Regioselectivity in the intramolecular allyl transfer reaction catalysed by electrogenerated nickel complexes: Influence of metal ions

Article abstract of DOI:10.1016/S0040-4020(02)01184-5

The intramolecular transfer of allyl moieties from substituted allyl aryl ethers to carbonyl groups has been studied by electrosynthesis, in a nickel-catalysed reaction. The influence of metal ions such as Mg²⁺, Zn²⁺ and Al³⁺ has been examined. Regioselectivity towards the branched isomer was better with Zn²⁺ than with Mg²⁺ ions, but it was higher in the absence of added metal ions.

Full text of DOI:10.1016/S0040-4020(02)01184-5

Tetrahedron 58 (2002) 9289–9296
Regioselectivity in the intramolecular allyl transfer reaction
catalysed by electrogenerated nickel complexes:
influence of metal ions
Delphine Francoa and Elisabet Dun˜achb
,
*
a
ˆ
`
´
´
Laboratoire Aromes, Syntheses et Interactions, CNRS, UMR 6001, Universite de Nice-Sophia Antipolis, 06108 Nice cedex 2, France
´ ´
Laboratoire de Chimie Bio-Organique, CNRS, UMR 6001, Universite de Nice-Sophia Antipolis, 06108 Nice cedex 2, France
b
Received 14 March 2002; revised 1 June 2002; accepted 11 September 2002
Abstract—The intramolecular transfer of allyl moieties from substituted allyl aryl ethers to carbonyl groups has been studied by
electrosynthesis, in a nickel-catalysed reaction. The influence of metal ions such as Mg2þ, Zn2þ and Al3þ has been examined.
Regioselectivity towards the branched isomer was better with Zn2þ than with Mg2þ ions, but it was higher in the absence of added metal
ions. q 2002 Elsevier Science Ltd. All rights reserved.
1. Introduction
We studied an intramolecular version of the electrochemical
allylation of carbonyl compounds in which the allyl moiety
is issued from the cleavage of allyl ethers as summarised in
compound 1a (Eq. (1)).12
The allylation of carbonyl compounds is a widely used
reaction for C–C bond formation. Generally, a previously
prepared allylic organometallic reagent is added to a
carbonyl group to form the homoallylic alcohol.1 In the
presence of non-symmetrically substituted allyl groups,
branched or linear regioisomers can be obtained, depending
on the nature of the metal reagent and on the reaction
conditions.2 Thus, high regioselectivities towards the
branched isomer have been reported with allyl Grignard
reagents and with most of the allylic organometallic
reagents,3,4 although a reversed regioselectivity has been
observed with allyltin5 or allylbarium reagents.6,7 The
addition of metal ions can influence,3 and even reverse8 the
regioselectivity.
ð1Þ
This allyl transfer reaction allows the synthesis of alcohol–
phenol derivatives in moderate to good yields. These highly
functionalised compounds can undergo further reactivity,
such as the intramolecular cyclisation to benzopyrans. The
allyl transfer is catalysed by an electrogenerated Ni(0)
complex, formed from the in situ reduction of the0 cationic
Ni(II) complex, Ni(bipy)23þ, 2BF42((bipy¼2,2 -bipyri-
dine).13 The homoallylic alcohol–phenol, 2a, was obtained
in up to 87% yield.
In the field of electrosynthesis, we have been interested in
the intramolecular allylation of carbonyl compounds. The
electrochemical allylation of carbonyl compounds with allyl
chlorides and acetates has been reported.9,10 Regio-
selectivity has been examined in the case of the direct
electrochemical allylation,10 as well as in the case of the
nickel-catalysed allylation in the presence of a Zn anode.11
In both cases, the branched regioisomer was predominant.
The influence of metal ions such as Mg2þ, Zn2þ or Al3þon
the regioselectivity of the intramolecular allyl transfer
process has not been addressed before. We present here our
results on the regioselectivity of the electrochemical
intramolecular allylation process with derivatives of type
1, with non-symmetrically substituted allyl groups. We also
present the study of the influence of several metal ions, such
as Mg2þ, Zn2þ and Al3þ, on the regioselective outcome of
the reaction.
Keywords: allylation; electrochemical; regioselectivity; nickel; metal ions.
*
ˆ
Corresponding author. Address: Parc Valrose Laboratoire Aromes,
Synthese et Interactions, Universite de Nice-Sophia Antipolis, Parc
Valrose, F-06108, Nice cedex 2, France. Tel.: þ33-4-92-07-61-42; fax:
þ33-4-92-07-61-51; e-mail: dunach@unice.fr
0040–4020/02/$ - see front matter q 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(02)01184-5

9290
D. Franco, E. Dun˜ach / Tetrahedron 58 (2002) 9289–9296
2. Results and discussion
transfer reaction (yield of 99%) with a ratio 2d/3d of 99:1.
In this case a very selective process occurred. The branched
regioisomer 2d was obtained as a 45:55 mixture of
stereoisomers.
Allyl ethers 1b–1f were used as model compounds and
were prepared by allylation of ortho-hydroxybenzaldehyde
and ortho-hydroxyacetophenone with the corresponding
allyl bromides in K2CO3/DMF. The electrolyses were
carried out in a single-compartment cell at room tempera-
ture, with 10 mol% of the Ni(II) catalyst with respect to the
substrate.
The electrolysis of methyl ketones 1e and 1f, with a
Mg/stainless steel couple of electrodes, afforded no allylated
compound. Only the C–O cleavage of the allyl group
occurred, with the formation of 5 in 90 and 50% yields,
respectively.
2.1. Allylation in the presence of Mg21 ions
The results in Table 1 indicate that the main regioisomer
corresponds to the branched compound 2, and the regio-
selectivities are highly dependent on the type of substituent.
The cinnamyl group presents the lower selectivity, whereas
the presence of a single methyl substituent in the crotyl
group of 1d affords almost exclusively the branched isomer.
The [Ni(bipy)3]2þ2BF42-catalysed electrolysis of substrate
1b in the presence of a Mg anode and a stainless-steel
cathode led to a 53% yield of intramolecular allyl transfer
compounds, with a 85:15 ratio of branched/linear regio-
isomers, 2b and 3b (Eq. (2) and Table 1). The reaction was
run in DMF, at room temperature and at constant current
intensity. The complete consumption of 1b was reached
after 2.4 F mol21 electrolysis and the main by-product was
salicylaldehyde 4, formed in 14% yield. Compound 4 is
issued from a Ni-catalysed deallylation process but without
allyl transfer. The presence of the nickel catalyst was
essential for the reaction; in its absence, the elecroreduction
of 1a, 1b or 1c led mainly to 4, with low conversions. These
results are summarised in Table 1.
When comparing our data with the regioselective outcome
of the reaction of allyl Grignards with carbonyl com-
pounds,2 the results of the chemical and the electrochemical
reactions are similar: the formation of the branched
regioisomer is favoured in both cases, with slightly better
regioselectivities in the Ni-catalysed electrochemical
process.
2.2. Allylation in the presence of Zn21 ions
In order to compare the influence of the nature of the metal
ions issued from the anodic oxidation and present in
solution, several electrochemical intramolecular allylation
reactions were carried out with substrates 1 in the presence
of Zn and Al anodes. The results are presented in Tables 2
and 3.
The use of a Zn anode also enabled the process of allyl
transfer from ethers 1, as shown in entry 1, Table 2, with the
formation of 2a in 63% yield. Some elimination from the
homoallyl alcohol 2a to the diene 6a occurred, resulting in
an overall allyl transfer of 90% from 1a.
The electrolysis of 1b (entry 2) led to a 90% allyl transfer
with a 2b/3b ratio of 89:11. Cinnamyl derivative 1c afforded
91% allylation with a 2c/3c ratio of 83:17 (entry 4). The
electrolysis of the crotyl ether derivative 1d with a Zn anode
led mainly to 2d, in a 99:1 regioselectivity (entry 6).
Compound 2d was obtained as a 79:21 mixture of
stereoisomers.
ð2Þ
The electrochemical procedure in a single-compartment cell
is a one-step allyl ether cleavage—allyl transfer process, in
which there is no need for the previous preparation of the
metal-allyl species. At the anode, a magnesium rod is
oxidised to Mg2þ ions, and at the cathode the Ni(II) catalyst
precursor is reduced to Ni(0), the active catalytic species.
The mechanistic aspects of this reaction are discussed
hereafter.
With a Zn anode, the ratios towards the branched isomer
increased in the reactions involving 1b–1d, as compared to
the use of a Mg anode, and regioselectivities higher than
83% were attained for isomer 2.
To check if the concentration of Zn2þ ions in solution had a
beneficial effect on the regioselective outcome of the allyl
transfer, the electrolysis of 1b was carried out in the
presence of an excess of Zn2þ. Thus, a pre-electrolysis of a
DMF solution of 1,2-dibromoethane with n-Bu4NþBF42 in
the presence of a Zn anode for 1.85 equiv. of Zn2þ towards
1b, was followed by the standard electrolysis. The results
(entry 3) led to a ratio of branched to linear isomers of
89:11. However, the reaction was not complete until the
The electrolysis of the cinnamyl ether derivative, 1c, under
the same conditions, led to 2c and 3c in 70% yield, with a
67:33 ratio of branched to linear regioisomers. Only traces
of 4 were obtained.
The electrolysis of 1d led to an almost quantitative allyl

D. Franco, E. Dun˜ach / Tetrahedron 58 (2002) 9289–9296
9291
Table 1. Nickel-catalysed electrochemical allyl transfer from substrates 1 in DMF, with a Mg/stainless steel couple of electrodes
Entry
Substrate
F mol21
Products (%)
Yield of allyl transfer (2þ3) (%)
Ratio 2/3
1
2.1
87
–
2
2.4
53
85:15
3
2.5
70
67:33
4
2.9
99
99:1
5
6
3.0
2.0
5 (90%)
5 (50%)
–
–
–
–
b
See Section 4. One indeterminated diastereoisomer. c 2d as a 45:55 ratio of stereoisomers.

9292
D. Franco, E. Dun˜ach / Tetrahedron 58 (2002) 9289–9296
Table 2. Nickel-catalysed electrochemical allyl transfer from substrates 1 in DMF, with a Zn/stainless steel couple of electrodes
Entry
Substrate
F mol21
Products (%)
Yield of allyl transfer (2þ3) (%)
Ratio 2/3
1
1a
4.6
90
–
2
3
1b
3.0
3.7
2b (80%)
90
90
88:12
89:11
3b (10%)
2b (80%)b
3b (10%)
4 (traces)
2c (75%)c
3c (16%)
4 (7%)
1ba
4
5
6
1c
2.6
2.0
5.0
91
73
87
83:17
70:30
99:1
1cd
1d
2c (51%)e
3c (22%)
4 (27%)
2d (86%)f
3d (1%)
4 (13%)
7
1e
2.7
40
.99:1
8
1f
3.3
19
,1:99
See Section 4.
a
Reaction with the addition of 1.85 equiv. of Zn2þ to the reaction medium before electrolysis.
2b was found partially under its reduced form on the double bond, ratio 60:20.
2c obtained together with the corresponding dehydrated product in a 73:27 ratio.
b
c
d
e
f
Reaction run in two-compartment cell with the addition of 5 equiv. of Zn2þ to the reaction medium before electrolysis.
2c obtained together with the corresponding dehydrated product in a 20:80 ratio.
2d as a 79:21 mixture of diastereomers.
passage of 3.7 F mol21. Thus, comparing entries 2 and 3,
the effect of the added Zn2þ ions does not seem to influence
the regioselectivity.
linear ratio of allylated compounds reached 70:30. The
excess of Zn2þ ions did not allow to get a higher ratio of 2c
(compare entries 4 and 5). Moreover, under these conditions
the reaction was less selective: compound 4 was obtained in
27% yield and the branched regioisomer 2c was partially
dehydrated to the corresponding diene. We could conclude
that the excess of Zn2þ ions did not favour the selectivity of
the allyl transfer.
The high electricity consumption in several reactions run in
the presence of a Zn anode is most probably due to the
parallel process of reduction of the Zn2þ ions present in the
medium. Indeed, the reduction potentials of Ni(II) and Zn2þ
are close enough (21.2 and 21.4 V, respectively) to reduce
the Zn2þ ions when their concentration progressively
increases.
The partial dehydration of isomer 2 was observed in the case
of 2a (entry 1) and 2c (entry 5) with a Zn anode, but not with
a Mg anode. For 2c, out of the 75% of the branched isomer
formed, 55% of 2c and 20% of the corresponding
dehydrated diene were obtained after 2.6 F mol21
electrolysis.
A different electrolysis was carried out in the presence of
excess Zn2þ ions, in a two-compartment cell at controlled
potential of 21.3 V vs SCE, corresponding to the reduction
potential of the Ni(II) species. Allyl ether 1c was thus
electrolysed in the presence of 5 equiv. of added Zn2þ ions
at the cathodic compartment (entry 5). The branched to
The intramolecular allyl transfer from allyl ethers to ketone
groups could be obtained in the presence of a Zn anode

D. Franco, E. Dun˜ach / Tetrahedron 58 (2002) 9289–9296
9293
Table 3. Nickel-catalysed electrochemical allyl transfer from substrates 1
in DMF, with a A1/stainless steel couple of electrodes
We also carried out the electrolysis of 1b in the presence of
the nickel catalyst but in the absence of other metal ions.
Such reaction was run in a two-compartment cell at 21.3 V
vs SCE, for 2 F mol21. The yield of the allyl transfer
products 2b and 3b was of 89% and the ratio 2b/3b was
91:9. This result is interesting and reveals that high
regioselectivity can be obtained in the absence of metal
ions such as Mg2þ, Zn2þ or Al3þ. However, from the points
of view of the simplicity and time of the electrolysis, the
reactions carried out in a one-compartment cell with a metal
anode under constant intensity conditions are more efficient.
Entry Substrate F mol21
Products
(%)
Yield of
allyl transfer
(2þ3)
Ratio 2/3
(%)
1
2
1a
4.3
3.3
2a (90%)
4 (traces)
2b (24%)
3b (37%)
4 (4%)
90
61
–
1ba
39:61
3
4
1c
2.0
4.0
2c (25%)
3c (15%)
4 (30%)
2d (40%)
3d (traces)
4 (22%)
40
40
62:38
99:1
Other controlled-potential electrolyses were carried out in
order to better evaluate the role of the metal ions in the
efficiency of the reaction and/or in the recycling of the
nickel species. Thus, the controlled-potential electrolysis of
a 1:1 mixture of Ni(II) and 1a was carried out at 21.46 V in
DMF, in a two-compartment cell in the absence of other
metal ions. The reaction was run up to the consumption of
2 F mol21 of 1a, and 2a was obtained in 36% yield. A
similar but catalytic controlled-potential electrolysis with a
1:10 molar ratio of Ni(II) to 1a at 21.75 V led to 32% of 2a.
The same catalytic electrolysis in the presence of Mg2þ ions
(one equiv. vs 1a) led to 85% of 2a after 2 F mol21. These
results indicate that the presence of Mg2þ ions facilitate the
recycling of the nickel species and enhance the overall
efficiency of the catalytic reaction.
1d
See Section 4.
a
Recovered 1b, 9%.
(entries 7 and 8, Table 2). Thus, compounds 1e and 1f could
be allylated regioselectively. In the case of 1e, only the
branched 2e was obtained in 40% yield, together with non-
allylated 5. In contrast, the electrolysis of 1f led to linear 3f
and to 5 in 19 and 37% yields, respectively. This result is in
sharp contrast with the regioselectivity obtained in the case
of 1c. A high regioselectivity towards the linear isomer in
the case of cinnamyl derivatives has already been reported
for the coupling of diethyl ketone and cinnamyl chloride in
the presence of a Zn anode.11 On the other hand, the crotyl
or the prenyl zinc reagents when reacting with aliphatic or
aromatic aldehydes form almost exclusively the branched
alcohol.2
2.4. Mechanistic aspects
From a mechanistic point of view, chemical, electrochemi-
cal and theoretical studies on the nickel-catalysed cleavage
of allyl ethers in substrates such as 1a, with a further
intramolecular allyl transfer process, have been recently
described in the presence of a Mg anode.14 However, no
indication on the role of the anodically generated metal
cations (e.g. Mg2þ) in the catalytic cycle has been reported.
The Mg2þ ions were proposed be involved in a transme-
tallation step after the allylation, thus enabling the recycling
of the nickel species.14 The experimental results presented
here, concerning the effect of metal ions on the regio-
selectivity of the allyl transfer, clearly indicate that these
different cations have to be present and play an important
role in the allyl transfer step.
2.3. Allylation in the presence of Al31 ions
The electrolyses of 1a–1d were also carried out in the
presence of a consumable Al anode, giving rise to Al3þ ions
in solution. The results are shown in Table 3. The allyl
transfer from 1a to 2a was efficient, affording 90% yield of
2a (entry 1), although passivation of the Al anode could
occur in some cases. The reaction of 1b with an Al rod as the
anode led to a 39:61 ratio of branched to linear isomers in
61% yield. Unexpectedly, the presence of Al3þ ions
reversed the regioselectivity as compared to the effect of
Mg2þ or Zn2þ ions for the same substrate.
Cyclic voltammetry of the Ni(II) system was carried out and
revealed that the Ni(II)/Ni(0) reduction step occurs
reversibly at 21.2 V vs SCE in a DMF/n-Bu4BF4 solution.
The direct reduction of substrates 1 in the absence of the
Ni(II) complex takes place beyond 22.0 V. This reduction
potential is in agreement with the observed low reduction
potential of some functionalised allyl ethers (22.4 V)15 or
that of benzyl ethers (22.3 V).16 The addition of 1a to the
Ni(II) solution caused irreversibility of the NI(II)/Ni(0)
reduction peak and a further one-electron reduction peak
appeared, among other changes, at 21.4 V.
With substrate 1c the yield of intramolecular allylation was
of 40% with a branched to linear ratio of 62:38, and 30% of
4 were also obtained. The same reaction with 1d allowed a
very selective allylation process (99% regioselectivity) in
40% yield, with 22% of 4.
Intramolecular allyl transfer to ketones was not efficient
with an Al anode. Unless for 1a, the reactions carried out
with an Al anode were less efficient from the point of view
of the allylation yields. The regioselectivities were very
dependent on the nature of the substrate. Whereas substrate
1d afforded almost exclusively the branched regioisomer in
the presence or either Mg2þ, Zn2þ or Al3þ ions, with 1b and
1c the results were influenced by the different ions, even
allowing for a reversal of the regioselectivity in the case of
According to these data, we propose the mechanism as
shown in Scheme 1. The active complex Ni(0) species
formed by a two-electron reduction of the Ni(II)(bipy)32þ
complex react with substrate 1 to form a (p-allyl)Ni(II)
intermediate, A. Intermediate A is further reduced by one
electron (at 21.4 V) to afford a (p-allyl)Ni(I)2 complex,
1b and Al3þ
.

9294
D. Franco, E. Dun˜ach / Tetrahedron 58 (2002) 9289–9296
Scheme 1.
B.14 A transmetallation is proposed to occur between the
anodically generated species M2þ (Zn2þ, Mg2þ or Al3þ
and could even reverse the ratio of branched to linear
isomers. The reaction was very sensitive to the nature of the
allyl substituents and thus, the crotyl group could be
transferred regiospecifically under the different conditions
tested.
)
and B to give C. This transmetallation is proposed before
the intramolecular allylation of the carbonyl fragment
occurs. The Ni(I)þ species is liberated and can regenerate
the active catalytic species Ni(0) after one-electron electro-
reduction. The (p-allyl)M2þ(phenate) intermediate C
undergoes allyl transfer to the carbonyl group to give the
branched and linear homoallylic alcoholate-phenates D-1
and D-2, respectively. The final hydrolysis step affords the
regioisomers 2 and 3. The possibility of a mixed mechanism
in which a bimetallic Ni(I)/M2þ species would be involved
in the allyl transfer step cannot be ruled out.
Our results indicated on one hand that the presence of metal
ions favoured the efficiency of the reaction and the
regeneration of the catalyst. However, concerning the
regioselective outcome of the allyl transfer, the reactions
run in the absence of metal ions afforded the highest
regioselectivity.
4. Experimental
3. Conclusions
4.1. General procedure for one-compartment cell
electrolyses
In conclusion, the nickel-catalysed intramolecular transfer
of allyl groups from substituted allyl ethers to carbonyl
compounds affords regioselectively the branched isomer in
almost all cases. The best regioselectivities were obtained
without added metal ions or in the presence of Zn2þ. The
presence of Al3þ afforded the lower yields and selectivities,
In a glass cell such as described in Ref. 17 were introduced
20 ml of freshly distilled DMF, n-Bu4NþBF42 (1022 M),
[Ni(bipy)3]2þ, 2BF42 (0.1 mmol) and the substrate 1
(1 mmol),
prepared
from
the
corresponding

D. Franco, E. Dun˜ach / Tetrahedron 58 (2002) 9289–9296
9295
ortho-hydroxybenzaldehyde or ortho-hydroxyacetophenone
by stirring at 508C with the corresponding substituted allyl
bromide and potassium carbonate in DMF. The solution was
stirred at room temperature and electrolysed at constant
current of 60 mA (current density of 0.2 A dm22 and
5–15 V between the rod anode (Mg, Zn or Al electrode)
and a stainless steel cathode, up to the total consumption of
the starting material (checked by GC analysis of aliquots),
unless electrode passivation occurred. After evaporation of
the DMF under vacuum, the crude mixture was hydrolysed
with HCl 0.1 M saturated with NaCl, up to pH 1–2 and
extracted with Et2O. The organic layers, dried over MgSO4,
were filtered off and evaporated. The major products 2 and 3
were purified by SiO2 column chromatography with
pentane–ether (90/10) as eluent and analysed by NMR,
mass spectroscopy and IR.
4.2.4. 2-(10-Hydroxy-40-methylpent-30-enyl)phenol, 3b.
1H NMR (CDCl3, 200 MHz): 7.68 (1H; s large); 7.20–
6.66 (4H; m); 6.05 (1H; d; J¼9.0 Hz); 4.93 (1H; dd; J¼7.0,
7.0 Hz); 4.75 (1H; s large); 4.64 (1H; s large); 2.50 (2H; dd;
J¼9.0, 7.0 Hz); 1.72 (3H; s); 1.61 (3H; s). 13C NMR
(CDCl3, 50.3 MHz): 167.9; 142.2; 132.6; 129.7; 128.4;
117.4; 113.8; 111.6; 82.9; 47.9; 38.9; 30.5. MS: m/z
192(Mþ); 174; 157; 145; 122(100%); 104; 77; 65; 51; 39.
0
0
0
1
4.2.5. 2-(1 -Hydroxy-2 -phenylbut-3 -enol)phenol, 2c. H
NMR (CDCl3, 200 MHz): 7.92 (1H; s large); 7.85–6.8 (9H;
m); 5.91 (1H; ddd; J¼17.0, 10.2, 7.8 Hz); 5.02 (1H; dd;
J¼10.2, 1.4 Hz); 4.88 (1H; dd; J¼17.0, 1.4 Hz); 4.82 (1H;
d; J¼17.0 Hz); 3.83 (1H; dd; J¼8.6, 8.2 Hz). 13C NMR
(CDCl3, 50.3 MHz): 155.6; 141.1; 137.9; 129.2; 129.1;
129.0; 128.7; 128.5; 127.4; 126.8; 126.6; 119.9; 117.6;
117.2; 78.2; 57.1. MS: m/z 222(Mþ218); 145; 115; 91; 77;
63; 51; 39(100%).
4.2. General procedure for two-compartment cell
electrolyses
4.2.6. 2-(20-Phenylbut-10,30-diene)phenol (dehydrated
form of 2c). 1H NMR (CDCl3, 200 MHz): 7.98 (1H; s
large); 7.85–6.8 (9H; m); 6.45 (1H; s); 6.40 (1H; m); 5.0
(1H; dd; J¼10.2, 1.5 Hz); 4.92 (1H; dd; J¼17.2, 1.5 Hz).
13C NMR (CDCl3, 50.3 MHz): 163.6; 155.4; 141.1; 137.6;
133.5; 129.2; 129.1; 129.0; 128.7; 128.5; 127.4; 126.8;
126.6; 119.9; 117.6; 117.2. MS: m/z 222(Mþ); 145; 115; 91;
77; 63; 51; 39(100%).
Both compartments were filled with a DMF solution (50 ml
each) of n-Bu4NþBF42 (1 g, 3 mmol) under inert atmos-
phere. The Ni(II) complex (0.1 mmol) and 1b (0.1 mmol)
were added to the cathodic compartment. The electrolyses
were run at 208C at the desired controlled potential (21.3 V
vs SCE) and was stopped when the current was negligible.
The work-up was the same as described above, the reaction
being followed by GC.
0
0
0
1
4.2.7. 2-(1 -Hydroxy-4 -phenylbut-3 -enol)phenol, 3c. H
NMR (CDCl3, 200 MHz): 8.28 (1H; s large); 7.85–6.8 (9H;
m); 6.57 (1H; d; J¼17.0 Hz); 6.41–6.12 (1H; m); 4.92 (1H;
dd; J¼7.1, 7.1 Hz); 3.55 (1H; s); 2.75 (2H; m). 13C NMR
(CDCl3, 50.3 MHz): 155.3; 141.0; 138.0; 137.8; 130.8;
130.3; 129.1; 127.6; 127.0; 125.8; 120.3; 119.7; 118.7;
117.0; 74.5; 57.3. MS: m/z 222(Mþ218); 194; 165; 145;
128; 107; 91; 77; 63; 51; 39(100%).
4.2.1. 2-(10-Hydroxybut-30-enyl)phenol, 2a. To facilitate
analysis, 2a was also converted into its methyl ether by
treatment of the crude mixture with methyl iodide. 1H NMR
(CDCl3, 200 MHz; relative integration, multiplicity,
coupling constants in Hz): 7.90 (1H; s); 7.10 (1H; ddd;
J¼16.3, 1.7, 1.7 Hz); 6.90 (1H; dd; J¼16.8, 1.5 Hz); 6.85–
6.70 (2H; m); 5.84 (1H; dddd; J¼17.1, 10.5, 7.1, 7.1 Hz);
5.3 (1H; dd; J¼10.9, 1.1 Hz); 5.25 (1H; J¼17.1, 1.1 Hz);
4.87 (1H; dd; J¼7.5, 7.5 Hz); 2.90 (1H; s); 2.60 (2H; dd;
J¼7.6, 7.7 Hz). 13C NMR (CDCl3, 50.3 MHz): 155.5;
134.0; 129.0; 127.2; 126.6; 119.9; 119.3; 117.3; 74.7;
42.2. MS: m/z 164(Mþ); 146; 131; 121; 107; 91; 77; 65;
43(100%). IR (KBr): 3366; 3071; 1235; 755 cm21. HRMS
calcd for C10H12O2: 164.083730; found: 164.083091.
4.2.8. 2-(10-Hydroxy-20-methylbut-30-enol)phenol, 2d
1
(mixture of diastereomers 54:45). First diastereomer: H
NMR (CDCl3, 200 MHz): 9.20 (1H; s); 7.20–6.65 (4H; m);
5.92–5.71 (1H; ddd; J¼16.6, 10.1, 6.8 Hz); 5.28 (1H; d;
J¼10.1 Hz); 5.09 (1H; d; J¼16.6 Hz); 4.78 (1H; d;
J¼5.4 Hz); 2.68 (1H; qd; J¼6.8, 5.4 Hz); 1.45 (1H; s);
1.10 (3H; d; J¼6.8 Hz). 13C NMR (CDCl3, 50.3 MHz):
156.1; 140.4; 129.0; 126.8; 126.3; 120.5; 119.6; 117.8; 77.2;
44.8; 16.9. MS: m/z 160(Mþ218); 145; 131; 123 (100%);
115; 107; 95; 77; 65; 51; 39; 27; 18.
4.2.2. 2-(20,20-Dimethyl-10-hydroxybut-30-enyl)phenol,
2b. 1H NMR (CDCl3, 200 MHz): 7.9 (1H; s); 7.9–7.7
(3H; m); 7.1 (1H; ddd; J¼8, 2.1 Hz); 5.9–5.75 (1H; dd;
J¼13, 10 Hz); 5.3 (1H; dd; J¼13, 1.1 Hz); 5.25 (1H; dd;
J¼10, 1.1 Hz); 4.5 (1H; s); 3.1 (1H; s); 1.1 (3H; s); 1.0 (3H;
s). 13C NMR (CDCl3, 50.3 MHz): 156.6; 144.9; 130.0;
129.0; 128.9; 122.1; 118.7; 117.5; 115.2; 92.4; 83.0; 24.7;
20.2. MS: m/z 174(Mþ218); 159; 144; 123; 95; 77; 65; 51;
39(100%). IR (KBr): 3363; 2963; 2931; 2874; 1499; 1456;
1
Second diastereomer: H NMR (CDCl3, 200 MHz): 9.20
(1H; s); 7.20–6.65 (4H; m); 5.92–5.71 (1H; ddd; J¼16.6,
10.1, 4.7 Hz); 5.28 (1H; d; J¼10.1 Hz); 5.09 (1H; d;
J¼16.6 Hz); 4.49 (1H; d; J¼8.5 Hz); 2.63 (1H; qd; J¼8.5,
4.7 Hz); 1.45 (1H; s); 0.90 (3H; d; J¼4.7 Hz). 13C NMR
(CDCl3, 50.3 MHz): 155.8; 140.2; 128.9; 128.4; 125.3;
128.1; 119.6; 117.3; 79.3; 44.5; 13.8. MS: m/z 178 (Mþ);
160; 145; 131; 123 (100%); 115; 105; 95; 77; 65; 51.
1288; 1246; 754 cm21
.
4.2.3. 2-(20,20-Dimethyl-10-hydroxybutyl)phenol (reduced
1
form of 2b). H NMR (CDCl3, 200 MHz): 8.38 (1H; s
large); 7.26–6.70 (4H; m); 4.65 (1H; s); 1.43 (2H; q;
7.5); 0.95 (3H; s); 0.90 (3H; s); 0.88 (3H; t; 7.5). 13C
NMR (CDCl3, 50.33 MHz): 129.7; 129.4; 128.8; 120.0;
118.7; 117.4; 84.0; 40.0; 30.7; 22.6; 22.3; 8.3. MS: m/z
194(Mþ); 176; 161; 147; 123(100%); 107; 91; 77; 65; 51;
43; 27.
4.2.9. 2-(10-Hydroxy-10-methyl-2-20-dimethyl-30-but-
1
enyl)phenol, 2e. H NMR (CDCl3, 200 MHz): 9.59 (1H; s
large); 7.16–7.12 (1H; m); 6.99–6.95 (1H; m); 6.79–6.85
(2H; m); 6.02 (1H; dd; J¼17.4, 10.9 Hz); 5.21 (1H; dd;
J¼10.9, 1.3 Hz); 5.16 (1H; dd; J¼17.4, 1.3 Hz); 2.75 (1H; s
large); 1.65 (3H; s); 1.08 (3H; s); 1.07 (3H; s). 13C NMR

9296
D. Franco, E. Dun˜ach / Tetrahedron 58 (2002) 9289–9296
6. Yanagisawa, A.; Habaue, S.; Yamamoto, H. J. Am. Chem. Soc.
1991, 113, 8955–8956.
(CDCl3, 50.3 MHz): 156.8; 144.1; 129.2; 128.9; 126.9;
118.4; 117.6; 115.1; 83.3; 46.3.
7. Yanagisawa, A.; Yamada, Y.; Yamamoto, H. Synlett 1997,
1090–1092.
0
0
1
4.2.10. 2-(Buta-1 ,3 -dienyl)phenol, 6a. H NMR (CDCl3,
200 MHz): relative integration, multiplicity, coupling con-
stants in Hz) 8.80 (1H; s large); 7.85–6.8 (4H; m); 6.60 (1H;
d; 17.1); 6.26 (1H; ddd; J¼17.0, 10.2, 7.1 Hz); 5.44 (1H; dd;
J¼17.1, 7.1 Hz); 5.24 (1H; dd; J¼17.0, 1.5 Hz); 5.07 (1H;
dd; J¼10.2, 1.5 Hz). 13C NMR (CDCl3, 50.3 MHz): 163.5;
155.5; 134.0; 129.0; 127.2; 126.6; 119.9; 119.3; 117.3;
111.8. MS: m/z 146(Mþ); 133(100%); 115; 105; 91; 77; 65;
51.
8. Guo, B. S.; Doubleday, W.; Cohen, T. J. Am. Chem. Soc. 1987,
109, 4710–4711. (b) Cohen, T.; Bhupathy, M. Acc. Chem.
Res. 1989, 22, 152–161.
9. Tokuda, M.; Satoh, S.; Suginome, H. J. Org. Chem. 1989, 54,
5608–5613.
10. Sibille, S.; D’Incan, E.; Leport, L.; Messebiau, M. C.;
´
Perichon, J. Tetrahedron Lett. 1987, 28, 55–58.
´
11. Durandetti, S.; Sibille, S.; Perichon, J. J. Org. Chem. 1989, 54,
2198–2204.
12. Franco, D.; Olivero, S.; Dun˜ach, E. Electrochim. Acta 1997,
42, 2159–2164.
References
´
13. Dun˜ach, E.; Perichon, J. J. Organomet. Chem. 1988, 352,
239–246.
14. Franco, D.; Wenger, K.; Antonczak, S.; Cabrol-Bass, D.;
1. Yamamoto, Y.; Asao, N. Chem. Rev. 1993, 93, 2207–2293.
2. Courtois, G.; Miginiac, P. J. Organomet. Chem. 1974, 69,
1–44.
`
Dun˜ach, E.; Rocamora, M.; Gomez, M.; Muller, G. Chem.
Eur. J. 2002, 8, 664–672.
3. Cohen, T.; Guo, B.-S. Tetrahedron 1986, 42, 2803–2808.
4. Araki, S.; Itoh, H.; Butsugan, Y. J. Organomet. Chem. 1988,
347, 5–10.
15. Bhuvaneswari, N.; Venkatachalamand, C. S.; Balasubrama-
nian, K. K. J. Chem. Soc., Chem. Commun. 1994, 1177–1178.
16. Santiago, E.; Simonet, J. Electrochim. Acta 1975, 20,
853–858.
5. Keck, G. E.; Abbott, D. E.; Boden, E. P.; Enholm, E. J.
Tetrahedron Lett. 1984, 25, 3927–3930. (b) Yamamoto, Y.;
Maruyama, K.; Matsumoto, K. J. Chem. Soc., Chem. Commun.
1983, 489–490.
´ ´ ´
17. Chaussard, J.; Folest, J. C.; Nedelec, J. Y.; Perichon, J.; Sibille,
S.; Troupel, M. Synthesis 1990, 369–387.