The Manganese-Catalyzed Cross-Coupling Reaction and the Influence of Trace Metals

Article abstract of DOI:10.1002/ejoc.201701005

The substrate scope of the MnCl₂-catalyzed cross-coupling between aryl halides and Grignard reagents has been extended to several methyl-substituted aryl iodides by performing the reaction at elevated temperature in a microwave oven. A radical clock experiment revealed the presence of an aryl radical as an intermediate leading to the proposal of an S_RN1 pathway for the coupling. The mechanistic information gave rise to suspicion about two previously published cross-coupling reactions catalyzed by manganese(II) salts. As a result, the coupling between aryl halides and organostannanes as well as between aryl halides and amines were revisited. Both reactions were found impossible to reproduce without the addition of small amounts of palladium or copper and are therefore not believed to be catalyzed by manganese.

Full text of DOI:10.1002/ejoc.201701005

A Journal of
Accepted Article
Title: The Manganese-Catalyzed Cross-Coupling Reaction and the
Influence of Trace Metals
Authors: Carola Santilli, Somayyeh Sarvi Beigbaghlou, Andreas
Ahlburg, Giuseppe Antonacci, Peter Fristrup, Per-Ola Norrby,
and Robert Madsen
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Org. Chem. 10.1002/ejoc.201701005
Link to VoR: http://dx.doi.org/10.1002/ejoc.201701005
Supported by

10.1002/ejoc.201701005
European Journal of Organic Chemistry
FULL PAPER
The Manganese-Catalyzed Cross-Coupling Reaction and the
Influence of Trace Metals
Carola Santilli,[a] Somayyeh Sarvi Beigbaghlou,[a,b] Andreas Ahlburg,[a] Giuseppe Antonacci,[a] Peter
Fristrup,[a] Per-Ola Norrby,[c,d] and Robert Madsen*[a]
Abstract: The substrate scope of the MnCl2-catalyzed cross-
coupling between aryl halides and Grignard reagents has been
extended to several methyl-substituted aryl iodides by performing
the reaction at elevated temperature in a microwave oven. A radical
clock experiment revealed the presence of an aryl radical as an
intermediate leading to the proposal of an SRN1 pathway for the
coupling. The mechanistic information gave rise to suspicion about
two previously published cross-coupling reactions catalyzed by
manganese(II) salts. As a result, the coupling between aryl halides
and organostannanes as well as between aryl halides and amines
were revisited. Both reactions were found impossible to reproduce
without the addition of small amounts of palladium or copper and are
therefore not believed to be catalyzed by manganese.
coupling between Grignard reagents and vinyl/aryl halides.[3]
The reactions are carried out in THF solution with 3 – 10% of
MnCl2 at a temperature between 0 °C and rt.[3] Very recently, we
investigated the substrate scope in detail for the coupling with
aryl halides and showed that the reaction was limited to aryl
chlorides with cyano or ester groups in the para or ortho
position.[3a] The Grignard reagent, on the other hand, could be
either an aryl- or an alkylmagnesium halide.[3a] The mechanism
was also investigated by a radical clock experiment where an
aryl radical was identified as an intermediate leading to the
proposal of an overall SRN1 pathway for the coupling.[3a] Besides
the reaction with Grignard reagents, aryl halides have also
undergone substitution with other groups in the presence of
manganese catalysts. Aryl and vinyl iodides have been coupled
with aryl, vinyl and alkynyl tributylstannanes in the presence of
MnBr2.[4] Furthermore, aryl halides have been coupled with aryl
boronic acids and alkyl acrylates in the presence of manganese
deposited on heterogeneous supports although the actual
catalysts are less well defined in these cases.[5] In addition to C-
C bond formation C-N bonds have also been formed where
MnCl2 has been presented as a catalyst for connecting aryl
halides and amines.[6]
A constant concern in the development of catalytic reactions
with new metals is the possible presence of trace amounts of
other metals which may then be the actual catalyst for the
transformation.[7] For the cross-coupling reaction minute
quantities of palladium or copper impurities have in some cases
been responsible for a transformation which was otherwise
believed to be either metal free or catalyzed by a different
metal.[8]
Herein, we describe our further development of the MnCl2-
catalyzed coupling between aryl halides and Grignard reagents.
The substrate scope in the halide has been extended beyond
cyano- and ester-activated substrates by performing the reaction
at elevated temperature in a microwave oven. In addition, we
describe our attempts to reproduce two previously published
manganese-catalyzed coupling reactions[4,6] where we believe
trace amounts of other metals serve as the actual catalyst.
Introduction
Manganese is one of the most abundant and cheapest metals in
the periodic table. Manganese is also present in all living
systems and constitutes
a A
relatively non-toxic metal.[1]
significant number of manganese-catalyzed homogeneous
reactions have therefore been developed over the past decade
in order to replace the expensive and toxic platinum group
metals in the same reactions or to develop entire new
transformations.[2] This is also true for the manganese-catalyzed
cross-coupling reaction to form C-C and C-N bonds where
manganese(II) salts have been employed as the catalysts.
Progress, however, has been slower in the development of
these transformations and some reactions are poorly understood.
Several groups have studied the MnCl2-catalyzed cross-
[a]
C. Santilli, S. S. Beigbaghlou, A. Ahlburg, G. Antonacci, P. Fristrup,
R. Madsen
Department of Chemistry
Technical University of Denmark
2800 Kgs. Lyngby (Denmark)
E-mail: rm@kemi.dtu.dk
http://www.kemi.dtu.dk
[b]
[c]
[d]
S. S. Beigbaghlou
Faculty of Chemistry
Kharazmi University of Tehran (Tarbiat Moalem University)
49 Mofateh Avenue, Tehran 15719 (Iran)
P.-O. Norrby
Department of Chemistry and Molecular Biology
University of Gothenburg
Kemigården 4, 412 96 Göteborg (Sweden)
P.-O. Norrby
Pharmaceutical Sciences
Results and Discussion
Bromobenzene and p-tolylmagnesium bromide in THF solution
were selected as the substrates in a 1:2 ratio for the initial
studies with 10% of MnCl2 since no cross-coupling occurred in
this case at room temperature or upon refluxing the reaction
mixture.[3a] However, heating the solution in a microwave oven at
180 °C produced the desired heterocoupling product in 40%
yield with homocoupling of the Grignard reagent and
AstraZeneca
Pepparedsleden 1, 431 83 Mölndal (Sweden)
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201701005
European Journal of Organic Chemistry
FULL PAPER
dehalogenation of the halobenzene as the major side reactions
(Table 1, Entry 1). Increasing or decreasing the temperature
gave slightly lower yields (Entries 2 – 5) and the same was
observed when MnCl2 was replaced with MnF2, MnBr2 and MnI2
(Entries 6 – 8). With one equivalent of MnCl2 the coupling yield
increased to 49% at 160 °C (Entry 9).
8[f]
H
Me
Me
H
THF[d]
160
160
160
160
140
120
100
160
120
120
120
120
2
28
49
23
60
57
70[h]
69
65
75
33
0
9
H
THF[d]
2
10
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Et2O[d]
Et2O[g]
Et2O[g]
Et2O[g]
Et2O[g]
2-MeTHF[g]
Et2O[g]
Et2O[g]
Et2O[g]
Et2O[g]
1
11
H
1
Since dehalogenation is the major side reaction, the cross-
coupling was also investigated with p-bromotoluene and
phenylmagnesium bromide. The latter is now prepared in Et2O
and initially no improvement was observed in the yield (Entry 10).
However, increasing the concentration of the Grignard reagent
from 1 M to 3 M raised the yield to 60 – 70% depending on the
temperature and the reaction time (Entries 11 – 14). A similar
result was obtained when the same concentration of the
Grignard reagent in 2-methyltetrahydrofuran (2-MeTHF) was
used (Entry 15), which underlines the importance of the
concentration to suppress the dehalogenation. The Schlenk
equilibrium in Et2O favors the monomeric ArMgX while Ar2Mg +
MgX2 becomes more preferred in THF.[9] However, the Schlenk
equilibrium can shift very fast and is therefore not believed to be
responsible for the different reactivities in THF and Et2O. Notably,
4% of 4,4’-dimethylbiphenyl was also formed in entry 13 arising
from homocoupling of the aryl halide. In the other entries small
traces of this homocoupling product was also observed, but it
was not further quantified. Replacing p-bromotoluene with p-
iodotoluene gave an additional improvement in the outcome
while p-chlorotoluene resulted in a lower yield (Entries 16 and
17). A control experiment without MnCl2 gave no conversion of
the starting materials and as a result no cross-coupling,
dehalogenation and homocoupling were observed (Entry 18).
Similarly, no cross-coupling occurred when an aryl triflate was
treated with a Grignard reagent under the reaction conditions.
Consequently, a 3 M solution of the Grignard reagent in Et2O
and a temperature of 120 °C were selected for general use since
it affords a reasonable reaction time in the microwave oven
(Entry 13). For comparison, the experiment in Entry 13 was also
performed by conventional heating in an oil bath overnight which
resulted in 62% yield of 4-methylbiphenyl (Entry 19).
12
H
2
13
H
5
14
H
12
2
15
H
16[i]
17[j]
18[k]
19[l]
H
5
H
5
H
5
H
18
62
[a] Conditions: aryl bromide (2 mmol), arylmagnesium bromide (4 mmol),
MnCl2 (0.2 mmol), decane (1 mmol, internal standard) and solvent (8 mL, i.e.
Grignard concentration 0.5 M) in a closed vial with microwave heating. [b] GC
yield. [c] With MnF2 instead of MnCl2. [d] 4 mL solvent (Grignard concentration
1 M). [e] With MnBr2 instead of MnCl2. [f] With MnI2 instead of MnCl2. [g] 1.3
mL solvent (Grignard concentration 3 M). [h] 4,4’-Dimethylbiphenyl (4%) was
also formed. [i] With p-iodotoluene. [j] With p-chlorotoluene. [k] Without MnCl2.
[l] Performed in an oil bath.
With the optimized procedure available the substrate scope
could now be explored in further detail with different aryl
bromides and iodides (Table 2). 4-Methylbiphenyl was isolated
in 66% yield from the reaction between p-iodotoluene and
phenylmagnesium bromide (Entry 1). Dehalogenation of the aryl
iodide was responsible for the remaining conversion of the
starting halide. With p-bromotoluene as the aryl halide the yield
of 4-methylbiphenyl decreased to 47% (Entry 2). m-Iodotoluene
afforded 3-methylbiphenyl in 50% yield (Entry 3) while the same
reaction with m-bromotoluene only gave about 20% yield (result
not shown). This again illustrates the lower yield with the aryl
bromide as compared to the aryl iodide. o-Iodotoluene furnished
the cross-coupling product in 34% yield (Entry 4) with
dehalogenation of the starting material as the main side reaction.
Aryl iodides with two methyl substituents in the meta and/or para
position gave the corresponding biphenyl compounds in 77%
and 62% yield (Entries 5 and 6). Lower yields were obtained
with methoxy and dimethylamino groups in the meta or para
positions due to dehalogenation of the aryl halide (Entries 7 –
10). p-Chloroiodo- and p-fluoroiodobenzene were also reacted
with phenylmagnesium bromide, but only about 10% of the
desired biphenyl compounds were obtained in these two cases
(result not shown). Very small traces of the homocoupling
product from the aryl halide were detected in several entries, but
not further quantified. The studies show that aryl iodides are the
preferred coupling partners and decent yields can be obtained
with simple methyl substituted substrates while other
substituents afford lower yields.
Table 1. Optimization of MnCl2-catalyzed cross-coupling.[a]
Entry
1
R
H
H
H
H
H
H
H
R'
Solvent
THF
T [°C]
180
200
160
140
120
160
160
t [h]
2
Yield [%][b]
Me
Me
Me
Me
Me
Me
Me
40
29
33
34
21
10
29
2
THF
1
3
THF
2
4
THF
2
5
THF
14
2
6[c]
7[e]
THF[d]
THF[d]
2
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201701005
European Journal of Organic Chemistry
FULL PAPER
but-1-ene was reacted with phenylmagnesium bromide under
the optimized conditions (Scheme 1).[10] The reaction gave a
mixture of the cyclization products 1 – 3 in a combined yield of
Table 2. Cross-coupling with different aryl halides.[a]
41%. In addition, the cross-coupling product
4 and the
dehalogenation product 5 were obtained in 5% and 13% yield,
respectively. Again, the results provide a strong indication for the
involvement of an aryl radical. The formation of olefins in radical
clock experiments has previously been observed and depends
on the ease by which the generated radicals are trapped by the
solvent.[11]
Entry
1
Ar‒X
Ar‒Ph
Yield [%][b]
66[c]
2
3
47[c]
50[c]
4
5
6
34[c]
77
62[c]
7
26
Scheme 1. Aryl radical trapping experiments.
8
9
23
33
These results lead to the proposal of the same SRN
1
mechanism as in our previous cross-coupling with
chlorobenzonitriles (Scheme 2).[3a] The reaction is initiated by
single electron transfer to the aryl halide to afford radical anion 6
and the most likely one electron donor is the triphenylmanganate
complex 7 which is a known radical initiator[12] and is readily
formed from MnCl2 and phenylmagnesium bromide.[13] The
Grignard reagent appears to be unable to initiate the reaction
since the direct cross-coupling between an aryl halide and an
arylmagnesium halide was shown to proceed without the
formation of an aryl radical.[14] Subsequent loss of the halide
furnishes the aryl radical 8 which upon reaction with an aryl
nucleophile gives rise to the biphenyl radical anion 9. The
Grignard reagent and the triphenylmanganate complex 7 can
both serve as the aryl nucleophile where the latter is a softer
nucleophile than the former.[13] The triphenylmanganate complex
is the most likely nucleophile since a Grignard reagent has never
been shown to react with an aryl radical. Final single electron
transfer from the radical anion 9 to the starting aryl halide closes
the catalytic cycle. The high temperature is most likely required
in the present case since the electron-donating substituents
destabilize radical anion 6 and make aryl radical 8 less
electrophilic[15] as opposed to the electron-withdrawing
substituents in our earlier study.[3a] For the same reason, the
homocoupling product from the aryl halide and the
10
28
[a] Conditions: aryl bromide (2 mmol), phenylmagnesium bromide (4 mmol),
MnCl2 (0.2 mmol), and Et2O (1.3 mL, i.e. Grignard concentration 3 M) in a
closed vial with microwave heating at 120 °C for 5 h. [b] Isolated yield. [c] Yield
based on NMR since isolated product not completely pure.
The substrate scope does not provide any information about
the mechanism, but based on our recent analysis of the coupling
with chlorobenzonitriles,[3a] the reaction may proceed by a
radical pathway. In fact, the small amounts of a homocoupling
product from the aryl halide may result from dimerization of an
intermediate aryl radical. Therefore, two experiments were
conducted in order to trap this radical. First, the coupling in
Table 1, Entry 16 was repeated in the presence of 10 equiv. of
cyclohexa-1,4-diene (Scheme 1). This afforded 4-methylbiphenyl
in only 7% GC yield while the dehalogenation product was now
obtained in 56% yield. The substantial dehalogenation in the
presence of the 1,4-diene indicates the involvement of an aryl
radical. To trap this species with an olefin 4-(2-bromophenyl)-
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201701005
European Journal of Organic Chemistry
FULL PAPER
dehalogenation product with cyclohexa-1,4-diene were observed
in the present investigation and not in our previous study.[3a]
transformation with CuI. The experiments with MnBr2 and CuI
were performed several times, but the results were the same.
The employed solvent, reactants and metal salts were analyzed
by inductively coupled plasma mass spectrometry (ICP-MS)
which showed the presence of several metal impurities such as
lead, mercury and chromium. However, palladium and nickel
were not identified in any of the samples beyond the detection
limit, i.e. 1 ppm for Pd and 5 ppm for Ni. No conversion of the
iodide occurred when the reaction in Scheme 3 was carried out
with 10% of NiCl2 instead of MnBr2. However, with 10% of
Pd(OAc)2 the coupling proceeded to give 4-methylbiphenyl in
52% GC yield together with the two homocoupling products.
Several experiments were then performed with lower amounts of
Pd(OAc)2 and even with 0.003% of Pd(OAc)2 was it possible to
obtain 42% yield of the cross-coupling product. Lowering the
amount further to 0.0004% of Pd(OAc)2, however, resulted in no
conversion of the aryl halide. Consequently, we do not believe
the reported cross-coupling reaction[4] is catalyzed by MnBr2 or
CuI, but instead it may be mediated by very small amounts of
palladium in the starting materials.
Scheme 2. Proposed mechanism for MnCl2-catalyzed cross-coupling.
The coupling between aryl halides and amines was
described to take place in either DMSO or water with 5 – 20% of
MnCl2 as the catalyst and a temperature of 130 – 135 °C.[6]
Trans-1,2-diaminocyclohexane or proline was employed as the
ligand and a base (Cs2CO3, K3PO4 or NaOtBu) was also
added.[6] Regioisomeric products resulting from a benzyne
intermediate were reported with NaOtBu as the base[6b] while no
regioisomers were disclosed with Cs2CO3 or K3PO4.[6a,c] MnCl2 of
either >99% or 99.99% purity was used in the published
reactions,[6] but no analysis for trace elements was performed on
any of the components in the transformations. We selected the
coupling between m-iodotoluene and pyrazole for our
experiment since the reaction was reported to give 75% yield
under the conditions in Scheme 4[6c] and it would be possible to
pinpoint a possible benzyne intermediate in this case.
The discovery of a radical pathway in these MnCl2-catalyzed
couplings made us revisit two previously published procedures
for manganese-catalyzed cross-couplings. In 1997 a MnBr2-
catalyzed method was presented for coupling of aryl iodides and
arylstannanes[4] while in 2009 – 2012 three papers described the
MnCl2-catalyzed condensation between aryl halides and
amines.[6] No mechanistic information was presented in any of
these publications and based on the reactants and the
conditions it appeared doubtful that radical pathways were
involved. Consequently, we decided to repeat the experiments in
these publications in an attempt to understand the puzzling
reactivity.
The coupling between aryl iodides and –stannanes was
described to take place under the conditions shown in Scheme 3
where the addition of one equivalent of NaCl was essential
(although it could be replaced with KCl).[4] The coupling was
reported to give a lower yield with MnCl2 while no coupling was
observed with MnI2 or when using aryl bromides or triflates as
substrates.[4] No information was provided about the purity of the
reagents that were used to carry out these reactions.[4]
Scheme 4. Reported MnCl2-catalyzed coupling between aryl halides and
amines.[6c]
We repeated the experiment in Scheme 4 under the exact
same conditions and with redistilled substrates, ligand and
deionized water. The yield of 1-(m-tolyl)-1H-pyrazole was 6 –
8% in our hands depending on the source of MnCl2 and with
unreacted starting materials remaining. No isomers of the
product were formed which excludes an aryne pathway.
Interestingly, the yield increased to 45% when deionized water
was replaced with ordinary and undistilled tap water. When the
reaction in Scheme 4 was done in the absence of MnCl2 (and
with deionized water) the yield of 1-(m-tolyl)-1H-pyrazole
decreased to 2 – 3%. These experiments indicate that MnCl2 is
Scheme 3. Reported MnBr2-catalyzed coupling between aryl halides and -
stannnes.[4]
When we repeated the experiment in Scheme 3 under the
exact same conditions we observed no conversion of the aryl
iodide and no formation of 4-methylbiphenyl. The coupling in
Scheme 3 was also reported to give 81% yield with 10% of CuI
as a catalyst instead of MnBr2.[4] Again, we detected no
conversion of p-iodotoluene when we carried out the
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201701005
European Journal of Organic Chemistry
FULL PAPER
General Procedure for Cross Coupling: MnCl2 (25 mg, 0.2 mmol) was
placed in a predried microwave vial (with a liquid volume allowance
between 0.5 mL and 2 mL) equipped with a magnetic stirrer and then
sealed with a rubber septum. The vial was evacuated and refilled three
times with nitrogen through a syringe. The aryl halide (2 mmol) and
decane (1 mmol, internal standard) were placed in the vial followed by
addition of 3 M phenylmagnesium bromide (4 mmol) in diethyl ether
under a flow of nitrogen. The reaction vial was sealed with a cap and
placed in the microwave reactor at 120 °C for 5 h. The mixture was
quenched with a saturated solution of ammonium chloride. The phases
were separated and the aqueous phase extracted with diethyl ether (3 x
5 mL). The combined organic phases were dried over Na2SO4, filtered
and evaporated in vacuo to give the crude product, which were purified
by flash chromatography eluting with pentane or pentane containing 0.5
– 10% EtOAc.
not able to catalyze the coupling between aryl halides and
amines, but instead traces of another element is most likely
responsible for the transformation. It has previously been shown
that 0.01% of CuCl2 is able to catalyze the coupling between
iodobenzene and pyrazole in 88% yield under very similar
conditions.[16] In fact, when we performed the reaction in
Scheme 4 with 0.01% of CuCl2•H2O instead of MnCl2, the yield
of 1-(m-tolyl)-1H-pyrazole increased to 78% while 29% was
obtained with a 0.001% loading of CuCl2•H2O. The solvent and
the reactants were again analyzed by ICP-MS, but no copper
impurities were found beyond the detection limit which was 20
ppm for MnCl2, 0.02 ppm for m-iodotoluene and K3PO4, and
0.005 ppm for pyrazole, deionized water and the ligand.
Pyrazole, although, contained 0.02 ppm of copper before the
amine was distilled from EDTA. In all, we do not believe the
reported C-N bond formations[6] are catalyzed by MnCl2, but
most likely minute amounts of copper in the reagents and the
solvent are responsible for the observed transformations.
4-Methyl-1,1’-biphenyl:[18] Table 2, Entries 1 and 2. Prepared from p-
iodotoluene and phenylmagnesium bromide according to the general
procedure. The residue was purified by flash column chromatography
eluting with pentane to yield a mixture of biphenyl and 4-methyl-1,1’-
biphenyl. The yield of the latter was determined by 1H NMR. 1H NMR
(400 MHz, CDCl3):  = 7.52–7.49 (m, 2 H), 7.42–7.32 (m, 4 H), 7.28–7.21
(m, 1 H), 7.18–7.14 (m, 2 H), 2.31 (s, 3 H) ppm. 13C NMR (100 MHz,
CDCl3):  = 141.3, 138.5, 137.1, 129.6, 128.8, 127.1, 127.1, 21.2 ppm.
MS: m/z = 168.00 [M]+.
The reaction in Scheme 4 has also been published with 10%
of CoCl2•6H2O and 20% of N,N’-dimethylethylenediamine
instead of MnCl2 and trans-1,2-diaminocyclohexane where a
75% yield of 1-(m-tolyl)-1H-pyrazole was obtained.[17] In our
hands and with purified reactants the yield was only 6% which
also in this case indicates that the added cobalt salt is not the
true catalyst for the reported transformation.
3-Methyl-1,1’-biphenyl:[18] Table 2, Entry 3. Prepared from m-
iodotoluene and phenylmagnesium bromide according to the general
procedure. The residue was purified by flash column chromatography
eluting with pentane to yield a mixture of biphenyl and 3-methyl-1,1’-
biphenyl. The yield of the latter was determined by 1H NMR. 1H NMR
(400 MHz, CDCl3):  = 7.62–7.59 (m, 2 H), 7.48–7.33 (m, 6 H), 7.19–7.17
(m, 1 H), 2.44 (s, 3 H) ppm. 13C NMR (100 MHz, CDCl3):  = 141.5, 141.4,
138.5, 128.8, 128.8, 128.1, 127.3, 124.4, 21.7 ppm. MS: m/z = 168.05
[M]+.
Conclusions
In summary, the substrate scope of the MnCl2-catalyzed
coupling between aryl halides and Grignard reagents has been
extended to methyl-substituted aryl iodides by performing the
reaction a microwave oven. An aryl radical was identified by a
radical clock experiment and the coupling is therefore also in this
case believed to proceed by a SRN1 mechanism. The role of
MnCl2 is most likely to react with the Grignard reagent to provide
a softer nucleophile which can also serve as a one-electron
donor. The mechanistic information caused suspicion about two
previously published manganese-catalyzed cross-coupling
reactions. In fact, control experiments revealed that
manganese(II) salts are not able to catalyze the coupling
between aryl halides and organostannanes/amines. The results
illustrate the importance of mechanistic experiments as well as
analyses for trace metals when developing new metal catalysts
for known catalytic transformations.
2-Methyl-1,1’-biphenyl:[19] Table 2, Entry 4. Prepared from o-
iodotoluene and phenylmagnesium bromide according to the general
procedure. The residue was purified by flash column chromatography
eluting with pentane to yield a mixture of biphenyl and 2-methyl-1,1’-
biphenyl. The yield of the latter was determined by 1H NMR. 1H NMR
(400 MHz, CDCl3):  = 7.46–7.40 (m, 2 H), 7.37–7.32 (m, 3 H), 7.28–7.23
(m, 4 H), 2.28 (s, 3 H) ppm. 13C NMR (100 MHz, CDCl3):  = 142.1, 142.1,
135.5, 130.4, 129.9, 129.3, 128.2, 127.3, 126.9, 125.9, 20.6 ppm. MS:
m/z = 168.05 [M]+.
3,4-Dimethyl-1,1’-biphenyl:[20] Table 2, Entry 5. Prepared from 1-iodo-
3,4-dimethylbenzene and phenylmagnesium bromide according to the
general procedure. The residue was purified by flash column
chromatography eluting with pentane to yield the desired product as a
colorless oil (75%). 1H NMR (400 MHz, CDCl3):  = 7.60–7.58 (m, 2 H),
7.45–7.39 (m, 3 H), 7.36–7.31 (m, 2 H), 7.22 (d, J = 7.7 Hz, 1 H), 2.35 (s,
3 H), 2.32 (s, 3 H) ppm. 13C NMR (100 MHz, CDCl3):  = 141.4, 139.0,
137.0, 135.8, 130.2, 128.8, 128.6, 127.1, 127.0, 124.6, 20.1, 19.6 ppm.
MS: m/z = 182.00 [M]+.
Experimental Section
3,5-Dimethyl-1,1’-biphenyl:[18] Table 2, Entry 6. Prepared from 1-iodo-
3,5-dimethylbenzene and phenylmagnesium bromide according to the
general procedure. The residue was purified by flash column
chromatography eluting with pentane to yield a mixture of biphenyl and
3,5-methyl-1,1’-biphenyl. The yield of the cross-coupling product was
General: All the reactions were performed in a Biotage microwave
reactor and monitored by gas chromatography on a Shimadzu GCMS-
QP2010S instrument fitted with an Equity 5, 30 m × 0.25 mm × 0.25 m
column. Flash column chromatography separations were performed on
silica gel 60 (40 – 63 m). NMR spectra were recorded on a Bruker
Ascend 400 spectrometer. Chemical shifts were measured relative to the
signals of residual CHCl3 (H = 7.26 ppm) and CDCl3 (C = 77.16 ppm).
Analyses for trace metals by ICP-MS was performed by ALS Denmark
A/S.
1
determined by 1H NMR. H NMR (400 MHz, CDCl3):  = 7.63–7.59 (m, 2
H), 7.49–7.43 (m, 2 H), 7.39–7.33 (m, 2 H), 7.24 (br s, 1 H), 7.24 (br s, 1
H), 2.41 (s, 6 H) ppm. 13C NMR (100 MHz, CDCl3):  = 141.6, 141.4,
138.4, 129.0, 128.8, 127.3, 127.2, 125.2, 21.6 ppm. MS: m/z = 182.00
[M]+.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201701005
European Journal of Organic Chemistry
FULL PAPER
3-Methoxy-1,1’-biphenyl:[19] Table 2, Entry 7. Prepared from m-
iodoanisole and phenylmagnesium bromide according to the general
procedure. The residue was purified by flash column chromatography
eluting with pentane/EtOAc (10:0.05) to yield the desired product as
white solid (26%). 1H NMR (400 MHz, CDCl3):  = 7.61–7.58 (m, 2 H),
7.46–7.42 (m, 2 H), 7.38–7.33 (m, 2 H), 7.20–7.18 (m, 1 H), 7.14–7.13
(m, 1 H), 6.92–6.89 (m, 1 H), 3.87 (s, 3 H) ppm. 13C NMR (100 MHz,
CDCl3):  = 160.1, 142.9, 141.3, 128.9, 128.9, 127.6, 127.3, 119.8, 113.0,
112.8, 55.5 ppm. MS: m/z = 184.00 [M]+.
4-Phenylbut-1-ene (5):[26] 1H NMR (400 MHz, CDCl3):  = 7.53–7.12 (m,
5 H), 5.88 (ddt, J = 16.9,10.2, 6.6 Hz, 1 H), 5.05–5.04 (m, 1 H), 5.00 (dd,
J = 10.2, 1.6 Hz, 1 H), 2.75–2.71 (m, 2 H), 2.42–2.37 (m, 2 H) ppm. 13
C
NMR (100 MHz, CDCl3):  = 142.0, 138.2, 128.6, 126.2, 125.9, 115.0,
35.7, 35.5 ppm. MS: m/z = 132.05 [M]+.
Acknowledgements
4-Methoxy-1,1’-biphenyl:[18] Table 2, Entry 8. Prepared from p-
iodoanisole and phenylmagnesium bromide according to the general
procedure. The residue was purified by flash column chromatography,
eluting with pentane/EtOAc (10:0.05) to yield the desired product as
white solid (23%). 1H NMR (400 MHz, CDCl3):  = 7.57–7.52 (m, 4 H),
7.44–7.40 (m, 2 H), 7.33–7.28 (m, 1 H), 7.00–6.97 (m, 2 H), 3.86 (s, 3 H)
ppm. 13C NMR (100 MHz, CDCl3):  = 159.3, 141.0, 133.9, 128.9, 128.3,
126.9, 126.8, 114.3, 55.5 ppm. MS: m/z = 184.00 [M]+.
We thank the Danish Council for Independent Research ‒
Technology and Production Sciences for financial support (grant
1335-00153).
Keywords: Cross-coupling • Grignard reagent • Manganese •
Radical reactions • Trace analysis
N,N-Dimethyl-[1,1’-biphenyl]-3-amine:[21] Table 2, Entry 9. Prepared
from 3-bromo-N,N-dimethylaniline and phenylmagnesium bromide
according to the general procedure. The residue was purified by flash
column chromatography eluting with pentane/EtOAc (10:1) to yield the
desired product as colorless oil (33%). 1H NMR (400 MHz, CDCl3):  =
7.64–7.61 (m, 2 H), 7.47–7.44 (m, 2 H), 7.38–7.32 (m, 2 H), 7.00–6.97
(m, 2 H), 6.80–6.77 (m, 1 H), 3.03 (s, 6 H) ppm. 13C NMR (100 MHz,
CDCl3):  = 151.1, 142.4, 142.4, 129.5, 128.7, 127.5, 127.2, 116.1, 111.8,
111.8, 40.9 ppm. MS: m/z = 197.05 [M]+.
[1]
[2]
A. B. Santamaria, S. I. Sulsky, J. Toxicol. Environ. Health A 2010, 73,
128‒155.
(a) J. R. Carney, B. R. Dillon, S. P. Thomas, Eur. J. Org. Chem. 2016,
3912‒3929. (b) W. Liu, L. Ackermann, ACS Catal. 2016, 6, 3743‒3752.
(c) D. A. Valyaev, G. Lavigne, N. Lugan, Coord. Chem. Rev. 2016, 308,
191‒235.
[3]
(a) G. Antonacci, A. Ahlburg, P. Fristrup, P.-O. Norrby, R. Madsen, Eur.
J. Org. Chem. (2017) DOI: 10.1002/ejoc.201700981. (b) G. Cahiez, O.
Gager, F. Lecomte, Org. Lett. 2008, 10, 5255‒5256. (c) M. Rueping, W.
Ieawsuwan, Synlett 2007, 247‒250. (d) G. Cahiez, F. Lepifre, P.
Ramiandrasoa, Synthesis 1999, 2138‒2144. (e) M. Alami, P.
Ramiandrasoa, G. Cahiez, Synlett 1998, 325‒327.
N,N-Dimethyl-[1,1’-biphenyl]-4-amine:[22] Table 2, Entry 10. Prepared
from 4-bromo-N,N-dimethylaniline and phenylmagnesium bromide
according to the general procedure. The residue was purified by flash
column chromatography eluting with pentane/EtOAc (10:1) to yield the
desired product as white solid (28%). 1H NMR (400 MHz, CDCl3):  =
7.59–7.52 (m, 4 H), 7.43–7.39 (m, 2 H), 7.29–7.25 (m, 1 H), 6.84–6.82
(m, 2 H), 3.01 (s, 6 H) ppm. 13C NMR (100 MHz, CDCl3):  = 150.1, 141.3,
128.8, 127.9, 126.4, 126.1, 113.0, 40.8 ppm. MS: m/z = 197.00 [M]+.
[4]
[5]
S.-K. Kang, J.-S. Kim, S.-C. Choi, J. Org. Chem. 1997, 62, 4208‒4209.
(a) J. Qiao, W. Zhu, G. Zhuo, H. Zhou, X. Jiang, Chin. J. Catal. 2008,
29, 209‒211. (b) S. Iyer, V. V. Thakur, J. Mol. Catal. A: Chem. 2000,
157, 275‒278.
[6]
(a) F.-F. Yong, Y.-C. Teo, Synlett 2012, 2106‒2110. (b) F.-F. Yong, Y.-
C. Teo, Tetrahedron Lett. 2010, 51, 3910‒3912. (c) Y.-C. Teo, F.-F.
Yong, C.-Y. Poh, Y.-K. Yan, G.-L. Chua, Chem. Commun. 2009,
6258‒6260.
Radical Clock Experiment in Scheme 1: Commercially available 4-(2-
bromophenyl)-but-1-ene was treated with MnCl2 and phenylmagnesium
bromide as described above in the general procedure to afford an
inseparable mixture of compounds 1 – 3 and 5 which were characterized
by NMR and the yield determined with an internal standard. In addition,
[7]
[8]
I. Thomé, A. Nijs, C. Bolm, Chem. Soc. Rev. 2012, 41, 979‒987.
(a) R. B. Bedford, M. Nakamura, N. J. Gower, M. F. Haddow, M. A. Hall,
M. Huwe, T. Hashimoto, R. A. Okopie, Tetrahedron Lett. 2009, 50,
6110‒6111. (b) S. L. Buchwald, C. Bolm, Angew. Chem. Int. Ed. 2009,
48, 5586‒5587; Angew. Chem. 2009, 121, 5694‒5695. (c) H. Plenio,
Angew. Chem. Int. Ed. 2008, 47, 6954‒6956; Angew. Chem. 2008, 120,
7060‒7063. (d) R. K. Arvela, N. E. Leadbeater, M. S. Sangi, V. A.
Williams, P. Granados, R. D. Singer, J. Org. Chem. 2005, 70, 161‒168.
J. Tammiku-Taul, P. Burk, A. Tuulmets, J. Phys. Chem. A 2004, 108,
133‒139.
compound
4 was isolated as a mixture with biphenyl and again
characterized by NMR.
1-Methylindan (1):[23] 1H NMR (400 MHz, CDCl3):  = 7.53–7.12 (m, 4 H),
3.20 (sextet, J = 7.2 Hz, 1 H), 3.02–2.79 (m, 2 H), 2.36–2.28 (m, 1 H),
1.66–1.57 (m, 1 H), 1.31 (d, J = 6.9 Hz, 3 H) ppm. 13C NMR (100 MHz,
CDCl3):  = 148.9, 144.0, 128.4, 126.2, 124.5, 123.3, 39.6, 34.9, 31.6,
20.0 ppm. MS: m/z = 132.05 [M]+.
[9]
[10] Recently, alkyl radicals were trapped in radical clock experiments for
the MnBr2-catalyzed borylation of alkyl chlorides, see: T. C. Atack, S. P.
Cook, J. Am. Chem. Soc. 2016, 138, 6139‒6142.
1-Methyleneindan (2):[24] 1H NMR (400 MHz, CDCl3):  = 7.53–7.12 (m,
4 H), 5.47 (t, J = 2.6 Hz, 1 H), 5.09–5.08 (m, 1 H), 3.02–2.98 (m, 2 H),
2.93–2.79 (m, 2 H) ppm. 13C NMR (100 MHz, CDCl3):  = 150.7, 146.9,
141.2, 128.4, 126.5, 125.5, 120.7, 102.6, 31.3, 30.2 ppm.
[11] C. Ha, J. H. Horner, M. Newcomb, T. R. Varick, B. R. Arnold, J. Lusztyk,
J. Org. Chem. 1993, 58, 1194‒1198.
[12] J. Nakao, R. Inoue, H. Shinokubo, K. Oshima, J. Org. Chem. 1997, 62,
1910‒1911.
3-Methyl-1H-indene (3):[24] 1H NMR (400 MHz, CDCl3):  = 7.53–7.12 (m,
4 H), 6.22–6.21 (m, 1 H), 3.34–3.33 (m, 2 H), 2.20–2.18 (m, 3 H) ppm.
13C NMR (100 MHz, CDCl3):  = 146.2, 144.5, 140.1, 128.9, 126.2, 124.6,
123.7, 119.0, 37.8, 13.2 ppm.
[13] G. Cahiez, C. Duplais, J. Buendia, Chem. Rev. 2009, 109, 1434‒1476.
[14] N. Uchiyama, E. Shirakawa, T. Hayashi, Chem. Commun. 2013, 49,
364‒366.
[15] F. De Vleeschouwer, V. Van Speybroeck, M. Waroquier, P. Geerlings,
F. De Proft, Org. Lett. 2007, 9, 2721‒2724.
2-(But-3-en-1-yl)-1,1’-biphenyl (4):[25] 1H NMR (400 MHz, CDCl3):  =
7.63–7.24 (m, 9 H), 5.73 (ddt, J = 16.9,10.2, 6.6 Hz, 1 H), 4.94 (q, J = 1.7
Hz, 1 H), 4.91–4.89 (m, 1 H), 2.72–2.68 (m, 2 H), 2.26–2.20 (m, 2 H)
ppm. 13C NMR (100 MHz, CDCl3):  = 142.1, 142.0, 139.4, 138.3, 130.2,
129.3, 128.2, 127.5, 126.9, 125.9, 114.8, 35.3, 32.7 ppm. MS: m/z =
208.00 [M]+.
[16] P.-F. Larsson, A. Correa, M. Carril, P.-O. Norrby, C. Bolm, Angew.
Chem. Int. Ed. 2009, 48, 5691‒5693; Angew. Chem. 2009, 121,
5801‒5803.
[17] Y.-C. Teo, Adv. Synth. Catal. 2009, 351, 720‒724.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201701005
European Journal of Organic Chemistry
FULL PAPER
[18] H. Yang, L. Zhang, L. Jiao, Chem. Eur. J. 2017, 23, 65‒69.
[19] Q.-X. Liu, K.-Q. Cai, Z.-X. Zhao, RSC Adv. 2015, 5, 85568‒85578.
[20] F. Gaviña, A. M. Costero, A. M. González, J. Org. Chem. 1990, 55,
2060‒2063.
[24] M. R. Friedfeld, M. Shevlin, G. W. Margulieux, L.-C. Campeau, P. S.
Chirik, J. Am. Chem. Soc. 2016, 138, 3314‒3324.
[25] N. Uchiyama, E. Shirakawa, T. Hayashi, Chem. Commun. 2013, 49,
364‒366.
[21] D. Heijnen, J.-B. Gualtierotti, V. Hornillos, B. L. Feringa, Chem. Eur. J.
2016, 22, 3991‒3995.
[26] A. Chatterjee, S. H. H. Eliasson, K. W. Törnroos, V. R. Jensen, ACS
Catal. 2016, 6, 7784‒7789.
[22] A. Ohtsuki, K. Yanagisawa, T. Furukawa, M. Tobisu, N. Chatani, J. Org.
Chem. 2016, 81, 9409‒9414.
[23] W. F. Bailey, M. J. Mealy, J. Am. Chem. Soc. 2000, 122, 6787‒6788.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201701005
European Journal of Organic Chemistry
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
Layout 1:
FULL PAPER
Text for Table of Contents
Key Topic*
Author(s), Corresponding Author(s)*
Page No. – Page No.
Title
((Insert TOC Graphic here: max.
width: 5.5 cm; max. height: 5.0 cm;
NOTE: the final letter height should
not be less than 2 mm.))
*one or two words that highlight the emphasis of the paper or the field of the study
Layout 2:
FULL PAPER
Manganese Catalysis
Carola Santilli, Somayyeh Sarvi
Beigbaghlou, Andreas Ahlburg,
Giuseppe Antonacci, Peter Fristrup, Per-
Ola Norrby, Robert Madsen*
Page No. – Page No.
The substrate scope of the MnCl2-catalyzed coupling between aryl halides and
Grignard reagents has been extended to unactivated halides by performing the
reaction in a microwave oven. An aryl radical was identified as an intermediate
leading to the proposal of an SRN1 pathway. Two previously published cross
coupling reactions with manganese catalysts were revisited and found impossible to
reproduce without the assistance of small amounts of palladium or copper.
The Manganese-Catalyzed Cross-
Coupling Reaction and the Influence
of Trace Metals
This article is protected by copyright. All rights reserved.