This document is downloaded from DR‑NTU (https://dr.ntu.edu.sg)Nanyang Technological University, Singapore.

Phase‑Transfer and Ion‑Pairing Catalysis ofPentanidiums and Bisguanidiniums

Zong, Lili; Tan, Choon‑Hong

2017

Zong, L., & Tan, C.‑H. (2017). Phase‑Transfer and Ion‑Pairing Catalysis of Pentanidiums andBisguanidiniums. Accounts of Chemical Research, 50(4), 842‑856.

https://hdl.handle.net/10356/87360

https://doi.org/10.1021/acs.accounts.6b00604

© 2017 American Chemical Society. This is the author created version of a work that hasbeen peer reviewed and accepted for publication by Accounts of Chemical Research,American Chemical Society. It incorporates referee’s comments but changes resultingfrom the publishing process, such as copyediting, structural formatting, may not bereflected in this document. The published version is available at:[http://dx.doi.org/10.1021/acs.accounts.6b00604].

Downloaded on 08 Jul 2021 05:08:56 SGT

1

Phase transfer and ion pairing catalysis of

pentanidiums and bisguanidiniums

Lili Zong and Choon-Hong Tan*

Division of Chemistry and Biological Chemistry, School of Physical and Mathematical Sciences,

Nanyang Technological University, 21 Nanyang Link, Singapore 637371

CONSPECTUS: Catalysts accelerate biological processes and organic reactions in a controlled

and selective fashion. There are continuing efforts in asymmetric catalysis to develop efficient

catalysts with broad reaction scope and industrial practicability. Amongst the various modes of

asymmetric catalysis, phase transfer catalysis has attracted intense interest due to its ease to scale

up and low catalyst loading. Chiral quaternary ammonium and phosphonium salts are well-

studied classes of chiral phase transfer catalysts and they typically are comprised of sp3-

hybridized quaternary onium salts. In this Account, we describe our recent attempts to develop

N-sp2 hybridized guanidinium-type salts as efficient phase transfer catalysts as well as ion pair

catalysis based on N-sp2 hybridized bisguanidinium-type salts.

The sp2-quaternized ammonium salts, pentanidiums, which contain five nitrogen atoms in

conjugation, displayed remarkable phase transfer catalytic efficiency. We have shown that

pentanidium can catalyze Michael additions of tert-butyl glycinate-benzophenone Schiff base

with various ,-unsaturated acceptors, such as vinyl ketones, acrylates and chalcones in high

2

enantioselectivities. The structurally amendable pentanidium phase transfer catalysts supply

diverse reactivity and selectivity to various other organic transformations, such as α-

hydroxylation of 3-substituted-2-oxindoles, Michael addition of 3-alkyloxindoles with vinyl

sulfone and alkylation reactions of sulfenate anions and dihydrocoumarins. Pentanidium salts are

applicable to enantioselective transformations on a preparative scale at low catalyst loading,

allowing for the synthesis of a broad range of enantiopure compounds. From computational and

experimental results, we also proposed that halogenated pentanidium catalysts participated in

halogen bonding and this contributed to the excellent stereocontrol in alkylation reactions.

Subsequently, we articulated that chiral cations can direct functional anions besides basic anions

in traditional Brønsted basic phase transfer reactions, including metal-centered anions. We

identified dicationic bisguanidinium as an excellent ion pairing catalyst, first demonstrating that

bisguanidinium formed an ion pair with permanganate and directed the anion in enantioselective

dihydroxylation and oxohydroxylation of a,β-unsaturated esters. This initial success led us to

explore chiral cationic ion pairing catalysis as a general mode of catalysis. This mode of catalysis

is at the interphase between organocatalysis, phase transfer catalysis and organometallic

catalysis. We then identified bisguanidinium diphosphatobisperoxotungstate and bisguanidinium

dinuclear oxodiperoxomolybdosulfate ion pairs as the active catalysts in enantioselective

sulfoxidations using aqueous H2O2 as oxidant. The structure of bisguanidinium dinuclear

oxodiperoxomolybdosulfate ion pair was elucidated using single crystal X-ray analysis.

Bisguanidinium-catalyzed sulfoxidations emerged as a practical methodology for the synthesis of

enantioenriched sulfoxides including armodafinil and lansoprazole, which are commercial drugs.

Finally, we are also able to show that pentanidium and bisguanidinium hypervalent silicates are

intermediates in enantioselective alkylations using silylamide as Brønsted probase.

3

1. INTRODUCTION

Guanidine, the functional group on the side chain of arginine, is a strong base and is protonated

over a wide pH range. Thus, in physiological conditions, it usually exists as guanidinium. The

guanidinium motif is known to play crucial roles in substrate recognition at the active site of

enzymes through non-covalent interactions such as hydrogen bonding.1 Interactions between

guanidinium and functional groups such as phosphates, carboxylates and enolates are also of

considerable interest in bioorganic chemistry and host-guest macromolecular chemistry.2 Since

the pioneering works of Chinchilla,3

Lipton4

and Corey5

in the 1990s, numerous exciting

outcomes have been achieved by utilizing chiral guanidines as general Brønsted base catalysts.6

Our group has reported a variety of chiral guanidine-catalyzed asymmetric reactions and these

results have been summarized in several reviews.7 This Account will emphasize our recent

efforts in using guanidinium-type and bisguanidinium-type salts 1, 2 and 3 as phase transfer8 and

ion pairing catalysts (Figure 1).

Figure 1. Guanidinium-type salts as phase transfer and ion pairing catalysts.

2. PENTANIDIUM-CATALZYED ENANTIOSELECTVE PHASE

TRANSFER REACTIONS

In our attempt to enhance the basicity of the guanidine catalyst 1, structurally novel pentanidine

containing five nitrogen atoms in conjugation was designed. The hypothesis was that with

NN

N

N N

N

R1

R2

R1

R2

Ph

PhPh

Ph

Bisguanidinium

R1, R2 = benzyl

N

N

N

R1

Ph

PhN

N

R2

Ph

Ph

Pentanidium

R1, R2 = alkyl, benzyl

R1 R2

ClNH

N

NH

Bicyclic guanidinium

I ClCl1.HI 2 3

4

increased conjugation, it might render pentanidine with increased basicity and less acidic

nucleophiles can be activated. Although the pentanidine did not exhibit enhanced basicity,

serendipitously, we found that its fully alkylated salt, pentanidium salts 2, is an excellent phase

transfer catalyst (Figure 1). We prepared the pentanidium salts 2 from commercially available

chiral diamines and tetra-methylated pentanidium 2a can be obtained in five steps with good

chemical yield. Generally, pentanidium salts 2 can be synthesized through the coupling between

imidazolinium chloride and a guanidine partner under basic conditions (Scheme 1). Single-

crystal X-ray diffraction analysis of pentanidium 2a reveals that the five nitrogen atoms are not

coplanar and the resulting two five-membered rings are in a twisted spatial arrangement (Figure

2).9

Scheme 1. Synthesis of pentanidium salts 2 through imidazolinium chloride.

2 R1, R2

2a R1 = R2 = Me

R1 =

tBu

tBu

R2 =

X = H, F, Cl, Br, IX

tBu

tBu

R1 =

tBu

tBu

R2 =

Cl

2b

2c-g

2h-k

X = Cl, Br, I, OMe

R1 = R2 =

tBu

tBu

X

Imidazolinium chloride

+MeCN reflux

Et3N

N

N

Cl

R1

Ph

Ph

R1

ClHN N

N

R2

Ph

Ph

R2

N

N

N

R1

Ph

PhN

N

R2

Ph

Ph

Pentanidium

R1, R2 = alkyl, benzyl

R1 R2

Cl 2

Guanidine

5

Figure 2. Crystal structure of tetra-methylated pentanidium 2a.

2.1 Pentanidium-catalyzed Michael addition of tert-butylglycinate Schiff base

Scheme 2. Pentanidium 2a-catalyzed Michael addition of Schiff base.

When we investigated the propensity for pentanidium to function as phase transfer catalyst, we

found that pentanidium 2a is an excellent catalyst for the Michael addition of tert-butylglycinate-

benzophenone Schiff base 4.9 Pentanidium 2a displayed excellent reactivity; with only 2.0 mol%

of catalyst, the reaction with methyl vinyl ketone 5a was completed within 3 h, providing the

Cs2CO3 (5.0 equiv.)

mesitylene, -20 oC

NPh2C OtBu

O

+

R2

O

R1

N

CPh2

tBuO2C

HH

4

2a (2.0 mol%)

R1 R2

O

Me

O

N

CO2tBu

Ph2C Et

O

N

CO2tBu

Ph2C n-Bu

O

N

CO2tBu

Ph2C

Ph

O

N

CO2tBu

Ph2C OEt

O

N

CO2tBu

Ph2C OBn

O

N

CO2tBu

Ph2C

ON

CPh2

tBuO2C

HHO

N

CPh2

tBuO2C

HH

5a-l6a-l

6a, 86% yield, 91% ee 6b, 92% yield, 93% ee 6c, 97% yield, 93% ee

F

ON

CPh2

tBuO2C

HH

MeO

ON

CPh2

tBuO2C

HHO

N

CPh2

tBuO2C

HHO

N

CPh2

tBuO2C

HH

6d, 50% yield, 88% ee 6e, 71% yield, 97% ee 6f, 80% yield, 96% ee

6g, 98% yield, 92% ee 6h, 89% yield, 90% ee 6i, 89% yield, 85% ee

6j, 95% yield, 90% ee 6k, 92% yield, 90% ee

S

6l, 71% yield, 87% ee

O

6

adduct 6a in 86% yield and in 91% ee (Scheme 2). The generality of the method was

successfully demonstrated using different Michael addition acceptors such as vinyl ketones,

acrylates and chalcones. It was found that loading of the catalyst could be further lowered to 0.05

mol% for a gram-scale reaction (Scheme 3). The Michael addition between Schiff base 4 and

chalcone 5m proceeded with high stereoselectivity providing the adduct 6m in 90% ee and as a

single diastereomer; it was transformed to pyrrolidine derivative 7 in simple steps.

Scheme 3. Preparative-scale reaction with low catalyst loading.

Figure 3. Pentanidium-catalyzed reactions using prochiral intermediates.

The aforementioned result indicates that chiral pentanidium cation can discriminate efficiently

between the two enantiotopic faces of prochiral ester enolate. Further investigations reveal that

amide enolate, sulfenate anion and cyclic ester enolate are also suitable prochiral nucleophiles

(Figure 3).

NPh2C OtBu

O

+Cs2CO3 (2.5 equiv.)

mesitylene, -20 oC

24 h

2a (0.05 mol%)

5m, Ar = 4-ClC6H4 6m1.16 g, 87% yield dr> 99:1, 90% ee

4 2.5 mmol

NH

Ph

Ar

CO2tBu

7

Ph

O

Ar Ph

O

Ar

N

CPh2

tBuO2C

HH

NPh2C OtBu

O

NO

R2

R1

ester enolate

amide enolate

Hydroxylation using molecular oxygen

Michael addition to vinyl sulfone

Michael addition to a,b-unsaturatedcarbonyl compounds

cyclic ester enolate

O O

R1

Alkylation using alkyl bromides

S

S

O

Alkylation to form sulfoxides

sulfenate anion

a)

c)

b)

d)

7

2.2 Pentanidium-catalyzed -hydroxylation of 3-substituted-2-oxindoles

The amide enolate (Figure 3b) generated from 3-substituted 2-oxindole has been widely

employed as key intermediate for the synthesis of enantioenriched 3,3-disubstituted oxindoles.

These oxindoles are frequently observed as the core structure in natural products and extensively

investigated in drug discovery programs due to their potent antibacterial and anticancer

activities.10

Catalytic enantioselective -oxidation of 3-substituted-2-oxindole is a direct and

attractive approach to access 3-substituted-3-hydroxy-2-oxindole core structures.11

In 2008, Itoh

reported the preparation of -hydroxyoxindoles using a Cinchonidinium catalyst under phase

transfer conditions.12

The product was obtained with moderate enantioselectivity and triethyl

phosphite was required to reduce the peroxide intermediate. Encouraged by the efficiency and

excellent stereocontrol of pentanidium, we investigated the -hydroxylation of 3-substituted-2-

oxindole 7 with air as oxidant (Scheme 4a).13

Pentanidium 2b, bearing bulkier side arms was

developed for this reaction (Scheme 1). Initially, we were perplexed that no reductant such as

triethyl phosphite was required and -hydroxyoxindoles were obtained directly with high

enantioselectivities. Using isotope labelling experiments with 0.55 equivalent of 18

O2, the role of

molecular oxygen was verified (Scheme 4b). Mass spectrometry was employed to characterize

the isolated hydroxyloxindole [18

O]-8a and 84% level of 18

O incorporation was observed.

8

Scheme 4. a) Pentanidium 2b-catalyzed -hydroxylation of 3-substituted-2-oxindoles, b) isotope

labelling experiment and c) effect of the amount of oxygen on chemoselectivity.

We found that the amount of molecular oxygen in the reaction affects the amount of

hydroperoxide oxindole formed (Scheme 4c). When the reaction was conducted with excess

oxidant by using an O2 balloon, hydroperoxide oxindole 9a was obtained in 85% yield with good

enantioselectivity. Hydroperoxide oxindole 9a and hydroxyloxindole 8a were found to have the

same absolute configuration (R). To gain insights into the reaction, racemic hydroperoxide

oxindole 9a was prepared and used as oxidant in the -hydroxylation of oxindole 7a in the

absence of air (Scheme 5a). With 5 mol % of pentanidium 2b and two equivalents of racemic 9a

as oxidant, the -hydroxylated product (R)-8a was achieved with 76% ee, while the remained

hydroperoxide oxindole was determined to be (S)-9a with 51% ee. On the basis of the

experimental results (Scheme 4c and Scheme 5a), we proposed a two-step reaction that includes

a kinetic resolution step (Scheme 5b). In the first step, in the presence of pentanidium 2b and

base, the enolate generated from 3-substituted-2-oxindole 7a, adds to O2 in an enantioselective

fashion to give hydroperoxide oxindole (R)-9a. This is followed by a kinetic resolution step, in

NO

PMB

2b (5 mol%)air (0.5 equiv. O2)

50% aq. KOH

toluene, -60 oC12-96 h

No reductant

R2

RN

O

PMB

R2

R

HO

NO

PMB

MeHO

80% yield, 95% ee

NO

PMB

MeHO

75% yield, 86% ee

NO

PMB

MeHO

80% yield, 98% ee

F MeO

NO

PMB

n-BuHO

72% yield, 85% ee

NO

PMB

HO

78% yield, 91% ee

NO

PMB

HO

78% yield, 94% ee

NO

PMB

2b (5.0 mol%),

0.55 equiv. O2

conditionsMe N

O

PMB

MeHO

2b (1.0 mol%),

excess O2

conditionsN

O

PMB

MeHOO

7a

8a80% yield, 95% ee

a) b)

c)

9a85% yield, 80% ee

NO

PMB

2b (5 mol%),

0.55 equiv.18O250% aq. KOH

toluene, -60 oC72 h, 87% yield

95% ee

Me

NO

PMB

MeHO

7a [18O]-8a

84%18O incorporation

7 major 8

8a 8b 8c

8d 8e 8f

9

which hydroperoxide oxindole (R)-9a is reduced by a second enolate of 3-substituted-2-oxindole

7a, furnishing hydroxyloxindole product (R)-8a with improved enantiopurity.

Scheme 5. a) Racemic hydroperoxide 9a as oxidant and b) proposed reaction pathway involving

a kinetic resolution process.

2.3 Pentanidium-catalyzed Michael addition of 3-alkylsubstituted oxindoles

The direct Michael addition of 3-substituted oxindole to activate alkenes is an attractive

approach to prepare 3,3-disubstitued oxindoles containing a quaternary carbon stereocenter.

Recently, successful attempts using commercially available vinyl sulfones as Michael acceptors

have been reported using organocatalysis.14

However, only limited examples of highly

enantioenriched 3,3-dialkyl oxindoles have been obtained.15

We found that pentanidium 2 are

efficient phase transfer catalysts for Michael addition of 3-alkyloxindoles 7 to phenyl vinyl

sulfone 10 at low catalyst loading (Scheme 6).

NO

PMB

NO

PMB

MeHOO

7a1.0 equiv.

rac-9a2.0 equiv.

toluene, -60 oC72 h, no O2

S = 5

(S,S)-2b (5.0 mol%)50% aq. KOH

NO

PMB

MeHO

(R)-8a76% ee

NO

PMB

MeHOO

(S)-9a51% ee

+

Pentanidium 2b

NO

PMB

Me2b

Me

7a

base

+

NO

PMB

MeHOO

(R)-9a

NO

PMB

Me2b

(S)-9a + (R)-8a

kinetic resolution process

O2

a)

b)

10

Scheme 6. Effect on enantioselectivities using halogenated pentanidiums.

Scheme 7. Pentanidium 2f-catalyzed Michael addition of 3-alkylsubstituted oxindoles.

Pentanidium 2c, with four 3,5-di-tert-butylbenzyl side-arms, was first tested and a moderate

level of enantioselectivity was obtained (Scheme 6).16

Further modification, by installing

different halogens to the two benzyl groups, led to an enhancement in enantioselectivity and a

high level of enantiocontrol was achieved with brominated-pentanidium 2f or iodinated-

NO

PMB

Me

7a

SO2Ph+

10

2c-g (0.25 mol %)

K3PO4 (10 equiv.)

CPME, -60 oCN

O

PMB

Me SO2Ph

R2 =

2c X = H 2d X = F2e X = Cl2f X = Br2g X = I

X

tBu

tBu

N

N

N

R1

Ph

PhN

N

R2

Ph

Ph

2Pentanidium

R1 R2

Cl

R1 = 3,5-(tBu)2C6H3CH2

11a

X ee (%)

7678808390

substituent variation

increasing enantioselectivity

NO

PMB

Me SO2Ph

NO

Bn

Et SO2Ph

NO

PMB

SO2Ph

NO

PMB

SO2PhPh

99% yield, 95% ee

NO

PMB

SO2Ph

NO

PMB

SO2Ph

77% yield, 94% ee

85% yield, 92% ee 78% yield, 98% ee

N

Me

O

SO2PhEtO

O

MeO

99% yield, 90% ee

NO

PG

R1

SO2Ph+

2f (0.25 mol %)

K3PO4(10 equiv.)

CPME/m-xylene(1:2)

-60 oC

R2

NO

PG

R1

R2

SO2Ph

91% yield, 95% ee

99% yield, 99% ee

N

Me

MeON

MeH

CH3

12an analogue of esermethole

11a 11b 11c

11d 11e 11f

7 10 11

11g

11

pentanidium 2g. With 0.25 mol% of pentanidium 2f, the reaction between a series of 3-

alkylsubstituted oxindoles 7 and phenyl vinyl sulfone 10 proceeded in a highly enantioselective

manner to furnish corresponding Michael adducts 11 with high functional group tolerance

(Scheme 7). Gram quantity of oxindole 11g can be easily obtained from scale-up experiments.

The adduct 11g then underwent simple transformations to provide pyrroloindoline derivative 12,

an analogue of the natural product precursor esermethole.17

2.4 Pentanidium-catalyzed alkylation of sulfenate

Sulfenate anion18

is an unstable reaction intermediate but it can be generated in situ; it has a

nucleophilic sulfur center, which can be used for the construction of S-C bond. This intermediate

can be used to prepare sulfoxide in a complementary way to the classical Andersen method and

direct sulfoxidation method. There are a few methods to generate sulfenate anion in situ,

including the use of base, fluoride or heat (Scheme 8).19

Perrio and co-workers reported a phase

transfer alkylation of sulfonate using Cinchonidinium catalyst and it provided chiral sulfoxides

with moderate enantioselectivity.20

Using pentanidium 2c as catalyst, we investigated the

enantioselective benzylation of 2-thienyl sulfenate generated in situ from sulfinyl methyl esters

13a (Scheme 9).21

After optimization of reaction parameters, such as sulfenate anion precursors,

solvent, inorganic base and temperature, the enantioselectivities remained moderate. Several

pentanidiums with different steric or electronic patterns were prepared and it was found that

levels of enantioselectivities significantly increases with the introduction of halogen to the

pentanidiums (2h-2j). On the contrary, pentanidium 2k, containing OMe substituted side arms,

results in significant deterioration of enantioselectivity.

12

Scheme 8. Methods for the in situ generation of sulfenate.

Scheme 9. Development of halogenated pentanidiums for sulfenate alkylation.

The generality of this reaction was systematically evaluated with a variety of sulfenate

precursor 13 and alkyl halides using halogenated pentanidiums 2i or 2j. By trapping 2-thienyl

sulfenate with various alkyl bromides, various sulfoxides 14a-i bearing benzyl, alkyl, allylic,

propargylic, 2-naphthylmethyl groups were produced in good enantioselectivities. When thienyl

or benzothienyl sulfenates were utilized, sulfoxides 14j-o were obtained in high yields and

excellent enantioselectivities (Scheme 10). It is noteworthy that the reaction between sulfenate

and m-xylene dibromide furnished bis-sulfoxide 14k in good yield (87%, dl/meso = 4.75:1) and

excellent enantioselectivity. Sulfenates functionalized with furyl, benzofuryl and benzimidazole

moieties were also smoothly converted to their corresponding heterocyclic sulfoxides 14p-s.

RS

O

EWGEWG+

a) Base-mediated retro-Michael elimination

EWG = SO2Ph, CN, COR', CO2R'

RS

O

TMS+

b) Fluoride-mediated retro-Michael elimination

RS

O

+

c) Heat-mediated elimination

RS

Obase

fluoride

heat

RS

O

RS

O

TMSF +

R =

2c X = H 2h X = Cl2i X = Br2j X = I2k X = OMe

X

tBu

tBu

N

N

N

R

Ph

PhN

N

R

Ph

Ph

2Pentanidium

R R

Cl

X ee (%)

6179889025substituent

variation

S

O

CO2Me67 wt% aq. CsOH

CPME, -60 oC

SS S

2c, 2h-2k (1.0 mol %)BnBr (1.2 equiv.)

13a 14a

O

13

Remarkably, benzimidazole sulfoxide 14s, reminiscence to the drug esomeprazole,22

was

obtained in excellent enantioselectivity.

Scheme 10. Halogenated pentanidium 2i or 2j-catalyzed alkylation of sulfenate anion.

Preliminary mechanistic studies on the retro-Michael/alkylation reaction found that the base-

promoted in situ generated sulfenate anion, could undergo sulfur alkylation, oxygen Michael

addition or alkaline hydrolysis, depending on reaction conditions (Figure 4a). When a less active

alkylating reagent, benzyl chloride, was used, only oxa-Michael adduct 15 was obtained even in

the presence of pentanidium 2c. On the contrary, sulfoxide 14a was produced through sulfur

alkylation using iodinated-pentanidium 2j. These experimental studies indicated that iodinated-

RS

O

CO2Me

2i or 2j (1.0 mol %), R'Br or R'I

saturated aq. CsOH, CPME/Et2O (1:3)

-70 oC or -40 oC

RS

R'

O

13 14

S

O

S SMe

O

S S

O

S S

O

S

S

O

S S

O

S S

O

S S

O

S

S

O

S

CF3

S

O

S S

O

S

Me

S

O

S S

O

S

S

O

O

Me

Cl

S

O

O

S

O

O

Cl

S

O

N

N

Me

Me

S

O

S S

O

SMe Me

Me

Me

S

O

S

Me

14a87% yield92% ee

14b65% yield77% ee

14c66% yield81% ee

14d69% yield81% ee

OMe

14e82% yield92% ee

14f77% yield82% ee

14g83% yield94% ee

14h84% yield90% ee

14l85% yield91% ee

14m93% yield95% ee

14n89% yield92% ee

14o83% yield95% ee

14p95% yield81% ee

14q73% yield81% ee

14s99% yield90% ee

Me

MeO

14r83% yield79% ee

14i84% yield92% ee

14j85% yield91% ee

14k87% yield99% ee

Me

14

pentanidium 2j activated and stabilized both the electrophilic alkylating reagents and the

sulfenate.

Figure 4. a) Diverse reaction pathways of sulfenate anion, b) optimized most R-TS and c)

proposed mechanistic model.

Computational studies were performed to obtain the most stable transition state (TS) structure

for the R- and S-sets of the TS at M06/BS1:UFF calculated with ONIOM method in

Gaussian09A2. It was found that the R-TS, which leads to the experimental product 14a with R

configuration, is more stable than the S-TS by 1.2 kcal mol-1

in terms of the Gibbs free energy

(Figure 4b). Theoretical study on stereoselectivity is consistent with experimental observations.

Further analysis of the non-covalent interactions between iodinated-pentanidium 2j and the

substrates revealed that Br-I halogen bonding23

between the leaving Br of benzyl bromide and

iodide on the iodinated-pentanidium 2j, plays a key role to stabilize the TS (Figure 4c).

S

O

S

sulfenate anion

BnCl or BnBr

2j

Alkaline hydrolysis

15

OCO2MeSS

base

2c

CO2MeBnCl,Oxa-Michael addition

Sulfur alkylation to 14a

SS

O

N

IX

d-d+

X= Br, Cl

Proposed Mechanistic Model

Pentanidium 2j

Halogen bond

c)

a) Diverse reaction pathways of sulfenate anion b) Optimized most stable (in terms of G) R-TS

15

2.5 Pentanidium-catalyzed alkylation of dihydrocoumarins

Scheme 11. a) Silylamide BSA as a probase, b) effect of halogenated pentanidiums and c)

identification of silyl ketene acetal 18 as an intermediate.

We found that halogenated pentanidiums 2h-j exhibited excellent stereocontrol in the

alkylation of dihydrocoumarins 16, using silylamide as the probase (Scheme 11).24

In this

strategy, a base with strong basicity but weak nucleophilicity would be generated in situ and

consumed immediately for the formation of an enolate (Scheme 11a).25

The potential

background and side reactions are thus suppressed when employing this transient strong base.

The approach will allow the utilization of base-sensitive substrates and reagents, consequently

expanding the scope of asymmetric phase transfer reactions. For instance, dihydrocoumarins

16a, which will undergo alkaline hydrolysis in the presence of alkali metal alkoxides or

R = 2c X = H 2h X = Cl2i X = Br2j X = I

X

tBu

tBu

N

N

N

R

Ph

PhN

N

R

Ph

Ph

2Pentanidium

R R

Cl

X ee (%)

40919793substituent

variation

O O

Ph

2c, 2h-2j (10 mol %)BnBr (2.0 equiv.)

CsF (4.0 equiv.)

BSA (5.0 equiv.)

THF, -40 oC, 24 hBn

Ph

OO

16a 17a

Si

SiN

O

Bis(trimethylsilyl)acetamide (BSA)

chiral cation catalyst

SiN

O

(CH3)3SiF

[chiral cation]

probase

strong basicity weak nucleophilicity

a)

b)

c)O O

Ph

2i (5.0 mol %), BnBrCsF (2.0 equiv.)

BSA (3.0 equiv.)

THF, -40 oC

83% yield, 95% eeBn

Ph

OO

16a 17a

O OTMS

Ph

LiHMDSTMSCl

THF, 0 oC

2i (5.0 mol %), BnBr

CsF (3.0 equiv.)

THF, -40 oC

55% yield, 91% ee

F

18

16

hydroxides, due to the labile lactone moiety, was smoothly converted to the alkylated product

17a in high yield and excellent enantioselectivity. Silyl ketene acetal 18 was identified as a key

intermediate via NMR analysis of the crude reaction mixture. This was further verified by the

direct benzylation of pre-prepared silyl ketene acetal 18 (Scheme 11c). This reaction is suitable

for lactones including dihydrocoumarins, 3,4-dihydro-2H-benzo[h]chromen-2-ones and 1,2-

dihydro-3H-benzo[f]chromen-3-ones (Scheme 12).

Scheme 12. Halogenated pentanidium 2i-catalyzed enantioselective alkylation of lactones.

3. BISGUANIDINIUM-CATALZYED ION PAIRING

ENANTIOSELECTIVE REACTIONS

In 2013, we found that bicyclic guanidinium salt 1·HI can operate under phase transfer

conditions for the alkylation of 3-substituted 2-oxoindoles (Scheme 13).26

Recently, we

+

O O

Bn

O O

Bn

O

R1

O O

R1

O

R2 R2

O O

Ph

16b-m 17b-m

85% yield, 90% ee

81% yield, 93% ee 78% yield, 94% ee

O O

Ph

84% yield, 90% ee

17c 17d

17f 17g

O O

Bn

85% yield, 87% ee

17b

O

OBn

80% yield, 93% ee

17k

Ph

O O O O

Bn Bn

75% yield, 95% ee 95% yield, 90% ee

17h 17i O O

Bn

85% yield, 96% ee

17jMeO

O OO

OEtBn

87% yield, 86% ee

17e

Br R3

O

O

Ph

Bn

89% yield, 95% ee

17lBr

THF, -40 oC

R3

O

O

Ph

Bn

80% yield, 96% ee

17m

MeO

2i (5.0 mol %)CsF (2.0 equiv.)BSA (3.0 equiv.)

Ph Ph

17

developed dicationic bisguanidinium 3, which was initially designed with the intention to work

with dianions or multiple anions. Bisguanidinium salts27

(BG) 3 features two guanidinium

moieties linked with various spacers (Scheme 14). For ease of synthesis, commercially available

cyclic secondary diamines such as piperazine were examined as spacers between the two chiral

imidazolinium salts. The dicationic bisguanidinium salts 3 were thus achieved in excellent yields

and are amenable to modification. We hope that the dicationic moiety will lead us to discover

new chemistries.

Scheme 13. Guanidinium 1HI-catalyzed phase-transfer alkylation reaction.

Scheme 14. General route to bisguanidinium salts 3 with piperazine as spacer.

In conventional cationic phase transfer catalysis, most reactions are Brønsted base reactions,

which the functional anion is usually either hydroxide or carbonate. Occasionally, other

functional anions such as cyanide (CN) and hypochlorite (ClO

) are used in phase transfer

Strecker or oxidation reactions respectively. An innovative example is the use of (hypo)iodite

NO

Me

Ph

7

Br CO2Me+

19

1.HI (10 mol %)

K2HPO4 (3.0 equiv.)

ZnI2 (0.5 equiv.)

mesitylene, 0 oC, 96 h

NO

Me

Ph CO2MeNH

N

NHI

93% yield94% ee

20a

Imidazolinium chloride

+MeCN reflux

Et3N

N

N

Cl

R2

Ph

Ph

R1

Cl

NN

N

N N

N

R1

R2

R1

R2

Ph

PhPh

Ph

3Bisguanidinium

ClCl

HN

NH

Spacer

(S,S)-3a R1, R2 = CH2Ar

(S,S)-3b R1, R2 = CH2Arp-F

(S,S)-3c R1, R2 = CH2Aro-F(S,S)-3d R1 = CH2Arp-F R2 = CH2Aro-F

tBu

tBu

tBu

tBu

tBu

tBu

F F

Ar Arp-F Aro-F

18

(IO), generated in situ from iodide, by Ishihara for the phase transfer oxidative

cycloetherification.28

However, these reactions are but a subset of a more general strategy of

chiral cationic ion pairing catalysis, which is defined as reactions using a catalyst compose of an

organic chiral cation and an inorganic anionic salt (Figure 5).29

In theory, all functional inorganic

anions are possible counterparts to the chiral cations including organometallic anions. This mode

of catalysis is complementary to the anion-directed catalysis using chiral phosphates.30

Motivated by the desire to exploit more cation-directed anionic inorganic reagents, we began to

systematically investigate metal-centered anions, beginning with metal oxides.

Figure 5. Chiral cationic ion pairing catalysis.

3.1 Bisguanidinium-catalyzed dihydroxylation and oxohydroxylation of -aryl acrylates

Permanganate (MnO4) oxidation of alkenes under phase transfer condition was first reported

using stoichiometric amount of a Cinchonidinium salt.31

Moderate enantioselectivities and low

yields were reported. The lack of catalytic activity of Cinchonidinium phase transfer catalyst is

ascribed to its decomposition under the oxidation conditions. In the early stage of our research,

ChiralCationicIonPairCatalysis

Inorganicbasesi.e.,hydroxide,carbonate

Otherinorganicsaltsi.e.,cyanide,iodide

Metalanionicspeciesi.e.,permanganate,tungstate,molybdate

RadicalGeneratingSalts

i.e.,CAN(futurework)

Organo-

catalysis

PhaseTransferCatalysis Organo-

metallicCatalysis

19

we conducted simple experiments to determine the stability of dicationic bisguanidiniums in the

presence of large excess KMnO4; we found that bisguanidiniums 3a-d (Scheme 14) were

compatible with the oxidation conditions and were recoverable after the reactions. With 2.0

mol% BG (S,S)-3d and 20 wt% aqueous potassium iodide (KI), the oxidation of the ,-

unsaturated ester 21a proceeded smoothly to afford diol 22a with excellent enantioselectivity.32

The yield is moderate as a side product 23a is formed due to C-C cleavage (Scheme 15).

Generally, substrates bearing electron-rich aryl groups results in higher enantioselectivities than

electron-deficient ones. We were surprised that thioether group is well tolerated in this

methodology. We also demonstrated that this methodology is highly selective for ,-

unsaturated esters over simple alkenes.

Scheme 15. Bisguanidinium 3d-catalyzed dihydroxylation of -aryl acrylates.

Trisubstituted enoates were also investigated for the permanganate-mediated alkene oxidation

(Scheme 16). With 2.0 mol% of BG (S,S)-3c, a mixture of Z,E-trisubstituted enoates 24a

provided two diastereoisomers, diols 25a and 25a’ in high enantioselectivities with a significant

Ar COOtBu

2.0 mol% (S,S)-3d

KMnO4 (1.5 equiv.)

20 wt% aq. KI

TBME, -60 oC or -70 oC

Ar COOtBu

OH

HO

Ar COOtBu

O

21a-k 22a-k 23a-k

+

Ph

Me

MeO

Me

Me

Et

O

O

MeS

22k60% yield92% ee

22a65% yield92% ee

22c72% yield85% ee

22e67% yield89% ee

22f64% yield90% ee

22j71% yield86% ee

22i63% yield86% ee

22h71% yield86% ee

MeO

22d65% yield96% ee

MeO

OMe

22j63% yield94% ee

Me 22b62% yield90% ee

20

amount of the cleavage product observed. In an attempt to improve the yield of the desired diols,

various additives were examined. When the pH of the reaction was lowered through the addition

of acetic acid, 2-hydroxy-3-oxocarboxylic ester 26a was unexpectedly obtained exclusively in

high yield and high enantioselectivity (Scheme 16). The observed high enantioselectivity

indicates that both Z- and E-trisubstituted enoates 24a were transformed to the same enantiomer

26a under these conditions. The reaction pathways to diols and C-C cleavage products were

efficiently suppressed and we named this transformation, oxohydroxylation. Substrates bearing

alkyl, thienyl, furyl, alkene moieties are well tolerated and a series of 2-hydroxy-3-oxocarboxylic

esters 26a-n with high levels of enantioselectivities were achieved. Enantioenriched -

hydroxymalonates 26o-p were prepared using methoxy-substituted enoates E-24o-p as starting

materials.

COOtBu

O

OMe

OH

COOtBu

O

OMe

OH

COOtBu

O

OMe

OH

COOtBu

O

OMe

OH

COOtBu

nC5H11 O

OMe

OH

COOtBu

O

OMe

OH

COOtBu

O

OMe

OHi-Pr

COOtBu

nC11H23 O

OMe

OH

COOtBu

O

OMe

OH

BnO

COOtBu

O

OMe

OH

COOtBu

MeO O

OH

COOtBu

MeO O

OMe

OH

COOtBu

O

OMe

OHMeO

COOtBu

O

OH

S

Ar COOtBu

2.0 mol% (S,S)-3c

KMnO4 (1.5 equiv.)

AcOH (3.5 equiv.)

TBME/H2O (20:1)

-60 oC

R

Ar COOtBu

OR

OH

2426

R = alkyl (Z, E mixtures)R = OMe (E isomer)

26a85% yield92% ee

26b92% yield93% ee

26c79% yield92% ee

26d75% yield92% ee

26e98% yield94% ee

26f93% yield91% ee

26g85% yield93% ee

26h88% yield94% ee

26i73% yield91% ee

26j77% yield91% ee

26k49% yield96% ee

26l77% yield84% ee

26o88% yield74% ee

26p87% yield97% ee

R = alkyl, OMe

COOtBu

O

OHO

O26n

81% yield94% ee

COOtBu

O

OH

O 26m47% yield87% ee

Ar COOtBu

OHR

OH

25Not detected

21

Scheme 16. Bisguanidinium 3c-catalyzed oxohydroxylation of trisubstituted enoates.

Scheme 17. a) Selectivity between TBA+MnO4

− and KMnO4, b) cyclic manganate diester 27 as

an intermediate and c) proposed ion pairing working model.

A control reaction was attempted using 1.5 equivalent of tetrabutylammonium permanganate,

an oxidant soluble in organic solvent (Scheme 17a). The reaction provided high level of chiral

induction, which led us to propose that rate acceleration through ion pairing rather than the phase

transfer step is crucial for asymmetric induction. Cyclic manganate diester 2733

is widely

recognized as the common intermediate in permanganate reaction and we proposed that it is

Ph COOtBu

(S,S)-3d (2.0 mol%) 1.5 equiv. oxidant

TBME/20 wt% aq. KI

-60 oC

Ph COOtBu

OH

HO

83% ee with TBA+MnO4-

92% ee with KMnO4

a)

R CHOAr COOtBu

OC-C Cleavage

ArO

MnO

COOtBu

O O

H

R

COOtBu

Ar

OH

HOCOOtBu

Ar

OH

R

OR

Dihydroxylation

H+

Oxohydroxylation

22 or 25 26

27

+

b)

21a

c)

MnO O

O O

Mn OO

O O-

COOtBu

Ar

R

Mn OO

O O-

Ar

COOtBu

R

MnO O

O O

Ar

-O

OtBuR COOtBu

Ar

R

OMn

O

O O-

BG2+

BG2+

accelerated pathway

MnO O

O O

ArO-

OtBu

R

BG2+

low energy TS I

high energy TS II

BG2+

COOtBu

Ar

R

OMn

O

O O-BG2+

KMnO4BG2+

phase transfer

Re

COOtBu

Ar

R

E, Z

Si

oxidationproducts

Ar

COOtBu

R

E, Z

22a

´

22

through this intermediate that different products are obtained, via diverse pathways (Scheme

17b). We found that the isolated diols 25 were not further oxidized to 26 under oxohydroxylation

conditions, indicating that the diols are not intermediates in oxohydroxylation.

The absolute configuration of the α-chiral center of oxidation products 26, derived from both

Z- and E-trisubstituted enoates 24, is identical. On the basis of Lee’s prior work34

and our

experimental results, a working model of reaction acceleration through the formation of an

intimate ion pair between bisguanidinium and enolate anion in the transition states was proposed

(Scheme 17c). The results reveal that rate acceleration is mainly attributed to transition state

stabilization through strong electrostatic interaction between dicationic BG and enolate anion.

The disclosure of this catalytic asymmetric permanganate oxidation opens up new paradigms for

chiral cation-directed catalysis. Exploration of other functional anions will undoubtedly lead to

the discovery of interesting new transformations for asymmetric synthesis.

3.2 Bisguanidinium diphosphatobisperoxotungstate-catalyzed sulfoxidation

Peroxotungstate species have been reported to serve as catalyst for oxidation reactions such as

epoxidation35

and sulfoxidation.36

Since peroxotungstate are anionic species, we hypothesized

that their reactions can be accelerated and modulated using chiral cations. We found that with 2.0

mol% of BG (S,S)-3a and 2.0 mol% of Ag2WO4, oxidation of heterocyclic sulfides 28a went

smoothly to produce sulfoxides 29a with excellent enantioselectivities (Scheme 18).37

Notably,

the amount of additive, NaH2PO4 or NH4H2PO4 is crucial to achieve high yield and high

stereoinduction. A series of enantioenriched sulfoxides 29a-q bearing benzimidazole,

benzothiazole, pyridine, thiophene moieties were obtained, which are of potential interest for

23

drug development. The practical utility was further illustrated in the preparation of proton-pump

inhibitor (S)-Lansoprazole 29l.

Scheme 18. Bisguanidinium diphosphatobisperoxotungstate-catalyzed sulfoxidation.

Experimentally, we found that a ratio of greater than 2 equivalent of dihydrogen phosphate

(H2PO4−) with respect to tungstate is essential for achieving high enantioselectivity. Raman

spectra of the active catalyst was obtained (Figure 6) and an experimentally peak observed at 711

cm-1

corresponds well to a peak at 713 cm-1

in the computed Raman spectra of

28 29

(S,S)-3a (2.0 mol%)

Ag2WO4 (2.0 mol% or 5.0 mol%)

NaH2PO4 or NH4H2PO4 (10 mol%)

35%wt H2O2 (1.05 equiv.)

solvent, 0 oC

N

N S

Me

29a96% yield92% ee

N

N S

MeCl

29c81% yield97% ee

N

N S

MeF

29d78%yield 89%ee

N

N S

MeNO2

29e74%yield88%ee

N

N S

Me

29f 95%yield

99%ee

N

N S

Me

29g 84%yield

92%ee

N

N S

Me

29h72%yield 95%ee

O

O

N

N S

Me

29i71%yield82%ee

N

N S CO2Et

Me

29j 92%yield 90%ee

N

N S CO2tBu

Me

29k 86%yield

80%ee

CN

ArS R

ArS R

N

S S

29m78%yield 93%ee

N

S S

29o85%yield92%ee

N

S S

29n70%yield 91%ee

S

29p76%yield 80%ee

S S

29q53%yield94%ee

N

O

N

HN S

29l81% yield 90% ee

N

O

CF3

S

29r79%yield90%ee

N

N S

Me

29b94% yield92% ee

C6F5

Me

O

OOO

O O O

OOO

O O O

OOO

O O

24

[{PO2(OH)2}2{WO(O2)2}]2-

. This peak is attributed to the twisting of P-O-H group in the

phosphate ligand. A more thorough analysis of the [{PO2(OH)2}2{WO(O2)2}]2-

revealed that an

intramolecular hydrogen bond interaction between the two phosphate ligands is important for

this peak; absence of such interaction will shift the vibrational frequency towards 800 cm-1

. We

thus proposed that the active anion is diphosphatobisperoxotungstate

[{PO2(OH)2}2{WO(O2)2}]2-

(Scheme 19a). Based on both computational and experimental

results, a simple working model is proposed detailing the formation of

diphosphatobisperoxotungstate and its transport from the aqueous to the organic phase (Scheme

19b).

Figure 6. Experimental Raman spectrum of [{PO2(OH)2}2{WO(O2)2}]2-

.

25

Scheme 19. a) Diphosphatobisperoxotungstate 30 and b) proposed working model.

3.3 Bisguanidinium dinuclear oxodiperoxomolybdosulfate-catalyzed sulfoxidation

Similar to peroxotungstates, peroxomolybdates38

are well-known catalysts for oxidation

reactions.39

However, they are typically complex mixtures and it is widely recognized as the

challenging obstacle for elaborating them into highly enantioselective catalyst. In the attempt to

prepare enantiopure sulfoxides of 2-sulfinyl esters through alkylation of sulfenate anions using

-halogenated carboxylate, an unsatisfactory outcome was observed. Thus, direct sulfoxidation

strategy was adopted to gain access to 2-sulfinyl ester sulfoxides 31a, which can be further

elaborated to a commercial drug armodafinil. We found that with 2.5 mol % Na2MoO4·2H2O and

potassium hydrogen sulfate (KHSO4), the oxidation proceeded efficiently with 1.0 mol% of BG

(S,S)-3a and H2O2 as oxidant (Scheme 20a).40

Its practical utility was successfully demonstrated

in the gram-scale synthesis of (R)-modafinil (armodafinil) using a 0.25 mol% loading of (R,R)-

3a (Scheme 20b).

BG2+

Ag2WO4

organic phase

water phase[{PO2(OH)2}2{WO(O2)2}]2-

BG2+2Cl-

W

OP(O)(OH)2

O

OP(O)(OH)2O

O

O O

2-Ar

S

R

ArS R

O

BG2+[{PO2(OH)2}2{WO(O2)2}]2-

W

OP(O)(OH)2

O

OP(O)(OH)2O

O

O

2-

H2O2

W

O

OO

O

O

(HO)2(O)PO

OP(O)(OH)2

30Diphosphatobisperoxotungstate

2-

[{PO2(OH)2}2{WO(O2)2}]2-

+ H2O2 + NaH2PO4 (aq.)

a) b)

26

Scheme 20. a) Catalytic molybdate-mediated sulfoxidation and b) preparative scale synthesis of

armodafinil.

The reactive catalytic species can be prepared by mimicking the reaction conditions in the

absence of sulfide substrate (Scheme 21). The active catalyst was collected as a pale-yellow

precipitate through filtration and the structure of the catalyst, (R,R)-3a-[Mo], was confirmed

using X-ray analysis (Figure 7)41 and 95Mo NMR.42 The achiral anionic molybdenum species

[(2-SO4){Mo2O2(2-O2)2(O2)2}]2-

is revealed by X-ray crystallography to be embedded within

the chiral cavity formed by two side arms of the chiral bisguanidinium dication.

Scheme 21. Identification of the active catalyst.

S CO2MePh

Ph

SPh

Ph

(R,R)-3a (0.25 mol%)Na2MoO4.2H2O (2.5 mol%)

ent-31a91% yield91% ee

35% aq. H2O2 (1.05 equiv.)

KHSO4 (0.5 equiv.)

nBu2O (0.05 M), rt, 8 h

30a5 mmol

MeOH (2 M)rt, 24 h

NH3(10.0 equiv.)

Armodafinil, 1.19 g95% yield, 91% ee

32

CO2Me

O

SPh

Ph

O

NH2

Ob)

S CO2MePh

Ph

SPh

Ph

(S,S)-3a (0.25 mol%)Na2MoO4.2H2O (2.5 mol%)

31a99% yield94% ee

35% aq. H2O2 (1.05 equiv.)

KHSO4 (0.5 equiv.)

iPr2O (0.05 M), 0 oC, 1 h

30a0.2 mmol

CO2Me

Oa)

(R,R)-3a

BG2+[2Cl]2-

(1.0 mol%)

0.04 mmol

(R,R)-3a-[Mo]

BG2+[(m2-SO4){Mo2O2(m2-O2)2(O2)2}]2-

91% yield

Na2MoO4·2H2O (2.5 mol%), 35% aq. H2O2 (1.0 equiv.)

KHSO4 (0.5 equiv.) or H2SO4 (0.25 equiv.)Et2O (2 mL), rt, 2 h

27

Figure 7. X-Ray crystallographic structure of [BG]2+

[(2-SO4){Mo2O2(2-O2)2(O2)2}]2-

(R,R)-

3a-[Mo](ellipsoids at 50% probability)

The catalyst (R,R)-3a-[Mo] displayed excellent catalytic activity in sulfoxidation with H2O2 as

oxidant (Scheme 22). (R,R)-3a-[Mo] was used directly as a stoichiometric oxidant and it

provided sulfoxide ent-31a in 90% yield and 80% ee. This result confirm that (R,R)-3a-[Mo] is

the actual oxidant and high level of enantiodiscrimination can be achieved. However, using just

0.25 equivalent of (R,R)-3a-[Mo] led to the formation of ent-31a in 50% yield with 31% ee;

demonstrating that two out of four peroxo moieties on (R,R)-3a-[Mo] are active oxygen donors

as two equivalents of active oxygen from (R,R)-3a-[Mo] are transferred to the sulfides. On the

basis of the experimental results, a possible mechanistic model was proposed (Scheme 23).43

It is

proposed that the second oxygen transfer is slower and less enantioselective than the first. In the

presence of H2O2, the dimeric structure of the catalyst will be maintained and it is this structure

that provides the highly effective enantiofacial discrimination.

NN

N

N N

N

R1

R2

R1

R2

Ph

PhPh

Ph

OS

O

O O

Mo

OO

Mo

O

O

O

OO

OO

O

[(m2-SO4){Mo2O2(m2-O2)2(O2)2}]2-

(R,R)-3a-[Mo]

2-

[Mo]2-

[Mo]2- =

S CO2MePh

Ph

SPh

Ph

(R,R)-3a-[Mo] (x equiv.)

35% aq. H2O2 (y equiv.)

ent-31aiPr2O, rt, 8 h30a

CO2Me

O

x = 0.01, y = 1.05 x = 1.0, y = 0 x = 0.25, y = 0

yield 95%, ee 91%yield 90%, ee 80%yield 50%, ee 31%

28

Scheme 22. Mechanistic studies by using [BG]2+

[(-SO4)Mo2O2(-O2)2(O2)2]2(R,R)-3a-[Mo].

Scheme 23. Proposed catalytic cycle of bisguanidinium dinuclear oxodiperoxomolybdosulfate.

3.4 Bisguanidinium-catalyzed alkylation of cyclic ketones

Previously, we used the silylamide probase strategy with pentanidium as catalyst, to achieve

the highly enantioselective alkylation of lactones24

. For cyclic and linear ketones, we found that

bisguanidiniums are more suitable. The probase can act as a silylation reagent to generate silyl

enol ether, which is a key intermediate for the alkylation. With 10 mol% of BG (S,S)-3a and 2

equivalent of probase, bis(tert-butyldimethylsilyl)acetamide (BTBSA), a variety of -benzyl-1-

indanones and -benzyl-tetralones 32a-i can be smoothly converted to alkylation product 33a-i

in high yields and enantioselectivities (Scheme 24a).24

TBS enol ether 34 was prepared from -

benzyl-1-indanone 32a and submitted to the alkylation condition (Scheme 24b). The

enantioselectivity and yield obtained were similar to the conditions using probase directly,

demonstrating that TBS enol ether is an intermediate in the reaction. For the alkylation of linear

BG2+

BG2+

H2O2

sulfide

sulfoxide

BG2+

sulfidesulfoxide

first oxygen transfer ishighly enantioselective

second oxygen is less enantioselective

[O]

[O]

(R,R)-3a-[Mo]

formation of B was inhibited in the presence of

excess H2O2

OS

O

O O

Mo

OO

Mo

O

O

O

OO

OO

O

2-

OS

O

O O

Mo

OO

Mo

O

OO

OO

O

2-

O

OS

O

O O

Mo

OO

Mo

O

OO

O

2-

O

O

BA

29

ketones, such as simple phenyl ethyl ketone, due to their low reactivity, the corresponding

alkylation reaction can only be achieved via the preparation of their silyl enol ethers.

Scheme 24. a) Bisguanidinium 3a-catalyzed enantioselective alkylation of cyclic ketones using

probase strategy and b) identification of silyl enol ether as an intermediate.

CsF (5.0 equiv.)

Et2O, -40 oC

O(S,S)-3a (10 mol%)BTBSA (2.0 equiv.)

32a-n 33b-un

n = 1,2

O

Ph

98% yield, 95% ee

33a

O

Ph

99% yield, 89% ee

33b

S

O

Ph

95% yield, 85% ee

33c

O

O

98% yield, 91% ee

33d

BrO

95% yield, 95% ee

33e

S

O

99% yield, 90% ee

33f

CF3

OMe

O

O

Ph

F

99% yield, 92% ee

33i

O

CF3

75% yield, 90% ee

33j

O

87% yield, 92% ee

33k

S

O

O

CF3

87% yield, 77% ee

33lOMe

Br R3

O

Ph

90% yield, 98% ee

33g

F3C

O

Ph

96% yield, 92% ee

33hCl

R2

n = 1,2

R1

OTBS

NTBS

(BTBSA)

O

n

R1R3

b)

a)

O

Ph

PhBrO

Ph

Bn

32a 33a

(S,S)-3a (10 mol%)

CsF(5.0 equiv.)

Et2O, -40 oC

(S,S)-3a (10 mol%)BTBSA (2.0 equiv.)

CsF( 5.0 equiv.)

Et2O, -40 oC yield 98%, ee 95%

PhBrKHMDSTBSCl

34 84% yield, 94% eeTHF, -78 oC

OTBS

Ph

R2

30

Scheme 25. Proposed working model with BG hypervalent silicates ion pair.

Based on the experimental results, a working model for enantioselective alkylation using

silylamide as probase is putatively proposed (Scheme 25). When the probase is activated with

fluoride, BG silylamide is first formed followed by BG enolate A. Silyl enol ether B is

subsequently formed via silylation with another silylamide. We found that the steric features of

the probase affects strongly the enantioselectivities and bis(tert-butyldimethylsilyl)acetamide

(BTBSA) afforded better results than bis(trimethylsilyl)acetamide (BSA). A BG hypervalent

silicates ion pair C is proposed to be the key intermediate, which is responsible for the

enantiofacial differentiation of the incoming electrophiles.

4. CONCLUSION AND PERSPECTIVES

Over the past few years, we have developed N-sp2 hybridized guanidinium-type salts as

efficient phase transfer catalysts. We have shown that these are general catalysts for a variety of

transformations and reactions can be performed at a preparative scale with low loading of

catalyst. In some cases, commercially important compounds are prepared as a demonstration of

Br R3

BG2+[2Cl]2-

probase

N

O

Si

Si

N

O

Si

BG2+

O

Bn

CsF

O

Bn

-BG2+

O

Bn

Si

BG2+[2Cl]2-

CsF

C

O

Bn

Si

F

L

LL

F

BG2+O

Bn

N

O

Si

BG2+

chiral organic base

chiral organic base

A

B

R3

32a

33a

2-

31

industrial applicability. We have also articulated a new mode of catalysis at the interface of

phase transfer catalysis, organocatalysis and organometallic catalysis. We describe the possibility

of activating and modulating organometallic catalysts through ionic interactions and not through

the use of ligands. Chiral cationic ion pairing catalysis is in its infancy and we hope that more

colleagues will find this approach refreshing and join us in exploring and defining its scope.

AUTHOR INFORMATION

Corresponding Author

*E-mail: choonhong@ntu.edu.sg

Notes

The authors declare no competing financial interest.

Biographical Information

Lili Zong was born in Henan (China) in 1985. She received her B.S. in engineering from Harbin

Institute of Technology (China) in 2006. She studied organic chemistry under the supervision of

Professor Yixiang Cheng during her M.Sc. in Nanjing University (China). In 2010, she started

her graduate studies at National University of Singapore under the tutelage of Professor Choon-

Hong Tan. She is currently a postdoctoral fellow in the laboratory of Professor Stefan Matile at

University of Geneva.

Choon-Hong Tan was born in Singapore in 1971. He received his BSc(Hons) First Class from

the National University of Singapore in 1995. He obtained financial support from Trinity College

(Cambridge), the Cambridge Commonwealth Trust and Tan Kah Kee Postgraduate Scholarship

to pursue his PhD and he graduated from the University of Cambridge in 1999. Following that,

he carried out postdoctoral training at Harvard University. He was a Research Associate at

Harvard Medical School before joining the National University of Singapore in 2003 to start his

32

independent career. He joined Nanyang Technological University (Singapore) in 2012 and was

promoted to Full Professor in 2016.

ACKNOWLEDGMENT

We are grateful to the technical and intellectual contributions of our co-workers whose names are

listed in the relevant references. Financial support for the research was provided by National

University of Singapore and Nanyang Technological University (Singapore).

REFERENCES

(1) Riordan, J. F. Arginyl residues and anion binding sites in proteins. Mol. Cell. Biochem.

1979, 26, 71-92.

(2) (a) Müller, G.; Riede, J.; Schmidtchen, F. P. Host-Guest Bonding of Oxoanions to

Guanidinium Anchor Groups. Angew. Chem., Int. Ed. 1988, 27, 1516-1518. (b) Berger, M.;

Schmidtchen, F. P. Zwitterionic Guanidinium Compounds Serve as Electroneutral Anion Hosts.

J. Am. Chem. Soc. 1999, 121, 9986-9993.

(3) Chinchilla, R.; Nájera, C.; Sánchez-Agulló, P. Enantiomerically pure guanidine-catalysed

asymmetric nitroaldol reaction. Tetrahedron: Asymmetry 1994, 5, 1393-1402.

(4) Iyer, M. S.; Gigstad, K. M.; Namdev, N. D.; Lipton, M. Asymmetric Catalysis of the

Strecker Amino Acid Synthesis by a Cyclic Dipeptide. J. Am. Chem. Soc. 1996, 118, 4910-4911.

(5) Corey, E. J.; Grogan, M. J. Enantioselective Synthesis of α-Amino Nitriles from N-

Benzhydryl Imines and HCN with a Chiral Bicyclic Guanidine as Catalyst. Org. Lett. 1999, 1,

157-160.

(6) (a) Terada, M.; Nakano, M.; Ube, H. Axially Chiral Guanidine as Highly Active and

Enantioselective Catalyst for Electrophilic Amination of Unsymmetrically Substituted 1,3-

Dicarbonyl Compounds. J. Am. Chem. Soc. 2006, 128, 16044-16045. (b) Shen, J.; Nguyen, T. T.;

Goh, Y. P.; Ye, W. P.; Fu, X.; Xu, J. Y.; Tan, C.-H. Chiral bicyclic guanidine-catalyzed

enantioselective reactions of anthrones. J. Am. Chem. Soc. 2006, 128, 13692-13693. (c) Fu, X.;

Jiang, Z.; Tan, C.-H. Bicyclic guanidine-catalyzed enantioselective phospha-Michael reaction:

synthesis of chiral beta-aminophosphine oxides and beta-aminophosphines. Chem. Commun.

2007, 5058-5060. (d) Ye, W. P.; Jiang, Z. Y.; Zhao, Y. J.; Goh, S. L. M.; Leow, D.; Soh, Y. T.;

Tan, C.-H. Chiral bicyclic guanidine as a versatile Bronsted base catalyst for the enantioselective

Michael reaction of dithomalonates and beta-keto thioesters. Adv. Synth. Catal. 2007, 349, 2454-

2458. (e) Leow, D. S.; Lin, S. S.; Chittimalla, S. K.; Fu, X.; Tan, C.-H. Enantioselective

protonation catalyzed by a chiral bicyclic guanidine derivative. Angew. Chem., Int. Ed. 2008, 47,

5641-5645. (f) Jiang, Z. Y.; Pan, Y. H.; Zhao, Y. J.; Ma, T.; Lee, R.; Yang, Y. Y.; Huang, K. W.;

Wong, M. W.; Tan, C.-H. Synthesis of a Chiral Quaternary Carbon Center Bearing a Fluorine

Atom: Enantio- and Diastereoselective Guanidine-Catalyzed Addition of Fluorocarbon

Nucleophiles. Angew. Chem., Int. Ed. 2009, 48, 3627-3631. (g) Liu, H. J.; Leow, D.; Huang, K.

W.; Tan, C.-H. Enantioselective Synthesis of Chiral Allenoates by Guanidine-Catalyzed

Isomerization of 3-Alkynoates. J. Am. Chem. Soc. 2009, 131, 7212-7213. (h) Wang, J. M.; Chen,

33

J.; Kee, C. W.; Tan, C.-H. Enantiodivergent and gamma-Selective Asymmetric Allylic

Amination. Angew. Chem., Int. Ed. 2012, 51, 2382-2386.

(7) (a) Leow, D.; Tan, C.-H. Chiral Guanidine Catalyzed Enantioselective Reactions. Chem.

Asian J. 2009, 4, 488-507. (d) Leow, D.; Tan, C.-H. Catalytic Reactions of Chiral Guanidines

and Guanidinium Salts. Synlett 2010, 1589-1605. (e) Wang, C.; Goh, C. M. T.; Xiao, S. H.; Ye,

W. P.; Tan, C.-H. Enantioselective Protonation Catalyzed by Chiral Bronsted Bases. J. Synth.

Org. Chem. Jpn. 2013, 71, 1145-1151.

(8) (a) Shirakawa, S.; Maruoka, K. Recent Developments in Asymmetric Phase-Transfer

Reactions. Angew. Chem., Int. Ed. 2013, 52, 4312-4348. (b) Kaneko, S.; Kumatabara, Y.;

Shirakawa, S. A new generation of chiral phase-transfer catalysts. Org. Biomol. Chem. 2016, 14,

5367-5376. (c) Albanese, D. C. M.; Foschi, F.; Penso, M. Sustainable Oxidations under Phase-

Transfer Catalysis Conditions. Org. Process Res. Dev. 2016, 20, 129-139.

(9) Ma, T.; Fu, X. A.; Kee, C. W.; Zong, L. L.; Pan, Y. H.; Huang, K. W.; Tan, C.-H.

Pentanidium-Catalyzed Enantioselective Phase-Transfer Conjugate Addition Reactions. J. Am.

Chem. Soc. 2011, 133, 2828-2831.

(10) Fonseca, G. O.; Cook, J. M. Modern Methods for Total Synthesis of Important Oxindole

Alkaloids. Org. Chem. Insights 2016, 6, 1-55.

(11) Peddibhotla, S. 3-Substituted-3-hydroxy-2-oxindole, an Emerging New Scaffold for Drug

Discovery with Potential Anti-Cancer and other Biological Activities. Curr. Bioact. Compd.

2009, 5, 20-38.

(12) Sano, D.; Nagata, K.; Itoh, T. Catalytic Asymmetric Hydroxylation of Oxindoles by

Molecular Oxygen Using a Phase-Transfer Catalyst. Org. Lett. 2008, 10, 1593-1595.

(13) Yang, Y.; Moinodeen, F.; Chin, W.; Ma, T.; Jiang, Z.; Tan, C.-H. Pentanidium–Catalyzed

Enantioselective α-Hydroxylation of Oxindoles Using Molecular Oxygen. Org. Lett. 2012, 14,

4762-4765.

(14) Li, H.; Song, J.; Liu, X.; Deng, L. Catalytic Enantioselective C−C Bond Forming

Conjugate Additions with Vinyl Sulfones. J. Am. Chem. Soc. 2005, 127, 8948-8949.

(15) (a) Ohmatsu, K.; Kiyokawa, M.; Ooi, T. Chiral 1,2,3-Triazoliums as New Cationic

Organic Catalysts with Anion-Recognition Ability: Application to Asymmetric Alkylation of

Oxindoles. J. Am. Chem. Soc. 2011, 133, 1307-1309. (b) Zhu, Q.; Lu, Y. Stereocontrolled

Creation of All-Carbon Quaternary Stereocenters by Organocatalytic Conjugate Addition of

Oxindoles to Vinyl Sulfone. Angew. Chem., Int. Ed. 2010, 49, 7753-7756.

(16) Zong, L.; Du, S.; Chin, K. F.; Wang, C.; Tan, C.-H. Enantioselective Synthesis of

Quaternary Carbon Stereocenters: Addition of 3-Substituted Oxindoles to Vinyl Sulfone

Catalyzed by Pentanidiums. Angew. Chem., Int. Ed. 2015, 54, 9390-9393.

(17) Pallavicini, M.; Valoti, E.; Villa, L.; Resta, I. New asymmetric synthesis of (−)-

esermethole. Tetrahedron: Asymmetry 1994, 5, 363-370.

(18) (a) O'Donnell, J. S.; Schwan, A. L. Generation, structure and reactions of sulfenic acid

anions. J. Sulfur Chem. 2004, 25, 183-211. (b) Schwan, A. L.; Söderman, S. C. Discoveries in

Sulfenic Acid Anion Chemistry. Phosphorus, Sulfur Silicon Relat. Elem. 2013, 188, 275-286.

(19) (a) Caupène, C.; Boudou, C.; Perrio, S.; Metzner, P. Remarkably Mild and Simple

Preparation of Sulfenate Anions from β-Sulfinylesters: A New Route to Enantioenriched

Sulfoxides. J. Org. Chem. 2005, 70, 2812-2815. (b) Maitro, G.; Vogel, S.; Sadaoui, M.; Prestat,

G.; Madec, D.; Poli, G. Enantioselective Synthesis of Aryl Sulfoxides via Palladium-Catalyzed

Arylation of Sulfenate Anions. Org. Lett. 2007, 9, 5493-5496. (c) Foucoin, F.; Caupène, C.;

34

Lohier, J.-F.; Sopkova de Oliveira Santos, J.; Perrio, S.; Metzner, P. 2-(Trimethylsilyl)ethyl

Sulfoxides as a Convenient Source of Sulfenate Anions. Synthesis 2007, 2007, 1315-1324.

(20) Gelat, F.; Jayashankaran, J.; Lohier, J.-F.; Gaumont, A.-C.; Perrio, S. Organocatalytic

Asymmetric Synthesis of Sulfoxides from Sulfenic Acid Anions Mediated by a Cinchona-

Derived Phase-Transfer Reagent. Org. Lett. 2011, 13, 3170-3173.

(21) Zong, L.; Ban, X.; Kee, C. W.; Tan, C.-H. Catalytic Enantioselective Alkylation of

Sulfenate Anions to Chiral Heterocyclic Sulfoxides Using Halogenated Pentanidium Salts.

Angew. Chem., Int. Ed. 2014, 53, 11849-11853.

(22) Cotton, H.; Elebring, T.; Larsson, M.; Li, L.; Sörensen, H.; von Unge, S. Asymmetric

synthesis of esomeprazole. Tetrahedron: Asymmetry 2000, 11, 3819-3825.

(23) Gilday, L. C.; Lang, T.; Caballero, A.; Costa, P. J.; Félix, V.; Beer, P. D. Angew. Chem.,

Int. Ed. 2013, 52, 4356-4360.

(24) Teng, B.; Chen, W.; Dong, S.; Kee, C. W.; Gandamana, D. A.; Zong, L.; Tan, C.-H.

Pentanidium- and Bisguanidinium-Catalyzed Enantioselective Alkylations Using Silylamide as

Brønsted Probase. J. Am. Chem. Soc. 2016, 138, 9935-9940.

(25) Inamoto, K.; Okawa, H.; Taneda, H.; Sato, M.; Hirono, Y.; Yonemoto, M.; Kikkawa, S.;

Kondo, Y. Organocatalytic deprotonative functionalization of C(sp2)-H and C(sp

3)-H bonds

using in situ generated onium amide bases. Chem. Commun. 2012, 48, 9771-9773.

(26) Chen, W.; Yang, W.; Yan, L.; Tan, C.-H.; Jiang, Z. Bicyclic guanidinium-catalyzed

enantioselective phase-transfer alkylation: direct access to pyrroloindolines and furoindolines.

Chem. Commun. 2013, 49, 9854-9856.

(27) Fu, X.; Loh, W. T.; Zhang, Y.; Chen, T.; Ma, T.; Liu, H. J.; Wang, J. M.; Tan, C.-H.

Chiral Guanidinium Salt Catalyzed Enantioselective Phospha-Mannich Reactions. Angew.

Chem., Int. Ed. 2009, 48, 7387-7390.

(28) Uyanik, M.; Okamoto, H.; Yasui, T.; Ishihara, K. Quaternary Ammonium (Hypo)iodite

Catalysis for Enantioselective Oxidative Cycloetherification. Science 2010, 328, 1376-1379.

(29) Brak, K.; Jacobsen, E. N. Asymmetric Ion-Pairing Catalysis. Angew. Chem., Int. Ed.

2013, 52, 534-561.

(30) (a) Hamilton, G. L.; Kang, E. J.; Mba, M.; Toste, F. D. A Powerful Chiral Counterion

Strategy for Asymmetric Transition Metal Catalysis. Science 2007, 317, 496-499. (b) Phipps, R.

J.; Hamilton, G. L.; Toste, F. D. The progression of chiral anions from concepts to applications

in asymmetric catalysis. Nat. Chem. 2012, 4, 603-614. (c) Mahlau, M.; List, B. Asymmetric

Counteranion-Directed Catalysis: Concept, Definition, and Applications. Angew. Chem., Int. Ed.

2013, 52, 518-533.

(31) Bhunnoo, R. A.; Hu, Y.; Lainé, D. I.; Brown, R. C. D. An Asymmetric Phase-Transfer

Dihydroxylation Reaction. Angew. Chem., Int. Ed. 2002, 41, 3479-3480.

(32) Wang, C.; Zong, L.; Tan, C.-H. Enantioselective Oxidation of Alkenes with Potassium

Permanganate Catalyzed by Chiral Dicationic Bisguanidinium. J. Am. Chem. Soc. 2015, 137,

10677-10682.

(33) (a) Fatiadi, A. J. The Classical Permanganate Ion: Still a Novel Oxidant in Organic

Chemistry. Synthesis 1987, 1987, 85-127. (b) Dash, S.; Patel, S.; Mishra, B. K. Oxidation by

permanganate: synthetic and mechanistic aspects. Tetrahedron 2009, 65, 707-739.

(34) Lee, D. G.; Brown, K. C. Oxidation of hydrocarbons. 11. Kinetics and mechanism of the

reaction between methyl (E)-cinnamate and quaternary ammonium permanganates. J. Am. Chem.

Soc. 1982, 104, 5076-5081.

35

(35) Sato, K.; Aoki, M.; Ogawa, M.; Hashimoto, T.; Noyori, R. A Practical Method for

Epoxidation of Terminal Olefins with 30% Hydrogen Peroxide under Halide-Free Conditions. J.

Org. Chem. 1996, 61, 8310-8311.

(36) Sato, K.; Hyodo, M.; Aoki, M.; Zheng, X.-Q.; Noyori, R. Oxidation of sulfides to

sulfoxides and sulfones with 30% hydrogen peroxide under organic solvent- and halogen-free

conditions. Tetrahedron 2001, 57, 2469-2476.

(37) Ye, X.; Moeljadi, A. M. P.; Chin, K. F.; Hirao, H.; Zong, L.; Tan, C.-H. Enantioselective

Sulfoxidation Catalyzed by a Bisguanidinium Diphosphatobisperoxotungstate Ion Pair. Angew.

Chem., Int. Ed. 2016, 55, 7101-7105.

(38) Dickman, M. H.; Pope, M. T. Peroxo and Superoxo Complexes of Chromium,

Molybdenum, and Tungsten. Chem. Rev. 1994, 94, 569-584.

(39) (a) Shi, X.; Wei, J. Preparation, characterization and catalytic oxidation properties of bis-

quaternary ammonium peroxotungstates and peroxomolybdates complexes. Appl. Organomet.

Chem. 2007, 21, 172-176. (b) Chakravarthy, R. D.; Ramkumar, V.; Chand, D. K. A molybdenum

based metallomicellar catalyst for controlled and selective sulfoxidation reactions in aqueous

medium. Green Chem. 2014, 16, 2190-2196.

(40) Zong, L.; Wang, C.; Moeljadi, A. M. P.; Ye, X.; Hirao, H.; Tan, C.-H. Bisguanidinium

dinuclear oxodiperoxomolybdosulfate ion pair-catalyzed enantioselective sulfoxidation. Nat.

Commun. 2016, 7, 13455.

(41) Salles, L.; Robert, F.; Semmer, V.; Jeannin, Y.; Bregeault, J. M. Novel di- and trinuclear

oxoperoxosulfato species in molybdenum(VI) and tungsten(VI) chemistry: The key role of pairs

of bridging peroxo groups. Bull. Soc. Chim. Fr. 1996, 133, 319-328.

(42) Nardello, V.; Marko, J.; Vermeersch, G.; Aubry, J. M. 90

Mo NMR and kinetic studies of

peroxomolybdic intermediates involved in the catalytic disproportionation of hydrogen peroxide

by molybdate ions. Inorg. Chem. 1995, 34, 4950-4957.

(43) Thompson, D. J.; Cao, Z.; Judkins, E. C.; Fanwick, P. E.; Ren, T. Peroxo-dimolybdate

catalyst for the oxygenation of organic sulfides by hydrogen peroxide. Inorg. Chim. Acta 2015,

437, 103-109.