dihaloaromatics with alkyl nucleophiles employing iron
catalysis. Although the selective functionalization of symme-
trical dihaloaromatics using iron has been demonstrated,11
the selective substitution of dihalogenated heterocycles
with electronically or sterically dissimilar CꢀX bonds has
been limited to the pyrimidine and purine nuclei.12 Whether
chemoselectivity could be achieved with a broader range of
dihaloaromatics such as pyridones, pyridines, quinolines,
isoquinolines, and carbocycles remained unanswered. Addi-
tionally, for substrates containing electronically similar CꢀX
bonds, we investigated if factors beyond bond distortion
energy may direct the selectivity in such transformations.
Table 1. Selected Optmization Studies
entrya
catalyst
solvent
THF
temp
yieldb
1
none
rt
0%
2
15 mol % MnBr2
15 mol % Mn(acac)2
15 mol % Mn(acac)3
15 mol % Cu2O
THF
rt
4%
3
THF
rt
3%
4
THF
rt
2%
€
Extensive investigations by Furstner et al. toward
5
THF
rt
1%
Fe-catalyzed cross-couplings between haloaromatics and
alkyl magnesium reagents revealed that Grignards contain-
ing higher alkyls (those possessing β-hydrogens) were essen-
tial for efficient alkyl transfer.13 This is in contrast to vinyl
bromides where incorporation of a methyl group can be
achieved.7a,8 Despite this difference, we postulated that diha-
loaromatics possessing lower LUMO energies could prove to
be more reactive substrates for selective incorporation of this
moiety. The evaluation of this hypothesis is reported herein.
Evaluation of the cross-coupling reaction on model
substrate 1a (Table 1) revealed that the coupling between
MeMgBr and 3,5-dibromopyridone in the absence of a
catalyst or in the presence of selected copper, manganese,
palladium, nickel, or cobalt salts resulted in the formation
of pyridone 2a in low yield (Table 1, entries 1ꢀ11). In
contrast, incorporation of FeCl3 or Fe(acac)3 provided
significantly higher yields of the desired product 2a (entries
12 and 13). Further evaluation of the reaction was con-
ducted with Fe(acac)3 due to its relative ease of handling.14
Lowering the catalyst loading to 3.75 mol % furnished
pyridone 2a in 76% yield (entry 14). Previous reports of
Fe-catalyzed cross-coupling reactions demonstrated that the
6
15 mol % CuI
THF
rt
2%
7
15 mol % Co(acac)3
THF
rt
8%
8
15 mol % NiCl2 DME
THF
rt
26%
13%
30%
9%
3
9
15 mol % Ni(dppf)Cl2
15 mol % Ni(PPh3)2Cl2
15 mol % Pd(OAc)2
15 mol % FeCl3
THF
rt
10
11
12
13
14
15
16
17
18
THF
rt
THF
rt
THF
rt
52%
56%
76%
44%
75%
44%
82%c
15 mol % Fe(acac)3
3.75 mol % Fe(acac)3
3.75 mol % Fe(acac)3
3.75 mol % Fe(acac)3
3.75 mol % Fe(acac)3
3.75 mol % Fe(acac)3
THF
rt
THF
rt
THF/NMP
THF
rt
0 °C
0 °C
rt
THF/NMP
THF
a All reactions were conducted on a 0.75 mmol scale using 1.0 equiv of
MeMgBr in THF at a concentration of 0.15 M where the Grignard reagent
was added via syringe pump over 7.5 min. b Unless otherwise indicated,
HPLC determined assay yields. c Isolated yield using 1.15 equiv of MeMgBr.
cosolvent N-methylpyrrolidone (NMP) can provide higher
yields.15 Incorporation of this cosolvent revealed that at 0 °C
and rt, the inclusion of NMP lowered the efficiency of the
transformation (compare entries 14ꢀ17). An increase in the
quantity of Grignard to 1.15 equiv led to the isolation of
the desired product 2a in 82% yield (entry 18).16
Selective incorporation of a methyl group on dihaloaro-
matics was then evaluated with a broader range of electro-
philes (Table 2). In THF alone, dichloroisoquinolines 1b
and 1c provided products in high selectivity with a pre-
ference for substitution at C-1 in 91% and 92% yield,
respectively (entries 2 and 3). Dichloropyridine 1d required
the use of 4:1 THF/NMP to obtain the desired product in
synthetically useful yields. Substitution at C-2 was favored
affording compound 2d in 77% yield (entry 4). 4,7-
Dichloroquinoline 1e and 2,6-dichlorobenzothiazole 1f,
both substrates that possess a halogen on separate rings,
led to selective alkyl transfer at the more electron-deficient
ring to afford 2e in 96% yield and 2f in 73% yield (entries 5
and 7). Of particular note, the reaction with 1e was
successful on 20 mmol scale (entry 6, 93% yield), demon-
strating the practicality of this methodology.
(6) (a) Garcia, Y.; Schoenebeck, F.; Legault, C. Y.; Merlic, C. A.;
Houk, K. N. J. Am. Chem. Soc. 2009, 131, 6632. (b) Legault, C. Y.;
Garcia, Y.; Merlic, C. A.; Houk, K. N. J. Am. Chem. Soc. 2007, 129,
12664. (c) Fairlamb, I. J. S. Chem. Soc. Rev. 2007, 36, 1036. For
prediction of selectivity in Pd-catalyzed cross-couplings with di- and
trihaloaromatics using NMR spectroscopy, see: (d) Handy, S. T.;
Zhang, Y. Chem. Commun. 2006, 299.
(7) (a) Tamura, M.; Kochi, J. K. J. Am. Chem. Soc. 1971, 93, 1487. (b)
Neumann, S. M.; Kochi, J. K. J. Org. Chem. 1975, 40, 599. (c) Smith,
R. S.; Kochi, J. K. J. Org. Chem. 1976, 41, 502.
€
(8) (a) Kauffman, T. Angew. Chem., Int. Ed. 1996, 35, 386. (b) Furstner,
A.; Krause, H.; Lehmann, C. E. Angew. Chem., Int. Ed. 2006, 45, 440.
€
(9) (a) Furstner, A.; Leitner, A. Angew. Chem., Int. Ed. 2002, 41, 609.
€
For a leading review by these authors, see: (b) Sherry, B. D.; Furstner, A.
Acc. Chem. Res. 2008, 41, 1500.
(10) (a) Czaplik, W. M.; Mayer, M.; von Wangelin, A. J. Angew.
Chem., Int. Ed. 2009, 48, 607. For a leading review by these authors, see:
ꢀ
(b) Czaplik, W. M.; Mayer, M.; Cvengros, J.; von Wangelin, A. J.
ChemSusChem 2009, 2, 396.
€
ꢁ
(11) Furstner, A.; Leitner, A.; Mendez, M.; Krause, H. J. Am. Chem.
Soc. 2002, 124, 13856.
(12) (a) Hocek, M.; Dvorakova, H. J. Org. Chem. 2003, 68, 5773. (b)
ꢀꢁ
ꢁ
€
ꢁꢀ
Scheiper, B.; Bonnekessel, M.; Krause, H.; Furstner, A. J. Org. Chem.
^
ꢀ
ꢁ
2004, 69, 3943. (c) Cerna, I.; Pohl, R.; Klepetarova, B.; Hocek, M. Org.
(15) (a) Cahiez, G.; Avedissian, H. Synthesis 1998, 1199. (b) Kleimark,
ꢀ
ꢁ
ꢁꢀ
ꢁ
nska, L.; Klepetarova, B.;
Lett. 2006, 8, 5389. (d) Kubelka, T.; Slavetı
Hocek, M. Eur. J. Org. Chem. 2010, 2666.
´
€
J.; Hedstrom, A.; Larsson, P.-F.; Johansson, C.; Norrby, P.-O. Chem-
CatChem 2009, 1, 152.
€
(13) Furstner, A.; Martin, R.; Krause, H.; Seidel, G.; Goddard, R.;
(16) When excess Grignard was added, the LC/MS trace suggested
the formation of dimethylated and dimeric products. However, these
products were not isolated and characterized in the present study.
Meanwhile, iron salts from three different suppliers gave similar results
in the cross-coupling reaction between MeMgBr and pyridone 1a.
Lehmann, C. W. J. Am. Chem. Soc. 2008, 130, 8773.
(14) Lots from three different suppliers provided equivalent results.
For trace metal analysis of the lot used for the majority of experiments
presented in this work, see Supporting Information.
Org. Lett., Vol. 15, No. 14, 2013
3699