Very fast CO2 response and hydrophobic properties of novel poly(ionic liquid)s

Article abstract of DOI:10.1039/b912379b

A series of polymerizable tetraalkylammonium ionic liquids based on [2-(methacryloyloxy)ethyl]dimethylheptyl ammonium cation and bis(trifluoromethylsulfonyl)imide, nonafluoro-1-butanesulfonate, dodecylbenzenesulfonate, heptadecafluorooctanesulfonate, 4,4,

Full text of DOI:10.1039/b912379b

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
Very fast CO2 response and hydrophobic properties of novel
poly(ionic liquid)s†
Placido G. Mineo,a Letizia Livoti,b Marco Giannetto,c Antonino Gulino,a Sandra Lo Schiavob
b
*
and Paola Cardiano
Received 23rd June 2009, Accepted 18th August 2009
First published as an Advance Article on the web 19th October 2009
DOI: 10.1039/b912379b
A series of polymerizable tetraalkylammonium ionic liquids based on
[2-(methacryloyloxy)ethyl]dimethylheptyl ammonium cation and bis(trifluoromethylsulfonyl)imide,
nonafluoro-1-butanesulfonate, dodecylbenzenesulfonate, heptadecafluorooctanesulfonate,
4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,11-heptadecafluoroundecanoate anions, as well as their
corresponding homo- and copolymers, have been synthesized and characterized by means of 1H NMR,
TG-DTA, DSC, MALDI-TOF, viscosimetry and XPS investigations. Hydrophobic and CO2
sensing properties of the poly(ionic liquid)s have been explored by dynamic contact angle and quartz
crystal microbalance measurements. The CO2 sensing behavior of present polymers is very
remarkable as they are featured by extremely rapid and completely reversible response without any
memory-effect. Best results, in terms of sensitivity, have been obtained for [2-
(methacryloyloxy)ethyl]dimethylheptyl ammonium nonafluoro-1-butanesulfonate-based
homopolymer. Tensiometric data show good hydrophobic properties with qadv > 90ꢀ for all the
polymers under study except the one involving 4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,11-
heptadecafluoroundecanoate anion (qadv ¼ 78.3ꢀ); receding contact angles, representative of the most
hydrophilic portions of the polymers surface, lie in the range 22–54ꢀ and can be further improved
by choosing the proper long-chained N-alkyl groups.
and microgels synthesis for the entrapment of enzymes,19 nano-
Introduction
particles stabilization.20,21 As a new application of poly(ionic
liquid)s (pILs), we recently reported on the synthesis
and hydrophobic properties of a series of pILs obtained by
free-radical polymerization of monomeric methacrylic trime-
thylethylammonium-based ILs.22 Dynamic contact angle inves-
tigations showed that most of them are featured by very
promising advancing contact angles (qadv > 110ꢀ), while the
receding ones were very low (qrec < 20ꢀ), indicating that on the
surface very low and high energy domains are present.
Room temperature ionic liquids (ILs)1 are a class of compounds
exhibiting various unusual and attractive properties (i.e. very low
vapour pressure, large electrochemical window, wide liquid
phase range, thermal stability over a wide temperature range,
nonflammability, recyclability, high ion density, broad solvating
ability, environmentally friendliness, etc.). Their unique chemical
and physical properties, resulting from the proper combination
of organic cations and organic or inorganic anions, make them
absolutely unique and incomparable to other compounds, hence
their application is very broad as they have been used as reaction
media,2,3 catalysts,4 thermal fluids,5 energetic materials,6 fuel
cells,7,8 sensoristic applications,9–13 to name only a few. Despite
the huge amount of reports and patents concerning ILs, rela-
tively few papers have been published on ILs covalently bound to
a polymeric network and on their applications. These mainly
concern gas separation and sensoristic applications,14
polyelectrolyte15–17 and conductive flexible films synthesis,18 gels
As a continuation of the cited work, in this paper we report on
the synthesis of a series of novel poly(ionic liquid)s involving the
long-chained dimethylethylheptyl ammonium cation and the
bis(trifluoromethylsulfonyl)imide, nonafluoro-1-butanesulfonate,
dodecylbenzenesulfonate,
heptadecafluorooctanesulfonate,
4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,11-heptadecafluoroundecanoate
anions, entrapped into a methacrylic backbone. The new mate-
rials have been characterized by means of TG-DTA and DSC
analysis, as well as by solution NMR, MALDI-TOF and XPS
spectroscopy. The aim of this paper is twofold: on one hand, as
a continuation of the previous paper, to optimize the pILs
hydrophobic properties in terms of receding contact angles and
to clarify the role played by the cation on the hydrophobicity of
the resulting materials. Tensiometric measurements show that
the substitution of a methyl group on the quaternary nitrogen
with a heptylic one improves hydrophobic behavior of the
resulting films. The second aim, on the other hand, is to explore
the potentiality of the synthesized pILs as sorption materials for
CO2 sensing. It is well known that ionic liquids have remarkable
aDipartimento di Scienze Chimiche, University of Catania, Viale A. Doria 6,
95125 Catania, Italy
bDipartimento di Chimica Inorganica, Chimica Analitica e Chimica Fisica,
University of Messina, Salita Sperone 31 S. Agata, 98166 Messina, Italy.
E-mail: p.cardiano@unime.it
cDipartimento di Chimica Generale ed Inorganica, Chimica Analitica,
Chimica Fisica, University of Parma, Viale G. P. Usberti 17/A, Campus
Universitario, 43124 Parma, Italy
† Electronic supplementary information (ESI) available: Synthetic
procedures and characterization data. See DOI: 10.1039/b912379b
This journal is ª The Royal Society of Chemistry 2009
J. Mater. Chem., 2009, 19, 8861–8870 | 8861

View Article Online
CO2 sorption capacity and they have been found to behave as
non-volatile and reversible CO2 absorbents;9–13 in addition,
polymers prepared from ionic liquid monomers have shown even
higher CO2 sorption capability23 than the parent room temper-
ature ionic liquids, with faster response in terms of absorption/
desorption rates. It has been also found that ammonium-based
poly(ionic liquid)s have better CO2 sorption capacity than imi-
dazolium-based pILs.24,25 This paper reports also on quartz
crystal microbalance (QCM) investigations that clearly evidence
the remarkable CO2 sorption capabilities of the above new pILs
as well as the relationship between the sorption properties and
anion type.
measurements were performed on coated glass microscopy slides
(26 ꢂ 76 ꢂ 1 mm by Prestige) using the Wilhelmy method by
means of a KSV Sigma 700 tensiometer at the speed of 2 mm minꢁ1
in ultra pure water. The glass slides were accurately cleaned by
sulfuric acid–K-dichromate cleaning solution before use, then
they were coated by dip coating method immersing the plates in
7.5 wt% acetone solution (for homopolymers P1 and P2),
7.5 wt% chloroform solution (for homopolymer P3), 3.75 wt%
acetone solution and 7.5 wt% acetone solution (for copolymers
C1-4 and C1-5, respectively) at a constant rate (20 mm minꢁ1),
without delay time between immersion and withdrawn. After the
immersion cycles, the coated slides were aged for 10 days at 50 ꢀC
and then contact angle measurements have been carried out. The
contact angles correspond to the average of the measurements
performed on five glass slides treated in the same way. X-Ray
photoelectron spectra (XPS) were measured at 45ꢀ take-off angle,
relative to the surface plane, with a PHI 5600 Multi Technique
System (base pressure of the main chamber 2 ꢂ 10ꢁ10 Torr).
The P1 sample was not suitable for XPS analysis because of
prolonged sample out-gassing. The spectrometer is equipped
with a dual anode X-ray source; a spherical capacitor analyzer
(SCA) with a mean diameter of 279.4 mm; an electrostatic lens
system Omni Focus III. The nominal analyzer resolution was set
to 400 meV. Samples were mounted on Pt stubs. Spectra were
excited with Al-Ka radiation. Structures due to satellite Ka2
radiation have been subtracted from the spectra before the data
processing. The XPS peak intensities were obtained after Shirley
background removal.28 Procedures to account for steady state
charging effect have been described elsewhere.29 Samples were
left overnight in the XPS antechamber for sample out-gassing
before spectra collection. Experimental uncertainties in binding
energies lie within ꢃ0.4 eV. Some spectra were deconvoluted by
fitting the experimental profiles with a series of symmetrical
Gaussian envelopes, after subtraction of the background. The
agreement factor, R ¼ [S(Fo ꢁ Fc)2/S(Fo)2]1/2, after minimization
of the function S(Fo ꢁ Fc)2 converged to R values #0.035.30
Quartz crystal microbalance (QCM) experiments were carried
out on quartz/chrome/gold PQCs (piezoelectric quartz crystals)
with a fundamental frequency of 7.98 ꢃ 0.005 MHz, using a CHI
430 (CH Instrument, Texas) electrochemical workstation
supplied with QCM unit and managed by dedicated software.
The deposition of polymeric ionic liquids on the PQC units was
performed by simple drop-casting procedure. For this purpose
10 mL of the properly diluted solutions of P1–P3, C1-4 and C1-5
were deposed on the gold surface of the device and the solvent
was slowly allowed to evaporate under atmosphere saturated
with the solvent itself. The proper dilution factor of each casting
solution was determined starting from the concentrations
already employed for dip coating by progressive dilution until
reaching a suitable piezoelectric response. As a result, the dilu-
tion factors used for drop-casting solutions were 1 : 10 for P1 and
C1-4, and 1 : 20 for P2, P3 and C1-5. Retention properties of the
polymeric films towards CO2 were evaluated by exposing the
coated PQCs to a stream of a mixture of N2 (carrier gas) and
CO2. The composition of the stream conveyed to the detection
chamber was varied ranging from 100% of carrier gas to 100% of
CO2. For this purpose the inlet of the pure gases was set by
means of two flow regulators, while a flowmeter was employed in
order to assure a constant total flux (1.2 L minꢁ1) in the detection
Experimental
Materials
1-Bromoheptane,
[2-(methacryloyloxy)ethyl]dimethylamine,
lithium bis(trifluoromethylsulfonyl)imide, potassium non-
afluoro-1-butanesulfonate, sodium dodecylbenzenesulfonate,
potassium heptadecafluorooctanesulfonate, potassium 4,4,5,5,
6,6,7,7,8,8,9,9,10,10,11,11,11-heptadecafluoroundecanoate and
a,a0-azoisobutyronitrile (AIBN) were purchased from Sigma
Aldrich and used as received.
Measurement
1H NMR spectra were performed with a Bruker AMX R-300
spectrometer operating at 300.13 MHz. Thermogravimetric
analyses were performed by means of Perkin-Elmer Pyris Dia-
mond TG-DTA in the temperature range between 25 and 800 ꢀC,
under nitrogen or air atmosphere (50 mL minꢁ1) and with
a heating rate of 10 C minꢁ1. Differential scanning calorimetry
ꢀ
experiments were performed by using a TA Q100 instrument
equipped with a refrigerant cooling system (RCS) in the
temperature range between ꢁ90 and 50 ꢀC, with a heating rate of
10 ꢀC minꢁ1, under an anhydrous N2 atmosphere (60 mL minꢁ1).
The DSC measurements between 30 and 500 ꢀC were performed
with a Mettler DSC 20S, under an anhydrous N2 atmosphere
(60 mL minꢁ1), with a heating rate of 10 ꢀC minꢁ1. Inherent
viscosities of the polymers (hinh ¼ ln hr/c; c ¼ 0.5 g dLꢁ1) were
measured in a Desreux–Bischoff suspended level viscometer, in
DMSO (samples P1, P2 and C1-5) or DMF (samples P3 and C1-4,
see later) at 50 ꢀC. The positive MALDI-TOF mass spectra were
acquired by a Voyager DE (PerSeptive Biosystem) using a delay
extraction procedure (25 kV applied after 2600 ns with a poten-
tial gradient of 454 V mmꢁ1 and a wire voltage of 25 V) with ion
detection in linear mode.26 The instrument was equipped with
a nitrogen laser (emission at 337 nm for 3 ns) and a flash AD
converter (time base 2). To avoid fragmentation of the polymers,
the laser irradiance was slightly above threshold (ca. 106 W cmꢁ2).
The MALDI experiments were performed by loading on the
sample plate a 0.1 mmol sample and 40 mmol matrix trans-3-
indoleacrylic acid (IAA) or trans-2-[3-(4-tert-butylphenyl)-2-
methyl-2propenylidene]-malonitrile (DCTB), with DMSO
or DMF as solvent. Both 5,10-di(p-dodecanoxyphenyl)-15,20-
di(p-hydroxyphenyl) porphyrin (C68H78N4O4, 1014 Da), tetra-
kis(p-dodecanoxyphenyl)porphyrin (C92H126N4O4, 1350 Da)
and a PEG sample of known structure were used as external
standards for m/z calibration.27 Dynamic contact angle
8862 | J. Mater. Chem., 2009, 19, 8861–8870
This journal is ª The Royal Society of Chemistry 2009

View Article Online
cell. After the deposition of each coating the PQC unit was placed
in the detection chamber and the frequencimetric signal was
allowed to equilibrate under a constant flow of N2 (1.2 L minꢁ1).
When a satisfactory stability of the signal was reached, the
composition of the stream was progressively varied increasing
the percentage of CO2. The frequency shift was acquired for at
least 5 different levels of CO2, repeating the acquisition three
times for each explored concentration level. In order to evaluate
the sensitivity of the studied polymers the frequency shifts
were plotted versus the percent concentration of CO2 and linear
regression was carried out. The linearity of the response was
checked by means of Mandel test. Syntheses information as well
1
as H NMR and FT-IR data, elemental analyses and contact
angle results are available as ESI†.
Results and discussion
Bulk alkylation of [2-(methacryloyloxy)ethyl]dimethylamine by
1-bromoheptane affords, after 15 days, the expected product [2-
(methacryloyloxy)ethyl]dimethylheptylammonium bromide (M0)
with an acceptable yield (Fig. 1). The metathesis reactions of the
above precursor with the anions bis(trifluoromethylsulfonyl)imide,
nonafluoro-1-butanesulfonate, dodecylbenzenesulfonate, hepta-
decafluorooctanesulfonate and 4,4,5,5,6,6,7,7,8,8,9,9,10,10,
11,11,11-heptadecafluoroundecanoate, in water, lead to the ionic
liquids
[2-(methacryloyloxy)ethyl]dimethylheptylammonium
bis(trifluoromethylsulfonyl)imide (M1), [2-(methacryloyloxy)
Fig. 2 Synthesis of M1–M5 monomers.
ethyl]dimethylheptylammonium nonafluoro-1-butanesulfonate
(M2),
[2-(methacryloyloxy)ethyl]dimethylheptylammonium
(M3), [2-(methacryloyloxy)ethyl]
dodecylbenzenesulfonate
affected by the presence of a different anion, as already
observed;22 furthermore, the spectra of M3 and M5 are featured
by the presence of the additional arylic and alkylic resonances
due to the anions, with the correct integral ratio with respect to
the cation. FT-IR investigations carried out on M1–M5 defi-
nitely demonstrate that the exchange reactions have been
successful; in particular the anion exchange is clearly observed by
the appearance of new and typical bands corresponding to
bis(trifluoromethylsulfonyl)imide anion (1352, 1196, 1130 and
1057 cmꢁ1), to nonafluoro-1-butanesulfonate anion (1269, 1143,
1052 and 904 cmꢁ1), to dodecylbenzenesulfonate anion (2926,
2855, 1033 and 1007 cmꢁ1), to heptadecafluorooctanesulfonate
anion (1261, 1151, 1036 and 875 cmꢁ1) and to 4,4,5,5,6,6,7,7,8,8,
9,9,10,10,11,11,11-heptadecafluoroundecanoate anion (1571,
1405, 1370 and 875 cmꢁ1).
dimethylheptylammonium heptadecafluorooctanesulfonate (M4)
and [2-(methacryloyloxy)ethyl]dimethylheptylammonium 4,4,5,5,
6,6,7,7,8,8,9,9,10,10,11,11,11-heptadecafluoroundecanoate (M5),
with yields ranging from 46 to 98%, as described in ESI† (Fig. 2).
All the cited species have been characterized by means of
elemental analysis, 1H NMR solution spectroscopy, FT-IR,
DSC, and TG-DTA. 1H NMR investigations support the
alkylation as well as the exchange reaction; in particular the
spectrum of M0 clearly show the nitrogen quaternarization as
the signal at 3.37 ppm, ascribed to methylenic group adjacent to
Br in the parent compound 1-bromoheptane, is shifted to
3.58 ppm upon nitrogen alkylation. As expected, the resonances
of the other three N-alkyl groups are involved in a downfield shift
as well. As far as exchange reactions are concerned, all the
observed peaks strongly suggest the formation of M1–M5 species
although the proton signals attributable to the cations are poorly
Thermal properties of M2, M3 and M4, that are solids at room
temperature, and M1 and M5, that are viscous liquids, have been
Table 1 Thermal properties of monomers M1–M5
Residueb (%)
a
Sample
Tg/ꢀC
Tmelt/ꢀC
Tonset/ꢀC
Tmax /ꢀC
M1
M2
M3
M4
M5
ꢁ58.3
—
—
—
ꢁ64.2
—
82.2
—
95.7
—
336
314
276
308
181
390
361
343
372
217
0
1.2
1.5
2.5
0
a
b
Tmax: temperature of maximum rate of degradation. Residue at 800
ꢀC under N2 flow.
Fig. 1 Synthesis of M0.
This journal is ª The Royal Society of Chemistry 2009
J. Mater. Chem., 2009, 19, 8861–8870 | 8863

View Article Online
Fig. 4 Thermogravimetric traces of M0–M5 under nitrogen flow.
Fig. 3 DSC trace of M2.
ꢀ
upon heating, M5 starts to decompose at 181 C, at a tempera-
ture lower than the one displayed by M0 (Tonset ¼ 203 ꢀC)
(Fig. 4). Thermal degradation of M1 and M2 is featured by two
main endothermic processes displaying a comparable rate of
volatilization herewith found to range between 12 and 18% minꢁ1.
On the contrary, monomers M3 and M4 show a main decom-
position process (Tmax 343 and 372 ꢀC, respectively) occurring at
a higher rate with a huge mass loss, while the other thermal
effects are quite slower. M5 decomposes under nitrogen with two
closely occurring endothermic effects; the main one, character-
ized by a high volatilization rate (22% minꢁ1) with a weight loss
of 51%, culminates at 217 ꢀC. From the comparison of the data
collected in Table 1, it can be stated that M5, among the studied
species, is the most vulnerable on heating; this evidence suggests
that the presence of the carboxylate anion can be responsible for
the above behavior. The hypothesis that M5 may undergo
decarboxylation during thermogravimetric scanning, as already
observed for other ionic liquids,31 has been confirmed by col-
lecting FT-IR spectra on the solid after heating at 120, 170 and
220 ꢀC. These investigations confirm that decarboxylation occurs
as the signal at 1571 cmꢁ1, due to the COOꢁ moiety, disappears
after the heating at 220 ꢀC. Moreover, thermogravimetric
investigations carried out in oxidative conditions do not show
notable differences with those recorded under nitrogen, as
already observed.32
explored by DSC and TG-DTA investigations. Thermal prop-
erties of monomers are summarized in Table 1. DSC scannings,
under nitrogen atmosphere, have been carried out on M1–M5
thus allowing us to explore their phase transitions. In particular,
in order to investigate the compounds already liquid at room
temperature, the scanning has been performed not only by
continuous heating (30 toꢀ500 ꢀC) but also by heating and
cooling cycles (ꢁ90 to 150 C). In the latter temperature range
M0, M1 and M5 do not melt; on the contrary, at low tempera-
tures, they are featured by a solid amorphous phase that shows
a glass transition at ꢁ58.3, ꢁ64.2 and ꢁ34.5 ꢀC, respectively. In
addition the observed glass transition temperatures (Tg) are not
followed by any other thermal transition indicating that the cited
compounds are not able to crystallize upon cooling. Straight
ꢀ
DSC scanning, starting from 30 C, on M2 shows a melting at
92 ꢀC. At the same time, whenever the sample is cooled and
heated from ꢁ90 ꢀC up to 150 ꢀC (Fig. 3), no Tg is detectable but
an exothermic effect, due to a reorganization of the solid in a crys-
talline material, is distinguishable at 63.4 ꢀC (DHricr ¼ 13.69 J gꢁ1);
the cited crystalline solid so-formed further undergoes melting at
82.2 ꢀC (DHmelt ¼ 21.54 J gꢁ1). For this compound very likely the
ꢀ
glass transition occurs at a temperature less than ꢁ90 C. The
ordered nature of solid M2 has been confirmed by means of
XRD (trace not reported) clearly showing sharp peaks due to
crystalline phases. Compound M3 does not show any
thermal transition in the studied range. Moreover, M4 displays
a melting endothermic signal at 95.75 ꢀC corresponding to
a DHmelt ¼ 44.16 J gꢁ1 but no Tg in the whole investigated
temperature range. XRD analysis on such a monomer confirmed
a high degree of crystallinity.
Free-radical homopolymerization of monomers M1–M3 in
bulk or in CHCl3, following the work-up method described
in ESI†, leads to the formation of the homopolymers poly[2-
(methacryloyloxy)ethyl]dimethylheptylammonium bis(trifluoro-
methylsulfonyl)imide
(P1),
poly[2-(methacryloyloxy)ethyl]
dimethylheptylammonium nonafluoro-1-butanesulfonate (P2)
and poly[2-(methacryloyloxy)ethyl]dimethylheptylammonium
dodecylbenzenesulfonate (P3), with yields of about 75–80%.
As homopolymerization of M4 and M5 employing the above
experimental conditions led to insoluble materials, thus
hindering their structural characterization as well as the evalua-
tion of their hydrophobic properties and CO2 sorption capacity,
copolymerization reactions by employing 1 : 1 ratio of M1 and M4
as well as of M1 and M5 were carried out in order to obtain soluble
copolymers. In particular, the cited reactions in CHCl3 with AIBN
as free-radical initiator afford poly[2-(methacryloyloxy)
TG-DTA investigations allowed us to explore the thermal
behavior of M1–M5 either in air or under nitrogen flow. As
already observed, thermal behavior of M1–M5 is largely influ-
enced by the nature of the anion. In particular, thermogravi-
metric analyses under inert atmosphere show the first
degradation step, strictly connected to the thermal stability of the
studied species, whose temperature follows the M1 > M2 z M4
> M3 > M5 trend. The mentioned processes occur with differ-
ences in the rate of volatilization, being much faster for M1, M2
and M5. By comparing the above data with the thermal stability
of the parent M0, it appears that while M1–M4 are more stable
ethyl]dimethylheptylammonium
bis(trifluoromethylsulfonyl)
imide-co-[2-(methacryloyloxy)ethyl]dimethylheptylammonium
8864 | J. Mater. Chem., 2009, 19, 8861–8870
This journal is ª The Royal Society of Chemistry 2009

View Article Online
heptadecafluorooctanesulfonate (C1-4) and poly[2-(methacryloy-
loxy)ethyl]dimethylheptylammonium bis(trifluoromethylsulfonyl)
imide-co-[2-(methacryloyloxy)ethyl]dimethylheptylammonium 4,4,
5,5,6,6,7,7,8,8,9,9,10,10,11,11,11-heptadecafluoroundecanoate
(C1-5) with good yields; the above copolymers display a slight
solubility in acetone only. Both homopolymers as copolymers
is lower than the one displayed under nitrogen; P2 and P3 show
in fact the first degradation step at Tonset 234 and 204 ꢀC,
respectively. As far as copolymers C1-4 and C1-5 are concerned,
their thermal behavior is mainly influenced by the presence of M1
meaning an increased stability of the cited copolymers towards
heating with respect to M4 and M5. The stabilization effect
against heating due to the presence of M1 occurs also for C1-5
that starts to degrade at Tonset 201 ꢀC. The investigations carried
out in oxidative environment on C1-5 show a comparable
behavior to that displayed under nitrogen atmosphere while C1-4
starts to decompose at 219 ꢀC, a temperature considerably lower
than the first thermal event occurring in nitrogen. DSC analysis
carried out on polymeric samples shows the presence of glass
transition for P1 (ꢁ44 ꢀC) and C1-5 (ꢁ23.9 ꢀC) only. The higher
Tg observed for C1-5 with respect to the one showed by P1 can be
ascribed to the presence of the repetitive unit M5 into the poly-
meric network that may induce a perturbation of the amorphous
structure of the homopolymer.
1
have been characterized by means of solution H NMR, FT-
IR, TG-DTA, DSC, MALDI-TOF, viscosimetry and XPS
measurements. In particular in NMR spectra the olefinic protons
due to the methacrylic moiety are hardly detectable after poly-
merization. In addition, in FT-IR spectra no signals at 988 and
1638 cm–1 due to C]C bonds are observable, while CO
stretchings of the polymers P1, P2, P3 and C1-4 are shifted
towards a higher frequency with respect to the corresponding
monomers. On the contrary, for C1-5 copolymer, CO stretching
is featured by an intermediate frequency between those displayed
by the monomers.
Thermogravimetric investigations for homo- and copolymers,
both under nitrogen and air flow, have been carried out. Fig. 5
shows the thermogravimetric traces of polymers in nitrogen
atmosphere. By comparing TGA data of M1–M3 with the
behavior of the corresponding homopolymers P1–P3, it appears
that the thermal stabilities of the discussed species do not show
significant variations upon polymerization (Table 2). For these
polymers PDT decreases following the trend P1 > P2 > P3 with
the corresponding Tmax of 378, 361 and 346 ꢀC as well as weight
losses of 74, 84 and 52%, respectively. Thermogravimetric
patterns for P1–P3 are quite similar being characterized by two
main steps. Thermooxidative stability (data not shown) of P1–P3
The poor solubility of the materials under study in THF
(solvent used as eluent in our GPC system) as well as the strong
ionic interactions between these products and the stationary
phase hinder the macromolecules elution thus hampering GPC
investigations. Therefore the macromolecular nature of the
homo- and copolymers has been ascertained by means of vis-
cosimetry, MALDI-TOF and XPS experiments. The viscosity
experiments suggest the polymeric nature of the studied samples.
In fact, homo- and copolymers show hinh values lying in the 4–0.5
range (Table 2). The characterization of polymers was also per-
formed through MALDI-TOF spectra that show signals
consistent with the presence of oligomeric poly(ionic liquid)s. In
particular, the positive MALDI-TOF mass spectrum of P1
(Fig. 6) is featured by two series of peaks: at m/z value of 1394 +
(n ꢂ 535) (+) with n ¼ 0–7, due to oligomers (from trimers to
decamers) bearing isobutyronitrile and vinyl (and/or hydrogen)
as end groups (arising from chain transfer phenomena); at m/z
value of 1462 + (n ꢂ 535) (++) with n ¼ 0–7, due to oligomers
(from trimers to decamers), detected as M+, with isobutyronitrile
fragments as end groups (deriving from chain coupling
phenomena). Furthermore, the ions of (+) and (++) are detected
as M+ species due to loss of a [N(SO2CF3)2]ꢁ fragment from the
corresponding oligomeric species.
The mass spectrum of P2 (Fig. S1, ESI†) consists essentially of
two series of peaks, detected as M+, at m/z values 1921 + (n ꢂ 555)
Fig. 5 Thermogravimetric traces of P1–P3, C1-4 and C1-5 under
nitrogen flow.
Table 2 Properties of homopolymers P1–P3 and copolymers C1-4 and
C1-5
Sample
hinh
Tg/ꢀC
Tonset/ꢀC
PDTa/ꢀC
Residueb (%)
P1
P2
P3
C1-4
C1-5
4.00
4.06
1.17
0.76
0.46
ꢁ44.0
—
—
—
ꢁ23.9
333
291
284
317
201
378
361
346
386
219
1.4
1.8
3.3
0
2.9
a
PDT: polymer degradation temperature, temperature of maximum rate
of degradation. b Residue at 800 ꢀC under N2 flow.
Fig. 6 MALDI-TOF mass spectrum of P1.
J. Mater. Chem., 2009, 19, 8861–8870 | 8865
This journal is ª The Royal Society of Chemistry 2009

View Article Online
Table 4 XPS derived atomic concentration analysis for C1-4
with n ¼ 0–6 and 1929 + (n ꢂ 555) with n ¼ 0–6, respectively.
Both the ion species are detected as M+ species due to loss of
a [C4F9SO3]ꢁ fragment from the corresponding oligomeric
species. It is worth noticing that the first series of peaks corre-
sponds to structures relative to oligomers (from tetramers to
decamers) having hydrogen and vinyl as end groups, while the
latter corresponds to chains (from tetramers to decamers) having
an isobutyronitrile and hydrogen as end groups. Little signals
due to oligomers having isobutyronitrile fragments as end groups
(deriving from chain coupling phenomena) are also present. The
presence of oligomeric species which do not contain iso-
butyronitrile as end groups, although unexpected, can be
ascribed to thermal polymerization of M2, in analogy with other
methacrylate systems,33 which undergo spontaneous self-poly-
merization via a biradical growth mechanism. In order to verify this
hypothesis a thermal bulk polymerization of M2 (ꢄ95 ꢀC, 60 min)
without employment of chemical initiator was attempted.
Working up of the final mixture led to a low amount of product
(ca. 10%), which was analyzed by MALDI-TOF mass spec-
trometry. This affords spectra displaying a series of enveloped
signals indicative of oligomer formation (from dimer to pen-
tamer) and having the same structure of species at m/z values
1921 + (n ꢂ 555) and detected as M+ (spectrum omitted for
brevity).
XPS atomic concentration (%), functional
group
Shell, B.E./eV
C 1s, 285.0
C 1s, 286.6
C 1s, 288.9
C 1s, 291.5
C 1s, 293.7
O 1s, 532.4 broad
N 1s, 339.4
N 1s, 402.7
S 2p, 168.3
F 1s, 689.6
23.0
13.6
4.0
7.4
1.5
13.1
1.1
2.2
2.6
18 (CH2, CH3)
8 C–N+, 2 CO, 2 COR
2 CF3–S, 1 CF2–S
1 –CF2–CF2–CF2–CF2–CF2–CF2–
1 –CF3
11 O
1 Nꢁ
2 N+
3 S
31.5
23 F
Table 5 XPS derived atomic concentration analysis for C1-5
XPS atomic concentration (%), functional
group
Shell, B.E./eV
C 1s, 285.0
C 1s, 286.6
C 1s, 288.9
C 1s, 291.7
23.2
11.7
3.2
20 (CH2, CH3)
8 C–N+, 2 CO, 2 COR
2 CF3–S, –1 COOꢁ
1 –CF2–CF2–CF2–CF2–CF2–CF2–
8.7
CF2
1 –CF3
10 O
1 Nꢁ
2 N+
2 S
C 1s, 293.6
1.8
10.3
1.4
2.6
1.8
O 1s, 530.8–533.9
N 1s, 339.3
N 1s, 402.7
S 2p, 168.5
The mass spectrum of P3 (spectrum not reported for brevity)
shows signals due to two series of peaks (from trimers to deca-
mers) assigned to oligomers bearing isobutyronitrile and vinyl
(and/or hydrogen) as end groups (m/z 1486 + (n ꢂ 581)), and
with isobutyronitrile fragments as end groups (m/z value of
1554 + (n ꢂ 581)). The ions are detected as M+ species due to loss
F 1s, 689.7
34.3
23 F
ꢁ
of a CH3(CH2)11Ph–SO3 fragment, from the corresponding
Probably, the absence of oligomeric species as well as the
impossibility to be desorbed/ionized hampered the MALDI-
TOF analysis of C1-4 and C1-5 samples. Therefore, copolymeric
compositions were obtained by XPS experiments. Tables 3–5
collect XPS atomic concentration data for samples P2, C1-4 and
C1-5, once the relevant atomic sensitivity factors have been
accounted for.36–38 It is evident that concentration results always
match the polymer formula composition. P2 has been used as
a blank test and Table 3 shows the obtained XPS atomic
composition. It shows a N/S atomic ratio ¼ 1, a F/S ratio of 9 : 1
and a C/F ratio of 19 : 9, as expected on the basis of the formula
oligomeric species.
It is worth mentioning that the MALDI-TOF signals detected
for the pILs under study arise from the oligomeric species,
whereas the signals due to the rest of the polymeric chains cannot
be visible due to an intrinsic limitation of the technique
employed. In particular, the MALDI-TOF analysis of high
polydisperse polymers (MWD > 1.2) suffers from some instru-
mental and experimental limitations:34,35 (i) the decrease of
sensitivity and resolution by increasing the ion molecular mass;
(ii) the decrease of ion abundance when investigating higher
molecular mass distribution (i.e. MALDI-TOF technique gives
signals proportional to the molar abundance of the revealed
polymeric chain).
Table 3 XPS derived atomic concentration analysis for P2
XPS atomic concentration (%), functional
group
Shell, B.E./eV
C 1s, 285.0
C 1s, 286.7
C 1s, 289.0
C 1s, 291.3
C 1s, 293.7
O 1s, 532.2
O 1s, 533.9
N 1s, 402.7
S 2p, 168.1
F 1s, 689.6
25.8
17.4
3.0
5.8
2.6
11.2
3.0
3.0
2.8
9 (CH2, CH3)
4 C–N+, 1 CO, 1 COR
1 CF2–S
1 –CF2–CF2–
1 –CF3
4 O
O (COR)
N+
1 S
Fig. 7 Al-Ka excited XPS of P2 at 45ꢀ electron take-off angle: (a) in the
25.4
9 F
O 1s energy range; (b) in the C 1s energy range.
8866 | J. Mater. Chem., 2009, 19, 8861–8870
This journal is ª The Royal Society of Chemistry 2009

View Article Online
composition. In addition, on the basis of the binding energies, it
is also possible to distinguish between two kinds of oxygen and
five kinds of carbon atoms (Fig. 7). In particular, the three
sulfonate oxygens (–SO3ꢁ) and the carbonyl oxygen (–CO–) lie at
532.2 eV whilst the ester oxygen (–CO–O*–R) lies at 533.9 eV.38,39
Accordingly, deconvolution of the experimental profile gave
a 4 : 1 area ratio, in agreement with this assignment. Deconvo-
lution of the rich C 1s spectrum reveals five components with an
intensity ratio of 9 : 6 : 1 : 2 : 1. According to Table 3, both
binding energies and atomic concentration analysis allow for
a straightforward spectrum interpretation. In fact, ionizations of
the –CH2 and –CH3 backbone lie at 285.0 eV and add up for 9
carbon atoms; the –C–N+, –CO, and –COOR ionizations, in tune
with literature data, fall at 286.7 eV and count for 6 carbon
atoms; the –CF2–S group lies at 289.0 eV; ionization of the
–CF2–CF2– function lies at 291.3 eV and, finally, that of the CF3
group is at 293.7 eV. The sharp N 1s peak lies at 402.7 eV as
expected for the positive N+ function.30,38–40
on the particular charge state thus ranging from 530.8 eV for the
–SO2– groups to 534.0 for the –COOR groups. The fluorine 1s
component shows a broad peak at 689.6–689.7 eV. Finally, the S
2p states show a peak in the 168.1–168.5 eV energy range.
CO2 sensing
On the basis of literature data, the study of the sensoristic appli-
cations of monomeric and polymeric ionic liquids based on tet-
ralkylammonium cations is particularly interesting considering
their capabilities for CO2 absorption.24,25 It is well established that
the affinity of ILs for CO2 is related to their structure in terms of
cations, anions and substituents, although the nature of the
anionic moiety seems to play a major role.41–44 Bearing this in
mind, the potentiality of the synthesized polymers as sensing
materials for CO2 has been explored by means of QCM
measurements. As the polymers under study are featured by
a common cationic part and five different anionic counterparts, we
have explored the variation of CO2 absorption by changing the
anion type. It is worth noticing that the studied anionic moieties
differ for functional group itself (imide, sulfonate and carboxylate)
as well as for the fluorination degree of the alkyl chain, if present.
The frequencimetric responses of the coating obtained by the five
studied polymers were compared in order to investigate the rela-
tions between structure and retention properties.
Tables 4 and 5 show XPS atomic concentration analyses and
ionization binding energies for C1-4 and C1-5 copolymers. XPS
results evidence that the obtained copolymer compositions
parallel the starting equimolar monomers (M1 + M4 or M1 + M5)
reaction mixtures. In fact, the atomic F/S/N/O/C ratios are 23 : 3 :
3 : 11 : 40 and 23 : 2 : 3 : 10 : 43 for C1-4 and C1-5, in tune with the
expected 50 : 50 monomer composition. As already observed for
the P2 blank test, a careful deconvolution of the C 1s band
envelope (Fig. 8a) reveals for both C1-4 and C1-5 copolymers
five components: the first centered at 285.0 eV is due to
the aliphatic –CH2– and –CH3 backbone. The second at
286.6–286.7 eV is due to the carbon of the C–N+, –C]O and
–C*OOR groups. The third band centered at 288.9–289.0 eV is
consistent with ionization energies of CF2–S and, in the case of
the C1-5 sample, with the –COOꢁ groups. The fourth component
at 291.3–291.7 accounts for the carbon centers of the CF2–
groups and, finally, the fifth component at 293.6–293.7 eV
belongs to the –CF3 moiety.38 Spectra of the nitrogen 1s core
level show binding energies in tune with the formal charge of the
nitrogen groups (Fig. 8b). Therefore, quaternized N+ gives an
XPS peak at 402.7 eV whilst negatively charged nitrogen shows
a peak at 339.3–339.4 eV. These features nicely account for the
two kinds of nitrogen depending on the particular sample. The
O 1s signal (Fig. 8c) lies at binding energies strongly dependent
Fig. 9 Frequencygram recorded over progressive increase of the CO2
percentage in the mixture conveyed to the QCM cell for P2.
Fig. 8 Al-Ka excited XPS of C1-5 at 45ꢀ electron take-off angle: (a) in the C 1s energy range; (b) in the N 1s energy range; (c) in the O 1s energy range.
This journal is ª The Royal Society of Chemistry 2009 J. Mater. Chem., 2009, 19, 8861–8870 | 8867

View Article Online
Table 6 Sensitivity values for homo- and copolymers exposed to
different CO2 concentrations
Sensitivity/Hz (percent
concentration)ꢁ1
Sample
P1
P2
P3
C1-4
C1-5
0.629 ꢃ 0.007
1.56 ꢃ 0.02
0.88 ꢃ 0.02
0.40 ꢃ 0.01
0.53 ꢃ 0.01
procedures. Nevertheless the advantage in terms of response time
is even more evident, considering that, in the study concerning
the use of tetramethylammonium fluoride as sensing coating,
response times ranging from 15 min to 5 h are reported.
Fig. 10 Relationship between the frequency shift and the CO2
percentage, obtained by coating the PQC with pIL P2.
As far as the polymers under study are concerned, best results,
in terms of sensitivity, were obtained with P2, as established from
the comparison among the responses displayed by P1–P3 as well
as C1-4 and C1-5 (data not shown). The sensitivity values
obtained with all studied polymers are summarized in Table 6. It
is worth noticing that the best response of P2 was obtained
although the dilution of the casting solution (1 : 20) was higher
with respect to the factor employed for P1 (1 : 10). These results
indicate that the retention properties of polymer P2 are
remarkably higher if compared with the performance of the other
studied compounds. Furthermore, the comparison between the
responses shown by copolymers C1-4 and C1-5, only differing
for the co-monomers M4 and M5, does not evidence appreciable
differences in terms of CO2-philicity. Monomers M4 and M5
are associated by analogous polyfluorinated chains and differ for
the anionic functional group (carboxylate vs. sulfonate). The
comparability of the sensing properties evidenced by the corre-
sponding coatings indicates that the nature of the anionic func-
tional group does not influence the CO2-philicity of the
polymeric ionic liquids based on polyfluorinated chains.
Conversely, a comparison between the responses shown by
coatings P2 and P3 shows remarkable differences in terms of
sensitivity, referable to the CO2-philicity. Particularly, P2
showed a sensitivity almost double with respect to P3, so indi-
cating the fundamental role played by the nature of the alkylic
chain. Homopolymers P2 and P3 are, in fact, featured by the
same anionic moiety (sulfonate), with different alkylic chains,
unfluorinated for P3 and completely fluorinated for P2. This
observation is supported by the findings of other studies,42,48
aimed to investigate about the CO2-philicity of differently fluo-
rinated ionic liquids. As previously stated, the properties of
homopolymers P4 and P5 cannot be directly investigated
because of their insolubility. For this reason the coatings
obtained with copolymers C1-4 and C1-5 have been character-
ized. The role played by the extension of the polyfluorinated
chains (P2 vs. P4) cannot be investigated for the same reason. A
comparison between the retention properties of the polymers
with respect to the parent monomers was also attempted, but the
QCM measurements could not be performed since, after the
deposition of the monomer coatings, the stability of the fre-
quencimetric signal was not reached under a constant flow of N2
(1.2 L minꢁ1). This behavior is probably ascribable to progressive
stripping of the monomeric coatings caused by the flow of the
carrier gas (bleeding).
In order to evaluate the repeatability and the reversibility of
the responses, the coated piezoelectric quartz crystals (PQCs)
were alternately exposed to N2–CO2 mixtures and to pure carrier
gas. All the polymeric coatings showed excellent responses
featured by quickness and reversibility without any memory-
effect. A quick and complete restoring of the baseline signal was
always obtained, with all coatings, when pure carrier gas was
conveyed to the detection chamber. It is worth mentioning that
the response time was, in all cases, lower than a second, a value
limited by the actual time required for adjusting the gas flows.
This behavior is easily visible as shown by the responses obtained
during the characterization of P2, reported as an example in
Fig. 9. The calibration line obtained by interpolating the
frequencimetric data is reported in Fig. 10. In order to estimate
the exceptional rapidity of the response, a comparison of the
performance of P2 with the literature data available for other
coatings used for the development of piezoelectric sensors,
whether or not based on ionic liquids, was performed. Baltus and
co-workers41 performed a QCM study on the CO2-philicity of 1-
n-butyl-3-methylimidazolium bis[trifluoromethylsulfonyl]amide
and other related compounds. From the data reported in the
paper it is possible to calculate both the detection limit, using
the regression line,45 and the sensitivity. The first value is
comparable to that obtained with P2 (6%) under similar exper-
imental conditions. The above detection limit values, in spite of
different sensitivities, indicate that, even if the frequency shifts
obtained by Baltus et al.41 are lower than those obtained with P2,
the signal-to-noise ratio is comparable. Conversely, a noticeable
advantage offered by P2 lies in the response time which, as
already stated, is in the order of a few seconds while in the case of
the above mentioned study the stability of the signal is achieved
about 3 min after adjusting the flow of CO2 and N2. In our
opinion this aspect is fundamental for the realization of real-time
responding sensors.
Concerning other materials employed for the realization of
CO2 responding piezoelectric devices, Gomes and co-workers
reported about the sensing properties of coatings based on
N,N,N0,N0-tetrakis 2-hydroxyethylethylenediamine46 or tetra-
methylammonium fluoride.47 A direct comparison between the
sensitivity of these sensors and the one displayed by the coatings
under study cannot be performed because of the different
experimental conditions employed for the calibration
8868 | J. Mater. Chem., 2009, 19, 8861–8870
This journal is ª The Royal Society of Chemistry 2009

View Article Online
Table 7 Dynamic contact angles measured by Wilhelmy balance on
coated glass slides
fluorooctanesulfonate, 4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,11-hep-
tadecafluoroundecanoate have been synthesized and characterized.
Our previous investigation on the evaluation of hydrophobic
properties of tetraalkylammonium poly(ionic liquid)s was very
promising; as a result, the current work has been focused on
assessing the variation of such hydrophobic behavior of ionic
liquid-based polymers obtained by substituting a methyl group
on the quaternary nitrogen with an heptylic one. This strategy
allowed us to improve the receding contact angles. Furthermore,
as ILs and pILs are featured by a considerable CO2-philicity, the
potential applications of the synthesized polymers as sensing
materials for CO2 have been explored by means of QCM
measurements. Regardless of the structure–property relation-
ship, the studied polymeric coatings showed very promising
properties for sensor applications. Frequencimetric responses
were very rapid, reversible and no memory-effect occurred. The
quickness of the response is a fundamental quality parameter of
chemical sensors: P2-based sensors showed a response time lower
than a second, limited by time required for adjusting the N2 and
CO2 flows. Considering also the cheapness of the materials and
the simplicity of the coating procedure, these materials are
excellent candidates for development of sensors.
ꢀ
ꢀ
qrec/
Sample
qadv
/
P1
P2
P3
C1-4
C1-5
97.3 ꢃ 0.7
108.8 ꢃ 0.4
100.0 ꢃ 0.6
113.2 ꢃ 0.8
78.3 ꢃ 0.9
47.6 ꢃ 1.4
49.0 ꢃ 1.3
54.0 ꢃ 0.9
42.0 ꢃ 1.5
22.2 ꢃ 0.3
Hydrophobic properties of homo- and copolymers
Hydrophobicity and wettability properties of polymeric
compounds P1–P3, C1-4 and C1-5 have been investigated by
means of dynamic advancing (qadv) and receding (qrec) contact
angle measurements, using Wilhelmy method. Advancing and
receding contact angles, as well as the difference between the two
cited values (hysteresis), are essential to evaluate the hydro-
phobic behavior of films. Dynamic contact angle values,
collected in Table 7, have been obtained from measurements
performed on microscope glass slides properly coated by films of
P1–P3, C1-4 and C1-5. From these data it can be inferred that
P1, P2, P3 and C1-4 display a hydrophobic behavior, being their
qadv > 90ꢀ, providing a trend that follows P1 < P3 < P2 < C1-4.
On the contrary, C1-5 shows a less hydrophobic behavior, with
a qadv of 78.3ꢀ. The most remarkable hydrophobic attitude has
been displayed by P2 and C1-4 coatings with advancing contact
angles of 108.8 and 113.2ꢀ, respectively. The tensiometric inves-
tigations on P2 and C1-4 also show receding contact angles of
49ꢀ and 42ꢀ, representative of the more hydrophilic portion of
the considered coatings. Molecular chains of C1-4, in particular,
are featured by an intrinsic heterogeneity, due to the monomer
differences so that the complexity of the organization of side
chains may rise leading to a growing hysteresis value. The high
hysteresis experienced for all the studied films may be explained
in terms of surfaces heterogeneity where hydrophobic and
hydrophilic microdomains coexist. As an example, for C1-5 the
low values of both qadv and qrec suggest the preferential orien-
tation of hydrophilic domains at the solid–air interface, with the
consequent reduction of the coating hydrorepellent activity. It is
worth mentioning that by comparing the contact angle values of
References
1 P. Wasserschied and W. Keim, Angew. Chem., Int. Ed., 2000, 39,
3772–3789.
2 J. Dupont, R. F. De Souza and P. A. Z. Suarez, Chem. Rev., 2002,
102, 3667–3692.
3 P. Wasserscheid and T. Welton, Ionic Liquid in Synthesis, Wiley-
VCH, Weinheim, 2003.
4 T. Welton, Chem. Rev., 1999, 99(8), 2071–2084.
5 M. E. Van Valkenburg, R. L. Vaughn, M. Williams and J. S. Wilkes,
Thermochim. Acta, 2005, 425, 181–188.
6 M. Smiglak, A. Melten and R. D. Rogers, Acc. Chem. Res., 2007, 40,
1182–1192.
7 E. G. Yanez, S. R. Gratz and M. J. Baldwin, Anal. Chem., 2001, 73,
3838–3844.
8 H. Nakamoto and M. Watanabe, Chem. Commun., 2007, 24, 2539–
2541.
9 P. Scovazzo, D. Havard, M. McShea, S. Mixon and D. Morgan,
J. Membr. Sci., 2009, 327, 41–48.
10 K. R. Seddon, J. Chem. Technol. Biotechnol., 1997, 68, 315–316.
11 X. Jin, L. Yu, D. Garcia, R. X. Ren and X. Zeng, Anal. Chem., 2006,
78, 6980–6989.
12 L. Yu, D. Garcia, R. X. Ren and X. Zeng, Chem. Commun., 2005,
2277–2279.
P1–P3
with
the
three
[2-(methacryloyloxy)ethyl]-
trimethylammonium polymers already investigated,22 the
substitution of a methyl group with an heptylic one on the
quaternarized nitrogen does not affect qadv data to a great extent,
while considerably increasing the qrec with values ranging from
47.6ꢀ to 54ꢀ. This evidence supports the hypothesis previously
proposed where the tetraalkylammonium moiety was suggested
as the most responsible for low receding contact angle values; the
hydrophobic properties of such polymers can be then improved
by choosing the proper N-alkyl groups.
13 P. S. Kulkarni, L. C. Branco, J. G. Crespo and C. A. M. Afonso,
Chem.–Eur. J., 2007, 13, 8470–8477.
14 Y. Shen and M. Radosz, PCT Int. Appl. WO, 2006/026064 A2, 2006.
15 W. Ogihara, S. Washiro, H. Nakajima and H. Ohno, Electrochim.
Acta, 2006, 51, 2614–2619 and references therein.
16 R. Marcilla, J. A. Blazquez, J. Rodriguez, J. A. Pomposo and
D. Mecerreyes, J. Polym. Sci., Part A: Polym. Chem., 2004, 42,
208–212.
17 R. Marcilla, J. A. Blazquez, R. Fernandez, H. Grande, J. A. Pomposo
and D. Mecerreyes, Macromol. Chem. Phys., 2005, 206, 299–304.
18 S. Washiro, M. Yoshizawa, H. Nakajima and H. Ohno, Polymer,
2004, 45, 1577–1582.
Conclusions
19 R. Marcilla, M. Sanchez-Paniagua, B. Lopez-Ruiz, E. Lopez-
Carbacos, E. Ochoteco, H. Grande and D. Mecerreyes, J. Polym.
Sci., Part A: Polym. Chem., 2006, 44, 3958–3965.
20 X. Mu, J. Meng, Z. Li and Y. Kou, J. Am. Chem. Soc., 2005, 127,
9694–9695.
21 R. Marcilla, M. L. Curri, P. D. Cozzoli, M. T. Mart´ınez, I. Loinaz,
H. Grande, J. A. Pomposo and D. Mecerreyes, Small, 2006, 2, 507–
512.
A series of poly(ionic liquid)s obtained by free-radical polymer-
ization of monomeric ionic liquids based on the polymerizable
cation [2-(methacryloyloxy)ethyl]dimethylheptylammonium and
on the anions bis(trifluoromethylsulfonyl)imide, nonafluoro-1-
butanesulfonate,
dodecylbenzenesulfonate,
heptadeca-
This journal is ª The Royal Society of Chemistry 2009
J. Mater. Chem., 2009, 19, 8861–8870 | 8869

View Article Online
22 P. Cardiano, P. G. Mineo, F. Neri, S. Lo Schiavo and P. Piraino,
J. Mater. Chem., 2008, 18, 1253–1260.
35 P. Mineo, D. Vitalini and E. Scamporrino, Macromolecules, 1997, 30,
5285–5289.
23 J. B. Tang, W. L. Sun, H. D. Tang, M. Radosz and Y. Shen,
Macromolecules, 2005, 38, 2037–2039.
24 J. B. Tang, W. L. Sun, H. D. Tang, H. Plancher, M. Radosz and
Y. Shen, Chem. Commun., 2005, 3325–3327.
36 D. Briggs and G. Beamson, Anal. Chem., 1992, 64, 1729–1736.
37 D. Briggs and G. Beamson, Anal. Chem., 1993, 65, 1517–1523.
38 D. Briggs, in Practical Surfaces Analysis, ed. D. Briggs and M. P.
Seah, Wiley VCH, Weinheim, 2nd edn., 1995, vol. 1.
ꢀ
39 A. Gulino, F. Lupo, G. G. Condorelli, P. Mineo and I. Fragala,
Chem. Mater., 2007, 19, 5102–5109.
25 J. B. Tang, H. D. Tang, W. L. Sun, M. Radosz and Y. Shen, Polymer,
2005, 46, 12460–12467.
26 P. Mineo, E. Scamporrino, D. Vitalini, R. Alicata and S. Bazzano, Rapid
Commun. Mass Spectrom., 2005, 19, 2773–2779 and references therein.
27 D. Vitalini, P. Mineo and E. Scamporrino, Macromolecules, 1996, 29,
5520–5528.
28 M. Repoux, Surf. Interface Anal., 1992, 18, 567–570.
ꢀ
29 A. Gulino, S. La Delfa, I. Fragala and R. G. Egdell, Chem. Mater.,
40 L. Motiei, M. Altman, T. Gupta, F. Lupo, A. Gulino, G. Evmenenko,
P. Dutta and M. E. van der Boom, J. Am. Chem. Soc., 2008, 130,
8913–8915.
41 R. E. Baltus, B. H. Culbertson, S. Dai, H. Luo and D. W. DePaoli,
J. Phys. Chem. B, 2004, 108, 721–727.
42 J. L. Anderson, J. K. Dixon and J. F. Brennecke, Acc. Chem. Res.,
2007, 40, 1208–1216.
43 J. L. Anthony, J. L. Anderson, E. J. Maginn and J. F. Brennecke,
J. Phys. Chem. B, 2005, 109, 6366–6374.
1996, 8, 1287–1291.
30 A. Gulino, G. G. Condorelli, P. Mineo and I. Fragala,
Nanotechnology, 2005, 16, 2170–2175.
ꢀ
31 M. Smiglak, J. D. Holbrey, S. T. Giffin, W. M. Reichert,
R. P. Swatloski, A. R. Katritzky, H. Yang, D. Zang, K. Kirichenko
and R. D. Rogers, Green Chem., 2007, 9, 90–98.
32 H. L. Ngo, K. LeCompte, L. Hargens and A. B. McEwen,
Thermochim. Acta, 2000, 357–358, 97–102.
44 A. Yokozeki, M. B. Shiflett, C. P. Junk, L. M. Grieco and T. Foo,
J. Phys. Chem. B, 2008, 112, 16654–16663.
45 J. Vial and A. Jardy, Anal. Chem., 1999, 71, 2672–2677.
46 M. T. Gomes, A. C. Duarte and J. P. Oliveira, Anal. Chim. Acta, 1996,
327, 95–100.
33 J. Lingnau, M. Stickler and G. Meyerhoff, Eur. Polym. J., 1980, 16,
785–791.
47 M. T. Gomes, T. A. Rocha, A. C. Duarte and J. P. Oliveira, Anal.
Chim. Acta, 1996, 335, 235–238.
34 P. Mineo, G. Montaudo, D. Vitalini and E. Scamporrino, Rapid
Commun. Mass Spectrom., 2006, 10, 1551–1559.
48 P. Reveendran and S. L. Wallen, J. Phys. Chem. B, 2003, 107, 1473–
1477.
8870 | J. Mater. Chem., 2009, 19, 8861–8870
This journal is ª The Royal Society of Chemistry 2009