Table 2 Synthesis of γ-hydroxy-α,β-acetylenic esters starting from
2,2,6,6-tetramethylpiperidine with MeLi in the same 2-MeTHF
prior to its transfer via cannula to the reaction flask: due to the
high stability of organolithiums in 2-MeTHF,34 the 1.2 M sol-
ution of LTMP could be used without any detrimental effect.
When the reaction was performed at −20 °C a significant
decrease of the yield was observed (entry 17). This temperature
effect may be related to the well known lability of the intermedi-
ate propiolate nucleophile.2 The use of 2-MeTHF was crucial for
the success of the procedure, leading to best results compared to
both toluene (entry 21) and other ether-type solvents (entries
18–20). Under the optimized reaction conditions, THF afforded
the desired alcohol 2a in only 67% yield, and moreover the reac-
tion crude was contaminated by the presence of impurities prob-
ably formed by the lithiation processes on THF. In fact, as
previously reported, among the several advantages that 2-
MeTHF shows compared to THF, there is the possibility to
perform organometallic reactions at higher temperatures in the
former, without observing decomposition of the solvent, as a
consequence of its improved stability towards organolithiums.34
It is worth mentioning the straight relationship existing among
base employed, solvent and temperature: in fact, as reported very
recently by Eisch and co-workers47 the treatment of benzal-
dehyde with LTMP in hexane at 0 °C affords benzyl benzoate as
a consequence of carbene formation (generated by the C–H
lithiation of the formyl group), followed by the addition of this
carbene to a second molecule of benzaldehyde. In contrast, by
using our conditions (2-MeTHF at −40 °C) no lithiation of ben-
zaldehyde was observed whatsoever, and thus the procedure
shows an excellent chemoselective profile by directing the action
of LTMP exclusively on the 2-bromoacrylic ester. Furthermore,
no other side-reactions (metal–halogen exchange, addition to the
electrophilic alkene) were observed. This latter point was funda-
mental for the development of our synthetic strategy: in fact, as
observed by Marino and Linderman48 such substrates may
undergo lithiation–Cu-transmetallation reactions, or may afford
cyclopropanes as a consequence of the addition of the organo-
metallic (Li or Mg) to 2 equiv. of 2-haloacrylic ester.49 Further-
more, the use of non-organometallic bases such as lithium tert-
butoxide did not afford any product even at 0 °C (entry 22).
Thus, these preliminary results of our investigations confirm
the effectiveness of replacing normally used ether-type solvents
like THF with its greener substitute 2-MeTHF. In particular, we
were delighted to confirm its excellent stability in a high basic
environment at temperatures that were not prohibitive (<−78 °C)
for industrial purposes. Remarkably, it is now recognized as a
non-genotoxic and non-mutagenic solvent,50 so that it is ideal
for scale-up processes.51 The use of 2-MeTHF enormously
improves the work-up of the reactions; in fact, its limited solubi-
lity in water allows the recovery of the reaction product without
adding any other organic solvent. In the procedure described
herein, we used 2-MeTHF distilled under simple atmospheric
conditions, thus avoiding the use of hazardous methodologies
based on ketyl formation (Na or K, benzophenone).
different alkyl 2-haloacrylic estersa
Entry
X
R
Reaction time [h]
Productb [%]
1
2
3
Br
Br
Br
Cl
Cl
Et (1b)
tBu (1c)
Bn (1d)
Me (1e)
Me (1e)
1.5
2.5
1.5
6
2b (93)
2c (91)
2d (94)
2a (49)
2a (54)
4
5c
6
a Conditions: alkyl-2-haloacrylate (1 equiv.); LTMP (2 equiv.). b Isolated
yields. c Reaction run at −20 °C.
In fact, our system showed significant superiority compared to
previously described protocols: in fact, Villeneuve and Tam
could not obtain more than 45% using LiHMDS in the deproto-
nation of ethyl propiolate at −70 °C.52 Remarkably, the dehydro-
halogenation–addition reaction dramatically decreased when
using the corresponding methyl 2-chloroacrylate because of the
lower tendency of a chloroalkene to undergo an elimination reac-
tion, even running the reaction at a higher temperature (entries
4–5).
Subsequently, a series of different aldehydes were used to
check the scope of the reaction on the aldehyde side. As shown
in Table 3, aromatic aldehydes bearing electron withdrawing
groups gave the best yields in the shortest reaction times (entries
1–4, 7–8). In fact, our procedure led to better yields compared to
previous reported methods; for example, the use of silver acety-
lide afforded compound 4a in only 61% yield,53 while our proto-
col gave 93% after only 1 h.
On the other hand, the presence of an electron donating group
(entry 5) increased the reaction times up to 6 h, but no loss of
efficiency of the protocol was observed even in the presence of a
sterically demanding aldehyde (entry 6). Again, the method-
ology confirmed its chemoselectivity also in the presence of a
bis-electrophilic substrate, providing compound 4b in 94% yield,
without any detection of the product derived by the attack of the
transient propiolate on the carboxylic ester (entry 2). The system
performed very well with heteroaromatic aldehydes, thus
expanding the substrate scope (entries 9–10). Moreover, the pro-
tocol can also be used with a non-enolizable aliphatic aldehyde,
although in this case a longer reaction time was required (entry
11).
To conclude our study on the applicability of this strategy, we
turned our attention to an α,β-unsaturated aldehyde like cinna-
maldehyde: the first attempt under our classical conditions
afforded 4l in only 67% yield after 12 h. Pleasingly, as shown in
entry 12, we could improve the yield up to 95% simply by
adding LiBr (1.50 equiv.) to the reaction mixture. As previously
reported by our group,54 this task strongly enhances the attack of
an organolithium on the carbonylic carbon of this type of
aldehyde.
With the optimized conditions in hand, we turned our atten-
tion to different 2-bromoacrylate esters to evaluate the scope of
the reaction. As shown in Table 2, the protocol is widely appli-
cable to such esters giving excellent yields in all cases, including
those substrates bearing either a sterically hindered tert-butyl
ester or a benzyl ester (entries 2–3).
A possible mechanistic pathway that accounts for the observed
reaction is shown in Scheme 2. Thus, the first equivalent of
This journal is © The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 1859–1863 | 1861