Authors Accepted Manuscript

Carbon nanotube sponges as a solid-phase

extraction adsorbent for the enrichment and
determination of polychlorinated biphenyls at trace
levels in environmental water samples
Lei Wang, Xia Wang, Jia-Bin Zhou, Ru-Song Zhao
www.elsevier.com/locate/talanta

PII:
DOI:
Reference:

S0039-9140(16)30497-0
http://dx.doi.org/10.1016/j.talanta.2016.07.005
TAL16702

To appear in: Talanta

Received date: 5 April 2016
Revised date: 21 June 2016
Accepted date: 2 July 2016
Cite this article as: Lei Wang, Xia Wang, Jia-Bin Zhou and Ru-Song Zhao,
Carbon nanotube sponges as a solid-phase extraction adsorbent for the
enrichment and determination of polychlorinated biphenyls at trace levels in
environmental
water
samples, Talanta,
http://dx.doi.org/10.1016/j.talanta.2016.07.005
This is a PDF file of an unedited manuscript that has been accepted for
publication. As a service to our customers we are providing this early version of
the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting galley proof before it is published in its final citable form.
Please note that during the production process errors may be discovered which
could affect the content, and all legal disclaimers that apply to the journal pertain.

Carbon nanotube sponges as a solid-phase extraction adsorbent for

the enrichment and determination of polychlorinated biphenyls at
trace levels in environmental water samples
Lei Wang a,b, Xia Wang a,*, Jia-Bin Zhou b, Ru-Song Zhao a,**
a

Key Laboratory for Applied Technology of Sophisticated Analytical Instruments of

Shandong Province, Analysis and Test Center, Shandong Academy of Sciences, Jinan
250014, China

School of Resources and Environmental Engineering, Wuhan University of Technology,

Wuhan 430070, China

xia-w21cn0@163.com (Xia Wang)

zhaors1976@126.com (Ru-Song Zhao)
*

Corresponding author at: Shandong Academy of Sciences, Jinan, China. Tel.:

+86-531-82605340; Fax: +86-531-82964889

Abstract
Carbon nanotube (CNT) sponges has recently attracted considerable attention in numerous
fields because of its excellent properties, such as high porosity, light weight, and large
surface area. The potential of CNT sponges for the solid-phase extraction (SPE) of organic
pollutants at trace levels was investigated in this study for the first time. Seven
polychlorinated

biphenyls

(PCBs)

were

selected

as

analytes,

and

gas

chromatographytandem mass spectrometry was employed for the detection. We optimized

important parameters that may influence the efficiency of SPE, including the kind and
volume of elution solvent, sample pH, and sample flow rate and volume. Under optimized
conditions, low limits of detection (0.721.98 ng L1), wide range of linearity (101000 ng
1

L1) and good repeatability (2.696.85%, n = 5) were obtained. CNT sponges exhibited
higher extraction performance than other adsorbent materials under the optimized conditions.
Real environmental water samples were analyzed, and satisfactory recoveries (81.1119.1%)
were achieved. All these results demonstrated that CNT sponges are suitable SPE material
for the enrichment and sensitive determination of PCBs at trace levels.

Keywords: Carbon nanotube sponges; Solid-phase extraction; Polychlorinated biphenyls;

Gas chromatography-tandem mass spectrometry.

1. Introduction
Polychlorinated biphenyls (PCBs) had been widely used in various industries. PCBs are
relatively stable, i.e., extremely difficult to be oxidized, reduced, or otherwise chemically
transformed naturally [1,2]. These compounds can bio-magnify through the food chain and
then pose a threat on human health [3,4]. Although PCB production has been banned since
the late 1970s, humans are still exposed to these pollutants because of they have been widely
distributed and are highly stable [5]. Thus, the determination of PCBs in the environment is
urgently required for environmental impact and risk assessment [6]. Direct analysis of PCBs
by modern chromatographic techniques, such as gas chromatographymass spectrometry
(GC-MS) [7] or GCelectron capture detection [8], is usually difficult because of the low
concentrations or the complexity of sample matrixes. Therefore, suitable sample pretreatment
steps are necessary prior to chromatographic determination [9].

To date, various pretreatment techniques have been developed for the extraction and
enrichment of PCBs at trace levels in the environment. These techniques include
liquidliquid extraction (LLE) [10], ultrasonic extraction (USE) [11], microwave-assisted
extraction (MAE) [12], liquid-phase microextraction (LPME) [13,14], dispersive
liquidliquid microextraction (DLLME) [15], solid-phase microextraction (SPME) [16], and
solid-phase extraction (SPE) [17]. Among these sample pretreatment techniques, LLE is the
oldest sample preparation technique, which often requires large amounts of toxic solvent and
time-consuming procedures. Compared with LLE, SPE offers several evident advantages,
such as low consumption of solvents, high enrichment factors, simplicity, and ease of
operation [1820]. The selection of suitable adsorbents for SPE is crucial in obtaining high
enrichment efficiency of targeted analytes. Multiwalled carbon nanotubes (MWCNTs)
[18,19], bamboo charcoal [20], magnetic nanoporous carbon [21], C18-bonded silica [22],
expanded graphite [23], and hydrofluoric acid-etched sand particles [24] have been
investigated as SPE adsorbents.
Carbon nanotubes (CNTs) form unique nanostructures, thus enabling the use of the
material in various applications, such as supercapacitors [25,26], artificial muscles [27], and
superhydrophobic coatings [28]. A sponge-like material named CNT sponges has recently
attracted considerable attention in numerous fields because of its excellent properties, such as
high porosity (>98%), large surface area, great sorption capacity and rate, and strong
oleophilic and hydrophobic properties [2931]. Furthermore, compared with other materials,
CNT sponges show advantages, such as high flexibility of structure and thermal stability [32].
CNT sponges are promising candidates for environmental applications, such as adsorption,
3

filtration, and separation, owing to their excellent properties, such as light weight, high
porosity, and large surface area. Gui et al. used highly porous CNT sponges as an efficient
adsorbent material for oil and organic reagents [33], in which the adsorption capacities were
higher than 100 g g1 [33]. Li et al. reported that porous sponge-like materials can scavenge
and absorb large-area oil spill from water surface [32]. Then, the applications have been
extended to the removal of organic dyes and nanoparticles from water by using the sponges
as filtration membranes [32].
All applications of CNT sponges mentioned above mainly focus on the removal of
nanoparticles and environmental pollutants from water samples and not on the enrichment
and analysis of pollutants at trace levels in environmental water samples. To the best of our
knowledge, the enrichment and analysis of pollutants in aqueous samples by using CNT
sponges as SPE adsorbent has not been reported. This study aims to investigate the feasibility
of using CNT sponges as an adsorbent for the enrichment and analysis of organic pollutants
at trace levels in water samples. Seven PCBs were selected as analytes, and GCMS/MS was
used for detection. Surface morphology of CNT sponges was characterized using scanning
electron microscopy (SEM) and transmission electron microscopy (TEM). The important
parameters affecting the performance of SPE have been investigated and optimized in detail.
Moreover, the possible extraction mechanism of CNT sponges was discussed. Finally, the
developed method was used for the analysis of PCBs in real environmental water samples.

2. Experimental
2.1 Reagents and materials
CNT sponges, MWCNTs, and single-walled CNTs were purchased from XFNANO

Materials Tech Co., Ltd (Nanjing, China). Activated carbon was purchased from
Sigma-Aldrich (USA), and the Oasis HLB 6 cc (200 mg) extraction cartridge was sourced
from Waters Technology Co., Ltd. (Milford, USA). Methanol (HPLC grade), n-hexane
(HPLC grade), and dichloromethane (pesticide grade) were supplied by Oceanpak Alexative
Co., Ltd. (Sweden). Acetone (HPLC grade) was obtained from Duksan Pure Chemicals Co.,
Ltd. (Korea). Other reagents were of analytical grade unless otherwise stated. A soil standard
substance (GBW08307) was obtained from Research Center for Eco-Environmental Sciences,
Chinese Academy of Sciences (Beijing, China). PCBs standard solution with 10 g mL1
concentration of each component was purchased from AccuStandard, Inc. (New Haven,
USA).

The

solution

2,2,5,5-tetrachlorobiphenyl

contained
(PCB-52),

2,4,4-trichlorobiphenyl

2,2,4,5,5-pentachlorobiphenyl

(PCB-28),
(PCB-101),

2,3,4,4,5-pentachlorobiphenyl (PCB-118), 2,2,3,4,4,5-hexachlorobiphenyl (PCB-138),

2,2,4,4,5,5-hexachlorobiphenyl

(PCB-153)

and

2,2,3,4,4,5,5-heptachlorobiphenyl

(PCB-180). A stock solution (100 g L1) was prepared in methanol. Fresh working solutions
were prepared daily by appropriate dilution of the stock solution using double-distilled water.
All solutions were stored at 4 C in the dark prior to use.
2.2 Instrumentation
In this work, a gas chromatography system (436 GC, Bruker, USA) coupled with a
triple quadrupole mass spectrometer (SCION TQ, Bruker, USA) was used in multiple
reaction monitoring (MRM) mode. GC separation was conducted with a fused silica
DB-5MS capillary column (30 m 0.32 mm 0.25 m) (Agilent Technologies, USA). Oven
temperature was held at 150 C for 1 min, programmed at 10 C min1 to 290 C, and was

held for 4 min. The GC carrier gas was helium, and the flow rate was 1 mL min1. The
injection port temperature and transfer line temperature were 280 C and 290 C, respectively.
Other GC-MS parameters, such as retention time, precursor ions, product ions, and collision
energy, are presented in Table 1S. In the experiment, external standard calibration was
employed. Six different concentrations (10, 20, 50, 100, 500, and 1000 ng L1) in methanol
were used as calibration standards.
The micrographs of CNT sponges were recorded by SEM (Carl Zeiss MicroImaging Co.,
Ltd., Germany) and high-resolution TEM (HRTEM, JEM-2100, JEOL, Japan). An
accelerated surface area and porosimetry system (ASAP2020 HD88, Micromeritics, USA)
was used for the determination of specific surface area and pore size distribution.
An adsorbent material packed cartridge (CNT sponges, MWCNTs, SWCNTs, activated
carbon, or Oasis HLB) was prepared by modifying a Waters Oasis SPE C18 cartridge (60 mg,
3 mL, polypropylene) [1820]. An 21.8 mg portion of adsorbent material (CNT sponges,
MWCNTs, SWCNTs, activated carbon, or Oasis HLB) was packed into the cartridge after
evacuating the C18 packing of the cartridge. The 1.5 mm polypropylene upper and lower
partition remained at both ends of the cartridge to keep the adsorbent packing in place.
Afterwards, the outlet tip of the cartridge was connected to a model SHB-3 (III) vacuum
pump (Great Wall Scientific Industrial and Trade Co., Ltd., Zhengzhou, China), and the inlet
end was connected to a polytetrafluoroethylene suction tube. The other end of the PTFE tube
was inserted into working solutions or water samples. The SPE system was washed with
ample methanol prior to use.
2.3 SPE procedure
6

An SPE cartridge packed with CNT sponges was washed with 6 mL of methanol and 6
mL of ultrapure water. Then, working solutions or water samples were passed through the
SPE cartridge at a constant flow rate. Subsequently, the SPE cartridge was dried by negative
pressure for 5 min. After this step, target compounds were eluted by dichloromethane in an
optimum volume. The eluent was blown with a mild stream of nitrogen gas to dryness at
room temperature. The residue was dissolved in 100 L of methanol, and 1 L of the solution
was injected for analysis.
2.4 Water samples
Four real environmental water samples, including ground water, pond water, tap water,
and mine water, were used to verify the proposed method. Ground water was collected from
Wangzhuangxincun (Taian, China). Pond water was taken from Shandong Academy of
Sciences (Jinan, China). Tap water was collected from our laboratory (Jinan, China). Mine
water was obtained from a gold mine in Laiwu (Shandong, China). The water samples were
analyzed immediately after filtering through 0.45 m micropore membranes. If water
samples can not be analyzed immediately, they must be stored in brown glass at 4 C and
extracted within 7 days. The extract should be analyzed within 40 days.
2.5 Method comparison
Method validation was performed through detecting the concentration of PCBs in a soil
standard substance (GBW08307). In short, soil standard substance (0.3 g) was extracted
ultrasonically with 6.0 mL acetonitrile for 3 h and the extracted solution was collected by
centrifugation [34]. Then the supernatant was blown with a mild stream of nitrogen gas to 1
mL. 1 mL of soil sample solution was diluted into 100 mL distilled water and loaded for SPE
procedure. Then, the extract was analyzed by GCMS/MS.
7

3. Results and Discussion

3.1 Characterization of CNT sponges
Fig. 1a shows macroscopic, monolithic CNT sponges (1 cm 1 cm 1 cm, 10 mg cm3)
synthesized by chemical vapor deposition (CVD). The SEM image in Fig. 1b shows the
presence of CNT sponges as an extensive, interconnected three-dimensional (3D) framework.
The sponges are composed of hollow CNTs with inner diameters of 10 to 20 nm, external
diameters of 30 to 50 nm (Fig. 1c), and lengths of tens to hundreds of micrometers (Fig. 1b).
These dimensions indicate multiple layers of CNTs existing in the cm-thick sponge [32]. In
addition, the porosity of CNT sponges was approximately 99%. The BET surface area of
CNT sponges was determined by N2 adsorptiondesorption isotherms as 99.56 m2 g1 (Fig.
1d). The large surface areas of CNT sponges mainly originate from their mesoporous and
reticular structures. The average pore size and pore volume were 56.51 nm and 0.14 cm3 g1
(Fig. 1d), respectively. These excellent properties, such as light weight, high porosity, and
large surface area, make CNT sponges a promising adsorbent candidate for the SPE of
environmental pollutants in water samples.
3.2 Optimization of experimental parameters
To assess the enrichment potential of CNT sponges as SPE adsorbent, we selected seven
PCBs, including PCB-28, PCB-52, PCB-101, PCB-123, PCB-138, PCB-153, and PCB-180,
as model pollutants. GCMS/MS was used for the detection. Important parameters that affect
SPE performance, such as the kind and volume of elution solvent, sample pH, and sample
flow rate and volume, have been investigated and optimized in detail to achieve high
sensitivity and good repeatability.

In general, the eluent performs an important function in the desorption efficiency of

target analytes in the SPE process [20]. Different eluents exhibit varying elution power. In
this experiment, three kinds of eluents, including acetone, dichloromethane, and hexane,
were investigated. From the experimental results shown in Fig. 2a, we can evidently see that,
with the other experimental conditions kept constant, the seven PCBs could be eluted by
dichloromethane more easily. Therefore, dichloromethane was selected as eluent in the
subsequent experiments.
To achieve complete elution of target analytes from the SPE cartridge, we should
optimize the eluent volume [24]. In this work, a series of experiments was designed on the
basis of changing the volume of dichloromethane from 2 mL to 10 mL. When the experiment
was finished, another 10 mL of dichloromethane was employed for passing through the CNT
sponges cartridge prior to the next SPE procedure to remove residues that may be adsorbed
on the cartridge. From the experimental results shown in Fig. 2b, we can see that the peak
area of PCBs increased with the increase in dicholoromethane volume between 2 and 8 mL.
In addition, almost no change in peak area was observed when the volume of
dichloromethane increased up to 10 mL. Therefore, 8 mL of dichloromethane was used in
further experiments.
Sample pH is an important parameter in the SPE process because the value determines
the existing state of analytes and possibly affects extraction efficiency [20]. In this
experiment, the influence of sample pH on the extraction efficiency of PCBs has been
evaluated in the range of 3 to 11. The experimental results demonstrate that virtually no
variations in peak area occurred in the pH range from 3 to 11. This finding may be due to
9

stable properties of PCBs; thus, ionization, hydrolysis, or degradation did not occur [1, 2]. So
as to save time, sample pH value adjustment was unnecessary in the following experiments.
In the SPE procedure, the flow rate of the sample not only affects the enrichment
efficiency of analytes but also controls extraction time [19]. In this work, the sample flow
rate was investigated and optimized in the range between 2 and 9 mL min1. The
experimental results in Fig. 2c show that the peak area was almost unchanged when the flow
rate was 23 mL min1 and began to decrease when the flow rate exceeded 3 mL min1.
Therefore, to obtain high extraction efficiency and save time, we selected 3 mL min1 flow
rate in the subsequent experiments.
Sample volume can affect the extraction efficiency of target analytes. In this paper, the
volume was investigated in the range of 25- 600 mL. The experimental data exhibited in Fig.
2d, and it was found that almost no obvious variations of the recoveries occurred when the
sample volume increased from 25 to 400 mL for the 7 PCBs. The recoveries of these PCBs
decreased with the increase of sample volume from 400 to 600 mL, which showed that the
breakthrough volume was 400 mL. In other words, CNT sponges exhibit good adsorption for
PCBs in the range of 25 to 400 mL. A smaller sample may save time, but sensitivity should
be higher with a larger sample volume. Based on comprehensive consideration, 100 mL was
selected as the optimal volume.
3.3 Comparison of CNT sponges with other adsorbents
CNT sponges were compared with other adsorbents, such as MWCNTs, SWCNTs,
activated carbon and Oasis HLB. These adsorbents were used to extract 600 ng L-1 PCBs
from water samples under their own optimal conditions. The extraction efficiencies of three

10

carbon-based nano-materials (CNT sponges, SWCNTs and MWCNTs) are higher than those
of activated carbon and Oasis HLB. The specific surface area of CNT sponges was lower
than that of the other adsorbents, however, CNT sponges achieved the highest extraction
efficiencies among the five adsorbents. The excellent SPE performance of CNT sponges can
be explained as follows: first, like other carbon-based nano-materials, hexagonal arrays of
carbon atoms in graphite sheets of carbon nanotubes surface of the material have a strong
interaction with PCBs [18]. Second, compared with SWCNTs and MWCNTs with 1D
framework, CNT sponges with the special 3D framework show great advantages in the
aspects of the extraction and interception of organic pollutants. The 3D interconnected
porous structure formed by entangled nanotubes can be more effective in trapping organic
pollutants molecules by physisorption from water samples [32]. In addition, the specific
surface area can only reflect the ability of an adsorbent to adsorb nitrogen, and does not
really reflect its adsorption ability to organic pollutants.
3.4 Analytical performance
Analytical data under the above optimal conditions are listed in Table 1. The linear
range was 10 to 1000 ng L1 for the seven PCBs, with correlation coefficients (R) ranging
from 0.9936 to 0.9988. The limits of detection (LODs, S/N = 3) and limits of quantification
(LOQs, S/N = 10) were 0.72-1.98 ng L1 and 2.42 - 6.60 ng L1 in pure water, respectively.
Intra-day precisions (relative standard deviations, RSDs) were in the range of 2.69 to 6.85%
for the seven PCBs (60 ng L1, n = 5). As can be seen in Table 2, CNT sponges achieved
similar precisions (2.696.85%) and lower LODs (0.721.98 ng L1) than other materials
(Oasis HLB, MWCNTs, SWCNTs, and activated carbon) (1.510.2%, 3100 ng L1) under

11

the same experimental conditions. Evidently, the performance comparison with other
adsorbents demonstrated the extraction advantages of CNT sponges.
From the above-mentioned results, we evidently see that, as adsorbent, CNT sponges
exhibit excellent performance for PCBs extraction. This excellent performance is attributed
to several possible explanations. First, hydrophobic effect performs an important function in
the SPE process. Pristine sponges are hydrophobic, with a contact angle of about 156 for
water droplets, and the high hydrophobicity of CNT sponges is beneficial to the
hydrophobichydrophobic interaction between hydrophobic CNT sponges and hydrophobic
PCBs [3539] (Table 3). Second, the adsorption of PCBs to CNT sponges also results from
the hollow tube structure, electronic properties, and large surface area of CNTs. The
hexagonal arrays of carbon atoms in graphene sheets of the CNT surface show a strong
interaction with the two benzene rings of PCBs (Table 3) [40]. Third, given that the
molecular diameters of PCBs (0.88261.0639 nm) (Table 3) [41] are considerably smaller
than the inner diameter of hollow CNTs (1020 nm), the extraction efficiency is likely based
on the molecular sieving effect [42].
3.5 Analysis of the PCBs in water samples
To demonstrate the applicability of CNT sponges as SPE adsorbent, four real
environmental water samples, including ground water, pond water, tap water, and mine water,
were analyzed by the proposed method. Blank experiments show that PCB-28 and PCB-52
were present in the mine water. In order to investigate the feasibility of the method, a soil
standard substance (GBW08307) was used. The analytical results obtained from our method
are very close to the given values (Table 4). We spiked the four water samples with three

12

different concentrations (12, 30, and 60 ng L1) and analyzed the samples using the proposed
method. Spiked recoveries shown in Table 4 were 81.1 to 119.1%. These finding are
satisfactory, as the values are in the range of 80120%. The above results indicate that the
developed method is highly sensitive, highly reliable, and useful in the analysis of
real-environmental water samples. Typical chromatograms of real-environment water
samples are shown in Fig. 4.
4. Conclusion
In this study, the feasibility of CNT sponges as adsorbent for SPE of PCBs at trace
levels in environmental water samples was investigated in detail. CNT sponges exhibited
higher extraction performance than other adsorbent materials under the same conditions. The
material exhibited low LODs, wide range of linearity, and good repeatability for PCBs under
optimized conditions. The method was validated using several real water samples, in which
satisfactory results were obtained. All these results demonstrate that CNT sponges are
suitable SPE materials for the enrichment and sensitive determination of PCBs at trace levels
in real-environment water samples.
Acknowledgements
This study was supported by the National Natural Science Foundation of China
(21477068), Natural Science Foundation of Shandong Province (ZR2015YL003), and Key
Research and Development Program of Shandong Province (2015GSF117011). The authors
thank Professor Anyuan Cao, Hongwei Zhu and Chih-Feng Wang for their good suggestion
in adsorption mechanism.
Appendix A. Supplementary data

13

Supplementary data associated with this article can be found, in the online version, at
the supplementary material.

References
[1] International Program on Chemical Safety (IPCS), Polychlorinated Biphenyls and
Terphenyls, Geneva: World Health Organization, 1993.
[2] B. G. Bennett, Exposure of man to environmental PCBsAn exposure commitment
assessment, Sci. Total. Environ. 29 (1983) 101-111.
[3] X.-H. Bi, S.-G. Chu, Q.-Y. Meng, X.-B. Xu, Movement and retention of polychlorinated
biphenyls in a paddy field of Wen Tai area in China, Agr. Ecosyst. Environ. 89 (2002)
241252.
[4] M. Aresta, P. Caramuscio, L. De Stefano, T. Pastore, Solid state dehalogenation of PCBs
in contaminated soil using NaBH4, Waste. Manage. 23 (2003) 315-319.
[5] Q.-L. Li, X. Wang, Y.-L. Liu, X.-F. Chen, M.-L. Wang, R.-S. Zhao, Feasibility of
metalorganic nanotubes [Cu3(3-O)(-OH)(triazolate)2]+-coated fibers for solid-phase
microextraction of polychlorinated biphenyls in water samples, J. Chromatogr. A 1374
(2014) 58-65.
[6] J. Hassan, Mojtaba Shamsipur, Extraction of ultra traces of polychlorinated biphenyls in
aqueous

samples

using

suspended

liquid-phase

microextraction

and

gas

chromatography-electron capture detection, Environ Monit Assess 185 (2013)

36373644.

14

[7] Q.-L. Li, L.-L. Wang, X. Wang, M.-L. Wang, R.-S. Zhao. Magnetic metal-organic
nanotubes: an adsorbent for magnetic solid-phase extraction of polychlorinated
biphenyls from environmental and biological samples, J. Chromatogr. A 1449 (2016)
3947.
[8] Q.-X. Zhou, Y.-R. Huang, G.-H. Xie, Investigation of the applicability of highly ordered
TiO2 nanotube array for enrichment and determination of polychlorinated biphenyls at
trace level in environmental water samples, J. Chromatogr. A 1237 (2012) 2429.
[9] X. Cao, J. Chen, X. Ye, F. Zhang, L. Shen, W. Mo, Ultrasound-assisted magnetic SPE
based on Fe3O4-grafted graphene for the determination of polychlorinated biphenyls in
environmental water samples, J. Sep. Sci.21-22 (2013) 3579-3585.
[10] S. Mori, Identification and determination of phthalate esters in river water by
high-performance liquid chromatography, J. Chromatogr. A 129 (1976) 53-60.
[11] L. L. Ma, S. G. Chu, X. B. Xu, Organic contamination in the greenhouse soils from
Beijing suburbs, China, J. Environ. Monit. 5 (2003) 786-790.
[12] M. A. Shannon, P. W. Bohn, M. Elimelech, J. G. Georgiadis, B. J. Marins, A. M.
Mayes, Science and technology for water purification in the coming decades, Nature
452 (2008) 301-310.
[13] R. Batlle, C. Nern, Application of single-drop microextraction to the determination of
dialkyl phthalate esters in food stimulants, J. Chromatogr. A 1045 (2004) 29-35.
[14] E. Psillakis, N. Kalogerakis, Hollow-fibre liquid-phase microextraction of phthalate
esters from water, J. Chromatogr. A 999 (2003) 145-153.

15

[15] F. Rezaei, A. Bidari, A. P. Birjandi, M. R. M. Hosseini, Y. Assadi, Development of a

dispersive liquid liquid microextraction method for the determination of
polychlorinated biphenyls in water, J. Hazard. Mater. 158 (2008) 621-627.
[16] Y.-L. Liu, X.-F. Chen, X. Wang, J.-B. Zhou, R.-S. Zhao, Sensitive determination of
polychlorinated biphenyls in environmental water samples by headspace solid-phase
microextraction with bamboo charcoal@iron oxide black fibers prior to gas
chromatography with tandem mass spectrometry, J. Sep. Sci. 37 (2014) 14961502.
[17] M.-Z. Fu, H.-Z. Xing, X.-F. Chen, F. Chen, C.-M. Lawrence Wu, R.-S. Zhao, C.-G.
Cheng, Ultrathin-shell boron nitride hollow spheres as sorbent for dispersive solid-phase
extraction of polychlorinated biphenyls from environmental water samples, J.
Chromatogr. A 1369 (2014) 181-185.
[18] Y. Cai, G. Jiang, J. Liu, Q. Zhou, Multiwalled Carbon Nanotubes as a Solid-Phase
Extraction Adsorbent for the Determination of Bisphenol A, 4-n-Nonylphenol, and
4-tert-Octylphenol, Anal. Chem. 75 (2003) 2517-2521
[19] Q.-X. Zhou, J.-P. Xiao, W.-D. Wang, Using multi-walled carbon nanotubes as solid
phase extraction adsorbents to determine dichlorodiphenyltrichloroethane and its
metabolites at trace level in water samples by high performance liquid chromatography
with UV detection, J. Chromatogr. A 1125 (2006) 152158.
[20] R.-S. Zhao, X. Wang, J.-P. Yuan, J.-M. Lin, Investigation of feasibility of bamboo
charcoal as solid-phase extraction adsorbent for the enrichment and determination of
four phthalate esters in environmental water samples, J. Chromatogr. A 1183 (2008)
15-20
16

[21] X. Liu, C. Wang, Q. Wu, Z. Wang, Metal-organic framework-templated synthesis of

magnetic nanoporous carbon as an efficient absorbent for enrichment of phenylurea
herbicides, Anal. Chim. Acta 870 (2015) 67-74
[22] M. Lyytikainen, J. V. K. Kukkonen, M. J. Lydy, Analysis of pesticides in water and
sediment under different storage conditions using gas chromatography, Arch. Environ.
Contam. Toxicol. 44 (2003) 437-444.
[23] C. Blasco, C. M. Lino, Y. Picao, A. Pena, G. Font, M. I. N. Silveira, Determination of
organochlorine pesticide residues in honey from the central zone of Portugal and the
Valencian community of Spain, J. Chromatogr. A 1049 (2004) 155-160.
[24] H.-Z. Xing, X.-F. Chen, X. Wang, M.-L. Wang, R.-S. Zhao, Feasibility of hydrofluoric
acid-etched sand particles for enrichment and determination of polychlorinated
biphenyls at trace level in environmental water samples, Anal. Bioanal. Chem. 406
(2014) 3787-3793
[25] D. N. Futaba, K. Hata, T. Yamada, T. Hiraoka, Y. Hayamizu, Y. Kakudate, O. Tanaike, H.
Hatori, M. Yumura, S. Iijima, Shape-engineerable and highly densely packed
single-walled carbon nanotubes and their application as super-capacitor electrodes, Nat.
Mater. 5 (2006) 987-994.
[26] R. N. Das, B. Liu, J. R. Reynolds, A. G. Rinzler, Engineered macroporosity in
single-wall carbon nanotube films, Nano Lett. 9 (2009) 677-683.
[27] J. E. Trancik, S. C. Barton, J. Hone, Transparent and Catalytic Carbon Nanotube Films,
Nano Lett. 8 (2008) 982-987.

17

[28] K. K. S. Lau, J. Bico, K. B. K. Teo, M. Chhowalla, G. A. J. Amaratunga, W. I. Milne, G.

H. Mckinley, K. K. Gleason, Superhydrophobic carbon nanotube forests, Nano Lett. 3
(2003) 1701-1705.
[29] M. S. Mauter, M. Elimelech, Environmental applications of carbon-based nanomaterials,
Environ. Sci. Technol. 42 (2008) 5843-5859.
[30] B. Pan, B. Xing, Environ. Adsorption mechanisms of organic chemicals on carbon
nanotubes, Sci. Technol. 42 (2008) 9005-9013.
[31] X.-C. Gui, J.-Q. Wei, K.-L. Wang, A.-Y. Cao, H.-W. Zhu, Y. Jia, Q.-K. Shu, and D.-H.
Wu, Carbon nanotube sponges, Adv. Mater. 22 (2010) 617-621.
[32] H.-B. Li, X.-C. Gui, L.-H. Zhang, S.-S. Wang, C.-Y. Ji, J.-Q. Wei, K.-L. Wang, H.-W.
Zhu, D.-H. Wu and A.-Y. Cao, Carbon nanotube sponge filters for trapping
nanoparticles and dye molecules from water, Chem. Commun. 46 (2010) 79667968.
[33] X.-C. Gui, H.-B. Li, K.-L. Wang, J.-Q. Wei, Y. Jia, Z. Li, L.-L. Fan, A.-Y. Cao, H.-W.
Zhu, D.-H. Wu, Recyclable carbon nanotube sponges for oil absorption, Acta Mater. 59
(2011) 47984804.
[34] J. Zhang, W.-P. Zhang, T. Bao, Z.-L. Chen, Polydopamine-based immobilization of
zeolitic imidazolate framework-8 for in-tube solid-phase microextraction, J.
Chromatogr. A 1388 (2015) 916.
[35] Y.-Y. Wu, C.-X. Yang, X.-P. Yan, Fabrication of metal-organic framework MIL-88B
films on stainless steel fibers for solid-phase microextraction of polychlorinated
biphenyls, J. Chromatogr. A 1334 (2014) 1-8.

18

[36] W. A. Bruggeman, J. V. D. Steen, O. Hutzinger, Reversed-phase thin-layer

chromatography of polynuclear aromatic hydrocarbons and chlorinated biphenyls:
Relationship with hydrophobicity as measured by aqueous solubility and octanol-water
partition coefficient, J. Chromatogr. A 238 (1982) 335-346.
[37] T. H. Risby, T. B. Hsu, S. S. Sehnert, P. Bhan, Physicochemical parameters of individual
hexachlorobiphenyl congeners, Environ. Sci. Technol. 24 (1990) 1680-1687.
[38] R. A. Rapaport, S. J. Eisenreich, Chromatographic determination of octanol-water
partition coefficients (Kow's) for 58 PCB polychlorinated biphenyl congeners, Environ.
Sci. Technol. 18 (1984) 163-170.
[39] A. C. D Kock, D. A. Lord, A simple procedure for determining octanol-water patition
coefficients using reverse phase high performance liquid chromatography (RPHPLC),
Chemosphere 16 (1987) 133-142.
[40] Y.-Q. Cai, G.-B. Jiang, J.-F. Liu, Q.-X. Zhou, Multiwalled carbon nanotubes as a
solid-phase extraction adsorbent for the determination of bisphenol A, 4-n-nonylphenol,
and 4-tert-octylphenol, Anal. Chem. 75 (2003) 2517-2521.
[41] Y.-X E, T. Wang, L. Li, D.-Z. ChenA theoretical study of the molecular structure of
polychlorinated biphenyls, J. Beijing Univers. Chem. Tech. 33 (2006) 81-84.
[42] L.-Q. Yu, X.-P. Yan, Covalent bonding of zeolitic imidazolate framework-90 to
functionalized silica fibers for solid-phase microextraction, Chem. Commun. 21 (2013)
2142-2144.

19

Fig. 1 (a) Photo of the CNT sponges; (b) SEM image of the CNT sponges; (c) TEM image of
the CNT sponges; (d) N2 adsorption-desorption isotherms of the CNT sponges.

Fig. 2 Optimization of the experimental parameters. (a) Effect of the kind of eluent.
Conditions: spiked concentrations, 60 ng L-1; sample volume, 100 mL; eluent volume, 8 mL;
sample pH, 7; flow rate, 5.0 mL min1. (b) Effect of eluent volume. Conditions: spiked
concentrations, 60 ng L-1; eluant, dichloromethane; sample volume, 100 mL; sample pH, 7;
flow rate, 5.0 mL min1. (c) Effect of sample flow rate. Conditions: spiked concentrations, 60
ng L-1; sample volume, 100 mL; eluant, dichloromethane; eluent volume, 8 mL; sample pH,
7. (d) Effect of sample volume. Conditions: spiked concentrations, 60 ng L-1; eluant,
dichloromethane; eluent volume, 8 mL; sample pH, 7; flow rate, 3.0 mL min1.

Fig. 3 Comparison of CNT sponges with other adsorbent materials. Conditions: spiked
concentrations, 600 ng L-1; sample volume, 100 mL; eluant volume, 8 mL; flow rate, 3.0 mL
min-1; sample pH, no adjustment. The eluant of Oasis HLB, SWCNTs, activated carbon and
MWCNTs cartridge: n-hexane; The eluant of CNT sponges cartridge:

dichloromethane.

Fig. 4 Typical chromatograms of real environment water samples. Ground water (a), ground
water spiked at 12 ng L-1 (b), 30 ng L-1 (c) and 60 ng L-1 (d).
Table 1 Analytical parameters of the proposed method
Analytes Linear range (ng L-1)

PCB-28

10-1000

LOD (ng L-1)

0.9953

0.95

20

LOQ (ng L-1) RSD-Intra-day (%,n=5)

3.17

6.34

PCB-52

10-1000

0.9938

0.79

2.63

3.16

PCB-101

10-1000

0.9982

1.98

6.60

6.85

PCB-118

10-1000

0.9960

1.94

6.47

3.59

PCB-138

10-1000

0.9988

0.72

2.42

2.69

PCB-153

10-1000

0.9951

1.67

5.55

4.67

PCB-180

10-1000

0.9936

1.93

6.43

4.90

Table 2 SPE methods comparison for PCBs analysis

Adsorbent

Linear range (g L-1) LODs (ng L-1) RSD (%)

Oasis HLB

0.4-80

52.3-89.1

3.53-8.21

MWCNTs

0.2-50

20.3-36.7

2.78-9.62

SWCNTs

0.1-30

5.2-20.5

4.67-7.65

Activated carbon

0.2-80

11.2-18.6

5.01-7.90

CNT sponges

0.01-1

0.72-1.98

2.69-6.85

Table 3 Chemical structure and physical-chemical properties of PCBs

PCBs

Structure

PCB-28

Boiling point (
C)

log Kow
[35-39]

Molecular diameter
(nm)a

330.3

5.92

1.0639

21

PCB-52

344.9

6.26

0.9142

PCB-101

371.0

6.85

0.8826

PCB-118

390.2

7.12

1.0629

PCB-138

400.0

7.22

1.0624

PCB-153

376.8

7.69

1.0614

PCB-180

424.3

7.81

1.0614

Caculated by Gaussian 98 (estimated value).

Table 4 Analytical results of environmental water samples

Analytes
Ground Found (ng

PCB28

PCB52

N.D.d

N.D.

PCB101 PCB118 PCB138 PCB153 PCB180

N.D.
22

N.D.

N.D.

N.D.

N.D.

water

L-1)

Recoverya
90.23.4 89.20.8 91.14.2 86.26.5 88.75.9 88.16.5 105.46.6
(%)
Recoveryb
87.52.7 89.04.3 94.73.6 102.47.1 105.13.1 108.64.2 118.60.5
(%)
Recoveryc
102.90.4 106.42.3 113.63.4 104.30.9 115.62.9 112.45.2 108.93.8
(%)
Pond Found (ng
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
water
L-1)
Recoverya
91.52.9 87.44.5 86.73.0 86.36.4 86.25.8 83.96.5 104.14.2
(%)
Recoveryb
91.40.6 84.81.2 92.33.5 103.16.9 101.17.1 101.36.7 117.07.9
(%)
Recoveryc
97.91.3 104.93.3 119.81.7 107.65.9 116.07.6 110.97.3 115.37.2
(%)
Tap Found (ng
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
water
L-1)
Recoverya
90.63.8 86.46.4 85.36.3 84.04.7 81.16.8 82.26.7 107.95.9
(%)
Recoveryb
95.45.8 94.12.9 96.73.9 102.31.9 105.75.3 108.45.3 119.16.8
(%)
Recoveryc
92.75.6 98.92.0 110.15.7 100.84.4 109.26.0 111.83.4 107.04.8
(%)
Mine Found (ng
5.82
12.97
N.D.
N.D.
N.D.
N.D.
N.D.
water
L-1)
Recoverya
83.62.8 98.27.1 78.53.1 90.22.0 95.24.8 88.45.3 99.15.4
(%)
Recoveryb
78.40.6 80.31.8 93.26.1 98.36.2 102.35.8 82.14.7 87.43.2
(%)
Recoveryc
88.55.9 85.23.3 94.14.4 93.54.2 103.55.1 93.85.1 95.45.6
(%)
Soil
Found (ng
21232 8118
5715
6128
4618
359
8.12.3
standard
g-1)
Given (ng
substance
23551 8933
489
7221
4012
336
6.81.8
g-1)
a

Recovery of spiked 12 ng L-1.

Recovery of spiked 30 ng L-1.

Recovery of spiked 60 ng L-1.

Not detected.

23

Highlights

Investigate the feasibility of CNT sponges as a novel SPE adsorbent for PCBs.

It exhibited higher extraction performance and lower LODs than other adsorbents.

Establish a SPE method for the enrichment and analysis of PCBs at trace levels.

This study widened the application range of CNT sponges.

24

25

26