Mechanistic investigation of chiral phosphoric acid catalyzed asymmetric baeyer-villiger reaction of 3-substituted cyclobutanones with H 2O2 as the oxidant

Article abstract of DOI:10.1002/chem.200902698

The mechanism of the chiral phosphoric acid catalyzed Baeyer-Villiger (B-V) reaction of cyclobutanones with hydrogen peroxide was investigated by using a combination of experimental and theoretical methods. Of the two pathways that have been proposed for the present reaction, the pathway involving a peroxyphosphate intermediate is not viable. The reaction progress kinetic analysis indicates that the reaction is partially inhibited by the y-lactone product. Initial rate measurements suggest that the reaction follows Michaelis-Menten-type kinetics consistent with a bifunctional mechanism in which the catalyst is actively involved in both carbonyl addition and the subsequent rearrangement steps through hydrogen-bonding interactions with the reactants or the intermediate. High-level quantum chemical calculations strongly support a two-step concerted mechanism in which the phosphoric acid activates the reactants or the intermediate in a synergistic manner through partial proton transfer. The catalyst simultaneously acts as a general acid, by increasing the electrophilicity of the carbonyl carbon, increases the nucleophilicity of hydrogen peroxide as a Lewis base in the addition step, and facilitates the dissociation of the OH group from the Criegee intermediate in the rearrangement step. The overall reaction is highly exothermic, and the rearrangement of the Criegee intermediate is the rate-determining step. The observed reactivity of this catalytic B-V reaction also results, in part, from the ring strain in cyclobutanones. The sense of chiral induction is rationalized by the analysis of the relative energies of the competing diastereomeric transition states, in which the steric repulsion between the 3-substituent of the cyclobutanone and the 3- and 3'-substituents of the catalyst, as well as the entropy and solvent effects, are found to be critically important.

Full text of DOI:10.1002/chem.200902698

FULL PAPER
DOI: 10.1002/chem.200902698
Mechanistic Investigation of Chiral Phosphoric Acid Catalyzed Asymmetric
Baeyer–Villiger Reaction of 3-Substituted Cyclobutanones
with H2O2 as the Oxidant
Senmiao Xu,[a] Zheng Wang,[a] Yuxue Li,*[a] Xumu Zhang,[b] Haiming Wang,[a] and
Kuiling Ding*[a]
Abstract: The mechanism of the chiral
phosphoric acid catalyzed Baeyer–Vil-
liger (B–V) reaction of cyclobutanones
with hydrogen peroxide was investigat-
ed by using a combination of experi-
mental and theoretical methods. Of the
two pathways that have been proposed
for the present reaction, the pathway
through hydrogen-bonding interactions
with the reactants or the intermediate.
High-level quantum chemical calcula-
tions strongly support a two-step con-
certed mechanism in which the phos-
phoric acid activates the reactants or
tion step, and facilitates the dissocia-
tion of the OH group from the Criegee
intermediate in the rearrangement
step. The overall reaction is highly exo-
thermic, and the rearrangement of the
Criegee intermediate is the rate-deter-
mining step. The observed reactivity of
this catalytic B–V reaction also results,
in part, from the ring strain in cyclobu-
tanones. The sense of chiral induction
is rationalized by the analysis of the
relative energies of the competing dia-
stereomeric transition states, in which
the steric repulsion between the 3-sub-
stituent of the cyclobutanone and the
3- and 3’-substituents of the catalyst, as
well as the entropy and solvent effects,
are found to be critically important.
the intermediate in
a
synergistic
manner through partial proton transfer.
The catalyst simultaneously acts as a
general acid, by increasing the electro-
philicity of the carbonyl carbon, in-
creases the nucleophilicity of hydrogen
peroxide as a Lewis base in the addi-
involving
a
peroxyphosphate inter-
mediate is not viable. The reaction
progress kinetic analysis indicates that
the reaction is partially inhibited by
the g-lactone product. Initial rate
measurements suggest that the reaction
follows Michaelis–Menten-type kinetics
consistent with a bifunctional mecha-
nism in which the catalyst is actively in-
volved in both carbonyl addition and
the subsequent rearrangement steps
Keywords: asymmetric catalysis
·
Baeyer–Villiger reaction · density
functional calculations · kinetics ·
organocatalysis
Introduction
The Baeyer–Villiger (B–V) oxidation, which converts ke-
tones into the corresponding esters or lactones, is of great
importance in organic synthesis due to its predictable regio-
and stereoselectivity.[1] Since its discovery by Baeyer and
Villiger in 1899,[2] considerable progress has been made to
understand the mechanism of the reaction and to use this
reaction in preparative chemistry.[3–6] In the vast majority of
cases, B–V reactions have been performed with an organic
peracid, such as m-chloroperbenzoic acid (mCPBA), as the
oxidant. However, in some cases, hydrogen peroxide has
been successfully employed in combination with a catalyst;
this offers a number of advantages in terms of environmen-
tal and economical concerns. For the B–V oxidation with a
peroxycarboxylic acid in the absence of acid/base catalysis,
it is generally accepted that the reaction proceeds through a
[a] Dr. S. Xu, Prof. Dr. Z. Wang, Prof. Dr. Y. Li, H. Wang,
Prof. Dr. K. Ding
State Key Laboratory of Organometallic Chemistry
Shanghai Institute of Organic Chemistry
Chinese Academy of Sciences, 354 Fenglin Road
Shanghai 200032 (P.R. China)
Fax : (+86)21-6416-6128
E-mail: kding@mail.sioc.ac.cn
liyuxue@mail.sioc.ac.cn
[b] Prof. Dr. X. Zhang
Department of Chemistry and Chemical Biology
Center of Molecular Catalysis, Rutgers
The State University of New Jersey, 610 Taylor
Piscataway, New Jersey 08854 (USA)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200902698.
Chem. Eur. J. 2010, 16, 3021 – 3035
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3021

two-step mechanism. Initially, the peracid attacks the car-
bonyl carbon of the ketone or aldehyde, leading to the for-
mation of a tetrahedral perhemiketal species, known as the
Criegee intermediate.[7] Then, the intermediate undergoes
an intramolecular migration of an alkyl or aryl substituent
from the central carbon to the proximal oxygen of the per-
indicated that the catalyzed reactions proceed through a
Criegee intermediate that contains a five-membered chelate
ring involving the tin center, which effectively activates both
reactants in the addition step and facilitates the departure of
hydroxyl group in the following rearrangement. Corma and
co-workers have studied the B–V oxidation of cyclohexa-
none with hydrogen peroxide catalyzed by the Sn beta-zeo-
lite using a combination of molecular mechanics, quantum
chemical calculations, and spectroscopic and kinetic tech-
niques.[26] The zeolite active site was found to consist of two
catalytic centers: the Lewis acidic Sn atom for activation of
cyclohexanone by coordination, and an adjacent basic
ACHTUNGTRENNUNGester moiety, resulting in the production of the ester (or lac-
tone) and a carboxylic acid. The B–V oxidation with hydro-
gen peroxide is also often hypothesized to proceed in a simi-
lar manner, but the reaction needs some kind of catalyst for
sufficient reactivity and selectivity, and various catalysts
have been used, including homogeneous Brønsted acids,[8–12]
heterogeneous solid acids,[13] flavin-based organocata-
lysts,[14,15] Lewis acidic transition-metal complexes,[5] and
Lewis acidic metal incorporated into molecular sieves.[6] De-
spite such a general understanding, however, in many cases
the detailed mechanism of the reaction has been found to
be unclear and some key issues are still debated. For exam-
ple, the migration has often been proposed to be the rate-
determining step, which involves a concerted formation of a
ꢀ
oxygen atom of the Sn OH group for interaction with H2O2
by hydrogen-bond formation. It should be noted that from
these studies valuable insights can be obtained on the mo-
lecular pathways of the reactions, however, the lack of ex-
perimental data on many calculated systems might also
hinder further mechanistic understanding and corrobora-
tion.
On the other hand, catalytic asymmetric B–V oxidation of
cyclic ketones provides a rapid access to various chiral lac-
tones, which are valuable synthetic intermediates and have
found widespread biochemical applications. Thus, it is not
surprising to see that a continuous effort has been made to
search for more efficient chiral catalysts, either natural or
synthetic, to accomplish this reaction in an enantioselective
manner.[28] In this arena, enzymatic systems are highly enan-
tioselective, but, by their nature, exhibit substrate specifici-
ty.[29,30] Chemical catalysts offer an attractive alternative, es-
pecially reactions using environmentally benign oxidants
such as H2O2, but they are still limited in number and are
often less enantioselective than their biological counter-
parts.[5] Since the first reports on metal-catalyzed asymmet-
ric B–V reactions in 1994,[31] various chiral metal catalysts
based on CuII,[32] PtII,[33] TiIV,[34] CoIII,[35] ZrIV,[36] MgII,[37]
AlIII,[38] and PdII[39] have been developed in recent years for
this transformation. However, so far, very few of them could
attain enantioselectivity at a level comparable to enzymes,
and wide substrate generality has not yet been achieved in
each case. Complementary to metal catalysis, several chiral
organocatalysts based on organoseleniums,[40] flavins,[15]
amino acids,[41] and organophosphoric acids[42] have also
been demonstrated to be a useful and green alternative for
the conversion. With regards to mechanistic aspects of the
catalytic B–V reactions, Strukul and co-workers have inves-
tigated some PtII and PdII complexes of achiral diphosphines
using a combination of kinetic and spectroscopic tech-
niques.[43,44] However, to the best of our knowledge, the
chiral Brønsted acid catalysis of B–V reactions has never
been studied in great detail, and there is no computational
study that has considered catalytic enantioselective B–V re-
actions.
ꢀ
ꢀ
C O bond with the simultaneous cleavage of the O O
bond. However, some studies have shown that the carbonyl
addition can also be rate determining and can depend on
the reaction conditions as well as the reactants.[3,16–18] For
the carbonyl addition step, there is still no consensus on
whether the protonation of the carbonyl oxygen and the car-
bonyl addition occur in a stepwise or concerted manner.[19]
Besides, there are some controversies with regards to the
potential effect of the acid, which is generated as a byprod-
uct during the reaction and may catalyze both steps.[18] Over-
all, the B–V reaction cannot be explained by a single mech-
anism and the detailed mechanism of the reaction varies
with the catalyst, substituent effect, solvent, and/or acidity.[3]
The mechanism of the B–V reaction has also been the
subject of a number of theoretical studies. Most of these
studies have investigated B–V reactions with peroxycarbox-
ylic acids as oxidants.[17–23] Remarkably, even though calcula-
tions indicated that the reactions can be catalyzed by gener-
al acid catalysis, through hydrogen-bond formation, there
are still some contradictory results in the mechanism eluci-
dation with regards to the molecularity and the mode of ac-
tivation.[21,23] For the mechanism elucidation of the B–V oxi-
dation with hydrogen peroxide, only a few computational
studies have been published.[24–27] Carlqvist and co-workers
studied the uncatalyzed and BF3-assisted B–V oxidation of
acetone with hydrogen peroxide using high-level ab initio
and DFT methods and found that both steps in the uncata-
lyzed reaction have very high activation barriers, which is in
agreement with the experimental observation that B–V re-
actions of ketones with hydrogen peroxide does not occur to
a significant extent without catalytic activation.[24] It was
found that the Lewis acidic BF3 facilitates both steps of the
reaction and significantly lowers the activation barrier for
the rate-determining step (from 49.0 to 17.0 kcalmolꢀ1).
Sever and Root investigated several possible mechanisms
for Sn- and Ti-catalyzed B–V oxidations of acetone with hy-
drogen peroxide using a DFT method.[25] The calculations
We have previously reported that chiral organophosphoric
acids based on enantiopure 1,1’-bi-2-naphthol (BINOL)[42]
derivatives are competent catalysts for the enantioselective
B–V oxidation of 3-substituted cyclobutanones with aqueous
H2O2 (30%) as the oxidant, affording the corresponding g-
3022
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 3021 – 3035

Asymmetric Baeyer–Villiger Reactions
FULL PAPER
lactones in excellent yields with enantiomeric excess (ee)
values up to 93% (Scheme 1).[42] However, further attempts
to expand the strategy to other less-strained cyclic ketones
indicate that only a single molecule of the phosphoric acid is
involved in the catalytic cycle.
Then what is the exact identity of the catalytic species?
On the basis of the commonly accepted B–V mechanism for
peroxycarboxylic acids, we have proposed the involvement
of a peroxyphosphate intermediate, whereas List and Mꢁller
suggested an alternative bifunctional mechanism involving
noncovalent interactions between the phosphoric acid and
the reactants or the Criegee intermediate.[46] Peroxyphos-
phoric acid would presumably be generated in situ by the
reaction of phosphoric acid with hydrogen peroxide, and a
route involving this intermediate would have the advantage
of being similar to the well-known mechanism of the stoi-
chiometric B–V oxidation using various peracids. For exam-
ple, Ogata and co-workers have performed an elegant study
on the kinetics of the B–V reaction of acetophenones with
permonophosphoric acid, which was prepared in situ by the
reaction of P2O5 with H2O2 in MeCN.[47] Analogously, perse-
leninic acids have also often been invoked as the true active
species for a number of B–V oxidations using hydrogen per-
oxide with various organoselenium catalysts.[9,10,12] On the
other hand, the bifunctional nature of the phosphoric acid in
some asymmetric catalytic reactions has been well docu-
mented in the literature.[48–50] It was found that phosphoric
acid has a proton of considerable acidity, whereas the pres-
ence of the Lewis basic phosphoryl moiety, in proximity to
the acidic proton, may potentially allow for bifunctional cat-
alysis, that is, simultaneous activation of both electro- and
nucleophilic reaction partners.[51] Therefore, two possible
pathways for the B–V reaction, in which a single molecule
of phosphoric acid can be involved in the catalytic cycle
either as a peroxyphosphoric acid (path A) or as a bifunc-
tional general acid (path B) were considered for the present
mechanistic study (Scheme 2).
An inspection of paths A and B in Scheme 2 reveals two
important features that may be potentially useful in the dif-
ferentiation of the routes. First, new signals in the 31P NMR
spectrum would be expected upon formation of the peroxy
species in path A, whereas in path B, the chemical shift of
the 31P NMR signal of the catalyst may not change signifi-
cantly considering the relatively weak noncovalent interac-
tions. Second, the enantioselectivity of the reaction proceed-
ing by path A should be independent of the oxidant used
(R’OOH, R’=H or alkyl), since a common Criegee inter-
mediate I is involved in the reaction, whereas the reverse
should be true for path B, in which the alkyl substituent
group in R’OOH persists near the reactive sites in both
steps of the catalysis. It should be noted that when using al-
kylhydroperoxides as the oxidants, both the formation of
the peroxyphosphoric acid in path A (Scheme 2) and the
corresponding peroxyester (not shown in path A of
Scheme 2) are possible by release of alcohols or H2O, re-
spectively. However, the peroxyester is expected to be inac-
tive for B–V reactions because it lacks an acidic proton. In
fact, comparison of the 31P NMR spectra of the phosphoric
acid 1a taken in CDCl3 before and after the addition of
15 equivalents of a 30% aqueous solution of H2O2 reveals a
Scheme 1. BINOL-derived chiral phosphoric acid catalyzed enantioselec-
tive B–V reaction of cyclobutanone derivatives.
(cyclopentanone, cyclohexanone, or diphenylketone) were
unsuccessful, with no conversion being detected at ambient
temperature. To rationalize the origin of the reactivity dif-
ferences among the ketone substrates and to understand
some key issues of the reaction (including the plausible
mechanistic pathway(s) and the kinetics, the effects of the
chiral phosphoric acid, as well as the controlling factors with
respect to the enantioselectivity), we undertook a combined
experimental and theoretical study of the B–V oxidation of
cyclobutanones with aqueous H2O2 catalyzed by chiral phos-
phoric acids. Insights into the mechanism are gained by the
nonlinear effect (NLE), NMR spectroscopic analysis, and ki-
netic analysis of the system, whereas further details regard-
ing the geometries and energy profiles of the transition
structures and intermediates are provided by a theoretical
study using quantum chemical calculations. The results ob-
tained in this study underline the acid/base bifunctional
nature of the organophosphoric acid catalyst, which plays a
key role in facilitating both steps in the catalytic cycle
through hydrogen-bond formation with the reactants and
the intermediate.
Results and Discussion
The peroxyphosphoric acid pathway and the general acid
pathway: The NLE has been recognized as an important
phenomenon that contains rich mechanistic information in
enantioselective catalysis, and has often been used as a pow-
erful diagnostic tool to probe the nature, and especially the
aggregation behavior, of the catalytic species involved.[45]
With regards to the nature of the catalytic species involved
in the title reaction, we have previously reported an NLE
investigation of this system.[42] The study has shown that the
ee values of the product are proportional to those of the cat-
alyst, clearly indicating the absence of an NLE in this reac-
tion. Moreover, changes in catalyst concentration at room
temperature do not result in substantial impact on the enan-
tioselectivity of the catalysis.[42] These phenomena strongly
Chem. Eur. J. 2010, 16, 3021 – 3035
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3023

K. Ding, Y. Li et al.
(Figure 2).
The
calculated
Gibbs activation barriers are all
higher than 33 kcalmolꢀ1 (cal-
culated at the B3LYP/6-311+
G** level with solvent effect
taken into account), which
means that no peroxyphosphor-
ic acid would form under the
reaction conditions (ambient or
lower temperatures). On the
basis of these experimental
findings and the calculation re-
sults, the possibility of the reac-
tion proceeding by way of a
peroxyphosphoric acid species
(path A) can be safely excluded
from further considerations.
Kinetic studies: To gain further
insights into the mechanism of
the B–V reaction, an extensive
kinetic study was carried out.
Kinetic data were obtained
from the B–V oxidation of 3-n-
Scheme 2. Two plausible pathways for the chiral phosphoric acid catalyzed B–V oxidation of 3-substituted cy-
clobutanones by using R’OOH (R’=H or alkyl) as oxidants.
Table 1. Asymmetric B–V reaction of 2a by using different oxidants in
the presence of catalyst 1b.[a]
slight upfield change in chemical shift (d=4.29 to 4.60 ppm)
after standing for 24 h at RT, suggesting that the hypotheti-
cal formation of the peroxyphosphate species in path A is
unlikely under the reaction conditions (Figure 1).[52] Further-
more, different oxidants were employed to assess the enan-
tioselectivity for the catalytic B–V reaction of 3-phenyl cy-
clobutanone 2a with 1b as the catalyst under otherwise
identical conditions. The results have shown that the ee
values of the lactone 3a vary significantly with the oxidant
employed (Table 1), which is consistent with path B
(Scheme 2). In addition, three computational models were
studied to estimate the activation barrier for the oxidation
of the 5,5’,6,6’,7,7’,8,8’-octahydro-1,1’-bi-2-naphthol (H8-
BINOL)-derived phosphoric acid (1c) to the corresponding
peroxy species with or without the assistance of water
Entry
Oxidant
Time [h]
Yield[b] [%]
ee[c] [%]
1
2
3
30% H2O2
TBHP
CHP
12
36
36
99
87
92
71
17
15
[a] All reactions were carried out at room temperature with [2a]=0.1m
and 1.5 equiv of oxidant. [b] The yield of isolated product. [c] The enan-
tiomeric excess of 3a was determined by chiral HPLC. TBHP=tert-butyl
hydroperoxide, CHP=cumene hydroperoxide.
hexylcyclobutanone (2b) by 30% aqueous H2O2 using cata-
lyst 1b, and the reaction was performed at 278C in dichloro-
methane (Scheme 3). The consumption of 2b was monitored
by gas chromatography with n-dodecane as an internal stan-
dard.
Because the reaction medium is biphasic (dichlorome-
thane/water), possible influences of mass transfer limitations
of the reagents back and forth between the phases has to be
taken into account, and it is necessary to take precautions to
ensure that the reaction rate is not diffusion controlled.
Indeed, reaction profiles obtained under different stirring
Figure 1. Comparison of 31P NMR spectra of the phosphoric acid 1a in
the presence (top; d=4.292 ppm) or absence of H2O2 (bottom; d=
4.579 ppm).
3024
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 3021 – 3035

Asymmetric Baeyer–Villiger Reactions
FULL PAPER
the stirring rates were fixed to
1000 rpm in subsequent kinetic
experiments.
To obtain a comprehensive
picture of the kinetic behavior
for the B–V oxidation, a reac-
tion progress kinetic analysis
was performed by following the
elegant strategy developed by
Blackmond.[53] By monitoring
the entire course of the reaction
under synthetically relevant
conditions, such
a
method
allows for a rapid kinetic analy-
sis of a complex catalytic reac-
tion with a minimal number of
experiments, and can provide
significant information about
complicating issues such as cat-
alyst incubation, catalyst deacti-
vation, or product inhibition.
According to the principles de-
scribed by Blackmond,[53] hy-
drogen peroxide was the reac-
tant used in excess, a quantity
defined as the difference in
total molar concentration be-
tween the two substrates for
the present case. Although the
whole amount of hydrogen per-
oxide is partitioned between
the aqueous and organic
phases,[44,54] the total molar con-
centration of H2O2 in the whole
system is used here instead of
its actual concentration in the
organic phase. This does not
affect the meaning of the re-
sults, as is clear from a similar
kinetic study by Strukul and co-
workers on the PtII-catalyzed
alkene epoxidation with H2O2
in chlorinated solvents.[55] For
Figure 2. The optimized structures with selected bond lengths (in ꢂ) of the reaction complex R and the corre-
sponding transition states (TS) for the oxidation of the phosphoric acid 1c by hydrogen peroxide, with or with-
out the assistance of water. The geometries were fully optimized at the B3LYP/6-311+G** level. The relative
energies DGB3LYPꢀSol (in bold) and DGMP2ꢀSol (in italic) are in kcalmolꢀ1 (for the details of the calculation, see
the Theoretical Calculations section).
the present reaction, the rate is now plotted against the con-
centration of cyclobutanone 2b in Figure 3, line a, and indi-
cates that the reaction is approximately first order in cyclo-
butanone. However, when the reaction was carried out with
an increased initial cyclobutanone and hydrogen peroxide
concentration, but at the same excessive [H2O2], the new
plot does not overlay to the original one and shows a some-
what lowered rate (line b in Figure 3). This behavior indi-
cates that the total concentration of active catalysts varies
with the reaction progress, as a result of either catalyst deac-
tivation or product inhibition. This was further investigated
by adding a certain amount of lactone product 3b to the re-
action system, while keeping the other conditions identical
to those of the first experiment. Indeed, the approximate
Scheme 3. The model B–V reaction for kinetic studies.
rates (Figure S2 in the Supporting Information) clearly indi-
cate some influences of mass transfer on the reaction under
a stirring rate lower than 900 rpm, since the plots deviate
significantly from each other in these cases. Nevertheless,
the plots are essentially reproducible at a stirring rate of
about 1000 rpm, indicating that mass transfer under such cir-
cumstances would no longer be rate determining. Therefore,
Chem. Eur. J. 2010, 16, 3021 – 3035
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3025

K. Ding, Y. Li et al.
Figure 3. Plots of reaction rates against various initial cyclobutanone con-
centrations, standard conditions: [1b]=5 mm in 2 mL of CH2Cl2, T=
300K, stirring rate=1000 rpm. D: [2b]=0.10m, [H2O2]=0.15m,
[H2O2]excess =0.05m; &: [2b]=0.15m, [H2O2]=0.20m, [H2O2]excess =0.05m;
*: [2b]=0.10m, [H2O2]=0.15m, [H2O2]excess =0.05m, [3b]=0.05m.
Figure 4. Plot of initial rate against [H2O2] in the presence of 1b (5 mm)
at 278C (300K) ([2b]=0.1m in CH2Cl2).
by using the B–V reaction of cyclobutanone 2b (0.1m in
CH2Cl2) with H2O2 (0.15m) at 278C. As shown in Figure 5,
an approximately linear correlation between the reaction
overlap of the new data curve (line c in Figure 3) with that
shown in Figure 3, line b attests that product inhibition is re-
sponsible in this case (see also Figure S3 in the Supporting
Information). This is reasonable considering that the lactone
carbonyl group can compete with cyclobutanone for hydro-
gen-bonding interactions with the catalyst. However, in such
a case, the effective concentration of the active catalyst
would change considerably with the reaction progress, which
would complicate the kinetic analysis.
On the basis of the reaction process data mentioned
above, the initial rate analysis was used to further assess the
kinetic aspects of the present reaction to minimize the influ-
ence of product inhibition. As can be seen from Figures S5–
S7 in the Supporting Information, the initial rate remains
nearly constant for a typical reaction profile within 3 min
when the conversion was less than 20%, and can be deter-
mined with reasonable accuracy. Points at the end of the run
when the catalyst had presumably been trapped in part by
the product were discarded. Accordingly, kinetic data were
obtained by varying in turn the initial concentrations of
H2O2, the substrate, and the catalyst. To this end, rate stud-
ies were conducted on a series of B–V reactions in CH2Cl2
at 300K, and aliquots were withdrawn periodically from the
reaction mixture for GC analysis. The dependence of initial
rate on the concentration of hydrogen peroxide was first ex-
amined with the B–V reaction of 2b (0.1m in CH2Cl2) and
concentrations of H2O2 ranging from 0.10 to 0.30m in the
presence of 5 mol% catalyst. Interestingly, the rates of the
cyclobutanone conversion were found to be approximately
independent of H2O2 concentration under the conditions
tested (Figure 4). As far as the mechanism is concerned, the
independence of substrate conversion and the H2O2 concen-
tration implied that the step involving H2O2, either the asso-
ciation with the catalyst or the nucleophilic attack on the
carbonyl carbon, should be a relatively facile one in the cat-
alytic cycle.
Figure 5. Plot of initial rate of 2b (0.1m in CH2Cl2) with H2O2 (0.15m)
against [1b] (1.25 to 5 mm) at 278C (300K). y=3.34428x+0.00069, R2 =
0.9936.
rate and the concentration of 1b was obtained, reflecting a
nearly first-order dependence on catalyst. This is consistent
with the aforementioned results from NLE studies, which
suggested that only a single catalyst molecule is involved in
the rate-determining step. The small intercept on the y axis
indicates that the contribution from a noncatalyzed oxida-
tion of cyclobutanone is negligible, again confirming that
the reaction does not occur without a catalyst.[42]
Finally, the influence of the concentration of 2b (0.04–
0.12m in CH2Cl2) on the reaction rate was examined in the
presence of 1b (2.5 mm) and H2O2 (0.15m) at 300K to ascer-
tain the kinetic order with respect to the ketone substrate.
The profiles indicate no incubation period in the reaction
(Figure S7 in the Supporting Information), which suggests
that catalyst preactivation is not required for the reaction to
occur. Figure 6 depicts the plot of initial rate against the
concentraction of 2b (conversion <20%), which indicates
Subsequently, the rate dependence with respect to the ini-
tial concentration of catalyst 1b (1.25–5 mm) was evaluated
3026
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 3021 – 3035

Asymmetric Baeyer–Villiger Reactions
FULL PAPER
reversible formation of a hydrogen-bonded termolecular
array cat.–H2O2–S. An intra-supramolecular nucleophilic
attack of H2O2 on the cyclobutanone within the array results
in the formation of the catalyst-bonded Criegee intermedi-
ate, which undergoes a subsequent migration–rearrangement
to afford the lactone with the release of the catalyst. How-
ever, the simple kinetic analysis here does not distinguish
whether the addition to the carbonyl to form the Criegee in-
termediate (INT) or its subsequent rearrangement to the
lactone is the rate-determining step, because the rate equa-
tion would be the same in both cases. This issue will be fur-
ther clarified by quantum chemical calculations (see below).
This overall reaction sequence is reminiscent of the Mi-
chaelis–Menten two-step enzyme kinetics typical of those
encountered in enzyme-catalyzed biochemical reactions.
Indeed, a better correlation was obtained in the present cat-
alytic system by plotting the experimental data according to
the double-reciprocal equation developed by Lineweaver
and Burk.[57] For the sequence outlined in Scheme 4, postu-
lating the steady-state concentration of the Criegee inter-
mediate (INT) and using material conservations can lead to
the double-reciprocal expression given in Equation (2) after
simple algebraic manipulations.
Figure 6. Plot of initial reaction rate against [2b] (from 0.04 to 0.12 m in
CH2Cl2) in the presence of H2O2 (0.15m) and 1b (2.5 mm) at 278C
(300 K). y=0.079x+0.00098; R2 =0.9823.
that the rate for the process is approximately first order in
the ketone substrate.
Therefore, the B–V reaction of 2b with H2O2 catalyzed by
1b at 278C (300 K) in CH2Cl2 proceeds by approximately
following the initial rate law shown in Equation (1). It
should be emphasized that this is an empirical approximate
initial rate equation. In this case, the g-lactone concentration
is zero at the beginning of the reaction, and its concentra-
tion effect is not obvious in the kinetic equation.
1
k1 þ kꢀ1 þ k2
k1k2½cat:ꢂ0
kꢀ1 þ k2
a
¼
þ
¼ b þ
ð2Þ
Rate
k1k2K½cat:ꢂ0½Sꢂ
½Sꢂ
The constants k1, kꢀ1, k2, and K are rate or equilibrium con-
stants of the elementary steps defined in Scheme 4, and
[cat.]0 and [S] are the initial concentrations of the catalyst
and cyclobutanone substrate, respectively. Based on the data
from the initial rate measurement, an excellent linear rela-
tionship (with an R2 value of 0.997) is obtained by fitting the
experimental data into the double-reciprocal equation
[Eq. (2); Figure 7]. Therefore, the initial rate of the reaction
is strictly first order in the catalyst, but shows a more com-
plex dependence on the concentration of the cyclobutanone
substrate. This can be seen from an equation of rate depen-
1
1
0
ð1Þ
Rate ꢁ k½1 bꢂ ½2 bꢂ ½H2O2ꢂ
On the basis of these kinetic results, we propose a refined
catalytic cycle as outlined in Scheme 4. Addition of more
H2O2 does not lead to a rate enhancement, which is consis-
tent with a rapid saturation of the catalytic sites with H2O2
through hydrogen-bonding association at the beginning of
the reaction. This pre-equilibrium should be a fast and ki-
netically indistinguishable process. It is worth noting that
chiral organophosphoric acids are known to be amphiphil-
ic,[56] with a hydrophilic acidic interior being sandwiched be-
tween the hydrophobic binaphthalene skeleton and the 3-
and 3’-substituent groups. Although the effective concentra-
tion of H2O2 in dichloromethane is expected to be very
small compared with that of cyclobutanone due to mutual
immiscibility,[44,54] a hydrogen-bonding complex cat.–H2O2 is
preferentially formed probably at the interface between the
aqueous and organic phases. Further interaction of this com-
plex with the cyclobutanone substrate S would lead to the
AHCTUNGTREGdNNUN ence on substrate concentration [Eq. (3)], which is further
Figure 7. The Lineweaver–Burk plot of the B–V reaction of 2b in the
presence of H2O2 (0.15m) and 1b (2.5 mm) in CH2Cl2 at 300K. y=
9.59x+14.92; R2 =0.9973.
Scheme 4. A refined catalytic cycle based on kinetic studies.
Chem. Eur. J. 2010, 16, 3021 – 3035
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3027

K. Ding, Y. Li et al.
deduced from Equation (2), in which a and b are 9.59 and
14.92, respectively, for the present catalytic system (see
Figure 7).
fer hydrogenations,[59,60] Strecker reactions,[61] and hydro-
phosphonylation.[62–63]
Computational details: All calculations in this work were
performed with density functional theory (DFT)[64] and
MP2[65] implemented in the Gaussian 03 program.[66] For the
exploration of the reaction pathway shown in Scheme 5 and
the following reactivity calculations, (R)-H8-BINOL-derived
phosphoric acid (1c) was used as the model catalyst, in
which the initial conformation of the binaphthyl rings was
kept the same as that in the H8-BINOL/Ti complex, which
has been characterized by X-ray crystallography.[67] Models
were constructed and fully optimized at the B3LYP[68]/6-
311+G** level. For each optimized structure, a harmonic
vibrational frequency calculation was carried out and ther-
mal corrections were made. All structures were shown to be
either transition states (with one imaginary frequency) or
local minima (with no imaginary frequency). The solvent
effect was estimated with IEFPCM[69] (UAHF atomic radii)
method in chloroform (e=4.9) using the gas-phase opti-
mized structures. The MP2/6-311+G** single-point calcula-
tion was performed to obtain more accurate energies.
½Sꢂ
ð3Þ
Rate ¼
b½Sꢂ þ a
At a low concentration of the ketone substrate (e.g.,
[2b]=0.030–0.125 m as shown in Figure 6), the term b[S]
ranges from 0.45 to 1.94, which is much smaller than the a
value (9.59). Thus, the denominator b[S]+aꢁa, so that
under low concentrations of 2b the rateꢁ[S]/a. This implies
that the rate is approximately proportional to the concentra-
tion of cyclobutanone in a relatively dilute solution, such as
the cases shown in Figure 6. Accordingly, the intercept b on
the y axis of the Lineweaver–Burk plot is an indication of
kinetic behavior sensitivity of the ketone with the change of
its concentration. Equation (1) can be regarded as an ap-
proximation of Equation (2) when the cyclobutanone con-
centration is low enough, and Equation (2) is more rigorous
mathematically and more accurately reflects the kinetic be-
havior of the catalytic system.
Theoretical calculations: To gain further insights into the
mechanism, a theoretical study was carried out to explore
the geometries and energetics of the key species as well as
to rationalize the reactivity/enantioselectivity of the cataly-
sis. A plausible mechanistic pathway is shown in Scheme 5,
in which the H8-BINOL-derived phosphoric acid acts as a
bifunctional catalyst throughout the catalytic sequence to
form hydrogen-bonding networks with the reactants, Crie-
gee intermediate, or the lactone product. The BINOL-de-
rived phosphoric acids contain both a Brønsted acidic site
(acid proton) and a Lewis basic site (phosphoryl oxygen) in
proximity, and have been proposed to activate simultaneous-
ly both electrophilic and nucleophilic reaction partners by
hydrogen-bond formation in a number of reactions.[49–51,58]
Such a bifunctional catalysis pattern has also been demon-
strated to be viable in several recent theoretical studies on
the mechanisms of phosphoric acid catalyzed nucleophilic
addition reactions of imines, including Hantzsch ester trans-
The optimized structures for the species proposed in
Scheme 5 are depicted in Figure 8, along with the calculated
Gibbs free energy profiles for the corresponding mechanistic
pathway. Two types of relative free energies (in kcalmolꢀ1),
both with the hydrogen-bonded termolecular complex R as
the zero reference point, are shown in Figure 8. The
DGB3LYPꢀSol values correspond to the relative free energies
calculated with the B3LYP/6-311+G** method plus the sol-
vent effect, whereas DGMP2ꢀSol are MP2/6-311+G** single-
point energies plus the thermal free energy corrections and
the solvent effect at the B3LYP/6-311+G** level.
The reactant complex (R): For the initial step of the catalytic
B–V reaction, the calculation starts with a termolecular re-
actant complex R generated by hydrogen-bonding interac-
tions of the H8-BINOL-derived phosphoric acid with cyclo-
butanone and hydrogen peroxide. The phosphoric acid tends
to donate the proton Ha to the carbonyl Oe of the cyclobuta-
Scheme 5. Proposed two-step mechanism for the phosphoric acid (1c) catalyzed B–V reaction.
3028
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 3021 – 3035

Asymmetric Baeyer–Villiger Reactions
FULL PAPER
Figure 8. The optimized structures with selected bond lengths (in ꢂ) of the reactant complex (R), the transition states (TS1 and TS2), the intermediate
(INT) and the product (P). The geometries were fully optimized at the B3LYP/6-311+G** level. The relative free energies DGB3LYPꢀSol (in bold) and
DGMP2ꢀSol (in italic) are in kcalmolꢀ1
.
none, with an Oe···Ha distance of 1.644 ꢂ and a nearly linear
carbonyl group, in which the negative charge on the carbon-
yl oxygen is effectively trapped by protonation. As a result,
the carbonyl double bond is elongated from 1.212 ꢂ in R to
1.275 ꢂ in TS1. Moreover, the proton Hb of hydrogen per-
oxide is transferred to the phosphoryl oxygen, rendering hy-
ꢀ
O Ha···Oe geometry (174.78), suggesting a relatively strong
intermolecular hydrogen-bonding interaction. As a result,
the electrophilic character of the carbonyl carbon in R be-
comes more pronounced than that in the free cyclobuta-
none. Simultaneously, the phosphoryl oxygen Ob starts to
accept a hydrogen (Hb) from the hydrogen peroxide, thus
leading to an increase in the nucleophilicity of the oxidant.
The reactants are brought in proximity in the hydrogen-
bonding network, with a relative orientation that is favora-
ble for the addition step.
ꢀ
drogen peroxide more nucleophilic than in R. The two P O
bond lengths in TS1 are almost identical (1.511 and 1.519 ꢂ,
ꢀ ꢀ
respectively), indicating a delocalized O P O bond with a
bond length that is typically between the P=O double bond
[62]
ꢀ
(ca. 1.47 ꢂ) and the P O single bond (ca. 1.51–1.55 ꢂ).
Thus, these data clearly indicate that the bond-forming and
proton-transfer processes occur in a concerted manner
through an eight-membered ring structure in TS1 as shown
in Scheme 5 and Figure 8.
The transition state for the formation of the Criegee inter-
mediate (TS1): For the addition of hydrogen peroxide to the
cyclobutanone within the hydrogen-bonding network, a
moderate activation barrier in terms of Gibbs free energy
The phosphoric acid bound Criegee intermediate (INT): Ac-
cording to the B3LYP and MP2 calculations, the formation
of the Criegee intermediate is expected to be reversible
under the experimental conditions (with the barriers lower
than 12.5 kcalmolꢀ1). In the tetrahedral peroxide moiety of
was found on the potential energy surface (DGB3LYPꢀSol
=
13.8 kcalmolꢀ1 and DGPM2ꢀSol =12.0 kcalmolꢀ1). This is rea-
sonable considering the dual activation mode for the cataly-
sis. TS1 retains a similar gross structure as that of R, but
some critical changes have occurred around the phosphoric
acid moiety. In TS1, the proton of the phosphoric acid (Ha)
has been largely transferred to the carbonyl oxygen atom
ꢀ
INT, the peroxide carbonyl carbon bond is fully formed,
ꢀ
ꢀ
whereas the O O bond and the two C C bonds involving
the carbonyl carbon are slightly elongated relative to those
in R, indicating an onset weakening of these bonds. The
proton transfer of the addition step has been accomplished,
resulting in the regeneration of the phosphoric acid, which
is engaged in hydrogen-bonding interactions with the perox-
ꢀ
(Oe) of the cyclobutanone (Ha Oe distance of 1.064 ꢂ), re-
sulting in protonation of the latter. Concurrent with this
ꢀ
process is the incipient formation of the Oc C bond by nu-
cleophilic attack of hydrogen peroxide on the cyclobutanone
Chem. Eur. J. 2010, 16, 3021 – 3035
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3029

K. Ding, Y. Li et al.
ide moiety of INT through a nine-membered ring (Scheme 5
and Figure 8).
is effectively irreversible. Thus the overall reaction is kineti-
cally controlled, and for the asymmetric B–V reaction of 3-
substituted cyclobutanones, this migration will create a
chiral center in the product and will determine the enantio-
selectivity. The relatively strong binding of lactone to phos-
phoric acid has important implications for the kinetics of the
reaction, and this would provide an explanation of the prod-
uct inhibition as discussed above for the results of kinetic
studies.
Overall, a remarkable feature of the calculated mecha-
nism is the relay of the protons Ha and Hb, which are shut-
tling back and forth between the reacting species and the
catalyst by hydrogen-bonding interactions, thus facilitating a
concerted bond-making and -breaking process for both steps
of the catalysis. This reaction pathway is consistent with the
kinetic results and the mechanistic proposal by List and
Mꢁller,[46] and it is similar to the mechanism of carboxylic
acid catalyzed B–V oxidation using peracids proposed by
Alvarez-Idaboy and co-workers.[19,22,23]
The transition state for the migration step (TS2): The subse-
quent B–V rearrangement of the INT is also a concerted
step that is facilitated by the proton transfer between the
Criegee intermediate and the phosphoric acid moiety. In
TS2, the distal hydroperoxo oxygen Od from the intermedi-
ate is accepting a proton Hb from the phosphoric acid to
form water as a good leaving group, thus facilitating the het-
ꢀ
erolytic cleavage of the O O bond by stabilization of the
developing negative charge on Od. Simultaneously, the
ꢀ
methylene group that is anti-periplanar to Oc Od migrates
ꢀ
with the incipient formation of an O C bond (2.153 ꢂ) and
ꢀ
the rupture of C C bond (1.770 ꢂ). The migration in turn
facilitates the formation of the C=O double bond, and the
proton on the oxygen is ready to be transferred to the phos-
phoryl group of the catalyst. Animation of the imaginary
frequency value of TS2 at 820.5i confirms that the rear-
rangement is a concerted process, with the critical bond
forming and breaking being in synergy with the hydrogen
migrations. The activation barrier for the rearrangement of
INT is significantly higher than its formation, and thus will
control the overall rate of the B–V reaction. The activation
barrier of <20 kcalmolꢀ1 for this rate-determining step is
compatible with the mild reaction conditions employed in
the experimental studies.
Rationalization of the reactivities of different ketones towards
the B–V reaction: The B–V reactions of acetone, cyclopenta-
none, and cyclohexanone were also examined theoretically,
promoted by 1c and using hydrogen peroxide as the oxi-
dant. Because both the electrophilicity and hydrogen-bond-
ing capability of these ketones can be expected to be similar
to those of cyclobutanone, similar reactivity for the addition
was assumed for both cyclic and acyclic ketones. Only the
transition structures and the activation barriers for the rate-
determining step of these reactions are calculated, and the
results are shown in Figure 9. The geometries are fully opti-
mized at the B3LYP/6-311+G** level. The barriers
(DGB3LYPꢀSol) for the rate-determining step of the B–V reac-
The catalyst-bound g-lactone product (P): The formation of
the lactone is accomplished without loss of hydrogen bond-
ing to the phosphoric acid, leading to the catalyst-bound g-
lactone product complex P. The migration is calculated to
be strongly exothermic (>87 kcalmolꢀ1), indicating this step
Figure 9. The calculated structures with selected bond lengths (in ꢂ) and relative energies (in kcalmolꢀ1) (DGB3LYPꢀSol (in bold) and DGMP2ꢀSol (in italic))
of the reactant complexes (R) and the transition states TS2 for the B–V reactions of acetone, cyclopentanone and cyclohexanone.
3030
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 3021 – 3035

Asymmetric Baeyer–Villiger Reactions
FULL PAPER
tions of acetone, cyclopentanone, and cyclohexanone (31.2,
27.0, and 25.5 kcalmolꢀ1, respectively) are remarkably
higher than that of cyclobutanone (20.2 kcalmolꢀ1), which is
consistent with their inactivity in experimental observations.
This is not surprising considering the larger ring strain in cy-
rived phosphoric acid catalyst with bulky 3,3’-pyrenyl groups
(1b) and 3-phenylcyclobutanone (2a) substrate were chosen
for the calculation study, due to the high enantioselectivities
obtained by using such a catalyst and also for adequate
modeling of the substituent effect. Complete models of the
competing transition states corresponding to the transition
state of the rate-determining step (TS2) were constructed
according to the designations as shown in Figure 10 and
Scheme 6. To rationalize the high enantioselectivity, multi-
site (three-point) interactions[59] can be expected between
the catalyst and the substrate in transition-state structures,
that is, the two stabilizing hydrogen bonds as discussed
above and an additional interaction between the substrate
and the pyrenyl group(s) of the catalyst. It is a challenging
task to predict the enantioselectivity of such large systems,
in which the weak nonbonding interactions are predominant
and the energy differences between transition states are ex-
pected to be very small. To achieve a better description of
the weak nonbonding interactions, the PBE1PBE method[70]
was used instead of B3LYP in the enantioselectivity calcula-
tions. The 6-311+G** basis set was used for the reacting
centers of the structure in Scheme 6 (central part encircled
in the dashed line, but not including the bi(H4-naphthyl)
backbone atoms), whereas 6-31G* was used as the basis set
for the remaining atoms. Taking the H4-naphthyl rings in the
catalyst as the reference planes, each pyrenyl group has two
possible configurations, that is, syn or anti with respect to
the phosphoric acid moiety. Taking the two different orien-
tations of the phenyl group (up and down) in the cyclobuta-
none into consideration, for catalyst 1b there are eight
viable competing transition structures, four leading to the
product with R configuration (R-syn-anti, R-anti-syn, R-anti-
anti, and R-syn-syn) and the other four to the product with
ꢀ
clobutanone weakens the C C bond and makes the migra-
tion of the methylene group relatively more facile. In the re-
ꢀ
actant complex containing cyclobutanone, the C C bond
lengths are 1.523/1.517 ꢂ (Figure 8), whereas in acetone, cy-
ꢀ
clopentanone, and cyclohexanone, the C C bond lengths
are 1.504/1.504, 1.518/1.511, and 1.512/1.510 ꢂ, respectively
(Figure 9).Therefore, the present B–V reaction is facilitated
by the large ring strain in cyclobutanone.
The origin of the enantioselectivity: The mechanistic calcula-
tion revealed that activation of both substrate and H2O2 by
phosphoric acids constitutes the basis for these B–V reac-
tions of cyclobutanones, and the catalyst is capable of stabi-
lizing the TS2 of the rate-determining, stereocenter-forming,
alkyl-migration step by partial proton transfer. A chiral
center is created in this migration step, thus the ratio of R
and S isomers in the lactone product should necessarily be
dependent on the energetic differences between the compet-
ing diastereomeric transition states of this step. In the exper-
imental investigation of the reaction, it has been found that
the presence of bulky fused aryl substituents at the 3- and
3’-positions of the BINOL-derived phosphoric acid is crucial
for high enantioselectivity, and variations of the stereoelec-
tronic properties of these aryl substituents resulted in pro-
nounced effects.[42] On the other hand, changing the sub-
stituent at the 3-position of the cyclobutanone substrate
from an aryl to an alkyl group resulted in a considerable de-
crease in enantioselectivity. Therefore, a (R)-H8-BINOL-de-
Figure 10. The eight optimized transition states with the relative free energies including solvent effect DGTotal2 (in kcalmolꢀ1).
Chem. Eur. J. 2010, 16, 3021 – 3035
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3031

K. Ding, Y. Li et al.
tures, their energy gaps could lead to the formation of (R)-
lactone with an ee value of around 73% (DGTotal2, 298 K),
which agrees very well with the experimental result (71% ee
at RT).[42]
As shown in Table 2, DE(6-311+G**), ꢀTDS, and DDGsol are
in the ranges of 1.0, 1.7, and 2.1 kcalmolꢀ1, respectively.
Therefore, the entropy and solvent effects are decisive fac-
tors in the determination of the energy sequence of the
eight transition states. For the transition structures leading
to S products shown in Figure 10, S-anti-anti and S-syn-anti
are less stable than S-anti-syn and S-syn-syn. The phenyl
groups of the cyclobutanone in the first two transition states
have developed very strong crowdedness with the pyrenyl
group on the right side. On the contrary, in the last two
cases the phenyl substituent of the cyclobutanone is stag-
gered with the pyrenyl unit, which, being more relaxed, is
accordingly more favorable in terms of entropy (ꢀ2.143,
ꢀ0.952 vs. ꢀ5.694, ꢀ5.054 calmolꢀ1 K-1). Although S-syn-anti
has a lower DE(6ꢀ311+G**) value (ꢀ0.7 kcalmolꢀ1), the entropy
loss of this crowded structure increases the free energy by
1.5 kcalmolꢀ1. Here, the lower DE(6-311+G**) may indicate a
stable nonbonding interaction between the phenyl group of
the cyclobutanone and the pyrenyl group, however, the en-
tropy loss has larger effect on the total free energy. Similar-
ly, the relative stabilities of the four R-type transition states
can be easily understood. The phenyl group of the cyclobu-
tanone overlaps with the pyrenyl group on the right side in
R-syn-syn and R-anti-syn transition states, but these groups
are staggered in the R-syn-anti and R-anti-anti transition
states. Two factors may account for the relative stabilities of
the most favorable R-syn-anti and S-anti-syn transition struc-
tures: 1) In the S-anti-syn transition state, the phenyl group
of the cyclobutanone is to some extent congested with the
pyrenyl group of the catalyst, whereas in the R-syn-anti tran-
sition state the phenyl group of the cyclobutanone is in
some degree staggered with the pyrenyl group of the cata-
lyst. Therefore, the R-syn-anti transition state is more re-
laxed and more favorable than the S-anti-syn transition state
in terms of entropy (0.000 vs. ꢀ2.143 calmolꢀ1 K-1). 2) Com-
pared with the S-anti-syn transition state, the pyrenyl group
on the left in R-syn-anti is opened more widely, and accord-
ingly the polar phosphoric acid is more exposed to the polar
solvent, leading to favorable
Scheme 6. The model used to calculate the enantioselectivity (6-311+
G** basis set for the part of the structure encircled in the dashed line,
but not including the binaphthyl backbone atoms, and the 6-31G* basis
set for the remaining part).
S configuration (S-syn-anti, S-anti-syn, S-anti-anti, and S-
syn-syn). These eight transition structures were fully opti-
mized and checked by harmonic vibrational frequency calcu-
lations at the PBE1PBE/(6-311+G** and 6-31G*) level, and
the solvent effect was estimated in each case by the same
method as described in the previous section. The single-
point energies of the eight optimized transition structures
have also been calculated at the PBE1PBE/6-311+G**
level. The optimized transition structures are shown in
Figure 10, and the corresponding relative energies (DE), rel-
ative single-point energies at the PBE1PBE/6-311+G**
level (DE(6ꢀ311+G**)), relative entropies (DS,calmolꢀ1 K-1), the
contribution of DS to DG (ꢀTDS, T=298K), the relative
free energies in gas phase (DGgas, 298K), the relative solva-
tion free energy (DDGsol), the total relative free energies
(DGTotal(DGTotal=DGgas +DDGsol)), and the total relative free
energies (DGTotal2) corrected by DE(6ꢀ311+G**) single-point en-
ergies are listed in Table 2. Gratifyingly, the calculated enan-
tioselectivity is consistent with the experimental result. In
the presence of the R catalyst, the most favorable transition
structure is found to be R-syn-anti, which is 1.3 kcalmolꢀ1
more stable than S-anti-syn, the most favorable transition
state among the (S)-producing configurations. Assuming a
Boltzmann distribution of the eight transition-state struc-
solvation
effects
(0.0
vs.
Table 2. Various energy values for the optimized transition states.[a]
0.5 kcalmolꢀ1). It is notable
that the DGgas of S-syn-syn is
0.2 kcalmolꢀ1 lower than that of
R-syn-anti. However, the rela-
tive solvation free energy of the
[c]
[f]
[g]
[h]
[i]
Transition states
DE[b]
DE(6ꢀ311+G**)
DS[d]
ꢀTDS[e]
DGgas
DDGsol
DGTotal
DGTotal2
R-syn-anti
R-anti-anti
R-syn-syn
S-anti-syn
S-syn-syn
R-anti-syn
S-anti-anti
S-syn-anti
0.0
0.0
0.0
ꢀ0.1
ꢀ1.1
0.3
0.000
0.0
0.8
1.1
0.6
0.3
1.0
1.7
1.5
0.0
0.7
0.7
0.0
0.3
1.0
0.5
1.5
0.5
0.5
2.1
0.0
1.0
1.7
1.1
1.3
1.9
2.5
3.0
0.0
0.9
1.2
1.3
1.4
1.4
2.6
3.2
ꢀ2.767
ꢀ3.720
ꢀ2.143
ꢀ0.952
ꢀ3.278
ꢀ5.694
ꢀ5.054
ꢀ0.6
0.1
0.6
S-syn-syn
is
1.5 kcalmolꢀ1
ꢀ0.6
0.5
0.1
ꢀ0.9
ꢀ0.5
ꢀ0.2
0.0
1.4
higher than that of the R-syn-
anti. These data clearly indicate
that the solvent effect plays a
significant role in the control of
the enantioselectivity. There-
fore, the sense of asymmetry in-
duction will be determined not
0.2
ꢀ0.7
2.0
0.9
[a] All energies are relative values in kcalmolꢀ1 for the structures optimized at the PBE1PBE/(6-311+G**
and 6-31G*) level. [b] The relative energy. [c] Relative single-point energy. [d] Relative entropy [calmolꢀ1 K-1].
[e] Contribution of DS to DG (at 298 K). [f] Relative free energy in gas phase (at 298 K). [g] Relative solvation
free energy. [h] Total relative free energy (DGTotal =DGgas +DDGsol)). [i] Total relative free energies corrected
by DE(6-311+G**) single-point energy.
3032
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 3021 – 3035

Asymmetric Baeyer–Villiger Reactions
FULL PAPER
only by the entropy effect, but also critically by the solvent
effect, which has been known to greatly influence the migra-
tion aptitude of the B–V reaction of some substrates.[18,71] In
fact, both the catalytic efficiency and the asymmetric induc-
tion in our phosphoric acid catalyzed B–V oxidation of 2a
using aqueous H2O2 were found to be strongly dependent
on the nature of solvents used.[42]
vation of reactants disclosed in the reaction, that is, the cata-
lyst with two different binding sites in close proximity,
seems to be a common feature in several mechanistic
models proposed for B–V reactions, and a close resem-
blance in mechanism can be found between the title reac-
tion and catalysis by carboxylic acids,[19,22,23] metal ions,[25] or
zeolites,[26] and thus may represent a promising approach for
future catalyst development.
Conclusion
Experimental Section
The mechanism of the chiral phosphoric acid catalyzed B–V
reaction of cyclobutanones has been elucidated by a com-
bined experimental and theoretical study. It was found that
the involvement of a peroxyphosphate intermediate is not a
viable pathway for the present reaction. The reaction fol-
lows the Michaelis–Menten kinetics typical of those encoun-
tered in enzyme-catalyzed biochemical reactions, and the re-
sulting g-lactone product retards the reaction by competing
hydrogen bonding with the catalyst. Under the biphasic re-
action conditions, the amphiphilic BINOL–phosphoric acid
catalyst is saturated with hydrogen peroxide, resulting in
zero order kinetics in hydrogen peroxide. The initial rate of
the reaction is first order in the catalyst, but shows a more
complex dependence on the concentration of the cyclobuta-
none substrate. These results are fully consistent with the
proposed reaction sequence, in which the catalyst is actively
involved in both the carbonyl addition and the subsequent
rearrangement steps through hydrogen-bonding interactions
with the reactants or the intermediate.
High-level quantum chemical calculations provide strong
support for the two-step concerted mechanism for the cata-
lytic reaction in which the phosphoric acid plays the role of
a bifunctional catalyst to activate the reactants or the inter-
mediate in a synergistic manner through partial proton
transfer. The catalyst simultaneously increases the electro-
philicity of the carbonyl carbon as a general acid and the nu-
cleophilicity of hydrogen peroxide as a Lewis base in the ad-
dition step, and facilitates the dissociation of the poor leav-
ing group OH from the Criegee intermediate in the rear-
rangement step. The overall reaction is highly exothermic,
leading to the stable hydrogen-bonded lactone product. The
calculated reaction profile suggests that the rearrangement
of the Criegee intermediate is the rate-determining step,
with an activation barrier of around 20 kcalmolꢀ1, which is
compatible with the reaction conditions. The observed reac-
tivity of the present catalytic B–V reaction is also caused in
part by the ring strain in cyclobutanones. The sense of chiral
induction is rationalized by the analysis of the relative ener-
gies of the competing diastereomeric transition states, in
which the steric repulsion between the cyclobutanone deriv-
ative and the catalyst, as well as the entropy and solvent ef-
fects, are found to be key impact factors.
General methods: All reactions were performed under an air atmos-
phere, using oven-dried glassware and magnetic stirring. Dichlorome-
thane was distilled from powdered CaH2 under argon prior to use. 3-Phe-
nylcyclobutanone (2a),[72] 3-n-hexylcyclobutanone (2b),[72] and chiral
phosphoric acids 1a,b[73] were prepared according to literature proce-
dures. Aqueous H2O2 (30%) was purchased from a commercial source.
NMR spectra were recorded on a Varian Mercury 300 (1H: 300 MHz;
31P: 121 MHz) or Varian 400-MR (1H: 400 MHz; 31P: 161 MHz) spec-
trometer in CDCl3. Chemical shifts are expressed in ppm with TMS as an
internal standard (d=0 ppm) for 1H NMR data. Coupling constants, J,
are listed in hertz. 31P NMR data are reported in terms of chemical shift
(d, ppm) with 85% H3PO4 as the internal standard. HPLC analysis was
carried out on a JASCO PU 2089 instrument with a UV 2075 detector.
GC analysis was performed on an Agilent 6890N instrument with an HP-
5 column. Unless stated otherwise, all solvents were purified and dried
according to standard methods prior to use. Analytical GC was carried
out using a HP-5 column (30 m, 0.316 mm i.d., 0.25 mm film) and N2 as
carrier gas.
General procedures for the chiral phosphoric acid catalyzed asymmetric
B–V reaction of 3-phenylcyclobutanone with different oxidants: A 5 mL
Schlenk tube was charged with 1a (7.6 mg, 10 mol%, 0.01 mmol), 3-phe-
nylcyclobutanone 2a (0.1 mmol), and CHCl3 (1 mL). The mixture was
stirred at room temperature and then oxidant (1.5 equiv) was added at
room temperature. The reaction mixture was stirred for 12–36 h at room
temperature. The reaction was quenched with an aqueous solution of
Na2SO3 and the product 3a was isolated by column chromatography on
silica gel with petroleum ether/ethyl acetate=6/1 as the eluent. The
enantiomeric excess was determined by HPLC analysis on a Chiralcel
AD-H column.
Typical procedure for the kinetic analyses: A 5 mL Schlenk tube contain-
ing a solution of 1b (7.6 mg, 0.01 mmol), the internal standard (n-dodec-
ane, 34 mg), and 3-n-hexylcyclobutanone (30.8 mg, 0.2 mmol) in CH2Cl2
(2 mL) was immersed in an oil bath and its internal temperature allowed
to stabilize at 278C. The mixture was stirred at 278C for 5–10 min under
air. An aliquot was taken at this point to establish the initial cyclobuta-
none/dodecane ratio, then a 30% aqueous solution of H2O2 (34 mL,
1.5 equiv) was added quickly in one portion. Aliquots of 5–10 mL were
taken at the specified intervals, diluted with THF (1 mL) (note: no reac-
tion occurs in THF!),[42] and submitted to GC analysis. Experiments to
examine the mass transfer effect with different stirring rates, product in-
hibition, and the initial rates of the B–V reactions were also conducted in
a similar manner.
Acknowledgements
Financial support from the National Natural Science Foundation of
China (no. 20872168, 20632060, 20821002), the Chinese Academy of Sci-
ences, the Major Basic Research Development Program of China (Grant
No. 2010CB833300), the Science and Technology Commission of Shang-
hai Municipality, and Merck Research Laboratories is gratefully acknowl-
edged.
The insightful understanding of the mechanistic aspects of
present catalytic asymmetric B–V oxidation may aid in the
design of new catalytic systems for chiral Brønsted acid cat-
alyzed reactions. The multiple interaction mode for the acti-
Chem. Eur. J. 2010, 16, 3021 – 3035
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3033

K. Ding, Y. Li et al.
S. Valencia, Nature 2001, 412, 423–425; c) A. Corma, M. T. Navarro,
L. T. Nemeth, M. Renz, Chem. Commun. 2001, 2190–2191; d) M.
Renz, T. Blasco, A. Corma, V. Fornꢃs, R. Jensen, L. Nemeth, Chem.
Eur. J. 2002, 8, 4708–4717; e) A. Corma, M. T. Navarro, M. Renz, J.
Catal. 2003, 219, 242–246.
[1] G. R. Krow, Org. React. 1993, 43, 251–798.
[2] A. Baeyer, V. Villiger, Ber. Dtsch. Chem. Ges. 1899, 32, 3625–3633.
[3] M. Renz, B. Meunier, Eur. J. Org. Chem. 1999, 737–750.
[4] a) C. Bolm in Advances in Catalytic Processes, Vol. 2 (Ed.: M. P.
Doyle), JAI Press, Greenwich, 1997; b) G.-J. ten Brink, I. W. C. E.
Arends, R. A. Sheldon, Chem. Rev. 2004, 104, 4105–4123.
[5] G. Strukul, Angew. Chem. 1998, 110, 1256–1267; Angew. Chem. Int.
Ed. 1998, 37, 1198–1209.
[6] For a recent review, see: a) C. Jimꢃnez-Sanchidriꢄn, J. R. Ruiz, Tet-
rahedron 2008, 64, 2011–2026; for leading references in catalytic
heterogeneous B–V reactions, see: b) A. Corma, L. T. Nemeth, M.
Renz, S. Valencia, Nature 2001, 412, 423–425; c) A. Corma, M. T.
Navarro, L. T. Nemeth, M. Renz, Chem. Commun. 2001, 2190–
2191; d) M. Renz, T. Blasco, A. Corma, V. Fornes, R. Jensen, L.
Nemeth, Chem. Eur. J. 2002, 8, 4708–4717; e) A. Corma, M. T. Nav-
arro, M. Renz, J. Catal. 2003, 219, 242–246.
[27] Q. P. Long, H. B. Ji, S. S. Lu, J. Mol. Struct. 2009, 917–938, 117–127.
[28] M. D. Mihovilovic, F. Rudroff, B. Grotzl, Curr. Org. Chem. 2004, 8,
1057–1069.
[29] S. M. Roberts, P. W. H. Wan, J. Mol. Catal. B 1998, 4, 111–136.
[30] M. D. Mihovilovic, Curr. Org. Chem. 2006, 10, 1265–1287.
[31] a) C. Bolm, G. Schlingloff, K. Weickhardt, Angew. Chem. 1994, 106,
1944–1946; Angew. Chem. Int. Ed. Engl. 1994, 33, 1848–1849; b) A.
Gusso, C. Baccin, F. Pinna, G. Strukul, Organometallics 1994, 13,
3442–3451.
[32] a) C. Bolm, G. Schlingloff, J. Chem. Soc. Chem. Commun. 1995,
1247–1248; b) C. Bolm, T. K. K. Luong, G. Schlingloff, Synlett 1997,
1151–1152; c) Y. G. Peng, X. M. Feng, K. B. Yu, Z. Li, Y. Z. Jiang,
C. H. Yeung, J. Organomet. Chem. 2001, 619, 204–208.
[33] a) C. Paneghetti, R. Cavagnin, F. Pinna, G. Strukul, Organometallics
1999, 18, 5057–5065; b) M. Colladon, A. Scarso, G. Strukul, Synlett
2006, 3515–3520.
[7] R. Criegee, Justus Liebigs Ann. Chem. 1948, 560, 127–135.
[8] a) K. Sato, M. Hyodo, J. Takagi, M. Aoki, R. Noyori, Tetrahedron
Lett. 2000, 41, 1439–1442; b) A. Berkessel, M. R. M. Andreae, Tet-
rahedron Lett. 2001, 42, 2293–2295.
[9] G.-J. ten Brink, J.-M. Vis, I. W. C. E. Arends, R. A. Sheldon, J. Org.
Chem. 2001, 66, 2429–2433.
[34] M. Lopp, A. Paju, T. Kanger, T. Pehk, Tetrahedron Lett. 1996, 37,
7583–7586.
[10] G.-J. ten Brink, J.-M. Vis, I. W. C. E. Arends, R. A. Sheldon, Tetrahe-
dron 2002, 58, 3977–3983.
[11] A. Berkessel, M. R. M. Andreae, H. Schmickler, J. Lex, Angew.
Chem. 2002, 114, 4661–4664; Angew. Chem. Int. Ed. 2002, 41, 4481–
4484.
[12] a) H. Ichikawa, Y. Usami, M. Arimoto, Tetrahedron Lett. 2005, 46,
8665–8668; b) M. A. Goodman, M. R. Detty, Synlett 2006, 1100–
1104.
[13] a) S. E. Jacobson, F. Mares, P. M. Zambri, J. Am. Chem. Soc. 1979,
101, 6938–6946; b) R. T. Taylor, L. A. Flood, J. Org. Chem. 1983,
48, 5160–5164; c) Z. B. Wang, T. Mizusaki, T. Sano, Y. Kawakami,
Bull. Chem. Soc. Jpn. 1997, 70, 2567–2570; d) J. Fischer, W. F. Hçld-
erich, Appl. Catal. A 1999, 180, 435–443.
[35] T. Uchida, T. Katsuki, Tetrahedron Lett. 2001, 42, 6911–6914.
[36] a) A. Watanabe, T. Uchida, K. Ito, T. Katsuki, Tetrahedron Lett.
2002, 43, 4481–4485; b) A. Watanabe, T. Uchida, Y. Irie, T. Katsuki,
Proc. Natl. Acad. Sci. USA 2004, 101, 5737–5742.
[37] C. Bolm, O. Beckmann, A. Cosp, C. Palazzi, Synlett 2001, 1461–
1463.
[38] a) C. Bolm, O. Beckmann, T. Kuhn, C. Palazzi, W. Adam, P. B. Rao,
C. R. Saha-Mçller, Tetrahedron: Asymmetry 2001, 12, 2441–2446;
b) C. Bolm, J.-C. Frison, Y. Zhang, W. D. Wulff, Synlett 2004, 1619–
1621; c) J.-C. Frison, C. Palazzi, C. Bolm, Tetrahedron 2006, 62,
6700–6706.
ˇ
´
[39] A. V. Malkov, F. Friscourt, M. Bell, M. E. Swarbrick, P. Kocovsky, J.
Org. Chem. 2008, 73, 3996–4003.
[14] a) C. Mazzini, J. Lebreton, R. Furstoss, J. Org. Chem. 1996, 61, 8–9;
b) C. Mazzini, J. Lebreton, R. Furstoss, Heterocycles 1997, 45, 1161–
1167.
[15] S. Murahashi, S. Ono, Y. Imada, Angew. Chem. 2002, 114, 2472–
2474; Angew. Chem. Int. Ed. 2002, 41, 2366–2368.
[16] a) D. A. Singleton, M. J. Szymanski, J. Am. Chem. Soc. 1999, 121,
9455–9456; b) S. Kobayashi, H. Tanaka, H. Amii, K. Uneyama, Tet-
rahedron 2003, 59, 1547–1552.
[40] Y. Miyake, Y. Nishibayashi, S. Uemura, Bull. Chem. Soc. Jpn. 2002,
75, 2233–2237.
[41] G. Peris, S. J. Miller, Org. Lett. 2008, 10, 3049–3052.
[42] S. Xu, Z. Wang, X. Zhang, X. Zhang, K. Ding, Angew. Chem. 2008,
120, 2882–2885; Angew. Chem. Int. Ed. 2008, 47, 2840–2843.
[43] a) M. D. Del Todesco Frisone, F. Pinna, G. Strukul, Organometallics
1993, 12, 148–156; b) R. Gavagnin, M. Cataldo, F. Pinna, G. Strukul,
Organometallics 1998, 17, 661–667.
[17] L. Reyes, M. Castro, J. Cruz, M. Rubio, J. Phys. Chem. A 2005, 109,
3383–3390.
[44] A. Brunetta, P. Sgarbossa, G. Strukul, Catal. Today 2005, 99, 227–
232.
[18] F. Grein, A. C. Chen, D. Edwards, C. M. Crudden, J. Org. Chem.
2006, 71, 861–872.
[19] J. R. Alvarez-Idaboy, L. Reyes, N. Mora-Diez, Org. Biomol. Chem.
2007, 5, 3682–3689.
[45] a) C. Girard, H. B. Kagan, Angew. Chem. 1998, 110, 3088–3127;
Angew. Chem. Int. Ed. 1998, 37, 2922–2959; b) T. Satyanarayana, S.
Abraham, H. B. Kagan, Angew. Chem. 2009, 121, 464–503; Angew.
Chem. Int. Ed. 2009, 48, 456–494.
[20] a) V. A. Stoute, M. A. Winnik, I. G. Csizmadia, J. Am. Chem. Soc.
1974, 96, 6388–6393; b) Y. Okuno, Chem. Eur. J. 1997, 3, 210–218;
c) R. Cꢄrdenas, R. Cetina, J. Lagfflnez-Otero, L. Reyes, J. Phys.
Chem. A 1997, 101, 192–200; d) R. Cꢄdenas, L. Reyes, J. Lagfflnez-
Otero, R. Cetina, J. Mol. Struct. 2000, 516–558, 211–225; e) H. Han-
nachi, N. Anoune, C. Arnaud, P. Lantꢃri, R. Longeray, H. Cherm-
ette, J. Mol. Struct. 1998, 440–473, 183–191; f) C. Lehtinen, V. Neva-
lainen, G. Brunow, Tetrahedron 2000, 56, 9375–9382.
[21] S. Yamabe, S. Yamazaki, J. Org. Chem. 2007, 72, 3031–3041.
[22] J. R. Alvarez-Idaboy, L. Reyes, J. Cruz, Org. Lett. 2006, 8, 1763–
1765.
[23] J. R. Alvarez-Idaboy, L. Reyes, J. Org. Chem. 2007, 72, 6580–6583.
[24] P. Carlqvist, R. Eklund, T. Brinck, J. Org. Chem. 2001, 66, 1193–
1199.
[25] a) R. R. Sever, T. W. Root, J. Phys. Chem. B 2003, 107, 10521–
10530; b) R. R. Sever, T. W. Root, J. Phys. Chem. B 2003, 107,
10848–10862.
[46] B. List, S. Mꢁller, Synfacts 2008, 640.
[47] Y. Ogata, K. Tomizawa, T. Ikeda, J. Org. Chem. 1978, 43, 2417–
2419.
[48] D. Uraguchi, M. Terada, J. Am. Chem. Soc. 2004, 126, 5356–5357.
[49] D. Uraguchi, K. Sorimachi, M. Terada, J. Am. Chem. Soc. 2005, 127,
9360–9361.
[50] a) S. J. Connon, Angew. Chem. 2006, 118, 4013–4016; Angew. Chem.
Int. Ed. 2006, 45, 3909–3912; b) T. Akiyama, J. Itoh, K. Fuchibe,
Adv. Synth. Catal. 2006, 348, 999–1010.
[51] a) T. Akiyama, Chem. Rev. 2007, 107, 5744–5758; b) M. Terada,
Chem. Commun. 2008, 4097–4112; c) L.-Z. Gong, X.-H. Chen, X.-
Y. Xu, Chem. Eur. J. 2007, 13, 8920–8926; d) A. G. Doyle, E. N. Ja-
cobsen, Chem. Rev. 2007, 107, 5713–5743.
[52] After many trials we were unable to synthesize the peroxy deriva-
tive of the BINOL phosphoric acid or its tert-butyl peroxyphosphate
analogue. Nevertheless, the analogue diethyl tert-butyl phosphate
and diethyl tert-butyl peroxyphosphate can be easily synthesized ac-
cording to literature procedures: a) N. Yoshihiro, A. Yasushi, Chem.
Pharm. Bull. 1986, 34, 3121–3129; b) G. Sosnovsky, E. H. Zaret, J.
[26] a) M. Boronat, A. Corma, M. Renz, G. Sastre, P. M. Viruela, Chem.
Eur. J. 2005, 11, 6905–6915; b) A. Corma, L. T. Nemeth, M. Renz,
3034
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 3021 – 3035

Asymmetric Baeyer–Villiger Reactions
FULL PAPER
Org. Chem. 1969, 34, 968–970. 31P NMR spectra indicate a down-
field change in chemical shift upon peroxyphosphate formation, see
Figure S1 in the Supporting Information.
Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani,
N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J.
Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C.
Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari,
J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cio-
slowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaro-
mi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng,
A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.
Chen, M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Wall-
ingford CT, 2004.
[53] a) D. G. Blackmond, Angew. Chem. 2005, 117, 4374–4393; Angew.
Chem. Int. Ed. 2005, 44, 4302–4320; b) J. S. Mathew, M. Klussmann,
H. Iwamura, F. Valera, A. Futran, E. A. C. Emanuelsson, D. G.
Blackmond, J. Org. Chem. 2006, 71, 4711–4722.
[54] A. Zanardo, F. Pinna, R. A. Michelin, G. Strukul, Inorg. Chem.
1988, 27, 1966–1973.
[55] M. Colladon, A. Scarso, P. Sgarbossa, R. A. Michelin, G. Strukul, J.
Am. Chem. Soc. 2007, 129, 7680–7689.
[56] T. Dorn, A.-C. Chamayou, C. Janiak, New J. Chem. 2006, 30, 156–
167.
[57] H. Lineweaver, D. Burk, J. Am. Chem. Soc. 1934, 56, 658–666.
[58] For examples: a) T. Akiyama, H. Morita, J. Itoh, K. Fuchibe, Org.
Lett. 2005, 7, 2583–2585; b) M. Terada, K. Machioka, K. Sorimachi,
Angew. Chem. 2006, 118, 2312–2315; Angew. Chem. Int. Ed. 2006,
45, 2254–2257; c) Y.-X. Jia, J. Zhong, S.-F. Zhu, C.-M. Zhang, Q.-L.
Zhou, Angew. Chem. 2007, 119, 5661–5663; Angew. Chem. Int. Ed.
2007, 46, 5565–5567; d) Q.-X. Guo, H. Liu, C. Guo, S.-W. Luo, Y.
Gu, L.-Z. Gong, J. Am. Chem. Soc. 2007, 129, 3790–3791; e) M.
Rueping, W. Leawsuwan, A. P. Antonchick, B. J. Nachtsheim,
Angew. Chem. 2007, 119, 2143–2146; Angew. Chem. Int. Ed. 2007,
46, 2097–2100; f) Q. Kang, Z.-A. Zhao, S.-L. You, J. Am. Chem.
Soc. 2007, 129, 1484–1485; g) Q. Kang, X.-J. Zhen, S.-L. You, Chem.
Eur. J. 2008, 14, 3539–3542; h) X. Wang, B. List, Angew. Chem.
2008, 120, 1135–1138; Angew. Chem. Int. Ed. 2008, 47, 1119–1122.
[59] L. Simon, J. M. Goodman, J. Am. Chem. Soc. 2008, 130, 8741–8747.
[60] T. Marcelli, P. Hammar, F. Himo, Chem. Eur. J. 2008, 14, 8562–
8571.
[61] L. Simon, J. M. Goodman, J. Am. Chem. Soc. 2009, 131, 4070–4077.
[62] M. Yamanaka, T. Hirata, J. Org. Chem. 2009, 74, 3266–3271.
[63] F. Q. Shi, B. A. Song, Org. Biomol. Chem. 2009, 7, 1292–1298.
[64] a) P. Hohenberg, W. Kohn, Phys. Rev. B 1964, 136, B864–B871;
b) W. Kohn, L. J. Sham, Phys. Rev. 1965, 140, A1133–A1138.
[65] a) M. Head-Gordon, J. A. Pople, M. J. Frisch, Chem. Phys. Lett.
1988, 153, 503–506; b) M. J. Frisch, M. Head-Gordon, J. A. Pople,
Chem. Phys. Lett. 1990, 166, 275–280; c) M. J. Frisch, M. Head-
Gordon, J. A. Pople, Chem. Phys. Lett. 1990, 166, 281–289.
[66] Gaussian 03, Revision D.01, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgom-
[67] K. M. Waltz, P. J. Carroll, P. J. Walsh, Organometallics 2004, 23, 127–
134.
[68] a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652; b) C. T. Lee,
W. T. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785–789.
[69] a) E. Cances, B. Mennucci, J. Tomasi, J. Chem. Phys. 1997, 107,
3032–3041; b) M. Cossi, V. Barone, B. Mennucci, J. Tomasi, Chem.
Phys. Lett. 1998, 286, 253–260.
[70] a) J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77,
3865–3868; b) J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett.
1997, 78, 1396–1396.
[71] C. Lehtinen, V. Nevalainen, G. Brunow, Tetrahedron 2001, 57, 4741–
4751.
[72] L. R. Krepski, A. Hassner, J. Org. Chem. 1978, 43, 2879–2882.
[73] For the synthesis of racemic BINOL, see: a) K. Ding, Y. Wang, L.
Zhang, Y. Wu, T. Matsuura, Tetrahedron 1996, 52, 1005–1010; for
the resolution of BINOL, see: b) Y. Wang, J. Sun, K. Ding, Tetrahe-
dron 2000, 56, 4447–4451; c) H. Du, Y. Wang, J. Sun, J. Meng, K.
Ding, Tetrahedron Lett. 2002, 43, 5273–5276; for the synthesis of
H8-BINOL, see: d) H. Guo, K. Ding, Tetrahedron Lett. 2000, 41,
10061–10064; for the synthesis of BINOL and H8-BINOL deriva-
tives, see: e) J. Long, J. Hu, X. Shen, B. Ji, K. Ding, J. Am. Chem.
Soc. 2002, 124, 10–11; for the synthesis of chiral phosphoric acids,
see: f) J. Inanaga (Tosoh Corporation, Japan), EP1134209, 2001.
Received: September 30, 2009
A
Published online: January 27, 2010
Chem. Eur. J. 2010, 16, 3021 – 3035
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3035