CB[8]-based rotaxane as a useful platform for sensitive detection and discrimination of explosives

Article abstract of DOI:10.1039/c3sc51132d

Based on a naphthalene-threaded cucurbit[8]uril (CB[8]) rotaxane structure on a solid substrate, a new strategy for rapid, fully reversible, and highly sensitive detection of a broad class of explosives was developed by using one receptor. Due to the unique confinement effect and size exclusion of the CB[8] cavity, it is found that the intercalation of an explosive compound in the constructed rotaxane can significantly influence the photophysical property of the naphthalene core in the confined nanocavity of CB[8]. Dependent on the electronic structures and the sizes of explosive compounds, the fluorescence of the naphthalene core would be quenched or enhanced to different extents, leading to the direct detection and discrimination of distinctively different groups of trace explosives in the vapor phase, especially including the challenging aliphatic nitro-organics (RDX, HMX and PETN). Control experiments were performed to show the different sensing behaviors between the common organic vapors and nitrate-based explosives, which made it easy to realize the discrimination between target analytes and interferents. Due to the surface-attached sensing elements, a rapid response was also achieved in this system. Moreover, the non-covalent nature of the resulting heteroternary complex indicates that the trapped target molecules in the rotaxane structure are facilely removable by simply washing, demonstrating an excellent regeneration of the constructed explosive sensors for real-world application. The performed experiments suggested that the rotaxane structure-based sensing protocol opened a new way to develop a new kind of explosive sensors enabling a richer identification of threats.

Full text of DOI:10.1039/c3sc51132d

Chemical Science
View Article Online
View Journal
EDGE ARTICLE
CB[8]-based rotaxane as a useful platform for sensitive
detection and discrimination of explosives†
Cite this: DOI: 10.1039/c3sc51132d
Wei Zhu, Weina Li, Chen Wang, Jiecheng Cui, Haowei Yang, Yin Jiang
*
and Guangtao Li
Based on a naphthalene-threaded cucurbit[8]uril (CB[8]) rotaxane structure on a solid substrate, a new
strategy for rapid, fully reversible, and highly sensitive detection of a broad class of explosives was
developed by using one receptor. Due to the unique confinement effect and size exclusion of the
CB[8] cavity, it is found that the intercalation of an explosive compound in the constructed rotaxane
can significantly influence the photophysical property of the naphthalene core in the confined
nanocavity of CB[8]. Dependent on the electronic structures and the sizes of explosive compounds,
the fluorescence of the naphthalene core would be quenched or enhanced to different extents,
leading to the direct detection and discrimination of distinctively different groups of trace explosives
in the vapor phase, especially including the challenging aliphatic nitro-organics (RDX, HMX and
PETN). Control experiments were performed to show the different sensing behaviors between the
common organic vapors and nitrate-based explosives, which made it easy to realize the
discrimination between target analytes and interferents. Due to the surface-attached sensing
elements, a rapid response was also achieved in this system. Moreover, the non-covalent nature of
the resulting heteroternary complex indicates that the trapped target molecules in the rotaxane
structure are facilely removable by simply washing, demonstrating an excellent regeneration of the
constructed explosive sensors for real-world application. The performed experiments suggested that
the rotaxane structure-based sensing protocol opened a new way to develop a new kind of
explosive sensors enabling a richer identification of threats.
Received 26th April 2013
Accepted 12th June 2013
DOI: 10.1039/c3sc51132d
www.rsc.org/chemicalscience
remains a great challenge.3 Additionally, since their physico-
chemical properties differ from each other signicantly, the
1 Introduction
Rapid detection and discrimination of hidden explosives are
very important aspects concerning homeland security, envi-
ronmental and humanitarian safety.1 In this respect, nitrated
aromatics (e.g. TNT and DNT), nitramines (e.g. RDX and HMX),
and nitrate esters (e.g. PETN) are three classes of the most
commonly used energetic compounds,2 which are directly
associated with modern bombs in terrorist attacks or other
criminal intents. In the past years, although numerous methods
have been proposed for sensing these explosives, most of the
works are limited to the detection of nitroaromatic
compounds.2 In contrast, due to their low-volatility and weak
binding to the sensory materials, which arises from their highly
three dimensional structures and lack of p–p interaction,
directly detecting aliphatic nitro-organics such as RDX, HMX
and PETN, in particular in the vapor phase, is very difficult and
development of a sensor for broad-class explosive detection is
also a serious challenge.3a Many current sensors are only
specic for one class of explosives, which reduce their overall
utility. Therefore, developing a chemosensor that has a broad
responsive range for different sorts of explosives would present
a signicant advance for this eld, as it would enable a richer
identication of threats.3a
Cucurbit[n]urils (CB[n], n ¼ 5–8 glycoluril units) are a class of
barrel-shaped macrocyclic hosts possessing a hydrophobic
cavity accessible through both identical carbonyl-fringed
portals for guest molecules.4 These unique structural features
make them very useful as versatile synthetic receptors and
building blocks for construction of supramolecular architec-
tures and functional chemical systems.5 In particular, in
comparison with CB[6] and CB[7] that typically bind to one
equivalent guest molecule, the larger homologue CB[8] is able to
simultaneously accommodate two guests. A more remarkable
characteristic of CB[8] is that the binding can occur sequentially
and the resulting CB[8]$X complex of the rst guest (X) then can
serve as a receptor for the second guest (Y).6 Taking advantage
of the rare binding ability of CB[8] to form a heteroternary
Key Lab of Organic Optoelectronic and Molecular Engineering, Department of
Chemistry, Tsinghua University, Beijing, 100084, P. R. China. E-mail: lgt@mail.
tsinghua.edu.cn; Tel: +86-010-62792905
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c3sc51132d
This journal is ª The Royal Society of Chemistry 2013
Chem. Sci.

View Article Online
Chemical Science
Edge Article
2 Experimental section
Synthesis of 2,6-bis[2-(2-(2-hydroxyethoxy)ethoxy)ethoxy]-
naphthalene
2,6-Dihydroxynaphthalene (2.29 g, 14.0 mmol) was added to a
suspension of K2CO3 (5.94 g, 42.9 mmol) in dry DMF (50 ml).
The suspension was heated to 70 ꢀC under nitrogen and stirred
for a further 30 min. A solution of 2-(2-(2-chloroethoxy)ethoxy)-
ethoxyethanol (5.0 g, 62.3 mmol) in DMF (10 ml) was added
dropwise over 1 h. Heating was continued for a further 3 days.
The mixture was cooled, and the solid inorganic residues were
ltered off in vacuo and washed with DMF. The DMF was
removed in vacuo, and the resultant solid was taken up into
CH2Cl2 (50 ml) and washed with saturated Na2CO3 (2 ꢁ 30 ml)
and H2O (2 ꢁ 30 ml). The organic layer was dried (MgSO4),
ltered under gravity, and the CH2Cl2 was removed in vacuo.
MS: C22H32O8 (M ¼ 424.49), ESI-ms(+): m/z (+) ¼ 447.29
[M + Na]+, 1H NMR (300 MHz, CDCl3): d 7.70 (2H, d, J ¼ 9 Hz),
7.28 (2H, S), 7.13(2H, d, J ¼ 9 Hz), 4.17 (4H, t, J ¼ 4 Hz), 3.80
(4H, t, J ¼ 4 Hz), 3.61 (4H, t, J ¼ 3 Hz), 3.56 (4H, t, J ¼ 3 Hz), 3.48
(4H, t, J ¼ 4 Hz), 3.43 (4H, t, J ¼ 4 Hz), 3.37 (2H, broad, S).
Scheme 1 Schematic illustration of the construction of CB[8]-based rotaxane on
solid substrate for selective detection and discrimination of explosives in the vapor
phase.
Synthesis of 2,6-bis(2-(2-(2-(prop-2-ynyloxy)ethoxy)ethoxy)-
ethoxy)naphthalene (NT) and 2-(2-(2-((6-(2-(2-(2-(prop-2-yn-
1-yloxy)ethoxy)ethoxy)ethoxy)naphthalene-2-yl)oxy)ethoxy)-
ethoxy)ethanol (NT-1)
complex, in this work, we reported a novel strategy for selective
detection and discrimination of explosives based on naphtha-
lene-core-threaded CB[8] rotaxane structure (Scheme 1). To
achieve a good sensing system with fast response and good
sensitivity, the synthesized CB[8]-based rotaxane structure that
is generated by a three-step approach was rst graed onto the
solid substrate surface. During the detection process, interest-
ingly, it was found that the intercalation of the explosive
compound (the third component) in the constructed rotaxane
would signicantly inuence the photophysical property of the
naphthalene core in the conned nanocavity of CB[8].
Depending on the electronic structures and the sizes of
explosive compounds, the uorescence of the naphthalene
core would be quenched or enhanced to different extent,
leading to the direct detection and discrimination of distinc-
tively different groups of trace explosives in the vapor phase,
especially including the challenging aliphatic nitro-organics
(RDX, HMX and PETN). The sensing mechanism toward the
uorescence quenching was attributed to form the host
stabilized charge transfer (CT) complexes, whereas, enhance-
ment effect was probably attributed to the ‘rigidifying effect’
where the encapsulated bulky guest molecule limited the
motions (vibration, torsion etc.) of the naphthalene core and
reduced its nal nonradiative decay processes and thus
increased the intensity of the uorescence. Control experi-
Solid NaH (362 mg, 15.1 mmol) was added to the solution of 2,6-
bis[2-(2-(2-hydroxyethoxy)ethoxy)ethoxy]naphthalene (1.60 g,
3.78 mmol) in dry DMF (50 ml). The resulting suspension was
ꢀ
stirred at 23 C until the evolution of gas ceased. A solution of
propargyl bromide in xylene (80%) by weight, (4.20 ml,
37.8 mmol) was injected using a syringe. The mixture was stir-
red at 23ꢀ for 4 days. The reaction was stopped by the addition
of MeOH, and the solvent were removed in vacuo. Crude brown
oil was rst ltered through a plug of silica, eluting with Me2CO,
and then subjected to chromatography on silica (hexane/
EtOAc ¼ 50/50) to provide a white solid. MS: C28H36O8
1
(M ¼ 500.24), ESI-ms(+): m/z (+) ¼ 523.27 [M + Na]+, H NMR
(300 MHz, CDCl3): d 7.61 (2H, d, J ¼ 9 Hz), 7.16 (2H, d, J ¼ 9 Hz),
7.10 (2H, S), 4.20–4.23 (8H, m), 3.92 (4H, t, J ¼ 5 Hz), 3.37–3.77
(m, 8H), 3.70–3.72 (m, 8H), 2.42 (2H, t, 2 Hz). MS: C25H34O8
(M ¼ 462.53), ESI-ms(+): m/z (+) ¼ 485.2 [M + Na]+, 1H NMR
(300 MHz, CDCl3): d 7.61 (2H, d, J ¼ 9 Hz), 7.16 (2H, d, J ¼ 9 Hz),
7.10 (2H, S), 4.20–4.23 (6H, m), 3.92 (4H, t, J ¼ 5 Hz), 3.70–3.80
(14H, m), 3.62 (2H, t, 4 Hz), 3.48 (1H, s), 2.42 (1H, t, 2 Hz).
Synthesis of 3-azidopropyltriethoxysilane
ments were performed to show the different sensing behaviors 3-chloropropyltriethoxysilane (5.0 g, 20.85 mmol), sodium azide
between the common organic vapors and nitrate-based (5.0 g, 76.92 mmol) and 60 ml of dry DMF were heated to 90 ꢀC
explosives, which made it easy to realize the discrimination under a N2 atmosphere for 8 h. The boiling materials were
between target analytes and interferents. Moreover, the non- removed by distillation under reduced pressure, aer which 50
covalent nature of the resulting heteroternary complex indi- ml of diethyl ether was added to the cooled mixture and the
cates that the trapped target molecules in the rotaxane struc- precipitated salts were removed by ltration. Diethyl ether
ture are facilely removable by simply washing, demonstrating removal and the oil were subjected to chromatography on silica
1
an excellent regeneration of the constructed explosive sensors to provide as a colorless liquid. H NMR (300 MHz, CDCl3): d
for real-world application.
3.83 (q, J ¼ 7 Hz, 6H; CH3CH2O), 3.27 (t, J ¼ 7 Hz, 2H; CH2N3),
Chem. Sci.
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Edge Article
Chemical Science
1.72 (m, 2H; SiCH2CH2), 1.23 (t, J ¼ 7 Hz, 9H, CH3CH2O), 0.70
(m, 2H; SiCH2).
Fabrication of CB [8]-NT lm
A mixture of CB[8] (2 ml, 0.1 mM), methyl viologen (0.3 ml,
4 mM) and NT (50 ml, 2 mM) were heated at 65ꢀ for 10 min, then
CuSO4 (15 ml, 1 mM), sodium ascorbate (15 ml, 5 mM) and DMF
(200 ml) were added to form a solution. Aer an azide-quartz
wafer was immersed into the solution for a 30 min reaction, the
CB[8]-MV2+-NT lm was formed. Before the detection process,
the molecules MV2+ were extracted to form the desired
structure.
Fabrication of NT lm as reference
A mixture of deionized water and NT (50 ml, 2 mM) were heated
at 65ꢀ for 10 min, then CuSO4 (15 ml, 1 mM), sodium ascorbate
(15 ml, 5 mM) and DMF (200 ml) were added to form a solution.
Aer an azide-quartz wafer was immersed into the solution for
30 min reaction, the NT lm was formed.
Fig. 1 (A) ESI-MS spectrum of the formed CB[8]$NT$MV host–guest complexes
in water; (B) UV-vis spectrum of the formed CB[8]$NT$MV complex in solution in
comparison with that of NT$MV without CB[8]; (C) 1H NMR spectra (600 MHz,
D2O/DMF-d7) of NT and methyl viologen in the presence of the CB[8] cavity
(1 : 1 : 1 equiv.) (top), methyl viologen (middle), and NT (bottom) in solution of
D2O/DMF-d7.
Explosives detection
The prepared lm was inserted into a glass vials (10 ml) at RT
containing solid analytes and cotton gauze, which prevented
the direct contact of the lm with the analyte and helped to
maintain a constant saturated vapor pressure. Aer exposure at
the specied time, the uorescence spectra were measured
immediately at an excitation wavelength of 239 nm.
b-protons of the viologen moiety in the CB[8]-NT-MV system
shied upeld from 8.98 to 8.89, and 8.44 to 8.23, respectively.
Also, three different protons of the naphthalene moiety shied
upeld from 7.92 to 6.59, 7.47 to 5.93, and 7.32 to 5.73,
respectively. These results clearly conrmed that the synthe-
sized naphthalene derivate is exclusively threaded in CB[8]
assisted by the viologen molecule and a strong CT complex with
the viologen moiety inside the cavity of CB[8] formed in solu-
3 Results and discussion
3.1 Characterization
Three-step approach was employed for the construction of CB tion.9 These results provided a prerequisite for the construction
[8]-based rotaxane structure on solid substrate, as shown in of the desired CB[8] rotaxane on solid surface. Finally, an azide-
Scheme 1. Scheme S1 in the ESI† shows the detailed procedure. modied quartz wafer was immersed into the solution of the
Firstly, as a threaded molecule 2,6-dioxynaphthalene derivative formed pseudorotaxane CB[8]$NT$MV for surface immobiliza-
(NT) carrying alkyne-terminated glycol chains was synthesized tion using catalyzed azide-alkyne cycloaddition. It should be
by using a standard synthetic method. Then, naphthalene NT noted that, besides the addition of a catalytic amount of CuSO4
was mixed with equimolar CB[8] along with excess methyl and sodium ascorbate, the addition of a small amount of DMF
viologen (MV) in water to form a water-soluble pseudorotaxane (5%) in the reaction medium is crucial for successful imple-
CB[8]$NT$MV. Under this condition, the equilibrium lies mentation of the mentioned click reaction. The probable reason
predominantly or overwhelming in favor of the pseudorotaxane may be that the presence of DMF offers a favorable environment
CB[8]$NT$MV owing to its extremely high overall binding for alkyne units to extend into the reaction medium. In addi-
constant up to 1012 Mꢂ2 in aqueous solution.7 In fact, only the tion, to ensure the anchoring of both terminal alkyne moieties
mass peak of the formed heteroternary complex was detected in of pseudorotaxane CB[8]$NT$MV on the surface simulta-
the performed sample without the signature signal of the neously, a lower concentration of the pseudorotaxane (10ꢂ6 M)
naphthalene NT (Fig. 1A), clearly indicating that all naphtha- was used in our case. Indeed, under optimized preparation
lene is exclusively threaded in CB[8], assisted by viologen conditions, the construction of the desired NT-threaded rotax-
molecule. Upon the mixing of these three components, the ane on a solid substrate was achieved. Aer the extraction of MV
color of the solution was changed from colorless to light purple molecules, the uorescence of the naphthalene core in the
(Fig. S1†). UV/vis spectroscopy also showed the typical forma- cavity of CB[8] was restored (Fig. 2A), and the resulting CB[8]$NT
tion of the charge transfer (CT) interaction at 524 nm between complex can act as a receptor for the reversible binding of
MV and naphthalene mediated by CB[8] (Fig. 1B), conrming proper guest molecules as demonstrated by using MV (Fig. 2B).
that the naphthalene core of the compound NT stayed in the
In our work, as a control experiment, the synthesized
cavity of CB[8].8 To further conrm the formation of the 1 : 1 : 1 naphthalene derivative without CB[8] was also anchored on the
inclusion complex, NMR technique has been employed. In surface using the same procedure, whereas none of the
comparison with the 1H NMR of MV (Fig. 1C), the a-and phenomena described above were observed. Herein, both of the
This journal is ª The Royal Society of Chemistry 2013
Chem. Sci.

View Article Online
Chemical Science
Edge Article
Fig. 3 Comparison of the high-resolution XPS spectra of the N1s C 1s and O1s of
azide-modified substrate before (A–C) and after (A0–C0) click reaction.
Fig. 2 (A) Fluorescence evolution of CB[8]$NT$MV structure on a substrate with
the gradual removal of the trapped MV molecule; (B) reversible change of the
fluorescence of the CB[8]-NT-MV structure on the substrate upon trapping and
removal of an MV molecule; (C) fluorescence emission of the film with the
rotaxane structure where the naphthalene core is in the confined nanocavity of
CB[8]; (D) fluorescence emission of the film with pseudorotaxane structure after
the extraction of MV molecule; (E) schematic illustration of the extraction process
to demonstrate the unstable pseudorotaxane structure.
structure was observed as shown in Fig. 3A. The double-peak
structure is indicative of the azide as the smaller peak is seen at
404.5 eV and the larger peak is seen at 400.3 eV, in agreement
with previously reported results for azide peak location.11 The
ratio of areas of the two peaks is not exactly 1 : 2 as expected for
surface azides, the difference in areas can be attributed to the
well-known degradation of the middle nitrogen during the XPS
terminal alkyne groups of the pseudorotaxane CB[8]$NT$MV analysis process. The C1s signal (Fig. 3B) was deconvoluted and
that were simultaneously attached on the surface were the key tted to two peaks assigned to C–C/C–H (284.6 eV), and C–N3
factor to obtain a stable structure for analyte sensing. To prove (285.7 eV). The O1s signal (Fig. 3C) was tted to one peak
this point, the attachment of one terminal alkyne moiety (NT-1) (532.8 eV), which was consistent with the presence of the Si–O
of pseudorotaxane CB[8]$(NT-1)$MV onto the substrate was species on the surface. All these results prove a good graing of
carried out. In this case, the extraction of the MV molecule to azide functional groups on the substrate surface by treatment
form the (NT-1)-threaded CB[8] structure also caused the using 3-azidopropyltriethoxy silane. The subsequent click
removal of the CB[8] macrocycle, due to the weak hydrophobic reaction with the formed pseudorotaxane CB[8]$NT$MV led to a
interactions between the naphthalene core and CB[8] cavity. As clear change in the XPS spectra (Fig. 3A0–C0). The N1s signal of
shown in Fig. 2D, the characteristic emission peak at 420 nm the azide group became weaker, and the N1s signal (400.3 eV) of
which represents the presence of the NT–NT excimer the 1,2,3-triazole formed during the cycloaddition click reaction
appeared,10 indicating that the naphthalene cores were no is observed (Fig. 3A0). Further, the carbon signal (288.5 eV) of the
longer isolated by the CB host molecule and thus the formation formed 1,2,3-triazole is also detectable (Fig. 3B0). The character-
of the unstable and undesired pseudo-rotaxane structure istic XPS signals of the threaded CB[8] molecules are found at
(Fig. 2E).
289.1 eV for C–N (Fig. 3A0), at 289.1 eV for C]O (Fig. 3B0) and at
The attachment of the constructed sensor structure on a 531.9 eV for C]O (Fig. 3C0), respectively, which are consistent
solid substrate was rstly conrmed by QCM technique. As with the values in literature.12 Following the similar data pro-
shown in Fig. S2,† it was found that, upon addition of the cessing method in Gopalan's work,13 the surface density of CB[8]-
complex in the reaction medium, a fast negative shi (ca. 3 Hz) based rotaxanes was calculated to be about 5.61 ꢁ 1013 molecules
of frequency was observed, indicative of the immobilization of per cm2, less than the monolayer coverage.
the threaded complex (CB[8]$NT$MV) onto the azide-function-
The presence of the threaded CB[8] molecule on the surface
alized substrate. Moreover, the constructed rotaxane structure was further corroborated by Fourier-transform Infrared Reec-
was further conrmed by using XPS technique. Fig. 3 shows the tion Absorption Spectroscopy (FT-IRRAS). The characteristic
comparison of XPS data of the azide-modied quartz wafer C–N and C]O stretch absorptions of the immobilized CB[8]
before and aer the click reaction described above. In the case molecules are clearly evident at 1442 cmꢂ1 and 1714 cmꢂ1
,
14
of the used substrate, the characteristic azide double-peak respectively (Fig. S3 in ESI†). Also, the dominating peak at
Chem. Sci.
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Edge Article
Chemical Science
1105 cmꢂ1, which was assigned to the C–O–C stretch, conrms formation of host stabilized charger transfer complexes, the
the existence of the NT molecule. It should be noted that the constructed sensor shows a fast response. Unexpectedly, the
XPS signal of the terminal alkyne is not detectable in the per- presence of trace aliphatic explosives like RDX and HMX
formed XPS analysis, indicating that under the optimized click induced the opposite response of the optical property of the
reaction conditions described above both terminal alkyne naphthalene, and instead of quenching, uorescence
groups of the pseudorotaxane CB[8]$NT$MV were simulta- enhancement to varying degrees was detected (Fig. 4C). In the
neously attached on the surface, affording the desired rotaxane case of bulky PETN, however, the uorescence of the naphtha-
structure as shown in Scheme 1. Indeed, thoroughly washing lene core in CB[8] remains unchanged even with prolonged
the modied substrate using different solvents didn't lead to exposure (Fig. 4D). As expected for the surface sensing
any changes of the XPS signals of the attached CB[8] molecules. elements, all described responses are fast and take several
seconds. Compared with the reported explosive sensors, our
constructed sensor also exhibits a good sensitivity.1–3 In the case
3.2 Explosive detection and discrimination
of TNT in vapor phase, the detection limit could reach ppt level,
as shown in Fig. S6.† Moreover, the trapped explosive molecules
in the rotaxane structure can be facilely removed by simply
washing, and the sensing ability of the constructed rotaxane is
completely regenerable (Fig. 4F). As a control experiment, the
synthesized naphthalene derivative (NT) without CB[8] macro-
cycle was attached on the substrate using the same preparation
procedure in this work. The distinctly dissimilar sensing
behavior described above was not observed in the control case
(Fig. S5†), indicating that the conned nanospace provided by
CB is critical for the sensing and discrimination of explosives in
the constructed rotaxane structure. In our work, the secondary
effect arising from the impurity in the used explosives was
excluded. All the obtained results were reproducible.
Usually, the observed uorescence attenuation and
enhancement can be explained by the donor–acceptor electron-
transfer mechanism. Scheme S2 in ESI† displays the band gap
energies of the donor (naphthalene NT) and the acceptors (TNT,
DNT, PETN, RDX, HMX), which were calculated by using
density-functional theory (DFT) at the B3LYP/6-31G(d) level for
the optimized geometries.15 It is clear that for all the tested
analytes the energy gaps are large enough to provide an
appropriate thermodynamic driving force and photoinduced
electron transfer should be possible between the NT and
explosive molecules resulting in uorescence quenching of the
former (NT). In practice, however, naphthalene is hardly used
for the detection of explosives, and an effective quenching
effect is not detectable as conrmed in our control experi-
ments (Fig. S5†). In contrary, in the case of the naphthalene
threaded CB[8] rotaxane (Scheme 1) which is a kind of host
stabilized charge transfer complex, the connement of the
analyte (e.g. TNT or DNT) inside the CB[8] nanocavity may
promote the close association of aromatic explosive molecules
with a naphthalene core through p–p interactions, and
thereby facilitates an efficient electron transfer process
resulting in uorescent quenching of naphthalene. Different
from aromatic explosives, another mechanism should be
responsible for the observed uorescence enhancement in the
case of aliphatic counterparts. Due to their three-dimensional
molecular structure and lack of p–p interaction and subse-
quently tight binding interaction,3 the restricted nanospace
provided by CB[8] cannot lead to an efficient orbital overlap of
the naphthalene core and the trapped aliphatic explosives
necessary for electron transfer. But, the incorporation of such
a bulky guest in the constructed CB[8] rotaxane reduces the
The naphthalene core in the conned nanocavity of CB[8]
retains its optical properties as in bulky solution (Fig. S4†) and
exhibits strong uorescence emission in the range from 350 nm
to 450 nm. Interestingly, we found that the naphthalene core-
threaded CB[8] rotaxane structure on solid surface can serve as a
unique receptor for selective detection and discrimination of
explosives. In our work, the CB[8] rotaxane-attached substrates
were exposed to saturated vapors of different explosives at
room temperature, including TNT (5.8 ꢁ 10ꢂ6 torr) and DNT
(1.1 ꢁ 10ꢂ4 torr),1c for varied periods of time (10 s, 30 s, 60 s,
120 s, 300 s, 600 s), and PETN (1.4 ꢁ 10ꢂ8 torr), RDX (4.6 ꢁ 10ꢂ9
torr) and HMX (8.0 ꢁ 10ꢂ11 torr),1c for varied periods of time
(0 min, 1 min, 3 min). As shown in Fig. 4, depending on the
explosive used, the constructed rotaxane shows a clearly
different optical response. For nitroaromatics (TNT and DNT),
the uorescence of the naphthalene core was gradually
quenched with increased exposure (Fig. 4A and B). Due to the
Fig. 4 Time-dependent fluorescence intensity of the NT-threaded CB[8] rotax-
ane on solid substrate upon exposure to saturated vapor TNT (A), DNT (B), HMX
(C) and PETN (D) at room temperature; (E) the detected fluorescence attenuation
and enhancement of TNT, DNT, HMX, RDX and PETN for time of 3 min; (F)
regeneration of the constructed sensors in case of TNT.
This journal is ª The Royal Society of Chemistry 2013
Chem. Sci.

View Article Online
Chemical Science
Edge Article
free-volume of the CB[8] conned space, and actually exerts a
similar rigidifying effect on the motions (vibrations, torsions
etc.) of the naphthalene core in the CB[8] cavity. Then, non-
radiative decay processes should be slowed and the fraction of
excited species decaying radiatively should increase, resulting
in increased intensity of uorescence.16 Clearly, the degree of
the observed enhancement should be dependent on the size of
the included guest. Besides the connement effect, the size
exclusion effect of the carbonyl-fringed portals in CB[8]
structure provides another mechanism of molecular selec-
tivity. The guest molecules that are smaller than the CB[8]
apertures can be included, but larger molecules cannot. In
Table 1, the molecular size of a series of guest molecules and
CB[8] had been given. Indeed, in agreement with the obtained
3
˚
results the explosives with smaller size, e.g. TNT (167.75 A ),
3
3
3
˚
˚
˚
DNT (150.12 A ), HMX (201.21 A ), RDX (152.51 A ) can get into
the nanocavity of the CB[8] macrocycle, leading to different
changes of the photophysical properties of the naphthalene
core depending on the electronic structures and the molecular
sizes of the explosives. Conversely, for larger molecules like
Fig. 5 Optimized structures of CB[8]-NT-nitrated explosives (TNT, DNT, RDX and
HMX) based on MM calculations.
3
˚
PETN (244.69 A ), the size exclusion of CB[8] makes no
opportunity for such an analyte to enter the interior of CB[8] the NT molecule was replaced by 2,6-dihydroxynaphthalene for
and blockades the optical response of the naphthalene inside. clarity.
In addition, large naphthalenediimide molecule, which is a
To provide new evidence for the proposed mechanism
well-known effective quencher for naphthalene,17 was also described above, thioavin T (ThT), a well-known chromophore
employed to test the responsive behavior of the protected for its dramatic uorescent enhancement upon binding to a
naphthalene in CB[8] cavity.
restricted host space,18 was employed to construct sensor
As expected, due to its larger molecular size, the size exclu- device. It is found that signicant enhancement of sensing
sion of CB[8] inhibits such an analyte from entering the interior performance was detected with ThT-based structure in Fig. S9.†
of CB[8] and blockades the optical response of the naphthalene In comparison with the initial result, uorescence response
inside (Fig. S7†). This result implies that, based upon molecular increased by 434% in the case of RDX. This result clearly indi-
size and shape, the described ndings can be also utilized to cates that the strategy described in our paper could afford more
discriminate a couple of different classes of compounds besides sensitive and competent sensors with enhanced discrimination
explosives. In Fig. 5, the optimized structures of CB[8]-NT-TNT, of trace explosives when appropriate exible chromophores are
DNT, RDX, and HMX had been calculated based on MM integrated with CB[8] macrocycle.
calculations. It was found that due to the enhanced intermo-
In this work, control experiments were performed by using
lecular interactions by CB cavity, the optimized structure of the conventional organic vapors (e.g. toluene, THF, CH2Cl2 and
ternary structure proved the possibility of our CB[8]-NT system ethanol) as interferents. We found that the used organic vapors
for explosive sensing. Moreover, the formation of complexation could induce, to some extent, the change of the uorescence
of CB[8]-naphthalene-RDX and CB[8]-naphthalene-HMX could intensity of the protected naphthalene in CB[8] cavity
also be conrmed by ESI-MS measurements (Fig. S8†), where (Fig. S10†). However, quite different from the phenomena
observed in the cases of explosives, the uorescence of the
protected naphthalene in CB[8] immediately recovered
upon the exposure of the used sensors to air. This result indi-
Table 1 Molecular size of the guest molecules and CB[8]
cates that the detection of the explosives can be easily
discriminated from the interferent organic vapors.
˚
a/A
˚
b/A
˚
c/A
The preliminary results described above are encouraging.
Based on the unique connement effect and size exclusion of
CB[8] in the constructed rotaxane structure, direct detection
and discrimination of distinctively different groups of trace
explosives in the vapor phase are realized by using one single
host structure, especially including the challenging aliphatic
nitro-organics (RDX, HMX and PETN). The surface-anchored
sensing element is especially favorable to sensing efficiency.
Compared with the reported explosive sensors, our constructed
sensor also exhibited a good sensitivity. However, like most of
other sensors reported in the literature, it may still be difficult to
TNT
DNT
RDX
HMX
7.91
7.97
6.91
8.83
8.82
7.83
6.53
6.28
7.24
9.14
7.74
2.85
2.74
4.74
4.84
6.80
1.00
PETN
Naphthalene-
diimide
CB[8]a
10.03
6.9
8.8
9.1
a
The values quoted for a, b, and c for CB[8] take into account the van der
Waals radii of the relevant atoms. The size of other molecules are
determined based on the optimized structure which are calculated by
using density-functional theory at the B3LYP/6-31G(d) level.
Chem. Sci.
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Edge Article
Chemical Science
use one single sensor or method to realize the detection and
discrimination of all the explosives in a real complicated envi-
ronment. Mimicking the mammalian olfactory system, the
construction of sensor array using the developed sensor as
sensor element may be the best and most reliable way to achieve
the detection and discrimination in a complicated environ-
ment. The related work is ongoing in our lab. Additionally,
although our new concept was tested with CB[8], in principle it
should be used as a general and effective sensing scheme for
creating various sensors in other synthetic molecular-sized
nanocavities, for example in metal-organic frameworks (MOFs).
P. Anzenbacher Jr, L. Mosca, M. A. Palacios, G. V. Zyryanov
and P. Koutnik, Chem.–Eur. J., 2012, 18, 12712–
12718.
2 Y. Salinas, R. Martinez-Manez, M. D. Marcos, F. Sancenon,
A. M. Costero, M. Parra and S. Gil, Chem. Soc. Rev., 2012,
41, 1261–1296.
3 (a) M. E. Germain and M. J. Knapp, Chem. Soc. Rev., 2009, 38,
2543–2555; (b) A. Lan, K. Li, H. Wu, D. H. Olson, T. J. Emge,
W. Ki, M. Hong and J. Li, Angew. Chem., 2009, 121, 2370–
2374; Angew. Chem., Int. Ed., 2009, 48, 2334–2338; (c)
A. D. Hughes, I. C. Glenn, A. D. Patrick, A. Ellington and
E. V. Anslyn, Chem.–Eur. J., 2008, 14, 1822–1827; (d)
T. L. Andrew and T. M. Swager, J. Am. Chem. Soc., 2007,
129, 7254–7255; (e) M. E. Germain and M. J. Knapp, J. Am.
Chem. Soc., 2008, 130, 5422–5423; (f) S. Pramanik,
C. Zheng, X. Zhang, T. J. Emge and J. Li, J. Am. Chem. Soc.,
2011, 133, 4153–4155.
4 (a) J. W. Lee, S. Samal, N. Selvapalam and K. Kim, Acc. Chem.
Res., 2003, 36, 621–630; (b) L. Lagona, P. Mukhopadhyay,
S. Chakrabarti and L. Isaacs, Angew. Chem., 2005, 117,
4922–4949; Angew. Chem., Int. Ed., 2005, 44, 4844–4871; (c)
D. T. Dang, Chem. Sci., 2012, 3, 2679–2684; (d)
F. Biedermann, U. Rauwald, J. M. Zayed and
O. A. Scherman, Chem. Sci., 2011, 2, 279–286; (e) R. Hota,
K. Baek, G. Yun, Y. Kim, H. Jung, K. M. Park, E. Yoon,
T. Joo, J. Kang, C. G. Park, S. M. Bae, W. S. Ahn and
K. Kim, Chem. Sci., 2013, 4, 339–344.
5 (a) W. M. Nau and O. A. Scherman, Isr. J. Chem., 2011, 51,
492, special issue for cucurbit[n]uril chemistry; (b)
M. M. Boyle, R. A. Smaldone, A. C. Whalley,
M. W. Ambrogio, Y. Y. Botros and J. F. Stoddart, Chem.
Sci., 2011, 2, 204–210.
6 (a) H. J. Kim, H. Heo, W. S. Jeon, E. Lee, J. Kim, S. Sakamoto,
K. Yamaguchi and K. Kim, Angew. Chem., 2001, 113, 1574–
1577; Angew. Chem., Int. Ed., 2001, 40, 1526–1529; (b)
V. Ramalingam and A. R. Urbach, Org. Lett., 2011, 13,
4898–4901.
4 Conclusions
In summary, based on a NT-threaded CB[8] rotaxane structure
on solid substrate, a new strategy for rapid, fully reversible, and
highly sensitive detection of a broad class of explosives was
developed by using one receptor. Due to the unique conne-
ment effect and size exclusion of CB[8] nanocavity, it is found
that, depending on the electronic structures and the sizes of
explosives, the photophysical property of the naphthalene core
in the conned CB[8] cavity shows different changes upon
contact with the tested analyte, leading to the direct detection
and discrimination of different groups of trace explosives,
including the challenging aliphatic nitro-organics (RDX, HMX
and PETN). In recent years, a uorescence-based sensing
scheme has been extensively explored for the detection of
explosives. However, chemosensors capable of detection and
discrimination of a broad range for explosives classes, espe-
cially in the vapor phase, are rare.3e Therefore, we believe that
our ndings would open a new way to design and develop a new
kind of explosive sensors enable a richer identication of
threats.
Acknowledgements
We gratefully acknowledge the nancial support from the
National Science Foundation of China (no. 21025311, 21121004,
21261130581 and 91027016), Ministry of Education
(2011Z01014), MOST Program (2011CB808403, 2013CB834502)
and the transregional project (TRR61). We also thank Prof. Feng
Liu in Perking University for the QCM characterization.
7 (a) U. Rauwald and O. A. Scherman, Angew. Chem., 2008, 120,
3904–3907; Angew. Chem., Int. Ed., 2008, 47, 3950–3953; (b)
Y. Liu, Y. Yu, J. Gao, Z. Wang and X. Zhang, Angew. Chem.,
2010, 122, 6726–6729; Angew. Chem., Int. Ed., 2010, 49,
6576–6579.
8 A. Trabolsi, M. Hmadeh, N. M. Khashab, D. C. Friedman,
M. E. Belowich, N. Humbert, M. Elhabiri, H. A. Khatib,
A. Albrecht-Gary and J. F. Stoddart, New J. Chem., 2009, 33,
254–263.
9 D. Zou, S. Andersson, R. Zhang, S. Sun, B. Akermark and
L. Sun, J. Org. Chem., 2008, 73, 3775–3783.
Notes and references
1 (a) S. W. Thomas, G. D. Joly and T. M. Swager, Chem. Rev.,
2007, 107, 1339–1386; (b) D. T. McQuade, A. E. Pullen and
T. M. Swager, Chem. Rev., 2000, 100, 2537–2574; (c)
J. C. Sanchez and W. C. Trogler, J. Mater. Chem., 2008, 18, 10 J. S. Yang and T. M. Swager, J. Am. Chem. Soc., 1998, 120,
3143–3156; (d) S. J. Toal and W. C. Trogler, J. Mater. Chem.,
11864–11873.
2006, 16, 2871–2883; (e) T. Naddo, Y. Che, W. Zhang, 11 S. Prakash, T. M. Long, J. C. Selby, J. S. Moore and
K. Balakrishnan, X. Yang, M. Yen, J. Zhao, J. S. Moore and
M. A. Shannon, Anal. Chem., 2007, 79, 1661–1667.
L. Zang, J. Am. Chem. Soc., 2007, 129, 6978–6979; (f) Y. Che, 12 Q. An, G. Li, C. Tao, Y. Li, Y. Wu and W. Zhang, Chem.
D. E. Gross, H. Huang, D. Yang, X. Yang, E. Discekici, Commun., 2008, 1989–1991.
Z. Xue, H. Zhao, J. S. Moore and L. Zang, J. Am. Chem. Soc., 13 P. Paoprasert, J. W. Spalenka, D. L. Peterson, R. E. Ruther,
2012, 134, 4978–4982; (g) A. Ponnu, N. Y. Edwards and
E. V. Anslyn, New J. Chem., 2008, 32, 848–855; (h)
R. J. Hamers, P. G. Evans and P. Gopalan, J. Mater. Chem.,
2010, 20, 2651–2658.
This journal is ª The Royal Society of Chemistry 2013
Chem. Sci.

View Article Online
Chemical Science
Edge Article
14 K. Kim, J. W. Lee, Y. H. Ko and K. Kim, Chem. Commun.,
2004, 848–849.
A. C. Bhasikuttan, H. Pal and J. Mohanty, Chem. Commun.,
2011, 47, 9959–9971.
15 (a) A. D. Becke, J. Chem. Phys., 1993, 98, 5648; (b) C. Lee, 17 S. V. Bhosale, C. H. Jani and S. J. Langford, Chem. Soc. Rev.,
W. Yang and R. G. Parr, Phys. Rev. B, 1988, 37, 785. 2008, 37, 331–342.
16 (a) L. E. Kreno, K. Leong, O. K. Farha, M. Allendorf, R. P. Van 18 L. M. Tolbert, A. Baldridge, J. Kowalik and K. M. Solntsev,
Duyne and J. T. Hupp, Chem. Rev., 2012, 112, 1105–1125; (b) Acc. Chem. Res., 2012, 45, 171–181.
Chem. Sci.
This journal is ª The Royal Society of Chemistry 2013