DOI: 10.1002/chem.201503494
Communication
&
Carboxylation
The Renaissance of an Old Problem: Highly Regioselective
Carboxylation of 2-Alkynyl Bromides with Carbon Dioxide
Bukeyan Miao,[a] Gen Li,[b] and Shengming Ma*[b, c]
However, in zinc-mediated reactions, the selectivity control
Abstract: A steric effect-controlled, zinc-mediated carbox-
ylation of different 2-alkynyl bromides under an atmos-
pheric pressure of CO2 has been developed by careful
tuning of different reaction parameters, including the
metal, solvent, temperature, and additive. 2-Substituted
2,3-allenoic acids were afforded from primary 2-alkynyl
bromides, whereas the carboxylation of secondary 2-al-
kynyl bromides yielded 3-alkynoic acids in decent yields. A
rationale for the observed regioselectivity has been pro-
posed.
has not yet been established.[11a,13] Herein we report such
a carboxylation of 2-alkynyl bromides in the presence of CO2
and zinc with an excellent regioselectivity by applying the
steric effect of R2, demonstrating a completely different regio-
selectivity control to that with aldehydes or ketones (R1;
Scheme 1c).
Our initial attempt at the reaction of 2-heptynylic bromide
1a and indium powder with CO2 (provided by a balloon),
mediated with LiCl in THF led to a complicated product mix-
ture (Table 1, entry 1). Screening of other metal powders, in-
cluding Mn, Al and Bi, failed to yield either allenoic acid 2a or
3-alkynoic acid 3a with recovery of starting material 1a
(Table 1, entries 2–4). After some trial and error, we observed
that the reaction of 1a with CO2 mediated by zinc powder af-
forded 61% yield of allenoic acid 2a. However, 3-heptynoic
acid 3a was also formed in 17% yield, as confirmed by
1H NMR spectroscopic analysis of the crude product (Table 1,
entry 5). The reaction in DMF or DMSO gave lower yields and
selectivities to 2a/3a (Table 1, entries 6 and 7). Ethereal sol-
vents such as dioxane and methyl tert-butyl ether (MTBE) gave
a higher selectivity to 2a/3a, albeit with unsatisfactory yields
(Table 1, entries 8 and 9). Pleasingly, a dramatic solvent effect
was observed when dimethyl ether (DME) was used; allenoic
acid 2a was afforded in 73% yield with an allene/alkyne (2a/
3a) selectivity of 6:1 (Table 1, entry 10). Replacing LiCl with
LiOAc gave much lower yields (Table 1, entry 11). Finally, it was
exciting to observe that the ratio between allene 2a and
alkyne 3a was improved to >99:1 when the reaction was run
at 658C in the presence of LiCl, affording 2a in 71% yield
(Table 1, entries 12 and 13). The yield dropped to 59% in the
absence of LiCl at 658C, with a similar regioselectivity (Table 1,
entry 14).
Despite its extensive applications in constructing various differ-
ent kinds of heterocyclic compounds,[1–5] synthetic approaches
towards 2,3-allenoic acids are still underdeveloped. The estab-
lished approach is hydrolysis of 2,3-allenoates, but this method
suffers from poor atom- or step-economy, the use of toxic
carbon monoxide, and unsatisfying regioselectivity between
the allenoic acid and 3-alkynoic acid in some cases (Sche-
me 1a).[1c,6–8] We envisioned that the carboxylation of propar-
gylic derivatives with carbon dioxide would be the most
straightforward approach for the synthesis of allenoic acids.[9]
Interestingly, despite comprehensive studies on the reaction of
propargylic/allenylic metallic species with aldehydes and ke-
tones,[10,11] no careful study of this reaction with CO2 has, to
our knowledge, been reported to date.[12] For the reaction of
aldehydes or ketones, controlling the selectivity between alle-
nylic and propargylic alcohols is always a challenge; the struc-
tures of propargylic derivatives and carbonyl compounds code-
termined the selectivity.[10b,11b] In Al-, In-, and Ga-mediated reac-
tions, terminal propargylic bromides selectively afforded ho-
mopropargylic alcohols, whereas non-terminal 2-alkynyl bro-
mides selectively produced 2,3-allenyl alcohols (Scheme 1b).[10]
After this optimization, the carboxylation of various different
primary 2-alkynyl bromides was tested with typical results
given in Table 2. 2-Alkyl-substituted 2,3-allenoic acids 2a–d
were afforded exclusively in moderate to good yields (Table 2,
entries 1–4). The carboxylation reaction may be easily scaled
up to 100 mmol, affording 10 g of 2a in a yield of 71% with
a selectivity of >99:1, showing its practicality (Table 2, entry 1).
The introduction of a bulkier isobutyl group and a CH2CH2Ph
group did not affect the yield or selectivity (Table 2, entries 5
and 6). Synthetically useful functional groups, such as halides
(F and Cl) and OMe, were tolerated. Allyl or benzyl group-sub-
stituted 2-alkynyl bromides 1j and 1k also gave the corre-
sponding terminal allenoic acids 2j and 2k in moderate yields
under the optimized conditions (Table 2, entries 10 and 11).
[a] B. Miao
Shanghai Key Laboratory of Green Chemistry and Chemical Process
Department of Chemistry, East China Normal University
3663 North Zhongshan Lu, Shanghai 200062 (P. R. China)
[b] G. Li, Prof. Dr. S. Ma
Department of Chemistry, Fudan University
220 Handan Lu, Shanghai 200433 (P. R. China)
[c] Prof. Dr. S. Ma
State Key Laboratory of Organometallic Chemistry
Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences
345 Lingling Lu, Shanghai 200032 (P. R. China)
Supporting information for this article is available on the WWW under
Chem. Eur. J. 2015, 21, 17224 – 17228
17224
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Miao, Bukeyan
