tubular structure facilitates the Cu deposition. In light of synergetic effect of PPy and Cu on the reduction of nitrite, the PPy-Cu electrode has been used as a sensitive sensor for nitrite determination. In addition, the drop-cast method is easy for batch fabrication and the obtained PPy-Cu electrode is very stable and reusable without loss of activity, showing a great potential for future sensing application.

Experiments

Reagents

Pyrrole (Py, ≥ 98%), polystyrene (PS, Mw = 280,000), and polyvinylidene fluoride (PVDF) were purchased from Sigma-Aldrich. Py was purified by reduced pressure distillation. N,N-Dimethylformamide (DMF), tetrahydrofuran (THF), and 1-methyl-pyrrolidone (NMP) were supplied by Lingfeng Reagent (Shanghai). All other chemicals (analytical grade) were obtained from Sinopharm Chemical Reagent Company and used as received. The ultrapure water (18.2 MΩ cm, 25 °C, Thermo Fisher Scientific) was used for aqueous solution preparation.

PS nanotube preparation

The PS nanofibers (about 150 nm in diameter) were prepared by electrospinning method analogous to that reported previously [36]. The PS nanofibers were sulfonated in a concentrated sulfuric acid (98%) for 4 h at 40 °C to facilitate the growth of PPy. Following the cleaning of sulfonated fibers with ultrapure water, they were soaked in various concentrations of Py ethanol solution (0.2%, 0.4%, 0.6%, 0.8%, v/v) for 24 h to adsorb Py monomer onto the PS surface. The oxidation polymerization of Py occurred upon the addition of FeCl3 solution to the above solution [n(FeCl3):n(Py) = 1:1]. The polymerization reaction lasted for 24 h at 4 °C. Subsequently, the inner PS core of PS/PPy fiber was removed by THF wash. The final PPy nanotubes were obtained after ethanol and water wash followed by vacuum dry (50 °C) for 10 h. PPy tubes synthesized by 0.2%, 0.4%, 0.6%, and 0.8% Py were named as PPy(0.2), PPy(0.4), PPy(0.6), and PPy(0.8), respectively.

Preparation of PPy-Cu modified glassy carbon electrode

A glassy carbon electrode (GCE; Ø = 3 mm) was first polished successively with 1.0 and 0.3 μm Al2O3 slurries and then sonicated with a mixture of water and ethanol after each polishing. Subsequently, GCE was dried with high purity nitrogen to obtain a mirror-like surface. A suspension solution of PPy nanotubes was prepared by dispersing PPy nanotubes and PVDF solvent (PPy:PVDF = 8:1, weight ratio) in 10 mL NMP and sonicating for 10 min. The PPy nanotube-modified glassy carbon electrode (PPy/GCE) was obtained by drop coating 5 μL of the above suspension onto a clean GCE surface and dry under N2 gas.

The deposition of copper nanoparticles on the PPy/GCE was performed by electrolysis at −0.7 V for 200 s in 0.1 M H2SO4 aqueous solution in the presence of 0.05 M CuSO4 and 1 mM KCl. Following the thorough wash by water, the PPy-Cu modified glassy carbon electrode (PPy-Cu/GCE) was obtained.

The effective active surface areas of GCE, PPy/GCE, and PPy-Cu/GCE were calculated to be 0.11, 0.20, and 0.18 cm2 by CV in 1 mM K3[Fe(CN)6] solution, based on the literature method [37]. The detailed calculation can be found in SI (Fig. S1). The slight decrease of surface area after Cu deposition is presumably due to the interaction of partial oxidized Cu ions in PPy-Cu with [Fe(CN)6]3− to form the deposit during CV scan [38].

Characterization method

Scanning electron microscopy (SEM; Hitachi, S-4800) and transmission electron microscopy (HRTEM; Talos F200S*) were used to observe the morphology of the nanomaterials. The composition of the fibers was studied by Fourier transform infrared spectroscopy (FT-IR; Nicolet 6700) and X-ray photo spectroscopy (XPS; AXIS Ultra DLD).

For TEM characterization, the PPy-coated GC plate was electrolyzed at −0.7 V for certain times (50, 100, 200, 300 s). After washing with copious 0.1 M H2SO4 and water successively, the PPy-Cu sample was carefully collected by scratching out of the GC surface and then dispersed in ethanol solution by sonification.

Analytical method

The electrochemical experiments were carried out using CHI 760D electrochemical workstation (Shanghai Chen Hua CHI Instrument Company) with a three-electrode system including a GCE working electrode, an Ag/AgCl (3 M KCl) reference electrode, and a Pt counter electrode.

The electrochemical behavior of the work electrodes was investigated by cyclic voltammetry (CV). All the CV signals reached a steady state in 5 cycles of scans. The current response of PPy-Cu/GCE to nitrite was recorded by i-t curve at −0.4 V.

Results and discussion

Formation and analysis of PPy nanotubes

Polymerization of Py and the formation of PPy nanotubes were characterized by FT-IR. The spectra of sulfonated PS, PS/PPy, and hollow PPy nanotubes are shown in Fig. 1. The absorption bands at 1494, 1452, 754, and 694 cm−1 correspond to the C=C framework of benzene, methylene, and methane groups in PS [39]. The absorption peaks at 1180, 1070, and 1035 cm−1 are related to the stretching vibration of sulfonic group [40, 41], indicating that the sulfonic groups have been successfully grafted to the PS fiber surface. The presence of the –SO3− ensures that the Py monomer can be adsorbed stably and uniformly on the surface of the PS nanofibers. After polymerization of Py, two new peaks appear at 1566 and 1550 cm−1, corresponding to the characteristic absorption peaks of C=C and C–C stretching vibration of PPy, respectively [42]. New absorption bands at 1386 and 968 cm−1 are found, which are ascribed to the C–N stretching vibration and the in-plane deformation vibration of N–H, respectively [42]. These confirm the formation of PPy polymer on the PS surface. After THF washing, the characteristic absorption bands of PS at 754 and 694 cm−1 disappear, demonstrating that the PS nanofibers are removed with THF washing. However, the absorption peaks at 1180, 1070, and 1035 cm−1 representing the sulfonic group on the PPy nanotubes still exist, indicating that some sulfonic ions are doped into the PPy chain. It concludes the PPy nanotubes in oxidation state are formed successfully.

Fig. 1
figure 1

FT-IR spectra of three different nanofibers. a Sulfonated PS nanofibers. b PS/PPy composites. c PPy nanotubes

Full size image

Figure 2A displays the morphology of PS nanofibers with an average diameter of 150 nm. After polymerization of Py, the fibers maintain the original continuous and uniform structure, but the surface becomes a bit coarser and rougher (Fig. 2B). Through THF etching, the tubular PPy is found clearly from the cross-section image of SEM (inset of Fig. 2C). This indicates the formation of PPy nanotubes.

Fig. 2
figure 2

SEM images of three different nanofibers. a PS nanofibers. b PS/PPy composites. c PPy nanotubes. Inset is the cross-section image (upper right)

Full size image

Various PPy nanotubes were obtained by varying the concentration of Py monomer (0.2%, 0.4%, 0.6%, 0.8%) during polymerization. Figure 3 presents the cross-section images of different PPy tubes. It can be seen from Fig. 3A that the surface of the PPy(0.2) nanotubes is grainy and rough with an average wall thickness of 80 nm. With increasing the concentration of Py, the obtained PPy nanotubes become well-defined, and the surface is uniform with some pores (Fig. 3B, C). It is possible that few PPy nuclei are formed at low concentration, leading to large particles, and thus the PPy nanotube is poor-defined. However, high concentration (0.8%) of Py leads to agglomeration of PPy particles and thick wall (Fig. 3D).

Fig. 3
figure 3

Cross-section SEM images of PPy nanotubes prepared with different concentrations of pyrrole: 0.2% (a), 0.4% (b), 0.6% (c), and 0.8% (d)

Full size image

Electrochemistry of PPy/GCE and Cu deposition on PPy/GCE

As demonstrated in SEM, various Py monomer concentrations lead to different morphologies of PPy nanotubes. CV is applied to investigate the redox properties of various PPy/GCEs. Figure 4A presents the CV curves of the PPy/GCEs coated with same mass but different PPy structure. Compared with bare GCE (curve a), there is a broad redox peak at round 0 V, which corresponds to the redox doping and dedoping of PPy. The reaction is described as below [43]:

$$ {PPy}^{+}\left({A}^{-}\right)+{e}^{-}\underset{Doping}{\overset{Dedoping}{\rightleftarrows }}{PPy}^0+{A}^{-} $$
(1)
Fig. 4
figure 4

CVs of bare GCE (a) and PPy/GCEs (b~e). b CVs of Cu/GCE (a) and PPy-Cu/GCEs (b~e). Curve (b~e): PPy(0.2), PPy(0.4), PPy(0.6), and PPy(0.8). Electrolyte: 0.1 M H2SO4, scan rate: 50 mV/s

Full size image

The PPy/GCE obtained with well-defined PPy nanotubes presents higher redox current, indicating the large specific surface and better electron transfer of the PPy nanotube (vide supra).

The PPy-modified GCE with large specific surface area is supposed to facilitate the deposition of Cu due to Cu-PPy interaction and reducing property of PPy [33, 44, 45]. The deposition of Cu on PPy/GCEs is performed in 0.1 M H2SO4 solution containing 0.05 M CuSO4 and 1 mM KCl. Figure 4B exhibits the CV curves of Cu/GCE and various PPy-Cu/GCEs. In the absence of PPy, very weak reduction and oxidation signals of Cu are found at about −0.16 and 0 V. It indicates that very low amount of Cu is deposited on bare GCE (curve a). In contrast, the redox signal of Cu is enhanced dramatically in the presence of PPy, and the highest peak is observed on PPy(0.6)-Cu/GCE, with the peak-peak separation of 200 mV. Under the polarization condition for Cu deposition, the oxidized PPy is reduced, and the reduced state PPy0 promotes the reduction of Cu [44]. The reaction can be described in the following equation.

$$ {PPy}^{+}+{e}^{-}\to {PPy}^0 $$
(2)
$$ 2{PPy}^0+{Cu}^{2+}\to 2{PPy}^{+}+ Cu $$
(3)

The PPy nanotube with large specific surface area facilitates the charge transfer. Therefore, the deposition of Cu on PPy nanotube electrode is highly promoted. The peak current and reversibility of signal are highly dependent on the morphology of PPy. The current intensity of Cu arrives at the maximum at the PPy(0.6)/GCE with the same coating mass of PPy tube, because the PPy(0.6) nanotube is much uniform and well-defined, compared to other PPy nanotubes (see SEM in Fig. 3). It concludes that the well-defined PPy tube with uniform surface possesses higher active sites, facilitating the adsorption and deposition of Cu, being consistent with the CV in Fig. 4A.

Effect of Cu deposition time

By changing the Cu deposition time, Cu nanoparticles with different sizes were obtained. Figure 5A–D shows the TEM images of PPy-Cu with various deposition times. Ultrafine Cu particles are uniformly decorated on the PPy surface. As the deposition time increases, there are more and more Cu particles observed and the particle size is slightly increasing (insets of patterns A-D). When the deposition time increases to 300 s, there is a little agglomeration between Cu particles (Fig. 5D). The relationship of particle size with deposition time is drawn in Fig. 5E. The average particle size is 4.2 ± 1.2 nm at the electrolysis time of 200 s. The uniform distribution of Cu and increase of Cu content with increasing deposition time have been demonstrated by TEM mapping and energy-dispersive X-ray analysis (details in Fig. S2 and Table S1 in supporting information).

Fig. 5
figure 5

TEM images of PPy-Cu at different deposition times of Cu of 50 s (a), 100 s (b), 200 s (c), and 300 s (d). Scale bar: 50 nm for a and b; 100 nm for c and d; 20 nm in the insets of a–d. e The average copper size at different deposition times

Full size image

In order to analyze the valence state of copper, XPS characterization of PPy-Cu sample was performed. Figure 6 shows the high-resolution scan spectrum of Cu2p. By least square fitting, Cu2p3/2 and Cu2p1/2 at 933.0 and 953.1 eV, corresponding to zero-valence Cu [46], are observed. In addition, four weak peaks are found at 934.6, 944.0, 955.1, and 963.7 eV, which are the characteristic peaks of Cu2p in CuO [47, 48], indicating the presence of a little CuO due to partial oxidation of Cu. The result shows that zero-valence Cu is dominated in PPy-Cu.

Fig. 6
figure 6

Cu2p high-resolution XPS scan of PPy-Cu

Full size image

Electrocatalytic reduction of nitrite

Electrode materials play a significant role on the reduction of nitrite. Figure 7 shows the CV curves of various electrodes in 0.1 M H2SO4 solution in the absence and presence of 0.5 mM NO2−. There is negligible response on GCE (a). And the catalytic response of Cu/GCE to the reduction of nitrite is not found due to very small deposition amount of Cu (b). On PPy/GCE (c), the reduction current increases slightly upon the addition of nitrite, while the PPy-Cu/GCE (d) shows a dramatic enhancement of the catalytic current at −0.3 V. It is known that PPy can catalyze the reduction of nitrite during the dedoping process of PPy [49]. Meanwhile, the Cu nanoparticles on PPy-Cu/GCE can also function as a catalyst for nitrite reduction [50]. The nitrite reduction mechanism on the PPy-Cu/GCE can be described with the following equations:

$$ {PPy}^{+}+{NO_2}^{-}\to {PPy}^{+}{NO_2}^{-} $$
(4)
$$ {PPy}^{+}+{e}^{-}\to {PPy}^0 $$
(5)
$$ {PPy}^0+{NO_2}^{-}+2{H}^{+}\to {PPy}^{+}+ NO\left({N}_2\right)+{H}_2O $$
(6)
$$ 3 Cu(0)+{NO_2}^{-}+8{H}^{+}\to 3 Cu(II)+{NH_4}^{+}+2{H}_2O $$
(7)
Fig. 7
figure 7

CVs of GCE (a), Cu/GCE (b), PPy/GCE (c), and PPy-Cu/GCE (d) in the absence (a) and presence (b) of 0.5 mM nitrite in 0.1 M H2SO4 recorded at 50 mV/s

Full size image

The thus-synthesized PPy is in its oxidation state (PPy+) which can accumulate and adsorb NO2− on the active sites of PPy+. Upon applying a negative potential, PPy+ is reduced into PPy0, which further catalyzed the reduction of nitrite into NO or N2 (Eq. 6). On the other hand, Cu (0) in the PPy-Cu composite not only has good conductivity to promote the electron transfer of PPy, but also functions as a reducing agent catalyzing the reduction of nitrite, as shown in Eq. 7 [51]. Therefore, the synergetic effect in the PPy-Cu composite contributes much to its high catalytic activity towards the reduction of nitrite.

As discussed previously, PPy formed by different concentrations of Py exhibits various morphologies, and the resultant PPy electrodes possess different electrochemical activities (Fig. 4). The catalytic sensitivity of the various PPy-Cu/GCEs to nitrite reduction is investigated by measuring the calibration curves through CV scan in different concentrations of nitrite. Figure 8 shows the sensitivity at different electrodes. Being consistent with the redox response of PPy on electrodes, the corresponding sensitivity to nitrite reduction increases with the Py monomer and arrives at the maximum on the PPy(0.6)-Cu/GCE. It indicates that the increase in the copper deposition amount on the well-defined PPy tubes led to the enhancement of catalytic sensitivity. However, poor-defined PPy morphology (Py 0.8%) is not favorable to Cu deposition, and thus, the low Cu deposition and low sensitivity to nitrite reduction are observed at PPy(0.8)-Cu/GCE. The PPy(0.6)-Cu/GCE has the best sensitivity compared to other electrodes. In the subsequent experiments, the PPy(0.6)-Cu/GCE is used for nitrite measurement.

Fig. 8
figure 8

The sensitivity for nitrite ions on various PPy-Cu/GCEs (PPy(0.2), PPy(0.4), PPy (0.6), and PPy (0.8))

Full size image

Effect of copper size and solution pH on nitrite reduction at the PPy-Cu/GCE

As discussed in TEM analysis, the fine Cu particles with different sizes can be obtained by adjusting the deposition times. Nitrite reduction on PPy-Cu/GCE with different sizes was investigated and the results are shown in Fig. 9. The catalytic reduction current of nitrite increases with the deposition time of Cu particles and arrives at the highest when PPy-Cu is obtained with the deposition time of 200 s. Longer deposition time leads to the decreased catalytic current, which is probably due to larger Cu particle size and the agglomeration of Cu (shown in Table S1). It concludes that the catalytic reduction of nitrite by PPy-Cu/GCE is closely related to the Cu amount and Cu particle size, which is consistent with the morphological changes in the TEM image of Fig. 5A–D.

Fig. 9
figure 9

The sensitivity of PPy-Cu/GCE to nitrite ions with different sizes of Cu

Full size image

The catalytic reduction performance of the PPy-Cu/GCE was studied in different concentrations of H2SO4 solution (different pH) and the corresponding catalytic sensitivities were obtained by CV. As shown in Fig. 10, the catalytic sensitivity of the PPy-Cu/GCE to nitrite reduction increases with pH decreasing at high pH range (from pH 2.5 to pH 1.22), while decreases when pH is lower than 1.22. It is reasonable that the presence of H+ facilitates the catalytic reduction of nitrite on PPy-Cu, as stated in Eqs. 6 and 7. However, excessive acidity leads to severe hydrogen evolution reaction, thereby reducing the catalytic performance of nitrite. In the following experiments, pH 1.22 solution was used for further measurement.

Fig. 10
figure 10

The sensitivity of PPy-Cu/GCE to nitrite ions in electrolyte with different pH values

Full size image

Amperometric determination of nitrite at the PPy-Cu/GCE

In order to further analyze the catalytic property of the PPy-Cu/GCE, the amperometric response is obtained in 0.1 M H2SO4 at a constant potential of − 0.4 V. Figure 11A shows the typical i-t curve with successive increments of nitrite under stirring condition. The reduction current increases steeply upon addition of nitrite and arrives at a platform in less than 3 s, indicating a rapid diffusion of nitrite to the PPy-Cu/GCE. A steady increase of current is observed in a wide concentration range from 0.1 to 0.7 μM. The corresponding calibration plots are shown in Fig. 11B, C with concentration ranging in 0.1~0.7 and 0.7~1040.7 μM, respectively. The corresponding calibration equations are calculated to be ipa (μA) = − 1.5688 C − 0.01487 (Ra2 = 0.9911) and ipc (μA) = − 0.0458 C − 0.77763 (Rc2 = 0.9996). The limit of detection is about 0.03 μM (S/N > 3). The analysis performances of this proposed material for nitrite determination together with other materials previously reported are listed in Table 1 for comparison. It demonstrates that the current PPy-Cu/GCE possesses the wide linear range and low detection limit compared with the state of the art [52,53,54,55,56].

Fig. 11
figure 11

a Amperometric responses of PPy-Cu/GCE on successive injection of nitrite into 0.1 M H2SO4. Applied potential: −0.4 V. The corresponding calibration plots of current response versus nitrite concentration at different concentrations in the range of 0 to 0.7 μM (b) and 0.7 to 1040.7 μM (c)

Full size image
Table 1 Comparison of catalytic performance of various nitrite sensors
Full size table

Anti-interference ability, stability, and reproducibility of PPy-Cu/GCE

The common interfering substances were investigated. Figure 12 is the amperometric response of PPy-Cu/GCE by adding various interfering substances (NaCl, MgCl2, CaCl2, Na2SO4, KCl, NaNO3, Na3PO4, ZnCl2, NH4Cl, and C6H12O6) successively to 0.1 M H2SO4 containing 0.05 mM nitrite. It can be seen that there is no obvious current response after adding 200-fold of the above substances (shown in Table S2). It concludes that PPy-Cu/GCE has good selectivity for nitrite measurement.

Fig. 12
figure 12

Amperometric responses of PPy-Cu/GCE on successive injection of 200-fold of NaCl, MgCl2, CaCl2, Na2SO4, KCl, NaNO3, Na3PO4, ZnCl2, NH4Cl, and C6H12O6 into 0.1 M H2SO4. Applied potential: −0.4 V

Full size image

The long-term stability and reproducibility of PPy-Cu/GCE were further evaluated. After leaving the electrode in air for 1 month, the current response can maintain 90% for 0.5 mM nitrite, manifesting that PPy-Cu/GCE has good stability. In addition, the relative standard deviation (RSD) for six individual PPy-Cu/GCEs is 8.8% by assaying 0.5 mM nitrite with the same method. For six consecutive parallel determinations in 0.5 mM nitrite using the same modified electrode, the RSD of the corresponding current response was 3.6%, indicating that the modified electrode has good reproducibility.

Real sample analysis

To evaluate the feasibility of the PPy-Cu/GCE for practical application, this electrode was applied to detect NO2− in lake water and well water samples. All the water samples were filtered through a 0.2-μm membrane and adjusted to pH 1.22 before detection. The results are shown in Table 2 (n = 3), which are in good agreement with the standard UV-Vis method (GB/T 5750.5-2006). The spiked water recovery experiment gives the recovery values of 90–97%. Therefore, the PPy-Cu-modified electrode has acceptable detection accuracy for NO2− in real water samples.

Table 2 NO2− determination (μM) and spiked recovery test in real samples (N = 3) on the PPy-Cu/GCE under the optimal conditions
Full size table

Conclusion

Ultrafine Cu-coated PPy nanotube has been fabricated with electrospun PS nanofibers as template followed by PPy polymerization and Cu electrochemical deposition. Py concentration has significant influence on the morphology and electrochemical properties of PPy nanotubes, thus influencing the deposition amount of Cu. The well-defined PPy tube promotes the deposition of Cu and ultrafine Cu nanoparticles can be fine-tuned by changing deposition time of Cu. The PPy(0.6)-Cu/GCEs with 4.2 ± 1.2 nm Cu particle size (200 s deposition time) present a highly sensitive electrocatalytic performance to the reduction of nitrite, with broad linear detection ranges and low detection limit. It has been applied successfully for the real water sample analysis. The high selectivity, stability, and repeatability of the PPy-Cu/GCEs demonstrate a great potential for nitrite sensing application.