Selective and sensitive fluoride detection through alkyne cruciform desilylation

Article abstract of DOI:10.1021/ol401120a

Desilylation of silylethynyl-substituted benzobisoxazole cruciforms can be achieved using stoichiometric amounts of fluoride, leading to a significant change in their UV-vis absorption and fluorescence. This response is observable at micromolar concentrations of fluoride, and, in the case of a triisopropylsilyl-substituted cruciform fluorophore, extraordinarily selective for fluoride over other small inorganic anions, including hydroxide, acetate, and phosphate.

Full text of DOI:10.1021/ol401120a

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
Selective and Sensitive Fluoride Detection
through Alkyne Cruciform Desilylation
ꢀ
ꢁ
Minyoung Jo, Jaebum Lim, and Ognjen S. Miljanic*
Department of Chemistry, University of Houston, 112 Fleming Building, Houston,
Texas 77204-5003, United States
miljanic@uh.edu
Received April 22, 2013
ABSTRACT
Desilylation of silylethynyl-substituted benzobisoxazole cruciforms can be achieved using stoichiometric amounts of fluoride, leading to a significant
change in their UVꢀvis absorption and fluorescence. This response is observable at micromolar concentrations of fluoride, and, in the case of a
triisopropylsilyl-substituted cruciform fluorophore, extraordinarily selective for fluoride over other small inorganic anions, including hydroxide, acetate,
and phosphate.
During the past decade, cross-conjugated cruciform
fluorophores1 have been explored as versatile sensors for
Brønsted and Lewis acids,1a,2 metals,3 anions,4 amines,4a,5
phenols,2b and other analytes. A great majority of these
sensing protocols are reversible in nature, based on either
coordinative or dynamically covalent interactions. While
reversibility generally presents a desirable feature in sensor
design, low association constants between the cruciforms
and their analytes can limit the sensitivity of analyte
detection. Conversely, operation of an irreversible and
high-yielding covalent reaction on a cruciform sensor
constitutes a pathway for highly sensitive detection, as an
analyte needs to be added only in close-to-stoichiometric
amounts relative to the highly dilute fluorescent sensor.
In this contribution, we illustrate this principle through
the first use of benzobisoxazole cruciforms1b,2b,d,4b,6 as
irreversible, specific, and sensitive sensors for fluoride.
Fluorine and silicon form a very strong bond (135 kcal
molꢀ1),7 and thus fluoride-induced desilylation reactions
proceed rapidly and irreversibly.8 Sensing of fluoride ions
through the cleavage of OꢀSi bonds (and the exploration
of reactivity of the resultant OꢀH groups) was achieved by
(1) (a) Zucchero, A. J.; McGrier, P. L.; Bunz, U. H. F. Acc. Chem.
Res. 2010, 43, 397–408. (b) Feldman, A. K.; Steigerwald, M. L.; Guo, X.;
Nuckolls, C. Acc. Chem. Res. 2008, 41, 1731–1741.
(2) (a) Schwaebel, T.; Trapp, O.; Bunz, U. H. F. Chem. Sci. 2013, 4,
ꢀ
ꢁ
273–281. (b) Lim, J.; Nam, D.; Miljanic, O. S. Chem. Sci. 2012, 3, 559–
563. (c) Davey, E. A.; Zucchero, A. J.; Trapp, O.; Bunz, U. H. F. J. Am.
Chem. Soc. 2011, 133, 7716–7718. (d) Lim, J.; Albright, T. A.; Martin,
ꢀ
ꢁ
B. R.; Miljanic, O. S. J. Org. Chem. 2011, 76, 10207–10219. (e) Spitler,
E. L.; Haley, M. M. Tetrahedron 2008, 64, 11469–11474. (f) Marsden,
J. A.; Miller, J. J.; Shirtcliff, L. D.; Haley, M. M. J. Am. Chem. Soc. 2005,
127, 2464–2476.
€
(3) (a) Hauck, M.; Schonhaber, J.; Zucchero, A. J.; Hardcastle, K. I.;
€
Muller, T. J.; Bunz, U. H. F. J. Org. Chem. 2007, 72, 6714–6725.
(b) Wilson, J. N.; Bunz, U. H. F. J. Am. Chem. Soc. 2005, 127, 4124–4125.
ꢁ
(6) (a) Lim, J.; Osowska, K.; Armitage, J. A.; Martin, B. R.; Miljanic,
ꢀ
ꢀ
(4) (a) Lirag, R. C.; Le, H. T. M.; Miljanic, O. S. Chem. Commun.
ꢁ
ꢁ
O. S. CrystEngComm 2012, 14, 6152–6162. (b) Osowska, K.; Miljanic,
ꢀ
ꢀ
ꢁ
2013, 49, 4304–4306. (b) Lim, J.; Miljanic, O. S. Chem. Commun. 2012,
48, 10301–10303. (c) Brombosz, S. M.; Zucchero, A. J.; Phillips, R. L.;
Vazquez, D.; Wilson, A.; Bunz, U. H. F. Org. Lett. 2007, 9, 4519–4522.
(5) (a) Kumpf, J.; Bunz, U. H. F. Chem.;Eur. J. 2012, 18, 8921–
8924. (b) Patze, C.; Broedner, K.; Rominger, F.; Trapp, O.; Bunz,
U. H. F. Chem.;Eur. J. 2011, 17, 13720–13725. (c) McGrier, P. L.;
Solntsev, K. M.; Miao, S.; Tolbert, L. M.; Miranda, O. R.; Rotello,
V. M.; Bunz, U. H. F. Chem.;Eur. J. 2008, 14, 4503–4510.
O. S. Chem. Commun. 2010, 46, 4276–4278. (c) Klare, J. E.; Tulevski, G.
S; Nuckolls, C. Langmuir 2004, 20, 10068–10072. (d) Klare, J. E.;
Tulevski, G. S.; Sugo, K.; de Picciotto, A.; White, A. K.; Nuckolls, C.
J. Am. Chem. Soc. 2003, 125, 6030–6031. (e) Tlach, B. C.; Tomlinson,
A. L.; Bhuwalka, A.; Jeffries-EL, M. J. Org. Chem. 2011, 76, 8670–8681.
(7) See: Yoder, C. Wired Chemist: Common Bond Energies (D) and
Bond Lengths (r).http://www.wiredchemist.com/chemistry/data/bond-
energies-lengths, accessed July 8, 2013.
r
10.1021/ol401120a
XXXX American Chemical Society

Figure 1. Frontier molecular orbitals (FMOs) of silylated cruciforms 1ꢀ3. Spatial separation of FMOs is most pronounced in
compound 3.
Kim and Swager,9 and subsequently used by other groups,
including in the context of physiologically relevant10 sen-
sing of fluoride in aqueous solutions. We speculated that
fluoride could also be sensed by cleaving the CꢀSi bond
in silylated alkynes, and that the resultant change in the
electronics of the triple bond would be readily transmitted
through the conjugated circuit.
In order to assess the potential of silylated benzobisox-
azole cruciforms as fluoride sensors, we first computationally
evaluated the frontier molecular orbitals (FMOs) of trimethyl-
silyl (TMS)-substituted compounds 1ꢀ3 (Figure 1) using the
Gaussian 09W11 software package at the B3LYP/3-21G
(8) The strong SiꢀF bond was also utilized in fluoride detection via
the formation of hypervalent silicon complexes. See: (a) Lenormand, H.;
Goddard, J. P.; Fensterbank, L. Org. Lett. 2013, 15, 748–751. (b) Kim,
Y.; Kim, M.; Gabbaı, F. P. Org. Lett. 2010, 12, 600–602. (c) Yamaguchi,
S.; Akiyama, S.; Tamao, K. J. Am. Chem. Soc. 2000, 122, 6793–6794.
(d) Tamao, K.; Hayashi, T.; Ito, Y. J. Organomet. Chem. 1996, 506, 85–
91. (e) Jung, M. E.; Xia, H. Tetrahedron Lett. 1988, 29, 297–300.
(9) Kim, T.-H.; Swager, T. M. Angew. Chem., Int. Ed. 2003, 42, 4803–
4806.
Scheme 1. Synthesis of Fluorophores 1ꢀ3 and 7
(10) See, inter alia: (a) Ke, B.; Chen, W.; Ni, N.; Cheng, Y.; Dai, C.;
Dinh, H.; Wang, B. Chem. Commun. 2013, 49, 2494–2496. (b) Zheng, F.;
Zeng, F.; Yu, C.; Hou, X.; Wu, S. Chem.;Eur. J. 2013, 19, 936–942.
(c) Kim, D.; Singha, S.; Wang, T.; Seo, E.; Lee, J. H.; Lee, S.-J.; Kim,
K. H.; Anh, K. H. Chem. Commun. 2012, 48, 10243–10245. (d) Baker,
M. S.; Phillips, S. T. Org. Biomol. Chem. 2012, 10, 3595–3599. (e) Bao,
Y.; Liu, B.; Wang, H.; Tian, J.; Bai, R. Chem. Commun. 2011, 47, 3957–
3959. (f) Kim, S. Y.; Park, J.; Koh, M.; Park, S. B.; Hong, J.-I. Chem.
Commun. 2009, 4735–4737. (g) Jiang, X.; Vieweger, M. C.; Bollinger,
J. C.; Dragnea, B.; Lee, D. Org. Lett. 2007, 9, 3579–3582.
(11) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.;
Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.;
Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Keith, T.; Kobayashi, R.; Normand, J.;
Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.;
Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.;
Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.;
Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.;
Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador,
P.; Dannenberg, J. J.; Dapprich, S.;Daniels, A. D.; Farkas, O.; Foresman,
J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, Revision B.01;
Gaussian, Inc.: Wallingford, CT, 2010.
level of theory. Graphical representations of highest occu-
pied (HOMOs) and lowest unoccupied molecular orbitals
(LUMOs) of 1ꢀ3 are shown in Figure 1. In phenyl-sub-
stituted cruciform 1, FMOs are largely overlapping, suggest-
ing that desilylation should have a similar effect on both
orbitals and thus lead to just a small change in the
B
Org. Lett., Vol. XX, No. XX, XXXX

Figure 2. Normalized fluorescence intensity plots for the titrations of 10ꢀ7 M THF solutions of 1 (A), 2 (B), 3 (C), and 7 (D) with a
10ꢀ4ꢀ10ꢀ2 M solution of TBAþFꢀ. Excitation wavelengths were 303 nm for 1, 360 nm for 2, and 306 nm for 3 and 7.
HOMOꢀLUMO gap. Similar situation is observed in
cruciform 2 as well; in its case, the two FMOs were mostly
positioned along the vertical benzobisoxazole axis but
were still largely overlapping. However, in the pyridyl-based
compound 3, spatial separation of FMOs is noticeable:
HOMO is positioned along the electron-richer horizontal
axis, while the LUMO chiefly resides along the pyridine-
bearing vertical axis. This orbital distribution suggested that
the desilylation of 3 should impact HOMO and LUMO
unevenly and thus constitute the basis of an optical response
of cruciform 3 to fluoride addition.
Encouraged by this computational insight, we syn-
thesized TMS-substituted cruciforms 1ꢀ3, as well as com-
pound 7, the triisopropylsilyl (TIPS)-substituted analogue
of 3 (Scheme 1). All four silylated cruciformswere expected
to desilylate under the influence of fluoride, but cruciform
7 was the only one that was expected to be desilylated
exclusively with fluoride; compounds 1ꢀ3 could also be
deprotected by hydroxide or other basic species.12 Starting
with brominated benzobisoxazoles 4ꢀ6 (Scheme 1),2d So-
nogashira coupling with TMS-acetylene produced corre-
sponding silylated compounds 1ꢀ3 in moderate yields.
Triisopropylsilyl-substituted cruciform 7 was prepared in
54% yield by reacting 6 with TIPS-acetylene. Compounds
1ꢀ3 and 7 are yellow to dark orange powders, with strong
fluorescence in solution.
Optical response of silylated benzobisoxazoles to anions
was evaluated by UVꢀvis absorption and fluorescence
spectroscopy. Dilute solutions (10ꢀ5 M for absorption
and 10ꢀ7 M for fluorescence measurements) of 1ꢀ3
and 7 in THF were titrated with concentrated solutions
(10ꢀ2 to 10ꢀ4 M) of tetrabutylammonium fluoride, chlor-
ide, bromide, iodide, hydroxide, acetate, phosphate, and
nitrate. In Figure 2, fluorescence titration spectra for addi-
tion of fluoride are shown. The absorption titration spectra
for fluoride and all (absorption and fluorescence) spectra
for the titration of 1ꢀ3 and 7 with other anions are given in
the Supporting Information.
This behavior is consistent with the broadening of the
HOMOꢀLUMO gap, which in turn is explained by the
dominant stabilization of the HOMO upon desilylation.13
The emission of compound 1 shifted by approximately
ꢀ5 nm upon exposure to ∼100 equiv of fluoride, and those
changes are very similar to those observed upon the addi-
tion of hydroxide anion (see Supporting Information).
Small changes in absorption and fluorescence can be ratio-
nalized through the relatively similar spatial distribution of
FMOs of 1, while the nonspecific response comes from the
easy cleavage of the TMS group, which can be deprotected
using either hydroxide or fluoride anions. Similarly, com-
pound 2 showed noticeable but relatively small shifts in
absorption and fluorescence spectra, and these shifts were
induced by fluoride and hydroxide alike.
Cruciforms 3 and 7 exhibited the most diagnostic shifts
in absorption (see Supporting Information) and especially
fluorescence upon addition of fluoride, with emission
maxima shifting by ꢀ15 nm in the case of 7. After the
addition of an excess of fluoride (>50 equiv), the fluor-
escence profiles of 3 and 7 appeared largely identical,
which is logical, as both compounds desilylate to the
same terminal diyne.14
The crucial difference between 3 and 7 resided in the
lability of their silyl groups. Thus, the trimethylsilyl group
of 3 can be cleaved with fluoride, but also with hydroxide,
acetate, and phosphate, as evidenced by their similar
fluorescence titration profiles shown in the Supporting
Information.15 On the other hand, the more stable TIPS
protecting group is cleaved only with fluoride (on the time
scale of our experiments), and thus only fluoride induces
the noteworthy shifts in the fluorescence spectrum shown
(13) (a) Ensslin, W.; Bock, H.; Becker, G. J. Am. Chem. Soc. 1974, 96,
2757–2762. See also: (b) Albright, T. A.; Burdett, J. K.; Whangbo, M.-H.
Orbital Interactions in Chemistry; Wiley: Hoboken, NJ, 2013.
(14) The time required to develoꢀp full emission response to Fꢀ is
dependent on the concentration of F , but even for low concentrations
of this analyte (50 μM), the desilylation reaction of cruciform 7 appears
>90% complete in <1 min. See Supporting Information for plots of
emission spectra of a mixture of cruciform 7 and Fꢀ as a function of time.
(15) Desilylation of 3 still proceeds most readily with fluoride, as
evidenced by the number of anion equivalents needed to induce sig-
nificant shifts in fluorescence: 20ꢀ50 equiv with Fꢀ, and approximately
10ꢀ20 times more with HOꢀ, AcOꢀ, and PO43ꢀ. See Supporting
Information for details.
Following the addition of fluoride, the fluorescence emis-
sion maxima of all four fluorophores underwent a blue shift.
(12) Wuts, P. G. M.; Greene, T. W. Greene’s Protective Groups in
Organic Synthesis; Wiley: Hoboken, NJ, 2007; Chapter 8.
Org. Lett., Vol. XX, No. XX, XXXX
C

in Figure 2D. In order to quantify the selectivity of
fluoride detection with 7, we plotted the relative changes
in the intensity of its fluorescence emission at 381 nm,
following the addition of 100 equiv of different anions
(Figure 3). Fluoride addition enhances the emission of
7 at 381 nm by more than 1.5 times; in contrast, chloride,
bromide, hydroxide, acetate, nitrate, and phosphate all
changed the emission of 7 by less than 4%. Iodide anion
somewhat quenched the fluorescence of 7 (ꢀ17.1%), but
without significant shifts in the position of emission max-
ima. This behavior did not interfere with fluoride detection,
as it required ∼1000 equiv of iodide to decrease fluorescence
intensity by approximately 70%.16
low as 50 μM and are highly specific for this anion. They
operate through the desilylation of silylated alkynes, and are
thus rare among other desilylation-based fluoride sensors,
which are primarily exploring the cleavage of the SiꢀO bond.
Figure 4. Changes in emission colors of dilute solutions of
cruciforms 3 (top) and 7 (bottom) upon addition of 10 equiv
of the corresponding anions. Excitation wavelength was 365 nm.
More generally, we believe that this work demonstrates
the versatility of the “cruciform approach” to sensing.
As benzobisoxazole and other conjugated cruciforms can
be very easily induced to spatially isolate their FMOs,
creation of a viable sensor is highly predictable. Synthetic
attention can instead be focused on the relatively small
modifications that can change sensor specificity and selec-
tivity dramatically, allowing expedient adaptation of a
cruciform sensor to a wide variety of analytes. This strategy
is perhaps best illustrated in the current work, where
silylated alkyne cruciforms, which are in fact common
synthetic precursors to more complex alkynes, can be used
as selective sensors of their own desilylation. Our current
work is focusing on expanding the domain of chemical
sensing through cruciforms.
Figure 3. Relative changes in the intensity of emission of 7 at
381 nm upon addition of 100 equiv of corresponding analytes.
Excitation wavelength was 306 nm, and the percentage values
were calculated as %increase = (I100 ꢀ I0)/I0, where I100 and I0 are
intensities of fluorescence with 100 and 0 equiv of analyte,
respectively.
Acknowledgment. This work was financially supported
by the University of Houston (UH), the Welch Founda-
tion (Award E-1768), the National Science Foundation
(Award CHE-1151292), the donors of the American Che-
mical Society Petroleum Research Fund (Award 50390-
DNI10), and the Texas Center for Superconductivity at
The fluorescence emission changes that accompany
the addition of fluoride and other anions are visible by
naked eye for sensors 3 and 7 (Figure 4). The switch of
the emission color from cyan to purpleis clearly observable
for fluoride and hydroxide in the case of 3, and only for
fluoride when cruciform 7 is used.
In conclusion, this study has shown that silylated ben-
zobisoxazole cruciforms can serve as selective and sensitive
fluorescent detectors for the fluoride anion. Their emission
changes in response to fluoride addition; these changes are
spectroscopically observable at fluoride concentrations as
ꢀ
UH. O.S.M. is a Cottrell Scholar of the Research Corpora-
tion for Science Advancement. Prof. Thomas A. Albright
(UH) is acknowledged for insightful advice.
Supporting Information Available. Synthetic procedures,
full spectroscopic characterization, anion titration, and
computational data for 1ꢀ3 and 7. This material is available
free of charge via the Internet at http://pubs.acs.org.
(16) Similar behavior was observed for all examined cruciform
fluorophores upon exposure to iodide and can be rationalized through
heavy atom-induced collisional quenching.
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX