Scaffold-optimized dendrimers for the detection of the triacetone triperoxide explosive using quartz crystal microbalances

Article abstract of DOI:10.1002/cplu.201100080

In conclusion, the preparation of novel polyphenylene-type dendrimers led to affinity materials superior to the gold standard for TATP tracing by QCM. With incorporation of pyrenyl and cyanophenyl units not only the affinity to TATP was dramatically enhan

Full text of DOI:10.1002/cplu.201100080

DOI: 10.1002/cplu.201100080
Scaffold-Optimized Dendrimers for the Detection of the Triacetone
Triperoxide Explosive Using Quartz Crystal Microbalances
Daniel Lubczyk,[a] Matthias Grill,[b] Martin Baumgarten,[b] Siegfried R. Waldvogel,*[a] and Klaus Mꢀllen*[b]
Triacetone triperoxide (TATP; Scheme 1) is an explosive made
from common household chemicals. The sensitivity towards
shock, friction, and electrostatic ignition makes this explosive
frequency QCM (ca. 200 MHz) can be coated with affinity mate-
rials and enable the detection of airborne sources on a pico-
gram scale within seconds.[7] If a suitable coating is chosen,
then high sensitivity and selectivity is given. The affinity mate-
rial deposited on the electrode surface of the QCM creates a
signal based on the desired chemisorption upon binding of
the analyte.[8a–c] In general, this interaction is interfered with an
unselective physisorption,[8b] which is often dominated by hu-
midity. However, the key for a powerful sensor based on QCM
requires coating exhibiting a high affinity and selectivity for
the desired analyte.[9] The enhanced quality of such coatings
facilitates subsequent application in sensor arrays involving
several QCM.
Consequently, the affinity material plays an utmost and cru-
cial role which is the key within this interdisciplinary sensor
project. The supramolecular interaction of the analyte with the
affinity material exploits the reversible nature of noncovalent
binding and provides a fast recovery of the sensor in an air
stream.
Scheme 1. Structure of triacetone triperoxide (TATP) and the reference den-
drimer 1.[10]
In previous studies, we identified a variety of suitable affinity
materials for TATP with a complementary cross-affinity to inter-
fering compounds.[6] Among these coatings, polyphenylene
dendrimers turned out to exhibit the strongest binding capa-
bility for TATP while the signals for H2O and H2O2 are very low.
The best properties were found with 1 (Scheme 1). The out-
standing affinity was attributed to the complementary voids in
the dendrimer and binding through CH–p- and dipole–dipole
interactions on the pyridine moieties at the interior of the den-
dritic core–shell architecture.
extremely dangerous.[1] TATP and related compounds have ex-
perienced significant attention as a threat in terrorist attacks.[2]
It is considered as the prime material among home-made ex-
plosives.[1,3] Compared to other explosives TATP exhibits an un-
usually high vapor pressure of approximately 68–78 ppm
under ambient conditions.[4] Therefore, the direct tracing of air-
borne material should be possible. The majority of analytical
methods for the detection of TATP relies on the analysis of
H2O2 after hydrolysis or dissociation of the explosive.[5a–f] For
such a detection of TATP, two major drawbacks occur: First, ini-
tial removal of H2O2 which is ubiquitous in our daily life;
second, longtime sampling of analyte in order to have suffi-
cient material for the detection process. These approaches are
time-consuming and not feasible for a low-cost sensor technol-
ogy. Only few methods directly detect TATP.[5g–i] Recently, our
research group has demonstrated that quartz crystal microba-
lances (QCM) are a suitable low-cost technology platform for a
continuously working sensor to trace TATP.[6] High fundamental
Herein, we present a significant improvement in affinity and
selectivity by molecular variation of the new dendritic poly-
phenylene-structures by equipping the inner shells with di-
verse substituents (pyrenyl, pyridyl, nitro, cyano, and amide
moieties). Dendrimers represent a unique class of macromole-
cules because they offer a branched architecture with multiple
end groups and a defined molecular structure.[11] In the past
decades a broad variety of repetitive reaction sequences have
been employed for the construction of such systems.[12] In con-
trast to other dendrimers containing conformationally very
flexible single bonds, the rigidity of the dendritic polypheny-
lene scaffold limits the conformational freedom because only
confined rotation around the biaryl axis can occur.[13] Owing to
their unique shape persistence and monodispersity, polyphen-
ylene dendrimers proved to be perfect candidates for host–
guest chemistry.[14]
[a] D. Lubczyk, Prof. Dr. S. R. Waldvogel
Institut fꢀr Organische Chemie, Johannes Gutenberg-Universitꢁt Mainz
Duesbergweg 10-14, D-55128 Mainz (Germany)
Fax: (+49)613-39-26777
E-mail: waldvogel@uni-mainz.de
[b] Dr. M. Grill, Prof. Dr. M. Baumgarten, Prof. Dr. K. Mꢀllen
Max-Planck-Institut fꢀr Polymerforschung
Ackermannweg 10, D-55128 Mainz (Germany)
Fax: (+49)6131-379350
For the studies a collection of new polyphenylene dendrim-
ers was synthesized containing a variety of different substitu-
ents at the internal branches of the dendrimer architecture.
The incorporated moieties are mostly polar and electron with-
drawing. The synthetic pathway to these TATP attractive den-
E-mail: muellen@mpip-mainz.mpg.de
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cplu.201100080.
102
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2012, 77, 102 – 105

drimers exploits a reliable and established strategy.[15] Particular
attention has to be given to the cyclopentadienone starting
materials S2–S6 (Scheme 2) which provide upon cycloaddition
and subsequent extrusion of carbon monoxide branching into
the dendrimer scaffold. The tetraarylcyclopentadienones S2–S6
were prepared starting from the corresponding 1,2-diketones
in a Knoevenagel condensation.[16] The molecular variation of
this building block is focused onto the positions 2 and 5 of the
cyclopentadienone (marked as R in Scheme 2), whereas the re-
maining aryl moieties are equipped with alkyne moieties. After
removal of the silyl groups, an iterative installation of subse-
quent molecular shells was performed. It was demonstrated
that tetrahedral polyphenylene dendrimers based on a tetra-
phenylmethane core form cavities,[17] which display promising
shape persistent voids for the uptake of different guests. This
fundamental design was maintained throughout our studies.
After successful synthesis of the branching units S2–S6, itera-
tive assembly of specific dendritic scaffolds was accomplished
by a repetitive Diels–Alder approach and subsequent desilyla-
tion sequences. Thus dendritic macromolecules could be pre-
pared divergently to the third generation. As depicted in
Scheme 2, the sixteen terminal alkynes in the periphery of 11,
were then capped by reaction with tetraphenylcyclopentadie-
none. The amide dendrimers 7 were obtained by amidation of
the corresponding esters 6. This gives access to 2b–7b, which
exhibit a nonpolar outer shell. The novel G2 dendrimers 4a
contain a doubled number of in-
corporated pyridine entities per
generation compared to 1. This
aspect was of particular interest
because the performance of pre-
vious affinity materials for quartz
micro balances was attributed to
this specific moiety. With an in-
creased content an ameliorated
inclusion behavior towards TATP
was anticipated.
For evaluation of the affinity
capabilities of each dendrimer,
quartz crystal microbalances
with a fundamental frequency of
195 MHz were coated with 2–7.
The coating was done by an
electrospray protocol.[18] This
particular method guarantees a
highly reliable and reproducible
deposition of material in the
nanogram range. For determina-
tion of the affinities to a respec-
tive analyte, concentration pro-
files were measured at 358C.
Therefore, a nitrogen flow was
enriched with a defined partial
pressure of analyte gas and
passed on the coated QCM.
From concentration profiles the
adsorption isotherms can be ob-
tained.[17a] The slope of such an
isotherm is directly proportional
to the affinity of the analyte to
the coating material.[19] All den-
drimers, except 2a, 6a, and 4b,
could be applied as coating ma-
terial for the QCM. The dendrim-
ers 2a, 6a, and 4b displayed vis-
coelastic effects during the coat-
ing procedure. Consequently, it
was not possible to obtain a ho-
mogeneous coating using these
Scheme 2. Synthetic pathway towards substituted polyphenylene dendrimers.
compounds. Furthermore, the
ChemPlusChem 2012, 77, 102 – 105
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chempluschem.org
103

standard of the previous studies 1 (Scheme 1) and polypheny-
lene analogue (12, structure in the Supporting Information)
were investigated. With 12 the effect of the functionalizations
of the inner shells was explored. The affinities of the dendrimer
coatings for TATP are displayed in Figure 1. Dendrimers 4a, 5,
2b, 3b, and 7b show a significantly higher affinity for TATP
nation of TATP over potentially interfering compounds. Thus, a
selection of analytes was screened, including acetone and hy-
drogen peroxide, as starting materials for the preparation of
TATP. If no strong acid is present TATP will not be in equilibri-
um with the starting materials. Bis-tert-butyl peroxide served
as a lipophilic and structurally related peroxide. Water repre-
sents an omnipresent compound in real-life scenarios. From
the individual affinities the selectivity (S) in respect to TATP
was calculated:[6,20]
Affinityanalyte
AffinityTATP
S ¼
Only the dendrimers with a considerably higher affinity to-
wards TATP than 1 are depicted with their selectivity in
Figure 2. The relative intensity for TATP with the quantity of 1.0
is not displayed in Figure 2. Almost all selected dendrimers,
except 2b and 12, show the same low affinity for acetone. The
lowest affinities for bis-tert-butyl peroxide are seen for the den-
drimers 5 and 3b. In addition to this, 5 exhibits the lowest
Figure 1. Slopes of adsorption isotherms for TATP. Best known system poly-
phenylen-pyridyl dendrimer 1 is indicated by the horizontal line.
than reference material 1. The effect of different dendrimer
generations can be observed for two sets of affinity materials
(3a vs. 3b and 7a vs. 7b, respectively). These pairs reveal that
the G3 dendrimers, with the larger scaffolds and more pro-
nounced voids, provide the higher affinity. Interestingly, the
pyrene-functionalized G2 compound 5 ranges among the best
affinity materials despite the expanded aromatic substituents
inside the dendrimer. Owing to the spatial demand of the pyr-
enyl moieties the large voids in the dendritic scaffold should
be more structured. The polycyclic aromatic groups represent
electron-rich walls, which provide better niches for accommo-
dating the nonpolar analytes by CH–p contacts (see figures ob-
tained by computational chemistry in the Supporting Informa-
tion). Consequently, the relatively small G2 compound 5 is ca-
pable of accumulating TATP with an outstanding efficiency. An
enhanced affinity for TATP by a factor of two seems at a first
sight not significant but improvement of such a highly potent
system facilitates the later sensor applications tremendously.
Additional aspects that influence the dendrimer affinities to-
wards analytes are: 1) their electronic properties and thus CH–
p interactions between the dendrimer and incorporated guest,
and 2) the possibility of hydrogen bonds. The electron-defi-
cient nature of the inner shells has an obvious positive effect
on the affinity. Therefore, nitro-, cyano-, and amide-modified
G3 systems (2b, 3b, and 7b) reveal higher affinities than the
unmodified dendrimer 12. In addition, less electron-rich spe-
cies show dramatically enhanced long term stability because
the QCM are operated at ambient conditions. The coating will
consist of a few molecular layers, which might be oxidatively
degraded.
Figure 2. Selectivity of the best affinity materials normalized on TATP. For
TATP the relative intensity is S=1.0 and is omitted for clarity.
cross-affinities for water as well as hydrogen peroxide. This re-
markable selectivity of 5 is not only attributed to the absence
of heteroatoms because the same applies to 12 which show
the highest affinity towards the four interference analytes. The
dendrimers 5 and 3b combine the highest affinity for TATP
with the lowest cross-sensitivities, at which 5 has the smaller
affinity especially for water, which is the most common inter-
ference for artificial noses. Consequently, compound 5 is the
most promising candidate for application in a TATP sensor.
In conclusion, the preparation of novel polyphenylene-type
dendrimers led to affinity materials superior to the gold stan-
dard for TATP tracing by QCM. With incorporation of pyrenyl
and cyanophenyl units not only the affinity to TATP was dra-
matically enhanced but also the selectivity over interfering
compounds was ameliorated allowing an exquisite discrimina-
tion for TATP. Furthermore, the downsized dendrimers 4a and
5 are capable of outperforming larger structures in terms of
overall TATP affinity, which was so far believed to be the spe-
cialty of widely branched nanoparticles.[6] Thus, the short syn-
For a sensor application the affinity materials should exhibit
both, affinity and selectivity. The latter is necessary for discrimi-
104
www.chempluschem.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2012, 77, 102 – 105

Dordrecht, 2007, p. 217; i) U. Willer, W. Schade, Anal. Bioanal. Chem.
2009, 395, 275–282.
[6] D. Lubczyk, C. Siering, J. Lçrgen, Z. B. Shifrina, K. Mꢀllen, S. R. Waldvo-
gel, Sens. Actuators B 2010, 143, 561–566.
[7] C. Kreutz, J. Lçrgen, B. Graewe, J. Bargon, M. Yoshida, Z. Fresco, J. M.
Frechet, Sensors 2006, 6, 335–340.
[8] a) A. Janshoff, H.-J. Galla, C. Steinem, Angew. Chem. 2000, 112, 4164–
4195; Angew. Chem. Int. Ed. 2000, 39, 3213–3237; b) F. L. Dickert, P. A.
Lieberzeit in Springer Series on Chemical Sensors and Biosensors, Vol. 5
(Eds., C. Steinem, A. Janshoff), Springer-Verlag, Berlin, Heidelberg, 2007;
c) J. W. Loergen, C. Kreutz, J. Bargon, P. Krattiger, H. Wennemers, Sens.
Actuators B 2005, 107, 366–371.
[9] L. Baldini, M. Melegari, V. Bagnacani, A. Casnati, E. Dalcanale, F. Sansone,
R. Ungaro, J. Org. Chem. 2011, 76, 3720–3732; B. Graewe, A. Rang, C. A.
Schalley, J. Haubrich, J. Bargon, Sens. Actuators B 2006, 119, 302–307.
[10] Z. B. Shifrina, M. S. Rajadurai, N. V. Firsova, K. Mꢀllen, Macromolecules
2005, 38, 9920–9932.
thesis of small highly functionalized dendrimers is more cost
and time efficient compared to the preparation of
G4 dendrimers. The nonpolar and pyrenyl containing dendri-
mer 5 displays best overall characteristics for the detection of
TATP, thus representing a superior material for device construc-
tion. In an open device the limit of detection by 5 is 3s=
31 ppb TATP. For the supramolecular recognition of TATP the
conformational restrictions of the voids in the dendrimers
seems to be crucial by providing more defined incorporation
sites.
Acknowledgements
This study was supported by the ENQUETE consortium financed
by the state of North-Rhine Westphalia. Support by the DFG
through SFB 625 is gratefully acknowledged.
[11] F. Vçgtle, G. Richardt, N. Werner, in Dendrimer Chemistry, Wiley-VCH,
Weinheim, 2009.
[12] a) G. R. Newkome, C. N. Moorefield, F. Vçgtle in Dendritic Molecules: Con-
cepts, Synthesis Perspectives, Wiley-VCH, Weinheim, 1996; b) G. R. New-
kome, C. D. Shreiner, Polymer 2008, 49, 1–173; c) G. R. Newkome, C. D.
Shreiner, Chem. Rev. 2010, 110, 6338–6442; d) A. D. Tomalia, H. Baker,
J. R. Dewald, M. Hall, G. Kallos, S. Martin, G. Roeck, J. Ryder, P. Smith,
Polym. J. 1985, 17, 117–132; e) J. F. Jansen, H. W. Peerlings, E. M. de
Brabander-Van den Berg, E. W. Meijer, Angew. Chem. 1995, 107, 1321–
1324; Angew. Chem. Int. Ed. Engl. 1995, 34, 1206–1209.
Keywords: dendrimers · detection · explosives · host–guest
systems · triacetone triperoxide
[1] R. Meyer, J. Koehler, A. Homburg in Explosives, 5th Edition, Wiley-VCH,
Weinheim, 2002, p. 346.
[13] U. M. Wiesler, A. J. Berresheim, F. Morgenroth, K. Mꢀllen, Macromolecules
2001, 34, 187.
[2] a) E. J. Staples in Electronic Noses & Sensors for Detection of Explosives,
(Eds., J. W. Gardner, J. Yinon), NATO Science Series II. Mathematics, Phys-
ics and Chemistry Vol. 159, Kluwer, Dordrecht, 2004, p. 247; b) J. C. Oxley
in Detection of Liquid Explosives and Flammable Agents in Connection
with Terrorism (Eds., H. Schubert, A. Kuznetsov), Springer, Dordrecht,
2007.
[3] J. Zabicky in The Chemistry of Peroxides, (Eds., Z. Rappoport), Wiley-VCH,
West Sussex, 2006.
[4] J. C. Oxley, J. L. Smith, W. Luo, J. Brady, Propellants Explos. Pyrotech.
2009, 34, 539–543; J. C. Oxley, J. L. Smith, K. Shinde, J. Moran, Propel-
lants Explos. Pyrotech. 2005, 30, 127–130.
[5] a) D. F. Laine, C. W. Roske, I. F. Cheng, Anal. Chim. Acta 2008, 608, 56–
60; b) D. Lu, A. Cagan, R. A. Munoz, T. Tangkuaram, J. Wang, Analyst
2006, 131, 1279–1281; c) R. Schulte-Ladbeck, M. Vogel, U. Karst, Anal.
Bioanal. Chem. 2006, 386, 559–565; d) J. C. Sanchez, W. C. Trogler, J.
Mater. Chem. 2008, 18, 5134–5141; e) S. Malashikhin, N. S. Finney, J. Am.
Chem. Soc. 2008, 130, 12846–12847; f) H. Lin, K. S. Suslick, J. Am. Chem.
Soc. 2010, 132, 15519–15521; g) M. A. Walter, U. Panne, M. G. Weller,
Biosensors 2011, 1, 93–106; h) W. Schade, C. Bauer, R. Orghici, S. Wald-
vogel, S. Bçrner in Detection of Liquid Explosives and Flammable Agents
in Connection with Terrorism (Eds., H. Schubert, A. Kuznetsov), Springer,
[14] R. E. Bauer, C. G. Clark, K Mꢀllen, New J. Chem. 2007, 31, 1275–1282.
[15] a) R. E. Bauer, A. C. Grimsdale, K. Mꢀllen, Top. Curr. Chem. 2005, 245,
253–286; b) M. Baumgarten, T. Qin, K. Mꢀllen in Designing Dendrimers
(Eds., S. Campagna, P. Ceroni, F. Puntoriero), Wiley, Hoboken, 2011.
[16] U.-M. Wiesler, T. Weil, K. Mꢀllen, Top. Curr. Chem. 2001, 212, 1–40.
[17] a) M. Schlupp, T. Weil, A. J. Berresheim U. M. Wiesler, J. Bargon, K.
Mꢀllen, Angew. Chem. 2001, 113, 4124–4129; Angew. Chem. Int. Ed.
2001, 40, 4011–4015; b) E. V. Andreitchenko, R. E. Bauer, C. Kreutz, M.
Baumgarten, J. Bargon, K. Mꢀllen, Macromolecules 2008, 41, 548–558.
[18] J. B. Fenn, Angew. Chem. 2003, 115, 3999–4024; Angew. Chem. Int. Ed.
2003, 42, 3871–3894.
[19] V. M. Mirsky in Artificial Receptors for Chemical Sensors (Eds., V. M.
Mirsky, A. K. Yatsimirsky), Wiley-VCH, Weinheim, 2011.
[20] H.-J. Schneider, A. K. Yatsimirsky in Artificial Receptors for Chemical Sen-
sors (Eds., V. M. Mirsky, A. K. Yatsimirsky), Wiley-VCH, Weinheim, 2011.
Received: December 18, 2011
Published online on January 25, 2012
ChemPlusChem 2012, 77, 102 – 105
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chempluschem.org
105