Advanced Organic Chemistry III 1. Stereochemistry of ...
Transcript of Advanced Organic Chemistry III 1. Stereochemistry of ...
– 1 –
Advanced Organic Chemistry III — stereochemistry in organic chemistry Contents 1. Stereochemistry of Organic Compounds 2. Stereocontrol in Organic Reactions 3. Stereocontrol in Catalytic Reactions Books 1. Stereochemistry of Organic Compounds
E. L. Eliel, S. H. Wilen, Wiley, 1993. (ISBN978-0471016700)
2. Organic Stereochemistry M. J. T. Robinson, Oxford University Press, 2000. (ISBN 978-0198792758) (豊田 真司 訳 「立体化学入門」, 化学同人, 2002)
3. La Stereochimie Organique H. B. Kagan, Press Universitaires de France, 1975. (小田 順一 訳 「有機立体化学」, 化学同人, 1981. (ISBN 978-4759800821))
4. Stereoelectronic Effects A. J. Kirby, Oxford University Press, 2000. (ISBN 978-0198558934) (鈴木 啓介 訳 「立体電子効果」, 化学同人, 1999.)
5. Stereoselectivity in Organic Symthesis G. Procter, Oxford University Press, 1998. (ISBN 978-0198559573) (林 民生, 小笠原 正道 訳 「有機反応の立体選択性」, 化学同人, 2001.)
1. Stereochemistry of Organic Compounds (1) Structure and size of organic compounds
(a) Bond length bond length (Å) bond length (Å) bond length (Å) bond length (Å) C–H 1.09 C–F 1.38 C–C 1.53 C=C 1.34 C–C 1.53 C–Cl 1.77 C–C= 1.50 C=C= 1.31 C–N 1.47 C–Br 1.94 C–Cº 1.46 =C=C= 1.28 C–O 1.43 C–I 2.14 =C–C= 1.48 C–F 1.38 =C–Cº 1.43 CºC 1.20 C=O 1.22 ºC–Cº 1.38 C–C (in C6H6) 1.39
• Bond length is defined as the distance between two nuclei. • Distortion scarcely affects bond length.
(b) Bond angle • Bond angle is defined with the positions of three nuclei. • Typically, the bond angles involved with sp-, sp2-, and sp3-hybridized carbons are 180°,
120°, and 109.5°, respectively. ・ Distortion hardly restricts the bond angles. Therefore, the C–C–C bond angle in
cyclopropane is allowed to be 60°. However, the strain does not highly distort the angle of two sp3 orbitals. The C–C bond in cyclopropane is constituted with a banana shaped orbital.
(c) Dihedral angle
• Bond angle is defined with the positions of four nuclei. • Shape of molecule can be defined with bond lengths, bond angles, and dihedral angles. • In stable conformation, each dihedral angle should be 60°, 180°, or 120° (if the molecule
have sp2-hybridized atom). (see Conformation of organic molecules)
(d) Van der Waals radius Bondi, A. J. Phys. Chem. 1964, 68, 441. atom H C N O F P S Cl Br I radii (Å) 1.20 1.70 1.55 1.52 1.47 1.80 1.80 1.75 1.85 1.98
CH3 2.0 Å C6H6 (thickness) 1.7 Å
• Generally, van der Waals radius is useful for considering the bulkiness of substituent.
• However, van der Waals volume, which is calculated from the radii, does not contains the effect of bond rotation.
• Solvent accessible surface area is sometimes used for evaluating the accessibility of a reagent to a substrate.
e
Cf dC
b a
c
e
f d
b a
c
θdihedral angle
– 2 –
(e) A-value
R A-value R A-value R A-value R A-value R A-value H 0.00 CºCH 0.41 F 0.25 OMe 0.55 CHO 0.56 Me 1.74 CH=CH2 1.49 Cl 0.53 O(t-Bu) 0.75 COMe 1.02 Et 1.79 Ph 2.8 Br 0.48 OSiMe3 0.74 CO2H 1.4 i-Pr 2.21 SiH3 1.45 I 0.47 NHMe 1.29 CO2Me 1.2 Cy 2.2 SiMe3 2.5 OH 0.60 NMe2 1.53 CN 0.2 t-Bu 4.7 SnMe3 1.0 NH2 1.23 PMe2 1.5 CF3 2.4 Sn(i-Pr)3 1.10 NH3+ 1.7 PPh2 1.8
• A-value is widely used for discussing the steric effect of substituent. • In general, A-value correlates with the size of substituent. Furthermore, the value
includes the effect of bond rotation. Therefore, the value reflects the steric environment around the atom bearing the substituent.
• A-value is affected by bond length (C–R) as well as by van der Waals radius.
(f) Tips to consider steric effect of a substituent (i) Where is the most important for the related steric repulsion (near or far)?
• If you will consider the steric effect around the atom , t-Bu or 2,6-xylyl must be larger
than -CºC(t-Bu) group. • If you would like to create a steric hindrance far from , -CºC(t-Bu) should be larger
than t-Bu and 2,6-xylyl.
(ii) Snapshot or motion blur? (in the case of aryl group)
(top view) (side view) (snapshot) (motion blur)
• Snapshot may be preferable when you consider the steric hindrance in dynamic phenomena.
• Motion blur may be suitable for thermodynamic phenomena. • In snapshot steric effect, phenyl group can be regarded as a smaller substituent than
methyl one. • In motion blur steric effect, phenyl group can be regarded as a larger substituent than
secondary alkyl ones.
(g) Steric parameters other than A-value (i) Taft steric constant (Es)
Taft, R. W. J. Am. Chem. Soc. 1952, 74, 2729; Dubois, J.-E. Tetrahedron, 1978, 34, 3553.
• Taft steric constant was based on the average ratio of the rate constants between the acetate and carboxylate solvolyses or related esterifications under acidic conditions. As a premise, the acid-catalyzed hydrolysis is considered to be scarcely affected by the polar effect of the substituent.
• Similarly, steric constant had been defined with the solvolyses and esterifications
involving o-substituted benzoates.
(ii) Charton parameters (n) Charton, M. J. Am. Chem. Soc. 1969, 91, 615; ibid., 1975, 97, 1552; J. Org. Chem. 1976, 41, 2217.
• Charton parameter is the difference in van der Waals radus between a substituent and hydrogen atom. The radii of substituents are rv,min in the following figure.
(iii) Interference value (𝑰𝟑𝟒𝟎𝐗–𝐇)
Sternhell, S. J. Am. Chem. Soc. 1980, 102, 5618.
• Interference value is an index of the steric repulsion between the o-substituent and proton in biaryl compounds.
∆𝐺‡ = 𝐼()*+–, + 𝐼()*-–,
• The rotational barrier (DG‡) of 2,2’-disubstituted biaryl compounds can be predicted with sum of two interference values.
(iv) Sterimol parameters (B1, B5, L etc.) Verloop, A. In Drug Design Vol. III, Ariens, E. J., Ed. Academic Press, 1976; p 133.
• Sterimol parameters describe the dimension of substituents and are based on space-filling model, which indicates the van der Waals surface of molecule. The parameter is orginally used in a computer program, STERIMOL.
RR
Geq GaxA value = –ΔG0 = Gax – Geq (in kcal/mol)
MeMe
Me
Me
Me
Me
MeMevs vs
Me
Me
Cl
R CO2R’ + H2O R CO2H + R’OHk k: rate constant for any R
k0: rate constant for R = Me
δ : sensitivity factor to steric effectEs: steric constant
log(k/k0)δ
Es =
cat. H+
CH HH
CH
Hrv,min
rv,max l
rv,||ν = rv,min – rH
top view side view
r: van-der-Waals-based radiusl: covalent radius
X H
YH
X H
YH
ΔG‡I X–H340
I Y–H340
– 3 –
• Parameters B indicate widths of the substituent and are based on the axis of the bond between the parent skeleton and substituent. B1 and B5 are the smallest and largest width, respectively.
• Parameter L indicates length of the substituent from the parent skeleton.
(2) Chiral Compounds and Enantiomers • Chiral molecule cannot overlap with its mirror image. Therefore, chiral compounds have
no symmetry plane (s) or rotation-reflection axis (Sn). • Achiral compounds are the compounds that are not chiral. • "Chiral molecule" means the molecule possessing any chirality. • "Chiral compound" means the mass of chiral molecules. • "Racemic compound (racemate)" means the 1:1 mixture of both enantiomers (R and S).
Each molecule constituting a racemic compound should be "chiral molecule". • "Optically active compound" means the chiral compound other than racemate. A mixture
of enantiomers is often categorized as optically active compounds, even if R:S is 51:49. However, the term often indicates pure enantiomer (enantiopure compound or optically pure compound).
• "Enantiomer" originally means the mirror image of a chiral molecule (e.g. (S)-2-octanol is the enantiomer of (R)-2-octanol.). The term often indicates the enantiopure compound, whose absolute configuration has been known.
• "Diastereomer" means any stereoisomers other than enantiomer. • "Epimer" means stereoisomers bearing only one chiral center with the opposite absolute
configuration. The configuration of other chiral centers in the epimer must be identical to the original compound.
(a) Central chirality
• The chirality caused by asymmetric atom is called "central chirality". • Asymmetric atom is not limited to carbon (e.g. N, S, P, etc.). Lone pair can function as a
substituent and is regarded as atomic number 0 in CIP rule (see 3–5).
• Spiro compound 6 is chiral, although they seem to have no asymmetric atom. However, compound 7 is achiral because its tetrahydrofuran ring is symmetry plane.
• The molecules possessing symmetry plane (see 8, 9) or rotation-reflection axis (see 10)
must be achiral, even if they have asymmetric atoms (e.g. meso compound).
• Metal complexes are possible to be chiral when they are tetrahedral, trigonal bipyramidal,
or octahedral.
(b) Axial chirality • Axial chirality takes two forms: atropisomerism and isomerism in allenes. Atropisomerism • Atropisomerism is caused by inhibition of free rotation of a single bond. The inhibition of
free rotation is often observed in the biaryl compounds bearing four ortho-substituents or some sterically hindered carboxamides.
• A racemate of axially chiral compound can be resolved into each enantiomer when the
rotation barrier is over 20 kcal/mol. • Axial chirality will be stable at room temperature when the rotation barrier is over 30
kcal/mol. • Herical chirality is originated from the accumulation of axial chirality.
Isomerism in allenes • Each terminal carbon in C=C=C is plane because it is sp2-hybridized. One of the planes
is perpendicular to another because the internal carbon in C=C=C is linear and hybridized in sp manner to form two C–C double bonds. Therefore, substituted allenes are possible to be chiral as with biaryls.
• The chirality is seen in some spiro bicyclic compounds and alkylidene cycloalkanes.
MeMe
Me: van der Waals surface of CH3 (rvdw = 2.0 Å)
H
: van der Waals surface of H (rvdw = 1.2 Å)
: the atom bonding to the substituent
top view side view
LB1
B5
d
Ca
bc
d
Ca
bc =
d
Ca
cb
180°
NH2
Me CO2HH
+N
Ph MePh
N
N
1 2 3
Me PPh
OMe:
Ph SMe
O
:
4 5
O O OO =180°
O
O
6O7
Me MeOH
OHMe Me
OHOH =Me Me
HO OH
8 9O
OO
Me O
Me
10
b
aa
bb
aa
b
=180°
a
ba
b a ≠ b
OHOH
N
O
Pht-Bu
NO
Me
Me
– 4 –
(c) Planar chirality
• Planar chirality appears when a planar molecule loses its symmetry plane by bridging with short tether or forming p-complex.
• The chirality is seen in some cyclophanes, metallocenes, and trans-cycloalkenes.
(d) Notations of absolute configuration
(i) R/S notation • According to CIP rule, determine the priority of each substituent involved with the chirality. Central chirality
(in the case of spiro compounds)
Axial chirality i) Determine the priority of the two substituents on each atom involved with the chiral axis. ii) The sequence of substituents becomes a > b > c > d or c > d > a > b. Both sequences
result in the same configuration. iii) Put the substituent with the lowest priority (d or b) on the location far from you. Confirm
the direction of a → b → c (or c → d → a).
iv) To distinguish the axial chirality from others, prefix 'a' (or sufix 'a') is sometimes attached
to the chiral descriptor R or S (e.g. (aS)-2,3-pentadiene or (Sa)-2,3-pentadiene)
Planar chirality (cyclophanes, trans-cycloalkenes etc.) i) Determine ‘pilot atom’ from the bridging tether. The pilot atom is out of the plane and
closest to the plane. There are two candidates in general. The pilot atom is the candidate binding to the in-plane atom with higher priority in CIP rule.
ii) Among the atoms in the chiral plane, the atom binding the pilot atom is assigned to 'a'. The next atom is 'b', and the third atom is 'c'. If there are two candidates, the atom with higher priority is assigned to 'c' (or 'b').
iii) View the molecule from the pilot atom.
iv) To distinguish the planar chirality from others, prefix 'p' (or sufix 'p') is sometimes
attached to the chiral descriptor R or S (e.g. (pS)-(E)-cyclooctene or (Sp)-(E)-cyclooctene)
Planar chirality (metallocenes) • Three rules have been proposed to assign the stereochemical descriptor, R or S, for the
planar chiral metallocenes. One is based on central chirality and proposed by Prelog, Cahn, and Schloegl (Rule 3). Others are based on planar chirality and proposed by Ugi and Schloegl, independently (Rule 1 and 2).
• Unfortunately, the rules 1 and 2 resulted in configuration different from each other. However, rules 2 and 3 reach the same result in most cases.
Rule 1 i) View the substituted Cp ligand from the side opposite to the metal atom. ii) Confirm the direction of a → b.
Ugi, I. J. Am. Chem. Soc. 1970, 92, 5389.
Rule 2 i) Assign the metal atom to the pilot atom. ii) Regard the centroid of the substituted Cp ligand as atom 'a'. iii) The cyclopentadienyl atom bearing the substituent with the highest priority is 'b'. iv) The ortho-atom with higher priority is assigned to 'c'. v) View the Cp ring from the pilot atom, and then confirm the direction of a → b → c.
Schloegl, K. J. Organomet. Chem. 1986, 300, 219.
Rule 3 i) Treat chirality of metallocene as central chirality of the cyclopentadienyl carbon with
the highest priority. ii) Consider the metal atom to bond to the carbons in Cp ring.
•a
b
a
ba ≠ b
CO2H
H
HHO2C
HMe
HO2C
H
FeCO2Me
Br
CO2H
=
a
Cb
cd
a > b > c > d
a
b c
d
aC
a'
b b'
a > b
1 2
3 4
a
b a'
b'
d
ca
b a > b, c > d
=a
b dc
b
ac
d
Br cb a
Br: pilot atom
Feb
a Fe b
a
90°
a > b
FeBr
Mec
b Br
Me
•a
a → b → c clockwise ··· R anticlockwise ··· S
a → a' → b clockwise ··· R anticlockwise ··· S
a → b → c clockwise ··· R anticlockwise ··· S
a → b → c clockwise ··· R anticlockwise ··· S
a → b clockwise ··· R anticlockwise ··· S
a → b → c clockwise ··· R anticlockwise ··· S
– 5 –
iii) The absolute configuration can be similarly assigned to R or S with the rule for central chirality.
Schloegl, K. Forschr. Chem. Forsch. 1966, 6, 479.
(ii) P/M notation • P/M notation is based on the chirality of helix (helicity). • The descriptor P is used for right-handed helicity, and M is used for left-handed helicity
(from front to backward).
• The notation can be applied to axial and planar chiralities. • In the case of planar chirality, pR and pS configurations correspond to P and M,
respectively.
(iii) D/L notation
• D/L notation is sometimes used for the stereochemistries of carbohydrates and a-amino acids.
• Descriptors D and L strongly relate to the stereochemistries of (R)-(+)- and (S)-(–)-glyceraldehyde, respectively. D and L, must be smaller in size than other characters.
(iv) Notations based on the direction of optical rotation
• Descriptor (+) or (–) corresponds to the direction of optical rotation. (+) and (–) are used for dextrorotatory and levorotatory compounds, respectively.
• Descriptors d and l are equivalent to (+) and (–), respectively. • dl notation is not related to D/L notation.
(e) Characteristic properties of chiral compounds • Most properties of an enantiopure compounds are identical to those of its enantiomer.
Exceptionally, the direction of optical rotation is different. • However, an enantiopure compound is different in most properties from its racemate.
(i) Crystals of racemate • Three types of crystalline racemate are known, as follows: racemic conglomerate,
racemic compound, and psudoracemate. Most racemates preferentially to form racemic compounds.
Racemic conglomerate • Racemic conglomerate is a mechanical 1:1 mixture of R crystals and S crystals. Each
crystal in the mixture is homochiral (constits of a sole enantiomer). • Preferential formation of racemic conglomerate requires that the interaction between R
and R (or S and S) is stronger than that between R and S. • Racemic conglomate is much rarer than racemic compound. • Chiral compounds, which preferentially form racemic conglomerate, can be resolved into
each enantiomer through preferential crystallization without any other chiral source. Racemic compound • In racemic compound (true racemate), each crystal contains R and S molecules in 1:1
ratio. The both enantiomers form a racemic pair in a unit cell. • Most chiral compounds prefer the formation of racemic compound to that of racemic
conglomerate. In many cases, a chiral molecule has stronger affinity for its enantiomer than for itself.
• Nevertheless, homochiral crystals are preferentially obtained from the compound with relatively high enantiomeric excess
Psudoracemate • In crystals of psudoracemate, both enantiomers coexist in an unordered manner. • This type of chiral compound is very rare.
FeBr
Me
C Br
Me
Fe2
1
3
4
Fe
CBrC
C4
3Me
1
2
(P)-[6]helicene (M)-[6]helicene
c d
d
ca
ba > b, c > d
a
b
90°a
b
d cor
For axial chirality
For planar chirality
a
bdc
Br
• The a–b–c–d unit can be regarded as a helix.• View the helix from a.• Confirm the direction of the helix.
HOOH
CHO
D-glyceraldehyde
HOOH
CHO
L-glyceraldehyde
CHOOHH
CH2OH
CHOHHO
CH2OH
CO2HHH2N
CH2OHL-serine
OOH
HOHOHO
OH
CHOOHH
HO HH OH
HO HOH
=
D-glucose
1 → 2 → 3 clockwise ··· R anticlockwise ··· S
from before backward clockwise ··· P anticlockwise ··· M
– 6 –
(ii) Melting point
• Melting point of chiral compound is affected by its enantiomeric excess. Each type of crystalline racemate exhibits characteristic behavior in solid-liquid phase transition.
• In racemic compound, racemate is generally higher in melting point than its enantiopure form.
Binary phase diagrams describing the melting behavior of a) 1.2-diphenylethane-1,2-diol (racemic conglomerate); b) 3-fluoromandelic acid (racemic compound); c) 2,3-diacetoxybutane (psudoracemate).
(f) Resolution of chiral compounds (i) Use of resolving agent
• Although an enantiomer is impossible to be separated from its racemate, an epimer is physically separable from the mixture of diastereomers.
• Therefore, formation of diastereomers with an optically active resolving agent is very useful for the optical resolution of racemates.
Crystallization of diastereomeric salts • This method is useful for resolution of chiral carboxylic (phosphonic or sulfonic) acids,
amines, and phosphine oxides. • Acidic functional group (e.g. -CO2H, -SO3H) can readily form salts with basic functional
groups (e.g. -NH2). The acid–base pair is easier to crystallize from organic solvent than each compound, because the salt is generally less soluble.
• Treating racemic carboxylic acid 1 with stoichiometric enantiopure amine 2 gives the mixture of ammonium salt 3RR and 3SR. If 3RR is less soluble than 3SR, 3RR can selectively be obtained through crystallization. Purity of the crystals can be enhanced by recrystallization. Enantiopure (R)-1 will be obtained by decomposing the pure 3RR in hand with acid.
• Equimolar resolving agent (to racemate) is requires for the resolution controlled by type 1 thermodynamics. Halfmolar resolving agent may be enough for an efficient resolution, if it is controlled by type 2 or 3 thermodynamics.
• Representative resolving agents
Crystallization or chromatographic separation after derivatization • This method is useful for the resolution of chiral alcohols and ketones. It is applicable
for the resolution of chiral (secondary or primary) amines and acids. • In the resolution of a racemic alcohol 4 (e.q. R and S), the racemate is esterified with an
enantiopure acid anhydride 5 (e.g. R), which is the resolving agent. The esterification will give a mixture of diastereomeric esters 6RR and 6SR (RR and SR). Both diastereomers can be separated by column chromatography or crystallization. Each pure diastereomer in hand can be transformed to (R)- or (S)-alcohol through hydrolysis.
RRRRRR
SSSSSS
RSRSRS
SRSRSR
racemic conglomerate racemic compound
RRSRSS
SRRSSR
psudoracemate(true racemate)
147
120
100 (+) 100 (–)0% ee
(a)122
9794
100 (+) 100 (–)0% ee
(b)
43
28
100 (+) 100 (–)0% ee
(c)
mp
(°C
)
mp
(°C
)
mp
(°C
)
CO2HMeO
PhH
CO2HMeO
HPh
+ H2NH
Me
PhR
R
S
CO2–MeO
PhHR
· H2NH
Me
PhR+
CO2–MeO
HPhS
· H2NH
Me
PhR+
3RR
3SR
(R)-1
(S)-12
(crystallize)
(in solution)
H3O+CO2H
MeO
PhHR
(R)-1
R + S
S + S
R•S (less soluble)
(more soluble)
Type 1R + S
S + S
R•S (less soluble)
(soluble)
Type 2
S•S S•S
R + S
S + S
R•SType 3
S•S
N
O
O
N
H
H
HH
R
R
for acids
brucine (R = OMe)strychnine (R = H)
N
NOH
OMe
quinine
NOH
N
OMequinidine Me
NH2
Me
NH2
for bases
HO2C CO2HOH
OH
tartaric acid
HO2C CO2HOBz
OBz
HO2C CO2H
OH
malic acid
CO2H
OH
mandelic acid
O
Me Me
HO3S
camphorsulfonic acid
OP
O OOH
MeOH
PhHO
H
H
O
OMe
OHHPh
R
S
RR+
(R)-4
(S)-4HO2C
O
O
Ph
MeH
5HO2C
O
O
Ph
HMe
R S
6RR 6SR
(separable with chromatography or crystallization)
hydrolysis
(R)-4 (S)-4
– 7 –
• A racemic chiral ketone or aldehyde can be resolved through the reaction with an enantiopure primary amine, which gives a mixture of diasteromeric imines.
• Representative resolving agents
(ii) Chiral HPLC
• A racemate is resolved to each enantiomer through the column chromatography with a chiral stationary phase, which is typically composed of a chiral compound and silica gel.
• Various preparative chiral HPLC columns (20 mm f) are commercially available.
(iii) Preferential crystallization • Racemic conglomerate can be resolved through recrystallization without resolving agent,
if you have its homochiral single crystal. • To a saturated hot solution of the racemate, its homochiral crystal (e.g. R) is installed.
The same enantiomer (R) will be crystallized in preference to the antipode (S). The resulting mother liquor will be S-enriched. The recrystallization of the mother liquor will preferentially form the crystals of S-compound.
(iv) Kinetic resolution • In general, chiral reagents and catalysts exhibit different reaction rate for each
enantiomer of substrates. If the reaction of the R-enantiomer is faster than that of S-isomer, the recovered substrate should be S-enriched. Therefore, the reaction of a racemate with a chiral reagent (or through asymmetric catalysis) is usable for the resolution.
• The efficiency of kinetic resolution (s) is expressed by the ratio of the reaction rate of
each enantiomer (kR/kS). s = kR/kS = ln[(1–C)(1-ee)]/ln[(1–C)(1+ee)] = ln[1–C(1+ee’)]/ ln[1–C(1–ee’)]
C: conversion ee = enatiomeric excess of unreacted substrate ee’ = enatiomeric excess of product
• Examples
Marshall, J. A. Org. Synth. 2005, 82, 43.
Sharpless, K. B. J. Am. Chem. Soc. 1981, 103, 6237.
(g) Assignment of absolute configuration (i) Chemical transformation
• In this method, stereochemically unknown samples are transformed into suitable compounds for assigning the absolute configuration. The products are analyzed with X-ray diffraction, optical rotation or chiral HPLC analysis.
a) The target molecule is reacted with a known optically active compound (e.g. through ester condensation) to be transformed into a crystalline compound. Its crystal is subjected to X-ray structure analysis. The stereochemistry of the target can be assigned with the crystal structure by using the absolute configuration of the installed compound.
b) The target molecule is transformed to a known optically active compound, although
possibility of racemization must be considered during the transformation. The specific rotation of the resulting product is compared with the reported [a]D. Alternatively, comparison of the retention times in the chiral HPLC analyses is also useful for the assignment of absolute configuration.
c) The authentic sample of the target molecule is synthesized from a known optically active
compound. The specific rotation or the chiral HPLC retention time of the authentic sample allow us to assign the absolute configuration.
for alcohols
O
O
O
H
H
O
O
OH
H
H
Me
Me Me
NCO
(for urea derivatization)
(for lactor derivatization)
(for ester derivatization)
O
O
CO2HMe Me
Me OH
O
O
Me
MeMe Me Me Me
H
for ketones and aldehydes
Noe reagent
EtO2C CO2EtOH
OH
diethyl tartarate
NNH2
OMe
SAMPMeHN NHMe
Ph Ph
kR R&S
(50:50)
+ aR + a R a
kSS + a S a
(fast)
(slow)
if kR >> kS R a&S (almost enantiopure)
Me
OH
TMS
Amano Lipase AK
OAc , pentane Me
OH
TMS49%
Me
OAc
TMS47%
C6H13
OH Ti(O-i-Pr)4 & (+)-DIPT
t-BuOOH C6H13
OH
C6H13
OH
Os = 83
R1 R2
OH
(target)
Ph COCl
OMe
S
Ph O
O
R1R1MeO
SX-ray crystal structure
a
b Compare the stereochemistry of chiral center a with that of chiral center b.
target other compound(Its [α]D has been reported.)
Measure its optical rotation and caluculate [α]D.
Compare it with the [α]D reported in the literature.
other compound(Its [α]D has been reported.)
authenticsample optical rotation.
retention time in chiral HPLC.Measure its
Compare the value with that of the target.
– 8 –
(ii) Crystallography (Bijvoet method) • Commonly, X-ray single crystal analysis gives no information about the absolute
configuration of the chiral compound. However, the configuration can be determined with the crystallography, when the compound has one heavy-weight atom.
• The heavy-weight atom induces anomalous X-ray scattering, which caused that intensity F(h k l) is not equal to that of F(–h –k –l) in a chiral crystal.
• Flack parameter (x) is widely used for estimating the absolute configuration of crystal structure.
I(h k l) = (1–x)|F(h k l)|2 + x|F(–h –k –l)|2 x: Flack parameter I: the square of the scaled observed structure factor F: the calculated structure factor
If x is almost 0, the crystal structure has the correct absolute configuration. If the crystal structure has the inverted configuration, x should be almost 1.
• Bijvoet method is applicable for chiral compounds bearing no heavy-weight atom. In this case, a protective group containing a heavy-weight atom (e.g. p-bromobenzoyl) is installed to the target molecule.
(iii) Advanced Mosher method and its related methods see: Riguera, R. Chem. Rev. 2004, 104, 17.
Advanced Mosher method • Advanced Mosher method is empirical, but very useful for determining the absolute
configuration of an unknown chiral secondary alcohol. This method is applicable for the chiral secondary alkyl primary amines.
(original) Mosher, H. S. J. Am. Chem. Soc. 1973, 95, 512.
(alcohol) Kusumi, T.; Kakisawa, H. J. Am. Chem. Soc. 1991, 113, 4092. (amine) Kusumi, T.; Kakisawa, H. Tetrahedron Lett. 1991, 32, 2939.
• Prepare two samples of the secondary alcohol. Each Sample is esterified with (R)- or (S)-MTPA to prepare (R)- and (S)-MTPA esters (MTPA is 3,3,3-trifluoro-2-methoxy-2-phenylpropionic acid). For each proton in the secondary alcohol moiety, calculate the difference between the 1H NMR chemical shifts of (S)- and (R)-MTPA esters (Dd = dS – dR) as shown in above figure.
• In MTPA esters, their CF3 groups are located at the pseudo eclipse of the proton attaching to the chiral carbon of the secondary alcohol. The aromatic ring in MTPA induces upfield shift of the resonances of the protons in the secondary alcohol moiety.
Therefore, if Dd is negative, the proton should be located over the aromatic ring of (S)-MTPA.
Trost method • Trost method uses O-methylmandelic acid instead of MTPA. The procedure is very
similar to advanced Mosher method. • O-Methylmandelic acid is easier to react with secondary alcohols than MTPA, but may
be racemized during the esterification. • In the O-methylmandelates, their OMe groups are located at the pseudo eclipse of the
proton attaching to the chiral carbon of the secondary alcohol.
Trost, B. M. J. Org. Chem. 1986, 51, 2370.
PGME method • PGME (phenylglycine methyl ester) method is useful for determining the absolute
configuration of the chiral a-carbon of carboxylic acid. • In the PGME amides, their CO2Me groups are located at the pseudo eclipse of the a-
proton of the chiral carboxylic acid.
Kusumi, T. Tetrahedron Lett. 1995, 36, 1853; J. Org. Chem. 2000, 65, 397.
(iv) Circular dichroism (CD) spectrum • Circular dichroism is active for only optically active compounds. • Positive Cotton effect means that an enantiomer gives a positive peak in its CD spectrum. Octant rule • This method is empirical and useful for
determining the absolute configuration of cyclic ketones.
• First, consider the most stable conformation for the target molecule. Geometry optimization with MO or MM may be useful for the consideration.
• Space around the carbonyl group is divided into eight sectors as shown in figure (a).
R1
H
O
O
OMe
CF3
R2
S
(S)-MTPA ester
upfield shift
R1
H
O
O
MeO
CF3
R2
R
(R)-MTPA ester
upfield shift
Δδ = δS – δR
If Δδ > 0, the proton belongs to R2.If Δδ < 0, the proton belongs to R1.
R1
H
O
O
H
OMeR2
R
(R)-O-methylmandelate
upfield shift
R1
H
O
O
H
OMeR2
S
upfield shift
Δδ = δS – δR
If Δδ > 0, the proton belongs to R1.If Δδ < 0, the proton belongs to R2.
(S)-O-methylmandelate
R1
H
R2
(S)-PGME amide
upfield shift
R1
HNH
H
CO2MeR2
R
upfield shift
NH
H
CO2Me
O O
S
(R)-PGME amide
Δδ = δS – δR
If Δδ > 0, the proton belongs to R2.If Δδ < 0, the proton belongs to R1.
Octant rule for standard ketones. (a) Signs of the sectors in a left-handed Cartesian coordinate system; (b) projection of the rear sectors (z < 0).
O
x
z
y
O
x
y
(a) (b)
– 9 –
• In octant rule, the atoms in the + sectors induce a positive Cotton effect. The atoms in the – sectors induce a negative Cotton effect.
• The Cotton effect can be predicted as above. Compare the prediction with the observed CD spectrum (n→p* (C=O), ca. 300 nm).
Exciton chirality method • This method is theoretical and useful for determining the absolute configuration of
cycloalkanediol or diamine. Two equivalent chromophores (e.g. p-Me2NC6H4CO2-) in a molecule are required for the assignment.
• The two chromophores cause the split of their CD peak. One is negative and the other is positive.
• When the longer wavelength band of the split peak describes a positive Cotton effect and the shorter one describes a negative Cotton effect, the two chromophores arranged in a right-handed helix.
(h) How to analyze enantiomeric excess
• Percentage of enantiomeric excess (% ee) is commonly used for representing the ratio of enantiomers. (% ee (R-enriched) = ([R] – [S])/([R] + [S]) × 100)
(i) Specific rotation • Specific rotation is usable for measuring enantiomeric excess, if specific rotation of
enantiopure compound has been reported. % ee = [a]D (sample)/[a]D (100% ee) × 100
• However, the specific rotation is significantly affected by inpurity, solvent, and temperature. Sometimes, concentration of the sample affects the specific rotation.
(ii) Chiral HPLC and GC • Chiral HPLC or GC analysis is widely used for measuring enantiomeric excess. • Chiral HPLC is effective for relatively polar compounds (alcohol, ketone, ester, amide
etc.) with a UV active moiety (p-bond). • Chiral GC is usable for less polar compounds (hydrocarbon, ether etc.) as well as above
compounds. UV active moiety is not required, but the sample must be volatile.
(iii) Chiral shift reagent • Prepare NMR sample of the target compound in CDCl3 or C6D6 and measure 1H NMR.
Add a small amount of a shift reagent to the NMR sample, and then measure 1H NMR again. Repeat the process several times until a proton signal completely splits. The enantiomeric excess of the sample is calculated from the ratio of the integrals of each split peak.
• Representative chiral shift reagents
(3) Diastereomers (a) Notations of diastereomeric stereochemistry
(i) E/Z notation for alkenes • As with R/S notation, E/Z notation unambiguously defines the geometrical configuration
of alkenes. The notation is based on CIP rule. i) Determine the priority of the two substituents on each atom involved with the C–C
double bond. ii) Confirm the positional relationship of each substituent with higher priority.
(ii) cis/trans notation
• Cis/Trans notation is ambiguous, but often used for indicating the geometrical configuration of alkenes or the relative configuration of cyclic skeletons including two chiral centers.
• Priority of each substituent is determined with common sense, which causes the ambiguity (CIP priority or main chain in IUPAC nomenclature?).
• In the case of alkene geometry, cis and trans configurations correspond to Z and E, respectively.
(iii) erythro/threo notation for relative configuration
• Erythro/threo notation is ambiguous, but often used for indicating the relative configuration of acyclic vicinal chiral centers. Erythro and threo forms are strongly related to the stereochemistry of erythrose and threose.
• If the substituents on each chiral carbon are arranged in a similar manner, the configuration is threo. Otherwise, the configuration is erythro.
OH
NMe2
OO
Ar
O
O
ArOHOH
Ar
O
Cl==
HO
HO
NMe2Δε
λ
O
OM
Rf
Me
Me
Me 3
Rf = CF3 (tfc) or C3F7 (hfc)M = Yb or Eu
NO2
HNO
OHN
N
N
O
O
HN
HN
O
OM(tfc)3 or M(hfc)3
a
b
c
d
a > b, c > d
a
b
c
d
a
b
c
da > b, c > d
a
b
c
d
a
d
b
c
a
d
b
c
positional relationship between a and c the same side ··· Z opposite side ··· E
positional relationship between a and c the same side ··· cis opposite side ··· trans
– 10 –
(iv) syn/anti notation for relative configuration
• Syn/anti notation is less ambiguous than erythro/threo notation and applicable to various acyclic compounds.
• Draw the structure of the target compound with zig-zag projection. If two substituents are arranged on the same face, the configuration is syn. Otherwise, the configuration is anti.
(v) R*/S* (or RS/SR) notation for relative configuration
• R/S notation is applicable in relative configuration. This notation is unambiguous. • Assign the absolute configurations of all chiral center in an enantiomer. Attach an
asterisk to each R or S descriptor. • (2R*, 3S*) is equivalent to (2S*, 3R*).
(vi) l/u notation for relative configuration
• Determine the absolute configurations (R/S or P/M) of all chiral centers, and then arrange them according to the numbering rule in IUPAC nomenculture.
• If the absolute configurations of two neighboring chiral centers are identical or similar each other, the relative configuration is l (like). Otherwise, u (unlike) must be used as the stereochemical descriptor. For example, uu-1,2,3,5-hexanetetraol is equivalent to (2R*,3S*,5R*)-1,2,3,5-hexanetetraol.
(b) How to deal with diastereoisomeric substituents in CIP rule
In CIP rule ① Atomic number larger → higher priority ② Isotope atomic mass: larger → higher priority ③ Alkene stereochemistry priority: Z > E (cis > trans)
In most cases, the priority of two geometrically isomeric groups can be determined by Z/E system. However, the system is useless to 4-substituted alkylidenecyclohexanes, although they can be regarded as axially chiral molecules. In this case, the alkene configuration is determined from the substituent bearing the chiral center.
④ Diastereomeric groups priority: l > u (⑤ Enantiomeric groups R > S or M > P?)
(c) Assignment of diastereomer
(i) Crystallography • If the sample is crystalline solid, its X-ray crystal structure analysis is the most reliable.
(ii) NMR a) 1,2-Disubstituted alkene
nOe (nuclear Overhauser effect) difference spectrum • Positive nOe will be observed between two substituents, which are located in the cis
position each other. • NOESY should be avoid to use for the purpose.
Vicinal H–H coupling (3JH–H) • In general, the coupling constant between the two alkenyl the alkenyl protons is 10–13
Hz when the protons are arranged with cis configuration. In the trans isomer, the coupling constant is 15–17 Hz.
• The above rule is useful only when a carbon atom attaches to each alkenyl carbon. Heteroatom substituents causes decrease in the coupling constant.
b) Saturated four- and five-membered ring
NOESY or nOe difference spectrum • Use only transannular nOe. Sometimes, nOe is observed between vicinal hydrogens
with trans configurations, because these protons couple with each other.
CHOOH
OHD-erythrose HO
CHOOH
OHD-threose HO
==
==
CH2OH
H OH
OHC
H OH
CH2OH
H OH
OHC
HO H
b'
c' a'
b
c a
b'
c' a'
b
a c
2
1
33
1
2
2
1
33
1
2
Me
OH OH
OH2 4
5Me 2,4-syn, 2.5-anti
Me OHOH
OHOH(2R*,3S*,5R*)-1,2,3,5-hexanetetraol
= (2S*,3R*,5S*)-1,2,3,5-hexanetetraol
O OHOH
HOHOHO
CHO
OH
OH
OH
OHHO RR
R S
D-glucose uul-2,3,4,5,6-pentahydroxyhexanal
CO2H
H Me
H Me
HO2C
CO2Hcistrans
12
34S
H H H CHR2 R2HC CHR2 H OCH3
HOCH3
O?
CH3
OCH3O
?
H H
C C3JH–H = 10–13 Hz
H C
C H3JH–H = 15–17 Hz
a → b → c vs a' → b' → c' the same direction ··· erythro opposite direction ··· threo
– 11 –
c) Saturated six-membered ring
Chemical shift • In general, an axial proton appears in higher field (ca. 0.5 ppm) than its corresponding
equatorial proton.
Vicinal H–H coupling • Karplus equation: 3JH–H = A cos2q + B cos q + C (q: dihedral angle)
(e.g. A = 9.4, B = –1.4, C = 0.4 (for hydrocarbons)) • 3JHax–Hax is 8–14 Hz, because dihedral angle q for Hax–C–C–Hax is 180°. 3JHax–Heq and
3JHeq–Heq is 2–4 Hz, because the related dihedral angles are 60°. • Higher magnetic field NMR spectrometer should be used for assigning the relative
configuration of cyclohexane derivative. Assign as many peaks as possible by using not only 1D-NMR but also H–H cosy, HMBC. HSQC etc.
• Assign the stereochemistry by using each coupling constant in a similar way to solving a puzzle.
d) Acyclic compound see: Bifulco, G.; Riccio, R. Chem. Rev. 2007, 107, 3744. Universal methods • JBCA (J-based configuration analysis) see, Murata, M. J. Org. Chem. 1999, 64, 866. • UDB (Universal NMR Database) see, Kishi, Y. Org. Lett. 1999, 1, 2177 and 2188. • Use of quantum mechanical calculation
1,2- or 1,3-diols (diamine) • Protect the diol with acetone (or 2,2-dimethoxy propane, phosgene etc.) to give the
corresponding acetonide. Analyze the cyclic acetal with NOESY, nOe difference spectrum, H–H coupling constants etc.
1,3-diols • Acetonides of syn-1,3-diols have chair conformation, while those of anti-1,3-diols prefer
twist boat conformation. Therefore, the relative configuration of 1,3-diols can be elucidated with the 13C NMR resonances of the methyl groups of their acetonides.
Evans, D. A. Tetrahedron Lett. 1990, 31, 7099.
(4) Conformation of organic compounds (a) Acyclic saturated compounds
(i) Ethane • The staggered conformer (q = 60° or 180°) is more stable than the
eclipsed one (q = 0° or 120°) (DE‡ = 2.9 kcal /mol). • The eclipsed conformer is a transition state and not energy minimum. • The sC–H–s*C–H interaction stabilizes the staggered conformer. • The steric repulsion between the vicinal protons scarcely contributes to
rotational barrier. (ii) Butane
• Butane has two energy minima: one is anti, and the other is gauche. • Anti conformer is more stable than gauche one (DE = 0.9 kcal /mol). • Each conformer, including unstable conformers, can be named with the dihedral angle q
as shown below.
DFT-calculated energy diagram of the rotational isomerisms in ethane, propane, butane, and 1,2-difluoroethane at M06-2X/6-311++G(d,p) level. The diffuse functions are required for considering s-s* interaction.
(iii) Pentane • The free rotations of C2–C3 and C3–C4 should be considered for the discussion of the
conformation of pentane. • Conformer aa is the most stable conformer of pentane. • Conformers g+g– and g–g+ are the least stable. The energies of these conformers have
been estimated to lie over 3.3 kcal/mol above that of conformer aa. • The steric repulsion is called ‘syn-pentane interaction.’
abd
c
• The nOe between a and d (red) is reliable for determing the relative configulation.• The nOe between a and b (blue) should be avoid to use for the purpose, because an nOe is sometimes observed between a and c.
R1 R2
OH OH
R1 R2
O O
Me Me
R1 R2
OH OH
R1 R2
O O
Me Me
==
==
OO
R1
R2 Me
Me
OO
R1
R2
Me
MeOOR2
R1
H
H
MeMe
δ 24.6 ± 0.76
δ 19.4 ± 0.21
δ 30.0 ± 0.15
δ 98.1 ± 0.83
δ 100.6 ± 0.25syn-1,3-diol
anti-1,3-diol
Ra
bb
ccd
RC A
B
a: synperiplanar (ecripsed, –30° < θ < 30°)b: synclinal (gauche, 30° < θ < 90° (±))c: anticlinal (90° < θ < 150° (±))d: antiperiplanar (anti, 150° < θ < 180° (±))
θ
-1.0
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
-180 -120 -60 0 60 120 180
ethane (R,R'=H) propane (R=Me, R '=H) butane (R,R'=Me) 1,2-d ifluoroethane (R,R'=F)
HH
R’
H
R
H
θ
antiecripsed gauchegauche
dihehral angle (deg)
ΔE
(kca
l/mol
)
The s–s* interation between the vicinal C–H bonds.
H
H
σ*σ
– 12 –
(left) DFT-calculated contour diagram of the rotational isomerism in pentane at M06-2X/6-311G(d,p) level. (right) The most and least stable structures of pentane.
(iv) Effect of s–s* or n–s* interaction • Conformation involved with C(sp3)–C(sp3) bond is affected by the hyperconjugation
between vicinal C–X bonds. In the hyperconjugation, one of the s orbital of C–X bond donates electrons to s*-orbital of another C–X.
• Electron donating ability: nN > nO > sC–C, sC–H > sC–X (X: N > O > S > Hal) • Electron accepting ability: p*C=O > s*C–Hal > s*C–O > s*C–N > s*C–C, s*C–H • Gauche effect: In some compounds, gauche conformer is preferable to anti conformer.
• Anomeric effect: The ratio of a- and b-glucose is 38:62 in water. The unusual ratio (A
value of OH is 0.60) is caused by anomeric effect, the interaction between nO and s*C–O orbital.
(b) Acyclic unsaturated compounds
(i) X–CH2–CH=Y (Y = CH2, O) • Energy minima appear when the C(sp3)–X (or H) bond overlaps to the C=Y bond in the
Newman projection. Another conformer, in which the C–X bond overlaps to the C(sp2)–H bond, is a torsional barrier.
• Steric repulsion between X and Y is often considerable in the eclipsed conformer (1,3-allylic strain). Therefore, the gauche conformer is generally preferable to the eclipsed one.
• The gauche or eclipsed conformation is stable in substituted benzenes.
Allylic strain • 1,2-Allylic strain is caused by the steric repulsion between the allylic substituent R and
the a substituent R’. The strain is unfavorable for the gauche conformer. • 1,3-Allylic strain is caused by the steric repulsion between R and the cis-b substituent R”.
The strain • Ordinarily, 1,3-allylic strain much affects the conformation of olefinic compound.
DFT-calculated energy diagram of the rotational isomerisms in propene, 1-butene, 2-methyl-1-butene, (Z)-pentene, and (Z)-2-methyl-2-pentene at B3LYP/6-311++G(d,p) level. M06-2X may underestimate the 1,2-allylic strain.
(ii) X=CH–CH=Y • s-Trans conformer (q = 180°) is the most stable conformer in the conjugated unsaturated
bond system, when there is no significant steric repulsion in the molecule. • s-Cis conformer (q = 0°) is less stable than the s-trans one, although it is mostly another
stable conformer when X and/or Y is O or N (e.g. a,b-unsaturated carbonyl compounds). • In 1,3-butadiene (X, Y = CH2), the gauche conformer is more stable than to the s-cis one
because of the steric repulsion between cis-hydrogen atoms on 1- and 4-positions. The s-cis conformer is the transition state of the bond rotation between the two gauche conformers.
• In highly substituted 1,3-butadienes, even s-trans structure is often distorted from the planar structure.
0 60 120 180 –120 –60 0
60
120
180
–120
–60
0 (kcal/mol)
θ(C3–C4)
θ(C
2–C
3)
H 4
H
1
2H H
5
H 5
H
2
3H H
1
θ(C2–C3)
θ(C3–C4)
aa
g+g–
g–g+
g+
g+
g–
g–
H
HMe
Me
HH
Me
MeH
H
HH
conformer aa(most stable)
conformer g+g– or g–g+(most unstable)
syn-pentaneinteraction
(a) (b)
CH2FCH2F:
F
FHH
HH F
HHH
HF
ΔE = – 2.0–2.6 kcal/mol
OHOHO
HOOH
HO OHOHO
HOOH
HO
α-glucose β-glucoseOH
OnO
σ*C–O
X
HH
Y
H
H
XH
Y
HH
XH
Y
Hbisecting
(rotational barrier)gauche
(a stable confomer)eclipsed
(less stable than gauche)
X
H
H Y
HFelkin–Anh-type
(not stable)
R”R
HHH
R”H
HHR
1,3-allylic strain (A(1,3)-strain)
R
HH
H
RH
R’R’1,2-allylic strain (A(1,2)-strain)
1
2(α) 1(β)3
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
-180 -120 -60 0 60 120 180
propene (R, R ', R" =H) 1-butene (R=Me, R', R"=H) 2-methyl-1-butene (R,R'=Me, R"=H)(Z)-2-pentene (R, R"= Me, R '=H) (Z)-2-methyl -2-pentene (R, R', R"=Me)
ΔE
(kca
l/mol
)
dihehral angle (deg)
R’
RH
H
R” H
θ
A(1,2)
A(1,3)
ecripsed gauchegauche bisecting bisecting
– 13 –
DFT-calculated energy diagram of the rotational isomerisms in 1,3-butadiene, 2,3-dimethyl-1,3-butadiene, acrolein, methyl vinyl ketone, and 3-methyl-3-propen-2-one at M06-2X/6-311++G(d,p) level.
(iii) Biaryls • In the DFT calculation of biphenyl (non-substituted), the dihedral angle (q) between its
phenyl rings is 41° in the optimized structure, although the coplanar conformation was observed in the crystal structure analysis.
• The p-conjugation between two rings may less affects the conformation than the steric repulsion between ortho-protons.
• The ortho-substituent leads to the increase in the torsion angle. In 2-(tert-butyl)biphenyl, one aryl ring is almost perpendicular to another in the stable conformation.
DFT-calculated energy diagram of the rotational isomerisms in ortho-substituted biaryls at M06-2X/6-311++G(d,p) level.
(iv) Carboxamides • C–N bond of carboxamide possesses double bond character, because structure 2
remarkably contributes to the resonance hybrid. Therefore, the rotation of C–N bond is relatively slow.
• In NMR analysis, a set of signals are often observed because the Z–E isomerization of amide is slow in NMR time scale.
• The E–Z transformation of amide proceeds through the transition state involving sp3-hybridized nitrogen atom.
(c) 4-Membered rings
• In 4-membered rings, planar structure is unfavorable because torsional strain is maximum. • Puckered structure is preferable to planar one in order to escape the strain. The 4-
membered ring is released from the ecripsed conformation. • The angle of pucker is 28° and the barrier of ring inversion is 1.45 kcal/mol in cyclobutane.
(d) 5-Membered rings
• Conformation analysis of 5-membered rings is significantly complicated as compared to those of 4- and 6-membered rings, because their conformation is very flexible.
• Stable conformations of cyclopentane are envelope and half-chair conformations. The former is more stable than latter, but the energy difference of both conformers is very small (0.5 kcal/mol).
• The envelope conformer is Cs-symmetry, and the half-chair conformer is C2-symmetry. • Equilibrium between envelope and half-chair conformers is remarkably affected by
substituents.
Y
H
X
H
X YH
Y
X
H
XY
s-ciss-trans
Y
H
X
H
X Y
gauche
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
-180 -120 -60 0 60 120 180
1,3-butadiene (1) 2,3-d imethyl -1,3-butadiene (2) acro lein (3)methyl vinyl ketone (4) 3-methyl-3-propen-2-one (5)
X
Y
R’
R
θ
s-transs-cisgauche
ΔE
(kca
l/mol
)
dihehral angle (deg)
H H Me Me1 2
O
H H3
O
Me H4
O
Me Me5
-1.0
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
-180 -120 -60 0 60 120 180
biphenyl 2-fluoro-1,1 '-biphenyl 2-methoxy-1,1'-b iphenyl2-ch loro-1,1'-biphenyl 2-methyl-1,1'-b iphenyl 2-tert-butyl -1,1'-biphenyl
coplanar coplanar
dihehral angle (deg)
ΔE
(kca
l/mol
)
θ
R
O
R1N
R2
R3 –O
R1N+
R2
R3
1 2
ON
H
Me
HΔE‡ = 20.6 kcal/mol
ON
Me
Me
HΔE‡ = 21.3 kcal/mol
5
6
7
84
3
1
2
2
1
4
37
8
6
5
ΔE‡ = 1.45 kcal/mol
envelope
σ
C2
half-chair
– 14 –
(e) 6-Membered rings (i) Saturated rings
• Stable conformations of saturated 6-membered rings are chair and twist boat. Boat configuration is the transition state of the interconversion between two twist boat conformers.
• In most cases, chair conformer is more stable than twist boat conformer. Rarely, twist boat is preferable to chair in order to avoid large 1,3-diaxial interaction.
• Stability of conformer of substituted 6-membered rings is affected by A-value, 1,3-diaxial
interaction, and the presence of gauche conformation etc.
Fused ring systems • Bicyclic system of cis-fused cyclohexane, such as cis-decaline, is possible to invert. In
contrast, ring inversion of the trans-isomer is impossible.
(ii) Cyclohexene
• The most stable conformation of cyclohexene is half-chair.
• In 1,6-disubstituted cyclohexenes, substituent R' at the 6-position induces considerable
A(1,2)-strain if it occupies the pseudo-equatorial position. To avoid the strain, R' tends to occupy the psudo-axial position.
(iii) Cyclohexanone
• Cyclohexanones prefer chair conformation as with cyclohexanes. • The conformation must be considered when we discuss the dipole of cyclohexanones
and the reactivity of their carbonyl groups.
• •‡‡
chair half-chair twist boat boatR
R
axialequatorial
cis-decaline trans-decaline
==
half-chair
pseudo-axialpseudo-equatorial
R'R'
H HRR A(1,2)-strain
more stable
R R 1,3-diaxial interaction
– 15 –
2. Stereocontrol in Organic Reactions 2-1. General (1) Principles in selective organic reactions
(a) Curtin–Hammett principle Selectivities are controlled by the relative energies of two transition states. The difference
of reactants or products does not affect the selectivity. • Curtin-Hammett principle is useful for considering selectivities of organic reactions, when
the reaction forms different products (C and D) from two substrates (or intermediates) (A and B) in rapid equilibrium with one another (eq. 1).
• The selectivity, [C]/[D], is equal to exp(–DDG‡/RT). Therefore, the selectivity depends on only DDG‡, although DDG‡ = DG2‡ – (DG1‡ + DG).
• Even if the reactant (or intermediate) A is less stable than B, product C would be the major product when TS1 is more stable than TS2.
• Avoid to discuss the selectivity with the structures of reactants (or intermediates).
Seeman, J. I. Chem. Rev., 1983, 83, 83.
(b) Hammond’s postulate If two states, as for example, a transition state and an unstable intermediate, occur
consecutively during a reaction process and have nearly the same energy content, their interconversion will involve only a small reorganization of the molecular structure.
The postulate is useful for predicting the property of the transition state in highly exothermic or endothermic process (Figs a and b). Do not apply the postulate to the reactions, in which the reactant is similar in potential energy to the product (Fig c). (i) Highly exothermic reaction
• The energy diagram of a highly exothermic step (Fig a) indicates that the reactant reaches the transition state at early stage. Therefore, the transition state is similar in structure and energy to the reactant.
・ Selectivities, including stereoselectivity, would be affected by the relative energy of each reactant (or its conformer).
(ii) Highly endothermic reaction • The energy diagram of a highly endothermic step (Fig b) indicates that the reactant
reaches the transition state at late stage. Therefore, the transition state is similar in structure and energy to the product.
・ Selectivities, including stereoselectivity, would be affected by the relative energy of each possible product.
Hammond, G. S. J. Am. Chem. Soc., 1955, 77, 334.
See also, Leffler, J. E. Science, 1953, 117, 340.
(2) Stereoselective reaction and prochirality (a) Enantioselective reaction
(i) General • Enantioselective reaction is the reaction that produces an optically active compound from
an achiral or racemic starting material with an optically active reagent or catalyst. • It is ordinary that an achiral compound is converted into a 1:1 mixture of R- and S-
products, if the reaction creates a new chiral center. Both pathways leading to R- and S-products proceed through enantiomeric transition states, which have the same energy level each other.
• The optically active reagent or catalyst can discriminate the prochirality of the substrate.
• Commonly, the stereoselectivity in the enantioselective reaction, which is called
enantioselectivity, is equivalent to the enantiomeric excess of the optically active product. • There are two types of stereodiscrimination in enantioselective reaction. One involves
the discrimination of enantiotopic faces; another involves that of enantiotopic groups.
A BC Dk1 K k2
(1)
AB
reactantsproduct
product
CD
ΔG
ΔG1‡ΔG2‡
ΔΔG‡TS1
TS2
G
Ereactant
product
(a) highly exothermic reaction
E
reactant
product
(b) highly endothermic reaction
TS‡ TS‡
similar similarE
reactant product
(c) other
TS‡
O
HPh+ MeMgBr
O
H PhMe– a Me–b
OMgBr
Me PhH
OMgBrHPh Meor
1 3a2
R S
3b
TSa
1 + 2 3a
E
1 + 2 3b
E
TSbO
PhH Me
δ–δ–O
PhHMe
δ–δ–
– 16 –
(ii) Prochirality of achiral compounds Enantiotopic face (re/si) • In general, achiral unsaturated molecules have a symmetry plane involving the double
bond. The plane has two faces. If the attack of a reagent on each face leads to forming different enantiomer, the face is called ‘enantiotopic face.’
・ Prochirality of enantiotopic face can be indicated with descriptor re or si. The prochirality is assigned as follows: i) Determine the priority of the three substitutents on the atom, which will become a chiral
center after the enantioselective reaction. ii) View the face from above. The face is called re-face when the direction of a → b →
c is clockwise. Otherwise, the face is si-face.
Enantiotopic group (or atom) (pro-R/pro-S) • When two equivalent groups (or atoms) are attached on a sp3-hybridized carbon, the
reaction occurring on one of the groups results in forming a new chiral center. Different enantiomers will be obtained from the reactions on each group. The two equivalent groups are called 'enantiotopic group'.
• Prochirality of enantiotopic group can be indicated with descriptor pro-R or pro-S. The prochirality is assigned as follows: i) Determine the priority of the four substituents on the atom, which will become a chiral
center after the enantioselective reaction. Here, one of the enantiotopic groups is tentatively assume to be dominant over another.
ii) Determine the R/S configuration with the above tentative priority. When the configuration is R, the prochirality of the substituent with higher priority is assigned to pro-R. Otherwise, the prochirality is pro-S.
(iii) Discrimination of enantiotopic faces
• In most enantioselective reactions, a chiral reagent or catalyst discriminates two enantiotopic faces of the unsaturated bond in the substrate (or intermediate) to give the enantio-enriched product (See the figure in the left column).
• After the stereoselective reaction, no unreacted unsaturated bond remains in the resulting product. Therefore, the enantioselectivity of the reaction reflects the selection of the enantiotopic faces.
Examples
(iv) Differentiation of enantiotopic groups (or atoms)
• In some enantioselective reactions, a chiral reagent or catalyst discriminates two enantiotopic groups of a substrate to give an enantio-enriched product.
• After forming the chiral product, one of the enantiotopic groups remains in the product. The desired product will further react with the reagent, if the chiral reagent or catalyst exhibits low stereoselectivity. However, the overreaction may lead to increasing in the enantiomeric excess of the product. The enantiomeric excess of the product does not reflect the selection of the enantiotopic groups.
Sih, C. J. J. Am. Chem. Soc. 1984, 106, 3695.
See also, Schreiber, S. L. J. Am. Chem. Soc., 1987, 109, 1525.
(b) Diastereoselective reaction (i) General
• Diastereoselective reaction produces a new chiral center on the substrate, which has a chiral center. The diastereoselectivity reflects the ratio of the diastereoisomeric products.
• The stereogenic center in the substrate can control the stereochemistry of the reaction (i. e. diastereoselectivity). Therefore, an achiral reagent can selectively produce a single diastereomer in diastereoselective reaction.
bc
a
a
b
c'c
MeO2C HHHO
=
Me
H
CO2MeCO2Me
1
1
3
4
pro-Spro-R
MeMe
MeMeMe BH
2 H2O2, OH–
MeMe OH
98.4% ee
L-proline cat.NR2 O OH
H
O
Me
Me
Me
O
H+
Me Me
Me
MeH anti/syn = 24/1
>99% ee
AcO OAcMeMe PLE
AcO OHMeMe
(α = k1/k2 = 2.47 ± 0.36)at 36% yield ··· 80% eeat 15% yield ··· 95% ee
c
c
ba
d
c
ba
c
d
ba
k1
k2
k3
k4
d
d
ba
pro-R
pro-S
S
(achiral)R1
2
3
H
OPh H
Me+ MeMgBr Me– a Me–b
2
Ph MeMe
OH
or Ph MeMe
OH
1
O
H
PhMe
3a 3b
a → b → c clockwise ··· re-face anticlockwise ··· si-face
If c > c', the tentative configuration is S. The prochirality of c is pro-S. The prochirality of c' is pro-R.
If the pro-R-c is more reactive than pro-S-c in 1, (S)-2 is obtained as the major product (k1 > k2).
The minor product (R)-2 still has pro-R-c. The group c would be more reactive than pro-S-c remaining in (S)-2. Therefore, (R)-2 would be easier to be converted into achiral 3 than (S)-2 (k4 > k3).
PhO
H 1
2
3si-face
PhO
Hre-face
PhHCNHAc
Me
1
2 3
re-face
– 17 –
• The stereoselectivity in the diastereoselective reaction is often evaluated with
diastereomeric excess (de, %), which is the difference between percentages of major and minor products, (de (%) = 100([major product] – [minor product])/([major product] + [minor product]).
• There are two manners of stereocontrol in diastereoselective reaction. One is the stereocontrol by the chirality of the substrate; another is that by the chirality of the reagent or catalyst.
How to predict the major product • With transition state: In principle, the stereoselectivity is controlled by the difference
in energy between favorable and unfavorable transition states. Comparing the two possible transition states is useful for predicting the major product. However, it is not easy to predict the structure and energy of each transition state.
• WIth structure of chiral substrate: The structure of substrate sometimes strongly relates to the stereoselectivity. When the reaction is known to proceed through early transition state (in Hammond postulate), the major product can be predicted with the steric hindrance imagined from the most stable conformation of the substrate.
• With structures of diastereomeric products (or intermediates): The difference in thermodynamic stability between the possible diastereomeric products sometimes strongly relates to the stereoselectivity. When the reaction is known to proceed through late transition state (in Hammond postulate), the major product can be predicted with the thermodynamic stability of each possible product.
• If the stereochemical effect of the chiral substrate is predominant over that of the reagent, the reaction should be categorized as substrate-controlled reaction.
• If the effect of the chiral reagent or catalyst is predominant over that of the substrate, the reaction should be categorized as a reagent-controlled reaction.
• When the chiral substrate and reagent (or catalyst) are comparable in the effect on stereoselectivity, the direction of asymmetric induction should be discussed with great care.
(ii) Substrate-controlled reaction ・ In the substrate-controlled reaction, the stereochemistry of the substrate predominantly
controls the stereoselectivity. ・ All diastereoselective reactions are substrate-controlled reactions when they are
conducted with an achiral reagent (or catalyst).
Nelson, D. A. Tetrahedron Lett. 1986, 27, 3091.
Evans, D. A. Tetrahedron Lett. 1985, 26, 6005.
• Choice of the achiral reagent may bring about reversal of the stereoselectivity. However, such a reaction should be treated as a substrate-contorolled reaction, because the interaction between the reagent and the substrate must strongly affect the stereoselectivity.
Oshima, K.; Utimoto, K. Tetrahedron, 1993, 49, 11169.
(review) Reiser, O. Chem. Rev., 1999, 99, 1191.
(iii) Double stereodifferentiation (review) S. Masamune, Angew. Chem. Int. Ed. Engl., 1985, 24, 1. • In the reaction between two chiral molecules, both of the substrates affect the
stereoselectivity. The effect of the combination of two chiral molecules is called double strereodifferentiation.
• Diastereoselectivity of the reaction would enhance, when the directions of both asymmetric induction are parallel. The combination of the substrate and reagent is called matched pair. If the directions are opposite each other, the stereochemical combination is called mismatched pair.
Diasteleoselective aldol reaction controlled by chiral aldehyde
TSa
1 + 2 3a
E
1 + 2 3b
E
TSbO
H Meδ–
δ–
Me HPh
OHMe
δ–δ–
HMePh
OMe
OAr Et2O, 23°C
Zn(BH4)2OH
Me
OAr
OHMe
OAr1 2
Ar = 2-MeOC6H4Ar = 2-(t-Bu)C6H4
1/2 = >99/11/2 = <1/99
Et
OH
TBSO
OBzMe
Et
OH
TBSO
OBzMe
[Rh] cat.
H2, CH2Cl2, 25°CEt
OH
TBSO
OBzMe
1 2[Rh] = Rh(dppb)+
[Rh] = Rh[(R)-binap]+[Rh] = Rh[(S)-binap]+
1/2 = 89/111/2 = 97/31/2 = 92/8
t-Bu
OCO2Et
MeMeOH, 0°C
NaBH4, MClnt-Bu
OHCO2Et
Met-Bu
OHCO2Et
Me1 2MCln = noneMCln = MnCl2
1/2 = 10/901/2 = 95/5
HCy
Me
O+
OLiMe OTBS
Me Me
Cy
Me
OH
Me
OOTBS
MeMe
Cy
Me
OH
Me
OOTBS
MeMe2 3
2/3 = 1/2.71
SR S
– 18 –
Diasteleoselective aldol reaction controlled by chiral enolate
Aldol reaction with matched pair
Aldol reaction with mismatched pair
Masamune, S. Angew. Chem. Int. Ed. Engl., 1980, 19, 557.
(iv) Reagent-controlled reaction • In the reagent-controlled reactions, the stereochemistry of the reagent (or catalyst)
predominantly affects the stereoselectivity of the reaction. • A new chiral center possessing the desired configuration can be created on a chiral
substrate independent of its stereochemistry by choosing the enantiomeric reagent (or catalyst).
Substrate-controlled epoxidation
Enantioselective epoxidation (Katsuki-Sharpless oxidation)
Reagent-controlled epoxidation
Masamune, S.; Sharpless, K. B. J. Am. Chem. Soc., 1982, 47, 1373.
See also, Masamune, S. Angew. Chem. Int. Ed. Engl., 1985, 24, 1.
(c) How to control stereochemistry (i) Chiral reagent
• In the reaction producing a chiral compound from an achiral substrate, the product may be obtained as an optically active form, if the reagent is modified with an enantiopure substituent.
• The optically active reagent is called chiral reagent. • The reaction with the chiral reagent proceeds enantioselectively. The product does not
contain the chiral substituent on the chiral reagent. • Stoichiometric or excess amount of the chiral reagent is required to obtain the desired
chiral product with high ee.
Noyori, R. J. Am. Chem. Soc., 1979, 101, 3129; 1984, 106, 6709; 1984, 106, 6717.
(ii) Chiral catalyst • The chiral product may be obtained as an optically active form through a catalytic reaction,
if the catalyst is modified with an enantiopure substituent or compound. • The optically active catalyst and the enantiopure ligand are called chiral catalyst and
chiral ligand, respectively. The enantioselective reaction through a chiral catalyst is called catalytic asymmetric reaction.
• In the catalytic asymmetric reaction, a prochiral substrate (S) interacts with a chiral catalyst (C*) to form a chiral intermediate (S–C*), which is often called catalyst–substrate complex. The complex reacts with another substrate (or reagent) (R) to form the desired product (S–R). The chiral catalyst is regenerated and then activates S again.
• The reagents employed in catalytic asymmetric reactions are achiral.
Ph H
O+
OLiMe OTBS
Ph MePh
OH
Me
OOTBS
Ph Me
S Ph
OH
Me
OOTBS
Ph MeS R
4 5 65/6 = 1/3.5
OLiMe OTBS
Ph Me
S
4
Cy
Me
OH
Me
OOTBS
Ph Me
Cy
Me
OH
Me
OOTBS
Ph Me7 8
R S
7/8 = 1/8
1 +
OLiMe OTBS
Me Ph
R
ent-4
Cy
Me
OH
Me
OOTBS
MePh
Cy
Me
OH
Me
OOTBS
MePh9 10
R S
9/10 = 1.5/1
1 +
OHO
OTi(O-i-Pr)4
TBHP OHO
OO
OHO
OO
1 2 1/2 = 2.3/1
OHBnOTi(O-i-Pr)4, (+)-DET
TBHP, –20°COHBnO O 99/1
OHO
O
OHO
OO
Ti(O-i-Pr)4, (–)-DET
TBHP, –20°C
Ti(O-i-Pr)4, (+)-DET
TBHP, –20°C OHO
OO
1
2
1/2 = 90/1
1/2 = 1/22
mismatched
matched
THPO
THPO
CO2Me
O
Me(S)-BINAL-H
THF, –100 to –78°C
THPO
THPO
CO2Me
OH
Me15S
15S/15R = 100:0
OAl
O H
OEt
–
Li+
(S)-BINAL-H
– 19 –
Kowles, W. S. J. Am. Chem. Soc., 1975, 97, 2567; Acc. Chem. Res. 1983, 16, 106.
(iii) Chiral auxiliary • An achiral or racemic substrate can be modified with an enantiopure compound in order
to obtain the enantio-enriched product, when it has a reactive functional group, e.g. carboxylate, alcohol, or amine. The enantiopure compound for the chiral modification is called chiral auxiliary.
• The chiral auxiliary can be removed with hydrolysis etc. after the desired stereoselective reaction. The used chiral auxiliary can be recycled.
• In this case, the substrate has the chiral center stemming from the auxiliary. Therefore, the stereoselective reaction involving the chiral auxiliary is not enantioselective reaction, but diastereoselective one.
• It is relatively easy to design the key intermediate involving the substrate and chiral auxiliary and to predict the stereochemistry of the product. Therefore, the stereoselective reaction using a chiral auxiliary is frequently used in organic synthesis as compared to other methods.
Evans, D. A. J. Am. Chem. Soc., 1981, 103, 2127.
2-2. Stereochemistry in the nucleophilic addition (1) Acyclic 1,2-asymmetric induction
• Nowadays, the nucleophilic addition is believed to mostly proceed through the transition state proposed by Felkin and Ahn. However, several transition state models have been proposed and used for illustrating the stereochemistry of the nucleophilic addition. This section describes not only the Felkin-Ahn model but also the other transition state models.
(a) Felkin–Ahn model • In a nucleophilic addition, the nucleophile
donates its lone pair to the p* orbital of carbonyl group.
• In the transition state, the electrons shared between the nucleophile and p* orbital conjugate the s* orbital of Ca–X bond.
• The hyperconjugation stabilizes the transition state.
• In the transition state, the angle of Nu–C–O (q) is ca. 110°. • If the carbonyl substrate has a chiral center on its a-carbon,
the largest substituent (L) is located on the antiperiplanar position of Nu.
• To avoid the steric hindrance to the attack of the nucleophile to the carbonyl, the middle (M) and smallest substituent (S) are respectively located by the carbonyl oxygen and carbonyl substituent R in the favorable transition state.
(b) Polar-Felkin–Ahn model • In a C–X bond, the large electronegativity of X causes the decrease in the energy level of
its s* orbital. The effect enhances the ability to accept electrons through the hyperconjugation.
• Therefore, an additional rule is required for predicting or explaining the stereochemistry of the nucleophilic addition, when the electrophilic substrate has a polar substituent on its a-carbon of carbonyl group.
• The polar substituent X, e.g. OMe, Cl, etc., prefers the antiperiplanar position of the nucleophile Nu in the Felkin-Ahn transition state.
• To avoid the steric hindrance to the attack of the nucleophile to the carbonyl, the large (L) and small substituent (S) are respectively located by the carbonyl oxygen and carbonyl substituent R in the favorable transition state.
SC*
S C*R
prochiralsubstrate
catalyst–substratecomplex
(diastereoisomeric)
S C*R
chiral catalyst
S Rchiral
product
enantioselective
CO2Na
NHAc
0.05% cat. {Rh(cod)[(R,R)-DiPAMP]}BF4
50% MeOH, 25°C, 4 h+ H2(4 atm)
CO2Na
NHAc
PP PhPh
MeO
OMe
(R,R)-DiPAMP
96% ee
ON
OOMe
i-PrR H
O+
Bu2BOTf
i-Pr2NEtON
OO
Mei-Pr
OH
RH2O2
LiOH
O
Me
OH
R OH
>99% ee (2S,3R)(R = Bu, i-Pr, Ph)
OHN
O
i-Pr
+
R O
X
Nuθ
OCα
R
X
Nu
:π*
σ*
hyperconjugation
Newman projection
R OS M
L
Nu
Felkin–Ahn model(L > M > S)
O RS M
L
Nu
(favorable) (unfavorable)
R OS L
X
Nu
polar-Felkin–Ahn model(L > S, X = polar substituent)
O RS L
X
Nu
(favorable) (unfavorable)
– 20 –
(c) Cram’s rule • Cram’s rule is an empirical method for predicting the
stereochemistry of the nucleophilic addition. • It is assumed that the largest substituent (L) prefers to stay
in the eclipsed position of carbonyl substituent R. • The nucleophile Nu attacks on the carbonyl carbon from the
side of the smallest substituent S. • In the case of ketone or aldehyde bearing a polar substituent
X on the a-carbon, the polar substituent prefers to stay in the eclipsed position of carbonyl substituent R because of the dipole interaction between C=O and Ca–X (Cornforth model).
(2) Chelation-controlled 1,2-asymmetric induction • Stereoselectivity in the nucleophilic addition is often controlled by a metal
atom, when the substrate has a Lewis basic heteroatom (e.g. -OR) on its a-carbon and the metal atom has sufficient Lewis acidity.
• The heteroatom and the carbonyl oxygen interact with the metal atom to form a chelate complex as shown in the right figure. The coordination of carbonyl oxygen to metal enhances the reactivity of the substrate.
• The nucleophile Nu attacks on the carbonyl carbon from the side of the smaller substituent S.
• In general, the chelation-controlled reaction proceeds with the opposite stereoselectivity to the Felkin–Ahn-controlled reaction.
Yamamoto, H. J. Am. Chem. Soc. 1988, 110, 3588.
Keck, G. E. Tetrahedron Lett. 1984, 25, 265.
(3) 1,3-Asymmetric induction (a) Chelation-controlled 1,3-asymmetric induction
• Stereoselectivity in the nucleophilic addition is sometimes controlled by the stereochemistry of the b-carbon of the substrate, which has a Lewis basic heteroatom on its b-carbon.
• The substrate readily forms a 6-membered chelate complex with a Lewis acid possessing multiple coordination sites (e.g. TiCl4, SnCl4) as shown in the right figure. The nucleophilic addition will occur from the chelate intermediate.
• The nucleophile Nu accesses the carbonyl carbon from above to avoid the steric interaction with the pseudo-axial proton at the a-position.
Reetz, M. T. J. Am. Chem. Soc. 1983, 105, 4833.
Narasaka, K. Tetrahedron 1984, 40, 2233.
(b) Non-chelation-controlled 1,3-asymmetric induction • The nucleophilic addition (Sakurai-Hosomi reaction) provides the 1,3-anti-product with high
diastereoselectivity when BF3 is used as a promoter, although BF3 cannot form the chelate intermediate.
• In the favorable transition state, substituent R1 on the b-carbon would be located at the antiperiplanar position of the formyl group.
• The carbonyl double bond faces the opposite direction to the C–X bond to avoid the dipole repulsion.
• The nucleophile Nu attacks on the carbonyl carbon from the opposite side to the b-carbon.
Reetz, M. T. Tetrahedron Lett. 1984, 25, 729; Evans, D. A. Tetrahedron Lett. 1994, 35, 8537.
MeH
O+ BuMgBr
–78°C
Me
OH
BuMe
OH
Bu
syn anti81%syn : anti = 89 : 11
OR
O
H SnBu3+Lewis acid (LA)
OR
OH
OR
OHsyn anti
LA = BF3•OEt3, R = TBSLA = BF3•OEt3, R = BnLA = MgBr2, R = TBSLA = MgBr2, R = Bn
CH2Cl2
syn : anti = 9 : 91 (Felkin-controlled)syn : anti = 39 : 61syn : anti = 21 : 79syn : anti = >250 : 1 (chelation-controlled)
Me
OBnCHO + SiMe3
TiCl4CH2Cl2–78°C
Me
OBn OH
anti : syn = 95 : 5
Ph
OH O
Ph
Bu3B
THFPh
O O
Ph
Bu2B NaBH4
–78°C Ph
OH OH
Ph Ph
OH OH
Phsyn antisyn : anti = 98 : 2
Me
OBnCHO + SiMe3
BF3
CH2Cl2–78°C
Me
OBn OH
anti : syn = 91 : 9
R
O
L
MSNu Nu
(favorable) (unfavorable)
Cram's rule
R
O
X
LSNu Nu
(favorable) (unfavorable)
Conforth model
O RX
L
S+M
Nu
chelation model(L > S, X = OR etc.)
O TiCl4OBn
HH
HMe
H
Nu
Nu
(favorable)
(unfavorable)
ORH H
XR1
H
Nu
ORH H
XR1
H
Nu
(favorable)
ROH H
XR1
H
(unfavorable)
Nu
– 21 –
(4) Stereochemistry in the reaction of cyclohexanone • The stereoselectivity of the nucleophilic addition of cyclohexanone is strongly affected by
the substituent and/or the size of the nucleophile. • Small nucleophiles prefer the axial attack in general. Large ones prefer the equatorial
attack.
Nu = LiAlH4 (R = H) 90 : 10 Nu = Li(s-Bu)3BH (L-selectride) (R = H) 12 : 88 Nu = MeLi (R = Me) 21 : 79 Nu = MeLi (with MAD) (R = Me) 99 : 1
Yamamoto, H. J. Am. Chem. Soc. 1985, 107, 4573.
2-3. Stereochemistry in the reaction of enolates (1) Stereochemistry in the formation of enolates
• Enolates are generated from carbonyl compounds through the deprotonation of the a-proton with a base. The enolates are possible to possess geometric isomerism. The stereochemistry can be controlled by metal and/or reaction conditions.
(a) Lithium enolate • The deprotonation often proceeds through the conformation in which a Ca–H bond is
perpendicular to the carbonyl group, because the s-orbital of the C–H bond is required to interact with p*-orbital of the C=O bond in order to form the C=C bond in the resulting enolate.
• Repulsion between R1 and R2 in the carbonyl substrate controls the E/Z selectivity of the enolate.
Ireland, R. E. J. Org. Chem. 1991, 56, 650.
• The deprotonation is believed to proceed through Ireland’s transition state model, when lithium dialkylamide was used as the base.
• Higher reaction temperature is advantageous to the formation of E-enolate. Z-Enolate may be kinetically preferable, and E-enolate may be thermodynamically preferable.
Xie, L. J. Org. Chem. 1997, 62, 7516.
Ireland, R. E. J. Am. Chem. Soc. 1976, 98, 2868.
t-BuO
+ Nu t-Bu OH
R
t-Bu R
OH
equatorial-attackaxial-attackt-Bu
ONu
Nu(nucleophile)
Me OEt
O 1. LDA, solvent
2. TBSCl MeOTBS
OMeMe
OTBS
OMe
in THF–45% DMPUin THF
Z : E = 93 : 7Z : E = 6 : 94
Z E
N NMeO
Me
DMPU
O R1
H
R2 HO R1
H
H R2R1 R2O base (B) B: B:
(favorable) (unfavorable)
O–
R2R1
Z-enolate
Et
OMe
R2NLi
THF
with (t-Bu)(TMS)NLi at 23°Cwith Ph(TMS)NLi at –78°C
Et
OLiMe
E : Z = 94 : 6E : Z = 7 : 93
t-Bu
OMe
R2NLi
THF
with (t-Bu)(TMS)NLi at 23°Cwith Ph(TMS)NLi at –78°C
t-Bu
OLiMe
E : Z = 11 : 89E : Z = 0 : 100
R1 R2O
O
H
NLi
R1
L1
L2R2
H
O
H
NLi
R1
L1
L2H
R2
‡
L1L2NLi
‡
(if R1 is large, L1 is small)
(if R1 is small, L1 is large) R1
R2
OLi
R1 R2OLi
Z-enolate
E-enolate
– 22 –
(b) Soft enolization T. Mukaiyama, Chem. Lett. 1976, 559: Bull. Chem. Soc. Jpn. 1980, 53, 174. • Ketones and thioesters can be transformed into their enolates in the presence of a tertiary
amine and a Lewis acid (soft enolization). The interaction between the carbonyl oxygen and the Lewis acid facilitates the deprotonation of the a-hydrogen.
• The soft enolization is useful for preparing boron, tin, silicon, titanium, magnesium enolates. • Stereochemistry of the boron enolate can be controlled by choosing the boron Lewis acid.
Z-Enolate will be selectively formed when two alkyl substituents on the boron atom are relatively small.
• Triflate or iodide is favorable for the formation of Z-enolate. Chloride is favorable for the formation of E-enolate.
Brown, H. C. J. Org. Chem. 1993, 58, 147.
(2) Stereochemistry in the reaction of chiral enolates with electrophiles (a) Cyclohexanone enolate
• The enolates of cyclohexanones take a half-chair conformation. The substituent on the cyclohexene ring prefers an equatorial position.
• An electrophile preferentially approaches the enolate from above in order to avoid the steric hindrance of the pseudo-axial proton.
Kuehne, M. E. J. Org. Chem. 1970, 35, 171.
(b) Evans’s oxazolidinone • Optically active oxazolidinones, which are readily prepared from b-amino alcohols and
phosgene, are widely used as the chiral auxiliaries for carboxylate enolates. • In general, the N-acyloxazolidinone gives its Z-enolate through deprotonation. • The carbonyl oxygen of the oxazolidinone can bond to the metal atom of the enolate. The
chelation fixes the conformation of the enolates. • Electrophiles preferentially approaches the enolate to avoid the steric hindrance of the
substituent R’. • Various electrophiles are possible to react with the chiral enolate to give the optically active a-substituted carboxylates with high enantiomeric excesses.
(R–X) Evans, D. A. J. Am. Chem. Soc. 1982, 104, 1737; 1990, 112, 8215.
(ArSO2N3) Evans, D. A. J. Am. Chem. Soc. 1990, 112, 4011. (BocN=NBoc) Evans, D. A. Tetrahedron 1988, 44, 5525.
(Davis reagent) Evans, D. A. J. Am. Chem. Soc. 1985, 107, 4346.
(b) Enders’s hydrazone (SAMP/RAMP) • SAMP and RAMP are prepared from (S)-proline and (R)-glutamic acid, respectively.
These hydrazones can form chiral hydrazones with an achiral ketone or aldehyde. • The hydrazones are readily deprotonated with lithium dialkylamides to give their
azaenolates. Coordination of the methoxy group to lithium fixes the conformations of the azaenolates.
• Electrophiles react with the azaenolates from the same side of the MeO group to avoid the steric hindrance of the pyrrolidine ring.
Enders, D. Angew. Chem., Int. Ed. Engl. 1976, 15, 549; (review) Tetrahedron 2002, 58, 2253.
O
Ph MeEt3N
BOTfO
Ph Me
9-BBN (c-Hex)2BCl
Et3N
O
PhMe
(c-Hex)2B
Z E
Z : E = >97 : <3 Z : E = <3 : >97
O
R1 Me
O
R1 Me
BR2
R2 X
R22BX
O
R1 Me
BX
R2R2
OR1
H
H MeBR2
2XR3N
OR1
H
Me HBR2
2XR3N
(if R2 is small)
(if R2 is large)
OBR22
R1 Me
Z-enolate
OBR22
R1
MeE-enolate
NR3
NR3
‡
‡
OCO2Me
t-Bu
LiNH2
C6H623 °C
MeO2C
Ht-BuH
H OLi
I Me
I Me
Me I t-BuMeO2C
Me
O
OCO2Me
t-Bu
Me=
NO
OR
O
R'
R2NM
orTiCl4, i-Pr2NEt
(M = Li, Na, K)NO
OR
O
R'
MEl (electrophile)
NO
OR
O
R'El
>90% de(R' = Bn or i-Pr)
El:R X
(X = Br or I)ArSO2N3 BocN NBoc N
O
Ph SO2Ph
O
R1 R2 N
R1 R2
N
MeOSAMP N
R2R1
N
OLi
Me
R2OR2O
El
El
LiNR2 El N
R1 El
N
MeO
R2
O3
O
R1 R2
El
NNH2
OMe
SAMP
NNH2
OMe
RAMP
– 23 –
2-4. Stereoselective aldol reaction (1) Problems in aldol reaction
• Aldol reaction is the reaction of an enolate with an aldehyde or a ketone, giving b-hydroxy carbonyl compound (aldol).
• Chemoselectivity
• Regioselectivity in the formation of enolate
These problems are solved by preformed enolate. Formation of more substituted enolate is thermodynamically favored. Formation of less substituted enolate is kinetically favored.
Hause, H. O. J. Org. Chem. 1969, 34, 2324. (modified method) 第 4 版 実験化学講座 24, p 154.
• Diastereo- and enantioselectivities (stereoselectivity)
(2) Stereocontrol of vicinal chiral centers of the aldol product • In the aldol reactions of prochiral enolates with carbonyl compounds, the aldol products have
vicinal chiral centers. The stereochemistry of the vicinal chiral centers is often controllable.
(a) Aldol reactions through Zimmerman-Traxler transition state • The electrophilic substrate is activated by the interaction between its carbonyl oxygen and
the metal atom of enolate, when the enolate has a strong Lewis acidic metal. The complexation allows the intramolecular aldol reaction through 6-membered transition state (Zimmerman–Traxler transition state).
• In the aldol reactions through Zimmerman–Traxler transition state, Z-enolates selectively yield syn-product. Anti-aldols are preferentially obtained from E-enolates.
Brown, H. C. J. Am. Chem. Soc. 1989, 111, 3441.
Zimmerman, H. E. J. Am. Chem. Soc. 1956, 79, 1920.
(b) Aldol reactions through open transition state model • If the metal atom of enolate cannot interact with the carbonyl oxygen, their aldol reaction
proceeds through the antiperiplanar transition state, in which the dipole of C=O is arrayed to avoid the repulsion of the dipole of C–OSi.
• Both of E- and Z-enolates selectively give the syn-aldol product, because the stereochemistry is mainly controlled by the steric repulsion between R2 and R3.
Noyori, R. J. Am. Chem. Soc. 1981, 103, 2106; 1983, 105, 1598.
H3C CH3
O
H3C
O
H+
H3C
OH
CH3
O
H3C
OH
CH3
O
CH3
H3C
OH
H
O
H3C
OH
H
O
CH3
H3C
O
Ph
O
H+ CH3 Ph
OH OPh
OH
CH3
O
CH3CH3
MeO
MeOSiMe3
MeOSiMe3
(thermodynamically favored)(kinetically favored)
(1) LDA, THF, –78°C (2) Me3SiCl, –78°C to rtEt3N, Me3SiCl, DMF, 130°C, 90 h
99 : 112 : 88
Ph
O
Ph
O
H+ CH3
Ph
OH
Ph
O
CH3
Ph
OH
Ph
O
CH3
Ph
OH
Ph
O
CH3
Ph
OH
Ph
O
CH3
Ph
OMe
9-BBN-Cl
i-Pr2NEt
(c-Hex)2BCl
Et3N
Ph
OBMe
Ph
OB
Me
>99% (Z)
>99% (E)
1. PhCHO
2. H2O2
1. PhCHO
2. H2O2Ph
OH
Ph
O
Me
Ph
OH
Ph
O
Me 98% syn
95% anti
R1 R2O M
O
MO
R1
HR2
H
R3 O
MO
R1
HR2
R3
HZ-enolate
+ R3CHO
‡
R3
OH
R1
O
R2
(favorable) (unfavorable)syn-aldol
R1
R2
O M
O
MO
R1
R2
H
H
R3 O
MO
R1
R2
H
R3
HE-enolate
+ R3CHO
‡
R3
OH
R1
O
R2
(favorable) (unfavorable)anti-aldol
Ph
OSiMe3Me + PhCHO
F– Me
H
H
OPh
Me3SiO
Ph
‡
Ph
OH
Ph
O
Me
Ph
OSiMe3
Me
+ PhCHOF– Me
H
H
OPh
Ph
Me3SiO
‡
Ph
OH
Ph
O
Me
syn : anti = 95 : 5
syn : anti = 94 : 6
R2
H
R3
OH
SiO
R1
(unfavorable)
– 24 –
(3) Stereocontrol with the chiral center of the electrophilic substrate (aldehyde) • In the aldol reactions of chiral a-substituted aldehydes, the stereochemistry is controlled by
Felkin–Ahn or chelate model.
(a) Felkin–Ahn (FA) controlled aldol reaction
Heathcock, C. H. J. Am. Chem. Soc. 1983, 105, 1667.
(b) Polar-Felkin–Ahn (PFA) controlled aldol reaction
Heathcock, C. H. J. Am. Chem. Soc. 1987, 109, 3353.
(c) Chelation-controlled aldol reaction
Reetz, M. T. Tetrahedron Lett. 1984, 25, 729.
(d) Harmony or Conflict between FA model and Zimmerman–Traxler TS • E-enolates are coupled with a-chiral aldehydes to give 2,3-anti-3,4-syn aldol products in
high stereoselectivities. Meanwhile, the stereochemical control is not easy in the aldol reaction of Z-enolates with a-chiral aldehydes.
• The reaction of Z-enolate: In the transition state based on Felkin–Ahn and Zimmerman–Traxler models, syn-pentane interaction will be seen between the enolate a-substituent and the middle-sized a-substituent of aldehyde. The repulsion allows the aldol reaction to proceed through non-Felkin–Ahn transition state leading to the formation of 2,3-syn-3,4-anti-product. The undesired reaction may disturb the selectivity.
R = Ph (syn, syn) : (syn, anti) = 81 : 19 R = c-C6H11 (syn, syn) : (syn, anti) = 27 : 73 R = CH2OAc (syn, syn) : (syn, anti) = 21 : 79
Roush, W. J. Org. Chem. 1991, 56, 4151.
• The reaction of E-enolate: No significant steric repulsion is found in the Felkin–Ahn-based transition state leading to the 2,3-anti-3,4-syn product. Meanwhile, the transition state to the 2,3-anti-3,4-anti product possesses syn-pentane interaction. Therefore, the aldol reaction generally proceeds through Felkin-Ahn transition state to give 2,3-anti-3,4-syn product with high diastereoselectivity.
Roush, W. J. Org. Chem. 1991, 56, 4151; Evans, D. A. J. Am. Chem. Soc. 1995, 117, 9073.
OTBS
t-BuPh
MeH
O+
BF3•Et2OO
t-Bu
OHPh
Me
O
t-Bu
OHPh
Mesyn : anti = 24 : 1
(Felkin–Ahn) (anti-Felkin–Ahn)
OLi
t-BuPh
OMeH
O+
O
t-Bu
OHPh
OMe
O
t-Bu
OHPh
OMesyn : anti = 17 : 83
(anti-Felkin–Ahn) (Felkin–Ahn)
OTBS
t-BuMe
OBnH
O+
TiCl4 or SnCl4O
t-Bu
OHMe
OBn
O
t-Bu
OHMe
OBn–78°C
syn : anti = >95 : 5(anti-Felkin–Ahn) (Felkin–Ahn)
OLi
Me Me
TMSO Me
HMe
RO O
Me Me
TMSO
Me
OHR
Me
O
Me Me
TMSO
Me
OHR
MeFelkin product (syn,syn) anti-Felkin product (syn,anti)
23
4
O
MO
HMe
RO
R'
H
Me
H
H
Me
R
O
MOR
H
MeR'
H
Me
H
+OM
R' MeZ
OH
Me
RO
MeR'
syn, syn
OH
Me
RO
MeR'
2,3-syn, 3,4-anti
syn-pentaneinteraction
Felkin (matched pair)
anti-Felkin (mismatched pair)
Felkin product(anti,syn)
anti-Felkin product(anti,anti)
93 : 7
O
HMe
i-Pr
OPMB i-Pr
OBCy2
MeOH
Mei-Pr
OPMB
i-Pr
O
Me
OH
Mei-Pr
OPMB
i-Pr
O
Me
23
4
O
MO
HMe
RO
R'
Me
H
H
H
Me
R
O
MOR
H
MeR'
Me
H
H
+OM
R'Me
E
OH
Me
RO
MeR'
anti, anti
OH
Me
RO
MeR'
2,3-anti, 3,4-syn
Felkin (matched pair)
anti-Felkin (mismatched pair)
– 25 –
(e) Harmony or Conflict between PFA model and Zimmerman–Traxler TS • If the aldol reaction of chiral a-oxy aldehyde proceeded through polar-Felkin–Ahn TS, the
reaction of E-enolate should be higher stereoselective than that of Z-enolate. However, the former is actually difficult to proceed with high selectivity as compared to the latter.
• From the observation, modified Cornforth TS was proposed for rationalizing the stereoselectivity.
• The reaction of Z-enolate: Although the reaction proceeds with high stereoselectivity to give 2,3-syn-3,4-anti product, large steric repulsion caused by syn-pentane interaction can be found between the enolate a-substituent and the middle-sized a-substituent of aldehyde in the transition state based on polar-Flekin–Ahn model. Cornforth model is proposed in order to rationalize the high selectivity.
• The reaction of E-enolate: The polar-Felkin–Ahn-based transition state leading to 2,3-
anti-3,4-anti product has no syn-pentane interaction. However, the aldol reaction provides the desired product with low stereoselectivity. No significant steric repulsion is not found in the Cornforth-based transition state.
Evans, D. A. Angew. Chem. Int. Ed. 2003, 42, 1761.
(4) Stereocontrol with the stereogenic center of the enolate substrate (a) Reaction of Z-enolate bearing a chiral center at its a-position
• The Z-enolate of chiral ketone 1 reacts with an aldehyde to give the 1,2-syn-2,4-syn aldol product with high stereoselectivity.
Masamune, S. J. Am. Chem. Soc. 1981, 103, 1566.
Evans, D. A. J. Am. Chem. Soc. 1991, 113, 1047.
• According to Evans, the large-sized substituent RL on the a-carbon locates the antiperiplanar position of the forming bond to avoid the steric repulsion between RL and the aldehyde. In the transition state, the middle-sized substituent RS is oriented to the enolate C=C bond to escape the steric repulsion toward the enolate metal M.
• The chiral center of b-position scarcely affects the stereochemistry of the aldol reaction.
Evans, D. A. J. Am. Chem. Soc. 1991, 113, 1047.
i-Pr
OBMe +
O
H Me
OTBSi-Pr
O
Me
OHMe
OTBSi-Pr
O
Me
OHMe
OTBSFelkin anti-Felkin98 : 2
Z
23
4
HOSi
RO
+OM
R' MeZ
(polar-Felkin-Ahn model) (Cornforth model)
M
OO
H
R
SiOMe
HH
R'M
OO
OSi
H
RMe
HH
R'
vs R'
O
Me
OHR
OSi
i-Pr
OB
Me
+
O
H Me
OTBSi-Pr
O
Me
OHMe
OTBSi-Pr
O
Me
OHMe
OTBSFelkin anti-Felkin21 : 79
E 23
4
HOSi
RO
+OM
R'Me
E
(polar-Felkin-Ahn model) (Cornforth model)
M
OO
H
R
SiOH
MeH
R'M
OO
OSi
H
RH
MeH
R'
vs R'
O
Me
OHR
OSi
O
MeTBSO
O
RH+
(c-C5H9)BOTf(i-Pr)2NEt
O
MeTBSOR
OH O
MeTBSOR
OH
(+ others)R = PhR = EtR = i-Pr (Bu2BOTf)
75 : 1>100 : 1>100 : 1
124
syn
syn
1
RL
RS
OMMe
Z
RCHO OR
H
OC
RL
HH
MRSOC
= RL
RS
O
MeR
OH
1,2-syn, 2,4-syn
O
M
Me
H
H
RL
RS
O
MO
Me
H
RS
R
HH
RL
= OC
RL
RSH
MH
OC
RL
RS
O
MeR
OH
(favorable)
(unfavorable)
O
MeMe
Me
Me
TBSO O
H Me
Me
+TiCl4Et3N
or(i-Pr)2NEt
C5
C5 ··· RC5 ··· S
O
MeMe
Me
Me
TBSO OHMe
Me
O
MeMe
Me
Me
TBSO OHMe
Me(+ others)
C5
95 : 595 : 5
– 26 –
(b) Reaction of E-enolate bearing a chiral center at its a-position • The E-enolate of a chiral ketone 2 reacts with an aldehyde to give the 1,2-anti, 2,4-syn
aldol product with high stereoselectivity.
Evans, D. A. Tetrahedron 1992, 48, 2127.
see also, Gennari, C.; Paterson, I. Tetrahedron 1993, 49, 685.
• Sometimes, the aldol reactions of E-enolate selectively gives 1,2-anti, 2,4-anti aldol product.
Evans, D. A. Tetrahedron 1992, 48, 2127.
(5) Asymmetric aldol reaction of the enolate modified with a chiral auxiliary (a) Syn-selective Evans asymmetric aldol reaction
• Chiral N-alkanoyloxazolidinones react aldehydes to yield syn-aldol products with high stereoselectivities. Boron or titanium Lewis acid is commonly used for the enolization of the alkanoyl group.
• In general, the N-alkanoyl group is converted to the corresponding Z-enolate, which react with an aldehyde through Zimmerman-Traxler transition state.
• It is noteworthy that the aldehyde preferentially reacts the re-face of the enolate bearing (4S)-oxazolidinone to give the (2S,3R) product. Other electrophiles, e.g. haloalkane, attack the si-face in general.
• When the oxazolidinone derived from (1S,2R)-norephedirine is used as the chiral auxiliary, the aldol reaction selectively gives the syn-(2R,3S)-aldol.
Evans, D. A. J. Am. Chem. Soc. 1980, 113, 1047.
Evans, D. A. J. Am. Chem. Soc. 1991, 113, 1047.
• Possible reaction pathways and transition states:
• In the transition state of the asymmetric aldol reaction, the carbonyl oxygen of the
oxazolidinone is dissociated from the metal (B or Ti) of the enolate. The C=O of the chiral auxiliary is oriented to the antiperiplanar position of the C–O bond of enolate in order to cancel these dipoles.
• The aldehyde attacks the enolate to avoid the steric hindrance of the substituent of chiral oxazolidinone.
• In the titanium-mediated aldol reaction of N-alkanoyloxazolidinethione 1, each stereoisomer of the syn-aldol product can be selectively prepared by choosing amine base and Lewis acid stoichiometry.
• The aldol reaction of 1 gives (2S)-aldol 3, which corresponds to the Evans aldol product, when it is conducted with 2 eq. of TiCl4 and 1.1 eq. of monoamine, such as i-Pr2NEt.
(c-C6H11)2BClEt3N
94 : 696 : 4
O
MeMe
Me
Me
TBSO O
H Me
Me
+C5
C5 ··· RC5 ··· S
O
MeMe
Me
Me
TBSO OHMe
Me
O
MeMe
Me
Me
TBSO OHMe
Me(+ others)
C5
124
2
O
MOR
H
Me
HH
RL
RS
O
MO
H
Me
RS
R
HH
RL
OC
RL
RSMe
MH
OC
RL
RS
OM
Me
E RCHO= RL
RS
O
MeR
OH
1,2-anti, 2,4-syn
= RL
RS
O
MeR
OH
(favorable)
(unfavorable)
OC
RL
HMe
MRSOC
NO
OO
Me
O
MeBn
Xq
+O
i-PrH(c-C6H11)2BCl
Et3NXq
O
Me
O
Mei-Pr
OH
Xq
O
Me
O
Mei-Pr
OH
2,4-anti 2,4-syn84 : 16
O N
OMe
O
i-Pr+ R CHO
O N
O
Me
O
i-PrBu2BOTf
i-Pr2NEt
R
OH
O N
OMe
O
Me
O N
O
Me
O
R
OH
Ph MePhR = Bu, i-Pr, Ph
syn : anti = > 99 : 1>99% ee (2S,3R)
syn : anti = > 99 : 1>99% ee (2R,3S)
O N
OMe
O
Bn
TiCl4TMEDA
i-PrCHOO N
O
Me
O
Bn
i-Pr
OH
others
98 : 2
S
O
BO
H
R
BuBu
N
OOi-Pr
MeH
ON
OMe
i-Pr
O
BBu2
O N
OMe
BBu2O
i-Pr
O
BOH
R
Bu
Bu
N
MeH
OO i-Pr
O N
O
Me
O
i-Pr
R
OH
O N
O
Me
O
i-Pr
R
OH
Z
Z : E = >100 : 1
RCHO
RCHO
O N
OMe
O
i-Pr
Bu2BOTf
i-Pr2NEt
– 27 –
• The antipode 2 is obtained from the reaction with 1 eq. of TiCl4 an
d 1.1 eq. of diamine base, such as TMEDA or spartein. In this case, the sulfur atom of C=S forms a chelate with the titanium atom in the transition state.
Crimmins, M. T. J. Am. Chem. Soc. 1997, 119, 7883.
See also, Thornton, E. R. J. Org. Chem. 1991, 56, 2489.
(b) Anti-selective Evans asymmetric aldol reaction • The aldol reaction of N-alkanoyloxazolidinone is promoted in the presence of TMSCl by a
magnesium(II) catalyst, giving the anti-aldol product. • The N-acyloxazolidinone 1a is selectively converted into the aldol product 2 with 2R-
configuration. Meanwhile, the anti-(2S)-aldol 3 is selectively obtained from the reaction of the thiazolidinethione 1b.
• The Z-enolate is exclusively formed in the soft enolization with the magnesium Lewis acid. However, the magnesium-catalyzed aldol reaction selectively affords the anti-aldol products, because it proceeds through boat transition states.
(X = O) Evans, D. A. J. Am. Chem. Soc. 2002, 124, 392.
(X = S and PM3 study on TS) Evans, D. A. Org. Lett. 2002, 4, 1127.
(c) Abiko-Masamune chiral auxiliary • In contrast to imides, carboxylate esters are readily converted to E- and Z-enolates by
choosing boron triflate and amine. • N-(Arenesulfonyl)-N-benzylnorephedrines are sometimes employed as the chiral auxiliary
for the asymmetric aldol reaction. It is noteworthy that the chiral auxiliary allows the selective formation of E-enolate and the anti-selective asymmetric aldol reaction. Bulky boron triflate (c-Hex)2BOTf and small amine Et3N is of choice for the anti-selective aldol reaction.
• The chiral auxiliary is possible to use for the syn-selective asymmetric aldol reaction. Use of small boron triflate and bulky amine allows the syn-selective aldol reaction.
(anti-selective) Abiko, A.; Masamune, S. J. Am. Chem. Soc. 1997, 119, 2586.
(syn-selective) Abiko, A.; Masamune, S. Tetrahedron Lett. 1998, 39, 1873. Abiko, A.; Masamune, S. J. Org. Chem. 2002, 67, 5250; 安孫子淳, 有機合成化学協会誌, 2003, 61, 24.
O N
OMe
S
Bn
i-PrCHOO N
O
Me
S
Bn
i-Pr
OHRTiCl4, R3N
CH2Cl2O N
O
Me
S
Bn
i-Pr
OHS
& anti
1 2 3
TiCl4 (2 eq.), i-Pr2NEt (1.1 eq.)TiCl4 (1 eq.), (–)-spartein (2.5 eq.)
2 : 3 : anti = 95 : 0 : 52 : 3 : anti = 2 : 97 : 1
X N
OMe
X
Bn
+ PhCHO
1. MgY2 (10%) Et3N, TMSCl2. H+
X N
O
Me
X
Bn
OH
Ph
2a : 3a (& others) = 32 : 12b : 3b (& others) = 1 : 19
X N
O
Me
X
Bn
OH
Ph
X = O (1a)X = S (1b)
MgCl2MgBr2•Et2O
2 3
O
O Me
N
PhMe
ArBn
1. R2BOTf, R'3N
2. i-PrCHO
O
OMeN
PhMe
ArBn
i-Pr
OH O
OMeN
PhMe
ArBn
i-Pr
OH
Ar = Mes
Ar =
R = c-Hex, R'3N = Et3N
R = Bu, R'3N = i-Pr2NEt
anti : syn = >98 (96% de) : 2
anti : syn = 6 : >94 (94% de)
anti synO
TiO
H
R
SN
Oi-Pr
MeH Cl
ClCl
– 28 –
2-5. Stereochemistry in electrophilic addition to alkenes (1) General
(a) Overview • Electrophilic addition to alkenes starts from the interaction between the C–C double bond
and an electrophilic (cationic) species (El+). • El+ enhances the electrophilicity of the alkene to facilitate the attack of the nucleophilic
(anionic) species (Nu–) through an SN2-like pathway. • Therefore, the reaction proceeds with anti-selectivity.
(b) Electrophilic addition to cyclohexenes (Fürst-Plattner rule)
• In the epoxide opening, the nucleophilic species site-selectively attacks the carbon to produce the diaxial conformer (Fürst-Plattner rule), because the ring opening is similar to the reversed reaction of the intramolecular nucleophilic substitution of halohydrins.
• The Fürst-Plattner rule is applicable to the prediction or discussion of the stereo- and
regioselectivity in the electrophilic addition, which proceeds through 3-memered ring intermediate, e.g. cyclic bromonium intermediate.
Oxy-mercuration
Pasto, D. J. J. Am. Chem. Soc. 1970, 92, 7480.
(2) Stereoselective iodocarboxylation to acyclic alkenes (a) Intermolecular iodoacetoxylation of allylic alcohols
Chemberlin, A. R. Tetrahedron 1984, 40, 2297.
• The iodoacetoxylation proceeds through the ring-opening of cyclic iodonium intermediate. Stereochemistry of the electrophilic addition is anti.
• Potential energies of the iodonium intermediates are affected by the electrostatic interaction between iodine atom and hydroxy group and the steric repulsion between 3-membered ring and a-substituent.
(b) Iodolactonization of allylic alcohols (intramolecular reaction)
Chemberlin, A. R. J. Am. Chem. Soc. 1983, 105, 5819.
Reaction pathway
• In the intramolecular reaction, the p-complex between iodine and alkene rapidly reacts with
the intramolecular carboxylate before the formation of cyclic iodonium intermediate. • The reaction proceeds through more thermodynamically stable p-complex to give 2
preferentially. Chemberlin, A. R.; Hehre, W. J. J. Am. Chem. Soc. 1987, 109, 672.
b
a
d
c El+a
bc
d
El+
ab
cd
El+
Nu–
b
Nu
a El
cd
1 2
El
Nuab
dc
or/and
path 1 path 2
OOt-Bu
H
H
Nut-Bu=
OHt-Bu
Nu
t-Bu NuOHab
(path a, favorable) (path b, unfavorable)
OO
t-Bu
H
H
Nu
t-Bu=
Nu
t-Bu
OH
t-Bu OHNu
(path a, favorable) (path b, unfavorable)
ab
Rt-Bu
Hg(OAc)2
THF, H2Ot-Bu
HgOAc
OHR
t-BuOH
HgOAcR
R = HR = Me
41%100%
48%0%
Rt-Bu
Hg+Rt-Bu
Hg+
(only R = H)
t-Bu
Hg+
OHH
(R = H, Me)
t-Bu
OH
Hg+
R
Bu
OH
MeBu
OH
Me
OAc
I
Bu
OH MeBu
OH
Me
OAc
I
I2, AcOH
98 : 2
94 : 6
HO
HBuH
I+
MeH
–OAc
HO
BuH
H
I+RR
(unfavorable)
HO
HBuH
I+
HMe
–OAc
I2, AcOH
OHHO2C Me
I2, NaHCO3
THF, H2O O
I
Me
OH
O
H
95 : 5
O
I
Me
OH
O
H
OHH
HMeH
CO2–
HHOH
MeH
CO2–
HHOH
MeH
CO2–
II
OHH
HMeH
CO2–
II
H H
HO I+
CO2–
H
OHH
I+
CO2–
MeH
HMe
O
HO
OH
I
Me
O
HO
OH
I
Me
favored conformation
disfavored conformation
disfavored π-complex
favored π-complex
favored iodonium ion
disfavored iodonium ion
fast
slow
– 29 –
(3) Epoxidation of alkenes (a) Epoxidation of 3-substituted cycloalkenes
• Commonly, the oxidant accesses the carbon–carbon double bond through the pathway to avoid the steric hindrance of 3-substituent to give the trans product.
• The stereoselectivity is possible to reverse if the substituent can work as hydrogen donor or acceptor. The functional group attracts the oxidant through hydrogen bond and/or electrostatic interaction and is named ‘directing group.’
Stereocontrol with steric repulsion
Stereocontrol with directing group
Ganem, B. Tetrahedron Lett. 1985, 26, 4895.
(b) Hydroxy-group-directed epoxidation of acyclic alkene • m-Chloroperbenzoic acid (mCPBA) is useful for diastereoselective epoxidation of the allylic
alcohols bearing a cis-g-substituent. However, the substrate scope is very limited.
Sharpless, K. B. Tetrahedron Lett. 1979, 109, 4733.
• Metal complexes, VO(acac)2 and Mo(CO)6, are known to catalyze the epoxidation with t-butyl hydroperoxide (t-BuOOH). In particular, VO(acac)2 facilitates the reaction of olefinic substrates bearing a hydroxy group. The vanadium complex readily forms the ester with the hydroxy group. The epoxidation proceeds intramolecularly.
Sharpless, K. B. J. Am. Chem. Soc. 1973, 95, 6136.
• The hydroxy-directed diastereoselective epoxidation with vanadium catalyst is widely applicable to the oxidation of various chiral alkenols.
Epoxidation of homoallylic alcohols
Mihelich, E. D. J. Am. Chem. Soc. 1981, 103, 7690.
Epoxidation of bishomoallylic alcohols
Kishi, Y. J. Am. Chem. Soc. 1979, 100, 2933.
(4) Hydroboration (a) Overview
• Hydroboration proceeds through a 4-center transition state. Therefore, the reaction proceeds with syn-selectivity.
R
C CR
H H H=
[O]unfavorable
[O]
[O]O
H
H
RSteric control Stereocontrol with directing group
X
C CX
H H H=
[O]favorable
[O]
[O]O
H
H
X
OTBS mCPBA
CH2Cl2, 0°COTBSO
H
HOTBSO
H
H
6.7 : 1
OH OHOH
HOHO
H
H
24 : 150 : 1
CH2Cl2
[O]
[O] = mCPBA[O] = CF3CO3H
R1 Me
OHR2 CH2Cl2
mCPBA R Me
OHR
OR Me
OHR
O
R1 = Me, R2 = HR1 = H, R2 = MeR1 = R2 = Me
64 : 3695 : 595 : 5
αγ
OH OHOH
HOHO
H
H[O]
[O] = peracid[O] = t-BuOOH & cat. Mo(CO)6[O] = t-BuOOH & cat. VO(acac)2
krel = 0.55krel = 4.5krel = >200
92 : 898 : 298 : 2
R3 R2OH
R1t-BuOOH
VO(acac)2
VOR1
H
R2
H
Ln ORO
H
HR3 R3 R2
OH
R1O
>100 : 1R3 R2
OH
R1O
OHAr
Me Et
EtMe
t-BuOOH
VO(acac)2 OAr
Me
Et OH
EtMe
(Ar = p-MeOC6H4) dr = 8 : 1
b
a
d
c BH3a
bcd
H2B H
ab
cd
H2B H
‡
H2B
ab
H
cd =
H2B
ab H
cd
– 30 –
(b) Acyclic hydroboration controlled with A(1,3)-strain • Stereochemistry of the hydroboration of alkenes bearing A(1,3)-strain is controlled by the
stereogenic center at the allylic position.
Kishi, Y. J. Am. Chem. Soc. 1979, 101, 259.
• In the transition state of the hydroboration, the hydrogen of the chiral center is oriented to the synperiplanar position of the C=C bond to avoid the A(1,3)-strain.
• In the favorable transition state, the largest substituent is oriented to the antiperiplanar position of the forming B–C bond.
Houk, K. N. Tetrahedron 1984, 40, 2257.
(c) Acyclic hydroboration controlled by A(1,2)-strain • Alkenes bearing A(1,2)-strain also can undergo the stereoselective hydroboration controlled
by the allylic chiral center. • The stereoselective hydroboration of 2-propenyl group was used for the synthesis of
Lynomycin A. The syn-product was selectively obtained from the reaction with BH3, while dialkylborane, such as 9-BBN, selectively yielded anti-product.
• In the case of BH3, A(1,2)-strain controls the conformation of the transition state. Therefore,
the hydrogen on the chiral center is oriented to the synperiplanar position of the methyl in the favorable transition state, because the allylic strain overcomes A(1,3)-strain. The borane reagent attacks the C–C double bond to avoid the steric hindrance of the substituent L.
• In the case of 9-BBN, the steric repulsion between the borane reagent and the substituent
S overcomes the A(1,2)-strain. Therefore, the hydrogen on the chiral center is oriented to the synperiplanar position of the C–C double bond in the favorable transition state.
Evans, D. A. J. Am. Chem. Soc. 1995, 117, 3448.
OMe Me
OBn1. B2H6
2. H2O2, OH– OMe Me
OBn
OH
Hdr = 8 : 1
OMe
OMe
Me
Me
OH
1. B2H6
2. H2O2, OH– OMe
OMe
Me
OH
MeOH
Hdr = 12 : 1
HS H
L
H
B
CH2OR
HH
MeH
L H
S
H
BCH2OR
HH
Me
A(1,3)-strain
(favorable) (unfavorable)
L > SHH S
L
H
BMe
HH
CH2OR
OOMe H
MeOMe
Me
MeN
O
Me
OH
MeO
O
Bn
R
OH
MeOH
= R
R
OH
MeOH
1. R2BH
2. H2O2, OH–H H
R2BH = BH3•SMe2R2BH = 9-BBN
85% (92 : 8)60% (<5: 95)
BS H
L
CH2Me
H
BH S
L
CH2Me
H
BH L
S
CH2Me
H
(favorable) (unfavorable)L > S
A(1,2)-strain
HH
HH
HH
(favorable) (unfavorable)L > S
BS H
L
CH2Me
H RR
BL H
S
CH2Me
H RR
BH S
L
CH2Me
H RR
repulsion
– 31 –
2-7. Nucleophile Trajectory — Baldwin’s Rule (1) Trajectory of nucleophile
• In the reaction between a nucleophile and electrophile, the LUMO of electrophile interacts with the HOMO of nucleophile, which is usually an unshared electron pair.
• Trajectory of radical is similar to that of nucleophile in most radical reactions, because the radical species provide its unpaired electron to LUMO of another substrate.
(a) Attack on sp3-carbon (tetragonal) • LUMO of electrophile is the s* antibonding orbital of C–X bond, in which X is a leaving
group. Therefore, the nucleophile (Nu) interacts with the s*(C–X) orbital in the nucleophilic attack on the electrophilic sp3 carbon.
• The nucleophile approaches the carbon from the opposite side of X to form C–Nu bond (Walden inversion).
• The angle between trajectory of the nucleophile and the C–X bond, a, should be 180°.
(b) Attack on sp2-carbon (trigonal)
• LUMO of electrophilic double bond is the p* antibonding orbital, therefore its lobe on the sp2 carbon atom interacts with the nucleophile to form the C–Nu bond.
• Trajectory of the nucleophile forms an angle of 105±5° with the C=X bond to maximize the interaction between the lone pair and the p* lobe (Bürgi–Dunitz angle).
Bürgi, H. B.; Dunitz, J. D. Tetrahedron 1974, 30, 1563.
(c) Attack on sp-carbon (digonal) • A p* antibonding orbital of the triple bond is the LUMO of electrophilic substrate in this case.
As with the above trigonal reaction, the nucleophile interacts with the p* orbital. • In the original paper of Baldwin’s rule (vide infra), the nucleophile approaches the
electrophilic carbon from the same side of X (a = 60°). • Recently, the trajectory was revised to a = 120°.
(2) Baldwin’s rule (a) Baldwin’s rule
• Baldwin’s rule is a guideline for predicting the results of nucleophilic cyclizations. • For intramolecular addition of nucleophile to unsaturated bond, the regioselectivity can be
predicted with the rule. • The rule is applicable to intramolecular radical cyclization. • In favored reactions, the internal nucleophile can be positioned on the above trajectory. Baldwin’s rule • In 1976, Baldwin described in his original paper, as follows:
Rule 1: Tetrahedral systems (a) 3 to 7-Exo-Tet are all favoured processes with many literature precedents. (b) 5 to 6-Endo-Tet are disfavoured.
Rule 2: Trigonal systems (a) 3 to 7-Exo-Trig are all favoured processes with many literature precedents. (b) 3 to 5-Endo-Trig are disfavoured; 6 to 7-Endo-Trig are favoured.
Rule 3: Digonal systems (a) 3 to 4-Exo-Dig are disfavoured processes; 5 to 7-Exo-Dig are favoured. (b) 3 to 7-Endo-Dig are favoured.
• Recently, the rule 3 was revised by Alabugin with DFT simulation as well as experimental results. Rule 3 (revised): Digonal systems
(a) 3 to 4-Exo-Dig are borderline; 5 to 7-Exo-Dig are favorable. (b) 3 to 4-Endo-Dig are unfavorable; 5 to 6-Endo-Trig are borderline; 7-Endo-Trig is
favorable. ring size of product 3 4 5 6 7
Tet (sp3) Exo ○ ○ ○ ○ ○ Endo ✕ ✕ ✕ ✕ ?
Trig (sp2) Exo ○ ○ ○ ○ ○ Endo ✕ ✕ ✕ ○ ○
Dig (sp) Exo ○/✕a ○/✕a ○ ○ ○ Endo ✕b ✕b ○/✕b ○/✕b ○
○ and ✕ indicate favorable and disfavorable processes, respectively. ○/✕ is borderline or problematic case. a The reactions are disfavored in the original rule. b The reactions are favored in the original rule.
xxx is ‘Tet’, ‘Trig’, or ‘Dig’, when Y–C is a single, double, or triple bond.
Baldwin, J. E. J. Chem. Soc., Chem. Commun. 1976, 734. (revised rule) Alabugin, I. V. J. Am. Chem. Soc. 2011, 133, 12608.
XXNu–α
Nu X–α = 180°
Nu : σ*(C–X)
X
Nu–
α α = 105±5° X–Nu
C X
Nu:
π*(C=X)
X
Nu–
αα = 120°(?)
NuX–
C X
Nu:
π*(C X)
?
–XY X–
YX–
YX–
YX–
Y3-Exo-xxx 4-Exo-xxx 5-Exo-xxx 6-Exo-xxx 7-Exo-xxx
X–Y
X–Y
–XY
X–Y
X–Y
3-Endo-xxx 4-Endo-xxx 5-Endo-xxx 6-Endo-xxx 7-Endo-xxx
– 32 –
Examples
• Eq. (2) suggested us that the rule is limited to the nucleophile of second-row element.
Baldwin, J. E. J. Chem. Soc., Chem. Commun. 1976, 736.
Bailey, W. F. J. Am. Chem. Soc. 1993, 115, 3080.
Knochel, P. Angew. Chem. Int. Ed. 2000, 39, 2488.
(b) Baldwin’s rule for enolate • Cyclizations involving an enolate intermediate can be classified into two categories, exo or
endo, on the subject of the enolate C–C double bond. These reactions are designated (enolexo)- and (enolendo)-cyclizations, respectively.
• Trajectory of enolate nucleophile is more restricted than those of other nucleophiles, because the nucleophilic electrons delocalized in p-orbitals
• The following rule had been proposed for intermolecular enolate alkylations (Exo-Tet) or aldol reactions (Exo-Trig).
ring size of product 3 4 5 6 7
Exo-Tet Enolexo ○ ○ ○ ○ ○ Enolendo ✕ ✕ ✕ ○ ○
Exo-Trig Enolexo ○ ○ ○ ○ ○ Enolendo ✕ ✕ ✕ ○ ○
○ and ✕ indicate favorable and disfavorable processes, respectively.
Baldwin, J. E. Tetrahedron 1982, 38, 2939.
Examples
Baldwin, J. E. Tetrahedron 1982, 38, 2939.
MeO2C
H2NO
OMe
HNMeO2C CO2Me
HNMeO2C
O5-Exo-Trig 5-Endo-Trig
100% 0%CO2Me
HS SCO2Me
5-Endo-Trig
NaOMeMeOH, 65 °C
(2)
(1)
Ph
I
t-BuLi
Et2O, C5H12, –78 °C1. warm2. MeOH
Ph4-Exo-Dig
93%
Bu
NH2
KO(t-Bu)
NMP, rt NH
Bu 5-Endo-Dig78%
–O YY–O O
(Enolendo)-Exo-Tet
O
(Enolexo)-Exo-Tet
–O YY–O O
(Enolendo)-Exo-Trig
Y–O
(Enolexo)-Exo-Trig
Y–
O
OO KOH
MeOH, rt
O
O6-(Enolendo)-Exo-Trig
77%
OO
O
O
5-(Enolendo)-Exo-Trig0%
O
Ph OPh
O45
3
5
5 KOH
MeOH, reflux
O
Ph
Ph O
6
6-(Enolendo)-Exo-Trig: Enolendo: Enolexo
– 33 –
2-7. Pericyclic reactions (1) Pericyclic Reaction
• Pericyclic reaction is the reactions through a cyclic transition state, in which the electrons move round the circle concertedly and simultaneously. Bond cleavages and formations took place at the same time in the pericyclic reaction.
• Pericyclic reactions can be classified into four categories, (a)–(d), as follows:
(a) Cycloaddition • In cycloadditions, two or more unsaturated molecules (or p-components, see Woodward–
Hofmann rule) forms two or more bonds between molecules at each terminus of the components to yield the cyclic adduct.
Examples
(b) Electrocyclic reaction
• Electrocyclic reaction is the cyclization of conjugated p-system (e.g. 1,3,5-hexatriene) containing two or more multiple bonds or its retro-reaction (ring-opening). The cyclization forms a new s-bond between the termini of the component.
Examples
(c) Sigmatropic rearrangement (i) [m,n]-Sigmatropic rearrangement (m, n ≠ 1)
• [m,n]-Sigmatropic rearrangement is the intramolecular reaction between two p-components, which are connected through one s-component (three s-bonds). The s-component is cleaved and a s-bond is concurrently formed between the termini of the p-components.
Examples
(ii) [1,n]-Sigmatropic rearrangement (n ≠ 1)
• [1,n]-Sigmatropic rearrangement is the internal reaction between p- and s-components, which are connected through a s-bond. The s-component is cleaved and the eliminated terminal atom bonds to the terminal of the p-component.
Examples
+Δ
Diels–Alder (diene) reaction
Ozonolysis
+Δ
O–
O+
O+
OO
O O
O+–O
[3+2]-cycloaddition
OO
O
retoro-[3+2]-cycloaddition [3+2]-cycloaddition
Cheletoropic reactionO
SO
+ S OO
Δ or hν
Δ or hν
Electrocyclic reaction
Proposed biosynthesis of endiandric acids
HO2CHH
HO2C RHO2C PhR
R
endiandric acid DNazarov cyclization
8π 6π
R’
RO
H+ R’
ROH+ 4π
OHR’
R
H +
OR
R’– H+
Cope rearrangement
Claisen rearrangement
Δ
Oxy-Cope rearrangement
–O
Δ
–O OHC
O Δ
H
O OH O O
Fisher indole synthesis
NHNH2 O R
R’+
H+
NH
HN R
R’ Δ
HNHN R
R’
NH
R
R’
[2,3]-Wittig rearrangementR
O EWG
base R
O EWG–
–O EWG
R
EWG = electron-withdrawing group
H+
[1,7]-Sigmatropic rearrangementH
HO R
1
234
5
6
7 H
HO R
AcOHO
[1,3]-Sigmatropic rearrangement
– 34 –
(d) Group transfer reaction • In group transfer reaction, one or more atoms (or groups) are transferred from one
molecule to another. Examples
(2) Frontier Molecular Orbital (FMO) Theory (a) p-Molecular orbital in conjugate system
• Phases of p-orbitals in various p-conjugation system
• Electrons were placed into the orbitals on Hund’s rule. • The highest energy orbital, which is occupied with two electrons, is called HOMO (highest
occupied molecular orbital). • The lowest energy orbital, which has no electron, is called LUMO (lowest unoccupied
molecular orbital). • The orbital having an unpaired electron is called SOMO (singly occupied molecular orbital).
(b) Frontier molecular orbital (FMO) theory • In the reaction between two molecules, the HOMO of one (nucleophilic or electron-rich)
substrate interacts with LUMO of the other (electrophilic or electron-deficient) substrate. • A bond forms between the atoms having the largest coefficient in the HOMO and LUMO. • A pair of molecular orbitals having the same phase leads to the bond formation. The
orbital pair generates a new bonding molecular orbital. • A pair of molecular orbitals, whose phases are different from each other, leads to the
generation of an antibonding molecular orbital. Fukui, K. J. Chem. Phys. 1952, 20, 722.
(3) Woodward–Hoffmann Rule A ground-state pericyclic change is symmetry-allowed when the total number of (4q+2)s and
(4r)a components is odd (q, r = integral number). Woodward, R. B.; Hoffmann, R. Angew. Chem. Int. Ed. 1969, 8, 781.
• Here, ‘component’ means a bond or orbital taking part in a pericyclic reaction as a single unit.
e.g. s2a p4s w0s etc. (molecular orbital)(Number of electrons)(reaction manner)
(a) How to use Woodward–Hoffmann rule ① Draw the equation of the pricyclic reaction including its TS. ② Choose all components (bonds or orbitals) involved with the pericyclic reaction. ③ Assign each component to p, s, or w. (non-essential)
p p bond s s bond w single orbital (e.g. lone pair or carbocation)
④ Count the number electrons in each component. ⑤ Assign each component to s or a.
• p component s The component forms new two bonds on the
same face at both ends. (suprafacial) a The component forms new two bonds on
opposite faces at both ends. (antarafacial) • s component
s The component forms new two bonds with the large lobe of sp3 orbital on each atom. (suprafacial)
a The component forms a bond with the large lobe of sp3 orbital on an atom. On the other atom, the small lobe of sp3 orbital is used for the formation of another bond. (antarafacial)
• w component This component can participate in either a suprafacial or antarafacial manner.
⑥ Count the number N of (4q+2)s and (4r)a components. • Thermal reaction
If N = odd, the pericyclic reaction is allowed. If N = even, the pericyclic reaction is forbidden.
• Photoreaction If N = even, the pericyclic reaction is allowed. If N = odd, the pericyclic reaction is
forbidden.
Ene reaction
H+
Δ H
Diimide reductionR
RNN
H
H+
Δ NN
H
R
R
H+
antibonding
bonding
E
•••••••
ψ1
ψ2
ψ1
ψ2
ψ3
ψ1
ψ2
ψ3
ψ4
ψ1
ψ2
ψ3
ψ4
ψ5
ψ1
ψ2
ψ3
ψ4
ψ5
ψ6
suprafacial antarafacial
suprafacial antarafacial
– 35 –
(4) Cycloadditions (a) Diels–Alder reaction
(i) General • Diels–Alder reaction is the [4+2] cycloaddition of 1,3-dienes and alkenes (dienophile).
Commonly, the cycloaddition proceeds through the concerted interaction between the HOMO of diene and the LUMO of dienophile.
• In common Diels-Alder reaction, the reaction rate is enhanced by installing an electron-withdrawing group to the alkene. Also, electron-donating group in the diene substrate accelerates the formation of [4+2] cycloadduct.
• The cycloaddition is possible to proceed with substrate combination of electron-deficient 1,3-diene and electron-rich alkene. The cycloaddition is called inverse-electron-demand Diels-Alder reaction.
(ii) Stereochemistry in Diels-Alder reaction
• Diels–Alder reaction is stereospecific. For example, the [4+2] cycloaddition of trans-alkene specifically produces trans-4,5-disubstituted cyclohexene. Meanwhile, cis-alkene provide cis-4,5-disubstituted cyclohexene.
• Endo-rule: Commonly, the cycloaddition of cyclic 1,3-diene with dienophile selectively
give endo-product, because increasing the interaction between the HOMO and LUMO is preferable for the Diels–Alder reaction.
• The exo isomer is obtained as a major product when the reaction proceeds under thermodynamic control, because the cycloaddition is reversible at high temperature.
(iii) [4+2] Cycloaddition of substrates other than diene or alkene.
• Component containing a carbon–heteroatom multiple bond can work as diene or dienophile substrate in Diels–Alder reaction. The [4+2] cycloaddition is called hetero-Diels–Alder reaction.
• Carbon dioxide and molecular nitrogen can work as the dienophile component in the
[4+2] cycloaddition. Its retro reaction is often used for the in-situ generation of difficult-to-use compounds, such as 1,3-butadiene and ortho-quinodimethanes.
(b) [m+n] Cycloadditions • A variety of unsaturated functional groups, including 1,3-dipolar and carbene-like
compounds, can participate the pericyclic cycloaddition as diene components. • Each component commonly reacts in a suprafacial–suprafacial manner. (i) 1,3-Dipolar cycloaddition ([3+2] cycloaddition)
• 1,3-Dipole is a compound with four delocalized p-electrons in which a charge separation is found in a major resonance structure. which
(iii) Unusual cycloadditions [2+2] cycloaddition of ketones
(c) Discussion with FMO theory: cycloaddition (i) Thermal [4+2] cycloaddition
Possible HOMO–LUMO interactions • The interaction between HOMO of 1 and LUMO of 2 in TS.
Ψ1
Ψ2*Ψ3*
Ψ2
Ψ2
Ψ3*
Ψ1
Ψ2*
EWGEDG
Ψ2
Ψ3*
Ψ1
Ψ2*
EDGEWG
Diels-Alder reactionInverse-electron-demand
Diels-Alder reaction
a
b
c
d+
a
b
c
d& regioisomers
a
b
c
d
O
O
O
+OO
O
‡
OOO
endo-1
OO
O
Δ
exo-1
N
R1
R2
+ N R2
R1
aza-Diels–Alder reaction
oxo-Diels–Alder reactionO
R+ O
R
O R+
O R
OHC+
CHO1 2
– 36 –
The orbitals on C1 and C4 overlap those on C6 and C5 with the same phase in TS,
respectively. Therefore, the thermal cycloaddition is ‘allowed’. • The interaction between LUMO of 2 and HOMO of 1 in TS.
As with the interaction between HOMO of 1 and LUMO of 2, the orbitals on C1 and C4
overlap those on C6 and C5 with the same phase, respectively. This orbital interaction scarcely affects common Diels–Alder reaction. However, the inverse-electron-demand Diels–Alder reaction (vide infra) was governed by this interaction.
(ii) [4+2] Photocycloaddition • In photo-induced pericyclic reaction, one of the substrates (diene or dienophile in Diels–
Alder reaction) is excited by the photo-irradiation. In the excited molecule, an electron in HOMO transfers to LUMO, giving two SOMOs.
• The higher and lower SOMOs interact with the LUMO and HOMO of the other substrate, respectively.
• In frontier molecular orbital theory, the interaction between HOMOs or LUMOs must be
considered for photo reactions.
In both HOMO–HOMO and LUMO–LUMO interactions, the orbital on C4 overlaps that on C5 with the same phase. Meanwhile, the orbital on C1 overlaps that on C6 with the opposite phase. Therefore, the photocycloaddition is ‘forbidden’.
(b) Discussion with Woodward–Hoffmann rule: cycloadditions (i) Diels–Alder reaction
This reaction is classified into [p4s + p2s] cycloaddition. N = 1 (p2s) → allowed (thermal)
(ii) retro Diels–Alder reaction
This reaction is classified into retro [p2s + s2s + s2s] cycloaddition. N = 3 (p2s, 2×s2s) → allowed (thermal)
(iii) [2+2] cycloaddition
This reaction is classified into [p2s + p2s] cycloaddition. N = 2 (2×p2s) → forbidden (thermal)
allowed (photo)
OO
HOMO (1) LUMO (2) CHO12
3 4 567
‡
OO
LUMO(1) HOMO (2) CHO12
3 4 567
‡
HOMO
LUMOhν
1 1* 2exciplex product
OO
‡
LUMO (1*) LUMO (2)
12
3 4 567 No Reaction
OO
‡
HOMO (1*) HOMO (2)
12
3 4 567 No Reaction
a
bc
d
e f
g h+
b a
f
d
h
e
gc
c
d
a
b
gf
h
e
‡π4s
π2s
b a
f
d
h
e
gc
=σ2s
σ2s
π2s+
a b
c d
a b
c d+
a b
c d
a b
c d
‡a
c c
ab b
d dπ2s π2s
– 37 –
(5) Electrocyclic Reaction (a) Electrocyclic reaction of neutral molecules
• All electrocyclic reactions are allowed thermally as well as photochemically. • However, the photocyclizations proceed in opposite stereochemistry to yield the
diastereoisomeric products of the thermal cyclization products.
• An electrocyclic reaction is classified into disrotatory or conrotatory reaction from the
viewpoint of stereochemistry. • In disrotatory reactions, one group rotates clockwise and the other group rotates anti-
clockwise. • In conrotatory reactions, the two groups taking part in the bond formation or cleavage rotate
in the same direction: both clockwise or both anticlockwise.
(b) Woodward–Hofmann rule in electrocyclic reaction
• Disrotatory reaction is suprafacial. Conrotatry reaction is antarafacial.
equation thermal reaction photoreaction
4e system
con close: [p4a] open: [p2s + s2a] dis close: [p4s]
open: [p2s + s2s]
6e system
dis close: [p6s] open: [p4s + s2s] con close: [p6a]
open: [p4s + s2a]
8e system
con close: [p8a] open: [p6s + s2a] dis close: [p8s]
open: [p6s + s2s]
con: conrotatory, dis: disrotatory
• Any electrocyclic reactions are symmetry-allowed. However, the reaction is restricted by steric or geometrical factor. (e.g. formation of trans-olefin etc.)
(c) Electrocyclic reaction of ionic molecules
• Ring-opening [w0a + s2s] electrocyclic reaction
This ring-opening proceeds in a disrotatory manner. The reaction is formally 2e-system,
when it was supposed to proceed through cyclopropyl cation. • Nazarov cyclization ([p4a] electrocyclization)
This electrocyclic reaction is conrotatory, because it proceeds through pentadienyl-
cation-like intermediate, which is 4e system. • Ring-opening [p4s + s2s] electrocyclic reaction
This ring-opening reaction is disrotatory, because the reaction involves the interaction
between the bridged cyclopropyl C–C s-bond and the allyl anion, which contains four p-electrons.
Ph
PhΔ
Ph
Ph
OO O
O
Ph
Ph
O
O
H
H
R'O
Me
RMeΔ hν
R'O
Me
RMe
R'O
Me
RMe
HH
–10 °C, 30 h 20 °C, 8 hH
H
9 °C, 155 h 40 °CH
H
R
R
R
R
conrotatory
disrotatory
R
RR
R
D
D
D
D
D
H
D
H180 °C
H H
D
D
H
H
Cl– Cl–
H
H
+
+ ClCl–Cl=
OMe3Si OMe3Si FeCl3–
+FeCl3
H H
OMe3SiFeCl3–
+
O
H H
H
Ph Ph
NaOR
Ph Ph
–
Ph Ph
–
Ph Ph Ph Ph
or
– 38 –
(6) Sigmatropic Rearrangement (a) [m,n]-Sigmatropic rearrangements
(i) [3,3]-Sigmatropic (Cope, Claisen) rearrangement
Structures of possible TSs
• Chair-form is commonly preferable to boat-form. • When the dissociating bond in TS is treated as s2s, this rearrangement is [s2s + p2s + p2s],
in which N = 3. The thermal sigmatropic reaction is symmetry-allowed. • When the dissociating bond in TS is treated as s2a, this rearrangement is [s2a + p2a + p2s],
in which N = 1. The thermal sigmatropic reaction is symmetry-allowed. Stereochemistry of Cope rearrangement
(ii) [2,3]-Sigmatropic (Wittig) rearrangement
• This rearrangement is [s2s + p2s + w2s], in which N = 3. • If the [2,3]-rearrangement proceed through the corresponding benzyl cation intermediate,
it will be [s2s + p2s + w0s], in which N = 2. The reaction will be symmetry-forbidden.
(b) [1,n]-Sigmatropic hydrogen shifts
• The [1,n]-hydrogen shift is possible to proceed through either suprafacial or antarafacial
pathway. The stereochemistry of the product depends on the reaction pathway.
(i) Stereochemistry of [1,n]-hydrogen shift in Woodward–Hoffmann rule
• The [1,n]-sigmatropic hydrogen shift is [s2s + p(n–1)s] or [s2s + p(n–1)a]. • The C–H bond participating in the reaction is always s2s. • If n = 4p + 1, the reaction proceeds through suprafacial
pathway. • If n = 4p – 1, the reaction proceeds through antarafacial
pathway. • Symmetry-allowed [1,n]-hydrogen shifts can be restricted by steric or geometrical reason.
equation thermal reaction photoreaction
[1,3]-H shift
a [s2s + p2a] s [s2s + p2s]
[1,5]-H shift
s [s2s + p4s] a [s2s + p4a]
[1,7]-H shift
a [s2s + p6a] s [s2s + p6s]
[1,9]-H shift
s [s2s + p8s] a [s2s + p8a]
a: antarafacial, s: suprafacial
(ii) Examples • Suprafacial pathway in [1,5]-sigmatropic hydrogen shifts was proved by the observed
stereochemistry of the rearrangement product.
1
1'
m
n
1
1'
m
n
[m,n]
O O
π2s π2s
σ2s
π2s π2s
σ2s
chair-form boat-form
180 °C
18 h
‡E
E
‡ E
Z
240 °C
24 h
97%
97%
‡ E
Z
‡E
E
OPh
BuLi
OPh– Ph
σ2s
π2sω2s
‡
Ph OH
H
R
H
R R
H
[1,5]-hydrogen shift [1,7]-hydrogen shift
R
H
HH
H
a
b
path a path b
suprafacial antarafacial
H H13
H1
5 H
H1
7 H
H1
9 H
DEtHMe Me
Me MeH
DEtR E E S =
Me HD
MeEt
MeEtDMe H=
MeMeEtD H
Et HMe
MeZ R R Z
D
Me HMe
DEtE R
not observedEt Me
HDMe
Z S
Hπ4s
σ2sHπ6a
σ2s
– 39 –
• Antarafacial pathway in [1,7]-sigmatropic hydrogen shifts was proved by the observed stereochemistry and deuterium distribution of the recovered substrate.
(c) [1,n]-Sigmatropic alkyl shifts
• Antarafacial pathway is allowed for the s component in [1,n]-alkyl shifts, while [1,n]-hydrogen shift limits the s component to suprafacial.
(i) [1,3]-Sigmatropic alkyl shift
• [1,3]-Sigmatropic alkyl-shift is possible to proceed, but the stereogenic center on the alkyl
group inverts.
(ii) Stereochemistry of [1,n]-alkyl shift in Woodward–Hoffmann rule • In thermal reaction with 4p+2 electron system, the configuration of the stereogenic center
in R is retained when R suprafacially migrates to the reaction terminus (si, ar). • In thermal reaction with 4p electron system, the configuration inverts when R suprafacially
migrates to the reaction terminus. • In thermal reaction with 4p electron system, the configuration is retained when R
antarafacially migrates to the reaction terminus. equation thermal reaction photoreaction
2e system
sr ai
[s2s + w0s] [s2a + w0a]
sr ai
[s2s + w0a] [s2a + w0s]
4e system
si ar
[s2a + p2s] or [s2a + w2s] [s2s + p2a] or [s2s + w2a]
sr ai
[s2s + p2s] or [s2s + w2s] [s2a + p2a] or [s2a + w2a]
6e system
sr ai
[s2s + p4s] [s2a + p4a]
si ar
[s2a + p4s] [s2s + p4a]
8e system
si ar
[s2a + p6s] [s2s + p6a]
sr ai
[s2s + p6s] [s2a + p6a]
p component: a ··· antarafacial, s ··· suprafacial s component: r ··· retention, I ··· inversion
Examples
RDHMe
OH
MeR
D
OH
MeMeH RH
MeD
OH
Me
de
acb
σ2s
π2s
e
d
a
c b
[σ2s + π2s] symmetry-forbidden
de
acb
σ2sπ2a
e
d [σ2s + π2a] symmetry-alloweda c
b(but sterically impossible)
de
acb
σ2a
π2s
e
d
a
cb [σ2a + π2s] symmetry-allowed
R+[1,2]
R–
R[1,2] R+
[1,3] [1,4]
R–[1,4] R [1,5] R + [1,6]
RR – [1,6] [1,7] R+
[1,8]
D
OAcAcOD
H
H inversion
D
HRD
D
RD
H DR
D
H
[1,3]
[1,5] [1,5]
H
D
R
D
R
H D
DR
D
D
H