Mechanism and enantioselectivity of [zinc(diamine)(diol)]-catalyzed asymmetric hydrosilylation of ketones: DFT, NMR and ECD studies

Article abstract of DOI:10.1002/ejoc.201200992

NMR and ECD measurements together with density functional theory computations were used to analyze the mechanism of [Zn(diamine)(diol)]-catalyzed ketone hydrosilylation. Of the three possible pathways, the one that assumes formation of a Zn-hydride specie

Full text of DOI:10.1002/ejoc.201200992

FULL PAPER
DOI: 10.1002/ejoc.201200992
Mechanism and Enantioselectivity of [Zinc(diamine)(diol)]-Catalyzed
Asymmetric Hydrosilylation of Ketones: DFT, NMR and ECD Studies
Jadwiga Gajewy,[a] Jacek Gawronski,[a] and Marcin Kwit*[a]
Keywords: Reaction mechanisms / Density functional calculations / Hydrosilylation / Enantioselectivity / Zinc
NMR and ECD measurements together with density func-
tional theory computations were used to analyze the mecha-
nism of [Zn(diamine)(diol)]-catalyzed ketone hydrosilylation.
and the substrate, took place through Zn-activation of the
carbonyl group. The absolute stereochemistry of the final
product predicted on the basis of our proposal is in agree-
Of the three possible pathways, the one that assumes forma- ment with the available experimental data. It appears that
tion of a Zn-hydride species acting as an active catalyst ap- the most important factor that controls stereochemistry of the
pears energetically most favorable. This conclusion is in con- whole process is preorganization of the reaction substrates
trast to a previously proposed mechanism that assumed the
reaction between the [Zn(diamine)(diol)] catalyst, the silane,
by formation of a NH···O=C hydrogen bond between the
catalyst and the substrate.
lyze asymmetric hydrosilylation of aryl alkyl ketones with
enantiomeric excess of the product of up to 89 %.
Introduction
Referring to the idea of “asymmetric activation”,[30–38]
Ushio and Mikami have expanded the [Zn(diamine)]-cata-
lyzed hydrosilylation method (Scheme 1).[39] Whereas con-
versions of substrates were usually quantitative, the enan-
tiomeric excesses of the products varied, reaching 96 % ee
for aromatic ketones substituted with bulky alkyl groups,
when diamines of rigid structure were used together with
aliphatic or aromatic diols or biphenols.[39,40]
Hydrosilylation reactions are widely used in organic syn-
thesis for the addition of Si–H bonds across carbon–carbon
or carbon–heteroatom multiple bonds.[1] To date, numerous
methods utilizing various transition metals[1–18] in com-
plexes with diphosphanes,[19,20] phosphor-sulfur[21] ligands,
or N-heterocyclic carbenes[11,12] have been developed for
asymmetric hydrosilylation of prochiral ketones and imines.
The drawbacks of these systems are high cost and the need
for careful and elaborate preparation of the catalyst. In the
light of the above facts, [Zn(diamine)]-catalyzed hydro-
silylation, introduced to organic synthesis by Mimoun and
co-workers,[2a] appears highly promising and advantageous
as a method for metal-catalyzed asymmetric hydrosilylation
of ketones and imines.[1] Improvements of the yield and
enantioselectivity of the method could be made either by
judicious choice of a chiral ligand or a change of the reac-
tion conditions.[22,23] The amine ligands are based on either
1,2-diamino-1,2-diphenylethane,
pentane, or 1-phenylethylamine.
trans-1,2-diaminocyclo-
Enantiomerically pure trans-1,2-diaminocyclohexane
(DACH) is particularly valuable because it can be used as
a highly flexible source of diamine structures as well as in
various chiral complexes with N,N- or N,S-chelating li-
gands.[24–27] We[24] and others[28] have recently reported that
chiral hexamine macrocycles (trianglamines)[29] derived
from DACH, in complexes with diethylzinc, efficiently cata-
Scheme 1. Asymmetric activation provides a highly effective cata-
lyst for enantioselective hydrosilylation of ketones.
Recently, Bette et al.[41,42] reported the Zn-promoted di-
rect reduction of various ketones to the corresponding
alcohols with polymethylhydrosiloxane (PMHS) in protic
solvents. Although the chemical yields of the reactions were
quantitative, only moderate enantioselectivities (ee values
up to 55 %) were obtained by using a variety of enantiopure
diamine ligands. The [Zn(diamine)]-catalyzed hydro-
silylation also worked for activated imines;[43] when carried
out in protic solvents this process was reported by us to
[a] Department of Chemistry, A. Mickiewicz University,
Grunwaldzka 6, 60 780 Poznan, Poland
Fax: +48-61-8291505
E-mail: Marcin.Kwit@amu.edu.pl
Homepage: www.amu.edu.pl
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201200992.
Eur. J. Org. Chem. 2013, 307–318
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
307

J. Gajewy, J. Gawronski, M. Kwit
FULL PAPER
Scheme 2. Mechanisms of [Zn(diamine)(diol)]-catalyzed hydrosilylation of prochiral ketones proposed by Ushio and Mikami[39] and Bette
et al.[42]
provide amines with high chemical yields and almost manding methods that can be used for DFT calculations.
100 % ee, using trianglamine as a chiral ligand.[44]
However, these authors pointed out that the standard hy-
Two alternate catalytic cycles were proposed by Bette et brid GGA functionals, such as B3LYP[50] and PBE0,[51] give
al. for ketone hydrosilylation reactions carried out in protic comparable results to those of newer DFT methods and
solvent. In both cycles the role of PMHS is to generate a their use for organic and organometallic reaction mecha-
[Zn(diamine)(H)(OR)] species in the first step of the cata- nisms is still recommended.[48]
lytic cycle, which acts as the reducing agent in the following
In contrast to a number of other zinc-catalyzed reac-
steps (see Scheme 2 and Scheme A1 in the Supporting In- tions[52–56] and Re,[57] Mo,[58] W, [59] Ru,[60] Cu,[61] and Rh[62]
formation).[42] The main difference between the two cata- catalyzed hydrosilylations of C=C or C=O bonds, which
lytic cycles is in the structures of the proposed transition have been studied by means of ab initio and DFT methods,
states, which may be either four-membered or six-mem- the more convenient Zn-promoted hydrosilylation of
bered; however, these assumptions were not confirmed by ketones has, to the best of our knowledge, not been the
calculations.[42]
The catalytic cycle proposed by Ushio and Mikami for ies.[2a,22c,24]
subject of exhaustive experimental and theoretical stud-
the [Zn(diamine)(diol)]-catalyzed hydrosilylation reaction
Because the reaction mechanism is highly dependent on
carried out in nonpolar solvents (Scheme 2) involves initial the catalytic system, the presence of any additive, and the
formation of a tetradentate [Zn(diamine)(diol)] complex nature of the solvent, the originally proposed mechanism
(Step 1), followed by cleavage of one of the two ZnII–O for [Zn(diamine)]-catalyzed hydrosilylation[2a] may not be
bonds after addition of ketone and silane in Steps 2 and 3 the same as those proposed for the reactions carried out in
(see Scheme A2 in the Supporting Information).[39] A sim- the presence of an alcohol.[22,39,42] As part of our interest
ilar mechanism was proposed recently by Marinos et al.[23] in developing experimental methods for enantioselective
for hydrosilylation of acetophenone catalyzed by Zn com- transformations, we sought to determine the factors that
plexes comprising two hard (oxygen) and one soft (sulfur) control stereoselectivity of the hydrosilylation reac-
donor atoms in combination with a diamine ligand.
tion.[24,40,44] To this end, we carried out experimental and
Despite current progress in the development of new ana- theoretical studies on the mechanism of [Zn(diamine)-
lytical methods for tracking the reaction paths and, in this (diol)]-catalyzed hydrosilylation of prochiral ketones, based
way, deducting a probable mechanism, much of the mech- on NMR and electronic circular dichroism (ECD) measure-
anistic understanding of reactions comes from computa- ments as well as on DFT calculations of the possible path-
tional investigations or from studies involving both experi- ways of reaction. Because all the synthetic results have been
mental and complementary computational explorations.[45] published elsewhere,[40] in this contribution we focus on the
Over the last two decades, DFT has become a method of results of computational and spectroscopic measurements.
choice for cost-effective treatment of chemical systems with
high accuracy.[46] Recent advances, especially in the treat-
ment of dispersion effects,[47] now offer more reliable mod-
els of reaction profiles and stereoselectivities.[45] Simón and
Goodman[48] revealed that calculations that make use of
meta-GGA functionals with ultrafine integration grids (as
Results and Discussion
Models and Computational Details
As a model diamine for calculations, we have chosen
implemented in the Gaussian software)[49] are probably N,NЈ-dimethylethylenediamine (dmea, 1). For experimental
both the most reliable and the most computationally de- NMR and ECD measurements, we used (R,R)-trans-N,NЈ-
308
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 307–318

Asymmetric Hydrosilylation of Ketones
dibenzyldiaminocyclohexane (diBnDACH, 2; Scheme 3) as simple as possible, we focus here only on the structures ob-
the chiral ligand. Ethylene glycol (3), (R,R)-2,3-butanediol tained with the use of the PBE0/6-311G(d,p) method. Re-
(4), or 2,2Ј-biphenol (5) were used as activators of the cata- maining computational results and short comments on the
lytic system (Scheme 3). Diamine 1 and activators 3–5 were performance of the computational methods are included in
subsequently employed for acetophenone hydrosilylation the Supporting Information.
reactions with the use of diphenylsilane (Ph2SiH2) or tri-
Relative energies (unit kcalmol–1) reported here refer to
ethoxysilane [(EtO)3SiH] as reducing agents.[40] For the cal- Gibbs free energies (ΔΔG)[48] computed at the PBE0/6-
culations, diols 4 and 5 were replaced by simple ethylene 311G(d,p) level and, unless stated otherwise, are given rela-
glycol (3), acetophenone was the substrate, and trimethylsil- tive to the substrates. For all calculated structures and tran-
ane (Me3SiH) was used as a model reducing agent. Because, sition states, descriptor “a” refers to the formation of O-
in this work, diamine (M)-1 represents the rigid structure silylated methylcarbinol 6 of S absolute configuration,
of DACH derivative, in all calculations we assumed a nega- whereas “b” refers to R absolute configuration of 6. Ad-
tive (H)N–C–C–N(H) torsion angle (–60°, M helicity), that ditionally, any possible reaction pathways having high cal-
directly corresponds to the experimentally determined culated reaction barriers are only briefly mentioned in the
(H)N–C–C–N(H) torsion angle of DACH derivatives.[29b] text (see the Supporting Information for further details).
The initial product of the reaction was O-silylated carbinol 6.
ECD spectra of 7 (Scheme 3) were calculated at the TD-
Zn complex 7, which is readily formed upon mixing ZnEt2, B2LYP/6-311++G(d,p) and TD-B2LYP/def2-TZVPP level
2 and 5 in an inert atmosphere, was chosen as a model com- for all structures optimized at the DFT/6-311G(d,p) level,
pound to assess the reliability of computational methods.
and Boltzmann averaged by taking into account conformers
of 7 ranging from 0 to 2.0 kcalmol–1 in relative energies,
following a generally accepted protocol.[63] All calculations
were carried out with the use of Gaussian 09[49] or Turbom-
ole 6.0 packages.[64]
Mechanism of [Zn(diamine)(diol)]-Catalyzed
Hydrosilylation of Acetophenone
The use of computational methods allows abstraction of
mechanistic concepts from material reality, therefore the
number of possible reaction paths may be large. However,
experimental observations can be taken into account to
support the proposed mechanism. First, negligible stereo-
chemical effect of a diol activator on the performance of
the active catalyst has been found by us for a large number
Scheme 3. Model compounds used in this study.
Computational investigations of possible mechanisms of of diamine and activator combinations.[40] This means that
[Zn(diamine)(diol)]-catalyzed asymmetric hydrosilylation of whereas the activator significantly increases the efficiency
ketones were carried out according to the protocol pro- of the catalytic system, it has little effect on the absolute
posed by Simón and Goodman.[48] Preliminary calculations configuration of the product, which is controlled primarily
were carried out at the PBE0/6-31G(d) level,[51] followed by by the chirality of the diamine.[24,40,44] Furthermore, ketone
higher accuracy calculations with the use of B3LYP,[50] hydrosilylation does not take place with the use of tertiary
PBE0, and M06[63] hybrid functional, all in conjunction mono- or diamines as chiral ligands, hence the presence of
with the enhanced 6-311G(d,p) basis set.[49]
a N–H bond is mandatory for the success of the reaction
All calculations using the M06 functional were carried carried out in nonpolar solvent.[24,40,44] We have not ob-
out with the ultrafine integration grid as implemented in served any N-silylated amines as products, even when the
Gaussian09[49] and abbreviated here as M06-UF.[48] Ad- reactions were conducted with stoichiometric amounts of
ditionally, for all calculated structures at any of the DFT all reagents (1 mmol scale) and the second step of the reac-
levels, single-point calculations were performed with the use tion (basic hydrolysis) was omitted. When used alone, di-
of the M06-UF/6-311++G(2d,2p) method to obtain more amine, diol, diethylzinc, and [Zn(diol)] species were not cat-
accurate energies (Esp). Calculations performed with the use alytically active.[40] The order of addition of the Zn-catalyst
of the M06-UF/6-311G(d,p) method were treated as a refer- precursors (zinc source, diamine and diol) to the reaction
ence and allowed the reliability of the calculations carried mixture was not important and, in all cases, the same active
out with the use of PBE0 and B3LYP hybrid functionals to species of the general formula [Zn(diamine)(diol)] were ap-
be assessed. Except for numerical values, the results ob- parently formed.[39,40,42]
tained with the use of PBE0 and B3LYP functionals were
On the basis of precedents available in the recent litera-
not significantly different to those obtained from calcula- ture[39,42] and the results of our calculations, we assume four
tions employing the M06-UF functional [all in conjunction steps are involved in the catalytic cycle: (1) formation of
with the 6-311G(d,p) basis set]. To keep the discussion as an active catalytic species from Zn-catalyst precursors; (2)
Eur. J. Org. Chem. 2013, 307–318
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
309

J. Gajewy, J. Gawronski, M. Kwit
FULL PAPER
preorganization of the substrate and the active catalyst;
The aim of this work is to provide answers to important
(3) reduction (hydrosilylation) of the substrate; (4) dissoci- questions: (i) what is the structure of the active catalyst,
ation of the product with recovery of the active catalyst.
since contradictory experimental results were so far ob-
tained[23,39,42] (Step 1); (ii) what is the role of silane in the
catalytic cycle (Step 1 or Step 3); (iii) which factors control
the stereostructure of the final product (Step 2 or Step 3);
(iv) how is the catalytically active species recovered (Step 4),
and (v) which step is rate determining.
A catalytic cycle incorporating the experimental observa-
tions and assumptions made earlier (Steps 1–3) is shown in
Scheme 4.
Step 1 – Formation of Catalytically Active Species
In the first step of the catalytic cycle, formation of the
active species 8 [Zn(diamine)(diol)] takes place (Scheme 4).
The reaction is exothermic and irreversible (Er
=
–30.89 kcalmol–1). The stereostructure of 8 may be ana-
lyzed in terms of helicity of the diamine and the diol moie-
ties and the configuration of the substituents at the nitrogen
atoms. For the assumed M helicity of diamine, there are
four C2-symmetrical diastereomers of 8 having either M or
P helicity of the dialkoxide moiety (and either RN,RN or
SN,SN absolute configuration of the nitrogen atoms).
Homohelical (M,M) structures[65] were found to be unstable
by calculations at any level of the DFT methods used. Of
the two (M,P)-heterohelical structures, the one charac-
terized by SN absolute configurations at the nitrogen atoms
(8; shown in Figure 1) is energetically preferred (ΔΔG =
1.61 kcalmol–1, which corresponds to 95 % abundance of 8
in the reaction mixture) over that characterized by RN abso-
lute configurations (denoted as 8Ј, 5 % of abundance, see
the Supporting Information, Figure B1).
Each of the five-membered rings adopts the lowest-en-
ergy half-chair conformation (SN,SN,Mdiamine,Pdiol)-8 with
N-methyl substituents occupying equatorial positions. Cal-
culated lengths of the Zn–N and Zn–O bonds in 8 are 2.05,
2.11 and 1.85, 1.91 Å, respectively. A longer Zn–O bond
(1.91 Å) is found for the bond perpendicular to the average
plane formed by the second oxygen and the two nitrogen
atoms. This suggests that [Zn(diamine)(diol)] species 8
forms a complex with the structure of a trigonal pyramid,
rather than of a regular tetrahedron. The lengths of the Zn–
N bonds are in reasonable agreement with those experimen-
tally determined by X-ray diffraction analysis, which range
from 2.08 to 2.23 Å (shorter bonds were found in complexes
in which the zinc atom forms an additional Zn–O
bond).[2,41,66]
The transformation of 8 into [Zn-hydride] species 9
(shown in Figure 1) required activation energy Ea(8–9)
=
6.03 kcalmol–1, relative to precursor complex 8 (Figure 1
and Figure B2 in the Supporting Information). [Zn-
Hydride] species 9 is characterized by the presence of a
hydrogen bond (length 1.90 Å) between one of the NH pro-
tons and the oxygen atom.
According to the calculations, formation of 9 (from 8
and HSiMe3) is an exothermic process, with a reaction en-
ergy of –21.70 kcalmol–1. Transition state TS8–9 resembles
a trigonal bipyramidal pentacoordinate Zn complex in
Scheme 4. Formation of the active complexes 8 and 9 (Step 1) and
possible further steps (2 and 3) of acetophenone coordination and
reduction.
310
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 307–318

Asymmetric Hydrosilylation of Ketones
Figure 2. Sections of the 1H NMR spectra measured after addition
to the [D8]toluene solution of (R,R)-2 (a) equimolar amounts of
[D8]toluene diethylzinc solution and (R,R)-4 (b), (EtO)3SiH (c) and
acetophenone (d).
Figure 1. Structures of catalytically active Zn-species 8 (a) [Zn-
hydride] complex 9 (b) and transition state TS8–9 structure (c) ob-
tained at the M06-UF/6-311++G(2d,2p)//PBE0/6-311G(d,p) level
of theory; some hydrogen atoms were omitted for clarity; distances
are in Å.
After addition of the silane, a sharp signal at δ = 4.50
ppm appeared, which did not vanish within two hours. It is
known that in deuterated aromatic solvents, a 1H NMR
signal due to the Zn-H appears in the region δ = 4.4–
4.6 ppm.[42,68–71] The intensity of the Zn-H signal was re-
duced after addition of acetophenone and increased again
after addition of another equivalent of the silane.
When Ph2SiH2 was used instead of (EtO)3SiH, both Zn-
H and Si-H signals were observed at δ = 4.50 and δ = 5.02
ppm, respectively, as sharp signals. Additionally, the signals
of the NHCH2Ar benzylic protons (doublet of doublets at
δ = 3.62 ppm) were multiplied, probably due to the presence
which the incoming hydride and one of the oxygen atoms
adopt apical positions, and both nitrogen atoms and the
second oxygen atom remain equatorial.
NMR and ECD Measurements
Despite much effort, we were not able to isolate any spe- of rotamers of these groups (see Figure C3 in the Support-
cies of the formula [Zn(diamine)(diol)] as crystals that were ing Information). It is significant that the [Zn-hydride] spe-
suitable for X-ray analysis, however, their formation could cies was stable even when deuterated methanol was added
be demonstrated through the use of both NMR and ECD to the NMR tube.
spectroscopy.[63a]
When Ph2SiD2 was used instead of Ph2SiH2, the signals
1
For H NMR measurements, we used (R,R)-2 as a di- at δ = 4.50 (Zn-H) and 5.02 ppm (Si-H) were not observed.
amine, (R,R)-4 as an activator, Ph2SiH2 or (EtO)3SiH as Within two hours, the 1H NMR spectra (measured at
reducing agent, and acetophenone as substrate. All mea- 15 min intervals) did not show any change and we did not
surements were carried out in anhydrous and degassed [D8]- observe the appearance of the Zn-H signal at δ = 4.50 ppm.
toluene solutions and were conducted by stepwise addition This suggests that there is no exchange between Zn-D and
of equimolar amounts of the reagents to the NMR tube, N-H.
under rigorously inert atmosphere. The measured 1H NMR
spectra are shown in Figure 2 and Figures C1, C2, and C3 further confirmed by the ECD measurements. To the best
in the Supporting Information. of our knowledge, ECD spectra of chiral [Zn(diamine)-
The formation of the [Zn(diamine)(diol)] complex was
After addition of one equivalent of diethylzinc solution (diol)] complexes formed in situ have not been reported to
to a solution of 2, a significant change in the shape of sig- date. In addition, the results of our experiments enabled us
nals was observed in the spectral region 0 to 2 ppm, due to to test the performance of DFT methods, which were fur-
the presence of the ethyl groups. After mixing 2 with dieth- ther used to understand the reaction mechanism. The calcu-
ylzinc and 4, evolution of ethane took place and a sharp lation methods were evaluated indirectly by a comparison
signal at δ = 0.81 ppm was observed, which is characteristic of the calculated ECD spectra with the experimental data
for ethane.[67]
(see the Supporting Information).[72]
Eur. J. Org. Chem. 2013, 307–318
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
311

J. Gajewy, J. Gawronski, M. Kwit
FULL PAPER
ECD spectra were measured in anhydrous and degassed ment of the diamine (M) and biphenyl moiety (P, which
cyclohexane (Figure 3, a), using 2, ZnEt2, and 5 as the sub- corresponds to S configuration), and C2-symmetry of mole-
strates to provide complex 7 (Scheme 3) according to cule 7 is energetically preferred (Figure 3, d).[65] The prefer-
Scheme 1, at a substrate concentration of 10–4 m.
ence of heterohelical over homohelical conformations of 8
is in full agreement with the experimental and theoretical
data obtained for 7.
A strong confirmation of the preferred structure of com-
plex 7 (denoted as 7a) has been obtained by comparison of
the experimental and Boltzmann averaged calculated ECD
spectra (Figure 3, a, and Supporting Information, Figures
D1–D5). Because the experimental and calculated ECD
spectra are in good agreement, the preferred structure of 7a
is considered highly probable.
The best agreement with the experimental data and the
most consistent results were obtained when the averaged
ECD spectrum was calculated at the B2LYP/def2-TZVPP
level for structures optimized by the M06-UF/6-
311++G(2d,2p)//PBE0/6-311G(d,p) method (see the Sup-
porting Information, Figures D3 and D5).
Since within six hours we did not observe any changes
in the ECD spectra, we can conclude that complex 7 is
stable in an inert atmosphere.
Step 2 – Coordination of Substrate to the Catalyst
The second step, preorganization of substrate and the
catalyst, may determine the stereochemistry of the product.
Coordination of acetophenone to the active catalysts may
occur in two ways: either by Zn activation of the C=O
group, accompanied by rearrangement of the ligands
around the metal center to form complexes 11 or 13 (Paths
Figure 3. Induced circular dichroism observed during titration of
cyclohexane solution of (R,R)-2 (dash-dotted line) by diethylzinc
solution (dashed line) followed by solution of 5 (solid line) and
Boltzmann averaged ECD spectrum calculated at the TD-B2LYP/ A2 or C, Scheme 4), or by formation of the NH···O=C hydro-
def2-TZVPP level for 7 (dashed line) (a). Possible structures of
gen bond (complexes 10 and 12, Paths A1 or B, Scheme 4).
complexes 7: square planar (b) tetrahedral (c), and the lowest-en-
Paths D make links between Paths A1 and B or A2 and C
ergy conformer of 7 calculated at the M06-UF/6-311++G(2d,2p)//
and provide [Zn-hydride] complexes 10 and 11 from 12 and
PBE0/6-311G(d,p) level (d). Some hydrogen atoms were omitted
for clarity; distances are in Å.
13, respectively. The main difference between Paths A and
D is the sequence of events. In the case of Paths D, aceto-
Due to high conformational flexibility of both 2 and its phenone is coordinated to complex 8 either through
complex with ZnEt2, we did not observe any significant NH···O=C hydrogen bond formation to provide complex 12
changes in the measured ECD spectra after addition of or by Zn activation of the C=O group to provide complex
ZnEt2 solution to the solution of 2. The situation changed 13. Both 12 and 13 further react with the silane to provide
rapidly when one equivalent of 5 was introduced into the complexes 10 and 11, respectively. Note that the structures
cell. We observed the appearance of strong CD signals in containing acetophenone can exist as pairs of diastereomers
the region between 320 and 185 nm (Figure 3, a), whereas (denoted here as “a” or “b”), differing in relative orienta-
no ECD spectrum was observed when 2 and 5 were mixed tion of the ketone moiety. In the case of complexes 10 and
separately.
11, facial orientation of Zn-hydride complex and ketone di-
Because 2,2Ј-biphenol may be treated as a chiral, but dy- rectly determines the stereochemistry of the product (hy-
namically racemizing molecule,[73] the CD bands originate dride attacks si or re face of the ketone). On the other hand,
from induced deracemization of the biphenyl moiety.[74] in the case of 12 and 13, the stereochemistry of the products
This suggests a tetrahedral rather than a square planar ar- depend not only on the relative orientation of the ketone
rangement of the N- and O-ligands around the zinc atom in but also on the direction of hydride attack. One can imagine
7 (see Figure 3, b and c). DFT calculations of the possible two possibilities – the first, when activated by an oxygen
structures of 7 confirmed this hypothesis. To achieve a atom silane releases hydride, and the second, when the re-
planar structure of 7 the cyclohexane ring had to be de- lease of hydride takes place without silane activation. The
formed, whereas in tetrahedral complexes 7 the chair-like latter possibility provides the product of opposite configu-
conformation of the cyclohexane skeleton was retained, ration to that obtained with silane activation.
however, the formation of two diastereomers, differing in
Calculations of atomic charges through the use of the
the helicity of the biphenyl moiety, was possible. According NBO method[75,76] clearly show that the zinc ion in [Zn(di-
to the results of our computations, heterohelical arrange- amine)(diol)] complex 8 carries a more positive charge
312
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 307–318

Asymmetric Hydrosilylation of Ketones
(1.60) than the amine hydrogen atom (0.41). In the case of
9, the zinc atom is again more positive (1.40) than any of
the (N)H atoms (0.37). This suggests that direct binding of
the acetophenone carbonyl group (oxygen atom carries a
negative charge –0.56) to the zinc ion (Paths A2 and C,
Scheme 4) is energetically preferred over the formation of
a NH···O=C hydrogen bond (Paths A1 and B, Scheme 4).
However, postulated formation of complexes 13 (from 8)
and 11 (from 9) by zinc···O=C coordination requires re-
arrangement of the ligands around the metal center to give
the complexes with the structure of a trigonal bipyramid,
in which the zinc coordination number reaches five.[77]
Binding acetophenone to the active complexes 8 or 9,
through the formation of NH···O=C hydrogen bond, pre-
serves the zinc coordination number four and the spatial
arrangement of ligands around the central atom in 10 and
12 is the same as in the mother complexes. The preferential
substrate coordination mode may be readily evaluated by
direct comparison of their calculated relative energies (see
Table 1, Figure 4 and Figure E1 in the Supporting Infor-
mation).
Figure 4. Structures of low-energy complexes 10 and 11 calculated
at the M06-UF/6-311++G(2d,2p)//PBE0/6-311G(d,p) level (some
hydrogen atoms were omitted for clarity; distances and bond
lengths are in Å).
Table 1. Relative energies (ΔΔG, in kcalmol–1) calculated at the
M06-UF/6-311++G(2d,2p)//PBE0/6-311G(d,p) level for individual
structures of complexes 9–13.
DFT calculation levels employed here were we able to lo-
calize transition states for the formation of 11 directly from
13 (Path D2) even when we explored the possibility of a
stepwise mechanism (i.e., silane could approach 13, link to
it, and then convert into 11), thus this path is not consid-
ered. Isomeric complexes 12 and 13 do not differ signifi-
cantly in energy, however, in the light of the experimental
results (1H NMR spectra) and the theoretical data collected
in Table 1, their presence in the reaction mixture seems to
be improbable. Thus, complexes 12 and 13 and Paths B and
C may be excluded from further considerations (see the
Supporting Information for details).
We may conclude that formation of 10 from 9 (Path A1)
appears to be favored over the formation of 10 from 12
(Path D1). Complexes 11 and Path A2 do not play signifi-
cant roles in the reaction mechanism; calculated relative en-
ergies of complexes 11 were over 2 kcalmol–1 higher than
those calculated for complexes 10, which corresponds to
negligible participation of 11 in 10–11 isomer equilibrium.
We conclude that activation of the carbonyl group by direct
interaction with the zinc ion is evidently not favored in the
present case.
Complex
ΔΔG[a]
ΔΔGrel
9
–21.70
–18.62
–18.26
–16.31
–16.22
–4.56
–4.68
–4.00
–4.93
–
10a
10b
11a
11b
12a
12b
13a
13b
0.00[b]
0.36[b]
2.30[b]
2.39[b]
0.36[c]
0.25[c]
0.93[c]
0.00[c]
[a] Relative to the sum of energies of 8, acetophenone, and HSiMe3.
[b] Relative to the lowest-energy 10a. [c] Relative to the lowest-
energy 13b.
A comparison of the Gibbs free energy values, calculated
relative to the sum of starting reactants (i.e., complex 8,
acetophenone, and trimethylsilane), within all possible
structures 10–13 shows clearly a strong energetic preference
of [Zn-hydride] complexes 10 over all other possible struc-
tures. Due to a large distance between acetophenone and
the rest of the molecule, structures 10b and 10a show only
a small energy difference (ΔΔG = 0.36 kcalmol–1). The cal-
culated barrier of conversion of 10a into 10b is small and
amounts to 2.64 kcalmol–1.
Step 3 – Ketone Reduction
The formation of 10a and 10b may take place directly
The third step involves reduction of the acetophenone
from 12a and 12b in the reaction with silane (Path D1, C=O bond, which may proceed through direct transfer of
Scheme 4). Activation energies calculated for the transition the hydride from Zn to the activated C=O group (Paths A1
states TS12a–10a and TS12b–10b are Ea(12a–10a)
=
and A2, Scheme 4).[42] According to our results, hydrogen
10.98 kcalmol–1 and Ea(12b–10b) = 26.33 kcalmol–1, and they transfer from zinc to the C=O group is best described by
are higher than the calculated activation energy for the a six-membered transition state, stabilized by NH···O(=C)
transformation of complex 8 into the [Zn-hydride] species 9 hydrogen-bonding (see Figure 5 and the Supporting Infor-
(6.03 kcalmol–1) (see Figure E2 in the Supporting Infor- mation, Figure E3) to give intermediate complexes 14.
mation). The differences in activation energies Ea(12a–10a) These complexes remain tetracoordinate, and other struc-
and Ea(12b–10b) are due to steric crowding in the transition tural features that characterize 14a and 14b include one sig-
state of the reaction 12b Ǟ 10b. However, at none of the nificantly shorter (ca. 0.2 Å) N–Zn bond and planarity of
Eur. J. Org. Chem. 2013, 307–318
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
313

J. Gajewy, J. Gawronski, M. Kwit
FULL PAPER
Figure 5. Structures and Gibbs free energy changes (ΔΔG, kcalmol–1) during hydride transfer from zinc to the carbonyl group and
isomerization of intermediate product 14. Distances and bond lengths are in Å, some hydrogen atoms were omitted for clarity.
the C–N(CH3)–Zn fragment, due to a change of the type activation energies for the formation of 15 from 11 are ca.
of N–Zn bond from coordinating to covalent. Activation 7 kcalmol–1 higher than those calculated for the reaction 10
energies calculated for reactions 10Ǟ14 were Ea(10a–14a)
=
Ǟ 14 (see the Supporting Information, Figure E3). Ad-
11.30 kcalmol–1 and Ea(10b–14b) = 11.10 kcalmol–1, respec- ditionally, Path A2 shows a preference for the formation of
tively, and transformation of 10 into 14 is an exothermic the (R)-product (sub-path A2b, see the Supporting Infor-
reaction (Er is ca. –7 kcalmol–1).
mation, Figure E3), in disagreement with the experimental
Primary products 14a and 14b may subsequently un- data. Note that in the case of Paths B and C (Scheme 4),
dergo 1-phenylethoxy ligand migration to form more stable which have been excluded from consideration, regardless of
complexes 15a and 15b, through a four-membered transi- the silane modus operandi (i.e., with or without activation),
tion state. Calculated activation energies are small, the reduction requires higher activation energies (see the
Ea(14a–15a)
=
3.27 kcalmol–1
and
Ea(14b–15b)
=
Supporting Information, Figure E4) than direct intramolec-
1.90 kcalmol–1. The rearrangement of 14 to 15 is an exo- ular hydrogen transfer from Zn to the C=O group. The
thermic and irreversible process and the estimated Er is ap- preference of intramolecular over intermolecular reaction
proximately –10 kcalmol–1.
remains in full agreement with experimental data.
In the case of Path A1, calculated energy profiles show a
preference for the formation of the (S)-product (sub-path
A1a), in agreement with the experimental data.
According to one proposal,[42] intramolecular hydrogen
Step 4 – Dissociation of the O-Silylated Product 6 and
Recovery of the Catalyst
There are three ways to convert the silylated product 15
transfer could take place via a four-membered transition into 6 (Scheme 5). First is an intermolecular trans-silylation
state in which the carbonyl group is activated directly by with recovery of the initial [Zn(diamine)(diol)] complex 8
the Lewis acid (Path A2 in Scheme 4). However, the pop- and silylated carbinol 6, through the formation of an inter-
ulation of starting complexes 11 in the 10–11 isomer equi- mediate complex 16 (Path E, Scheme 5). An alternative
librium is negligible and, more importantly, the calculated path assumes a displacement of one of the alkoxy ligands
314
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 307–318

Asymmetric Hydrosilylation of Ketones
present in 15 by the hydride, analogous to the formation of
[Zn-hydride] species 9 from 8 (Paths F, Scheme 5). Due to
the presence of two nonequal alkoxy ligands in complex 15,
formation of two different complexes 17 and 18 of the gene-
ral formula [Zn(diamine)(H)(OR)]Me3SiORЈ may be envis-
aged. An alternative to the above-mentioned possibilities is
represented in Path G (Scheme 5), starting from complexes
14, which may undergo direct reaction with a silane to pro-
duce 9 and 6, through an intermediate complex 17.
Figure 6. Calculated energy profiles (ΔΔG, in kcalmol–1) for Paths
E and F (a) and calculated structures of transition states TS15a–16a
and TS15b–16b (b). Energies (in kcalmol–1) are calculated relative to
the energies of starting complexes 15 (15a and 15b), in parentheses
are shown energies calculated for paths “b”. Some hydrogen atoms
were omitted for clarity (distances and bond lengths are in Å).
42.59 kcalmol–1 (Path F1a), Ea(15b–17b) = 42.72 (Path F1b),
Ea(15a–18a) = 45.73 kcalmol–1 (Path F2a), and Ea(15b–18b)
=
5.92 kcalmol–1 (Path F2b).
Alternatively, O-silylated carbinol 6 may by displaced
from complex 14 by the silane, as was assumed for Path
G. However, intermolecular reaction between silane and 14
requires higher activation energy than intermolecular shift
of the alkoxide ligand leading to complexes 15 (see the Sup-
porting Information, Figure E5). Calculated activation en-
ergies for intermolecular reactions are 28.64 and
31.05 kcalmol–1, which are approximately ten times higher
than activation energies required for intramolecular shift of
1-phenylethoxy ligand. Thus, due to the higher activation
barriers, all of the intermolecular processes may be consid-
ered improbable.
Scheme 5. Possible pathways for the release of product 6 and
recovery of the active catalyst.
Conclusions
The energy profile calculated for paths E–G as well as
calculated transition states TS15a–16a and TS15b–16b are
The energy profile normalized to the sum of energies of
shown in Figure 6 (remaining transition states are shown in all substrates is shown in Figure 7 (see also Figures A1 and
Figure E5 included in the Supporting Information). A2 in the Supporting Information). The diagram clearly
Path E assumes intramolecular transfer of the trimethyl- shows that Path A1, which assumes formation of the [Zn-
silyl (TMS) group from the dialkoxide ligand to 1-phenyl- hydride] species 10 and then intramolecular hydride transfer
ethoxy anion connected to the zinc ion, as postulated pre- from Zn to the C=O group of the substrate, is energetically
viously by Ushio and Mikami.[39] The energy barrier values more favored than any intermolecular reduction. Alterna-
are Ea(15a–16a) = 20.74 and Ea(15b–16b) = 26.12 kcalmol–1, tively, preorganized complexes 10 can be formed indirectly
respectively, for Path Ea and Path Eb and both reactions from complexes 12. The absolute energies of TS8–9 and
are endoergic (Figure 6). The size of these energy barriers TS12–10 seem to be comparable, however, when considering
depend on the configuration of the alcohol product; a the intrinsic activation energy (difference between the en-
higher energy barrier is calculated when the TMS group ergy of the transition state and the preceding minimum),
is transferred to the (R)-1-phenylethoxy ligand. In the less the activation energy of the reaction 8Ǟ9 (Path A,
sterically crowded four-membered transition state TS15a–16a Scheme 4) is much smaller than the activation energy of the
(Figure 6, b) there is no repulsion between 1-phenylethoxide reaction leading from 12 to 10 (Path D1, Scheme 4).
methyl group and one of the methyl groups of the TMS
The two possible transition states for Paths A1 and A2
group.
(TS10–14 and TS11–15, respectively) differ in structure and
Calculated barriers for intermolecular reactions between the manner of binding acetophenone to the catalyst, the
silane and 15 (Paths F) are higher than that for intramolec- one characterized by a (N)H···O=C hydrogen bond between
ular trans-silylation (Path E), i.e., Ea(15a–17a)
=
the catalyst and the substrate and a six-membered structure
Eur. J. Org. Chem. 2013, 307–318
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
315

J. Gajewy, J. Gawronski, M. Kwit
FULL PAPER
Figure 7. Calculated energy profile (ΔΔG, in kcalmol–1) of all possible reaction pathways (shown in Schemes 3 and 4) leading to the
product of S absolute configuration (in parentheses are the values of energies calculated for the intermediates and transition states leading
to the product of R absolute configuration), and structures of stable intermediates and transition states. The energy values are given
relative to the sum of total energies of 8, free acetophenone, and trimethylsilane molecule (for a full version see the Supporting Infor-
mation).
is energetically preferred. This confirms the preferred for- source, (R,R)-diMeDACH as a chiral diamine, ethanediol
mation of the [Zn-hydride] species, acting as an effective as an activator, and triethoxysilane as a reducing agent[40]
reducing agent, in agreement with experimental measure- was 40%, which is in reasonable agreement with the calcu-
ments.[42] One can say that the silane is a spectator rather lated value.
than an actor in the third step (reduction) of the catalytic
cycle.
In this paper we presented a comprehensive analysis and
rationalization of the observed enantioselectivity of zinc-
For the last step of the catalytic cycle, the preference for catalyzed hydrosilylation, with ligands derived from (R,R)-
intramolecular over intermolecular reaction has been deter- DACH, which provide preferentially the carbinol of S con-
mined. On the basis of calculated energy profiles, direct for- figuration. Our analysis is based on the DFT calculated
mation of any hydride species from 15 or 14 is improbable, low-energy structures of the [Zn(ligand)(OR)(H)]-aceto-
mostly for steric reasons. Of all the individual steps of Path phenone precursors, as well as on the structures of the tran-
A(A1), the intramolecular shift of the TMS group has the sition states leading to the formation of the chiral products.
highest activation energy and controls the kinetics of the The outcome of our analysis is the identification of the for-
whole cycle.
mation of the NH···O(=C) hydrogen bond, which is crucial
Preorganization of the substrates in complexes 10 is of for preorganization of the reactants and determines the spa-
primary importance for control of the enantioselectivity of tial arrangement of both the catalyst and the ketone. In
the whole process. The stereochemical outcome of Path A1 this way the enantioselectivity of the whole hydrosilylation
is in excellent agreement with the experimental data; the use process could be rationally accounted for. Other pathways
of DACH derivatives as catalysts provides 1-phenylethanol of acetophenone reduction either do not predict the prod-
of S absolute configuration.
uct of S absolute configuration or are characterized by
Based on a previously proposed methodology,[78] we were higher activation barriers than intramolecular hydrogen
able to predict the enantiomeric excess of the final product transfer from zinc to the C=O group going through the six-
from the energy difference between the transition states. In membered transition state. Additionally, we were able to ra-
the case of TS10a–14a and TS10b–14b, the predicted enantio- tionalize the manner of active catalyst recovery; of the three
meric excess of (S)-1-phenylethanol is about 17%. Enantio- possibilities considered, the one that assumes intramolecu-
meric excess experimentally found for hydrosilylation of 4- lar transfer of the silyl group to 1-phenylethyloxy ligand is
methylacetophenone through the use of ZnEt2 as a zinc energetically the most favored.
316
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 307–318

Asymmetric Hydrosilylation of Ketones
[21] a) D. A. Evans, F. E. Michael, J. S. Tedrow, K. R. Campos, J.
Am. Chem. Soc. 2003, 125, 3534; b) N. Khiar, M. P. Leal, R.
Navas, J. F. Moya, M. V. G. Pérez, I. Fernández, Org. Biomol.
Chem. 2012, 10, 355.
[22] a) S. Liu, J. Peng, H. Yang, Y. Bai, G. Lai, Tetrahedron 2012,
68, 1371; b) K. Junge, K. Möller, B. Wendt, S. Das, D. Gördes,
K. Thurow, M. Beller, Chem. Asian J. 2012, 7, 314; c) M.
Kahnes, H. Görls, L. González, M. Westerhausen, Organome-
tallics 2010, 29, 3098; d) S. Enthaler, K. Schröder, S. Inoue, B.
Eckhardt, K. Junge, M. Beller, M. Drieß, Eur. J. Org. Chem.
2010, 4893; e) S. Enthaler, B. Eckhardt, S. Inoue, E. Irran, M.
Driess, Chem. Asian J. 2010, 5, 2027.
Supporting Information (see footnote on the first page of this arti-
cle): Schemes A1–A2, Figures A1–E5, spectra obtained during
NMR experiments, calculated UV and ECD spectra for complex
7, relative energies, selected structural parameters and Cartesian
coordinates of all species and comments on the performance of the
DFT methods.
Acknowledgments
This work was supported by the National Science Centre (Poland),
grant number 2011/03/B/ST5/01011. All calculations were per-
formed at the Poznan´ Supercomputing and Networking Center.
[23] N. A. Marinos, S. Enthaler, M. Driess, ChemCatChem 2010, 2,
846.
[24] J. Gajewy, M. Kwit, J. Gawronski, Adv. Synth. Catal. 2009, 351,
1055.
[1] B. Marciniec, Hydrosilylation, Springer, Berlin, 2009.
[2] a) H. Mimoun, J. Y. de Saint Laumer, L. Giannini, R. Scopel-
liti, C. Floriani, J. Am. Chem. Soc. 1999, 121, 6158; b) H. Mim-
oun, J. Org. Chem. 1999, 64, 2582; c) X.-F. Wu, Chem. Asian
J. 2012, 7, 2502.
[3] H. Nishiyama, K. Itoh, in: Catalytic Asymmetric Synthesis
(Ed.: I. Ojima), Wiley-VCH, Weinheim, Germany, 2000, chap-
ter 2.
[4] H. Nishiyama, in: Comprehensive Asymmetric Catalysis (Eds.:
E. N. Jacobsen, A. Pfaltz, H. Yamamoto), Springer, Berlin,
1999, vol. 1, chapter 6.3.
[5] H. Nishiyama, in: Transition Metals for Organic Synthesis
Building Blocks and Fine Chemicals (Eds.: M. Beller, C. Bolm),
Wiley-VCH, Weinheim, Germany, 2004, vol. 2, chapter 1.4.2.
[6] J.-F. Carpentier, V. Bette, Curr. Org. Chem. 2002, 6, 913.
[7] O. Riant, N. Mostefaï, J. Courmarcel, Synthesis 2004, 2943.
[8] T. Hayashi, Acc. Chem. Res. 2000, 33, 354.
[9] A. K. Roy, Adv. Organomet. Chem. 2008, 55, 1.
[10] S. Diez-González, S. P. Nolan, Acc. Chem. Res. 2008, 41, 349.
[11] a) L. H. Gade, V. César, S. Bellemin-Laponnaz, Angew. Chem.
2004, 116, 1036; Angew. Chem. Int. Ed. 2004, 43, 1014; b) V.
César, S. Bellemin-Laponnaz, H. Wadepohl, L. H. Gade,
Chem. Eur. J. 2005, 11, 2862; c) H. Songa, Y. Liua, D. Fana,
G. Zib, J. Organomet. Chem. 2011, 696, 3714.
[12] C. Song, C. Ma, Y. Ma, W. Feng, S. Ma, Q. Chai, M. B.
Andrus, Tetrahedron Lett. 2005, 46, 3241.
[13] B. H. Lipshutz, K. Noson, W. Chrisman, J. Am. Chem. Soc.
2001, 123, 12917.
[14] a) S. Sirol, J. Courmarcel, N. Mostefaï, O. Riant, Org. Lett.
2001, 3, 4111; b) N. Mostefaï, S. Sirol, J. Courmarcel, O. Riant,
Synthesis 2007, 1265; c) J. Courmacel, N. Mostefaï, S. Sirol, S.
Chopin, O. Riant, Isr. J. Chem. 2002, 41, 231; d) D.-W. Lee, J.
Yun, Tetrahedron Lett. 2004, 45, 5415; e) J. Wu, J.-X. Ji,
A. S. C. Chan, Proc. Natl. Acad. Sci. USA 2005, 102, 3570; f)
M. L. Kantam, S. Laha, J. Yadav, P. R. Likhar, B. Sreedhar, S.
Jha, S. Bhargava, M. Udayakiran, B. Jagadeesh, Org. Lett.
2008, 10, 2979.
[15] M. L. Kantam, S. Laha, J. Yadav, P. R. Likhar, B. Sreedhar,
B. M. Choudary, Adv. Synth. Catal. 2007, 349, 1797.
[16] a) B. H. Lipshutz, A. Lower, K. Noson, Org. Lett. 2002, 4,
4045; b) B. H. Lipshutz, K. Noson, W. Chrisman, A. Lower, J.
Am. Chem. Soc. 2003, 125, 8779.
[17] N. S. Shaikh, S. Enthaler, K. Junge, M. Beller, Angew. Chem.
2008, 120, 2531; Angew. Chem. Int. Ed. 2008, 47, 2497.
[18] J. Yun, S. L. Buchwald, J. Am. Chem. Soc. 1999, 121, 5640.
[19] a) H. Brunner, R. Becker, G. Riepl, Organometallics 1984, 3,
1354; b) F. Yu, X.-C. Zhang, F.-F. Wu, J.-N. Zhou, W. Fang,
J. Wu, A. S. C. Chan, Org. Biomol. Chem. 2011, 9, 5652; c) X.-
C. Zhang, F.-F. Wu, S. Li, J.-N. Zhou, J. Wu, N. Li, W. Fang,
K. H. Lam, A. S. C. Chan, Adv. Synth. Catal. 2011, 353, 1457;
d) K. Junge, B. Wendt, D. Addis, S. Zhou, S. Das, M. Beller,
Chem. Eur. J. 2010, 16, 68; e) J.-T. Issenhuth, S. Dagorne, S.
Bellemin-Laponnaz, C. R. Chim. 2010, 13, 353.
[25] S. Gérard, Y. Pressel, O. Riant, Tetrahedron: Asymmetry 2005,
16, 1889.
[26] M. Bandini, M. Melucci, F. Piccinelli, R. Sinisi, S. Tommasi,
A. Umani-Ronchi, Chem. Commun. 2007, 4519.
[27] E. Santacruz, G. Huelgas, S. K. Angulo, V. M. Mastranzo, S.
Hernández-Ortega, J. A. Avina, E. Juaristi, C. Anaya de Par-
rodi, P. J. Walsh, Tetrahedron: Asymmetry 2009, 20, 2788.
[28] D. Savoia, A. Gualandi, H. Stoeckli-Evans, Org. Biomol. Chem.
2010, 8, 3992.
[29] a) J. Gawronski, H. Kołbon, M. Kwit, A. Katrusiak, J. Org.
Chem. 2000, 65, 5768; b) J. Gawronski, K. Gawronska, J. Gra-
jewski, M. Kwit, A. Plutecka, U. Rychlewska, Chem. Eur. J.
2006, 12, 1807.
[30] K. Mikami, M. Terada, T. Korenaga, Y. Matsumoto, M. Ueki,
R. Angelaud, Angew. Chem. 2000, 112, 3676; Angew. Chem.
Int. Ed. 2000, 39, 3532.
[31] K. Mikami, S. Matsukawa, Nature 1997, 385, 613.
[32] T. Ohkuma, H. Doucet, T. Pham, K. Mikami, T. Korenaga,
M. Terada, R. Noyori, J. Am. Chem. Soc. 1998, 120, 1086.
[33] S. Matsukawa, K. Mikami, Tetrahedron: Asymmetry 1997, 8,
815.
[34] K. Ding, A. Ishii, K. Mikami, Angew. Chem. 1999, 111, 519;
Angew. Chem. Int. Ed. 1999, 38, 497.
[35] K. Mikami, M. Terada, in: Comprehensive Asymmetric Cataly-
sis (Eds.: K. Mikami, M. Terada), Springer, Berlin, 1999; vol.
3, pp. 1143–1174.
[36] L. Pu, H.-B. Yu, Chem. Rev. 2001, 101, 757.
[37] K. Mikami, R. Angelaud, K. L. Ding, A. Ishii, A. Tanaka, N.
Sawada, K. Kudo, M. Senda, Chem. Eur. J. 2001, 7, 730.
[38]
A. M. Costa, C. Jimeno, J. Gavenonis, P. J. Carroll, P. J. Walsh,
J. Am. Chem. Soc. 2002, 124, 6929.
[39]
[40]
H. Ushio, K. Mikami, Tetrahedron Lett. 2005, 46, 2903.
J. Gajewy, J. Gawronski, M. Kwit, Monatsh. Chem. 2012, 143,
1045.
[41]
[42]
V. Bette, A. Mortreux, C. Lehmann, J.-F. Carpentier, Chem.
Commun. 2003, 332.
V. Bette, A. Mortreux, D. Savoia, J.-F. Carpentier, Adv. Synth.
Catal. 2005, 347, 289.
[43]
[44]
B.-M. Park, S. Mun, J. Yun, Adv. Synth. Catal. 2006, 348, 1029.
J. Gajewy, J. Gawronski, M. Kwit, Org. Biomol. Chem. 2011,
9, 3863.
[45]
[46]
[47]
P. H.-Y. Cheong, C. Y. Legault, J. M. Um, N. Çelebi-Ölçüm,
K. N. Houk, Chem. Rev. 2011, 111, 5042.
W. Koch, M. C. Holthasuen, A Chemist’s Guide to Density
Functional Theory; Wiley-VCH: Weinheim, 2001.
a) Y. Zhao, D. G. Truhlar, J. Chem. Theory Comput. 2007, 3,
289; b) Y. Zhao, D. G. Truhlar, J. Chem. Theory Comput. 2005,
1, 415.
[48]
[49]
L. Simón, J. M. Goodman, Org. Biomol. Chem. 2011, 9, 689.
M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B.
Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li,
H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Son-
nenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hase-
[20] H. Nishiyama, S. Yamaguchi, M. Kondo, K. Itoh, J. Org.
Chem. 1992, 57, 4306.
Eur. J. Org. Chem. 2013, 307–318
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
317

J. Gajewy, J. Gawronski, M. Kwit
FULL PAPER
gawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai,
T. Vreven, J. A. Montgomery Jr., J. E. Peralta, F. Ogliaro, M.
Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Starov-
erov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell,
J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, N. J.
Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Ad-
amo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev,
A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Mar-
tin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador,
J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B.
Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian 09,
revision A.02, Gaussian, Inc., Wallingford CT, 2009.
[65] M. Kwit, U. Rychlewska, J. Gawronski, New J. Chem. 2002,
26, 1714.
[66] a) A. Johansson, E. Wingstrand, M. Håkansson, J. Organomet.
Chem. 2005, 690, 3846; b) V. K. Muppidi, P. S. Zacharias, S.
Pal, Inorg. Chem. Commun. 2005, 8, 543; c) U. P. Singh, P.
Tyagi, S. Pal, Inorg. Chim. Acta 2009, 362, 4403; d) T. M.
McCormick, S. Wang, Inorg. Chem. 2008, 47, 10017; e) Y.
Wang, H. Fu, F. Shen, X. Sheng, A. Peng, Z. Gu, H. Ma, J. S.
Ma, J. Yao, Inorg. Chem. 2007, 46, 3548; f) A. Lalehzari, J.
Desper, C. J. Levy, Inorg. Chem. 2008, 47, 1120; g) S. G. Roh,
J. U. Yoon, J. H. Jeong, Polyhedron 2004, 23, 2063; h) S. J.
Birch, S. R. Boss, S. C. Cole, M. P. Coles, R. Haigh, P. B.
Hitchcock, A. E. H. Wheatley, Dalton Trans. 2004, 3568.
[67] G. R. Fulmer, A. J. M. Miller, N. H. Sherden, H. E. Gottlieb,
A. Nudelman, B. M. Stoltz, J. E. Bercaw, K. I. Goldberg, Orga-
nometallics 2010, 29, 2176.
[50]
[51]
a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648; b) C. Lee, W.
Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785; c) A. D. Becke,
Phys. Rev. A 1988, 38, 3098; d) J. P. Perdew, Phys. Rev. B 1986,
33, 8822.
[68] a) A. J. De Koning, J. Boersma, G. J. M. van der Kerk, J. Or-
ganomet. Chem. 1980, 195, 1; b) G. V. Goeden, K. G. Caulton,
J. Am. Chem. Soc. 1981, 103, 7354; c) E. C. Ashby, A. B. Goel,
Inorg. Chem. 1981, 20, 1096.
[69] a) N. A. Bell, A. L. Kassyk, J. Organomet. Chem. 1988, 345,
245; b) T. L. Neils, J. M. Burtlich, Inorg. Chem. 1989, 28, 1607;
c) N. A. Bell, A. L. Kassyk, Inorg. Chim. Acta 1996, 250, 345.
[70] B. Gutschank, S. Schultz, D. Bläser, R. Boese, C. Wölper, Or-
ganometallics 2010, 29, 6133.
a) P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996,
77, 3865; b) C. Adamo, V. Barone, J. Chem. Phys. 1999, 110,
6158.
J. Vàzquez, M. A. Perics, F. Maseras, A. Lledós, J. Org. Chem.
2000, 65, 7303.
J. Rudolph, C. Bolm, P. O. Norrby, J. Am. Chem. Soc. 2005,
127, 1548.
F. Wang, J.-M. Wang, M. Li, Struct. Chem. 2009, 20, 129.
M. Yamakawa, R. Yamakawa, Organometallics 1999, 18, 128.
[52]
[53]
[54]
[55]
[56]
[71] S. Schulz, T. Eisenmann, D. Schuchmann, M. Bolte, M. Kirch-
ner, R. Boese, J. Spielmann, S. Z. Harder, Z. Naturforsch. B
2009, 64, 1397.
N. Qi, R.-Z. Liao, J.-G. Yu, R. Z. Liu, J. Comput. Chem. 2010,
31, 1376.
[57] K. A. Nolin, J. R. Krumper, M. D. Pluth, R. Bergman, F. D.
Toste, J. Am. Chem. Soc. 2007, 129, 14684.
[58] a) M. Dress, T. Strassner, Inorg. Chem. 2007, 46, 10850; b) E.
Peterson, A. Y. Khalimon, R. Simionescu, L. G. Kuzmina,
J. A. K. Howard, G. I. Nikonov, J. Am. Chem. Soc. 2009, 131,
908.
[72] M. Kwit, M. D. Rozwadowska, J. Gawronski, A. Grajewska, J.
Org. Chem. 2009, 74, 8051.
[73] E. L. Eliel, S. H. Wilen, Stereochemistry of Organic Compounds,
Wiley, New York 1994.
[74] For some examples of induced electronic circular dichroism,
´
see: a) J. Sciebura, P. Skowronek, J. Gawronski, Angew. Chem.
[59] X.-H. Zhang, L. W. Chung, Z. Lin, Y.-D. Wu, J. Org. Chem.
2008, 73, 820.
2009, 121, 7203; Angew. Chem. Int. Ed. 2009, 48, 7069; b) J.
Gawronski, M. Kwit, K. Gawronska, Org. Lett. 2002, 4, 4185;
c) S. Hosoi, M. Kamiya, T. Ohta, Org. Lett. 2001, 3, 3659.
[75] E. D. Glendening, A. E. Reed, J. E. Carpenter, F. Weinhold,
NBO Version 3.1.
[76] a) F. Weinhold, in: Encyclopedia of Computational Chemistry
(Eds.: P. v. R. Schleyer, N. L. Allinger, T. Clark, J. Gasteiger,
P. A. Kollman, H. F. Schaefer III, P. R. Schreiner), Wiley,
Chichester, UK, 1998, vol. 3, pp. 1792–1811; b) A. E. Reed,
L. A. Curtiss, F. Weinhold, Chem. Rev. 1988, 88, 899; c) F.
Weinhold, J. E. Carpenter, in: The Structure of Small Molecules
and Ions (Eds.: R. Naaman, Z. Vager), Plenum, New York,
1988, pp. 227–236; d) F. Weinhold, in: Computational Methods
in Organic Photochemistry: Molecular and Supramolecular Pho-
tochemistry (Ed.: A. G. Kutateladze), Taylor & Francis/CRC
Press, Boca Raton, 2005, pp. 393–476.
[60] L. W. Chung, Y.-D. Wu, B. M. Trost, Z. T. Ball, J. Am. Chem.
Soc. 2003, 125, 11578.
[61] T. Gathy, D. Peeters, T. Leyssens, J. Organomet. Chem. 2009,
694, 3943.
[62] a) S. Sakaki, M. Sumimoto, M. Fukuhara, M. Sugimoto, H.
Fujimoto, S. Matsuzaki, Organometallics 2002, 21, 3788; b) I.
Ojima, T. Kogure, M. Kumagai, S. Horiuchi, Y. Sato, J. Or-
ganomet. Chem. 1976, 122, 83; c) G. Z. Zheng, T. H. Chan,
Organometallics 1995, 14, 70; d) C. Reyes, A. Prock, W. P. Gier-
ing, Organometallics 2002, 21, 546; e) N. Schneider, M. Finger,
C. Haferkemper, S. Bellemin-Laponnaz, P. Hofmann, L. H.
Gade, Angew. Chem. 2009, 121, 1637; Angew. Chem. Int. Ed.
2009, 48, 1609; f) N. Schneider, M. Finger, C. Haferkemper, S.
Bellemin-Laponnaz, P. Hofmann, L. H. Gade, Chem. Eur. J.
2009, 15, 11515.
[63] a) Comprehensive Chiroptical Spectroscopy (Eds.: N. Berova,
P. L. Polavarapu, K. Nakanishi, R. W. Woody), John Wiley &
Sons, Hoboken, USA, 2012; b) J. Gawronski, M. Kwit, D. R.
Boyd, N. D. Sharma, J. F. Malone, A. F. Drake, J. Am. Chem.
Soc. 2005, 127, 4308; c) M. Kwit, N. D. Sharma, D. R. Boyd,
J. Gawronski, Chem. Eur. J. 2007, 13, 5812; d) M. Kwit, J. Ga-
wronski, D. R. Boyd, N. D. Sharma, M. Kaik, R. A.
More O’Ferrall, J. S. Kudavalli, Chem. Eur. J. 2008, 14, 11500.
[64] R. Ahlrichs, M. Baer, M. Haeser, H. Horn, C. Koelmel, Chem.
Phys. Lett. 1989, 162, 165 see also www.turbomole-gmbh.com
for an overview of the TURBOMOLE program.
[77] a) A. L. Singer, D. A. Atwood, Inorg. Chim. Acta 1998, 277,
157; b) Y.-H. Chin, O. dos Santos, J. W. Canary, Tetrahedron
1999, 55, 12069; c) A. V. Wiznycia, J. Desper, C. J. Levy, Can.
J. Chem. 2009, 87, 224; d) M. E. Germain, T. R. Vargo, P. G.
Khalifah, M. J. Knapp, Inorg. Chem. 2007, 46, 4422; e) E. C.
Escudero-Adán, J. Benet-Buchholz, A. W. Kleij, Eur. J. Inorg.
Chem. 2009, 3562; f) G. A. Morris, H. Zhou, C. L. Stern, S. T.
Nguyen, Inorg. Chem. 2001, 40, 3222.
[78] D. H. Ess, J. Kister, M. Chen, W. R. Roush, J. Org. Chem. 2009,
74, 8626, and references cited therein.
Received: July 25, 2012
Published Online: November 22, 2012
318
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 307–318