Cruciforms as functional fluorophores: Response to protons and selected metal ions

Article abstract of DOI:10.1021/ja061112e

The photophysics of dialkylamino- and/or pyridine-containing functional chromophores, 1,4-distyryl-2,5-bis(ethynylaryl)benzenes (cruciforme) was investigated; their fluorescence quantum yields and emissive lifetimes were determined. Depending upon their s

Full text of DOI:10.1021/ja061112e

Published on Web 08/19/2006
Cruciforms as Functional Fluorophores: Response to
Protons and Selected Metal Ions
Anthony J. Zucchero, James N. Wilson, and Uwe H. F. Bunz*
Contribution from the School of Chemistry and Biochemistry, Georgia Institute of Technology,
770 State Street, Atlanta, Georgia 30332
Received February 15, 2006; E-mail: uwe.bunz@chemistry.gatech.edu
Abstract: The photophysics of dialkylamino- and/or pyridine-containing functional chromophores, 1,4-distyryl-
2,5-bis(ethynylaryl)benzenes (cruciforms) was investigated; their fluorescence quantum yields and emissive
lifetimes were determined. Depending upon their substituents, the frontier molecular orbitals (FMOs) of
these cruciforms are either congruent, i.e., HOMO and LUMO occupy the same real space, or disjoint, i.e.,
the HOMO is located on one branch of the cruciform while the LUMO is located on the second one. Donor-
acceptor substitution leads to a disjoint FMO pattern, while the parent 1,4-distyryl-2,5-bis(phenylethynyl)-
benzene shows congruent FMOs. The photophysics of the cruciforms was investigated upon addition of
either an excess of trifluoroacetic acid or an excess of selected metal (Mg2+, Ca2+, Mn2+, Zn2+
)
trifluoromethanesulfonate salts. Addition of either metal ions or protons led to analogous but not identical
changes in the spectroscopic properties of the investigated cruciforms. The collected data suggest that
the metals bind preferentially at the aniline nitrogen and not at the electron-rich arene. The spatially separated
FMOs permit the independent manipulation of the HOMO and the LUMO of such cruciforms. If the branches
contain metal-complexing moieties, metal binding leads to either a hypsochromic or a bathochromic shift
in emission via interaction of the metal cations with either the HOMO or the LUMO.
Introduction
are an integral part of a dye, functional chromophores result.
Pyridines, bipyridines, and terpyridines as well as aromatic
Functional chromophores contain embedded recognition
elements as integral parts of their π-system. Such chromophores
are innately qualified to serve as active elements in sensors or
sensor arrays and in advanced materials applications. Functional
chromophores are different from functionalized chromophores,
where recognition elements are merely appendages affixed to a
fluorescent core. Functional chromophores undergo analyte-
induced shifts in absorption and emission; the binding modulates
electronic properties due to the interweaving of π-system and
recognition element. Prominent examples are Zn-ESIPT sensors
and related phores,1 bipyridine-containing conjugated polymers,2
tetraethynylethenes (TEEs) and diethynylethenes (DEEs),3 and
oligophenyleneethynylene-embedded 2,6-bis(1,9-methylbenz-
imidazolyl)pyridines.4 Whenever basic nitrogens or phenolates
dialkylamines, phenolates, and phosphines, inter alia, are good
candidates to render a chromophore “functional”. Embedded
bases can be either strongly electron accepting, such as
pyridines, or strongly electron donating, such as dialkylanilines,
permitting the rational design of functional chromophores with
modular photophysical responses. If donor and acceptor units
are spatially separated, it will be possible to address HOMO
and LUMO independently by Lewis acids or metal cations. A
chromophore system will result that naturally displays a two
stage response or differential binding to an electron deficient
analyte. Cruciforms 1-7 are functional chromophores, with
incorporated basic nitrogens; they should show potential in the
fabrication of functional sensor arrays for metal ions.5
To fully understand and appreciate the subtleties and issues
associated with the sensory response of these cruciform chro-
mophores, we report the steady-state photophysics and the
emissive lifetimes of 1-7 and their response to the addition of
protons and the triflates of Mg2+, Ca2+, Mn2+, and Zn2+. We
selected these metal cations as they are all biologically active,
+2 charged ions of similar size. Large shifts in absorption,
(1) (a) Henary, M. M.; Wu, Y. G.; Fahrni, C. J. Chem.sEur. J. 2004, 10,
3015-3025. (b) Pond, S. J. K.; Tsutsumi, O.; Rumi, M.; Kwon, O.; Zojer,
E.; Bredas, J. L.; Marder, S. R.; Perry, J. W. J. Am. Chem. Soc. 2004, 126,
9291-9306.
(2) Wang, B.; Wasielewski, M. R. J. Am. Chem. Soc. 1997, 119, 12-21.
Bangcuyo, C. G.; Rampey-Vaughn, M. E.; Quan, L. T.; Angel, S. M.; Smith,
M. D.; Bunz, U. H. F. Macromolecules 2002, 35, 1563-1568. (c) Pautzsch,
T.; Klemm, E. Macromolecules 2002, 35, 1569-1575.
(3) Gobbi, L.; Seiler, P.; Diederich, F. Angew. Chem., Int. Ed. 1999, 38, 674-
678. (b) Diederich, F. Chem. Commun. 2001, 219-227. (c) Tykwinski, R.
R.; Gubler, U.; Martin, R. E.; Diederich, F.; Bosshard, C.; Gunter, P. J.
Phys. Chem. B 1998, 102, 4451-4465. (d) Tykwinski, R. R.; Schreiber,
M.; Carlon, R. P.; Diederich, F.; Gramlich, V. HelV. Chim Acta 1996, 79,
2249-2281. (e) Hilger, A.; Gisselbrecht, J. P.; Tykwinski, R.; Boudon,
C.; Schreiber, M.; Martin, R.; Lu¨thi, H. P.; Gross, M.; Diederich, F. J.
Am. Chem. Soc. 1997, 119, 2069-2078. (f) Mitzel, F.; Boudon, C.;
Gisselbrecht, J. P.; Seiler, P.; Gross, M.; Diederich, F. HelV. Chim. Acta
2004, 87, 1130-1157. (g) Gisselbrecht, J. P.; Moonen, N. N. P.; Boudon,
C.; Nielsen, M. B.; Diederich, F.; Gross, M. Eur. J. Org. Chem. 2004,
2959-2972.
(4) (a) Knapton, D.; Rowan, S. J.; Weder, C. Macromolecules 2006, 39, 651-
657. (b) Iyer, P. K.; Beck, J. B.; Weder, C.; Rowan, S. J. Chem. Commun.
2005, 319-321.
(5) (a) Lavigne, J. J.; Anslyn, E. V. Angew. Chem., Int. Ed. 1999, 38, 3666-
3669. (b) Wright, A. T.; Anslyn, E. V.; McDevitt, J. T. J. Am. Chem. Soc.
2005, 127, 17405-17411. (c) Rakow, N. A.; Suslick, K. S. Nature 2000,
406, 710-713. Suslick, K. S.; Rakow, N. A.; Sen, A. Tetrahedron 2004,
60, 11133-11138. (d) Albert, K. J.; Lewis, N. S.; Schauer, C. L.; Sotzing,
G. A.; Stitzel, S. E.; Vaid, T. P.; Walt, D. R. Chem. ReV. 2000, 100, 2595-
2626.
9
11872
J. AM. CHEM. SOC. 2006, 128, 11872-11881
10.1021/ja061112e CCC: $33.50 © 2006 American Chemical Society

Cruciforms as Functional Fluorophores
A R T I C L E S
Scheme 1. Synthesis of Cruciforms by a Combination of Horner
and Sonogashira Methods
fluorescence, quantum yields, and emissive lifetimes are ob-
served upon the addition of protons or metal cations. The
changes of the cruciforms’ photophysics upon exposure to
protons and to metal ions are comparable and suggest that
binding of either species occurs at the same molecular loci. The
observed photophysics can be explained by the position and
the spatial arrangement of the frontier molecular orbitals (FMO)
and their interaction with cations.
right with applications that range from fluorescence lifetime
imaging to their use as cofactors in fluorescent antibodies.8-11
If metal-complexing substituents are attached to the aromatic
core, bis(aminostyryl)benzenes function as metal-sensing spe-
cies, particularly if 15-crown[5] substituents are appended via
an aniline nitrogen.9 Significant blue-shift in the absorption is
observed upon binding of various metal cations, but the emissive
lifetimes remain almost unchanged.
Similarly, bis(arylethynyl)benzenes were investigated as
ligands for metal-organic networks and as model compounds
for the respective conjugated polymers.12 Their fluorescence
lifetimes (table in the Supporting Information) (τ) range from
τ ) 0.32 ns to τ ) 2.5 ns. Most of the emissive lifetimes are
around τ ) 1 ns. The herein investigated cruciforms 1-8 are
1,4-distyrylbenzenes, yet their photophysical behavior and their
spectroscopic responses to metal cations are significantly
expanded.
Results and Discussion
Synthesis. Cruciform fluorophores13-16 are π-systems in
which two alkene and two alkyne arms are placed in a 1,4-
2,5-relationship on a central benzene ring. The synthesis of
cruciforms 1-3, 5, and 7 has been described in ref 13, while
the synthesis of cruciforms 4, 6, and 8a is presented in the
Supporting Information. All of the cruciforms are made from a
common precursor (Scheme 1); 2,5-bis(bromomethyl)-1,4-
diiodobenzene is reacted with triethyl phosphite in an Arbuzov
reaction. A Horner17 olefination of the resulting bis(phosphonate)
(7) (a) Yang, J. S.; Chiou, S. Y.; Liau, K. L. J. Am. Chem. Soc. 2004, 126,
2518-2527. (b) Yang, J. S.; Liau, K. L.; Wang, C. M.; Hwang, C. Y. J.
Am. Chem. Soc. 2004, 126, 12325-12335. (c) Yang, J. S.; Liau, K. L.;
Tu, C. W.; Hwang, C. Y. J. Phys. Chem. A 2005, 109, 6450-6456. (d)
Yang, J. S.; Lin, Y. D.; Chang, Y. H.; Wang, S. S. J. Org. Chem. 2005,
70, 6066-6073.
(8) Holzer, W.; Penzkofer, A.; Gong, S. H.; Bradley, D. D. C.; Long, X.; Bleyer,
A. Chem. Phys. 1997, 224, 315-326.
(9) Barlow, S.; Risko, C.; Coropceanu, V.; Tucker, N. M.; Jones, S. C.; Levi,
Z.; Khrustalev, V. N.; Antipin, M. Y.; Kinnibrugh, T. L.; Timofeeva, T.;
Marder, S. R.; Bredas J. L. Chem. Commun. 2005, 764-766.
(10) Matsushita, M.; Meijler, M. M.; Wirsching, P.; Lerner, R. A.; Janda, K.
D. Org. Lett. 2005, 7, 4943-4946.
(11) (a) Bastiaens, P. I. H.; Squire, A. Trends Cell Biol. 1999, 9, 48-52. (b)
Lakowicz, J. R.; Szmacinski, H.; Nowaczyk, K.; Berndt, K. W.; Johnson,
M. Anal. Biochem. 1992, 202, 316-330. (c) Pepperkok, R.; Squire, A.;
Geley, S.; Bastiaens P. I. H. Curr. Biol. 1999, 9, 269-272. (d) Mayr, T.;
Igel, C.; Liebsch, G.; Klimant, I.; Wolfbeis, O. S. Anal. Chem. 2003, 75,
4389-4396.
(12) (a) Evans, O. R.; Lin, W. B. Acc. Chem. Res. 2002, 35, 511-522. (b) Lin,
W. B.; Evans, O. R.; Xiong, R. G.; Wang, Z. Y. J. Am. Chem. Soc. 1998,
120, 13272-13273.
1,4-Distyrylbenzene was first synthesized in 1917 by Kauf-
mann through the reaction of benzylmagnesium bromide with
terephthalic aldehyde followed by acid-catalyzed dehydration.
Wittig, Horner, and Heck routes were developed subsequently.6
The distyrylbenzene scaffold allows one to investigate changes
in photophysical properties upon the attachment of donor and
acceptor species.7 Distyrylbenzenes find use as blue-emitting
laser dyes and as model compounds for poly(para-phenyle-
nevinylene)s (PPV) but are attractive chromophores in their own
(13) (a) Wilson, J. N.; Josowicz, M.; Wang, Y. Q.; Bunz, U. H. F. Chem.
Commun. 2003, 2962-2963. (b) Wilson, J. N.; Bunz, U. H. F. J. Am. Chem.
Soc. 2005, 127, 4124-412. (c) Gerhardt, W. W.; Zucchero, A. J.; Wilson,
J. N.; South, C. R.; Bunz, U. H. F.; Weck, M. Chem. Commun. 2006, 2141-
2143.
(6) (a) Kaufmann, H. Chem. Ber. 1917, 50, 515-529. (b) Meier, H. Angew.
Chem., Int. Ed. Engl. 1992, 31, 1399-1420. (c) Meier, H.; Lehmann, M.
Angew. Chem., Int. Ed. 1998, 37, 643-645. (d) Meier, H.; Gerold, J.;
Kolshorn, H.; Baumann, W.; Bletz, M. Angew. Chem., Int. Ed. 2002, 41,
292-295. (e) Deb, S. K.; Maddux, T. M.; Yu, L. P. J. Am. Chem. Soc.
1997, 119, 9079-9080. (f) Maddux, T. M.; Li, W. J.; Yu, L. P. J. Am.
Chem. Soc. 1997, 119, 844-845. (g) Wang, S. J.; Oldham, W. J.; Hudack,
R. A.; Bazan, G. C. J. Am. Chem. Soc. 2000, 122, 5695-5709. (h) Bazan,
G. C.; Oldham, W. J.; Lachicotte, R. J.; Tretiak, S.; Chernyak, V.; Mukamel,
S. J. Am. Chem. Soc. 1998, 120, 9188-9200. (i) Oldham, W. J.; Lachicotte,
R. J.; Bazan, G. C. J. Am. Chem. Soc. 1998, 120, 2987-2988. (j) Scherf,
U.; Mu¨llen, K. Synthesis 1992, 23-38. (k) Schenk, R.; Gregorius, H.;
Meerholz, K.; Heinze, J.; Mu¨llen, K. J. Am. Chem. Soc. 1991, 113, 2634-
2647. (l) Scherf, U. Top. Curr. Chem. 1999, 201, 163-222.
(14) (a) Marsden, J. A.; Miller, J. J. Shirtcliff, L. D.; Haley, M. M. J. Am. Chem.
Soc. 2005, 127, 2464-2476. (b) Marsden, J. A.; Haley, M. M. J. Org.
Chem. 2005, 70, 10213-10226. (c) Marsden, J. A.; Miller, J. J.; Haley,
M. M. Angew. Chem., Int. Ed. 2004, 43, 1694-1697.
(15) (a) Hu, K.; Zhu, P. W.; Yu, Y.; Facchetti, A.; Marks, T. J. J. Am. Chem.
Soc. 2004, 126, 15974-15975. (b) Niazimbetova, Z. I.; Christian, H. Y.;
Bhandari, Y. J.; Beyer, F. L.; Galvin, M. E. J. Phys. Chem. B 2004, 108,
8673-8681.
(16) Klare, J. E.; Tulevski, G. S.; Sugo, K.; de Picciotto, A.; White, K. A.;
Nuckolls, C. J. Am. Chem. Soc. 2003, 125, 6030-6031.
9
J. AM. CHEM. SOC. VOL. 128, NO. 36, 2006 11873

A R T I C L E S
Zucchero et al.
in THF as solvent with NaH as base affords 1,4-diiodo-2,5-
distyrylbenzenes. The two aromatic iodides are substituted
by alkyne modules in a Sonogashira-Hagihara coupling,18
performed in piperidine with (Ph3P)2PdCl2/CuI as catalyst sys-
tem to give the cruciforms 1-7 and 8a in good to excellent
yields.
with λmax emission in these speciessthe lower energy the
emission, the lower of the fluorescence quantum yield. The
decrease of emission quantum yield upon red shift of the
emission is not unexpected, as nonradiative pathways (vibronic
coupling) are more accessible in compounds with a smaller band
gap.19 However, in 1-7, there is no clean numerical correlation
between band gap and emission quantum yields.
Spectroscopic Properties of the Cruciforms 1-7 in Dichlo-
romethane. We have recently shown that 1, 3, and 7 are useful
as fluorescent ligands for metal ions such as zinc, magnesium,
and silver.13b Dual emission results if 7 is used as a ligand,
because metal complexation can successively occur on the
dibutylaniline and then on the pyridine units, or vice versa. The
pyridine rings are the less basic ligands and only bind through
the free nitrogen lone pairs. For the anilines, it is not readily
apparent whether metal cations bind directly to the nitrogen or
to the π-face of the electron-rich arene. We investigated the
optical properties of cruciforms 1-7 and examined the change
of their photophysical properties upon addition of either acid
or an excess of zinc ions. Figures 1 and 3 and Tables 1-4 show
selected spectroscopic data (absorption, emission, Stokes shift,
radiative lifetime, quantum yields) for cruciforms 1-7 in
The normalized emission spectra of 1-7 and 8a (in hexanes)
are displayed in Figure 3. Their emissions appear blue-shifted
in hexanes relative to those recorded in dichloromethane; their
Stokes shifts are also smaller than the corresponding Stokes
shifts in dichloromethane. The cruciforms 1-3 and 8a show a
vibronic progression ranging from 1097 to 1302 cm-1, associ-
ated with the CdC-stretching mode of the distyrylbenzene arms.
In these cruciforms, this numerical value is somewhat smaller
than in simple aminostilbenes (1300 ( 80 cm-1), probably due
to increased conjugation.6a In 1-3, the intensity of the main
band is larger than the intensity of the vibronic progression while
in 8a both are of the same intensity. A classic interpretation
would assert that the 0,0 transition is more prevalent than the
0,1 transition; concomitantly, in the excited state, the distortion
around the stilbene double bonds would be only modest, because
distortion correlates with an increased intensity of the vibronic
progression. It is not clear if the classic picture holds for 1-3
and 8a; their Stokes shifts are quite large and the main band
might already be the vibronic 0,1 or 0,2 transition. Cruciforms
1-3 and 8 feature spatially delocalized HOMOs and LUMOs,
and their emission spectra in hexanes are similar to those
observed for stilbenes with analogous substituent patterns. Upon
examination of the emission spectra of 4-7 in hexanes, one
notices a bathochromic shift and a loss of vibronic fine structure.
The more red-shifted the emission, the less vibronic fine
structure is visible. For 4 and 7 there is still a shoulder, but in
5 and 6 this shoulder has largely disappeared. The increasing
donor-acceptor character in 4-7 and the concomitantly in-
creased internal charge-transfer character of the excited state
makes it more distorted and explains the observed features (or
the lack thereof). We hypothesize that the local separation of
the FMOs in 4-7 is characterized by a loss of fine structure in
their fluorescence spectra taken in nonpolar solventssa con-
venient heuristic instrument to characterize novel cruciforms
(vide infra).
If donor substituents are attached to the distyrylbenzene arms
and acceptor substituents to the arylethynyl arms, the coefficient
distribution of the FMOs changes. The donor and acceptor
substituents localize the FMOs on the respectiVe arylethynyl
and styryl branches. The chromophores in these donor-
acceptor-substituted cruciforms show a disjoint FMO structure.
On the other hand, cruciforms devoid of the donor-acceptor
pattern display FMOs that are spatially superimposable upon
one another; we propose to call cruciforms with such an FMO
arrangement congruent. The transition between congruent and
disjoint is gradual but coincides with the occurrence of the
intense charge transfer band that dominates their absorption
spectra and the loss of fine structure in the emission spectra
taken in nonpolar solvents. Organic chromophores with a disjoint
FMO structure are quite rare but have recently been found in
solution and upon addition of either metal (Mg2+, Ca2+, Mn2+
,
Zn2+) triflates or upon protonation by trifluoroacetic acid. Figure
2 shows a photograph of solutions of 1-7 under a blacklight.
The top row depicts the chromophores alone, while in the middle
and bottom rows the same chromophores are presented after
addition of trifluoroacetic acid and zinc ions, respectively. The
color changes of the brightly fluorescent cruciforms are
spectacular; we expect these functional chromophores to show
a significant potential for array applications, when equipped with
additional auxiliary appendages.5
The absorption and emission spectra and the radiative
lifetimes of 1-3, 5, and 7 are discussed in detail. The optical
properties of 4 and 6 are included in Tables 1-4 but are similar
to those of 5 and do not warrant separate discussion. Absorption
and emission spectra of 4 and 6 are included in the Supporting
Information. Figure 1 shows the absorption and emission spectra
of 1-3, 5, and 7 in dichloromethane and hexanes.
The absorption spectra of the either donor- or acceptor-
substituted cruciforms 1 and 2 are similar in shape, possessing
one main transition around 330 nm and a second one at higher
wavelengths that appears as a shoulder. The donor and the
donor-acceptor cruciforms 3-7 (discussed here are 3, 5, and
7) show a prominent, broad absorption feature around 450 nm
and a second one with an additional shoulder at lower
wavelengths. The low-energy band is intense in 4-7 and is due
to a charge-transfer interaction of the donor and the acceptor
groups. The emission spectra of cruciforms 1-7 in dichlo-
romethane are broad and featureless with emission maxima
ranging from 444 to 590 nm. The strongly donor-acceptor-
substituted cruciforms 4-7 show red-shifted emission (561-
590 nm) when compared to that of 1-3 (444-519 nm). The
quantum yield of 1-3 in halogenated solvents is in the range
of 0.3-0.7, while the donor-acceptor-substituted cruciforms
4-7 display quantum yields that are lower and range from 0.05
to 0.11. The quantum yield is qualitatively inversely correlated
(17) Horner, L.; Hoffmann, H.; Wippel, H. G.; Klahre, G. Chem. Ber. 1959,
92, 2499-2505.
(18) (a) Negishi, E.; Anastasia, L. Chem. ReV. 2003, 103, 1979-2017. (b)
Sonogashira, K. J. Organomet. Chem. 2002, 653, 46-49. (c) Bunz, U. H.
F. Chem. ReV. 2000, 100, 1605-1645.
(19) (a) Tolbert, L. M.; Nesselroth, S. M.; Netzel, T. L.; Raya, N.; Stapleton,
M. J. Phys. Chem. 1992, 96, 4492-4496. (b) Englman, R.; Jortner, J. Mol.
Phys. 1970, 18, 145-154. (c) Caspar, J. V.; Meyer, T. J. J. Phys. Chem.
1983, 87, 952-957.
9
11874 J. AM. CHEM. SOC. VOL. 128, NO. 36, 2006

Cruciforms as Functional Fluorophores
A R T I C L E S
Figure 1. Normalized absorption and emission spectra of 1-3, 5, and 7. Left: Absorption and emission spectra of 1-3, 5, and 7 in hexanes (blue) and
dichloromethane (green). The emissions of 1-3 show vibronic spacing in hexanes. Right: Emission spectra of 1-3, 5, and 7 in dichloromethane (purple)
and upon exposure to excess TFA (black), Mg2+ (green), Zn2+ (blue), Mn2+ (yellow), and Ca2+ (orange). The features in the emission spectra of 1 and 7
at 400-450 nm are due to artifacts from TFA. y-axis: arbitrary units of intensity.
diarylpyrazolines and were exploited by Fahrni et al. as sensory
frameworks.20 Generally, a disjoint FMO structure, i.e., spatially
separated HOMO and LUMO, permits the independent tuning
of the oxidation and reduction potential (or HOMO and LUMO)
of such fluorophores. If one examines pars pro toto the disjoint
FMOs of 7, one notes that there is orbital overlap in the central
benzene ring, but the overall overlap integral is not large; the
HOMO-LUMO transition is symmetry-allowed and not de-
9
J. AM. CHEM. SOC. VOL. 128, NO. 36, 2006 11875

A R T I C L E S
Zucchero et al.
Table 1. Absorption and Emission Data for Cruciforms 1-7 and 8a in Dichloromethane and Hexanes
λ
max abs in
λ
max emission in
λ
max abs in
λ
max emission in
Stokes shift in
Stokes shift in
vibronic progressn
1
1
1
cruciform
CH2Cl2 (nm)
CH2Cl2 (nm)
hexanes (nm)
hexanes (nm)
CH2Cl2 (cm-
)
hexanes (cm-
)
(solvent) (cm-
)
1
330
370
456
324
365
427
448
5097
8373
3978
7445
1097 (hexanes)
395 sh
330
361 sh
336
436
390 sh
325
2
3
444
519
422
445
469
498
5178
7781
4632
7073
2375
8799
1225 (hexanes)
1242 (hexanes)
353
332
422
3668
10494
447 sh
334
4
5
337
561
567
509
4748
4277
9089
352
348
10584
443
418
479 sh
296
470 sh
296
516
5458
11099
4659
9606
331
333
348
345
433
416
472 sh
340
461 sh
321
6
590
567
514
496
6470
10752
12463
4641
8714
9615
361
344
427
478 sh
355
415
460
7
295
288
5247
11265
12393
3368
9164
10142
333
330
346
341
437
425
472 sh
331
460 sh
327
8a
419
441
409
432
2872
6345
2650
6131
1191 (CH2Cl2 )
1302 (hexanes)
374
369
396 sh
390 sh
Table 2. Emissive Data for Cruciforms 1-7 upon the Addition of Protons and Selected Metal Cations
λ
max emission
λ
max emission with
λ
max emission with
λ
max emission with
λmax emission with
excess of Ca2+ (nm)
cruciform
CH2Cl2 (nm)
excess of H+ (nm)
λ
max emission with excess of Zn2+ (nm)
excess of Mg2+ (nm)
excess of Mn2+ (nm)
1
2
3
4
456
444
519
561
555
532
425
420
438
430
580
543
429
421
440
429
581
542
429
418
440
413
431
422
432
530
579
542
429
418
422
414
430
421
435
418
442
526
576
536
429
419
440
417
431
422
432
414
433
560
5
6
7
567
590
567
430
509
429
518
Table 3. Quantum Yields of 1-7 upon the Addition of Protons and Selected Metal Cations
quantum yield
CHCl3
quantum yield with
excess of H+
quantum yield with
quantum yield with
quantum yield with
quantum yield with
+
+
+
+
cruciform
excess of Zn2
excess of Mg2
excess of Mn2
excess of Ca2
1
2
3
4
5
6
7
0.70
0.67
0.31
0.15
0.09
0.05
0.11
NA
0.31
0.18
0.62
0.80
0.84
0.58
0.11
0.49
0.28
0.79
0.85
0.94
0.86
0.23
0.46
0.22
0.69
0.86
0.89
0.79
NA
NAa
NAa
0.76
NAa
NAa
NAa
NAa
0.06
0.44
0.65
0.68
0.38
0.06
a NA: not available due to the presence of two emitting species in solution.
generate in the one-electron approximation. The disjoint FMOs
explain the unusually long emissive lifetime as a consequence
of this weakly Franck-Condon allowed transition. The large
Stokes shifts in the disjoint cruciforms likewise support the
assumption of a poor overlap of the ground with the excited
state and corroborate the results obtained from the lifetime
measurements.
Spectroscopic Data for Cruciforms 1-7 upon Addition
of Trifluoroacetic Acid. If the FMO-localizing branches of the
cruciforms contain free electron pairs such as in dialkylamino
or pyridine units, the optical properties of such chromophores
or luminophores should change upon metal ion complexation
or protonation, directly shifting the HOMO or the LUMO
positions. That is obserVed in cruciforms (Figures 1 and 2). To
systematically explore this sensory response, the effect of
protonation on the photophysics of 1-7 was examined. Upon
(20) Fahrni, C. J.; Yang L. C.; VanDerveer, D. G. J. Am. Chem. Soc. 2003,
125, 3799-3812.
9
11876 J. AM. CHEM. SOC. VOL. 128, NO. 36, 2006

Cruciforms as Functional Fluorophores
A R T I C L E S
Table 4. Radiative Lifetimes of 1-7 upon the Addition of Protons and Metal Cationsa
radiative lifetime
radiative lifetime
(ns) with excess of H+
radiative lifetime
(ns) with excess of Zn2
radiative lifetime
(ns) with excess of Mg2
radiative lifetime
(ns) with excess of Mn2
radiative lifetime
(ns) with excess of Ca2
+
+
+
+
cruciform
τ
(ns) CHCl3
τ
τ
τ
τ
τ
1
4.1
3.9
3.9
3.8
3.7
3.1
3.6
6.7
7.6
2.6
2.9
3.5
2.9
3.1
2.8
3.1
2.5
3.0
2.9
1.8
2.1
2.0
3.9
4.0
5.0
4.9
3.4
2
3
4
6.8
6.9
3.1
2.8
3.1
2.7
2.9
2.6
2.9
3.7
3.1
3.3
2.1
2.0
6.9
3.0
2.9
7.0
2.9
2.8
6.4
2.8
2.7
1.7
4.2
4.4
5
6
4.4
4.2
3.9
4.0
2.9
3.4
2.9
3.0
2.7
2.9
7
4.3
4.3
2.1
1.3
2.6
a The lifetimes were monoexponential and fitted with a single decay function. In the case of Ca2+ the lifetime of the bound species was measured.
Figure 3. Emission of 1-7 and 8a in hexanes. Emission maxima are
presented in Table 1.
Upon protonation, cruciforms 1 and 2 show bathochromic
shifts and 3-6 hypsochromic shifts, while 7 shows a blue shift
followed by a red shift with decreasing the pH (Figure 5). The
unusual response exhibited by 7 can be rationalized by
qualitatively examining relative energies of the FMOs as the
solution pH is decreased. The absorption and emission maxima
in unprotonated 7 are well correlated with the relative HOMO
and LUMO energies obtained from the ab initio calculations.
The FMOs are disjoint (Figure 4), and binding of a positive
ion to the aniline branch, where the HOMO resides, will lead
to a stabilization of the HOMO but will leave the LUMO
unchanged (Figure 6). The electronic transitions (absorption and
emission) will be shifted toward shorter wavelengths. If a cation
binds now to the pyridine moieties, the LUMO will be stabilized
but the position of the HOMO will be barely influenced. As a
consequence, the HOMO-LUMO gap will shrink and both
absorption and emission will be red-shifted (Figure 6).
Figure 2. Emission of 1-7 in chloroform: (top) 1-7 without any additive;
(middle) after addition of trifluoroacetic acid; (bottom) after addition of
zinc triflate. The difference in the appearance of 7 and 1 in the presence of
zinc triflate as compared to the pictures taken in the presence of trifluoractic
acid is due to the inherent sensitivity profile of the used camera. All pictures
were taken under a black light at a maximum emission wavelength of 365
nm.
addition of an excess of trifluoroacetic acid (TFA) to the
solutions of the cruciforms 1-7 in dichloromethane or chloro-
form, the absorptions of 1 and 2 experience a red shift, while
the absorptions of 3-7 experience a significant blue shift (see
absorption spectra included in the Supporting Information).
These shifts are mirrored in the emission spectra of 1-7, where
upon protonation of 1 and 2 large red shifts are observed, while
in 3-7 blue shifts occur. In 3 and 7 the effect is moderate;
however, in 4-6 protonation results in a dramatic blue shift in
excess of 150 nm. At the same time the fluorescence quantum
yields change; in 1 and 2 the quantum yields drop from 0.7 to
0.06, while in 4-6 the quantum yields increase from 0.05-
0.06 to 0.4-0.7 upon protonation. In the most extreme cases, 5
and 6, the quantum yields increase by a factor of 7.5 upon
protonation. Overall, there is a crude and qualitative correlation
with the energy gap law.19
To investigate the effect of protonation on the spectra of 7
in more detail, we performed a titration in THF/methanol; 7 is
attractive as it contains both pyridine and dibutylamino substi-
tutents (Figure 5). Upon addition of acid, the emission of 7
em
(λmax ) 598 nmsnote that, in THF/methanol, the emission
of 7 is red-shifted when compared to the emission recorded in
dichloromethane) diminishes and is replaced by an emission at
445 nm, corresponding to 7 with dibutylamino-protonated arms
9
J. AM. CHEM. SOC. VOL. 128, NO. 36, 2006 11877

A R T I C L E S
Zucchero et al.
Figure 4. Top: Frontier molecular orbitals of 8b calculated with B3LYP 6-31G**//6-31G** using Spartan. Left: HOMO (-5.17 eV) of 8b. Right: LUMO
(-2.00 eV) of 8b. Bottom: Frontier orbitals of 7 calculated with B3LYP 6-31G**//6-31G** using Spartan. HOMO (-4.63 eV) (left) and LUMO (-2.07
eV) (right) are localized on the different branches of the molecule.
Figure 5. Protonation of 7 in methanol with trifluoroacetic acid. Emission maxima are 598 nm at pH 7.1, 437 nm at pH 2.7, and 542 nm at pH 0.34.
(λmaxem ) 445 nm). At pH 3.6, an approximate 1:1 ratio of the
dibutylamino-protonated species (λmax ) 445 nm) and the
arms and at the pyridine rings. From the above discussion we
conclude that either one or both pyridine rings could be
protonated, depending upon the difference in pKa values of the
monoprotonated pyridine and the diprotonated pyridine species.
In the second case, the pKa values of all four basic sites would
be largely independent and successive double protonation of
the dibutylamino and the pyridine group would occur.
em
nonprotonated species (cruciform 7) is found. At a pH of 2.7
all of the nonprotonated 7 has disappeared, and only protonated
species emitting at 445 nm are observed.21 Diethylaniline has a
pKa of 6.67, while pyridine displays a pKa of 5.25; from the
pH-dependent optical data we conclude that (a) the dibutylamino
group is protonated first and (b) both dibutylamino groups are
protonated independently. If that were not the case, i.e., if
monoprotonation would happen, one would expect the disap-
pearance of 50% of the neutral species to be complete at pH
6.6 and not at 3.6.
To address the problem of the stoichiometry, we investigated
the NMR spectra of 2 and 3, in deuterated chloroform, and
added TFA. Upon addition of TFA to either of the cruciforms
we see only one set of sharp signals assigned to a symmetrical
structure, suggesting in both cases that the respectively formed
species is doubly protonated at the aniline or at the pyridine
(Figures 7 and 8). In the case of a putative monoprotonated
species, we would have expected to see a more complex and/
or broadened NMR pattern. From the emission spectra of 1 and
Upon further protonation, a second isosbestic point is found
and an emission centered at 542 nm grows in. We attribute this
species to a cruciform 7 that is protonated at the dibutylamino
(21) For a collection of pKa values, see: http://www.zirchrom.com/organic.htm.
9
11878 J. AM. CHEM. SOC. VOL. 128, NO. 36, 2006

Cruciforms as Functional Fluorophores
A R T I C L E S
Figure 6. Sequential protonation or reaction of zinc ions with 7.
Figure 7. Proton NMR spectra of compound 3: (top) after addition of an excess of zinc triflate; (niddle) after addtion of an excess of trifluoracetic acid;
(bottom) 3 in chloroform-d. All spectra were taken at 500 MHz.
3 (Figure 1) we know that protonation produces one discrete
species each, i.e., the NMR spectra taken under identical
conditions cannot represent an equilibrium mixture of different
protonated species. We conclude that in 7 we first doubly
protonate the dibutylamino groups and then, in a second step,
doubly protonate the pyridine units, suggesting that the pKa
values do not change much upon protonation of one end of the
respective arm of the cruciform (Figure 6).
If one examines the fluorescence spectrum of 5 upon addition
of trifluoroacetic acid in dichloromethane, there is a residual
emission of the unprotonated species visible, suggesting that
full protonation of the aniline sites is only achieved upon
addition of a large excess of the acid. In 5 that is not too
surprising, as the four CF3 groups on the bis(phenylethynyl)-
benzene branch will decrease the basicity of the dibutylaniline
sites in a spillover process.
found. The compound 2 shows an increase in lifetime from 3.8
to 7.2 ns upon protonation. These significant changes make the
cruciforms potentially useful for fluorescence lifetime imaging
(FLIM) applications.11 In a broadly general sense, the emissive
lifetimes increase upon decreasing the HOMO-LUMO gap and
decreasing orbital overlap in the protonated species.
Addition of Zinc Triflate to Solutions of Cruciforms in
Dichloromethane. The exposure of functional cruciforms
toward zinc and other metal ions leads to chromic changes in
absorption and emission.13 It was of interest to examine more
carefully the nature of the cation/cruciform interaction. More
specifically, it was imperative to determine the locus of metal
binding. In the case of the pyridines the binding site must be
the lone electron pair. However, in case of cruciforms that
contain dialkylaniline subunits, the metal cations could either
bind to the electron-rich aromatic π-face or directly through
the nitrogen of the dialkylamino group. Examples of metal
cations binding to electron-rich aromatic π-faces are known;
therefore, the question for the location of the metal cations is
The change in radiative lifetime upon protonation varies from
moderate for 5 (4.4 to 4.0 ns) to significant for 7 (4.4 to 1.8
ns). Both increase and decrease of the radiative lifetimes are
9
J. AM. CHEM. SOC. VOL. 128, NO. 36, 2006 11879

A R T I C L E S
Zucchero et al.
Figure 8. Proton NMR spectra of compound 2: (top) after addition of an excess of zinc triflate; (middle) after addtion of an excess of trifluoracetic acid;
(bottom) 2 in chloroform-d. All spectra were take at 400 MHz.
not trivial.22 We selected zinc, a metal of importance in
biology,1,23 because it is in the d10 configuration and without
optical transition in the visible.
ures 7 and 8) results that are comparable to those we have
obtained for the protonation of these cruciforms, suggesting
again the similarity of the processes involved. Binding of the
zinc through the electron-rich aromatic π-face does not seem
to be important in solution, but in the solid state the situation
could be different.
How could one distinguish between zinc binding through
the aniline nitrogens or through the electron-rich ring? In the
reaction of cruciforms with TFA, the proton will bind to the
free electron pair of the nitrogen; i.e., protonation will occur
either at the pyridine-nitrogen or at the dibutylaniline. If the
photophysical properties of the cruciforms upon exposure to
protons and to zinc triflate are significantly different, one would
conclude that zinc ions and protons would occupy different
binding sites. Figures 1-3 and Tables 1-4 show, howeVer, that
changes in spectroscopic properties, quantum yields, and
emissiVe lifetimes of 1-7 are similar upon addition of zinc
triflate and of TFA. The discussion of the photophysical
properties of 1-7 upon addition of TFA holds as well for the
photophysics resulting from the addition of zinc salts. In both
cases, the positive charge is the determining issue. Addition
of zinc triflate to the pyridine-containing cruciforms leads to
a somewhat larger red-shift in emission and to higher fluores-
cence quantum yields when compared to the values obtained
for protonation of the cruciforms. These findings may indi-
cate a stronger binding of the divalent zinc ions than of the
protons to the pyridine units of the cruciforms. However,
the overall similarities of the series of photophysical data,
obtained for protonation vs exposure to zinc ions, suggests
that the zinc ions bind directly through the aniline nitrogens.
We investigated the 1H NMR spectra of the model cruci-
forms 2 and 3 upon addition of zinc triflate and observe (Fig-
Fluorescence Response of Cruciforms to the Addition
of the Triflates of Mg2+, Ca2+, and Mn2+. With the effects
of the addition of proton acids and zinc ions secured, we
were curious if other metal cations in their +2 oxidation state
would affect the fluorescence of the cruciforms differently.
We chose Mg2+, Ca2+, and Mn2+, all as their triflates as suit-
able representatives, because they are similar to zinc ions in
their overall properties. Manganese and magnesium ions have
roughly the same size as zinc ions, and only calcium is bigger
(Zn2+ 88 pm, d10; Mn2+ 89 pm, d5; Mg2+ 86 pm, d0; Ca2+ 114
pm, d0).24 The main difference is the increase in hardness when
going from Zn2+ to Mg2+ and the presence of five unpaired
electrons in Mn2+. We were interested to see if the presence of
the unpaired electrons would lead to a quenching of the
fluorescence of the cruciforms and how increasing hardness and
size would influence binding to the cruciforms. Figure 1 and
Tables 2-4 show the pertinent results of the binding of all four
metal cations to 1-7. We will discuss the metal responses of
the cruciforms 1-3, 5, and 7 in more detail. Upon exposure to
the set of four metals, cruciforms 1-3 show large but metal
independent shifts, similar to those obtained for zinc ions. In
cruciform 2 the response is somewhat more varied with the
emission maxima ranging from 536 to 543 nm upon addition
of the metal cations. In 1 and 2, the addition of Mg2+, Mn2+
,
(22) (a) Hirsch, K. A.; Wilson, S. R.; Moore, J. S. Inorg. Chem. 1997, 36, 2960-
2968. (b) Venkataraman, D.; Lee, S.; Moore, J. S.; Zhang, P.; Hirsch, K.
A.; Gardner, G. B.; Covey, A. C.; Prentice, C. L. Chem. Mater. 1996, 8,
2030-2040. (c) Hirsch, K. A.; Wilson, S. R.; Moore, J. S. J. Am. Chem.
Soc. 1997, 119, 10401-10412.
(23) Suh, S. W.; Jensen, K. B.; Jensen, M. S.; Silva, D. S.; Kesslak, P. J.;
Danscher, G.; Frederickson, C. J. Brain Res. 2000, 852, 274-278.
and Zn2+ transforms all of the fluorophore into a metal-bound
species, while in the case of Ca2+ only partial binding occurs.
(24) Values from http://www.scescape.net/∼woods/.
9
11880 J. AM. CHEM. SOC. VOL. 128, NO. 36, 2006

Cruciforms as Functional Fluorophores
A R T I C L E S
One can observe bound and unbound species. The more
electron-rich 2 binds better to Ca2+ ions and the ratio of the
emission intensity bound/unbound cruciform is considerably
larger than in 1.
be discerned under ideal conditions. This is surprising, because
the cruciforms were not designed to be metal sensory com-
pounds.
Conclusions
Compound 5 with its strongly electron-withdrawing triflu-
oromethyl substituents shows a different response for Zn2+ than
for Mg2+, Mn2+, and Ca2+, as an additional blue-shifted feature
is visible at 413-417 nm. The Zn2+-induced emission is broad,
featureless, and centered at 429 nm. Even saturated solutions
of Ca(OTf)2 leave a fraction of 5 unbound. The pyridine-
dibutylamino cruciform 7 shows the most varied metal response.
Both zinc and magnesium fully complex 7 but with different
emission maxima (Zn2+, 518 nm; Mg2+, 530 nm). Manganese
only binds partially to 7. A feature at 418 nm is due to
complexation of the dibutylamino groups while the second
feature at 526 nm represents the fully, 4-fold-complexed 7. In
the case of Ca2+ we also see two different features. One
corresponds to the uncomplexed cruciform, while the second
one is observed at 414 nm and represents 7 in which only the
two dibutylamino units are calcium-bound.
All of the cruciform-metal ion complexes are fluorescent.
Their quantum yields are shown in Table 3. We were not able
to determine the emission quantum yield of the Ca complexes,
because we always (with exception of 3) observe bound and
unbound species. Generally, the Mg2+ complexes show the
highest quantum yields, followed by the Mn2+ complexes. The
Zn2+ complexes are the least fluorescent ones, only to be
undercut by the protonated cruciforms. The reason for the metal-
based variations in the fluorescence quantum yield is not clear,
and it is not reflected (Table 4) by changes in the radiative
lifetimes. Here all of the different metal complexes of one
cruciform show similar fluorescence lifetimes.
The cruciforms 1-7 display radiative lifetimes considerably
longer than those observed for distyrylbenzenes. The poor spatial
orbital overlap of HOMO and LUMO in some cruciforms leads
to a weakly Franck-Condon-allowed transition; the quantum
yields correlate with the inverse of the emission wavelength.
We have investigated the responses of spectroscopic properties
of the cruciforms 1-7 to TFA and to the triflates of Mg2+, Ca2+
,
Mn2+, and Zn2+: In cruciforms the (dibutylamino)anilines are
complexed first, because they are more basic than the pyridine
sites, in accordance with pKa values from literature sources.21
Addition of protons or metal ions results in similar changes in
the optical and NMR spectroscopic properties of all of the
investigated cruciforms, suggesting that the metal ions and the
protons bind to the same sites, viz. the free electron pairs of
the dialkylaniline and the pyridine nitrogens. The parallel results
observed for the addition of protons and metal ions suggests
that protons can be used to qualitatively and conveniently assess
the photophysical sensory responses observed in operationally
functional fluorophores such as cruciforms, when positive charge
is the major contributor to the change in fluorescence.25
In future contributions, we will report upon the binding
constants of different cruciforms to Zn2+ and other metal cations
and explore the use of water-soluble cruciforms in sensory
applications. Suitable cruciforms will be recruited as members
in sensor array libraries for the detection of metal cations. Last
but not least, these functional chromophores show a wealth of
unexpected photophysical properties that makes them attractive
objects of study on their own.
From these results a self-consistent picture emerges. The
cruciforms can be ordered in their metal binding power as 3 >
2 ≈ 5 ≈ 7 > 1. The electron-releasing quality of the binding
element seems to play the largest role. The metals can also be
ranked in the power in which they bind the cruciforms, Mg2+
≈ Zn2+ > Mn2+ > Ca2+. The lesser binding ability of Ca2+ as
compared to the other metal cations is due to its increased size
and decreased charge density. Mg2+, Zn2+, and Mn2+ have all
the same size and charge but are d0, d5, and d10 configured,
respectively. Zn2+, the softer metal ion, binds better to the
pyridines than Mn2+, which is expected. The second row ion,
Mg2+, the smallest and the hardest of all of the investigated
ones, binds well both to the pyridine as well as to the
dibutylamino groups. It was surprising for us that the d5-
configured Mn2+ did not lead to quenching of the fluorescence
of any of the cruciforms, because we expected that unpaired
electrons lead to spin-orbit coupling in the metal-cruciform
complexes. Overall, the responses of the different metal cations
to the different cruciforms are varied and the four cations can
Acknowledgment. We thank the National Science Foundation
(Grants DMR 0138948, 2002-03, and DMR 0454471, 2004-
05) and Georgia Institute of Technology for financial support,
Prof. Laren Tolbert and Prof. Christoph Fahrni for helpful
discussions, and Prof. Rob Dickson for the use of the CAROM
lifetime spectroscopy setup. We thank Sandeep Patel for help
with the lifetime measurements.
Supporting Information Available: Experimental details for
the synthesis of cruciforms 4, 6, and 8a as well as details for
the photophysical experiments and the NMR experiments. This
material is available free of charge via the Internet at
http://pubs.acs.org.
JA061112E
(25) We thank an anonymous reviewer for this insight as well as the use of the
term “operationally functional”.
9
J. AM. CHEM. SOC. VOL. 128, NO. 36, 2006 11881