Catalytic asymmetric formation of δ-Lactones from Unsaturated acyl halides

Article abstract of DOI:10.1002/chem.200902896

Previously unexplored enantiopure zwitterionic ammonium dienolates have been utilized in this work as reactive intermediates that act as diene components in hetero-Diels-Alder reactions (HDAs) with aldehydes to produce optically active δ-lactones, subunits of numerous bioactive products. The dienolates were generated in situ from E/Z mixtures of a,b- unsaturated acid chlorides by use of a nucleophilic quinidine derivative and Sn (OTf)₂ as co-catalyst. The latter component was not directly involved in the cycloaddition step with aldehydes and simply facilitated the formation of the reactive dienolate species. The scope of the cycloaddition was considerably improved by use of a complex formed from Er- (OTf)₃ and a simple commercially available norephedrine-derived ligand that tolerated a broad range of aromatic and heteroaromatic aldehydes for a cooperative bifunctional Lewis-acid-/ Lewis-base-catalyzed reaction, providing a,b-unsaturated d-lactones with excellent enantioselectivities. Mechanistic studies confirmed the formation of the dienolate intermediates for both catalytic systems. The active Er^III complex is most likely a monomeric species. Interestingly, all lanthanides can catalyze the title reaction, but the efficiency in terms of yield and enantioselectivity depends directly on the radius of the Ln ^III ion. Similarly, use of the pseudolanthanides Sc^III and Y^III also resulted in product formation, whereas the larger La ^III and other transition metal salts, as well as main group metal salts, proved to be inefficient. In addition, various synthetic transformations of 6- CCl₃- or 4-silyl-substituted α,β-unsaturated d-lactones, giving access to a number of valuable δ-lactone building blocks, were investigated.

Full text of DOI:10.1002/chem.200902896

FULL PAPER
DOI: 10.1002/chem.200902896
Catalytic Asymmetric Formation of d-Lactones from
Unsaturated Acyl Halides
Paolo S. Tiseni[b] and Renꢀ Peters*[a]
Dedicated to Professor Franz Effenberger on the occasion of his 80th birthday
Abstract: Previously unexplored enan-
cloaddition was considerably improved
by use of a complex formed from Er-
AHCTUNGRETG(NNUN OTf)3 and a simple commercially
lytic systems. The active ErIII complex
tiopure zwitterionic ammonium dieno-
lates have been utilized in this work as
reactive intermediates that act as diene
components in hetero-Diels–Alder re-
actions (HDAs) with aldehydes to pro-
duce optically active d-lactones, sub-
units of numerous bioactive products.
The dienolates were generated in situ
from E/Z mixtures of a,b-unsaturated
acid chlorides by use of a nucleophilic
is most likely a monomeric species. In-
terestingly, all lanthanides can catalyze
the title reaction, but the efficiency in
terms of yield and enantioselectivity
depends directly on the radius of the
LnIII ion. Similarly, use of the pseudo-
lanthanides ScIII and YIII also resulted
in product formation, whereas the
larger LaIII and other transition metal
salts, as well as main group metal salts,
proved to be inefficient. In addition,
various synthetic transformations of 6-
CCl3- or 4-silyl-substituted a,b-unsatu-
rated d-lactones, giving access to a
number of valuable d-lactone building
blocks, were investigated.
available norephedrine-derived ligand
that tolerated a broad range of aromat-
ic and heteroaromatic aldehydes for a
cooperative bifunctional Lewis-acid-/
Lewis-base-catalyzed reaction, provid-
ing a,b-unsaturated d-lactones with ex-
cellent enantioselectivities. Mechanistic
studies confirmed the formation of the
dienolate intermediates for both cata-
quinidine derivative and SnACTHNUTRGENUG(N OTf)2 as
co-catalyst. The latter component was
not directly involved in the cycloaddi-
tion step with aldehydes and simply fa-
cilitated the formation of the reactive
dienolate species. The scope of the cy-
Keywords: bifunctional catalysis ·
cycloaddition · hetero-Diels–Alder
reaction · lactones · lanthanides
Introduction
over-engineered, sensitive substrates that are tedious and ex-
pensive to prepare.
The quest for operationally simple catalytic methodologies
that permit the diversity-oriented asymmetric synthesis of
densely functionalized molecules from simple precursors is a
considerable challenge for modern synthetic chemistry. In
this regard, the discovery and application of new reactive in-
termediates in association with chiral catalyst systems repre-
sents an attractive tool with which to overcome the need for
The Diels–Alder reaction is recognized as one of the most
useful and powerful synthetic transformations, because its
application not only leads to great increases in molecular
complexity, but can also result in structures that lend them-
selves to additional amplification of complexity. Of the dif-
ferent versions of this methodology, the hetero-Diels–Alder
reaction (HDA) has found considerable relevance.[1] The ap-
plication of chiral Lewis acids has been the most frequently
employed strategy to achieve stereocontrol in HDA reac-
tions with carbonyl derivatives. In particular, the [4+2] cy-
cloaddition of Danishefsky-type dienes with different alde-
hydes to prepare enantioenriched pyran-4-one derivatives
has been widely investigated.[2–9]
[a] Prof. Dr. R. Peters
Institut fꢁr Organische Chemie, Universitꢂt Stuttgart
Pfaffenwaldring 55, 70569 Stuttgart (Germany)
Fax : (+49)711-685-64321
E-mail: rene.peters@oc.uni-stuttgart.de
The structurally related Brassard-type dienes [Brassardꢀs
[b] Dr. P. S. Tiseni
diene: 1,3-dimethoxy-1-(trimethylsilyloxy)buta-1,3-diene[10]
]
ETH Zꢁrich, Hçnggerberg, Laboratory of Organic Chemistry
Wolfgang-Pauli-Strasse 10, 8093 Zꢁrich (Switzerland)
provide direct access to b-methoxy-substituted a,b-unsatu-
rated d-lactones. d-Lactones constitute an exceptionally
widespread structural motif in natural and synthetic com-
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200902896.
Chem. Eur. J. 2010, 16, 2503 – 2517
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2503

pounds displaying a wide range of biological activity. In
many cases they show high efficacy as antibacterial,[11] anti-
viral (e.g. HIV protease inhibitors),[12] anticancer,[13] immu-
nosuppressive,[14] or cholesterol-lowering (HMGR inhibi-
tors) agents.[15] The majority of statin drugs—such as Lipitor
and Zocor, for example—contain either a b-hydroxy-d-lac-
tone moiety or the corresponding open-chain d-hydroxy car-
boxylate form.[16] In addition, d-lactones are very useful
building blocks for the synthesis of bioactive compounds.[17]
Unfortunately, the synthesis, isolation, purification, and
application of Brassard-type dienes is complicated by their
moisture- and acid-sensitive natures, which has limited their
utility in synthetic transformations. In addition, so far only a
few examples of the use of vinylketene acetals as diene com-
ponents in highly enantioselective HDA reactions[18–19] have
been reported, because these dienes are prone to be de-
stroyed by Lewis acids. All reported HDA examples have,
to the best of our knowledge, been restricted to the use of
Brassard-type dienes and require extended reaction times to
provide useful yields.
lyzed asymmetric synthesis of b-lactones from ketenes
through the intermediate formation of zwitterionic eno-
lates.[22]
Scheme 1. Proposed formation of zwitterionic dienolate intermediates for
[4+2] cycloaddition with aldehydes.
Vinylketenes had previously
not been useful for catalytic
asymmetric Diels–Alder reac-
tions, as a result of their varia-
ble tendency to undergo [2+2]
cycloadditions[23] and due to
their inherent and (in compari-
son with non-conjugated ke-
tenes) significantly increased in-
stability (rapid dimerization
and polymerization, sensitivity
towards moisture, air, acids, and
bases).[24–25] However, because
of the considerable homology
of 3 to vinylketene acetals, it
was anticipated that the inter-
mediate and rapid formation of
these dienolates should over-
come these problems. Early
studies had shown that vinylke-
tenes are rapidly trapped by
various nucleophiles.[26]
Asymmetric ammonium eno-
lates, generated in situ through
the reaction between a chiral
Here we present a different HDA concept for the synthe-
sis of d-lactones 6, circumventing the use of vinylketene ace-
tals. Our work was based on the hypothesis that vinylke-
tenes 2, which are formed in situ by dehydrohalogenation
from a,b-unsaturated acid chlorides 1 (Scheme 1),[20] might
be trapped and at the same time activated as diene compo-
nents for Diels–Alder reactions by enantiopure tertiary
amines, thus forming enantiomerically pure zwitterionic di-
enolates of type 3.[21] It was believed that a species of this
type should be reactive enough, when adopting an s-cis con-
formation, to undergo [4+2] cycloadditions with aldehydes.
Our investigations were inspired by the tertiary-amine-cata-
tertiary amine catalyst and ketene or a ketene equivalent,[27]
have been widely exploited as reactive intermediates for the
synthesis of optically active compounds. The first studies in
this direction date back to the early 1960s, when Pracejus re-
ported that enantiomerically enriched esters can be obtained
by addition of alcohols to ketenes in the presence of brucine
as a chiral catalyst.[28]
2504
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 2503 – 2517

Catalytic Asymmetric Formation of d-Lactones
FULL PAPER
Results and Discussion
significantly improved with all investigated Lewis acids,[29]
and the yield of 6a increased to >70% in the presence of
At the outset, the model reaction between 3,4-dimethyl-
pent-2-enoyl chloride (1a) and the highly electron-deficient
aldehyde chloral (5) was investigated (Scheme 2). Initial ex-
20 mol% variously of Er
ACHUTGTNERN(GUN OTf)3 (76%), SnACHTUNGTRNEN(UGN OTf)2 (75%),
Sc(OTf)3 (74%), or Nd(OTf)3 (72%) when the acid chloride
A
ACHTUNGTRENNUNG
was slowly added by syringe pump over a period of 30 min.
The enantiomeric excesses were determined to be
82(Æ1)% in all experiments regardless of the nature of the
Lewis acid co-catalyst. Because the enantioselectivity does
not depend upon the presence or absence of the metal tri-
flate salt, we assume that the Lewis acid is not directly in-
volved in the cycloaddition step and simply facilitates the
dehydrochlorination of 1. Increasing the temperature to 08C
Scheme 2. Proof of concept.
gave similar results [70% yield with Er
whereas decreasing the temperature to À408C resulted in a
dramatic decrease in yield (22% with Er(OTf)3, ee 84%).
ACHTUNGRTEN(NUNG OTf)3, ee=80%],
AHCTUNGTRENNUNG
periments in acetonitrile at À158C with stoichiometric
amounts of quinuclidine or triethylamine, acting both as a
base and as an achiral nucleophilic catalyst, provided the
racemic d-lactone 6a in moderate yield.
Investigation of the co-catalyst loadings (Table 2) showed
that reducing the amount of the Lewis acid to 10 mol% had
almost no negative effect in the case of SnACTHNURTGNENG(U OTf)2 (entries 1
The combination of trimethylsilylquinidine (TMSQd, 4a,
1.0 equiv) and iPr2NEt (Hꢁnigꢀs base, 2.0 equiv) in toluene
at À158C gave the product with an ee value of 82%, albeit
in very low yield (18%, Table 1, entry 1), which was as-
cribed to incomplete conversion of 1a (1H NMR monitor-
ing). In order to facilitate the deprotonation process and to
activate the aldehyde substrate, various metal triflate salts
Table 2. Investigation of catalyst and Lewis acid loadings for the model
reaction between 1a and 5.
MACHTUNGTRENNUNG(OTf)n were investigated as Lewis acid co-catalysts (en-
tries 2–18). The deprotonation process—visible through the
Entry TMSQd (x equiv) SnACHTNUTGRNEUNG
(OTf)2 (y equiv) Yield[a] [%] ee[b] [%]
1
2
3
4
5
6
1
1
1
1
0.4
0.2
0.2
0.1
0.2
0.1
0.05
0.014
0.1
0.1
0.1
0.1
75
72
65
60
72
65
78[d]
70
81
82
82
82
81
81
82
81
formation of a precipitate of [iPr2NHEt]Cl in toluene—was
Table 1. Investigation of the effects of various Lewis acids in the
TMSQd-catalyzed model reaction between 1a and 5.
7[c]
8[c]
[a] Yield determined from the mass of the crude product in combination
with 1H NMR with DMSO as internal standard. [b] ee determined by
HPLC. [c] Compound 1a was added by syringe pump over 110 min.
[d] Isolated yield after column chromatography.
Entry
Lewis acid
–
Yield[a] [%]
ee[b] [%]
1
18
75
66
62
62
68
61
63
68
56
74
55
72
65
57
61
76
68
82
81
83
82
81
81
81
81
81
81
81
82
82
82
82
82
82
82
and 2), and even with as little as 1.4mol% the reaction was
still relatively efficient (entry 4). Decreasing the amount of
TMSQd (4a) necessitated a slower addition of the acid chlo-
ride in order to avoid massive polymerization (entries 6 and
7). Through the action of 4a (20 mol%) and SnACHTUNGTRENNUNG(OTf)2
(10 mol%) with addition of 1a to the reaction mixture over
110 min, 6a was obtained in 78% isolated yield.
Similar reaction conditions were applied to several alter-
native substrates 1b–f, bearing different branched or un-
branched aliphatic, alicyclic, or aromatic groups R1
(Table 3). The d-lactones 6a–f were formed in good yields
and with ees of up to 95% (entries 1–7). Both the enantiose-
lectivities and the degrees of conversion of the acid chlor-
ides depended primarily upon the steric bulk of R1. Whereas
with the unbranched Et substituent the ee was only moder-
ate (entry 2), the values were significantly higher and prepa-
ratively useful with branched or aromatic substituents (en-
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Sn
LiOTf
Mg(OTf)2
Al(OTf)3
In(OTf)3
Bi(OTf)3
Zr(OTf)4
Cu(OTf)2
Zn(OTf)2
Sc(OTf)3
(OTf)3
Nd(OTf)3
Sm(OTf)3
Eu(OTf)3
Gd(OTf)3
Er(OTf)3
Yb(OTf)3
ACHTUNGTRENNUNG(OTf)2
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
YACHTUNGTRENNUNG
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
[a] Yield determined from the mass of the crude product in combination
with 1H NMR with DMSO as internal standard. [b] ee determined by
HPLC.
Chem. Eur. J. 2010, 16, 2503 – 2517
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2505

R. Peters and P. S. Tiseni
Table 3. Preparation of a,b-unsaturated d-lactones 6.
Entry
1
R1
X
Y
Yield[a] [%]
ee[b] [%]
1
2
1a
1b
1c
1d
1e
1e
1 f
1g
1g
1h
1i
iPr
Et
iBu
cHex
tBu
20
20
40
20
20
40
20
100
40
10
10
20
10
10
20
10
30
20
30
30
30
78
60
73
75
58
80
73
54
43
47
51
61
82
54
70
83
95
95
81
96
96
92
97
97
3[c]
4
5
6
7
8
tBu
Ph
Et3Si
Et3Si
BnMe2Si
nPr3Si
nBu3Si
9[d]
10
11
12
100
100
100
Scheme 3. Determination of the absolute configuration by chemical cor-
relation: synthesis of ent-6 f via the known b-lactone 7.
1j
[a] Isolated yields. [b] ee determined by HPLC. [c] Catalyst 4b, bearing
an OSi(iPr)3 group instead of the OMe group on the quinoline part, was
ACHTUNGTRENNUNG
used (see Scheme 1 for the catalyst structure). [d] Compound 1g was
added by syringe pump over 220 min.
1
À
tries 3–7). Notably, even in the case of 1b (R =Et) the C C
bond formation occurred with complete regioselectivity.
With the more bulky iBu or tBu substituents, higher catalyst
and co-catalyst loadings were required in order to obtain
high conversions (entries 3 and 6).
Scheme 4. Preparation of a,b-unsaturated acid chlorides 1g–j bearing
silyl substituents at their b-positions.
Both a 1.5:1 mixture of (Z)- and (E)-3-phenylbut-2-enoyl
chloride (1 f) and the geometrically pure E-configured sub-
strate afforded nearly identical yields and ee values
The carboxylic acid functionalities were subsequently
smoothly converted into the corresponding acid chlorides.
To provide acceptable yields in the cycloaddition step, sto-
ichiometric quantities of the tertiary amine catalyst had to
be used in order to overcome incomplete conversion of the
acid chlorides. Presumably the bulky trialkylsilyl substitu-
ents hamper the deprotonation of the reactive methyl
group. Nonetheless, the obtained enantioselectivities were
excellent.
As a general trend, the remote control of the enantiose-
lectivity for the reactions presented in Table 3 depends
largely on the steric bulk of R1. A working hypothesis to ex-
plain the enantioselectivity was developed with the aid of
MMFF calculations performed to figure out the preferred
conformations of the cisoid zwitterionic dienolates
(Scheme 5). Because the re face (with regard to the C1-eno-
late atom) is shielded by the quinoline and the OTMS moi-
eties, the aldehyde will attack from the better accessible si
face, by either a concerted or a stepwise pathway.[33] In the
preferred orientation of the aldehyde in the transition state
the residue R1 and the CCl3 group should point away from
each other to avoid unfavorable steric interactions, thus ex-
plaining the large influence of R1.
[TMSQd (0.4 equiv), SnACTHNUTRGNE(UNG OTf)2 (20 mol%); (E)-1 f: 68%
yield, 81% ee; (Z/E)-1 f: 71% yield, 81% ee], showing that
the configuration of the double bond does not have any sig-
nificant impact on the reaction outcome.
To determine the absolute configuration of the generated
stereocenter by chemical correlation, ent-6 f was prepared
by an alternative reaction pathway (Scheme 3).
The R-configured b-lactone 7 was obtained with 88% ee
by a reported procedure[30] and subsequently ring-opened
under Friedel–Crafts conditions to provide a b-hydroxyke-
tone.[31] Peterson olefination of the TMS-protected deriva-
tive 8 afforded 9 as a mixture of two geometrical isomers.
After removal of the silyl group the E isomer cyclized under
acidic conditions to yield the targeted lactone ent-6 f, where-
as the Z isomer remained in the acyclic form. HPLC analy-
sis showed that this lactone ent-6 f had the opposite configu-
ration to the product obtained through the TMSQd/Sn-
ACHTUNGTRENNUNG(OTf)2-catalyzed [4+2] cycloaddition between the acid chlo-
ride 1 f and chloral (5), which was thus assigned as the S
enantiomer.
To extend the synthetic value of the methodology it was
also applied to substrates 1g–j bearing trialkylsilyl moieties
(Table 3, entries 8–12). The required carboxylic acids were
easily prepared by Ru-catalyzed hydrosilylation of tetrolic
acid by a procedure developed by Trost et al. (Scheme 4).[32]
At least two similar mechanisms may be envisaged for the
formation of the reactive intermediate 3. The first one in-
volves Lewis acid activation of the acid chloride for an
attack of the nucleophilic catalyst, thus forming the acylam-
monium species 13 (Scheme 6), which would be less Lewis
2506
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 2503 – 2517

Catalytic Asymmetric Formation of d-Lactones
FULL PAPER
Scheme 5. Proposed explanation of the remote stereocontrol.
Scheme 7. Alternative II for the reaction mechanism, involving a vinylke-
tene intermediate (L.A.=Lewis acid).
acid chlorides in the presence of triethylamine have been re-
ported.[26c]
An experiment with NEt3 in toluene in the absence of a
Lewis acid (Table 4, entry 1), affording a 20:80 mixture of
the a,b-unsaturated ester 14 and the b,g-unsaturated isomer
15, served as a reference in our studies. Contrary to the pre-
vious results by Truce and Bailey, MS analysis of the crude
material indicated that the a,b-unsaturated compound was
also partially deuterated (16% of monodeuterated ester
[D1]-14), whereas the b,g-unsaturated ester 15 was mainly
monodeuterated ([D1]-15, 61%). Small amounts of the bis-
deuterated esters [D2]-14 and [D2]-15 were also detected.
Under the elaborated conditions of the catalytic asymmet-
Scheme 6. Alternative I for the reaction mechanism, involving an acylam-
monium intermediate (L.A.=Lewis acid).
ric cycloaddition [TMSQd (4a, 0.2 equiv) and Sn
ACHTUNGTRENNUNG(OTf)2
(0.1 equiv), entry 2], the amount of a,b-unsaturated isomer
basic than the acid chloride.
Table 4. Deuteration experiments for the esterification of the a,b-unsaturated acid chloride 1 f.
The Lewis acid might hence be
released. Subsequently iPr2NEt
would deprotonate the g-posi-
tion, generating the zwitterionic
dienolate 3, which would under-
go a cycloaddition with the di-
enophile when adopting the s-
cis conformation. Because the
Lewis acid does not display any
influence on the selectivity of
the reaction, it is most probably
not coordinated to the zwitter-
ionic intermediate.
X
Catalyst
(equiv)
Base
(equiv)
14/
[D0]-
[D1]-
[D2]-
[D0]-
[D1]-
[D2]-
15[a]
14[b]
14[b]
14[b]
15[b]
15[b]
15[b]
1
2
–
NEt3 (2)
–
20:80 82
67
16
30
2
3
31
24
61
67
8
9
0.1 4a (0.2)
iPr2NEt (2) 31:69
1
[a] Determined by H NMR. [b] The percentages of nondeuterated (D0), monodeuterated (D1), and bisdeuter-
ated (D2) ester in 14 and 15 were determined by GC-MS.
The
second
alternative
(Scheme 7) involves dehydrohalogenation of the acid chlo-
ride, initiated by the stoichiometric base and assisted by the
Lewis acid, thus forming the vinylketene derivative 2. This
species is again less Lewis basic than the acid chloride, thus
releasing the Lewis acid. After attack of the nucleophilic
catalyst on the ketene, the zwitterionic dienolate intermedi-
ate 3 is formed.
was increased to 31%. MS analysis showed that the a,b-un-
saturated ester 14 contained a larger proportion of the mon-
odeuterated compound [D1]-14 (30 vs. 16%), whereas in the
case of 15 only a slight increase in the degree of deuteration
was observed.
As a further control experiment, a mixture of undeuterat-
ed esters 14 and 15 was subjected to the reaction conditions
To provide more information about the reaction mecha-
nism, deuteration experiments with monodeuterated metha-
nol (MeOD) were carried out. Related studies directed to-
wards investigation of the alcoholysis of a,b-unsaturated
[TMSQd (0.2 equiv), SnACTHNGUTERN(NUG OTf)2 (0.1 equiv), DIEA (2 equiv)]
in the presence of MeOD (2 equiv). The esters were recov-
ered undeuterated, showing that no deuterium incorporation
occurs through H/D exchange after product formation.
Chem. Eur. J. 2010, 16, 2503 – 2517
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2507

R. Peters and P. S. Tiseni
The generation mainly of the a-monodeuterated b,g-unsa-
turated ester [D1]-15 can be readily explained in terms of a
vinylketene intermediate as depicted in Scheme 8.
a range of a,b- and b,g-unsaturated deuterated esters are
thus generated (pathways B–E). In addition, deuterated
methanol is not the only D+ source in the acid/base equili-
bria, because a considerable amount of protonated Hꢁnigꢀs
base is formed during the reaction and can undergo H/D ex-
change with MeOD. From the experimental data it is diffi-
cult to draw a definite conclusion about the intermediate
formation of a vinylketene species. However, the existence
of a zwitterionic dienolate intermediate has been estab-
lished.
The trichloromethyl moiety in lactones 6 is a synthetically
versatile functional group that can be readily modified into
several valuable functionalities. Basic hydrolysis of 6a with
LiOH in water at 608C gave the carboxylic acid 16 without
significant racemization (Scheme 10).
Scheme 8. Generation of monodeuterated ester [D1]-15 via a vinylketene
intermediate.
The formation of the monodeuterated a,b-unsaturated
ester [D1]-14, however, as well as the observation of bisdeu-
terated species for both isomers, is consistent with a parallel
mechanism involving a series of acid/base equilibria based
on an acylammonium/zwitterionic dienolate couple
(Scheme 9).
Scheme 10. Synthetic modification of the trichloromethyl group.
The hydrolysis occurred with inversion of the stereocen-
ter, in close analogy to literature precedence for the hydrol-
ysis of CCl3-substituted b-lactones and trichloromethylcarbi-
nols,[34] due to the intermediate formation of the dichloroep-
oxide 20, which is ring-opened through an SN2 mechanism
(Scheme 11).
Scheme 9. General mechanistic picture of the equilibria involving the
acylammonium/dienolate couple.
Whereas the formation of the undeuterated a,b-unsaturat-
ed ester [D0]-14 should mainly occur through the acylammo-
nium salt 13 (reaction pathway A), the generation of the
monodeuterated [D1]-14 and the bisdeuterated compounds
provides compelling evidence for the existence of the
common zwitterionic dienolate intermediate 3, which could
be generated either by deprotonation of 13 or by attack of
the nucleophilic catalyst on a vinylketene. The equilibration
between 3 and 13 allows the incorporation of a variable
number of deuterium atoms. Upon trapping of 3 by MeOD
Scheme 11. Inversion of the configuration during the hydrolysis of 6a.
2508
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 2503 – 2517

Catalytic Asymmetric Formation of d-Lactones
FULL PAPER
Table 5. Diastereoselective preparation of b-hydroxy d-lactones 23 by
alkyl migration/oxidation.
Selective partial reductive removal of the chlorine atoms
of the trichloromethyl group was achieved by treatment
with tributyltin hydride (Scheme 10).[35] Performing the re-
duction in THF at 608C afforded the dichloro derivative
17,[36] whereas in toluene at reflux the monochloro deriva-
tive 18 was obtained. The latter was smoothly hydrolyzed to
give the primary alcohol 19. The configuration of the stereo-
center was also inverted in that case, as verified by direct
oxidation of compound 19 to carboxylic acid 16 with Jonesꢀ
reagent (CrO3 in H2SO4, 8n, Scheme 12). The final confir-
mation of the inversion was obtained by conversion of 19
into ent-18, with the opposite configuration to the material
obtained by radical reduction of 6a (Scheme 12).
Entry Substrate R’
R’’ Yield 22[a] [%] Yield 23[a] [%] d.r.[b] [%]
1[c]
2
3
6g
6h
6i
Et
Et
44
70
85
65
61
>99:1
25:1[c]
>99:1
>99:1
Bn Me 92
nPr nPr 38
nBu nBu 41
4
6j
[a] Isolated yield. [b] Determined by 1H NMR of isolated 23. [c] d.r. of
crude 22h and crude 23h=7:1.
drolyzed, presumably because the free carboxylate moiety
intramolecularly activates the silyl group.[32]
Different migration behavior was observed for the benzyl-
and alkyl-substituted d-lactones 6g–j: whereas the benzyl-
substituted compound 6h directly afforded the ring-opened
product 22h, in the alkyl migration the products 21 were
converted into the seco-acids 22 only slowly under the reac-
tion conditions. For that reason, an additional hydrolysis
step was required to accomplish
Scheme 12. Determination of the absolute configuration of the stereocen-
ter by chemical correlation.
Additional derivatization at the 4-position is possible
through modification of the silyl moieties in 6g–j. Fleming
et al.[37] reported that a fluoride-triggered 1,2-migration of a
phenyl substituent on silicon to an adjacent electrophilic
carbon can be induced by TBAF (tetrabutylammonium fluo-
ride) treatment of an a,b-disilyl enone. Migration of allyl
groups was first reported by Jung and Piizzi, who used an E-
the
subsequent
oxidation
(Table 5).
The configurations were un-
ambiguously determined by
NOE experiments (Figure 1)
and are consistent with the
configured b-silyl enone as substrate.[38] The C Si bond
À
Figure 1. NOE connectivities
for 23g.
could be oxidatively cleaved under Tamao conditions to
yield a homoallylic alcohol. Similarly, Salvatori et al. investi-
gated b-silyl-substituted a,b-unsaturated aldehydes.[39] Re-
cently, Trost and co-workers[32] reported related 1,2-migra-
tions for Z-configured b-silyl-substituted a,b-unsaturated ke-
tones or esters. The migration reactions were found to be se-
lective for aromatic, benzylic, and allylic residues, with
simple alkyl groups not being transferred.[40]
mechanism
Scheme 13.
depicted
in
Because the si face of the C=
C double bond is shielded by
For the b-silyl-substituted d-lactones 6g–j it was envisaged
that analogous migration reactions might be performed ste-
reoselectively, taking advantage of the sterically demanding
CCl3 residues in their d-positions, thus leading to the diaste-
reoselective formation of O-substituted quaternary stereo-
centers (Table 5). Treatment with TBAF (2.0 equiv) trig-
gered the migration of benzyl or simple alkyl substituents.
Silanols 22 were subsequently oxidized to provide the
almost diastereomerically pure tertiary alcohols 23 by slight-
ly modified Tamao oxidation conditions. To accomplish this
transformation, the lactone moiety in 21 first had to be hy-
Scheme 13. Proposed explanation for the diastereoselectivity.
Chem. Eur. J. 2010, 16, 2503 – 2517
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2509

R. Peters and P. S. Tiseni
the bulky trichloromethyl group, the addition occurs mainly
from the opposite face, generating the new stereocenter
with the S configuration (Scheme 13). It is likely that the ac-
celerated ring-opening step in the case of the benzyl sub-
strate 6h might compete with the fluoride-triggered migra-
tion, thus explaining the lower stereoselectivity in that case
(dr of the crude product 7:1), because the migration at the
acyclic stage would be expected to proceed without efficient
stereocontrol.
The formation of d-lactones can also be accomplished in
the case of a,a-dichlorinated aldehydes, as exemplified for
2,2-dichloropropanal[41] (24a). Although the use of metal tri-
flate co-catalysts did not lead to high levels of conversion of
24a, use of an excess of LiClO4 resulted in smooth reactions
(Table 6, entries 1–4).[42]
hemilabile coordination, still permitting sufficient reactivity
for nucleophilic trapping of a vinylketene intermediate.[47]
Although ErIII complexes with aliphatic b- or g-amino al-
cohols possessing tertiary amino groups have to the best of
our knowledge never previously been described in the litera-
ture, ErACTHNUTRGENNG(U OTf)3 was initially chosen as the lanthanide source
for these investigations, owing to the combination of: a) the
comparatively low price of Er, which is linked to its impor-
tance for the telecommunications industry,[48] and b) its rela-
tively small ionic radius (as a consequence of the lanthanide
contraction),[49] which was regarded as advantageous for
achieving a rigid cycloaddition transition state. The assump-
tion of a cooperative bifunctional Lewis acid/Lewis base ac-
tivation mechanism was initially supported by the observa-
tion that ErACHTNUGRTENUNG(OTf)3 did not abolish the nucleophilicity of pyr-
idine in the model reaction in THF/toluene depicted in
Table 7 (entry 1), whereas no reaction took place in the ab-
sence either of nucleophile or of Lewis acid.
Table 6. Preparation of a,b-unsaturated d-lactones 25 from dichlorinated
aldehydes 24.
Table 7. Screening of 1,2-amino alcohol derivatives for the Lewis-acid-/
Lewis-base-catalyzed formation of the a,b-unsaturated d-lactone 28aA.[a]
Entry
1
R1
R2
Y
Yield[a] [%]
ee[b] [%]
1
2
3
4
5
6
1a
1d
1e
1 f
1a
1a
iPr
Me
Me
Me
Me
Et
2.0
1.5
1.5
1.5
3.0
3.0
67
56
79
57
35
29
74
76
71
72
60
32
cHex
tBu
Ph
iPr
iPr
27
NR2
–
R1
R2 R3 R4 Yield[b] [%] ee[c] [%]
1
2
3
4
5
pyr
27a NMe2
27b
27c
27d
–
–
–
–
87
–
H
H
H
H
H
H
H
H
Ph Me 27
Ph Me 35
Ph Me 32
Ph Me 27
74
95
96
57
nPr
N
N
N
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
[a] Isolated yields. [b] ee determined by HPLC.
AHCTUNGTRENNUNG
6
7
27e
27 f
H
H
H
H
Ph Me 14
44
–
Ph Ph
0
Non-activated or less activated aldehydes such as benzal-
dehyde or p-nitrobenzaldehyde did not furnish the target
products. The development of bifunctional Lewis acid/Lewis
base catalysts through which the reactivities of the dienolate
and of the aldehyde could be controlled simultaneously thus
appeared to be required.[43] The combination of a Lewis acid
and a Lewis base successfully united in one catalytic system
has recently found numerous applications in asymmetric cat-
alysis, due to synergistic activation of both the electrophilic
and the nucleophilic substrates, often allowing high reaction
rates and excellent chirality transfer.[44]
8
9
27g
27h
N
N
N
N
N
N
H
H
H
Me
TMS
Ph Ph Me <5
6
–
32
À34
–
H
H
H
H
H
Ph
Me
H
0
10
10 27i
11 27j
12 27k
Ph Me 24
Ph Me
0
13 27l
N
H
Ph Me 11
À33
[a] Compound 1a was slowly added by syringe pump over 120 min (1:1
stoichiometry of both substrates). Stirring was continued for an addition-
al 150 min. [b] Yield determined from the mass of the crude product in
1
combination with H NMR with MeNO2 as internal standard. [c] ee deter-
mined by HPLC. Negative values indicate that the S-configured product
was preferentially formed.
Our studies were based upon the hypothesis that a lantha-
nide Lewis acid, offering an exceptionally high number of
coordination sites, should be advantageous for binding both
of the aldehyde and the dienolate, plus additional ligands to
control reactivity and stereoselectivity.[45–46]
To create a cycloaddition transition state with a high level
of organization the nucleophilic catalyst should be directly
connected to the Lewis acid template. We envisaged that an
oxophilic lanthanide should bind strongly to an alkoxide
moiety whereas a tertiary amino group should enter into
With N-methylephedrine (27a), the d-lactone 28aA was
formed with a promising ee of 74% (entry 2), although the
yield was low. Replacement of the NMe2 group by a pyrroli-
dine unit (27b) not only enhanced the reactivity (yield=
35%), but also resulted in an ee value of 95% (entry 3). A
piperidine ring (ligand 27c) provided similar results, where-
2510
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 2503 – 2517

Catalytic Asymmetric Formation of d-Lactones
FULL PAPER
Table 9. Optimization of the model reaction between 1a and 26A.
as the azetidine 27d was far less selective (entries 4–5). The
nucleophilicity of the tertiary amino group is essential, as it
was demonstrated in entries 6 and 7, in which the steric ac-
cessibility and the electron density of the amino group are
diminished with the consequence of reduced or no reactivity.
For ligands containing a tertiary or primary alcohol moiety
the title reaction was retarded (entries 8 and 9), in the latter
case presumably due to ligand O-acylation. An N-substitut-
ed stereocenter is required for high stereocontrol and, sur-
prisingly, also for sufficient reactivity (entry 10). A methyl-
protected hydroxy group in the ligand impeded high enan-
tioselectivity (entry 11), whereas TMS protection gave no
product at all (entry 12). Entry 13 demonstrates that O-acy-
lated amino alcohols, such as in 27l, cannot be significantly
involved in the catalytic cycle. Compound 27l was also
never detected in the reaction mixture or the crude product
with 27b as ligand. This indicates: a) that the oxygen atom
binds strongly to the ErIII ion, and b) that this bond must be
inert under the reaction conditions.
t [min]
T [8C]
X
Y
Z
Yield[a] [%]
ee[b] [%]
1
2
3
4
5
6
7
8
120
30
30
30
30
30
30
30
À15
À15
À10
À10
À10
À10
À10
À10
1.1
1.1
1.1
1.5
1.5
1.5
1.5
1.5
2.0
2.0
2.0
2.0
2.5
2.5
2.5
2.5
0.4
0.4
0.4
0.4
0.4
0.2
0.1
0.05
38
42
52
58
56
49
30
95
95
95
95
95
91
84
[a] Yield determined from the mass of the crude product in combination
with 1H NMR with MeNO2 as internal standard. [b] ee determined by
HPLC.
Different bases were investigated as alternatives to
iPr2NEt (Table 8). Although with the more bulky PMP
(1,2,2,6,6-pentamethylpiperidine) only minute amounts of
ErACTHUNTRGENNG(U OTf)3 from 1.1 to 1.5 equiv (entry 4), and the amount of
iPr2NEt from 2.0 to 2.5 equiv (entry 5), whereas the amount
of the chiral nucleophilic ligand could be decreased to 10–
20 mol% without seriously affecting the reaction outcome
(entries 6 and 7). Amino alcohol loadings lower than
10 mol% resulted in reduced yields and enantioselectivities
(entry 8).
Table 8. Investigation of different bases for the model reaction between
1a and 26A.
In contrast with the catalyst system described above, the
second-generation Lewis acid/Lewis base catalyst system
provided excellent enantioselectivities irrespective of the
size of the substituent R1 at the 3-position in the acid chlo-
ride 1 (Table 10, entries 1–6). Unbranched alkyl groups such
as Me or Et, a- or b-branched alkyls such as iPr and iBu,
alicyclic substituents such as cyclohexyl, or aromatic groups
such as Ph all furnished ee values of ꢀ94% with PhCHO as
test substrate. Although the yield was low with 3-methylbut-
2-enoyl chloride 1l, which is notoriously highly sensitive to-
wards polymerization under basic reaction conditions, in all
other cases in which 26A was employed as heterodienophile
the yields were synthetically useful.
Entry
Base (x equiv)
Yield[a] [%]
ee[b] [%]
1
2
3
4
5
6
iPr2NEt (2.5)
PMP (2.5)
MDC (2.5)
DMC (2.5)
NaH (1.3)
35
traces
33
59
0
95
95
89
2
–
–
NaHMDS (2.5)
0
[a] Yield determined from the mass of the crude product in combination
with 1H NMR with MeNO2 as internal standard. [b] ee determined by
HPLC.
The reactions generally provided excellent enantioselec-
tivities with all kinds of aromatic aldehydes regardless of
their electronic or steric natures (entries 7–21, ee 88–
96%).[51] Whereas electron donors such as o-OMe or p-Me
resulted in lower yields (entries 7 and 8), electron-withdraw-
ing substituents such as Cl, Br, NO2, or CF3 enhanced the
reactivity (entries 10–17) relative to PhCHO and also per-
mitted the use of a,b-unsaturated enals (entry 18). The sub-
stitution pattern of the aldehyde is less important; o-, m-, or
p-substituted systems were all well tolerated. Even the em-
ployment of electron-rich heterocycles such as furan- or thi-
ophenecarbaldehydes (entries 19–21) provided the desired
products highly enantioselectively, albeit in low yields. In
contrast, neither enolizable nor non-enolizable aliphatic al-
dehydes, nor non-activated a,b-unsaturated enals, were tol-
erated by the catalyst system.[52]
product were obtained (entry 2), methyl dicyclohexyl amine
(MDC) provided the desired product in comparable yield
and with slightly lower selectivity than obtained with
iPr2NEt (entry 3). In the case of the sterically less hindered
dimethyl cyclohexyl amine (DMC, entry 4) the reactivity
was considerably improved, but with complete loss of enan-
tioselectivity, indicating that DMC is nucleophilic enough to
catalyze a non-enantioselective background reaction. Strong
bases such as NaH or NaHMDS did not afford any targeted
product (entries 5&6).
To enhance the yield to a synthetically useful level, it was
necessary to decrease the addition time of 1a from 120 to
30 min (Table 9, entries 1 and 2)[50] and to raise the reaction
temperature from À15 to À108C (entry 3), the amount of
Chem. Eur. J. 2010, 16, 2503 – 2517
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2511

R. Peters and P. S. Tiseni
Table 10. Scope and limitations of the title reaction.[a]
28 R1
R2
27b [equiv] Yield[b] [%] ee[c] [%]
1
2
3
4
5
6
7
8
9
aA iPr
kA Me
bA Et
cA iBu
Ph
Ph
Ph
Ph
0.2
0.2
0.2
0.2
0.2
0.1
0.2
0.2
0.2
0.1
0.2
0.1
0.2
0.1
0.2
0.2
0.1
56
24
62
54
65
64
26
30
55
77
78
71
77
70
91
72
87
62
23
40
46
95
95
95
98
96
94
94
94
95
88
93
92
95
93
91
88
93
92
94
95
96
Figure 2. Investigation of a potential nonlinear effect for the formation of
28 fA under the conditions of Table 10, entry 6.
dA cHex Ph
fA Ph
fB Ph
fC Ph
fD Ph
Ph
o-MeOC6H4
p-MeC6H4
2-naphthyl
o-ClC6H4
m-ClC6H4
p-ClC6H4
m-BrC6H4
p-BrC6H4
o-O2NC6H4
p-O2NC6H4
m-F3CC6H4
10 fE Ph
11 fF Ph
12 fG Ph
13 fH Ph
14 fI
Ph
15 fJ Ph
16 fK Ph
17 fL Ph
18 fM Ph
19 fN Ph
20 fO Ph
21 fP Ph
p-O2NC6H4CH=CH 0.2
2-furyl
2-thiophenyl
2-(3-Br)-thiophenyl 0.2
0.1
0.2
[a] The acid chlorides were slowly added by syringe pump over 30 min
(1:1 stoichiometry of both substrates). Stirring was continued for an addi-
tional 120 min. [b] Isolated yield. [c] ee determined by HPLC.
Scheme 14. Proposed catalytic cycle for the ErIII-/amino-alcohol-catalyzed
formation of d-lactones from a,b-unsaturated acid chlorides 1 and alde-
hydes 26.
The complexation of 27b seems to proceed very rapidly,
because identical results were obtained by either preforma-
tion of the catalyst for 15 or 150 min at RT or by addition of
the ligand to the reaction mixture without any precoordina-
tion. Even if the alcohol moiety was first deprotonated with
NaH at RT for 1 h the outcome was identical in terms both
of yield and of enantioselectivity. Limited data relating to
the Er catalyst structure are available because ErIII com-
plexes are paramagnetic, thus precluding NMR investiga-
tions, and attempts to obtain X-ray quality crystals have
failed. ESI-MS experiments were performed with different
clic intermediate. Turnover is achieved by an intramolecular
acylation leading to 29. ClÀ ions generated from 1 are as-
sumed to be the reason for the need for stoichiometric
amounts of ErACTHNUTRGNEUNG
(OTf)3, because the coordination of ClÀ might
deactivate the catalyst species and additional ErIII might be
necessary as a ClÀ trap. This is supported by the fact that
ErCl3 cannot catalyze the title reaction, whereas use of mix-
tures of ErACHTUNTRGNEUNG(OTf)3 and ErCl3 results in considerably de-
Er
G
creased reactivity.
species could be identified from the MS spectra.
As discussed above, compelling evidence for the existence
of a zwitterionic dienolate intermediate in the case of the
TMSQd (4a)/SnACHTNUTRGNEUNG(OTf)2 catalyst system could be obtained by
deuteration experiments with MeOD. The formation of the
monodeuterated a,b-unsaturated ester [D1]-14 and the bis-
deuterated a,b- and b,g-unsaturated esters [D2]-14 and [D2]-
15 was interpreted in terms of an equilibrium between a di-
enolate and an acylammonium species, allowing the incorpo-
ration of a variable number of deuterium atoms. Analogous
experiments were carried out with the bifunctional ligand
ErIII is known to prefer high coordination numbers, typi-
cally seven to 10.[53] We therefore assume that the ligand
and both substrates all bind to the same metal center. The
absence of a nonlinear effect (Figure 2) indicates that higher
aggregates are most likely not involved.
The results presented in Table 10 are in accordance with a
mechanism in which the reversibly binding Lewis basic
amino site forms a nucleophilic dienolate that strongly binds
to the metal ion in 30, resulting in a highly organized transi-
tion state for
a
vinylogous aldol addition reaction
27b/ErACTHUNGTRENNU(G OTf)3 system (Table 11).
(Scheme 14).[54] A concerted [4+2] cycloaddition pathway as
suggested in Scheme 5 appears to be less likely in this case
because it would have to proceed through a strained tricy-
The reference experiment with ErIII/NEt3 (entry 1) fur-
nished isomer 15 in large excess. MS analysis of the crude
material showed the presence of mono- and polydeuterated
2512
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 2503 – 2517

Catalytic Asymmetric Formation of d-Lactones
FULL PAPER
Table 11. Deuteration experiments for the esterification of the a,b-unsaturated acid chloride 1 f.
A
correlation between the
ionic radius of the lanthanide
cation and the reaction out-
come was found (Table 12).
For lanthanides the ionic
radii decrease linearly with in-
creasing atomic numbers (“lan-
thanide contraction”).[45] Lan-
thanide cations with smaller
ionic radii than Er and Ho (en-
27b
(equiv)
Base
(equiv)
14/
[D0]-
[D1]-
[D2]-
[D3]-
[D0]-
[D1]-
[D2]-
[D3]-
15[a]
14[b]
14[b]
14[b]
14[b]
15[b]
15[b]
15[b]
15[b]
1
2
–
0.2
TEA (3)
DIEA (4) 81:19 98
7:93
68
24
2
5
0
3
0
39
32
44
59
14
7
3
2
1
[a] Determined by H NMR. [b] The percentages of nondeuterated ([D0]), monodeuterated ([D1]), bisdeuterat- tries 1–3) following Er in the
ed ([D2]), and trisdeuterated ([D3]) ester in 14 and 15 were determined by GC-MS.
Periodic Table of the elements
Table 12. Dependence of the reaction outcome on the ionic radius of the
lanthanideACTHNUTRGENUG(N III) cation.
esters for both 14 and 15. The results obtained with ligand
27b (0.2 equiv) and Er(OTf)3 (1.1 equiv) differ considerably
AHCTUNGTRENNUNG
(entry 2). The a,b-unsaturated ester 14 was formed in
excess, but the deuterium incorporation was negligible. With
regard to isomer 15, the percentage of monodeuterated
ester was increased relative to the NEt3 experiment. Signifi-
cant amounts of the dideuterated compound and small
amounts of the trideuterated compound were also detected
in the crude material. As a control experiment, a mixture of
undeuterated esters 14 and 15 was subjected to the reaction
M
Effective ionic radius[a] [ꢄ]
Yield[b] [%]
ee[c] [%]
1
2
3
4
5
6
7
8
Lu
Yb
Tm
Er
Ho
Dy
Tb
Gd
Eu
Sm
Nd
Pr
0.977
0.985
0.994
1.004
1.015
1.027
1.040
1.053
1.066
1.079
1.109
1.126
1.143
64
63
62
62
62
46
30
24
25
26
13
11
15
95
96
96
95
94
88
77
70
66
55
34
34
34
conditions [ligand 27b (0.2 equiv), ErACTHNUTRGENUG(N OTf)3 (1.1 equiv),
iPr2NEt (4 equiv)] in the presence of MeOD (2 equiv). The
esters were recovered undeuterated, showing that no deute-
rium incorporation occurs through H/D exchange after
product formation.
Because the nondeuterated a,b-unsaturated ester 14 is
formed preferentially, the reaction presumably occurs to
some degree via an acylammonium intermediate, generated
upon attack of the tertiary amine moiety of the ligand on
the Lewis-acid-activated acid chloride 32 (Scheme 15).
9
10
11
12
13
Ce
[a] For M3+ with coordination number 8. [b] Yield determined from the
mass of the crude product in combination with 1H NMR with MeNO2 as
internal standard. [c] ee determined by HPLC.
all displayed catalytic activity similar to that of ErACHTNUTRGNEUNG(OTf)3, af-
fording the d-lactone 28 fA with comparable yields and
enantioselectivities. In contrast, starting from Dy (entries 6–
13), as the ionic radius of the metal center increased, a dra-
matic reduction in the reactionꢀs efficiency and the ee values
was observed, until a plateau was reached with NdIII triflate
(entries 11–13). The tendency is illustrated in Figure 3.
The results with the pseudolanthanide triflates of the
group III elements Sc, Y, and La (Table 13) are in agree-
ment with these observations. The relatively small ScIII fur-
nished the desired product with low yield and enantioselec-
tivity (entry 1), whereas YIII performed similarly to DyIII
(entry 2), which has a comparable ionic radius. In contrast,
the larger LaIII failed to afford any product (entry 3).
Scheme 15. General mechanistic alternatives, either through an acylam-
monium intermediate or through a vinylketene.
Presumably, a relatively small ionic radius of the lantha-
nide is required for the formation of a rigid transition state
in which both reaction components (aldehyde and dienolate
in 30; Scheme 14) are arranged in close proximity. Looser
transition states are probably obtained with the larger lan-
thanides, resulting in lower stereocontrol and disfavoring
The acylammonium salt might be trapped either by
MeOD or, more likely, by methoxide linked to the oxophilic
Er (intermediate 33). The significant amount of bisdeuterat-
ed isomer [D2]-15 can only be explained by formation of di-
enolate 34.
Chem. Eur. J. 2010, 16, 2503 – 2517
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2513

R. Peters and P. S. Tiseni
coordination both of the alkoxide moiety and of the tertiary
amino group, whereas in the absence of base coordination is
presumably relatively labile. Addition of benzaldehyde does
not change the ligand and aldehyde spectrum, showing that
the lifetime of the corresponding Lu/aldehyde complex is
most likely very brief on the NMR timescale. However,
complexation of benzaldehyde was established in the case of
the paramagnetic ErIII, resulting under the same conditions
in a shift from 9.97 to 9.27 ppm at À208C in CDCl3.
An obvious limitation of the bifunctional ErIII/Lewis base
catalyst system is the need for stoichiometric amounts of the
Lewis acid in order to obtain preparatively useful yields. Of
the investigated alternative lanthanides, ytterbium[56] showed
a peculiar activity, as shown in Table 14.
Figure 3. Influence of the ionic radius of LnIII on the title reaction [data
from Table 12; square: ee (%), triangle: yield (%)].
Table 13. Dependence of the reaction outcome on the ionic radii of
pseudolanthanideACHTUNGTRENNUNG(III) cations.
Table 14. Effect of the amount of YbACTHNURGTNEUNG(OTf)3.
M
Effective ionic radius[a] [ꢄ]
Yield[b] [%]
ee[c] [%]
X
Yield[a] [%]
ee[b] [%]
1
2
3
Sc
Y
La
0.870
1.019
1.160
33
46
0
51
92
–
1
2
3
4
5
6
1.5
1.1
0.8
0.6
0.4
0.2
65
63
62
62
54
19
96
96
95
93
87
73
[a] For M3+ with coordination number 8. [b] Yield determined from the
mass of the crude product in combination with 1H NMR with MeNO2 as
internal standard. [c] ee determined by HPLC.
[a] Yield determined from the mass of the crude product in combination
with 1H NMR with MeNO2 as internal standard. [b] ee determined by
HPLC.
the product formation. Another parameter associated with
the ionic radius is the coordination number (CN) of the
metal. It might be expected that fewer ligands should be
packed around the central metal ion as the ionic radius de-
creases. In fact, such a tendency is indeed observed for some
classes of lanthanide complexes
The amount of metal triflate could be reduced to
0.6 equiv without loss of catalytic activity (entries 1–4),
(e.g. the hydrated Ln3+ ions),
but several exceptions to this
rule are known and the differ-
ence in the CN across the series
of lanthanide complexes is usu-
ally small (8 vs. 9 or 10).[55]
NMR complexation studies
were performed with the dia-
magnetic Lu3+ (Figure 4). Ad-
dition of the triflate salt
(1 equiv) to the amino alcohol
resulted in a shift of the ligand
signals and broadened peaks.
Upon addition of Hꢁnigꢀs base
(1 equiv) the signals became
sharp again and protonation of
base was observed. The two sig-
nals for the diastereotopic
ligand NCH2 protons were split
and the new signals were signif-
Figure 4. 1H NMR complexation studies in [D8]THF at RT. Spectrum 1: ligand 27b. Spectrum 2: ligand
27b+LuIII triflate (1:1). Spectrum 3: ligand 27b+LuIII triflate+iPr2NEt (1:1:1). Spectrum 4: ligand 27b+LuIII
icantly shifted, indicating the triflate+iPr2NEt+PhCHO (1:1:1:1).
2514
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 2503 – 2517

Catalytic Asymmetric Formation of d-Lactones
FULL PAPER
the system had been cooled to À158C, a solution of chloral (5, 97 mL,
1 mmol) in toluene (2 mL) was added. After an additional 10 min a solu-
tion of the corresponding acid chloride 1 (1 mmol) in toluene (2 mL) was
added by syringe pump over 120 min. The reaction mixture was allowed
to stir for an additional 3 h, and aqueous HCl (1n, 6 mL) was then added
to quench the reaction. MTBE (20 mL) was added and the organic phase
was washed with aqueous HCl (1n, 2ꢅ10 mL) and with brine (10 mL).
After drying over MgSO4 and filtration, the solvent was removed under
reduced pressure.
whereas use of 0.4 equiv resulted in a somewhat lower yield
and enantioselectivity (entry 5).[57–58] Use of lower amounts
resulted in poor yields and moderate ee values (entry 6).
The new reaction seems to be specific for metalACTHNUTRGENUG(N III) tri-
flates of the (pseudo)lanthanide series. For instance, the cor-
responding salts of Sr2+, Ba2+, Hf4+ (also having noble gas
configuration and capable of adopting high coordination
numbers), Al3+, In3+, Sn2+, Cu2+, or Zn2+ failed to give any
d-lactone product in combination with ligand 27b.
General procedure for the formation of the a,b-unsaturated d-lactones
28 from aldehydes 26: Dry Er
tion flask in a glove-box. THF (0.9 mL), toluene (0.8 mL), and a solution
of (1R,2S)-1-phenyl-2-(pyrrolidin-1-yl)propan-1-ol (27b, 0.034–
ACHTUNGRTEN(NUNG OTf)3 (0.51 mmol) was placed in the reac-
0.068 mmol) in THF (0.3 mL) were added and the mixture was stirred
for 15 min at room temperature. After the system had been cooled to
À108C, iPr2NEt (0.85 mmol) and the corresponding aldehyde
(0.34 mmol) were successively added. A solution of the corresponding
acid chloride 1 (0.34 mmol) in toluene (0.5 mL) was then added by sy-
ringe pump over 30 min. After stirring for an additional 2 h at À108C,
the reaction mixture was filtered through a short plug of silica gel (2 cm,
hexanes/ethyl acetate 1:1). The solvent was removed under reduced pres-
sure and the residue was purified by flash chromatography (hexanes/
ethyl acetate).
Conclusion
We have developed tertiary-amine-catalyzed enantioselec-
tive [4+2] cycloadditions of a,b-unsaturated acid chlorides 1
and the electron-poor aldehyde chloral (5), which proceeded
through the formation of previously unexplored enantiopure
zwitterionic ammonium dienolates. The substituent at the b-
position in compounds 1 could be varied to a large degree,
and the trichloromethyl group allowed several useful func-
tional groups to be installed at the d-position. b-Hydroxy-d-
lactones possessing O-substituted quaternary stereocen-
ters[59] at the b-position were diastereoselectively synthe-
sized through 1,2-migration reactions starting from the d-
lactones 6g–j containing trialkylsilyl substituents at the b-
position. The scope of the cycloadditions was considerably
improved by use of a novel ErIII/Lewis base catalyst system
that tolerated a broad range of aromatic and heteroaromatic
aldehydes, providing direct access to d-lactone building
blocks with generally excellent enantioselectivities. Our re-
sults show that: a) ErIII and the amino alcohol ligand form
Acknowledgements
This work was financially supported by Novartis (one year Ph.D. fellow-
ship to P.S.T.), ETH Zꢁrich, F. Hoffmann–La Roche, and the Universitꢂt
Stuttgart. We thank Nicolas Dietl (TU Berlin) for technical assistance
(semester internship at ETHZ).
[1] Review about catalytic asymmetric hetero Diels–Alder reactions:
K. A. Jørgensen, Angew. Chem. 2000, 112, 3702; Angew. Chem. Int.
Ed. 2000, 39, 3558.
[2] M. Bednarski, S. Danishefsky, J. Am. Chem. Soc. 1983, 105, 6968.
[3] a) Q. Gao, T. Maruyama, M. Mouri, H. Yamamoto, J. Org. Chem.
1992, 57, 1951; b) E. J. Corey, C. L. Cywin, T. D. Roper, Tetrahedron
Lett. 1992, 33, 6907.
[4] a) K. Maruoka, T. Itoh, T. Shirasaka, H. Yamamoto, J. Am. Chem.
Soc. 1988, 110, 310; b) K. Maruoka, H. Yamamoto, J. Am. Chem.
Soc. 1989, 111, 789.
[5] G. E. Keck, X. Y. Li, D. Krishnamurthy, J. Org. Chem. 1995, 60,
5998.
[6] S. E. Schaus, J. Branalt, E. N. Jacobsen, J. Org. Chem. 1998, 63, 403.
[7] a) A. G. Dossetter, T. F. Jamison, E. N. Jacobsen, Angew. Chem.
1999, 111, 2549; Angew. Chem. Int. Ed. 1999, 38, 2398; b) E. R.
Jarvo, B. M. Lawrence, E. N. Jacobsen, Angew. Chem. 2005, 117,
6197; Angew. Chem. Int. Ed. 2005, 44, 6043.
[8] Y. Yamashita, S. Saito, H. Ishitani, S. Kobayashi, J. Am. Chem. Soc.
2003, 125, 3793.
[9] J. Long, J. Hu, X. Shen, B. Ji, K. Ding, J. Am. Chem. Soc. 2002, 124,
10.
À
an inert Er O bond precluding O-acylation, b) both the
Lewis acid and the nucleophilic amino moiety are essential
for product formation, and c) the catalyst is most likely a
monomeric species. Mechanistic studies confirmed the for-
mation of the dienolate intermediates for both catalytic sys-
tems. Interestingly, all lanthanides were found to be capable
of promoting the title reaction, but the efficiency in terms of
yield and enantioselectivity was directly dependent on the
radius of the LnIII ion. Similarly, use of the pseudolantha-
nides ScIII and YIII resulted in product formation, whereas
that of the larger LaIII and other transition metal salts, as
well as main group metal salts, proved to be inefficient. A
key characteristic of the new bifunctional catalyst system is
its simplicity, because the commercially available nucleophil-
ic amino alcohol ligand can be prepared from inexpensive
norephedrine[60] in a single step.[61]
[10] J. Savard, P. Brassard, Tetrahedron Lett. 1979, 20, 4911.
[11] Y. Kato, Y. Ogawa, T. Imada, S. Iwasaki, N. Shimazaki, T. Kobaya-
shi, T. Komai, J. Antibiot. 1991, 44, 66.
[12] F. E. Boyer, J. V. N. V. Prasad, J. M. Domagala, E. L. Ellsworth, C.
Gajda, S. E. Hagen, L. J. Markoski, B. D. Tait, E. A. Lunney, A. Pal-
ovsky, D. Ferguson, N. Graham, T. Holler, D. Hupe, C. Nouhan, P. J.
Tummino, A. Urumov, E. Zeikus, G. Zeikus, S. J. Gracheck, J. M.
Sanders, S. VanderRoest, J. Brodfuehrer, K. Iyer, M. Sinz, S. V.
Gulnik, J. W. Erickson, J. Med. Chem. 2000, 43, 843.
Experimental Section
General information, additional procedures, NMR spectra, and HPLC
chromatograms are given in the Supporting Information.
General procedure for the formation of the chloral-derived a,b-unsatu-
rated d-lactones 6: Dry Sn
(OTf)2 (41.7 mg, 0.1 mmol) was placed in the
[13] a) M. Kobayashi, K. Higuchi, N. Murakami, H. Tajima, S. Aoki, Tet-
rahedron Lett. 1997, 38, 2859; b) X.-P. Fang, J. E. Andersen, C.-J.
Chang, J. L. McLaughlin, J. Nat. Prod. 1991, 54, 1034; c) D. S. Lewy,
C.-M. Gauss, D. R. Soenen, D. L. Boger, Curr. Med. Chem. 2002, 9,
reaction flask in a glove-box. Subsequently, toluene (3.6 mL), a solution
of trimethylsilylquinidine (TMSQd, 4a, 79.3 mg, 0.2 mmol) in toluene
(2.4 mL), and iPr2NEt (331 mL, 2.0 mmol) were successively added. After
Chem. Eur. J. 2010, 16, 2503 – 2517
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2515

R. Peters and P. S. Tiseni
2005; d) S. J. Shaw, K. F. Sundermann, M. A. Burlingame, D. C.
Myles, B. S. Freeze, M. Xian, I. Brouard, A. B. Smith III, J. Am.
Chem. Soc. 2005, 127, 6532; e) R. Peters, M. Althaus, A.-L. Nagy,
Org. Biomol. Chem. 2006, 4, 498.
Dreiding, Helv. Chim. Acta 1983, 66, 2330; g) D. A. Jackson, M.
Rey, A. S. Dreiding, Tetrahedron Lett. 1983, 24, 4817; h) L. Lombar-
do, Tetrahedron Lett. 1985, 26, 381; i) K. Ando, E. Tsuji, Y. Ando, N.
Kuwata, J. Kunitomo, M. Yamashita, S. Ohta, S. Kohno, Y. Ohishi,
Org. Biomol. Chem. 2004, 2, 625.
[14] K. Yasui, Y. Tamura, T. Nakatani, K. Kawada, M. Ohtani, J. Org.
Chem. 1995, 60, 7567.
[15] A. Endo, J. Med. Chem. 1985, 28, 401.
[27] Review articles: a) S. France, D. J. Guerin, S. J. Miller, T. Lectka,
Chem. Rev. 2003, 103, 2985; b) M. J. Gaunt, C. C. C. Johansson,
Chem. Rev. 2007, 107, 5596.
[16] E. S. Istvan, J. Deisenhofer, Science 2001, 292, 1160.
[17] J. Mulzer, E. ꢆhler, Chem. Rev. 2003, 103, 3753.
[28] a) H. Pracejus, Justus Liebigs Ann. Chem. 1960, 634, 9; b) H. Prace-
jus, Fortschr. Chem. Forsch. 1967, 8, 493; c) H. Pracejus, G. Kohl,
Justus Liebigs Ann. Chem. 1969, 722, 1.
[29] The combination of cinchona alkaloid derivatives and metal triflate
salts was previously successfully applied to [2+2]-cycloadditions of
ketenes. See: S. France, M. H. Shah, A. Weatherwax, H. Wack, J. P.
Roth, T. Lectka, J. Am. Chem. Soc. 2005, 127, 1206 and [22e].
[30] B. G. Jackson, Swiss Patent CH681302A5, 1993.
[18] a) Q. Fan, L. Lin, J. Liu, Y. Huang, X. Feng, G. Zhang, Org. Lett.
2004, 6, 2185; b) L. Lin, Q. Fan, B. Qin, X. Feng, J. Org. Chem.
2006, 71, 4141; c) L. Lin, Z. Chen, X. Yang, X. Liu, X. Feng, Org.
Lett. 2008, 10, 1311; for an alternative through a vinylogous Mu-
kaiyama aldol reaction, see: d) G. Bluet, B. Bazꢇn-Tejeda, J. M.
Campagne, Org. Lett. 2001, 3, 3807; e) X. Moreau, B. Bazꢇn-Tejeda,
J. M. Campagne, J. Am. Chem. Soc. 2005, 127, 7288.
[19] For the described example and selected references about hydrogen-
bonding activation, see a) H. Du, D. Zhao, K. Ding, Chem. Eur. J.
2004, 10, 5964. For computational studies and selected references re-
lated to this topic, see: b) X. Zhang, H. Du, Z. Wang, Y. Wu, K.
Ding, J. Org. Chem. 2006, 71, 2862.
[31] T. Fujisawa, T. Ito, K. Fujimoto, M. Shimizu, H. Wynberg, E. G. J.
Staring, Tetrahedron Lett. 1997, 38, 1593.
[32] B. M. Trost, Z. T. Ball, J. Am. Chem. Soc. 2004, 126, 13942.
[33] Theoretical studies of HDA reactions have revealed that the mecha-
nism can change from a concerted non-synchronous to a stepwise
mechanism depending on the substituents on the reacting species
and on the reaction conditions. See for example: a) L. F. Tietze, J.
Fennen, E. Anders, Angew. Chem. 1989, 101, 1420; Angew. Chem.
Int. Ed. Engl. 1989, 28, 1371; b) M. A. McCarrick, Y.-D. Wu, K. N.
Houk, J. Am. Chem. Soc. 1992, 114, 1499; c) M. A. McCarrick, Y.-D.
Wu, K. N. Houk, J. Org. Chem. 1993, 58, 3330; d) B. S. Jursic, Z.
Zdravkovski, J. Phys. Org. Chem. 1994, 7, 641; e) L. F. Tietze, A.
Schuffenhauer, P. R. Schreiner, J. Am. Chem. Soc. 1998, 120, 7952.
[34] a) H. Wynberg, E. G. J. Staring, J. Chem. Soc. Chem. Commun.
1984, 1181; b) see ref. [22n].
[35] a) C. E. Song, J. K. Lee, S. H. Lee, S. Lee, Tetrahedron: Asymmetry
1995, 6, 1063; b) F. M. Koch, R. Peters, Synlett 2008, 1505.
[36] For the synthetic utility of dichloromethyl groups see e.g.: a) K.-i.
Sato, S. Akai, N. Sugita, T. Ohsawa, T. Kogure, H. Shoji, J. Yoshi-
mura, J. Org. Chem. 2005, 70, 7496; b) H. Arasaki, M. Iwata, D.
Nishimura, A. Itoh, Y. Masaki, Synlett 2004, 546; c) M. Shimizu, T.
Fujimoto, X. Liu, H. Minezaki, T. Hata, T. Hiyama, Tetrahedron
2003, 59, 9811.
[20] G. B. Payne, J. Org. Chem. 1969, 34, 1341.
[21] Preliminary results have been published as communication: P. S.
Tiseni, R. Peters, Angew. Chem. 2007, 119, 5419; Angew. Chem. Int.
Ed. 2007, 46, 5325.
[22] Selected examples: a) D. Borrmann, R. Wegler, Chem. Ber. 1967,
100, 1575; b) H. Wynberg, E. G. J. Staring, J. Am. Chem. Soc. 1982,
104, 166; c) R. Tennyson, D. Romo, J. Org. Chem. 2000, 65, 7248;
d) C. Zhu, X. Shen, S. G. Nelson, J. Am. Chem. Soc. 2004, 126, 5352;
e) M. A. Calter, O. A. Tretyak, C. Flaschenriem, Org. Lett. 2005, 7,
1809; reviews on the use of ketenes in asymmetric synthesis:
f) R. K. Orr, M. A. Calter, Tetrahedron 2003, 59, 3545; g) D. H.
Paull, A. Weatherwax, T. Lectka, Tetrahedron 2009, 65, 6771; re-
views on the history of ketene chemistry: h) T. T. Tidwell, Eur. J.
Org. Chem. 2006, 563; i) T. T. Tidwell, Angew. Chem. 2005, 117,
5926; Angew. Chem. Int. Ed. 2005, 44, 5778; use of acyl bromide
enolates: j) T. Kull, R. Peters, Angew. Chem. 2008, 120, 5541;
Angew. Chem. Int. Ed. 2008, 47, 5461; use of sulfonyl chlorides in-
stead of ketenes or acyl halides: k) M. Zajac, R. Peters, Chem. Eur.
J. 2009, 15, 8204; l) M. Zajac, R. Peters, Org. Lett. 2007, 9, 2007;
m) F. M. Koch, R. Peters, Angew. Chem. 2007, 119, 2739; Angew.
Chem. Int. Ed. 2007, 46, 2685; n) F. M. Koch, R. Peters, Synlett 2008,
1505.
[37] I. Fleming, T. W. Newton, V. Sabin, F. Zammattio, Tetrahedron 1992,
48, 7793.
[38] M. E. Jung, G. Piizzi, J. Org. Chem. 2002, 67, 3911.
[39] L. A. Aronica, F. Morini, A. M. Caporosso, P. Salvatori, Tetrahedron
Lett. 2002, 43, 5813.
[23] a) R. L. Danheiser, H. Sard, J. Org. Chem. 1980, 45, 4810; b) J. M.
Berge, M. Rey, A. S. Dreiding, Helv. Chim. Acta 1982, 65, 2230;
c) W. S. Trahanovsky, B. W. Surber, M. C. Wilkes, M. Preckel, J. Am.
Chem. Soc. 1982, 104, 6779; d) G. Barbaro, A. Battaglia, P. Gior-
gianni, J. Org. Chem. 1987, 52, 3289; e) R. B. Gammill, T. M. Judge,
G. Philips, Q. Zhang, C. G. Sowell, B. W. Cheney, S. A. Mizsak,
L. A. Dolak, E. P. Seest, J. Am. Chem. Soc. 1994, 116, 12113; f) D.
Collomb, A. Doutheau, Tetrahedron Lett. 1997, 38, 1397; g) J. F.
Loebach, D. M. Bennett, R. L. Danheiser, J. Org. Chem. 1998, 63,
8380; h) D. M. Bennett, I. Okamoto, R. L. Danheiser, Org. Lett.
1999, 1, 641; vinylketene imine [4+2] cycloaddition: i) E. Sonveaux,
L. Ghosez, J. Am. Chem. Soc. 1973, 95, 5417; allenylketene [4+2] cy-
cloaddition: j) W. H. Huang, T. T. Tidwell, Synthesis 2000, 457.
[24] T. T. Tidwell, Ketenes, 2nd ed., Wiley, New York, 2006.
[25] 2-(Trimethylsilyl)vinylketene has been reported to be a remarkable
stable vinylketene and has been used for non-enantioselective HDA
reactions, see ref. [23h]. Attempts to extend the reaction scope to
unactivated carbonyl compounds failed even employing Lewis acids
to promote the cycloaddition, which only caused decomposition of
the ketene.
[40] Review: G. Jones, Y. Landais, Tetrahedron 1996, 52, 7599.
[41] R. Verhꢈ, N. De Kimpe, L. De Buyck, N. Schamp, Synthesis 1975,
455.
[42] With a residue R2 in 24 bulkier than Me, enantioselectivity de-
creased to a large degree (e.g., R2 =Et: 60% ee with 1a; R2 =nPr:
32% ee with 1a). Similarly, glyoxylates and other 1,2-dicarbonyl di-
enophiles resulted in poor enantioselectivity.
[43] For the application of bifunctional catalysts in [2+2] cycloadditions
of ketenes, see for example: reference [22d,e,29] and a) V. Gnanade-
sikan, E. J. Corey, Org. Lett. 2006, 8, 4943; b) Y.-M. Lin, J. Boucau,
Z. Li, V. Casarotto, J. Lin, A. N. Nguyen, J. Ehrmantraut, Org. Lett.
2007, 9, 567.
[44] Dual activation catalysis review: J.-A. Ma, D. Cahard, Angew.
Chem. 2004, 116, 4666; Angew. Chem. Int. Ed. 2004, 43, 4566.
[45] Review about the use of lanthanides in asymmetric catalysis: K.
Mikami, M. Terada, H. Matsuzawa, Angew. Chem. 2002, 114, 3704;
Angew. Chem. Int. Ed. 2002, 41, 3554.
[46] For lanthanide complexes acting as bifunctional catalysts, see, for
example: a) M. Shibasaki, H. Sasai, T. Arai, Angew. Chem. 1997,
109, 1290; Angew. Chem. Int. Ed. Engl. 1997, 36, 1236; b) M. Shiba-
saki, M. Kanai, K. Funabashi, Chem. Commun. 2002, 1989; c) M.
Kanai, N. Kato, E. Ichikawa, M. Shibasaki, Synlett 2005, 1491; d) M.
Shibasaki, S. Matsunaga, Chem. Soc. Rev. 2006, 35, 269.
[26] a) D. H. R. Barton, G. Quinkert, J. Chem. Soc. 1960, 1; b) P. W.
Hickmott, J. R. Hargreaves, Tetrahedron 1967, 23, 3151; c) W. E.
Truce, P. S. Bailey Jr. , J. Org. Chem. 1969, 34, 1341; d) P. W. Hick-
mott, G. J. Miles, G. Sheppard, R. Urbani, C. T. Yoxall, J. Chem.
Soc. Perkin Trans. 1 1973, 1514; e) R. Gelin, S. Gelin, R. Dolmazon,
Bull. Soc. Chim. Fr. 1973, 1409; f) D. A. Jackson, M. Rey, A. S.
[47] Preliminary results have been published as communication: P. S.
Tiseni, R. Peters, Org. Lett. 2008, 10, 2019.
2516
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 2503 – 2517

Catalytic Asymmetric Formation of d-Lactones
FULL PAPER
[48] A. Bellemare, Prog. Quantum Electron. 2003, 27, 211.
[49] R. D. Shannon, Acta Crystallogr. Sect. A 1976, 32, 751.
[50] Slow addition of the acid chloride substrates over 30 min is recom-
mended to maintain a low vinylketene concentration so as to mini-
mize di- and oligomerization.
[51] The absolute configuration was determined by chemical correlation
using a similar synthetic route as the one described in Scheme 3 (see
the Supporting Information).
[52] The enolizable dihydrocinnamaldehyde and cyclohexylcarbaldehyde
as well as the non-enolizable pivaldehyde resulted in complete alde-
hyde decomposition.
[53] Selected examples: a) Y. Ren, S. Chen, G. Xie, S. Gao, Q. Shi,
Inorg. Chim. Acta 2006, 359, 2047; b) R. O. Freire, E. V. do Monte,
G. B. Rocha, A. M. Simas, J. Organomet. Chem. 2006, 691, 2584.
[54] S. E. Denmark, J. R. Heemstra Jr. , G. L. Beutner, Angew. Chem.
2005, 117, 4760; Angew. Chem. Int. Ed. 2005, 44, 4682.
[57] Due to its employment in optical fibers, ErIII is currently about 4–5
times less expensive than YbIII
.
[58] [YbACHTNUGTRNEGNU(NTf2)3] gave no conversion, probably because of a poor solubil-
ity in the THF/toluene solvent system.
[59] For other approaches by our group to selectively form quaternary
stereocenters, see: a) D. F. Fischer, A. Barakat, Z.-q. Xin, M. E.
Weiss, R. Peters, Chem. Eur. J. 2009, 15, 8722; b) S. Jautze, R.
Peters, Angew. Chem. 2008, 120, 9424; Angew. Chem. Int. Ed. 2008,
47, 9284; c) Z.-q. Xin, D. F. Fischer, R. Peters, Synlett 2008, 1495;
d) D. F. Fischer, Z.-q. Xin, R. Peters, Angew. Chem. 2007, 119, 7848;
Angew. Chem. Int. Ed. 2007, 46, 7704; e) R. Peters, C. Diolez, A.
Rolland, E. Manginot, M. Veyrat, Heterocycles 2007, 72, 255; f) R.
Peters, M. Althaus, C. Diolez, A. Rolland, E. Manginot, M. Veyrat,
J. Org. Chem. 2006, 71, 5783; g) see ref. [13e].
[60] Both enantiomers of norephedrine or 27b are commercially avail-
able at a comparable price.
[61] a) J. Kang, J. W. Lee, J. I. Kim, J. Chem. Soc. Chem. Commun. 1994,
2009; b) D. Zhao, C.-Y. Chen, F. Xu, L. Tan, R. Tillyer, M. E.
Pierce, J. R. Moore, Org. Synth. 2000, 77, 12.
[55] S. Cotton, Lanthanide and Actinide Chemistry, Wiley, New York,
2006, and references therein.
[56] A mixture of YbIII and ligand 27b has been described for asymmet-
ric aldol-Tishchenko reactions: J. Mlynarski, B. Rakiel, M. Stodulski,
A. Suszczynska, J. Frelek, Chem. Eur. J. 2006, 12, 8158.
Received: October 20, 2009
Published online: January 14, 2010
Chem. Eur. J. 2010, 16, 2503 – 2517
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2517