Iridium-catalysed allylic substitution: Stereochemical aspects and isolation of IrIII complexes related to the catalytic cycle

FULL PAPER

Iridium-Catalysed Allylic Substitution: Stereochemical Aspects and Isolation of

Ir^IIIComplexes Related to the Catalytic Cycle

[a]

´

Björn Bartels, Cristina Garcıa-Yebra, Frank Rominger, and Günter Helmchen*

Dedicated to Professor Gottfried Huttner on the occasion of his 65th birthday

Keywords: Iridium / Allyl complexes / NMR spectroscopy / Allylic substitution / Asymmetric catalysis

Ir-catalysed allylic alkylations of enantiomerically enriched

monosubstituted allylic acetates proceed with up to 87% re-

tention of configuration using P(OPh)₃as ligand. High regio-

ylbenzylamide was identified as a suitable nucleophile for

allylic aminations. Of particular importance is the use of li-

thium chloride as an additive, generally leading to increased

and enantioselectivity of up to 86% ee in asymmetric allylic enantioselectivities. Two (π-allyl)Ir^IIIcomplexes were charac-

alkylations of achiral or racemic substrates is achieved with

monodentate phosphorus amidites as ligands. Lithium N-tos-

terised by X-ray crystal structure analysis and spectroscopic

data.

Introduction

mory of the chiral sense of the stereogenic centre in the

branched substrate 1 is lost during the reaction.

Asymmetric transition-metal-catalysed allylic substitu-

tion is a useful reaction in organic synthesis.^[1]The selecti-

vity of this reaction is a function of many factors, for

example the metal ion, ligands, nucleophile and substituents

of the allyl system. Over the last few years, work has been

directed at finding catalysts that favour the formation of

branched, chiral products 3 in the substitution of monosub-

stituted allylic substrates 1 and (E)-2 (Scheme 1). With pa-

lladium complexes this is so far only possible for special

cases.^[2]With Mo- or W-based catalysts, branched products

are generally preferred, with higher levels of both regio- and

enantioselectivity reported for substrates with R ϭ aryl.^[3]

It is assumed that in these cases the reactions proceed via

π-allyl complexes, which can isomerize^[3d]via πϪσϪπ re-

arrangement or related processes in such a way that me-

Substitutions catalysed by Rh, Fe or Ru complexes differ

from the above as they generally proceed for branched sub-

strates 1 with a high degree of conservation of enantiomeric

excess (cee). It has been proposed that these reactions pro-

ceed via σ-allyl or π-allyl complexes, which isomerize

slowly.^[4]As a consequence, from enantiomerically enriched

substrates 1, even with achiral ligands enantiomerically en-

riched products 3 are obtained via reactions proceeding

with double inversion.

Not many examples of Ir^I-catalysed allylic substitutions

have been published so far. In the last few years, it has been

shown that catalysts prepared by combining [IrCl(COD)]₂

with a strong π-acceptor ligand — for example triphenyl-

phosphite — in situ give rise to excellent regioselectivities in

favour of the branched product 3 for both aryl- and alkyl-

substituted allylic substrates.^[5]In 1997, we described the

first enantioselective Ir^I-catalysed allylic alkylations of aryl-

allyl acetates (E)-2, which gave high levels of regio- and en-

antioselectivity in favour of product 3 when the bidentate

phosphinooxazoline L1 (Figure 1) was used as the chiral

ligand.^[6]Better results were later achieved with monodent-

ate chiral strong π-acceptor ligands.^[7]Thus, branched prod-

ucts 3 are obtained in Ir^I-catalysed allylic alkylations of

both monoaryl- and monoalkylallyl acetates, (rac)-1 and

(E)-2, with ligands such as phosphites and phosphorus ami-

dites [(R)-L2]. It was also found that Ir^I-catalysed allylic

alkylations of enantiomerically enriched acetates 1 with

achiral ligands occurred with a noticeable ‘‘memory effect’’,

i.e. partial conservation of the enantiomeric purity of the

starting substrates. For example, a sample of (R)-1 with

91% ee furnished (R)-3 with 51% ee [R ϭ (CH₂)₂Ph].^[7a]

Scheme 1. General scheme for the Ir^I-catalysed allylic alkylation of

monosubstituted allylic acetates

[a]

Organisch-chemisches Institut der Universität Heidelberg,

Im Neuenheimer Feld 270, 69120 Heidelberg, Germany

Fax: (internat.) ϩ49-6221/544-205

E-mail: en4@popix.urz.uni-heidelberg.de

Eur. J. Inorg. Chem. 2002, 2569Ϫ2586  WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002 1434Ϫ1948/02/1010Ϫ2569 $ 20.00ϩ.50/0

2569

´

B. Bartels, C. Garcıa-Yebra, F. Rominger, G. Helmchen

FULL PAPER

Scheme 3. Synthesis of aldehydes 5; TPAP: (nPr₄N)[RuO₄]; MOM:

methoxymethyl; DMAP: 4-dimethylaminopyridine; NMO: N-

methylmorpholine N-oxide

Figure 1. Ligands used in Ir^I-catalysed substitutions

Enantiomerically pure substrates 1a؊c were prepared

either by esterification of commercially available enantio-

merically pure alcohols (6a and 6c) or by enzyme (Novozym

435^) catalysed kinetic resolution of rac-6b with vinyl acet-

ate (1b) (Scheme 4).^[12]The absolute configuration of (R)-

6b was confirmed by X-ray crystal analysis of the corres-

ponding ester 12 of (1S)-camphanic acid (Figure 2).

This observation led us to consider the formation of π- as

well as (σ-allyl)Ir^IIIcomplexes as intermediates, which un-

dergo slow isomerization.^[4d,8]

Herein we present a full account of our studies con-

cerning the various factors affecting the course of the Ir^I-

catalysed allylic alkylation and amination as well as the

isolation and full characterization of Ir^IIIcomplexes related

to the proposed intermediate species, in order to contribute

to the elucidation of the still unclear mechanisms of these

processes.

Results and Discussion

Synthesis of Allylic Substrates

The branched substrates 1 were prepared by reaction of

aldehydes 5 with vinylmagnesium bromide and esteri-

fication of the resulting alcohols 6. The linear substrates

(E)-2 were obtained by WittigϪHornerϪEmmons ole-

fination to give the (E)-α,β-unsaturated ester 7 which, after

reduction to alcohols 8 and esterification, gave the acetates

(E)-2 (Scheme 2).

Scheme 4. Kinetic resolution of the alcohol 6b by enzyme-catalysed

esterification and confirmation of the absolute configuration of

the products

Substrates 1e,f and (E)-2f were prepared in a similar way;

the requisite aldehydes^[9]were not commercially available

Substrates (Z)-2 were prepared as described in Scheme 5.

and were prepared according to the procedure described in (Z)-2e was obtained from the commercially available (Z)-2-

Scheme 3, involving formation of the orthoester 10 by reac- buten-1,4-diol (13) via reaction with trimethyl orthofor-

tion of diol 9 with trimethyl orthoformate, reduction of the mate, reduction of the resulting orthoester 14, and sub-

orthoester with DIBAL-H^[10]and, finally, oxidation of the sequent esterification.^[10]The synthesis of the homologous

resulting alcohol 11 with NMO/TPAP.^[11]

alkene (Z)-2f was performed starting from 3-butyn-1-ol (16)

Scheme 2. Synthesis of substrates 1 and (E)-2; DIBAL-H: Diisobutylaluminium hydride; DMAP: 4-dimethylaminopyridine

2570 Eur. J. Inorg. Chem. 2002, 2569Ϫ2586

Iridium-Catalysed Allylic Substitution

FULL PAPER

Table 1. Allylic alkylation of acetates 1 and (E)-2 catalysed by Ir^I

complexes according to Scheme 1^[a]

Entry Substrate Ligand L/Ir

T

Time Yield [%]^[b]Ratio^[c]

[°C] [h] 3 and (E)-4 3:(E)-4

1

2

3

4

5

6

7

8

9

10

11

12

13

14

1a

P(OPh)₃1:1 25

3

99

98

15

98

89

58

99

66

0

60

99

0

90

98:2

32:68

64:36

95:5

89:11

Ϫ

83:17

95:5

Ϫ

95:5

Ϫ

0

25

PPh₃

1:1 25

(E)-2a P(OPh)₃1:1 25

3

(E)-2a

1b

Ϫ

0

65

24

3

22

3

18

PPh₃

1:1 65

P(OPh)₃1:1 25

1b

Ϫ

0

25

1b

PPh₃

1:1 25

1:1 55

1b

(E)-2b P(OPh)₃1:1 25

(E)-2b

1b

Ϫ

0

25

PPh₃

1:1 25

Figure 2. Crystal structure of ester 12 prepared by reaction of the

enantiomerically pure alcohol (R)-6b with (1S)-camphanic acid

chloride

P(OPh)₃2:1 25

^[a]All reactions were carried out in THF on a 0.5 mmol scale (0.125

) with 2 equiv. of NaCH(CO₂Me)₂and 2 mol % of [IrCl(COD)]₂.

[b]

[c]

Isolated yields are given. Determined by GC/MS.

Ir^Igave the best results (compare entries 7 and 14). In this

respect, it is significant that only one complex is formed by

reaction of [IrCl(COD)]₂with two equivalents (L/Ir^Iϭ 1:1)

of P(OPh)₃(³¹P NMR: δ ϭ 86.77 ppm),^[14]while an excess

of ligand gives rise to the formation of additional species.^[5b]

It is also remarkable that while the regioselectivities ob-

tained from the reaction of the branched substrates are gen-

erally high, the linear substrates give rise to lower degrees

of regioselectivity, with PPh₃showing a particularly strong

‘‘memory effect’’ (entries 6 and 10).

Scheme 5. Synthesis of substrates (Z)-2e and (Z)-2f (CSA: cam-

phorsulfonic acid)

It has been reported that the geometry of the double

bond of the linear substrates is also a factor determining

the regioselectivity of Ir^I-catalysed allylic substitutions.^[15]

Using substrates 1e, 1f and 2f (Scheme 1), the dependence

on the geometry of the double bond in combination with

the influence of σ-acceptor substituents in the substrate was

studied. The results (Table 2) clearly show that the allylic

which, after a three-step sequence, afforded the (Z)-alkene

19, which was then esterified to give acetate (Z)-2f.^[13]

Factors Affecting the Rate and Regioselectivity of Ir^I-

Catalysed Allylic Alkylation

Allylic alkylations of aryl- and alkyl-substituted allyl alkylations of (Z)-2e and (Z)-2f preferentially yield (Z)-4e

acetates 1 and (E)-2 catalysed by [IrCl(COD)]₂and com- and (Z)-4f as products, respectively. It is also remarkable

binations of this complex with PPh₃or the strong π-ac- that the (Z)-substrates react distinctly faster than the (E)-

ceptor ligand P(OPh)₃were initially investigated. The re- substrates. Furthermore, it is apparent from Table 1 and 2

sults are summarized in Table 1.

that regioselectivity (3:4) is a function of the substituent at

Concerning the rate of the reaction, the allylic alkylation the allylic moiety, giving rise to the order Ph Ͼ (CH₂)₂Ph

of linear substrate (E)-2a was slightly slower than that of Ͼ (CH₂)₂OMOM Ͼ CH₂OMOM, reflecting the ability of

the isomeric branched substrate 1a when [IrCl(COD)]₂the substituent to stabilize a positive charge.

without additional ligand was used as catalyst (Table 1, ent-

Some of the results described above contribute to the un-

ries 2 and 5). Differences in reaction rates between derstanding of the mechanism of the Ir^I-catalysed allylic al-

branched and linear substrates have been observed for the kylation. Generally, the following steps are to be consid-

Rh^I-catalysed reaction and they have been attributed to a ered: a) oxidative addition of the substrate to the metal

S_N2Ј mechanism operating in the oxidative addition step.^[4d]centre and formation of π- and/or σ-allyl complexes, b) iso-

While the combination of [IrCl(COD)]₂with PPh₃led to merization of the intermediates via πϪσϪπ rearrangement

slow reactions and low degrees of regioselectivity (entries 6, or similar processes, and c) attack of a nucleophile

9 and 13), the π-acceptor ligand P(OPh)₃gave rise to the (Scheme 6). If π-allyl complexes are the product determin-

shortest reaction times along with optimal yields and regio- ing intermediates, it is expected that cationic character fa-

selectivities in favour of the branched product 3 (entries 1, vours attack at the substituted allylic terminus. It is, there-

4, 7 and 11).^[5a][5b]Furthermore, a 1:1 ratio of P(OPh)₃and fore, clear that the rate and regioselectivity of the nucleo-

Eur. J. Inorg. Chem. 2002, 2569Ϫ2586

2571

´

B. Bartels, C. Garcıa-Yebra, F. Rominger, G. Helmchen

FULL PAPER

Steric Course of the Substitution and Memory Effects

It has already been reported that Rh^I- and Ir^I-catalysed

Table 2. Influence of the configuration of the double bond of the

substrates on the regioselectivity of the substitution^[a]

allylic substitutions show

a noticeable ‘‘memory ef-

Entry Substrate

R

T

Time Yield Ratio^[b]E/Z

[°C] [h] [%]

fect’’,^[4d,7a,19]i.e. enantiomerically enriched branched sub-

strates 1 yield substitution products with retention of con-

figuration and partial conservation of enantiomeric purity.

Rather than aiming at high degrees of conservation we were

interested in finding conditions that would maximally erode

enantiomeric purity in order to achieve asymmetric induc-

tion via dynamic kinetic resolution. To this end, the influ-

ence of reaction conditions, such as concentration of the

nucleophile, temperature, influence of additives and ligand,

were studied in allylic alkylations of enantiomerically en-

riched acetates 1 using [IrCl(COD)]₂combined in situ with

achiral ligands P(OPh)₃and L3ϪL5 as catalysts. The results

are summarized in Table 3.

3:4 in 4^[b]

1

2

3

4

5

1e

(Z)-2e

1f

CH₂OMOM

CH₂CH₂OMOM 25 18

25 18

24

91

65

94

95

80

77:23 99:1

0:100 8:92

93:7 99:1

97:3 99:1

12:88 1:99

4

(E)-2f CH₂CH₂OMOM 25 18

(Z)-2f CH₂CH₂OMOM 25

4

^[a]All reactions were carried out in THF on a 0.5 mmol scale (0.125

) with 2 equiv. of NaCH(CO₂Me)₂, 2 mol % [IrCl(COD)]₂and 4

mol % P(OPh)₃. Determined by GC/MS.

[b]

The results allow the following conclusions: (a) phos-

phorus amidite L3 gives rise to maximal erosion of enanti-

omeric purity (entries 7 and 8); (b) the highest degree of

conservation of enantiomeric purity is induced by

[IrCl(COD)]₂without additional ligand (entry 11); (c) sur-

prisingly, there is almost no influence of the concentration

of the nucleophile (cf. entries 1 and 2). It was expected that

an increase would lead to a higher degree of conservation

of enantiomeric purity because of acceleration of the nucle-

ophilic attack relative to isomerization of allyl complexes;

(d) halide ions are known to accelerate the isomerization of

allyl complexes via σ-allyl complexes.^[20]Indeed, addition

of LiCl induces an enhanced degree of racemisation (entries

5, 6 and 8), (e) finally, increase of the reaction temperature

also favours racemisation (entry 6).

Scheme 6. Ir^I-catalysed allylic substitution of substrates 1 and 2 via

σ- and π-allyl intermediates

Concerning the steric course of the process, it was of in-

terest to establish whether the Ir^I-catalysed reactions pro-

ceed with double inversion or retention. To this end, reac-

tions with substrates 20 and 21 were investigated

(Scheme 7).^[21]These substrates do not allow a double in-

version process because in 20 oxidative addition to Ir^Iand

in 21 nucleophilic attack at the (allyl)Ir^IIIintermediate are

both prevented by steric hindrance. Indeed, no reaction oc-

curred under reaction conditions typical for allylic al-

kylations.^[22]Control experiments showed that under the

same conditions the reaction of NaCH(CO₂Me)₂with

cyclopentenyl acetate proceeds to completion. These results

demonstrate that the Ir^I-catalysed allylic alkylation pro-

ceeds via a double inversion process.

philic attack would be higher with π-acceptor ligands such

as P(OPh)₃and olefins (such as COD) than with σ-donor

ligands such as PPh₃.^[16]It has been argued that nucleo-

philic attack on analogous Rh^IIIcomplexes is determined

by the location of the acceptor ligand in the trans position

to the higher substituted allyl terminus.^[17]This argument

neglects the fact that due to the strong trans influence of

both an allyl and a phosphorus-based ligand, these ligands

strongly prefer a mutual cis disposition. Furthermore, there

is no marked influence of the steric bulk of the phosphorus

ligand on the regioselectivity.^[5b]

The fact that no (Z)-products were obtained from (E)-

substrates while these are the major products in the allylic

alkylation of (Z)-substrates is highly significant and is also

found in the reaction catalysed by Fe complexes.^[4b]As (E)-

substrates would form syn π-allyl intermediates and (Z)-

substrates anti π-allyl intermediates, these results clearly

show that the syn-anti isomerization between these species

is a slow process (Scheme 6). Considering a generally close

analogy of Ir^I- and Rh^I-catalysed allylic substitutions, it is

of interest to note that reactions of isolated anti (π-al-

lyl)Rh^IIIcomplexes with carbon nucleophiles occur at the

less substituted terminus of the allylic moiety, giving rise

exclusively to the formation of (Z)-olefin complexes.^[18]On

the other hand, there are no facts which would rule out a

reaction course via σ-complexes.

Scheme 7. Ir^I-catalysed allylic substitutions

2572

Eur. J. Inorg. Chem. 2002, 2569Ϫ2586

Iridium-Catalysed Allylic Substitution

Table 3. Influence of ligand, reaction conditions and addition of lithium chloride on allylic substitutions of substrates 1aϪ1c^[a]

FULL PAPER

Entry

Substrate

ee [%] of 1

(conf.)

Ligand

Time

[h]

c_Nu

[]

Additive

(equiv.)

Yield

[%]

Ratio

ee [%]^[b]

(conf.)

cee

3:(E)-4^[b]

[%]^[c]

1

2^[d]

3^[d]

4

1b

1a

1c

95 (R)

98 (S)

91 (R)

98 (S)

91 (R)

98 (S)

91 (R)

98 (S)

94 (R)

99.8 (R)

P(OPh)₃

L3

3

18

3

0.250

0.125

0.250

؊

95

79

98

97

71

91

98

61

99

93

67

98

91

95:5

96:4

95:5

93:7

91:9

99:1

79:21

84:16

53:47

91:9

95:5

54 (R)

60 (R)

65 (S)

51 (R)

49 (R)

38 (S)

39 (R)

34 (R)

70 (S)

66 (R)

85 (S)

56 (S)

85 (R)

57

63

66

56

54

39

43

37

71

73

87

60

85

5

LiCl (1.0)

6^[e]

7

؊

8

9

10

11

12

13

L3

LiCl (1.0)

L4

؊

L5

Ϫ

18

3

P(OPh)₃

[a]

All reactions were carried out in THF at a 0.5 mmol scale (0.125 ) with 2 mol % [IrCl(COD)]₂and 4 mol % ligand, at room

[b]

[c]

temperature.

Determined by HPLC (entries 1Ϫ12) or GC (entry 13).

Retention or cee (conservation of enantiomeric excess) is

[d]

[e]

defined as [ee(product)/ee(substrate)]ϫ100. Ratio substrate/nucleophile 1:1. Temperature: 50 °C.

To gain further insight, the alkylation of the symmetric- for tetracoordinate planar Pd complexes.^[23]However, for

ally substituted acetate 22 was studied (Scheme 8).^[7a]In octahedral Ir^IIIor Rh^IIIcomplexes the trans disposition of

this reaction enantiomerically pure (R)-22 gave the product an allyl and a phosphorus ligand, i.e. B1, is destabilised

(R)-23 with 71% ee.

by their strong trans influence.^[24]A complex B2 with cis

disposition of these ligands is expected to be more stable.

Rescue of the argument, i.e. control of the substitution via

the trans effect operating in B1, would require further as-

sumption of a dynamic system with strong kinetic prefer-

ence for attack of a nucleophile at the trans complex B1.

Also conceivable is a reaction via a (π-allyl)metal com-

plex with syn-anti configuration of the allyl ligand. Accord-

ing to precedents in Pd-catalysed allylic substitutions, such

a complex might react preferentially at the terminus carry-

ing the anti substituent.^[25]In view of the results for the Ir^I-

catalysed allylic substitution of monosubstituted (Z)-sub-

strates (vide infra), i.e. preferential formation of linear

product, this possibility appears unlikely.^[15]

Scheme 8. Allylic substitution using a substrate with a potentially

symmetric (π-allyl)Ir^IIIintermediate

This result rules out the intermediacy of a symmetric (π-

allyl)Ir^IIIcomplex or a fast dynamic equilibrium between

isomeric complexes, as is typical in palladium chemistry.

Possible intermediates (Figure 3) are a σ-complex A,

formed by direct oxidative addition with inversion, or a cor-

responding π-complex B with a nonsymmetric ligand

sphere. A third type of intermediate conceivable is a nondy-

namic (σϩπ)-allyl complex C formed by S_N2Ј attack of an

Ir^Ispecies. Complexes of this type have been proposed for

Rh^I-catalysed allylic substitutions which proceed with al-

most complete conservation of enantiomeric purity.^[4d]In

principle, proposals Β and C are identical as they are simply

different ways of describing a nonsymmetric π-allyl com-

plex. Assessing such intermediates is fairly straightforward

Asymmetric Catalysis

The results in Table 3 show that isomerization among the

various intermediate species is fastest with phosphorus ami-

dites as ligands and, therefore, these appeared particularly

suited for asymmetric Ir^I-catalysed allylic substitutions. In

preliminary experiments we found that branched substrates

yield higher degrees of enantio- as well as regioselectivity

than linear substrates. Accordingly, reactions with branched

substrates were studied preferentially. Using (R)-L2 as the

chiral ligand, the dependence of the enantioselectivity on

the reaction conditions and substrate structure were invest-

igated. These results are displayed in Table 4.

The ratio of ligand to iridium is an important parameter.

In analogy to the earlier finding (Table 1), a 1:1 ratio was

optimal (compare entries 2Ϫ4). In view of the previous dis-

cussion concerning allylic intermediates it is significant that

a decrease of temperature leads to a lowering of enantiose-

lectivity (cf. entries 2, 5 and 6). This is probably caused

by a slow isomerization of intermediary allyl complexes. A

further aspect to take into account is the influence of the

Figure 3. Possible intermediates in the allylic substitution of acetate

(R)-22

Eur. J. Inorg. Chem. 2002, 2569Ϫ2586

2573

´

B. Bartels, C. Garcıa-Yebra, F. Rominger, G. Helmchen

FULL PAPER

Table 4. Variation of reaction parameters in asymmetric allylic alkylations using (R)-L2 as chiral ligand^[a]

Entry

X

Solvent

Base

Ratio

P/Ir

Additive^[b]

T

Time

Yield

[%]^[c]

3:(E)-4^[d]

ee of 3

[%] (R)^[e]

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

OAc

Cl

THF

ether

toluene

CH₂Cl₂

THF

NaH

nBuLi^[g]

nBuLi

Cs₂CO₃

BSA^[i]

NaH

0

1

2

4

1

Ϫ

room temp.

0 °C

3 h

6 h

1 h

3 h

3 d

3 h

7 d

3 h

5 d

3 h

66

92

66

5

89:11

99:1

Ϫ^[f]

Ϫ

69

68

Ϫ^[f]

47

68

0

25

0

25

0

74

98

88

47

41

65

91

10

30

91

76

13

95

15

99

83

97

69

99:1

98:2

60:40

85:15

82:18

99:1

98:2

99:1

98:2

86:14

99:1

Ϫ^[f]

99:1

98:2

65 °C

room temp.

50 °C

OCO₂Et

OPO(OEt)₂

OAc

53

60

3

ZnCl₂

Ϫ

NMe₄Br

N(nBu)₄Cl

LiF

LiCl

LiBr

[h]

29

20

49

25

58

86

80

83

room temp.

[a]

[b]

Nucleophiles were prepared by treatment of CH₂(CO₂Me)₂with base over a period of 30 min at room temperature in THF.

One

equivalent (relative to substrate) of additive was added before the nucleophile. ^[c]Isolated yields. ^[d]Determined by GC/MS. ^[e]Determined

[f]

[g]

by HPLC. Not determined.

The nucleophile was generated by addition of nBuLi at Ϫ78 °C and keeping the reaction mixture at

room temperature for 30 min. ^[h]The nucleophile was generated by reacting the base with dimethyl malonate in THF for 30 min at reflux

[i]

temperature. BSA ϭ N,O-bis(trimethylsilyl)acetamide; KOAc (catalytic quantity) was added to start the reaction.

leaving group on rate, regio- and enantioselectivity (cf. ent- 1a. It is quite clear that further progress in this area will

ries 2, 7Ϫ9). Substrates with better leaving groups than depend on establishing conditions for further enhanced

acetate give rise to very low selectivities. As of now, the rates of isomerization of allylic intermediates.

best results are obtained with acetates. Of the solvents used,

tetrahydrofuran was found to be the most suitable (entries

2, 10Ϫ12).

Variation of the cation (entries 13Ϫ17) did not lead to an

Ir^I-Catalysed Allylic Amination

Ir^I-Catalysed allylic aminations have so far only been de-

improvement of the enantioselectivity. This was surprising scribed for amines as nucleophiles.^[15]These reactions re-

because an improvement had been achieved previously with quire polar solvents and/or high concentration and rela-

zinc derivatives in a closely related reaction.^[7b]In contrast, tively high temperature. We have now found that lithium N-

addition of anhydrous lithium bromide or lithium chloride tosylbenzyl amide, already successfully employed in Rh^I-

resulted in an improved regio- and enantioselectivity (up to catalysed allylic aminations,^[19a,26]reacts under the condi-

86% ee) (entries 20Ϫ22).

tions typical for malonates (Scheme 9). Some preliminary

Using the reaction conditions optimized for 1b, allylic results obtained in allylic aminations of both branched and

substitutions with a set of systematically varied substrates linear substrates with this nucleophile are presented here

were investigated (Table 5). In every case addition of LiCl (Table 6). With triphenylphosphite as ligand, only moderate

led to markedly improved enantioselectivities; this is prob- yields and regioselectivities were obtained in the amination

ably due to an acceleration of the isomerization of the inter- of substrate 1b (entry 1). With the chiral phosphorus amid-

mediary allyl complexes. A most surprising result was the ite (R)-L2 as ligand (entries 2Ϫ3) yield (up to 97 %) and

low enantioselectivity obtained with the phenyl derivative regioselectivity (up to 99:1) were improved for the branched

2574

Eur. J. Inorg. Chem. 2002, 2569Ϫ2586

Iridium-Catalysed Allylic Substitution

FULL PAPER

Table 5. Asymmetric allylic substitutions with substrates 1aϪf using (aR)-L2 as chiral ligand^[a]

Entry

Substrate

R

Additive^[b]

Time

[h]

Yield

[%]

Ratio^[c]

3:(E)-4

ee [%]^[d]

(config.)

1

2

3

4

5

6

7

8

9

1c

1a

1b

1d

1e

1f

Me

Ph

Ϫ

LiCl

Ϫ

LiCl

Ϫ

LiCl

Ϫ

LiCl

Ϫ

3

99

98

97

92

83

41

31

93

90

94

78

95:5

94:6

98:2

99:1

94:6

95:5

99:1

98:2

31 (R)

57 (R)

8 (S)

Ph

4 (R)

CH₂CH₂Ph

iPr

69 (R)

86 (R)

66 (R)

74 (R)

49 (R)

78 (R)

31 (R)

70 (R)

3

48

3

iPr

CH₂OMOM

CH₂CH₂OMOM

10

11

12

LiCl

Ϫ

LiCl

1f

3

[a]

All reactions were carried out in THF at 25 °C on a 0.5 mmol scale [0.125 ] with 2 equiv. NaCH(CO₂Me)₂, 2 mol % [IrCl(COD)]₂

and 4 mol % (R)-L2; a general procedure (II) for the Ir^I-catalysed allylic alkylation is described in the Exp. Sect. (see below).

equivalent (relative to substrate) of additive was added before addition of the nucleophile. Determined by GC/MS. Determined by

HPLC or GC (for details see Exp. Sect.).

One

[b]

[c]

[d]

product 27b. Further work is required in order to improve meric complexes of the type [IrCl(COD)L].^[27]Thus, with

the enantioselectivity of this reaction.

the chiral ligand (R)-L2, the complex [IrCl(COD){(R)-L2}]

(29) was prepared and fully characterised by NMR spectro-

scopy and X-ray diffraction. The crystal structure of 29 to-

gether with a selection of bond lengths is shown in Figure 4.

Scheme 9. Ir^I-catalysed allylic amination of (rac)-1b and (E)-2b

Table 6. Ir-catalysed allylic amination of acetates 1b and (E)-2b^[a]

Entry

Substrate

Ligand

Yield

[%]

Ratio

ee [%]^[b]

(config.)

27:28^[b]

1

2

3

1b

(E)-2b

P(OPh)₃

(R)-L2

47

97

47

77:23

99:1

85:15

Ϫ

11 (S)

13 (S)

[a]

The reaction was carried out on a 0.5 mmol scale in THF (0.1

) (18 h, room temp.), with 2 equiv. of LiN(CH₂Ph)Ts, 2 mol %

[IrCl(COD)]₂and 4 mol % ligand. Determined by HPLC.

˚

Figure 4. X-ray structure of complex 29; selected bond lengths (A):

[b]

Ir(1)ϪC(1) 2.114(6), Ir(1)ϪC(2) 2.118(6), Ir(1)ϪC(5) 2.222(5),

Ir(1)ϪC(6) 2.222(5), Ir(1)ϪP(1) 2.2363(13), Ir(1)ϪCl(1) 2.358(1),

C(1)ϪC(2) 1.434(8), C(5)ϪC(6) 1.371(8)

Characterization of Ir^Iand Ir^IIIComplexes Related to the

Catalytic Cycle

Both Ir^I- and σ- or (π-allyl)Ir^IIIcomplexes are expected

to participate in the catalytic process. Therefore, the isola-

tion and structural characterization of such complexes is of

great interest. We have prepared a variety of these com-

pounds by stoichiometric reactions of the type comprising

the catalytic cycle.

As expected, the coordinated double bond [C(1)-C(2)]

trans to the chloro ligand is noticeably longer than that

[C(5)-C(6)] trans to the phosphorus due to the greater trans

influence of the phosphorus ligand, which is translated into

a weakening of the IrϪC(5,6) bonds and a stronger C(5)-

C(6) double bond character.

Preparation of (π-Allyl)Ir^IIIComplexes

A Chiral Ir^IComplex Containing a Chiral Phosphorus

Amidite

Oxidative addition is a necessary step in the cata-

Reaction of ligand L with complex [IrCl(COD)]₂pro- lytic process. Therefore, we attempted to generate the cor-

ceeds with cleavage of the chloride bridges to give mono- responding (allyl)Ir^IIIintermediates by direct reaction of al-

Eur. J. Inorg. Chem. 2002, 2569Ϫ2586

2575

´

B. Bartels, C. Garcıa-Yebra, F. Rominger, G. Helmchen

FULL PAPER

lyl acetates with the Ir^Iprecursors [IrCl(COD)]₂or com-

plexes [IrCl(COD)(L)]. Surprisingly, despite long reaction

times and a large excess of the allyl acetate, the expected

oxidative addition reaction, which proceeds readily in palla-

dium chemistry,^[1c]did not take place here.

We observed that addition of the nucleophile NaCH-

(CO₂Me)₂to a solution of the catalyst generally produced a

change of colour from pale orange to pale yellow. This

seemed to indicate that the true catalyst is produced by re-

action with the nucleophile. Accordingly, the stoichiometric

reaction of [IrCl(COD)P(OPh)₃],^[14]formed by treatment of

[IrCl(COD)]₂with triphenylphosphite (1:2), with NaCH-

(CO₂Me)₂(1 equiv.) was explored. The ³¹P{¹H} NMR spec-

trum of the mixture, recorded in [D₈]THF after one hour

at room temperature, showed a singlet (δ ϭ 86.3 ppm), cor-

responding to the initial complex, and two sets of doublets

Figure 5. Crystal structure of complex 30; selected bond lengths

˚

(A) with estimated standard deviations: Ir(1)ϪC(1)/C(1B) 2.11(3)/

2.31(5), Ir(1)ϪC(2)/C(2B) 2.24(2)/2.31(2), Ir(1)ϪC(3)/C(3B)

2.45(1)/2.48(1), C(1)ϪC(2)

ϭ 1.44(1), C(1B)ϪC(2B) 1.44(2),

C(2)ϪC(3) 1.35(1), C(2B)ϪC(3B) 1.34(2)

3

of which one (δ ϭ 124.5 and 81.4 ppm, J_P,Pϭ 74 Hz) was

˚

longer than the Ir(1)ϪC(12) bond [2.176(8) A] due to the

trans influence of the COD double bond C(1)ϪC(2) (Fig-

ure 6). Concerning the bond lengths within the allyl moiety,

identified as belonging to the previously reported orthomet-

allation product [Ir{η²-(P,C)-P(OC₆H₄)(OPh)₂}(COD)-

{P(OPh)₃}].^[14]The second set of doublets (δ ϭ 104.3 and

˚

3

it is remarkable that the C(10)ϪC(11) bond [1.36(1) A] is

85.1 ppm, J_P,Pϭ 43 Hz) indicates coordination of two

considerably shorter than the C(11)ϪC(12) bond [1.46(1)

equivalents of ligand per metal centre. Attempts to isolate

and characterise this complex were unsuccessful.

˚

A], i.e. the allyl ligand is highly unsymmetrical and dis-

torted towards a σ-complex.

Next, the oxidative addition of allyl halides to

[IrCl(COD)]₂was studied. The reaction of (E)-cinnamyl

chloride with [IrCl(COD)]₂and allyl bromide with

[IrBr(COD)]₂gave complexes 30 and 31, respectively, in up

to 80% yield (Scheme 10).^[28]In the presence of phos-

phorus-derived ligands such as P(OPh)₃or P(C₆F₅)₃the

same compounds were formed, which demonstrates the

very strong coordination of COD. Complexes 30 and 31

were fully characterised by spectroscopic methods and X-

ray diffraction.

Figure 6. Crystal structure of complex 31; selected bond lengths

˚

(A) with estimated standard deviations: Ir(1)ϪC(10) 2.321(8),

Ir(1)ϪC(11) 2.241(9), Ir(1)ϪC(12) 2.176(8), C(10)ϪC(11) 1.36(1),

C(11)ϪC(12) 1.46(1), C(1)ϪC(2) 1.39(1), C(3)ϪC(4) 1.53(1),

C(5)ϪC(6) 1.38(1), C(7)ϪC(8) 1.53(1)

Scheme 10. Oxidative addition of allyl halides to Ir^Icomplexes

X-ray Crystal Structures of (π-Allyl)Ir^IIIComplexes 30 and

31

Complexes 30 and 31 gave yellow single crystals from pattern of the allyl group is quite common for Ir^[30]or

Asymmetry in π-allyl systems as a consequence of either

an asymmetric set of ligands or an asymmetric substitution

dichloromethane/diethyl ether solutions.

Mo^[31]complexes and distinctly more pronounced than in

The crystal of complex 30 (Figure 5) contains a ca. 1:1 Pd complexes.^[32]For example, Fryzuk et al. have found a

mixture of endo and exo isomers.^[29]As a consequence, the situation similar to that of 30 for the (π-allyl)Ir^IIIcomplexes

structural data of the allyl moieties are of low precision. [Ir(η³-CH₂CHCHMe)(H){N(SiMe₂CH₂PPh₂)₂}]

The configuration of the complex is as expected on the ba- [Ir(η³-CH₂CHCHCN)(H){N(SiMe₂CH₂PPh₂)₂}],

and

which

sis of trans influences of the ligands. The syn configuration display σ-character for the substituted allylic carbon and

of the allyl ligands is typical for aryl-allyl ligands.

double bond character for the CHCH₂moiety. The authors

The allyl ligand of complex 31 was found to be in the describe this situation as a σ-π²-type allyl ligand.^[30d]Many

˚

exo disposition. The Ir(1)ϪC(10) bond [2.321(8) A] is additional examples of such ligands have been

2576

Eur. J. Inorg. Chem. 2002, 2569Ϫ2586

Iridium-Catalysed Allylic Substitution

FULL PAPER

reported.^{[8e,30bϪ30d,31]}Obviously, the previously discussed

‘‘enyl (σϩπ)’’ species proposed as intermediates in Rh^I-

catalysed allylic substitutions (cf. Scheme 8) belong to this

type.^[4d]

NMR Spectroscopic Study of (π-Allyl)Ir^IIIComplexes

(π-Allyl)Ir^IIIcomplexes generally show slow isomeri-

sation compared to π-allyl complexes of the first and second

row late transition metals such as Ni, Pd, Fe, Ru and

Co.^[8,30]In contrast, complexes 30 and 31 display broad

peaks in their ¹H NMR spectra at room temperature, indic-

ating that dynamic processes are taking place.

The ¹H and ¹³C{¹H} NMR spectra of 30 at temperatures

in the range Ϫ60 to Ϫ40 °C show a 2:1 ratio of two species

which, on the basis of the proton coupling constants, are

the syn-π-allyl complexes found in the crystal structure (cf.

1

Figure 7. ¹H,¹H EXSY spectrum of complex 30 recorded at Ϫ40

°C (CDCl₃, 500.13 MHz); cross peaks chosen for the calculations

are enlarged; A and B denote the minor and major isomer, respect-

ively

Scheme 10). The H NMR spectrum of 30 at 50 °C shows

one set of signals as a consequence of fast equilibration;

coalescence was found at ca. 0 °C.

For both isomers of complex 30, the resonances of the

allylic protons appear in the range δ ϭ 4Ϫ6.5 ppm (¹H

NMR spectrum at Ϫ60 °C). The syn-configuration of the

allyl ligand is apparent from the values of the coupling con-

stant (³J_2,3ϭ 14 Hz) for both isomers, indicating the anti-

disposition of 3-H (for numbering see Scheme 10). An un-

usual feature is a marked downfield shift of 3-H for the

minor (δ ϭ 6.33 ppm) relative to the major isomer (δ ϭ

4.93 ppm).

Table 7. ¹H,¹H EXSY data for the determination of the isomeri-

sation rate constant k

[a]

[b]

t_m[ms] X_A

X_B

I_AAI_BB

I_AB

I_BA

r

k [s^Ϫ1

]

25

50

100

0.43

0.42

0.41

0.57 1.00 1.77 0.065 0.069 23.74 3.37

0.58 1.00 1.88 0.150 0.147 10.32 3.89

0.59 1.00 2.08 0.309 0.302 5.16

3.92

The ¹³C{¹H} NMR spectroscopic data (recorded at Ϫ60

°C) were analysed with the help of HMQC and HMBC ^[a]Measured at Ϫ40 °C in CDCl₃. ^[b]The molar fraction X_iis given

experiments. Resonances at δ ϭ 40.9 and 42.6 ppm for the

methylene carbon atoms (C-1) of the major and minor iso-

mer, respectively, are characteristic of the CH₂group of (π-

allyl)Ir^IIIsystems; σ-bound allylic CH₂groups typically ap-

pear in the range δ ϭ 0Ϫ25 ppm. Resonances at δ ϭ 110.6

and 102.2 ppm for C-2 of the major and minor isomer, re-

spectively, are also in the range typical for π systems; σ

systems display values greater than δ ϭ 130 ppm. The res-

onances for the substituted carbon (-CHPh) appear at δ ϭ

111 and 123 ppm for the major and minor isomers, respect-

ively. The latter value is unusually low compared to other

(η³-CH₂CHCHPh)Ir^IIIor (η³-CH₂CHCHPh)Pd com-

plexes.^[8d,33]

by the formula

1

complex 30. Coalescence was found at ca. 50 °C. The H

NMR resonances are found in the expected range. In the

¹³C{¹H} NMR spectrum (Ϫ60 °C), one of the CH₂groups

shows resonances in the expected range (δ ϭ 44.6 and

44.9 ppm) while the other appears at unusually low field

(δ ϭ 79.7 and 89.6 ppm), for the major and minor isomer,

respectively. The latter values are very unusual for (π-al-

lyl)Ir^IIIcomplexes,^{[8aϪ8d,30c,36]}although values at similarly

low field are known for (π-allyl)Pd complexes.^[33]Similar to

complex 30, the resonances of C-2 appear at fairly low field,

δ ϭ 114.5 and 107.9 ppm for the minor and major isomer,

respectively.

The rate of interconversion of the isomers of 30 was de-

1

termined as follows. By recording H-¹H EXSY spectra at

varied mixing times (t_m) at Ϫ40 °C, the magnetisation trans-

fer between chosen atoms present in both isomers, denoted

as A (minor isomer) and B, was measured (Figure 7).^[34]

From the 2D NMR spectra a correlation between the allylic

protons in the isomers was chosen to perform the calcula-

tions. Applying Equation (1) (I_ii, I_ijintegral values of diag-

onal and cross-peaks, respectively, X_imolar fraction)^[35]to

the data summarised in Table 7 a value of k ϭ 4 Hz at Ϫ40

(1)

Activity of Complex 30 in Stoichiometric Reactions with

Nucleophiles and in Catalysis

The low-temperature stoichiometric reaction of complex

°C was determined for the rate constant.

30 with dimethyl 2-sodiomalonate as nucleophile in the

1

The H and ¹³C{¹H} NMR spectra of complex 31 meas- presence of chiral ligand (R)-L2 was investigated. The H

ured at Ϫ60 °C show two isomeric species in a 9:1 ratio; and ³¹P NMR spectra of the reaction mixture, recorded at

these are probably the endo and exo isomers in analogy to Ϫ60 °C, show fast formation of the branched substitution

Eur. J. Inorg. Chem. 2002, 2569Ϫ2586

2577

´

B. Bartels, C. Garcıa-Yebra, F. Rominger, G. Helmchen

FULL PAPER

gel, grade 60 (0.04Ϫ0.063 mm). ¹H, ¹³C and ³¹P NMR spectra were

recorded on Bruker DRX 200, DRX 300 or DRX 500 instruments.

¹H NMR chemical shifts are relative to residual non-deuterated

solvent in CDCl₃(δ ϭ 7.26 ppm) or CD₂Cl₂(δ ϭ 5.32 ppm). ¹³C

NMR shifts are given relative to the solvents CDCl₃(δ ϭ 77.0

ppm) and CD₂Cl₂(δ ϭ 53.8 ppm) and ³¹P NMR shifts are relative

to 85% H₃PO₄(δ ϭ 0.00 ppm). MS: JEOL, JMS-700. FAB: JEOL,

JMS-700; matrix: 4-nitrobenzyl alcohol (NBA) or 4-nitrophenyl oc-

tyl ether (NPOE). Optical rotation: PerkinϪElmer P 241. GC:

Hewlett Packard HP 5890 with Chiraldex γ-CD TA column (30

product 3a and the complex {IrCl(COD)[(R)-L2]} (29) in a

1:1 ratio. After workup compound 29 was isolated in 32%

yield and 3a in 64% yield. A GC and HPLC investigation

revealed that 3a had been formed with 99% regioselectivity

and 0% ee. The lack of enantioselectivity shows that nucleo-

philic attack at the allyl ligand is faster than coordination of

the chiral ligand by displacement of a COD double bond.

Complexes 30 and 31 were found to possess catalytic ac-

tivity. For example, allylic alkylations of substrates 1 and

(E)-2 with NaCH(CO₂Me)₂gave very similar results with m ϫ 0.25 mm). GC/MS: Hewlett Packard HP 5890/7972 instru-

ment with a HP 5 column (crosslinked methylsilicon, 25 m ϫ

0.2 mm, helium). HPLC: Hewlett Packard HP 1090 with DAICEL

Chiralcel ODH column (25 cm ϫ 0.46 cm) in combination with

DAICEL Chiralcel ODH precolumn (5 cm ϫ 0.46 cm). Elemental

analyses: Microanalytical laboratory of the Organisch-Chemisches

Institut, Universität Heidelberg. The compounds (aR)-L2,^[37]L5^[38]

and [IrCl(COD)]₂,^[39]were prepared according to published proced-

ures. The allylic esters were prepared by reaction of the correspond-

ing alcohols with acetic anhydride [1a,^[40]1b,^[41]1c,^[42]1d,^[43](E)-

2a,^[40](E)-2b,^[44](E)-2c,^[42]], methyl chloroformate (25b)^[45]or di-

ethyl chlorophosphate (26b). Compounds 3a,^[40]3c,^[4d,42]6b,^[12a]

7bϪ8b,^[46]8d,^[47]7a,^[48]7cϪ7f,^[48]20Ϫ21,^[21a]and 23^[4d]have been

described and characterised previously. The enantiomerically pure

allylic acetates (S)-1b,^[12a](R)-1b,^[12a,49]and 22^[50]were obtained by

enzyme-catalysed partial esterification of the corresponding ra-

cemic alcohols. Compounds 14 and 15,^[10]and 18^[13]were prepared

according to previously published methods. 1-(S)-Phenylpropen-2-

ol [(S)-6a], (R)-buten-2-ol [(R)-6c], tetra(n-propyl)ammonium per-

ruthenate (TPAP), [P(NMe₂)₃] and L4 are commercially available.

catalyst 30/(R)-L2 (4 mol %) and catalyst [IrCl(COD)]₂/(R)-

L2 (see Table 8). In agreement with the results obtained in

the stoichiometric reaction described above, small amounts

of racemic 3a (HPLC) were formed in allylic alkylations of

substrates 1b and (E)-2b upon catalysis with 30/(R)-L2 as a

consequence of attack of the nucleophile at the allyl ligand

present in complex 30. In this reaction, complex 29 is also

formed, which acts as the catalyst of the process.

Table 8. Allylic alkylations of substrates 1 and (E)-2 catalysed by

mixtures of 30 and L2 or [IrCl(COD)L2] (29)^[a]

Entry Substrate Catalyst^[b]

Time Yield [%] Ratio ee [%] of 3

[h] 3 and 4 3:4 (config.)^[c]

1

2

3

1a

30/L2 [1:1]

[IrCl(COD)L2]

30/L2 [1:1]

18

3

18

3

85

98

38

99

80

92

12

54

98:2

92:8

95:5

98:2

99:1

98:2

9 (S)

8 (S)

34 (R)

37 (R)

67 (R)

69 (R)

(E)-2a

4

(E)-2a [IrCl(COD)L2]

5^[d]

6

1b

(E)-2b

30/L2 [1:1]

[IrCl(COD)L2]

30/L2 [1:1]

General Procedure I. Acetylation of the Alcohols: A solution of the

alcohol (10 mmol), DMAP (12 mg, 0.10 mmol) and NEt₃(4.2 mL,

30 mmol) in CH₂Cl₂(10 mL) was cooled to 0 °C. Freshly distilled

Ac₂O (1.9 mL, 20 mmol) was added dropwise and the mixture was

stirred for 18 h at room temperature. Water (20 mL) was then ad-

ded and the aqueous layer extracted with CH₂Cl₂(3 ϫ 5 mL). The

combined organic layers were washed with satd. NH₄Cl solution

(20 mL), dried (Na₂SO₄) and concentrated in vacuo. Products were

purified by flash chromatography.

7^[d]

8

18

3

50:50 40 (R)

95:5 43 (R)

(E)-2b [IrCl(COD)L2]

All the reactions were performed at 25 °C in THF with 0.5 mmol

of allyl substrate [0.125 ], 1 mmol of NaCH(CO₂Me)₂and 4

[b]

mol % of [Ir]. When catalyst [IrCl(COD)L2] was used, this was

formed in situ by reaction of [IrCl(COD)]₂(2 mol %) with (R)-L2

(4 mol %). ^[c]Determined by HPLC. ^[d]Small quantities of 3a were

also formed.

1-Methoxymethoxymethylallyl Acetate (1e): Crude 1e was prepared

according to general procedure I from alcohol 6e (740 mg,

5.6 mmol) and was purified by flash column chromatography (silica

gel, petroleum ether/ethyl acetate 8:1). Yield: 58% (570 mg). ¹H

NMR (300.13 MHz, CDCl₃, 25 °C): δ ϭ 2.10 (s, 3 H, COCH₃),

3.36 (s, 3 H, CH₃OCH₂), 3.63Ϫ3.67 [m, 2 H, CH₂CH(OAc)], 4.63

Conclusions

Our results demonstrate that it is possible to achieve

modest to high levels of enantioselectivity in Ir^I-catalysed

asymmetric alkylations of monoalkylallyl acetates. Pres-

ently it is necessary to use racemic, branched allylic derivat-

2

4

3

(s, 2 H, OCH₂O), 5.26 [ddd, J_H,Hϭ 1.1, J_H,Hϭ 1.1, J_H,H

ϭ

2

4

10.7 Hz, 1 H, syn-(ϭCH₂)], 5.34 [ddd, J_H,Hϭ 1.1, J_H,Hϭ 1.1,

³J_H,H

17.3 Hz, 1 H, anti-(ϭCH₂)], 5.44Ϫ5.48 [m, 1 H,

CH(OAc)], 5.83 (ddd, ³J_H,Hϭ 6.3, J_H,Hϭ 10.7, ³J_H,Hϭ 17.3 Hz,

ϭ

3

ives as substrates. Further progress will probably be 1 H, ϭCH) ppm. ¹³C{¹H} NMR (75.47 MHz, CDCl₃, 25 °C): δ ϭ

20.9 (COCH₃), 55.1 (CH₃OCH₂), 68.4 [CH₂CH(OAc)], 73.0

[CH(OAc)], 96.2 (OCH₂O), 117.9 (ϭCH₂), 133.0 (ϭCH), 170.0

(CϭO) ppm. MS (CI, isobutane): m/z (%) ϭ 175 (10) [M^ϩϩ 1],

143 (90) [M^ϩϪ OCH₃], 113 (100) [M^ϩϪ OMOM]. HRMS (EI)

for C₇H₁₁O₃[M^ϩϪ OCH₃]: calcd. 143.0708; found 143.0685.

achieved on the basis of more detailed mechanistic investi-

gations. In particular, further investigations of the structure

and dynamic properties of (allyl)Ir^IIIcomplexes are re-

quired.

1-(2-Methoxymethoxyethyl)allyl Acetate (1f): Crude 1f was pre-

pared according to general procedure I from alcohol 6f (1.5 g,

10.3 mmol) and purified by flash chromatography (silica gel, petro-

leum ether/ethyl acetate 6:1). Yield: 86% (1.7 g). ¹H NMR

Experimental Section

General: All reactions were carried out using dry solvents under an

argon atmosphere. TLC: Macherey & Nagel Polygram Sil G/UV

precoated sheets, treatment with I₂or aqueous KMnO₄solution

for visualization of spots. Column chromatography: Fluka silica

(300.13 MHz, CDCl₃, 25 °C):

δ

ϭ 1.85Ϫ1.99 [m, 2 H,

CH₂CH(OAc)], 2.07 (s, 3 H, CH₃CO), 3.35 (s, 3 H, CH₃OCH₂),

3

3.56 (t, J_H,H

ϭ

6.4 Hz, 2 H, MOMOCH₂), 4.60 (s, 2 H,

Eur. J. Inorg. Chem. 2002, 2569Ϫ2586

2578

Iridium-Catalysed Allylic Substitution

FULL PAPER

2

3

CH₃OCH₂), 5.19 [ddd, J_H,Hϭ 1.1, ⁴J_H,Hϭ 1.1, J_H,Hϭ 10.5 Hz,

[M^ϩϪ CH₂OCH₃Ϫ CO₂], 45 (100) [CH₂OCH₃^ϩ]. HRMS (EI) -

2

4

3

1 H, synϪ(ϭCH₂)], 5.27 [ddd, J_H,Hϭ 1.1, J_H,Hϭ 1.1, J_H,H

ϭ

for C₆H₉O₃[M^ϩϪ CH₂OCH₃]: calcd. 129.0552; found 129.0557.

17.2 Hz, 1 H, antiϪ(ϭCH₂)], 5.80 (ddd, ³J_H,Hϭ 6.5, ³J_H,Hϭ 10.5,

³J_H,Hϭ 17.2 Hz, 1 H, ϭCH) ppm. ¹³C{¹H} NMR (75.47 MHz,

CDCl₃, 25 °C): δ ϭ 21.0 (COCH₃), 34.1 [CH₂CH(OAc)], 55.0

(CH₃OCH₂), 63.4 (MOMOCH₂), 71.8 [CH(OAc)], 96.3 (CH₃OCH-

₂OCH₂), 116.7 (ϭCH₂), 136.0 (ϭCH), 170.0 (CϭO) ppm. C₉H₁₆O₄

(188.22): calcd. C 57.43, H 8.57; found C 57.23, H 8.58.

(Z)-5-Methoxymethoxypent-2-enyl Acetate [(Z)-2f]: Crude (Z)-2f

was prepared from alcohol 19 (5.0 g, 34.2 mmol) according to gen-

eral procedure I and purified by flash column chromatography (sil-

ica gel, petroleum ether/ethyl acetate 9:1). Colourless oil. Yield:

1

77% (4.9 g). H NMR (300.13 MHz, CDCl₃, 25 °C): δ ϭ 2.03 (s,

3

3 H, CH₃CO), 2.38 (dt, J_H,Hϭ 6.6, J_H,Hϭ 6.6 Hz, 2 H, MOM-

3

5-Phenylpent-2-enyl Acetate [(E)-2b]: Crude 2b was prepared ac-

cording to general procedure I from alcohol 8b (6.0 g, 37.0 mmol)

and purified by flash chromatography (silica gel, petroleum ether/

OCH₂CH₂), 3.32 (s, 3 H, OCH₃), 3.53 (t, J_H,Hϭ 6.6 Hz, 2 H,

3

MOMOCH₂), 4.58 (s, 2 H, CH₃OCH₂), 4.60 (d, J_H,Hϭ 5.6 Hz,

2 H, CH₂OAc), 5.52Ϫ5.76 (m, 2 H, ϭCH) ppm. ¹³C{¹H} NMR

(75.47 MHz, CDCl₃, 25 °C): δ ϭ 20.9 (CH₃CO), 28.1 (MOM-

ethyl acetate 97:3). Yield: 83% (6.3 g). ¹H NMR (300.13 MHz,

3

CDCl₃, 25 °C): δ ϭ 2.08 (s, 3 H, COCH₃), 2.41 (dt, J_H,Hϭ 6.7, OCH₂CH₂), 55.1 (OCH₃), 60.2 (CH₂OAc), 66.8 (MOMOCH₂),

³J_H,Hϭ 8.1 Hz, 2 H, PhCH₂CH₂), 2.74 (br. t, ³J_H,Hϭ 7.4 Hz, 2 H,

96.3 (CH₃OCH₂), 125.4, 131.1 (ϭCH), 170.8 (CϭO) ppm.

4

3

PhCH₂), 4.54 (dd, J_H,Hϭ 0.9, J_H,Hϭ 6.4 Hz, 2 H, CH₂OAc), C₉H₁₆O₄(188.22): calcd. C 57.43, H 8.57; found C 57.27, H 8.72.

5.58Ϫ5.89 (m, 2 H, ϭCH), 7.19Ϫ7.34 (m, 5 H, Ph) ppm. ¹³C{¹H}

General Procedure II. Ir^I-Catalysed Allylic Alkylation: A solution

NMR (75.47 MHz, CDCl₃, 25 °C): δ ϭ 20.8 (COCH₃), 33.9 (CH₂),

of [IrCl(COD)]₂(6.7 mg, 0.01 mmol) in THF (2 mL) was treated

with substrate (0.5 mmol) and then ligand (0.02 mmol). The re-

sulting solution was stirred for 5 min. If the reaction was run with

additive, this was added now (0.5 mmol) and the solution stirred

35.2 (CH₂), 64.1 (CH₂OAc), 124.4 (ϭCHCH₂OAc), 126.1 (Ph),

128.2 (Ph), 128.3 (Ph), 135.2 (CHϭCHCH₂OAc), 141.4 (Ph), 170.6

(CϭO) ppm. C₁₃H₁₆O₂(204.27): calcd. C 76.44, H 7.90; found

C 76.26, H 7.85.

for a further 5 min. After that, a freshly prepared solution of di-

4-Methylpent-2-enyl Acetate [(E)-2d]: Crude 2d was prepared ac-

cording to general procedure I from alcohol 8d (4.6 g, 46.5 mmol)

and purified by flash chromatography (silica gel, petroleum ether/

methyl 2-sodiomalonate, prepared by suspending sodium hydride

(24 mg, 1.0 mmol) in THF (2 mL) and dropwise addition of di-

methyl malonate (115 µL, 1.0 mmol), was added. The mixture was

stirred under the stated reaction conditions, then water (4 mL) was

diethyl ether 95:5). Yield: 89% (5.9 g). ¹H NMR (300.13 MHz,

3

CDCl₃, 25 °C): δ ϭ 1.00 (d, J_H,Hϭ 7.0 Hz, 6 H, CH₃), 2.06 (s, added and the mixture extracted with diethyl ether (3 ϫ 5 mL). The

3 H, COCH₃), 2.31 [br. sext., ³J_H,Hϭ 7.0 Hz, 1 H, CH(CH₃)₂], 4.50 combined organic layers were washed with a satd. NH₄Cl solution

(d, ³J_H,Hϭ 6.6 Hz, 2 H, CH₂OAc), 5.50 (ddt, ⁴J_H,Hϭ 1.1, ³J_H,Hϭ

(10 mL), dried (Na₂SO₄) and concentrated in vacuo. The crude

3

4

6.6, J_H,Hϭ 15.5 Hz, 1 H, ϭCHCH₂OAc), 5.74 (ddt, J_H,Hϭ 1.1, product was purified by flash column chromatography to give sub-

³J_H,Hϭ 6.2, J_H,Hϭ 15.5 Hz, 1 H, CHϭCHCH₂OAc) ppm.

¹³C{¹H} NMR (75.47 MHz, CDCl₃, 25 °C): δ ϭ 21.0 (COCH₃),

22.0 (two CH₃), 30.7 [CH(CH₃)₂], 65.4 (CH₂OAc), 120.9

stitution products 3 and 4 as colourless oils.

3

2-(1-Phenethylallyl) Dimethylmalonate (3b): Compound 3b was ob-

tained by allylic alkylation of substrate 1b (535 mg, 2.6 mmol) ac-

cording to general procedure II [ligand: P(OPh)₃, reaction time:

3 h, room temperature]. After purification (silica gel, petroleum

ether/ethyl acetate 97:3), a 95:5 mixture of 3b and 4b was obtained.

Yield: 91% (650 mg). [α]²_D⁰ϭ 11.4 (c ϭ 0.63, CHCl₃) for 3b with

93% ee (R). HPLC: DAICEL Chiralcel ODH column, length:

25 cm ϩ 5 cm precolumn, flow: 0.5 mL·min^Ϫ1, eluent: n-hexane/

iPrOH (99.5:0.5); 3b: t_R(R) ϭ 32.2 min, t_R(S) ϭ 34.3 min; (E)-4b:

(CHCH₂OAc), 143.2

(CHϭCHCH₂OAc), 170.9 (CϭO)

ppm. C₈H₁₄O₂(142.20): calcd. C 67.57, H 9.92; found C 67.69,

H 10.07.

(E)-5-Methoxymethoxypent-2-enyl Acetate [(E)-2f]: Crude (E)-2f

was prepared according to general procedure I from alcohol 8f

(1.2 g, 8.2 mmol) and purified by flash column chromatography

(silica gel, petroleum ether/diethyl ether 5:1). Colourless oil. Yield:

97% (1.5 g). ¹H NMR (300.13 MHz, CDCl₃, 25 °C): δ ϭ 2.06 (s,

t

_Rϭ 49.6 min. ¹H NMR (300.13 MHz, CDCl₃, 25 °C): δ ϭ

1.57Ϫ1.88 (m, 2 H, PhCH₂CH₂), 2.46Ϫ2.77 (m, 2 H, PhCH₂), 2.84

3 H, COCH₃), 2.37 (dt, ³J_H,Hϭ 6.6, ³J_H,Hϭ 12.1 Hz, 2 H, MOM-

3

OCH₂CH₂), 3.35 (s, 3 H, CH₃O), 3.58 (t, J_H,Hϭ 6.6 Hz, 2 H, (dddd, ⁴J_H,Hϭ 3.7, ³J_H,Hϭ 9.0, ³J_H,Hϭ 9.5, ³J_H,Hϭ 9.3 Hz, 1 H,

4

3

MOMOCH₂), 4.51 (dd, J_H,Hϭ 1.1, J_H,Hϭ 6.2 Hz, 2 H,

CH₂OAc), 4.62 (s, 2 H, CH₃OCH₂), 5.61Ϫ5.72 (m, 1 H, CH),

CHCHϭCH₂), 3.44 [d, J_H,Hϭ 9.0 Hz, 1 H, CH(CO₂CH₃)₂], 3.69

(s, 3 H, OCH₃), 3.71 (s, 3 H, OCH₃), 5.11Ϫ5.20 (m, 2 H, ϭCH₂),

5.74Ϫ5.85 (m, 1 H, CH) ppm. ¹³C{¹H} NMR (75.47 MHz, CDCl₃, 5.73 (ddd, J_H,Hϭ 9.3, J_H,Hϭ 10.7, J_H,Hϭ 16.6 Hz, 1 H, CHϭ

3

25 °C): δ ϭ 20.8 (COCH₃), 32.5 (MOMOCH₂CH₂), 55.0 (CH₃O), CH₂), 7.13Ϫ7.31 (m, 5 H, Ph) ppm. ¹³C{¹H} NMR (75.47 MHz,

64.8 (MOMOCH₂or CH₂OAc), 66.6 (MOMOCH₂or CH₂OAc),

CDCl₃, 25 °C): δ ϭ 33.2 (PhCH₂CH₂), 33.9 (PhCH₂), 43.7

96.2 (CH₃OCH₂), 125.7 (CH), 132.2 (CH), 170.6 (CϭO) ppm. (CHCHϭCH₂), 52.1 (CH₃), 52.2 (CH₃), 56.7 [CH(CO₂CH₃)₂],

C₉H₁₆O₄(188.22): calcd. C 57.43, H 8.57; found C 57.21, H 8.56.

117.9 (ϭCH₂), 125.8 (Ph), 128.2 (Ph), 128.3 (Ph), 137.6 (CHϭ

CH₂), 141.6 (Ph), 168.3 (CϭO), 168.5 (CϭO) ppm. C₁₆H₂₀O₄

(276.33): calcd. C 69.55, H 7.30; found C 69.34, H 7.33.

(Z)-4-Methoxymethoxybut-2-enyl Acetate [(Z)-2e]: Crude (Z)-2e

was prepared from alcohol 15 (6.8 g, 51.8 mmol) according to gen-

eral procedure I and purified by flash column chromatography (sil-

2-(1-Isopropylallyl) Dimethylmalonate (3d): Ester 3d was prepared

ica gel, petroleum ether/ethyl acetate 6:1). Colourless oil. Yield: by allylic alkylation of substrate 1d (850 mg, 6.0 mmol) according

69% (6.2 g). ¹H NMR (300.13 MHz, CDCl₃, 25 °C): δ ϭ 2.03 (s, to general procedure II [ligand: P(OPh)₃, reaction time: 5 days,

3

3 H, CH₃CO), 3.30 (s, 3 H, OCH₃), 4.14 (d, J_H,Hϭ 5.7 Hz, 2 H, room temperature]. After purification by flash chromatography (sil-

3

MOMOCH₂), 4.60 (s, 2 H, CH₃OCH₂), 4.63 (d, J_H,Hϭ 6.0 Hz,

ica gel, petroleum ether/ethyl acetate 98:2), a 93:7 mixture of 3d

2 H, CH₂OAc), 5.62Ϫ5.80 (m, 2 H, ϭCH) ppm. ¹³C{¹H} NMR and 4d was obtained. Yield: 69% (890 mg). [α]²_D⁰ϭ 0.07 (c ϭ 0.54,

(75.47 MHz, CDCl₃, 25 °C): δ ϭ 20.7 (CH₃CO), 55.2 (OCH₃), 60.1

CHCl₃) for 3d with 66% ee (R). GC: Chiraldex γ-CD TA column,

(CH₂OAc), 62.6 (MOMOCH₂), 95.6 (CH₃OCH₂), 126.7, 130.2 (ϭ length: 30 m, 100 kPa helium, 50Ǟ100 °C at 1 °C·min^Ϫ1, then

CH), 170.7 (CϭO) ppm. MS (EI): m/z (%) ϭ 174 (1) [M^ϩ], 129 (8) 20 min at 100 °C; 3d: t_R(R) ϭ 44.8 min, t_R(S) ϭ 46.3 min; (E)-4d:

[M^ϩϪ CH₂OCH₃], 99 (6) [M^ϩϪ CH₂OCH₃Ϫ CH₂O], 85 (32)

Eur. J. Inorg. Chem. 2002, 2569Ϫ2586

t

_Rϭ 56.7 min. ¹H NMR (300.13 MHz, CDCl₃, 25 °C): δ ϭ 0.82

2579

´

B. Bartels, C. Garcıa-Yebra, F. Rominger, G. Helmchen

FULL PAPER

(d, ³J_H,Hϭ 7.0 Hz, 3 H, CH₃), 0.89 (d, ³J_H,Hϭ 7.0 Hz, 3 H, CH₃),

1.71 [dqq, J_H,Hϭ 5.2, J_H,Hϭ 7.0, J_H,Hϭ 7.0 Hz, 1 H, (80 mg). GC/MS (Z)-4e: t_Rϭ 12.79 min; (E)-4e: t_Rϭ 12.85 min.

9:1) an 8:92 mixture of (E)-4e and (Z)-4e was obtained. Yield: 65%

3

CH(CH₃)₂], 2.64 (ddd, J_H,Hϭ 5.2, J_H,Hϭ 9.6, J_H,Hϭ 9.9 Hz,

(Z)-4: ¹H NMR (300.13 MHz, CDCl₃, 25 °C): δ ϭ 2.65 [dd,

1 H, CHCHϭCH₂), 3.56 [d, J_H,Hϭ 9.6 Hz, 1 H, CH(CO₂CH₃)₂], ³J_H,Hϭ 7.4, ³J_H,Hϭ 7.4 Hz, 2 H, CH₂CH(CO₂Me)₂], 3.34 (s, 3 H,

3.67 (s, 3 H, OCH₃), 3.71 (s, 3 H, OCH₃), 5.05 [dd, J_H,Hϭ 1.8, CH₃OCH₂), 3.40 [t, J_H,Hϭ 7.4 Hz, 1 H, CH(CO₂Me)₂], 3.70 (s,

³J_H,Hϭ 16.9 Hz, 1 H, anti-(ϭCH₂)], 5.10 [dd, ²J_H,Hϭ 1.8, ³J_H,Hϭ 6 H, CO₂CH₃), 4.11 (d, ³J_H,Hϭ 6.2 Hz, 2 H, MOMOCH₂), 4.59 (s,

10.1 Hz, 1 H, syn-(ϭCH₂)], 5.67 (ddd, J_H,Hϭ 9.9, J_H,Hϭ 10.1,

³J_H,H16.9 Hz, 1 H, CHϭCH₂) ppm. ¹³C{¹H} NMR NMR (50.32 MHz, CDCl₃, 25 °C): δ ϭ 26.4 [CH₂CH(CO₂Me)₂],

(75.47 MHz, CDCl₃, 25 °C): δ ϭ 17.5 (CH₃), 21.2 (CH₃), 29.0 51.3 [CH(CO₂Me)₂], 52.5 (CO₂CH₃), 55.1 (CH₃OCH₂), 62.6 (MO-

3

2

3

2 H, CH₃OCH₂), 5.40Ϫ5.72 (m, 2 H, CHϭCH) ppm. ¹³C{¹H}

ϭ

[CH(CH₃)₂], 50.3 (CHCHϭCH₂), 52.2 (OCH₃), 52.4 (OCH₃), 54.9 MOCH₂), 95.7 (CH₃OCH₂), 128.2 (ϭCH), 129.0 (ϭCH), 169.1 (2s,

[CH(CO₂CH₃)₂], 118.6 (ϭCH₂), 134.8 (CHϭCH₂), 168.7 (CϭO),

169.0 (CϭO) ppm. C₁₁H₁₈O₄(214.26): calcd. C 61.66, H 8.47;

found C 61.47, H 8.43.

CϭO) ppm. HRMS (EI) for C₁₀H₁₅O₅[M^ϩϪ OCH₃]: calcd.

215.0919; found 215.0892.

(Z)-[2-(5-Methoxymethoxypent-2-enyl)] Dimethylmalonate [(Z)-4f].

Ester (Z)-4f was obtained by allylic alkylation of substrate (Z)-

2f (752 mg, 4.0 mmol) according to the general method II [ligand:

P(OPh)₃, reaction time: 4 h, room temperature]. After purification

(silica gel, petroleum ether/ethyl acetate 9:1), a 6:1 mixture of (Z)-

4f and 3f was obtained. Yield: 80% (830 mg). GC/MS (Z)-4f: t_Rϭ

13.33 min. ¹H NMR (200.13 MHz, CDCl₃, 25 °C): δ ϭ 2.31 (dt,

2-[1-Methoxymethoxymethylallyl] Dimethylmalonate (3e): Ester 3e

was obtained by allylic alkylation of substrate 1e (43 mg,

0.25 mmol) according to general procedure II [ligand: P(OPh)₃, re-

action time: 18 h, room temperature]. After purification by flash

chromatography (silica gel, petroleum ether/ethyl acetate 9:1), a

77:23 mixture of 3e and 4e was obtained. Yield: 91% (56 mg).

[α]_D²⁴ϭ 23.3 (c ϭ 0.69, CHCl₃) for 3e with 78% ee (R). GC: Chiraldex

γ-CD TA column, length: 30 m, 100 kPa helium, 50Ǟ100 °C at 10

°C·min^Ϫ1, then 60 min at 100 °C; 3e: t_R(R) ϭ 54.1 min, t_R(S) ϭ

55.8 min; (E)-4e: t_Rϭ 63.5 min. ¹H NMR (300.13 MHz, CDCl₃,

25 °C): δ ϭ 3.06Ϫ3.23 [m, 1 H, CH(CHϭCH₂)] 3.32 (s, 3 H,

3

³J_H,Hϭ 6.6, J_H,Hϭ 6.6 Hz, 2 H, MOMOCH₂CH₂), 2.61 [dd,

³J_H,Hϭ 6.6, ³J_H,Hϭ 6.6 Hz, 2 H, CH₂CH(CO₂Me)₂], 3.29 (s, 3 H,

3

CH₃OCH₂), 3.32Ϫ3.39 [m, 1 H, CH(CO₂Me)₂], 3.48 (t, J_H,H

ϭ

6.6 Hz, 2 H, MOMOCH₂), 3.68 (s, 6 H, CH₃), 4.56 (s, 2 H,

CH₃OCH₂), 5.27Ϫ5.53 (m, 2 H, CHϭCH) ppm. ¹³C{¹H} NMR

(50.32 MHz, CDCl₃, 25 °C): δ ϭ 26.7 [CH₂CH(CO₂Me)₂], 27.7

(MOMOCH₂CH₂), 51.5 [CH(CO₂Me)₂], 52.3 (CH₃), 56.4

(CH₃OCH₂), 66.9 (MOMOCH₂), 96.2 (CH₃OCH₂), 126.5 (ϭCH),

129.1 (ϭCH), 169.2 (CϭO) ppm. C₁₂H₂₀O₆(260.29): calcd. C

55.37, H 7.74; found C 55.22, H 7.85.

3

CH₃OCH₂), 3.59 (d, J_H,Hϭ 6.3 Hz, 1 H, MOMOCH₂), 3.61 (d,

3

³J_H,Hϭ 5.5 Hz, 1 H, MOMOCH₂), 3.68 [d, J_H,Hϭ 8.1 Hz, 1 H,

CH(CO₂Me)₂], 3.70 (s, 3 H, Me), 3.73 (s, 3 H, Me), 4.57 (s, 2 H,

3

CH₃OCH₂), 5.11Ϫ5.21 (m, 2 H, ϭCH₂), 5.84 (ddd, J_H,Hϭ 8.8,

3

³J_H,Hϭ 10.3, J_H,Hϭ 17.3 Hz, 1 H, ϭCH) ppm. ¹³C{¹H} NMR

(75.47 MHz, CDCl₃, 25 °C): δ ϭ 43.8 [CH(CHϭCH₂)], 52.3 (Me),

52.5 (Me), 53.1 [CH(CO₂Me)₂], 55.3 (CH₃OCH₂), 68.4 (MOM-

OCH₂), 96.5 (CH₃OCH₂), 118.2 (ϭCH₂), 135.3 (ϭCH), 166.9 (Cϭ

O), 168.7 (CϭO) ppm. C₁₁H₁₈O₆(246.26): calcd. C 53.65, H 7.37;

found C 53.37, H 7.23.

1-Methoxymethoxybut-3-en-2-ol (6e): This compound was prepared

via intermediates (see Schemes 2 and 3) that were not fully charac-

terised.

11e: A solution of 2-methoxy-1,3-dioxolane (10e) (30.0 g, 0.288

mol) in 250 mL of dichloromethane was cooled to Ϫ70 °C. Then,

a pre-cooled (Ϫ70 °C) solution of DIBAL-H in hexane (346 mL,

0.346 mol) was added dropwise and the reaction mixture was

stirred for 90 min at Ϫ70 °C. Thereafter, the temperature was al-

lowed to rise slowly to 0 °C and water (20 mL) was added. The

resulting precipitate was dissolved by addition of aqueous NaOH

(20%, 300 mL) and the resulting solution was extracted with

dichloromethane (4 ϫ 50 mL). The combined organic layers were

washed with satd. NaCl solution (2 ϫ 80 mL), dried (Na₂SO₄) and

concentrated in vacuo to give an oil which was distilled [65 °C

(10 Torr)] to give 11e. Yield: 51% (15.2 g).^{[51] 1}H NMR

(300.13 MHz, CDCl₃, 25 °C): δ ϭ 2.44 (s, 1 H, OH), 3.38 (s, 3 H,

CH₃OCH₂), 3.67 (m, 2 H, CH₂OH or MOMOCH₂), 3.73 (m, 2 H,

CH₂OH or MOMOCH₂), 4.66 (s, 2 H, CH₃OCH₂) ppm. ¹³C{¹H}

NMR (75.47 MHz, CDCl₃, 25 °C): δ ϭ 55.1 (CH₃OCH₂), 61.9,

70.1 (CH₂OH and MOMOCH₂), 96.3 (CH₃OCH₂) ppm.

2-[1-(2-Methoxymethoxyethyl)allyl] Dimethylmalonate (3f): Ester 3f

was obtained by allylic alkylation of substrate 1f (94 mg, 0.5 mmol)

according to general procedure II [ligand: P(OPh)₃, reaction time:

18 h, room temperature]. After purification by flash chromato-

graphy (silica gel, petroleum ether/ethyl acetate 9:1), a 93:7 mixture

of 3f and 4f was obtained. Yield: 94% (122 mg). [α]²_D⁴ϭ Ϫ1.4 (c ϭ

0.76, CHCl₃) for 3f with 70% ee (R). HPLC: DAICEL Chiralcel

ODH column, length: 25 cm

0.5 mL·min^Ϫ1

eluent: n-hexane/iPrOH (95:5); 3f: t_R(R) ϭ

23.2 min, t_R(S) ϭ 24.6 min; (E)-4f:

_Rϭ 28.9 min. ¹H NMR

ϩ

5 cm precolumn, flow:

,

t

(300.13 MHz, CDCl₃, 25 °C): δ ϭ 1.54Ϫ1.66 (m, 1 H, MOM-

OCH₂CH₂), 1.76Ϫ1.87 (m, 1 H, MOMOCH₂CH₂), 2.94Ϫ2.97 [m,

1 H, CH(CHϭCH₂)], 3.45 [d, ³J_H,Hϭ 8.4 Hz, 1 H, CH(CO₂Me)₂],

3.39Ϫ3.58 (m, 2 H, MOMOCH₂), 3.70 (s, 3 H, Me), 3.75 (s, 3 H,

3

Me), 4.56Ϫ4.61 (m, 2 H, ϭCH₂), 5.69 (ddd, J_H,Hϭ 9.6, J_H,H

ϭ

3

9.6, J_H,Hϭ 16.9 Hz, 1 H, ϭCH), 3.34 (s, 3 H, CH₃OCH₂) ppm.

¹³C{¹H} NMR (75.47 MHz, CDCl₃, 25 °C): δ ϭ 32.1 (MOM-

OCH₂CH₂), 41.1 [CH(CHϭCH₂)], 52.4 (Me), 52.6 (Me), 55.2

(CH₃OCH₂), 56.6 [CH(CO₂Me)₂], 65.1 (MOMOCH₂), 96.4

(CH₃OCH₂), 118.0 (ϭCH₂), 137.3 (ϭCH), 168.4 (CϭO), 168.6

(CϭO) ppm. C₁₂H₂₀O₆(260.29): calcd. C 55.37, H 7.74; found C

55.28, H 7.63.

6e:

A

mixture of 11e (4.00 g, 37.7 mmol), NMO (6.63 g,

˚

56.6 mmol), activated molecular sieves (4 A) (18.86 g) and dichlor-

omethane (75 mL) was cooled to 0 °C, then TPAP (0.132 g,

0.38 mmol) was added slowly and the mixture stirred for 30 min.

After stirring for a further 2 h at room temperature the mixture

was filtered through a short silica column. The filtrate containing

the aldehyde 5e,^[52]was concentrated in vacuo and the residue was

used immediately for the next step.

(Z)-[2-(4-Methoxymethoxybut-2-enyl)] Dimethylmalonate [(Z)-4e]:

Ester (Z)-4e was obtained by allylic alkylation of substrate (Z)-

2e (87 mg, 0.5 mmol) according to general procedure II [ligand:

P(OPh)₃, reaction time: 4 h, room temperature]. After purification

by flash chromatography (silica gel, petroleum ether/ethyl acetate

A solution of aldehyde 5e (8 mL, 37.7 mmol) in THF (50 mL) was

added dropwise to a pre-cooled (0 °C) solution of vinylmagnesium

bromide in THF (1.0 , 41.5 mL, 41.5 mmol). The resulting mix-

ture was stirred for 1 h at 0 °C and 18 h at room temperature. Then,

2580

Eur. J. Inorg. Chem. 2002, 2569Ϫ2586

Iridium-Catalysed Allylic Substitution

FULL PAPER

water (ca. 50 mL) was added dropwise and the resulting precipitate (E)-5-Methoxymethoxypent-2-en-1-ol (8f): A solution of ester 7f

was dissolved by addition of 2  HCl. The solution was extracted (2.7 g, 14.4 mmol) in diethyl ether (50 mL) was cooled to Ϫ70 °C

with diethyl ether (4 ϫ 20 mL) and the combined organic layers and DIBAL-H (31.6 mL of 1.0  solution in hexane, 31.6 mmol)

were washed with brine (30 mL), dried (Na₂SO₄) and concentrated

in vacuo to give a yellow oil which was subjected to flash column

was slowly added via a dropping funnel. After stirring for 3.5 h at

Ϫ70 °C, water (10 mL) was added dropwise and the solution was

chromatography (silica gel, petroleum ether/ethyl acetate 1:1). allowed to warm to room temperature. Additional water (10 mL)

1

Yield: 10% (500 mg over 2 steps). H NMR (300.13 MHz, CDCl₃,

was added and the resulting precipitate was dissolved with 2  HCl

25 °C): δ ϭ 1.85 (s, 1 H, OH), 3.39 (s, 3 H, CH₃O), 3.47 [dd, solution (ca. 90 mL). The mixture was extracted with diethyl ether

2

³J_H,Hϭ 7.4, J_H,Hϭ 10.6 Hz, 1 H, CH₂CH(OH)], 3.67 (dd, (4 ϫ 20 mL), the combined organic layers washed (2 ϫ 30 mL satd.

2

³J_H,Hϭ 3.2, J_H,Hϭ 10.6 Hz, 1 H, CH₂CH(OH)], 4.29Ϫ4.34 [m, NH₄Cl solution), dried (Na₂SO₄) and concentrated in vacuo to give

3

1 H, CH(OH)], 4.67 (s, 2 H, CH₃OCH₂), 5.22 [d, J_H,Hϭ 10.6 Hz,

a colourless oil which was subjected to flash column chromato-

3

1 H, synϪ(ϭCH₂)], 5.29 (d, J_H,Hϭ 17.3 Hz, 1 H, antiϪ(ϭCH₂)], graphy (silica gel, diethyl ether/petroleum ether) to give 8f with

3

5.86 (ddd, J_H,Hϭ 5.5, J_H,Hϭ 10.6, J_H,Hϭ 17.3 Hz, 1 H,

Ͼ95% purity (GC/MS). Yield: 66% (1.4 g, 9.5 mmol). ¹H NMR

ϭCH) ppm. ¹³C{¹H} NMR (75.47 MHz, CDCl₃, 25 °C): δ ϭ 55.3 (300.13 MHz, CDCl₃, 25 °C): δ ϭ 1.56 (s, 1 H, OH), 2.31Ϫ2.42 (m,

3

(CH₃O), 71.4 [CH(OH)], 72.5 [CH₂CH(OH)], 96.8 [CH₃OCH₂)],

116.3 (ϭCH), 136.4 (ϭCH₂) ppm. MS (CI, isobutane): m/z (%) ϭ

133 (15) [M^ϩϩ 1].

2 H, MOMOCH₂CH₂), 3.36 (s, 3 H, CH₃O), 3.59 (t, J_H,H

ϭ

6.6 Hz, 2 H, MOMOCH₂), 4.02Ϫ4.13 (m, 2 H, CH₂OH), 4.63 (s,

2 H, CH₃OCH₂), 5.71Ϫ5.77 (m, 2 H, ϭCH) ppm. ¹³C{¹H} NMR

(75.47 MHz, CDCl₃, 25 °C): δ ϭ 32.6 (MOMOCH₂CH₂), 55.2

(CH₃O), 63.7, 67.1 (CH₂OH and MOMOCH₂), 96.4 (CH₃OCH₂),

129.3, 131.1 (both ϭCH) ppm. MS (EI): m/z (%) ϭ 145 (1) [M^ϩ

Ϫ H], 129 (1) [M^ϩϪ OH], 115 (10) [M^ϩϪ OCH₃], 45 (100)

[H₃COCH₂^ϩ]. HRMS (EI) for C₈H₁₄O₂[M^ϩϪ (OϭCH₂)]: calcd.

116.0837; found 116.0837.

5-Methoxymethoxypent-1-en-3-ol (6f): This compound was pre-

pared via intermediates (see Schemes 2 and 3) that were not fully

characterised.

10f: Trimethyl orthoformate (151 mL, 1.38 mol) was dropwise ad-

ded to a stirred solution of diol 9f (50.0 mL, 0.69 mol) and cam-

phorsulfonic acid monohydrate (1.73 g, 6.91 mmol) in dichlorome-

thane (300 mL), and the mixture was stirred for 20 h at room tem-

perature. The solvent was then removed under reduced pressure

and the residue distilled to give 10f (b.p. 148Ϫ152 °C) as a colour-

less oil.^[53]Yield: 45% (37.02 g). ¹H NMR (300.13 MHz, CDCl₃,

25 °C): δ ϭ 1.71Ϫ1.73 (m, 2 H, CH₂CH₂O), 3.36 (s, 3 H, CH₃O),

3.74Ϫ3.77 (m, 2 H, CH₂CH₂O), 4.13Ϫ4.15 (m, 2 H, CH₂CH₂O),

5.18 [s, 1 H, CH(OCH₃)] ppm. ¹³C{¹H} NMR (75.48 MHz, CDCl₃,

25 °C): δ ϭ 24.7 (CH₂CH₂O), 52.5 (CH₃), 61.7 (CH₂CH₂O), 110.2

(OCHO) ppm

(R)-5-Phenyl-1-penten-3-yl (1S,4R)-4,7,7-Trimethyl-3-oxo-2-oxab-

icyclo[2.2.1]heptane-1-carboxylate (12): (R)-5-phenyl-1-penten-3-ol

[(R)-6b)] (0.50 g, 3.1 mmol) (prepared by enzyme-catalysed kinetic

resolution of rac-6b)^[12a]followed by 864 µL (6.2 mmol) of triethyl-

amine were added dropwise to a cooled (0 °C) solution of (1S)-

camphanic acid chloride (671 mg, 3.1 mmol) in anhydrous dichloro-

methane (10 mL). The solution was stirred at room temperature

for 18 h and left to stand for 7 days. Colourless crystals precipitated

from the mixture, were filtered off and washed with diethyl ether.

The combined filtrates were concentrated in vacuo to give crude

12. Crystals suitable for X-ray diffraction were obtained by recrys-

tallization from ethyl acetate. Yield: 84% (890 mg). M. p. 55Ϫ58 °C.

11f: Treatment of 10f (15 g, 0.127 mol) with DIBAL-H (153 mL,

1  in hexane, 0.153 mol) in 200 mL dichloromethane, according to

the procedure described above for 11e, gave crude monoester 11f,

which was purified by distillation (85Ϫ87 °C, 10 Torr) as a colour-

less oil.^[54]Yield: 40% (6.16 g). H NMR (200.13 MHz, CDCl₃, 25

1

3

°C): δ ϭ 1.83 (tt, J_H,Hϭ 6.0, J_H,Hϭ 6.0 Hz, 2 H, CH₂CH₂OH),

3

2.41 (s, 1 H, OH), 3.34 (s, 3 H, CH₃O), 3.70 (t, J_H,Hϭ 6.0 Hz,

2 H, MOMOCH₂or CH₂OH), 3.74 (t, ³J_H,Hϭ 5.9 Hz, 2 H, MOM-

OCH₂or CH₂OH), 4.60 (s, 2 H, OCH₂O) ppm. ¹³C{¹H} NMR

(50.32 MHz, CDCl₃, 25 °C): δ ϭ 32.1 (CH₂CH₂OH), 55.2 (CH₃O),

61.0, 66.1 (CH₂OH and MOMOCH₂), 96.4 (OCH₂O) ppm

6f: Monoester 11f (4.9 g, 41.37 mmol) was transformed into alde-

hyde 5f and this reacted in situ with vinylmagnesium bromide (1 

in THF, 45.8 mmol), by following an analogous procedure to the

one described above for the preparation of 6e.^[55]After purification

¹H NMR (300.13 MHz, CDCl₃, 25 °C): δ ϭ 0.98 (s, 3 H, 7Ј-CH₃),

3

by flash column chromatography (silica gel, petroleum ether/ethyl 1.07 (s, 3 H, 7Ј-CH₃), 1.12 (s, 3 H, 5Ј-H), 1.71 (ddd, J_3Јa,2Јaϭ 4.4,

3

acetate 2:1) 6f was obtained as an oil. Yield: 29% over 2 steps

²J_3Јa,3Јbϭ 9.6, J_3Јa,2Јbϭ 13.6 Hz, 1 H, 3Ј-Ha), 1.87Ϫ2.15 (m, 4 H,

(17.5 g). ¹H NMR (300.13 MHz, CDCl₃, 25 °C): δ ϭ 1.74Ϫ1.91

4-H, 3Ј-Hb, 2Ј-Ha), 2.44 (ddd, ³J_2Јb,3Јbϭ 4.4, ²J_2Јb,2Јaϭ 10.7,

[m, 2 H, CH₂CH(OH)], 2.32 (s, 1 H, OH), 3.37 (s, 3 H, CH₃O), ³J_2Јb,3Јaϭ 13.6 Hz, 1 H, 2Ј-Hb), 2.67 (ddd, ³J_5,4aϭ 3.3, ³J_5,3ϭ 6.6,

3.60Ϫ3.80 (m, 2 H, MOMOCH₂), 4.31Ϫ4.36 [m, 1 H, CH(OH)], ³J_5,4bϭ 9.6 Hz, 2 H, 5-H), 5.25 (ddd, J_1a,1bϭ 1.1, J_1a,3ϭ 1.1,

2

4

2

4

2

4

4.62 (s, 2 H, CH₃OCH₂), 5.13 [ddd, J_H,Hϭ 1.4, J_H,Hϭ 1.4,

³J_1a,2ϭ 10.5 Hz, 1 H, 1-Ha), 5.32 (ddd, J_1b,1aϭ 1.1, J_1b,3ϭ 1.1,

³J_H,Hϭ 10.5 Hz, 1 H, syn-(ϭCH₂)], 5.28 [ddd, J_H,Hϭ 1.4, ³J_1b,2ϭ 17.1 Hz, 1 H, 1-Hb), 5.36Ϫ5.45 (m, 1 H, 3-H), 5.83 (ddd,

2

⁴J_H,Hϭ 1.4, J_H,Hϭ 17.2 Hz, 1 H, anti-(ϭCH₂)], 5.89 (ddd, ³J_2,3ϭ 6.6, J_2,1aϭ 10.5, J_2,1bϭ 17.1 Hz, 1 H, 2-H), 7.13Ϫ7.32

3

³J_H,Hϭ 5.6, J_H,Hϭ 10.5, J_H,Hϭ 17.2 Hz, 1 H, ϭCH) ppm.

¹³C{¹H} NMR (CDCl₃, 75.47 MHz, 25 °C):

ϭ 36.3 °C): δ ϭ 9.66 (C-5Ј), 16.80 (7Ј-CH₃), 28.95 (C-3Ј), 30.73 (C-4Ј),

[CH₂CH(OH)], 55.4 (CH₃O), 65.5 (MOMOCH₂), 71.6 [CH(OH)], 31.31 (C-5), 35.75 (C-4), 54.19, 54.82 (C-4Ј and C-7Ј), 75.82 (C-3),

96.5 (CH₃OCH₂), 114.6 (ϭCH₂), 140.5 (ϭCH) ppm. MS (CI, 91.05 (C-1Ј), 118.16 (C-2), 126.11 (C-9), 128.32, 128.49 (C-7 and

(m, 5 H, arom. H) ppm. ¹³C{¹H} NMR (75.47 MHz, CDCl₃, 25

3

δ

isobutane): m/z (%) ϭ 147 (100) [M^ϩϩ 1], 129 (40) [M^ϩϩ 1 Ϫ C-8), 135.51 (C-1), 140.87 (C-6), 166.81 (C-8Ј), 178.21 (C-6Ј) ppm.

H₂O].

MS (FAB): m/z (%) ϭ 343 (42) [M^ϩϩ 1], 199 (46) [{M^ϩϩ 1} Ϫ

Eur. J. Inorg. Chem. 2002, 2569Ϫ2586

2581

´

B. Bartels, C. Garcıa-Yebra, F. Rominger, G. Helmchen

FULL PAPER

(PhCH₂CH₂)

Ϫ

(CH-CHϭCH₂)], 181 (36) [{M^ϩ

ϩ

1}

Ϫ

25 cm ϩ 5 cm precolumn, flow: 0.5 mL min^Ϫ1, eluent: n-hexane/

(PhCH₂CH₂) Ϫ (CHO)-CHϭCH₂)], 153 (60) [{M^ϩϩ 1} Ϫ iPrOH (95:5); 27b: t_R(S) ϭ 23.8 min, t_R(R) ϭ 27.9 min; (E)-28b:

(PhCH₂CH₂) Ϫ CH(OCO) Ϫ (CHCH₂)]. C₂₁H₂₆O₄(342.44): calcd.

C 73.66, H 7.65; found C 73.49, H 7.64.

t_Rϭ 38.2 min.

27b: ¹H NMR (300.13 MHz, CDCl₃, 25 °C): δ ϭ 1.53Ϫ1.75 (m,

2 H, PhCH₂CH₂), 2.44 (s, 3 H, CH₃), 2.43Ϫ2.53 (m, 2 H, PhCH₂),

(Z)-5-Methoxymethoxypent-2-en-1-ol (19): A mixture of Lindlar

catalyst (1.2 g), quinoline (0.33 mL, 357 mg, 2.77 mmol), alcohol

18 (5.9 g, 40.1 mmol) and methanol (340 mL) was stirred under a

H₂atmosphere for 15 hours at room temperature. The catalyst was

removed by filtration through celite and the brown filtrate was con-

centrated in vacuo. The residue was purified by flash column chro-

matography (silica gel, petroleum ether/ethyl acetate 2:1) to afford

a colourless oil. Yield: 69% (4.1 g). ¹H NMR (300.13 MHz, CDCl₃,

2

3

4.10 (d, J_H,Hϭ 15.4 Hz, 1 H, PhCH₂N), 4.37 (ddd, J_H,Hϭ 6.6,

³J_H,Hϭ 6.6, ³J_H,Hϭ 7.7 Hz, 1 H, CHN), 4.60 (d, ²J_H,Hϭ 15.4 Hz,

4

2

3

1 H, PhCH₂N), 4.99 [ddd, J_H,Hϭ 1.1, J_H,Hϭ 1.5, J_H,H

ϭ

4

2

17.3 Hz, 1 H, anti-(ϭCH₂)], 5.10 [ddd, J_H,Hϭ 1.1, J_H,Hϭ 1.5,

³J_H,Hϭ 10.7 Hz, 1 H, syn-(ϭCH₂)], 5.47 (ddd, J_H,Hϭ 6.6,

3

³J_H,Hϭ 10.7, J_H,Hϭ 17.3 Hz, 1 H, CHϭ CH₂), 6.78Ϫ6.84 (m,

3

2 H, arom. H), 7.08Ϫ7.21 (m, 3 H, arom. H), 7.24Ϫ7.35 (m, 5 H,

arom. H), 7.38Ϫ7.45 (m, 2 H, arom. H), 7.70 [d, J_H,Hϭ 8.7 Hz,

3

4

3

25 °C): δ ϭ 2.32 (ddt, J_H,Hϭ 1.5, J_H,Hϭ 6.2, J_H,Hϭ 6.2 Hz,

2 H, MOMOCH₂CH₂), 2.68 (br. s, 1 H, OH), 3.29 (s, 3 H, OCH₃),

3.50 (t, ³J_H,Hϭ 6.2 Hz, 2 H, MOMOCH₂), 4.09 (d, ³J_H,Hϭ 6.6 Hz,

2 H, CH₂OH), 4.55 (s, 2 H, CH₃OCH₂), 5.46Ϫ5.57 (m, 1 H, ϭCH),

5.62Ϫ5.77 (m, 1 H, ϭCH) ppm. ¹³C{¹H} NMR (75.47 MHz,

CDCl₃, 25 °C): δ ϭ 27.8 (MOMOCH₂CH₂), 55.1 (OCH₃), 57.8

(CH₂OH), 66.5 (MOMOCH₂), 96.2 (CH₃OCH₂), 128.8, 130.9 (ϭ

CH) ppm.

2 H, m-(1-SO_2-4_-Me-C₆H₄)] ppm. ¹³C{¹H} NMR (75.48 MHz,

CDCl₃, 25 °C): δ ϭ 21.5 (CH₃), 32.6 (PhCH₂), 34.1 (PhCH₂CH₂),

48.2 (PhCH₂N), 60.3 [CH(NTs)(Bzl)], 118.6 (ϭCH₂), 125.7, 127.3,

127.5, 128.2, 128.3, 128.5, 128.6, 129.6 (all CH arom.), 135.8 (CHϭ

CH₂), 138.1, 138.1, 141.6, 143.1 (s, all C arom.) ppm. MS (FAB):

m/z (%) ϭ 406 (60) [M^ϩϩ H], 300 (100) [M^ϩϪ PhCH₂CH₂], 262

(99) [M^ϩϪ Ts]. C₂₅H₂₇NO₂S (405.55): calcd. C 74.04, H 6.71, N

3.45, S 7.91; found C 74.03, H 6.74, N 3.51, S 7.76.

5-Phenyl-1-penten-3-yl Diethylphosphonate (26b): A solution of al-

cohol 6b (2.0 g, 12.3 mmol) in THF (40 mL) was cooled to Ϫ78 °C

and nBuLi (13 mL, 1.6  in hexane, 20.9 mmol) was added slowly

with a syringe. After stirring for 30 min at Ϫ78 °C, diethyl chloro-

phosphate (5.35 mL, 6.36 g, 36.9 mmol) was added. The mixture

was stirred for 18 h, while the temperature was allowed to slowly

rise from Ϫ78 °C to 25 °C. Then, water (40 mL) was added and

the mixture was extracted with diethyl ether (3 ϫ 20 mL). The com-

bined organic layers were washed with satd. NaCl solution

(50 mL), dried (Na₂SO₄) and concentrated in vacuo. The crude

product was purified by flash column chromatography (silica gel,

petroleum ether/ethyl acetate 2:1) to give 26b as a yellow oil. Yield:

64% (2.3 g). ¹H NMR (300.13 MHz, CDCl₃, 25 °C): δ ϭ 1.32 (t,

(5,7-Dioxa-6-phosphadibenzo[a,c]cyclohepten-6-yl)dimethylamine

(L3): A mixture of 2,2Ј-dihydroxybiphenyl (1.00 g, 5.4 mmol),

NH₄Cl (8.0 mg, 0.1 mmol) and [P(NMe₂)₃] (976 µL, 5.4 mmol) and

toluene (15 mL) was stirred for 18 h at 80 °C and then concentrated

in vacuo to give a yellow oil which was purified by flash column

chromatography (silica gel, petroleum ether/ethyl acetate 98:2).

Colourless oil. Yield: 29% (400 mg). ¹H NMR (500.13 MHz,

3

CDCl₃, 25 °C): δ ϭ 2.66 (d, J_H,Pϭ 9.4 Hz, 6 H, CH₃), 7.18 (d,

³J_H,Hϭ 8.0 Hz, 2 H, arom. H), 7.22Ϫ7.26 (m, 2 H, arom. H),

3

7.32Ϫ7.37 (m, 2 H, arom. H), 7.46 (br. d, J_H,Hϭ 8.0 Hz, 2 H,

arom. H) ppm. ¹³C{¹H} NMR (125.76 MHz, CDCl₃, 25 °C): δ ϭ

2

35.9 (d, J_P,Cϭ 10.7 Hz, CH₃), 121.9 (arom. C), 124.5 (arom. C),

2

129.2 (arom. C), 129.6 (arom. C), 131.1 (d, J_P,Cϭ 2.8 Hz, arom.

3

³J_H,Hϭ 7.0 Hz, 3 H, CH₃), 1.33 (t, J_H,Hϭ 7.0 Hz, 3 H, CH₃),

C), 151.5 (arom. C) ppm. ³¹P NMR (202.46 MHz, CDCl₃, 25 °C):

δ ϭ 149.74 (s) ppm. C₁₄H₁₄NO₂P (259.24): calcd. C 64.86, H 5.44,

N 5.40, P 11.95; found C 64.95, H 5.37, N 5.45, P 11.91.

1.87Ϫ2.14 (m, 2 H, CH₂CH), 2.65Ϫ2.74 (m, 2 H, PhCH₂), 4.10 (br.

3

q, J_H,Hϭ 6.6 Hz, 4 H, OCH₂), 4.74Ϫ4.85 [m, 1 H, CHO-

3

P(O)(OEt)₂], 5.25 (d, J_H,Hϭ 10.3 Hz, 1 H, ϭCH₂), 5.34 (d,

3

³J_H,Hϭ 17.3 Hz, 1 H, ϭCH₂), 5.88 (ddd, J_H,Hϭ 6.6, J_H,H

ϭ

[IrCl(η⁴-1,5-COD){(aR)-L2}] (29): THF (4 mL) was added to a

mixture of [IrCl(COD)]₂(20 mg, 0.03 mmol) and phosphorus

amidite (R)-L2 (21.3 mg, 0.06 mmol) at room temperature. Forma-

tion of complex 29 was observed by ¹H and ³¹P NMR spectroscopy

within 10 min. Recrystallization from diethyl ether furnished or-

3

10.3, J_H,Hϭ 17.3 Hz, 1 H, ϭCH), 7.15Ϫ7.35 (m, 5 H, Ph) ppm.

¹³C{¹H} NMR (75.47 MHz, CDCl₃, 25 °C): δ ϭ 16.1 (2 CH₃), 31.0

(CH₂CHOR), 37.5 (PhCH₂), 63.7 (2 OCH₂), 79.2 (CHOR), 117.5

(ϭCH₂), 126.0 (Ph), 128.4 (Ph), 128.4 (Ph), 136.7 (ϭCH), 141.3

(Ph) ppm. ³¹P NMR (121.49 MHz, CDCl₃, 25 °C): δ ϭ Ϫ1.40 ppm.

HRMS (FAB) for C₁₄H₂₄O₄P [M^ϩϩ 1]: calcd. 299.1412; found

299.1440.

1

ange crystals of 29. Yield: 90% (37.5 mg). H NMR (500.13 MHz,

CDCl₃, 25 °C): δ ϭ 1.22Ϫ1.44 (m, 2 H, CH₂-COD), 1.74Ϫ1.84 (m,

2 H, CH₂-COD), 1.81Ϫ1.91 (m, 2 H, CH₂-COD), 1.93Ϫ2.04 (m,

N-Benzyl-4-methyl-N-(1-phenethylallyl) Benzenesulfonamide (27b): 1 H, CH₂-COD) 2.09Ϫ2.19 (m, 1 H, CH₂-COD), 2.20Ϫ2.30 (m, 1

3

An LHMDS solution (4.0 mL, 1.0  in THF, 4.0 mmol) was added

to a solution of N-tosylbenzylamine (1.1 g, 4.2 mmol) in THF

H, CH-COD), 2.45Ϫ2.52 (m, 1 H, CH-COD), 2.79 (d, J_P,H

ϭ

10.7 Hz, 6 H, CH₃), 3.35Ϫ3.42 (m, 1 H, CH-COD), 5.30Ϫ5.45 (m,

3

(12 mL). The mixture was stirred for 30 min at room temperature 1 H, CH-COD), 7.27 (dd, J_H,Hϭ 8.4, J_H,Hϭ 8.7 Hz, 2 H, CH

3

and then added to a substrate/catalyst solution prepared as follows:

arom.), 7.30 (d, J_H,Hϭ 8.7 Hz, 1 H, CH arom.), 7.32 (d, J_H,Hϭ

3

[IrCl(COD)]₂(26.8 mg, 0.04 mmol) was dissolved in THF (4 mL) 8.4 Hz, 1 H, CH arom.) 7.42 (d, J_H,Hϭ 8.4 Hz, 1 H, CH arom.),

3

and substrate 1b (2.0 mmol) and then triphenylphosphite (21.7 µL,

7.39Ϫ7.47 (m, 2 H, CH arom.), 7.89 (d, J_H,Hϭ 7.4 Hz, 1 H, CH

3

0.08 mmol) were added. The reaction mixture was stirred for 18 h arom.), 7.91 (d, J_H,Hϭ 8.7 Hz, 1 H, CH arom.) 7.95 (d, J_H,H

ϭ

at room temperature and then, treated with water (20 mL) and ex-

tracted with diethyl ether (3 ϫ 20 mL). The combined organic

layers were washed with satd. NaHCO₃(1 ϫ 50 mL) and satd.

8.35 Hz, 2 H, CH arom.), 8.01 (d, ³J_H,Hϭ 9.0 Hz, 1 H, CH arom.)

ppm. ¹³C{¹H} NMR (125.77 MHz, CDCl₃, 25 °C): δ ϭ 29.7 (d,

⁴J_P,Cϭ 1.9 Hz, CH₂-COD), 29.0 (d, J_P,Cϭ 2.8 Hz, CH₂-COD),

4

NaCl solutions (1 ϫ 50 mL) and dried over Na₂SO₄. After removal 32.5 (d, J_P,Cϭ 2.8 Hz, CH₂-COD), 34.2 (d, J_P,Cϭ 3.8 Hz, CH₂-

3

of the solvent in vacuo a yellow oil was obtained which was sub-

COD), 38.2 (d, J_P,Cϭ 10.36 Hz, CH₃), 52.2 (d, J_P,Cϭ 1.9 Hz,

3

jected to flash column chromatography (silica gel, petroleum ether/ CH-COD), 55.3 (d, J_P,Cϭ 1.88 Hz, CH-COD), 102.7 (d, J_P,C

ϭ

3

ethyl acetate 97:3) to give a 79:21 mixture of 27b and 28b. Yield:

17.9 Hz, CH-COD), 103.22 (d, J_P,Cϭ 19.1 Hz, CH-COD), 121.1

47% (382 mg), colourless oil. [α]²_D⁴ϭ 7.55 (c ϭ 0.58, CHCl₃) for 27b (CH arom.), 122.1 (d, J_C,Pϭ 1.88 Hz, C-arom.), 122.9 (d, J_C,P

ϭ

4

with 11% ee (S). HPLC: DAICEL Chiralcel ODH column, length:

2.86 Hz, C-arom.), 123.9 (d, J_C,Pϭ 2.83 Hz, CH-arom.), 125.2,

2582

Eur. J. Inorg. Chem. 2002, 2569Ϫ2586

Iridium-Catalysed Allylic Substitution

FULL PAPER

126.1, 126.5, 126.8, 127.1, 128.3, 128.4, 129.9, 130.3 (s, all CH

COD), 33.6 (CH₂-COD), 37.9 (CH₂-COD), 42.6 (allyl CH₂), 80.8

(CH-COD), 81.7(CH-COD) 89.4 (CH-COD), 92.6 (CH- COD),

102.2 (allyl CH), 123.0 (allyl CHPh), 128.5, 128.9, 130.8 (C₆H₅),

147.6 (C_ipso-C₆H₅) ppm. C₁₇H₂₁Cl₂Ir (488.52): calcd. C 41.80, H

3

arom.), 131.0, 131.6 (both s, C-arom.), 132.2 (d, J_C,Pϭ 1.88 Hz,

2

C-arom.), 132.5 (s, C-arom.), 148.4 (d, J_C,Pϭ 4.7 Hz, OC-arom.),

2

149.5 (d, J_C,P

ϭ

12.9 Hz, OC-arom.) ppm. ³¹P NMR

(202.46 MHz, CDCl₃, 25 °C): δ ϭ 116.3 (s) ppm. C₃₀H₃₀ClIrNO₂P 4.34; found C 41.84, H 4.37.

(694.4): calcd. C 51.79, H 4.35, N 2.01, P 4.46; found C 52.01, H

4.47, N 2.06, P 4.40.

[Ir(Br)₂(η³-C₃H₅)(η⁴-1,5-COD)] (31). Method A: A solution of

[Ir(η³-CH₂CHCHPh)(Cl)₂(η⁴-1,5-COD)] (30): Cinnamyl chloride

(83 µL, 91 mg, 0.596 mmol) was added to solution of

[IrCl(COD)]₂(50 mg, 0.07 mmol) and allyl bromide (28.6 µL,

36.0 mg, 0.3 mmol) in THF (1 mL) was stirred for 18 h at 50 °C.

A yellow precipitate of 31 formed which was washed with diethyl

ether and recrystallised from dichloromethane/pentane. Yield:

81% (60 mg).

a

[IrCl(COD)]₂(100 mg, 0.149 mmol) in THF (2 mL) and the mix-

ture stirred for 18 h at 50 °C. A yellow precipitate of 30 was formed,

which was isolated by filtration, washed with hexane and dried in

vacuo. Yield: 82% (119.3 mg). A 2:1 mixture of two isomers was

observed in solution at between Ϫ60 and Ϫ40 °C.

Method B: [IrCl(COD)]₂(150 mg, 0.2 mmol) was treated with an

excess of anhydrous LiBr (155.1 mg, 1.8 mmol) in acetone (15 mL).

After stirring for 18 h, the solvent was removed under vacuum and

the product [IrBr(COD)]₂extracted three times with dichlorome-

thane. Removal of the solvent gave an orange oil which was further

treated with allyl bromide (57 µL, 79.8 mg, 0.7 mmol). After 18 h,

a crystalline yellow precipitate had formed, which was isolated by

filtration and washed with diethyl ether. Yield: 66% (145 mg).

NMR spectroscopy at Ϫ60 °C showed a 9:1 mixture of two iso-

mers.

Major isomer: ¹H NMR (300.13 MHz, CD₂Cl₂, Ϫ60 °C): δ ϭ

1.45Ϫ1.65 (m, 1 H, CH₂-COD), 1.65Ϫ1.83 (m, 1 H, CH₂-COD),

2.25Ϫ2.55 (m, 3 H, CH₂-COD), 2.60Ϫ3.00 (m, 2 H, CH₂-COD),

2.95Ϫ3.15 (m, 1 H, CH₂-COD), 3.72 3.87 (m, 1 H, CH-COD), 4.07

3

[d, J_H,Hϭ 8.8 Hz, 1 H, anti-(allyl CH₂)], 4.45 [d, J_H,Hϭ 6.8 Hz,

1 H, syn-(allyl CH₂)], 4.59Ϫ4.86 (m, 3 H, CH-COD), 4.93 [d,

³J_H,Hϭ 14.0 Hz, 1 H, anti-(allyl CHPh)], 6.06 (ddd, J_H,Hϭ 6.8,

3

³J_H,Hϭ 8.8, ³J_H,Hϭ 14.0 Hz, 1 H, allyl CH), 7.20Ϫ7.90 (m, 5 H, -

C₆H₅) ppm. ¹³C{¹H} NMR (75 MHz, CD₂Cl₂, Ϫ60 °C): δ ϭ 26.6 Major isomer: ¹H NMR (500.13 MHz, CD₂Cl₂, Ϫ60 °C): δ ϭ

(CH₂-COD), 28.6 (CH₂-COD), 35.5 (CH₂-COD), 37.6 (CH₂-

1.46Ϫ1.59 (m, 1 H, CH₂-COD), 1.75Ϫ1.88 (m, 1 H, CH₂-COD),

COD), 40.9 (allyl CH₂), 81.9 (CH-COD), 83.0 (CH-COD), 92.7 2.23Ϫ2.35 (m, 1 H, CH₂-COD), 2.40Ϫ2.50 (m, 1 H, CH₂-COD),

(CH-COD), 93.3 (CH-COD), 110.6 (allyl CH), 111.1 (allyl CHPh), 2.50Ϫ2.61 (m, 1 H, CH₂-COD), 2.60Ϫ2.72 (m, 1 H, CH₂-COD),

3

129.0, 130.3, (C₆H₅), 134.8 (C_ipso-C₆H₅) ppm.

3.00Ϫ3.10 (m, 1 H, CH₂-COD), 3.12 [d, J_H,Hϭ 13.3 Hz, 1 H,

anti-(allyl CH₂)], 3.22Ϫ3.34 (m, 1 H, CH₂-COD), 3.33Ϫ3.41 (m, 1

Minor isomer: ¹H NMR (300.13 MHz, CD₂Cl₂, Ϫ60 °C): δ ϭ

1.26Ϫ1.36 (m, 1 H, CH₂-COD), 1.45Ϫ1.65 (m, 1 H, CH₂-COD), H, CH-COD), 3.83 [d, ³J_H,Hϭ 10.0 Hz, 1 H, anti-(allyl CH₂)], 4.48

3

1.93Ϫ2.10 (m, 1 H, CH₂-COD), 2.25Ϫ2.55 (m, 2 H, CH₂-COD),

[d, J_H,Hϭ 6.7 Hz, 1 H, syn-(allyl CH₂)], 4.75Ϫ4.84 (m, 1 H, CH-

3

2.60Ϫ3.00 (m, 3 H, CH₂-COD), 4.13Ϫ4.26 (m, 1 H, CH-COD), COD), 4.93 [d, J_H,Hϭ 8.7 Hz, 1 H, syn-(allyl CH₂)], 4.95Ϫ5.03

4.31Ϫ4.55 (m, 2 H, CH-COD and allyl CH₂), 4.59Ϫ4.86 (m, 2 H, (m, 1 H, CH-COD), 5.10Ϫ5.17 (m, 1 H, CH-COD), 5.74Ϫ5.87 (m,

CH-COD and allyl CH₂), 5.22Ϫ5.32 (m,

1

H, CH-COD), 1H. allyl CH) ppm. ¹³C{¹H} NMR (125.76 MHz, CD₂Cl₂, Ϫ60

3

5.55Ϫ5.74 (m, 1 H, allyl CH), 6.33 [d, J_H,Hϭ 14.1 Hz, 1 H, anti-

°C): δ ϭ 26.4, 27.2, 36.4, 37.1 (all CH₂-COD), 44.6 (allyl CH₂),

(allyl CHPh)], 7.20Ϫ7.90 (m, 5 H, -C₆H₅) ppm. ¹³C{¹H} NMR 83.2, 85.5 (both CH-COD), 79.7 (allyl CH₂), 91.8, 92.7 (both CH-

(75 MHz, CD₂Cl₂, Ϫ60 °C): δ ϭ 27.0 (CH₂-COD), 28.8 (CH₂- COD), 114.5 (allyl CH) ppm.

Table 9. Crystal data and details of the structure refinement for compounds 12, 29, 30, and 31

Compound

12

29

30

31

Empirical formula

Molecular weight [g·mol^Ϫ1

Crystal system

Space group

C₂₁H₂₆O₄

342.42

Monoclinic

P2₁

C₃₀H₃₀ClIrNO₂P

695.17

Orthorhombic

P2₁2₁2₁

C₁₇H₂₁Cl₂Ir

488.44

Monoclinic

P2₁/n

C₁₁H₁₇Br₂Ir

501.27

Orthorhombic

Pna2₁

]

Z

a [A]

b [A]

c [A]

2

4

˚

7.6989(3)

12.4705(5)

10.0936(4)

99.888(1)

954.68(7)

0.08

1.07 and 0.80

.64 ϫ .13 ϫ .13

2.0 to 25.0

8026

11.9440(1)

14.2192(1)

15.2164(2)

Ϫ

2584.26(4)

5.36

0.80 and 0.46

.22 ϫ .12 ϫ .05

2.0 to 27.5

27013

6.8597(1)

20.9350(3)

11.1414(2)

97.724(1)

1585.48(4)

8.75

0.67 and 0.58

.14 ϫ .08 ϫ .06

2.0 to 25.6

11653

15.6809(3)

7.4221(1)

10.5656(2)

Ϫ

1229.68(4)

17.32

0.77 and 0.41

.50 ϫ .04 ϫ .02

2.6 to 27.5

11961

β [deg.]

Volume [A ]

3

˚

Absorption coefficient [mm^Ϫ1

Max. and min. transmission

Crystal size [mm³]

]

Theta range [deg.]

Reflections collected

Independent reflections

3350 (R_intϭ 0.052)

2568

330

5926 (R_intϭ 0.075)

5087

343

2761 (R_intϭ 0.034)

2301

209

2816 (R_intϭ 0.066)

2494

154

Observed reflections [I Ͼ 2σ(I)]

Refined parameters

Goodness-of-fit on F²

1.06

0.99

1.09

1.00

Final R1 index [I Ͼ 2σ(I)]

Final wR2 index [I Ͼ 2σ(I)]

Absolute structure parameter

0.043

0.086

Ϫ1.4(12)

0.15/Ϫ0.17

0.030

0.050

Ϫ0.018(6)

0.93/Ϫ0.72

0.023

0.048

Ϫ

0.027

0.058

0.015(15)

1.03/Ϫ1.23

Ϫ3

˚

Largest diff. peak/hole [e A

]

0.60/Ϫ0.74

Eur. J. Inorg. Chem. 2002, 2569Ϫ2586

2583

´

B. Bartels, C. Garcıa-Yebra, F. Rominger, G. Helmchen

FULL PAPER

Minor isomer: ¹H NMR (500.13 MHz, CD₂Cl₂, Ϫ60 °C): δ ϭ

1.09Ϫ1.20 (m, 1 H, CH₂-COD), 1.57Ϫ1.67 (m, 1 H, CH₂-COD),

2.11Ϫ2.20 (m, 1 H, CH₂-COD), 2.24Ϫ2.35 (m, 1 H, CH₂-COD),

2.40Ϫ2.50 (m, 1 H, CH₂-COD), 2.69Ϫ2.79 (m, 1 H, CH₂-COD),

2.93Ϫ3.00 (m, 1 H, CH₂-COD), 3.00Ϫ3.10 (m, 1 H, CH₂-COD),

4.23Ϫ4.30 (m, 1 H, CH-COD), 4.62Ϫ4.70 (m, 3 H, CH-COD and

allyl CH₂), 4.75Ϫ4.84 (m, 1 H, CH-COD), 4.90Ϫ4.97 (m, 1 H, allyl

CH), 5.20Ϫ5.30 (m, 2 H, allyl CH₂), 5.58Ϫ5.63 (m, 1 H, CH-COD)

ppm. ¹³C{¹H} NMR (125.76 MHz, CD₂Cl₂, Ϫ60 °C): δ ϭ 26.7,

27.2, 35.3, 37.7 (all CH₂-COD), 44.9 (allyl CH₂), 83.6, 85.4, 87.1

(all CH-COD), 89.6 (allyl CH₂), 93.9 (both CH-COD), 107.9 (allyl

CH) ppm. C₁₁H₁₇Br₂Ir (501.28): calcd. C 26.36, H 3.42; found C

26.30, H 3.56.

Acknowledgments

This work was supported by the Deutsche Forschungsgemeinschaft

and the Fonds der Chemischen Industrie. C. G.-Y. thanks the Alex-

ander von Humboldt Foundation for a Postdoctoral Fellowship

and the European Community for a Marie Curie Fellowship, pro-

gram ‘‘Improving Human Research Potential and the Socio-economic

Knowledge Base’’ under contract number HPMF-CT-2000-00905.

We thank T. D. Weiβ for NMR measurements, Prof. M. Reggelin,

Darmstadt, for advice concerning EXSY spectroscopy, T. Leβmann

and A. Spieβ for preparation of substrates, and Degussa AG for

iridium salts.

Stoichiometric Reaction of Complex 30 with Dimethyl 2-Sodioma-

lonate in the Presence of (R)-L2: A 0.24  solution of NaCH-

(CO₂Me)₂(0.4 mL, 0.1 mmol) in THF was syringed into a Schlenk

tube and the solvent was removed under vacuum. To the residual

white solid were added complex 30 (49.9 mg, 0.1 mmol) and the

stoichiometric quantity of ligand (R)-L2 (36.2 mg, 0.1 mmol), then

[D₈]THF/CD₂Cl₂(2:1; 1.5 mL) was added at Ϫ78 °C. The mixture

was stirred for a few minutes at this temperature to give a clear

light yellow solution that was transferred into a pre-cooled (Ϫ60

°C) NMR tube. The ¹H and ³¹P NMR spectra of this solution,

recorded at Ϫ60 °C, showed complete reaction to give exclusively

a 1:1 mixture of rac-3a and complex 29 which were separated by

flash column chromatography (silica gel, hexane/diethyl ether 9:1).

Yield: 64% (16 mg) of 3a (purity Ͼ 99% by GCMS) and 32%

(22.6 mg) of 29 as orange crystals.

[1] [1a]

B. M. Trost, D. L. Van Vranken, Chem. Rev. 1996, 96,

395Ϫ422. ^[1b]T. Hayashi, Catalytic Asymmetric Synthesis (Ed.:

I. Ojima), VCH, New York 1993, p. 325.

[1c]

B. M. Trost, C.

Lee, Catalytic Asymmetric Synthesis (Ed.: I. Ojima), 2nd ed.,

Wiley-VCH, Weinheim, Germany, 2000. ^[1d]A. Pfaltz, M. Laut-

ens, Comprehensive Asymmetric Catalysis IϪIII 1999, 2,

833Ϫ884.

[2] [2a]

S.-L. You, X.-Z. Zhu, Y.-M. Luo, X.-L. Hou, L.-X. Dai, J.

Am. Chem. Soc. 2001, 123, 7471Ϫ7472. ^[2b]R. Pretot, A. Pfaltz,

Angew. Chem. 1998, 110, 337Ϫ339; Angew. Chem. Int. Ed.

´ ˆ

[2c]

1998, 37, 323Ϫ325.

Soc. 1998, 120, 9074Ϫ9075.

Uozumi, Chem. Commun. 1997, 561Ϫ562.

B. M. Trost, F. D. Toste, J. Am. Chem.

[2d]

T. Hayashi, M. Kawatsura, Y.

[2e]

T. Hayashi, A.

Ohno, S.-J. Lu, Y. Matsumoto, E. Fukuyo, K. Yanagi, J. Am.

Chem. Soc. 1994, 116, 4221Ϫ4226. ^[2f]T. Hayashi, K. Kishi, A.

Yamamoto, V. Ito, Tetrahedron Lett. 1990, 31, 1743Ϫ1746.

Mo:

[3]

[3a]

B. M. Trost, K. Droga, I. Hachiya, T. Emura, D. L.

X-ray Crystallographic Studies: Crystals of compound 12 were ob-

tained as needles by slow evaporation of solvent from a saturated

solution in ethyl acetate. Single crystals of complex 29 were grown

as orange needles by slow concentration of a saturated solution in

diethyl ether. Single crystals of complexes 30 and 31 were grown,

both as yellow needles, by slow diffusion of diethyl ether into a

saturated solution in dichloromethane.

Hughes, S. Krska, R. A. Reamer, M. Palucki, N. Yasuda, P. J.

Reider, Angew. Chem. 2002, 114, 2009Ϫ2012; Angew. Chem.

[3b]

Int. Ed. 2002, 41, 1929Ϫ1932.

B. M. Trost, I. Hachiya, J.

[3c]

Am. Chem. Soc. 1998, 120, 1104Ϫ1105.

F. Glorius, A.

[3d]

Pfaltz, Org. Lett. 1999, 1, 141Ϫ144. W:

G. C. Lloyd-Jones,

A. Pfaltz, Angew. Chem. 1995, 107, 534Ϫ536; Angew. Chem.

Int. Ed. Engl. 1995, 34, 462Ϫ464. Isomerization is slow for W

complexes.

[4]

[4a]

[4b]

Fe:

Y. Xu, B. Zhou, J. Org. Chem. 1987, 52, 974Ϫ977.

The measurements were made on a Bruker SMART-CCD dif-

U. Eberhardt, G. Mattern, Chem. Ber. 1988, 121, 1531Ϫ1534.

fractometer with graphite monochromated Mo-K_αradiation

[4c]

D. Enders, B. Jandeleit, S. von Berg, G. Raabe, J. Runsink,

˚

(0.71073 A) at 200 K. Frames corresponding to a sphere of data

[4d]

Organometallics 2001, 20, 4312Ϫ4332. Rh:

P. A. Evans, J.

were collected. 20 s exposures of 0.3 degree scans in omega were

taken. An absorption correction was applied using SADABS (G.

M. Sheldrick, SADABS, Bruker AXS, Inc.: Madison, WI, 1996)

based on the laue symmetry of the reciprocal space. The structures

were solved by direct methods and expanded using Fourier tech-

niques. All non-hydrogen atoms were refined with anisotropic dis-

placement parameters, most hydrogen atoms could be located in

the Fourier map, in 12 all hydrogens were refined isotropically, in

29 and 30 only the allylic hydrogens and the olefinic hydrogens of

the COD were refined isotropically, in 30 the allylic hydrogens were

not included in the structure model due to the disorder, all the rest

of the hydrogen atoms was taken into account at calculated posi-

tions. The full-matrix least-squares refinement against F²con-

verged. All calculations were performed using the SHELXTL crys-

tallographic software package of Bruker (G. M. Sheldrick,

SHELXTL V5.10, Bruker AXS, Inc.: Madison, WI, 1997). For de-

tails and crystal data see Table 9.

[4e]

D. Nelson, J. Am. Chem. Soc. 1998, 120, 5581Ϫ5582. Ru:

Y. Morisaki, T. Kondo, T.-A. Mitsudo, Y. Watanabe, Organo-

[4f]

metallics 1999, 18, 4742Ϫ4746.

B. M. Trost, P. L. Fraisse,

Z. T. Ball, Angew. Chem. 2002, 114, 1101Ϫ1103; Angew. Chem.

Int. Ed. 2002, 41, 1059Ϫ1061.

[5] [5a]

R. Takeuchi, M. Kashio, Angew. Chem. 1997, 109,

[5b]

268Ϫ270; Angew. Chem. Int. Ed. Engl. 1997, 36, 263Ϫ265.

R. Takeuchi, M. Kashio, J. Am. Chem. Soc. 1998, 120,

8647Ϫ8655. ^[5c]R. Takeuchi, Polyhedron 2000, 19, 557Ϫ561. ^[5d]

R. Takeuchi, K. Tanabe, Angew. Chem. 2000, 112, 2051Ϫ2054;

Angew. Chem. Int. Ed. 2000, 39, 1975Ϫ1978.

[6]

J. P. Janssen, G. Helmchen, Tetrahedron Lett. 1997, 38,

8025Ϫ8026.

[7] [7a]

[7b]

B. Bartels, G. Helmchen, Chem. Commun. 1999, 741Ϫ742.

K. Fuji, N. Kinoshita, K. Tanaka, T. Kawabata, Chem.

Commun. 1999, 2289Ϫ2290.

[8]

For characterised σ- and π-allyl complexes see: ^[8a]K. D. John,

K. V. Salazar, B. L. Scott, R. T. Baker, A. P. Sattelberger, Chem.

Commun. 2000, 581Ϫ582.

Scott, R. T. Baker, A. P. Sattelberger, Organometallics 2001, 2,

296Ϫ304.

1990, 9, 2428Ϫ2430.

Park, Organometallics 2000, 19, 4043Ϫ4050.

A. Ibers, J. Am. Chem. Soc. 1979, 101, 4201Ϫ4212.

Muetterties, K. D. Tau, J. F. Kirner, T. V. Harris, J. Stark, M.

[8b]

K. D. John, K. V. Salazar, B. L.

CCDC-181254 (12), -181255 (29), -181256 (30) and -181257 (31)

contain the supplementary crystallographic data for this paper.

These data can be obtained free of charge via www.ccdc.cam.ac.uk/

conts/retrieving.html (or from the CCDC, 12 Union Road, Cam-

bridge CB2 1EZ, UK; Fax: (internat.) ϩ44-1223/336-033; E-mail:

deposit@ccdc.cam.ac.uk)

[8c]

J. B. Wakefield, J. M. Stryker, Organometallics

[8d]

C. S. Chin, D. Chong, S. Lee, Y. J.

[8e]

T. H. Tulip, J.

[8f]

E. L.

2584

Eur. J. Inorg. Chem. 2002, 2569Ϫ2586

Iridium-Catalysed Allylic Substitution

FULL PAPER

The oxidative addition of allyl halides to Ir^Icompounds is a

general method for the synthesis of (allyl)Ir^IIIcomplexes, see: J.

M. OЈConnor, in Science of Synthesis, vol. 1 (Ed.: M. Lautens),

Thieme, Stuttgart, 2001, p. 717.

[28]

R. Thompson, V. W. Day, Organometallics 1982, 1, 1562Ϫ1567.

[8g]

J. Müller, B. Passon, J. A. Pickardt, J. Organomet. Chem.

1982, 236, C11ϪC14. ^[8h]P. Powell, J. Organomet. Chem. 1983,

252, 205Ϫ208.^[8i]O. Boutry, E. Gutierrez, A. Monge, M. C.

´

[29]

[30]

´

Nicasio, P. J. Perez, E. Carmona, J. Am. Chem. Soc. 1992,

114, 7288Ϫ7290.

The terms endo and exo are defined in Scheme 10.

[30a]

See for example:

J. A. Kaduk, A. T. Poulos, J. A. Ibers, J.

[9]

Fast polymerization of aldehyde 5e prevented the preparation

of acetate (E)-2e.

M. Takasu, Y. Naruse, H. Yamamoto, Tetrahedron Lett. 1988,

29, 1947Ϫ1950.

S. V. Ley, J. Norma, W. Griffith, P. Marsden, Synthesis 1994,

639Ϫ666.

Organomet. Chem. 1977, 127, 245Ϫ260 and references therein.

[30b]

[30c]

J. S. Merola, Organometallics 1989, 8, 2975Ϫ2977.

M.

[10]

[11]

D. Fryzuk, L. Huang, N. T. McManus, P. Paglia, S. L. Rettig,

[30d]

G. S. White, Organometallics 1992, 11, 2979Ϫ2990.

M. D.

Fryzuk, X. Gao, S. J. Rettig, J. Am. Chem. Soc. 1995, 117,

3106Ϫ3117.

[31]

[12] [12a]

A very similar bonding situation as in 31 has been reported for

the complex [(η⁵-C₅H₅)Mo(η³-C₃H₅)(NO)(I)]: J. W. Faller, D.

F. Chodosh, D. Katahira, J. Organomet. Chem. 1980, 227Ϫ231.

M. Kollmar, B. Goldfuss, M. Reggelin, F. Rominger, G.

Helmchen, Chem. Eur. J. 2001, 7, 4913Ϫ4927.

M. Kollmar, H. Steinhagen, J. Janssen, B. Goldfuss, S. Mali-

novskaya, F. Rominger, G. Helmchen, Chem. Eur. J. in press.

H. Friebolin, Basic One- and Two-Dimensional NMR Spectro-

scopy, Wiley-VCH, Weinheim, 1998.

T. Itoh, E. Akasaki, K. Kudo, S. Shirakami, Chem. Lett.

2001, 262Ϫ263. ^[12b]A Sharpless kinetic resolution of the alco-

hol 6b has also been published: R. P. Discordia, D. C. Dittmer,

J. Org. Chem. 1990, 55, 1414Ϫ1415.

[32]

[33]

[34]

[13]

[14]

L. E. Overman, D. Lesuisse, M. Hashimoto, J. Am. Chem. Soc.

1983, 105, 5373Ϫ5379.

´

R. B. Bedford, S. Castillon, P. A. Chaloner, C. Claver, E. Fer-

´

nandez, P. B. Hitchcock, A. Ruiz, Organometallics 1996, 15,

3990Ϫ3997.

[35] [35a]

[15] [15a]

[15b]

J. Jelner, B. H. Meier, P. Bachmann, R. R. Ernst, J. Chem.

R. Takeuchi, N. Shiga, Org. Lett. 1999, 2, 265Ϫ267.

[35b]

Phys. 1979, 71, 4546Ϫ4553.

Chem. Rev. 1990, 90, 935Ϫ967.

C. L. Perrin, T. J. Dwyer,

R. Takeuchi, N. Ue, K. Tanabe, K. Yamashita, N. Shiga, J.

Am. Chem. Soc. 2001, 123, 9525Ϫ9534.

For a study of the influence of donor capacity of ligands on

[36] [36a]

[16]

K. E. Schwiebert, J. M. Stryker, Organometallics 1993, 12,

[36b]

[16a]

600Ϫ602.

C. Tejel, M. A. Ciriano, A. J. Edwards, F. J.

the stability of (allyl)Rh complexes see:

Vrieze, J. Organomet. Chem. 1967, 9, 527Ϫ536.

H. C. Volger, K.

[16b]

Lahoz, L. A. Oro, Organometallics 2000, 19, 4968Ϫ4976.

R. Hulst, N. K. de Vries, B. L. Feringa, Tetrahedron: Asym-

metry 1994, 5, 699Ϫ708.

M. Peer, J. C. de Jong, M. Kiefer, T. Langer, H. Rieck, H.

Schell, P. Sennhenn, J. Sprinz, H. Steinhagen, B. Wiese, G.

Helmchen, Tetrahedron 1996, 52, 7547Ϫ7583.

H. C.

[37]

[38]

Volger, K. Vrieze, J. Organomet. Chem. 1967, 9, 537Ϫ548.

G. Poli, G. Giambastiani, Chemtracts: Org. Chem. 2001, 14,

204Ϫ216.

J. Wolf, H. Werner, Organometallics 1987, 6, 1164Ϫ1169.

E. B. Tjaden, J. M. Stryker, Organometallics 1992, 11,

16Ϫ18.

[17]

[18] [18a]

[18b]

[39]

[40]

J. L. Herde, J. C. Lambert, C. V. Senoff, Inorg. Synth. 1974,

[19] [19a]

15, 18Ϫ20.

P. A. Evans, J. E. Robinson, J. D. Nelson, J. Am. Chem.

[19b]

J. L. Lehmann, G. C. Lloyd-Jones, Tetrahedron 1995, 51,

8863Ϫ8874; resolution of enantiomers of 3a by HPLC: Chira-

cel ODH column, length: 25 cm ϩ 5 cm precolumn, flow:

0.5 mL·min^Ϫ1, eluent: n-hexane/iPrOH (99.5:0.5); 3a: t_R(R) ϭ

28.9 min, t_R(S) ϭ 31.6 min; (E)-4a: t_Rϭ 45.1 min. [α]²_D⁴ϭ 27.2

(c ϭ 0.85, CHCl₃) for 3a with 86 % ee (R).

Soc. 1999, 121, 6761Ϫ6762.

Am. Chem. Soc. 2000, 122, 5012Ϫ5013.

P. A. Evans, D. K. Leahy, J.

[19c]

P. A. Evans, J. E.

Robinson, K. K Moffett, Org. Lett. 2001, 3, 3269Ϫ3271.

[20] [20a]

T. A. Mobley, C. Shade, R. G. Bergman, Organometallics

[20b]

1998, 17, 3574Ϫ3587.

Chem. Commun. 1998, 2321Ϫ2322.

Baumann, A. Togni, Tetrahedron: Asymmetry 1997, 8,

G. C. Lloyd-Jones, S. C. Stephen,

[20c]

U. Bruckhardt, M.

[41]

[42]

E. Keinan, E. Bosh, J. Org. Chem. 1986, 51, 4006Ϫ4016.

A. L. J. Beckwith, A. A. Zavitas, J. Am. Chem. Soc. 1986, 108,

8230Ϫ8234; analysis of enantiomers of 3c by GC: Chiraldex γ-

CD TA column, length: 30 m, 100 kPa helium, flow

80 mL·min^Ϫ1, 50Ǟ100 °C at 1 °C·min^Ϫ1, then 20 min at 100

[20d]

155Ϫ159.

Commun. 1998, 217Ϫ218.

nou, M. Lautens, Angew. Chem. Int. Ed. 2002, 41, 26Ϫ47.

M. Kawatsura, Y. Uozumi, T. Hayashi, Chem.

[20e]

For a recent review see K. Fag-

[21] [21a]

J.-C. Fiaud, J.-Y. Legros, J. Org. Chem. 1987, 52,

[21b]

´

´ ´

D. Dvorak, I. Stary, P. Kocovsky, J. Am.

°C; 3c: t_R(R) ϭ 35.7 min, t_R(S) ϭ 36.7 min; (E)-4c: t_Rϭ

1907Ϫ1911.

43.8 min. [α]²_D⁷ϭ 12.1 (c ϭ 1.23, CH₂Cl₂) for 3c with 85 %

ee (R).

Chem. Soc. 1995, 117, 6130Ϫ6131.

[22]

0.5 mmol of the corresponding substrate (20, 21 or cyclopen-

tenyl acetate) was treated with 1 mmol of NaCH(CO₂Me)₂in

the presence of [IrCl(COD)]₂(0.01 mmol) and P(OPh)₃

(0.02 mmol) in THF (4 mL). After 3 h at room temperature

and a further 15 h under reflux, the reaction mixture was

worked up.

B. Goldfuss, U. Kazmaier, Tetrahedron 2000, 56, 6493Ϫ6496.

Trans influence or ‘‘transphobia’’ can strongly affect the

stability and reactivity of organometallic compounds. See for

[43]

[44]

[45]

Y. D. Ward, L. A. Villanueva, G. D. Alfred, S. C. Payne, M.

A. Semones, L. S. Liebeskind, Organometallics 1995, 14,

4132Ϫ4156.

(E)-2b has been prepared previously by a different route: Z.

Rappoport, S. Winstein, W. G. Young, J. Am. Chem. Soc. 1972,

94, 2320Ϫ2329.

H. Matsuhashi, S. Asai, K. Hirabayashi, Y. Hatanaka, A.

Mori, T. Hiyama, Bull. Chem. Soc. Jpn. 1997, 70, 1943Ϫ1952.

M. T. Nunez, V. S. Martın, J. Org. Chem. 1990, 55, 1928Ϫ1932.

[23]

[24]

[46]

[47]

´

example: J. Vicente, J. A. Abad, A. D. Frankland, M. C. Ramı-

rez de Arellano, Chem. Eur. J. 1999, 5, 3066Ϫ3075 and refer-

L. A. Gorthey, M. Vairamani, C. Djerassi, J. Org. Chem. 1984,

ences therein.

49, 1511Ϫ17.

[25] [25a]

˚

[48]

[49]

B. Akermark, S. Hansson, A. Vitagliano, J. Am. Chem.

´

S. W. Baldwin, J. Aube, P. M. Gross, Tetrahedron Lett. 1987,

Soc. 1990, 112, 4587Ϫ4588. ^[25b]M. Sjögren, S. Hansson, P.-O.

Norrby, B. Akermark, M. E. Cucciolito, A. Vitagliano, Or-

28, 179Ϫ182.

˚

For the preparation of (R)-1b the reaction was stopped at a

conversion of 60%, the unconverted alcohol isolated by flash

column chromatography and acetylated. (S)-1b: 36%, 97.7% ee;

(R)-1b: 40%, 94.8% ee. [α]²_D⁰ϭ Ϫ1.5 (c ϭ 1.17, CHCl₃) for 1b

with 95.5% ee (R). HPLC: DAICEL Chiralcel ODH column,

length: 25 cm ϩ 5 cm precolumn, flow: 0.5 mL·min^Ϫ1, eluent:

n-hexane/iPrOH (99.5:0.5); 1b: t_R(R) ϭ 20.1 min, t_R(S) ϭ

22.6 min.

[25c]

ganometallics 1992, 11, 3954Ϫ3964.

Zumpe, Angew. Chem. 2000, 112, 805Ϫ807; Angew. Chem. Int.

U. Kazmeier, F. L.

Ed. 2000, 39, 802Ϫ804.

[26]

[27]

P. A. Evans, J. E. Robinson, Org. Lett. 1999, 1, 1929Ϫ1931.

Survey on Ir complexes: G. J. Leigh, R. L Richards, Compre-

hensive Organometallic Chemistry, Vol. 5, Pergamon, Oxford,

1982.

Eur. J. Inorg. Chem. 2002, 2569Ϫ2586

2585

´

B. Bartels, C. Garcıa-Yebra, F. Rominger, G. Helmchen

FULL PAPER

[50]

T. E. Wessel, J. A. Berson, J. Am. Chem. Soc. 1994, 116,

E. Eliel, L. Giza, A. Chester, J. Org. Chem. 1968, 33,

3754Ϫ3758.

Compound 11f has been prepared previously by a different

route: V. Bertine, F. Lucchesini, M. Pocci, A. de Munno, Hete-

rocycles 1995, 41, 675Ϫ688.

Compound 5f has been prepared previously by a different

route: A. B. Reitz, S. O. Nortey, B. E. Maryanoff, J. Org. Chem.

1987, 52, 4191Ϫ4202.

495Ϫ505.

[51]

[54]

[55]

Compound 11e was prepared previously by another route: M.

F. Shostakovskii, N. V. Kuznetsov; Y. B. Zaretskaya, Bull.

Acad. Sci. USSR Div. Chem. Sci. (Engl. Trans.) 1963,

830Ϫ831; Izv. Akad. Nauk SSSR, Otd. Khim. Nauk 1963,

922Ϫ923.

[52]

Compound 5e was prepared previously by another route: J.

Roos, F. Effenberger, Tetrahedron: Asymmetry 1999, 10,

2817Ϫ2828.

A different synthesis for compound 10f was first described by:

Received April 2, 2002

[I02168]

[53]

2586

Eur. J. Inorg. Chem. 2002, 2569Ϫ2586

Article Doi

Iridium-catalysed allylic substitution: Stereochemical aspects and isolation of IrIII complexes related to the catalytic cycle

DOI: 10.1002/1099-0682(200210)2002:10<2569::AID-EJIC2569>3.0.CO;2-5

Source and publish data:

Authors:

Article abstract of DOI:10.1002/1099-0682(200210)2002:10<2569::AID-EJIC2569>3.0.CO;2-5

Full text of DOI:10.1002/1099-0682(200210)2002:10<2569::AID-EJIC2569>3.0.CO;2-5

Products guided by the article

R&D Labs maybe for 501909-58-6

Relevant to this article

Hot Product