A stable enol in small ring systems: Clear differentiation between penta- and tri-valency of phosphorus atoms

Article abstract of DOI:10.1039/b809877h

The first stable enols in 1,2-dihydrophosphetes 6 and 10 were synthesized and structurally characterized with intermolecular hydrogen bonds to phosphoryl groups in 10-membered dimeric structures; in contrast, trivalent analogue 9 exists in keto-form, where such stabilization by hydrogen bonds is not feasible. The Royal Society of Chemistry.

Full text of DOI:10.1039/b809877h

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
A stable enol in small ring systems: clear differentiation between
penta- and tri-valency of phosphorus atomsw
Alex S. Ionkin,* William J. Marshall, Brian M. Fish, Matthew F. Schiffhauer and
Charles N. McEwen
Received (in Cambridge, UK) 11th June 2008, Accepted 15th August 2008
First published as an Advance Article on the web 22nd September 2008
DOI: 10.1039/b809877h
The first stable enols in 1,2-dihydrophosphetes 6 and 10 were
synthesized and structurally characterized with intermolecular
hydrogen bonds to phosphoryl groups in 10-membered dimeric
structures; in contrast, trivalent analogue 9 exists in keto-form,
where such stabilization by hydrogen bonds is not feasible.
tion of sterical hindrance, affording cisoidal orientation of the
tert-butyl and trimethylsilyl groups along the P–C bond in
3 (Scheme 1).
Surprisingly X-ray analysisz reveals that 6 exists in a stable
enolic form. The enol moiety is stabilized by intermolecular
hydrogen bonds to phosphoryl groups, affording a ten-mem-
bered dimer in the solid state (Fig. 1). According to the
Enols of carbonyl compounds have an important and
historical role in organic chemistry and especially in biochemi-
1
cal processes. Since Fuson et al.’s original discovery of
1
3
C NMR spectrum of 6, it also exists in an enol form in
solution. There are two down-field doublets at 134.30 ppm
2
1
sterically hindered stable enols, particular attention has been
2
paid to the search for enols in strained carbonyl molecules. In
with JPC = 14.1 Hz, and at 154.60 ppm with JPC = 80.9 Hz
corresponding to the enol form of 6.
this communication, we describe a new enolization in sterically
hindered four-membered carbonyl compounds that is
promoted by phosphorus atoms. Enolization enthalpy of the
parent cyclobutanone in the gas phase was found to be
1,2-Dihydrophosphetes were reported to undergo retro-
cyclo-opening, and can react as masked 1-phosphabuta-
8
dienes. Thus, we carried out a flash-pyrolysis of 3 in an
attempt to generate 5. Unexpectedly, the product of ring
expansion, 7, and tris(trimethylsilyl)phosphine 8 were isolated
as distillable components. Chromatography of the residue
gave 1,2-dihydrophosphete 9, which contains a trivalent phos-
phorus atom. According to X-ray analysis,z compound 9
exists in the keto-form. Selective oxidation of 9 by pyridine
N-oxide resulted in penta-valent phosphorus derivative 10. In
contrast to 9, X-ray analysisz of 10 shows that it exists in the
enol form (Fig. 2).
ꢀ
1
1
6.0 kcal mol , which points to the difficulty of forming
3
unsaturated cyclic enols in all-carbon four-membered rings.
The coupling reaction between lithium bis(trimethylsilyl)-
phosphide (1) and Z-2-tert-butyl-4,4-dimethyl-pent-2-enoyl
chloride (2) led to the formation of 2,3-di-tert-butyl-1-tri-
methylsilanyl-4-trimethylsilanyloxy-1,2-dihydrophosphete (3)
stereoselectively. The presumptive mechanism of formation
of 3 involves a 1,3-shift of a trimethylsilyl group from
phosphorus to oxygen in intermediate 4, affording transient
Solid state structures of 9 and 10 are consistent with
1
structures observed in solution. The C NMR spectrum of
3
1
-phosphabutadiene (5), which then undergoes electrocyclic
ring closure affording 3 (Scheme 1).
-Phosphabutadienes are among the least stable phos-
9 has two doublets of carbonyl groups at 211.10 ppm with
1
phabutadienes, although a few stable examples have been
4
described. Generated in situ, 1-phosphabutadienes undergo
4
b
intermolecular [2 + 2]- and [4 + 2]-cycloadditions. The ring
closure paths (disrotatory or conrotatory) in the parent buta-
diene have great importance to organic chemistry and
5
numerous studies have been devoted to this field. So far,
6
there are few theoretical studies of 1,2-dihydrophosphetes.
Compound 3 is a liquid. In order to establish the orientation in
this four-membered ring, a sample of 3 was exposed to air to
obtain a more crystalline substance. Hydrolysis and oxidation
of 3 to 6 should take place without loss of stereochemistry
(
7
Scheme 2). The ring closure in 5 took place with minimiza-
DuPont Central Research & Development, Experimental Station,
Wilmington, Delaware 19880-500, USA.
E-mail: alex.s.ionkin@usa.dupont.com; Fax: (+1) 302-695-1672
w Electronic supplementary information (ESI) available: Crystallo-
graphic cif-files of 6, 7, 9, 10, and synthesis of compounds 3, 6, 7, 8,
9, and 10. CCDC 691153–691156. For ESI and crystallographic data
in CIF or other electronic format, see DOI: 10.1039/b809877h
Scheme 1
5
432 | Chem. Commun., 2008, 5432–5434
This journal is ꢁc The Royal Society of Chemistry 2008

View Article Online
Fig. 2 Stabilization of enol 10 through PQOꢂ ꢂ ꢂH bonds. Struc-
tural data of the independent molecule on the right are listed in
Table 1.
˚
0
.479 A for 10. C1–P–C3 internal angles are between 75.151
and 78.031, which are close to the theoretically calculated
8
number of 741 for such heterocycles. Enol derivatives 6 and
˚
0 have single C–O bond lengths at 1.3401(10) A and
1
˚
.344(6) A. Ketone derivative 9 clearly has a double bond
1
between oxygen and carbon at 1.1951(19) A. Enols 6 and 10
˚
˚
have double C–C bonds at 1.3624(10) A and 1.361(7) A.
˚
Scheme 2
˚
Ketone 9 has only single C–C bonds in the ring at 1.531(2) A
˚
and 1.563(2) A. Intermolecular PQOꢂ ꢂ ꢂO distances in the
˚
enol derivatives are 2.590(3) A for 6 and 2.65(1) A for 10,
˚
1
1
1
J
PC = 29.5 Hz, and at 211.50 ppm with JPC = 76.3 Hz. The
3
which are typical distances for hydrogen bonds, consistent
˚
with the average value of 2.68 A in the CCDC database
C NMR spectrum of 10 has one doublet of a carbonyl group
1 31
at 218.00 ppm with J = 53.9 Hz. The P NMR spectra
PC
(
Table 1).
of 1,2-dihydrophosphetes are influenced by the type of
substituent at the phosphorus atom, not by ring-size. Com-
pounds 6 and 10 contain a phosphoryl group and have typical
˚
Table 1 Selected bond lengths (A) and angles (1) for 6, 9 and 10
1
chemical shifts for phosphine oxides at 4.17 ppm (d, JPH
=
95.6 Hz) and 17.9 ppm correspondingly. Trivalent compound
has a chemical shift of 104.00 ppm, which is often associated
6
9
10
4
9
P–C1
C1–C2
C2–C3
C3–P
1.7741(8) 1.8982(17) 1.774(5)
1.3624(10) 1.531(2)
1.5441(11) 1.563(2)
1.8379(8) 1.8794(16) 1.851(5)
1.3401(10) 1.1951(19) 1.344(6)
with acylphosphines. Trivalent compound 3, with a direct
bond between phosphorus and silicon, has an upfield signal
at ꢀ40.16 ppm, attributed to mono-silylphosphines.
1.361(7)
1.524(7)
C1–O
P–O
Pꢂ ꢂ ꢂC2
The 1,2-dihydrophosphete rings in 6, 9 and 10 are nearly
planar, with deviation of the phosphorus atom from the
1.4966(7) N/A
2.2862(8) 2.526
1.490(4)
2.282(5)
Transannular interaction
C1–P–C3
P–C1–C2
C1–C2–C3
C2–C3–P
PQOꢂ ꢂ ꢂO–C
˚
C–C–C plane at 0.365 A for 6, 0.156 A for 9, and 0.501 and
˚
78.03(4) 75.15(7)
92.65(5) 94.28(10) 92.5(4)
102.81(6) 96.26(12) 102.5(4)
84.59(5) 93.98(9)
2.590(3) N/A
76.8(2)
84.5(3)
2.64(1) and 2.65(1)
The formation of intramolecular hydrogen bonds to nitrogen
in Mannich/Schiffs bases and to phosphoryl/thiophosphoryl
groups in five-membered rings are known to increase the
9
enol form in keto–enol tautomerism. In contrast, C–O–Hꢂ ꢂ ꢂP
hydrogen bonds to trivalent phosphorus are weak, although
not unprecedented, and have been described in a few cases
1
0
˚
3.03–3.50 A). In our case the strong intermolecular hydrogen
(
bonding in the sequences C–O–Hꢂ ꢂ ꢂOQP of 6 and 10 is
responsible for stabilizing the enol forms, thus revealing the
dramatic differences between the effects of pentavalent and
trivalent phosphorus atoms in four-membered rings.
Fig. 1 Stabilization of enol 6 through PQOꢂ ꢂ ꢂH bonds.
This journal is ꢁc The Royal Society of Chemistry 2008
Chem. Commun., 2008, 5432–5434 | 5433

View Article Online
Notes and references
1 (a) S. V. Ley, Pure Appl. Chem., 2005, 77, 1115; (b) A. J. M. Carpy,
P. P. Haasbroek and D. W. Oliver, Med. Chem. Res., 2004, 13, 565;
(
c) G. S. Coumbarides, J. Eames, S. Ghilagaber, N. Weerasooriya
z Crystal data for all compounds were collected using a Bruker ApexII
CCD system using Mo radiation equipped with a graphite monochro-
mator and low temperature apparatus. The structures were solved and
refined using the Shelxtl suite of programs.
Crystal data for 6: C11 P, M = 216.25, triclinic, a =
and Y. Yohannes, Recent Res. Dev. Org. Bioorg. Chem., 2002, 5,
117; (d) Z. Rappoport, J. Frey, M. Sigalov and E. Rochlin, Pure
Appl. Chem., 1997, 69, 1933; (e) Z. Rappoport and S. E. Biali, Acc.
Chem. Res., 1988, 21, 442; (f) H. Hart, Chem. Rev., 1979, 79, 515;
(g) A. J. Kresge, Chem. Soc. Rev., 1996, 25, 275; (h) A. J. Kresge,
Acc. Chem. Res., 1990, 23, 43.
2 (a) R. C. Fuson, J. Corse and C. H. McKeever, J. Am. Chem. Soc.,
1940, 62, 3250; (b) R. C. Fuson and C. A. Sperati, J. Am. Chem.
Soc., 1941, 63, 2643.
3 (a) X.-M. Zhang, D. Malick and G. A. Petersson, J. Org. Chem.,
1998, 63, 5314; (b) J. Harnisch, J. Am. Chem. Soc., 1979, 101, 3370.
4 (a) R. Appel, B. Niemann, W. Schuhn and N. Siabalis, J. Orga-
nomet. Chem., 1988, 347, 299; (b) R. Appel, H. Kunze and F.
Knoch, Chem. Ber., 1984, 117, 3151; (c) B. A. Boyd, R. J. Thoma,
W. H. Watson and R. H. Neilson, Organometallics, 1988, 7, 572.
5 M. Aoyagi and Y. Osamura, J. Am. Chem. Soc., 1989, 111, 470.
6 S. M. Bachrach and M. Liu, J. Org. Chem., 1992, 57, 209.
7 A. Marinetti and D. Carmichael, Chem. Rev., 2002, 102, 201.
11
21 2
H O
˚
˚
˚
6
.1970(4) A, b = 8.5935(5) A, c = 11.9624(7) A, a = 79.0472(10)1,
3
˚
b = 85.1087(11)1, g = 80.2040(10)1, V = 615.40(6) A , T = ꢀ100 1C,
ꢀ
space group = P1, Z = 2, 26 437 reflections measured, 3801 unique
int = 0.025) which were used in all calculations, R indices(all data)
R1 = 0.032, wR2 = 0.086.
Crystal data for 7: C28
(
R
H
56
O
2
P
2
Si
2
, M = 542.85, trigonal, a =
3
˚
˚
˚
1
7.1720(18) A, c = 60.643(7) A, V = 15 486(3) A , T = ꢀ100 1C,
ꢀ
space group R3c, Z = 18, 55 549 reflections measured, 3863 unique
int = 0.048) which were used in all calculations, R indices(all data)
R1 = 0.062, wR2 = 0.142.
Crystal data for 9: C22
(
R
39 2
H O P, M = 366.50, orthorhombic, a =
3
˚
˚
˚
˚
1
1.3952(6) A, b = 11.9039(6) A, c = 33.5993(17) A, V = 4557.7(4) A ,
T = ꢀ100 1C, space group Pbca, Z = 8, 78 472 reflections measured,
626 unique (Rint = 0.042) which were used in all calculations,
R indices(all data) R1 = 0.064, wR2 = 0.143.
Crystal data for 10: C22 P, M = 382.50, orthorhombic, a =
5
8
(a) N. H. Tran Huy and F. Mathey, Tetrahedron Lett., 1988, 29,
077; (b) K. M. Doxsee and G. S. Shen, J. Am. Chem. Soc., 1989,
111, 9129.
3
39 3
H O
3
˚
˚
˚
˚
4
4.018(15) A, b = 9.170(3) A, c = 23.309(8) A, V = 9409(5) A , T =
ꢀ
100 1C, space group Pbcn, Z = 16, 61 459 reflections measured, 7841
9 (a) A. Koll, A. Karpfen and P. Wolschann, J. Mol. Struct., 2006,
790, 55; (b) I. L. Odinets, Ya. A. Vereshchagina, O. I. Artyushin,
R. M. Kalyanova, T. A. Mastryukova, E. A. Ishmaeva, G. R.
Fattakhova, D. V. Chachkov and E. G. Yarkova, Russ. Chem.
Bull., 2003, 52, 638; (c) Y. X. Lei, D. Casarini, G. Cerioni and Z.
Rappoport, J. Org. Chem., 2003, 68, 947; (d) L. D. Quin and J. A.
Caputo, Chem. Commun. (London), 1968, 1463; (e) F. Mathey, G.
Muller and H. Bonnard, Bull. Soc. Chim. Fr., 1972, 10, 4021.
unique (Rint = 0.22) which were used in all calculations, R indices (all
data) R1 = 0.173, wR2 = 0.235. The overall quality of the structure
was limited by the thin data crystal (B0.58 ꢃ 0.24 ꢃ 0.03 mm) which
gave a weak smeared diffraction pattern especially at high angles. The
structure also suffers from disorder in the ethylene groups although
they were well resolved using a system of restraints for the locations of
ethylene and tert-butyl atoms (SADI) as well as thermal parameters
ˇ
10 P. Stepnicka and I. Cisarova, New J. Chem., 2002, 26, 1389.
ˇ ´
ˇ
(EADP). Other thermal parameter restraints (SIMU, ISOR, DELU)
11 G. Sheldrick, Shelxtl Software Suite, Version 5.1, Bruker AXS
Corp., Madison, Wisconsin, 1996.
12 This is DuPont contribution # 8869.
were tried but gave unreasonable thermal parameters. Because of the
disorder, bond lengths around the ethylene groups should be inter-
preted cautiously.
1
2
5
434 | Chem. Commun., 2008, 5432–5434
This journal is ꢁc The Royal Society of Chemistry 2008