Palladium-catalyzed C-allylation of benzoins and an NHC-catalyzed three component coupling derived thereof: Compatibility of NHC- and Pd-catalysts

Article abstract of DOI:10.1021/ol801644f

(Chemical Equation Presented) A large range of benzoins was successfully applied as C-nucleophiles in the palladium-catalyzed allylic alkylation with several allyl acetates, resulting in functionalized tertiary homoallylic alcohols. A number of unsymmetrical benzoins can be coupled with high levels of regio- and chemoselectivity. Finally, the challenging compatibility of free N-heterocyclic carbenes with a palladium catalyst has been utilized in a number of metal- and organocatalyzed three-component coupling reactions.

Full text of DOI:10.1021/ol801644f

ORGANIC
LETTERS
2008
Vol. 10, No. 19
4243-4246
Palladium-Catalyzed C-Allylation of
Benzoins and an NHC-Catalyzed Three
Component Coupling Derived Thereof:
Compatibility of NHC- and Pd-Catalysts
Raphae¨l Lebeuf,§ Keiichi Hirano,§ and Frank Glorius*
Organisch-Chemisches Institut der Westfa¨lische Wilhelms-UniVersita¨t Mu¨nster,
Corrensstrausse 40, 48149 Mu¨nster, Germany
glorius@uni-muenster.de
Received July 19, 2008
ABSTRACT
A large range of benzoins was successfully applied as C-nucleophiles in the palladium-catalyzed allylic alkylation with several allyl acetates,
resulting in functionalized tertiary homoallylic alcohols. A number of unsymmetrical benzoins can be coupled with high levels of regio- and
chemoselectivity. Finally, the challenging compatibility of free N-heterocyclic carbenes with a palladium catalyst has been utilized in a number
of metal- and organocatalyzed three-component coupling reactions.
The benzoin condensation is one of the oldest reactions in
organic chemistry, and early on,1 it has been catalyzed by
N-heterocyclic carbenes (NHCs). In the last couple of years,
NHCs have become popular organocatalysts for a number
of transformations.2,3 Many attempts have been made to
utilize benzoins as synthetic building blocks. Whereas
O-allylation without epimerization can be realized using
silver oxide as base,4 no enantiomerically pure C-allylated
product has been obtained directly from benzoin through
desymmetrization of its prochiral enolate.5 This is in stark
contrast to the achievements and the versatility of the
palladium-catalyzed asymmetric allylic alkylation.6 Soft
C-nucleophiles like malonates and ketoesters are especially
suitable nucleophiles, and we reasoned that benzoins, which
possess an acidic C-H bond, should exhibit similar properties.
Moreover, we envisaged that the combination of NHC-
catalyzed umpolung2,7 and palladium-catalyzed (asymmetric)
allylic alkylation would allow a three component coupling
and pave the way for new synthetic opportunities. In this
process, two successive C-C bond formations would result
in the formation of functionalized tertiary homoallylic
alcohols from simple aldehydes and allylacetates. However,
since NHCs also act as ligands and bind to metal complexes,
which can result in the poisoning of both catalysts, the
compatibility of a free NHC and a transition-metal catalyst
and their parallel functioning is difficult to put into practice.
§ These two authors have contributed equally to this work.
(1) Ukai, T.; Tanaka, R.; Dokawa, T. J. Pharm. Soc. Jpn. 1943, 63,
296–300.
(2) Excellent reviews of NHCs in organocatalysis: (a) Enders, D.;
Niemeier, O.; Henseler, A. Chem. ReV. 2007, 107, 5606–5655. (b) Marion,
N.; Díez-Gonzalez, S.; Nolan, S. P. Angew. Chem., Int. Ed. 2007, 46, 2988–
(5) C-allylation has been described on few examples: Heine, H.-G.
Liebigs Ann. Chem. 1970, 735, 56–64.
3000. (c) Enders, D.; Balensiefer, T. Acc. Chem. Res. 2004, 37, 534–541
.
(3) For applications of NHCs as ligands in catalysis, see: (a) N-
Heterocyclic Carbenes in Synthesis; Nolan, S. P., Ed.; Wiley-VCH:
Weinheim, Germany, 2006. (b) N-Heterocyclic Carbenes in Transition Metal
Catalysis; Glorius, F., Ed.; Topics in Organometallic Chemistry, Vol 21;
(6) (a) Trost, B. M.; Crawley, M. L. Chem. ReV. 2003, 103, 2921–2943.
(b) Lu, Z.; Ma, S. Angew. Chem., Int. Ed. 2008, 47, 258–297.
(7) (a) Hirano, K.; Piel, I.; Glorius, F. AdV. Synth. Catal. 2008, 350,
984–988. (b) Burstein, C.; Tschan, S.; Xie, X.; Glorius, F. Synthesis 2006,
2418–2439. (c) Burstein, C.; Glorius, F. Angew. Chem., Int. Ed. 2004, 43,
6205–6208. (d) Glorius, F.; Hirano, K. Ernst Schering Found. Symp. Proc.
Springer: Berlin, 2007
.
(4) Wildemann, H.; Du¨nkelmann, P.; Mu¨ller, M.; Schmidt, B. J. Org.
Chem. 2003, 68, 799–804.
2008, 2, 159–181
.
10.1021/ol801644f CCC: $40.75
Published on Web 09/03/2008
2008 American Chemical Society

Thus, our study commenced with the palladium-catalyzed
allylic alkylation of preformed benzoins.
improved the yield (entry 9). The same reaction in THF and
toluene12 led to 78 and 54% isolated yield respectively, with
excellent regioselectivity in both cases. Alternatively, DBU
was found to be an efficient base and fair alternative to KH
(Table 1, entry 10 and 11). However, the resulting reaction
is slower and more limited to electron-poor benzoins (Vide
infra). Under these conditions, the formation of the C-
allylated product through Claisen rearrangement from the
O-allylated species can not be excluded.13
Allylic Alkylation of Benzoins. The C-allylation of
benzoins would provide an attractive alternative to the
nucleophilic attack of an allylmetal species on the corre-
sponding diones.8 Interestingly, despite a report on the use
of benzoin as an O-nucleophile in palladium-catalyzed allylic
alkylation,9 no corresponding C-allylated compounds with
a free hydroxy group in R-position to a carbonyl function
have been described.10 Early on in our study on the allylation
of benzoin, we realized that the nature of the base and the
allyl source is critical for success (Table 1).11 Use of
With these conditions in hand, we screened a variety of
benzoins (Table 2). Excellent regioselectivities and good
Table 2. Allylation of Differently Substituted Benzoins
Table 1. Optimization of Reaction Conditions for the Allylation
of Benzoin (1)a
entry cat. base
4
t (°C), time GC-MS 1/2/3 yieldb 2/3
1
2
3
4
5
6
7
8
9
5ac
5ac
5ac
5ac
5ac
5a
no base 4a rt, 16 h
Cs2CO3 4a rt, 16 h
K3PO4 4b rt, 40 h
Cs2CO3 4b rt, 40 h
34:0:66
4:5:91
33:77:0
20:80:0
10:31:59
13:77:10
n.d.
n.d.
n.d.
n.d.
n.d.
NaH
KHd
KHd
4b rt, 24 h
n.d.
4b rt, 0.5 h
59% (89:11)
69%e (83:17)
81% (>95:5)
88% (>95:5)
70% (>95:5)
89% (>95:5)
5a
4b -78 to rt, 4 h
4b -78 to rt, 4 h
4bg -15, 1 h
4bg rt, 16 h
5b+6 KHd
n.d.
5b+6 KHf
n.d.
10 5b+6 DBU
n.d.
11h 5b+6 DBU
4bg rt, 16 h
n.d.
a Reaction conditions: benzoin (0.5 mmol), catalyst (1 mol %), ligand (2.5
mol %), base (1.1 equiv), allyl source (1.1 equiv), DCM (0.1 M). b Isolated
yield. c Five mol %. d KH and benzoin were premixed for 20 min and added
to the reaction mixture. e 17% of benzil was isolated. f KH was added last.
g Two equiv. h t-AmylOH as solvent (0.5 M).
a Generally, ratios of C- and O-allylated in the isolated fraction are the
same as in crude NMR, the products being inseparable. b Plus 4%
bisallylated product. c Ratio of 8/9 before purification: 77:23 (based on
NMR). d Plus 11% bisallylated product.
allylethylcarbonate (4a) without base (ethoxide is generated)
confirmed the reported tendency for O-allylation by using
this electrophile (entries 1,2). The use of allylacetate (4b)
results in the preferred formation of the C-allylated product
2.11 As the reaction rate was rather low with cesium
carbonate (entry 4), we tested stronger bases such as hydride,
and contrary to the use of NaH (entry 5), the desired isomer
2 was predominantly formed with KH (entries 6-9). At the
outset of our study, benzoin was stirred with KH for 20 min
to realize full deprotonation, and the resulting enolate was
added to the other reactants. The reaction proceeded very
quickly; however, the isolated yield of 59% (89:11 C/O ratio)
was not optimal yet. Lowering the temperature (entry 7) was
slightly beneficial, but the major advance was made by
changing the catalyst system. Several ligands were screened
and iPr-PHOX (6) was found to be the optimal one, providing
the product in very high selectivity of >95:5 (entry 8).
Finally, addition of the base at a later stage significantly
yields were obtained when electron withdrawing groups are
present and with five-membered heteroaromatics. Somewhat
more O-allylated products were observed in case of elec-
tronrich substituents and ortho substitution. The pyridine
derivative 7m was not found to provide any product.
Gratifyingly, more highly substituted allylacetates can also
be employed under similar conditions (Table 3). Reactions
are generally slower and in some cases do not occur at all at
(8) (a) Nishigaichi, Y.; Suzuki, A.; Takuwa, A. Tetrahedron Lett. 2006,
48, 211–214. (b) Sinha, P.; Roy, S. Organometallics 2004, 23, 67–71. (c)
Kang, S.-K.; Baik, T.-G.; Jiao, X.-H. Synth. Commun. 2002, 32, 75–78.
(d) Nair, V.; Jayan; C. N.; Ros, S. Tetrahedron 2001, 57, 9453–9459. (e)
Gewald, R.; Kira, M.; Sakurai, H. Synthesis 1996, 111–115. (f) Masuyama,
Y.; Tsunoda, T.; Kurusu, Y. Chem. Lett. 1989, 9, 1647–1650.
4244
Org. Lett., Vol. 10, No. 19, 2008

Table 3. Allylation of Benzoin using Substituted Allylacetates
Table 4. Allylation of Aryl-Alkyl Hydroxyketones
entry
12/12′
R1
R2
product
yield (%)
1
2
3
4
5
6
0/100
78/22
78/22
72/28
72/28
100/0
H
Me
Me
i-Pr
i-Pr
CHCH2
H
H
Ph
H
Ph
13a
13b
13c
13d
13e
13f
0
55
43a
58
52
0
H
a Mixture of 83:17 of C-allylated regioisomers.
of stable complexes.3 So far, only one NHC acting as an
organocatalyst has been shown to be compatible with a
transition metal, in an allylation/Stetter tandem reaction.16
Other catalysts involving allylpalladium species have been
used together with chiral phosphoric acids,17a secondary
amines17b,c for enamine formation from aldehydes, rhodium
for activation of R-ketonitrile,17d chiral phase transfer catalyst
for iminoester,17e stoichiometric amount of phosphines for
activation of enones,17f and boron for alcohol transfer.17g
Our study commenced with an investigation of different
types of NHCs (Figure 1). The goal was to find an NHC
a Ratio of 85:15 C/O-allylation. b With substrate 7k, dr ) 60:40; ee )
8% and 11%; with substrate 1: 64% yield, dr ) 57:43, ee not determined.
c Plus 41% mixture of other isomers.
-15 °C (TLC control). Therefore, the standard protocol
began at -15 °C but also included stirring at ambient
temperature overnight. Products were obtained in modest to
good yields.
Thus far, the enantioselectivities obtained for the products
of Tables 2 and 3 were quite low and did not exceed 26%
i
using the enantiomerically pure Pr-PHOX ligand. BINAP,
bisoxazolines and Trost’s ligand gave even lower yields and
enantioselectivities.
To get access to a larger variety of products, we next
investigated the use of unsymmetrical benzoin substrates,
which could lead to four possible isomers (Table 4). These
benzoins are readily available through published cross
benzoin condensations14 and exist as tautomeric mixtures
12/12′ as indicated in Table 4. The efficiency of the reaction
depends on the substituents. Whereas R-hydroxyacetophe-
none 12a and the R-hydroxyvinylketone 12f led to complete
degradation, the other alkyl-aryl hydroxyketones offer the
product regio- and chemoselectively in moderate yield. This
constitutes a more direct route to these interesting allyl
hydroxyketones.15
Figure 1. NHCs tested for tandem benzoin/allylation reactions.
that is catalytically active and that does not irreversibly bind
to the palladium catalyst, thereby poisoining both catalysts.
For this screening, we used a tertiary alcohol as solvent and
DBU, a standard base for carbene formation, since under
these conditions the allylation step was shown to work (Table
1, entry 11). We were pleased to find that thiazolidinylidenes
Having shown the feasibility of these benzoin R-allyla-
tions, we turned our attention to the tandem benzoin
formation/allylation catalysis.
Tandem Catalytic System. Althought tandem catalysis
allows reaching molecular complexity from simple substrates
in a one pot procedure, few examples are described in the
literature due to the problem of compatibility of many
catalytic systems. This is particularly true for the use of free
N-heterocyclic carbenes as organocatalysts in the presence
of another transition-metal based catalyst, since NHCs can
readily act as ligands for the metal and lead to the formation
(9) Schmidt, B.; Nave, S. AdV. Synth. Catal. 2006, 348, 531–537.
(10) (a) Moreno-Man˜as, M.; Pleixats, R.; Villaroya, M. J. Org. Chem.
1990, 55, 4925–4928. (b) Kuwano, R.; Ito, Y. J. Am. Chem. Soc. 1999,
121, 3236–3237. (c) Moorlag, H.; De Vries, J. G.; Kaptein, B.; Schoemaker,
H. E.; Kamphuis, J.; Kellogg, R. M. Rec. TraV. Chim. Pays-Bas 1992, 111,
129–137. (d) Evans, P. A.; Lawler, M. J. Am. Chem. Soc. 2004, 126, 8642–
8643. (e) Yan, X.-X.; Liang, C.-G.; Zhang, Y.; Hong, W.; Cao, B.-X.; Dai,
L.-X.; Hou, X.-L. Angew. Chem., Int. Ed. 2005, 44, 6544–6546.
Org. Lett., Vol. 10, No. 19, 2008
4245

are tolerant of the metal catalyst and proved to be the most
efficient (see Supporting Information). Recently, we have
introduced the thiazolium salt E for the diastereoselective
synthesis of trifluoromethylated γ-butyrolactones conjugate
umpolung.7a In our study, the two new derivatives thereof
F and G were found to be the most efficient organocatalysts,
being tolerant of the palladium cocatalyst.
Table 5. Tandem Benzoin/Allylation Reaction
Two optimal sets of conditions were identified by this
initial study. Electron-poor as well as electron-rich benzal-
dehyde derivatives could be reacted in this three component
coupling catalyzed in parallel by an NHC as organocatalyst
and a palladium complex (Table 5). In addition, heterocyclic
aldehydes like 2-furfuraldehyde and 2-thiophenecarbaldehyde
led to the products in good yield. In the case of electronrich
aromatics somewhat more forcing conditions like a higher
reaction temperature of 55 °C were required (condition B).
In all cases, only the C-allylation could be observed. This
compatibility of an NHC and a transition-metal catalyst
represents a major advance and should be useful for related
processes.
In conclusion, we have demonstrated the potential of
palladium-catalyzed allylic alkylation of benzoins and related
compounds. Remarkably high regio- and chemoselectivities
were obtained in many cases, whereas the enantioselectivity
of this reaction is still troublesome. Moreover, the rare
compatibility of NHC catalysts with a transition-metal
NHC’s
salt
isolated
yield
entry
aryl
C6H5-
p-F-C6H4-
p-Cl-C6H4-
condition
8
1
2
3
4
5
6
7
8
9
A
A
A
A
A
B
B
B
B
G
F
F
G
G
G
G
G
G
2
56%
76%
73%
83%
89%
69%
64%
76%
32%
8a
8b
8k
8l
8n
8c
8h
8d
2-furyl-
2-thiophenyl-
m-Me-C6H4-
p-Me-C6H4-
m-CH3O-C6H4-
p-CH3O-C6H4-
catalyst has been demonstrated in a multicomponent cou-
pling. The investigation of other dually catalyzed reactions
using NHCs is ongoing in our laboratory.
Acknowledgment. We thank Mrs. Karin Gottschalk
(WWU Mu¨nster) for skillful technical assistance and Mrs.
Isabel Piel (WWU Mu¨nster) for the synthesis of catalyst G.
Generous financial support by the Deutsche Forschungsge-
meinschaft (Priority program organocatalysis), the Fonds der
Chemischen Industrie, the Deutscher Akademischer Aus-
tausch Dienst (fellowship for K.H.) is gratefully acknowl-
edged. In addition, this research was also generously
supported by the Alfried Krupp Prize for Young University
Teachers of the Alfried Krupp von Bohlen and Halbach
Foundation.
(11) It is stated in ref 9 that the O-allylated product 3 decomposes
partially under GC conditions. Although this problem has to be considered,
we initially based our optimization on the results of GC-MS analysis (entries
1-6 of Table 1).
(12) Benzoin is almost insoluble in toluene, and the reaction time was
increased to two hours.
(13) Wood, J. L.; Moniz, G. A. Org. Lett. 1999, 1, 371–374.
(14) Stetter, H.; Da¨mbkes, G. Synthesis 1976, 403–404.
(15) (a) Trost, B. M.; Xu, J.; Reichle, M. J. Am. Chem. Soc. 2007, 129,
282–283. (b) Enholm, E. J.; Moran, K. M.; Whithley, P. E.; Battiste, M. A.
J. Am. Chem. Soc. 1998, 120, 3807–3808. (c) Soai, K.; Ishizaki, M. J. Org.
Chem. 1986, 51, 3290–3295.
(16) Nemoto, T.; Fukuda, T.; Hamada, Y. Tetrahedron Lett. 2006, 47,
4365–4368.
(17) (a) Mukherjee, S.; List, B. J. Am. Chem. Soc. 2007, 129, 11336–
11337. (b) Ibrahem, I.; Co´rdova, A. Angew. Chem., Int. Ed. 2006, 45, 1952–
1956. (c) Bihelovic, F.; Matovic, R.; Vulovic, B.; Saicic, R. N. Org. Lett.
2007, 9, 5063–5066. (d) Sawamura, M.; Sudoh, M.; Ito, Y. J. Am. Chem.
Soc. 1996, 118, 3309–3310. (e) Nakoji, M.; Kanayama, T.; Okino, T.;
Takemoto, Y. J. Org. Chem. 2002, 67, 7418–7423. (f) Jellerichs, B. G.;
Kong, J. R.; Krische, M. J. J. Am. Chem. Soc. 2003, 125, 7758–7759. (g)
Trost, B. M.; McEachern, E. J.; Toste, F. D. J. Am. Chem. Soc. 1998, 120,
12702–12703.
Supporting Information Available: Experimental details,
products characterization, and spectra of the new compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
OL801644F
4246
Org. Lett., Vol. 10, No. 19, 2008