Enols of substituted cyanomalonamides

Article abstract of DOI:10.1021/jo070877h

(Chemical Equation Presented) Twenty open-chain mono-, di-, and trialkyl and aryl-N-substituted cyanomalonamides R²R¹NCOCH-(CN) CONHR³ were prepared. In solution, signals for both amide and a single enol are mostly observed, despite the potential for E and Z isomeric enols. The equilibrium (K_Enol) values between the amides and the enols were determined in different solvents by NMR spectra. They decrease on increasing the polarity of the solvent in the order CDCl₃ ～ C₆D₆ > THF-d₈ (CD₃)₂CO > CD₃CN > DMF-d₇ > DMSO-d₆. For the R¹R²NCOCH(CN)CONHR³ system when R¹ = R² = H, Me or R¹ = H, R² = Me, K _Enol for R³ follows the order: C₆F₅ > Ph ≥ An ≥ i-Pr ≥ t-Bu, and for R¹, R²:H, H > Me, H > Me, Me in all solvents. A unique feature is the appreciable % enol in DMSO-d₆ when R¹ = R² = H, in contrast with enol systems with other electron-withdrawing groups (EWGs). Calculations (B3LYP/6-31G**) corroborate the higher K_Enol values for less alkyl-substituted systems, showing that in the most stable conformer when R¹ = H, R² = R³ = Me the N-hydrogens are closer to the CN group. The order of promoting substituents for enol of amide formation is CONH₂ > CO₂CH₂CF₃ > CO₂Me > CONHMe. The solid-state structures of the isolated species, determined by X-ray crystallography, were either amides or enols, and a higher K_Enol(CDCl₃) value does not ensure a solid enol structure. In no system were both solid species isolated. The X-ray structures of the enols were temperature-dependent. In most cases, the difference between the O-H and O...H bond lengths at low temperature were appreciable, but they become closer at the higher temperature. Similar tendency for either the C=C/C-C or the C-O/C=O bonds was observed. This is ascribed to a hydrogen shift between two regioisomeric enols in an asymmetric double-well potential, which becomes faster at a higher temperature. Calculations show that the enol structures are nonsymmetrical, resembling the lower temperature structures, even when they are chemically symmetrical, but the energy differences between the two regioisomers are <1 kcal. The hydrogen bonds in the enol moiety are strong, with O...O distances <2.45 A, and are resonance-assisted hydrogen bonds. IR spectra in solution and the solid state qualitatively corroborate the NMR and X-ray structure determination.

Full text of DOI:10.1021/jo070877h

Enols of Substituted Cyanomalonamides
Ahmad Basheer,† Hiroshi Yamataka,‡ Salai Cheettu Ammal,‡ and Zvi Rappoport*,†
Department of Organic Chemistry, The Hebrew UniVersity, Jerusalem 91904, Israel, and Department of
Chemistry, Rikkyo UniVersity, Nishi-Ikebukuro 3-34-1, Toshima-ku, Tokyo 171-8501, Japan
zr@Vms.huji.ac.il
ReceiVed April 26, 2007
Twenty open-chain mono-, di-, and trialkyl and aryl-N-substituted cyanomalonamides R2R1NCOCH-
(CN)CONHR3 were prepared. In solution, signals for both amide and a single enol are mostly observed,
despite the potential for E and Z isomeric enols. The equilibrium (KEnol) values between the amides and
the enols were determined in different solvents by NMR spectra. They decrease on increasing the polarity
of the solvent in the order CDCl3 ∼ C6D6 > THF-d8 > (CD3)2CO > CD3CN > DMF-d7 > DMSO-d6.
For the R1R2NCOCH(CN)CONHR3 system when R1 ) R2 ) H, Me or R1 ) H, R2 ) Me, KEnol for R3
follows the order: C6F5 > Ph g An g i-Pr g t-Bu, and for R1, R2:H, H > Me, H > Me, Me in all
solvents. A unique feature is the appreciable % enol in DMSO-d6 when R1 ) R2 ) H, in contrast with
enol systems with other electron-withdrawing groups (EWGs). Calculations (B3LYP/6-31G**) corroborate
the higher KEnol values for less alkyl-substituted systems, showing that in the most stable conformer
when R1 ) H, R2 ) R3 ) Me the N-hydrogens are closer to the CN group. The order of promoting
substituents for enol of amide formation is CONH2 > CO2CH2CF3 > CO2Me > CONHMe. The solid-
state structures of the isolated species, determined by X-ray crystallography, were either amides or enols,
and a higher KEnol(CDCl3) value does not ensure a solid enol structure. In no system were both solid
species isolated. The X-ray structures of the enols were temperature-dependent. In most cases, the difference
between the O-H and O‚‚‚H bond lengths at low temperature were appreciable, but they become closer
at the higher temperature. Similar tendency for either the CdC/C-C or the C-O/CdO bonds was
observed. This is ascribed to a hydrogen shift between two regioisomeric enols in an asymmetric double-
well potential, which becomes faster at a higher temperature. Calculations show that the enol structures
are nonsymmetrical, resembling the lower temperature structures, even when they are chemically
symmetrical, but the energy differences between the two regioisomers are <1 kcal. The hydrogen bonds
in the enol moiety are strong, with O‚‚‚O distances <2.45 Å, and are resonance-assisted hydrogen bonds.
IR spectra in solution and the solid state qualitatively corroborate the NMR and X-ray structure
determination.
Introduction
constants for the enolization, KEnol (or pKEnol ) -log KEnol),
under various conditions.
In our previous work on enols of carboxylic acid amides,1
we had shown that long-lived enols of amides 1a can be
obtained if the amide 2a T 2b carries two â-electron-
withdrawing groups (EWGs) Y, Y′. This is mainly due to the
contribution of the dipolar structure, in which the negative
charge is delocalized as in 1b over the two â-EWGs (eq 1 in
Scheme 1). This enabled us to calculate the equilibrium
From our data when Y, Y′ ) CO2R,1a,c CN,1b or SO2R2
groups, enolization is always on the amide carbonyl. In contrast,
(1) (a) Mukhopadhyaya, J. K.; Sklenak, S.; Rappoport, Z. J. Am. Chem.
Soc. 2000, 122, 1325. (b) Mukhopadhyaya, J. K.; Sklenak, S.; Rappoport,
Z. J. Org. Chem. 2000, 65, 6856. (c) Lei, Y. X.; Cerioni, G.; Rappoport, Z.
J. Org. Chem. 2001, 66, 8379. (d) Lei, Y. X.; Casarini, D.; Cerioni, G.;
Rappoport, Z. J. Org. Chem. 2003, 68, 947. (e) Lei, Y. X.; Casarini, D.;
Cerioni, G.; Rappoport, Z. J. Phys. Org. Chem. 2003, 16, 525. (f) Basheer,
A.; Rappoport, Z. J. Org. Chem. 2004, 69, 1151.
† The Hebrew University.
‡ Rikkyo University.
10.1021/jo070877h CCC: $37.00 © 2007 American Chemical Society
Published on Web 06/09/2007
J. Org. Chem. 2007, 72, 5297-5312
5297

Basheer et al.
SCHEME 1
SCHEME 2
either a single minimum or a double minimum with a rapid
equilibration between two structures,7 this question is meaning-
less. Finally, we want to compare the effect of an amido group
with that of other EWGs as enolization promoting groups.
We have therefore prepared >20 Y-substituted malonamides
and studied their enolization and the structures and properties
of their enols under various conditions. Since a single activating
group on an amide is insufficient to generate an observable enol,
Y should be another EWG. In order to avoid a possible
involvement of the group Y as an alternative enolization site
or as a hydrogen-bonding partner (cf. the nitro group in 3), we
studied the cyanomalonamide systems 4/5/6 with Y ) CN (eq
3 in Scheme 3). Previous experience showed that cyano is
effective in promoting enol formation1b,e and that enolization
on the cyano nitrogen to form a ketenimine does not compete
with enolization on a geminal amide carbonyl. Moreover, the
linear small cyano group is expected to reduce steric effects
and not to be involved in intramolecular H-bonds with the enolic
OH since the CN nitrogen and OH are too remote. Systems
with other Y’s will be discussed elsewhere.
when enolization can take place on a Y or Y′ group rather than
on the amide, both calculations3 and experiment1b indicate that
it will take place on them, for example, when Y is a CdO group
of a ketone, except in special cases. Preferred enolization on a
â-NO group was deduced from calculations.4 With a NO2 group,
it is not easy from the experimental data obtained so far to
distinguish between enolization on the amide or on the NO2
group,1b although calculations4 and experimental data for
nitromalonamide5 indicate enolization on the amide group.
Intramolecular hydrogen bonds play an important role in
directing the enolization process, by contributing to increased
enol stability.1 This is shown by the lower field position of the
OH in the more stable isomer of the E/Z enol pairs,1c-f as well
as in the short O‚‚‚O distances when one oxygen is of the enol
and the other is of a â-ester group.1c-f This is demonstrated in
eq 2 in Scheme 2, which shows an enol stabilized by an ester
group where the E isomer is preferred since an O‚‚‚H-O is
stronger than an O‚‚‚H-N H-bond.6 In the solid state at low
temperature, the hydrogen is located on the amide carbonyl
rather than on the ester carbonyl.1c This is corroborated by
calculations and by the appearance of two enols in solution when
Y, Y′ are two different ester groups1c-f but of only one enol
when the two ester groups are identical.1c
Results and Discussion
Synthesis. The 20 cyanomalonamides were prepared by the
reaction of the anion of substituted cyanoacetamide with aryl
or alkyl isocyanates R3NCO (eq 4 in Scheme 4). This gave the
amides (4) or their enols (5 or 6) with at most three nitrogen
substituents, that is, R4 ) H (eq 3).
1
NMR Data. H and 13C NMR spectroscopies were used to
probe the structures of the enols/amides in solution and to
calculate the KEnol values. Both species were observed in the
NMR spectra in most solvents. Although there are potentially
both E- and Z-enols as well as regioisomeric enols (due to the
two amido carbonyls), only signals for an apparent one enol
species were observed. The enols were identified unequivocally
1
by a very low-field H δ(OH) signal at δ 16.16-19.02 which
Several reasons exist to study a system with two amides as
â-EWG groups (i.e., Y-substituted malonamides, YCH-
(CONR1R2)(CONR3R4)). A rare example of a known stable
solid enol of amide is the formally symmetric nitromalonamide
35 having an extremely short O(1)‚‚‚O(2) nonbonding distance
of 2.384 Å and a nonsymmetrical hydrogen bond.
was observed for all of the enols investigated in most solvents
(Table 1). The Table shows (a) that electron withdrawal shifts
the OH to higher field, (b) that with two hydrogens on one
nitrogen the OH is at a higher field, (c) that mostly in C6D6 the
OH is at the highest field (e.g., for 5d/6d δ(OH) ) 19.02), and
(d) that in some cases the OH is observed also in DMSO-d6
and DMF-d7 (e.g., in 5l-p/6l-p). Generally, the δ(OH) values
are at a lower field than for the enols of diester 1, Y, Y′ )
CO2R, CO2R′ or of a cyanoester 1, Y ) CN, Y′ ) CO2R.1 The
δ(NH) of the enol is at a lower field than that of the isomeric
(2) Basheer, A.; Rappoport, Z. Presented in the 9th European Symposium
on Organic Reactivity (ESOR 9), Oslo, Norway, July 12-17, 2003; Abstr.
OR 59, p 128, and the 17th IUPAC Conference on Physical Organic
Chemistry, Shanghai, China, August 15-20, 2004; Abstr. IL-10, p 84.
(3) Sklenak, S.; Apeloig, Y.; Rappoport, Z. J. Am. Chem. Soc. 1998,
120, 10359.
Its stability contrasts the small percent of enolization of system
1a/2a, R ) H, R′ ) Ph, Y ) NO2, Y′ ) CO2Et.1b The effect
of the four N-substituents R1-R4 appearing on both CR and Câ,
which have opposite electronic demands on enolization on the
(4) Yamataka, H.; Ammal, S. C. Unpublished results.
(5) Simonsen, O.; Thorup, N. Acta Crystallogr., Sect. B 1979, 57, 1177.
(6) Grabowski, S. J. J. Phys. Org. Chem. 2004, 17, 18.
(7) (a) Schuster, P.; Zundel, G.; Sandorfy C. In The Hydrogen Bond.
Recent DeVelopments in Theory and Experiments; North Holland: Am-
sterdam, 1976. (b) Hibbert, F.; Emsley, J. AdV. Phys. Org. Chem. 1990,
26, 255.
KEnol values, is interesting. The enolization site is of interest if
the two amide groups differ. However, if the hydrogen bond is
5298 J. Org. Chem., Vol. 72, No. 14, 2007

Enols of Substituted Cyanomalonamides
SCHEME 3
SCHEME 4
amide, and the δ(NH)[amide] - δ(NH)[enol] difference is
higher in the nonpolar solvents.
tively, compared with the two amide CdO groups at 157.58-
162.97 ppm. The amide CN signals appear in the C/H coupled
spectra as doublets (J ) 10.6-11.3 Hz) at a higher field
(113.04-116.12 ppm) than the enol CN singlets at 115.01-
121.08 ppm. Finally, the coupled Câ doublets (J ) 136.4-138.9
Hz) of the amides appear at 42.46-48.92 ppm, a lower chemical
shift than that of the enol Câ singlets at 53.51-57.88 ppm.
Substituent and Solvent Effects on KEnol Values. Table 1
displays KEnol values for 19 cyanomalonamides in 5-7 solvents
and for the 4f/5f/6f system in 3 solvents. As with all the other
activated enols of amides previously studied,1 increase in the
dielectric constant of the solvent results in a lower % enol. This
is ascribed to the higher polarity of the amides than of the
internally hydrogen bonded, that is, internally solvated, and
hence of lower polarity enols, which consequently require less
external solvation.1c A similar explanation was suggested for
the solvent effect on KEnol of 1,3-dicarbonyl compounds whose
enols can form intramolecular hydrogen bonds.8 In contrast,
simple enols of ketones, aldehydes, or 1,3-diketones incapable
of forming strong hydrogen bonds display higher % enol in the
more polar and hydrogen-bonding solvents.9
The spectra also show the two enol NH signals, as well as
the CH and the two NH signals of the amide. The NH signals
were identified by their relative integrations compared with the
OH and CH signals. An example of the spectrum of the 4e/5e/
6e system displaying both the Z-enol (see below) and the amide
(Figure 1) is simple despite the potential problems mentioned
above. Separate signals for the amide and enol are observed,
so that they do not equilibrate rapidly on the NMR time scale
of the experiment. Unidentified signals are observed for the
polyaryl-substituted 4f/5f/6f and 4t/5t/6t and are ascribed to
decomposition; 4k/5k/6k display signals which may be due to
their derived anion formed in THF-d8 (18%), CD3CN (39%),
and DMF-d7 (100%).
The enol/amide ratios were calculated from the integrations
of pairs of signals, such as OH versus CH, the Ph-H,
MeOC6H4, or NH signals for the aryl-substituted systems, and
i-Pr or t-Bu signals of both species for the N-alkyl-substituted
systems. The differences between the percentages determined
for a species by different methods were mostly 2-4%. The
average values were used to obtain the KEnol values given in
Table 1. The relative integrations of all signals are given in
Table S1 in the Supporting Information.
However, a remarkable feature of the enols of cyano- (and
nitro-)10 malonamides is that several of them, especially when
singly N-substituted (Table 1 and see below), are observable in
The 13C NMR spectral data (Table S3 in the Supporting
(8) Floris, B. In The Chemistry of Enols; Rappoport, Z., Ed.; Wiley:
Chichester, 1990; Chapter 4, p 147.
1
Information) are in line with the H NMR data (Tables 1 and
(9) (a) Miller, A. R. J. Org. Chem. 1976, 41, 3599. (b) Koelle, U.; Forsen,
S. Acta Chem. Scand. Ser. A 1974, 28, 531. (c) Rochlin, E.; Rappoport, Z.
J. Am. Chem. Soc. 1992, 114, 230.
(10) The chemistry of nitromalonamide and derivatives will be discussed
elsewhere.
S1) and provide a qualitative characterization for the amides
and their enols. Two pairs of low-field signals were observed.
The lower pair for the enol appears at 170.62-173.18 and
171.17-175.16 ppm and is ascribed to CdO and CR, respec-
J. Org. Chem, Vol. 72, No. 14, 2007 5299

Basheer et al.
TABLE 1. δ(OH), δ(CH), and δ(NH) Spectral Data and R1, R2, and R3 Effect on KEnol for the R3NHCOCH(CN)CONR1R2/
R3NHC(OH)dC(CN)CONR1R2/R3NHCOC(CN)dC(OH)NR1R2 (4/5/6) System in Several Solvents at Room Temperature
Enol
Amide
compound
solvent
δ(OH)
δ(NH)1
δ(NH)2
δ(CH)
δ(NH)1
δ(NH)2
% enol
KEnol
pKEnol
4a/5a/6a
CDCl3
C6D6
17.79
18.41
18.18
18.07
17.88
7.67
7.96
8.52
8.50
7.91
4.80
4.32
5.05
5.31
5.00
5.57
5.43
4.75
4.43
5.06
5.00
5.51
5.39
4.84
4.10
5.26
5.15
5.83
5.69
4.61
4.28
4.80
5.03
4.80
5.23
5.12
4.51
4.01
4.70
9.16
9.49
9.64
9.66
8.94
10.62
10.45
9.14
9.43
9.56
8.73
10.48
10.34
9.40
9.37
9.85
8.76
10.89
10.73
6.68
6.92
7.31
7.47
6.85
8.35
8.36
6.59
6.87
7.19
80
80
46
18
13
0
4.0
4.0
-0.60
-0.60
0.07
0.66
0.83
g1.7
g1.7
-0.60
-0.55
0.10
THF-d8
(CD3)2CO
CD3CN
DMF-d7
DMSO-d6
CDCl3
0.85
0.22
0.15
e0.02
e0.02
4.0
3.55
0.79
0.15
0.052
e0.02
13.3
19.0
3.35
0.64
0.18
0.08
4.0
4.88
1.70
0.39
0.35
0.08
0.04
2.70
4.0
0.89
0.39
0.09
0.02
e0.02
2.85
0.09
e0.02
9.0
0
4b/5b/6b
4c/5c/6c
4d/5d/6d
17.88
18.61
18.19
18.02
7.66
7.97
8.51
7.86
8.45
80
78
44
13
5
C6D6
THF-d8
CD3CN
DMF-d7
DMSO-d6
CDCl3
0.83
1.28
0
g1.7
-1.12
-1.28
-0.52
0.19
0.75
1.12
-0.60
-0.69
-0.23
0.41
0.45
1.12
1.38
-0.43
-0.60
0.05
0.41
1.06
16.82
17.46
17.46
16.95
17.39
17.28
18.18
19.02
18.53
18.51
18.36
18.59
18.42
18.24
18.91
18.37
7.44
7.65
8.81
7.76
9.81
9.91
5.62
5.89
6.37
6.47
6.00
7.40
5.49
5.66
5.85
5.98
93
95
77
39
15
7
80
83
63
28
13
7
C6D6
THF-d8
CD3CN
DMF-d7
DMSO-d6
CDCl3
C6D6
THF-d8
(CD3)2CO
CD3CN
DMF-d7
DMSO-d6
CDCl3
4
4e/5e/6e
73
80
47
28
8
2
0
74
8
0
C6D6
THF-d8
(CD3)2CO
CD3CN
DMF-d7
DMSO-d6
CDCl3
CD3CN
DMSO-d6
CDCl3
18.30
18.40
5.80
6.55
4.72
5.20
5.07
4.54
4.96
5.49
4.63
3.78
4.60
4.65
5.08
4.90
4.56
5.60
4.49
4.54
4.59
5.04
4.85
4.69
3.39
4.78
4.79
5.29
5.07
4.39
6.70
8.22
8.15
8.17
8.17
9.37
9.21
9.18
9.68
8.95
10.53
10.38
8.96
7.36
8.85
9.56
8.77
10.42
10.24
9.25
8.88
9.84
8.87
10.77
10.60
6.88
1.66
g1.7
-0.45
1.06
g1.7
-0.95
-1.06
-0.72
-0.03
0.37
4f/5f/6f
17.52
17.39
7.10
6.33
4g/5g/6g
17.65
18.34
18.18
17.89
18.21
18.16
17.71
18.47
17.70
18.25
17.96
18.31
a
16.70
17.29
17.35
16.91
17.03
17.09
17.73
18.70
18.44
18.11
18.39
18.26
17.80
18.65
18.27
18.10
10.82d
18.15
7.40
7.40
8.58
7.87
9.59
9.61
7.16
7.08
7.09
8.53
7.76
9.56
9.52
7.16
6.73
8.89
7.70
9.87
9.81
5.37
5.45
6.56
5.99
7.58
7.57
5.31
5.04
5.91
5.60
7.40
6.65
6.04
5.32
7.32
6.51
8.05
f
f
90
92
84
52
30
18
90
90
85
84
57
31
16
97
97
96
79
69
56
92
100
90
54
46
25
84
93
63
20
0
C6D6
6.12
7.77
7.00
8.38
8.33
7.23
5.85
f
7.76
7.08
8.38
8.29
6.94
5.55
7.82
7.02
8.25
8.42
7.14
f
7.59
6.95
8.32
8.15
7.01
6.75
7.81
7.45
8.35
8.09
11.5
5.25
1.08
0.43
0.22
9.0
THF-d8
CD3CN
DMF-d7
DMSO-d6
CDCl3
0.66
4h/5h/6h
5.76
5.00
5.68
7.23
6.42
7.46
7.97
6.06
4.92
7.61
6.68
8.25
8.42
5.74
5.45
6.91
6.24
7.67
7.67
5.58
5.30
7.01
6.22
8.35
f
-0.95
-0.95
-0.75
-0.72
-0.12
0.35
C6D6
9.0
CD2Cl2
THF-d8
CD3CN
DMF-d7
DMSO-d6
CDCl3
5.67
5.25
1.33
0.45
0.19
32.3
32.3
24
3.76
2.23
1.27
11.5
0.72
4i/5i/6i
-1.51
-1.51
-1.38
-0.58
0.35
-0.11
-1.06
e-1.7
-0.95
-0.07
0.07
C6D6
THF-d8
CD3CN
DMF-d7
DMSO-d6
CDCl3
4j/5j/6j
4k/5k/6k
C6D6
g50
THF-d8
CD3CN
DMF-d7
DMSO-d6
CDCl3
C6D6
THF-d8
CD3CNb
4.26
4.38
4.77
4.59
4.32
3.62
4.30
4.33
10.82d
4.59
7.49
6.88
8.32
8.13
6.81
6.09
7.51
6.79
7.40
7.91
9.0
1.17
0.85
0.33
5.25
13.3
3.32
0.49
0.48
-0.72
-1.12
-0.52
0.31
g1.7
0.69
a
c
DMF-d7
DMSO-d6
e0.02
0.21
17
5300 J. Org. Chem., Vol. 72, No. 14, 2007

Enols of Substituted Cyanomalonamides
TABLE 1. Continued
Enol
Amide
compound
solvent
δ(OH)
δ(NH)1
δ(NH)2
δ(CH)
δ(NH)1
δ(NH)2
% enol
KEnol
pKEnol
4l/5l/6l
CDCl3
C6D6
17.32
17.70
17.35
17.29
17.40
17.00
17.54
17.54
17.43
17.56
17.35
16.23
16.57
16.42
16.16
16.29
15.63
17.53
18.32
17.89
17.80
18.00
17.76
17.59
18.25
17.61
17.71
17.96
17.53
17.27
17.75
17.76
17.41
7.30
7.36
8.53
7.88
9.56
9.53
4.84
8.52
7.81
9.44
9.38
7.00
6.51
8.88
7.79
9.82
9.74
5.36
5.10
6.55
5.90
7.79
8.13
5.43
5.26
5.87
5.68
6.52
6.51
7.04
7.29
9.18
7.93
5.48
4.01
4.72
6.36
8.28
7.90
7.23
7.23
6.29
8.20
8.20
5.68
4.09
7.50
6.59
8.31
8.24
5.36
4.18
6.93
6.08
7.48
f
5.43
4.17
6.96
6.12
7.87
7.68
7.28
7.08
9.26
8.28
100
100
97
83
69
43
94
91
81
74
45
100
100
100
95
92
88
100
100
100
84
78
40
100
100
94
66
53
33
100
100
87
54
0
0
94
93
69
g50
g 50
32.3
4.88
2.23
0.75
15.97
10.11
4.26
e-1.7
e-1.7
-1.51
-0.69
-0.35
0.12
THF-d8
CD3CN
DMF-d7
DMSO-d6
THF-d8
(CD3)2CO
CD3CN
DMF-d7
DMSO-d6
CDCl3
4.58
4.64
5.07
4.68
4.73
4.82
4.61
5.03
4.84
9.66
8.85
10.54
10.36
10.84
9.55
8.73
10.53
10.24
7.39
6.71
8.28
7.37
f
4m/5m/6m
4n/5n/6n
-1.19
-1.0
f
7.06
f
7.72
-0.63
-0.45
0.087
e-1.7
e-1.7
e-1.7
-1.28
-1.06
-0.87
e-1.7
e-1.7
e-1.7
-0.72
-0.55
0.18
e-1.7
e-1.7
-1.2
-0.29
-0.05
0.31
e-1.7
e-1.7
-0.83
-0.07
g1.7
g1.7
-1.19
-1.12
-0.35
2.85
0.82
g50
C6D6
g50
g50
13
11.2
7.33
g50
g50
g50
THF-d8
CD3CN
DMF-d7
DMSO-d6
CDCl3
4.83
5.03
5.06
8.83
10.78
10.60
6.81
8.99
8.99
4o/5o/6o
4p/5p/6p
4q/5q/6q
4r/5r/6r
4s/5s/6s
4t/5t/6t
C6D6
THF-d8
CD3CN
DMF-d7
DMSO-d6
CDCl3
4.38
4.80
4.59
6.67
8.19
7.60
6.80
f
3.36
5.25
3.55
0.67
g50
g50
15.7
1.94
1.13
0.49
g50
g50
6.69
1.17
e0.02
e0.02
15.67
13.29
2.23
C6D6
THF-d8
CD3CN
DMF-d7
DMSO-d6
CDCl3
4.78
4.37
4.78
4.58
7.30
6.27
7.77
7.89
7.35
6.67
7.68
7.56
C6D6
THF-d8
CD3CN
DMF-d7
DMSO-d6
CDCl3
4.89
4.97
5.61
5.30
f
5.51
4.72
4.83
5.29
5.12
4.41
f
9.90
8.88
11.12
10.75
8.72
8.26
9.66
8.72
8.00
10.40
6.88
6.71
7.50
6.85
8.09
8.04
9.75
8.70
10.96
10.60
17.94
18.41
18.33
18.06
7.29
6.91
8.91
8.07
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
C6D6
THF-d8
CD3CN
DMF-d7
DMSO-d6
CDCl3
0
0
e0.02
e0.02
8.09
15.67
5.67
0.79
0.61
0.23
g50
g50
1.63
0.15
e0.02
e0.02
g1.7
g1.7
-0.91
-1.1
-0.75
0.10
17.76
18.77
18.38
18.18
18.43
18.29
17.69
18.62
18.39
17.96
5.38
5.42
6.60
6.00
7.50
7.51
6.03
6.54
7.70
6.79
89
94
85
44
38
19
100
100
62
13
0
C6D6
THF-d8
CD3CN
DMF-d7
DMSO-d6
CDCl3
4.27
4.36
4.71
4.54
0.21
0.36
e-1.7
e-1.7
-0.21
0.83
g1.7
g1.7
C6D6
THF-d8
CD3CN
DMF-d7
DMSO-d6
4.78
4.84
5.50
5.18
8.78
8.00
9.50
9.29
0
a Ionization of 18%. b Ionization of 39%. c Ionization of 100%. d Signal ascribed to H+. e NHR3 ) NR1R2. f Not observed.
significant percentage even in DMSO-d6 and DMF-d7. Previous
studies with other amide/enol systems in these solvents either
showed no enol or ionization to the enolate ion when substituted
with strong EWGs.1c-e Likewise, contrary to the very low %
enol previously observed in CD3CN,1c-e the cyanomalonamides
display 8-95% enol in CD3CN.
enol changes from 73 to 80% (KEnol ) 2.2-4.0) for the NMe2
derivative to 84-92% (KEnol ) 5.3-11.5) for the NHMe
derivatives and reaches 100% (KEnol g 50) for the NH2
derivatives. Within the first group, the % enol is nearly the same
for R3 ) i-Pr, p-An, and Ph and somewhat lower for R3 ) t-Bu
and CPh3. The highest % enol is for R3 ) C6F5.
The dependence of KEnol on R1, R2 in all solvents is regular
and appreciable. The % enol increases when the number of Me
(and other alkyls) groups decreases and follows the order: R1,
R2 ) H, H > Me, H > Me, Me (Table 1, a-p). The
cyanomalonamides are not soluble enough for determining KEnol
values in CCl4. However, KEnol values in CDCl3 show systematic
substituent effects (Table 1). For the NR1R2 derivatives, the %
For R3 ) t-Bu in CDCl3, C6D6, and (CD3)2CO, and less in
THF-d8, the % enol resembles the values for R3 ) i-Pr, whereas
the values in DMSO-d6 are somewhat lower. With R3 ) Ph,
the values in the less polar solvents resemble those for R3 )
t-Bu. In DMF-d7 and DMSO-d6, the enol was not observed for
R1R2 ) Me2, but its percentage was significant, being 69 and
43 for R1R2 ) H2.
J. Org. Chem, Vol. 72, No. 14, 2007 5301

Basheer et al.
FIGURE 1. 1H NMR spectrum of 4e/5e/6e in CDCl3.
In THF-d8, a hydrogen-bonding solvent of a moderate polar-
ity, the % enol is lower than that in CDCl3, but the same trend
for R1R2 is obtained, that is, 44-77% enol (KEnol ) 0.79-3.35)
for NMe2, 76-90% enol (KEnol ) 3.2-9.0) for NHMe, and 94-
100% enol (KEnol ) 15.7 to g50) for NH2. The highest % enol
is for i-Pr for NH2 and NHMe, and the second highest is for
NMe2. In the high dielectric constant DMSO-d6, all the R3,
NMe2 systems give no enol, similarly to almost all systems stud-
ied previously,1 although with R3 ) i-Pr or C6F5, signals inter-
preted as those of 4 and 7% enol were observed. However, for
the R3, NHMe systems, the % enol is already 16-25% (KEnol
) 0.19-0.33), except for R3 ) C6F5, which displays 56% enol
(KEnol ) 1.27). For the R3, NH2 system, the % enol is 33-45%
(KEnol ) 0.49-0.82), except when R3 ) C6F5 (88% enol, KEnol
) 7.33, Table 1). In CD3CN, there is 8-13% enol for the NMe2
(39% for R3 ) C6F5), 20-57% enol for the NHMe (79% for R3
) C6F5), and 66-84% enol for the NH2 (95% for R3 ) C6F5).
C6D6 gave very similar % enol to CDCl3 for several R3, NMe2
systems, except when R3 ) t-Bu (higher % in CDCl3). The
difference, in the same direction, is higher for the R3 ) i-Pr,
t-Bu, NHMe systems. For the NH2 systems, the % enol is 100%
for all derivatives (Table 1). Interestingly, DMF-d7 consistently
displays higher % enol (53-78%, R1R2 ) H2) than DMSO-d6
(33-45%, R1R2 ) H2) and lower % than CD3CN (66-95%,
R1R2 ) H2). The % enol in (CD3)2CO for selected systems
shows that it is between THF-d8 and CD3CN (Table 1).
The % enol is higher for R3 ) C6F5 than for R3 ) Ph, p-An
(aromatic substituents) and for R3 ) i-Pr than for R3 ) t-Bu or
CPh3 (aliphatic substituents). The effect of R3 on % enol is
relatively small, and when R1 ) R2 ) H, Me or R1 ) H, R2 )
Me, it follows the order: C6F5 > Ph g An g i-Pr g t-Bu. In
summary, the solvent effect is similar to that found earlier with
other amide/enol systems1 where increase in the dielectric
constant of the solvent results in a decreased KEnol value which
follows the order: CDCl3 ∼ C6D6 > THF-d8 > (CD3)2CO >
CD3CN > DMF-d7 > DMSO-d6.
Three symmetrical systems where the substituents R1,R2 )
R3,R4 are H, Ph (4r/5r/6r), H, i-Pr (4s/5s/6s), or H, CHPh2 (4t/
5t/6t) were investigated (Table 1). Symmetry does not increase
the enol stability compared with the slightly less symmetric
systems with R1,R2 ) HPh, R3,R4 ) NHMe or R1,R2 ) NHPh,
R3,R4 ) NHC6F5 and R1,R2 ) NHPr-i, R3,R4 ) NHMe
substituted pairs. In most cases, the less symmetrical system
gives higher KEnol (e.g., KEnol ) 2.23 when R1,R2 ) R3,R4 )
HPh compared with KEnol ) 6.69 when R1,R2 ) R3,R4 ) HPh,
HC6F5 or KEnol ) 5.25 when R1,R2 ) R3,R4 ) HPh, HMe in
THF-d8) regardless if the steric effect is increased or decreased.
The “exceptions” are cases where KEnol is very high (>90%)
and its error is therefore large. KEnol value for the NHMe, NHPr-i
(11.5 in CDCl3) derivative is slightly higher than that for the
NHMe, NHBu-t (5.52 in CDCl3) one in all solvents, and we
ascribe it to a small steric effect (Table 1). Similar effect was
observed with the NMe2, NHPr-i (4.0 in CDCl3) and NMe2,
NHBu-t (2.7 in CDCl3) pairs. However, for the NHCHPh2,
NHCHPh2 system, the KEnol value is g50 in CDCl3 and C6D6,
a higher value than for the NHPr-i, NHPr-i (4s/5s/6s) (8.09 in
C6D6) system, but the value of 0% enol in DMF-d7 and DMSO-
d6 was lower than the values of 38 and 19% enol for the latter
system. A high steric effect is expected for the 4f/5f/6f system,
where R1,R2 ) Me2, R3 ) CPh3. However, the KEnol values of
2.85 in CDCl3 and 0.09 in CD3CN do not show an extreme
steric effect. For an explanation see below.
Calculated pKEnol Values. Calculated pKEnol values were
found in the past to be somewhat sensitive to the basis set
5302 J. Org. Chem., Vol. 72, No. 14, 2007

Enols of Substituted Cyanomalonamides
TABLE 2. Calculated (B3LYP/6-31G**) pKEnol Values at 298.15 K
and Bond Lengths (in Å) of Enols Derived from
R1R2NCOCH(Y)CONR3R4 at 0 K
Y
R1
R2
R3
R4
O-H
O‚‚‚H
O‚‚‚O pKEnol
CN
CN
CN
CN
CN
CN
CN
CN
CN
CN
CN
H
H
H
Me
H
H
H
H
H
H
Me
1.050
1.054
1.067
1.454
1.437
1.383
1.452
1.436
1.444
1.400
1.428
1.409
1.485
1.505
1.504
2.445
2.437
2.400
2.443
2.437
2.440
2.411
2.423
2.423
2.461
2.471
2.474
-5.7
-4.3
-2.1
-5.1
-4.3
-4.2
-4.1
-4.7
-4.19
-4.21
-5.19
5.8
Me
Me
H
i-Pr
Me
Me
H
H
H
H
H
Me Me
H
H
H
H
i-Pr 1.050
i-Pr 1.055
i-Pr 1.051
i-Pr 1.060
H
Me
Ph
An
C6F5
C6F5
H
Me Me
1.048
1.065
1.038
1.033
1.035
Me
Me
H
H
H
H
H
H
H
H
NO2
H
H
1.086a 1.344a 2.380a
-8.5
a Obs. O-H 1.01, O‚‚‚H 1.44, O-H‚‚‚O 2.384 Å (rt); 1.11, 1.28, 2.3730
Å (100 K).
used.3,11 The calculations in this work are at the B3LYP/6-
31G** level, in order to facilitate comparisons with previous
calculations.1a,3,11
The calculated values for R1-R4 for 13 systems of special
interest are given in Table 2. For the symmetrical system with
R1, R2 ) R3, R4, the pKEnol value decreases with the increased
methyl substitution on the two amido groups, being -5.7 for
the parent cyanomalonamide, -4.3 for its N,N′-dimethyl deriva-
tive, and -2.1 for the N,N,N′,N′-tetramethyl derivative. When
one amide derivative is fixed as a CONHPr-i and the other one
is changed from CONH2 to CONHMe to CONMe2, the same
trend is observed, but the values become much closer: pKEnol
) -5.1, -4.2, and -4.1, respectively. The trend is identical
with that found experimentally, indicating that it is not due
mainly to solvation, although the observed values are more
negative. The similarity of the calculated KEnol values for the
5d/6d and 5j/6j systems is due to the fact that the hydrogen of
the NHPr-i moiety is located near the CN in both systems.
These systems display systematic changes in the O-H, O‚‚‚H,
and O‚‚‚O distances (Table 2). The O-H bond elongates from
1.050 (R1-R4 ) H), via 1.054 (R1, R3 ) H, R2, R4 ) Me) to
1.067 (R1-R4 ) Me), and the O‚‚‚H bond shortens, respectively,
from 1.454, via 1.437 to 1.383 Å. The O‚‚‚O distance decreases
from 2.445 for the 4H to 2.400 Å for the 4Me compound. Again,
the symmetrical NHPr-i derivative behaves similarly to the
NHMe derivative. Table 2 also presents the calculated data for
the enols of YCH(CONH2)2 for Y ) H, NO2. Compared with
Y ) CN, R1-R4 )H, O-H is shorter and O‚‚‚H is longer for
Y ) H, and O-H is longer and O‚‚‚H is shorter for Y ) NO2.
The symmetry of the hydrogen bond, reflected by the calculated
FIGURE 2. Calculated structures (B3LYP/6-31G**) for several
conformations of MeNHCOC(CN)dC(OH)NHMe: (A) Me groups
close to the O-H‚‚‚O bond; (B) Me groups close to the CN group;
(C) one Me close to the O-H‚‚‚O bond and one close to the CN.
other symmetrical conformation (B) with two Me groups on
the side of the CN (Figure 2B). The two other conformations
are (C1) with Me on the amide side and Me on the CN side
(Figure 2C), and (C2) Me on the enol side and Me on the CN
side which are 3.36 and 3.72 kcal/mol, respectively, less stable
than conformer A. An indication for the steric repulsion is given
by the dihedral angle between the C-CdC and C-CN planes.
This angle is 180° for systems with at least one H near the CN,
for example, for conformers A, C1, and C2 and for the
unsubstituted cyanomalonamide (R1-R4 ) H), meaning that
the CN and the six-membered enolic ring are in the same plane.
In contrast, when two Me groups reside close to the CN as in
conformer B or the tetra-Me compound (R1-R4 ) Me), the
dihedral angles are 169 and 163°, respectively. The distances
between the CN carbon and the Me hydrogen close to it are
2.747 and 2.746 Å in B, 2.746 Å in C1, 2.751 Å in C2, and
2.747 and 2.510 Å in the tetra-Me compound.
∆
ÃΗ
) d(O‚‚‚H)-d(O-H), is the lowest for the most stable
enol with pKEnol of -8.5 when Y ) NO2 (0.258 Å) and the
highest for the least stable enol with Y ) H (pKEnol )5.8, ∆ÃΗ
) 0.469 Å). The values for the di- and tetra-Me systems follow
this trend, the more negative the pKEnol value, the lower the
∆
ÃΗ
value.
The energies of the four stable conformers of the enol with
R1, R2 ) R3, R4 ) Me, H were calculated. The most stable
conformer (A) is a symmetrical one with two Me on the side
of the hydrogen bond and two hydrogens on the side of the CN
(Figure 2A), which is 5.92 kcal/mol more stable than the most
stable amide conformer. It is 9.9 kcal/mol more stable than the
The energy difference between conformers A and B, which
translates to pKEnol of ca. 3 for B, seems inconsistent with the
pKEnol of -2.1 for the tetramethyl derivative (Table 2). If only
the effect operating for B applies, the observed value obtained
(11) Mishima, M.; Matsuoka, M.; Lei, Y. X.; Rappoport, Z. J. Org. Chem.
2004, 69, 5947.
J. Org. Chem, Vol. 72, No. 14, 2007 5303

Basheer et al.
TABLE 3. ∆G Values (kcal/mol) for the Isodesmic Reactions 5-8
is at least 6 kcal/mol too negative. We ascribe this to hydrogen
bonding in the corresponding malonamides. The bond for the
Me4 derivative cannot be stabilized by an intramolecular
N-H‚‚‚OdC hydrogen bond, whereas that of B has such
stabilization. In summary, the relatiVely small difference of 3.6
pKEnol units between the tetra-H and tetra-Me compared with
7.2 units between conformers A and B is due to a combination
of destabilization of the enols by electronic and steric effects
of the Me groups and destabilization of the tetramethyldiamide
by lack of H-bonding.
These results point to a source of a possible up to 1 kcal/mol
difference between the calculated and observed pKEnol values.
The calculated values are for the difference between the most
stable conformers of the amide and the enol. In solution, there
may be rapid equilibria between more than two conformers with
the most stable one, which may give a more negative calculated
pKEnol value.
These energy differences between conformers and the re-
semblance of Kenol of the 4f/5f/6f and the 4s/5s/6s systems with
the bulky CPh3 and CHPh2 substituents to those of enols with
smaller substituents suggest that in the conformations of these
enols the bulky groups occupy positions remote from the CN
group, so that steric effects are less pronounced than they may
be for the higher energy conformers. Moreover, the diamides
in these cases can form intramolecular hydrogen bonds.
Steric effects clearly affect the preferred conformation and
presumably the pKEnol values. In either the observed or the
calculated preferred conformation, the two N-substituents
adjacent to the CN group are the smaller ones, usually two
hydrogens.
R1R2 ) R3R4
eq
∆G
H, H
5
6
7
8
5
6
7
8
5
6
7
8
-0.7
36.1
-10.4
26.4
0.8
37.5
-9.0
27.8
-2.2
29.7
-9.8
22.1
H, Me
Me, Me
TABLE 4. Calculated pKEnol Values (B3LYP/6-31G**) for Some
Activated Amides Y′YCHCONH2
Y′/Y)
CN
CONH2
CO2Me
NO2
NO
H
CN
CO2Me
CONH2
11.0a
-0.2
5.8b
-5.7
-5.8
1.4
5.3
(-5.5)c
-4.4
-5.5
(-5.7)c
(-5.8)c
-8.5d
-14.2e
a Exp. 9.22 (H2O). b Exp. 9.1 (H2O). c Appears twice in the table. d -6.7
for enol on the NO2 group. e -18.9 for enol on the NO group.
difference between eqs 5 and 6 or eqs 7 and 8. The values are
very high, being 36.8 kcal/mol for the H, H and H, Me systems
and 31.9 for the Me, Me systems. This gives the pKEnol values
of Table 2 when the literature values for acetamide and N,N-
dimethylacetamide at the B3LYP/6-31G** level of 28.9 and
28.8 kcal/mol3 are considered. (b) The major effect is indeed
Isodesmic Reactions. When presenting eq 1, we suggested
that the effect of EWGs on KEnol is mainly due to enol
stabilization by structure 1b. However, amide destabilization
probably due to repulsion between the lone pairs on the nitrogen
and the oxygen carbonyl was found to be considerable in several
previously published systems.1b We probed this question
together with the N-substituent effect by isodesmic reactions
5-8, which eliminate the interaction between the “functional”
and “substituent” amide or enol groups.
R1R2NCOCH(CN)CONR3R4 + CH2dC(OH)NR1R2 h
R1R2NCOC(CN)dC(OH)NR3R4 + CH3CONR1R2 (9)
stabilization of the enol, whereas destabilization of the amide
amounts to a lower percent of the overall effect. This differs
from systems which are activated by two cyano substituents.1c
(c) The overall effect of the methyl groups is not systematic.
Whereas the ∆G9 value is the smaller of the two Me groups,
the values for H, H and H, Me are identical. Clearly, the effect
of the two Me groups is smaller, but the effect of one H and
one Me group on the enol stabilization is the highest. Specu-
latively, this may be due to steric effect between the two Me
groups and the CN in the planar enol form. The effect on the
amide is the smallest for the H, Me system.
R1R2NCOCH(CN)CONR3R4 + CH4 h
CH3CONR1R2 + R4R3NCOCH2CN (5)
R1R2NCOC(CN)dC(OH)NR3R4 + CH4 h
CH2dC(OH)NR1R2 + R4R3NCOCH2CN (6)
Carboxamide as a Promoting Group for Enol Formation.
The high % enol in DMSO and DMF, the lower field δ(OH),
and the short H-bond (see below) give the impression that an
amido group is a better enol formation promoter than the
previously studied ester or even CN groups. Although there is
R1R2NCOCH(CN)CONR3R4 + H2CdCH2 h
CH3CONR1R2 + R4R3NCOC(CN)dCH2 (7)
no unequivocal correlation between the σ- value of a substituent
-
and its ability to promote enolization, we note that the σp- (σR
)
R1R2NCOC(CN)dC(OH)NR3R4 + H2CdCH2 h
CH2dC(OH)NR1R2 + R4R3NCOC(CN)dCH2 (8)
values of CO2R, CN, and CONH2 groups are 0.74, 1.02, and
0.62 (0.31, 0.26, 0.23), respectively,12 so that in our systems,
the weaker negative charge delocalizing substituent by resonance
(CONH2) is the better enol formation promoter.
Comparison of some calculated pKEnol values for EWG-
substituted amides including malonamides in the gas phase are
of interest (Table 4). For enolization of malonamide, methyl
malonamate, and cyanoacetamide, pKEnol ) 5.8, 5.3, and 11.0,
The calculations were conducted by transfer of an amide or
an enol group to either CH4 or H2CdCH2. The three sym-
metrical systems R1, R2 ) R3, R4 )H, H, H, Me, and Me, Me
were used. In the data given in Table 3, positive values indicate
that the “reactants” are more stable than the “products” and vice
versa. The following conclusions arise from Table 3. (a) The
stabilization of the malonamide/its enol pair compared with the
acetamide/its enol pair is obtained from eq 9, which is the
(12) Charton, M. In The Chemistry of Organolithium Compounds;
Rappoport, Z., Ed.; Wiley: Chichester, 2004; Vol. 1, Chapter 7, p 267.
5304 J. Org. Chem., Vol. 72, No. 14, 2007

Enols of Substituted Cyanomalonamides
TABLE 5. Comparison of “Enol Promoting” Groups in
RNHCOCH(CN)COX
respectively. This is due to the feasibility for hydrogen-bond
stabilization in the two former compounds which is lacking in
the cyanoacetamide. Williams’ experimental values in water13
are higher, being 9.1 (malonamide) and 9.22 (cyanoacetamide).13b
On replacing one of the hydrogens in YCH2CONH2 with a CO2-
Me group, KEnol increases for Y ) CN, CONH2, or CO2Me by
10.8-11.5 orders of magnitude, and the ester-substituted
malonamide is slightly more enolic than the other derivatives.
In the system YCH(CN)CONH2 for Y ) CO2Me and CONH2,
the % enol is nearly the same but is substantially less for Y )
CN, probably due to the lack of hydrogen bonding in the enol.
For malonamide itself, pKEnol ) 5.8. With an additional
carbamido group, the amide is still more stable than the enol
(pKEnol ) 1.4), due to the nonplanarity of the enol resulting
from steric repulsion between two NH2 moieties. However,
when other EWG substituents replace the H, the enol becomes
much more stable (Table 4): with CN or CO2Me groups, pKEnol
R
X
solvent
KEnol
g50
δ(OH)
14.37
Ph
OCH2CF3
CDCl3
THF-d8
CD3CN
CDCl3
THF-d8
CD3CN
CDCl3
THF-d8
CD3CN
CDCl3
5
1.44
9.0 (g50)
5.25 (32)
1.08 (4.9)
g50
10
2.33
9
14.42
14.04
17.65 (17.32)
18.18 (17.35)
17.98 (17.29)
14.37
14.00
14.04
17.71
Ph
NHMe(H)
OCH2CF3
NHMe(H)
An
An
THF-d8
CD3CN
CDCl3
CD3CN
CDCl3
THF-d8
CD3CN
DMF-d7
CDCl3
CD3CN
DMF-d7
CDCl3
5.7 (13.3)
1.33 (4.3)
g50
18.25 (17.54)
17.96 (17.43)
14.37
i-Pra
i-Pr
OCH2CF3
NHMe(H)
5.3
13.98
11.5 (g50)
9.0 (g50)
11.7 (5.25)
0.85 (3.55)
48
17.73 (17.53)
18.44 (17.89)
18.11 (18.39)
18.39 (18.00)
14.81
14.93
14.95
14.37
) -5.7 and -5.8. O2NCH(CO2NH2)2 is more enolic: pKEnol
)
i-Pra
i-Pra
OCH3
-8.5 and enolization on the nitro group is calculated to be 1.8
pKEnol units less favorable (pKEnol ) -6.7).10 This trend is
reversed for the nitroso-substituted system when the pKEnol
values are -14.2 and -18.9 for enolization on the amide and
the NO group, respectively. Indeed, NO compounds give the
oxime in CH-NO systems and are stable only in the absence
of H.14
1.8
1.9
OCH2CF3
g50
THF-d8
DMF-d7
g50b
c
a From ref 1e. b Probably ionizes completely. c Ionizes completely.
There are few other cyanodiamide systems that can be
experimentally compared with the analogous cyanoester-amide
systems. The effect of a CO2Me group in NCCH(CO2Me)-
CONHR cannot be compared when R ) Ph since the signals
for the enol and the amide coalesce at rt1b and the KEnol value
is unavailable. However, comparison is possible when R )
i-Pr.1e From Table 5 for COX, KEnol(CONH2) > KEnol(CO2Me)
> KEnol(CONHMe) in both CDCl3 and CD3CN. Comparison
of δ(OH) values shows that the OH signal is at a >3 ppm lower
field in the diamides than in the ester amides. This applies to
the other amide/ester comparisons below. However, δ(OH) in
the diester amides is much closer to that in the cyanodiamides.
From the KEnol values, CO2CH2CF3 is a better enol formation
promoter1c than the N-methylated CONHMe in CDCl3 and in
CD3CN, but the relationship is inverted in one of the two cases
studied in THF-d8. However, a non-N-methylated CONH2 group
is somewhat more activating than CO2CH2CF3, and the effect
is larger in THF for R ) Ph, An, whereas for R ) i-Pr the two
groups show similar effect in CD3CN. This is in line with the
structure of the isolated crystals, and is there a correlation
between the % enol in a certain solvent (e.g., the pKEnol in
CDCl3) and the structure of the isolated solid species, similar
to that found previously?1c (b) Do the CdC bond lengths in
the enols elongate as observed with other enols substituted by
EWGs,1 and if so, to what extent? (c) What is the substituent
effect on the preferred enolization site in systems carrying two
different CONRR′ groups, and are the two possible regioisomers
formed in one system? (d) Is the configuration of the solid enol
E or Z, and is it determined by intramolecular hydrogen bonding
as found previously?1c-f (e) Can we isolate crystals of both
isomeric enol and amide under different conditions. (f) What
are the substituent and temperature effects on the nature and
symmetry of the hydrogen bond? (g) What is the effect of a
possible resonance-assisted hydrogen bond on crystallographic
parameters of the enol? (h) Are NHR groups which are not
intramolecularly hydrogen bonded, intermolecularly bonded?
Concerning question (h), the enols take advantage of hydrogen-
bonding stabilization when possible. Intramolecular hydrogen
bonds between OH and O and NRR′ groups are discussed below,
but when these cannot be formed, intermolecular H-bonds are
formed. The CN group is incapable of forming an intramolecular
H-bond in our systems due to its geometry, but a common
feature is that the cyano group’s nitrogen on one enol molecule
is hydrogen bonded to the N-H group of another enol
molecule.1b,d Figure 3 shows such an intermolecular dimer for
5o, with a CN‚‚‚H-N hydrogen bond length of 2.193 Å, an
N‚‚‚N nonbond distance of 3.030 Å, and a N‚‚‚H-N angle of
158.85°. The O‚‚‚O nonbonding distance in the intramolecular
O-H‚‚‚Od hydrogen bond is much shorter, 2.408 Å, and hence
much stronger. A similar intermolecular CN‚‚‚H-N hydrogen
bond in enol MeCO2C(CN)dC(OH)NHPh1b is of approximately
the same length (2.07 Å).
calculations above. Finally, KEnol(CO2CH(CF3)2) > KEnol
-
(CONHMe),1c although we cannot conclude unequivocally that
K
Enol(CO2CH(CF3)2) > KEnol(CONH2) since the hexafluoroester
derivative ionizes in DMF-d7.1b However, the inequality seems
to hold, as judged by indirect comparison of other values in
Table 5. We conclude that there is no special effect in
malonamides on KEnol values compared with ester-substituted
systems, except for the relatively high KEnol values in DMSO
and DMF.
Crystallographic Data. We conducted an extensive crystal-
lographic structure determination of many amide/enol systems
in order to answer the following questions: (a) What is the
(13) (a) Williams, D. L. H.; Xia, L. J. Chem. Soc., Chem. Commun.
1992, 985; J. Chem. Soc., Perkin Trans. 2 1993, 1429. (b) Eberlin, A.;
Williams, D. H. L. J. Chem. Soc. 2002, 1316.
(14) (a) Smith, M. B.; March, J. March’s AdVanced Organic Chemistry,
5th ed.; Wiley: New York, 2001; p 76. (b) The oxime was identified as
the nitrosation product of methylmalonic acid: Barry, R. H.; Hartung, W.
H. J. Org. Chem. 1947, 12, 460.
We determined 13 structures of amide/enol systems by X-ray
crystallography. Although the % enol in CDCl3 is high (g80%)
for all of them, four of them display an amide structure, and
nine display an enol structure in the solid (Table 6). In none of
J. Org. Chem, Vol. 72, No. 14, 2007 5305

Basheer et al.
Table 6 gives selected crystallographic data for the 13 amide/
enol structures. The full data are in the crystallographic
information files in the Supporting Information. Interesting
parameters are the bond lengths for the pairs of C-C, C-O,
and O-H bonds, their differences ∆CO, ∆CC, and ∆OH, the
nonbonded O‚‚‚O distance, and the O(1)HO(2) hydrogen bond
angle. Also given is the % enol in CDCl3, which is our least
polar solvent and hence the one that mostly “resembles” an
isolated molecule, and δ(OH) in CDCl3. The calculated (B3LYP/
6-31G**) bond lengths and angles for several of the compounds
for which experimental data are available are given in Table 6
in bold font.
As expected, in the four amide structures (entries 3, 5, 6,
and 10), the ∆CC and ∆CO values are small and within the
combined experimental errors. Only enol 5j/6j has a low R factor
(0.12), and the hydrogen position is of low accuracy, but the
enol structure is clear. Seven of these structures were determined
at a single temperature, but those of 6a were determined at 293
and 123 K and of 5h at 295 and 123 K. The major temperature-
dependent difference is in the O-H‚‚‚O bond. At 123 K, the
two O-H bonds were somewhat longer than the quoted
“normal” O-H bond of 0.967 Å,16 being 1.05 Å (6a) and 1.02
Å (5h). The corresponding O‚‚‚H lengths d6 were 1.39 and 1.47
Å, respectively. At the higher temperature, there is an approach
to O-H bond equalization for 6a (Figure 5), whereas for 5h,
the bond lengths are closer than at 123 K but still differ
significantly (1.17 and 1.30 Å).
FIGURE 3. The intermolecular CN‚‚‚H-N bond in the solid structure
of enol 5o. (The term “_3” stands for the (crystallographic) operation
-x, -y, 1-z needed to produce the second molecule.)
our systems did we isolate separate crystals of the amide and
the enol. We had recently isolated separate crystals of the
cyanothiomalonamide MeNHCSCH(CN)CONHMe and its enol,15
a related system to our compounds.
For all systems with an amide structure, R1 ) Me, whereas
R2 ) Me, R3 ) p-An, i-Pr, or R2 ) H, R3 ) Ph, t-Bu (4b, 4d,
4g, and 4k with 80, 80, 90, and 84% enol, respectively, in
CDCl3). These values are mostly lower than those for the other
species having an enol structure. Consequently, there is no strong
correlation between the % enol in solution and the solid-state
structure. The solid-state structure is not necessarily that of the
main species observed in solution, in contrast with our previous
experience with the diester-substituted amides.1c,f The X-ray
structure of 4k is given in Figure 4.
A similar trend of higher differences (higher ∆OH) at the lower
temperature is deduced from the collected single temperature
∆OH values for other enols: -0.48 (6i at 173 K), -0.50 (5s )
6s at 123 K), -0.13 (5n at 123 K), and -0.05 (6q at 295 K).
This is not always the case: a low value of -0.196 (for 5o at
100 K) and a relatively high value of -0.45 (for 6c) at 295 K
were also found. We realize that, due to the general difficulty
in locating the hydrogen, especially at the higher temperature,17
the ∆OH parameter is a priori of a lower accuracy than the ∆CC
and ∆CO parameters. The observed O-H‚‚‚O bond angles are
153-164°.
The Solid-State Enols. (a) Observed Structures and
Temperature Effect. The CONMe2/CONHPr-i system is 80%
enolic in CDCl3 but gives amide crystal structure at 298 K. In
contrast, when the CONMe2/CONHPh analogue (6a) is crystal-
lized from CDCl3, EtOAc/petroleum ether mixture, or DMF-
d7/H2O mixture, it displays at 293 K an enol structure (Figure
5), whose O-H bond is formed on the NMe2-bound carbonyl.
The two O-H(O‚‚‚H) distances of the hydrogen-bonded enol
moiety O-H‚‚‚O are d5 ) O-H ) 1.15 Å, d6 ) H‚‚‚O ) 1.26
Å; that is, the difference ∆OH ()d6 - d5) of 0.11 Å is not very
large. The O‚‚‚O distance of 2.38 Å is very short, and the
O-H‚‚‚O angle is 158°. The enolic CdC bond is elongated to
1.417 Å, and the C-O bond is a single bond at 1.293 Å.
Consequently, the two C-C bonds around the cyano-bound
carbon differ by only 0.08 Å, whereas the formal CdO and
C-O bonds differ by 0.042 Å.
(b) Calculated Structures. The eight calculated structures
(“Cal” in Table 6) reasonably match the X-ray determined
structures, remembering that the calculations are for a single
molecule in the gas phase and for a static molecule. A dynamic
process leads to averaging of the bond lengths of two isomeric
static structures (see below) and is partially responsible for the
difference between some of the observed and calculated values.
The calculated O-H‚‚‚O bond angles are 153.0-156.4°, that
is, mostly somewhat lower than the X-ray values. Other
calculations are discussed below.
If the tendency for bond equalization is due to a rapid
hydrogen shift between two regioisomeric enols (see below), it
should be reflected in a similar tendency for temperature-
dependence bond equalization in the C-C or C-O bonds, that
is, to lower ∆CC and ∆CO values at a higher temperature. The
problem with this probe is that the difference in the relevant
From the data for 6a at a lower temperature of 123 K (Figure
5), the main feature is that the OH is still on the same former
carbonyl, but the O-H bond is somewhat elongated (∆OH
)
0.34 Å), and the H-bond is much less symmetric. The bond
angle of 159.5(19)° is nearly the same as that in 293 K.
Interestingly, ∆CO ) (d(C-O) - d(CdO) ) C(1)O(1)-
C(3)O(2) ) d1 - d4) remains nearly the same, whereas the C-C
bonds differ by only ∆CC ) (d(CdC) - d(C-C) ) C(1)C(2)-
C(2)C(3) ) d2 - d3) ) 0.012 Å. A similar temperature effect
is shown also for the CONHMe/ CONHAn-p (5h) enol.
(16) Allen, F. H.; Kennard, O.; Watson, D. G.; Brammer, L. E.; Orpen,
A. G.; Taylor, R. J. Chem. Soc., Perkin Trans. 2 1987, S1.
(17) (a) Dunitz, J. D. X-ray Analysis and the Structure of Organic
Molecules; Cornell University Press: Ithaca, NY, 1979. (b) Allen, F. H.;
Bellard, S.; Brice, M. D.; Cartwright, B. A.; Doubleday, A.; Higgs, H.;
Hummelink, T. W. A.; Hummelink-Peters, B. G. M. C.; Kennard, O.;
Motherwell, W. D. S.; Rodgers, J. R.; Watson, D. G. Acta Crystallogr.
1979, B35, 2331. (c) Allen, F. H.; Davies, J. E.; Galloy, J. E.; Johnson, J.
J.; Kennard, O.; Macrae, C. F.; Mitchell, G. F.; Smith, J. M.; Watson, D.
G. J. Chem. Inf. Comput. Sci. 1991, 31, 187.
(15) Basheer, A.; Rappoport, Z. Org. Lett. 2006, 8, 5931.
5306 J. Org. Chem., Vol. 72, No. 14, 2007

Enols of Substituted Cyanomalonamides
J. Org. Chem, Vol. 72, No. 14, 2007 5307

Basheer et al.
FIGURE 4. ORTEP structure of amide 4k.
FIGURE 6. Different PESs for the hydrogen bond.
temperature, but at 295 K, it is at the limit of the combined
experimental errors. For 6c at 295 K, the difference is significant
as shown by the ∆OH values. For 5j/6j, 5n, and 6q at 295 K,
the ∆CO values are close to zero considering their errors, whereas
for 5s ) 6s at 123K and for 6i at 173 K, ∆CO is significant, and
for 5o at 100 K, the lengths differ slightly.
Gilli suggested that the coordinate of the antisymmetric
vibration of the enolone moiety Q ) d1 - d2 + d3 - d4 ) ∆CO
- ∆CC is a better measure than the individual ∆CC and ∆CO
terms and tabulated a variety of Q values for many enol
systems.20a,c,g Remarkably low Q values of 0.013, 0.015, 0.005,
and 0.008 were found for 5j/6j, 5n, 5o, and 6q, respectively,
whereas the eight other values for seven enols were 0.061 (
0.006. The calculated seven values are higher, being 0.083 (
0.012.
A major concern in the solid-state structure is the nature of
the hydrogen bond. The two amido groups are structurally
mostly similar and sometimes identical, raising the question of
symmetry of the bond. The nature of the hydrogen bond is
usually discussed in terms of the four potential energy surfaces
(PES) shown in Figure 6.19 When the energies of the two
possible hydrogen-bond arrangements formed by a proton shift
between the oxygens differ significantly, the PES is an asym-
metric double-well potential (Figure 6a). If the energies are
similar, the symmetric double-well potential PES (Figure 6b)
is obtained. A decrease of the proton-transfer barrier leads to
the low barrier hydrogen bond PES (Figure 6c), and if the barrier
is very small or disappears, a single-well potential PES (Figure
6d) is obtained. The stronger the hydrogen bond, the lower the
barrier. The hydrogen bond strength is related to the nonbonded
O‚‚‚O distance, and distances of <2.5 Å are ascribed to strong
hydrogen bonds.19 In our enols, the values are 2.38-2.46 Å
(Table 6), the bonds are strong hydrogen bonds, and the barriers
for hydrogen transfer are presumably low.
FIGURE 5. ORTEP structure of enol 6a at 293 K (no brackets) and
123 K [in square brackets].
bond lengths is a priori not large. The CdC bond length is 1.332
Å, and in enones CâdCR-CdO, it is 1.340 Å, whereas the
central formal single bond length is 1.464 Å.16 However, our
enols are dipolar push-pull alkenes with an important contribu-
tion from structure HO(RR’N)C+_C-(CN)CONR′′R′′′, so that
d2 . 1.332 Å, and in many activated enols similar to 5 and 6,
it is usually 1.41-1.42 Å,1 and the ∆CC value should be
relatively small even at low temperature.18 The C-O bond in
enols is 1.333 Å, and a CdO bond in the enones is 1.222 Å.16
However, in the dipolar structure, the CdO double bond is
elongated due to some single bond character, and ∆CO is
expected to decrease. Indeed, even at the low temperature, the
∆
CC
and ∆CO terms are relatively small. Only in four of the
enols is the ∆CC value higher than the combined experimental
error in d2 and d3, and in three remaining cases, it is <0.01 Å.
Only for 5h, ∆CC of 0.03 Å at 123 K becomes 0.01 Å at 295 K.
Note that the two bond lengths of 6a are identical within their
combined errors at 123 K but are somewhat different at 298 K.
Consequently, the ∆CC probe alone is of little help in studying
the dynamic process.
A similar analysis of the CdO and C-O bonds is more
encouraging. In contrast with the ∆CC values of -0.05 to -0.021
Å for all systems, except of -0.039/-0.037 for 5h, the ∆CO
values are larger, being 0.003-0.055 Å, indicating a potentially
higher change with the temperature. For 6a, the two C-O bonds
differ significantly, but the ∆CO is almost temperature-
independent. For 5h, ∆CO is somewhat higher at the lower
(19) Perrin, C. L.; Nielson, J. B. Annu. ReV. Phys. Chem. 1997, 48, 511.
(20) (a) Gilli, G.; Belluci, V.; Ferretti, V.; Bertolasi, V. J. Am. Chem.
Soc. 1989, 111, 1023. (b) Gilli, G.; Bertolasi, V. In The Chemistry of Enols;
Rappoport, Z., Ed.; Wiley: Chichester, 1990; Chapter 13, p 713. (c)
Bertolasi, V.; Gilli, P.; Ferretti, V.; Gilli, G. J. Am. Chem. Soc. 1991, 113,
4917. (d) Bertolasi, V.; Gilli, P.; Ferretti, V.; Gilli, G. J. Chem. Soc., Perkin
Trans. 2 1997, 945. (e) Gilli, P.; Bertolasi, V.; Ferretti, V.; Gilli, G. J. Am.
Chem. Soc. 2000, 122, 10495. (f) Bertolasi, V.; Gilli, P.; Ferretti, V.; Gilli,
G. New J. Chem. 2001, 25, 408. (g) Gilli, P.; Bertolasi, V.; Pretto, L.;
Ferretti, V.; Gilli, G. J. Am. Chem. Soc. 2004, 126, 3845.
(18) Note that the hydrogen bond strongly reduces the polarity of the
enols, as shown by the solvent effect on pKEnol values.
5308 J. Org. Chem., Vol. 72, No. 14, 2007

Enols of Substituted Cyanomalonamides
FIGURE 7. Resonance-assisted hydrogen bond of two equilibrating structures.
Following Gilli’s analysis for the structurally related enols
of 1,3-diketones,20 and the temperature effect on the bond
lengths, the enol is depicted as two structures in rapid equilib-
rium, each of which is stabilized by resonance, that is, being a
resonance-assisted hydrogen-bonded structure (Figure 7).20
As shown in Figure 7, both enols have a Z configuration,
regardless whether hydrogen migration takes place. If it occurs,
together with the known rapid rotation around the formal single
CR-Câ bond character in the zwitterionic enol,1c four enols can
be potentially observed, that is, E,Z pairs of each of the two
regioisomeric enols on the two amido carbonyls. The observa-
tion of only a single enol, in contrast with the observation of
E,Z pairs for enols RNHC(OH)dC(CO2R′)CO2R,1c,f indicates
much less stable E structures. This is not surprising since the
only intramolecular hydrogen bonds possible in them are
O‚‚‚H-N and N‚‚‚H-N bonds, which are weaker than the
O‚‚‚H-O bond.6
The temperature effect on ∆OH, ∆CC, and ∆CO values (Table
6) may suggest an apparent more symmetric structure at the
higher temperature. This is ascribed to a rapid proton shift
between the regioisomeric enol structures (Figure 7). At a low
temperature, this process is relatively slow, and mainly one static
structure of the double minimum surface is observed. At a higher
temperature, the hydrogen transfer rate approaches the measure-
ment time scale and we see higher “formal” symmetry at 293
and 295 K due to partial or complete averaging of the bond
lengths of the two equilibrating structures (cf. Figure 6c). The
error in the X-ray determination of the hydrogen position calls
for a corroboration of this suggestion, which is obtained from
the DFT calculations (Table 6), which show in all cases at 0 K
a nonsymmetrical structure, even for 5s ) 6s, which is a
chemically symmetrical system.
The calculated O-H bond lengths for 6a and 5h are
somewhat longer than the normal value for an O-H bond, and
the O-H and O‚‚‚H bond lengths are much closer to the
experimental values at 123 K than at 293 K; for 6i, they fit the
values at 173 K considering the experimental error and for 5n
the calculated ∆OH(cal) > ∆OH(exp) even at 100 K. Due to this
error, the comparison between the calculated and experimental
C-C and CdC bond lengths is less satisfactory, although the
calculated CdC bond is longer (or identical for 5n) than the
experimental value, whereas the trend for the C-C bond is less
clear. Consistently, ∆CO(cal) > ∆CO(exp). The conclusion from
all these data supports the assumption of a nonsymmetrical
hydrogen bond with a slow hydrogen movement between two
regioisomeric enols at the low temperature and a faster shift at
the higher temperature. We assume a faster hydrogen shift in
solution, and since our measurements were at room temperature,
a fast mutual isomerization of enols 5 and 6 takes place in
solution.
compounds show the two parameters in this region (Table 6),
but good correlation between them was not observed. The four
compounds with δ(OH) ) 17.7-17.8 display O‚‚‚O distances
of 2. 38-2.46 Å, and the lowest δ(OH) ) 16.23 corresponds
to O‚‚‚O of 2.415 Å. This can be understood since the three
lower δ(OH) values in Table 6 are when R3 ) C6F5, an EWG.
Both the electron withdrawal and the ring current may affect
the δ(OH) and probably do not affect the O‚‚‚O in a similar
way.
The Enolization Site. The enolization site in the crystals of
solid enols is deduced from X-ray crystallography. Due to our
assumption of the rapid proton shift, there is little meaning to
the “site of enolization” at rt for several systems. No general
rule for all X-ray determined structures can be deduced from
Table 6, but the calculated energy difference between the two
“regioisomeric” enols is low. Remembering these limitations,
three of our five systems bearing an aryl group on one nitrogen
and a single alkyl or two alkyls on the other nitrogen enolize
on the alkyl-substituted amido group, and two carrying an
electron-withdrawing p-anisyl group (5h) and an electron-
donating C6F5 group (5n) enolize near the aryl site. When an
i-PrNH group competes with H2N or MeNH group, enolization
is near the i-PrNH.
The energy differences between regioisomeric enols of
cyanomalonamides were calculated for seven cases (Table S4)
and account for the lack of a generalization for the enolization
site. The differences are very small, being 0.01-0.7 kcal/mol
for 5j, 5m, 6c, 6d, 6n, and 6o over 6j, 6m, 5c, 5d, 5n, and 5o,
respectively. Hence, in half of the cases, there is ca. 1:1 ratio
of both isomers, and in the other cases, there is at least 23% of
the minor isomer. The barriers for the regioisomer intercon-
version are presumably low, based on the temperature effect in
the solid enols and the O‚‚‚O values, and hence both species
exist in appreciable amounts in solution. Since the amide is
sometimes formed when the enol is g80% of the mixture in
solution, and with an energy difference of <0.6 kcal/mol
between the two isomers, either of them can be crystallized,
depending on differences in crystal packing. Consequently, it
is not surprising that the calculations of Table S4 match two
X-ray determined isomers, do not match it in other two cases,
and the amide is formed in another case.
IR Spectra (a) of the Solid Species. The solid-state structures
of systems for which we lack X-ray crystallographic data were
analyzed by IR spectra of the solid species in KBr or in nujol.
The absorptions were very similar in both media. The regions
of interest are those of the CN and the OH. The 2100-2300
cm-1 region is that applied for detecting CN stretch, although
usually, the common range studied is 2260-2200 cm-1, with
stronger absorption toward its lower end for conjugated cyano
groups.21 In cyano carbanions, the wavenumbers are lower than
those of conjugated nitriles.21b Our 20 solid systems studied
show two types of behavior. Five show mainly or exclusively
a signal in the narrow range of 2252-2257 cm-1, and 4k
displayed absorption at 2264 cm-1. These are saturated CN
group stretchings and compounds 4b, 4d, 4g, and 4k whose
O‚‚‚O and δ(OH) Values. A short O‚‚‚O distance in the
solid and a low-field δ(OH) in solution are regarded as probes
of strong hydrogen bonds.20 A correlation between the two
parameters is close to linear, and distances of 2.38-2.46 Å are
associated with δ(OH) values of >16 ppm.20c All of our
J. Org. Chem, Vol. 72, No. 14, 2007 5309

Basheer et al.
TABLE 7. IR Absorptions for R3NHCOCH(CN)CONR1R2 (KBr,a cm-1
)
R3
R1
R2
compd
νCN
νCO
νNH
νOH
2560
νother
Group Ab
t-Bu
t-Bu
Ph
t-Bu
i-Pr
H
H
H
Me
Me
Me
H
4p
4k
4g
4e
4d
4b
2255.5
1715, 1684
1697, 1653
1702, 1663
∼1700, 1682
∼1670
3392, 3310, 3076
3328, 3126, 3061
3323, 3281, 3209,3147, 3110, 3056
3761, 3287, 3078
2751
Mec
Mec
Me
2264
2254
2253.5
2252
2257
2660
1966
2600
2550, 2655
2680
Mec,d
Mec
3299, 3077
p-An
1663
3486, 3417, 3313, 3256
1899, 2046, 2530
Group Bf
i-Pr
H
H
H
H
H
H
H
H
5o
2200, 2148g
2200, 2149g
2202
1643
1637
1648
1648
-
1633
1673
1645
1682
1640
1650
1651
1651
3416, 3336, 3243
3437, 3342, 3309
3422, 3319, 3248
3420, 3324, 3048
3340, 3313
3324, 3214, 3025
3392, 3322, 3287, 3208, 3136, 3069
3411, 3251
2698
2699
2550, 2675
2680
2474, 1811
1765, 2464
C6F5
p-An
Ph
5n
5/6m
5l
2201, 2180g
2195, 2166g
2213, 2161g
2197, 2253
2207,2195h
2198,2254g
2189, 2137g
2196, 2137g
2196,2144g
2204,2192h
1788
i-Pr
H
H
H
H
H
H
H
Me
Me
Me
Me
Mee
Ph
CHPh2
i-Pr
Ph
5j
6i
2685
2670
2600
2674
1917, 2475
2468
1887
ca. 1900
1880,1900
1945,2435
C6F5
p-An
C6F5
CHPh2
i-Pr
Ph
Ph
C6F5
4h
5q
5t
5/6s
5/6r
5g
3289
3338
2685
3489, 3406, 3257, 3057g
3293, 3195, 3065
3241
Me
Me
2630
2660
1800,1906,2470
1901,2454
5i
a Also in nujol. b Amide according to IR. c Amide according to X-rays. d With D2O: new signals at 2415, 2450 cm-1
.
e Enol according to X-rays. f Enol
or mostly enol according to IR. g Weak. h Medium.
TABLE 8. IR Absorptions (cm-1) of Selected Formal
R1R2NHCOCH(CN)CONHR3 in CHCl3 Solution
X-ray diffraction was determined, belong to Group A of Table
7. The correspondence between X-ray structure and IR data of
the solid enables the determination, with some confidence, of
the solid-state structure of a species whose X-ray data are
unavailable.
R3
R1 R2 compd
νCN
νCO
νNH
p-An
C6F5
H
H
H
H
5/6m
5/6n
2194
2206
1670 3430, 3392
1648 3420 (br),
3356, 3302
All of the three solid amides with one t-BuNHCO group and
a second H2NCO, MeNHCO, or Me2NCO group (100, 84, and
73% enol in CDCl3, respectively) belong to this group. They
t-Bu
H
H
5/6p
2192
1698 (w) 3524, 3483,
3416 (br)
i-Pr
C6F5
t-Bu
H
H
H
i-Pr 4/5/6o 2187, 2259 (w)
1697 3420, 3327
1644 3435, 3331
1716 3420, 3347
Me
5/6i
2197
also display CO signals at 1697-1715 and 1653-1684 cm-1
.
,
Me 4/5/6t
2169, 2253
The other three compounds display signals at 1663-1702 cm-1
Ph Me Me 4/5/6a
C6F5 Me Me 5/6c
t-Bu Me Me 4/5/6e
2188, 2255 1654, 1700 3409, 3299
2194 1648, 1717 3403, 3252
the approximate ranges of solid primary and secondary amide
absorptions.21a
2183, 2254
1655 3419, 3307
All Group A compounds display 3-6 absorptions in the
3486-3056 cm-1 range, which are assigned to the several amido
NH groups. Table 7 give the CN, NH, and CdO absorptions in
KBr.
When D2O is added to solid 4d, new signals appear at 2415
and 2450 cm-1, which are ascribed to the labeled CONDPr-i
and OD groups.
small signals which appear in the 1800 and 1900 cm-1 ranges
for most compounds were not identified.
Solid compounds 4h and 5t, which display CN signals in
both the saturated and the unsaturated regions, apparently
contain significant percentages of both the amide and enol.
IR Spectra (b) in CHCl3 Solution. In CHCl3, the CONH2-
substituted systems and some other derivatives have very low
solubility, resulting in low-intensity signals. Almost all com-
pounds display conjugated CN signals at 2169-2206 cm-1. A
weak signal for a saturated CN does not appear in the CONH2
derivatives, but a medium or weak signal at 2252 ( 3 cm-1 is
observed in most other derivatives. CO signals at 1655-1715
cm-1 are observed when the saturated CN signal is significant.
There are signals >3050 cm-1 for all compounds, but an
extensive band at 3200-2500 cm-1, including the O-H‚‚‚O
peak (see above), was not observed, except for a weak signal
at 2669 cm-1 for 5i. Consequently, the spectra are not
informative in relation to the hydrogen bonding.
Most of the compounds (Group B, Table 7) display a strong
signal at 2189-2207 cm-1 (mostly at the narrow range of
2195-2200 cm-1) and at 2213 cm-1 in one case. These are in
the range and even lower than that ascribed to unsaturated
nitriles, such as the enols. Each of them displays an additional
weak signal at 2136-2185 cm-1 which may be relevant to the
intermolecular N-H/CN H-bonds in the solid. They also display
signals which may be the amide I N-H bending signal at 1633-
1651 cm-1 21a Several signals appear at >3047 cm-1 which we
.
ascribe to NH stretchings, but even in the enols, there is still
one amido group which absorbs >3050 cm-1. The weak
absorption in the range of 2560-2699 cm-1 is ascribed to the
strong O-H‚‚‚O bonds, based on the weak22a νOH of 250022b
or 2566-2675 cm-1 for enols of 1,3-diketones.20c Additional
Table 8 gives CN, CO, and NH absorptions in CHCl3 for
selected enols/amides, and Table S5 in the Supporting Informa-
tion gives additional IR data. Enol signals are observed in all
(21) (a) Silverstein, R. M.; Bassler, G. C.; Morrill, T. C. Spectrometric
Identification of Organic Compounds, 4th ed.; Wiley: Chichester, 1981;
pp 129-130. (b) Juchnovski, I. N.; Binev, I. G. In The Chemistry of
Functional Groups, Supplement C: The Chemistry of Triple-bonded
Functional Groups; Patai, S., Rappoport, Z., Eds.; Wiley: Chichester, 1983;
Chapter 4, p 107.
(22) (a) Kopteva, T. S.; Shigorin, D. N. Russ. J. Phys. Chem. (Engl.
Transl.) 1974, 48, 532. (b) Vener, M. V. In Hydrogen-Transfer Reactions;
Hynes, J. T., Klinman, J. P., Limbach, H.-H., Schowen, R. L., Eds.; Wiley-
VCH: Weinheim, Germany; Vol. 1, p 273.
5310 J. Org. Chem., Vol. 72, No. 14, 2007

Enols of Substituted Cyanomalonamides
and overnight at rt. No precipitate was formed, even after addition
of NaOH (0.4 g) and additional stirring for 48 h. The mixture was
extracted with AcOEt (2 × 100 mL), the organic phase was dried
(Na2SO4), and the solvent was evaporated, leaving an oil which
gave a crystalline compound after standing overnight. After washing
with ether, 14.2 g (0.125 mol, 63%) of N,N-dimethylcyanoaceta-
mide, mp 68 °C (lit.29 62 °C; lit.30 58 °C), was formed: 1H NMR
(CDCl3, 298 K) δ 2.94 (3H, s, Me), 3.01 (3H, s, Me), 3.50 (2H, s,
CH2).
Reaction of N,N-Dimethylcyanoacetamide with Phenyl, p-
Anisyl, Pentafluorophenyl, Isopropyl, and tert-Butyl Isocyanates.
The procedure described for phenyl isocyanate was also applied
for the other four isocyanates: To a stirred suspension of Na (0.25
g, 11 mmol) in dry THF (50 mL) was added N,N-dimethylcyano-
acetamide (1.12 g, 10 mmol). After 4 h, all the Na reacted, hydrogen
evolution was ceased, and the sodium salt of the amide precipitated
as a white solid. Phenyl isocyanate (1.08 mL, 10 mmol) was added,
the solid was dissolved, and the solution turned orange-yellow. The
mixture was refluxed overnight, and the solvent was evaporated,
leaving the sodium enolate. It was dissolved in DMF (10 mL), and
the solution was added dropwise with stirring and cooling to a cold
2 N HCl solution (100 mL). The white oily solid formed was
extracted with ethyl acetate (2 × 100 mL). The organic layer was
dried (Na2SO4), most of the solvent was evaporated, and to the
remaining 10 mL was added petroleum ether (5 mL), and the
solution was kept overnight in the refrigerator. The white solid
obtained was filtered and dried, giving 1.28 g (5.54 mol, 55%) of
the enol/amide mixture 4a/5a/6a, mp 125-126 °C. Anal. Calcd
for C12H19N3O2: C, 62.34; H, 5.63; N, 18.18. Found: C, 61.97;
H, 5.94; N, 18.11.
Suitable crystals of 6a for X-ray diffraction were obtained by
dissolving the crude solid in ethyl acetate, adding petroleum ether,
and keeping the solution in the refrigerator for 1 week: 1H NMR
(CDCl3 298 K) displayed signals for 4:1 enol/amide mixture δ (enol)
δ 3.23 (6H, s, 2Me), 7.13 (1H, t, J ) 7.3 Hz, p-H), 7.33 (2H, t, J
) 8.3 Hz, m-H), 7.43 (2H, d, J ) 7.9 Hz, o-H), 7.67 (1H, br s,
NH), 17.79 (1H, s, OH); δ (amide) 3.02 (s, Me), 3.18 (s, Me),
4.80 (s, CH), 7.13 (t, p-H, overlaps), 7.33 (t, m-H, overlaps), 7.53
(d, J ) 7.9 Hz, o-H), 9.16 (br s, NH); 13C NMR (CDCl3, 298 K)
δ (enol) 38.33 (q, J ) 140.0 Hz, Me), 55.60 (s, Câ), 120.25 (s,
CN), 121.19 (d, J ) 162.3 Hz), 124.77 (dt, Jd ) 163.0 Hz, Jt )
6.8 Hz), 128.89 (dd, J1 ) 161.5 Hz, J2 ) 8.0 Hz), 136.59 (t, J )
8.8 Hz), 171.09 (m), 172.06 (d, J ) 3.0 Hz); δ (amide) 36.71 (q,
J ) 143.2 Hz, Me), 38.15 (q, J ) 137.7 Hz, Me), 44.08 (d, J )
138.9 Hz, CH), 113.40 (d, J ) 11.3 Hz, CN), 120.25 (d, J ) 163.8
Hz), 125.27 (dt, Jd ) 162.3 Hz, Jt ) 7.5 Hz), 128.93 (dd, J1 )
161.5 Hz, J2 ) 8.0 Hz), 136.61 (overlaps, 157.63 (d, J ) 8.3 Hz),
161.34 (s).
The 4-methoxyphenyl derivative (4b/5b/6b), mp 104 °C, was
prepared similarly in 52% (1.37 g) yield. The pentafluorophenyl
derivative (4c/5c/6c) was obtained similarly in dry ether as the
solvent. The precipitate was obtained after the acidification without
extraction by EtOAc. The product, mp 128-129 °C, was obtained
in 49% (0.78 g) yield. The i-Pr (4d/5d/6d) and t-Bu (4e/5e/6e)
derivatives, mp 114-115 and 142-143 °C, respectively, were
prepared similarly in 20 and 25% yield, respectively, except that
acidification gave no solid, which was obtained only after extraction
with EtOAc and evaporation of the solvent. Crystals of the i-Pr
derivative for X-ray diffraction were obtained by cooling its solution
in EtOAc/petroleum ether. The triphenylmethyl derivative (4f/5f/
6f), mp 157-159 °C, was prepared similarly in 74% yield. The
spectral and analytical data are given in Tables 1, S2, S3, and S6.
Reaction of N-Methylcyanoacetamide with Phenyl, p-Anisyl,
Pentafluorophenyl, Isopropyl, and tert-Butyl Isocyanates. The
solutions, and additional amide signals are observed in several
of them, as judged by similar CN absorptions which are similar
to those in the solid state. Comparison with the data in Table 1
shows that only unsaturated CN absorptions appear for com-
pletely enolic systems in CHCl3, whereas saturated CN absorp-
tions appear for systems which display both enol and amide in
solution.
Conclusions. All of the N-substituted cyanomalonamides
studied form the Z-enols in solution. A rare feature is the
presence of an appreciable % enol in DMF-d7 and DMSO-d6.
The temperature-dependent bond length changes in the X-ray
structure suggest a dynamic proton transfer between regioiso-
meric enols in the solid state. The short O‚‚‚O nonbonded
distances indicate a strong hydrogen bond, and calculations
indicate it to be nonsymmetrical. Substituent and solvent effects
were determined. The isolated crystals are not necessarily those
of the more abundant species in solution.
Experimental Section
1
General Methods. Melting points, H and 13C NMR and IR
spectra were measured as described previously.23 All of the
commercial precursors and solvents were purchased from Aldrich.
N-Methylcyanoacetamide. To methyl cyanoacetate (19.8 g, 0.2
mol) at -10 °C was added a solution of 40% aqueous methylamine
(25 mL, 0.29 mol) dropwise with stirring, while retaining the
temperature below 0 °C. A white solid was obtained, and the
mixture was stirred for an additional 1 h at 0 °C and for 3 h at
room temperature. Most of the solvent was evaporated, the
remaining solid was filtered and dissolved in ethyl acetate (200
mL), the solution was dried (Na2SO4), and the solvent was
evaporated, giving 15.7 g (0.16 mol, 80%) of N-methylcyano-
acetamide, mp 108 °C (lit.24 101 °C): 1H NMR (DMSO, rt) δ 2.60
(3H, d, J ) 4.6 Hz, Me), 3.58 (2H, s, CH2), 8.14 (1H, br s, NH).
N-Isopropylcyanoacetamide. To methyl cyanoacetate (9.9 g,
0.1 mol) at 0 °C was added isopropylamine (8.5 mL, 0.1 mol)
dropwise at <10 °C. After stirring for 1 h, water (10 mL) and NaOH
(0.1 g) were added and the white solid N-isopropylcyanoacetamide
obtained (4.8 g, 0.04 mol, 40%), mp 82 °C (lit.24 65 °C; lit.25 73-
74 °C), was filtered: 1H NMR (CDCl3, rt) δ 1.19 (6H, d, J ) 6.6
Hz, Me), 3.35 (2H, s, CH2), 4.06 (1H, octet, J ) 7.0 Hz, CH),
6.10 (1H, br s, NH); 13C NMR (CDCl3, rt) δ 22.28 (q, J ) 126.8
Hz, Me), 25.91 (t, J ) 136.9 Hz, CH2), 42.62 (d, J ) 140.4 Hz,
CH), 114.84 (s, CN), 159.98 (s, CdO).
Cyanoacetanilide was prepared similarly from methyl cyano-
acetate and aniline, mp 196-8 °C (lit.26 198 °C; lit.27 203-205
°C): 1H NMR (CDCl3, 298 K) δ 3.50 (2H, s, CH2), 7.20 (1H, t, J
) 7.3 Hz, Ph-H), 7.37 (2H, t, J ) 8.0 Hz, Ph-H), 7.51 (2H, d, J
) 8.1 Hz, Ph-H), 7.67 (1H, br s, NH).
N-Diphenylmethylcyanoacetamide,28 mp 123-5 °C, was pre-
pared similarly from methyl cyanoacetate and diphenylmethyl
amine: 1H NMR (CDCl3, 298 K) δ 3.89 (2H, s, CH2), 6.21 (1H,
d, J) 7.9 Hz, CH), 6.73 (1H, d, J ) 7.0 Hz, NH), 7.22-7.38 (10H,
m, 2Ph).
N,N-Dimethylcyanoacetamide. A 40% aqueous solution of
Me2H (40 mL, 0.35 mol) was added to methyl cyanoacetate (19.8
g, 0.2 mol) at -10 °C. The mixture was stirred for 1 h at <0 °C
(23) Frey, J.; Rappoport, Z. J. Am. Chem. Soc. 1996, 118, 3994 and
5169.
(24) Shukla, J. S.; Bhatia, P. J. Indian Chem. Soc. 1978, 55, 281.
(25) Cossey, A. L.; Harris, R. L. N.; Huppatz, J. L.; Phillips, J. N. Aust.
J. Chem. 1976, 29, 1039.
(26) Gorobets, N. Y.; Yousefi, B. H.; Behrooz, H.; Belaj, F.; Kappe, C.
O. Tetrahedron 2004, 60, 8633.
(27) Diago-Meseguer, J.; Palomo-Coll, A. L.; Fernandez-Lizarbe, J. R.;
Zugaza-Bilbao, A. Synthesis 1980, 547.
(28) Lafton, L. BE Patent, 866163 19781020; Chem. Abstr. 1979, 90,
103986p.
(29) Ben Cheikh, A.; Chuche, J.; Manisse, N.; Pommelet, J. C.; Netsch,
K. P.; Lorencak, P.; Wentrup, C. J. Org. Chem. 1991, 56, 970.
(30) Tkachenko, V. V.; Tregub, N. G.; Knyazev, A. P.; Mezheritskii,
V. V. RostoV. Gos. Zh. Org. Khim. 1990, 26, 638; Chem. Abstr. 1990, 113,
152224.
J. Org. Chem, Vol. 72, No. 14, 2007 5311

Basheer et al.
reaction is demonstrated for isopropyl isocyanate. To Na (0.25 g,
11 mmol) in dry THF (60 mL) was added N-methylcyanoacetamide
(0.98 g, 10 mmol) with stirring, and the mixture was refluxed for
24 h, until all the Na was consumed. An orange salt was obtained.
A solution of isopropyl isocyanate (1 mL, 10 mmol) in dry THF
(20 mL) was added dropwise to the mixture. The solid was mostly
dissolved, and the solution was stirred under reflux for 8 h. Removal
of the solvent left a brown solid which was dissolved in DMF (10
mL), and the orange solution was added dropwise to a cold 2 N
HCl solution (100 mL). The liquid mixture was extracted with ethyl
acetate (3 × 100 mL) and washed with water (3 × 100 mL), the
organic phase was dried (Na2SO4), and the solvent was evaporated,
giving 1.18 g (6.45 mmol, 65%) of the solid yellow product (4j/
5j/6j), mp 133-134 °C. Anal. Calcd for C8H13N3O2: C, 52.46; H,
7.10; N, 22.95; Found: C, 52.74; H, 7.02; N, 23.13.
The t-Bu analogue (4k/5k/6k), mp 138 °C, was obtained similarly
in 67% yield and crystallized from EtOAc. In the analogous reaction
with the aryl isocyanates, the solid was obtained in the acidification
stage without extraction by EtOAc. It was washed with water (300
mL).
The product from phenyl isocyanate (4g/5g/6g), mp 140 °C, was
obtained in 59% yield. Crystals for X-ray diffraction were obtained
after standing for a long time in a 1:20 DMF/H2O mixture. The
4-methoxyphenyl derivative (4h/5h/6h), mp 148 °C, was obtained
in 64% yield, and crystals for X-ray diffraction were obtained from
CDCl3. The pentafluorophenyl derivative (4i/5i/6i), mp 170 °C, was
obtained in 51% yield. Crystals for X-ray diffraction were obtained
by slow evaporation of the EtOAc solvent at rt. The spectral and
analytical data for all derivatives are given in Tables 1, S2, S3,
and S6.
Reaction of Cyanoacetamide with Phenyl, p-Anisyl, Pen-
tafluorophenyl, Isopropyl, and tert-Butyl Isocyanates. Except for
small differences, the procedure is similar to that described above
for the reaction of N-methylcyanoacetamide. Longer reaction times
of 48-72 h were required for obtaining the sodium salt. The color
of the mixtures is mustard yellow. The mixtures were refluxed
overnight.
The N-phenyl derivative (4l/5l/6l), mp 176-177 °C, the N-p-
anisyl derivative (4m/5m/6m), mp 172 °C, and the N-pentafluo-
rophenyl derivative (4n/5n/6n), mp 168 °C, were obtained in 28,
62, and 66% yield, respectively, and crystallized from EtOAc. The
N-i-Pr derivative (4o/5o/6o), mp 166 °C, was obtained in 39% and
crystallized from EtOAc/petroleum ether. The N-t-Bu derivative
(4p/5p/6p), mp 162 °C, was obtained in 26% yield. Spectral and
analytical data for all compounds are given in Tables 1, S2, S3,
and S6.
Reaction of N-Isopropylcyanoacetamide with Isopropyl Iso-
cyanate. To a suspension of Na (250 mg, 11 mmol) in dry THF
(50 mL) was added N-isopropylcyanoacetamide (1.26 g, 10 mmol),
and the mixture was stirred overnight at rt until the Na disappeared.
To the yellow solid formed was added a solution of isopropyl
isocyanate (1 mL, 10 mmol) in dry THF (20 mL) during 20 min,
after which the solid was dissolved giving an orange solution. It
was refluxed overnight, and the solvent was evaporated, giving the
sodium enolate salt: 1H NMR (DMSO-d6, 298 K) δ 1.00 (12H, d,
J ) 6.5 Hz, Me), 3.80 (2H, octet, J ) 6.6 Hz, CH-i-Pr), 4.97 (1H,
br s, NH), 8.77 (1H, br s, NH).
6s) (1.08 g, 5.1 mmol, 51%), mp 178 °C. Crystals for X-ray
diffraction were obtained by slow evaporation of a CDCl3 solution
of the compound. Anal. Calcd for C10H17N3O2: C, 56.87; H, 8.06;
1
N, 19.90. Found: C, 56.42; H, 8.01; N, 19.40. H NMR (CDCl3,
rt) δ (91% enol) 1.19 (12H, d, J ) 6.5 Hz, Me), 4.08 (2H, octet,
J ) 6.7 Hz, CH-i-Pr), 5.29 (2H, d, J ) 5.0 Hz, NH), 17.77 (1H, s,
OH); δ (9% amide) 1.19 (overlap), 4.08 (overlap), 4.31 (s, CH),
6.77 (s, NH).
A similar procedure was applied for preparing 4r/5r/6r, from
N′-phenylcyanoacetanilide and phenyl isocyanate, 4t/5t/6t from
N-diphenylmethyl cyanoacetamide and diphenylmethyl isocyanate,
and 4q/5q/6q from N-phenylcyanoacetanilide and pentafluorophenyl
isocyanate. The spectral and analytical data for these compounds
are given in Tables 1, S2, S3, and S6.
An analogous reaction of N-tert-butylcyanoacetamide gave the
solid Na salt in the first step which did not react with tert-butyl
isocyanate.
Computation Methods. The compounds investigated were
optimized by using the Gaussian 98 program31 at the hybrid density
functional B3LYP/6-31G** method,32 which was extensively used
in recent calculations of enols of carboxylic acid derivatives and
shown to give pKEnol values reliable within 2 pKEnol units.3,33 All
optimized structures including higher energy conformers were
verified by means of their Hessian matrices to be local minima on
the potential energy surfaces. KEnol values were calculated from
the free energy differences between the amide and enol forms at
298.15 K.
Acknowledgment. We are indebted to the Israel Science
Foundation for support, and to Dr. Shmuel Cohen from the
Department of Inorganic and Analytical Chemistry, The Hebrew
University, for the X-ray structure determination.
Supporting Information Available: Table S1 with relative
1
integration of signals, Tables S2 and S3 with H and 13C NMR
data for all compounds, Table S4 with calculated energies of the
regioisomeric enols, Table S5 with IR spectral data, and Table S6
with microanalyses and mp of all new compounds. The full
crystallographic data for the compounds in Table 6 are available
as cifs. The material is available free of charge via the Internet at
http://pubs.acs.org.
JO070877H
(31) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A, Jr.;
Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A.
D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi,
M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.;
Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick,
D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.;
Ortiz, J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi,
I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson,
B.; Chen, W.; Wong, M. W.; Andres, J. L.; Gonzalez, C.; Head-Gordon,
M.; Replogle, E. S.; Pople, J. A. Gaussian 98, revision A.6; Gaussian,
Inc.: Pittsburgh, PA, 1998.
(32) (a) Becke, A. D. Phys. ReV. 1988, A38, 3098. (b) Lee, C.; Yang,
W.; Parr, R. G. Phys. ReV. 1988, B37, 785.
(33) (a) Apeloig, Y.; Arad, D.; Rappoport, Z. J. Am. Chem. Soc. 1990,
112, 9131. (b) Sklenak, S.; Apeloig, Y.; Rappoport, Z. J. Chem. Soc., Perkin
Trans. 2 2000, 2269. (c) Yamataka, H.; Rappoport, Z. J. Am. Chem. Soc.
2000, 122, 9818. (d) Rappoport, Z.; Lei, Y. X.; Yamataka, H. HelV. Chem.
Acta 2001, 84, 1401.
The salt was dissolved in DMF (10 mL), and the solution was
added dropwise to a cold 2 N HCl solution (100 mL). The white
oily solid obtained was solidified after stirring for 5 min, filtered,
and washed with cold water (200 mL), giving the product (4s/5s/
5312 J. Org. Chem., Vol. 72, No. 14, 2007