Detailing hydrogen bonding and deprotonation equilibria between anions and urea/thiourea derivatives

Article abstract of DOI:10.1021/jo702458f

(Chemical Equation Presented) Spectrophotometric and ¹H NMR titrations of N-methoxyethyl-N′'-(4-nitrophenyl)thiourea (3) by Bu ₄-NOAc show that in DMSO deprotonation of the receptor and formation of a hydrogen-bonded complex with ani

Full text of DOI:10.1021/jo702458f

Detailing Hydrogen Bonding and Deprotonation Equilibria between
Anions and Urea/Thiourea Derivatives
Carol Pe´rez-Casas and Anatoly K. Yatsimirsky*
Facultad de Qu´ımica, UniVersidad Nacional Auto´noma de Me´xico, 04510 Me´xico D.F., Mexico
anatoli@serVidor.unam.mx
ReceiVed NoVember 15, 2007
1
Spectrophotometric and H NMR titrations of N-methoxyethyl-N′-(4-nitrophenyl)thiourea (3) by Bu4-
NOAc show that in DMSO deprotonation of the receptor and formation of a hydrogen-bonded complex
with anion proceed simultaneously but in MeCN deprotonation requires the participation of the second
acetate anion. The formation constants of hydrogen-bonded complexes were determined from titrations
in the presence of added acetic acid, which suppressed deprotonation. These constants together with
independently measured stability constants of (AcO)2H- complexes were employed for a rigorous numerical
analysis of titration results in the absence of added acid, which allowed us to determine the equilibrium
deprotonation constants as well as pKa values for 3 in both solvents. Although 3 appeared to be a weaker
acid than AcOH in both solvents, it can be deprotonated by acetate in dilute solutions when the
concentration of liberated acetic acid is low enough. With disubstituted N,N-bis(methoxyethyl)-N′-(4-
nitrophenyl)thiourea 4 only deprotonation equilibrium is observed. In contrast, both parent urea derivatives
1 and 2 cannot be deprotonated by acetate anions. Independent of the real type of equilibrium, whether
it is a deprotonation or a hydrogen bonding, titration plots always can be satisfactorily fitted to a formal
1:1 binding isotherm. A relationship between apparent “binding constants” and real equilibrium constants
of hydrogen bonding association and deprotonation processes is discussed.
Introduction
sensors is their ability to signal the interaction with anions in
an easily detectable and quantifiable way. For this reason, the
discovery of capacity of some anions to abstract protons from
sufficiently acidic receptors converting them into brightly
colored deprotonated forms attracted much attention.2-5 Al-
though in this process the receptor behaves essentially as an
acid-base indicator, there is a significant selectivity to the type
Anion recognition by hydrogen-bonding receptors is an area
of intensive current research.1 A particularly important charac-
teristic of anion receptors required for their applications as
* To whom correspondence should be addressed. Tel: 55 5622 3813. Fax:
55 5616 2010.
(1) Reviews: (a) Amendola, V.; Bonizzoni, M.; Esteban-Go´mez, D.;
Fabbrizzi, L.; Licchelli, M.; Sanceno´n, F.; Taglietti, A. Coord. Chem. ReV.
2006, 250, 1451. (b) Gale, P. A. Coord. Chem. ReV. 2003, 240, 1. (c) Beer,
P. D.; Gale, P. A. Angew. Chem., Int. Ed. 2001, 40, 486. (d) Sessler, J. L.;
Gale, P. A.; Cho, W.-S. Anion Receptor Chemistry; Royal Society of
Chemistry: Cambridge, UK, 2006. (e) Gale, P. A. Acc. Chem. Res. 2006,
39, 465. (f) Beer, P. D.; Hayes, E. J. Coord. Chem. ReV. 2003, 240, 167.
(g) Davis, A. P. Coord. Chem. ReV. 2006, 250, 2939. (h) Schmidtchen, F.
P.; Berger, M. Chem. ReV. 1997, 97, 1609. (i) Schmidtchen, F. P. Coord.
Chem. ReV. 2006, 250, 2918.
(2) (a) Amendola, V.; Esteban-Go´mez, D.; Fabbrizzi, L.; Licchelli, M.
Acc. Chem. Res. 2006, 39, 343.(b) Boiocchi, M.; Del Boca, L.; Esteban-
Go´mez, D.; Fabbrizzi, L.; Licchelli, M.; Monzani, E. J. Am. Chem. Soc,
2004, 126, 16507. (c) Boiocchi, M.; Del, Boca, L.; Esteban-Go´mez, D.;
Fabbrizzi, L.; Licchelli, M.; Monzani, E. Chem. Eur. J. 2005, 11, 3097. (d)
Esteban-Go´mez, D.; Fabbrizzi, L.; Licchelli, M. J. Org. Chem. 2005, 70,
5717. (e) Esteban-Go´mez, D.; Fabbrizzi, L.; Licchelli, M.; Monzani, E.
Org. Biomol. Chem. 2005, 3, 1495. (f) Bonizzoni, M.; Fabbrizzi, L.;
Taglietti, A.; Tiengo, F. Eur. J. Org. Chem. 2006, 16, 3567.
10.1021/jo702458f CCC: $40.75 © 2008 American Chemical Society
Published on Web 02/21/2008
J. Org. Chem. 2008, 73, 2275-2284
2275

Pe´rez-Casas and Yatsimirsky
of anion, and several systems based on this phenomenon have
been developed for practical anion sensing.6
characterized in terms of a meaningless “binding” constant
formally calculated by fitting the absorbance changes as a
function of added anion concentration to the equation for 1:1
binding isotherm and the real K2 value has never been estimated.
Also, in situations when a second anion is needed to induce
the deprotonation, the fact that the complex A2H- has a
sufficiently high formation constant has never been proved,
although the reaction is usually performed in highly polar
solvents like DMSO where such complexes may be rather
unstable. Finally, a situation when both formation of a hydrogen-
bonded complex and deprotonation processes occur simulta-
neously was not yet analyzed quantitatively. This last situation
is important because the highest stability of hydrogen-bonded
complexes is reached for couples of proton donor and proton
acceptor of equal basicity,7 but for such a couple K2 ) 1 and
deprotonation by 50% should be observed already in equimolar
mixture of anion and receptor. However, in fact, the degree of
deprotonation is a more complex function of equilibrium
constants and reaction conditions. It follows from eqs 4 and 5
that the ratio of concentrations of deprotonated and hydrogen
bonded forms is given by eq 7.
A recently developed general scheme of processes involved
in anion-receptor interaction considers three types of reactions.2
If basicity of an anion A- is insufficient to induce deprotonation
of the receptor RH, one observes formation of a hydrogen-
bonded complex R-H‚‚‚A- (eq 1) manifested in a red shift of
the receptor UV absorption band and a downfield shift or often
disappearance of NMR signals of receptor protons involved in
the hydrogen bonding.
RH + A- a RHA-
(1)
If basicity of an anion if high enough to deprotonate the
receptor (eq 2), one observes appearance of a new intense
absorption bond in the visible range of the electronic spectrum
attributed to the deprotonated receptor, disappearance of NMR
signals of abstracted receptor protons, and an upfield shift of
signals of adjacent receptor protons.
RH + A- a R- + AH
(2)
In a borderline case, an anion initially forms a hydrogen-
bonded complex, but with sufficiently high excess of added
anions the deprotonation occurs due to formation of hydrogen
bonded anion dimers A2H- (eq 3), which shifts the equilibrium
2 to the right. This last situation is observed most often with
fluoride anion and also was described with acetate anions.2a,b,e
[R-]/[RHA-] ) K2/K1[AH]
(7)
Obviously, strong hydrogen bonding (large K1) will reduce
the degree of deprotonation. At the same time, if there is no
added acid the concentration of AH will be equal to the
concentration of deprotonated form, which in its turn is
proportional to total receptor concentration. Therefore, depro-
tonation will be more significant in more dilute receptor
solutions where less acid is produced and the “binding constant”
formally calculated for the deprotonation reaction will depend
on the receptor concentration.
In this paper, possible approaches to analyze all of these
aspects are illustrated by performing a detailed study of
interactions between acetate anions and a set of receptors 1-4
of structural type frequently employed in design of anion-sensing
molecules in two most often used solvents DMSO and MeCN.
Both UV-vis and NMR titrations were analyzed numerically
in terms of a complete set of coexisting equilibria 1-3. An
approach to determine separately the equilibrium constant for
the hydrogen-bonding reaction by performing titrations in the
presence of added acid was applied. Special attention was paid
to consistency of spectroscopic manifestations of one or another
type of interaction, which follow from the results of UV-vis
and NMR titrations.
AH + A- a A2H-
The respective equilibrium constants are given by eqs 4-6.
(3)
K1 ) [RHA-]/[RH][A-]
K2 ) [R-][AH]/[RH][A-]
K3 ) [A2H-]/[AH][A-]
(4)
(5)
(6)
Although this scheme in general is well justified, a quantita-
tive description of equilibria involved still requires more
detailing. In particular, the deprotonation process is always
(3) (a) Evans, L. S.; Gale, P. A.; Light, M. E.; Quesada, R. Chem.
Commun. 2006, 965. (b) Brooks, S. J.; Edwards, P. R.; Gale, P. A.; Light,
M. E. New J. Chem. 2006, 30, 65.
(4) (a) Gunnlaugsson, T.; Glynn, M.; Hussey, G. M.; Kruger, P. E.;
Pfeffer, F. M. Coord. Chem. ReV. 2006, 250, 3094. (b) Pfeffer, F. M.;
Gunnlaugsson, T.; Jensen, P.; Kruger, P. E. Org. Lett. 2005, 7, 5357. (c)
Gunnlaugsson, T.; Kruger, P. E.; Jensen, P.; Pfeffer, F. M.; Hussey, G. M.
Tetrahedron Lett. 2003, 44, 8909. (d) Quinlan, E.; Matthews, S. E.;
Gunnlaugsson, T. J. Org. Chem. 2007, 72, 7497.
(5) (a) Lin, C.; Simov, V.; Drueckhammer, D. G. J. Org. Chem. 2007,
72, 1742. (b) Kim, Y. J.; Kwak, H.; Lee, S. J.; Kwon, H. J.; Nam, S. H.;
Lee, K.; Kim, C. Tetrahedron 2006, 62, 9635. (c) Yen, Y. P.; Chen, C. L.;
Fu, T. M.; Wu, C. Y.; Lin, C. Y. Aust. J. Chem. 2006, 59, 805. (d) He, X.;
Hu, S.; Liu, K.; Guo, Y.; Xu, J.; Shao, S. Org. Lett. 2006, 8, 333. (e) Sun,
X. H.; Li, W.; Xia, P. F.; Luo, H.-B.; Wei, Y.; Wong, M. S.; Cheng, Y.-
K.; Shuang, S. J. Org. Chem. 2007, 72, 2419. (f) Lin, C.; Selvi, S.; Fang,
J.; Chou, P.; Lai, C.; Cheng, Y. J. Org. Chem. 2007, 72, 3537. (g) Jose, D.
A.; Kumar, D. K.; Ganguly, B.; Das, A. Tetrahedron Lett. 2005, 46, 5343.
(h) Fillaut, J.; Andrie´s, J.; Perruchon, J.; Desvergne, J.; Toupet, L.; Fadel,
L.; Zouchoune, B.; Saillard, J. Inorg. Chem. 2007, 46, 5922. (i) Jose, D.
A.; Kar, P.; Koley, D.; Ganguly, B.; Thiel, W.; Ghosh, H. N.; Das, A.
Inorg. Chem. 2007, 46, 5576. (j) Cui, Y.; Mo, H.; Chen, J.; Niu, Y.; Zhong,
Y.; Zheng, K.; Ye, B. Inorg. Chem. 2007, 46, 6427.
(6) (a) Palacios, M. A.; Nishiyabu, R.; Marquez, M.; Anzenbacher, P. J.
Am. Chem. Soc. 2007, 129, 7538. (b) Yen, Y.; Ho, K. Tetrahedron Lett.
2006, 47, 7357.
(7) Shan, S.; Loh, S.; Herschlag, D. Science 1996, 272, 97.
2276 J. Org. Chem., Vol. 73, No. 6, 2008

Hydrogen Bonding and Deprotonation Equilibria
FIGURE 1. (A) Spectrophotometric titration of 4.8 × 10-5 M 1 by Bu4NOAc in DMSO. Arrows show the directions of spectral changes at
1
increasing Bu4NOAc concentrations (0-4 mM). (B) Partial H NMR spectra of 5 mM 1 in DMSO-d6 recorded with increased amounts of Bu4-
NOAc.
TABLE 1. Summary of Equilibrium Constants for Hydrogen Bonding or/and Deprotonation of Receptors 1-4 and Acetic Acid by Acetate
Anions
K1, M-1
K2
K3, M-1
receptor
DMSO
MeCN
DMSO
MeCN
DMSO
MeCN
1
2
3
4
(1.09 ( 0.05) × 103
(1.14 ( 0.03) × 103
(2.15 ( 0.07) × 104
(1.06 ( 0.06) × 105
40 ( 10
14 ( 2
0.050 ( 0.05
2.0 ( 0.2
0.020 ( 0.002
0.040 ( 0.004
AcOH
125 ( 4
(9.7 ( 0.2) × 103
Results and Discussion
see that signals of ureido protons H1 and H2 undergo a strong
downfield shift by ca. 3 ppm, signals of H3 also undergo a small
downfield shift by 0.1 ppm attributable to a through-space
interaction with hydrogen bound acetate anion,2c and signals
of H4 undergo also small, but upfield shift due to an inductive
shielding effect of the bound anion. Fitting of the titration profile
for H2 protons, Figure 1S(B) (Supporting Information), to the
equation similar to 8, but rewritten for NMR titration, gives K1
) 720 ( 50 M-1 reasonably close to the value found by more
precise spectrophotometric titration. Similar pattern was ob-
served also in MeCN, but in this case, K1 was too large to be
reliably calculated form NMR results obtained at high receptor
concentration.
Interaction of receptor 2 with acetate in both solvents also
proceeds as hydrogen-bonding association (Figure 2S, Support-
ing Information), but with expectedly much smaller K1 values
found by NMR titrations (Table 1).
Association between Acetic Acid and Acetate Anion. The
association constants K3 ) 180 M-1 in DMSO and 1600-4000
M-1 in MeCN have been determined recently by potentiometric
titrations with a glass electrode and a modified calomel
electrode.9 Since the use of a glass electrode in pure nonaqueous
solvents is not free of possible systematic errors and the
calculations were based on literature values of pKa of picric
acid in respective solvents used as a standard for the electrode
Urea Derivatives 1 and 2. Arylurea derivatives lacking
strong electron-accepting substituents usually cannot be depro-
tonated by anions.2a In line with this, we observed with com-
pounds 1 and 2 only formation of hydrogen-bonded complexes
with acetate. Figure 1A illustrates the course of spectrophoto-
metric titration of 1 by Bu4NOAc in DMSO, which shows a
typical pattern for formation of a hydrogen-bonded complex.
The absorption maximum of 1 at 351 nm shifts to 365 nm with
two isosbestic points at 356 and 288 nm. The spectral course
of interaction of 1 with acetate anions in MeCN was similar
but all spectra were shifted to shorter wavelengths by ca. 20
nm.
Fitting the profiles of absorbances (Abs) measured at several
fixed wavelengths vs [Bu4NOAc] to the eq 8 for a 1:1 binding
isotherm (subscript T stands for total concentration, Abs0 is the
initial absorbance measured in the absence of acetate, ∆ꢀ is the
difference in molar absorptivities between the complex and free
receptor)8 gives Kassoc ) K1 shown in Table 1. Examples of
such fittings are shown in Figure 1S(A) (Supporting Informa-
tion). As expected for a hydrogen-bonded association, larger
K1 is observed in a less polar solvent.
Abs ) Abs0 + 0.5∆ꢀ([RH]T + [AcO-]T + 1/Kassoc
(([RH]T + [AcO-]T + 1/Kassoc)2 - 4[RH]T[AcO-]T)0.5) (8)
-
(8) Schneider, H.-J.; Yatsimirsky, A. K. Principles and Methods in
Supramolecular Chemistry; John Wiley and Sons: Chichester, U.K., 2000.
(9) Kozak, A.; Czaja, M; Chmurzyn´ski, L. J. Chem. Thermodyn. 2006,
38, 599. Czaja, M; Makowski, M.; Chmurzyn´ski, L. J. Chem. Thermody-
namics 2006, 38, 606.
Results of NMR titration confirm this type of interaction.
Figure 1B shows the partial 1H NMR spectra of 1 recorded with
increased amounts of acetate in DMSO-d6, from which one can
J. Org. Chem, Vol. 73, No. 6, 2008 2277

Pe´rez-Casas and Yatsimirsky
FIGURE 2. (A) Spectrophotometric titration of 7.0 × 10-5 M 3 by Bu4NOAc (0-0.8 mM) and Bu4NOH (0-0.4 mM) in DMSO. (B)
Spectrophotometric titration of 5.0 × 10-5 M 3 by Bu4NOAc (0-3 mM) in DMSO in the presence of 1.9 mM AcOH. Arrows show the directions
of spectral changes at increasing Bu4NX concentrations.
1
FIGURE 3. (A) Partial H NMR of 24 mM 3 with increased amounts of added Bu4NOAc in DMSO-d6. (B) Titration profile for H3, solid line is
the fitting curve to the equation for 1:1 complexation.
calibration, we considered it worthy to determine K3 values in
significant excess of Bu4NOAc over 1 and of AcOH over Bu4-
NOAc, a simple eq 9 can be used to fit the profile of absorbance
at a fixed wavelength vs [AcOH] obtained in the presence of a
constant concentration of Bu4NOAc. In eq 9, A01 and A00 are
the initial absorbance of the mixture of 1 with Bu4NOAc and
the absorbance of 1 alone respectively, and Kobs ) K3/(1 +
K1[Bu4NOAc]).
these solvents again by an independent technique. An attempt
1
to measure K3 by H NMR titration of acetic acid with Bu4-
NOAc was unsuccessful because of strong broadening and
disappearance of the signal of the carboxylic group proton
already on addition of ca. 5% molar equiv of acetate ions.
Therefore, we employed the effect of acetic acid on the inter-
action of acetate anions with 1, which was not complicated by
the deprotonation process, for an indirect determination of K3.
In preliminary experiments, we found that addition of up to
0.1 M AcOH did not affect UV and NMR spectra of 1 in DMSO
and MeCN. At the same time, addition of AcOH to a solution
containing a mixture of 1 with an excess of acetate anion
resulted in a restoration of the original spectrum of 1. This can
be attributed to displacement of equilibrium 1 to the left due to
consumption of free acetate anions by acetic acid affording the
complex (AcO)2H- via reaction 3. Under conditions of a
A ) (A01 + A00Kobs[AcOH])/(1 + Kobs[AcOH]) (9)
Figure 3S (Supporting Information) illustrates the titration
profiles obtained in DMSO (3 mM Bu4NOAc added) and MeCN
(0.14 mM Bu4NOAc added), which fit to eq 9 with observed
equilibrium constants 29.1 ( 0.1 and 2720 ( 50 M-1
,
respectively. With the above K1 values, one obtains from these
results the equilibrium constants K3 for the reaction 3 given in
Table 1, which are reasonably close to the above-reported
2278 J. Org. Chem., Vol. 73, No. 6, 2008

Hydrogen Bonding and Deprotonation Equilibria
increase in intensity of these new signals, disappearance of
initially observed signals, and change of color from orange to
deep red probably due to second deprotonation of 3. Apparently
in the presence of 1 equiv of hydroxide both singly and a small
amount of doubly deprotonated forms coexist in a slow
equilibrium. It seems therefore that the first deprotonation affects
very little the positions of aromatic signals apparently because
of strong delocalization of the negative charge. As a result of
this, the shift in H3 proton signal induced by a through space
interaction with the hydrogen bonded acetate anion becomes a
predominant effect.
It is worth noting that while results of spectrophotometric
titration clearly show only deprotonation by acetate but do not
show any spectral characteristics attributable to hydrogen
bonding, the results of NMR titrations on the contrary show
only hydrogen bonding interaction without any indication of
deprotonation (broadening is often observed during formation
of hydrogen-bonded complexes, see e.g., refs 2c and 11). As
will be shown below, the reason for this apparent discrepancy
is that titration experiments by these two techniques are
performed in very different concentration ranges of the receptor.
A possible way to reveal the formation of expected hydrogen-
bonded complexes under conditions of spectrophotometric
titration is to suppress deprotonation by adding a small amount
of acetic acid. Indeed, when the spectrophotometric titration of
3 was performed in the presence of added 1.9 mM AcOH, the
spectral course of titration (Figure 2B) corresponded to those
expected for the formation of a hydrogen-bonded complex (cf.
Figure 1A). Fitting of the titration profiles between 370 and
400 nm to the eq 8 gives after a small correction for the
complexation of acetate with added acetic acid the value of K1
given in Table 1. This value is close to that estimated from the
NMR titration (see above) and also is close to K1 for urea
derivative 1. It should be noted in this connection that more
acidic thiourea derivatives do not always form more stable
hydrogen-bonded complexes than corresponding urea deriva-
tives. In fact, even formation of less stable complexes with
thioureas was reported and attributed to different conformations
of thio and oxo derivatives.12
FIGURE 4. Titration curve for 7.0 × 10-5 M 3 (absorbance at 482
nm) with Bu4NOAc in DMSO. Solid line: formal fit to eq 8. Dashed
line: fit to eqs 1-3 (see text for details).
values. Apparently, the formation of (AcO)2H- in DMSO is
insignificant below 0.01 M acetate, but in MeCN the association
is significant already in mM range of concentrations of acetate
anions. Interestingly, the formation constant for (AcO)2H- in
MeCN is much higher than that in chloroform, K3 ) 861 M-1 10
.
Thiourea Derivatives 3 and 4. Additions of acetate anions
to more acid thiourea derivatives induced deprotonation in both
solvents. Spectrophotometric titration of 3 by Bu4NOAc in
DMSO (Figure 2A) shows the appearance of a new band at
482 nm characteristic of deprotonation of the receptor. Indeed,
titration with Bu4NOH produces the same band (Figure 2A,
inset), however, of a larger intensity. Another difference is that
the titration with acetate anions proceeds with two rather
approximate isosbestic points at 380 and 280 nm, but titration
with hydroxide involves two different clearly observed isosbestic
points at 395 and 289 nm. This comparison indicates that the
deprotonation by acetate anions is incomplete and the interaction
with acetate may involve an additional product, most probably
a hydrogen-bonded complex.
Thus, we conclude from above results that the formation of
hydrogen-bonded complex between 3 and acetate, reaction 1,
and deprotonation of 3, reaction 2, proceed simultaneously.
Let us consider now the spectrophotometric titration profile
for 3 obtained in the absorption maximum of the deprotonated
form at 482 nm, Figure 4. The concentration of acetate anions
during the titration experiment is always below 1 mM, and
therefore, the degree of complexation of acetic acid liberated
in the step 2 by acetate anions is always less than 10%, and
step 3 can be neglected. Combining equilibria 1 and 2, one
obtains for the concentration of deprotonated form the eq 10.
Results of NMR titration, Figure 3A, indeed show charac-
teristics expected for formation of a hydrogen-bonding complex.
Signals of NH protons broaden, but the signal of H1 proton
is still detectable and shows a characteristic for hydrogen-
bonding strong downfield shift. Signals of aromatic protons
also are shifted in the same way as in titration of 1: H3
downfield and H4 upfield. Fitting of the titration profiles to the
equation for 1:1 association (Figure 3B illustrates the fitting
for protons H3) gives an approximate value of K1 ) 1000 (
(K2/[AcOH])[AcO-]
1 + (K1 + K2/[AcOH])[AcO-]
250 M-1
.
[R-] ) [R]T
(10)
Titration with strong Bu4NOH base, which definitely leads
to deprotonation, shows a completely different pattern: signals
of NH protons disappear on addition of ca. 0.1 equiv of
hydroxide, and signals of aromatic protons undergo a very small
upfield shift by 0.05 ppm and lose their multiplicity in the
presence of 1 equiv of hydroxide (Figure 4S, Supporting
Information). Also, two new upfield-shifted weak signals appear
in the spectrum. Further addition of Bu4NOH leads to a strong
Equation 10 predicts a hyperbolic titration curve similar to
that for the case of a simple 1:1 complexation process, which,
however, should be strongly distorted since concentration of
(11) Dahan, A.; Ashkenazi, T.; Kuznetsov, V.; Makievski, S.; Drug, E.;
Fadeev, L.; Bramson, M.; Schokoroy, S.; Rozenshine-Kemelmakher, E.;
Gozin, M. J. Org. Chem. 2007, 72, 2289.
(12) Roussel, C.; Roman, M.; Andreoli, F.; Del Rio, A.; Faure, R.;
Vanthuyne, N. Chirality 2006, 18, 762.
(10) Mascal, M.; Marjo C. E.; Blake, A. J. Chem. Commun. 2000, 1591
J. Org. Chem, Vol. 73, No. 6, 2008 2279

Pe´rez-Casas and Yatsimirsky
acetic acid grows in the course of titration, and therefore, the
apparent “binding constant” is variable. Nevertheless, the
titration curve fits surprisingly well to eq 8 derived for a simple
1:1 complexation reaction (Figure 4, solid line) with an apparent
“binding constant” Kapp ) (1.41 ( 0.05) × 104 M-1 1 order of
magnitude larger than K1. In order to apply physically meaning-
ful eq 10 for the fitting, one needs to resolve mass balance
equations for acetate anions and acetic acid and to substitute
total concentrations of these components for their free concen-
trations in eq 10. This leads to too complicated expression and
therefore one need to use a more powerful numerical fitting
program such as, e.g., Hyperquad, to find the equilibrium
constants by iterations. We found it possible however to apply
a simplified procedure, which allows one to control better the
iteration steps and gives the same final results.
To analyze the titration results in terms of a complete scheme
involving all three reactions 1-3, the species distribution
diagram was calculated by using the HYSS 2000 (Hyperquad
Simulation and Speciation) program13 with values of K1 and
K3 determined above and with K2 ) 1 as a first approximation.
The calculated concentrations of deprotonated form of 3 at
FIGURE 5. Titration curves for receptor 4 (absorbance at 492 nm)
with Bu4NOAc in DMSO at different concentrations: 0.03 mM (open
triangles), 0.11 mM (solid triangles), 0.03 mM in the presence of 0.65
mM AcOH (open squares), and 4.4 mM (solid squares, NMR titration,
right and upper scales,). Solid lines: formal fits to eq 8. Dashed lines:
fits to eqs 2 and 3.
variable [Bu4NOAc] were multiplied by the molar absroptivity
of the deprotonated form (ꢀR ) 1.8 × 104 M-1 cm-1
)
-
determined under conditions of complete deprotonation of the
receptor by an excess of tetrabutylammonium hydroxide and
thus obtained absorbance vs [Bu4NOAc] profile was compared
with the experimentally obtained profile shown in Figure 4. The
value of K2 was varied, initially with a step ∆ log K2 ) 0.5 and
then with a step ∆ log K2 ) 0.05 in the range where calculated
and experimental profiles were most close to each other, until
the best fit shown by a dashed line in Figure 4 was reached.
With this manual iteration procedure we obtained K2 ) 0.050
( 0.05.
The equilibrium constant for the reaction 2 can be expressed
as a ratio of acid dissociation constants of the receptor and acetic
acid
K2 ) KaRH/KaAcOH
(11)
or in the logarithmic form
log K2 ) pKaAcOH - pKaRH
(12)
The simulation of the titration profile with thus obtained
equilibrium constants shows that the “saturation” observed in
Figure 4 is to some extent apparent. The fraction of deprotonated
form of the receptor reaches 39% at the end of titration (0.8
mM acetate) and further doubling in acetate concentration
increases it to only 43%, with what looks like saturation within
dispersion of experimental measurements. However, at concen-
trations of Bu4NOAc of 10 mM, the equilibrium 3 will be
already significant, and further increase in acetate concentration
must displace the equilibrium of deprotonation to the right.
Surprisingly, we did not observe this, and the degree of
deprotonation did not surpass ca. 50% even at concentrations
of acetate ions up to 20 mM. A similar effect was also observed
in MeCN and will be discussed below.
The species distribution diagram calculated for conditions of
the NMR titration experiment (Figure 3) predicts that concentra-
tion of the deprotonated form does not exceed 5% of total 3 at
highest employed concentration of acetate anions. This explains
why the NMR titration shows only formation of the hydrogen-
bonded complex. The reason for the low degree of deprotonation
is that there is a high concentration of 3 and therefore also a
high concentration of acetic acid which would be liberated
during significant deprotonation. Note in this connection that
the ratio of concentrations of deprotonated and hydrogen bonded
forms depends on the absolute concentration of acetic acid (eq
7) rather than on the ratio of concentrations of Bu4NOAc and
receptor.
RH
From the last relationship, one can estimate pKa ) 13.6
for 3 from the experimentally measured log K2 ) -1.3 and
AcOH
known pKa
) 12.3 in DMSO.14 Obviously, 3 is a weaker
acid than AcOH; nevertheless, deprotonation by acetate may
be nearly quantitative at sufficiently high dilution of the receptor.
The pKa for receptor 1 may be estimated assuming that the effect
of substitution of O for S in 3 is equal to the difference in pKa
values for urea (26.9) and thiourea (21) in DMSO.15 Such an
RH
estimate predicts a very high value of pKa ) 19.4 for 1 and
the respective value for log K2 ) -7.1. The calculated degree
of deprotonation of 5 × 10-5 M 1 in the presence of 0.01 M
acetate anions is only 0.5%. The deprotonation of unsubstituted
thiourea by acetate would have log K2 ) -8.8 and cannot be
observed even with very high concentration of added acetate
anions.
Very weak hydrogen bonding observed with N,N-disubsti-
tuted derivative 2 (Table 1) allows one to expect that in DMSO
the deprotonation reaction 2 may be the sole process at low
acetate concentrations for the receptor 4. Indeed, the spectral
course of titration of 4 by Bu4NOAc shows the appearance of
intense band at 492 nm (Figure 5S, Supporting Information)
and is similar to that observed during titration by Bu4NOH.
Titration curves for 4 at two different concentrations of the
receptor are shown in Figure 5. Like in the case of 3, they both
can be fitted satisfactorily to the eq 8 with apparent “binding
(14) Izutsu, K. Acid-Base Dissociation Constants in Dipolar Aprotic
SolVents; Blackwell Scientific Publications: Oxford, 1990.
(15) Bordwell, F. G. Ji, G. J. Am. Chem. Soc. 1991, 113, 8398.
(13) Alderlghi, L.; Gans, P.; Ienco, A.; Peters, D.; Sabatini, A.; Vacca,
A. Coord. Chem. ReV. 1999, 184, 311.
2280 J. Org. Chem., Vol. 73, No. 6, 2008

Hydrogen Bonding and Deprotonation Equilibria
FIGURE 6. (A) Spectrophotometric titration of 6.4 × 10-5 M 3 by Bu4NOAc in MeCN. Solid lines: spectra recorded at 0-0.5 mM acetate.
Dashed lines: spectra recorded at 0.5-25 mM acetate. (B) Titration of 6.4 × 10-5 M 3 by Bu4NOH (0-0.15 mM) in MeCN. Arrows show the
directions of spectral changes at increasing Bu4NX concentrations.
constants” Kapp ) (3.6 ( 0.4) × 104 M-1 for 0.03 mM 4 and
Kapp ) (1.9 ( 0.3) × 104 M-1 for 0.11 mM 4. NMR titration
of 4.4 mM 4 in DMSO-d6 shows a pattern typical for
deprotonation process (Figure 6S, Supporting Information), and
formal fitting of the titration profile for proton H3 (Figure 5,
solid squares) gives Kapp ) (4.1 ( 0.2) × 102 M-1. Thus, all
titration profiles can be satisfactorily fitted to the 1:1 binding
isotherm, but with “binding constants”, which vary by a factor
of 100 depending on the concentration of the receptor. At the
same time, analysis in terms of eqs 2 and 3 allows one to fit all
data with a single value of K2 ) 2.0 ( 0.2 and the value of K3
acid to 1:1 binding isotherm and subsequent correction of Kapp
in accordance with eq 13. Surprisingly, we see at the same time
that the shape of the titration curve is not significantly affected
by a large variation in Kapp, i.e., Kassoc the eq 8. Thus, these
apparent “binding constants” may be used for, e.g., comparative
purposes, but obviously titrations of different receptors must
be performed at the same concentration.
The values of Kapp obtained in the absence of added acid are
proportional to the inverse concentration of 4 in accordance with
a simple expression Kapp ) (1.7 ( 0.3)/[4]. This dependence
reflects the fact that the concentration of acetic acid liberated
in the system during titration is proportional to total concentra-
tion of the receptor. Interestingly, the proportionality coefficient
is close to K2. On the other hand, an attempt to substitute the
ratio K2/[AcOH] in the eq 10 with a constant term K2/[3] )
710 M-1 by analogy with results for 4 fails. By substituting
K2/[AcOH] ) constant ) 710 M-1 together with K1 ) 1140
M-1 in eq 10, one obtains Kapp ) 1850 M-1, a value 7.5-fold
lower than that observed experimentally. At the same time, at
the end of titration, the ratio K2/[AcOH] indeed should be close
to 700 M-1 since according to eq 10 the “saturation” of the
titration profile is observed at [R-]/[R]T ) (K2/[AcOH])/(K1 +
K2/[AcOH]) and the experimentally observed saturation at 39%
of deprotonation requires K2/[AcOH] ) 730 M-1. Large
observed Kapp results, therefore, from a very low concentration
of acetic acid at the beginning of the titration.
given above (the fitting curves are shown as dashed lines in
RH
Figure 5). With K2 ) 2.0, one obtains from the eq 12 pKa
)
12.0 for 4. The reason of higher acidity of 4 as compared to 3
is not clear; probably it is related to some conformational
changes induced by second substitution at the nitrogen atom.
Perhaps the most extraordinary aspect of these titration
experiments is a very good quality of the fitting to obviously
incorrect model of 1:1 complex formation (solid lines in Figures
4 and 5), even surpassing in some cases the fitting quality to
experimentally justified deprotonation model (dashed lines in
Figures 4 and 5). It follows from the eq 5 that when reaction 2
is the dominant process the expression for Kapp takes the form
Kapp ) [R-]/[RH][AcO-] ) K2/[AcOH]
(13)
and therefore, Kapp must vary by a factor of 10 during the
titration experiment, which covers an interval of conversion of
RH into R- from ca. 10% to ca. 90%. A possible way to make
Kapp really constant is to perform the titration in the presence
of an excess of AcOH over the receptor. The titration plot shown
by open squares in Figure 5 was obtained in the presence of
0.65 mM AcOH, 20 times in excess over 4. As expected, it fits
very well to eq 8, and the fitting gives in this case Kapp ) (3.8
( 0.4) × 103 M-1, reasonably close to the expected value of
Kapp ) 2/[AcOH] ) 3.1 × 103 M-1. Thus, in principle, true
values of K2 for deprotonation may be obtained by simple fitting
of titrations performed in the presence of small amounts of added
Interaction of 3 with acetate in MeCN also involves both
formation of hydrogen-bonded complex and deprotonation, but
in this case the reactions are not simultaneous and deprotonation
process requires participation of the second acetate anion. The
spectral course of the titration in MeCN is shown in Figure
6A. In the range of acetate concentrations 0-0.5 mM, the
spectral changes are typical for formation of the hydrogen-
bonded complex: the absorption maximum of free receptor at
340 nm is shifted to 366 nm through an isosbestic point at 346
nm, but at higher acetate concentrations a new bond at 452 nm
characteristic of the deprotonation process is developed with
two isosbestic points at 397 and 284 nm. Titration by Bu4NOH,
J. Org. Chem, Vol. 73, No. 6, 2008 2281

Pe´rez-Casas and Yatsimirsky
FIGURE 7. Partial 1H NMR spectra of 3 in MeCN-d3 in the presence of increased amounts of (A) Bu4NOAc with 9 mM 3 and (B) Bu4NOH with
15 mM 3.
Figure 6B, gives a more intense band at 459 nm indicating an
incomplete deprotonation of 3 by acetate anions. A shorter
wavelength of maximum observed in the presence of acetate
ions is due to superimposing of the band of the deprotonated
form with the absorption band of the hydrogen-bonded complex.
After deconvolution of the final spectrum in Figure 6A the
maximum appears at 460 nm. The spectra for titration by
hydroxide initially do not pass through an isosbestic point
indicating that the reaction with hydroxide may be a more
complicated process than just a simple one-step deprotonation.
Results of NMR titration of 9 mM 3 in MeCN by Bu4NOAc
(0-0.075 M) indicate only formation of a 1:1 hydrogen-bonded
complex (Figure 7A): signals of H1 and H2 are shifted downfield
by ca. 4.5 ppm, H3 also is shifted downfield by 0.2 ppm, and
H4 undergoes a very small upfield shift by 0.05 ppm. There is
no indication of deprotonation even in the presence of a 10-
fold excess of acetate anions over 3. Titration with hydroxide
shows a typical deprotonation pattern, Figure 7B: signals of
NH protons disappear and signals of aromatic protons are shifted
FIGURE 8. Apparent molar absorptivity (ꢀapp) of 3 at 460 nm in MeCN
vs [Bu4NOAc] at different concentrations of the receptor: 1.25 mM
(solid triangles), 0.064 mM (open squares), 0.013 mM (solid squares).
upfield. However, the shift of H4 signal is nearly complete
already in the presence of 0.5 mol equivalent of the base and
the signal of H3 undergoes a strong broadening under the same
Solid lines are formal fits to the eq 8 and dashed lines are theoretical
conditions. In spectrophotometric titration, Figure 6B, the spectra
start to pass through an isosbestic point after approximately 50%
of deprotonation. All these observations indicate that the course
of the reaction between 3 and hydroxide is changed after half
of the receptor becomes deprotonated. Probably this is due to
formation of a hydrogen-bonded complex R2H- between anion
of 3 and its neutral form analogous to the (AcO)2H- complex,
which finally becomes completely deprotonated at sufficiently
high hydroxide concentration.
Fitting of absorbance vs acetate concentration profiles to the
eq 8 at low acetate concentrations give K1 ) (6 ( 1) × 104
M-1. In order to get a more accurate value of K1, titration of 3
was performed in the presence of added 1.2 mM AcOH, which
suppress completely the deprotonation process (Figure 7S,
Supporting Information). Fitting of these results gives a smaller
observed value of Kassoc ) 8400 ( 500 M-1, from which after
profiles calculated in accordance with Schemes 1-3 with equilibrium
constants given in the text.
correction for the complexation of acetate with added acetic
acid one obtains K1 value given in Table 1.
The titration profiles at 460 nm obtained at different total
concentrations of 3 are shown in Figure 8. They all can be
satisfactorily fitted to the eq 8 with formal association constants
Kapp equal to 180 ( 10, 850 ( 70, and 7600 ( 100 M-1 for
1.25, 0.064, and 0.013 mM 3, respectively. The degree of
deprotonation at “saturation” also increases on going to more
dilute receptor solutions but never reaches 100% (the molar
absroptivity of the deprotonated form at 460 nm found from
results with hydroxide equals 18700 M-1 cm-1). The manual
fitting to the scheme involving eqs 1-3 is shown by dashed
2282 J. Org. Chem., Vol. 73, No. 6, 2008

Hydrogen Bonding and Deprotonation Equilibria
lines. Obviously, in all cases the saturation occurs much earlier
than predicted theoretically, although results at low acetate
concentrations are fitted reasonably well with K2 ) 0.020 (
0.002 M-1. By using the eq 11 with known pKa 22.3 of acetic
acid in MeCN,14 one obtains pKa ) 24.0 for 3 in this solvent.
Thus, we again observe a significant deprotonation of the
receptor by acetate anion, which is a base nearly 2 orders of
magnitude weaker than anion of 3.
Deprotonation constants K2 differ in two solvents very little;
just by a factor of 2.5 (see Table 1). Therefore, the fact that in
MeCN deprotonation requires the second acetate anion while
in DMSO it proceeds simultaneously with formation of the
hydrogen-bonded complex is not due to a decreased acidity of
the receptor in MeCN. The reason for a different behavior in
these solvents is that in MeCN the stability of the hydrogen-
bonded complex RHA- is much higher than in DMSO with K1
value larger by a factor of 102 (Table 1). Therefore, deproto-
nation via step 2 cannot compete with the complex formation
until the concentration of acetate surpasses 0.1 mM, and it starts
to remove efficiently acetic acid liberated in step 2 by converting
it into a (AcO)2H- complex with stability constant K3 also much
larger in MeCN than in DMSO. Thus, less polar MeCN solvent
is less favorable for the direct deprotonation because of a higher
stability of the hydrogen-bonded complex, but this is compen-
sated by a larger stability of (AcO)2H- complex in MeCN as
compared with DMSO, and deprotonation becomes possible at
higher concentrations of acetate. Such “compensation” is not
observed in chloroform, however, where the stability of
hydrogen-bonded complexes of anions with urea derivatives is
higher than in MeCN (see, e.g., ref 1g) but the stability of
(AcO)2H- complex is lower (see above). As a result, deproto-
nation was never reported in this solvent.
FIGURE 9. Titration curve for 5.5 × 10-5 M 4 (absorbance at 476
nm) with Bu4NOAc in MeCN. Dashed line: fit to eqs 2 and 3. Solid
line - formal fit to eq 8.
concentration of the receptor. A smaller than expected decrease
in Kapp may be attributed to higher concentrations of acetate in
the NMR titration experiment and, consequently, a stronger shift
in the deprotonation equilibrium due to higher degree of
complexation of acetic acid into (AcO)2H- species.
Conclusions
The results of this study point to several often overlooked
important aspects of interactions between receptors possessing
relatively acid proton donor groups and basic anions like acetate,
fluoride, or hydrophosphate, capable both to complex them via
hydrogen bonding and to deprotonate them. The degree of
deprotonation depends on the concentration of the receptor and
is larger in more dilute solutions. As a result, an NMR titration
performed at high receptor concentration may feel only hydrogen
bonding, while a spectrophotometric titration of much more
dilute receptor may clearly show the deprotonation. The
deprotonation equilibrium can be shifted to a higher yield of
the deprotonated receptor by complexation of liberated acid with
an excess of anion, but in case of acetate the equilibrium
constant for the formation of the complex (AcO)2H- is
significant only in less polar MeCN solvent. A simple way to
measure the equilibrium constant for the formation of the
hydrogen-bonded complex is to perform the titration in the
presence of added acid, which suppresses the deprotonation
process. In order to obtain the equilibrium constant for the
deprotonation reaction, a more complicated procedure needs to
be employed. We found that a reasonably simple approach
consists in a manual iteration procedure employing HYSS
program for species distribution calculations. Interestingly, a
formal fit to the equation for a 1:1 complexation reaction always
is good and allows one to estimate an apparent “binding
constant”, which, however, lacks any meaning and is a function
of total receptor concentration.
The decrease in Kapp and the yield of the deprotonated form
on going to more concentrated receptor solutions is due to higher
concentrations of liberated acetic acid, as was the case in DMSO.
The simulated species distribution diagram for the conditions
of NMR titration predicts the maximum degree of deprotonation
less than 10%. This explains why NMR data show only the
formation of the hydrogen-bonded complex.
We did not find a satisfactory explanation of the fact that
deprotonation by acetate anions is never complete and the
titration profiles tend to “saturate” when approximately half of
the receptor becomes deprotonated (see Figure 8), even at lowest
concentrations of 3. Probably a part of the problem is mentioned
above with possibility of formation of stable R2H- species,
which may be difficult to further deprotonate by acetate in
contrast to much stronger hydroxide base.
Reaction of acetate anions with 4 in MeCN proceeds as a
simple complete deprotonation in accordance with both spec-
trophotometric and NMR titration results. A typical titration
profile at the absorption maximum of the deprotonated form is
shown in Figure 9, and its formal fit to the eq 8 gives Kapp
)
(3.21 ( 0.02) × 103 M-1. The manual fit to a scheme involving
reactions 2 and 3 shown by the dashed line gives K2 ) 0.040
( 0.004 and pKa ) 23.7 for 4 in MeCN. As in DMSO, 4 is a
stronger acid than 3, but the effect is much smaller.
Fitting of the results of NMR titration of 12 mM 4 in MeCN
gives the same K2 value and Kapp ) 60 ( 7 M-1 (Figure 8S,
Supporting Information). The much smaller “binding constant”
in this experiment of course is related to higher receptor
concentration, although in this case a 50-fold decrease in Kapp
is not inversely proportional to a 220-fold increase in the
Experimental Section
N-Methoxyethyl-N′-(4-nitrophenyl)urea 1. To 2-methoxyethy-
lamine (0.50 g, 6.65 mmol) in 20 mL of dry CH2Cl2, was added
4-nitrophenyl isocyanate (1 g, 6.65 mmol), and the mixture was
J. Org. Chem, Vol. 73, No. 6, 2008 2283

Pe´rez-Casas and Yatsimirsky
stirred overnight under nitrogen. The solvent was removed under
reduced pressure and the residue triturated sequentially with MeOH
and hexane. The product was collected by filtration to give 1 (1.13
g) as a pale yellow powder in 71% yield: mp 141-143 °C; IR
Hz, 2H), 8.17 (d, J ) 9.23 Hz, 2H), 9.69 (s, 1H); alkyl protons are
overlapping with the residual water protons of the solvent; 13C NMR
(DMSO-d6) δ 52.38, 59.15, 70.79, 123.48, 124.64, 142.93, 147.27,
181.10; MS (FAB, m/z) 314 (M+); Anal. Calcd for C13H19N3O4S
(313.37): C, 49.83; H, 6.11; N, 13.41. Found: C, 49.81; H, 6.12;
N, 13.43.
1
(KBr) 3330, 1644 cm-1; H NMR (DMSO-d6) δDMSO 3.40 (t, J )
5.56 Hz, 2H), 6.59 (s, 1H), 7.61 (d, J ) 8.14 Hz, 2H), 8.14 (d, J
) 8.11 Hz, 2H), 9.32 (s, 1H); lkyl protons are overlapping with
the residual water protons of the solvent; 13C NMR (DMSO-d6) δ
57.92, 70.94, 116.76, 125,16, 140.41, 147.10, 154.37; MS(FAB,
m/z) 240 (M+). Anal. Calcd for C10H13N3O4 (239.23): C, 50.21;
H, 5.48; N, 17.56. Found: C, 50.20; H,5.48; N,17.55.
Spectrophotometric and 1H NMR Titrations. The absorption
spectra were recorded after additions of aliquots of Bu4NOAc or
Bu4NOH stock solutions in respective solvents to a 10-5-10-4
M
receptor solution in DMSO or MeCN in a quartz cuvette placed in
a compartment of a diode array spectrophotometer thermostated at
25 ( 0.1 °C with a recirculating water bath. NMR titrations were
performed on a 300 MHz spectrometer with more concentrated
stock solutions in the respective deuterated solvents adding aliquots
of them to 5-20 mM receptor solutions in DMSO-d6 or CH3CN-
d3 directly to NMR tubes. In order to test possible autoassociation
of receptors 1-4 their NMR spectra were recorded in a range of
concentrations from 3 to 25 mM in both solvents and signals of all
protons were found unchangeable. Nonlinear least-squares fits of
the experimental results to the eq 8 or its analogue for NMR titration
were performed by using the Microcal Origin version 5 program.
N-Methoxyethyl-N′-(4-nitrophenyl)thiourea 3. A procedure
similar to that used for 1 was applied with a yield of 75%: mp
1
121-123 °C; IR (KBr) 3296, 3198, 1308 cm-1; H NMR (300
MHz DMSO-d6) δDMSO 3.50 (t, J ) 5.19 Hz 2H), 3.68 (br s, 2H),
7.87 (d, J ) 9.16 Hz, 2H), 8.18 (d, J ) 9.17 Hz, 2H), 8.27 (br s,
1H), 10.22 (br s, 1H); alkyl protons are overlapping with the residual
water protons of the solvent; 13C NMR (DMSO-d6) 43.53, 58.00,
69.67, 120.31, 124.51, 141.80, 146.39, 180.09; MS(FAB, m/z) 256
(M+). Anal. Calcd for C10H13N3O3S (255.29): C, 47.05; H, 5.13;
N, 16.46. Found: C, 47.09; H, 5.12; N, 16.47.
N,N-Bis(methoxyethyl)-N′-(4-nitrophenyl)urea 2: mp 109-
1
111 °C; IR (KBr) 3322, 1654 cm-1; H NMR (DMSO-d6) δDMSO
Acknowledgment. Carol Pe´rez-Casas thanks DGAPA UNAM
3.52 (dt, J ) 21.75, 4.85 Hz, 2H), 7.66 (d, J ) 9.36 Hz, 2H), 8.11
(d, J ) 9.31 Hz, 2H), 9.00 (s, 1H); alkyl protons are overlapping
with the residual water protons of the solvent; 13C NMR (DMSO-
d6) δ 47.17, 58.30, 70.72, 118.30, 124.77, 140.83, 124.24, 154.54;
MS (FAB, m/z) 298 (M+). Anal. Calcd for C13H19N3O5 (297.31):
C, 52.52; H, 6.44; N, 14.13. Found: C, 52.55; H,6.43; N,14.13.
N,N-Bis(methoxyethyl)-N′-(4-nitrophenyl)thiourea 4: mp 123-
for the postdoctoral fellowship.
Supporting Information Available: General experimental
1
methods; H and 13C NMR spectra for obtained compounds. This
material is available free of charge via the Internet at http://pubs.
acs.org.
1
125 °C; IR (KBr) 3436, 3268, 1322 cm-1; H NMR(DMSO-d6)
δDMSO 3.64 (t, J ) 5.19 Hz, 2H), 3.98 (br s, 2H), 7.63 (d, J ) 9.28
JO702458F
2284 J. Org. Chem., Vol. 73, No. 6, 2008