Hydrogen-bond-assisted activation of allylic alcohols for palladium-catalyzed coupling reactions

Article abstract of DOI:10.1002/cssc.201300723

We report direct activation of allylic alcohols using a hydrogen-bond-assisted palladium catalyst and use this for alkylation and amination reactions. The novel catalyst comprises a palladium complex based on a functionalized monodentate phosphoramidite ligand in combination with urea additives and affords linear alkylated and aminated allylic products selectively. Detailed kinetic analysis show that oxidative addition of the allyl alcohol is the rate-determining step, which is facilitated by hydrogen bonds between the alcohol, the ligand functional group, and the additional urea additive. Hydrogen Bond Rule(s): Direct activation of allylic alcohols and subsequent alkylation and amination reactions are reported. The new catalyst is based on functionalized palladium and phosphoramidite ligands to allow hydrogen bond-assisted activation. Kinetic data are in line with this mechanism as the oxidative addition is the rate-determining step.

Full text of DOI:10.1002/cssc.201300723

CHEMSUSCHEM
FULL PAPERS
DOI: 10.1002/cssc.201300723
Hydrogen-Bond-Assisted Activation of Allylic Alcohols for
Palladium-Catalyzed Coupling Reactions
Yasemin Gumrukcu, Bas de Bruin, and Joost N. H. Reek*[a]
We report direct activation of allylic alcohols using a hydro-
gen-bond-assisted palladium catalyst and use this for alkylation
and amination reactions. The novel catalyst comprises a palladi-
um complex based on a functionalized monodentate phos-
phoramidite ligand in combination with urea additives and af-
fords linear alkylated and aminated allylic products selectively.
Detailed kinetic analysis show that oxidative addition of the
allyl alcohol is the rate-determining step, which is facilitated by
hydrogen bonds between the alcohol, the ligand functional
group, and the additional urea additive.
Introduction
Palladium-catalyzed allylic substitution reactions are efficient
and widely used synthetic methods in organic synthesis for
CÀC and CÀX (X: O,N,S) bond formation.[1] Until recently, acti-
vated allylic reagents that are substituted with carboxylates,[2]
halides,[3] or phosphates[4] have been mainly used as activated
allyl alcohol reagents (Scheme 1a); however, these reagents re-
quire an additional synthetic step compared to the use of allyl-
ic alcohols and the conversion results in the formation of stoi-
chiometric amounts of waste (salts). In this context, direct con-
version of allyl alcohols is an important challenge. In addition,
numerous renewable compounds are available that could be
converted into interesting intermediates for the fine-chemical
and pharmaceutical industry by direct substitution of the allylic
alcohol. A commonly used approach is preactivation of allylic
alcohols with a stoichiometric or catalytic amount of a Lewis-
The development of a waste-free direct catalytic transforma-
tion of allyl alcohol was initially reported by Ozawa et al. They
developed (p-allyl)palladium complexes bearing sp2-hybridized
phosphorus ligands capable of direct conversion of allylic alco-
hols without any activators, producing water as the only by-
product.[11] Afterwards, several Pd catalysts were reported, for
various related applications. For example Tamaru and co-work-
ers focused on allylic alkylation reactions with indole deriva-
tives[7] while Beller and co-workers broadened the scope for
primary aliphatic amines and electron deficient heterocycles.[12]
Notably, the catalytic systems have different methods to acti-
vate allylic alcohols. Sarkar reported that Pd complexes with
bisphosphine ligands are successful due to the high p-acidity
of the phosphorous ligands,[13a] whereas Oshima proposed that
hydrogen-bonding between water and allylic alcohols is key in
decreasing the activation energy barrier of these processes.[13b]
The latter report caught our attention as such supramolecular
interactions have recently become a more general tool to con-
trol transition-metal-catalyzed reactions. Supramolecular inter-
actions are powerful tools to generate catalysts in situ by self-
assembly of functionalized ligand building blocks,[14] prevent-
ing intricate ligand synthesis, and utilize H-bond interactions
for proper orientation of the substrate via the functional
groups to tune the activity and selectivity of the catalyst.[15]
The latter strategy was recently also applied in metalloradical
carbene transfer reactions using supramolecular H-bond donor
appended porphyrin complexes.[16] When functional ligand
building blocks are used for the self-assembly of bidentate li-
gands, these functional groups can also interact with function-
al groups of the substrate. Breit and co-workers employed bi-
dentate ligands based on self-assembly strategies using phos-
phine ligands with complementary H-bonding motifs, and they
found that these functional groups activate allyl alcohols by
hydrogen bonding.[17] We reported simple ligand building
blocks that formed heterobidentate ligand complexes through
a single hydrogen bond and determined that additional func-
tional groups in the ligand give rise to H-bonding with the
substrate, an interaction that appeared crucial in the enantio-
acid activator such as As2O3,[5] B2O3,[6] BEt3,[7] BPh3,[8] SnCl2,[9] or
[10]
Ti(O-iPr)4
(Scheme 1b), but this method is still associated
with substantial waste formation in most cases.
Scheme 1. General nucleophilic substitution reactions at allylic reagents (a),
and the direct activation of allylic alcohols with the aid of additives (b).
[a] Y. Gumrukcu, Prof. Dr. B. de Bruin, Prof. Dr. J. N. H. Reek
Van’t Hoff Institute for Molecular Sciences
University of Amsterdam
Science Park 904, 1098 XH, Amsterdam (The Netherlands)
Fax: (+31)20-525-5604
E-mail: j.n.h.reek@uva.nl
Supporting Information for this article is available on the WWW under
http://dx.doi.org/10.1002/cssc.201300723.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2014, 7, 890 – 896 890

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
Table 1. Catalyst optimization for allylic alkylation reactions.[a]
Entry
Ligand
Yield[b] [%]
1
2
3
4
5
6
7
8
1
65(3)[d]
53(3)[d]
21
1/phenyl urea[c]
2
2/phenyl urea[c]
30
3
trace
trace
62
3/phenyl urea[c]
4
Scheme 2. Activation of allylic alcohols with H-bond donor ligand and 1,3-di-
ethylurea.
4/phenyl urea[c]
90
selective hydrogenation of Roche ester precursors.[18] Inspired
by these results we decided to study these simple ligand build-
ing blocks in the activation of allyl alcohols.
[a] Reaction conditions: 5a (0.128 mmol), 6a (0.128 mmol), 3 mol%
[(n3-allyl)Pd(cod)]BF4, 6 mol% ligand, toluene (0.1 mL), 808C, 20 h.
[b] Yields are determined via 1H NMR where mesitylane used as an exter-
nal standard. [c] 3 mol% phenyl urea. [d] Bis-cinnamylether formed as
a side product.
Herein, we report a novel catalyst that operates through co-
operative action of the metal and H-bonding interactions be-
tween the ligand and the substrate (Scheme 2). We also ex-
plored the effect of adding urea moieties as additives to fur-
ther steer the activity and selectivity of the catalyst. Thus far
palladium-catalyzed allylic alkylation reactions with monoden-
tate phosphoramidites have been limited to activated allylic
halides or acetates as substrates, where they display high
enantio- and regioselectivity.[19] This is the first report wherein
phosphoramidite-based palladium complexes are used in the
direct conversion of allyl alcohols. We here disclose that a com-
bination of phosphoramidite homocomplexes and urea build-
ing blocks provides an efficient and selective catalyst system
for direct activation of allylic alcohols towards linear alkylated
and aminated products by enabling complementary H-bond-
ing between the catalyst, the substrate, and the urea moiety.
First, we examined commercially available monophos, 1, and
triphenylphosphine, 2, ligands for catalytic activity. The stron-
ger p-accepting monophos enabled higher conversions than
PPh3 (Table 1, entries 1 and 3). The addition of phenyl urea
hardly affected the product yields when using homo-com-
plexes of ligand 1 and 2 (entries 2 and 4). Additionally, we ob-
served formation of bisdicinnamylether as a side product up to
3% when monophos 1 was applied as the ligand. The ana-
logues ligand with the H-bond donor, isophos ligand 3, did
not show any activity, regardless of the presence of urea (en-
tries 5 and 6). Interestingly, the amino-acid-substituted phos-
phoramidite 4 that has both H-bond-donor and -acceptor
groups, resulted in 62% yield of the product without forma-
tion of any side products (entry 7). Remarkably, in the presence
of phenyl urea as additive the complex based on this ligand
(4) led to full conversion and 90% yield of the single linear al-
kylated product (entry 8). Notably, phenyl urea as additive only
shows increased conversions if a ligand with functional groups
is used. This suggests that the functional groups of the ligand
form a more complex hydrogen-bond pattern with the urea
and the substrate, which leads to the activation of the allylic
alcohol (Scheme 2).
Results and Discussion
To study the concept of allyl alcohol activation with hydrogen
bonds the alkylation of cinnamyl alcohol with indole was stud-
ied as a model reaction. Simple phosphoramidite and phos-
phine ligands were explored as ligands to form the palladium
based
catalyst
by
in situ
mixing
with
[(h3-allyl)Pd(cod)]BF4 (Scheme 3). The ligands varied in electron-
ic properties (p-accepting and electron-donating abilities), 3
has a hydrogen-bond donor and in ligand 4 both hydrogen-
bond-accepting and -donating groups are present. Reactions
were carried out in the absence and presence of 3 mol%
phenyl urea as a potential additional H-bond donor for activa-
tion of the alcohol.
After having established a working protocol based on
simple building blocks, we carried out CÀC bond formation re-
actions for a series of indole derivatives with various allylic al-
cohols (Table 2). For this series of reactions 1,3-diethylurea was
used as additive instead of phenyl urea, as 1,3-diethylurea has
a better solubility. We have ex-
plored several urea derivatives
with variable substituents, and it
was found that 1,3-diethylurea
gave rise to the highest activity
(see Supporting Information).
With these optimized reaction
conditions, the catalyst system
(ligand 4+1,3-diethylurea) was
Scheme 3. Phosphoramidite- and phosphine-type of ligands used for catalyst optimization.
able to convert a wide variety of
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2014, 7, 890 – 896 891

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
to functional groups at the indole derivatives. N-Methyl, 5b,
C2-methyl, 5c, and C5-methoxy, 5d, substituted indole deriva-
tives selectively produced linear alkylated products in high
yields (entries 2–4). Notably, C3-methyl substituted indole, 5e,
did not form any allylation product (entry 5), indicating that
the reaction is selective for C3-allylation products. The sub-
strate scope was further extended for various aliphatic allylic
alcohols that showed moderate isolated yields. Most likely, the
Table 2. Allylic alkylation reactions of indole derivatives with various allyl-
ic alcohols.[a]
Entry
Nucleophile
Substrate
Product
Yield[b]
[%]
Table 3. Allylic amination reactions of primary and secondary amines
with various allylic alcohols.[a]
1
5a
6a
7aa
42
2
3
5b
5c
6a
6a
7ba
7ca
69
77
Entry
1
Nucleophile
Substrate
Product
Yield[b] [%]
8a
6a
8aa
78(7)[c]
4
5
5d
5e
6a
6a
7da
7ea
78
2
3
4
5
8b
8c
8d
8e
6a
6a
6a
6a
8ba
8ca
8da
8ea
74(8)[c]
76
n.r.
95
6
7
5a
5a
6b
6c
7ab
7ac
7ad
97
22
87
6
7
8 f
6a
6a
8 fa
55
85
8g
8ga
8ad
8
9
5a
5a
6d
6e
7ae
26 (9:3:1)[c]
19 (9:3:1)[c]
7ad–7ae
8
9
8a
8a
6d
6e
8ae
8ad–8ae
55 (9:1:1)[d]
64 (9:1:1)[d]
10
5a
6 f
7af
21
10
11
8a
8a
6g
6h
8ag
8ag
61
54
[a] Reaction conditions: 5a (0.5 mmol), 6a (0.75 mmol), 3 mol%
[(h3-allyl)Pd(cod)]BF4, 3 mol% 1,3-diethylurea, 6 mol% ligand 4, toluene
(2.5 mL), 808C, 20 h. [b] Isolated yields. [c] The ratio of the linear product
to the branched and bis-allylated ones.
12
8a
6 f
8af
84
substrates in high yields and with high selectivity for the linear
alkylated products. Primary and secondary aromatic allylic alco-
hols 6a and 6b afforded C3-selective allylated indoles up to
97% yield (entries 1 and 6). The reaction has a high tolerance
[a] Reaction conditions: 8a (0.5 mmol), 6a (0.75 mmol), 3 mol%
[(h3-allyl)Pd(cod)]BF4, 3 mol% 1,3-diethylurea, 6 mol% ligand 4, toluene
(2.5 mL), 808C, 20 h. [b] Isolated yields. [c] Yield of the disubstituted
amines. [d] Ratio of linear products to branched and bis-allylated ones.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2014, 7, 890 – 896 892

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
low boiling point of these substrates and products
led to losses during work-up or during the reaction.
Primary unsubstituted allyl alcohol, 6c, and cyclic
secondary aliphatic alcohol, 6 f, afforded single linear
products (entries 7 and 10, whereas, methyl substi-
tuted linear and branched allylic alcohols, 6d and
6e, provided the same allylated products with trace
amounts of di-substituted indole as 9:3:1, linear to
branched and di-allylated product ratio (entries 8 and
9).
Table 4. Allylic amination reactions of primary and secondary amines with various al-
lylic alcohols[a]
Entry Nucleophile Substrate
Product (E/Z)[b]
Yield[c]
[%]
1
2
8a
8a
6k
6l
8ak 1:1.6
8ak 1:1.6
74
57
Henceforth, we studied the direct activation of al-
lylic alcohols with primary and secondary amines.
The urea assisted Pd catalyst (using ligand 4+1,3-di-
ethylurea) afforded high yields and selectivity for
linear monosubstituted products with aromatic and
aliphatic allyl alcohols (Table 3). Importantly, electron
deficient nucleophiles also afforded the desired prod-
ucts in high yields with traces of bis-allylated amines
(entries 1–3). Electron-rich secondary amines, which
are highly nucleophilic, formed the linear aminated
products with excellent yields (entries 4, 5). Primary
alkyl amines gave mono-allylated product in moder-
ate yield (entry 6) whereas morpholine, 8g, produced
the desired product in excellent yield (entry 7). The
expected products with aliphatic allyllic alcohols ob-
tained with moderate yields. The methyl substituted
allylic alcohols, 6d and 6e, formed the mixture of
linear and branched products together with di-allylat-
ed (entries 8, 9). Dimethyl substituted primary and
secondary allylic alcohols, 6g and 6h, generated only
the linear product. No branched or bisallylated prod-
3
4
8a
8a
6m
6n
8am 1:1.6:1.7:2.5
8am 1:1.4:1.6:2.4
90
96
5
6
8g
8g
6k
6l
8gk 1:1.9
8gk 1:1.8
81
93
7
8
8g
8g
6m
6n
8gm 1: 3.3.2.5
8gm 1: 6.7: 5.0
57
57
[a] Reaction conditions: 8a (0.5 mmol), 6a (0.75 mmol), 3 mol% [(h3-allyl)Pd(cod)]BF4,
3 mol% 1,3-diethylurea, 6 mol% ligand 4, toluene (2.5 mL), 808C, 20 h. [b] The E/Z
isomeric ratio of the products were determined by GC. [c] Isolated yields.
ucts were detected in these reactions (entries 10, 11). Also, sec-
ondary cyclic aliphatic alcohol afforded the monosubstituted
product with excellent yields (entry 12).
The application of this novel catalyst system was further
studied in the direct activation of terpenols (Table 4). These are
long-chain hydrocarbons with OH substituents that are essen-
tial oils found in nature. Terpenols are important building
blocks for the preparation of the terpenes and intermediates
for natural product synthesis. Functionalization of these valu-
able alcohols with amine derivatives leads to intermediates rel-
evant for the synthesis of biologically active compounds. Excel-
lent yields were obtained when these alcohols were reacted
using the combined Pd/Ligand 4/1,3-diethylurea catalytic
system. Linear and branched isomeric alcohols, geraniol and li-
nalool, 6k and 6l, showed full conversions with primary aro-
matic and secondary aliphatic amines, with high selectivity for
the monoallylated linear products (entries 1–2 and 5–6). Addi-
tionally, longer chain alcohols, farnesol and neralidol, 6m and
6n, showed similar activity and regioselectivity (entries 3–4
and 7–8). Interestingly, the linear and the branched isomeric al-
cohols result in formation of the same linear products. This
result suggests that these reactions go through the same inter-
mediate for the corresponding allylic alcohols. Indeed, analysis
of the reaction mixtures with ESI-MS show the formation of
the same p-allylic intermediate during the conversion of either
the linear or branched alcohols (see Supporting Information).
Scheme 4. The model allylic amination reaction of the kinetic studies.
Next, we studied the mechanism of the urea-assisted nucle-
ophilic substitution of allylic alcohols. For the kinetic studies
we chose the N-methylaniline, 8d, as a nucleophile in combi-
nation with cinnamyl alcohol, 6a, as the catalyst system produ-
ces a single product with full conversion and 95% isolated
yield (Table 3, entry 4)—(Scheme 4). We studied the effect of
the concentration of the alcohol substrate, the nucleophile,
the Pd catalyst and urea additive by monitoring the reaction in
time and analyzing the data with reaction progress kinetics.[20]
The reaction progress was monitored by GC and initial rates
were calculated from the slope of curve taken at the beginning
of the reaction. These data were utilized to determine the
order of the each reactant (Figure 1). We observed first order
kinetics in alcohol and zero order kinetics in nucleophile con-
centration (Figure 1a). These results are consistent with litera-
ture data where it was suggested that the oxidative addition is
the rate determining step for the allylic substitution of allyl al-
cohol substrates.[21] As expected, the kinetic data reveal a first
order kinetics in Pd catalyst. At high Pd concentrations we ob-
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2014, 7, 890 – 896 893

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
Figure 2. Same excess experiments A) Comparison of run a: [6a]=0.24m,
[8d]=0.36m and run b: [6a]=0.18m, [8d]=0.30m, by reaction rate versus
concentration of allylic alcohol, 6a. B) Comparison of run a with product
added reaction run c: [6a]=0.18m, [8d]=0.30m, [8ad]=0.06m.
remains constant from beginning to the end. Accordingly, two
reactions with same excess values with different absolute initial
concentrations of the substrates were performed. From the
plot of these two reactions (reaction rate versus substrate con-
centration) it is clear that these do not overlay, having a slower
reaction rate for that with the highest concentration (Fig-
ure 2A). This indicates that we have catalyst decomposition or
product inhibition. According to the Blackmond protocol, an
additional experiment was performed with the same excess
and in the presence of product. The concentrations were
chosen as such that the amount of substrate is identical to
that in experiment b, and the amount of product plus sub-
strate is identical to that in experiment a. As is clear from the
Figure 2B, the plots of reaction a and c are now nicely overlay-
ing, indicating that the reaction suffers from product
inhibition.
Figure 1. Kinetic analysis of amination reaction (Scheme 3) a) Rate versus
substrate concentration of cinnamyl alcohol, 6a, and N-Methylaniline, 8d.
b) Rate versus Pd catalyst concentration. c) Rate versus 1,3-diethylurea
concentration.
served deviations from first order kinetics, which may indicate
catalyst decomposition under these conditions (Figure 1b). Im-
portantly, from the kinetic data it is clear that the presence of
the urea additive increases the activity of the catalyst and the
effect is stronger at higher concentration urea (Figure 1c).
Above urea concentrations of 24 mm the reaction rate does
not increase further. Notably, the reactions in the absence of
urea gave less reproducible yields, indicating that the urea ad-
ditive is also important to obtain reproducible results.
The perfect overlap observed in Figure 2B also shines light
on the effect of water as additive under these conditions.
In principle water, which is released during the reaction, could
play a role in the activation of allylic alcohols, as has been ob-
served for other systems.[13b] However, in our system water is
not playing a major role, as otherwise the experimental kinetic
curves of experiment b and c would not have over-layed. Full
overlap of these kinetic curves suggests that the water that is
formed during the reaction is neither participating significantly
in product inhibition nor in allyl alcohol activation.
Recently, Blackmond developed progress kinetic analysis as
a new methodology to analyze kinetic data.[20] In such ap-
proach the kinetic data are used to generate rate vs substrate
concentration plots, from which one can directly see the order
of the reaction. Interestingly, reactions that have different start-
ing concentrations of the substrate should overlay in the event
that there is no catalyst deactivation or product inhibition, im-
portant issues that are visualized in plots from two experi-
ments. For reactions that have two different substrates such as
in the current study, experiments are designed in which the
excess of one of the two substrates remains the same, also al-
lowing to make overlay plots to judge product inhibition/cata-
lyst deactivation events. The parameter called the ‘excess’ is
defined as initial concentration difference of two reactants that
Based on the kinetic data we propose the mechanism for
direct activation of allylic alcohols for the amination reactions
as depicted in Scheme 5. In situ formed cationic phosphorami-
dite palladium precursor 9 reacts with the nucleophile to gen-
erate species 12. Alkene exchanges leads to the formation Pd
intermediate 10. Subsequently, oxidative addition leads to the
allyl intermediate 11, which can undergo nucleophilic attack to
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2014, 7, 890 – 896 894

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
Further calculations should demonstrate the effect of these hy-
drogen bonds on the transition states that lead to 11.
Conclusions
We report a new, efficient, regioselective Pd catalyst system for
direct activation of allylic alcohols that can be applied in alkyla-
tion and amination reactions. The catalyst has a broad sub-
strate scope and is tolerant towards different functional
groups. The phosphoramidite ligand with phenyl alanine
moiety has functional groups that form supramolecular inter-
actions with the substrate. In addition, the 1,3-diethylurea
building also forms hydrogen bonds with the allylic alcohol,
which leads to activation of the allylic alcohols. Detailed kinetic
studies for amination reactions show that this reaction is first
order in allyl alcohol, urea, and palladium, supporting that oxi-
dative addition is the rate determining step and this step facili-
tated by urea. This simple catalyst system opens up new syn-
thetic routes to functionalize allylic alcohols with a minimum
of waste formation, including those based on biorenewable
terpenols.
Scheme 5. Proposed mechanism for amination reactions of allylic alcohols.
Experimental Section
form the product. Product inhibition is explained by the equi-
librium between 12 and 10. Zero order kinetics in nucleophile
and first order kinetics in allylic alcohol, Pd and 1,3-diethylurea
indicate that the oxidative addition is the rate determining
step. This oxidative addition reaction is facilitated by the urea
additive. Preliminary DFT calculations suggest that both in 10
and 10’ a hydrogen bond is formed between the substrate
and the ligand, whereas in 10’ also hydrogen bonds are
formed between the urea N-H and the alcohol (Figure 3).
General procedure for alkylation and amination reactions: Allylic al-
cohol, 6a, (0.5 mmol) and nucleophile, 8a, (0.75 mmol) added to
the mixture of 3 mol% [(h3-allyl)Pd(cod)]BF4, 3 mol% 1,3-diethylur-
ea and 6 mol% ligand 4 in toluene (2.5 mL). Reactions were com-
pleted in 20 h at 808C. Products were isolated in pure form after
column chromatography.
Acknowledgements
This research has been performed within the framework of the
CatchBio program. The authors gratefully acknowledge the sup-
port of the Smart Mix Program of the Netherlands Ministry of
Economic Affairs and the Netherlands Ministry of Education, Cul-
ture and Science.
Keywords: alkylation · aminations · homogeneous catalysis ·
hydrogen bonding · palladium
[1] a) B. M. Trost, T. J. Fullerton, J. Am. Chem. Soc. 1973, 95, 292–294;
b) B. M. Trost, D. L. van Vranken, Chem. Rev. 1996, 96, 395–422.
[2] a) B. M. Trost, Acc. Chem. Res. 1980, 13, 385–393; b) J. E. Baeckvall, Acc.
Chem. Res. 1983, 16, 335–342; c) M. Tsuji, I. Minami, Acc. Chem. Res.
1987, 20, 140–145; d) B. M. Trost, Angew. Chem. 1989, 101, 1199–1219;
Angew. Chem. Int. Ed. Engl. 1989, 28, 1173–1192; e) W. Oppolzer, Angew.
Chem. 1989, 101, 39–53; Angew. Chem. Int. Ed. Engl. 1989, 28, 38–52;
f) J. Tsuji, Synthesis 1990, 739–749; g) B. M. Trost, Pure Appl. Chem.
1992, 64, 315–322; h) J. E. Backvall, Pure Appl. Chem. 1992, 64, 429–
437; i) C. G. Frost, J. Howarth, J. M. J. Williams, Tetrahedron: Asymmetry
1992, 3, 1089–1122; j) G. Giambastiani, G. Poli, J. Org. Chem. 1998, 63,
9608–9609; k) Y. Uozumi, H. Danjo T. Hayashi, J. Org. Chem. 1999, 64,
3384–3388.
[3] a) M. Sakamoto, I. Shimizu, A. Yamamoto, Bull. Chem. Soc. Jpn. 1996, 69,
1065–1078; b) R. D. Connell, T. Rein, B. Akermark, P. Helquist, J. Org.
Chem. 1988, 53, 3845–3849.
[4] a) F. E. Ziegler, A. Kneisley, R. T. Wester, Tetrahedron Lett. 1986, 27, 1221–
1224; b) F. E. Ziegler, R. T. Wester, Tetrahedron Lett. 1986, 27, 1225–
Figure 3. DFT-calculated intermediates of 10 and 10’, displaying hydrogen
bonds that are crucial according to kinetic analysis of reaction mechanism
(Scheme 5).
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2014, 7, 890 – 896 895

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
1228; c) F. E. Ziegler, W. T. Cain, A. Kneisley, E. P. Stirchak, R. T. Wester, J.
Am. Chem. Soc. 1988, 110, 5442–5452.
[5] X. Lu, L. Lu, J. Sun, J. Mol. Catal. 1987, 41, 245–251.
b) B. Breit, Angew. Chem. 2005, 117, 6976–6986; Angew. Chem. Int. Ed.
2005, 44, 6816–6825; c) A. J. Sandee, J. N. H. Reek, Dalton Trans. 2006,
3385–3391; d) J. Meeuwissen, J. N. H. Reek, Nat. Chem. 2010, 2, 615–
621.
[6] X. Lu, X. Jiang, X. Tao, J. Organomet. Chem. 1988, 344, 109–118.
[7] a) M. Kimura, T. Tomizawa, Y. Horino, S. Tanaka, Y. Tamaru, Tetrahedron
Lett. 2000, 41, 3627–3629; b) M. Kimura, Y. Horino, R. Mukai, S. Tanaka,
Y. Tamaru, J. Am. Chem. Soc. 2001, 123, 10401–10402; c) M. Kimura, M.
Futamata, K. Shibata, Y. Tamaru, Chem. Commun. 2003, 234–235; d) M.
Kimura, M. Futamata, R. Mukai, Y. Tamaru, J. Am. Chem. Soc. 2005, 127,
4592–4593; e) B. M. Trost, J. Quancard, J. Am. Chem. Soc. 2006, 128,
6314–6315.
[15] a) S. Das, C. D. Incarvito, R. H. Crabtree, G. W. Brudvig, Science 2006, 312,
1941–1943; b) P. Dydio, W. I. Dzik, M. Lutz, B. de Bruin, J. H. N. Reek
Angew. Chem. 2011, 123, 416–420; Angew. Chem. Int. Ed. 2011, 50, 396–
400; Angew. Chem. Int. Ed. 2011, 50, 396–400; c) P. Dydio, J. H. N. Reek,
Angew. Chem. 2013, 125, 3970–3974; Angew. Chem. Int. Ed. 2013, 52,
ˇ
3878–3882; d) T. Smejkal, D. Gribkov, J. Geier, M. Keller, B. Breit, Chem.
ˇ
Eur. J. 2010, 16, 2470–2478; e) T. Smejkal, B. Breit, Angew. Chem. 2008,
´
[8] a) I. Stary, I. G. Stara, P. Kocovsky, Tetrahedron Lett. 1993, 34, 179–182;
120, 317–321; Angew. Chem. Int. Ed. 2008, 47, 311–315.
´
b) I. Stary, I. G. Stara, P. Kocovsky, Tetrahedron 1994, 50, 529–537.
[16] a) W. I. Dzik, X. Xu, X. P. Zhang, J. N. H. Reek, B. de Bruin, J. Am. Chem.
Soc. 2010, 132, 10891–10902; b) H. Lu, W. I. Dzik, X. Xu, L. Wojtas, B. de
Bruin, X. P. Zhang, J. Am. Chem. Soc. 2011, 133, 8518–8521; c) W. I. Dzik,
X. P. Zhang, B. de Bruin, Inorg. Chem. 2011, 50, 9896–9903; d) W. I. Dzik,
J. N. H. Reek, B. de Bruin, Chem. Eur. J. 2008, 14, 7594–7599; e) S. F. Zhu,
X. Xu, J. A. Perman, X. P. Zhang, J. Am. Chem. Soc. 2010, 132, 12796–
12799; f) Y. Chen, X. P. Zhang, J. Org. Chem. 2004, 69, 2431–2435.
[17] I. Usui, S. Schmidt, M. Keller, B. Breit, Org. Lett. 2008, 10, 1207–1210.
[18] P. A. R. Breuil, F. W. Patureau, J. N. H. Reek, Angew. Chem. 2009, 121,
2196–2199; Angew. Chem. Int. Ed. 2009, 48, 2162–2165.
[19] a) T. Hayashi, Acc. Chem. Res. 2000, 33, 354–362; b) M. D. K. Boele,
P. C. J. Kamer, M. Lutz, A. L. Spek, J. G. de Vries, Piet W. N. M. van Leeu-
wen, G. P. F. van Strijdonck, Chem. Eur. J. 2004, 10, 6232–6246.
[20] D. G. Blackmond, Angew. Chem. 2005, 117, 4374–4393; Angew. Chem.
Int. Ed. 2005, 44, 4302–4320.
[9] Y. Masuyama, J. P. Takahara, Y. Kurusu, J. Am. Chem. Soc. 1988, 110,
4473–4474.
[10] a) K. Itoh, N. Hamaguchi, M. Miura, M. Nomura, J. Chem. Soc. Perkin
Trans. 1 1992, 2833–2835; b) T. Satoh, M. Ikeda, M. Miura, M. Nomura, J.
Org. Chem. 1997, 62, 4877–4879; c) S. C. Yang, C. W. Hung, J. Org.
Chem. 1999, 64, 5000–5001; d) S. C. Yang, Y. C. Tsai, Y. J. Shue, Organo-
metallics 2001, 20, 5326–5330; e) Y. J. Shue, S. C. Yang, H. C. Lai, Tetrahe-
dron Lett. 2003, 44, 1481–1485.
[11] a) F. Ozawa, H. Okamoto, S. Kawagishi, S. Yamamoto, T. Minami, M. Yosh-
ifuji, J. Am. Chem. Soc. 2002, 124, 10968–10969; b) F. Ozawa, T. Ishiya-
ma, S. Yamamoto, S. Kawagishi, H. Murakami, Organometallics 2004, 23,
1698–1707.
[12] a) D. Banerjee, R. V. Jagadeesh, K. Junge, H. Junge, M. Beller, Chem-
SusChem 2012, 5, 2039–2044; b) D. Banerjee, R. V. Jagadeesh, K. Junge,
H. Junge, M. Beller, Angew. Chem. 2012, 124, 11724–11728; Angew.
Chem. Int. Ed. 2012, 51, 11556–11560; c) Yi-C. Hsu, K. H. Gan, S. C. Yang,
Chem. Pharm. Bull. 2005, 53, 1266–1269; d) B. Sundararaju, M. Achard,
C. Bruneau, Chem. Soc. Rev. 2012, 41, 4467–4483.
[21] T. Ohshima, Y. Miyamoto, J. Ipposhi, Y. Nakahara, M. Utsunomiya, K. Ma-
shima, J. Am. Chem. Soc. 2009, 131, 14317–14328.
[13] a) R. Ghosh, A. Sarkar, J. Org. Chem. 2011, 76, 8508–8512; b) H. Kinoshi-
ta, H. Shinokubo, K. Oshima, Org. Lett. 2004, 6, 4085–4088.
[14] For supramolecular ligand strategies see: a) M. J. Wilkinson, P. W. N. M.
van Leeuwen, J. N. H. Reek, Org. Biomol. Chem. 2005, 3, 2371–2383;
Received: July 19, 2013
Revised: November 1, 2013
Published online on January 16, 2014
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2014, 7, 890 – 896 896