Download Development of Multi-Component Reactions using Catalytically Generated Allyl Metal Reagents

Development of Multi-Component
Reactions using Catalytically Generated
Allyl Metal Reagents
Nicklas Selander
Licentiate Thesis
Department of Organic Chemistry
Stockholm University, Sweden
June 2008
Phil. Licentiate seminar will be held Friday 27 June 2008 at 10.00 in Magnélisalen
Kemiska Övningslaboratoriet Svante Arrhenius väg 12A
Abstract
This licentiate thesis is based on the development of catalytic reactions for the synthesis
and application of organometallic reagents. By use of palladium pincer-complex
catalysts, we have developed an efficient procedure for the synthesis of allylboronates
starting from allylic alcohols. These reactions were further extended by including various
one-pot multi-component reactions, using the in situ generated allylboronates.
Furthermore, novel unsymmetrical palladium pincer-complexes were synthesized and
studied in auto-tandem catalysis.
II
Table of Contents
ABSTRACT............................................................................................................................................ II
TABLE OF CONTENTS .....................................................................................................................III
LIST OF PUBLICATIONS ................................................................................................................. IV
LIST OF ABBREVIATIONS ................................................................................................................V
1. INTRODUCTION .............................................................................................................................. 1
1.1 MULTI-COMPONENT REACTIONS .................................................................................................... 1
1.2 ALLYLIC ORGANOMETALLIC REAGENTS ......................................................................................... 2
1.2.1 ALLYLBORONATE REAGENTS IN ORGANIC SYNTHESIS ................................................................ 3
1.2.2 SYNTHESIS OF ALLYLBORONATES ............................................................................................... 4
1.3 PALLADIUM PINCER-COMPLEXES ................................................................................................... 5
2. PALLADIUM PINCER-COMPLEX-CATALYZED BORYLATION OF ALLYLIC
ALCOHOLS (PAPERS I-II).................................................................................................................. 7
2.1 ALLYLIC ALCOHOLS AS SUBSTRATES IN THE CATALYTIC BORYLATION REACTION ....................... 7
2.2 SELECTIVITY AND SUBSTITUENT EFFECTS IN THE BORYLATION REACTION .................................... 9
2.3 SOLVENT EFFECTS – METHANOL AS CO-SOLVENT .......................................................................... 9
2.4 ACTIVATION OF THE HYDROXY GROUP ........................................................................................ 10
2.5 STRATEGIES FOR FINE-TUNING THE CATALYTIC ACTIVITY OF PINCER-COMPLEXES .................... 11
2.5.1 SYNTHESIS AND CHARACTERIZATION OF METHOXY SUBSTITUTED PINCER-COMPLEXES .......... 12
2.5.2 APPLICATION OF METHOXY SUBSTITUTED PINCER-COMPLEXES IN CATALYTIC ARYLATION OF
VINYL EPOXIDE AND BORYLATION OF CINNAMYL ALCOHOL ............................................................. 12
3. ONE-POT ALLYLATION OF CARBONYL COMPOUNDS VIA TRANSIENT
ALLYLBORONATES GENERATED BY PALLADIUM-CATALYZED SUBSTITUTION OF
ALLYLIC ALCOHOLS (PAPERS III-V).......................................................................................... 15
3.1 ONE-POT ALLYLATION OF ALDEHYDES ........................................................................................ 16
3.2 ONE-POT ALLYLATION OF KETONES ............................................................................................ 20
3.3 ONE-POT SYNTHESIS OF Α-AMINO ACIDS VIA IN SITU GENERATED ALLYLBORONATES................. 21
3.4 COUPLING OF CATALYTICALLY GENERATED ALLYLBORONATES WITH IN SITU HYDROLYZED
ACETALS............................................................................................................................................. 24
3.5 MECHANISTIC ASPECTS OF THE ONE-POT ALLYLATION OF CARBONYL COMPOUNDS .................. 28
4. SYNTHESIS AND TANDEM CATALYTIC PERFORMANCE OF SYMMETRICAL AND
UNSYMMETRICAL SULFUR-CONTAINING PALLADIUM PINCER-COMPLEXES (PAPER
VI) .......................................................................................................................................................... 32
4.1 SYNTHESIS OF THE UNSYMMETRICAL PC(H)S-PINCER PRO-LIGAND AND THE PCS-COMPLEXES . 33
4.2 CATALYTIC PERFORMANCE OF SULFUR- AND PHOSPHOROUS-BASED PINCER-COMPLEXES AS PRECATALYST IN AN ALDOL CONDENSATION ........................................................................................... 35
4.3 SCS- AND PCS-PINCER-COMPLEXES AS CATALYSTS FOR TANDEM CATALYSIS ........................... 36
4.4 MECHANISTIC CONSIDERATIONS OF THE TANDEM COUPLING REACTION ..................................... 39
5. GENERAL CONCLUSIONS AND OUTLOOK ........................................................................... 42
6. ACKNOWLEDGEMENTS ............................................................................................................. 43
7. REFERENCES.................................................................................................................................. 44
III
List of Publications
The thesis is based on the following papers, referred to by their roman numerals 1-VI.
I. V. J. Olsson, S. Sebelius, N. Selander and K. J. Szabó: Direct Boronation of Allyl
Alcohols with Diboronic Acid Using Palladium Pincer-Complex Catalysis. A
Remarkably Facile Allylic Displacement of the Hydroxy Group under Mild
Reaction Conditions. J. Am. Chem. Soc. 2006, 128, 4588-4589.
II. J. Aydin, N. Selander and K. J. Szabó: Strategies for Fine-tuning the Catalytic
Activity of Pincer-Complexes. Tetrahedron Lett. 2006, 47, 8999-9001.
III. N. Selander, S. Sebelius, C. Estay and K. J. Szabó: Highly Selective and Robust
Palladium-Catalyzed Carbon-Carbon Coupling between Allyl Alcohols and
Aldehydes via Transient Allylboronic Acids. Eur. J. Org. Chem. 2006, 40854087.
IV. N. Selander, A. Kipke, S. Sebelius and K. J. Szabó: Petasis Borono-Mannich
Reaction and Allylation of Carbonyl Compounds via Transient Allyl Boronates
Generated by Palladium Catalyzed Substitution of Allyl Alcohols. An Efficient
One-Pot Route to Stereodefined α-Amino Acids and Homoallyl Alcohols. J.
Am. Chem. Soc. 2007, 129, 13723-13731.
V. N. Selander and K. J. Szabó: Single-pot Triple Catalytic Transformations Based
on Coupling of in situ Generated Allyl Boronates with in situ Hydrolyzed
Acetals. Chem. Commun. 2008, DOI: 10.1039/b804920c.
VI. M. Gagliardo, N. Selander, N. C. Mehendale, G. v. Koten, R. J. M. Klein-Gebbink
and K. J. Szabó: Catalytic Performance of Symmetrical and Unsymmetrical
Sulfur-Containing Pincer Complexes: Synthesis and Tandem Catalytic
Activity of the First PCS-Pincer Palladium Complex. Chem. Eur. J. 2008, 14,
4800-4809.
Not included in this thesis:
N. Selander and K. J: Szabó: Efficient Synthesis of α-Amino Acids via
Organoboronate Reagents. ACS Symposium Series: “Current Frontiers in Asymmetric
Synthesis and Application of alpha-Amino Acids. Eds: V. A. Soloshonok and K. Izawa,
2008, in press.
IV
List of Abbreviations
cat
catalyst
dba
dibenzylideneacetone
DMSO
dimethylsulfoxide
L. A.
Lewis acid
MeOH
methanol
p-TsOH
para-toluenesulfonic acid
RT
room temperature
THF
tetrahydrofuran
TS
transition state
V
1. Introduction
One of the most attractive tools in modern organic chemistry is the use of multi-step onepot procedures1-7 to construct densely functionalized compounds starting from simple
precursors. This strategy is widely adopted by Nature, but in traditional organic synthesis,
complex molecules are usually built up by stepwise formation of the individual bonds in
the target structure. By using the multi-component approach, in which several reaction
steps are performed in the same reaction vessel, the amount of solvents, waste and labor
can be reduced. Thus, the higher efficiency is favorable from both an environmental and
economical point of view. In particular, multi-component synthesis involving transition
metal catalysis is a powerful combination of synthetic efficiency and highly selective
transformations. Since the catalytically generated intermediates (such as organometallic
compounds) need not to be isolated, but react with other components present in the
reaction mixture, cumbersome purification of highly reactive species can be avoided.
This thesis is focused on development of selective multi-component reactions in which an
allyl metal reagent is generated in situ by palladium-catalysis.
1.1 Multi-Component Reactions
In the past decades, the development of effective multi-component based synthesis has
played an important role to achieve high atom economy and sustainable chemistry.1-7 The
major challenges in this field are compatibility between the reagents and catalysts present
to prevent catalyst inhibition and unwanted side-reactions. Furthermore, the timematching of the distinct reactions are important to avoid high concentrations of reactive
intermediates which can lead to decomposition or side-reactions. The classifications
found in the literature are often non-consistent and distinguishing terms such as domino,
cascade, one-pot and tandem reactions (or catalysis) sometimes cause confusion. Several
definitions have been proposed,2-6 and the most important ones are presented here.
The term domino reactions was first defined by Tietze2 as “a process involving two or
more bond-forming transformations (usually C-C bonds) which take place under the
same reaction conditions without adding additional reagents and catalysts, and in which
the subsequent reactions result as a consequence of the functionality formed in the
previous step”. Cascade reactions usually refer to multiple domino sequences.
When multi-component reactions involve transition metal-catalyzed processes, additional
definitions are found.3-6 A thorough taxonomy of tandem catalysis has been reviewed by
Fogg and co-workers.4 Accordingly, in tandem or domino catalyses, all catalytic species
must be present from the outset. When only one catalytic transformation is performed,
followed by another stoichiometric transformation, the tandem concept is not valid, but
the process is a domino reaction. The definition of domino reactions proposed by Tietze2
can also be extended to catalytic protocols where the transformations are performed by a
single catalytic mechanism. Tandem catalysis is defined as coupled catalyses in which
sequential catalytic elaborations of the substrate occurs via two or more mechanistically
distinct processes. Three different categories are proposed; Orthogonal, assisted and
auto-tandem catalysis.
1
In orthogonal tandem catalysis, two different catalysts transform the substrate
simultaneously. If one catalyst is able to perform two or more mechanistically distinct
processes in concert, auto-tandem catalysis is operating. The catalyst can also be
triggered by an additional reagent to be able to perform a second catalytic transformation,
referred to as assisted tandem catalysis (Figure 1).4
Are all precatalysts present at outset?
No
One-pot multicatalytic reaction
Yes
Is more than one catalytic mechanism required?
No
Yes
Is a single (pre)catalyst used?
No
Cat. A
Cat. A
Domino
A' + B
B'
( cascade ) A
catalysis
Mech. A
Mech. A
Cat. A
A
Orthogonal catalysis
Yes
Cat. B
A'
Mech. A
No
Is a trigger used to change the mechanism
or to transf orm the catayst?
Yes
Auto-tandem
catalysis
Trigger
Cat. A
Assisted-tand em catalysis A
A'
Mech. A
Cat. A (A')
B'
Mech. B
Cat. B
A'
Mech. A
Mech. B
Cat. A
A
B'
B'
Mech. B
TANDEM CATALYSES
Figure 1. Overview of one-pot processes involving multiple catalytic transformations.
Throughout this thesis, we employ this subset of definitions when applicable.
1.2 Allylic Organometallic Reagents
Allylic organometallic reagents are extensively used as carbanion equivalents in organic
synthesis. Their additions to aldehyde, ketone and imine electrophiles provide a valuable
route to homoallylic alcohols and amines (Scheme 1). 8-14
O
OH
R2 2 H
R1
M
1
NR3
R2
4
R1
NHR 3
R2 3 H
R2
5
R1
Scheme 1. Allylation of aldehydes and imines using allylic organometallic reagents.
These C-C bond forming reactions are complementary to the aldol reactions, and the
introduction of a double bond in the products offers an important synthetic advantage. A
wide array of metals has been utilized with different levels of selectivity.
2
1.2.1 Allylboronate Reagents in Organic Synthesis
The main focus of this thesis is directed to the synthesis and highly selective
transformations of allylboronates. The allylation of aldehydes by allylboronates leading
to homoallylic alcohols is an important C-C bond forming reaction. Several attractive
properties of allylboronates account for their popularity as synthetic intermediates in
organic synthesis.8-62 Their high diastereoselectivity in the coupling reactions with
aldehydes is predictable and often superior to other allylating reagents. Moreover, the
allylboronates are easy to handle and less sensitive to oxidation and rearrangements than
allylboranes such as 6a. The stability and reactivity of allylboron derivatives have been
studied by Brown and co-workers.63 It was concluded that the stability and reactivity are
highly dependent on the groups directly attached to the boron atom. Thus, the oxygen
substituents found in allylboronic acids (6b) and allylboronic acid esters (6c) are
important to decrease the electrophilicity of boron by donation (nπ-pπ) of electrons into
the unoccupied pπ-orbital of boron. The pinacol allylboronic esters (6c) are more stable
than the corresponding allylboronic acids (6b), and can be purified by column
chromatography. The allylboronic acids (6b) are not stable under solvent-free
conditions30,47 but can be transformed into the corresponding potassium
trifluoro(allyl)borates (6d),43,44,64,65 which are isolatable air- and moisture stable species
(Figure 2).
OH
B
OH
<
B
6a
O
B
<
6b
<
O
BF3K
6c
6d
Figure 2. Relative stability of allylboronic derivatives.
The high diastereoselectivity in the coupling of allylboronates with aldehydes is a result
of an internal activation of the aldehyde functionality by the empty pπ-orbital of boron.
The addition proceeds via a compact six-membered ring transition state, where the
geometry of the olefin bond determines the diastereoselectivity66 (Scheme 2).
Type II
L.A.
O
Type I
OR
B
OR
RE
RCHO
2
RZ
6e
RE
OR
B OR
OH
R O
R E RZ
7a
4
RZ
R
RE
R
H
M
RZ
7b M: Si, Sn
Scheme 2. Allylation of aldehydes via type I (7a) and type II (7b) mechanisms.
Allylboronates add to aldehydes by a “type I” (7a) mechanism67 with high
diastereoselectivity. Thus, E-allylboronate gives the anti product 4, and Z-allylboronate
gives the syn isomer of 4. In contrast, allylic silanes and stannanes require an external
Lewis acid activation via an open transition state 7b (type II), reducing the
diastereoselectivity and predictability.
3
Although allylboronates are self-activating in their coupling with aldehydes, Hall and coworkers found that the addition of a Lewis acid catalyzes the reaction.16-20,23,24,27,28,40 It
was shown that the Lewis acid interacts with the lone-pairs of the boronate oxygen (7c)
and thereby renders the boron more electron deficient (Figure 3). By using chiral
Brønsted acids,23,27 moderate levels of enantioselectivity were induced. Alternatively,
chirality can be induced by employment of chiral diol-based boronates.17,19,20,31,40,47,49,61
RE
OR
B OR
R O
L.A
RZ 7c
Figure 3. Proposed Lewis acid activation in the allylation of aldehydes.
Allylation reactions using imines as electrophiles, obtaining homoallylic amines have
also been reported.29,31,45,58,68 Perhaps the most elegant and synthetically powerful
development in this field is based on in situ generation of imines. Petasis and coworkers69-73 have shown that vinylboronic acids add to imines formed in situ from
various amines (9) and glyoxylic acid (2a), to give un-natural α-amino acid derivatives
(10). Virtually all types of organoboronates (8) undergo the Petasis reaction (Also
referred to as the Petasis Borono-Mannich reaction), including vinyl-,70,72,74,75 aryl-72,74,76
and allylboronates46,51,52 (Scheme 3). Very recently, Schaus77 reported an asymmetric
version of the Petasis reaction of vinylboronates.
R-B(OH) 2 + R'2NH + HOOC-CH=O
8
9
2a
NR´2
R
10
COOH
Scheme 3. General scheme for the Petasis reaction affording α-amino acids.
Allylation of ketones is generally much slower than the allylation of aldehydes, and the
selectivity is highly dependent on the relative size of the ketone substituents. Chelating
groups such as amines in α-position to the ketone functionality were shown to enhance
the reactivity.41,42 Furthermore, asymmetric allylation of ketones, using chiral
phosphorus-based ligands57 or diols have been reported.59,62 Kobayashi and co-workers
have also found that InI catalyzes the allylboration of ketones. It was shown that
allylboronates could be added to a variety of ketones under mild conditions.53 The scope
of this protocol was recently extended to include N-acylhydrazones as electrophiles.55,56
1.2.2 Synthesis of Allylboronates
Several methods for the synthesis of allylboronates have been reported.15,29,31,38,48,49,63,78102
The boronate functionality can either be introduced directly into the substrate (direct
methods), or an organoboronate can be modified to construct an allylic system (indirect
methods). In the context of this thesis, the direct methods are the most relevant, but also
more synthetically interesting and general.
4
Simple, non-functionalized allylboronates such as allyl- and crotylboronates can readily
be prepared by addition of reactive allylic organometallic reagents 1 such as allyllithium78-82
allyl-Grignard-49,63,78,83 or allylpotassium48,82,84-89 reagents to trialkoxyborates (11).
With substituted allylic reagents, the stereo- and regioselectivity is sometimes low, giving
a mixture of products 6f and 6g by metallotropic rearrangement (Scheme 4).
R2
M
B(OR)3
1
R 1
M = Li, Mg, K
OR
B
OR +
2
R
11
R1
OR
B
OR
R1 R2
6g
6f
Scheme 4. Synthesis of allylboronates using highly reactive allyl metal reagents.
However, the harsh conditions and the stereochemical instability66 associated with this
type of reagents, limits the synthetic scope of the procedure. A more versatile strategy in
the synthesis of functionalized allylboronates is the use of transition metal based catalytic
transformations.29-31,38,90-103 Miyaura and co-workers96 have shown that allylic acetates 12
can be employed in a palladium-catalyzed substitution reaction using
bis(pinacolato)diboron 14 to yield functionalized allylboronic esters 6h (Scheme 5).
R
OAc
12
[Pd(dba)2]cat
O
O
B B
O 14 O
R
L
Pd
L
13
R
O
B
R
O
6h
+
15
Scheme 5. Palladium-catalyzed substitution of allylic acetates.
The reaction proceeds via mono-allyl palladium intermediate 13. A drawback with this
process is the formation of varying amounts of the dimer 15, resulting from
transmetallation of the product to the mono-allyl palladium intermediate 13 followed by
reductive allyl-allyl coupling. Notably, formation of this by-product was not observed
when the borylation was carried out in the presence of an electrophile.29,31,38
1.3 Palladium Pincer-Complexes
The development of highly selective and effective transition metal catalysts is an
important field in organic chemistry.104-108 A key property of such catalysts is a welldefined metal-ligand bonding, which can be achieved by the use of so called “pincercomplexes”.109-121 A plethora of pincer-complexes with several different metals have
been synthesized, and this thesis is focused on the application of palladium pincercomplexes (16) as highly selective catalysts in organic transformations (Figure 4).
5
R
BF4PhSe
Pd
Cl
SePh MeS
16a
Pd SMe
NCMe
BF4PhS
16b
Pd PPh 2
NCMe
16c
Figure 4. Selected palladium pincer-complexes applied in this thesis.
Due to the tri-coordinated ligand in the palladium pincer-complexes, the air-, moistureand thermal stability is high. Moreover, the single free coordination site is important to
give robust, selective and predictable catalytic properties. Several ligands with the
general structure [C6H3(CH2E)2-2,6]- have been utilized. Usually, the pincer-complexes
are classified according to the hetero-atoms (E) in the side-arms of the pincer-ligand and
the ipso atom (e.g. complex 16a119,122 is a SeCSe-pincer, E=Se). Another attractive
feature of the pincer-complexes is the tunability of the ligand. By varying the side-arms,
counter ions and aromatic backbones, the reactivity and selectivity in catalytic
transformations can be altered. Furthermore, asymmetric induction by chiral
ligands,123,124 and immobilization of the catalyst on a solid support125,126 have also been
reported.
A large number of synthetic applications using palladium pincer-complexes can be found
in the literature. In the Szabó group, palladium pincer-catalysts have been used in the
synthesis and transformations of organometallic reagents.30-35,37,110,123,124,127-135 As pointed
out in Chapter 1.2, allylboronates are highly important reagents in organic synthesis and
therefore selective methods for preparation of functionalized allylboronates are of high
interest. It was shown30 that palladium pincer-complex 16a catalyses the boronate
transfer from tetrahydroxydiboron (19) to allylic acetates 12, vinyl cyclopropanes 17 and
aziridines 18. The initial products of these highly selective reactions are the allylboronic
acids 6i, but due to their instability under solvent-free conditions, 30,47 they where instead
isolated as the corresponding potassium trifluoro(allyl)borates 6j (Scheme 6).
R
OAc
12
EtOOC COOEt
PhSe
17
Ph
OH
B
OH
Q
HO
Ts
N
18
Pd SePh
Cl
16a
[ ]cat
OH
B B
HO 19 OH
6i
KHF2
Q
BF3K
6j
Scheme 6. Palladium pincer-complex-catalyzed synthesis of functionalized allylborates.
A clear advantage of the rigid pincer architecture in these transformations is that the allylallyl coupling product 15 (Scheme 5) can be completely avoided.
6
2. Palladium Pincer-Complex-Catalyzed Borylation of Allylic Alcohols (Papers I-II)
As mentioned in the introduction, allylboronates are important synthetic intermediates in
organic synthesis. However, many of the existing protocols require either harsh
conditions or rather exotic precursors. Thus, the functional group tolerance is sometimes
low, and the poor availability of the starting materials limits the synthetic scope.
Therefore, development of selective methods to prepare functionalized allylboronates
from easy accessible starting materials is highly important. Allylic alcohols are one of the
most attractive and readily available substrates that undergo palladium-catalyzed allylic
displacement reactions.10,107,136-143 Unfortunately, the hydroxy group is one of the most
reluctant leaving group in substitution reactions; and therefore, application of harsh
reaction conditions, use of Lewis acids or other additives are required in these
processes.136,143 However, employment of these reaction conditions and reagents
precludes the efficient synthesis of unstable allyl metal derivatives, such as
allylboronates, in high isolated yields.
2.1 Allylic Alcohols as Substrates in the Catalytic Borylation Reaction
Previously, the Szabó group30 has shown that allylic acetates could be used as precursors
in the palladium pincer-complex-catalyzed synthesis of allylboronic acids (Scheme 6). In
these reactions, DMSO was used as solvent. Inspired by these results, we studied the
possibility to use allylic alcohols as substrates instead of allylic acetates. We found that
under the same reaction conditions used for allylic acetates, the conversions of the allylic
alcohols were low. However, using methanol as a co-solvent we observed that the
conversions of the allylic alcohols were significantly accelerated.
Very recently, we found33 that the hydroxy to boronate substitution can be achieved
under mild conditions using a wide variety of allylic alcohols 20 and diboronic acid (also
called tetrahydroxydiboron, 19) with 5 mol% pincer-complex 16a in a mixture of DMSO
and MeOH (Scheme 7).
R
HO
OH
20
+
OH
B B
HO 19 OH
PhSe
Pd
SePh
Cl
[16a]cat
R
B(OH) 2
21a-j
DMSO/MeOH
KHF2
R
BF3K
22a-j
Scheme 7. Direct borylation of allylic alcohols.
This method can be used for regio- and stereoselective synthesis of functionalized
allylboronic acids 21a-j. As allylboronic acids are not sufficiently stable under solvent
free conditions,30,47 compounds 21a-j were converted30,43-45,65,99,144 to their potassium
trifluoro(allyl)borate derivatives 22a-j, which are useful precursors in Lewis acid and
7
transition metal-catalyzed allylation reactions.32,44,144 Because of the mild reaction
conditions and the use of the highly selective (and readily available37,119) pincer-complex
catalyst 16a, the products 22a-j were isolated in high yields (Table 1). Commonly used
palladium(0) sources, such as Pd2(dba)3 and Pd(PPh3)4, were found to be inefficient as
catalysts in the presented reactions.
Table 1. Palladium-catalyzed direct borylation of allylic alcohols.a
Entry
1
Ph
Ph
4
OH
20a
OH
2c
3
Cond. [oC/h]
Substrate
C5H11
C3H7
5
40/16
Allylboronic Acid
Ph
40/7
50/16
B(OH)2
C5H11
21b
OH
50/16
C3H7
20d
OH
B(OH)2
C3H7
OH
40/21
BnO
21f
HO
OH
20/16
HO
94
BF3K
B(OH) 2
B(OH) 2
22f
HO
87
BF3K
74
22g
OH
87
OH
B(OH)2
21h
COOMe
10e
BF3K
22h
COOMe
COOMe
50/24
20j
82
OH
B(OH)2
21i
OH
EtOOC
20k
40/20
92
BF3K
40/16
20i
98
22e
BnO
21g
OH
20h
HO
BF3K
22d
B(OH)2
60/16
90
BF3K
21e
BnO
22c
B(OH)2
21d
OH
BF3K
22b
21c
50/16
20g
11e
92
86
22a
C5H11
20c
20f
9d
BF3K
22a
20b
OH
6
8c
Ph
21a
20e
7
B(OH)2
21a
Yieldb
Product
EtOOC
21j
a
EtOOC
B(OH) 2
BF3K
22i
22j
BF3K
77
Unless otherwise stated, the reactions of 19 and the corresponding substrates 20 were conducted in the
presence of 16a (5 mol% ) in a mixture of DMSO and MeOH. After the indicated reaction times, aqueous
KHF2 was added. b Isolated yield. c MeOH was used as solvent. d A mixture of DMSO and water was used
as solvent. e p-toluenesulfonic acid 2.5 mol% (entry 11) or 5 mol% (entry 10) was used as co-catalyst.
8
2.2 Selectivity and Substituent Effects in the Borylation Reaction
The regioselectivity in the borylation reactions (Scheme 7, Table 1) was excellent, as
isomeric allylic alcohols 20a and 20b give the same regioisomer 22a. Similarly, the
branched alcohol 20c and the linear alcohol 20d provide linear allylboronates 22b and
22c, respectively. Unsubstituted cyclic alcohol 20f reacted smoothly providing the
corresponding boronate 22e in excellent yield. Allylic substitution in the presence of
hydroxy or benzyloxy substituents leads to an increased reactivity, as substitution of
20g-i proceeded at lower temperature (20-40 °C) than the corresponding reaction of the
alkyl substituted analogues 20c-f (40-60 °C). The relatively low temperature is also
essential for the high isolated yields of the products, since allylic hydroxyboronates
22g-h very easily undergo hydroxy-boronate elimination to give the corresponding 1,3diene. Compounds 20g-h reacted with excellent regioselectivity to give linear products
20f-g. Interestingly, substitution of 20h resulted in mono-borylated product 22g, and thus
the reaction can be used for desymmetrization of bisallyl alcohols.
Borylation of cyclic substrate 20i provides a single diastereomer 22h indicating that the
reaction is both regio- and stereoselective. The substitution pattern of 22h clearly shows
that the reaction proceeds with allylic rearrangement and with anti substitution. For
acyclic primary dialcohol 20h the borylation reaction affords the corresponding linear
product 22g, however in case of cyclic analogue 20i, the directing effect of the hydroxy
group leads to formation of the 1,2-substituted product 22h. Formation of this
regioisomer is particularly surprising, since usual palladium-catalyzed reactions via allylpalladium complexes bearing electron-withdrawing substituents usually gives the other
regioisomer, the 1,4-substituted product.107,145-147
In the presence of ester groups (20j-k), the rate of the borylation reaction drops
considerably. However, we found that addition of catalytic amounts (3-5 mol %) of
strong acid, such as p-toluenesulfonic acid (p-TsOH) accelerated the conversion of 20j-k,
affording the corresponding borylated products 22i and 22j. Again, 22i was obtained as a
single diastereomer, clearly indicating that the borylation reactions proceed with high
trans regioselectivity.
2.3 Solvent Effects – Methanol as Co-solvent
We found that a DMSO/MeOH mixture is indispensable to achieve high reaction rates
and high yields.33 When the reactions were performed in DMSO solely, low conversions
(about 5-20%) of the allylic alcohols were observed. On the other hand, catalyst 16a is
poorly soluble in methanol, and therefore, only the most reactive allylic alcohols 20h-i
react using methanol as solvent.
As mentioned above, allylic hydroxyboronates tend to decompose to 1,3-dienes. For 20h
this decomposition could to a large extent be avoided using methanol as solvent,
increasing the yield of the borylation reaction. We found that in a DMSO/water mixture
the conversion of 20i is slower than that in methanol; however, the corresponding
boronated product 21h/22h could be obtained in high yield, indicating that 21h is stable
in the presence of water as a co-solvent.
9
2.4 Activation of the Hydroxy Group
In order to gain insight into the activation of the hydroxy group, we carried out
competitive borylation experiments monitored by 1H-NMR (Figure 5). We have
previously shown30 that cinnamyl acetate can be converted into cinnamylboronic acid
21a in the presence of catalytic amounts of 16a and 19 (Scheme 6). Surprisingly, under
the same reaction conditions, cinnamyl alcohol 20a was converted significantly faster to
boronic acid 21a, than cinnamyl acetate. Thus, 20a was converted quantitatively to 21a in
8 hours, while its acetate derivative was still present in the reaction mixture after 11
hours. Addition of 5 mol% of p-TsOH had a dramatic effect on the rate of the reaction, as
borylation of 20a was complete in two hours, accounting for a 4-fold acceleration of the
reaction. Allylboronic acid 21a was surprisingly stable even in the presence of p-TsOH,
as in 9 hours only 5% of 21a was decomposed. The S-shaped curve obtained for
cinnamyl alcohol (•) indicates that the reaction requires an induction period, such as
activation of the hydroxy group. Alternatively, an auto-catalytic process with respect to
increasing acid concentration operates (Figure 5).
formation of 21a [%]
100
50
t [hours]
0
2
4
6
8
10
12
Figure 5. Formation of cinnamylboronic acid 21a from cinnamyl alcohol 20a (• , ∆) and
cinnamyl acetate (□) using 19 and 5 mol % of 16a in DMSO-d6/MeOH-d4 at 55 °C.
Effects of addition of 5 mol% p-TsOH (∆).
The above studies clearly indicate that under the employed reaction conditions, the
hydroxy group of the alcohol is converted to an excellent leaving group, which is easier
to displace than the acetate functionality. A possible explanation is that 19 acts as a
Lewis-acid catalyst by interacting with the free electron-pairs of the oxygen of 20a-k. A
similar activation is suggested in the Tamaru reaction, employing BEt3 for activation of
allylic alcohols.136-138 On the other hand, boronic acids are far less efficient Lewis-acids
than alkyl (or halo) boranes. Therefore, we envision another type of activation of the
hydroxy group, involving formation of allylboronic acid ester 24 (Scheme 8). This
esterification is probably facilitated by inclusion of a methanol molecule in the sixmembered ring transition state (23) of the process. In 24, the hydroxy group is converted
to a better leaving group, and moreover, the cleavage of the B-B bond is also facilitated
by coordination of the water molecule produced in the esterification. Application of
catalytic amounts of p-TsOH is supposed to catalyze the ester formation.
10
R
Me
O
H
H
OH
20
+
+ MeOH
R
HO
OH
B B
HO 19 OH
- MeOH R
O
OH
24
O
O H
B
OH
B(OH)2
23
B
OH2
B(OH) 2
Scheme 8. Proposed activation of the allylic alcohols.
In summary, we have shown that allylic alcohols can be converted to functionalized
allylboronates using diboronic acid 19 and catalytic amounts of pincer-complex 16a. The
above presented robust catalytic transformation employs the least expensive allylic
precursors reported for synthesis of functionalized allylboronates, and therefore this
method is also a good example for synthesis of highly value added chemicals.
2.5 Strategies for Fine-Tuning the Catalytic Activity of Pincer-Complexes
As discussed in Chapter1.3, one of the most important features of pincer-complexes is the
strong terdentate coordination between the metal and the pincer-ligand. Thus, when a
pincer-complex is the active catalyst in a transformation, the ligand remains tightly bound
to the metal center for the entire reaction, which allows an efficient transmission of the
ligand effects to the catalyst. Accordingly, the activity and selectivity of pincer-complex
catalysts are expected to be efficiently fine-tuned by substituents in the pincer-ligand.
However, the electronic effects of the ligand may also depend upon the location of the
substituent in the pincer-complex catalyst.
Our previous results30,33,130 suggested that the catalytic activity of 16a can be increased
for several catalytic transformations by increasing the electron density on palladium.
These studies inspired us to prepare several analogues (16d-f) of 16a (Figure 6) to
investigate the electronic effects of electron donating methoxy substituents upon the
catalytic activity of the complex in the borylation of allylic alcohols and the arylation of
vinyl epoxides.130
OMe
PhSe
Pd
Cl
16d
OMe
SePh
Se
MeO
Pd
Cl
Se
Se
16e
OMe MeO
Pd
Cl
16f
Se
OMe
Figure 6. Methoxy substituted palladium pincer-complexes applied in this study.
In complex 16d, the methoxy substituent is located para to the metal, while in 16e the
side arms are substituted. Expecting a synergic effect on the methoxy substitution, we
also prepared complex 16f, substituted in the para position and also on the side-arms.
11
2.5.1 Synthesis and Characterization of Methoxy Substituted Pincer-Complexes
Pincer-complexes 16d-f were synthesized by a slightly modified version of the procedure
reported by Yao and co-workers.119 Dibromoxylene derivatives 25a-b were reacted with
the appropriate diselenide (26a or 26b) to obtain pro-ligands 27b-d, which underwent a
transcyclometallation reaction111,114 with 28 to give complexes 16d-f in good to excellent
yields (Scheme 9).
R
R
R'
Br
26a R = H
26b R' = OMe
Br
25a R = H
25b R = OMe
N
Pd
Se
2
Se
Se
Cl
2
28
NaBH4
AcOH
R'
16d-f
yield 70-96%
R'
27b R = OMe, R' = H
27c R = H, R´= OMe
27d R = R' = OMe
Scheme 9. Synthesis of methoxy substituted palladium pincer-complexes.
Characterization of the complexes by 77Se-NMR spectroscopy revealed an interesting
trend. The 77Se-NMR shift values obtained for 16a132 (427.1 and 424.9 ppm) and 16d
(427.1 and 425.7 ppm) were almost identical, while methoxy substitution of the sidearms led to an increase of the shielding of the selenium nuclei (16e, 420.5 and 419.3
ppm; 16f, 420.7 and 419.1 ppm). This indicates that only methoxy substitution of the
side-arms affects the electron-density of the selenium atoms. Note that the two 77Se-NMR
shift values for each complex 16a,d-f is a consequence of the two diastereomeric forms
of the complexes (about 1:1 ratio).
2.5.2 Application of Methoxy Substituted Pincer-Complexes in Catalytic Arylation
of Vinyl Epoxide and Borylation of Cinnamyl Alcohol
Previously, we have shown that pincer-complex 16a catalyzes the ring opening of vinyl
epoxides and aziridines with organoboronic acids.130 In this reaction, an electron-rich
palladium centre was expected to increase the catalytic activity of the applied pincercomplex. Therefore, we investigated the effects of the methoxy substituted pincer-ligands
on the catalytic ring opening of vinyl epoxide 29a with phenylboronic acid (30)
(Scheme10).
Ph
O
+ PhB(OH)2
29a
30
OH
[16a,d-f ]cat
THF/H2O Cs2CO3
Ph
Ph
31
Scheme 10. Palladium pincer-complex-catalyzed opening of vinyl epoxide.
12
The progress of the reaction (formation of 31) was monitored by 1H-NMR spectroscopy
(Figure 7).
100
+
+ + + + + + + + + + + + +
Formation of 31 [%]
+
+
50
+
+
x
0 +x
0
x x
1
2
x
x
x
x
x
x
x
x
x
x
x
x
x
x
t [hours]
3
4
5
6
7
8
9
10
Figure 7. Catalytic phenylation of 29a affording 31 at room temperature.
Catalysts (2.5 mol %): 16a (•), 16d (+), 16e (x), and 16f (■).
Under the reaction conditions applied, the complete conversion of 29a to 31 required
about 10 hours using the parent catalyst 16a. Application of 16d led to a dramatic (4fold) acceleration of the process, which was completed in only two hours. Surprisingly,
methoxy substitution on the side-arms of the catalyst led to a weak deactivating effect.
Complex 16e proved to be slightly less reactive than the parent complex 16a, and the trimethoxy substituted catalyst 16f was less efficient than the mono-methoxy complex 16d.
Subsequently, we also studied the rate of borylation of cinnamyl alcohol 20a with
diboronic acid 19, in the presence of catalytic amounts of 16a, d and f (Table 1, entry 1
and Scheme 11).
Ph
OH
20a
+
HO
HO
B B
19
OH
[16a,d,f] cat
OH
DMSO/MeOH
Ph
B(OH)2
21a
Scheme 11. Palladium pincer-complex-catalyzed borylation of cinnamyl alcohol.
The reaction was completed in about 8 hours using the parent catalyst 16a. Similar to the
catalytic arylation of 29a, the borylation reaction was significantly accelerated using
para-methoxy complex 16d. Although the acceleration in the borylation reaction was less
extensive than in the ring-opening process, the catalytic transformation with the methoxy
complex 16d proceeded about twice as fast as with the parent catalyst 16a. On the other
hand, the reaction rate was unaffected by the methoxy substituents on the side-arms, as
the reaction was completed about as quickly with 16f as with 16d (Figure 8).
13
100
+ + + + + + + + + + + +
Formation of 21a [%]
+
50
+
+
+ + +
0 + +
0
2
+
t [hours]
4
6
8
10
12
Figure 8. Catalytic borylation of 20a to obtain cinnamylboronic acid 21a at 55°C.
Catalysts (5 mol%): 16a (•), 16d (+), and 16f (■).
The above results clearly indicate that the catalytic activity of pincer-complex 16a can be
efficiently fine-tuned by para-substitution of the complexes. The 77Se-NMR shifts (see
above) indicate that methoxy substitution of the side arms (16e and 16f) increases the
electron density on the selenium atom. Nevertheless, complexes 16e and 16f do not
display higher reactivity than the corresponding analogues 16a and 16d. Accordingly, the
most efficient fine-tuning of the catalytic activity of pincer-complexes can be achieved by
para-substitution of the aromatic ring, while substitution of the side-arms has much
weaker effects on the catalytic activity. These results are also in line with previous
studies on the electronic and magnetic properties of pincer-complexes. Van Koten and
co-workers111 have shown that the 195Pt-NMR shifts and the natural charges on the
platinum atom are in excellent correlation with the Hammet substituent constants of the
para-substituents in platinum pincer-complexes.
In summary, we have shown that in arylation and borylation reactions, the catalytic
activity of pincer-complex 16a can be effectively increased by para-methoxy substitution
of the catalyst (16d). On the other hand, methoxy substitution of the side-arms does not
increase the catalytic activity of the complexes or even has a slight deactivation effect.
These substituent effects probably apply to other pincer-complex-catalyzed substitution
reactions, in which charge accumulation on the metal atom is required to increase the
catalytic activity of the complex.
14
3. One-Pot Allylation of Carbonyl Compounds via Transient Allylboronates
Generated by Palladium-Catalyzed Substitution of Allylic Alcohols (Papers III-V)
The synthetic applications of allylboronates as allylating reagents were discussed in
Chapter 1.2. An important feature of these reagents is their highly selective and effective
carbon-carbon formation in coupling reactions with carbonyl compounds such as
aldehydes, ketones and imines. As a consequence, there is a high current demand for new
efficient synthetic procedures providing functionalized allylboronates. However,
development of these procedures raise two important synthetic issues: (i) the synthetic
procedures have to be highly stereo- and regioselective, and have to tolerate many
functionalities; and (ii) the prepared functionalized allylboronates are required to be
stable under the purification and isolation procedures, yet sufficiently reactive in the
desired allylation reactions. The development of transition metal based synthesis of
functionalized allylboronates29-31,38,90-103 is an important strategy to address problem (i)
above. Considering the relatively low stability of the functionalized allylboronates, their
isolation and purification often becomes a major issue [(ii) above], which may encumber
the synthetic application of allylboronates. Allylboronic acids are more reactive allylating
reagents46,63 than allylboronate esters, however, these species rapidly decompose
(probably oxidize) under solvent free conditions.30,31,34,47 An attractive solution for the
above purification issues is the development of one-pot procedures, in which the transient
allylboronates are not isolated but directly reacted with the corresponding allyl
acceptors.29,31,99,103,148 Development of these procedures requires highly selective
formation of allylboronates and carefully designed reaction conditions to avoid undesired
side products in the multi-component processes.
In Chapter 2, a versatile method for preparation of functionalized allylboronates, starting
from allylic alcohols, was presented. We have found35,37 that this efficient borylation
procedure can be integrated in a one-pot sequence for stereo- and regioselective carboncarbon bond coupling of allylic alcohols 20 and aldehydes 2, ketones 32 or in situ
generated imines (from 9 and 2a) (Scheme 12).
R2-CHO
2b-e
OH
R2
R1 4a-i
OH + 14/19
R1
20
O
O
B B
14
[16a,b] cat
[p -TsOH] cat
O
HO
O
HO
B B
19
OH
R1
B(OH)2
21
not isolated
OH
R 3R4CO
32a-b
[InI]cat
R 5-NH2
9a-c
HOOC CHO
2a
OH
R4
R3
R1 4j-l
NHR 5
COOH
R1 10a-h
Scheme 12. One-pot transformations of in situ generated allylboronic acids.
15
The selective coupling reactions proceed via palladium pincer-complex- (16a,b) (Figure
4.) catalyzed generation of transient allylboronic acids 21 using bis(pinacolato)diboron
(14) or diboronic acid (19) as boronate source. These transient intermediates are proved
to be reasonably acid, base, alcohol, water and air stable species, which allows a high
level of compatibility with the reaction conditions of the allylation of various
aldehyde/ketone and imine electrophiles. The regio- and stereoselectivity of the reaction
is excellent, as almost all products are formed as single regio- and stereoisomers.
3.1 One-Pot Allylation of Aldehydes
Allylation of aldehydes using allylic alcohol as allylating reagent is a highly efficient and
attractive synthetic route to functionalized homoallylic alcohols. Allylic alcohols are one
of the least expensive and most accessible allylating agents, and therefore carbon-carbon
couplings between allylic alcohols and aldehydes represent excellent examples for value
added synthesis. However, a usual problem with these types of reactions is the activation
of the allylic carbon-oxygen bond of the alcohol functionality. Many excellent solutions
for this problem were presented by using palladium-catalyzed substitution9,10,13 of the in
situ activated hydroxy group of the allylic alcohols. The most important strategies for
activation of the hydroxy group136,149 involve application of SnCl2,139,150-152 BEt3,153-155
Et2Zn156 and various indium salts.157-159 As these transformations involve a two step
procedure using Lewis acids and/or reductive organometallic reagents, the functional
group tolerance of the reaction may be problematic. Another problem is often the
unsatisfactorily low stereoselectivity of the reaction.
The majority of these selectivity problems can be avoided by converting the hydroxy
group of the allylic alcohol to a boronate. In these reactions the hydroxy group is
activated by diboronic acid (Chapter 2.4), which does not affect the usual functional
groups, and therefore the reaction tolerates carbonyl, cyano, aromatic halogenide and
nitro groups. As the allylation proceeds via a cyclic six-membered ring TS (Scheme
2),15,18,66 the coupling of the transient allylboronate with aldehydes proceeds with a
remarkably high stereoselectivity. In fact, all the presented coupling reactions of
aldehydes with in situ generated allylboronates provided a single diastereomer without
traces of the other isomer. Considering the allylation reactions of aldehydes with allylic
alcohols, we have also extended the synthetic scope of the reactions, by application of
commercially easily available diboronate 14 and a new SCS-catalyst, 16b, to accelerate
the rate of the reactions (Scheme 12).
The coupling reaction between the corresponding allylic alcohol 20 and aldehyde 2 can
be performed as an operationally simple one-pot, one-step procedure. Accordingly, in a
typical reaction, reactants 20, 14/19 and 2 and catalysts 16a-b119,122,135 (5 mol%) and pTsOH (5-20 mol%) were dissolved in a mixture of DMSO and methanol (1:1) at the
outset of the reaction. After the allotted reaction times given in Table 2, the coupling
products (4a-i) were isolated in good to excellent yields (Table 2).
16
Table 2. Representative entries for allylation of aldehydes via transient allylboronates.a
Entry
1
b
Alcohol
Aldehyde Diboron Cat.
O
Ph
OH
2f
20a
20a
3
20a
4
20a
Ph
2b
2b
O
5
19
16a
40/16
19
16a
40/48
14
16a
50/24
19
16a
40/16
Product
OH
OH
C5H11
16b
14
16a
19
2e
OH
Br
4c
OH
50/36
69
C5H11 4d
OH
7
8 BnO
87
87
O
20e
96
50/24
20c
OH
4a
4a
4b
OH
Ph
2d
88
e
93
2d
6g
Ph
Ph
14
Yield
4a
OH
Br
O
20a
Cond.d
Ph
2b
2c
c
2b
16a
77
50/16
NO2
4e
OH
19
16a
Ph
40/16
20g
NO2
96
4f
BnO
OH
9g
20g
2d
19
16a
50/36
78
BnO
OH
10
H
2b
19
16a
50/36
4g
OH
Ph
93
11
COOMe
20j
20j
2b
14
16a
50/48
4h
92
12
20j
2b
14
16b
50/6
4h
OH
87
2b
19
16a
60/16
Ph
82
13
OH
20l
COOMe 4h
4i
In a typical reaction 20, 14/19 (1.2 equiv.) and 2 (1.2 equiv.) were dissolved in a DMSO/MeOH mixture
in the presence of catalytic amounts of 16a/b and p-TsOH (each 5 mol%). b When 14 was applied, 8.0
equiv. of water and 20 mol% of p-TsOH were used. c Catalyst. d Temperature/time [°C]/[h].
e
Isolated yield. f Without addition of p-TsOH . g 3.0 equiv. of aldehyde was used.
a
17
We have found that primary, secondary, cyclic and acyclic alcohols react with an
excellent stereo- and regioselectivity to give the branched homoallylic products
exclusively. In the described processes, the homoallylic alcohol products 4a-d,f-h were
obtained as single diastereomers. The reactions from acyclic allylic alcohols 20a,c,g, give
the anti products, as the transformations proceed via allylboronic acids 21a,b,f (See
Table 1) in which the geometry of the double bond is trans. The reaction proceeds readily
and with high diastereoselectivity even for cyclic allylic alcohols as 20j. In these
reactions the double bond in the cyclic allylboronic acid intermediate 21i (Table 1) has
cis geometry, and thus the new carbon-carbon bond forms with syn diastereoselectivity. It
is remarkable that starting from 20j, a selective tandem stereoinduction can be achieved,
providing the product as a single diastereomer out of the four possible (Table 2, entry
10). The excellent stereoselectivity is a consequence of the highly stereoselective
formation of the allylboronic acid intermediate 21i (Table 1) and the subsequent selective
coupling with the aldehyde. Furthermore, the high selectivity was maintained even for
tertiary alcohols. The reaction of alcohol 20e with aldehyde 2e resulted in formation of
the branched isomer 4e exclusively, creating a new quaternary carbon center (Table 2,
entry 7).
As shown in Chapter 2, the transient allylboronic acids derived from the alcohols in
Table 2 can be isolated and fully characterized. Probably, the only exception is the
dienylboronic acid intermediate (21l) formed by borylation of 20l (Table 2, entry 13),
which resisted all attempts to isolation. However, due to the one-pot/one-step conditions,
the unstable 21l intermediate was allowed to react directly with 2b after its formation,
and thus the corresponding product 4i could be isolated in high yield.
As previously communicated,30,33 the efficient borylation procedures are based on
application of diboronic acid 19, which is a highly efficient reagent in the pincercomplex-catalyzed transformation of allylic alcohols (Chapter 2) and other easily
accessible substrates. Furthermore, the only by-product using 19 is non-toxic boric acid.
Unfortunately, the poor commercial availability of 19 can be considered as a limiting
factor and therefore we studied the possibility to replace 19 with commercially available
bis(pinacolato)diboron 14. In general, diboronic reagent 14 performs much less
efficiently in the pincer-complex-catalyzed borylation reactions, than 19 does.30
Accordingly, using 14 under the standard reaction conditions described above, the
conversion into the homoallylic product 4 is low. We have reasoned that the in situ
hydrolysis of 14 to 19 could help to improve the performance of the bispinacolato reagent
14 in the coupling reactions. We have previously reported33 that addition of water does
not inhibit the robust borylation process of allylic alcohols. However, the catalytic
reactions are slightly slowed down, and hence the in situ hydrolysis of 14 to 19 has to be
done by addition of as small amount of water as possible. Our optimization studies
showed that the best results were achieved by addition of 8 equivalents of water and
increasing the amount of p-TsOH to 20 mol%. This modification did not affect the
isolated yields in the coupling reaction significantly. However, the required reaction
times were usually longer with 14, than with 19 (c.f entries 1 and 3, or 10-11 in Table 2).
This is probably because of the decelerating effect of water in the borylation process.
18
Although the coupling reactions can be carried out in the absence of p-TsOH (Table 2,
entry 2), the reaction proceeds much faster when catalytic amounts of p-TsOH is present
in the reaction mixture (c.f. Table 2, entries 1-2). As pointed out in Chapter 1-2,21,24,28,33
the addition of p-TsOH has an important catalytic effect in the borylation reaction as well
as in the allylation of aldehydes. Accordingly, in the procedures when bispinacolato
derivative 14 is used as borylation agent, p-TsOH catalyses three distinct processes under
the applied one-pot conditions: (i) hydrolysis of 14 to 19 in the presence of water, (ii)
borylation of alcohol 20 to allylboronate 21 and (iii) allylation of aldehyde 2 with 21.
Some structural features in the alcohol substrates, such as the presence of carboxy
functionality and the cyclic topology of the substrate (20j), leads to a decrease in
reactivity.33 As application of 14 in place of 19 leads to a further slowdown of the
processes, some of the reactions required extended reaction times (e.g. Table 2 entry 11).
We attempted to accelerate these reactions by employing other catalysts. It was found
that application of SCS-catalyst 16b135 instead of 16a119 led to a much faster coupling
reaction, without significant decrease of the yields. For example, the coupling reaction of
cyclic substrate 20j with benzaldehyde (2b) using 14 as boronate source required 48 h
using 16a as catalyst (Table 2, entry 11), while the same reaction with 16b could be
completed in only 6 h (Table 2, entry 12).
The obtained stereodefined and densely functionalized homoallylic products involve
functionalities that directly can be employed in further catalytic processes. For example,
applying 4-pentenal (2d) as aldehyde component with alcohols 20a, 20c and 20g we
could perform an operationally simple one-pot synthesis of stereodefined 1,7-dienes (4c,
4d and 4g). These products can easily be cyclized (Scheme 13), affording stereodefined
cyclohexene derivatives (33a-c) using Grubbs RCM procedure. 50,160
OH
[34]cat
CH2Cl2
Z
OH
Z
4c
Z= Ph
33a
yield (%)
96
4d
Z= C5H11
33b
89
4g
Z= CH2OBn
33c
97
Mes N N Mes
Cl
Ph
Ru
Cl
PCy3
34
Scheme 13. Synthesis of stereodefined cyclohexenes from the products of the one-pot
allylation of 2d with various allylic alcohols.
In conclusion, functionalized regio- and stereodefined homoallylic alcohols may easily be
prepared from aldehydes and allylic alcohols via catalytic generation of allylboronates in
a one-pot sequence. The reaction displays a high functional group tolerance and broad
synthetic scope, involving primary, secondary and tertiary alcohols with both cyclic and
acyclic architecture. By a slight change of the reaction conditions, the one-pot procedures
were performed using commercially readily available bis(pinacolato)diboron (14) in
place of diboronic acid (19). Furthermore, a sulfur-based pincer-complex catalyst 16b
was introduced as a complement for the very efficient and easily accessible complex 16a.
19
3.2 One-Pot Allylation of Ketones
Although allylboronates do not react directly with ketone substrates,10,13 these
transformations can be performed by activation of one of the reactants.42,53,57 Recently,
Kobayashi and Schneider53,55,56 reported an operationally simple and elegant way to
activate of allylboronates by using InI as catalyst. We have found that application of InI
catalysis can also be employed for activation of allylboronic acids generated in situ from
allylic alcohols (Table 3).
Table 3. One-pot allylation of ketones using InI as co-catalyst.a
Entry
Ketone
Alcohol
O
1 Ph
OH
Ph
32a
20a
2
c
Diboron Cat.b Cond.
19
16a
50/16
4j
72
e
OH
O
20a
19
16a
50/16
73
Ph
4k
OH
OH
C5H11
20c
Ph
Ph
32b
3
Yield d
Product
OH
32b
19
16a
50/16
86
C5H11
4l
In a typical reaction 20, and 19 (1.2 equiv.) were dissolved in a DMSO/MeOH mixture in the presence of
catalytic amounts of 16a and p-TsOH (each 5 mol%). After the borylation was completed, 32a/b (1.2
equiv) and InI (20 mol%) were added. b Catalyst. c Temperature/time [°C]/[h]. d Isolated yield. e Formation
of 10 % of the syn diastereomer was also observed.
a
The reaction was conducted similarly to the allylation of aldehydes, except that the
ketone substrate (32a-b) and InI (20 mol%) was added after completion of the borylation
step. This slight modification was necessary to avoid deactivation of catalyst 16a by InI
in the borylation reaction. The limited number of reactions we have carried out with
alcohols 20a and 20b with ketones 32a-b suggest a broad synthetic scope. The
regioselectivity of the process is excellent, as the branched allylic isomer with a new
quaternary carbon center is formed as the only product. The diastereoselectivity of the
coupling of 20a and acetophenone (32a) is still high (Table 3, entry 1), as the two
diastereomers are formed in a 9:1 ratio. As expected,10,13,161 this level of stereoselectivity
is somewhat lower than in the corresponding reaction with benzaldehyde (Table 2, entry
1). The high regio- and stereoselectivity and the high yield in the presented reactions
clearly shows that the one-pot allylation reactions using alcohols as allyl sources can
easily be extended to ketone substrates as well.
Although, the above results clearly demonstrate the benefits of the operationally simple
multi-component coupling of easily available allylic alcohols and carbonyl compounds,
we wished to further extend the synthetic scope of the reaction by integrating the Petasis
Borono-Mannich reaction46,51,52,69-77as a crucial segment in our one-pot approach.
20
3.3 One-Pot Synthesis of α-Amino Acids via in situ Generated Allylboronates
Petasis and co-workers69-73 have reported a new powerful three component method for
coupling of alkenylboronic acids with amines and aldehydes to obtain homoallylic
amines, including α-amino acids.162-169 This efficient coupling reaction have found many
applications,70,72,74,170-173 and recently, it was also extended by Kobayashi and Thadani
and their co-workers46,51,52 to coupling of allylboronates with aldehydes and ammonia.
These reactions proceed with high regio- and stereoselectivity, however the synthetic
scope is limited by the poor availability of functionalized allylboronates. Therefore, we
decided to extend the three-component Petasis Borono-Mannich reaction to a four
component coupling, involving in situ generation of the allylboronate component. As far
as we know, allylic alcohols were not previously used as reagents in the allylation of
imine substrates. Furthermore, the Petasis Borono-Mannich reactions were not carried out
with in situ generated organoborane substrates either.
We found that several elements of the above described coupling of alcohols and
aldehydes can be employed in combination with the Petasis reaction. The most important
difference was that amines 9a-c and glyoxylic acid (2a) were not added at the outset of
the coupling reaction, but after completion of the borylation process. Addition of the
amine component in the beginning of the reaction inhibited the formation of allylboronate
21. Thus, the synthesis of α-amino acids 10a-h from allylic alcohols 20 could be
performed as a sequential one-pot reaction, where the in situ generated allylboronic acids
were coupled with in situ formed imines 35a-c (Scheme 14).
O
OH +
H
O
2a
R'
OH
20
NR
-H2O
R-NH2
9a-c
14/19 +
[16a,b] cat
[p -TsOH]cat
H
OH not isolated
O
35a-c
NHR
R'
B(OH)2
21
not isolated
R'
COOH
10a-h
Scheme 14. In situ formation of imines used in the Petasis reaction
with in situ generated allylboronates.
The structural diversity of the amino acid products 10a-h is created by the various
substituents of the allylboronates 21 that can be prepared from simple precursors without
cumbersome isolation and handling of the allylboronate reagents. First we studied the
substrate scope of the reactions. It was found that primary (20a,g), secondary (20c,k,m)
and tertiary (20e) alcohols perform equally well in the reactions, and that benzyloxy
(20g) and carbethoxy/methoxy (20k,m) groups are tolerated (Table 4).
21
Table 4. Petasis Borono-Mannich reaction with in situ generated allylboronates.a
Entry
1
Alcohol
c
d
e
Diboronb Cat. Cond. A Cond. B
Amine
Ph
Ph
OH
H 2N
20a
Ph
19
16a
40/16
HN
25/16
9a
2
3
H 2N
OH
40/16
25/16
9b
19
16a
50/16
HN
25/8
9a
9a
19
14
16b
16b
50/4
50/4
25/16
25/16
19
16a
50/16
25/8
10c
10c
OMe
20c
H 2N
HN
9c
9a
52
C5H11 10d
Ph
19
16a
50/16
HN
25/24
Ph
8 BnO
OH
HN
9a
19
16a
40/8
25/16
Ph
COOH
10f
20g
BnO
OH
EtOOC
19
16a
50/16
25/16
MeOOC
9a
19
16a
50/16
25/16
Ph
COOH
78
10g
EtOOC
O
OH
75
Ph
HN
9a
20k
10
60
COOH
10e
Ph
20e
9
OMe
83
77
COOH
OH
7
75
C5H11 10c
20c
20c
6
Ph
COOH
20c
4
5
78
COOH
10b
Ph
Ph
9a
C5H11
16a
76
COOH
10a
Ph
HN
19
Yield
Ph
Ph
20a
f
Product
Ph
Ph
N
Ph
80
COOH
10h
20m
In a typical reaction 20 and 14/19 (1.2 equiv.) were dissolved in a DMSO/MeOH mixture in the presence
of catalytic amounts of 16a/b and p-TsOH (each 5 mol%). After the allotted times (cond. A), 9 (2.0 equiv.)
was added followed by addition of 2a (1.5 equiv.) and the reaction was continued for the times and
temperatures given in column cond. B. b When 14 was applied, 8.0 equiv. of water and 20 mol% of
p-TsOH were used. c Catalyst. d Temperature/time [°C]/[h] for the borylation reaction e Temperature/time
[°C]/[h] for the allylation reaction f Isolated yield. All products except 10h were isolated as HCl salts.
a
22
In the four-component coupling reactions we observed a stereo- and regioselectivity
equivalent to the analogue coupling reaction of allylic alcohols and aldehydes. Hence, αamino acid derivatives (10a-h) were formed as single regio- and stereoisomers from
readily available alcohols and amines. The benzhydryl (10a,c,e-h) and anilyl (10b)
derivatives could be purified by silica gel chromatography with addition of HCl to the
eluent. However, methoxy-anilyl derivative 10d displayed some tendency for
decomposition and therefore it was purified by recrystallization. Accordingly, the
presented reactions provide an easy access to stereodefined analogues and homologues of
natural amino acids, such as phenylalanine (10a,b), isoleucine (10c,d), valine (10e),
serine (10f), glutamic acid (10g) and pyroglutamic acid (10h).
An interesting domino reaction can be triggered by using ester functionalized allylic
alcohols 20k and 20m as precursors. When ethyl ester derivative 20k was reacted under
the standard reaction conditions, glutamic acid analogue 10g was formed with high
selectivity and in high yield. However, upon heating, product 10g underwent lactam
formation reaction, affording pyroglutamic acid derivative 10h. Interestingly, starting
from methyl ester derivative 20m, the corresponding glutamic acid analogue (10g) could
not be isolated, as it underwent spontaneous cyclization affording 10h as the final product
of the process. Accordingly, the reaction of 20m, 19, 9a and 2a triggers a spectacular
four-step cascade, which can be performed as a one-pot reaction (Scheme 15).
Ph
OH
ROOC
HN
1) 19/[16a]ca t/[p -TsOH]c at
Ph
COOH
2) 9a + 2a
20k: R=Et
20m: R=Me
NH2
10g, R=Et
yield 78 %
ROOC
COOH
HOOC
L-Glutamic acid
R=Me
O
N
Lactam formation
O
Ph
Ph
COOH
10h yield 80 %
NH
COOH
L-Pyroglutamic acid
Scheme 15. Lactam formation in the Petasis reaction of ester substituted allylic alcohols,
obtaining glutamic and pyroglutamic acid derivatives.
In the coupling reactions we have employed benzyl and aryl amines 9a-c, which proved
to be beneficial to avoid allylic rearrangement of the products. According to Kobayashi
and co-workers,52 branched homoallylic α-amino acids with unprotected amino groups
show a tendency for rearrangement to the corresponding linear isomers via aza-Cope
rearrangement. This process is most extensive for products formed with quaternary
carbon center,52 such as 10e (Table 4, entry 7), but even in this case, formation of the
linear isomer could be avoided completely.
23
Similarly to the coupling reactions of alcohols and aldehydes (Table 2), the four
component coupling reactions proceed smoothly when diboronic acid 19 is replaced with
bis(pinacolato)diboron 14 (Table 4, entry 5). The required modifications of the reaction
conditions, such as addition of water (8 equiv.) and increasing the catalyst loading of pTsOH (20 mol%) gave comparable isolated yields (c.f. Table 4, entries 3-5). Use of
diboronic acid 19 (either directly or in situ hydrolyzed from 14) to generate the transient
allylboronic acids 21 is probably an important factor to obtain high yield and reactivity in
the Petasis reaction. Thadani and co-workers46 have pointed out that the analogue
coupling reactions of (isolated) allylboronates with ammonia and aldehydes proceed most
efficiently when allylboronic acids are used instead of allylboronic esters. This can be
explained by the fact63 that allylboronic acids are more reactive in allylation reactions,
than the allylboronic ester analogues.
To conclude this part, we have demonstrated that the palladium pincer-complexcatalyzed borylation of allylic alcohols can be combined with the Petasis BoronoMannich reaction. The products of these one-pot multi-component reactions are densely
functionalized stereodefined α-amino acids which are useful drug intermediates162-169
3.4 Coupling of Catalytically Generated Allylboronates with in situ
Hydrolyzed Acetals
As a part of our research program focused on development of one-pot reactions based on
catalytically generated allylboronates, we have studied the possibility to use in situ
hydrolyzed acetals as electrophiles. This approach is particularly useful when the acetals
are more stable or more easily accessible than the requisite aldehyde. Moreover, this
method gives access to novel stereodefined homoallylic alcohols, epoxides and amino
alcohols. Indeed, we have found that under the reaction conditions of pincer-complex(16a-b) catalyzed borylation with bis(pinacolato)diboron (14) or diboronic acid (19),
various acetals, 36a-e can readily be hydrolyzed by water in the presence of catalytic
amounts of p-TsOH. (Scheme 16).
R
OH
14/19 +
[16a,b]cat
20
R
OAc
12
EtOOC COOEt
+
OR''
R''O
R'
36
[p-TsOH]cat
DMSO/MeOH/H2O
OH
R'
R
4m-u
17
Scheme 16. One-pot allylation of in situ hydrolyzed acetals.
In these processes, a broad variety of allylic substrates, including allylic alcohols (20),
acetates (12) and vinyl cyclopropane (17), can be used. The overall reaction results in
functionalized, regio- and stereodefined homoallylic alcohols (4). A typical reaction
could be performed as a real one-pot process by mixing diboronate 14/19, pincer-catalyst
16a/b, acetal 36 and allylic substrate 12/17/20 in a mixture of DMSO, methanol and
water to obtain the homoallylic alcohol products (4) in good to excellent yields (Table 5).
24
Table 5. Selective one-pot allylation of in situ hydrolyzed acetals.a
Entry
1
Substrate
Acetal
OEt
OH
Ph
EtO
20a
2
3d
4e
NHTs
MeO
MeO
70/24
16a
16a
16a
Ts
N
92
89
4n
OH
50/16
81
4o
Ph
OH
Cl
36c
Cl
Yieldc
4m
70/24
Ph
OMe
20a
16a
Product
OH
NHTs
Ph
OH
OMe
20a
Cond.b
36a
OEt Ts
N
EtO
36b
20a
Cat
70/24
Ph
36d
78
4p
O
5 e,f
20a
36d
16a
70/24
OH
H
6
36a
16b
Ph
OH
29b
NHTs
76
70/24
COOMe 20j
4q
COOMe
O
7
68
36a
12a
16b
70/16
16a
70/24
4q
73
O
8
OH
OMe
36e
20n
OAc
9g
MeO
36a
OH
NHTs
70/24
12b SiMe3
11
12b
78
SiMe3 4s
OH
10g
4r
OH
16b
74
36b
16b
70/24
Ts
N
73
SiMe3 4t
OH
Ts
N
EtOOC COOEt
36b
16b
17
70/24
COOEt
69
COOEt 4u
Unless otherwise stated, the allylic substrate, 14 (1.2 equiv.) and 36 (1.2 equiv.) were dissolved in a
DMSO/MeOH/H2O mixture in the presence of catalytic amounts of 16a/b (5 mol%) and p-TsOH (20
mol%). b Temperature/time [°C]/[h].c Isolated yield d Sequential one-pot reaction was performed e 50
mol%) p-TsOH was used f After in situ formation of 4p, KOH was added. g 19 and 20 mol% LiOAc was
used in a sequential one-pot reaction.
a
25
The use of acetals is advantageous from a synthetic point of view, since the aldehydes
formed by hydrolysis of these acetals are either difficult to access (36a,b), unstable (36d)
or difficult to handle (36c,e). Protected amino aldehydes 36a,b reacted smoothly under
standard reactions conditions, providing ready access to various functionalized
stereodefined amino alcohols 4m,n,q,s-u. Chloro substituted acetal 36d was used to
prepare chlorohydrins such as 4p. By addition of base, chlorohydrin 4p can easily be
converted to the corresponding epoxide 29b (Table 5, entry 5) without isolation of 4p.
Thus, this procedure, involving hydrolysis of acetal 36d, borylation of alcohol 20a, their
coupling to obtain 4p and the subsequent epoxide formation (29b), can be considered as a
four-step, one-pot sequence. Another synthetically important transformation can be
achieved using dimethoxy methane 36e, which is easier to handle and less toxic than
formaldehyde. The overall result of this one-pot reaction is a homologization of the
allylic substrate with allyl shift (Table 5, entry 8).
Although the majority of the reactions could be performed as the above described real
one-pot procedure, a sequential approach had to be applied in a few cases. Acrylaldehyde
formed by hydrolysis of 36c inhibited catalyst 16, and therefore 36c was added after
completion of the catalytic borylation process (Table 5, entry 3). Likewise, silyl
substituted substrate 12b undergoes rapid Peterson elimination even under mild acidic
conditions, and consequently, p-TsOH and acetals 36a-b had to be added after borylation
of 12b (Table 5, entries 9-10). Under the neutral borylation conditions,
bis(pinacolato)diboron (14) was not suitable as a boronate source, and therefore in the
reactions with 12b, diboronic acid 19 was used. Interestingly, the corresponding
allylboronate30 with an allylsilane functional group is stable under the mildly acidic
conditions of the acetal hydrolysis and the densely functionalized products 4s-t were
isolated in high yields.
The above described procedure involves three individual reactions performed in concert.
The first process is based on our catalytic borylation of allylic substrates (Chapter 2), to
give allylboronates. As previously discussed, these processes are accelerated by catalytic
amounts of p-TsOH, and can also be performed in the presence of water. These reaction
components can also be used for the in situ hydrolysis of acetals 36. The aldehydes
formed in this process immediately react with the allylboronate in a highly regio- and
stereoselective process (Scheme 17).
[p-TsOH]cat
OR''
R''O
R
OH
20
R
OAc
12
EtOOC COOEt
H2O
R'
36
14/19 +
[16a,b] cat
[p -TsOH]cat
DMSO/MeOH/H2O
R'CH=O not isolated
2
OH
R
B(OH)2
21
not isolated
R'
R 4m-u
17
Scheme 17. One-pot coupling of catalytically generated allylboronates
and in situ hydrolyzed acetals.
26
This approach ensures that unstable or highly reactive aldehydes formed from 36 do not
accumulate in the reaction mixture, leading to a smooth reaction without decomposition
or catalyst inhibition. Accordingly, the homoallylic product 4 is formed as a single regioand stereoisomer, without significant formation of by-products, indicating that both the
borylation and allylation processes are highly selective. Overall, the reaction can be
classified as cooperative concerted catalysis3-6 involving a catalytic action by palladium
pincer-complex 16 on the allylic substrate and by p-TsOH on both the allylic substrate
and on acetal 36, followed by coupling of the transient allylboronate and aldehyde
intermediates. Accordingly, the presented new one-pot procedure involves three perfectly
synchronized catalytic actions, of which two are catalyzed by p-TsOH and one is
catalyzed by pincer-complex 16. The stereo- and regiodefined products obtained in the
above described operationally simple procedures are useful intermediates for advanced
organic synthesis. This can be demonstrated by synthesis of stereodefined heterocyclic
compounds174-177 from 4n,t,u using Hoveyda-Grubbs catalyst160,178 37 (Scheme 3). This
ring-closing metathesis reaction does not affect the stereocenters of the dienes, thus it can
be employed for synthesis of stereodefined azepines 38a-c, which are analogues of potent
glycosidase and protein kinase inhibitors.174-176
OH Ts
N
R
OH
[37]cat
R
4n R=Ph
4t R=CH2SiMe3
4u R=CH2CH(COOEt) 2
NTs
yield (%)
38a
95
38b
92
38c
94
Mes N
Ru
Cl
O
i-Pr
N Mes
Cl
37
Scheme 18. Synthesis of stereodefined azepines by ring-closing metathesis.
In summary, we have devised a new one-pot procedure for allylation of in situ
hydrolyzed acetals using easy accessible substrates, such as allylic alcohols, acetates and
vinylcyclopropanes. This procedure involves three individual processes under the same
reaction conditions: borylation of the allylic substrates, hydrolysis of the acetals and
highly selective carbon-carbon bond formation by the in situ generated allylboronates and
aldehydes. This sequence of fully compatible cooperative processes involves three
perfectly synchronized catalytic actions by p-TsOH and palladium-pincer-complex 16.
The reaction is particularly useful for in situ generation and application of unstable and
sensitive aldehydes. Because of the one-pot approach, all discrete reactions share the
same solvent and at least two purification steps can be avoided.
27
3.5 Mechanistic Aspects of the One-Pot Allylation of Carbonyl Compounds
In this chapter, several applications of catalytically generated allylboronates have been
presented. All of these one-pot reactions are based on the versatile palladium pincercomplex-catalyzed borylation of allylic substrates presented in Chapter 2. In this section,
some general mechanistic aspects of these transformations will be discussed. Although
the in depth mechanistic details of the distinct chemical reactions involved in the above
one-pot procedures are not fully understood, the consecutive processes and their
connectivity can be described based on the above studies and literature reports on
allylation reactions with allylboronates.9,10,13-15,52,66,70,72,179
Accordingly, the coupling reaction of allylic alcohols and corresponding electrophiles is
initiated by the activation of the hydroxy functionality of 20 with diboronic acid 19 as
proposed in Chapter 2. Diboronic acid 19 is either present as a reactant or it is generated
by hydrolysis of 14. It was found that under the reaction conditions when 14 is used as a
boronate source (i.e. in the presence of 8 equiv. of water and 20 mol% of p-TsOH),
immediate formation of free pinacol can be observed. Hydrolysis of 14 to 19 under these
conditions is probably an equilibrium process, as after extended reaction times the ratio
of the boron bound and free pinacol (3:1) is about constant. We have previously
suggested (Chapter 2) that the activation of the hydroxy group of 20 takes place by
formation of boronic ester 39 (Figure 9). This is a very mild way of activation compared
to the analogue SnCl2,139,150-152 BEt3,153-155 Et2Zn156 and indium157-159 based methods,
which explains the broad synthetic scope and high functional group tolerance of the
above presented reactions.
R 2-CHO
2
+ H2O
1
R
1
R
21
B(OH)2
+ B(OH) 3
PhSe
Pd
SePh
L
16g
B2pin2
14
O
OH
B B
HO
OH
39
[p -TsOH]cat
H2O
B2(OH) 4
19
OH
R2
4
+ B(OH) 3
Allylation
Activation
Borylation
R1
OH
20
R1
R1
R1
40
O
40
HO
B L + H2O
PhSe
Pd
SePh
O
B L
HO
B(OH)2
16h
Figure 9. Plausible mechanism for the allylation of aldehydes using allylic alcohols.
28
The detailed mechanism of the catalytic formation of 21 from 39 is a subject of ongoing
and future studies, however, we strongly suggest that the catalytic transfer of the boronate
group from diboronic acid 19 or its derivatives (such as 14) proceeds by a similar
mechanism as the analogue stannylation process. Previously, Szabó and co-workers127,131
have shown that hexamethylditin (dimetallic analogue of 19) readily reacts with pincercomplexes under stoichiometric conditions to give an η1-coordinated mono-stannyl
pincer-complex, which is the tin analogue of 16h. Subsequent DFT modeling studies
have shown that the stannyl group is easily transferred from palladium to the allylic
position of unsaturated carbons.131 Accordingly, we suppose that the boronate group is
transferred from diboronate 19 to catalyst 16g, affording η 1-boronato complex 16h.
Subsequently, the boronato group substitutes the activated hydroxy group of 40 to give
allylboronic acid 21 and boric acid. In the absence of aldehydes or other electrophiles,
these allylboronic acid derivatives (21) can also be isolated, as described in Chapter 2.
Using functionalized allylic alcohols, the borylation reaction takes place with an excellent
stereo- and regioselectivity, which is very important for the subsequent allylation
processes. The borylation reaction shows a clear preference for formation of the linear
allylboronates, as both primary (e.g. 20a) and secondary (e.g. 20b) allylic alcohols give
the corresponding terminally borylated products (Chapter 2).
The next step of the process is the allylation of aldehyde 2, which proceeds with an
excellent stereo- and regioselectivity, as the reaction takes place via a six-membered ring
TS (Scheme 2).13,15,66,180 The very high stereoselectivity is probably explained by the fact
that in the TS, the carbonyl oxygen of the aldehyde (2) and the boron atom of the
allylboronate comes to a very close proximity.66,180 As the metal atoms (Si, Sn, Zn, etc) in
other widely used allyl metal species are derived from higher periods of the periodical
table, such close proximity between the reactants in the TS of the allylation cannot be
realized.66 Therefore, the stereoselectivity in the allylation of electrophiles is much higher
with allylboronates than with any other allyl metal species. Obviously, the high
stereoselectivity of formation of the homoallylic products 4 is based on the availability of
regio- and stereodefined allylboronic acids (21), which is delivered by the efficient
pincer-complex-catalyzed borylation reaction. Moreover, in this chapter, we have also
shown that acetals can be readily used as aldehyde sources. The in situ hydrolysis of the
acetals is fully compatible with the catalytic borylation process, giving access to other
functionalized electrophiles.
Using ketones as substrates, Kobayashi and Schneider53 suggested that allylboronates are
activated by InI followed by electrophilic coupling with the carbon atom of the keto
functionality. In the presented reactions (Chapter 3.2, Table 3) involving ketones 32a-b
allylboronic acid 21 is supposed to be activated in a similar way, and the TS geometry of
the allylation is probably similar to that of the corresponding reaction with aldehydes.
While the coupling of allylic alcohols with aldehydes is based on three stoichiometric
components and involves two core processes (Figure 9), the use of allylic alcohols as
substrates in the Petasis Borono-Mannich reaction is a four component (20, 14/19, 9 and
2a) reaction in at least three different processes: the borylation (20 → 21), imine
formation (9 + 2a → 35) and allylation (21+ 35 → 10) (Figure 10).
29
HOOC-CH=NR2
R 2-NH2 + HOOC CHO
9
35
2a
Imine f ormation
+ H2O 1
R
21
B(OH) 2
+ B(OH) 3
16g
NHR2
COOH
R1
Allylation
Borylation
10
+ B(OH) 3
R1
O
40
16h
B L + H2O
HO
Figure 10. Allylboronates derived from allylic alcohols in the Petasis reaction.
The borylation process takes place similarly to the above described mechanism.
However, as mentioned in Chapter 3.3, the amine component 9 and glyoxylic acid 2a
cannot be added to the reaction mixture at the outset. Addition of 9 inhibits the borylation
process, probably by coordination to the palladium atom of pincer-complex 16g. This
coordination hinders the transmetallation of diboronate 19 to form 16h, which prevents
formation of 21. On the other hand, addition of 2a alone leads to allylation of the
aldehyde functionality providing the homoallylic alcohol as product. Formation of imine
35 and its addition to organoboronates is still a subject of mechanistic studies.15,170
Similarly to the studies with isolated vinyl and allylboronates, we have also observed that
addition of 2a and 9 separately gives higher yields than addition of imine 35 formed in a
separate process. This suggests precoordination of 9 to the boron atom of 21. Then, this
imine-boronate complex undergoes the allylation process affording product 10.
Coordination of 9 to 21 may explain the fact that in the presented reactions, we could not
observe allylation of aldehyde 2a, whereas the in situ formation of 35 is an equilibrium
process. Considering that the regio- and stereoselectivity in the formation of amino acids
(10) is similar to the allylation of aldehydes, we suppose a similar type of TS structure in
both processes.
In the one-pot reactions we use p-TsOH as an additive, which was an important key for
an efficient conversion of some allylic alcohols (Chapter 2). The acid may also be
beneficial in the allylation reactions. Although allylboronates are self-activating with
aldehydes, recent studies revealed16,17,40,181 that Lewis acids accelerate this reaction via
activation of the allylic component as discussed in Chapter 1. This activation may be
particularly important for allylation of deactivated aldehydes with sterically hindered
allylboronates. We have also shown that the sulfur-based catalyst 16b, in which the
counter ion is loosely coordinated, can be used to obtain faster conversions. A possible
explanation of this acceleration effect can be ascribed to a facile transmetallation of the
diboronic reagent due to the more accessible palladium atom in 16b. Another possibility
is that 16b acts as a Lewis-acid catalyst in the allylation process.
30
In this Chapter it was shown that homoallylic alcohols and amino acids can efficiently be
prepared from allylic alcohols via transient allylboronates in a one-pot sequence. The
high level of compatibility of the borylation procedure with allylation of various
electrophiles forms the basis of the above described reactions. This compatibility relies
on three main factors: (a) the high selectivity of the pincer-complex catalyst, which does
not undergo further reactions with the allylboronic acid products or other components; (b)
use of 14/19 as boronate source, allowing that the only by-product of the borylation
reaction is unreactive (and non-toxic) boric acid; (c) the high reactivity of the in situ
formed allylboronic acids, which is accompanied with a fairly high chemical inertness,
such as their acid, base, water and air stability and tolerance of alcohol as co-solvent. The
reactions have a broad synthetic scope involving cyclic and acyclic alcohols, aromatic
and aliphatic aldehydes, ketones and in situ formed imines. The mild conditions for
activation of the allylic alcohols allow a wide range of functionalities, resulting in
densely functionalized products. Due to the high regio- and stereoselectivity both in the
formation of the allylboronates and in the allylation reaction, the final products are
obtained as single regio- and stereoisomers.
31
4. Synthesis and Tandem Catalytic Performance of Symmetrical and
Unsymmetrical Sulfur-Containing Palladium Pincer-Complexes (Paper VI)
As pointed out in Chapter 1.3, palladium pincer-complexes are important as robust
catalysts in various synthetically important organic reactions, including C–C and C–X
bond formation reactions.109-121 One of the most attractive features is the tunability
(Chapter 2.5) of the ligand as the metal center can be controlled by both the nature of the
E-donor and the various substituents on the pincer-ligand (Figure 11). The catalytic
activity of pincer-complexes has for example been studied in the aldol reaction126,135,182186
and in transformations involving allyl metal compounds (Chapter 2-3).30,34,123,127129,131,134,135,187
In these catalytic transformations, exclusively symmetrical aryl ECEpincer-metal catalysts (Figure 11a) were applied, due to the stability, modularity and
relatively simple availability of these species. It is well documented that the σ-donor/πacceptor properties of the hetero-atoms in the pincer-complexes influence the catalytic
activity.129,188,189 For example, the allylation of aldehydes and imines are readily
catalyzed by PCP-pincer palladium complexes, whereas complexes with σ-donating
hetero-atoms (NCN and SeCSe) displayed lower (if any) catalytic activity.32,128,129,187
Interestingly, formation of allyl metal species proceeds smoothly using NCN, SCS and
SeCSe complexes, however PCP complexes display a very low activity.30,33-35,127,129,131
These observations prompted us to synthesize and explore the catalytic activity of a novel
class of pincer palladium catalysts by combining the σ-donor/π-acceptor effects of the
pincer-ligands. A possible way to accomplish this goal is to use the unsymmetrical ECE’pincer-complexes shown in Figure 11b.
Z
Z
a)
b)
E
Pd
E
E
L
E = NR2, PR2, AsR 2,
OR, SeR or SR
Pd
E'
L
E = NR2, or SR
E' = PR2
Figure 11. General formula for a) symmetrical and b) unsymmetrical pincer-complexes.
The first unsymmetrical PCN-pincer palladium complexes were reported by Dupont and
co-workers115,190 These complexes, generated by chloropalladation of hetero-substituted
alkynes, proved to be highly reactive catalyst precursors in coupling of arylboronic acids
with aryl chlorides. Aryl based, unsymmetrical (PCN) pincer-metal (i.e., rhodium)
complexes have been reported by Milstein and co-workers,191,192 and very recently, an
excellent method for synthesis of PCN-pincer palladium complexes was reported by
Song and co-workers.193
32
In order to study the composite effects of electron-withdrawing and -donating side arms
in palladium-catalyzed aldol reactions and transformations involving allyl metal
compounds, we have designed a new aryl-based pincer-complex catalyst. Considering the
synthetic aspects of obtaining such a complex (see below), we chose phosphorous as πacceptor hetero-atom and sulfur as the σ-donor, obtaining a PCS- pincer-complex.
(Figure 11b, E = SPh, E’ = PPh2) We used the corresponding palladium complex as
catalyst in aldol condensation of methyl isocyanoacetate with benzaldehyde as well as in
tandem catalytic transformation of allyl chloride or vinyloxirane substrates with aldehyde
and sulfonimine reagents. The major focus of this study is on the mechanistic aspects of
having two electronically different donor atoms (phosphorous and sulfur) present in the
pincer-ligand. Moreover, it was of interest to see whether the PCS-pincer palladium
complex would combine the different activities and selectivities of the PCP- and SCSpincer palladium complexes in the respective C-Sn and C-C coupling reactions discussed
and could thus lead to a catalytic tandem reaction. This latter aspect is particularly
interesting since previous attempts to perform tandem catalytic reactions with a single,
symmetrical ECE-pincer-complex failed, and synthetically useful procedures could only
be performed by the simultaneous use of two different (NCN and PCP) complexes.129 In
the reference reactions, we employed symmetrical SCS- and in some cases PCP-catalysts
under the same reaction conditions as the PCS-pincer palladium catalyst. The catalytic
performance of the SCS-complexes have previously been unexplored in synthesis and
transformations of organometallic compounds. In this study, we were able to present
some new interesting applications with this type of catalysts.
4.1 Synthesis of the Unsymmetrical PC(H)S-pincer Pro-ligand
and the PCS-Complexes
The synthesis of symmetrical pincer arene ligands typically starts from α,α’-dibromometa-xylene (25a) by nucleophilic displacement of the benzylic bromines (Chapter 2). By
applying of excess of NR2-, PR2- SR- or other nucleophiles, the corresponding
symmetrical ECE-pincer-ligand can be obtained in usually excellent yields.109-121 As the
two bromines in 25a are symmetrically equivalent, synthesis of unsymmetrical ECE’pincer-ligands requires selective and sequential nucleophilic substitution of the benzylic
leaving groups, which is obviously a more challenging task. Pro-ligand 43 was prepared
in three steps from commercially available α,α’-dibromo-meta-xylene (25a). Reaction of
25a with one equivalent of in situ prepared Li–PPh2(BH3) yielded a mixture containing
monophosphine 41 (63%), diphosphine [C6H4[CH2P(BH3)Ph2]2-1,3] (12%) together with
unreacted 25a (25%). This mixture of compounds was used without further purification.
Full conversion of the benzylic bromides into thioethers was achieved by treatment of the
mixture with an excess of thiophenol in the presence of K2CO3. Separation of the
resulting borane-protected PC(H)S pro-ligand 42 from the symmetrical SC(H)S- and
PC(H)P-pincer arenes was carried out by column chromatography. Subsequent
deprotection of the diphenylphosphino-BH3 moiety in 42 with HBF4·OEt2, followed by a
basic work-up, gave the PC(H)S-pincer arene 43 as an air- and moisture sensitive white
solid (Scheme 19).
33
Br
Br
BH3
PPh2
Br
LiPPh2 . BH3
+
THF
–78 oC to rt
+
PPh2
BH3
PPh2
BH3
41
25a
Br
THF
rt
Br
PhSH, K2CO3
18-Crown-6
BH3
PPh2
SPh
+
SPh
+
PPh2
BH3
PPh2
BH3
SPh
42 yield 51 %
CH2Cl2
rt
1) HBF4 . Et2O
2) aq. NaHCO3
SPh
PPh2
43 yield 74%
Scheme 19. Synthesis of the PC(H)S-pincer arene 43.
The cationic PCS-Pd(II) complex 16c was prepared via electrophilic C-H activation upon
treatment of 43 with [Pd(NCMe)4](BF4)2.194 Complex 16c was readily converted into the
neutral complex 16i by treatment with excess NaCl in an acetonitrile/water mixture. Both
complexes were isolated as air-stable, yellow solids, which can be stored without change
for months in air at ambient temperature (Scheme 20).
SPh
SPh
[Pd(NCMe)4](BF4)2
PPh 2
43
MeCN
RT
Pd NCMe
PPh 2
16c yield 70%
BF4
SPh
NaCl
MeCN/H2O
RT
Pd
PPh2
16i yield 95%
Scheme 20. Synthesis of ionic 16c and neutral 16i palladium pincer-complexes.
34
Cl
4.2 Catalytic Performance of Sulfur- and Phosphorous-based Pincer-Complexes as
Pre-catalyst in an Aldol Condensation
The catalytic activities of the new PCS-pincer-complex 16i, and those of the known SCSand PCP-pincer-complexes 16j195,196 and 16k,197 were compared as Lewis acid precatalysts in the aldol condensation reaction of methyl isocyanoacetate (44) with
benzaldehyde 2b (Scheme 21). The data reported in Scheme 21 show that PCS complex
16i is the fastest catalyst (entry 1), with a trans/cis ratio of the product 45 comparable to
that obtained when the reaction is carried out with SCS-pincer palladium complex 16j
(entry 2). In contrast with these results, the PCP-catalyst 16k is the slowest, also giving a
different cis-trans-ratio of the products 45a/b (Scheme 21).
Ph2P
Pd
Cl
16i
SPh
O
OMe
H + CN
Ph
O
44
2b
PhS
Pd
Cl
16j
SPh
Ph2P
[Pd]cat / i Pr2EtN
Ph
CH2Cl2
RT
H
entry [Pd]cat
1
2
16i
16j
16k
3
H
Pd
Cl
16k
O
PPh2
Ph
H
+
O
N
N
CO2Me MeO2C H
trans
cis
TOF
57%
59%
45b
43%
41%
82%
18%
17
45a
75
46
Scheme 21. Aldol condensation of benzaldehyde and methyl isocyanoacetate. The turn
over frequency (TOF) is given as mol of product per mol of catalyst in the first hour
Previous studies126,189 have shown that 16k is the real catalyst for this reaction, whereas
the SCS-pincer palladium complex 16j is in fact a pre-catalyst.126 This is indicated by the
fact that complex 16j126 as well as the PCS-complex 16i (but not 16k) can be converted
by the methyl isocyanoacetate substrate 44 into the corresponding 1:1 isocyanide
insertion products 46a and 46b (Scheme 22).
E
E
E'
O
OMe
MeO
N
OMe
Pd Cl + 2 CN
O
CH2Cl2
RT
44
16i E = SPh E' = PPh2
16j E = SPh E' = SPh
CN
Pd
E'
Cl
46a E = SPh E' = PPh 2
46b E = SPh E' = SPh
Scheme 22. Insertion of 44 into the Pd-C bond of 16i and 16j126.
35
O
Insertion of an isocyanide into the M–C bond of a cyclometallated compound has been
extensively studied126,188,189 and has also been documented for ECE-pincer palladium-d8
complexes (NCN188,189 and SCS126,195,196). The fact that PCP-pincer-complex 16k appears
to be stable towards insertion189 has been ascribed to the stronger P-Pd coordination of
the two ortho-(diphenylphosphino)methyl substituents relative to the Pd–N and Pd–S
bonding in the corresponding NCN- and SCS-pincer palladium(II) complexes. Recently,
it was shown126,188,189 that a likely reaction mechanism for this insertion reaction involves
coordination of the isocyanide as a ligand cis to the aryl C-atom in the symmetrical ECEpincer palladium complex (E = NMe2 or SPh), which may only occur through prior
decoordination of one or both of the ortho-ligands. These findings prompted us to also
study the reactivity of PCS complex 16i towards alkyl isocyanide 44 in CH2Cl2. It was
found that PCS complex 16i readily reacted with isocyanide 44 to afford quantitatively
complex 46a (Scheme 22), incorporating two isocyanoacetate molecules. One of the
isocyanoacetates has inserted into the Pd–C bond, whereas the second one is η1-Ccoordinated to the palladium centre. Complex 46a was characterized by elemental
analysis and by IR- and NMR spectroscopy. The rapid insertion of 44 into the Pd–C bond
of complex 16i contrasts with the stability of 16k towards this reagent and has been
ascribed to the more labile coordination of the Pd-S bond in 16j compared to that of the
Pd-P one. As a consequence, in 46a it is the S-donor ligand that is the non-coordinated
one.
With regard to the mechanistic details of the aldol reaction catalyzed by 16i it seems
likely that prior Pd-S bond cleavage occurs with η1-coordination of the incoming methyl
isocyanoacetate (44). This is followed by insertion of the isocyanide into the Pd–C bond,
which is followed by η1-coordination of the second molecule of 44 to give 46a (cf.
reaction in Scheme 22). Subsequently, it is the η1-coordinated isocyanoacetate (44)
molecule that then undergoes the aldol reaction. Consequently, the resulting
isocyanoacetate (44) insertion compounds (46a-b) are the active catalysts in the aldol
reaction in cases of PCS (16i) and SCS (16j) which in fact are true pre-catalysts. In
contrast, the PCP-pincer-complex 16k follows a different reaction coordinate, involving
halide displacement and η1-C coordination of a methyl isocyanoacetate (44).189
The catalytic results obtained with complexes 16i, 16j, and 16k in this aldol condensation
show how changes in the nature of the pincer-ligand donor arms can affect the reactivity
of the corresponding metal complexes. In this specific case, changing the arms from
symmetrical sulfur hetero-atoms (16j) to symmetrical phosphorous hetero-atoms (16k)
alters the direct interaction/reaction of these complexes with 44, and as a result alters the
mechanism of the catalyzed reaction, resulting in different activities (TOFs and product
selectivities).
4.3 SCS- and PCS-Pincer-Complexes as Catalysts for Tandem Catalysis
As shown in Chapter 2 and in several other studies,30,33,34,129 NCN- and SeCSe-pincer
palladium complexes are excellent catalysts for preparation of allyl metal species, such as
allylstannanes (Scheme 23, 47 → 49a) and allylboronates. However, PCP-pincer
palladium complexes are less efficient catalysts in these processes. On the contrary,
allylation of aldehydes and imines (Scheme 23, 49a → 4/5) can be performed more
efficiently with PCP-catalysts than with the corresponding NCN- or SeCSe-catalysts.129
36
Accordingly, we anticipated that with PCS-pincer palladium catalyst 16c it should be
possible to perform both catalytic processes in tandem. Indeed, a tandem catalytic
coupling of allyl chlorides (47a and 47b) and vinyloxirane (29c) with aldehyde 2e or
sulfonimine 3a reagents could be achieved with hexamethylditin (48) in the presence of
catalytic amounts (5 mol%) of 16c or 16l (Scheme 23, Table 6).
BF4
BF4
[Pd]cat =
R
Ph2P
Pd
SPh
NCMe
16c
Cl + Me3Sn-SnMe3
47
48
[Pd]cat
PhS
Pd SPh
NCMe
16l
R
SnMe3
49a
[Pd]cat
Ar-CH=Z
2e Z = O
3 Z = NSO2Ph
ZH
R
NO2
4 or 5
Scheme 23. Tandem catalytic approach for stannylation of allyl chlorides followed by
allylation of aldehydes and imines.
These reactions resulted in formation of homoallylic alcohols (4v-x) and homoallylic
amines (5a-b) via intermediate formation of allylstannanes (49). These processes could
be performed under mild conditions (40°C) as genuine one-pot reactions in a single
operational step. Accordingly, all reagents were mixed together at the outset of the
reaction, and after the allotted reaction time (16h) the products (4 or 5) were isolated. The
reactions with allyl chloride (47a) proceeded cleanly and with relatively high isolated
yields of 63-78% (Table 6, entries 1 and 3). When a substituted allyl chloride (such as
cinnamyl chloride, 47b) was used, the yields dropped slightly (59-60%), but they still
remained respectable, considering that the product is formed in a tandem catalytic
process (Table 6, entries 5 and 7).
The diastereoselectivity of the reaction is poor in the case of allylation of 4nitrobenzaldehyde (2e, entry 5), but it is rather promising for allylation of sulfonimine 3a
(entry 7). The major diastereomer is the anti form in the case of allylation of 2e and the
syn form in that of allylation of 3a. Previously, we have studied128,134 the roles of steric
and electronic interactions in determining the stereoselectivity in pincer-complexcatalyzed allylation reactions. These studies have clearly shown that the allylation of
aldehydes (such as 2e) proceeds with anti selectivity, while the opposite (syn selectivity)
occurs for the corresponding process with imine substrates (such as 3a). These previous
observations128,134 are in perfect agreement with the stereoselectivity observed in the
presented processes catalyzed by 16c and 16l (Table 6).
37
Table 6. Tandem catalytic transformations of allylic substrates.a
Entry
Substrate
Electrophile
Cat.
Cond.b
16c
40/16
Product
OH
O
Cl
1
2
47a
47a
3
47a
4
5
2e
2e
NSO2Ph
3a
3a
47a
Ph
Cl
2e
NO2
16l
40/16
16c
40/16
4v
4v
NHSO2Ph
NO2
16l
40/16
16c
40/16
5a
5a
OH
Ph
47b
6
4w
2e
47b
16l
40/16
Yieldc
drd
63
-
80
-
78
-
73
-
59
3:1
74
9:1
60
1:5
48
1:4
46
5:3
57
3:1
NO2
NO2
NO2
4w
NHSO2Ph
7
47b
3a
16c
40/16
Ph
8
47b
O
9[e]
10 [e]
a
d
29c
29c
3a
16l
40/16
2e
16c
20/16
5b
5b
OH
HO
2e
16l
20/16
4x
4x
NO2
NO2
In a typical reaction, 5 mol% catalyst was used in THF. b Temperature/time [°C]/[h].c Isolated yield
Diastereomeric ratio: anti/syn. e LiOAc (10 mol%) and water (2 equiv) were also added to the reaction.
As previously reported,110,129 symmetrical PCP-pincer palladium complexes do not
catalyze the formation of the intermediate allylstannanes (Scheme 23, 47 → 49), and so
these complexes alone cannot be used as catalysts for coupling of allyl chlorides and
aldehydes in the presence of hexamethylstannane 48. Therefore, reactions with PCP
complex 16k as reference catalyst were not performed in this study. On the other hand we
also studied the catalytic performance of SCS-pincer-complex 16l. To our delight, this
complex showed a high catalytic activity, comparable to that of PCS-pincer-complex 16c
(entries 2, 4, 6 and 8). The high activity of SCS complex 16l in the tandem catalytic
reactions is surprising, because symmetrical ECE-pincer-complexes with σ-donor heteroatoms (e.g. NCN) usually perform poorly128,134,187 in the allylation step (Scheme 23,
49 → 4/5). The isolated yields obtained with the SCS-pincer-complex 16l are particularly
high for allylation of aldehyde 2e, while the stereoselectivity using 47b as allyl substrate
is even higher than with the PCS-pincer-complex 16c (entry 6). However, in the
allylation reaction of sulfonimine 3a the PCS-pincer-complex 16c clearly outperforms
38
16l. The homoallylic amines 5a and 5b (entries 3-4 and 7-8) are formed in higher yields
with PCS-catalyst 16c (78 and 60%) than with the SCS-catalyst 16l (73 and 48%). In
particular, when the tandem reaction of 47b and 3a was catalyzed with SCS-catalyst 16l
(entry 8), a substantial amount of cinnamyl stannane (49a R = Ph) was present in the
crude reaction mixture. This finding indicates that the allylation process (Scheme 23,
49 → 5b) of the imine by cinnamyl stannane proceeds very slowly in the presence of 16l
as catalyst.
We have also succeeded in performing a tandem catalytic coupling reaction of
vinyloxirane (29c) with 2e (entries 9 and 10). This reaction proceeds under reaction
conditions similar to those used for the allylation with allyl chlorides, except that
catalytic amounts of LiOAc and water were added to the reaction mixture. Without this
addition of LiOAc and water, vinyloxirane (29c) is decomposed by the catalysts,
probably through opening of the epoxy functionality. The coupling reaction of aldehyde
2e with vinyloxirane (29c) leads to formation of an unprotected homoallylic diol (4x),
which is a useful synthetic intermediate. As far as we know, this reaction represents the
first example of a palladium-catalyzed coupling of a vinyloxirane with an aldehyde to
afford homoallylic alcohol 4x.
4.4 Mechanistic Considerations of the Tandem Coupling Reaction
On the basis of the above results and our previous mechanistic and modeling
studies128,131,134 it is possible to propose a plausible catalytic cycle for the tandem C-Sn
and C-C processes (Scheme 24). The first catalytic process is formation of the
allylstannane intermediate 49 from allyl chloride 47 (or vinyloxirane 29c) and
hexamethylstannane 48. This catalytic reaction is initiated by transmetallation of catalyst
16m with 47 to afford complex 16n and trimetyltin chloride. Formation of a trimethyltinpalladium pincer-complex under similar reaction conditions has been observed for NCNpincer palladium complexes.131 Complex 16n undergoes a nucleophilic substitution
reaction with allyl chloride, furnishing allylstannane 49 and recovering PCS-pincer
palladium complex 16m. The regenerated catalyst 16m is able to undergo
transmetallation with allylstannane 49 to give η1-allylpalladium complex 16o. Similar
transmetallation reactions were previously reported for PCP-pincer-complexes and
allylstannanes.128 In those studies it was concluded128,187 that complexes with heteroatoms exhibiting π-accepting properties in the ortho-substituents (such as symmetrical
PCP-pincer-complexes) undergo more facile transmetallation than the corresponding
analogues with strongly σ-electron donating ortho-ligands (such as NCN). Considering
this, the presence of at least one phosphorous atom in 16c probably allows an efficient
transmetallation. On the other hand, the relatively low catalytic activity of 16l in the
allylation step (entry 8) may be due to a slow transmetallation of the stannane
intermediate 49 to the palladium atom, which is attached to an electron-rich SCS-pincerligand. According to our mechanistic and modeling studies128,134 these η1-allylpalladium
complexes readily react with aldehyde or imine electrophiles (2e or 3a). Thus
electrophilic substitution by complex 16o gives 50, which subsequently undergoes
decomplexation to provide the homoallylic alcohol or amine products (4 or 5) and to
regenerate catalyst 16m once again. In this way complex 16m performs as an autotandem catalyst4 controlling two bond formation processes: the stannylation (Sn-C bond
39
formation) of the allyl chloride and the electrophilic substitution (C-C bond formation) of
the allylstannane intermediate. Our studies show that the stannylation process (47 → 49)
using PCS-catalyst 16c is much faster than the catalytic allylation (49 → 4/5). It was
found that in the presence of 16c under the applied reaction conditions (entry 1), but in
the absence of an electrophile (e.g. 2e), allyl chloride (47a) was converted into
allylstannane (49a) in 30 min, while the full reaction time required for coupling of 48a
with 2e is 16 h. This indicates that the catalytic generation of the allyl stannane (49) is
completed before significant progress in the allylation has taken place (Scheme 24).
Ar-CH=Z
2e or 3a
Me3SnCl
Ph2P
Pd
SPh
Cl
R
Me3SnCl
47
SnMe3
R
49
16o
R
Carbon-carbon
coupling
Ph2P
Pd
Stannylation
SPh
Ph2P
Pd
SPh
SnMe3
16n
L
16m
ZH
Ph 2P
Me3Sn
Pd
Z
Ar
SPh
Cl
Ar
H2O
R
4 or 5
(SnMe3)2
+ Me3SnOH
Me3SnCl
48
R 50
Scheme 24. Proposed catalytic cycle of the auto-tandem reaction.
In summary, aryl-based, unsymmetrical PCS-pincer palladium complexes 16c and 16i
were prepared in relatively simple four and five step synthesis sequences, respectively
(Scheme 20). These PCS-pincer-complexes, as well as SCS-pincer-complex 16j, proved
to be efficient catalysts in the aldol reaction between benzaldehyde and methyl
isocyanoacetate, but are obviously pre-catalysts because the methyl isocyanoacetate
reactant converts the pincer-catalyst into a ketimine-palladium catalyst. In the aldol
condensation reaction the PCS complex gave higher turnover numbers than its PCP and
SCS counterparts. The combination of a sulfur and a phosphorus hetero-atom in the side
arms, such as in complex 16c, results in a synergetic effect. This effect involves two
major components: a weak coordination of the sulfur atom allowing insertion of 44, as
well as a strong coordination of the phosphorus atom, which increases the rate of the
catalytic process.
40
In the auto-tandem catalytic coupling of allyl chlorides or vinyloxirane with aldehyde or
sulfonimine substrates, both 16c and 16l are the true catalytic species. Comparative
studies using symmetrical PCP- and SCS-pincer palladium complexes show that a strong
synergistic effect is also operative in these catalytic reactions. In the tandem catalytic
stannylation and allylation process of allyl chlorides (Table 6) the PCS complex performs
much better than the PCP analogue.129 Although, the SCS complex has a surprisingly
high catalytic activity, the PCS complex is more efficient in allylation of sulfonimines. In
these processes, the PCS-catalyst unifies the attractive catalytic features of the
symmetrical analogues: firstly, a high catalytic activity in the stannylation reaction, in
which symmetrical complexes with electron-donating hetero-atoms and a weaker
coordination interaction to the palladium centre (e.g. NCN or SCS) performs well, and
secondly, the more efficiently catalyzed allylation of the electrophile by less electron rich
pincer-complexes in which the hetero-atom has a stronger interaction with palladium (i.e.
as in PCP).
41
5. General Conclusions and Outlook
In this thesis, it was demonstrated that palladium pincer-complexes can be used as highly
selective catalysts for synthesis and transformations of organometallic reagents. A novel
and versatile method for the preparation of functionalized allylboronates starting from
allylic alcohols was presented. Because of the mild conditions and reagents used, this
procedure could successfully be combined with several one-pot domino transformations.
The basis of these protocols is the clean transformations performed by the highly
selective palladium pincer-catalysts. We have shown that the in situ generated
allylboronates can be isolated as potassium trifluoro(allyl)borates (A), reacted with
ketone (B) or aldehyde (C) electrophiles, or in the Petasis Borono-Mannich reaction (D).
The one-pot approach of these transformations is an important key to avoid isolation and
purification of the reactive allyl metal intermediates. The products of these reactions are
stereodefined densely functionalized homoallylic alcohols or amino acid derivatives
(Scheme 25).
R
[Pd]cat.
O
NHR'
O
O
OH
R'
OH
B
OH
R
D
R
B2(OR)4
OH
R'NH2 + H
COOH
OH
R'
C
B
KHF2
O
R'
BF3K
OH
R''
[InI]cat
R
R
A
R
R'
R''
Scheme 25. Overview of the presented reactions.
Moreover, a new procedure for the synthesis of novel unsymmetrical pincer-complexes
was presented. These complexes were able to perform an auto-tandem catalytic
transformation of allyl chlorides via allylstannanes which were used as allylating reagents
in the same reaction vessel.
We believe that the clean and selective transformations performed by the studied
palladium pincer-complexes can be extended further to involve other type of domino and
tandem reactions. The possibility to induce chirality in the above presented reactions is a
challenge in future projects. In order to develop such processes, an important factor is to
gain a higher insight in the mechanism of the reactions.
42
6. Acknowledgements
I would like to thank:
My research advisor Prof. Kálmán J. Szabó for inspiration and excellent guidance
Prof. Jan-Erling Bäckvall for showing interest in this thesis
My co-workers Dr. Sara Sebelius, Vilhelm J. Olsson, Juhanes Aydin, Dr. Marcella
Gagliardo, Andreas Kipke, Cesar Estay and Helena Larsson
My Dutch co-workers Dr. Marcella Gagliardo, Dr. Nilesh C. Mehendale, Prof. Gerard
van Koten and Prof. Robertus J. M. Klein Gebbink, Utrecht University, The Netherlands
All the past and present members of the Szabó group
Vilhelm J. Olsson and Dr. Niclas Solin for proof-reading
The TA staff and Kemiska biblioteket
All colleagues at the Department of Organic Chemistry
For the generous financial support:
“The IDECAT Network of Excellence of the European Community”
“K & A Wallenbergs Stiftelse”
“John Söderbergs Stiftelse”,
“Astra Zeneca”
“Svenska Kemistsamfundet”
“Ångpanneföreningens Forskningsstiftelse”
My family and friends for support
43
7. References
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
(15)
(16)
(17)
(18)
(19)
(20)
(21)
(22)
(23)
(24)
(25)
(26)
(27)
(28)
(29)
(30)
(31)
(32)
(33)
(34)
(35)
(36)
(37)
Zhu, J.; Bienaymé, H. Multicomponent Reactions; Wiley-VCH: Weinheim, 2005.
Tietze, L. F. Chem. Rev. 1996, 96, 115.
Poli, G.; Giambastini, G. J. Org. Chem. 2002, 67, 9456.
Fogg, D. E.; Santos, E. N. d. Coord. Chem. Rev. 2004, 248, 2365.
Wasilke, J.-C.; Obrey, S. J.; Baker, R. T.; Bazan, G. C. Chem. Rev. 2005, 105, 1001.
Lee, J. M.; Na, Y.; Han, H.; Chang, S. Chem. Soc. Rev. 2004, 33, 302.
D´Souza, D. M.; Müller, T. J. J. Chem. Soc. Rev. 2007, 36, 1095.
Roush, W. R. In Comprehensive Organic Synthesis; Trost, B. M., Flemming, I., Eds.;
Pergamon Press: New York, 1991; Vol. 2, p 1.
Yamamoto, Y.; Asao, N. Chem. Rev. 1993, 93, 2207.
Marshall, J. A. Chem. Rev. 2000, 100, 3163.
Bloch, R. Chem. Rev. 1998, 98, 1407.
Zanoni, G.; Pontiroli, A.; Marchetti, A.; Vidari, G. Eur. J. Org. Chem. 2007, 3599.
Denmark, S. E.; Fu, J. Chem. Rev. 2003, 103, 2763.
Otera, J. Modern Carbonyl Chemistry; Wiley: Weinheim, 2000.
Hall, D. G. Boronic Acids; Wiley: Weinheim, 2005.
Kennedy, J. W. J.; Hall, D. G. J. Am. Chem. Soc. 2002, 124, 11586.
Lachance, H.; Lu, X.; Gravel, M.; Hall, D. G. J. Am. Chem. Soc. 2003, 125, 10160.
Kennedy, J. W. J.; Hall, D. G. Angew. Chem. Int. Ed. 2003, 42, 4732.
Kennedy, J. W. J.; Hall, D. G. J. Org. Chem. 2004, 69, 4412.
Gravel, M.; Lachance, H.; Lu, X.; Hall, D. G. Synthesis 2004, 1290.
Yu, S. H.; Ferguson, M. J.; McDonald, R.; Hall, D. G. J. Am. Chem. Soc. 2005, 127,
12808.
Gao, X.; Hall, D. G.; Deligny, M.; Favre, A.; Carreaux, F.; Carboni, B. Chem. Eur. J.
2006, 12, 3132.
Rauniyar, V.; Hall, D. G. Angew. Chem. Int. Ed. 2006, 45, 2426.
Elford, T. G.; Arimura, Y.; Yu, S. H.; Hall, D. G. J. Org. Chem. 2006, 72, 1276.
Peng, F.; Hall, D. G. J. Am. Chem. Soc. 2007, 129, 3070.
Peng, F.; Hall, D. G. Tetrahedron Lett. 2007, 48, 3305.
Rauniyar, V.; Hall, D. G. Synthesis 2007, 3421.
Hall, D. G. Synlett 2007, 1644.
Sebelius, S.; Wallner, O. A.; Szabó, K. J. Org. Lett. 2003, 5, 3065.
Sebelius, S.; Olsson, V. J.; Szabó, K. J. J. Am. Chem. Soc. 2005, 127, 10478.
Sebelius, S.; Szabó, K. J. Eur. J. Org. Chem. 2005, 2539.
Solin, N.; Wallner, O. A.; Szabó, K. J. Org. Lett. 2005, 7, 689.
Olsson, V. J.; Sebelius, S.; Selander, N.; Szabó, K. J. J. Am. Chem. Soc. 2006, 128,
4588.
Sebelius, S.; Olsson, V. J.; Wallner, O. A.; Szabó, K. J. J. Am. Chem. Soc. 2006, 128,
8150.
Selander, N.; Sebelius, S.; Estay, C.; Szabó, K. J. Eur. J. Org. Chem. 2006, 4085.
Olsson, V. J.; Szabó, K. J. Angew. Chem. Int. Ed. 2007, 46, 6891.
Selander, N.; Kipke, A.; Sebelius, S.; Szabó, K. J. J. Am. Chem. Soc. 2007, 129,
13723.
44
(38)
(39)
(40)
(41)
(42)
(43)
(44)
(45)
(46)
(47)
(48)
(49)
(50)
(51)
(52)
(53)
(54)
(55)
(56)
(57)
(58)
(59)
(60)
(61)
(62)
(63)
(64)
(65)
(66)
(67)
(68)
(69)
(70)
(71)
(72)
(73)
(74)
(75)
(76)
Ahiko, T.; Ishiyama, T.; Miyaura, N. Chemistry Letters 1997, 811.
Ishiyama, T.; Miyaura, N. J. Organomet. Chem. 2000, 611, 392.
Ishiyama, T.; Ahiko, T.-a.; Miyaura, N. J. Am. Chem. Soc. 2002, 124, 12414.
Kabalka, G. W.; Narayana, C.; Reddy, N. K. Tetrahedron Lett. 1996, 37, 2181.
Pace, R. D.; Kabalka, G. W. J. Org. Chem. 1995, 60, 4838.
Batey, R. A.; Thadani, A. N.; Smil, D. V. Tetrahedron Lett. 1999, 40, 4289.
Batey, R. A.; Thadani, A. N.; Smil, D. V.; Lough, A. J. Synthesis 2000, 990.
Li, S.-W.; Batey, R. A. Chem. Commun, 2004, 1382.
Dhudshia, B.; Tiburcio, J.; Thadani, A. N. Chem. Commun, 2005, 5551.
Roush, W. R.; Walts, A. E.; Hoong, L. K. J. Am. Chem. Soc. 1985, 107, 8186.
Roush, W. R.; Ando, K.; Powers, D. B.; Palkowitz, A. D.; Halterman, R. L. J. Am.
Chem. Soc. 1990, 112, 6339.
Roush, W. R.; Hoong, L. K.; Palmer, M. A. J.; Park, J. C. J. Org. Chem. 1990, 55,
4109.
Heo, J.-N.; Micalizio, G. C.; Roush, W. R. Org. Lett. 2003, 5, 1693.
Sugiura, M.; Hirano, K.; Kobayashi, S. J. Am. Chem. Soc. 2004, 126, 7182.
Sugiura, M.; Mori, C.; Hirano, K.; Kobayashi, S. Can. J. Chem. 2005, 83, 937.
Schneider, U.; Kobayashi, S. Angew. Chem. Int. Ed. 2007, 46, 5909.
Fujita, M.; Nagano, T.; Schneider, U.; Hamada, T.; Ogawa, C.; Kobayashi, S. J. Am.
Chem. Soc 2008, 130, 2914.
Kobayashi, S.; Konishi, H.; Schneider, U. Chem. Commun. 2008, 2313.
Schneider, U.; Chen, I.-H.; Kobayashi, S. Org. Lett. 2008, 10, 737.
Wada, R.; Oisaki, K.; Kanai, M.; Shibasaki, M. J. Am. Chem. Soc. 2004, 126, 8910.
Wada, R.; Shibuguchi, T.; Makino, S.; Oisaki, K.; Kanai, M.; Shibasaki, M. J. Am.
Chem. Soc. 2006, 128, 7687.
Lou, S.; Moquist, P. N.; Schaus, S. E. J. Am. Chem. Soc. 2006, 128, 12660.
Lou, S.; Moquist, P. N.; Schaus, S. E. J. Am. Chem. Soc. 2007, 129, 15398.
Roush, W. R.; Hoong, L. K.; Palmer, M. A. J.; Straub, J. A.; Palkowitz, A. D. J. Org.
Chem. 1990, 55, 4117.
Wu, T. R.; Shen, L.; Chong, J. M. Org. Lett. 2004, 6, 2701.
Brown, H. C.; Racherla, U. S.; Pellechia, P. J. J. Org. Chem. 1990, 55, 1868.
Darses, S.; Genet, J.-P. Chem. Rev. 2008, 108, 288.
Vedejs, E.; Chapman, R. W.; Fields, S. C.; Lin, S.; Schrimpf, M. R. J. Org. Chem.
1995, 60, 3020.
Hoffmann, R. W. Angew. Chem. Int. Ed. Engl. 1982, 21, 555.
Denmark, S. E.; Weber, E. J. Helv. Chim. Acta 1983, 66, 1655.
Chataigner, I.; Zammattio, F.; Lebreton, J.; Villiéras, J. Synlett 1998, 275.
Petasis, N. A.; Akritopoulou, I. Tetrahedron Lett. 1993, 34, 583.
Petasis, N. A.; Zavialov, I. A. J. Am. Chem. Soc. 1997, 119, 445.
Petasis, N. A.; Goodman, A.; Zavialov, I. A. Tetrahedron 1997, 53, 16463.
Petasis, N. A.; Zavialov, I. A. J. Am. Chem. Soc. 1998, 120, 11798.
Petasis, N. A. Aust. J. Chem. 2007, 60, 795.
Jourdan, H.; Gouhier, G.; Hijfte, L. V.; Angibaud, P.; Piettre, S. R. Tetrahedron Lett.
2005, 46, 8027.
Kumagai, N.; Muncipinto, G.; Schreiber, S. L. Angew. Chem. Int Ed. 2006, 45, 3635.
Petasis, N. A.; Patel, Z. D. Tetrahedron Lett. 2000, 41, 9607.
45
(77)
(78)
(79)
(80)
(81)
(82)
(83)
(84)
(85)
(86)
(87)
(88)
(89)
(90)
(91)
(92)
(93)
(94)
(95)
(96)
(97)
(98)
(99)
(100)
(101)
(102)
(103)
(104)
(105)
(106)
(107)
(108)
(109)
(110)
Lou, S.; Schaus, S. E. J. Am. Chem. Soc. 2008, 130, 6922.
Blais, J.; L´Honoré, A.; Soulié, J.; Cadiot, P. J. Organomet. Chem. 1974, 78, 323.
Tsai, D. J. S.; Matteson, D. S. Tetrahedron Lett. 1981, 22, 2751.
Hoffmann, R. W.; Kemper, B. Tetrahedron 1984, 40, 2219.
Hoffmann, R. W.; Kemper, B.; Metternich, R.; Lehmeier, T. Liebigs Ann. Chem.
1985, 2246.
Hoffmann, R. W.; Kemper, B. Tetrahedron Lett. 1981, 22, 5263.
Stürmer, R. Angew. Chem. Int. Ed. Engl. 1990, 29, 59.
Hoffmann, R. W.; Zeiss, H.-J. J. Org. Chem. 1981, 46, 1309.
Hoffmann, R. W.; Kemper, B. Tetrahedron Lett. 1982, 23, 845.
Hoffmann, R. W.; Zeiss, H.-J. Angew. Chem. Int. Ed. Engl. 1979, 18, 306.
Roush, W. R.; Grover, P. T.; Lin, X. Tetrahedron Lett. 1990, 31, 7563.
Roush, W. R.; Grover, P. T. Tetrahedron Lett. 1990, 31, 7567.
Roush, W. R.; Grover, P. T. Tetrahedron 1992, 48, 1981.
Satoh, M.; Nomoto, Y.; Miyaura, N.; Suzuki, A. Tetrahedron Lett. 1989, 30, 3789.
Ishiyama, T.; Yamamoto, M.; Miyaura, N. Chem. Commun. 1996, 2073.
Suginome, M.; Nakamura, H.; Matsuda, T.; Ito, Y. J. Am. Chem. Soc. 1998, 120,
4248.
Suginome, M.; Matsuda, T.; Yoshimoto, T.; Ito, Y. Org. Lett. 1999, 1, 1567.
Ishiyama, T.; Kitano, T.; Miyaura, N. Tetrahedron Lett. 1998, 39, 2357.
Hoffmann, R. W.; Sander, T.; Hense, A. Liebigs Ann. Chem. 1993, 771.
Ishiyama, T.; Ahiko, T.-A.; Miyaura, N. Tetrahedron Lett. 1996, 37, 6889.
Murata, M.; Watanabe, S.; Masuda, Y. Tetrahedron Lett. 2000, 41, 5877.
Ramachandran, P. V.; Pratihar, D.; Biswas, D.; Srivastava, A.; Reddy, M. V. R. Org.
Lett. 2004, 6, 481.
Kabalka, G. W.; Venkataiah, B.; Dong, G. J. Org. Chem. 2004, 69, 5807.
Ito, H.; Ito, S.; Sasaki, Y.; Matsuura, K.; Sawamura, M. J. Am. Chem. Soc. 2007, 129,
14856.
Ito, H.; Kawakami, C.; Sawamura, M. J. Am. Chem. Soc. 2005, 127, 16034.
Clegg, W.; Johann, T. R. F.; B. Marder, T.; Norman, N. C.; Orpen, A. G.; Paekman,
T. M.; Quayle, M. J.; Rice, C. R.; Scott, A. J. J. Chem. Soc., Dalton Trans. 1998,
1431.
Carosi, L.; Hall, D. G. Angew. Chem. Int. Ed. 2007, 46, 5913.
Hegedus, L. S. Transition Metals in the Synthesis of Complex Organic Molecules;
University Science Books: Mill Valley, California, 1999.
Crabtree, R. H. The Organometallic Chemistry of the Transition Metals; Wiley:
Hoboken, 2005.
Tsuji, J. Transition Metal Reagents and Catalysts. Innovation in Organic Synthesis;
Wiley: Chichester, 2000.
Tsuji, J. Palladium Reagents and Catalysts. New Perspectives for the 21st Century;
Wiley: Chichester, 2004.
Negishi, E.; Meijre, A. d. Organopalladium Chemistry for Organic Synthesis; Wiley:
New York, 2002.
Morales-Morales, D.; Jensen, C. M. The Chemistry of Pincer Compounds; Elsevier:
Amsterdam, 2007.
Szabó, K. J. Synlett 2006, 811.
46
(111) Slagt, M. Q.; Rodríguez, G.; Grutters, M. M. P.; Gebbink, R. J. M. K.; Klopper, W.;
Jenneskens, L. W.; Lutz, M.; Spek, A. L.; Koten, G. v. Chem. Eur. J. 2004, 10, 1331.
(112) Albrecht, M.; Koten, G. v. Angew. Chem. Int. Ed. 2001, 3750.
(113) Rodríguez, G.; Albrecht, M.; Schoenmaker, J.; Ford, A.; Lutz, M.; Spek, A. L.;
Koten, G. v. J. Am. Chem. Soc. 2002, 124, 5127.
(114) Dupont, J.; Consorti, C. S.; Spencer, J. Chem. Rev. 2005, 105, 2527.
(115) Dupont, J.; Pfeffer, M.; Spencer, J. Eur. J. Inorg. Chem. 2001, 1917.
(116) Beletskaya, I. P.; Cheprakov, A. V. J. Organomet. Chem. 2004, 689, 4055.
(117) Boom, M. E. v. d.; Milstein, D. Chem. Rev. 2003, 103, 1759.
(118) Singleton, J. T. Tetrahedron 2003, 59, 1837.
(119) Yao, Q.; Kinney, E. P.; Zheng, C. Org. Lett. 2004, 6, 2997.
(120) Bergbreiter, D. E.; Osburn, P. L.; Liu, Y.-S. J. Am. Chem. Soc. 1999, 121, 9531.
(121) Dupont, J.; Beydoun, N.; Pfeffer, M. J. Chem. Soc. Dalton Trans. 1989, 1715.
(122) Yao, Q.; Sheets, M. J. Org. Chem. 2006, 71, 5384.
(123) Wallner, O. A.; Olsson, V. J.; Eriksson, L.; Szabó, K. J. Inorg. Chim. Acta 2006, 359,
1767.
(124) Aydin, J.; Kumar, K. S.; Sayah, M. J.; Wallner, O. A.; Szabó, K. J. J. Org. Chem.
2007, 72, 4689.
(125) Mehendale, N. C.; Bezemer, C.; Walree, C. A. v.; Gebbink, R. J. M. K.; Koten, G. v.
J. Mol. Cat. 2006, 257, 167.
(126) Mehendale, N. C.; Sietsma, J. R. A.; Jong, K. P. d.; Walree, C. A. v.; Klein-Gebbink,
R. J. M.; Koten, G. v. Adv. Synth. Catal. 2007, 349, 2619.
(127) Kjellgren, J.; Sundén, H.; Szabó, K. J. J. Am. Chem. Soc. 2004, 126, 474.
(128) Solin, N.; Kjellgren, J.; Szabó, K. J. J. Am. Chem. Soc. 2004, 126, 7026.
(129) Wallner, O. A.; Szabó, K. J. Org. Lett. 2004, 6, 1829.
(130) Kjellgren, J.; Aydin, J.; Wallner, O. A.; Saltanova, I. V.; Szabó, K. J. Chem. Eur. J.
2005, 11, 5260.
(131) Kjellgren, J.; Sundén, H.; Szabó, K. J. J. Am. Chem. Soc. 2005, 127, 1787.
(132) Wallner, O. A.; Szabó, K. J. J. Org. Chem. 2005, 70, 9215.
(133) Aydin, J.; Selander, N.; Szabó, K. J. Tetrahedron Lett. 2006, 47, 8999.
(134) Wallner, O. A.; Szabó, K. J. Chem. Eur. J. 2006, 12, 6976.
(135) Aydin, J.; Kumar, K. S.; Eriksson, L.; Szabó, K. J. Adv. Synth. Catal. 2007, 349,
2585.
(136) Tamaru, Y. Eur. J. Org. Chem. 2005, 2647.
(137) Kimura, M.; Futamata, M.; Mukai, R.; Tamaru, Y. J. Am. Chem. Soc. 2005, 127,
4592.
(138) Kimura, M.; Horino, Y.; Mukai, R.; Tanaka, S.; Tamaru, Y. J. Am. Chem. Soc. 2001,
123, 10401.
(139) Takahara, J. P.; Masuyama, Y.; Kurusu, Y. J. Am. Chem. Soc. 1992, 114, 2577.
(140) Ozawa, F.; Okamoto, H.; Kawagishi, S.; Yamamoto, S.; Minami, T.; Yoshifuji, M. J.
Am. Chem. Soc. 2002, 124, 10968.
(141) Ozawa, F.; Ishiyama, T.; Yamamoto, S.; Kawagishi, S.; Murakami, H.
Organometallics 2004, 23, 1698.
(142) Starý, I.; Stará, I. G.; Kocovský, P. Tetrahedron 1994, 50, 529.
(143) Bergbreiter, D. E.; Weatherford, D. A. J. Chem. Soc., Chem. Commun. 1989, 883.
(144) Batey, R. A.; Quach, T. D. Tetrahedron Lett. 2001, 42, 9099.
47
(145) Comprehensive Organic Synthesis; Godleski, S. A., Ed.; Pergamon Press: New York,
1991; Vol. 4.
(146) Szabó, K. J. Chem. Soc. Rev. 2001, 30, 136.
(147) Bäckvall, J.-E.; Nyström, J.-E.; Nordberg, R. E. J. Am. Chem. Soc. 1985, 107, 3676.
(148) Goldberg, S. D.; Grubbs, R. H. Angew. Chem. Int. Ed. 2002, 41, 807.
(149) Muzart, J. Tetrahedron 2005, 61, 4179.
(150) Masuyama, Y.; Takahara, J. P.; Kurusu, Y. J. Am. Chem. Soc. 1998, 110, 4473.
(151) Carde, L.; Llebaria, A.; Delgado, A. Tetrahedron Lett. 2001, 42, 3299.
(152) Banerjee, M.; Roy, S. J. Mol. Cat. A 2005, 246, 231.
(153) Kimura, M.; Tomizawa, T.; Horino, Y.; Tanaka, S.; Tamaru, Y. Tetrahedron Lett.
2000, 41, 3627.
(154) Mukai, R.; Horino, Y.; Tanaka, S.; Tamaru, Y.; Kimura, M. J. Am. Chem. Soc. 2004,
126, 11138.
(155) Zhu, S.-F.; Yang, Y.; Liu, B.; Zhou, Q.-L. Org. Lett. 2005, 7, 2333.
(156) Kimura, M.; Shimizu, M.; Shibata, K.; Tazoe, M.; Tamaru, Y. Angew. Chem. Int. Ed.
2003, 42, 3392.
(157) Araki, S.; Kamei, T.; Hirashita, T.; Yamamura, H.; Kawai, M. Org. Lett. 2000, 2,
847.
(158) Hirashita, T.; Kambe, S.; Tsuji, H.; Omori, H.; Araki, S. J. Org. Chem. 2004, 69,
5054.
(159) Jang, T.-S.; Keum, G.; Kang, S. B.; Chung, B. Y.; Kim, Y. Synthesis 2003, 775.
(160) Trnka, T. M.; Grubbs, R. H. Acc. Chem. Res. 2001, 34, 18.
(161) Yashuda, M.; Hirata, K.; Nishio, M.; Yamamoto, A.; Baba, A. J. Am. Chem. Soc.
2002, 124, 13442.
(162) Kazmaier, U.; Zumpe, F. L. Angew. Chem. Int. Ed. 1999, 38, 1468.
(163) Zumpe, F. L.; Kazmaier, U. Synlett 1998, 1199.
(164) Kazmaier, U. Angew. Chem. Int. Ed. 1994, 33, 998.
(165) Kazmaier, U.; Krebs, A. Angew. Chem. Int. Ed. 1995, 34, 2012.
(166) Bakke, M.; Ochta, H.; Kazmaier, U.; Sugai, T. Synthesis 1999, 1671.
(167) Mock, M. L.; Michon, T.; Hest, J. C. M. v.; Tirrell, D. A. ChemBiochem 2006, 7, 83.
(168) Ellis, T. K.; Ueki, H.; Yamada, T.; Ohfune, Y.; Soloshonok, V. A. J. Org. Chem.
2006, 71, 8572.
(169) Soloshonok, V. A.; Cai, C.; Hruby, V. C.; Meervelt, L. V.; Yamazaki, T. J. Org.
Chem. 2000, 65, 6688.
(170) Schlienger, N.; Bryce, M. R.; Hansen, T. K. Tetrahedron 2000, 56, 10023.
(171) Naskar, D.; Roy, A.; Seibel, W. L.; Portlock, D. E. Tetrahedron Lett. 2003, 44, 8865.
(172) Sieber, J. D.; Morken, J. P. J. Am. Chem. Soc. 2006, 128, 74.
(173) Southwood, T. J.; Curry, M. C.; Hutton, C. A. Tetrahedron 2006, 62, 236.
(174) Nakamura, I.; Yamamoto, Y. Chem. Rev. 2004, 104, 2127.
(175) Fürstner, A.; Thiel, O. R. J.Org. Chem. 2000, 65, 1738.
(176) Li, H.; Blériot, Y.; Chantereau, C.; Mallett, J.-M.; Sollugoub, M.; Zhang, Y.;
Rodríguez-García, E.; Vogel, P.; Jiménez-Barbero, J.; Sinaÿ, P. Org. Biomol. Chem.
2004, 2, 1492.
(177) Lee, H. K.; Im, J. H.; Jung, S. H. Tetrahedron 2007, 63, 3321.
(178) Garber, S. B.; Kingsbury, J. S.; Gray, B. L.; Hoveyda, A. H. J. Am. Chem. Soc. 2000,
122, 8168.
48
(179)
(180)
(181)
(182)
(183)
(184)
(185)
(186)
(187)
(188)
(189)
(190)
(191)
(192)
(193)
(194)
(195)
(196)
(197)
Kennedy, J. W. J.; Hall, D. G. Angew. Chem. Int. Ed. 2003, 4732.
Gung, B. W.; Xue, X.; Roush, W. R. J. Am. Chem. Soc. 2002, 124, 10692.
Rauniyar, V.; Hall, D. G. J. Am. Chem. Soc. 2004, 126, 4518.
Gorla, F.; Togni, A.; Venanzi, L. M. Organometallics 1994, 13, 1607.
Longmire, J. M.; Zhang, X.; Shang, M. Organometallics 1998, 17, 4374.
Motoyama, Y.; Kawakami, H.; Shimozono, K.; Aoki, K.; Nishiyama, H.
Organometallics 2002, 21, 3408.
Stark, M. A.; Richards, C. J. Tetrahedron Lett. 1997, 38, 5881.
Gosiewska, S.; Veld, M. H. i.; Pater, J. J. M. d.; Bruijnincx, P. C. A.; Lutz, M.; Spek,
A. L.; Koten, G. v.; Gebbink, R. J. M. K. Tetrahedron Asymm. 2006, 17, 674.
Solin, N.; Kjellgren, J.; Szabó, K. J. Angew. Chem. Int. Ed. 2003, 42, 3656.
Zografidis, A.; Polborn, K.; Beck, W.; Markies, B. A.; Koten, G. v. Naturforsch., B
1994, 49, 1494.
Mehendale, N. C., Ph.D. Thesis: Silica Immobilized Pincer-Metal Complexes.
Catalysis, Recycling, and Retrospect on Active Species, 2007,
Utrecht University, The Netherlands, 2007.
Rosa, G. R.; Ebeling, G.; Dupont, J.; Monterio, A. L. Synthesis 2003, 2894.
Gandelman, M.; Vigalok, A.; Shimon, L. J. W.; Milstein, D. Organometallics 1997,
16, 3981.
Gandelman, M.; Konstantinovski, L.; Rozenberg, H.; Milstein, D. Chem. Eur. J.
2003, 9, 2595.
Gong, J.-F.; Zhang, Y.-H.; Song, M.-P.; Xu, C. Organometallics 2007, 26, 6478.
Yuan, J.-C.; Lu, S.-J. Organometallics 2001, 20, 2697.
Dupont, J.; Pfeffer, M.; Daran, J. C.; Jeannin, Y. Organometallics 1987, 6, 899.
Dupont, J.; Pfeffer, M. J. Chem. Soc. Dalton Trans. 1990, 3193.
Gorla, F.; Venanzi, L. M.; Albinati, A. Organometallics 1994, 13, 43.
49