Allyl alcohols and organocerium reagents, II. - Mechanism and extensibility of the reaction

Article abstract of DOI:10.1002/(SICI)1099-0690(200001)2000:1<99::AID-EJOC99>3.0.CO;2-1

Alkylcerium reagents add to the multiple bonds of allyl and propargyl alcohols in good yields and under mild conditions. The double bond can be reduced with lithium aluminum hydride in the presence of cerium trichloride. The regiochemistry of the attack depends on electronic factors.

Full text of DOI:10.1002/(SICI)1099-0690(200001)2000:1<99::AID-EJOC99>3.0.CO;2-1

FULL PAPER
Allyl Alcohols and Organocerium Reagents, II[°]
Mechanism and Extensibility of the Reaction
Giuseppe Bartoli,[a] M. Cristina Bellucci,[a] Marcella Bosco,[a] Renato Dalpozzo,*[b]
Antonio De Nino,[b] Letizia Sambri,[a] and Antonio Tagarelli[b]
Keywords: Allyl alcohols / Organocerium reagents / Alkenes / Addition reactions / Cerium
Alkylcerium reagents add to the multiple bonds of allyl and
hydride in the presence of cerium trichloride. The regio-
propargyl alcohols in good yields and under mild conditions. chemistry of the attack depends on electronic factors.
The double bond can be reduced with lithium aluminum
double bonds.[5] These organocerium reagents promise to
overcome most of the drawbacks of the previously reported
Introduction
The observation that organometallic compounds can add
to allyl alcohols dates back many years.[1] This reaction has
not yet found general synthetic applications, however, since
it is restricted to a few reactants and substrates, and often
requires extreme reaction conditions. Moreover the regio-
and stereochemical outcomes of the reaction are not easily
generalized.
reactions as the regiochemistry of the reaction is easily pre-
dictable and the reaction conditions are very mild.
In this paper we report on the extension of this reaction
to triple bonds and aryl-substituted double bonds, and the
chance to reduce the double bond of allyl alcohols by me-
ans of the reducing system LiAlH4 and CeCl3. Moreover,
on attempting to explain the reaction pathway we realized
that many different cerium species are involved, each with
a peculiar reactivity pattern. The exact nature of organocer-
ium compounds is still unknown and has proved to be re-
sistant to all attempts at elucidation,[6][7] although the dif-
ferent reactivity of cerium species, which appear at first
sight to be very similar, under different generation con-
ditions, has been demonstrated by Reetz.[8] For example, in
the reaction of aldehydes with organocerium compounds,
the same [CH3Ce(OiPr)3]Ϫ species gave different yields or
chemoselectivities when generated from CeCl3/3iPrOH/
3BuMgBr/CH3Li, Ce(OiPr)3/CH3Li or CeCl3/3iPrOH/
4CH3MgBr.
Two mechanisms were proposed for the reaction of Grig-
nard reagents. The first one involves an intramolecular re-
arrangement of an alkenoxy(alkyl)magnesium derivative,[2]
followed by an electrophilic attack of the magnesium atom
at the olefinic linkage and transfer of the alkyl framework
to the farthest terminus of the double bond to form a five-
membered magnesacyclic derivative. In such a mechanism
the greater the electronic density available in the RϪMg
bond, the more readily the alkyl group would migrate to
the olefinic bond. A very similar mechanism was also pro-
posed for the addition of organolithium reagents to allyl
alcohols in the presence of TMEDA as catalyst.[3] The
TMEDA should promote a more facile coordination of RLi
to the alcoholate function, removing the counterion from
the oxygen, followed by electrophilic addition of the lithium
to the double bond and transfer of the alkyl framework to
give a noncyclic dianonic species. The second mechanism
involves a magnesium halide catalysis in which the mag-
nesium atom, acting as an electrophile, weakens the double
bond and favors an intermolecular addition of the Grig-
nard reagent, releasing a magnesium halide molecule.[3]
Recently we reported that alkylcerium reagents add to
the double bond of aliphatic, straight chain allyl alcohols
to give alkenes with elimination of water.[4] These findings
open up an interesting and new window on the synthetic
application of organocerium reagents which, until now,
were employed mainly in the addition to electrophilic
Results and Discussion
The reaction was carried out as reported previously,[4]
i.e. the lithium alcoholate, preformed from the alcohol and
lithium hydride at 0°C, was added to the organocerium re-
agent, also preformed from a threefold excess of organoli-
thium with respect to dry cerium trichloride, at Ϫ78°C. The
reaction was allowed to stand at Ϫ20°C overnight before
quenching with 4% aqueous HCl and submitting the mix-
ture to the usual workup.
As shown in Table 1, the reaction works well with both
allyl (entries 1Ϫ12) and propargyl alcohols (entries 13Ϫ15)
with a large variety of substituents at the positions R1, R2
and R3. Among the organocerium species, the most re-
markable finding from this reaction is the addition of hy-
drides under the same experimental conditions (entries 1,
8, 15). The regiochemistry of the reaction is easily pre-
dicted, except for the addition of hydrides. In fact, in all
the reactions with R1 ϭ alkyl or H (entries 2Ϫ7), the alkyl
[a]
Dipartimento di Chimica Organica ЈA. ManginiЈ,
viale Risorgimento 4, I-40136 Bologna, Italy
[b]
`
Dipartimento di Chimica, Universita della Calabria,
I-87030 Arcavacata di Rende (Cosenza), Italy
Fax: (internat.) ϩ 39-98/449-2044
E-mail: dalpozzo@pobox.unical.it
Eur. J. Org. Chem. 2000, 99Ϫ104
WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
1434Ϫ193X/00/0101Ϫ99 $ 17.50ϩ.50/0
99

G. Bartoli, M. C. Bellucci, M. Bosco, R. Dalpozzo, A. De Nino, L. Sambri, A. Tagarelli
Table 1. Reaction of allyl and propargyl alcohols with various alkylcerium reagents at Ϫ78°C
FULL PAPER
[a]
[b]
Mixture LiAlH4/CeCl3. Ϫ See ref.[4]
framework of the organocerium reagent binds to the double might be tentatively described as [(R4)3CeCl2ORЈ]3Ϫ or
bond terminus farthest from the alcohol function, while [(R4)3CeORЈ]Ϫ (A, Scheme 1).
with R1 ϭ aryl (entries 9Ϫ14), it is linked to the nearest
one.
Species such as [R4Ce(ORЈ)3]Ϫ and the addition of R4Li
to RЈOCeCl2 are discarded for the following reasons. It has
been reported that Ce(ORЈ)3 species are able to form “ate”
complexes at Ϫ78°C.[6][8] However, the preparation of ce-
rium octen-3-olate (7a) at room temperature, in analogy
with MehorotraЈs procedure for lanthanide alcoholates,[9]
followed by addition of butyllithium (1:3:3 cerium trichlo-
ride/alcoholate/butyllithium ratio) at Ϫ78°C gave no reac-
tion. Moreover, the ability of cerium reagents to undergo
1,2-addition to α,β-unsaturated carbonyl compounds 6 is
well-known.[5] The intermediates of these reactions are al-
lylcerium alcoholates described as ROCeCl2 (B) and they
do not undergo further alkylation reactions with the excess
of organometallic reagent always present in the reaction
mixture.
The optimum reaction conditions require a molar ratio of
1:1:3 for alcoholate, cerium trichloride and organolithium
reagent, while a molar ratio of 1:1:1 leaves large amounts
of unreacted starting materials,[4] thus indicating that the
“R4CeCl2” species commonly reported in the reaction of
organocerium reagents[5k] cannot be the reactive species.
Denmark has reported that an [(R4)3CeCl3]3Ϫ species may
be present at Ϫ78°C.[7]
The formation of a negatively charged oxygen-containing
species is essential for the reaction to proceed. In fact, the
allyl ethers 5 do not react with organocerium reagents and
starting material is quantitatively recovered. If the al-
coholate 7 is not preformed by treatment with lithium hy-
dride, it can be formed in situ from deprotonation of the
alcohol by the organometallic reagent and thus a further
equivalent is required.[4]
On the other hand, some abnormal results in the reac-
tions of organocerium species can now be rationalized. The
only reported complete reduction of an unsaturated com-
Coordination to the oxygen atom of the cerium species, pound is the formation of 3-(2-hydroxyphenyl)-1-propanol
in analogy with the proposal of Eisch for lithium or mag- from the cerium-mediated lithium aluminum hydride re-
nesium,[2] should be the key intermediate step in the reac- duction of coumarin (Equation B, Scheme 2).[10] This result
tion. However, although magnesium forms a classical salt, can be rationalized supposing the intervention of the al-
cerium should form an “ate” complex, and assuming that coholate function at the aromatic ring which is formed after
the species proposed by Denmark is correct,[7] the complex the first reduction step. While the aliphatic oxygen is al-
100
Eur. J. Org. Chem. 2000, 99Ϫ104

Allyl Alcohols and Organocerium Reagents, II
FULL PAPER
Scheme 1. General mechanism of the reaction
ready busy, the aromatic one may form an “ate” complex R1 is an alkyl group, the “ate” complex transfers the R4
similar to A (C, Scheme 2) which should promote the se- alkyl group to the farthest terminus of the double bond
cond addition of hydride. Moreover, we have found the for- (Scheme 1, pattern a). At the same time, the cerium species
mation of 4,7,7-trimethyl-2,4-octanediol as a by-product[11] still bound to the oxygen is able to cause spontaneous elim-
in the cerium-mediated addition of vinyllithium to 4- ination of the alkene in an SN2Ј-type mechanism (Scheme
hydroxy-2-pentanone,[5g] although 4,7,7-trimethyl-2-oc- 1, pattern b). No evidence for the formation of an anionic
tanol was not observed in the reaction of 2-pentanone un- intermediate which might dehydrate during acidic workup,
der the same experimental conditions (Scheme 3).
such as that proposed by Felkin for the TMEDA-catalyzed
This result provides further evidence for our hypothesis. addition of RLi to allyl alcohols,[3] was found. In fact, all
In fact, since vinyllithium was prepared by metal-halogen attempts to trap anionic intermediates either by quenching
exchange from an excess of tert-butyllithium and vinyl bro- the reaction with deuterium oxide or by performing tandem
mide, the diol arises from addition of the adventitious tert- reactions were unsuccessful. The transfer most likely occurs
butylcerium to the allyl alcoholate intermediate promoted from above or below the double bond plane due to the bet-
by the second hydroxy function, which acts as the alkoxy ter overlap of the appropriate molecular orbitals, therefore
function able to give the “ate” complex.
the substitution at the carbon atom carrying the hydroxy
The “ate” complex can evolve following different patterns function (entries 1Ϫ4, 7, 12) or both cis and trans substi-
depending on the substitution at the double bond. When tution of the double bond (entries 5, 6) cannot influence
Eur. J. Org. Chem. 2000, 99Ϫ104
101

G. Bartoli, M. C. Bellucci, M. Bosco, R. Dalpozzo, A. De Nino, L. Sambri, A. Tagarelli
FULL PAPER
Scheme 2. Abnormal behaviour in the reduction of coumarin[10]
Scheme 3. Abnormal addition of tBuLi to β-hydroxy ketones[11]
the reaction. However, when both R1 and R3 are not hydro- terium oxide, but tandem reactions with benzaldehyde or
gen, as in compound 1f, a 10:1 mixture of diastereomers is styrene oxide were unsuccessful.
obtained. When the group to be transferred is the small
With iodine as the electrophile a reaction occurs, but the
hydride ion, it links to the terminus of the double bond expected iodo derivative is not recovered. The products of
nearest to the hydroxyl function without elimination the reaction are an alkene and an aldehyde, most likely aris-
(Scheme 1, pattern c), leading to a carbanionic intermedi- ing from β-elimination from the iodo derivative (Scheme 4).
ate, as demonstrated by quenching the reaction with deut- To the best of our knowledge, no similar reactions are re-
erum oxide; 1-d-3-octanol (d-4a) was isolated in yields simi- ported for other β-iodoalcoholates.
lar to those reported in Table 1 (entry 1). When R1 is a
In conclusion the addition reaction of organoceriums to
phenyl group, pattern c is also followed by the reaction allyl alcohols appears to be widely applicable. It represents
(table: entries 8Ϫ12), but it is likely that electronic factors a vast improvement in the chemistry of the addition of or-
due to the presence of a charge-stabilizing substituent (the ganometallic reagents to functionalized double bonds. In
phenyl group) reverse the regiochemistry. As expected, the fact, at present, this reaction seems to have no drawbacks
anionic intermediate can be trapped by reaction with deu- and, moreover, employs easily available starting materials
102
Eur. J. Org. Chem. 2000, 99Ϫ104

Allyl Alcohols and Organocerium Reagents, II
FULL PAPER
Scheme 4. Attempted trapping of anionic intermediates with S2
terized by NMR, IR and mass spectroscopy and gave physical data
identical to those reported in the literature for 6-dodecene,[12] 3-
butylcyclohexene,[13] 3-phenylcyclohexene,[13] 3-propyl-1-hep-
tene,[14] 4-phenyl-3-octene,[15] 2-(phenylmethyl)-1-hexanol,[16] 3-
phenylpropanol,[17] 2-methyl-3-phenylpropanol,[18] (E)-2-methyl-3-
phenylprop-2-en-1-ol,[19] (E)-2-butyl-3-phenylprop-2-en-1-ol.[20] 3-
Octanol and cinnamyl alcohol (from reaction between 1g and 2a)
were identical to commercial samples.
and mild experimental conditions. Studies are in progress
to understand how long the chain between the oxygen atom
and the double bond might be.
These results have also shown that the reactivity of or-
ganoceriums is very complex. The organocerium species
shown to be the same in many reaction schemes for the sake
of simplicity are actually very different and behave differ-
ently. The discovery of the exact nature of these compounds
is a very stimulating field of organometallic chemistry.
2-(Phenylmethyl)-1-octanol: 1H NMR (200 MHz, CDCl3): δ ϭ 0.87
(t, J ϭ 6.7 Hz, 3 H, CH3), 1.22Ϫ1.37 (m, 10 H, 5 CH2), 1.76Ϫ1.78
(m, 1 H, CH), 1.84 (br. s, 1 H, OH), 2.61 (ABX, JAB ϭ 15, JAX
ϭ
3.2; JBX ϭ 3.8 Hz, 2 H, CH2Ph), 3.48 (d, J ϭ 5.2 Hz, 2 H,
CH2OH), 7.14Ϫ7.28 (m, 5 H, Ar). Ϫ 13C NMR (50 MHz, CDCl3):
δ ϭ 14.30 (CH3), 22.87 (CH2), 27.14 (CH2), 29.81 (CH2), 30.95
(CH2), 32.04 (CH2), 37.84 (CH2Ph), 42.76 (CH), 64.96 (CH2OH),
125.99 (CH, Ar), 128.44 [2 ϫ CH, Ar], 129.38 [2 ϫ CH, Ar], 141.07
(C, Ar). Ϫ IR (film) ν˜ ϭ 3331 (OH) cmϪ1. Ϫ MS: m/z (%) ϭ 220
(8), 202 (15), 131 (16), 117 (43), 104 (78), 91 (100), 77 (7).
Experimental Section
General: 1H NMR spectra were recorded at 200 MHz or 300 MHz
and 13C NMR at 50.4 MHz or 75.5 MHz with a Varian Gemini 200
instrument or a Bruker WM300 instrument, respectively. Chemical
shifts are given in ppm. from Me4Si. Coupling constants are given
in Hz. Assignment of the signals in the 13C NMR spectra was made
by DEPT experiments. IR spectra were recorded with a PerkinϪ
Elmer Paragon 1000 PC FTIR spectrometer. MS were recorded
with HP 5972 workstation.
1
2-(Phenylmethyl)-1-phenyl-1-hexanol: mixture of diastereomers: H
NMR (200 MHz, CDCl3): δ ϭ 0.78 (t, J ϭ 6.71 Hz, 3 H, CH3),
1.12Ϫ1.40 (m, 6 H, 3 CH2), 1.89 (br. d, J ϭ 2.47 Hz, 1 H, OH),
1.96Ϫ2.08 (m, 1 H, CH), 2.66 (ABX, JAB ϭ 13.76, JAX ϭ 4.58,
JBX ϭ 8.47 Hz, 2 H, CH2Ph), 4.60Ϫ4.80 (m, 1 H, CHOH),
7.10Ϫ7.40 (m, 10 H, Ar). Ϫ 13C NMR (50 MHz, CDCl3) major
diastereomer: δ ϭ 13.87 (CH3), 22.75 (CH2), 29.07 (CH2) 29.16
(CH2), 35.01 (CH2Ph), 46.97 (CH), 75.57 (CHOH), 125.63 (CH,
Ar), 126.39 (CH, Ar), 127.27 [2 ϫ CH, Ar], 128.12 [2 ϫ CH, Ar],
128.22 [2 ϫ CH, Ar], 129.27 [2 ϫ CH, Ar] 141.16 (C, Ar), 143.76
(C, Ar). Minor diastereomer (detectable peaks): δ ϭ 15.19 (CH3),
27.45 (CH2), 29.24 (CH2) 36.84 (CH2), 47.22 (CH2Ph), 65.76 (CH),
74.70 (CHOH). Ϫ IR (film) ν˜ ϭ 3408 (OH) cmϪ1. Ϫ MS: m/z
(%) ϭ 268 (1), 250 (12), 193 (9), 162 (11), 107 (100), 91 (41), 79 (25).
THF was dried by refluxing over sodium wire/benzophenone until
the blue color of benzophenone ketyl persisted, and then distilling
into a dry receiver under nitrogen atmosphere. 1-Octen-3-ol (1a),
cyclohex-2-en-1-ol (1b), cis and trans-hex-2-en-1-ol (c-1c and t-1c)
and trans-cynnamyl alcohol (1e) are commercially available (Ald-
rich). 2-Phenylpent-1-en-3-ol (1d) (α-styrylmagnesium bromide and
propionaldehyde),
trans-1-phenylcinnamyl
alcohol
(1f)
(phenyllithium and trans-cinnamaldehyde), 3-phenylpropargyl al-
cohol (1g) (lithium phenylacetylide and formaldehyde) were ob-
tained by classical Grignard reactions from the reactants in brack-
ets.
Preparation of Cerium Allyl Alcoholate: Lithium hydride
(16.5 mmol) was poured into a THF solution of alcohol 1a
(15 mmol) with stirring at 0°C under a nitrogen atmosphere. After
1 h the mixture was added to a stirred suspension of dry cerium
chloride (5 mmol), prepared as above, by syringe at Ϫ78°C under
a nitrogen atmosphere. After 2 h a solution of BuLi (15 mmol) was
dropped from a funnel into the mixture at Ϫ78°C. The mixture
was then allowed to stir at 0°C and monitored by GC/MS and
TLC. After two days both chromatograms showed the peak for 1a
to be the only product.
Preparation of the Organocerium Reagents 2a؊d: Commercial ce-
rium(III) chloride heptahydrate (5.5 mmol) was placed in a flask
with a stir bar. The flask was heated in vacuo in an oil bath to
140°C/0.2 Torr for 2 h. Nitrogen was introduced while the flask
was still hot. The flask was cooled in an ice bath and dry THF was
introduced from a syringe. The suspension was stirred overnight at
room temperature. The resulting white slurry was then cooled to
Ϫ78°C and the titrated organolithium reagent (16.5 mmol) was ad-
ded dropwise from a syringe [or lithium aluminum hydride
(7.5 mmol) was poured in] and the mixture allowed to stir for 2 h.
Deuterium Oxide Quenching: The reaction between 1a and 2b was
quenched with D2O extracted with ether and the organic layer then
submitted to GC/MS analysis. No deuterium incorporation was
detected since the Mϩ/(Mϩ ϩ 1) ratio of 6-dodoecene remained
unchanged with respect to the standard procedure.
General Procedure for the Addition of Allyl Alcoholates to Organ-
ocerium Reagents: Lithium hydride (5.5 mmol) was poured into a
THF solution of allyl (or propargyl) alcohol 1a؊h (5 mmol) with
stirring at 0°C under a nitrogen atmosphere. After 1 h the mixture
was syringed into the organocerium reagent with stirring at Ϫ78°C
under a nitrogen atmosphere. The reaction was allowed to stand in The reaction between 1a and 2a was quenched with D2O extracted
a freezer for 24 h before quenching with 4% HCl solution, extrac-
tion with diethyl ether, and washing with water. The dried
(Na2SO4) extracts were concentrated under reduced pressure and
purified by flash chromatography on a silica gel column light petro-
leum (40Ϫ60°)/diethyl ether, 9:1 as eluent]. Yields of the recovered
products are listed in Table 1. All compounds were fully charac-
with ether and the organic layer then submitted to GC/MS analysis.
The peak corresponding to 3-octanol showed the following frag-
mentation pattern: m/z (%) ϭ 113 (2), 101 (17), 83 (51), 60 (100),
55 (76), 41 (47). 1-d-3-Octanol was isolated after column chroma-
tography (72% yield, oil) and submitted to NMR analysis: 1H
NMR (200 MHz, CDCl3): δ ϭ 0.83Ϫ0.95 (m, 5 H, CH3 ϩ CH2D),
Eur. J. Org. Chem. 2000, 99Ϫ104
103

G. Bartoli, M. C. Bellucci, M. Bosco, R. Dalpozzo, A. De Nino, L. Sambri, A. Tagarelli
FULL PAPER
1.20Ϫ1.60 (m, 10 H, 5 CH2), 1.88 (br. s, 1 H, OH), 3.46Ϫ3.57 (m,
and 52% yield respectively. 1-Phenyl-1-hexene was fully charac-
1 H, CHOH). Ϫ 13C NMR (50 MHz, CDCl3): δ ϭ 9.45 (t, J ϭ terized and gave physical data identical to those reported in the
19.3 Hz, CH2D), 13.89 (Me), 22.56 (CH2), 25.26 (CH2), 29.98, literature.[21] From the latter reaction a 51% yield of benzaldehyde
(CH2), 31.88 (CH2), 36.88 (CH2), 73.23 (CHOH)
(identical to the commercial product) was also recovered.
A reaction between 1e and 2e was quenched with D2O extracted
with ether and the organic layer then submitted to GC\MS analysis.
The peak corresponding to 2-(phenylmethyl)-1-octanol showed the
following fragmentation pattern: m/z (%) ϭ 221 (5), 203 (8), 131
(12), 117 (31), 104 (72), 92 (100). 2-(Phenyl-d-methyl)-1-octanol was
isolated after column chromatography (69% yield, oil) and submit-
[1]
`
M. Cherest, H. Felkin, C. Frajerman, C. Lion, G. Roussi, G.
Swierczewski, Tetrahedron Lett. 1966, 875.
[2]
[3]
J. J. Eisch, J. H. Merkeley, J. Am. Chem. Soc. 1979, 101, 1148.
`
H. Felkin, G. Swierczewski, A. Tambute, Tetrahedron Lett.
1969, 707.
[4]
[5]
R. Dalpozzo, A. De Nino, A. Tagarelli, G. Bartoli, M.C.
Bellucci, M. Bosco, L. Sambri, J. Org. Chem. 1998, 63, 9559.
[5a] T. Imamoto, Pure Appl. Chem. 1990, 62, 747. Ϫ [5b] T. Imam-
1
ted to NMR analysis: H NMR (200 MHz, CDCl3): δ ϭ 0.87 (t, 3
H, J ϭ 6.6 Hz, Me), 1.16Ϫ1.36 (m, 10 H, 5 CH2), 1.55 (t, J ϭ 6.4
Hz, 1 H, OH), 1.7Ϫ1.82 (m, 1 H, CH), 2.59 (d, J ϭ 6.5 Hz, 1 H,
CHD), 3.50 (d, J ϭ 5.3 Hz, 2 H, CH2OH), 7.16Ϫ7.30 (m, 5 H,
Ar). Ϫ 13C NMR (50 MHz, CDCl3): δ ϭ 14.24 (Me), 22.80 (CH2),
27.07 (CH2), 29.74 (CH2), 30.89 (CH2), 31.97 (CH2), 37.5 (t, J ϭ
19.5 Hz, CHD), 42.63 (CH), 64.92 (CHOH), 125.95 (CH, Ar),
128.40 [2 ϫ CH,Ar], 129.30 [2 ϫ CH, Ar].
oto, in Comprehensive Organic Chemistry, vol 1 (Eds.: B. M.
[5c]
Trost, I. Fleming), Pergamon, Oxford, 1991, 231. Ϫ
T. Im-
amoto, Lanthanides in Organic Synthesis, Academic Press, New
[5d]
York, 1994. Ϫ
Petrini, Tetrahedron Lett. 1994, 33, 8453. Ϫ
Sambri, E. Marcantoni, M. Bosco, Tetrahedron Lett. 1994, 33,
G. Bartoli, L. Sambri, E. Marcantoni, M.
[5e]
G. Bartoli, L.
[5f]
8651. Ϫ
G. Bartoli, L. Sambri, E. Marcantoni, M. Tambur-
ini, Angew. Chem. Int. Ed. Eng. 1995, 34, 2046. Ϫ [5g] G. Bartoli,
M. Bosco, J. Van Beck, L. Sambri, E. Marcantoni, Tetrahedron
Attempted Tandem Reaction: When formation of 6-dodecene from
the reaction between 1a and 2b was detected by GC/MS, styrene
oxide (6 mmol) or benzaldehyde (6 mmol) in THF was added.
Samples, treated as above, were submitted to GC/MS analyses. No
products arising from tandem addition to oct-1-en-3-ol were de-
tected. The only recognizable products by GC/MS were styrene ox-
ide, 6-dodecene, 1-phenyl-1-hexene (from addition of butyl car-
banion excess to styrene oxide) in the former reaction and benzal-
dehyde, 6-dodecene, and 1-phenylpentanol (from addition of butyl
carbanion excess to benzaldehyde) in the latter.
[5h]
Lett. 1996, 35, 1293. Ϫ
G. Bartoli, E. Marcantoni, M. Pe-
[5i]
trini, L. Sambri, Chem. Eur. J. 1996, 2, 913. Ϫ
G. Bartoli,
M. Bosco, L. Sambri, E. Marcantoni, F. Nobili, J. Org. Chem.
1997, 62, 4183. Ϫ [5j] G. Bartoli, M. Bosco, R. Dalpozzo, A. De
Nino, E. Marcantoni, L. Sambri, J. Org. Chem. 1998, 63, 3745.
[5k]
Ϫ
H.-J. Liu, K.-S. Shia, X. Shang, B.-Y. Zhu Tetrahedron
1999, 55, 3803.
[6]
[7]
[8]
C. Alcaraz, U. Groth, Angew. Chem. Int. Ed. Eng. 1997, 36,
2480.
S. E. Denmark, J. P. Edwards, O. Nicaise, J. Org. Chem. 1993,
58, 569.
M. T. Reetz, H. Haning, S. Stanchev, Tetrahedron Lett. 1992,
31, 6963.
When formation of 2-(phenylmethyl)-1-hexanol from the reaction
between 1e and 2b was detected by GC/MS, styrene oxide (6 mmol)
or benzaldehyde (6 mmol) in THF was added. Samples, treated as
above, were submitted to GC/MS analyses. No products arising
from tandem addition to oct-1-en-3-ol were detected. The only re-
cognizable products by GC/MS were styrene oxide, 2-(phe-
nylmethyl)-1-hexanol, 1-phenyl-1-hexene (from addition of butyl
carbanion excess to styrene oxide) in the former reaction and ben-
zaldehyde, 2-(phenylmethyl)-1-hexanol, and 1-phenylpentanol
(from addition of butyl carbanion excess to benzaldehyde) in the
latter.
[9]
R. C. Mehorotra, J. M. Batwara Inorg. Chem. 1970, 9, 2507.
S. Fukuzawa, T. Fujinami, S. Yamauchi, S. Sakay, J. Chem.
Soc., Perkin Trans. 1 1986, 1929.
[10]
[11]
[12]
[13]
[14]
G. Bartoli, M. Bosco, J. Van Beck, L. Sambri, R. Dalpozzo, E.
Marcantoni, unpublished results.
T. Hamatami, S. Matsubara, H. Matsuda, M. Schlosser, Tetra-
hedron 1988, 44, 2875.
C. C. Tseng, S. D. Paisley, H. L. Goering, J. Org. Chem. 1986,
51, 2884.
J-E. Baeckvall, M. Sellen, B. Grant, J. Am. Chem. Soc. 1990,
112, 6615.
[15]
[16]
B. Dunkelblum, Tetrahedron Lett. 1973, 669.
G. Boireau, D. Abenhaim, E. Henry-Basch, Tetrahedron 1980,
36, 3061.
[17]
[18]
Iodine Treatment of the Reaction Between 1e,f and 2b: When forma-
tion of 2-(phenylmethyl)-1-hexanol and 2-(phenylmethyl)-1-phenyl-
1-hexanol, from the reaction between 1e and 2b and 1f and 2b
respectively, were detected by GC/MS, a THF solution of iodine
(6 mmol) was added. The reaction was allowed to stand in a freezer
overnight and then treated with HCl (4%) and extracted with ether.
The organic layer was washed with HCl (4%), a sodium hydrogen
sulfite solution and water, then dried over dry sodium sulfate and
evaporated under reduced pressure. After column chromatography
from both reactions trans-1-phenyl-1-hexene was recovered in 64%
C. G. Screttas, M. Micha-Screttas, J. Org. Chem. 1982, 47, 3008.
S. E. Drewes, D. G. S. Malissar, G. H. P. Roos, Chem. Ber. 1993,
126, 2663.
[19]
G. W. Daub, J. P. Edwards, C. R. Okada, J. W. Allen, C. T.
Maxet, M. S. Wells, A. S. Goldstein, M. J. Dibley, C. J. Wang,
D. P. Ostercamp, S. Chung, P. S. Cunningham, M. A. Berliner,
J. Org. Chem. 1997, 62, 1976.
J. K. Crandall, F. Collonges, J. Org. Chem. 1976, 41, 4089.
S.-I. Murahashi, M. Yamamura, K.-I. Yanagisawa, N. Mita, K.
Kondo, J. Org. Chem. 1979, 44, 2408.
[20]
[21]
Received July 26, 1999
[O99456]
104
Eur. J. Org. Chem. 2000, 99Ϫ104