2.1 Materials

Pyrrole (C₄H₄NH), ferric chloride hexahydrate (FeCl₃[H₂O]₆), ammonium persulphate ((NH₄)₂S₂O₈), sodium dodecyl-sulphate (NaC₁₂H₂₅SO₄), deionised water (H₂O), resorcinol (C₆H₆O₂), formaldehyde (CH₂O), hydrochloric acid (HCl), acetonitrile(C₂H₃N), and sodium chloride (NaCl), Polyvinylidene fluoride (-(C₂H₂F₂)n-), N-methyl-pyrrolidone (C₅H₉NO), acetylene black (C), and Dimethyl-formamide (C₃H₇NO) were purchased from Sigma Aldrich, South Africa. All chemicals were used as received from the vendor.

2.2 Synthesis of polypyrrole (PPy)

Polypyrrole (PPy) was prepared by a chemical oxidation polymerisation method (Chitte et al., 2011). Ferric chloride was used as an oxidant, while sodium dodecyl sulphate was a dopant. 3.48 g of sodium dodecyl sulphate was dissolved in 100 mL of deionised water, 27.03 g of ferric chloride hexahydrate was also dissolved separately in 100 mL of deionised water. The two solutions were mixed in a 500 mL beaker and stirred for 15 min. 0.15 mol of pyrrole was then gradually added into the mixture dropwise and stirred for 4 hrs. A black precipitate of PPy was formed and filtered off. The PPy was then washed several times with deionised water and later dried in an oven for 12 hrs. at 60 °C. The synthesis procedure of PPy was as shown in Scheme 1.

Close

Scheme 1

Schematic of the Synthesis of Polypyrrole by Oxidative Polymerisation.

Export to PPT

2.3 Synthesis of carbon aerogel (CA)

Aerogel was first synthesised by the sol–gel condensation of formaldehyde and resorcinol. 0.337 g of resorcinol was dissolved in 11.5 mL of acetonitrile, 0.477 mL of formaldehyde was added, followed by 0.09 mL of a concentrated hydrochloric acid catalyst; the mixture was then stirred for 5 min. The resultant solution was then transferred into 50 mL centrifuge tubes and left to age for three days. The resultant sol–gel was frozen to −80 °C before freeze drying to form the aerogel. The aerogel was then carbonised in a tube furnace at 600 °C for 4 hrs. under nitrogen gas flow to give a carbon aerogel (Xu et al., 2012; Tannert). The schematics of the preparation of CA is shown in Scheme 2.

Close

Scheme 2

Schematic of carbon aerogel synthesis by the sol–gel condensation method.

Export to PPT

2.4 Preparation of CA-PPy composite slurry

The CA and PPy were homogenised to form a CA-PPy composite slurry using a method that has been previously reported (Yong-Jin Han et al., 2011). 4.0 g of PPy and 8.0 g of CA were thoroughly mixed. 8.0 g of the CA-PPy mixture, 1.0 g of Polyvinylidene fluoride (PVDF as a binder), and 1.0 g of carbon black (conductive material) mixture were then dissolved in a 100 mL N-methyl pyrrolidone (NMP) solution, and the mixture was stirred for 24 hrs to obtain a homogenous CA-PPy composite slurry. The ratio of the materials used to prepare the slurry is the standard adopted by most researchers (Rambabu et al., 2020; Zhao et al., 2017).

2.5 Coating of electrodes with CA-PPy composite slurry

The homogenised CA-PPy composite slurry was coated on activated carbon (AC) by the doctor blade drop coating method (Buczek et al., 2020). AC materials of 0.5 cm by 1.0 cm were first thoroughly washed in ethanol to remove any impurities. These were then dried in an oven at 80 °C for two hours, and the weights of the electrodes were then determined with a chemical balance. The CA-PPy composite slurry was then coated on the AC by the doctor blade drop coating method, it was then dried in a vacuum oven for 12 hrs at 100 °C. The weights of the modified electrodes are then determined on a chemical balance in anticipation to calculate the mass of CA-PPy composite on each of the AC electrodes. These were then applied as modified electrodes for desalination studies by the CDI method.

3.1 Electrochemical characterisation

The electrochemical characterisations, include cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS). Cyclic CV is an electrochemical method used to measure the current that is generated in an electrochemical cell under conditions where the applied voltage exceeds that predicted by the Nernst equation. CV tests are done by cycling the potential of a WE and measuring the resultant current. It is used to determine the oxidative and reductive properties of chemical substances. EIS is a very sensitive characterization method applied to establish an electrical response from chemical systems in a non-destructive manner. It investigates the response time of chemical systems using low amplitude alternating current (AC) voltages within a range of frequencies. A three-electrode system consisting of a working electrode (WE), reference electrode (RE), and a counter electrode (CE) was adopted for this study. A defined voltage was passed from the WE through an electrolytic solution and into the CE. This allows for a quantitative evaluation of small-scale chemical mechanisms at the electrode interface and within the electrolytic solution. The CV and EIS tests were done using a three-electrode Gamry potentiostat electrochemical workstation. The working electrode was an activated carbon sheet coated with the active material. A platinum wire was used as the counter electrode, and the reference electrode was an Ag-AgCl electrode. The electrolyte used for each experiment was a 1.0 M KCl solution. The potential window applied was −0.2 to + 0.6 V. The CV was done at different scan rates ranging from (5 to 100) mV/s. Electrochemical Impedance Spectra (EIS) was conducted over a frequency range of 0.1 Hz to 100 000 Hz. The applied potential for the EIS test is 0.22 V.

The specific capacitance (C_s) of the materials was evaluated from data of the CV experiment using Eq. (1) (Rambabu et al., 2020; O. logo et al., 2017)

Where the integral (∫idv) is the area of the CV spectra, m is the mass of CA-PPy (g), ΔV is the potential window (V), and S is the scan rate (mVs⁻¹).

3.2 CDI experiment for ion adsorption studies

The ion adsorption performance of the CA-PPy composite was conducted in the laboratory by the CDI method. Two AC electrodes of 0.5 cm by 1.0 cm were used as a cathode and anode, respectively. The anode was previously modified with the CA-PPy composite. These electrodes were then separated with a polythene material at a distance of 1.0 cm apart in a 100 mL beaker. 50 mL of NaCl solution was then transferred into the beaker (300 mg/L, 500 mg/L and 800 mg/L were used separately). An electrical conductivity meter was used to measure the conductance of the electrolyte throughout the experiment while it was constantly stirred with a magnetic stirrer at a speed of 200 rpm. The electrodes were then connected to a DC power source, and a potential of 1.2 V was applied.

On application of the potential, one of the electrodes becomes the cathode and the other the anode. The resultant polarization causes the cations Na⁺ and anions Cl^- to migrate to the cathode and anode. The ions were adsorbed on the surface of the electrodes. The process was allowed to continue until the electrodes were wholly loaded with ions (saturation). The electrodes were then transferred into another beaker containing deionised water, and the polarity at the electrodes was reversed for desorption to occur. This process was repeated several times, and the adsorption capacity and recyclability of the electrodes were monitored. The salt electrosorption capacity (A) of the electrodes was calculated using Eq. (2) (Alaei Shahmirzadi et al., 2018)

Where $\mathrm{\Delta }C$ $=$ Change in concentration (mg/L), $V=$ volume of NaCl solution used (L), $m$ $=$ mass of composite material on AC electrode (g).

The salt recovery efficiency (SRE) of the electrode material was also determined using methods previously reported. It was measured from the desalination conductivity data of the different concentrations of the NaCl solutions used for the study. Eq. (3) displays the formula used to determine the SRE (%).

Where Co = Initial conductivity of the solution, C = Final conductivity of the solution.

4.1 Instrument details

4.1.1

4.1.1 FT-IR characterisation

Figure 1 (a-d) shows the FT-IR spectra of the carbon aerogel, CA (Fig. 1a), PPy (Fig. 1b), and CA-PPy (Fig. 1 c and d) in evaluating functional groups. CA displayed several peaks at 3380 cm⁻¹, 2852 cm⁻¹, 2539 cm⁻¹, 2185 cm⁻¹, 2006 cm⁻¹, 1580 cm⁻¹, 1177 cm⁻¹ and 788 cm⁻¹ which corresponds to stretching vibration of O—H, ⚌CH_2, ≡CH, aromatic C—H, —C≡C—, C⚌C, C—O—C, and C⚌CH₂ groups as shown in Fig. 1(a), and confirms the formation of the carbon aerogel after calcination (Heslop, 2007; Kobayashi and Konishi, 2009; Mahani et al., 2018; Wang et al., 2015) shows the spectrum of PPy. The spectrum of PPy showed various peaks at (1513 cm⁻¹, 1325 cm⁻¹), (1154 cm⁻¹, 1049 cm⁻¹) and 809 cm⁻¹ which was attributed to the bending vibration of C⚌C, C—N, C—H and stretching C⚌CH₂ of the PPy ring (Alaei Shahmirzadi et al., 2018; Li et al., 2018; Wang et al., 2018). The 3382 cm⁻¹ and 2874 cm⁻¹ peaks were assigned to the stretching vibration of N—H and C-H₂ bonds (Oh et al., 2006; Ryoo et al., 2003). The CA-PPy peaks were observed at, 779 cm⁻¹, 1026 cm⁻¹, 1154 cm⁻¹, 1294 cm⁻¹, 1448 cm⁻¹, 1535 cm⁻¹, 1629 cm⁻¹, 1872 cm⁻¹, 2336 cm⁻¹, 2844 cm⁻¹, 2919 cm⁻¹ and 3114 cm⁻¹, which have PPy, and CA as shown in Fig. 1(c). The 779 cm⁻¹, 1026 cm⁻¹, 1154 cm⁻¹, 1294 cm⁻¹, 1448 cm⁻¹, 1535 cm⁻¹, 1629 cm⁻¹ showed sharp intense peaks shifted position due to the presence of PPy which is coated on CA (Samancı et al., 2021; Hebalkar et al., 2005).

Close

Fig. 1

(a): FTIR profile of CA (carbon aerogel). (b): FTIR profile of PPy (polypyrrole). (c): FTIR profile of CA/PPy (carbon aerogel and polypyrrole composite).

Export to PPT

The polypyrrole showed C—H wagging at 808 cm⁻¹ and 911 cm⁻¹ to confirm the presence of an extended carbon structure. C⚌N, C—N, and N—H stretching vibrations were seen at 1680 cm⁻¹, 1317 cm⁻¹ and 3550 cm^−1, respectively. Carbon to carbon double bonds were also seen at 1555 cm⁻¹ and 1487 cm⁻¹ to confirm the successful synthesis of polypyrrole. The spectrum of the CA-PPy composite the O—H, band was absent, a reduction in C—H stretch was observed and it also show the absence of some C—H bends. This indicates a successful binding of the functional sites of the CA and that of the PPy. Although, C⚌O, C—O, C—N bends remained visible on the FTIR spectrum for the CA-PPy composite at 1380 cm⁻¹, 1020 cm⁻¹ and 1005 cm^−1, respectively.

4.2 XRD characterisation

XRD characterisation was used to evaluate the crystal nature of the as-prepared materials CA, PPy and CA-PPy composite, as shown in Fig. 2(a-c). On the diffraction pattern of CA, two peaks were observed at 22.42^° and 43.79^°, which were assigned to the (0 0 2) and (1 0 1) planes. These peaks are due to the graphite nature of CA as shown in Fig. 2(a) (Farma et al., 2013; Xu et al., 2017). Fig. 2(b) shows the diffraction pattern of PPy, a broad peak at 23.82^°, is observed which corresponded to the (0 0 2) plane, and it confirm the amorphous nature of PPy (Li et al., 2019b). Fig. 2(c) is the pattern of CA-PPy, it shows a less intense peak at 24.47^° due to the graphitic properties of CA in the CA-PPy composite material. Other planes disappeared, suggesting the presence of a mixed structural form of the CA-PPy (Samancı et al., 2021).

Close

Fig. 2

(a): XRD profile of CA (carbon aerogel). (b): XRD profile of PPy (polypyrrole). (c): XRD profile of CA/PPy (carbon aerogel and polypyrrole composite).

Export to PPT

4.3 Raman characterisation

Raman spectroscopy was also employed to investigate the materials’ molecular structure, as shown in Fig. 3 The CA showed D and G bands at 1405 cm⁻¹ and 1605 cm⁻¹. In addition, the 2G band was absent in the aerogel, but it shows the 2D band at 2410 cm⁻¹. The D band represents disorder and defects present in carbon after preparation, while the G band represents the carbon stretches in the benzene rings. The carbon aerogel, D and G bands were seen at 1408 cm⁻¹ and 1609 cm⁻¹ (Lim et al., 2013; Skrzypek, 2021). The significant difference between the carbon aerogel and the ordinary aerogel is the 2G band at 1180 cm⁻¹ in the carbon aerogel to confirm stacked sp² and sp³ carbons in its ordered structure (Skrzypek, 2021).

Close

Fig. 3

(a): Raman spectroscopy of CA (carbon aerogel). (b): Raman spectroscopy of PPy (polypyrrole). (c): Raman spectroscopy of CA/PPy (carbon aerogel and polypyrrole composite).

Export to PPT

In addition, the 2D band of the CA shifted forward to 2470 cm⁻¹; this is higher than that at 2410 cm⁻¹ seen in the CA. The CA/PPy composite showed its D band at 1411 cm⁻¹ and G band at 1610 cm⁻¹. Its 2G and 2D bands were seen at 1181 cm⁻¹ and 2474 cm^−1, respectively (Samancı et al., 2021; Xu et al., 2017; Lim et al., 2013; Skrzypek, 2021). The I_d/I_g was additionally calculated to reflect the graphitisation level in the carbon aerogel. Before the carbonisation of the aerogel, the I_d/I_g ratio was 0.63. After carbonisation, the carbon aerogel showed a I_d/I_g ratio of 0.74, showing a higher degree of graphitisation in the carbon aerogel (Lim et al., 2013).

4.4 BET characterisation

The materials' surface area and porosity measurements were determined by N₂ adsorption/desorption isotherm. Fig. 4(a and b) show the isotherms of the CA-PPy composite material. According to Fig. 4(a) the CA-PPy composite typically shows an IUPAC Type IV isotherm with a H3 hysteresis loop. The N₂ adsorption at low pressure indicates that the material contains a mixture of mesopores and macropores. The BET measurement of the surface area of the CA-PPy composite material is 496 m²g⁻¹ with an average pore diameter of 3.9 nm. The Barrett-Joyner-Halenda (BJH) analysis of the material's pore size is shown in Fig. 4(b). It shows the hierarchical porosity of CA-PPy composite material and large mesopore distribution. The large specific surface area and mesoporous texture indicate that the as-prepared CA-PPy composite material possesses proper morphology suitable as electrode material in hybrid CDI cells.

Close

Fig. 4

(a): BET isotherm of CA- PPy (carbon aerogel and polypyrrole composite). (b): BJH Profile of pore size distribution of CA/PPy (carbon aerogel and polypyrrole composite).

Export to PPT

4.5 TGA characterisation

TGA was applied to study the thermal properties of CA, PPy, and CA-PPy. The thermograms of these materials are given in Fig. 5. The TGA analysis was conducted at 25 – 1000 °C and a heating rate of 10 °C/min. The CA showed thermal stability up to about 450 °C; at this temperature, a weight loss of 3 % was observed. Between 450 and 650 °C, the CA was significantly decomposed, and a mass loss of 97 % was recorded. The CA showed a two-step decomposition profile. The PPy and CA-PPy composite show similar thermal behaviour of a 3 % mass loss from 0 to 100 °C. They both exhibited a multi-step decomposition profile. The decomposition steps of the PPy are 3,10, 27, 9, and 3 %. A total of 52 % weight loss was recorded for the PPy, while that of the CA-PPy composite is 3,13, 22, 10, and 24 %. The PPy decomposed between 100 and 600 °C while the CA-PPy composite was at 100–700 °C. This data shows that the addition of the CA to the PPy slightly increased the thermal stability of the composite material and reduced the decomposition's starting temperature.

Close

Fig. 5

(a): Thermogram of CA (b): Thermogram of PPy (c): Thermogram of CA/PPy.

Export to PPT

4.6 TEM characterisation

The microstructures of CA and CA-PPy were studied from TEM images. Fig. 6(A-F) shows the TEM images. The CA (Fig. 6A and B) images show a disorderly structure with the presence of cavities. The polypyrrole (C and D) images show that the PPy has a more orderly spherical form. The CA-PPy composite images (E and F) shows a mixed structure distribution with cavities. The presence of the cavities in the structure of the CA-PPy composite material provides sites for ion intercalation in hybrid CDI operations (Cao et al., 2019b; Zhuo et al., 2019).

Close

Fig. 6

(a – b): TEM images of CA. (c – d): TEM images of PPy (e – f): TEM images of CA/PPy.

Export to PPT

4.7 SEM and EDAX characterisation

The microstructure of the materials was further investigated by SEM. The SEM images show a three-dimensional interconnected network of carbon aerogel flakes. The SEM images of the materials are shown in the Fig. 7(A-F). The EDAX spectrum of CA-PPy is displayed in Fig. 7G (insert). The carbon aerogels exist in different sizes with smooth surfaces, as seen in Fig. 8(A and B). The SEM images of Polypyrrole (Fig. 7C and D) shows a foam-like structure. Fig. 7(E and F) shows the SEM images of CA-PPy composite. The surface of the CA-PPy composite is rough compared to that of CA in Fig. 7(A and B); this suggests that the PPy is successfully coated on the CA in the CA-PPy composite.

Close

Fig. 7

(a – b): SEM images of CA. (c – d): SEM images of PPy. (e – f): SEM images of CA/PPy. (G – G¹): EDAX spectra showing the elemental composition of CA/PPy.

Export to PPT

Close

Fig. 7

(a – b): SEM images of CA. (c – d): SEM images of PPy. (e – f): SEM images of CA/PPy. (G – G¹): EDAX spectra showing the elemental composition of CA/PPy.

Export to PPT

Close

Fig. 8

(a): CV Profile of CA at 5 mV/s scan rate. (b): CV profile of CA/PPy at 5 mV/s scan rate.

Export to PPT

EDAX analysis was also employed to determine the elemental composition of the CA-PPy composite. The EDAX result of the CA-PPy composite is shown in Fig. 7G (insert). The chlorine in the material as reflected in the EDAX spectrum is most likely from the FeCl₃ oxidant used in the synthesis of the PPy.

4.8 CV and EIS tests of materials

The electrochemical properties of the materials were determined by cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS). The materials' CV profiles and Nyquist plots are shown in Figs. 8 and 9, respectively. The CV profile of CA, Fig. 8(a) and CA-PPy, Fig. 8(b) confirms that reversible reactions take place for the two samples. The smooth polygonal shape (without distinct anodic and cathodic peaks) of CA Fig. 8(a) shows its capacitive nature (Rambabu et al., 2020). In the CA-PPy CV profile, Fig. 8(b), the curves showing the anodic and cathodic current peaks are distinct and imply the presence of Faradaic current. This is indicative of the pseudo-capacitive property of the CA-PPy composite. The specific capacitance of the CA-PPy composite at 5 mV/s scan rate is 360.1F/g. This value agrees with that reported by Zhou H et al. 345F/g for CA-PPy. The specific capacitance of the CA at 5 mV/s is 115F/g. The increase in capacitance of the CA-PPy composite confirms that the addition of PPy to the CA improves its capacitance in the composite. The PPy is a battery material that has been reported to have a high energy density, hence it is used as a component of supercapacitors (Samancı et al., 2021; Kim et al., 2012). This supports the ability of PPy to enhance the capacitance of CA in the composite material. It also suggests that the CA/PPy composite material undergoes redox reactions resulting in pseudocapacitance (Samancı et al., 2021).

Close

Fig. 9

CV of CA/PPy at different scan rates.

Export to PPT

The CV profiles of the CA-PPy composite at different scan rates are also shown in Fig. 9. When the scan rate is increased to 50 mV/s, the shape of the voltammogram remains polygonal, with the cathodic and anodic current peaks intact. This means that an increase in the scan rate will not necessarily affect the ion adsorption/desorption mechanism at the electrical double layer of the electrode/electrolyte interface.

Fig. 10 shows the profile of the specific capacitance of the CA-PPy composite at different scan rates. A decrease in specific capacitance was observed at higher scan rates. At 5 mV/s, the specific capacitance is 360.1F/g and at 50 mV/s the scan rate decreased to 280.3F/g. The observation agrees with reports of similar research (Rambabu et al., 2020). The decrease in specific capacitance can be attributed to the insufficient time for ion diffusion into the gaps and inner pores of the CA-PPy composite electrode. The ion adsorption and desorption rates are slow at lower scan rates, and there isn’t sufficient time for ion diffusion. Hence the specific capacitance value is higher at lower scan rates.

Close

Fig. 10

Profile of the specific capacitance of CA-PPy at different scan rates.

Export to PPT

The electrochemical properties of the CA and CA-PPy composite were further investigated by EIS. The EIS studies were conducted at a potential of 0.22 V and a frequency range of 10⁵ Hz to 0.1 Hz. Fig. 11 shows the Nyquist plots of CA and CA-PPy. At a high frequency, the impedance acted as resistance, but as the frequency decreased, the impedance increased. The impedance line is about 45° to the imaginary axis, typical of porous electrode materials; it also confirms the dependence of the ions in the saline solution on frequency (Kandalkar et al., 2011). The charge transfer resistances were found to be 240 Ω for the CA sample, and 170 Ω for the CA-PPy composite. The CA-PPy composite depicted a smaller semi-circle than CA due to the improved charge transfer of electrons due to the presence of PPy. PPy is a known efficient electrical conductor due to its pi-pi delocalised bonds (Bose et al., 2011; Kulandaivalu et al., 2019). Therefore, the CA-PPy composite has a better charge transfer which could benefit ion transport, leading to an enhanced salt removal rate.

Close

Fig. 11

Nyquist plots of CA and CA-PPy at 50 mV/s scan rate.

Export to PPT

4.9 CDI desalination study

The CDI removal of Na⁺ and Cl^- by the CA-PPy composite was done by a batch mode experiment. Three different concentrations of saline solutions (300 mg/L, 500 mg/L, and 800 mg/L) were used, and the applied potential was 1.2 V. The anode was a 0.5 cm by 1.0 cm activated carbon modified with CA-PPy composite, while the cathode was a 0.5 cm by 1.0 cm activated carbon sheet. A 50 mL saline solution was transferred into a 100 mL beaker containing the two electrodes and a conductivity meter. The electrodes were then connected to a DC source. When the potential was applied, Na⁺ and Cl^- migrated to the anode and cathode, respectively. The experiment proceeded until the electrodes were saturated with the ions. The conductivity of the solution was monitored and recorded at intervals by the conductivity meter throughout the investigation. The saline solution was continually stirred with a magnetic stirrer to prevent the accumulation of charges in a particular region of the cell. The experimental setup is shown in Scheme 3.

Close

Scheme 3

Schematic of hybrid CDI desalination experimental set-up.

Export to PPT

On application of the potential, the conductivity of the solution dropped quickly for a while. A steady decline was observed until it became stable for about an hour, indicating ion adsorption saturation at the electrodes.

The experiment was repeated several times to evaluate the stability and ion retention capacity of the CA-PPy composite electrode after cycles of ion adsorption /desorption.

Fig. 12 (a) shows the profile of SAC vs Time for the different concentrations of NaCl solutions, while Fig. 12(b) shows that of the conductivity of NaCl solutions against time. Fig. 12(a) shows a sharp rise in SAC on applying the potential; this gradually reduces until saturation is reached. In Fig. 12(b), the conductivity of the NaCl solution decreases as ion adsorption proceeds. The salt recovery efficiency of the electrode material at the different concentrations of saline solutions was also determined from their conductivity values using Eq. (3) as displayed in Section 3.2.

Close

Fig. 12

(a) Graph of SAC vs Time. (b) Graph of Conductivity vs Time for the different concentrations.

Export to PPT

The electrochemical stability and ion retention capacity of the CA-PPy composite were evaluated by a study of several ion adsorption/desorption cycles. Fig. 13 shows the profile of the SAC and ion retention capacity after nine cycles of ion adsorption/desorption of (300, 500, and 800) mg/L of NaCl solutions.

Close

Fig. 13

Graph of SAC vs Ion Adsorption/Desorption Cycles.

Export to PPT

The SAC and ion retention values decrease gradually for each cycle, as shown in Table 1. This indicates the CA-PPy is an electrochemically stable electrode material. Table 2 compares the SAC and electrochemical properties of our material and that of similar materials that have been previously reported. The results indicate that the CA-PPy composite is a decent material for application in CDI desalination.