Dialkoxyphosphinyl-substituted enols of carboxamides

Article abstract of DOI:10.1021/jo0710679

(Chemical Equation Presented) Reactions of isocyanates XNCO (e.g., X = p-An, Ph, i-Pr) with (MeO)₂P(=O)CH₂CO₂R [R = Me, CF₃CH₂, (CF₃)₂CH] gave 15 formal "amides" (MeO)₂

Full text of DOI:10.1021/jo0710679

Dialkoxyphosphinyl-Substituted Enols of Carboxamides
Jinhua Song,† Hiroshi Yamataka,‡ and Zvi Rappoport*,†
Department of Organic Chemistry and the Lise Meitner MinerVa Center for Computational Quantum
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 May 23, 2007
Reactions of isocyanates XNCO (e.g., X ) p-An, Ph, i-Pr) with (MeO)2P(dO)CH2CO2R [R ) Me,
CF3CH2, (CF3)2CH] gave 15 formal “amides” (MeO)2P(dO)CH(CO2R)CONHX (6/7), and with (CF3-
CH2O)2P(dO)CH2CO2R [R ) Me, CF3CH2] they gave eight analogous amide/enols 17/18. X-ray
crystallography of two 6/7, R ) (CF3)2CH systems revealed Z-enols of amides structures (MeO)2P(d
O)C(CO2CH(CF3)2)dC(OH)NHX 7 where the OH is cis and hydrogen bonded to the OdP(OMe)2 group.
The solid phosphonates with R ) Me, CF3CH2 have the amide 6 structure. The structures in solution
were investigated by 1H, 13C, 19F, and 31P NMR spectra. They depend strongly on the substituent R and
the solvent and slightly on the N-substituent X. All systems displayed signals for the amide and the E-
and Z-isomers. The low-field two δ(OH) and two δ(NH) values served as a probe for the stereochemistry
of the enols. The lower field δ(OH) is not always that for the more abundant enol. The % enol, presented
as Kenol, was determined by 1H, 19F, and 31P NMR spectra, increases according to the order for R, Me <
CF3CH2 < (CF3)2CH, and decreases according to the order of solvents, CCl4 > CDCl3 ∼ THF-d8 >
CD3CN >DMSO-d6. In DMSO-d6, the product is mostly only the amide, but a few enols with fluorinated
ester groups were observed. The Z-isomers are more stable for all the enols 7 with E/Z ratios of 0.31-
0.75, 0.15-0.33, and 0.047-0.16 when R ) Me, CF3CH2, and (CF3)2CH, respectively, and for compounds
18, R ) Me, whereas the E-isomers are more stable than the Z-isomers. Comparison with systems where
the OdP(OMe)2 is replaced by a CO2R shows mostly higher Kenol values for the OdP(OMe)2-substituted
systems. A linear correlation exists between δ(OH)[Z-enols] activated by two ester groups and δ(OH)-
[E-enols] activated by phosphonate and ester groups. Compounds (MeO)2P(dO)CH(CN)CONHX show
e7.3% enol in CDCl3 solution. For [(MeO)2P(dO)]2CHCONHX, activated by two OdP(OMe)2 groups,
only the amides were observed in solution and in the solid. DFT calculations reproduce the general
effect of R on Kenol, but the correlation between observed and calculated Kenol values is not linear. The
roles of electron withdrawal by the activating phosphonate and ester groups, and the importance of N-H
and O-H hydrogen bonding to them in stabilizing the enols are discussed.
Introduction
and hence, the enol and related enols are unobservable by
spectroscopic techniques such as NMR at room temperature.
Increase of the stability by substitution with two â-bulky
aromatic substituents increases Kenol appreciably, but even then
the enol of amide is observable by NMR or UV spectroscopy
Simple enols of carboxamides are very unstable compared
with their isomeric amides. For example, the calculated equi-
librium constant Kenol ) [enol]/[amide] between the parent
acetamide CH3CONH2 and its enol CH2dC(OH)NH2 is 10-21.6 1
,
† The Hebrew University.
‡ Rikkyo University.
(1) Sklenak, S.; Apeloig, Y.; Rappoport, Z. J. Am. Chem. Soc. 1998,
120, 10359.
10.1021/jo0710679 CCC: $37.00 © 2007 American Chemical Society
Published on Web 08/31/2007
J. Org. Chem. 2007, 72, 7605-7624
7605

Song et al.
only rarely, and if at all, as a short-life or medium-life
intermediate of some kinetic but not thermodynamic stability.2,3
In recent years, we used a different method and increased the
stability of the enols by substituting Câ with two electron-
withdrawing groups (EWGs) by resonance Y and Y′. The
contributing dipolar push-pull structure 1b to the enol hybrid
1a/1b is much more stabilizing than the contributing structure
2b to the amide hybrid 2a/2b, and the right-hand side of eq 1
becomes frequently more stabilized than the left-hand side,
giving in favorable cases observable enols of amides with easily
measured Kenol values and isolable enols in several systems.
bonding ability of phosphonate vs ester group, and we want to
compare the dialkoxyphosphinyl-substituted systems with di-
esters like E-3 and Z-35,7a,b or with the cyanoesters Z-4.7a,b,8
These systems showed a systematic behavior related to Kenol
values, δ(OH) and δ(NH) values, the E/Z ratios, and the solid-
state structures, with the increase in the number of fluorine atoms
in the ester group(s). We want to find out if these features are
also reflected in dialkoxyphosphinyl ester activated systems.
Results and Discussion
Synthesis. Four groups (a-d) of “formal”15 amido phospho-
nate systems had been prepared by the general method used
previously for obtaining amides activated by Y, Y′.4,5,7-10
(a) Reactions of 2,2,2-trifluoroethyl or 1,1,1,3,3,3-hexafluoro-
2-propyl chloroacetate (obtained from chloroacetyl chloride with
2,2,2-trifluoroethanol or with 1,1,1,3,3,3-hexafluoro-2-propanol,
respectively) with trimethyl phosphite gave the phosphonate
esters 5b and 5c. The methyl ester 5a is commercial. The three
esters were converted to their conjugated bases with Na in THF
or in ether, and each of the anions was reacted with three
aromatic and two aliphatic isocyanates XNCO (X ) p-An, Ph,
C6F5, i-Pr, t-Bu) to give 15 of the formal amido phosphonate
esters (eq 2). The actual structures are either the amides 6a-o
or the E,Z-enols 7a-o or their mixtures which were determined
Enols of amides were isolated and their solid-state structures
determined when YY′ ) Meldrum’s acid4,5 or indandione6
moieties or when Y,Y′ ) CO2R,CO2R′′;4,5,7c CN,CN;8 CN,-
CO2R;7,8 CN,CONR1R,9 CN,CSNR1R2,10 NO2,CONR1R2,11 or
observed together with the amides in solution in these cases
and in low concentrations in other cases such as when Y,Y′ )
1
by H, 13C, 19F, and 31P NMR spectra.
CO2R,NO2 or SO2R,CO2R.12 When Y * Y′, E- and Z-enols
8
are frequently observed and their relative stabilities are deter-
mined to a significant extent by intramolecular hydrogen
bonding to the enol hydrogen. Kenol values were measured for
many enols as a function of Y,Y′, the amido N-substituents and
the solvent, and NMR spectra serve as important probe in the
structure assignment.
The present work investigates the possibility of generating
enols when Y,Y′ are two dialkoxyphosphinyl (RO)2PdO groups
or a combination of (RO)2PdO and CO2R or CN. There are
literature cases of enol formation activated by the presence of
dialkoxyphosphinyl groups, but these are enols of aldehydes
and ketones and not of carboxylic acids.13,14 We are interested
in the activating effect for enol formation and in hydrogen
(2) Frey, J.; Rappoport, Z. J. Am. Chem. Soc. 1996, 118, 3994.
(3) Rappoport, Z.; Frey, J.; Sigalov, M.; Rochlin, E. Pure Appl. Chem.
1997, 69, 1933.
(4) Mukhopadhyaya, J. K.; Sklenak, S.; Rappoport, Z. J. Am. Chem. Soc.
2000, 122, 1325.
(5) Lei, Y. X.; Cerioni, G.; Rappoport, Z. J. Org. Chem. 2001, 66, 8379.
(6) Song, J.; Rappoport, Z. The 10th European Symposium on Organic
Reactivity (ESOR 10), Rome, Italy, July 25-30, 2005, Abstract book, p
46.
(7) (a) Lei, Y. X.; Casarini, D.; Cerioni, G.; Rappoport, Z. J. Org. Chem.
2003, 68, 947. (b) Lei, Y. X.; Casarini, D.; Cerioni, G.; Rappoport, Z. J.
Phys. Org. Chem. 2003, 16, 525. (c) Basheer, A.; Rappoport, Z. J. Org.
Chem. 2004, 69, 1151.
(b) By reacting chloroacetonitrile with trimethyl phosphite
the dimethoxyphosphinylacetonitrile 8 was obtained, and reac-
tion with Na, followed by reaction with XNCO (X ) p-An,
Ph, C6F5, t-Bu) gave the corresponding formal cyanoami-
dophosphonates which can be the amides 9a-d or the enols
10a-d (eq 3).
(8) Mukhopadhyaya, J. K.; Sklenak, S.; Rappoport, Z. J. Org. Chem.
2000, 65, 6856.
(9) Basheer, A.; Yamataka, H.; Ammal, S. C.; Rappoport, Z. J. Org.
Chem. 2007, 72, 5297.
(13) (a) Breuer, E. In The Chemistry of Organophosphorus Compounds;
Hartley, F. R., Ed.; Wiley: Chichester, 1996; Vol. 4, Chapter 7, pp 694-
698. (b) Afarinkia, K.; Echenique, J.; Nyburg, S. C. Tetrahedron Lett. 1997,
38, 1663.
(10) Basheer, A.; Rappoport, Z. Submitted for publication.
(11) (a) Simonsen, O.; Thorup, N. Acta Crystallogr. Sect. B 1979, 57,
1177. (b) Basheer, A.; Rappoport, Z. Unpublished results.
(12) Basheer, A.; Rappoport, Z. The 17th IUPAC Conference on Physical
Organic Chemistry, Shanghai, China, August 15-20, 2004, Abstract IL-
10, p 84.
(14) There are slightly related phosphato-substituted enols in biological
systems such as phosphoenol pyruvate (Richard, J. P. In The Chemistry of
Enols; Rappoport, Z., Ed.; Wiley: Chichester, 1990; Chapter 11).
(15) We use the term “formal” since in solution the compounds exist an
amide/enols (E and Z) mixtures and in the solid state they may have an
amide or an enol structure.
7606 J. Org. Chem., Vol. 72, No. 20, 2007

Dialkoxyphosphinyl-Substituted Enols of Carboxamides
(c) Similarly to the reaction of group (a) above, reaction of
the methylenebisphosphonate 11 with the five isocyanates gave
five formal bisphosphonate amides which could be either the
amides 12a-e or the enols 13a-e (eq 4).
FIGURE 1. ORTEP drawing of the amide 6c.
Solid-State Structures. We were able to obtain suitable
crystals for X-ray diffraction of six of the amidophosphonates:
four belonging to series (a), one to series (c), and one to series
(d). We failed in obtaining suitable crystals in the cyano series
(b).
In series (a), we determined the solid-state structures of 6c/
7c with no fluorine atoms in the ester group, of 6g/7g having
three fluorine atoms in the ester group, and of 6k/7k and 6n/
7n, both having six fluorine atoms in the ester group. The
ORTEP drawing of 6c and 7k are given in Figures 1 and 2,
respectively, and selected crystallographic parameters of the four
compounds are given in Table 1. Other ORTEPs and the full
crystallographic parameters are given in the Supporting Infor-
mation.
The structures of the first two compounds with none or three
fluorines in the ester groups are those of the amides 6c and 6g.
Their bond lengths and angles, except for one bond length and
one angle, are identical within the experimental error, with
normal C-C and C-N bond lengths, while the short bonds of
1.18 and 1.21 Å are characteristic of a CdO of a C(CdO)X
moiety.17
(d) When commercial methyl bis(trifluoroethoxy)phosphi-
nylacetate 15 was refluxed with excess trifluoroethanol and
catalytic amount of H2SO4, the methyl ester group was
exchanged by a trifluoroethyl group to form the trifluoroethyl
bis(trifluoroethoxy)phosphinylacetate 16. Reaction of 15 or 16
with Na in ether, followed by reaction with the p-An, Ph, i-Pr,
and t-Bu isocyanates, gave the eight formal bis(trifluoroethoxy)-
phosphinyl amidoesters 17a-h or their enols 18a-h (eq 5).
In contrast, the hexafluoroester derivatives display the enol
structures 7k and 7n. The data for both compounds are in Table
1. Characteristic important features for Y,Y′-activated enols are
the elongated C(1)-C(2) bond, compared with a normal CdC
value, of 1.337 to 1.41-1.42 Å (indicating a dipolar structure)
which is only 0.01 Å shorter than the C(2)-C(3) bond and the
conversion of the amido CdO to a single C-O bond of 1.313-
1.321 Å, whereas the ester CdO bond length remains 1.21 (
0.004 Å. The angles around the C(1)-C(2) bond are mostly
around 120°. For compound 7k there are three independent
molecules in the unit cell. In two of them the N-p-An group
and the CdC bond are nearly coplanar, whereas they are nearly
perpendicular in the third molecule. Almost all parameters
involving heavy atoms are nearly the same.
The enol can have an E or Z-configuration. The enolic OH
was found to be cis to the OdP(OMe)2 group, rather than to
the CO2CH(CF3)2 in both enols. The nonbonded O(1)‚‚‚O(3)
distances of 2.488-2.513 Å indicate a moderate to strong O-
H‚‚‚OdP hydrogen bond. This resembles the configuration in
enols 3, R2 ) (CF3)2CH, where the hydrogen bonding is not to
the CO2CH(CF3)2 group.5
All 31 amidophosphonates except 9b16 are new compounds.
The OdP(OMe)2-substituted derivatives are not very stable.
The fluoro derivatives frequently decompose during crystal-
lization from 1:5-10 EtOAc-petroleum ether mixtures even
at ca. -20 °C. X-ray crystallography showed that the decom-
position product of 6f/7f is 19 and that of 6m/7m is 20; i.e.,
the fluorinated ester moiety was lost during the crystallization.
The crystallographic data of 19 and 20 (CIF) are provided in
the Supporting Information.
The hydrogen bond is asymmetric with O(1)-H and
O(3)‚‚‚H bond lengths or distances of 1.02(6) and 1.54(6)Å in
7n and 1.00(4) and 1.52 Å for one molecule of 7k, i.e., ∆OH
)
0.52 Å. However, ∆OH values are 0.68 and 0.74 Å in the other
two independent molecules of 7k. This may reflect the difficulty
in locating hydrogens in the X-ray diffraction. The O(1)HO(3)
angles are 158(4)-162(4)°; i.e., the hydrogen bond is not far
from being linear.
(16) Hamlet, Z.; Mychajlowskij, W. Chem. Ind. (London, UK) 1974, (20),
829.
(17) Allen, F. H.; Kennard, O.; Watson, D. G.; Brammer, L.; Orpen, A.
G.; Taylor, R. J. Chem. Soc., Perkin Trans. 2 1987, S1.
J. Org. Chem, Vol. 72, No. 20, 2007 7607

Song et al.
1
The H NMR δ(OH) and δ(NH) and δ(CH) values for the
enols and the amides are given in Table 2. It is noteworthy that
for the amide the hydrogens of the OCH3 of 6 and the CH2
hydrogens of the OCH2CF3 on P of 17 appear as two doublets,
3
each one with JPH ) ca. 11.5 Hz. This is ascribed to two
diastereotopic signals due to the chiral center. In ca. 15% of
the cases only one doublet was observed and this is attributed
to overlap of the two doublets. Two doublets appear also for
the carbons bearing these hydrogens (see below).
The full data are given in Table S1 in the Supporting
Information.
Selected 13C, 19F, and 31P NMR data are given in Tables 3
and 4. The full data are given in Tables S2 and S3 in the
Supporting Information. In all cases, small unidentified signals
or signals due to decomposition are omitted.
1
1H NMR. Table 2 gives the lowest field H NMR range
which includes the OH and NH signals of the enols and amides
and the higher field CH signals of the amides. For series (a)
and (d) there are two sets of signals with δ >8.40 ppm in CCl4,
CDCl3, THF-d8, and CD3CN (except for 7c, 7g, and 7h for
which few OH signals in some solvents were not observed,
perhaps due to fast intermolecular exchange), but none in
DMSO-d6. The lowest field set is at δ 12.92-16.84, and the
higher field set is at δ 8.40-12.00, and the values change
systematically depending on the nitrogen-X and the ester-R
substituents. Each signal in each group has a signal with identical
integration in the other group and by comparison with the
integration of the methyls of the p-anisyl, i-Pr, or t-Bu groups,
these signals are identified as one hydrogen signals. In parallel
with previously studied enol systems,4-10 they were identified
as the enol of amide OH (at the lowest field) and NH (at the
higher field) signals. Moreover, in parallel with systems E-3/
Z-35 the two signals in each groups belong to the E and Z
isomers. As with enols E-3/Z-3, the OH signal at the lowest
field for 7a-e belong to the same enol with the highest field
NH signal, whereas the “internal” signals between them belong
to the other isomer. The other two signals in Table 2 at a higher
field have similar integration in each species and are the NH
(at δ 6.42-10.54) and the CH (at δ 3.47-5.09) of the amide.
The latter is easily identified since the proton is split to a doublet
by the phosphorus.
FIGURE 2. ORTEP drawing of the enol 7k.
Similarly to the diester activated systems, there is an
additional N-H‚‚‚OdC hydrogen bond to the ester group, with
an N‚‚‚O nonbonding distances of 2.695(4)-2.726(3) Å, N-H
and O‚‚‚H distances of 0.72(3)-0.82(4) and 2.00(4)-2.11(3)
Å, and an N(1)HO(2) angles of 131(3)-142(4)°.
The structural change from an ester carrying 0 to 3 and then
6 fluorine atoms resembles the finding for the solid diester
systems, where the structure is that of an amide with 0 and 3
fluorines, but an enol for the 6-F compounds. In both cases the
activation for enol formation increases, but the hydrogen bond
accepting ability decreases with the increased electron-
withdrawal by the ester group.
Inspection of Table 2 leads to several generalizations. (a) The
lowest field OH and NH signals are consistently in THF-d8.
The order of the four signals as a function of the solvent at 298
K is THF-d8 > CD3CN > CDCl3 > CCl4, and the differences
between the extreme δ values, e.g., for δ(OH) of the E-enol,
are 0.42-0.65 ppm for the 6/7 series. Data for the other signals
and for the 17/18 series are given in Table 5a. (b) The N-aryl
substituent effect in the 6/7 series is not systematic. An increase
in the electron-donating ability of the aryl ring results in the
order p-An < Ph > C6F5 for δ(OH,E-enol), whereas for δ-
(OH,Z-enol) the order is p-An < Ph < C6F5. In the 17/18
systems the order is p-An < Ph. For N-aliphatic substituents
δ(OH) and δ(NH) values follow the order i-Pr < t-Bu. These
δ values for the N-alkyl substituents are always lower than for
the aromatic ones both in the 6/7 and 17/18 systems, with more
significant differences for the δ(NH) in the 6/7 series.
Structure in Solution. The probes for the structure(s) and
compositions of the mixtures in solution were H, 13C, 19F,
1
and 31P NMR spectra which were recorded for all systems in
five solvents of different polarity: CCl4, CDCl3, THF-d8,
CD3CN, and DMSO-d6, mostly at 298 K and occasionally at
lower temperatures. All spectra were used for qualitative
(c) The mechanistically most revealing data are the differences
due to the change of the ester group R within the series 6/7 and
17/18. All of the following quoted differences are for the same
solvent and N-substituent. The differences between members
of the 7a-e (R ) CO2Me) and the 7f-j (R ) CF3CH2) series
1
structural analysis, whereas H, 19F, and 31P spectra were also
used for quantitative determination of the compositions of the
various species in solution which were structure and solvent
dependent.
7608 J. Org. Chem., Vol. 72, No. 20, 2007

Dialkoxyphosphinyl-Substituted Enols of Carboxamides
TABLE 1. Selected X-ray Data for 7n, 7k, 6g, and 6c at Room Temperature
length, Å
bond
7n
7k
6g
6c
C(1)-C(2)
C(1)-N(1)
C(1)-O(1)
C(2)-C(3)
C(2)-P(1)
C(3)-O(2)
C(3)-O(6)
O(3)-P(1)
O(1)-H
1.418(5)
1.306(5)
1.318(5)
1.426(5)
1.754(3)
1.213(4)
1.375(4)
1.484(3)
1.02(6)
1.54(6)
2.513(4)
0.82(4)
2.00(4)
2.695(4)
1.414(4) [1.420(4)] (1.416(4))
1.306(4) [1.311(4)] (1.314(4))
1.321(4) [1.316(3)] (1.313(3))
1.427(4) [1.427(4)] (1.423(4))
1.751(3) [1.752(3)] (1.755(3))
1.206(4) [1.208(3)] (1.206(3))
1.393(3) [1.387(3)] (1.384(3))
1.487(2) [1.484(2)] (1.485(2))
1.00(4) [0.93(4)] (0.90(4))
1.52(4) [1.61(4)] (1.64(5))
2.488(3) [2.502(3)] (2.500(3))
0.85(3) [0.76(3)] (0.72(3))
2.08(3) [2.11(3)] (2.10(3))
2.716(3) [2.726(3)] (2.709(3))
1.537(5)
1.336(5)
1.217(5)
1.519(5)
1.810(4)
1.183(5)
1.333(5)
1.449(3)
1.524(2)
1.352(2)
1.213(2)
1.524(3)
1.8162(18)
1.180(3)
1.311(3)
1.4654(14)
O(3)-H
O(1)-O(3)
N(1)-H
O(2)-H
N(1)-O(2)
angle (deg)
bond angle
7n
7k
6g
6c
N(1)C(1)O(1)
N(1)C(1)C(2)
O(1)C(1)C(2)
C(1)C(2)C(3)
C(1)C(2)P(1)
C(3)C(2)P(1)
O(2)C(3)C(2)
O(2)C(3)O(6)
O(6)C(3)C(2)
O(1)HO(3)
114.9(4)
123.2(3)
121.9(3)
119.4(3)
119.3(3)
121.3(3)
128.4(3)
120.5(3)
111.2(3)
159(5)
114.7(3) [114.2(3)] (114.5(3))
123.6(3) [124.0(3)] (123.5(3))
121.7(3) [121.7(3)] (122.0(3))
119.1(3) [119.2(3)] (119.4(3))
119.1(2) [119.3(2)] (119.0(2))
121.8(2) [121.6(2)] (121.7(2))
129.7(3) [129.2(3)] (129.1(3))
120.3(3) [120.5(3)] (120.6(3))
110.0(3) [110.3(3)] (110.4(2))
161(4) [162(4)] (158(4))
125.2(4)
114.6(4)
120.2(3)
110.6(3)
109.6(2)
111.0(3)
126.1(4)
124.2(4)
109.7(3)
123.73(17)
113.89(15)
122.38(16)
109.61(15)
110.11(12)
111.50(13)
124.54(19)
124.8(2)
110.63(18)
N(1)HO(2)
142(4)
131(3) [138(3)] (142(3))
are nearly constant. For the δ(OH,E-enol) they are 0.95-1.09
ppm for the aromatic substituents. There are differences in the
values for N-Ar and N-Alk substituents and the average values,
as well as the other signals discussed below are given in Table
5b. The differences between the 7f-j (R ) CF3CH2) series and
the 7k-o (R ) (CF3)2CH) series are also approximately
constant, being 0.76-1.01 ppm. For the δ(OH, Z-enol) the
differences between the 7a-e and 7f-j series are negative,
being -0.16 to -0.27 ppm, and the differences between the
7f-j and 7k-o series show a large variation of -0.07 to -0.28
ppm.
groups (Table 5d). Remarkably, the differences between systems
7 and 18 for the δ(OH,E-enol) and the δ(NH,Z-enol) were very
small, being 0.02-0.23 (average for 27 pairs 0.12 ( 0.05) and
-0.13 to +0.03 (average for 32 pairs -0.03 ( 0.04), respec-
tively. The largest difference is for δ(OH,Z-enol) where the
values for series 7 are 0.84-1.14 ppm higher than for system
18 when R ) Me and 0.92-1.21 ppm when R ) CF3CH2. The
δ(NH,E-enol) is also higher for system 7 than for 18 by 0.18
-0.61 ppm when R ) Me and by 0.38-0.60 ppm when R )
CF3CH2 (Table 5d).
∆δ(OH)(E-enol - Z-enol) values for 18 are 2.33-2.67 when
R ) Me (average 2.53 ( 0.10). When R ) CF3CH2, the range
is 1.21-1.58 (average 1.42 ( 0.08). The ∆δ(NH)(E-enol -
Z-enol) values are negative: -0.63 to -0.94 ppm when R )
Me (average 0.77 ( 0.10 ppm) and -0.14 to -0.46 ppm when
R ) CF3CH2 (average -0.29 ( 0.10 ppm) (Table 5c).
For the δ(NH,E-enol) signals, the differences between the
7a-e and the 7f-j series are also negative , being -0.16 to
-0.27 ppm, and the differences between the 7f-j and 7k-o
series are -0.11 to -0.24 ppm. For the δ(NH,Z-enol), the
differences between the 7a-e and 7f-j series are positive, being
0.20-0.37 ppm, and the differences between the 7f-j and 7k-o
series are 0.15-0.32 ppm.
(d) The ∆δ(OH)(E-enol - Z-enol) values for the 7a-e (R
) Me) pairs also depend on the N-substituents, being 1.51-
1.83 ppm. Values for the N-Ar and N-Alk subgroups for these
and other systems and signals are given in Table 5c. For 7f-j
(R ) CF3CH2), the differences are much smaller: 0.27-0.76
ppm. For 7k-o (R ) (CF3)2CH), the values are negative, being
-0.20 to -1.11 ppm.
(f) For seven and eight compounds, δ(OH) and δ(NH) values
were determined also at 240 K in addition to 298 K. The
temperature effect depends on the isomer and the signal. In series
7, the values in ppm are as follows: E-enol OH +0.03 to -0.16
(average -0.04 ( 0.03), Z-enol OH 0.04-0.19 (average 0.10
( 0.02), E-enol NH 0.06-0.17 (average 0.11 ( 0.03), and
Z-enol NH -0.03 to +0.05 (average -0.01 ( 0.01). For the
cyano enols 10 the E-enol OH signal was not observed at all,
and it was observed only at 240 K for the Z-enol OH. For 10d
it decreased by 0.08 ppm from 298 to 240 K. In contrast, the
δ(NH) signals were observed at both temperatures, and the
temperature effect was significantly larger than in series 7 for
seven of the eight values. δ(NH) for the Z-enol increase at 240
K by 0.51, 0.59, 0.68, and 0.08 ppm and for the amide by 0.76,
0.66, 0.90, and 0.38 ppm. A similar 298K f 240K change in
CD3CN was small: -0.02 ( 0.04, -0.08 ( 0.02, -0.08 (
The ∆δ(NH)(E-enol - Z-enol) differences for 6/7(R ) Me)
are negative , -0.09 s -0.44 ppm. When R ) CF3CH2 the
values are positive, being 0.06-0.54. When R ) (CF3)2CH the
values are more positive, being 0.44-1.02 ppm. Table 5c gives
the smaller ranges for the sub-groups.
(e) Electron withdrawal from the phosphorus for the 18a-h
system carrying (CF3CH2O)2 groups was probed by comparison
with the corresponding 7a,b,d-g,i,j systems carrying (MeO)2
J. Org. Chem, Vol. 72, No. 20, 2007 7609

Song et al.
1
TABLE 2. Relevant H NMR Data for All Enol/Amide Systems in Five Solvents at 298 K
OH, ppm NH, ppm
δ
δ
δ
NH, ppm
amide
δ
CH,a ppm
amide
compd
solvent
E-enol
Z-enol
E-enol
Z-enol
6a/7a
CCl4
CDCl3
THF-d8
CD3CN
DMSO-d6
CCl4
CDCl3
CDCl3 (240 K)
THF-d8
CD3CN
DMSO-d6
CCl4
CDCl3
THF-d8
CD3CN
DMSO-d6
CCl4
CDCl3
THF-d8
CD3CN
DMSO-d6
CCl4
CDCl3
THF-d8
CD3CN
DMSO-d6
CCl4
CDCl3
THF-d8
CD3CN
CD3CN (255 K)
DMSO-d6
CCl4
CDCl3
CDCl3 (240 K)
THF-d8
16.03
16.35
16.63
16.49
14.44 (br)
14.57
15.10
10.93
11.02
11.53
11.26
11.22
11.46
11.62
11.42
8.59
9.05
9.23
8.84
10.09
8.77
9.13
9.40
9.38
8.98
10.23
8.62
8.96
9.30
8.76
10.35
6.61
7.00
7.16
6.88
7.93
6.64
7.07
7.11
6.84
7.67
8.53
8.88
9.18
8.74
9.00
10.15
8.86
9.05
9.33
9.38
8.92
9.25
10.29
8.71
8.89
9.39
9.33
9.71
8.68
10.45
6.64
6.90
7.19
6.91
8.01
6.63
6.92
7.04
7.09
6.76
6.92
7.79
8.41
8.82
9.24
8.71
8.98
10.21
8.79
8.73
9.22
9.39
8.85
10.34
8.61
8.79
9.73
9.37
8.64
10.54
3.64
4.22
4.26
4.28
4.51
3.72
4.21
4.33
4.31
4.31
4.57
3.80
4.30
4.42
4.41
4.71
3.51
4.01
4.03
4.04
4.27
3.47
3.99
3.98
3.99
4.34
3.77
4.31
4.43
4.41
4.47
4.68
3.99
4.36
ov
14.85
6b/7b
16.19
16.49
16.53
16.84
16.63
14.58
14.72
14.65
15.20
15.00
11.12
11.30
11.24
11.73
11.47
11.39
11.63
11.64
11.80
11.62
6c/7c
6d/7d
6e/7e
6f/7f
16.11
16.45
16.70
14.60 (br)
10.70
10.86
11.09
10.82
11.09
11.20
15.52
15.77
16.10
16.02
13.90
14.00
14.59
14.36
8.89
9.12
9.50
9.28
9.20
9.46
9.63
9.53
15.69
15.89
16.27
16.15
13.98
14.06
14.68
14.43
9.07
9.33
9.71
9.51
9.45
9.75
9.90
9.80
15.03
15.38
15.54
15.45
15.47
14.64
14.78 (br)
15.27
15.02
14.94
11.14
11.29
11.69
11.42
11.36
10.86
11.15
11.26
11.10
11.08
6g/7g
6h/7h
15.18
15.52
15.58
15.79
14.81
14.94 (br)
14.90
11.28
11.53
11.43
11.87
11.62
11.57
11.02
11.32
11.35
11.44
11.28
11.29
15.47
4.52
4.46
4.55
4.74
4.03
4.43
ov
CD3CN
CD3CN(240K)
DMSO-d6
CCl4
CDCl3
CDCl3 (240 K)
THF-d8
THF-d8 (240 K)
CD3CN
DMSO-d6
CCl4
CDCl3
THF-d8
CD3CN
DMSO-d6
CCl4
CDCl3
CDCl3 (240 K)
THF-d8
15.66
15.14
15.01
15.39
15.55
10.88
11.06
10.93
11.37
11.25
11.19
10.40
10.75
10.75
10.76
10.81
10.65
14.98
15.60
ov
15.75
15.56
4.79
4.57
4.89
ov
6i/7i
6j/7j
14.60
14.88
15.16
15.07
14.14
14.20
14.76
14.52
14.64
14.23
14.27 (br)
14.18
14.88
14.62
14.55
14.70
14.81
14.95 (br)
15.33 (br)
15.15
9.10
9.30
9.68
9.47
8.90
9.20
9.34
9.28
9.29
9.15
9.48
9.49
9.62
4.11
4.22
4.22
4.44
3.64
4.07
4.16
4.17
4.13
4.18
4.51
b
4.38
4.64
4.55
4.61
4.84
4.07
4.42
4.58
4.69
4.59
4.90
4.13
4.53
4.76
4.86
4.75
5.09
14.78
15.03 (t)
15.00
15.34
15.21
15.24
9.30
9.54
9.47
9.92
9.72
9.63
CD3CN
9.53
9.56
9.54
CD3CN (240 K)
DMSO-d6
CCl4
CDCl3
THF-d8
6k/7k
6l/7l
14.27
14.62
14.71
14.55
14.48
11.25
11.43
11.81
11.60
11.49
10.59
10.91
10.96
10.82
10.75
10.85
10.73
11.07
11.08
11.12
10.97
CD3CN
CD3CN (240 K)
DMSO-d6
CCl4
CDCl3
CDCl3 (240 K)
THF-d8
CD3CN
DMSO-d6
CCl4
CDCl3
CDCl3 (240 K)
THF-d8
15.03
14.40
14.75
14.79
14.87
14.66
15.03
15.12
14.99
15.54
11.43
11.59
11.46
12.00
11.76
15.32 (br)
6m/6m
14.23
14.60
14.61
14.65
14.55
15.30 (br)
15.34 (br)
15.26
15.76
15.60
11.11
11.26
11.14
11.53
11.34
10.11
10.47
10.50
10.51
10.35
CD3CN
DMSO-d6
7610 J. Org. Chem., Vol. 72, No. 20, 2007

Dialkoxyphosphinyl-Substituted Enols of Carboxamides
Table 2 (Continued)
δ
OH, ppm
δ
NH, ppm
δ
NH, ppm
amide
δ
CH,a ppm
amide
compd
solvent
CCl4
E-enol
13.91
14.22
14.20
14.37
14.22
Z-enol
14.36
E-enol
Z-enol
6n/7n
9.29
9.54
9.41
9.91
9.65
8.70
9.05
9.05
9.12
9.04
9.11
8.93
9.28
9.23
9.38
9.28
9.28
9.28
9.34
7.58
8.09
7.92
8.51
7.98
8.66
5.58
5.66
b
6.87
7.03
7.24
6.79
8.11
b
6.87
6.95
7.13
7.47
6.76
6.82
7.90
8.83
9.59
9.12
9.78
9.01
9.91
6.53
6.91
8.78
9.04
8.84
6.67
6.66
8.52
8.82
9.28
8.66
10.22/10.17
8.74
8.96
b
CDCl3
14.42
14.29
14.95
14.69
14.64
14.44
14.45
14.26
15.03
15.06
14.75
14.70
14.73
4.20
4.31
4.38
4.32
4.61
b
CDCl3 (240 K)
THF-d8
CD3CN
DMSO-d6
CCl4
6o/7o
14.07
14.34
9.50
9.72
9.55
10.10
9.89
9.89
9.83
CDCl3
4.13
4.19
4.37
4.45
4.29
4.36
4.68
4.33
4.57
4.46
4.64
4.45
4.65
3.98
4.17
3.79
3.91
3.98
3.56
3.47
ov
4.39
4.45
4.50
4.82
3.97
4.36
4.65
4.56
4.88
ov
CDCl3 (240 K)
THF-d8
THF-d8 (240 K)
CD3CN
CD3CN (255 K)
DMSO-d6
CDCl3
CDCl3 (240 K)
CDCl3
CDCl3 (240 K)
CDCl3
overlap Z-enol
14.53
14.88
14.35
14.32
9a/10a
9b/10b
9c/10c
9d/10d
13.34
13.45
CDCl3 (240 K)
CDCl3
13.07
12.99
CDCl3 (240 K)
CDCl3
CDCl3
CDCl3
CDCl3
12a/13a
12b/13b
12c/13c
12d/13d
12e/13e
17a/18a
CDCl3
CCl4
15.95
16.21
16.47 (br)
13.30
13.57
14.03 (br)
10.34
10.59
10.96
10.77
11.22
11.44
11.59
11.40
CDCl3
THF-d8
CD3CN
DMSO-d6
CCl4
17b/18b
17c/18c
17d/18d
17e/18e
17f/18f
16.14
16.39
13.47 (br)
13.73 (br)
10.51
10.77
11.15
10.96
11.39
11.62
11.78
11.60
CDCl3
THF-d8
CD3CN
DMSO-d6
CCl4
9.52
8.84
10.38
6.42
6.75 (d)
7.25 (d)
6.72
8.11/8.03 (d)
6.47
6.73
7.09
6.63
7.87
8.48
8.45
9.32
8.61
10.29
8.28
8.62
9.45
8.75
10.42
b
6.48
7.32 (d)
6.71
8.20 (d)
6.71
6.47
7.19
6.62
7.97
15.46 (t)
15.67 (t)
15.95
12.92
13.16
13.62
13.41
8.40
9.28
9.56 (d)
9.68
CDCl3
8.72 (d)
9.01
4.19
4.31
4.26
4.58
3.65
4.10
4.29
4.23
4.64
ov
THF-d8
CD3CN
DMSO-d6
CCl4
15.79
8.90
9.57
15.63
15.75
16.12
15.94 (br)
13.00
13.19
13.71
13.46 (br)
8.59
8.89
9.22
9.13
9.53
9.77
9.94
9.85
CDCl3
THF-d8
CD3CN
DMSO-d6
CCl4
14.99
15.24
15.40
15.28
13.53
13.70
14.19
13.94
10.57
10.75
11.09
10.91
10.92
11.15
11.25
11.08
CDCl3
ov
THF-d8
CD3CN
DMSO-d6
CCl4
4.65
4.60
4.97
ov
15.16
15.39
15.64
15.38
13.70
13.86
14.26 (br)
14.09
10.74
10.93
11.28
11.10
11.08
11.31
11.42
11.26
CDCl3
ov
THF-d8
CD3CN
DMSO-d6
CCl4
4.69
4.64
5.03
b
ov
ov
4.37
4.74
ov
17g/18g
17h/18h
14.55
14.78
15.05
14.92
13.14
13.28
13.76
13.54
8.65 (d)
8.89
9.01 (d)
9.29
9.41
9.31
9.16 (br)
9.28
9.52
9.67
9.56
CDCl3
THF-d8
CD3CN
DMSO-d6
CCl4
9.22
9.09
14.74
14.90
15.21
15.01
13.23
13.32
13.84
13.59
8.85
9.06
9.42
9.27
CDCl3
ov
THF-d8
CD3CN
DMSO-d6
4.46
4.36
4.80
a ov: overlaps another signal. b The signal is too weak to be observed.
0.02, and -0.01 ( 0.03 ppm for three or four cases for the
E-enol OH, Z-enol OH, E-enol NH, and Z-enol NH, respectively.
The magnitude of the temperature effect in THF-d8 was larger,
but only two compounds were studied.
J. Org. Chem, Vol. 72, No. 20, 2007 7611

Song et al.
a
TABLE 3. Selected 13C NMR Data for Several Enol/Amide Systems in CDCl3
compd
solvent
T (K)
species
Câ
CR
ACH2O-POb
CO2R
6a/7a
6b/7b
6d/7d
CDCl3
CDCl3
CDCl3
298
298
298
amide
amide
E-enol
Z-enol
amide
E-enol
Z-enol
amide
E-enol
Z-enol
amide
E-enol
Z-enol
Z-enol
E-enol
Z-enol
amide
Z-enol
amide
amide
amide
amide
amide
Z-enol
E-enol
amide
Z-enol
E-enol
amide
Z-enol
E-enol
E-enol
Z-enol
amide
E-enol
Z-enol
amide
E-enol
Z-enol
amide
amide
E-enol
Z-enol
E-enol
Z-enol
Z-Enol
Z-enol
amide
Z-enol
amide
E-enol
Z-enol
amide
E-enol
Z-enol
amide
E-enol
Z-enol
amide
amide
amide
54.2
54.41
60.4
58.3
158.7
158.9
172.0
170.9
160.0
171.9
170.8
158.2
170.93
171.98
159.24
171.2
170.7
170.93
173.2
172.1
158.9
170.17
156.4
156.9
158.5
160.2
159.4
169.79
171.59
157.78
169.86
171.14
158.20
170.16
171.47
172.8
171.77
158.1
173.0
171.9
159.1
173.4
172.3
159.4
160.6
172.3
171.7
171.9
171.1
171.8
171.3
158.9
171.8
159.9
172.74
170.92
157.71
172.89
171.10
158.63
173.27
171.55
159.83
159.99
160.61
54.4
54.42
52.3
52.8
54.20, 54.13
52.69
53.16
54.61, 54.54
52.32
52.85
166.2
166.2
174.7
170.0
166.1
172.4
167.5
164.0
172.09
167.31
163.92
170.1
165.4
165.46
169.9
165.2
162.9
164.91
6g/7g
6i/7i
CDCl3
CDCl3
CDCl3
298
298
298
61.81
60.43
53.96
60.85
58.46
58.24
62.4
54.29, 54.22
52.70
53.15
6k/7k
59.8
60.24
6l/7l
6m/7m
CDCl3
CDCl3
298
240
53.20
61.5
53.7
6n/7n
CDCl3
CDCl3
CDCl3
CDCl3
CDCl3
CDCl3
CDCl3
240
298
298
298
298
298
298
57.57
39.2
39.2
48.3
47.4
53.16
9b/10b
9d/10d
12a/13a
12c/13c
12e/13e
17a/18a
55.4, 55.3
55.4, 55.0
54.3
54.4
54.0
63.20
62.41
63.47, 63.30
62.86
62.52
48.3
59.57
61.16
54.44
59.89
61.81
54.18
58.43
60.55
61.4
58.66
53.51
61.3
58.9
54.2
61.6
58.7
53.65
53.2
61.6
58.6
60.7
58.0
57.9
169.14
174.39
165.35
166.80
171.82
162.44
166.69
171.69
171.83
166.8
166.8
172.3
167.6
164.1
172.6
167.2
163.6
163.8
169.6
164.8
169.6
165.1
165.0
164.6
161.8
163.8
17e/18e
CDCl3
298
17g/18g
6j/7j
CDCl3
CCl4
298
298
62.41
61.99
51.5
52.0
53.43
52.45
52.96
54.42, 54.29
51.21
51.92
53.18, 53.06
54.1
51.5
52.1
52.7
53.1
52.1
52.1
53.5, 53.4
52.9
54.2, 54.1
61.60
62.05
62.95, 62.51
62.29
62.73
63.40, 63.07
61.68
CDCl3
298
298
THF-d8
DMSO-d6
CCl4
298
298
6o/7o
CDCl3
240
THF-d8
CD3CN
240
255
57.1
52.2
58.6
52.8
DMSO-d6
298
298
162.3
17d/18d
CCl4
59.89
57.88
54.09
60.15
58.31
54.83
60.58
58.37
54.41
54.52
54.20
173.84
168.24
165.04
174.42
169.34
165.46
174.66
168.85
164.62
164.86
164.91
CDCl3
298
298
THF-d8
62.21
63.07, 62.84
62.96, 62.72
62.93, 62.67
CD3CN
DMSO-d6
298
298
a Some signals are missing due to too weak intensity or to a difficulty in identification. b A ) H or CF3.
(g) The range of the δ(NH) signal for the amides is mostly
DMSO-d6 . THF-d8 > CDCl3 > CCl4. In ppm for N-Ar
compounds it is at 8.28-9.52 (in DMSO-d6 10.09-10.54) and
for N-alkyl compounds 6.42-7.47 (in DMSO-d6 7.67 - 8.20).
How can the configuration of each group of species be
deduced from these data? This was easy for series 3 and 4 where
the orders of increased electron withdrawal of the ester R groups
and the decreasing hydrogen bond accepting abilities
were (CF3)2CH > CF3CH2 > CH3.5,7 The values of δ(OH) and
δ(NH) followed these values. In the present systems where
OdP(OMe)2 and OdP(OCH2CF3)2 groups are present the
changes in δ values are not monotonous, and hence, we present
them in Table 2 according to the E and the Z species to which
they belong, and the reasoning for this is given below.
The two geometric isomers to be considered are the E isomers
(E-21-E-25) and the Z isomers (Z-21-Z-25). As in previous
7612 J. Org. Chem., Vol. 72, No. 20, 2007

Dialkoxyphosphinyl-Substituted Enols of Carboxamides
TABLE 4. Selected 19F and 31P NMR Data of All Enol/Amide
Systems at 298K
substituent R. Qualitatively, the data suggest that the relative
order changes when the number of fluorine atoms in the ester
(series a) or in the phosphonate (series d) group changes. The
presence of two hydrogen bonds in each enol complicates the
issue. The easiest way to unequivocally evaluate the δ(OH)
values of both the PdO‚‚‚H-O and the PdO‚‚‚H-N bonds in
our system would be to obtain the values for compounds 13a-e
(series c) where both hydrogen bonds are present simultaneously.
19F NMR in CDCl3
E-enol Z-enol amide
31P NMR in CCl4
compd
E-enol Z-enol amide
6a/7a
6b/7b
6c/7c
6d/7d
6e/7e
6f/7f
6g/7g
6h/7h
6i/7i
6j/7j
6k/7k
6l/7l
6m/7m
6n/7n
6o/7o
26.10 33.36 17.00
25.84 33.29 16.92
24.95 31.96 15.71
26.55 33.87 17.21
26.81 34.15 17.46
24.05 31.47 16.3
23.79 31.42 16.22
22.94 30.29 15.34
24.53 31.87 16.37
24.64 32.12 15.60
22.62 30.09 15.57
22.33 30.07 15.43
21.52 29.17 14.90
22.89 30.38 15.69
23.06 30.66 16.00
13.64
-74.81 -74.78 -74.41
-74.8 -74.77 -74.4
-74.85 -74.81 -74.52
-74.9 -74.87 -74.5
-74.89 -74.86 -74.48
-74.25
-74.27
-74.24
-74.34
-74.32
-73.81 (m)
-73.85
-73.87
-73.89
-73.88
9a/10aa
9b/10ba
9c/10ca
9d/10da
12a/13aa
12b/13ba
12c/13ca
12d/13da
12e/13ea
13.34
12.18
14.19
17.57
17.43
16.4
17.81
18.05
Since the OdP(OMe)2 group is a better hydrogen bond
acceptor than a CO2Me group18 (see below), we assumed that
the doubly OdP(OMe)2-activated systems 13 will show a higher
Kenol than the doubly activated CO2Me system 3, R1 ) R2 )
Me, which show ca. 7% enol in CDCl3 when X ) Ph.4 However,
1
17a/18ab -75.90 -75.88 -76.16
17b/18bb -75.93 -75.91 -76.22
26.05 31.92 17.01
25.78 31.79 16.89
the H NMR spectra of all the five 12a-e/13a-e systems in
CDCl3 at either 298 or 240 K displayed no signal g10 ppm.
The possibility that the OH is lost in the baseline due to a
coalescence process as found for dC(CN)CO2Me activated
systems8 is excluded since the signal should then be observed
at the lower temperature, in contrast to the observation.
Moreover, in the 13C NMR C/H coupled spectra of all five
systems, the Câ carbon at 47.4-48.4 ppm appears as a doublet
of triplets due to coupling with both the geminal hydrogen (1JCH
) 135.1-135.5 Hz) and the geminal phosphorus (1JCP ) 127.5-
129.1 Hz). No vinylic carbon was observed. Likewise, the 31P
NMR spectra displayed only one signal at 16.40-17.81 ppm.
Consequently, all systems display exclusively the amide struc-
tures 12a-e in solution. This is also the solid-state structure.
X-ray diffraction of 12d/13d showed the amide 12d. Its CIF is
in the Supporting Information.
17c/18cb -76.01 -75.98 -76.23/-76.24 26.50 32.41 17.14
17d/18db -75.94 -75.91 -76.16/-76.17 26.71 32.63 17.33
17e/18eb -76.02 -76.00 -76.19
17f/18fb -76.04 -76.01 -76.21
17g/18gb -76.18 -76.13
24.28 30.2
16.06
24.03 30.08 15.93
24.64 30.61 16.20
24.88 30.87 16.50
17h/18hb -76.15 -76.12 -76.29
17e/18ec -74.92 -74.86 -74.75
17f/18fc -74.94 -74.87 -74.48
17g/18gc -75.06 -75.01 -74.61
17h/18hc -75.04 -74.99 -74.59
a In CDCl3. b δ(CF3) of the OdP(CO2CH2CF3)2 group. c δ(CF3) of the
CO2CH2CF3 group.
studies, the intramolecular hydrogen bondings CdO‚‚‚H-O and
PdO‚‚‚H-O and CdO‚‚‚H-N and PdO‚‚‚H-N H-bonds
should be very important in determining the configurations. A
priori, the latter two N-H‚‚‚O H-bonds are weaker than the
O-H‚‚‚O bonds. A previous observation was that the stronger
the bond, the higher is the corresponding δ value (i.e., at a lower
field). Another observation is that a bond to a certain ester group
in 3 has a nearly constant δ(OH) regardless of the second
activating ester group.5,7b
We ascribe the absence of e2% (the limit of detection) of
the enols 13 to the fact that the two geminal OdP(OMe)2 groups
create a too sterically congested situation for generating the enol
in which both groups should preferably be in the same plane.
This is reflected in the calculated structure of enol 13b (Figure
3) which clearly show that the two OdP(OMe)2 groups are
above and below the dC(OH)NHPh plane; i.e., the double bond
is twisted and the conjugation necessary for the enol formation
(eq 1) is reduced.
Based on the observation in systems E- and Z-3 that δ(OH)
for a O-H‚‚‚OdC-OR hydrogen bonding is nearly constant
when R is constant,5 but that it is reduced by ca. 1 ppm when
R changes first for Me f CF3CH2 and then further by ca. 0.7
ppm for the change CF3CH2 f (CF3)2CH, we expect a similar
change in δ(OH) for the 21 f 22 f 23 (or 24 f 25) changes,
provided that the OdP(OR′)2 group has a minor influence on
this hydrogen bond. This was indeed found for the assigned
E-enol, where the change R ) Me f CF3CH2 f (CF3)2CH
[E-21 f E-22 f E-23] result in corresponding average changes
in δ(OH) of 1.02-0.83 ppm for N-Ar and of 0.92-0.76 ppm
for N-Alk substituents (Table 5b). The changes of δ(OH) for
The question for our systems concerns the relative strengths
of PdO‚‚‚H-O vs CdO‚‚‚H-O bonds as a function of the
(18) Modro, A. M.; Modro, T. Can. J. Chem. 1999, 77, 890.
J. Org. Chem, Vol. 72, No. 20, 2007 7613

Song et al.
TABLE 5. δ(OH) and δ(NH) Differences
(a) Maximum Solvent Effect on the Values
δ(OH,THF-d8) - δ(OH,CCl4)
δ(NH,THF-d8) - δ(NH,CCl4)
compd
E
Z
E
Z
7
18
0.42-0.65
0.41-0.52
0.46-0.70
0.49-0.71
0.39-0.64
0.52-0.63
0.37-0.47
0.33-0.41
(b) Differences as a Function of the Ester Group
∆δ(OH,E-enol)
∆δ(OH,Z-enol)
∆δ(NH,E-enol)
∆δ(NH,Z-enol)
N-sub-
stituent
δ(7a-e) -
δ(7f-j)
δ(7f-j) -
δ(7k-o)
δ(7a-e) -
δ(7f-j)
δ(7f-j) -
δ(7k-o)
δ(7a-e) -
δ(7f-j)
δ(7f-j) -
δ(7k-o)
δ(7a-e) -
δ(7f-j)
δ(7f-j) -
δ(7k-o)
Ar
range
0.95-1.09
0.76-1.01
-0.17 to
-0.07 to
-0.16 to
-0.23 to
0.25-0.37
0.24-0.32
-0.27
-0.28
-0.27
-0.11
avg (no.
of pairs)
1.02 ( 0.04
(11)
0.83 ( 0.07
(10)
-0.21 ( 0.02
(7)
-0.16 ( 0.08
(8)
-0.21 ( 0.04
(10)
-0.16 ( 0.03
(12)
0.32 ( 0.04
(12)
0.28 ( 0.03
(12)
Alk
range
avg (no.
of pairs)
0.86-0.95
0.66-0.86
-0.16 to -0.25
-0.13 to -0.22
-0.16 to -0.24
-0.17 to -0.24
0.20-0.30
0.15-0.25
0.92 ( 0.02
0.76 ( 0.07
-0.20 ( 0.03
-0.19 ( 0.03
-0.21 ( 0.02
-0.20 ( 0.02
0.27 ( 0.02
0.22 ( 0.02
(8)
(8)
(8)
(8)
(8)
(8)
(8)
(8)
(c) Differences between E- and Z-Enols
∆δ(OH)(E-enol - Z-enol)
∆δ(NH)(E-enol - Z-enol)
N-sub-
stituent
7a-e
7f-j
7k-o
18a-da
18e-ha
7a-e
7f-j
7k-o
18a-da
18e-ha
Ar
range
1.51-1.78 0.27-0.60 -0.33 to
-1.11
2.33-2.67 1.21-1.58 -0.09 to
-0.44
0.14-0.54 0.52-1.02 -0.63 to
-0.94
-0.14 to
-0.46
avg (no. 1.63 ( 0.07 0.42 ( 0.10 -0.59 ( 0.30
of pairs) (9) (7) (12)
Alk range 1.51-1.83 0.40-0.76 -0.20 to -0.58
2.53 ( 0.10 1.42 ( 0.08 -0.21 (0.10
0.36 ( 0.11 0.79 ( 0.13 -0.77 ( 0.11 -0.29 ( 0.10
(13)
(16)
(11)
(12) (12) (16) (16)
-0.13 to -0.42 0.06-0.34 0.44-0.79
-0.29 ( 0.07
(8)
avg (no. 1.68 ( 0.08 0.56 ( 0.09 -0.39 ( 0.12
of pairs)
0.19 ( 0.07 0.60 ( 0.08
(8)
(8)
(8)
(8)
(8)
(d) Differences between Systems 7 and 18
∆δ(OH) (Z-enol)
∆δ(NH) (E-enol)
δ(7a-e)b -
δ(18a-d)
δ(7f-j)c -
δ(18e-h)
δ(7a-e)b -
δ(18a-d)
δ(7f-j)c -
δ(18e-h)
range
avg (no. of pairs)
0.84-1.14
0.98 ( 0.06 (13)
0.92-1.21
1.04 ( 0.06 (16)
0.18-0.61
0.47 ( 0.08 (16)
0.38-0.60
0.50 ( 0.06 (16)
a Including all four N-substituents b Excluding 7c. c Excluding 7h.
and -0.13 ( 0.02 ppm, respectively, supporting this conclusion
for both O-H‚‚‚OdC-OR (R ) Me, CF3CH2) hydrogen bonds.
This applies to all solvents studied, although the differences in
CCl4 are smaller (-0.06 ( 0.01 and -0.04 ( 0.01, respec-
tively).
The δ(NH) for the Z-isomers should behave similarly since
the NH is hydrogen bonded to the OdC-OR. They are 0.32
(0.27) and 0.28 (0.22) ppm for N-Ar(N-Alk) groups for the
changes in R of Me f CF3CH2 f (CF3)2CH, Z-21 f Z-22 f
Z-23 (or Z-24 f Z-25). The differences resemble the related
changes in systems 3 and 4.5,7a,b
By a similar argument, δ(OH) of the Z-enols and δ(NH) of
the E-enols should also be nearly constant since in both there
are identical O-H‚‚‚OdP(OMe)2 and N-H‚‚‚OdP(OMe)2
hydrogen bonds in series 7, provided that the CO2R group does
not influence the δ(OH). However, both δ(OH,Z-enol) and δ-
(NH,E-enol) increase by ca. 0.20 ppm for each of the changes
Me f CF3CH2 f (CF3)2CH in R. This secondary effect which
is still unclear, may arise from transmitting the EW effect of R
group through CdC double bond.
The reduction of the δ(OH) values of the Z-enol by the CF3
groups should be more pronounced on reducing the basicity of
the hydrogen-bonded PdO, when the MeO groups of 7 are
replaced by the CF3CH2O groups of 18. Indeed, a significant
average decrease of 0.98 and 1.04 ppm for the change when R
of the CO2R is Me or CF3CH2 was found. For the weaker
N-H‚‚‚OdP hydrogen bond of the E-enol the effect should be
FIGURE 3. Sideview of the B3LYP/6-31G**-calculated structure of
13b which shows the nonplanarity of the [OdP(OMe)2]2CdC moiety.
system 4 for the Me f CF3CH2 f (CF3)2CH changes are ca.
1.0 and 0.75 ppm, respectively.7a,b From this analogy in solution
the E-enol is the one with O-H‚‚‚OdC-OR hydrogen bond with
the δ values assigned to it in Table 2. Further support arises
from comparison of the analogous systems 7 and 18 (E-21 and
E-24 and E-22 and E-25). For the E-enol the ∆δ(OH) values
are negligible, indicating that the OH is not bonded to the
PdO group, since δ(OH) for 18 in which the PdO group is a
weaker hydrogen bond acceptor than 7 should have been lower
than for 7. Moreover, these differences apply for both the 18a-
d/7a,b,d,e and the 18e-h/7f,g,i,j pairs which are -0.12 ( 0.02
7614 J. Org. Chem., Vol. 72, No. 20, 2007

Dialkoxyphosphinyl-Substituted Enols of Carboxamides
significant but smaller, and values of 0.47 and 0.50 for Me and
CF3CH2, half of those given above were found for δ(OH,E-
enol) and δ(NH,Z-enol) were observed.
negative δ(F) values is amide (less negative) > Z-enol > E-enol.
The differences between the two enols are small, between 0.02
and 0.07 ppm.
13C, 19F and 31P NMR Data. (a) 13C Data. Selected 13C
NMR chemical shifts are given in Table 3. The relevant signals
are Câ and CR of the E- and Z- enols and the amide, the CH2 in
the (R′O)2PO group and the ester carbonyl. The δ values are
given only in CDCl3 for several compounds, except for three
systems for which the values are given in five solvents in order
to demonstrate the solvent effect. The full data are provided in
Table S2 of Supporting Information.
For the hexafluoro derivatives 6k-o/7k-o only two signals
are observed in all solvents except DMSO-d6 due to overlap of
the E-enol and Z-enol signals. The δ(F) values for the amides
are less negative. In the H-F coupled spectrum all signals are
split to doublets (3JHF ) 4.9 - 7.2 Hz) by the methine hydrogen.
For 6l/7l two amide signals are observed in the non-coupled
spectra in DMSO-d6 (see above), which appear as multiplets in
the coupled spectrum. As in the 1H NMR spectrum, 6n-o/7n-o
display the Z-enol signals in DMSO-d6, as well as traces of
Z-7k and Z-7l.
The amide is recognized by its CH signal which appears at
the narrow range of 52.2-54.4 ppm in CDCl3 (52.0-54.83 ppm
in all solvents) and in the C/H coupled spectra as a doublet of
doublets due to both C-H and C-P coupling (1JCH ) 127.9-
The δ(F) values are less negative for the hexa-F than for the
tri-F of the three species.
1
1
133.4 Hz, JCP )120.8-136.6 Hz for 6/7 and JCP ) 129.4-
141.3 Hz for 17/18). In contrast, the enol Câ which appears at
56.8-64.2 ppm, is easily distinguishable since it appears as a
For systems 17a-d/18a-d there is only one type of fluorine
signal in the (CF3CH2O)2PdO group. For 17e-h/18e-h there
are two types of fluorine signals, for (CF3CH2O)2PdO and CF3-
CH2OCO. The values for the former group in 17a-h/18a-h
are more negative than those for the latter ones in 6f-j/7f-j.
1
doublet, being coupled to the P with a JCP ) 201-211.8 Hz
for 6/7 and 212.1-222.1 Hz for 17/18. The N-substituent effect
on δ(Câ) is C6F5 > Ph > p-An > Alk. CR of the enols appears
(c) 31P NMR Data. Selected 31P NMR data in CCl4 only are
given in Table 4 based on H3PO4 as a reference. The complete
data are in Table S3 in the Supporting Information. 31P NMR
is a better quantitative probe than 19F NMR since the signals
are much better separated. Indeed, the relative integration of
the 31P signals is closer in most cases to those of the 19F spectra
2
as a doublet at δ 170.2-173.8 with JCP ) 16.3-20.3 Hz.
Consequently, ∆R,â, which may be regarded as a crude measure
of the polarity of the “CdC” bond of the enol (cf. structure
1b) is >113 ppm, a value higher than that found for the
corresponding diester systems.5,7b,c The amide NCdO signal is
2
1
a doublet at δ 157.4-161.5 ppm with JCP ) 4.5-6.4 Hz.
than to those of the H NMR spectra (Table S4). The three
Assignment of signals to the E- and Z-enols is based on their
relative intensities and comparison with the intensities in the
1H NMR spectra. Consistently, Câ, CR, and CO2R of the E-enol
are at a lower field than for the Z-enol, and this feature can be
used as diagnostic. For CR the differences are 1.0 ( 0.2 for
7c-o and 1.5 ( 0.3 for 18a-g. For Câ the differences are 2.3
( 0.3 for 7c-o and 1.8 ( 0.1 for 18a-g and for δ(CO2R)
they are larger, being 4.8 ( 0.1 ppm for 7c-o and 5.13 ( 0.10
for 18a-g. We ascribe these differences to the stronger O-H‚
‚‚OdC hydrogen bond to CO2R in the E-enol than in the weaker
N-H‚‚‚OdC hydrogen bond in the Z-enol. This is supported
by the opposite differences in the 31P NMR (Table 4) where
δ(31P) is higher by 6 - 7.5 ppm for the Z-enol where the PdO
is hydrogen bonded in the stronger O-H‚‚‚OdP hydrogen bond
than in the E-enol with the N-H‚‚‚OdP bond. This is due to
electron-withdrawal by the hydrogen bond from the MdO (Md
C, P), which increases with its strength. The ester CO signals
are at a lower field than the amide signals. The CH2OPO signals
of 6 are at the same region as the amide CH, and the CH2 of
CF3CH2OPO of 17 is at lower field than the amide CH. In the
uncoupled spectra these signals of the amide appear mostly as
signals observed in most cases were identified by their integra-
tion and splitting pattern in the H-P coupled spectrum. In the
6/7 series the δ(P) signal of the Z-enol appears at the lowest
field of 29.17 - 34.78 ppm in CDCl3 (28.92 - 35.04 ppm in
all solvents), followed by the E-enol signal at 20.38 - 27.73
ppm as an heptet for 6/7 or quintet for 17/18, due to coupling
with the two MeO groups. For the amide δ(P) is at a higher
field of 12.62-18.67 ppm. The R-H-coupled signals display a
nine-line multiplet for 6/7 and eight-line multiplet for 17/18
due to overlap of few lines. The difference in ppm ∆δ ) δ(Z-
enol) - δ(E-enol) is appreciable, and slightly increase with the
increase in the number of fluorine atoms in the ester group.
For 7a-e it is 7.15 ( 0.19, for 7f-j it is 7.32 ( 0.12 and for
7k/7o it is 7.59 ( 0.19. In the (CF3CH2O)2PdO-substituted
systems 18, the differences are smaller and within the combined
experimental errors: 5.92 ( 0.04 for 18a/18d and 5.99 ( 0.03
for 18e/18h.
The Cyano-Activated Systems. The four cyano-activated
1
systems 9a-d/10a-d were analyzed by H NMR spectra. At
298 K in CDCl3 the lowest field signals were for the amides:
at 8.83-9.12 ppm when X ) Ar (9a-c) and at 6.53 ppm for X
) t-Bu (9d). The amides were the major species as judged from
their CH signals. In addition, small signals at 7.58-7.98 ppm
(10a-c) and 5.58 ppm (10d) with 0.9-7.1% intensities of the
amide signals were observed. On cooling to 240 K, these signals
shifted to 8.09-8.66 and 5.66 ppm, and small new signals
(except for 9c) with identical intensity appeared at ca. 13.4 ppm.
In analogy with other systems we ascribe the new low field
signal to an enolic OH, and the higher field signal to the NH of
the enol. The assignment is corroborated by the 31P NMR spectra
which display one small broad signal for X ) Ph, p-An and an
heptet (3JPH ) 11.57 Hz) for X ) C6F5 at a low field of 27.58-
28.97 ppm, with intensities of 3.7-7.3% of the intensity of the
main amide species 9a-c at 12.84-13.64 ppm (d of heptets,
2JHP ) 24.25-24.96 Hz, 3JHP ) 11.44-11.68 Hz). The 1H and
two doublets in 6 or as two quartets of doublets in 17 (2JCP
)
ca. 6.7 Hz). The doubling of the signals is ascribed to the Câ
chiral center and helps in unequivocally assigning these signals
to the amide.
(b) 19F NMR Data. Selected δ(19F) NMR values in CDCl3
using CFCl3 as the reference are given in Table 4. The complete
data are in Table S3 in the Supporting Information. In the
noncoupled spectrum of the trifluoro derivatives 6f-j/7f-j three
singlets are observed in all solvents, except for DMSO-d6. In
DMSO-d6, only one signal, for the amide, is observed, except
for 6i-j/7i-j where traces of the Z-enols were also observed.
In the H-F coupled spectrum all signals are triplets (3JHF
)
7.2-10.9 Hz). The signals are identified by comparison with
1
the relative integrations in the H NMR data. The order of the
J. Org. Chem, Vol. 72, No. 20, 2007 7615

Song et al.
31P NMR spectra enable calculation of Kenol values at 298 K of
0.01-0.08 in CDCl3.
(CF3CH2O)2PdO groups is also appreciable: by 8.5-15-fold
in CDCl3 and by 4.9-6.1-fold in CCl4, THF-d8, and CD3CN.
The change caused by introducing six fluorine atoms in the
phosphonate group is shown by comparing Kenol values for
corresponding compounds in series 6/7 and 17/18. Kenol (17/
18) > Kenol (6/7) when R is Me (CF3CH2) by average values of
9.0 (4.3) in CCl4, 10.3 (10) in CDCl3, 2.5 (1.7) in THF-d8 and
5.0 (3.1) in CD3CN. The increase in Kenol on increasing the
number of fluorine atoms in the carboxylic ester or the
phosphonate ester fit the expectation of higher % enol with
higher electron-withdrawal by the activating Y,Y′ groups. The
effect is smaller in the phosphonate ester (despite the six fluorine
atoms) than by a carboxylic ester. Tentatively, the steric
hindrance of the former group offsets the EWG effect. The small
effect in THF-d8 may be connected with its hydrogen bonding
ability.
The t-Bu derivative 9d/10d already shows at 298 K in CDCl3
or in 2:1 CCl4/CDCl3 a broad signal at 13.22 ppm with 6.5%
intensity of the amide signal at 6.53 ppm, as well as a signal of
the same intensity at 5.58 ppm. On lowering the temperature
the low field signal sharpens. Both signals were ascribed to the
enol 10d. In the 31P NMR spectrum a signal at 29.97 ppm (hept,
3JPH ) 11.66 Hz) with 6.6% of the intensity of the amide signal
2
3
at 14.19 ppm (d of hept, JPH ) 24.57 Hz, JPH ) 11.65 Hz)
was observed.
The less sensitive 13C NMR spectra in CDCl3 had shown
signals consisting only with the amide 9 structures. The CH
carbons at 38.5-39.2 ppm were doublets of doublets due to
coupling with both the geminal hydrogen and phosphorus (1JCH
1
) 135.1-135.5 Hz, JCP ) 127.9-129.1 Hz).
The very low Kenol values of the CN phosphonate systems
in CDCl3 apparently contrast the higher activation of a
(MeO)2PdO than of a CO2Me group in other systems. The
larger steric effect for (MeO)2PdO suggested above for the
absence of enols 13 in solution, although (MeO2C)2CdC(OH)-
NHPh is observable in CDCl3,4 may partially account for the
effect.
The behavior of the cyano-activated systems resembles that
of some (CN)CO2R activated systems Z-4 with N-Ar ) Ph or
p-BrC6H4, in the absence of an enolic OH signal at rt and its
appearance on cooling.7a The reason for this may be an exchange
(and coalescence of signals) of the amide and the enol at rt.
However, these systems are fully enolic both in the solid state
and in solution. Apparently, the activation by a single (MeO)2Pd
O group is lower than that of a CO2Me group when both appear
together with a CN group.
The N-aryl substituent effect for compounds Z-4 gives a
Hammett log Kenol vs σ correlation with a small negative slope,
7a
i.e., electron-donating substituents increase Kenol
.
For com-
Kenol and E/Z Values. Solvent and Substituent Effects. The
pounds 6/7 the effect is not large and less systematic. The
changes from 6-7f to 6-7h and from 6-7k to 6-7m parallel
those for compounds 4, but for 6-7a to 6-7c Kenol first decrease
and then increase again. For pairs of N-p-An and N-Ph in 17/
18 Kenol is higher for the more electron donating p-An.
enols/amide distributions and the derived Kenol values and E/Z
ratios are given in Table 6 for all of the systems, except 12/13
in all of the solvents studied. The ratios were determined by
integration of the NH signals, and occasionally of the amide
CH signals in the 1H NMR spectra, by integration of the three
31P signals, and when available of the three 19F NMR signals.
There is a close similarity (within 1%) in the product distribu-
For N-alkyl substituents Kenol(i-Pr) > Kenol(t-Bu), and Kenol
-
(i-Pr) is mostly the highest among all the N-substituted systems,
as found for systems 3 where for Kenol i-Pr > p-An > Ph.5
1
tions derived from the H and 31P spectra, which is somewhat
better than with the 19F derived values since the 19F signals are
less spread. The enols/amide ratios are averages of the two first
values, or of the three (in parentheses) values. The integrations
for all compounds are given in Table S4 in the Supporting
Information.
The three signals are observed for all systems, except for the
bisphosphonates 12/13 where only the amides 12 were observed.
They were observed in all solvents except for DMSO-d6 in
which mostly, but not always, only the amide signal was
observed.
The solvent effect for 17 out of the 23 systems for which
data are available in five solvents follow the order for Kenol of
CCl4 > CDCl3 > THF-d8 > CD3CN > DMSO-d6. This order
differs from the order of δ(OH) or δ(NH) at 298 K which is
THF-d8 > CD3CN > CDCl3 > CCl4 (Table 2).
Changing the substituent in the ester group of systems 6/7
causes a significant change in Kenol. The amide is the major
species for the CO2Me esters in all solvents, the enols
predominate in the less polar solvents for the CO2CH2CF3 esters,
and the enols are the major species in all solvents , except
DMSO-d6 for the CO2CH(CF3)2 esters. Roughly, Kenol increases
by an order of magnitude when R in the ester changes from
Me to CF3CH2, and by a factor of 7 when R changes from CF3-
CH2 to (CF3)2CH. The changes in log Kenol are roughly the same
but in the opposite direction, as those in δ(OH, ppm) for changes
in these R’s. Obviously, when the Kenol is very high (>100),
the numbers and their ratios are very approximate, since a
fraction of a percent change in one component changes Kenol
enormously.
The Z-enol predominates in all the E-7/Z-7 systems; i.e., E/Z
ratios are <1. They decrease when R in CO2R carries more
fluorine atoms, e.g., in CDCl3 E/Z ) 0.75 (N-Ph) and 0.58 (N-
i-Pr) when R) Me, 0.33 (N-Ph) and (0.30) (N-i-Pr) for R )
CF3CH2 and 0.16 (N-Ph), and 0.13 (N-i-Pr) when R ) (CF3)2-
CH. In each case the ratios are higher in the chlorinated than in
the more polar solvents. In contrast, for E-18/Z-18, R ) CO2-
Me the E/Z ratio >1, but for R ) CF3CH2 it is <1. The other
characteristic features are the same as for the E-7/Z-7 systems.
There is a delicate balance between several effects to
determine the stability difference between the E- and Z-enols.
The E/Z ratio in series 7 is <1. When the (MeO)2PdO groups
in 7 are replaced by the more EW (CF3CH2O)2PdO groups in
18 the O-H‚‚‚OdP hydrogen bond in Z-18 become weaker
and the effect is more pronounced than on the N-H‚‚‚OdP
bond in E-18. Consequently, the E/Z ratio increases and becomes
>1. For 18e-h the same effect also increase the E/Z ratio over
that for 7a-e, but since R ) CF3CH2, the E/Z value for 7f-j
is already lower than for 7a-e, and the increased E/Z value for
18e-h remains <1.
Relationship between the Stability of the E/Z Isomers and
the Chemical Shifts. In compounds 3 and 4 the order of δ-
(OH) and δ(NH) values as a function of R was the same as the
order of stability of the species and correlations were observed
between δ(OH) and δ(NH) and Kenol values.5 The higher the
δ(OH), the higher the Kenol value. This is not the case in the
present systems. The ∆δ(OH)(E-enol - Z-enol) values change
from high positive values of 1.68 (N-Ar and N-Alk) for 7a-e
to 0.42 (N-Ar) and 0.56 (N-Alk) for 7f-j to negative values of
-0.68 (N-Ar) and -0.39 (N-Alk) for 7k-o. Consequently, there
is an approximate additivity of ca. 1 ppm for addition of 3 and
The increase in Kenol on changing the substituent from R )
Me to CF3CH2 in the ester group for systems 17/18 with the
7616 J. Org. Chem., Vol. 72, No. 20, 2007

Dialkoxyphosphinyl-Substituted Enols of Carboxamides
TABLE 6. Percentage of E-Enol, Z-Enol, and Amide, Kenol, and E/Z-Enol Ratio for the Amide/Enols at 298 K in Different Solventsa
compd
solvent
E-Enol
20.7
4.35
4.0
0.6
Z-Enol
27.8
5.85
7.5
1.2
amide
Kenol
E/Z
6a/7a
CCl4
CDCl3
THF-d8
CD3CN
DMSO-d6
CCl4
51.5
0.94
0.11
0.13
0.74
89.8
88.5
98.2
0.74
0.53
0.50
0.018
e0.01
0.61
0.098
0.087
0.014
e0.01
0.76
0.11
0.06
0.014
e0.01
1.11
0.16
0.41
0.041
e0.01
0.66
0.10
0.21
0.024
e0.01
100
6b/7b
6c/7c
6d/7d
6e/7e
6f/7f
13.8
3.8
2.95
23.9
5.1
5.05
62.3
91.1
92.0
98.65
100
56.8
90.05
94.30
98.60
0.58
0.75
0.58
0.50
CDCl3
THF-d8
CD3CN
DMSO-d6
CCl4
0.45
0.90
17.3
25.9
0.66
0.73
0.31
0.33
CDCl3
4.20
1.35
0.35
5.75
4.35
1.05
THF-d8
CD3CN
DMSO-d6
CCl4
100
18.4
34.1
47.5
86.4
71.0
96.1
100
60.1
90.55
82.5
97.65
100
10.0 (10.2)
44.25 (44)
53.5 (54.1)
87.2 (86.9)
0.54
0.58
0.44
0.44
CDCl3
5.0
8.8
1.2
8.6
20.2
2.7
THF-d8
CD3CN
DMSO-d6
CCl4
15.0
3.45
5.3
0.8
24.9
6.0
12.2
0.60
0.58
0.43
0.52
CDCl3
THF-d8
CD3CN
DMSO-d6
CCl4
1.55
19.6 (21.3)
13.35 (14.5)
8.6 (8.7)
70.4 (68.5)
42.4 (41.5)
37.9 (37.2)
10.55 (10.6)
9.0 (8.8)
0.28 (0.31)
0.32 (0.35)
0.23 (0.23)
0.21 (0.24)
CDCl3
1.26 (1.27)
0.87 (0.85)
0.15 (0.15)
e0.02
6.8 (6.7)
1.07 (1.09)
0.54 (0.56)
0.11 (0.11)
THF-d8
CD3CN
DMSO-d6
CCl4
2.25 (2.5)
100
6g/7g
6h/7h
6i/7i
19.7 (20.4)
12.7 (14.1)
6.25 (7.1)
1.9 (1.9)
67.5 (66.6)
38.9 (38.1)
28.8 (28.6)
8.2 (8.3)
12.8 (13.0)
48.4 (47.8)
64.95 (64.3)
89.9 (89.8)
0.29 (0.31)
0.33 (0.37)
0.22 (0.25)
0.23 (0.23)
CDCl3
THF-d8
CD3CN
DMSO-d6
CCl4
100
e0.02
17.2
67.5
15.3 (15.3)
51.9 (52.2)
68.1 (67.5)
90.8 (90.5)
5.53 (5.53)
0.93 (0.92)
0.47 (0.48)
0.10 (0.10)
0.25
CDCl3
11.0 (11.4)
3.9 (3.8)
1.2
37.1 (36.4)
28.0 (28.7)
8.0
0.30 (0.31)
0.14 (0.13)
0.15
THF-d8
CD3CN
DMSO-d6
CCl4
100
e0.02
19.6 (20.4)
14.85 (15.2)
11.0 (11.5)
4.8 (4.7)
71.9 (71.2)
48.7 (48.9)
60.0 (60.5)
22.9 (23.1)
4.2 (4.2)
70.4 (69.7)
42.0 (42.2)
50.1 (50.2)
14.5 (14.4)
2.7 (2.7)
86.8
80.6
80.8
55.2
4.1
86.2
79.9
74.6
47.9
8.5 (8.2)
36.45 (35.9)
29 (28)
10.8 (11.2)
1.74 (1.79)
2.45 (2.57)
0.38 (0.39)
0.04
7.62 (7.62)
1.21 (1.24)
1.48 (1.53)
0.21 (0.21)
0.028 (0.028)
61.5 (57.8)
14.9 (15.7)
7.6 (7.4)
0.27 (0.29)
0.30 (0.31)
0.18 (0.19)
0.21 (0.21)
b
CDCl3
THF-d8
CD3CN
DMSO-d6
CCl4
72.3 (72.2)
95.8 (95.8)
11.6 (11.6)
45.2 (44.6)
40.3 (39.5)
82.4 (82.4)
97.3 (97.3)
1.6 (1.7)
6j/7j
18.0 (18.7)
12.8 (13.2)
9.6 (10.3)
3.1 (3.2)
0.26 (0.27)
0.30 (0.31)
0.19 (0.21)
0.21 (0.22)
b
CDCl3
THF-d8
CD3CN
DMSO-d6
CCl4
6k/7k
6l/7l
11.6
13.1
7.6
0.13
0.16
0.094
0.089
CDCl3
6.3 (6)
THF-d8
CD3CN
DMSO-d6
CCl4
11.6 (11.9)
39.9 (40.4)
95.9 (95.4)
1.9 (2)
4.9
1.51 (1.48)
0.043 (0.048)
51.6 (49)
11.9
12.9
7.0
0.14
0.16
0.094
0.094
CDCl3
7.2 (7.1)
12.9 (13.1)
4.43 (4.32)
1.10 (1.12)
0.025 (0.026)
31.3 (26.8)
8.52 (8.71)
3.08 (3.05)
0.68 (0.70)
(0.032)c
THF-d8
CD3CN
DMSO-d6
CCl4
18.4 (18.8)
47.6 (47.2)
97.3 (97.5)
3.1 (3.6)
10.5 (10.3)
24.5 (24.7)
59.5 (58.7)
96.9c
4.5
2.7
87.3
77.8
70.9
6m/7m
6n/7n
6o/7o
9.6
11.7
4.6
0.11
0.15
0.065
0.047
CDCl3
THF-d8
CD3CN
DMSO-d6
CCl4
1.8
38.7
3.1c
10.3
11.25
6.9
89.5
85.55
90.35
76.6
25.0
89.6
83.1
85.9
0.2
ca. 496
0.11
0.13
0.08
0.08
CDCl3
3.2 (3.23)
3.25 (3.17)
17.3 (16.8)
75.0 (75.1)
0.4
30.25 (29.93)
29.92 (30.55)
4.78 (4.95)
0.33 (0.33)
THF-d8
CD3CN
DMSO-d6
CCl4
6.1
10.0
11.6
7.15
5.6
249
0.11
0.14
0.08
0.09
CDCl3
5.1 (5.2)
17.9 (18.2)
13.4 (14.0)
2.42 (2.48)
0.18 (0.18)
THF-d8
CD3CN
DMSO-d6
6.95 (6.67)
29.2 (28.7)
84.6(84.4)
65.2
15.4 (15.6)
J. Org. Chem, Vol. 72, No. 20, 2007 7617

Song et al.
Table 6 (Continued)
compd
solvent
CDCl3
E-enol
Z-enol
amide
Kenol
E/Z
9a/10a
7.2
92.8
0.08
2.3d
4.0
97.7d
96.0
0.02d
0.04
9b/10b
9c/10c
CDCl3
CDCl3
1d
99d
0.01d
e0.01e
0.04e
0.01d
0.07
0.9e
99.1e
96c
4c
1.4d
6.6
98.6d
9d/10d
CDCl3
93.4
5d
95d
0.05d
6.58 (6.63)
1.05 (1.04)
0.34 (0.33)
0.09 (0.09)
17a/18a
CCl4
55.0
31.8
13.2 (13.1)
48.8 (48.9)
74.8 (75.0)
91.8 (91.5)
100
16.0 (16.0)
51.8 (52.4)
81.5 (81.4)
93.2 (92.7)
100
8.5 (8.6)
36.2 (36.1)
54.7 (54.2)
82.7 (82.0)
100
17.2 (17.0)
47.6 (48.2)
64.3 (64.7)
89.4 (89.0)
100
2.2
6.0
1.72
CDCl3
THF-d8
CD3CN
DMSO-d6
CCl4
31.1 (29.5)
14.0
20.1 (21.6)
11.2
1.55 (1.37)
1.25
4.5 (4.8)
3.7 (3.7)
1.22 (1.30)
e0.02
17b/18b
17c/18c
17d/18d
17e/18e
17f/18f
55.1
28.9
5.25 (5.25)
0.93 (0.91)
0.23 (0.23)
0.07 (0.08)
1.91
CDCl3
THF-d8
CD3CN
DMSO-d6
CCl4
CDCl3
THF-d8
CD3CN
DMSO-d6
CCl4
CDCl3
THF-d8
CD3CN
DMSO-d6
CCl4
CDCl3
THF-d8
CD3CN
DMSO-d6
CCl4
30.1 (31.4)
10.7
18.1 (16.2)
7.8
1.66 (1.94)
1.37
4.1
2.7
1.52
e0.02
56.5 (55.7)
36.7 (38.8)
23.1
35.0 (35.7)
27.1 (25.1)
22.2
10.7 (10.6)
1.76 (1.77)
0.83 (0.85)
0.21 (0.22)
e0.02
4.80 (4.88)
1.10 (1.07)
0.55 (0.55)
0.11 (0.12)
e0.02
44.5
1.61 (1.51)
1.35 (1.55)
1.04
9.5
7.8
1.22
51.5 (52.1)
30.2 (30.0)
19.2
31.3 (30.8)
22.2 (21.8)
16.5
1.65 (1.69)
1.36 (1.38)
1.16
5.9
4.7
1.26
41.4
40.6
21.3
11.6
56.4
53.4
38.9
21.0
0.73
0.76
0.55
0.55
15.7
1.51
0.48
e0.02
32.3
11.5
39.8
67.4
100
42.7
40.7
17.7
9.7
54.3
51.3
30.2
16.0
3.0
8.0
52.1
74.3
0.79
0.79
0.59
0.61
CDCl3
THF-d8
CD3CN
DMSO-d6
CCl4
0.92
0.35
100
e0.02
499
17g/18g
17h/18h
40.1
38.0
27.0
17.3
59.7
56.9
55.5
35.3
6.1
55.7
53.7
44.3
25.1
0.2
5.1
0.67
0.67
0.49
0.49
CDCl3
THF-d8
18.6
b
17.5
4.71
CD3CN
DMSO-d6
CCl4
47.4
1.11
93.9
0.06
40.2
36.6
23.9
14.1
4.1
9.7
23.4
0.72
0.68
0.54
0.56
CDCl3
9.31
THF-d8
CD3CN
DMSO-d6
31.8
60.8
100
2.14
0.64
e0.02
a Data for system 12/13, which forms the amide exclusively are not given. The ratios are averages of integrations of the 1H (NH) signals and 31P signals
1
(without parentheses) and of the H, 31P, and 19F values (in parentheses) for the fluoro-containing compounds. b Based on OH and NH of amide due to a
1
1
partial overlap of the NH signals of E- and Z-enols. c Based on 31P NMR. No enol was observed by H NMR. d Measured at 240K. e Based on H NMR.
6 fluorine atoms in the ester group. Likewise, the sign of ∆δ-
(NH)(E-enol - Z-enol) also changed from -0.29 (N-Ar) and
-0.36 (N-Alk) for 7a-c to 0.36 (N-Ar) and 0.19 (N-Alk) for
7f-j to 0.71 (N-Ar) and 0.58 (N-Alk) for 7k-o. There is less
than an additive change in ∆δ(NH) of ca. 0.6 and 0.37 ppm
for the 0F f 3F f 6F change.
From Table 6 the Z-enol is always formed in excess, except
for 17a-d/18a-d and hence, it is the more stable enol. This
preferred stability increases with the increased number of
fluorine atoms in the CO2R.
Both facts fit a single pattern. The shift in ∆δ(OH) with the
number of fluorines in R of E-7 is due to the decreasing
hydrogen bond accepting ability by the CdO oxygen resulting
since R is better EWG. The effect on the Z-enol should be
smaller since the same PdO group is hydrogen bonded to the
O-H in all systems. In practice δ(OH) is slightly increasing,
probably the O-H bond is more acidic due to the increased
electron-withdrawal from this bond by the remote ROCdO
group. The weaker hydrogen bond to O-H in the E-isomer
reduces its stability together with the slightly increased hydrogen
bond strength of the Z-isomer which increases its stability,
results in the parallel decrease of the E/Z ratios.
A complete analysis of the relative stabilities should take into
account the other weaker hydrogen bonds to the N-H present
in each enol. The decreased strength of the O-H‚‚‚OdM (M
) P, C) hydrogen bonds is accompanied by an increased
strength of the N-H‚‚‚OdM (M ) C, P) hydrogen bond, so
that the overall stability of the species reflects the energy
difference between the two hydrogen bonds. This is the reason
why the E/Z ratios, translated to energy differences of 0.17-
1.8 kcal/mol for 6/7 and 17e-h/18e-h and -0.3 to -0.13 kcal/
mol for 17a-d/18a-d are not large, enabling to see both E-7
and Z-7 (and E-18 and Z-18) in solution. When one hydrogen
bond cannot be formed, as in the cyano esters, due to the linear
geometry of the CN group, only the Z-isomer is observed in
solution.
The similarity between the enols of diesters with 0, 3, and 6
F atoms and the dialkoxyphosphinyl esters with a similar pattern
of fluorine substitution is shown in a plot of the literature δ-
7618 J. Org. Chem., Vol. 72, No. 20, 2007

Dialkoxyphosphinyl-Substituted Enols of Carboxamides
) 0.9995). The inversion in the positions of δ(E-enol) and δ-
(Z-enol) is reflected by the negative ∆δ(OH) value when R )
(CF3)2CH.
Comparison of (R′O)2PdO and CO2R Groups. The dif-
ference in electron-withdrawing power of phosphonate and ester
groups and of hydrogen bonding abilities of the phosphoryl and
carbonyl groups affects the δ(OH), δ(NH), E/Z enol ratios and
Kenol values, and their differences from those of compounds 3
and 4. Evaluation of these differences is therefore necessary
for understanding the behavior of our systems.
A single phosphonate ester is a better electron-withdrawing
and somewhat better negative charge delocalizing substituent
than a single methyl ester as judged by various types of Taft’s,
-
Hammett’s, and Brown’s σ parameters. For (MeO)2PO σR, σR
,
σm, σp-, σp, and σp+ are 0.21, 0.33, 0.46, 0.87, 0.59, and 0.73,
respectively, and for CO2Me they are 0.11, 0.30, 0.36, 0.74,
0.44, and 0.49, respectively.19a The σ* values are 2.45
((MeO)2PO)19b and 2.00 (CO2Me).19c
FIGURE 4. Plot of δ(OH, Z-enol) of the diesters 3 (N-substituents )
Ph, p-An, p-MeC6H4, and p-BrC6H45 vs δ(OH, E-enol) of the dialkoxy-
phosphinyl ester systems 7 and 18 (N-substituents ) Ph, p-An, and
C6F5) in CDCl3 at 298 K. δ(OH, Z-enol) ) 1.14δ(OH, E-enol) - 1.161
(R2 ) 0.9933).
The PdO group is a well-known strong hydrogen bond
acceptor,20 e.g., in an intramolecular hydrogen bond with OH
20c
in HO(CH2)nP(dO)Ph2 or in (RO)2P(dO)CH2CH(OH)R′,21
but only a few works address the question of the relative
hydrogen bond accepting ability in competition of a PdO with
a CdO group. Deshielding of the P in 2-hydroxyalkylphospho-
nates was ascribed to a decreased π-donation from intramo-
lecular hydrogen bonding between the PdO oxygen and the
OH group.21a PdO‚‚‚H-O bonding was also found in polymers
and it is assumed that in the hydrogen bond the PdO becomes
1
softer.21b By conformation analysis based on H NMR spectra
the “PdO” is favored over CdO in an intramolecular interaction
with a â-OH group.22 From pKa’s of PdO and CdO groups it
was deduced that protonation of the former in ArCOPO(OH)-
OPh is preferred over that of latter.23
Quantitative data on hydrogen bond parameters of the
(MeO)2PdO and CO2Me groups for use in LFERs indicate that
the former is a better hydrogen bond acceptor.24 The only
quantitative data relevant to our system comes from the IR ∆ν
shift measurements of the formation constants of the hydrogen
bond adducts between phenol and RCH2R′ (R ) R′ ) MeCO,
MeO2C, (EtO)2PdO) which give data on intermolecular com-
petition, and for R ) (EtO)2PO, R′ ) CO2Et for intramolecular
FIGURE 5. Plot of the average % E-enol in the enols mixture for
7a-e,f-j,k-o vs the average ∆δ(OH) ) δ(E-enol) - δ(Z-enol) in
CDCl3 at 298 K: (A) N-Ar groups, %E-enol ) 13.71∆δ + 18.13 (R2
) 0.9947); (B) N-alkyl groups, %E-enol ) 12.47∆δ + 13.73 (R2 )
0.9995).
(19) (a) Charton, M. In The Chemistry of Cyclobutanes; Rappoport, Z.,
Liebman, J. F., Eds.; Wiley: New York, 2005; Chapter 10, p 441. (b)
Katzhendler, J.; Ringel, I.; Karaman, R.; Zaher, H.; Breuer, E. J. Chem.
Soc., Perkin Trans. 2 1997, 341. (c) Taft, R. W., Jr. In Steric Effects in
Organic Chemistry; Newman, M. S., Ed.; Wiley: New York, 1956; Chapter
13, p 619.
(20) (a) Corbridge, D. C. E. Phosphorus. An Outline of its Chemistry,
Biochemistry and Technology, 4th ed.; Elsevier: Amsterdam, 1990; Chapter
14.1. (b) Gramstad, T. Spectrochim. Acta 1963, 19, 1363. (c) Asknes, G.;
Albriktsen, P. Acta Chem. Scand. 1966, 22, 1866. (d) Gramstad, T.;
Storesund, H. J. Spectrochim. Acta Part A 1970, 26A, 426. (e) Bel’skii, V.
E.; Bakeeva, R. F.; Kudryavtseva, L. A.; Kurguzova, M. A.; Ivanov, B. E.
Zh. Obshch. Khim. (Eng. Transl.) 1975, 43, 2568. (f) Raevskii, O. A.;
Gilyazov, M. M.; Levin, Y. A., Zh. Obshch. Khim. 1978, 48, 1053. (g)
Bel’skii, V. E.; Ashrafullina, L. Kh. Zh. Obshch. Khim. (Engl. Transl.) 1979,
49, 1968.
(21) (a) Genov, D. G.; Tebby, J. C. J. Org. Chem. 1996, 61, 2454. (b)
Alexandratos, M.; Zhu, X. Macromolecules 2005, 38, 5981.
(22) (a) Belciug, M. P.; Modro, A. M.; Modro, T. A. J. Phys. Org. Chem.
1992, 5, 787. (b) Belciug, M. P.; Modro, P. A.; Wessels, P. J. J. Phys. Org.
Chem. 1993, 6, 523. (c) Bourne, S.; Modro, A. M.; Modro, T. A.
Phosphorus, Sulfur Silicon 1995, 102, 83.
(23) Ta-Shma, R.; Schneider, H.; Mahajana, M.; Katzhendler, J.; Breuer,
E. J. Chem. Soc., Perkin Trans. 2 2001, 1404.
(24) (a) Abraham , M. H.; Duce, P. P.; Pirov, D. V.; Barratt, D. G.;
Morris, J. J.; Taylor, P. J. J. Chem. Soc., Perkin Trans. 2 1989, 1355. (b)
Hickey, J. P.; Passino-Reader, D. R. EnViron. Sci. Technol. 1991, 25, 1753.
(c) Raevsky, O. A.; Grigor’ev, V. Y.; Kireev, D. B.; Zefirov, N. S. Quant.
Struct. Act. Relat. 1992, 11, 49.
(OH, Z-enol) of the diesters series vs the δ(OH, E-enol) in the
phosphonate series (Figure 4). In both cases, the OH is hydrogen
bonded to a CO2R group, while the other group (Me ester or P
ester) is constant. The excellent linearity (R2 ) 0.9933) and
the slope of 1.14 indicate a similar behavior of the diesters and
of the phosphonate esters.
A similar argument as for ∆δ(OH) applies also for the
∆δ(NH)(E-enol - Z-enol) except that due to the weaker
N-H‚‚‚OdM (M ) P, C) bonds the effects are smaller.
Although the lower field signal is not always that for the
more stable enol, we may expect a correlation between the
abundance of one enol in the enols mixture and ∆δ(OH)
between the two isomers, since both parameters are affected
by the number of fluorine atoms in the ester group R. Indeed,
a plot of the average percentage of the E-enol (100% [E-enol]/
([E-enol] + [Z-enol]) vs ∆δ(OH) ) δ(E-enol) - δ(Z-enol) in
CDCl3 at 298 K is linear, but depends on the N-substituent X.
Figure 5 shows two parallel lines, one for the aryl (Ar ) p-An,
Ph) substituents and one for the alkyl (Alk ) i-Pr, t-Bu)
substituents, with slopes of 13.7 (R2 ) 0.9947) and 12.47 (R2
J. Org. Chem, Vol. 72, No. 20, 2007 7619

Song et al.
TABLE 7. Comparison of Kenol Values for Dialkoxyphosphinyl-
In most cases, Kenol[(MeO)2PdO] > Kenol[MeOCdO] when
the other group is constant. From the data quoted above the
OdP(OMe)2 group is a better H-bond acceptor and EWG and
hence a better enol stablizing group than CO2Me, and more so
when the number of fluorines in the other ester group increases.
Such comparison may be misleading since the Kenol values
are composite. A dissection to the values for the E- and
Z-configurations is required and similar configurations of the
(MeO)2PdO and CO2Me derivatives should be compared. The
major Z-enol of the phosphonate has a similar configuration to
the major E-enol of the ester because both (MeO)2PdO and
CO2Me are cis to the OH. The KenolP(Z)/KenolE(E) values (data
not shown) also support the conclusion that a OdP(OMe)2 is a
better enol stabilizing group than CO2Me.
and Ester-Activated Carboxamides in Different Solvents at 298 K
(MeO)2P(dO)CH(CO2R)CONHX and
MeOCOCH(CO2R)CONHX
P
E
E
R
X
solvent
Kenol
Kenol
KenolP/Kenol
Me
Ph
CDCl3
CCl4
CDCl3
CD3CN
CCl4
CDCl3
CD3CN
CCl4
CDCl3
CD3CN
CCl4
CDCl3
CD3CN
CCl4
CDCl3
CD3CN
CCl4
0.098
1.10
0.16
0.041
9.0
1.26
0.15
6.8
1.07
0.11
10.8
1.74
0.38
61.5
0.05a
1.4b
1.96
0.79
1.6
i-Pr
p-An
Ph
0.1b
e0.01b
3.5c
g4.1
2.57
2.42
g7.5
2.96
3.06
g5.5
1.07
1.74
g38
e1.23
3.04
g75
3.91
3.39
g55
CH2CF3
0.52c
e0.02c
2.3c
0.35c
e0.02c
10.1b
1b
i-Pr
p-An
Ph
We did not find σ values for the CO2CH2CF3 and CO2CH-
(CF3)2 groups, but the gas-phase data show that ∆G° acidity
values for CH2(CO2Me)2 and CH2(CO2Me)CO2CH2CF3 are
341.3 and 333.5, respectively. We have no data for the
hexafluoro-substituted ester, but for CH2(CN)CO2R the ∆G°
values are 334.5, 324.7 and 317.5 for R ) Me, CF3CH2 and
CH(CF3)2, respectively.25 Hence, the fluorinated esters are
expected to be equal or better than OdP(OMe)2 as enol
promoting groups.
e0.01b
CH(CF3)2
g50c
14.9
1.5
51.6
12.9
4.9c
e0.02c
13.2c
CDCl3
CD3CN
3.8c
1.10
e0.02c
(CF3CH2O)2P(dO)CH(CO2R)CONHX and
(CF3CH2O)COCH(CO2R)CONHX
P
6-F
In CCl4 and CDCl3, Kenol < Kenol3-F, Kenol
(these
P
E
E
R
X
solvent
Kenol
Kenol
KenolP/Kenol
designations relate to the enols of fluorinated esters). In most
3-F
6-F
Me
p-An
CCl4
6.58
1.05
0.09
5.25
0.93
0.07
10.7
1.76
0.21
44.5
15.7
0.48
32.3
11.5
3.5c
0.52c
e0.02c
2.3c
1.88
2.02
g4.5
2.28
2.66
g3.5
1.06
1.76
g21
e0.89
2.15
6
cases, the KenolP/Kenol
and KenolP/Kenol
ratios are 0.12-
CDCl3
CD3CN
CCl4
CDCl3
CD3CN
CCl4
CDCl3
CD3CN
CCl4
CDCl3
CD3CN
CCl4
0.28 and 0.022-0.045, respectively, and even close to zero due
to complete enolization of some fluorinated ester derivatives.
Consequently, the fluorinated esters are equal or better than Od
P(OMe)2 as enol stabilizing groups. However, in CD3CN the
KenolP/Kenol3-F values are 1.48-4.8 (for constant CO2Me group
they are >0.6 and 0.75 for X ) Ph and p-An, respectively).
KenolP/Kenol6-F values are g0.48. The values in CD3CN (includ-
ing KenolP/KenolE) are much larger than the corresponding values
in CCl4 and CDCl3. It is not simple to explain the difference in
term of solvation since four species (enol and amide in two
solvents) are involved in the comparison. In the more polar
DMSO-d6, enols of the N-i-Pr and N-t-Bu derivatives of
(MeO)2P(dO)CH(CO2CH(CF3)2)CONHX are observed at 298
K with Kenol ) 0.33 and 0.19, respectively. In DMSO-d6/CCl4
(v/v) mixtures the % of E- and Z-enol of 6o decreases on
addition of DMSO-d6. At 0.08:1, 0.5:1, and 1:1 ratios, the %
E-enol and % Z-enol are ca. 7.8 and 84.5, 0 and 61.2, 0 and
43.1, respectively. The data also indicate that the Z-isomer is
more stable than the E-isomer and is even present in DMSO-
d6.
Ph
0.35c
e0.02c
10.1b
1.0b
i-Pr
p-An
Ph
e0.01b
CH2CF3
g50
7.3c
0.08c
32c
1.01
1.72
7
CDCl3
CD3CN
6.7c
0.35
0.05c
a ref 4. b ref 7b. c ref 5.
competition.18 It was concluded that the PdO group is a strong
hydrogen bond acceptor and with respect to phenol it is ca. 100-
fold more effective than a CdO group. As shown above, this
preference is an important factor in determining the configu-
ration of our enols.
DFT Calculations. The relative free energies and geometries
of the amide and the two enols (cf. Table S5 in the Supporting
Information) and the Kenol values were calculated at B3LYP/
6-31G** for the N-Ph derivatives 6b/7b, 6g/7g and 6l/7l, for
the cyano(phosphonate) derivative 9b/10b and for the bis-
(phosphonate)acetanilide 12b/13b. The enol with a hydrogen-
bonded OH to the CdO of ester group is the most stable species
E-7b, being 1.4 kcal/mol more stable than the enol Z-7b and
4.1 kcal/mol more stable than the most stable amide conformer
6b. The pKenol values at 298 K are -3.0 and -2.0 for the E-
and Z- enols, respectively. Substitution by three or six fluorine
atoms inverts the stability and the Z-7g is 1.7 kcal/mol more
stable than E-7g and 7.4 kcal/mol more stable than the amide.
The pKenols are -5.4 and -4.2, respectively. Likewise, Z-7l is
0.5 and 6.9 kcal/mol more stable than E-7l and amide 6l,
respectively and the derived pKenols are -5.1 and -4.7. In
Dialkoxyphosphinyl vs Ester Comparison. There are many
systematic Kenol values, E/Z ratios and structural information
from our previous work on diester-activated enols 3,4,5,7c which
are analogs of systems 6/7, 9/10, and 17/18. This enables
comparisons with systems such as 3, R1 ) CO2Me, and
evaluation of the effect of, e.g., replacing a CO2Me group by a
(MeO)2PdO group. Comparative data are given in Table 7
P
E
where Kenol and Kenol are the equilibrium constants for the
compared phosphonate and ester groups, respectively.
For a system with a constant CO2Me group, the Kenol(Od
P(OMe)2)/Kenol(CO2Me) ratios are ca. 1.96 for X ) Ph in CDCl3,
and 0.79, 1.6 and G4.1 for X ) i-Pr in CCl4, CDCl3, and CD3-
CN, respectively. For a constant CO2CH2CF3 group, the Kenol
-
(OdP(OMe)2)/Kenol(CO2Me) ratios in the same solvents are
2.96, 3.06, G5.5 when X ) Ph, 2.57, 2.42, and G7.5 when X
) p-An, and 1.07, 1.74, and G38 when X ) i-Pr, respectively.
For a constant CO2CH(CF3)2 group, the ratios in CCl4, CDCl3,
and CD3CN are 3.91, 3.39, and G55 for X ) Ph, and E1.23,
3.04, and G75 for X ) p-An, respectively.
(25) Mishima, M.; Matsuoka, M.; Lei, Y. X.; Rappoport, Z. J. Org. Chem.
2004, 69, 5947.
(26) Frey, J.; Rappoport, Z. Z. J. Org. Chem. 1997, 62, 8327.
7620 J. Org. Chem., Vol. 72, No. 20, 2007

Dialkoxyphosphinyl-Substituted Enols of Carboxamides
contrast, the amides 9b and 12b are calculated to be the more
stable species, by 1.6 and 5.7 kcal/mol than 10b and 13b,
respectively.
In order to evaluate the effect of the (MeO)2PdO and CO2-
Me reactions on both components of the enolization equilibria
the energies of the isodesmic reactions 6-9 were calculated
when R ) (MeO)2PdO and CO2Me. In eqs 6 and 8, the
CONHPh group is transferred from the amide 6 to CH4
(MeO)2POCHRCONHPh + CH4 h
CH3CONHPh + (MeO)2POCH2R (6)
(MeO)2POCRdC(OH)NHPh + CH4 h
CH2dC(OH)NHPh + (MeO)2POCH2R (7)
FIGURE 6. Correlations of the observed pKenol(total) or the observed
pK(E-enol) vs the calculated pK(E-enol): (A) calcd pK(E-enol) vs obsd
pKenol(total), y ) 0.798x - 3.923 (R2 ) 0.9414); (B) calcd pK(E-enol
vs obsd pK(E-enol), y ) 1.019x - 4.54 (R2 ) 0.9451).
)
(MeO)2POCHRCONHPh + CH2dCH2 h
CH3CONHPh + (MeO)2POCRdCH2 (8)
slopes of 0.80 (R2 ) 0.9414) and 1.02 (R2 ) 0.9451),
respectively (Figure 6).
(MeO)2POCRdC(OH)NHPh + CH2dCH2 h
CH2dC(OH)NHPh + (MeO)2POCRdCH2 (9)
Conclusions. Carboxamides (R′O)2P(dO)CH(CO2R)CONHX
(R ) Me, CF3CH2, (CF3)2CH, R′ ) Me, CF3CH2) appear in
CCl4, CDCl3, THF-d8, and CD3CN solutions as mixtures of
amide and the E- and Z-enols, and in DMSO-d6 mostly as the
amides. δ(1HO) values of the E-enols when R ) Me are at the
lowest field and ∆δ(OH) (E-enol - Z-enol) are at ca. 1.6 ppm
to a higher field. When R ) CF3CH2 and (CF3)2CH ∆δ(OH)
decrease by ca. 1 and 2 ppm, respectively, and for R ) (CF3)2-
CH δ(OH, Z-enol) > δ(OH, E-enol). The Z-enols are more
abundant (stable) than the E-enol by 0.17-1.8 kcal/mol for 6/7
and 17e-h/18e-h and less abundant than the E-enol by -0.3
to -0.13 kcal/mol for 17a-d/18a-d. Except when R′ is
CF3CH2, the % of the E-enol in the E/Z mixtures is linear
with ∆δ(OH) value. Analysis of the δ(OH), δ(NH), δ(13C),
δ(19F), and δ(31P) values indicate that for each enol there
are two simultaneously hydrogen bonds O-H‚‚‚OdM and
N-H‚‚‚OdM with M ) C and M ) P. The hydrogen bond
strengths follow the order for PdO > CdO, they decrease when
the number of F atoms in R increase, and they are responsible
for the configurations of the enols. The solid-state structure when
R ) R′ ) Me is that of the amide, but when R ) (CF3)2CH, R′
= Me it is the Z-enol. [(MeO)2PO]2CHCONHX form only the
amides in solution, and (MeO)2POCH(CN)CONHX form only
the Z-enol in e7.3%. DFT calculations confirm the experimental
trends, but the Kenol values differ somewhat from the experi-
mental values.
The general message is that the effect of OdP(OR′)2 groups
on the extent of enolization of a CONHX group is mostly, but
not always, consistent with conclusions obtained earlier with
other activating groups, especially CO2R. The changes in Kenol
values by fluorine substitution, or by the change in the solvent
are determined by the EW ability and especially by the hydrogen
bond accepting ability of the activating group. The latter effect
when applies to the simultaneous hydrogen bonding to the OH
and NH groups of the enol predominates in determining the
E/Z enol ratios. On the other hand, caution should be exercised
in extrapolation from well understood systems to slightly less
understood ones, e.g., from the close analogy between Od
P(OR′)2, CO2R′-substituted systems and two ester systems where
a single OdP(OMe)2 group is a better activator to enol formation
than a single CO2Me, to a doubly activated system where two
CO2Me groups are better activators than two OdP(OMe)2
groups. Likewise, a CN, OdP(OMe)2 activated system is less
enolic than a CN, CO2Me activated system. Hence, effects such
as steric crowding should be also taken into account. The overall
or to CH2dCH2, formally eliminating the interaction between
the enolizing amide and the activating groups. In eqs 7 and 9
the transfer is similar from the precursor enol.
The ∆G values for R ) (MeO)2PdO are 0.0, 24.5, 0.8 and
25.3 kcal/mol for eqs 6-9, respectively, and the corresponding
values for R ) CO2Me are -2.2, 32.1, -7.6 and 26.7 kcal/
mol, respectively. Hence, the main stabilizing interaction in the
6b/7b system is of the enol species, which is larger for R )
CO2Me than for R ) (MeO)2PdO. Equation 10, which is the
difference between eqs 6 and 7, or 8 and 9 gives the calculated
energetics of the equilibria between 6b and 7b relative to the
enolization of acetanilide.
(MeO)2POCHRCONHPh + CH2dC(OH)NHPh h
CH3CONHPh + (MeO)2POCRdC(OH)NHPh (10)
The ∆G values of reaction 10 are -24.5 and -34.3 kcal/
mol for R ) (MeO)2PdO and CO2Me, respectively, i.e.,
enolization of the bis phosphonate derivative is much less
favored than that of the phosphonate ester derivative, in line
with the experimental observations.
From the isodesmic data the difference arises from the
interaction between the CdC bond and the MeOCO (vs
(MeO)2PdO) rather than from the interaction between the Cd
C-OH and the MeOCO (vs (MeO)2PdO) (see eq 8), i.e., the
[(MeO)2PdO]2CdC moiety is less stable than the [(MeO)2Pd
O](CO2Me)CdC moiety.
Note that the increase in Kenol on substituting with fluorine
atoms follows the experimental trend, although the close
similarity between the CF3 and the (CF3)2 derivatives is not
observed in the experiment. Few attempted correlations between
the observed and the calculated pKenol values for enols 7b, 7g,
7l, and 10b were attempted. The correlation between pKenol
-
(calcd) and the observed pKenol is close to half a parabola for
the four available points when either the overall experimental
pKenol(total) () -log[K(E-enol) + K(Z-enol)]) or the observed vs
calculated pK(E-enol) or pK(Z-enol) are used. Attempted correla-
tions of observed pKenol (total) vs calculated pKenol(total) and
observed pK(Z-enol) vs calculated pK(Z-enol) and observed pKenol
(total) vs calculated pK(Z-enol) gave R2 values of 0.639, 0.696,
and 0.754, respectively. In contrast, a linear correlation of the
observed Kenol(total) or the observed K(E-enol) vs the calculated
K(E-enol) value for the three available points are linear with
J. Org. Chem, Vol. 72, No. 20, 2007 7621

Song et al.
TABLE 8. Analytical Data and Melting Points for All Amide/Enol Systems
calcd, %
H
found, %
H
compd
formula
mp, °C
C
N
C
N
6a/7a
C13H18NO7P
108-110
85-8
47.13
47.84
36.83
40.45
42.70
42.11
42.28
33.99
35.82
37.82
38.54
38.44
31.88
32.75
34.53
48.32
49.25
36.87
43.55
40.95
41.03
32.67
34.08
36.26
38.55
38.46
32.77
34.54
35.90
35.66
30.59
32.18
5.44
5.32
2.81
6.74
7.12
4.26
4.07
2.18
5.07
5.44
3.43
3.20
1.71
3.97
4.32
5.03
4.85
2.23
6.85
5.58
5.45
3.20
6.67
7.00
3.45
3.23
4.00
4.35
2.82
2.59
3.21
3.53
4.23
4.65
3.58
5.24
4.98
3.51
3.79
3.05
4.18
4.01
3.00
3.20
2.66
3.47
3.36
9.40
10.45
7.82
11.29
3.67
3.90
3.18
4.42
4.23
3.00
3.20
3.47
3.36
2.62
2.77
2.97
2.89
47.40
47.97
36.76
40.44
42.83
42.45
42.52
34.20
35.65
38.13
38.83
38.76
32.14
33.03
34.73
48.72
49.28
37.03
43.68
41.04
41.17
32.74
34.22
35.98
38.72
38.63
32.89
34.43
35.74
35.81
30.30
31.24
5.44
5.37
2.88
6.69
7.12
4.35
4.34
2.26
5.18
5.58
3.41
3.13
1.92
3.95
4.03
5.13
4.91
2.29
6.81
5.57
5.43
3.01
6.77
7.07
3.33
3.03
3.73
4.12
2.71
2.66
3.03
3.17
4.16
4.44
3.43
5.04
4.92
3.31
3.46
2.88
4.23
3.78
2.94
3.12
2.56
3.44
3.36
9.10
10.20
7.69
11.14
3.65
3.69
2.89
4.40
4.18
2.90
2.94
3.21
3.23
2.56
2.61
2.88
2.27
6b/7b
C12H16NO6P
6c/7c
C12H11F5NO6P
C9H18NO6P
C10H20NO6P
98-101
53-5
6d/7d
6e/7e
73-5
6f/7f
C14H17F3NO7P
C13H15F3NO6P
C13H10F8NO6P
C10H17F3NO6P
C11H19F3NO6P
C15H16F6NO7P
C14H14F6NO6P
C14H9F11NO6P
C11H16F6NO6P
C12H18F6NO6P
C12H15N2O5P
C11H13N2O4P
C11H8F5N2O4P
C9H17N2O4P
C13H21NO8P2
C12H19NO7P2
C12H14F5NO7P2
C9H21NO7P2
C10H23NO7P2
C15H16F6NO7P
C14H14F6NO6P
C11H16F6NO6P
C12H18F6NO6P
C16H15F9NO7P
C15H13F9NO6P
C12H15F9NO6P
C13H17F9NO6P
138-140
61-3
6g/7g
6h/7h
104-7
46-7
6i/7i
6j/7j
60-3
6k/7k
6l/7l
6m/7m
6n/7n
6o/7o
92-4
74-7
104-7
90-2
113-6
155-7
120-2
152-4
164-6
152-4
109-111
112-4
86-8
9a/10a
9b/10b
9c/10c
9d/10d
12a/13a
12b/13b
12c/13c
12d/13d
12e/13e
17a/18a
17b/18b
17c/18c
17d/18d
17e/18e
17f/18f
17g/18g
17h/18h
62-4
96-8
98-100
101-2
96-8
82-4
94-95
69-70
67-9
mL) and kept overnight at 4 °C. The solution was extracted with
EtOAc, washed with brine, and dried over Na2SO4. After evapora-
tion of the solvent the residue was crystallized from EtOAc-
petroleum ether (40-60 °C) to afford 0.25 g (38%) of 6a, 0.30 g
(56%) of 6d, or 0.21 g (37%) of 6e as light yellow or white
powders.
(b) No precipitate was formed for 6b,c in the solution, and the
solution was directly added to ice-cooled 2 N HCl (40 mL), and
then followed procedure (a) above in order to obtain 0.26 g (43%)
of 6b or 0.28 g (36%) of 6c as a white or pale yellow powder.
Analyses and NMR data are given in Table 8 and Tables S1-
S3 in Supporting Information.
effects affect the relative Kenol(OdP(OR′)2)/Kenol(CO2R) ratios
for enol formation.
An important conclusion is that the previously used rule that
the lowest δ(OH) value is always associated with the most
abundant isomer does not always apply. Analysis of the
appropriate hydrogen bonds should replace the formal “rule”.
The present work also demonstrates the ability of inverting
the order of stability of E- and Z-isomers by proper substitution.
They also show that the OdP(OR′)2 substituent is a suitable
activator for enolization since no competing enolization on the
PdO group was found.
2,2,2-Trifluoroethyl Chloroacetate. The compound was pre-
pared by two methods.
Experimental Section
General Methods. Melting points were uncorrected, 1H and 13
C
(a) Chloroacetic acid (9.45 g,100 mmol) was refluxed for 12 h
in benzene (30 mL) containing excess 2,2,2-trifluoroethanol (11 g,
110 mmol) and 98% H2SO4 (6 g) in a 100 mL round flask attached
to a Dean-Stark trap. The solution was washed with saturated Na2-
CO3 (3 × 20 mL) and brine (3 × 10 mL) and dried (Na2SO4).
Distillation of the crude product afforded 2.4 g (13.6%) of the
colorless ester, bp 124-126 °C.
(b) To a solution of 2,2,2-trifluoroethanol (26.4 g, 264 mmol)
in CH2Cl2 (70 mL) was added dropwise triethylamine (24.32 g,
240 mmol) during 15 min, and then chloroacetyl chloride (27.15
g, 240 mmol) in CH2Cl2 (70 mL) was added dropwise during 30
min. The solution was refluxed for 2 h and stirred overnight at rt.
The formed triethylammonium salt was removed by filtration, and
the filtrate was washed successively with 2 N HCl (2 × 50 mL)
and saturated Na2CO3 (2 × 20 mL) and dried (Na2SO4). The solvent
was evaporated, the crude product was distilled and the fraction
with boiling range of 120-126 °C (23 g, 54%) was collected.
1H NMR (CDCl3, 298K) δ: 4.56 (q, 3JHF ) 8.28 Hz, 2H, CH2-
CF3), 4.19 (s, 2H, CH2Cl). The compound is relatively unstable,
and satisfactory elemental analysis could not be obtained.
2,2,2-Trifluoroethyl Phosphonoacetate. A mixture of 2,2,2-
trifluoroethyl chloroacetate (23 g, 130 mmol) and trimethyl
NMR spectra were recorded as described previously.8
Solvents and Materials. CCl4, CDCl3, THF-d8, CD3CN, DMSO-
d6, and the precursors for the synthetic work were purchased from
a commercial supplier and were used without further purification.
Analytical and Spectral Data. Melting points and microanalysis
data are given in Table 8.
Selected 1H, 13C, 19F, and 31P NMR spectra in the various solvents
studied are given in Tables 2-4, and the full data are given in
Tables S1-S3 in the Supporting Information.
Reaction of Dimethyl Phosphonoacetate with Organic Iso-
cyanates XNCO To Form 6a-e/7a-e. Na pieces (48 mg, 2.1
mmol) were added to the solution of dimethyl phosphonoacetate
(364 mg, 2 mmol) in dry THF (5 mL) for 6d,e or in Et2O (20 mL)
for 6a at 0 °C, and the mixture was stirred for ca. 6 h at 0 °C and
for additional 18 h at rt. To this solution the isocyanate (2 mmol)
in THF (1 mL) was added dropwise at 0 °C, and the mixture was
stirred for about 12 h and for additional 12 h at rt. (a) For 6a and
6d-e, the precipitate formed in the reaction solution was filtered,
and washed with little Et2O to give the Na salts of 6a (0.72 g), of
6d (0.45 g) or of 6e (0.32 g). A solution of the salt in DMF (2-4
mL) was added dropwise to ice-cooled 2 N HCl solution (20-40
7622 J. Org. Chem., Vol. 72, No. 20, 2007

Dialkoxyphosphinyl-Substituted Enols of Carboxamides
phosphite (17.8 g, 144 mmol) was heated at ca. 80 °C for 48 h.
The volatile components were evaporated under reduced pressure
and the residue was purified by column chromatography (silica,
EtOAc) giving 15.6 g (48%) of a liquid colorless 2,2,2-trifluoroethyl
liquid 1,1,1,3,3,3-hexafluoro-2-propyl chloroacetate. 1H NMR
(CDCl3, 298 K) δ: 5.70 (hept, 3JHF ) 6.01 Hz, 1H, CH), 3.71 (d,
3JPH ) 11.41 Hz, 6H, CH3O), 3.08 (d, 2JPH ) 22.00 Hz, 2H, CH2).
19F NMR (CDCl3, 298 K) δ: -74.45 (d, 3JHF ) 6.1 Hz, CF3). 31
P
1
3
2
3
phosphonoacetate. H NMR (CDCl3, 298 K), δ: 4.51 (q, JHF
)
NMR (CDCl3, 298 K) δ: 19.00 (t of hept, JPH ) 21.54 Hz, JPH
8.33 Hz, 2H, CH2CF3), 3.80 (d, 3JPH ) 12.26 Hz, 6H, CH3O), 3.59
) 11.58 Hz). 13C NMR (CDCl3, 298 K) δ: 162.8 (d of m, 2JCP
)
2
(d, JPH ) 21.62 Hz, 2H, CH2P). 19F NMR (CDCl3, 298 K) δ:
6.8 Hz, CO), 120.1 (q of m, JCF ) 283.0 Hz, CF3), 66.9 (d of hept,
-74.60 (t, 3JHF ) 7.9 Hz, CF3). 31P NMR (CDCl3, 298 K) δ: 20.59
JCH ) 151.0 Hz, JCF ) 35.1 Hz, CH), 53.1 (q of d, JCH ) 148.8
2
3
2
(t of hept, 2JPH ) 21.61 Hz, JPH ) 11.10 Hz). 13C NMR (CDCl3,
Hz, JCP ) 6.3 Hz, CH3O), 33.3 (t of d, JCH ) 131.8 Hz, JCP
)
298 K) δ: 164.2 (d of m, 2JCP ) 6.1 Hz, CO), 122.6 (q of t, JCF
)
133.7 Hz, CH-PO).
277.1 Hz, 2JCH ) 4.8 Hz, CF3), 61.0 (t of q, JCH ) 151.3 Hz, 2JCF
) 37.0 Hz, CH2CF3), 53.3 (q of d, JCH ) 148.6 Hz, 2JCP ) 6.4 Hz,
CH3O), 32.8 (d of t, JCP ) 135.1 Hz, JCH ) 130. 9 Hz, CH2PO).
Anal. Calcd for C6H10F3O5P: C, 28.80; H, 4.00. Found: C,
28.75; H, 4.08.
Anal. Calcd for C7H9F6O5P: C, 26.42; H, 2.83. Found: C, 26.26;
H, 2.88.
Reaction of 1,1,1,3,3,3-Hexafluoro-2-propyl Phosphonoacetate
with Organic Isocyanates To Form 6k-o. The procedure with
phenyl isocyanate is also representative of the reactions of the p-An,
C6F5, t-Bu, and i-Pr isocyanates.
Reaction of 2,2,2-Trifluoroethyl Phosphonoacetate with Or-
ganic Isocyanates To Give 6f-j. Na pieces (48 mg, 2.1 mmol)
were added to a solution of 2,2,2-trifluoroethyl phosphonoacetate
(500 mg, 2 mmol) in dry THF (4 mL) at 0 °C, and the solution
was stirred for ca. 6 h and for an additional 8 h at rt. To this solution
was added the isocyanate solution (2 mmol) in THF (1 mL)
dropwise at 0 °C, and the mixture was stirred for ca. 18 h and an
additional 6 h at rt. (a) For 6f and 6h, the solution was then poured
into ice-cooled 2 N HCl solution (50 mL) and kept overnight at
4 °C. The precipitate formed was filtered, washed with cold water,
and dried in air giving 0.36 g (45%) of a white powder of 6f or
0.54 g (59%) of a light yellow powder of 6h. A white analytical
sample was crystallized from EtOAc-petroleum ether (40-60 °C).
(b) For 6g and 6i, the solution was poured into ice-cooled 2 N
HCl solution (50 mL) and then extracted with CH2Cl2, washed with
brine, and dried (Na2SO4). Evaporation of the solvent afforded 0.65
g (88%) of a light yellow powder of 6g, or 0.35 g (52%) of a light
yellow oil of 6i, which solidified at -20 °C after standing for a
long time. A colorless analytical sample was crystallized from
EtOAc-petroleum ether (40-60 °C).
Sodium (48 mg, 2.1 mmol) was added to the solution of
1,1,1,3,3,3-hexafluoro-2-propyl phosphonoacetate (636 mg, 2 mmol)
in dry THF (4 mL) at 0 °C, and the mixture was stirred for 12 h.
A solution of phenyl isocyanate (238 mg, 2 mmol) in THF (1 mL)
was added dropwise at 0 °C, and the mixture was stirred for ca. 12
h and additional 12 h at rt. The precipitate was removed by filtration,
and the filtrate was poured into ice-cooled 2 N HCl solution (50
mL) and kept overnight at 4 °C. The precipitate formed was filtered,
washed with cold water and dried in air, giving 0.48 g (55%) of a
white powder of 6l which was crystallized from EtOAc-petroleum
ether (40-60 °C). 6k and 6m-o were obtained similarly and their
yields, analyses and NMR data are given in Table 8 and Tables
S1-S3 in the Supporting Information.
Dimethoxyphosphinylacetonitrile. A mixture of chloroaceto-
nitrile (6.04 g, 80 mmol) and trimethyl phosphite (10.92 g, 88
mmol) was heated at ca. 80 °C for 48 h. The volatile component
was evaporated under reduced pressure and the residue was purified
by column chromatography (silica, EtOAc), giving 3.6 g (30%) of
colorless dimethoxyphosphinylacetonitrile which was solidified after
storing at -20 °C.
(c) For 6j, the precipitate formed in the solution was filtered,
washed with little Et2O to give the Na salt of 6j (0.25 g, 34%).
The salt was dissolved in DMF (1 mL) and added dropwise to an
ice-cooled 2 N HCl solution (10 mL) and kept overnight at 4 °C.
The solution was extracted with EtOAc, washed with brine and
dried (Na2SO4). The solvent was evaporated and the residue was
crystallized from EtOAc-petroleum ether (40-60 °C), giving 0.19
g (27%) of a white powder of 6j.
Anal. Calcd for C4H8NO3P: C, 32.21; H, 5.37; N, 9.40. Found:
C, 32.82; H, 5.54; N, 8.66.
3
1H NMR (CDCl3, 298 K) δ: 3.69(d, JPH ) 11.50 Hz, 6H,
2
CH3O), 2.85(d, JPH ) 21.10 Hz, 2H, CH2); 13C NMR (CDCl3,
2
2
298 K) δ: 112.3 (d of t, JCP ) 11.4 Hz, JCH ) 10.9 Hz, CN),
2
53.5 (q of d, JCH) 149.2 Hz, JCP ) 6.7 Hz, CH3O), 14.7 (d of t,
JCP ) 143.7 Hz, JCH ) 135.6 Hz, CH2); 13P NMR (CDCl3, 298 K)
2
3
Analyses and NMR data are given in Table 8 and Tables S1-
S3 in the Supporting Information.
δ: 17.03 (d of hept, JPH ) 21.62 Hz, JPH ) 11.29 Hz, PO).
Reaction of Dimethoxyphosphinylacetonitrile with Organic
Isocyanates To Form 9a-d. Na (72 mg, 3.1 mmol) was added to
a solution of dimethoxyphosphinylacetonitrile (447 mg, 3 mmol)
in dry THF (5 mL) at 0 °C and the solution was stirred for ca. 6 h
and then for additional 8 h at rt. To this solution the isocyanate
(447 mg, 3 mmol) in THF (1 mL) was added dropwise at 0 °C,
and the mixture was stirred for ca. 12 h and for additional 12 h at
rt. It was then poured into ice-cooled 2 N HCl solution (60 mL)
and kept for a few hours at 4 °C. (a) For 9a and 9c, a precipitate
was formed in the aqueous solution, filtered, washed with cold
water, and dried in air. Crystallization from EtOAc-petroleum ether
(40-60 °C) afforded 0.19 g (21%) of white cotton-like solid 9a or
0.56 g (51.7%) of the yellow solid 9c.
1,1,1,3,3,3-Hexafluoro-2-propyl Chloroacetate. To the solution
of 1,1,1,3,3,3-hexafluoro-2-propanol (24.5 g, 146 mmol) in CH2-
Cl2 (65 mL) at 0 °C was added dropwise triethylamine (14.7 g,
146 mmol) during 15 min. Chloroacetyl chloride (18.3 g, 162 mmol)
in CH2Cl2 (65 mL) was then added dropwise during 1 h, and the
solution was stirred at 0 °C for ca. 6 h and then overnight at rt.
The formed triethylammonium salt was removed by filtration, and
the filtrate was washed successively with 2 N HCl solution (2 ×
30 mL) and saturated Na2CO3 solution (2 × 20 mL) and dried (Na2-
SO4). The solvent was evaporated, the crude product was distilled
and the fraction with boiling range of 104-108 °C (21.6 g, 61%)
was collected.
3
1H NMR (CDCl3, 298 K) δ: 5.78 (hept, JHF ) 5.93 Hz, 1H,
(b) For 9b and 9d, no precipitate was formed in the aqueous
solution, and the solution was extracted with EtOAc and dried (Na2-
SO4), and the solvent was evaporated. The remaining solid was
crystallized from EtOAc-petroleum ether (40-60 °C), giving 0.51
g (63%) of a light yellow solid of 9b, or 0.44 g (58.5%) of a light
yellow powder of 9d. Crystallization was from EtOAc.
CHCF3), 4.24 (s, 2H, CH2Cl). 13C NMR (CDCl3, 298 K) δ: 164.8
1
(m, CO), 120.1 (q of m, JCF ) 282.3 Hz, CF3), 67.6 (d of hept,
2
1
1JCH ) 151.2 Hz, JCF ) 35.1 Hz, CH), 39.5 (t, JCH ) 152.8 Hz,
CH2). The compound is relatively unstable, and satisfactory
elemental analysis could not be obtained.
Analyses and NMR data are given in Table 8 and Tables S1-
1,1,1,3,3,3-Hexafluoro-2-propyl Phosphonoacetate. A mixture
of 1,1,1,3,3,3-hexafluoro-2-propyl chloroacetate (21.6 g, 88 mmol)
and trimethyl phosphite (12.0 g, 97 mmol) was heated at ca. 80 °C
for ca. 48 h. The volatile components were evaporated under
reduced pressure, and the residue was purified by column chro-
matography (silica, EtOAc) giving 9.64 g (34%) of the colorless
S3 in the Supporting Information.
General Procedure for the Reaction of Tetramethyl Meth-
ylenebisphosphonate with Isocyanates to Form 12a-e. Na (72
mg, 3.1 mmol) was added to a solution of tetramethyl methyl-
enebisphosphonate (696 mg, 3 mmol) in dry Et2O (20 mL) at rt
J. Org. Chem, Vol. 72, No. 20, 2007 7623

Song et al.
and stirring continued until its complete disappearance. The organic
isocyanate (3 mmol) in Et2O (3 mL) was added dropwise at 0 °C,
and the mixture was stirred for ca. 12 h at 0 °C and additional 12
h at rt. The solution was filtered and the Na salt was washed quickly
with a small amount of Et2O and suspended in Et2O (10 mL). An
ice-cooled 2 N HCl solution (50 mL) was poured into the
suspension, which was then extracted with chloroform (3 × 20 mL),
dried (MgSO4) and the solvent was evaporated. The crude residue
was crystallized from EtOAc-petroleum ether (40-60 °C), giving
the desired compound (12a-e). Analyses and NMR data are given
in Table 8 and Tables S1-S3 in the Supporting Information.
2,2,2-Trifluoroethyl Bis(2,2,2-trifluoroethyl)phosphonoace-
tate. To a solution of bis(2,2,2-trifluoroethyl) phosphonoacetate (5
g, 15.7 mmol) in 2,2,2-trifluoroethanol (32 g, 314 mmol) was added
98% sulfuric acid (0.8 g), and the solution was refluxed overnight.
After cooling, the solution was added to saturated Na2CO3 solution
(30 mL), and the organic layer at the bottom was separated. The
aqueous layer was extracted twice with chloroform, and the
combined organic layer was dried over anhydrous Na2SO4. Evapo-
ration gave a 1:0.81 mol/mol mixture of the precursor and the
desired esters (based on 1H NMR). The mixture was separated by
column chromatography (silica gel, 2:3 EtOAc/petroleum ether).
The first fraction was collected and evaporation afforded 1.6 g
(26%) of 16 as a colorless oil.
noacetate (386 mg, 1 mmol) in dry Et2O (10 mL) at 0 °C and stirring
continued until all of the Na disappeared. The organic isocyanate
(1 mmol) in Et2O (1 mL) was added dropwise at 0 °C, and the
mixture was stirred for ca. 24 h at 0 °C. The solvent was evaporated
and the remainder (except for 17g) was washed with chloroform
for 17e,f or with benzene for 17h and suspended in chloroform.
Then 32% HCl (10-20 mL) containing ice (20 g) was poured into
the suspension which was extracted with chloroform (3 × 20 mL)
and dried (MgSO4).The residue was crystallized from EtOAc-
petroleum ether (40-60 °C) and gave the desired compound.
Analyses and NMR data are given in Table 8 and Tables S1-S3
in the Supporting Information.
Computation Methods. The compounds investigated were
optimized by using the Gaussian 98 program27 at the hybrid density
functional B3LYP/6-31G** method28 which was extensively used
in recent calculations of enols of carboxylic acid derivatives and
shown to give pKenol values reliable within 2 pKenol units.29 All
optimized structures including higher energy conformers were
confirmed by frequency analysis 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, to Dr. Shmuel Cohen from the Depart-
ment of Inorganic and Analytical Chemistry, The Hebrew
University, for the X-ray structure determination, and to
Professors Tony Kirby, Marvin Charton, and Eli Breuer for
discussions.
Anal. Calcd for C8H8F9O5P: C, 24.89; H, 2.09. Found: C, 25.69;
H, 2.18.
1H NMR (CDCl3, 298 K) δ: 4.53 (q, 3JHF ) 8.25 Hz, 2H, CO2-
2
CH2CF3), 4.45 (m, 4H, CF3CH2OPO), 3.26 (d, JPH ) 21.54 Hz,
2H, CH2). 19F-H coupled NMR (CDCl3, 298 K) δ: -74.75 (t,
3JHF ) 8.09 Hz, 3F, CO2CH2CF3), -76.37 (t, 3JHF ) 7.91 Hz, CF3-
Supporting Information Available: Tables S1-S3 with com-
plete 1H, 13C, 19F, and 31P NMR data, Table S4 with relative
integration of signals for all compounds, and Table S5 for the
calculated geometries of enols and amides. Full crystallographic
data for the compounds in Table 1 and compounds 12d, 17b, 19,
and 20 (CIF). This material is available free of charge via the
Internet at http://pubs.acs.org.
2
CH2OPO). 13C NMR (CDCl3, 298 K) δ: 163.2 (dm, JCP ) 4.8
2
Hz, CO), 122.4 + 122.3 + 122.2 (q of t, JCF ) 277.2 Hz, JCH
)
)
2
2
4.6 Hz, CF3), 62.7 (tqd, JCH ) 152.6 Hz, JCF ) 38.4 Hz, JCP
2
5.6 Hz, CF3CH2OPO), 61.3 (t of q, JCH ) 151.8 Hz, JCF ) 37.3
Hz, CO2CH2CF3), 33.4 (d of t, JCP ) 144.5 Hz, JCH ) 131.5 Hz,
CH2PO).
General Procedure for the Reaction of Bis(2,2,2-trifluoroet-
hyl) Methoxycarbonylmethyl Phosphonate with Isocyanates to
Give 17a-d. Na (48 mg, 2.1 mmol) was added to a solution of
bis(2,2,2-trifluoroethyl) methoxycarbonylmethyl phosphonate (636
mg, 2 mmol) in dry Et2O (20 mL) at rt with stirring which continued
until complete disappearance of the Na. The organic isocyanate (2
mmol) in Et2O (2 mL) was added dropwise at 0 °C, and the mixture
was stirred for ca. 12 h at 0 °C and additional 12 h at rt. Ice-
cooled 2 N HCl solution (40 mL) was poured into the solution,
which was then extracted with chloroform (3 × 20 mL) and dried
(MgSO4), and the solvent was evaporated. The crude residue was
crystallized with EtOAc-petroleum ether (40-60 °C) giving 17b-
d. The product 17a of p-methoxyphenyl isocyanate was crystallized
from CCl4-petroleum ether. Analyses and NMR data are given in
Table 8 and Tables S1-S3 in the Supporting Information.
General Procedure for the Reaction of 2,2,2-Trifluoroethyl
Bis(2,2,2-trifluoroethyl) Phosphonoacetate with Isocyanates To
Give 17e-h. Na (25 mg, 1.1 mmol) was added with stirring to a
solution of 2,2,2-trifluoroethyl bis(2,2,2-trifluoroethyl)phospho-
JO0710679
(27) Gaussian 98, Revision A.6: 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,
Inc., Pittsburgh, PA, 1998.
(28) (a) Becke, A. D. Phys. ReV. 1988, A38, 3098. (b) Lee, C.; Yang,
W.; Parr, R. G. Phys. ReV. 1988, B37, 785.
(29) Apeloig, Y.; Arad, D.; Rappoport, Z. J. Am. Chem. Soc. 1990, 112,
9131. Sklenak, S.; Apeloig, Y.; Rappoport, Z. J. Chem. Soc., Perkin Trans.
2 2000, 2269. Yamataka, H.; Rappoport, Z. J. Am. Chem. Soc. 2000, 122,
9818. Rappoport, Z.; Lei, Y. X.; Yamataka, H. HelV. Chem. Acta 2001,
84, 1401.
7624 J. Org. Chem., Vol. 72, No. 20, 2007