Multimetallic catalysed cross-coupling of aryl bromides with aryl triflates

Article abstract of DOI:10.1038/nature14676

The advent of transition-metal catalysed strategies for forming new carbon-carbon bonds has revolutionized the field of organic chemistry, enabling the efficient synthesis of ligands, materials, and biologically active molecules. In cases where a single metal fails to promote a selective or efficient transformation, the synergistic cooperation of two distinct catalysts - multimetallic catalysis - can be used instead. Many important reactions rely on multimetallic catalysis, such as the Wacker oxidation of olefins and the Sonogashira coupling of alkynes with aryl halides, but this approach has largely been limited to the use of metals with distinct reactivities, with only one metal catalyst undergoing oxidative addition. Here, we demonstrate that cooperativity between two group 10 metal catalysts - (bipyridine)nickel and (1,3-bis(diphenylphosphino)propane)palladium - enables a general cross-Ullmann reaction (the cross-coupling of two different aryl electrophiles). Our method couples aryl bromides with aryl triflates directly, eliminating the use of arylmetal reagents and avoiding the challenge of differentiating between multiple carbon-hydrogen bonds that is required for direct arylation methods. Selectivity can be achieved without an excess of either substrate and originates from the orthogonal reactivity of the two catalysts and the relative stability of the two arylmetal intermediates. While (1,3-bis(diphenylphosphino)propane)palladium reacts preferentially with aryl triflates to afford a persistent intermediate, (bipyridine)nickel reacts preferentially with aryl bromides to form a transient, reactive intermediate. Although each catalyst forms less than 5 per cent cross-coupled product in isolation, together they are able to achieve a yield of up to 94 per cent. Our results reveal a new method for the synthesis of biaryls, heteroaryls, and dienes, as well as a general mechanism for the selective transfer of ligands between two metal catalysts. We anticipate that this reaction will simplify the synthesis of pharmaceuticals, many of which are currently made with pre-formed organometallic reagents, and lead to the discovery of new multimetallic reactions.

Full text of DOI:10.1038/nature14676

LETTER
doi:10.1038/nature14676
Multimetallic catalysed cross-coupling of aryl
bromides with aryl triflates
Laura K. G. Ackerman1, Matthew M. Lovell1 & Daniel J. Weix1
The advent of transition-metal catalysed strategies for forming 3; 70% yield), in preference to the dimers biphenyl (product 4; 17%)
new carbon-carbon bonds has revolutionized the field of organic
chemistry, enabling the efficient synthesis of ligands, materials,
and biologically active molecules1–3. In cases where a single metal
fails to promote a selective or efficient transformation, the syn-
ergistic cooperation4 of two distinct catalysts—multimetallic
catalysis—can be used instead. Many important reactions rely
on multimetallic catalysis5–10, such as the Wacker oxidation of
olefins6–8 and the Sonogashira coupling of alkynes with aryl
halides9,10, but this approach has largely been limited to the use
of metals with distinct reactivities, with only one metal catalyst
undergoing oxidative addition11,12. Here, we demonstrate that
cooperativity between two group 10 metal catalysts—(bipyridine)-
nickel and (1,3-bis(diphenylphosphino)propane)palladium—
enables a general cross-Ullmann reaction (the cross-coupling of
two different aryl electrophiles)13–15. Our method couples aryl bro-
mides with aryl triflates directly, eliminating the use of arylmetal
reagents and avoiding the challenge of differentiating between
multiple carbon–hydrogen bonds that is required for direct aryla-
tion methods16,17. Selectivity can be achieved without an excess of
either substrate and originates from the orthogonal reactivity of
the two catalysts and the relative stability of the two arylmetal
intermediates. While (1,3-bis(diphenylphosphino)propane)palla-
dium reacts preferentially with aryl triflates to afford a persistent
intermediate, (bipyridine)nickel reacts preferentially with aryl
bromides to form a transient, reactive intermediate. Although
each catalyst forms less than 5 per cent cross-coupled product in
isolation, together they are able to achieve a yield of up to 94 per
cent. Our results reveal a new method for the synthesis of biaryls,
heteroaryls, and dienes, as well as a general mechanism for the
selective transfer of ligands between two metal catalysts. We
anticipate that this reaction will simplify the synthesis of pharma-
ceuticals, many of which are currently made with pre-formed
organometallic reagents1–3, and lead to the discovery of new multi-
metallic reactions.
and bianisole (product 5; 10%). When each catalyst was allowed to
react independently with the two aryl electrophiles, however, no
such ‘cross-selectivity’ was observed. (Bpy)NiBr2 formed exclusively
biphenyl (4 in Fig. 2a) from bromobenzene (2) before reacting with
4-methoxyphenyltriflate (1) (Fig. 2c), while (dppp)PdCl2 was unreac-
tive under these conditions, consuming only a minimal amount of
substrate and forming trace amounts of benzene, anisole, bianisole
(5), and product (3) (Fig. 2d). Only when the two catalysts were com-
bined did a selective reaction occur.
To ensure that the cross-product observed was the result of synergy
between the two proposed metal catalysts, we examined the reactivity
of alternative catalysts that could be formed in situ via exchange of the
supporting ligands dppp and bpy21. Our results (Table 1, entries 5, 7–9)
demonstrate that the two alternative catalysts, (dppp)NiBr2 and
(bpy)PdCl2, are poorly reactive and form primarily biphenyl (4) rather
than cross-product (3). These findings are consistent with product
arising from our proposed mechanism, and with symmetrical biaryl
products arising from ‘mismatched’ metal–ligand combinations.
Interestingly, when all of the reagents were added together at the begin-
ning of the reaction, the results were nearly identical to reactions in
whichpre-formed catalysts wereused(Table1, entry3 versusentry10).
Examination of different ligands for both nickel and palladium
revealed that while many nitrogen-based ligands were effective, relatively
few bisphosphines worked well. Nickel complexes derived from biden-
tate ligands (2,29-bipyridine, 1,10-phenanthroline, 2-pyridylimidazole)
5 mol% (bpy)NiBr2
5 mol% (dppp)PdCl2
1
2
TfO
OMe
Br
OMe
Zn (2 equiv.)
KF (1 equiv.)
DMF, 40 °C, 12 h
+
77% yield of 3
>10:1 selectivity
1 equiv. each
We envisaged that a general solution to the cross-Ullmann reaction
could be realized through multimetallic catalysis if, first, each of the
two catalysts activated only one of the two substrates (1 and 2 in Fig. 1);
second, selective transmetallation (the transfer of ligands from one
metal to another) could be achieved (overlap of circles in Fig. 1);
and third, the catalysts were redox compatible18. Based on Hayashi’s
work19, we chose an electron-rich palladium(0) (Pd) catalyst with a
bidentate phosphine ligand, 1,3-bis(diphenylphosphino)propane
(dppp), to react preferentially with aryltrifluoromethylsulfonate esters
(aryl triflates) over aryl bromides19. From our own studies20, we chose a
bipyridine (bpy) nickel(0) (Ni) catalyst to react selectively with aryl
bromides in preference to aryl triflates.
Ni(II
)
Pd(II
)
Ar2 OTf
Ar1
Br
Ar2
Ar1
Br
TfO
Pd(0)
Ni(0)
Pd
Ni
ZnBrOTf
Pd(II
)
Ni(II
)
Ar1 Ar2
Ar1 Ar2
Zn(0)
TfO Br
We began our investigation of this multimetallic system by com-
bining a 1:1 mixture of bromobenzene (substrate 2) and 4-methoxy-
phenyltriflate (substrate 1) in the presence of bothof the catalysts and a
zinc reducing agent (Fig. 2b). Remarkably, we observed a high select-
Figure 1
|
A general cross-Ullmann reaction catalysed by a combination of
nickel and palladium. 1, 4-methoxyphenyltriflate; 2, bromobenzene; 3, the
main cross-product, 4-methoxybiphenyl. Although a nickel(0/II) andpalladium
(0/II) cycle is depicted, alternative mechanisms are also possible, such as a
ivity for formation of the cross-product, 4-methoxybiphenyl (product nickel(I/III) cycle30. OTf, triflate; DMF, N,N-dimethylformamide (solvent).
1Department of Chemistry, University of Rochester, Rochester, New York 14627-0216, USA.
0 0 M O N T H 2 0 1 5
| V O L 0 0 0 | N A T U R E | 1
G
2015 Macmillan Publishers Limited. All rights reserved

RESEARCH LETTER
Figure 2
|
Selectivities of nickel and palladium
a
b
catalysts. a, The substrates and three possible
products. 1, 4-methoxyphenyltriflate (Ar-OTf);
2, bromobenzene (Ph-Br); 3, 4-methoxybiphenyl
(Ph-Ar); 4, biphenyl (Ph-Ph); 5, bianisole (Ar-Ar).
b, (dppp)PdCl2 and (bpy)NiBr2 (10 mol% each);
c, (bpy)NiBr2 (10 mol%); and d, (dppp)PdCl2
(10 mol%) were used along with approximately
0.5 mmol of each starting material. Concentrations
of products were determined by gas chromato-
graphy analysis, corrected with an internal
standard. The low mass balance in c is due to the
formation of benzene (0.13 mmol) from
0.6
0.4
0.2
0
Ph–Br
Ar–OTf
Ph–Ph
Ar–Ar
TfO
OMe
Ni, Pd, or Ni+Pd
OMe
OMe
3
4
5
1
2
Ph–Ar
Zn (2 equiv.)
DMF, 40 ºC
+
Br
MeO
~1 equiv. each
0
100
200
300
400
500
600
700
Time (min)
c
d
hydrodehalogenation of substrate 2.
0.6
0.4
0.2
0
0.6
0.4
0.2
0
0
100
200
300
400
500
600
700
0
100
200
300
400
500
600
700
Time (min)
Time (min)
or a tridentate ligand (2,29:69,20-terpyridine) all formed reasonably entries 11 and 12 versus entries 2 and 3; see also Supplementary
selective multimetallic catalytic systems with (dppp)PdCl2 (47–69% Information, Chart S6). Together, these results suggest that the trans-
yield; Extended Data Fig. 1). By contrast, only two bisphosphine metallation step between the two catalysts is inherently selective.
ligands besides dppp were effective: 1,2-bis(diphenylphosphino)ben-
zene (56% yield) and (Z)-1,2-bis(diphenylphosphino)ethylene (53%
yield). The performance of dppp might arise from improved desired
reactivity with palladium or from diminished undesired reactivity
with nickel. We also expected the ratio of nickel to palladium to be
important to selectivity, considering the vastly different reactivities of
the individual complexes, but we found that reactions using 2/1, 1/1,
or 1/2 Pd/Ni ratios provided similar yields and selectivities (Table 1,
In many cases, the ratio of substrates also had minimal effect on the
final yield of product. In previously reported cross-Ullmann reactions
catalysed by a single metal, two or more equivalents of one reactant
were required to produce yields that were consistently greater than
60%, and ratios closer to 1/1 usually resulted in dramatically lower
yields22. However, in our multimetallic cross-Ullmann reaction, an
excess of aryl triflate did not dramatically alter the yield of cross-
product, although in some cases an excess of aryl bromide did improve
our results. For example, three equivalents of 3-bromothiophene
improved the cross-coupled yield from 64% to 85% (18 in Fig. 3).
However, an excess of aryl bromide did not improve yields with pyr-
idine substrates.
Table 1
|
Conditions for the multimetal-catalysed cross-Ullmann
reaction
The selectivity and rate of the reactions were also improved by the
addition of potassium fluoride (KF). In all examples studied, the addi-
tion of one equivalent of KF accelerated the rate of the couplings, and,
with the exception of reactions with pyridyl chlorides and activated
aryl bromides, a higher selectivity for cross-product over dimerization
was observed (Extended Data Fig. 2). In the couplingof bromobenzene
with 4-methoxyphenyl triflate (Table 1, entry 1), 89% of substrate 2
was consumed at 24 hours, producing a 62% yield of product 3. In the
presence of KF, however, 91% conversion of 2 occurred in 11 hours
with a 77% yield of 3 (Table 1, entry 2). The beneficial effect of KF seen
here is consistent with observations of other cross-coupling reactions
that are catalysed by nickel23 and palladium24,25. In nickel catalysis, KF
can reduce the amount of dimeric products formed; in palladium
catalysis, KF can alter the oxidative addition selectivity for carbon–
triflate bonds over carbon–X bonds through the formation of a pal-
ladium ‘ate’ complex. In our preliminary studies, it appears that both
the potassium ion and the fluoride ion play a role in enhancing cata-
lysis. Although optimization was performed in an inert-atmosphere
glove box, reactions could be set up on the bench top as long as
standard air-free techniques were used (Table 1, entry 13). Reactions
run without added catalysts did not consume starting materials (entry
14), demonstrating that zinc itself is slow to react with the aryl electro-
philes under these conditions and arguing against an intermediary role
of arylzinc reagents in this cross-Ullman reaction. (For further details
on reaction condition optimization, see Supplementary Information,
Charts S4–S9).
OMe
3
4
w mol% NiBr2(diglyme)
x mol% bpy
y mol% PdCl2
1
+
2
z mol% dppp
+
Zn
DMF
+
MeO
OMe
5
Entry
w
x
y
z
3*
4*
5*
1{
2
3
4
5
6
7
8
9
10{
11
12
131
14
5
5
5
5
77
62
70
3
0
6
1
4
1
67
61
66
65
0
7
21
17
4
15
42
1
4
14
10
1
0
6
0
1
0
12
14
13
10
0
5
10
0
5
10
0
10
10
0
5
10
10
10
0
5
10
10
0
0
10
5
0
10
10
0
10
10
5
10
10
0
0
5
5
10
0
10
10
5
7
5
16
13
20
18
12
0
10
5
10
10
0
10
10
5
10
0
10
0
Reactions were run on a 0.5 mmol scale in 2 ml DMF.
*Yields were determined by gas chromatography and are expressed as area per cent, uncorrected.
Remaining yields consist of unreacted starting material or hydrodehalogenation byproducts.
{One equivalent of KF was added.
{Ligands and metal salts were not individually pre-stirred to ensure ligation. All solid reagents were
added together, followed by solvent and liquids.
Applying these reaction conditions to a variety of aryl electro-
philes demonstrates the generality of this method (Fig. 3). Both
1Instead of being set up in a glove box, the reaction was set up on the bench top using standard inert
atmosphere techniques.
2
|
N A T U R E
|
V O L 0 0 0
|
0 0 M O N T H 2 0 1 5
G
2015 Macmillan Publishers Limited. All rights reserved

LETTER RESEARCH
Figure 3
|
Substrate scope. a, Reactions of aryl
a
Bromides and triflates
R
and vinyl bromides with aryl and vinyl triflates.
b, Reactions of other aryl electrophiles. c, Relative
reactivity of the palladium and nickel catalysts. To
generate products 18, 23, 24 and 25, no KF was
used. For product 18, yield was increased from 64%
to 85% when three (rather than one) equivalents
of 3-bromothiophene were used. For product
19, yield was increased from 60% to 75% when
two equivalents of 2-bromocyclohexenone were
used. The reaction to produce product 3 was
replicated independently by three different
researchers in our group using the procedure given
in the Supplementary Information.
5 mol% NiBr2(diglyme), 5 mol% bpy
5 mol% PdCl2, 5 mol% dppp
R′
R
R′
Br
+
TfO
Zn (2 equiv.)
KF (1 equiv.)
DMF (2 ml), 40 °C
1 equiv.
1 equiv.
MeO
OMe
OMe
OMe
3
6
7
8
78% yield
85% yield
94% yield
87% yield
OMe
F
N
OMe
O
OMe
O
O
9
10
11
12
75% yield
61% yield
71% yield
63% yield
O
MeO
Cl
CN
13
14
15
16
53% yield
71% yield
71% yield
61% yield
O
O
17
59% yield
S
18
19
B
CN
64% yield
85% yield
60% yield
75% yield
O
b
Other aryl electrophiles
R
5 mol% NiBr2(diglyme), 5 mol% bpy
mol% PdCl2, 5 mol% dppp
R′
R
R′
5
Br
+
TfO
Zn (2 equiv.)
KF (1 equiv.)
DMF (2 ml), 40 °C
1 equiv.
1 equiv.
F3C
O
OMe
S
OMe
O
N
N
20
21
22
25
X = I, R= SO2CF3
47% yield
X = Cl, R= SO2CF3
75% yield
X = Cl, R= SO2CF3
74% yield
X = Cl, R= SO2CF3
61% yield
c
Relative reactivity of catalysts
OMe
N
OTf
Br
Cl
(bpy)Ni
I
Br
>
>
>
>
>
23
X = Cl, R = SO2CF3 59% yield
24
X = Cl, R = SO2C4F9 54% yield
(dppp)Pd
I
OTf
>
Cl
electronically similar and electronically disparate aryl bromides and strong affinity for naphthyl substrates that is not shared by the
triflates couple with good yields, but the coupling of a highly acti- palladium catalyst26. The higher yield of the coupled product
vated substrate with a deactivated substrate results in a low yield. obtained when using naphthyl chloride rather than naphthyl iodide
Aryl substrates with ortho-substitution can be coupled (Fig. 3a, 8 suggested the possibility of coupling other activated aryl chlorides
and 14), and the synthesis of 2,29-disubstituted biaryls and 2,29- with this system. Indeed, an aryl chloride with a sulfonyl group in
binaphthyls can also be achieved (Fig. 3a, 9 and 15). The tolerance the para position was successfully coupled with high selectivity to
of this multimetallic method to different functional groups—includ- give product 22. By contrast, reactions of unactivated aryl chlorides
ing amine, ether, ester, aldehyde, sulfone, and acetal groups—is formed large amounts of symmetric biaryl from the aryl triflate.
Bromothiophene and pyridyl halides and triflates coupled with reas-
onable yield, including the synthesis of 2,39-bipyridine (product 25).
Finally, the chemistry could be extended to the synthesis of diene
(product 19) from a-bromocyclohexenone and cyclohexenyl triflate.
To our knowledge, this is the first cross-Ullmann reaction to form a
diene product.
promising. In addition, boronic acid ester containing product 17
was formed without competing Suzuki coupling, despite the pres-
ence of palladium and KF, demonstrating complementarity with
established cross-coupling methods. Although the combination of
other halides and pseudohalides under these reaction conditions
resulted in lower yields (see Supplementary Information, Charts
S1–S3), we found several synthetically useful exceptions. While reac-
Although we have not yet conducted a detailed study of the mech-
tions with aryl iodides generally resulted in rapid formation of biaryl anism, current evidence is consistent with our original proposal
(see Supplementary Information, Chart S3; both palladium and (Fig. 1). We propose that, after the two arylmetal intermediates are
nickel react with aryl iodides more quickly than with aryl triflates), formed by oxidative addition, a mechanism analogous to the ‘persist-
naphthyl iodide could be coupled to form product 20 with 47% yield ent radical effect’ allows for the observed cross-selectivity in transme-
(Fig. 3b). Similarly, 2-naphthyl chloride could be coupled to form tallation27,28. In the persistent radical effect, one radical is unreactive
product 21 with 75% yield, suggesting that the nickel catalyst has a with itself and stable, while the other radical is highly reactive and
0 0 M O N T H 2 0 1 5
| V O L 0 0 0 | N A T U R E | 3
G
2015 Macmillan Publishers Limited. All rights reserved

RESEARCH LETTER
unselective. In our system, (dppp)Pd(Ar2) triflate is stable to further
reaction with itself (Fig. 2d) and accumulates in solution. By contrast,
(bpy)Ni(Ar1) bromide is a transient intermediate that quickly
reacts with itself (Fig. 2c) or with (dppp)Pd(Ar2) triflate (Fig. 2b).
Selectivity arises from the relative concentrations of these two com-
plexes in solution, similar to what occurs during multimetallic reac-
tions with group 11 organometallic intermediates, which are generally
slow to homocouple11. Consistent with our observations (Table 1), the
selectivity of this cross-Ullmann reaction is predicted to be independ-
ent of the amount of palladium and nickel added. Furthermore, our
proposed mechanism suggests that the formation of (bpy)Ni(Ar1)
bromide would limit the rate of product formation. Therefore, a higher
concentration of nickel, but not of palladium, should result in a higher
rate of product formation. We found this to be the case: reactions that
were run with a higher concentration of nickel were faster (24% yield
versus 7% yield at 4 hours for 25 mM versus 12.5 mM nickel), but the
concentrationof palladium didnot affect the rate of product formation
(11% yield versus 7% yield at 4 hours for 25 mM versus 12.5 mM
palladium). We are now working to further illuminate the mechanism
of this reaction and the origin of the cross-selectivity.
´
11. Perez-Temprano, M. H., Casares, J. A. & Espinet, P. Bimetallic catalysis using
transition and group 11 metals: an emerging tool for C–C coupling and other
reactions. Chem. Eur. J. 18, 1864–1884 (2012).
12. Ko, S., Kang, B. & Chang, S. Cooperative catalysis by Ru and Pd for the direct
coupling of a chelating aldehyde with iodoarenes or organostannanes. Angew.
Chem. Int. Edn 44, 455–457 (2005).
13. Ullmann, F. & Bielecki, J. Ueber Synthesen in der Biphenylreihe. Chem. Ber. 34,
2174–2185 (1901).
14. Fanta, P. E. The Ullmann synthesis of biaryls. Synthesis 1974, 9–21 (1974).
´
15. Hassan,J., Sevignon, M.,Gozzi, C., Schulz,E. & Lemaire, M. Aryl-arylbondformation
one century after the discovery of the Ullmann reaction. Chem. Rev. 102,
1359–1470 (2002).
16. Ackermann, L. Modern Arylation Methods (Wiley, 2009).
17. Stuart, D. R. & Fagnou, K. The catalytic cross-coupling of unactivated arenes.
Science 316, 1172–1175 (2007).
18. Weber, D. & Gagne, M. R. Pd(0)/Au(
catalyzed homo-coupling of Arylgold(
5172–5174 (2011).
19. Kamikawa, T. & Hayashi, T. Control of reactive site inpalladium-catalyzed Grignard
cross-coupling of arenes containing both bromide and triflate. Tetrahedr. Lett. 38,
7087–7090 (1997).
20. Everson, D. A., Jones, B. A. & Weix, D. J. Replacingconventionalcarbon nucleophiles
with electrophiles: nickel-catalyzed reductive alkylation of aryl bromides and
chlorides. J. Am. Chem. Soc. 134, 6146–6159 (2012).
21. Panteleev, J., Zhang, L. & Lautens, M. Domino rhodium-catalyzed alkyne arylation/
palladium-catalyzed N arylation: a mechanistic investigation. Angew. Chem. Int.
Edn 50, 9089–9092 (2011).
22. Gomes, P., Fillon, H., Gosmini, C., Labbe, E. & Perichon, J. Synthesis of
unsymmetrical biaryls by electroreductive cobalt-catalyzed cross-coupling of aryl
halides. Tetrahedron 58, 8417–8424 (2002).
23. Hatakeyama, T., Hashimoto, S., Ishizuka, K. & Nakamura, M. Highly selective biaryl
cross-coupling reactions between aryl halides and aryl Grignard reagents: a new
catalyst combination of N-heterocyclic carbenes and iron, cobalt, and nickel
fluorides. J. Am. Chem. Soc. 131, 11949–11963 (2009).
I) redox incompatibilities as revealed by Pd-
I)-complexes. Chem. Commun. 47,
After a century of developing strategies for the formation of
biaryls, a general method to selectively form unsymmetrical biaryls
directly from two aryl electrophiles has been realized. In addition to
the immediate utility of this transformation, the general principle of
selective transmetallation between a persistent catalytic intermediate
and a transient, reactive organometallic intermediate should spur the
development of new multimetallic reactions beyond the cross-Ullman
coupling, such as C–H arylation29.
´
´
24. Littke, A. F., Dai, C. & Fu, G. C. Versatile catalysts for the Suzuki cross-coupling of
arylboronic acids with aryl and vinyl halides and triflates under mild conditions.
J. Am. Chem. Soc. 122, 4020–4028 (2000).
25. Proutiere, F. & Schoenebeck, F. Solvent effect on palladium-catalyzed cross-
coupling reactions and implications on the active catalytic species. Angew. Chem.
Int. Edn 50, 8192–8195 (2011).
Online Content Methods, along with any additional Extended Data display items
andSource Data, are available in the onlineversion of the paper; references unique
to these sections appear only in the online paper.
26. Rosen, B. M. et al. Nickel-catalyzed cross-couplings involving carbon–oxygen
Received 18 February; accepted 10 June 2015.
Published online 17 August 2015.
bonds. Chem. Rev. 111, 1346–1416 (2011).
27. Studer, A. The persistent radical effect in organic synthesis. Chem. Eur. J. 7,
1159–1164 (2001).
28. Fischer, H. The persistent radical effect: a principle for selective radical reactions
and living radical polymerizations. Chem. Rev. 101, 3581–3610 (2001).
29. Durak, L. J. & Lewis, J. C. Iridium-promoted, palladium-catalyzed direct arylation of
unactivated arenes. Organometallics 33, 620–623 (2014).
1. Magano, J. & Dunetz, J. R. Large-scale applications of transition metal-catalyzed
couplings for the synthesis of pharmaceuticals. Chem. Rev. 111, 2177–2250
(2011).
2. Busacca, C. A., Fandrick, D. R., Song, J. J. & Senanayake, C. H. The growing impact of
catalysis in the pharmaceutical industry. Adv. Synth. Catal. 353, 1825–1864 (2011).
3. Roughley, S. D. & Jordan, A. M. The medicinal chemist’s toolbox: an analysis of
reactions used in the pursuit of drug candidates. J. Med. Chem. 54, 3451–3479
(2011).
30. Montgomery, J. in Organometallics in Synthesis: Fourth Manual (ed. Lipshutz, B. H.)
319–428 (John Wiley & Sons, 2013).
Supplementary Information is available in the online version of the paper.
4. Allen, A. E. & MacMillan, D. W. C. Synergistic catalysis: a powerful synthetic strategy
for new reaction development. Chem. Sci. 3, 633–658 (2012).
5. Shibasaki, M. & Yamamoto, Y. Multimetallic Catalysis in Organic Synthesis (Wiley,
2004).
6. Smidt, J. et al. Olefinoxydation mit Palladiumchlorid-Katalysatoren. Angew. Chem.
74, 93–102 (1962).
7. Takacs, J. & Jiang, X. The Wacker reaction and related alkene oxidation reactions.
Curr. Org. Chem. 7, 369–396 (2003).
8. Michel, B. W., Steffens, L. D. & Sigman, M. S. The Wacker oxidation. Org. React. 84,
75–414 (2014).
Acknowledgements This work was supported by the NIH (R01 GM097243). L.K.G.A
acknowledges both an NSF Graduate Fellowship (NSF DGE-1419118) and an Elon
Huntington Hooker Fellowship (Univ. Rochester). D.J.W is an Alfred P. Sloan Research
Fellow and a Camille Dreyfus Teacher Scholar. We thank Z. Melchor (Univ. Rochester)
for preliminary exploration of coupling electron-poor aryl halides, and S. Wu for
assistance with graphics.
Author Contributions D.J.W. conceived the idea. L.K.G.A. performed experiments and
analysed the results with assistance from M.M.L. Both D.J.W. and L.K.G.A. proposed the
experiments, discussed the data, and wrote the manuscript.
9. Sonogashira, K., Tohda, Y. & Hagihara, N. A convenient synthesis of acetylenes:
catalytic substitutions of acetylenic hydrogen with bromoalkenes, iodoarenes and
bromopyridines. Tetrahedr. Lett. 16, 4467–4470 (1975).
Author Information Reprints and permissions information is available at
www.nature.com/reprints. The authors declare no competing financial interests.
Readers are welcome to comment on the online version of the paper. Correspondence
and requests for materials should be addressed to D.J.W. (daniel.weix@rochester.edu).
´
10. Chinchilla, R. & Najera, C. The Sonogashira reaction: a booming methodology in
synthetic organic chemistry. Chem. Rev. 107, 874–922 (2007).
4
|
N A T U R E
|
V O L 0 0 0
|
0 0 M O N T H 2 0 1 5
G
2015 Macmillan Publishers Limited. All rights reserved

LETTER RESEARCH
(a)
(b)
OMe
OMe
10 mol% PdCl2, 10 mol% ligand
10 mol% PdCl2, 10 mol% dppp
OMe
OMe
Br
Br
10 mol% NiBr2•3H2O, 10 mol% bpy
10 mol% NiBr2•3H2O, 10 mol% ligand
+
+
Zn (2 equiv)
DMF (2 mL), 60 °C, 24 h
Zn (2 equiv)
DMF (2 mL), 60 °C, 24 h
TfO
TfO
1 equiv
1 equiv
1 equiv
1 equiv
100%
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
1
2
3
4
5
6
7
8
9
10 11 12 13
14
61
5
15
47
9
16
59
10
21
10
0
17
69
8
18
53
12
29
1
19
54
10
23
4
Product
Bianisole
Biphenyl
Anisole
Product
Bianisole
Biphenyl
Anisole
12 28 53 61 26 56 22 11 21
10 12
6
0
8
1
2
6
9
2
2
8
4
5
5
7
9
3
0
12
0
28 14 24 18 31 21 53 36 37 44 24
18
7
29
1
20
1
2
6
7
6
0
0
7
9
0
0
3
5
8
8
0
0
3
5
11 11
Benzene
Benzene
24 48
29 11
35 34
9
13 24 28
9
4
2
2
9
Aryl triflate
Aryl triflate
26
0
0
0
6
0
0
0
3
0
0
0
30 30 42 44
0
9
0
0
3
0
Bromobenzene
Bromobenzene
0
13 15 14
0
1
0
0
0
0
Ph2P
PPh2
Ph2P
PPh2
PPh2
Ph2P
PPh2
Ph2P
PPh2
MeO
OMe
L1
L2
L6
L3
L4
N
N
N
N
N
N
N
PPh2
PPh2
PPh2
O
PPh2
L14
L15
L18
L16
Ph2P
PPh2
PPh2
H
N
L5
L7
L8
N
N
N
N
N
N
PPh2
Fe
L17
L19
PPh2
Me2P
PMe2
Et2P
PEt2
(iPr)2P
P(iPr)2
(Cy)2P
P(Cy)2
L9
L10
L11
L12
L13
Extended Data Figure 1
catalysts. Several different phosphine ligands for palladium (a) and amine
|
Varying ligands on the nickel and palladium
different amine ligands were effective (b). Reactions conducted with
L15, L17, and L18 used 5 mol% catalyst loadings, were heated to 40 uC, and
ligands for nickel (b) were investigated. Although selectivity and yield of cross- were monitored for 64 h. GC, gas chromatography; area per cent, uncorrected.
product were sensitive to the identity of the phosphine ligand (a), a variety of
G
2015 Macmillan Publishers Limited. All rights reserved

RESEARCH LETTER
OMe
OMe
5 mol% PdCl2, 5 mol% dppp
5 mol% PdCl2, 5 mol% dppp
(a)
OMe
(c)
OMe
Br
Br
5 mol% NiBr2•3H2O, 5 mol% bpy
5 mol% NiBr2•3H2O, 5 mol% bpy
+
+
KF (X equiv), Zn (2 equiv)
Additive (1 equiv), Zn (2 equiv)
TfO
TfO
DMF (2 mL), 40 °C, 24 h
DMF (2 mL), 40 °C, 24 h
1 equiv
1 equiv
1 equiv
1 equiv
100%
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
0.20
0.0 equiv
0.5 equiv 1.0 equiv 2.0 equiv 3.0 equiv
equiv
37
1
No Salt
KF
79
8
LiF
66
10
15
0
CsF
27
19
9
KI
35
28
37
0
KBr
34
26
36
1
KOAc
Product
62
14
21
1
79
5
77
4
79
9
72
12
13
0
Product
62
14
21
1
46
24
29
0
Bianisole
Biphenyl
Anisole
Bianisole
Biphenyl
13
0
12
0
7
10
0
9
1
Anisole
0
5
Benzene
3
2
3
2
2
2
Benzene
3
3
4
30
11
0
1
1
1
Aryl triflate
0
28
20
1
9
0
0
Aryl triflate
Bromobenzene
0
0
3
0
2
0
Bromobenzene
0
0
0
0
0
0
0
0
0
0
0
PhBr
ArOTf
(b)
(d)
PhBr
ArOTf
Additive
No Salt
KF
t (h)
Remaining Remaining
Equiv KF
t (h)
Remaining Remaining
(% yield)
(% yield)
3
24
3
42
5
47
10
31
0
(% yield)
(% yield)
0
0.2
0.5
1
11
24
11
24
11
24
11
24
11
24
11
24
30
5
38
10
36
28
15
1
30
0
24
3
35
20
13
0
LiF
39
0
39
3
24
3
CsF
3
37
11
37
0
24
3
0
0
9
KI
5
0
0
24
3
0
2
1
0
KBr
0
23
2
0
0
24
3
0
3
0
0
KOAc
2
26
0
0
0
24
0
Extended Data Figure 2
|
The effect of potassium and fluoride salts on the palladium multimetallic system, we included various amounts of this
additive under standard reaction conditions. The resulting selectivity and rate
presence of potassium fluoride (KF) in cross-couplings has been demonstrated data are shown. c, d, Once the advantageous role of KF was confirmed,
to improve the product yield and rate of reactions while minimizing the other potassium and fluoride salts were tested. The resulting selectivity and rate
formation of dimeric products. This ‘KF effect’ can be attributed to a variety of data are shown. The presence of potassium resulted in faster reaction rates,
selectivity of the multimetal-catalysed cross-Ullmann reaction. The
different factors, including the formation of ‘ate’ complexes, the removal of a
halide from a metal complex, or the formation of a fluoride-bridged
complex, which could facilitate an oxidative addition or transmetallation
step23,25. a, b, To investigate whether KF could be beneficial for the nickel and
while the presence of fluoride reduced the amount of bianisole and biphenyl
byproducts, improving the yield of cross-product. (Reactions without additives
took 32 h to complete.)
G
2015 Macmillan Publishers Limited. All rights reserved