Diastereoselective cobalt-mediated [2 + 2 + 2] cycloadditions of substituted linear enediynes phosphine oxides: Scope and limitations

Article abstract of DOI:10.1021/jo026212r

Variously substituted linear enediynes phosphines oxides possessing the double bond at either the terminal or internal position and with the phosphine oxide appended onto the alkyne or the alkene terminus have been prepared. Their cobalt(I)-mediated cyclizations produce the η⁴-complexed tricyclic compounds in high yields. The endo/exo selectivity depends on both the position of the phosphine oxide on the enediyne and the position of the double bond in the tether. With chiral phosphine oxides, a certain degree of induction was observed, and depending on the substituents on the phosphorus atom, the diastereoselectivity can reach 74%. Up to now, it is the highest level reported for such a cyclization in which a stereogenic center is created. Regarding all of our results, two reaction pathways involving an initial coordination of the cobalt moiety on the chelating site of the substituent have been suggested to explain the observed selectivities.

Full text of DOI:10.1021/jo026212r

Dia ster eoselective Coba lt-Med ia ted [2 + 2 + 2] Cycloa d d ition s of
Su bstitu ted Lin ea r En ed iyn es P h osp h in e Oxid es: Scop e a n d
Lim ita tion s
Franck Slowinski, Corinne Aubert,* and Max Malacria*
Universite´ Pierre et Marie Curie (Paris 6), Laboratoire de Chimie Organique de Synthe`se associe´ au
CNRS, Tour 44-54, 2e`me e´tage, Case 229, 4 Place J ussieu 75252 Paris Cedex 05
malacria@ccr.jussieu.fr
Received J uly 18, 2002
Variously substituted linear enediynes phosphines oxides possessing the double bond at either the
terminal or internal position and with the phosphine oxide appended onto the alkyne or the alkene
terminus have been prepared. Their cobalt(I)-mediated cyclizations produce the η4-complexed
tricyclic compounds in high yields. The endo/exo selectivity depends on both the position of the
phosphine oxide on the enediyne and the position of the double bond in the tether. With chiral
phosphine oxides, a certain degree of induction was observed, and depending on the substituents
on the phosphorus atom, the diastereoselectivity can reach 74%. Up to now, it is the highest level
reported for such a cyclization in which a stereogenic center is created. Regarding all of our results,
two reaction pathways involving an initial coordination of the cobalt moiety on the chelating site
of the substituent have been suggested to explain the observed selectivities.
In tr od u ction
dicarbonyl, [CpCo(CO)2], catalysis for the cyclotrimer-
ization of alkynes, and the [2 + 2 + 2] cyclization of
various polyunsaturated systems.
The [2 + 2 + 2] cycloaddition reaction of three
unsaturated partners is nowadays one of the most
powerful synthetic tools for forming several carbon-
carbon bonds in a single chemical transformation.1
Particularly, cobalt(I)-mediated intramolecular cycliza-
tions allow the formation of polycyclic compounds, in a
highly regio-, chemo-, and stereoselective fashion, start-
ing from acyclic polyunsaturated precursors. A large
range of syntheses of natural and unnatural compounds
combined with their biological and/or theoretical interest
have been proposed by Vollhardt,2 ourselves,3 and others4
who successfully developed the cyclopentadienylcobalt
The ultimate development of this reaction would be its
asymmetric version, especially for the selective formation
of stereogenic centers in the construction of polycyclic
compounds. To our knowledge, there are few examples
of asymmetric [2 + 2 + 2] cycloaddition reactions.
Vollhardt reported one example of asymmetric [2 + 2 +
2] cyclization promoted by chiral cyclopentadienylcobalt
complexes, however, with quite low diastereomeric ex-
cesses (16-28%)5 and also some cyclizations of enediynes
possessing a racemic stereogenic center in the tether
between both unsaturated moieties.6 Besides, Mori pub-
lished the use of Ni(COD)2 and chiral ligands to realize
asymmetric [2 + 2 + 2] cyclizations.7 This cycloaddition
involves a facial stereodifferentiation between two enan-
tiotopic groups and the selective formation of a nickela-
cyclopentadiene, which produces isoindolines and iso-
quinolines having benzylic chiral centers; the enantiomeric
(1) (a) Schore, N. E. In Comprehensive Organic Synthesis; Trost, B.
M., Fleming, I., Paquette, L. A., Eds.; Pergamon Press: Oxford, 1991;
Vol. 5, p 1129-1162. (b) Grotjahn, D. B. In Comprehensive Organo-
metallic Chemistry II; Abel, E. W., Stone, F. G. A., Wilkinson, G.,
Hegedus, L., Eds.; Pergamon Press: Oxford, 1995; Vol. 12, p 741-
770. (c) Malacria, M.; Aubert, C.; Renaud, J . L. In Science of
Synthesis: Houben-Weyl Methods of Molecular Transformations; Lau-
tens, M., Trost, B. M., Eds.; Georg Thieme Verlag: Stuttgart, 2001;
Vol. 1, pp 439-530.
(2) (a) Chang, C. A.; Francisco, C. G.; Gadek, T. R.; King, J . J . A.;
Sternberg, E. D.; Vollhardt, K. P. C. In Organic Synthesis Today and
Tomorrow; Trost, B. M., Hutchinson, C. R., Eds.; Pergamon Press:
Oxford, 1981; pp 71-83. (b) Vollhardt, K. P. C. Angew. Chem., Int.
Ed. Engl. 1984, 23, 536-556. (c) Vollhardt, K. P. C. Pure Appl. Chem.
1985, 57, 1819-1826. (d) Sternberg, E. D.; Vollhardt, K. P. C. J . Org.
Chem. 1984, 49, 1564-1573. (e) Gadek, T. R.; Vollhardt, K. P. C.
Angew. Chem., Int. Ed. Engl. 1981, 20, 802-803. (f) Malacria, M.;
Vollhardt, K. P. C. J . Org. Chem. 1984, 49, 5010-5012. (g) J ohnson,
E. J .; Vollhardt, K. P. C. J . Am. Chem. Soc. 1991, 113, 381-382. (h)
Germanas, J .; Aubert, C.; Vollhardt, K. P. C. J . Am. Chem. Soc. 1991,
113, 4006-4008. (i) Eichberg, M. J .; Dorta, R. L.; Grotjahn, D. B.;
Lamottke, K.; Schmidt, M.; Vollhardt, K. P. C. J . Am. Chem. Soc. 2001,
123, 9324-9337. (j) Vollhardt, K. P. C. Pure Appl. Chem. 1993, 65,
153-156. (k) Eickmeier, C.; Holmes, D.; J unga, H.; Matzger, A. J .;
Scherhag, F.; Shim, M.; Vollhardt, K. P. C. Angew. Chem., Int. Ed.
1999, 38, 800-804 and references therein.
(3) (a) Aubert, C.; Buisine, O.; Petit, M.; Slowinski, F.; Malacria,
M. Pure Appl. Chem. 1999, 71, 1463-1470 and all references therein.
(b) Llerena, D.; Buisine, O.; Aubert, C.; Malacria, M. Tetrahedron 1998,
54, 9373-9392. (c) Slowinski, F.; Aubert, C.; Malacria, M. Eur. J . Org.
Chem. 2001, 3491-3500.
(4) (a) Stara´, I. G.; Stary, I.; Kolla´rovic, A.; Teply, F.; Saman, D.;
Tichy, M. J . Org. Chem. 1998, 63, 4046-4050. (b) Sigman, M. S.;
Fatland, A. W.; Eaton, B. E. J . Am. Chem. Soc. 1998, 120, 5130-5131.
(c) Varela, J . A.; Castedo, L.; Maestro, M.; Mah´ıa, J .; Saa´, C. Chem.
Eur. J . 2001, 7, 5203-5213.
(5) (a) Halterman, R. L.; Vollhardt, K. P. C. Tetrahedron Lett. 1986,
27, 1461-1464. (b) Halterman, R. L.; Vollhardt, K. P. C. Organome-
tallics 1988, 7, 883-892.
(6) Clinet, J . C.; Dun˜ach, E.; Vollhardt, K. P. C. J . Am. Chem. Soc.
1983, 105, 6710-6712.
(7) (a) Sato, Y.; Nishimata, T.; Mori, M. J . Org. Chem. 1994, 59,
6133-6135. (b) Sato, Y.; Nishimata, T.; Mori, M. Heterocycles 1997,
44, 443-457.
10.1021/jo026212r CCC: $25.00 © 2003 American Chemical Society
Published on Web 12/04/2002
378
J . Org. Chem. 2003, 68, 378-386

Substituted Linear Enediynes Phosphine Oxides
SCHEME 1. P r ep a r a tion of En ed iyn es 2a a n d 2b a
excesses reached 73%. Another example concerns the
enantioselective synthesis of optically active helicen
compounds and also involved a nickel(0)-mediated in-
tramolecular [2 + 2 + 2] cocyclization of triynes in the
presence of chiral phosphine ligands.4a,8
For several years, we have been concerned in our
laboratory with the developments of the cobalt version
of this reaction, particularly the finding of new unsatur-
ated partners such as allenes3b,9 and also the utilization
of this cyclization in the synthesis of polycyclic natural
products.3a Recently, we turned our attention to the
asymmetric version of this cyclization based on the use
of chiral linear enediynes. Our initial efforts were focused
on the identification of the most appropriate substituent
and its most judicious position on the polyunsaturated
precursors.10 We showed that the level of diasteroselec-
tivity of the [2 + 2 + 2] cyclization of linear enediynes
could be improved over those reported in the literature
by introducing substituents such as ester, sulfoxide, or
phosphine oxide groups at the terminal positions of either
the triple or the double bond of the enediynes. The ester
and phosphine oxide groups emerged as the most promis-
ing substituents for the asymmetric study in terms of
stability of the complexed cycloadducts and yields of the
cyclization. However, using chiral esters allowed the level
of the induction to remain quite low.3c On the contrary,
in our first attempts with chiral phosphine oxide groups,
we observed, by a proper choice of the substituents on
the phosphorus atom, a highly stereoselective induction.11
Although many syntheses of optically active phosphine
oxides have been proposed for their use as chiral ligands
in organometallic catalysts, only a few examples of
asymmetric induction have been reported with their use
as chiral auxiliaries. They were involved in 1,3-dipolar
cycloadditions,12 Diels-Alder13 and Pauson-Khand13d
reactions, and conjugate additions.
a
Conditions: (a) n-BuLi, -78 °C, THF. (b) ClP(O)Ph2, from -78
°C to room temperature.
SCHEME 2. P r ep a r a tion of P r ecu r sor 2ca
a
Conditions: (a) (i) n-BuLi, -78 °C, THF; (ii) Me3SiCl, from
-78 °C to room temperature, yield of 6 ) 98%. (b) cat. PTSA,
MeOH, 7: 97%. (c) (COCl)2, DMSO, Et3N, from -78 °C to room
temperature, CH2Cl2, yield of 8 ) quant. (d) CBr4, PPh3, CH2Cl2,
yield of 9 ) 96%. (e) n-BuLi, -78 °C, THF, yield of 10 ) quant. (f)
(i) n-BuLi, -78 °C, THF; (ii) ClP(O)Ph2, from -78 °C to room
temperature, yield of 11 ) 61%. (g) MeMgBr, CuCl, Et2O, rt, yield
of (Z)-12 ) 32%, yield of (E)-12 ) 48%. (h) KF, DMSO, rt, yield of
(Z)-2c ) 95%, yield of (E)-2c ) 99%.
We report herein the full details of the preparation of
phosphine oxide-substituted linear enediynes and the
scope and limitations of their cyclization, particularly our
investigations aimed at the improvement of the stereo-
selective induction.
in Scheme 1 from the tridec-1-ene-6,12-diyne 1a and
tetradec-1-ene-7,13-diyne 1b.2d Deprotonation of the lat-
ter with n-BuLi, followed by addition of diphenylphos-
phinyl chloride, provided 2a and 2b in 55 and 61% yields.
With the same reaction, tetradec-7-ene-1,13-diyne3c 3 led
to 4a in 51% yield.
The precursors (E)-2c and (Z)-2c were prepared as
presented in Scheme 2 from the previously described
2-trideca-6,12-diynyloxy-tetrahydropyran 5.3c The se-
quence (silylation of the triple bond, cleavage of the THP
ether, and Swern oxidation) furnished aldehyde 8 in 95%
overall yield. Then, chain extension14 afforded triyne 10
in 98% yield. After introduction of the diphenylphosphine
oxide substituent on the terminal triple bond, the 1,4-
addition of methyl Grignard in the presence of cuprous
chloride15 provided 12 in 49% yield as a 3/2 separable
mixture of E/Z stereoisomers. By analogy to the litera-
ture,16 the stereochemistry of the double bond was
determined by 13C NMR on the basis of the C-P coupling
constants of the vinylic methyl. Other 1,4-addition pro-
Resu lts a n d Discu ssion
Various enediynes possessing double bonds at either
the terminal or internal position in the chain were
prepared. The phosphine oxide substituents were in all
cases appended onto the alkyne or the alkene terminus.
Syn t h esis of Su b st it u t ed E n ed iyn es P h osp h in e
Oxid es. Enediynes 2a and 2b were prepared as outlined
(8) Stara´, I. G.; Stary, I.; Kollarovic, A.; Teply, F.; Vyskocil, S.;
Saman, D. Tetrahedron Lett. 1999, 40, 1993-1996.
(9) Buisine, O.; Aubert, C.; Malacria, M. Synthesis 2000, 985-989.
(10) Slowinski, F.; Aubert, C.; Malacria, M. Tetrahedron Lett. 1999,
40, 707-710.
(11) Slowinski, F.; Aubert, C.; Malacria, M. Tetrahedron Lett. 1999,
40, 5849-5852.
(12) (a)Brandi, A.; Cannavo, P.; Pietrusiewicz, K. M.; Zablocka, M.;
Wieczorek, M. J . Org. Chem. 1989, 54, 3073-3077. (b) Brandi, A.;
Cicchi, S.; Goti, A.; Pietrusiewicz, K. M.; Zablocka, M.; Wisniewski,
W. J . Org. Chem. 1991, 56, 4383-4388. (c) Brandi, A.; Cicchi, S.; Goti,
A.; Koprowski, M.; Pietrusiewicz, K. M. J . Org. Chem. 1994, 59, 1315-
1318 and references therein.
(13) (a) Pietrusiewicz, K. M.; Zablocka, M.; Monkiewicz, J . J . Org.
Chem. 1984, 49, 1522-1526. (b) Maffei, M.; Buono, G. New. J . Chem.
1988, 12, 923-930. (c) Mutti, S. Ph.D. Thesis, P. M. Curie University:
Paris, 1991. (d) Sabelle, S. Ph.D. Thesis, P. M. Curie (Paris 6): Paris,
1996.
(14) Corey, E. J .; Fuchs, P. L. Tetrahedron Lett. 1972, 3769-3772.
(15) Aguiar, A. M.; Irelan, J . R. S. J . Org. Chem. 1969, 34, 4030-
4031.
(16) Savignac, P.; Teulade, M. P.; Aboujaoude, E. E.; Collignon, N.
Synth. Commun. 1987, 17, 1559-1568.
J . Org. Chem, Vol. 68, No. 2, 2003 379

Slowinski et al.
SCHEME 5. P r ep a r a tion of En ed iyn e 16
SCHEME 3. P r ep a r a tion of P h osp h in es Oxid es
13a ,ba
SCHEME 6. P r ep a r a tion of th e Ch ir a l-Su bstitu ted
En ed iyn e Th iop h osp h in e Oxid e 15h
a
Conditions: (a) P(O)Cl3, NEt3, toluene, ∆. (b) (i) t-BuCl, AlCl3,
CH2Cl2; (ii) HCl, H2O. (c) MeOH, NaH, THF, rt.
TABLE 1. Coba lt(I)-Med ia ted Cycliza tion of En ed iyn es
2a -f
SCHEME 4. P r ep a r a tion of Ch ir a l-Su bstitu ted
En ed iyn e P h osp h in es Oxid es 15a -g
yield (endo/exo)e
SM
n
methoda
R1
R2
R3
17a -f
2a
2b
3
4
4
3
A
A
B
A
B
B
B
P(O)Ph2
P(O)Ph2
H
SiMe3
SiMe3
SiMe3
H
H
H
H
H
88% (37/63)
96% (25/75)
32% (100/0)
85% (0/100)
85% (46/54)
92% (50/50)
76% (33/67)
2cb
2d c
2d d
2ed
2fd
P(O)Ph2 Me
H
H
H
H
H
H
H
H
4
4
a
Method A: CpCo(CO)2 (1 equiv), toluene, hν, ∆. Method B:
CpCo(CO)2 (1 equiv), decane, ∆. (E)-2c. c From ref 3c. From ref
b
d
2d. Endo/exo ratio was determined both by 1H NMR on the basis
d
of the integration of Cp-protons and by 31P NMR.
cedures were tried, but the use of both cyanocuprates and
CuBr‚Me2S failed to deliver the desired products. Finally,
desilylation of 12 furnished (Z)-2c and (E)-2c in 95 and
99% yields, respectively.
in yields from 13 to 74% (Scheme 4). It is noteworthy
that the formation of 15c and 15d results from the
alkylation of 1b with 13d . In the same manner, the
enediyne 3 was alkylated with 13f giving 16 in 31% yield
(Scheme 5).
In addition, the reaction of Lawesson’s reagent20 with
15g quantitatively provided thiophosphine oxide 15h
(Scheme 6).
Cycliza tion s of Su bstitu ted En ed iyn es P h osp h in e
Oxid es. The cyclizations of the enediynes were conducted
using two methods, A and B (Table 1). When the enediyne
is substituted at the alkyne terminus, both methods could
be used to produce the cycloadducts without affecting the
endo/exo ratio. However, method A was usually preferred
for better yields and shorter reaction times. Precursors
2a ,b afforded red-brown complexes as mixtures of in-
separable endo/exo diastereoisomers of 17a ,b in very good
yields. It is interesting to note that the isolated complexes
are highly stable, since they could be purified on silica
gel with nondegassed solvents. On the contrary, the
behavior of the enediyne substituted at the vinylic
position is totally different. Indeed, method B is the only
effective procedure and (E)-2c is the only stereoisomer
that reacts under these conditions, furnishing the very
unstable cycloadducts 17c even upon purification on
The alkylating chiral phosphine oxide reagents 13a ,b
were prepared as outlined in Scheme 3. Compound 13a
was obtained in 83% yield from the base-catalyzed
condensation of the chiral diamine17 with phosphorus
oxychloride. The sequence (condensation of PCl3 with
t-BuCl followed by an acidic aqueous workup18 and then
alkylation of the resulting tert-butyldichlorophosphine
oxide with sodium methoxide) provided 13b in 70%
overall yield. Finally, the chiral chlorophosphine oxides
13c-g (for their representation, see Scheme 4) were
prepared as previously reported in the literature.19
The preparation of substituted chiral enediynes phos-
phine oxides was achieved using the same protocol as
above.
Deprotonation of 1a or 1b with n-BuLi, followed by
addition of the chiral chlorophosphine oxide 13a -g,
provided the corresponding enediynes 14f-g and 15a -g
(17) Horner, L.; Dickerhof, K. Liebigs Ann. Chem. 1984, 1240-1257.
(18) Crofts, P. C.; Kosolapoff, G. M. J . Am. Chem. Soc. 1953, 75,
3379-3383.
(19) (a) 13c and 13d : J ohnson, C. R.; Elliott, R. C.; Penning, T. D.
J . Am. Chem. Soc. 1984, 106, 5019-5020. (b) 13e: Alexakis, A.; Aujard,
I.; Kanger, T.; Mangeney, P. Org. Synth. 1999, 76, 23-36. (c) 13f:
Korpiun, O.; Lewis, R. A.; Chickos, J .; Mislow, K. J . Am. Chem. Soc.
1968, 90, 4842-4846. (d) 13g: Crofts, P. C.; Parker, D. M. J . Chem.
Soc. C 1970, 332-336.
(20) Pedersen, B. S.; Scheibye, S.; Nilsson, N. H.; Lawesson, S. O.
Bull. Soc. Chim. Belg. 1978, 87, 223-228.
380 J . Org. Chem., Vol. 68, No. 2, 2003

Substituted Linear Enediynes Phosphine Oxides
TABLE 2. Coba lt(I)-Med ia ted Cycliza tion of En ed iyn es
4a ,b
TABLE 3. Cycliza tion s of En ed iyn es 14f,g a n d 15a -g
Bea r in g Ch ir a l P h osh in e Oxid es
SM
R
method
yield (endo/exo) 18a ,b
SM
n
yield (%)
complex
endo (de %)
exo (de %)
4a
P(O)Ph2
H
A
B
77% (100/0)
63% (0/100)
4b2e
14f
14g
15a
15c
15d
15e
15f
15g
3
3
4
4
4
4
4
4
99
93
94
83
88
100
80
100
19f
19g
20a
20c
20d
20e
20f
20g
35(27)
39(56)
24(2)
33(24)
31(34)
22(25)
31(42)
29(58)
65(11)
61(20)
76(6)
67(29)
69(36)
78(15)
69(30)
71(60)a
alumina with degassed solvents. In addition, whatever
the procedure used, (Z)-2c led to intractable materials.
The endo/exo stereochemical assigments of 17a -c
relied on spectral interpretations analogous to those of
the parent compounds 17d -f. These are based on the
characteristic chemical shifts of the angular hydrogen
atom 10-H, which is more shielded in the exo diastereo-
mer than in the endo one. Similarly, the 13C resonance
at the endo-methyne carbon atom C-10 occurs 8-9 ppm
downfield from the analogous exo-methyne carbon atom.
a
Assigned structure of the major exo-20g diastereomer was
confirmed by X-ray analysis.
TABLE 4. In flu en ce of Solven t on th e Cycliza tion of 15g
Several features in these cyclizations are noteworthy,
in particular when compared to those reported in the
literature.2d,3c,21 The substitution of the trimethylsilyl
group of the enediyne (2d ) by a phosphine oxide (2a )
resulted in a slight improvement of the diastereoselec-
tivity (37/63 vs 46/54, while the cyclization of 2d is totally
diastereoselective under thermal conditions). In contrast,
the diastereoselectivity of the cyclization of 2b was really
affected, relative to the parent compounds 2e and 2f, in
favor of the exo complexes. Finally, the cyclization of (E)-
2c is totally diastereoselective in favor of the endo
complexes, albeit the yield is quite moderate.
solvent
toluene
benzene
THF
yield (%)
endo (de %)
exo (de %)
100
100
97
100
74
29(58)
30(64)
29(72)
28(66)
22(69)
23(62)
24(58)
71(60)
70(64)
71(74)
72(60)
78(70)
77(57)
76(61)
dioxane
t-butylmethyl ether
cyclohexane
diethyl ether
100
97
The cyclization of enediyne 4a was carried out accord-
ing to method A and led to the very stable cycloadduct
18a in 77% yield as a unique endo diastereomer as it was
also observed for the ester group.3c This total diastereo-
selectivity is completely opposite to that reported for the
cyclization of 4b, which exhibits hydrogen at the alkyne
termini2e (Table 2).
TABLE 5. Cycliza tion of 14f a n d 14g in THF
The endo stereochemical assignment was attributed by
1H NMR on the basis of the chemical shifts of the angular
hydrogens at δ ) 2.1 ppm, which are consistent with a
cis relationship between those and the cobalt moiety.
SM
yield (%)
complex
endo (de %)
exo (de %)
14f
14g
100
100
19f
19g
36(30)
39(58)
64(8)
61(21)
To define any degree of chiral induction during the
cyclization, we then investigated the behavior of linear
enediynes 14f-g and 15a -g bearing chiral phosphine
oxides on the alkyne. In all cases, the reactions were
performed using method A and the cycloadducts were
obtained in high yields with the same endo/exo selectivity
as that observed for 2a (Table 3).
The results from 15a to 15g showed that the diaster-
eomeric excess for either the exo diastereomer or the endo
one is higher when the stereogenic centers are closer to
the triple bond. The best excesses were observed when
the phosphorus atom was stereogenic, especially for 15f,
which bears a bulky tert-butyl group on phosphorus. The
difference of selectivity between f and g is noticeable
either for the formation of the tricyclic 6,6,5- (19f vs 19g)
or 6,6,6-compounds (20f vs 20g), the tert-butyl group
bringing an important sterical discrimination.
We also checked the influence of the solvent and ran a
series of experiments with 15g according to procedure A
(Table 4). The table shows an increase in the diastereo-
meric excesses in refluxing THF or tert-butylmethyl
ether. However, neither the boiling point nor the polarity
of the solvent is the determining factor for the improve-
ment of the excesses. Indeed, in ether or dioxane, the
excesses are almost the same as in toluene, benzene, or
cyclohexane and, in addition, the reaction did not proceed
at 0 °C in THF.
Surprisingly, the formation of tricyclic 6,6,5-compounds
19f,g in THF was not accompanied by an increase in the
diastereoselectivity related to toluene (Table 5)
In addition, substitution of the methyl group by a
methoxy on the phosphorus atom did not bring any
(21) Sternberg, E. D.; Vollhardt, K. P. C. J . Am. Chem. Soc. 1980,
102, 4839-4841.
J . Org. Chem, Vol. 68, No. 2, 2003 381

Slowinski et al.
TABLE 6. Cycliza tion of 15b a n d 15h in THF
to the group used in the literature (SiMe3, H), the
phosphine oxide is a Lewis base. Thus, we suppose that
the first step of the cyclization that is responsible for the
selectivity could be a Lewis acid/base coordination be-
tween the cobalt moiety and the coordinating site of the
substituent. Then, when the phosphine oxide is at the
alkyne terminus (Scheme 8), the second step could be the
complexation of the internal triple bond that is the closest
and the electronically richest unsaturation of the tether
(I). Consequently, to the coordination of both alkynes, the
triple bond bearing the substituent is in the coordination
sphere of the cobalt, and therefore the oxidative addition
leads to the cobaltacyclopentadiene (II). The latter could
react with the terminal double bond via a [4 + 2] reaction
to afford the major exo complexes. However, a second
pathway has to be considered: after coordination and a
ligand exchange, which could be influenced by the
irradiation, the double bond would be complexed to the
metal (III). Thus, the oxidative coupling could give a
cobaltacyclopentene (IV). Then, the insertion of the
second triple bond into the cobalt-carbon bond followed
by a reductive elimination anti to the six-membered ring
could afford the minor endo complexes.
SM
yield (%)
complex
endo (de %)
exo (de %)
15b
15h
80
29
20b
20h
27(66)
34(58)
73(68)
66(64)
SCHEME 7. Cycliza tion of En ed iyn e 16 in THF
improvement to the diastereoselectivity of the cyclization
(Table 6). On the other hand, no increase in the diaster-
eoselectivity has been shown when a sulfur atom replaced
the oxygen on phosphorus, as was the case for the [3 +
2] cyclizations.12b
Finally, the cyclization of 16 in THF according to
method A led to 21 as only one diastereomer (endo), as
was observed for 4a , but the diastereomeric excess
remained moderate (30%) (Scheme 7).
According to the literature,2b,d two major reaction
pathways may be considered to account for the selectivity
of the cyclizations. One involves the binding of two alkyne
units and leads to a metallacyclopentadiene, which then
reacts with the appended alkene via a Diels-Alder
reaction to give the cycloadduct. The stereochemistry is
set either in the formation of the metallocyclopentadiene
(exo complexation) or during the Diels-Alder step (exo
transition state). In the second pathway, after the initial
formation of a metallacyclopentene, subsequent insertion
of the second alkyne furnishes a metallacycloheptadiene
that affords by a reductive elimination the cycloadduct.
The stereocontrol could occur in these different steps, but
it appears more likely to occur during the formation of
the metallacyclopentene.
This initial coordination could also explain the major
endo selectivity observed, when the phosphine oxide is
at the alkene terminus, the pathway via the cobalta-
cyclopentene in that case being major.
Similarly, these pathways could validate the total endo
selectivity observed when the double bond is internal
(Scheme 9). Indeed, after the initial coordination, the
complexation of the second unsaturation might be ex-
pected to be governed by electronic as well as stereoelec-
tronic factors. Therefore, the complexation of the double
bond would probably be favored over that of the other
triple bond, which would have given a macrocycle.
Oxidative addition would give cobaltacyclopentene V,
which in turn would rearrange to the endo cycloadduct.
It is obvious that a ligand exchange and the subsequent
steps led to the same cobaltacycloheptadiene VI and then
to the endo cycloadduct.
To demonstrate the initial coordination we proposed,
we ran some additional experiments with 2b in the
presence of a Lewis acid such as ZnCl2. For example, we
added 1 equiv of ZnCl2 to precoordinate the phosphine
oxide group before adding CpCo(CO)2 and carrying out
the reaction. Indeed, the cyclization led to 17b in 78%
yield as a 48:52 mixture of exo/endo cycloadducts. The
Looking back at the results we obtained, it appears
that the position of the phosphine oxide group on the
enediyne (acetylenic vs vinylic) could be an additional
controlling feature for the endo/exo selectivity. Compared
SCHEME 8. P h osp h in e Oxid e a t th e Alk yn e Ter m in u s: P ossible P a th w a ys for th e F or m a tion of th e
En d o/Exo Cycloa d d u cts
382 J . Org. Chem., Vol. 68, No. 2, 2003

Substituted Linear Enediynes Phosphine Oxides
SCHEME 9
SCHEME 10
when the double bond was internal and an acetylenic
phosphine oxide was present or when the phosphine
oxide is at the terminal vinylic position, but in the latter
case, the yield is moderate and the adduct very unstable.
When chiral phosphine oxides were employed, we
observed a certain degree of induction during the cycliza-
tion. Depending of the substituents on the phosphorus
atom, especially if the latter is stereogenic, the diaster-
eoselectivity can reach 74%, and up to now, this level is
the highest reported in such a cyclization. In relation with
the position of the phosphine oxide on the enediyne, two
reaction pathways could be proposed to explain the exo/
endo selectivity, an initial coordination between the
cobalt moiety and the chelating site of the substituent
appearing to be the determining factor for the selectivity.
Even if it is impossible to distinguish between these
pathways, they are consistent with our results and bring
additional information concerning the mechanism of the
[2 + 2 + 2] cycloaddition of enediynes.
loss of selectivity means that the postulated coordination
is indeed effective.
The length of the tether does not really influence the
endo/exo selectivity, because this ratio is almost identical
for the formation of the 6,6,6- and 6,6,5-cycloadducts (in
the case of P(O)Ph2, 25/75 vs 37/63). However, with chiral
phosphine oxide, the diastereomeric excesses are quite
different: for example, the de for exo-19g complexes is
21%, whereas it reaches up 74% for exo-20g. This
difference could be explained by the two transition states
of the [4 + 2] reaction, TS-1 and TS-2, leading to the
corresponding cycloadducts (Scheme 10).
The most favored approach of the polyunsaturated
partners in which the nonbonded interactions and A1,3
strains were minimized is the exo approach in which the
tether is in an envelop conformation for TS-1 and a
chairlike for TS-2. Probably, the chairlike conformation
allows a better overlapping of the π orbitals and develops
stronger interactions with the chiral auxiliary than the
envelop conformation and gives rise to a better facial
diastereodifferentiation. For the endo complexes, the
difference between the de is not sufficiently significant
to invoke an effect of the length of the tether on transition
states TS-3 and TS-4.
Exp er im en ta l Section
Syn th esis of Su bstitu ted a n d Ch ir a l Su bstitu ted Lin -
ea r En ed iyn e P h osp h in e Oxid es. See Supporting Informa-
tion.
Gen er a l P r oced u r e for th e Cycliza tion of En ed iyn es.
In most cyclizations, the complexes were obtained as an
inseparable mixture of endo/exo cycloadducts (four com-
pounds). In some cases, one couple of cycloadducts is fully
described; for the other case, we gave only characteristic data.
Meth od A. CpCo(CO)2 (0.37 mmol, 1.2 equiv) was added
to a boiling solution of the enediyne (0.31 mmol, 1 equiv) in
toluene (9 mL), degassed by three freeze-pump-thaw cycles,
and irradiated (light from a projector lamp, ELW, 300 W, 80%
of its power). The reaction was monitored by TLC, and after
completion (about 1 h), the reaction mixture was purified by
flash chromatography either on deactivated alumina (4% H2O)
or on silica gel.
Meth od B. CpCo(CO)2 (0.37 mmol, 1.2 equiv) was added
to a boiling solution of the enediyne (0.31 mmol, 1 equiv) in
decane (9 mL) and degassed by three freeze-pump-thaw
cycles, and the reaction mixture was stirred at reflux over-
night. Then, the solution was purified by flash chromatography
on deactivated alumina (4% H2O) or on silica gel.
Com p lexes 17a . Yield ) 88%. Method A. Exo (character-
istic data): 1H NMR (400 MHz, C6D6) δ 4.93 (s, 5H), 3.87 (dt,
J ) 17.3, 5.6 Hz, 1H), 1.22 (dd, J ) 8.2, 2.5 Hz, 1H), 0.69 (m,
1H); 13C NMR (100 MHz, C6D6) δ 136.3 (J ) 99.1 Hz), 136.1
(J ) 91.0 Hz), 133.2 (J ) 8.1 Hz), 132.0 (2C, J ) 10.1 Hz),
131.8 (2C), 130.7 (J ) 8.1 Hz), 128.5 (J ) 13.2 Hz), 128.3 (J )
10.1 Hz), 127.7 (2C, J ) 11.1 Hz), 94.1 (J ) 8.1 Hz), 89.1 (J )
12.1 Hz), 83.2 (5C), 76.8, 48.6 (J ) 91 Hz), 40.9 (J ) 12.1 Hz),
36.4 (J ) 10.1 Hz); 31P NMR (162 MHz, C6D6) 37.5. E n d o
(characteristic data): 1H NMR (400 MHz, C6D6) δ 4.85 (s, 5H),
4.19 (dt, J ) 12.7, 4.6 Hz, 1H), 1.19 (dd, J ) 8.2, 3.1 Hz, 1H),
Con clu sion
In summary, the cobalt(I)-mediated cyclization of
linear enediynes bearing a phosphine oxide at the acety-
lenic position provided predominantly exo cycloadducts
in high yields. Those cycloadducts as well as the endo
ones are very air-stable and can easily be purified and
handled. On the opposite, the cyclization was totally
diastereoselective in favor of the endo complexes either
J . Org. Chem, Vol. 68, No. 2, 2003 383

Slowinski et al.
0.58-0.47 (m, 1H), -0.43 (dd, J ) 11.2, 9.1 Hz); 13C NMR (100
MHz, C6D6) δ 136.7 (J ) 103.1 Hz), 136.4 (J ) 87.0 Hz), 132.5
(2C, J ) 8.1 Hz), 132.2 (2C, J ) 10.1 Hz), 131.2 (J ) 12.1 Hz),
130.4 (2C), 128.5 (J ) 13.2 Hz), 128.2 (J ) 8.1 Hz), 127.9 (J )
9.1 Hz), 95.0 (J ) 4.0 Hz), 88.8 (J ) 12.1 Hz), 83.2 (5C), 75.8,
52.6 (J ) 8.1 Hz), 32.8 (J ) 6.1 Hz); 31P NMR (162 MHz, C6D6)
δ 37.0. (En d o + Exo) 17a (unattributed): 1H NMR (400 MHz,
C6D6) δ 8.22-8.16 (m, 2H), 8.14-7.96 (m, 4H), 7.92-7.86 (m,
2H), 7.24-7.12 (m, 12H), 2.51-2.21 (m, 4H), 2.20-1.28 (m,
23H), 1.10-0.95 (m, 2H); 13C NMR (100 MHz, C6D6) δ 36.0,
32.5, 31.0, 30.0, 28.4, 28.3, 27.7, 26.9, 26.8, 24.1, 23.9, 23.5,
uted): 13C NMR (100 MHz, C6D6) δ 132.2 (J ) 8.3 Hz), 131.9,
131.8, 131.6 (J ) 9.9 Hz), 131.4 (J ) 8.3 Hz), 130.9 (J ) 9.1
Hz), 130.8 (2C), 130.7 (J ) 8.3 Hz), 130.6 (J ) 7.6 Hz), 130.5
(J ) 5.3 Hz), 130.3 (J ) 4.5 Hz), 130.2 (J ) 4.5 Hz), 128.7,
128.3 (J ) 10.6 Hz), 128.2 (J ) 10.6 Hz), 128.2, 128.1(2C),
127.8, 34.3, 34.2, 34.1, 33.9, 33.8, 33.6, 32.9, 32.7, 32.5, 32.4,
31.1, 31.0, 29.8, 29.7, 27.7, 27.6, 27.5, 27.3, 27.2, 27.1, 27.0,
26.9, 25.6, 25.4, 24.0, 23.7, 23.5, 23.3, 23.2, 23.1, 23.0, 22.9;
1H NMR (400 MHz, C6D6) δ 8.08-7.87 (m, 8H), 7.35-7.10 (m,
12H), 4.21-3.87 (m), 3.67 (dq, J ) 10.2, 6.6 Hz), 2.61 (m),
2.40-1.19 (m), 1.04-0.82 (m).
23.2 (2C); IR (neat) 3060, 2940, 1480, 1435, 1235 cm-1
.
Com p lexes 19g. Yield ) 93%. Method A. En d o (charac-
teristic data). 1H NMR (400 MHz, C6D6) δ: (major) 4.78 (s,
5H), 2.13 (m, 1H), 1.45 (d, J ) 23.8 Hz, 3H), 1.21 (d, J ) 12.2
Hz, 9H), -0.52 (dd, J ) 9.7, 1.6 Hz, 1H); (minor) 4.77 (s, 5H),
2.21 (m, 1H), 1.43 (d, J ) 24.4 Hz, 3H), 1.26 (d, J ) 10.2 Hz,
9H), -0.65 (ddd, J ) 9.7, 4.6, 1.0 Hz, 1H). 13C NMR (100 MHz,
C6D6) δ: (major) 92.2 (J ) 8.1 Hz), 88.0 (J ) 10.1 Hz), 83.3
(5C), 74.4, 51.6 (J ) 83.0 Hz), 35.2 (J ) 58.6 Hz), 53.1 (J )
8.1 Hz), 26.7 (3C), 13.1 (J ) 68.7 Hz); (minor) 94.3 (J ) 8.2
Hz), 88.0 (J ) 10.1 Hz), 83.2 (5C), 75.0, 51.8 (J ) 10.1 Hz),
51.6 (J ) 83.0 Hz), 35.2 (J ) 58.6 Hz), 26.6 (3C), 13.1 (J )
68.7 Hz). 31P NMR (162 MHz, C6D6) δ: (major) 55.8; (minor)
Com p lexes 17b. Yield ) 96%. Method A. Exo (character-
istic data): 1H NMR (400 MHz, C6D6) δ 8.21-8.13 (m, 2H),
8.05-8.00 (m, 1H), 7.95-7.87 (m, 2H), 7.30-7.10 (m, 5H), 4.98
(s, 5H), 3.95 (d, J ) 2.6 Hz, 1H), 2.31 (t, J ) 5.1 Hz, 1H), 1.86
(m, 1H), 1.53 (m, 1H), 1.00 (t, J ) 7.1 Hz), 2.53-0.90 (m, 12H),
0.71 (m, 1H), 0.51 (m, 1H); 13C NMR (100 MHz, C6D6) δ 142.2
(2C, J ) 113.2 Hz), 136.4 (2C, J ) 13.9 Hz), 132.6 (2C, J )
9.0 Hz), 131.7 (2C, J ) 8.8 Hz), 130.3 (2C), 127.6 (2C, J )
10.3 Hz), 90.8 (J ) 12.1 Hz), 83.3 (5C), 73.4, 46.5 (J ) 93 Hz),
41.6 (J ) 10.1 Hz), 33.6; 31P NMR (162 MHz, C6D6) δ 36.1.
En d o (characteristic data): 1H NMR (400 MHz, C6D6) δ 4.83
(s, 5H), 2.17 (m, 1H); 13C NMR (100 MHz, C6D6) δ 130.1 (2C),
128.2 (2C, J ) 12.4 Hz), 95.4 (J ) 5.3 Hz), 91.3 (J ) 12.1 Hz),
83.8 (5C), 73.1, 48.6 (J ) 10.1 Hz); 31P NMR (162 MHz, C6D6)
δ 36.2. (En d o + Exo) 17b (unattributed): 13C NMR (100 MHz,
C6D6) δ 36.7, 36.5 (2C), 33.8, 33.4, 33.3, 32.8, 30.1, 30.0, 29.7,
28.9, 28.3, 28.1, 27.9, 27.5, 26.9, 26.6, 26.5, 26.4, 23.5, 23.3,
1
55.8. Exo (characteristic data). H NMR (400 MHz, C6D6) δ:
(major) 4.79 (s, 5H), 1.46 (d, J ) 23.9 Hz, 3H), 1.18 (d, J )
13.2 Hz, 9H), 0.71 (m, 1H); (minor) 4.82 (s, 5H), 1.42 (d, J )
24.3 Hz, 3H), 1.27 (d, J ) 13.2 Hz, 9H), 0.64 (m, 1H). 13C NMR
(100 MHz, C6D6) δ: (major) 95.5 (J ) 6.1 Hz), 87.7 (J ) 12.1
Hz), 83.6 (5C), 77.0, 52.6 (J ) 78.0 Hz), 41.9 (J ) 11.1 Hz),
35.4 (J ) 68.7 Hz), 27.3 (3C), 12.9 (J ) 66.7 Hz); (minor) 96.8
(J ) 6.1 Hz), 88.3 (J ) 10.1 Hz), 83.4 (5C), 76.6, 52.6 (J )
78.0 Hz), 40.4 (J ) 10.1 Hz), 35.4 (J ) 68.7 Hz), 27.3 (3C),
12.9 (J ) 66.7 Hz). 31P NMR (162 MHz, C6D6) δ: (major) 57.4;
(minor) 57.4. (En d o + Exo) 19g (unattributed): 1H NMR (400
MHz, C6D6) δ 4.03 (m, 1H), 2.10-2.60 (m), 1.53-2.08 (m),
0.95-1.15 (m), 0.5-0.78 (m); 13C NMR (100 MHz, C6D6) δ 36.0,
35.5, 34.6, 34.3, 34.2, 33.4, 33.3 (3C), 32.8, 32.5 (2C), 31.2, 30.2,
30.0, 29.8, 29.1, 28.5, 28.2, 28.0, 27.9, 27.7, 27.4, 27.1, 24.0,
23.9 (2C), 23.4, 23.2 (2C), 22.7, 22.6; IR (C6D6) 2920, 2850,
19.1 (2C); IR (neat) 3060, 2940, 2860, 1480, 1435, 1235 cm-1
.
Com p lex en d o-17c. Yield ) 32%. Method B: 1H NMR (400
MHz, C6D6) δ 8.26-8.00 (m, 4H), 7.36-7.16 (m, 6H), 4.60 (s,
5H), 2.77 (d, 1H, J ) 10.8 Hz), 2.03 (s, 3H), 2.00-0.82 (m,
17H); 13C NMR (100 MHz, C6D6) δ 138.0 (2C, J ) 88.2 Hz),
131.8 (2C, J ) 8.1 Hz), 131.5 (2C, J ) 9.1 Hz), 131.0 (2C),
130.6 (2C, J ) 9.1 Hz), 128.2 (2C, J ) 7.1 Hz), 94.0, 93.7, 92.7,
82.2 (5C), 51.9 (J ) 8.8 Hz), 46.7 (J ) 73 Hz), 43.4, 41.5, 31.4,
30.6, 27.6, 25.4 (2C), 25.2 (J ) 9.1 Hz), 24.1, 23.4; 31P NMR
(162 MHz, C6D6) δ 28.9.
1440, 1360, 1285, 1140 cm-1
.
Com p lex en d o-18a . Yield ) 77%. Method A: 1H NMR (400
MHz, C6D6) δ 8.35-8.30 (m, 2H), 8.23-8.18 (m, 2H), 8.10-
8.04 (m, 2H), 7.32-7.12 (m, 4H), 4.78 (s, 5H), 4.37 (d, J ) 5.6
Hz, 1H), 3.62 (d, J ) 12.7 Hz, 1H), 2.21 (tt, J ) 12.0, 3.0 Hz,
1H), 2.01 (td, J ) 12.0, 3.0 Hz, 1H), 1.85-1.19 (m, 13H), 1.10-
0.95 (m, 1H), 0.83-0.76 (m, 1H); 13C NMR (100 MHz, C6D6) δ
136.2 (J ) 99.1 Hz), 134.2 (J ) 101.1 Hz), 132.8 (J ) 8.1 Hz),
132.3 (J ) 8.1 Hz), 132.0 (J ) 8.1 Hz), 131.9 (J ) 12.1 Hz),
131.3, 131.2, 128.5 (J ) 8.1 Hz), 128.4 (J ) 9.1 Hz), 128.3
(J ) 11.1 Hz), 127.8 (J ) 11.1 Hz), 84.0 (J ) 44.5 Hz), 82.3
(5C), 79.1 (J ) 12.1 Hz), 77.7 (J ) 24.3 Hz), 77.4, 49.2, 46.7,
40.7, 37.9, 34.0, 31.7, 31.3, 28.9, 27.9, 27.8; 31P NMR (162 MHz,
Com p lexes 20a . Yield ) 94%. Method A. En d o (charac-
teristic data). 1H NMR (400 MHz, C6D6) δ: (major) 4.68 (s,
5H); (minor) 4.70 (s, 5H). 13C NMR (100 MHz, C6D6) δ: (major)
143.3 (2C), 96.4 (J ) 6.1 Hz), 92.0 (J ) 16.2 Hz), 72.4, 51.3
(J ) 170 Hz), 83.5 (5C), 49.8 (J ) 4.1 Hz); (minor) 142.7 (2C),
96.4 (J ) 6.1 Hz), 92.0 (J ) 16.2 Hz), 83.6 (5C), 72.4, 51.3
(J ) 170 Hz), 49.8 (J ) 4.1 Hz). 31P NMR (162 MHz C6D6) δ:
(major) 42.2; (minor) 41.8. Exo (characteristic data). 1H NMR
(400 MHz, C6D6) δ: (major) 4.82 (s, 5H); (minor) 4.81 (s, 5H).
13C NMR (100 MHz C6D6) δ: (major) 144.0 (2C), 95.6 (J ) 8.1
Hz), 90.4 (J ) 14.2 Hz), 83.0 (5C), 72.4, 50.6 (J ) 115 Hz),
40.4 (J ) 7.1 Hz); (minor) 144.1 (2C), 95.5 (J ) 6.1 Hz), 90.4
(J ) 14.2 Hz), 82.9 (5C), 72.4, 50.6 (J ) 115 Hz), 40.4 (J ) 7.1
Hz). 31P NMR (162 MHz, C6D6) δ: (major) 42.9; (minor) 42.8.
(En d o + Exo) 20a (unattributed): 1H NMR (400 MHz, C6D6)
δ 8.00-7.15 (m, 40H), 5.48-5.19 (2m), 4.41-4.26 (m), 2.78-
2.32 (m), 1.64 (d, J ) 7.1 Hz, 3H), 1.28 (d, J ) 7.0 Hz, 3H),
2.21-0.5 (m); 13C NMR (100 MHz, C6D6) δ 126.7-128.69 (m,
40C), 50.3-53.5 (m, 8C), 36.9, 36.8 (2C), 36.7 (2C), 35.4, 35.3,
35.2, 35.1, 34.0, 33.9, 33.8, 33.3, 33.0, 32.8, 32.7 (2C), 29.7 (2C),
29.4 (2C), 28.9, 28.8, 27.5, 27.4, 27.3, 27.2, 27.1, 27.0 (2C), 26.9
(2C), 26.8, 26.7 (2C), 26.6, 26.5, 24.1, 24.0 (2C), 23.9 (2C), 23.8,
23.7, 16.7-17.9 (m, 8C).
C6D6) δ 31.0; IR (C6D6) 3060, 2920, 1440, 1235 cm-1
.
Com p lexes 19f. Yield ) 99%. Method A. En d o (charac-
teristic data). 1H NMR (400 MHz, C6D6) δ: (major) 4.86 (s,
5H) 1.41 (d, J ) 14.8 Hz, 3H), -0.43 (dd, J ) 12.2, 9.2 Hz,
1H); (minor) 4.76 (s, 5H), 1.69 (d, J ) 12.7 Hz, 3H), -0.54 (dd,
J ) 11.2, 9.2 Hz, 1H). 13C NMR (100 MHz, C6D6) δ: (major)
137.8 (J ) 183.4 Hz), 94.5 (J ) 4.6 Hz), 89.1 (J ) 12.1 Hz),
83.1 (5C), 74.9, 52.5 (J ) 9.8 Hz), 50.4 (J ) 57.0 Hz), 16.0
(J ) 47.8 Hz); (minor) 137.8 (J ) 183.4 Hz), 94.4 (J ) 5.3 Hz),
88.6 (J ) 12.0 Hz), 82.9 (5C), 75.6, 51.8 (J ) 9.8 Hz), 50.4
(J ) 57.0 Hz), 17.3 (J ) 46.5 Hz). 31P NMR (162 MHz, C6D6)
δ: (major) 36.0; (minor) 40.2. Exo (characteristic data). 1H
NMR (400 MHz, C6D6) δ: (major) 4.88 (s, 5H), 2.49 (td, J )
17.3, 6.1 Hz, 1H), 1.70 (d, J ) 12.7 Hz, 3H); (minor) 4.82 (s,
5H), 1.56 (d, J ) 11.7 Hz, 3H); 13C NMR (100 MHz, C6D6) δ:
(major) 137.7 (J ) 178.8 Hz), 93.4 (J ) 6.1 Hz), 89.0 (J ) 11.4
Hz), 82.7 (5C), 76.7, 50.6 (J ) 56.1 Hz), 40.9 (J ) 12.1 Hz),
18.5 (J ) 48.9 Hz); (minor) 138.2 (J ) 171.3 Hz), 92.5 (J ) 7.6
Hz), 88.9 (J ) 11.4 Hz), 82.8 (5C), 76.8, 50.0 (J ) 55 Hz), 40.6
(m, J ) 12.1 Hz), 13.9 (J ) 45.5 Hz); 31P NMR (162 MHz, C6D6)
δ: (major) 37.3; (minor) 37.8. (En d o + Exo) 19f (unattrib-
Com p lexes 20b. Yield ) 80%. Method A. En d o (charac-
teristic data). 1H NMR (400 MHz, C6D6) δ: (major) 4.86 (s,
5H), 3.75 (d, J ) 9.6 Hz, 3H), 2.56 (ddd, J ) 16.3, 12.2, 5.1
Hz, 1H), 1.30 (d, J ) 7.2 Hz, 9H), 0.52 (td, J ) 12.2, 3.6 Hz,
1H), -0.05 (ddd, J ) 12.2, 7.1, 1.6 Hz, 1H); (minor) 4.81 (s,
5H), -0.08 (ddd, J ) 12.4, 6.9, 1.2 Hz, 1H). 13C NMR (100
MHz, C6D6) δ: (major) 83.3 (5C), 48.5 (J ) 8.1 Hz), 33.8 (J )
8.1 Hz), 27.0; no characteristic data for the minor endo
compound. 31P NMR (162 MHz, C6D6) δ: (major) 60.6; (minor)
384 J . Org. Chem., Vol. 68, No. 2, 2003

Substituted Linear Enediynes Phosphine Oxides
1
66.2. Exo (characteristic data). H NMR (400 MHz, C6D6) δ:
72.5, 59.5 (J ) 8.1 Hz), 41.8 (J ) 159.7 Hz), 40.0 (J ) 10.1
Hz), 32.8 (J ) 8.1 Hz), 29.1 (J ) 5.1 Hz), 14.2. 31P NMR (162
MHz, C6D6) δ: (major) 48.7; (minor) 48.4. (En d o + Exo) 20d
(unattributed): 1H NMR (400 MHz, C6D6) δ 7.26-7.09 (m,
12H), 3.36-3.03 (m), 4.23-4.11 (3t, J ) 4.6 Hz, 3H), 3.96-
3.80 (4t, J ) 5.1 Hz, 4H), 2.94, 1.14 (m), 0.92-0.59 (2m); 13C
NMR (100 MHz, C6D6) 127.4 (6C), 127.2 (4C), 126.9 (4C), 126.3
(4C), 37.0 (2C), 36.9, 36.8, 36.7, 36.5, 35.8, 34.0 (2C), 33.7 (2C),
28.5, 28.4, 28.3, 28.2, 27.6, 27.3 (2C), 26.8 (2C), 26.7, 26.6 (2C),
26.4, 23.9, 23.8, 23.7, 23.6, 23.5 (2C), 23.3, 23.1.
(major) 4.99 (s, 5H), 3.73 (d, J ) 9.6 Hz, 3H), 3.96-3.87 (m,
1H), 2.47 (ddd, J ) 15.8, 11.7, 4.6 Hz, 1H), 1.29 (d, J ) 9.6
Hz, 9H), 0.74 (t, J ) 9.6 Hz, 1H), 0.61 (ddd, J ) 13.2, 8.1, 2.0
Hz, 1H); (minor) 4.92 (s, 5H), 3.64 (d, J ) 9.5 Hz, 3H), 1.31
(d, J ) 9.4 Hz, 9H). 13C NMR (100 MHz, C6D6) δ: (major)
96.2 (J ) 10.1 Hz), 89.6 (J ) 12.1 Hz), 82.8 (5C), 72.6, 50.8
(J ) 113.2 Hz), 51.1, 41.1 (J ) 9.1 Hz), 36.5, 36.3 (J ) 86.1
Hz), 33.7, 33.3 (J ) 6.1 Hz), 29.2, 27.2 (3C), 26.8, 26.5, 26.3,
23.9, 23.5; no characteristic data for the minor exo compound.
31P NMR (162 MHz, C6D6) δ: (major) 60.4; (minor) 66.1.
(En d o + Exo) 20b (unattributed): 1H NMR (400 MHz, C6D6)
δ 2.28-2.11 (m), 2.06-1.84 (m), 1.68-1.01 (m); IR (C6D6) 2920,
Com p lexes 20e. Yield ) 100%. Method A. En d o (charac-
teristic data). 1H NMR (400 MHz, C6D6) δ: (major) 4.72 (s,
5H), 4.20-3.80 (m, 2H), 2.34 (d, J ) 9.2 Hz, 6H); (minor) 4.65
(s, 5H), 4.20-3.80 (m, 2H), 2.74 (d, J ) 7.5 Hz, 6H). 13C NMR
(100 MHz, C6D6) δ: (major) 140.3 (J ) 10.1 Hz), 139.6 (J )
4.0 Hz), 97.2 (J ) 8.1 Hz), 91.2 (J ) 14.2 Hz), 83.3 (5C), 75.4
(J ) 6.1 Hz), 74.4 (J ) 4.3 Hz), 73.4, 47.9 (J ) 11.1 Hz), 45.2
(J ) 142.0 Hz), 33.1, 30.3 (6.1 Hz); (minor) 140.3 (J ) 10.1
Hz), 139.6 (J ) 4.0 Hz), 97.2 (J ) 8.1 Hz), 91.0 (J ) 14.2 Hz),
84.0 (5C), 75.4 (J ) 6.1 Hz), 74.4 (J ) 4.3 Hz), 72.9, 48.8 (J )
11.1 Hz), 45.8 (J ) 155.0 Hz), 33.1, 31.0. 31P NMR (162 MHz,
C6D6) δ: (major) 48.3; (minor) 46.3. Exo (characteristic data).
1H NMR (400 MHz, C6D6) δ: (major) 4.86 (s, 5H), 4.20-3.80
(m, 2H), 2.73 (d, J ) 10.4 Hz, 6H), 0.85 (m, 1H); (minor) 4.75
(s, 5H), 4.20-3.80 (m, 2H), 2.54 (d, J ) 10.6 Hz, 6H). 13C NMR
(100 MHz, C6D6) δ: (major) 139.9 (J ) 8.1 Hz), 139.3 (J ) 4.0
Hz), 96.5 (J ) 10.1 Hz), 90.4 (J ) 14.2 Hz), 82.9 (5C), 74.2
(J ) 4.1 Hz), 73.7, 73.6 (J ) 4.2 Hz), 48.1 (J ) 141.6 Hz), 40.6
(J ) 11.1 Hz), 32.9, 30.1 (J ) 7.1 Hz); (minor) 139.9 (J ) 8.1
Hz), 139.3 (J ) 4.0 Hz), 95.7 (J ) 8.1 Hz), 90.4 (J ) 14.2 Hz),
83.7 (5C), 74.2 (J ) 4.1 Hz), 73.6 (J ) 4.2 Hz), 72.6, 47.3 (J )
145.6 Hz), 41.3 (J ) 11.1 Hz), 32.6, 31.3 (J ) 4.1 Hz). 31P NMR
(162 MHz, C6D6) δ: (major) 47.6; (minor) 45.8. (En d o + Exo)
20e (unattributed): 1H NMR (400 MHz, C6D6) δ 7.25-6.95 (m,
40H), 2.33-1.10 (m), 0.79-0.32 (m); 13C NMR (100 MHz, C6D6)
δ 127.8-128.9 (mult, 40C), 37.2, 37.1, 37.0, 36.8, 34.1, 33.8,
33.2, 33.1, 31.7 (2C, J ) 8.1 Hz), 29.4 (2C), 27.6, 27.4, 27.2,
27.1, 27.0, 26.9 (2C), 26.8 (2C), 26.7, 26.5, 24.0 (2C), 23.9 (2C),
23.7 (2C), 23.5 (2C), 23.4, 23.3, 23.2, 22.8, 22.7.
Com p lexes 20f. Yield ) 80%. Method A. En d o (charac-
teristic data). 1H NMR (400 MHz, C6D6) δ: (major) 4.84 (s,
5H), 1.57 (m, 3H), 2.09 (m, 1H); (minor) 4.74 (s, 5H), 1.57 (m,
3H), 2.09 (m, 1H). 13C NMR (100 MHz, C6D6) δ: (major) 95.1
(J ) 4.1 Hz), 91.3 (J ) 12.1 Hz), 83.0 (5C), 72.3, 48.5 (J ) 8.1
Hz), 48.1 (J ) 87.0 Hz), 15.2 (J ) 64.7 Hz); (minor) 95.1 (J )
4.1 Hz), 91.3 (J ) 12.1 Hz), 83.1 (5C), 72.3, 48.1 (J ) 87.0
Hz), 48.5 (J ) 8.1 Hz), 15.1 (J ) 65.1 Hz). 31P NMR (162 MHz,
C6D6) δ: (major) 37.5; (minor) 37.7. Exo (characteristic data).
1H NMR (400 MHz, C6D6) δ: (major) 4.95 (s, 5H), 1.74 (m,
3H), 0.68 (m, 1H), -0.13 (m, 1H); (minor) 4.90(s, 5H), 1.74 (m,
3H), 0.68 (m, 1H), -0.22 (m, 1H). 13C NMR (100 MHz, C6D6)
δ: (major) 134.4 (J ) 101.1 Hz), 132.3 (2C, J ) 10.1 Hz), 131.5,
128.5 (2C, J ) 12.1 Hz), 94.8, 94.3 (J ) 4.1 Hz), 90.4 (J ) 10.1
Hz), 82.6 (5C), 72.0, 48.2 (J ) 91.2 Hz), 40.4 (J ) 12.1 Hz),
17.2 (J ) 74.8 Hz); (minor) 134.4 (J ) 101.1 Hz), 94.8, 94.3
(J ) 4.1 Hz), 90.4 (J ) 10.1 Hz), 82.4 (5C), 71.9, 48.2 (J )
91.2 Hz), 40.1 (J ) 12.1 Hz), 17.1 (J ) 76.8 Hz). 31P NMR (162
MHz, C6D6) δ: (major) 36.1; (minor) 36.2. (En d o + Exo) 20f
(unattributed): 1H NMR (400 MHz, C6D6) δ 8.04-7.97 (m, 4H),
7.94-7.74 (m, 4H), 7.18-7.02 (m, 12H), 4.18-3.92 (m), 2.04-
1.18 (m), 0.66-0.57 (m); 13C NMR (100 MHz, C6D6) δ 132.0
(2C, J ) 102.1 Hz), 131.9, 131.8, 130.9, 130.8 (2C), 130.6 (4C),
130.3 (4C), 39.1, 36.8 (2C), 36.5, 36.4 (2C), 33.7, 33.6, 33.5,
30.7, 30.1, 29.2, 28.9, 28.6, 28.4 (2C), 28.3, 27.8 (2C), 27.4 (2C),
27.1, 26.7, 26.4 (2C), 26.3, 26.2, 24.1, 23.7 (2C), 23.5 (2C), 23.4
(2C), 23.2 (2C), 18.9, 18.5.
2850, 1440, 1360, 1285 cm-1
.
Com p lexes 20c. Yield ) 83%. Method A. En d o (charac-
teristic data). 1H NMR (400 MHz, C6D6) δ: (major) 5.82 (d,
J ) 6.8 Hz, 1H), 4.82 (s, 5H), 2.60 (d, J ) 9.2 Hz, 3H), 2.02
(m, 1H), 0.58 (d, J ) 6.3 Hz, 3H), 0.09 (ddd, J ) 11.7, 6.6, 1.0
Hz, 1H); (minor) 5.83 (d, J ) 6.8 Hz, 1H), 4.89 (s, 5H), 2.61 (d,
J ) 9.3 Hz, 3H), 0.58 (d, J ) 6.3 Hz, 3H). 13C NMR (100 MHz,
C6D6) δ: (major) 138.0 (J ) 10.1 Hz), 96.4 (J ) 10.1 Hz), 92.4
(J ) 14.2 Hz), 83.2 (5C), 78.4, 72.9, 61.7 (J ) 8.1 Hz), 48.2
(J ) 12.1 Hz), 39.2 (J ) 93.0 Hz), 33.5 (J ) 8.1 Hz), 28.9 (J )
8.1 Hz), 15.3; (minor) 138.3 (J ) 10.1 Hz), 96.7 (J ) 10.1 Hz),
92.1 (J ) 14.2 Hz), 83.6 (5C), 78.3, 72.6, 61.7 (J ) 8.1 Hz),
47.9 (J ) 10.1 Hz), 39.1 (J ) 95.0 Hz), 33.3 (J ) 8.1 Hz), 29.1
(J ) 6.1 Hz), 15.4. 31P NMR (162 MHz, C6D6) δ: (major) 50.2;
(minor) 48.3. Exo (characteristic data). 1H NMR (400 MHz,
C6D6) δ: (major) 5.73 (d, J ) 6.6 Hz, 1H), 4.93 (s, 5H), 2.83 (d,
J ) 8.8 Hz, 3H), 0.69 (m, 1H), 0.53 (d, J ) 6.7 Hz, 3H); (minor)
5.74 (d, J ) 6.6 Hz, 1H), 5.01 (s, 5H), 2.84 (d, J ) 8.9 Hz, 3H),
1.92 (m, 1H), 0.69 (m, 1H), 0.56 (d, J ) 8.7 Hz, 3H), 0.24 (dd,
J ) 12.7, 7.1 Hz, 1H). 13C NMR (100 MHz, C6D6) δ: (major)
138.1 (J ) 10.1 Hz), 96.0 (J ) 8.1 Hz), 91.5 (J ) 14.2 Hz),
82.9 (5C), 78.4, 72.7, 61.5 (J ) 6.1 Hz), 43.4 (J ) 157.8 Hz),
40.1 (J ) 12.1 Hz), 34.1 (J ) 8.1 Hz), 29.7 (J ) 4.0 Hz), 15.3;
(minor) 138.3 (J ) 10.1 Hz), 96.3 (J ) 8.1 Hz), 91.6 (J ) 14.2
Hz), 83.1 (5C), 78.3, 72.6, 61.3 (J ) 6.1 Hz), 42.2 (J ) 164.0
Hz), 40.1 (J ) 12.1 Hz), 34.1 (J ) 8.1 Hz), 30.1 (J ) 2.1 Hz),
15.0. 31P NMR (162 MHz, C6D6) δ: (major) 49.0; (minor) 47.4.
(En d o + Exo) 20c (unattributed): 1H NMR (400 MHz, C6D6)
δ: 7.29-6.98 (m, 20H), 4.44-4.09 (m), 4.04-3.79 (m), 3.26-
3.19 (m), 2.99-0.44 (m); 13C NMR (100 MHz, C6D6) δ 128.3
(6C), 127.7 (6C), 126.2 (4C), 126.0 (4C), 37.0, 36.8, 36.7, 36.6,
33.9, 33.7 (2C), 28.5 (2C), 28.1(2C), 27.5 (2C), 27.3 (2C), 26.9,
26.8, 26.7, 26.6 (2C), 26.4, 24.0, 23.9, 23.8, 23.7, 23.6 (2C), 23.5
(2C), 23.3, 23.1, 22.7.
Com p lexes 20d . Yield ) 88%. Method A. En d o (charac-
teristic data). 1H NMR (400 MHz, C6D6) δ: (major) 7.46 (d,
J ) 7.6 Hz, 2H), 5.28 (d, J ) 6.1 Hz, 1H), 4.78 (s, 5H), 2.46 (d,
J ) 10.2 Hz, 3H), 2.14 (m, 1H), 0.70 (d, J ) 6.6 Hz, 3H); (minor)
7.34 (d, J ) 7.6 Hz, 2H), 5.30 (d, J ) 5.1 Hz, 1H), 4.88 (s, 5H),
2.32 (d, J ) 9.6 Hz, 3H), 2.24 (m, 1H), 2.14 (m, 1H), 0.73 (d,
J ) 6.6 Hz, 3H), 0.03 (ddd, J ) 13.0, 7.9, 2.0 Hz, 1H). 13C NMR
(100 MHz, C6D6) δ: (major) 138.4, 127.9, 96.4 (J ) 8.1 Hz),
92.2 (J ) 14.2 Hz), 83.5 (5C), 82.1, 72.3, 59.6 (J ) 8.1 Hz),
48.3 (J ) 14.2 Hz), 42.4 (J ) 153.7 Hz), 32.4 (J ) 8.1 Hz),
28.3 (J ) 7.1 Hz), 14.7; (minor) 138.5, 128.0, 96.5 (J ) 6.1
Hz), 92.0 (J ) 16.2 Hz), 83.7 (5C), 81.8, 72.0.3, 59.2 (J ) 8.1
Hz), 48.2 (J ) 12.1 Hz), 41.8 (J ) 159.7 Hz), 32.2 (J ) 10.1
Hz), 28.2 (J ) 4.0 Hz), 15.0. 31P NMR (162 MHz, C6D6) δ:
(major) 49.6; (minor) 48.8. Exo (characteristic data). 1H NMR
(400 MHz, C6D6) δ: (major) 7.49 (d, J ) 7.6 Hz, 2H), 5.25 (d,
J ) 4.6 Hz, 1H), 4.89 (s, 5H), 2.75 (d, J ) 9.1 Hz, 3H), 0.79 (d,
J ) 6.6 Hz, 3H), 0.75 (m, 1H); (minor) 7.59 (d, J ) 7.1 Hz,
2H), 5.24 (d, J ) 6.1 Hz, 1H), 5.00 (s, 5H), 2.45 (d, J ) 10.2
Hz, 3H), 1.90 (m, 1H), 0.81 (d, J ) 5.1 Hz, 3H), 0.65 (m, 1H),
0.13 (ddd, J ) 12.7, 7.0, 1.5 Hz, 1H). 13C NMR (100 MHz, C6D6)
δ: (major) 138.3, 95.2 (J ) 8.1 Hz), 91.4 (J ) 14.2 Hz), 83.1
(5C), 82.0, 72.6, 59.9 (J ) 8.1 Hz), 42.4 (J ) 153.7 Hz), 40.2
(J ) 12.1 Hz), 33.1 (J ) 8.1 Hz), 29.3 (J ) 5.1 Hz), 14.8; (minor)
138.4, 95.5 (J ) 6.1 Hz), 91.3 (J ) 16.2 Hz), 83.3 (5C), 81.7,
Com p lexes 20g. Yield ) 100%. Method A. En d o (charac-
teristic data): 1H NMR (400 MHz, C6D6) δ: (major) 4.75 (s,
5H), 1.97 (m, 1H), 1.30 (d, J ) 13.2 Hz, 3H), 1.20 (d, J ) 13.2
Hz, 9H), -0.29 (dd, J ) 12.2, 7.1 Hz, 1H); (minor) 4.76 (s, 5H),
1.27 (d, J ) 13.2 Hz, 3H), 1.20 (d, J ) 13.2 Hz, 9H), 1.19 (m,
1H). 13C NMR (100 MHz, C6D6) δ: (major) 97.6 (J ) 4.0 Hz),
J . Org. Chem, Vol. 68, No. 2, 2003 385

Slowinski et al.
90.2 (J ) 10.1 Hz), 83.5 (5C), 71.6, 49.8 (J ) 80.9 Hz), 48.8
(J ) 8.1 Hz), 35.3 (J ) 62.7 Hz), 34.4, 27.0 (3C), 13.0 (J )
66.7 Hz); (minor) 94.6 (J ) 4.0 Hz), 90.4 (J ) 10.1 Hz), 83.5
(5C), 71.8, 49.8 (J ) 80.9 Hz), 47.7 (J ) 10.1 Hz), 35.3 (J )
62.7 Hz), 34.4, 27.1 (3C), 13.0 (J ) 66.7 Hz). 31P NMR (162
MHz, C6D6) δ: (major) 55.6; (minor) 55.6. Exo (characteristic
5.1 Hz), 3.42-2.93 (m), 2.80-2.49 (m), 2.40-0.86 (m); IR (C6D6)
2920, 2850, 1440, 1360, 1285, 1140 cm-1
.
Com p lex en d o-21. Yield ) 72%. Method A. Characteristic
data. 1H NMR (400 MHz, C6D6) δ: (major) 8.17-8.12 (m, 1H),
7.93-7.89 (m, 1H), 7.21 (m, 3H), 7.79 (s, 5H), 4.55 (d, J ) 4.6
Hz, 1H), 3.43 (d, J ) 12.7 Hz, 1H), 2.04 (tt, J ) 11.7, 2.6 Hz,
1H), 1.41 (d, J ) 14.8 Hz, 3H), 0.67 (qd, J ) 12.2, 3.0 Hz, 2H);
(minor) 8.07-8.03 (m, 1H), 7.93-7.89 (m, 1H), 7.21 (m, 3H),
4.58 (s, 5H), 4.27 (d, J ) 5.6 Hz, 1H), 3.34 (d, J ) 11.7 Hz,
1H), 2.07 (tt, J ) 11.0, 2.0 Hz, 1H), 1.82 (d, J ) 12.7 Hz, 3H),
0.44 (qd, J ) 11.7, 3.0 Hz, 2H). 13C NMR (100 MHz, C6D6) δ:
(major) 138.1 (J ) 95.0 Hz), 131.7, 131.5 (J ) 10.1 Hz), 130.7
(J ) 3.0 Hz), 128.5 (J ) 13.1 Hz), 128.2 (J ) 8.1 Hz), 83.3,
82.2 (5C), 77.6 (J ) 66.6 Hz), 77.5 (J ) 11.1 Hz), 77.1 (J ) 6.1
Hz), 49.2, 46.6, 41.0, 37.8, 33.8, 32.8, 31.2, 29.4, 27.9, 27.7,
15.8 (J ) 103.1 Hz); (minor) 133.6 (J ) 125.4 Hz), 131.9 (J )
9.1 Hz), 131.7, 130.8 (J ) 3.0 Hz), 128.3 (J ) 11.1 Hz), 127.9
(J ) 11.1 Hz), 83.3, 82.1 (5C), 77.9 (J ) 10.1 Hz), 77.5 (J )
60.7 Hz), 76.9 (J ) 10.1 Hz), 49.6, 46.6, 40.7, 38.1, 33.5, 32.8,
31.0, 30.4, 28.1, 27.6, 17.4 (J ) 72.3 Hz). 31P NMR (162 MHz,
C6D6) δ: (major) 32.3; (minor) 33.1. en d o-21 (unattributed):
1H NMR (400 MHz, C6D6) δ 1.93-0.95 (m, 28H); IR (C6D6)
1
data). H NMR (400 MHz, C6D6) δ: (major) 4.87 (s, 5H), 1.49
(d, J ) 11.2 Hz, 3H), 1.18 (d, J ) 13.2 Hz, 9H), 0.65 (m, 1H),
-0.32 (dd, J ) 12.7, 7.1 Hz, 1H); (minor) 4.89 (s, 5H), 1.45 (d,
J ) 11.2 Hz, 3H), 1.18 (d, J ) 13.2 Hz, 9H), 0.60 (m, 1H). 13C
NMR (100 MHz, C6D6) δ: (major) 97.0 (J ) 6.1 Hz), 89.1 (J )
12.1 Hz), 83.3 (5C), 72.4, 53.4 (J ) 80.9 Hz), 41.9 (J ) 10.1
Hz), 35.3 (J ) 62.7 Hz), 34.5, 26.9 (3C), 13.4 (J ) 62.7 Hz);
(minor) 94.2 (J ) 6.1 Hz), 89.8 (J ) 10.1 Hz), 83.0 (5C), 71.5,
53.2 (J ) 80.9 Hz), 40.1 (J ) 10.1 Hz), 35.3 (J ) 62.7 Hz),
34.6, 26.8 (3C), 13.4 (J ) 62.7 Hz). 31P NMR (162 MHz, C6D6)
δ: (major) 55.2; (minor) 55.2. (En d o + Exo) 20g (unattrib-
uted): 1H NMR (400 MHz, C6D6) δ 4.38 (s), 4.34 (s), 4.30 (t,
J ) 6.6 Hz), 4.08 (td, J ) 16.8, 6.6 Hz), 3.97 (m), 2.57-2.38
(m), 2.27-2.05 (m), 1.99-0.79 (m), 0.54-0.28 (m); 13C NMR
(100 MHz, C6D6) δ 36.8, 36.4 (3C), 33.9, 33.8, 33.5 (2C), 32.9
(2C), 29.1 (2C), 28.9 (2C), 28.3 (2C), 27.3 (2C), 26.8 (2C), 26.6
(2C), 26.4 (2C), 23.9 (2C), 23.8 (2C), 23.5 (2C), 23.4 (2C); IR
(C6D6) 2920, 2850, 1440, 1360, 1285 cm-1. Anal. Calcd for
2920, 2850, 1440, 1300, 1290, 1230, 1195 cm-1
.
C
24H36CoOP: C, 66.97; H, 8.43. Found: C, 66.89; H, 8.49.
Com p lexes 20h . Yield ) 29%. Method A. Very unstable
Ack n ow led gm en t. Financial support was provided
by CNRS and MRES.
compound. En d o (characteristic data). 1H NMR (400 MHz,
C6D6) δ: (major) 4.67 (s, 5H), 1.48 (d, J ) 12.7 Hz, 3H), 1.18
(d, J ) 15.8 Hz, 9H); (minor) 4.38 (s, 5H), 1.47 (d, J ) 12.2
Hz, 3H), 1.17 (d, J ) 16.8 Hz, 9H). 31P NMR (162 MHz, C6D6)
δ: (major) 59.0; (minor) 60.0. Exo (characteristic data). 1H
NMR (400 MHz, C6D6) δ: (major) 4.90 (s, 5H), 1.59 (d, J )
13.2 Hz, 3H), 1.26 (d, J ) 17.8 Hz, 9H); (minor) 4.57 (s, 5H),
1.34 (d, J ) 14.8 Hz, 3H), 1.24 (d, J ) 14.7 Hz, 9H). 31P NMR
(162 MHz, C6D6) δ: (major) 65.7; (minor) 62.0. (En d o + Exo)
20h (unattributed): 1H NMR (400 MHz, C6D6) δ 3.54 (t, J )
Su p p or tin g In for m a tion Ava ila ble: Procedure and full
description of compounds 2a , 2b, 4a , 6-12, (E)- and (Z)-2c,
13a , 13b, 14f, 14g, 15a -g; 1H and 13C NMR spectra for
compounds 2a , 4a , 13a , 13b, 15c, 15d ; and 31P NMR spectra
for 15c and 15d . This material is available free of charge via
the Internet at http://pubs.acs.org.
J O026212R
386 J . Org. Chem., Vol. 68, No. 2, 2003