Organocatalytic stereoisomerization versus alkene isomerization: Catalytic asymmetric synthesis of 1-hydroxy-trans -2,5-diphenylphospholane 1-oxide

308▌

FEATURE ARTICLE

feature article

Organocatalytic Stereoisomerization versus Alkene Isomerization: Catalytic

Asymmetric Synthesis of 1-Hydroxy-trans-2,5-diphenylphospholane 1-Oxide

Organocatalytic Asymmetric Isomerization

Lukas Hintermann,*^aMarco Schmitz,^aOleg V. Maltsev,^aPanče Naumov^b,c

a

Department Chemie, Technische Universität München, Lichtenbergstr. 4, 85748 Garching bei München, Germany

Fax +49(89)28913669; E-mail: lukas.hintermann@tum.de

b

New York University Abu Dhabi, PO Box 129188, Abu Dhabi, UAE

c

Institute for Chemical Research and the Hakubi Center for Advanced Research, Kyoto University, Uji, Kyoto 611–0011, Japan

Received: 22.09.2012; Accepted after revision: 28.11.2012

reported (vide infra).⁶In contrast to double-bond shifts,

Abstract: The potential for an organocatalytic asymmetric stereo-

stereoisomerization reactions proceed with retention of

isomerization or alkene isomerization as atom-economic reaction

bond connectivity and order. The equilibration of meso

with minimal structural change was investigated. The McCormack

cycloaddition of 1,4-diarylbuta-1,3-dienes with (dialkylamino)di-

and chiro diastereomers [Scheme 1 (a)] is an example.⁷

chlorophosphane and aluminum trichloride gives meso-2,5-diaryl- There is no obvious enthalpic driving force for such iso-

1-(dialkylamino)-1-oxo-2,5-dihydro-1H-phospholes, which were

identified as suitable substrates for asymmetric isomerization to

(1R,5R)-2,5-diaryl-1-(dialkylamino)-1-oxo-4,5-dihydro-1H-phos-

feasible in the absence of a driving force.⁸The situation

pholes in the presence of bifunctional organocatalysts (cinchona al-

kaloids, Takemoto catalyst) in up to 91% ee and quantitative yield.

desmic reactions. Equilibrium mixtures of meso and chiro

isomers are obtained, and asymmetric catalysis is hardly

changes, once the groups X are joined to a small cycle

[Scheme 1 (b)].

The substrate range and the mechanism of the catalysis were stud-

ied. The reaction involves proton abstraction by the base, but a pri-

mary deuterium KIE is absent. Enriched (1R,5R)-1-(diethylamino)-

1-oxo-2,5-diphenyl-4,5-dihydro-1H-phosphole was hydrolyzed to

(5R)-1-hydroxy-1-oxo-2,5-diphenyl-4,5-dihydro-1H-phosphole,

which was hydrogenated diastereoselectively under dissolving met-

al conditions to give (2R,5R)-1-hydroxy-1-oxo-2,5-diphenylphos-

pholane (Fiaud’s acid) in preference to meso-1-hydroxy-1-oxo-2,5-

diphenylphospholane. An asymmetric catalytic total synthesis of

Fiaud’s acid, which is a building block for chiral phospholane syn-

thesis, has been realized in five steps from thiophene, using nickel-

catalyzed Wenkert arylation, McCormack cycloaddition, asymmet-

ric dihydro-1H-phosphole isomerization, hydrolysis, and diastereo-

selective hydrogenation.

a)

X

R

X

R

X

R

meso

chiro

(X)

(Y)

b)

(X)

(Y)

or

R

chiro-trans

meso-cis

Ph

c)

Key words: asymmetric catalysis, isomerization, heterocycles,

H₂N

Ph

NH₂

Ph

phosphorus, stereoselective synthesis

Base

H

N

Ph

amarine

isoamarine

Introduction

Catalytic and stereoselective isomerization reactions with

retention of the carbon skeleton¹are attractive synthetic

tools that can increase molecular complexity with full

atom economy. Examples include alkene isomerizations,²

including rearrangements of allyl ethers or amines to enol

ethers or enamines, respectively.³The rhodium-catalyzed

asymmetric isomerization of nerylamine has gained

prominence as the key step of an industrial synthesis of

(–)-menthol.^3,4The manner of breaking and making C–H

bonds in these reactions is not readily transferred to or-

ganocatalysis. Alternatively, several alkene and alkyne

isomerizations proceed via carbanions.⁵In principle, such

reactions are amenable to asymmetric organocatalysis by

chiral organic bases; indeed, the first examples have been

d)

H⁺

Base

Ph

Me₂N

Ph

P

H₂O

HO

O

Me₂N

O

1

2

3

Scheme 1 Examples of hypothetic or established stereoisomeriza-

tion reactions amenable to asymmetric catalysis

The meso/chiro transformation becomes a cis/trans isom-

erization that profits from releasing the eclipsing strain of

the R groups in several ring systems. This principle is seen

at work in the base-mediated isomerization of amarine to

isoamarine⁹which forms the basis of a stereoselective

synthesis of racemic 1,2-diphenylethylenediamine

[Scheme 1 (c)].¹⁰With the aim of developing an asymmet-

ric catalytic stereoisomerization reaction, we planned to

study organocatalytic meso/chiro or cis/trans isomeriza-

SYNTHESIS 2013, 45, 0308–0325

Advanced online publication: 09.01.2013

0

3

9

-

7

8

1

4

3

7

-

2

1

0

X

DOI: 10.1055/s-0032-1316835; Art ID: SS-2012-T0745-FA

FEATURE ARTICLE

Organocatalytic Asymmetric Isomerization

309

tions. Model reactions of the type shown in Scheme 1 (b) 3 as the key step [Scheme 1 (d)]. The racemic acid 1 can

were defined: meso-heterocycles including acceptor be resolved by crystallization with quinine.¹²Our attempts

groups X/Y should be capable of undergoing chiral base- to catalyze the transformation of 2 to 3 with chiral bases

catalyzed stereoisomerization via deprotonation of acidi- like the cinchona alkaloids were not successful.¹³As re-

fied α-CH bonds. The attractive reaction shown in ported in our communication, we next turned to the more

Scheme 1 (c) appeared unsuitable due to the limits of the activated meso-2,5-dihydro-1H-phosphole 4, a precursor

base strength of typical organocatalysts.¹¹Through a data- of 2,¹²and found its asymmetric organocatalytic double-

base search for suitably activated heterocycles, we found bond isomerization to the chiral 4,5-dihydro-1H-phos-

Fiaud’s synthesis of (±)-1-hydroxy-1-oxo-trans-2,5-di- phole 5 [Scheme 2 (a)].¹⁴Deng and co-workers have re-

phenylphospholane (1) using a base-catalyzed isomeriza- cently realized a related asymmetric isomerization of

tion of meso-2 to chiral trans-1-amino-1-oxophospholane achiral 3-butenolides to enantioenriched 2-butenolides

Biographical Sketches█

Lukas Hintermann was with Keisuke Suzuki. He cipient of the medal of the

born in Switzerland in 1972. started independent re- ETH, the Heisenberg fel-

He studied chemistry at search at RWTH Aachen lowship of the DFG, and the

ETH Zurich, obtaining a University in Germany, Friedrich-Wilhelm-Preis of

Ph.D. in 2000 for work on hosted by Carsten Bolm and the RWTH Aachen. His re-

asymmetric catalytic fluo- supported by an Emmy search interests are centered

rination performed with Noether program of the on developing catalytic re-

Antonio Togni. In 2001– DFG. In 2009, he accepted actions towards solving spe-

2002 he visited the Tokyo his current professorship at cific synthetic problems,

Institute of Technology as a the Technische Universität and for applications in sus-

JSPS postdoctoral fellow München (TUM). He is re- tainable organic synthesis.

Marco Schmitz was born in diploma in 2004. Starting in ceived a Ph.D. degree in

Hamm, Germany, in 1977. 2005, he performed re- 2009. Since 2009, he has

He studied chemistry at search in the group of Prof. been working as Regulatory

RWTH Aachen University, Hintermann on the de- Affairs Manager in the No-

Germany, which he con- rivatization of cinchona al- vartis/Sandoz-Hexal group

cluded with a thesis in tech- kaloids and asymmetric at Salutas Pharma GmbH.

nical chemistry with Prof. catalytic synthesis of phos-

Marcel Liauw, receiving his pholanes, for which he re-

Oleg Maltsev was born in istry Russian Academy of Hintermann at TUM as

Kumertau, Russian Federa- Sciences, where he earned HFSP postdoctoral fellow.

tion, in 1985. He studied his Ph.D. in 2011 with work He is recipient of the Gold

chemistry at D. I. Men- on recoverable chiral ionic Medal of the Russian Acad-

deleyev

University

of liquids as enantioselective emy of Sciences for Best

Chemical Technology of organocatalysts with Prof. Scientific Works of Young

Russia and N. D. Zelinsky Sergey G. Zlotin. In 2011, Scientists in Organic Chem-

Institute of Organic Chem- he joined the group of Prof. istry (2012).

Panče Naumov was born in group leader in the National took on positions at New

Macedonia, in 1975. As a Institute for Materials Sci- York University Abu Dhabi

scholar of the Japanese ence in Japan (2004–2007), and Kyoto University. His

Government, he acquired and associate professor at research interests revolve

his Ph.D. with Prof. Yuji Osaka University (2007– around solid-state and struc-

Ohashi at Tokyo Institute of 2012) where he established tural chemistry, with focus

Technology in 2004. He new laboratories for solid- on photochemistry and reac-

was a research fellow and a state research. In 2012 he tivity.

Synthesis 2013, 45, 308–325

310

L. Hintermann et al.

FEATURE ARTICLE

[Scheme 2 (b)].¹⁵Both our and Deng’s reaction rely on a

prototropic rearrangement, induced by a bifunctional or-

ganocatalyst, where the driving force is the generation of

a conjugated system.

a)

b)

215 °C

10 h

Ph

+

PhPCl₂

Ph

P

Ph

6

1. AlCl₃, DCE

80 °C, 48 h

a)

Ph

19%

P

OH

2. NaHCO₃(aq)

EDTA

O

Ph

N

7

N

H

83% ee

cryst.

Ph

P

Ph

(5–10 mol%)

Scheme 3 McCormack cycloaddition of (E,E)-1,4-diphenylbuta-

1,3-diene (6) with PhPCl₂: (a) a phosphole is obtained under thermal

conditions;¹⁸(b) the use of a Lewis acid allows isolation of meso-2,5-

dihydro-1H-phosphole 7.

P

R₂N

O

R₂N

O

MeCN, 60 °C

95% ee

(R = Et)

meso-4

(–)-(5R)-5

The substrate scope of the aminophosphenium cycloaddi-

tion was extended to various R₂NPCl₂²⁰(Table 1) or diene

structures (Table 2). Either Et₂NPCl₂^20b(entry 2) or a pi-

peridino derivative (entry 3) gave 1-amino-1-oxo-2,5-di-

hydro-1H-phosphole with yields comparable to those

described for Me₂NPCl₂(entry 1).^12bThe recrystallized or

chromatographically purified products were pure

trans,trans-diastereomers, but the crude cycloaddition

mixture of 4b contained the minor cis,cis-diastereomer²¹

in a ratio of trans,trans/cis,cis 16:1.

OH

b)

OH

N

O

N⁺

H

^-O

O

R

O

up to

94% ee

(0.1–20 mol%)

CH₂Cl₂, –20 °C

R

Scheme 2 Examples of organocatalytic, asymmetric double-bond

isomerizations. (a) Hintermann et al.¹⁴(b) Deng et al.¹⁵

Table 1 McCormack Cycloadditions of 1,4-Diphenylbutadiene

Here we present an extended study of the asymmetric cat-

alytic meso-2,5-dihydro-1H-phosphole isomerization

chemistry, including an extension of the substrate range of

asymmetric dihydro-1H-phosphole isomerization and the

introduction of bifunctional thioureas as suitable cata-

lysts. In addition, deuterium labeling has been used to

study the mechanism of the asymmetric reaction. Finally,

a stereoselective hydrogenation of (5R)-1-hydroxy-1-

oxo-2,5-diphenyl-4,5-dihydro-1H-phosphole to (2R,5R)-

1-hydroxy-1-oxo-2,5-diphenylphospholane (1) was

found; incorporating parallel work on Wenkert aryla-

tions,¹⁶we have realized an asymmetric catalytic total

synthesis of Fiaud’s acid 1 from thiophene in only five

steps.

1) AlCl₃, R₂NPCl₂

CH₂Cl₂, 0 °C, 16 h

P

Ph

O

R₂N

2) NaHCO₃(aq)

NTA, 0 °C, 4 h

6

4

Entry

NR₂

Product

Yield (%)

1

2

3

4

NMe₂

NEt₂

4a

4b

74^a

83

76

71

piperidino

4c

Ni-Pr₂

4d

Y

P

Results and Discussion

5

NCy₂

57^b

O

Cy₂N

Synthesis of New 2,5-Diaryl-2,5-dihydro-1H-phos-

pholes by McCormack Cycloaddition

8e (Y = OOH)

9e (Y = OH)

The McCormack cycloaddition¹⁷of (E,E)-1,4-diphenyl-

buta-1,1-diene (6) with dichloro(phenyl)phosphane re-

quires high temperatures, producing 1,2,5-triphenyl-1H-

phosphole [Scheme 3 (a)].¹⁸Milder reaction conditions

were realized by Fiaud and co-workers by choosing the

aminophosphenium cycloaddition variant.¹⁹They reacted

diene 6 with Me₂NPCl₂/AlCl₃to obtain 4a selectively af-

ter hydrolytic workup (Table 1, entry 1).¹²We likewise

added AlCl₃to the reaction of PhPCl₂with 6 and obtained

the desired meso-1-oxo-1,2,5-triphenyl-2,5-dihydro-1H-

phosphole (7), though in low, unoptimized yield [Scheme

3 (b)].

^aData from ref. 12b.

^bMixture of 8e/9e, 89:11.

The sterically demanding i-Pr₂NPCl₂gave 4d reliably

only if the reaction and workup were performed under ar-

gon (entry 4). When the workup was performed in air, a

mixture of the desired 4d with two unexpected side-prod-

ucts 8d and 9d was obtained (Scheme 4). Both 8d and 9d

are products of a diastereoselective autoxidation; stirring

a solution of pure 4d in CH₂Cl₂for prolonged times in air

only returned 9d, but no peroxide 8d. The observed autox-

idation to peroxide may have occurred already in the

McCormack reaction at the stage of the hydrolytically la-

Synthesis 2013, 45, 308–325

FEATURE ARTICLE

Organocatalytic Asymmetric Isomerization

311

bile intermediate (Scheme 4), or under the specific condi-

tions of the reaction workup.

HO

O

HO

O₂(air)

Ph

NR₂

P

+

Ph

reaction

workup

P

Cl

O

R₂N

R = i-Pr

R = Cy

8d

8e

9d

9e

Scheme 4 Generation of autoxidation products in the workup of the

McCormack cycloadditions with bulky R₂NPCl₂

Figure 2 X-ray crystal structure of 4-hydroxy-4,5-dihydro-1H-phos-

phole 9d, determined from a racemic twin

The structures of both 8d and 9d were supported by X-ray

diffraction analysis.²²The crystals of the hydroperoxide

8d (Figure 1) are twinned, but the apparent twining comes

from the two conformations that are present in a 50:50 ra-

tio; crystallographically, 8d is considered a twin, while

chemically it is a racemate.

release of strain present in 4 between the NR₂and cis-phe-

nyl groups by generation of the planarized allylic radical

in the H-abstraction with oxygen. Spontaneous autoxida-

tion of activated allylic C–H bonds leading to isolation of

allylic hydroperoxides or alcohols is well documented.²⁵

Variation of the Aryl Groups

The McCormack cycloaddition scope was also extended

by varying the aryl groups of the diene component. 1,4-

Diarylbuta-1,3-dienes 10 were available by Wenkert ary-

lation of thiophene with ArMgX, according to our opti-

mized protocol.^16,26Cycloaddition of the dienes with

Et₂NPCl₂/AlCl₃proceeds readily if the aryl groups in di-

ene 10 are either devoid of ortho-aryl substituents (Table

2, entries 1–3), or only contain a single ortho substituent

(entries 4 and 5). Cycloaddition of 1,4-bis(1-naphthyl)bu-

ta-1,3-diene (10f) gave the regular product trans,trans-

11f accompanied by the minor cis,cis-diastereomer (entry

6). The most hindered bis-mesityl diene 10g gave low

yields under standard conditions (entry 7).

Figure 1 Two molecules in the asymmetric unit of the crystal of the

peroxide 8d; hydrogen atoms are omitted for clarity. The β angle is

very close to 90°, and the unit cell emulates orthorhombic symmetry.

There are two independent molecules in the asymmetric unit (Z′ = 2,

Z = 4).

There are two molecules in the asymmetric unit, where the

peroxide hydroxyl group of one molecule is strongly hy-

drogen bonded to the P=O oxygen of the other, with an

O···O distance of 2.652 Å. The P=O group of the first mol-

ecule and the peroxide of the second are also hydrogen

bonded (O···O = 2.691 Å) to peroxide and P=O groups of

adjacent molecules, respectively, thereby extending the

packing into infinite linear catemers. The structure of the

alcohol 9d, determined from a racemic twin (Figure 2),

contains a single molecule in the asymmetric unit.

Table 2 Synthesis of 1-Amino-2,5-diaryl-1-oxo-2,5-dihydrophos-

pholes

1) AlCl₃, Et₂NPCl₂

CH₂Cl₂, 0 °C, 16 h

Ar

Et₂N

Ar

P

Ar

O

2) NaHCO₃(aq)

NTA, 0 °C, 4 h

10

11

Entry

Ar

Product

Yield (%)

Similar to 8d, the hydroxyl group in 9d donates a proton

(O···O = 2.746 Å) to a P=O group of an adjacent molecule

with opposite disposition, but the P=O group also receives

a proton from the same molecule, thereby forming centro-

1

2

3

4

5

6

7

4-MeC₆H₄

11a

79

4-t-BuC₆H₄

11b

70

symmetric R²(12) dimers. The isopropyl group is disor-

4-MeO-3,5-t-Bu₂C₆H₂

2-MeC₆H₄

11c

74

2

dered even at 120 K. Structurally related 4,5-dihydro-1H-

11d

84

phosphol-4-ols have been described.^23,24

2-EtC₆H₄

11e

63

Attempted synthesis of 2,5-dihydro-1H-phosphole 4e

from Cy₂NPCl₂returned mainly hydroperoxide 8e as mix-

ture with alcohol 9e (Table 1, entry 5; Scheme 4). The sen-

sitivity towards autoxidation apparently increases with

the bulk of the NR₂group. This could be due to favorable

1-naphthyl

11f + 11f′^a

11g

43 + 7^a

24

2,4,6-Me₃C₆H₂

^a11f′ is the cis,cis-diastereomer.

Synthesis 2013, 45, 308–325

312

L. Hintermann et al.

FEATURE ARTICLE

Table 4 Solvent Effect on Asymmetric Olefin Isomerization^a

Asymmetric Catalytic Isomerization

Evaluation of the Reaction Parameters

cinchonine

Ph

NEt₂

P

solvent, 50 °C, 16–24 h

O

Addition of an amine base to a solution of meso-1-amino-

2,5-diaryl-1-oxo-2,5-dihydro-1H-phospholes 4 or 11 in-

duces isomerization to the more stable conjugated 4,5-di-

hydro-1H-phospholes 5 or 12.^14,27Chiral bases were

screened in CDCl₃solution for their efficiency and enan-

tioselectivity in the reaction (Table 3). The use of CDCl₃

as solvent allowed for simple ¹H and ³¹P NMR analysis of

the conversion and enantioselectivity, the latter after addi-

tion of quinine as chiral shift reagent (vide infra).

NEt₂

4b

5b

Entry

Solvent

Cat. (mol%) Conv. (%)

ee (%)

1

CH₂Cl₂

THF

30

20

30

20

10

91

64 (5R)

55 (5R)

62 (5R)

59 (5R)

39 (5R)

44 (5R)

61 (5R)

72 (5R)

74 (5R)

73 (5R)

70 (5R)

72 (5R)

74 (5R)

2

100

94

3

EtOAc

acetone

t-AmOH

toluene

o-C₆H₄Cl₂

MeCN

EtCN

4

100

46

5

Table 3 Screening of Chiral Bases in the Isomerization of meso-Di-

hydro-1H-phosphole 4b^a

6

98

catalyst (10 mol%)

7

100

90

Ph

NEt₂

P

70 °C, CDCl₃, 16 h

O

NEt₂

8

4b

5b

9

Entry

Catalyst

Conv.^b(%)

ee^c(%)

10

11

12

13

14

100

89

1^d

2

cinchonine

100

50

94

6

54 (5R)

39 (5R)

47 (5R)

33 (5S)

26 (5R)

10 (5R)

8 (5S)

PhCN

2′-butylcinchonine

dihydrocinchonine

(–)-N-methylephedrine

(S)-diphenylprolinol

(DHQD)₂Pyr

DMF

3

MeNO₂

EtNO₂

4

97

^aGeneral: 4b (0.1 mmol), solvent (1 mL). Conversion and ee deter-

5

12

9

mined by ³¹P NMR with quinine or chiral HPLC.

6^e

7

strychnine

10

78

A screening of solvents was next performed with cincho-

nine as the most efficient catalyst among the cinchona al-

kaloids (Table 4). The asymmetric catalysis proceeded

readily in polar, nonprotic (entries 1–4), or aromatic sol-

vents (entries 6 and 7), but less successfully in protic sol-

vents (entry 5). However, best results were clearly

achieved in dipolar aprotic media including nitriles (en-

tries 8–11), DMF (entry 12), or nitroalkanes (entries 13

and 14). Acetonitrile was chosen as suitable reaction me-

dium that can be recycled by simple vacuum evaporation.

Dipolar solvents are suitable media for microwave accel-

erated reactions,²⁸and those were also tested (Table 5). At

50 °C, the thermal reaction requires 16 hours for complete

conversion (entry 1), whereas the microwave reaction was

stopped after 4 hours, having reached about half conver-

sion (entry 2). The enantioselectivity with either heating

modes was almost the same. Reactions at 70 °C were

completed in a similar time range (entries 3 and 4). In

spite of identical nominal temperature readings, the mi-

crowave reaction proceeded faster and with lower

enantioselectivity. Presumably, the same nominal temper-

ature readings correspond to a higher internal temperature

in case of the microwave heating.²⁹Since microwave

heating offered no particular advantage, catalyses were

henceforth performed under thermal (oil bath) conditions.

8^f

L-proline

15 (5S)

Me

N

OH

Ph

9^g

85

49 (5R)

NMe

^aAll reactions performed at 70 °C with 4b in CDCl₃(10 L/mol) and

10 mol% of catalyst for 16 h.

^bDetermined by ¹H NMR.

^cDetermined by ³¹P NMR with quinine.

^d20 mol%.

^e5 mol%.

^f30 mol%.

^g50 mol%.

Bifunctional catalysts with a basic site and a hydrogen

bond donor group are most successful, e.g. β-amino alco-

hols (Table 3, entries 1–5) from the cinchona (entries 1–

3) or ephedra (entry 4) alkaloid families. Ethers of cincho-

na alkaloids (entry 6) or other alkaloids lacking a hydro-

gen bond donor (entry 7) were neither very active nor

selective catalysts. Some success was also achieved with

an amino acid (entry 8) or a guanidinyl alcohol (entry 9)

as bifunctional catalysts, though these leads were not in-

vestigated further.

Synthesis 2013, 45, 308–325

FEATURE ARTICLE

Organocatalytic Asymmetric Isomerization

313

Table 5 Comparison of Microwave and Thermal Heating^a

ral organocatalyst families, including Takemoto’s thio-

urea-amines (Table 7).³¹Indeed, the standard Takemoto

catalyst (R,R)-14a induced an isomerization at compara-

ble rates to the cinchona alkaloids, but with a higher enan-

tioselectivity (88%) in favor of (5S)-5b (entry 1). The

sense of induction in this case is inverted relative to cin-

chonine. An advantage of the thiourea-amines is their

amenability to structural variation. In a first example, the

use of piperidino derivative 14b (entry 2) gave 5b with an

even higher 91% ee. Mixed results were obtained when

applying catalysts 14 to the more difficult substrate 11f

(entries 3 and 4). These first results with structurally vari-

able bifunctional catalysts are promising and imply that

additional optimization to find suitable catalyst/substrate

matches will be possible.

cinchonine (10 mol%)

Ph

NEt₂

P

MeCN

O

NEt₂

4b

5b

Entry Heating

Temp (°C) Time (h) Conv.^b(%) ee^b(%)

1

2

3

4

oil bath

MW

50

16

4

100

59

78.5 (5R)

78.1 (5R)

76.7 (5R)

71.2 (5R)

50^c

70

oil bath

MW

5

100

70^c

3

^aConditions: MeCN (10 L/mol), cinchonine (10 mol%). Microwave

experiments: p ≤ 5 bar, P = 300 W.

^bConversion and ee determined by HPLC.

^cIR temperature measurement.

Mechanism of the Asymmetric Isomerization

The catalyst screening had revealed the importance of a

bifunctional activation mode. A strong hydrogen-bonding

interaction between the catalyst OH and the sub-

strate/product P=O group can be assumed [Figure 3 (a)].

It is responsible for a very simple determination of the en-

antiomeric excess of the rearranged 1-oxo-4,5-dihydro-

1H-phospholes 5, 12, and 13, which consists of recording

³¹P NMR spectra of the latter in the presence of 1–5 equiv-

alents of quinine in CDCl₃solution [Figure 3 (b)], though

for more accurate determinations, chiral HPLC in the nor-

mal phase mode was also suitable.³²

Asymmetric Isomerization of New meso-2,5-Diaryl-

2,5-dihydro-1H-phospholes

The new meso-dihydro-1H-phospholes 4 and 11 were cat-

alytically isomerized with cinchonine (Table 6). Com-

pared to 4b (entries 1 and 2) sterically more demanding

NR₂groups decreased enantioselectivity in the rearrange-

ment (entries 3 and 4).

The presence of ortho substituents on the aryl group also

lowered enantioselectivity (entries 5–8) and led to slower

rearrangements, which was counteracted by increasing

the reaction temperature. Generally, substrates with steri-

cally demanding aryl groups require higher reaction tem-

peratures and longer reaction times. The para-tert-butyl-

substituted 11b (entry 9) and the meta/para-substituted

11c (entry 10) gave a reasonably selective rearrangement

with similar enantiomeric excesses like the unsubstituted

reference substrate 4b. This indicates that steric bulk in

the remote meta and para positions is better tolerated and

sterically does not affect the transition state. The reduced

reactivity of these substrates then must be due to a nega-

tive electronic effect of electron-donating groups, which

would agree with a carbanionic reaction mechanism (vide

infra). Finally, the 1-oxo-2,5-dihydro-1H-phosphole 7

showed great promise by rearranging neatly to 13 with a

relatively high enantiomeric excess (entry 11), which

shows that the reaction is by no means limited to phos-

phinic acid amides.

Figure 3 (a) Presumable complexation of 1-oxo-4,5-dihydro-1H-

phospholes by quinine as hydrogen bond donor; (b) ee determination

of 1-amino-1-oxo-4,5-dihydro-1H-phospholes by ³¹P NMR spectros-

copy with quinine as chiral shift reagent in CDCl₃[er 69.6:30.4, 39%

ee (5R)].

Similar enantiomeric excess determinations by NMR

have been reported for 1-oxo-dihydro-1H-phosphole with

chiral amides as H-bond donors.³³Based on the impor-

tance of hydrogen bonding and basicity, we devised a

working hypothesis for the mechanism which involved

the precomplexation of substrate and catalyst via a hydro-

gen bond, followed by stereoselective deprotonation of

C–H at C2 or C5. In case of rate-limiting deprotonation, a

primary deuterium isotope effect might be expected.

Thus, placing a deuterium label at C2/5 of 4b was of

greatest interest. Our Wenkert cross-coupling synthesis of

1,4-diarylbuta-1,3-dienes¹⁶provided a practical access to

labeled substrates, as shown in Scheme 5. Thiophene was

Bifunctional Thioureas as Catalysts

The results obtained so far were promising, but they im-

plied that higher enantioselectivity cannot be achieved

with the cinchona alkaloids as catalysts. Being natural

products, they lack the structural variability of modular

synthetic catalysts. However, the bifunctionality pattern

of cinchona alkaloids³⁰is also found in fully synthetic chi-

Synthesis 2013, 45, 308–325

314

L. Hintermann et al.

FEATURE ARTICLE

Table 6 Catalytic Isomerization of meso-2,5-Dihydro-1H-phospholes^a

cinchonine (10 mol%)

MeCN

Ar

NR₂

P

O

NR₂

5 or 12

4 or 11

Entry

1

Product^b

Temp (°C)

Time (h)

20

Yield (%)

ee (%)

79

5b

50

98

P

O

NEt₂

2^c

5b

50

70

20

5

97

98

78

77

2a

P

3

5c

70

16

96

66

O

N

4

P

5d

70

16

96

63

60

47

NⁱPr₂

NEt₂

O

5^d

12d

220

6

7

12e

12g

90

180

220

92

68

41

37

P

O

NEt₂

P

O

NEt₂

8

9

12f

90

80

170

100

86

87

30

64

P

O

NEt₂

P

12b

tBu

O

NEt₂

P

10

12c

13

80

50

100

16

94

71

77

MeO

OMe

O

NEt₂

tBu

11^e

P

O

Ph

^aConditions: MeCN (10 L/mol); ee determined by ³¹P NMR with quinine as shift reagent.

^bThe absolute configuration of the excess enantiomer is presumably (5R) in all cases by analogy to that of 5b.

^c0.3-mol scale, catalyst (5 mol%).

^dCH₂Cl₂(0.3 L/mol) was added for solubility reasons.

^eStarting material is 1-oxo-2,5-dihydro-1H-phosphole 7.

Synthesis 2013, 45, 308–325

FEATURE ARTICLE

Organocatalytic Asymmetric Isomerization

315

Table 7 Asymmetric Catalytic Rearrangement of meso-1-Oxo-2,5-dihydro-1H-phospholes with Bifunctional Thiourea Catalysts^a

CF₃

thiourea catalysts:

catalyst 14

(10 mol%)

Ar

S

P

14a R = Me

14b R = {CH₂}₅

P

O

NEt₂

O

MeCN

NEt₂

N

CF₃

5/12

H

4/11

NR₂

Entry

Substrate

Catalyst

Product

Temp (°C)

Time (h)

Conv. (%)

ee (%)

P

1

2

4b

14a

14b

50

20

16

>99

88 (5S)

91 (5S)

Et₂N

O

5b

P

3

11f

14a

50

168

7

8 (5S)

O

Et₂N

12f

4

5

11f

14a

14b

90

144

92

24

20

11 (5S)

66 (5S)

12f

^aConditions: MeCN (10 L/mol), catalyst (10 mol%). Conversion and ee determined by chiral HPLC or ³¹P NMR with quinine.

dilithiated³⁴and quenched with D₂O, to give 2,5-[D₂]- loss of the label is readily explained by exchange with the

thiophene with high deuterium incorporation at C2/5 catalyst hydroxy group and residual water in the reaction

(Scheme 5). The low chemical yield is due to the volatility mixture. The rearrangement is suprafacial, the deuteron

of thiophene. Wenkert arylation¹⁶proceeded as usual to (²H⁺) being removed from C2 and re-attached to C4 from

give 1,4-diphenyl-1,4-dideuteriobuta-1,3-diene ([D₂]-6) the same face of the heterocycle. There is no need to pos-

with specific deuterium incorporation. Eventually, tulate a concerted, thermally forbidden mechanism, be-

McCormack cycloaddition gave the labeled 2,5-dihydro- cause face-selectivity can result from binding of the

1H-phosphole [D₂]-4b (Scheme 5).

catalyst as proton-shuttle to the substrate via the P=O

group (vide infra).

PhMgCl

1) s-BuLi

Ni(acac)₂(3%)

IMes-HCl (7.5%)

85% D

2 x 98% D

TMEDA

D

2% D

cinchonine

H

4

D

S

D

Ph

D

Ph

(10 mol%)

88% D

2) D₂O

PhMe

80 °C, 20 h

D

2

S

1

Ph

P

Ph

5

21% (97% D)

(volatile)

P

O

MeCN

O

Et₂N

50 °C, 20 h

[D₂]-5b (91%)

83% ee (5R)

[D₂]-4b

D

1) Cl₂PNEt₂, AlCl₃

CH₂Cl₂, 0 °C – r.t.

Ph

D

Ph

Et₂N

Ph

P

Ph

Scheme 6 Catalytic isomerization of deuterium-labeled 4b

O

2) NaHCO₃, NTA

H₂O, 0 °C

[D₂]-4b

66% (97% D)

[D₂]-6

38% (≥95% D)

The more surprising result concerns the kinetics of the

catalytic rearrangement, where the absence of any notable

deuterium kinetic isotope effect (KIE) is evident (Figure

4). The kinetic curves of the rearrangement of 2,5-[H₂]-4b

vs. 2,5-[D₂]-4b at 50 °C (Figure 4) or also 70 °C³⁵show a

close match. There is a small but significant isotope effect

on enantioselectivity of the reaction product, both at 50 °C

(81% ee vs. 83% ee with deuterated substrate) and 70 °C

(76% ee vs. 79% with deuterated substrate).

Scheme 5 Synthesis of deuterium-labeled dihydro-1H-phosphole 4b

Substrate [D₂]-4b was submitted to the catalysis condi-

tions at 50 °C and the rearrangement product isolated

(Scheme 6). As expected, the deuterium labels show up at

C4 and C5 of 5b, with a recovery of 85% (H_Si, trans to Ph)

and 2% (H_Re) at C4, and a recovery of 88% at C5. Partial

Synthesis 2013, 45, 308–325

316

L. Hintermann et al.

FEATURE ARTICLE

H

N

O

N

H

Ph

H

Ph

+

O

P

HO

P

H

Ph

4b

Ph

NEt₂

A

N⁺

H

O

N⁺

O

H

Ph

N⁺

Ph

O

H

-

H

O

P

C

O

-

H

-

H

Ph

NEt₂

P

O

D

Ph

P

Figure 4 Kinetic reaction progress curve (determined by HPLC) of

the catalytic isomerization 4b → 5b (10 mol% of cinchonine, 10

L/mol MeCN) recorded with 4b or 2,5-[D₂]-4b at 50 °C. For results

at 70 °C, see the Supporting Information.

NEt₂

B

Ph

NEt₂

N

O

Ph

O

N

H

This clear-cut result was surprising since a primary KIE of

up to 6 might have been expected for a rate-limiting de-

protonation of H–C2/5,³⁶the step which almost certainly

initiates the overall prototropic rearrangement. Potential

explanations for the absence of a primary KIE have been

considered: first, coincidental canceling of two KIEs for

the deprotonation (at C2/5) and reprotonation (at C4) step

might occur. However, both of those KIEs should be pri-

mary, and none is expected to be inverse, excluding this

explanation. Thus, the absence of a primary KIE means

that initial deprotonation is not rate-limiting. A second ex-

planation then assumes reversible initial deprotonation,

followed by a rate-limiting, irreversible conformational

change at the stage of the allylic carbanion, followed by

fast reprotonation. However, it is not evident what kind of

conformational restriction could be responsible for a rela-

tively large energy barrier. The third and in our opinion

most probable explanation (cf. Scheme 7) involves a se-

ries of ion-pairs as intermediates. We assume fast, revers-

ible initial deprotonation within a hydrogen-bonded

adduct A to a tight ammonium–carbanion pair B, which

collapses back to A at a comparatively fast rate. The slow-

er, competing dissociation of A to C (which may be either

a high-lying intermediate or a transition state on the way

from B to D) is the rate-limiting as well as enantio-deter-

mining step, explaining the absence of a primary KIE

(Scheme 7). A similar explanation for the absence of a pri-

mary KIE was offered in case of the base-catalyzed hy-

drogen exchange of tritiated phenylacetylene in water,

where, according to Kresge,³⁷‘proton transfer is fast, and

subsequent separation of the proton transfer products, the

acetylide ion and the conjugate acid of the catalyzing

base, is slow.’³⁷

+

P

O

H

P

NEt₂

(5S)-5b

HO

Ph

NEt₂

E

Scheme 7 Suggested mechanism of prototropic rearrangement via

fast, reversible initial deprotonation; the catalyst is hydrogen-bonded

to the substrate via a hydrogen bond as handle. The pathway to (5S)-

5b is shown.

this step could still proceed enantioselectivity-determin-

ing. Another labeling experiment is consistent with this

interpretation: a solution of 4b in [D₄]-MeOH with a drop

of triethylamine as catalyst was analyzed by H and ³¹P

1

NMR spectroscopy over several days (Figure 5).

Figure 5 H/D exchange in 4b and isomerization to 5b in [D₄]-MeOH

at 22 °C with triethylamine (44 mol%); H/D-exchange is given as per-

centage amount of exchanged hydrogen of H2/5 (two positions) in re-

maining 4b. Isomerization is given as percentage conversion to 5b.

Once ion-pair B dissociates to C, it either returns via B to

A by protonation at C2, or rearranges to the contact ion-

pair D, followed by a (practically) irreversible protonation

at C4 to give adduct E. Since the rate-limiting step is dis-

sociation of B to C, our initial assumption of an enantio-

determining selective deprotonation of 4b at the diastereo-

topic H–C2/5 positions is not necessarily correct, though

Dihydro-1H-phosphole 4b initially exchanged H for D at

C2/5 ca. 7 times faster than it rearranged to 5b, which sup-

ports the assumption of a reversible initial deprotonation

to an ion pair. It also suggests that the solvent-separated

allylic ion-pair C (Scheme 7) will exchange H/D with the

solvent, and this exchange is intrinsically favored at C2 by

a factor of ca. 14 (twice the relative rate of H/D exchange,

Synthesis 2013, 45, 308–325

FEATURE ARTICLE

Organocatalytic Asymmetric Isomerization

317

since there are two exchangeable positions) over proton- notably without alkene isomerization [Scheme 9 (c)]. En-

ation at C4. This is reminiscent of the kinetically favored antioenriched 1-amino-1-oxo-4,5-dihydro-1H-phosphole

protonation of dienolates at C2 over C4,³⁸and the pre- 5b was hydrolyzed to cyclic phosphinic acid 19 overnight

ferred deprotonation of 1-oxophospholanes at the C–H with relative ease, but under harsher conditions than 15

bond cis to the P=O group.³⁹Pakulski et al. have described [Scheme 8 (d)]. The enantiomeric excess of cyclic phos-

a reaction related to ours, in which quinidine (or cincho- phinic acid 19 was retained in the hydrolysis, as shown by

nine) mediated elimination of a meso-dihydro-1H-phos- ³¹P NMR spectroscopy of the product in the presence of

phole epoxide gave

a

4-hydroxy-4,5-dihydro-1H- quinine as chiral shift reagent. As reported in our commu-

phosphole [Scheme 8 (a)] with the same sense of induc- nication,¹⁴phosphinic acid 19 could be fractionally crys-

tion as observed in our catalytic reaction [Scheme 8 (b)].²⁴tallized to ≥95% ee. The enantiopurity of the acid was

Adaptation of their transition-state model to our reaction conveniently analyzed by ³¹P NMR spectroscopy with

is shown in Scheme 8 (c). Even though deprotonation is quinine as the chiral shift reagent.

not rate-limiting in our catalysis, one can assume that the

interactions present in Scheme 8 (c) will also be important

in the structurally similar ion-pair C (Scheme 7), and in

the transition state of Pakulski et al., explaining the same

sense of induction observed in both reactions.

d(³¹P)/ppm

a)

EtOH

3 M HCl

Ph

R₂N

Ph

66.8

P

r.t., 16 h

O

HO

O

83%

75%

15a R = Me

15b R = Et

rac-1

a)

O

OH

cinchonine (1 equiv)

b)

c)

glycol

H_Si

H_Re

6 M HCl

CH₂Cl₂, 20 °C

90 d, 77%

P

68.0

P

Ph

Et₂N

Ph

P

O

Ph

47% ee

P

O

Ph

O

Ph

HO

O

140 °C, 24 h

80%

17

16

b)

glycol

cinchonine (0.1 equiv)

Ph

H_Re

Ph

H_Si

6 M HCl

Ph

Et₂N

Ph

Et₂N

Ph

P

MeCN, r.t.

16 h, 99%

P

69.8

60.1

O

120 °C, 24 h

95%

Et₂N

O

4b

18

4b

(5R)-5b

81% ee

d)

MeOH

c)

6 M HCl

Ph

Et₂N

(5R)-5b

Ph

P

N

O

60 °C, 16 h

91%

H

(5R)-19

O

N

O

P

H

Ph

H

Ph

Scheme 9 Conditions for the hydrolysis of 1-amino-1-oxophos-

NEt₂

pholes

Scheme 8 Sense of induction in desymmetrization reactions of 1-

oxo-2,5-dihydro-1H-phospholes: (a) reaction in ref. 24; (b) asymmet-

ric catalytic dihydro-1H-phosphole isomerization (this work); (c)

adaption of the steric model of Pakulski et al. to the reaction of the

current work.²⁴

The Asymmetric Catalytic Synthesis of Fiaud’s Acid

Diastereoselective Reduction of 2,5-Diphenyl-4,5-di-

hydro-1H-phosphole Derivatives

2,5-Diphenylphosphole-Derived Acids

Having developed an asymmetric catalytic access to 5b

and cyclic phosphinic acid 19 from thiophene, we wished

to extend this synthetic sequence to an asymmetric total

synthesis of Fiaud’s acid 1, which is a useful building

block in chiral phosphane ligand synthesis.⁴⁰Diastereose-

lective hydrogenation of 4,5-dihydro-1H-phosphole 5b,

followed by hydrolysis should lead to the desired target

(Scheme 10).

Having found a route to enantioenriched 1-amino-1-oxo-

phospholes, we investigated their hydrolysis to cyclic

phosphinic acids in aqueous mineral acid.^19dThe influ-

ence of stereochemistry and degree of saturation was ana-

lyzed in the 2,5-diphenylphosphole series (Scheme 9).

Fiaud had already reported the hydrolysis of racemic

cis,trans-1-amino-1-oxo-2,5-diphenylphospholane 15 to

1 under relatively mild conditions [Scheme 9 (a)].¹²Hy-

drolysis of the diastereomer trans,trans-16, obtained by

hydrogenation of 4b, to meso phosphinic acid 17 required

more forcing conditions, presumably because of steric

hindrance towards the attack of water [Scheme 9 (b)]. The

meso-2,5-dihydro-1H-phosphole 4b hydrolyzed to meso

phosphinic acid 18 under equally forcing conditions, but

H⁺

H₂

Ph

Et₂N

Ph

Et₂N

Ph

P

O

H₂O

O

OH

(R,R)-1

5b

15b

Scheme 10 Projected transformation of 5b to phosphinic acid 1

Synthesis 2013, 45, 308–325

318

L. Hintermann et al.

FEATURE ARTICLE

Table 8 Diastereoselective Hydrogenation of 5b^a

However, the conformation of 1-amino-1-oxo-4,5-di-

hydro-1H-phosphole 5b is not favorable for attack of any

hydrogenating reagent from the top-face (Re-attack at C2;

Figure 6). Instead, favored hydrogenation from the lower

face opposing the phenyl and NEt₂groups should lead to

achiral meso-1-amino-phospholane 16.

+

Ph

NEt₂

Ph

NEt₂

Ph

NEt₂

P

O

5b

15b

16

Entry Reagent Solvent Temp (°C) Time (h) Ratio 15/16

1

2

3

4

5

6

Pd/C, H₂EtOH

NaBH₄EtOH

NaBH₄/I₂THF

r.t.

0

4

1:75

3

1:2^b

0

6

1:1.2^c

1.21:1^d

1.3:1

1.87:1

Mg

MeOH r.t.

MeOH

MeOH –20

4

0

2

15

^aReactions were run with enriched (5R)-5b (81% ee) to ≥99% con-

version; 15b was 81% ee unless otherwise indicated.

^b24% conversion.

Figure 6 Diastereoselectivity of hydrogenation of 5b. The modeled

(Chem3D, MM2) conformation of (5S)-5b has the NEt₂group in an

axial position. The (desired) attack from the top face is blocked, in-

stead, attack from the lower face leads to meso-16.

^c43% conversion.

^dProduct was 67% ee.

ly formed carbanionic intermediate was quenched by ad-

dition of ammonium chloride at low temperature to

prevent base-catalyzed isomerization on warming. The

desired 2,5-trans-1 was obtained in excess, and with re-

tention of enantiomeric excess (Table 9). Sodium (entry

1) gave superior results to other metals (entries 2–4). Vari-

ation of the acid quencher influenced the diastereoselec-

tivity of the reaction; use of alcohols or water for

quenching generated side-products of the Birch type (en-

tries 5–7). Several ammonium salts were good quenchers;

interestingly, there was a marked counterion effect of the

ammonium salt on the diastereoselectivity of the reduc-

tion (entries 1, 8–12).

In agreement with the steric arguments, hydrogenation of

5b over a palladium catalyst produced almost exclusively

meso-16 (Table 8, entry 1). Reagents which hydrogenate

via hydride transfer gave mixtures of diastereomers in

comparable amounts (entries 2 and 3). Reduction with

magnesium in methanol, a reaction proceeding by step-

wise transfer of e^–and H⁺, inverted the diastereoselectivi-

ty (entries 4–6). At ambient temperature the reduction was

accompanied by loss of enantiomeric excess, probably

due to base-mediated isomerization of side-product meso-

16 to give rac-15b. This side-reaction is suppressed at

lower temperatures (entries 5 and 6), where increased dia-

stereoselectivities were also observed.

Finally, the diastereoselective reduction of acid 19 to

Fiaud’s acid 1 was best performed with sodium in liquid

ammonia at –78 °C, followed by quenching of the reaction

mixture with ammonium sulfate. The results of Table 9

Hydrogenation of cyclic phosphinic acid 19 was also test-

ed and found to show the same global preference for the

undesired meso-product 17, but an intrinsically higher se-

lectivity for the desired product (Scheme 11).

+

may be interpreted such that reactions with free NH₄ions

give the more stable product 1, whereas weaker and hin-

dered acids, including strongly ion-paired ammonium

salts, produce higher amounts of meso-diastereomer. The

preferential generation of the thermodynamic stereoiso-

mer in dissolving metal reductions is established⁴¹and has

been attributed to pyramidalization of the carbanion to-

wards the incipient C–H bond of the more stable product

diastereomer.⁴¹In analogy, pyramidalization of the inter-

mediary sp²-hybridized⁴²carbanion A (Scheme 12) in di-

rection of the more stable 2,5-trans arrangement of phenyl

groups could account for the observed results. The ion

pairing of an ammonium cation with oxido groups at

phosphorus (Scheme 12, B/C) allows for an assisted pro-

ton transfer. Reaction from ion pair B to D will be faster

than from C to E, because it profits from carbanion py-

ramidalization, which may conceptually be the same as

energetic product development control in the reaction to

Pd/C, H₂

Ph

+

P

EtOH

3 h, r.t.

O

OH

HO

O

HO

O

1

17

(5R)-19

13:87

>99%

(1:6.7)

Scheme 11 Hydrogenation of cyclic phosphinic acid 19

Concluding from these experiments, 19 should be the pre-

ferred substrate, and a dissolving metal reduction at low

temperature should provide the best approach to further

direct the diastereoselectivity of the reduction towards the

chiro product. Sodium in liquid ammonia recommended

itself for improved reduction experiments. Phosphinic

acid 19 was cleanly and quickly reduced by dissolving

metals in liquid ammonia at –78 °C (Table 9). The initial-

Synthesis 2013, 45, 308–325

FEATURE ARTICLE

Organocatalytic Asymmetric Isomerization

319

Table 9 Dissolving Metal Reduction of Dihydro-1H-phospholes^a

Ph

P

1. M, NH₃(l)

2 Na⁺

+

Ph

O

OH

Ph

P

2. HX

2 Na

Ph

O

OH

O

OH

HO

19

P

O

A

NH₃(l)

19

81% ee

1

17

Entry

Metal

Time (h)

Quencher

Ratio 1/17

NH₄X

– NaX

1

Na

Ca

K

0.1

1.5

0.2

0.1

0.2

0.1

NH₄Cl

H₂O

5.6:1

2:1^b

4:1

H

O

P

H

N

2

Ph

H

O

P

Ph

3^c

4

Ph

C

H

O

B

H

Ph

Li

2.5:1

2.3:1^b

6.7:1^b

3.2:1^b

2.2:1

5:1

H

N

H

O

H

5

Na

6

MeOH

t-BuOH

NH₄OAc

NH₄Br

NH₄NO₃

NH₄PF₆

H⁺

E

D

O

+

1

17

7

H

Ph

P

Ph

8

Ph

H

O

H

O

H

9

Scheme 12 Dissolving metal reduction of phosphinic acid 19

10

11

12

6:1

4:1

(NH₄)₂SO₄

7:1

^aReactions performed in liquid NH₃at –78 °C with excess Na (5–20

equiv). Reactions went to >99% conversion and close to quantitative

yield of crude material. Product ee was determined by ³¹P NMR (vide

infra); it remained 81 ± 3% in all examples.

^bBirch reduction side-products were formed.

^c90% conversion.

the more stable product. Proton transfer from the lower

face will become more favorable for sterically large

quencher acids, which prefer attack opposite to the Ph–C5

group. Strongly ion-paired ammonium salts behave like

sterically large acids in that respect.

Association of Cyclic Phosphinic Acids Observed by

³¹P NMR Spectrometry

The ³¹P NMR spectrum of crude samples of enriched

Figure 7 ³¹P NMR (162 MHz, CDCl₃) analysis of the dr (1/17) and

ee of samples of 1 in the absence of chiral additives. Enriched samples

of acid 1, as obtained from dissolving metal reduction of enriched 19

phosphinic acid 1 in CDCl₃solution displayed three peaks

instead of the expected two for 1 and 17 (Figure 7). After (81% ee), were dissolved in CDCl₃.

addition of quinine as chiral shift reagent, the ³¹P NMR

displayed a new set of three peaks in an unchanged ratio.

Asymmetric Catalytic Total Synthesis of 1-Hydroxy-

1-oxo-trans-2,5-diphenylphospholane

We conclude that either enantiomer of 1 and meso-acid 17

display a single ³¹P NMR peak in the absence of a chiral

shift reagent;⁴³in fact, it is the enriched acid 1 which

serves as internal shift reagent. The success of this exper-

iment relies on the use of enriched samples and on the

strong hydrogen-bonding interactions of phosphonic

acids,⁴⁴which is at the same time sufficiently dynamic to

produce time-averaged peaks.

The diastereoselective dissolving metal reduction of 19

provides the missing link to a total synthesis of Fiaud’s

cyclic phosphinic acid 1 (Scheme 13). The sequence starts

from thiophene that is converted into (E,E)-1,4-diphenyl-

buta-1,3-diene (6) by a modified Wenkert arylation.¹⁶The

McCormack cycloaddition first used by Fiaud stereose-

lectively gives meso-1-amino-1-oxo-2,5-dihydro-1H-

phosphole 4b, which is isomerized in an asymmetric or-

ganocatalytic process to enantioenriched 1-amino-1-oxo-

4,5-dihydro-1H-phosphole 5b, currently in up to 91% ee.

Synthesis 2013, 45, 308–325

320

L. Hintermann et al.

FEATURE ARTICLE

cinchonine

or 14b

Et₂NPCl₂

AlCl₃

PhMgCl

1. Na, NH₃(l)

–78 °C

(5 mol%)

[Ni]-cat.

Ph

P

Ph

Et₂N

Ph

P

X

O

S

2. (NH₄)₂SO₄

PhMe

80 °C

CH₂Cl₂

0 °C, 20 h

O

MeCN

20 h, 50 °C

O

OH

(R,R)-1

5b (X = NEt₂)

19 (X = OH)

6 (70%)

4b (83%)

HCl (aq)

MeOH

80 °C

54%, 98% ee

(90%, up to 91% ee)

Scheme 13 Overview of the five-step asymmetric catalytic total synthesis of 1-hydroxy-1-oxo-trans-2,5-diphenylphospholane

After hydrolysis to phosphinic acid 19, dissolving metal solute configuration assignments of all intermediates

reduction with diastereoselective protonation gives (Scheme 14).

Fiaud’s acid 1, which is recrystallized to high diastereo-

meric and enantiopurity. For example, a cinchonine-cata-

lyzed isomerization of 4b gave 5b with 80% ee, which

Conclusions

was hydrolyzed and reduced without further purification.

The crude product 1 (80% ee, dr 6:1) thus obtained was re- The study presented here started from a purely conceptual

crystallized from methanol to give pure 1 (>98% ee) in question: how can we realize an asymmetric organocata-

54% yield.

lytic stereoisomerization? We chose an approach of isom-

erizing cyclic meso substrates to give the chiral isomers

with release of conformational strain, and identified meso-

1-oxo-2,5-dihydro-1H-phosphole described by Fiaud and

Structure Correlation Studies

A series of reactions was performed (Scheme 14) to cor- co-workers¹²as suitable model substrates. The project

relate the absolute configuration (of the excess enantio- took a different turn, when meso-1-amino-1-oxo-2,5-di-

mer) of all our intermediates with Fiaud’s acid 1, whose hydro-1H-phosphole 4b isomerized (quite predictably, in

configuration and optical rotation are known.¹²An en- hindsight) to the conjugated 1-amino-1-oxo-4,5-dihydro-

riched sample of 1-amino-1-oxo-4,5-dihydro-1H-phos- 1H-phosphole 5b. This reaction illustrates the utility of

phole (–)-(5R)-5b as obtained from a cinchonine- asymmetric catalytic isomerizations, and its investigation

catalyzed isomerization of 4b was hydrolyzed to phos- has exposed us to aspects of phospholane stereochemistry,

phinic acid 19. This compound displayed the notable phe- which have led to new insights. A catalytic asymmetric

nomenon of changing the sign of optical rotation in synthesis of Fiaud’s acid 1 by means of an organometallic

solution, depending on the solvent:⁴⁵in methanol, it was cross coupling (Wenkert arylation), an asymmetric or-

laevorotatory, in dichloromethane dexrotatory.⁴⁶In addi- ganocatalysis and a diastereoselective contrasteric hydro-

tion, the magnitude of [α]_Din CH₂Cl₂was also dependent genation as key steps has been realized. Practical large-

on sample concentration, pointing to an association equi- scale applications will profit from further optimization of

librium. A sample of (–)-(5R)-5b was reduced with mag- the catalyst for asymmetric isomerization, and by finding

nesium metal in methanol (cf. Table 8, entry 5) and the even more selective hydrogenation conditions. The route

resulting mixture of (+)-15b and meso-1-amino-1-oxo- 6 → 4b → 5b → 19 was readily scaled up to 100-gram

phospholane 16 separated by chromatography. Either hy- quantities in each step. The dissolving metal reduction

drolysis of (+)-(R,R)-15b or dissolving metal reduction of and in particular the acid quench is more problematic,

(5R)-19 produced (+)-(R,R)-1,^12bthereby assuring the ab- since temperature control is important to achieve high se-

Na/NH₃

6 M HCl

MeOH

Ph

NEt₂

Ph

P

(NH₄)₂SO₄

O

OH

O

OH

(5R)-5b (80% ee)

(5R)-19 (81% ee)

(R,R)-1 (81% ee)

[α]_D= –141 (c = 0.6, CH₂Cl₂)

[α]_D= –115 (c = 0.6, CH₂Cl₂)

[α]_D= +126 (c = 0.6, CH₂Cl₂)

[α]_D= + 98 (c = 0.3, CH₂Cl₂)

[α]_D= –22 (c = 0.6, MeOH)

[α]_D= +79 (c = 0.6, CH₂Cl₂)

[α]_D= +58 (c = 0.6, MeOH)

@76% ee

(S,S)-1 (>99.6% ee)

[α]_D= –102.7 (c = 0.6, CH₂Cl₂)

Lit.:

Mg, MeOH

6 M HCl

glycol

98%

Ph

NEt₂

P

O

15b (31%) + 16 (53%)

separate from 16

(2R,5R)-15b (80% ee)

[α]_D= +46 (c = 0.6, CH₂Cl₂)

[α]_D= +43 (c = 0.6, MeOH)

Scheme 14 Structure correlation of intermediates in asymmetric catalytic dihydro-1H-phosphole isomerization chemistry with Fiaud’s acid 1

Synthesis 2013, 45, 308–325

FEATURE ARTICLE

Organocatalytic Asymmetric Isomerization

321

celite, the aqueous phase was decanted off and extracted with

CH₂Cl₂(2 ×). The combined organic phases were washed with sat.

NaHCO₃, 1 M HCl, and aq NaCl. The solution was dried (MgSO₄),

filtered, and evaporated to give the crude product. Excess diene was

occasionally removed by dissolving the crude in MeOH and filter-

ing off diene. The crude product was purified by crystallization

(cold acetone–hexanes, 1:1; by dissolving in acetone, overlayering

with hexanes, and storing the mixture at –20 °C in a fridge).

lectivity. The fundaments for asymmetric syntheses of

new, more bulky 1-hydroxy-1-oxo-2,5-diarylphospholanes

have been laid. Their preparation and use in asymmetric

Brønsted acid catalysis is a promising endeavor. We note

that the initial quest for an enantioselective catalytic meso

to chiro isomerization [cf. Scheme 1 (a)–(c)] remains an

open challenge for new catalytic asymmetric stereo-

isomerization chemistry.

(1s,2R,5S)-1-(Diethylamino)-1-oxo-2,5-diphenyl-2,5-dihydro-

1H-phosphole (4b)

Synthesis according to GP1:¹⁴from Et₂NPCl₂(8.7 g, 0.05 mol, 1.1

equiv), AlCl₃(6.667 g, 0.05 mol, 1.1 equiv) and (E,E)-1,4-diphen-

ylbuta-1,3-diene (6, 9.386 g, 0.046 mol, 1 equiv) in CH₂Cl₂(120

mL). After crystallization (acetone–hexane, 1:1), 4b (12.3 g, 83%)

was obtained as a white powder that slowly turned yellow on stand-

ing in air. The material was stored in a fridge under argon.

Reactions with air- and water-sensitive substances were performed

in Schlenk glassware under argon. Reaction progress was followed

by TLC. Standard workup consisted in quenching with H₂O or

aqueous salt solutions, extraction with a suitable solvent, followed

by drying of the organic phase (MgSO₄). Filtration and evaporation

of the filtrate gave crude product that was purified by flash column

chromatography (silica gel, 0.040–0.063 mm) with a positive pres-

sure of 0.2 bar.

Large-scale synthesis: Et₂NPCl₂(90 g, 0.517 mol) was slowly add-

ed to AlCl₃(68.94 g, 0.517 mol) in CH₂Cl₂(200 mL) at 0 °C with

stirring. The mixture was stirred for 45 min at r.t. until the solids had

dissolved and then cooled to 0 °C. A chilled solution of (E,E)-1,4-

diphenylbuta-1,3-diene (6, 96.96 g, 0.47 mol) in CH₂Cl₂(700 mL)

was slowly added and the mixture stirred for 19 h at 0 °C. The reac-

tion was quenched by slowly transferring into a stirred, ice-cooled

mixture of NTA (98.82 g, 0.60 mol) in sat. aq NaHCO₃(1500 mL)

and further stirred for 4 h at 0 °C. The organic phase was collected,

and the aqueous phase extracted with CH₂Cl₂(2 ×) The combined

organic phases were washed with sat. NaHCO₃, aq 2.4 M HCl, and

aq NaCl. The solution was dried (MgSO₄), filtered, and evaporated

to give a crude yellowish product (126.37 g, 83%); ratio of diaste-

reomers 16:1 (³¹P NMR). Crystallization (acetone–hexane) gave a

first crop (86.17 g), and another crystallization of the mother liquors

gave a second crop (22.84 g) of pure product (total: 109.01 g, 72%);

dr ≥ 50:1.

MeCN: HPLC-grade, used without purification. CH₂Cl₂: distilled

over CaH₂. DMF: dry grade, used without purification. Et₂O:

predried over KOH, distilled over Na/benzophenone. EtOH: HPLC

quality (abs), used without purification. MeOH: HPLC grade, used

without purification. THF: predried over KOH, distilled over

Na/benzophenone. Toluene: distilled over Na/benzophenone. Sol-

vents for chromatography (EtOAc, CH₂Cl₂, t-BuOMe, hexanes)

were technical grade and distilled before use. Chemicals were ob-

tained from commercial suppliers and used without further purifica-

tion. Grignard reagents were prepared according to general

procedures and titrated using salicylaldehyde phenylhydrazone in-

dicator.⁴⁷The following compounds were synthesized according to

literature procedures: 1,3-dimesitylimidazolium chloride and 1,3-

bis(2,6-diisopropylphenyl)imidazolium chloride.⁴⁸

Microwave reactions were performed in a microwave reactor with

continuous irradiation and adjustable power (0–300 W) to keep the

temperature of the reaction vessel at a constant value.

Mp 160–165 °C; R_f= 0.23 (EtOAc–hexanes 2:1).

IR (KBr): 2978 (s), 1596 (m), 1495 (s), 1380 (s), 1218 (vs), 1027

(vs), 952 cm^–1(s).

Analytical techniques: Analytical HPLC was performed on chiral

columns (250 × 4.6 mm) of the type: Chiralpak OD (10 μm), Chira-

cel OJ (10 μm). Melting points were measured on a metal heating

block with a digital thermometer (thus: corrected values). NMR

spectroscopy: Measured at r.t. ¹H NMR shifts relative to TMS (δ =

0) or alternatively solvent signals (CDCl₃, δ = 7.26; DMSO, δ =

2.50; CD₃OD, δ = 3.34) as internal standards. ¹³C NMR spectra

were referenced to solvent signals (CDCl₃, δ = 77.0; DMSO, δ =

39.7, CD₃OD, δ = 49). ³¹P NMR measured with ¹H decoupling. IR

spectra: were either measured in KBr pellets, CHCl₃solution, or as

a film. Intensities of bands: vs = very strong (0–20%), s = strong

(21–40%), m = medium (41–60%), w = weak (61–80%), vw = very

weak (81–90%).

¹H NMR (400 MHz, CDCl₃): δ = 0.13 (t, J = 7.1 Hz, 6 H), 2.42 (ψ-

dq, J = 9.1, 7.1 Hz, 4 H), 4.33 (dd, J = 18.3, 0.7 Hz, 2 H), 6.51 (dd,

J = 28.8, 0.8 Hz, 2 H), 7.20–7.35 (m, 10 H, aryl).

¹³C NMR (100 MHz, CDCl₃): δ = 13.1 (d, J_PC= 2.5 Hz, CH₃), 38.0

(d, J_PC= 2.3 Hz, CH₂), 50.1 (d, J_PC= 70.5 Hz, CH), 126.6 (d, J_PC

=

3.8 Hz, CH), 127.3 (d, J_PC= 5.7 Hz, CH), 128.5 (d, J_PC= 2.5 Hz,

CH), 130.8 (d, J_PC= 16.6 Hz, CH), 135.7 (d, J_PC= 8.3 Hz, C).

³¹P NMR (162 MHz, CDCl₃): δ = 70.8.

MS (ESI): m/z (%) = 325 (2) M⁺, 206 (100), 91 (14).

Anal. Calcd for C₂₀H₂₄NOP: C, 73.82; H, 7.43; N, 4.30. Found: C,

73.37; H, 7.36; N, 4.49.

Syntheses of meso-2,5-Diaryl-2,5-dihydro-1H-phospholes

The (dialkylamino)dichlorophosphanes used are known com-

pounds and were prepared from PCl₃and amine (2 equiv) according

to literature procedures.^20a,b,491,4-Diarylbuta-1,3-dienes for this

study were prepared as described.¹⁶1,4-Diphenylbuta-1,3-diene

was also obtained commercially.

(1r,2R,5S)-1-(Diethylamino)-1-oxo-2,5-diphenyl-2,5-dihydro-

1H-phosphole

Minor diastereomer.

¹H NMR (400 MHz, CDCl₃): δ = 1.08 (t, J = 7.1 Hz, 6 H), 3.11 (ψ-

dq, J = 10.4, 7.0 Hz, 4 H), 3.90 (dd, J = 5.5, 1.1 Hz, 2 H), 6.13 (dd,

J = 25.6, 1.1 Hz, 2 H), 7.2–7.4 (m, 10 H).

(1s,2R,5S)-2,5-Diaryl-1-(dialkylamino)-1-oxo-2,5-dihydro-1H-

phospholes; General Procedure (GP1)

³¹P NMR (162 MHz, CDCl₃): δ = 60.6 (s).

To a stirred suspension of AlCl₃(1.1 equiv) in anhyd CH₂Cl₂under

argon, a solution of (dialkylamino)dichlorophosphane (1.1 equiv)

in CH₂Cl₂was added dropwise with stirring. The mixture was

stirred for 45 min at r.t. to give a greenish solution. After cooling to

0 °C, a precooled solution of 1,4-diarylbuta-1,3-diene (1 equiv) in

CH₂Cl₂was added and the mixture stirred overnight at 0 °C. The

red-brown mixture was poured carefully into precooled (0 °C) mix-

ture of nitrilotriacetic acid (NTA) and sat. aq NaHCO₃. The two-

phase mixture was stirred for 4 h at 0 °C. After filtration through

Asymmetric Catalytic Isomerization of meso-Dihydro-1H-

phospholes; General Procedure (GP2)

A solution of the meso-2,5-dihydro-1H-phosphole 4, 7, or 11 (0.10

mmol) and catalyst (5–10 mol%) in MeCN (1–2 mL) was stirred for

16–48 h at 50–70 °C. After completion of the reaction (TLC control,

typically EtOAc–hexanes, 2:1) the mixture was cooled and MeCN

was removed in vacuo. The residue was dissolved in EtOAc and the

solution washed with aq 2 M HCl (2 ×). The organic phase was

dried (MgSO₄). Filtration and evaporation of the filtrate gave a res-

Synthesis 2013, 45, 308–325

322

L. Hintermann et al.

FEATURE ARTICLE

idue that was either used as received or further purified by flash col-

umn chromatography.

Synthesis of 2,5-Diphenyl-1H-phosphole-Derived Phosphinic

Acids

Analysis: The enantiomeric excess of the reaction product was de-

termined either by chiral HPLC or by ³¹P NMR in CDCl₃with ad-

dition of quinine (1–5 molar equiv).

meso-1-Hydroxy-1-oxo-2,5-diphenylphospholane (17)

A mixture of (1s,2S,5R)-1-(diethylamino)-1-oxo-2,5-diphenylphos-

pholane (16)⁵⁰(4.11 g, 12.55 mmol), EtOH (8 mL), H₂O (5 mL) and

concd HCl (10 mL, 120 mmol) in a round-bottom flask with a

mounted condenser (air cooling) was heated at 120 °C (oil bath) for

20 h; most of the EtOH had evaporated and hydrolysis was incom-

plete (TLC, EtOAc). Ethylene glycol (6 mL) and concd HCl (2 mL)

were added and the mixture refluxed at 140 °C (oil bath) for a fur-

ther 20 h. The mixture was cooled, diluted with H₂O and extracted

with CH₂Cl₂. The aqueous phase was extracted with CH₂Cl₂(2 ×).

The combined organic phases were washed with H₂O (2 ×) and

evaporated to dryness. The crystalline residue was redissolved in

CH₂Cl₂, diluted with cyclohexane, and slowly evaporated in a rota-

tory evaporator to give a suspension of crystals. Filtration and wash-

ing of the solid with cyclohexane gave a first crop of colorless

crystals that was dried in an oven (2.413 g, 71%). The mother liquor

and washing phases were combined and crystallized once more

(CH₂Cl₂–cyclohexane) to give a second crop (328 mg, 9.5%); total

yield: 2.741 g (80%). Note: Non-optimized conditions. The hydro-

lysis required an elevated temperature (130–140 °C); use of an eth-

ylene glycol–H₂O–concd HCl mixture from the start is

recommended.

Recycling: The evaporated solvent (MeCN) can be directly reused.

The catalyst can be precipitated from the aqueous acidic washing

phase by addition of excess aq NH₃; it is isolated by filtration, or by

extraction with CH₂Cl₂and evaporation.

Example for a small-scale screening reaction: The catalyst (cincho-

nine; 2.9 mg, 0.01 mmol) and 4b (32.5 mg, 0.10 mmol) were stirred

in MeCN (2 mL) at 50 °C for 16 h. The solvent was removed in vac-

uo and the residue taken up in EtOAc (5 mL). After washing with

aq 2 M HCl (2 × 1 mL) and H₂O (1 ×), the organic phase was dried

(MgSO₄), filtered, and evaporated to give 5b (31.8 mg, 98%) as a

slightly yellow solid.

(1R,5R)-1-(Diethylamino)-1-oxo-2,5-diphenyl-4,5-dihydro-1H-

phosphole (5b)

Synthesis according to GP2:¹⁴from 4b (0.500 g, 1.54 mmol) and

cinchonine (0.045g, 0.154 mmol, 10 mol%) in MeCN (30 mL) at 50

°C for 40 h. Evaporation of the solvent gave 5b (0.49 g, 98%) as a

slightly yellow solid; 80% ee (HPLC).

Large-scale synthesis of 5b: To a solution of 4b (97.4 g, 0.30 mol)

in MeCN (2.5 L), cinchonine (4.416 g, 0.015 mol, 5 mol%) was

added and the mixture stirred for 20 h at 50 °C. The solution was

cooled to r.t. and the solvent recycled by evaporation using a ro-

tavapor. The solid residue was dissolved in EtOAc (ca. 2 L), the so-

lution washed with 1 M HCl (2 × 100 mL) and the aqueous phases

were retained for catalyst recycling. The EtOAc solution was dried

(MgSO₄), filtered, and evaporated in vacuo. The residue consisted

of 5b (94.5 g, 97%) as a slightly yellow solid; 78% ee (HPLC).

HPLC (Chiralcel-OJ, n-heptane–i-PrOH–H₂O,³²90:10 + 0.25%;

flow = 0.7 mL min^–1, λ = 230 nm): t_R= 10.6 [minor, (5S)], 12.1 min

[major, (5R)]; mp 152–155 °C (enriched sample, 80% ee); R_f= 0.3

(EtOAc–hexanes 2:1); [α]_D–141.1 (c 0.595, CH₂Cl₂; 80% ee); [α]_D

–115.1 (c 0.595, MeOH; 80% ee).

Mp 207–208 °C (droplet formation and sublimation from 185 °C;

compare Lit.⁵¹169–171 °C); R_f= <0.1 (EtOAc).

IR (KBr): 3431 (m), 3028 (m), 2946 (m), 2868 (w), 1709 (w), 1597

(m), 1494s, 1449 (m), 1252 (s), 1199 (s), 1075 (m), 1034 (m), 974

(s), 764 (s), 694 (s), 526 (s), 484 cm^–1(s).

¹H NMR (400 MHz, CDCl₃): δ = 2.20–2.36 (m, 4 H), 3.09–3.24 (m,

2 H), 7.07–7.21 (m, 10 H), 10.22 (s, 1 OH).

¹³C NMR (100 MHz, CDCl₃): δ = 28.0 (d, J_PC= 14.0 Hz, CH₂), 43.3

(d, J_PC= 86.1 Hz, CH), 126.5 (d, J_PC= 2.8 Hz, CH), 128.1 (d, J_PC

=

5.3 Hz, CH), 128.3 (d, J_PC= 2.2 Hz, CH), 135.9 (d, J_PC= 6.9 Hz, C).

³¹P NMR (161.9 MHz, CDCl₃): δ = 68.0 (s).

MS (EI): m/z (%) = 272 (100) [M⁺], 168 (25), 104 (90).

IR (KBr): 2973 (s), 2921 (m), 2875 (m), 1597 (s), 1493 (s), 1453 (s),

1376 (s), 1201 (vs), 1023 (vs), 948 (vs), 854 (m), 765 (vs), 697 (vs),

635 (m), 517 (vs), 475 cm^–1(s).

Anal. Calcd for C₁₆H₁₇O₂P (272.28): C, 70.58; H, 6.29. Found: C,

70.84; H, 6.21.

¹H NMR (300 MHz, CDCl₃): δ = 0.54 (t, J = 7.1 Hz, 6 H), 2.54–2.91

(m, 4 H, NCH₂Me), 2.98 (ddd, J = 18.6, 6.8, 2.8 Hz, 1 H, H4), 3.17

(tdd, J = 18.8, 8.5, 3.9 Hz, 1 H, H4), 3.76 (ddd, J = 22.6, 8.4, 7.0 Hz,

1 H, H5), 7.14–7.41 (9 H, 8 H_Arl+ H5), 7.66–7.74 (m, 2 H_Arl).

meso-1-Hydroxy-1-oxo-2,5-diphenyl-2,5-dihydro-1H-phos-

phole (18)

A solution of 4b (0.260 g, 0.80 mmol) in ethylene glycol (10 mL)

and 6 M HCl (10 mL) was heated to 120 °C for 24 h. The cooled

mixture was diluted with H₂O and extracted with CH₂Cl₂(3 ×). The

organic phase was washed with a small volume of H₂O and dried

(MgSO₄). Evaporation gave 18 (0.206 g, 95%) as a white crystalline

powder.

¹³C NMR (75 MHz, CDCl₃): δ = 13.6 (d, J_PC= 1.9 Hz, CH₃), 32.0

(d, J_PC= 14.6 Hz, CH₂), 37.9 (d, J_PC= 4.5 Hz, CH₂), 44.9 (d, J_PC

=

82.1 Hz, CH), 126.6 (CH), 126.6 (d, J_PC= 5.2 Hz, CH), 127.7 (d, J_PC

= 5.0 Hz, CH), 128.2 (CH), 128.5 (d, J_PC= 2.5 Hz, CH), 128.6 (CH),

Mp 221–223 °C; R_f= 0.24 (EtOAc–MeOH, 3:1).

133.7 (d, J_PC= 9.4 Hz, C), 136.4 (d, J_PC= 5.7 Hz, C), 137.9 (d, J_PC

100.2 Hz, C), 142.8 (d, J_PC= 33.1 Hz, CH).

=

IR (KBr): 3442 (w), 3030 (m), 1657 (m), 1493 (m), 1218 (vs), 972

(vs), 744 (vs), 696 (vs), 539 (s), 495 cm^–1(s).

¹H NMR (300 MHz, CDCl₃): δ = 3.87 (d, J = 13.3 Hz, 2 H), 6.13 (d,

J = 29.9 Hz, 2 H), 7.04–7.11 (m, 4 H_Arl), 7.19–7.28 (m, 6 H_Arl), 7.73

(br s, 1 H, OH).

³¹P NMR (120 MHz, CDCl₃): δ = 60.6.

MS (EI, 70 eV): m/z (%) = 325 (61) M⁺, 310 (56), 253 (49), 205

(13), 91 (11), 72 (100).

Anal. Calcd for C₂₀H₂₄NOP + 0.85 H₂O: C, 70.51; H, 7.60; N, 4.11.

Found: C, 70.56; H, 7.17; N, 4.11.

¹³C NMR (75 MHz, CDCl₃): δ = 47.6 (d, J_PC= 84.2 Hz, CH), 126.9

(d, J_PC= 3.0 Hz, CH), 128.3 (d, J_PC= 5.0 Hz, CH), 128.4 (d, J_PC

=

Enantiomeric enrichment by crystallization: A sample of 5b (0.300

g; 80% ee) was dissolved in acetone (20 mL) and the solution over-

layered with hexanes (20 mL). After standing for 16 h at –10 °C, the

solid (racemate) was filtered off and the filtrate evaporated to leave

(1R,5R)-5b (0.213 g, 71%); 95% ee.

2.5 Hz, CH), 131.3 (d, J_PC= 18.2 Hz, CH), 134.6 (d, J_PC= 8.5 Hz,

C).

³¹P NMR (120 MHz, CDCl₃): δ = 69.8.

MS (EI, 70 eV): m/z (%) = 270 (17) M⁺, 206 (100), 191 (14), 104

(21).

Anal. Calcd for C₁₆H₁₅O₂P + 0.2 H₂O: C, 70.17; H, 5.67. Found: C,

70.37; H, 5.52.

Synthesis 2013, 45, 308–325

FEATURE ARTICLE

Organocatalytic Asymmetric Isomerization

323

25

(5R)-1-Hydroxy-1-oxo-2,5-diphenyl-4,5-dihydro-1H-phos-

phole (19)

Mp 128–131 °C (82% ee); R_f= 0.26 (EtOAc–hexanes, 2:1); [α]_D

+46.4 (c 0.6, CH₂Cl₂; 80% ee); [α]_D²⁵+42.5 (c 0.61, MeOH; 80%

ee).

Regular-scale synthesis:¹⁴A sample of 5b (0.26 g, 0.80 mmol),

which had been recrystallized for enantiomeric enrichment, was

dissolved in MeOH (10 mL). After the addition of 6 M HCl (10

mL), the mixture was stirred for 16 h at 60 °C. After cooling and di-

lution with a small volume of H₂O, the mixture was extracted with

CH₂Cl₂(3 ×). The organic phase was washed with a small volume

of H₂O, dried (MgSO₄), and evaporated to leave 19 (0.197 g, 91%)

as a colorless powder; 95% ee (³¹P NMR).

IR (KBr): 3427 (vs), 2937 (vs), 1598 (m), 1496 (s), 1382 (s), 1175

(vs), 1027 (vs), 937 (s), 761 (vs), 696 cm^–1(vs).

¹H NMR (400 MHz, CDCl₃): δ = 0.65 (t, J = 7.1 Hz, 6 H), 2.06–2.33

(m, 2 H), 2.38–2.57 (m, 2 H), 2.60–2.85 (m, 4 H), 3.10 (td, J = 12.9,

6.8 Hz, 1 H), 3.64 (ddd, J = 24.6, 12.7, 7.4 Hz, 1 H), 7.19–7.38 (m,

10 H_Arl).

Large-scale synthesis: To a solution of (5R)-5b (94.5 g, 0.291 mol;

78% ee) in MeOH (1.6 L), aq 6 M HCl (1.6 L) was added and the

mixture heated to 60 °C for 20 h with mechanic stirring. The mix-

ture was cooled and extracted with several portions of CH₂Cl₂. The

organic phase was dried (MgSO₄). After filtration, the solvents were

evaporated in vacuo and the solid residue suspended in a small vol-

ume of chilled MeOH. Filtration and drying under high vacuum

gave 19 (62.1 g, 79%) as a colorless crystalline material; 77% ee

(³¹P NMR). The enantiomeric excess was determined by ³¹P NMR

of a solution of 19 (5.5 mg) and quinine (37 mg, 5 equiv) in CDCl₃

(0.5 mL); δ(³¹P) = 51.14 (major), 50.79 (minor).

¹³C NMR (100 MHz, CDCl₃): δ = 14.0 (d, J_PC= 2.3 Hz, CH₃), 27.9

(d, J_PC= 9.0 Hz, CH₂), 30.2 (d, J_PC= 11.4 Hz, CH₂), 37.9 (d, J_PC

=

3.1 Hz, CH₂), 44.3 (d, J_PC= 74.9 Hz, CH), 47.7 (d, J_PC= 76.5 Hz,

CH), 126.3 (d, J_PC= 2.6 Hz, CH), 126.6 (d, J_PC= 2.3 Hz, CH), 127.4

(d, J_PC= 4.9 Hz, CH), 128.3 (d, J_PC= 2.2 Hz, CH), 128.4 (d, J_PC

=

1.8 Hz, CH), 128.9 (d, J_PC= 5.0 Hz, CH), 136.5 (d, J_PC= 5.1 Hz, C),

137.0 (d, J_PC= 4.7 Hz, C).

³¹P NMR (161 MHz, CDCl₃): δ = 59.3.

MS (EI, 70 eV): m/z (%) = 327 (42) M⁺, 312 (30), 206 (15), 120

(14), 104 (100), 91 (15), 72 (88).

Anal. Calcd for C₂₀H₂₆NOP: C, 73.37; H, 8.00; N, 4.28. Found: C,

73.16; H, 8.09; N, 4.21.

Mp 237–239 °C (77% ee); R_f= 0.26 (EtOAc–MeOH, 3:1); [α]_D

+126.4 (c 0.595, CH₂Cl₂; 76% ee); [α]_D+98.4 (c 0.28, CH₂Cl₂; 76%

ee); [α]_D–21.9 (c 0.59, MeOH; 76% ee).

(2R,5R)-1-Hydroxy-1-oxo-trans-2,5-diphenylphospholane (Fi-

aud’s Acid, 1)

IR (KBr): 3439 (w), 3050 (m), 1601 (s), 1494 (s), 1449 (m), 1265

(m), 977 (vs), 755 (vs), 694 (vs), 533 (s), 476 cm^–1(m).

¹H NMR (250 MHz, CDCl₃): δ = 2.84 (dddd, J = 18.6, 6.5, 2.6, 1.1

Hz, 1 H), 3.13 (dddd, J = 22.0, 18.5, 8.6, 3.8 Hz, 1 H), 3.46 (ddd,

J = 18.6, 8.6, 6.6 Hz, 1 H), 7.11 (ddd, J = 45.5, 3.7, 2.6 Hz, 1 H),

7.21–7.37 (m, 8 H_Ph), 7.51–7.58 (m, 2 H_Ph), 10.49 (br s, 1 H).

¹H NMR (300 MHz, CD₃OD): δ = 2.89 (ddd, J = 18.7, 6.8, 2.7 Hz,

1 H), 3.20 (dddd, J = 21.8, 18.7, 8.7, 3.8 Hz, 1 H), 3.54 (ddd, J =

18.4, 8.7, 6.7 Hz, 1 H), 7.19–7.43 (m, 9 H, 8 H_Arl+ H3), 7.65–7.73

(m, 2 H_Arl).

By Na/NH₃reduction of 19: Na (0.17 g, 0.370 mmol, 20 equiv) was

dissolved in liquid NH₃(ca. 50 mL) at –78 °C. A solution of

(1R,5R)-19 (0.10 g, 0.370 mmol; 81.5% ee) in THF (2 mL) was add-

ed and the mixture was stirred for 5 min at –78 °C. The reaction was

quenched by addition of solid (NH₄)₂SO₄(1.00 g, 7.57 mmol, 20.5

equiv), causing the blue solution of the mixture to disappear. After

warming to r.t. with evaporation of NH₃into the hood, the residue

was suspended in a small volume of 2 M HCl and extracted with

CH₂Cl₂. After evaporation of the solvent, a white solid remained

(0.0978 g, 98%; ratio 1/17 7.7:1; 82% ee). After crystallization (hot

MeOH, ca. 3 mL/mmol), pure product (0.0539 g, 54%) was ob-

tained as colorless solid; ≥ 98% ee.

¹³C NMR (75 MHz, CD₃OD): δ = 35.1 (d, J_PC= 17.3 Hz, CH₂), 44.4

(d, J_PC= 93.7 Hz, CH), 127.5 (d, J_PC= 5.7 Hz, CH), 128.0 (d, J_PC

=

2.7 Hz, CH), 129.4 (CH), 129.6 (CH), 129.7 (d, J_PC= 3.6 Hz, CH),

129.8 (CH), 134.4 (d, J_PC= 9.7 Hz, C), 137.4 (d, J_PC= 118 Hz, C),

137.6 (d, J_PC= 6.1 Hz, C), 143.9 (d, J_PC= 35.6 Hz, CH).

³¹P NMR (101 MHz, CDCl₃): δ = 66.39 (s, minor, 12%), 66.28 (s,

major, 88%; 76% ee).

³¹P NMR (120 MHz, CD₃OD): δ = 60.1 (s).

By hydrolysis of 15b: To a solution of (2R,5R)-1-(diethylamino)-1-

oxo-2,5-diphenylphospholane (15b, 0.30 g, 0.916 mmol; 81% ee) in

MeOH (10 mL), 6 M HCl (10 mL) was added and the mixture

stirred for 16 h at 60 °C. The aqueous layer was extracted with

CH₂Cl₂and the organic layer washed with a small volume of H₂O,

dried (MgSO₄), filtered, and evaporated. The residue (0.225 g,

75%) consisted of a white solid; 81% ee (³¹P NMR).

MS (EI, 70 eV): m/z (%) = 270 (100) M⁺, 206 (24), 192 (10), 103

(10), 91 (12).

Hydrolysis in ethylene glycol: To a sample of (1R,5R)-15b (116 mg,

0.35 mmol) in ethylene glycol (5 mL), aq 6 M HCl (5 mL) was add-

ed and the mixture stirred at 90 °C for 12 h. The aqueous layer was

extracted with CH₂Cl₂(3 × 20 mL). The organic phase washed with

H₂O (2 × 10 mL), dried (MgSO₄), filtered, and evaporated to leave

the product (94.2 mg, 98%) as a white solid.

Anal. Calcd for C₁₆H₁₅O₂P + 0.1 H₂O: C, 70.64; H, 5.63. Found: C,

70.64; H, 5.66.

Diastereoselective Reduction of 4,5-Dihydro-1H-phospholes

24

[α]_D+79.2 (c 0.615, CH₂Cl₂; 76% ee); [α]_D+57.9 (c 0.615,

(2R,5R)-1-(Diethylamino)-1-oxo-2,5-diphenylphospholane

(15b)

MeOH; 76% ee).

Prepared by reduction of (1R,5R)-5b (1.00 g, 3.07 mmol, 81% ee)

with Mg powder (1.49 g, 30.7 mmol, 20 equiv) in MeOH (31 mL,

10 mL/mmol): to a solution of 5b in MeOH at 0 °C, Mg powder was

added and the suspension stirred for 7 h at 0 °C. After quenching

with a small volume of both 2 M HCl and H₂O, the product was ex-

tracted with CH₂Cl₂. The organic phase was dried (MgSO₄), fil-

tered, and evaporated. Separation of the crude by column

chromatography (EtOAc) gave 15b (0.362 g, 36%) as a white solid;

82% ee (³¹P NMR, HPLC). The meso-isomer 16, which was also

formed in this reaction, was not isolated. In a second reaction with

(1R,5R)-5b (482 mg, 1.48 mmol) and Mg (>0.72 g) in MeOH (30

mL) at 0 °C, both meso-16 (257 mg, 53%) and (2R,5R)-15b (148.4

mg, 31%) were isolated.

¹H NMR (250 MHz, CDCl₃): δ = 1.91–2.15 (m, 2 H), 2.19–2.52 (m,

2 H), 3.01–3.26 (m, 2 H), 7.09–7.36 (m, 10 H), 9.54 (s, 1 H).

¹³C NMR (100 MHz, CDCl₃): δ = 28.4 (d, J_PC= 12.2 Hz, CH₂), 45.3

(d, J_PC= 88.0 Hz, CH), 126.6 (CH), 128.2 (d, J_PC= 5.2 Hz, CH),

128.3 (CH), 135.7 (C).

³¹P NMR (161 MHz, CDCl₃): δ = 66.8 (s).

Analytical data correspond to those reported.^12b

Analysis of enantiomeric excess by ³¹P NMR: 1 (4.8 mg, 0.018

mmol) and quinine (33 mg, 0.10 mmol) were dissolved in CDCl₃

(0.5 mL). δ(³¹P) = 52.07 [minor, (S,S)-1], 51.71 [major, (R,R)-1].

Synthesis 2013, 45, 308–325

324

L. Hintermann et al.

FEATURE ARTICLE

Y.; Daran, J.-C.; Fiaud, J.-C. Tetrahedron 2002, 58, 5895.

(c) Fiaud, J.-C.; Legros, J.-Y. Tetrahedron Lett. 1991, 32,

5089.

Acknowledgment

This project was funded in part by the DFG (SPP 1179 and Emmy-

Noether-Programm to L.H.). P.N. thanks Prof. S. W. Ng, University

of Malaya, for the advice with the refinement of the structures of 8d

and 9d.

(13) A stoichiometric asymmetric protonation approach to the

isomerization of 2 to 3 had already been reported:

(a) Guillen, F.; Moinet, C.; Fiaud, J.-C. Bull. Soc. Chim. Fr.

1997, 134, 371. (b) For a related asymmetric alkylation

approach: Blake, A. J.; Hume, S. C.; Li, W.-S.; Simpkins, N.

S. Tetrahedron 2002, 58, 4589.

Supporting Information for this article including experimental

procedures, product characterization and copies of NMR spectra is avai-

lable online at http://www.thieme-connect.com/ejournals/toc/synthesis

(14) Hintermann, L.; Schmitz, M. Adv. Synth. Catal. 2008, 350,

1469.

S

u

p

onritgInformatio

n

S

u

p

orntigInformaoti

n

(15) Wu, Y.; Singh, R. P.; Deng, L. J. Am. Chem. Soc. 2011, 133,

12458.

(16) Hintermann, L.; Schmitz, M.; Chen, Y. Adv. Synth. Catal.

2010, 352, 2411.

References

(1) This definition of isomerization encompasses stereoiso-

merizations and tautomerization (bond order isomerization),

but not rearrangements, in order to point out the minimal

change to the substrate in the course of the reaction.

(2) (a) Larsen, C. R.; Grotjahn, D. B. J. Am. Chem. Soc. 2012,

134, 10357. (b) Frauenrath, H.; Brethauer, D.; Reim, S.;

Maurer, M.; Raabe, G. Angew. Chem. Int. Ed. 2001, 40, 177.

(c) Fu, G. In Modern Rhodium-Catalyzed Organic

Reactions; Evans, P. A., Ed.; Wiley-VCH: Weinheim, 2005.

(d) Akutagawa, S. Top. Catal. 1997, 4, 271. (e) Otsuka, S.;

Tani, K. Synthesis 1991, 665.

(17) (a) McCormack, W. B. US 2663737, 1953. (b) McCormack,

W. B. Org. Synth. 1963, 43, 73.

(18) (a) Campbell, I. G. M.; Cookson, R. C.; Hocking, M. B.;

Hughes, A. N. J. Chem. Soc. 1965, 2184. (b) Ezzell, B. R.;

Freedman, L. D. J. Org. Chem. 1970, 35, 241. (c) Lukas, B.;

Roberts, R. M. G.; Silver, J.; Wells, A. S. J. Organomet.

Chem. 1983, 256, 103.

(19) (a) Cowley, A. H.; Kemp, R. A.; Lasch, J. G.; Norman, N.

C.; Stewart, C. A. J. Am. Chem. Soc. 1983, 105, 7444.

(b) Cowley, A. H.; Kemp, R. A.; Lasch, J. G.; Norman, N.

C.; Stewart, C. A.; Whittlesey, B. R.; Wright, T. C. Inorg.

Chem. 1986, 25, 740. (c) SooHoo, C. K.; Baxter, S. G. J. Am.

Chem. Soc. 1983, 105, 7443. (d) Polniaszek, R. P. J. Org.

Chem. 1982, 57, 5189.

(3) (a) Mantilli, L.; Mazet, C. Chem. Lett. 2010, 40, 341.

(b) Uma, R.; Crévisy, C.; Grée, R. Chem. Rev. 2003, 103, 27.

(c) Van der Drift, R. C.; Bouwman, E.; Drent, E.

J. Organomet. Chem. 2002, 650, 1.

(4) (a) Inoue, S.; Takaya, H.; Tani, K.; Otsuka, S.; Sato, T.;

Noyori, R. J. Am. Chem. Soc. 1990, 112, 4897. (b) Tani, K.;

Yamagata, T.; Akutagawa, S.; Kumobayashi, H.; Taketomi,

T.; Takaya, H.; Miyashita, A.; Noyori, R.; Otsuka, S. J. Am.

Chem. Soc. 1984, 106, 5208.

(20) (a) King, R. B.; Sadanani, N. D. Synth. React. Inorg. Met.-

Org. Chem. 1985, 15, 149. (b) Issleib, K.; Seidel, W. Chem.

Ber. 1959, 92, 2681.

(21) In this abbreviated nomenclature for cyclic compounds, cis

and trans indicate the relative position of the aryl group at

C2 and/or C5 to the central oxo-substituent at phosphorus,

which is of higher priority than the amino group.

(22) The crystal structures have been deposited at the Cambridge

Crystallographic Data Centre and allocated the deposition

numbers CCDC 901663 (8d) and 901662 (9d).

(23) Bodalski, R.; Janecki, T.; Galdecki, Z.; Glowka, M. L.

Phosphorus Sulfur Relat. Elem. 1982, 14, 15.

(24) Pakulski, Z.; Koprowski, M.; Pietrusiewicz, M. Tetrahedron

2003, 59, 8219.

(25) (a) Wang, G.-W.; Lu, Q.-Q.; Xia, J.-J. Eur. J. Org. Chem.

2011, 4429. (b) Seiser, T.; Cramer, N. Chem. Eur. J. 2010,

16, 3383. (c) Smith, A. B.; Kingery-Wood, J.; Leenay, T. L.;

Nolen, E. G.; Sunazuka, T. J. Am. Chem. Soc. 1992, 114,

1438. (d) Wenkert, E.; Greenberg, R. S.; Kim, H.-S. Helv.

Chim. Acta 1987, 70, 2159. (e) Carter, M. J.; Fleming, I.;

Percival, A. J. Chem. Soc., Perkin Trans. 1 1981, 2415.

(f) Alder, K.; Günzl-Schumacher, M. Chem. Ber. 1959, 92,

822.

(5) Olah, G. A.; Molnár, A. Hydrocarbon Chemistry, 2nd ed;

Wiley Interscience: New York, 2003, 177.

(6) For related asymmetric organocatalytic alkyne/allene

isomerization reactions: (a) Liu, H.; Leow, D.; Huang, K.-

W.; Tan, C.-H. J. Am. Chem. Soc. 2009, 131, 7212. (b) Oku,

M.; Arai, S.; Katayama, K.; Shioiri, T. Synlett 2000, 493.

(c) Inokuma, T.; Furukawa, M.; Suzuki, Y.; Kimachi, T.;

Kobayashi, Y.; Takemoto, Y. ChemCatChem 2012, 4, 983.

(d) Jacobs, T. L.; Dankner, D. J. Org. Chem. 1957, 22, 1424.

(7) A classic example would be the conversion of tartaric acid

into mixtures containing meso- and rac-tartaric acid by

heating in water at 170 °C: Pasteur, L. C. R. Hebd. Seances

Acad. Sci. 1852, 35, 176.

(8) For a discussion of the feasibility of asymmetric catalysis of

equilibrium reactions, see: (a) Hintermann, L. In

Asymmetric Synthesis — More Methods and Applications;

Christmann, M.; Bräse, S., Eds.; Wiley-VCH: Weinheim,

2012, 109. (b) Hintermann, L.; Dittmer, C. Eur. J. Org.

Chem. 2012, 5573.

(9) Japp, F. R.; Moir, J. J. Chem. Soc., Trans. 1900, 608.

(10) Williams, O. F.; Bailar, J. C. J. Am. Chem. Soc. 1959, 81,

4464.

(11) Isomerization of amarine to isoamarine requires excess

NaOH in ethylene glycol at 155 °C.¹⁰Attempted

isomerization of N-tosyl-amarine with cinchona alkaloids

was not successful (Hintermann, L., unpublished result).

The approach is pointless for free amarine, since isoamarine

racemizes under relatively mild conditions: Busacca, C. A.;

Bartholomeyzik, T.; Cheekoori, S.; Grinberg, N.; Lee, H.;

Ma, S.; Saha, A.; Shen, S.; Senanayake, C. H. J. Org. Chem.

2008, 73, 9756.

(26) Wenkert, E.; Leftin, M. H.; Michelotti, E. L. J. Chem. Soc.,

Chem. Commun. 1984, 617.

(27) Precedence for base-mediated 2,5-dihydro-1H-phosphole to

4,5-dihydro-1H-phosphole rearrangements: (a) Keglevich,

G.; Forintos, H.; Keseru, G. M.; Hegedus, L.; Toke, L.

Tetrahedron 2000, 56, 4823. (b) Tolmachev, A. A.; Ivonin,

S. P.; Kharchenko, A. V.; Lampeka, R. D.; Kozlov, E. S.

J. Gen. Chem. USSR (Engl. Transl.) 1991, 61, 852.

(28) (a) Kappe, O. Angew. Chem. Int. Ed. 2004, 43, 6250.

(b) Loupy, A. Microwaves in Organic Synthesis; Wiley-

VCH: Weinheim, 2002.

(29) Obermayer, D.; Gutmann, B.; Kappe, C. O. Angew. Chem.

Int. Ed. 2009, 48, 8321.

(12) (a) Guillen, F.; Fiaud, J.-C. Tetrahedron Lett. 1999, 40,

2939. (b) Guillen, F.; Rivard, M.; Toffano, M.; Legros, J.-

(30) Hiemstra, H.; Wynberg, H. J. Am. Chem. Soc. 1981, 103,

417.

Synthesis 2013, 45, 308–325

FEATURE ARTICLE

Organocatalytic Asymmetric Isomerization

325

(31) (a) Connon, S. J. Chem. Commun. 2008, 2499. (b) Okino,

T.; Hoashi, Y.; Takemoto, Y. J. Am. Chem. Soc. 2003, 125,

12672.

(32) The NMR method of ee determination tends to

underestimate enantioselectivity. Concerning the chiral

HPLC analysis of 5b, see: Hintermann, L. J. Org. Chem.

2007, 72, 9790.

(33) Pakulski, Z.; Demchuk, O. M.; Kwiatosz, R.; Osinski, P. W.;

Swierczynska, W.; Pietrusiewicz, K. M. Tetrahedron:

Asymmetry 2003, 14, 1459.

(34) Chadwick, D. J.; Willbe, C. J. Chem. Soc., Perkin Trans. 1

1977, 887.

M. E.; Jackson, M.; Lennon, I. C.; Klosin, J.; Abboud, K. A.

J. Org. Chem. 2008, 73, 775.

(41) Barton, D. H. R.; Robinson, C. H. J. Chem. Soc. 1954, 3045.

(42) Polniaszek, R. P.; Foster, A. L. J. Org. Chem. 1991, 56,

3137.

(43) For another NMR method of ee determination that also relies

on the chirality of the sample itself, see: Feringa, B. L.;

Smaardijk, A.; Wynberg, H. J. Am. Chem. Soc. 1985, 107,

4798.

(44) Nolde, C.; Schürmann, M.; Mehring, M. Z. Anorg. Allg.

Chem. 2007, 633, 142.

(45) Precedence: Suga, T.; Ohta, S.; Aoki, T.; Hirata, T. Chem.

Lett. 1985, 1331.

(35) See the Supporting Information for additional illustration or

examples.

(46) In our communication,¹⁴we stated in Scheme 3 that the signs

of the optical rotation were the same in CH₂Cl₂and MeOH

solution. While this is true for 5b, 15b and 1, it is wrong for

19. The same scheme also gave the wrong sign of rotation

for Fiaud’s acid 1; the enantiomer (2R,5R)-1 is dextro-

rotatory in both MeOH (this work) and CH₂Cl₂.^12bThe

configurational assignments in that paper are not affected by

these errors

(36) Westheimer, F. Chem. Rev. 1961, 61, 265.

(37) Kresge, A. J. Acc. Chem. Res. 1975, 8, 354.

(38) (a) Duhaime, R. M.; Weedon, A. C. J. Am. Chem. Soc. 1987,

109, 2479. (b) Malhotra, S. K.; Ringold, H. J. J. Am. Chem.

Soc. 1965, 87, 3228.

(39) Lampin, J. P.; Mathey, F.; Bartet, B. Bull. Soc. Chim. Fr.

1971, 317.

(40) (a) Hols, J.; Gensow, M.-N.; Zayas, O.; Börner, A. Curr.

Org. Chem. 2007, 11, 61. (b) Galland, A.; Paris, J. M.;

Schlama, T.; Guillot, R.; Fiaud, J.-C.; Toffano, M. Eur. J.

Org. Chem. 2007, 863. (c) Pilkington, C. J.; Zanotti-Gerosa,

A. Org. Lett. 2003, 5, 1273. (d) Jackson, M.; Lennon, I. C.

Tetrahedron Lett. 2007, 48, 1831. (e) Dobrota, C.; Toffano,

M.; Fiaud, J.-C. Tetrahedron Lett. 2004, 45, 8153. (f) Fox,

(47) Love, B. E.; Jones, E. G. J. Org. Chem. 1999, 64, 3755.

(48) Hintermann, L. Beilstein J. Org. Chem. 2007, 3, 22.

(49) Maerkl, G.; Alig, B. J. Organomet. Chem. 1984, 273, 1.

(50) Obtained by hydrogenation of 4b over Pd/C with H₂(g),

compare with ref. 12b.

(51) Hum, G.; Wooler, K.; Lee, J.; Taylor, S. D. Can. J. Chem.

2000, 78, 642.

Synthesis 2013, 45, 308–325

Article Doi

Organocatalytic stereoisomerization versus alkene isomerization: Catalytic asymmetric synthesis of 1-hydroxy-trans -2,5-diphenylphospholane 1-oxide

DOI: 10.1055/s-0032-1316835

Source and publish data:

Authors:

Article abstract of DOI:10.1055/s-0032-1316835

Full text of DOI:10.1055/s-0032-1316835

Products guided by the article

R&D Labs maybe for 1428255-72-4

Relevant to this article

Hot Product