Effects of extended aryl-substituted bisoxazoline ligands in asymmetric synthesis - Efficient synthesis and application of 4,4′-bis(1-naphthyl)-, 4,4′-bis(2-naphthyl)- and 4,4′-bis(9-anthryl)-2,2′- isopropylidenebis(1,3-oxazolines)

Article abstract of DOI:10.1002/ejoc.200500566

The steric influence of extended aryl substituents on 2,2′-bis(1,3- oxazollne) ligands was investigated in a series of asymmetric catalytic reactions such as Mukaiyama aldol and Michael reactions, hetero-Diels-Alder processes, and allylic alkylation react

Full text of DOI:10.1002/ejoc.200500566

FULL PAPER
DOI: 10.1002/ejoc.200500566
Effects of Extended Aryl-Substituted Bisoxazoline Ligands in Asymmetric
Synthesis – Efficient Synthesis and Application of 4,4Ј-Bis(1-Naphthyl)-, 4,4Ј-
Bis(2-Naphthyl)- and 4,4Ј-Bis(9-Anthryl)-2,2Ј-isopropylidenebis(1,3-oxazolines)
Hester L. van Lingen,[a,c] Floris L. van Delft,[a] Roy P. M. Storcken,[a] Koen F. W. Hekking,[a]
Anouk Klaassen,[a] Jan J. M. Smits,[a] Patrycja Ruskowska,[b] Jadwiga Frelek,[b] and
Floris P. J. T. Rutjes*[a]
Keywords: Asymmetric catalysis / Ligand design / Bisoxazoline ligands / Non-proteinogenic amino acids / Chiral Lewis
acids
The steric influence of extended aryl substituents on 2,2Ј-
bis(1,3-oxazoline) ligands was investigated in a series of
asymmetric catalytic reactions such as Mukaiyama aldol and
Michael reactions, hetero-Diels–Alder processes, and allylic
alkylation reactions. 4,4Ј-(2-Naphthyl)- and 4,4Ј-(9-anthryl)-
substituted isopropylidene-bridged 2,2Ј-bis(1,3-oxazolines)
were synthesized and their enantioselective and catalytic
properties in combination with different metals evaluated.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2005)
Introduction
Bisoxazoline (box) ligands[1] in combination with transi-
tion metals have been shown to be excellent catalysts in a
wide range of asymmetric transformations such as Diels–
Alder[2] and hetero-Diels–Alder reactions[3] and aldol con-
densations.[4] So far, many variations at the core of the box
ligand have been carried out. Although a wide range of dif-
ferently substituted bisoxazoline ligands have been synthe-
sised over the years, only a few ligands are currently com-
monly used, which are commercially available or easily pre-
pared from the corresponding amino acids (A, Figure 1).
In case of alkyl substituents on C4, a common trend
shows that increase of steric bulk of the substituent en-
hances the enantioselectivity. Thus, generally higher
enantioselectivities are obtained with tBu-box than with
iPr- or Me-box. Very recently, the group of Moreno-Manas
accomplished the synthesis of a new and even bulkier 4-
alkyl-substituted bisoxazoline ligand, adamantyl-box
(Adam-box).[5] It was found that the use of this ligand re-
sulted in three different reactions in similar enantio-
selectivity as previously observed for tBu-box (in all in-
stances Ͼ95% ee). In addition, Evans and co-workers have
investigated the contribution of electronic effects of the R-
substituent, using a small series of para-substituted Ph-box
ligands (B).[3d] Although no clear electronic effect was ob-
Figure 1. Different types of bisoxazoline ligands.
served, possible dipole–dipole and Van der Waals attrac-
tions may still be partially responsible for the observed
stereoselectivity, since these interactions do not vary with
the π-donor capacity of the phenyl group.
Recently, Pagenkopf and co-workers investigated the
electronic and steric tuning of the aryl substituent by syn-
thesising a small series of o-alkyloxyaryl-substituted bisoxaz-
olines.[6] They employed these ligands in the addition of a
rather bulky silyl enol ether to methyl pyruvate and other
glyoxylate esters (vide infra). They reported a remarkable
increase in both yield and selectivity compared to the Ph-
box ligand, especially when the triflate counterion was re-
placed by chloride. However, their attempts to increase the
[a] Institute for Molecules and Materials, Radboud University
Nijmegen,
Toernooiveld 1, 6525 ED Nijmegen, The Netherlands
[b] Institute of Organic Chemistry, Polish Academy of Sciences,
Warsaw, Poland
[c] Current address: NV Organon,
P. O. Box 20, 5340 BH Oss, The Netherlands
Eur. J. Org. Chem. 2005, 4975–4987
© 2005 Wiley-VCH Verlag GmbH & Co. KgaA, Weinheim
4975

F. P. J. T. Rutjes et al.
FULL PAPER
enantioselectivity by increasing steric effects were unsuc-
cessful.
Despite the versatility of box ligands that have been syn-
thesised over the past decades, there is not a single ligand
that gives optimal results in every type of conversion. For
example, in some reactions alkyl-substituted box ligands
provide better results, whereas in other reactions the aryl-
substituted analogues are the most suitable ligands. Surpris-
ingly, the steric influence of extended aryl substituents has
only once been studied by Desimoni and Faita and they
observed that replacement of a phenyl by a 2-naphthyl sub-
stituent at the 4-position of the box ligand resulted in a
slight increase of enantioselectivity.[7]
Figure 2. Crystal structure of (S)-1-Np-box (1a).
A new class of more complex bisoxazoline ligands D,
bearing an additional ether substituent at the position 5 of
both oxazoline rings, was investigated by Pericas and
Muller.[8] In palladium-catalysed allylic alkylations, the
phenyl-substituted ligand gave similar results as obtained
with Ph-box. Extension of the phenyl substituent to a 1-
naphthyl group resulted in a similar excellent enantio-
selectivity and enhanced yield.
Recently, we developed a straightforward and efficient
route towards 4-substituted bisoxazoline ligands as depicted
in Scheme 1.[9] We have demonstrated the viability of this
route for the new (S)-1-naphthyl-substituted ligand (1a) and
observed some increase in enantioselectivity when we ap-
plied this ligand in a small range of Lewis acid-catalysed
reations. The synthesis of (S)-1-Np-box (1a) was ac-
complished in 60% overall yield starting from (S)-(1-naph-
thyl)glycine (2a) and in 37% overall yield from 1-naph-
thaldehyde (4a). The crystal structure of (S)-1a was recently
determined, thus confirming its structure unambiguously
(Figure 2).[10]
Figure 3. (1-Naphthyl)-, (2-naphthyl)-, (9-anthryl)bisoxazoline li-
gands, respectively.
Results and Discussion
Ligand Synthesis
We applied our previously established chemoenzymatic
pathway[9] to prepare (2-naphthyl)bisoxazoline (2-Np-box,
1b), despite the fact that a synthesis of this ligand was al-
ready accomplished by the group of Desimoni and Faita.[7]
The corresponding (2-naphthyl)glycine amide (3b) was
formed upon treatment of 2-naphthaldehyde (4b) with
modified Strecker conditions, followed by partial hydrolysis
of the resulting aminonitrile 5b to give Schiff base 6b
(Scheme 2). Acidic hydrolysis of the Schiff base resulted in
formation of the crystalline HCl salt of the desired amino
acid amide, which was then, after treatment with base, ex-
tracted from the basic water layer with dichloromethane to
give the free amino acid amide 3b in 70% overall yield from
2-naphthaldehyde.[11]
Subjection of racemic 3b·HCl to an -specific aminopep-
tidase from the microorganism Ochrobactrum anthropi
NCIMB 40321[12] (pH = 6.5, T = 50 °C) led to a mixture
Scheme 1. Retrosynthesis of (S)-1-Np-box.
In this contribution, we show the expansion of our set of the corresponding (R)-amide 3b and (S)-acid 2b
of ligands with the (2-naphthyl)- and the even more ex- (Scheme 3). The work-up consisted of removal of the whole
tended (9-anthryl)bisoxazoline (Figure 3) and applied these cell rests by filtration, followed by lowering of the pH to 2
ligands in a range of Lewis acid-catalysed reactions. We also in order to precipitate (S)-acid 2b (46% yield). Unfortu-
aimed to make a more extensive evaluation of the steric nately, we were unable to determine the enantiopurity at
influence of the aryl substituents and to investigate the this point due to the insolubility of the acid. Raising the
scope and limitations of these ligands.
pH to 9, filtration and subsequent extraction of the aque-
4976
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2005, 4975–4987

Aryl-Substituted 2,2Ј-Bis(1,3-oxazoline) Ligands in Asymmetric Synthesis
FULL PAPER
Initially, we aimed to synthesise the 9-anthryl-substituted
bisoxazoline, which has hitherto not been described in lit-
erature, via an identical approach to that shown above for
the two naphthyl-substituted bisoxazolines (Scheme 1).
However, from the start we encountered severe problems
when applying the same sequence to the exceptionally bulky
and lipophilic 9-anthraldehyde. Especially the insolubility
of the precursors and resulting products in the modified
Strecker reaction led to slow conversions and low yields.
Moreover, we envisaged that (9-anthryl)glycine amide might
lead to additional problems during the subsequent enzy-
matic resolution, since it was expected to be even less solu-
ble in the aqueous buffers than its naphthyl counterpart.
Therefore, we changed our strategy into a classical resolu-
tion approach to obtain optically pure 2-(9-anthryl)glycin-
ol. Starting from benzyl carbamate (9) and glyoxylic acid
monohydrate (10), the spontaneously formed N,O-acetal
was methylated in acidic methanol to give 11 in two smooth
steps according to the literature.[13] Next, this carbamate
was allowed to react with anthracene via the corresponding
N-acyliminium ion (Scheme 4).[14] Optimisation studies re-
vealed that propionic acid as a co-solvent was superior to
formic and acetic acid in this particular case. In addition,
the reaction time had to be restricted to one hour in order
to avoid decomposition of both anthracene and the electro-
philic aromatic substitution product 12 by sulfuric acid.
Scheme 2. Synthesis of (2-naphthyl)glycine amide.
ous layer afforded (R)-amide 3b in 43% yield with 80% ee.
Optically active (2-naphthyl)glycine (2b) gave access to (S)-
2-Np-box (1b) in the same fashion as described previously
(Scheme 3). Since (S)-(2-naphthyl)glycinol is a known com-
pound,[7] comparison of the optical rotation of the LiAlH4
reduction product 7b indicated that the enantiopurity of the
(S)-(2-naphthyl)glycine obtained by enzymatic resolution
was excellent (Ͼ95% ee). Base-catalysed cyclisation of di-
amide 8b resulted in the desired enantiopure 2-Np-box li-
gand, which could be readily purified via recrystallisation.
Scheme 3. Synthesis of (S)-(2-naphthyl)bisoxazoline (1b).
Eur. J. Org. Chem. 2005, 4975–4987
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4977

F. P. J. T. Rutjes et al.
FULL PAPER
Scheme 4. Synthesis of 2-(9-anthryl)glycinol.
Synthesis of the racemic HCl salt of amino alcohol 7c was (90% ee) and tR = 44.1 min (98% ee).[18] Assignment of the
accomplished via LiBH4 reduction[15] of methyl ester 12, absolute configuration of both enantiomers was achieved
subsequent deprotection of the amine in the presence of by circular dichroism (CD) measurements. In contrast with
LiOH and an acidic work-up (Scheme 4). Surprisingly, the known derivatisation methods to elucidate the absolute ste-
free amino alcohol was relatively unstable, but the decom- reochemistry of the 9-anthryl amino alcohol, we chose to
position could be significantly slowed down by conversion use a newly developed spectroscopic technique. Recently, it
into the corresponding HCl salt. This suggests that the in- was found that circular dichroism may provide a convenient
stability of HCl-free 7c is caused by the presence of a basic alternative for the direct determination of the absolute con-
nitrogen atom near the anthracene moiety. This is in agree- figuration of optically active amino alcohols.[19] In this ap-
ment with findings of Stack and co-workers, who observed proach, the amino alcohols are converted into cottonogenic
that in the synthesis of 9-aminomethylanthracene there are derivatives via in situ complexation with dirhodium tetra-
large variations in the yield,[16] a result that in our view acetate, which then also acts as the auxiliary chromophore.
could well be explained by a similar decomposition event.
The absolute stereochemistry was assigned based on the
Despite several attempts using the so-called “Dutch Res- Cotton effect arising in the CD spectra at around 620 nm;
olution” protocol,[17] involving treatment with families of this band probably originates from the π*(Rh–Rh) Ǟ
chiral resolving agents, no successful resolution of racemic σ*(Rh–O) transition. A negative Cotton effect would indi-
2-(9-anthryl)glycinol into both enantiomers was achieved. cate that the amino alcohol belongs to the -series, which
Therefore, the amino alcohol was resolved using preparative corresponds to the (S)-configuration and vice versa.[19] The
chiral HPLC (Chiralpak AD column) to afford both absolute configuration of both enantiomers of 2-(9-anthryl)-
enantiomers with excellent enantiopurity: tR = 39.3 min glycinol was determined according to this method: the en-
Scheme 5. Synthesis of (S)-(9-anthryl)bisoxazoline (1c).
4978
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2005, 4975–4987

Aryl-Substituted 2,2Ј-Bis(1,3-oxazoline) Ligands in Asymmetric Synthesis
FULL PAPER
Figure 4. CD spectra of (S)-2-(1-naphthyl)glycinol, (S)-2-(2-naphthyl)glycinol, (R)- and (S)-2-(9-anthryl)glycinol complexed to
[Rh(OAc)4]2.
antiomer with tR = 39.3 min (90% ee) was found to be (R)- rings was formed while the other alcohol remained unre-
2-(9-anthryl)glycinol and than with tR = 44.1 min (98% ee) acted. Alternatively, treatment with (diethylamino)sulfur
was found to be (S)-2-(9-anthryl)glycinol. To compare the trifluoride (DAST) led to smooth double cyclisation to give
Cotton effect of (S)- and (R)-2-(9-anthryl)glycinol with the desired (S)-9-Ant-box (1c) in 93% yield.[21] During sub-
other aryl-substituted amino alcohols, the spectra of (S)-1- sequent catalysis experiments, we noticed that this ligand is
naphthyl and (S)-2-naphthyl were also recorded in the pres- relatively unstable, similar to what was found for 2-(9-an-
ence of dirhodium tetraacetate (Figure 4).
thryl)glycinol.
The enantiomerically pure (S)-2-(9-anthryl)glycinol (7c)
was converted into the corresponding (S)-9-Ant-box (1c) in
two steps (Scheme 5). First, acylation of (S)-2-(9-anthryl)-
glycinol resulted in diamide (S)-8c in good yield. Under the
conditions reported by Evans,[20] i.e. in the presence of two
Catalysis Results
As a starting point, our aryl-substituted ligands were
equivalents of tosyl chloride, only one of the oxazoline- tested in the tandem oxa-Michael addition-Friedel–Craft
Eur. J. Org. Chem. 2005, 4975–4987
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4979

F. P. J. T. Rutjes et al.
FULL PAPER
alkylation reaction, in combination with Mg(OTf)2 in tolu- able enantiopurity (35% ee) was observed. A significant in-
ene at 0 °C and in the presence of 10 mol-% of N,N-di- crease to 45% ee was seen in the presence of 9-Ant-box in
methyl-p-toluidine (Table 1).[22] The 1-Np-box ligand CH2Cl2 (entry 5), despite partial decomposition of the li-
clearly resulted in the best enantioselectivity (74% ee), gand during catalysis. Nevertheless, the tBu-box ligand
whereas in the presence of both Ph- and 2-Np-box, the gives the best enantioselectivity in both solvents. Addition-
product was formed with remarkably better yield, but a ally, we observed that the stereochemical outcome of this
slight decrease in enantioselectivity (65% ee and 62% ee, reaction depends on the nature of the ligand used. The use
respectively). Despite the more bulkiness of the 9-Ant-box, of Ph- and 1-Np-box in THF led to a reversal of facial
the use of this ligand in combination with Mg(OTf)2 re- selectivity compared to tBu-box. Remarkably, 2-Np-box af-
sulted in a drastic decrease in enantioselectivity (20% ee). forded the hetero-Diels–Alder adduct with the same config-
This was due to the limited stability of the 9-Ant-box ligand uration as that obtained in the presence of tBu-box.
at 0 °C, i.e. the ligand decomposed during reaction (ca.
Table 2. Results of the hetero-Diels–Alder reaction.
20 h). Our results show that tBu-box is not an appropriate
ligand for this reaction, since tandem product 16 was ob-
tained with only 18% ee (entry 5). Surprisingly, the use of
Ph-box resulted in the same enantiomer of 16 as when tBu-
box was used, whereas all other aryl-substituted box ligands
yielded the product as the opposite enantiomer.
Table 1. Results of the tandem oxa-Michael addition-Friedel–Craft
alkylation reaction.
Entry Ligand
THF
CH2Cl2
yield [%][a] ee [%][b]
yield [%][a] ee [%][b]
1
2
3
4
5
tBu-box[c]
99
99
90
Ͼ95
–
Ͼ95 (S)
35 (R)
22 (R)
36 (S)
–
nd
nd
nd
–
Ͼ98 (S)
13 (S)
19 (S)
–
Ph-box[c]
1-Np-box
2-Np-box
9-Ant-box
82
45 (S)
[a] Isolated yields after chromatography. [b] Enantiomeric excesses
were determined by GC on a chiral γ-CD column (110 °C). [c] Our
results correspond well to literature data.[23]
Another type of reaction in which methyl pyruvate was
applied as the substrate in combination with [Cu(OTf)2box]
complexes is the Mukaiyama aldol reaction of methyl pyr-
uvate (17) with silyl ketene acetal 20.[9] Going from Ph-box
to 2-Np-box, the effect of the steric influence is clearly
shown by an increase of the ee, both in THF and in CH2Cl2
(Table 3, entries 2 and 4). In contrast, the use of 1-Np-box
in combination with Cu(OTf)2 only resulted in a decrease
of enantioselectivity compared to Ph-box (entries 2 and 3).
Entry
Ligand
Toluene
yield [%][a]
ee [%]
1
2
3
4
5
Ph-box
75
39
67
nd
77
65 (2R,4S)
74 (2S,4R)
62 (2S,4R)
20 (2S,4R)
18 (2R,4S)
1-Np-box
2-Np-box
9-Ant-box
tBu-box
[a] Isolated yields.
Furthermore, we applied the aryl-substituted ligands in The results show that 1-Np-box is not an appropriate ligand
the hetero-Diels–Alder reaction of methyl pyruvate with for this Mukaiyama aldol reaction. However, the use of 2-
DanishefskyЈs diene (18).[9] In both THF and CH2Cl2, we Np-box results an increase of enantioselectivity in THF
observed no significant differences between Ph-, 1-Np- and compared to Ph-box (31% vs. 24% ee) and a significant
2-Np-box (Table 2, entries 2–4). Replacement of Ph-box by increase in enantioselectivity in CH2Cl2 (60% vs. 46% ee).
1-Np-box resulted in an increase in asymmetric induction The highest ee (Ͼ94%) was observed in both solvents when
in CH2Cl2 (13% to 19% ee, respectively) and a decrease in tBu-box was used as ligand (entry 1). These results suggest
THF (35% to 22% ee, respectively). Comparing 2-Np-box that in both reactions discussed above, i.e. the Diels–Alder
to Ph-box in THF, a similar excellent yield and a compar- and the aldol reaction, the use of aryl-substituted box li-
4980
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2005, 4975–4987

Aryl-Substituted 2,2Ј-Bis(1,3-oxazoline) Ligands in Asymmetric Synthesis
FULL PAPER
gands does not have a crucial influence on the stereochemi- Table 4. Results of the allylation of dimethyl malonate.
cal outcome of the reaction.
Table 3. Results of the Mukaiyama aldol reaction.
Entry
Ligand
Yield [%][a]
ee [%][b]
1
2
3
4
tBu-box
0
–
Entry Ligand
THF
CH2Cl2
Ph-box[c]
1-Np-box
2-Np-box
Ͼ95
Ͼ95
Ͼ95
96 (R)
97 (R)
97 (R)
yield [%][a] ee [%][b] yield [%][a] ee [%][b]
1
2
3
4
tBu-box[c] 89
Ph-box[c]
73
1-Np-box nd
2-Np-box Ͼ95
97 (S)
24 (S)
19 (S)
31 (S)
Ͼ80
Ͼ80
75
94 (S)
46 (S)
3 (R)
[a] Isolated yields after chromatography. [b] Enantiomeric excesses
were determined by chiral HPLC analysis on a Chiralpak AD col-
umn (hexane/iPrOH, 90:10, 0.8 mL/min). [c] Our results corre-
spond well to literature data.[8]
Ͼ95
60 (S)
[a] Isolated yields after chromatography. [b] Enantiomeric excesses
were determined by chiral HPLC analysis on a Chiralcel OD col-
umn (hexane/IPA, 99:1, 0.5 mL/min) or by GC analysis using a
chiral β-CD column (150 °C). [c] Our results correspond well to
literature data.[28a]
Table 5. Results of the Mukaiyama–Michael reaction.
Subsequently, we applied our series of aryl-substituted
box ligands in the allylic substitution of (E)-1,3-diphenyl-2-
propenyl acetate, since Ph-box in combination with Pd is
known to be an excellent catalyst for this reaction.[8] Using
an in situ generated malonate nucleophile from dimethyl
malonate (23), bis(trimethylsilyl)acetamide (BSA) and a
catalytic amount of potassium acetate, the allylic substitu-
tion adduct 24 was formed with excellent enantioselectivity
(Table 4). Surprisingly, in the presence of tBu-box the reac-
tion did not proceed at all, whereas the use of Ph-, 1-Np-
and 2-Np-box afforded indeed the desired product in high
yield (Ͼ95%) and with excellent enantioselectivity (Ͼ95%
ee).
Previously, we observed a significant enhancement in
enantioselectivity in the Mukaiyama Michael reaction of
silyl ketene acetal 20 with Knoevenagel adduct 25 catalysed
by [Cu(SbF6)2box] complexes in the presence of 1,1,1,3,3,3-
hexafluoro-propan-2-ol (HFIP), depicted in Table 5.[9] In-
deed, we were pleased to find that catalysis by both 1-Np-
and 2-Np-box afforded the addition product with good
enantioselectivity (70% ee and 79% ee, respectively) being
a significant improvement compared to Ph-box (52% ee).
In this reaction, we also observed a reversal of enantiofacial
selectivity: in this case, all aryl-substituted box ligands af-
forded the opposite enantiomer than tBu-box. Again, the
highest enantioselectivity was achieved using tBu-box as the
Entry
Ligand
Yield [%][a]
ee [%][b]
1
2
3
4
tBu-box[c]
Ph-box[c]
1-Np-box
2-Np-box
99
67
55
nd
Ͼ90 (R)
56 (S)
70 (S)
79 (S)
[a] Isolated yields after chromatography. [b] Enantiomeric excesses
were determined by chiral HPLC analysis on a Chiralpak AD col-
umn (heptane/iPrOH, 98:2, 1.0 mL/min). [c] Our results correspond
ligand (Ͼ90% ee). From these results, it may be postulated well to literature data.[24]
Eur. J. Org. Chem. 2005, 4975–4987
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4981

F. P. J. T. Rutjes et al.
FULL PAPER
that the use of extended aryl-substituted bisoxazolines is
more promising in reactions, where the reaction centre is
relatively remote from the coordination site.
Conclusions
The (S)- and (R)-enantiomers of (2-naphthyl)glycine
While exploring the scope of aryl-substituted bisoxaz- were accessible through enzymatic resolution using an -
oline ligands, we observed a reversal of enantiofacial selec- specific aminopeptidase in good yield and excellent ee.
tivity between differently substituted box ligands. Other These synthetically versatile amino acids could also be ef-
groups have also addressed this phenomenon, which was fectively transformed into the corresponding enantiopure
first reported by the group of Jørgensen[25] and is usually bisoxazoline ligands. The (S)-9-Ant-box ligand could not
encountered in combination with copper(). Although be prepared in an analogous fashion, but instead was ob-
these groups have suggested a rationale over the years, so tained from the corresponding racemic amino alcohol,
far there is no agreement on a suitable explanation. Evans which was resolved into both enantiomers by chiral prepar-
suggested a mechanism in which π-stabilisation of the prod- ative HPLC. Subsequently, assignment of the absolute con-
uct-like transition state is responsible for the observed re- figuration of the anthryl-substituted amino alcohols was
versal of selectivity.[3d,26] However, Jørgensen and co- achieved by circular dichroism measurements in the pres-
workers addressed the observed effect to the more flexible ence of dirhodium tetraacetate.
character of the Ph-box compared to tBu-box complexes
Our studies towards the application of aryl-substituted
due to interconversion between square planar and tetrahe- box ligands afforded several interesting results. Both in the
dral conformations and thereby “flipping” of the oxazoline Mukaiyama aldol and in the hetero-Diels–Alder reaction,
substituent between pseudo-axial and pseudo-equatorial extension of the aryl-substituent did not lead to any dra-
positions.[27] These preliminary conclusions were derived matic improvement, although in the latter reaction, the use
from X-ray analyses and PM3-modelling experiments. Ad- of 9-Ant-box resulted in an enhancement of the enantio-
ditional EPR measurements unambiguously confirm the selectivity (45% ee compared to 13% ee for Ph-box). In the
dissimilarity in conformation of the CuII complexes of tBu- tandem reaction, 1-Np-box gave the best results (74% ee),
and Ph-box.[28]
while the Ph- and 2-Np-box also afford the tandem product
We also observed that the stereochemical outcome of the with good enantioselectivity (65% and 62% ee, respec-
reaction is strongly dependent on the ligand and that the tively). A second example in which the aryl-substituted li-
facial selectivity even changed by going from one aryl li- gands were superior, was the allylic substitution reaction.
gand to another. In order to acquire more insight into this While no reaction occurred using tBu-box, excellent yields
phenomenon, we performed several experimental and theo- (Ͼ95%) and enantioselectivities (Ͼ95%) were observed
retical investigations on the geometry of our [CuIIbox] com- using Ph-, 1-Np- and 2-Np-box. A significant effect of en-
plexes. In Figure 5, the crystal structures of the [CuCl2(1- largement of the aryl substituent on the enantioselectivity
Np-box)] and the [CuCl2(2-Np-box)] are depicted, which was observed in the Mukaiyama Michael reaction, in which
are also quite different from each other. Comparing their both 1-Np and 2-Np-box resulted in a clear improvement
geometry with those of the corresponding tBu- and Ph-box of the enantiopurity. This result once more demonstrates
complexes,[29] we observed that the [CuCl2(1-Np-box)] com- the beneficial influence of extension of the aryl substituent
plex strongly resembles the [CuCl2(tBu-box)] complex, in reactions where the distance between reaction centre and
whereas the [CuCl2(2-Np-box)] complex is more similar to coordination site is relatively large.
the [CuCl2(Ph-box)] complex. These different conforma-
Despite the use of X-ray crystallography to acquire more
tions could to some extent explain the occasional reversal details about the exact geometry of our [CuIIbox] com-
in facial selectivity in our experiments, but in order to be plexes, we could not find a conclusive explanation for the
able to give a conclusive explanation more detailed studies strong dependency of the stereochemical outcome of the
will have to be carried out.
reaction on a particular ligand.
Figure 5. X-ray structures of [CuCl2(1-Np-box)] (left) and [CuCl2(2-Np-box)] (right).
4982
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2005, 4975–4987

Aryl-Substituted 2,2Ј-Bis(1,3-oxazoline) Ligands in Asymmetric Synthesis
FULL PAPER
13CNMR (50 MHz, CDCl3, 25 °C): δ = 173.7, 163.2, 136.4, 135.2,
133.1, 132.8, 131.3, 128.5, 128.3, 128.3, 127.8, 127.4, 126.2, 126.0,
125.9, 124.7, 76.9 ppm. HRMS(EI) calcd. for C19H16N2O
288.1263, found 288.1256. Rf (Et2O/heptane, 1:1) = 0.12.
Experimental Section
General Remarks: All reactions were carried out under an atmo-
sphere of dry nitrogen or argon. Standard syringe techniques were
applied for the transfer of dry solvents and air- or moisture-sensi-
tive reagents. Flash chromatography was performed with Acros Or-
ganic silica gel (0.035–0.070 mm). Rf values are obtained using thin
layer chromatography (TLC) on silica gel-coated plates (Merck 60
(2-Naphthyl)glycine Amide (3b): To a solution of 6b (72.9 g,
0.25 mol) in acetone (3.00 L) was added concentrated HCl
(24.7 mL, 0.24 mol). After stirring for 1 h at room temp., a precipi-
tate was formed, which was collected by filtration, washed with
acetone and dried in vacuo. The amide 6b was isolated as its HCl
salt (61.7 g, 88%): 1HNMR ([D6]DMSO, 200 MHz): δ = 7.99–7.60
(m, 7 H), 5.02 (s, 1 H). Extraction of an aqueous solution of the
HCl salt (pH, 10, 500 mL) with CH2Cl2 (3×200 mL) afforded the
HCl-free product as an off-white, crystalline solid (41.2 g, 86%
F254) with the indicated solvents. Melting points were determined
using a Büchi melting point B-545 apparatus and are uncorrected.
IR spectra were recorded on an ATI Mattson Genesis Series FTIR
spectrometer. NMR spectra were recorded with Bruker DMX 200
and DMX300 spectrometers and with a Varian Inova 400 MHz
spectrometer with TMS as internal standard. Optical rotations
were determined with a Perkin–Elmer 241 polarimeter. Mass spec-
tra were measured with a Fisons (VG) Micromass 7070E apparatus.
Elemental analyses were carried out by using the CHNS-O EA
1108 element analyser from Carlo–Erba Instruments. CD spectra
were recorded in the 250–800 nm range, at room temperature, with
a Jasco 715 spectropolarimeter using chloroform solutions in cells
of 2, 10 or 20 mm path length (spectral band with 2 nm, sensitivity
10×10–6 or 20×10–6 ∆A-unit/nm). Depending on the S/N ratio the
λ-scan speed was 0.2 or 0.5 nm/s. For CD standard measurements
the solid chiral amino alcohol (1–5 mg, ca. 0.003 ) was dissolved
in a stock solution of the dirhodium tetraacetate (6–7 mg, ca.
0.002 ) in chloroform so that the molar ratio of the stock complex
to ligand was about 1:1.5. HPLCЈs were run using a Shimadzu
CLASS-VP instrument, equipped with a Chiralpak AD column
and a Chiralcel OD column. For chiral GC, a Hewlett–Packard
6890 instrument was used, equipped with a GammaDEX 120TM
fused silica capillary column of Supelco (30 m, 0.25 mm ID,
0.25 µm film thickness) with H2 as carrier gas.
from 3b): m.p. 163 °C. IR: ν = 3366, 3297, 3053, 1660, 1641, 1576,
˜
1405, 1294 cm–1. 1HNMR (300 MHz, CD3CN, 25 °C): δ = 7.94–
7.78 (m, 4 H), 7.58–7.42 (m, 3 H), 6.89 (br. s, 1 H), 5.81 (br. s, 1
H), 4.56 (s, 1 H) ppm. 13CNMR (75 MHz, CD3CN, 25 °C): δ =
129.2, 128.8, 128.6, 127.3, 127.0, 126.8, 126.2, 118.4, 61.0 ppm.
HRMS(EI) calcd. for C12H12N2O 200.0950, found 200.0945. Rf
(Et2O/heptane, 1:1) = 0.06.
Enzymatic Synthesis of (S)-(2-Naphthyl)glycine [(S)-2b] and (R)-(2-
Naphthyl)glycine Amide [(R)-3b]: A solution of 3b (78.8 g, 0.39 mol)
in H2O (3.00 L) was brought to pH 6.5 using aqueous KOH. After
addition of aqueous ZnSO4 (1.13 g, 3.9 mmol) and a cell-free ex-
tract containing the -amidase from Ochrobactrum anthropi
NCIMB 40321 in water (20 mL), water was added to attain a 2%
suspension of the amide. The mixture was shaken for 18 h 30 at
50 °C. After cooling to room temp., the mixture was brought to
pH 2.0 using aqueous H2SO4. The mixture was filtered to afford
acid (S)-2b as a white solid (44.8 g including approximately 20%
salts, 46%): since the amino acid was not soluble in any common
solvent (D2O, CD3CN, CD3OD, [D6]DMSO), at this stage no spec-
troscopic data could be recorded. HRMS(EI) calcd. for
C12H11N2O 201.0790, found 201.0783. The resulting clear filtrate
was brought to pH 9 by addition of aqueous KOH (5 ), so that a
precipitate was formed, which was collected by filtration, affording
amide (R)-3b (25.8 g) as a white solid. The aqueous layer was then
concentrated in vacuo to a volume of 1.5 L (pH, 9) and extracted
in portions of 400 mL with CH2Cl2 (3×200 mL). The organic layer
was dried (MgSO4) and concentrated in vacuo, affording amide
(R)-3b as a white solid (7.2 g). The combined batches afforded
amide (R)-3b as a white solid (33.0 g, 43%): the ee was determined
by 1H NMR using (S,S)-DBTAAN[9] to be 80%: 1HNMR
(300 MHz, CD3CN, 25 °C): δ = 7.94–7.78 (m, 4 H), 7.58–7.42 (m,
3 H), 6.89 (br. s, 1 H), 5.81 (br. s, 1 H), 4.56 (s, 1 H) ppm.
Materials: Solvents were distilled from appropriate drying agents
immediately prior to use. Unless stated otherwise, all chemicals
were purchased and used as such. 2-Naphthaldehyde was distilled
before use. (S,S)-DBTAAN, and a cell-free extract containing the
Ochrobactrum anthropi NCIMB 40321 -amidase were kindly pro-
vided by DSM. Dowex 50W×4 (Fluka) was used for ion exchange
chromatography with 2  NH4OH as the eluent and 2  HCl as
regeneration agent. Aryl-substituted β,γ-unsaturated α-keto esters
14 was kindly provided by prof. Jørgensen.
Crystal Structure Determination of (S)-1-Np-box [(S)-1a]:[10]
C29H26N2O2, M = 434.52, crystallises in the monoclinic space
group P21, with a = 11.2514(17) Å, b = 7.7746(13) Å, c =
14.283(3) Å, V = 1148.8(3) Å3, T = 293(2) K, Z = 2, µ =
0.079 mm–1; 22993 reflections were measured, giving 24020 inde-
pendent, of which 2986 with I Ͼ 2σ(I) were used in the refinements.
The internal agreement had R = 0.0731, the final R = 0.0450, Rw
= 0.0782, GOF = 1.035, for the significant reflections.
(S)-2-(2-Naphthyl)glycinol [(S)-7b]: To a suspension of acid (S)-2b
(5.00 g, 25.0 mmol) in THF (350 mL) at 0 °C was added LiAlH4
(10.2 g, 268 mmol) in portions. After refluxing for 18 h, the mixture
was cooled to 0 °C and quenched through addition of H2O (9 mL),
aqueous NaOH (4 , 9 mL) and H2O (20 mL) over a period of
15 min. After stirring for 30 min at room temp., the mixture was
filtered. The residue was washed with EtOAc (3×225 mL) and the
combined filtrates were concentrated in vacuo to yield amino
alcohol (S)-7b as a yellow foam (3.72 g, 81%): [α]2D5 = +39 (c =
N-Benzylidene-(2-naphthyl)glycine Amide (6b): To a solution of
NaCN (15.5 g, 0.32 mol) in methanol (250 mL) was added concen-
trated ammonia (25% in H2O, 225 mL), NH4Cl (16.9 g, 0.32 mol)
and 2-naphthaldehyde (49.2 g, 0.32 mol). After stirring overnight
at room temp., the reaction was complete according to TLC (Et2O/
heptane, 1:1). To the reaction mixture was added toluene (300 mL),
aqueous NaOH (1 , 315 mL, 0.32 mol) and benzaldehyde
(32.0 mL, 0.31 mol). After stirring for 1 h, a precipitate was
formed, which was collected by filtration, washed with heptane and
dried in vacuo. Benzaldimine 6b was isolated as an off-white solid
(72.9 g, 81% from 2-naphthaldehyde) and used directly for the next
0.59, CHCl3) [ref.[7] for (S)-7b [α]2D5 = +33.9 (c = 1.39, CHCl3)]. IR:
1
ν = 3360–2863, 2358, 2336, 1597, 1048 cm–1. HNMR (300 MHz,
˜
CDCl3, 25 °C): δ = 7.87–7.72 (m, 4 H), 7.52–7.40 (m, 3 H), 4.20
3
2
(dd, 3J(H,H) = 4.5, 8.1 Hz, 1 H), 3.82 (dd, JH,H = 4.5, JH,H
=
3
2
10.5 Hz, 1 H), 3.63 (dd, JH,H = 8.1, JH,H = 10.5 Hz, 1 H) ppm.
13CNMR (75 MHz, CD3OD, 25 °C): δ = 139.4, 133.0, 132.6, 128.0,
127.5, 127.4, 125.9, 125.6, 124.9, 124.5, 67.7, 57.4 ppm.
step: m.p. 148 °C. IR: ν = 3443, 3286–3058, 1684, 1640, 1580,
˜
1359 cm–1. HNMR (200 MHz, CDCl3, 25 °C): δ = 8.37 (s, 1 H), N,NЈ-Bis[(S)-2-hydroxy-1-(2-naphthyl)ethyl]-2,2-dimethylpropanedi-
1
7.82–7.48 (m, 12 H), 7.13 (s, 1 H), 5.63 (s, 1 H), 5.17 (s, 1 H) ppm. amide [(S)-8b]: A solution of amino alcohol (S)-7b (5.75 g,
Eur. J. Org. Chem. 2005, 4975–4987
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4983

F. P. J. T. Rutjes et al.
30.8 mmol) and Et3N (22 mL, 159 mmol) in CH2Cl2 (200 mL) was C25H21NO4 399.1471, found 399.1471. Rf (EtOAc/heptane, 2:1) =
FULL PAPER
cooled to 0 °C. A solution of 2,2-dimethylmalonyl dichloride[20a]
(2.69 g, 15.8 mmol) in CH2Cl2 (70 mL) was added via a cannula.
After stirring for 16 h, CH2Cl2 (200 mL) was added and the mix-
ture was washed with aqueous HCl (1 , 300 mL). The aqueous
layer was extracted with CH2Cl2 (3×200 mL). The combined or-
ganic layers were washed with saturated aqueous NaHCO3
(300 mL) and the aqueous layer was extracted with CH2Cl2
(200 mL). The combined organic layers were washed with brine
(300 mL) and the aqueous layer was extracted with CH2Cl2
(200 mL). The combined organic layers were dried (MgSO4) and
concentrated in vacuo. The resulting brown solid was recrystallised
from EtOAc, yielding diamide (S)-8b as an off-white solid (3.74 g,
52%): m.p. 160 °C. [α]2D5 = +122 (c = 0.55, CHCl3). [α]2D5 = +99 (c
= 0.20, 96% EtOH) [ref.[7] for (S)-8b [α]2D5 = +85.0 (c = 0.20, 95%
0.45.
2-(9-Anthryl)-2-[(benzyloxycarbonyl)amino]ethanol (13): To a solu-
tion of methyl ester 12 (0.70 g, 1.76 mmol) in THF (20 mL) at 0 °C
was added LiBH4 (2.0  in THF, 1.7 mL, 3.4 mmol). After stirring
for 18 h at reflux, the reaction was quenched with EtOAc (2 mL)
and saturated aqueous NaHCO3 (3 mL). After addition of brine
(30 mL), the reaction mixture was extracted with CH2Cl2
(3×30 mL). The organic layer was dried (MgSO4) and concen-
trated in vacuo. The off-white solid obtained was crystallised from
CH2Cl2/hexane, yielding CBz-protected amino alcohol 13 as a
white, crystalline solid (0.59 g, 90%): m.p. 134 °C. IR: ν = 3322,
˜
3057, 1695, 1533, 1525, 1251, 1063, 1032 cm–1. 1HNMR (400 MHz,
3
CDCl3, 25 °C): δ = 8.46 (s, 1 H), 8.38 (d, JH,H = 8.9 Hz, 2 H),
8.07–8.02 (m, 2 H), 7.58–7.52 (m, 2 H), 7.52–7.44 (m, 2 H), 7.38–
7.27 (m, 5 H), 6.38 (dt, JH,H = 4.7, 8.2 Hz, 1 H), 5.76 (d, JH,H
4.7 Hz, 1 H), 5.21–5.07 (m, 2 H), 4.71–4.59 (m, 1 H), 4.15–4.07 (m,
1 H), 3.49 (br. s, 1 H) ppm. 13CNMR (75 MHz, CDCl3, 25 °C): δ
= 155.2, 135.7, 131.5, 129.5, 128.7, 128.4, 128.3, 128.1, 127.9, 126.5,
125.1, 124.8, 123.8, 67.5, 67.2, 54.8 ppm. HRMS(EI) calcd. for
C24H21NO3 371.1521, found 371.1521. Rf (EtOAc/MeOH, 9:1) =
0.69.
EtOH)]. IR: ν = 3378, 2976, 2868, 2487, 1632, 1411, 1044 cm–1.
˜
3
3
=
1HNMR (300 MHz, CD3OD, 25 °C): δ = 7.74–7.61 (m, 8 H), 7.44–
3
3
7.32 (m, 6 H), 5.22 (dd, JH,H = 4.8, 7.5 Hz, 2 H), 3.91 (dd, JH,H
= 5.1, 2JH,H = 11.4 Hz, 2 H), 3.84 (dd, 3JH,H = 7.5, 2JH,H = 11.4 Hz,
2 H), 1.55 (s, 6 H) ppm. 13CNMR (75 MHz, CD3OD, 25 °C): δ =
175.4, 137.9, 134.4, 133.9, 128.9, 128.6, 128.3, 126.8, 126.5, 126.3,
125.6, 65.8, 57.3, 51.6, 24.4 ppm.
2,2-Bis[(S)-4-(2-naphthyl)-1,3-oxazolin-2-yl]propane [(S)-1b]: To a
solution of (S)-8b (1.02 g, 2.20 mmol), DMAP (29 mg, 0.24 mmol)
and triethylamine (1.52 mL, 10.9 mmol) in CH2Cl2 (55 mL) was
added a solution of tosyl chloride (1.00 g, 5.25 mmol) in CH2Cl2
(25 mL) via cannula. After stirring at room temp. for 90 min, the
mixture was washed with saturated aqueous NH4Cl (85 mL) and
the aqueous layer was extracted with CH2Cl2 (3×85 mL). The
combined organic layers were washed with saturated aqueous
NaHCO3 (170 mL), the aqueous layer was extracted with CH2Cl2
(3×130 mL). The combined organic layers were dried (MgSO4)
and concentrated in vacuo to yield the crude product as an off-
white solid. Purification by column chromatography (EtOAc/hep-
tane, 4:6) resulted in isolation of (S)-1b as a white solid (0.53 g,
55%). The product was purified by recrystallisation from EtOAc:
m.p. 172 °C [ref.[7] for (S)-1b m.p. 173–174 °C]. [α]2D5 = –246 (c =
0.55, CHCl3) [ref.[7] for (S)-1b [α]2D5 = –233 (c = 0.55, CHCl3)]. IR:
2-(9-Anthryl)glycinol·HCl (7c·HCl): To a solution of CBz-protected
amino alcohol 13 (50 mg, 0.14 mmol) in EtOH (1.5 mL) was added
aqueous LiOH (2.0 , 1.5 mL, 3.0 mmol). After stirring for 2 h at
reflux, the reaction was quenched with saturated aqueous NH4Cl
(5 mL). The reaction mixture was extracted with EtOAc
(2×10 mL). The combined organic layers were dried (MgSO4) and
concentrated in vacuo, yielding amino alcohol rac-7c as an off-
white salt (33 mg, quant): IR: ν = 3380–2840, 2244, 1942, 1618,
˜
1
1447, 1038 cm–1. HNMR (300 MHz, CDCl3, 25 °C): δ = 8.46 (d,
3JH,H = 7.8 Hz, 2 H), 8.28 (s, 1 H), 7.95–7.86 (m, 2 H), 7.44–7.19
(m, 4 H), 5.51 (dd, JH,H = 5.0, 9.5 Hz, 1 H), 4.30 (t, JH,H
10.3 Hz, 1 H), 3.84 (dd, JH,H = 5.0, JH,H = 10.9 Hz, 1 H) 2.81
(br. s, 3 H) ppm. 13CNMR (75 MHz, CDCl3, 25 °C): δ = 131.4,
129.3, 128.2, 127.8, 127.2, 126.7, 125.5, 124.5, 66.0, 53.0 ppm. A
solution of rac-7c in EtOAc was added to aqueous HCl (2 ) to
form the corresponding HCl salt rac-7c·HCl as an off-white solid
3
2
=
3
2
ν = 3049, 2977, 2939, 2897, 1653 cm–1. 1HNMR (300 MHz, CDCl ,
˜
3
(38 mg, 80%): m.p. 186 °C. IR: νmax = = 3399, 3205, 2950, 2894,
˜
3
25 °C): δ = 7.82–7.66 (m, 8 H), 7.47–7.31 (m, 6 H), 5.41 (dd, JH,H
1627, 1519, 1057 cm–1. 1HNMR (300 MHz, CD3OD, 25 °C): δ =
8.63 (s, 1 H), 8.38 (d, J = 9.0 Hz, 2 H), 8.12 (d, J = 8.4 Hz, 2 H),
7.72–7.63 (m, 2 H), 7.58–7.50 (m, 2 H), 5.95 (dd, J = 4.8, 9.6 Hz,
1 H), 4.64 (dd, J = 9.9, 11.7 Hz, 1 H), 4.09 (dd, J = 4.8, 12.0 Hz,
1 H) ppm. 13CNMR (75 MHz, CD3OD, 25 °C): δ = 132.7, 131.3,
131.2, 130.8, 128.3, 126.0, 125.4, 123.7, 63.3, 54.2 ppm. HRMS(EI)
calcd. for C16H15NO 237.1154, found 237.1155. Rf (EtOAc/hep-
tane, 3:1) = 0.2–0.5. Separation of the enantiomers was ac-
complished by preparative chiral HPLC analysis (chiralpak AD
column; heptane with 0.5% EtOH and 0.2% Et3N): the enantio-
meric excess was determined by HPLC on a Chiralpak AD column
2
3
2
= 7.5, JH,H = 9.9 Hz, 2 H), 4.74 (dd, JH,H = 8.1, JH,H = 9.9 Hz,
2 H), 4.26 (t, JH,H = 7.8 Hz, 2 H), 1.75 (s, 6 H) ppm. 13CNMR
3
(75 MHz, CDCl3, 25 °C): δ = 170.2, 139.4, 133.1, 132.7, 128.5,
127.7, 127.4, 125.9, 125.6, 125.4, 124.4, 75.4, 69.8, 39.2, 24.7 ppm.
HRMS(EI) calcd. for C29H26N2O2 434.1994, found 434.1988.
Methyl 2-(9-Anthryl)-2-(benzyloxycarbonylamino)acetate (12): To a
solution of 11[13] (0.93 g, 3.7 mmol) and anthracene (0.65 g,
3.6 mmol) in propionic acid (18 mL) at 0 °C was added concen-
trated sulfuric acid (2.0 mL, 10% v/v) dropwise and stirred for 1 h.
The slightly yellow solution was poured into an ice/saturated aque-
ous NaHCO3 mixture (100 mL) and extracted with CH2Cl2
(3×85 mL). The organic layer was dried (MgSO4) and concen-
trated in vacuo. Purification by column chromatography (EtOAc/
heptane, 1:5) resulted in isolation of methyl ester 12 as an off-white
(heptane with 5% EtOH and 0.2% EtNH2, 0.5 mL/min): tR
=
39.3 min (R)-7c·HCl (90% ee): [α]2D5 = +27 (c = 0.28, MeOH); tR
= 44.1 min (S)-7c·HCl (98% ee): [α]2D5 = –28 (c = 0.25, MeOH).
N,NЈ-Bis[(S)-2-hydroxy-1-(9-anthryl)ethyl]-2,2-dimethylpropanedi-
amide [(S)-8c]: To a solution of amino alcohol (S)-7c·HCl (40 mg,
solid (1.0 g, 72%): m.p. 160 °C. IR: ν = 3369, 3051, 2958, 1728,
˜
1
1714, 1518, 1306, 1221 cm–1. HNMR (300 MHz, CDCl3, 25 °C): 0.17 mmol) and Et3N (59 µL, 0.42 mmol) in CH2Cl2 (2 mL) at 0 °C
3
δ = 8.50 (s, 1 H), 8.37–8.24 (m, 2 H), 8.05 (d, JH,H = 8.4 Hz, 2 was added 2,2-dimethylmalonyl dichloride[20a] (14 mg, 84 µmol).
H), 7.63–7.52 (m, 2 H), 7.52–7.44 (m, 2 H), 7.35–7.29 (br. s, 5 H),
6.99–6.93 (m, 1 H), 6.04–5.97 (m, 1 H), 5.19–5.03 (m, 2 H), 3.64
After stirring for 16 h, the reaction mixture was quenched with
H2O (1 mL), CH2Cl2 (10 mL) was added and the mixture was
(s, 3 H) ppm. 13CNMR (75 MHz, CDCl3, 25 °C): δ = 172.1, 155.4, washed with aqueous HCl (1 , 5 mL). The aqueous layer was ex-
135.4, 131.1, 129.3, 129.0, 128.7, 128.0, 127.7, 127.6, 127.2, 126.7,
124.6, 122.8, 67.0, 52.7, 52.1 ppm. HRMS(EI) calcd. for
tracted with CH2Cl2 (5 mL). The combined organic layers were
washed with H2O (10 mL) and the aqueous layer was extracted
4984
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2005, 4975–4987

Aryl-Substituted 2,2Ј-Bis(1,3-oxazoline) Ligands in Asymmetric Synthesis
FULL PAPER
with CH2Cl2 (5 mL). The combined organic layers were washed
with saturated aqueous NaHCO3 (10 mL) and the aqueous layer
was extracted with CH2Cl2 (5 mL). The combined organic layers
were dried (MgSO4) and concentrated in vacuo. Purification by
column chromatography (acetone/CH2Cl2, 1:4 to 1:2) resulted in
isolation of (S)-8c as an off-white solid (74 mg, 77%): [α]2D5 = –55
(17) (25 µL, 0.28 mmol) was added and the mixture was cooled to
–78 °C. After addition of Danishefsky’s diene 18 (72 µL,
0.34 mmol), the mixture was warmed up to room temp. overnight.
Et2O (10 mL) was added and the mixture was washed using satu-
rated aqueous NaHCO3 (10 mL), H2O (10 mL) and brine (10 mL).
The organic layer was dried (MgSO4) and concentrated in vacuo,
(c = 1.6, MeOH). IR: νmax = = 3500–3100, 2924, 2250, 1658, 1502, affording 19 as a slightly yellow oil (47 mg, 99%): the ee was deter-
˜
1260, 1031 cm–1. HNMR (300 MHz, CDCl3, 25 °C): δ = 8.39 (s,
mined by GC to be Ͼ95%: chiral γ-CD column (110 °C), tR
1
3
2 H), 8.32 (d, J = 9.0 Hz, 4 H), 7.97 (d, JH,H = 8.7 Hz, 4 H), 7.75 (minor) = 15.7 min (R), tR(major) = 16.4 min (S). 1HNMR
3
3
(d, JH,H = 5.7 Hz, 2 H), 7.57–7.36 (m, 6 H), 6.56–6.47 (m, 2 H),
(300 MHz, CDCl3, 25 °C): δ = 7.54 (d, JH,H = 12.6 Hz, 1 H), 5.58
3
2
3
2
4.61 (dd, JH,H = 8.7, JH,H = 11.7 Hz, 2 H), 4.26 (br. s, 2 H), 4.05 (d, JH,H = 12.6 Hz, 1 H), 3.68 (s, 3 H), 3.00 (d, JH,H = 16.5 Hz,
3
2
2
(dd, JH,H = 3.9, JH,H = 11.7 Hz, 2 H), 1.38 (s, 6 H) ppm.
13CNMR (75 MHz, CDCl3, 25 °C): δ = 131.5, 129.5, 129.0, 128.7,
126.5, 124.8, 123.5, 66.8, 54.1, 49.9, 23.8 ppm. Rf (EtOAc/heptane,
2:1) = 0.29.
1 H), 2.68 (d, JH,H = 16.5 Hz, 1 H), 1.49 (s, 3 H) ppm.
Representative Experimental Procedure for the Mukaiyama Aldol
Reaction to Thioacetate 21: To a flame-dried Schlenk tube was
added Cu(OTf)2 (10 mg, 28 µmol) and (S)-tBu-box (9.0 mg, 31
µmol). The mixture was dried under vacuum for 0.5 h and THF
(1.0 mL) was added. After stirring for 1 h, methyl pyruvate (17,
25 µL, 0.28 mmol) was added and the mixture was cooled to
–78 °C. After addition of silyl ketene acetal 20[30] (83 µL,
0.34 mmol), the mixture was warmed up to room temp. overnight.
Et2O (10 mL) was added and the mixture was washed using satu-
rated aqueous NaHCO3 (10 mL), H2O (10 mL) and brine (10 mL).
The organic layer was dried (MgSO4) and concentrated in vacuo.
Flash chromatography (EtOAc/heptane, 1:9 to 1:4) afforded 21 as
a colourless oil (58 mg, 89%): the ee was determined by HPLC to
be 97%: OD column; hexane/iPrOH, 99:1; 0.5 mL/min; tR(minor)
2,2-Bis[(S)-4-(9-anthryl)-1,3-oxazolin-2-yl]propane [(S)-1c]: To
a
solution of diamide (S)-8c (32 mg, 56 µmol) in CH2Cl2 (1 mL) at
–78 °C was added (diethylamino)sulfur trifluoride (22 µL,
0.17 mmol). After stirring at room temp. for 10 min, CH2Cl2
(10 mL) was added, the mixture was washed with saturated aque-
ous NaHCO3 (5 mL) and the aqueous layer was extracted with
CH2Cl2 (5 mL). The combined organic layers were washed with
saturated aqueous H2O (5 mL), the aqueous layer was extracted
with CH2Cl2 (5 mL). The combined organic layers were dried
(MgSO4) and concentrated in vacuo to yield the crude product as
an off-white solid. Purification by column chromatography
(EtOAc/heptane, 1:3 to 1:2) resulted in isolation of (S)-1c as a white
= 14.9 min (R), tR(major) = 16.0 min (S). 1HNMR (200 MHz,
2
CDCl3, 25 °C): δ = 3.80 (s, 3 H), 3.72 (br. s, 1 H), 3.08 (d, JH,H
=
solid (28 mg, 93%), which was directly used for catalysis: IR:
2
15.9 Hz, 1 H), 2.83 (d, JH,H = 15.8 Hz, 1 H), 1.45 (s, 9 H), 1.41
1
ν
˜
max = = 3500–3050, 3047, 2982, 2921, 1649, 1145 cm–1. HNMR
(s, 3 H) ppm.
(300 MHz, CDCl3, 25 °C): δ = 8.42 (s, 2 H), 8.37 (d, 3JH,H = 9.9 Hz,
2
Representative Experimental Procedure for the Allylic Substitution
Reaction to Acetate 24: To a flame-dried Schlenk tube was added
[Pd(η3-C3H5)Cl]2 (4.8 mg, 13 µmol) and (S)-Ph-box (13 mg, 39
µmol). The mixture was dried under vacuum for 0.5 h and distilled
anhydrous CH2Cl2 (1.0 mL) was added. After stirring for 1 h, ace-
tate 22[31] (101 mg, 0.48 mmol), malonate 23 (0.14 mL, 1.2 mmol),
bis(trimethylsilyl)acetamide (BSA) (0.30 mL, 1.8 mmol) and KOAc
(2 mg, 20 µmol) were added and the mixture was stirred overnight.
Et2O (10 mL) was added and the mixture was washed using satu-
rated aqueous NH4Cl (10 mL), H2O (10 mL) and brine (10 mL).
The organic layer was dried (MgSO4) and concentrated in vacuo.
Flash chromatography (EtOAc/heptane, 1:15) afforded 24 as a
colourless oil (ca, 150 mg, Ͼ95%): the ee was determined by HPLC
to be 96%: AD column; hexane/iPrOH, 90:10; 0.8 mL/min;
tR(minor) = 17.3 min (S), tR(major) = 20.8 min (R). 1HNMR
4 H), 8.00–7.94 (m, 4 H), 7.39–7.34 (m, 8 H), 6.81 (t, JH,H
=
3
2
10.8 Hz, 2 H), 5.01 (dd, JH,H = 8.4, JH,H = 11.4 Hz, 2 H), 4.74
(dd, JH,H = 8.4, 10.8 Hz, 2 H), 1.84 (s, 6 H) ppm. 13CNMR
3
(75 MHz, CDCl3, 25 °C): δ = 169.4, 131.5, 130.0, 128.6, 128.3,
126.7, 125.7, 124.5, 123.7, 73.8, 39.6, 24.5 ppm. Rf (EtOAc/heptane,
1:3) = 0.23.
Representative Experimental Procedure for the Tandem Reaction to
Chromane 16: To a flame-dried Schlenk tube was added Mg(OTf)2
(8.0 mg, 0.025 mmol) and (S)-1-Np-box (12 mg, 0.028 mmol). The
mixture was dried in vacuo for 0.5 h and distilled anhydrous tolu-
ene (0.5 mL) was added. After stirring for 0.5 h, 15 (58 mg,
0.25 mmol) and N,N-dimethyl-p-toluidine (3.6 µL, 0.025 mmol)
were added and the mixture was cooled to 0 °C. After addition of
14 (55 µL, 0.50 mmol), the mixture was stirred overnight at 0 °C.
Flash chromatography (Et2O/pentane, 1:4) afforded 16 as a colour-
less oil (77 mg, 89%): the ee was determined by HPLC to be 73%:
3
(300 MHz, CDCl3, 25 °C): δ = 7.31–7.13 (m, 10 H), 6.45 (d, JH,H
3
= 15.9 Hz, 1 H), 6.30 (dd, JH,H = 8.4, 15.6 Hz, 1 H), 4.25 (dd,
3
3JH,H = 8.4, 10.8 Hz, 1 H), 3.93 (d, JH,H = 10.8 Hz, 1 H), 3.69 (s,
OD column; hexane/iPrOH, 95:5; 1.0 mL/min; tR(major)
=
12.9 min, tR(minor) = 14.4 min. [α]2D5 = +0.28 (c = 0.63, CHCl3,
3 H), 3.51 (s, 3 H) ppm.
73% ee). 1HNMR (400 MHz, CDCl3, 25 °C): δ = 7.31 (dt, 3JH,H
=
Representative Experimental Procedure for the Mukaiyama Michael
Reaction to Diester 26: To a flame-dried Schlenk tube was added
CuCl2 (7.3 mg, 54 µmol) and (S)-tBu-box (17 mg, 59 µmol). The
mixture was dried under vacuum for 0.5 h and distilled anhydrous
CH2Cl2 (2.0 mL) was added. After stirring overnight, the solvent
was removed in vacuo. The resulting fluorescent green solids were
dissolved in CH2Cl2 (0.54 mL) and the flask was protected from
light. After addition of AgSbF6 (37 mg, 0.11 mmol), the mixture
was stirred for 4 h at room temp. The white solids (AgCl) were
removed by filtering the solution through an oven-dried glass pi-
pette tightly packed with cotton. The resulting turquoise solution
contained Cu(SbF6)2-tBu-box catalyst complex (0.1  in CH2Cl2).
3
2.0, 8.4 Hz, 2 H), 7.19 (dt, JH,H = 2.0, 8.4 Hz, 2 H), 6.60 (dd,
3
3JH,H = 0.8, 8.8 Hz, 1 H), 6.45–6.41 (m, 2 H), 4.43 (d, JH,H
=
3
2
2.0 Hz, 1 H), 4.27 (dd, JH,H = 6.0, JH,H = 13.2 Hz, 1 H), 3.90 (s,
3
2
3 H), 3.74 (s, 3 H), 2.38 (td, JH,H = 2.0, JH,H = 13.2 Hz, 1 H),
2.24 (dd, JH,H = 5.6, JH,H = 13.2 Hz, 1 H) ppm. 13CNMR
(100 MHz, CDCl3, 25 °C): δ = 170.3, 159.8, 152.4, 142.3, 132.9,
130.4, 130.1, 129.1, 117.4, 108.8, 102.1, 101.9, 94.6, 55.6, 55.5, 53.8,
36.9, 36.8 ppm. HRMS [M + Na] calcd. 371.0662; found 371.0665.
3
2
Representative Experimental Procedure for the Hetero-Diels–Alder
Reaction to Pyran 19: To a flame-dried Schlenk tube was added
Cu(OTf)2 (10 mg, 28 µmol) and (S)-tBu-box (9.0 mg, 31 µmol). The
mixture was dried under vacuum for 0.5 h and distilled anhydrous
THF (1.0 mL) was added. After stirring for 1 h, methyl pyruvate
Alkylidene malonate 25[32] (60 mg, 0.27 mmol) and 1,1,1,3,3,3-
hexafluoro-2-propanol (HFIP) (50 µL, 0.46 mmol) were dissolved
Eur. J. Org. Chem. 2005, 4975–4987
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4985

F. P. J. T. Rutjes et al.
FULL PAPER
[4]
[5]
For an example, see: D. A. Evans, J. M. Janey, Org. Lett. 2001,
3, 2101–2103.
J. Clariana, J. Comelles, M. Moreno-Manas, A. Vallribera, Tet-
rahedron: Asymmetry 2002, 13, 1551–1554.
J. C.-D. Le, B. L. Pagenkopf, Org. Lett. 2004, 6, 4097–4099.
S. Crosignani, G. Desimoni, G. Faita, P. P. Righetti, Tetrahe-
dron 1998, 54, 15721–15730.
M. A. Pericas, C. Puigjaner, A. Riera, A. Vidal-Ferran, M. Go-
mez, F. Jimenez, G. Muller, M. Rocamora, Chem. Eur. J. 2002,
8, 4164–4178.
in toluene (0.7 mL) and the mixture was cooled to –78 °C. After
addition of the Cu(SbF6)2-tBu-box catalyst solution (0.1  in
CH2Cl2, 0.54 mL, 54 µmol), the mixture was stirring for 1 h at
–78 °C. Silyl ketene acetal 20[31] (0.10 mL, 0.41 mmol) was added
and the mixture was warmed up to room temp. overnight. The
mixture was concentrated in vacuo. Flash chromatography (EtOAc/
heptane, 1:5) afforded 26 as a white solid (94 mg, 99%): the ee was
determined by HPLC to be Ͼ90%: OD column; hexane/iPrOH,
[6]
[7]
[8]
98:2; 1.0 mL/min; tR(major)
= 14.1 min (R), tR(minor) =
15.6 min(S). 1HNMR (200 MHz, CDCl3, 25 °C): δ = 7.34–7.19 (m,
[9]
H. L. van Lingen, J. K. W. van de Mortel, K. F. W. Hekking,
F. L. van Delft, T. Sonke, F. P. J. T. Rutjes, Eur. J. Org. Chem.
2003, 19, 317–324.
5 H), 3.95 (dt, 3JH,H = 5.4, 10.0 Hz, 1 H), 3.80 (t, JH,H = 10.0 Hz,
3
1 H), 3.75 (s, 3 H), 3.49 (s, 3 H), 3.01–2.84 (m, 2 H), 1.31 (s, 9 H)
[10]
ppm.
Reference numbers CCDC-255207 (for, 1-Np-box), CCDC-
206608 {for [CuCl2(1-Np-box)]}, and CCDC-255209 {for
[CuCl2(2-Np-box)]} contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
a) L. B. Wolf, T. Sonke, K. C. M. F. Tjen, B. Kaptein, Q. B.
Broxterman, H. E. Schoemaker, F. P. J. T. Rutjes, Adv. Synth.
Catal. 2001, 343, 662–674; b) T. Sonke, B. Kaptein, W. H. J.
Boesten, Q. B. Broxterman, J. Kamphuis, H. E. Schoemaker,
F. Formaggio, C. Toniolo, F. P. J. T. Rutjes, in: Stereoselective
Biocatalysis (Ed.: R. N. Patel), Marcel Dekker, New York,
2000, p. 23–58.
W. J. J. van den Tweel, T. J. G. M. van Dooren, P. H. de Jonge,
B. Kaptein, A. L. L. Duchateau, J. Kamphuis, Appl. Microbiol.
Biotechnol. 1993, 39, 296–300.
R. M. Williams, D. J. Aldous, S. C. Aldous, J. Org. Chem. 1990,
55, 4657–4663.
D. Ben-Ishai, I. Sataty, Z. Bernstein, Tetrahedron 1976, 32,
1571–1573.
H.-S. Chong, K. Garmestani, L. H. Bryant Jr., M. W. Brech-
biel, J. Org. Chem. 2001, 66, 7745–7750.
D. E. Stack, A. L. Hill, C. B. Diffendaffer, N. M. Burns, Org.
Lett. 2002, 4, 4487–4490.
Crystal Structure Determination of [CuCl2(1-Np-box)]:[10]
C29H26Cl2CuN2O2, M = 569, crystallises in the orthorhombic
space group P212121, with a = 10.066(1) Å, b = 12.920(2) Å, c =
19.424(2) Å, V = 2526(1) Å3, T = 120 K, Z = 4, µ = 1.106 mm–1;
30861 reflections were measured, giving 7316 independent, of
which 6592 with I Ͼ 3σ(I) were used in the refinements. The in-
ternal agreement had R = 0.087, the final R = 0.028, Rw = 0.035,
GOF = 1.21, for the significant reflections.
Crystal Structure Determination of [CuCl2(2-Np-box)]:[10]
C29H26Cl2CuN2O2, M = 568.96, crystallises in the orthorhombic
space group P212121, with a = 11.7629(12) Å, b = 11.7891(6) Å, c
= 19.4027(17) Å, V = 2690.7(4) Å3, T = 208(2) K, Z = 4, µ =
1.039 mm–1; 78222 reflections were measured, giving 4708 indepen-
dent, of which 4050 with I Ͼ 2σ(I) were used in the refinements.
The internal agreement had R = 0.0623, the final R = 0.0465, Rw
= 0.1155, GOF = 1.110, for the significant reflections.
[11]
[12]
[13]
[14]
[15]
[16]
[17]
Acknowledgments
R. M. Kellogg, J. W. Nieuwenhuijzen, K. Pouwer, T. R. Vries,
Q. B. Broxterman, R. F. P. Grimbergen, B. Kaptein, R. M.
La Crois, E. de Wever, K. Zwaagstra, A. C. van der Laan, Syn-
thesis 2003, 10, 1626–1638.
Gerard Metselaar, Dr. Jos Hulshof and Erik van Echten (Syncom,
Groningen, The Netherlands) are kindly acknowledged for the
Dutch Resolution experiments and the separation by preparative
chiral HPLC. Etienne Gijsberts and Marcel A. H. Moelands are
gratefully acknowledged for their contribution.
[18]
[19]
Preparative chiral HPLC analysis of the racemic mixture on a
chiralpak AD column (heptane with 0.5% EtOH and 0.2%
Et3N) yielded both enantiomers of amino alcohol 7c.
a) J. Frelek, Tetrahedron: Asymmetry 1999, 10, 2809–2816; b) J.
Frelek, J. Jazwinski, M. Masnyk, P. Ruskowska, R. Szmigielski,
Tetrahedron: Asymmetry 2005, in press; c) For a recent review,
see: J. Zhang, A. E. Holmes, A. Sharma, N. R. Brooks, R. S.
Rarig, J. Zubieta, J. W. Canary, Chirality 2003, 15, 180–189.
a) D. A. Evans, G. S. Peterson, J. S. Johnson, D. M. Barnes,
K. R. Campos, K. A. Woerpel, J. Org. Chem. 1998, 63, 4541–
4544; b) D. A. Evans, K. A. Woerpel, M. M. Hinman, M. M.
Faul, J. Am. Chem. Soc. 1991, 113, 726–728.
[1] For reviews of C2-bisoxazoline Lewis acid catalysts, see, for
example: a) H. A. McManus, P. J. Guiry, Chem. Rev. 2004, 104,
4151–4202; b) D. Rechavi, M. Lemaire, Chem. Rev. 2002, 102,
3467–3494; c) J. S. Johnson, D. A. Evans, Acc. Chem. Res.
2000, 33, 325–335; d) K. A. Jørgensen, M. Johannsen, S. Yao,
H. Audrain, J. Thorhauge, Acc. Chem. Res. 1999, 32, 605–613;
e) A. K. Ghosh, P. Mathivanen, J. Cappiello, Tetrahedron:
Asymmetry 1998, 9, 1–45; f) A. Pfaltz, Acc. Chem. Res. 1993,
26, 339–345.
[2] For recent examples, see: a) H. Usuda, A. Kuramochi, M.
Kanai, M. Shibasaki, Org. Lett. 2004, 6, 4387–4390; b) P. O’Le-
ary, N. P. Krosveld, K. P. de Jong, G. van Koten, R. J. M.
Klein Gebbink, Tetrahedron Lett. 2004, 45, 3177–3180; c) T.
Rovis, D. A. Evans, Prog. Inorg. Chem. 2001, 50, 1–50; d) J. M.
Fraile, J. I. Garcia, M. A. Harmer, C. I. Herrerias, J. A.
Mayoral, J. Mol. Cat. A., Chem. 2001, 165, 211–218.
[3] For recent examples, see: a) A. Bayer, M. M. Endeshaw, O. R.
Gautun, J. Org. Chem. 2004, 69, 7198–7205; b) M. Kurosu,
J. R. Porter, M. A. Foley, Tetrahedron Lett. 2004, 45, 145–148;
c) A. K. Ghosh, M. Shirai, Tetrahedron Lett. 2001, 42, 6231–
6233; d) D. A. Evans, J. S. Johnson, E. J. Olhava, J. Am. Chem.
Soc. 2000, 122, 1635–1649; e) K. A. Jørgensen, Angew. Chem.
Int. Ed. 2000, 39, 3558–3588.
[20]
[21]
a) A. J. Phillips, Y. Uto, P. Wipf, M. J. Reno, D. R. Williams,
Org. Lett. 2000, 2, 1165–1168; b) P. Lafargue, P. Guenot, J.-P.
Lellouche, Heterocycles 1995, 41, 947–958.
[22]
[23]
[24]
[25]
H. L. van Lingen, W. Zhuang, T. Hansen, F. P. J. T. Rutjes,
K. A. Jørgensen, Org. Biomol. Chem. 2003, 1, 1953–1958.
S. Yao, M. Johannsen, H. Audrain, R. G. Hazell, K. A.
Jørgensen, J. Am. Chem. Soc. 1998, 120, 8599–8605.
D. A. Evans, T. Rovis, M. C. Kozlowski, C. W. Downey, J. S.
Tedrow, J. Am. Chem. Soc. 2000, 122, 9134–9142.
M. Johannsen, K. A. Jørgensen, J. Org. Chem. 1995, 60, 5757–
5762.
J. S. Johnson, D. A. Evans, Acc. Chem. Res. 2000, 33, 325–335.
J. Thorhauge, M. Roberson, R. G. Hazell, K. A. Jørgensen,
Chem. Eur. J. 2002, 8, 1888–1898.
a) D. A. Evans, C. S. Burgey, M. C. Kozlowski, S. W. Tregay, J.
Am. Chem. Soc. 1999, 121, 686–699; b) D. A. Evans, M. C.
[26]
[27]
[28]
4986
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2005, 4975–4987

Aryl-Substituted 2,2Ј-Bis(1,3-oxazoline) Ligands in Asymmetric Synthesis
FULL PAPER
Kozlowski, C. S. Burgey, D. W. C. MacMillan, J. Am. Chem.
Soc. 1997, 119, 7893–7894.
[31] R. M. Williams, J. A. Hendrix, J. Org. Chem. 1990, 55, 3723–
3728.
[29] Reference numbers for [CuCl2(tBu-box)] and [CuCl2(Ph-box)]
complexes in the Cambridge Crystallographic Data Centre:
CCDC-166462 and CCDC-166464, respectively.
[30] D. A. Evans, M. C. Kozlowski, J. A. Murry, C. S. Burgey, K. R.
Campos, B. T. Connell, R. J. Stapless, J. Am. Chem. Soc. 1999,
121, 669–685.
[32] A. Garcia, L. Castedo, D. Dominguez, Tetrahedron 1995, 51,
8585–8598.
Received: July 27, 2005
Published Online: October 10, 2005
Eur. J. Org. Chem. 2005, 4975–4987
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4987