Pd-PEPPSI-IPent: Low-temperature negishi cross-coupling for the preparation of highly functionalized, tetra-ortho-substituted biaryls

Article abstract of DOI:10.1002/anie.200906811

(Chemical Equation Presented) Cool couplings: Complex, hindered biaryls have been prepared at temperatures ranging from 0°C to room temperature, or with gentle heating. The Pd-PEPPSI-IPent catalyst (see scheme) nicely couples starting materials containing acidic moieties and routinely prepares biaryl derivatives where one or both rings comprising the biaryl are heterocyclic. Ar¹=hindered aryl or heteroaryl, Ar²=unactivated aryl or heteroaryl.

Full text of DOI:10.1002/anie.200906811

Communications
DOI: 10.1002/anie.200906811
Synthetic Methods
Pd-PEPPSI-IPent: Low-Temperature Negishi Cross-Coupling for the
Preparation of Highly Functionalized, Tetra-ortho-Substituted
Biaryls**
Selꢀuk C¸ alimsiz, Mahmoud Sayah, Debasis Mallik, and Michael G. Organ*
Biaryls are important motifs seen in the structures of many
biologically active compounds and organic materials. Thus,
the development of efficient bond-forming procedures
between sp2-hybridized carbon atoms has been pursued
intensively by both academic and industrial scientists.[1] The
synthesis of biaryl compounds began with the century-old
Ullmann reaction[2] and has evolved into a wide variety of Ni-
and Pd-catalyzed cross-coupling procedures.[3] Organotin
(Stille–Migita) and organoboron (Suzuki–Miyaura) reagents
have been used most widely as the transmetalating partner in
cross-coupling reactions; organozincs (Negishi coupling) are
the most reactive partners, but have been employed to a lesser
extent owing primarily to their increased basicity and
moisture sensitivity.[4]
In 2001, Dai and Fu reported the first general protocol for
Pd-catalyzed Negishi coupling between aryl zinc reagents and
aryl/heteroaryl chlorides, and gave good yields using [Pd{P-
(tBu)3}2] (1; Scheme 1) in THF/NMP at 1008C.[5] In 2004,
Milne and Buchwald prepared a variety of sterically bulky
biaryls that contained heterocycles with many functional
groups by coupling aryl/heteroaryl halides with aryl zinc
reagents that were prepared in situ using the hindered biaryl
ligand 2 in conjunction with [Pd2(dba)3] at 708C.[6] In 2006,
our research group developed a user-friendly Negishi proto-
col capable of cross-coupling aryl zinc halides with aryl
bromides, chlorides, and triflates in excellent yields using Pd-
PEPPSI-IPr (4) under mild conditions.[7] Similarly, in 2008,
Knochel and co-workers reported a one-pot Negishi coupling
protocol employing 4 to form biaryls and heterobiaryls from
aryl/heteroaryl zinc reagents generated in situ and aryl
bromides, chlorides, or triflates under mild conditions.[8] In a
series of publications, Knochel and co-workers also reported
Negishi protocols for coupling zinc reagents with aryl halides
bearing relatively acidic hydrogen atoms using Pd(OAc)2 in
conjunction with Buchwaldꢀs SPhos ligand (3).[9]
Scheme 1. Previously employed catalysts in the Negishi cross-coupling
reaction and Pd-PEPPSI-IPent. Cy=cyclohexyl, dba=trans,trans-diben-
zylideneacetone.
sterically bulky aryl bromides/chlorides with aryl boronic
acids at 658C employing KOtBu/tBuOH.[10] Despite the
widespread use of boronic acids for Suzuki–Miyaura coupling
reactions, their tendency to form boroxines, frequent need for
recrystallization of the aryl boronic acids prior to use, and
competitive protodeboronation under coupling conditions
remains problematic.[11] Conversely, Negishi reactions are an
attractive alternative because they typically requires milder
reaction conditions and, although organozincs are quite basic,
they are highly tolerant of various functional groups.[12]
Herein, we investigate the use of 5 in Negishi cross-coupling
reactions for the synthesis of very challenging and structurally
diverse biaryl compounds.
The coupling of 2-mesitylzinc halide (7) with 2,6-
dimethyl-1-bromobenzene (8) was thought to be a good
substrate pairing to develop general coupling conditions.
Although these substrates have no functionality, this pair led
to the formation of a tetra-ortho-substituted biaryl that is still
beyond the capability of most catalysts. In a direct comparison
of catalysts 4 and 5 that was conducted at room temperature, 5
completed the reaction in about 4 hours, meanwhile 4 turned
over continuously, but at a much slower rate (Figure 1).
To shed some light on the performance of 4, and to
compare the reactivity of 5 with different ligand systems that
are known to be highly effective at cross-coupling (e.g. 2 and
3),[13] we systematically modified the model reaction to expose
the differences (Figure 2). Under the optimized reaction
conditions in THF/NMP (2:1),[7] catalyst 5 displayed 80%
Recently, we have shown Pd-PEPPSI-IPent (5) to be an
excellent catalyst for the Suzuki–Miyaura cross-coupling of
[*] Dr. S. C¸alimsiz, M. Sayah, Dr. D. Mallik, Prof. M. G. Organ
Department of Chemistry, York University
4700 Keele Street, Toronto, ON, M3J 1P3 (Canada)
Fax: (+1)416-736-5936
E-mail: organ@yorku.ca
Homepage: http://www.yorku.ca/organ/
[**] This work was supported by the NSERC (Canada) and the ORDCF
(Ontario).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200906811.
2014
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 2014 –2017

Angewandte
Chemie
Figure 1. Effect of reaction time on the model Negishi reaction utilizing
Pd-PEPPSI complexes 4 and 5. Conversions were determined by GC/MS
analysis against a calibrated internal standard (undecane). Reactions were
performed in duplicate. NMP=N-methylpyrrolidine, THF=tetrahydro-
furan.
conversion into the cross-coupled product 9 after 2.5 hours.
Conversely, when 4 was employed approximately 10% of 9
was observed, while ligand 2 or 3 (used in conjunction with
[Pd2(dba)3])[13b] produced only a trace amount of product.
Notably, unreacted aryl bromide 8 accounted for the entire
balance of the reaction mixture for all catalysts. Ligands 2 or 3
have been well engineered for highly effective reductive
elimination, but phosphines are not as electron-rich as N-
heterocyclic carbene (NHC) ligands; taken together, this
suggests that oxidative addition is rate limiting with 2 and 3.
This suggestion is supported by the observation that heating
to 708C, and with all else held constant, leads to complete
consumption of 8 with ligands 2 and 3. Interestingly, heating
also led to significant homocoupling of 8 (i.e. 10) with all
catalysts except 5, which still converted 95% of 8 into 9. Also,
based on the aryl zinc compound (i.e. 7), catalysts 2, 3, and 4
all provided approximately 40% of the homocoupled organo-
zinc product 11 (based on the consumption of 7, i.e. 2 equiv of
7 to make 1 equiv of 11), which is not accounted for on
Figure 2. Formation of the expected product 9 and significant
amounts of 11, suggests that a second transmetalation step
may be operative that competes with reductive elimination
leading to the homocoupled products, as was suggested
recently by Lei and co-workers.[14] Moreover, in the absence
of NMP, reactions performed at 708C were less effective with
all catalysts. Notably, significant disproportionation of 8,
which led to the formation of 10, was observed for the first
time using 5, as was the reduction of 8 (to provide 12) with
catalysts 2, 3, and 4. Even though there was again a significant
amount of 11 formed (ca. 40%) with 2, 3, and 4, there was not
enough of it formed to account for the lack of consumption of
the oxidative addition partner 8.
Figure 2. Catalyst, temperature, and solvent effects in the coupling of
mesitylzinc bromide (7) and 1-bromo-2,6-dimethylbenzene (8). Percent
conversion is based on 8 and determined by GC/MS analysis against a
calibrated internal standard (undecane). Reactions were performed in
duplicate.
reactive functional groups (Table 1). Under standard reaction
conditions using 5, aryl bromides/chlorides containing phe-
nols protected with alkyl, alkoxy, pinacol boronic ester, TBS,
acetyl, and benzyl groups were coupled efficiently at room
temperature or under mild heating. Acidic moieties including
anilines (19), phenols (20), alcohols (21), and amides (22)
were well tolerated. With a few exceptions, catalyst 4
provided considerably lower yields of cross-coupled products.
Remarkably, catalyst 5 was able to generate 90% of 9 and
80% of 13 when the reaction was run at 08C and allowed to
slowly warm to 68C.
In light of the importance of heterocyclic compounds,[15]
we examined the coupling between heteroaryl halides and
hindered aryl zinc reagents (Table 2). Varieties of heterocyclic
chlorides/bromides were coupled in excellent yields; these
included pyrazine (26), quinoline (29, 31), sterically bulky
isoxazole (23) and pyrazole (24), as well as substituted
pyrimidine (28), pyridazine (25), and pyridines (27, 30, 32).
We then focused on the coupling of heteroaryl zinc reagents
with aryl bromides/chlorides as well as heteroaryl bromides/
chlorides (Table 3). Catalyst 5 effectively coupled 2-pyridyl
(33–36), 4-isoquinolinyl (37, 38), 2-thiophenyl (39), 2-thia-
zolyl (40), and 5-ethoxycarbonyl-2-furyl (41–43) zinc reagents
with a variety of aryl- and heteroaryl halides at room
temperature or under mild heating.[16]
Following the promising initial results with Pd-PEPPSI-
IPent, 4 and 5 were evaluated in the Negishi coupling of aryl
zinc reagents (prepared in situ) with oxidative additions
partners bearing considerable steric bulk and/or various
Angew. Chem. Int. Ed. 2010, 49, 2014 –2017
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2015

Communications
Table 1: Negishi cross-couplings of aryl zinc reagents with aryl halides.
Table 2: Negishi cross-coupling reactions of aryl zinc reagents with
heteroaryl halides.
No.
X
T
t
Product
Yield [%] Yield [%]
[8C] [h]
with 4[a] with 5[a]
Entry
X
T
[8C]
Product
Yield
[%][a]
1
2
Cl 23
Cl 23
8
8
9
3
quant. (90)[b]
1
2
3
4
Br
Br
Cl
Cl
23
23
23
23
23
24
25
26
82
13
11
96 (80)[b]
quant.
quant.
quant.
3
Br 40 24 14
1
73
4
5
Cl 50 24 14
1
71
97
Br 23 16 15
14
5
Br
Br
Cl
Br
Cl
Cl
23
23
23
60
60
60
27
28
29
30
31
32
85
89
96
98
95
91
6
6
7
Br 23
4
16
63
30
87
80
7
Br 23 16 17
8
8
9
Cl 45 24 18
90
90
9
Br 50 24 19
1[c]
57[c]
10
18[d]
95[d]
[a] Yields of isolated products are the average of two runs. Boc=tert-
10 Br 50 24 20
butoxycarbonyl.
Pd-PEPPSI-IPent (5) has proven to be an excellent
catalyst for the Negishi cross-coupling procedure, and was
used to prepare an impressive array of biaryl and heterobiaryl
compounds bearing various functional groups and/or con-
gested steric bulk in excellent yields under the mildest
conditions yet reported for such difficult substrates. These
results, in addition to our earlier report about the perfor-
mance of 5 in Suzuki–Miyaura cross-couplings, reinforce the
notion that conformationally flexible steric bulk is intricately
linked to the performance of NHC-based palladium catalysts
in cross-coupling reactions.[10,17] We are currently studying the
origin of these effects. Further, the coupling that leads to the
11 Br 50 24 21
14[c]
82[c]
12 Br 40 24 22
60[d]
69[d]
[a] Yields of isolated products are the average of two runs. [b] Reaction was
performed at 0–68C with 1.6 equivalents of ZnCl2. [c] ArMgBr (2.6 equiv),
ZnCl2 (3 equiv). [d] NaH (1.0 equiv). Bn=benzyl, TBS=tert-butyldime-
thylsilyl.
2016 www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 2014 –2017

Angewandte
Chemie
Table 3: Negishi cross-couplings of heteroaryl zinc reagents with aryl
halides and heteroaryl halides.
Keywords: biaryls · cross-coupling · heteroaryls ·
Negishi reaction · palladium
.
[1] a) C. Ivica in Synthesis of Biaryls, Elsevier, Oxford, 2004; b) G.
Bringmann, C. Gꢁnther, M. Ochse, O. Schupp, S. Tasler in
Progress in Chemistry of Organic Natural Products, Vol. 82
(Eds.: W. Herz, H. Falk, G. W. Kirby, R. E. Moore), Springer,
Vienna, 2001; c) J. P. Corbet, G. Mignani, Chem. Rev. 2006, 106,
2651 – 2710.
[2] a) F. Ullmann, J. Bielecki, Ber. Dtsch. Chem. Ges. 1901, 34, 2174;
b) J. Hassan, M. Sꢂvignon, C. Gozzi, E. Schulz, M. Lemaire,
Chem. Rev. 2002, 102, 1359 – 1469.
Entry
1
X
T
[8C]
Product
Yield
[%][a]
Br
23
23
60
23
70
33
34
35
36
37
85
98
64
61
77
[3] A. de Meijere, F. Diederich in Metal-Catalyzed Cross-Coupling
Reactions, 2nd ed., Wiley-VCH, Weinheim, 2004.
2
3
4
5
Cl
Br
Br
Br
[4] Handbook of Organopalladium Chemistry for Organic Synthesis
(Ed.: E. I. Negishi), Wiley-Interscience, New York, 2002.
[5] C. Dai, G. C. Fu, J. Am. Chem. Soc. 2001, 123, 2719 – 2724.
[6] J. E. Milne, S. L. Buchwald, J. Am. Chem. Soc. 2004, 126, 13028 –
13032.
[7] M. G. Organ, S. Avola, I. Dubovyk, N. Hadei, E. A. B. Kantchev,
C. J. OꢀBrien, C. Valente, Chem. Eur. J. 2006, 12, 4749 – 4755.
[8] S. Sase, M. Jaric, A. Metzger, V. Malakhov, P. Knochel, J. Org.
Chem. 2008, 73, 7380 – 7382.
[9] a) G. Manolikakes, C. M. Hernandez, M. A. Schade, A. Metzger,
P. Knochel, J. Org. Chem. 2008, 73, 8422 – 8436; b) G. Manoli-
kakes, M. A. Schade, C. M. Hernandez, H. Mayr, P. Knochel,
Org. Lett. 2008, 10, 2765 – 2768; c) G. Manolikakes, J. Li, P.
Knochel, Synlett 2009, 681 – 686.
[10] M. G. Organ, S. C¸ alimsiz, M. Sayah, K. H. Hoi, A. J. Lough,
Angew. Chem. 2009, 121, 2419 – 2423; Angew. Chem. Int. Ed.
2009, 48, 2383 – 2387.
[11] a) M. Genov, A. Almorꢃn, P. Espinet, Chem. Eur. J. 2006, 12,
9346 – 9352; b) S. T. Handy, Y. Zhang, H. Bregman, J. Org.
Chem. 2004, 69, 2362 – 2366; c) H. Chaumeil, S. Signorella, C.
Le Drian, Tetrahedron 2000, 56, 9655 – 9662; d) T. Watanabe, N.
Miyaura, A. Suzuki, Synlett 1992, 207 – 210.
6
7
Br
Br
70
23
38
39
70
quant.[b]
8
9
Cl
Cl
60
23
40
41
91[c]
79
[12] a) E. Erdik, Organozinc Reagents in Organic Synthesis, CRC
Press, Boston, 1996; b) Organozinc Reagents,
A Practical
Approach, (Eds.: P. Knochel, P. Jones), Oxford, New York, 1999.
[13] a) To the best of our knowledge 2 is the most reactive,
commercially available ligand in the Buchwald family of steri-
cally crowded ligands for Negishi cross-coupling; b) To ensure
the formation of the active Pd complex formation when 2 or 3
were used in conjunction with [Pd2(dba)3], the reaction was
stirred in THF for 10 minutes before the addition of the
Grignard reagent; colour changes from purple to dark orange
signified formation of the complex. Reactions with 4 and 5 were
conducted under the same conditions.
10
11
Br
Br
23
23
42
43
quant.
91
[a] Yields of isolated products are the average of two runs. [b] Solvent
system: THF/NMP (2:1); [c] ZnCl2 (4.5 equiv).
[14] Q. Liu, Y. Lan, J. Liu, G. Li, Y. D. Wu, A. Lei, J. Am. Chem. Soc.
2009, 131, 10201 – 10210.
[15] Symposium abstract “Heterocyclic Chemistry: Synthesis Chem-
ical Biology and Drug Discovery” for the 42nd IUPAC Congress
August 2009, Glasgow, UK.
[16] When heteroaryl zinc reagents were used, the addition of
1.0 equivalent of ZnCl2 was beneficial. On the other hand, when
NMP was employed as the cosolvent, significant decreases in the
reaction rates were observed.
[17] G. Altenhoff, R. Goddard, C. W. Lehmann, F. Glorius, Angew.
Chem. 2003, 115, 3818 – 3821; Angew. Chem. Int. Ed. 2003, 42,
3690 – 3693.
formation of 9 and 13 at temperatures close to 08C indicates
that even milder reaction conditions are possible when using 5
to prepare complex biaryl motifs.
Received: December 3, 2009
Revised: January 16, 2010
Published online: February 16, 2010
Angew. Chem. Int. Ed. 2010, 49, 2014 –2017
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2017