Electrosynthesis of aromatic aldehydes by palladium-catalyzed carbonylation of aryl iodides in the presence of formic acid

Article abstract of DOI:10.1002/(SICI)1099-0690(199906)1999:6<1471::AID-EJOC1471>3.0.CO;2-E

The palladium-catalyzed electrocarbonylation of aryl halides performed in the presence of formic acid under one atmosphere of carbon monoxide affords aromatic aldehydes in good to high yields.

Full text of DOI:10.1002/(SICI)1099-0690(199906)1999:6<1471::AID-EJOC1471>3.0.CO;2-E

FULL PAPER
Electrosynthesis of Aromatic Aldehydes by Palladium-Catalyzed
Carbonylation of Aryl Iodides in the Presence of Formic Acid
[
a]
[a]
[b]
[b]
[c]
Italo Carelli,* Isabella Chiarotto, Sandro Cacchi, Paola Pace, Christian Amatore,
[c]
[c]
Anny Jutand, and Gilbert Meyer
Keywords: Palladium / Homogeneous catalysis / Carbon monoxide / Aromatic aldehyde / Formic acid / Carbonylation /
Electrochemistry
The palladium-catalyzed electrocarbonylation of aryl halides
performed in the presence of formic acid under one
atmosphere of carbon monoxide affords aromatic aldehydes
in good to high yields.
It has been recently reported that vinyl- and aryl halides protonation step is faster than its reaction with CO to af-
or -triflates react under electroreducing conditions in the ford the corresponding acylpalladium complex (vide infra).
[
1][2]
presence of palladium catalysts to produce biaryls,
aro- A dramatic change occurred when performing the reaction
[
1][3]
[4]
matic
and α,β-unsaturated carboxylic acids, alkenes, at a less negative potential in the presence of formic acid
[
5]
and arenes. As part of our ongoing program devoted to (Eq. 2, Table 1).
the development of new applications of the palladium-cata-
lyzed electrosynthesis, we now report our preliminary re-
sults on the preparation of aromatic aldehydes from aryl
(
2)
halides and carbon monoxide under electrochemical reduc- Table 1. Palladium-catalyzed electrosynthesis of aromatic aldehydes
from aryl iodides and carbon monoxide in the presence of formic
acid (Eq. 2)
ing conditions.
The synthesis of aryl aldehydes from aryl halides and CO
is formally a hydride transfer.[ We have reported that un-
der electrochemical reduction (Eq. 1), the reactivity of aryl
halides is reversed since the formation of aryl aldehydes re-
sults from a proton transfer. nBu NHSO and HClO were
6]
4
4
4
used as proton sources.[
7]
(
1)
The initial concept was based on electrochemical re-
duction of acylpalladium intermediates, [ArCOPdI], gener-
ated in situ from aryl iodides, palladium(0), and carbon
monoxide, followed by the protonation of the resultant
[
7]
anionic species. However, aromatic aldehydes were ob-
[a]
Electrolysis potential vs. SCE. Ϫ [b] Relative to initial aryl iodide.
For example, when the electrochemical reduction of 4-
tained in low yield.[ For example, the electrolysis at
7]
Ϫ2
Ϫ3
Ϫ1.7 V vs. SCE of 4-iodoanisole (10 mol dm in DMF
Ϫ3
iodoanisole was carried out at Ϫ1.5 V vs. SCE under one
atmosphere of CO in the presence of 10 mol-% of
containing nBu NBF 0.2 mol dm ) in the presence of 10
4
4
mol-% of PdCl (Ph P) , 1.5 equiv. of nBu NHSO and one
2
3
2
4
4
[
PdCl (Ph P) ] and 2 equiv. of HCO H, p-anisaldehyde was
2 3 2 2
atmosphere of CO at 22°C gave anisole as the major prod-
uct (46% yield), together with minor amount (14%) of the
desired p-anisaldehyde. Apparently, when the potential is
too negative, the reduction of the arylpalladium(II) com-
plex (formed initially by the oxidative insertion of pal-
ladium(0) into the carbonϪhalogen bond) followed by a
isolated in 78% yield. These conditions were applied to a
variety of aryl iodides and the preparative results are sum-
marized in Table 1.[ Aryl iodides bearing electron-donat-
ing substituents gave the best results, with a slight decrease
of the yields when they contain ortho-substituents (Table 1,
compare entries 2 and 3). Lower yields were obtained with
aryl iodides containing electron-withdrawing substituents
(Table 1, entry 9), most probably because of competitive
overreduction. This view seems to be supported by the cy-
clic voltammetry of p-EtOCOϪC H ϪCHO, which shows
8]
[
[
[
a]
b]
c]
Universita degli Studi “La Sapienza”, Dipartimento di Ingegne-
ria Chimica dei Materiali delle Materie Prime e Metallurgia
7
, via Castro Laurenziano, I-00185 Rome, Italy
Universita degli Studi “La Sapienza”, Dipartmento di Studi di
Chimica e Tecnologia delle Sostanze Biologicamente Attive
6
4
5
, P. le A. Moro, I-00185 Rome, Italy
that it is more easily reduced than the starting aryl iodide.
The reduction of [PdCl (Ph P) ] at the very beginning of
the electrolysis affords a palladium(0) complex which acti-
Ecole Normale Sup e´ rieure, D e´ partement de Chimie, CNRS
UMR 8640
2
3
2
24 Rue Lhomond, F-75231 Paris Cedex 5, France
Eur. J. Org. Chem. 1999, 1471Ϫ1473

WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
1434Ϫ193X/99/0606Ϫ1471 $ 17.50ϩ.50/0
1471

I. Carelli, I. Chiarotto, S. Cacchi, P. Pace, C. Amatore, A. Jutand, G. Meyer
FULL PAPER
vates aryl iodides by an oxidative addition. Kinetic investi-
gation showed that the rate of the oxidative addition of
0
[
Pd (Ph P) ] with PhI was considerably slower in the pres-
3
4
Ϫ3
ence of CO (e.g.: for [PhI] ϭ 20 mmol dm , t1/2 ϭ 2 s in
the absence of CO; t1/2 ϭ 310 s in the presence of CO at
atmospheric pressure). This fact suggests that the nature of
the usual reactive species in the oxidative addition, viz.
0
[
Pd (Ph P) ], changes in the presence of carbon monoxide,
3
2
very likely because of the formation of a palladium(0) com-
plex ligated to carbon monoxide (Eq. 3).
(
3)
Thus the oxidative addition overall rate decreases either
0
because [Pd (Ph P) ] is less available, or because
3
2
0
[
Pd (Ph P) (CO)] is less reactive owing to its lower nucleo-
3 2
Scheme 1
philicity, or both. This could not be kinetically discrimi-
nated because the too low solubility of CO at atmospheric
pressure in DMF (ca. 3Ϫ4 mmol dm ) prevented any fine
Ϫ3
quantitative investigation of the kinetics of the oxidative ad-
dition. Indeed, our amperometric method used to monitor
(
4)
0
the kinetics requires concentrations of [Pd (Ph P) ] of at
3
4
Ϫ3
least 2 mmol dm to allow precise currents measurements.
Therefore, at this stage, we cannot decide whether the CO
group is already present on the palladium(0) center when
supplying the reaction medium with formate ions. This
prompted us to investigate the role of palladium(0) com-
plexes in the generation of formate ions from formic acid.
the oxidative addition takes place or the oxidative addition
0
to [Pd (Ph P) ] precedes the coordination of carbon mon-
0
3
2
[Pd (Ph P) ] was found to react with formic acid in DMF
3 4
oxide. Whatever the real mechanism may be, it remains that
acylpalladium(II) intermediates are formed in the pal-
ladium-catalyzed reaction of aryl iodides with carbon mon-
at room temperature. The reaction was monitored by con-
ductimetry as well as by cyclic voltammetry and P-NMR
spectroscopy. Conductivity increased with time so as to
reach a plateau whose value depended on the amount of
formic acid (Figure 1). Thus ionic species were formed by
this reaction which is an equilibrium.
31
[
7]
oxide. It is worthwhile to note that the electrolysis poten-
tials are less negative that the reduction potential of acylpal-
ladium(II) complexes.[ Consequently, these complexes
cannot be reduced to anionic species as it was established
in our previous work (Eq. 1).[ Under the present con-
ditions, they can be converted into the corresponding alde-
hydes only through the intermediacy of σ-acylϪσ-formate
palladium(II) complexes formed by the nucleophilic attack
7]
7]
[6]
of formate on the acylpalladium(II) complexes (Scheme
[6]
1
). Subsequent elimination of CO2, followed by the re-
ductive elimination of palladium(0) species from the result-
ant acylpalladium hydride affords the aldehyde and regener-
ates the palladium(0) complex.
However, the dissociation constant of formic acid in
DMF is too low to enable it to be the direct source of for-
mate ions. Accordingly, 4-iodoanisole was recovered in 90%
yield when it was treated in DMF (containing nBu NBF
Figure 1. Conductance measurements of solutions of [Pd(Ph
3
P)
4
]
Ϫ3
(
(
2 mmol dm ) and n equivalents of HCO
O) n ϭ 1; (ϩ) n ϭ 100
2
H in DMF, at 20 °C.
4
4
Ϫ3
0
%
.2 mol dm ) for 24 h at 60 °C with formic acid, 10 mol-
0
of [Pd (PPh ) ] under one atmosphere of carbon monox-
Concomitantly, cyclic voltammetry indicated that a re-
ducible palladium(II) complex was formed (E ϭ Ϫ0.99 V
3
4
ide. p-Anisaldehyde was obtained in only 6% yield.
p
[9a]
Nor could formic acid be the indirect source of formate vs. SCE, compare Figures 2aϪc). The same species could
ions by electrochemical reduction of its protons. Indeed, the be characterized by its 31P-NMR signal at δ ϭ 24.27 vs.
[9b]
electrolyses were carried out at a potential of Ϫ1.5 V, H PO (external standard).
Since the equilibrium con-
3
4
whereas the electrochemical reduction of formic acid under centrations of the ionic species were lower in the presence
0
the same conditions, in the absence of palladium, occurred of excess PPh , this suggests a reaction between Pd (Ph P)
3
3
3
at Ϫ2.0 V (Eq. 4).
and formic acid presumably via the low ligated complex
0
Therefore, it is apparent that under our electrolytic con- [Pd (Ph P) ] to afford a formate anion and a cationic pal-
3
2
ditions, the palladium must play a second crucial role in ladium(II) hydride (Eq. 5). The equilibrium constant K has
1472
Eur. J. Org. Chem. 1999, 1471Ϫ1473

Electrosynthesis of Aromatic Aldehydes by Palladium-Catalyzed Carbonylation
FULL PAPER
a potential where the direct reduction of formic acid to for-
mate (Eqs. 4) cannot occur.
To sum up, the palladium catalyst plays a triple role: i)
activation of the aryl iodide by oxidative addition, ii) acti-
vation of CO to form an acylpalladium complex and iii)
activation of formic acid to generate formate ions. The ten-
tative mechanism of the electrosynthesis of aldehydes is out-
lined in Scheme 1
Acknowledgments
This work has been supported by the Consiglio Nazionale Delle
Ricerche (CNR, Italy), the Ministero dellЈUniversita e della
Ricerca Scientifica e Technologica (MURST, Italy), the Centre
National de la Recherche Scientifique (CNRS, France), the Minis-
t e` re de lЈEducation Nationale et de la Recherche et de la Technolo-
gie and the Ecole Normale Sup e´ rieure (France).
[
[
1]
2]
A. Jutand, S. N e´ gri, A. Mosleh, J. Chem. Soc., Chem. Commun.
992, 1729Ϫ1730.
S. Torii, H. Tanaka, K. Morizaki, Tetrahedron Lett. 1985, 26,
655Ϫ1658.
1
Figure 2. Cyclic voltammetry performed in DMF (containing
1
Ϫ3
nBu
4
NBF
4
, 0.3 mol dm ) at a steady glassy carbon disk electrode
[3] 3a]
S. Torii, H. Tanaka, T. Hamatani, K. Morizaki, A. Jutand,
Ϫ1
0
(
(
(
(
i.d. ϭ 2 mm) with a scan rate of 0.2 V s . 20 °C. a) [Pd (Ph
3
P)
4
]
[3b]
F. Pflüger, J. F. Fauvarque, Chem. Lett. 1986, 169Ϫ172. Ϫ
C. Amatore, A. Jutand, F. Khalil, M. Nielsen, J. Am. Chem.
Soc. 1992, 114, 7076Ϫ7085.
Ϫ3
Ϫ3
2 mmol dm ) alone. b) [Pd(Ph
P)
3 4
] (2 mmol dm ) and HCO H
2
Ϫ3
Ϫ3
0
0.4 mol dm ): reduction after one hour reaction. c) [Pd (Ph
3
P)
4
]
Ϫ3
2 mmol dm ) and HCO
2
H (0.4 mol dm ): oxidation after one
[4]
[5]
A. Jutand, S. Negri, Synlett 1997, 719Ϫ721.
hour reaction.
I. Chiarotto, I. Carelli, S. Cacchi, P. Pace, J. Electroanal. Chem.
1
995, 385, 235Ϫ239.
[
[
[
6]
7]
8]
T. Okano, N. Harada, J. Kiji, Bull. Soc. Chim. Jpn. 1994, 67,
329Ϫ2332.
2
(
5)
C Amatore, E. Carr e´ , A. Jutand, H. Tanaka, S. Torii, I. Chiar-
otto, I. Carelli, Electrochimica Acta 1997, 42, 2143Ϫ2152.
The electrosynthesis of aldehydes was performed in a three-elec-
trode cell containing 40 mL of anhydrous DMF and nBu NBF
been estimated from conductivity measurements: K ϭ
4
4
Ϫ3
Ϫ5
Ϫ3
(0.2 mol dm ) as supporting electrolyte. The cathode was a
glassy carbon electrode and the anode a platinum wire. The
reference electrode was a saturated calomel electrode. The cell
was charged with 0.5 mmol of aryl iodide, 10 mol-% of the cata-
lyst and 2 equivalents of formic acid. Electrolyses were per-
formed at constant potential (see Table 1) under one atmos-
phere of CO at 60°C. After acidic hydrolysis, the mixture was
extracted with diethyl ether. Crude reaction mixtures were then
analyzed by GC analysis and yields were calculated by using
authentic samples as external standards.
[9] [9a]
5
·10 mol dm , at 0 °C.
On the basis of its reduction potential (Ϫ0.99 V vs. SCE),
ϩ
[
HPd(Ph P) ] is reducible under our electrolytic con-
3 2
ditions to palladium(0) complex (Eq. 6, Figure 2b).
(
6)
Consequently, the equilibrium sketched in Eq. 5 is con-
tinuously displaced to the right side to generate a small but
constant flux of formate in the reaction medium. The result
is that formate ions are generated at an adequate rate by
this combination of chemical and electrochemical steps, at
The identical reduction peak was observed in the reaction
0
[9b]
31
of acetic acid and [Pd (Ph P) )]. Ϫ
3
4
The identical P-NMR
signal was observed in the reaction of acetic acid and
0
[
3 4
Pd (Ph P) ].
Received December 23, 1998
[O98585]
Eur. J. Org. Chem. 1999, 1471Ϫ1473
1473