Palladium-catalyzed dienylations of chelated enolates

Article abstract of DOI:10.1002/ejoc.200800519

Isomerization-free reactions of dienyl carbonates with chelated amino acid ester enolates at -78 °C provide important information concerning the mechanism of these dienylations. The formation of regioisomeric products can be explained by competing SN2/SN2′ reactions, and the product distribution can be influenced by the proper choice of the reaction conditions. Chiral allylic substrates show a significant transfer of chirality. Wiley-VCH Verlag GmbH & Co. KGaA, 2008.

Full text of DOI:10.1002/ejoc.200800519

FULL PAPER
DOI: 10.1002/ejoc.200800519
Palladium-Catalyzed Dienylations of Chelated Enolates
Sankar Basak[a] and Uli Kazmaier*[a]
Keywords: Allylations / Amino acids / Chelates / Dienylations / Enolates / Palladium
Isomerization-free reactions of dienyl carbonates with che-
lated amino acid ester enolates at –78 °C provide important
information concerning the mechanism of these dienylations.
The formation of regioisomeric products can be explained by
competing SN2/SN2Ј reactions, and the product distribution
can be influenced by the proper choice of the reaction condi-
tions. Chiral allylic substrates show a significant transfer of
chirality.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2008)
preferentially at the terminal position, whereas reaction via
a cationic SN1 transition state should provide a higher ratio
of the branched product. The SN1 transition state can be
favoured by using electron-withdrawing groups in the li-
gand, for example if phosphanes are replaced by phos-
phites.[5]
Introduction
(π-Allyl)metal complexes are of significant importance in
modern organic synthesis. Among the different metals used,
palladium plays a dominant role,[1] and the allylic alky-
lation, also called the Tsuji–Trost reaction, is by far the
most popular application.[2] (π-Allyl)palladium complexes
B are formed from allylic substrates such as A1 and A2,[3]
which can be attacked in the next step either at the terminal
position, giving rise to the linear product C1, or at the in-
ternal position, which provides the branched product C2
(Scheme 1). In general, attack of the nucleophile occurs at
the sterically least hindered position, and therefore the lin-
ear product C1 is the preferred one.[2]
According to the simplified model shown in Scheme 1,
one should expect more or less the same product distribu-
tion, independent of the starting material used, as long as
the same π-allyl intermediate B is formed. Interestingly,
quite often this is not the case. Especially with unsymmetric
substrates, such as A1 and A2, a substrate-dependent prod-
uct ratio is observed. Depending on the reaction conditions,
branched substrate A2 can give a higher ratio of branched
product C2 than the linear substrate A1.[6] This so-called
“memory effect” has intensively be studied, and several ex-
planations were given.[7] For example, in the presence of
chloride ions an unsymmetrical π-allyl complex BЈ can be
formed from substrates A2 (Scheme 2). Based on the strong
trans effect of the phosphorus atom, the phosphane ligand
should be located trans to the leaving group. Because nucleo-
philic attack occurs also preferentially trans to the phos-
phorus atom, this trans effect can significantly contribute
to the memory effect (regioretention).[8]
Scheme 1. Simplified mechanism of the Tsuji–Trost reaction.
In general, for an asymmetric version the branched prod-
uct C2 is of greater interest, because the linear C1 in most
cases is achiral, as long as symmetric nucleophiles such as
malonates are used. Therefore, much efforts have been
made to obtain higher ratios of branched product. Pfaltz et
al. could show, that the regioselectivity of the nucleophilic
attack can be influenced by “tuning” the reaction mecha-
nism.[4] Nucleophilic attacks in an SN2-type fashion occurs
Scheme 2. Formation of unsymmetrical (π-allyl)palladium com-
plexes.
[a] Institut für Organische Chemie, Geb. C4.2, Universität des
Saarlandes,
Another important class of memory effects, also ob-
served for unsymmetrical π-allyl complexes, is more com-
plex. It is observed that the product distribution not only
depends on the regio- but also the stereochemistry of the
starting material.[6b] This effect was investigated by Norrby
66041 Saarbrücken
Fax: +49-681-3022409
E-mail: u.kazmaier@mx.uni-saarland.de
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Eur. J. Org. Chem. 2008, 4169–4177
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4169

S. Basak, U. Kazmaier
FULL PAPER
et al.[9] and recently also by our group.[10] Scheme 3 shows served with other transition metals such as molybdenum,[16]
a more detailed mechanism of the allylation using unsym- and iridium.[17] In principle, this result can be explained by
metrical substrates A1–A3.
the formation of (π-allyl)palladium complex E1, which can
be attacked at the two different allylic positions giving rise
to products F1 and F2. More detailed investigations by
Trost et al.[18] and simultaneously by Bäckvall et al.[19] indi-
cated, that with these substrates D the situation is much
more complex. With respect, that with sodium malonates a
mixture of substitution products F1 and F2 (not F3) is
formed from D1 in a ratio of 2–4:1 (depending on the ma-
lonate), one might expect a product mixture F2/F3 from
substrate D3 (via E3). Dienyl substrate D2 should be able
to form both π-allyl complexes E1 and E3, and therefore,
all three products F can be expected. In addition, D2 should
be able to form an allyl anti-vinyl complex,[9] analogous to
B2 (not shown here for clarity reasons), which should favor
the formation of F2.[6b] But the results obtained are quite
different. Both, D1 and D2 give rise to F1 (major product)
and F2 (minor product), while D3 provided all three prod-
ucts (F2 as major isomer). This clearly indicates that the
initially formed π-allyl complex E3 can isomerize (at least
in part) to π-allyl complex E1, probably via the σ-allyl inter-
mediate E2 (π-σ-π isomerization). Interestingly, not only
the structure of the malonate has an influence on the re-
gioselectivity, but also the counterion.[18] Replacement of
the sodium ion by caesium or tetrahexylammonium gave a
completely different result. All these substrates D1–D3 gave
more or less the same product ratio F1–F3, indicating that
under these conditions all (allyl)palladium intermediates
E1–E3 are in full equilibrium (Scheme 4).
Scheme 3. Detailed mechanism of allylic alkylation.
Whereas A1 gives rise to the syn complex B1, the anti
complex B2 is obtained from cis substrate A3.[11] The
branched substrate A2 is able to form both, the syn and the
anti complex, and the ratio depends on the conformation
of the substrate in the ionization step. Therefore, a strong
influence of the leaving group can be observed in certain
cases.[9] Especially for terminal allyl complexes the interme-
diates B1 and B2 can equilibrate rapidly by π-σ-π isomer-
ization.[12] If isomerization is fast compared to nucleophilic
attack, product distribution results from the equilibrium
mixture in which the syn complex as the thermodynamically
favoured one is enriched. In this case the product ratio is
nearly independent of the allylic substrate used (Scheme 1).
On the other hand, if nucleophilic attack is faster than the
equilibrium, or at least in the same range, one can observe
memory effects based on the different ratio and reactivity
of the allyl complexes B1 and B2.[10,13]
Another explanation for the product ratio formed from
D3 might be SN2Ј-type processes. Both steps, the oxidative
addition (on substrates D1 and D3, but not D2) as well as
the nucleophilic attack, can also occur in an SN2Ј-type fash-
ion. In addition to the “normal” allylic alkylation, this
might explain the formation of all three products,[18b] and
it is extremely difficult to differentiate between these mech-
anistic options. Under certain circumstances, also isomer-
ization of the products can be observed.[14]
Even more complicated is the situation if higher unsatu-
rated substrates such as D are used. Early investigations by
Trost et al. using dienyl substrate D1 and amines or sodium
malonates as nucleophiles showed that in the palladium-
catalyzed reaction nucleophilic attack occurred preferen-
An interesting observation was made by Bäckvall et
tially at the least hindered terminal position (F1),[14] while al.,[19] that PBu3 instead of PPh3 gives a significantly higher
with tungsten complexes the branched product F2 was ob- ratio of linear product F1. This was explained by a dimin-
tained preferentially.[15] The same regioselectivity is also ob- ished carbenium ion character at the π-allyl carbon atom in
Scheme 4. Allylic alkylations with dienyl substrates D.
4170
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 4169–4177

Palladium-Catalyzed Dienylations of Chelated Enolates
the π-allyl complex, which favours attack at the terminal the memory effect,[10] or to differentiate between SN2Ј-type
position via an SN2-type transition state.[14] On the other reactions and π-σ-π isomerization of type E allyl com-
hand, replacement of the phosphanes by phosphites or re- plexes.[25]
lated ligands gives preferentially rise to branched product
as illustrated by a recent example using a chiral 1-oxazol-
inyl-1Ј-phosphanylferrocene complex.[20]
In all examples investigated so far, either symmetrical C-
nucleophiles or amines were used, generating only one
stereogenic center in the allyl fragment if branched products
are obtained (F2, F3; R ϶ H). To the best of our knowl-
edge, there is no report on diastereoselective reactions using
unsymmetrical nucleophiles.
Scheme 5. Allylic alkylations of chelated enolate G (Tfa = trifluo-
roacetyl).
For some time, our group is investigating chelated amino
acid ester enolates G as nucleophiles in transition-metal-
catalyzed allylic alkylations (Scheme 5).[21] The great advan-
Results and Discussion
To obtain results comparable to those previously ob-
tage of these nucleophiles is their high reactivity. Therefore, tained with malonates, we first investigated the reaction of
they react under much milder conditions (already at –78 °C) the three isomeric allyl substrates 1–3 (Table 1).[25] Accord-
than the generally used nucleophiles (such as malonates). ing to Scheme 4, these substrates should form the three dif-
As a result, these were the first C-nucleophiles showing no ferent products 4–6 in different ratios, depending on the π-
π-σ-π isomerization of complexes obtained from 1,3-disub- allyl complexes E1–E3 (R = Me) involved. An excess of
stituted substrates.[22] For example, cis-configured sub- nucleophile was used to obtain high reaction rates and to
strates can react either with conservation of the olefin ge- suppress isomerization as much as possible.[26] We started
ometry (stereoretention), or selectively at the more reactive our investigation with the linear acetate 1a which gave the
anti position,[13] depending on the substrate used. Very re- allylation products in high yield as a mixture of all three
cently, we could also show that in reactions of chiral sub- isomers. The linear product was formed as a single isomer
strates H the nucleophilic attack on the terminal π-allyl in- [both double bonds (E)]. Interestingly, the branched iso-
termediate could be controlled by the stereogenic centre in mers 5 and 6 were formed in equal amounts which indicates
the substrate.[23] This clearly indicates, that with these enol- that either π-σ-π isomerization occurs or the nucleophilic
ates even terminal (π-allyl)palladium complexes can react attack occurs in an SN2Ј-type fashion.
without significant isomerization[24] which makes them
In general, allylic carbonates are more reactive than car-
good candidates to investigate mechanistical details such as boxylates, and in our previous investigations isomerization
Table 1. Allylic alkylations of dienyl substrates 1–3.
Entry
Substrate
X
Reaction conditions
Yield [%]
Ratio
4/5/6
dr (anti/syn)[a]
5
6
1
2
3
4
5
6
7
1a
1b
2b
3b
1b
2b
3b
OCOMe
OCOOEt
OCOOEt
OCOOEt
OCOOEt
OCOOEt
OCOOEt
–78 °C to room temp., overnight
–78 °C, 16 h
89
92
64
87
83
25
41
46:26:28
57:22:21
54[b]:31:15
36:13:51
37:20:43
35[c]:16:49
31:12:57
92:8
97:3
90:10
95:5
92:8
88:12
94:6
65:35
65:35
33:67
50:50
61:39
42:58
35:65
–78 °C, 16 h
–78 °C, 16 h
room temp.
room temp.
room temp.
[a] Diastereomeric ratio determined by GC of the crude product mixture. [b] Product contains 36% (E/Z) isomer 4Ј (GC). [c] Product
contains 12% (E/Z) isomer 4Ј (GC).
Eur. J. Org. Chem. 2008, 4169–4177
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4171

S. Basak, U. Kazmaier
FULL PAPER
could mostly be suppressed by using carbonates, while with to be the thermodynamical one. In principle, especially at
acetates isomerization (at least in part) was observed. higher temperature these two complexes can equilibrate by
Therefore, we next allowed allylic carbonate 1b to react at π-σ-π isomerization, and the equilibrium probably is on the
–78 °C (Entry 2). Again, the yield was excellent, which indi- side of K1. Both complexes can be attacked at the two al-
cates that the reaction proceeds very well at this low tem- lylic positions (a,b) and as proposed by Trost, also at the
perature. The selectivity towards the linear product 4 was SN2Ј position (c).
increased, but surprisingly the branched products 5 and 6
Nucleophilic attack at the positions a and aЈ provide
again were formed as a 1:1 mixture. Interestingly, under both the same product 5, whereas attacks b and bЈ should
these conditions product 5 was obtained nearly as a single give rise to branched products 6 and 6Ј. The formation of
diastereomer (97% ds), while the isomeric products 6 were 6 [with an (E) double bond] is a result of attack b, whereas
formed in a 2:1 ratio. The lower selectivity in this case is not the product 6Ј from attack bЈ with a (Z)-configured double
surprising, because for a nucleophilic attack at the central bond is not observed. But this is not surprising. Work by
position, the two substituents at this allylic carbon atom are Åkermark et al. clearly indicates, that the anti position (at-
too similar for significant stereodifferentiation.
tack aЈ) is the more reactive one,[13] and in our case, the
Next, we switched to the regioisomeric substrate 2b and SN2Ј attack (cЈ) at the terminal position is obviously signifi-
subjected it to the same reaction conditions (Entry 3). The cantly favoured over bЈ. This explains the relatively high
yield was lower with this substrate, but the linear product 4 amount of (Z)-configured product 4Ј. The major product 4
was again obtained as the major product, surprisingly in is obtained from the analogous attack c on the thermody-
this case as a 2:1 mixture of 4 and the (Z) isomer 4Ј (Fig- namical complex K1. The high amount of 5 (obtained from
ure 1); 5 was formed as the major isomer of the two both complexes K1 and K2) and of 4Ј (obtained only from
branched products, again with good selectivity.
K2) indicates that the complexes K1 and K2 are formed
in comparable amounts, and that isomerization under our
reaction conditions is slow compared to nucleophilic attack.
But if isomerization of the complexes K1 and K2 can be
neglected, one should not expect a significant isomerization
between the two regioisomeric π-allyl complexes E1 and E3
(Scheme 4). Therefore, allylic substrate 3b was investigated
next (Entry 4), because this substrate should give rise to a
mixture of these complexes (which might be close to the
equilibrium mixture). In high yield the branched product 6
was formed as the major isomer (51%). In addition to a
potential memory effect, this product can be obtained by
direct nucleophilic attack on both isomeric complexes E1
and E3.
Figure 1. Minor linear product 4Ј.
The relatively high amount of 5 (compared to 6) might
result (at least in part) from the memory effect, but on the
other hand it might also be the result of the reaction mecha-
nism (Scheme 6). Previous investigations by Norrby[9] and
our group[10] showed that especially methyl-substituted al-
lylic substrates can give significant amounts of anti com-
plexes,[11] depending on the leaving groups used.
Because this product mixture probably arises from a mix-
ture of regioisomeric complexes, one might expect similar
results under conditions where isomerization occurs. There-
fore, all these substrates were allowed to react also at room
temperature (Entries 5–7). And indeed, the results obtained
are relatively similar. Only the yields differ significantly,
which is the result of several side reactions occurring at
higher temperature. But these reactions were not optimized,
because we were only interested in mechanistical infor-
mation. Obviously, at room temperature all π-allyl com-
plexes are more or less in equilibrium. This assumption is
also supported by the observation that the amount of linear
product 4 as well as the (Z) isomer 4Ј is significantly re-
duced (Entry 6 vs. 3). The fact that 4Ј still is observed indi-
cates that even at room temperature isomerization is not
complete and not so much faster than nucleophilic attack.
Therefore, isomerization of the π-allyl intermediates can be
ruled out at temperatures as low as –78 °C. The somewhat
lower selectivities observed with allylic acetate 1a (Entry 1)
might be caused by a competing isomerization, based on an
Scheme 6. Formation of isomeric products 4–6 from allylic sub-
strate 2.
With this in mind one can propose a similar situation ionization of acetates at higher temperatures.
also for dienyl substrates such as 2. Depending on the con- Because the leaving group obviously might have an influ-
formation in the ionization step, both π-allyl complexes K1 ence, not only on the ratio isomerization/nucleophilic at-
(syn/syn) and K2 (anti/syn) can be formed, with K1 expected tack, but also on the mode of ionization, we investigated
4172
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 4169–4177

Palladium-Catalyzed Dienylations of Chelated Enolates
Table 2. Influence of the leaving group on the product distribution.
[a] Diastereomeric ratio and enantiomeric excess determined by GC after catalytic hydrogenation. [b] Ar: 2,4-dichlorophenyl.
this influence more detailed in reactions of substrates 7. strates 7b and 7e. If isomerization occurs, one should ob-
These substrates have the great advantage that in principle serve a complete loss of chirality, because the σ-allyl inter-
only two products should be obtained resulting from a nu- mediate M is achiral (Scheme 7). Nucleophilic attack a then
cleophilic attack at either the external position (giving rise provides a 1:1 mixture of the two enantiomers 8 and ent-8.
to conjugated diene 8) or internal, providing product 9 with If no isomerization occurs, loss of chirality can also be ex-
only one stereogenic center at the α-position. Attack under pected by nucleophilic attack in an SN2-type fashion (c),
SN2 conditions also gives rise to 8, although in this case the but attacks a and c should show different kinetics, and
formation of (Z) product 8Ј can be expected. The results therefore at least in part a chirality transfer should be ob-
obtained from the different leaving groups are summarized served.
in Table 2.
As expected, the carbonates 7a–c were superior to car-
boxylates, while the differences between the carbonates were
minimal. The selectivity slightly decreased with the elec-
tron-donor ability of the alkyl group, and best results were
obtained with the methyl carbonate. More obvious was the
effect for the carboxylates, and the selectivity directly corre-
lated with the pKa value of the corresponding carboxylic
acid. Whereas the 2,4-dichlorobenzoate 7g (pKa = 2.67)
gave results very close to the carbonates (Entry 7), the piva-
loate 7d (pKa = 5.05) showed a mixture of isomers which is
close to the results obtained at room temperature (Entry 4).
Probably with this leaving group isomerization occurs (at
least in a significant ratio). The acetate (pKa = 4.76) and
benzoate (pKa = 4.20) gave results between these extremes.
Unfortunately, we were not able to introduce better leav-
ing groups, such as phosphates, because the required sub-
strates were not stable and decomposed immediately. It
should also be mentioned, that the substrates 7 are rela-
tively sensitive. For example, during purification of 7b by
Scheme 7. Formation of isomeric products from optically active 7.
silica gel chromatography, we observed scrambling of the
leaving group, presumably by an ionic mechanism.[26] For-
To prove this option, we subjected the corresponding al-
tunately, there was no scrambling of the leaving group in lyl alcohol to an enzymatic kinetic resolution by using an
the crude substrate, and it was found sufficiently pure by immobilised Candida antarctica lipase (Novozym 435). In
NMR spectroscopy. Therefore, we decided to use these vinyl acetate we obtained the required acetate (R)-7e with
crude substrates without purification.
66% ee. The remaining alcohol (91% ee) was esterified with
To investigate whether the acetate undergoes isomeriza- ethyl chloroformate to (S)-7b and was used without further
tion, we repeated our experiments with optically active sub- purification. And indeed, as expected, the acetate (R)-7e
Eur. J. Org. Chem. 2008, 4169–4177
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4173

S. Basak, U. Kazmaier
FULL PAPER
Table 3. Allyic alkylations with various dienyl substrates 10.
[a] Diastereomeric ratio and enantiomeric excess determined by GC of the crude product mixture. [b] Product contains less than 2% (E/
Z) isomer 12aЈ (GC). [c] Product contains 2–3% (E/Z) isomer 12aЈ (GC). [d] Product contains 3% (E/Z) isomer 12aЈ (GC).
yielded the allylation products in complete racemic form the electron-rich derivative 10e gave a higher ratio of
(Entry 8), whereas the corresponding carbonate (S)-7b gave branched product as well as SN2Ј, while the electron-poor
enantiomerically enriched products (Entry 9).
substrate 10f provided the conjugated diene 11f preferen-
A relatively high ee was obtained for the minor syn dia- tially, but the effects are only moderate. With electron-rich
stereomer (syn-8). Probably, this isomer is formed from the heterocyclic systems, such as the furane derivative, a similar
corresponding anti/syn complex K2 (Scheme 6). The higher situation is found as with the p-methoxy substrate 10e.
reactivity of the anti position should favour this position vs.
the SN2Ј position, which should result in a higher ee. The
moderate ee for the branched product 9 results from the
unselective attack at the internal position, comparable to
the low diastereoselectivities obtained with substrates 1–3.
To show the generality of these observations, we investi-
gated a range of other substrates. The results obtained are
collected in Table 3. With the methyl/ethyl-substituted sub-
strate 10a the results were comparable to those obtained
with the dimethylated substrates. But in this case, the SN2
and the SN2Ј attacks gave different products, and therefore
a high degree of chirality transfer was observed, especially
under “isomerization-free” conditions (Entry 4). At higher
temperature, again the relative amount of the branched
product 13 was increased. Probably because of a conjuga-
tion effect, the amount of the SN2Ј products 12 is slightly
Conclusions
We have shown that in dienylations of chelated enolates
isomerization processes of the (π-allyl)palladium complexes
involved can be suppressed and that the formation of iso-
meric products is the result of competing SN2/SN2Ј reac-
tions. By choosing proper reaction conditions (kinetic or
thermodynamic control) the product ratio can be manipu-
lated, at least in part. A high degree of chirality transfer
can be observed in reactions of chiral allylic substrates.
Experimental Section
General Remarks: All reactions were carried out in oven-dried
reduced in reactions of the phenyl-substituted substrates glassware (100 °C) under argon or nitrogen. All solvents were dried
before use. THF, diethyl ether, hexane and toluene were distilled
from sodium/benzophenone, CH2Cl2 from P2O5, MeOH from acti-
vated Mg, and stored over molecular sieves. Et3N and pyridine
were distilled from KOH. The products were purified by flash
chromatography on silica gel (32–63 µm). Mixtures of EtOAc and
hexane or diethyl ether were generally used as eluents. Analysis by
TLC was carried out on commercially precoated Polygram® SIL-G/
UV 254 plates (Macherey–Nagel). Visualization was accomplished
with UV light, KMnO4 solution or iodine. Melting points were
determined with a Büchi melting point apparatus and are uncor-
10c and 10d. Especially with the chiral carbonates, the selec-
tivities and the chirality transfers were excellent. Aryl sub-
strates are best suited to vary the electronic properties of
the substrates. While electron-donating groups on the aro-
matic ring should favour an SN1-type reaction by stabilizing
a
potential carbeniumion intermediate, electron-with-
drawing groups should destabilize these intermediates, pre-
ferring an SN2-type reaction. Therefore, we also investi-
gated the p-methoxy derivative 10e as well as the p-fluoro
substrate 10f. Compared to the unmodified substrate 10c, rected. 1H and 13C NMR spectroscopic analysis was performed
4174
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 4169–4177

Palladium-Catalyzed Dienylations of Chelated Enolates
with Bruker AC-400 or DRX-500 spectrometers. Chemical shifts allylic alkylation using 1b as a diastereomeric mixture. anti-5 (se-
1
are reported in ppm relative to CHCl3 as an internal reference. If
selected signals are provided for the other isomeric products, these
signals are extracted from the NMR spectra of the isomeric mix-
tures. Enantiomeric and diastereomeric excesses were determined
with Varian Chrompack CP 3380 and GC 3400 gas chromato-
graphs with a Chira-Si--Val (25 mϫ0.25 mm) and an FS-Cyclo-
dex-β-I/P capillary column (25 mϫ0.25 mm), respectively. Helium
was used as carrier gas. Enantiomeric excesses were also deter-
mined by analytical HPLC with a Trentec Reprosil-100 Chiral-NR
8 µm column (flow rate: 1–2 mL/min) and a Shimazu diode-array
UV detector. HRMS analyses were performed with a Finnigan
MAT 95S. Elemental analyses were carried out at the Institute of
Organic Chemistry, Saarland University, Germany.
lected signals): H NMR (400 MHz, CDCl3): δ = 6.71 (br. s, 1 H,
NH), 6.29 (dt, J = 16.8, 10.4 Hz, 1 H, CH=CH2), 6.16–6.04 (m, 1
H, CH=CH), 5.53 (dd, J = 15.6, 8 Hz, 1 H, CH=CH), 5.22–5.06
(m, 2 H, CH=CH2), 4.58–4.48 (m, 1 H, CHN), 2.96–2.82 (m, 1
H, CHCH3), 1.10 (d, J = 6.8 Hz, 3 H, CHCH3) ppm. 13C NMR
(100 MHz, CDCl3): δ = 168.7 (s, COO), 136.2 (d, CH=CH2), 133.2
(d, CH=CH), 132.2 (d, CH=CH), 117.4 (t, CH=CH2), 57.1 (d,
CHN), 39.5 (d, CHCH3), 16.0 (q, CHCH3) ppm. syn-5 (selected
1
signals): H NMR (400 MHz, CDCl3): δ = 1.14 (d, J = 6.8 Hz, 3
H, CHCH3) ppm. 13C NMR (100 MHz, CDCl3): δ = 35.9 (d,
CHCH3), 16.0 (q, CHCH3) ppm. GC [Chirasil--Val, T0 (30 min)
= 80 °C, 1 °C/min to T = 180 °C]: tR(anti-5) = 58.28 min, tR(anti-
5) = 59.13 min, tR(syn-5) = 59.57 min, tR(syn-5) = 60.41 min.
General Procedure for Palladium-Catalyzed Dienylations: Hexa-
methyldisilazane (225 mg, 1.40 mmol) was dissolved in dry THF
(2 mL) in a Schlenk flask under Ar or N2. To this solution nBuLi
(1.6  in hexane, 0.80 mL, 1.25 mmol) was added dropwise at
–78 °C. The reaction mixture was stirred at this temperature for
30 min, before the cooling bath was removed. Stirring was contin-
ued for further 15 min. In a second Schlenk flask, ZnCl2 (102 mg,
0.74 mmol) was carefully dried in vacuo with a heat gun. After
cooling to room temperature, it was dissolved in dry THF (2 mL),
(؎)-tert-Butyl (4E)-2-(Trifluoroacetamido)-3-vinyl-4-hexenoate (6):
Ester 6 was obtained as one of the minor products in the allylic
alkylation using 1b as a diastereomeric mixture. Selected signals:
1H NMR (400 MHz, CDCl3): δ = 6.71 (br. s, 1 H, NH), 5.83–5.70
(m, 1 H, CH=CH2), 5.63–5.49 (m, 1 H, CHCH3), 5.42–5.31 (m, 1
H, CH=CH), 5.22–5.06 (m, 2 H, CH=CH2), 4.58–4.48 (m, 1 H,
CHN), 3.32–3.17 (m, 1 H, CHCH), 1.70 (d, J = 6.8 Hz, 3 H,
CHCH3) ppm. 13C NMR (100 MHz, CDCl3): δ = 168.4 (s, COO),
135.3 (d, CH=CH2), 126.9 and 126.8 (2 d, CH=CH), 120.0 and
and a solution of tert-butyl N-(trifluoroacetyl)glycinate (114 mg, 119.9 (2 t, CH=CH2), 55.95 and 55.92 (2 d, CHN), 49.7 and 49.6
0.50 mmol) in dry THF (2 mL) was added. This mixture was cooled
to –78 °C, the freshly prepared cooled LHMDS solution was added
dropwise, and stirring was continued for 45 min. In a third flask
(2 d, CHCH), 17.9 (q, CHCH3) ppm. GC [Chirasil--Val, T0
(30 min) = 80 °C, 1 °C/min to T = 180 °C]: tR(6, diast. 1) =
50.97 min, tR(6, diast. 2) = 51.31 min, tR(6, diast. 1) = 51.93 min,
[Pd(allyl)Cl]2 (1.8 mg, 5.0 µmol, 2 mol-%) and PPh3 (3.9 mg, tR(6, diast. 2) = 52.20 min.
14.87 µmol) were dissolved in dry THF (0.5 mL). After this solu-
(؎)-tert-Butyl (4E,6E)-3-Methyl-2-(trifluoroacetamido)-4,6-nona-
tion was stirred at room temperature for 15 min, it was added to
the chelated enolate at –78 °C, followed by a cold solution of the
conjugated dienyl substrate (0.25 mmol) in dry THF (1 mL). The
mixture was warmed slowly to room temperature overnight. The
solution was diluted with diethyl ether, and 1  aq. KHSO4 solu-
tion was added. The layers were separated, and the aqueous layer
was extracted with diethyl ether. The combined organic extracts
were dried (Na2SO4), and the solvent was evaporated under re-
duced pressure. The crude product was purified by flash
chromatography on silica gel (hexane/ethyl acetate).
dienoate (11a): According to the general procedure, 11a was ob-
tained as major product from allylic carbonate 10a (50 mg,
0.25 mmol) in an overall yield (mixture of isomers) of 85% (71 mg,
0.21 mmol) as a colourless oil after flash chromatography (hexanes/
EtOAc, 98:2). anti-11a: 1H NMR (400 MHz, CDCl3): δ = 6.67 (br.
d, J = 8.0 Hz, 1 H, NH), 6.07 (ddd, J = 14.8, 10.1, 0.8 Hz, 1 H,
CH=CH), 6.11–5.93 (m, 1 H, CH=CH), 5.71 (dt, J = 14.8, 6.4 Hz,
1 H, CH=CH), 5.37 (dd, J = 14.8, 7.5 Hz, 1 H, CH=CH), 4.49
(dd, J = 8.4, 4.4 Hz, 1 H, CHN), 2.95–2.80 (m, 1 H, CHCH3), 2.09
(dqd, J = 7.5, 7.4, 1.0 Hz, 2 H, CH2CH3), 1.48 [s, 9 H, C(CH3)3],
(؎)-tert-Butyl (4E,6E)-2-(Trifluoroacetamido)-4,6-octadienoate (4): 1.09 (d, J = 6.8 Hz, 3 H, CHCH3), 1.00 (t, J = 7.5 Hz, 3 H,
According to the general procedure, 4 was obtained as major prod-
uct from allylic carbonate 1b (42.5 mg, 0.25 mmol) in an overall
yield (mixture of isomers) of 92% (70 mg, 0.23 mmol) as a colour-
less oil after flash chromatography (hexanes/EtOAc, 98:2). 1H
CH2CH3) ppm. 13C NMR (100 MHz, CDCl3): δ = 168.9 (s, COO),
157.0 (q, J = 37.4 Hz, CON), 136.8 (d, CH2CH), 132.9 (d,
CH=CH), 128.8 (d, CH=CH), 128.4 (d, CH=CH), 115.7 (q, J =
286 Hz, CF3), 83.2 [s, C(CH3)3], 57.2 (d, CHN), 39.4 (d, CHCH3),
NMR (400 MHz, CDCl3): δ = 6.87 (br. s, 1 H, NH), 6.16–6.01 (m, 27.9 [q, C(CH3)3], 25.5 (t, CH2CH3), 16.33 (q, CHCH3), 13.3 (q,
1 H, CH=CH), 6.10–5.94 (m, 1 H, CH=CH), 5.71–5.59 (m, 1 H, CH2CH3) ppm. syn-11a (selected signals): 1H NMR (400 MHz,
CH=CH), 5.38–5.27 (m, 1 H, CH=CH), 4.58–4.48 (m, 1 H, CHN), CDCl3): δ = 6.78 (br. d, 1 H, NH), 4.46 (dd, J = 8.4, 5.2 Hz, 1 H,
2.75–2.50 (m, 2 H, CHCH2), 1.74 (d, J = 6.8 Hz, 3 H, CHCH3), CHN), 2.77–2.62 (m, 1 H, CHCH3), 1.49 [s, 9 H, C(CH3)3], 1.12
1.48 [s, 9 H, C(CH3)3] ppm. 13C NMR (100 MHz, CDCl3): δ = (d, J = 6.8 Hz, 3 H, CHCH3) ppm. 13C NMR (100 MHz, CDCl3):
169.2 (s, COO), 156.4 (q, J = 37.4 Hz, CON), 135.4 (d, CH=CH),
δ = 40.0 (d, CHCH3), 16.37 (q, CHCH3) ppm. GC (Chirasil--
130.6 (d, CH=CH), 129.7 (d, CH=CH), 122.3 (d, CH=CH), 115.6 Val, 100 °C, isothermic): tR(anti-11a) = 111.20 min, tR(anti-11a) =
(q, J = 286 Hz, CF3), 83.4 [s, C(CH3)3], 52.8 (d, CHN), 34.8 (t,
CHCH2), 27.9 [q, C(CH3)3], 18.0 (q, CHCH3) ppm. GC [Chirasil-
-Val, T0 (30 min) = 80 °C, 1 °C/min to T = 180 °C]: tR(4) =
115.24 min, tR(syn-11a) = 117.25 min, tR(syn-11a) = 121.01 min.
HRMS (mixture of isomers) (CI): calcd. for C16H25NO3F3 [M +
1]+ 336.1786, found 336.1819. C16H24NO3F3 (335.37) (mixture of
isomers): calcd. C 57.30, H 7.21, N 4.18; found C 56.96, H 7.33,
N 4.34.
72.04 min, tR(4)
= 72.73 min, tR(4Ј) = 72.91 min, tR(4Ј) =
73.43 min. HRMS (mixture of isomers) (CI): calcd. for
C14H20NO3F3 [M]+ 307.1395, found 307.1407; calcd. for
C14H21NO3F3 [M + 1]+ 308.1473, found 308.1484. C14H20NO3F3
(307.31) (mixture of isomers): calcd. C 54.72, H 6.56, N 4.56; found
C 54.55, H 6.66, N 4.80.
(؎)-tert-Butyl
(4E,6E)-3-Ethyl-2-(trifluoroacetamido)-4,6-octa-
dienoate (12a): Ester 12a was obtained as one of the minor products
in the allylic alkylation using 10a. 1H NMR (400 MHz, CDCl3)
(selected signals): δ = 6.67 (br. d, J = 8 Hz,1 H; NH), 5.76–5.62
(m, 1 H, CHCH3), 5.21 (dd, J = 14.8, 9.2 Hz, 1 H, CH=CH), 4.58
(dd, J = 8.8, 4.4 Hz, 1 H, CHN), 2.62–2.46 (m, 1 H, CHCH2), 1.75
(؎)-tert-Butyl (4E)-3-Methyl-2-(trifluoroacetamido)-4,6-heptadieno-
ate (5): Ester 5 was obtained as one of the minor products in the
Eur. J. Org. Chem. 2008, 4169–4177
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4175

S. Basak, U. Kazmaier
J. Ansell, M. Wills, Chem. Soc. Rev. 2002¸ 31, 259–268.
a) T. Hayashi, A. Yamamoto, T. Hagihara, J. Org. Chem. 1986,
51, 723–727; b) M. P. T. Sjögren, S. Hansson, B. Åkermark, A.
Vitagliano, Organometallics 1994, 13, 1963–1971; c) T. Hayashi,
M. Kawatsura, Y. Uozumi, J. Am. Chem. Soc. 1998, 120, 1681–
1687; d) A. J. Blacker, M. L. Clarke, M. S. Loft, J. M. J. Wil-
liams, Org. Lett. 1999, 1, 1969–1971.
a) B. M. Trost, R. C. Bunt, J. Am. Chem. Soc. 1996, 118, 235–
236; b) G. C. Lloyd-Jones, S. C. Stephen, Chem. Eur. J. 1998,
4, 2539–2549; c) G. C. Lloyd-Jones, S. C. Stephen, M. Murray,
C. P. Butts, S. Vyskocil, P. Kocovsky, Chem. Eur. J. 2000, 6,
4348–4357; d) G. Poli, C. Scolastico, Chemtracts 1999, 12, 837–
845; e) G. C. Lloyd-Jones, Synlett 2001, 161–183; f) I. J. S. Fair-
lamb, G. C. Lloyd-Jones, S. Vyskocil, P. Kocovsky, Chem. Eur.
J. 2002, 8, 4443–4453; g) L. Gouriou, G. C. Lloyd-Jones, S.
Vyskovil, P. Kocovsky, J. Organomet. Chem. 2003, 687, 525–
537; h) B. J. Lussem, H.-J. Gais, J. Org. Chem. 2004, 69, 4041–
4052; i) J. W. Faller, N. Sarantopoulos, Organometallics 2004,
23, 2179–2185.
a) J. Sprinz, M. Kiefer, G. Helmchen, M. Reggelin, G. Huttner,
O. Walter, L. Zsolnai, Tetrahedron Lett. 1994, 35, 1523–1526;
b) J. M. Brown, D. I. Hulmes, P. J. Guiry, Tetrahedron 1994, 50,
4493–4506; c) P. E. Blöchl, A. Togni, Organometallics 1996, 15,
4125–4132; d) K. Fagnou, M. Lautens, Angew. Chem. 2002,
114, 26–49; Angew. Chem. Int. Ed. 2002, 41, 26–47; e) G. C.
Lloyd-Jones, S. C. Stephen, Chem. Commun. 1998, 2321–2322;
f) B. Goldfuss, U. Kazmaier, Tetrahedron 2000, 56, 6493–6496.
P. Fristrup, T. Jensen, J. Hoppe, P.-O. Norrby, Chem. Eur. J.
2006, 12, 5352–5360.
FULL PAPER
[5]
[6]
(d, J = 6.4 Hz, 3 H, CHCH3), 1.47 [s, 9 H, C(CH3)3], 1.41–1.20 (m,
2 H, CH2CH3), 0.91 (t, J = 7.3 Hz, 3 H, CH2CH3) ppm. 13C NMR
(100 MHz, CDCl3) (selected signals): δ = 169.08 (s, COO), 134.5
(d, CH=CH), 130.7 (d, CH=CH), 129.7 (d, CH=CH), 126.9 (d,
CH=CH), 56.15 (d, CHN), 47.4 (d, CHCH2), 24.1 (t, CH2CH3),
17.9 (q, CHCH3), 11.7 (q, CH2CH3) ppm. GC (Chirasil--Val,
100 °C, isothermic): tR(anti-12a)
89.49 min, tR(syn-12a) = 89.49 min, tR(syn-12a) = 93.23 min.
(؎)-tert-Butyl (4E,6E)-3-Ethyl-2-(trifluoroacetamido)-4,6-octa-
dienoate (12aЈ): Ester 12aЈ was obtained as one of the minor prod-
ucts in the allylic alkylation using 10a. 1H NMR (400 MHz,
CDCl3) (selected signals): δ = 6.39 (dd, J = 15.2, 11.2 Hz,
CH=CH), 5.90 (dd, J = 15.2, 15.2 Hz, CH=CH), 5.59–5.43 (m,
CH=CH), 5.34–5.28 (m, CH=CH), 2.22–2.14 (m, 2 H, CH2CH3)
= 85.79 min, tR(anti-12a) =
[7]
ppm. GC (Chirasil--Val, 100 °C, isothermic): tR(12Ј)
101.95 min, tR(12Ј) = 105.23 min.
=
(؎)-tert-Butyl
(4E)-3-[(1E)-1-Propenyl]-2-(trifluoroacetamido)-4-
[8]
heptenoate (13a): Ester 13a was obtained as one of the minor prod-
ucts in the allylic alkylation using 10a. 1H NMR (400 MHz,
CDCl3) (selected signals): δ = 6.78 (br. d, 1 H, NH), 5.61–5.44 (m,
1 H, CHCH3), 4.46 (dd, J = 8.4, 5.2 Hz, 1 H, CHN), 3.25–3.10 (m,
1 H, CHCH), 2.14–1.99 (m, 2 H, CH2CH3), 1.69 (d, J = 6.2 Hz, 3
H, CH CH3), 1.46 [s, 9 H, C(CH3)3] ppm. GC (Chirasil--Val,
100 °C, isothermic): tR(13a, diast. 1) = 59.37 min, tR(13a, diast.
2) = 60.48 min, tR(13a, diast. 1) = 61.23 min, tR(13a, diast. 2) =
62.13 min.
[9]
[10]
[11]
U. Kazmaier, D. Stolz, K. Krämer, F. Zumpe, Chem. Eur. J.
2007, 13, 6204–6211.
The syn/anti terminology is used to describe the orientation of
the substituent R relative to the hydrogen atom at the central
position of the allyl complex.
Supporting Information (see footnote on the first page of this arti-
cle): Experimental details and analytical data of all new com-
pounds.
[12]
a) P. R. Auburn, P. B. Mackenzie, B. Bosnich, J. Am. Chem.
Soc. 1985, 107, 2033–2046; b) P. B. Mackenzie, J. Whelan, B.
Bosnich, J. Am. Chem. Soc. 1985, 107, 2046–2054; c) P. Corrad-
ini, G. Maglio, A. Musco, G. Paiaro, Chem. Commun. 1996,
618–619; d) N. Solin, K. J. Szabo, Organometallics 2001, 20,
5464–5471; e) J. Yu, M. J. Gaunt, J. B. Spencer, J. Org. Chem.
2002, 67, 4627–4629; f) M. Ogasawara, K.-i. Takizawa, T. Hay-
ashi, Organometallics 2002, 21, 4853–4861.
Acknowledgments
Financial support by the Deutsche Forschungsgemeinschaft and
the Fonds der Chemischen Industrie is gratefully acknowledged. S.
B. thanks the Alexander von Humboldt Foundation for a postdoc-
toral fellowship.
[13]
[14]
a) B. Åkermark, S. Hansson, A. Vitagliano, J. Am. Chem. Soc.
1990, 112, 4587–4588; b) M. Sjögren, S. Hansson, P.-O.
Norrby, B. Åkermark, M. E. Cucciolito, A. Vitagliano, Orga-
nometallics 1992, 11, 3954–3964; c) P.-O. Norrby, B. Åkermark,
F. Hæffner, S. Hansson, M. Blomberg, J. Am. Chem. Soc. 1993,
115, 4859–4867.
a) B. M. Trost, E. Keinan, J. Org. Chem. 1979, 44, 3451–3457;
b) B. M. Trost, E. Keinan, J. Am. Chem. Soc. 1978, 100, 7779–
7781.
B. M. Trost, M. Lautens, M.-H. Hung, C. S. Carmichael, J.
Am. Chem. Soc. 1984, 106, 7641–7643.
B. M. Trost, S. Hildbrand, K. Dogra, J. Am. Chem. Soc. 1999,
121, 10416–10417.
a) R. Takeuchi, K. Tanabe, Angew. Chem. 2000, 112, 2051–
2054; b) G. Lipowsky, G. Helmchen, Chem. Commun. 2004,
116–117.
a) B. M. Trost, R. C. Bunt, Tetrahedron Lett. 1993, 34, 7513–
7616; b) B. M. Trost, R. C. Bunt, J. Am. Chem. Soc. 1998, 120,
70–79.
Y. I. M. Nilsson, P. G. Andersson, J.-E. Bäckvall, J. Am. Chem.
Soc. 1993, 115, 6609–6613.
W.-H. Zheng, N. Sun, K.-L. Hou, Org. Lett. 2005, 7, 5151–
5154.
[1] a) L. S. Hegedus, Organische Synthese mit Übergangsmetallen,
VCH, Weinheim, 1995; b) U. Kazmaier, M. Pohlman, “Cross
Coupling Reactions via π-Allyl Intermediates” in Metal-Cata-
lyzed C–C and C–N Coupling Reactions (Eds.: A. de Meijere,
F. Diederich), Wiley-VCH, Weinheim, 2004, pp. 531–583, and
references therein.
[2] a) G. Consiglio, R. M. Waymouth, Chem. Rev. 1989, 89, 257–
276; b) B. M. Trost, D. L. Van Vranken, Chem. Rev. 1996, 96,
395–422; c) G. Helmchen, J. Organomet. Chem. 1999, 576, 203–
214; d) A. Pfaltz, M. Lautens in Comprehensive Asymmetric
Catalysis I–III (Eds.: E. N. Jacobsen, A. Pfaltz, H. Yamamoto),
Springer Verlag, Berlin, 1999, pp. 833–884; e) L. Acemoglu,
J. M. J. Williams in Handbook of Organopalladium Chemistry
for Organic Synthesis, vol. 2 (Eds.: E.-I. Negishi, A. de Mei-
jere), John Wiley, New York, 2002, pp. 1689–1705; f) J. Tsuji in
Handbook of Organopalladium Chemistry for Organic Synthesis,
vol. 2 (Eds.: E.-I. Negishi, A. de Meijere), John Wiley, New
York, 2002, pp. 1669–1688; g) B. M. Trost, M. L. Crawley,
Chem. Rev. 2003, 103, 2921–2944; h) B. M. Trost, C. Jiang,
Org. Lett. 2003, 5, 1563–1565; i) B. M. Trost, J. Xu, J. Am.
Chem. Soc. 2005, 127, 2846–2847; j) B. M. Trost, M. K. Bren-
nan, Org. Lett. 2006, 8, 2027–2030.
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[3] a) G. Poli, G. Giambastiani, A. Heumann, Tetrahedron 2000,
56, 5959–5989; b) J. W. Faller, J. C. Wilt, J. Parr, Org. Lett.
2004, 6, 1301–1304.
[4] a) R. Prétôt, A. Pfaltz, Angew. Chem. 1998, 110, 337–339; An-
gew. Chem. Int. Ed. 1998, 37, 323–325; b) G. Helmchen, A.
Pfaltz, Acc. Chem. Res. 2000, 33, 336–345.
a) U. Kazmaier, F. L. Zumpe, Angew. Chem. 1999, 111, 1572–
1574; Angew. Chem. Int. Ed. 1999, 38, 1468–1470; b) U. Kaz-
maier, S. Maier, F. L. Zumpe, Synlett 2000, 1523–1535; c) T. D.
Weiss, G. Helmchen, U. Kazmaier, Chem. Commun. 2002,
1270–1271; d) U. Kazmaier, Curr. Org. Chem. 2003, 7, 317–
4176
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 4169–4177

Palladium-Catalyzed Dienylations of Chelated Enolates
328; e) U. Kazmaier, D. Stolz, Angew. Chem. 2006, 118, 3143–
3146; Angew. Chem. Int. Ed. 2006, 45, 3072–3075 and refer-
ences cited therein.
Lindner, U. Kazmaier, Adv. Synth. Catal. 2005, 347, 1687–
1695.
[24] U. Kazmaier, K. Krämer, J. Org. Chem. 2006, 71, 8950–8953.
[22] a) U. Kazmaier, F. L. Zumpe, Angew. Chem. 2000, 112, 805– [25] S. Basak, U. Kazmaier, Org. Lett. 2008, 10, 501–504.
807; Angew. Chem. Int. Ed. 2000, 39, 802–804; b) U. Kazmaier,
F. L. Zumpe, Eur. J. Org. Chem. 2001, 4067–4076; c) U. Kaz-
maier, M. Pohlman, Synlett 2004, 623–626; d) U. Kazmaier, M.
Pohlmann, T. Lindner, J. Org. Chem. 2004, 69, 6909–6912.
[23] a) U. Kazmaier, T. Lindner, Angew. Chem. 2005, 117, 3368–
3371; Angew. Chem. Int. Ed. 2005, 44, 3303–3306; b) T.
[26] a) J. J. Gajewski, J. Jurayj, D. R. Kimbrough, M. E. Gande, B.
Ganem, B. K. Carpenter, J. Am. Chem. Soc. 1987, 109, 1170–
1186; b) G. C. Lloyd-Jones, S. C. Stephen, Chem. Eur. J. 1998,
4, 2539–2549.
Received: May 28, 2008
Published Online: July 7, 2008
Eur. J. Org. Chem. 2008, 4169–4177
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4177