Hypervalent Silicon:
Bonding, Properties and Synthetic Utility
R
R
Si R
R
R
R Si R
R
R
MacMillan Group Meeting
Ian Storer
20th July, 2005
R
R
R
Si
R
R
R
Hypervalent Silicon: Reactivity and Application in C-C Bond Formation
presentation outline
! Introduction to hypervalent silicon chemistry
! comparison of silicon with carbon – reactivity & coordination
! physical and chemical reactivity characteristics of hypervalent silicon complexes
! bonding considerations
! Asymmetric C(sp3) – C(sp3) bond forming reactions – Organocatalytic Lewis base catalysis
! Conceptual origins – Sakurai allylation reaction (mid 1980s)
! Asymmetric addition of allyl silanes – Kobayashi, Denmark, Kocovsky (1994-present)
! Asymmetric synthesis of aldol products – Denmark (1994-present)
! C(spn) – C(sp2) bond forming reactions
! Pd cross-coupling – Denmark (1994-present)
Useful Reviews:
! Hypervalent Silicon as a Reactive Site in Selective Bond-Forming Processes. Rendler, S.,
Oestreich, M., Synthesis. 2005, 11, 1727-1747.
! Carbon-Carbon Bond Forming Reactions Mediated by Silicon Lewis Acids. Dilman, A. D., Ioffe, S.
L., Chem. Rev. 2003, 103, 733-772.
! Comparison of Phosphorus and Silicon: Hypervalency, Stereochemistry and Reactivity. Holmes,
R. R., Chem. Rev. 1996, 96, 927-950.
! Reactivity of Penta- and Hexacoordinate Silicon Compounds and Their Role as Reaction
Intermediates. Chult, C., Corriu, R. J. P., Reye, C., Young, J. C., Chem. Rev. 1993, 93, 1371-1448.
! Corriu, R. J. P.; Perz, R.; Réye, C. Tetrahedron, 1983, 39, 999.
Introduction to Silicon Chemistry
physical properties
! Physical characteristics
! Silicon comes directly below carbon in periodic table – atomic no. = 14 (3s2 3p2)
Electronegativity
(Allred-Rochow scale)
Si
1.7
H
2.1
C
2.5
Cl
3.0
N
3.0
O
3.5
F
4.0
!-Bond strengths (kcal/mol)
Average Bond Lengths (Å)
C–C
C–Si
Si–Si
83
76
53
C–C
C–Si
1.54
1.87
C–H
Si–H
83
76
C–O
Si–O
1.43
1.66
C–O
Si–O
86
108
C–N
Si–N
83
76
C–F
Si–F
116
135
Silicon forms very strong bonds to oxygen and fluorine. Much of organosilicon chemistry is
driven by the formation of these bonds at the expense of weaker bonds
!
Silicon does not form very stable multiple bonds, as the large 3p orbital on Si does not overlap
well with the 2p orbital on C, O or N
!
"-Bond strengths (kcal/mol)
Si=C, Si=O and Si=N are generally not found
C=C
C=Si
65
36
Coordination of Silicon vs Carbon
common coordinations
! Carbon (2s2 2p2): adopts 3- and 4- coordinate complexes
L
R
C
L
L
–L
R
C
L
R
L
L
C
L
R
L
sp3
3-coordinate
carbocation
R
C
L
+L
sp2
C
L
sp
L
L
5-coordinate
(Texas)
4-coordinate
! Carbon is unable to access hypervalent complexes
! Silicon is below carbon in the periodic table but is capable of very different bonding characteristics
! Silicon (3s2 3p2): adopts 4-, 5- and 6- coordinate complexes
R
Si L
L
3-coordinate
siliconium ion
–L
L
Si L
R
L
4-coordinate
sp3
L
+L
R
Si L
L
L
5-coordinate
+L
L
R
L
Si
L
L
L
6-coordinate
+L
L
R
L
Si
L
L
L
L
7-coordinate
! How does silicon access these higher coordination complexes? – need to consider the bonding options available
! What are the reactivity profiles of the different coordination states?
Hypervalent Silicon : Pentavalent and Hexavalent Complexes
Chemical Reactivity
! Silicon can adopt 4-, 5- and 6- coordinate complexes
! 4-coordinate = electrophile, ! 5-coordinate = electrophile & nucleophile,
L
+L
Si L
R
L
L
R
Si L
L
L
! 6-coordinate = nucelophile
L
+L
R
L
Si
L
L
L
4-coordinate
5-coordinate
6-coordinate
poor Lewis acid
poor R nucleophile
good Lewis acid
good R nucleophile
not Lewis acidic
v good R nucleophile
! Increasing !+ at silicon
! Increasing !– at ligands L andR
! Increasing Lewis acidity
L = negatively charged or neutral silaphilic ligands such as
F, Cl, O-alkyl, O-aryl (good Lewis bases)
R = H or C(spn) (n = 1-3)
nucleophilicity of R = ability of R–transfer
! Electron density at Si decreases with increased coordination, causing the electropositive character
(Lewis acidity) of the Silicon centre to be increased
! How does the extracoordination or hypervalency originate – vacant d-orbitals on Si combined with
the effect of "* (Si–L) orbitals
Bonding to Silicon - How are 5 or 6 Bonds Accommodated?
Valence Shell Electron Pair Repulsion Theory (VSEPR)
! Theory to account for molecular bond geometries
! Predicted hybrid orbitals for 2–6 coordinate compounds
linear
sp
trigonal planar
sp2
tetrahedral
sp3
trigonal bipyramid
sp3d
octahedral
sp3d2
! Tetrahedral coordination – sp3 rehybridization
H
p
hybridize
sp3
H
C
H
H
sp3
tetrahedral
s
! Trigonal bipyramidal 5-coordination – sp3d rehybridization?
d
dp
H
H
hybridize
C
H
p
sp2
H
H
sp2dp
trigonal bipyramid
s
Gillespie, R. J. Chem. Soc. Rev., 1992, 21, 59.
Michael, F. Evans Group Seminar: Hypercoordinate Main Group Compounds, 1999.
The Role of d–Orbitals in Main-Group Compounds
pentavalent compounds
! How do 3s, 3p and 3d orbitals really hybridize to permit pentavalency?
d
d
dp
>200 kcal/mol
p
hybridize
hybridize
p
sp2
sp2
s
sp2p
sp2dp
! The d-orbitals must be close enough in energy to the s and p orbitals to mix favourably
! The 3sp2dp hybridization would come at a massive energetic cost of >200 kcal/mol rendering this
unlikely to ever occur – sp2p hybridization is likely to occur preferentially
! 3d orbitals are still involved to a limited extent
! The d-orbitals have been essential for complete computation of all main group compounds
! Their role appears to be confined to that of polarization of the p-orbitals – d-orbital occupation of <0.3e
+
p
=
0.0–0.3 d
'polarized p-orbital'
Michael, F. Evans Group Seminar: Hypercoordinate Main Group Compounds, 1999.
Hypervalent Main Group Complexes
The Existence of 3c–4e Bonds
! How do 3s, 3p and 3d orbitals really hybridize to permit pentavalency?
d
5-coordinate
p
sp2
p
hybridize
sp2
p
sp2
sp2
3 x sp2
1 x 3c–4e with p
s
sp2p
! The 3c–4e bond – MO consideration
! The filled non-bonding MO has all of the elcetron density on
!"
the ligand atoms
! Consequently ligands/atoms that stabilize electron density
(electronegative) promote 3c–4e bonding.
! This is why virtually all hypervalent compounds bond F, Cl, OR!
n
! Bent's rule: Electronegative elements prefer bonds with more p-character
! Result: Hypervalent compounds are more d+ at the central atom (Si) -
more Lewis acidic
!
Hypervalent Silicon
bond order and charge distribution
! Consider Bond Order and charge distribution
sp3
X
X
Si
X
1x 3c–4e
B.O. = 0.83
+X–
X
X
X
Si X
X
X
2x 3c–4e
–
+X–
X
2–
X Si X
X
X
X
B.O. = 1
B.O. = 0.75
B.O. = 0.66
Charges:
X=0
Si = 0
Charges:
Xax = –0.25
Xeq = –0.17
Si=0
Charges:
X = –0.33
Si = 0
! Study of charge distribution
Complex
Coordination
Si charge
Ligand charge
SiH4
SiH5–
4
5
+0.63
+0.84
–0.16
–0.29(eq), –0.49(ax)
SiH3F
SiH3F2–
4
5
+1.10
+1.26
–0.15(H), –0.67(F)
–0.26(H), –0.74(F)
SiF4
SiF4.NH3
SiF4.2NH3
4
5
6
+1.434
+1.470
+1.463
–0.358
–0.397(F, eq), –0.385(F, ax), +0.084(NH3)
–0.463(F), +0.196(NH3)
Voronkov, M. Top. Curr. Chem. 1986, 131, 99.
Corriu, R. Chem. Rev. 1993, 93, 1371.
Reactivity at Silicon - Parameters
Lewis acidity of Tetracoordinate Si
! Activation by Lewis bases
4-coordinate
bond angle 109.5˚
relatively weakly
Lewis acidic
!
!
5-coordinate
L
R
Si L
R
L
R
bond angles:
eq-ax 90˚
eq-eq 120˚
Si L
L
L
limited to electronegative ligands on Si and/or strongly Lewis basic Nu
Lewis base catalyzed activation- Denmark
! Activation driven by 'strain release Lewis acidity'
! Cyclobutane 'strained' silacycles – Myers (1992), Denmark (1993)
bond angle 90˚
'strained cycle'
relatively strongly
Lewis acidic
Si L
L
!
bond angles:
eq-ax 90˚
eq-eq 120˚
'strain released'
L
Si
L
L
Reactive driving force = release of ring strain
! Cyclocycles 'strained' with diamine ligands – Leighton (late 1990s–present)
bond angle ~90˚
long Si-N bonds
'strained cycle'
relatively strongly
Lewis acidic
NR
NR
RN
Si L
L
RN
Si
L
L
L
bond angles:
eq-ax 90˚
eq-eq 120˚
'strain released'
! Long Si–N bonds distort the 5-membered ring – strained cycle
Hypervalent Silicon
historical perspective
! Silicon compounds with a coordination number higher than four have been known since the early
19th century
2–
F
F
F
Si
F
NH2
F
F
Si
F
F
NH2
F
F
2–
Gay-Lussac, J. L.; Thenard, L. J. Mémoires de Physique et de Chimie de la Société d'Arcueil, 1809, 2, 317.
Davy, J. Phil. Trans. Roy. London, 1812, 102, 352.
! Mechanism of fluoride deprotection of silyl ethers
Me
O
Si
Me
Me
Me
TBAF
F NBu4
O
Si
F
Me
Me
work–up
NBu4
–Me3SiF
pentavalent Si
! Driving force: the strength of Si–F bond (135 kcal/mol)
OH
Lewis Base Activation : Hydride Transfer
racemic hydrosilation
! Hydrosilation – Corriu (1981)
HSi(OEt)3 (1.0 equiv.)
CsF (1.0 equiv.)
O
Ph
Me
OH
Ph
neat, 0 ˚C
80%
Me
! Probable mechanism
O
H
Lewis base
Si OEt
EtO
OEt
activation
H
EtO
F
EtO
Me
Cs
5-coordinate
good L.A.
F
EtO
Ph
Si OEt
OEt
F
OEt
Si
O
H
Me
Ph
OEt
Cs
EtO
Si OEt
O
F
Me
Cs
work-up
OH
Ph
Ph
6-coordinate
good H– donor
Boyer, J.; Corriu, R. J. P.; Perz, R.; Reye, C. Tetrahedron, 1981, 37, 2165.
Corriu, R. J. P.; Perz, R.; Reye, C. Tetrahedron, 1983, 39, 999.
Me
Hypervalent Silicon : Hydride Transfer
asymmetric hydrosilation
! Hosomi (1988)
HSi(OMe)3 (1.5 equiv.)
cat. (0.4 mol%)
O
Ph
Me
THF, RT
OH
Ph
Me
89%
52% ee
Kohra, S; Hayashida, H.; Tominaga, Y.; Hosomi, A. Tetrahedron Lett., 1988, 29, 89.
N
Li
OLi
catalyst
! Hosomi discovered that lithium alkoxides can act as reversible binding Lewis bases
! Development of highly catalytic processes (0.4 mol%)
! Kagan (1997)
HSi(OMe)3 (1.0 equiv.)
cat. (10 mol%)
O
Ph
Me
TMEDA (2 equiv.), Et2O, 0 ˚C
OH
Ph
Schiffers, R.; Kagan, H. B. Synett., 1997, 1175.
Me
80%
61% ee
OLi
OH
catalyst
! Problem: Product alkoxide can also act as a Lewis base catalyst – alternative low ee reaction
Hypervalent Silicon : Lewis Base Catalyzed Allylation
Seminal Contributions
! Sakurai (1988)
OH
O
O
Ph
H
Me
! In 1987 Corriu, Hosomi and Sakurai all
independently reported racemic Lewis Base
promoted allylation using fluorides and
alkoxides to activate tetravalent silanes.
Ph
O
Si O
O
Me
82%
88:12 dr
5-coordinate
Lewis acidic
E:Z = 88:12
O
Si
Me
O
Li+
O
O
O
Ph
6-coordinate
allyllic nucleophile
Kira, M.; Kobayashi, M..; Sakurai, H. Tetrahedron Lett., 1987, 28, 4081.
Kira, M.; Sato, K.; Sakurai, H. J. Am. Chem. Soc., 1988, 110, 4599.
! Dual activation of both electrophile and nucleophile via a 6-membered closed TS.
X
X
Si X
X-
X X
X Si X
Nu
X X
X Si X
Nu
increasing !"# conjugation
! Coordination of electron rich ligands increases the !"# conjugation. (13C NMR evidence)
Sakurai, H. Synlett, 1989, 1.
Kobayashi's Observation: DMF Promotes Allylations
Seminal Contributions
! Dramatic solvent effect!
R1
CH2Cl2
Et2O
benzene
THF
DMF
CH2Cl2 +
1 equiv DMF
OH
O
R2
H
Yield
Solvent
solvent, 0 ˚C
SiCl3
R1
R2 R3
R3
trace
trace
trace
trace
90%
68%
! Important discovery: Neutral Lewis bases such as DMF can activate trichlorosilanes!!
! Reaction is stereospecific
OH
O
SiCl3
Me
DMF, 0 ˚C
97:3 anti:syn
H
Me
H
97:3 E:Z
OH
O
SiCl3
H
DMF, 0 ˚C
>1:99 anti:syn
H
Me
Me
>1:99 E:Z
! High diastereoselectivity indicative of a rigid closed transition state
Kobayashi, S.; Nishio, K. Tetrahedron Lett. 1993, 34, 3453-3456.
Kobayashi, S.; Nishio, K. J. Org. Chem. 1994, 59, 6620-6628.
Rationale Behind the Kobayashi Allylation
diastereoselective closed transition state
!
29Si
NMR Chemical Shifts of (Z)-crotyltrichloro silane in various solvents
OH
O
SiCl3
H
Ph
solvent, 0 ˚C
H
Ph
Me
Me
97% yield
>99:1 anti:syn
Solvent
Si NMR
CDCl3
CD3CN
C6D6
THF-d8
DMF-d7
HMPA
+8.0
+8.6
+7.9
+8.5
–170
–22
! Kobayashi proposes a 6-coordinate Si, closed transition state
H
O
Ph
Me
H
SiCl3
H
97:3 E:Z
O
DMF
H
Me
N
Me
SiCl3 Me
OH
Ph
O
H Ph
! Dual activation of both electrophile and nucleophile via a 6-membered closed TS.
Kobayashi, S.; Nishio, K. Tetrahedron Lett. 1993, 34, 3453-3456.
Kobayashi, S.; Nishio, K. J. Org. Chem. 1994, 59, 6620-6628.
Me
Denmark Expands the Potential
Phosphoramides
! Re-examined common Lewis Base additives
OH
O
SiCl3
Ph
23 ˚C, 1M
H
Lewis base
solvent
DMF (1 eq)
HMPA (1 eq)
HMPA (1 eq)
HMPA (1 eq)
HMPA (0.1 eq)
HMPA (0.1 eq)
benzene - d6
benzene - d6
CDCl3
acetonitrile - d3
benzene - d6
THF - d8
Ph
conversion
time
yield
83
n/a
63
63
60
80
70 h
n/a
4 min
4 min
46 h
124 h
n/a
77%
85%
86%
n/a
n/a
Me2N
P
Me2N
O
NMe2
Hexamethylphosphoramide
(HMPA)
!
In contrast to Kobayashi, Denmark found that HMPA gave superior results to DMF
!
Reaction rate solvent dependent
!
DMSO (sulfoxide) and pyridine N-oxide were tried, but were not compatible with the reaction
! Can an asymmetric HMPA analogue be developed to catalyze the enantioselective variant?
Denmark, S. E.; Coe, D. M.; Pratt, N. E.; Griedel, B. D. J. Org. Chem. 1994, 59, 6161-6163.
Early Lewis Base Catalyzed Asymmetric Allylations
! Denmark - chiral phosphoramides
OH
O
Ar
H
ee 21-65%
yield 67-81%
t = 6-24 h
23 ˚C, 1M
SiCl3
Ar
catalyst 0.1-1 equiv
Me
Me
N
R
O
Me
N
O
P
P
N
N
R
N
R
N
N
N
Me
Me
O
P
R
1
N R
1
Me R
catalysts
! Assumed a similar associative mechanism of action to DMF
R2N
O
Ph
SiCl3
O
cat.
H
H
Me
NR2
P
SiCl3
OH
NR2
Ph
O
H Ph
Low enantioselectivities, generally high loadings of catalyst only aromatic aldehydes
tolerated – but a huge leap in the right direction
!
Denmark, S. E.; Coe, D. M.; Pratt, N. E.; Griedel, B. D. J. Org. Chem. 1994, 59, 6161-6163.
Early Lewis Base Catalyzed Asymmetric Allylations
Iseki also tried chiral HMPA and DMF
! Iseki (1996) - chiral phosphoramides
OH
O
Ar
R1
H
SiCl3
THF, –60 ˚C
Ar
catalyst 0.1-0.2 equiv
R2
ee 77-88%
yield 72-90%
t = 72-168 h
H
N
O
P
R1, R2 = H, Me
N
N
Pr
Pr
Low enantioselectivities, generally high loadings of catalyst and only
aromatic aldehydes were tolerated
!
Iseki, K. Tetrahedron Lett. 1996, 37, 5149-5150.
Iseki, K. Tetrahedron. 1997, 53, 3513-3526.
! Iseki (1998) - chiral formamides
O
SiCl3
R
Me
OH
H
–78 ˚C, CH2Cl2
R
catalyst 0.2-0.4 equiv
ee 68-95%
yield 51-84%
t = 7-14 days
Ph
N
H
R=alkyl only
!
Me
good to excellent enantioselectivities, only aliphatic aldehydes tolerated
Iseki, K.; Mizuno, S.; Kuroki, Y.; Kobayashi, Y. Tetrahedron Lett. 1998, 39, 2767-2770.
Iseki, K.; Mizuno, S.; Kuroki, Y.; Kobayashi, Y. Tetrahedron 1999, 55, 977-988.
Ph
O
Denmark – Could the Allylation Strategy be Applied to Aldol Reactions?
aldol of trichloro silylketene acetals
! Transferring the concept
A
R4
O
H
R3
R1
Direct Translation?
C
R4
O
H
SiCl3
C
R2
P
A
B
O
R2
O
R3
R1
P
B
C
SiCl3
O
! Investigated HMPA as a reaction additive
O
But
OSiCl3
H
O
O
solvent, –80 ˚C
But
OMe
Lewis base
solvent
none
none
none
toluene - d8
CD2Cl2
THF - d8
HMPA (0.1 eq)
CD2Cl2
conversion
time
18
50
69
120 min
120 min
120 min
100
<3 min
OMe
Me2N
P
Me2N
O
NMe2
Hexamethylphosphoramide
(HMPA)
! HMPA catalyzed the trial reactions, but a potential background reaction was evident
Denmark, S. E.; Winter, S. B. D.; Su, X.; Wong, K. -T. J. Am. Chem. Soc. 1996, 118, 7404-7405.
Enantioselective Catalysis by Phosphoramides
Me
Ph
aldol reactions
N
O
P
Ph
N
N
Me
! Variation in the trichlorosilyl enol ether component
O
SiCl3
H
R
5 mol % catalyst
O
CH2Cl2, –78 ˚C
Ph
O
R
OH
93 - 98% yield
Ph
49 - 87% ee
R = Me, nBu, iBu, iPr, tBu, Ph, CH2OSiMe3tBu
! Variation in the aldehyde component
O
Bu
SiCl3
5 -10 mol% catalyst
O
H
R
CH2Cl2, –78 ˚C
O
Bu
OH
79 - 95% yield
R
84 - 92% ee
R = cinnamyl, a-methylcinnamyl, naphthyl, 4-phenyl-phenyl, cyclohexyl, tBu
! Trichlorosilyl enol ethers are prepared from the corresponding trimethylsilyl enol ethers by treatment
with SiCl4 and catalytic Hg(OAc)2
Denmark, S. E., et al., J. Org. Chem., 1998, 63, 918.
Enantioselective Catalysis by Phosphoramides
Me
Ph
syn and anti aldol reactions
N
O
P
Ph
N
N
Me
! Propionate aldol with various aldehydes
O
SiCl3
Me
Ph
15 mol % catalyst
O
O
OH
+
H
R
CH2Cl2, –78 ˚C
6 - 8 hours
Ph
R
89 - 97% yield
3:1 - 18:1 syn:anti
84 - 96% ee
Me
R = cinnamyl, naphthyl, phenyl, tolyl, crotyl
The uncatalyzed reaction at 0° C is slightly anti selective (2:1)
! Enforced E-enol silane aldol reaction: anti selective
O
SiCl3
10 mol % catalyst
O
+
H
R
CH2Cl2, –78 ˚C
2hours
O
OH
R
94 - 98% yield
61:1 - 99:1 anti:syn
94 - 98% ee
R = cinnamyl, a-methylcinnamyl, naphthyl, phenyl
! The uncatalyzed reaction at 0 ˚C is highly syn selective (5:1 - 49:1)
Denmark, S. E., et al., J. Am. Chem. Soc., 1996, 118, 7404-7406.
Denmark, S. E., et al., J. Am. Chem. Soc., 1997, 119, 2333-2334.
Enantioselective Catalysis by Phosphoramides
preliminary rational for the aldol stereochemistry
! Uncatalyzed Reactions
! E-enolsilane
Cl
H
R
O
O
Cl
Si
Cl
! Z-enolsilane
O
OH
slow
R
Cl
H
Me
R
O
O
uncatalyzed
Cl
Si
O
slow
Cl
uncatalyzed Ph
R
Me
Ph
pseudoboat TS
pentacoordinate Si
pseudoboat TS
pentacoordinate Si
syn aldol
OH
anti aldol
! Catalyzed Reactions - An initial proposal
! E-enolsilane
O
! Z-enolsilane
OH
Cl
R
fast
catalyzed
anti aldol
Me
Ph
Ph
N
N Cl
P
N
R1
O
Si O
O
Me
Cl
R3
H
O
fast
2
RR
catalyzed
Hexacoordinate Si proposed
R
Ph
Me
syn aldol
! Chair-like transition state accounts for observed diastereo- and enantioselectivities
! Precise coordination geometry was not known at this stage
Denmark J. Am. Chem. Soc. 1997, 119, 2333-2334.
OH
Enantioselective Catalysis by Phosphoramides
Intriguing Observations - New TS Required!
! Diastereoselectivity dependent on catalyst loading
Me
Ph
OSiCl3
O
P
O
Ph
O
N
N
O
OH
loading conc. syn:anti yield
OH
R1
R1
Me
R1
H
N
syn aldol
E-enolsilane
anti aldol
10 mol% 0.5 M
5 mol% 0.5 M
2 mol% 0.5 M
1:14
1:10
1:2.4
94%
90%
84%
! Catalyst change give a complete switch in diastereoselectivity
R
Ph
OSiCl3
Ph
E-enolsilane
O
P
O
H
N
R1
N
O
O
OH
N
OH
R1
R1
R
10 mol% cat,
CH2Cl2, –78 ˚C
syn aldol
loading
R
10 mol%
10 mol%
Me
Ph
anti aldol
! Anti-diastereoisomer dominates with less bulky ligands and high concentrations
! Syn-diastereoisomer dominates with more bulky ligands and low concentrations
Denmark J. Am. Chem. Soc. 1998, 120, 12990.
syn:anti yield
1:60
97:1
94%
90%
Enantioselective Catalysis by Phosphoramides
Intriguing Observations - New TS Required!
! A productive experiment
R
Ph
OSiCl3
O
O
P
O
Ph
H
N
R1
N
N
syn aldol
O
N
anti aldol
Ph
Ph
N
P
Ph
R1
R
Me
N
OH
R1
E-enolsilane
Ph
O
OH
O
P
N
Ph
Me
N
N
Ph
! Positive non-linear effect using less bulky catalyst
! A new mechanistic postulate is needed
! Conclusion: 2 pathways operating (a) TS contains 1 molecule of catalyst - syn product
(b) TS contains 2 molecules of catalyst - anti product
! R=Me catalyst operates via mix of (a) and (b).
! R=Ph catalyst operates via mostly (a)
Denmark J. Am. Chem. Soc. 1998, 120, 12990.
Denmark: Asymmetric Lewis Base Catalyzed Aldol Reaction
Refined Mechanism: Dissociative
! Pathway to anti aldol - double coordination
Me
Ph
OSiCl3
Ph
N
R2N
N
R2N
NR2 Cl
Me
R1
R2N
P
R2N
E-enolsilane
+
NR2
O
P
O
H
N
O
O
R1
O
Si O
O
Cl
R
OH
Cl–
R1
R2
R3
H
Hexacoordinate Si proposed
(Octahedral)
Good Enantioselectivity
anti aldol
(92% ee)
! Pathway to syn aldol –!mono coordination
Ph
Ph
OSiCl3
Ph
R1
N
Ph
N
Cl
H
O
P
O
H
N
R
O
O
Si
+
Cl
O
R2N
O
NR2
OH
Cl–
R1
NR2
E-enolsilane
Pentacoordinate Si proposed
(Trigonal bipyramidal)
Poor Enantioselectivity
! Ligand binding forces Cl– dissociation. Experimental evidence:
! Bu4NCl retards the reaction rate - common ion effect
! Bu4NOTf and Bu4NI accelerate the rate by increasing ionic strength
Denmark J. Am. Chem. Soc. 1998, 120, 12990.
syn aldol
(50% ee)
Revisiting the Phosphoramide Asymmetric Allylations
Kinetic Study
! Do multiple mechanisms operate here?
OH
O
SiCl3
Ar
Me
H
N
23 ˚C, 1M
O
P
Ar
N
N
Me
! Positive non-linear relationship with catalyst ee - seen similar indicators in the aldol work!
! Proposal: The 1.77th order in catalyst is due to competing mechanisms involving 1 or 2
prosphoramides on silicon
Enantioselective Allylation Mechanism
Dissociative Mechanism
Me
! Two competing transition states
R2N
R2N
R2N
P
+
NR2
R2N
NR2 Cl
Cl–
O
Si O
O
Cl
Me
Ph
H
Ph
+
Cl
H
O
Cl
Si
O
Ph
H
Hexacoordinate Si proposed
(Octahedral)
N
N
23 ˚C, 1M
SiCl3
O
P
OH
O
Ph
N
R2N
NR2
Cl–
NR2
Pentacoordinate Si proposed
(Trigonal bipyramidal)
! Enantioselectivity is likely lower because the cationic octahedral TS gives the opposite facial
selectivity to the cationic trigonal bipyramidal TS.
! Would a new bidentate phosphoramide avoid the problem of mono-coordination in allyations?
Denmark J. Am. Chem. Soc. 2000, 122, 12021.
Denmark J. Am. Chem. Soc. 2001, 122, 9488.
Lewis Base Catalyzed Asymmetric Allylations
pyridine N-oxides
! Nakajima - chiral N-oxides
OH
O
SiCl3
R
H
R=alkyl only
!
–78 ˚C, CH2Cl2
ee 28-92%
yield 27-91%
t=6h
R
catalyst 0.1 equiv
iPr2NEt (5 equiv)
Me
Me
good to excellent enantioselectivities, only aliphatic aldehydes tolerated
! Proposed transition state model - cationic hexacoordinate
R2
Ar
O
H
+
Cl
R1
R3
Cl
Si
O
N
O
N
Cl–
Me
Me
!
Bidentate chelation to give a 6-coordinate octahedral silicon
!
Kocovsky later made a library of related catalysts and optimized this reaction further
Nakajima, M.; Saito, M.; Shiro, M.; Hashimoto, S.-I. J. Am. Chem. Soc. 1998, 120, 6419-6420
N
N
O
O
Allylation: Solving the Dual Mechanism Problem
bidentate catalysts
! Testing a range of asymmetric bisphosphonamides
OH
O
N
23 ˚C, 1M
SiCl3
Ph
Me
Me
O
P
Ph
H
P
N
N
Me Me
ee
catalyst
ee
a
b
c
d
e
0%
35%
17%
65%
46%
f
g
h
i
56%
18%
87%
67%
N
H
R1
R
H
R = Ph, naph
SiCl3
R2
R1=
H
P
H
N
H
N
H, Me
2
R = H, Me
N
Me Me
N
O
P
N
N
O
P
N
N
O
P
N
Me
5N
N
P
N
nN
N
Me
H
H
57-92% yield
OH
Me
CH2Cl2, –78 ˚C
5 mol% cat.
N
O
g: n=4 h: n=5
i: n=6
f:
! Using the new bisphosphonamide
O
O
Me
H
nN
a: n=2 b: n=3
c: n=4 d: n=5
e: n=6
catalyst
H
N
O
80-94% ee
Ph
R1
R2
Denmark J. Am. Chem. Soc. 2000, 122, 12021.
Denmark J. Am. Chem. Soc. 2001, 122, 9488.
>99:1 diastereomers
H
H
Crossed-Aldol Reactions of Aldehydes
application of the bidentate catalyst
! Z-enolsilane - syn products
O
R
1) 5 mol% cat.
CHCl3, –78 ˚C
OSiCl3
H
Me
2) MeOH
Me
Me
OH
N
OMe
N
R
OMe
ee
anti:syn
yield
Ph
cinnamyl
crotyl
phenyl propargyl
dihydrocinnamyl
81%
42%
5%
7%
8%
2:98
1:99
1:99
2:98
5:95
95%
86%
85%
98%
47%
O
P
N
Me Me
OH
R
O
P
5N
N
N
Me Me
bisphosphoramide catalyst
! High levels of diastereoselectivity
! Generally poor ee
! E-enolsilane - anti products
R
1) 5 mol% cat.
CHCl3, –78 ˚C
OSiCl3
O
H
2) MeOH
Me
OH
OMe
OMe
R
OH
R
ee
anti:syn
yield
Ph
cinnamyl
crotyl
phenyl propargyl
dihydrocinnamyl
59%
26%
52%
76%
66%
99:1
99:1
98:2
98:2
99:1
97%
88%
91%
99%
79%
! High levels of diastereoselectivity
! Low ee
Denmark Angew. Chem. Int. Ed., 2001, 40, 4759.
Limitations of Aldol Methodology to This Point
! Trichlorosilanes are difficult to prepare and handle
1) LDA, –78 ˚C
2) Bu3SnCl
O
Me
Bu3Sn
60%
OMe
1) SiCl4 (6 equiv)
2) distill 20 ˚C
O
OMe
OSiCl3
65%
OMe
Denmark, S. E.; Winter, S. B. D.; Su, X.; Wong, K.-T. J. Am. Chem. Soc. 1996, 118, 7404-7405.
OTMS
Hg(OAc)2 (1 mol%)
R
SiCl4, CH2Cl2
OSiCl3
O
XHg
R
R
~60%
R = iPr, nBu, iBu, tBu, TBSOCH2, Ph, CH2CH
Denmark, S. E.; Winter, S. B. D.; Su, X.; Wong, K.-T. J. Org. Chem. 1998, 63, 9517-9523.
! Possible solution to this problem?
R2N
R2N
O
R2N
P
NR2
SiCl4
NR2
Lewis Base
Poor Lewis acid
+
NR2
P
O
Cl
Cl–
O Si
Cl
R2N P
Cl
R2N NR
2
Qu: Is this a useful Lewis Acid?
Hypervalent Silicon : Chiral Lewis Acids
covalently bonded chiral backbones
! Jorgensen (1998)
O
O
N
10 mol% cat.
O
CD3CN
–40 ˚C, 1 h
95%
Me
O
O
N
Si
NCCD3
O
B(C6F5)4
10% ee
endo:exo = >95:5
catalyst
First example of an asymmetric silicon-based Lewis Acid
Johannsen, M.; Jorgensen, K. A.; Helmchen, G. J. Am.Chem. Soc, 1998, 120, 7636.
! Ghosez (2000)
O
10 mol% cat.
OMe
PhMe
–100 ˚C, 94%
O
OMe
Me
54% ee
endo:exo = 99:1
Mathieu, B.; de Fays, L.; Ghosez, L. Tetrahedron Lett., 2000, 41, 9561.
Me
catalyst
Me
OMe
Si
Me
NTf2
Hypervalent Silicon : Chiral Lewis Acids
non-covalently bonded chiral backbones
! Ihara (2000) - resolution
O
OTBS
3.0 equivs amine
TBSOTf
CH2Cl2, –78 ˚C
H
MeO2C
NR2
Me
N
H
MeO2C
stoich reagent
SiMe2tBu
R
Me
O
Amine needed in stoichiometric quantities due to protonation
Ihara, L. et al. Chem. Commun., 2000, 1739.
! Chiral Lewis Base activation
L
R
Si L
L
poor LA
+LB
–LB
LB
R
Si L
L
L
L = electronegative ligand
(Cl)
reactive LA
(LA/LB complex)
Stoichiometric in Si, but catalytic in chiral Lewis base
Hypervalent Silicon : Chiral Lewis Acids
Chiral Lewis Base activated Lewis Acids
! Denmark (2001–2005)
Me
Me
N
O
Nu
Ph
H
OH
cat. LA*
Ph
N
O
O
P
P
N
Me Me
Nu
5 N
N
N
Me Me
bisphosphoramide catalyst
Cl
Cl
Si Cl
Cl
4-coordinate
poor LA
MeN
(R2N)2P
NMe
O
P(NR2)2
O
Cl Si
Cl
Cl
5-coordinate
good LA
Cl–
MeN
(R2N)2P
NMe
O
P(NR2)2
Cl
O
Si
Cl
O
Cl
Ph
6-coordinate
activated
electrophile
Cl–
Hypervalent Silicon : Chiral Lewis Acids
Me
Me
N
Denmark's bisphosphoramide complex
N
P
P
N
Me Me
5 mol% cat.
SiCl4 (1.1 eq)
Ph
H
Bu3Sn
5 N
N
N
Me Me
bisphosphoramide catalyst
! Denmark (2001) – Asymmetric propargylation
O
O
O
CH2Cl2, –78 ˚C
OH
81%
97% ee
Ph
Denmark, S. E.; Wynn, T. J. Am. Chem. Soc. 2001, 123, 6199.
! Denmark (2002) – Addition of silyl ketene acetals
O
Ph
OTBS
H
OMe
5 mol% cat.
SiCl4 (1.1 eq)
CH2Cl2, –78 ˚C
OH
O
Ph
97%
93% ee
OMe
! Diastereoselective addition of silyl ketene acetals
O
Ph
OTBS
H
OMe
Me
E:Z = 95:5
E:Z = 12:88
5 mol% cat.
SiCl4 (1.1 eq)
CH2Cl2, –78 ˚C
93% yield
73% yield
dr = 99:1
dr = 99:1
OH
O
Ph
OMe
product independent
of acetal geometry
Me
ee = 99%
ee = 99%
Indicative of the expected open (acyclic) transition state
Denmark, S. E.; Wynn, T., Beutner, G. L. J. Am. Chem. Soc. 2002, 124, 13405.
Hypervalent Silicon : Chiral Lewis Acids
Me
Me
N
Aldol diastereoselectivity
N
O
O
P
P
N
Me Me
■ Denmark (2005) – Diastereoconvergent aldol
O
Ph
OTBS
H
OMe
Me
E:Z = 95:5
E:Z = 12:88
N
Me Me
bisphosphoramide catalyst
5 mol% cat.
SiCl4 (1.1 eq)
OH
O
Ph
CH2Cl2, –78 ˚C
93% yield
73% yield
5 N
N
OMe
product independent
of acetal geometry
Me
dr = 99:1
dr = 99:1
ee = 99%
ee = 99%
 Stoichiometric in Lewis acid: Ligand acceleration by catalytic Lewis base
Denmark, S. E.; Beutner, G. L.; Wynn, T.; Eastgate, M. D. J. Am. Chem. Soc., 2005, 127, 3774-3789.
Hypervalent Silicon : Chiral Lewis Acids
Me
Me
N
Denmark's bisphosphoramide complex
N
O
O
P
P
5 N
N
Me Me
N
N
Me Me
bisphosphoramide catalyst
! Denmark (2003) – Addition of silyl enol ethers
O
Ph
1 mol% cat.
SiCl4 (1.1 eq)
OTBS
iPr
H
OH
O
iPr
Ph
CH2Cl2, –78 ˚C
99%
98% ee
Denmark, S. E.; Heemstra J. R. jr. Org. Lett., 2003, 5, 2303.
! Denmark (2003) – Addition of vinylogous enol ether
O
Ph
OTBS
H
Me
OtBu
1 mol% cat.
SiCl4 (1.1 eq)
CH2Cl2, –78 ˚C
O
OH
OtBu
Ph
Me
92%
89% ee
99:1 dr
99:1 !:"
Denmark, S. E.; Beutner, G. L. J. Am. Chem. Soc., 2003, 125, 7800.
! Denmark (2003) – Addition of isocyanides
O
Ph
H
C
N
tBu
1 mol% cat.
SiCl4 (1.1 eq)
CH2Cl2, –78 ˚C
OH
NHtBu
Ph
O
Denmark, S. E.; Fan, Y. J. Am. Chem. Soc., 2003, 125, 7825.
96%
98% ee
Hypervalent Silicon : Chiral Lewis Acids
Denmark's bisphosphoramide complex
Me
Me
N
N
O
P
N
Me Me
■ Denmark (2005) – Calculated structure
O
P
5 N
N
N
Me Me
bisphosphoramide catalyst
benzaldehyde-silyl cation complex ortimized in GAMESS (US) at PM3 level
Denmark, S. E.; Beutner, G. L.; Wynn, T.; Eastgate, M. D. J. Am. Chem. Soc., 2005, 127, 3774-3789.
Hypervalent Silicon : Hydride Transfer
asymmetric hydrosilation
Me
! Kocovsky (2004)
N
Ph
HSi(OMe)3 (1.5 equiv.)
cat. (10 mol%)
Ph
Me
CHCl3, –20 ˚C
HN
Ph
Me
Ph
Me
94%
92% ee
iPr
Me
N
6-coordinate Si
Me
Me
H
catalyst
Cl
O Si H
Cl
O
Me
N
N
H
N
O
O
Cl
H
NH
H-bonding
H
! Works well for limited range of methyl ketones
! TS postulated to involve a 6-coordinate Si
Me
!"! interaction
Me
Me
Proposed activated complex
hexavalent Silicon
Malkov, A. V. Mariani, A.; MacDougall, K. N.; Kocovsky, P.; . Org. Lett., 2004, 6, 2253.
Summary and Conclusions
! Over the past 10 years the power of hypervalent silanes has begun to be harnessed
! Denmark in particular had made a large contribution to the development of this field both
in terms of reaction development and mechanistic rationale
! Problem: Difficult to accurately assess the precise nature, coordination and
conformation of the active catalyst and transition state orientation