Some Observations on the Mechanism of Diorganocuprate 1,4-Addition Reactions with α,β-Unsaturated Ketones: Effects of Diethyl Ether in Reactions of Butylcoppers in Toluene

Article abstract of DOI:10.1021/jo970316y

Mixtures of butyllithium and copper iodide prepared in toluene react with α,β-unsaturated ketones predominantly in a 1,2-fashion. Addition of 2 equiv of diethyl ether to the system results in a dramatic preference for the 1,4-product, as found normally with diorganocuprates in ethereal solvents. These results are in accordance with the theoretical predictions that a suitable ligand is required to facilitate the stabilization of a formally copper(III) reaction intermediate on the 1,4-addition mechanistic pathway.

Full text of DOI:10.1021/jo970316y

J . Org. Chem. 1997, 62, 4629-4634
4629
Som e Obser va tion s on th e Mech a n ism of Dior ga n ocu p r a te
1,4-Ad d ition Rea ction s w ith r,â-Un sa tu r a ted Keton es: Effects of
Dieth yl Eth er in Rea ction s of Bu tylcop p er s in Tolu en e
Celia L. Kingsbury and Robin A. J . Smith*
Department of Chemistry, University of Otago, P.O. Box 56, Dunedin, New Zealand
Received February 19, 1997X
Mixtures of butyllithium and copper iodide prepared in toluene react with R,â-unsaturated ketones
predominantly in a 1,2-fashion. Addition of 2 equiv of diethyl ether to the system results in a
dramatic preference for the 1,4-product, as found normally with diorganocuprates in ethereal
solvents. These results are in accordance with the theoretical predictions that a suitable ligand is
required to facilitate the stabilization of a formally copper(III) reaction intermediate on the 1,4-
addition mechanistic pathway.
Sch em e 1
In tr od u ction
The 1,4-addition (conjugate) reaction of lithium dior-
ganocuprates (R2CuLi) with R,â-unsaturated ketones
(enones) is a well established synthetic protocol1 for
creating a carbon-carbon σ bond relatively remote from
the carbonyl group. Despite substantial efforts dating
from 1965,2 the detailed mechanism of the organocuprate
1,4-addition reaction remains uncertain. Indeed, a gen-
eralized reaction mechanism may not be possible as even
slight modification of the reaction conditions (e.g. solvent,
temperature, additives) can significantly influence the
reaction outcome. Mechanistic investigations are made
even more difficult by the absence of structural knowl-
edge of the reacting species in solution. Nonetheless, the
current belief is that the 1,4-addition reaction proceeds
via a lithium-coordinated enone (1) to a cuprate(3d) f
alkene(π*) complex (2) and finally to an enolate (4),3-7
as illustrated in Scheme 1.
The intermediates 1, 2, and 4 have been directly
observed and characterized by low temperature NMR
studies.4,5 The conversion of 2 to 4 has been proposed to
proceed via a formal copper(III) intermediate species
(3).7-9 This situation arises from the incorporation of
three covalently bound anionic organic groups about the
copper center. To the best of our knowledge, direct
observation of 3 has not been achieved and some skepti-
cism has been expressed about the viability of a copper-
(III) intermediate.10
and found a local energy minimum provided the copper
center was stabilized by donors such as water or am-
monia. These donors were used as model ligands for
reactions performed in the presence of ethereal solvents
and/or nitrogen ligands. The energy changes involved
in the fragmentation of the oxygen-coordinated trimeth-
ylcopper species to methylcopper and ethane were also
estimated, and a metallocyclopropane was found to be
an energetically viable transition state. It was concluded
that the oxygen-coordinated trimethylcopper species was
kinetically stable although thermodynamically much less
stable than the reaction products. Snyder12 has recently
modeled a series of similar compounds viewed as “key
parts of the potential energy surface” for the 1,4-addition
reaction. It was found that without solvent (e.g. dimethyl
ether) stabilization, a trialkyl copper species had little
or no independent existence and was therefore presum-
ably unlikely to serve as an intermediate. However, with
appropriate ligand support a viable intermediate was
located. As a result of these studies an overall mecha-
nism was proposed for the 1,4-addition reaction which
involves, at one stage, solvent stabilization of the copper-
(III) species 3, e.g. as 5, which subsequently undergoes
facile reductive elimination.
Schleyer et. al.11 have carried out ab initio calculations
on trimethylcopper, Cu(CH3)3, as a simple model for 3,
X Abstract published in Advance ACS Abstracts, J une 15, 1997.
(1) Recent reviews: Perlmutter, P. Conjugate Addition Reactions
in Organic Synthesis; Pergamon Press: Oxford, 1992. Lipshutz, B. H.;
Sengupta, S. Org. React. 1992, 41, 135-631. Kozlowski, J . A. In
Comprehensive Organic Chemistry; Trost, B. M., Fleming, I., Eds.;
Pergamon Press: Oxford, 1991; Vol. 4, Chapter 1.4. Chapdelaine, M.
J .; Hulce, M. Org. React. 1990, 38, 225-653.
(2) House, H. O.; Respess, W. L.; Whitesides, G. M. J . Org. Chem.
1966, 31, 3128-3141.
(3) Hallnemo, G.; Olsson, T.; Ullenius, C. J . Organomet. Chem. 1984,
265, C22. Hallnemo, G.; Olsson, T.; Ullenius, C. J . Organomet. Chem.
1985, 282, 133-144.
(4) Bertz, S. H.; Smith, R. A. J . J . Am. Chem. Soc. 1989, 111, 8276-
8277.
The notion of considering intermediates 3 and 5 as
formal copper(III) species has been discussed13 and
(5) Vellekoop, A. S.; Smith, R. A. J . J . Am. Chem. Soc. 1994, 116,
2902-2913.
(10) e.g. Collman, J . P.; Hegedus, L. S. Principles and Applications
of Organotransition Metal Chemistry; University Science Books: Mill
Valley, California, 1980.
(11) Dorigo, A. E.; Wanner, J .; Schleyer, P. von R. Angew. Chem.,
Int. Ed. Engl. 1995, 34, 476-478.
(12) Snyder, J . P. J . Am. Chem. Soc. 1995, 117, 11025-11026.
(6) House, H. O.; Fischer, W. F. J . Org. Chem. 1968, 33, 949-956.
(7) Krauss, S. R.; Smith, S. G. J . Am. Chem. Soc. 1981, 103, 141-
148.
(8) Casey, C. P.; Cesa, M. J . Am. Chem. Soc. 1979, 101, 4236-4244.
(9) Corey, E. J .; Boaz, N. Tetrahedron Lett. 1985, 26, 9015-6018.
S0022-3263(97)00316-2 CCC: $14.00 © 1997 American Chemical Society

4630 J . Org. Chem., Vol. 62, No. 14, 1997
Kingsbury and Smith
alternative views such as “copper(I) complexes whose
bonds are dominated by ionic contributions, but whose
chromatogram of 7 showed a series of up to five peaks
assumed to be the various isomeric dehydration products.
The relative amount of 7 was derived by the summation
of the integrals of all these peaks.
structures are determined by a nonvanishing degree of
12
2
2
covalent Cu(dx -y ) ligand binding” have been expressed.
This paper reports experimental findings from some
reactions which relate to this aspect of the mechanism.
The focus of this research was to prepare an organocopper
species in a hydrocarbon solvent and to investigate its
reactivity with enones using stoichiometric amounts of
the typical, heteroatom donor diethyl ether (Et2O). The
amount of 1,4-addition product was expected to be an
indicator of cuprate activity. An alternative, (1,2-) ad-
dition, product was also anticipated and this could be
derived from 1 and/or 2 without evoking a copper(III)
intermediate.
Addition of 2 equiv of BuLi to a grey suspension of CuI
in toluene at -78 °C gave no apparent reaction. On
warming to -40 °C, the temperature at which Bu2CuLi
was formed in Et2O solvent, no change was noted. It was
found that reaction at -20 °C for 15 min was necessary
to create a visible change in the toluene solvent system.
Addition of 6 (1.0 equiv) to this black mixture recooled
to -78 °C and reaction at -40 °C for 30 min resulted in
complete consumption of the starting material. The
product was predominantly 7 (98 mol %) with minor
amounts (2 mol %) of 8. Numerous repetitions of this
experiment gave similar results and confirmed that the
1,4-/1,2-addition product ratio in toluene was completely
different to that in Et2O.
A series of experiments adding various amounts of
Et2O to the 1:2 CuI/BuLi mixture in toluene were then
carried out using a standard protocol of stoichiometry,
time, temperature, and order of addition (see Experi-
mental Section). Only a slight excess (10%) of the
organometallic over enone was required to consume all
starting material. A slight excess of CuI to BuLi was
used in these reactions to allow for all free BuLi to be
reacted. The amount of Et2O used in these studies is
expressed in terms of molar equivalents (ME). The basic
ME unit is equal to half the amount of BuLi used in each
experiment. The results from varying the amount of
Et2O are presented in Figure 1.
Lithium dimethylcuprate could not be used for this
work, as it is not possible to remove the final traces of
ethereal solvents from the cuprate14 nor the methyl-
lithium precursor.15 For these reasons the copper species
examined was based on butyllithium, which is com-
mercially available in a hexane solution. Toluene has
been used previously as a solvent component in lithium
dimethylcuprate reactions16 (necessarily including Et2O)
and was utilized as the solvent in these studies. 4,4a,
5,6,7,8-Hexahydro-4a-methyl-2(3H)-naphthalenone (6) was
chosen as the primary reaction substrate as the organo-
cuprate reactivity,4,17 and electrochemistry18 has been
thoroughly examined previously in these laboratories.
Resu lts a n d Discu ssion
Butyl addition to 6 gives the 1,2- and the 1,4-addition
products, 7 and 8 (Scheme 2). Reference samples of 7
Sch em e 2
and 8 were prepared using standard procedures. Com-
pound 7 was obtained by the addition of 6 to BuLi
followed by careful isolation to avoid dehydration. The
1,4-addition product 8 was synthesized, in 89% isolated
yield, by reaction of lithium dibutylcuprate (Bu2CuLi)
F igu r e 1. Effect of Et2O on the amount of 1,4-/1,2-addition
to 6 with 1:2 CuLi/BuLi.
Reaction of the 1:2:1 CuI/BuLi/Et2O mixture in toluene
with 6 consistently gave approximately 7 mol % 8, i.e.
comparable to that obtained without any Et2O. However,
reaction using 2 ME of Et2O gave 97 mol % 8, i.e. similar
to that observed with Et2O as the solvent. There was a
dramatic, progressive increase in the amount of 8 pro-
duced from reactions carried out with between 1 and 2
ME of Et2O. Increasing the amount of Et2O above 2 ME
had no deleterious effects on the amount of 8 produced.
Other enones were reacted with the 1:2 CuI/BuLi
mixture in toluene to establish whether the Et2O effect
on the 1,4-/1,2-addition ratio was unique to the sterically
and electronically demanding substrate 6. The results
with 2-cyclohexen-1-one (9), 3,5,5-trimethyl-2-cyclohexen-
1-one (10), and trans-4-phenyl-3-buten-2-one (11) are
presented in Table 1, together with those from similar
reactions with 6 for comparison purposes.
1
with 6 in Et2O. At room temperature, the H NMR and
13C NMR spectra of 8 showed many broad resonances;
however, spectra obtained at 90 °C showed significantly
sharper resonances. This phenomenon is indicative of
the conformational mobility associated with the labile,
cis-fused, dialkyl ring junction.19 Gas chromatography
(GC) has been employed for determining the yields of
methyl addition products from 6,20 and an adaptation of
this methodology was used in this work. The GC
(13) Snyder, J . P. Angew. Chem., Int. Ed. Engl. 1995, 34, 80-81.
(14) Besace, V.; Berlan, J .; Pourcelot, G.; Huche, M. J . Organometal.
Chem. 1983, 247, C11-C13.
(15) Wakefield, B. J . The Chemistry of Organolithium Compounds;
Pergamon Press: New York, 1974; pp 4-10, and references therein.
(16) Hallnemo, G.; Ullenius, C. Tetrahedron 1983, 39, 1621-1625.
Posner, G. H. Org. React. 1972, 19, 56 cf. ref 51.
(17) Smith, R. A. J .; Vellekoop, A. S. Tetrahedron 1989, 45, 517-
522.
The 1,2-addition product was predominant in all reac-
tions without Et2O present. Only in the case with the
very reactive enone, 9, was a significant amount of 1,4-
addition observed. In contrast to 6, reactions with 9, 10,
or 11 and the 1:2:1 CuI/BuLi/Et2O mixture gave signifi-
(18) Smith, R. A. J .; Hannah, D. J . Tetrahedron Lett. 1980, 21,
1081-1084.
(19) Marshall, J . A.; Fanta, W. I.; Roebke, H. J . Org. Chem. 1966,
31, 1016-1020. Marshall, J . A.; Roebke, H. J . Org. Chem. 1968, 33,
840-843.
(20) Bertz, S. H.; Smith, R. A. J . Tetrahedron 1990, 46, 4091-4100.

Mechanism of Diorganocuprate 1,4-Addition Reactions
J . Org. Chem., Vol. 62, No. 14, 1997 4631
Ta ble 1. Effect of Et2O on 1,4-/1,2-Ad d ition Ra tios w ith
Va r iou s En on es
obtained. This result eliminates the possiblity of signifi-
cant direct reaction between BuLi and toluene under
these reaction conditions. The small amount of recovered
starting material 6 may be the result of enolate formation
or simply reflect the precision of BuLi delivery. Exami-
nation of the 1:1 CuI/BuLi system was undertaken to
allow for the formation of butylcopper (BuCu) and
resulted in a dark mixture indicating at least some
interaction between the two reagents. Reaction of this
material with 6 (entry 2) gave predominantly 7 together
with 6 and a small amount of 8. By comparison with
the BuLi reaction (entry 1) it appeared that some
modification of the BuLi by the CuI had occurred which
manifested itself in the form of less 7 and some, albeit
minor, 1,4-addition. Reaction of the 1:2 CuI/BuLi mix-
ture with 6 (entry 3) gave a product distribution compa-
rable to that obtained with BuLi, except a detectable
amount of 8 was evident. This reaction was repeated but
with 2 equiv of 6. The result (entry 4) indicated that at
least 88% of the butyl units were active, with a strong
1,2-addition preference. These reactivity experiments
indicate that the black CuI/2BuLi/toluene mixture reacts
as if it consists of varying amounts of BuCu and BuLi.
A similar set of reactions were run but with fixed
amounts of Et2O and are summarized in Table 3.
products (%)a
enone
w (ME)
1,2-addition
1,4-addition
6
9
10
11
6
0
0
0
0
1
1
1
1
2
2
2
2
98b
66
95
90
93b
27
78
73
3b
1
2b
34
5
10
7b
9
73
22
27
97b
99
95
95
10
11
6
9
10
11
5
5
a
b
Obtained by 1H NMR unless stated otherwise. From GC
analysis.
cant amounts of 1,4-addition, particularly with 9. The
variation of 1,4-/1,2-addition ratio with enone using 1 ME
of Et2O is consistent with the enone/methylcuprate
relative reactivity series reported previously.5 Over-
whelming preference for the 1,4-addition product was
observed in all reactions containing 2 ME of Et2O. These
results clearly indicate the generality of the stoichiomet-
ric effect of Et2O on the butylcopper system in toluene.
There are two possible segments in the proposed
reaction where the presence of Et2O could be influential
on the 1,4-addition proclivity. The ligand may be neces-
sary to allow the formation of the cuprate reagent and
also, as discussed above, be required for intermediate
stabilization. The large amounts of 1,2-addition observed
in the toluene system with small amounts of Et2O may
indicate that, under these reaction conditions, either the
cuprate species was formed but was 1,2-directing or,
alternatively, free BuLi was present. In a typical ethe-
real diorganocuprate reaction system only one of the alkyl
groups is transferred to the organic substrate and the
second alkyl group is generally unavailable, presumably
as an alkylcopper(RCu) species. Alkylcoppers are con-
sidered to be unreactive, particularly in Et2O, and even
in THF require the assistance of Lewis acids to become
effective alkyl donors.21 Hence, normally the addition of
2 ME of enone to a cuprate gives equal amounts of the
1,4-addition product and recovered starting material. In
order to establish the amount of active butyl group in
CuI/BuLi mixtures in toluene a series of experiments
with different ratios of BuLi, CuI, and 6 were carried out,
and the results are presented in Table 2.
Ta ble 3. Rea ction of Va r iou s Cu I/Bu Li/Et2O Mixtu r es
w ith 6 in Tolu en e
amount
(mmol)
products (mol %)
w
(ME)
entry
x
y
z
6
7
8
1
2
3
1.0 1.0 1.0
1.0 2.0 1.0
1.0 2.0 1.5
5
1
1
95
0
2
4
95
85
1
5
13
(0.03 mmol) (1.28 mmol) (0.19 mmol)
4
5
1.0 2.0 1.0
1.0 2.0 1.5
5
5
1
33
3
6
96
61
(0.49 mmol) (0.09 mmol) (0.92 mmol)
Reaction of 6 with 1:1:5 CuI/BuLi/Et2O gave predomi-
nantly starting material, together with small amounts
of both 7 and 8 (entry 1). This result is in concordance
with established alkylcopper reactivity without Lewis
acid activation and contrasts to the previous result in
toluene alone (Table 2, entry 2). Two experiments were
conducted using the 1:2:1 CuI/BuLi/Et2O mixture. As
shown in Figure 1, reaction with 1 equiv of 6 gave
predominantly 7 (entry 2). The addition of 1.5 equiv of
6 (entry 3) gave a product distribution comparable to that
obtained without Et2O (Table 2, entry 4) with ca. 74% of
the butyl groups being transferred. A substantial amount
of 8 was observed from the reaction of 6 with the 1:2:5
CuI/BuLi/Et2O mixture (entry 4). This result was an-
ticipated based on the previous work (cf. Figure 1) which
showed high 1,4-selectivity with these amounts of Et2O.
The total amount of active butyl groups available in this
reaction was determined by using 1.5 equiv of 6 (entry
5). The recovery of 0.5 equiv of 6 along with high
amounts of 8 is consistent with a “normal” cuprate. From
these studies it can be concluded that traditional butyl-
copper (BuCu, Bu2CuLi) properties can be established in
toluene provided a certain minimum amount of Et2O is
also present.
Ta ble 2. Rea ction of Va r iou s Cu I/Bu Li Mixtu r es w ith 6
in Tolu en e
amount (mmol)
products (mol %)
entry
x
y
z
6
7
8
1
2
3
4
0.0 1.0 1.0
1.0 1.0 1.0
1.0 2.0 1.0
1.0 2.0 2.0
3
14
1
97
79
98
86
0
7
1
2
12
(0.24 mmol) (1.72 mmol) (0.04 mmol)
Reaction of BuLi with 6 (entry 1) was included as a
reference experiment, and a large amount of 7 was
It became clearly important to establish the nature of
the species obtained by mixing CuI and BuLi in toluene,
and a method for distinguishing organolithium from
organocopper reactivity was required to clarify this issue.
(21) Matsuzawa, S.; Horiguchi, Y.; Nakamura, E.; Kuwajima, I.
Tetrahedron 1989, 45, 349-362.

4632 J . Org. Chem., Vol. 62, No. 14, 1997
Kingsbury and Smith
Early mechanistic research on the 1,4-addition reaction
tended to focus on an electron transfer process, and
molecules similar to those previously utilized for mecha-
nistic studies of metal/ammonia reductions22 were used
to demonstrate the electron transfer potential of orga-
nocuprates toward enones. These molecules were de-
signed to trap any increase in electron density at the
â-carbon by an internal alkylation process.23 Specifically,
the establishment of organocopper properties in a variety
of systems was related to the isolation of 13 from reaction
with 12.24 Increased electron density on the alkenyl
carbons has been demonstrated4 by NMR techniques in
the Cu(3d)falkene(π*) complex from 6 so, even in the
current mechanistic scenario which does not involve
electron transfer, the use of internal trapping experi-
ments remains valid.
12. Reaction of Bu2CuLi prepared in Et2O solvent with
12 (entry 4) was included as a reference experiment, and
the detection of 0.47 mmol of 13 indicated a typical
baseline result for a “normal” cuprate. Reaction of the
1:2 CuI/BuLi mixture with 12 in the absence of Et2O
(entry 5) gave a significant amount of 13, but the
predominant species was 14. The occurrence of 13
indicated that an organocopper species was formed, at
least to some extent in toluene alone. Reaction of 12 with
a 1:2:2 CuI/BuLi/Et2O mixture (entry 6) gave a product
distribution comparable to that obtained from reaction
in Et2O solvent (entry 4) and is fully consistent with
traditional diorganocuprate activity.
A feature of this work was that with 1 equiv of Et2O
relatively small amounts of 1,4-addition were evident,
especially with the relatively unreactive enones.
A
reasonable explanation is that 1 equiv of Et2O was being
consumed preferentially and any additional Et2O was
then effective in enhancing cuprate-like reactivity. If
reactions with CuI/BuLi proceed to completion then 1
equiv of LiI is produced which would be expected to be
highly attracted to the Et2O. Thus it seemed possible
that the first equivalent of Et2O was required to seques-
ter the LiI and the additional amounts affected the 1,4-
reactivity by Cu(III) intermediate stabilization. To test
this idea, reactions were carried out with additional LiI
in 1:2 CuI/BuLi mixtures containing various amounts of
Et2O, and the results are presented in Table 5.
This probe/trap sequence was applied to the system of
interest in this work in order to gain evidence for
organocuprate activity. Experiments were carried out
using a variety of BuLi/CuI ratios in both toluene and
Et2O with a 10% excess of organometallic reagent relative
to the amount of 12. The reaction mixtures were
1
analyzed by H NMR and ratios of products determined
Ta ble 5. Rea ction of 1:2:w Cu I/Bu Li/Et2O a n d LiI
Mixtu r es w ith 6
by comparing the diagnostic resonances of 12 (5.84 ppm,
1H) and 13 (2.64 ppm, 2H). The amount of the 1,2-
addition product 14 in reaction mixtures containing 12
was calculated by difference. The results from these
experiments are depicted in Table 4.
amount (ME)
products (mol %)
entry
LiI
(w)Et2O
6
7
8
1
2
0.0
1.0
1.0
1.0
1.0
2
2
3
2
2
0
0
5
0
4
3
35
3
42
2
97
65
92
58
94
3
Ta ble 4. Rea ction of Va r iou s Cu I/Bu Li/Et2O Mixtu r es
w ith 12 in Tolu en e
4a
5b
a
b
LiI added with CuI. LiI added after Et2O.
In contrast to the reaction of 1:2:2 CuI/2BuLi/Et2O
which gives predominantly 8 (entry 1), the presence of 1
ME of LiI resulted in the formation of significant
amounts of 7 (entry 2). A high 1,4-/1,2-addition product
ratio was reestablished by increasing the amount of Et2O
(entry 3). These observations are in complete accordance
with the idea that the LiI removes some Et2O from the
reaction scene, but once this is complete then 1,4-
activation can be achieved by any residual Et2O. In the
reactions, summarized in entries 2 and 3 of Table 5, LiI
was added in the reaction sequence before Et2O. It was
established that addition of LiI to CuI at the initial stage
of the reaction had the same effect (entry 4). Curiously,
addition of the LiI after the Et2O did not influence the
reaction product distribution (entry 5) using reaction
times comparable to the rest of this work.
amount (mmol)
products (mmol)
entry
x
y
solvent w (ME)
12
13
14
1
2
3
4
5
6
0.0
1.2
1.2
1.2
1.2
1.2
1.1
1.1
1.1
2.2
2.2
2.2
toluene
toluene
toluene
Et2O
0
0
2
0
0
2
0.51
0.37
0.00 0.34
0.09 0.46
>0.95 <0.05 0.00
0.45
0.20
0.59
0.47 0.00
0.16 0.50
0.40 0.00
toluene
toluene
Reaction of 12 with BuLi did not yield any 13 (entry
1); however, the amount of 14 obtained did give an
indication of the maximum yield attainable under these
reaction conditions. Reaction of 12 with a 1:1 CuI/BuLi
mixture (entry 2) gave some 13, but once again the major
products were 14 and 12. These results confirm that
copper is indeed required to obtain at least some 13.
Reaction of 12 with a 1:1:2 CuI/BuLi/Et2O mixture gave
nearly complete recovery of the starting material (entry
3). This provides further evidence that BuCu, formed
in an ethereal environment, is unreactive even toward
Con clu sion
These results indicate that the reactivity of the CuI/
BuLi mixtures in toluene closely resembles that of a
mixture of various amounts of BuLi and BuCu. The
addition of only 2 ME of Et2O to this mixture is sufficient
to produce reactivity similar to that observed from a
normal cuprate. The method by which Et2O affects CuI/
BuLi mixtures in toluene cannot be ascertained with
absolute confidence from these results. Et2O may acti-
vate an existing species already present in the toluene
(22) Stork, G.; Rosen, P.; Goldman, N.; Coombs, R. V.; Tsuji, J . J .
Am. Chem. Soc. 1965, 87, 275-286.
(23) Smith, R. A. J .; Hannah, D. J . Tetrahedron 1979, 35, 1183-
1189.
(24) Hannah, D. J .; Smith, R. A. J .; Teoh, I.; Weavers, R. T. Aust.
J . Chem. 1981, 34, 181-188.

Mechanism of Diorganocuprate 1,4-Addition Reactions
J . Org. Chem., Vol. 62, No. 14, 1997 4633
purified by continuous extraction with THF.26 Anhydrous
lithium iodide was prepared according to the literature
method.27
mixture or create an entirely new species and, in either
event, influence 1,4-reactivity. Either scenario is possible
but the clear indication from this work is that at least
some reactive organocopper is created in toluene without
Et2O. A second area of influence is in the facilitation of
collapse of the π-complex by stabilization involving
coordination of the copper(III) intermediate as predicted
by the theoretical results. The observations reported in
this paper are totally in accord with this feature.
2-Cyclohexenone (9), 3,5,5-trimethyl-2-cyclohexenone (10),
and trans-4-phenyl-3-buten-2-one (11) were commercial samples.
1-Butyl-2-cyclohexen-1-ol,28 3-butylcyclohexanone,28 3-butyl-
3,5,5-trimethylcyclohexanone,29 trans-3-methyl-1-phenyl-1-
hepten-3-ol,30 and 4-phenyl-2-octanone31 were identified from
authentic samples or spectra. 4,4a,5,6,7,8-Hexahydro-4a-
methyl-2(3H)-naphthalenone (6)32 and cis-4,4a,5,6,7,8-hexahy-
dro-5-(mesyloxy)-4a-methyl-2(3H)-naphthalenone (12)26 were
prepared as described previously.
Reactions involving the use of moisture and/or oxygen
sensitive compounds were performed on a double manifold
system under an atmosphere of argon. Manipulations of all
sensitive materials and dry solvents were carried out using
double-ended needle techniques. Cooling baths were obtained
by the addition of liquid nitrogen to appropriate solvent
mixtures. The -78 °C bath was prepared from a mixture of
ethanol and acetone, and the warmer baths were prepared
from mixtures of ethanol and water. Organometallic reactions
were routinely quenched at -40 °C with an aqueous mixture
of ammonium chloride/ammonia (NH4+/NH3), prepared by
mixing saturated ammonium chloride and 25% ammonia in a
ratio of 25:4.
Illustrated in Scheme 3 is a summary of the possible
effects of Et2O on the reactivity of the 1:2 CuI/BuLi
mixture in toluene.
Sch em e 3
Gen er a l Met h od s for Or ga n om et a llic R ea ct ion s.
(Note: details of solvent, enone, and Et2O are given with each
example.)
Meth od A. BuLi (1.1 mmol) was added to dry degassed
solvent (10 mL) in a dry Schlenk tube at -78 °C. The solution
was stirred for 10 min, and then enone (1.0 mmol) was added
by syringe over 5-10 s. The mixture was stirred at -40 °C
for 30 min and then quenched and worked up as described
below.
Meth od B. CuI (0.228 g, 1.2 mmol) was weighed into a
dry Schlenk tube, evacuated for 5 min, and then filled with
argon. Dry, degassed Et2O (10 mL) was added, and the
resulting white suspension was cooled to -78 °C, with stirring.
BuLi (2.2 mmol) was added, and the mixture was stirred for
30 min. Over this time period the mixture darkened to a grey
suspension. The mixture was then transferred to a -40 °C
cold bath and stirred for 5 min, resulting in a black mixture.
This mixture was recooled to -78 °C before enone (1.0 mmol)
was added by syringe over 5-10 s, resulting in a slight khaki
hue. The mixture was stirred for 30 min at -40 °C and then
worked up as below.
In the absence of Et2O, the system reacts as a mixture
of BuLi and BuCu, which may be 1,2-addition reactive
under these conditions. The addition of appropriate
amounts of Et2O shifts the equilibrium toward a cuprate.
Reaction of this material with an enone is reversible5 and
leads eventually to the π-complex. The Et2O then
stabilizes the trialkylcopper intermediate species to give
the ultimate enolate product.
Direct, spectroscopic observation of the reagent(s) and
intermediates in the presence of stoichiometric amounts
of Et2O may further resolve the nature of these species,
and these studies are currently being pursued in these
laboratories.
Exp er im en ta l Section
Meth od C. CuI (0.228 g, 1.2 mmol) was weighed into a
dry Schlenk tube, evacuated for 5 min, and then filled with
argon. Dry, degassed toluene (10 mL) was added and the
resulting white suspension was cooled to -78 °C, with stirring.
BuLi (2.2 mmol) was added, and the mixture was stirred for
30 min, transferred to a -20 °C cold bath, and stirred for 15
min. Over this time period the suspension darkened through
grey to a black color. The mixture was again cooled to -40
°C, inoculated with various amounts of Et2O, and stirred for
10 min. This mixture was recooled to -78 °C before enone
(1.0 mmol) was added by syringe over 5-10 s. The mixture
was stirred for 30 min at -40 °C and then quenched and
worked up as below.
Meth od D. As for Method C, except using BuLi (1.1 mmol).
In reactions that were analyzed by GC an internal standard
(1.0 mmol) was added by syringe after the enone addition at
-78 °C.
Wor k u p P r otocols. Wor k u p 1. The total reaction mix-
ture was poured into a separatory funnel, and the reaction
Schlenk tube was washed consecutively with NH4+/NH3 (10
mL) and solvent (10 mL). Reactions containing copper were
Gen er a l Meth od s a n d Ma ter ia ls. Column chromatog-
raphy was performed using Merck 9385 silica gel or Riedel-
de Hae¨n basic alumina S (modified to activity II by addition
of 3% H2O). A 1:60 ratio of crude material to adsorbent was
used. NMR spectra were recorded in CDCl3 solution, unless
otherwise indicated. Numbers in brackets directly after
chemical shift values reflect the relative isomeric ratio. Gas
chromatographic (GC) analyses were carried out using a 10
m, 0.25 mm internal diameter, 0.25 µm film thickness, DB1
fused silica column. Helium was used as the carrier gas at a
flow rate of 1.5 mL min-1 with a 50:1 split.
Toluene was distilled from sodium and stored over sodium
wire. Anhydrous diethyl ether (Et2O) (Aldrich, 99+%) was
stored over sodium wire. For organometallic reactions, the
solvents were deoxygenated by alternate application of vacuum
and argon. n-Butyllithium (BuLi) was obtained (Aldrich) as
a 1.6 mol L-1 hexane solution and analyzed by the Gilman
double titration method.25 Free alkali was determined by
adding BuLi (1.00 mL) to a solution of 3-bromo-1-propene (0.5
mL) at -40 °C, in either Et2O (10 mL) or toluene (10 mL).
The toluene reaction was quenched with ethanol. Commercial
copper iodide (CuI) (Aldrich, 99.999%) was finely ground and
(28) Vellekoop, A. S. Ph.D. Thesis (University of Otago, 1992).
(29) Lipshutz, B. H.; Ellsworth, E. L.; Siahaan, T. J . J . Am. Chem.
Soc. 1989, 111, 1351-1358.
(25) Gilman, H.; Cartledge, F. K. J . Organometal. Chem. 1964, 2,
(30) Imamoto, T.; Kusumoto, T.; Tawarayama, Y.; Sugiura, Y.; Mita,
T.; Hatanaka, Y.; Yokoyama, M. J . Org. Chem. 1984, 49, 3904-3912.
(31) Yamamoto, Y.; Yamamoto, S.; Yatagai, H.; Ishihara, Y.; Maruya-
ma, K. J . Org. Chem. 1982, 47, 119-126.
447-454.
(26) Marshall, J . A.; Huffman, W. F.; Ruth, J . A. J . Am. Chem. Soc.
1972, 94, 4691-4696.
(27) Taylor, M. D.; Grant L. R. J . Am. Chem. Soc. 1955, 77, 1507-
(32) Heathcock, C. H.; Ellis, J . E.; McMurry, J . E.; Coppolino, A.
Tetrahedron Lett. 1971, 4995-4996.
1508.

4634 J . Org. Chem., Vol. 62, No. 14, 1997
Kingsbury and Smith
vigorously shaken until a deep blue color developed and all
solids had dissolved. The layers were then separated. The
aqueous layer was solvent extracted (2 × 20 mL), and the
combined organic extracts were washed with water (50 mL)
and then dried over anhydrous magnesium sulfate (MgSO4).
The mixture was filtered after approximately 15 min, and
solvents were evaporated under reduced pressure.
Wor k u p 2. After quenching, two aliquots (≈0.5 mL) were
removed from the organic phase and were added to separate
sample tubes, each containing NH4+/NH3 (≈0.5 mL), and
diluted with hexane (2 mL). The mixtures were shaken until
all solid had dissolved and then allowed to settle, and the
organic phase was analyzed by GC.
2-Bu t yl-2,3,4,4a ,5,6,7,8-oct a h yd r o-4a -m et h yl-2-n a p h -
th a len ol (7). Method A (three times scale), Et2O (30 mL), 6
(465 µL, 3.0 mmol), workup 1, hexane, but dried over anhy-
drous sodium sulfate (Na2SO4). The 1H NMR of the crude
product indicated a small amount of 6 still remained, and this
material was chromatographed on basic alumina (activity II).
Elution with 15% Et2O/hexane gave 7 (0.526 g, 2.37 mmol,
79%) of a colorless oil, as an inseparable mixture of isomeric
allylic alcohols. Anal. Calcd for C15H26O: C, 81.02; H, 11.79%.
Found: C, 80.71; H, 11.48%. IR νmax (cm-1) (film) 3354 (OH),
1657 (CdC). 1H NMR δ (ppm) 5.22 (1H, broad s, H-1), 2.3-
1.1 (19H, m) 1.10 (2), 1.02 (1) (3H, s, H-9), 0.92 (3H, t, J ) 6.9
Hz, H-13). Workup by standing over anhydrous magnesium
sulfate or chromatography on silica gel or basic alumina
(activity I) resulted in dehydration of 7.
Rea ction of th e 1:2:w Cu I/Bu Li/Et 2O Mixtu r es w ith 6.
All reactions were run using method C, Et2O (w(ME)), 6 (155
µL, 1.0 mmol), and workup 2, and products were analyzed by
GC. The results, graphically presented in Figure 1:
(w)Et2O
products (mol %)
(ME)
7
8
0.00
1.00
1.25
1.50
1.75
2.00
3.00
5.00
98
93
67
44
37
3
2
7
33
56
63
97
98
96
2
4
Rea ction s of 1:2:w Cu I/Bu Li/Et2O Mixtu r es w ith 9-11.
All reactions were run using method C, Et2O (w(ME)), enone
(1.0 mmol). Reactions were worked up via workup 1, dried
over Na2SO4, and analyzed by 1H NMR. Product percentages
from mixtures were ascertained from diagnostic 1H NMR
resonances or from integrated area excesses. The results are
given in Table 3.
Stoich iom etr ic Stu d ies on Cu I/Bu Li Mixtu r es w ith 6.
All reactions were adjusted to 1.0 mmol scale and worked up
as for workup 2. The results are given in Tables 2 and 3.
cis-8a-Bu tyl-3,4,4a,5,6,7,8,8a-octah ydr o-4a-m eth yl-2(1H)-
n a p h th a len on e (8). Method B, 6 (155 µL, 1.0 mmol), workup
Bu Li. Method A, toluene, and 6 (155 µL, 1.0 mmol).
1:1 Cu I/Bu Li. Method D and 6 (155 µL, 1.0 mmol).
1
1, hexane. The H NMR of the crude product showed a small
amount of 7. The product was chromatographed on silica gel,
and elution with dichloromethane (CH2Cl2) gave 8 (0.179 g,
89%) as a colorless oil. Anal. Calcd for C15H26O: C, 81.02; H,
11.79%. Found: C, 80.78; H,12.07%. IR νmax (cm-1) (film) 1715
(CdO). 1H NMR (300 MHz) δ (ppm) 2.4-2.1 (4H, m), 1.6-1.1
1:2 Cu I/Bu Li. (a) Method C and 6 (155 µL, 1.0 mmol). (b)
Method C and 6 (310 µL, 2.0 mmol).
1:1:5 Cu I/Bu Li/Et2O. Method D, Et2O (523 µL, 5.0 mmol),
and 6 (155 µL, 1.0 mmol).
1:2:1 Cu I/Bu Li/Et2O. (a) Method C, Et2O (105 µL, 1.0
mmol), and 6 (155 µL, 1.0 mmol). (b) Method C, Et2O (105
µL, 1.0 mmol), and 6 (233 µL, 1.5 mmol).
1:2:5 Cu I/Bu Li/Et2O. (a) Method C, Et2O (523 µL, 5.0
mmol), and 6 (155 µL, 1.0 mmol). (b) Method C, Et2O (523
µL, 5.0 mmol), and 6 (233 µL, 1.5 mmol).
Rea ction of Cu I/Bu Li Mixtu r es w ith 12. All reactions
were carried out in toluene with 12 (0.258 g, 1.0 mmol) added
as a CH2Cl2 solution (2 mL). Addition of 12 caused a yellow/
orange precipitate to appear, which stopped stirring. Workup
1, CH2Cl2, was followed by analysis using 1H NMR. The
results are listed in Table 4.
1
(16H, m), 1.05 (3H, s, H-9), 0.89 (3H, t, J ) 6.9 Hz, H-13). H
NMR (300 MHz) (90 °C) (toluene-d8) δ (ppm) 2.17 (1H, d, J )
14.1 Hz, H-1), 2.09 (2H, d, J ) 14.2 Hz, H-3), 1.91 (1H, d, J )
14.1 Hz, H-1), 1.69 (1H, m, H-3), 1.4-0.9 (15H, m), 0.85 (3H,
t, J ) 6.9 Hz, H-13), 0.80 (3H, s, H-9). 13C NMR (75MHz) δ
(ppm) 213.0 (C-2), ≈47 (very broad) 42.8, 38.1, 35.6 (broad),
34.2, 33.9 (broad), 30.1 (broad), 24.9, 23.6, 22.7, 21.6, 21.2, 14.1.
13C NMR (75 MHz) (90 °C) (toluene-d8) δ (ppm) 208.6 (C-2),
47.0 (CH2), 42.9 (C), 38.1 (CH2), 36.2 (CH2), 36.0 (C), 34.7
(CH2), 34.5 (CH2), 30.8 (CH2), 25.5 (CH2), 24.1 (CH2), 22.8
(CH3), 22.1 (CH2), 21.7 (CH2), 14.1 (CH3).
1-Bu tyl-3,5,5-tr im eth yl-2-cycloh exen -1-ol. Method C, 10
(150 µL, 1.0 mmol), workup 1, hexane, dried over anhydrous
Bu Li. Method A gave 12 (0.131 g, 0.51 mmol) and cis-2-
butyl-2,3,4,4a,5,6,7,8-octahydro-5-(mesyloxy)-4a-methyl-2-naph-
thalenol (14) (0.116 g, 0.34 mmol) as an inseparable mixture
1
Na2SO4. The H NMR analysis gave 1-butyl-3,5,5-trimethyl-
1
2-cyclohexen-1-ol (95%), H NMR δ (ppm) 5.32 (1H, broad s,
H-2), and 3-butyl-3,5,5-trimethylcyclohexanone (5%), from the
integrated areas. The crude material was chromatographed
on alumina, and elution with 25% Et2O/hexane gave the
alcohol (0.161 g, 82%) as a colorless oil. Anal. Calcd for
C13H24O: C, 79.53; H, 12.32%. Found: C, 79.23; H, 12.46%.
IR νmax (cm-1) (film) 3386 (OH) 1670 (CdC). 1H NMR δ (ppm)
5.32 (1H, broad s, H-2), 1.8-1.6 (5H, m), 1.5-1.2 (9H, m), 1.04
(3H, s), 0.97 (3H, s), 0.90 (3H, t, J ) 7.1 Hz). 13C NMR (50
MHz) δ (ppm) 135.2 (C-3), 126.0 (C-2), 71.4 (C-1), 47.6, 44.4,
44.0, 31.7, 29.9, 27.4, 25.7, 24.0, 23.2, 14.1.
1
of isomers, H NMR δ (ppm) 3.1-3.0 (3H, s, OMs), 0.90 (3H,
t, J ) 7.0 Hz, H-13).
1:1 Cu I/Bu Li. Method D gave 12 (0.095 g, 0.37 mmol) and
6-methyltricyclo[4.4.0.01,5]-9-decanone (13) (0.015 g, 0.09 mmol),
1H NMR δ (ppm) 2.64, 2.63, (2H, AB system, J AB ) 18.6 Hz,
H-1), 1.06 (3H, s, H-11), in agreement with an authentic
sample,17 and 14 (0.147 g, 0.46 mmol).
1:1:2 Cu I/Bu Li/Et 2O. Method D, Et2O (230 µL, 2.2 mmol).
Bu 2Cu Li/Et2O. Method B.
Rea ction s of Or ga n ocop p er Rea gen ts w ith 6. Products
from organometallic reactions with 6 were analyzed by GC
with n-dodecane (n-C12) (227 µL, 1.0 mmol) as the internal
standard (workup 2). The column was heated from 90 to 190
°C at a rate of 15 °C min-1. Retention times (tR) and response
factors (Rf) for 6-8 are given in the following table.
1:2 Cu I/Bu Li. Method C.
1:2:2 Cu I/Bu Li/Et2O. Method C, Et2O (230 µL, 2.2 mmol).
Rea ction s of 1:2:w Cu I/Bu Li/Et2O w ith 6 Con ta in in g
LiI. All reactions were run using method C, Et2O (w(ME)),
LiI (0.147 g, 1.0 ME), and 6 (155 µL, 1.0 mmol), and the
reactions were worked up via workup 2. The results are given
in Table 5.
compound
tR (min)
Rf
Ack n ow led gm en t. C.L.K. thanks the Research
Committee of the New Zealand Universities Grants
Committee for a postgraduate scholarship.
n-C12
2.0
3.7
4.2-5.4
6.1
-
6
7
8
1.30
0.98
0.92
J O970316Y