Tsuji-Trost allylation of CH acids in supercritical carbon dioxide: Advantages and problems

Article abstract of DOI:10.1016/j.mencom.2013.03.009

The Pd(PPh₃)₄-catalyzed reaction between α-cyano or β-oxo carboxylates and allyl acetate in supercritical carbon dioxide with the K₂CO₃-18-crown-6 system as a base affords exhaustive allylation products in high yields, whereas malononitrile and acetylacetone of higher CH acidity form the mixtures of mono- and diallylated derivatives in moderate yields.

Full text of DOI:10.1016/j.mencom.2013.03.009

Mendeleev
Communications
Mendeleev Commun., 2013, 23, 84–85
Tsuji–Trost allylation of CH acids in supercritical
carbon dioxide: advantages and problems
Andrei A. Vasil’ev, Ilya V. Kuchurov and Sergei G. Zlotin*
N. D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, 119991 Moscow, Russian Federation.
Fax: +7 499 135 5328; e-mail: zlotin@ioc.ac.ru
DOI: 10.1016/j.mencom.2013.03.009
The Pd(PPh ) -catalyzed reaction between a-cyano or b-oxo carboxylates and allyl acetate in supercritical carbon dioxide with the
3
4
K CO –18-crown-6 system as a base affords exhaustive allylation products in high yields, whereas malononitrile and acetylacetone
2
3
of higher CH acidity form the mixtures of mono- and diallylated derivatives in moderate yields.
1
2
Previously, we have pioneered in the Tsuji–Trost allylation in
upercritical carbon dioxide (scCO ) using the asymmetric coupling
X
X
X
Y
i or ii
s
2
Y
Y
of dimethyl malonate and prochiral 3-acetoxy-1,3-diphenylpropene
as an example. The greatest challenge in such a processing was
to provide effective CH-acid deprotonation because traditional
deprotonating agents such as NaH and LDA (for full preliminary
CH-acid deprotonation) or the BSA–AcONa system (for gradual
deprotonation in the course of reaction) in organic solvents seem
incompatible with the electrophilic character of carbon dioxide.
Luckily, we used cesium carbonate for the gradual deprotonation
of dimethyl malonate in scCO thus allowing the coupling to
proceed well at 40–75°C and 110–240 atm (an optimum pressure
was 170 atm); the application of chiral ligands resulted in products
with ee up to 90%. However, a change from active 3-acetoxy-
,3-diphenylpropene to other allylating agents led to a low con-
1a–e
2a–e
3a–e
a X = Y = CO Et
2
b X = CO2Et, Y = C(O)Me
c X = CO2Et, Y = CºN
d X = Y = C(O)Me
e X = Y = CºN
Scheme 1 Reagents and conditions: i, AllOAc, [Pd], base, scCO (170 atm),
D, 18 h (Table 1); ii, AllOAc, Pd(PPh ) , K CO , 18-crown-6, CH Cl , room
temperature, 18 h.
2
2
3
4
2
3
2
2
1
transformed into their diallylated derivatives 3b,c in good yields
(Table 1, entries 3, 4, 7–9). Not only Cs CO but also much less
‡
1
2
3
version; this was likely due to the insufficient steady-state concen-
trations of both reaction species.
expensive K CO can be used as a base for this purpose. Note
2 3
that, in the case of cyanoacetate 1c, the 18-crown-6 additive was
not so crucial (Table 1, entries 8, 9) as in the case of acetoacetate
1b (entry 4). Lowering the temperature to 50°C resulted in
Here, we used the CH acids that are stronger than dialkyl
3
malonate 1a (pK ~ 16.4) and whose deprotonation with a hetero-
a
g
eneous base in almost nonpolar scCO occurs more readily, which
2
should increase the steady-state concentration of a carbanion.
Scheme 1 and Table 1 summarize the results obtained using allyl
acetate (AllOAc) and trivial Pd(PPh ) as a catalyst. Based on our
previous findings, the reaction pressure was 170 atm in all cases. At
Table 1 Allylation of CH acids 1a–e in supercritical CO .a
2
†
GC data (%)
Substrate
(pKa)
3
4
Entry
Base
T/°C
3
1
1
2
3
7
5°C, malonate 1a was predominantly monoallylated with Cs CO
2
3
1
2
3
4
5
6
7
8
9
1a (16.4) Cs CO3
75
75
12
100
–
84
–
4
–
2
used as a base, whereas no reaction occurred with a less soluble
system of K CO and a 18-crown-6 additive (Table 1, entries 1, 2).
The formation of product 2a can be rationalized based on essen-
1a
K CO /18-crown-6 75
2
3
2
3
1b (14.4) Cs CO3
–
100
100 (87)c
2
1b
1b
1b
K2CO3/18-crown-6b 75
–
–
3
tially greater values of pK (>18) for C-substituted malonates.
a
K CO /18-crown-6 50
–
80
52
–
20
2
3
As expected, ethyl acetoacetate 1b (pK ~ 14.4) and ethyl
a
K CO /18-crown-6 75d
–
48
2
3
cyanoacetate 1c (pK ~ 13.1) of higher CH acidity were readily
a
1c (13.1) Cs CO3
75
–
100
100 (91)c
2
e
1c
1c
1c
1c
K CO /18-crown-6 75
–
–
2
3
†
Allylation of CH acids in scCO . Catalyst Pd(PPh ) (23 mg, 0.02 mmol),
2
3 4
K CO3
75
–
–
100
2
CH acid 1a–e, 4 or 5 (1 mmol), allyl acetate (0.3 ml, ~3 mmol), K CO3
2
10
11
12
13
14
K CO /18-crown-6 50
44
3
35
44
21
2
3
(
345 mg, 2.5 mmol) and 18-crown-6 (13 mg) were placed in a 10 ml
K CO /18-crown-6 75d
53
2
3
autoclave open to air (in some experiments, other bases or catalysts were
used in equivalent amounts; when the CH acids were 4 and 5, the base
amount was halved). The vessel was filled with scCO to a total pressure
of 60 atm. The mixture was equilibrated at a reaction temperature (30 min);
1d (13.3) K CO /18-crown-6 75
–
(18)c (23)c
(6)c (22)c
2
3
1e (11.0) K CO /18-crown-6 75
–
2
3
2
1e
K CO /18-crown-6 75e
–
–
(36)c
2
3
then, additional CO was pumped to adjust a pressure of 75 atm. After this,
a
2
1 mmol of CH acid 1, 3 mmol of AllOAc, 2.5 mmol of a base, 2 mol%
the mixture was stirred for 18 h at a specified temperature. The vessel was
cooled and slowly depressurized. The autoclave contents were treated with
CH Cl , filtered through silica gel and concentrated. The residue was
Pd(PPh ) , 5 mol% 18-crown-6 (if taken), 10 ml autoclave, 170 atm scCO ,
3
4
2
18 h heating. b No reaction without 18-crown-6. c Isolated preparative yields
d
e
are given in parentheses. Catalyst, Pd dba /2 dppe. The same loading in a
2
2
2
3
analyzed by GC and/or subjected to column chromatography (silica gel,
gradient 0®6% EtOAc in light petroleum) to afford products 2, 3, 6, 7, 9
whose NMR-spectroscopic data were close to those reported earlier (see
Online Supplementary Materials).
2 ml autoclave.
‡
In cases of ethyl acetoacetate 1b and acetylacetone 1d, O-allylation
products were not detected.
©
2013 Mendeleev Communications. All rights reserved.
– 84 –

Mendeleev Commun., 2013, 23, 84–85
Me
incomplete conversion, and monoallylation products 2b,c were
detected (entries 5, 10). The catalyst prepared from Pd dba and
Me
CN
2
3
1c
Me
+
OAc
1
,2-bis(diphenylphosphino)ethane (dppe) in CH Cl was less
2 2
25% total
Me
CO Et
2
effective than Pd(PPh ) (entries 6, 11).
3
4
§
8
9
Similar reactions of acetylacetone 1d and malononitrile 1e
afforded the mixtures of mono- (2d,e) and diallylation (3d,e)
products in moderate total yields (Table 1, entries 12, 13). In these
cases, GC analysis could not be used to monitor the composition
of the mixtures because substrates 1d,e and monoallylated com-
pounds 2d,e were undetectable (moreover, they can occur as
potassium derivatives due to their high acidity). Therefore, the
preparative yields of isolated materials were used to characterize
the efficiency of these processes. The moderate yields of products
Me Me
Me
CN
CN
CN
+
+
Me
CO Et
2
Me CO Et
Me CO Et
2
2
9
'
9''
9'''
Scheme 3
2
d,e and 3d,e can be attributed to the lower nucleophilicity of 1d,e-
In the conclusion, the Tsuji–Trost reaction can be performed
and 2d,e-derived carbanions. This result allowed us to conclude
in scCO , although it turned to be very substrate-dependent under
2
that the dependence of the yield of an allylation product on the
studied conditions. However, the use of scCO in such a cross-
2
CH acidity of a XCH Y substrate approaches a maximum value
coupling can be promising in industry because the advantages
2
8
at pK 13–14: less acidic substrates are difficult to deprotonate
of scCO as a reaction medium are well-recognized.
a
2
while more acidic ones form low-nucleophilicity carbanions.
When malononitrile 1e was allylated in an autoclave of five times
smaller volume with the same loading of reactants (to diminish
dilution), the yield of product 3e was somewhat higher (entry 14),
however, it remained far from quantitative.
This work was supported by the Presidium of the Russian
Academy of Sciences (the basic research program PRAS-8).
Online Supplementary Materials
Supplementary data associated with this article can be found
in the online version at doi:10.1016/j.mencom.2013.03.009.
4
5
Substrates 4 (pK ~ 7.5) and 5 (pK ~ 10.15), which are
a
a
more acidic than 1d,e but have only one active proton, afforded
allylation derivatives 6, 7 in good yields (Scheme 2).
The prenylation of cyanoacetate 1c with 3-acetoxy-3-methyl-
References
but-1-ene 8 (Scheme 3) in scCO gave in a total yield of 25% a
2
1 (a) S. E. Lyubimov, I. V. Kuchurov, A. A. Vasil’ev, A. A. Tyutyunov, V. N.
Kalinin, V. A. Davankov and S. G. Zlotin, J. Organomet. Chem., 2009,
694, 3047; (b) S. E. Lyubimov, I. V. Kuchurov, A. A. Vasil’ev, S. G. Zlotin
and V. A. Davankov, Mendeleev Commun., 2010, 20, 143.
2 (a) B. M. Trost, Acc. Chem. Res., 1996, 29, 355; (b) B. M. Trost and D. L.
Van Vranken, Chem. Rev., 1996, 96, 395; (c) Z. Lu and S. Ma, Angew.
Chem. Int. Ed., 2008, 47, 258.
mixture of two expected normal (9) and iso (9') products and iso-
pentenyl analogues 9'' and 9''' bearing C H substituents of the
5
9
same skeleton with other positions of double bonds and linkage
1
sites (as judging from H NMR-spectroscopic and GC-MS data)
in a ratio of ~18:32:21:29. Similar isomerization within prenyl-
type p-allylic metal complexes leading to such by-products was
3
F. G. Bordwell, Acc. Chem. Res., 1988, 21, 456.
6
indicated earlier; it seems possible to occur at the elevated
4 I. A. Koppel, J. Koppel, V. Pihl, I. Leito, M. Mishima, V. M. Vlasov, L. M.
Yagupolskii and R. W. Taft, J. Chem. Soc., Perkin Trans. 2, 2000, 1125.
7(a)
temperature of our experiment. Previously, we have reported
5
6
V. H. Naringrekar and V. J. Stella, J. Pharm. Sci., 1990, 79, 138.
(a) T. Cuvigny, M. Julia and C. Rolando, J. Organomet. Chem., 1985, 285,
that the K CO -promoted prenylation of cyanoacetate 1c in DMF
2
3
at 20°C quantitatively afforded only isomers 9 and 9' in ratios
depending on the ligand applied.
Note that substrates 1a–e and 4 in dichloromethane at 20°C
give exhaustive allylation products in very good yields (for the
conditions, see ref. 7 and Online Supplementary Materials).
3
95; (b) T. Cuvigny and M. Julia, J. Organomet. Chem., 1986, 317, 383.
7
(a) A. A. Vasil’ev, S. E. Lyubimov, E. P. Serebryakov, V. A. Davankov,
M. I. Struchkova and S. G. Zlotin, Izv. Akad. Nauk, Ser. Khim., 2010, 592
(Russ. Chem. Bull., Int. Ed., 2010, 59, 605); (b) A. A. Vasil’ev, S. E.
Lyubimov, E. P. Serebryakov, V.A. Davankov and S. G. Zlotin, Mendeleev
Commun., 2009, 19, 103; (c)A.A.Vasil’ev,S.E.Lyubimov,E.P.Serebryakov,
V. A. Davankov and S. G. Zlotin, Mendeleev Commun., 2012, 22, 39.
CN
Ph
i
CN
Ph
8 (a) D. J. Cole-Hamilton, Adv. Synth. Catal., 2006, 348, 1341; (b) E. J.
Beckman, J. Supercrit. Fluids, 2004, 28, 121; (c) J. Durand, E. Teuma
and M. Gomez, C. R. Chimie, 2007, 10, 152; (d) T. Ikariya andY. Kayaki,
Catal. Surv. Jpn., 2000, 4, 39; (e) P. G. Jessop, J. Supercrit. Fluids, 2006,
8
4%
CO2Et
CO Et
2
4
6
3
8, 211.
O
O
i
CO2Et
6
4%
CO2Et
5
7
Scheme 2 Reagents and conditions: i,AllOAc, Pd(PPh ) , K CO , 18-crown-6
,
3
4
2
3
scCO (170 atm), 75°C, 18 h.
2
§
1
Judging from H NMR spectra, monoallylacetylacetone 2d in CDCl3
exists as a mixture of enol and keto forms in a 1:1 ratio.
Received: 24th January 2013; Com. 13/4055
–
85 –