Diastereoselective hydrogen-transfer reactions: An experimental and DFT study

Article abstract of DOI:10.1002/chem.201300377

Radical reductions of halogenated precursors bearing a heterocycle exo (α) to the carbon-centered radical proceed with enhanced anti-selectivity, a phenomenon that we termed exocyclic effect . New experimental data and DFT calculations at the BHandHLYP/TZVP level demonstrate that the origin of the exocyclic effect is linked to the strain energy required for a radical intermediate to reach its reactive conformation at the transition state (ΔE^≠_strain). Furthermore, radical reductions of constrained THP systems indicate that high 2,3-anti inductions are reached only when the radical chain occupies an equatorial orientation. Hydride deliveries to different acyclic substrates and calculations also suggest that the higher anti-selectivities obtained with borinate intermediates are not related to the formation of a complex mimicking an exocycle. From a broader standpoint, this study reveals important conformational factors for reactions taking place at a center vicinal to a heterocycle or an α-alkoxy group. Copyright

Full text of DOI:10.1002/chem.201300377

DOI: 10.1002/chem.201300377
Diastereoselective Hydrogen-Transfer Reactions: An Experimental and DFT
Study
FranÅois Godin,[a, b] Michel Prꢀvost,[a, b] Serge I. Gorelsky,[c] Philippe Mochirian,[a, b]
Maud Nguyen,[a, b] Frꢀdꢀrick Viens,[a, b] and Yvan Guindon*[a, b]
Abstract: Radical reductions of halo-
genated precursors bearing a heterocy-
cle exo (a) to the carbon-centered radi-
cal proceed with enhanced anti-selec-
tivity, a phenomenon that we termed
“exocyclic effect”. New experimental
data and DFT calculations at the
BHandHLYP/TZVP level demonstrate
that the origin of the exocyclic effect is
linked to the strain energy required for
a radical intermediate to reach its reac-
tive conformation at the transition
state (DE°strain). Furthermore, radical
reductions of constrained THP systems
indicate that high 2,3-anti inductions
are reached only when the radical
chain occupies an equatorial orienta-
tion. Hydride deliveries to different
acyclic substrates and calculations also
suggest that the higher anti-selectivities
obtained with borinate intermediates
are not related to the formation of a
complex mimicking an exocycle. From
a broader standpoint, this study reveals
important conformational factors for
reactions taking place at a center vici-
nal to a heterocycle or an a-alkoxy
group.
Keywords: borinate
·
conforma-
tional analysis · density-functional
calculations · diastereoselectivity ·
radical reactions
Introduction
Creating stereogenic centers using free-radical intermediates
has been a topic of great interest.[1–3] Our group contributed
to this field[4] by developing efficient stereoselective tandem
strategies involving radical reactions with anionic processes
that allowed the efficient synthesis of complex fragments of
biologically relevant ionophores,[5] such as zincophorin
(1a)[6] or narasin (1b)[7] (Figure 1).
The strategies that we have developed involved carbon-
center radicals (generated from a mixture of halogenated
precursors, such as 2a,b and 4a,b) flanked on one side by an
ester and, on the other, by a stereogenic center bearing an
Figure 1. Structure of zincophorin (1a) and narasin (1b).
ꢀ
electron-withdrawing group. Substrates with a C O bond at
C3 embedded in a ring were reduced with higher selectivi-
ties than comparable acyclic substrates having a small alkyl
group at that position [Eq. (1) vs. Eq. (2)].[8]
This magnification of selectivity with cyclic systems was
therefore termed the exocyclic effect. Taking advantage of
this effect allowed for the reliable synthesis of 2,3-anti propi-
[a] F. Godin, Dr. M. Prꢀvost, Dr. P. Mochirian, M. Nguyen, F. Viens,
Prof. Dr. Y. Guindon
Dꢀpartement de Chimie, Universitꢀ de Montrꢀal
C.P. 6128, Succursale Centre-Ville, Montrꢀal
Quꢀbec, H3C 3J7 (Canada)
[b] F. Godin, Dr. M. Prꢀvost, Dr. P. Mochirian, M. Nguyen, F. Viens,
Prof. Dr. Y. Guindon
Bio-organic Chemistry Laboratory
Institut de Recherches Cliniques de Montrꢀal (IRCM)
110 avenue des Pins Ouest, Montrꢀal, Quꢀbec, H2W 1R7 (Canada)
Fax : (+1)514-987-5789
E-mail: yvan.guindon@ircm.qc.ca
[c] Dr. S. I. Gorelsky
Department of Chemistry and Center for
Catalysis Research and Innovation, University of Ottawa
10 Marie Curie, Ottawa, Ontario, K1N 6N5 (Canada)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201300377.
9308
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 9308 – 9318

FULL PAPER
onate relationships[9] found within numerous ionophores
bearing tetrahydropyran (THP) rings or tetrahydrofuran
rings (THF).[5,8a–c,e,f]
To overcome the limitations observed with less substituted
acyclic compounds, such as 4a,b, we next studied the reduc-
bearing an exocycle [Eq. (1) vs. Eq. (2)] or a borinate pro-
tecting group [Eq. (5)]. The impact of axial or equatorial
orientation of the chain bearing the radical exo to a THP
was also examined with constrained bicyclic systems. Im-
pressive correlation between the experimental results and
theoretical calculations were found and improved our under-
standing of the reactivity of these radical intermediates. We
show that the exocyclic and the borinate effects, which lead
to an increase of diastereoselectivity, originate from the alle-
viation of destabilizing interactions with the incoming hy-
dride in the transition state leading to the major product
(2,3-anti), whereas the presence of a bulkier side chain at C3
increases the energy of the transition state leading to the
minor isomer (2,3-syn).
ꢀ
tion of C3 C5 diol radical precursors temporarily protected
as acetonides or benzylidenes to take advantage of the exo-
cyclic effect [Eq. (3)].[8c] The high 2,3-anti inductions ob-
tained then encouraged us to develop single-pot methodolo-
gies with boron Lewis acids to form in situ cyclic boronates
and borinates [Eq. (4) and Eq. (5)].[4f,8f] Aluminum and
magnesium Lewis acids, which were demonstrated to allow
the formation of endocyclic radical intermediates by com-
ꢀ
plexation of the ester and the C3 alcohol group, afforded
selectively
the
complementary
2,3-syn
inductions
[Eq. (6)].[4a,f,10] The stereocontrol brought by these radical
reduction strategies proved to be highly reliable. By combin-
ing them with anionic Mukaiyama aldol reactions, a compre-
hensive approach to the preparation of polypropionates[4f]
was elaborated for the synthesis of all possible 16 polypropi-
onate stereopentads as single diastereoisomers from a-chiral
b-hydroxyaldehydes.[4d]
Results and Discussion
The diastereoselectivity enhancement for the radical reduc-
tion of systems bearing an exocycle at C3 was first hypothe-
sized to be linked to the conformational biases imposed at
ꢀ
ꢀ
the C3 O and C3 C4 bonds (Scheme 1). The chain at C3
Scheme 1. Acyclic and exocyclic radical intermediates 4c, 2c, 11–12c
(generated from bromide precursors 4a,b, 2a,b or 11–12a,b).
bearing the radical in the exocyclic systems should occupy
preferably either equatorial (eq) or axial (ax) positions of
low energy chair-like conformations of the THP ring 2c
(Scheme 1).[11] Although it could reasonably be assumed
that the equatorial position is preferred for steric factors, we
could not rule out completely the possibility that a radical
center might prefer to react in the axial position. In order to
gain experimental insight on the impact of equatorial or
axial orientations of the C3 chain bearing the radical on a
THP ring, 6,7-trans-octahydrochromen derivatives 11a;11b
and 12a;12b were prepared (Scheme 2) and then submitted
to hydrogen-transfer reactions with trialkyltin hydride
(Table 1).[12]
The reaction sequence to access selectively either the 3,7-
cis or 3,7-trans-6,7-trans-octahydrochromen radical precur-
sors is presented in Scheme 2. The olefin 13, obtained from
copper-catalyzed Grignard addition to cyclohexene oxide,
was first treated with OsO4/NaIO4 to give the correspond-
ing aldehyde in equilibrium with its lactol isomer. Wittig
Reported herein is the first transition state DFT study
that investigates the origin of the anti-selectivity observed
for these types of radical hydrogen-transfer reactions. We
were particularly interested in elucidating the factors re-
sponsible for the enhanced stereocontrol with substrates
[13]
Chem. Eur. J. 2013, 19, 9308 – 9318
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9309

Y. Guindon et al.
products (1:5.2, 18a/18b) was induced from the axial precur-
sors 12a,b (Table 1, entries 2 and 4). The chain bearing the
radical equatorial therefore provides higher 2,3-anti selectiv-
ities than when oriented axially—an intriguing observation
that will be further examined in silico.
Description of the computational method: All geometry op-
timizations were performed using standard gradient techni-
ques and tight SCF convergence in Gaussian 09[15] with the
BHandHLYP functional.[16] This functional was demonstrat-
ed[17] to be well suited for radical systems.[18] Transition state
calculations were performed with Me3SnH. This hydride led
to experimental selectivities comparable to Bu3SnH (vide
infra). The effective core potential of Hay and Wadt[19]
LANL2DZ supplemented with a single set of d-type polari-
zation functions was used for tin (d(z)Sn =0.20),[18b,d,20] to-
gether with the valence triple-z TZVP basis set[21] for all
other atoms. For each reaction, the difference in Gibbs free
energies of activation (DDG°) was calculated between the
lowest energy anti- and syn-predictive TS, and the ratio of
products was calculated from Equation (7).
Scheme 2. Synthesis of radical precursors 11a,b and 12a,b. Reaction con-
ditions: a) OsO4, 2,6-lutidine, NaIO4, 1,4-dioxane/H2O (3:1), RT;
b) Ph3PC(Me)=CO2Et, CH2Cl2, RT; c) NBSac, MeCN, 08C to RT;
d) Ac2O, pyridine/CH2Cl2 (1:1), 08C to RT; e) SnCl4, 16, CH2Cl2, ꢀ788C.
olefination provided the a,b-unsaturated ester 14a,b (10:1,
E/Z), whereas treatment with acetic anhydride afforded a
mixture of the corresponding acetals 15a,b (3:1). The insep-
arable mixture of 14a,b was then treated with N-bromosac-
charin (NBSac)[14] to provide a mixture of C2 tertiary bro-
mides 11a;11b with high 3,7-cis selectivity. Alternatively,
acetals 15a,b were treated with tetrasubstituted enoxysilanes
½2,3-antiꢂ
DDG° ¼ RT ꢁ ln
ð7Þ
½2,3-synꢂ
[4e]
To determine Gibbs free energies in solution, geometry
optimizations were performed using the IEFPCM implicit
solvation model.[22] More details concerning the computa-
tional method used are provided in the Experimental Sec-
tion and in the Supporting Information.
16 and SnCl4 to afford C2 tertiary bromides 12a;12b with
excellent 3,7-trans stereocontrol.
Table 1. Radical reductions of 6,7-trans-octahydrochromen radical pre-
cursors 11a,b and 12a,b.
Entry R3SnH
Solvent
d.r. (a/b)[a] Yield [%][b]
DFT study of THP radical intermediates: The different
energy barriers of the radical ground state were evaluated
to support the hypothesis that the radical reduction were
under Curtin–Hammett conditions.[23] THP radical inter-
mediate 2c displayed a low ground state conformer with an
equatorial radical chain and two possible E- and Z-enol rad-
icals (2c-E and 2c-Z), resulting from radical delocalization
1
2
Ph3SnH
Bu3SnH
PhMe
DCM
>20:1 72
>20:1 74
3
4
Ph3SnH
Bu3SnH
PhMe
DCM
1.8:1 76
[c]
1:5.2
–
[a] Product ratios were determined by 1H NMR analysis of the crude re-
action mixture. [b] Yields of isolated products. [c] Complete conversion
1
was observed by H NMR spectroscopy.
The radical precursors 11a,b (bearing the C3-chain equa-
torial) was reduced with an excellent ratio favoring the 2,3-
anti product 17a (>20:1), whereas a strikingly poor ratio
(1.8:1, 18a/18b) was obtained for substrates 12a,b with the
radical-bearing chain forced to occupy an axial orientation
(Table 1, entries 1 and 3). It is noteworthy that when per-
forming the reduction with Bu3SnH in DCM, high 2,3-anti
selectivity was maintained with equatorial precursors 11a,b,
whereas an inversion of selectivity in favor of the 2,3-syn
ꢀ
Figure 2. Energy barriers of unsubstituted THP radical 2c: a) C1 C2 iso-
merization of E-Z enol radical, and b) equatorial (eq)–axial (ax) chairs
ring flip. Gibbs free energies [kcalmolꢀ1] in toluene are reported relative
to the energy of 2c-eq-E.
9310
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 9308 – 9318

Diastereoselective Hydrogen-Transfer Reactions
FULL PAPER
in the ester moiety (Figure 2).[24] Transition state A involved
in the possible equilibration between these equatorial enol
radicals displays a 10.24 kcalmolꢀ1 barrier that was deter-
mined by the QST3 method in Gaussian (Figure 2).[25] The
THP ring flip through half-chair TS B was located at
10.62 kcalmolꢀ1 for the E isomer. These energy barriers are
considered to allow ground state equilibration at rates sig-
nificantly higher than the intermolecular hydride delivery,
which occurred with transition state energies greater than
16.3 kcalmolꢀ1 at ꢀ788C.[26]
fore also represent an important factor to explain why anti-
predictive TS C and syn-predictive TS H are significantly
lower in energy than the other transition states (Figure 3).[28]
The DFT study allowed to confirm that TS C would be
preferred because of optimal intramolecular dipole–dipole
[24c,32]
ꢀ
minimization and s-donation of the C3 C4 bond
and
revealed the conformation of the lowest syn-predictive TS
H. The presence of more severe allylic-1,2 strain (A1,2) in
the latter could also account for the lower energy of anti-
predictive TS C. According to the DDG° calculated between
C-2c-eq[33] and H-2c-eq in toluene (2.31 kcalmolꢀ1) excel-
lent selectivities in favor of the 2,3-anti product 2a are ex-
pected at ꢀ788C (>380:1), which is in agreement with the
experimental >50:1 ratio determined by GLC analysis of
the crude reaction mixture (Figure 4a).[8a] This DDG° value
also correlates with the calculated and experimental ratios
obtained for bicyclic substrate 11c with the radical chain
constrained in the equatorial position, in toluene or DCM
(Figure 4b).
Transition states analysis of THP intermediates with an
ꢀ
equatorial radical chain: C2 C3 Fischer projections of the
low energy transition state conformers found for the hydride
delivery to radical intermediates are presented at
Figure 3.[28] The reported energies in toluene were obtained
ꢀ
Figure 3. Fischer projections of the C2 C3 conformers of the lowest
energy transition states. Gibbs free energies in parentheses [kcalmolꢀ1] in
toluene were obtained for the unsubstituted THP radical equatorial
2c-eq (generated from 2a,b) for the E and Z isomers and are reported
relative to the energy of TS C-E.[27]
for the unsubstituted THP 2c with the C3-radical chain in
ꢀ
ꢀ
ꢀ
the equatorial orientation (R R’= ACTHNUGTRNEG(UN CH2)4 ). The average
ꢀ
ꢀ
C2 H and Sn H Wiberg bond orders (0.3 and 0.5, respec-
tively) suggest that the reduction occurs in a relatively early
TS.[29] Starting from an almost planar conformation (0.68) at
the ground state (2c-E or -Z, Figure 2), the C3 center of the
radical intermediate experiences approximately 108 of pyra-
midalization[30] at the TS.[31] The allylic-1,3 strain (A1,3) mini-
mization observed for the radical intermediate could there-
Figure 4. Lowest energy transition states of hydride delivery to: a) unsub-
stituted THP radical 2c-eq, and b) equatorial bicyclic radical 11c. Gibbs
free energies [kcalmolꢀ1] in toluene are reported in parentheses relative
to the energy of 2c-eq-E. Atomic distances [ꢂ] and Wiberg bond orders
ꢀ
ꢀ
for Sn H and C2 H interactions are shown for the TS structures.
Chem. Eur. J. 2013, 19, 9308 – 9318
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9311

Y. Guindon et al.
Transition states analysis of THP intermediates with an
axial radical chain: The calculated energies obtained with
the bicyclic system 12c, locking the radical chain in the axial
position, correlate well with the poor anti-diastereoselectivi-
ty obtained in toluene and the reversal of selectivity towards
the syn-product in DCM (Figure 5).
tance of considering solvents when studying these radical re-
actions in silico.[34]
In summary, the analysis of cyclic systems 11c and 12c in-
dicated the crucial impact of the conformation (equatorial
vs. axial) of the C3-chain bearing the radical to rationalize
the increase in diastereoselectivity. We then sought to identi-
fy other factors that could explain why acyclic substrates
generally lead to a decrease in 2,3-anti selectivity [Eq (1) vs.
Eq. (2)].
Transition states analysis of acyclic substrate 4c with a sec-
ondary C3 side chain: The acyclic radical substrate 4c was
more challenging to analyze because it required the exami-
ꢀ
ꢀ
ꢀ
ꢀ
nation of all the possible C1 C2, C2 C3, C3 C4 and C4’ O
conformations (Figure 6). All the corresponding rotational
energy barriers for 4c were calculated to be significantly
lower than the energy required to reach the transition states,
therefore allowing for Curtin–Hammett conditions to be
met.[35] In the lowest energy ground state conformation 4c-E
(0 kcalmolꢀ1), it is interesting to note that the methoxy
ꢀ
group (C4’ O) is pointing downward. This conformation is
unlikely to be preferred at the transition state, because
severe steric clash would occur with the incoming hydride.[36]
Figure 5. Transition states of hydride delivery to axial bicyclic radical
12c. Gibbs free energies [kcalmolꢀ1] in toluene are reported in parenthe-
ses relative to the energy of 2c-ax-E.
The analysis of the lowest energy transition states C and
H indicates that the chair conformation is distorted slightly
ꢀ
to orient the radical chain (C2 C3) away from the fully
axial position (Figure 5). This could alleviate partially two
ꢀ
ꢀ
high energy pseudo syn-pentane-like interactions (Me C2
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ ꢀ
C3 C4 C5 and Me C2 C3 O C7) in C-12c-ax and one
ꢀ
ꢀ
ꢀ
ꢀ
pseudo syn-pentane interaction (Me C2 C3 C4 C5) in
H-12c-ax. The presence of an additional pseudo syn-pentane
interaction in C-12c-ax should result in a more significant
destabilization of the anti-predictive TS. The optimal intra-
molecular dipole minimization in TS C, resulting from the
ꢀ
antiperiplanar orientation of the ester group (C2 C1) rela-
ꢀ
tive to the C3-alkoxy group (C3 O), however, could explain
that the 2,3-anti product remains preferred in low dielectric
media. In a solvent with a higher dielectric constant (e.g.,
DCM), this stabilization could become less important and
increase the relative energy of C versus H, which would be
consistent with the observed reversal to 2,3-syn selectivity in
DCM for bicyclic compound 12c. This highlights the impor-
Figure 6. Transition states of acyclic radical intermediate 4c (from 4a,b)
with a methoxy at C3 O. Gibbs free energies [kcalmolꢀ1] in DCM are re-
ꢀ
ported relative to the energy of 4c-E.
9312
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 9308 – 9318

Diastereoselective Hydrogen-Transfer Reactions
FULL PAPER
Indeed, in the lowest energy transition states C1-4c
DFT study of borinate intermediates: Another important
aspect of acyclic radical reductions is the impact of installing
a borinate at the C3 position of radical intermediate (I,
Figure 8). These boron intermediates are synthetically more
useful than methoxyether groups, since they are convenient
to install and to cleave in situ. The high anti-selectivity ob-
served for the reduction of borinate intermediates in the
context of polypropionate synthesis [R=Me, Figure 8 and
Eq. (5)] led us to suggest that a complex (II, Figure 8) was
mimicking the exocyclic THP [Eq (1)]. The borinate at C3
could also induce the formation of complex III, resulting
from the intramolecular complexation to the carbonyl of the
ester. The hydrogen transfer reaction on such intermediate
should, however, lead to the 2,3-syn product, as suggested
by Equation (6).[39]
ꢀ1
ꢀ1
ꢀ
(20.13 kcalmol ) and C2-4c (19.93 kcalmol ), the C4’ O
rotamer adopts a conformation to orient the C4’ away from
the hydride reagent. The 0.52 kcalmolꢀ1 free energy differ-
ence (DDG°) between the lowest syn-predictive TS H-4c
and C2-4c in DCM predicts an anti-selectivity of 3.8:1
[Eq. (7)] in agreement with the experimental selectivity
(1.5:1).
The origin of the exocyclic effect: Acyclic system 4c adopts
a reactive conformation leading to 2,3-anti product that
presents destabilizing interactions not found in the exocyclic
system 2c (Figures 6 and 4a). This was demonstrated by
energy decomposition of the transition state energy (DE°)
using
the
activation
strain
model[37,38]
(DE° =
DE°
strainA
strainA
tion energy (DE°int) and the strain energy (DE°strain) of the
hydride were found to be marginally different in these two
systems, which reflects the fact that no severe steric clash
with the incoming hydride was found.[35] The strain energy
gap of the radical for reaching the anti-predictive TS C or
C2 (Figure 7), however, is markedly greater in acyclic sys-
tems (3.30 vs. 6.30 kcalmolꢀ1; 2c vs. 4c), whereas the
DE°
in the syn-predictive TS H for both systems are
strain
comparable in energy (5.35 vs. 5.67 kcalmolꢀ1; 2c vs. 4c).
The additional interaction for radical intermediate 4c (i.e.,
C4’ with C4 and C5 with C2) in the lowest anti-predictive
TS (C2), could in part explain this increase in relative
energy. Linking C4’ and C5 as an exocycle would provide
high 2,3-anti inductions by avoiding these unfavorable inter-
actions in the anti-predictive TS, which leads to an enhanced
DDG° gap relative to acyclic intermediate 2c.
Figure 8. Various complexation modes of C3,C5-alkoxy radical intermedi-
ates and resulting outcome (2,3-anti or 2,3-syn) of hydride transfer reac-
tion.
To determine if the formation of an exocyclic complex
was a prerequisite to reach high 2,3-anti selectivity, radical
precursors 19a,b (R=BBu2) were prepared and reduced
after in situ formation of the corresponding borinate inter-
mediate (Table 2, entry 2). A marginally higher 2,3-anti ratio
was obtained with the latter, as compared with the corre-
sponding C3-methyl protected 4a,b (4.2:1 vs. 1.5:1, entry 2
vs. 1). Radical precursors 20–23a,b with secondary side
chain at C3 and bearing an alkoxy group at C5 were then
examined under the standard reducing conditions (Table 2,
entries 3–6). The borinate intermediates furnished signifi-
cantly higher 2,3-anti selectivity than their C3-methoxy pro-
tected counterparts.
These results were puzzling, since the higher selectivity
obtained with C5-benzyloxy suggested the possible involve-
ment of an exocyclic intermediate, but we were reluctant to
invoke the complexation of a bulky silyloxy ether group by
boron, as these protecting groups are generally reported to
prevent complexation.[40] We have, therefore, decided to in-
vestigate these questions by the DFT calculations reported
herein.
In the presence of a tertiary alkyl group at C3 (Table 2,
entry 8), 2,3-anti product 32a was formed selectively from
25a,b, as it was the case with the corresponding C3-methoxy
precursor 24a,b (entry 7). The high level of diastereoselec-
tivity for precursors with bulkier alkyl side chain falls
beyond the limit of detection, and it is thus not possible to
determine if an enhancement is obtained between 24a,b,
Figure 7. The origin of the exocyclic effect. The E°
energies of the
strain
radical [kcalmolꢀ1] are reported relative to the energies of the respective
ground state structures of 2c-eq-E and 4c-E. The arrows indicate the tra-
jectory of the incoming hydride for TS C or C2 (anti-predictive) and TS
H (syn-predictive).
Chem. Eur. J. 2013, 19, 9308 – 9318
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9313

Y. Guindon et al.
Table 2. Radical reductions of acyclic radical precursors 4a,b and 19–
25a,b.
Table 3. Radical reductions of radical precursors 33–36a,b.
Entry[a]
R
R’
d.r. (a/b)[b]
Yield [%][c]
Entry
R
Solvent
d.r. (a/b)[a]
Yield [%][b]
1 (33a)
2 (34a)
OBn
OTBDPS
OBn
MOM
MOM
CF3CO
9-BBN
1.3:1 (37)
1.5:1 (38)
2.3:1 (39)
3:1 (28)
87
80
89
82
1[c] (4a,b)
Me
BBu2
DCM
DCM
1.5:1 (5)
4.2:1 (26)
89
41
2[d] (19a,b)
3 (35a)
4[d] (36a,b)
OBn
[a] X=Br (entries 1–3) and X=I (entry 4). [b] Product ratios were deter-
mined by 1H NMR analysis of the crude reaction mixture. [c] Yields of
isolated products. [d] Borinate (R’=9-BBN) was generated in situ from
secondary alcohol 21a,b (R’=H). The secondary alcohols were isolated
(R’=H); see the Supporting Information for details.
3 (20a)
Me
BBu2
DCM
DCM
1.2:1 (27)
9:1 (28)
75
86
4[d] (21a,b)
the second lowest TS C1 for methoxy-substituted acyclic in-
termediate 4c (Figure 6). The planar trigonal conformation
of the borinate likely reduces its interaction with the C5
methyl group in C-19c. The bond lengths and conformation
obtained by calculations are consistent with the X-ray data
obtained from a related borinate compound.[42] The lowest
energy transition states C-19c and H-19c were calculated to
have a DDG° of 0.21 kcalmolꢀ1, which would predict selec-
tivities of 1.7:1, slightly below the experimental value with
Bu3SnH (4.2:1, Table 2, entry 2). Since Me3SnH[43] was used
5 (22a)
Me
BBu2
DCM
DCM
1.4:1 (29)
13:1 (30)
80
70
6[d] (23a,b)
7[e,f] (24a,b)
8[d,e] (25a,b)
Me
BEt2
toluene
DCM
>20:1 (31)
>20:1 (32)
87
71
[a] Product ratios were determined by 1H NMR analysis of the crude re-
action mixture. [b] Yields of isolated products. [c] Ref. [8c]. [d] Borinate
(R=BBu2 or BEt2) was generated in situ from secondary alcohol (R=
H). The secondary alcohols were isolated (R=H); see the Supporting In-
formation for details. [e] X=Br (entry 7) and X=I (entry 8).
[f] Ref. [44].
25a,b and 9a,b [Table 2, entries 7–8 and Eq. (5)]. Neverthe-
less, these results indicate that borinate intermediates are re-
duced with high anti-selectivity through an acyclic mode
when a tertiary side chain is present at C3.
The last important point that we explored experimentally
is the possibility that borinates provide greater polarization
ꢀ
of the C3 O bond by delocalization of oxygen lone pairs in
the empty p-orbital of boron. The presence of an electron-
withdrawing group at C3 was previously shown to be man-
datory for 2,3-anti selectivity.[3c,32] Substrates protected with
electron-withdrawing groups, such as MOM[41] or trifluoroa-
cetate, were synthesized to examine if they would also in-
crease 2,3-anti inductions (Table 3). None of these sub-
strates, however, proceeded with levels of selectivity compa-
rable to the ones obtained with borinates (Table 3, en-
tries 1–3 vs. Table 2, entries 4 and 6). Finally, an erosion of
diastereoselectivity was observed when the size of the sub-
stituents was increased on the borinate (Table 3, entry 4).
Transition state analysis for the reductions of borinate inter-
mediates: The reduction of the borinate radical intermediate
19c was analyzed in silico (Figure 9a). The preferred TS
Figure 9. Transition states of acyclic radical intermediates 19c (in situ
ꢀ
from 19a,b) with a diethylborinate (BEt2) at C3 O. Gibbs free energies
[kcalmolꢀ1] of TS structures in DCM are reported relative to the energy
ꢀ
ꢀ
ꢀ
G
of 19c-E.
9314
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 9308 – 9318

Diastereoselective Hydrogen-Transfer Reactions
FULL PAPER
in our calculations, we decided to verify experimentally if
the hydrogen transfer gave similar results than with
Bu3SnH, which was used throughout this study with acyclic
substrates. The reduction of compound 19a,b was also eval-
uated with Ph3SnH at ꢀ788C (Figure 9). The level of anti-se-
lectivity obtained with the former (4.6:1) was similar to the
ratio Bu3SnH provided (4.2:1), whereas reduction with the
bulkier hydride (Ph3SnH) led to somewhat higher selectivity
(5.9:1, Figure 9). These results suggest that the use of a bori-
nate as a protecting group of the alkoxy group at C3 in-
creases the 2,3-anti ratio as compared to corresponding
methyl ether. A more extensive in silico evaluation of the
borinate effect was, however, necessary to understand this
selectivity trend.
tivity than the corresponding substrate bearing a C3-ethyl
group (Table 2, entry 2 vs. 4). To determine if complexation
of the C5-alkoxy was at the origin of this modest, but yet,
significant enhancement of selectivity (II, Figure 8), transi-
tion state calculations were performed with radical 40c
(Figure 10). The anti- or syn-predictive TS obtained only dis-
ꢀ
played a weak B O5 interaction and these structures were
not the lowest in energy (C1 vs. C2 and H1 vs. H2). More-
over, these calculations indicated that complexation of the
boron with the ester (III, Figure 8) is likely not involved in
the syn-predictive TS. The enhancement of anti-selectivity
with borinates is therefore not linked to the formation of a
complex and may result from the particular conformation of
these intermediates (Table 2, entries 4 and 6).
Exocyclic complexation versus acyclic reduction of borinate
intermediates bearing a C5-alkoxy group: It was observed
experimentally that substrates with a secondary side chain
at C3 and an alkoxy group at C5 provide higher anti-selec-
Transition state analysis of radical precursors with a tertiary
C3 side chain: The transition state analysis performed on
24c (Figure 11) revealed that the lowest energy anti-predic-
ꢀ
tive TS C-24c adopts a staggered C4’ O conformation, simi-
lar to TS C1-4c (fC4’-O-C3-H3 ꢀ1.58; Figure 6). One could
expect that this unfavorable conformation would raise the
energy of the anti-predictive TS C, leading to lower selectiv-
ity. The analysis of the lowest energy syn-predictive TS
H-24c, however, indicates that an unfavorable syn-pentane-
like interactions is present between the C5 and C2-methyl
group (Figure 11). The resulting DDG° value therefore re-
Figure 10. Borinate intermediates 40c in TS C (anti-predictive) or H
(syn-predictive) with conformations analogous to I or II (Figure 8). Ener-
gies [kcalmolꢀ1] in DCM are reported relative to the energy of C2-40c
Figure 11. Transition states of acyclic radical intermediate 24c (from
24a,b) with a methoxy at C3 O. Gibbs free energies [kcalmolꢀ1] of TS
ꢀ
ꢀ
and B O5 atomic distances [ꢂ] are shown for the TS structures.
structures in toluene are reported relative to the energy of 24c-E.
Chem. Eur. J. 2013, 19, 9308 – 9318
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9315

Y. Guindon et al.
mains high (1.58 kcalmolꢀ1)[35] and correlates with the high
selectivity (>20:1) measured experimentally at ꢀ788C
(Figure 11).[44] It can, therefore, be proposed that a bulkier
C3 alkyl side chain generates high diastereoselectivity by
raising the energy of the syn-predictive TS H. Various tin
hydrides (Me3SnH and Ph3SnH) were examined for the re-
duction of 24c at 258C (Figure 11). As for the reduction of
19c (Figure 9), Me3SnH provided selectivities comparable to
Bu3SnH (6.3:1 vs. 5.8:1).
Figure 13. The DE°
of radical 24c versus 25c in TS C (anti-predic-
strain
tive). Electronic energies [kcalmolꢀ1] are reported relative to the ener-
gies of the respective ground state structures of 24c-E and 25c-E.
Calculations for borinate radical intermediate 25c indicat-
ed a sufficient DDG° value to account for the selective for-
mation of the 2,3-anti product 32a (Figure 12). The excellent
Conclusion
The study of rigid bicyclic 6,7-trans-octahydrochromen radi-
cal precursors demonstrated the critical impact of position-
ing the radical-bearing chain equatorial to take advantage of
the exocyclic effect. When the radical is forced to occupy an
axial position on a THP ring, severe loss of anti-selectivity
occurs. DFT calculations of TS structures indicate that this
phenomenon can be attributed to important interactions be-
tween the C2-methyl group and THP ring centers (C5 and
C7). Functionalized THP systems, which are expected to
impose a bias to the orientation of the radical chain, are
under active investigation and will be reported in due
course.
The increased conformational energy required for acyclic
systems to reach their anti-predictive TS explains why they
undergo radical reductions with lower selectivities than exo-
cyclic THP. When a tertiary side chain is present at C3, the
syn-predictive TS are significantly destabilized by additional
syn-pentane interactions, resulting in a sufficient DDG° to
attain high selectivities. Experimental and computational
evidences indicate that high anti-selectivities with borinate
intermediates are not linked to the formation of a complex
mimicking the exocycle. These species prefer to adopt a
planar geometry that could facilitate the entry of the incom-
ing hydride in the anti-predictive TS.
In summary, this study answers fundamental questions re-
lated to the successful reduction of radical substrates. From
a more general perspective, the different conformational
factors identified herein are likely to influence other reac-
tions taking place at a center vicinal to a heterocycle or an
a-alkoxy group.
Figure 12. Transition states of acyclic radical intermediate 25c (in situ
ꢀ
from 25a,b) with a diethylborinate (BEt2) at C3 O. Gibbs free energies
[kcalmolꢀ1] of TS structures in DCM are reported relative to the energy
of 25c-E.
anti-selectivity are rationalized, as for 24c, by significant de-
stabilizing interaction of syn-TS resulting from the presence
of a bulky C3-alkyl side chain.[35] It is noteworthy that the
borinate and methoxy group adopt conformations in anti-
predictive TS C that remove severe interactions with the in-
Experimental Section
Supporting Information: The Supporting Information includes the experi-
mental procedures, physical characterization and NMR spectra (1H and
13C) of radical precursors (11–12a,b, 19–23a,b, 25a,b, 33–35a,b) and re-
duced products (17–18a,b, 26–30a,b, 32a,b, 37–39a,b); and optimized ge-
ometries, energies and Cartesian coordinates for all ground and transition
states structures.
coming hydride. The DE°
cost of borinate 25c, however,
strain
is significantly lower than for methoxy 24c (4.68 vs. 5.80,
Figure 13), likely due to pseudo-allylic-1,3 strain (A1,3) mini-
mization between the alkyl chain attached to the sp2 boron
and the vicinal C3 stereogenic center in 25c.[45] Radical 24c
Computational method for the conformational analysis of radical inter-
mediates: All geometry optimizations were performed using standard
gradient techniques and tight SCF convergence in Gaussian 09[15] with
the BHandHLYP exchange-correlation (XC) functional[16] using restrict-
ed (RBHandHLYP) and unrestricted (UBHandHLYP) methods for
closed- and open-shell systems, respectively.[46] It was previously demon-
strated[17] that BHandHLYP functional performed better for radical sys-
ꢀ
in TS C rather presents a fully eclipsed C3 O conformer.
The particular conformational bias of borinate group could
find application in other reactions in which interactions with
an alkoxy-protecting group need to be avoided.
9316
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 9308 – 9318

Diastereoselective Hydrogen-Transfer Reactions
FULL PAPER
tems than other XC functionals,[18] disparity often associated with inade-
quate treatment of the exchange terms.[47] The effective core potential of
Hay and Wadt[19] LANL2DZ supplemented with a single set of d-type
polarization functions was used for tin (d(z)Sn=0.20)[18b,d,20] together
with the valence triple-z TZVP basis set[21] for all other atoms. Ground
state (GS) radicals were evaluated by unconstrained optimization of all
possible rotating bonds including both E- and Z-enol radical conforma-
tions. Staggered transition states (TS) for hydride delivery using Me3SnH
(anti- and syn-predictive TS) were considered for each conformation of
methyl ester radical intermediates. Unscaled harmonic frequency calcula-
tions at the appropriate temperature were performed for each optimized
structure to characterize it as an energy minimum or a TS structure and
to calculate Gibbs free energies (DG°). To calculate Gibbs free energies
in solution, geometry optimizations and harmonic frequency calculations
were performed with the IEFPCM model[22] and the UFF atomic radii for
either toluene or dichloromethane. Wiberg bond indices[48] were calculat-
ed using the natural bonding orbitals[49] as implemented in Gaussi-
an 09.[15] The stereochemistry of the radical precursor at C2 was demon-
strated experimentally to be unrelated to the stereochemical outcome of
chard, C. Meyer, J. Cossy, J. Org. Chem. 2004, 69, 4626–4647; c) K.
Komatsu, K. Tanino, M. Miyashita, Angew. Chem. 2004, 116, 4441–
4445; Angew. Chem. Int. Ed. 2004, 43, 4341–4345; d) T. J. Harrison,
S. Ho, J. L. Leighton, J. Am. Chem. Soc. 2011, 133, 7308–7311;
e) Z. L. Song, A. G. Lohse, R. P. Hsung, Nat. Prod. Rep. 2009, 26,
560–571; f) G. Sabitha, R. Srinivas, J. S. Yadav, Synthesis 2011,
1484–1488.
[7] a) Y. Kishi, S. Hatakeyama, M. D. Lewis, in Frontiers of Chemistry
(Ed.: K. J. Laidler), Pergamon, Oxford, 1982; b) J. A. Tino, M. D.
Lewis, Y. Kishi, Heterocycles 1987, 25, 97–104.
[8] a) Y. Guindon, C. Yoakim, V. Gorys, W. W. Ogilvie, D. Delorme, J.
Renaud, G. Robinson, J. F. Lavallꢀe, A. Slassi, G. Jung, J. Rancourt,
K. Durkin, D. Liotta, J. Org. Chem. 1994, 59, 1166–1178; b) Y.
Guindon, A. Slassi, J. Rancourt, G. Bantle, M. Bencheqroun, L.
Murtagh, E. Ghiro, G. Jung, J. Org. Chem. 1995, 60, 288–289; c) Y.
Guindon, A. M. Faucher, E. Bourque, V. Caron, G. Jung, S. R.
Landry, J. Org. Chem. 1997, 62, 9276–9283; d) Y. Guindon, G. Jung,
B. Guꢀrin, W. W. Ogilvie, Synlett 1998, 213–220; e) Y. Guindon,
Z. P. Liu, G. Jung, J. Am. Chem. Soc. 1997, 119, 9289–9290; f) J. P.
Bouvier, G. Jung, Z. P. Liu, B. Guꢀrin, Y. Guindon, Org. Lett. 2001,
3, 1391–1394.
[9] C. H. Schiesser, M. A. Skidmore, Chem. Commun. 1996, 1419–1420.
[10] Y. Guindon, J. Rancourt, J. Org. Chem. 1998, 63, 6554–6565.
[11] a) E. L. Eliel, N. L. Allinger, S. J. Angyal, G. A. Morrison, Confor-
mational Analysis, Wiley, New York, 1965; b) E. L. Eliel, S. H.
Wilen, L. N. Mander, Stereochemistry of Organic Compounds, Wiley,
New York, 1994; c) U. Salzner, P. V. Schleyer, J. Org. Chem. 1994,
59, 2138–2155; d) E. Juaristi, Conformational Behaviour of Six-
Membered Rings, VCH, Weinheim, 1995; e) C. W. Beckett, K. S.
Pitzer, R. Spitzer, J. Am. Chem. Soc. 1947, 69, 2488–2495.
[12] Detailed experimental procedures and characterization for all new
compounds are reported in the Supporting Information.
f]
radical reduction reaction.[4a,c,d, It was thus reasonable to consider the
same radical intermediates generated from both halide isomers at the C2
position at the transition state.
Acknowledgements
The authors wish to express their gratitude to the Natural Sciences and
Engineering Research Council of Canada (NSERC) for financial support.
Fellowship support (B2) from Fonds Quꢀbꢀcois de la Recherche sur la
Nature et les Technologies (FQRNT) to F.G. is also gratefully acknowl-
edged. We also thank Prof. Tom K. Woo for providing access to the high-
performance computing facilities at the University of Ottawa.
[13] a) R. Pappo, D. S. Allen, R. U. Lemieux, W. S. Johnson, J. Org.
Chem. 1956, 21, 478–479; b) W. S. Yu, Y. Mei, Y. Kang, Z. M. Hua,
Z. D. Jin, Org. Lett. 2004, 6, 3217–3219.
[14] a) D. Urankar, I. Rutar, B. Modec, D. Dolenc, Eur. J. Org. Chem.
2005, 2349–2353; b) B. Zajc, Synth. Commun. 1999, 29, 1779–1784.
[15] Gaussian 09, Revision A.02, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, 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. Hasegawa,
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. Staroverov, R. Kobayashi, J.
Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar,
J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A.
Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, ꢃ.
Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian,
Inc., Wallingford, CT, 2009.
[1] a) D. P. Curran, W. Shen, J. C. Zhang, T. A. Heffner, J. Am. Chem.
Soc. 1990, 112, 6738–6740; b) N. A. Porter, I. J. Rosenstein, R. A.
Breyer, J. D. Bruhnke, W. X. Wu, A. T. Mcphail, J. Am. Chem. Soc.
1992, 114, 7664–7676; c) J. G. Stack, D. P. Curran, S. V. Geib, J.
Rebek, P. Ballester, J. Am. Chem. Soc. 1992, 114, 7007–7018;
d) M. P. Sibi, J. G. Ji, Angew. Chem. 1996, 108, 198–200; Angew.
Chem. Int. Ed. Engl. 1996, 35, 190–192; e) M. P. Sibi, J. G. Ji, J. Org.
Chem. 1996, 61, 6090–6091.
[2] a) J. H. Wu, R. Radinov, N. A. Porter, J. Am. Chem. Soc. 1995, 117,
11029–11030; b) M. Murakata, T. Jono, Y. Mizuno, O. Hoshino, J.
Am. Chem. Soc. 1997, 119, 11713–11714; c) N. A. Porter, J. H. L.
Wu, G. R. Zhang, A. D. Reed, J. Org. Chem. 1997, 62, 6702–6703.
[3] a) D. J. Hart, H. C. Huang, R. Krishnamurthy, T. Schwartz, J. Am.
Chem. Soc. 1989, 111, 7507–7519; b) B. Giese, M. Bulliard, H. G.
Zeitz, Synlett 1991, 425–427; c) Y. Guindon, J. F. Lavallꢀe, L. Bois-
vert, C. Chabot, D. Delorme, C. Yoakim, D. Hall, R. Lemieux, B. Si-
moneau, Tetrahedron Lett. 1991, 32, 27–30.
[16] BHandHLYP functional is a hybrid functional defined as a half-and-
half combination of HF and Becke exchange with Lee–Yang–Parr
correlation functional: 0.5ꢄEHXF +0.5ꢄEXLSDA +0.5ꢄDEXBecke88 +ELCYP
(Gaussian 09).
[17] M. Mohr, H. Zipse, D. Marx, M. Parrinello, J. Phys. Chem. A 1997,
101, 8942–8948.
[18] a) T. Morihovitis, C. H. Schiesser, M. A. Skidmore, J. Chem. Soc.
Perkin Trans. 2 1999, 2041–2047; b) H. Matsubara, C. H. Schiesser,
Org. Biomol. Chem. 2003, 1, 4335–4341; c) H. Matsubara, C. T.
Falzon, I. Ryu, C. H. Schiesser, Org. Biomol. Chem. 2006, 4, 1920–
1926; d) H. Matsubara, I. Ryu, C. H. Schiesser, J. Org. Chem. 2005,
70, 3610–3617; e) H. Matsubara, I. Ryu, C. H. Schiesser, Org.
Biomol. Chem. 2007, 5, 3320–3324; f) S. H. Kyne, C. H. Schiesser,
H. Matsubara, Org. Biomol. Chem. 2007, 5, 3938–3943; g) S. H.
Kyne, C. H. Schiesser, H. Matsubara, J. Org. Chem. 2008, 73, 427–
434.
[4] a) Y. Guindon, K. Houde, M. Prꢀvost, B. Cardinal-David, S. R.
Landry, B. Daoust, M. Bencheqroun, B. Guꢀrin, J. Am. Chem. Soc.
2001, 123, 8496–8501; b) Y. Guindon, M. Prꢀvost, P. Mochirian, B.
Guꢀrin, Org. Lett. 2002, 4, 1019–1022; c) P. Mochirian, B. Cardinal-
David, B. Guꢀrin, M. Prꢀvost, Y. Guindon, Tetrahedron Lett. 2002,
43, 7067–7071; d) J.-F. Brazeau, P. Mochirian, M. Prꢀvost, Y. Guin-
don, J. Org. Chem. 2009, 74, 64–74; e) J.-F. Brazeau, A. A. Guil-
bault, J. Kochuparampil, P. Mochirian, Y. Guindon, Org. Lett. 2010,
12, 36–39; f) P. Mochirian, F. Godin, I. Katsoulis, I. Fontaine, J.-F.
Brazeau, Y. Guindon, J. Org. Chem. 2011, 76, 7654–7676; g) F.
Godin, I. Katsoulis, E. Fiola-Masson, S. Dhambri, P. Mochirian, Y.
Guindon, Synthesis 2012, 44, 474–488.
[5] M. M. Faul, B. E. Huff, Chem. Rev. 2000, 100, 2407–2473.
[6] Four total syntheses of zincophorin (1a) have been reported to
date: a) S. J. Danishefsky, H. G. Selnick, M. P. Deninno, R. E. Zelle,
J. Am. Chem. Soc. 1987, 109, 1572–1574; b) M. Defosseux, N. Blan-
Chem. Eur. J. 2013, 19, 9308 – 9318
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9317

Y. Guindon et al.
[19] W. R. Wadt, P. J. Hay, J. Chem. Phys. 1985, 82, 284–298.
[20] R. Nakayama, H. Matsubara, D. Fujino, T. Kobatake, S. Yoshida, H.
Yorimitsu, K. Oshima, Org. Lett. 2010, 12, 5748–5751.
[21] A. Schꢅfer, C. Huber, R. Ahlrichs, J. Chem. Phys. 1994, 100, 5829–
5835.
[22] M. Cossi, G. Scalmani, N. Rega, V. Barone, J. Chem. Phys. 2002,
117, 43–54.
[23] J. I. Seeman, Chem. Rev. 1983, 83, 83–134.
4040; c) A. Diefenbach, F. M. Bickelhaupt, J. Phys. Chem. A 2004,
108, 8460–8466; d) A. Diefenbach, G. T. de Jong, F. M. Bickelhaupt,
J. Chem. Theory Comput. 2005, 1, 286–298; e) G. T. de Jong, F. M.
Bickelhaupt, ChemPhysChem 2007, 8, 1170–1181; f) A. P. Bento,
F. M. Bickelhaupt, J. Org. Chem. 2008, 73, 7290–7299; g) I. Fernꢆn-
dez, F. M. Bickelhaupt, F. P. Cossio, Chem. Eur. J. 2009, 15, 13022–
13032; h) W. J. van Zeist, F. M. Bickelhaupt, Org. Biomol. Chem.
2010, 8, 3118–3127.
[24] a) L. M. Wu, H. Fischer, Helv. Chim. Acta 1983, 66, 1588; b) B.
Giese, W. Damm, F. Wetterich, H. G. Zeitz, Tetrahedron Lett. 1992,
33, 1863–1866; c) B. Giese, W. Damm, F. Wetterich, H. G. Zeitz, J.
Rancourt, Y. Guindon, Tetrahedron Lett. 1993, 34, 5885–5888.
[25] M. D. Fernꢆndez-Alonso, J. Canada, J. Jimenez-Barbero, G. Cuevas,
ChemPhysChem 2005, 6, 671–680.
[26] This transition state energy was determined to be the lowest for sub-
strate 2c calculated with Me3SnH using BHandHLYP/TZVP level
of theory.
[27] Modeling was performed on methyl ester analogue of 2.
[28] All these transition states (anti- and syn-predictive TS) were consid-
ered for hydride delivery using Me3SnH for each conformation (E
and Z) of methyl ester radical intermediates.
[29] G. S. Hammond, J. Am. Chem. Soc. 1955, 77, 334–338.
[30] This angle was determined from the plane formed by C3-C1-C2 and
[38] a) C. Y. Legault, Y. Garcia, C. A. Merlic, K. N. Houk, J. Am. Chem.
Soc. 2007, 129, 12664–12665; b) D. H. Ess, K. N. Houk, J. Am.
Chem. Soc. 2008, 130, 10187–10198; c) S. I. Gorelsky, D. Lapointe,
K. Fagnou, J. Am. Chem. Soc. 2008, 130, 10848–10849; d) C. B. Shi-
nisha, R. B. Sunoj, J. Am. Chem. Soc. 2010, 132, 12319–12330.
[39] B. Guꢀrin, C. Chabot, N. Mackintosh, W. W. Ogilvie, Y. Guindon,
Can. J. Chem. 2000, 78, 852–867.
[40] a) G. E. Keck, S. Castellino, Tetrahedron Lett. 1987, 28, 281–284;
b) D. A. Evans, B. D. Allison, M. G. Yang, C. E. Masse, J. Am.
Chem. Soc. 2001, 123, 10840–10852.
[41] S. Kiyooka, M. Shiinoki, K. Nakata, F. Goto, Tetrahedron Lett. 2002,
43, 5377–5380.
[42] M. Drouin, A. G. Michel, Y. Guindon, W. Ogilvie, Acta Crystallogr.
Sect. C 1993, 49, 75–77.
[43] a) A. E. Finholt, A. C. Bond, K. E. Wilzbach, H. I. Schlesinger, J.
Am. Chem. Soc. 1947, 69, 2692–2696; b) R. H. Fish, H. G. Kuivila,
I. J. Tyminski, J. Am. Chem. Soc. 1967, 89, 5861–5868.
[44] Y. Guindon, B. Guꢀrin, J. Rancourt, C. Chabot, N. Mackintosh,
W. W. Ogilvie, Pure Appl. Chem. 1996, 68, 89–96.
ꢀ
C2 C3.
[31] The reduced products were pyramidalized to nearly 188 at C3 in the
ground state.
[32] Y. Guindon, C. Yoakim, R. Lemieux, L. Boisvert, D. Delorme, J. F.
Lavallꢀe, Tetrahedron Lett. 1990, 31, 2845–2848.
[33] The following nomenclature was used throughout the manuscript:
TS-radical intermediate-orientation. For example, C-2c-eq corre-
sponds to radical intermediate 2c in the conformation of TS C while
oriented equatorially (eq).
[45] The strain energy of the hydride (DE°
[hydride]) and the inter-
strain
action energy (DE°int) were found to be marginally different in
these two systems. See the Supporting Information for details.
[46] W. J. Hehre, L. Radom, P. v. R. Schleyer, P. A. Pople, Ab Initio Mo-
lecular Orbital Theory, Wiley, New York, 1986.
[34] J. M. Lee, P. Helquist, O. Wiest, J. Am. Chem. Soc. 2012, 134,
14973–14981.
[35] See the Supporting Information for details.
[36] D. A. Evans, V. J. Cee, S. J. Siska, J. Am. Chem. Soc. 2006, 128,
9433–9441.
[47] V. Tschinke, T. Ziegler, Theor. Chim. Acta 1991, 81, 65–78.
[48] K. B. Wiberg, Tetrahedron 1968, 24, 1083–1096.
[49] E. D. Glendening, A. E. Reed, J. E. Carpenter, F. Weinhold, Theo-
retical Chemistry Institute and Department of Chemistry, University
of Wisconsin, Madison, WI.
[37] a) F. M. Bickelhaupt, J. Comput. Chem. 1999, 20, 114–128; b) A.
Diefenbach, F. M. Bickelhaupt, J. Chem. Phys. 2001, 115, 4030–
Received: January 30, 2013
Published online: June 3, 2013
9318
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 9308 – 9318