Abstract

Activated carbon particle electrodes modified by oxygen or nitrogen groups could be promising electrode candidates for capacitive deionization (CDI) processes. In this work, activated carbon particle electrodes were modified by phosphoric acid, nitric acid, urea, melamine, and zinc chloride to enhance desalination of an aqueous electrolytic solution. The modified activated carbon particles were characterized by scanning electron microscopy, energy dispersive X-ray spectroscopy, Fourier transform infrared spectroscopy, Brunauer–Emmett–Teller measurements and electrochemical scanning. The electrodes with oxygen or nitrogen groups on the surface exhibited a much higher desalination capacity and charge efficiency than those without any surface modification. Particularly, the activated carbon particle electrode modified by phosphoric acid exhibited a desalination capacity of 15.52 mg/g at 1.4 V in 500 mg/L NaCl solution, which was approximately eight times that of the unmodified electrode (2.46 mg/g). The enhancement was attributed to a higher specific capacitance, a lower electrochemical impedance and an increase in oxygen or nitrogen-containing groups on the surface.

INTRODUCTION

Electrosorption technology (EST), also called capacitive deionization (CDI), is a novel and economical water treatment technology in which ions are attracted to the surface of the electrodes when an electric potential is applied (Welgemoed & Schutte 2005; Oren 2008; Foo & Hameed 2009). Currently, the most successful desalination technologies include multi-effect distillation, multi-stage flashing, electrodialysis, and reverse osmosis (Gabelich et al. 2002; Lee et al. 2006; Xu et al. 2008; Zou et al. 2008). In these desalination technologies, the pretreatment of raw water in multi-effect distillation cannot reach standard, which results in scaling and low energy transfer efficiency. The operation cost and energy consumption are high, because multi-stage flashing requires a large amount of fluid transportation and a large heat transfer area (Gabelich et al. 2002). Electrodialysis is based on a high voltage to force directional ion transfer, which leads to electrode corrosion (Dong et al. 2017). Reverse osmosis is a technology that uses a semi-permeable membrane to purify water. The maintenance and operating costs are high because of the restriction of the semi-permeable membrane and operating pressure (Zou et al. 2008). In addition, the regeneration of reverse osmosis membrane will produce secondary pollution. One can see that the development of traditional desalination technologies is constrained by high energy consumption, secondary pollution, and low efficiency. Therefore, it is important to find low energy consumption and environmentally friendly desalination technology.

CDI is a non-Faradaic process and its operation remains limited by the electrochemical stability of water, translating to a cell voltage of 1.1–1.5 V (Yu et al. 2013; Kim et al. 2018). Carbonaceous electrodes are one of the research hotspots of CDI technology. Pan's group (Zhan et al. 2011) prepared activated carbon fibers and carbon nanofiber (ACF/CNF) composite electrode materials by chemical vapor deposition, and the maximum electrosorption capacity was 17.19 mg/g. The maximum electrosorption capacity of nitrogen-doped graphene was 4.81 mg/g, which is much higher than that (3.85 mg/g) of unmodified graphene (Xu et al. 2015). Qiu's group (Wang et al. 2013; Wu et al. 2016) used low-cost and high mesoporous activated carbon or 4-vinylpyridine as a modifier; the electrosorption capacities were 4.60 mg/g and 9.60 mg/g. Wang et al. (2007) used carbon nanotubes (CNT) as electrosorption electrodes and observed a capacity of 9.35 mg/g for sodium chloride. Wang et al. (2015) optimized the electrosorption device with a maximum electrosorption capacity of 5.00 mg/g.

Recently, activated carbon has been a research hotspot because of its low cost, large specific surface area, and high mechanical strength (Huang et al. 2014; Niu et al. 2015; Yeh et al. 2015; Cheng et al. 2019). Currently, activated carbon features a low capacitance and leads to a low removal rate, which limits the industrial application of CDI. Surface modification of activated carbon particle electrode is regarded as an effective method to solve this problem. Surface modification mainly contains heteroatom doping, metal oxide-modified carbon material, and chemical treatment of carbon material (Cheng et al. 2019). Heteroatom doping mainly contains N-doped carbon material (Kurak & Anderson 2009), multi-element doping (Ma et al. 2017), and ion-doped carbon material (Zhang et al. 2016b). In fact, not all elements are suitable for heteroatom doping. N, P, S, and Cl are commonly selected. TiO2, MnO2, and ZnO are utilized to modify carbon electrodes in the method of metal oxide-modified carbon material (Schmidt-Mende & MacManus-Driscoll 2007; Chang et al. 2011; Kim et al. 2014; Zhao et al. 2017). Chemical treatment of carbon material is a common method. It mainly includes acid treatment and alkali treatment (Cheng et al. 2019). HNO3, H2SO4 (Huang et al. 2014; Niu et al. 2015) and KOH (Yeh et al. 2015) are commonly used. In this work, H3PO4, HNO3, CON2H4, C3N3(NH2)3, and ZnCl2 were used as modifiers to study the effects of oxygen or nitrogen-containing groups on the surface morphology, specific surface area, and specific capacitance of activated carbon. Electrochemical analysis was carried out and the relationship between the functional groups and the electrosorption capacity was studied, so as to improve electrosorption technology for water treatment.

METHODS

Treatment of activated carbon particles

First, to remove impurities, the activated carbon particles (Fuzhou Yihuan Carbon Co., Ltd, China) were boiled in deionized water and washed with deionized water until the conductivity was less than 10 μS/cm. The particle sizes of the raw activated carbon are shown in Table 1 and Figure 1.

Table 1

Data of activated carbon particles' size and volume percent

Size (nm)Volume percent (%)
58.77 24.1 
68.06 49.1 
78.82 25.9 
91.28 0.9 
Size (nm)Volume percent (%)
58.77 24.1 
68.06 49.1 
78.82 25.9 
91.28 0.9 
Figure 1

Distribution by size of activated carbon particles.

Figure 1

Distribution by size of activated carbon particles.

Next, the activated carbon particles were dried at 60 °C (Machunda et al. 2009; Jiang et al. 2018). Next, phosphoric acid, nitric acid, urea, melamine, and zinc chloride were used (Shanghai Titan Technology Co., Ltd, China) to modify the activated carbon particles according to Table 2 (Zhang et al. 2016a). Finally, the activated carbon particles were washed with deionized water until the filtrate became neutral and then were dried at 60 °C. These particles are referred to as modified activated carbon particles.

Table 2

Activated condition

ModifiersConcentrationSoaking time (h)Soaking temperature (°C)
Phosphoric acid (P) 50 wt% 36 50 
Nitric acid (O) 5 mol/L 36 50 
Urea (U) 1,000 g/L 36 50 
Melamine (M) 15 g/L 36 50 
Zinc chloride (Zn) 80 g/L 36 50 
ModifiersConcentrationSoaking time (h)Soaking temperature (°C)
Phosphoric acid (P) 50 wt% 36 50 
Nitric acid (O) 5 mol/L 36 50 
Urea (U) 1,000 g/L 36 50 
Melamine (M) 15 g/L 36 50 
Zinc chloride (Zn) 80 g/L 36 50 

Fabrication of the carbon electrode

The electrode material was a mixture of activated carbon particles, conductive carbon black (Fuzhou Yihuan Carbon Co., Ltd, China), and polyvinylidene fluoride (PVDF, Arkema Chemical Co., Ltd, France) at a mass ratio of 8:1:1 (Jiang et al. 2018). First, PVDF was dissolved in dimethylacetamide (DMAC, Shanghai Titan Technology Co., Ltd, China) and placed into an ultrasonic reactor until it dissolved. Then, the activated carbon particles and conductive carbon black were added. The mixture was stirred for 12 hr until it was uniformly mixed. Lastly, the slurry was uniformly cast onto a graphite sheet (Shandong Baofeng Carbon Co., Ltd, China) with a blade to form electrodes with dimensions of 4 × 4 cm. The casted electrodes were dried at 60 °C for 4 hr and then placed in a vacuum oven for 4 hr to remove the DMAC residual on the electrode.

Electrosorption experiments

The desalination experiments were performed in a self-made CDI system, consisting of a CDI cell, a peristaltic pump (BT100-1 L, Lange Constant Flow Pump Co., Ltd, China), a conductivity meter (Multi 350i, Beijing Yised Technology Development Co., Ltd, China) and an external power supply (MS305D, Guangdong Meisheng Electronics Co., Ltd, China). The CDI module included four graphite sheets coated with activated carbon particles, and the distance between the positive and negative electrodes was 2 mm. The CDI module and experimental device are shown in Figure 2.

Figure 2

Experimental device: (a) flat operating mode and (b) activated carbon particle electrode electrosorption device.

Figure 2

Experimental device: (a) flat operating mode and (b) activated carbon particle electrode electrosorption device.

During desalination, a solution of sodium chloride (Shanghai Titan Technology Co., Ltd, China) at different concentrations (100–2,000 mg/L) was prepared. The corresponding conductivity was determined. One hundred milliliters of the NaCl solution at 500 mg/L was continuously recycled in the system at the flow rate of 5 mL/min by a peristaltic pump. The conductivity of the solution is 1,007 μS/cm. The value of pH is 6.38. The potential applied to the electrode was 1.4 V. The conductivity of the solution at a given time was recorded, by which the performance of the electrode can be calculated according to the following equation: 
formula
where C0 and Ce are the initial concentration and the solution concentration at adsorption equilibrium, respectively (mg/L). V is the volume of the electrolyte (L), m is the mass of active materials on the cathode and anode (g). Es is the electrosorption capacity.

Measurements

The specific surface area of the samples was determined using a fully automatic microporous physical adsorption instrument (ASAP2020, Micromeritics, USA) based on the Brunauer–Emmett–Teller (BET) method. Fourier transform infrared spectroscopy (FTIR, Thermo Nicolet Corporation, USA) measurements were carried out to identify the chemical structure and functional groups on the activated carbon particles. Scanning electron microscopy (SEM, S-3400N, Hitachi, Japan) and energy dispersive X-ray spectroscopy (EDS, Falion 60S, EDAX, USA) were applied to obtain the morphology and the elemental compositions of the activated carbon particles, respectively.

Electrochemical measurements were carried out using a conventional three-electrode system on an electrochemical workstation (CHI610D, Shanghai Chenhua Instrument Co., Ltd, China) at room temperature. In the measurements, the activated carbon particle and modified activated carbon particle electrodes were used as the working electrode. The working electrode is made of unmodified activated carbon particles or modified activated carbon particles coated on a glassy carbon electrode with a diameter of 3 mm, while a platinum net and an Ag/AgCl electrode (Shanghai Yueci Co., Ltd, China) were used as the counter and reference electrodes, respectively. A 0.5 mol/L NaCl electrolyte solution was used.

Cyclic voltammetry (CV) measurements were performed in a potential range of −0.4 V to 0.6 V at a scan rate of 10 mV/s and 50 mV/s (Feng et al. 2018). The specific capacitance was calculated as follows: 
formula
where Cs is the specific capacitance of the active material (F/g), I is the instantaneous current (A), m is the mass of the active material (g), Uf and Ui represent the potential limits at low and high levels (V), and υ is the potential scan rate (mV/s).

Electrochemical impedance spectroscopy (EIS) measurements were performed at open circuit in the frequency range of 0.1 Hz to 100,000 Hz.

Tafel measurements were performed within the potential of range of −1 V to 1 V.

RESULTS AND DISCUSSION

Surface properties of activated carbon particles

Figure 3 shows SEM images of activated carbon particles. It can be seen from Figure 3 that the morphology of the modified activated carbon has a certain degree of change. The surface becomes rough and uneven. The surface of activated carbon particles modified by phosphoric acid is rough and uneven. The number of particles on the surface of the activated carbon particles modified by nitric acid and urea are obviously increased, the surface is eroded, resulting in corrosion and collapse of the pore structure of activated carbon particles, forming a rough surface morphology structure. The activated carbon particles modified with zinc chloride and melamine are loose.

Figure 3

Scanning electron micrograph (SEM) of different activated carbon particles (×5,000).

Figure 3

Scanning electron micrograph (SEM) of different activated carbon particles (×5,000).

N, Cl, Zn, P, O, and other elements can be identified on the surfaces of the modified activated carbon particles, mainly due to the introduction of oxygen-containing, nitrogen-containing, and phosphorus-containing functional groups, as displayed in Table 3.

Table 3

EDS elemental distribution of different activated carbon particles

Main elements of sampleC (%)O (%)N (%)P (%)Zn (%)Cl (%)
80.19 11.37 
39.72 32.09 18.63 
70.24 18.50 2.69 
57.17 18.27 13.77 
78.42 12.47 3.11 
Zn 65.68 9.41 6.94 7.67 
Main elements of sampleC (%)O (%)N (%)P (%)Zn (%)Cl (%)
80.19 11.37 
39.72 32.09 18.63 
70.24 18.50 2.69 
57.17 18.27 13.77 
78.42 12.47 3.11 
Zn 65.68 9.41 6.94 7.67 

The change in functionality on the activated carbon particles' surfaces after chemical modification was evidenced in FTIR measurements. Figure 4 presents the FTIR spectra of the activated carbon particles before and after chemical modification. The peak at 3,430 cm−1 is the stretching vibration of hydroxyl (-OH) group. The intensity was greatly increased after the modification with phosphoric acid, nitric acid, and urea. The peak was also enhanced when the activated carbon particles were modified by melamine and zinc chloride. These results indicate that the hydroxyl radicals on the surface of activated carbon particles were enhanced by different acid/base modifiers. After the phosphoric acid modification, a new peak appeared at 980 cm−1, due to the stretching vibration of the P-O-C or P-OH group (Jiang et al. 2018). In addition, a weak peak appeared at 1,080 cm−1, which may be ascribed to the P = O stretching vibration. After nitric acid modification, a strong peak appeared at 1,380 cm−1, which was related to the symmetric stretching of nitro groups bonded with alkane chains (Huang et al. 2014). In addition, there was a peak at 1,380 cm−1, for activated carbon particles modified by urea and melamine. After the modification with melamine, hydrocarbon bands appeared at 2,913 cm−1 and 2,842 cm−1, due to the antisymmetric and symmetric stretching vibrations of CH2, indicating that hydrocarbons were indeed introduced by melamine modification. After the modification with urea, a strong peak around 1,159 cm−1 appeared, which was the stretching vibration peak of C-N. There was little change after the modification by zinc chloride, which indicated that zinc chloride may only affect the surface morphology but had little effect on changing the functional group. In summary, it can be seen from the FTIR analysis that, except for zinc chloride, the other modifiers had a certain influence on the number and type of surface functional groups of the activated carbon particles.

Figure 4

FTIR spectra of the different activated carbon particles.

Figure 4

FTIR spectra of the different activated carbon particles.

The pore size and distribution of the activated carbon particles' materials were studied with N2 adsorption/desorption isotherms, as shown in Figure 5. The surface area and pore diameter of the activated carbon particles are given in Table 4. The modified activated carbon particles had a smaller specific surface area. In general, the electrosorption of electrolytic ions on activated carbon particles closely corresponds to the specific surface area. The higher the specific surface area, the larger the electrosorption capacity. Accordingly, in this study, the desalination improvement by the modification of activated carbon particles was not realized by increasing the surface area (Huang et al. 2014). In addition, the results showed that the physisorption isotherms for all activated carbon particles' samples followed a Langmuir monolayer type (Type 1), typically found in correlation with micropores. Figure 5(b) shows that the pore size distribution of samples is mainly concentrated in the range of micropores (<2 nm), but some are also in the mesopore range (2–50 nm). A higher specific surface area can be obtained for the micropores, and mesopores are conducive to the rapid transmission of ions, leading to a higher adsorption quantity (Porada et al. 2012).

Table 4

Microstructure parameters of different activated carbon particles

SampleSBET (m2/g)SLangmuir (m2/g)St-plot (m2/g)DBET (nm)DBJH (nm)Vtotal (cm3/g)
2,006.13 2,657.60 1,782.17 1.9085 2.4067 0.957 
1,907.45 2,548.18 1,679.14 1.9184 2.3927 0.915 
1,199.65 1,563.17 1,068.61 1.8822 2.5685 0.565 
1,918.74 2,518.13 1,709.93 1.8737 2.4067 0.899 
1,278.53 1,698.70 1,154.11 1.8699 2.3889 0.598 
Zn 1,928.33 2,552.56 1,694.67 1.8969 2.3996 0.915 
SampleSBET (m2/g)SLangmuir (m2/g)St-plot (m2/g)DBET (nm)DBJH (nm)Vtotal (cm3/g)
2,006.13 2,657.60 1,782.17 1.9085 2.4067 0.957 
1,907.45 2,548.18 1,679.14 1.9184 2.3927 0.915 
1,199.65 1,563.17 1,068.61 1.8822 2.5685 0.565 
1,918.74 2,518.13 1,709.93 1.8737 2.4067 0.899 
1,278.53 1,698.70 1,154.11 1.8699 2.3889 0.598 
Zn 1,928.33 2,552.56 1,694.67 1.8969 2.3996 0.915 
Figure 5

(a) N2 adsorption/desorption isotherm and (b) the pore size distribution of different activated carbon particles.

Figure 5

(a) N2 adsorption/desorption isotherm and (b) the pore size distribution of different activated carbon particles.

Electrochemical properties of the activated carbon particles

To investigate the electrochemical properties of activated carbon particles' materials as electrodes, CV, EIS, and Tafel measurements were conducted in a 0.5 M NaCl solution.

The electrochemical performances of the activated carbon particle electrodes before and after chemical modification were obtained by CV measurements with a conventional three-electrode system. Figure 6(a) and 6(b) show the CV curves of electrodes at different scan rates. These curves were similar to the rectangular shape in the potential range of −0.4 V to 0.6 V. The enclosed areas are large and no significant redox peaks are observed. The results show that there is no Faraday electron transfer on the surface of the electrode. The charge transfer mainly occurs in the double layer and the material has a good charge transfer rate and electrochemical performance (Cai et al. 2017). In addition, the results indicate that only electrosorption/desorption occurs on the electrode. Electrosorption is a reversible process. The ideal shape of the CV curve is rectangular. However, in this case, the shape of the CV curve is not a standard rectangle. This might be attributed to the internal polarization resistance, according to the ‘aperture matching principle’ (Ghadimi et al. 2011). In addition, Faradaic reaction is another reason for this (Oh et al. 2006). Figure 6(c) shows the specific capacitance of activated carbon particle electrodes at 50 and 10 mV/s, and the order of the specific capacitance is P > O > U > M > Zn > B.

Figure 6

Cyclic voltammograms obtained using the activated carbon particles and the modified activated carbon particles as the electrodes: (a) CV curve at 50 mv/s, (b) CV curve at 10 mv/s, and (c) the specific capacitance of activated carbon particle electrodes as a function of the scan rate.

Figure 6

Cyclic voltammograms obtained using the activated carbon particles and the modified activated carbon particles as the electrodes: (a) CV curve at 50 mv/s, (b) CV curve at 10 mv/s, and (c) the specific capacitance of activated carbon particle electrodes as a function of the scan rate.

Figure 7 shows the potential of zero charge of electrode materials at 1 mV/s in 4.3 mM deaerated NaCl. The surface charge could be positively or negatively enhanced by chemical modification. In this study, the potential of zero charge of modified activated carbon electrodes has been positively enhanced by chemical modification, due to the introduction of negative functional groups (Gao et al. 2014, 2015, 2017).

Figure 7

Measurement of zero charge potential of the activated carbon particles and the modified activated carbon particles as the electrodes at 1 mV/s in 4 mM deaerated NaCl.

Figure 7

Measurement of zero charge potential of the activated carbon particles and the modified activated carbon particles as the electrodes at 1 mV/s in 4 mM deaerated NaCl.

The Nyquist plots for the activated carbon particles and the modified activated carbon particles as the electrodes are illustrated in Figure 8. Clearly, all the plots contained two parts. (1) An incomplete semicircle at high frequency, which presents the reaction impedance at the electrode–solution interface. This semicircle is mainly caused by the oxidation-reduction reaction of metal impurities and oxygen-containing groups in the electrode materials. The larger the diameter of the semicircle, the larger the reaction resistance. (2) The non-straight line in the low-frequency zone, which is indicative of the diffusion impedance. The longer the diagonal, the larger the diffusion impedance (Jia et al. 2018). The diameter of the activated carbon particle electrode modified by zinc chloride increases obviously in the high-frequency region, indicating that the reaction impedance of the electrode is larger than that of the other electrodes. The oblique lines in the low-frequency region are biased towards the imaginary axis, showing that the diffusion impedance is the capacitive impedance. The linear slope of the modified activated carbon particle electrodes are higher, which demonstrates that the modified activated carbon particle electrodes exhibit a faster ion-diffusion process.

Figure 8

Nyquist plots for the activated carbon particle and the modified activated carbon particle electrodes.

Figure 8

Nyquist plots for the activated carbon particle and the modified activated carbon particle electrodes.

The corresponding Tafel plots are shown in Figure 9. The fitting parameters are illustrated in Table 5. The higher the corrosion potential, the stronger the corrosion resistance of the electrode. The lower the corrosion current, the smaller the corrosion rate. In general, the corrosion potential of the modified activated carbon particle electrodes is obviously higher than that of the unmodified electrodes, which indicates that the modification of the activated carbon particles enhanced the corrosion resistance of the electrodes. In addition, the larger the corrosion resistance, the larger the diameter of the semicircle in the Nyquist plot. In Figure 8, the activated carbon particle electrode modified by zinc chloride has the largest diameter, and the unmodified activated carbon particle electrode has the smallest diameter. Accordingly, their corrosion potential is the largest and the smallest, respectively, which is in accordance with the results in the Nyquist plots.

Table 5

Fitting parameters of the Tafel curve

SamplesBPOUMZn
Corrosion potential (V) −0.479 −0.436 −0.224 −0.249 −0.275 −0.189 
Corrosion current density (A/cm20.00386 0.00514 0.00650 0.00462 0.00375 0.00646 
SamplesBPOUMZn
Corrosion potential (V) −0.479 −0.436 −0.224 −0.249 −0.275 −0.189 
Corrosion current density (A/cm20.00386 0.00514 0.00650 0.00462 0.00375 0.00646 
Figure 9

Tafel curves of the activated carbon particle and the modified activated carbon particle electrodes.

Figure 9

Tafel curves of the activated carbon particle and the modified activated carbon particle electrodes.

Desalination performance of the activated carbon particle electrodes

The electrosorptive desalination performance of the activated carbon particle and the modified activated carbon particle electrodes was studied at 1.4 V in 500 mg/L NaCl. Figure 10(a) depicts the electrosorption/desorption performance of the activated carbon particle and the modified activated carbon particle electrodes. As observed, in the electrosorption stage, there is a dramatic increase in the adsorption capacity, showing the removal of salt ions from the aqueous solution. In addition, the activated carbon particle electrode modified by phosphoric acid has the largest ion electrosorption capacity, with a desalination capacity of 15.52 mg/g. The capacity of electrosorption is much larger than that of the electrodes mentioned at the beginning, such as the nitrogen-doped graphene, and activated carbon particles modified by 4-vinylpyridine. The order of the electrosorption capacity is P > O > U > M > Zn > B, which is in accordance with the specific capacitance. The regeneration of the electrodes was also essential for the electroadsorption, which can be regenerated by applying a 0.0 V voltage. The adsorption capacity decreased dramatically during the desorption process, and the recovery rate was above 90%.

Figure 10

Electrosorption/desorption profiles of the activated carbon particle and the modified activated carbon particle electrodes: (a) adsorption capacity of the activated carbon particle and modified activated carbon particle electrodes and (b) the applied voltage on the electrodes.

Figure 10

Electrosorption/desorption profiles of the activated carbon particle and the modified activated carbon particle electrodes: (a) adsorption capacity of the activated carbon particle and modified activated carbon particle electrodes and (b) the applied voltage on the electrodes.

Regenerability and stability of the activated carbon particle electrodes

The regenerability and stability of the electrodes with unmodified and modified activated carbon particles in the CDI process were investigated. The regenerability of the electrodes was carried out through three charge–discharge cycles in a 500 mg/L NaCl solution. The results are illustrated in Figure 11. In addition, cyclic desalination capacity of the activated carbon particle and the modified activated carbon particle electrodes in 10 cycles are shown in Table 6. The modified activated carbon particle electrodes presented a relatively high and stable desalination capacity, demonstrating that the addition of modifiers improves the stability and regeneration of the electrodes.

Table 6

Cyclic desalination capacity of the activated carbon particle and the modified activated carbon particle electrodes in 10 cycles

12345678910
2.46 1.76 1.41 2.11 1.76 1.41 1.05 1.05 0.70 0.70 
15.52 13.96 15.13 15.13 14.35 14.74 14.35 13.96 13.58 13.19 
10.43 9.31 9.68 9.68 10.05 9.31 9.68 9.31 8.56 9.31 
9.22 8.48 8.48 8.48 8.11 7.74 8.11 7.74 7.37 7.74 
7.14 5.26 5.63 6.39 6.39 6.01 5.63 5.63 5.26 5.26 
Zn 4.63 3.56 3.92 3.92 4.27 3.92 3.56 3.917 3.56 3.56 
12345678910
2.46 1.76 1.41 2.11 1.76 1.41 1.05 1.05 0.70 0.70 
15.52 13.96 15.13 15.13 14.35 14.74 14.35 13.96 13.58 13.19 
10.43 9.31 9.68 9.68 10.05 9.31 9.68 9.31 8.56 9.31 
9.22 8.48 8.48 8.48 8.11 7.74 8.11 7.74 7.37 7.74 
7.14 5.26 5.63 6.39 6.39 6.01 5.63 5.63 5.26 5.26 
Zn 4.63 3.56 3.92 3.92 4.27 3.92 3.56 3.917 3.56 3.56 
Figure 11

Cyclic desalination capacity of the activated carbon particle and the modified activated carbon particle electrodes.

Figure 11

Cyclic desalination capacity of the activated carbon particle and the modified activated carbon particle electrodes.

Charge efficiency and energy consumption of different activated carbon particle electrodes

The charge efficiency and energy consumption of the electrodes with unmodified and modified activated carbon particles in the CDI process are calculated in Table 7. The modified activated carbon particle electrodes presented a relatively high charge efficiency and low energy consumption.

Table 7

Charge efficiency and energy consumption of the activated carbon particle and the modified activated carbon particle electrodes

SamplesCharge efficiency (%)Energy consumption (Wh/g)
15.2 0.320 
46.6 0.224 
45.8 0.227 
43.9 0.237 
42.2 0.247 
Zn 34.6 0.301 
SamplesCharge efficiency (%)Energy consumption (Wh/g)
15.2 0.320 
46.6 0.224 
45.8 0.227 
43.9 0.237 
42.2 0.247 
Zn 34.6 0.301 

CONCLUSIONS

Activated carbon particle electrodes with a high specific surface area and different functional groups have been facilely fabricated through a simple chemical method. The modified activated carbon particle electrodes exhibited an excellent electrosorption capacity. Particularly, the activated carbon particles modified by phosphoric acid exhibited a desalination capacity of 15.52 mg/g at 1.4 V in 500 mg/L NaCl solution, which was around eight times that of the unmodified activated carbon particles (2.46 mg/g). In addition, the modified activated carbon particle electrodes also presented great regenerability, stability, charge efficiency, and a relatively low energy consumption. This is attributed to nitric acid and phosphoric acid treatments that increased the amount of oxygen-containing groups. Urea and melamine treatments increased the amount of nitrogen-containing groups on the activated carbon particles' surface. Thus, increasing the specific capacitance of the activated carbon particle electrodes results in stronger electrosorption of electrolytic ions on the modified activated carbon particles.

ACKNOWLEDGEMENTS

This work was supported by the National Natural Science Foundation of China (NSFC) (21876050, 51273196), the National Key Research and Development Plans of Special Project for Site Soils (2018YFC1800600) and the Special Fund from State Key Joint Laboratory of Environment Simulation and Pollution Control (18K10ESPCT).

REFERENCES

REFERENCES
Cai
W.
Yan
J.
Hussin
T.
Liu
J.
2017
Nafion-AC-based asymmetric capacitive deionization
.
Electrochimica Acta
225
,
407
415
.
Cheng
Y.
Hao
Z.
Hao
C.
Deng
Y.
Li
X.
Li
K.
Zhao
Y.
2019
A review of modification of carbon electrode material in capacitive deionization
.
RSC Advances
9
(
42
),
24401
24419
.
Gabelich
C. J.
Tran
T. D.
Suffet
I. H.
2002
Electrosorption of inorganic salts from aqueous solution using carbon aerogels
.
Environmental Science and Technology
36
(
13
),
3010
3019
.
Gao
X.
Omosebi
A.
Holubowitch
N.
Landon
J.
Liu
K.
2017
Capacitive deionization using alternating polarization: effect of surface charge on salt removal
.
Electrochimica Acta
233
,
249
255
.
Ghadimi
A.
Saidur
R.
Metselaar
H. S. C.
2011
A review of nanofluid stability properties and characterization in stationary conditions
.
International Journal of Heat and Mass Transfer
54
(
17–18
),
4051
4068
.
Jiang
S.
Wang
H.
Xiong
G.
Wang
X.
Tan
S.
2018
Removal of nitrate using activated carbon-based electrodes for capacitive deionization
.
Water Supply
18
(
6
),
2028
2034
.
Kurak
K.
Anderson
A.
2009
Nitrogen-treated graphite and oxygen electroreduction on pyridinic edge sites
.
Journal of Physical Chemistry C
113
(
16
),
6730
6734
.
Ma
X.
Song
X.
Ning
G.
Hou
L.
Kan
Y.
Xiao
Z.
Li
W.
Ma
G.
Gao
J.
Li
Y.
2017
S-doped porous graphene microspheres with individual robust red-blood-cell-like microarchitecture for capacitive energy storage
.
Industrial & Engineering Chemistry Research
56
(
34
),
9524
9532
.
Machunda
R. L.
Jeon
H.
Lee
J. K.
Lee
J.
2009
Effects of acid treatment on the improvement of specific capacitance and deionization efficiency of porous carbon electrodes
.
Water Science and Technology: Water Supply
9
(
2
),
159
165
.
Oh
H.
Lee
J.
Ahn
H.
Jeong
Y.
Kim
Y.
Chi
C.
2006
Nanoporous activated carbon cloth for capacitive deionization of aqueous solution
.
Thin Solid Films
515
(
1
),
220
225
.
Porada
S.
Weinstein
L.
Dash
R.
Van Der Wal
A.
Bryjak
M.
Gogotsi
Y.
Biesheuvel
P. M.
2012
Water desalination using capacitive deionization with microporous carbon electrodes
.
ACS Applied Materials and Interfaces
4
(
3
),
1194
1199
.
Schmidt-Mende
L.
MacManus-Driscoll
J. L.
2007
Zno – nanostructures, defects, and devices
.
Materials Today
10
(
5
),
40
48
.
Wang
S.
Wang
D.
Ji
L.
Gong
Q.
Zhu
Y.
Liang
J.
2007
Equilibrium and kinetic studies on the removal of NaCl from aqueous solutions by electrosorption on carbon nanotube electrodes
.
Separation and Purification Technology
58
(
1
),
12
16
.
Wang
G.
Qian
B.
Dong
Q.
Yang
J.
Zhao
Z.
Qiu
J.
2013
Highly mesoporous activated carbon electrode for capacitive deionization
.
Separation and Purification Technology
103
,
216
221
.
Welgemoed
T. J.
Schutte
C. F.
2005
Capacitive deionization technology™: an alternative desalination solution
.
Desalination
183
(
1
),
327
340
.
Xu
X.
Pan
L.
Liu
Y.
Lu
T.
Sun
Z.
2015
Enhanced capacitive deionization performance of graphene by nitrogen doping
.
Journal of Colloid and Interface Science
445
,
143
150
.
Yu
J.
Qin
J.
Kekre
K. A.
Viswanath
B.
Tao
G.
Seah
H.
2013
Impact of operating conditions on performance of capacitive deionisation for reverse osmosis brine recovery
.
Journal of Water Reuse and Desalination
4
(
2
),
59
64
.
Zhang
L.
Lu
Z.
Li
D.
Ma
J.
Song
P.
Huang
G.
Liu
Y.
Cai
L.
2016a
Chemically activated graphite enhanced oxygen reduction and power output in catalyst-free microbial fuel cells
.
Journal of Cleaner Production
115
,
332
336
.
Zhao
C.
Lv
X.
Li
J.
Xie
T.
Qi
Y.
Chen
W.
2017
Manganese oxide nanoparticles decorated ordered mesoporous carbon electrode for capacitive deionization of brackish water
.
Journal of the Electrochemical Society
164
(
13
),
505
511
.