Gold-catalyzed efficient preparation of linear α-lodoenones from propargylic acetates

Article abstract of DOI:10.1021/ol070637o

Only 2 mol % of Au(PPh₃)NTf₂ is needed to convert readily accessible propargylic acetates into versatile linear α-iodoenones in good to excellent yields. This reaction is easy to execute and has broad substrate scope. Good to excellent Z-selectivities are observed in the cases of aliphatic propargylic acetates derived from aldehydes.

Full text of DOI:10.1021/ol070637o

ORGANIC
LETTERS
2007
Vol. 9, No. 11
2147-2150
Gold-Catalyzed Efficient Preparation of
Linear -Iodoenones from Propargylic
Acetates
r
Meng Yu, Guozhu Zhang, and Liming Zhang*
Department of Chemistry, UniVersity of NeVada, Reno, NeVada 89557
lzhang@chem.unr.edu
Received March 15, 2007
ABSTRACT
Only 2 mol % of Au(PPh3)NTf2 is needed to convert readily accessible propargylic acetates into versatile linear
r-iodoenones in good to
excellent yields. This reaction is easy to execute and has broad substrate scope. Good to excellent Z-selectivities are observed in the cases
of aliphatic propargylic acetates derived from aldehydes.
R-Iodoenones have become important synthetic intermediates
and can be readily converted into enones with various
R-carbon substituents, including alkyl, alkenyl, alkynyl, and
aryl groups, via transition metal-catalyzed cross-coupling
reactions.1
Various methods for the synthesis of R-iodoenones have
been developed. For example, they can be prepared from
enones using a combination of I2 and a nucleophilic base
(e.g., pyridine,2 DMAP,3 and quinuclidine3). However, this
Michael-addition-based method generally does not work well
with linear enones.4 In addition, the enone starting material
may not be readily available. Iodination of C-Si5 bonds with
ICl or C-Sn6 bonds with I2 can afford linear R-iodoenones
as well, although rather elaborate starting materials are
required. Interestingly, propargylic alcohols can be converted
into linear R-iodoenones by reacting with N-iodosuccinimide
(NIS) in the presence of a catalytic amount of hydroxy-
(tosyloxy)iodobenzene. Although this would be a versatile
method for R-iodoenone synthesis, it is limited only to
secondary alkynols, and moreover, only a few special
examples were reported.7 It is our conclusion that there is
still a lack of efficient synthetic methods for linear
R-iodoenones. Herein, we report a gold-catalyzed facile
synthesis of R-iodoenones from readily accessible propargylic
acetates.
Propargylic acetates can often be prepared in one pot from
aldehydes/ketones, terminal alkynes, and acetic anhydride
and have served as versatile substrates for Au catalysis.8,9
We have recently developed novel Au-catalyzed tandem
(1) For review, see: Negishi, E. J. Organomet. Chem. 1999, 576, 179-
194.
(2) (a) Johnson, C. R.; Adams, J. P.; Braun, M. P.; Senanayake, C. B.
W.; Wovkulich, P. M.; Uskokovic, M. R. Tetrahedron Lett. 1992, 33, 917-
918. (b) Sha, C. K.; Huang, S. J. Tetrahedron Lett. 1995, 36, 6927-6928.
(c) Djuardi, E.; Bovonsombat, P.; Nelis, E. M. Synth. Commun. 1997, 27,
2497-2503.
(3) Krafft, M. E.; Cran, J. W. Synlett 2005, 1263-1266.
(4) Reference 3 examined cases of linear enones using a combination of
I2 and DMAP. However, only a methyl group at the â-position was shown
to lead to good results. Indeed, in our hands, oct-6-en-5-one reacted very
sluggishly and only 30% conversion was observed in 2 days.
(5) Alimardanov, A.; Negishi, E. Tetrahedron Lett. 1999, 40, 3839-
3842.
(6) Bellina, F.; Carpita, A.; Ciucci, D.; Desantis, M.; Rossi, R.
Tetrahedron 1993, 49, 4677-4698.
(7) (a) Bovonsombat, P.; Mcnelis, E. Tetrahedron 1993, 49, 1525-1534.
(b) Angara, G. J.; Mcnelis, E. Tetrahedron Lett. 1991, 32, 2099-2100.
(8) For selected recent reviews, see: (a) Stephen, A.; Hashmi, A. S. K.;
Hutchings, G. J. Angew. Chem., Int. Ed. 2006, 45, 7896-7936. (b) Zhang,
L.; Sun, J.; Kozmin, S. A. AdV. Synth. Catal. 2006, 348, 2271-2296. (c)
Widenhoefer, R. A.; Han, X. Eur. J. Org. Chem. 2006, 4555-4563. (d)
Ma, S.; Yu, S.; Gu, Z. Angew. Chem., Int. Ed. 2006, 45, 200-203. (e)
Jimenez-Nunez, E.; Echavarren, A. M. Chem. Commun. 2007, 333-346.
(f) Patil, N. T.; Yamamoto, Y. ARKIVOC 2007, 6-19. (g) Marion, N.;
Nolan, S. P. Angew. Chem., Int. Ed. 2007, 46, 2750-2752.
10.1021/ol070637o CCC: $37.00
© 2007 American Chemical Society
Published on Web 04/28/2007

reactions of propargylic esters, proposed to involve 3,3-
rearrangement of propargylic esters and subsequent activation
of the in situ-generated carboxyallenes to form highly
reactive Au-containing oxocarbenium A (Scheme 1). This
Table 1. Gold-Catalyzed Reactions of Propargylic Acetate 1
with NIS to Form R-Iodoenone 2
Scheme 1. Proposed Synthesis of R-Iodoenones via
Au-Catalyzed Reactions of Propargylic Esters
yieldb
entrya
catalyst
conditions
(%)
Z/Ec
3:2
6:1
3:1
>99:1
>99:1
>99:1
45:1
1
2
3
4
5
2 mol % of Au(PPh3)NTf2 anhydrous acetone
2 mol % of Au(PPh3)NTf2 wet CH2Cl2
2 mol % of Au(PPh3)NTf2 MeNO2
95
85
78
35e
d
2 mol % of Au(PPh3)NTf2 acetone/H2O (40:1)
2 mol % of Au(PPh3)NTf2 acetone/H2O (200:1) 61
2 mol % of Au(PPh3)NTf2 acetone/H2O (400:1) 80
2 mol % of Au(PPh3)NTf2 acetone/H2O (800:1) 95f
6
7
8
9
10
2 mol % of AuIII g
2 mol % of PtCl2
no catalyst
ClCH2CH2Clh
toluenei
acetone/H2O (800:1)
95
6:1
12.5 >99:1
0j
a Reaction concentration was 0.05 M. b Estimated by 1H NMR using
diethyl phthalate as internal reference. c The geometries of enone 2 were
determined by NOESY 1D experiments. d Regular, without drying. e 17%
of acetate 1 was left unreacted. f 89% isolated yield. g Dichloro(pyridine-
2-carboxylato)gold(III). h Heating at 80 °C for 0.5 h. i Heating at 80 °C
for 2 h. j No reaction.
design led to the development of efficient synthetic methods
for highly functionalized 2,3-indoline-fused cyclobutane,10
cyclopentenones,11 R-alkylidene-â-diketones,12 and alkenyl
enol esters/carbonates.13 A notable observation in some of
these studies is that the Au-C(sp2) bond in A can react with
intramolecular electrophiles14 such as iminiums10 and acti-
vated acyl groups.12 We surmise that this Au-C(sp2) bond
could also react with electrophilic iodine intermolecularly,15
leading to efficient formation of R-iodoenones upon hy-
drolysis (Scheme 1). Notably, there is still much need to
explore the synthetic potential of nucleophilic organogolds.16
We began to examine the reaction of oct-3-yn-2-yl acetate
(1) and NIS in the presence of Au catalysts (Table 1). Stable
Au(PPh3)NTf2 was first used for its ease of preparation and
handling. Gratifyingly, with 2 mol % of the catalyst, the
desired reaction did happen, and R-iodoenone 2 was formed
in anhydrous acetone in excellent yield, although the Z/E-
selectivity was marginal (entry 1).17 Attempts to improve
the stereoselectivity met with limited success when different
solvents were used (entries 2 and 3). Interestingly, a small
amount of water in acetone seemed to enhance the Z/E-
selectivity dramatically, and only the Z-isomer of 2 was
observed when a mixture of acetone and H2O (40:1) was
used, although the yield of 2 was undesirably low (entry 4).18
Decreasing the amount of H2O, however, led to increased
yields of 2 without compromising the Z/E-selectivity (entries
5 and 6). Finally, when the ratio of H2O and acetone was
1:800 (about 1.4 equiv of H2O to 1), R-iodoenone 2 was
formed in 95% yield with excellent Z-selectivity (entry 7).
Although previous studies5,6 show that stereoisomerization
to the generally more stable Z-isomers of R-iodoenones
happened during reaction, we did not observe noticeable
change of Z/E ratio during and after the reaction at 0 °C.
Other Au catalysts (e.g., entry 8) and PtCl2 (entry 9) gave
less desirable results, and no reaction was observed without
any catalyst (entry 10).
(9) For selected examples of Au catalysis, see: (a) Fukuda, Y.; Utimoto,
K. J. Org. Chem. 1991, 56, 3729-3731. (b) Hashmi, A. S. K.; Schwarz,
L.; Choi, J.-H.; Frost, T. M. Angew. Chem., Int. Ed. 2000, 39, 2285-2288.
(c) Arcadi, A.; Di Giuseppe, S.; Marinelli, F.; Rossi, E. AdV. Synth. Catal.
2001, 343, 443-446. (d) Wei, L.; Li, C. J. J. Am. Chem. Soc. 2003, 125,
9584-9585. (e) Nieto-Oberhuber, C.; Munoz, M. P.; Bunuel, E.; Nevado,
C.; Cardenas, D. J.; Echavarren, A. M. Angew. Chem., Int. Ed. 2004, 43,
2402-2406. (f) Luzung, M. R.; Markham, J. P.; Toste, F. D. J. Am. Chem.
Soc. 2004, 126, 10858-10859. (g) Shi, Z.; He, C. J. Am. Chem. Soc. 2004,
126, 5964-5965. (h) Zhang, L.; Kozmin, S. A. J. Am. Chem. Soc. 2005,
127, 6962-6963. (i) Buzas, A.; Gagosz, F. J. Am. Chem. Soc. 2006, 128,
12614-12615. (j) Zhang, Z.; Liu, C.; Kinder, R. E.; Han, X.; Qian, H.;
Widenhoefer, R. A. J. Am. Chem. Soc. 2006, 128, 9066-9073. (k) Lin,
M.-Y.; Das, A.; Liu, R.-S. J. Am. Chem. Soc. 2006, 128, 9340-9341. (l)
Engel, D. A.; Dudley, G. B. Org. Lett. 2006, 8, 4027-4029.
(10) Zhang, L. J. Am. Chem. Soc. 2005, 127, 16804-16805.
(11) Zhang, L.; Wang, S. J. Am. Chem. Soc. 2006, 128, 1442-1443.
(12) Wang, S.; Zhang, L. J. Am. Chem. Soc. 2006, 128, 8414-8415.
(13) Wang, S. Z.; Zhang, L. M. Org. Lett. 2006, 8, 4585-4587.
(14) Alternative to the Au-C(sp2) bond attacking electrophiles directly,
electrophilic reaction with the C-C double bond followed by facile
elimination of Au is possible. For examples with such proposed mechanism,
see: (a) Nakamura, I.; Sato, T.; Yamamoto, Y. Angew. Chem., Int. Ed.
2006, 45, 4473-4475. (b) Jimenez-Nunez, E.; Claverie, C. K.; Nieto-
Oberhuber, C.; Echavarren, A. M. Angew. Chem., Int. Ed. 2006, 5452-
5455.
Using the optimized reaction conditions in Table 1, entry
7, the scope of this reaction was then studied. We first
examined propargylic esters derived from various aldehydes.
(16) For selected examples of Au-C bond nucleophilicity, see: (a) Yao,
X. Q.; Li, C.-J. Org. Lett. 2006, 8, 1953-1955. (b) Hashmi, A. S. K.; Frost,
T. M.; Bats, J. W. J. Am. Chem. Soc. 2000, 122, 11553-11554. (c) Shi,
Z.; He, C. J. Am. Chem. Soc. 2004, 126, 13596-13597. (d) Dube, P.; Toste,
F. D. J. Am. Chem. Soc. 2006, 128, 12062-12063. (e) Wang, S.; Zhang,
L. J. Am. Chem. Soc. 2006, 128, 14274-14275.
(17) The formation of 2 is presumably due to adventitious H2O in
anhydrous acetone or the succinimide anion as the nucleophile.
(18) The role of H2O in enhancing the stereoselectivity is not clear,
though it is likely related to the rate of hydrolysis (Scheme 1).
(15) Iodination of the Au-C(sp2) bond generated via other methods by
NIS is known, although only a few examples have been reported. For
selected references, see: (a) Buzas, A.; Gagosz, F. Org. Lett. 2006, 8, 515-
518. (b) Buzas, A.; Gagosz, F. Synlett 2006, 2727. (c) Kirsch, S. F.; Binder,
J. T.; Crone, B.; Duschek, A.; Haug, T. T.; Libert, C.; Menz, H. Angew.
Chem., Int. Ed. 2007, 46, 23102313.
2148
Org. Lett., Vol. 9, No. 11, 2007

3f) worked well, affording corresponding products in excel-
lent yields. Noteworthy is that no iodination of the benzene
ring in 3f was observed. Interestingly, ester 3h with
2-acetoxyethyl at the propargylic position yielded R-iodo-
dienone 4h only in its Z-form in 80% yield (entry 8). The
elimination of acetic acid in this reaction likely happened
after the formation of the iodoenone moiety. In contrast, ester
3i with one more CH2 spacer led to the expected product,
R-iodoenone 4i, in 83% yield. Substrates related to 4h and
4i with the terminal hydroxyl group protected by a TBS
group instead led to rather complicated mixtures, and crude
1H NMRs indicated that desilylation happened. While the
reaction conditions are considered mild, Au(PPh3)NTf2 seems
to be acidic enough to cause facile desilylation and elimina-
tion of acetic acid (entry 8).
Table 2. Reaction Scope for the Formation of
â-Monosubstituted R-Iodoenones
Extension of this chemistry to propargylic acetates derived
from ketones was delightfully successful. Hence, as shown
in Table 3, acetates prepared from linear ketones such as
Table 3. Reaction Scope for the Formation of
â,â-Disubstituted R-Iodoenones
a The substrate concentration was 0.05 M. b Isolated yield. c 10 mol %
of AgNTf2 was added, and the reaction time was 3 h. d The reaction took
3 h. e The reaction finished in 0.5 h. f 10 mol % of AgNTf2 was added.
g Elimination of acetic acid happened during the reaction.
As shown in Table 2, these substrates all reacted well, and
good to excellent yields of the corresponding R-iodoenones
with â-monosubstitution were obtained. When R1 and R2 are
both alkyl groups, high to excellent stereoselectivities
favoring the Z-isomer were observed (entries 1-3, 8, and
9). With a cyclohexyl group at the propargylic position (entry
3), the reaction was slow and incomplete even after extended
time. The addition of 10 mol % of AgNTf2 did help to drive
the reaction to completion, presumably due to its scavenging
of iodide and thus keeping Au(PPh3)NTf2 reactive.19 Notably,
in the cases of aryl-containing substrates 3d-g, the Z/E-
selectivities were diminished (entries 4-7); moreover, E-4e
was isolated as the major product. Substrates derived from
aromatic aldehydes with either electron-withdrawing (e.g.,
CF3 in 3g) or electron-donating substituents (i.e., MeO in
a The substrate concentration was 0.05 M. b Isolated yield. c Z/E ) 1:2.2.
acetone (i.e., 5a) and 3-methylbutan-2-one (i.e., 5e) led to
excellent yields of â,â-disubstituted R-iodoenones with
marginal stereoselectivity in the reaction of 5e (entry 2).20
Similarly, this reaction worked well with substrates derived
from cyclic ketones including cyclopentanone (entry 2),
cyclohexanone (entries 3, 6, and 7), and cycloheptanone
(entry 4). In addition, different substituents at the alkyne
terminus, including primary (entries 1-5) and secondary
(19) Indeed, precipitates, presumably AgI, were observed during the
reaction. Without the addition of AgNTf2, no precipitate formed.
Org. Lett., Vol. 9, No. 11, 2007
2149

(entry 6) alkyl groups as well as a phenyl group (entry 7),
were allowed. Remarkably, no elimination of acetic acid from
acetates 5 to form corresponding enynes was observed,
indicating that Au(PPh3)NTf2 prefers activating the C-C
triple bond to coordinating to the acetoxy group.
product but with opposite stereoselectivities. In order to gain
insights into the mechanism, reactions were run in an NMR
tube and the Z/E ratios were monitored by 1H NMR. Hence,
when propargylic ester 3c with all alkyl substituents was the
substrate, a constant Z/E ratio of 19:1 was observed during
the reaction, while the intermediate carboxyallene was not
detected. This supports path b, and the allene intermediate
presumably reacts fast with the AuI catalyst. Quite the
opposite, when ester 3e with a phenyl group was the
substrate, an opposite stereoselectivity with the E-isomer
slightly favored was detected invariably throughout the
reaction (Z/E ) 1:2). Interestingly, in this case, 90% of
substrate 3e was converted into its corresponding carboxy-
allene in 5 min, while the completion of the reaction took
30 min. Moreover, when a 1:1 mixture of 3e and its
corresponding carboxyallene23 was mixed with NIS without
the AuI catalyst, the carboxyallene reacted to yield (E)-4e
predominantly (Z/E ) 1:6 throughout the reaction) while 3e
remained unchanged. These observations suggest that 3e
reacted via both path a and path b with the former
presumably dominant. As a preliminary conclusion out of
these studies, aryl-containing esters such as 3d, 3e, 3f, and
3g may all undergo reactions via path a to a significant
extent, leading to diminished stereoselectivities; in contrast,
aliphatic propargylic esters yield the Z-isomers selectively
likely via path b predominantly.
Two possible competing mechanistic paths for this reaction
are proposed in Scheme 2 for the case of propargylic acetates
Scheme 2. Proposed Competing Mechanistic Paths for the
Formation of Iodoenone 4
In summary, an efficient synthesis of linear R-iodoenones
from readily accessible propargylic acetates was developed.
This reaction works well not only with substrates derived
from aldehydes but also with those from ketones. Good to
excellent Z-selectivities were observed in the case of aliphatic
propargylic acetates.
derived from aldehydes. Thus, carboxyallene 7, generated
via AuI-catalyzed 3,3-rearrangement, can either react with
NIS directly (path a) or be further activated by the AuI
catalyst to generate Au-containing cation A (path b). The
former path will lead to oxocarbenium B with the iodo group
trans to R1 (i.e., E-isomer) selectively due to the approach
of NIS toward the more electron-rich enolic C-C double
bond from the less hindered face (i.e., the face of H), while
the latter would afford C with the iodo group cis to R1 (i.e.,
Z-isomer) as our previous studies10-13 showed repeatedly a
cis relationship between Au(PPh3) and R1 in intermediate
A.21,22 If subsequent hydrolysis is facile and does not allow
stereoisomerization, these two paths will yield the same
Acknowledgment. This work is supported by the Uni-
versity of Nevada, Reno, and ACS PRF (#43905-G1).
Supporting Information Available: Experimental pro-
cedures and full spectroscopic data for all new compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
OL070637O
(21) Compared to the average C(sp3)-C(sp2) bond length (1.50 Å), the
AuI-C(sp2) bond is much longer (2.04 Å by averaging 8 hits from
Cambridge Structural Database), which may be the reason for the preference
for the cis relationship between Au(PPh3) and R1.
(22) Retention of double bond geometry was observed when alkenylgold
intermediates were iodinated with NIS. For references, see refs 15a and
15b.
(20) The stereoisomers of 6e are inseparable, and the assignment of 1H
NMR signals to the stereoisomers is based on the observation that the
Z-isomers in other cases consistently have chemical shifts at lower fields.
(23) Prepared by heating acetate 3e with 10 mol % of AgClO4 in
refluxing 2-butanone for 2 h.
2150
Org. Lett., Vol. 9, No. 11, 2007