Structural and reactivity insights in Mg-Zn hybrid chemistry: Zn-I exchange and Pd-catalysed cross-coupling applications of aromatic substrates

Article abstract of DOI:10.1039/c4sc01263a

Expanding the synthetic potential of Mg-Zn hybrid organyl reagents (generated via transmetallation reactions), this study uncovers a versatile approach, involving a sequence of direct Zn-I exchange and Pd catalysed cross-coupling reactions which grants access to a wide range of asymmetric bis(aryls). By combining X-ray crystallography with ESI-MS and non-deuterium NMR spectroscopic studies, new light is shed on the heterobimetallic constitution of the intriguing organometallic species [(THF)₄MgCl ₂Zn(tBu)Cl] (1) and [{Mg₂Cl₃(THF) ₆}⁺{ZntBu₃}^-] (2), formed through transmetallation of tBuMgCl with n equivalent amounts of ZnCl₂ (n = 1 and 3 respectively). Operating by cooperative effects, alkyl-rich hybrid 2 can effectively promote direct Zn-I exchange reactions with aromatic halides in short periods of time at room temperature in THF solution. The structural elucidation of key organometallic intermediates involved in some of these Zn-I exchanges, provides new reactivity insights into how these bimetallic systems operate. Thus, while the reaction of 2 with 3 equivalents of 2-iodoanisole (3b) gives magnesium dizincate [{Mg(THF)₆}²⁺{Zn(o-C ₆H₄-OMe)₃}₂^-] (4) which demonstrates the 3-fold activation of the tBu groups attached to Zn in 2, using 2-iodobenzonitrile (3i), only two tBu groups react with the substrate, affording [(THF)₄MgCl(NC-o-C₆H₄)ZnI(o-C ₆H₄-CN)(THF)] (7). In 7 Mg and Zn are connected by an aryl bridge, suggesting that the formation of contacted ion-pair hybrids may have a deactivating effect on the outcome of the Zn-I exchange process. A wide range of homoleptic tris(aryl) zincate intermediates have been prepared in situ and used as precursors in Pd catalysed cross-coupling reactions, affording bis(aryls) 6a-s in excellent yields under mild reaction conditions without the need of any additive or polar cosolvent such as NMP or DMI.

Full text of DOI:10.1039/c4sc01263a

Chemical
Science
View Article Online
View Journal | View Issue
EDGE ARTICLE
Structural and reactivity insights in Mg–Zn hybrid
chemistry: Zn–I exchange and Pd-catalysed cross-
coupling applications of aromatic substrates†‡
Cite this: Chem. Sci., 2014, 5, 3552
Tobias D. Bluemke,a William Clegg,b Pablo Garcıa-Alvarez,c Alan R. Kennedy,a
´
d
a
b
a
*
Konrad Koszinowski, Matthew D. McCall, Luca Russo and Eva Hevia
Expanding the synthetic potential of Mg–Zn hybrid organyl reagents (generated via transmetallation
reactions), this study uncovers a versatile approach, involving a sequence of direct Zn–I exchange and
Pd catalysed cross-coupling reactions which grants access to a wide range of asymmetric bis(aryls). By
combining X-ray crystallography with ESI-MS and non-deuterium NMR spectroscopic studies, new light
is shed on the heterobimetallic constitution of the intriguing organometallic species [(THF)4MgCl2Zn(tBu)
Cl] (1) and [{Mg2Cl3(THF)6}+{ZntBu3}ꢀ] (2), formed through transmetallation of tBuMgCl with n equivalent
amounts of ZnCl2 (n ¼ 1 and 3 respectively). Operating by cooperative effects, alkyl-rich hybrid 2 can
effectively promote direct Zn–I exchange reactions with aromatic halides in short periods of time at
room temperature in THF solution. The structural elucidation of key organometallic intermediates
involved in some of these Zn–I exchanges, provides new reactivity insights into how these bimetallic
systems operate. Thus, while the reaction of 2 with 3 equivalents of 2-iodoanisole (3b) gives magnesium
dizincate [{Mg(THF)6}2+{Zn(o-C6H4-OMe)3}2ꢀ] (4) which demonstrates the 3-fold activation of the tBu
groups attached to Zn in 2, using 2-iodobenzonitrile (3i), only two tBu groups react with the substrate,
affording [(THF)4MgCl(NC-o-C6H4)ZnI(o-C6H4-CN)(THF)] (7). In 7 Mg and Zn are connected by an aryl
bridge, suggesting that the formation of contacted ion-pair hybrids may have a deactivating effect on
the outcome of the Zn–I exchange process. A wide range of homoleptic tris(aryl) zincate intermediates
have been prepared in situ and used as precursors in Pd catalysed cross-coupling reactions, affording
bis(aryls) 6a–s in excellent yields under mild reaction conditions without the need of any additive or
polar cosolvent such as NMP or DMI.
Received 2nd May 2014
Accepted 13th June 2014
DOI: 10.1039/c4sc01263a
www.rsc.org/chemicalscience
and reactivity studies in organolithium chemistry have exposed
the rich structural diversity exhibited by these reagents.4 All
Introduction
these polar compounds are used daily worldwide in synthetic
laboratories and industrial processes and have found numerous
applications.5 Many of these important reagents (oen depicted
by a single simple formula such as RM, RMX; M ¼ Li, Mg, Zn,
Cu; X ¼ halide) are, in fact, dynamic multi-component mixtures in
solution which can contain one, or more reactive organome-
tallic species, in different aggregated and/or solvated states.6
Thus, understanding and controlling the complex constitution
of these reagents, not only in the solid state but even more
importantly, in solution, is crucial since structural variations
such as solvation and aggregation can oen nely tune their
reactivity and selectivity.7
The reactivity of any organometallic reagent is inherently linked
to its structure or structures in solution.1 For example, it is well
known, that the composition of organomagnesium reagents
(Grignard reagents) in solution is best described in terms of the
Schlenk equilibrium.2 Similar equilibria have also been
proposed and reported for organozinc and organoaluminium
reagents.3 In addition, a multitude of structural, spectroscopic
aWestCHEM, Department of Pure and Applied Chemistry, University of Strathclyde,
Glasgow, G1 1XL, UK. E-mail: eva.hevia@strath.ac.uk
bSchool of Chemistry, Newcastle University, Newcastle upon Tyne, NE1 7RU, UK
c
´
´
´
Departamento de Quımica Organica e Inorganica-IUQOEM, Universidad de Oviedo-
Adding a new level of complexity to this area, recent studies
have revealed that in many cases the reactivity of these
compounds can be greatly affected by the synthetic method
chosen for their preparation. For example, magnesium reagents
prepared by oxidative metal insertion react generally poorer in
Pd-catalysed cross coupling reactions, than the same reagent,
prepared by an iodine–magnesium exchange reaction.8 Lei has
CSIC, E-33071 Oviedo, Spain
dInstitut fur Organische und Biomolekulare Chemie, Georg-August Universitat
¨
¨
¨
¨
Gottingen, Tammannstraße 2, 37077 Gottingen, Germany
† Electronic supplementary information (ESI) available: Full experimental details,
X-ray crystallographic data and copies of the NMR and ESI-MS spectra. CCDC
1000413 and 1000414. For ESI and crystallographic data in CIF or other
electronic format see DOI: 10.1039/c4sc01263a
‡ Dedicated to the memory of Professor Ken Wade.
3552 | Chem. Sci., 2014, 5, 3552–3562
This journal is © The Royal Society of Chemistry 2014

View Article Online
Edge Article
Chemical Science
reported a dramatic reactivity difference for arylzinc ArZnCl organometallic species, which were assigned to the single metal
reagents in Ni-catalysed homo-coupling reactions, depending components tBu2Zn and tBuMgCl. Preliminary reactivity
on whether they are prepared by a transmetallation reaction studies showed that THF solutions of 2 undergo a metal–iodine
from lithium-, or from magnesium reagents.9 A similar effect exchange reaction with 4-iodotoluene allowing the isolation of a
has been noticed by Bedford in iron-catalysed cross-coupling magnesium tris(aryl)zincate intermediate resulting from a
reactions of bis(aryl)zinc reagents with benzyl halides.10 three-fold activation of all the tBu groups present in hybrid 2.
Furthermore, Knochel has shown that zinc reagents prepared
Building on these preliminary ndings, here we provide a
by transmetallation reaction starting from organo- comprehensive study, exploring the synthetic applications of
a
magnesiums, are generally more nucleophilic towards carbonyl Mg–Zn-hybrid reagents, generated via salt-metathesis (trans-
functionalities, than zinc reagents, prepared by oxidative metallation) reactions in metal–halogen exchange and Pd cat-
insertion of zinc powder.11
alysed cross-coupling processes. Considering the complex
In situ metathesis (or transmetallation) approaches, where a composition of 2 in THF solutions, where at least three main
metal ionic salt (such as MgX2, ZnX2, AlX3; X ¼ halide) is reacted species can be detected (tBu3Znꢀ, tBu2Zn and tBuMgCl), this
with a high polarity organometallic reagent (commonly RLi or study assesses which of these components could be responsible
RMgX) constitute common methodologies to prepare low for the exchange reaction, or, in other words, whether the
polarity organometallic reagents.12 These reactions are favoured process taking place is a genuine halogen–zinc exchange or
by the concomitant formation of LiX or MgX2 ionic species alternatively a halogen–magnesium exchange reaction (fol-
which in solution should coexist with the newly generated lowed by a transmetallation step). Seeking an answer to this
organometallic species. Although their presence in the reaction fundamental question and in order to get a more detailed
media has oen been overlooked, a urry of recent reports has picture of the organometallic species participating in such
highlighted that, far from being mere spectators, these salts can transmetallation mixtures, a series of comparative reactivity
greatly inuence the constitution and reactivity of the newly studies have been undertaken, in combination with multinu-
formed organometallic reagent.13 Some of these studies propose clear NMR spectroscopic and electrospray ionization (ESI) mass
the involvement of mixed-aggregates, resulting from the co- spectrometric (MS) analysis.
complexation of the low-polarity organometallic reagent and
Furthermore, we present a new synthetically useful protocol,
the salt byproduct (in a metastable molecular form distinct allowing a stepwise sequence of transmetallation (generation of
from the precipitated ionic lattice form), as a plausible expla- Mg–Zn hybrid 2 in situ), iodine–zinc exchange with an aryl iodide
nation for the observed salt-effects.14 Notwithstanding, it (formation of a tris(aryl) intemediate) and Pd catalysed cross-
should be noted that tangible proof on the formation of these coupling reactions to be carried out. The scope and limitations
mixed-metal species is limited and little information is avail- of this new iodine–zinc exchange, as well as trapping and
able regarding their constitutions (either in solution or in the characterization of key reactions intermediates, which provide
solid state).15
new reactivity insights into how these bimetallic systems oper-
Shedding some light on this intriguing area, in a preliminary ate, are also presented in this article.
report studying the seemingly straightforward metathesis reac-
tions of ZnCl2 with tBuMgCl, we disclosed the formation of novel
Results and discussion
mixed-metal salt intermediates, Mg–Zn hybrids.16 Depending on
Assessing the reactivity of n equivalent RMgCl–ZnCl2 (n ¼ 1, 3)
the reaction stoichiometry employed bimetallic species
[(THF)4MgCl2Zn(tBu)Cl] (1) and [{Mg2Cl3(THF)6}+{ZntBu3}ꢀ] (2)
could be isolated and structurally dened (Scheme 1).
combinations in metal–iodine exchange processes
Different stoichiometric ratios of freshly titrated, commercially
In the solid state, tris(alkyl)zincate 2 exists as a solvent- available tBuMgCl solution and ZnCl2 (1 : 1 and 3 : 1) were
separated ion pair (SSIP). However, NMR spectroscopic analysis prepared in THF and the resulting organometallic mixtures
of re-dissolved crystals of 2 in d8-THF revealed that this were tested in the halogen–metal exchange reaction using
compound appears to be in equilibrium with at least two other 4-iodoanisole (3a) as typical model substrate. Alkyl magnesium
compounds are known to undergo halogen–metal exchange
reactions.17 In accordance with this, tBuMgCl showed some
reactivity in the exchange reaction with 3a, although this
exchange proceeded slowly at 0–20 ꢁC and aer 24 h only 27% of
4-iodoanisole was consumed. No exchange reaction was
observable within the rst 30 minutes (Fig. 1A). Interestingly, on
adding 1 equivalent of ZnCl2 to tBuMgCl, this limited (but
observable) exchange reactivity was completely suppressed.
Similarly, a 1 : 1 mixture of iPrMgCl and ZnCl2 showed no
reactivity with 3a, whereas iPrMgCl on its own reacts efficiently
with this substrate at 0 ꢁC within only 5–10 minutes. These
ndings suggest that under these conditions the trans-
metallation reaction between the two alkyl magnesium reagents
Scheme 1 Synthesis of Mg–Zn hybrids 1 and 2 via salt-metathesis
reactions.
(RMgCl; R ¼ tBu, iPr) and ZnCl2 must be complete forming
This journal is © The Royal Society of Chemistry 2014
Chem. Sci., 2014, 5, 3552–3562 | 3553

View Article Online
Chemical Science
Edge Article
Table 1 Direct comparison of the reactivity of Grignard reagents
(RMgX) and the corresponding salt-metathesis mixtures RMgCl–ZnCl2
(3 : 1) in the halogen–metal exchange reactions with 3a
Entry
Ra
RMgClb
RMgCl–ZnCl2 (3 : 1)b,c
1
2
3
4
5
Me
Et
nBu
iPr
tBu
2%
40%
12%
86%
0%
19%
80%
>95%
>95%
>95%
a
Commercially available THF solutions of alkylmagnesium reagents
b
were used, as obtained. Conversion (depletion) of 4-iodoanisole
determined by GC-analysis of hydrolysed reaction aliquots with an
internal standard (tetradecane). The exchange reagent was prepared
in situ by mixing RMgCl (3.0 mmol) and ZnCl2 (1.0 mmol) in THF.
Fig. 1 Reaction of 4-iodoanisole (3a, 3 mmol) with (A): tBuMgCl
(3 mmol); (B): tBuMgCl (3 mmol) + ZnCl2 (3 mmol); (C): tBu2Zn
(1.5 mmol); (D): tBuMgCl (3 mmol) + ZnCl2 (1 mmol); (E): tBu2Zn
(1.5 mmol) + tBuMgCl (1.5 mmol); conversions have been determined
by GC-analysis of hydrolyzed reaction aliquots using an internal
standard (tetradecane) monitoring the depletion of 3a. The error is
estimated to be ꢂ3%.
c
readily achieved aer 10 min reaction time. Nevertheless, the
combination of iPrMgCl with ZnCl2 (3 : 1) led again to an
accelerated exchange, affording complete conversion of 3a in
the same time (Table 1, Entry 4). These ndings revealed that
halide-rich Mg–Zn hybrid 1, a contacted ion pair (CIP) structure,
which is not reactive enough for the desired transformation
(Fig. 1B).
Halide-free tBu2Zn (puried by sublimation)18 did not
undergo any halogen–zinc exchange reaction (Fig. 1C). Also,
combining tBu2Zn with 2 molar equivalents of anhydrous
MgCl2 did not facilitate the exchange-reaction. In contrast, on
treating 3a with a 3 : 1 mixture of tBuMgCl–ZnCl2 (presumably
forming tris(alkyl) hybrid 2 as a major species in solution) a
rapid exchange reaction took place leading to almost full
the
in
situ
formed
ate
complexes
(presumably
[{Mg2Cl3(THF)6}+{R3Zn}ꢀ]) are signicantly more reactive in
halogen–metal exchange processes than the corresponding
Grignard reagents from which they are derived. The effect is
most pronounced in the case of the tertiary alkylmagnesium
reagent tBuMgCl where the greatest acceleration is observed
(Table 1, Entry 5).
NMR spectroscopic studies
conversion to the corresponding aryl organometallic product, To acquire a deeper understanding of the constitution of the
ꢁ
within only 10–15 minutes at 0 C (Fig. 1D). Comparing these organometallic species involved in these transmetallation
exchange rates, the 3 : 1 mixture reacted approximately reactions multinuclear (1H and 13C) studies on tBuMgCl–ZnCl2
500 times faster with 4-iodoanisole than did tBuMgCl.19 Tri- mixtures were undertaken. Previously we have shown that when
s(alkyl)zincate 2 can also be obtained by the combination of isolated crystals of [{Mg2Cl3(THF)6}+{ZntBu3}ꢀ] (2) are dissolved
tBu2Zn with one molar equivalent of tBuMgCl. In fact, such in deuterated THF, this Mg–Zn hybrid is in dynamic equilib-
mixtures displayed the same high reactivity in the exchange rium with its homometallic components.16 It should be noted
reaction (Fig. 1E).
that although 2 is the dominant species present in solution, this
Interestingly, a similar accelerating effect upon addition of equilibrium is highly sensitive to its concentration in these
ZnCl2 was observed for other simple alkyl Grignard reagents THF-d8 solutions and under diluted conditions, the dispro-
(Table 1). Reactions of 3a with RMgCl (where R ¼ Me, Et, Bu, portionation of 2 into tBuMgCl and tBu2Zn becomes much
iPr), as well as the corresponding 3 : 1 mixtures of RMgCl–ZnCl2 more favourable.21 A solvent effect has also been noticed. Thus
were studied under the same conditions. Primary alkylmagne- in non-polar deuterated benzene, 2 exists as a single species
1
sium reagents (RMgX; R ¼ Me, Et, Bu) are generally inefficient displaying one signal in its H NMR for the tBu groups (d 1.12
exchange reagents. In contrast, the mixing of these magnesium ppm) along with two multiplets at 1.38 and 3.69 ppm corre-
1
reagents with ZnCl2 in the aforementioned 3 : 1 stoichiometry sponding to the THF ligands bonded to Mg. The H and 13C-
prior to the addition of 3a, led in all three cases to a faster NMR spectra of crystalline 2 in THF-d8 solution and a freshly
exchange reaction (19–100%, Table 1, Entries 1–3). The prepared mixture of tBuMgCl and ZnCl2 in a 3 : 1 stoichiometry
secondary alkylmagnesium reagent iPrMgCl is a well-known proved to be almost identical, showing that Mg–Zn hybrid 2 is
halogen–magnesium exchange reagent.20 Even in the case of an the main component in solution and is not only formed during
electron rich aromatic iodide (3a) a conversion of 86% was crystallization.
3554 | Chem. Sci., 2014, 5, 3552–3562
This journal is © The Royal Society of Chemistry 2014

View Article Online
Edge Article
Chemical Science
In order to obtain further knowledge of the species in solu- of the above-mentioned equilibrium towards the production of
tion, the constitution of these tBuMgCl–ZnCl2 mixtures was also bimetallic 2 which can subsequently react with more substrate.
studied by collecting their NMR spectra in non-deuterated THF
(No Deuterium-NMR).22 This approach avoids the tedious
removal and replacement of the solvent by expensive THF-d8
ESI mass spectrometric studies
Complementary insight into the nature of species involved in
these reaction mixtures was obtained from ESI-mass spec-
trometry. Contrasting with NMR spectroscopy, this technique
selectively probes the charged components present in solu-
tion.25 Some of our previous studies have shown that ESI-MS can
be a useful tool for the detection of solvent-separated organo-
zincate anions present in THF solutions.26,14e Thus, mixtures of
n tBuMgCl (n ¼ 1 and 3) and ZnCl2 in THF were prepared, and
diluted aliquots (ꢃ25 mM) were introduced into the ESI-MS
spectrometer (refer to ESI† for experimental details).27
tBuMgCl–ZnCl2 in a 1 : 1 ratio. Upon addition of 1 equiva-
lent of ZnCl2 to a solution of tBuMgCl in THF, the formation of
tert-butyl zincate anion {tBu3Zn3Cl4}ꢀ is observed (Fig. 2) as the
predominant species as well as {tBu2Zn2Cl3}ꢀ (m/z ¼ 351) and
{tBuZnCl2}ꢀ (m/z ¼ 195, off-scale in Fig. 2).28 The formation of
{tBu2Zn2Cl3}ꢀ can be rationalized as the stepwise complexation
of tBuZnCl to {tBuZnCl2}ꢀ, with the equilibrium lying on the
side of the higher aggregate (note that the concentration
enrichment during the ESI process supposedly shis the equi-
librium toward higher aggregation states compared to the
situation in the sample solution). The detection of these zincate
anions29 contrasts with our previous studies on THF solutions
of neutral zinc species RZnCl (R ¼ Bu, Bn, Ph) which afforded
only trace amounts of alkyl or arylzincate anions upon
which could potentially alter or inuence the composition of
the organometallic species present in solution (see ESI† for
1
details). H and 13C NMR spectra of tBuMgCl (0.86 M in THF)
show two sets of tert-butyl resonances (see Table 2) consistent
with the presence of two different aggregates in solution.23 On
adding 1 molar equivalent of ZnCl2, both spectra simplify
substantially, showing a single tBu-containing species (Table 2,
Entry 3). The 13C NMR spectrum shows an informative reso-
nance at 21.8 ppm assignable to the quaternary C of the tBu
group, which is further downeld than those observed in the
related Grignard precursor (15.4 and 14.8 ppm) but comparable
with that found in neutral tBu2Zn (25.1 ppm, Table 2, Entry 2).
These spectra are consistent with a complete transmetallation
reaction, forming hybrid [(THF)4MgCl2Zn(tBu)Cl] (1) which in
solution retains its bimetallic constitution.24 Interestingly,
when a 3 : 1 tBuMgCl–ZnCl2 mixture is studied using NoD NMR
experiments, the 1H and 13C spectra are almost identical to
those previously discussed in deuterated THF solutions, with
tris(tert-butyl) zincate 2 co-existing in solution with its two
single metal components (Table 2, Entry 4). Following a similar
trend to that observed for 1, the {ZntBu3}ꢀ anion shows a singlet
at 0.82 ppm in the 1H NMR spectrum whereas the quaternary C
of the tBu groups resonates at 23.3 ppm.
The presence of these three different organometallic species
in equilibrium poses the question whether the reaction of this
mixture with 3-iodoanisole (3a) is a genuine Zn–I exchange
process, or alternatively, is the result of a Mg–I exchange reac-
tion, to form an arylmagnesium intermediate which in turn can
undergo transmetallation with tBu2Zn or unreacted ZnCl2.
However, considering the comparative reactivity studies shown
above (Fig. 1 and Table 1, Entry 1), this last scenario appears to
be very unlikely, as the in situ generated zincate complex 2 is
signicantly more reactive in the halogen-exchange process
than tBuMgCl. This higher reactivity should affect the position
1
Table 2 Chemical shifts (ppm) of the H and 13C{1H} NMR spectra of
tBuMgCl, tBu2Zn and mixtures n eq. tBuMgCl + ZnCl2 (n ¼ 1 and 3) in
THF solutions
1
Entry RMa
d H (Me)b
d
13C (M–C)c
d
13C (Me)c
1
2
3
4
tBuMgCl
tBu2Zn
0.86, 0.85
0.97
0.95
15.4, 14.8
25.1
21.8 (br)
35.5, 34.8
32.5
tBuMgCl + ZnCl2
33.5
3 tBuMgCl + ZnCl2 0.88, 0.82d, 24.8, 23.3d,
35.9d, 35.2,
34.4, 32.5
0.80 (br)
15.0, 14.5
a
The in situ mixtures were prepared by mixing freshly titrated tBuMgCl
b
with ZnCl2 in THF in the indicated ratio. d/ppm, 500 MHz, 27 ꢁC,
c
downeld THF resonance (3.62) was used as a reference. d/ppm, 125
MHz, 27 ꢁC, downeld THF resonance (68.03) was used as a reference.
d
Resonances belonging to Mg–Zn hybrid 2, which is the major
Fig. 2 Negative-ion mode (top) and positive-ion mode (bottom) ESI
mass-spectra of a 25 mmolar solution of ZnCl2 and tBuMgCl in THF.
species present in solution.
This journal is © The Royal Society of Chemistry 2014
Chem. Sci., 2014, 5, 3552–3562 | 3555

View Article Online
Chemical Science
Edge Article
negative-ion mode ESI,14e indicating that spontaneous dispro- isolated Mg–Zn hybrid 2. Interestingly, although the tris(alkyl)-
portionation of RZnCl in neat THF is not favourable. These zincate anion is not the major species present in solution, the
results provide further support for the presence of mixed-metal equilibrium can be shied towards its almost exclusive forma-
species (Mg–Zn hybrids) in the seemingly simple metathetical tion by introducing an additional equivalent of tBuMgCl
reactions between RMgCl and ZnCl2. Furthermore, highlighting making four in total (see Fig. S16, ESI†).
the complexity of these reactions, a comparison with the spectra
obtained for transmetalation mixtures of lithium alkyl tBuLi
and ZnCl2 in a 1 : 1 ratio, revealed that although in both cases
zincate anions are detected, the composition of the latter is
Applying Mg–Zn hybrids to direct Zn–I exchange and Pd-
catalysed cross-coupling reactions: structural authentication
of key reaction intermediates
more complicated, containing several mononuclear and dinu-
clear species including {ZnCl3}ꢀ and {tBu2LiZn2Cl4}ꢀ which are
Encouraged by the reactivity studies with 3a (vide infra) which
absent in the tBuMgCl–ZnCl2 mixture.26a These signicant
differences in the compositions of these salt-metathesis
mixtures (which at a supercial understanding would be
expected to generate tBuZnCl along with either MgCl2 and LiCl),
illustrate the important role that inorganic salts can play in
controlling the overall constitution of the newly formed
organometallic reagent. Moreover, they also contribute to
understanding the different reactivities noted in the literature
for RZnX reagents depending on the organometallic source
(RMgX or RLi) employed in their preparation.9,30
show that the metal–halogen exchange process occurs almost
quantitatively when treated with a tBuMgCl–0.33 ZnCl2 mixture
(Fig. 1D), the isolation of the organometallic intermediate
previous to the hydrolysis step was attempted, upon cooling the
reaction mixture. While this led to the successful isolation of a
microcrystalline solid unfortunately it could not be employed
for an X-ray crystallographic study. Notwithstanding, the same
approach, changing the organic halide to the related 2-iodoa-
nisole (3b), led to the isolation of novel tris(aryl) Mg–Zn hybrid
[{Mg(THF)6}2+{Zn(o-C6H4-OMe)3}22ꢀ] (4) in an isolated 28%
yield whose structure was determined by X-ray crystallography
(Fig. 4).
Turning to the positive ion mode spectra, several THF-
+
solvated magnesium ions ([MgCl(THF)3]+, [Mg2Cl3(THF)4–5
]
and [Mg3Cl5(THF)5–6]+) were detected in solution (Fig. 2,
bottom). Such Mg-cations have been reported before in the solid
state, as well as in solution.6d,31 The series of cations can be
rationalized by stepwise co-complexation of neutral MgCl2 and
the [MgCl]+ cation and it is reasonable to assume that they
might exist in a dynamic equilibrium in solution. Interestingly,
no zinc cations, or cations containing tBu-groups, have been
detected though zinc cationic species are known.32
tBuMgCl–ZnCl2 in a 3 : 1 ratio. The negative-ion mode ESI
mass spectrum of an in situ prepared THF-solution of 3 equiv-
alents of tBuMgCl and ZnCl2 revealed the presence of three new
zincate species {tBu2ZnCl}ꢀ, {tBu3Zn}ꢀ and {tBu4Zn2Cl}ꢀ (m/z ¼
213, 235 and 392 respectively, Fig. 3). All these zincate anions
are consistent with a higher degree of alkylation and it should
be noted that none of them occur in the 1 : 1 mixtures. The
spectra recorded for a crystalline sample of isolated 2 proved to
be identical to that shown in Fig. 3 (see Fig. S15, ESI†), which is
consistent with the same reactivity observed in the Zn–I
exchange reactions with 3a for the 3 : 1 in situ mixture and the
Reinforcing the results gleaned from the reactivity and
spectroscopic studies, the structural elucidation of 4 provides
further conrmation of the occurrence of a Zn–I exchange
process, where each tert-butyl group of Mg–Zn hybrid 2 reacts
with one equivalent of the aromatic substrate, to generate a new
tris(aryl) zincate intermediate. Revealing a new structural
modication, 4 comprises an octahedral magnesium dication
solvated by six molecules of THF, balanced by two tris(aryl)
zincate anions. Interestingly, MgCl2 is not included in the
constitution of this bimetallic intermediate, despite being
present in solution as well as being a component of 2. A plau-
sible explanation could be the presence of several cationic
species in equilibrium in THF solution (as hinted by the positive
ion-mode ESI-MS spectra, see for example, Fig. 2, bottom), with
the homoleptic dication {Mg(THF)6}2+ crystallising preferen-
tially in the case of 4.33 Although the isolated yield of 4 is modest
(28%), 1H NMR monitoring of the reaction of isolated crystals of
2 with three equivalents of 3b in deuterated THF at room
temperature showed the almost quantitative formation of 4
(>90% conversion) along with tBuI within 20 minutes.34
Fig. 4 Molecular structure of [{Mg(THF)6}2+{Zn(o-C6H4OMe)3}2ꢀ] (4)
Fig. 3 Negative-ion mode ESI mass-spectrum of a 25 mmolar solu- with 30% probability ellipsoids. Hydrogen atoms and minor THF
tion of ZnCl2 and 3 tBuMgCl in THF.
disorder components have been omitted for clarity.
3556 | Chem. Sci., 2014, 5, 3552–3562
This journal is © The Royal Society of Chemistry 2014

View Article Online
Edge Article
Chemical Science
The high atom economy of this transformation (employing 60–65% yield aer successful cross-coupling with iodobenzene
all three tBu arms of 2) contrasts with previous studies using (4a, Entries 4–6). In the case of iodotoluene derivatives, the
related lithium zincates [LiZntBu3] and [Li2ZntBu4], where isolated yields were lower due to the formation of biphenyl as
equimolar amounts of the organic halide and the mixed-metal side product arising from a homo-coupling process. Generally,
reagent are required.35 Under substoichiometric conditions during the exchange reactions, the use of electron-rich
these mixed Li–Zn species are sluggish to react, which is substrates (iodotoluene, iodoanisole) seemed to favor this
attributed to their degradation, due to the competing reaction competing process; whereas no homo-coupling products were
of the tBu groups on the zincate with the generated tBuX to form detected when aromatic iodides bearing electron-withdrawing
isobutene.36 Also Knochel has shown that ZnAr2 intermediates substituents were used.
can be prepared in situ by adding substoichiometric amounts of
Nitriles are important synthetic building blocks and the
lithium acetylacetonate, Li(acac), to ZniPr2 in the presence of iodine–zinc exchange is perfectly compatible with the cyano
several functionalised aromatic iodides (Ar-I); however the use group of iodobenzonitriles (3g,h). Illustrating the functional
of a highly polar combination of solvents (Et2O–NMP in a group tolerance of this approach, despite its enhanced reac-
1 : 10 ratio; NMP ¼ N-methylpyrrolidone) is required to favour tivity, tris(alkyl)zincate 2 did not attack the nitrogen containing
the reactions.37
unsaturated functional group under the employed reaction
Since the Mg–Zn hybrid approach grants access to homo- conditions and the cross-coupling with iodobenzene (5a)
leptic tris(aryl) zincates in an efficient manner, under mild provided the coupling products (6g,h) in good yields (71–75%,
reaction conditions (room temperature, short reaction times, in Entries 7–8).42,43 Interestingly, when this method was applied to
the absence of additives), we then tested their applications as 2-iodobenzonitrile (3i), the relevant coupling product 6i could
precursors in Pd-catalysed Negishi cross-coupling reactions, only be obtained in a modest 31% yield. 1H NMR monitoring of
one of the most powerful methods for C(sp2)–C(sp2) bond the reaction of isolated crystals of Mg–Zn hybrid 2 with three
formation in synthesis.38 Informative mechanistic studies on equivalents of 3i revealed that only two of three alkyl groups on
this fundamental process have revealed that the trans- the zincate are active towards the Zn–I exchange with this
metallation of the substituent in the organozinc precursor to substrate (see Fig. S31 and S32, in ESI†). This reduced reactivity
the Pd catalyst may be the rate-determining step of the catalytic of 2 towards 3i is rather surprising as in general aromatic
cycle.39 Interestingly, Organ has recently shown that mixed- substrates containing strongly electron-withdrawing groups are
halide–alkyl zincates are much more efficient transmetallating more activated towards metal–halogen exchange processes,
reagents in these processes than neutral RZnX reagents, especially if the substitution of the halide is at the ortho position
although the solvent choice also plays a major role in these of these groups.44 Furthermore, aer 2 hours at room temper-
processes and without polar NMP or DMI the couplings do not ature, the remaining tert-butyl group attached to zinc starts to
take place.40
be consumed by reaction with the generated tBuI furnishing the
To explore the scope of our Mg–Zn hybrid-mediated bimetallic intermediate [(THF)4MgCl(NC-o-C6H4)ZnI(o-C6H4-
approach, several aryl iodides (3) were treated with a 3 : 1 CN)(THF)] (7), along with concomitant release of isobutene
mixture of tBuMgCl–ZnCl2 in THF for 2 h, before adding iodo- (Scheme S1 in ESI†).
benzene (5a) as an organic electrophile, in the presence of
X-ray crystallographic studies established the contacted ion
2.5 mol% of PdCl2(dppf) (dppf ¼ 1,10-bis(diphenylphosphino)- pair (CIP) structure 7 (Fig. 5) where both metals are connected
ferrocene).41 The preliminary cross-coupling results proved very by an ortho-metallated benzonitrile fragment which coordinates
promising as several biphenyls could be isolated in good yields unsymmetrically to the bimetallic system, bonding through its
(60–79%) under non-optimized reaction conditions (Scheme 2 ortho-C (C12 in Fig. 5) to Zn and through its N atom (N2 in
and Table 3, Entries 1–8).
Fig. 5) to Mg. Magnesium completes its distorted octahedral
In general, the substitution pattern of the aryl iodide did not coordination by binding to a Cl atom and to four THF molecules
show any signicant inuence on the efficiency of the iodine– whereas distorted tetrahedral Zn binds to a terminal ortho-
zinc exchange or the cross-coupling conditions. Thus, p-, ¼o- metallated benzonitrile fragment, an iodine atom and a
and m-iodoanisole (3a–c) furnished the corresponding phenyl- solvating THF molecule. The CIP structure of this hybrid
substituted anisole derivatives 6a–c smoothly in a similar yield contrasts sharply with the SSIP structures previously
(76–79%; Table 3, Entries 1–3). Similarly, o-, m- and p-iodoto- determined for precursor 2 and also the Zn–I exchange inter-
luene derivatives (3d–f) afforded asymmetric biphenyls 6d–f in mediate 4.
This structural feature could shed some light on the lack of
activation observed for the third tert-butyl group present in 2
when reacted with 3i. A recent theoretical study into Zn–I
exchange reactions of lithium tris(alkyl)zincates (LiZnR3)
revealed that in the most energetically favoured pathway the
rst step of the reaction is coordination of the substrate
through its iodine atom to lithium, activating the C–I bond of
the arene which facilitates the Zn–I exchange reaction.45 If a
similar mechanism could be taking place for Mg–Zn hybrid 2, 3i
would initially coordinate to the cationic Mg centre, allowing
Scheme 2 Synthesis of asymmetric bis(aryls) 6a–h by sequential
direct Zn–I exchange and Pd-catalysed Negishi cross-coupling
reaction with PhI (5a).
This journal is © The Royal Society of Chemistry 2014
Chem. Sci., 2014, 5, 3552–3562 | 3557

View Article Online
Chemical Science
Edge Article
Table 3 Iodine–zinc exchange reaction using various aryl iodides (3) Table 3 (Contd.)
and subsequent cross-coupling reactions with electrophiles (5) using
Substrate
Electrophilea
Product/yieldb
Pd-catalysis, leading to bis(aryls) of type 6
Substrate
Electrophilea
Product/yieldb
14
15
16
1
2
3
4
5
6
PhI (5a)
PhI (5a)
PhI (5a)
PhI (5a)
PhI (5a)
PhI (5a)
17
18
19
7
PhI (5a)
a
0.9–1.0 equivalent of electrophile was used. b Isolated yield of product
1
estimated to be of >95% purity according to GC and H-NMR analysis.
c
d
2.5 mol% of PdCl2(dppf) was used. 0.9 mol% of PEPPSI-iPr was
used. e No reaction was observed even aer heating to 50 ꢁC for 24 h.
8
9
PhI (5a)
PhI (5a)
10
11
12
Fig.
5 Molecular structure of [(THF)4MgCl{NC-o-C6H4}ZnI(o-
C6H4CN)(THF)] (7) with 50% probability ellipsoids. Hydrogen atoms
have been omitted for clarity.
anionic {ZntBu3}ꢀ to react with the activated C–I bond.
Assuming that the Zn–I exchange reactions of the three tert-
butyl groups in 2 occur in a sequential manner, aer the
successful exchange of two of the alkyl groups, the formation of
a putative CIP intermediate [(THF)4MgCl(NC-o-C6H4)Zn(tBu)-
(o-C6H4–CN)(THF)]46 may hinder the coordination of the third
13
3558 | Chem. Sci., 2014, 5, 3552–3562
This journal is © The Royal Society of Chemistry 2014

View Article Online
Edge Article
Chemical Science
molecule of substrate to magnesium. Furthermore, even if the with variable amounts of ZnCl2. In all cases mixed-metal
remaining equivalent of 3i was able to bind to Mg, the less magnesium zincates (Mg–Zn hybrids) are formed. Focussing on
exible CIP structure of this intermediate may prevent the metal–halogen exchange reactions, a fundamental tool for the
approach of the remaining tert-butyl ligand close enough to the functionalisation of aromatic substrates, we report that SSIP
C–I bond for the exchange reaction to occur.
hybrid [{Mg2Cl3(THF)6}+{ZntBu3}ꢀ] (2) can effectively promote
A similar rationale could also explain the complete lack of direct Zn–I exchange reaction of aromatic halides in short
reactivity observed for CIP zincate [(THF)4MgCl2Zn(tBu)Cl] (1) periods of time at room temperature in THF without the need of
when reacted with 3a (vide supra) as well as the previously polar solvent additives. By switching on cooperative effects, the
reported use of highly polar cosolvents,37 since the latter should three tBu groups attached to Zn are active towards the exchange,
favour the formation of more reactive SSIP species.
accessing synthetically valuable tris(aryl)zincate anions. Con-
Regarding the scope of this approach to prepare asymmetric trastingly, under similar reaction conditions described neither
bis(aryls), it was found that aryl bromides could also be used as homometallic component tBuMgCl or ZntBu2 undergo MꢀI
electrophiles. A cross-coupling reaction occurred smoothly exchange. In situ studies, using a range of RMgCl reagents,
between 3a and 4-bromobenzonitrile (5b) utilizing three reveal that in all cases the transmetallation step is faster than
different commonly used Pd-catalysts (PEPPSI-iPr,47 Pd(OAc)/S- the magnesium–iodine exchange process. Contrastingly, using
Phos,48 PdCl2dppf41). No signicant reactivity differences were a CIP halide-rich alkyl Mg–Zn hybrid [(THF)4MgZn(tBu)Cl3](1),
observed between the employed catalyst systems and the biaryl the metal–halogen exchange process is totally suppressed.
6j was isolated in 57–81% yield (Entry 10). Upon treatment with
No Deuterium-NMR and ESI-MS spectroscopic studies
hybrid 2 the ester functionalized aryl iodide 3j could be readily probing the organometallic intermediates in these trans-
converted to a zinc reagent and subsequent Negishi cross- metallation mixtures show that in THF solutions, mixed Mg–Zn
coupling afforded 6j in 76% yield (Entry 11).49 Ethyl 3-iodo- compounds are the dominant species present, although in the
benzoate (3k) reacted in a similar way and cross-coupling with case of tBuMgCl/0.33 equivalents of ZnCl2, it appears that
3-bromo-benzotriuoride (5c) furnished the functionalized hybrid 2 is in equilibrium with its homometallic counterparts.
biaryl in 86% yield (Entry 12). Contrastingly ethyl 2-iodo- Notwithstanding, considering that neither tBuMgCl or ZntBu2
benzoate was compatible with the halogen–zinc exchange, but can promote M ꢀ I exchange reactions, it can be expected that
reacted only sluggishly in cross-coupling reactions with PEPPSI- when these mixtures are confronted with the aromatic
iPr (1.5 mol%) as catalyst.50 Interestingly, the 1,2-dihalo- substrate, this equilibrium will lie toward the formation of 2.
substituted benzene derivative (3l) proved to be stable aer Structural evidence for the success of this reaction was gained
successful halogen–metal exchange and did not undergo elim- by the isolation of the magnesium dizincate intermediate
ination forming an aryne under the present reaction conditions. [{Mg(THF)6}2+{Zn(o-C6H4-OMe)3}2ꢀ] (4), which in agreement
Whereas ortho- and para-chloro-iodobenzene (3l,m) could be with the reactivity studies demonstrates the high atom economy
readily transformed into zinc reagents and used for further of the Zn–I exchange process. Interestingly, using 2-iodo-
functionalization (Entries 13 and 14), the zinc reagent derived benzonitrile (3i), only two tBu groups react with the substrate,
from m-chloro-iodobenzene (3n) was reluctant to undergo affording [(THF)4MgCl(NC-o-C6H4)ZnI(o-C6H4-CN)(THF)] (7),
cross-coupling reactions, even aer prolonged heating to 50 ꢁC where Mg and Zn are connected by a bridging aryl ligand. These
(Entry 15).51 Fluorine-containing aromatic compounds play an ndings suggest that the formation of contacted ion-pair
important role in pharmaceutical science.52 Thus, the reactions intermediates may have a detrimental effect on the efficiency of
of different uorine-containing aromatic iodides (3o–q) were the Zn–I exchange process.
tested under these reaction conditions. Pleasingly these
Expanding the synthetic utility of these heterobimetallic
substrates could be readily converted to the zincate reagent and systems, we have applied this approach to a wide range of
used efficiently in cross-coupling reactions affording products substituted aromatic substrates, generating homoleptic tri-
6p–r in excellent yields (75–80%, Entries 16–18). Furthermore, s(aryl) zincate intermediates which in turn can react efficiently
the heterocyclic substrate 2-iodothiophene (3r) reacted in an in Negishi cross-coupling reactions under mild reaction
exchange reaction furnishing a heterocyclic zincate reagent. Its conditions without the need of a polar co-solvent.
reaction with 3-chloro-iodobenzene (5f) furnished the func-
tionalized thiophene 6s in 61% yield aer column chromato-
graphical purication (Entry 19).53
Experimental section
Typical procedure for the iodine–zinc exchange using in situ
generated 2 and subsequent cross-coupling reaction
Conclusions
A Schlenk ask was charged with freshly titrated tBuMgCl
By combining transmetallation reactions with metal–halogen solution (0.96 M in THF, 3.12 mL, 2.99 mmol) and a solution of
exchanges processes we report an efficient and versatile method ZnCl2 in THF (1.0 M, 1.0 mmol, 1.0 mL) was added via a syringe
which grants access to magnesium tris(aryl) zincate reagents. at 0 ꢁC. The resulting clear solution was stirred for 5 min at 0 ꢁC
Firstly, combining X-ray crystallography with NMR spectro- before 4-iodoanisole (3a, 702 mg, 2.99 mmol) was added. The
scopic and ESI MS studies, we have shed some new light on the mixture was stirred for 30 min at 0 ꢁC and subsequently
complex constitution of the organometallic intermediates 4-bromobenzonitrile (5b, 546 mg, 2.99 mmol) and PdCl2dppf
involved in the transmetallation reactions of Grignard reagents (54 mg, 0.075 mmol) were added. Aer stirring for 2 h at 20 ꢁC
This journal is © The Royal Society of Chemistry 2014
Chem. Sci., 2014, 5, 3552–3562 | 3559

View Article Online
Chemical Science
Edge Article
the reaction was quenched by addition of aq. HCl (2 M, 5 mL),
the phases separated and the aqueous phase was extracted
three times with ethyl acetate. The combined organic extracts
were dried with MgSO4 and the solvent was removed in vacuo.
The crude product was subjected to column-chromatographic
purication (SiO2, hexane–EtOAc ¼ 10 : 1) furnishing 6j as an
off-white solid (505 mg, 81%). For analytical data see ESI.†
2001, 3, 1793; (e) G. N. Khairallah, C. C. L. Thum,
D. Lesage, J.-C. Tabet and R. A. J. O'Hair, Organometallics,
2013, 32, 2319; (f) A. Harrison-Marchand and F. Mongin,
Chem. Rev., 2013, 113, 7470.
7 For selected examples see: (a) L. Gupta, A. C. Hoepker,
K. J. Singh and D. B. Collum, J. Org. Chem., 2009, 74, 2231;
(b) Y. Ma, A. C. Hoepker, L. Gupta, M. F. Faggin and
D. B. Collum, J. Am. Chem. Soc., 2010, 132, 15610; (c)
L. Salvi, J. G. Kim and P. J. Walsh, J. Am. Chem. Soc., 2009,
Acknowledgements
¨
131, 12483; (d) M. Granitzka, A.-C. Poppler, E. K. Schwarze,
We thank the European Research Council (E.H.), the Royal
Society (E.H.), EPSRC (W.C.) and the Deutsche For-
schungsgemeinscha (K.K.) for sponsoring this research. K.K.
also thanks Prof. Dietmar Stalke for the access to his glove box.
We also thank Prof. Robert E. Mulvey for his insightful
comments.
D. Stern, T. Schulz, M. John, R. Herbst-Irmer, S. K. Pandey
and D. Stalke, J. Am. Chem. Soc., 2012, 134, 1344; (e)
´
J. A. Garden, D. R. Armstrong, W. Clegg, J. Garcıa-Alvarez,
E. Hevia, A. R. Kennedy, R. E. Mulvey, S. D. Robertson and
´
L. Russo, Organometallics, 2013, 32, 5481; (f) T. Tatic,
S. Hermann and D. Stalke, Organometallics, 2012, 31, 5615;
(g) H. Zong, H. Huang, J. Liu, G. Bian and L. Song, J. Org.
Chem., 2012, 77, 4645.
8 G. Manolikakes and P. Knochel, Angew. Chem., Int. Ed., 2009,
48, 205.
9 L. Jin, C. Liu, J. Liu, F. Hu, Y. Lan, A. S. Batsanov,
J. A. K. Howard, T. B. Marder and A. Lei, J. Am. Chem. Soc.,
2009, 131, 16656.
Notes and references
1 M. Schlosser, Organoalkali Chemistry, in Organometallics in
Synthesis: Third Manual, ed. M. Schlosser, John Wiley &
Sons, Inc., Hoboken, NJ, USA, 2013.
2 (a) The Chemistry of Organomagnesium Compounds, ed. Z.
Rappoport and I. Marek, Patai Series, Wiley, Chichester, 10 R. B. Bedford and M. C. Wilkinson, Chem. Commun., 2009,
UK, 2008; (b) For an excellent historical essay see:
D. Seyferth, Organometallics, 2009, 28, 1598.
3 (a) The chemistry of organozinc compounds, ed. Z. Rappoport
600.
11 A. Metzger, S. Bernhardt, G. Manolikakes and P. Knochel,
Angew. Chem., Int. Ed., 2010, 49, 4665.
and I. Marek, Patai Series, Wiley, Chichester, UK, 2006; (b) 12 (a) P. Knochel and P. Jones, in Organozinc Reagents: A
´
J. Lewinski and A. E. H. Wheatley, Top. Organomet. Chem.,
Practical Approach, ed. L. H. Harwood and C. J. Moody,
Oxford University Press, Oxford, 1999; (b) T. Klatt, K. Groll
and P. Knochel, Chem. Commun., 2013, 49, 6953; (c) H. Gao
and P. Knochel, Synlett, 2009, 8, 1321; (d) P. Knochel and
R. D. Singer, Chem. Rev., 1993, 93, 2117; (e) A. Inoue,
K. Kitagawa, H. Shinokubo and K. Oshima, J. Org. Chem.,
2001, 66, 4333.
2013, 41, 1.
4 For selected reference see: (a) E. Carl and D. Stalke in Lithium
Compounds in Organic Synthesis: From Fundamentals to
Applications, ed. R. Luisi and V. Capriati, Wiley-VCH Verlag
& Co, Weinheim, Germany, 2014; (b) H. J. Reich, Chem.
Rev., 2013, 113, 7130; (c) H. J. Reich, J. Org. Chem., 2012,
´
77, 5471; (d) F. Pate, N. Duguet, H. Oulyadi, A. Harrison- 13 E. Hevia and R. E. Mulvey, Angew. Chem., Int. Ed., 2011, 50,
´
Marchand, C. Fressigne, J.-Y. Valnot, M.-C. Lasne and
6448 and references therein.
J. Maddaluno, J. Org. Chem., 2007, 72, 6982; (e) D. Li, 14 (a) G. T. Achonduh, N. Hadai, C. Valente, S. Avola,
I. Keresztes, R. Hopson and P. G. Williard, Acc. Chem. Res.,
2009, 42, 270; (f) D. B. Collum, Acc. Chem. Res., 1992, 25,
448; (g) J. S. DePue and D. B. Collum, J. Am. Chem. Soc.,
1988, 110, 5518; (h) J. F. Remenar and D. B. Collum, J. Am.
Chem. Soc., 1998, 120, 4081.
C. J. O'Brien and M. G. Organ, Chem. Commun., 2010, 46,
4109; (b) J. F. G. A. Jansen and B. L. Feringa, J. Chem. Soc.,
Chem. Commun., 1989, 741; (c) M. Hatano, S. Suzuki and
K. Ishihara, J. Am. Chem. Soc., 2006, 128, 9998; (d)
K. Murakami, H. Yorimitsu and K. Oshima, J. Org. Chem.,
2009, 74, 1415; (e) J. E. Fleckenstein and K. Koszinowski,
Organometallics, 2011, 30, 5018.
5 (a) F. Totter and P. Rittmeyer, Organolithium Compounds—
Industrial Applications and Handling, in Organometallics in
Synthesis: A Manual, ed M. Schlosser, John Wiley & Sons, 15 For selected references see: (a) D. Stern, N. Finkelmeier and
Inc., Hoboken, NJ, USA, 1994; (b) G. Wu and M. Huang,
Chem. Rev., 2006, 106, 2596; (c) J. Clayden, Organolithiums:
Selectivity for Synthesis, Tetrahedron Organic Chemistry
Series, Elsevier Science Ltd, Oxford, UK, 2002, vol. 23.
6 For selected references see: (a) C. Su, R. Hopson and
P. G. Williard, J. Am. Chem. Soc., 2014, 136, 3246; (b) C. Su,
R. Hopson and P. G. Williard, J. Org. Chem., 2013, 78,
11733; (c) Y. Ma, C. E. Stivala, A. M. Wright, T. Hayton,
J. Liang, I. Keresztes, E. Lobkovsky, D. B. Collum and
D. Stalke, Chem. Commun., 2011, 47, 2113; (b) P. Garcia-
Alvarez, D. V. Graham, E. Hevia, A. R. Kennedy, J. Klett,
R. E. Mulvey, C. T. O'Hara and S. Weatherstone, Angew.
Chem., Int. Ed., 2008, 47, 8079; (c) D. R. Armstrong,
P. Garcia-Alvarez, A. R. Kennedy, R. E. Mulvey and
J. A. Parkinson, Angew. Chem., 2010, 122, 3253; (d)
D. R. Armstrong, W. Clegg, P. Garcia-Alvarez,
A. R. Kennedy, M. D. McCall, L. Russo and E. Hevia,
Chem.–Eur. J., 2011, 17, 8333.
A. Zakarian, J. Am. Chem. Soc., 2013, 135, 16853; (d) 16 E. Hevia, J. Z. Chua, P. Garcia-Alvarez, A. R. Kennedy and
S. Sakamoto, T. Imamoto and K. Yamaguchi, J. Org. Chem.,
M. D. McCall, Proc. Natl. Acad. Sci. U. S. A., 2010, 107, 5294.
3560 | Chem. Sci., 2014, 5, 3552–3562
This journal is © The Royal Society of Chemistry 2014

View Article Online
Edge Article
Chemical Science
17 (a) Commercially available Grignard solutions are most
Energy Environ. Sci., 2012, 5, 9100; (c) N. Pour, Y. Gofer,
D. T. Major and D. Aurbach, J. Am. Chem. Soc., 2011, 133,
6270.
likely prepared by oxidative addition of Mg into the
organic
halide;
(b)
P.
Knochel,
W.
Dohle,
N. Gommermann, F. F. Kneisel, F. Kopp, T. Korn, 32 R. B. Bedford, N. J. Gower, M. F. Haddow, J. N. Harvey,
I. Sapountzis and V. A. Vu, Angew. Chem., Int. Ed., 2003, 42,
4302; (c) C. Christophersen, M. Begtrup, S. Ebdrup,
H. Petersen and P. Vedsø, J. Org. Chem., 2003, 68, 9513.
18 P. C. Andrikopoulos, D. R. Armstrong, H. R. L. Barley,
W. Clegg, S. H. Dale, E. Hevia, G. W. Honeyman,
A. R. Kennedy and R. E. Mulvey, J. Am. Chem. Soc., 2005,
127, 6184.
J. Nunn, R. A. Okopie and R. F. Sankey, Angew. Chem., Int.
Ed., 2012, 51, 5435.
33 The reaction of 2 with three equivalents of 4-iodotoluene
allowed the isolation of [{Mg2Cl3(THF)6}+{Zn(p-Me-
C6H4)3}ꢀ], see ref. 16. Similarly to the reaction with 3b,
tris(aryl) zincate anions are formed, conrming the
successful Zn–I exchange process but in this case, the
19 Exchange rates are roughly estimated to be tBuMgCl: ca.
compound crystallizes preferentially with
a dinuclear
0.02% minꢀ1; hybrid 2: ca. 10% minꢀ1
.
cation {Mg2Cl3(THF)6}+ which can be envisaged as a co-
¨
20 L. Boymond, M. Rottlander, G. Cahiez and P. Knochel,
complex
of
neutral
MgCl2(THF)x
and
cationic
Angew. Chem., Int. Ed., 1998, 37, 1701.
21 Analysis on the integration of the H NMR resonances of a 34 H NMR spectrum showed four aromatic resonances at d
{MgCl(THF)6ꢀx}+.
1
1
0.22 M solution of 2 in deuterated THF showed a ratio of
{tBu3Zn}ꢀ : tBu2Zn : tBuMgCl of 8 : 1 : 1 whereas at the
lower concentration 0.07 M the ratio was 4 : 3 : 3.
22 T. R. Hoye, B. M. Eklov, T. D. Ryba, M. Voloshin and L. J. Yao,
Org. Lett., 2004, 6, 953.
23 The presence of more than one species in solution is
consistent with the complicated solution behaviour that
these reagents can exhibit. For example, see ref. 6d.
7.16 (d), 7.13 (d), 7.02 (t) and 6.59 (dd), along with a
singlet at 3.70 ppm for compound 4, which were
signicantly shied from those of substrate 3b (d 7.27
(dd), 7.25 (dt), 6.99 (t), 6.88 (ddd) and 3.75 (s)).
35 (a) D. Tilly, F. Chevallier, F. Mongin and P. C. Gros, Chem.
Rev., 2014, 114, 1207; (b) T. Furuyama, M. Yonehara,
S. Arimoto, M. Kobayashi, Y. Matsumoto and
M. Uchiyama, Chem.–Eur. J., 2008, 14, 10348.
24 NMR data observed for this in situ mixture are almost 36 Interestingly, when the n-butyl derivative Li2ZnBu4$TMEDA
identical to those reported for isolated crystals of 1 in
deuterated THF, see ref. 16.
25 (a) A. Putau and K. Koszinowski, Organometallics, 2011, 30,
4771; (b) K. Koszinowski, J. Am. Chem. Soc., 2010, 132,
6032; (c) B. H. Lipshutz, J. Keith and D. J. Buzard,
is employed, zincation of bromo-pyridines can be achieved
using 1/3 equivalents of the bases. See: N. T. T. Chau,
M. Meyer, S. Komagawa, F. Chevalier, Y. Fort,
M. Uchiyama, F. Mongin and P. C. Gros, Chem.–Eur. J.,
2010, 16, 12425.
Organometallics, 1999, 18, 1571; (d) A. Putau, M. Wilken 37 F. F. Kneisel, M. Dochnahl and P. Knochel, Angew. Chem.,
and K. Koszinowski, Chem.–Eur. J., 2013, 19, 10992. Int. Ed., 2004, 43, 1017.
26 (a) K. Koszinowski and P. Bohrer, Organometallics, 2009, 28, 38 (a) E. Negishi in Handbook of Organopalladium Chemistry for
¨
´
´
100; (b) A. Hernan-Gomez, E. Herd, E. Hevia, A. R. Kennedy,
P. Knochel, K. Koszinowski, S. M. Manolikakes,
R. E. Mulvey and C. Schnegelsberg, Angew. Chem., Int. Ed.,
2014, 53, 2706.
Organic Synthesis, Wiley-Interscience, New York, 2002, vol. 1,
pp. 229–248; (b) Metal-Catalysed Cross-Coupling Reactions,
ed. A. de Meijere and F. Diederich, Wiley-VCH, Weinheim,
2nd edn, 2004.
27 ESI
information to NMR spectroscopic studies. The ESI mass 40 L. C. McCann, H. N. Hunter, J. A. Clyburne and M. G. Organ,
spectra obtained for the 1 : 1 and 3 : 1 tBuMgCl–ZnCl2 Angew. Chem., Int. Ed., 2012, 51, 7024.
mixtures showed a higher diversity of zincate anions than 41 (a) G. Cahiez, F. Lepifre and P. Ramiandrasoa, Synthesis,
mass
spectrometry
provided
complementary 39 J. Li, L. Jin, C. Liu and A. Lei, Chem. Commun., 2013, 49, 9615.
the corresponding NMR spectra. This might be due to
equilibration/aggregation processes which occur too fast to
be resolved on the NMR time scale.
1999, 2138; (b) E. Riguet, M. Alami and G. Cahiez,
J. Organomet. Chem., 2001, 624, 376; (c) V. L. Blair,
W. Clegg, B. Conway, E. Hevia, A. R. Kennedy, J. Klett,
R. E. Mulvey and L. Russo, Chem.–Eur. J., 2008, 14, 65.
28 The m/z ratio is given for the most abundant isotopologue.
The suggested sum formulae are based on comparison of 42 For the possible addition of highly reactive organozinc
calculated and measured isotope patterns, as well as
collision-induced decomposition (CID) experiments, see
ESI.†
reagents to cyano groups see: P. Knochel and
J. F. Normant, J. Organomet. Chem., 1986, 309, 1.
43 For related examples of mixed Mg–Zn reagents acting as
nucleophiles towards ketones see: (a) D. R. Armstrong,
W. Clegg, P. Garcia-Alvarez, M. D. McCall, L. Nuttall,
A. R. Kennedy, L. Russo and E. Hevia, Chem.–Eur. J.,
2011, 17, 4470; (b) M. Hatano, S. Suzuki and
K. Ishihara, J. Am. Chem. Soc., 2006, 128, 9998; (c)
M. Hatano and K. Ishihara, Synthesis, 2008, 1647; (d)
M. Hatano, O. Ito, S. Suzuki and K. Ishihara, J. Org.
Chem., 2010, 75, 5008.
29 Traces of another species (m/z ¼ 445), could be detected and
attributed to a magnesium–zinc hybrid [tBu2Zn2MgCl5]ꢀ,
which can be rationalized by formal complexation of
MgCl2 to the dinuclear chloro-zincate.
30 C. Stude and B. Breit, Angew. Chem., Int. Ed., 2008, 47, 5451.
31 (a) Y. Vestfried, O. Chusid, Y. Goffer, Y. Aped and
D. Aurbach, Organometallics, 2007, 26, 3130; (b) Y.-S. Guo,
F. Zhang, J. Yang, F.-F. Wang, Y. NuLi and S.-I. Hirano,
This journal is © The Royal Society of Chemistry 2014
Chem. Sci., 2014, 5, 3552–3562 | 3561

View Article Online
Chemical Science
Edge Article
44 P. Knochel, W. Dohle, N. Gommermann, F. F. Kneisel,
F. Kopp, T. Korn, I. Sapountzis and V. A. Vu, Angew. Chem.,
Int. Ed., 2003, 42, 4302.
45 S. Nakamura, C. Y. Liu, A. Muranaka and M. Uchiyama,
Chem.–Eur. J., 2009, 15, 5686.
biphenyls see: (a) M. G. Organ, S. Çalimsiz, S. Mahmoud,
K. H. Hoi and A. J. Lough, Angew. Chem., Int. Ed., 2010, 48,
2383. For structure–reactivity relationships in cross-
coupling reactions see: (b) Z. B. Dong, G. Manolikakes,
L. Shi, P. Knochel and H. Mayr, Chem.–Eur. J., 2010, 16, 248.
46 As shown in Scheme S1 in the ESI,† the formation of this 51 3-Chloroiodobenzene (3n) reacts well as electrophile in
intermediate can be detected spectroscopically but
attempts to crystallise this species led to the isolation of 7,
resulting from the reaction of [(THF)4MgCl(NC-o-C6H4)
cross-coupling reactions showing that oxidative insertion
and reductive elimination are usually occurring without
any problem. This suggests some sort of deactivation
which presumably hinders the transmetalation reaction to
the active Pd-complex.
Zn(tBu)(o-C6H4-CN)(THF)]
produced tBuI.
with
the
concomitantly
47 C. J. O'Brien, E. A. B. Kantchev, C. Valente, N. Hadei, 52 For the importance of uorinated aromatics in
G. A. Chass, A. Lough, A. C. Hopkinson and M. G. Organ,
Chem.–Eur. J., 2006, 12, 4743; M. G. Organ, S. Calimsiz,
M. Sayah, K. H. Hoi and A. J. Lough, Angew. Chem., 2009,
121, 2419.
pharmaceutical sciences see: (a) A. Frischmuth, A. Unsinn,
¨
K. Groll, H. Stadtmuller and P. Knochel, Chem.–Eur. J.,
2012, 18, 10234–10238; (b) C. Wang, Y.-B. Yu, S. Fan and
X. Zhang, Org. Lett., 2013, 15, 5004 and references therein.
48 (a) E. R. Strieter, D. G. Blackmond and S. L. Buchwald, J. Am. 53 Attempted exchange reactions using pi-decient N-
Chem. Soc., 2003, 125, 13978; (b) S. D. Walker, T. E. Barder,
J. R. Martinelli and S. L Buchwald, Angew. Chem., Int. Ed.,
2004, 43, 1871.
heterocycles such as 2-iodopyridine, or 2-bromopyridine
were unsuccessful. We assume heteroatom coordination is
disturbing the exchange process. Similarly, aryl bromides
are not reacting with 2, however, treatment of 3-
bromobenzotriuoride with a mixture of iPrMgCl–ZnCl2
(3 : 1) led to bromine–zinc exchange (86% conversion, 3 h).
49 E. Negishi, Acc. Chem. Res., 1982, 15, 340.
50 Yields for the cross-coupling reaction remained usually
below 30%. For the importance of ortho substituted
3562 | Chem. Sci., 2014, 5, 3552–3562
This journal is © The Royal Society of Chemistry 2014