Scheme 1 displays the synthetic procedures of Cu-doped PANI porous films and corresponding devices. The Cu-doped PANI porous films were obtained by in situ electrochemical depositions of aniline molecules in the H2SO4 aqueous solution. The Au porous substrate layer has two functions: One is to provide a conductive framework and external electrode for the electrode of IR electrochromic device; second, the gold metal has a high reflectivity in the IR range, which can effectively reflect the IR energy through the working electrode of PANI films. When polymerized in acid dopant, the PANI begins to aggregate on the surface of the Au/PES porous membrane. Meanwhile, the color of the membrane surface changes from golden to yellowish-green and to dark green as the polymerization charge increases.
To reveal the composition information, the XRD patterns are collected. As shown in Fig. 1, all the samples present a wide peak centered at 18.5°, which could be from the amorphous structure of the PES substrate . According to the previous reports, the different diffraction peak usually relates to the mono-distributions of the periodicity between the polymer backbone chains . For the pure PANI, a broad peak at 2θ = 20° is assigned to the periodicity parallel to the polymer chains of PANI (020). The Cu-doped PANI film shows a diffraction peak at 2θ = 25°, indicating the periodicity perpendicular to the polymer backbone chain (200) . With the introduction of copper ions, the PANI signal peaks from Cu-doped PANI porous films become stronger, suggesting that Cu ions stimulate an increase in crystalline fraction and formation of Cu/PANI co-crystals. Furthermore, it is found that two signal peaks from Cu3(SO4)2(OH)2·4H2O and (NH3)2Cu(NO3)2 exist in the prepared Cu-doped PANI porous films, as marked by the diamonds and clubs (PDF# Cu3(SO4)2(OH)2·4H2O and PDF# (NH3)2Cu(NO3)2). It indicates the possible existence of Cu–N coordination bonds in Cu-doped PANI porous films, meantime the simultaneous doping of sulfuric and nitric acid ions. The signal peaks of (NH3)2Cu(NO3)2 from PANI-1 and PANI-2 porous films are the sharpest among the prepared PANI porous films, which may have a great impact on the subsequent electrochromic properties.
To analyze the chemical bonds in the PANI films, the Raman spectra were measured in the range of 100–2000 cm−1. In Fig. 2, all the Cu-doped PANI porous films present the signals at the same positions. Specifically, the characteristic peaks observed at 1622 and 1572 cm−1 are assigned to the C–C stretching vibration of the benzenoid ring and C=C stretching vibration of the quinoid ring, respectively. The peaks at 1480, 1313, and 1252 cm−1 in the spectra are attributed to the C=N stretching vibration of the quinoid ring and the C–N stretching vibrations of the benzenoid ring and quinoid ring, respectively. The C–N·+ stretching vibration of more delocalized polaronic structures is represented by the peak at 1343 cm−1 (Fig. 2b), and the benzenoid ring deformation in polarons is suggested by the signal at 868 cm−1 . Compared with the Raman spectrum of the PANI-0 porous film, the Cu-doped PANI porous films show a stronger and sharper absorption peak at 1343 cm−1. As expected, this reveals that a large number of polarons and bipolarons delocalized on the PANI chains with the introduction of copper ions. The peaks at 1198 cm−1 and 812 cm−1 represent the C-H in-plane bending vibration and out-plane bending vibration in the quinoid ring . Note that the intensity of the peak at 1198 cm−1 increases with the introduction of copper ions, suggesting the increased oxidation degree of Cu-doped PANI porous films.
The geometric structure of the material is critical to optical response performance since the ion transfer between the electrode material and electrolyte plays a decisive role during the electrochemical redox process. Herein, the morphologies of PANI porous films are investigated by SEM images. As shown in Additional file 1: Fig. S2, similar to that of the PES and Au/PES membranes with a linear porous network structure, pure PANI film presents a fibrous structure, which constructs a reticular membrane. The porous structure benefits to electrolyte infiltration and ion transmission. The surface morphology of the Cu-doped PANI porous films is exhibited in Fig. 3a–d. Figure 3e–h shows the cross sections of the prepared Cu-doped PANI films. Obviously, all the PANI layers with a thickness of 40 µm are tightly attached to the Au/PES porous membrane, and uniform voids can be found in the porous structures. Additional file 1: Figure S1 shows the polymerization current as a function of time for prepared PANI porous films (i–t curves). During the entire electrodeposition process, the aggregation process of PANI on the porous films can be described in three stages. At the beginning of the polymerization reaction, it is mainly performed through linear polymerization. PANI is gradually filled on the walls of the porous membrane, constructing a reticular membrane similar to the morphology Au/PES membrane. Subsequently, with increasing polymerization time, the polymerization mode was changed from linear polymerization to radial polymerization. The PANI film grows perpendicular to the polymer backbone chain, and the cavities of the porous films are gradually filled by PANI particles. Finally, large particles and rodlike PANI appear on the surface of the films and are accumulated. Moreover, with the addition of copper concentration, the surface morphologies of Cu-doped PANI porous films become more compact and rougher (Fig. 3), indicating that the cavities of the porous films were gradually filled by PANI nanoparticles and crystal complexes visible on the XRD patterns.
To present the surface roughness of PANI porous films, the AFM was used to investigate the topography of Cu-doped PANI films. Additional file 1: Figure S3 depicts 2D and 3D images of different copper contents for Cu-doped PANI film in area 5 × 5 µm2. The ridges and the valleys on the surface of porous films can be observed from the layers of prepared films. The roughness Ra values are calculated to be 69.8 nm, 85.5 nm, 115.7 nm, and 135.4 nm for PANI-0.5, PANI-1, PANI-2, and PANI-3 porous films, respectively. The energy-dispersive spectrometer (EDS) mapping images of Cu-doped PANI film surface are shown in Fig. 4a, c–f. The results show that the elements of carbon, oxygen, copper, and nitrogen are uniform distribution on the surface of PANI-2 porous films. Compared to that in spectrum 4/6, the higher copper content in spectrum 5 indicates that the copper element is distributed uniformly in the prepared films, rather than the local larger nanoparticles, as shown in Fig. 4b.
To reveal the elements’ electronic state, XPS analyses were conducted. As shown in Additional file 1: Fig. S4, all the Cu-doped PANI films show the presence of carbon, nitrogen, oxygen, sulfur, and copper element. As a comparison, the copper signals are missing in PANI-0 films. To analyze the chemical valence state of the copper elements and disclose their role on polarons, the high-resolution Cu 2p spectra are exhibited in Fig. 5. Benefitting from the sensitivity of X-ray photoelectron energy spectra toward the element coordination environment, the copper signals from all the Cu-doped PANI films can be fitted into two components with the presence of two satellite peaks. The binding energies near 934.98 and 955.14 eV are originated from the characteristic peaks of Cu 2p3/2 and Cu 2p1/2 from Cu (II) species, while the two peaks located at 932.3 and 952.05 eV correspond to the Cu 2p3/2 and Cu 2p1/2 from Cu(I) species, respectively [29,30,31]. Therefore, the binding energies near 933 and 953 eV are originated from the characteristic peaks of Cu 2p3/2 and Cu 2p1/2 from Cu(δ) species. The Cuδ+ (+ 1 < δ < + 2) with a lower oxidation state could be from Cu–N coordination bonds, and the Cu (II) species come from the metal salt. These are inconsistent with those from the above XRD. As shown in Table 1, with the increase in Cu content, the Cuδ+ species of prepared PANI films shift to lower binding energy and then to higher energy. Such behavior suggests that the introduction of copper ions not only facilitates protonation of quinonoid imine, but also forms Cu–N coordination compounds in polymer films. As reported, Cu–N coordination bonds were also catalyzed REDOX reactions . Therefore, the Cu–N coordination bonds formed in PANI-2 can obtain the best electrochromic performance.
As the coordinate atoms, the fine analysis of the N 1 s core-level spectra was performed. As shown in Fig. 6a–e, all the spectra from PANI films can be fitted into three major components located at 398.5, 399.5, 401.1, and 402.2 eV, which are attributed to the quinonoid imine (=N–), benzenoid amine (-NH-), protonated amine (-NH2+), and protonated imine (= NH+), respectively . Compared to PANI-0 porous film, the quinonoid imine (=N–) peak disappears in the XPS N 1 s spectra from Cu-doped PANI films. This indicates that the copper ions doping induced the transformation of quinonoid imines in PANI films. The =N– structure in the Cu-doped PANI films is converted to a positively charged –N+·– structure, corresponding to the two peaks in the XPS spectra with binding energy greater than 400 eV (i.e., protonated amine and protonated imine) . Specifically, with the increase in copper concentration from 0.005 M to 0.02 M, the percentage of the protonated amine and protonated imine increases from 50 to 66%, while the content of benzenoid amine decreases. It hints that the initial introduction of copper ions promotes protonation, leading to the formation of protonated amine. When the concentration is further increased to 0.03 M, PANI-3 porous film experiences a reverse trend, in which the content of protonated amine and protonated imine decreases to 52%.
As shown in Raman spectra, with the introduction of copper ions, the intensity of the peak at 1198 cm−1 increases, suggesting the increased oxidation degree of Cu-doped PANI porous films. Therefore, the decreased number of polarons for PANI-3 may be caused by the excessive oxidation degree of PANI-3, hindering the formation of carriers. Moreover, the dense microstructure of PANI-3 porous film was not conducive to carrier transport. These results indicate that the proper doping of copper ions for PANI may achieve the formation of Cu–N coordination, which facilitates the generation of polarons and bipolarons. However, exceeding the content of copper ions could format superfluous crystals complexes to hinder the delocalization of polarons and bipolarons. As mentioned above, the formation and the elimination of polarons and bipolarons delocalized on the PANI chains are the direct and most critical factors in realizing excellent emissivity modulation performance. Thus, based on the maximum number of polarons and bipolarons, the PANI-2 porous film is expected to exhibit the best infrared electrochromic performance.
According to the above structural analysis, it is anticipated the obvious differences in infrared emission modulation on the Cu-doped PANI films prepared with different copper ions concentrations. Cyclic voltammetric measurements were performed to preliminarily determine the electrochemical performance of the Cu-doped PANI porous films. As displayed in Fig. 7, the CV curves of Cu-doped PANI porous film present a pair of redox peaks corresponding to the state of emeraldine salt (ES) and leucoemeraldine (LE). The electrochemical active area under the CV curves is increasing with the increase in copper ions concentration. The PANI-2 porous film shows the largest electrochemical active area, suggesting the maximum amount of charge transfer and the potential optimal infrared electrochromic regulation performance.
As reported, the IR electrochromic properties of PANI-based films are directly related to their existential states [35, 36]. Hence, the potentials of − 0.25 V and 0.5 V were chosen on account of the films’ states completely transforming between LE and ES. Chronoamperometric (CA) experiment was conducted to investigate the IR electrochromic performance of PANI-2 and PANI-0 porous films. As shown in Fig. 7b, the color of the film materials changed from yellow to atrovirens in the voltage positive conversion process. For the PANI-2 porous film, the response time, defined as the times required for achieving 95% change of the full current density, from coloring state to bleaching state (the bias voltage − 0.25 V) can be calculated to be 0.7 s. When the bias voltage of 0.5 V is applied, the response time is determined to be 1.5 s (Additional file 1: Fig. S5(a)), much superior to that of pure PANI and previous reports [13, 14, 20, 37]. This can be reasonable by the reticular structure that provides more ion channels and a larger reactive area, improving the electrochemical reaction rate. The faster response time of the PANI-2 porous film suggests a more sensitive device. Furthermore, the response time of the present PANI-2 porous film is, respectively, about 1.3 s and 2.7 s for the bleaching state and the coloring state after 50 cycles of durability test, as shown in Fig. 7b and Additional file 1: Fig. S5(b). Therefore, the prepared PANI-2 films not only show excellent response ability, but also good stability.
To quantitatively analyze and compare the IR emissivity change of PANI-based films, the Fourier transform infrared spectra were studied in the whole wavelength range from 2.5 to 25 µm and the values of ε are calculated according to Eqs. (1) and (2). Figure 8 presents the emittance curves of Cu-doped PANI porous films at potentials of − 0.25 V and 0.5 V. The emittance curves at 0.5 V gradually rise with the copper ions concentration changing from 0.005 to 0.02 M, while a decrease is observed for the further increasing to 0.03 M. This trend is inconsistent with the number of polarons and bipolarons obtained in XPS results. As discussed above, the copper ions can activate the part of the inactive PANI film. Therefore, the Cu-doped PANI films, especially for PANI-1 and PANI-2 films, have low IR emissivity at -0.25 V. Similar to the emittance curves, the ∆ε between the curves at − 0.25 to 0.5 V presents a change in volcanic patterns. The maximum ∆ε from the PANI-2 porous film is 0.35 in ranges of wavelength 8–12 µm. By contrast, the PANI-0, PANI-0.5, PANI-1, and PANI-3 only provide the ∆ε of 0.1, 0.12, 0.21, and 0.06, respectively, as shown in Fig. 8e. The decreased ∆ε of PANI-3 may be caused by the dense microstructure of PANI-3 porous film, hindering the transmission of carriers [35, 38]. Therefore, the PANI-2 porous film exhibits the best-infrared modulation performance in this work (∆ε in the wavelength ranges of 2.5 to 25 µm). It indicates that the evolution of the emittance curves largely depends on the copper ions concentration and microstructures of PANI on the porous films.
Based on the best Δε value, the PANI-2 porous film is selected as the functional layer to assemble IR electrochromic (EC) device. Figure 9a–d exhibits the digital photographs of the Cu-doped PANI IR EC device at different voltages. When the applied voltage is fixed at 0.3 V, the device presents a color change from yellow to green. As the applied voltage increases to 0.5 V, the color of the device changes to dark green. When the applied voltage reaches 0.8 V, the device presents atrovirens. The results demonstrate that the Cu-doped PANI IR devices can tunably blend with green or yellow backgrounds under different voltages.
To explore the application potential of the Cu-doped PANI IR EC device in optical and thermal management, the IR emissivity of the EC device was also investigated in the wavelength ranges of 3–5 µm and 8–12 µm at the different voltages. As shown in Fig. 9e, the evolution of ε in two wavelength ranges shows a similar tendency. That is, the emissivity gradually increases along with the applied voltage. When the applied voltage reaches 0.5 V, the emissivity of the device increases a little and tends to be stable as a result of the PANI layer in the ES state. Note that the present △ε of 0.32 in 8–12 µm is much larger than that (0.24) from the previous pure sulfuric acid-doped PANI device . In addition, the response time and stability of the device are also shown in Fig. 9f. The coloring time of the device is about 5.74 s (1st cycle) and 6 s (50th cycle), while the fading time is about 3.77 s (1st cycle) and 5 s (50th cycle). Meanwhile, the current density of the device still retains less than 0.3 mA cm−2 loss after the 50 cycles. Therefore, the present IR electrochromic device exhibits superior tenability and good electrochromic stability.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.