Iron-catalyzed reaction of propargyl sulfides and trimethylsilyldiazomethane

Full text of DOI:10.1021/jo010247u

5256
J . Org. Chem. 2001, 66, 5256-5258
catalyzed reaction of propargyl sulfides with trimethyl-
silyldiazomethane.
Ir on -Ca ta lyzed Rea ction of P r op a r gyl
Su lfid es a n d Tr im eth ylsilyld ia zom eth a n e
Rosalind Prabharasuth and David L. Van Vranken*
Resu lts a n d Discu ssion
Department of Chemistry, University of California,
Irvine, California 92697
Syn t h esis of Su b st r a t es. Propargyl sulfides 1a -d
and 2 were synthesized by alkylation of the corresponding
thiolates in yields ranging from 58 to 94% (Scheme 2).
Sulfides 1a -d were prepared from propargyl bromides,
whereas substrate 2 was prepared from a propargyl
mesylate.9 The termini of 1a and 1b were functionalized
by deprotonation with 1.1 equiv of n-butyllithium and
addition of benzaldehyde at -78 °C.10 The resulting
propargyl alcohols 3a and 3b (not shown) were then
protected by acetylation. Propynoate 5 was prepared by
formation of the Grignard salt of alkyne 1b followed by
acylation with methyl chloroformate (Scheme 3).11 Since
xylyl derivative 1b lacked polar functionality, the conver-
sion to ester 4b was essential for chromatographic
purification of products of the iron-catalyzed reaction.
dlvanvra@uci.edu
Received March 7, 2001
In tr od u ction
Iron is the most abundant metal in the cosmos, yet
catalytic applications of iron in C-C bond formation are
rare. Some examples include Fischer-Tropsch chemistry,
ene carbocyclizations,1 cross-coupling reactions of Grig-
nard reagents,2 and some limited examples of cyclopro-
panation.3 Bach and co-workers have recently shown that
iron efficiently catalyzes the addition/[2,3]-rearrangement
of allyl sulfides and BocN3, leading to C-N bond forma-
tion.4 This process was later shown to be efficient with
allyl sulfides, iron catalysts, and trimethylsilyldi-
azomethane, leading to C-C bond formation (Scheme 1).5
Unlike reactions with traditional metal catalysts such
as rhodium, copper, cobalt, or palladium, reactions with
ferrous salts give good yields without syringe pump
addition of the diazo compound or a large excess of
reagents.
Sch em e 2
Sch em e 1
Sch em e 3
The reaction of allyl sulfides, iron salts, and Me3-
SiCHN2 generates R-silyl sulfides that can be unveiled
as aldehydes or used directly in Peterson olefination
reactions.6 An important property of silicon, the ability
to enhance the nucleophilicity of pi systems, is not
expressed in the homoallylsilane products. In contrast,
homoallenylsilanes, available from a [2,3]-rearrangement
(Scheme 1), are poised to take advantage of the silicon
effect.7 Copper and rhodium-catalyzed reactions of pro-
pargyl derivatives8 with diazomethane and R-diazoesters
have been previously reported; this work describes the
efficient formation of homoallenylsilanes by the iron-
Rea ction Con d ition s. Propargyl sulfide 1c was sub-
jected to the iron-catalyzed addition/rearrangement reac-
tion with Me3SiCHN2. To ensure reliable results in these
reactions, the propargyl sulfide substrate first had to be
heated for 1.5 h in 1,2-dichloroethane with the 5 mol %
catalyst precursor, typically dppeFeCl2, prior to addition
of Me3SiCHN2. Shorter incubation times may be effective,
but were not investigated. A 2.5 equiv portion of Me3-
SiCHN2 was added as a 2.0 M solution in hexanes, and
the reaction was stirred at reflux (3 h) until TLC
* To whom correspondence should be addressed. Fax: (949) 824-
8571.
(1) Takacs, J . M.; Weidner, J . J .; Newsome, P. W.; Takacs, B. E.;
Chidambaram, R.; Shoemaker, R. J . Org. Chem. 1995, 60, 3473.
(2) (a) Smith, R. S.; Kochi, J . K. J . Org. Chem. 1976, 41, 502. (b)
Cahiez, G.; Avedissian, H. Synthesis 1998, 1199. (c) Yanagisawa, A.;
Nomura, N.; Yamamoto, H. Synlett 1991, 513.
(3) (a) Seitz, W. J .; Saha, A. K.; Hossain, M. M. Organometallics
1993, 12, 2604. (b) Wolf, J . R.; Hamaker, C. G.; Djukic, J . P.; Kodadek,
T.; Woo, L. K. J . Am. Chem. Soc. 1995, 117, 9194.
(8) (a) Pourcelot, G.; L., V.; Cadiot, P. Bull. Soc. Chim. Fr. 1975,
1281. (b) Grieco, P. A.; Meyers, M.; Finkelhor, R. S. J . Org. Chem. 1974,
39, 119. (c) Doyle, M. P.; Bagheri, V.; Claxton, E. E. J . Chem. Soc.,
Chem. Commun. 1990, 46. (d) Kido, F.; Abiko, T.; Kato, M. Bull. Chem.
Soc. J pn. 1992, 65, 2471. (e) J ung, M. E.; Pontillo, J . Org. Lett. 1999,
1, 367.
(4) (a) Bach, T.; Ko¨rber, C. Tetrahedron Lett. 1998, 39, 5015. (b)
Bach, T.; Ko¨rber, C. J . Org. Chem. 2000, 65, 2358. (c) Bach, T.;
Schlummer, B.; Harms, K. Synlett 2000, 1330.
(5) Carter, D. S.; Van Vranken, D. L. Org. Lett. 2000, 2, 1303.
(6) Block, E.; Aslam, M. Tetrahedron 1988, 44, 281.
(9) Denmark, S. E.; Harmata, M. A.; White, K. S. J . Org. Chem.
1987, 52, 4031.
(10) Dai, W.-M.; Wu, J . L.; Fong, K. C.; Lee, M. Y. H.; Lau, C. W. J .
Org. Chem. 1999, 64, 5062.
(11) Buttery, J . H.; Moursounidis, J .; Wege, D. Aust. J . Chem. 1995,
48, 593.
(7) Hatakeyama, S. J . Syn. Org. Chem. J pn. 1997, 55, 793.
10.1021/jo010247u CCC: $20.00 © 2001 American Chemical Society
Published on Web 06/07/2001

Notes
J . Org. Chem., Vol. 66, No. 15, 2001 5257
Ta ble 1. Ir on -Ca ta lyzed of Rea ction w ith P r op a r gyl
Su lfid es a n d Me3SiCHN2
Sch em e 4
indicated complete consumption of starting material.
Under these conditions, the desired homoallenylsilane 6a
was isolated in 76% yield. An additional product, diene
6b, was obtained as a 2:1 mixture of E and Z isomers.
The stereochemistry of the Z isomer was confirmed by
the presence of a 15% NOE (2 s mixing time) from the
C2 methyl to the C1 proton. Diene 6b probably results
from a second addition of Me3SiCHN2 to the homoallenyl
sulfide moiety of 6a as depicted in Scheme 4. While a
third addition/rearrangement is possible, the correspond-
ing products were not formed in sufficient amounts to
isolate. Although the intermediates involved in this
catalytic reaction have not been established, they may
be analogous to intermediates involved in cyclopropana-
tion by CpFe(CO)2dCH2, except that the process de-
scribed in this work is catalytic in iron.12 When the
stoichiometry of Me3SiCHN2 was reduced from 2.5 to 1.3
equiv, the starting material was still completely con-
sumed and the yield of allene 6a increased to 85%; only
a trace of diene 6b was formed.
When the addition/rearrangement is carried out with
alkyne 1c using catalytic FeBr2, the allene 6a was
isolated in 65% yield. Even though phosphine ligand is
not required in the reaction there are two reasons to use
dppeFeCl2 rather than simple ferrous salts. First, FeBr2
is sufficiently hygroscopic that it requires handling under
argon or nitrogen. Second, simple ferrous halides have
low solubility in chlorinated solvents, even when the
thiother is present.
Su bstitu en t Effects. When the terminal alkyne 1b
was subjected to the rearrangement conditions, the
reaction stopped after about 1.5 h even though about 10%
of the starting material remained. Addition of an extra
0.25 equiv of Me3SiCHN2 led to consumption of the
remaining starting material. The desired product 7a was
isolated in 73% yield, accompanied by 11% of diene 7b
(not shown) resulting from a second addition/rearrange-
ment. Thus, the extra methyl in substrate 1c was
probably beneficial in minimizing diene formation. Ad-
ditionally, aryl thioethers should be weaker donors than
benzyl thioethers, yet the aryl thioether 1d gives a yield
similar to that of the benzyl thioether 1c.
but as 1:1 mixtures of diastereomers. The juxtaposition
of allylic acetate and allylsilane functional groups in
products 9 and 10 may make them useful for palladium-
catalyzed trimethylenemethane cycloaddition reactions.13
In contrast to 4a and 4b, the stereogenic center of
substrate 2 is proximal to the sulfur atom rather than
distal. However, reaction of substrate 2 proceeds with a
mere 2:1 diastereocontrol. As with substrate 1a , the
reaction stopped after 1.5 h leaving a substantial amount
of starting material. Addition of an extra 0.75 equiv. Me3-
SiCHN2 helped to consume the starting material, but
probably led to further consumption of the allene product.
Ultimately, chiral allene 11 was isolated in 48% yield.
Finally, the sensitive propynoate 5 was a poor substrate
for the reaction. Allene product 12 was isolated in only
18% yield from a mixture of products.
Sch em e 5
Formation of the initial ylide complex should play an
important role in diastereoselection. The sulfur lone pairs
of substrates 4a and 4b are diastereotopic, but are far
from the stereogenic center. The chiral acetates 4a and
4b both gave good yields of allenes 9 and 10, respectively,
Me3SiCHN2 is more efficient than ethyl diazoacetate
in the iron-catalyzed addition/rearrangement reaction
(Scheme 5). Dimerization is the primary competing
(12) (a) Brandt, S.; Helquist, P. J . Am. Chem. Soc. 1979, 101, 6473.
(b) Mccormick, F. B.; Gladysz, J . A. J . Organomet. Chem. 1981, 218,
C57.
(13) Trost, B. M. Angew. Chem., Int. Ed. Engl. 1986, 25, 1.

5258 J . Org. Chem., Vol. 66, No. 15, 2001
Notes
pathway that reduces the yield of allene product.5,14 The
bulk of Me3SiCHN2 makes dimerization less favorable
compared to sulfonium ylide formation. In contrast, the
smaller ethyl diazoacetate is more likely to form dimer.
(7.7 mL) was added dppeFeCl2 (0.020 g, 0.038 mmol), and the
resulting solution was heated to reflux. After 1.5 h, Me3SiCHN2
(0.48 mL, 0.96 mmol, 2.0 M solution in hexanes) was added, and
the solution was maintained at reflux for 1 h. The reaction
mixture was cooled to room temperature, filtered through a plug
of silica gel (80:20 ethyl acetate/hexanes), and concentrated in
vacuo to afford a brown oil. Purification by silica gel chroma-
tography (85:15 hexanes/benzene) gave allene 6a as a clear
colorless oil (0.190 g, 85%).
Con clu sion s
In summary, propargyl sulfides were shown to be
efficient partners for the iron-catalyzed addition/rear-
rangement reaction with Me3SiCHN2. Yields of the
allenyl R-silylsulfides were generally between 48 and
90%, except for propynoate 5, which gave a low yield.
Adding Me3SiCHN2 after premixing the catalyst with the
sulfide substrate gave significantly higher yields and
lower reaction times than comparable reactions without
premixing. While dppeFeCl2 was a convenient catalyst
precursor, phosphine ligand was not required for the
reaction. Larger substituents on the alkyne gave allene
products in higher yields, presumably because they
disfavor a second addition/rearrangement. Low diaste-
reoselection was observed when a stereogenic center was
adjacent to sulfur, while no diastereoselectivity was seen
when stereogenic centers were far. Yields of allene
products were enhanced when bulky Me3SiCHN2 was
used compared to the smaller ethyl diazoacetate.
6a : Rf ) 0.33 (85:15 hexanes/benzene); IR (thin film) 2954,
1951, 841 cm-1; 1H NMR (400 MHz, CDCl3) δ 7.21 (dd, J ) 8.8,
2.8 Hz, 2H), 6.82 (dd, J ) 8.8, 2.8 Hz, 2H), 4.66 (dqd, J ) 10.0,
3.2, 0.4 Hz, 1H), 4.62 (dqd, J ) 10.0, 3.2, 0.8 Hz, 1H), 3.80 (s,
3H), 3.60 (s, 2H), 2.59 (m, 1H), 1.74 (t, J ) 3.2 Hz, 3H), 0.04 (s,
9H); 13C NMR (125 MHz, CDCl3) δ 207.5, 158.4, 130.4, 130.2,
113.6, 97.6, 73.7, 55.2, 36.3, 35.6, 16.8, -2.0; LRMS (EI) m/z
(relative intensity) 292(29), 171(51), 121(99), 73(100); HRMS (CI/
NH3) m/z calcd for C16H24OSSi (M+) 292.1317, found 292.1320.
Anal. Calcd for C16H24OSSi: C, 65.70; H, 8.27. Found: C, 65.47;
H, 8.18.
Ack n ow led gm en t. We thank the National Science
Foundation (CHE 9623903). Additional support was
provided by Eli Lilly and the Dupont Aid to Education
Program. D.L.V.V. is a fellow of the Alfred P. Sloan
Foundation.
Exp er im en ta l Section
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedures, 1H NMR spectra, and other characterization data for
all new compounds. NOE spectra for stereochemical assign-
ment of dienes 6b. This material is available free of charge
via the Internet at http://pubs.acs.org.
Gen er a l P r ep a r a tion of Allen es. To a solution of propargyl
sulfide 1c (0.160 g, 0.77 mmol) in anhydrous 1,2-dichloroethane
(14) (a) Carter, D. S.; Van Vranken, D. L. Tetrahedron Lett. 1999,
40, 1617. (b) Aggarwal, V. K.; Ferrara, M.; Hainz, R.; Spey, S. E.
Tetrahedron Lett. 1999, 40, 8923.
J O010247U