Binding of carboxylic acids by fluorescent pyridyl ureas

Article abstract of DOI:10.1021/jo101730w

Fluorescent pyrid-2-yl ureas were prepared by treating halogenated 2-aminopyridines with hexyl isocyanate, followed by Sonogashira coupling with arylacetylenes. The sensors emit light of ～360 nm with quantum yields of 0.05-0.1 in acetonitrile solution. Ad

Full text of DOI:10.1021/jo101730w

pubs.acs.org/joc
Binding of Carboxylic Acids by Fluorescent Pyridyl Ureas
Lisa M. Jordan,† Paul D. Boyle,‡ Andrew L. Sargent,*,† and William E. Allen*,†
†Department of Chemistry, Science and Technology Building, East Carolina University, Greenville, North
Carolina 27858-4353, United States, and ‡Department of Chemistry, North Carolina State University,
Raleigh, North Carolina 27695-8204, United States
allenwi@ecu.edu; sargenta@ecu.edu
Received September 2, 2010
Fluorescent pyrid-2-yl ureas were prepared by treating halogenated 2-aminopyridines with hexyl
isocyanate, followed by Sonogashira coupling with arylacetylenes. The sensors emit light of ∼360 nm
with quantum yields of 0.05-0.1 in acetonitrile solution. Addition of strong organic acids (pKa < 13 in
CH3CN) shifts the fluorescence band to lower energy, and clean isoemissive behavior is observed.
Fluorescence response curves (i.e., F/F0 vs [acid]total) are hyperbolic in shape for CCl3COOH and
CF3COOH, with association constants on the order of 103 M-1 for both acids. 1H NMR titrations and
DFT analyses indicate that trihaloacetic acids bind in ionized form to the receptors. Pyridine protonation
disrupts an intramolecular H-bond, thereby unfolding an array of ureido NHdonors for recognition of the
corresponding carboxylates. Methanesulfonic acid protonates the sensors, but no evidence for conjugate
base binding at the urea moiety is found by NMR. An isosteric control compound that lacks an integrated
pyridine does not undergo significant fluorescence changes upon acidification.
Introduction
protonate the pyridine ring,2 causing the (thio)urea moiety to
change orientation and present two syn NH groups suitable for
anion binding.3 Acids that are too weak to protonate the base do
not bring about conformational switching, and are recognized
in their undissociated state at the peripheral NH and CO units.4
For this work, pyridyl ureas were incorporated into diarylace-
tylene reporter units5 to generate a series of functional fluoro-
phores. The response of the sensors to a series of organic acids
Recognition of carboxylic acids is complicated by their poten-
tial to dissociate. Indicator compounds rely upon such behavior,
changing color in response to the hydrogen ions so generated,
yet they do not typically feature designated sites for binding of
the corresponding conjugate bases. Pyrid-2-yl ureas and thio-
ureas are suitable platforms for the development of acid recep-
tors, as their conformational flexibility allows them to recognize
the ionized or molecular forms of RCOOH. In organic solution,
pyridyl ureas adopt a folded shape in which one of the ureido
NH groups is intramolecularly hydrogen bonded to the hetero-
cyclic N atom.1 Strong acids such as CF3COOH or HBF4
(3) (a) Carroll, C. N.; Berryman, O. B.; Johnson, C. A.; Zakharov, L. N.;
Haley, M. M.; Johnson, D. W. Chem. Commun. 2009, 2520–2522. (b) Byrne,
P.; Lloyd, G. O.; Clarke, N.; Steed, J. W. Angew. Chem., Int. Ed. 2008, 47,
5761–5764. (c) Catalgirone, C.; Bates, G. W.; Gale, P. A.; Light, M. E. Chem.
Commun. 2008, 61–63. (d) Custelcean, R.; Jiang, D.; Hay, B. P.; Luo, W.; Gu,
B. Cryst. Growth Des. 2008, 8, 1909–1915. (e) Ghosh, K.; Masanta, G.;
Chattopadhyay, A. P. Tetrahedron Lett. 2007, 48, 6129–6132.
(4) Goswami, S.; Jana, S.; Dey, S.; Sen, D.; Fun, H.-K.; Chantrapromma,
S. Tetrahedron 2008, 64, 6426–6433.
(5) (a) Jo, J.; Lee, D. J. Am. Chem. Soc. 2009, 131, 16283–16291. (b)
Yamaguchi, Y.; Shimoi, Y.; Ochi, T.; Wakamiya, T.; Matsubara, Y.;
Yoshida, Z. J. Phys. Chem. A 2008, 112, 5074–5084. (c) Wong, J.; Masson,
P.; Nicoud, J.-F. Polym. Bull. 1994, 32, 265–272.
(1) (a) McGhee, A. M.; Kilner, C.; Wilson, A. J. Chem. Commun. 2008,
344–346. (b) Chien, C.-H.; Leung, M.-K.; Su, J.-K.; Li, G.-H.; Liu, Y.-H.;
Wang, Y. J. Org. Chem. 2004, 69, 1866–1871. (c) Corbin, P. S.; Zimmerman,
S. C.; Thiessen, P. A.; Hawryluk, N. A.; Murray, T. J. J. Am. Chem. Soc.
2001, 123, 10475–10488.
(2) Rashdan, S.; Light, M. E.; Kilburn, J. D. Chem. Commun. 2006, 4578–
4580.
8450 J. Org. Chem. 2010, 75, 8450–8456
Published on Web 11/16/2010
DOI: 10.1021/jo101730w
r
2010 American Chemical Society

Jordan et al.
JOCArticle
SCHEME 1
was evaluated in acetonitrile solution, where the extent of acid
ionization is low.
Hb (Nureido Npyridyl = 2.70 A). Close approach between the
3 3 3
hexyl chain and the phenyl group, presumably resulting from
packing forces, is accompanied by rotation of the ring by 32°
relative to the plane defined by the pyridine heterocycle. Mole-
cules of 6-H are further organized into dimers via pairs of
H-bonds at the external NHa and CO units.1c,4 The hydrogen
bonding present in solid samples of 6-H/CH3/CF3 is stronger
than that of the 5-isomers, as judged by IR stretching frequen-
cies of the NH groups. Comparing 6-H to 5-H, for example, the
Results and Discussion
Synthesis and Structure of the Sensors. The desired fluoro-
phores were prepared on a ∼0.5 g scale in two steps without
chromatographic purification (Scheme 1). Commercially avai-
lable halogenated arylamines 1 were treated with hexyl isocya-
nate in THF at reflux to afford crude ureas 2-4. These inter-
mediates were purified by washing CH2Cl2 solutions with 5%
HCl to remove unreacted amines. Using a modular approach
to impart electronic variability in the fluorophore, substituted
ethynylarenes were coupled to 2-4 under Cu-free Sonogashira
conditions to provide the isomeric sets of products 5and 6,6 and
nonpyridinic control compound 7. In all cases, analytically
pure materials crystallized out of the reaction mixtures upon
cooling.
former has H-bonded NH-stretches at 3214 and 3122 cm-1
while the latter’s appear at 3224 and 3134 cm-1
,
.
In acetonitrile solution, the proton NMR chemical shifts of
ureido proton Hb in 5-H/CH3/CF3 (8.35, 8.35, and 8.30 ppm,
respectively) and 6-H/CH3/CF3 (8.54, 8.58, and 8.46 ppm)
indicate that the sensors retain a folded conformation that is
not available to control compound 7(δHb=5.23 ppm). Previous
work with 1-butyl-3-(4-methylpyridin-2-yl)urea has established
the shift of this resonance to be concentration-independent in
CDCl3.1c In CD3CN, no changes in δHb occurred as a solution
of 6-H was diluted from 2.1ꢀ10-3 M to 9.0ꢀ10-5 M. Taken
together, the NMR and IR data are consistent with the closed
forms of the 6-arylethynyl isomers being marginally more stable
than those with 5-substitution. Heating disrupts the intramolec-
ular H-bonding to a small extent. When a 6.02 ꢀ 10-3
M
solution of 5-H in CD3CN was warmed from 50 to 90 °C in a
flame-sealed tube, the Hb signal moved upfield by 0.2 ppm.
Fluorescence Response to Acids. In the absence of acids,
sensors 5-6show a single featureless fluorescence band between
350 and 370 nm in N2-saturated acetonitrile (Table 1). Fluores-
cence quantum yields (ΦF) for electron-rich systems 5-H/CH3
and 6-H/CH3 are higher than those of the trifluoromethyl-
substituted analogues. Incremental addition of trichloroacetic
acid (pKa in CH3CN=10.757) or trifluoroacetic acid (12.65)
causes bathochromic shifts of 49 and 64 nm in the emission
profiles of 5-H and 5-CH3, respectively (Figure 2a), with
modest drops in ΦF that qualitatively obey the energy gap law.8
FIGURE 1. View of 6-H in the crystal. Thermal ellipsoids are
scaled to the 30% probability level.
Crystals of 6-H suitable for X-ray diffraction were grown
from acetonitrile in the absence of acids. Figure 1 shows one of
the two crystallographically independent molecules found in the
unit cell. The sensor adopts a compact, hairpin-like conforma-
tion stabilized by an intramolecular hydrogen bond involving
€
(7) (a) Eckert, F.; Leito, I.; Kaljurand, I.; Kutt, A.; Klamt, A.; Diedenhofen,
€
M. J. Comput. Chem. 2009, 30, 799–810. (b) Kutt, A.; Leito, I.; Kaljurand, I.;
Soovali, L.; Vlasov, V. M.; Yagupolskii, L. M.; Koppel, I. A. J. Org. Chem.
€
2006, 71, 2829–2838.
(6) Compound numbers in Scheme 1 were assigned so that sensors with
arylacetylene units attached to position 5 of the pyridine ring begin with “5”,
and those featuring C-6 substitution begin with “6”.
(8) (a) Spitler, E. L.; Shirtcliff, L. D.; Haley, M. M. J. Org. Chem. 2007,
72, 86–96. (b) Resch-Genger, U.; Li, Y. Q.; Bricks, J. L.; Kharlanov, V.;
Rettig, W. J. Phys. Chem. A 2006, 110, 10956–10971.
J. Org. Chem. Vol. 75, No. 24, 2010 8451

Jordan et al.
JOCArticle
TABLE 1. Photophysical Data in CH3CN
abs λmax (nm)
ε (M-1 cm-1
)
em λmax (nm)
ΦF
em λmax,b þ acid (nm)
ΦF,b þ acid
a
5-H
295
297
317
319, 271
321, 274
320, 275
301
30500
36100
33600
19500, 18400
19500, 16200
18700, 19500
34400
352
356
366
360
357
367
359
0.084
0.096
0.060
0.11
0.084
0.056
0.046
401
420
374
393
410
374
359
0.083
0.065
0.11
0.25
0.10
5-CH3
5-CF3
6-H
6-CH3
6-CF3
7
0.082
0.043
aReferenced to 7-amino-4-methylcoumarin (ΦF = 0.33 in CH3OH). bMeasured at the completion of titrations with CH3SO3H, CCl3COOH,
or CF3COOH.
FIGURE 2. Emission spectra of selected sensors in CH3CN during titration with CF3COOH (0 f 1000 equiv): (a) 5-CH3 (1.32 ꢀ 10-6 M),
λ
ex = 300 nm; (b) 6-H (1.56 ꢀ 10-6 M), λex = 320 nm; and (c) 5-CF3 (1.41 ꢀ 10-6 M), λex = 320 nm.
With 6-H and 6-CH3 the red shifts are less pronounced. Proto-
nation studies of fluorescent tetrakis(arylethynyl)benzenes have
shown that electronic communication differs between pyridyl N
atoms and arylethynyl linkers that lie ortho or meta to each
other.8a Quantum yields for all three of the 6-isomers increase
upon acidification; the most extreme case is shown in Figure 2b.
Sensors 5-CF3 (Figure 2c) and 6-CF3 experience a red shift of
less than 10 nm. Compound 7, which lacks an intended
protonation site, is unaffected by acidification.
Protonation of pyridine-containing fluorophores can give
rise to new emission bands at longer wavelengths, by making
internal charge transfer (ICT) from a donor to the heterocycle
energetically favorable.9 Frontier molecular orbitals for repre-
sentative fluorophores 5-CH3 and 5-CF3 were visualized by
using time-dependent DFT10 in a solvent field corresponding to
CH3CN. In both of these hosts the HOMO and LUMO are
uniformly delocalized over the urea-diarylacetylene π system,
with electronic transition maxima calculated to appear at 328
and 338 nm, respectively. However, the next-highest MO for
5-CH3 (LUMOþ1) resides exclusively on the pyridine ring
(Figure 3). As observed for other embedded-base fluoro-
phores,11 protonation should stabilize the localized orbital
to a greater extent than the spatially dispersed HOMO
or LUMO. This is borne out by the putative complex
Hþ 5-CH3 CF3COO-, in which Hþ and CF3COO- ions are
3
3
bound at the pyridyl N atom and urea NH groups, respectively.
Here, the pyridine-centered MO is predicted to lie below the
delocalized one in energy, and emission from Hþ 5-CH3
3
3
CF3COO- is bathochromically shifted to 355 nm. In con-
trast, the calculated λmax for trifluoromethyl-containing com-
plex Hþ 5-CF3 CF3COO- (340 nm) is almost identical with
3
3
that expected for the free base.
Fluorescence spectra of the six sensors were integrated from
their respective isoemissive points to 500 nm during titrations
with CCl3COOH and CF3COOH. The resultant F values
increase by a factor of 4-6 for 5-H/CH3 and 6-H/CH3, and
approximately double for 5-CF3 and 6-CF3. Normalized
response curves (i.e., F/F0 vs [acid]total) are hyperbolic for all but
6-CF3, for which they have weak and distorted sigmoidal charac-
ter (see the Supporting Information). Sensitivity is low. Addition
of an approximately 1000-fold excess of each acid is required to
achieve saturation. The 1:1 association constants12,13 (Table 2)
show that the sensors are not selective for trichloroacetic acid,
even though it is significantly more ionized than CF3COOH in
acetonitrile. Indeed, the DFT-calculated binding energies for
Hþ 5-H CCl3COO- and Hþ 5-H CF3COO- in an acetoni-
3
3
3
3
(9) (a) Liu, Y.; Han, M.; Zhang, H.-Y.; Yang, L.-X.; Jiang, W. Org. Lett.
2008, 13, 2873–2876. (b) Pischel, U.; Heller, B. New J. Chem. 2008, 32, 395–
400. (c) Kozhevnikov, D. N.; Shabunina, O. V.; Kopchuk, D. S.; Slepukhin,
P. A.; Kozhevnikov, V. N. Tetrahedron Lett. 2006, 47, 7025–7029. (d) Ihmels,
H.; Meiswinkel, A.; Mohrschladt, C. J.; Otto, D.; Waidelich, M.; Towler, M.;
White, R.; Albrecht, M.; Schnurpfeil, A. J. Org. Chem. 2005, 70, 3929–3938.
(10) (a) Delley, B. J. Chem. Phys. 1990, 92, 508–517. (b) Delley, B. J.
Chem. Phys. 2000, 113, 7756–7764. (c) Becke, A. D. J. Chem. Phys. 1988, 88,
1053–1062. (d) Tsuneda, T.; Hirao, K. Chem. Phys. Lett. 1997, 268, 510–520.
(e) Listorti, A.; Esposti, A. D.; Kishore, R. S. K.; Kalsani, V.; Schmittel, M.;
Armaroli, N. J. Phys. Chem. A 2007, 111, 7707–7718.
trile dielectric continuum, -9.3 and -9.6 kcal/mol, respectively,
(12) (a) Hyperbolic fitting: Causey, C. P.; Allen, W. E. J. Org. Chem.
2002, 67, 5963–5968. (b) Sigmoidal fitting: Tian, Y.; Su, F.; Weber, W.;
Nandakumar, V.; Shumway, B. R.; Jin, Y.; Zhou, X.; Holl, M. R.; Johnson,
R. H.; Meldrum, D. R. Biomaterials 2010, 31, 7411–7422. (c) Connors, K. A.
Binding Constants; Wiley: New York, 1987.
(13) Association constants derived from ratiometric treatment of the
fluorescence data were unreasonably large. For a system that gives consistent
Kassoc values from both direct and ratiometric fits, see: Baruah, M.; Qin, W.;
(11) (a) Zucchero, A. J.; McGrier, P. L.; Bunz, U. H. F. Acc. Chem. Res.
2010, 43, 397–408. (b) Zucchero, A. J.; Wilson, J. N.; Bunz, U. H. F. J. Am.
Chem. Soc. 2006, 128, 11872–11881.
ꢀ
Vallee, R. A. L.; Beljonne, D.; Rohand, T.; Dehaen, W.; Boens, N. Org. Lett.
2005, 7, 4377–4380.
8452 J. Org. Chem. Vol. 75, No. 24, 2010

Jordan et al.
JOCArticle
FIGURE 4. Chemical shifts of selected sensors in CD3CN during
titration with trihaloacetic acids: (a) 5-CH3 (5.74 ꢀ 10-3 M) þ
CF3COOH and (b) 6-CF3 (5.88 ꢀ 10-3 M) þ CCl3COOH.
FIGURE 3. Calculated frontier molecular orbitals of 5-CH3 (a) before
and (b) after complexation with CF3COOH.
TABLE 2. Fluorescence Response of Sensors 5-6 to Increasing
Concentrations of Acetic Acids in CH3CNa
acid does not alter the position of the fluorescence band of any
of the sensors. Integrated intensities, obtained over the entire
emission range, decrease linearly with [CH3COOH]. The aver-
age Stern-Volmer quenching constant (KSV) with 5-H/CH3/
CCl3COOH
CF3COOH
CHCl2COOH
CH3COOH
5-H
5000
4500
2300
600
1000
1100b
3900
5200
2600
800
1900
1000b
quenching
quenching
∼200c
quenching
quenching
quenching
quenching
quenching
quenching
5-CH3
5-CF3
6-H
6-CH3
6-CF3
CF3 is 46 ( 4 M-1 16 An identical KSV value (44 ( 1 M-1) was
.
∼200c
found using the 6-isomers. Dichloroacetic acid causes fluores-
cence enhancement in 5-CF3, 6-H, and 6-CH3, though only
6-CH3 binds the acid strongly enough to display saturation
behavior, and acts as a simple quencher toward the other fluoro-
phores. Thus, the pKa of this sensor class lies between about 13
and 16, in good agreement with the pKa of 2-aminopyridine,
14.47.17
630
quenching
aNumerical values are averaged association constants (in M-1) derived
from at least two replicate experiments. Unless otherwise indicated, binding
isotherms were fit to a hyperbolic function as described in the Experimental
b
c
Section. Sigmoidal plots of F/F0 vs log[acid]total. Negligible curvature in
binding isotherms.
Proton NMR Response to Acids. 1H NMR binding
studies were performed in CD3CN for 5-H/CH3/CF3 with
CCl3COOH and CF3COOH. The chemical shifts from a
typical titration of 5-CH3 þ trifluoroacetic acid are pre-
sented in Figure 4a. Only pyridyl proton H4, which is more
than three bonds removed from the expected sites of pro-
tonation and carboxylate recognition, moves in a smooth
asymptotic manner. Both H3 and H6 subtly reverse direction
at >1 equiv of acid, while the reaction of H-bonded urea
proton Hb is more dramatic. Initially this resonance under-
goes a small increase in δ indicative of more robust H-bond-
ing, presumably to CF3COO-. At higher [CF3COOH] the
favor the acid with the more basic carboxylate ion. Kassoc values
must reflect the composite thermodynamics of both N-proton-
ation and hydrogen bonding to conjugate bases, so carboxylate
binding makes a significant contribution to the overall stabilities
of complexes.
Additional titrations were performed with CH3COOH (pKa
in CH3CN=23.51) and CHCl2COOH (15.814) to delineate the
minimum acid strength necessary to “activate” 5-6.15 Acetic
(14) Hayakawa, Y.; Iwase, T.; Nurminen, E. J.; Tsukamoto, M.; Kataoka,
M. Tetrahedron 2005, 61, 2203–2209.
(15) In acetonitrile, 5-H and 6-CH3 do not display “off-on” fluorescence
response toward CH3COO- or Cl- (as tetrabutylammonium salts). Acetate ion
acts as a quencher, while chloride causes no change in F or λem. For reviews of
urea-based anion receptors, see: (a) Steed, J. W. Chem. Soc. Rev. 2010, 39, 3686–
3699. (b) Li, A.-F.; Wang, J.-H.; Wang, F.; Jiang, Y.-B. Chem. Soc. Rev. 2010, 39,
3729–3745. (c) Gale, P. A. Chem. Soc. Rev. 2010, 39, 3746–3771. (d) Amedola, V.;
Fabbrizzi, L.; Mosca, L. Chem. Soc. Rev. 2010, 39, 3889–3915.
´
(16) Sadowska, M.; Miller, E.; Wysocki, S.; Rodrıguez-Muniz, G. M.;
Costero, A. M.; Wandelt, B. J. Lumin. 2010, 130, 1085–1091.
€
€
€
(17) Kaljurand, I.; Kutt, A.; Soovali, L.; Rodima, T.; Maemets, V.; Leito,
I.; Koppel, I. A. J. Org. Chem. 2005, 70, 1019–1028.
J. Org. Chem. Vol. 75, No. 24, 2010 8453

Jordan et al.
JOCArticle
FIGURE 5. Calculated chemical shifts (ppm, in CH3CN dielectric field) of 5-CH3 and its proposed complexes with CF3COOH.
shift of Hb decreases significantly. Addition of CCl3COOH
results in similar qualitative behavior. Maximum shifts of Hb
are 0.05 ppm lower, on average, than those measured during
titrations with CF3COOH, reflecting the poorer H-bond ac-
ceptor ability of trichloroacetate. In every case the data from H4
fit cleanly to a 1:1 stoichiometry model,18 but with apparent
binding constants that are an order of magnitude lower than
ones derived from fluorescence experiments. Given the com-
plex behavior of H3, H6, and Hb, the discrepancies likely arise
from the presence of additional solution equilibria.19 Attempts
to describe the shifts of H3 by including both 1:1 and 1:2 sensor/
acid stoichiometries in the fitting routine brought the K1:1
values more in line with those from fluorescence, but large
errors in K1:2 rendered the overall fits unreliable. Sensor self-
association, which has been observed for other heterocyclic
ureas in the NMR concentration regime,1c was determined to
be a minor factor. Diluting a CD3CN solution of 6-H from
2.1 ꢀ 10-3 M to 9.0 ꢀ 10-5 M caused the ureido Ha signal to
progress upfield by 0.03 ppm.
Hþ 6-CF3 failed to identify obvious alternative sites for recogni-
3
tion of CX3COO-;then the H-bonding from Hb to CXCOO-
must be weak. Notably, sigmoidal fluorescence behavior is
elicited from all six sensors by CH3SO3H, the conjugate base
of which is a significantly poorer hydrogen bond acceptor than
the trifluoroacetates. The pKa of methanesulfonic acid is 9.97 in
CH3CN, and its tetrahedral sulfonate is not an optimal steric
match to the planar H-bonding array presented by a urea
group. F values of 5-6 increase until 10-20 equiv of CH3SO3H
are present, confirming that the fluorophores are protonated
by this acid. Normalized response curves have asymmetrical
“S”-shapes, which resolve into sigmoidal outputs typical of acid/
base titrations by plotting F/F0 as a function of log[CH3SO3-
H]total (see the Supporting Information). An NMR study of
5-CF3 showed only upfield shifts in Hb during acidification
with CH3SO3H in acetonitrile. Therefore, it appears that sensor
protonation and unfolding can occur without concomitant
binding of the conjugate base at the urea moiety.
Simulated 1H NMR spectra were generated for the lowest-
energy DFT structures of 5-CH3 and Hþ 5-CH3 CF3COO-.
Conclusions
3
3
Fluorescence output of the present pyridyl urea-based recep-
tors is intimately linked to the strength of the organic acids added
to them, though members of this class do not report solely on
[Hþ]. Trichloroacetic and trifluoroacetic acids, at equivalent
concentrations, induce similar “off-on” emission enhancements
despite their pKa difference of nearly 2 orders of magnitude. The
relatively greater basicity of CF3COO- leads to tighter H-bond-
ing at the urea NH groups, compensating for the parent acid’s
lower degree of ionization, and driving complexation.
The acid-binding mode at the center of Figure 5 correctly
predicts the behavior of urea Hb at the outset of the titrations.
This proton is calculated to become more deshielded as its
intramolecular interaction with pyridine is replaced by an inter-
molecular one with trifluoroacetate (δcalc = 9.07f10.41 ppm).
Furthermore, the complex features downfield shifts in pyridyl
H3 and H4, and an upfield change in H6, as experimentally
observed. Upon removing the bound CF3COO- guest, the com-
puted shifts of cation Hþ 5-CH3 follow the directional trends of
3
the actual titrations at high [CF3COOH]. Excess acid may facili-
tate decomplexation of carboxylate by allowing for formation of
Experimental Section
species like CF3COO- (HOOCF3)n, with favorable hard
1-Hexyl-3-(5-iodopyridin-2-yl)urea (2). A round-bottomed
flask containing 15 mL of dry THF was charged with 2-amino-5-
iodopyridine (1.14 g, 5.18 mmol) and hexyl isocyanate (0.66 g,
5.2 mmol). Under N2, the mixture was heated to reflux with stirring
for 72 h. The solution was allowed to cool to rt and the THF was
removed by rotary evaporation. The resulting yellow oil was
dissolved in CH2Cl2 and washed with 5% HCl then saturated
NaHCO3. The organic layer was dried over Na2SO4, filtered, and
evaporated to afford 1.39 g (77%) of 2as a cream-colored solid. This
material was used in the subsequent step without further purifica-
tion. Mp 94-96 °C; 1H NMR (CDCl3) δ 0.90 (t, 3H), 1.3-1.7 (m,
8H), 3.36 (q, 2H), 6.74 (d, 1H), 7.80 (dd, 1H), 8.34 (s, 1H), 9.01 (br s,
1H), 9.11 (s, 1H); 13C NMR (CDCl3) δ 14.0, 22.6, 26.6, 30.0, 31.5,
39.9, 81.5, 114.2, 145.9, 151.8, 152.6, 155.9; HRMS (ESI) calcd for
C12H19IN3O (M þ H)þ 348.0567, found 348.0520.
3 3 3
-hard acid-base character.20
The sensor/acid combinations that produced anomalous
sigmoidal fluorescence curves were also examined by NMR.
Treatment of 6-CF3 with CCl3COOH or CF3COOH causes
all three pyridyl protons to shift downfield in a manner
consistent with protonation (Figure 4b), although natural popu-
lation analysis of pyridyl N atomic charges suggests 6-CF3 is less
basic than other sensors. Urea Hb does not experience an in-
crease in chemical shift early in the titrations, but instead tracks
uniformly upfield. Assuming that a carboxylate binding model
akin to that shown at the center of Figure 5 is operative for
this system;examination of the electrostatic potential map of
1-(6-Bromopyridin-2-yl)-3-hexylurea (3). The procedure for 2 was
followed, using 0.76 g (4.4 mmol) of 2-amino-6-bromopyridine and
0.57 g (4.5 mmol) of hexyl isocyanate. After workup, 1.12 g (85%)
of 3 was obtained as light yellow crystals. This material was used in
the subsequent step without further purification. Mp 78-80 °C;
(18) Hynes, M. J. J. Chem. Soc., Dalton Trans. 1993, 311–312.
(19) dos Santos, C. M. G.; McCabe, T.; Watson, G. W.; Kruger, P. E.;
Gunnlaugsson, T. J. Org. Chem. 2008, 73, 9235–9244.
ꢀ
ꢀ
(20) Perez, P.; Toro-Labbe, A.; Contreras, R. J. Phys. Chem. A 2000, 104,
5882–5887.
8454 J. Org. Chem. Vol. 75, No. 24, 2010

Jordan et al.
JOCArticle
1H NMR (CDCl3) δ 0.90 (t, 3H), 1.3-1.7 (m, 8H), 3.39 (q, 2H),
6.89 (d, 1H), 7.02 (dd, 1H), 7.42 (t, 1H), 8.84 (br s, 1H), 9.48 (s, 1H);
13C NMR (CDCl3) δ 14.1, 22.6, 26.7, 29.6, 31.6, 40.0, 110.3, 120.0,
137.9, 140.1, 153.5, 155.5; HRMS (ESI) calcd for C12H19BrN3O
(M þ H)þ 300.0706, found 300.0800.
112.2, 122.1, 125.4, 126.7, 131.7, 140.7, 149.5, 152.9, 155.8; FTIR
(ATR, solid) ν 3338, 3224, 3133, 1671 cm-1; UV/vis (CH3CN) λmax
(ε,M-1 cm-1) 317 (33600). Anal. Calcd for C21H22F3N3O: C, 64.77;
H, 5.69; N, 10.79. Found: C, 64.72; H, 5.64; N, 10.71.
1-Hexyl-3-(6-(phenylethynyl)pyridin-2-yl)urea (6-H). The
procedure for 5-H was followed, using 0.60 g (2.0 mmol) of 3. After
6 h, stirring was stopped and the orange solution was allowed to cool
to rt for 18 h in the dark. The precipitated product was collected by
filtration, washed with ice-cold CH3CN, and dried under vacuum to
1-Hexyl-3-(4-iodophenyl)urea (4). The procedure for 2 was
followed, using 1.97 g (8.99 mmol) of 4-iodoaniline and 1.37 g
(10.8 mmol) of hexyl isocyanate in 30 mL of THF. After workup,
3.09 g (99%) of 4 was obtained as an off-white solid. Recrystalliza-
tion from absolute ethanol provided an analytical sample as color-
1
yield 0.34 g (53%) of 6-H as yellow plates. Mp 153-154 °C; H
1
less plates. Mp 166-168 °C; H NMR (CDCl3) δ 0.89 (t, 3H),
NMR (CDCl3) δ 0.83 (t, 3H), 1.3-1.5 (m, 4H), 1.47 (m, 2H), 1.66
(m, 2H), 3.42 (q, 2H), 6.81 (d, 1H), 7.11 (d, 1H), 7.39 (m, 3H), 7.55
(m, 3H), 8.31 (s, 1H), 9.30 (br s, 1H); 13C NMR (CDCl3) δ 14.0,
22.6, 26.7, 29.6, 31.6, 39.9, 88.3, 89.0, 111.6, 120.3, 122.2, 128.4,
129.0, 131.9, 138.2, 139.5, 153.4, 155.8; FTIR (ATR, solid) ν 3358
1.2-1.4 (m, 6H), 1.52 (m, 2H), 3.24 (q, 2H), 4.55 (br s, 1H), 6.13 (s,
1H), 7.10 (d, 2H), 7.60 (d, 2H); 13C NMR (DMSO-d6) δ 13.9, 22.0,
26.0, 29.6, 31.0, 83.3, 119.8, 137.1, 140.5, 154.9; HRMS (ESI) calcd
for C13H20IN2O (M þ H)þ 347.0615, found 347.0633.
1-Hexyl-3-(5-(phenylethynyl)pyridin-2-yl)urea (5-H). A pressure
tube containing 10 mL of acetonitrile and a stir bar was charged with
2 (0.69 g, 2.0 mmol), phenylacetylene (0.22 g, 2.1 mmol), and
piperidine (0.86 g, 10 mmol). With stirring, N2 was bubbled into
the suspension for 5 min, followed by the addition of 0.047 g (0.040
mmol) of tetrakis(triphenylphosphine)palladium(0). The pressure
tube was immediately sealed and lowered into an oil bath that was
preheatedto85°C. The mixture clarified over several minutes. After
90 min, stirring was stopped and the tube was allowed to cool to rt
for 18 h in the dark. During this time, the mixture deposited a cream
colored precipitate. The solid was collected by filtration, washed
with ice-cold CH3CN, and dried under vacuum to afford 0.58 g
(wk), 3214, 3122, 1677 cm-1; UV/vis (CH3CN) λmax (ε, M-1 cm-1
)
271 (18400), 319 (19500). Anal. Cacld for C20H23N3O: C, 74.74; H,
7.21; N 13.07. Found: C, 74.77; H, 7.21; N, 12.95.
1-Hexyl-3-(6-(p-tolylethynyl)pyridin-2-yl)urea (6-CH3). The
procedure for 5-H was followed, using 0.541 g (1.81 mmol) of
3 and 0.211 g (1.82 mmol) of p-tolylacetylene. The reaction clarified
after several minutes. After 6 h, stirring was stopped and the tube
was allowed to cool to rt for 18 h in the dark. The yellow solution
deposited a precipitate that was collected via filtration and washed
with ice-cold CH3CN. Drying under vacuum afforded 0.37 g (60%)
of 6-CH3 as a pale yellow solid. Mp 145-146 °C; 1HNMR(CDCl3)
δ 0.87 (t, 3H), 1.30 (m, 4H), 1.44 (m, 2H), 1.65 (m, 2H), 2.39 (s, 3H),
3.42 (q, 2H), 6.88 (d, 1H), 7.08 (dd, 1H), 7.18 (d, 2H), 7.44 (d,
2H),7.53 (t, 1 H), 8.94 (s, 1H), 9.32 (br s, 1H); 13C NMR (CDCl3) δ
14.1, 21.6, 22.6, 26.8, 29.7, 31.6, 39.9, 87.9, 89.3, 111.6, 119.1, 120.2,
129.2, 131.8, 138.2, 139.3, 139.6, 153.5, 156.1; FTIR (ATR, solid) ν
3364 (wk), 3218, 3110, 1679 cm-1; UV/vis (CH3CN) λmax (ε, M-1
cm-1) 274 (16200), 321 (19500); HRMS (ESI) calcd for C21H26N3O
(M þ H)þ 336.2076, found 336.2006. Anal. Calcd for C21H25N3O:
C, 75.19; H, 7.51; N, 12.53. Found: C, 74.95; H, 7.47; N, 12.68.
1-Hexyl-3-(6-((4-(trifluoromethyl)phenyl)ethynyl)pyridin-2-yl)-
urea (6-CF3). The procedure for 5-H was followed, using 0.60 g
(2.0 mmol) of 3 and 0.37 g (2.2 mmol) of 1-ethynyl-4-(trifluoro-
methyl)benzene. After 6 h, stirring was stopped and the mixture
was allowed to cool to rt for 18 h in the dark. The precipitate was
collected via filtration, washed with ice-cold CH3CN, and dried
under vacuum to give 0.44 g (56%) of 6-CF3 as white needles. Mp
166-167 °C; 1H NMR (CDCl3) δ 0.83 (t, 3H), 1.30 (m, 4H), 1.47
(m, 2H), 1.63 (m, 2H), 3.42 (q, 2H), 6.84 (d, 1H), 7.14 (d, 1H), 7.61
(m, 5H), 8.36 (s, 1H), 9.23 (br s, 1H); 13C NMR (CDCl3) δ 14.0,
22.6, 26.8, 29.6, 31.6, 39.9, 87.3, 90.4, 112.1, 120.6, 125.4, 132.2,
138.3, 138.9, 153.4, 155.6; FTIR (ATR, solid) ν 3366 (wk), 3220,
3118, 1682 cm-1; UV/vis (CH3CN) λmax (ε, M-1 cm-1) 275
1
(89%) of 5-H as a lustrous, off-white solid. Mp 131-133 °C; H
NMR (CDCl3) δ 0.91 (t, 3H), 1.3-1.5 (m, 4H), 1.5-1.7 (m, 4H),
3.39 (q, 2H), 6.73 (d, 1H), 7.36 (m, 3H), 7.52 (m, 2H), 7.70 (dd, 1H),
7.86 (s, 1H), 8.35 (d, 1H), 9.17 (br s, 1H); 13C NMR (CDCl3) δ 14.0,
22.6, 26.7, 29.8, 31.5, 40.0, 85.9, 91.2, 111.2, 113.1, 122.8, 128.4,
128.5, 131.5, 140.7, 149.2, 152.1, 155.3; FTIR (ATR, solid) ν 3342,
3224, 3134, 1667 cm-1; UV/vis (CH3CN) λmax (ε, M-1 cm-1) 295
(30500); HRMS (ESI) calcd for C20H24N3O (M þ H)þ 322.1919,
found 322.1790. Anal. Calcd for C20H23N3O: C, 74.74; H, 7.21; N,
13.07. Found: C, 74.62; H, 7.12; N, 12.80.
1-Hexyl-3-(5-(p-tolylethynyl)pyridin-2-yl)urea (5-CH3). The
procedure for 5-H was followed, using 0.512 g (1.47 mmol) of
2, 0.171 g (1.47 mmol) of p-tolylacetylene, 0.65 g (7.6 mmol) of
piperidine, and 0.034 g (0.030 mmol) of Pd(PPh3)4 catalyst. The
reaction mixture clarified over several minutes and after 1 h a
precipitate had formed. Stirring was stopped at 90 min and the
tube was allowed to cool to rt for 18 h in the dark. The
precipitate was collected by filtration and washed with ice-cold
CH3CN. Drying under vacuum gave 0.39 g (80%) of 5-CH3 as a
white solid. Mp 145-147 °C; 1H NMR (CDCl3) δ 0.90 (t, 3H),
1.2-1.5 (m, 6H), 1.60 (m, 2H), 2.38 (s, 3H), 3.38 (q, 2H), 6.70 (d,
1H), 7.17 (d, 2H), 7.41 (d, 2H), 7.67 (s, 1H), 7.68 (d, 1H), 8.34 (s,
1H), 9.17 (br s, 1H); 13C NMR (CDCl3) δ 14.1, 21.5, 22.6, 26.7,
29.9, 31.5, 40.0, 85.2, 98.6, 111.1, 113.4, 119.7, 129.2, 131.4,
138.7, 140.7, 149.2, 152.0, 155.2; FTIR (ATR, solid) ν 3352,
(19500), 320 (18700). Anal. Calcd for C21H22F3N3O 0.25(H2O):
3
C, 64.03; H, 5.76; N, 10.67. Found: C, 64.27; H, 5.53; N, 10.67.
1-Hexyl-3-(4-(phenylethynyl)phenyl)urea (7). The procedure for
5-H was followed, using 0.70 g (2.0 mmol) of 4. Stirring was stopped
after 90 min and the tube was allowed to cool to rt for 18 h in the
dark. The precipitate was collected by filtration, washed with ice-
cold CH3CN, and dried under vacuum to afford 0.46 g (71%) of 7as
yellow needles. Mp 134-142 °C dec; 1H NMR (CDCl3) δ 0.89 (t,
3H), 1.2-1.4 (m, 6H), 1.52 (m, 2H), 3.25 (q, 2H), 4.75 (br s, 1H),
6.43 (s, 1H), 7.32 (t, 5H), 7.46 (d, 2H), 7.51 (dd, 2H); 13C NMR
(CDCl3) δ 14.0, 22.6, 26.6, 30.0, 31.5, 40.6, 88.8, 89.2, 118.0, 119.9,
123.4, 128.1, 128.3, 131.5, 132.6, 138.8, 155.1; UV/vis (CH3CN) λmax
(ε, M-1 cm-1) 301 (34400). Anal. Calcd for C21H24N2O: C, 78.71;
H, 7.55; N, 8.74. Found: C, 78.80; H, 7.51; N, 8.64.
3223, 3130, 1670 cm-1; UV/vis (CH3CN) λmax (ε, M-1 cm-1
)
297 (36100); HRMS (ESI) calcd for C21H26N3O (M þ H)þ
336.2076, found 336.2094. Anal. Calcd for C21H25N3O: C,
75.19; H, 7.51; N, 12.53. Found: C, 75.17; H, 7.43; N, 12.39.
1-Hexyl-3-(5-((4-(trifluoromethyl)phenyl)ethynyl)pyridin-2-yl)-
urea (5-CF3). The procedure for 5-H was followed, using 0.63 g
(1.8 mmol) of 2, 0.32 g (1.9 mmol) of 1-ethynyl-4-(trifluoromethyl)-
benzene, 0.77 g (9.0 mmol) of piperidine, and 0.042 g (0.036 mmol)
of Pd(PPh3)4 catalyst. After 90 min, stirring was stopped and the
tube was allowed to cool to rt for 18 h in the dark. The yellow preci-
pitate was collected via filtration, washed with ice-cold CH3CN, and
dried under vacuum to give 0.55 g (79%) of 5-CF3 as a fluffy, off-
Fluorescence Titrations. Sensors 5-7 were dissolved in spectro-
photometric grade CH3CN to concentrations of 1.5 ꢀ 10-6-2.0 ꢀ
10-6 M, and an electronic absorption spectrum was acquired for
each to ensure that the absorbance was less than 0.05. A 3.0 mL
sample was transferred to a quartz cuvette containing a small
magnetic stir bar. The solution was excited near its predetermined
1
white solid. Mp 160-162 °C; H NMR (CDCl3) δ 0.91 (t, 3H),
1.2-1.5 (m, 6H), 1.63 (m, 2H), 3.40 (q, 2H), 6.89 (d, 1H), 7.62 (s,
4H), 7.73 (dd, 1H), 8.36 (s, 1H), 9.08 (s, 1H), 9.22 (br s, 1H); 13
C
NMR (CDCl3) δ 14.1, 22.6, 26.7, 29.9, 31.5, 40.0, 88.5, 89.7, 111.7,
J. Org. Chem. Vol. 75, No. 24, 2010 8455

Jordan et al.
JOCArticle
λ
max (i.e., 300 or 320 nm), and an emission spectrum was recorded.
urea protons until at least 5 equiv of acid were present in the
tube. Association constants were calculated by using the non-
linear fitting program WinEQNMR.18
Aliquots of organic acid solutions, prepared in CH3CN to be up to
300 times more concentrated than that of the sensor solution, were
added to the cuvette and the contents were stirred for 1 min before
being scanned. This process was repeated until the fluorescence
spectra ceased to evolve. For sensor/acid combinations that dis-
played fluorescence enhancement, the integrated emission intensity
(F) was computed from the isoemissive point to 500 nm for each
spectrum. Data were normalized by dividing each F value by F0,
then were plotted as a function of [acid]total. Derivation of binding
constants varied depending upon the appearance of the normalized
response curves so obtained. Hyperbolic data sets were fit to the
Computational Modeling. Routine geometry optimizations
were run through the numerical DFT program DMol3 and
employed the Becke-Tsuneda-Hirao gradient-corrected exchange-
correlation functional and a double-numerical plus polariza-
tion basis set.10 NMR chemical shifts and excitation/emission
energies were calculated by using the GIAO and time-dependent
density-functional theory (TDDFT) methods, respectively, as
implemented in Gaussian 09. In both cases, molecular geome-
tries were optimized at the same level of theory in which the
properties were evaluated: B3LYP/6-31G** and B3LYP/tzvp
for the NMR and TDDFT calculations, respectively, performed
in the presence of a CPCM reaction field with the dielectric equal
to that of acetonitrile.
equation F/F0 = {1 þ (kcomplex/ksensor)Kassoc[acid]total}/(1 þ Kassoc
-
[acid]total), as previously described.12a Sigmoidal curves were fit to F/
F0 = 1 þ {{(Fmax/F0) - 1}/{1 þ (Kassoc[acid]total)p}}, where Fmax is
the integrated intensity at highest acid concentration, p is a slope
parameter, and [acid]total is placed on a logarithmic scale.12b During
iterative fitting, values of Kassoc and p were allowed to vary freely.
For cases where weak binding was apparent (little or no curvature in
binding isotherms), values were obtained by linear fitting to F/F0 =
Acknowledgment. W.E.A. thanks the ECU Division of
Research and Graduate Studies for a Research Development
Award. A.L.S. thanks the ECU CACS and NSF (CNS-
0619285). L.M.J. was supported by a Burroughs-Wellcome
Fellowship for 2008-2009. Purchase of NMR and mass
spectrometers was made possible by the NSF (CRIF-
0077988 and MRI-0521228, respectively).
K
assoc[CHCl2COOH]total. For acetic acid or dichloroacetic acid
acting as a simple quencher, Stern-Volmer constants were calcu-
lated from the best fit line to the equation F0/F = 1 þ KSV[acid]total
.
1H NMR Titrations. With sonication and gentle heating,
sensors were dissolved in CD3CN to concentrations of 0.0056-
0.0060 M. A 1.00 mL sample was transferred to an NMR tube.
Aliquots of acid solutions, prepared to be ∼30ꢀ more concen-
trated than that of the host, were injected into the tube. After
each addition the tube was capped and inverted several times,
and a proton NMR spectrum was acquired. This process was
repeated while monitoring the chemical shift of the pyridyl and
1
Supporting Information Available: H/13C NMR spectra for
all new compounds, crystallographic details, representative
1
titration curves from fluorescence and H NMR, and atomic
coordinates of DFT-calculated structures. This material is
available free of charge via the Internet at http://pubs.acs.org.
8456 J. Org. Chem. Vol. 75, No. 24, 2010