Demonstration of promoted zinc schlenk equilibria, their equilibrium values and derived reactivity

Article abstract of DOI:10.1002/chem.200601739

The presence of promoted Schlenk equilibria for organozinc halide species has been explicitly demonstrated by ¹³CNMR studies. Thus, addition of methylaluminoxane (McAlO)_n, MAO, to RZnX (R = Et, Bn, ArCH ₂, (CH₂)

Full text of DOI:10.1002/chem.200601739

FULL PAPER
DOI: 10.1002/chem.200601739
2462
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 2462 – 2472

FULL PAPER
Demonstration of Promoted Zinc Schlenk Equilibria, their Equilibrium
Values and Derived Reactivity
Alexander J. Blake, Jonathan Shannon, John C. Stephens, and Simon Woodward*[a]
Abstract: The presence of promoted
Schlenk equilibria for organozinc
halide species has been explicitly dem-
onstrated by 13C NMR studies. Thus,
3.0 equiv MAO). Use of RZnX/MAO
mixtures allows copper-catalysed 1,4-
addition to 2-cyclohexenone to be ach-
ieved, but a competing cascade reac-
tion (two subsequent Michael additions
and an intramolecular aldol reaction)
leads to novel tetracyclic by-products
(characterised crystallographically in
one case). Activation of EtZnCl is also
achieved by ZnMe2 addition and the
presence of intermediate EtZnMe was
observed by 13C NMR spectroscopy (at
equilibrium, Kꢁ1). Asymmetricconju-
gate addition in this system can be re-
alised (up to 92%ee for additions to 2-
cyclohexenone).
addition
of
methylaluminoxane
(MeAlO)n, MAO, to RZnX (R=Et,
Bn, ArCH2, (CH2)3CO2Et; X=Cl, Br)
leads to the formation of ZnR2 and
ZnX2·MAO. For EtZnCl, equilibration
of ZnEt2 and ZnX2·MAO is rapid at
ꢀ358C; a K value of 0.19mꢀ1 indicates
the equilibrium favours ZnEt2 (0.75–
Keywords: aluminum
· diorgano-
zinc · organozinchalide · Schlenk
equilibria · structure elucidation ·
zinc
Introduction
observations were made by using ZnMe2 as an apparent
halide scavenger (2 generated in 84%ee). The success of
these approaches implied that the zinc Schlenk equilibrium
[Eq. (1)] was being modified in processes that can be repre-
sented by equilibria shown in Equations (2) and (3). Al-
though the suggestion has been made that the equilibrium
constant for the formation of ZnR2 in the unpromoted zinc
Schlenk case [Eq. (1)] is vanishingly small,[2] essentially
nothing is known about such processes.[3] As an alternative,
one could suggest that the observed reactivity in Scheme 1
is due simply to the presence of the Lewis acids MAO and
ZnMe2, which activate substrate 1 to a point at which the
RZnX species become adequate nucleophiles.[4] It is there-
fore important to demonstrate explicitly that the equilibria
shown in Equations (2) and (3) do operate, to measure their
equilibrium constants and to show other catalytic procedures
that could also use this new approach for ZnR2 formation.
Recently we described SN2’ displacement of the allylic chlo-
ride 1, and related species, in which ZnEt2 leads to highly
stereoselective formation of 2 (Scheme 1).[1]
The use of a methylaluminoxane [(MeAlO)n, MAO] addi-
tive was crucial to the realisation of high ee values in these
reactions. It was postulated that removal of the kinetic
EtZnCl by-product (which leads to unselective SN2’ reac-
tions) by MAO was instrumental in achieving this. Similar
K<0:002
ð1Þ
*
2 RZnX )ꢀ R2Zn þ ZnX2
Scheme 1. Asymmetric SN2’ reactions of 1 employing recycling of EtZnCl
to avoid the “low ee” reaction manifold.
2 RZnX þ ðAlMeOÞn Ð R2Zn þ ZnX2 Á ðAlMeOÞn
RZnX þ ZnMe2 Ð RZnMe þ MeZnX
ð2Þ
ð3Þ
[a] Prof. A. J. Blake, J. Shannon, Dr. J. C. Stephens, Prof. S. Woodward
School of Chemistry, The University of Nottingham
University Park, Nottingham, NG7 2RD (UK)
Fax : (+44)115-951-3564
Diorganozincreagents (ZnR ) are valuable synthons that
2
are more reactive than organozinc halides (RZnX) and un-
E-mail: simon.woodward@nottingham.ac.uk
dergo transmetalation reactions much more readily.[5] How-
Chem. Eur. J. 2007, 13, 2462 – 2472
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2463

S. Woodward et al.
ever very few diorganozinc species are commercially avail-
able (R=Me, Et, nBu, Ph) and new, technically simple, gen-
eral routes to their preparation are highly desirable. To the
best of our knowledge, in contrast to the known magnesium
Schlenk processes,[6] no quantitative study of zinc-based
Schlenk equilibria has ever appeared in the literature.
Results and Discussion
Figure 1. Chemical shift behaviour of the time averaged 13C NMR spectra
of the ZnCH2 signal of the species EtZnY (Y=Cl, Et) at 228C. a) Pure
EtZnCl; b) EtZnY mixtures; c) pure ZnEt2.
Promotion of zinc Schlenk equilibria by MAO: The mixture
of (AlMeO)n species known as MAO is a strong halide scav-
enger and we believed it would act as a “latent Lewis
acid”[7] promoting the forward equilibrium given in Equa-
tion (1) by sequestering the ZnCl2. This is example of anti-
crown behaviour. Anti- and inverse-crown complexes are
species in which multiple Lewis acids surround an occluded
Lewis base.[8] The term anti-crown is used if the Lewis base
has soft, or intermediate hard–soft acid–base (HSAB)[9]
character (e.g., Br, I), of which poly-mercury bound species
are the best known (e.g., 3,[10] Scheme 2). If the occluded
Lewis base has much harder HSAB character, usually an
anionicoxygen donor, then additional HSAB-mahtced
Lewis acids are usually present (e.g., Na, K, Mg as in the in-
verse-crown species 4,[11] Scheme 2).
Figure 2. Calculated percentage of ZnEt2 on addition of MAO to a 0.5m
solution of EtZnCl.
tectable in pure 0.5m EtZnCl (in THF), indicating that
Equilibrium (1) is significantly promoted by the presence of
MAO. Interestingly, new signals at approximately ꢀ13, 8.5
and 9.5 ppm are also present in the EtZnCl/MAO mix: they
ꢀ
clearly arise from the anti/inverse-crown co-product MAO
ZnCl2 [Eq. (2)], but the available spectroscopic data does
not allow its structure to be determined.
Scheme 2. Representative examples of anti-crown 3 and inverse-crown 4
Lewis acids. The B10H10C2 ligands of 3 are shown schematically.
ꢀ
ꢁ
Ambient temperature 13C NMR studies showed that
MAO promotes the zincShclenk equilibrium [Eq. (2)]
through anti/inverse-crown ZnCl2 behaviour. Stepwise addi-
tion of MAO (0.75–3.0 equiv) to solutions of EtZnCl in
THF (0.5m) at 228C results in fast exchange of all the ethyl
substituents, leading to observation of an averaged ZnCH2
methylene signal in the 13C NMR spectrum (dobs). As the
MAO concentration increases, the value of dobs increases
from dC =ꢀ0.4 ppm (for pure EtZnCl, dX) towards that for
pure ZnEt2 dC =2.8 ppm (dR2) indicating an increased popu-
lation of ZnEt2 (behaviour shown schematically in Figure 1).
This chemical shift change can be modelled by using
Equations (4) and (5) ([Zntot] and [Altot] are the total con-
centration of all zinc and aluminium species present in the
system, see Experimental Section). This procedure gives
K=0.19Æ0.05mꢀ1. Putting the K value back into Equa-
tion (2) allows the percentage speciation to be determined
(Figure 2). At 0.5m equimolar quantities of EtZnCl and
MAO, about 20% of the mixture is converted to ZnEt2.
Under identical conditions, no ZnEt2 is spectroscopically de-
½Zntot
Š
dobsꢀdX
dR2ꢀdX
½ZnR2Š ¼
K ¼
ð4Þ
ð5Þ
2
2
½ZnR2Š
1
2 Á
ð½ZntotŠꢀ2½ZnR2ŠÞ ð½AltotŠꢀ½ZnR2ŠÞ
Despite the clear presence of ZnEt2 in the MAO-promot-
ed EtZnCl [Eq. (2)], the addition of such mixtures to 2-cy-
clohexenone leads mainly to uncharacterised oligomeric spe-
cies in THF (Scheme 3) rather than the desired 5a. In none
of the cases was more than a trace amount of the 1,4-methyl
addition product detected. Changing the organozinc reagent
to the commercially available PhCH2ZnBr (0.5m, THF)
gave similarly low yields of 3-benzylcyclohexanone 5b
(Table 1, entry 3). Attempted optimisation, varying the tem-
perature (ꢀ30, ꢀ50 or ꢀ658C), copper source (library of 11)
or added ligands (library of 50) had little effect on the yield
of 5b in THF. In all cases complete conversion was attained
within 60min, but without improvement in the yield of 5b.
The optimal catalyst was found to be copper(I) thiophene-
2464
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 2462 – 2472

Promoted ZincEquilibria
FULL PAPER
Figure 3. Molecular structure and numbering scheme for tetracyclic oligo-
mer 6.
Scheme 3. Organozinc addition to 2-cyclohexenone (in D, Ar=4-Me, 4-
iPr, 4-Br).
tional enone starting material is
known to be a significant com-
petition pathway. Mixed diaste-
reomers of the derived so-
called “dimer”[14] products (i.e.,
those formed by protonation of
the enolate structure within 7,
Scheme 4) are commonly seen
as high-boiling impurities in the
Table 1. MAO-promoted RZnX reactions with 2-cyclohexenone.[a]
Run RZnX Source
[equiv]
Solvent
T
Ligand
[4 mol%]
MAO
1,4-Yield
[%]
0[e]
Oligomers
[%][d]
6 [%][c]
[oC]
[equiv][b]
[c]
1
2
3
4
5
none
THF
ꢀ30
ꢀ30
ꢀ30
ꢀ65
ꢀ30
P
P
P
P
N
5
0
5
5
3
63
39
80
68
–
–
–
–
B PhCH2ZnBr (2.8) THF
B PhCH2ZnBr (2.8) THF
B PhCH2ZnBr (2.8) THF
B PhCH2ZnBr
(0.33)
0[e] (5b)
16[e] (5b)
8[e] (5b)
n.d.[f]
THF
–
n.d.
20–25
(6b)
–
–
–
–
gas
chromatography
(GC)
6
7
8
9
C PhCH2ZnBr (2.0) DME
C PhCH2ZnBr (2.0) DME
C PhCH2ZnBr (2.0) DME
D 4-MePhCH2ZnBr DME
(2.0)
ꢀ30
ꢀ30
ꢀ30
ꢀ30
P
A
6
3
6
6
53[e] (5b)
53 (5b)
70 (5b)
60 (5c)
43
traces of 2-cyclohexenone de-
rived reaction mixtures. Never-
theless, the formation of tetra-
cyclic trimer products, such as
6, does not seem to have been
noted previously. Such behav-
iour is rare, but could be de-
scribed as a MIMIMIRC pro-
cess (MI=Michael; RC=Ring
Closing) using the nomencla-
ture of Posner.[15] Traces of a re-
lated product, derived from 1-
n.d.
n.d.
n.d.
10
11
12
D 4-iPrPhCH2ZnBr DME
(2.0)
D 4-BrPhCH2ZnBr DME
(2.0)
ꢀ30
ꢀ30
ꢀ30
–
–
–
6
6
6
70 (5d)
72 (5e)
20 (5 f)
n.d.
n.d.
n.d.
–
–
E EtO2C-
(CH2)3ZnBr (2.0)
DME
10 (6 f)
ACHTREUNG
[a] Standard conditions: [Cu(TC)]/RZnBr/cyclohexenone: 0.01:4:2.0 mmol; DME (or THF)/toluene 3:4 v/v;
total volume DME or THF (12 mL); total volume toluene (including that added with any MAO, 16 mL); 1 h
reaction time. [b] Added as toluene solution (10 wt.%, AlMeO). [c] Isolated yield after chromatography
unless otherwise indicated. [d] Estimated from reaction mass balance. [e] Determined by GC on a BP20
column using a tridecane internal standard (25 mL). [f] Not determined.
carboxylate[13] [Cu(TC)] (2 mol%) at ꢀ308C by using either
no ligand or PACHTREUNG(OPh)3 (4 mol%). Due to this initial resist-
ance to optimisation, attempts were made to identify the
nature of the oligomeric species formed in these reactions.
Fortunately, the major oligomericprodutc derived from
PhCH2ZnBr could be crystallised and X-ray studies revealed
the tetracycle 6b (R=CH2Ph) (Figure 3). Remarkably, 6b is
isolated as a single diastereoisomer (in 20–25% yield) re-
sulting from three consecutive conjugate additions and a
1,2-addition termination step, generating seven contiguous
stereocentres. The cascade in Scheme 4 is a probable explan-
ation of its origin. Reinvestigation of the EtZnCl/MAO/cy-
clohexenone mixtures by mass spectrometry revealed the
presence of the ethyl analogue 6a in the oligomericmixture;
however, this could not be isolated as a pure crystalline
solid.
In poorly chemoselective catalytic 1,4-additions of orga-
nometallics, capture of the initial (kinetic) enolate by addi-
Scheme 4. Proposed mechanism for the formation of 6 (R=CH2Ph).
Chem. Eur. J. 2007, 13, 2462 – 2472
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2465

S. Woodward et al.
cyanocyclohexene, have been claimed,[16] but in this case no
supporting evidence was presented. In our original MAO-
then the approximation that [Zn
valid and K can be estimated by using Equation (5). Under
these conditions,
plot of [CAobs]2/ [CAobs])2
([Zntot]ꢀ2
against ([Altot]ꢀ[CAobs]) is expected to be a straight line of
G
promoted SN2’ ZnEt2 additions to the allylichcloride
1
a
(Scheme 1), the choice of solvent dramatically affected both
the yield and stereoselectivity, both being favoured by the
use of dimethoxyethane (DME). We believed that this sol-
vent might also stabilise the initial kineticenolates, prevent-
ing the cascade reaction leading to 6. Therefore, solutions of
ArCH2ZnBr in DME were prepared by direct, and very
ACHTREUNG
gradient K; here [Zntot] and [Altot] are again the total con-
centration of all zinc and aluminium species present in the
system at a given composition, and [CAobs] is the measured
concentration of 5b based on its GC yield. The derived plot
of Equation (5) was found to be linear (correlated coeffi-
cient=0.995) for 1.5–9.0 equivalents of MAO to
PhCH2ZnBr and gives K=4.8Æ0.8mꢀ1. This value is signifi-
cantly higher than the K=0.19mꢀ1 determined from the
NMR experiments conducted on EtZnCl/MAO. Such behav-
ꢀ
clean, insertion of zinc into the C Br bond of ArCH2Br.
Use of DME for the formation of RZnI species was first re-
ported by Zakharin and Okhlobystin,[17] but this approach
has been rarely used, despite the fact that the higher boiling
point of DME (858C) is beneficial to the insertion rate.
Indeed, DME as solvent, in the absence of any added
ligand, allows synthetically useful yields of the conjugate-ad-
dition products 5a–f to be realised (Table 1, entries 6–10),
presumably due to increased stability of the initial enolates.
However, use of other polyether or donor solvents was not
as profitable (diglyme, triglyme and DMF gave 5–37% yield
of 5b). The PhCH2ZnBr/MAO procedure is much more ef-
iour cannot be caused by under estimating the [ZnACHTREU(GN CH2Ph)2]
by equating this to [5b] (i.e., not allowing for a minor oligo-
merisation component). Equation (5) leads to an overesti-
mation of K, as the concentration of MAO found in the re-
action mixture is less than expected. The kinetic enolate
formed from the conjugate addition to 2-cyclohexenone is a
potent Lewis base and competitively binds MAO in the re-
action mixture (K2, Scheme 5). Such behaviour lowers the
fective than use of pure ZnACTHER(UNG CH2Ph)2 (the latter gave <20%
conversion under the conditions investigated), and appears
to be one of the best methods for 1,4-benzylicaddition. One
functionalised organozinc ester BrZnACHTER(UNG CH2)3CO2Et could be
added to 2-cyclohexenone in poor yield (Table 1, entry 12).
However, in this case a significant side product was the de-
rived cascade oligomer 6 f, which could be isolated (10%).
The formation of 6 f was favoured by the residual THF pres-
ent in the commercial organometallic used. Attempts to
carry out the copper-catalysed reaction of 2-cyclohexenone
and PhCH2ZnBr by using only 0.75 equivalents of MAO
were unsuccessful, producing only traces of the addition
product 5b (4%). In fact, the yield of the MAO-promoted
reaction of PhCH2ZnBr and 2-cyclohexenone clearly dem-
onstrates saturation behaviour (Figure 4) with respect to the
MAO concentration.
Scheme 5. Competitive MAO binding by both the kinetic enolate product
and the starting 2-cyclohexenone.
effective concentration of MAO available to promote the
zincSchlenk equilibrium [Eq. (2)] and leads to the overesti-
mation of K when Equation (5) us used. A similar, but less
effective, MAO competitive binding to 2-cyclohexenone
also exists (K3, Scheme 5). It will also be seen that the reac-
tion sequence in Scheme 5 prevents Le Chatelierꢁs principle
from operating as although irreversible enolate formation
should drive the equilibrium in Equation (2) to completion
(through ZnR2 removal), this is self-limited. Generation of
the enolate product also removes MAO (through K2), lower-
ing the MAO concentration available to promote the initial
zinc Schlenk process [Eq. (2)].
Although it is clear that the reaction sequences in
Scheme 5 will account for Equation (5) over reporting the
value of K, deconvolution of the additional equilibrium con-
stants K2 and K3 could not be attained. 13C NMR studies of
solutions of MAO and 2-cyclohexenone in THF at low tem-
peratures are not possible due to the presence of multiple
broad signals arising from the presence of a mixture of
quadrupolar aluminium MAO species (27Al, I=5/2). Under
fast exchange conditions at ambient temperature, K3 could
Figure 4. Yield of 1,4-addition product 5b against MAO present in reac-
tion mixture for the reaction of 2-cyclohexenone with PhCH2ZnBr.
If equilibrium in Equation (2) is the only process in which
MAO participates, then the observed yield of 5b estimates
the quantity of ZnACHTREUNG(CH2Ph)2 present in the reaction mixture
(as only this traps onto 2-cyclohexenone). As long as the
DME solvent minimises the quantity of oligomers formed,
2466
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 2462 – 2472

Promoted ZincEquilibria
FULL PAPER
not be independently measured due to competing 2-cyclo-
hexenone polymerisation and some formation of the methyl
analogue of 6. Similarly, K2 could not be independently
measured at low temperature due to broad signals; attempt-
ed quantification at ambient temperature led to decomposi-
tion only. As an additional complication, the literature prep-
ZnEt2, and a coalescence temperature of ꢀ458C (Figure 5).
We propose that this exchange takes place through the tran-
sient intermediate [EtZnACTHER(UGN m-Cl)ACHTREU(GN m-Et)ZnEt] 8.
arative routes[18] to Zn
ACHTREU(NG CH2Ph)2 (required to form the eno-
late in Scheme 5) lead to impure samples containing a
second species, which could be characterised as PhCH2ZnCl.
By using an Et2O solvent rather than 1,4-dioxane the forma-
tion of pure Zn
A
Relevant 13C NMR signals from this system are summar-
ised in Table 2 along with the organozincspecies studied in
this investigation for comparison. A mixture of EtZnCl
(0.26m) and ZnMe2 (0.26m) results in little apparent reac-
tion at ambient temperature; signals assignable to both the
EtZnCl (dC =2.8, 12.5 ppm) and ZnMe2 (dC =10.2 ppm) are
present. However, cooling the mixture reveals a fluxional
process leading to the observation (at ꢀ1008C) of eight sig-
binding of the added MAO by the kineticenolate product
formed by the conjugate addition; therefore, it is necessary
to use a significant excess of MAO in the chemistry of
Scheme 3. An attempted in situ quench of the kinetic eno-
late, in which TMSCl, or BEt3 was added to prevent it from
binding to MAO, had no effect on the reaction.
Promotion of zinc Schlenk equilibria by other added ZnR2
species: In a seminal 1971 paper, Evans and co-workers re-
ported 1H NMR studies of ethyl exchange between ZnEt2
and EtZnCl.[2] We extended these investigations by using
13C NMR spectroscopy, as it was anticipated this would pro-
vide better spectral separation in the subsequent more com-
plicated reaction mixtures. A combination of ZnEt2 (0.25m)
and ZnCl2 (0.15m) in THF led to clean formation of EtZnCl
(0.3m) containing residual ZnEt2 (0.1m); no other apprecia-
Table 2. 13C NMR shifts [ppm] for the species RZnY species (R=Me,
Et; Y=Cl, Me, Et).
Species
Solvent
T [8C] c [m]
Signals [ppm] (assignment)
ZnEt2
THF
22
22
22
ꢀ100
0.25
0.26
0.3
12.5 (Me), 2.8 (CH2)
ꢀ10.2 (Me),
ZnMe2
EtZnCl
EtZnCl
THF/toluene[a]
THF
13.2 (Me), ꢀ0.4 (CH2)
13.9 (Me), ꢀ0.6 (CH2)
THF
0.3
EtZnMe THF/toluene[a] ꢀ100
0.26[b] 14.0 (CH2Me), 3.1 (CH2Me),
ꢀ11.1 (ZnMe)
1
ble species were detected. The complimentary H NMR in-
tegrals of the sample are consistent with the complete for-
mation of EtZnCl, in which there is a very low K value for
the equilibrium in Equation (1). Variable-temperature (VT)
studies (+22 to ꢀ1008C) confirmed exchange with the free
MeZnCl THF/toluene[a] ꢀ100
0.26[b] ꢀ14.8 (ZnMe)
[a] 13% toluene v/v in THF. [b] Total concentration of all Zn species
present.
Figure 5. Variable-temperature partial 13C NMR spectra of EtZnCl in the presence of ZnEt2.
Chem. Eur. J. 2007, 13, 2462 – 2472
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2467

S. Woodward et al.
nals, one of trace and batch dependent intensity (see foot-
note of Figure 6). Based on the data of Table 1 we assign
this behaviour to scrambling of all of the zinc substituents
(Scheme 1) via the 4-membered transition state 9.
generated in these NMR procedures, is confirmed by its in-
teraction with 2-cyclohexenone in the presence of CuACHTREUNG(OTf)2
and Feringaꢁs phosphoramidite (10)[21] (Table 3). Modifica-
tion of the solvent mixture results in >90%ee values and
minimal methyl transfer, consistent with the presence of
EtZnMe. Control reactions including EtZnCl give negligible
yields of conjugate addition product as expected.
Although the transfer of “Et” could be readily attained,
attempts to promote other “R” groups to transfer from the
RZnMe species were complicated by a competing “Me”
transfer to the enone. For example, BuZnMe was readily
prepared from BuZnCl and ZnMe2, but only 16–23% yield
The trace and variable signal at dC =2.8 ppm is assigned
to very small amounts of ZnEt2 (generated by a second
round of Schlenk redistribution between EtZnMe and
EtZnCl). Comparison of the 13C NMR data of the four
major zinc species indicates that their ratio is essentially
1:1:1:1, implying a K value of ꢁ1 for Equation (6) (R=Et,
[Zntot]=0.26m). Although the equilibrium in Equation (3)
has previously been investigated by calorimetry,[19] this is the
first confirmation of the formation of mixed diorganozinc
species by NMR spectroscopy.
Table 3. Asymmetric addition of EtZnCl to 2-cyclohexenone under pro-
moted equilibrium conditions.[a]
EtZnCl/ZnMe2
THF/toluene
Yield 5a/%[b]
ee/%[b]
½EtZnMeŠ½MeZnClŠ
ð6Þ
K ¼
ꢁ 1
1:0
1:3
1:0
1:3
1:1
1:3
2:1
2:1
1:9
1:9
2:98
2:98
<2
5
4
80
67
60
<2
26
<2
83
85
92
½EtZnClŠ½ZnMe2Š
As both RZnCl (R=Me, Et) species are relatively un-
reactive, we believed it would be possible to selectively uti-
lise EtZnMe alone in subsequent catalytic reactions. Fur-
[a] EtZnCl prepared from ZnCl2 and neat ZnEt2. Addition to 2-cyclohex-
enone (0.5 mmol) with Cu(OTf)2 (2 mol%) and (R,S,S)-10 (4 mol%) in
3.9 mL of the solvent mix: THF/toluene; ꢀ308C, 1 h. [b] Determined by
GC on a Lipodex-A column using nonane as an internal standard
(25 mL) for yield calculations; (R)-product observed in all cases.
ꢀ
ther, as Zn Me bonds are among the strongest in organo-
AHCTREUNG
zincchemistry ( D0 ꢁ68 kcalmolꢀ1), selective transfer of only
“Et” should also be viable, demonstrating the use of methyl
as a “non-transferable group”.[20] The utility of EtZnMe,
Figure 6. Variable-temperature 13C NMR spectra of EtZnCl in the presence of ZnMe2. The species marked “*” is assigned to the CH2 group of ZnEt2
generated in small quantities through secondary disproportionation of EtZnMe.
2468
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 2462 – 2472

Promoted ZincEquilibria
FULL PAPER
ZnMe2 (2.0m toluene solution; Aldrich), ZnEt2 (neat 98%; Strem),
MAO (10% w/w in toluene; Aldrich).
of the butyl 1,4-addition product was realised under the con-
ditions of Table 3. The mass balance of these reactions has
found to be the equivalent 1,4-methyl addition product (4–
20%) and unreacted 2-cyclohexenone. Attempts to use
“TMSCH2” as an alternative non-transferable group by re-
NMR investigations: NMR studies were carried out on a JEOL-270 MHz
spectrometer (270.2 MHz for 1H; 67.9 MHz for 13C) at temperatures be-
tween ꢀ1008C and RT. Studies were conducted in undeuterated THF
dried by distillation from sodium/benzophenone. Peak shape optimisation
was attained by manual optimisation of the repeated FID shape and
area. All NMR scales were referenced to the lower frequency signal of
THF (dH =1.80 ppm; dC =26.7 ppm). All preparations were carried out
under argon using flame-dried glassware. Key 13C NMR shifts are report-
ed in Table 2. The systems were allowed to equilibrate for 0.25–16 h at
RT, but in all cases identical results were attained irrespective of the in-
cubation period.
acting BuZnCl and ZnACHTREU(NG CH2TMS)2 produced a mixed diorga-
nozinc species of very low reactivity. It could be concluded,
at present, that the best approach to using promoted zinc
Schlenk equilibria is to use the MAO procedure given
above.
NMR studies of EtZnCl: Ethyl exchange between ZnEt2 and EtZnCl
was characterised by using VT 13C NMR spectroscopy. Fused ZnCl2
(102 mg, 0.75 mmol) was suspended in THF (5.0 mL) and neat ZnEt2
(48 mL, 92 mg, 0.75 mmol) and the mixture equilibrated at RT (16 h). A
sample (0.7 mL, 0.3m in EtZnCl) was transferred by syringe to a flame-
dried NMR tube under argon. An equivalent sample containing excess
ZnEt2 was prepared by addition of ZnEt2 (80 mL, 154 mg, 1.25 mmol) to
fused ZnCl2 (102 mg, 0.75 mmol) and THF (5.0 mL), leading to a sample
that contained, nominally, EtZnCl (0.3m) and ZnEt2 (0.1m). The VT
13C NMR spectra of this mixture are shown in Figure 5. At ꢀ858C the
ratio of 1H NMR methylene integrals for EtZnCl:ZnEt2 were 1.50–1.58
[a value of 1.50 is expected for K=0 in Equation (1)]. The system
reached coalescence at ꢀ458C.
NMR studies of the reaction of EtZnCl with ZnMe2: A nominal solution
of EtZnCl (1.50 mmol, 0.3m) was prepared from the reaction of neat
ZnEt2 (48 mL, 92 mg, 0.75 mmol) to fused ZnCl2 (102 mg, 0.75 mmol) in
THF (5.0 mL). The equilibrated EtZnCl solution was treated at RT with
ZnMe2 (0.75 mL, 1.50 mmol, 2.0m, toluene) and the mixture stirred at
RT (16 h). A sample (0.7 mL) was transferred to a flame-dried NMR
tube under argon that nominally contained EtZnCl (0.26m) and ZnMe2
(0.26m) in a 13:87 v/v mixture of toluene/THF. The VT 13C NMR behav-
iour of this sample is shown in Figure 6.
Conclusion
13C NMR studies have demonstrated explicitly that the
Lewis acids ZnMe2 and MAO cause disproportionation of
RZnX reagents into RZnMe and ZnR2 respectively. The
promoted equilibria favour the formation of diorganozinc
species by at least two orders of magnitude more than the
unpromoted case [Eq. (1)]. Modelling of both the NMR
data and chemical quenching data are both in accord with
the equilibria in Equations (2) and (3) and showed K values
from 0.2 to 4.5mꢀ1. Significant populations of RZnMe or
ZnR2 are achieved and these can be used in 1,4-conjugate
addition chemistry. However, the resulting kinetic enolate
can also participate in competing Michael additions to the
starting enone, resulting in previously uncharacterised cas-
cade products. Using excess MAO can overcome this prob-
lem as the enolate is captured by the MAO. This approach
is presently preferable to forming mixed RZnACHTRE(UNG RNT) species
NMR studies of the reaction of EtZnCl with MAO: Direct NMR analysis
(RNT =non-transferable group). Although the latter species
are easily prepared and characterised by 13C NMR spectros-
copy, further investigation is needed to attain chemospecific
of mixtures of organozincspeices and MAO (AlMeO)
solutions are
n
complicated by the oligomeric nature of the MAO and the quadrupole
nature of the aluminium nucleus (27Al, 100%, I=5/2). Below ꢀ108C
overlapped broad signals were observed (1H NMR studies) or the pres-
ence of multiple signals (13C NMR studies). We believe the latter is due
to zinc species interacting with the differing MAO oligomers. Fast ex-
change conditions at RT or 08C were therefore employed. A nominal
1.25m solution of EtZnCl (25.0 mmol) was prepared from the reaction of
neat ZnEt2 (1.28 mL, 12.50 mmol) to fused ZnCl2 (1.7 g, 12.50 mmol) in
THF (18.7 mL), and the resulting solution was stirred at RT (1 h). After
this time, 2 mL of the solution was transferred into a second flame-dried
Schlenk and MAO (0.25–3.00 mL, 10% w/w in toluene, ~0.38–4.52 mmol
of “AlMeO”) was added, followed by toluene such that the total volume
of toluene added was consistently 3.0 mL. The mixture was then stirred
at RT (15 min). A sample (0.7 mL) was then transferred to a flame-dried
NMR tube under argon that nominally contained 0.5m total zincspecies
and 0.076–0.90m total aluminium species in a 3:2 v/v mixture of toluene/
THF. Equivalent runs at 228C and with greater amounts of MAO (up to
3.0 equiv) produced similar spectra (not shown) except that the final
1,4-addition of just “R” from RZn
rently underway.
ACHTRE(UNG RNT) species. This is cur-
Experimental Section
General: All experiments were carried out under an argon atmosphere
by using standard Schlenk techniques. Solvents were distilled from suita-
ble drying agents prior to use: THF and DME from sodium/benzophe-
none, 2-cyclohexenone was distilled and stored under argon. 1H and
13C NMR spectra were recorded by using either a Bruker AM400 or a
JEOL EX270 spectrometer, details of the NMR monitoring experiments
are given below. For all samples, d values were referenced to residual sol-
vent peaks. All J values are in Hz. Infrared spectra were recorded by
using a Perkin Elmer 1600 Series FTIR machine. Mass spectra were re-
corded by using electron ionisation (EI) or chemical ionisation (CH4, CI)
techniques using a Micromass 70E machine. GC analyses was performed
by using a Varian 3380 gas chromatograph through a Lipodex A or BP20
column, from which the yield was determined by using an appropriately
calibrated internal standard. TLC was performed on Merck Silicagel 60
F254+366 precoated plates (0.25 mm) silica. The plates were visualised by
the use of a combination of ultraviolet light (254 and 366 nm) and/or
aqueous potassium permanganate or anisaldehyde and then heat. Liquid
chromatography was by forced flow (flash chromatography) including the
solvent systems indicated by using silica gel 60 (220–240 mesh) supplied
chemical shift of the Zn
A
value K was determined for added MAO concentrations of up to 0.9m
through Equations (4) and (5). A plot of Equation (5) (using [ZnR]2
(R=Et) attained from Equation (4) using the crude NMR data) gave
K=0.19Æ0.05 (correlation coefficient 0.986).
Preparation of EtZnCl on a synthetic scale: To a dried Schlenk tube
under an argon atmosphere containing ZnCl2 (0.95 g, 7.0 mmol) dried
THF (12.3 mL) was added. The solution was stirred for 10min and then
neat Et2Zn (0.72 mL, 7.0 mmol) was added dropwise. The resulting
cloudy solution was stirred for 1 h and then the THF was removed under
high vacuum. The resulting white solid was then dissolved in the desired
solvent system and stirred for 20 min to give a 1.07m solution.
by Fluka. The compound [Cu(TC)] was prepared by
a literature
method,[13] all other compounds were commercial products, specifically:
Chem. Eur. J. 2007, 13, 2462 – 2472
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2469

S. Woodward et al.
Preparation of EtZnMe from EtZnCl and Me2Zn on a synthetic scale:
To a dried Schlenk tube under an argon atmosphere was added EtZnCl
(1.95 mL, 2.1 mmol) followed by dropwise addition of Me2Zn (2m tolu-
ene solution). The resulting cloudy solution was stirred for 30min and
then diluted if necessary to give a 0.41m solution.
(m, 1H), 1.48–1.42 ppm (m, 1H); 13C NMR (100 MHz, CDCl3): dC =
211.7, 136.3, 135.7, 129.1 (d), 129.0 (d), 47.9, 42.5, 41.4, 41.0, 30.9, 25.1,
21.0 ppm; IR (CHCl3): n˜ =2926, 2863, 1706, 1603, 1448, 1347, 1314 cmꢀ1
;
HRMS (EI): calcd for C14H18ONa+, 225.1255; found: 225.1236 [M+Na]+.
3-[(4-Isopropylphenyl)methyl]cyclohexanone (5d, R=CH2C6H4iPr-4):
Yield 322.5 mg (70%). 1H NMR (400 MHz, CDCl3, 258C): dH =7.14 (d,
J=8.0, 2H; Ar), 7.04 (d, J=8.0, 2H; Ar), 2.88 (sept, J=6.9, 1H; CH-
General procedure for the 1,4-addition of EtZnMe to 2-cyclohexenone,
preparation of 5a: A dried Schlenk tube was charged with CuACHTRE(UNG OTf)2
(3.62 mg, 0.01 mmol) and (R,S,S)-10 (10.79 mg, 0.02 mmol). Dry solvent
(0.3 mL) was introduced, the stirred mixture cooled to ꢀ308C and stirred
for 30min before EtZnMe (0.41m) was added. The resulting solution was
stirred for a further 30min and 2-cyclohexenone (48.5 mL, 0.5 mmol) was
added dropwise. Stirring was continued for 1 h at ꢀ308C, then the reac-
tion was quenched by cautious addition of aqueous HCl (2 mL, 2m).
Nonane (25 mL) was added and the organiclayer was passed through a
plug of silica and the yield determined by quantitative GC analysis. (R)-
3-Ethylcyclohexanone (5a) was attained, 60% (92%ee). Colourless oil;
1H NMR (400 MHz, CDCl3, 258C): dH =2.48–2.25 (m, 3H), 2.11–1.92 (m,
3H), 1.79–1.62 (m, 2H), 1.49–1.29 (m, 3H), 0.94 ppm (t, J=7.6, 3H;
CH2Me); 13C NMR (100 MHz, CDCl3): dC =212.2, 47.9, 41.5, 40.6, 30.9,
29.3, 25.3, 11.2 ppm. These data are concordant with published values
and the spectrum identical to an authentic sample.[21] The enantiomeric
excesses were determined by chiral GC (Lipodex A, isothermal 758C):
(R)-2 9.9 min; (S)-2 10.2 min.
AHCTRE(UNG CH3)), 2.60 (ABX, J=13.6, 6.5, 2H; CH2Ar), 2.40–2.20 (m, 3H), 2.09–
2.01 (m, 3H), 1.93–1.82 (m, 1H) 1.68–1.55 (m, 1H), 1.42–1.36 (m, 1H),
1.24 ppm (d, J=6.9, 6H; 2CH3); 13C NMR (100 MHz, CDCl3): dC =211.8,
146.7, 136.7, 129.0 (d), 126.4 (d), 47.8, 42.6, 41.5, 41.0, 33.7, 31.0, 25.2,
24.1 ppm (d); IR (CHCl3): n˜ =2961, 2928, 2871, 1705, 1510, 1449, 1347,
1314, 1055 cmꢀ1; HRMS (EI): calcd for C16H22ONa+: 253.1568; found:
253.1564 [M+Na]+.
3-[(4-Bromophenyl)methyl]cyclohexanone (5e, R=CH2C6H4Br-4): Yield
329.0 mg (62%). 1H NMR (400 MHz, CDCl3, 258C): dH =7.41–7.37 (m,
2H; Ar), 7.00–6.97 (m, 2H; Ar), 2.61–2.52 (m, 2H; CH2Ar), 2.37–2.31
(m, 2H), 2.27–2.20 (m, 1H), 2.06–1.98 (m, 3H), 1.87–1.82 (m, 1H) 1.66–
1.55 (m, 1H), 1.37–1.33 ppm (m, 1H); 13C NMR (100 MHz, CDCl3): dC =
211.1, 138.3, 131.4 (d), 130.7 (d), 120.0, 47.7, 42.2, 41.3, 40.6, 30.7,
24.9 ppm; IR (CHCl3): n˜ =2931, 2870, 1707, 1592, 1488, 1347, 1314, 1073,
1012 cmꢀ1; HRMS (EI): calcd for C13H15BrO: 266.0306; found: 266.0304
[M]+.
Ethyl-4-(3-oxocyclohexyl)butanoate (5 f, R=(CH2)3CO2Et): 1H NMR
(400 MHz, CDCl3, 258C): dH =4.09 (q, J=7.2, 2H), 2.44–2.18 (m, 3H),
2.25 (t, J=7.6, 2H), 2.04–1.95 (m, 2H), 1.92–1.55 (m, 5H), 1.39–1.25 (m,
3H), 1.22 ppm (t, J=7.2, 3H); 13C NMR (100 MHz, CDCl3): dC =211.8,
173.5, 60.3, 48.0, 41.5, 38.8, 36.0, 34.3, 31.1, 25.2, 22.1, 14.3 ppm; HRMS
(EI): calcd for C12H20O3+: 212.1412; found: 212.1407 [M]+. These data
are concordant with published values.[23]
Preparation of ArCH2ZnBr in DME: A dried Schlenk tube was charged
with granular zincpowder (1.83 g, 28.0 mmol; Arocs, 200 mesh,
99.9999% purity), heated under vacuum with a hot-air gun and backfilled
with argon. The evacuation/heating procedure was repeated three times.
Anhydrous DME (26.3 mL) and TMSCl (71 mL, 0.56 mmol) were intro-
duced and the mixture stirred vigorously at RT (30 min). The mixture
was cooled to 08C and the appropriate benzylicbromide (14.0 mmol)
added dropwise. The mixture was stirred at RT until no starting bromide
was present by TLC (1 h). Stirring was stopped and the supernatant
ArCH2ZnBr solution (0.5m in DME) removed by syringe once the
excess zinc powder had settled. Equivalent preparations in THF were
readily attained.
Representative preparation of tetracycle 6b: A dried Schlenk tube was
charged with [Cu(TC)] (7.6 mg, 0.01 mmol) and dry THF (4.0 mL). The
stirred mixture was cooled to ꢀ308C and MAO (10 wt% in toluene,
6 mmol, 6 mL) added. The yellow reaction mixture was stirred for 10min
at ꢀ308C after which BnZnBr (1.3 mL, 0.5m, THF, 0.65 mmol) was
added followed by 2-cyclohexenone (194 mL, 2.0 mmol). Stirring was con-
tinued for 1 h at ꢀ308C and the reaction was then quenched by cautious
addition of aqueous HCl (8 mL, 2m). The reaction mixture was extracted
with dichloromethane (2”30 mL), the organic phase was washed with
water (25 mL), brine (25 mL), dried (MgSO4) and the solvent evaporat-
ed. The product was isolated by flash chromatography using light petrole-
um (b.p. 40–608C)/Et2O (3:1) (Rf =0.20) then recrystallised using di-
chloromethane/hexane 58.5 mg (24%). Positional assignments use the no-
menclature of Figure 3. Geminal, anti, syn, gauche and long-range cou-
plings are indicated gemJ, antiJ, synJ , gauJ and 1,4J respectively. 1H NMR
(400 MHz, CDCl3, 258C): dH =7.28 (app. t, J=7.3, 2H; Phm), 7.20 (app. t,
J=7.3, 1H; Php), 7.16 (app. d, J=7.3, 2H; Pho), 4.41 (d, 1,4J=2.0, 1H;
OH), 3.07 (t, antiJ=11.2, 1H; HC7), 2.95 (dd, gemJ=13.2, gauJ=2.2, 1H;
axHC2), 2.64 (m, antiJ=11.2, 10.4, plus unresolved couplings;1H; axHC5),
2.51 (d, synJ=10.4, 1H; axHC6), 2.49–1.76 (m, 14H; mixed ring CH and
CH), 1.69–1.52 (m, 3H; mixed CH2), 1.35 (m, gemJ=13.2, plus unresolved
couplings,1H; CH2Ph), 1.23 (ddd, antiJ=10.4, synJ=8.8, 1,4J=2.0, 1H;
axHC13), 1.18–1.04 (m, 2H; ring CH2), 0.85–0.72 ppm (m, 1H; ring CH2);
13C NMR (100 MHz, CDCl3): dC =219.1, 213.4, 140.5, 129.1, 128.2, 125.9,
73.7, 55.9, 55.1, 45.0, 44.9, 43.1, 41.8, 41.4, 40.7, 35.6, 34.2, 31.5, 29.2, 28.2,
27.5, 20.7, 20.2 ppm; IR (CHCl3): n˜ =3605, 2934, 2864, 1703, 1600, 1454,
1391, 1004 cmꢀ1; HRMS (EI): calcd for C25H32O3Na+: 403.2249; found:
403.2244 [M+Na]+.
General procedure for 1,4-addition of ArCH2ZnBr to 2-cyclohexenone
(leading to 5b–f): A dried Schlenk tube was charged with [Cu(TC)]
(7.6 mg, 0.01 mmol) and dry DME (4.0 mL). The stirred mixture was
cooled to ꢀ308C and MAO (10 wt% in toluene, 6.0–24.0 mmol, 3–
12 equiv) added followed by toluene such that the total volume of tolu-
ene added remained 16.0 mL. The yellow reaction mixture was stirred for
10 min at ꢀ308C after which RZnBr (8.0 mL, 0.5m, DME, 4.0 mmol) was
added followed by 2-cyclohexenone (194 mL, 2.0 mmol). Stirring was con-
tinued for 1 h at ꢀ308C after which time the reaction was quenched by
cautious addition of aqueous HCl (8 mL, 2m). The reaction mixture was
extracted with Et2O (2”25 mL), the organicphase was washed with
water (25 mL) and brine (25 mL), and dried (MgSO4) and the solvent
evaporated. The product was isolated by flash chromatography using
light petroleum (b.p. 40–608C)/Et2O (7:1 or 5:1).
Alternatively for the purpose of GC analysis, the same reaction was car-
ried out using 2-cyclohexenone on a 0.5 mmol scale. The reaction mixture
was quenched as above, followed by the addition of tridecane (C13H28)
(25 mL). The organiclayer was passed through a plug of silica and then
analysed by GC. Isolated yields were within 5% of those determined by
GC.
3-Benzylcyclohexanone (5b, R=CH2Ph): Yield 271.1 mg (72%);
1H NMR (400 MHz, CDCl3, 258C): dH =7.26–7.14 (m, 5H; Ar), 2.64–2.58
(m, 2H; CH2Ar), 2.41–2.00 (m, 6H), 1.86–1.78 (m, 1H), 1.62–1.55 (m,
1H), 1.42–1.35 ppm (m, 1H); 13C NMR (100 MHz, CDCl3): dC =211.7,
139.4, 129.1 (d), 128.4 (d), 126.2, 47.9, 43.0, 41.4, 41.0, 30.9, 25.1 ppm;
HRMS (EI): calcd for C13H16ONa+: 211.1099; found: 211.1105 [M+Na]+;
GC (BP20, initial temperature 1208C for 15 min, then ramp 208Cminꢀ1
to 1808C) retention time 29.8 min. These data are concordant with pub-
lished values.[22]
X-ray crystal structure determination of 6b: Due to the very small size of
the available crystals, it was necessary to collect diffraction data on Sta-
tion 9.8 of the high-intensity Daresbury Synchrotron Radiation Source.
Colourless needle, 0.08”0.01”0.01 mm3, monoclinic, space group P21/c;
a=19.6178(13), b=5.3724(4), c=19.0540(12) Š; b=97.207(2)8; V=
1992.3(2) Š3; 1calcd =1.269 gcmꢀ3; l=0.6727 Š; 2qmax =608; fine-slice w
3-[(4-Tolyl)methyl]cyclohexanone
(5c,
R=CH2C6H4Me-4):
Yield
scans; T=120(2) K; 22629 reflections measured, 6374 unique (Rint =
242.2 mg (60%). 1H NMR (400 MHz, CDCl3, 258C): dH =7.11 (d, J=8.9,
2H; Ar), 7.03 (d, J=8.9, 2H; Ar), 2.60 (d, J=6.8, 2H; CH2), 2.34 (s, 3H;
Me), 2.42–2.26 (m, 3H), 2.11–2.02 (m, 3H), 1.92–1.88 (m, 1H), 1.70–1.64
0.041) used in refinement; corrections for Lorentz, polarisation and ab-
sorption (m=0.081 mmꢀ1, T range 0.993–0.999) effects were applied. The
structure was solved by direct methods using SHELXS97[24] and refined
2470
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 2462 – 2472

Promoted ZincEquilibria
FULL PAPER
by full-matrix least squares using SHELXL97,[25] 254 parameters, H
atoms placed geometrically and thereafter refined using a riding model,
R1 [4719 F=4s(F)]=0.0505, wR2 [all 6375 F2]=0.144, refinement on F2,
D1max, min =0.37, ꢀ0.27 eŠꢀ3. CCDC-604125 contains the supplementary
crystallographic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
Tetracycle 6 f (R=(CH2)3CO2Et): 1H NMR (400 MHz, CDCl3, 258C):
dH =4.38 (s, 1H; OH), 4.12 (qd, J=7.2, 0.8, 2H), 2.98 (t, J=11.2, 1H),
2.48–2.36 (m, 3H), 2.46 (d, J=10.4, 1H), 2.33–2.21 (m, 4H), 2.18–2.02
(m, 2H), 2.05–1.38 (m, 14H), 1.26 (t, J=7.2, 3H), 1.15–1.04 (m, 3H),
0.95–0.75 ppm (m, 2H); 13C NMR (100 MHz, CDCl3): dC =219.1, 213.5,
173.6, 73.5, 60.3, 55.8, 55.0, 44.9, 44.7, 43.1, 41.8, 41.3, 34.6, 34.2, 33.8,
33.5, 31.6, 29.2, 28.2, 27.5, 21.9, 20.7, 20.3, 14.3 ppm; IR (CHCl3): n˜ =
3505, 2936, 2864, 1709, 1602, 1455, 1372, 1045 cmꢀ1; HRMS (EI): calcd
for C24H36O5+: 404.2563; found: 404.2568 [M]+.
weighted average of the number of “R” groups in each environment al-
lowing for the presence of two “R” groups in ZnR2 leading to Equa-
tion (A8):
2 RZnXþMAO Ð ZnR2þZnX2 Á MAO
dX½RZnXŠþdR22½ZnR2Š
ðA7Þ
ðA8Þ
dobs
¼
½Zntot
Š
For which Equations (A3), (A4) and (A2) are all valid.
As:
½ZnR2Š ¼ ½ZnX2 Á MAOŠ
ðA2Þ
Preparation of ZnACHTREUNG
(CH2Ph)2: Literature preparations[18] of this compound
were found to generate samples significantly contaminated with
PhCH2ZnCl, especially after exposure to 1,4-dioxane (as recommend-
ed[18]). The following procedure was found to be more reliable: A dried
Schlenk tube under an argon atmosphere was charged with benzylmagne-
sium chloride (8.0 mL, 1.0m Et2O solution, 8.0 mmol), and ZnCl2
(4.0 mL, 1.0m Et2O solution, 4.0 mmol) was added at RT over 10 min
through a cannula to give a white suspension. The suspension was stirred
for 24 h at RT and then filtered through Celite to give a 0.25m colourless
filtrate. 1H NMR[18] (500 MHz, C6D6, 258C): dH =7.32–7.28 (m, 2H),
7.25–7.21 (m, 2H), 7.02–6.98 (m, 1H), 2.05 ppm (s, 2H); 13C NMR
(125 MHz, C6D6): dC =151.0, 127.9, 126.4, 119.7, 21.2 ppm.
Therefore substituting Equation (A4) into (A8) leads directly to Equa-
tion (4).
Acknowledgements
This work was supported by the European Union Framework 6 Pro-
gramme (Project: FP6-505267-1; LIGBANK, J.C.S.) and we thank the
EPSRC-funded synchrotron crystallography service and its director Pro-
fessor W. Clegg for collection of single-crystal diffraction data at Dares-
bury SRS Station 9.8. J.S. thanks the EPSRC for its contribution to a stu-
dentship. We acknowledge Dr. Chris Frost (University of Bath) for useful
discussions.
Appendix
Derivation of Equation (5): Under the conditions of Equilibrium (2) the
equilibrium constant is expected to be Equation (A1):
[1] a) P. J. Goldsmith, S. Woodward, Angew. Chem. 2005, 117, 2275–
2277; Angew. Chem. Int. Ed. 2005, 44, 2232–2234.
[2] Evans estimated K<0.002m based on the non-observation of ZnEt2
in the 1H NMR spectra of EtZnI; D. F. Evans, G. V. Fazakerley, J.
Chem. Soc. A 1971, 182–183.
½ZnR2Š½ZnX2 Á MAOŠ
K ¼
ðA1Þ
2
½RZnXŠ ½MAOŠ
[3] a) Boersma detected only EtZnCl in equimolar mixtures of ZnEt2/
ZnCl2: J. Boersma, J. G. Noltes, Tetrahedron Lett. 1966, 7, 1521–
1525; b) the species PhZnMe has been prepared in situ by admix-
ture of ZnPh2+ZnMe2; P. I. Dosa, G. C. Fu, J. Am. Chem. Soc. 1998,
120, 445–446; c) P. I. Dosa, J. C. Ruble, G. C. Fu, J. Org. Chem.
1997, 62, 444–445.
As every turnover over of the equilibrium produces equimolar quantities
of ZnR2 and [ZnBr2·MAO] then:
½ZnR2Š ¼ ½ZnX2 Á MAOŠ
ðA2Þ
[4] a) H. Yorimitsu, K. Oshima, Angew. Chem. 2005, 117, 4509–4513;
Angew. Chem. Int. Ed. 2005, 44, 4435–4439.
[5] a) P. Knochel, J. J. A. Perea, P. Jones, Tetrahedron 1998, 54, 8275–
8319; b) P. Knochel, P. Jones, Organozinc Reagents: A Practical Ap-
proach, Oxford Press, 1999.
[6] a) R. I. Yousef, B. Walfort, T. Ruffer, C. Wagner, H. Schmidt, R.
Herzog, D. Steinborn, J. Organomet. Chem. 2005, 690, 1178–1191;
b) J. Tammiku-Taul, P. Burk, A. Tuulmets, J. Phys. Chem. A 2005,
108, 133–139; c) P. E. M. Allen, S. Hagiass, C. Mair, E. H. Williams,
Ber. Bunsen-Ges. Phys. Chem. 1984, 88, 623–626.
The total amount of zincin the system must be a ocnstant, defining
Equation (A3). By using the relationship given in Equation (A2) an ex-
pression for the concentration of RZnBr can be defined [Eq. (A4)]:
½ZntotŠ ¼ ½RZnXŠþ½ZnR2Šþ½ZnX2 Á MAOŠ
½RZnXŠ ¼ ½ZntotŠꢀ2½ZnR2Š
ðA3Þ
ðA4Þ
[7] E. Zurek, T. Ziegler, Prog. Polym. Sci. 2004, 29, 107–148.
[8] a) For a review on inverse-crown chemistry see: R. E. Mulvey, Orga-
nometallics 2006, 25, 1060–1075; b) for information on anti-crown
chemistry see: I. A. Tikhonova, K. I. Tugashov, F. M. Dolgushin,
A. A. Yakovenko, B. N. Strunin, P. V. Petrovshii, G. G. Furin, V. B.
Shur, Inorg. Chim. Acta 2006, 359, 2728–2735; c) V. B. Shur, I. A.
Tikhonova, Russ. Chem. Bull. 2004, 52, 2539–2554.
Similarly, the total amount of aluminium present in the system must be
constant leading directly to Equations (A5) and (A6):
½AltotŠ ¼ ½MAOŠþ½ZnX2 Á MAOŠ
½AltotŠ ¼ ½MAOŠþ½ZnR2Š
ðA5Þ
ðA6Þ
[9] S. Woodward, Tetrahedron 2002, 58, 1017–1050.
[10] Y. Xiaoguang, Z. Zhiping, C. B. Knobler, M. F. Hawthorne, J. Am.
Chem. Soc. 1993, 115, 193–195.
[11] a) A. R. Kennedy, R. E. Mulvey, R. B. Rowlings, Angew. Chem.
1998, 110, 3321- 3323; Angew. Chem. Int. Ed. 1998, 37, 3180–3183;
b) A. R. Kennedy, R. E. Mulvey, B. A. Roberts, R. B. Rowlings,
C. L. Raston, Chem. Commun. 1999, 353–354.
Substitution of Equations (A2), (A4) and (A6) into (A1) leads directly
to Equation (5).
Derivation of Equation (4): Under the high temperature limit of Equa-
tion (A7) the observed chemical shift of the ZnCH2 carbon must be the
Chem. Eur. J. 2007, 13, 2462 – 2472
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2471

S. Woodward et al.
[12] Attained by use of K=0.19 in Equation (5), attaining [ZnEt2] by
use of the Excel “solver” function, see: J. E. Billio, Excel for Chem-
ists, 2nd ed., Wiley, NY, 2001.
[13] G. D. Allred, L. S. Liebeskind, J. Am. Chem. Soc. 1996, 118, 2748–
2749.
[14] P. K. Fraser, S. Woodward, Chem. Eur. J. 2003, 9, 776–783.
[15] G. H. Posner, J. P. Mallomo, A. Y. Black, Tetrahedron 1981, 37,
3921–3926.
[16] R. L. Cargill, D. F. Bushey, J. J. Good, J. Org. Chem. 1979, 44, 300–
301.
[17] L. I. Zakharkin, O. Y. Okhlobystin, Izv. Akad. Nauk SSSR Ser.
Khim. 1963, 193.
[18] V. Weissig, R. Beckhaus, U. Banasiak, T. H. Z Thiele, Z. Anorg.
Allg. Chem. 1980, 467, 61.
[19] H. Nehl, W. R. Scheidt, J. Organomet. Chem. 1985, 289, 1–8.
[20] a) C. A. Ogle, J. B. Human, Tetrahedron: Asymmetry 2003, 14,
3281–3283; b) F. K. Zinn, E. C. Ramos, J. V. Comasseto, Tetrahe-
dron Lett. 2001, 42, 2415–2417; c) P. Jones, C. K. Reddy, P. Knochel,
Tetrahedron 1998, 54, 1471–1490; d) M. van Klaveren, E. S. M. Pers-
son, D. M. Grove, J.-E. Bäckvall, G. van Koten, Tetrahedron Lett.
1994, 35, 5931–5934; e) B. H. Lipshutz, M. Koerner, D. A. Parker,
Tetrahedron Lett. 1987, 28, 945–948; f) B. H. Lipshutz, J. A. Kozlow-
ski, D. A. Parker, S. L. Nguyen, K. E. McCarthy, J. Organomet.
Chem. 1985, 285, 437–447.
[21] M. P. Bertrand, L. Feray, R. Nouguier, P. Perfetti, J. Org. Chem.
1999, 64, 9189–9193.
[22] A. Yanagisawa, S. Habaue, K. Yasue, H. Yamamoto, J. Am. Chem.
Soc. 1994, 116, 6130–6141.
[23] P. Knochel, P. Jones, C. K. Reddy, Tetrahedron 1998, 54, 1471–1490.
[24] G. M. Sheldrick, Acta Crystallogr. Sect. A 1990, 46, 467–473.
[25] SHELXL97, G. M. Sheldrick, University of Gçttingen, Gçttingen,
Germany, 1998.
Received: December 4, 2006
Published online: February 7, 2007
2472
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 2462 – 2472