Adsorption of hexavalent chromium by polyacrylonitrile-based porous carbon from aqueous solution

Owing to the unique microporous structure and high specific surface area, porous carbon could act as a good carrier for functional materials. In this paper, polyacrylonitrile (PAN)-based porous carbon materials (PPC-0.6-600, PPC-0.8-600, PPC-0.6-800 and PPC-0.8-800) were prepared by heating KOH at 600°C and 800oC for the removal of Cr(VI) from aqueous solution. The adsorbent was characterized by the techniques of Fourier transform infrared spectroscopy (FT-IR), elementary analysis, scanning electron microscopy (SEM), X-ray photoelectron spectroscopy (XPS) and N2 adsorption techniques. The results showed that the adsorption capacity increased with decreasing pH value of the initial solution. The adsorption capacity of Cr(VI) on PPC-0.8-800 was much greater than that on other materials, and maximum adsorption capacity were calculated to be 374.90 mg g−1. Moreover, PPC-0.8-800 had superior recyclability for the removal of Cr(VI) from wastewater, about 82% of its initial adsorption capacity was retained even after five cycles. The result of kinetic simulation showed that the adsorption of Cr(VI) on the PAN-based porous carbon could be described by pseudo-second-order kinetics. The adsorption process was the ionic interaction between protonated amine groups of PPC and HCrO4- ions.


Introduction
In recent years, pollution of surface and underground water resources with toxic Cr(VI) has become a major environmental problem attracting much more attention to develop and operated at 10 kV. FTIR measurements were performed using a Fourier Transform Infrared Spectrometer (Nicolet FT-IR 380) with KBr as background over the range of 4000-400 cm −1 . The X-ray photoelectron spectra (XPS) were recorded with an ESCALAB 250 (Thermo Electron), the X-ray excitation was provided by a monochromatic Al Kα (1486.6 eV) source.

Adsorption experiments
A series of different concentrations of Cr(VI) solution were prepared by dissolving K 2 Cr 2 O 7 in distilled water. The pH value was adjusted by drop-wise addition of 1 mol l −1 HCl or 1 mol l −1 NaOH solution. The concentration of Cr(VI) was measured by UV-vis spectroscopy at λ = 540 nm via photometric diphenylcarbohydrazide method. Colour indicator was prepared with 2 g l −1 1,5diphenylcarbohydrazide.
Adsorption experiments were performed as follows: the influence of pH on Cr(VI) adsorption onto PPC was performed by varying solution pH value from 1.0 to 6.0. First, 0.05 g PPC-0.6-600, PPC-0.6-800, PPC-0.8-600 and PPC-0.8-800 were put into 100 ml beaker which contained 50 ml of 421.80 mg l −1 Cr(VI) solution. Then the solution was stirred at 150 r.p.m. at 25°C for 24 h. Kinetic experiment was conducted at Cr(VI) 235.31 mg l −1 initial concentration and pH = 2. The solution samples were removed at different time period. Equilibrium experiment was performed using different concentrations of Cr(VI), including 123.33 mg l −1 , 231.48 mg l −1 , 288.78 mg l −1 , 448.70 mg l −1 , 614.50 mg l −1 , 740.00 mg l −1 and 861.00 mg l −1 , and pH = 2. PPC adsorption amount (q t ) can be calculated as follows: where q t (mg g −1 ) represents the amount of Cr(VI) adsorbed onto PPC, C 0 and C t represent the Cr(VI) concentration in solution at initial time and at time t (mg l −1 ), respectively, V(L) represents the solution volume and m(g) represents the mass of adsorbent.

SEM analysis
SEM images of the raw and treated samples are shown in figure 1. It can be seen that PPC-0.6-800 maintains a certain fibrous morphology, which is similar to PAN fibres. The fibrous structure was broken when the ratio of KOH/PPC was 0.8 which suggests many structural defects are formed after KOH activation, and a large number of macroholes were observed in their external surface as a result of burning off some carbon species. Comparing with PPC-0.8-600 and PPC-0.8-800, the higher the reaction temperature is, the deeper the degree of etching.

BET analysis
N 2 adsorption-desorption isotherms of porous carbons are shown in figure 2a; the isotherms of PPC-0.6-600, PPC-0.6-800, PPC-0.8-600 and PPC-0.8-800 exhibited the type I isotherm characteristic of microporous materials. The pore-size distribution plot of the PPC in figure 2b shows that the average diameter of micropores is in the range of 0.8-2 nm, while that of mesopores is in the ranges of 2-3 nm. It has been reported that the micropores can provide a large number of adsorption sites, and the mesopores and macropores can facilitate the transfer of heavy metal ions (e.g. Cr, Pb, Cd, etc.) [30]. Obviously, the unique hierarchical porous structure of PPC is desirable for efficient and rapid removal of Cr(VI), leading to the improvement of the adsorption capacity of the adsorbent [31,32]. Moreover, PPC-0.8-800 had the maximum specific surface area and pore volume, which were 2151.42 m 2 g −1 and 1.109 cm 3 g −1 , respectively, as shown in   in a carbon matrix. However, N content in the porous carbon decreased with temperature (table 1), which indicates the activation process is enhanced and N is eliminated by transforming into gas at temperatures higher than 600°C. In this work, KOH took a role as an activation agent for micropore formation during carbonization processing.

FTIR analysis
The FTIR spectra of samples are shown in figures 3 and 4. The broad band appearing at 3444 cm −1 was attributed to the -OH groups stretching vibration. With the increase of ratio of KOH and temperature, the intensity of the peak became weaker and weaker, which indicated that the hydroxyl group was neutralized during the activation process. The bands appearing at 2925 cm −1 were attributed to C-H stretching vibrations in aliphatic -CH 2 . The relative strong peaks at 1625 cm −1 and 1380 cm −1 , which could be ascribed to C=N stretching and N-C=O-skeletal vibration. The weak bands at around 1023 cm −1 and 956 cm −1 might be ascribed to -C-O stretching. The existence of the -OH, N-C=O-and -C-O-bonds can be attributed to the reaction of carbon with KOH during the activation process [33]. In general, as the activation temperature and KOH ratio increase, adsorbents, surface functional groups almost were destroyed and the peak intensity of each functional group gradually weakened.

Effect of pH on Cr(VI) adsorption
The effect of pH value (1-6) on the adsorption capacity of PPC was investigated, and the results are shown in figure 5. It can be found that the adsorption capacity of PPC increases with the decrease of pH value. And precipitations of chromium occur when pH value is higher than 6, which is the reason that absorption is not studied beyond pH value of 6     However, more and more negative charge formed on the surface of the adsorbent with increasing pH value. Anyhow, the pH value of industrial wastewater and domestic wastewater is generally about 2, which means that pH value of the optimum sorption was 2.0. In addition, comparing the adsorption capacity of the PPC-0.6-800 and PPC-0.8-800, it can be seen that the adsorption capacity of the former is higher than that of the latter, although the latter has higher specific area and pore volume. The abundant N-containing groups in the PPC-0.6-800 may be the reason. Furthermore, the dominant species of Cr ion in solution was HCrO 4 − . The chromate anion interacts strongly with the positively charged ions of the PPCs. So, the reasons for the pH effect on the adsorption capacity of chromium can be expressed as follows [35]:   pseudo-first-order model pseudo-second-order model adsorbents  figure 7. It is observed that Cr(VI) uptake increases with contact time and levels out within 120 min in all samples. PPC-0.8-800 shows the maximal rate of adsorption within the 120 min due to its high surface areas and abundant N-containing groups. Moreover, its porous structure provided more active groups, such as -NH and -OH groups, exposed in the aqueous solution, and was beneficial to chromium ion diffusion and adsorption. Figure 7 revealed the effect of contact time on Cr(VI) adsorption onto PPC. The data suggest that the adsorption capacity increases rapidly in the first 60 min, then levels out. The initial rapid increase in adsorption amount may be due to many vacant sites available at the initial time interval; as a result there was an increased concentration gradient of adsorbate between solution and adsorbent [36]. Generally speaking, the initial adsorption is rapid, because the adsorption involves a surface reaction process. Then, a slower adsorption could be due to the gradual decrease of the available adsorption site [37]. Adsorption kinetics was modelled by the first-order, and the second-order, rate equation expressed as follows:

Adsorption kinetics
and

2)
where k 1 is the rate constant of adsorption (min −1 ); k 2 is the second-order constant (g mg −1 min −1 ); q t and q e are Cr(VI) adsorption amount and maximum adsorption amount. The obtained kinetic parameters are listed in  coefficients higher than 0.92 which indicates that the adsorption kinetics is not diffusion controlled but chemical adsorption. The result further confirms that the PPCs possess a high adsorption rate for Cr(IV), which is mainly attributed to its unique hierarchical porous structure combined with a high specific surface area.

Desorption of Cr(VI)
The cycling stability of the PPC-0.8-800 was investigated by performing five cycles of Cr(VI) adsorption and desorption. Considering that the adsorbed Cr(VI) on PPC-0.8-800 can be efficiently removed in acidic condition by reinforcing the electrostatic effect between PPC-0.8-800 and Cr(VI), the regeneration experiment was conducted in 5% NaOH aqueous solution, as shown in figure 8. The PPC-0.8-800 remains almost 82% of its initial adsorption capacity after five cycles, indicating its superior recyclability for the removal of Cr(VI) from wastewater. It is, however, hard to understand that the removal efficiency after the third cycle is lower than that after fourth cycle.

Comparison of the results of present study with literature ones
To justify the viability of the prepared PPC-0.8-800 as effective adsorbents for Cr(VI) removal, the adsorption capacity of PPC-0.8-800 on Cr(VI) was compared with the efficiency of other low-cost adsorbents found in the literature with similar batch studies. Table 4 shows a summary of Cr(VI) removal capacity (mg g −1 ), at optimum pH and maximum concentration of Cr(VI) used (mg l −1 ) for various adsorbents in previous studies. The adsorption capacity of PPC-0.8-800 is larger than other materials. Hence, PPC-0.8-800 can be considered to be viable adsorbent for the removal of Cr(VI) from dilute solutions.

Adsorption mechanism
To fully understand the adsorption mechanisms of chromium species on the surface of the PPCs, XPS spectra were obtained for the PPC-0.8-800 before and after chromium adsorption (denoted as PPC-0.8-800-Cr) at a solution pH = 2. It has been well demonstrated that when an adsorbate was adsorbed on an adsorbent through chemical interactions, the chemical state of the atoms involved on the surface of the adsorbent could be changed, resulting in different XPS spectra for the same atom before and after the adsorption took place [38].
The   The N 1s spectra of PPC-0.8-800 and PPC-0.8-800-Cr composites can be separated into three peaks at binding energies 398.54 eV, 400.09 eV and 400.99 eV, corresponding to the nitrogen of the species N=C, -N-C-and -NH + -, respectively. Compared with PPC-0.8-800, the relative intensity of both amino group -N=C and -NH + decreased but -N-C-increased in PPC-0.8-800-Cr,which may be attributed to N-Cr.
To clarify the adsorption mechanism of Cr(VI) on PPC, XPS changes of PPC-0.8-800 before and after adsorption are plotted in figure 9e. Two new energy bands were observed after Cr adsorption, which were at 577.5 eV and 587 eV. These two energy bands corresponded to the binding energies of Cr (2p3/2) and Cr (2p1/2), respectively [39]. The presence of Cr (2p3/2) and Cr (2p1/2) confirms the existence of both Cr(III) and Cr(VI) on the surface of PPC-0.8-800. The presence of Cr(III) on the surface of PPC-0.8-800 suggests that some adsorbed Cr(VI) is reduced to Cr(III). The reduction may be attributed to the presence of a positive -NH + group in PPC-0.8-800 [40]. The fraction of Cr(VI) and Cr(III) in adsorbents is calculated using the peak fitting method by subtracting baseline and integration. The integral area ratio of the peaks of Cr(VI) and Cr(III) was calculated to be 1 : 7, which indicates that 87.5% of the total adsorbed Cr(VI) is reduced to Cr(III) on the surface of PPC-0.8-800. The result agrees with other literature studies [41,42]. Some of the adsorbed Cr(VI) ions are reduced to Cr(III) with the oxidation of hydroxyl groups which caused the formation of additional carboxylic groups [43]. Because of this reduction and chelation, the most abundant form of chromium was Cr(III) on the surface of sorbents. When the acidity of solution decreases, the surface amine groups are hard to charge. In this case, the electrostatic attraction between the adsorbent and Cr(VI) ions decreases, and the removal efficiency falls.
In addition, the specific surface area and pore-size distribution of the materials have a great influence on the adsorption of chromium ions. PPC-0.8-800 with large specific surface area and abundant microporous structure has a large number of adsorption sites, which increase the adsorption of chromium ions. In summary, PPC-0.8-800 has great performance to remove chromium under the combination of chemisorption and physical adsorption.

Conclusion
In summary, the nitrogen-doped porous carbon for chromium removal in aqueous solution based on the PAN was prepared by carbonization and KOH activation. Activated carbons deriving from PANfibre are high in pores centred at the supermicropore region. Porous structure contributed to the high specific surface area, and made more nitrogen groups contact with ions in aqueous solution. Meanwhile, the N-containing groups displayed a great role in Cr(VI) removal. The maximum removal capacity for chromium of 374.90 mg g −1 was found at pH = 2.0 with PPC-0.8-800. Compared with other low-cost adsorbents [44], the removal ability and cycling stability were higher.
Data accessibility. Estimation of parameters have been stated throughout the body of the paper and included in the reference section. The graphs were produced using the Origin software that is available from http://www.originlab. com/.
Authors' contributions. B.F. carried out the molecular laboratory work, participated in data analysis, carried out sequence alignments, participated in the design of the study and drafted the manuscript; S.Q. carried out the statistical analyses; B.F. collected field data; W.S. and L.S. conceived of the study, designed the study, coordinated the study and helped draft the manuscript. All authors gave final approval for publication.