Self-flocculated powdered activated carbon with different oxidation methods and their influence on adsorption behavior

Journal of Hazardous Materials 304 (2016) 222–232

Contents lists available at ScienceDirect

Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat

Self-flocculated powdered activated carbon with different oxidation methods and their influence on adsorption behavior Zailin Gong a , Shujin Li b , Jun Ma a,∗ , Xiangdong Zhang b,∗∗ a b

State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology, Heilongjiang Province, Harbin 150001, China College of Chemistry, Liaoning University, Liaoning Province, Shenyang 110000, China

h i g h l i g h t s

g r a p h i c a l

• Powdered activated carbon (PAC) is

Powdered activated carbon (PAC) has been treated with two different oxidation methods and grafted with poly (N-isopropylacrylamide) (PNIPAM). The new oxidation method of thermal treatment followed by acidificating enables the carboxyl-rich PAC with large surface area. The PNIPAM graft PAC synthesized from the carboxyl-rich PAC will have good adsorption behavior of bisphenol A and self-flocculation effect with rapid response to temperature.

• • • •

treated with two different oxidation methods. The oxidated PAC is grafted with poly (N-isopropylacrylamide). The grafted PAC via new oxidation method has good adsorption behavior of bisphenol A. The grafted PAC has excellent selfflocculation effect. The grafted PAC has high application in the water remediation required pre-heating.

a r t i c l e

i n f o

Article history: Received 9 August 2015 Received in revised form 15 October 2015 Accepted 19 October 2015 Available online 26 October 2015 Keywords: Powdered activated carbon Oxidation Adsorption Poly(N-isopropylacrylamide) Self-flocculation

a b s t r a c t

a b s t r a c t The commercial powdered activated carbon (PAC) has been selectively oxidized by two methods. The two oxidized methods are wet oxidation with ammonium persulfate and thermal treatment after acidification with hydrochloride acid, respectively. The two oxidized PAC were then functionalized with thermoresponsive poly (N-isopropylacrylamide) (PNIPAM) in aqueous solution at ambient temperature. Comparing the two oxidized PAC products and their grafted derivatives, the oxidized PAC modified with thermal treatment after acidification shows larger surface area of 1184 m2 /g and better adsorption of bisphenol A. Its derivative also exhibits relatively large surface area and adsorption capacity after grafted with PNIPAM. The maximum surface adsorption capacity simulated under Langmuir Models reached 156 mg/g. In addition, the grafted PAC products show self-flocculation behaviors with rapid response to temperature because of the thermal phase transition and entanglement behaviors of PNIPAM. The present study provides a new way to obtain carboxyl-rich activated carbon with large surface area and better adsorption capacity. The retrievable grafted PAC with good self-flocculation effect responsive to temperature will have high potential application in water remediation which requires pre-heating and emergency water treatment in the wild. © 2015 Published by Elsevier B.V.

1. Introduction ∗ Corresponding author. Fax: +86 415 86283010. ∗∗ Corresponding author. E-mail addresses: [email protected] (J. Ma), [email protected] (X. Zhang). http://dx.doi.org/10.1016/j.jhazmat.2015.10.039 0304-3894/© 2015 Published by Elsevier B.V.

With the development of separation technology, there has been increasing number of water treatment methods required

Z. Gong et al. / Journal of Hazardous Materials 304 (2016) 222–232

pre-heating, such as distillation, membrane distillation (MD) etc. Especially, MD is well known as the emerging thermally-driven separation process, which is based on the vapor transport across the hydrophobic microporous membrane driven by the vapor pressure gradient across the membrane and poses a lot of promise in desalination, water and wastewater treatment. [1] However, the loose pore structure of the microporous membrane results in the inefficient removing of organic pollutants from water during membrane distillation process, which also occurs during distillation process. Moreover, in the cases of drinking in the wild or purifying the water in swimming pool, a relatively high resultant temperature is needed, which also presents many difficulties in the removing of organic pollutants [2]. Activated carbon adsorption, which has a low capital cost and does not produce much toxic intermediates [3], seems to be a promising adsorbent for organic pollutants removal. Comparing the kinds of carbon adsorbent, powdered activated carbon (PAC) is much better than that of granular activated carbon (GAC) because of the larger surface areas [4–6]. However, the small particle size makes the retrieval of PAC from water difficult, and the residual PAC in the water will result in the secondary pollution. To recycle PAC, there are two obstacles need to be tackled, including the retrieve of PAC from the water and the regeneration of PAC. There have been a few studies on the regeneration of PAC [7], however, no research is on making the PAC retrievable. As the activated carbon could be functionalized [8], the retrieval limitation could be solved by modifying PAC, such as the PAC with reversible flocculation function upon the changing of temperature. The reversible flocculation effect in response to temperature can be achieved by grafting thermoresponsive polymer on the PAC surface. As for surface modification, the carboxyl groups play a key role in providing anchoring sites for further surface modification by grafting or surface reaction [9,10]. There are many methods to create carboxyl group on carbon surface, and most of them are the wet oxidation in a solution of oxidant, such as H2 O2 , HNO3 , (NH4 )2 S2 O8 , NaOCl, KClO3 , KIO4 , KMnO4 , AgNO3 etc. at relatively lower temperature (20–120 ◦ C) [11–14]. However, of all these wet oxidation reactions, the decrease of the surface areas and pore volumes of PAC is inevitable which will lower the adsorption performance for hydrophobic organics. Previous report [15] shows that carboxylrich carbon sphere can be obtained via a simple approach by the thermal treatment in air. In this report, we prepare the carboxylrich group functioned PAC by the thermal treatment based on the acidificating PAC firstly. Surface-initiated atom transfer radical polymerization (SiATRP) is one of the most frequently used methods to covalently connect polymer chains to the solid surface [16–19], which can be employed in fabricating thermoresponsive PAC. We found that the modification of PAC surface by Si-ATRP has rarely been reported [9,19]. Furthermore, the carboxyl functionalized PAC in these articles is created by wet oxidation method which is known for the destruction of the pore structure of the PAC. Additionally, the influences of the oxidation and grafting reaction on surface areas and adsorption capacity of the PAC are also not discussed. As one of the most widely studied thermoresponsive polymers, Poly (N-isopropylacrylamide) (PNIPAM) and modified PNIPAM are frequently used based on their potential applications in biological medicine field [20–22]. In recent years, using PNIPAM as flocculants has also been reported. [23] Compared with the conventional polymeric flocculants, the PNIPAM capped particles show hydrophobicity with the temperature higher than the LCST of 32 ◦ C and thus will deposit in water; however, the hydrophilicity transition from hydrophobicity with the temperature lower than LCST provides the modified particles re-dispersing ability [24–26]. In this report, the PAC of thermoresponsive self-flocculation ability would be prepared by grafting PNIPAM on PAC surface via Si-ATRP.

223

Among all the organic pollutants, Bisphenol A, one kind of endocrine-disrupting chemicals (EDCs) [27,28], is widely used as an intermediate in the production of polycarbonates, epoxy resins, and other plastics. Bisphenol A has been detected in all kinds of environmental water. Moreover, many in vitro and in vivo assays have confirmed that bisphenol A increased the incidence of infertility, genital tract abnormalities and breast cancer [29,30]. Accordingly, there is an urgent need to develop effective technology to remove bisphenol A from aquatic environment. In this paper, carboxyl functionalized powdered activated carbon (PAC) were obtained by two oxidation methods. After that, thermoresponsive PNIPAM graft PAC (PAC-PNIPAM) was synthesized in aqueous solution under ambient temperature by Si-ATRP. The surface and textual properties of the obtained carboxyl functionalized PAC and PNIPAM graft PAC were investigated. Furthermore, the influence of oxidation methods and the graft process was examined by studying the adsorption ability of bisphenol A on these obtained PAC. Additionally, the thermoresponsive selfflocculation effect of PAC-PNIPAM was investigated to estimate the retrievable performance. 2. Experimental 2.1. Materials Bisphenol A (99+% purity) [C15 H16 O2 ] was purchased from Aladdin Chemical Company. The largest distance of bisphenol A molecule is 0.94 nm, and the first deprotonation of bisphenol A started at around pH 8.0 and the second one at around pH 9.0. In view of the high value of logKow (3.32), the stock solution of bisphenol A was prepared with methanol which was adsorbed slightly onto activated carbons in water [31]. Then the working solution of bisphenol A was obtained by diluting the stock solution with pure water. Monomer N-isopropylacrylamide (NIPAM, 98%, Aladdin) was used directly without further purification. In this paper, Copper (I) bromide (CuBr), 2-bromoisobutyryl bromide (BMPB, 98%), and N,N,N ,N ,N -pentamethyldiethylenetriamine (PMDETA) were used as catalyst, initiator, and ligand, respectively, and all of them were obtained from Sigma–Aldrich and used without further purification. The 4-dimethylamiopryidine (DMAP), triethylamine (TEA) and 3-aminopropyltriethoxysilane (APTES) were supplied by Aladdin. Powdered activated carbon (PAC) was purchased from Aladdin and dried at 120 ◦ C in vacuum for at least 12 h (h) before use. Ammonium persulfate and other reagents, organic solvents were purchased from Sinopharm. 2.2. Sample preparation Carboxyl functioned PAC (PAC-COOH) was prepared by two methods, and Fig. 1 presents the preparation procedures of PACCOOH and PNIPAM graft PAC. 2.2.1. Preparation of PAC-COOH via thermal treatment The acidification of PAC was needed for low-temperature heat treatment in air, which was performed as follows. 3.0 g of PAC was added to a round-bottomed flask containing 50 mL of 12.0 mol L−1 hydrochloride acid and dispersed by ultrasonic for 30 min (min). Subsequently, the mixture was transferred into an oil bath and heated at 60 ◦ C for 3 h. The resulting mixture was filtered and thoroughly washed with deionized water until neutral. The resultant product was dried under vacuum at 110 ◦ C and denoted as PACOH. PAC-COOH was obtained by the thermal treatment of PAC-OH. The PAC-OH was heated to the desired temperature of 300 ◦ C at a rate of 5 ◦ C min−1 under air atmosphere and kept for 5 h. The resultant product was washed with deionized water and subsequently with ethanol until neutral. Then, the product was dried at

224

Z. Gong et al. / Journal of Hazardous Materials 304 (2016) 222–232

Fig. 1. Preparation of PAC-PNIPAM(1) and PAC-PNIPAM(2).

50 ◦ C under vacuum for 12 h, and the final product was denoted as PAC-COOH(1). 2.2.2. Preparation of PAC-COOH via wet oxidation The PAC-COOH sample was also prepared using the wet oxidation method in comparison with the PAC-COOH(1). The oxidation condition is acquired according to the previous report of Liu [12]. The oxidized PAC was prepared by adding 3.0 g of PAC into a round-bottomed flask containing 50 mL of ammonium persulfate (2.0 mol L−1 ) in 1.0 mol L−1 sulfuric acid. Then, the mixture was dispersed in an ultrasonic bath for 30 min and heated in an oil bath at 60 ◦ C for 3 h. The final oxidized PAC was obtained by filtering and thoroughly washing the above mixture until neutral. Finally, the PAC-COOH(2) was produced by drying the oxidized PAC under vacuum at 110 ◦ C. 2.2.3. Synthesis of amino-functionalized PAC The amino-functionalized PAC was prepared as follows. First, a homogeneous suspension was produced by mixing one of the kinds of PAC-COOH and 0.1 mL triethanolamine in 50 mL of ethanol. The above suspension was transferred into a 150 mL flask equipped with a magnetic stir bar followed by heating to 60 ◦ C. Then, 60 ␮L of ammonium hydroxide and 1.5 mL of deionized water was injected into the flask. Subsequently, 0.94 mL of APTES (4.0 mmol) was added dropwise. The resultant mixture was stirred at 50 ◦ C for 12 h and cooled to room temperature. The sediment was obtained by centrifuging the crude product at 8000 rpm and the supernatant was discarded. After that the sediment was re-dispersed in ethanol and centrifuged once more. The above purification cycle was repeated thrice to remove the excess APTES thoroughly. The resultant amino-functionalized PAC were denoted as PAC-NH2 (1) (synthesized by PAC-COOH(1)) and PAC-NH2 (2) (synthesized by

PAC-COOH(2)) and were finally redispersed in 50 mL of anhydrous tetrahydrofuran (THF) for storage.

2.2.4. Synthesis of 2-bromoisobutyrate-functionalized PAC The above THF suspension of amino-functionalized PAC (50 mL) (PAC-NH2 (1) or PAC-NH2 (2)) was mixed with DMAP (0.056 g, 0.47 mmol) and TEA (1.4 mL, 10 mmol) in a flask. After cooling to 0 ◦ C, 2-bromoisobutyryl bromide (BMPB) (0.68 mL, 5.0 mmol) was added dropwise. The solution was stirred at 0 ◦ C for 60 min and then at room temperature for 24 h. The resultant mixture was purified and isolated following the similar procedures as those described for the synthesis of amino-functionalized PAC. Finally, APTES–BMPB immobilized PAC were obtained by drying the final product under vacuum at 80 ◦ C for 24 h and denoted as PAC-Br(1) (synthesized by PAC-NH2 (1)) and PAC-Br(2) (synthesized by PAC-NH2 (2)). 2.2.5. Preparation of surface graft PAC-PNIPAM The surface graft PAC-PNIPAM was prepared via surfaceinitiated ATRP technique. Firstly, PAC-Br(1) or PAC-Br(2) (400 mg), PMDETA (78.75 mL, 0.5 mmol), NIPAM (0.8 g, 7.1 mmol) and water (8 mL) were added to a Schlenk flask equipped with a magnetic stir bar. After dispersing by sonification, the above mixture was degassed by three freeze–pump–thaw cycles. In the frozen state, CuBr(50 mg, 0.5 mmol) was added under the protection of N2 flow. The flask was then subjected to two additional freeze–pump–thaw cycles. Stirring was started immediately after the thawing. After polymerization for 6 h, the emulsion was centrifuged and subsequently dried under vacuum at 80 ◦ C for 24 h, and then the PNIPAM immobilized PAC was prepared. The resultant products were denoted as PAC-PNIPAM(1) (synthesized from PAC-Br(1)) and PAC-PNIPAM(2) (synthesized from PAC-Br(2)).

Z. Gong et al. / Journal of Hazardous Materials 304 (2016) 222–232

2.3. Characterization The existence of PNIPAM on the PAC surface was verified by FTIR technique which was conducted on an AVATER-360B FTIR spectrometer (Nicolet Co., USA) in the range from 400 to 4000 cm−1 . [7] The graft degree of NIPAM was quantitatively characterized with TGA which was performed using a Pyris 6 thermogravimetric analysis (PerkinElmer, USA) at a heating rate of 10 ◦ C min−1 from room temperature to 800 ◦ C. The surface morphologies of the composites were examined using a JEOL-6701F scanning electron microscope. The surface carboxyl functional group was determined by Boehm titrations [32]. The absorbance of bisphenol A was detected by UV–vis spectrophotometer (Beijing Purkinje General Instrument Co., CN). The dispersion properties of the composites were observed by optical photographs and the turbidity was detected by a WGZ2000 turbidity meter (Beijing Warwick Industrial Science and technology Co., CN). Nitrogen adsorption–desorption isotherms were obtained with a Micromeritics ASAP2010 instrument (USA) at −196 ◦ C. 2.4. Adsorption experiments The adsorption ability of samples was determined by detecting the adsorbance of bisphenol A (BPA) on the powder surface. The experiments were conducted in an orbital shaker at about 140 rpm. The maximum absorbance of BPA at 276.5 nm was employed as the standard. The experimental temperature was 25 ± 0.1 ◦ C with the pH of 7.0 ± 0.1. In the adsorption experiments, 5 mg of sample was placed into a 200 mL flask containing 100 mL of BPA solution with an initial BPA concentration of 3 mg/L, 5 mg/L, 10 mg/L, 15 mg/L, 20 mg/L and 25 mg/L, respectively. Samples were taken at specific times up to 600 min for PAC and PAC-COOH(2), 720 min for PACCOOH(1), 3600 min for PAC-PNIPAM(1) and PAC-PNIPAM(2). The adsorption of BPA was calculated according to Eq. (1): qe =

(C0 − Ce ) V M

(1)

where qe (mg/g) is the equilibrium adsorption amount, C0 (mg/L) and Ce (mg/L) are the initial and equilibrium concentrations of BPA in the solution, respectively. V (L) is the solution volume, and M (g) is the mass of the adsorbent. 3. Results and discussion 3.1. Sample characterization The FT-IR spectra of the modified activated carbon samples prepared with different oxidation methods is shown in Fig. 2 and Fig. 3. Each FT-IR band of the modified activated carbon samples is supplied in the Supplemental materials (Table 1). Fig. 2 shows the FTIR spectra of PAC-COOH (a), PAC-NH2 (b), PAC-Br (c) and PAC-PNIPAM (d) prepared by the method of thermal treatment. As shown in Fig. 2, the characteristic peak at 3420 cm−1 is observed in Fig. 2(a), which is attributed to the stretching vibration of OH group and H2 O. However, the characteristic peak shift to about 3415 cm−1 at Fig. 2(b–d) is due to the stretching vibration of N H group. And, the peaks observed at about 2926 cm−1 in Fig. 2(b–d) are due to the introduction of CH3 group. The oxidation process can be certified by the observation of characteristic peak at 1731 cm−1 , which is from the stretching vibration of C O in COOH group in (a). The attachment of 3-aminopropyltriethoxysilane can be certified by the observation of characteristic peak at 1107 cm−1 , which is from the stretching vibration of C Si O in (b). The stretching vibration of C O at 1627 cm−1 and bending vibration of N H at 1544 cm−1 in amide group is from the 2-bromoisobutyryl bromide in (c). For PNI-

225

PAM graft PAC (d), the much more distinct peaks at 1637 cm−1 and 1544 cm−1 for the amide group verify the polymerization of NIPAM. Fig. 3 shows the FTIR spectra of PAC-COOH (a), PAC-Br (b) and PAC-PNIPAM (c) prepared by the normal wet oxidation using ammonium persulfate. In comparison with Fig. 2, the characteristic stretching peak of C O from carboxyl at 1718 cm−1 becomes stronger in Fig. 3(a–c). The similar characteristic peaks corresponding to that in Fig. 2 indicate the successful preparation of samples. 3.2. Grafting degrees determination The thermal decomposition behavior of the initial PAC and the prepared samples by the method of thermal treatment followed by acidificating (samples from (b) to (d)) and normal wet oxidation using ammonium persulfate (samples from (b ) to (d )) are estimated by the TGA detection, which is present in Fig. 4. Fig. 4(a) shows that the retention for the initial PAC is about 80%. However, the retention for PAC-COOH(1) and PAC-COOH(2) are 72% and 61%, respectively. Results indicate that the thermal stability of the PAC has been reduced after oxidation. Compared with the oxidation by ammonium persulfate, thermal treatment followed by acidificating provides the PAC-COOH much better thermal stability because of the relatively less destruction of pore structure. TGA analysis could also be used to quantitatively obtain the grafting degree of PNIPAM from the activated carbon [17]. For the PAC-PNIPAM(1) and PAC-PNIPAM(2), a sharp decrease of weight retention is observed at 400–500 ◦ C, which is due to the decomposition of the grafted PNIPAM above the thermal decomposition temperature. The weight retention is 53% and 45% at 800 ◦ C, respectively. Using the weight retention for PAC-Br(1) (69 wt%) and PAC-Br(2) (60 wt%) at 800 ◦ C as a reference, the PNIPAM weight contents, relative to that of the PAC-COOH(1) (72 wt%) and PAC-COOH(2) (61 wt%) cores, are calculated as 24% and 26%. Results mean that the grafting degrees of PAC-PNIPAM(1) and PAC-PNIPAM(2) are similar. 3.3. Surface morphology determination Fig. 5 shows the SEM images of the modified activated carbon samples. Apparent changes can not be observed on the surface morphology of PAC-COOH(1) in comparison with PAC. However, loose surface structure can be observed on PAC-COOH(2), especially with a larger amplifications of 370,000 times. Results illustrate that the original pore structure has been destroyed after oxidation with ammonium persulfate. For the PAC-PNIPAM(1) and PAC-PNIPAM(2), the surface morphology of the grafted PAC has been changed significantly compared with PAC-COOH(1) and PACCOOH(2). Especially for PAC-PNIPAM(2), the initial loose surface of PAC-COOH(2) has been grafted with thin layer of polymer and the contour of the polymer is very obvious. 3.4. Boehm titration and nitrogen adsorption-desorption experiment The textural and the surface characteristics of the samples are listed in Table 1. Compared with the initial PAC, the amounts of carboxyl group of PAC-COOH(1) and PAC-COOH(2) have been increased from 1.4 mmol/g to 2.9 mmol/g and 3.9 mmol/g, illustrating both the two oxidation methods can increase the amounts of carboxyl group on PAC. The changing on the pore structure of PAC after oxidation and graft of PNIPAM is accurately estimated by nitrogen adsorption-desorption isotherm. For the initial PAC, the Brunauer–Emmett–Teller (BET) surface area and the pore volume are 1326 m2 /g and 0.9017 cm3 /g, which decrease to 758 m2 /g and 0.5269 cm3 /g after the oxidation of ammonium persulfate, respectively. Results indicate that the pore structure of the PAC is

226

Z. Gong et al. / Journal of Hazardous Materials 304 (2016) 222–232

a

c

1384

1107

2500

1384 1456 1544 1637 1554

1627 1620

3000

1610

1710

3500

2926

3410

4000

2926 2926

3415 3415

d

1731

3420

Intensity（a.u.）

b

2000

1500

1000

500

-1

Wave number(cm ) Fig. 2. Fourier transform infrared (FTIR) spectra of PAC-COOH(1) (a), PAC-NH2 (1) (b), PAC-Br(1) (c), PAC–PNIPAM(1) (d).

(b)

1107

1459 1544

1385

3000

1381

1624 1712

2958

3408

1381 1458 1544 1624

1718

3410

(c)

4000

1610

1718

3420

Intensity（a.u.）

(a)

2000

1000 -1

Wave number(cm ) Fig. 3. FTIR spectra of PAC-COOH(2) (a), PAC-Br(2) (b) and PAC-PNIPAM(2) (c).

Table 1 Textural and surface carboxyl group characteristics of the modified activated carbon samples. Sample

BET surface area(m2 /g)

BJH pore diameter(nm)

Total pore volume(cm3 /g)

Carboxyl group (mmol/g)

PAC PAC-COOH(1) PAC-COOH(2) PAC-PNIPAM(1) PAC-PNIPAM(2)

1326 1184 758 458 20

3.71 3.82 3.82 3.71 3.82

0.90 0.82 0.53 0.34 0.03

1.40 2.90 3.90 0.08 0.32

Z. Gong et al. / Journal of Hazardous Materials 304 (2016) 222–232

227

105 100 95

Weight retention(%)

90 85

a)80%

80 75

b) 72% c) 69%

70 65

b')61% c')60%

60 55

d)53%

50

d')45%

45 40 200

400

600

800

1000

o

Temperature( C) Fig. 4. TGA of PAC (a), PAC-COOH(1) (b), PAC-Br(1) (c), PAC-PNIPAM(1) (d) and PAC-COOH(2) (b ), PAC-Br(2) (c ), PAC-PNIPAM(2) (d ).

destroyed by the ammonium persulfate, resulting in the decrease of the surface area and pore volume. However, the decrease of the surface area and pore volume is limited for PAC after thermal treatment followed by acidificating, indicating the new oxidation method has little influence on the initial pore structure. After the grafting of the PNIPAM, the surface area and pore volume have both decreased. However, the surface area for PAC-PNIPAM(1) is much larger than that of PAC-PNIPAM(2) with similar grafting degree. These results indicate that the oxidation is fatal to the grafting of PAC.

3.5. Adsorption of bisphenol A The adsorbance of BPA is used to estimate the adsorption performance of the two modified PAC samples. The adsorption values of BPA on the various activated carbons are shown in Fig. 6 and summarized in Table 2. For the non-grafted PAC, PAC and PAC-COOH(1) show the similar adsorption ability of BPA, while the PAC-COOH(2) shows the very different adsorption ability. As shown in Fig. 6 and

Table 2, the equilibrium capacity of bisphenol A onto PAC is the highest, little decreased on PAC-COOH(1) and the lowest onto PACCOOH(2). The different adsorption abilities may be attributed to the oxidation modification. Firstly, the oxidation treatment decreases the surface areas of the original activated carbons, which reduces the adsorptive sites of bisphenol A. Secondly, a large amount of carboxyl groups are introduced on the surface of activated carbons, which will decrease the adsorption rate of BPA because the BPA is negatively charged in water. However, the adsorption ability for PNIPAM graft PAC is different from that of non-grafted PAC. And, PAC-PNIPAM(1) shows a much better equilibrium capacity of bisphenol A than PAC-PNIPAM(2) as shown in Table 2. As the two grafted PAC-PNIPAM have the similar grafting degree, the difference in the equilibrium capacity can also be attributed to the different oxidation modification method. The PNIPAM influences the adsorption of BPA on PAC from two aspects. On one hand, the grafted PNIPAM will decrease the surface areas of the activated carbon which reduces the adsorptive sites of bisphenol A. On the other hand, the grafted PNIPAM could form

Fig. 5. SEM images of PAC (a and a ), PAC-COOH(1) (b and b ), PAC-COOH(2) (c and c ), PAC-PNIPAM (d and d ) and PAC-PNIPAM(2) (e and e ) with amplifications of 3300 times (a–e) and 37,000 times (a –e ).

228

Z. Gong et al. / Journal of Hazardous Materials 304 (2016) 222–232

Fig. 6. Adsorption rate of BPA at the concentration of 3 mg/L (a), 5 mg/L (b), 10 mg/L (c), 15 mg/L (d), 20 mg/L (e), 25 mg/L (f) on PAC, PAC-COOH(1), PAC-COOH(2), PACPNIPAM(1) and PAC-PNIPAM(2).

hydrogen bonds with the BPA, which will promote the adsorption performance. As a result, the PAC-PNIPAM with relatively small surface areas can also have the good adsorption performance of BPA. The adsorption of bisphenol A on PAC-PNIPAM(1) would be carried out in two steps. In the first initial 200 min, the BPA is adsorbed by the superficial pores. After that, the internal pores and the grafted PNIPAM are assumed playing the major role in the adsorption process.

To better elucidate the mechanism of BPA on the modified PAC samples, the commonly used Langmuir (Eq. (2)) and Freundlich models (Eq. (3)) are adopted to simulate the experimental data. Fig. 7 shows the linearized form and the nonlinear form of Freundlich and Langmuir isotherms.

Non − Linear form : qe =

qm KL Ce 1 1 1 Linear form: = + 1 + K L Ce qe (KL qm ) Ce qm (2)

Z. Gong et al. / Journal of Hazardous Materials 304 (2016) 222–232

229

Table 2 Adsorption capacity of PAC, PAC-COOH(1), PAC-COOH(2), PAC-PNIPAM(1) and PAC-PNIPAM(2) for BPA at different concentrations. Concentration, C(mg/L)

Adsorption capacity Qe (mg/g)

Adsorption Rate W (%)

0.51 0.86 3.18 5.61 8.26 11.95

49.88 82.90 136.80 187.80 234.76 261.00

83.1 82.9 68.1 62.6 58.7 52.2

3 5 10 15 20 25

0.56 0.95 2.97 5.97 9.60 13.46

48.84 81.02 140.60 180.54 208.00 230.75

81.4 81.0 70.3 60.2 52.0 46.2

PACPNIPAM(1)

3 5 10 15 20 25

0.54 1.20 4.12 8.63 12.80 17.20

49.14 76.00 118.00 131.00 148.08 156.00

81.9 76.0 58.8 43.5 37.0 31.2

PACCOOH(2)

3 5 10 15 20 25

2.32 3.93 8.45 13.25 18.07 23.02

13.54 21.34 30.98 34.99 38.60 39.60

22.6 21.3 15.5 11.7 9.7 7.9

PACPNIPAM(2)

3 5 10 15 20 25

2.84 4.81 9.72 14.72 19.62 24.60

3.12 3.90 5.60 5.70 7.60 8.00

5.2 3.9 2.8 1.9 1.9 1.6

Before adsorption (C0 )

After adsorption (Ce )

PAC

3 5 10 15 20 25

PACCOOH(1)

1/n

Non − Linear form : qe = KF Ce

Linear formlog(qe ) =

log (KF ) + 1 nlog (Ce ) (3)

In Eq. (2), Ce (mg/L) and qe (mg/g) are the equilibrium adsorption concentrations in the solution and amount on the adsorbents, respectively. KL (L/mg) describes the intensity of the adsorption process, and qm (mg/g) reflects maximum adsorption capacity. In Eq. (3), the constant KF (mg/g)(L/g)n is related to the bonding energy and represents the general capacity of adsorbate adsorbed onto adsorbents for a unit equilibrium concentration. 1/n, between 0 and 1, reflects the adsorption intensity or surface heterogeneity. The parameters, obtained through the fitting of the experimental data by both models, are listed in Table 3. Different regression methods have been used to simulate the experimental data, this is because the traditional linear method doesnot take advantage to minimize the error distribution if the experimental error appears in both X and Y direction and just reports the Y data with respect to the X based on the linearized form we use [33]. Under such conditions, it would be more rational and reliable to interpret adsorption data through a process of linear and nonlinear regression. The relatively higher R2 (determination coefficient) values conform that nonlinear form is a better regression method to simulate the experimental data for the Freundlich expression. For the Langmuir expression, the R2 values of higher than 0.96 for the initial PAC and its modified derivatives also show the applicability of the nonlinear regression method. For the different models, the relatively higher R2 values of Langmuir isotherms in compare with that of Freundlich isotherms confirm the Langmuir model as a better-fitting isotherm for

the experimental data of BPA onto the modified PAC, such as PAC-COOH(1), PAC-COOH(2), PAC-PNIPAM(1). As the Langmuir equation is derived under the assumption of monolayer coverage, the high R2 values for both the linearized form and the nonlinear form imply that monolayer adsorption has occurred. From the Langmuir model, the maximum adsorption capacity of PAC, PAC-COOH(1), PAC-COOH(2), PAC-PNIPAM(1) and PAC-PNIPAM(2) can also be elucidated. The qm of PAC-PNIPAM(1) is higher than 156 mg/g for both regression methods, which is larger than the half of the initial PAC, meaning that the grafted PAC with a relative large adsorption capacity has been obtained.

3.6. Self-flocculation of the grafted PAC After grafting with PNIPAM, the PAC will have self-flocculation ability in response to the temperature because of the unique property of PNIPAM and the entanglement between the polymer chains and the activated carbon particles. Heating the hybrid PAC-PNIPAM above the LCST of PNIPAM, the grafted PNIPAM changed from hydrophilic to hydrophobic because of the phase separation, and then bigger hybrids are formed due to the bridging adsorption caused by the entanglement of the grafted polymer chains. Therefore, the hybrids agglomerate to a bigger size of flocs and settle down in the water and show the self-flocculation effect. With the temperature lower than the LCST of PNIPAM, the transition from hydrophobicity to hydrphilicity results in the re-disperse of PAC-PNIPAM hybrids aqueous solution for the next adsorption process. And, the self-flocculation effect of PAC-PNIPAM(1) and PAC-PNIPAM(2) with temperature is illustrated by Fig. 8. Because of the self-flocculation effect, it’s very easy to drain the upper layer of the supernatant with a pipeline apparatus higher than the level of flocs. Thus, the powder activated carbon could be retrieved without

230

Z. Gong et al. / Journal of Hazardous Materials 304 (2016) 222–232

Fig. 7. Adsorption isotherms of BPA over PAC (a), PAC-COOH(1) (b), PAC-COOH(2) (c) and PAC-PNIPAM(1) (d) fitted by different regression methods of the Langmuir equation (A, A ) and Freundlich equation (B, B ).

Table 3 Parameters of adsorption isotherms modeled by the Langmuir equation and Freundlich equation. Adsorbents

Fitting method

Langmuir

Freundlich

qm (mg/g)

KL (L/mg)

R

KF (mg/g)(L/g)n

n

R2

2

PAC

Linear Non-linear

285.71 319.44

0.45 0.29

0.981 0.965

78.3249 82.7758

1.988 2.115

0.979 0.981

PACCOOH(1)

Linear Non-linear

264.55 264.86

0.42 0.40

0.991 0.991

74.8110 82.9656

2.143 2.459

0.958 0.973

PACCOOH(2)

Linear Non-linear

54.05 49.86

0.15 0.18

0.988 0.992

10.4472 12.0650

2.194 2.517

0.933 0.938

PACPNIPAM(1)

Linear Non-linear

156.25 161.67

0.84 0.72

0.994 0.982

66.4048 70.6682

3.104 3.458

0.957 0.963

Table 4 Removal percentages of PAC-PNIPAM(1) and PAC-PNIPAM(2) in aqueous solutions changing with the phase transition. Initial concentration

10 g/L 8 g/L 4 g/L 2 g/L 1 g/L 0.5 g/L 0.2 g/L 0.1 g/L 0.05 g/L 0.01 g/L

PAC-PNIPAM(1)

PAC-PNIPAM(2)

Turbidity

Concentration (mg/L)

Removal percentage

Turbidity

Concentration (mg/L)

Removal percentage

80 119 51 65 45 48 14 2 1 1

55.4 81.9 35.7 45.2 31.6 33.7 10.5 2.4 1.7 1.7

99.5% 99.0% 99.1% 97.7% 96.8% 93.3% 94.7% 97.6% 96.6% 83.0%

115 45 40 60 38 47 13 2 1 1

79.2 31.6 28.2 41.8 26.9 33.0 9.9 2.4 1.7 1.7

99.2% 99.6% 99.3% 97.9% 97.3% 93.4% 95.1% 97.6% 96.6% 83.0%

Z. Gong et al. / Journal of Hazardous Materials 304 (2016) 222–232

231

Fig. 8. Illustrative diagram of the self-flocculation effect with temperature.

Fig. 9. Optical photographs of PAC-PNIPAM(1) (a) and PAC-PNIPAM(2) (A) in aqueous solution at 25 ◦ C for 30 min, (b–g) and (B–G) are PAC-PNIPAM(1) and PAC-PNIPAM(2) in aqueous solution at 40 ◦ C for 3 s, 5 s, 10 s, 20 s, 40 s, 60 s, respectively, (h) and (H) are the heated PAC-PNIPAM(1) and PAC-PNIPAM(2) back to 25 ◦ C, (I) and (I ) are the initial PAC in aqueous solution at 25 ◦ C and 40 ◦ C for 30 min.

the solid-liquid separating equipment and the separation process could be optimized. The dispersion photographs of PAC-PNIPAM(1), PAC-PNIPAM(2) and the initial PAC in aqueous solution varied with temperatures are shown in Fig. 9. The initial concentration of the samples is 5 g L−1 , and the pH value is 7.0 ± 0.1. The initial PAC are well dispersed in aqueous solution at the both temperature of 25 ◦ C and 40 ◦ C for 30 min, and no flocculation can be observed during the process, as shown in Fig. 9(I) and (I ). In the meantime, the PAC-PNIPAM(1) and PAC-PNIPAM(2) can also be well dispersed in aqueous phase at ambient temperature for 30 min as shown in Fig. 9(a) and (A), indicating the hydrophilic character of the hybrids. However, selfflocculation is observed by heating the temperature to 40 ◦ C as shown in Fig. 9(b) and (B), which can be illustrated as the result of the formation of bigger hydrophobic hybrids agglomerate because of the bridging adsorption. Apparently, quick flocculation effect is observed on PAC-PNIPAM hybrids above the LCST of PNIPAM, as shown in Fig. 9(c)–(g) and (C)–(G). However, decreasing the temperature back to below 32 ◦ C and stirring again, well dispersed

PAC-PNIPAM(1) and PAC–PNIPAM(2) are obtained again, as shown in Fig. 9(h) and (H). The unique self-flocculation effect makes grafted PAC have high potential application in the adsorption process for the water remediation which needs pre-heating. As for the pre-treatment of the membrane distillation, the grafted PAC with relatively large adsorption capacity could be used to adsorb the organic pollutants before the membrane distillation. After heating the solution to above 32 ◦ C, the grafted PAC will settle down in the water and the heated supernatant will be used for membrane distillation. The precipitated PAC-PNIPAM can be dispersed again in the next injected water below 32 ◦ C for the next adsorption. Moreover, the thermoresponsive PAC could also be used in the emergency water treatment in the wild. The grafted PAC would adsorb the hazardous substance in the water and then settle down after heating the solution. Eventually, the heated supernatant will be used for drinking. To accurately estimate the self-flocculation performance of the PAC-PNIPAM(1) and PAC-PNIPAM(2), turbidity experiment is conducted to calculate the residual concentrations of the grafted PAC

232

Z. Gong et al. / Journal of Hazardous Materials 304 (2016) 222–232

after flocculation. The linear plot of the turbidity with the concentration of the grafted PAC is shown in the Supplemental materials (Fig. 1). The relational expression is y = 1.47114x−1.50273 (y is the turbidity, x is the concentration (mg/L)). The residual concentration of the PAC-PNIPAM(1) and PAC-PNIPAM(2) in aqueous solution after flocculation can be calculated with this expression. In the experiment, the initial concentration of PAC-PNIPAM(1) and PAC-PNIPAM(2) changes from 0.01 g/L to 10 g/L. The turbidity test was conducted using the supernatant after heating the aqueous solution to 40 ◦ C for 40 s, and the residual concentration was calculated. The residual concentration and removal percentage of the PAC-PNIPAM(1) and PAC-PNIPAM(2) changing with the phase transition is shown in Table 4. As shown in Table 4, the remaining concentration of the grafted PAC after flocculation is less than 82 mg/L with the initial concentration from 0.01 g/L to 8 g/L. Results certify that both PAC-PNIPAM(1) and PAC-PNIPAM(2) have perfect self-flocculation effect. In addition, the PAC-PNIPAM will have high potential applications in the water treatment. 4. Conclusions Thermal treatment followed by acidificating with hyfrochloric acid was used to obtain carboxyl-rich activated carbon with large surface area and good adsorption behavior. The oxidized PAC was grafted with thermoresponsive PNIPAM to obtain new grafted PAC. Compared with the grafted PAC synthesized by traditional oxidized PAC (PAC-PNIPAM(2)), the new grafted PAC (PAC-PNIPAM(1)) have the thermoresponsive self-flocculation effect with rapid response to temperature and good adsorption behavior of bisphenol A. These effects for the new grafted PAC would make the system feasible in the water purification areas. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jhazmat.2015.10. 039. References [1] X. Li, C. Wang, Y. Yang, X. Wang, M. Zhu, B.S. Hsiao, Dual-biomimetic superhydrophobic electrospun polystyrene nanofibrous membranes for membrane distillation, ACS Appl. Mater. Interfaces 6 (2014) 2423–2430. [2] J. Sutton, C. League, T.L. Sellnow, D.D. Sellnow, Terse messaging and public health in the midst of natural disasters: the case of the boulder floods, Health commun. 30 (2015) 135–143. [3] J.P. Chen, S.N. Wu, K.H. Chong, Surface modification of a granular activated carbon by citric acid for enhancement of copper adsorption, Carbon 41 (2003) 1979–1986. [4] W.T. Tsai, C.W. Lai, T.Y. Su, Adsorption of bisphenol—A from aqueous solution onto minerals and carbon adsorbents, J. Hazard. Mater. B 134 (2006) 169–175. [5] K.J. Choi, S.G. Kim, C.W. Kim, Effects of activated carbon types and service life on removal of endocrine disrupting chemicals: amitrol, nonylphenol, and bisphenol—A, Chemosphere 58 (2005) 1535–1545. [6] I. Bautista-Toledo, M.A. Ferro-García, J. Rivera-Utrilla, A. Bisphenol, Removal from water by activated carbon effects of carbon characteristics and solution chemistry, J. Environ. Sci. Technol. 39 (2005) 6246–6250. [7] F. Salvador, N. Martin-Sanchez, R. Sanchez-Hernandez, M.J. Sanchez-Montero, C. Izquierdo, Regeneration of carbonaceous adsorbents. Part II: chemical, microbiological and vacuum regeneration, Micropor. Mesopor. Mater. 202 (2015) 277–296. [8] S.A. Carabineiro, T. Thavorn-Amornsri, M.F. Pereira, J.L. Figueiredo, Adsorption of ciprofloxacin on surface-modified carbon materials, Water Res. 45 (2011) 4583–4591.

[9] Y.H. Liu, X.L. Miao, J. Zhu, Z.B. Zhang, Z.P. Cheng, X.L. Zhu, Polymer-grafted modifi cation of activated carbon by surface-initiated AGET ATRP, Macromol. Chem. Phys. 213 (2012) 868–877. [10] J.Z. Zhu, B.L. Deng, J. Yang, D.C. Gang, Modifying activated carbon with hybrid ligands for enhancing aqueous mercury removal, Carbon 47 (2009) 2014–2025. [11] Z. Jiang, Y. Liu, X. Sun, F. Tian, F. Sun, C. Liang, et al., Activated carbons chemically modified by concentrated H2 SO4 for the adsorption of the pollutants from wastewater and the dibenzothiophene from fuel oils, Langmuir 19 (2003) 731–736. [12] B.K. Pradhan, N.K. Sandle, Effect of different oxidizing agent treatments on the surface properties of activated carbons, Carbon 37 (1999) 1323–1332. [13] P. Vinke, E.M. van der, M. Verbree, A.F. Voskamp, H. van Bekkum, Modification of the surfaces of a gasactivated carbon and a chemically activated carbon with nitric acid, hypochlorite, and ammonia, Carbon 32 (1994) 675–686. [14] N. Li, X.L. Ma, Q.F. Zha, K. Kim, Y.S. Chen, Ch. Sh. Song, Maximizing the number of oxygen-containing functional groups on activated carbon by using ammonium persulfate and improving the temperature-programmed desorption characterization of carbon surface chemistry, Carbon 49 (2011) 5002–5013. [15] Z. Chen, L.J. Ma, S.Q. Li, J.X. Genga, Q. Songa, J. Liu, C.L. Wang, H. Wang, J. Li, Z. Qin, S.J. Li, Simple approach to carboxyl-rich materials through low-temperature heat treatment of hydrothermal carbon in air, Appl. Surf. Sci. 257 (2011) 8686–8691. [16] Q. Xiao, S. He, L. Liu, X. Guo, K. Shi, Z. Du, B.L. Zhang, Coating of multiwalled carbon nanotubes with crosslinked silicon-containing polymer, Compos. Sci. Technol. 68 (2008) 321–328. [17] Y.Z. Jin, C. Gao, H.W. Kroto, T. Maekawa, Polymer-grafted carbon spheres by surface-initiated atom transfer radical polymerization, Macromol. Rapid Commun. 26 (2005) 1133–1139. [18] H. Kong, C. Gao, D.Y. Yan, Controlled functionalization of multiwalled carbon nanotubes by in situ atom transfer radical polymerization, J. Am. Chem. Soc. 126 (2004) 412–413. [19] P. Liu, T.M. Wang, Surface-initiated atom transfer radical polymerization of hydroxyethyl acrylate from activated carbon powder with homogenized surface groups, Surf. Rev. Lett. 14 (2007) 269–275. [20] S.Q. Liu, Y.W. Tong, Y.Y. Yang, Incorporation and in vitro release of doxorubicin in thermally sensitive micelles made from poly(N-isopropylacrylamide-coN,N-dimethylacrylamide)-b-poly(d,l-lactide-co-glycolide) with varying compositions, Biomaterials 26 (2005) 5064–5074. [21] S.R. Sershen, S.L. Westcott, N.J. Halas, J.L. West, Temperature-sensitive polymer-nanoshell composites for photothermally modulated drug delivery, J. Biomed. Mater. Res. 51 (2000) 293–298. [22] R. Yoshida, K. Uchida, Y. Kaneko, K. Sakai, A. Kikuchi, Y. Sakurai, T. Okano, Comb-typegrafted hydrogels with rapid deswelling response to temperature changes, Nature 374 (1995) 240–242. [23] S. Sakohara, R. Hinago, H. Ueda, Compaction of TiO2 suspension by using dual ionic thermosensitive polymers, Sep. Purif. Technol. 63 (2008) 319–323. [24] S. Sakohara, K. Nishikawa, Compaction of TiO2 suspension utilizing hydrophilic/hydrophobic transition of cationic thermosensitive polymers, J. Colloid Interface Sci. 278 (2004) 304–309. [25] P.W. Zhu, D.H. Nappe, Studies of aggregation kinetics of polystyrene latex sterically stabilized by poly(N-isopropylacrylamide), Phys. Rev. E Stat Phys. Plasmas, Fluids, Relat. Interdiscip. Top. 50 (1994) 1360–1366. [26] E. Burdukova, H. Ishida, L.N.J.P. O’shea, G.V. Franks, Temperature controlled surface hydrophobicity and interaction forces induced by poly (N-isopropylacrylamide), J. Colloid Interface Sci. 342 (2010) 586–592. [27] U.S. Environmental Protection Agency (EPA). Endocrine Disruptor Screening and Testing Advisory Committee (EDSTAC) Final Report. Washington, DC: U.S. EPA, 1998. [28] R. Rudel, Predicting health effects of exposure to compounds with estrogenic activity: methodological issues, Environ. Health Perspect. 105 (1997) 655–663. [29] J. Q. Sui, Y.S. Huang, X.F. Liu, G.B. Chang, S.B. Ji, T. Deng, G. Yu Xie, Rapid removal of bisphenol A on highly ordered mesoporous carbon, J. Environ. Sci. 23 (2011) 177–182. [30] T. Suzuki, Y. Nakagawa, I. Takano, K. Yaguchi, K. Yasuda, Environmental fate of bisphenol A and its biological metabolites in river water and their xeno-estrogenic activity, Environ. Sci. Technol. 38 (2004) 2389–2396. [31] D.M. Giusti, R.A. Conway, C.T. Lawson, Activated carbon adsorption of petrochemicals, J. Water Pollut. Control Fed. 46 (1974) 947–965. [32] H.P. Boehm, Some aspects of the surface chemistry of carbon blacks and other carbons, Carbon 32 (1994) 759–769. [33] K.Y. Foo, B.H. Hameed, Insights into the modeling of adsorption isotherm systems, Chem. Eng. J. 156 (2010) 2–10.