Unexpected effect of the fluorine atom on the optimal ligand-to-palladium ratio in the enantioselective Pd-catalyzed allylation reaction of fluorinated enol carbonates

Article abstract of DOI:10.1039/b803097a

The enantioselective Pd-catalyzed allylation reaction of fluorinated allyl enol carbonates is presented; a key feature of this transformation is the important effect of the ligand-to-palladium ratio on the enantioselectivity of the α-fluoroketones, since using a ligand excess (L/Pd ratio = 1.25: 1) led to moderate results (30-76% ee), while using a L/Pd ratio <1: 1.67 (to as low as 1: 4) allowed the desired products to be obtained with high enantiopurity (up to 94% ee). The Royal Society of Chemistry.

Full text of DOI:10.1039/b803097a

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Unexpected effect of the fluorine atom on the optimal ligand-to-
palladium ratio in the enantioselective Pd-catalyzed allylation reaction of
fluorinated enol carbonatesw
´
Etienne Belanger, Clement Houze, Nicolas Guimond, Katy Cantin and Jean-Franc¸ ois Paquin*
´
´
´
Received (in College Park, MD, USA) 25th February 2008, Accepted 15th April 2008
First published as an Advance Article on the web 15th May 2008
DOI: 10.1039/b803097a
The enantioselective Pd-catalyzed allylation reaction of fluori-
nated allyl enol carbonates is presented; a key feature of this
transformation is the important effect of the ligand-to-palladium
ratio on the enantioselectivity of the a-fluoroketones, since using
a ligand excess (L/Pd ratio = 1.25 : 1) led to moderate results
(30–76% ee), while using a L/Pd ratio o1 : 1.67 (to as low as
1 : 4) allowed the desired products to be obtained with high
enantiopurity (up to 94% ee).
nates, where the L/M ratio has no influence on enantioselec-
tivity. Furthermore, this phenomenon is only observed with
fluorinated allyl enol carbonates and not with other fluori-
nated substrates (silyl enol ethers or b-keto allyl esters), thus
highlighting the differences among various enolate precursors.
ð1Þ
The development of novel methods to synthesize fluorinated
molecules is an active field of research due to the importance
of these compounds in medicinal chemistry and other related
fields.1 A practical method to generate complex fluorinated
compounds is to functionalize simple and readily available
fluorinated substrates. However, this approach is not always
trivial, since fluorinated substrates often show different and
sometimes unexpected reactivities compared to their non-
fluorinated analogues.2 As a result, reaction conditions devel-
oped for non-fluorinated substrates are not always optimal for
fluorinated examples.
We initially examined the reaction of allyl enol carbonate 1
catalyzed by Pd2(dba)3 (2.5 mol%) and ligand 28 (eqn (1)).
Using excess ligand (6.25 mol%, L/Pd = 1.25 : 1), the desired
product, 3, was obtained in 85% yield and low enantiomeric
excess (59% ee). Efforts to optimize the enantiopurity by
varying classic factors were unsuccessful.9 This was surprising,
since the successful allylation of non-fluorinated allyl enol
carbonates has been reported under similar conditions.6,10
A significant advance was made when probing the effect of
the ligand-to-palladium ratio (L/Pd, Fig. 1). While excess (i.e.
L/Pd 41 : 1) or equimolar (i.e. L/Pd = 1 : 1) amounts of
chiral ligand were detrimental to the enantioselectivity
(60–64% ee), using less than 1 equiv. of chiral ligand relative
to the metal (i.e. L/Pd = 1 : 1.33 or 1 : 1.1) resulted in an
improved enantiomeric excess (80–83% ee). High enantio-
selectivities (92–93% ee) were obtained by further decreasing
the L/Pd ratio.11,12 Interestingly, a L/Pd ratio as low as 1 : 4
Our group is interested in using fluorinated enolates as
nucleophiles in allylic alkylation reactions3,4 for the synthesis
of chiral organofluorine compounds. We have recently re-
ported the first enantioselective Pd-catalyzed allylation of
fluorinated silyl enol ethers.5,6 A variety of cyclic tertiary
a-fluoroketones1c,7 were prepared in good yields and with
excellent enantioselectivities (up to 95% ee). Following this
work, we became interested in exploring an alternative ap-
proach to similar products by using fluorinated enol carbo-
nates, such as 1, as fluorinated enolate precursors. Carbonates
would have a number of added benefits over the silyl enol
ethers, including increased stability and simplified reaction
conditions. Herein, we report an enantioselective Pd-catalyzed
allylation reaction of fluorinated allyl enol carbonates
(eqn (1)). We also report the discovery of an unusual metal-
catalyzed reaction where only a low ligand-to-metal ratio
(L/M o1 : 1.67) is necessary in order to obtain high enantio-
selectivities. The importance of the L/M ratio for fluorinated
carbonates drastically contrasts with non-fluorinated carbo-
Canada Research Chair in Organic and Medicinal Chemistry,
De´partement de chimie, Universite´ Laval, 1045 avenue de la
´
Medecine, Quebec, QC, Canada G1V 0A6. E-mail:
jean-francois.paquin@chm.ulaval.ca
´
w Electronic supplementary information (ESI) available: General ex-
perimental procedures, specific details of representative reactions, and
the isolation and spectroscopic information of the new compounds
prepared. See DOI: 10.1039/b803097a
Fig. 1 Effect of L/Pd ratio on the enantioselectivity in the allylation
of allyl enol carbonate 1 (cf. eqn (1)).
ꢀc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 3251–3253 | 3251

View Article Online
(i.e. 1.25 mol% ligand) could be used without affecting the
yield or enantioselectivity.13
Using these conditions, we next explored the scope of the
reaction. A variety of cyclic allyl enol carbonates were exam-
ined, including tetralone derivatives (Table 1, entries 1–7),
thiochromanone (entries 8–9), indanone (entries 10–11), and
benzosuberone (entries 12–13). It is interesting to note that, in
all cases, the yields of a-fluoroketones were similar, regardless
of the L/Pd ratio, whereas a substantial improvement in
the enantiomeric excess was observed with an L/Pd ratio
o1 : 1.67. For example, reducing the amount of chiral ligand
from 6.25 to 1.25 mol% for the reaction of a thiochromanone
carbonate (entries 8–9) resulted in an enantiomeric excess
improvement from 42 to 92% ee! Finally, the reaction of an
acyclic substrate (entries 14–15) resulted in a moderate
enantioselectivity regardless of the L/Pd ratio. This may be
related to the fact that the initially generated Z palladium
enolate rapidly equilibrates to an E/Z mixture prior to
allylation.
Scheme 1 Allylation of a non-fluorinated allyl enol carbonate.
While this reaction is quite general for a variety of cyclic
fluorinated allyl enol carbonates, we were intrigued to verify
whether the drastic effect of the L/Pd ratio on the enantio-
selectivity was specific to fluorinated substrates. We therefore
subjected methyl-substituted carbonate 4 to the reaction con-
ditions and varied the amount of chiral ligand (Scheme 1).14
Regardless of the L/Pd ratio, the desired ketone, 5, was
isolated in similar yields and with an identical enantiomeric
excess (92% ee), suggesting that the fluorine is playing a
Scheme 2 Allylation of alternative fluorinated enolate precursors.
critical role in the allylation of fluorinated allyl enol carbo-
nates.15
Having established the importance of both the fluorine
atom and the L/Pd ratio in this reaction, we next examined
the effect of the L/Pd ratios in reactions with other fluorinated
enolate precursors (which had previously employed a L/Pd
41 : 1 to achieve high enantiomeric excesses) (Scheme 2). We
first revisited the allylation of fluorinated silyl enol ethers.5
The reaction of 6 using a L/Pd of 1 : 4 afforded a-fluoroketone
3 with a similar result (79%, 91% ee) compared to the reaction
with an excess of ligand (L/Pd = 1.25 : 1). We then examined
the reaction of a-fluoro-b-keto ester 7, which is an isomeric
equivalent of fluorinated allyl enol carbonate 1.16 Using a
L/Pd ratio of 1 : 4 gave a lower yield of 3 compared to a L/Pd
ratio of 1.25 : 1, but a similarly high enantiomeric excess.
These results clearly highlight the fact that, unlike non-fluori-
nated enolate precursors, different fluorinated enolate precur-
sors have different reactivity patterns and should not be
considered equivalent.6,10,16a
Table 1 Reactions of various fluorinated allyl carbonatesa
Entry Substrate
Product
L/Pd
ratio
Result
Yield
(%)b
ee
(%)c
1
2
3
1 : 4
1 : 1.67 97
1.25 : 1 85
93
92
93
59
4
5
1 : 4 87
1.25 : 1 98
92
57
6
7
1 : 4 92
1.25 : 1 86
94
54
In conclusion, we have detailed the first enantioselective Pd-
catalyzed allylation reaction of fluorinated allyl enol carbo-
nates. In addition, we have discovered an important and
uncommon effect of the L/Pd ratio on the enantioselectivity,
whereby a L/Pd ratio o1 : 1.67 (and as low as 1 : 4) is
necessary in order to obtain the a-fluoroketones in high
enantiopurity (up to 94% ee). Although these conditions are
presently limited to cyclic fluorinated allyl enol carbonates, the
use of a L/Pd ratio o1 : 1 in other catalytic asymmetric
reactions would have tremendous benefits. Notably, chiral
ligands are often more expensive and less accessible than metal
catalysts, and therefore reducing the amount of ligand in any
given reaction may result in lower costs.17 Detailed mechan-
istic studies are currently under way to explain this unusual
finding.
8
9
1 : 4 90
1.25 : 1 93
92
42
10
11
1 : 1.67 91
1.25 : 1 90
82
30
12
13
1 : 4 83
1.25 : 1 87
88
76
14
15
1 : 1.67 58
1.25 : 1 75
34
36
a
5 mol% Pd (2.5 mol% of Pd2(dba)3) was used in all the reactions,
along with the appropriate amount of ligand (see ESI for details).
This work was supported by the Canada Research Chair
Program, the Natural Sciences and Engineering Research
Council of Canada, the Canada Foundation for Innovation,
b
c
Isolated yield. Determined by chiral HPLC.
ꢀc
This journal is The Royal Society of Chemistry 2008
3252 | Chem. Commun., 2008, 3251–3253

View Article Online
Merck Frosst Centre for Therapeutic Research and the
Universite Laval. OmegaChem Inc. is thanked for a generous
8. (a) G. Helmchen and A. Pfaltz, Acc. Chem. Res., 2000, 33, 336;
(b) J. M. Williams, Synlett, 1996, 705.
´
9. Factors studied include: solvents (THF, dioxane, CH2Cl2,
CH3CN), chiral ligands ((R,R)-Trost ligand ((1R,2R)-(+)-1,2-
diaminocyclohexane-N,N0-bis(20-diphenylphosphinobenzoyl)),
(R)-(S)-JOSIPHOS ((R)-(ꢁ)-1-[(S)-2-(diphenylphosphino)ferroce-
gift of L-tert-leucine.
Notes and references
nyl]ethyldicyclohexylphosphine)
(R)-BINAP),
temperature
1. (a) T. Hiyama, in Organofluorine Compounds: Chemistry
and Applications, ed. H Yamamoto, Springer-Verlag, Berlin,
Heidelberg, New York, 2000, pp. 1–23 and pp. 137–182;
(b) A. M. Thayer, Chem. Eng. News, 2006, 84, 15; (c) J.-A. Ma
and D. Cahard, Chem. Rev., 2004, 104, 6119; (d) C. Bobbio and
V. Gouverneur, Org. Biomol. Chem., 2006, 4, 2065; (e) K. L.
(RT–40 1C) and concentration (0.033–0.1 M).
10. (a) B. M. Trost and J. Xu, J. Am. Chem. Soc., 2005, 127,
2846; (b) B. M. Trost and J. Xu, J. Am. Chem. Soc., 2005, 127,
17180.
11. For other examples showing the importance of the L/Pd ratio on
the enantioselectivity in asymmetric allylic alkylation reactions,
see: (a) X. Cheng and K. K. Hii, Tetrahedron: Asymmetry, 2003,
14, 2045; (b) D.-R. Hou, J. H. Reibenspies and K. Burgess, J. Org.
Chem., 2001, 66, 206; (c) A. M. Porte, J. Reibenspies and
K. Burgess, J. Am. Chem. Soc., 1998, 120, 9180; (d) K. Burgess
and A. M. Porte, Tetrahedron: Asymmetry, 1998, 9, 2465.
12. For examples showing the importance of the L/M ratio on the
enantioselectivity in other transformations, see: (a) R. Rasappan,
M. Hager, A. Gissibl and O. Reiser, Org. Lett., 2006, 8, 6099;
(b) G. C. Lloyd-Jones and C. P. Butts, Tetrahedron, 1998, 54, 901.
13. All the reactions reported in Fig. 1 afforded 3 in good yield
(473%). Low conversion (o20%) was observed when a L/Pd
ratio of o1 : 4 was used, whereas no product was detected without
the chiral ligand being present.
J. Kirk, J. Fluorine Chem., 2006, 127, 1013; (f) J.-P. Begue and
´ ´
D. J. Bonnet-Delpon, J. Fluorine Chem., 2006, 127, 992;
(g) S. Purser, P. R. Moore, S. Swallow and V. Gouverneur, Chem.
Soc. Rev., 2008, 37, 320.
2. For example, see: (a) M. Marhold, A. Buer, H. Hiemstra,
J. H. van Maarseveen and G. Haufe, Tetrahedron Lett., 2004,
45, 57; (b) F. Grein, A. C. Chen, D. Edwards and C. M. Crudden,
J. Org. Chem., 2006, 71, 861; (c) S. Kobayashi, H. Tanaka,
H. Amii and K. Uneyama, Tetrahedron, 2003, 59, 1547;
(d) C. M. Crudden, A. C. Chen and L. A. Calhoun, Angew.
Chem., Int. Ed., 2000, 39, 2851; (e) T. Ernet, A. H. Maulitz,
E.-U. Wurthwein and G. J. Haufe, J. Chem. Soc., Perkin Trans. 1,
¨
2001, 1929; (f) W. Heitz and A. Knebelkamp, Macromol. Rapid
Commun., 1991, 12, 69; (g) S. J. Getty and W. T. J. Borden, J. Am.
´
Chem. Soc., 1991, 113, 4334; (h) J. Banus, H. J. Emeleus and R. N.
14. The reaction of 4 using an excess of ligand under similar condi-
tions (Pd2(dba)3 (2.5 mol%), ligand 2 (6.25 mol%), THF (0.033
M), 10 1C, 2 h) has been reported previously, and the desired
product, 5, was obtained with comparable results (87%, 91% ee).
See ref. 6.
J. Haszeldine, J. Chem. Soc., 1951, 60.
3. For reviews on asymmetric allylic alkylation reactions, see:
(a) J.-F. Paquin and M. Lautens, in Comprehensive Asymmetric
Catalysis: Supplement 2, ed. E. N. Jacobsen, A Pfaltz and
15. Other non-fluorinated substrates were also examined. The allyl
enol carbonate derived from a-tetralone was reacted under
our conditions (L/Pd = 1 : 4 and L/Pd = 1.25 : 1). However,
none of the desired product was isolated in either case.
Interestingly, when using a L/Pd ratio of 1 : 4, only starting
material was recovered; when a L/Pd ratio of 1.25 : 1 was used, a
mixture of various side-products were observed, with only traces
of the desired product being detected. This transformation has
been previously conducted using an excess of a chiral ligand (see
ref. 10a). In addition, the allyl enol carbonate derived from
2-chloro-1-tetralone was also tested. In this case, a small increase
in enantioselectivity was observed from L/Pd = 1.25 : 1 (34% ee)
to L/Pd = 1 : 4 (49% ee).
16. (a) E. C. Burger, B. R. Barron and J. A. Tunge, Synlett, 2006,
2854; (b) J. T. Mohr, D. C. Behenna, A. M. Harned and
B. M. Stoltz, Angew. Chem., Int. Ed., 2005, 44, 6924;
(c) M. Nakamura, A. Hajra, K. Endo and E. Nakamura, Angew.
Chem., Int. Ed., 2005, 44, 7248.
17. For example, in the PHOX ligand series (phosphinooxazolines), only
i-Pr-PHOX is commercially available: (S)-t-Bu-PHOX has to be
synthesized in four steps from L-tert-leucine (see ref. 8). For our
system, and if we use i-Pr-PHOX for the sake of comparison, the
ligand is seven times more expensive than the metal: (S)-i-Pr-PHOX
(578 US$ gꢁ1 ꢂ 216US$ mmolꢁ1, 2007 Strem catalogue) vs.
Pd2(dba)3 (162 US$ gꢁ1 ꢂ 30 US$ mmolꢁ1, 2007 Strem catalogue).
H
Yamamoto, Springer-Verlag, Berlin, 2004, pp. 73–95;
(b) A. Pfaltz and M. Lautens, in Comprehensive Asymmetric
Catalysis, ed. E. N. Jacobsen, A Pfaltz, H Yamamoto, Springer,
New York, 1999, vol. 2, pp. 833–884; (c) B. M. J. Trost, J. Org.
Chem., 2004, 69, 5813; (d) B. M. Trost and C. Lee, in Catalytic
Asymmetric Synthesis, ed. I Ojima, Wiley-VCH, New York, 2nd
edn, 2000, pp. 593–649; (e) G. J. Helmchen, J. Organomet. Chem.,
1999, 576, 203; (f) B. M. Trost, Chem. Pharm. Bull., 2002, 50, 1;
(g) B. M. Trost and D. L. Van Vranken, Chem. Rev., 1996, 96,
395; (h) B. M. Trost, Acc. Chem. Res., 1996, 29, 355.
4. For a recent review on the use of pre-formed enolates in Pd-
catalyzed allylic alkylation reactions, see: M. Braun and T. Meier,
Synlett, 2006, 661.
´
5. E. Belanger, K. Cantin, O. Messe, M. Tremblay and J.-F. Paquin,
´
J. Am. Chem. Soc., 2007, 129, 1034.
6. For a related reaction of non-fluorinated substrates using similar
conditions, see: (a) D. C. Behenna and B. M. Stoltz, J. Am. Chem.
Soc., 2004, 126, 15044; (b) J. A. Keith, D. C. Behenna, J. T. Mohr,
S. Ma, S. C. Marinescu, J. Oxgaard, B. M. Stoltz and
W. A. Goddard III, J. Am. Chem. Soc., 2007, 129, 11876.
7. (a) V. A. Brunet and D. O’Hagan, Angew. Chem., Int. Ed., 2008,
47, 1179; (b) G. K. S. Prakash and P. Beier, Angew. Chem., Int.
Ed., 2006, 45, 2172; (c) P. M. Pihko, Angew. Chem., Int. Ed., 2006,
45, 544.
ꢀc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 3251–3253 | 3253