Pyrene-modified quartz crystal microbalance for the detection of polynitroaromatic compounds

Article abstract of DOI:10.1021/ac200755c

The synthesis of a dithiol-functionalized pyrene derivative is reported, together with studies of interactions between this receptor (and other related pyrenes) and nitroaromatic compounds (NACs), in both solution and in the solid state. Spectroscopic analysis in solution and X-ray crystallographic analysis of cocrystals of pyrene and NACs in the solid state indicate that supramolecular interactions lead to the formation of defined π-π stacked complexes. The dithiol-functionalized pyrene derivative can be used to modify the surface of a gold quartz crystal microbalance (QCM) to create a unique π-electron rich surface, which is able to interact with electron poor aromatic compounds. For example, exposure of the modified QCM surface to the nitroaromatic compound 2,4-dinitrotoluene (DNT) in solution results in a reduction in the resonant frequency of the QCM as a result of supramolecular interactions between the electron-rich pyrenyl surface layer and the electron-poor DNT molecules. These results suggest the potential use of such modified QCM surfaces for the detection of explosive NACs.

Full text of DOI:10.1021/ac200755c

ARTICLE
pubs.acs.org/ac
Pyrene-Modified Quartz Crystal Microbalance for the Detection of
Polynitroaromatic Compounds
Rajendran Vaiyapuri, Barnaby W. Greenland, Joanne M. Elliott,* Wayne Hayes, Roger A. Bennett,
Christine J. Cardin, Howard M. Colquhoun, Haitham Etman, and Claire A. Murray
Department of Chemistry, The University of Reading, Whiteknights, Reading, RG6 6AD, U.K.
S Supporting Information
b
ABSTRACT: The synthesis of a dithiol-functionalized pyrene
derivative is reported, together with studies of interactions
between this receptor (and other related pyrenes) and nitroaro-
matic compounds (NACs), in both solution and in the solid
state. Spectroscopic analysis in solution and X-ray crystallo-
graphic analysis of cocrystals of pyrene and NACs in the solid
state indicate that supramolecular interactions lead to the
formation of defined πꢁπ stacked complexes. The dithiol-
functionalized pyrene derivative can be used to modify the
surface of a gold quartz crystal microbalance (QCM) to create a unique π-electron rich surface, which is able to interact with electron
poor aromatic compounds. For example, exposure of the modified QCM surface to the nitroaromatic compound 2,4-dinitrotoluene
(DNT) in solution results in a reduction in the resonant frequency of the QCM as a result of supramolecular interactions between
the electron-rich pyrenyl surface layer and the electron-poor DNT molecules. These results suggest the potential use of such
modified QCM surfaces for the detection of explosive NACs.
he design and synthesis of chemosensors has become an
increasingly active area of research over recent years. Efforts
during/after exposure to an analyte. For example, variations in
the electrochemical response of metal19 or composite surfaces20
have been found to offer both sensitivity and selectivity in NAC
detection. Liedberg and co-workers produced a quartz crystal
microbalance (QCM) with a surface coating of 2,4-dinitrotoluene
(DNT) and monoclonal antibodies which was able to detect
TNT as a change in the resonant frequency of the QCM through
competitive binding of TNT to the antibody.21 More recently
Cerruti and co-workers have demonstrated a QCM coated with a
poly(ethylene-co-glycidyl methacrylate) (PEGM) peptide con-
jugate that could detect TNT in water.22 In both these examples,
the receptor layers were accessed through extensive and expen-
sive multistep synthesis. For example, the PEGM-peptide con-
tained 16 amino acid residues, and even though it was found
that by altering the amino acid sequence it was possible to
differentiate between the two closely related NACs, DNT
and TNT, the approach may not prove to be the most cost-
effective.
T
have focused on producing sensitive and selective detection
devices for a diverse range of analytes, for example, to improve
medical diagnosis1 and for real-time pollution control.2 The
pressing, if regrettable, requirement to produce robust methods
for detecting trace levels of explosive compounds is a major
objective in modern society. Nitroaromatic compounds (NACs)
are commonly used as explosives, and to be able to detect these
compounds on a short time scale in a reliable fashion is highly
desirable for both civilian and military security.
One of the most widely produced and environmentally prevalent
explosives is 2,4,6-trinitrotoluene (TNT), which is not only a
high explosive but is also toxic and a known mutagen upon
ingestion.3 However, detection methods tend to focus on
identifying a range of NACs, and these procedures are often
solution-based as TNT has very low volatility (∼2 ꢀ 10ꢁ4 Torr
at 25 °C).4 For example, several of the most highly developed
chemosensors for NACs rely on a measurable change in the
optical properties of a polymer upon contact with the explosive.
Thus, the fluorescence emission spectra of a range of conjugated
polymers including poly(acetylenes)5 and poly(p-phenylene-
vinylenes)6 among others7ꢁ15 have been shown to be extremely
sensitive to trace levels of NACs. Alternative systems that
produce a readily visualized optical readout have also been
developed based on surface functionalized silica nanoparticles,16
quantum dots,17 and metal nanoparticles.18
Burattini and co-workers recently reported the one-step
synthesis of a pyrene-functionalized copolyimide that was able
to detect NACs both in solution and in the solid state.23 Sensing
was achieved by exploiting the well-known24,25 πꢁπ stacking
interactions between π-electron rich pyrene residues and π-elec-
tron poor species such as NACs. Complex formation between
these species results in fluorescence quenching of the pyrenyl
Received: March 25, 2011
Accepted: July 1, 2011
Published: July 15, 2011
A conceptually distinct approach to NAC detection is to
measure the change in the physical properties of a substrate
r
2011 American Chemical Society
6208
dx.doi.org/10.1021/ac200755c Anal. Chem. 2011, 83, 6208–6214
|

Analytical Chemistry
ARTICLE
Scheme 1. Schematic of the Pyrenyl-Modified Gold QCM Surface and the Subsequent Intercalation of a NAC via πꢁπ Stacking
Interactionsa
a R = NO2 (TNT); R = H (DNT).
emission, which provides a convenient visual readout that has
been investigated for NAC detection previously.23,26
THF (75 mL) under argon. After 2 h at room temperature, the
resulting white precipitate was filtered off and 1-pyremethyl-
We report here a new strategy for producing a novel QCM
based NAC detector. The strategy involved the synthesis of a
pyrene functionalized dithiol derivative that could be used to
modify the surface of a gold quartz crystal. It was hypothesized
that upon exposure to a gold surface, the pyrenyl derivative would
anchor to the surface via the formation of goldꢁthiol bonds to
produce a stable modified surface capable of interacting with
NACs in solution. This approach is based on the well-known
theory27 that describes the formation of self-assembled mono-
layers. Upon exposure to π-electron poor NACs, it was hypothe-
sized that complementary πꢁπ stacking interactions with the π-
electron rich pyrenyl modified surface would lead to an increase
in the mass of the surface layer and a concomitant reduction in
the resonant frequency of the QCM; Scheme 1 represents this
concept. A significant advantage of this approach, unlike many
related QCM and polymer-based sensors,21,22 is that the key
NAC receptor group is accessed efficiently in only one synthetic
step from commercially available materials. The results presented
in this paper demonstrate the successful formation and applica-
tion of such a novel modified surface for NAC detection and is of
particular relevance to the field of sensor development given the
simplicity of the approach.
amine HCl (2.41 g, 9.00 mmol), triethylamine (5.6 mL, 40 mmol),
3
and chloroform (150 mL) were added to the filtrate. After 16 h,
the solvent was removed under reduced pressure and the crude
product was washed with copious amounts of water to remove
triethylamine hydrochloride salt and the residue was suspended
in chloroform and filtered. The solvent was removed under
reduced pressure, and the crude product was washed with
methanol followed by ethyl acetate to give the title compound
(1) as a pale yellow solid 1.73 g (46%). mp = 154.5ꢁ155.5 °C;
MS ESI: (C25H26ONS2) = 420.1450, found m/z = 420.1448. 1H
(400 MHz, CDCl3): δ (ppm) = 8.25ꢁ8.11 (m, 5 H), 8.07ꢁ7.93
(m, 4 H), 5.81 (br t, J = 5.0 Hz, 1H), 5.17ꢁ5.08 (m, 2 H), 3.46
(q, J = 6.5 Hz, 1H), 3.10ꢁ2.98 (m, 2H), 2.36ꢁ2.29 (m, 1H),
2.20 (t, J = 7.4 Hz, 2H), 1.83ꢁ1.73 (m, 1H), 1.71ꢁ1.59 (m, 2H
and 2H), 1.50ꢁ1.33 (m, 2H). 13C NMR (100 MHz, CDCl3): δ
(ppm) = 172.3, 131.2, 131.1, 130.7, 129.0, 128.2, 127.5, 127.3,
127.2, 126.1, 125.4, 125.3, 125.0, 124.7, 124.6, 122.8, 56.4, 41.9,
40.1, 38.4, 36.4, 34.5, 28.8, 25.4. FT-IR: ν/cmꢁ1 3290 (amide
NꢁH), 2937 (methylene CꢁH), 1638 (amide CdO), 1558
(aromatic CdC), 843 (aromatic CꢁH).
Instrumental Analyses. 1H and 13C NMR spectra were
recorded on a Bruker Nanobay 400 MHz spectrometer operating
at 400 and 100 MHz for 1H and 13C nuclei, respectively.
Chemical shifts are reported in parts per million (ppm) relative
to tetramethylsilane (0.00 ppm). Electrospray mass spectra were
recorded using a ThermoFisher Scientific Orbitrap XL instru-
ment. FT-IR spectra were recorded on a Perkin-Elmer Spectrum
100 series spectrometer. The samples were visualized under
natural light and with UV light (254ꢁ365 nm). Fluorescence
spectra were carried out using a Varian Eclipse fluorescence spec-
trophotometer. Single crystal X-ray analysis was carried out on an
Oxford Diffraction X-Calibur CCD diffractometer using CuꢁKR
radiation. Structure solution and refinement were carried out
using the CRYSTALS suite of programs. Melting points were
determined on a Stuart SMP10 melting point instrument and are
uncorrected.
’ EXPERIMENTAL SECTION
Chemicals. 1-Pyrenemethylamine hydrochloride (95%), R-
lipoic acid (g99%), dicyclohexylcarbodiimide (DCC), N-hydroxy
succinimide (NHS), and anhydrous triethylamine were purchased
from Aldrich and used as received. Solvents and sulfuric acid
(concentrated) were purchased from Fisher and used as received
except THF, which was distilled from sodium and benzophe-
none, and chloroform, which was distilled from calcium hydride
prior to use. Argon and nitrogen gases were purchased from BOC
Gases and used as received. Deionized water was collected from a
Milli-Q purification system.
Synthesis of 5-(1,2-Dithiolan-3-yl)-N-(pyren-1-ylmethyl)-
pentanamide [pyrenyl derivative (1)]. R-Lipoic acid (2.60 g,
10.0 mmol), N-hydroxy succinimide (1.44 g, 12.5 mmol) and
dicyclohexylcarbodiimide (2.58 g, 12.5 mmol) were stirred in
The XPS experiments were performed on a modified ESCA2
system from Vacuum Generators UK.28,29 The experiment was
performed under UHV conditions (base pressure 1 ꢀ 10ꢁ9 Torr),
6209
dx.doi.org/10.1021/ac200755c |Anal. Chem. 2011, 83, 6208–6214

Analytical Chemistry
ARTICLE
Scheme 2. Synthesis of Pyrenyl Derivative (1)a
a Reagents and conditions: NHS, DCC, THF/CHCl3, 18 h, 46%.
and data were acquired using an unmonochromated Al KR
(1486.6 eV) X-ray source (10 kV, 400 W). The binding energies
were calibrated with respect to the Au 4f7/2 peak at 84.0 eV. The
takeoff angle between the sample surface and the energy analyzer
was 40°. The survey scan was taken at 50 eV pass energy, and the
high-resolution detailed scans were taken at 20 eV for Au 4f, S 2p,
C 1s, N 1s, and O 1s. The data were analyzed using CasaXPS
(software package) by fitting a Shirley background and using a
standard GaussianꢁLorentzian mixed product curve.
used as a structurally related but safer alternative to TNT).
Simple mixing of the essentially colorless solutions of pyrenyl
derivative (1) and DNT produced a yellow solution, indicative
of a charge transfer absorption arising from complementary πꢁπ
stacking between the components.30,31 In addition, irradiation of
a solution of pyrenyl derivative (1) by a standard laboratory UV
lamp (310 nm) produced a vibrant blue fluorescence, which is
typical for emission from pyrenyl derivatives. This emission was
completely quenched on addition of a solution of DNT, provid-
ing further evidence for complex formation (see the Supporting
Information, Figures S1a and S1b for images and fluorescence
spectra).
A more quantitative analysis of complex formation between
the pyrenyl derivative (1) and DNT was achieved by an NMR
titration study. A fixed concentration of DNT (5 mM) was titrated
with increasing concentrations of pyrenyl derivative (1) (0ꢁ10 mM),
and an NMR spectrum was taken after each sequential addition.
Figure 1a shows that as the concentration of pyrenyl derivative
(1) increased, a progressive upfield shift for the protons asso-
ciated with the DNT occurred from 8.86 to 8.84 ppm (see the
Supporting Information, Figure S2, for the NMR spectra). This is
a consequence of the equilibrium position for the system moving
from a less bound state (lower degree of complexation) to a more
bound state as the titration progresses. The upfield shift in
resonance frequency for the DNT protons during complexation
demonstrates that these protons reside within the shielding
region generated by the π electron system of the pyrenyl ring.32
The only relative orientation of these two molecules that can
produce this effect is when the pyrene and DNT form a classic
face-to-face π-stack.33
A QCM200 instrument and piezoelectric quartz crystals (Au/
Ti with AT-cut, 5 MHz, 2.54 cm diameter) were obtained from
Lambda Photometrics, U.K., and used for all experiments. The
electrode surface area in contact with the liquid was 1.37 cm2.
Surface Modification and NAC Detection. Surface modifica-
tion experiments were carried out in the following manner. Crystals
were first cleaned by dipping into piranha solution: concentrated
H2SO4/30% H2O2, 3:1 v/v. Caution! Procedure to be carried
out in a fume hood and with extreme care: formation of piranha
solution is exothermic and violently reacts with organic compounds.
This cleaning process was carried out at room temperature for
∼15 s followed by rinsing with copious amounts of deionized
water (this was repeated three times). The crystals were then
dried in a gentle stream of nitrogen gas. Prior to further use, the
crystals were subject to electrochemical cleaning (by cyclic
voltammetry between ꢁ0.3 and 1.8 V vs Ag/AgCl (1 M) for
10 cycles) in aqueous 0.5 M sulfuric acid solution (prepared
using deionized water).
A clean crystal was then immersed in 100 mL of chloroform at
28 °C. The frequency was stabilized, and a 2 mL solution of
pyrenyl derivative (1) in chloroform was injected (resulting in an
overall concentration of 2.0 mM), and the solution was stirred for
∼30 s. Any change in crystal frequency was then recorded for 3 h;
however, the immersion was continued for 24 h. Finally the
modified crystal surface was rinsed with copious amounts of
chloroform followed by water and dried with a gentle stream of
nitrogen gas.
A modified crystal was then immersed in 100 mL of chloro-
form at 28 °C. The frequency was stabilized, and a 2 mL solution
of DNT in chloroform was injected (resulting in an overall
concentration of 2.0 mM) and the solution was stirred for ∼30 s.
Any change in crystal frequency was then recorded for up to
2000 s.
The binding constant for this complex was found to be very
low, (∼1 Mꢁ1) as determined by regression analysis using the
software Dynafit.34,35 Over the concentration range studied,
complexation/decomplexation remained fast on the NMR time
scale as indicated by the single signal observed for the DNT
resonances in all measurements. The method of continuous vari-
ation (Job’s plot analysis36) was used to investigate the binding
stoichiometry between pyrenyl derivative (1) and DNT, and as
shown in Figure 1b, this was found to be 1:1 (see the Supporting
Information, Figure S2, for NMR spectra).
The formation of a face-to-face π-stack between pyrenyl
1
derivative (1) and DNT was verified through H NMR spec-
troscopy; however, it was not possible to obtain single crystals
suitable for X-ray structure determination to verify this interac-
tion in the solid state, perhaps as a consequence of the structural
asymmetry of the compounds. To overcome this problem, the
commercially available NAC, 3,5-dinitrobenzonitrile (DBN), was
selected as an electronically related but more symmetric analogue of
DNT.23 In addition, pyrene was selected as an appropriate analogue
of the pyrenyl derivative (1), because the thiol-containing side chain
of pyrenyl derivative (1) does not influence the electronic interac-
tions that are present during complex formation.
’ RESULTS AND DISCUSSION
The synthesis of pyrenyl derivative (1) needed to functiona-
lize the QCM surface was achieved by reaction of R-lipoic acid
(2) with pyrenemethylamine HCl (3) under conventional cou-
3
pling conditions. Scheme 2 summarizes the reagents and condi-
tions used.
It was first necessary to verify the key interaction underpinning
the sensor design by investigating the nature of the complex
formed between NACs and the pyrenyl derivative (1) (DNT was
6210
dx.doi.org/10.1021/ac200755c |Anal. Chem. 2011, 83, 6208–6214

Analytical Chemistry
ARTICLE
Figure 1. (a) Chemical shift of DNT proton as a function of pyrenyl
derivative (1) concentration, measured at a fixed concentration of DNT
(5 mM) and (b) binding stoichiometry (Job plot).
Additional NMR binding studies performed between pyrenyl
derivative (1) and DBN (see the Supporting Information, Figure
S3a for the binding constant determination study and Job’s plot
analysis and Figure S3b in the Supporting Information for the
NMR spectra) revealed a similar upfield shift in the resonant
frequency for the protons on DBN as seen for DNT. In addition,
Job’s plot analysis verified complexation between DBN and
pyrenyl derivative (1) proceeded with a binding stoichiometry
of 1:1. These data show that DNT and DBN form similar face-to-
face π-stacked complexes with pyrenyl derivative (1) in solution
and so may be expected to form similar solid state structures.
Cocrystallization of DBN and pyrene afforded deep yellow
single crystals suitable for X-ray crystallographic analysis, and the
results are summarized in Figure 2aꢁc (for full data see the
Supporting Information, S4, S5, and S6). Analysis of the cocrys-
tals of DBN and pyrene revealed extended stacks comprising
alternating layers of π-electron rich pyrene and π-electron poor
DBN molecules separated by ∼3.42 Å, in good agreement with
related structures.37
Figure 2. Single crystal X-ray structure of the 1:1 complex between
DBN and pyrene: (a) ball and stick representation (side view), (b) space
filling representation (side view), and (c) ball and stick representation
(end view along the axis of the πꢁπ stack).
Hence surface modification was attempted by the addition of a
2.0 mM solution of pyrenyl derivative (1) in chloroform to the
resonance-stabilized crystal surface. The deposition of pyrenyl
derivative (1) could be followed readily by monitoring the
frequency of the QCM with respect to time. Figure 3a shows
that upon exposure to pyrenyl derivative (1), the frequency of the
QCM decreased rapidly, indicating successful surface modifica-
tion. An average decrease in frequency (Δf) of ꢁ30 ( 6 Hz (95%
CI quoted, σ = 5, n = 3) was observed over a period of 3 h. It is
known that the formation of high-quality adlayers requires long
6211
dx.doi.org/10.1021/ac200755c |Anal. Chem. 2011, 83, 6208–6214

Analytical Chemistry
ARTICLE
molecules cmꢁ2 are in evidence;27 hence, the surface coverage
determined in this study may appear a little high. However, it should
be remembered that pyrenyl derivative (1) is significantly larger in
cross section than an alkane thiol molecule, and also the QCM
surface used was rough (indicating that the actual surface area is
significantly larger than the geometric area). Indeed the roughness
factor (Rf) of the QCM surface was determined electrochemically
by cyclic voltammetry in aqueous sulfuric acid39 to be ∼2, and if this
real surface area is taken into account then the coverage in this study
corresponds to 3.85 ꢀ 1014 molecules cmꢁ2. This suggests that
perpendicular packing of the pyrenyl derivative (1) at the QCM
surface is a feasible model.
Further characterization of the modified surface was therefore
necessary, and X-ray photoelectron spectroscopy (XPS) was
carried out (for spectra, see the Supporting Information, Figure
S7). The cleaned, uncoated, Au surface exhibited a small C signal,
trace O and no N or S whereas after modification well-resolved
signals corresponding to the presence of C, N, O, and S on the Au
surface were observed, indicating successful surface modification.
The C 1s peak was observed to contain three components:
aromatic, methylene, and carbonyl carbons at 284.7 eV (CdC),
285.1 eV (CꢁC), and 288.1 eV (CdO), respectively.40 The
normalized areas of the three peaks were 0.62, 0.31 and 0.07,
respectively, and these are in good agreement with the expected
theoretical values of 0.64, 0.32, and 0.04. For O 1s a single broad
peak at 531.8 eV was observed, and this was assigned to the
carbonyl oxygen atom.40 The N 1s binding energy was observed
at 399.7 eV, consistent with the presence of an amide group.40 As
a result of the intrinsic low photoionization cross section, the
signal-to-noise ratio for the S 2p peaks was low and the spinꢁ
orbit split components were not clearly resolved. Nevertheless,
the binding energy for S 2p3/2 was determined to be 163.3 eV
when the data was fit with a spin orbit split 2p1/2 component
1.2 eV higher in energy at 164.5 eV (with the appropriate 2:1
intensity ratio). The 2p3/2 binding energy of 163.3 eV is high for a
chemisorbed species and is more closely associated with dithio-
late sulfur atoms in an unbound state on the gold surface.41ꢁ43
Overall these results support surface modification and indicate
that pyrenyl derivative (1) is present at the QCM surface,
probably in both a bound and an unbound state.
In order to clarify the nature of the adsorbed layer, atomic
force microscopy (AFM) was attempted on the modified crystal
surface. Unfortunately these attempts were unsuccessful, pre-
sumably as a result of the high Rf of the underlying QCM surface.
However, an STM study of the assembly of large π-conjugated
aromatic thiols at a gold (111) surface has been reported else-
where in the literature,44 and it was shown that a molecule similar
to pyrenyl derivative (1), 1-(11-mercaptoundecyl)pyrene, actu-
ally forms a disordered layer at room temperature, where
clustering of the molecules was observed.
In terms of understanding the nature of the adsorbed layer,
another factor to consider is that the solubility of pyrenyl
derivative (1) in chloroform is not high and as has been reported
previously, low solubility can lead to the formation of multilayer
structures.27 On the basis of the XPS analysis, and by considera-
tion of literature reports,43 it therefore seems likely that the
modified surface is composed of an adsorbed layer of dithiol
molecules where additional dithiol molecules lie on top of the
adsorbed layer or partially penetrate into the adsorbed later.
Subsequently, the pyrenyl derivative (1) modified surfaces
were tested in solution by exposure to a 2.0 mM solution of DNT
in chloroform, and the frequency was again monitored with
Figure 3. a) Frequency change observed at QCM surface following
addition of 2.0 mM of pyrenyl derivative (1) to chloroform, (b)
frequency change observed at a modified QCM surface following
addition of 2.0 mM DNT, and (c) frequency change observed at a bare
QCM surface following addition of 2.0 mM DNT.
exposure times,27 and so even though the frequency of the crystal
appeared to be stable this time exposure was continued for 24 h.
Postmodification, the surfaces were rinsed and dried prior to
further use. The Sauerbrey equation38 (eq 1) may be applied to
an adsorbed thin rigid layer and relates the observed frequency
change (Δf) with mass change per geometric electrode area (Δm),
ꢁ Δf ¼ ꢁ Cf Δm
ð1Þ
Here, Cf is a crystal sensitivity factor and it has a value of 56.6 Hz
μgꢁ1 cm2 for a 5 MHz AT-cut crystal. From this equation, the
deposited mass per geometric surface area and the average number
of molecules per geometric surface area were calculated to be 536
ng cmꢁ2 and 7.69 ꢀ 1014 molecules cmꢁ2, respectively. It has been
reported that in a close packed alkane thiol monolayer, ∼4.5 ꢀ 1014
6212
dx.doi.org/10.1021/ac200755c |Anal. Chem. 2011, 83, 6208–6214

Analytical Chemistry
ARTICLE
respect to time. Figure 3b shows that upon exposure to DNT, the
frequency decreased rapidly. An average decrease in frequency
(Δf) of ꢁ1.7 ( 0.7 Hz (95% CI quoted, σ = 0.6, n = 3) was
observed. This drop in frequency indicates an increase in mass of
the surface of the QCM presumably through interaction of DNT
with the pyrenyl modified surface.
Figure 3c shows data for the control experiment. In order to
exclude the possibility of nonspecific interaction of DNT with the
underlying gold surface, freshly cleaned and frequency-stabilized
QCMs were also exposed to a 2.0 mM solution of DNT in
chloroform. This resulted in very little change in frequency (Δf).
An average decrease in frequency (Δf) of ꢁ0.3 ( 0.3 Hz (95% CI
quoted, σ = 0.3, n = 3) was observed. This control study revealed
that there is little interaction between DNT and an unmodified
gold surface under these experimental conditions.
The gold surface modified with pyrenyl derivative (1) clearly
responds to DNT in solution (the observed signal was signifi-
cantly different to that shown by an unmodified surface). This
response may be related to the degree of order of the modified
surface, and attempts to improve the surface order by optimiza-
tion of the modification process followed by analysis of the effects
this may have upon NAC detection are now in progress.
support of C.A.M.. We thank the Royal Society/Wolfson Foun-
dation for a laboratory refurbishment grant.
’ REFERENCES
(1) He, Q. W.; Miller, E. W.; Wong, A. P.; Chang, C. J. J. Am. Chem.
Soc. 2006, 128, 9316–9317.
(2) Liu, Y.; Tang, Y. H.; Barashkov, N. N.; Irgibaeva, I. S.; Lam,
J. W. Y.; Hu, R. R.; Birimzhanova, D.; Yu, Y.; Tang, B. Z. J. Am. Chem. Soc.
2010, 132, 13951–13953.
(3) Won, W. D.; Disalvo, L. H.; Ng, J. Appl. Environ. Microbiol. 1976,
31, 576–580.
(4) Pushkarsky, M. B.; Dunayevskiy, I. G.; Prasanna, M.; Tsekoun,
A. G.; Go, R.; Patel, C. K. N. Proc. Natl. Acad. Sci. U.S.A. 2006,
103, 19630–19634.
(5) Liu, Y.; Mills, R. C.; Boncella, J. M.; Schanze, K. S. Langmuir
2001, 17, 7452–7455.
(6) Chang, C. P.; Chao, C. Y.; Huang, J. H.; Li, A. K.; Hsu, C. S.; Lin,
M. S.; Hsieh, B. R.; Su, A. C. Synth. Met. 2004, 144, 297–301.
(7) Saxena, A.; Fujiki, M.; Rai, R.; Kwak, G. Chem. Mater. 2005,
17, 2181–2185.
(8) Yang, J. S.; Swager, T. M. J. Am. Chem. Soc. 1998, 120, 11864–11873.
(9) Yang, J. S.; Swager, T. M. J. Am. Chem. Soc. 1998, 120, 5321–5322.
(10) Zhao, D.; Swager, T. M. Macromolecules 2005, 38, 9377–9384.
(11) Narayanan, A.; Varnavski, O. P.; Swager, T. M.; Goodson, T.
J. Phys. Chem. C 2008, 112, 881–884.
’ CONCLUSIONS
A quartz crystal microbalance (QCM) modified with pyrenyl
residues is capable of detecting the nitro aromatic compound 2,4-
dinitrotoluene (DNT) in solution. Sensing was measured as a
reduction in the resonant frequency of the QCM upon exposure
to DNT, which occurs as a result of an increase in the mass of the
QCM through interactions between the pyrenyl surface layer and
the DNT. Analysis of the interactions between pyrene and its
derivatives with NACs by spectroscopic analysis in solution and
X-ray crystallographic analysis of cocrystals in the solid state
suggest that formation of a defined πꢁπ stacked complex is the
dominant interaction between the NAC and the modified gold
surface.
(12) Sohn, H.; Calhoun, R. M.; Sailor, M. J.; Trogler, W. C. Angew.
Chem., Int. Ed. 2001, 40, 2104–2105.
(13) Sohn, H.; Sailor, M. J.; Magde, D.; Trogler, W. C. J. Am. Chem.
Soc. 2003, 125, 3821–3830.
(14) Toal, S. J.; Magde, D.; Trogler, W. C. Chem. Commun. 2005,
5465–5467.
(15) Sanchez, J. C.; DiPasquale, A. G.; Rheingold, A. L.; Trogler,
W. C. Chem. Mater. 2007, 19, 6459–6470.
(16) Gao, D. M.; Wang, Z. Y.; Liu, B. H.; Ni, L.; Wu, M. H.; Zhang,
Z. P. Anal. Chem. 2008, 80, 8545–8553.
(17) Shi, G. H.; Shang, Z. B.; Wang, Y.; Jin, W. J.; Zhang, T. C.
Spectrochim. Acta, Part A: Mol. Biomol. Spectrosc. 2008, 70, 247–252.
(18) Dasary, S. S. R.; Singh, A. K.; Senapati, D.; Yu, H. T.; Ray, P. C.
J. Am. Chem. Soc. 2009, 131, 13806–13812.
(19) Filanovsky, B.; Markovsky, B.; Bourenko, T.; Perkas, N.;
Persky, R.; Gedanken, A.; Aurbach, D. Adv. Funct. Mater. 2007, 17,
1487–1492.
(20) Hrapovic, S.; Majid, E.; Liu, Y.; Male, K.; Luong, J. H. T. Anal.
Chem. 2006, 78, 5504–5512.
’ ASSOCIATED CONTENT
S
Supporting Information. Photographs of solutions of
b
DNT, DNT + pyrenyl derivative (1), and pyrenyl derivative (1)
under natural and UV light (Figure S1a) and fluorescence quench-
(21) Larsson, A.; Angbrant, J.; Ekeroth, J.; Mansson, P.; Liedberg, B.
Sens. Actuators, B: Chem. 2006, 113, 730–748.
1
ing spectra (Figure S1b); H NMR spectra used for binding
constant/stoichiometry determination (Figure S2); pyrenyl deriva-
tive (1) + DBN binding constant/stoichiometry determination
(Figure S3a) and 1H NMR spectra (Figure S3b); X-ray structure of
the cocrystal: crystal data, structure solution, and refinement (S4);
X-ray cocrystal data in CIF file format (electronic, S5) and Checkcif
report (electronic, S6); XPS spectra of the QCM modified surface
(Figure S7). This material is available free of charge via the Internet
at http://pubs.acs.org.
(22) Cerruti, M.; Jaworski, J.; Raorane, D.; Zueger, C.; Varadarajan,
J.; Carraro, C.; Lee, S. W.; Maboudian, R.; Majumdar, A. Anal. Chem.
2009, 81, 4192–4199.
(23) Burattini, S.; Colquhoun, H. M.; Greenland, B. W.; Hayes, W.;
Wade, M. Macromol. Rapid Commun. 2009, 30, 459–463.
(24) Roesky, H. W.; Andruh, M. Coord. Chem. Rev. 2003, 236,
91–119.
(25) Claessens, C. G.; Stoddart, J. F. J. Phys. Org. Chem. 1997, 10,
254–272.
(26) Bai, H.; Li, C.; Shi, G. Q. Sens. Actuators, B: Chem. 2008, 130,
777–782.
’ AUTHOR INFORMATION
(27) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.;
Whitesides, G. M. Chem. Rev. 2005, 105, 1103–1169.
(28) McCavish, N. D.; Bennett, R. A. Surf. Sci. 2003, 546, 47–56.
(29) Bennett, R. A.; Mulley, J. S.; Newton, M. A.; Surman, M.
J. Chem. Phys. 2007, 127, 7.
Corresponding Author
*E-mail: j.m.elliott@reading.ac.uk
’ ACKNOWLEDGMENT
(30) Foster, R. In Organic Charge Transfer Complexes; Academic
Press Inc.: London, 1969; p 33ꢁ93.
We thank the University of Reading for a studentship in
support of R.V., EPRSC for support of B.W.G. (Grant EP/
G026203/1) and H.E., and Diamond Light Source Ltd. for
(31) Greenland, B. W.; Burattini, S.; Hayes, W.; Colquhoun, H. M.
Tetrahedron 2008, 64, 8346–8354.
6213
dx.doi.org/10.1021/ac200755c |Anal. Chem. 2011, 83, 6208–6214

Analytical Chemistry
ARTICLE
(32) Colquhoun, H. M.; Greenland, B. W.; Zhu, Z. X.; Shaw, J. S.;
Cardin, C. J.; Burattini, S.; Elliott, J. M.; Basu, S.; Gasa, T. B.; Stoddart,
J. F. Org. Lett. 2009, 11, 5238–5241.
(33) Hunter, C. A.; Sanders, J. K. M. J. Am. Chem. Soc. 1990,
112, 5525–5534.
(34) Kuzmic, P. Anal. Biochem. 1996, 237, 260–273.
(35) Bell, J. H.; Pratt, R. F. Inorg. Chem. 2002, 41, 2747–2753.
(36) Berkessel, A.; Adrio, J. A.; Huettenhain, D.; Neudorfl, J. M.
J. Am. Chem. Soc. 2006, 128, 8421–8426.
(37) Lee, Y. H.; Liu, H.; Lee, J. Y.; Kim, S. H.; Kim, S. K.; Sessler, J. L.;
Kim, Y.; Kim, J. S. Chem.—Eur. J. 2010, 16, 5895–5901.
(38) Sauerbrey, G. Z. Phys., A: Hadrons Nuclei 1959, 155, 206–222.
(39) Trasatti, S.; Petrii, O. A. Pure Appl. Chem. 1991, 63, 711–734.
(40) NIST X-ray Photoelectron Spectroscopy Database, version 3.5;
National Institute of Standards and Technology: Gaithersburg, MD,
2003, http://srdata.nist.gov/xps/.
(41) Nuzzo, R. G.; Zegarski, B. R.; Dubois, L. H. J. Am. Chem. Soc.
1987, 109, 733–740.
(42) Wen, X. G.; Linton, R. W.; Formaggio, F.; Toniolo, C.;
Samulski, E. T. J. Phys. Chem. A 2004, 108, 9673–9681.
(43) Castner, D. G.; Hinds, K.; Grainger, D. W. Langmuir 1996,
12, 5083–5086.
(44) Xu, Q. M.; Ma, H.; Yip, H.; Jen, A. K. Y. Nanotechnology 2008,
19, 11.
6214
dx.doi.org/10.1021/ac200755c |Anal. Chem. 2011, 83, 6208–6214