Origins of enantioselectivity in the chiral Bronsted acid catalyzed hydrophosphonylation of imines

Article abstract of DOI:10.1039/b815008g

The results of an experimental and ONIOM-based computational investigation of the mechanism and the origins of enantioselectivity in the asymmetric synthesis of α-amino phosphonates by an enantioselective hydrophosphonylation of imines catalyzed by chiral

Full text of DOI:10.1039/b815008g

View Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Origins of enantioselectivity in the chiral Brønsted acid catalyzed
hydrophosphonylation of imines†
Fu-Qiang Shi* and Bao-An Song
Received 28th August 2008, Accepted 5th January 2009
First published as an Advance Article on the web 5th February 2009
DOI: 10.1039/b815008g
The results of an experimental and ONIOM-based computational investigation of the mechanism and
the origins of enantioselectivity in the asymmetric synthesis of a-amino phosphonates by an
enantioselective hydrophosphonylation of imines catalyzed by chiral Brønsted acids are reported. It
was found that the enantioselectivity observed in the enantioselective hydrophosphonylation of the
imine with a benzothiazole moiety was poor. A detailed computational study with a two-layer ONIOM
(B3LYP/6–31G(d)/AM1) method on the mechanism of the investigated reaction was carried out to
explore the origins of the enantioselectivity. Calculations indicate that the investigated reaction is a
two-step process involving proton-transfer and nucleophilic addition, which is the stereo-controlling
step. The investigated reaction prefers a di-coordination pathway to a mono-coordination pathway. The
different enantioselectivities exhibited by three kinds of catalyst and two kinds of nucleophile were
rationalized. Calculations indicate that si-facial attack is higher in energy than re-facial attack by only
0.1 kcal/mol, which accounts well for the low ee value observed in the enantioselective
hydrophosphonylation of the imine with a benzothiazole moiety. The energy barrier for
phosphonate–phosphite tautomerism catalyzed by chiral Brønsted acid in toluene is only 1.8 kcal/mol,
which could explain why the investigated reaction can take place at room temperature.
of optically active a-amino phosphonates.16,17 Recently, Akiyama,
Introduction
et al. reported a simple and efficient synthetic protocol catalyzed
by chiral Brønsted acid 1 to asymmetrically synthesize a-amino
phosphonates.18 a-Amino phosphonates were obtained in moder-
ate yields (Scheme 1). The enantioselectivity could be improved
to 90% in cases where diisopropylphosphite was added to imines
The development of efficient methods of accessing complex chiral
molecules with biological activities continues to be one of the
most challenging topics in modern synthetic organic chemistry.1
Although a number of metal-based Lewis acid catalysts have been
developed to activate carbon–carbon and carbon–nitrogen double
bonds, metal-free chiral organocatalysts, such as chiral Brønsted
acids,2–4 are generally preferred because of their stability toward
water and oxygen.
=
having o-X-C6H4-CH CH substituents at the carbon (where X is
an electron-withdrawing group).
a-Amino phosphonic acids and their derivatives play an
important role in chemistry because they often exhibit intriguing
biological activities. For example, they can be used as inhibitors of
enzymes such as EPSP synthase and HIV protease.5 Recently, a lot
of a-amino phosphonates containing heterocycle moieties, such
as thiophene, furan, pyrrole, 1,3,4-thiadiazole, and benzothiazole,
were found to possess a wide range of antitumor, antiviral, and
antifungal properties and to have been widely used as insecticides
and herbicides.6–12
Diastereoselective addition of phosphite derivatives to chiral
imines,13 chiral Lewis acid- or chiral Brønsted acid-catalyzed
enantioselective addition of phosphites to imines,14 and other
methods15 have been successfully developed for the preparation
Scheme 1
Among the test phosphoric acids 1, 1a exhibited the lowest
enantioselectivity (23% ee) and 1c exhibited the highest enantiose-
lectivity (43% ee) and reactivity (99% yield). The enantioselectivity
increased when the nucleophile was changed from diethyl phos-
phonate (43% ee in toluene) to diisopropyl phosphonate (52% ee
in m-xylene).18 The study also found that the presence of an
o-hydroxy moiety decreased the enantioselectivity (88% ee vs 39%
ee). This indicates that the structures of imine and catalyst play an
important role in affecting the reaction’s enantioselectivity.
With the information in hand, we proceeded to synthesize
enantiopure a-amino phosphonates with heterocycle moieties
Key Laboratory of Green Pesticide and Agricultural Bioengineering, Ministry
of Education, Center for Research and Development of Fine Chemicals,
Guizhou University, Huaxi District, Guiyang 550025, PR China. E-mail:
fcc.fqshi@gzu.edu.cn
† Electronic supplementary information (ESI) available: Discussion of
computational methods; ONIOM evaluation; selected topological param-
eters at bond critical points in int1 and TS4; optimized structures of
the investigated system; Cartesian coordinates of the stationary points
discussed in the text. See DOI: 10.1039/b815008g
1292 | Org. Biomol. Chem., 2009, 7, 1292–1298
This journal is
The Royal Society of Chemistry 2009
©

View Online
Scheme 2
to investigate their biological activities. However, we found the
enantioselectivity of the chiral Brønsted acid, 1c, catalyzed enan-
tioselective hydrophosphonylation of imines with a benzothiazole
moiety was quite poor (Scheme 2).19
energies in solution were calculated at the B3LYP/D95(d,p) level
on the geometries optimized in the gas phase with the conductor-
like polarizable continuum model (CPCM) using UAKS radii.29
The dielectric constant in the CPCM calculations was set to
e = 2.379 to simulate toluene as the solvent medium used
in the experiments. The wave function was obtained at the
B3LYP/D95(d,p) level for the topological analysis. Topological
analysis of the electron densities at bond critical points was
performed with the AIM2000 program.30 In what follows, the
discussed energies are relative Gibbs free energies in toluene
without zero-point energy correction.
This finding motivated us to study the mechanism and the
origins of enantioselectivity in the enantioselective hydrophos-
phonylation of imines catalyzed by chiral Brønsted acids in
detail. Recently, Yamanaka, Akiyama et al. reported a combined
experimental and theoretical study on the chiral Brønsted acid-
catalyzed enantioselective Mannich-type reaction of ketene silyl
acetals with aldimines.23 The theoretical study elucidated that the
two-point hydrogen bonding interaction resulted in the preference
for a di-coordination pathway and the reaction proceeded through
protonation followed by nucleophilic attack via a nine-membered
cyclic TS. It is more important that the origins of enantioselectivity
in the nucleophilic addition are well elucidated.20 On the other
hand, theoretical studies of the mechanism of organocatalytic and
organometallic reactions have been confirmed to be helpful in
understanding reaction results21 and to give guidance for the future
design of new catalysts and new reactions.22
Inspired by this, we carried out a detailed computational
study on the mechanism of the chiral Brønsted acid-catalyzed
enantioselective hydrophosphonylation of aldimine. The present
mechanistic study will provide an in-depth understanding of the
details of the chiral Brønsted acid-catalyzed C–P bond formation
and explain the origins of the low enantioselectivity in the
enantioselective hydrophosphonylation of imine with a benzoth-
iazole moiety. The mechanistic insights could also be useful for
understanding other Brønsted acid-catalyzed reactions23 and to
guide the design of new chiral Brønsted acids for asymmetric
C–O, C–N, and C–P bond formations.24
Computational results and discussion
In this section, we will first discuss the mechanism of the chiral
Brønsted acid-catalyzed enantioselective hydrophosphonylation
of aldimines and explore the two possible pathways, mono-
coordination and di-coordination. Then, we will investigate the
substituent effects on the enantioselectivity. Finally, we will explore
the origins of the low enantioselectivity in the enantioselective hy-
drophosphonylation of the imine with a benzothiazole moiety. In
addition, we will discuss the phosphonate–phosphite tautomerism
catalyzed by chiral Brønsted acid.
Mechanism of the chiral Brønsted acid-catalyzed enantioselective
hydrophosphonylation of aldimine
Fig. 1 shows the computed potential energy surfaces for the
Brønsted acid-catalyzed enantioselective hydrophosphonylation
of aldimine via a di-coordination pathway (two substrates form
hydrogen bonds with different oxygen atoms of the Brønsted acid)
in solution. The optimized structures involved in this reaction
are given in Fig. 2. In the first step of the hydrophosphonylation
of the aldimine, the chiral Brønsted acid, aldimine, and diethyl
phosphite first form a hydrogen-bond complex int1, which is
stabilized mainly by three favorable hydrogen bonding interactions
and whose stability is also affected by the steric hindrance between
the 3,3¢-aryl substituents of the Brønsted acid and the substrates.
Calculations indicate that (S)-int1 is slightly more stable than
(R)-int1. In (R)-int1, one strong stabilizing interaction is the
O3–H3 ◊ ◊ ◊ O2 hydrogen bonding interaction between the Brønsted
acid catalyst and diethyl phosphite, as demonstrated by the
Computational methods
The layer treatment (two-layered ONIOM method) implemented
in the Gaussian 98 program package25 was used to obtain a
description of the potential energy surface (PES). The model is
divided into a high layer that is treated at the B3LYP/6-31G(d)
level of theory26,27 for the critical bond cleavage atoms and a low
layer that is treated with the AM1 method.28 Selected carbon–
oxygen and carbon–carbon bonds were cut and saturated with
hydrogen link atoms during the generation of the model subsystem
to avoid possible chemical artifacts. Frequency calculations at the
same level were performed to confirm each stationary point as
either a minimum or a transition structure (TS). Single-point
˚
H3 ◊ ◊ ◊ O2 distance (1.795 A) and nearly linear O3–H3 ◊ ◊ ◊ O2 geom-
etry (177.4◦). Another stabilizing interaction is the O1–H1 ◊ ◊ ◊ N
hydrogen bonding interaction, which is also strong indicated
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 1292–1298 | 1293
©

View Online
C2–H2 ◊ ◊ ◊ O2 interaction between the Brønsted acid and the
aldimine. The topological analyses at the bond critical point
show this interaction is a normal hydrogen bond and their
charge densities and Laplacians fall into the proposed range
of 0.002–0.035 au and 0.014–0.139 au, respectively.31 The O1–
H1 ◊ ◊ ◊ N and O3–H3 ◊ ◊ ◊ O2 interactions in (S)-int1 are stronger
than the corresponding ones in (R)-int1, which is demonstrated
˚
by the corresponding hydrogen bond distances (1.440 A vs.
˚
˚
˚
1.460 A and 1.779 A vs. 1.795 A). The negative Laplacian at
the bond critical point between H1 and N (-0.025 au) indicates
that the H1 ◊ ◊ ◊ N interaction is partly covalent in nature.32 The
C4–H4 ◊ ◊ ◊ O2 interaction in (S)-int1 is stronger than the C2–
H2 ◊ ◊ ◊ O2 in (R)-int1, which is demonstrated by the corresponding
Fig. 1 Potential energy surfaces of the chiral Brønsted acid-catalyzed
enantioselective hydrophosphonylation of aldimine by a di-coordination
pathway.
˚
˚
intermolecular distances (2.266 A vs. 2.337 A). However, (S)-
int1 is only more stable than (R)-int1 by 0.3 kcal/mol, which
implies that the steric hindrance may be more significant in (S)-
int1. These complexation processes are endergonic by 9.3 and
9.0 kcal/mol and exothermic by 22.0 and 23.7 kcal/mol in terms of
electronic energy, respectively, which is due to the entropy penalty
for bringing three molecules together (see ESI).†
˚
by both the H1 ◊ ◊ ◊ N distance (1.460 A) and O1–H1 ◊ ◊ ◊ N angle
(168.2◦). The charge density and its Laplacian at the bond critical
point between H1 and N are 0.101 and -0.002 au, respectively.
The charge density and its Laplacian at the critical point between
H3 and O2 (0.034 and 0.106 au) indicate that the O3–H3 ◊ ◊ ◊ O2
interaction is weaker than the O1–H1 ◊ ◊ ◊ N interaction, which
is consistent with the distance difference between H3 ◊ ◊ ◊ O2 and
H1 ◊ ◊ ◊ N. The large charge density shows that the O1–H1 ◊ ◊ ◊ N
hydrogen bond is stronger than normal hydrogen bonds31 and
has some ionic characteristics. The O1–H1 ◊ ◊ ◊ N angle in (R)-
int1 deviates from 180◦ by about 11.8◦, which results from the
Complex int1 leads to intermediate int3 via a proton-transfer
transition structure TS2. The protonation of aldimine generates
a zwitterionic complex of the iminium salt and is very easy, as
demonstrated by the activation free energy (0.2 kcal/mol for (R)-
int1 and 0.4 kcal/mol for (S)-int1). Both intermediates, int3, are
more stable than the corresponding reactants, int1. This is due
to the fact that int3 is more compact than int1. For example,
the H3 ◊ ◊ ◊ O2 distance in (S)-int3 is shorter than that in (S)-int1
Fig. 2 Optimized structures of the chiral Brønsted acid-catalyzed enantioselective hydrophosphonylation of aldimine by a di-coordination pathway.
˚
Bond lengths are in A.
1294 | Org. Biomol. Chem., 2009, 7, 1292–1298
This journal is
The Royal Society of Chemistry 2009
©

View Online
˚
˚
(1.698 A vs. 1.779 A). Obviously, the steric hindrance of the 3,3¢-
aryl substituents also contributes to the stability increase of int3.
The aldimine moiety in int3 is flatter than that in int1, which
results from less steric hindrance. These proton-transfer processes
are exergonic by 3.5 and 5.4 kcal/mol, respectively. Though the
hydrogen bonding interactions in (S)-int3 are stronger than those
in (R)-int3, calculations show that (R)-int3 is more stable than
(S)-int3 by 1.6 kcal/mol. This further indicates that steric factors
also contribute toward the energy differences.
Intermediate int3 evolves to the products via a nine-membered
transition structure, TS4. This transformation is a concerted
process involving C–P bond formation and proton-transfer.
(R)-TS4 is more stable than its counterpart, (S)-TS4, by
5.2 kcal/mol. The structural properties allow us to examine the
origins of the instability of (S)-TS4 in detail. The shorter H1 ◊ ◊ ◊ O1
distance indicates that the N1–H1 ◊ ◊ ◊ O1 hydrogen bonding inter-
action in (R)-TS4 is stronger than that in (S)-TS4, however, the
C3–H3 ◊ ◊ ◊ O2 hydrogen-bonding interaction in (S)-TS4 is stronger
than that in (R)-TS4. Furthermore, the electrostatic attraction
between the lone pair electrons of the phosphorus atom in dialkyl
Fig. 3 Optimized stereo-controlling transition structures of the chi-
ral Brønsted acid-catalyzed enantioselective hydrophosphonylation of
˚
aldimine by a mono-coordination pathway. Bond lengths are in A.
to reach (R)-TS4 in the di-coordination pathway (5.9 kcal/mol).
In addition, (R)-TS4-mcp and (S)-TS4-mcp are higher in energy
than (R)-TS4 by 4.6 and 1.6 kcal/mol, respectively. This indicates
that the di-coordination pathway is more favorable. It is noted
that (R)-TS4-mcp is higher in energy than (S)-TS4-mcp, which
implies that more (S)-product should be formed instead. This is not
consistent with the experimental result. Therefore, the investigated
reactions must be carried out via the di-coordination pathway,
which is similar to the situation observed in the chiral Brønsted
acid-catalyzed enantioselective Mannich-type reaction of ketene
silyl acetals with aldimines.19
=
phosphite and the partly positive carbon atom in the C N double
bond in (S)-TS4 is stronger than that in (R)-TS4, which is
˚
˚
demonstrated by the C–P distances (2.345 A vs. 2.438 A) and
relative large electron density. However, the steric hindrance of the
3,3¢-aryl substituents in (S)-TS4 is more significant than that in
(R)-TS4, demonstrated by the aldimine moiety geometries. The
steric hindrance of the 3,3¢-aryl substituents in (S)-TS4 has forced
the aldimine moiety to deviate from the stable planar structure.
The benzene ring linking with the N atom is almost parallel to
one of the 3,3¢-aryl substituents to minimize the steric hindrance,
however, the other 3,3¢-aryl substituent causes significant steric
hindrance with the aldimine and diethyl phosphate. Therefore,
(S)-TS4 is less stable than its counterpart, (R)-TS4. This results
in the fact that re-facial attack is 5.2 kcal/mol more favored than
si-facial attack. The predicted ee for the above catalysts should
be >99%, however, the experimental ee is only moderate (up to
43%). Although the mismatch between experiment and theory is
relatively large, the calculations could well explain the trend of the
increase in ee predicted. The use of extended basis sets did not
change the trend (see ESI).† The O–H of the diethyl phosphite
Substituent effects on the enantioselectivity of the enantioselective
hydrophosphonylation of aldimine
To explore the issue of the enantioselectivity in the present
reaction, we compared the transition structures involving the
three Brønsted acid catalysts with different 3,3¢-aryl substituents
used in the experiment. The optimized transition structures and
relative stabilities are shown in Fig. 4. The energy difference
between si-facial attack and re-facial attack is as follows: TS4
> TS4-2 > TS4-3, which indicates that the catalyst 1c should
induce the best enantioselectivity among the three test catalysts.
This is in agreement with experimental results. Calculations
indicate that the energy difference between the two enantioselective
transition structures increased when the nucleophile was changed
from diethyl phosphite (5.2 kcal/mol) to diisopropyl phosphite
(6.8 kcal/mol), which is also consistent with the experiment.
Therefore, the substituent groups in both catalyst and imine
affect the enantioselectivities. The sterically repulsive interaction
between the 3,3¢-substituents of the catalyst and the reaction
substrates is sensitive to the nature of the substituents on the
aryl group and the dialkyl phosphite, therefore, the highest
enantioselectivity was observed when chiral Brønsted acid 1c and
diisopropyl phosphite were used. These results could well explain
the origins of enantioselectivity in the enantioselective hydrophos-
phonylation of aldimine catalyzed by chiral Brønsted acid.33
When the origin of the enantioselectivity in the chiral Brønsted
acid-catalyzed enantioselective hydrophosphonylation of aldimine
was clear, we investigated further why the enantioselectivity of
the 1c-catalyzed enantioselective hydrophosphonylation of imine
with a benzothiazole moiety was poor. The optimized transition
structures and relative stabilities are shown in Fig. 5. Calculations
indicate that (S)-TS4-5 is only 0.1 kcal/mol higher in energy
˚
moiety in (R)-TS4 is 1.027 A, indicating that proton transfer from
the diethyl phosphite moiety to the oxygen of the Brønsted acid
occurs later than the C–P bond formation. The shortest distance in
˚
TS4 between two phenyl rings is larger than 4.5 A, which indicates
that the aromatic stacking interaction is very weak. This concerted
process requires an activation free energy of 5.9 kcal/mol for
re-facial attack and 9.5 kcal/mol for si-facial attack. The (S)-
product is 2.4 kcal/mol higher in energy than the reactants and
the (R)-product is 0.3 kcal/mol higher in energy than the reactants.
So formation of the (R)-product is both thermodynamically and
kinetically favorable.
We also explored the possible mono-coordination (two sub-
strates form hydrogen bonds with the same oxygen atom of
the Brønsted acid) pathway. The optimized transition structures
of the stereo-controlling step for the Brønsted acid-catalyzed
enantioselective hydrophosphonylation of aldimines via a mono-
coordination pathway are shown in Fig. 3. The free energies,
required to reach (R)-TS4-mcp and (S)-TS4-mcp, are 6.4 and
8.3 kcal/mol respectively, which are greater than that required
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 1292–1298 | 1295
©

View Online
Fig. 4 Optimized selected structures of the chiral Brønsted-acid catalyzed enantioselective hydrophosphonylation of aldimine by a di-coordination
˚
pathway. Bond lengths are in A.
Fig. 5 Optimized stereo-controlling transition states of the chiral Brønsted-acid catalyzed enantioselective hydrophosphonylation of imine with a
˚
benzothiazole moiety. Bond lengths are in A.
than (R)-TS4-5, which can well account for the fact that the
enantioselectivity of the 1c catalyzed enantioselective hydrophos-
phonylation of imines with a benzothiazole moiety is poor. The
structural properties of (S)-TS4-5 and (R)-TS4-5 were examined
to investigate why an energy difference exists between them. The
O–H ◊ ◊ ◊ O hydrogen bonding and P–C electrostatic interactions in
(S)-TS4-5 are stronger than those in (R)-TS4-5, which is demon-
strated by the corresponding distances. However, the situation is
the opposite for the N–H ◊ ◊ ◊ O hydrogen bonding interaction. The
sterically repulsive interactions between the 3,3¢-substituents of
the catalyst and substrates in TS4-5 are not so significant as those
in TS4, demonstrated by the imine moiety geometries. The imine
moiety in (S)-TS4-5 is almost as planar as that in (R)-TS4-5,
which is probably caused by the relatively small thiazole moiety
and no C–H ◊ ◊ ◊ O hydrogen bond being formed between the thia-
zole moiety and catalyst in (S)-TS4-5. This indicates that the
sterically repulsive interactions between the 3,3¢-substituents of the
catalyst and substrates in (S)-TS4-5 are not significant, which is
not the case for (S)-TS4. Therefore, the energy difference between
si-facial attack and re-facial attack for TS4-5 is small and the
enantioselectivity of the 1c-catalyzed enantioselective hydrophos-
phonylation of imine with a benzothiazole moiety was poor.
Further experimental and theoretical investigations are ongoing.
Mechanism of phosphonate–phosphite tautomerism
It is known that phosphonate–phosphite tautomerism exists with
the phosphite form as the active nucleophile form and the
1296 | Org. Biomol. Chem., 2009, 7, 1292–1298
This journal is
The Royal Society of Chemistry 2009
©

View Online
phosphonate tautomer as the almost exclusively favored but non-
nucleophilic form.18,34 We found that the investigated reaction
could not be carried out at room temperature without Brønsted
acid. That is to say, the hydrophosphonylation of imines can take
place only if there is phosphite in the reaction and phosphonate–
phosphite tautomerism can not be carried out at room tem-
perature without a Brønsted acid catalyst. Furthermore, most
one-pot syntheses of a-amino phosphonates have to be carried
out under heating conditions. Therefore, we envisioned that the
Brønsted acid might be involved in the transition structure of
phosphonate–phosphite tautomerism. The optimized transition
structures and the energy barriers of the catalyst-free transition
structure TS-tau-1 and the Brønsted acid-catalyzed transition
structure TS-tau-2 are shown in Fig. 6. It can be seen from
Fig. 6 that, with its activation free energy of 33.7 kcal/mol, it
is difficult for the phosphonate–phosphite tautomerism to take
place without catalysis by a Brønsted acid. However, tautomerism
catalyzed by a Brønsted acid is likely, with an activation energy
of only 1.8 kcal/mol. This could well account for why the
hydrophosphonylation of imines catalyzed by Brønsted acid
could occur at room temperature or even lower temperature, but
hydrophosphonylation of imines without catalyst needs additional
energy (heat or microwave).
of imine has to be carried out under heating conditions. Further
design and synthesis of highly enantiopure a-amino phosphonates
with heterocycle moieties are in progress.
Acknowledgements
Supported by the National Science Foundation of China
(20802011), the Key Project of Chinese Ministry of Education
(208126), the Special Foundation of Government of Guizhou
Province (200818), and Talent Project of Guizhou University
(2007023).
References
1 (a) R. M. Williams and J. A. Hendrix, Chem. Rev., 1992, 92, 889–
917; (b) J.-A. Ma, Angew. Chem., Int. Ed., 2003, 42, 4290–4299; (c) H.
Gro¨ger, Chem. Rev., 2003, 103, 2795–2828; (d) K. Maruoka and T. Ooi,
Chem. Rev., 2003, 103, 3013–3028; (e) M. J. O’Donnell, Acc. Chem. Res.,
2004, 37, 506–517; (f) B. Lygo and B. I. Andrews, Acc. Chem. Res., 2004,
37, 518–525; (g) G. Desimoni, G. Faita and K. A. Jørgensen, Chem.
Rev., 2006, 106, 3561–3651; (h) J.-C. Kizirian, Chem. Rev., 2008, 108,
140–205.
2 (a) T. Akiyama, J. Itoh and K. Fuchibe, Adv. Synth. Catal., 2006, 348,
999–1010; (b) T. Akiyama, J. Itoh, K. Yokota and K. Fuchibe, Angew.
Chem., Int. Ed., 2004, 43, 1566–1568; (c) J. Itoh, K. Fuchibe and T.
Akiyama, Angew. Chem., Int. Ed., 2006, 45, 4796–4798; (d) T. Akiyama,
Y. Tamura, J. Itoh, H. Morita and K. Fuchibe, Synlett, 2006, 141–143;
(e) T. Akiyama, Y. Honma, J. Itoh and K. Fuchibe, Adv. Synth. Catal.,
2008, 350, 399–402; (f) J. Itoh, K. Fuchibe and T. Akiyama, Angew.
Chem., Int. Ed, 2008, 47, 4016–4018.
3 (a) D. Uraguchi and M. Terada, J. Am. Chem. Soc., 2004, 126, 5356–
5357; (b) D. Uraguchi, K. Sorimachi and M. Terada, J. Am. Chem.
Soc., 2004, 126, 11804–11805; (c) M. Terada, K. Machioka and K.
Sorimachi, Angew. Chem., Int. Ed., 2006, 45, 2254–2257; (d) M. Terada
and K. Sorimachi, J. Am. Chem. Soc, .2007, 129, 292–293; (e) M.
Terada, S. Yokoyama, K. Sorimachi and D. Uraguchi, Adv. Synth.
Catal., 2007, 349, 1863–1867; (f) M. Terada, K. Machioka and K.
Sorimachi, J. Am. Chem. Soc., 2007, 129, 10336–10337; (g) I. D.
Gridnev, M. Kouchi, K. Sorimachi and M. Terada, Tetrahedron Lett.,
2007, 48, 497–500.
4 (a) Q. Kang, Z.-A. Zhao and S.-L. You, J. Am. Chem. Soc., 2007, 129,
1484–1485; (b) Q. Kang, Z.-A. Zhao and S.-L. You, Adv. Synth. Catal.,
2007, 349, 1657–1660; (c) Q. Kang, Z.-A. Zhao and S.-L. You, Org.
Lett., 2008, 10, 2031–2034; (d) Q. Kang, X.-J. Zheng and S.-L. You,
Chem.–Eur. J., 2008, 14, 3539–3542; (e) M. Rueping, A. P. Antonchick
and T. Theissmann, Angew. Chem., Int. Ed., 2006, 45, 6751–6755; (f) M.
Rueping, A. P. Antonchick and T. Theissmann, Angew. Chem., Int. Ed,
2006, 45, 3683–3686; (g) M. Rueping, E. Sugiono and C. Azap, Angew.
Chem., Int. Ed., 2006, 45, 2617–2619; (h) M. Rueping, E. Sugiono and S.
A. Moreth, Adv. Synth. Catal., 2007, 349, 759–764; (i) M. Rueping and
C. Azap, Angew. Chem., Int. Ed., 2006, 45, 7832–7835; (j) M. Rueping,
W. Ieawsuwan, A. P. Antonchick and B. J. Nachtsheim, Angew. Chem.,
Int. Ed., 2007, 46, 2097–2100; (k) M. Rueping, E. Sugiono, C. Azap,
T. Theissmann and M. Bolte, Org. Lett., 2005, 7, 3781–3783; (l) M.
Rueping, E. Sugiono, T. Theissmann, A. Kuenkel, A. Kochritz, A.
Pews-Davtyan, N. Nemati and M. Beller, Org. Lett., 2007, 9, 1065–
1068; (m) M. Rueping, B. J. Nachtsheim, S. A. Moreth and M. Bolte,
Angew. Chem., Int. Ed., 2008, 47, 593–596; (n) H. Liu, L.-F. Cun,
A.-Q. Mi, Y.-Z. Jiang and L.-Z. Gong, Org. Lett., 2006, 8, 6023–6026;
(o) X.-H. Chen, X.-Y. Xu, H. Liu, L.-F. Cun and L.-Z. Gong, J. Am.
Chem. Soc., 2006, 128, 14802–14803; (p) Q.-X. Guo, H. Liu, C. Guo,
S.-W. Luo, Y. Gu and L.-Z. Gong, J. Am. Chem. Soc., 2007, 129, 3790–
3791.
Fig. 6 Optimized structures of TSs for phosphonate–phosphite tau-
tomerism. Bond lengths are in A.
˚
Conclusions
In this paper, we have reported a joint experimental and com-
putational investigation on the mechanism and the origins of
enantioselectivity in the enantioselective hydrophosphonylation
of imines catalyzed by chiral Brønsted acids. Calculations indicate
that the investigated reaction involves tandem proton transfer and
nucleophilic addition. The first proton transfer is very easy with
an activation free energy less than 1.0 kcal/mol. The formation of
the C–P bond and the second proton-transfer occur in an asyn-
chronous concerted process, which is the stereo-controlling step.
Calculations show the di-coordination pathway is favorable. The
energy differences between si-facial attack and re-facial attack are
significant for the hydrophosphonylation of aldimine. Calculations
indicate that re-facial attack of the imine with a benzothiazole moi-
ety is only favorable by 0.1 kcal/mol, which well accounts for the
low ee value observed in the enantioselective hydrophosphonyla-
tion. The energy barrier for phosphonate–phosphite tautomerism
catalyzed by chiral Brønsted acid in toluene is only 1.8 kcal/mol,
which can explain why the investigated reaction can be performed
at room temperature and the catalyst-free hydrophosphonylation
5 (a) P. Kafarski and B. Lejczak, Phosphorus, Sulfur Silicon Relat. Elem.,
1991, 63, 193–215; (b) R. Hirschmann, A. B. Smith, III, C. M. Taylor,
P. A. Benkovic, S. D. Taylor, K. M. Yager, P. A. Sprengeler and S.
J. Benkovic, Science, 1994, 265, 234–237; (c) W. W. Smith and P. A.
Bartlett, J. Am. Chem. Soc., 1998, 120, 4622–4628; (d) D. V. Patel, K.
Rielly-Gauvin, D. E. Ryono, C. A. Free, W. L. Rogers, S. A. Smith, J.
M. DeForrest, R. S. Oehl and E. W. Petrillo, J. Med. Chem., 1995, 38,
4557–4569; (e) B. Stowasser, K.-H. Budt, L. Jian-Qi, A. Peyman and
D. Ruppert, Tetrahedron Lett., 1992, 33, 6625–6628.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 1292–1298 | 1297
©

View Online
6 (a) S. Yang, X. W. Gao, C. L. Diao, B. A. Song, L. H. Jin, G. F. Xu,
G. P. Zhang, W. Wang, D. Y. Hu, W. Xue and P. Lu, Chin. J. Chem.,
2006, 24, 1581–1588; (b) G.-P. Zhang, B.-A. Song, W. Xue, L.-H. Jin,
D.-Y. Hu, Q.-Q. Wan, P. Lu, S. Yang, Q.-Z. Li and G. Liu, J. Fluorine
Chem., 2006, 127, 48–53; (c) Y.-S. Xu, K. Yan, B.-A. Song, G.-F. Xu, S.
Yang, W. Xue, D.-Y. Hu, P. Lu, G.-P. Ouyang, L.-H. Jin and Z. Chen,
Molecules, 2006, 11, 666–676; (d) B.-A. Song, Y.-L. Wu, S. Yang, D.-Y.
Hu, X.-Q. He and L.-H. Jin, Molecules, 2003, 8, 186–192; (e) D. Y. Hu,
Q. Q. Wan, S. Yang, B.-A. Song, P. S. Bhadury, L. H. Jin, K. Yan, F.
Liu, Z. Chen and W. Xue, J. Agric. Food Chem., 2008, 56, 998–1001.
7 L. H. Jin, B. A. Song, G. P. Zhang, R. Q. Xu, S. M. Zhang, X. W. Gao,
D. Y. Hu and S. Yang, Bioorg. Med. Chem. Lett., 2006, 16, 1537–1543.
8 N. Dieltiens, K. Moonen and C. V. Stevens, Chem.–Eur. J., 2006, 13,
203–214.
9 T. K. Olszewski, B. Boduszek, S. Sobek and H. Kozlowski, Tetrahedron,
2006, 62, 2183–2189.
10 P. Kafarski and B. Lejczak, Curr. Med. Chem. Anti-Cancer Agents,
2001, 1, 301–312.
11 T. Lintunen and J. T. Yli-Kauhaluoma, Bioorg. Med. Chem. Lett., 2000,
10, 1749–1750.
(h) Y. Huang, A. K. Unni, A. N. Thadani and V. H. Rawal, Nature,
2003, 424, 146–146; (i) B. M. Nugent, R. A. Yoder and J. N. Johnston,
J. Am. Chem. Soc., 2004, 126, 3418–3419; (j) N. T. McDougal, W. L.
Trevellini, S. A. Rodgen, L. T. Kliman and S. E. Schaus, Adv. Synth.
Catal., 2004, 346, 1231–1240; (k) N. Momiyama and H. Yamamoto, J.
Am. Chem. Soc., 2005, 127, 1080–1081; (l) A. K. Unni, N. Takenaka, H.
Yamamoto and V. H. Rawal, J. Am. Chem. Soc., 2005, 127, 1336–1337;
(m) K. Matsui, S. Takizawa and H. Sasai, J. Am. Chem. Soc., 2005, 127,
3680–3681.
22 S. Niu and M. B. Hall, Chem. Rev., 2000, 100, 353–406.
23 M. Yamanaka, J. Itoh, K. Fuchibe and T. Akiyama, J. Am. Chem. Soc.,
2007, 129, 6756–6764.
24 For recent theoretical studies of transition metal-catalyzed reactions,
see: (a) F.-Q. Shi, X. Li, X. Y. Xia, L. Zhang and Z.-X. Yu, J. Am.
Chem. Soc., 2007, 129, 15503–15512; (b) C. Nieto-Oberhuber, M. P.
Mun˜oz, E. Bun˜uel, C. Nevado, D. J. Ca´rdenas and A. M. Echavarren,
Angew. Chem., Int. Ed., 2004, 43, 2402–2406; (c) C. Nieto-Oberhuber,
S. Lo´pez, E. Jime´nez-Nu´n˜ez and A. M. Echavarren, Chem. Eur. J.,
2006, 12, 5916–5923; (d) C. Nieto-Oberhuber, S. Lo´pez, M. P. Mun˜oz,
D. J. Ca´rdenas, E. Bun˜uel, C. Nevado and A. M. Echavarren, Angew.
Chem., Int. Ed., 2005, 44, 6146–6148; (e) Y.-D. Wu and Z.-X. Yu, J.
Am. Chem. Soc., 2001, 123, 5777–5786; (f) Z.-X. Yu and K. N. Houk,
J. Am. Chem. Soc., 2003, 125, 13825–13830; (g) Z.-X. Yu, P. A. Wender
and K. N. Houk, J. Am. Chem. Soc., 2004, 126, 9154–9155; (h) O. N.
Faza, C. S. Lopez, R. Alvarez and A. R. de Lera, J. Am. Chem. Soc.,
2006, 128, 2434–2437.
25 Gaussian 98 (Revision A.11), M. J. Frisch, G. W. Trucks, H. B. Schlegel,
G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G. Zakrzaweski,
J. A. Montgomery, R. E. Stratmann, J. C. Burant, S. Dapprich, J.
M. Millam, A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J.
Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli,
C. Adamo, S. Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q.
Cui, K. Morokuma, D. K. Malick, A. D. Rabuck, K. Raghavachari,
J. B. Foresman, J. Ciolowski, J. V. Ortiz, B. B. Stefanov, G. Liu, A.
Liashenko, P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin, D.
J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, C.
Gonzalez, M. Challacombe, P. M. W. Gill, B. G. Johnson, W. Chen, M.
W. Wong, J. L. Andres, M. Head-Gordon, E. S. Replogle and J. Pople,
Gaussian, Inc., Pittsburgh, PA, 1998.
12 W. Liu, C. J. Rogers, A. J. Fisher and M. D. Toney, Biochemistry, 2002,
41, 12320–12328.
13 (a) K. M. Yager, C. M. Taylor and A. B. Smith, III, J. Am. Chem. Soc.,
1994, 116, 9377–9378; (b) A. B. Smith, III, K. M. Yager and C. M.
Taylor, J. Am. Chem. Soc., 1995, 117, 10879–10888; (c) I. M. Lefebvre
and S. A. Evans, Jr., J. Org. Chem, 1997, 62, 7532–7533; (d) F. A. Davis,
S. Lee, H. Yan and D. D. Titus, Org. Lett., 2001, 3, 1757–1760.
14 (a) H. Sasai, S. Arai, Y. Tahara and M. Shibasaki, J. Org. Chem.,
1995, 60, 6656–6657; (b) H. Gro¨eger, Y. Saida, S. Arai, J. Martens,
H. Sasai and M. Shibasaki, Tetrahedron Lett., 1996, 37, 9291–9292;
(c) H. Gro¨eger, Y. Saida, H. Sasai, K. Yamaguchi, J. Martens and M.
Shibasaki, J. Am. Chem. Soc., 1998, 120, 3089–3103; (d) Y. Saida, H.
Gro¨eger, W. Maison, N. Durot, H. Sasai, M. Shibasaki and J. Martens,
J. Org. Chem., 2000, 65, 4818–4825; (e) S. Kobayashi, H. Kiyohara, Y.
Nakamura and R. Matsubara, J. Am. Chem. Soc., 2004, 126, 6558–
6559.
15 (a) F. Hammerschmidt and M. Hanbauer, J. Org. Chem., 2000, 65,
6121–6131; (b) C. De´jugnat, G. Etemad-Moghadam and I. Rico-Lattes,
Chem. Commun., 2003, 1858–1859.
16 O. I. Kolodiazhnyi, Tetrahedron: Asymmetry, 1998, 9, 1279–1332.
17 G. D. Joly and E. N. Jacobsen, J. Am. Chem. Soc., 2004, 126, 4102–4103.
18 T. Akiyama, H. Morita, J. Itoh and K. Fuchibe, Org. Lett., 2005, 7,
2583–2585.
26 (a) S. Dapprich, I. Komiromi, K. S. Byun, K. Morokuma and M. J.
Frisch, J Mol. Struct. THEOCHEM, 1999, 461–462, 1–21; (b) W. Deng,
T. Vreven, M. J. Frisch and K. B. Wiberg, J Mol. Struct. THEOCHEM,
2006, 775, 93–99.
27 (a) R. Wieczorek and J. J. Dannenberg, J. Am. Chem. Soc., 2004, 126,
12278–12279; (b) R. Wieczorek and R. J. J. Dannenberg, J. Am. Chem.
Soc., 2004, 126, 14198–14205; (c) M. Tsai, Y. Xu and J. J. Dannenberg,
J. Am. Chem. Soc., 2005, 127, 14130–14131; (d) R. Wieczorek and J.
J. Dannenberg, J. Am. Chem. Soc., 2005, 127, 14534–14535; (e) R.
Wieczorek and J. J. Dannenberg, J. Am. Chem. Soc., 2005, 127, 17216–
17223.
28 (a) M. J. S. Dewar, E. G. Zoebisch, E. F. Healy and J. J. P. Stewart,
J. Am. Chem. Soc., 1985, 107, 3902–3909; (b) J.-M. Pons, M. Oblin,
A. Pommier, M. Rajzmann and D. Liotard, J. Am. Chem. Soc., 1997,
119, 3333–3338; (c) J. R. H. anke, G. Haufe, E.-U. Wurthwein and J.
H. Borkent, J. Am. Chem. Soc., 1996, 118, 6031–6035; (d) M. Sola, M.
Duran and J. Mestres, J. Am. Chem. Soc., 1996, 118, 8920–8924.
29 (a) V. Barone, M. Cossi and J. Tomasi, J. Comput. Chem., 1998, 19,
404–417; (b) Y. Takano and K. N. Houk, J. Chem. Theory Comput.,
2005, 1, 70–77.
30 (a) R. F. W. Bader, Atoms in Molecules: A Quantum Theory, Oxford
University Press, Oxford, 1990; (b) F. Biegler-Ko¨nig, J. Scho¨nbohm and
D. J. Bayles, J. Comput. Chem., 2001, 22, 545–559; (c) F. Biegler-Ko¨nig
and J. Scho¨nbohm, J. Comput. Chem., 2002, 23, 1489–1494.
31 (a) P. L. A. Popelier, J. Phys. Chem. A., 1998, 102, 1873–1878; (b) R. F.
W. Bader and H. Essen, J. Chem. Phys., 1984, 80, 1943–1960.
32 L. Sobczyk, S. J. Grabowski and T. M. Krygowski, Chem. Rev., 2005,
105, 3513–3560.
19 To a solution of an imine (27.1 mg, 0.1 mmol) and Brønsted acid 1c
(7.7 mg, 0.01 mmol) in toluene (1 mL) was added diethyl phosphite
(26 mL, 0.2 mmol). The reaction was stirred at room temperature for
24 h, after which the a-amino phosphonate was obtained through
direct purification of the reaction mixture by column chr◦omatography
on silica gel (petroleum ether–EtOAc, 2:1). mp: 147–148 C. 1H NMR
(500 MHz, CDCl3) d = 1.135 (3H, t, J = 7.1 Hz, CH2CH3), 1.326
(3H, t, J = 7.1 Hz, CH2CH3), 2.527 (3H, s, Ar-CH3), 3.860 (1H, m,
Ar-CHNH), 4.244 (4H, m, OCH2), 5.774 (1H, d, J = 11.1 Hz, NH),
6.984–7.368 (m, 7H, Ar-H). 13C NMR (125 MHz, CDCl3): 164.89,
159.85, 150.55, 130.51, 130.04, 129.32, 126.81, 124.61, 122.93, 122.09,
118.33, 115.55, 63.77, 49.67, 18.30, 16.51, 16.26. 19F NMR (470 MHz,
CDCl3):d -117.24. 31P NMR (202 MHz, CDCl3):d 20.59. Elemental
analysis: C, 55.87, H, 5.43, N, 6.86; calculated from C19H22FN2O3PS.
Observed: C, 55.80, H, 5.20, N, 6.76. HPLC: Chiralpak AD-H,
n-hexane-EtOH, 95:5, 1.0 mL min-1, 254 nm: tR(major) = 5.7 min,
tR(minor) = 6.4 min.
20 (a) S. J. Connon, Angew. Chem., Int. Ed., 2006, 45, 3909–3912; (b) M.
S. Taylor and E. N. Jacobsen, Angew. Chem., Int. Ed., 2006, 45, 1520–
1543; (c) C. Bolm, T. Rantanen, I. Schiffers and L. Zani, Angew. Chem.,
Int. Ed., 2005, 44, 1758–1763; (d) P. M. Pihko, Angew. Chem., Int. Ed.,
2004, 43, 2062–2064; (e) P. R. Schreiner, Chem. Soc. Rev., 2003, 32,
289–296.
21 (a) S. Mitsumori, H. Zhang, P. H.-Y. Cheong, K. N. Houk, F. Tanaka
and C. F. Barbas, III, J. Am. Chem. Soc., 2006, 128, 1040–1041; (b) G.
Vayner, K. N. Houk and Y.-K. Sun, J. Am. Chem. Soc., 2004, 126, 199–
203; (c) P. H.-Y. Cheong and K. N. Houk, J. Am. Chem. Soc., 2004,
126, 13912–13913; (d) F. R. Clemente and K. N. Houk, J. Am. Chem.
Soc., 2005, 127, 11294–11302; (e) A. G. Wenzel and E. N. Jacobsen, J.
Am. Chem. Soc., 2002, 124, 12964–12965; (f) N. T. McDougal and S.
E. Schaus, J. Am. Chem. Soc., 2003, 125, 12094–12095; (g) T. Okino, Y.
Hoashi and Y. Takemoto, J. Am. Chem. Soc., 2003, 125, 12672–12673;
33 It is found that the electronic energy of the sum of 2-component
complex and diethyl phosphite is larger than the corresponding
transition state TS4, which implies that the transition state involving
2-component complex and diethyl phosphite must be less stable than
TS4, therefore, the pathway involving 2-component complex can not
compete with those already considered.
34 Y. Saida, H. Gro¨ger, W. Maison, N. Durot, H. Sasai, M. Shibasaki and
J. Martens, J. Org. Chem., 2000, 65, 4818.
1298 | Org. Biomol. Chem., 2009, 7, 1292–1298
This journal is
The Royal Society of Chemistry 2009
©