Iptycene-based fluorescent sensors for nitroaromatics and TNT

Article abstract of DOI:10.1002/chem.201200469

Small-molecule fluorescent sensors (1-5) for the recognition of nitroaromatic compounds, such as 2,4-dinitrotoluene and the explosive TNT, were obtained by using a three-step dehydrohalogenation cycloaddition protocol. The interaction of the receptors and nitroaromatics was studied both in solution and in the solid state by using fluorescence spectroscopy and X-ray crystallography, respectively. It is shown that the iptycene receptors 1-5 provide a cavity suitable for binding nitroaromatic compounds in an edge-to-face mode, rather than simple ring-stacking interactions. The results obtained inspired us to develop an inexpensive, reliable and robust sensor for vapour detection of explosives. Polymer nanofibres are particularly suitable for the production of such TNT sensors as they accelerate the mass exchange between the polymer and the vapours of TNT. Quenching of the sensors took place within 1 min compared to 10 min for a glass-slide assay. Hence, the sensor performance can be improved by optimising the matrix material and morphology without resynthesising the sensor moieties. TNT is sensed! Fluorescent sensors based on the iptycene recognition motif were synthesised and studied with different nitroaromatic compounds. X-ray structures confirm the establishment of edge-to-face interactions. Fluorescence studies in solution and in polymer matrices allowed the development of polymer-nanofibre sensors for the detection of nitroaromatic compounds and TNT (see figure). Copyright

Full text of DOI:10.1002/chem.201200469

FULL PAPER
DOI: 10.1002/chem.201200469
Iptycene-Based Fluorescent Sensors for Nitroaromatics and TNT
Pavel Anzenbacher, Jr.,* Lorenzo Mosca, Manuel A. Palacios, Grigory V. Zyryanov, and
Petr Koutnik[a]
Abstract: Small-molecule fluorescent
sensors (1–5) for the recognition of ni-
troaromatic compounds, such as 2,4-di-
nitrotoluene and the explosive TNT,
were obtained by using a three-step de-
hydrohalogenation cycloaddition proto-
col. The interaction of the receptors
and nitroaromatics was studied both in
solution and in the solid state by using
fluorescence spectroscopy and X-ray
crystallography, respectively. It is
shown that the iptycene receptors 1–5
provide a cavity suitable for binding ni-
troaromatic compounds in an edge-to-
face mode, rather than simple ring-
stacking interactions. The results ob-
tained inspired us to develop an inex-
pensive, reliable and robust sensor for
vapour detection of explosives. Poly-
mer nanofibres are particularly suitable
for the production of such TNT sensors
as they accelerate the mass exchange
between the polymer and the vapours
of TNT. Quenching of the sensors took
place within 1 min compared to 10 min
for
a glass-slide assay. Hence, the
sensor performance can be improved
by optimising the matrix material and
morphology without resynthesising the
sensor moieties.
Keywords: explosives
·
fluores-
cence spectroscopy · nanofibres ·
polymers · sensors
Introduction
over, handheld instruments often suffer from diminished se-
lectivity to interfering analytes, a condition frequently en-
countered in the field or in high-volume screening, such as
landmine fields or in the case of shipping containers or cra-
tes.[2b]
Until recently, most of the research efforts in explosives
sensing were oriented towards colorimetric/fluorescent
single-molecule sensors, such as phosphole oxides,[3a] oligo-
pyrroles,[3b,c] and calix[4]arenes,[3d] and fluorescent polymers
in which fluorescence changes in the presence of nitroaro-
matic analyte. Interesting examples are anthracene-based H-
bonded supramolecular polymers,[4a] iptycenes-based poly-
phenylene ethynylenes (PPE)[4b] or polyphenylenebutadiyny-
lenes,[4c] polyacetylenes,[4d] polymetaloles,[4e] such as polysil-
The development of efficient, robust, reliable and portable
methods for the detection of explosives is an issue in global
security. The widespread use of nitro-compounds and nitro-
aromatics as explosive materials, both in civilian and milita-
ry applications, makes them susceptible to illicit use. More-
over, TNT and its analogues are massively used in the fabri-
cation of landmines and improvised explosive devices
(IEDs).[1] This poses various threats to the civilian popula-
tion as the landmines are expensive to remove and demining
requires a significant expertise. It is estimated that 2.5 mil-
lion new landmines and IEDs are fabricated every year and
around 70 people are killed or injured daily by these weap-
ons. The cost of disarming a single landmine is in the vicinity
of 1000 US dollars.
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
State-of-the-art detection in public and military structures
is performed by using trained canines, IR/Raman spectro-
scopy detectors, by X-raying the suspect containers, and
with ion mobility spectrometry, an analytical technique ca-
pable of detecting vapours of explosive materials as well as
narcotics or hazardous chemicals.[2a] Although these meth-
ods are considered highly efficient, their cost and portability
is usually limited, particularly for stand-off detection. More-
nylene (MEH-PPV),[4j] dendrimers[5a,b] and metal-organic
frameworks (MOFs).[6] However, many of these materials
are relatively costly to prepare on a large scale and, to date,
the only example of successful commercial application are
Swagerꢀs iptycene amplifying polymers,[4b] which led to the
development of the Fidoꢁ portable devices[7] by Nomadics,
Inc.
We previously reported[8] on our synthetic efforts of
small-molecule receptors 1–5 inspired by Swagerꢀs iptycene
polymers (Figure 1).[4b] Receptors 1–6 exhibit fluorescence
quenching upon coordination of nitroaromatics, presumably
due to a photoinduced electron transfer mechanism (PET).
To achieve the maximum efficiency of the PET process the
fluorogenic fragment must be appended close to the elec-
tron-rich receptor. In this case, upon receptor (sensor) exci-
tation, the electron in the LUMO of the receptor (donor)
can be transferred to the LUMO of the nitroaromatic ana-
[a] Dr. P. Anzenbacher, Jr., Dr. L. Mosca, Dr. M. A. Palacios,
Dr. G. V. Zyryanov, P. Koutnik
Center for Photochemical Sciences
Bowling Green State University
Bowling Green, OH 43403 (USA)
Fax : (+1)419-372-9809
E-mail: pavel@bgnet.bgsu.edu
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201200469.
Chem. Eur. J. 2012, 00, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

Figure 3. Analytes employed in this study. NB: nitrobenzene; 2,4-DNT:
2,4-dinitrotoluene; TNT: 2,4,6-trinitrotoluene; 1-NO2Napht: 1-nitronaph-
talene; DMDNB: 2,3-dimethyl-2,3-dinitrobutane; RDX: 1,3,5-trinitroper-
hydro-1,3,5-triazine.
Figure 1. Molecular structures of receptors 1–6.
lyte (acceptor). As a result, the excited state of the sensor
does not relax with emission of light, that is, its fluorescence
is quenched (Figure 2).
gation by means of X-ray crystallography. Further studies
were then performed by using fluorescence spectroscopy in
solution as well as in the solid state. The analytes tested in
this study are shown in Figure 3.
The considered nitroaromatics are of widespread use
amongst the commercial nitro-based explosives, either as
the main components or as manufacture by-products.
DMDNB is an aliphatic compound used as taggant for legal-
ly manufactured explosives; canines are particularly sensi-
tive to this molecule, less than a part per trillion.[7c] RDX is
another widespread explosive and is supposed to have a
poor or nil interaction with receptors 1–6.
Figure 2. Mechanism of electron transfer between the S1 state of receptor,
R, and the LUMO of a nitroaromatic compound, NA. a) The PET proc-
ess prevents radiative decay from the S1 state; b) back electron transfer
+
ÀC
C
from NA to R anion/cation radicals.
Results and Discussion
Synthesis: The synthesis of receptors 1–5 was previously re-
ported by our group.[8] To synthesise 1–5, we utilised a cyclo-
addition reaction of an aryne[10] and anthracene. Cadoganꢀs
procedure[11a,b] utilising a haloarene and tBuOK in the pres-
ence of anthracene is described to yield tert-butyl aryl ethers
as well as iptycene side-products (~2%). Revision of the
method allowed for an easy three-step synthesis of 1–5 in
20–50% yield.[12] The reaction mechanism is summarised in
Scheme 1. Compound 6 was prepared from pentiptycene
quinone[13] and lithium phenyl acetylide.[14]
As shown in Figure 1, different fluorogenic fragments
were
attached
in
the
1,4-positions
of
the
[1.1.1b.1.1]pentiptycene[9] central benzene ring. The steric
congestion caused by the ortho-substituents of the fluoro-
genic fragments and the iptycene bridgehead hydrogen
atoms is expected to constrain the 1,4-aryl fragments to
assume an almost perpendicular alignment with respect to
the central benzene ring. Unlike in the iptycene-phenyl
ethynylene polymers,[4b,c] this perpendicular arrangement of
the fluorogenic residues makes an effective conjugation with
the iptycene unlikely, thereby increasing the HOMO–
LUMO gap in 1–5. Conversely, the ethynylene moieties in 6
allow for efficient conjugation through the central benzene
ring. This conjugation is likely to decrease the corresponding
sensor LUMO energy, and potentially to reduce the driving
force of the PET process. To investigate this point, sensor 6
was prepared to show the improved sensing efficacy of 1–5
with a direct attachment of aryl moieties.
X-ray structures: Single crystals suitable for X-ray diffrac-
tion studies were obtained by slow evaporation of solutions
of receptors 5 in benzene and 3 in nitrobenzene. 5·C6H6
crystallises in the P–1 space group and the unit cell contains
one full and four quarter-units of 5·C6H6, shared with neigh-
bouring cells (Z=2).
As it is shown in Figure 4b, the dihedral angle of the thio-
phene rings with respect to the central plane of the iptycene
is 74.78, the two thiophenyl rings pertinent to the same 2,2’-
bisthiophene fragment are not perfectly coplanar, and show
a plane-to-plane angle of 1618. Figure 4a and b clearly show
that the benzene molecule fits into the cavity formed by the
receptor. Thus, the receptor, at least in the solid state, is
Previously published iptycene sensors were of polymeric
nature, which precluded insight into possible binding motifs
of the nitroaromatics and iptycene receptors. Because sen-
sors 1–5 are small molecules, an extensive effort was mount-
ed to grow single crystals that would enable binding investi-
&
2
&
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Iptycene-Based Fluorescent Sensors
FULL PAPER
Literature on the topic of
face-to-edge binding between
electron-rich faces and edges of
the nitroaromatics lends sup-
port to this binding mode.[15] In
any case, we can expect a stron-
ger interaction in the case of
the nitrobenzene compared to
the benzene guest, since the
former is an electron-poor aro-
matic thereby more susceptible
to form stronger bonds with an
electron-rich walls of the ipty-
cene receptor.
Fluorescence quenching stud-
ies: The photophysical proper-
ties of 1–6 were investigated by
Scheme 1. General synthesis of receptors 1–5.
using UV/Vis and fluorescence
spectroscopy.[16] The quantum
yields of 1–5 are relatively
modest (FF =0.10–0.20) owing to a low degree of conjuga-
tion and the presence of sulfur atoms that encourage inter-
system crossing to nonradiative triplet states. Sensors 1–5
showed dramatic fluorescence quenching in the presence of
nitroaromatic species compared to the very low response to
nitroaliphatic materials (DMDNB).
Figure 5 shows the results of the fluorimetric titration of
sensor 2 with TNT. The 1,4-bis(2-thienyl)iptycene was excit-
ed at 300 nm with the emission maximum at 345 nm. Upon
addition of TNT, the sensor displays a dramatic quenching
of fluorescence emission. The Stern–Volmer plot shows that
the quenching process is nonlinear with the increased con-
centration of the quencher. This supports the proposed
mechanism of quenching as a result of combined dynamic
and static quenching. The apparent value of the Stern–
Volmer constant (KSV ~3.3ꢃ103 mÀ1) calculated from the ini-
tial linear part of the isotherm is reported in Figure 5 and in
Table 1.[12,16]
Figure 4. Crystal structure of: a), b) 5·C6H6, and c), d) 3·C6H5NO2. Ther-
mal ellipsoids are represented at the 50% probability level. Panels (b)
and (d) show side-views of the host–guest complexes; the benzene and
the nitrobenzene molecules are shown in space-fill mode. The dihedral
angles, f, between the pendant groups and the iptycene are highlighted.
able to bind small-molecule aromatics, in an edge-to-face
fashion.[7b,14]
The host–guest complex 3·C6H5NO2 crystallises in the P–1
space group and the unit cell contains four units of the com-
plex (Z=4). Figure 4c shows a similar arrangement for the
complex 3·C6H5NO2 to that of 5·C6H6; the two benzothio-
phene residues lie almost perpendicular to the central ben-
zene ring of the iptycene with a dihedral angle of 81.98 (Fig-
ure 4d). The nitrobenzene molecule fits into the cavity and
also displays an edge-to-face binding mode.
Figure 5. Fluorimetric titration of sensor 2 (1.00ꢃ10À5 m in CH2Cl2) with
a standard solution of TNT (0.02m), lexc =300 nm. Emission spectra
before the addition of titrant and at the end of the titration are shown in
bold lines. Inset: Stern–Volmer plot corresponding to that titration.
Chem. Eur. J. 2012, 00, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
3
&
ÞÞ
These are not the final page numbers!

P. Anzenbacher, Jr. et al.
Table 1. Values of the apparent Stern–Volmer constants recorded for receptors 1–6 and the nitro-compounds used in this study.[a]
KSVa [mÀ1
]
analyte
1
2
3
4
5
6
NB
2,4-DNT
TNT
1-NO2Napht
DMDNB
4.0ꢃ103 (1)
3.57ꢃ103 (8)
2.76ꢃ103 (7)
1.11ꢃ104 (6)
2.3ꢃ102 (2)
2.2ꢃ103 (7)
2.35ꢃ103 (2)
3.0ꢃ103 (1)
1.75ꢃ103 (4)
6.96ꢃ103 (7)
1.43ꢃ103 (6)
91 (5)
577 (9)
4.5ꢃ103 (1)
3.3ꢃ103 (2)
5.10ꢃ103 (7)
81 (1)
4.43ꢃ103 (5)
1.83ꢃ103 (8)
6.9ꢃ103 (2)
1.29ꢃ103 (2)
1.01ꢃ103 (4)
6.8ꢃ103 (2)
17.6 (6)
1.19ꢃ103 (3)
1.33ꢃ103 (2)
1.34ꢃ104 (4)
38.3 (2)
167 (9)
79 (1)
[a] Calculated from titrations performed in CH2Cl2; excitation l=300 nm. Fitting errors were <10%. The value in parentheses represents the standard
deviation on the last digits.
Titrations of all the sensors (1–6) with the nitroaromatics
confirm that the expected recognition process is taking
place. The degree of divergence from the linear collision
quenching model is visualised as an upturn of the binding
isotherm. The nonlinearity is presumed to be due to the in-
crement of the contribution of the static binding. Supporting
our hypothesis is the fact that this nonlinearity is much
more pronounced for nitrobenzene and the least for TNT
(NB@2,4-DNT>TNT). We presume this reflects the ability
of the relatively small pentiptycene cleft to bind larger nitro-
aromatics, such as TNT and 2,4-DNT. Such weak binding is
perhaps the reason for failure of sensors to produce host–
guest crystals with TNT and 2,4-DNT. The Stern–Volmer
constants for DMDNB, a common detection taggant[17] used
Figure 6. Fluorimetric titration of sensor 1 (1.00ꢃ10À5 m in CH2Cl2) with
a standard solution of 2,4-DNT (0.02m), lexc =300 nm. Emission spectra
before the addition of titrant and at the end of the titration are shown in
in the legal production of explosives, are low compared to
the other aromatic compounds. Naturally, DMDNB cannot
form stable edge-to-face complexes with the iptycene recep-
tor. The perfect linearity of the Stern–Volmer plot for this
compound suggests a pure collisional quenching mechanism
(Figure 8).
bold lines. Inset: Stern–Volmer plot pertinent to the titration.
The titration of sensor 1 with 2,4-DNT is shown in
Figure 6. The behaviour of this receptor is similar to that of
sensor 2 in the presence of TNT. Again, the apparent Stern–
Volmer constant is reported. In Figure 7, the titration of re-
ceptor 3 with nitrobenzene is displayed. Also, in this case
the receptor displays a dramatic quenching of fluorescence
upon the addition of nitrobenzene.
Figure 8. Combined Stern–Volmer plots for sensor 3 and the analytes
used in this study. The DMDNB trace coincides with the diffusion limit
for the quenching process.
From Table 1 and Figure 8 one can appreciate the differ-
ence in the Stern–Volmer constants. As expected, DMDNB
shows only minor quenching activity as the absence of an ar-
omatic ring fragment renders any interaction with receptors
1–6 impossible. Also, the high energy of its LUMO forbids
any significant PET quenching. On the other hand, NB,
DNT, and TNT, display Stern–Volmer constants the differ-
ences for which appear to correlate with the volume and
Figure 7. Fluorescence titration of sensor 3 (1.00ꢃ10À5 m in CH2Cl2) with
a standard solution of nitrobenzene (0.02m), lexc =300 nm. Emission
spectra before the addition of titrant and at the end of the titration are
shown in bold lines. Inset: Stern–Volmer plot pertinent to the titration.
&
4
&
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Iptycene-Based Fluorescent Sensors
FULL PAPER
their corresponding fit into the pentiptycene cavity, vide
supra, as well as with their LUMO energies.[3a] Furthermore,
we verified our hypothesis regarding the face-to-edge bind-
ing motif by studying the interaction of RDX with receptors
2 and 3. As expected, the KSV (2: 32(Æ3)ꢃ102 mÀ1, 3:
16(Æ3)ꢃ102 mÀ1) values for RDX are consistent with a weak
interaction at the limit of the diffusional quenching mecha-
nism. We can explain the low affinity of RDX towards the
receptors by the absence of an aromatic ring required for
the edge-to-face interaction as well as the high LUMO
energy associated with the absence of aromaticity. The
stereo-electronic effects[18] of the RDX nitro-groups are as-
sumed to prevent an effective interaction of the hydrogen
atoms with the iptycene receptors.
Figure 9. Left-hand panel: CCD images of sensor chips by using receptors
2 and 5 embedded in polyurethane and illuminated with black light,
before exposure to vapours of 2,4-DNT (180 ppb),[20] TNT (10 ppb),[20]
and after recovery. Within 10 min after exposure the emission of both 2
and 5 appeared to be quenched (>80%) whereas the red emission from
the internal standard (tetraphenylporphyrin, HTPP) was largely unper-
turbed. The fluorescence is regenerated by removing the slide from the
chamber. Right-hand panel: series of images, taken by using a fluores-
cence optical microscope, showing the polymer nanofibres mat containing
sensor 5, upon exposure to 10 ppb vapours of TNT at different time inter-
vals. RGB: image in true colours (RGB), showing both blue and red
fibres. Blue: deconvolution of the blue channel showing the fibres con-
taining only receptor 5, quenching their fluorescence after recognition of
TNT. Red: red channel shows that the red fluorescent, HTPP fibres are
not affected by nitroaromatics.
Remarkably, the TNT quenching displayed by 1–5 sug-
gests that the efficacy of these small-molecule sensors is
comparable to, if not better than, the iptycene-based ampli-
fying polymers (KSV =1.17ꢃ103 mÀ1) or polysilole sensors
(KSV =4.34ꢃ103 mÀ1).[4e] In this context it is not surprising
that sensor 6 shows lower Stern–Volmer constants than 1–5.
The conjugation of the fluorophore to the iptycene, in fact,
lowers the LUMO of the receptor and renders the electron
transfer quenching less efficient.
Sensor-doped polymer films and nanofibres: One of the po-
tential disadvantages of small-molecule sensors compared to
the polymer ones is the fact that small molecules often lack
good film-forming properties. In fact, sensors 1–5 appear to
crystallise when deposited as films. This potential drawback
is circumvented by doping the sensors into a polymeric
matrix known to form good-quality films that could be ap-
plied on the surface of, for example, an optical probe. We
have solution-cast the polyurethane solutions (10% w/w) of
sensors 2 and 5 in wells of a glass chip to form 10 mm thick
films (Figure 9).
The chip also comprised tetraphenylporphyrin free-base
(HTPP) chosen as an internal standard, the fluorescence of
which is not affected by the presence of nitroaromatics as its
LUMO is below the LUMOs of the nitroaromatic com-
pounds used in this study. The chips were exposed to va-
pours of nitroaromatics at equilibrium. Vapour pressures of
2,4-DNT and TNT are 180 and 10 ppb, respectively.[19] As ex-
pected, the sensor films react differently to the 2,4-DNT and
TNT vapours. This is due to the different affinity of the
sensor–polymer films but also due to the difference in 2,4-
DNT and TNT vapour pressure. Perhaps most importantly,
the fluorescence is largely recovered when the chips are re-
equilibrated in air without the presence of the explosives.
As pointed out previously,[4] the main problem with mono-
lithic materials is the slow response to vapours of nitroaro-
matics. Here too, our films required about 15 min to achieve
a steady state and even longer for recovery. Although this
might be offset by pumping a stream of contaminated air
onto the sensor films, the results seemed to require further
improvement.
fashion the sensor materials into polymer nanofibre mats
hoping that the use of nanofibres with a large surface area
will increase the mass exchange rate from the vapour phase
to the polymer. The polyurethane nanofibres were electro-
spun[20] from a 0.03% (w/w) solution of the sensor in a 10%
(w/w) solution of polyurethane (Tecoflexꢄ). The sensor ma-
terial was electrospun as long, 300–400 nm in diameter
nanofibres. The blue fluorescence of nanofibre mats was
imaged by using a fluorescence microscope. Once again, the
red emission of the HTPP was used as an internal standard.
The observable quenching of 5 (blue nanofibres, Figure 9)
took place within 2–3 min, whereas the HTPP (red fibres)
was not significantly quenched. Once again, the fluorescence
was recovered upon re-equilibration in pure air.
The deconvolution of the RGB channels into separate
red, green, and blue channels allows for integration of non-
black (i.e., non-zero) pixels (Figure 9). The ratio of blue
(sensor 5)-to-red (HTPP) pixel counts serves as an output
value (Table 2). The blue-to-red ratios do not change signifi-
cantly between ten randomly selected sites in the mat. This
confirms that one can improve the sensor performance by
Table 2. Evolution of blue/red ratio values over time.[a]
t [min]
0
1
3
5
blue/red pixel
2.1835
1.8024
0.7866
0.1108
[a] Blue/red ratio values were calculated by deconvolution of the red and
blue channels in the images in Figure 9 followed by integration of non-
zero pixels in both the channels.
To optimise the mass exchange between the explosives-
contaminated air and the sensor material, we decided to
Chem. Eur. J. 2012, 00, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
5
&
ÞÞ
These are not the final page numbers!

P. Anzenbacher, Jr. et al.
chemical shifts (d/ppm) are referenced to the respective solvent and split-
ting patterns are designed as s (singlet), d (doublet), t (triplet), m (mul-
tiplet), and br (broad). EI DIP mass spectra were recorded by using a
Shimadzu QP5050A, MALDI-TOF was recorded with OmniFlex
(Bruker Daltonics).
optimising the matrix material and morphology without re-
synthesising the sensor moieties. It is conceivable that the
blue-to-red pixel ratio can be used as a digital output in de-
tection instruments.
Polymer-casted sensors: Studies utilising polyurethane matrices were per-
formed by using polymer films with incorporated sensors (2 and 5),
which were prepared by casting THF solutions containing sensors 2 and
5 (0.4% w/w sensor and 0.2% w/w for HTPP films as internal standard)
in TecoflexTM and TecoflexTM/polystyrene 2:8 (3.5% w/w) onto multi-well
chips. The fluorescence quenching of the polymer films with analyte was
carried out by placing the sensor chip in a tall sealed glass chamber in
which the analyte vapours had been equilibrated. The equilibrium pres-
sure of TNT at 228C is 10 ppb,[19] and is delivered to the sample by diffu-
sion. The quenching of the fluorescence was estimated by placing the
chamber on a UV scanner (Kodak Image Station 440CF) by using broad-
band UV epi-illumination centred at 365 nm. The emission in the blue
region was recorded by using a blue filter (band-pass filter 380–500 nm),
a red filter (long pass 580 nm). Recording blue and red fluorescence for
each image allows for a direct comparison of the sensor versus internal
standard (HTPP) fluorescence.
Conclusion
In summary, we have developed simple, small-molecule sen-
sors for aromatic nitro-compounds based on the iptycene re-
ceptor acting as the recognition moiety. Sensors 1–5 can
easily be prepared on a multigram scale by a robust two-
step synthetic protocol. Fluorogenic moieties attached to the
benzene core of the iptycene act as signalling units, by
switching off their fluorescence upon analyte recognition.
Fluorimetric titration experiments revealed that the sensors
1–6 show a significant response to nitroaromatics, particular-
ly to 2,4-DNT and TNT, and a poor response to aliphatic
nitro-compounds. Thus, sensors 1–5 can be used as new, in-
expensive yet sensitive compounds for the detection of ex-
plosives vapours.
Sensing with nanofibres: In a typical experiment a microscope slide with
electrospun fibres comprising sensor 5 and fibres comprising HTTP (in-
ternal standard) were developed in the vapour chamber. At 2, 5 and
10 min exposure the slides were removed and immediately covered with
a coverslip. The images of the sensor chips were captured by using a fluo-
rescence microscope with 365 nm epi-illumination. To prevent false posi-
tives from photobleaching every measurement was taken from a new
spot on the slide. Also, control experiments show that no significant pho-
tobleaching takes place within the first 2 min of exposure to UV light.
Nanofibre fabrication: Polymer solution (10% w/w, TecoflexTM and
0.03% w/w, 5) was loaded into a 1 mL disposable plastic syringe. A Ham-
ilton stainless steel needle with a 21 gauge was used. The needle was con-
nected to a high-voltage power supply (ES30P-5W, Gamma High Voltage
Research, Inc.). The polymer sensor solution was continuously dispensed
by using a syringe pump (Genie Plus, Kent Scientific) at a rate of
3 mLminÀ1. In a typical experiment a voltage of 5 kV was applied for
electrospinning. The distance between the needle tip and the collector
was 10 cm. The collector consists of two independent aluminium rods
connected to ground and a microscope slide (75ꢃ25 mm) between them.
The characterisation of the fibres was carried out by using epi-fluores-
cence optical microscope (DIALUX 20, Leitz/Leica) with CCD cooled
colour camera (Penguin 150CL, Pixera). In a typical sensing experiment
microscope slides containing electrospun fibres with 5 and fibres contain-
ing HTPP as internal standard were prepared for the TNT sensing. The
slides were placed inside a sealed glass chamber in which TNT (10 ppb)
and DNT vapours (180 ppb) were previously equilibrated at 238C. Micro-
graphs were taken at different times during the experiment. For this we
made use of an epi-fluorescence optical microscope with excitation cen-
tred at 365 nm. The RGB deconvolution and all the calculation over the
images were carried out by using ImageJ.
Materials suitable for coating surfaces with
a TNT
vapour-sensitive coating were achieved simply by doping
sensors 1–5 into polymer solutions. Particularly effective
were nanofibre sensor mats produced by electrospinning of
such sensor-doped polymer solutions. The use of polymer
nanofibre mats seems to be suitable for the production of
such sensor assays, as demonstrated recently by other re-
search groups.[21a,b] Experiments with polyurethane nanofi-
bres displayed high surface area and effective mass ex-
change between the sensor mat and the surrounding gas,
and resulted in an effective nitroaromatic vapour recogni-
tion in 2–3 min. This confirms that the efficiency and time-
line of the response may be tuned by using polymer matri-
ces and the way these matrices are processed.
Studies employing non-equilibrium conditions, that is,
pumping the contaminated headspace gas onto the sensor
nanofibres expecting to dramatically improve the temporal
resolution of the sensors, are in progress. Nonetheless, the
use of a colour channel (red-blue) deconvolution provides a
robust output signal not only for the explosives, but also as
a fail-safe in case of malfunction due to, for example, non-
specific quenching or extensive contamination, which would
affect both the red and blue signal levels.
Acknowledgements
Experimental Section
The financial support from BGSU, NSF (grants CHE-0750303, DMR-
1006761 and EXP-LA 0731153) are gratefully acknowledged.
General procedures and materials: All synthetic manipulations were per-
formed under dry argon atmosphere by using standard techniques. All
1
starting materials were purchased and used as received. H and 13C NMR
[1] Containing the threat from illegal bombings: an integrated national
strategy for marking, tagging, rendering inert, and licensing explosives
and their precursors, National Academy Press, Washington DC,
1998.
[2] a) http://www.smithsdetection.com/mmtd.php, last accessed: Febru-
ary, 2012; b) D. S. Moore, Rev. Sci. Instrum. 2004, 75, 2499–2512.
[3] a) K. Shiraishi, T. Sanji, M. Tanaka, ACS Appl. Mater. Interfaces
2009, 1, 1379–1382; b) D. S. Kim, V. M. Lynch, K. A. Nielsen, C.
spectra were recorded with a Bruker DRX-300 (300 MHz) spectrometer.
Emission spectra were acquired at the Edinburgh single photon counting
spectrofluorimeter (FL/FS 900). The spectra in solution were obtained at
room temperature by using a quartz cuvette with a path-length of 1 cm at
right angle detection. The absorbance OD of all samples used for fluores-
cence studies were equal to or below 0.1. Fluorescence quantum yields of
1–6 were determined relative to anthracene (FF =0.27 in ethanol). The
&
6
&
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Iptycene-Based Fluorescent Sensors
FULL PAPER
Johnsen, J. O. Jeppesen, J. L. Sessler, Anal. Bioanal. Chem. 2009,
395, 393–400; c) B. M. Rambo, J. L. Sessler, Chem. Eur. J. 2011, 17,
4946–4959; d) Y. H. Lee, H. Liu, J. Y. Lee, S. H. Kim, S. K. Kim,
J. L. Sessler, Y. Kim, J. S. Kim, Chem. Eur. J. 2010, 16, 5895–5901.
[4] a) B. Gole, S. Shanmugaraju, A. K. Bar, P. S. Mukherjee, Chem.
Commun. 2011, 47, 10046–10048; b) D. T. McQuade, A. E. Pullen,
T. M. Swager, Chem. Rev. 2000, 100, 2537–2574; c) D. Zhao, T. M.
Swager, Macromolecules 2005, 38, 9377–9384; d) Y. Liu, R. C. Mills,
J. M. Boncella, K. S. Schanze, Langmuir 2001, 17, 7452–7455; e) H.
Sohn, M. J. Sailor, D. Magde, W. C. Trogler, J. Am. Chem. Soc. 2003,
125, 3821–3830; f) H. Sohn, R. M. Calhoun, M. J. Sailor, W. C. Tro-
gler, Angew. Chem. 2001, 113, 2162–2163; Angew. Chem. Int. Ed.
2001, 40, 2104–2105; g) S. Tao, G. Li, H. Zhu, J. Mater. Chem. 2006,
16, 4521–4528; h) S. Burattini, H. M. Colquhoun, B. W. Greenland,
W. Hayes, M. Wade, Macromol. Rapid Commun. 2009, 30, 459–463;
i) F. Chu, L. Ye, J. Yang, J. Wang, Opt. Commun. 2012, 285, 1171–
1174.
[5] a) G. Tang, S. S. Y. Chen, P. E. Shaw, K. Hegedus, X. Wang, P. L.
Burn, P. Meredith, Polym. Chem. 2011, 2, 2360–2368; b) D. A.
Olley, E. J. Wren, G. Vamvounis, M. J. Fernee, X. Wang, P. L. Burn,
P. Meredith, P. E. Shaw, Chem. Mater. 2011, 23, 789–794.
[6] B. Gole, A. K. Bar, P. S. Mukherjee, Chem. Commun. 2011, 47,
12137–12139.
[7] a) J.-S. Yang, T. M. Swager, J. Am. Chem. Soc. 1998, 120, 5321–
5322; b) J.-S. Yang, T. M. Swager, J. Am. Chem. Soc. 1998, 120,
11864–11873; c) C. J. Cumming, C. Aker, M. Fisher, M. Fox, M. J. la
Grone, D. Reust, M. G. Rockley, T. M. Swager, E. Towers, V. Wil-
liams, IEEE T. Geosci. Remote 2001, 39, 1119–1128.
Bonded Functional Groups, Supplement C2 (Ed.: S. Patai) Wiley,
Chichester, 1994.
[11] a) J. I. G. Cadogan, J. K. A. Hall, J. T. Sharp, J. Chem. Soc. C 1967,
1860–1862; b) J. I. G. Cadogan, M. J. P. Harger, J. T. Sharp, J. Chem.
Soc. B 1971, 602–607.
[12] See the Supporting Information for detailed synthetic procedures
and photophysical aspects of nitroaromatics sensing.
[13] X.-Z. Zhu, C.-F. Chen, J. Org. Chem. 2005, 70, 917–924.
[14] E. A. Meyer, R. K. Castellano, F. Diederich, Angew. Chem. 2003,
115, 1244–1287; Angew. Chem. Int. Ed. 2003, 42, 1210–1250.
[15] F. J. Carver, C. A. Hunter, D. J. Livingstone, J. F. McCabe, E. M.
Seward, Chem. Eur. J. 2002, 8, 2847–2859.
[16] Principles of Fluorescence Spectroscopy, 3rd ed. (Ed.: J. R. Lako-
wicz) Springer, New York, 2006.
[17] S. W. Thomas III, J. P. Amara, R. E. Bjork, T. S. Swager, Chem.
Commun. 2005, 4572–4574.
[18] I. F. Shiskov, L. V. Vilkov, M. Kolonitz, B. Rozsondai, Struct. Chem.
1990, 1, 57–64.
[19] a) Vapor Pressures at 258C, Handbook of Physical Properties of Or-
ganic Chemicals (Ed.: D. R. Lide) CRC, Boca Raton, FL, 1997;
b) B. C. Dionne, D. P. Rounbehler, E. K. Achter, J. R. Hobbs, D. H.
Fine, J. Energ. Mater. 1986, 4, 447–472; c) R. B. Cundall, T. F.
Palmer, C. C. Wood, J. Chem. Soc. Faraday Trans. 1 1978, 74, 1339–
1345; d) J. M. Rosen, C. Dickinson, J. Chem. Eng. Data 1969, 14,
120–124.
[20] D. Li, Y. Xia, Adv. Mater. 2004, 16, 1151–1170.
[21] a) C. Zhang, Y. Che, X. Yang, B. R. Bunes, L. Zang, Chem.
Commun. 2010, 46, 5560–5562; b) Y. Yang, H. Wang, K. Su, Y.
Long, Z. Peng, N. Li, F. Liu, J. Mater. Chem. 2011, 21, 11895–11900.
[8] G. V. Zyryanov, M. A. Palacios, P. Anzenbacher, Jr., Org. Lett. 2008,
10, 3681–3684.
[9] H. Hart, S. Shamouilian, Y. Takehira, J. Org. Chem. 1981, 46, 4427–
4432.
[10] a) R. W. Hoffman, in Dehydrobenzene and Cycloalkynes, Academic
Press, New York, 1967; b) H. Hart, in The Chemistry of Triple-
Received: February 13, 2012
Revised: June 3, 2012
Published online: && &&, 0000
Chem. Eur. J. 2012, 00, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
7
&
ÞÞ
These are not the final page numbers!

P. Anzenbacher, Jr. et al.
TNT is sensed! Fluorescent sensors
based on the iptycene recognition
motif were synthesised and studied
with different nitroaromatic com-
pounds. X-ray structures confirm the
establishment of edge-to-face interac-
tions. Fluorescence studies in solution
and in polymer matrices allowed the
development of polymer-nanofibre
sensors for the detection of nitroaro-
matic compounds and TNT (see
figure).
Sensors
P. Anzenbacher, Jr.,* L. Mosca,
M. A. Palacios, G. V. Zyryanov,
P. Koutnik ..................... &&&& — &&&&
Iptycene-Based Fluorescent Sensors
for Nitroaromatics and TNT
&
8
&
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!