Recognition and sensing of biologically relevant anions in alcohol and mixed alcohol-aqueous solutions using charge neutral cleft-like glycol-derived pyridyl-amidothiourea receptors

Article abstract of DOI:10.1021/jo202332h

In this paper, the synthesis and the spectroscopic investigation of new colorimetric receptors for anions 3-6, possessing a glycol chain at the 4-position of the pyridyl ring, and 1 and 2, which lack such a chain, and the X-ray crystal structure of 2 is presented. Structures 3-6 are able to bind to anions in competitive media, such as alcohol or in a mixture of methanol and water, where the anion recognition gives rise to changes in the absorption spectra, which is red-shifted, in 1:1 or 1:2 (sensor/anion) stoichiometry. The anion recognition for 1 and 2 was also investigated in organic solvents and in a 4:1 mixture of DMSO/H₂O. The binding of 1 to anions such as acetate, phosphate, and fluoride was also evaluated using ¹H NMR in DMSO-d₆.

Full text of DOI:10.1021/jo202332h

Article
pubs.acs.org/joc
Recognition and Sensing of Biologically Relevant Anions in Alcohol
and Mixed Alcohol−Aqueous Solutions Using Charge Neutral Cleft-
Like Glycol-Derived Pyridyl−Amidothiourea Receptors
Rebecca M. Duke, Thomas McCabe, Wolfgang Schmitt, and Thorfinnur Gunnlaugsson*
School of Chemistry, Center for Synthesis and Chemical Biology, Trinity College Dublin, Dublin 2, Ireland
S
* Supporting Information
ABSTRACT: In this paper, the synthesis and the spectroscopic investigation of new colorimetric receptors for anions 3−6,
possessing a glycol chain at the 4-position of the pyridyl ring, and 1 and 2, which lack such a chain, and the X-ray crystal structure
of 2 is presented. Structures 3−6 are able to bind to anions in competitive media, such as alcohol or in a mixture of methanol and
water, where the anion recognition gives rise to changes in the absorption spectra, which is red-shifted, in 1:1 or 1:2 (sensor/
anion) stoichiometry. The anion recognition for 1 and 2 was also investigated in organic solvents and in a 4:1 mixture of DMSO/
1
H2O. The binding of 1 to anions such as acetate, phosphate, and fluoride was also evaluated using H NMR in DMSO-d6.
lack such a solubilizing group, Figure 1. Compounds 1−6 were
also developed with the view of binding anions through a
cooperative effect, where the pyridyl unit preorganized the two
amidothiourea moieties into a cleftlike12a arrangement. When
we commenced this study, compound 2 was first formed and
studied, as we had experienced significant changes in the UV−
vis absorption spectra of related systems. Because of this, a
greater number of anions were tested for 2 than 1. We were
only able to study 2 in detail in DMSO and various other
organic solvents as well as in a 4:1 mixture of DMSO/H2O
solution, while compound 1 was studied predominantly in
DMSO and in a 4:1 mixture of DMSO/H2O solution, with the
view of evaluating the effect of the substituent on the
amidothiourea moiety. Hence, these compounds function as
model compounds in the present discussion. In this paper, we
show that the incorporation of a polyethylene chain into the
pyridyl moieties of 3 and 4 greatly increases their solubility in
protic solvents, while such modification also facilitates the
incorporation of two pyridyl clefts, 5 and 6, into a single
structure, which should enable the detection of anions such as
organic phosphates. We also demonstrate that the anion
recognition can either be through the mechanism of hydrogen
bonding or via deprotonation, both of which give rise to
changes in the absorption spectra of these compounds.
INTRODUCTION
■
The design and study of luminescent and colorimetric sensors
for anions has become a very active area of research within the
field of supramolecular chemistry.1,2 While anions can be
bound to positively charged receptors, the use of charge-neutral
anion recognition moieties to achieve anion binding has been
the focus of many publications in recent years.3−5 Most of these
examples have been designed for use in organic solution, where
the anion recognition process occurs via hydrogen bonding. In
contrast, only a few examples of such recognition/sensing using
hydrogen bonding in more competitive aqueous media have
been published.5g We have developed both colorimetric- and
luminescent-based sensors for anions using urea- and thiourea-
based sensors.5,6 Recently, we have also developed such
fluorescent sensors possessing amidourea or amidothiourea
functionalities as anion receptors,7 but these binding motifs
have been shown, in particular by Jiang et al. and Gale et al., to
give rise to strong interactions with anions such as acetate and
phosphates.8 We have also incorporated these into colorimetric
sensors based on structures such as 1,8-naphthalimides,1,9
calixarenes10 and various aryl groups and investigated their
ability to sense anions in organic solutions such as DMSO and
CH3CN.
With the aim of further exploring the use of these binding
motifs for anion sensing, particularly in competitive aqueous
media, we have now developed the pyridyl−dicarboxyl systems
3−6, which possess a poly(ethylene glycol) chain at the 4-
postions of the pyridyl ring, and compounds 1 and 2,11 which
Received: November 17, 2011
Published: February 23, 2012
© 2012 American Chemical Society
3115
dx.doi.org/10.1021/jo202332h | J. Org. Chem. 2012, 77, 3115−3126

The Journal of Organic Chemistry
Article
Figure 1. Structures 1−6 developed in this study.
a
Scheme 1
a
Key: (a) (i) N2H4·H2O, MeOH, rt; (ii) CH3CN, reflux, 16 h and either 4-trifluoromethylphenyl isothiocyanate or p-nitrophenyl isothiocyanate,
giving 1 and 2, respectively. (b) (iii) 1-Iodo-2-(2- methoxyethoxy)ethane, K2CO3, DMF, 80 °C, 16 h; (iv) N2H4·H2O, EtOH, reflux, 16 h; (v) 4-
trifluoromethylphenyl isothiocyanate or p-nitrophenyl isothiocyanate, MeCN, reflux, 16 h, giving 3 and 4, respectively; (vi) di(iodoethoxy)ethane,
K2CO3, DMF, 80 °C, 16 h; (vii) N2H4·H2O, EtOH, reflux, 16 h; (viii) 4-trifluoromethylphenyl isothiocyanate or p-nitrophenyl isothiocyanate,
MeCN, reflux, 16 h, giving 5 and 6, respectively.
1 and 2 showed the presence of two rotamers in solution. In
the case of 2, three broad resonances occurred at 11.30, 10.37,
and 10.20 ppm, respectively, for the three N−H protons of the
major isomer. These were assigned using various 2D NMR
experiments and variable temperature NMR. In contrast, when
the 1H NMR of 2 was recorded in DMF-d7 at room
temperature only a single rotamer was observed.
The synthesis of 3−6 is shown in Scheme 1b and is based on
the same synthetic strategy as for 1 and 2, beginning with 9, the
diethyl ester of the commercially available chelidamic acid
(formed by refluxing in EtOH in the presence of a catalytic
amount of H2SO4),13 which was placed in DMF in the presence
of 1 equiv of K2CO3 and stirred at room temperature for 15
RESULTS AND DISCUSSION
■
Synthesis of 1−6 and X-ray Crystallographic Analysis
of 2. The synthesis of 1 was achieved in a few linear steps as
shown in Scheme 1a, where the commercially available 2,6-
pyridinedicarboxylic acid was first converted to its correspond-
ing diethyl ester 7,12a followed by reacting it in refluxing MeOH
with hydrazine monohydrate. This gave 2,6-pyridinedicarbox-
ylic acid, dihydrazide 8,12b which upon reaction with 4-
trifluoromethylphenyl isothiocyanate in CH3CN gave 1 as a
white solid in 77% yield. Compound 2 was formed in a similar
manner using 4-nitrophenyl isothiocyanate in 98% yield as a
pale yellow solid. The 1H NMR (600 MHz, DMSO-d6) of both
3116
dx.doi.org/10.1021/jo202332h | J. Org. Chem. 2012, 77, 3115−3126

The Journal of Organic Chemistry
Article
Figure 2. (A) X-ray crystal structure of 2. Solvent molecules have been removed for clarity. (B) Packing of 2 when viewed down the crystallographic
c-axis. Any solvent molecules and hydrogen atoms have been omitted for clarity.
Figure 3. (A) Changes in the absorption spectra of 2 (1 × 10−5 M) upon addition of H2PO4− (0 → 5 equiv) in DMSO. Inset: Relative changes in
−
the absorbance at 425 nm against equivalents of H2PO4 . (B) Changes in the absorption spectrum of 1 (1.9 × 10−5 M) upon titration with AcO−
(0→9 equiv) in MeCN (with 1% DMSO). Inset: Changes at 360 nm.
min, after which 1-iodo-2-(2- methoxyethoxy)ethane14 was
added and the resulting solution heated at 80 °C for 16 h. After
acidic workup, 10 was isolated as a yellow oil in 67% yield.
Treatment of 10 with excess hydrazine in EtOH under reflux
for 16 h resulted in the formation of 11 as a precipitate in 51%
yield, which was reacted further under reflux with 2 equiv of
either 4-trifluoromethylphenyl isothiocyanate or p-nitrophenyl
isothiocyanate in MeCN, giving 3 and 4, respectively, as off-
white precipitates in 66% and 98% yields, respectively.
For the synthesis of 5 and 6, the diester 9 was reacted with
di(iodoethoxy)ethane in the presence of K2CO3 in DMF at 80
°C to give 12 as a yellow oil in 73% yield after recrystallization
from EtOH. This intermediate was further reacted with
hydrazine in EtOH to give the tetra hydrazide 13 in 81%
yield as a white precipitate, which upon reaction with 4 equiv of
4-trifluoromethylphenyl isothiocyanate and p-nitrophenyl iso-
thiocyanate gave 5 and 6 as pale yellow solids in 68% and 42%
yields, respectively.
ray crystallographic diffraction analysis.15 The resulting
structure, was, however, not of good enough quality for
publication, but we were able to demonstrate the correct
connectivity from the data obtained. We were, however, also
able to grow crystals of 2 suitable for single-crystal X-ray
crystallographic analysis15 from a DMSO solution, the structure
of which is shown in Figure 2A. The combination of H-bonding
and π−π stacking also gave rise to an elaborate 3D network that
was perforated with ordered DMSO solvent-filled channels,
Figure 2B. Here, the unit cell contained two molecules of 2 and
six DMSO molecules (omitted from Figure 2B), all of which
were hydrogen bonding directly to the N−H protons within
the amidothiourea receptor. It was also clear, from these
structures that the thiourea protons of the receptor moieties in
both 2 and 6 occupy an anti conformation, but we had
previously observed such geometry in the crystal structures of
aminourea-based colorimetric sensors.5g
Spectroscopic Studies of Model Compounds 1 and 2
in Organic Solutions. As detailed above, the objective of the
current study was to develop water-soluble anion receptors and
test their anion recognition ability in competitive media, and
compounds 1 and 2 were chosen as model compounds for this
study, as both lack the water solubilizing groups introduced into
3−6. Compound 2, which contains a nitro substituent on each
of the aryl part of the two amide thiourea moieties, should
enable colorimetric changes, which should be visible to the
naked eye upon anion recognition. Several research groups16−21
have investigated the nature of such recognition processes in
organic polar aprotic solvents such as DMSO, CHCl3, or
The compounds were fully characterized (see the Exper-
1
imental Section). As an example, the H NMR (see Figures
S1−S6, Supporting Information, 400 MHz, DMSO-d6 for 1−6)
of 3 showed the presence of the three sets of N−H protons,
resonating at 11.25, 10.19, and 10.04 ppm for the amido and
the two thiourea protons, respectively, while using HR MALDI
MS gave an m/z peak for the [M + H]+ ion at 720.1533.
Recrystallization of sensors 3−6, using a variety of solvents,
generally resulted in the formation of poor quality crystals.
However, in the case of 6, crystals grown from a solution of
EtOH−DMSO were found to be of satisfactory quality for X-
3117
dx.doi.org/10.1021/jo202332h | J. Org. Chem. 2012, 77, 3115−3126

The Journal of Organic Chemistry
Article
CH3CN for hydrogen-bonding receptors and various anions.
We have shown that, in general, such binding interactions are
dominated by hydrogen-bonding interactions. However, we
have also shown that anion-induced deprotonations can occur,
−
particularly for anions such as F− and H2PO4 in such
solvents.5,6f,9,21 We have also demonstrated that it can be
difficult to distinguish between these two processes using either
UV−vis or fluorescence spectroscopic titrations, as the spectral
changes are often similar and, in some cases, the same for both
mechanisms. As such, various 1D and 2D 1H NMR techniques
have to be employed to elucidate the mechanism for such
anion−host interactions. However, the concentration difference
employed for these techniques differs greatly, which is an
additional disadvantage. Hence, one of the objectives of
developing 1 and 2 was also to attempt to distinguish between
these two events as these two systems possess different electron
withdrawing groups which are known to affect the acidity of the
hydrogen donating receptors. Anion receptors possessing nitro
groups can in particular give rise to deprotonation events.
Therefore, compound 1, possessing 4-trifluoromethylphenyl
moiety, which at the same time as being highly electron
withdrawing is also known to be less likely to induce
deprotonation events for such amidothioureas, was also
synthesized. Of these two systems, the discussion below will
focus mostly on the results obtained from the titrations of 2
with various anions.
Figure 4. Relative changes in the absorption spectra of 2 (1 × 10−5 M)
at 425 nm upon addition of various anions in DMSO.
the sensor, as a plateau was observed after the addition of ca. 2
equiv of each of these anions. Conversely, it was envisaged that
3−
ditopic anions such as HP2O7 may bridge both arms of the
sensor, thus giving rise to the observed 1:1 stoichiometry;
3−
hence, we next evaluated the ability of 1 to bind HP2O7
.
Again, similar changes were observed in the absorption spectra
of 1 as seen in Figure 3A upon binding to HP2O73−, and
analysis of the spectral changes indeed demonstrated that the
anion was bound in a 1:1 stoichiometry. Importantly the
spectral changes were not reversed upon the addition of
competitive hydrogen bonding solvents such as MeOH or
H2O, which strongly indicated that the anions could be
detected in a mixture of organic−aqueous solution.
The absorption spectra of 2 in DMSO (1 μM) showed a
broad absorption band centered at ca. 348 nm (ε = 21270 M−1
cm−1), which we assigned to the internal charge transfer (ICT)
character of the molecules, and a shoulder at shorter
wavelength at ca. 290 nm, assigned to the pyridyl unit and
higher energy bands associated with the nitrobenzene
chromophore, Figure 3A. Upon excitation of 2 at the 348 nm
band, the compound was found to be only slightly emissive,
and hence, the following investigation only focuses on the
anion-induced changes observed in the ground state of 2. The
compound was titrated with various anions (as their
In the case of 1, the titrations were carried out in DMSO/
CH3CN solution. Because of the nature of the receptor moiety,
the absorption band occurred at higher energy in comparison
to that seen for 2, with a λmax at 276 nm. However, upon
titrations with various anions such as AcO−, Figure 3B, the
formation of a red-shifted transition was observed with a λmax at
360 nm, and as in the case of 2, the binding was found to be in
a 1:2 (Host/Guest) stoichiometry (shown as inset in Figure
3B). However, analysis of the changes observed in the
absorption spectra of both 1 and 2 by fitting the changes
using the nonlinear regression analysis program SPECFIT, did
not result in an accurate determination of binding constants for
these changes, as these were all determined with large errors.22
Because of this we investigated the binding phenomenon using
NMR titrations.
tetrabutylammonium, TBA, salts) such as AcO− and H2PO4 ,
−
HP2O73−, and F−. For these titrations, upon increasing the
anion concentration the recognition of the anions induced a
bathochromic shift. In the case of phosphate, Figure 3A, a new
band was formed at longer wavelength, centered at 420 nm,
tailing into 560 nm, with the formation of an isosbestic point at
353 nm. These spectral changes demonstrated that the anion is
bound within the receptor cleft, which is in conjugation to the
p-nitrophenyl chromophore, and this binding enhances the
strength of the ICT character with a concomitant shift in the
absorption spectra to longer wavelengths. Importantly, these
changes were also, even at such low concentration of the
sensor, visible to the naked eye, changing from colorless to
yellow. Similar changes were also observed in the absorption
spectra of 1 using AcO− (Figure 3B) and F−, while only minor
changes were observed in the absorption spectra using Cl−, and
no changes were observed using Br−.
1H NMR Titration of 1 and 2 in DMSO-d6. The above
titrations showed that 1 and 2 could recognize various anions in
1:1 or 1:2 stoichiometries. While the changes in the absorption
spectra are indicative of hydrogen bonding by these anions, it is
not conclusive. Consequently, to shed light on the mechanism
1
of action here, we carried out H NMR titrations in DMSO-d6
on 1 and 2 using the same anions as employed above. The
1
changes in the H NMR spectra of 2 upon titration with
−
H2PO4 in DMSO-d6 (see the Supporting Information) gave
an indication of the general trend observed for these titrations.
−
The first addition of H2PO4 (0.2 equivalents, not shown)
resulted in the broadening of all the proton resonances. The
intensity of the N−H amide proton, 11.30 ppm, decreased, and
a new broad resonance was also observed slightly further
downfield at ca. 11.8 ppm. Both thiourea protons also
broadened before completely disappearing after the addition
of 2 equiv of anion, along with a new signal at 11.8 ppm. At 1
Analysis of the anion-induced changes in both 1 and 2
demonstrated that each of the amidothioureas was able to bind
to a single anion, giving rise to an overall 1:2 binding
stoichiometry, as shown as an inset in Figure 3A,B for the
titration with H2PO4 and AcO−, respectively. The relative
−
−
changes in the absorption spectra of 2 at 425 nm are plotted
equiv of H2PO4 a new signal also appeared at ca. 8.9 ppm,
against the added equivalents of these anions, Figure 4. These
which gradually sharpened, along with the amide N−H signal,
as the titration progressed. The resulting sharp resonances at
11.36 and 8.95 ppm were identified as N−H protons by
−
results indicate that F− and AcO− may, as did H2PO4 , form
hydrogen-bonded complexes with each amidothiourea arm of
3118
dx.doi.org/10.1021/jo202332h | J. Org. Chem. 2012, 77, 3115−3126

The Journal of Organic Chemistry
Article
Figure 5. (A) Stack plot of 2 (top) and resulting spectra upon addition of various equivalents of anions in (600 MHz, DMSO-d6). (B) Stack plot of
1
changes in the H NMR (400 MHz, DMSO-d6) spectra of 1 upon addition of various anions.
Figure 6. (A) Changes in the absorption spectra of 2 upon titration with AcO− in 4:1 DMSO/H2O solution. (B) Plot of the changes in the
absorption of 2 upon addition of various phosphate-based anions in 4:1 DMSO/H2O solution.
15N−1H COSY NMR, and both signals integrated for two
as the likely mechanism in organic solvents. With this in mind
we attempted to investigate the binding affinity of 1 and 2 in
aqueous solution, which would suppress any such deprotona-
tion mechanism.
protons each. In the case of the aromatic protons, the signals
for the pyridine protons, as well as the p-nitrophenyl protons at
ca. 8.22 ppm, were shifted upfield, while the p-nitrophenyl
signal at 7.90 ppm was shifted downfield. Overall, these types of
changes were indicative of deprotonation, as only four N−H
protons were identified in the final spectrum, which might
explain the difficulties encountered above when fitting the
changes in the absorption spectra. Additionally, similar
deprotonation patterns with pyrrole−amidothioureas have
also been reported by Gale et al.18 Figure 5A shows that
titration of 2 with AcO−, F−, AMP−, HP2O73−, and OH− (all as
TBA salts) also gave rise to a spectrum similar to that seen for
Spectroscopic Studies of Model Compounds 1 and 2
in 4:1 DMSO/H2O Solution. Because of the poor solubility of
1 and 2 in water or in alcoholic solutions, we were unable to
carry out UV−vis anion titrations on these sensors in water or
alcohol/aqueous solutions. Nevertheless, as 2 had been soluble
in DMSO, we consequently carried out a preliminary
investigation of 2 in an aqueous solution of DMSO where a
4:1 DMSO/H2O mixture was found to give the best solubility
for 2, and this was also employed for 1.
−
The changes observed in the absorption spectra of 2 were
similar to those observed above in DMSO solution where the
spectra were shifted to longer wavelengths upon interaction
H2PO4 above. Figure 5 also shows the number of equivalents
of each anion that were required to produce the spectra above.
−
Apart from H2PO4 , these results are in close agreement with
−
with anions such as AcO−, Figure 6A, H2PO4 or F−. Once
the number of equivalents required to reach a plateau in the
absorption titrations in DMSO, shown previously in Figure 4,
more, no clear isosbestic point was observed for these changes.
However a pseudo-isosbestic point was seen at ca. 340 nm in
Figure 6A. However, on all occasions the number of equivalents
required to reach a plateau (ca. 30 equiv) was higher than that
required in DMSO (ca. 2 equiv). In the case of hydrogen
bonding, this would be due to the competitive hydrogen-
bonding nature of water. However, if deprotonation occurred,
the higher concentration required would be due to the lower
1
for 2. In the case of 2, we also carried out H NMR titrations
using AMP, which showed that a large excess of AMP was
required to reach a plateau in comparison to the other anions.
Hence, in DMSO solution, the recognition of these anions is
most likely by hydrogen bonding between the receptor and the
anions, which then at higher concentrations, gives rise to a full
deprotonation event. In a similar manner the 1H NMR of 1 was
−
−
also recorded and the results in the presence of AcO−, H2PO4 ,
bascitiy of ions such as H2PO4 , compared to hydroxide, which
and F−, Figure 5B, again indicted that the final species had the
same common structure, and hence, supporting deprotonation
was also shown to give rise to a red shift in the absorption
spectra of 2. In a similar manner, the changes in the absorption
3119
dx.doi.org/10.1021/jo202332h | J. Org. Chem. 2012, 77, 3115−3126

The Journal of Organic Chemistry
Article
Figure 7. (A) Changes in the UV−vis absorption spectra of 3 in MeOH upon titration with acetate (TBAAc). (B) Species distribution diagram upon
fitting the titration of 3 with AcO− in MeOH to a 1:1 and 1:2 H/G binding model.
spectra of 1 was also affected upon binding to anions such as
AcO−, where a new transition was formed at 357 nm and a
concomitant color change from colorless to yellow was
observed. This clearly demonstrates that this receptor was
also able to bind anions in this competitive media, at the same
time as demonstrating that the nature of the electron-
withdrawing group has significant effect on the changes in
the absorption spectra.
occur, which is not desirable for sensing of biologically
important anions. However, the above measurements using in
particular 1 seem to indicate that this occurred to a lesser extent
than using the nitro substituent 2. Consequently, we made
sensors 3−6 with the view of improving their use in more
aqueous environments. However, judging from the above
binding trends and affinities, we also anticipated that the
binding of these modified versions would be lower in this
competitive media.
Unlike that seen for the DMSO, analysis of these changes
using the nonlinear regression analysis program SPECFIT did
result in good fits to the experimental data of 2 (with the
exception of F−) from which binding constants and speciation
distribution diagrams were determined. In the case of anions
Spectroscopic Studies of 3 and 4 in Alcoholic
Solutions. Having been unable to fully study the sensing
ability of 1 and 2 in competitive aqueous media, we embarked
on the synthesis of a more “aqueous friendly” versions, namely
3−6, the synthesis of which was described above. Because of
the improved solubility of these compounds the anion-binding
ability of 3 and 4 was first evaluated in MeOH and EtOH which
are highly competitive medias for such hydrogen-bonding anion
studies. Most of the discussion herein, however, focuses on the
use of 3, having the CF3 substituent, as the results above
suggested that the NO2 substituent (i.e., compound 4) would
be more likely to give rise to deprotonation of the
amidouthiourea protons due to a strong electron-withdrawing
effect, as had been seen for 2 above. However, the use of the
CF3 group also has its drawbacks as the absorption spectra is
blue-shifted in comparison to 2, as we saw for 1 above. The
absorption spectrum of 3 (ε = 26200 M−1 cm−1, 275 nm) when
recorded in MeOH is shown in Figure 7A and shows several
transitions centered at 210, 240, 275, and 350 nm. While the
transition centered at 350 nm was assigned to the ICT of the p-
CF3-phenyl chromophore, the band at ca. 275 nm was assigned
to the n−π* transitions of the pyridine moiety, and the band at
240 nm to the π−π* transitions of both the pyridine and the p-
CF3-phenyl moieties. Figure 7A also shows the changes
observed in the absorption spectra upon addition of acetate
at 350 nm. Here, an increase in the ICT absorption band with a
small concomitant decrease in the band centered at 240 nm was
observed. In contrast, no changes were observed in the pyridine
transition (275 nm) due to the interaction of AcO− with the
amidothiourea receptor moieties of 3. Similar effects were also
seen in EtOH solution when 3 was titrated with AcO−.
However, the changes were ca. 15% larger for the ICT band.
The changes in the absorption spectrum of 3 in MeOH at 350
nm as a function of the equivalents added are shown as an
insert in Figure 7A. As for 1 and 2, these binding interactions
were analyzed using the nonliner regression analysis program
SPECFIT, which showed that the data was best fit (see solid
line as inset in Figure 7A) to a 1:1 and then at higher
−
such as H2PO4 , the changes in the absorption spectra were
best fitted to both 1:1 and 1:2 Host/Guest stoichiometry, with
binding constants of log K1:1 = 4.24 ( 0.06) and log K1:2 = 4.07
−
( 0.04) determined for H2PO4 . These values are close in
magnitude, as expected for separated binding sites, one on each
arm of the sensor. Similarly, high binding constants were
determined for AcO−, with log K1:1 = 5.73 ( 0.06) and log K1:2
= 5.57 ( 0.04).
Having demonstrated that 2 was able to bind these
structurally simple anions, we next evaluated its ability further
for the sensing of various biologically relevant phosphates, such
as pyrophosphate, AMP, ADP, and ATP. Generally, with the
exception of ATP, where no changes were observed in the
absorption spectra, these titrations also gave rise to the
formation of a new transition in the absorption spectra at
long wavelengths with accompanied color changes, which were
visible to the naked eye. Moreover, the spectral changes
indicated a common mechanism of interaction of 2 with these
anions. In the case of pyrophosphate, the spectral changes
indicated a strong 1:1 binding interaction, which was also found
to be the case for anions such as AMP and ADP. These results
are summarized in Figure 6B as normalized changes in
absorption vs anion equivalents. As before, the binding
constants were determined for these interactions using
nonlinear regression analysis. However, the best results were
3−
observed for HP2O7 which gave a high binding constant for
the 1:1 binding, with log K1:1 = 6.53 ( 0.3). The affinity of 1
for ADP was also high, but here the sensor was found to bind
the HP2O73− in both 1:1 and 1:2 stoichiometries, with log K1:1
= 3.62 and log K1:2 = 3.8, respectively.
While these results clearly indicate that sensors 1 and 2 were
able to bind anions in highly competitive media, then the use of
DMSO limits their use for practical purposes. Moreover, this
high DMSO content can also favor a deprotonation event to
3120
dx.doi.org/10.1021/jo202332h | J. Org. Chem. 2012, 77, 3115−3126

The Journal of Organic Chemistry
Article
Figure 8. (A) Changes in the absorption spectrum of 3 (2.44 × 10−5 M) upon titration with F− (0→16 equiv) in MeOH. (B) Changes in the
absorption spectra of 3 upon titration with OH−. Inset: Relative changes at 350 nm upon titration with AcO− and OH− as a function of equivalents.
Figure 9. (A) Overall changes in the absorption spectra of 5 (1.55 × 10−5 M) in EtOH upon titration with H2PO4− (0 → 32 equiv). Inset: Changes
at 360 nm with 1:1 fit (0 → 20 equiv) obtained from SPECFIT. (B) Changes in the absorption spectra of 6 (1.1 × 10−5 M) upon titration with
AcO− in MeOH. Inset: Changes at 380 nm with 1:2 H/G fit obtained from SPECFIT.
concentrations the 1:2 (Host/Guest) complex formation
occurred, indicating that each of the amidothiourea moieties
bound to a single anion for each of the pyridyl units. This fit
resulted in binding constants of log K1:1 = 4.9 ( 0.1) and log
K1:2 = 3.0 ( 0.4). This is the expected binding mode for such
simple linear “Y”-shaped anions. In EtOH, the changes were
also best fitted to a 1:2 (Host/Guest) stoichiometry, which
gave a large 1:1 binding constant of log K1:1 = 6.0 ( 0.1). The
1:2 species, log K1:2 = 2.8 ( 0.3), was also found to be a minor
species throughout the titration; not reaching more than 30%
formation after the addition of 60 equiv of AcO−, thereby
demonstrating the effect the media has on recognition, Figure
7B.
the formation of an isosbestic point at ca. 266 nm, after the
addition of ca. 6 equiv of OH−.
Whereas large changes were seen for 2, possessing the nitro
substituent, only minor changes were seen in the absorption
spectra of 4 (ε = 36950 M−1 cm−1 at 339 nm) when similar
anion titrations were carried out in this media (see Figures S10
and S11, Supporting Information). This clearly suggests that
the chain on the 4-position of the pyridyl unit does influence
the sensing ability of these structures, as well as the nature of
the substituent and the media; and while this could have been
anticipated, the difference observed here is significantly greater
than we had foreseen. Nevertheless, we were able to determine
the binding constants for the binding of 4 to these anions. The
changes in the presence of AcO− and F− were fitted best to 1:1
and to 1:2 stoichiometry which gave binding constants of log
K1:1 = 5.60 ( 0.07) and log K1:2 = 3.6 ( 0.1) for AcO− and log
K1:1 = 4.8 ( 0.1) and log K1:2 = 3.6 ( 0.2) for F−. In
comparison to the fitting of 3 with AcO−, the species
distribution diagrams for 4 upon titration with both AcO−
and F− showed a larger contribution from the 1:2 species (see
Figures S12 and S13, Supporting Information). This could be
due to the presence of the nitro groups, which, as discussed
above, makes the N−H protons more acidic and, hence, better
hydrogen bond donors, or conversely, easier to deprotonate.
Spectroscopic Studies of 5 and 6 in Alcoholic
Solutions. In a similar manner, 5 and 6 were also titrated
using various anions in MeOH and in EtOH solutions.26
Similar changes were observed for 5 as were seen for 3 for
AcO− and F−. However, the analyses of changes induced by
Because of the high binding affinity observed for AcO−,
sensor 3 was also titrated with the TBA salts of F− and H2PO4
−
in MeOH. Changes similar to those observed for AcO−
occurred in the presence of F−; however, these changes were
smaller in magnitude (ca. 50%) and did not reach a plateau
until after the addition of ca. 5 equiv of the anion, Figure 8A.
The reason for this is most likely the high solvation energy
associated with F− in this media, which will make it very
difficult to overcome.
−
Surprisingly, the titration of 3 with H2PO4 resulted in a
minor increase in the ICT transition, which after the addition of
5 equiv began to decrease in absorption with a concomitant
minor bathochromic shift. Unfortunately, the changes in the
presence of F− and H2PO4− could not be accurately fitted using
nonlinear regression analysis. It is likely that this could possibly
be due to some contribution from a deprotonation event which
can give rise to “false-positive” responses, as we have discussed
above for 111,19 and 2. The occurrence of a possible
deprotonation of the sensor 3 by these anions was thus
investigated by using TBAOH in MeOH. The changes in the
absorption spectrum of 3, Figure 8B, were considerably larger
than those seen upon titration of 3 with AcO−, with significant
changes occurring at both the 240 and 350 nm transitions, with
−
H2PO4 , Figure 9, showed that only the 1:1 binding (log K1:1=
4.55 ( 0.03)) gave rise to the main changes in the absorption
spectra, with minor changes occurring up to the addition of two
equivalents (Figure S14, Supporting Information), as seen in
the Figure 9 inset. However, the largest changes occurred up to
the addition of 1 equiv of H2PO4− and reached a plateau at ca. 2
equiv. Sensor 5 was also titrated with AcO− in MeOH, which
3121
dx.doi.org/10.1021/jo202332h | J. Org. Chem. 2012, 77, 3115−3126

The Journal of Organic Chemistry
Article
−
Figure 10. (A) Overall changes in the absorption spectra of 3 (1.65 × 10−5 M) upon titration with H2PO4 (0 → 8.5 equiv) in 1:1 EtOH−H2O.
Inset: Changes at 350 nm with a 1:1 fit obtained from SPECFIT. (B) Species distribution diagram obtained from the fitting of the titration of 3 with
−
H2PO4 in 1:1 EtOH−H2O to a 1:1 binding stoichiometry.
gave rise to similar changes to those seen in the absorption
spectra of 1; these changes were best fitted to both 1:1 and to
1:2 stoichiometry using SPECFIT, which gave binding
constants of log K1:1 = 5.00 ( 0.04) and log K1:2= 3.1
( 0.1). The log K1:1 value was similar to that seen for 1 upon
titration with AcO− in MeOH (log K1:1= 4.86 ( 0.04)). Sensor
changes could not be fitted using SPECFIT, the evolving factor
analysis (EFA) estimated the presence of three colored species
in solution, including sensor 6. The other two species most
likely resulted from the deprotonation of the receptor moieties
of 6. In support of the latter explanation, the titration of 6 with
TBAOH resulted in large changes in the absorption spectrum,
the magnitude of which were similar to those seen in the
−
5 was also titrated with TBAOH and HSO4 in MeOH, which
24,25
3−
resulted in a large increase and decrease in the ICT transition of
the absorption spectra. This is most likely due to deprotonation
and protonation of the receptor moieties by these anions,
respectively.23
presence of HP2O7 .
Spectroscopic Studies of 3 and 4 in 1:1 Alcoholic/H2O
Solutions. Having investigated the changes of 3−6 in alcoholic
solutions, we next investigated these in a more competitive 1:1
EtOH:/H2O mixture. The absorption spectrum of 3 (ε =
25950 M−1 cm−1, 280 nm) in 1:1 EtOH−H2O solution was
similar to that observed in EtOH. As before, the addition of
Spectroscopic changes similar to to that seen for 1 were
observed upon titration of 6 with AcO−, but here the AcO− and
F− titrations were also best fitted to a 1:1 stoichiometry model,
which resulted in binding constants of log K1:1= 6.3 ( 0.3) and
log K1:1= 5.59 ( 0.2), respectively. This log K1:1 value was close
in magnitude to that determined for 1 with AcO− in EtOH.
Sensor 6 was also titrated with AMP, ADP, and ATP in EtOH.
In the case of AMP, slightly larger changes were seen in the
ICT transition of the absorption spectrum compared to those
anions such as AcO−, H2PO4 , and F− to this solution resulted
−
in a modulation of the ICT transition of the sensor to 3 (see
Figures S15 and S16, Supporting Information) in this
competitive media, resulted in a modulation of the ICT
transition of the sensor, which was found to be most
−
pronounced for the titration with H2PO4 , Figure 10A. The
seen for AcO−, H2PO4 , and F−. However, these changes did
insert in Figure 10A shows that 2 equiv of the anion were
necessary to reach a plateau; however, these changes were best
fitted to a 1:1 binding stoichiometry using SPECFIT, and in
this case, no 1:2 species was observed, Figure 10B. This clearly
demonstrated the sensing ability of these structures, even in
highly competitive media. Unlike that seen above, these
changes were best fitted to a 1:1 binding stoichiometry, with
a log K1:1 = 5.12 ( 0.09), indicating a strong binding affinity for
−
not reach a plateau until after the addition of ca. 80 equiv of
AMP, at which point the absorption of the AMP began to
overlap with the changes in the ICT transition of 6. The
changes in the presence of AMP were fitted using SPECFIT,
from which the best fit was for a 1:1 stoichiometry, giving log
K1:1 = 3.67 ( 0.03) for the binding of AMP. In comparison to
the changes in the presence of AMP, only minor changes
occurred in the presence of ADP, and titration with ATP did
not result in a decrease in the absorption of the ICT transition
as had been seen for the previous sensors. In the case of
hydrogen bonding, the absence of any significant spectral
changes suggested that ADP may be too large to interact in a
favorable manner with the sensor. This may indicate that a
cavity or “binding pocket” was formed by the sensor in
solution, such as that seen in the crystal structure of 6. The
increase in the absorption of the ICT transition of 6 upon
titration with anions in EtOH indicated a common mechanism
of anion interaction to that seen for sensors 1 and 2. The
−
−
H2PO4 . Similarly 4 gave log K1:1 = 5.46 ( 0.07) for H2PO4 .
If this species is due to the formation of a hydrogen-bonded
complex between the sensor and this anion, it is possible that
the 1:1 binding stoichiometry was favored due to the formation
of a hydrophobic cleft within the structure of 3, as previously
postulated by Jiang et al. for monoamidothiourea sensors.8,16
For AcO−, 2 equiv of the anion were required to reach a plateau
for 3 and 4; nevertheless, these changes were best fitted to 1:1
binding with log K1:1 = 5.58 ( 0.05) for 4. Despite the more
competitive solvent system employed here, the binding
constant determined for AcO− was similar to that determined
in MeOH for 3. It is, however, important to point out that the
pH of the 1:1 solution changed by ca. 0.2−0.3 pH units upon
addition of 10 equiv of these anions, indicating that these
anions were recognized dominantly by 3 and 4 through
hydrogen-bonding interactions. In the case of F−, ca. 15 equiv
was required to reach a plateau in 3, while the titration of 4
with F− required ca. 4 equiv. As above, the pH of the solution
did not change during the titration (see Table S1, Supporting
3−
spectral changes in the presence of HP2O7 were similar to
those observed upon titration of 1 with HP2O73− and TBAOH
in 1:1 EtOH−H2O solution. These changes were previously
assigned to deprotonation of the receptor moieties of 1, as a
large increase in the pH of the solution was measured.
Therefore, it is quite likely that the overall changes in the
3−
absorption spectrum of 6 upon titration with HP2O7 were
also due to deprotonation of the N−H protons. Although these
3122
dx.doi.org/10.1021/jo202332h | J. Org. Chem. 2012, 77, 3115−3126

The Journal of Organic Chemistry
Article
Figure 11. (A) Changes in the long-wavelength ICT band in the UV−vis absorption spectra of 3 in 1:1 EtOH/H2O upon titration with AMP. Inset:
Binding isotherm and fitted 1:1 data. (B) Changes in the IC- based absorption band of 4 in the presence of pyrophosphate in the same solvent
system.
Information). From these results, it is possible that the strong
interaction with H2PO4 may be due to the fact that the
over ATP, where the selectivity is in favor of AMP over ADP.
In a similar manner, compound 4 was titrated with various
anions, and in general, these changes were less than seen for 3,
confirming the same trend as seen for the structurally simple
anions above.
−
tetrahedral anion could more easily bridge both arms of the
sensor, in comparison to the planar AcO− anion or the small
spherical F− anion (also more hydrated in this solvent system).
In contrast to these changes, no changes were seen in the
absorption spectrum of 3 upon titration with Cl−, Br−, and I−.
The absorption spectra of 5 and 6 in 1:1 EtOH−H2O (and
1:1 MeOH−H2O) solutions, respectively, were similar to their
−
Because of the dominant 1:1 binding of AcO− and H2PO4
−
corresponding spectra in alcohol. Titration of 5 with H2PO4
resulted in a different pattern of change to that observed for the
previous sensors and a decrease in the absorptions centered at
210, 240, and 280 nm was observed. However, upon the
in aqueous solution, we also evaluated the sensitivities of these
sensors toward biologically relevant phosphates such as
pyrophophate (HP2O73−), AMP, ADP, and ATP for both 3
and 4. The changes seen for the titration of 3 with AMP and
HP2O73− are shown in parts A and B of Figure 11, respectively,
and clearly show that the long wavelength absorption band
underwent significant changes upon binding of both of these
anions, which on both occasions was significantly greater in
magnitude than seen for the more structurally simpler anions.
Analysis of these changes using nonlinear regression analysis
clearly showed the initial formation of a 1:1 complex, but after
the addition of 15 equiv, both 1:1 and 1:2 complexes were
formed in equal amount; and in fact, the overall spectral
−
addition of 0.8 equivalents of H2PO4 , a white precipitate was
visible in solution. This indicated that the initial decreases in
the absorption spectrum were also as a result of precipitation of
these sensors from solution. As discussed above, the solubility
of sensors 5 and 6 in alcohol were lower than that of 3 and 4,
and while it was anticipated that an “anion binding pocket” may
be formed in solution by the tetraamidothiourea sensors, such
interactions may also result in an increase of stacking
interactions in solution and, hence, lower the solubility of the
compounds. As the studies carried out on the tetraamidothiour-
ea sensors in alcohol showed more or less similar spectral
responses to their corresponding bis analogues, it was decided
not to investigate the anion binding capabilities of 5 and 6 in
more competitive media any further.27
changes were best fitted to a 1:2 stoichiometry with log K1:1
=
5.58 ( 0.07) and log K1:2 3.64 ( 0.09) for these changes.
However, the pH of this solution, after the addition of just
three equiv. of HP2O73−, changed by ca. 0.7 pH units, which
indicates that a deprotonation of the sensor by the anion may
be contributing to these spectral changes as well.
CONCLUSION
■
Similarly, significant spectral changes were seen for the
titration of 3 using AMP, ADP, and ATP. For AMP (Figure
11A) and ADP the ICT band of 3 increased in absorption in a
similar manner to that seen above in Figure 7, and the changes
were similar in magnitude; AMP giving slightly greater changes
in the absorption at 350 nm. The changes were best fitted to a
1:1 stoichiometry which gave log K1:1 = 4.02 ( 0.12) and 3.63
( 0.12) for these anions, respectively, showing that AMP was
bound with slight preference over ADP. Importantly, in
comparison to that seen for pyrophosphate, only minor pH
changes were observed (ca. 0.2 pH units) for these titrations.
This suggests that the changes observed are more likely due to
the formation of hydrogen-bonded complexes between the
sensor and these anions, rather than the occurrence of
deprotonation. In contrast, the titration of ATP gave rise to a
decrease in the λmax, and the pH of the solution had changed to
pH 4.5 ( 0.1) at the end point. Hence, from these results it is
possible to state that 3 can selectively recognize AMP and ADP
In this paper, we present the synthesis and discuss the
spectroscopic results from six new anion receptors, 1−6, based
on cleftlike structures where the anion recognition occurs
through the synergetic action of two or more amiothiourea
moieties with various anions. We also demonstrated that
caution must to be paid to the elucidation of the mechanism for
such anion sensing; as hydrogen bonding and deprotonation
can both contribute to the overall anion sensing mechanism.
We show that the structures possessing glycol chains, 3−6,
enable us to evaluate their use for anion recognition in alcohol
or competitive aqueous/alcohol solutions, while compounds 1
and 2, which lack these chains, could only be studied in organic
solutions such as DMSO. The structural modification of 3−6
proved particularly successful for both 3 and 4, which were
titrated with various anions in both alcohol and alcohol−water
mixtures, while the larger tetra amido-thiourea systems 5 and 6
were only partially soluble in alcohol−water mixtures and,
hence, were only titrated in alcohol solutions. The anion-
3123
dx.doi.org/10.1021/jo202332h | J. Org. Chem. 2012, 77, 3115−3126

The Journal of Organic Chemistry
Article
binding capabilities of 3 and 4 were initially assessed in either
the presence of anions more strongly than 3 and 4 due to the
cooperative binding of anions by all four receptor moieties
within a “binding pocket”. The main species observed upon
or both MeOH and EtOH solution by titrating these
compounds with AcO−, H2PO4 , and F−, respectively. The
−
fitting the titrations of 5 with AcO−, H2PO4 , F−, and AMP was
−
results from these titrations showed that both sensors
interacted most strongly with AcO−, with large 1:1 binding
constants determined from nonliner regression analysis, using
SPECFIT. Additionally, the control titrations carried out with
base suggested that these changes could be due to the
formation of hydrogen bonded complexes, as the changes in the
presence of base were far larger in magnitude. In comparison to
the changes observed in the absorption spectra of both 3 and 4
with AcO− and F−, the spectral changes in the presence of
a 1:1 stoichiometric species, as had been seen for the bis-
systems earlier. In the case of AcO−, the binding constant with
5 was comparable to that observed for 3 in EtOH. Hence, these
binding constants did not indicate a cooperative effect for the
recognition of AcO−. Unfortunately, the solubility of these
compounds in 1:1 alcohol−water solutions was lower than
anticipated, making the preparation of such solutions difficult.
Because of solubility issues and the fact that the general trends
of the spectroscopic changes were similar to those observed for
the 3 and 4, the anion-sensing capabilities of these sensors were
not investigated in mixed alcohol−water solutions. We are
currently developing other analogues of 3 and 4 with the view
of further improving their water solubility and evaluating their
ability to transport phosphate-based anions between different
solvent systems and through bilayers.
−
H2PO4 were different, whereby an increase in the absorption
followed by a decrease and bathochromic shift occurred. This
indicated that the nature of the hydrogen-bonded complex
formed with the tetrahedral anion was quite different to that of
AcO− and F−. Working towards achieving anion binding in
highly competitive media, compounds 3 and 4 were then
titrated with the aforementioned anions in mixtures of alcohol
and water solutions. These titrations resulted, on all occasions,
in an increase in the ICT transition of both 3 and 4 in the
EXPERIMENTAL SECTION
■
presence of AcO−, H2PO4 and F−. In the case of 3, strong
−
Diethyl 4-(2-(2-Methoxyethoxy)ethoxy)pyridine-2,6-dicar-
boxylate (10). Compound 9 (0.300 g, 1.26 mmol) and anhydrous
K2CO3 (0.174 g, 1.26 mmol) were stirred in anhydrous DMF (9 mL)
at 25 °C for 0.5 h. 1-Iodo-2-(2- methoxyethoxy)ethane (0.264 g, 1.14
mmol) was added and the solution was heated at 50 o C for 50 h. The
solvent was then removed under reduced pressure, and DCM (20 mL)
was added to the residue. The organic layer was washed with 1% acetic
acid (5 mL) and H2O (5 mL). The organic layer was dried over
MgSO4, and the solvent was removed under reduced pressure to give
10 as a white solid (0.263 g, 67%): mp 200−203 °C dec; HRMS (ES+)
calcd for [M + H]+ C16H24NO7 342.1548, found 342.1553; δH (400
MHz, CDCl3) 7.82 (2H, s, CHpy), 4.48 (4H, q, J = 7.0 Hz, CH2ester),
4.33 (2H, t, J = 4.5 Hz, CH2), 3.93 (2H, t, J = 4.5 Hz, CH2), 3.73−3.72
(2H, m, CH2), 3.59 (2H, t, J = 4.5 Hz, CH2) 3.40 (3 H, s, OCH3), 1.47
(6 H, t, J = 7.0 Hz, CH3); δC (100 MHz, CDCl3) 166.7 (q), 162.5
(CO), 150.1 (q), 114.4 (CH), 71.9 (CH2), 70.9 (CH2), 69.1
(CH2), 68.2 (CH2), 62.4 (OCH2), 59.1 (OCH3), 14.2 (CH3).
4-(2-(2-Methoxyethoxy)ethoxy)pyridine-2,6-dicarbohydra-
zide (11). Compound 10 (0.263 g, 0.770 mmol) and hydrazine
monohydrate (0.197 g, 6.16 mmol, 0.19 mL) were stirred at reflux in
EtOH (40 mL) overnight. The resulting precipitate was isolated by
suction filtration and washed with EtOH (2 × 20 mL) to give 11 as a
white solid (0.094 g, 51%): mp 137−140 °C; HRMS (ES+) calcd for
C12H19N5O5Na [M + Na]+ 336.1284, found 336.1278; δH (400 MHz,
DMSO-d6) 10.60 (2H, br s, NHamide), 7.60 (2H, s, CHpy), 4.62 (4H, br
s, NH2), 4.32 (2H, t, J = 4.1 Hz, CH2), 3.77 (2H, t, J = 4.5 Hz CH2),
3.60−3.58 (2H, m, CH2), 3.46−3.43 (2H, m, CH2), 3.23 (3H, s,
OCH3); δc (400 MHz, DMSO-d6) 166.9 (q), 161.6 (CO), 150.5
(q), 109.6 (CH), 71.2 (CH2), 69.7 (CH2), 68.4 (CH2), 68.0 (CH2),
58.0 (CH3); IR νmax (cm−1) 3489, 3321, 3100, 2904, 1651, 1596, 1536,
1511, 1442, 1365, 1348, 1285, 1257, 1165, 1135, 1112, 1096, 1048,
996, 933, 880, 826, 746, 688.
−
recognition of H2PO4 was shown to occur in 1:1 EtOH-H2O
solution, as a high binding constant was determined by fitting
the overall changes in the absorption spectra. In order to better
understand the anion interactions of 3 and 4 in 1:1 EtOH−
H2O and 1:1 MeOH−H2O solutions, spectroscopic pH
titrations were carried out. The results from these titrations
gave a further understanding of these sensors in solution and
the effect that pH had on the changes in the absorption spectra.
These experiments showed that the pKa1 values of these sensors
were quite low (ca. 6 for both 3 and 4). These values initially
caused concern that the receptor moieties of 3 and 4 may be
deprotonated by anions too easily. However, despite the low
pKa1 determined, only minor changes in the pH (ca. 0.3) were
−
measured upon the addition of anions such as AcO−, H2PO4
and F− to mixed alcohol−water solutions of 3 and 4. These
results supported the formation of hydrogen-bonded complexes
in these solvent systems. Additionally, the changes observed in
these titrations were fitted to 1:1 stoichiometries, while those in
100% alcohol were fitted to both 1:1 and to 1:2 stoichiometries.
This may indicate that there was a minor contribution from
deprotonation in the less competitive alcohol solutions.
Compounds 3 and 4 were also titrated with “larger” biologically
important phosphate-based anions, such as AMP, ADP, ATP,
3−
and HP2O73−. The largest changes were seen for HP2O7
,
3−
where the overall changes in the presence of HP2O7 were
assigned to some degree of deprotonation for both 3 and 4
(with a change in the pH of the solution of 0.7). In contrast to
these changes, smaller spectral changes were seen in the ICT
transitions of 3 and 4 upon addition of AMP and ADP. These
changes gave rise to minor changes in the pH of ca. 0.2 pH
units and in the case of 3 were best fitted to a 1:1
stoichiometry. In contrast to these results, the addition of
ATP resulted in a decrease in the ICT transition of the
absorption spectra of both 3 and 4 as well as a substantial
decrease in the pH. Titrations of the tetraamidothiourea
sensors, 5 and 6 with various anions also gave rise to an
increase in the ICT band in the absorption spectra, in a similar
manner to that seen for 3 and 4 in EtOH and MeOH,
respectively. However, the changes in the presence of AcO−,
4,4′-((Ethoxyethoxy)ethyl)bis[diethyl ester-2,6-pyridine]
(12). Compound 9 (0.200 g, 0.836 mmol) and anhydrous K2CO3
(0.115 g, 0.836 mmol) were stirred in anhydrous DMF (60 mL) at 25
°C for 0.5 h. Di(iodoethoxy)ethane (0.154 g, 0.418 mmol) was added,
o
and the solution was heated at 50 C for 50 h. The solvent was then
removed under reduced pressure, and DCM (50 mL) was added to
the residue. The organic layer was washed with 1% acetic acid (25 mL)
followed by H2O (25 mL). The organic layer was dried with MgSO4,
and the solvent was removed under reduced pressure to give yellow
oil. The crude product was recrystallized from EtOH giving 12 as a
white solid (0.147 g, 41%): mp 75−76 °C; HRMS (ES+) calcd for
C28H36N2O12Na [M + Na]+ 615.2166, found 615.2169; δH (400 MHz,
CDCl3) 7.78 (4H, s, CHpy), 4.42 (8H, q, J = 7.0 Hz, CH2OCH3), 4.28
(4H, t, J = 4.5 Hz, CH2), 3.88 (4H, t, J = 4.4 Hz, CH2), 3.72 (4H, s,
CH2), 1.41 (12 H, t, J = 7.2 Hz, CH3); δC (100 MHz, CDCl3) 166.7
−
H2PO4 , and F− were smaller in magnitude when compared to
those seen for 3. It was hoped that these sensors could detect
3124
dx.doi.org/10.1021/jo202332h | J. Org. Chem. 2012, 77, 3115−3126

The Journal of Organic Chemistry
Article
(q), 164.6 (CO), 150.1 (q), 114.4 (CH), 70.9 (CH2), 69.2 (CH2),
68.2 (CH2), 62.4 (CH2), 14.1 (CH3); IR νmax (cm−1) 3076, 2983,
2933, 2905, 1742, 1706, 1594, 1567, 1478, 1435, 1369, 1371, 1342,
1328, 1244, 1230, 1158, 1134, 1102, 1067, 1023, 994, 938, 887, 866,
817, 786, 732, 701.
4,4′-(Ethoxyethoxy)ethyl)bis[dihydrazide-2,6-pyridine] (13).
Compound 12 (0.500 g, 0.84 mmol) and hydrazine monohydrate
(0.540 g, 16.86 mmol) were refluxed in anhydrous EtOH (300 mL)
for 2 days. The white precipitate was collected by suction filtration
giving 13 as a white solid (0.367 g, 81%): mp 200−203 °C dec;δH
(400 MHz, DMSO-d6) 10.59 (4H, br s, NHamide), 7.59 (4H, s, CHpy),
4.61 (8H, br s, NH2), 4.32 (4H, br s, CH2), 3.78 (4H, br s, CH2), 3.62
(4H, s, CH2); δC (100 MHz, DMSO-d6) 167.4 (q), 162.1 (CO),
150.9 (q), 110.1 (CH), 70.4 (CH2), 69.0 (CH2), 68.5 (CH2); δN (60
MHz, DMSO-d6) 129.6, 135.6; IR νmax (cm−1) 3304, 3078, 2929,
1654, 1587, 1520, 1434, 1374, 1352, 1322, 1244, 1170, 1124, 1109,
1062, 1050, 992, 946, 912, 889, 823, 747.
General Procedure for the Synthesis of Amidothioureas. A
solution of the isocyanate (2 equivalents) in CH3CN was added to a
solution of the dihydrazide (1 equiv) in CH3CN. The reaction mixture
was refluxed under argon for 16 h. The precipitate was then isolated by
suction filtration and washed several times with warm CH3CN. The
resulting solid was dried under vacuum.
2,6-Bis[4-trifluoromethylphenyl(thioureidocarbomoyl)]-
pyridine (1). Compound 1 was synthesized according to above
procedure using 8 (0.150 g, 0.769 mmol) and 4-trifluoromethylphenyl
isothiocyanate (0.312 g, 1.53 mol) in CH3CN (50 mL), yielding 1 as a
white solid (0.360 g, 77%): mp >300 °C; HRMS (ES−) calcd for
C23H16N7O2S2F6 [M − H]− 600.0711, found 600.0686; δH (400 MHz,
DMSO-d6) 11.30 (2H, br s, NHamide), 10.24 (2H, br s, NHurea), 10.06
(2H, br s, NHurea), 8.21−8.17 (3H, m, Hpy), 7.67−7.74 (8H, m, Ar-H);
δC (100 MHz, DMSO-d6), 181.1 (CS), 162.8 (CO), 147.7 (q),
142.9 (q), 139.6 (CH), 126.1 (CH), 125.5 (CH), 125.4 (CH), 124.2
for C28H27F6N7O5S2: C, 46.73 H, 3.78 N, 13.57. Found: C, 46.43; H,
3.69; N, 13.57.
4-(2-(2-Methoxyethoxy)ethoxy)-2,6-bis[4-nitrophenyl-
(thioureidocarbomoyl)]pyridine(4). Compound 4 was synthesized
according to the above procedure using 11 (0.113 g, 1.03 mmol) and
4-nitrophenyl isothiocyanate (0.518 g, 2.88 mmol) in CH3CN (20
mL). A pale yellow solid was obtained (0.784 g, 98%): mp 188−190
°C; HRMS (ES+) calcd for C26H27N9O9S2Na [M + Na]+ 696.1271,
found 696.1240; δH (400 MHz, DMSO-d6) 11.38−11.25 (2H, br m,
NHamide), 10.68−10.22 (4H, br m, NHurea), 8.20 (4H, br s, Ar-H), 7.90
(4H, br s, Ar-H), 7.75 (2H, br s, Hpy), 4.39 (2H, br s, CH2,), 3.80 (2H,
br s, CH2), 3.61 (2H, m, CH2), 3.46 (2H, m, CH2), 3.24 (3H, s,
OCH3); δC (100 MHz, DMSO-d6) 181.1 (CS), 167.6 (q), 163.1
(CO), 155.6 (q), 150.2 (q), 146.0 (q), 125.3 (CH), 124.0 (CH),
111.5 (CH), 71.7 (CH2), 70.2 (CH2), 68.9 (CH2), 68.8 (CH2), 58.5
(CH3); IR νmax (cm−1) 3215, 2933, 1689, 1596, 1548, 1505, 1411,
1336, 1271, 1211, 1146, 1109, 1041, 996, 928, 878, 851, 775, 746, 698.
Anal. Calcd for C26H27N9O9S2: C, 46.35; H, 4.04; N, 18.71. Found: C,
46.19; H, 3.93; N, 18.72.
4,4′-(Ethoxyethoxy)ethyl)-2,6-bis-[4-trifluoromethylphenyl-
(thioureidocarbomoyl)]pyridine] (5). Compound 5 was synthe-
sized according to the above procedure using compound 12 (0.059 g,
0.114 mmol) and 4-trifluoromethylphenyl isothiocyanate (0.089 g,
0.447 mmol) and refluxed in CH3CN (5 mL) for 20 h. The resulting
precipitate was collected and dried by suction filtration, giving 5 as a
white solid (0.101 g, 68%): mp >300 °C; HRMS (ES+) calcd for
C52H45N14O8F12S4 [M + H]+ 1349.2236, found 1349.2284; δH (400
MHz, DMSO-d6) 10.99 (4H, br s, NHamide), 10.27 (4H, br s, NHurea),
10.11 (4H, br s, NHurea), 7.67−7.55 (20H, br m, Ar-H), 4.37−4.22
(4H, br m, CH2), 3.82 (4H, br s, CH2), 3.69 (4H, br s, CH2); δC (100
MHz, DMSO-d6) 180.9 (CS), 167.1 (q), 162.7 (CO), 149.1 (q),
142.6 (q), 125.4 (CH), 124.7 (CH), 124.8 (CH), 124.4 (CF3, J13 9
=
F
C
(CF3,J13 9 = 273.1 Hz), 122.2 (q); δN (60 MHz, DMSO-d6) 129.3,
C
F
273.1 Hz), 110.6 (q), 70.3 (CH2), 68.9 (CH2), 68.3 (CH2); IR νmax
(cm−1) 3133, 2956, 1702 (CO), 1602, 1548, 1480, 1416, 1324
(CS), 1278, 1158, 1106, 1066, 1017, 997, 883, 838.
124.2, 126.6; δF (162 MHz DMSO-d6) −61.0; IR νmax (cm−1) 3368,
3134, 2962, 1705, 1617, 1601, 1541, 1463, 1415, 1328, 1278, 1217,
1156, 1105, 1065, 1016, 1000, 931, 886, 838, 786, 745, 725, 685. Anal.
Calcd for C23H17F6N7O2S2: C, 45.92; H, 2.85; N, 16.30. Found: C,
45.79; H, 2.88; N, 16.10.
4,4′-(Ethoxyethoxy)ethyl)-2,6-bis[4-nitrophenyl-
(thioureidocarbomoyl)]pyridine] (6). Compound 6 was synthe-
sized according to the above procedure using compound 12 (0.106 g,
0.19 mmol) and 4-nitrophenyl isothiocyanate (0.142 g, 0.79 mmol)
and refluxed in CH3CN (5 mL) for 20 h. The precipitate was collected
and dried by suction filtration, giving 6 as a pale yellow solid (0.215 g,
45%): mp 178−179 °C; HRMS (ES+) calcd for C48H45N18O16S4 [M +
H] 1257.2144, found 1257.2189; δH (400 MHz, DMSO-d6) 11.00
(4H, br s, NHamide), 10.35 (4H, br s, NHurea), 10.32 (4H, br s, NHurea),
8.18−7.44 (20H, br m, Ar-H), 4.37−4.25 (4H, br m, CH2), 3.83 (4H,
br s, CH2), 3.64 (4H, br s, CH2); δC (100 MHz, DMSO-d6) 181.0
(CS), 167.7 (q), 163.2 (CO), 149.5 (q), 145.5 (q), 143.8 (q),
124.7 (CH), 123.7 (CH), 123.6 (CH), 68.9 (CH2), 69.4 (CH2), 70.8
(CH2); IR νmax (cm−1) 3215, 2933, 1689, 1596, 1548, 1505, 1411,
1336, 1271, 1211, 1146, 1109, 1041, 996, 928, 878, 851, 746, 698.
1,6-Bis[4-nitrophenyl(thioureidocarbomoyl)]pyridine (2).
Compound 2 was synthesized according to the above procedure
using 8 (0.281 g, 1.03 mmol, 1 equiv) and 4-nitrophenyl
isothiocyanate (0.518 g, 2.88 mmol, 2 equiv) and then refluxed for
16 h. The precipitate was isolated by suction filtration and washed with
acetonitrile to give a pale yellow solid, 0.784 g, 98%: mp 203.8−205
°C; δH (600 MHz, DMSO-d6) 11.30 (2H, br s, NHNHurea), 10. 37
(2H, br s, NHurea), 10.20 (2H, br s, NHurea), 8.31 (2H, br s, CHpy),
8.30 (1H, br s, CHpy), 8.22 (4H, br s, CHCNO2), 7.92 (4H, br s, Hz,
CHCNH); δC (100 MHz DMSO-d6) 181.1 (CS), 162.8 (CO),
147.8 (C3), 145.7 (CNH), 143.6 (CNO2), 139.8 (C5), 125.6 (C4),
125.2 (CHCNH), 123.9 (CHCNO2); υN (600 MHz DMSO-d6) 129.6,
127.7, 126.2; IR νmax (cm−1) 3128, 2962, 1704, 1597, 1549, 1506,
1469, 1345, 1278, 1219, 1157, 1002, 887, 851, 745, 698; HRMS (ES+)
calcd for C21H18N9O6S2 556.0821, found 556.0822 (M + H).
4 - ( 2 - ( 2 - M e t h o x y e t h o x y ) e t h o x y ) - 2 , 6 - b i s - [ 4 -
trifluoromethylphenyl(thioureidocarbomoyl)]pyridine (3).
Compound 3 was synthesized according to the above procedure
using 11 (0.113 g, 0.360 mmol) and 4-trifluoromethylphenyl
isothiocyanate (0.144 g, 0.720 mmol) in CH3CN (20 mL). A white
solid was obtained (0.172 g, 66.4%): mp 200−212 °C dec; HRMS
(MALDI+) calcd for C28H28N7O5F6S2 [M + H]+ 720.1498, found
720.1533; δH (400 MHz, DMSO-d6) 11.25 (2H, br s, NHamide), 10.19
(2H, br s, NHurea), 10.04 (2H, br s, NHurea), 7.67−7.74 (10H, br m,
Ar-H), 4.39 (2H, br s, CH2), 3.80 (2H, br s, CH2), 3.60 (2H, m, CH2),
3.46 (2H, m, CH2), 3.24 (3H, s, OCH3); δC (100 MHz, DMSO-d6)
181.5 (CS), 167.4 (q), 162.9 (CO), 150.2 (q), 143.2 (q), 126.0
ASSOCIATED CONTENT
■
S
* Supporting Information
General experimental description, synthesis and character-
1
ization of all intermediates, H and 13C NMR spectra of all
intermediates and final products, and various figures and tables
(see text for references). This material is available free of charge
via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
(CH), 125.5 (CH), 125.3 (CH), 124.6 (CF3, J13 9 = 273.1 Hz), 122.5
C
F
*E-mail: gunnlaut@tcd.ie.
(q), 111.3 (CH), 71.5 (CH2), 70.5 (CH2), 68.8 (CH2), 58.3 (OCH3);
δF (376 MHz, DMSO-d6) −61.0 (CF3); IR νmax (cm−1) 3139, 2947,
1711, 1602, 1583, 1479, 1321, 1277, 1108, 1066, 892, 839. Anal. Calcd
Notes
The authors declare no competing financial interest.
3125
dx.doi.org/10.1021/jo202332h | J. Org. Chem. 2012, 77, 3115−3126

The Journal of Organic Chemistry
Article
(9) Ali, H. D. P.; Quinn, S. J.; McCabe, T.; Kruger, P. E.;
Gunnlaugsson, T. New J. Chem. 2009, 33, 793.
(10) (a) Quinlan, E.; Matthews, S. E.; Gunnlaugsson, T. J. Org. Chem.
2007, 72, 7497. (b) Quinlan, E.; Matthews, S. E.; Gunnlaugsson, T.
Tetrahedron Lett. 2006, 47, 9333.
(11) Duke, R. M.; O’Brien, J. E.; McCabe, T.; Gunnlaugsson, T. Org.
Biomol. Chem. 2008, 6, 4089.
(12) This cleft unit has also been used by: (a) Wei, L.-H.; He, Y.-B.;
Wu, J.-L.; Meng, L.-Z.; Yang, X. Supramol. Chem. 2004, 16, 561.
(b) Newton, A. J. Am. Chem. Soc. 1943, 65, 320.
ACKNOWLEDGMENTS
■
We thank Science Foundation Ireland for an SFI RFP 2009
grant and HEA-PRTLI Cycle 3 and Cycle 4 (CSCB) and TCD
for financial support. We thank Dr. John O’Brian for helping
with NMR studies and Dr. Emma B. Veale and Dr. Jonathan A.
Kitchen for helpful discussions.
REFERENCES
■
(13) Vandevyver, C. D. B.; Chauvin, A. S.; Comby, S.; Bunzli, J. C.
Chem. Commun. 2007, 1716.
(14) Arnaud, A.; Belleney, J.; Boue, F.; Bouteiller, L.; Carrot, G.;
Wintgens, V. Angew. Chem., Int. Ed. 2004, 43, 1718.
(1) (a) Gale, P. A; Gunnlaugsson, T. Chem Soc. Rev. 2010, 39, 3595.
(b) Gale, P. A. Chem Soc. Rev. 2010, 39, 3746. (c) Steed, J. W. Chem.
Soc. Rev. 2010, 39, 3686. (d) Hay, B. P. Chem. Soc. Rev. 2010, 39, 3700.
(e) Li, A.-F.; Wang, J.-H.; Wang, F.; Jiang, Y.-B. Chem. Soc. Rev. 2010,
39, 3729. (f) Amendola, V.; Fabbrizzi, L.; Mosca, L. Chem. Soc. Rev.
2010, 39, 3889. (g) Duke, R. M.; Veale, E. B.; Pfeffer, F. M.; Kruger, P.
E.; Gunnlaugsson, T. Chem. Soc. Rev. 2010, 39, 3936. (h) Veale, E. B.;
Gunnlaugsson, T. Annu. Rep. Prog. Chem., Sect. B: Org. Chem 2010,
106, 376−406.
(2) (a) Wenzel, M.; Hiscock, J. R.; Gale, P. A. Chem. Soc. Rev 2012,
41, 480. (b) Gale, P. A.; Garcia-Garrido, S. E.; Garric, J. Chem. Soc. Rev.
́
2008, 37, 151. (c) Gunnlaugsson, T.; Glynn, M.; Tocci (nee Hussey),
G. M.; Kruger, P. E.; Pfeffer, F. M. Coord. Chem. Rev. 2006, 250, 3094.
(d) Steed, J. W. Chem. Commun. 2006, 2637. (e) Sessler, J. L.; Gale, P.
A.; Cho, W. S. Anion Receptor Chemistry; Royal Society of Chemistry:
Cambridge, UK, 2006.
(15) Crystallographic data for 2 (CCDC 649247) M = 789.99,
triclinic, a = 14.489(3) Å, b = 15.929(5) Å, c = 17.154(5) Å, α =
87.377(7)°, β = 68.142(6)°, γ = 81.774(8)°, V = 3636.5(17) Å3, T =
123 K, space group P-1, Z = 4, μ(Mo Kα) = 0.381 mm−1, 36636
reflections collected, 12 589 unique, (Rint = 0.0430), R = 0.0815, wR2
[I > 2σ(I)] = 0.2394. CCDC deposition no. 649247 for 6.
(16) (a) Jiang, Q.-Q.; Darhkijav, B.; Liu, H.; Wang, F.; Li, Z.; Jiang, Y.
B. Chem. Asian J. 2010, 5, 543. (b) Liu, W. X.; Yang, R.; Li, A. F.; Li,
Z.; Gao, Y. F.; Luo., X. X.; Ruan, Y. B.; Jiang, Y. B. Org. Biomol. Chem.
2009, 7, 4021. (c) Li, Z.; Wu, F. Y.; Guo, L.; Li, A. F.; Jiang, Y. B. J.
Phys. Chem. B 2008, 112, 7071. (d) Liu, W. X.; Jiang, Y. B. J. Org.
Chem. 2008, 73, 1124.
(17) (a) Veale, E. B.; Gunnlaugsson. J. Org. Chem. 2008, 73, 8073.
(b) dos Santos, C. M. G.; Gunnlaugsson, T. Dalton Trans. 2009, 4712.
(c) dos Santos, C. M. G.; Glynn, M.; McCabe, T.; De Melo, J. S. S.;
Burrows, H. D.; Gunnlaugsson, T. Supramol. Chem. 2008, 20, 407.
(18) (a) Caltagirone, C.; Bates, G. W.; Gale, P. A.; Light, M. E. Chem.
Commun. 2008, 61. (b) Evans, L. S.; Gale, P. A.; Light, M. E.; Quesada,
R. Chem. Commun. 2006, 965.
(19) Cametti, M.; Rissanen, K. Chem. Commun. 2009, 2809.
(20) Gale, P. A.; Hiscock, J. R.; Moore, S. J.; Caltagirone, C.;
Hursthouse, M. B.; Light, M. E. Chem.Asian J. 2010, 5, 555.
(21) (a) Gunnlaugsson, T.; Kruger, P. E.; Lee, T. C.; Parkesh, R.;
Pfeffer, F. M.; Hussey, G. M. Tetrahedron Lett. 2003, 44, 6575.
(b) Boiocchi, M.; Del Boca, L.; Gomez, D. E.; Fabbrizzi, L.; Licchelli,
M.; Monzani., E. J. Am. Chem. Soc. 2004, 50, 16507.
(3) (a) Pradros, P.; Quesada, R. Supramol. Chem. 2008, 20, 201.
(b) Filby, M. H.; Steed, J. W. Coord. Chem. Rev. 2006, 250, 3200.
(c) Gale, P. A.; Quesada, R. Coord. Chem. Rev. 2006, 250, 3219.
(d) Nguyen, B. T.; Anslyn, E. V. Coord. Chem. Rev. 2006, 250, 3118.
(e) Gunnlaugsson, T.; Ali, H. D. P.; Glynn, M.; Kruger, P. E.; Hussey,
G. M.; Pfeffer, F. M.; dos Santos, C. M. G.; Tierney, J. J. Fluoresc. 2005,
15, 287. (f) Martínez-Manez, R.; Sancenon, F. Chem. Rev. 2003, 103,
́ ́
̃
4419.
(4) (a) Ghosh, K.; Saha, I.; Masanta, G.; Wang, E. B.; Parish, C. A.
Tetrahedron Lett. 2010, 51, 343. (b) Ghosh, K.; Masanta, G.;
Chattopadhyay, A. P. Eur. J. Org. Chem. 2009, 26, 4515. (c) Lowe,
A. J.; Dyson, G. A.; Pfeffer, F. M. Eur. J. Org. Chem. 2008, 1559.
(d) Lowe, A. J.; Dyson, F. A.; Pfeffer, F. M. Org. Biomol. Chem. 2007,
5, 1343. Pfeffer, F. M.; Lim, K. F.; Sedgwick, K. J. Org. Biomol. Chem.
2007, 5, 1795. (e) Bates, G. W.; Triyanti; Light, M. E.; Albrecht, M.;
Gale, P. A. J. Org. Chem. 2007, 72, 8921. (f) Liu, S.-Y.; Fang, L.; He, Y.-
B.; Chan, W.-H.; Yeung, K.-T.; Cheng, Y.-K.; Yang, R.-H. Org. Lett.
2005, 7, 5825.
(5) (a) Veale, E. B.; Tocci, G. M.; Pfeffer, F. M.; Kruger, P. E.;
Gunnlaugsson, T. Org. Biomol. Chem. 2009, 7, 3447. (b) dos Santos, C.
M. G.; Gunnlaugsson, T. Dalton Trans. 2009, 4712. (c) Duke, R. M.;
Gunnlaugsson, T. Tetrahedron Lett. 2007, 48, 8043. (d) dos Santos, C.
M. G.; McCabe, T.; Watson, G. W.; Kruger, P. E.; Gunnlaugsson, T. J.
Org. Chem. 2008, 73, 9235. (e) dos Santos, C. M. G.; McCabe, T.;
Gunnlaugsson, T. Tetrahedron Lett. 2007, 48, 3135. (f) Ali, H. D. P.;
Kruger, P. E.; Gunnlaugsson, T. New. J. Chem. 2008, 32, 1153.
(g) Gunnlaugsson, T.; Kruger, P. E.; Jensen, P.; Tierney, J.; Ali, H. D.
P.; Hussey, G. M. J. Org. Chem. 2005, 70, 10875.
(6) (a) Duke, R. M.; Gunnlaugsson, T. Tetrahedron Lett. 2011, 52,
1503. (b) Pfeffer, F. M.; Kruger, P. E.; Gunnlaugsson, T. Org. Biomol.
Chem. 2007, 5, 1894. (c) Pfeffer, F. M.; Buschgens, A. M.; Barnett, N.
W.; Gunnlaugsson, T.; Kruger, P. E. Tetrahedron Lett. 2005, 46, 6579.
(d) Pfeffer, F. M.; Gunnlaugsson, T.; Jensen, P.; Kruger, P. E. Org. Lett.
2005, 7, 5357. (e) Gunnlaugsson, T.; Davis, A. P.; O’Brien, J. E.;
Glynn, M. Org. Biomol. Chem. 2005, 3, 48. (f) Gunnlaugsson, T; Davis,
A. P.; Hussey, G. M.; Tierney, J; Glynn, M. Org. Biomol. Chem. 2004,
2, 1856. (g) Gunnlaugsson, T.; Kruger, P. E.; Jensen, P.; Pfeffer, F. M.;
Hussey, G. M. Tetrahedron Lett. 2003, 44, 8909.
(22) The AcO− was also best fitted to a 1:1 stoichiometric model,
which resulted in a binding constant of log K1:1 = 6.3 ( 0.3).
However, the fit was not ideal; therefore, this value was considered
only an as estimation.
(23) No measurable changes were observed upon addition of group I
ions to these solutions.
(24) We also carried out a spectrophotometric pH titration on 1 and
2 in this 1:1 mixture, the fitting of which gave rise to the determination
of two pKa values of 5.99 ( 0.06) and 8.07 ( 0.05) for 1, while pKa =
5.8( 0.07) and 7.6 ( 0.06) was determined for 2. Consequently,
buffers such as MES and TRIS could not be used for these titrations.
Such pH titrations usually have significant effect on the photophysical
properties of such sensors, see: Gunnlaugsson, T.; Parker, D. Chem.
Commun. 1998, 511.
(25) In contrast, the addition of 10 equiv of TBAOH changed the pH
from 7.3 to 11 for 1.
(26) Sensors 5 and 6 also showed colorimetric changes similar to that
observed in the absorption spectra of 1 and 2 for these anions in
EtOH.
(27) It is worth pointing out that the addition of ATP to a solution of
6 in 1:1 MeOH−H2O also showed colorimetric detection of this anion
in a manner similar to that observed for 4.
(7) Duke, R. M.; Gunnlaugsson, T. Tetrahedron Lett. 2010, 51, 5402.
(8) (a) Yang, R.; Liu, W.-X.; Shen, H.; Huang, H.-H.; Jiang, Y.-B. J.
Phys. Chem. B 2008, 112, 5105. (b) Liu, W.-X.; Jiang, Y.-B. Org. Biomol.
Chem. 2007, 5, 1771. (c) Evans, L. S.; Gale, P. A.; Light, M. E.;
Quesada, R. New J. Chem. 2006, 30, 1019.
3126
dx.doi.org/10.1021/jo202332h | J. Org. Chem. 2012, 77, 3115−3126