Ti-catalyzed Barbier-type allylations and related reactions

DOI: 10.1002/chem.200802180

Ti-Catalyzed Barbier-Type Allylations and Related Reactions

Rosa E. Estꢀvez,^[a]Josꢀ Justicia,^[a]Btissam Bazdi,^[a]Noelia Fuentes,^[a]Miguel Paradas,^[a]

Duane Choquesillo-Lazarte,^[b]Juan M. Garcꢁa-Ruiz,^[b]Rafael Robles,^[a]

Andreas Gansꢂuer,^[c]Juan M. Cuerva,*^[a]and J. Enrique Oltra*^[a]

2774

ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2009, 15, 2774 – 2791

FULL PAPER

Abstract: Titanocene

(III) complexes,

jugated aldehydes proceed through a

coupling of a Ti^IV-bound ketyl radical

with an allyl radical. Reactions of ke-

tones with allylic halides take place by

the classical addition of an allylic or-

ganometallic reagent. The radical cou-

pling processes enable transformations

such as the highly regioselective a-pre-

nylation that are otherwise difficult to

achieve. The mild reaction conditions

and the possibility to employ titano-

cene complexes in only catalytic quan-

tities are highly attractive features of

our protocol. These unusual properties

have been taken advantage of for the

straightforward synthesis of the natural

products rosiridol, shikalkin, and 12-hy-

droxysqualene.

easily generated in situ from commer-

cial Ti^IVprecursors, catalyze Barbier-

type allylations, intramolecular crotyla-

tions (cyclizations), and prenylations of

a wide range of aldehydes and ketones.

The reaction displays surprising and

unprecedented mechanistic subtleties.

In cyclizations a fast and irreversible

addition of an allyl radical to a Ti^III-co-

ordinated carbonyl group seems to

occur. Intermolecular additions to con-

Keywords: allylation · homogene-

ous catalysis · natural products ·

synthetic methods · titanium

As long ago as 1899 Phillipe Barbier reported a coupling

reaction between a ketone (6-methyl-5-hepten-2-one) and

an alkyl halide (CH₃I) in the presence of a stoichiometric

quantity of magnesium metal,^[1]thus establishing the basis

Titanium, the seventh most abundant metal on earth, is

one of the cheapest transition metals and a lot of titanium

compounds are nontoxic and environmentally friendly.^[8]

Moreover, allyltitanium complexes have proven to be capa-

ble of reacting with carbonyl compounds with considerable

chemo-, regio-, diastereo-, and even enantioselectivity.^[8,9]

These reactions, however, require stoichiometric quantities

of the titanium complex, which is disadvantageous in the

case of enantioselective additions. Moreover, due to the

closed transition state of allylation reactions with organotita-

nium complexes, a-prenylations, which are important in nat-

ural product synthesis, cannot be realized.

À

for the one step C C bond-forming process currently known

as the Barbier reaction.^[2]The one-step strategy of this reac-

tion is often more convenient than the two-step one (involv-

ing the preparation of the organometallic reagent and subse-

quent coupling with the carbonyl derivative) characteristic

of Grignard-type processes. This is especially so in two

cases. First, with allylic halides the Grignard reagent can be

difficult to prepare in high yields.^[3]Second, cyclization reac-

tions can in principle be carried out efficiently only under

the Barbier-type conditions.

In this context we deemed that titanoceneACTHNUGRTENUNG(III) complexes

(such as [TiClCp₂] and others)^[10]might be used in a Barbi-

er-type strategy to transform allyl halides into allyl radicals,

which would subsequently react with a carbonyl compound

present in the medium. In this manner, new reactivity mani-

folds for addressing these regioselectivity issues might

become available.

Due to the considerable synthetic relevance of allylation

reactions,^[4]different transition metals have been assayed in

Barbier-type allylations,^[2,5]including Sn, Pb, In, Zn, Cr (the

Nozaki–Hiyama–Kishi allylation),^[6]and SmI₂(the samarium

Barbier reaction).^[7]Nevertheless, the use of stoichiometric

proportions of many of these metals has serious limitations

due to their toxicity, low solubility causing problems in the

reproducibility of results and/or high costs. Therefore, the

development of novel, safer, and more sustainable reactions

for the realization of Barbier-type allylations and related re-

actions remains an attractive and important goal.

Additionally, with the aid of titanocene-regenerating

agents such as 1 (generated by mixing Me₃SiCl and 2,4,6-

collidine),^[11]the process should become catalytic in titanium

(Scheme 1).^[12]Such a procedure is highly attractive for the

[a] R. E. Estꢁvez, Dr. J. Justicia, B. Bazdi, N. Fuentes, M. Paradas,

Dr. R. Robles, Dr. J. M. Cuerva, Prof. J. E. Oltra

Department of Organic Chemistry

University of Granada, Faculty of Sciences

C. U. Fuentenueva s/n 18071 Granada (Spain)

Fax : (+34)958-248437

E-mail: joltra@ugr.es

[b] Dr. D. Choquesillo-Lazarte, Prof. J. M. Garcꢂa-Ruiz

Laboratorio de Estudios Cristalogrꢃficos

IACT, CSIC-University of Granada

P. T. Ciencias de la Salud, 18100, Granada (Spain)

[c] Prof. A. Gansꢄuer

Kekulꢁ Institut fꢅr Organische Chemie und Biochemie

University of Bonn, Gerhard Domagk Strasse 1

53121 Bonn (Germany)

Supporting information for this article is available on the WWW

under http://dx.doi.org/10.1002/chem.200802180.

Scheme 1. Anticipated catalytic cycle for Ti-induced Barbier-type

allylations.

Chem. Eur. J. 2009, 15, 2774 – 2791

ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

2775

J. E. Oltra, J. M. Cuerva et al.

Table 1. Barbier-type allylation of carbonyl compounds 2–7 promoted by

stoichiometric proportions of [TiClCp₂].

development of enantioselective Barbier-type allylations

with enantiomerically pure titanium catalysts.

Electrophile

Halide

Product (yield)

À

Here, we disclose a comprehensive study of our novel C

bond-forming process that features Barbier-type

C

allylation, intramolecular crotylation (cyclization), and pre-

nylation reactions catalyzed by titanocene(III) reagents. We

8

AHCTUNGTRENNUNG

also describe the straightforward synthesis of the natural

products rosiridol, shikalkin, and 12-hydroxysqualene via a-

prenylations that are unprecedented with titanium.

8 or 9

8

Results and Discussion

Barbier-type allylations promoted and catalyzed by

8 or 9

titanocene

ACHTUNGTRENNUNG

bis(cyclopentadienyl)titaniumAHCUTNGTRENNUNG

be a useful single-electron-transfer agent to promote and

catalyze the homolytic ring opening of epoxides,^[13]pinacol

couplings of conjugated aldehydes,^[14]stereoselective cou-

plings between aldehydes and conjugated alkenals,^[15]Refor-

8

9

matsky-type processes,^[16]divergent C C bond-forming reac-

À

tions with modulation by Ni or Pd,^[17]and other free-radical-

based transformations, thus becoming a formidable tool in

organic synthesis.^[18][TiClCp₂] can be prepared by reaction

between TiCl₃and thallium cyclopentadienide,^[13a]or simply

generated in situ by stirring commercially available

[TiCl₂Cp₂] with Zn or Mn dust,^[10,13a]which is often the most

convenient procedure from a practical point of view. It is

known, however, that Zn is capable of promoting not only

Reformatsky reactions,^[19]but also Barbier-type allylations^[5]

and consequently might interfere in the titanium-mediated

process. So, as it is believed that without an activating agent

such as iodine or ZnCl₂Mn does not promote Barbier-type

allylations in THF,^[20,21]we chose this metal to generate

[TiClCp₂] in situ for the experiments described in this

report.

probably because the slow addition of the halide partner

minimizes the concentration of allyl radicals and thus re-

duces the possibility of radical–radical homocoupling side

reactions, which would lead to undesirable Wurtz-type by-

products.^[23]It is known, however, that aromatic and a,b-un-

saturated aldehydes are prone to pinacol coupling in the

presence of [TiClCp₂];^[14]a side reaction that might also

occur with conjugated ketones, although presumably at a

slower rate due to steric factors. Therefore, we deemed that

the experimental procedure should be changed for sub-

strates 4–7. In fact the best yields for alcohols 12 (75%), 13

(99%), 14 (85%), and 15 (92%; Table 1) were obtained by

the simultaneous addition of allyl halide and carbonyl com-

pound into a solution of [TiClCp₂] in THF. In this way the

formation of pinacol-coupling byproducts was minimized.

Allyl bromide (8) and allyl chloride (9) provided similar

yields for tertiary alcohols 13 and 15. In contrast, the use of

chloride 9 instead of bromide 8 provided substantially lower

yields for secondary alcohols 12 and 14, possibly because

the formation of an allyl radical from chloride 9 (BDE=

71.3 kcalmol^À1)^[24]was slower than from bromide 8 (BDE=

56.7 kcalmol^À1)^[24]and thus the fast pinacol coupling of con-

jugated aldehydes 4 and 6 predominated.

We started by exploring the experimental conditions suit-

able for optimizing the yields for the allylation of model car-

bonyl compounds, including aliphatic (2, 3), aromatic (4, 5),

and a,b-unsaturated aldehydes and ketones (6, 7), with allyl

halides 8 and 9, promoted by an excess of [TiClCp₂]

(2.5 equiv) at room temperature (Scheme 2).

Scheme 2. Barbier-type allylation of carbonyl compounds 2–7 promoted

by [TiClCp₂].

Organometallic catalysis plays an important role in both

laboratory and industrial organic synthesis.^[25]Therefore we

decided to assay a Ti-catalyzed version of our allylation pro-

cess. To this end we treated carbonyl compounds 2–7 with

halides 8 or 9 in the presence of a mixture of a substoichio-

metric proportion of [TiCl₂Cp₂] (0.2 equiv),^[10]relatively

The best yields^[22]for alcohols 10 (91%) and 11 (99%)

were obtained by adding the allylic halide slowly into a pre-

viously prepared solution of [TiClCp₂] and the correspond-

ing carbonyl compound (2 or 3) in THF (8 and 9 provided

similar yields; the best being set out in Table 1). This is

2776

www.chemeurj.org

ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2009, 15, 2774 – 2791

Barbier-Type Allylations

FULL PAPER

Table 2. TitanoceneACHTNUTGRNEUNG(III)-catalyzed Barbier-type allylation of aldehydes

16–21.

cheap Mn dust (8 equiv), and a combination of Me₃SiCl

(4 equiv) and 2,4,6-collidine (7 equiv) to form the titano-

cene-regenerating agent 1.^[26]Thus we obtained good-to-ex-

cellent yields of homoallylic alcohols 10–15 (Scheme 3). A

control experiment in the absence of titanium did not pro-

vide any coupling product.

Aldehyde

X^[a]

Product (yield)

8

9

Scheme 3. Titanocene-catalyzed allylation of 2–7. [a] Yield obtained em-

ploying 8. [b] Yield obtained employing 9.

8

The results summarized in Scheme 3, obtained with tita-

nocene quantities one order of magnitude lower than in the

stoichiometric procedure, supported the viability of the cata-

lytic version and pointed to the potential synthetic value of

the Ti-catalyzed Barbier-type allylation process. The high

yields obtained for allylation of ketones 3, 5, and 7 were es-

pecially intriguing, because some years ago Roy et al. re-

ported that ketones did not react in Barbier-type allylations

promoted by [TiClCp₂] under the reaction conditions they

used.^[27]Our results demonstrate however that under our

conditions [TiClCp₂] can promote and catalyze the Barbier-

type allylation of ketones to produce good yields of tertiary

homoallylic alcohols.

8

[a] Alkyl halide that provided the best yield. [b] Mixture syn/anti 2:3.

[c] 1:1 mixture of diastereomers.

Once we were confident about the synthetic potential of

the Ti-catalyzed method we decided to explore its scope,

limitations, and stereochemical behavior in more detail. To

this end we assayed the [TiClCp₂]-catalyzed reaction of al-

dehydes 16–21 (Table 2), ketones 28–34 (Table 3), and func-

tionalized carbonyl derivatives 42–46 (Table 4) with allylic

halides 8 and 9.

The results summarized in Table 2 suggested that the Ti-

catalyzed procedure might become a general method for the

allylation of aldehydes, including the selective 1,2-addition

to a,b-unsaturated aldehydes such as 20 and 21, with good

yields. Additionally, a modest stereoselectivity was observed

for the allylation of the a-substituted aldehyde 17 (syn/anti-

isomers ratio 3:2). Moreover, the yields we obtained for

these reactions employing stoichiometric proportions of

[TiClCp₂] were roughly similar to those presented in

Table 2, lending weight to the idea that this catalytic cycle is

effective.^[28]

The results summarized in Table 3 confirmed that the pro-

cedure was also useful for the allylation of aliphatic, aromat-

ic, and a,b-unsaturated ketones, including cyclic and acyclic

ones. Additionally, the yields obtained from Ti-catalyzed re-

actions were roughly similar to those obtained employing

stoichiometric proportions of [TiClCp₂],^[28]highlighting once

more the usefulness of the catalytic version. It is known that

in the allylation of 4-tert-butylcyclohexanone (30) by

Nozaki–Hiyama–Kishi^[29]and samarium Barbier^[30]proce-

dures, or with allylzinc^[31]and allylindium^[32]reagents, the

equatorial attack prevails. Our Ti-catalyzed allylation of 30,

on the other hand, led mainly to the product derived from

axial attack, which is comparable to the results obtained by

Reetz et al. using previously prepared allyltitanium re-

agents.^[33]This observation suggested that, for the intermo-

lecular allylation of ketones at least, an alternative mecha-

nism via an allyl-Ti^IVintermediate and subsequent nucleo-

philic attack to the carbonyl group could not be ruled out.

As we will see later, this seems to be the case for the inter-

molecular prenylation of ketones (for a detailed mechanistic

discussion see the above section devoted to Barbier-type

prenylation reactions). Additionally, the considerable stereo-

selectivity observed for the Ti-catalyzed synthesis of 41 fol-

lowed the same trend as the recently reported allylation of

cyclohexenone 34 with stoichiometric proportions of allyl-

magnesium and allylindium species.^[34]Our Ti-catalyzed pro-

cedure has the advantage that it does not require the prepa-

ration of stoichiometric quantities of any organometallic re-

agent. Finally, it should be noted that no cyclopropane rear-

rangement products were detected in the allylation of cyclo-

propyl ketone 33.^[35]

The results summarized in Table 4 indicate that under the

mild conditions used the Ti-catalyzed procedure is compati-

Chem. Eur. J. 2009, 15, 2774 – 2791

ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

2777

J. E. Oltra, J. M. Cuerva et al.

Table 4. TitanoceneACHTNUTGRNEUNG(III)-catalyzed Barbier-type allylation of functional-

ized carbonyl derivatives 42–46.

Table 3. TitanoceneACHTUNGTRENNUNG(III)-catalyzed Barbier-type allylation of ketones 28–

34.

X^[a]

Product (yield)

Substrate

X

Product (yield)

Ketone

9

8

9

8

[a] Mixture of a and b-allyl derivatives at a ratio of 2:1. [b] 3:2 mixture

of R and S epimers at C-1’.

8

9

Scheme 4. Chemoselective Ti-catalyzed allylation of aldehyde 2 in the

presence of ketone 3.

[a] Alkyl halide. [b] Mixture of axial and equatorial attack products in a

2:1 ratio. [c] Together with the main product 41, a minor proportion

(8%) of its C-1 epimer was obtained.

ble with different functional groups, including alkyl halides,

phenols, and ketals. Moreover, the easy lactonization that

occurred after the allylation of keto ester 43 suggested that

this procedure might become a useful tool for the synthesis

of g-lactones. Finally, the Ti-catalyzed allylation of benzoyl

chloride 46 gave an acceptable 56% yield of the considera-

bly labile, benzylic tertiary alcohol 51, thus highlighting the

mildness of our method.

Asymmetric catalysis plays a crucial role in contemporary

organic synthesis.^[37]Therefore, we decided to assay some

enantiomerically pure titanium catalysts to check the possi-

bility of achieving enantioselective allylation processes using

our Ti-catalyzed method. To this end we chose the commer-

cial Brintzinger complex dichloroACTHNUGTRNEUNG(R,R)-ethylenebis(4,5,6,7-

tetrahydro-1-indenyl)titanium(IV) (52) and Kaganꢆs com-

plex 53 prepared in our laboratory.^[38]

Selectivity is one of the most desirable properties for

novel methods in organic synthesis.^[36]Therefore we assayed

the capacity of this procedure for discriminating between al-

dehydes and ketones. The results of the competing experi-

ment depicted in Scheme 4 indicated that the Ti-catalyzed

allylation of aldehyde 2 was much faster than that of ketone

3. This phenomenon might be advantageously exploited for

the selective allylation of aldehydes in the presence of ke-

tones.

Allylation of 3,4,5-trimethoxybenzaldehyde (19) catalyzed

by 52 gave a 50% yield of (S)-(À)-25,^[39]with a 33% enan-

tiomeric excess (ee) (Scheme 5).^[40]Moreover, the allylation

of 2 catalyzed by 53 afforded an 80% yield of (S)-(À)-10,^[41]

with a 20% ee.^[40]Despite these moderate ee values, the op-

tical activity observed for products (À)-10 and (À)-25 dem-

onstrated for the first time that Ti-catalyzed Barbier-type

allylations can be conducted in an enantioselective manner

by using chiral titanium catalysts.^[12]

2778

www.chemeurj.org

ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2009, 15, 2774 – 2791

Barbier-Type Allylations

FULL PAPER

shift of its equatorial methyl group with that of sulfone 62

(vide infra)^[43]and was confirmed by the NOE signals ob-

served between the equatorial methyl group and the axial

hydrogen atoms H-2 and H-6b of acetyl derivative 57

(Scheme 6). These results revealed the stereoconvergent

character of this Ti-catalyzed cyclization process.

To gain information about the scope and limitations of

the cyclization process we prepared ketones 58 and 59 to-

gether with isomeric aldehydes 60 and 61 and treated them

with [TiClCp₂] (Table 5).

Table 5. Ti-promoted/catalyzed Barbier-type cyclization of aldehydes and

ketones.

Substrate

Products (yield)

Scheme 5. Ti-catalyzed enantioselective allylations of 2 and 19.

Ti-catalyzed intramolecular Barbier-type crotylations—ste-

ACHTUNGTRENNUNGreoconvergent cyclization of allylic halides: The g-regiose-

lectivity and considerable diastereoselectivity of intermolec-

ular carbonyl allylations with stoichiometric proportions of

crotyltitanium reagents are well documented.^[9,42]To the

best of our knowledge, however, there is no precedent for

intramolecular Barbier-type crotylations (cyclizations) cata-

lyzed by any metal. In this context, we decided to assay the

intramolecular version of our Ti-catalyzed procedure. Thus

we prepared isomeric crotyl-type bromides 54 and 55, and

treated them with

a substoichiometric proportion of

[TiClCp₂] at room temperature (Scheme 6). In both cases

[a] A stoichiometric proportion of [TiClCp₂] was used in this case. [b] A

substoichiometric proportion of [TiClCp₂] was used in this case.

Ti-catalyzed cyclization of ketones 58 and 59 led mainly

to products 62 and 63 respectively (Table 5), confirming the

usefulness of this procedure for the stereoselective synthesis

of not only cycloalkanols, but also piperidine derivatives.

Additionally, tertiary alcohol 62 crystallized from diethyl

ether and thus its structure and relative configuration could

be established unambiguously by X-ray diffraction analysis

(Figure 1). Moreover, the cyclization of sulfone 58 promoted

by a substoichiometric quantity of [TiClCp₂] gave a 38%

yield of 62 together with a minor proportion of 67, its C-1

epimer (12%). Therefore, we could compare the physical

properties of both diastereomers, including their NMR

data.^[43]

Scheme 6. Ti-catalyzed Barbier-type intramolecular allylation (cycliza-

tion) of 54 and 55 (double arrows indicate reciprocal NOEs observed for

57).

Intriguingly, the Ti-catalyzed cyclization of aldehydes 60

and 61 gave roughly 1:1 mixtures of isomers 65 and 66 with

an almost complete loss of stereoselectivity. These results

strongly suggested that the methyl groups of ketones 54, 55,

58, and 59 played a crucial role in controlling the stereo-

we obtained good yields of vinyl cycloalkanol 56 (82 and

94% respectively) with excellent stereoselectivity (the

1S*,2R* diastereomer 73 was not detected). The relative

configuration of 56 was assigned by comparing the chemical

Chem. Eur. J. 2009, 15, 2774 – 2791

ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

2779

J. E. Oltra, J. M. Cuerva et al.

immediately after the formation of the allylic radical 70.

Formation of the same radical intermediate (70) from both

coordinated species 68 and 69 results in the stereoconver-

gent nature of our process. Finally, the stereoselectivity of

the cyclization is governed by the pseudo-equatorial disposi-

tion of the methyl group in the transition state 71. This en-

forces the radical attack from the si face of the ketone. In

the case of aldehydes 60 and 61, which lack the methyl

group, radical attack can take place from both prochiral

faces of the carbonyl group, resulting in an unselective over-

all reaction.

According to the above mechanism, Ti^IIIplays a triple

role: it acts as a Lewis acid for carbonyl coordination, as a

halogen-atom-trapping reagent for the generation of allylic

radicals, and finally enforces the reductive termination of

the overall process.

Figure 1. Three-dimensional structure of alcohol 62 established by X-ray

diffraction analysis: C=gray, H=white, O=red, S=green.

chemistry of the cyclization process. The above observations

are the basis for our mechanistic proposal in Scheme 7 out-

lined for isomeric ketones 54 and 55.

In this context it is interesting to use a second, redox-inac-

tive, Lewis acid to compete with [TiClCp₂] for carbonyl co-

ordination, thus making the radical cyclization process re-

versible.^[44]In this case, the allylic radical formed by Ti^IIIab-

straction of the halogen atom could be trapped by a second

Ti^IIIspecies, giving rise to an organometallic crotyl–Ti^IVin-

termediate. It is known that in the presence of a Lewis acid

the reaction between crotyl–Ti^IVcomplexes and carbonyl

compounds provides mixtures of diastereomers.^[42a]There-

fore, the decrease in stereoselectivity in the cyclization of a

model ketone such as 55 can be regarded as a probe indicat-

ing the participation of an alternative mechanism via a

crotyl-Ti^IVintermediate, followed by nucleophilic attack on

the carbonyl group. As we expected, the treatment of 55

with a stoichiometric proportion of [TiClCp₂] (2 equiv)^[45]in

the presence of the Lewis acid [ErACTHNUTRGNE(UNG CF₃O₃S)₃] (2 equiv) gave

a mixture of isomers 56 (58% yield) and 73 (19%; d.r.=3:1;

Scheme 8). The minor isomer arises from the cyclization of

Scheme 8. Ti-promoted cyclization of 55 in the presence of Er^III

.

a crotyl–Ti^IVintermediate. Moreover, when we treated 55

Scheme 7. Proposed mechanism for Ti-catalyzed Barbier-type cyclization

of ketones (outlined for isomers 54 and 55).

with [TiClCp₂] in the presence of BF₃etherate, a harder

Lewis acid than [ErACTHNUTRGNEUNG

(CF₃O₃S)₃],^[46]we obtained a mixture of

56 (51%) and an increased proportion of 73 (33%; d.r.=

1.5:1), supporting the idea that the addition of Lewis acids

can shift the cyclization mechanism from the radical cou-

pling depicted in Scheme 7 to a nucleophilic g-attack by a

preformed crotyl–Ti^IVintermediate. The experiments de-

scribed above allowed us to isolate the equatorial alcohol 73

and compare its physical properties, including NMR data,

with those of the axial isomer 56.^[43]

Our analysis commences with the coordination of the car-

bonyl group by the dimer species [(TiClCp₂)₂] with the con-

comitant release of a monomer [TiClCp₂] species in close

proximity to the corresponding allylic halide.^[10]Thus, de-

spite of using substoichiometric proportions of the titanium

catalyst, the dimer species can account for both carbonyl co-

ordination and the subsequent halogen-atom abstraction

processes. Ti–carbonyl coordination is essential for the acti-

vation of the carbonyl group and for the irreversible termi-

nation of the 6-exo radical cyclization,^[44]which takes place

We subsequently assayed the cyclization of model ketone

55 with other titanium catalysts. Thus, the cyclization of 55

catalyzed by [Ti^IIICl₂Cp] (generated in situ by stirring com-

2780

www.chemeurj.org

ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2009, 15, 2774 – 2791

Barbier-Type Allylations

FULL PAPER

mercial [Ti^IVCl₃Cp] with Mn dust) gave a 64% yield of 56

(no 73 was detected) with the same stereoselectivity ob-

served for [TiClCp₂]. In contrast, when we employed

a- and g-addition products at ratios of about 1:4 and 2:3 re-

spectively (Table 6). However, Ti-promoted Barbier-type

prenylation of conjugated aldehydes 6, 20, 21, 78, and 79

dichloroACHTUNGTRENNUNG(R,R)-ethylenebis(4,5,6,7-tetrahydro-1-indenyl)tita-

nium(IV) (52) as pre-catalyst we obtained a mixture of dia-

stereomers (À)-56 (36% yield; 20% ee)^[47]and (À)-73

(29%; 19% ee)^[47]with a diastereomeric ratio (d.r.=1.2:1)

close to that obtained from the [TiClCp₂]-promoted cycliza-

tion in the presence of BF₃. These results suggest that the

bulky Ti^IIIcomplex formed from 52 has considerably lower

ketone-coordination ability than [TiClCp₂] or [TiCl₂Cp].

Moreover, it should be noted that the optical activity ob-

served for both (À)-56 and (À)-73 strongly supports the idea

that the titanium catalyst actually participates in the crucial

Table 6. Ti-promoted Barbier-type prenylation of aldehydes.

S^[a]

X^[b]

Products (yield)

Y^[c]

1:4

17

74

18

75

2:3

À

C C bond-forming step of the cyclization process.

To the best of our knowledge, this is the first metal-cata-

lyzed Barbier-type cyclization described to date.

6

75^[h]

1:0

Barbier-type prenylations promoted and catalyzed by

[TiClCp₂]: Isoprene units constitute the building blocks of

natural terpenoids, in which they are generally linked in a

“head-to-tail” manner, although occasionally they lie “head-

to-head” (e.g., the central bonds of squalene and phytoene).

In both cases, however, at least one of the isoprene units is

linked at the a-position.^[48]Therefore, a-prenylation reac-

tions might facilitate the chemical synthesis of this valuable

family of natural products. Unfortunately, few methods have

been described for carbonyl a-prenylation and they require

stoichiometric proportions of preformed organometallic

complexes derived from light rare-earth elements, relatively

expensive Sm, or reactive Ba, and generally provide mix-

tures containing variable amounts of g-addition byprod-

ucts.^[30,49]Moreover, it is known that both crotyl–Ti^IVand

prenyl–Ti^IVcomplexes are prone to add to carbonyl com-

pounds at the g-position.^[42]Nevertheless, we deemed that

because of the potential free-radical nature of the coupling

step of Ti^III-mediated Barbier-type allylations (see

Scheme 1) this kind of reaction might afford a convenient

procedure for the a-prenylation of carbonyl derivatives.^[50]

To check our hypothesis we treated decanal (2) with

prenyl bromide (74) and prenyl chloride (75) in the presence

of [TiClCp₂] (2.2 equiv; Scheme 9).

20

74^[h]

21

78

75^[h]

1:0

75^[h]

79

75^[h]

1:0

[a] S=substrate. [b] X=halide that provided the best yield. [c] Y=a/g-

prenylation products ratio. [d] 9:1 mixture of diastereomers. [e] 2:3 mix-

ture of diastereomers. [f] 7:3 mixture of diastereomers. [g] 1:1 mixture of

diastereomers. [h] A mixture of aldehyde and prenyl halide in THF was

slowly added into a THF solution of [TiClCp₂].

generated only the desired a-

prenylation products (84–88;

Table 6). Regio- and stereospe-

cific synthesis of product 84

showed the potential viability

of this procedure to build

“head-to-head” isoprene linkag-

es in the synthesis of terpe-

Scheme 9. Barbier-type prenylation of decanal (2) promoted by

[TiClCp₂].

noids.

We subsequently assayed the Ti-catalyzed version of our

prenylation process. Thus, we treated aldehydes 2, 6, 18, and

79 with halides 74 or 75 in the presence of a mixture of a

substoichiometric proportion of [TiCl₂Cp₂] (0.2 equiv),^[10]

Mn dust, and a combination of Me₃SiCl and 2,4,6-collidine

In both cases we obtained mixtures of a-addition (76) and

g-addition (77) products at a ratio of close to 1:1. We subse-

quently treated a-substituted (17) and b-substituted (18) al-

dehydes under similar conditions, thus obtaining mixtures of

Chem. Eur. J. 2009, 15, 2774 – 2791

ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

2781

J. E. Oltra, J. M. Cuerva et al.

to form the titanocene-regenerating agent 1.^[26]In this way

we obtained prenylation products 76, 77, 82–84, and 88

(Table 7). The yields and regiochemical trends (a/g-addition

dehydes, and especially to those sterically hindered by a-

substitution, such as 17, may be assumed to be considerably

slower than with conjugated aldehydes. Therefore, prenyl

radicals could accumulate and be eventually trapped by

[TiClCp₂] to form prenyl–Ti^IVintermediates. Consequently,

mixtures of a- and g-prenylation products would be ob-

tained, as was in fact observed.

The higher reactivity of conjugated aldehydes was con-

firmed by the following experiment. When we treated an

equimolar mixture of decanal (1 equiv) and conjugated alde-

hyde 79 (1 equiv) with prenyl bromide (1 equiv) in the pres-

ence of [TiClCp₂], we obtained only 88 (82% yield), the a-

prenylation product from 79. Products 76 and 77, potentially

deriving from the prenylation of decanal, were not detected.

This difference in reactivity must be put down to a substan-

tial difference between the activation energies (AEs) of the

radical coupling steps for conjugated and nonconjugated al-

dehydes. In turn, the difference in AE may well derive from

the different nature of the reactive intermediates involved

in each coupling step (Scheme 11).

Table 7. Ti-catalyzed Barbier-type prenylation of aldehydes.

S^[a]

X^[b]

Products (yield)

Y^[c]

2

18

6

74

76 (38%) + 77 (16%)

82 (36%)^[d]+ 83 (55%)^[d]

84 (83%)

7:3

2:3

1:0

75

74^[e]

79

88 (89%)

[a] S=substrate. [b] X=halide employed. [c] Y=a/g-prenylation prod-

ucts ratio. [d] 1:1 mixture of diastereomers. [e] A mixture of aldehyde

and prenyl halide in THF was slowly added into a THF solution of a sub-

stoichiometric proportion of [TiCl₂Cp₂], Mn dust, and 1.

ratios) were similar to those obtained by using stoichiomet-

ric amounts of [TiClCp₂] (Table 6), thus confirming the val-

idity of the Ti-catalyzed procedure. What is more, the Ti-cat-

alyzed version provided a slightly higher proportion of the

preferred a-prenylated isomer 76.

The dichotomy of the reaction is easily rationalized. The

a-prenylation products are derived from the coupling be-

tween a prenyl radical and a carbonyl compound,^[50]whereas

g-prenylation products originate from the nucleophilic

attack of an organometallic prenyl–Ti^IVintermediate, via a

cyclic transition state analogous to that reported by Sato

et al. (Scheme 10).^[51]

Scheme 11. Hypothetical reactive intermediates involved in the radical

coupling step for a) conjugated and b) non-conjugated aldehydes.

It is believed that conjugated aldehydes are reduced by

[TiClCp₂] to mesomerically stabilized ketyl-type radicals

that are also involved in the pinacol coupling reaction un-

dergone by these aldehydes in the presence of titanocene-

Scheme 10. Divergent pathways towards a- and g-addition products from

Ti-promoted Barbier-type prenylations.

AHCTUNGTRENNUNG

(III).^[14]Therefore, the prenyl radical coupling of conjugated

aldehydes (Scheme 11a) could have the character of a very

fast radical–radical coupling, with a very low AE.^[52]On the

other hand, [TiClCp₂] does not promote pinacol couplings

of nonconjugated aldehydes because the corresponding

ketyl-type radicals are not formed through single-electron

transfer from Ti^III. Therefore, in the radical coupling step of

Prenyl radical coupling to conjugated aldehydes is postu-

lated to be fast and irreversible. Thus, prenyl radicals would

not have the opportunity to be trapped by [TiClCp₂] to form

prenyl–Ti^IVderivatives and so only a-prenylation products

were obtained. Prenyl radical coupling to nonconjugated al-

2782

www.chemeurj.org

ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2009, 15, 2774 – 2791

Barbier-Type Allylations

FULL PAPER

nonconjugated aldehydes (Scheme 11b) the p-bond of the

Ti^III-coordinated carbonyl group has to be broken and con-

sequently the energy of transition state VI increases. So it

seems that in the prenyl radical coupling of nonconjugated

aldehydes Ti^IIIalso plays a triple role: it serves as a Lewis

acid to activate the aldehydes towards radical attack, as a

halide-abstracting reagent to generate prenyl radicals, and

finally serves as a radical terminator to render the radical

coupling step irreversible. In a-substituted aldehydes, such

as 17, carbonyl–Ti^IIIcoordination is less favored due steric

factors. Consequently the carbonyl p-bond is stronger and

the radical coupling step slower, thus providing an increased

proportion of g-prenylation products.

At this point, the possibility of two competing catalytic

cycles for Ti-mediated Barbier-type prenylations (and prob-

ably also for simple allylations) of nonconjugated aldehydes

should be considered. One of these, closely related to that

depicted in Scheme 1, would proceed through the irreversi-

ble coupling between prenyl radicals and Ti^III-coordinated

aldehydes, thus providing a-prenylation products. Another

would proceed via a prenyl–Ti^IVintermediate and subse-

quent nucleophilic attack upon the carbonyl group, thus

generating g-prenylation isomers (Scheme 12).

Scheme 13. Ti-promoted Barbier-type prenylation of ketone 3.

Finally, we assayed the a-prenylation of conjugated alde-

hyde 79 by using commercial complex 52 as a chiral titani-

um pre-catalyst. In this way we obtained a 21% yield of al-

cohol (À)-88 (29% ee).^[47]No g-prenylation product was de-

tected. The optical activity observed for (À)-88 confirmed

that the titanium catalyst actually participated in the crucial

À

C C bond-forming step of the intermolecular coupling pro-

cess.

To the best of our knowledge the Ti-catalyzed a-prenyla-

tion procedure described above is the first metal-catalyzed

a-prenylation method reported so far. The reaction pro-

ceeds at room temperature under mild conditions and in

many cases might be more convenient than the previously

reported carbonyl prenylation methods,^[30,49]which require

stoichiometric proportions of metal complexes that are not

always easy to prepare. Moreover, in the case of conjugated

aldehydes, the Ti-catalyzed method exclusively provides the

desired a-addition product (regiospecificity), which is unusu-

al among previously reported methods.

Ti-catalyzed synthesis of rosiridol, shikalkin, and 12-hydroxy-

squalene: The synthesis of natural products constitutes one

of the most demanding tests of the viability of a new

method in organic synthesis. Therefore we decided to try

out Ti-catalyzed prenylation procedure by synthesizing some

natural products. To this end we chose rosiridol (90), shikal-

kin (91), and 12-hydroxysqualene (92) as target molecules.

Rosiridol (90), a monoterpenoid isolated from different

plants, has in the past been synthesized by heating to reflux

a preformed prenyl–zinc complex with an isoprenic alde-

hyde closely related to 93 in THF (b.p. 668C) for 72 h.^[53a]

Shorter reaction times or lower temperatures provided a g-

prenylation product instead of the desired a-prenylation

one, suggesting that the coupling reaction gave the g-addi-

tion product, which, subject to prolonged heating, rear-

ranged itself into the a-addition isomer.^[53a]More recently,

Lindel et al. have synthesized rosiridol through a BCl₃-pro-

moted coupling between isoprenic aldehydes and a stoichio-

metric quantity of a previously prepared (from halides 74 or

75) prenyl–tin complex at À788C.^[53b]

Scheme 12. Alternative catalytic cycle for Barbier-type prenylation of

non-conjugated aldehydes.

We were subsequently interested in the regiochemistry of

Ti-promoted prenylation of ketones, which are intrinsically

subject to higher steric hindrance than aldehydes. Therefore

we treated 2-decanone (3) with prenyl bromide (74) in the

presence of a stoichiometric proportion of [TiClCp₂]. In this

way we obtained only g-prenylation product 89 (66% yield;

Scheme 13), whereas the corresponding a-prenylation

isomer was not detected.

Using our novel method, we directly obtained rosiridol

from halide 74 and aldehyde 93 (Scheme 14). The key Ti-

This result did not surprise us. It seems that carbonyl–Ti^III

coordination in ketones is weaker than in nonconjugated al-

dehydes. Therefore, prenyl radical coupling is much slower

and prenyl radicals are trapped by [TiClCp₂] to form

prenyl–Ti^IVderivatives. In consequence, a-prenylation prod-

ucts cannot be formed and this represents one of the main

limitations of the Ti-promoted a-prenylation method.

Scheme 14. Ti-catalyzed synthesis of rosiridol (90).

Chem. Eur. J. 2009, 15, 2774 – 2791

ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

2783

J. E. Oltra, J. M. Cuerva et al.

catalyzed, regiospecific a-prenylation step took place in 6 h

at room temperature. To the best of our knowledge this is

the first metal-catalyzed synthesis of rosiridol that proceeds

at room temperature and does not require stoichiometric

quantities of any organometallic reagent. This result high-

lights the usefulness of this procedure in constructing “head-

to-tail” isoprene linkages for the synthesis of terpenoids.

Shikalkin (91) is a racemic mixture of (R)-shikonin and

(S)-alkannin, metabolites derived from the roots of the ori-

ental medicinal herb Lithospermum erythrorhizon^[54]and the

European plant Alkanna tinctoria.^[55]These products have

proved to possess a plethora of pharmacological uses, in-

cluding anti-inflammatory, antibacterial, antifungal, antitu-

moral, analgesic and antipyretic, antithrombotic, immunosti-

mulatory, angiostatic, and wound healing properties.^[56]They

have, therefore, been synthesized by several researchers.^[55–57]

One of the most convenient synthesis, conducted by Torii

et al.,^[57b]proceeds via alcohol 96, prepared from the well-

known aldehyde 95^[57a]in three steps and produces an over-

all yield of 46%. In contrast, the Ti-catalyzed prenylation of

95 directly provided a 72% yield of 96. Thus we completed

the formal synthesis of shikalkin (Scheme 15) and improved

Scheme 16. Ti-catalyzed synthesis of 12-hydroxysqualene (92).

catalyzed synthesis of 12-hydroxysqualene, which proceeds

at room temperature and does not require stoichiometric

quantities of any organometallic reagent. This result con-

firms that our method can be employed for the coupling not

only of simple isoprene units, but also more complex poly-

prene moieties. This capacity might be usefully exploited for

the synthesis of higher terpenoids.

Conclusion

We have demonstrated that titanoceneACHTNUTRGNEUNG(III) complexes,

easily generated in situ from commercial Ti^IVprecursors,

can promote and catalyze Barbier-type allylations, intramo-

lecular crotylations (cyclizations), and prenylations of car-

bonyl compounds, including a wide range of aldehydes and

ketones. These reactions take place at room temperature

under mild conditions compatible with many functional

groups, provide good yields of open-chain and cyclic homo-

allylic alcohols, including heterocyclic derivatives, afford

moderate-to-high diastereoselectivity, and can be conducted

enantioselectively by using chiral titanium catalysts.

Ti-catalyzed Barbier-type cyclizations and intermolecular

reactions with conjugated aldehydes seem to be relatively

fast and irreversible radical-coupling processes. The former

probably proceeds through intramolecular coupling between

a crotyl-type radical and a Ti^III-coordinated carbonyl group.

The latter possibly occurs by intermolecular coupling be-

tween an allylic and a Ti^IV-bonded ketyl-type radical. Inter-

molecular reactions with ketones are the slowest processes

described here. Their mechanism presumably proceeds via

the generation of an organometallic allyl–Ti^IVintermediate

followed by nucleophilic attack upon the carbonyl deriva-

tive. Intermolecular reactions with nonconjugated aldehydes

are medium-rate processes in which both radical coupling

and organometallic attack (summarized in Schemes 1 and 12

respectively) would seem to participate.

As a probable consequence of the radical coupling mech-

anism, the Ti-catalyzed a-prenylation of conjugated alde-

hydes is regiospecific. This unusual property was advanta-

geously exploited for the straightforward synthesis of the

natural products rosiridol, shikalkin, and 12-hydroxysqua-

lene.

In recent years the development of enantioselective

allylation reactions has become an important goal in chemi-

cal synthesis.^[59]In this work we have paved the way to Ti-

catalyzed, enantioselective Barbier-type allylation, cycliza-

tion, and prenylation reactions. Nevertheless, currently

Scheme 15. Ti-catalyzed formal synthesis of shikalkin (91).

considerably upon Toriiꢆs procedure.

12-Hydroxysqualene (92), the major product from the bio-

transformation of presqualene diphosphate catalyzed by the

enzyme squalene synthase in the absence of NADPH, is ap-

parently produced by water attack upon the carbocation in-

termediate precursor of squalene biosynthesis.^[58]Yamamoto

and co-workers are responsible for the only chemical syn-

thesis of 92 described to date, by treating (E,E)-farnesal

(97) with an stoichiometric proportion of the preformed al-

lylic barium complex derived from (E,E)-farnesyl chloride

(98).^[49b]This reaction demands a very low temperature

(À958C) and affords a 75% yield of 92.

In our laboratory, Ti-catalyzed Barbier-type coupling be-

tween 97 and 98 at room temperature, directly provided a

76%

yield

of

(E,E,E,E)-12-hydroxysqualene

(92;

Scheme 16). No regio- or stereoisomers were detected, thus

demonstrating the regio- and stereospecific character of the

process. To the best of our knowledge, this is the first metal-

2784

www.chemeurj.org

ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2009, 15, 2774 – 2791

Barbier-Type Allylations

FULL PAPER

solution of KHSO₄and extracted with EtOAc. The organic layer was

washed with brine and dried (anhyd Na₂SO₄), and the solvent was re-

moved. Products 12–15, 25–27, 38–41, and 51 were purified by flash chro-

matography on silica gel (hexane/EtOAc) and characterized by spectro-

scopic techniques. Yields obtained are reported in Scheme 3 and

Tables 2–4. In some experiments, trimethylsilyl derivatives were ob-

served. In these cases, the residue was solved in THF (20 mL) and stirred

with Bu₄NF (10 mmol) for 2 h. The mixture was then diluted with

EtOAc, washed with brine and dried (anhyd Na₂SO₄), and the solvent

was removed.

available titanium complexes only afforded moderate ee. At

the moment, we are engaged in the rational design and syn-

thesis of novel, more efficient chiral titanium catalysts.

Experimental Section

General details: For all reactions with titanocene, solvents and additives

were thoroughly deoxygenated prior to use. The following known com-

pounds were isolated as pure samples and showed NMR spectra match-

ing those of the reported compounds: 10,^[60]11,^[17]12,^[60]13,^[42b]15,^[61]

22,^[62]23,^[63]24,^[17]25,^[64]27,^[65]37,^[66]38,^[67]39,^[68]41,^[34]47,^[69]48,^[70]49,^[71]

51,^[72]65,^[17]66,^[17]73,^[17]77,^[73]81,^[74]89,^[17]90,^[53]92,^[58]95,^[56]and 96.^[57b]

Ti-catalyzed chemospecific coupling of allyl chloride with decanal in the

presence of 2-decanone: Strictly deoxygenated THF (20 mL) was added

to a mixture of [TiCl₂Cp₂] (32 mg, 0.13 mmol) and Mn dust (282 mg,

5.13 mmol) under an Ar atmosphere and the suspension was stirred at

room temperature until it turned lime green (after about 15 min). Then,

General procedure for Ti^III-mediated Barbier-type allylations of noncon-

jugated aldehydes and ketones (GP1): Strictly deoxygenated THF

(20 mL) was added to a mixture of [TiCl₂Cp₂] (2.2 mmol) and Mn dust

(8 mmol) under an Ar atmosphere and the suspension was stirred at

room temperature until it turned lime green (after about 15 min). Then,

a solution of aldehyde (1 mmol) in THF (1 mL) was added. Subsenquent-

ly, a solution of allylic halide (2 mmol) in THF (1 mL) was slowly added

over a period of 1 h and the mixture was stirred for 6 h. The reaction was

quenched with brine and extracted with EtOAc. The organic layer was

washed with brine and dried (anhyd Na₂SO₄), and the solvent was re-

moved. Products were purified by flash chromatography on silica gel

(hexane/EtOAc) and characterized by spectroscopic techniques. Yields

obtained are reported in Table 1 (products 10 and 11) and in the Support-

ing Information (products 22–24, 35–37, and 47–50).

General procedure for Ti^III-mediated Barbier-type allylations of conju-

gated aldehydes and ketones (GP2): Strictly deoxygenated THF (20 mL)

was added to a mixture of [TiCl₂Cp₂] (2.2 mmol) and Mn dust (8 mmol)

under an Ar atmosphere and the suspension was stirred at room temper-

ature until it turned lime green (after about 15 min). Then, a solution of

aldehyde (1 mmol) and allylic halide (2 mmol) in THF (1 mL) was slowly

added over a period of 1 h and the mixture was stirred for 6 h. The reac-

tion was quenched with brine and extracted with EtOAc. The organic

layer was washed with brine and dried (anhyd Na₂SO₄), and the solvent

was removed. Products were purified by flash chromatography on silica

gel (hexane/EtOAc) and characterized by spectroscopic techniques.

Yields obtained are reported in Table 1 (products 12–15) and in the Sup-

porting Information (products 25–27, 38–41, and 51).

General procedure for Ti^III-catalyzed Barbier-type allylations of noncon-

jugated aldehydes and ketones (GP3): Strictly deoxygenated THF

(20 mL) was added to a mixture of [TiCl₂Cp₂] (0.2 mmol) and Mn dust

(8 mmol) under an Ar atmosphere and the suspension was stirred at

room temperature until it turned lime green (after about 15 min). Then,

a solution of aldehyde (1 mmol) and 2,4,6-collidine (7 mmol) in THF

(1 mL), and Me₃SiCl (4 mmol) were added. Subsequently, allylic halide

(2 mmol) in THF (1 mL) was slowly added over a period of 1 h and the

mixture was stirred for 6 h. The reaction was then quenched with saturat-

ed solution of KHSO₄and extracted with EtOAc. The organic layer was

washed with brine and dried (anhyd Na₂SO₄), and the solvent was re-

moved. Products 10–11, 22–24, 35–37, and 47–50 were purified by flash

chromatography on silica gel (hexane/EtOAc) and characterized by spec-

troscopic techniques. Yields obtained are reported in Scheme 3 and

Tables 2–4. In some experiments, trimethylsilyl derivatives were ob-

served. In these cases, the residue was solved in THF (20 mL) and stirred

with Bu₄NF (10 mmol) for 2 h. The mixture was then diluted with

EtOAc, washed with brine, and dried (anhyd Na₂SO₄), and the solvent

was removed.

General procedure for Ti^III-catalyzed Barbier-type allylations of conju-

gated aldehydes and ketones (GP4): Strictly deoxygenated THF (20 mL)

was added to a mixture of [TiCl₂Cp₂] (0.2 mmol) and Mn dust (8.0 mmol)

under an Ar atmosphere and the suspension was stirred at room temper-

ature until it turned lime green (after about 15 min). Then, a solution of

aldehyde (1 mmol), 2,4,6-collidine (7 mmol), and allylic halide (2 mmol)

in THF (2 mL), and Me₃SiCl (4 mmol) were slowly added and the solu-

tion was stirred for 6 h. The reaction was then quenched with saturated

a

solution of decanal (100 mg, 0.64 mmol), 2-decanone (100 mg,

0.64 mmol), and 2,4,6-collidine (621 mg, 5.13 mmol) in THF (2 mL), and

Me₃SiCl (279 mg, 2.56 mmol) were added. Subsequently, allyl chloride

(49 mg, 0.64 mmol) in THF (1 mL) was slowly added over a period of 1 h

and the solution was stirred for 6 h. The reaction was then quenched with

saturated solution of KHSO₄and extracted with EtOAc. The organic

layer was washed with brine and dried (anhyd Na₂SO₄), and the solvent

was removed. The residue was purified by flash chromatography

(hexane/EtOAc, 9:1) to yield 10 (97 mg, 72%), 11 (trace) and the starting

materials 2 (19 mg, 19%) and 3 (95 mg, 95%).

Preparation of compound 14: Following GP2 and GP4, compound 14

was obtained as a colorless oil; ¹H NMR (400 MHz, CDCl₃): d=5.84

(ddd, J=19.2, 13.6, 9.6 Hz, 2H), 5.14 (m, 3H), 4.41 (q, J=8.4 Hz, 1H),

2.31 (brt, J=9.2 Hz, 2H), 2.13 (brt, J=8.8 Hz, 2H), 2.06 (m, 2H), 1.71

(s, 6H), 1.63 ppm (s, 3H); ¹³C NMR (100 MHz, CDCl₃; DEPT): d=137.7

(C), 133.5 (CH), 130.6 (C), 126.0 (CH), 122.9 (CH), 116.8 (CH₂), 66.7

(CH), 41.2 (CH₂), 38.5 (CH₂), 25.34 (CH₂), 24.7 (CH₃), 16.7 (CH₃),

15.6 ppm (CH₃); HRMS FAB: m/z calcd for C₁₃H₂₂ONa [M+Na]⁺:

217.1724; found: 217.1567.

Preparation of compound 26: Following GP2 and GP4, compound 26

was obtained as a pale yellow oil; ¹H NMR (400 MHz, CDCl₃): d=5.58

(ddt, J=16.4, 10.6, 6.8 Hz, 1H), 5.15 (d, J=16.4 Hz, 1H), d 5.12 (d, J=

10.3 Hz, 1H), 4.30 (dd, J=10.6, 3.2 Hz, 1H), 2.67–2.59 (m, 1H), 2.32–

2.26 (m, 1H), 1.96–1.92 (m, 2H), 1.87 (s, 3H), 1.59–1.52 (m, 2H), 1.47–

1.39 (m, 2H), 1.16 (s, 3H), 1.00 ppm (s, 3H); ¹³C NMR (100 MHz,

CDCl₃; DEPT): d=138.3 (C), 135.4 (CH), 130.6 (C), 116.2 (CH₂), 69.2

(CH), 40.4 (CH₂), 38.9 (CH₂), 33.6 (C), 33.0 (CH₂), 27.6 (CH₃), 27.0

(CH₃), 20.1 (CH₃), 18.3 ppm (CH₂); HRMS FAB: m/z calcd for

C₁₃H₂₂ONa [M+Na]⁺: 217.3058; found: 217.0932.

Preparation of compound 35: Following GP1 and GP3, compound 35

was obtained as a colorless oil; ¹H NMR (400 MHz, CDCl₃): d=5.90

(ddd, J=20, 14, 10 Hz, 2H), 5.16 (m, 2H), 5.12 (m, 1H), 2.73 (brd, J=

9.6 Hz, 2H), 2.08 (m, 6H), 1.71 (s, 3H), 1.65 (s, 3H), 1.63 (s, 3H), 1.53

(m, 2H), 1.21 ppm (s, 3H); ¹³C NMR (100 MHz, CDCl₃; DEPT): d=

135.6 (C), 134.3 (CH), 131.6 (C), 124.5 (CH), 124.5 (CH), 118.7 (CH₂),

72.4 (C), 46.6 (CH₂), 41.8 (CH₂), 39.9 (CH₂), 26.9 (CH₃), 25.9 (CH₃), 22.7

(CH₂), 17.9 (CH₃), 16.2 ppm (CH₃), (one carbon signal was not ob-

served); HRMS FAB: m/z calcd for C₁₆H₂₈ONa [M+Na]⁺: 259.3861;

found: 259.2132.

Preparation of compound 36: Following GP1 and GP3, compound 36

was obtained as a colorless oil; ¹H NMR (400 MHz, CDCl₃): d=5.88

(ddt, J=17.3, 10.3, 7.7 Hz, 1H), 5.13 (brd, J=10.3 Hz, 1H), 5.10 (brd,

J=17.3 Hz, 1H), 2.17 (d, J=7.7 Hz, 2H), 1.80–1.40 ppm (m, 18H);

¹³C NMR (100 MHz, CDCl₃; DEPT): d=134.0 (CH), 118.7 (CH₂), 75.6

(C), 34.1 (CH₂), 26.8 (CH₂), 26.4 (CH₂), 23.7 (CH₂), 23.5 (CH₂), 21.2 ppm

(CH₂); HRMS FAB: m/z calcd for C₁₃H₂₄ONa [M+Na]⁺: 219.1827;

found: 219.1833.

Preparation of compound 40: Following GP2 and GP4, compound 40

was obtained as a pale yellow oil; ¹H NMR (400 MHz, CDCl₃): d=7.51

(d, J=7.7 Hz, 2H), 7.38 (t, J=7.7 Hz, 2H), 7.28 (t, J=7.7 Hz, 1H), 5.71

(ddt, J=13.7, 8.1, 6.7 Hz, 1H), 5.18 (d, J=13.7 Hz, 1H), 5.14 (d, J=

8.1 Hz, 1H), 2.82 (dd, J=13.8, 6.7 Hz, 1H), 2.64 (dd, J=13.7, 8.1 Hz,

1H), 1.32 (quint, J=7.1 Hz, 1H), 0.52 (m, 2H), 0.37 ppm (dt, J=10.9 Hz,

Chem. Eur. J. 2009, 15, 2774 – 2791

ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

2785

J. E. Oltra, J. M. Cuerva et al.

7.3 Hz, 2H); ¹³C NMR (100 MHz, CDCl₃; DEPT): d=146.3 (C), 133.8

(CH), 128.0 (CH), 126.7 (CH), 125.6 (CH), 119.4 (CH₂), 47.3 (CH₂), 21.7

(CH), 1.6 (CH₂), 0.6 ppm (CH₂) (one carbon signal was not observed);

IR (film) n_max=3472, 3007 cm^À1; HRMS EI: m/z calcd for C₁₃H₁₅O

[MÀH]⁺: 187.1123; found: 187.1122.

8.4 Hz, 2H). This compound is quite unstable and we could obtain nei-

ther ¹³C NMR nor HRMS data.

General procedure for Barbier-type cyclizations promoted by [TiClCp₂]

(GP6): Strictly deoxygenated THF (20 mL) was added to a mixture of

[TiCl₂Cp₂] (2.2 mmol) and Mn dust (8 mmol) under an Ar atmosphere

and the suspension was stirred at room temperature until it turned lime

green (after about 15 min). Then, a solution of substrate 58 (1 mmol) in

THF (1 mL) was slowly added over a period of 15 min and the mixture

was stirred for 6 h. The reaction was quenched with brine and extracted

with EtOAc. The organic layer was washed with brine and dried (anhyd

Na₂SO₄), and the solvent was removed. Product 62 (73%) was purified

by flash chromatography on silica gel (hexane/EtOAc 8:2) and character-

ized by spectroscopic techniques.

Preparation of compound 50 (3:2 mixture of diastereomers): Following

GP1 and GP3, compound 50 was obtained as a colorless oil; ¹H NMR

(400 MHz, CDCl₃): d=6.03–5.86 (m, 2H), 5.21–5.10 (m, 4H), 4.68–4.59

(m, 2H), 4.54 (d, J=2.6 Hz, 1H)*, 4.39 (d, J=2.6 Hz, 1H), 4.28–4.25 (m,

4H), 3.98–3.62 (m, 3H), 2.51–2.25 (m, 4H), 1.60 (brs, 2H; both OH),

1.59 (s, 3H)*, 1.58 (s, 3H), 1.55 (s, 3H)*, 1.50 (s, 3H), 1.45 (s, 3H), 1.41

(s, 3H)*, 1.39 (s, 3H)*, 1.38 ppm (s, 3H); ¹³C NMR (100 MHz, CDCl₃;

DEPT): d=136.2 (CH)*, 135.6 (CH), 117.9 (CH₂), 116.7 (CH₂)*, 109.4

(C)*, 109.2 (C), 108.7 (C), 108.5 (C)*, 104.8 (C), 104.4 (C)*, 75.4 (CH),

72.8 (CH)*, 72.3 (CH), 71.1 (CH), 70.8 (CH)*, 70.6 (C), 70.2 (CH)*, 61.5

(CH₂), 61.2 (CH₂)*, 35.9 (CH₂), 35.7 (CH₂)*, 26.6 (CH₃), 26.5 (CH₃)*,

26.0 (CH₃), 26.0 (CH₃)*, 25.9 (CH₃)*, 25.8 (CH₃), 24.2 (CH₃), 24.0 ppm

(CH₃)* (one carbon signal was not observed) (signals with asterisk corre-

spond to the minor diastereomer); HRMS FAB: m/z calcd for

C₁₅H₂₄O₆Na [M+Na]⁺: 323.3386; found: 323.3382.

General procedure for Barbier-type cyclizations catalyzed by [TiClCp₂]

(GP7): Strictly deoxygenated THF (20 mL) was added to a mixture of

[TiCl₂Cp₂] (0.2 mmol) and Mn dust (8 mmol) under an Ar atmosphere

and the suspension was stirred at room temperature until it turned lime

green (after about 15 min). Then, 2,4,6-collidine (7 mmol), and Me₃SiCl

(4 mmol) were added. Finally, a solution of the corresponding substrate

(54, 55, and 58–61; 1 mmol) in THF (1 mL) was slowly added over a

period of 15 min and the mixture was stirred for 6 h. The reaction was

quenched with brine and extracted with EtOAc. The organic layer was

washed with brine and dried (anhyd Na₂SO₄), and the solvent was re-

moved. Products 56, and 62–67 were purified by flash chromatography

on silica gel (hexane/EtOAc) and characterized by spectroscopic tech-

niques. Yields obtained are reported in Table 5 and the body text.

General procedure for the synthesis of allylic bromides 54, 55, and 58–61

(GP5): PBr₃(2 mmol) was added to a solution of the corresponding allyl-

ic alcohol (1 mmol; for the preparation of these allylic alcohols see Sup-

porting Information) in Et₂O (20 mL) at 08C. The mixture was stirred at

room temperature for 4 h. Then, the mixture was diluted with Et₂O,

washed with brine, and dried (anhyd Na₂SO₄) and the solvent was re-

moved. The residue was used in the next step without further purifica-

tion.

Preparation of compound 56: Following GP7, compound 56 was ob-

tained as a colorless oil; ¹H NMR (400 MHz, CDCl₃): d=5.91 (ddd, J=

17.1, 10.2, 7.2 Hz, 1H), 5.21 (dd, J=10.2, 0.9 Hz, 1H), 5.15 (dd, J=17.1,

0.9 Hz, 1H), 3.78 (s, 3H), 3.73 (s, 3H), 2.40–1.98 (m, 5H), 1.73 (dt, J=

14.1, 3.3 Hz, 1H), 1.54 (td, J=14.1, 4.5 Hz, 1H), 1.20 ppm (s, 3H); NOE-

diff. experiment: proton irradiated, (NOEs observed): H-7, (H₂–8, H₃–9);

¹³C NMR (75 MHz, CDCl₃; DEPT): d=172.7 (C), 171.8 (C), 137.9 (CH),

117.6 (CH₂), 69.4 (C), 54.9 (C), 52.9 (CH₃), 52.7 (CH₃), 46.7 (CH), 36.3

(CH₂), 31.8 (CH₂), 29.4 (CH₃), 26.6 ppm (CH₂); HRMS ES: m/z calcd for

C₁₃H₂₀O₅Na [M+Na]⁺: 279.1202; found: 279.1192; the relative configura-

tion of 56 was established on the basis of NOE-diff. experiments per-

formed on its acetyl derivative 57.

Preparation of allylic bromide 54: Following GP5, compound 54 was ob-

tained as a colorless oil (48 mg, 70%); ¹H NMR (400 MHz, CDCl₃): d=

5.75–5.70 (m, 1H), 5.39–5.33 (m, 1H), 3.99 (d, J=8.4 Hz, 2H), 3.70 (s,

6H), 2.65 (d, J=8 Hz, 2H), 2.44 (t, J=7.3 Hz, 2H), 2.12 (t, J=8.4 Hz,

2H), 2.11 ppm (s, 3H). This compound is quite unstable and we could

obtain neither ¹³C NMR nor HRMS data.

Preparation of allylic bromide 55: Following GP5, compound 55 was ob-

1

tained as a colorless oil (226 mg, 75%); H NMR (500 MHz, CDCl₃): d=

5.70 (dt, J=15.0, 5.7 Hz, 1H), 5.52 (dt, J=15.0, 5.7 Hz, 1H), 3.91 (d, J=

6.6 Hz, 2H), 3.70 (s, 6H), 2.62 (d, J=7.2 Hz, 2H), 2.44 (t, J=8.0 Hz,

2H), 2.12 (s, 3H), 2.11 ppm (t, J=7.5 Hz, 2H). This compound is quite

unstable and we could obtain neither ¹³C NMR nor HRMS data.

Preparation of compound 57: Treatment of 56 with acetyl chloride in the

presence of 4-dimethylaminopyridine (DMAP) generated compound 57

as a colorless oil; ¹H NMR (400 MHz, CDCl₃): d=5.83–5.68 (m, 1H),

5.22–5.10 (m, 2H), 3.77 (s, 3H), 3.70 (s, 3H), 3.02 (dt, J=14.8, 3.2 Hz,

1H), 2.29 (s, 3H), 2.26–2.19 (m, 2H), 2.15 (dt, J=11.2, 3.2 Hz, 1H), 2.03–

1.93 (m, 1H), 1.86–1.77 (m, 1H), 1.54 (s, 3H), 1.41 ppm (td, J=14.8,

3.6 Hz, 1H); NOE-diff. experiment: proton irradiated, (NOEs observed):

H-6b, (H₃–9, H-2, H-6a), H₃–9, (H-2, H-6a, H-6b) , H-6a, (H-6b, H₃–9);

¹³C NMR (125 MHz, CDCl₃; DEPT): d=172.3 (C), 171.6 (C), 167.3 (C),

137.0 (CH), 118.1 (CH₂), 83.2 (C), 54.4 (C), 52.9 (CH₃), 52.9 (CH₃), 49.2

(CH), 32.7 (CH₂), 31.3 (CH₂), 26.7 (CH₂), 25.6 (CH₃), 24.2 ppm (CH₃).

Preparation of allylic bromide 58: Following GP5, compound 58 was ob-

1

tained as a colorless oil (126 mg, 82%); H NMR (400 MHz, CDCl₃): d=

8.01 (brd, J=7.5 Hz, 4H) 7.72 (brt, J=7.5 Hz, 2H), 7.60 (t, J=7.5 Hz,

4H), 5.86 (m, 1H), 5.73 (m, 1H), 3.83 (d, J=8.3 Hz, 2H), 2.98 (dd, J=

6.4, 1.6 Hz, 2H), 2.92 (t, J=7.6 Hz, 2H), 2.50 (brt, J=7.6 Hz, 2H),

2.14 ppm (s, 3H); ¹³C NMR (100 MHz, CDCl₃; DEPT): d=206.1 (C),

136.7 (C), 135.1 (CH), 131.6 (CH), 129.4 (CH), 129.1 (CH), 126.4 (CH),

89.4 (C), 38.0 (CH₂), 30.2 (CH₃), 28.1 (CH₂), 25.8 (CH₂), 23.8 ppm (CH₂).

Preparation of disulfone 62: Following GP6 and GP7, compound 62 was

obtained as a colorless oil; ¹H NMR (500 MHz, CDCl₃): d=8.10 (d, J=

7.9 Hz, 2H), 8.00 (d, J=7.9 Hz, 2H), 7.69 (m, 2H), 7.58 (m, 4H), 5.80

(m, 1H), 5.19 (d, J=10.3 Hz, 1H), 5.15 (d, J=17.4 Hz, 1H), 2.92 (m,

1H), 2.68 (dt, J=14.1, 5.1 Hz, 1H), 2.62 (t, J=12.8 Hz, 1H), 2.31 (dt, J=

13.8, 4.5 Hz, 1H), 2.20 (dd, J=14.8, 2.1 Hz, 1H), 2.08 (dt, J=14.4,

3.2 Hz, 1H), 1.72 (ddd, J=8.8, 5, 2 Hz, 1H), 1.24 ppm (s, 3H); ¹³C NMR

(125 MHz, CDCl₃; DEPT): d=139.6 (CH), 139.2 (C), 138.7 (C), 137.1

(CH), 136.9 (CH), 134.1 (CH), 133.7 (CH), 131.1 (CH), 131.0 (CH),

120.8 (CH₂), 90.3 (C), 71.2 (C), 48.0 (CH), 37.6 (CH₂), 31.3 (CH₃), 29.8

(CH₂), 24.9 ppm (CH₂).

Preparation of allylic bromide 59: Following GP5, compound 59 was ob-

tained as a colorless oil (476 mg, 80%); H NMR (500 MHz, CDCl₃): d=

1

7.61 (d, J=8.3, 2H) 7.24 (d, J=8 Hz, 2H), 5.78 (m, 1 H-cis), 5.32 (m,

1H), 3.89 (d, J=8.4 Hz, 2H), 3.84 (d, J=7 Hz, 2H), 3.25 (t, J=7 Hz,

2H), 2.76 (t, J=7.2 Hz, 2H), 2.36 (s, 3H), 2.07 ppm (s, 3H); ¹³C NMR

(100 MHz, CDCl₃; DEPT): d=206.8 (C), 143.7 (C), 136.2 (C), 129.9

(CH), 129.5 (CH), 129.3 (CH), 127.3 (CH), 45.3 (CH₂), 43.6 (CH₂), 42.9

(CH₂), 30.3 (CH₃), 25.5 (CH₂), 21.6 ppm (CH₃).

Preparation of allylic bromide 60: Following GP5, compound 60 was ob-

tained as a colorless oil (72%); ¹H NMR (500 MHz, CDCl₃): d=9.75 (s,

1H), 5.76–5.69 (m, 1H), 5.57–5.49 (m, 1H), 3.91 (d, J=6.6 Hz, 2H), 3.76

(s, 6H), 2.66 (d, J=7.2 Hz, 2H), 2.47 (t, J=7.4 Hz, 2H), 2.19 ppm (t, J=

7.3 Hz, 2H). This compound is quite unstable and we could obtain nei-

ther ¹³C NMR nor HRMS data.

Preparation of piperidine 63: Following GP7, compound 63 was obtained

as a colorless oil; ¹H NMR (500 MHz, CDCl₃): d=7.64 (d, J=8.2 Hz,

2H), 7.32 (d, J=8.2 Hz, 2H), 5.83 (m, 1H), 5.20 (m, 2H), 3.28 (dd, J=

11.7, 3.4 Hz, 1H), 3.17 (m, 1H), 3.03 (m, 1H), 2.90 (dd, J=11.5, 6.7 Hz,

1H), 2.43 (s, 3H), 2.26 (m, 1H), 1.79 (m, 1H), 1.63 (m, 1H), 1.08 ppm (s,

3H); ¹³C NMR (125 MHz, CDCl₃; DEPT): d=143.6 (C), 135.3 (CH),

133.5 (C), 129.8 (CH) ,127.7 (CH), 118.8 (CH₂), 69.6 (CH), 50.9 (CH),

47.3 (CH₂), 43.2 (CH₂), 36.8 (CH₂), 29.8 (CH₂), 25.1 (CH₃), 21.6 ppm

Preparation of allylic bromide 61: Following GP5, compound 61 was ob-

tained as a colorless oil (65%); ¹H NMR (400 MHz, CDCl₃): d=9.76 (s,

1H), 5.79–5.72 (m, 1H), 5.42–5.37 (m, 1H), 3.96 (d, J=8.4 Hz, 2H), 3.73

(s, 6H), 2.74 (d, J=6.8 Hz, 2H), 2.51 (t, J=7.6 Hz, 2H), 2.21 (t, J=

2786

www.chemeurj.org

ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2009, 15, 2774 – 2791

Barbier-Type Allylations

FULL PAPER

(CH₃); HRMS FAB: m/z calcd for C₁₅H₂₁NO₃SNa [M+Na]⁺: 318.1139;

found: 318.1137.

troscopic techniques. Yields obtained are reported in Scheme 9 and

Scheme 13, and in Table 6.

General procedure for Barbier-type prenylations of nonconjugated alde-

Preparation of piperidine 64: Following GP7, compound 64 was obtained

as a colorless oil; H NMR (500 MHz, CDCl₃): d=7.61 (d, J=8 Hz, 2H),

hydes catalyzed by Ti^III

ACTHNUGRTENUNG(GP9): Strictly deoxygenated THF (20 mL) was

1

added to a mixture of [TiCl₂Cp₂] (0.2 mmol) and Mn dust (8 mmol)

under an Ar atmosphere and the suspension was stirred at room temper-

ature until it turned lime green (after about 15 min). Then, a solution of

carbonyl compound (1 mmol) and 2,4,6-collidine (7 mmol) in THF

(2 mL), and Me₃SiCl (4 mmol) were added. Subsequently, prenyl halide

(2 mmol) was slowly added and the solution was stirred for 6 h. The reac-

tion was then quenched with saturated solution of NaHCO₃and extract-

ed with EtOAc. The organic layer was washed with brine and dried

(anhyd Na₂SO₄), and the solvent was removed. Products 76, 77, 82, and

83 were purified by flash chromatography on silica gel (hexane/EtOAc)

and characterized by spectroscopic techniques. Yields obtained are re-

ported in Table 7. In some experiments, trimethylsilyl derivatives were

observed. In these cases, the residue was solved in THF (20 mL) and

stirred with Bu₄NF (10 mmol) for 2 h. The mixture was then diluted with

EtOAc, washed with brine and dried (anhyd Na₂SO₄), and the solvent

was removed, thus obtaining the corresponding homoallylic alcohols.

7.28 (d, J=8 Hz, 2H), 5.70 (m, 1H), 5.18 (dd, J=10.5, 1.3 Hz, 1H), 5.12

(d, J=17.4 Hz, 1H), 3.53 (td, J=13.4, 2.1 Hz, 1H), 3.48 (ddd, J=11.3,

4.3, 1.7 Hz, 1H), 2.61 (dt, J=11.8, 3.24 Hz, 1H), 2.48 (t, J=11.4 Hz, 1H),

2.40 (s, 3H), 2.26 (ddd, J=12, 8.5, 4.4 Hz, 1H), 1.67 (m, 2H), 1.15 ppm

(s, 3H); ¹³C NMR (125 MHz, CDCl₃; DEPT): d=143.7 (C), 134.7 (CH),

133.6 (C), 129.9 (CH), 127.8 (CH), 119.4 (CH₂), 116.9 (CH), 68.3 (CH),

49.6 (CH), 46.3 (CH₂), 42.3 (CH₂), 38.6 (CH₂), 28.7 (CH₃), 21.7 ppm

(CH₃) (one carbon signal was not observed); HRMS FAB: m/z calcd for

C₁₅H₂₁NO₃SNa [M+Na]⁺: 318.1139; found: 318.1137.

Preparation of disulfone 67: Following GP7, compound 67 was obtained

as a colorless oil; ¹H NMR (400 MHz, CDCl₃): d=8.10 (d, J=7.9 Hz,

2H), 8.00 (d, J=7.9 Hz, 2H), 7.72 (m, 2H), 7.58 (m, 4H), 5.75 (m, 1H),

5.25 (d, J=10.3 Hz, 1H), 5.20 (d, J=17.4 Hz, 1H), 2.98 (m, 1H), 2.68

(dt, J=14.1, 5.1 Hz, 1H), 2.50–2.25 (m, 3H), 1.80–1.50 (m, 1H), 1.30–

1.20 (m, 2H), 1.10 ppm (s, 3H).

Ti-promoted Barbier-type cyclizations of model allylic bromide 55 in the

presence of Lewis acids: Strictly deoxygenated THF (20 mL) was added

to a mixture of [TiCl₂Cp₂] (2 mmol) and Mn dust (8 mmol) under an Ar

atmosphere and the suspension was stirred at room temperature until it

turned lime green (after about 15 min). Then, a solution of substrate 55

General procedure for Barbier-type prenylations of conjugated aldehydes

promoted by Ti^III

ACTHNUGRTENUNG(GP10): Strictly deoxygenated THF (20 mL) was

added to a mixture of [TiCl₂Cp₂] (2.2 mmol) and Mn dust (8 mmol)

under an Ar atmosphere and the suspension was stirred at room temper-

ature until it turned lime green (after about 15 min). Then, a solution of

carbonyl compound (1 mmol) and prenyl halide (2 mmol) in THF (2 mL)

was slowly added and the solution was stirred for 6 h. The reaction was

then quenched with saturated solution of NaHCO₃and extracted with

EtOAc. The organic layer was washed with brine and dried (anhyd

Na₂SO₄), and the solvent was removed. Products 84–88 were purified by

flash chromatography on silica gel (hexane/EtOAc) and characterized by

spectroscopic techniques. Yields obtained are reported in Table 6.

(1 mmol) and the Lewis acid ([ErACHTNUTRGNEUNG(CF₃O₃S)₃] or BF₃etherate; 2 mmol) in

THF (1 mL) was added dropwise and the mixture was stirred for 6 h.

The reaction was quenched with brine and extracted with EtOAc. The or-

ganic layer was washed with brine and dried (anhyd Na₂SO₄), and the

solvent was removed. Products 56 and 73 were purified by flash chroma-

tography on silica gel (hexane/EtOAc 8:2) and characterized by spectro-

scopic techniques. Yields obtained are reported in the body text.

Barbier-type cyclizations of model allylic bromide 55 catalyzed by

[TiCl₂Cp]: Strictly deoxygenated THF (20 mL) was added to a mixture

of commercial [TiCl₃Cp] (0.2 mmol) and Mn dust (8 mmol) under an Ar

atmosphere and the suspension was stirred at room temperature until it

turned lime green (after about 15 min). Then, 2,4,6-collidine (7 mmol),

and Me₃SiCl (4 mmol) were added. Finally, a solution of 55 (1 mmol) in

THF (1 mL) was slowly added over a period of 15 min and the mixture

was stirred for 6 h. The reaction was quenched with brine and extracted

with EtOAc. The organic layer was washed with brine and dried (anhyd

Na₂SO₄), and the solvent was removed. Product 56 (64% yield) was puri-

fied by flash chromatography on silica gel (hexane/EtOAc 8:2).

General procedure for Barbier-type prenylations of conjugated aldehydes

catalyzed by Ti^III

ACHTNUTRGNEUNG(GP11): Strictly deoxygenated THF (20 mL) was added

to a mixture of [TiCl₂Cp₂] (0.2 mmol) and Mn dust (8 mmol) under an

Ar atmosphere and the suspension was stirred at room temperature until

it turned lime green (after about 15 min). Then, 2,4,6-collidine (7 mmol)

and Me₃SiCl (4 mmol) were added. Subsequently, a solution of carbonyl

compound (1 mmol) and prenyl halide (2 mmol) in THF (2 mL) was

slowly added over a period of 1 h and the solution was stirred for 6 h.

The reaction was then quenched with saturated solution of NaHCO₃and

extracted with EtOAc. The organic layer was washed with brine and

dried (anhyd Na₂SO₄), and the solvent was removed. Products 84 and 88

were purified by flash chromatography on silica gel (hexane/EtOAc) and

characterized by spectroscopic techniques. Yields obtained are reported

in Table 7. In some experiments, trimethylsilyl derivatives were observed.

In these cases, the residue was solved in THF (20 mL) and stirred with

Bu₄NF (10 mmol) for 2 h. The mixture was then diluted with EtOAc,

washed with brine and dried (anhyd Na₂SO₄), and the solvent was re-

moved, thus obtaining the corresponding homoallylic alcohols.

Barbier-type cyclizations of model allylic bromide 55 catalyzed by the

Brintzingerꢃs complex 52: Strictly deoxygenated THF (20 mL) was added

to a mixture of commercial Brintzingerꢆs complex (52; 0.2 mmol) and Mn

dust (8 mmol) under an Ar atmosphere and the suspension was stirred at

room temperature until it turned lime green (after about 15 min). Then,

2,4,6-collidine (7 mmol), and Me₃SiCl (4 mmol) were added. Finally, a so-

lution of 55 (1 mmol) in THF (1 mL) was slowly added over a period of

15 min and the mixture was stirred for 6 h. The reaction was quenched

with brine and extracted with EtOAc. The organic layer was washed with

brine and dried (anhyd Na₂SO₄), and the solvent was removed. The resi-

due was submitted to flash chromatography on silica gel (hexane/EtOAc

8:2) yielding products (À)-56 (36% yield, 20% ee) and (À)-73 (29%

yield, 19% ee).^[47]

Ti-catalyzed chemospecific coupling of prenyl bromide (74) with 2,5-di-

methoxybenzaldehyde (79) in the presence of decanal (2): Strictly deoxy-

genated THF (20 mL) was added to a mixture of [TiCl₂Cp₂] (299 mg,

1.2 mmol) and Mn dust (264 mg, 4.81 mmol) under an Ar atmosphere

and the suspension was stirred at RT until it turned lime green (after

about 15 min). Then, a solution of 2 (94 mg, 0.60 mmol), 79 (100 mg,

0.60 mmol), and 74 (90 mg, 0,60 mmol) in THF (2 mL) was slowly added

and the mixture was stirred for 6 h. The reaction was then quenched with

brine and extracted with EtOAc. The organic layer was dried (anhyd

Na₂SO₄) and the solvent was removed. Prenylation product 88 (82%

yield) and unchanged decanal were isolated by flash chromatography

(hexane/EtOAc, 8:2).

General procedure for Barbier-type prenylations of nonconjugated alde-

hydes and ketones promoted by Ti^III

ACTHNUTRGNEU(GN GP8): Strictly deoxygenated THF

(20 mL) was added to a mixture of [TiCl₂Cp₂] (2.2 mmol) and Mn dust

(8 mmol) under an Ar atmosphere and the suspension was stirred at

room temperature until it turned lime green (after about 15 min). Then,

a solution of carbonyl compound (1 mmol) in THF (1 mL) was added.

Subsequently, prenyl halide (2 mmol) in THF (1 mL) was slowly added

and the solution was stirred for 6 h. The reaction was then quenched with

saturated solution of NaHCO₃and extracted with EtOAc. The organic

layer was washed with brine and dried (anhyd Na₂SO₄), and the solvent

was removed. Products 76, 77, and 80–83 and 89 were purified by flash

chromatography on silica gel (hexane/EtOAc) and characterized by spec-

Preparation of compound 76: Following GP8 and GP9, compound 76

was obtained as a pale yellow oil; ¹H NMR (500 MHz, CDCl₃): d=5.17

(t, J=6.8 Hz, 1H), 3.60 (m, 1H), 2.15 (m, 2H), 1.75 (s, 3H), 1.65 (s, 3H),

1.26 (brs, 16H), 0.89 ppm (t, J=6.8 Hz, 3H); ¹³C NMR (125 MHz,

CDCl₃; DEPT): d=135.2 (C), 120.3 (CH), 71.8 (CH), 36.9 (CH₂), 36.3

(CH₂), 32.0 (CH₂), 29.7 (CH₂), 29.4 (CH₂), 29.3 (CH₂), 26.0 (CH₃), 25.9

Chem. Eur. J. 2009, 15, 2774 – 2791

ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

2787

J. E. Oltra, J. M. Cuerva et al.

(CH₂), 22.8 (CH₂), 18.1 (CH₃), 14.2 ppm (CH₃); HRMS FAB: m/z calcd

for C₁₅H₃₀ONa [M+Na]⁺: 249.2194; found: 249.2196.

¹³C NMR (100 MHz, CDCl₃; DEPT): d=149.9 (C), 149.8 (C), 139.4 (C),

139.2 (C), 134.97 (C), 134.8 (C), 122.7 (CH), 121.7 (CH), 120.3 (CH),

120.1 (CH), 108.6 (CH₂), 75.7 (CH), 75.5 (CH), 41.4 (CH), 41.2 (CH),

34.4 (CH₂), 34.2 (CH₂), 30.6 (CH₂), 30.4 (CH₂), 27.7 (CH₂), 27.5 (CH₂),

26.0 (CH₃), 25.9 (CH₃), 24.7 (CH₂), 24.1 (CH₂), 20.9 (CH₃), 20.8 (CH₃),

18.1 ppm (CH₃) (two carbon signals were not observed); HRMS FAB:

m/z calcd for C₁₅H₂₄ONa [M+Na]⁺: 243.1724; found: 243.1724.

Preparation of compound 80 (9:1 mixture of diastereomers): Following

GP8, compound 80 was obtained as a colorless oil. Only signals for the

major isomer are described; ¹H NMR (CDCl₃, 300 MHz): d=7.22 (m,

5H), 5.07 (t, J=7.2 Hz, 1H), 3.60 (m, 1H), 2.70 (quint, J=7.2 Hz, 1H),

1.97 (m, 2H), 1.64 (s, 3H), 1.50 (s, 3H), 1.27 ppm (d, J=6.8 Hz, 3H);

¹³C NMR (125 MHz, CDCl₃; DEPT): d=144.7 (C), 135.2 (C), 128.4

(CH), 127.7 (CH), 126.3 (CH), 120.2 (CH), 76.1 (CH), 45.3 (CH), 33.8

(CH₂), 25.9 (CH₃), 17.9 (CH₃), 16.4 ppm (CH₃); MS (70 eV): m/z (%):

204 (5) [M]⁺, 186 (5), 171 (4), 143 (8), 135 (47), 117 (48), 106 (93), 91

(94), 81 (84), 70 (100).

Preparation of compound 88: Following GP10 and GP11, compound 88

was obtained as a colorless oil; ¹H NMR (CDCl₃, 300 MHz): d=6.95 (d,

J=2.8 Hz, 1H), 6.75 (m, 2H), 5.22 (t, J=7.4 Hz, 1H), 4.87 (m, 1H), 3.79

(s, 3H), 3.76 (s, 3H), 2.61 (d, J=5.8 Hz, 1H), 2.42 (m, 2H), 1.71 (s, 3H),

1.60 ppm (s, 3H); ¹³C NMR (CDCl₃, 100 MHz; DEPT): d=153.8 (C),

150.6 (C), 134.7 (C), 133.6 (C), 120.5 (CH), 112.9 (CH), 112.4 (CH),

111.4 (CH), 70.3 (CH), 36.4 (CH₂), 36.3 (CH₂), 25.9 (CH₃), 17.9 ppm

(CH₃); HRMS FAB: m/z calcd for C₁₄H₂₀O₃Na [M+Na]⁺: 259.1412;

found: 259.1407.

Preparation of compound 82 (7:3 mixture of diastereomers): Following

GP8 and GP9, compound 82 was obtained as a colorless oil; ¹H NMR

(CDCl₃, 400 MHz): d=5.16 (dt, J=6.8, 1.2 Hz, 1H), 5.09 (t, J=6.8 Hz,

1H), 3.69 (m, 1H), 2.20–1.90 (m, 4H), 1.73 (s, 3H), 1.67 (s, 3H), 1.64 (s,

3H), 1.60 (s, 3H), 1.50–1.06 (m, 4H), 0.91 (d, J=6.4 Hz, 3H) 0.90 ppm

(d, J=6.4 Hz, 3H)*; ¹³C NMR (125 MHz, CDCl₃; DEPT): d=135.3 (C)*,

131.3 (C), 124.9 (CH), 120.3 (CH)*, 120.3 (CH), 69.9 (CH), 69.5 (CH)*,

44.5 (CH₂), 44.4 (CH₂)*, 38.0 (CH₂)*, 37.1 (CH₂)*, 36.9 (CH₂), 36.6

(CH₂), 29.5 (CH₃), 29.2 (CH₃)*, 26.0 (CH₃), 25.8 (CH₃), 25.6 (CH₂)*, 25.5

(CH₂), 20.4 (CH₃), 19.3 (CH₃)*, 18.1 (CH₃)*, 17.7 ppm (CH₃) (signals

with asterisk correspond to the minor diastereomer); HRMS FAB: m/z

calcd for C₁₅H₂₇O [MÀH]⁺: 223.2061; found: 223.2065.

Preparation of compound 83 (1:1 mixture of diastereomers): Following

GP8 and GP9, compound 83 was obtained as a colorless oil; ¹H NMR

(CDCl₃, 300 MHz): d=5.81 (dd, J=10.8, 2 Hz, 1H), 5.09 (m, 2H), 3.34

(d, J=10.8 Hz, 1H), 2.10–1.86 (m, 2H), 1.68 (s, 3H), 1.60 (s, 3H), 1.50–

1.10 (m, 4H), 1.00 (s, 6H), 0.94 (d, J=6.8 Hz, 3H), 0.88 ppm (d, J=

6.8 Hz, 3H); ¹³C NMR (125 MHz, CDCl₃; DEPT): d=145.5 (CH), 131.1

(C), 124.9 (CH), 124.8 (CH), 113.3 (CH₂), 76.0 (CH), 75.6 (CH), 41.7

(C), 41.5 (C), 39.0 (CH₂), 38.6 (CH₂), 38.4 (CH₂), 35.7 (CH₂), 29.8 (CH₃),

29.2 (CH₃), 25.7 (CH₂), 25.6 (CH₂), 25.3 (CH₂), 23.1 (CH₃), 23.0 (CH₃),

22.0 (CH₃), 21.9 (CH₃), 20.8 (CH₃), 18.8 (CH₃), 17.6 ppm (CH₃) (some

carbon signals were not observed); HRMS FAB: m/z calcd for C₁₅H₂₇O

[MÀH]⁺: 223.2061; found: 223.2065.

Barbier type a-prenylation of conjugated aldehyde 79 catalyzed by the

Brintzingerꢃs complex 52: Strictly deoxygenated THF (20 mL) was added

to

a mixture of commercial complex 52 (0.2 mmol) and Mn dust

(8 mmol) under an Ar atmosphere and the suspension was stirred at

room temperature until it turned lime green (after about 15 min). Then,

2,4,6-collidine (7 mmol) and Me₃SiCl (4 mmol) were added. Subsequent-

ly, a solution of 79 (1 mmol) and 74 (2 mmol) in THF (2 mL) was slowly

added over a period of 1 h and the solution was stirred for 6 h. The reac-

tion was then quenched with saturated solution of NaHCO₃and extract-

ed with EtOAc. The organic layer was washed with brine and dried

(anhyd Na₂SO₄), and the solvent was removed. The residue was solved in

THF (20 mL) and stirred with Bu₄NF (10 mmol) for 2 h. The mixture

was then diluted with EtOAc, washed with brine, and dried (anhyd

Na₂SO₄), and the solvent was removed. The residue was submitted to

flash chromatography on silica gel (hexane/EtOAc, 8:2) yielding product

(À)-88 (21% yield, 29% ee).^[47]

Preparation of aldehyde 93: A sample of MnO₂(188 mg, 2.16 mmol) was

added to

a solution of (E)-4-hydroxy-3-methylbut-2-enyl benzoate

(150 mg, 0.72 mmol) in CH₂Cl₂(50 mL), and the mixture was stirred for

8 h at RT. Then, the solution was filtered and the solvent was removed.

The residue was purified by flash chromatography (hexane/EtOAc, 8:2)

Preparation of compound 84: Following GP10 and GP11, compound 84

was obtained as a pale yellow oil; ¹H NMR (CDCl₃, 400 MHz): d=5.21

(d, J=8.5, 1H), 5.14 (m, 2H), 4.36 (ddd, J=14.1, 7.6, 1.3 Hz, 1H), 2.27

(m, 1H), 2.12 (m, 5H), 1.72 (s, 3H), 1.68 (s, 3H), 1.67 (s, 3H), 1.64 (s,

3H), 1.60 ppm (s, 3H); ¹³C NMR (CDCl₃, 100 MHz; DEPT): d=145.3

(C), 138.5 (C), 134.9 (C), 127.4 (CH), 124.5 (CH), 119.9 (CH), 68.6 (CH),

39.6 (CH₂), 36.6 (CH₂), 26.5 (CH₂), 26.0 (CH₃), 25.8 (CH₃), 18.1 (CH₃),

17.8 (CH₃), 16.7 ppm (CH₃); HRMS FAB: m/z calcd for C₁₅H₂₆ONa

[M+Na]⁺: 245.1881; found: 245.1876.

to yield aldehyde 93 (147 mg, 100%) as

a

colorless oil; ¹H NMR

(400 MHz, CDCl₃): d=9.47 (s, 1H), 8.06 (d, J=6.8 Hz, 2H), 7.57 (t, J=

7.6 Hz, 1H), 7.45 (t, J=8 Hz, 2H), 6.22 (dt, J=5.6, 1.2 Hz, 1H), 5.14 (d,

J=6 Hz, 2H), 1.85 ppm (s, 3H); ¹³C NMR (100 MHz, CDCl₃; DEPT):

d=194.2 (C), 166.2 (C), 145.8 (CH), 140.6 (C), 133.4 (CH), 129.7 (CH),

128.5 (CH), 61.3 (CH₂), 9.6 ppm (CH₃).

Ti-catalyzed prenylation of 93: Following the general procedure GP11,

alcohol 94 was obtained (84 mg, 63%) as a colorless oil; ¹H NMR

(400 MHz, CDCl₃): d=8.09 (dd, J=10, 1.2 Hz, 2H), 7.59 (t, J=10 Hz,

1H), 7.47 (t, J=10 Hz, 2H), 5.77 (t, J=9.2 Hz, 1H), 5.15 (t, J=9.6 Hz,

1H), 4.93 (d, J=9.2 Hz, 2H), 4.10 (t, J=8.4 Hz, 1H), 2.32 (t, J=9.2 Hz,

2H), 1.82 (s, 3H), 1.75 (s, 3H), 1.68 ppm (s, 3H); ¹³C NMR (100 MHz,

CDCl₃): d=166.6 (C), 143.1 (C), 135.4 (C), 132.9 (CH), 130.5 (C), 129.7

(CH), 128.4 (CH), 119.7 (CH), 119.5 (CH), 76.3 (CH), 61.5 (CH₂), 34.3

(CH₂), 25.9 (CH₃), 18.1 (CH₃), 12.6 ppm (CH₃); HRMS FAB: m/z calcd

for C₁₇H₂₂O₃Na [M+Na]⁺: 297.1469; found: m/z 297.1463.

Preparation of compound 85: Following GP10, compound 85 was ob-

tained as a pale yellow oil; ¹H NMR (300 MHz, CDCl₃): d=5.24 (t, J=

6.8 Hz, 1H), 4.23 (dd, J=10.6, 3.2 Hz, 1H), 2.65 (ddd, J=14.7, 9.4,

5.3 Hz, 1H), 2.14 (brd, J=4.5 Hz, 1H), 1.97–1.91 (m, 1H), 1.87 (s, 3H),

1.74 (s, 3H), 1.66 (s, 3H), 1.63–1.38 (m, 4H), 1.09 (s, 3H), 0.95 ppm (s,

3H); ¹³C NMR (75 MHz, CDCl₃; DEPT): d=139.6 (C), 134.5 (C), 131.4

(C), 121.9 (CH), 71.1 (CH), 40.0 (CH₂), 35.8 (CH₂), 34.7 (C), 34.1 (CH₂),

28.7 (CH₃), 28.1 (CH₃), 26.0 (CH₃), 21.2 (CH₃), 19.4 (CH₂), 18.1 ppm

(CH₃); IR (film) n_max=3447, 2927 cm^À1

C₁₅H₂₆O [M]⁺: 222.1984; found: 222.1986.

; HRMS EI: m/z calcd for

Synthesis of rosidirol (90): K₂CO₃(160 mg, 1.16 mmol) was added to a

solution of alcohol 94 (84 mg, 0.3 mmol) in MeOH (5 mL), and the mix-

ture was stirred for 24 h at RT. Then, AcOEt was added and the mixture

was washed with water, dried (anhyd Na₂SO₄) and the solvent was re-

moved. The crude was purified by flash chromatography (hexane/EtOAc,

1:1) to yield rosiridol (90)^[53](38 mg, 73%).

Preparation of compound 86: Following GP10, compound 86 was ob-

tained as a pale yellow oil; ¹H NMR (300 MHz, CDCl₃): d=5.43 (m,

1H), 5.09 (m, 1H), 3.97 (t, J=6.9 Hz, 1H), 2.43–2.08 (m, 8H), 1.72 (s,

3H), 1.62 (s, 3H), 1.28 (s, 3H), 0.83 ppm (s, 3H); ¹³C NMR (75 MHz,

CDCl₃; DEPT): d=149.9 (C), 134.5 (C), 120.3 (CH), 117.7 (CH), 74.4

(CH), 42.3 (CH), 41.1 (CH), 37.8 (C), 33.8 (CH₂), 31.9 (CH₂), 31.2 (CH₂),

26.2 (CH₃), 25.9 (CH₃), 21.5 (CH₃), 18.0 ppm (CH₃); MS (70 eV): m/z

(%): 202 (32) [MÀH₂O]⁺, 187 (12), 159 (30), 151 (62), 131 (46), 117 (75),

105 (76), 91 (100).

Ti-catalyzed prenylation of 95: Strictly deoxygenated THF (20 mL) was

added to a mixture of [TiCl₂Cp₂] (27 mg, 0.11 mmol) and Mn dust

(239 mg, 4.35 mmol) under an Ar atmosphere and the suspension was

stirred at room temperature until it turned lime green (after about

15 min). Then, a solution of aldehyde 95 (150 mg, 0.54 mmol), 2,4,6-

collidine (527 mg, 4.35 mmol), and 74 (162 mg, 1.09 mmol) in THF

(2 mL), and Me₃SiCl (232 mg, 2.14 mmol) were simultaneously added

and the solution was stirred for 6 h. The reaction was then quenched with

saturated solution of NaHCO₃and extracted with EtOAc. The organic

Preparation of compound 87 (1:1 mixture of diastereoisomers): Following

1

GP10, compound 87 was obtained as a colorless oil; H NMR (300 MHz,

CDCl₃): d=5.66 (brs, 1H), 5.10 (brt, J=10 Hz, 1H), 4.67 (s, 2H), 3.93 (t,

J=9 Hz, 1H), 1.73–2,24 (m, 9H), 1.71 (s, 6H), 1.62 ppm (s, 3H);

2788

www.chemeurj.org

ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2009, 15, 2774 – 2791

Barbier-Type Allylations

FULL PAPER

layer was washed with brine and dried (anhyd Na₂SO₄), and the solvent

was removed. The residue was submitted to flash chromatography

(hexane/EtOAc, 7:3) to give 96 (153 mg, 72%).

[10] Bis(cyclopentadienyl)titaniumACTHNGUTERNNU(G III) chloride, generated in situ by stir-

ring commercial [TiCl₂Cp₂] with Zn or Mn dust in THF, exists as an

equilibrium mixture of the monomer [TiClCp₂] and the dinuclear

species [(TiClCp₂)₂]; see: a) R. J. Enemærke, J. Larsen, T. Skrydstr-

up, K. Daasbjerg, J. Am. Chem. Soc. 2004, 126, 7853–7864; b) K.

Daasbjerg, H. Svith, S. Grimme, M. Gerenkamp, C. Mꢅck-Lichten-

feld, A. Gansꢄuer, A. Barchuck, F. Keller, Angew. Chem. 2006 , 118 ,

209 5 –2 098; Angew. Chem. Int. Ed. 2006 , 45 , 204 1 –2 044; c) A. Gan-

sꢄuer, A. Barchuk, F. Keller, M. Schmitt, S. Grimme, M. Geren-

kamp, C. Mꢅck-Lichtenfeld, K. Daasbjerg, H. Svith, J. Am. Chem.

Soc. 2007 , 129 , 135 9 –1 371; for clarityꢆs sake we usually represent

this complex as [TiClCp₂], except in the case in which the dimer

character of the species involved is relevant (see Scheme 7 and re-

lated discussion).

Synthesis of aldehyde 97: Dess-Martin periodinane (771 mg, 1.82 mmol)

was added to a solution of trans,trans-farnesol (200 mg, 0.909 mmol) in

CH₂Cl₂(15 mL). The suspension was stirred for 4 h at RT. Then, the sol-

vent was partially removed, Et₂O (20 mL) was added and the mixture

was washed with a 1:1 mixture of saturated solution of NaHCO₃and

10% Na₂S₂O₃and dried (anhyd Na₂SO₄), and the solvent was removed.

The residue was submitted to flash chromatography (hexane/EtOAc,

85:15) to give 97 (164 mg, 82%) as a colorless oil.

Ti-catalyzed “head to head” coupling between 97 and 98. Synthesis of

92: Strictly deoxygenated THF (20 mL) was added to a mixture of

[TiCl₂Cp₂] (23 mg, 0.09 mmol) and Mn dust (198 mg, 3.6 mmol) under an

Ar atmosphere and the suspension was stirred at room temperature until

it turned lime green (after about 15 min). Then, 2,4,6-collidine (436 mg,

3.6 mmol) in THF (1 mL) and Me₃SiCl (232 mg, 2.14 mmol) were simul-

taneously added. Subsequently, a solution of 97 (100 mg, 0.45 mmol) and

commercial 98 (217 mg, 0.9 mmol) in THF (2 mL) were slowly added and

the mixture was stirred for 6 h. The reaction was then quenched with sa-

turated NaHCO₃and extracted with EtOAc. The organic layer was

washed with brine and dried (anhyd Na₂SO₄), and the solvent was re-

moved. The residue was submitted to flash chromatography (hexane/

EtOAc, 8:2) to give 92 (145 mg, 76%) as a colorless oil.

[11] For protic titanocene-regenerating agents, see: a) A. Gansꢄuer, M.

Pierobon, H. Bluhm, Angew. Chem. 1998 , 110 , 10 7 –1 09; Angew.

Chem. Int. Ed. 1998 , 37 , 10 1 –1 03; b) A. Gansꢄuer, H. Bluhm,

Chem. Commun. 1998 , 214 3 –2 144; c) A. Gansꢄuer, H. Bluhm, M.

Pierobon, J. Am. Chem. Soc. 1998 , 120 , 1284 9 –1 2859. For aprotic

ones, see: d) A. F. Barrero, A. Rosales, J. M. Cuerva, J. E. Oltra,

Org. Lett. 2003 , 5 , 193 5 –1 938.

[12] A. Rosales, J. L. Oller-Lꢇpez, J. Justicia, A. Gansꢄuer, J. E. Oltra,

J. M. Cuerva, Chem. Commun. 2004 , 262 8 –2 629.

[13] For selected reports on epoxide openings promoted by stoichiomet-

ric [TiClCp₂], see: a) T. V. RajanBabu, W. A. Nugent, J. Am. Chem.

Soc. 1994 , 116 , 98 6 –9 97, and references therein; b) A. Fernꢃndez-

Mateos, E. Martꢂn de la Nava, G. Pascual-Coca, A. Ramos-Silvo, R.

Rubio-Gonzꢃlez, Org. Lett. 1999, 1, 607–609; c) A. F. Barrero, J. E.

Oltra, J. M. Cuerva, A. Rosales, J. Org. Chem. 2002 , 67 , 256 6 –2 571;

d) D. Leca, L. Fensterbank, E. Lacꢉte, M. Malacria, Angew. Chem.

2004 , 116 , 431 6 –4 318; Angew. Chem. Int. Ed. 2004 , 43 , 422 0 –4 222;

e) J. M. Cuerva, A. G. CampaÇa, J. Justicia, A. Rosales, J. L. Oller-

Lꢇpez, R. Robles, D. Cꢃrdenas, E. BuÇuel, J. E. Oltra, Angew.

Chem. 2006 , 118 , 564 8 –5 652; Angew. Chem. Int. Ed. 2006 , 45 , 552 2 –

5526; for selected reports on titanocene-catalyzed epoxide openings,

see reference [11] and: f) A. Gansꢄuer, T. Lauterbach, H. Bluhm,

M. Noltemeyer, Angew. Chem. 1999 , 111 , 311 2 –3 114; Angew.

Chem. Int. Ed. 1999 , 38 , 290 9 –2 910; g) A. Gansꢄuer, M. Pierobon,

H. Bluhm, Angew. Chem. 2002 , 114 , 334 1 –3 343; Angew. Chem. Int.

Ed. 2002 , 41 , 320 6 –3 208; h) A. Gansꢄuer, B. Rinker, M. Pierobon,

S. Grimme, M. Gerenkamp, C. Mꢅck-Lichtenfeld, Angew. Chem.

2003 , 115 , 381 5 –3 818; Angew. Chem. Int. Ed. 2003 , 42 , 368 7 –3 690;

i) J. Justicia, A. Rosales, E. BuÇuel, J. L. Oller-Lꢇpez, M. Valdivia,

A. Haꢈdour, J. E. Oltra, A. F. Barrero, D. J. Cꢃrdenas, J. M. Cuerva,

Chem. Eur. J. 2004 , 10 , 177 8 –1 788; j) A. Gansꢄuer, T. Lauterbach,

D. Geich-Gimbel, Chem. Eur. J. 2004 , 10 , 498 3 –4 990; k) J. Justicia,

J. E. Oltra, J. M. Cuerva, J. Org. Chem. 2004 , 69 , 580 3 –5 806; l) J.

Friedrich, M. Dolg, A. Gansꢄuer, D. Geich-Gimbel, T. Lauterbach,

J. Am. Chem. Soc. 2005 , 127 , 707 1 –7 077; m) J. Justicia, J. E. Oltra,

J. M. Cuerva, J. Org. Chem. 2005 , 70 , 826 5 –8 272; n) J. Justicia, J. L.

Oller-Lꢇpez, A. G. CampaÇa, J. E. Oltra, J. M. Cuerva, E. BuÇuel,

D. J. Cꢃrdenas, J. Am. Chem. Soc. 2005 , 127 , 1491 1 –1 4921.

Acknowledgements

We thank the Spanish Ministry of Education and Science (MEC) (proj-

ects CTQ2005-08402 and “Factorꢂa de Cristalizaciꢇn, Consolider-Ingenio-

2010”) and the “Junta de Andalucꢂa” (JA) (project P05-FQM-1111 and

aids to the group FQM339) for financial support. R.E.E. thanks MEC for

her fellowship. J.J. thanks MEC and the University of Granada for his

“Juan de la Cierva” contract. N.F. and M.P. thank JA for their fellow-

ships. D.C.L. thanks CSIC-EU for his I3P postdoctoral research contract.

We thank our English colleague Dr. J. Trout for revising our English text.

[1] P. Barbier, Compt. Rend. 1899, 128, 110–111.

[2] a) M. B. Smith, J. March, Marchꢀs Advanced Organic Chemistry, 5th

ed., Wiley, New York, 2001, p. 1210; b) M. B. Smith, Organic Synthe-

sis, 2nd ed., McGraw-Hill, New York, 2002, pp. 580–581.

[3] a) F. A. Carey, R. J. Sundberg, Advanced Organic Chemistry, Part B,

4th ed., Kluwer Academic/Plenum, New York, 2001, p. 458; b) C.

Blomberg, F. A. Hartog, Synthesis 1977 , 1 8 –3 0.

À

[4] Allylation reactions not only form C C bonds, but also incorporate

into products a new double bond that can be subsequently function-

alized.

[5] For a list of metals capable of promoting Barbier-type allylations,

see: R. C. Larock, Comprehensive Organic Transformations, 2nd

ed., Wiley-VCH, New York, 1999, pp. 1126–1133.

[14] For pinacol couplings promoted by stoichiometric [TiClCp₂], see:

a) Y. Handa, J. Inanaga, Tetrahedron Lett. 1987 , 28 , 571 7 –5 718; for

À

[6] For an excellent review of Cr-based C C bond-forming methods, in-

cluding Nozaki–Hiyama–Kishi reactions, see: a) A. Fꢅrstner, Chem.

Rev. 1999 , 99 , 99 1 –1 045; for a recent review of asymmetric Nozaki–

Hiyama–Kishi reactions, see: b) G. C. Hargaden, P. J. Guiry, Adv.

Synth. Catal. 2007 , 349 , 240 7 –2 424.

titanoceneACTHNUTRGENUG(N III)-catalyzed pinacol couplings, see: b) A. Gansꢄuer,

Chem. Commun. 1997 , 45 7 –4 58; c) A. Gansꢄuer, D. Bauer, J. Org.

Chem. 1998 , 63 , 207 0 –2 071; d) A. Gansꢄuer, D. Bauer, Eur. J. Org.

Chem. 1998 , 267 3 –2 676; e) T. Hirao, B. Hatano, M. Asahara, Y. Mu-

guruma, A. Ogawa, Tetrahedron Lett. 1998 , 39 , 524 7 –5 248; f) M. S.

Dunlap, K. M. Nicholas, J. Organomet. Chem. 2001 , 630 , 12 5 –1 31.

[15] R. E. Estꢁvez, J. L. Oller-Lꢇpez, R. Robles, C. R. Melgarejo, A.

Gansꢄuer, J. M. Cuerva, J. E. Oltra, Org. Lett. 2006, 8, 5433–5436.

[16] a) J. D. Parrish, D. R. Sheldon, R. D. Little, Org. Lett. 2003 , 5 , 361 5 –

3617; b) L. Sgreccia, M. Brandini, S. Morganti, A. Quintavalla, A.

Umani-Ronchi, P. G. Cozzi, J. Organomet. Chem. 2007 , 692 , 319 1 –

3197; c) R. E. Estꢁvez, M. Paradas, A. Millꢃn, T. Jimꢁnez, R.

Robles, J. M. Cuerva, J. E. Oltra, J. Org. Chem. 2008, 73, 1616–1619.

[7] For a review of SmI₂chemistry, including the samarium Barbier re-

action, see: a) H. B. Kagan, Tetrahedron 2003 , 59 , 1035 1 –1 0372; for

detailed mechanistic aspects of this reaction, see: b) D. P. Curran,

T. L. Fevig, C. P. Jasperse, M. J. Totleben, Synlett 1992 , 94 3 –9 61.

[8] For a recent review of Ti-based enantioselective synthesis, see: D. J.

Ramꢇn, M. Yus, Chem. Rev. 2006, 106, 2126–2208.

[9] For a recent overview of synthesis and reactivity of allyltitanium de-

rivatives, see: J. Szymoniak, C. Moꢈse in Titanium and Zirconium in

Organic Synthesis (Ed.: I. Marek), Wiley-VCH, Weinheim, 2002,

pp. 451–474.

Chem. Eur. J. 2009, 15, 2774 – 2791

ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

2789

J. E. Oltra, J. M. Cuerva et al.

[17] A. G. CampaÇa, B. Bazdi, N. Fuentes, R. Robles, J. M. Cuerva, J. E.

Oltra, S. Porcel, A. Echavarren, Angew. Chem. 2008, 120, 7625–

7629; Angew. Chem. Int. Ed. 2008, 47, 7515–7519.

for recent reviews highlighting the growing impact of asymmetric

catalysis, see: b) B. M. Trost, D. R. Fandrick, Aldrichimica Acta

2007, 40, 59–72; c) T. Ooi, K. Maruoka, Aldrichimica Acta 2007, 40,

77–86; for recent reviews on enantioselective radical processes, see:

d) M. P. Sibi, S. Manyem, J. Zimmerman, Chem. Rev. 2003 , 103 ,

326 3 –3 295; e) H. Miyabe, Y. Takemoto, Chem. Eur. J. 2007 , 13 ,

728 0 –7 286.

[18] For pertinent reviews, see: a) A. Gansꢄuer, H. Bluhm, Chem. Rev.

2000 , 100 , 277 1 –2 788; b) A. Gansꢄuer, M. Pierobon in Radicals in

Organic Synthesis, Vol. 2 (Eds.: P. Renaud, M. P. Sibi), Wiley-VCH,

Weinheim, 2001, pp. 207–220; c) A. Gansꢄuer, B. Rinker, Tetrahe-

dron 2002, 58, 7017–7026; d) A. Gansꢄuer, S. Narayan, Adv. Synth.

Catal. 2002 , 344 , 46 5 –4 75; e) A. Gansꢄuer, B. Rinker in Titanium

and Zirconium in Organic Synthesis (Ed.: I. Marek), Wiley-VCH,

Weinheim, 2002, pp. 435–450; f) A. Gansꢄuer, T. Lauterbach, S.

Narayan, Angew. Chem. 2003 , 115 , 571 4 –5 731; Angew. Chem. Int.

Ed. 2003 , 42 , 555 6 –5 573; g) J. M. Cuerva, J. Justicia, J. L. Oller-

Lꢇpez, B. Bazdi, J. E. Oltra, Mini-Rev. Org. Chem. 2006 , 3 , 2 3 –3 5;

h) J. M. Cuerva, J. Justicia, J. L. Oller-Lꢇpez, J. E. Oltra, Top. Curr.

Chem. 2006 , 264 , 6 3 –9 1; i) A. Gansꢄuer, J. Justicia, C.-A. Fan, D.

Worgull, F. Piestert, Top. Curr. Chem. 2007 , 279 , 2 5 –5 2.

[38] A. Gansꢄuer, H. Bluhm, B. Rinker, S. Narayan, M. Schick, T Lau-

terbach, M. Pierobon, Chem. Eur. J. 2003 , 9 , 53 1 –5 42.

[39] The absolute S configuration of (À)-25 was tentatively assigned by

comparing its optical rotation with those of (R)-(+)-1-phenyl-but-3-

en-1-ol, (R)-(+)-1-(3-methoxyphenyl)-but-3-en-1-ol, and (R)-(+)-1-

(4-methoxyphenyl)-but-3-en-1-ol; see: A. V. Malkov, P. Ramꢂrez-

Lꢇpez, L. Biedermannovꢃ, L. Ruliek, L. Dufkovꢃ, M. Kotora, F.

´

Zhu, P. Koꢊovsky, J. Am. Chem. Soc. 2008 , 130 , 534 1 –5 348.

[40] Enantiomeric excess was roughly determined with the aid of chiral

lanthanide NMR shift reagents; see: L. M. Sweeting, D. C. Crans,

G. M. Whitesides, J. Org. Chem. 1987 , 52 , 227 3 –2 276.

[41] W. R. Roush, L. K. Hoong, M. A. J. Palmer, J. C. Park, J. Org.

Chem. 1990 , 55 , 410 9 –4 117.

[19] M. W. Rathke, P. Weipert in Comprehensive Organic Synthesis,

Vol. 2 (Eds.: B. M. Trost, I. Fleming, C. H. Heathcock), Pergamon,

Oxford, 1991, pp. 277–299.

[20] G. Cahiez, P. Y. Chavant, Tetrahedron Lett. 1989 , 30 , 737 3 –7 376.

[21] Additionally, we performed a control experiment in the absence of

titanium by treating benzaldehyde with allyl bromide and Mn dust

in THF at RT, in which we recovered the starting aldehyde un-

changed, thus confirming that under our conditions Mn dust does

not promote Barbier-type allylations.

[22] All yields reported in this paper refer to isolated products.

[23] In analogy with that which occurs with SmI₂(see ref. [7b]), the reac-

tion between [Ti^IIIClCp₂]and an allylic halide presumably leads first

to [Ti^IVCl(X)Cp₂] (X=Cl or Br) and an allylic radical. In this way,

high concentrations of allylic radicals might facilitate radical-radical

homocouplings leading to Wurtz-type byproducts, with the conse-

quent fall in the yield of the desired carbonyl allylation product.

[24] L. Yu-Ran, Handbook of Bond Dissociation Energies in Organic

Compounds, CRC Press, Boca Raton, 2003.

[25] For an excellent overview of the use of organometallic species in or-

ganic synthesis, see: a) L. S. Hegedus, Transition Metals in the Syn-

thesis of Complex Organic Molecules, 2nd ed., University Science

Books, Sausalito, 1999; for recent overviews on organometallic cata-

lysts, see: b) Multimetallic Catalysts in Organic Synthesis (Eds.: M.

Shibasaki, Y. Yamamoto), Wiley-VCH, Weinheim, 2004; c) Catalysts

for Fine Chemical Synthesis. Metal Catalysed Carbon–Carbon Bond-

Forming Reactions, Vol. 3 (Eds.: S. M. Roberts, J. Xiao, J. Whittall,

T. E. Pickett), Wiley, New York, 2004.

[26] It should be noted that both the excess of Mn and 2,4,6-collidine

can be recovered at the end of the experiments by filtering and

simple acid-base extraction respectively. Subsequently, both recov-

ered collidine and Mn dust can be employed in further experiments.

[27] S. Jana, C. Guin, S. C. Roy, Tetrahedron Lett. 2004 , 45 , 657 5 –6 577.

[28] For the yields obtained in the allylation of aldehydes 16–21 and ke-

tones 28–34 promoted by stoichiometric proportions of [TiClCp₂],

see Supporting Information.

[42] a) M. T. Reetz, M. Sauerwald, J. Org. Chem. 1984 , 49 , 229 2 –2 295;

b) A. Kasatkin, T. Nakagawa, S. Okamoto, F. Sato, J. Am. Chem.

Soc. 1995 , 117 , 388 1 –3 882, and references therein.

[43] Equatorial methyl groups of products 56 (1.20 ppm) and 62

(1.24 ppm) came into resonance at lower fields than axial methyl

groups of the respective isomers 73 (1.13 ppm) and 67 (1.10 ppm).

Moreover, equatorial alcohols 73 and 67 showed higher polarity

than axial alcohols 56 and 62.

[44] Intramolecular additions of radicals to carbonyl groups generally

are unfavourable, reversible processes. Nevertheless, carbonyl coor-

dination with Ti^IIImakes the process relatively fast and irreversible.

For an interesting discussion on this subject, see: A. Fernꢃndez-

Mateos, P. Herrero Teijꢇn, L. Mateos Burꢇn, R. Rabanedo Cle-

mente, R. Rubio Gonzꢃlez, J. Org. Chem. 2007, 72, 9973–9982.

[45] In this experiment we used an stoichiometric proportion of

[TiClCp₂] to facilitate allylic radical trapping by this reagent.

[46] a) For an overview on the hard–soft acid–base theory, see referen-

ce [2b] pp. 84–93; for seminal works on this item, see: b) R. G. Pear-

son, J. Am. Chem. Soc. 1963 , 85 , 353 3 –3 539; c) R. G. Pearson, J.

Songstad, J. Am. Chem. Soc. 1967 , 89 , 182 7 –1 836.

[47] Enantiomeric excess measured by chiral HPLC.

[48] For an overview of the chemistry and biosynthesis of terpenoids,

see: J. Mann, R. S. Davidson, J. B. Hobbs, D. V. Banthorpe, J. B.

Harborne, Natural Products: Their Chemistry and Biological Signifi-

cance, Longman Scientific & Technical, Essex, 1994, pp. 289–359.

[49] a) B.-S. Guo, W. Doubleday, T. Cohen, J. Am. Chem. Soc. 1987 , 109 ,

471 0 –4 711; b) A. Yanagisawa, S. Habaue, K. Yasue, H. Yamamoto,

J. Am. Chem. Soc. 1994 , 116 , 613 0 –6 141; c) B. Hamann-Gaudinet,

J. L. Namy, H. B. Kagan, J. Organomet. Chem. 1998 , 567 , 3 9 –4 7;

d) S. Matsukawa, Y. Funabashi, T. Imamoto, Tetrahedron Lett. 2003 ,

44 , 100 7 –1 010.

[50] It is known that prenyl radicals are prone to react at the a-position;

see: S. Yamago, M. Hashidume, J. Yoshida, Tetrahedron 2002 , 58 ,

680 5 –6 813.

[29] Y. Okude, S. Hirano, T. Hiyama, H. Nozaki, J. Am. Chem. Soc.

1977 , 99 , 317 9 –3 181.

[30] P. Girard, J. L. Namy, H. B. Kagan, J. Am. Chem. Soc. 1980 , 102 ,

269 3 –2 698.

[51] F. Sato, K. Iida, S. Ijima, H. Moriya, M. Sato, J. Chem. Soc. Chem.

Commun. 1981 , 114 0 –1 141.

[31] M. Gaudemar, Tetrahedron 1976 , 32 , 168 9 –1 691.

[52] Activation energies for radical-radical coupling reactions (in which

À

[32] L. A. Paquette, P. C. Lobben, J. Org. Chem. 1998 , 63 , 560 4 –5 616.

[33] M. T. Reetz, R. Steinbach, J. Westermann, R. Peter, B. Wenderoth,

Chem. Ber. 1985 , 118 , 144 1 –1 454.

[34] L. Zhao, D. J. Burnell, Tetrahedron Lett. 2006 , 47 , 329 1 –3 294.

[35] For an interesting discussion on the rearrangement of cyclopropane

substituted ketyl radicals, see: J. P. Stevenson, F. J. Woodward, J. M.

Tanko, J. Am. Chem. Soc. 2002 , 124 , 427 1 –4 281.

[36] a) B. M. Trost, Science 1991 , 254 , 147 1 –1 477; b) B. M. Trost, Angew.

Chem. 1995 , 107 , 28 5 –3 07; Angew. Chem. Int. Ed. Engl. 1995 , 34 ,

25 9 –2 81.

[37] For a Nobel Lecture on this topic, see: a) R. Noyori, Angew. Chem.

2002 , 114 , 210 8 –2 123; Angew. Chem. Int. Ed. 2002 , 41 , 200 8 –2 022;

one C C bond is formed but none is broken) are very low (near to

zero for small alkyl radicals); see: T. H. Lowry, K. S. Richardson,

Mechanism and Theory in Organic Chemistry, 3rd ed., Harper &

Row, New York, 1987, p. 738.

[53] For previously reported synthesis of rosiridol, see: a) B.-C. Hong, J.-

H. Hong, Y.-C. Tsai, Angew. Chem. 1998 , 110 , 48 2 –4 84; Angew.

Chem. Int. Ed. 1998 , 37 , 46 8 –4 70; b) E. Schçttner, K. Simon, M.

Friedel, P. G. Jones, T. Lindel, Tetrahedron Lett. 2008, 49, 5580–

5582.

[54] a) R. Majima, C. Kuroda, Acta Phytochim. 1922, 1, 43–65; b) H. J.

Brockmann, Liebigs Ann. Chem. 1936, 521, 1–47.

[55] S. R. Pulley, B. Czakꢇ, Tetrahedron Lett. 2004 , 45 , 551 1 –5 514.

2790

www.chemeurj.org

ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 2009, 15, 2774 – 2791

Barbier-Type Allylations

FULL PAPER

[56] E. A. Couladouros, A. T. Strongilos, V. P. Papageorgiou, Z. F. Plyta,

Chem. Eur. J. 2002 , 8 , 179 5 –1 803, and references therein.

[63] R. Guyon, P. Villa, Bull. Soc. Chim. Fr. 1972, 1375–1384.

[64] H. C. Aspinall, J. S. Bissett, N. Greeves, D. Levin, Tetrahedron Lett.

2002 , 43 , 31 9 –3 21.

[65] Y. Nagano, A. Orita, J. Otera, Adv. Synth. Catal. 2003 , 345 , 64 3 –

646.

[66] B. M. Trost, M. J. Bogdanowicz, J. Am. Chem. Soc. 1973 , 95 , 532 1 –

5334.

[67] J. G. Kim, K. W. Waltz, I. F. Garcꢂa, D. Kwiatkowski, J. P. Walsh, J.

Am. Chem. Soc. 2004 , 126 , 1258 0 –1 2585.

[68] S. P. Jong, J. M. Fang, J. Org. Chem. 2001 , 66 , 353 3 –3 537.

[69] G. A. Molander, D. J. St. Jean, Jr., J. Org. Chem. 2002 , 67 , 386 1 –

3865.

[57] a) A. Terada, Y. Tanoue, A. Hatada, H. Sakamoto, Bull. Chem. Soc.

Jpn. 1987 , 60 , 20 5 –2 13; b) S. Torii, K. Akiyama, H. Yamashita, T.

Inokuchi, Bull. Chem. Soc. Jpn. 1995 , 68 , 291 7 –2 922; c) K. C. Nico-

laou, D. Hepworth, Angew. Chem. 1998 , 110 , 86 4 –8 66; Angew.

Chem. Int. Ed. 1998 , 37 , 83 9 –8 41; d) Q. Lu, W. Liu, J. Ding, J. Cai,

W. Duan, Bioorg. Med. Chem. Lett. 2002 , 12 , 137 5 –1 378; e) Q. Lu,

H. L. Tang, Q. S. Shao, J. C. Cai, Chin. Chem. Lett. 2008 , 19 , 17 2 –

174.

[58] a) D. Zhang, C. D. Poulter, J. Am. Chem. Soc. 1995 , 117 , 164 1 –1 642;

b) H.-U. Humpf, N. Berova, K. Nakanishi, M. B. Jarstfer, C. D.

Poulter, J. Org. Chem. 1995 , 60 , 353 9 –3 542; c) M. B. Jarstfer, D.-L.

Zhang, C. D. Poulter, J. Am. Chem. Soc. 2002 , 124 , 883 4 –8 845.

[70] S. Baskaran, I. Islam, S. Chandrasekaran, J. Org. Chem. 1990 , 55 ,

89 1 –8 95.

[59] For

a

recent overview of asymmetric allylation reactions, see:

[71] P. Dionne, B. T. Ngatchat, D. Poirier, Steroids 1997 , 62 , 67 4 –6 81.

[72] J. Barluenga, I. Pꢁrez-Sanchez, M. G. Suero, E. Rubio, J. Flꢇrez,

Chem. Eur. J. 2006 , 12 , 722 5 –7 235.

[73] T. Tsuji, S. Usugi, H. Yorimitsu, H. Shinokubo, S. Matsubara, K.

Oshima, Chem. Lett. 2002, 31, 2–3.

a) R. W. Hoffman in Asymmetric Synthesis—The Essentials (Eds.:

M. Christmann, S. Brꢄse), Wiley-VCH, Weinheim, 2007, pp. 27–31;

for a recent report on Ir-catalyzed enantioselective allylations, see:

b) I. S. Kim, M.-Y. Ngai, M. J. Krische, J. Am. Chem. Soc. 2008 , 130 ,

634 0 –6 341.

[74] C. Gosmini, Y. Rollin, J. Perichon, C. Wakselman, M. Tordeux, L.

Marival, Tetrahedron 1997 , 53 , 602 7 –6 034.

[60] T. Ishiyama, T. Ahiko, N. Miyaura, J. Am. Chem. Soc. 2002 , 124 ,

1241 4 –1 2415.

[61] W. Oroshnik, G. Karmas, R. A. Mallory, J. Am. Chem. Soc. 1954 , 76 ,

232 5 –2 329.

Received: October 21, 2008

[62] S. Kobayashi, K. Nishio, J. Org. Chem. 1994 , 59 , 662 0 –6 628.

Published online: January 21, 2009

Chem. Eur. J. 2009, 15, 2774 – 2791

ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

2791

Article Doi

Ti-catalyzed Barbier-type allylations and related reactions

DOI: 10.1002/chem.200802180

Source and publish data:

Authors:

Article abstract of DOI:10.1002/chem.200802180

Full text of DOI:10.1002/chem.200802180

Products guided by the article

R&D Labs maybe for 1152182-30-3

Relevant to this article

Hot Product