G Model
ARTICLE IN PRESS
CHROMA-355427; No. of Pages 37
Journal of Chromatography A, xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Review
Recent applications of carbon nanotube sorbents in Analytical Chemistry Bárbara Socas-Rodríguez a , Antonio V. Herrera-Herrera b , María Asensio-Ramos c , Javier Hernández-Borges a,∗ a Departamento de Química Analítica, Nutrición y Bromatología, Facultad de Química, Universidad de La Laguna (ULL), Avenida Astrofísico Francisco Sánchez s/n◦ , 38206 La Laguna (Tenerife), Spain b Servicio General de Apoyo a la Investigación (SEGAI), Universidad de La Laguna (ULL), Avenida Astrofísico Francisco Sánchez s/n◦ , 38206 La Laguna (Tenerife), Spain c Instituto Volcanológico de Canarias (INVOLCAN), Parque Taoro, 22 38400 Puerto de la Cruz, Tenerife, Spain
a r t i c l e
i n f o
Article history: Received 21 March 2014 Received in revised form 12 May 2014 Accepted 13 May 2014 Available online xxx Keywords: Carbon nanotubes Sorbents Solid-phase extraction Solid-phase microextraction Stir bar sorptive extraction Matrix solid-phase dispersion
a b s t r a c t Carbon nanotubes (CNTs) are still awakening scientists’ interest because of their inherent properties as well as their applications in a wide variety of fields. Regarding Analytical Chemistry, and although they have also been used as stationary phases in chromatography or pseudostationary phases in capillary electrophoresis, they have also found a particular place in sorbent-based extraction techniques. In fact, they are currently used as sorbents in solid-phase extraction, solid-phase microextraction, stir-bar sorptive extraction and matrix solid-phase dispersion, for analyte enrichment or storage, sample fractionation or clean-up as well as support for derivatization reactions. CNT surface is tuneable and, as a result, they can be suitably functionalized, aggregated or linked to other supports which increase their potential use as sorbents. They can also be arranged under different formats (cartridges, fibers, stir bars, disks, etc.) or even combined with magnetic nanoparticles, which clearly enlarge their applications. This review article overviews the most recent applications of CNTs as sorbent materials, covering the period from 2010 to early 2014. © 2014 Elsevier B.V. All rights reserved.
Contents 1. 2.
3.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CNTs modifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. CNTs functionalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.1. CNTs non-covalent functionalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.2. CNTs covalent functionalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Magnetic CNTs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. CNTs as sorbent supports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4. CNTs immobilization onto solid supports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Solid-phase extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.1. Conventional solid-phase extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.2. Dispersive solid-phase extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Solid-phase microextraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Membrane based microextractions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. Stir-bar sorptive extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5. Matrix solid-phase dispersion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
∗ Corresponding author. Tel.: +34 922 318 990; fax: +34 922 318 003. E-mail address: [email protected] (J. Hernández-Borges). http://dx.doi.org/10.1016/j.chroma.2014.05.035 0021-9673/© 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: B. Socas-Rodríguez, et al., J. Chromatogr. A (2014), http://dx.doi.org/10.1016/j.chroma.2014.05.035
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
G Model CHROMA-355427; No. of Pages 37
ARTICLE IN PRESS B. Socas-Rodríguez et al. / J. Chromatogr. A xxx (2014) xxx–xxx
2
4.
Conclusions and future trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Introduction We are currently living in the era of nanotechnology, in which challenging and interesting applications are continuously being proposed. Among new nanomaterials, the nanocarbon-based family, which includes graphene, fullerenes, nanotubes, nanodiamonds and nanohorns [1,2], has highly attracted researchers’ attention in many fields. Although all these new materials show interesting properties, carbon nanotubes (CNTs) studies have particularly triggered an exponential growth in nanoscience, being at the forefront of scientific research in physics, chemistry, material sciences, etc. As it is well known, CNTs are graphene sheet structures rolled up in the shape of a cylinder which can have an open end or a closed end depending on the synthetic procedure. In this sense, their synthesis and characterization, which are not easy tasks, have been recently reviewed by Liu et al. [3]. At present it is clear that the best methods for CNTs synthesis are still chemical vapor deposition (CVD), arc discharge and laser vaporization or laser ablation, with different variants/improvements (especially in catalyst preparation and new carbon sources). However, explanation for the growth mechanism of CNTs is still under a fair amount of controversy [3]. CNTs show very interesting properties that arise from one key feature: the combination of small size and immense surface area. Some of the most relevant properties are their outstanding tensile strength, high thermal conductivity and stability, high resilience, semiconducting and/or conducting electrical properties, etc. [4–6]. As a result, CNTs have been used in different disciplines. Some recent review articles have covered their specific applications in drug delivery [7,8], reproductive medicine [9], as biosensors [8], as scaffolds for tissue engineering [8], and for decontamination purposes [8], among others. CNTs are also playing an interesting role in Analytical Chemistry [10], in particular, for sample preparation. In this sense, and as it is well known, the use of sorbent-based techniques has found a very important place in sample preparation as a result of their high extraction capacity and selectivity. In general, solid-based sorbents can be used for different purposes depending on the physicochemical properties of both analyte and stationary phases and therefore, on the extraction mechanism/principle. Analyte enrichment or storage, sample fractioning or clean-up as well as support for derivatization reactions are some of the examples. Essentially, solid-phase extraction (SPE), matrix solid-phase dispersion (MSPD), solid-phase microextraction (SPME) and stir bar sorptive extraction (SBSE) comprise the sorbent-based techniques most commonly used, in which CNTs have also been applied. In this sense, CNTs are suitable sorbents for the extraction of both organic and inorganic analytes because of the previously mentioned combination of small size and extremely high surface area. On the one hand, they are somehow tuneable and they can be functionalized, aggregated or linked to other supports which may increase their affinity toward target compounds. On the other hand, they can be arranged under different presentations or formats (cartridges, disks, fibers, stir bars, solid suspensions, etc.) or even combined with magnetic-nanoparticles (m-NPs) which enlarge their applications. In previous articles developed by our group in 2010 [11] and 2012 [12], we reviewed the application of CNTs in SPE [11] as well as their more general application in Separation Science [12]. Following this trajectory, this article aims to overview the most
00 00 00 00
recent and challenging applications of CNTs sorbents, focusing on articles published in the period 2010 until early 2014. First, the different modifications of CNTs that may have an interesting effect on their use as sorbents will be presented, to continue with their most recent applications in the different sorbent-based extraction techniques used for Analytical Chemistry purposes. 2. CNTs modifications 2.1. CNTs functionalization CNTs functionalization allows the modification of their physical and chemical properties. Very often, CNTs surface is altered for this purpose or to change their selectivity, depending on the final required characteristics. This modification not only enlarges their potential but also enhances their solubility, which is extremely low in most solvents due the strong intertube van der Waals interactions [13,14]. The functionalization process frequently includes an acidic or an oxidative treatment that also reduces the impurities resulting from the synthetic procedure of CNTs. It can be done using either easy or complex methodologies to obtain covalent [15,16] or non-covalent [15,17,18] modified CNTs. On the one hand, covalent functionalization could be carried out by direct covalent sidewall functionalization with the molecule of interest or by indirect covalent functionalization with carboxylic groups previously introduced on their surface [15,16]. On the other hand, CNTs tend to form non-covalent aggregates via van der Waals forces, – stacking interactions, hydrogen bonds, electrostatic forces and hydrophobic interactions [15,17,18]. The combination of two or more of these interactions improves the stability and the selectivity of the system. A particular case of non-covalent functionalization is the endohedral filling of CNTs with atoms or small molecules. The encapsulation not only protects small molecules against the external environment and prevent from aggregation but also improves the dispersion stability of core-shell nanomaterials in a wide range of solvents [4]. From a detailed revision of the literature, it is quite clear that a high number of procedures for CNTs sidewall modifications have been published in the period covered by the present review in addition to their later application as sorbents (see Tables 1–5). With the aim of verifying CNTs functionalization and demonstrating that the sorptive structure has changed, a combination of different structural characterization techniques is frequently used. Among them, electronic microscopy (principally transmission electron microscopy, TEM), Fourier transform infrared (FTIR) spectroscopy, Raman spectroscopy, X-ray diffraction (XRD), thermo gravimetric analysis (TGA) and X-Ray photoelectron spectroscopy (XPS) are the most widely used. Electronic microscopy not only provides a 3D image of the material but also allows a semi-quantitative study of the elemental composition, depending on the instrument. FTIR and Raman are complementary techniques that allow confirmation of the molecular structure of liquid, solid or gas inorganic and organic samples through their covalent bonds. XRD supplies information about the crystal structure of compounds and imperfections or defects in the materials, while TGA allows deducing the introduction of changes in a compound by measuring the variation of decomposition temperatures. Finally, XPS provides information about the chemistry composition of CNTs surface. This set of
Please cite this article in press as: B. Socas-Rodríguez, et al., J. Chromatogr. A (2014), http://dx.doi.org/10.1016/j.chroma.2014.05.035
Sample
Au(III), Pd(II) Cd(II)
As, Bi, Cd, Pb, Hg, Ti
Soil (CRM), fish, shrimp, water and soil Distilled, tap, spring, river, and waste water, herbs and rice Skin whitening cosmetics
CNTs type
Sorbent amount (mg)
Elution
Determination technique
Recovery (%)
LODs
Comments
Reference
200 mL
MWCNT (–)/PPy nanocomposite MWCNTs (–)/tartrazine
200
HNO3 3 M (10 mL)
FAAS
99–103
1.1 g/L
–
[19]
200
HCl 2 M (10 mL)
FAAS
94–102
0.8–6.6 g/L
–
[31]
200
FAAS
98
250 extractions
4 PAHs
Sol–gel
[165]
[166]
[170]
[59]
[172]
[147]
[148]
ARTICLE IN PRESS
3 non-steroidal antiinflammatory drugs 5 aromatic amines
G Model
CHROMA-355427; No. of Pages 37
Analytes
B. Socas-Rodríguez et al. / J. Chromatogr. A xxx (2014) xxx–xxx
Please cite this article in press as: B. Socas-Rodríguez, et al., J. Chromatogr. A (2014), http://dx.doi.org/10.1016/j.chroma.2014.05.035
Table 3 Recent applications of CNTs as SPME coatings.
15
Sample
Sample amount
Coating
Fiber preparation method
Supporting substrate
Desorption
Determination technique
Recovery (%)
LODs
Comments
Reference
23 hydrocarbons
River and wastewater, soil
5 mL (water), 1.5 g (soil)
o-MWCNTs (–)
Physical coating
Stainless steel wire
Thermal (260 ◦ C, 15 min)
GC-FID
–
DI-SPME; use of PVC-THF viscose suspension as organic binder
[149]
2 amines, 1 alcohol, 2 carboxylic acids
Artificial water 10 mL
MWCNTs (–)
Physical coating
Stainless steel wire
Thermal (200 ◦ C, 4 min)
GC-FID
–
0.10–1.10 ng/L (water), 0.1–0.77 ng/kg (soil) 0.048–0.070 g/L
[150]
6 phenols
River water
10 mL
MWCNTs (–)-NH2
Physical coating
Stainless steel wire
Thermal (280 ◦ C, 5 min)
GC-FID
82–110
0.02–0.10 g/L
7 phenols
–
o-MWCNTs (–)/Si
Chemical bonding
0.005–0.05 g/L
Chemical bonding
GC-ECD
72–135
0.43–2.13 ng/L
8 multiclass
Artificial water 10 mL
MWCNTs (o.d.: 5–15 nm) MWCNTs (–)/PAN
Chemical bonding
Thermal (280 ◦ C, 5 min) Thermal (250 ◦ C, 5 min) Thermal (280 ◦ C, 2 min)
73–142
10 mL
Stainless steel wire Stainless steel wire Stainless steel wire
GC-FID
8 OCPs
River and wastewater River water
GC-FID
–
1.99–2.72 g/L
2 OPPs
River water
8 mL
o-MWCNTs (–)
Chemical bonding
Au wire
Thermal (200 ◦ C, 6 min)
GC-FID GC–MS
97–104
0.2–0.3 g/L
15 VOCs
Eucalyptus leaf
–
Thermal (250 ◦ C, 4 min)
GC–MS
–
–
River and rain water River water
10 mL
Chemical bonding, sol–gel and physical coating Chemical bonding and sol–gel Chemical bonding and sol–gel
Cu wire
7 PAHs
MWCNTs/PDMS and o-MWCNTs (–)/PDMS MWCNTs (o.d.: 0.8 nm)/TiO2 o-MWCNTs (o.d.: 20–40 nm)/PDMS
Electrochemically enhanced DI-SPME; use of nafion as organic binder; comparison with PA, PDMS/DVB, Car/PDMS and PDMS fibers DI-SPME; dopamine-Tris buffer solution as organic binder; comparison with other fibers. Life span: 110 extractions DI-SPME; Si interlayer by magnetron sputtering HS-SPME; corrosion of the metal wire with aqua regia HS-SPME; etching of the metal wire with HNO3 ; life span: >250 extractions HS-SPME; surface functionalization of the Au wire; life span: >100 extractions Comparison with PDMS and PDMS/silica NPs
Stainless steel wire Blade-shape substrate
GC-FID
93–121
0.002–0.004 g/L
HPLC-UV
64–90
1–2 g/L
5 phthalate esters
Mineral and tap water
10 mL
GC-FID
90–113
6 pyrethroid pesticides Acrylamide
Thermal (300 ◦ C, 5 min) ACN/water 70:30 (v/v) (1.2 mL, 100 min) Thermal (250 ◦ C, 25 min) Thermal (270 ◦ C, 4 min) Thermal (220 ◦ C, 3 min)
GC-ECD
8 aromatic hydrocarbons 4 aromatic hydrocarbons
Thermal (280 ◦ C, 2 min) Thermal (300 ◦ C, 5 min)
o-MWCNTs (–)/PPy Electropolymerization
Stainless steel wire
Pond, tap and 5 mL seawater 1.0 g Potato, chips, peanut, coffee, almond, red tea
MWCNTs (o.d.: 2 nm)/PPy SWCNTs (o.d.: 2 nm)/PPy
Electropolymerization
Stainless steel wire Stainless steel wire
Well, tap and wastewater Tap, sea and wastewater
o-MWCNTs (–)/PoPD o-SWCNTs (–)
Electropolymerization
10 mL 10 mL
Electropolymerization
Electrophoretic deposition
Stainless steel wire Pt wire
[152] [153] [154]
[[163]
[162]
DI-SPME; life span: >200 extractions DI-SPME; conditioning with HCl and polymerization life span: 110 extractions
[176]
0.05–0.1 g/L
HS-SPME; life span: 60 extractions
[155]
83–112
0.12–0.43 g/L
[156]
GC-ECD
69–87
0.26 g/L
GC-FID
81–112
0.02–0.09 g/L
GC-FID
75–105
0.005–0.026 g/L
DI-SPME; comparison with PDMS and PDMS/DVB fibers DI-SPME; comparison with Ppy, Ppy/MWCNTs and PDMS/CAR; life span: >300 extractions HS-SPME; life span: 90 extractions HS-SPME. Life span: >120 extractions
[177]
[157]
[158] [159]
ARTICLE IN PRESS
1.2 mL
[151] B. Socas-Rodríguez et al. / J. Chromatogr. A xxx (2014) xxx–xxx
3 phenolic compounds
G Model
Analytes
CHROMA-355427; No. of Pages 37
16
Please cite this article in press as: B. Socas-Rodríguez, et al., J. Chromatogr. A (2014), http://dx.doi.org/10.1016/j.chroma.2014.05.035
Table 3 (Continued)
G Model
CHROMA-355427; No. of Pages 37
Sample
Sample amount
Coating
Fiber preparation method
Supporting substrate
Desorption
Determination technique
Recovery (%)
LODs
4 EDCs
Tap and seawater
10 mL
o-SWCNTs (–)
Electrophoretic deposition
Pt wire
Mobile phase desorption
HPLC-DAD
82–97
0.32–0.52 g/L
F− , Cl− , Br− , NO3 − , SO4 2−
Distilled water
25 mL
o-SWCNTs (–)
Electrophoretic deposition
Pt wire
Water (200 L, potential assisted)
IC-ED
65–121
7 fluoroquinolone antibiotics
Urine and soil
15 mL (diluted urine) 2 g (soil)
o-MWCNTs (o.d.: 8–15 nm; i.d.: 3–5 nm)/MIPPy
Electrophoretic deposition and electropolymerization
Pt wire
MeOH/water/acetic HPLC-DAD acid 80:18:2 (v/v/v) (400 L, 15 min)
85–94 (urine), 90–96 (soil)
6 PAHs
Tap and well water
5 mL
o-MWCNTs (o.d.: 20–40 nm)
Electrodeposition
Stainless steel wire
Thermal (280 ◦ C, 15 min)
88–105
GC-FID
Comments
DI-SPME; comparison with PA fiber; life span: >120 extractions 0.06–0.26 g/L Electrochemically enhanced DI-SPME. Life span: 50 extractions 0.5–1.9 g/L (water Electrochemically samples) enhanced DI-SPME; comparison with PDMS fibers; life span: 50 extractions 0.03–0.07 g/L HS-SPME; life span: 30–40 extractions
Reference [160]
[161]
[138]
[60]
ACN: acetonitrile; CAR: carboxen; DAD: diode-array detector; DI-SPME: direct-immersion solid-phase microextraction; DPASV: differential pulse anodic stripping voltammetry; DVB: divinylbenzene; ECD: electron capture detector; ED: electrochemical detector; EDCs: endocrine disrupting compounds; FID: flame ionization detector; GC: gas chromatography; HS-SPME: head-space solid-phase microextraction; HPLC: high performance liquid chromatography; IC: ion chromatography; MeOH: methanol; MIPPy: molecularly imprinted polypyrrole; MS: mass spectrometry; NPs: nanoparticles; o-MWCNTs: oxidized multi-walled carbon nanotubes; o-SWCNTs: oxidized single-walled carbon nanotubes; PAHs: polycyclic aromatic hydrocarbons; PDMS: polydimethylsiloxane; PEG: poly(ethylene glycol); PoPD: poly-o-phenylenediamine; PPy: polypyrrole; UV: ultraviolet detector.
ARTICLE IN PRESS
Analytes
B. Socas-Rodríguez et al. / J. Chromatogr. A xxx (2014) xxx–xxx
Please cite this article in press as: B. Socas-Rodríguez, et al., J. Chromatogr. A (2014), http://dx.doi.org/10.1016/j.chroma.2014.05.035
Table 3 (Continued)
17
G Model CHROMA-355427; No. of Pages 37 18
ARTICLE IN PRESS B. Socas-Rodríguez et al. / J. Chromatogr. A xxx (2014) xxx–xxx
complementary techniques is often supplemented with studies of sorption capacity, including the obtaining of sorption isotherms or the development of kinetic and thermodynamic studies. 2.1.1. CNTs non-covalent functionalization Non-covalent functionalization involves the physical adsorption of molecules onto the sidewalls of CNTs and is governed by hydrophobic, van der Waals and/or electrostatic forces. Due the particular structure of CNTs, interaction with molecules with an aromatic system is particularly effective. For this reason, the formation of superstructures with biomolecules (as carbohydrates, proteins, enzymes and DNA), surfactants and polymers is very effective. In this sense, recent articles have reported the non-covalent modification of CNTs surface with polymers (such as polypyrrole (PPy) [19], polyethyleneimine [20], polydiphenylamine [21], poly(2-aminothiophenol) [22], polyaniline and poly(3,4-dioxythiophene) [23] or polyvinylalcohol [24]), ionic liquids (ILs, such as hexylmethylimidazolium hexafluorophosphate [HMIm][PF6 ] [25–27] or butylmethylimidazolium hexafluorophosphate [BMIm][PF6 ] [28]), biological structures (like biochar [29] or bacteria [30]) and other organic molecules (such as tartrazine [31], 4-(2-thiazolylazo)resorcinol [32], Aliquat 336 [33], poly(diallylmethylammoniumchloride) [34], di-(2ethylhexyl) phosphoric acid and tri-octyl phosphine oxide [35], tannic acid [36], or 5-(4 -dimethylamino-benzylidene)-rhodanine [37]). Compared to the chemical modification, this kind of functionalization could be operated under relatively mild reaction conditions maintaining the graphene-like structure of CNTs. Strategies adopted to modify CNTs with polymers include physical mixing in solution [20], in situ polymerization of monomers in the presence of CNTs [19,22,24,38] or surfactant-assisted processing of composites [21]. The combination of ILs and CNTs generates a soft material or gel with interesting properties which make them suitable for different analytical applications. For example, these gels are highly stable and they could be processed in almost any shape. Furthermore, they could retain their physical properties even under reduced pressures (due the negligible volatility of ILs) and facilitate electron transfer (due the large specific surface area of CNTs) [18]. The interaction mechanism between CNTs and ILs is believed to fall in the cation– and – interactions [18,27]. Regarding their use as sorbents, PoloLuque et al. [26] studied the sorption capacity of eight different soft materials obtained from the combination of [HMIm][PF6 ] with different types of CNTs (one type of single walled CNTs, SWCNTs, and seven types of multi-walled CNTs, MWCNTs, differing in dimensions). For this purpose, they mixed the suitable amount of CNTs and IL and ground them in an agate mortar with a pestle. The critical gel concentration was experimentally obtained since at higher concentrations the gel and the IL are clearly separated. Materials were characterized by Raman spectroscopy and their stability was also studied. Once all the materials were obtained, they were used to impregnate natural cotton fibers. Later, the same authors extended the methodology by using the same IL joined to coiled CNTs (pentagons and heptagons are incorporated in the structure) [27] or other MWCNTs [25]. Similarly Pourreza et al. [28] combined [BMIm][PF6 ] with MWCNTs by grinding in a mortar for about 1 h to impregnate pine sawdust. A very special application is the non-covalent functionalization of CNTs with bacteria [30,39]. For this purpose both living [30] and dead [30,39] Escherichia coli [30] or Pseudomonas aeruginosa [39] bacteria cells were mixed with CNTs and heated in an oven at a maximum temperature of 105 ◦ C. These sorbents were later used to extract several metals (Cd(II), Co(II), Cu(II), Ni(II) for E. Coli and Co(II), Cd(II), Pb(II), Mn(II), Cr(III), Ni(II) for P. aeruginosa) from water [30] or tomato leaves, bovine liver, boiled wheat, canned fish, black tea, lichen and natural waters [39]. Such manuscripts,
apparently anecdotal, demonstrate the potential of CNTs as supports for biosorption procedures. Although the introduction of an oxidation step is more common for covalent functionalization, it is also occasionally used for non-covalent modification to induce a more efficient interaction of CNTs with the modifier molecule. In this sense, Chen et al. [20] induced the creation of carboxylic groups to enhance electrostatic attraction between the positively charged protonated amines in the polyethyleneimine polymer and the carboxyl groups of oxidized-CNTs (o-CNTs) surface. Similarly, Tobiasz et al. [40] oxidized MWCNTs to enhance the interaction with 5dodecylsalicylaldoxime. 2.1.2. CNTs covalent functionalization Covalent functionalization of CNTs is relatively easy compared with other aromatic sp2 -bonded carbon structures such as graphite and graphene [18]. This fact is originated by the bending of the planar graphene sheet into a 3D carbon skeleton and the decrease in the orbital alignment [41]. Moreover, there should be a direct correlation between tube diameter and reactivity so that smaller CNTs are expected to be more reactive than bigger ones [42]. In the literature, reactions of halogenation, hydrogenation, radical addition, nucleophilic addition and cycloaddition with CNTs have been described [16,43]. This type of modifications involves the formation of a linkage between the functional groups and the carbon skeleton of nanotubes. As mentioned above, this phenomenon can occur by direct sidewall functionalization or indirect functionalization with carboxylic groups located on the defects of the carbon sheet. The first of them produces a change in carbon hybridization from sp2 to sp3 and, therefore, a loss of electronic conjugation, while the second takes place via the carboxylic groups at the open ends and holes of the sidewalls. Such carboxylic groups can be present on the as-grown CNTs; furthermore, they can be generated by putting CNTs through strong chemical conditions (usually with a strong acid like HNO3 , H2 SO4 and HCl or other oxidants such as H2 O2 , O3 , KMnO4 and NaOCl at high temperature) to introduce lactone, hydroxyl, carboxyl or carbonyl groups. Such reactions occur at a larger scale on their fullerene like tips, which are more reactive than their walls due the higher distortion of the planar framework of graphene [16]. Gas-phase oxidation mainly introduces hydroxyl and carbonyl surface groups, while oxidation in the liquid phase particularly increases the content of carboxylic acids [44,45]. The amount of carboxyl and lactone groups introduced on CNTs treated with HNO3 is higher than with other reagents [46]. Usually, the generated carboxylic groups are converted into acyl chloride and then undergo an esterification or amidation reaction [15]. It is worth mentioning that the use of some of the indicated oxidants, such as HNO3 , opens CNTs’ caps [47] and therefore, more surface area is available for either internal or external sorption. The principal drawback of covalent functionalization is that the graphitic structure of CNTs is disturbed, resulting in significant changes in their physical properties. In fact, it may even destroy the whole structure if functionalization is too high. A clear example for this issue is the recent work of Yang et al. [48] who studied the influence of different concentrations of NaOCl on the oxidation of MWCNTs with an outer diameter of 10–20 nm. Of the tested percentages of NaOCl (1, 3, 5 and 7%, w/v), the highest concentration destroyed CNTs structure and carbon appeared as larger agglomerated particles (see Fig. 1). These results were confirmed by TGA and FTIR analyses. In addition, they studied the influence of pH on z potentials and quantified the functional groups by Boehm’ titration. The adsorption capacity for Cu(II) was measured as well. Results demonstrated that adsorption increased when 5% (w/v) of NaOCl was used and fell with a 7% (w/v) of NaOCl. The oxidation process also provides a certain degree of clean-up of CNTs because metal impurities and amorphous carbon could be
Please cite this article in press as: B. Socas-Rodríguez, et al., J. Chromatogr. A (2014), http://dx.doi.org/10.1016/j.chroma.2014.05.035
Donor phase
CNTs type
Supporting substrate
Back-extraction
Determination technique
Recovery (%)
LODs
Comments
Reference
6 OPPs
Reservoir and drainage water, watermelon
o-MWCNTs (o.d.: 40–60 nm)
Stainless steel wire inserted in a PP HF impregnated by simple introduction of CNTs dispersed in 1-octanol
Desorption in MeOH (50 L, 5 min)
HPLC-DAD
83–101
0.1–0.3 g/L (water), 1.0–1.5 g/kg
HF-SPME combined with LLME
[61]
5 triazine herbicides
Tap, lake and drainage water, milk
15 mL of water or 15 g of the watermelon supernatant, pH 4–6, 5% (w/v) NaCl, 50 L chlorobenzene 15 mL, pH 6–8, 15% (w/v) NaCl, 30 L chlorobenzene
o-MWCNTs (o.d.: 40–60 nm)
Desorption in MeOH (50 L, 10 min)
HPLC-DAD
81–107
0.08–0.5 g/L
HF-SPME combined with LLME
[62]
5 carbamate pesticides
Apple
9 mL of the aqueous extract, pH 5.5, 16% (w/v) NaCl
MWCNTs (o.d.: 8–15 nm)
Desorption in MeOH (20 L, 25 min)
HPLC-DAD
92–113
0.09–6.00 ng/g
HF-SPME
[260]
2 alkaloids
Urine
10 mL, pH 14
MWCNTs (i.d.: 8–15 nm, o.d.: 30–50 nm)
Desorption in MeOH (80 L, 25 min)
HPLC-UV
91–94
0.7–0.9 g/L
HF-SPME
[261]
3 non-steroidal anti-inflammatory drugs
Urine
3 mL, pH 2
g-MWCNTs (o.d.: 10–20 nm)/PEG
Thermal desorption (280 ◦ C, 5 min)
GC-FID
8–12
0.03–0.07 g/L
HF-protectedSPME; comparison with HF-LPME and DI-SPME
[262]
Caffeic acid
Medicinal plants
15 mL of the aqueous extract, pH 2.5
o-MWCNTs (o.d.: 10–15 nm)
Retraction of acceptor phase (NaOH, pH 11, 6.5 L)
HPLC-UV
89–95
0.05 ng/L
Lumen filled with NaOH pH 11; comparison with HF-LPME
[63]
2 non-steroidal anti-inflammatory drugs
Plasma, urine, breast milk and wastewater
pH 7.4
o-MWCNTs (o.d.: 10–15 nm)
Stainless steel wire inserted in a PP HF impregnated by simple introduction of CNTs dispersed in 1-octanol Microsyringe needle inserted in a PP HF impregnated by simple introduction of CNTs dispersed in CTAB and water Microsyringe needle inserted in a PP HF impregnated by simple introduction of CNTs dispersed in CTAB and water PP HF impregnated in 1-octanol inserted in a PEG-g-MWCNTs coated fused-silica capillary PP HF impregnated by simple introduction of CNTs dispersed in 1-octanol PP HF impregnated by simple introduction of CNTs dispersed in 1-octanol
Retraction of acceptor phase (NaOH, pH 12, 20 L)
CE-UV
90–94
1–3 g/L
Lumen filled with NaOH pH 12; electrokinetic assisted extraction; comparison in the absence of CNTs and with HF-LPME
[263]
ARTICLE IN PRESS
Sample
G Model
CHROMA-355427; No. of Pages 37
Analytes
B. Socas-Rodríguez et al. / J. Chromatogr. A xxx (2014) xxx–xxx
Please cite this article in press as: B. Socas-Rodríguez, et al., J. Chromatogr. A (2014), http://dx.doi.org/10.1016/j.chroma.2014.05.035
Table 4 Recent applications of CNTs in membrane-based microextractions.
19
Sample
Donor phase
CNTs type
Supporting substrate
Back-extraction
Determination technique
Recovery (%)
LODs
Comments
Reference
Buprenorphine
Urine
4 mL, pH 2
o-MWCNTs (o.d.: 10–15 nm)
Retraction of acceptor phase (HCl, pH 2, 20 L)
CE-UV
92
1 g/L
MeOH
–
MWCNTs (–)
–
HPLC-UV
–
0.001–0.009 ng/L
Lumen filled with HCl pH 2; electrokinetic assisted extraction; comparison in the absence of CNTs Used for preconcentration of analytes from polar solvent
[237]
4 drugs
5 multiclass polar analytes
Water
200 mL
MWCNTs (–)
Retraction of acceptor phase (isopropyl acetate, 50 L)
HPLC-UV
–
–
Lumen filled with isopropyl acetate; comparison in the absence of CNTs and with o-CNTs
[265]
Phenobarbital
Wastewater
5 mL, pH 7
o-MWCNTs (o.d.: < 20 nm)
Desorption in MeOH (3 mL, 10 min)
HPLC-UV
102
0.32 g/L
Lumen filled with the CNTs/sol–gel solution
[64]
Diethylstilbestrol
Milk
1 mL, pH 3
o-MWCNTs (–)
Desorption in MeOH (50 L, 10 min)
HPLC-DAD
58–120
5.1 g/L
Lumen filled with 1-octanol
[65]
2 non-steroidal anti-inflammatory drugs
Tap and wastewater
10 mL, pH 3
o-MWCNTs (–)
PP HF impregnated by simple introduction of CNTs dispersed in 2-nitrophenyl octyl ether PP HF impregnated by simple introduction of CNTs dispersed in PVDF Nafion HF impregnated by simple introduction of CNTs dispersed in PVDF PP HF reinforced with CNTs/silica composite by sol–gel technology PP HF reinforced with CNTs/silica composite by sol–gel technology PP HF reinforced with CNTs/silica composite by sol–gel technology
Desorption in ACN (15 L, 25 min)
HPLC-DAD
70–114
0.40–4.48 g/L
[66]
3 phthalates
Juice, milk, carbonated drink, beer and wine Rice
1.0 mL, pH 3, 20% (w/v) NaCl
o-MWCNTs (–)
Desorption in diethyl ether (300 L, 10 min)
GC–MS
68–115
0.006–0.03 g/L
Lumen filled with 1-octanol; comparison with HF-SPME and HF-LPME Lumen remained empty
4 mL aqueous extract, pH 5, 1% (w/v) NaCl
o-MWCNTs (–)
DPASV
–
0.0073–0.025 g/L
Lumen filled with CNTs/sol–gel solution
[79]
20 mL of the aqueous extract, pH 8
o-MWCNTs (–)
Desorption in MeOH/HNO3 1:1 (v/v) (3 mL, 5 min) Desorption in MeOH (2 mL, 15 min)
HPLC-DAD
0.061–0.074 g/L
47–107
Lumen filled with CNTs/sol–gel solution
[67]
Rice, peanut and wheat
PP HF reinforced with CNTs by sol–gel technology
[266]
ARTICLE IN PRESS
2 aflatoxins
[264] B. Socas-Rodríguez et al. / J. Chromatogr. A xxx (2014) xxx–xxx
Pb(II), Cd(II), Cu(II)
PP HF reinforced with CNTs/silica composite by sol–gel technology PP HF reinforced with CNTs by sol–gel technology
G Model
Analytes
CHROMA-355427; No. of Pages 37
20
Please cite this article in press as: B. Socas-Rodríguez, et al., J. Chromatogr. A (2014), http://dx.doi.org/10.1016/j.chroma.2014.05.035
Table 4 (Continued)
G Model
CHROMA-355427; No. of Pages 37
Sample
Donor phase
CNTs type
Supporting substrate
Back-extraction
Determination technique
Recovery (%)
LODs
Comments
Reference
4 aromatic hydrocarbons
Wastewater and hair
o-MWCNTs (–)
Desorption in MeOH (400 L, 20 min) Desorption in MeOH (1 mL, 5 min)
74–87
0.5–0.7 ng/L
HPLC-DAD
–
–
1,2,4Trichlorobenzene, Cu(II)
Water
Crosslinking of CNTs/PVA suspension
–
GC-ECD GC–MS Spectrophotometry
–
–
Lumen filled with CNTs/sol–gel solution Lumen filled with CNTs/sol–gel solution; comparison with HF reinforced with SiO2 , nano TiO2 and MgO Comparison with PAC/PVA membrane; development of kinetics studies, batch sorption, batch uptake, and diaphragm cell breakthrough experiments
[68]
Wastewater and hair
PP HF reinforced with CNTs by sol–gel technology PP HF reinforced with CNTs by IL mediated sol–gel technology
GC-FID
6 multiclass pesticides
4.5 mL of the aqueous extract 5 mL of the aqueous extracts, pH 7
o-MWCNTs (o.d.: 10–15 nm), MWCNTs-NH2
SWCNTs
[267]
[268]
CE: capillary electrophoresis; CTAB: cetyltrimethylammonium bromide; DAD: diode-array detector; ECD: electron capture detector; FID: flame ionization detector; GC: gas chromatography; HF: hollow fiber; HF-SPME: hollowfiber solid-phase microextraction; HPLC: high performance liquid chromatography; LLME: liquid–liquid microextraction; MS: mass spectrometry; o-MWCNTs: oxidized-multi-walled carbon nanotubes; PAC: powdered activated carbon; PP: polypropylene; PVA: poly(vinyl alcohol); PVDF: polyvinylidene fluoride; UV: ultraviolet detector.
ARTICLE IN PRESS
Analytes
B. Socas-Rodríguez et al. / J. Chromatogr. A xxx (2014) xxx–xxx
Please cite this article in press as: B. Socas-Rodríguez, et al., J. Chromatogr. A (2014), http://dx.doi.org/10.1016/j.chroma.2014.05.035
Table 4 (Continued)
21
Sample amount
CNTs type
Stir-bar preparation method
Desorption
Determination technique
Recovery (%)
LODs
Comments
Reference
Pb (II), Cd (II)
Tap and river water
5 mL
o-MWCNTs (o.d.: 10–15 nm)
Sol–gel
1M HNO3 :MeOH 70:30 (v/v) (3 mL, 30 min)
DPASV-HDME
98–102
0.012–0.015 g/L
[173]
7 phenols
Lake water and soil
10 mL (water) 0.5 g (soil)
MWCNTs-DDM (–)
Sol–gel
MeOH/1 mM NaOH, 8:2 (v/v) (100 L, 25 min)
HPLC-UV
79–123% (water), 71.0–119.2% (soil)
0.14–1.76 g/L
Brilliant green
Fish pond water
5 mL
MWCNTs (o.d.: 10–15 nm)
Physical coating
MeOH (2 mL, 2 min)
UV/DAD
120
0.55 g/L
Ligand assisted pseudo-stir bar HF-SLME using sol–gel sorbent reinforced with CNTs; use of magnetic stoppers DDMMWCNTs/PDMS coated stir bar, comparison with PDMS and o-MWCNTs coated stir bar Pseudo-stir bar HF-SLME filled with MWCNTs dispersed in 1-octanol; use of magnetic stoppers
[175]
[174]
Matrix solid-phase extraction Analytes
Sample
Sample amount
CNTs type
Sorbent amount (mg)
Elution
Determination technique
Recovery (%)
LODs
Comments
Reference
9 OPPs
Apple, grape, banana, strawberry, tomato, cabbage, spinach and rape Butter
0.5 g
MWCNTs (i.d.: 40–60 nm)
1.0 g
MeOH (2.5 mL)
LC–MS/MS
71 –103
0.06 –0.15 g/kg
–
[271]
0.5 g
GMWCNTs (o.d.: 8–15 nm) MWCNTs (o.d.: 10–20 nm)
0.3 g GMWCNTs and 0.10 g MWCNTs
Acetone/ethyl acetate 1:1 (v/v) (20 mL)
GC–MS
85 –111
0.2 –1.3 g/kg
Derivatization with HFBA-ACN
[272]
8 hormones
ACN: acetonitrile; DAD: diode-array detector; DDM: 4,4 -diaminodiphenylmethane; DPASV: differential pulse anodic stripping voltammetry; GC: gas chromatography; GMWCNTs: graphitized multi-walled carbon nanotubes; HF-SLME: hollow-fiber solid–liquid microextraction; HFBA: heptafluorobutyric anhydride; HMDE: hanging mercury drop electrode; HPLC: high performance liquid chromatography; MeOH: methanol; MS: mass spectrometry; MWCNTs: multi-walled carbon nanotubes; UV: ultraviolet detector.
ARTICLE IN PRESS
Sample
B. Socas-Rodríguez et al. / J. Chromatogr. A xxx (2014) xxx–xxx
Analytes
G Model
Stir bar sorptive extraction
CHROMA-355427; No. of Pages 37
22
Please cite this article in press as: B. Socas-Rodríguez, et al., J. Chromatogr. A (2014), http://dx.doi.org/10.1016/j.chroma.2014.05.035
Table 5 Recent applications of CNTs as SBSE and MSPD sorbents.
G Model CHROMA-355427; No. of Pages 37
ARTICLE IN PRESS B. Socas-Rodríguez et al. / J. Chromatogr. A xxx (2014) xxx–xxx
23
Fig. 1. FE-SEM images for (a) raw CNTs, (b) 1%, (c) 3%, (d) 5% and (e) 7% of o-CNTs. Reprinted from [48] with permission of the Society of Chemical Industry.
eliminated. Such impurities removal depends on the intensity of the procedure [49]. However, certain features of CNTs (including the sorption capacity) could depend on the presence of the mentioned impurities. In this regard, Tian et al. [50] demonstrated that metal impurities dominate the sorption capacity of certain specific commercially available CNTs for Pb(II) in water. They acquired MWCNTs with an outer diameter of
ARTICLE IN PRESS
CHROMA-355427; No. of Pages 37
Journal of Chromatography A, xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Review
Recent applications of carbon nanotube sorbents in Analytical Chemistry Bárbara Socas-Rodríguez a , Antonio V. Herrera-Herrera b , María Asensio-Ramos c , Javier Hernández-Borges a,∗ a Departamento de Química Analítica, Nutrición y Bromatología, Facultad de Química, Universidad de La Laguna (ULL), Avenida Astrofísico Francisco Sánchez s/n◦ , 38206 La Laguna (Tenerife), Spain b Servicio General de Apoyo a la Investigación (SEGAI), Universidad de La Laguna (ULL), Avenida Astrofísico Francisco Sánchez s/n◦ , 38206 La Laguna (Tenerife), Spain c Instituto Volcanológico de Canarias (INVOLCAN), Parque Taoro, 22 38400 Puerto de la Cruz, Tenerife, Spain
a r t i c l e
i n f o
Article history: Received 21 March 2014 Received in revised form 12 May 2014 Accepted 13 May 2014 Available online xxx Keywords: Carbon nanotubes Sorbents Solid-phase extraction Solid-phase microextraction Stir bar sorptive extraction Matrix solid-phase dispersion
a b s t r a c t Carbon nanotubes (CNTs) are still awakening scientists’ interest because of their inherent properties as well as their applications in a wide variety of fields. Regarding Analytical Chemistry, and although they have also been used as stationary phases in chromatography or pseudostationary phases in capillary electrophoresis, they have also found a particular place in sorbent-based extraction techniques. In fact, they are currently used as sorbents in solid-phase extraction, solid-phase microextraction, stir-bar sorptive extraction and matrix solid-phase dispersion, for analyte enrichment or storage, sample fractionation or clean-up as well as support for derivatization reactions. CNT surface is tuneable and, as a result, they can be suitably functionalized, aggregated or linked to other supports which increase their potential use as sorbents. They can also be arranged under different formats (cartridges, fibers, stir bars, disks, etc.) or even combined with magnetic nanoparticles, which clearly enlarge their applications. This review article overviews the most recent applications of CNTs as sorbent materials, covering the period from 2010 to early 2014. © 2014 Elsevier B.V. All rights reserved.
Contents 1. 2.
3.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CNTs modifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. CNTs functionalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.1. CNTs non-covalent functionalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.2. CNTs covalent functionalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Magnetic CNTs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. CNTs as sorbent supports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4. CNTs immobilization onto solid supports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Solid-phase extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.1. Conventional solid-phase extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.2. Dispersive solid-phase extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Solid-phase microextraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Membrane based microextractions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. Stir-bar sorptive extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5. Matrix solid-phase dispersion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
∗ Corresponding author. Tel.: +34 922 318 990; fax: +34 922 318 003. E-mail address: [email protected] (J. Hernández-Borges). http://dx.doi.org/10.1016/j.chroma.2014.05.035 0021-9673/© 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: B. Socas-Rodríguez, et al., J. Chromatogr. A (2014), http://dx.doi.org/10.1016/j.chroma.2014.05.035
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
G Model CHROMA-355427; No. of Pages 37
ARTICLE IN PRESS B. Socas-Rodríguez et al. / J. Chromatogr. A xxx (2014) xxx–xxx
2
4.
Conclusions and future trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Introduction We are currently living in the era of nanotechnology, in which challenging and interesting applications are continuously being proposed. Among new nanomaterials, the nanocarbon-based family, which includes graphene, fullerenes, nanotubes, nanodiamonds and nanohorns [1,2], has highly attracted researchers’ attention in many fields. Although all these new materials show interesting properties, carbon nanotubes (CNTs) studies have particularly triggered an exponential growth in nanoscience, being at the forefront of scientific research in physics, chemistry, material sciences, etc. As it is well known, CNTs are graphene sheet structures rolled up in the shape of a cylinder which can have an open end or a closed end depending on the synthetic procedure. In this sense, their synthesis and characterization, which are not easy tasks, have been recently reviewed by Liu et al. [3]. At present it is clear that the best methods for CNTs synthesis are still chemical vapor deposition (CVD), arc discharge and laser vaporization or laser ablation, with different variants/improvements (especially in catalyst preparation and new carbon sources). However, explanation for the growth mechanism of CNTs is still under a fair amount of controversy [3]. CNTs show very interesting properties that arise from one key feature: the combination of small size and immense surface area. Some of the most relevant properties are their outstanding tensile strength, high thermal conductivity and stability, high resilience, semiconducting and/or conducting electrical properties, etc. [4–6]. As a result, CNTs have been used in different disciplines. Some recent review articles have covered their specific applications in drug delivery [7,8], reproductive medicine [9], as biosensors [8], as scaffolds for tissue engineering [8], and for decontamination purposes [8], among others. CNTs are also playing an interesting role in Analytical Chemistry [10], in particular, for sample preparation. In this sense, and as it is well known, the use of sorbent-based techniques has found a very important place in sample preparation as a result of their high extraction capacity and selectivity. In general, solid-based sorbents can be used for different purposes depending on the physicochemical properties of both analyte and stationary phases and therefore, on the extraction mechanism/principle. Analyte enrichment or storage, sample fractioning or clean-up as well as support for derivatization reactions are some of the examples. Essentially, solid-phase extraction (SPE), matrix solid-phase dispersion (MSPD), solid-phase microextraction (SPME) and stir bar sorptive extraction (SBSE) comprise the sorbent-based techniques most commonly used, in which CNTs have also been applied. In this sense, CNTs are suitable sorbents for the extraction of both organic and inorganic analytes because of the previously mentioned combination of small size and extremely high surface area. On the one hand, they are somehow tuneable and they can be functionalized, aggregated or linked to other supports which may increase their affinity toward target compounds. On the other hand, they can be arranged under different presentations or formats (cartridges, disks, fibers, stir bars, solid suspensions, etc.) or even combined with magnetic-nanoparticles (m-NPs) which enlarge their applications. In previous articles developed by our group in 2010 [11] and 2012 [12], we reviewed the application of CNTs in SPE [11] as well as their more general application in Separation Science [12]. Following this trajectory, this article aims to overview the most
00 00 00 00
recent and challenging applications of CNTs sorbents, focusing on articles published in the period 2010 until early 2014. First, the different modifications of CNTs that may have an interesting effect on their use as sorbents will be presented, to continue with their most recent applications in the different sorbent-based extraction techniques used for Analytical Chemistry purposes. 2. CNTs modifications 2.1. CNTs functionalization CNTs functionalization allows the modification of their physical and chemical properties. Very often, CNTs surface is altered for this purpose or to change their selectivity, depending on the final required characteristics. This modification not only enlarges their potential but also enhances their solubility, which is extremely low in most solvents due the strong intertube van der Waals interactions [13,14]. The functionalization process frequently includes an acidic or an oxidative treatment that also reduces the impurities resulting from the synthetic procedure of CNTs. It can be done using either easy or complex methodologies to obtain covalent [15,16] or non-covalent [15,17,18] modified CNTs. On the one hand, covalent functionalization could be carried out by direct covalent sidewall functionalization with the molecule of interest or by indirect covalent functionalization with carboxylic groups previously introduced on their surface [15,16]. On the other hand, CNTs tend to form non-covalent aggregates via van der Waals forces, – stacking interactions, hydrogen bonds, electrostatic forces and hydrophobic interactions [15,17,18]. The combination of two or more of these interactions improves the stability and the selectivity of the system. A particular case of non-covalent functionalization is the endohedral filling of CNTs with atoms or small molecules. The encapsulation not only protects small molecules against the external environment and prevent from aggregation but also improves the dispersion stability of core-shell nanomaterials in a wide range of solvents [4]. From a detailed revision of the literature, it is quite clear that a high number of procedures for CNTs sidewall modifications have been published in the period covered by the present review in addition to their later application as sorbents (see Tables 1–5). With the aim of verifying CNTs functionalization and demonstrating that the sorptive structure has changed, a combination of different structural characterization techniques is frequently used. Among them, electronic microscopy (principally transmission electron microscopy, TEM), Fourier transform infrared (FTIR) spectroscopy, Raman spectroscopy, X-ray diffraction (XRD), thermo gravimetric analysis (TGA) and X-Ray photoelectron spectroscopy (XPS) are the most widely used. Electronic microscopy not only provides a 3D image of the material but also allows a semi-quantitative study of the elemental composition, depending on the instrument. FTIR and Raman are complementary techniques that allow confirmation of the molecular structure of liquid, solid or gas inorganic and organic samples through their covalent bonds. XRD supplies information about the crystal structure of compounds and imperfections or defects in the materials, while TGA allows deducing the introduction of changes in a compound by measuring the variation of decomposition temperatures. Finally, XPS provides information about the chemistry composition of CNTs surface. This set of
Please cite this article in press as: B. Socas-Rodríguez, et al., J. Chromatogr. A (2014), http://dx.doi.org/10.1016/j.chroma.2014.05.035
Sample
Au(III), Pd(II) Cd(II)
As, Bi, Cd, Pb, Hg, Ti
Soil (CRM), fish, shrimp, water and soil Distilled, tap, spring, river, and waste water, herbs and rice Skin whitening cosmetics
CNTs type
Sorbent amount (mg)
Elution
Determination technique
Recovery (%)
LODs
Comments
Reference
200 mL
MWCNT (–)/PPy nanocomposite MWCNTs (–)/tartrazine
200
HNO3 3 M (10 mL)
FAAS
99–103
1.1 g/L
–
[19]
200
HCl 2 M (10 mL)
FAAS
94–102
0.8–6.6 g/L
–
[31]
200
FAAS
98
250 extractions
4 PAHs
Sol–gel
[165]
[166]
[170]
[59]
[172]
[147]
[148]
ARTICLE IN PRESS
3 non-steroidal antiinflammatory drugs 5 aromatic amines
G Model
CHROMA-355427; No. of Pages 37
Analytes
B. Socas-Rodríguez et al. / J. Chromatogr. A xxx (2014) xxx–xxx
Please cite this article in press as: B. Socas-Rodríguez, et al., J. Chromatogr. A (2014), http://dx.doi.org/10.1016/j.chroma.2014.05.035
Table 3 Recent applications of CNTs as SPME coatings.
15
Sample
Sample amount
Coating
Fiber preparation method
Supporting substrate
Desorption
Determination technique
Recovery (%)
LODs
Comments
Reference
23 hydrocarbons
River and wastewater, soil
5 mL (water), 1.5 g (soil)
o-MWCNTs (–)
Physical coating
Stainless steel wire
Thermal (260 ◦ C, 15 min)
GC-FID
–
DI-SPME; use of PVC-THF viscose suspension as organic binder
[149]
2 amines, 1 alcohol, 2 carboxylic acids
Artificial water 10 mL
MWCNTs (–)
Physical coating
Stainless steel wire
Thermal (200 ◦ C, 4 min)
GC-FID
–
0.10–1.10 ng/L (water), 0.1–0.77 ng/kg (soil) 0.048–0.070 g/L
[150]
6 phenols
River water
10 mL
MWCNTs (–)-NH2
Physical coating
Stainless steel wire
Thermal (280 ◦ C, 5 min)
GC-FID
82–110
0.02–0.10 g/L
7 phenols
–
o-MWCNTs (–)/Si
Chemical bonding
0.005–0.05 g/L
Chemical bonding
GC-ECD
72–135
0.43–2.13 ng/L
8 multiclass
Artificial water 10 mL
MWCNTs (o.d.: 5–15 nm) MWCNTs (–)/PAN
Chemical bonding
Thermal (280 ◦ C, 5 min) Thermal (250 ◦ C, 5 min) Thermal (280 ◦ C, 2 min)
73–142
10 mL
Stainless steel wire Stainless steel wire Stainless steel wire
GC-FID
8 OCPs
River and wastewater River water
GC-FID
–
1.99–2.72 g/L
2 OPPs
River water
8 mL
o-MWCNTs (–)
Chemical bonding
Au wire
Thermal (200 ◦ C, 6 min)
GC-FID GC–MS
97–104
0.2–0.3 g/L
15 VOCs
Eucalyptus leaf
–
Thermal (250 ◦ C, 4 min)
GC–MS
–
–
River and rain water River water
10 mL
Chemical bonding, sol–gel and physical coating Chemical bonding and sol–gel Chemical bonding and sol–gel
Cu wire
7 PAHs
MWCNTs/PDMS and o-MWCNTs (–)/PDMS MWCNTs (o.d.: 0.8 nm)/TiO2 o-MWCNTs (o.d.: 20–40 nm)/PDMS
Electrochemically enhanced DI-SPME; use of nafion as organic binder; comparison with PA, PDMS/DVB, Car/PDMS and PDMS fibers DI-SPME; dopamine-Tris buffer solution as organic binder; comparison with other fibers. Life span: 110 extractions DI-SPME; Si interlayer by magnetron sputtering HS-SPME; corrosion of the metal wire with aqua regia HS-SPME; etching of the metal wire with HNO3 ; life span: >250 extractions HS-SPME; surface functionalization of the Au wire; life span: >100 extractions Comparison with PDMS and PDMS/silica NPs
Stainless steel wire Blade-shape substrate
GC-FID
93–121
0.002–0.004 g/L
HPLC-UV
64–90
1–2 g/L
5 phthalate esters
Mineral and tap water
10 mL
GC-FID
90–113
6 pyrethroid pesticides Acrylamide
Thermal (300 ◦ C, 5 min) ACN/water 70:30 (v/v) (1.2 mL, 100 min) Thermal (250 ◦ C, 25 min) Thermal (270 ◦ C, 4 min) Thermal (220 ◦ C, 3 min)
GC-ECD
8 aromatic hydrocarbons 4 aromatic hydrocarbons
Thermal (280 ◦ C, 2 min) Thermal (300 ◦ C, 5 min)
o-MWCNTs (–)/PPy Electropolymerization
Stainless steel wire
Pond, tap and 5 mL seawater 1.0 g Potato, chips, peanut, coffee, almond, red tea
MWCNTs (o.d.: 2 nm)/PPy SWCNTs (o.d.: 2 nm)/PPy
Electropolymerization
Stainless steel wire Stainless steel wire
Well, tap and wastewater Tap, sea and wastewater
o-MWCNTs (–)/PoPD o-SWCNTs (–)
Electropolymerization
10 mL 10 mL
Electropolymerization
Electrophoretic deposition
Stainless steel wire Pt wire
[152] [153] [154]
[[163]
[162]
DI-SPME; life span: >200 extractions DI-SPME; conditioning with HCl and polymerization life span: 110 extractions
[176]
0.05–0.1 g/L
HS-SPME; life span: 60 extractions
[155]
83–112
0.12–0.43 g/L
[156]
GC-ECD
69–87
0.26 g/L
GC-FID
81–112
0.02–0.09 g/L
GC-FID
75–105
0.005–0.026 g/L
DI-SPME; comparison with PDMS and PDMS/DVB fibers DI-SPME; comparison with Ppy, Ppy/MWCNTs and PDMS/CAR; life span: >300 extractions HS-SPME; life span: 90 extractions HS-SPME. Life span: >120 extractions
[177]
[157]
[158] [159]
ARTICLE IN PRESS
1.2 mL
[151] B. Socas-Rodríguez et al. / J. Chromatogr. A xxx (2014) xxx–xxx
3 phenolic compounds
G Model
Analytes
CHROMA-355427; No. of Pages 37
16
Please cite this article in press as: B. Socas-Rodríguez, et al., J. Chromatogr. A (2014), http://dx.doi.org/10.1016/j.chroma.2014.05.035
Table 3 (Continued)
G Model
CHROMA-355427; No. of Pages 37
Sample
Sample amount
Coating
Fiber preparation method
Supporting substrate
Desorption
Determination technique
Recovery (%)
LODs
4 EDCs
Tap and seawater
10 mL
o-SWCNTs (–)
Electrophoretic deposition
Pt wire
Mobile phase desorption
HPLC-DAD
82–97
0.32–0.52 g/L
F− , Cl− , Br− , NO3 − , SO4 2−
Distilled water
25 mL
o-SWCNTs (–)
Electrophoretic deposition
Pt wire
Water (200 L, potential assisted)
IC-ED
65–121
7 fluoroquinolone antibiotics
Urine and soil
15 mL (diluted urine) 2 g (soil)
o-MWCNTs (o.d.: 8–15 nm; i.d.: 3–5 nm)/MIPPy
Electrophoretic deposition and electropolymerization
Pt wire
MeOH/water/acetic HPLC-DAD acid 80:18:2 (v/v/v) (400 L, 15 min)
85–94 (urine), 90–96 (soil)
6 PAHs
Tap and well water
5 mL
o-MWCNTs (o.d.: 20–40 nm)
Electrodeposition
Stainless steel wire
Thermal (280 ◦ C, 15 min)
88–105
GC-FID
Comments
DI-SPME; comparison with PA fiber; life span: >120 extractions 0.06–0.26 g/L Electrochemically enhanced DI-SPME. Life span: 50 extractions 0.5–1.9 g/L (water Electrochemically samples) enhanced DI-SPME; comparison with PDMS fibers; life span: 50 extractions 0.03–0.07 g/L HS-SPME; life span: 30–40 extractions
Reference [160]
[161]
[138]
[60]
ACN: acetonitrile; CAR: carboxen; DAD: diode-array detector; DI-SPME: direct-immersion solid-phase microextraction; DPASV: differential pulse anodic stripping voltammetry; DVB: divinylbenzene; ECD: electron capture detector; ED: electrochemical detector; EDCs: endocrine disrupting compounds; FID: flame ionization detector; GC: gas chromatography; HS-SPME: head-space solid-phase microextraction; HPLC: high performance liquid chromatography; IC: ion chromatography; MeOH: methanol; MIPPy: molecularly imprinted polypyrrole; MS: mass spectrometry; NPs: nanoparticles; o-MWCNTs: oxidized multi-walled carbon nanotubes; o-SWCNTs: oxidized single-walled carbon nanotubes; PAHs: polycyclic aromatic hydrocarbons; PDMS: polydimethylsiloxane; PEG: poly(ethylene glycol); PoPD: poly-o-phenylenediamine; PPy: polypyrrole; UV: ultraviolet detector.
ARTICLE IN PRESS
Analytes
B. Socas-Rodríguez et al. / J. Chromatogr. A xxx (2014) xxx–xxx
Please cite this article in press as: B. Socas-Rodríguez, et al., J. Chromatogr. A (2014), http://dx.doi.org/10.1016/j.chroma.2014.05.035
Table 3 (Continued)
17
G Model CHROMA-355427; No. of Pages 37 18
ARTICLE IN PRESS B. Socas-Rodríguez et al. / J. Chromatogr. A xxx (2014) xxx–xxx
complementary techniques is often supplemented with studies of sorption capacity, including the obtaining of sorption isotherms or the development of kinetic and thermodynamic studies. 2.1.1. CNTs non-covalent functionalization Non-covalent functionalization involves the physical adsorption of molecules onto the sidewalls of CNTs and is governed by hydrophobic, van der Waals and/or electrostatic forces. Due the particular structure of CNTs, interaction with molecules with an aromatic system is particularly effective. For this reason, the formation of superstructures with biomolecules (as carbohydrates, proteins, enzymes and DNA), surfactants and polymers is very effective. In this sense, recent articles have reported the non-covalent modification of CNTs surface with polymers (such as polypyrrole (PPy) [19], polyethyleneimine [20], polydiphenylamine [21], poly(2-aminothiophenol) [22], polyaniline and poly(3,4-dioxythiophene) [23] or polyvinylalcohol [24]), ionic liquids (ILs, such as hexylmethylimidazolium hexafluorophosphate [HMIm][PF6 ] [25–27] or butylmethylimidazolium hexafluorophosphate [BMIm][PF6 ] [28]), biological structures (like biochar [29] or bacteria [30]) and other organic molecules (such as tartrazine [31], 4-(2-thiazolylazo)resorcinol [32], Aliquat 336 [33], poly(diallylmethylammoniumchloride) [34], di-(2ethylhexyl) phosphoric acid and tri-octyl phosphine oxide [35], tannic acid [36], or 5-(4 -dimethylamino-benzylidene)-rhodanine [37]). Compared to the chemical modification, this kind of functionalization could be operated under relatively mild reaction conditions maintaining the graphene-like structure of CNTs. Strategies adopted to modify CNTs with polymers include physical mixing in solution [20], in situ polymerization of monomers in the presence of CNTs [19,22,24,38] or surfactant-assisted processing of composites [21]. The combination of ILs and CNTs generates a soft material or gel with interesting properties which make them suitable for different analytical applications. For example, these gels are highly stable and they could be processed in almost any shape. Furthermore, they could retain their physical properties even under reduced pressures (due the negligible volatility of ILs) and facilitate electron transfer (due the large specific surface area of CNTs) [18]. The interaction mechanism between CNTs and ILs is believed to fall in the cation– and – interactions [18,27]. Regarding their use as sorbents, PoloLuque et al. [26] studied the sorption capacity of eight different soft materials obtained from the combination of [HMIm][PF6 ] with different types of CNTs (one type of single walled CNTs, SWCNTs, and seven types of multi-walled CNTs, MWCNTs, differing in dimensions). For this purpose, they mixed the suitable amount of CNTs and IL and ground them in an agate mortar with a pestle. The critical gel concentration was experimentally obtained since at higher concentrations the gel and the IL are clearly separated. Materials were characterized by Raman spectroscopy and their stability was also studied. Once all the materials were obtained, they were used to impregnate natural cotton fibers. Later, the same authors extended the methodology by using the same IL joined to coiled CNTs (pentagons and heptagons are incorporated in the structure) [27] or other MWCNTs [25]. Similarly Pourreza et al. [28] combined [BMIm][PF6 ] with MWCNTs by grinding in a mortar for about 1 h to impregnate pine sawdust. A very special application is the non-covalent functionalization of CNTs with bacteria [30,39]. For this purpose both living [30] and dead [30,39] Escherichia coli [30] or Pseudomonas aeruginosa [39] bacteria cells were mixed with CNTs and heated in an oven at a maximum temperature of 105 ◦ C. These sorbents were later used to extract several metals (Cd(II), Co(II), Cu(II), Ni(II) for E. Coli and Co(II), Cd(II), Pb(II), Mn(II), Cr(III), Ni(II) for P. aeruginosa) from water [30] or tomato leaves, bovine liver, boiled wheat, canned fish, black tea, lichen and natural waters [39]. Such manuscripts,
apparently anecdotal, demonstrate the potential of CNTs as supports for biosorption procedures. Although the introduction of an oxidation step is more common for covalent functionalization, it is also occasionally used for non-covalent modification to induce a more efficient interaction of CNTs with the modifier molecule. In this sense, Chen et al. [20] induced the creation of carboxylic groups to enhance electrostatic attraction between the positively charged protonated amines in the polyethyleneimine polymer and the carboxyl groups of oxidized-CNTs (o-CNTs) surface. Similarly, Tobiasz et al. [40] oxidized MWCNTs to enhance the interaction with 5dodecylsalicylaldoxime. 2.1.2. CNTs covalent functionalization Covalent functionalization of CNTs is relatively easy compared with other aromatic sp2 -bonded carbon structures such as graphite and graphene [18]. This fact is originated by the bending of the planar graphene sheet into a 3D carbon skeleton and the decrease in the orbital alignment [41]. Moreover, there should be a direct correlation between tube diameter and reactivity so that smaller CNTs are expected to be more reactive than bigger ones [42]. In the literature, reactions of halogenation, hydrogenation, radical addition, nucleophilic addition and cycloaddition with CNTs have been described [16,43]. This type of modifications involves the formation of a linkage between the functional groups and the carbon skeleton of nanotubes. As mentioned above, this phenomenon can occur by direct sidewall functionalization or indirect functionalization with carboxylic groups located on the defects of the carbon sheet. The first of them produces a change in carbon hybridization from sp2 to sp3 and, therefore, a loss of electronic conjugation, while the second takes place via the carboxylic groups at the open ends and holes of the sidewalls. Such carboxylic groups can be present on the as-grown CNTs; furthermore, they can be generated by putting CNTs through strong chemical conditions (usually with a strong acid like HNO3 , H2 SO4 and HCl or other oxidants such as H2 O2 , O3 , KMnO4 and NaOCl at high temperature) to introduce lactone, hydroxyl, carboxyl or carbonyl groups. Such reactions occur at a larger scale on their fullerene like tips, which are more reactive than their walls due the higher distortion of the planar framework of graphene [16]. Gas-phase oxidation mainly introduces hydroxyl and carbonyl surface groups, while oxidation in the liquid phase particularly increases the content of carboxylic acids [44,45]. The amount of carboxyl and lactone groups introduced on CNTs treated with HNO3 is higher than with other reagents [46]. Usually, the generated carboxylic groups are converted into acyl chloride and then undergo an esterification or amidation reaction [15]. It is worth mentioning that the use of some of the indicated oxidants, such as HNO3 , opens CNTs’ caps [47] and therefore, more surface area is available for either internal or external sorption. The principal drawback of covalent functionalization is that the graphitic structure of CNTs is disturbed, resulting in significant changes in their physical properties. In fact, it may even destroy the whole structure if functionalization is too high. A clear example for this issue is the recent work of Yang et al. [48] who studied the influence of different concentrations of NaOCl on the oxidation of MWCNTs with an outer diameter of 10–20 nm. Of the tested percentages of NaOCl (1, 3, 5 and 7%, w/v), the highest concentration destroyed CNTs structure and carbon appeared as larger agglomerated particles (see Fig. 1). These results were confirmed by TGA and FTIR analyses. In addition, they studied the influence of pH on z potentials and quantified the functional groups by Boehm’ titration. The adsorption capacity for Cu(II) was measured as well. Results demonstrated that adsorption increased when 5% (w/v) of NaOCl was used and fell with a 7% (w/v) of NaOCl. The oxidation process also provides a certain degree of clean-up of CNTs because metal impurities and amorphous carbon could be
Please cite this article in press as: B. Socas-Rodríguez, et al., J. Chromatogr. A (2014), http://dx.doi.org/10.1016/j.chroma.2014.05.035
Donor phase
CNTs type
Supporting substrate
Back-extraction
Determination technique
Recovery (%)
LODs
Comments
Reference
6 OPPs
Reservoir and drainage water, watermelon
o-MWCNTs (o.d.: 40–60 nm)
Stainless steel wire inserted in a PP HF impregnated by simple introduction of CNTs dispersed in 1-octanol
Desorption in MeOH (50 L, 5 min)
HPLC-DAD
83–101
0.1–0.3 g/L (water), 1.0–1.5 g/kg
HF-SPME combined with LLME
[61]
5 triazine herbicides
Tap, lake and drainage water, milk
15 mL of water or 15 g of the watermelon supernatant, pH 4–6, 5% (w/v) NaCl, 50 L chlorobenzene 15 mL, pH 6–8, 15% (w/v) NaCl, 30 L chlorobenzene
o-MWCNTs (o.d.: 40–60 nm)
Desorption in MeOH (50 L, 10 min)
HPLC-DAD
81–107
0.08–0.5 g/L
HF-SPME combined with LLME
[62]
5 carbamate pesticides
Apple
9 mL of the aqueous extract, pH 5.5, 16% (w/v) NaCl
MWCNTs (o.d.: 8–15 nm)
Desorption in MeOH (20 L, 25 min)
HPLC-DAD
92–113
0.09–6.00 ng/g
HF-SPME
[260]
2 alkaloids
Urine
10 mL, pH 14
MWCNTs (i.d.: 8–15 nm, o.d.: 30–50 nm)
Desorption in MeOH (80 L, 25 min)
HPLC-UV
91–94
0.7–0.9 g/L
HF-SPME
[261]
3 non-steroidal anti-inflammatory drugs
Urine
3 mL, pH 2
g-MWCNTs (o.d.: 10–20 nm)/PEG
Thermal desorption (280 ◦ C, 5 min)
GC-FID
8–12
0.03–0.07 g/L
HF-protectedSPME; comparison with HF-LPME and DI-SPME
[262]
Caffeic acid
Medicinal plants
15 mL of the aqueous extract, pH 2.5
o-MWCNTs (o.d.: 10–15 nm)
Retraction of acceptor phase (NaOH, pH 11, 6.5 L)
HPLC-UV
89–95
0.05 ng/L
Lumen filled with NaOH pH 11; comparison with HF-LPME
[63]
2 non-steroidal anti-inflammatory drugs
Plasma, urine, breast milk and wastewater
pH 7.4
o-MWCNTs (o.d.: 10–15 nm)
Stainless steel wire inserted in a PP HF impregnated by simple introduction of CNTs dispersed in 1-octanol Microsyringe needle inserted in a PP HF impregnated by simple introduction of CNTs dispersed in CTAB and water Microsyringe needle inserted in a PP HF impregnated by simple introduction of CNTs dispersed in CTAB and water PP HF impregnated in 1-octanol inserted in a PEG-g-MWCNTs coated fused-silica capillary PP HF impregnated by simple introduction of CNTs dispersed in 1-octanol PP HF impregnated by simple introduction of CNTs dispersed in 1-octanol
Retraction of acceptor phase (NaOH, pH 12, 20 L)
CE-UV
90–94
1–3 g/L
Lumen filled with NaOH pH 12; electrokinetic assisted extraction; comparison in the absence of CNTs and with HF-LPME
[263]
ARTICLE IN PRESS
Sample
G Model
CHROMA-355427; No. of Pages 37
Analytes
B. Socas-Rodríguez et al. / J. Chromatogr. A xxx (2014) xxx–xxx
Please cite this article in press as: B. Socas-Rodríguez, et al., J. Chromatogr. A (2014), http://dx.doi.org/10.1016/j.chroma.2014.05.035
Table 4 Recent applications of CNTs in membrane-based microextractions.
19
Sample
Donor phase
CNTs type
Supporting substrate
Back-extraction
Determination technique
Recovery (%)
LODs
Comments
Reference
Buprenorphine
Urine
4 mL, pH 2
o-MWCNTs (o.d.: 10–15 nm)
Retraction of acceptor phase (HCl, pH 2, 20 L)
CE-UV
92
1 g/L
MeOH
–
MWCNTs (–)
–
HPLC-UV
–
0.001–0.009 ng/L
Lumen filled with HCl pH 2; electrokinetic assisted extraction; comparison in the absence of CNTs Used for preconcentration of analytes from polar solvent
[237]
4 drugs
5 multiclass polar analytes
Water
200 mL
MWCNTs (–)
Retraction of acceptor phase (isopropyl acetate, 50 L)
HPLC-UV
–
–
Lumen filled with isopropyl acetate; comparison in the absence of CNTs and with o-CNTs
[265]
Phenobarbital
Wastewater
5 mL, pH 7
o-MWCNTs (o.d.: < 20 nm)
Desorption in MeOH (3 mL, 10 min)
HPLC-UV
102
0.32 g/L
Lumen filled with the CNTs/sol–gel solution
[64]
Diethylstilbestrol
Milk
1 mL, pH 3
o-MWCNTs (–)
Desorption in MeOH (50 L, 10 min)
HPLC-DAD
58–120
5.1 g/L
Lumen filled with 1-octanol
[65]
2 non-steroidal anti-inflammatory drugs
Tap and wastewater
10 mL, pH 3
o-MWCNTs (–)
PP HF impregnated by simple introduction of CNTs dispersed in 2-nitrophenyl octyl ether PP HF impregnated by simple introduction of CNTs dispersed in PVDF Nafion HF impregnated by simple introduction of CNTs dispersed in PVDF PP HF reinforced with CNTs/silica composite by sol–gel technology PP HF reinforced with CNTs/silica composite by sol–gel technology PP HF reinforced with CNTs/silica composite by sol–gel technology
Desorption in ACN (15 L, 25 min)
HPLC-DAD
70–114
0.40–4.48 g/L
[66]
3 phthalates
Juice, milk, carbonated drink, beer and wine Rice
1.0 mL, pH 3, 20% (w/v) NaCl
o-MWCNTs (–)
Desorption in diethyl ether (300 L, 10 min)
GC–MS
68–115
0.006–0.03 g/L
Lumen filled with 1-octanol; comparison with HF-SPME and HF-LPME Lumen remained empty
4 mL aqueous extract, pH 5, 1% (w/v) NaCl
o-MWCNTs (–)
DPASV
–
0.0073–0.025 g/L
Lumen filled with CNTs/sol–gel solution
[79]
20 mL of the aqueous extract, pH 8
o-MWCNTs (–)
Desorption in MeOH/HNO3 1:1 (v/v) (3 mL, 5 min) Desorption in MeOH (2 mL, 15 min)
HPLC-DAD
0.061–0.074 g/L
47–107
Lumen filled with CNTs/sol–gel solution
[67]
Rice, peanut and wheat
PP HF reinforced with CNTs by sol–gel technology
[266]
ARTICLE IN PRESS
2 aflatoxins
[264] B. Socas-Rodríguez et al. / J. Chromatogr. A xxx (2014) xxx–xxx
Pb(II), Cd(II), Cu(II)
PP HF reinforced with CNTs/silica composite by sol–gel technology PP HF reinforced with CNTs by sol–gel technology
G Model
Analytes
CHROMA-355427; No. of Pages 37
20
Please cite this article in press as: B. Socas-Rodríguez, et al., J. Chromatogr. A (2014), http://dx.doi.org/10.1016/j.chroma.2014.05.035
Table 4 (Continued)
G Model
CHROMA-355427; No. of Pages 37
Sample
Donor phase
CNTs type
Supporting substrate
Back-extraction
Determination technique
Recovery (%)
LODs
Comments
Reference
4 aromatic hydrocarbons
Wastewater and hair
o-MWCNTs (–)
Desorption in MeOH (400 L, 20 min) Desorption in MeOH (1 mL, 5 min)
74–87
0.5–0.7 ng/L
HPLC-DAD
–
–
1,2,4Trichlorobenzene, Cu(II)
Water
Crosslinking of CNTs/PVA suspension
–
GC-ECD GC–MS Spectrophotometry
–
–
Lumen filled with CNTs/sol–gel solution Lumen filled with CNTs/sol–gel solution; comparison with HF reinforced with SiO2 , nano TiO2 and MgO Comparison with PAC/PVA membrane; development of kinetics studies, batch sorption, batch uptake, and diaphragm cell breakthrough experiments
[68]
Wastewater and hair
PP HF reinforced with CNTs by sol–gel technology PP HF reinforced with CNTs by IL mediated sol–gel technology
GC-FID
6 multiclass pesticides
4.5 mL of the aqueous extract 5 mL of the aqueous extracts, pH 7
o-MWCNTs (o.d.: 10–15 nm), MWCNTs-NH2
SWCNTs
[267]
[268]
CE: capillary electrophoresis; CTAB: cetyltrimethylammonium bromide; DAD: diode-array detector; ECD: electron capture detector; FID: flame ionization detector; GC: gas chromatography; HF: hollow fiber; HF-SPME: hollowfiber solid-phase microextraction; HPLC: high performance liquid chromatography; LLME: liquid–liquid microextraction; MS: mass spectrometry; o-MWCNTs: oxidized-multi-walled carbon nanotubes; PAC: powdered activated carbon; PP: polypropylene; PVA: poly(vinyl alcohol); PVDF: polyvinylidene fluoride; UV: ultraviolet detector.
ARTICLE IN PRESS
Analytes
B. Socas-Rodríguez et al. / J. Chromatogr. A xxx (2014) xxx–xxx
Please cite this article in press as: B. Socas-Rodríguez, et al., J. Chromatogr. A (2014), http://dx.doi.org/10.1016/j.chroma.2014.05.035
Table 4 (Continued)
21
Sample amount
CNTs type
Stir-bar preparation method
Desorption
Determination technique
Recovery (%)
LODs
Comments
Reference
Pb (II), Cd (II)
Tap and river water
5 mL
o-MWCNTs (o.d.: 10–15 nm)
Sol–gel
1M HNO3 :MeOH 70:30 (v/v) (3 mL, 30 min)
DPASV-HDME
98–102
0.012–0.015 g/L
[173]
7 phenols
Lake water and soil
10 mL (water) 0.5 g (soil)
MWCNTs-DDM (–)
Sol–gel
MeOH/1 mM NaOH, 8:2 (v/v) (100 L, 25 min)
HPLC-UV
79–123% (water), 71.0–119.2% (soil)
0.14–1.76 g/L
Brilliant green
Fish pond water
5 mL
MWCNTs (o.d.: 10–15 nm)
Physical coating
MeOH (2 mL, 2 min)
UV/DAD
120
0.55 g/L
Ligand assisted pseudo-stir bar HF-SLME using sol–gel sorbent reinforced with CNTs; use of magnetic stoppers DDMMWCNTs/PDMS coated stir bar, comparison with PDMS and o-MWCNTs coated stir bar Pseudo-stir bar HF-SLME filled with MWCNTs dispersed in 1-octanol; use of magnetic stoppers
[175]
[174]
Matrix solid-phase extraction Analytes
Sample
Sample amount
CNTs type
Sorbent amount (mg)
Elution
Determination technique
Recovery (%)
LODs
Comments
Reference
9 OPPs
Apple, grape, banana, strawberry, tomato, cabbage, spinach and rape Butter
0.5 g
MWCNTs (i.d.: 40–60 nm)
1.0 g
MeOH (2.5 mL)
LC–MS/MS
71 –103
0.06 –0.15 g/kg
–
[271]
0.5 g
GMWCNTs (o.d.: 8–15 nm) MWCNTs (o.d.: 10–20 nm)
0.3 g GMWCNTs and 0.10 g MWCNTs
Acetone/ethyl acetate 1:1 (v/v) (20 mL)
GC–MS
85 –111
0.2 –1.3 g/kg
Derivatization with HFBA-ACN
[272]
8 hormones
ACN: acetonitrile; DAD: diode-array detector; DDM: 4,4 -diaminodiphenylmethane; DPASV: differential pulse anodic stripping voltammetry; GC: gas chromatography; GMWCNTs: graphitized multi-walled carbon nanotubes; HF-SLME: hollow-fiber solid–liquid microextraction; HFBA: heptafluorobutyric anhydride; HMDE: hanging mercury drop electrode; HPLC: high performance liquid chromatography; MeOH: methanol; MS: mass spectrometry; MWCNTs: multi-walled carbon nanotubes; UV: ultraviolet detector.
ARTICLE IN PRESS
Sample
B. Socas-Rodríguez et al. / J. Chromatogr. A xxx (2014) xxx–xxx
Analytes
G Model
Stir bar sorptive extraction
CHROMA-355427; No. of Pages 37
22
Please cite this article in press as: B. Socas-Rodríguez, et al., J. Chromatogr. A (2014), http://dx.doi.org/10.1016/j.chroma.2014.05.035
Table 5 Recent applications of CNTs as SBSE and MSPD sorbents.
G Model CHROMA-355427; No. of Pages 37
ARTICLE IN PRESS B. Socas-Rodríguez et al. / J. Chromatogr. A xxx (2014) xxx–xxx
23
Fig. 1. FE-SEM images for (a) raw CNTs, (b) 1%, (c) 3%, (d) 5% and (e) 7% of o-CNTs. Reprinted from [48] with permission of the Society of Chemical Industry.
eliminated. Such impurities removal depends on the intensity of the procedure [49]. However, certain features of CNTs (including the sorption capacity) could depend on the presence of the mentioned impurities. In this regard, Tian et al. [50] demonstrated that metal impurities dominate the sorption capacity of certain specific commercially available CNTs for Pb(II) in water. They acquired MWCNTs with an outer diameter of
