Stabilization of Carboxylate Anion with a NH...O Hydrogen Bond: Facilitation of the Deprotonation of Carboxylic Acid by the Neighboring Amide NH Groups

Article abstract of DOI:10.1246/bcsj.77.321

The formation of the NH...O hydrogen bonds of carboxylic acids, 2,6-(t-BuCONH)₂C₆H₃COOH (1) and 2-t-BuCONH-6-MeC₆H₃COOH (2), carboxylate, [NEt ₄][2,6-(t-BuCONH)₂C₆H₃COO] (3), and a mixed complex, [N(n-Pr)₄][H{2,6-(t-BuCONH)₂C ₆H₃(COO)}₂] (4), were determined by X-ray structure analysis, ¹H NMR, and IR spectroscopy, both in the solid state and in solution. The amide NH group forms weak intramolecular hydrogen bond between the NH and O=C group, and no NH...OH hydrogen bond is formed in the carboxylic acid state. Carboxylate anion 3 forms a strong, intramolecular, partially-covalent hydrogen bond between NH...O ^- (anion). The strength of the NH...O hydrogen bonds is also maintained in a solution having a low dielectric constant. The pK_a values for 1 and 2, measured in a micellar solution, indicated that one of the important triggers for deprotonation is a direct interaction toward the oxygen atom of the OH group with the amide NHs.

Full text of DOI:10.1246/bcsj.77.321

Ó 2004 The Chemical Society of Japan
Bull. Chem. Soc. Jpn., 77, 321–329 (2004)
321
Stabilization of Carboxylate Anion with a NHÁÁÁO Hydrogen Bond:
Facilitation of the Deprotonation of Carboxylic Acid
by the Neighboring Amide NH Groups
Akira Onoda, Yusuke Yamada, Jiro Takeda, Yoshiki Nakayama, Taka-aki Okamura,
ꢀ
Mototsugu Doi, Hitoshi Yamamoto, and Norikazu Ueyama
Department of Macromolecular Science, Graduate School of Science, Osaka University,
Machikaneyama-cho 1-1, Toyonaka, Osaka 560-0043
Received June 24, 2003; E-mail: ueyama@chem.sci.osaka-u.ac.jp
The formation of the NHꢁꢁꢁO hydrogen bonds of carboxylic acids, 2,6-(t-BuCONH)2C6H3COOH (1) and 2-t-
BuCONH-6-MeC6H3COOH (2), carboxylate, [NEt4][2,6-(t-BuCONH)2C6H3COO] (3), and a mixed complex, [N(n-
1
Pr)4][H{2,6-(t-BuCONH)2C6H3(COO)}2] (4), were determined by X-ray structure analysis, H NMR, and IR spectrosco-
py, both in the solid state and in solution. The amide NH group forms weak intramolecular hydrogen bond between the NH
and O=C group, and no NHꢁꢁꢁOH hydrogen bond is formed in the carboxylic acid state. Carboxylate anion 3 forms a
ꢂ
strong, intramolecular, partially-covalent hydrogen bond between NHꢁꢁꢁO (anion). The strength of the NHꢁꢁꢁO hydrogen
bonds is also maintained in a solution having a low dielectric constant. The pKa values for 1 and 2, measured in a micellar
solution, indicated that one of the important triggers for deprotonation is a direct interaction toward the oxygen atom of the
OH group with the amide NHs.
The change in the pKa value during an enzymatic reaction is
significant for the regulating of the reactivity in the active cen-
1
ter of aspartate transcarbamylase (ATcase). In general, the pKa
value of Glu or Asp carboxylic acid in proteins has been con-
sidered to depend on the electrostatic interaction around the
COOH moiety. Warshel has reported that the enzyme stabilizes
ionized acids by combing two contributions: (i) the interaction
between the acid charge and the permanent dipoles of the en-
zyme, (VQꢀ), and (ii) the interaction between the acid’s charges
2
Chart 1.
and the induced dipoles of the enzyme, (VQꢀ). It has been the-
oretically demonstrated that such a model is available to the
perturbed pKa of carboxylic acid inside or outside of biological-
large sulfur pꢁ orbital can interact readily with the amide
ꢂ
NH group. A similar NHꢁꢁꢁO hydrogen bond is expected be-
ly important proteins.2–5 The active center of ATcase shows a
different pKa value between an N-(phosphonacetyl)-L-aspar-
tween the carboxylate oxygen and the amide NH, although
the pꢁ of the oxygen atom is smaller than that of the sulfur
ꢂ
6
atom. This paper reports on the properties of NHꢁꢁꢁO hydrogen
tate-binding form or a free one. Thus, we are motivated to ex-
bonds on the corresponding carboxylic acid oxygen using a ser-
ies of amidated benzoate compounds (Chart 1).
perimentally investigate the perturbation to pKa of carboxylic
acid groups by a local electrostatic interaction, such as amide
dipoles, which can be recognized as a model of interactions
with main chain of proteins, using synthetic the model com-
pounds.
Experimental
All solvents were purified by distillation before use. 2,6-Diami-
notoluene was synthesized by a method described in our previous
report.13
We have synthesized benzoic acid with bulky amide groups
at the 2,6-positions for analyses of perturbations from amide
groups to the –COOH moiety. The pKa value of simple carbox-
ylic acids varies with the solvent effect and the electrostatic in-
teraction, including a resonance effect in the case of benzoic
2
,6-(t-BuCONH)2C6H3CH3. To a THF solution (100 mL) of
ꢂ2
2
(
,6-diaminotoluene (2.52 g, 2:06 ꢃ 10 mol) and triethylamine
ꢂ2
12.0 mL, 8:24 ꢃ 10 mol) was added pivaloyl chloride (9.90
ꢂ2
ꢄ
mL, 8:24 ꢃ 10 mol) at 0 C. After 30 min, volatile materials
were removed under reduced pressure. Water was added and white
precipitates obtained were collected by filtration. The precipitates
7
acid derivatives. The contribution of the hydrogen bond to
lowering the pKa value has been proposed for cis-caronic acid
7–10
and salicylic acid.
In these cases, hydroxy groups were
1
were recrystallized by hot methanol. Yield 5.03 g (84%). H NMR
thought to direct to the carboxylic acid oxygen. Recently, we
reported the presence of a NHꢁꢁꢁS hydrogen bond between
amide NH and thiolate sulfur, which has been established using
(
(d, 2H, ArH), 1.95 (s, 3H, Me), 1.22 (s, 9H, t-Bu). C NMR (400
400 MHz, DMSO-d6) ꢂ 8.94 (s, 2H, NH), 8.11 (t, 1H, ArH), 7.00
1
3
MHz, DMSO-d ) ꢂ 175.97, 136.84, 130.79, 124.54, 118.82, 38.57,
6
1
11,12
crystallographic, IR and H NMR analyses.
The relatively
27.36, 12.61.
Published on the web February 10, 2004; DOI 10.1246/bcsj.77.321

3
22 Bull. Chem. Soc. Jpn., 77, No. 2 (2004)
Carboxylate Anion with a NHꢁꢁꢁO Hydrogen Bond
ꢂ4
2
,6-(t-BuCONH)2C6H3COOH (1). To an aqueous solution
tetrapropylammonium acetate (38.2 mg, 1:56 ꢃ 10 mol) were
ꢂ3
(
300 mL) of 2,6-bis(pivaloylamino)toluene (2.00 g, 6:89 ꢃ 10
dissolved in a mixed solvent (10 mL) of methanol/water (50:50).
After evaporation of the solvents, the obtained colorless oil was re-
crystallized from ethyl acetate. Yield 12 mg (19%). Anal. Calcd for
ꢂ2
mol) and magnesium sulfate (0.83 g, 6:90 ꢃ 10 mmol) was add-
ꢂ2
ed guranuated potasium permanganate (2.3 g, 1:46 ꢃ 10 mmol).
After vigorous stirring for 6 h, the brown precipitates were re-
moved by filtration and the filtrate was acidified by 12 M (1 M
C46H75N5O8 (H2O)0:5: C, 66.16; H, 9.17; N, 8.39%. Found: C,
ꢁ
66.02; H, 9.04; N, 8.41%.
t-BuCONHPh. To a THF solution of aniline (1.0 mL, 1:1 ꢃ
ꢂ3
=
1 mol dm ) hydrochloric acid. The obtained white needles were
collected by filtration and washed with water. Yield 1.27 g
57.6%). Anal. Calcd for C17H24N2O4: C, 63.73; H, 7.55; N,
ꢂ2
ꢂ2
10 mol) and Et3N (1.54 mL, 1:10 ꢃ 10 mol) was slowly added
ꢄ
(
pivaloyl chloride at 0 C. After stirring for 1 h, an aqueous
8
.74%. Found: C, 63.56; H, 7.62; N, 8.71%. MS (ESI) Calcd
ꢂ
NaHCO3 solution was added to the solution and THF was evapo-
rated under reduced pressure. The obtained white powder was col-
lected by filtration and recrystallized from hot MeOH. White nee-
dles were obtained (500 mg, 26%). Anal. Calcd for C11H15NO: C,
74.54; H, 8.53; N, 7.90%. Found: C, 74.31; H, 8.56; N, 7.91%.
(found) m=e: 2,6-(t-BuCONH)2C6H3COO , 319.2 (319.2).
1
H NMR (400 MHz, DMSO-d6) ꢂ 11.03 (s, 2H, NH), 7.92 (d,
1
3
2
H, ArH), 7.37 (t, 1H, ArH), 1.20 (s, 9H, t-Bu). C NMR (400
MHz, DMSO-d6) ꢂ 175.79, 169.01, 139.37, 131.22, 116.27,
11.77, 39.41, 27.16.
-t-BuCONH-6-MeC6H3COOH (2). To a THF solution (50
1
1
H NMR (400 MHz, DMSO-d6) ꢂ 9.13 (s, 1H, NH), 7.62 (d, 2H,
o-ArH), 7.27 (t, 2H, m-ArH), 7.02 (t, 1H, p-ArH), 1.23 (s, 9H, t-
2
1
3
mL) of 2-amino-6-methylbenzoic acid (2.0 g, 13 mmol), triethyl-
amine (2.7 mL, 19 mmol), and pivaloyl chloride (2.0 mL, 16
mmol) were added dropwise at 0 C. After stirring overnight at
Bu). C NMR (400 MHz, CDCl3) ꢂ 176.26, 137.89, 128.83,
124.08, 119.84, 39.68, 27.76.
ꢄ
Potentiometric Titration. pH measurements were performed
using a Horiba pH Meter M-8s. All pH standard solution, including
room temperature, the reaction mixture was concentrated under re-
duced pressure to give a brown oil. The oil was dissolved in 200
mL of ethyl acetate and 30 mL of water was added to the solution.
The organic layer was successively washed with water, 2% HCl
aqueous solution, water, and sat. NaCl aqueous solution, and dried
over anhydrous sodium sulfate. The oil, which was obtained by
concentration of the solution, was dissolved in hot n-hexane and
cooled to room temperature. Colorless crystals were obtained
and recrystallized from diethyl ether. Yield 1.43 g (46.0%). Anal.
ꢄ
a 0.1 M NaOH (f ¼ 1:006 at 20 C) solution were purchased from
ꢄ
Nacalai tesque. Each pKa measurement was performed at 21 C (ꢆ
ꢄ
1 C) three times. The concentrations of all solutions were 0.01 M.
An aqueous micellar solution was prepared as follows. The sam-
ples were dissolved in 0.5% volume (from total volume) of a
DMSO solution and a 10% volume of Triton X-100. To the mixed
solution was added water, and stirred at a warmed temperature to
mix homogeneously. The concentration of Triton X-100 micelle
ꢂ4
Calcd for C13H17NO3: C, 66.36; H, 7.28; N, 5.95%. Found: C,
was above a CMC of 2:5 ꢃ 10 M.
1
6
6.31; H, 7.14; N, 5.95%. H NMR (400 MHz, CDCl3) ꢂ 10.11
s, 1H, NH), 9.45 (s, 1H, OH), 8.34 (d, 1H, ArH), 7.39 (t, 1H,
Determination of pKa Values for Various Carboxylic Acids.
The equation below was used to calculate of the pKa values in
aqueous micellar solutions. When arenecarboxylic acid (Ar-
(
ArH), 7.00 (d, 1H, ArH), 2.57 (s, 3H, Me), 1.31 (s, 9H, t-Bu).
1
3
ꢂ
þ
C NMR (400 MHz, DMSO-d6) ꢂ 175.76, 169.29, 137.27,
30.28, 126.13, 124.66, 123.89, 120.04, 39.18, 27.16, 21.42.
COOH) is weak acid, [ArCOO ] is equal to [Na ] with [Ar-
ꢂ
1
COOH] = [ArCOOH]0 ꢂ [ArCOO ]. An equation, Ka
=
ꢂ
þ
þ
4-(t-BuCONH)C6H4COOH. To a THF solution (100 mL) of
[ArCOO ][H ]/[ArCOOH], leads to pKa = pH ꢂ log[Na ] +
ꢂ
4
-aminobenzoic acid (5.0 g, 36 mmol) was slowly added pivaloyl
log{[ArCOOH]0 ꢂ [ArCOO ]}. [ ] and [ ]0 refer to the final con-
ꢄ
chloride (4.4 g, 36 mmol) at 0 C. After stirring for 1 h, 4%
NaHCO3 aqueous solution (100 mL) was added and THF was re-
moved under reduced pressure. After being acidified to pH ꢅ 1 by
centration and the initial concentration, respectively. A 0.1 M
NaOH aquous solution was used to titrate a 0.01 M sample solution
ꢄ
at 21 C.
1
1
0% HCl aqueous solution, products were extracted with ethyl ace-
Physical Measurements. H NMR spectra in solutions were
1
tate. An oil layer was dried over sodium sulfate, and ethyl acetate
was removed under reduced pressure. The residue was recrystal-
lized from hot MeOH. Yield 3.0 g (38%). Anal. Calcd for
taken on a Jeol EX-270 spectrometer. H CRAMPS measurements
were performed on a CMX 300 spectrometer employing the BR 24
pulse sequence and MAS speed in the range of 1–1.5 kHz. IR spec-
tra were recorded on a Jasco FT-IR/8300 spectrometer. Samples
were prepared as KBr pellets or a CH2Cl2 solution.
C12H15NO3: C, 65.14; H, 6.83; N, 6.33%. Found: C, 64.91; H,
1
6
.84; N, 6.39%. H NMR (400 MHz, DMSO-d6) ꢂ 12.63 (s, 1H,
COOH), 9.43 (s, 1H, NH), 7.62 (d, 2H, ArH), 7.73 (d, 2H, ArH),
X-ray Structure and Determination. Single crystals of 2,6-(t-
BuCONH)2C6H3COOH (1), 2-t-BuCONH-6-MeC6H3COOH (2),
[NEt4][2,6-(t-BuCONH)2C6H3COO] (3), and [N(n-Pr)4][H{2,6-
(t-BuCONH)2C6H3(COO)}2] (4) were sealed in a glass capillary
for X-ray measurements. The measurements were performed at
1
1
.23 (s, 9H, t-Bu). 13C NMR (400 MHz, DMSO-d6) ꢂ 176.49,
66.55, 143.22, 129.75, 124.70, 119.01, 39.32, 27.02.
[
NEt4][2,6-(t-BuCONH)2C6H3COO] (3). Tetraethylammoni-
um acetate tetrahydrate (26.1 mg, 11.0 mmol) and 2,6-bis(pival-
oylamide)benzoic acid (32.0 mg, 10.0 mmol) were dissolved in
1
moved under reduced pressure. The obtained residue was dissolved
in a small amount of ethyl acetate to give colorless needles in 27%
yield (12 mg). Anal. Calcd for C25H43N3O4: C, 66.78; H, 9.64; N,
ꢄ
23 C on a Rigaku AFC7R or AFC5R diffractometer equipped with
a rotating anode X-ray generator. The radiation used was Mo Kꢃ
0 mL of MeOH. After stirring for 30 min, the solvents were re-
ꢀ
monochromatized with graphite (0.71069 A). The unit cell dimen-
sions were refined by 25 reflections. These standard reflections
were chosen and monitored with every 150 reflections, and did
not show any significant change. The structures were solved by a
direct method and expanded using Fourier techniques using teXsan
crystallographic software14 and SHELXL-97.15 All non-hydrogen
atoms were refined anisotropically. Hydrogen atoms were refined
only for their positions. Crystal data for 1: C17H24N2O4, 320.38,
monoclinic, space group P21=a, with a ¼ 8:894ð2Þ, b ¼
1
9
.35%. Found: C, 66.33; H, 9.69; N, 9.25%. H NMR (400 MHz,
DMSO-d6) ꢂ 12.26 (s, 1H, NH), 8.03 (d, 2H, m-ArH), 7.26 (t,
1
1
1
H, p-ArH), 3.19 (q, 8H, –CH2–), 1.23 (s, 9H, t-Bu), 1.15 (m,
1
3
2H, –CH3). C NMR (400 MHz, DMSO-d6) ꢂ 175.83, 169.62,
40.21, 130.18, 114.80, 111.57, 51.34, 38.37, 27.28, 7.08.
[
N(n-Pr)4][H{2,6-(t-BuCONH)2C6H3(COO)}2] (4).
2,6-
ꢂ4
ꢀ
Bis(pivaloylamino)benzoic acid (100 mg, 3:12 ꢃ 10 mol) and
15:095ð3Þ, c ¼ 13:415ð2Þ A, ꢃ ¼ 90:00, ꢄ ¼ 106:267ð15Þ,

A. Onoda et al.
Bull. Chem. Soc. Jpn., 77, No. 2 (2004)
323
a)
b)
H24
O12
H17
O3
O2
O11
C12
H1
H1
C1
H2
N2
C11
N1
C11
C16
C17
N1
C17
C27
O2
C12
O1
O1
c)
d)
C27
O2
N3
N2
H2
C16
O4
C11
O11
O12
H40
C12
H1
N1
H2
N2
C1
C11
N1
O3
H1
O1
C17
C17
O1
C27
O2
C12
C16
Fig. 1. ORTEP drawings of a) 2,6-(t-BuCONH)2C6H3COOH (1), b) 2-t-BuCONH-6-MeC6H3COOH (2), c) anion part of
NEt4][2,6-(t-BuCONH)2C6H3COO] (3), and d) [N(n-Pr)4][H{2,6-(t-BuCONH)2C6H3(COO)}2] (4).
[
3
3
ꢀ
ꢅ ¼ 90:00, V ¼ 1728:9ð6Þ A , Z ¼ 4, Dcalc ¼ 1:231 g/cm , no.
reflns used = 3970 (all data), R1 ¼ 0:061, wR2 ¼ 0:209, and
GOF = 1.12. Crystal data for 2: C13H17NO3, 235.28, monoclinic,
carboxylates.
In order to investigate the structural dif-
ference between carboxylic acid and carboxylate, the crystal
structures of 2,6-(t-BuCONH)2C6H3COOH (1) and 2-t-
BuCONH-6-MeC6H3COOH (2) as carboxylic acids, and
space group P21=a, with a ¼ 10:8443ð14Þ, b ¼ 10:4408ð14Þ, c ¼
ꢀ
1
1
2
1:5919ð10Þ A, ꢃ ¼ 90:00, ꢄ ¼ 93:943ð9Þ, ꢅ ¼ 90:00, V ¼
[
[
NEt4][2,6-(t-BuCONH)2C6H3COO] (3) as a carboxylate and
N(n-Pr)4][H{2,6-(t-BuCONH)2C6H3(COO)}2] (4) as a mixed
3
3
ꢀ
309:4ð3Þ A , Z ¼ 4, Dcalc ¼ 1:194 g/cm , no. reflns used =
569 (all data), R1 ¼ 0:063, wR2 ¼ 0:198, and GOF = 0.94. Crys-
complex of both states, were determined by X-ray analysis. The
location of amide groups near carboxylic acid or carboxylate
was shown to promote the formation of a NHꢁꢁꢁO hydrogen
bond between an oxygen atom and an amide NH. Figure 1
shows the crystal structures of 1, 2, 3, and 4. The selected bond
distances and bond angles are listed in Table 1.
tal data for 3: C25H43N3O4, 449.62, orthorhombic, space group
ꢀ
P212121, with a ¼ 15:136ð3Þ, b ¼ 17:019ð4Þ, c ¼ 10:417ð3Þ A,
3
3
ꢀ
V ¼ 2683:4ð11Þ A , Z ¼ 4, Dcalc ¼ 1:113 g/cm , no. reflns used
=
2973 (all data), R1 ¼ 0:057, wR2 ¼ 0:207, and GOF = 0.95.
Crystal data for 4: C23H37:50N2:50O4, 413.05, monoclinic, space
group C2=c, with a ¼ 30:838ð4Þ, b ¼ 12:063ð4Þ, c ¼ 13:793ð4Þ
3
The crystal structure of a carboxylic acid, 1, with two amide
0
ꢀ
ꢀ
A, ꢃ ¼ 90:00, ꢄ ¼ 104:618ð19Þ, ꢅ ¼ 90:00, V ¼ 4965ð2Þ A ,
3
groups at the o,o -positions has one C–O single bond (1.319(3)
Z ¼ 8, Dcalc ¼ 1:105 g/cm , no. reflns used = 3246 (all data),
ꢀ
ꢀ
A) and one C=O double bond (1.205(3) A). The two amide
R1 ¼ 0:070, wR2 ¼ 0:208, and GOF = 1.19.
NHs direct to each oxygen atom of carboxylic acids with two
ꢀ
Molecular Orbital Calculations. Ab initio molecular orbital
calculations for the carboxylic acid, 2-(t-BuCONH)C6H4COOH,
different NꢁꢁꢁO distances (2.576(2) and 2.614(3) A), corre-
ꢂ
sponding to the two different amide C–O bond distances. The
ꢀ
long C17=O1 distance (1.227(3) A) is due to the presence of
and the corresponding carboxylate, 2-(t-BuCONH)C6H4COO ,
were carried out with the assumption of an idealized coplanar
structure between amide and carboxylic acid or carboxylate planes
using the Gaussian 98 system of programs.16 Single point calcu-
lations with HF and MP2 levels of theory were performed using
the 6-31+G basis set. DFT calculations using Becke’s three-pa-
rameter hybrid functional with the correlation functional of Lee,
Yang, and Parr (B3LYP)17–19 were also performed using 6-
0
intramolecular N1–H1ꢁꢁꢁO11=C1 and intermolecular O12 –
0
H24 ꢁꢁꢁO1=C17 hydrogen bonds. The difference between the
ꢀ
ꢀ
N1–C17 (1.341(3) A) and N2–C27 (1.361(3) A) bond distances
is caused by a large bond order of the conjugated N1–C17 bond,
involving the inter- and intramolecular hydrogen bonds of the
amide group. The results indicate that the hydrogen-bond inter-
action of N2–H2ꢁꢁꢁO12H is weaker than that of N1–
H1ꢁꢁꢁO11=C1 in 1.
ꢀ
ꢀ
ꢀꢀ
31+G basis set. An estimation of the solvation energy in aque-
20,21
ous solution was taken into account by using the PCM method.
The crystal structure of 2, with one amide group at the o-po-
sition of the aromatic ring, indicates the location of the amide
group towards the C=O of the carboxyl group. The C–O bond
Results and Discussion
Crystal Structures of Arenecarboxylic Acids and Arene-

3
24 Bull. Chem. Soc. Jpn., 77, No. 2 (2004)
Carboxylate Anion with a NHꢁꢁꢁO Hydrogen Bond
Table 1. Selected Structural Parameters for 1, 2, 3, and 4
1
2
3
4
Carboxyl group
C1–O11
C1–O12
C1–O11
C1–O12
1.205(3)
1.319(3)
C1–O2
C1–O3
1.205(4)
1.294(4)
1.239(7)
1.244(8)
C1–O3
C1–O4
1.225(5)
1.289(5)
Amide group
C17–O1
C27–O2
N1–C17
N2–C27
1.227(3)
1.209(3)
1.341(3)
1.361(3)
C17–O1
N1–C17
1.224(4)
C17–O1
C27–O2
N1–C17
N2–C27
1.211(8)
1.230(8)
1.346(8)
1.340(8)
C17–O1
C27–O2
N1–C17
N2–C27
1.218(6)
1.204(6)
1.355(6)
1.350(6)
1.340(4)
Hydrogen bond distances and angles
N1ꢁꢁꢁO11
N2ꢁꢁꢁO12
2.576(2)
2.614(3)
140.9
N1ꢁꢁꢁO2
2.648(3)
N1ꢁꢁꢁO11
2.552(7)
2.569(7)
136.8
N1ꢁꢁꢁO3
2.613(4)
2.639(4)
132.5
N2ꢁꢁꢁO12
N2ꢁꢁꢁO4
N1H1ꢁꢁꢁO11
N2H2ꢁꢁꢁO12
N1ꢁꢁꢁO2
128.1
N1H1ꢁꢁꢁO11
N1H1ꢁꢁꢁO3
N2H2ꢁꢁꢁO4
135.2
N2H2ꢁꢁꢁO12
135.6
136.3
Angle between benzene ring and COO plane
29.8 15.7
1
7.8
26.5
Angle between benzene ring and amide plane
N1 amide
N2 amide
3.8
21.9
33.7
11.8
19.3
26.8
17.1
ꢀ
distances of carboxylic acid are 1.205(4) A for C1=O2 and
1
ꢂ1
ꢀ ꢀ
.294(4) A for C1–O3 in 2. The difference of 0.089 A in the
Table 2. Selected IR Bands (cm ) in the Amide Region of
1, 2, 3, and 4
two C–O bond distances indicates that the carboxylic acid con-
sists of one C=O double bond and one C–OH single bond. The
amide NH directs to the C=O oxygen atom to form an
NHꢁꢁꢁO=C hydrogen bond with an appropriate NꢁꢁꢁO distance
ꢆ(NH)
Solution
ꢆ(CO) (amide)
aÞ
aÞ
Solid
Solid
Solution
ꢀ
1
2
3
4
3407
3330
3024
3269
3444
3442
1694, 1636
1639
1664
1686
1688
1662
of 2.648(3) A.
The crystal structure of the anion part of 3, with two amide
0
bÞ
—
—
groups at the o,o -positions, indicates that two carboxylate C–O
bond distances, C1–O11 and C1–O12, are 1.239(7) and
cÞ
cÞ
1676
—
ꢀ
.244(8) A, respectively. This is because conjugation of the
1
a) In 10 mM dichloromethane solution at room temperature. b)
NH signal was broad and unassignable. c) Complex 4 decom-
poses in the solution.
ꢂ
two C–O bonds in COO anion. The two amide NHs form
NHꢁꢁꢁO–C hydrogen bonds to both oxygen atoms of the carbox-
ylate. Comparing the bond distances of both amide groups, the
ꢀ
ꢀ
bond length of C17=O (1.211(8) A) and C27=O2 (1.230(8) A)
are in a similar range of the N2 amide CO bond of 1. The N1–
NH and C=O amide regions for carboxylic acids, 2,6-(t-Bu-
CONH)2C6H3COOH (1) and 2-t-BuCONH-6-MeC6H3COOH
(2), and carboxylate, [NEt4][2,6-(t-BuCONH)2C6H3COO] (3)
in the solid state, compared with those IR bands in a dichloro-
ꢀ
C17 bond (1.346(8) A) and the N2–C27 bond distance
ꢀ
1.340(8) A) also indicate that both amide groups are involved
(
in NHꢁꢁꢁO hydrogen bonds. The torsion angles of N1H1ꢁꢁꢁ
methane solution. The free amide NH bands are known to ap-
ꢂ1 37
ꢄ
O11C1 and N2H2ꢁꢁꢁO12C1 are 6.1 and ꢂ33:3 , respectively,
pear in the range of 3400–3500 cm
ꢂ1
.
at 3407 cm in the solid state. The two amide C=O bands are
1 shows a free NH band
indicating that the N2H2ꢁꢁꢁO12 hydrogen bond gives a more
preferable geometry than the N1H1ꢁꢁꢁO11 one.
ꢂ1
observed at 1694 and 1636 cm . The former band is assigned
Figure 1d shows the crystal structure of a mixed complex,
to the amide whose NH is not hydrogen-bonded with carboxyl-
ic acid C=O. The latter band is assigned to the other, which is
intramolecularly hydrogen-bonded with carboxylic acid C=O,
and also intermolecularly hydrogen-bonded with O–HꢁꢁꢁO=C
interaction. 2 exhibits slightly shifted NH bands at 3330
[N(n-Pr)4][H{2,6-(t-BuCONH)2C6H3(COO)}2] (4), containing
both carboxylic acid and carboxylate states. Two carboxyl
ꢀ
groups share the same proton with a distance of 1.228(3) A.
ꢀ
Such a short OꢁꢁꢁO distance of 2.456 A clearly indicates that
ꢂ1
the complex forms a low-barrier hydrogen bond consisting of
a proton intermolecularly located between the two oxygen
atoms of the C–O anion groups in each carboxylate. Actually,
cm assignable to a weak NHꢁꢁꢁO=C (carboxylic acid) hydro-
ꢂ1
gen bond. The amide ꢆ(CO) bands are observed at 1639 cm
,
due to the amide C=O band participating in the intramolecular
NHꢁꢁꢁO=C (carboxylic acid) hydrogen bond and intermolecu-
larly OHꢁꢁꢁO=C hydrogen bond. In the carboxylate anion state,
ꢀ
the bond distance of C1–O3(H) (1.225(5) A) and C1–O4
ꢀ
(1.289(5) A) lies between those of C–O(H) and C–O bond in
ꢂ1
1
for various carboxylic acids compounds.
and 3. Similar low barrier hydrogen bonds have been found
the ꢆ(NH) band appears at 3024 cm , indicating that the pres-
22–36
ence of the strong NHꢁꢁꢁO hydrogen bonds in the solid state. 4
ꢂ1
IR Spectra of Carboxylic Acid and Carboxylate in the
Solid State and Solution. Table 2 lists the IR bands in the
also exhibits a shifted ꢆ(NH) band at 3269 cm . Thus, the IR
results clearly corroborate that the strength of the NHꢁꢁꢁO hy-

A. Onoda et al.
Bull. Chem. Soc. Jpn., 77, No. 2 (2004)
325
drogen bonds depends on the basicity of the carboxyl group,
even though little geometrical difference has been found in
X-ray analyses.
a)
The IR spectra of 1, 2, and 3 in 10 mM dichloromethane
were obtained at room temperature in order to investigate the
intramolecular interaction. A standard amide compound,
OH
NH
NH
ꢂ1
PhNHCO-t-Bu, gives a free amide NH band at 3451 cm
ꢂ1
and a corresponding free CO band at 1682 cm , without an in-
ter- or intramolecular interaction in the solution. Carboxylic
ꢂ1
acid, 2, indicates an NH band at 3442 cm and a CO band
ꢂ1
b)
at 1688 cm , whereas 1 also exhibits two bands at 3444 and
ꢂ1
1
686 cm . These bands of 1 and 2 in dichloromethane indicate
OH
NH
that the amide NH and CO groups are not involved in any
strong hydrogen bonding. We observed OH (carboxylate)
ꢂ1
stretching bands for 1 and 2 at ca. 3350 and 3500 cm , respec-
tively, as broad bands. Thus, the presence of an amide NH in-
teraction increases the acidity of the carboxyl OH proton. The
NH stretching in carboxylate 3 could not be detected in the
ꢂ1
2
anion appears at 1662 cm , shifting 20 cm from that of the
800–3600 cm region. The CO band of 3 in the carboxylate
ꢂ1
ꢂ1
c)
ꢂ1
standard PhNHCO-t-Bu. The 20 cm shift is due to the forma-
3
8
tion of a strong NHꢁꢁꢁO=C (carboxylate) hydrogen bond.
1
H NMR Spectra (CRAMPS) in the Solid State. Figure 2
shows the 1H NMR spectra of carboxylic acids, 2,6-(t-Bu-
CONH)2C6H3COOH (1) and 2-t-BuCONH-6-MeC6H3COOH
NH
(
2), and carboxylate, [NEt4][2,6-(t-BuCONH)2C6H3COO] (3)
and [N(n-Pr)4][H{2,6-(t-BuCONH)2C6H3(COO)}2] (4) in the
solid state. 1 shows two amide NH signals at 10.2 and 12.5
ppm assignable to the non-hydrogen bonding amide NH and
the intramolecularly NHꢁꢁꢁO=C hydrogen-bonded NH groups,
respectively. A carboxylic acid OH signal was observed at
d)
1
4.2 ppm. Similarly, an amide NH signal in 2 was observed
OH
20
NH
at 11.6 ppm, which is slightly involved in a weak intramolecu-
lar NHꢁꢁꢁO=C hydrogen bond. A carboxylic acid OH signal ap-
pears at 14.1 ppm, that is similar to a common chemical shift
for the carboxylic acid proton in the solution. Carboxylate 3 ex-
hibits one broad signal at 14.7 ppm due to the intramolecularly
hydrogen-bonded amide NH group in the solid state. Complex
25
15
10
ppm
5
0
-5
4
, consisting of carboxylic acid and carboxylate, shows a weak
1
hydrogen-bonded amide NH signal at 12.7 ppm and a shifted
OH signal at 20.2 ppm. This supports the presence of a low-bar-
rier hydrogen-bonded proton, as reported by the crystallograph-
ic analysis of various carboxylate–carboxylic acid complexes.
Thus, the low field shift of amide NH signals reflects the acidity
of the amide proton forming an NHꢁꢁꢁX hydrogen bond, as well
as the low field shift of OH signal with the increased acidity.
Fig. 2. H NMR CRAMPS spectra of a) 2,6-(t-BuCONH)2-
C H COOH (1), b) 2-t-BuCONH-6-MeC H COOH (2),
c) [NEt ][2,6-(t-BuCONH) C H COO] (3), and d) [N(n-
6
3
6
3
4
2
6
3
Pr)4][H{2,6-(t-BuCONH)2C6H3(COO)}2] (4) in the solid
state.
which exhibits no participation of a intramolecular hydrogen
39
1
The H shift of amide NH signal reflects extent of the hydrogen
bond of the carboxylic acid and the carboxylate.
bond. 2-t-BuCONH-6-MeC6H3COOH (2) also shows a large
ꢂ1
value of 0.0038 ppm K without a hydrogen bond. No partic-
1H NMR Spectra in Solution. In order to examine whether
amide NH in carboxylic acids 1 and 2 forms an NHꢁꢁꢁOH hydro-
gen bond in carboxylic COOH, the 1H NMR spectra were
measured in chloroform-d, which is a hydrogen bond-support-
ing solvent with a relatively low dielectric constant (" ¼ 7).
ipation of a NHꢁꢁꢁO (carboxylic acid COOH) hydrogen bond in
solution is consistent with the results of IR spectra in the amide
region.
On the other hand, carboxylate, [NEt4][2,6-(t-Bu-
CONH)2C6H3COO] (3), gives a smaller gradient value of
1
ꢂ1
The chemical shift of the amide NH H NMR signal is one of
0.00078 ppm K than a reported common gradient value for
the appropriate monitors for the detection of various hydrogen
bonds involving the amide NH. The temperature-dependent
gradient for 2,6-(t-BuCONH)2C6H3COOH (1) is estimated in
a linear peptide. The amide NH chemical shift of 3, (13.53
ppm), shifts downfield from that of 1, (10.19 ppm), as shown
in Fig. 3. These results indicate that 3 has a strong NHꢁꢁꢁO hy-
drogen bond in solution. Thus, the IR results in the solid state
and in solution reveal that the amide NH in carboxylic acid
ꢂ1
ꢄ
0
CDCl3 for a linear peptide is known to be 0.0024 ppm K
.0058 ppm K in the range of ꢂ30–50 C. The value in
ꢂ1
,

3
26 Bull. Chem. Soc. Jpn., 77, No. 2 (2004)
Carboxylate Anion with a NHꢁꢁꢁO Hydrogen Bond
Chart 2.
1
Fig. 3. H NMR spectra of a) 2,6-(t-BuCONH)2C6H3-
COOH (1) and b) [NEt4][2,6-(t-BuCONH)2C6H3COO]
(
3) in 10 mM CDCl3 solution.
Table 3. Selected Results of Ab initio Calculations
In gas phase
Charge
In aqueous solution
Overlap
population
Charge
Overlap
population
Oa
Ob
H (N)
a
O ꢁꢁꢁH (N)
a
a
Oa
Ob
H (N)
a
O ꢁꢁꢁH (N)
a
a
Models
5a
HF
MP2
B3LYP
HF
MP2
B3LYP
HF
MP2
B3LYP
HF
ꢂ0:734
ꢂ0:732
ꢂ0:590
ꢂ0:609
ꢂ0:609
ꢂ0:514
ꢂ0:863
ꢂ0:860
ꢂ0:684
ꢂ0:822
ꢂ0:824
ꢂ0:679
ꢂ0:536
ꢂ0:535
ꢂ0:442
ꢂ0:585
ꢂ0:585
ꢂ0:457
ꢂ0:648
ꢂ0:649
ꢂ0:540
ꢂ0:675
ꢂ0:674
ꢂ0:534
0.451
0.451
0.371
0.444
0.444
0.365
0.461
0.461
0.364
0.449
0.450
0.353
ꢂ0:0262
ꢂ0:0259
0.0020
0.0050
0.0050
0.0270
0.0666
0.0688
0.0875
0.0369
0.0353
0.0590
ꢂ0:764
ꢂ0:764
ꢂ0:627
ꢂ0:666
ꢂ0:666
ꢂ0:572
ꢂ0:999
ꢂ1:000
ꢂ0:842
ꢂ1:017
ꢂ1:017
ꢂ0:898
ꢂ0:607
ꢂ0:608
ꢂ0:508
ꢂ0:616
ꢂ0:616
ꢂ0:486
ꢂ0:758
ꢂ0:758
ꢂ0:661
ꢂ0:741
ꢂ0:741
ꢂ0:612
0.486
0.486
0.403
0.465
0.460
0.380
0.472
0.472
0.377
0.466
0.466
0.383
ꢂ0:0250
ꢂ0:0250
0.0036
0.0035
0.0035
0.0258
0.0273
0.0274
0.0549
0.0070
0.0070
0.0300
5b
6a
6b
MP2
B3LYP
state does not form a NHꢁꢁꢁO hydrogen bond, but the amide NH
in carboxylate anion state can inherently make a hydrogen
bond in solution. Furthermore, the carboxylic acid in the solid
state gives a weak NHꢁꢁꢁO=C hydrogen bond, but the bond is
not found in solution.
phase and in aqueous solution. The calculated overlap popula-
tions of the electron density for 5a and 5b, which are in the car-
boxylic acid state, are 0.0020 and 0.0270, respectively, in DFT
methods, whereas those of 6a and 6b in the carboxylate anion
state are 0.0875 and 0.0590, respectively. Similar tendencies
are also observed in HF and MP2 methods. The values of the
overlap population of 5a and 5b (in carboxylic acid state) in
aqueous solution are almost similar to that in the gas phase.
The carboxylate anion state (6a and 6b) gives a relatively
stronger covalency than that of the carboxylic acid state. The
covalency in aqueous solution was, however, decreased com-
pared with a gas phase. In a polar solvent, stabilization of the
carboxylate can also work but the random orientation of the di-
polar interaction of the polar solvent can not be as effective as
the amide groups with a readily formable orientation. The re-
sults indicate that the intramolecular NHꢁꢁꢁO hydrogen bond in-
teraction from amide NH is interfered by solvation, which can
Ab initio Calculations. Ab initio molecular orbital calcu-
lations of 2-(t-BuCONH)C6H4COOH (5a, 5b) and {2-(t-Bu-
CONH)C6H4COO} (6a, 6b) were carried out to investigate
the NHꢁꢁꢁO hydrogen bonding interaction. The calculation
ꢀ
ꢀ
was performed using the 6-31+G base set to obtain the over-
lap population for the carboxylic acid and the carboxylate as
simplified models. Although the crystal structure shows that
the amide plane is not coplanar to the aromatic ring, we use
model structures fixing the carboxyl and amide plane to a sim-
ilar plane of the benzene ring (Chart 2). The results are summa-
rized in Table 3. Increasing overlaps due to hydrogen bonding
were observed in the carboxylate anion state both in the gas

A. Onoda et al.
Bull. Chem. Soc. Jpn., 77, No. 2 (2004)
327
be explained by the previous report.40 The increased orbital
overlaps between (N)HꢁꢁꢁO indicate the presence of a partial
covalency between the amide NH proton and the O atom of
the carboxylate anion.
value of 2,6-(MeCONH)C6H3COOH is 3.2, which shifts 0.8
unit lower than that of C6H5COOH (4.0). Thus, we have also
observed that NHꢁꢁꢁO hydrogen bonds lower the pKa values in
an aqueous solution. Generally, the pKa values shift higher in
the more hydrophobic environment. Actually, the benzoic acid
derivatives without hydrogen bonds in an aqueous micellar so-
lution exhibit around 1 unit higher pKa than in an aqueous solu-
tion. For example, the pKa value of 4-(t-BuCONH)C6H4COOH
is 5.4 in a micellar solution and that of 4-(MeCONH)-
C6H4COOH is 4.1 in an aqueous solution. However, the hydro-
gen-bonded derivative, 1, shows similar pKa values to that in an
aqueous solution even though in a hydrophobic environment.
The results indicate that the pKa lowering effect of the NHꢁꢁꢁO
hydrogen bond is stronger under hydrophobic circumstances.
Formation of a NHÁÁÁO Hydrogen Bond between Amide
NH and Carboxylate Anion. The crystallographic analysis,
pKa Shift by NHÁÁÁO Hydrogen Bond. The pKa values of
various o-substituted benzoic acids (ArCOOH), such as 2,4,6-
(CH3)3C6H2COOH without an amide group, 2-t-BuCONH-6-
MeC6H3COOH (2) with one amide group, and 2,6-(t-Bu-
CONH)2C6H3COOH (1) with two amide groups near the car-
boxylic acid group, were then measured. The pKa measure-
ments were carried out in a micellar solution in order to exam-
ine the effect of amide NH on the pKa shift of the carboxylic
acid. A water-insoluble carboxylic acid allowed us to measure
the pKa value in an aqueous micellar solution. Although the mi-
cellar solution consisted of two heterogeneous layers: (hydro-
phobic and hydrophilic), a titration method was utilized be-
cause of rapid proton transfer. Thus, the effect of the NHꢁꢁꢁO hy-
drogen bond formed in the hydrophobic layer of the micellar
solution can be detected. Figure 4 summarizes the pKa values
of these carboxylic acids in an aqueous solution and in an aque-
ous micellar solution (10% Triton, 0.5% DMF). The pKa values
of C6H5COOH and 2,4,6-(CH3)3C6H2COOH are 4.6 and 4.8,
respectively. p-Substituted 4-(t-BuCONH)C6H4COOH gives
the value of 5.4. 1 and 2 which can form single and double hy-
drogen bond show lower shifted pKa values of 3.1 and 3.9, re-
spectively. Therefore, the experiment reveals that the o-substi-
tuted amide groups lower the pKa values of the carboxylic acid
group.
1
IR spectra, and the solid H NMR spectra (CRAMPS) of two
carboxylic acids, 2,6-(t-BuCONH)2C6H3COOH (1) and 2-t-
BuCONH-6-MeC6H3COOH (2), indicate that the amide NH
group forms a very weak intramolecular hydrogen bond be-
tween the NH and O=C group and the NHꢁꢁꢁOH hydrogen bond
in the solid state. The presence of a weak NHꢁꢁꢁOH hydrogen
bond of 1 and 2 is supported by the small bond order and a
somewhat positively charged OH oxygen obtained from ab ini-
ꢂ
tio MO calculations. The large bond order between NH and O
ꢂ
(anion) atoms was due to charge transfer from the O (anion)
oxygen, mainly to amide groups as estimated by the ab initio
MO calculations. The carboxylate anion 3 has a strong, intra-
molecular, partially covalent hydrogen bond between the
We measured the pKa values with the corresponding acetyl-
amino derivatives13 soluble in an aqueous solution. The pKa
ꢂ
NHꢁꢁꢁO (anion), as demonstrated by crystallographic, IR and
1
solid H NMR data. In a solution consisting of a low dielectric
constant solvent, the solution structures of 1 and 3 by H NMR
and IR spectroscopic analyses also indicate the formation of
1
In aqueous micellar soluton
H
O
O
ꢂ
H
O
NHꢁꢁꢁO (anion) hydrogen bond in a hydrogen bond-supporting
O
H
O
H
O
H
O
solvent, such as chloroform or dichloromethane.
O
O
O
H
N
H
N
H
N
Lower Shift of pKa Value by Neighboring Amide NH to
Carboxylic OH in Aqueous Micellar Solution. The pKa val-
ues of various carboxylic acids were determined in an aqueous
solution to obtain an accurate, reversible, deprotonating point
of COOH. However, to evaluate the effect of amide group ad-
jacent to the COOH group in the 2,6-substituted benzoic acids,
a low dielectric constant solvent is crucial during deprotonation
because of the necessity of maintaining the NHꢁꢁꢁO hydrogen
bond. Therefore, our pKa measurements were carried out in a
t-Bu
t-Bu
t-Bu
N
O
H
O
O
O
t-Bu
3.1
3.9
4.6
4.8
5.4
3
4
5
6
pKa
1
0% Triton X-100 aqueous micellar solution. Such a micellar
In aqueous soluton
H
O
solution addresses the above two requirements because a proton
diffusion rate from micelles to the water part in an aqueous mi-
cellar solution occurs quickly enough to record an accurate pKa
O
H
H
O
O
O
O
H
N
H
N
41,42
value.
In addition, the inner part in the micelles can support
Me
Me
N
O
H
ꢂ
the NHꢁꢁꢁO and NHꢁꢁꢁO=C hydrogen bonds in hydrophbic en-
vironments. The significant pKa values of 1 and 2 in an aqueous
micellar solution are associated with the formation of the
NHꢁꢁꢁOꢂ (anion) hydrogen bond after deprotonation. In this
case, the location of amide NH near the carboxylic acid OH
oxygen facilitates deprotonation of the COOH. In our com-
plexes, the amide group at the o-position allows an evaluation
of the chemical function of the NHꢁꢁꢁO hydrogen bond. This is
because the electronic effect of the amide group is negligible
(zero of Hammett ꢇp) when the p-position is substituted.
O
O
Me
3.2
4.0
4.1
3
4
5
6
pKa
Fig. 4. Summary of pKa values of series of benzoic acid
derivatives.

3
28 Bull. Chem. Soc. Jpn., 77, No. 2 (2004)
Carboxylate Anion with a NHꢁꢁꢁO Hydrogen Bond
One of the important triggers for deprotonation is a direct in-
2568 (1934).
11 N. Ueyama, N. Nishikawa, Y. Yamada, T. Okamura, S.
Oka, H. Sakurai, and A. Nakamura, Inorg. Chem., 37, 2415 (1998).
teraction toward the oxygen atom of the OH group with the
amide NH. Of course, the hydrogen bonding or electrostatic in-
teraction of OH with an electron donor, like OHꢁꢁꢁDonor is also
1
2
N. Ueyama, Y. Yamada, T. Okamura, S. Kimura, and A.
Nakamura, Inorg. Chem., 35, 6471 (1996).
Y. Yamada, N. Ueyama, T. Okamura, W. Mori, and A.
Nakamura, Inog. Chim. Acta, 275–276, 43 (1998).
14 teXsan: Crystal Structure Analysis Package, Molecular
Structure Corporation, 1985 & 1999.
G. M. Sheldrick, ‘‘SHELXL-97, Program for the Refine-
ment of Crystal ed,’’ Univeristy of Gottingen, Germany (1997).
M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
an important factor in facilitating deprotonation.4
3,44
Shan and
1
3
Herschlag have pointed out the importance of a charge rear-
rangement during deprotonation, in DMSO, of salicylic acid
with the hydrogen bond by the adjacent OH.45 Depronation fa-
cilitation of carboxylic acid by neighboring amide NH groups is
also thought to be important in understanding the controlled
pKa values of carboxylic acids in proteins.
1
5
1
6
M. A. Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A.
Montgomery, Jr., R. E. Stratmann, J. C. Burant, S. Dapprich, J.
M. Millam, A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas,
J. Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C.
Pomelli, C. Adamo, S. Clifford, J. Ochterski, G. A. Petersson, P.
Y. Ayalo, Q. Cui, K. Morokuma, D. K. Malick, A. D. Rabuck,
K. Raghavachari, J. B. Foresman, J. I. Cioslowski, J. V. Oritiz,
B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi,
R. Gomperts, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham,
C. Y. Peng, A. Nanayakkara, C. Gonzalez, M. Challacombe, P. M.
W. Gill, B. Johnson, W. Chen, M. W. Wong, J. L. Andres, M.
Head-Gordon, E. S. Replogle, and J. A. Pople, ‘‘Gaussian 98,
Revision A.1,’’ Gaussian, Inc., Pittsburgh, PA (1998).
Conclusion
The formation of the NHꢁꢁꢁO hydrogen bonds of carboxylic
acids, 2,6-(t-BuCONH)2C6H3COOH (1) and 2-t-BuCONH-6-
MeC6H3COOH (2), carboxylate, [NEt4][2,6-(t-BuCONH)2-
C6H3COO] (3) was determined both in the solid state and in
solution. Although an X-ray analysis indicates that the amide
NHs intramolecularly interact with O=C of the carboxyl group
in the –COOH state, the NH protons are not involved in the
ꢂ
strong NHꢁꢁꢁO hydrogen bonds. The –COO state only gives
ꢂ
the intramolecular, partially-covalent, strong NHꢁꢁꢁO (anion)
ꢂ
hydrogen bond. The strength of the NHꢁꢁꢁO hydrogen bonds
is also maintained in a solution with a low dielectric constant.
The pKa values for 1 and 2, measured in a micellar solution, in-
dicated that a direct interaction toward the oxygen atom of OH
group with the amide NH is important for lowering the pKa
values.
1
1988).
7
A. D. Becke, Phys. Rev. A: At., Mol., Opt. Phys., 38, 3098
(
1
1
8
9
A. D. Becke, J. Chem. Phys., 98, 5648 (1993).
C. Lee, W. Yang, and R. G. Parr, Phys. Rev. B: Condens.
Matter, 37, 7685 (1988).
S. Miertus, E. Scrocco, and J. Tomasi, J. Chem. Phys., 55,
17 (1981).
2
0
Support of this work by JSPS Fellowships [for A. O.; grant
306 (1999–2002)] and a Grant-in-Aid for Scientific Research
on Priority Area (A) (No 10146231) from the Ministry of Edu-
cation, Culture, Sports, Science and Technology, is gratefully
acknowledged.
1
2
21
22
23
24
S. Miertus and J. Tomasi, J. Chem. Phys., 65, 239 (1982).
J. C. Speakman and H. H. Mills, J. Chem. Soc., 1961, 1164.
L. Golic and J. C. Speakman, J. Chem. Soc. A, 1965, 2530.
A. L. Macdonald, J. C. Speakman, and D. Hadzi, J. Chem.
Soc., Perkin Trans. 2, 1972, 825.
A. L. Macdonald and J. C. Speakman, J. Chem. Soc., Perkin
Trans. 2, 1972, 942.
The crystallographic data can be obtained free of charge via
www.ccdc.cam.ac.uk/conts/retrieving.html (or from the
Cambridge Crystallographic Data Centre, 12, Union Road,
Cambridge, CB21EZ, UK; fax: (+44)1223-336-033; or
deposit@ccdc.cam.ac.uk).
2
5
2
2
6
7
I. Nahringbauer, Acta Chem. Scand., 23, 1653 (1969).
I. Leban and A. Rupnik, Acta Crystallogr., Sect. C, 48, 821
(1992).
2
8
J. Baran, M. Drozd, T. Lis, and H. Ratajczak, J. Mol.
Struct., 372, 151 (1995).
K. Birada, D. Dennis, V. A. Mackinnon, C. V. K. Sharma,
References
2
9
1
M. P. Glackin, J. B. Matthew, and N. M. Allewell, ‘‘Confor-
and M. J. Zaworotko, J. Am. Chem. Soc., 120, 11894 (1998).
30 D. Braga, A. Angeloni, E. Tagliavini, and F. Grepioni, J.
Chem. Soc., Dalton Trans., 1998, 1961.
mations and Forces in Protein Folding,’’ ed by B. T. Nall and K. A.
Dill, Amer. Assoc. Adv. Sci., New York (1991).
2
A. Warshel, Biochemistry, 20, 3167 (1981).
D. Bashford and M. Karplus, Biochemistry, 29, 10219
31 G. Chapuis, A. Zalkin, and D. H. Templeton, J. Chem.
Phys., 62, 4919 (1975).
32 C. J. Horan, P. E. Haney, C. L. Barnes, and R. Glaser, Acta
Crystallogr., Sect. C, 49, 1525 (1993).
3
1990).
4
(
A. Warshel, G. Naray-Szabo, F. Sussman, and J.-K. Hwang,
Biochemistry, 28, 3629 (1989).
J. Antosiewicz, J. A. MacCammon, and M. K. Gilson,
Biochemistry, 35, 7819 (1996).
G. K. Farrington, A. Kumar, and F. C. Wedler, J. Med.
Chem., 28, 1668 (1985).
33 C. Xiao-Ming and T. C. W. Mak, J. Mol. Struct., 221, 265
(1990).
34 A. J. A. R. Blankenstein and J. Kroon, Acta Crystallogr.,
Sect. B, 42, 291 (1986).
35 M. P. Bryn, C. J. Curtis, Y. Hsiou, S. I. Khan, P. A. Sawin,
S. K. Tendick, A. Terzis, and C. E. Strouse, J. Am. Chem. Soc.,
115, 9480 (1993).
5
6
7 S. Patai, ‘‘The Chemistry of Carboxylic Acids and Ester,’’
John & Wiley, London (1969).
8
9
0
J. Jones and F. G. Soper, J. Chem. Soc., 1936, 133.
W. Baker, Nature, 137, 236 (1936).
G. E. K. Branch and D. L. Yabroff, J. Am. Chem. Soc., 56,
36 S. Misaki, S. Kashino, and M. Haisa, Acta Crystallogr.,
Sect. C, 45, 917 (1989).
37 R. M. Silverstein, G. C. Bassler, and T. C. Morrill, ‘‘Spec-
1

A. Onoda et al.
Bull. Chem. Soc. Jpn., 77, No. 2 (2004)
329
trometric Identification of Organic Compounds,’’ Wiley, New
York (1981).
(2002).
42 F. Pina, M. J. Melo, S. Alves, R. Ballardini, M. Maestri, and
P. Passaniti, New J. Chem., 25, 747 (2001).
43 B. Brzezinski, B. Brycki, G. Zundel, and T. Keil, J. Phys.
Chem., 95, 8598 (1991).
44 B. Brzezinski, G. Schroeder, and G. Zundel, J. Chem. Soc.,
Perkin Trans. 2, 1992, 819.
3
8
T. Okamura, K. Sakaue, N. Ueyama, and A. Nakamura,
Inorg. Chem., 37, 6731 (1998).
E. S. Stevens, N. Sugawara, G. M. Bonora, and C. Toniolo,
J. Am. Chem. Soc., 102, 7048 (1980).
A. Warshel and A. Papazyan, Proc. Natl. Acad. Sci. U.S.A.,
3, 13665 (1996).
Q. Cui and M. Karplus, J. Am. Chem. Soc., 124, 3093
3
9
4
0
9
45 S.-O. Shan and D. Herschlag, J. Am. Chem. Soc., 118, 5515
(1996).
4
1