The Hosomi-Sakurai allylation in hexafluoroisopropanol: Solvent promotion effect

Article abstract of DOI:10.1016/j.mencom.2011.07.019

Hydrogen bond donating capacity of hexafluoroisopropanol was shown to be responsible for both the formation of its stoichiometric complexes with aldehydes and acetals and subsequent facilitation of their reactions with allylsilanes.

Full text of DOI:10.1016/j.mencom.2011.07.019

Mendeleev
Communications
Mendeleev Commun., 2011, 21, 226–228
The Hosomi–Sakurai allylation in hexafluoroisopropanol:
solvent promotion effect
a
a
b
Svetlana P. Panchenko, Sofia A. Runichina and Vasily V. Tumanov*
a
b
Moscow Chemical Lyceum, 111033 Moscow, Russian Federation. Fax: +7 495 362 3440
N. D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, 119991 Moscow,
Russian Federation. Fax: +7 499 135 5328; e-mail: vasilii_tumanov@mail.ru
DOI: 10.1016/j.mencom.2011.07.019
Hydrogen bond donating capacity of hexafluoroisopropanol was shown to be responsible for both the formation of its stoichiometric
complexes with aldehydes and acetals and subsequent facilitation of their reactions with allylsilanes.
1
‡
Our recent investigation revealed that hexafluoroisopropanol
HFIP) while used as a solvent is capable of promoting such
to electrophile (E). In our further studies we disclosed that
(
dilution of the reaction mixture with DCM (1:1, v/v) resulted in
a dead stop of the reaction. Similarly a dramatic retardation effect
was produced by the decrease of HFIP/E ratio from 8:1 to 1:1.
Hence one may conclude that not only the presence and amount of
the inert co­solvent but the ratio of E/HFIP as well could critically
affect the efficiency of the overall reaction outcome.
These results prompted us to undertake a more detailed study
of HFIP activation effects with an ultimate goal to get an answer
to a major question: whether it is possible to draw a borderline
between two most obvious facets of the HFIP promoting action,
namely, its role as an initiator and as a medium?
important C–C couplings as Mukaiyama and Hosomi–Sakurai
reactions in the absence of any additional catalysts (see examples
in Scheme 1).
The Mukaiyama reaction
Me Si
3
O
O
O
OSiMe3
HFIP
+
CH2Cl2,
room
MeO
MeO
temperature
As a model we have chosen the Hosomi–Sakurai reaction of
aldehydes 1 and 2 and acetals 3 and 4 employed as electrophiles
(E) with allylsilanes 5 and 6 (Nu). Our methodology was based
The Hosomi–Sakurai reaction
OMe
Me
HFIP
OMe Me
+
1
13
SiMe3
CH Cl ,
on the use of H (and to a lesser extent C) NMR spectroscopy
2
2
Ph
OMe
room
Ph
to register first the changes brought by the dissolving of a chosen
temperature
§
E component in neat HFIP and then, after an introduction of the
Scheme 1
respective Nu, to monitor the progress of allylation reaction.
First of all we had to address the problem of hydrogen bond
formation in the system E/HFIP. Hydrogen bonding may result
in changes of spectral parameters of both electrophile and HFIP.
According to the literature data formation of hydrogen bonded
species via an interaction of HFIP with various acceptors could
be detected by significant downfield shift of hydroxyl proton of
2
According to an ample literature evidence the above type
reactions as well as a plethora of other classical electrophilic trans­
formations require the application of strong Lewis or Brønsted
acid as the initiators. Numerous mechanistic studies revealed that
the mechanism of these reactions involves the activation of an
electrophile via its transformation into the covalent reactive carbo­
5
HFIP, however, no such data were available for aldehydes and
3
cationic intermediate under the action of the above initiators. In
acetals.
1
view of these data the disclosed property of HFIP seemed to be
rather bizarre since this solvent has never been listed among
Lewis acids and its Brønsted acidity is known to be quite low
We have established that H NMR spectra pattern of aldehydes
1 and 2 and acetals 3 and 4 dissolved in the neat HFIP did not
reveal any significant changes as compared to the basic spectra
(
pK = 9.3, cf. pK = 4.8 for acetic acid).
of these compounds in CDCl . At the same time the hydroxyl
a
a
3
Previously we have shown that in HFIP solution neither
proton signal of HFIP underwent a rather significant downfield shift
upon the addition of E component. The maximum deshielding was
aldehydes nor acetals undergo conversion into the respective
1
hydroxy­ or alkoxycarbenium ion intermediates. Hence it was
suggested that the promotion effect of HFIP in the reactions shown
in Scheme 1 could be ascribed to the ability of this solvent to serve
as a powerful hydrogen bond donor toward various acceptors in
†
The latter additive was needed due to limited solubility of some sub­
strates in pure HFIP. The ‘ideal’ ratio of the co­solvents was established
as 4:1 (v/v), respectively.
4
‡
conjunction with its high total polarity and low nucleophilicity.
For example, an interaction of isobutyric aldehyde (E) with methallyl­
silane (Nu) (E/Nu = 1:1.2) in the above solvent system proceeded readily
at ambient temperature to furnish the expected allylation product in a nearly
quantitative yield.
The observed unusual profile of the HFIP promotion effect
and its promising synthetic potential most certainly warranted
additional studies. Herein we present the results of our current
investigation in this area.
§
The performing of NMR studies should be commented separately. In the
present work well­miscible with HFIP substrates were employed and NMR
monitoring was accomplished for ‘pure’ mixtures – only HFIP and sub­
strate(s) – in order to keep the system properties unchanged and to exclude
1
According to the protocol described in the cited paper all
†
reactions were carried out in of 4:1 (v/v) HFIP–DCM solvent
–
3
system with total concentration of each reactant ~1 mol dm .
Thus, in the most cases HFIP was employed in 8–10 equiv. excess
any side influence. Deuterated co­solvent (CDCl ) was sometimes added
3
in a minimal amount as a chemical shift standard.
©
2011 Mendeleev Communications. All rights reserved.
– 226 –

Mendeleev Commun., 2011, 21, 226–228
t
CF3
CF3
SiMe Bu
2
HFIP,
2 equiv.
Me
SiMe2But
O
Me
O
H
O
O
HFIP,
equiv.
Me
+
1
Me
room
temperature
Me
Me
Me
O
Me
1
5
Me
Me
1
equiv.
1 equiv.
1
Me
t
Bu Me2Si
O
CF3
CF3
Me
HFIP,
O
SiMe2But
2
equiv.
O
+
O
HFIP,
equiv.
room
temperature
Me
1
H
2
5
O
1
equiv.
1 equiv.
Me
Me
H
2
Scheme 3
Me
order to trigger these reactions. However much to our surprise, the
process stopped at 60% conversion and no changes were detected
in H NMR spectra pattern upon the storage of the mixture for
24 h. Full conversion of starting materials was achieved only after
additional amount of HFIP (up to 5:1:1 ratio) was introduced
OMe
OMe
Me
1
OMe
OMe
HFIP,
equiv.
Me
1
H
Me
O
Me
(Scheme 4).
F3C
CF3
3
OMe
HFIP,
MeO
OMe
Me
5
equiv.
MeO
Me
HFIP,
equiv.
Ph
OMe
+
OMe
OMe
1
room
temperature
H
Me
Me3Si
O
Me
Ph
3
6
4
F3C
CF3
1 equiv.
1 equiv.
Scheme 2
HFIP,
MeO
1
OMe
5
equiv.
observed for 1:1 HFIP/E mixture, its value (Dd = 1.6±0.1 ppm)
being rather close to the data reported previously for a number of
hydrogen bonded complexes of HFIP with various acceptors.
+
MeO
room
temperature
Ph
Me3Si
5
,6
Ph
4
6
equiv.
1 equiv.
It is also known that deshielding of carbonyl resonance signals
in 13C NMR spectrum might be a useful criteria for the charac­
terization of hydrogen­bonding interactions of carbonyl group
Scheme 4
with hydrogen bond donors.6 We have established that C NMR
,7
13
spectra of the solutions of 1–4 in HFIP in the solute/solvent ratio
Thus, we came across a rather paradoxical phenomenon,
1
:1 reveal a rather significant downfield shift of the carbonyl or
namely, the dramatic acceleration of bimolecular reaction with
the decrease of reactants concentration. The above data could be
accounted for by suggesting an involvement of several molecules
of HFIP in the formation of the rate­determining transition state of
the Hosomi–Sakurai reaction in this solvent.
acetal carbon signals. The magnitude of these effects, Dd = 6–7 ppm
for the carbonyl of aldehydes 1 and 2 and Dd = 3–4 ppm for
the acetal carbon of 3 and 4 is close to the values reported earlier
for the valerolactone and related carbonyl derivatives. Further
increase of HFIP amount does not lead to any changes in C NMR
spectra of these compounds.
Hence the observed deshielding effects are not due to some
solvent effects and could be taken as an evidence in favor of stoi­
chiometric interaction of HFIP with E in 1:1 ratio (see Scheme 2).
Next, we have studied the dependence of efficiency of allylation
6
13
Certainly, the detailed kinetic and computational studies of this
process are required in order to verify the advanced suggestions.
Meanwhile, an evidence attesting to its validity could be found
in the relevant results obtained by the Berkessel group which
refer to the dramatic accelerating effect on the rate of olefin
epoxidation by hydrogen peroxide caused by the presence of
1
6
8
reaction on the amount of HFIP employed using H NMR spectra
HFIP (up to 10 ­fold rate increase). The authors disclosed that
under these conditions a rate order 2–3 with respect to the con­
centration of HFIP is observed and hence concluded that ‘two
to three molecules of HFIP are involved in the rate­determining
to monitor the reaction course. Initial experiments disclosed that
no reaction occurred at ambient temperature upon the addition of
allylsilane 5 to the 1:1 mixture of any electrophile of the set 1–4
1
8(a)
with HFIP. Moreover, H NMR data clearly indicated that no reac­
step of the kinetically dominant reaction path’.
tion between 1 and 5 took place even upon the reflux of 1:1:1
mixture HFIP/E/Nu overnight. At the same time, addition of 1 more
equivalent of HFIP to this mixture resulted in a rather fast forma­
The data obtained in the present work allowed us to clearly
discriminate two major factors responsible for the observed effect
of HFIP as a ‘magic’solvent capable of promoting Hosomi–Sakurai
reaction and a number of other Friedel–Crafts like electrophilic
1
tion of the expected allylation product (according to H NMR data
1
reaction went to the completion within 2 h at ambient temperature,
see Scheme 3). Similar observations were made for the interaction
of 2 with 5.
On the contrary, in case of allylation of acetals 3 and 4 with
allylsilane 6 no reaction startup was observed in 2:1:1 mixture
HFIP/E/Nu under the same conditions. Further increase of HFIP
amount in the reaction mixture (up to 3:1:1 ratio) was required in
transformations under acid­free conditions. Firstly, HFIP serves
as an activator of the starting electrophile due to formation of 1:1
hydrogen bonded complex E­HFIP. Next, HFIP participates as
an additional stoichiometric component facilitating the reaction
of the above complex with a nucleophilic reactant. In this case
the polar transition state stabilization could be achieved by the
involvement of the conglomerate of several hydrogen­bonded
–
227 –

Mendeleev Commun., 2011, 21, 226–228
HFIP molecules in a mode similar to that suggested by Berkessel
et al.
Thus, we may conclude that the underlying cause of the observed
unique activity of HFIP is due to the multifunctional pattern of its
donating capacity. Besides the direct binding with an electrophile
3 H. Mayr and G. Gorath, J. Am. Chem. Soc., 1995, 117, 7862 and references
cited therein.
8
4
5
(a) J.­P. Bégué, D. Bonnet­Delpon and B. Crousse, Synlett, 2004, 18;
b) I. A. Shuklov, N. V. Dubrovina and A. Börner, Synthesis, 2007, 2925.
(a) K. F. Pursell, J. A. Stikeleather and S. D. Brunk, J. Am. Chem. Soc.,
969, 91, 4019; (b) W. J. Middleton and R. V. Lindsey, Jr., J. Am. Chem.
Soc., 1964, 86, 4948.
(
1
(conventionally designated as a static effect) HFIP exhibits even
more important capacity to serve as a bank of hydrogen bonds
available for the use when required (dynamic effect).
6 O. Coulembier, D. P. Sanders, A. Nelson, A. N. Hollenbeck, H. W. Horn,
J. E. Rice, M. Fujiwara, P. Dubois and J. L. Hedrick, Angew. Chem. Int. Ed.,
2
009, 48, 5170.
(a) A. Fratiello and C. S. Stover, J. Org. Chem., 1975, 40, 1244;
b) T. Asakura, Macromol. Chem., 1981, 182, 1135.
(a) A. Berkessel and J. A. Adrio, J. Am. Chem. Soc., 2006, 128, 13412;
b) A. Berkessel, J. A. Adrio, D. Hüttenhain and J. M. Neudörfl, J. Am.
Chem. Soc., 2006, 128, 8421.
7
8
This work was supported by the Russian Foundation for Basic
Research (grant no. 09­03­00666).
(
(
References
1
M. O. Ratnikov, V. V. Tumanov and W. A. Smit, Angew. Chem. Int. Ed.,
008, 47, 9739.
2
2
(a) M. Santelli and J.­M. Pons, Lewis Acids and Selectivity in Organic
Chemistry, CRS Press, Inc., Boca Raton, 1996; (b) A. G. Doyle and E. N.
Jacobsen, Chem. Rev., 2007, 107, 5713.
Received: 29th November 2010; Com. 10/3633
–
228 –