Palladium-catalyzed synthesis of aromatic carboxylic acids with silacarboxylic acids

Article abstract of DOI:10.1021/ol4003465

Aryl iodides and bromides were easily converted to their corresponding aromatic carboxylic acids via a Pd-catalyzed carbonylation reaction using silacarboxylic acids as an in situ source of carbon monoxide. The reaction conditions were compatible with a wide range of functional groups, and with the aryl iodides, the carbonylation was complete within minutes. The method was adapted to the double and selective isotope labeling of tamibarotene.

Full text of DOI:10.1021/ol4003465

ORGANIC
LETTERS
2013
Vol. 15, No. 6
1378–1381
Palladium-Catalyzed Synthesis
of Aromatic Carboxylic Acids
with Silacarboxylic Acids
Stig D. Friis, Thomas L. Andersen, and Troels Skrydstrup*
Center for Insoluble Protein Structures (inSPIN), Interdisciplinary Nanoscience
Center (iNANO), Department of Chemistry, Aarhus University, Gustav Wieds Vej 14,
8000 Aarhus C, Denmark
ts@chem.au.dk
Received February 5, 2013
ABSTRACT
Aryl iodides and bromides were easily converted to their corresponding aromatic carboxylic acids via a Pd-catalyzed carbonylation reaction using
silacarboxylic acids as an in situ source of carbon monoxide. The reaction conditions were compatible with a wide range of functional groups, and
with the aryl iodides, the carbonylation was complete within minutes. The method was adapted to the double and selective isotope labeling of
tamibarotene.
Aromatic carboxylic acids and derivatives thereof are
important constituents of many bioactive compounds, and
hence methods for their synthesis under mild conditions
are in continuous demand.1 The Pd-catalyzed hydroxycar-
bonylation of aryl halides has emerged as a viable ap-
proach to such compounds,2 since it overcomes some of
the disadvantages of classical methodologies, which often
require harsh conditions or reactive intermediates that
may not be tolerant of or compatible with a number of
other functional groups.1c,3,4 The reaction also requires the
manipulation of a highly toxic gas, namely carbon mon-
oxide, and therefore methods that avoid the handling of
this gaseous reagent would be of high interest. One possi-
bility is to design reaction conditions that run with
substoichiometric CO, generating a reactive intermediate
that liberates the product and another molecule of CO.5
Another solution is to apply conditions that rely on the in
situ or ex situ generation of CO with carbon monoxide
releasing molecules (CORMs) that provide stoichiometric
amounts of CO. Such CORMs include Mo(CO)6,6 for-
mates,7 acid chlorides,8 and others.9
(1) (a) Kalgutkar, A. S.; Daniels, J. S. Chapter 3: Carboxylic Acids
and their Bioisosteres. In Metabolism, Pharmacokinetics and Toxicity of
Functional Groups: Impact of Chemical Building Blocks; Smith, D. A., Ed.;
RSC Publishing: Cambridge, 2010; p 99. (b) Maag, H. Prodrugs of
Carboxylic Acids. In Prodrugs; Stella, V., Borchardt, R., Hageman, M.,
Oliyai, R., Maag, H., Tilley, J., Eds.; Springer: New York, 2007; Vol. V, p
703. (c) Ogliaruso, M. A.; Wolfe, J. F. The synthesis of carboxylic acids
and esters and their derivatives. In The Chemistry of Acid Derivatives;
Patai, S., Ed.; John Wiley & Sons, Ltd.: 1979; p 267.
(2) For some recent reviews on Pd-catalyzed carbonylations, see: (a)
Wu, X.-F.; Beller Neumann, H.; Beller, M. Chem. Rev. 2013, 113, 1. (b)
Wu, X.-F.; Beller Neumann, H.; Beller, M. Chem. Soc. Rev. 2011, 40,
4986. (c) Magano, J.; Dunetz, J. R. Chem. Rev. 2011, 111, 2177. (d)
Grigg, R.; Mutton, S. P. Tetrahedron 2010, 66, 5515. (e) Brennfuhrer, A.;
Neumann, H.; Beller, M. Angew. Chem., Int. Ed. 2009, 48, 4114.
(3) (a) Roberts, G.; Shoppee, C. W. J. Chem. Soc. (Resumed) 1954,
3418. (b) Wnuk, S. F.; Chowdhury, S. M.; Garcia, P. I.; Robins, M. J.
J. Org. Chem. 2002, 67, 1816.
(4) For alternative routes to benzoic acids applying Ni-catalyzed
carboxylations of aryl halides with CO2, see: (a) Correa, A.; Martın, R.
J. Am. Chem. Soc. 2009, 131, 15974. (b) Fujihara, T.; Nogi, K.; Xu, T.;
Terao, J.; Tsuji, Y. J. Am. Chem. Soc. 2012, 134, 9106.
(5) Korsager, S.; Taaning, R. H.; Skrydstrup, T. J. Am. Chem. Soc.
2013, ASAP, DOI: 10.1021/ja3114032.
(6) For a recent review, see: Odell, L. R.; Russo, F.; Larhed, M.
Synlett 2012, 685.
(7) (a) Cacchi, S.; Fabrizi, G.; Goggiamani, A. Org. Lett. 2003, 5,
4269. (b) Lesma, G.; Sacchetti, A.; Silvani, A. Synthesis 2006, 594. (c)
Pri-Bar, I.; Buchman, O. J. Org. Chem. 1988, 53, 624.
r
10.1021/ol4003465
Published on Web 02/26/2013
2013 American Chemical Society

In this paper, we demonstrate that silacarboxylic acids
can operate as an efficient in situ CORM for the Pd-
catalyzed transformation of aryl iodides and bromides to
aromatic carboxylic acids. In particular, some of these reac-
tions are very fast and, in general, a high substrate scope is
demonstrated. Furthermore, we show that the methodology
can be adapted to the isotope labeling of benzoic acids.
In 2011, we published the application of crystalline
silacarboxylic acids as an efficient ex situ source of CO
in a number of Pd-catalyzed carbonylations with a two-
chamber system.10 In our search for new carbonylative coupl-
ing reactions, we recently attempted to utilize the silanoate
byproduct, produced via base-induced decarbonylation of
a silacarboxylic acid, as a coupling partner in a carbonylative
version of the HiyamaÀDenmark coupling to generate
benzophenone derivatives.11 This transformation was un-
successful in our hands, leading instead to the rapid genera-
tion of the corresponding aryl carboxylic acid 1 (Scheme 1).
the yield to 96% (entry 7),14 while changing to a Pd(II)
precatalyst provided poorer results (entries 8 and 9). Low-
ering the temperature to 40 °C with a slight increase in the
reaction time to 20 min also afforded an excellent yield of the
acid (entry 10), whereas a further decrease in the reaction
temperature or catalyst loading resulted in the incomplete
conversion of 4-iodoanisole to 1 (entries 11 and 12).
Table 1. Optimization Studies for the Pd-Catalyzed Carboxylic
Acid Synthesis from Aryl Iodidesa
time
yield
(%)b
entry
catalyst
ligand
solvent
(min)
1
Pd(dba)2
Pd(dba)2
Pd(dba)2
Pd(dba)2
Pd(dba)2
Pd(dba)2
Pd(dba)2
PdCl2
XantPhos
XantPhos
XantPhos
DPEphos
BINAP
THF
120
120
15
15
15
15
15
20
20
20
25
20
52
64
85
59
16
63
96
0
2
MeCN
Scheme 1. Initial Attempt Results
3
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
4
5
6c
PPh3
7d
8d
9d
10d,e
11d,f
12d,g
XantPhos
XantPhos
XantPhos
XantPhos
XantPhos
XantPhos
Pd(OAc)2
Pd(dba)2
Pd(dba)2
Pd(dba)2
80
95
85
60
Nevertheless, with the recognized usefulness of car-
boxylic acids and short reaction times observed, this result
encouraged us to optimize this undesired side reaction.
The carboxylation of 4-iodoanisole was found to pro-
ceed smoothly in nondried toluene using Pd(dba)2 and
XantPhos12 in the presence of the weakly alkaline potas-
sium trimethylsilanoate13 affording an 85% isolated yield
of carboxylic acid 1 after only 15 min at 50 °C (Table 1,
entry 3). Changing to other bidentate ligands (DPEphos,
BINAP) or PPh3 only retarded the reaction, and Xant-
Phos was therefore selected for further optimization. In-
creasing the amount of TMS-OK to 2.0 equiv boosted
a Reaction conditions: Under an ambient atmosphere, 4-iodoanisole
(0.5 mmol), catalyst (0.025 mmol), ligand (0.025 mmol), MePh2SiCO2H
(0.75 mmol), TMS-OK (0.75 mmol), and solvent (3.0 mL) were
mixed and heated to 50 °C in a closed 8 mL vial. b Isolated yield. c Ligand
(0.05 mmol). d TMS-OK (1.0 mmol). e 40 °C. f 30 °C. g Catalyst (0.010 mmol)
and ligand (0.010 mmol).
With the catalytic system at hand, operating at low
reaction temperatures, we set out to probe the scope and
limitations of this rapid carbonylation reaction. Initial
investigations were focused on the viability of a range of
different para-substituted aryl iodides as substrates for the
reaction. Gratifyingly, electron-deficient substrates also
worked providing compounds 2, 3, and 5À7 in excellent
yields (Scheme 2). Obtention of product 5 with a free
aldehyde demonstrates the mildness of the reactions con-
ditions. 4-Iodothioanisole and the Boc-protected iodoani-
line also led to high yields of the acids 8 and 10. On the
other hand, the transformation of 4-iodoaniline and 4-io-
dophenol failed under these conditions. High selectivity
was observed for substrates bearing either a bromide or a
boronic ester leading to compounds 9 and 11, respectively,
allowing for further carbon skeleton expansion with for
example the SuzukiÀMiyaura coupling.15
(8) (a) Hermange, P.; Lindhardt, A. T.; Taaning, R. H.; Bjerglund, K.;
Lupp, D.; Skrydstrup, T. J. Am. Chem. Soc. 2011, 133, 6061. (b) Taaning,
R. H.; Hermange, P.; Lindhardt, A. T.; Friis, S. D.; Skrydstrup, T. System
Providing Controlled Delivery of Gaseous CO for Carbonylation Reac-
tions. WO/2012/079583, 2011. (c) Hermange, P.; Gøgsig, T. M.; Lindhardt,
A. T.; Taaning, R. H.; Skrydstrup, T. Org. Lett. 2011, 13, 2444. (d) Nielsen,
D. U.; Taaning, R. H.; Lindhardt, A. T.; Gøgsig, T. M.; Skrydstrup, T.
Org. Lett. 2011, 13, 4454. (e) Xin, Z.; Gøgsig, T. M.; Lindhardt, A. T.;
Skrydstrup, T. Org. Lett. 2012, 14, 284.
(9) Fuji, K.; Morimoto, T.; Tsutsumi, K.; Kaiuchi, K. Angew. Chem.,
Int. Ed. 2003, 42, 2409.
(10) (a) Friis, S. D.; Taaning, R. H.; Lindhardt, A. T.; Skrydstrup, T.
J. Am. Chem. Soc. 2011, 133, 18114.
(11) Denmark, S. E.; Regens, C. S. Acc. Chem. Res. 2008, 41, 1486.
(12) Buchwald and co-workers demonstrated the usefulness of Xant-
Phos in Pd-catalyzed carbonylation reactions with a number of hetero-
atom nucleophiles. Martinelli, J. R.; Watson, D. A.; Freckmann,
D. M. M.; Barder, T. E.; Buchwald, S. L. J. Org. Chem. 2008, 73, 7102.
(13) Kagiya, T.; Sumida, Y. z.; Tachi, T. Bull. Chem. Soc. Jpn. 1970,
43, 3716.
3,4-Disubstituted benzoic acid 13 and 16 could be synthe-
sized in good-to-excellent yields, with no observation of de-
methylated byproducts. The conversion of ortho-substituted
aryl iodides was more sluggish and required a slight tem-
perature increase to 60 °C, in order to afford 14 and 15.
(14) Pressure measurement studies revealed that the reaction was
basically complete after 10 min (see Supporting Information).
Org. Lett., Vol. 15, No. 6, 2013
1379

With the di-o-fluoro-substituted aryl iodide, conversion
to the corresponding benzoic acid 17 provided a moderate
yield of 59%.
Table 2. Optimization of Reaction Conditions for the Pd-Cat-
alyzed Carboxylic Acid Synthesis from Aryl Bromidesa
Scheme 2. Scope of the Pd-Catalyzed Carboxylic Acid Synthesis
from Aryl and Hereroaryl Iodidesa
entry
solvent
toluene
activator
yield (%)b
1
TMS-OK
TMS-OLi
TMS-OK
TMS-ONa
TMS-OLi
TMS-OLi
TMS-OLi
TMS-OLi
TMS-OLi
42
18
43
37
87
56
5
2
toluene
3
dioxane
dioxane
dioxane
dioxane
dioxane
MeCN
4
5
6c
7c,d
8
45
39
9
2-methyl-THF
a Reaction conditions: In a glovebox under an argon atmosphere,
4-bromoanisole (0.5 mmol), Pd(dba)2 (0.025 mmol), XantPhos (0.025
mmol), MePh2SiCO2H (0.75 mmol), activator (1.0 mmol), and solvent
(3.0 mL) were mixed and heated to 80 °C for 19 h in a closed 8 mL vial.
b Isolated yield. c MePh2SiCO2H (0.85 mmol) and TMS-OLi (1.15 mmol).
d 60 °C.
With the conditions described in Table 2, entry 5, we
proceeded to examine the scope of the transformation of
organic bromides into carboxylic acids (Scheme 3). The
employment of substrates displaying electron-withdraw-
ing functionalities, including a nitro group in the para-
position, provided excellent yields of 2 and 7 from the cor-
a Reaction conditions: Under an ambient atmosphere, aryl/heteroaryl
iodode (0.5 mmol), Pd(dba)2 (0.025 mmol), XantPhos (0.025 mmol),
MePh2SiCO2H (0.75 mmol), TMS-OK (1.0 mmol), and toluene (3.0 mL)
were mixed and heated to 40 °C for 20 min in a closed 8 mL vial; isolated
yields given. b 0.063 M. c 60 °C. d 0.13 M. e 1 h.
Scheme 3. Scope of the Pd-Catalyzed Carboxylic Acid Synthesis
from Organic Bromidesa
Lastly, a selection of heteroaromatic substrates was ex-
amined, providing 18À20 in good yields. Purification of the
pyridine-containing compound 21 proved to be somewhat
cumbersome, explaining its lower isolated yields and con-
trasting the full conversionobservedby1H NMR analysis.16
We next turned our attention to the suitability of aryl
bromides for this transformation. Applying the optimized
conditions used in Scheme 2, albeit at 80 °C, provided a
42% yield of acid 1 after 19 h (Table 2, entry 1). Chang-
ing the solvent to dioxane produced a comparable yield
(entry 3), while exchanging the counterion of the activator
to lithium enhanced the yield of benzoic acid 1 to 87%
(entry 5). Increasing the amount of silacarboxylic acid and
lithium trimethylsilanoate did not furnish a better yield
(entry 6), whereas decreasing the reaction temperature to
60 °C almost shut down the reaction (entry 7). Switching to
acetonitrile or 2-methyltetrahydrofuran as the solvent led
to lower yields of 1 (entries 8 and 9).
a Reaction conditions: In a glovebox under an argon atmosphere,
organic bromide (0.5 mmol), Pd(dba)2 (0.025 mmol), XantPhos (0.025
mmol), MePh2SiCO2H (0.75 mmol), TMS-OLi (1.0 mmol), and dioxane
(3.0 mL) were mixed and heated to 80 °C for 16 h in a closed 8 mL vial;
isolated yields given. b 97% isolated yield when setup under an ambient
atmosphere with subsequent argon flushing. 1,3-Dibromobenzene
(0.25 mmol), Pd(dba)2 (0.013 mmol), XantPhos (0.013 mmol), 7%
3-bromobenzoic acid were observed.
(15) (a) Suzuki, A. J. Organomet. Chem. 1999, 576, 147. (b) Surry,
D. S.; Buchwald, S. L. Angew. Chem., Int. Ed. 2008, 47, 6338. (c) Yin, L.;
Liebscher, J. Chem. Rev. 2006, 107, 133. (d) Culkin, D. A.; Hartwig, J. F.
Acc. Chem. Res. 2003, 36, 234.
c
(16) Green, R. W.; Tong, H. K. J. Am. Chem. Soc. 1956, 78, 4896.
1380
Org. Lett., Vol. 15, No. 6, 2013

responding bromide.17 A tosylate and an amide were also
well tolerated providing 22 and 23 in yields of 99% and
77%, respectively. On the other hand, ethyl 4-bromobenzoate
yielded none of the desired product.
Scheme 5. Proposed Mechanism
Excellent yields were obtained in the synthesis of
m-trifluorobenzoic acid 24, the benzothiophenyl car-
boxylic acid 25, and diacid 27.18 Compatibility with a
terminal double bond was demonstrated by isolating 28 in
a 99% yield, with no detectable olefin migration. Finally, we
were pleased to observe that alternative substrates such as
a benzyl and vinyl bromide were also compatible with
these conditions. Hence, phenyl acetic acid 26 was iso-
lated in a 98% yield, whereas, with (Z)-2-bromobut-2-
ene, 97% of the R,β-unsaturated acid 29 could be isolated
with no observable change in stereochemistry.
The applicability of this chemistry was demonstrated in
the synthesis of the double 13C-labeled version of the
potent anticancer agent, tamibarotene.19 As illustrated in
Scheme 4, starting from 1-bromo-4-iodobenzene, the 13C-
carbonyl labeled benzoic acid 30 was synthesized in an
83% isolated yield using the 13C-labeled silacarboxylic acid.10
After amide formation, the bromide 31 was converted in an
excellent yield to double 13C-labeled tamibarotene 32.
take place by one of two pathways. Deprotonation and sub-
sequent 1,2-Brook rearrangement would generate CO and a
silanoate, which can either go on to release another molecule
of CO or participate as a nucleophile in the Pd-catalyzed
carbonylation.21 Alternatively, decarbonylation via SN2(Si)
with TMS-OMet as the nucleophile can also be envisaged,22
leading to CO and hydroxide formation. Equivalent to the
silanoate, this hydroxide ion can likely participate at several
different stages of the reaction. Subsequent CO insertion in-
to the PdÀC bond formed by oxidative addition, halideÀ
silanoate exchange at the Pd-metal center, and reductive
elimination would lead to the silylated benzoic acid. Release
of the carboxylate would then be achieved by siloxane
formation upon substitution with a silanoate nucleophile.
In summary, we have demonstrated how benzoic acids
can be synthesized from aryl and heteroaryl iodides and
bromides using palladium catalysis and by applying a
silacarboxylic acid as an in situ carbon monoxide source.
While displaying a wide substrate scope, the methodology
also proved to be convenient and high yielding. In parti-
cular, reaction times can be as short as a few minutes when
using aryl iodides, which suggests the possibility of apply-
ing this chemistry for the synthesis of 11C-labeled aromatic
carboxylic acids for potential PET investigations. Such
studies are ongoing and will be reported in due course.
Scheme 4. Double Carbonyl 13C-Isotope Labeling of Tamibar-
otene Employing MePh2Si13CO2Ha
a See Supporting Information for experimental details.
Acknowledgment. We are deeply appreciative of gener-
ous financial support from the Danish National Research
Foundation, the Danish Natural Science Research Coun-
cil, the Villum Foundation, the Carlsberg Foundation, the
Lundbeck Foundation, and Aarhus University.
A proposed mechanism for the conversion of aryl
bromides and iodides to benzoic acids utilizing the de-
scribed conditions is outlined in Scheme 5.20 Release of
carbon monoxide from the silacarboxylic acid can possibly
(17) We were delighted to find that it was not necessary to setup the
reaction in a glovebox under argon. This was demonstrated by the
excellent 97% isolated yield of 2 from an experiment whereby the flask
was just flushed with Ar.
Supporting Information Available. Experimental pro-
cedures and spectroscopic data. This material is available
free of charge via the Internet at http://pubs.acs.org.
(18) An additional 7% of the monoacid was observed.
(19) (a) Sanz, M. A.; Lo-Coco, F. J. Clin. Oncol. 2011, 29, 495. (b)
Miwako, I. Drugs Today (Barc.) 2007, 43, 563. (c) Tamibarotene is also
a candidate retinoid drug for Alzheimer’s disease. Holthoewer, D.;
Endres, K.; Schuck, F.; Hiemke, C.; Schmitt, U.; Fahrenholz, F.
Neurodegenerative Dis. 2012, 10, 224.
(21) Brook, A. G. Acc. Chem. Res. 1974, 7, 77.
(22) Igawa, K.; Kokan, N.; Tomooka, K. Angew. Chem., Int. Ed.
2010, 49, 728.
(20) (a) Lin, Y.-S.; Yamamoto, A. Organometallics 1998, 17, 3466.
(b) Barnard, C. F. J. Organometallics 2008, 27, 5402.
The authors declare no competing financial interest.
Org. Lett., Vol. 15, No. 6, 2013
1381