then, this transformation has been rarely used and no progress
has been made in regard to its utility.
conversion of 1 within 30 min, indicating that 1 is a
promising CO source for Pd-catalyzed fluorocarbonyla-
tion, in which KF is expected to serve a dual role not only
as a nucleophile for Pd-catalyzed carbonylation but also as
an activator for CO generation from 1.
Because the highly toxic nature of CO gas is the major
drawback of carbonylation reactions, CO-free carbonyla-
tion has been extensively researched. In the past decade in
particular, several compounds have been developed as
alternatives to toxic CO gas.5 Although successful hydro-
xy-,6 alkoxy-,7 and aminocarbonylation8 have been re-
ported, fluorocarbonylation of aryl halides without the
direct use of CO gas is not yet known.
Scheme 1. Decarbonylation of 1 with KFa
Recently, we reported novel and practical methods for
Pd-catalyzed alkoxycarbonylation and reductive carbony-
lation using a small excess amount (1.2À2.0 equiv) of a CO
source such asphenyl formates7aÀc and N-formylsaccharin
(1).9 In this approach, the decarbonylation of the CO
source using a mild base (e.g., NEt3 and Na2CO3) gener-
ates CO, which is then used in situ for the Pd-catalyzed
carbonylation to afford the corresponding phenyl esters or
aldehydes. Thus, this procedure provides a convenient
access to esters and aldehydes without concerns about
handling toxic CO gas. In particular, N-formylsaccharin
(1)10 has significant advantages such as low cost, good
availability, ease of handling, stability, and high reactivity
as a CO source. Owing to the low nucleophilicity of
saccharin generated from 1, the Pd-catalyzed carbonyla-
tions with various nucleophiles are expected to proceed
without any interference from the saccharin species. Our
continued interest in carbonylation reactions, particularly
in the conversion of aryl halides into the corresponding
acyl fluorides, prompted us to explore the Pd-catalyzed
fluorocarbonylation using 1 as the CO source.
a Reactions were conducted with 0.18 mmol of 1 and DMF-d7
(0.75 mL). b Conversion was determined using 1H NMR spectroscopy.
Encouraged by the above-mentioned results, the reac-
tion of bromobenzene (2a) with1.5 equiv of 1 and 2.5 equiv
of KF was carried out in the presence of a catalytic amount
of Pd(OAc)2 and ligand (P/Pd = 3) (Table 1). The use of a
monodentate phosphine ligand such as PPh3 and P(o-Tol)3
resulted in either low yield or none at all of the desired acyl
fluoride 3a (entries 1 and 2). Bidentate ligands such as DPPM,
DPPE, DPPP, DPPB, DPPF, DPPBz, and rac-BINAP were
ineffective for this transformation (entries 3À9). To our
delight, the use of a bulky monodentate phosphine,
P(t-Bu)3, and a bidentate ligand with bite angles of
102°À111°, such as DPEphos and Xantphos,11 improved
the catalytic activity (entries 10À12). Among the ligands
examined, Xantphos provided the best result, affording 3a
in 83% yield (entry 12).
Herein, we report a novel and practical method for Pd-
catalyzed fluorocarbonylation of aryl and alkenyl halides
using 1, which is an easily accessible and highly reactive
crystalline CO surrogate. The reactions using a near-
stoichiometric amount of the CO source (1.2 equiv) pro-
ceeded well and tolerated diverse functional groups. More-
over, the acyl fluorides obtained could be readily transformed
into various carboxylic acid derivatives such as carboxylic
acid, esters, thioesters, and amides in a one-pot procedure.
First, the decarbonylation of 1 with KF to generate CO
and saccharin was investigated (Scheme 1). Rapid decar-
bonylation was observed at 60 °C, leading to the complete
Table 1. Pd-Catalyzed Fluorocarbonylation of 2a with 1a
entry
ligand
PPh3
convb (%)
yieldb (%)
1
4
1
2
P(o-Tol)3
DPPM
DPPE
6
0
(6) (a) Cacchi, S.; Fabrizi, G.; Goggiamani, A. Org. Lett. 2003, 5,
4269. (b) Korsager, S.; Taaning, R. H.; Skrydstrup, T. J. Am. Chem. Soc.
2013, 135, 2891. (c) Berger, P.; Bessmernykh, A.; Caille, J.; Mignonac, S.
Synthesis 2006, 18, 3106.
3
0
0
4
1
0
5
DPPP
6
1
6
DPPB
10
14
0
3
(7) Recent examples of alkoxycarbonylation without using CO gas:
(a) Ueda, T.; Konishi, H.; Manabe, K. Org. Lett. 2012, 14, 3100. (b)
Ueda, T.; Konishi, H.; Manabe, K. Tetrahedron Lett. 2012, 53, 5171. (c)
Ueda, T.; Konishi, H.; Manabe, K. Org. Lett. 2012, 14, 5370. (d) Shang,
R.; Fu, Y.; Li, J.; Zhang, S.; Guo, Q.; Liu, L. J. Am. Chem. Soc. 2009,
131, 5738. (e) Fujihara, T.; Hosoki, T.; Katafuchi, Y.; Iwai, T.; Terao, J.;
Tsuji, Y. Chem. Commun. 2012, 48, 8012. (f) Schareina, T.; Zapf, A.;
Cotte, A.; Gotta, M.; Beller, M. Adv. Synth. Catal. 2010, 352, 1205.
(8) For a review on metal-catalyzed aminocarbonylation, see: Roy,
S.; Roy, S.; Gribble, G. W. Tetrahedron 2012, 68, 9867.
7
DPPF
4
8
DPPBz
rac-BINAP
0
9
20
57
55
86
12
42
51
83
10
11
12
P(t-Bu)3 HBF4
3
DPEphos
Xantphos
a Reactions were conducted with 0.6 mmol of 2a and anhydrous
DMF (2 mL). b Determined using GC.
(9) Ueda, T.; Konishi, H.; Manabe, K. Angew. Chem., Int. Ed. 2013,
52, 8611.
(10) Cossy et al. developed N-formylsaccharin as a new formylating
agent for amines: Cochet, T.; Bellosta, V.; Greiner, A.; Roche, D.;
Cossy, J. Synlett 2011, 13, 1920. N-Formylsaccharin is commercially
available from Tokyo Chemical Industry Co., Ltd. (TCI).
Further optimization of the reaction using Xantphos as
the ligand was conducted (Table 2). The P/Pd ratio was
B
Org. Lett., Vol. XX, No. XX, XXXX