Improvement of biodegradability for coking wastewater by selective adsorption of hydrophobic organic pollutants

Separation and Purification Technology 151 (2015) 23–30

Contents lists available at ScienceDirect

Separation and Purification Technology journal homepage: www.elsevier.com/locate/seppur

Improvement of biodegradability for coking wastewater by selective adsorption of hydrophobic organic pollutants Xubiao Yu a,b,c,⇑, Chaohai Wei b, Haizhen Wu d, Zhengming Jiang b, Ronghua Xu b a

Faculty of Architectural, Civil Engineering and Environment, Ningbo University, Ningbo 315211, PR China The Key Lab of Pollution Control and Ecosystem Restoration in Industry Clusters, Ministry of Education, Guangzhou 510006, PR China c The Belle W. Baruch Institute of Coastal Ecology and Forest Science, Clemson University, Georgetown 29440, USA d School of Bioscience and Bioengineering, South China University of Technology, Guangzhou, Guangdong 510006, PR China b

a r t i c l e

i n f o

Article history: Received 30 October 2014 Received in revised form 19 June 2015 Accepted 5 July 2015 Available online 6 July 2015 Keywords: Coking wastewater Selective adsorption Biodegradability Hydrophobic adsorption

a b s t r a c t As a typical industrial wastewater, coking wastewater is characterized as high organic load, complicated composition and strong biological inhibition. Based on the analysis between organic composition and toxicity contribution of coking wastewater, a selective removal for targeting hydrophobic contaminants was conducted by organic modified acid-vermiculites (Chlorotrimethylsilane modification, CTMS-V; and Chlorotriethysilane modification, CTES-V), powdered activated carbon (PAC) and XAD-16 polymer adsorbent. Binary solution containing hydrophilic (phenol) and hydrophobic (diethyl phthalate, DEP) compounds was employed to evaluate the performance of selective adsorption then coking wastewater was further tested. Results of binary adsorption indicated that the average adsorption percentages of DEP for PAC, XAD-16, CTMS-V and CTES-V were 32.6 ± 11.1%, 50.8 ± 17.4%, 68.7 ± 22.6%, 75.5 ± 27.4% (mean ± standard deviation, n = 5) at varying phenol:DEP ratios (10–100, mol:mol). Coking wastewater treated by PAC, XAD-16 CTMS-V and CTES-V had a decrease of chemical oxygen demand (COD) at 24.6 ± 3.0%, 11.3 ± 1.5%, 4.4 ± 0.08% and 6.8 ± 1.1%. For biochemical oxygen demand (BOD5), PAC and XAD-16 gave a decrease of 47.5 ± 5.6% and 7.8 ± 1.4%, whereas CTMS-V and CTES-V gave an increase of 36.7 ± 4.1% and 57.2 ± 4.6%, respectively. The resulting BOD5:COD were 0.21 ± 0.03, 0.31 ± 0.04, 0.42 ± 0.05 and 0.50 ± 0.05 after treatment by PAC, XAD-16, CTMS-V and CTES-V, respectively. Post incubation of treated coking wastewater showed that the residual TOC after ten-day incubation were 73.1 ± 1.5%, 77.9 ± 3.5%, 73.2 ± 3.7% and 59.2 ± 3.0% for PAC, XAD-16 CTMS-V and CTES-V, respectively. The non-p functional surface (i.e., –CH3 and –CH2CH3) of modified acid-vermiculites played a critical role for capturing hydrophobic compounds from the solution with high concentration of hydrophilic competitive compounds. Ó 2015 Elsevier B.V. All rights reserved.

1. Introduction Coking wastewater is generated in the process of coke making, especially the destructive distillation of coal under temperatures around 1000–1100 °C without oxygen. Due to the pyrolysis of coal, complicated composition of aromatic compounds such as phenols [1], amines [2], polycyclic aromatic hydrocarbons (PAHs) [3] and heterocyclic compounds [4] are generated as by-products and released to the water of coke quenching and gas washing. Coking wastewater is characterized as high organic load, complicated composition [5], strong bio-inhibition [6] and genotoxicity [7] for traditional biological treatment. The refractory contaminants such as PAHs and heterocyclic compounds have significant toxic and inhibitory effects on microbial activity such as anaerobic [8] and ⇑ Corresponding author at: Faculty of Architectural, Civil Engineering and Environment, Ningbo University, Ningbo 315211, PR China. E-mail address: [email protected] (X. Yu). http://dx.doi.org/10.1016/j.seppur.2015.07.007 1383-5866/Ó 2015 Elsevier B.V. All rights reserved.

nitrification [6] processes, thus coking wastewater was commonly reported with a low biodegradability (BOD5:COD < 0.3) [9]. The traditional industrial wastewater treatment, i.e., physicochemical pretreatment followed by activated sludge, encountered a series of challenges, e.g., long HRT, difficulty of sludge acclimation, weak resistance to impact load and high construction cost, for eliminating the hazardous compounds from coking wastewater. Many strengthening techniques were developed to overcome this problem. Enhanced biological treatments include zeolite-biological aerated [4,10], thermophilic anaerobic digestion [11] and biofilm systems [12], etc. The physicochemical improvements include supercritical water oxidation [13], electrochemical degradation [14], Fenton oxidation [5] and adsorption [15]. In addition, techniques combined with biological and physicochemical treatment were also studied [16,17]. The characteristics of coking wastewater, especially the chemical composition, should be fully considered before developing enhanced techniques. It has been reported that the

24

X. Yu et al. / Separation and Purification Technology 151 (2015) 23–30

easy-biodegradable contaminants such as phenols and amines account for the major part of coking wastewater, whereas the hydrophobic bio-inhibitory contaminants such PAHs, pyridine and indole only account for a minor part of total organic compounds [18]. Generally, phenols account for about 80% of the total COD amount in coking wastewater [18,19]. Song et al. analyzed coking wastewater from six coking plants [20]. Results showed that the sum of concentration of phenols and anilines was 23.8 ± 3.0 mg L1, but the sum of 16 PAHs’s concentration was 0.66 ± 0.10 mg L1. Our previous analysis of coking wastewater showed that the concentration of phenols was 178.5 ± 15.2 mg L1 [1], but the total amount of 18 PAHs (identified by US EPA) was in the range of 98.5 ± 8.9–216.0 ± 20.2 lg/L [21]. However, even though the amount of hydrophobic contaminants is much lower than that of hydrophilic compounds, their toxicity and inhibition for biodegradation were significantly higher. Adsorption is a quick and simple technique that has been widely used for pretreating industrial wastewater and polishing the biological effluent [22,23]. For coking wastewater, adsorbents such as activated carbon [2], activated coke [15] and bottom ash [24] have been studied. Most of these adsorbents were evaluated by the adsorption capacity for COD or TOC, but little emphasis has been put on the removal of critical toxic compounds and the improvement of biodegradability. For example, adsorption of organic compounds by carbonaceous adsorbents was driven by the p–p electron donor–acceptor (EDA) mechanism [25,26], thus it is difficult to employ them to remove hydrophobic aromatic pollutants with the competition of hydrophilic aromatic compounds. Ren et al. studied the competitive adsorption of phenol, aniline and n-heptane from tailrace coking wastewater [2]. Their results demonstrated that the hydrophobic compound (n-heptane) has a weak affinity toward activated carbon with the competition of hydrophilic compounds (phenol and aniline). Recent studies reported that the hydrophobic compounds such as PAHs had a strong affinity to the suspended microbial particles. Burmistrz and Burmistrz [27] investigated the fate of 16 PAHs and results indicated that 70% of PAHs in raw coking wastewater were adsorbed by suspended solid during biological treatment. Results of Zhang et al. [3] showed that 56–76% removal of high molecular weight PAHs was contributed by adsorption of sludge. Therefore, a hydrophobic adsorbent without the p–p EDA mechanism could be effective for this task. In this study, the hypothesis of improving biodegradability of coking wastewater by targeted removal of hydrophobic components was examined by different adsorbents. Non-p hydrophobic adsorbents modified by acid-vermiculite were used to selectively remove hydrophobic compounds. For comparison, powdered activated carbon (PAC), a widely used adsorbent in coking wastewater treatment, was selected as a representative of p functional adsorbents. The XAD-16 polymer adsorbent containing both the p and non-p functional group was also tested for detecting the mechanism involved in selective adsorption of hydrophobic compounds. Binary adsorption of phenol (hydrophilic) and diethyl phthalate (hydrophobic) was conducted to evaluate the capability of hydrophobic-targeted adsorption. Finally, coking wastewater was treated with different adsorbents and its biodegradability after treatment was investigated.

2. Materials and methods 2.1. Materials In this study, the natural vermiculite used for preparing clay-based adsorbents was provided by DingSheng Mining Co. LTD from Shijiazhuang, China. The chemical composition of this clay obtained by X-ray florescence (XRF) was 39.95% SiO2, 16.36%

Al2O3, 16.77% Fe2O3, 9.19% MgO, 5.24% K2O, 1.28% CaO, 1.76% TiO2, 0.14% Na2O, 0.20% MnO, 0.05% Cl, and 8.70% loss ignition. The powdered activated carbon (PAC) was supplied by Xinhua Activated Carbon Co. LTD, Shanxi Province, China. The parameters of PAC provided by the supplier are as follows: specific surface area: 1100 m2/g; iodine adsorb: >1000 mg/g; average particle size: 60 lm; ash content: <5%. The Amberlite XAD-16 hydrophobic polymer resin was purchased from Sigma–Aldrich (USA) and the porous parameters provided by the supplier are as follows: specific surface area: 900 m2/g; average pore diameter: 10 nm; total pore volume: 1.82 mL/g. The organic modifier for acid-vermiculite: Chlorotrimethylsilane (CTMS, CAS: 75-77-4) and Chlorotriethysilane (CTES, CAS: 994-30-9) were purchased from Aladdin-Reagent Co. (China); Phenol (log KOW: 1.50, CAS: 108-95-2) and diethyl phthalate (DEP, log KOW: 2.47, CAS: 84-66-2) were used as the representatives of hydrophilic and hydrophobic compounds, which were purchased from Sigma–Aldrich Co. (USA). Deionized water (>18.0 MX) used in the experiment was produced by Milli-Q pure water system (USA). 2.2. Coking wastewater Coking wastewater tested in this study was collected from the wastewater treatment plant (WWTP) of Shaoguan Steel Company, located in Guangdong province of China. The company’s affiliated coking plant has a coke production capability of 1.32 million metric tons per year and generates about 2000 m3 coking wastewater per hour. The biological treatment of the WWTP was designed as an anoxic–oxic–hydrolytic–oxic (A/O/H/O) system coupled with biological fluidized bed. Physicochemical pretreatments (i.e., air-floatation for oil removal, ferrous precipitation for cyanide and sulfide removal) and post-treatment (i.e., coagulation for suspended microbial products) were also employed for enhancing treatment effect. Ten liters of raw coking wastewater were collected from the intake of WWTP using acid-washed, glass bottles that were pre-rinsed with sample wastewater. The water samples were kept on ice during transportation from WWTP to the laboratory. Table 1 shows the main characteristics of coking wastewater studied in this work. 2.3. Preparation of adsorbents The preparation of vermiculite-based adsorbents tested in this study followed the previous methods [28]. Fig. 1 presents the synthetic route of CTMS and CTES modified acid-vermiculite. The main reaction of surface silanization took place through the condensation reaction between CTMS/CTES and surface silanol groups (Si-OH) of acid-vermiculite. The resulting hybrid organic–inorganic materials were abbreviated to CTMS-V and CTES-V, respectively. 2.4. Characterization of adsorbents FTIR spectras of adsorbents were recorded on a Nicolet 6700 spectrometer (ThermoNicolet, USA) in the 4000–400 cm1 region using KBr platelets. The nitrogen adsorption–desorption isotherms of the samples were measured using a Micrometrics ASAP 2020 M apparatus (Micrometrics, USA) at 196 °C. Prior to analysis, the samples of vermiculite were degassed for 4 h at 250 °C below 102 Pa. The parameters of specific surface areas and pores structure were calculated using the Micrometrics software. Pore size distribution of the porous materials was calculated using the Howarth–Kawazoe (HK) formalism for micropores and Barret–Joy ner–Halenda (BJH) method from the desorption branch for mesopores, respectively. Thermal gravimetric analysis was conducted by TG 209 F1 Thermogravimetric Analyzer (Netzsch, Germany).

25

X. Yu et al. / Separation and Purification Technology 151 (2015) 23–30 Table 1 Characteristics of raw coking wastewater derived from Shaoguan Steel Company, Guangdong, China.

a

pH

BOD5 (mg L1)

COD (mg L1)

TOC (mg L1)

NH+4–N (mg L1)

S2 (mg L1)

TCN (mg L1)

Phenolsa (mg L1)

10.6 ± 0.6

681.7 ± 35.3

2321.4 ± 14.3

1449.5 ± 11.3

281.4 ± 5.6

36.7 ± 1.3

9.8 ± 0.8

954.2 ± 3.6

Measured as total volatile phenols.

Fig. 1. Synthetic route of organic modified acid-vermiculites with Chlorotrimethylsilane and Chlorotriethysilane.

Ten mg of samples were heated from 25 to 1000 °C with a heating rate of 10 °C min1 and a nitrogen flow of 30 mL min1. 2.5. Adsorption experiments 2.5.1. Binary adsorption Binary solution with DEP and phenol was prepared to test the selective adsorption capability for hydrophobic organic compound with the competition of hydrophilic compounds. The binary solution was prepared with different phenol:DEP molar ratios: 10, 20, 40, 60 and 100. Adsorption experiments were conducted by adding 10.0 mg of adsorbents to 150 mL flasks containing 100 mL binary solutions. The flasks were shaken at 25 °C for 24 h. Control experiments were conducted for correction of possible removal of organic compounds by mechanisms other than adsorption. After centrifugation, the centrifugate were filtered through 0.45-lm filters (Membrana, Germany) prior to be analyzed by high performance liquid chromatography (HPLC). The amount of adsorption at equilibrium, qe (mg/g), was calculated by

qe ¼

ðC 0  C e Þ  V W

ð1Þ

where C0 and Ce (mg L1) are the liquid-phase concentrations of CR at initial and equilibrium states, respectively. V (L) is the volume of the solution and W (g) is the mass of adsorbent used. 2.5.2. Adsorption of coking wastewater The adsorption experiment for coking wastewater was conducted by adding 1.5 L of coking wastewater and 3.0 g adsorbents into a series of 2 L bottles, and then shaken for 24 h at 25 °C. After centrifugation, the centrifugate were filtered through 0.45-lm filters prior to be analyzed by total organic carbon (TOC), chemical oxygen demand (COD) and biochemical oxygen demand (BOD5). 2.5.3. Incubation of coking wastewater An incubation experiment was conducted to evaluate the biodegradability of coking wastewater with or without treatment by adsorbents. Activated sludge derived from aerobic tank of WWTP, Shaoguan, China was collected as the inoculum source for incubation. Three liter samples of biological sludge were firstly aerated for 2 h immediately after sampling, and then samples were centrifuged at 1500 rpm for 20 min at 4 °C. The supernatants were

stored in acid-washed bottles at 4 °C for further inoculation. Incubation experiments were carried out in sterile 500 mL flasks using the raw/treated coking wastewater from 2.4. Activated sludge supernatants were inoculated aseptically in flasks at 1% v/v. The flasks were then incubated in the dark at 21 ± 0.5 °C and maintained under agitation. 10 mL samples were obtained for monitoring the water qualities of COD, BOD and TOC at the first, second, fourth and tenth day. All the samples were filtered through 0.45-lm membrane before analysis. 2.6. Analytical methods The HPLC analysis was carried out on a LC-20AT series LC chromatographic system (Shimadzu, Japan) equipped with a vacuum degasser (DGU-20A5), a autosampler (SIL-20A), a thermostated column compartment (CTO-10ASVP), a pump (LC-20AT) and a UV–Vis detector (SPD-20AV). The analyses were performed on a Tracer excel 120 OctaDecilSilica-A column, 250 mm  4.6 mm, 5 lm particle size (Agilent, USA). The mobile phase consisted of a mixture of 80% (v/v) HPLC methanol (Merk) and ultra-pure water, with flow of 1 mL min1. The detector was set at 254 nm, the column temperature was maintained at 35 °C and injection volume was 10 lL with a draw speed of 1 mL min1. Phenol and DEP was determined at the retention time of 3.68 and 5.12 min, respectively. For quantification purposes calibration plots were performed under the instrumental conditions used. TOC concentration was measured by a TOC-VCPH carbon analyzer (Shimadzu, Japan). COD and BOD5 were measured according to standard methods (APHA, 2005). COD was measured by the closed reflux colorimetric method. Dilution water used for BOD5 measurement was seeded with the extraction of aerobic sludge from WWTP. HACH BODTrak II manometric respirometer (USA) was employed to determine the oxygen level of samples at 21 ± 0.5 °C. 3. Results 3.1. Characterization of modified acid-vermiculites Fig. 2 shows the characterization of FTIR (a) and thermal gravimetric curves (b) for modified acid-vermiculites. The organic

26

X. Yu et al. / Separation and Purification Technology 151 (2015) 23–30

functional groups grafted on the surface of acid-vermiculite can be observed with the FTIR spectrum. For CTMS-V, the absorption bands observed at 750 (c Si-(CH3)3) and 2958 cm1 (mas CH3) were associated with the functional groups of CTMS [29,30]; accordingly, the absorption bands observed at 2858 (ms CH2), 2927 (mas CH2) and 2958 (mas CH3) cm1 were attributed to the functional groups of CTES [30]. Thermal gravimetric analysis was used to examine the organic loading on acid vermiculite. It can be observed that the weight loss of water (<150 °C) for acid-vermiculite, CTMS and CTES was 4.5%, 2.2% and 1.7%, respectively. Weight loss in the temperature period of 150–1000 °C for acid-vermiculite, CTMS and CTES was 7.7%, 8.6% and 11.6%, respectively. Table 2 lists the porous parameters of acid-vermiculite before and after modification. The specific surface area (SBET) and total pore volume (Vt) of acid-vermiculite decreased after modification by CTMS and CTES.

For the pore size distribution, i.e., proportion of micropore and mesopore, there was no significant change after organic modification. 3.2. Adsorption of binary solute Fig. 3 shows the selective adsorption of DEP from phenol by different adsorbents. The molar ratios of phenol:DEP were controlled at the levels of 10, 20, 40, 60 and 100. Generally, the percentage of adsorbed DEP decreased with the increase of phenol in binary solution. At a phenol:DEP molar ratio of 10, the percentages of adsorbed DEP for PAC, XAD-16, CTMS-V and CTES-V were 30.9%, 78.8%, 84.6% and 99.8%, respectively. It can be observed that adsorbents showed different responses for the increasing proportion of phenol. PAC exhibited slight variation on percentages of adsorbed DEP under the phenol: DEP ratio of 60, then it decreased to 14.7% dramatically when phenol:DEP reached 100. XAD-16 resin showed a continuous decrease of DEP adsorption percentage with the increase of phenol: DEP ratio from 20 to 100. CTMS-V and CTES-V showed similar responses to the increase of phenol’s proportion. With the phenol:DEP ratio of 10, 20 and 40, adsorbed DEP’s percentages for CTMS-V were 84.6%, 76.2% and 91.6%, respectively; for CTES-V they were 99.8%, 90.2% and 94.0%, respectively. After phenol:DEP reached 60, adsorbed DEP’s percentages decreased significantly for both CTMS-V and CTES-V. At phenol:DEP of 100, adsorbed DEP’s percentages of CTMS-V and CTES-V were 36.9% and 37.8%, respectively. 3.3. Adsorption for coking wastewater Fig. 4 compares the BOD5, TOC, COD and BOD5:COD of coking wastewater before and after treatment with adsorbents. It can be observed that PAC showed the largest removal rates of TOC (24.5 ± 2.3%) and COD (24.6 ± 3.0%) among the four adsorbents. The TOC removal rates for XAD-16 resin, CTMS-V and CTES-V were 11.3 ± 1.5%, 4.4 ± 0.08% and 6.8 ± 1.1%, respectively; and the COD removal rates for XAD-16 resin, CTMS-V and CTES-V were 11.8 ± 1.3%, 4.4 ± 0.07% and 6.9 ± 1.2%, respectively. However, the results of BOD5 for these adsorbents were opposite. PAC and XAD-16 resin showed a 47.5 ± 5.6% and 7.8 ± 1.4% decrease of BOD5 compared with raw wastewater, whereas CTMS-V and CTES-V showed a 36.7 ± 4.1% and 57.2 ± 4.6% increase after the treatment, respectively. According to these results of BOD5:COD, the index of biodegradability for wastewater, BOD5:COD, had a 30.18 ± 6.7% decrease for samples treated with PAC, whereas XAD-16 resin, CTMS-V and CTES-V improved it to 4.1 ± 1.1%, 43.3 ± 8.1% and 69.3 ± 7.6% increase, respectively. 3.4. Effect of selective adsorption on biodegradability of coking wastewater

Fig. 2. Characterization of the modified acid-vermiculites: (a) FTIR spectra; (b) thermal gravimetric curves.

Fig. 5 showed the degradation kinetic plots of coking wastewater with and without treatment by adsorbents. Coking wastewater treated by PAC showed the lowest residual TOC of 84.2 ± 2.6% on the first incubation day. Coking wastewater without treatment

Table 2 Porous parameters calculated from N2 adsorption-desorption isotherms. Samples

SBET (m2/g)

Dav (nm)

Vt (cm3 g1)

Acid vermiculite CTMS-V CTES-V

528.0 358.4 361.0

2.89 2.83 2.74

0.349 0.250 0.247

Pore size distribution Vmicro (%)

Vmeso (%)

0.122 (35) 0.085(34) 0.087(35)

0.227 (65) 0.165(66) 0.160(65)

SBET: specific surface area; Dav: average pore diameter; Vt: total pore volume; Vmicro: volume of micropore; Vmeso: volume of mesopore.

X. Yu et al. / Separation and Purification Technology 151 (2015) 23–30

27

Fig. 3. Bisolute adsorption of DEP and phenol by different materials.

and treated by XAD-16 resin, CTMS-V and CTES-V had a residual TOC of 96 ± 1.1%, 97 ± 0.9%, 98 ± 0.7% and 97 ± 0.8% in the first incubation day, respectively. On the second incubation day, the residual TOC percentages for raw coking wastewater, PAC, XAD-16, CTMS-V and CTES-V decreased to 91.2 ± 2.7%, 79 ± 3.2%, 93.5 ± 2.3%, 85 ± 2.6% and 76 ± 2.4%, respectively. On the fourth and tenth incubation day, degradation of coking wastewater treated with PAC showed a slight decrease from 87.5 ± 2.5% to 85.3 ± 2.1%, and 77.1 ± 2.3%, respectively. However, the samples treated with XAD-16, CTMS-V and CTES-V still exhibited a significant decrease from 91.3 ± 2.8% to 77.5 ± 3.9%, 83.5 ± 2.4% to 73 ± 3.6% and 73.2 ± 2.2% to 59.6 ± 2.9, respectively.

4. Discussions 4.1. Evaluation of selective adsorption for DEP Fig. 4. BOD5, TOC, COD and BOD5:COD of coking wastewater before and after treatment with adsorbents (error bar represent the standard deviation, n = 3).

Fig. 5. Degradation kinetic of coking wastewater after treatment of selective adsorption (error bar represent the standard deviation, n = 3).

Phenol and DEP are the representative hydrophilic and hydrophobic organic pollutants in coking wastewater. To evaluate the adsorbents’ selective adsorption capability for hydrophobic compounds, binary solution of phenol and DEP was used to test the percentage of adsorbed DEP at different phenol:DEP ratios. According to the previous studies [31], strength of competition is positive correlated with the ratio of competitive solute and the affinity between competitive components and adsorbent. Results shown in Fig. 3 indicated that PAC had the weakest capability for adsorbing DEP from phenol among the tested adsorbents. In other words, phenol behaved the strongest competition for DEP in PAC’s adsorption system. This phenomenon was related to the adsorption mechanism of carbonaceous materials. It was well known that the adsorption of aromatic compounds on carbonaceous materials was controlled by the p–p EDA interaction [26]. Due to both phenol and DEP being single-ring aromatic compounds, phenol was able to perform a strong competitive effect on DEP’s adsorption. Therefore, PAC was not able to selectively adsorb DEP from the solution under the higher concentration of phenol. It should be noted that within the phenol:DEP ratio of 10-60, adsorbed DEP accounted for 35.3–44.5% of PAC’s total adsorption amount, which was unmatched with its proportion (about 1.6– 9.1%). This can be attributed to the hydrophobic interaction of activated carbon besides p–p EDA. Due to the graphene-like structure, PAC had hydrophobic surface characteristics and its strength was dependent on the graphene plane’s edge functionality [32].

28

X. Yu et al. / Separation and Purification Technology 151 (2015) 23–30

Though the substituents such as hydroxyl (–OH) and carboxyl (–COOH) can increase the hydrophilicity of PAC [26], the central site of graphene plane still prefer hydrophobic compounds by hydrophobic interaction. The monomer of XAD-16 polymer adsorbent consists of two phenyl groups (–C6H5) and two alkyls (butyl and propyl). Although the phenyl groups adsorb aromatic compounds by p–p EDA interaction, the alkyl groups can adsorb hydrophobic organic compounds via hydrophobic interaction. Binary adsorption showed that XAD-16 polymer adsorbent behaved a high removal for DEP (78.0%) at phenol:DEP ratio of 10, then it decreased continuously with the increase of phenol proportion. This result implies that with the increase of non-p hydrophobic groups, the affinity between phenol and adsorbent was decreased so that XAD-16 behaved with stronger selective adsorption toward DEP compared with PAC. However, due to its molecular characteristics, XAD-16 was sensitive to the ratio of phenol:DEP in binary solution. The rapid decrease of DEP uptake with the increase of phenol proportion demonstrated the p–p EDA interaction attributed to the phenyl groups in XAD-16 polymer. For the organic modified acid-vermiculites, non-p hydrophobic functional groups, i.e., methyl and ethyl were the dominant organic functional groups (Fig. 1). The grafted functional groups formed a hydrophobic surface between liquid–solid interface, thus hydrophobic interaction become the predominant mechanism for removing organic compounds. As a result, CTMS-V and CTES-V showed a significantly higher DEP uptake under phenol:DEP ratio of 60 compared with PAC and XAD-16. In this case, the surface of adsorbent was not site-specific for phenol molecule, and the competitive strength was decreased significantly. Moreover, it can be found that CTMS-V and CTES-V showed stable removal for DEP with the ratio of phenol:DEP ratio increasing from 10 to 40. This phenomenon demonstrates the high selectivity for DEP of organic modified vermiculites, which was highly driven by hydrophobic interaction and it was slightly affected even at a high concentration of hydrophilic competitive compounds. When phenol:DEP ratio reached 60, DEP uptake decreased significantly and more phenol replaced DEP to be adsorbed on the surface of CTMS-V and CTES-V. In a competitive adsorption system, the adsorption potential of components was related to their proportions [33,34]. Decrease of DEP uptake under high concentration of phenol implies that the absolute predominance of phenol’s ratio occupied more adsorption sites. Results also suggest that CTES-V exhibited higher DEP uptake compared with CTMS-V. From the result of Fig. 2(b), CTES-V had higher organic loading due to the longer alkyl chain of modifier, thus it can provide more hydrophobic adsorption sites for DEP. Above discussion indicate that at the certain solute ratio, the difference of adsorbent’s affinity toward solutes was the critical factor determining the strength of competitive adsorption. As a result of increase of non p–p function groups, XAD-16 behaved a higher selectivity for DEP compared with PAC. For modified vermiculites, non p–p function groups became exclusive and DEP’s selective adsorption increased dramatically. 4.2. Effects of selective adsorption on biodegradability of coking wastewater Coking wastewater is a type of organic industrial wastewater with distinctive features such as high concentration of pollutants, complex chemical composition, high toxicity and bioinhibitory. Among these complex organic matters, phenols and amines account for more than 80% of total organic carbons [18,19]. By contrast, the high toxic and recalcitrant organic matters such as PAHs, heterocyclic compounds, phthalate esters only account for less than 5% due to the low solubility [3]. Despite the

relatively low concentration, these contaminants contribute predominantly to the toxicity and inhibitory of coking wastewater. Thus, selective removal of these hydrophobic contaminants could be critical in improving the biodegradability of coking wastewater. Results showed that the capability of removing TOC and COD for adsorbents followed the order as: PAC > XAD-16 > CTES-V > CTMS-V. This was in accordance with the result of binary solutes adsorption: PAC and XAD-16 had higher total adsorption amounts than that of CTES-V and CTMS-V. However, coking wastewater treated with PAC and XAD-16 had a 47.5% and 7.8% decrease of BOD5, respectively. The decrease of BOD5 for PAC was 48.1% higher than that of COD. The unmatched decrease of BOD5 implies that coking wastewater treated by PAC was more recalcitrant for biodegradation. Results of binary adsorption indicated that PAC exhibited more affinity toward hydrophilic during the competitive adsorption. Therefore, the major contaminants removed by PAC from coking wastewater were the hydrophilic compounds, which had relatively high biodegradability. As a result, the proportion of hydrophobic compounds increased and the biodegradability of coking wastewater decreased accordingly. For XAD-16, the decrease of BOD5 was 31.2% lower than that of TOC, implying that XAD-16 removed more biological inhibitory compounds. By contrast, coking wastewater treated with CTMS-V and CTES-V showed a 36.8% and 57.2% increase of BOD5, respectively. Theoretically, BOD5 should have a corresponding decrease with TOC or COD for the single-solute solution. The blank test suggested that there was no significant release of organic compounds from modified acid-vermiculites, thus the reason should be attributed to the increase of biodegradability after adsorption. Results of binary adsorption indicated that organic modified acid-vermiculites had strong capability of selective adsorption for hydrophobic organic compounds with the competition of hydrophilic organic compounds. Therefore, by selective removal of hydrophobic inhibitory compounds, the biodegradability of coking wastewater was improved. BOD5:COD ratio is served as an index to evaluate biodegradability for wastewater more directly. Coking wastewater treated with PAC had a 30.4% decrease of BOD5:COD ratio, further demonstrating the decrease of biodegradability. For XAD-16, CTMS-V and CTES-V, BOD5:COD ratios were increased by 4.1%, 43.3% and 69.2% increase, respectively. The post incubation was conducted to further test the biodegradability of coking wastewater. Due to the high removal of total organic pollutants, coking wastewater treated with PAC showed the largest decrease (15.8%) among adsorbents on the first day. High concentration of organic pollutants (even the simple contaminants like phenol, etc.) in coking wastewater has a strong inhibitory effect on bacterial activity [6]. As a result, coking wastewater treated with XAD-16, CTMS-V and CTES-V showed a small decrease of TOC in the first day. On the second incubation day, coking wastewater treated with CTMS-V and CTES-V showed a much higher decrease of TOC compared with other samples. This phenomenon matches the results that hydrophobic inhibitory compounds can be removed selectively and biodegradability can be improved accordingly. From the second to the tenth incubation day, the raw coking wastewater and the sample treated with PAC showed a slight decrease, demonstrating that to further microbial degradation activity was inhibited by the inhibitory components. By contrast, coking wastewater treated with XAD-16, CTMS-V and CTES-V continued to be degraded during this period. The sample treated by CTES-V showed the largest degradation percentage (41.2%) for TOC among various materials during incubation period. The better performance of CTES-V compared with CTMS-V was due to the higher organic loading, which means stronger capture capability for hydrophobic contaminants.

X. Yu et al. / Separation and Purification Technology 151 (2015) 23–30

29

Fig. 6. Illustration of different adsorption process based on hydrophobic and p–p EDA interface.

4.3. Mechanism of selective adsorption for improving biodegradability of coking wastewater This mechanism of selective adsorption for hydrophobic organic compounds is shown in Fig. 6. Coking wastewater was a kind of typical organic industrial wastewater characterized as highly toxic and a strong bioinhibition. However, from the viewpoint of chemical composition, the major part of organic contaminants such as phenols and amines were simple and easy biodegraded compounds. Most of these pollutants can be categorized as the hydrophilic group. On the other hand, the bio-inhibitory and refractory contaminants such as PAHs and heterocyclic compounds can be categorized as the hydrophobic group, which account for a small relatively part of the total organic compounds. Adsorption is a quick and simple pretreatment to reduce the bio-inhibitory compounds before biological treatment. However, as a traditional and common adsorbent, activated carbon is not able to remove the hydrophobic pollutants selectively because its p–p EDA mechanism is equivalent to aromatic compounds regardless of the hydrophobicity. Particularly, the competition from hydrophilic compounds in coking wastewater was so strong that most of the adsorption sites were occupied by hydrophilic compounds. As shown in Fig. 6, although carbonaceous materials can remove more pollutants compared with other adsorbents, the proportion of bio-inhibitory components will be increased. This kind of treatment changes the composition of wastewater and cause biodegradability. The hydrophobic adsorbents shown in this study captured the organic compounds depending only on the hydrophobicity of target solutes, which had no relationship with the molecular structure. In this case, more hydrophobic compounds can be removed from coking wastewater and thus the biodegradability was improved despite the small reduction of total organic matters. Therefore, these two kinds of adsorbents change coking wastewater’s composition in opposite directions, which result in levels of

biodegradability. XAD-16 polymer adsorbent has the composite functional groups in its molecule structure and its performance for improving biodegradability of coking wastewater was between PAC and modified acid-vermiculites. The above results demonstrate that selective adsorption for complex industrial wastewater such as coking wastewater is more efficient for improving the biodegradability compared with the general adsorption, and it can potentially serve as a pretreatment according to the specific characteristics of various industrial wastewater.

5. Conclusion Based on the characteristics of coking wastewater, selective adsorption for hydrophobic organic contaminants was conducted aiming to improve the biodegradability of this refractory industrial wastewater. Organic modified acid-vermiculites with non-p functional groups showed significantly stronger selective adsorption capability for hydrophobic compounds compared with powdered activated carbon and XAD-16 polymer adsorbent. The non-p functional surface (i.e., –CH3 and –CH2CH3) played a critical role for capturing hydrophobic compounds from the solution with high concentration of hydrophilic competitive compounds. Coking wastewater treated with activated carbon showed a high removal of COD and TOC, but its BOD5:COD ratio decreased significantly indicating the decline of biodegradability. On the contrary, although the modified acid-vermiculites removed the TOC slightly, it was able to increase the BOD5:COD ratio significantly meaning an improvement for the biodegradability. Post incubation for coking wastewater demonstrated that selective adsorption resulted in higher removal of total organic compounds after a short adaptation period (1 day in this study) compared with activated carbon. Results shown in this work demonstrate that selective adsorption for high-loading industrial wastewater like coking wastewater

30

X. Yu et al. / Separation and Purification Technology 151 (2015) 23–30

could be an improvement for traditional adsorption techniques which mainly pursue high adsorption capability for total pollutants.

[15] [16]

Acknowledgements [17]

This study was supported by the National Science Foundation of China (No. 21207040, 51278199), Special disciplinary fund of Guangdong education department (2013CXZDA004) and Chinese Postdoctoral Science Foundation (No. 2012M521607). The authors are thankful to Mr. Reid Heaton (Clemson University) for the English revision of this manuscript. References [1] W.H. Zhang, C.H. Wei, C.H. Feng, Y. Ren, Y. Hu, B. Yan, C.F. Wu, The occurrence and fate of phenolic compounds in a coking wastewater treatment plant, Water Sci. Technol. 68 (2013) 433–440. [2] Y. Ren, T. Li, C.H. Wei, Competitive adsorption between phenol, aniline and nheptane in tailrace coking wastewater, Water Air Soil Poll. 224 (2013). [3] W.H. Zhang, C.H. Wei, X.S. Chai, J.Y. He, Y. Cai, M. Ren, B. Yan, P.A. Peng, J.M. Fu, The behaviors and fate of polycyclic aromatic hydrocarbons (PAHs) in a coking wastewater treatment plant, Chemosphere 88 (2012) 174–182. [4] Y.H. Bai, Q.H. Sun, R.H. Sun, D.H. Wen, X.Y. Tang, Bioaugmentation and adsorption treatment of coking wastewater containing pyridine and quinoline using zeolite-biological aerated filters, Environ. Sci. Technol. 45 (2011) 1940– 1948. [5] L.B. Chu, J.L. Wang, J. Dong, H.Y. Liu, X.L. Sun, Treatment of coking wastewater by an advanced Fenton oxidation process using iron powder and hydrogen peroxide, Chemosphere 86 (2012) 409–414. [6] Y.M. Kim, D. Park, D.S. Lee, J.M. Park, Inhibitory effects of toxic compounds on nitrification process for cokes wastewater treatment, J. Hazard. Mater. 152 (2008) 915–921. [7] Y.R. Dong, J.T. Zhang, Testing the genotoxicity of coking wastewater using Vicia faba and Hordeum vulgare bioassays, Ecotoxicol. Environ. Saf. 73 (2010) 944– 948. [8] W. Wang, H.J. Han, M. Yuan, H.Q. Li, Enhanced anaerobic biodegradability of real coal gasification wastewater with methanol addition, J. Environ. Sci. – China 22 (2010) 1868–1874. [9] X. Zhou, Y.X. Li, Y. Zhao, Removal characteristics of organics and nitrogen in a novel four-stage biofilm integrated system for enhanced treatment of coking wastewater under different HRTs, Rsc. Adv. 4 (2014) 15620–15629. [10] Y.H. Bai, Q.H. Sun, R. Xing, D.H. Wen, X.Y. Tang, Analysis of denitrifier community in a bioaugmented sequencing batch reactor for the treatment of coking wastewater containing pyridine and quinoline, Appl. Microbiol. Biotechnol. 90 (2011) 1485–1492. [11] W. Wang, W.C. Ma, H.J. Han, H.Q. Li, M. Yuan, Thermophilic anaerobic digestion of Lurgi coal gasification wastewater in a UASB reactor, Bioresour. Technol. 102 (2011) 2441–2447. [12] Y.M. Li, G.W. Gu, I. Zhao, H.Q. Yu, Y.L. Qiu, Y.Z. Peng, Treatment of coke-plant wastewater by biofilm systems for removal of organic compounds and nitrogen, Chemosphere 52 (2003) 997–1005. [13] X. Du, R. Zhang, Z.X. Gan, J.C. Bi, Treatment of high strength coking wastewater by supercritical water oxidation, Fuel 104 (2013) 77–82. [14] Y. Wang, S.J. Sun, G.F. Ding, H. Wang, Electrochemical degradation characteristics of refractory organic pollutants in coking wastewater on

[18]

[19]

[20]

[21]

[22] [23]

[24]

[25] [26] [27] [28]

[29]

[30]

[31]

[32]

[33]

[34]

multiwall carbon nanotube-modified electrode, J. Nanomater. 2012 (2012) 1–5. M.H. Zhang, Q.L. Zhao, X. Bai, Z.F. Ye, Adsorption of organic pollutants from coking wastewater by activated coke, Colloid Surface A 362 (2010) 140–146. P. Lai, H.Z. Zhao, M. Zeng, J.R. Ni, Study on treatment of coking wastewater by biofilm reactors combined with zero-valent iron process, J. Hazard. Mater. 162 (2009) 1423–1429. X.W. Jin, E.C. Li, S.G. Lu, Z.F. Qiu, Q. Sui, Coking wastewater treatment for industrial reuse purpose: combining biological processes with ultrafiltration, nanofiltration and reverse osmosis, J. Environ. Sci. – China 25 (2013) 1565– 1574. M. Han, G.K. Li, N. Sang, Y.R. Dong, Investigating the bio-toxicity of coking wastewater using Zea mays L. assay, Ecotoxicol. Environ. Saf. 74 (2011) 1050– 1056. H.L. Chen, G. Yang, Y.J. Feng, C.L. Shi, S.R. Xu, W.P. Cao, X.M. Zhang, Biodegradability enhancement of coking wastewater by catalytic wet air oxidation using aminated activated carbon as catalyst, Chem. Eng. J. 198 (2012) 45–51. G.X. Song, C.Y. Zhu, Y.M. Hu, J.M. Chen, H.F. Cheng, Determination of organic pollutants in coking wastewater by dispersive liquid–liquid microextraction/ GC/MS, J. Sep. Sci. 36 (2013) 1644–1651. W.H. Zhang, C.H. Wei, B. Yan, C.H. Feng, G.B. Zhao, C. Lin, M.Y. Yuan, C.F. Wu, Y. Ren, Y. Hu, Identification and removal of polycyclic aromatic hydrocarbons in wastewater treatment processes from coke production plants, Environ. Sci. Poll. Res. 20 (2013) 6418–6432. M. Ahmaruzzaman, Adsorption of phenolic compounds on low-cost adsorbents: a review, Adv. Colloid Interf. Sci. 143 (2008) 48–67. A.K. Jain, V.K. Gupta, A. Bhatnagar, Suhas, A comparative study of adsorbents prepared from industrial wastes for removal of dyes, Sep. Sci. Technol. 38 (2003) 463–481. W.L. Sun, Y.Z. Qu, Q. Yu, J.R. Ni, Adsorption of organic pollutants from coking and papermaking wastewaters by bottom ash, J. Hazard. Mater. 154 (2008) 595–601. B. Pan, B.S. Xing, Adsorption mechanisms of organic chemicals on carbon nanotubes, Environ. Sci. Technol. 42 (2008) 9005–9013. C. Moreno-Castilla, Adsorption of organic molecules from aqueous solutions on carbon materials, Carbon 42 (2004) 83–94. P. Burmistrz, M. Burmistrz, Distribution of polycyclic aromatic hydrocarbons in coke plant wastewater, Water Sci. Technol. 68 (2013) 2414–2420. X.B. Yu, C.H. Wei, L. Ke, H.Z. Wu, X.S. Chai, Y. Hu, Preparation of trimethylchlorosilane-modified acid vermiculites for removing diethyl phthalate from water, J. Colloid Interf. Sci. 369 (2012) 344–351. X.S. Zhao, G.Q. Lu, Modification of MCM-41 by surface silylation with trimethylchlorosilane and adsorption study, J. Phys. Chem. B 102 (1998) 1556–1561. T. Shigeno, M. Nagao, T. Kimura, K. Kuroda, Direct silylation of a mesostructured precursor for novel mesoporous silica KSW-2, Langmuir 18 (2002) 8102–8107. M.P.F. Fontes, P.C. Gomes, Simultaneous competitive adsorption of heavy metals by the mineral matrix of tropical soils, Appl. Geochem. 18 (2003) 795– 804. D.H. Lin, B.S. Xing, Adsorption of phenolic compounds by carbon nanotubes: role of aromaticity and substitution of hydroxyl groups, Environ. Sci. Technol. 42 (2008) 7254–7259. J.C. Crittenden, P. Luft, D.W. Hand, J.L. Oravitz, S.W. Loper, M. Arl, Prediction of multicomponent adsorption equilibria using ideal adsorbed solution theory, Environ. Sci. Technol. 19 (1985) 1037–1043. B.S. Xing, J.J. Pignatello, B. Gigliotti, Competitive sorption between atrazine and other organic compounds in soils and model sorbents, Environ. Sci. Technol. 30 (1996) 2432–2440.