In this paper, the sorption characteristics and mechanisms of phenanthrene and pyrene onto peat (PT) and surfactant-modified peat (MPT) were investigated. Sorption results fit closely to the Partition model and Freundlich model, the coefficient of determination (R2) were higher than 0.98 and 0.99, respectively. The contributions of partition and adsorption to the total sorption of phenanthrene and pyrene by PT and MPT were analyzed quantitatively. Results indicate that the sorption process is a combination of partition and adsorption, and partition plays a major role in the sorption process. The contribution of partition increased with the increasing of initial concentrations of polycyclic aromatic hydrocarbons. The sorption ability of phenanthrene and pyrene by PT and MPT followed the order of pyrene > phenanthrene. MPT has demonstrated potential as a promising new class of materials for environmental remediation of organic pollutants.

INTRODUCTION

Polycyclic aromatic hydrocarbons (PAHs), which are often generated during the incomplete burning of organics, are extremely harmful to humans due to their high toxicity and long-term persistence. Once released into the environment, PAHs can be transported long distances through air and water flows, and are difficult to biodegrade because of their chemical persistence and semi-volatile nature (Kong et al. 2012; Lindgren et al. 2014). Significant PAH contamination has been widely found in aquatic ecosystems (Qu et al. 2008). Current methods like chemical oxidation or biological treatment have sometimes been shown to be inefficient at removing trace levels of organic pollutants (Tan et al. 2009). Adsorption technology provides a simple approach for effective removal of organic pollutants from water and thus has a wide applicability in water pollution control. For the adsorptive removal of PAHs, various kinds of adsorbents have been investigated. As well as activated carbon, agricultural-derived adsorbents of natural origin such as bagasse, coconut shells, peat, coconut pith and wood fibre show prospects for application in the sorption of organic pollutants, due to their inherent sorption capability, biodegradability, ready availability and cost-effectiveness (Xia & Ball 1999; Pei et al. 2007). Peat is a porous biomaterial formed by the accumulation of partially decayed vegetation matter under acidic and anaerobic conditions. Fibric peats are the least decomposed peats, and comprise intact fibres. The intact fibre provides fibric peat with better mechanical strength when used as a sorbent. Peat has already been used in field applications for treatment of organic pollution from contaminated waters (Kalmykova et al. 2014). Moreover, the sorption performance of natural biomaterials can be improved through minor modification. Various functional groups have been grafted onto the surface of natural adsorbents to selectively adsorb different compounds (Motsa et al. 2012). Shen used a modified bentonite with a short-chain cationic surfactant to adsorb dissolved organic matter from water (Shen 2002). Zhou et al. (2008) studied the adsorption/desorption behaviour of PAHs (phenanthrene, anthracene, pyrene) on three kinds of modified charcoal which were different in structure, and found that the sorption of PAHs on charcoal is jointly controlled by partition and surface adsorption. Most studies reported in the literature were conducted on sorption performance for PAH removal, but quantitative descriptions of the relative contribution of the partition function and surface adsorption to the total sorption of organic compounds by natural and modified agricultural materials are rather limited. The high sorption affinity of pine needles with phenanthrene was reported by Chen et al. (Chen et al. 2008; Chen et al. 2011), and partition process was found to play a predominant role in PAH biosorption by plant residues. The combined adsorption and partition effects of adsorbents with varying fractions have not been clearly defined, including the different contributions of adsorption and partition to the sorption process. It is essential to understand the PAHs sorption behaviours and mechanisms to improve adsorbent design and sorption processes operation (Carmo et al. 2000; Kopinke et al. 2001). A previous study revealed that quaternary ammonium surfactant-modified peat is an effective sorbent for removing aqueous phenanthrene. Low temperature and neutral pH favoured phenanthrene adsorption (Zhou et al. 2012). In this paper, sorption isotherms of phenanthrene and pyrene on peat and cetyltrimethyl ammonium bromide (CTAB)-modified peat from water were compared. The relative contributions of partition and adsorption to the total sorption of phenanthrene and pyrene were quantified via an isotherm separation method. The effects of modification on the contributions are discussed to obtain a better understanding of sorption mechanisms and use that knowledge to assess PAHs environmental risks.

MATERIALS AND METHODS

Materials and reagents

A slightly decomposed fibric peat with an average degree of decomposition of 30% was obtained from Jixiang Peat Co. (Jilin, China). Phenanthrene (PHE) and pyrene (PYR) (> 98.9% purity), used as the model PAHs in this study, were purchased from AccuStandard Chemical Co. (CT, USA) and were used without further purification. The methanol used was high-performance liquid chromatography (HPLC)-grade and was obtained from TEDIA Chemical Co. (OH, USA). CTAB and other reagents were of analytical grade, obtained from Lingfeng Chemical Co. (Shanghai, China).

Analytical method

After sorption, the sample solution was filtered through an organic membrane filter (0.22 μm), after which the concentrations of PAHs were determined by HPLC (Hitachi D2000) with an Agilent (Santa Clara, CA, USA) LiChroshper PAH reverse phase column (Inertsil ODS-P sum 3 × 250 mm PAHs). The determining conditions were as follows: the injection volume was 50 μL, phenanthrene and pyrene were determined at monitoring wavelengths of 251 nm and 240 nm, respectively, the flow rate was 0.75 mL/min, and the column temperature was 30 °C. The equilibrium PAHs sorption capacity was calculated as follows: 
formula
1
where qe is the equilibrium sorption capacity (μg·g−1); c0 and ce are solute concentrations before and after sorption, respectively (μg·L−1); V is the solution volume (L); m is the amount of sorbent (g).

Preparation of CTAB-modified peat

The modification of peat was based primarily on ion exchange between cetyltrimethyl ammonium cations and polar functional groups in a peat matrix such as carboxyls and phenolic hydroxyls. The peat was washed several times with distilled water to remove any particles adhering to the surface and any water-soluble particles, then oven-dried at 65 °C. The peat used has a range of particle sizes and was sieved through a range of sieves; only particles that passed through a 0.25-mm mesh were used in this study. Following this operation, 1 g of oven-dried peat was poured into 500 mL conical flasks containing 250 mL of NaOH solution (0.01 mol·L−1), then shaken at 200 rpm for 1 h at 30 °C. The peats were then separated from the solution by suction filtration, washed with distilled water, and dried for 2 h at 65 °C. Then the base treated peats were obtained after sieving again. CTAB-modified peats were prepared by mixing 15.0 g of base treated peat and CTAB (1:1.108) in 1,000 mL conical flasks containing 500 mL of distilled water. The mixture was agitated at 200 rpm in an HHS-4 water bath oscillator (Shanghai Boxun, China) for 4 h at 35 °C. Then the peats were separated from the solution by suction filtration and washed with distilled water several times, then dried for 2 h at 65 °C. The CTAB-modified peat was termed surfactant-modified peat (MPT) Selected properties of the peats are listed in Table 1.

Table 1

Characteristics of PT and MPT

 Ultimate analysis     
 C% H% N% Ash content (%) BET surface area (m2·g−1Zeta potential (mv, pH 6.9) CTAB amount (mmol·g−1
PT 42.59 4.53 1.94 17.54 0.924 −43.6 – 
MPT 49.86 6.93 2.45 10.05 0.547 −31.0 0.364 
 Ultimate analysis     
 C% H% N% Ash content (%) BET surface area (m2·g−1Zeta potential (mv, pH 6.9) CTAB amount (mmol·g−1
PT 42.59 4.53 1.94 17.54 0.924 −43.6 – 
MPT 49.86 6.93 2.45 10.05 0.547 −31.0 0.364 

Characterization of the peats

Brunauer–Emmett–Teller (BET) data were determined by a NOVA-2000E surface area analyzer (Chen et al. 2011). Zeta potentials of mineral suspensions were obtained using a Dispersion Technologies Acoustic and Electroacoustic Spectrometer (DT1200, Bedford Hills, USA) equipped with titration unit, pH, conductivity and electroacoustic probes (Markiewicz et al. 2013). The CTAB amount was calculated according to the concentration of the CTAB solution before and after modification. The concentration of CTAB in the supernatant solution was estimated spectrophotometrically using methyl orange at 418 nm (Mishra & Panda 2005). Ash contents were determined following ASTM standard procedures (Wüst & Bustin 2001).

Batch mode sorption

Because of low water solubility, stock phenanthrene and pyrene solutions were made at high concentrations in methanol and then experimental solutions of the desired concentrations (concentrations of phenanthrene: 200, 400, 600, 800, 1000, 1200 μg·L−1; concentrations of pyrene: 40, 60, 80, 100, 120, 140 μg·L−1) were obtained by introducing different amounts of stock solutions into deionized water (pH 7.0) by glass syringe. Batch mode sorption was conducted in 100 mL screw cap vials with polytetrafluoroethylene sealer. For phenanthrene, 0.03 g of sorbent with 50 mL of solute-containing solution was added to the vials; for pyrene, 0.02 g of sorbent was used. The vials were shaken in a water bath (25 °C) oscillator for 24 h at 200 rpm. Then 2 mL of supernatant was collected to analyze the PAHs concentration. A control (blank) experiment confirmed that there was almost no adsorption on the equipment walls. The equilibrium sorption amount was calculated according to the equilibrium concentration. By fitting the equilibrium result to the partition equation and Freundlich equation, we analyzed the contributions of partition and adsorption to the total sorption of phenanthrene and pyrene.

Data analysis

Partition model

The Partition model is described by the following equation (Chen et al. 2008; Pavlović et al. 2014): 
formula
2
where Kd is the sorption capacity coefficient (L·g−1) (the slope); qe is the amount adsorbed at equilibrium (μg·g−1); ce is the equilibrium concentration of phenanthrene and pyrene (μg·L−1); b is sorption constant (the intercept). With the Partition model, the sorption equilibrium can be described by a partition coefficient (Kd) between the aqueous phase and the sorbent.

Freundlich model

The Freundlich model is an empirical isotherm model used for sorption onto heterogeneous surfaces or surfaces which support sites of varying affinities (Chen et al. 2011; Pavlović et al. 2014). The Freundlich isotherm model is described by the following equation: 
formula
3
 
formula
4
where Kf is the sorption capacity coefficient (L·μg−1); qe is the amount adsorbed at equilibrium (μg·g−1); n is the Freundlich exponent that describes the nonlinearity degree of sorption.

The contributions of partition and adsorption

In order to simplify the modeling and prediction of PAHs sorption to peat, the actual amount of conceptual sorption from water is comprised of two types of sorption: partition (QP) and adsorption (QA). Combined adsorption and partition models have been used to assess the contribution of peat to the PAHs sorption. They are related to the partition (linear) and adsorption (nonlinear) of PAHs in peat, respectively. The sorption amount is described by the following equation: 
formula
5
Sorption isotherms are described by the following equation: 
formula
6
The contributions of partition and adsorption are described by the following equations: 
formula
7
 
formula
8
where QT is the total amount adsorbed at equilibrium (μg·g−1); QP is the amount adsorbed at equilibrium by partition (μg·g−1); QA is the amount adsorbed at equilibrium by adsorption (μg·g−1); Kd is the sorption capacity coefficient (L·μg−1).

RESULTS AND DISCUSSION

Sorption isotherm

The Partition model and Freundlich model were used to quantify the sorption of phenanthrene and pyrene on to PT and MPT. The partition coefficient and model parameters of phenanthrene and pyrene by PT and MPT are presented in Table 2. The slope is Kd and the intercept is b. The coefficient of determination of partition isotherm fit is greater than 0.98, indicating that the sorption process fitted the Partition model well (as shown in Figure 1(a) and 1(b)). The sorption amount of PAHs on PT and MPT increased with increasing concentrations of PAHs. At a specified initial concentration, sorption amounts of PAHs on MPT is larger than that of PT, and the Kd value for MPT is larger than that for PT, which indicates that the sorption capacity of peat was notably improved through modification by CTAB. The attachment of CTAB to peat increased the aliphatic carbon content in the peat matrix and enhanced the hydrophobicity of the peat surface, which resulted in a greater sorption capacity for phenanthrene and pyrene (Qu et al. 2008). For a specified peat, the Kd value of pyrene sorption is larger than that of phenanthrene, which indicates that peat has a higher affinity to pyrene.

Table 2

Partition model and Freundlich model parameters of sorption isotherms of phenanthrene (PHN) and pyrene (PYR) on PT and MPT

   Partition model Freundlich model 
PAHs Sorbent T (°C) Kd (L/g) R2 Kf 1/n R2 
PHN PT 25 °C 3.388 0.984 9.393 0.8475 0.997 
MPT 19.342 0.982 59.665 0.7819 0.995 
PYR PT 25 °C 37.64 0.984 60.828 0.834 0.992 
MPT 146.67 0.987 185.22 0.8423 0.997 
   Partition model Freundlich model 
PAHs Sorbent T (°C) Kd (L/g) R2 Kf 1/n R2 
PHN PT 25 °C 3.388 0.984 9.393 0.8475 0.997 
MPT 19.342 0.982 59.665 0.7819 0.995 
PYR PT 25 °C 37.64 0.984 60.828 0.834 0.992 
MPT 146.67 0.987 185.22 0.8423 0.997 
Figure 1

Partition model fits to sorption isotherms of PHN (a) and PYR (b) on PT and MPT (pH = 7, 298K).

Figure 1

Partition model fits to sorption isotherms of PHN (a) and PYR (b) on PT and MPT (pH = 7, 298K).

The Freundlich sorption isotherms are given in Figure 2. Sorption coefficients and Freundlich model parameters of phenanthrene and pyrene by PT and MPT are presented in Table 2. The coefficient of determination (R2) for the Freundlich model were higher than 0.99. The sorption process was also perfectly described by the Freundlich isotherm model.

Figure 2

Freundlich model fits to sorption isotherms of PHN and PYR on PT and MPT (pH = 7, 298K).

Figure 2

Freundlich model fits to sorption isotherms of PHN and PYR on PT and MPT (pH = 7, 298K).

As shown in Table 2, the Kf values of phenanthrene and pyrene on MPT were higher than on PT, which also indicates that the sorption capacity of peat was notably improved through modification by CTAB. The Kf value of pyrene sorption is larger than that of phenanthrene for both peats, which indicates that peat has a higher sorption ability to pyrene, and this conclusion is consistent with that obtained from the Partition model, which can be explained by the following reason: the organic compounds, which have higher KOC values (organic carbon-normalized sorption coefficient), can be adsorbed much more easily by sorbents with same nature. Furthermore, it was found that the KOW value (octanol–water partition coefficient of PAH) was positive relative to their KOC value, the order of KOW was pyrene > phenanthrene, which was in accordance with the order of the KOC values (pyrene > phenanthrene), so pyrene can be adsorbed by peat more easily (Tang et al. 2010). The hydrophobic property of peat was enhanced due to the effect of CTAB. As a result, the Kf value of modified peat was enhanced, and the pyrene, which has a higher KOC value, was adsorbed much more easily.

The relationship between partition and adsorption

The Partition model and Freundlich model have been successfully used to predict the sorption of PAHs to peat. However, the observation of nonlinear sorption isotherms and competitive sorption behaviour of organic chemicals in Figure 1 questioned the hypothesis that peat is a homogeneous partition phase. Depending on the peat matrix and degree of alteration, peat can be described as a continuum of forms ranging from hard condensed carbon to soft biopolymer (Chen et al. 2008). Among the different forms of peat, the sorption mechanisms differ depending on the nature of the materials. Meng et al. (2006) investigated the contributions of adsorption and partition to the sorption of PAHs by study of the sorption of PAHs on particles from the Yellow River. They found adsorption is predominant in the sorption of Benzo(a)pyren, and the contribution of adsorption has a decreasing trend with the increase of the particle content. In this study, both partition and adsorption contributed to the sorption of PAHs onto PT and MPT. The peat matrix is expected to behave as an adsorbent and the grafted surfactant (CTAB) as a partition (absorption) phase; adsorption is typically nonlinear, whereas partition is essentially linear. The contributions of partition and adsorption to the total sorption of phenanthrene and pyrene by PT and MPT are shown in Table 3. The relationships between the initial concentrations of phenanthrene and pyrene and the contribution rate of partition and adsorption to the total sorption of phenanthrene and pyrene by PT and MPT are shown in Figure 3(a) and 3(b). Plots were drawn as a function of concentration to explore the relation between QP/QA and the initial concentration, where c0 represents the initial concentration of PAHs; QP/QA represents the contribution rate of partition and adsorption. When phenanthrene was adsorbed, Qp/QA ranged from 2.5–6.3 for PT and 2.2–7.6 for MPT. When it comes to pyrene, the rates changed to 1.9–5.9 and 2.3–8.9, respectively. Obviously, Qp/QA is always higher than 1, which indicates that partition contributed much more in the sorption process. Qp/QA values increased with increasing concentrations of PAHs, indicating that the contribution of partition became bigger with increasing concentrations of PAHs. At the same time, Qp/QA for pyrene was higher than that for phenanthrene. This may be attributed to the higher KOW of pyrene which allows pyrene to access the hydrophobic surfaces more easily through partition (Tang et al. 2010).

Table 3

Functions of partition sorption amount (QP) and adsorption amount (QA) with concentrations of phenanthrene (PHN) and pyrene (PYR) on PT and MPT

PAHs Sorbent Total sorption (QT) (μg/g) Partition sorption (QP) (μg/g) Adsorption (QA) (μg/g) 
PHN PT y = 9.393Ce0.8475 y = 3.388Ce y = 9.393Ce0.8475 − 3.388Ce 
MPT y = 59.665Ce0.7819 y = 19.342Ce y = 59.665Ce0.7819 − 19.342Ce 
PYR PT y = 60.828Ce0.834 y = 37.64Ce y = 60.828Ce0.834 − 37.64Ce 
MPT y = 185.22Ce0.8423 y = 146.67Ce y = 185.22Ce0.8423 − 146.67Ce 
PAHs Sorbent Total sorption (QT) (μg/g) Partition sorption (QP) (μg/g) Adsorption (QA) (μg/g) 
PHN PT y = 9.393Ce0.8475 y = 3.388Ce y = 9.393Ce0.8475 − 3.388Ce 
MPT y = 59.665Ce0.7819 y = 19.342Ce y = 59.665Ce0.7819 − 19.342Ce 
PYR PT y = 60.828Ce0.834 y = 37.64Ce y = 60.828Ce0.834 − 37.64Ce 
MPT y = 185.22Ce0.8423 y = 146.67Ce y = 185.22Ce0.8423 − 146.67Ce 
Figure 3

Effect of initial concentrations of PHN (a) and PYR (b) on functions of partition sorption amount (QP) and adsorption amount (QA) (pH = 7, 298K).

Figure 3

Effect of initial concentrations of PHN (a) and PYR (b) on functions of partition sorption amount (QP) and adsorption amount (QA) (pH = 7, 298K).

So partition plays a major role in the sorption process of pyrene. The more hydrophobic the pollutant, the higher the sorption capacity obtained and this is comparable to the results obtained by Zhang et al. using clay mineral (Zhang et al. 2010). Compared to the sorption to modified peat, the partition of hydrophobic PAHs to the polar surface of clay mineral is less due to the competition of water molecules.

CONCLUSION

In this study, we investigated the sorption of phenanthrene and pyrene (as model PAHs) to peat from water. The Partition model and Freundlich model were applied to fit the sorption curve of phenanthrene and pyrene on peat. Both models can describe the sorption isotherms very well. The sorption capacity for MPT has been improved due to the synergistic effect between partition of the interlayer organic phase and surface adsorption. Because pyrene has a higher KOC value, peat has a higher sorption ability for pyrene than that for phenanthrene. Partition and adsorption effects were quantified via an isotherm separation method. At any PAHs concentration, the partition function plays a leading role in the sorption process. The Qp/QA value increases gradually and the contribution of partition became higher with the increase of PAHs concentrations. The sorption abilities of peat follow the order: pyrene > phenanthrene. The Qp/QA value of pyrene is higher than that of phenanthrene on the same material, indicating that partition plays a major role in the sorption of pyrene on peat. These observations are crucial for modifying biosorbents and illustrating the sorption behaviour of PAHs.

ACKNOWLEDGEMENTS

The authors would like to express their sincere gratitude to the National Natural Science Foundation of China (51208201) and Shanghai Educational Development Foundation (11CG32) for their financial support of this study.

REFERENCES

REFERENCES
Carmo
A. M.
Hundal
L. S.
Thompson
M. L.
2000
Sorption of hydrophobic organic compounds by soil materials: Application of unit equivalent Freundlich coefficients
.
Environmental Science & Technology
34
(
20
),
4363
4369
.
Kong
H.
He
J.
Wu
H.
Wu
H.
Gao
Y.
2012
Phenanthrene removal from aqueous solution on sesame stalk-based carbon
.
CLEAN–Soil, Air, Water
40
(
7
),
752
759
.
Kopinke
F.-D.
Georgi
A.
Mackenzie
K.
2001
Sorption of pyrene to dissolved humic substances and related model polymers. 1. Structure-property correlation
.
Environmental Science & Technology
35
(
12
),
2536
2542
.
Lindgren
J. F.
Hassellöv
I.-M.
Dahllöf
I.
2014
PAH Effects on meio-and microbial benthic communities strongly depend on bioavailability
.
Aquatic Toxicology
146
,
230
238
.
Meng
L.
Xia
X.
Yu
H.
2006
Adsorption and partition of PAHs on particles of the Yellow River
.
Environmental Science
27
(
5
),
892
897
.
Mishra
S. K.
Panda
D.
2005
Studies on the adsorption of Brij-35 and CTAB at the coal–water interface
.
Journal of Colloid and Interface Science
283
(
2
),
294
299
.
Pavlović
D. M.
Ćurković
L.
Blažek
D.
Župan
J.
2014
The sorption of sulfamethazine on soil samples: Isotherms and error analysis
.
Science of the Total Environment
,
497
,
543
552
.
Pei
Z.-G.
Shan
X.-Q.
Liu
T.
Xie
Y.-N.
Wen
B.
Zhang
S.
Khan
S. U.
2007
Effect of lead on the sorption of 2,4,6-trichlorophenol on soil and peat
.
Environmental Pollution
147
(
3
),
764
770
.
Tang
X.
Zhou
Y.
Xu
Y.
Zhao
Q.
Zhou
X.
Lu
J.
2010
Sorption of polycyclic aromatic hydrocarbons from aqueous solution by hexadecyltrimethylammonium bromide modified fibric peat
.
Journal of Chemical Technology and Biotechnology
85
(
8
),
1084
1091
.
Wüst
R. A. J.
Bustin
R. M.
2001
Low-ash peat deposits from a dendritic, intermontane basin in the tropics: a new model for good quality coals
.
International Journal of Coal Geology
46
(
2–4
),
179
206
.
Zhang
J.
Séquaris
J.-M.
Narres
H.-D.
Vereecken
H.
Klumpp
E.
2010
Effect of organic carbon and mineral surface on the pyrene sorption and distribution in Yangtze River sediments
.
Chemosphere
80
(
11
),
1321
1327
.
Zhou
Z.
Wu
W.
Li
Y.
2008
Sorption and desorption behaviours of three PAHs by charcoals
.
Journal of Agro-Environment Science
27
(
2
),
813
819
.