Low-pressure hydrogenation of arenecarboxylic acids to aryl aldehydes

Article abstract of DOI:10.1002/adsc.201000418

A highly effective palladium catalyst has been developed that allows the selective hydrogenation of arenecarboxylic acids to the aryl aldehydes in the presence of pivalic anhydride already at 5 bar hydrogen pressure. With the new catalyst, diversely functionalized aromatic and heteroaromatic aldehydes are conveniently accessible from the corresponding carboxylic acids in a single reaction step without any overreduction to the alcohols.

Full text of DOI:10.1002/adsc.201000418

COMMUNICATIONS
DOI: 10.1002/adsc.201000418
Low-Pressure Hydrogenation of Arenecarboxylic Acids to Aryl
Aldehydes
Lukas J. Gooßen,a,* Bilal Ahmad Khan,a Thomas Fett,b and Matthias Treub
a
FB Chemie – Organische Chemie, TU Kaiserslautern, Erwin-Schroedinger-Strasse, Geb. 54, 67663 Kaiserslautern,
Germany
Fax : (+49)-631-205-3921; e-mail: goossen@chemie.uni-kl.de
Boehringer Ingelheim RCV GmbH & Co. KG, Dr.-Boehringer-Gasse 5-1, 1121 Wien, Austria
b
Received: May 28, 2010; Published online: September 7, 2010
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/adsc.201000418.
Abstract: A highly effective palladium catalyst has
been developed that allows the selective hydrogena-
tion of arenecarboxylic acids to the aryl aldehydes
in the presence of pivalic anhydride already at 5 bar
hydrogen pressure. With the new catalyst, diversely
functionalized aromatic and heteroaromatic alde-
hydes are conveniently accessible from the corre-
sponding carboxylic acids in a single reaction step
without any overreduction to the alcohols.
drides undergoes a palladium-catalyzed hydrogena-
tion to give the aldehydes along with pivalic acid
(Scheme 1). The reaction mechanism involves an oxi-
dative addition of the mixed anhydrides to the Pd(0)
center.[7] The high steric demand of the tert-butyl
group dictates the regiochemistry of this step, so that
the acylpalladium(II) pivalate is almost exclusively
formed. This complex reacts with molecular hydrogen
to liberate the aldehyde along with pivalic acid.
Keywords: aldehydes; carboxylic acids; catalysis;
hydrogenation; palladium
The conversion of carboxylic acid derivatives into al-
dehydes is a frequently used transformation in organic
synthesis.[1] It is usually performed using stoichiomet-
ric amounts of metal- or metal hydride-based reduc-
ing agents.[2,3,4] However, most protocols involve an
additional reaction step for the preparation of activat-
ed carboxylic acid derivatives. Also, unwanted side re-
actions, such as the overreduction to alcohols and the
reduction of other functional groups, are notoriously
hard to suppress. The transition metal-catalyzed hy-
Scheme 1. Hydrogenation of arenecarboxylic acids.
In the original protocol, palladium complexes of
drogenation of acid chlorides (Rosenmund reduc- PPh3 and other triarylphosphines were employed as
tion)[5] is more atom-economic, but overreduction to catalyst. However, they are only moderately active, so
the alcohols can be avoided only with elaborate flow- that high hydrogen pressures of at least 30 bar are
through reaction layouts, which are impractical for necessary for good turnover. The Yamamoto alde-
small-scale syntheses. This is why many chemists still hyde synthesis thus requires high-pressure equipment,
prefer to first reduce the carboxylic acids all the way which is unavailable in many preparative labs
to the alcohols, and then to selectively reoxidize them throughout industry and academia. This may be the
to the aldehydes using Swern-type reactions.
main reason why this elegant transformation has so
A new concept for the direct reduction of carboxyl- far found surprisingly little application in organic syn-
ic acids to aldehydes was introduced by Yamamoto thesis. The alternative use of hypophosphite salts as
et al.[6] In his single-step protocol the carboxylic acids reducing agents is more practical for lab-scale applica-
are activated in situ by treatment with pivalic anhy- tions, but waste-intensive and usually gives lower
dride. The resulting equilibrium mixture of anhy- yields.[8]
2166
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 2166 – 2170

Low-Pressure Hydrogenation of Arenecarboxylic Acids to Aryl Aldehydes
There clearly remained a need for a new generation conversion of the carboxylic acid and high yields of
of more active Pd catalysts that would allow perfor- the aldehyde were finally achieved with dicyclohexyl-
mance of the Yamamoto aldehyde synthesis below phenylphosphine, a ligand seldom used in palladium
10 bar hydrogen pressures which can be reached with catalysis (entry 10).
laboratory-scale hydrogenation equipment and indus-
trial-multi purpose reactors.
With dicyclohexylphenylphosphine, the model reac-
tion proceeded well even when reducing the hydrogen
We began our search for a more active catalyst by pressure to 5 bar. At this low pressure, several Pd pre-
studying the hydrogenation of 4-methoxybenzoic acid cursors were tested (entries 12–16). The best yields
(1a) in the presence of pivalic anhydride and 1 mol% were achieved with palladium acetylacetonate, but
of various palladium catalysts in THF. The model other Pd(II) salts or Pd(0) precursors can be used as
system was chosen based on the fact that this elec- well. A slightly lower conversion to 2a was observed
tron-rich benzoic acid is of particularly low reactivity. when reducing the reaction temperature to 608C
Using Yamamotoꢁs conditions [PdACHTUNTGRNEUNG(OAc)2/PPh3, 30 bar (entry 17). Several other solvents were tested as a re-
H2], 4-methoxybenzaldehyde (2a) was formed only in placement for THF. We were pleased to find that not
low yield (Table 1, entry 1). At a reduced H2 pressure only DMF, but also non-toxic acetone (entry 18) and
of 15 bar, the activity of this catalyst was unsatisfacto- diethyl ketone were effective solvents. With toluene
ry. However, a systematic investigation of various and acetonitrile, inferior results were obtained.
phosphine ligands revealed that with a more electron-
rich triarylphosphine, moderate yields are achieved PdACHTUNGTRENNUNG
Under the optimal conditions, using 1 mol% of
(acac)2, 5 mol% of dicyclohexylphenylphosphine,
even at this reduced pressure (entries 2 and 3). Steri- acetone (2 mL), pivalic anhydride (3 equiv.), 808C,
cally demanding, electron-rich trialkylphosphines such 20 h, 4-methoxybenzoic acid was smoothly converted
as PCy3 and some of Buchwaldꢁs dicyclohexylbiaryl- to the corresponding aldehyde in 92% yield at a hy-
phosphines gave even better results (entries 4–9). Full drogen pressure of only 5 bar. In this screening, the
catalyst and ligand, although expensive, were used in
1 and 5 mol% loadings, respectively, to ensure good
conversion of a range of substrates. On these small
Table 1. Optimization of the catalyst system.[a]
scales, further reduction of the catalyst loading result-
ed in decreased yields.
The reactions can thus be carried out using a stan-
dard hydrogenation reactor, for example, a glass auto-
clave. Most industrial reactors are also certified for
pressures of up to 5 bar. The new catalyst system is
active even at ambient pressures: With an increased
catalysts loading of 3 mol% and using DMF as a more
high-boiling solvent, 73% yield was reached.
Having thus established a reliable low-pressure pro-
tocol for the Yamamoto aldehyde synthesis, we next
tested its generality by applying it to the hydrogena-
tion of various carboxylic acids. As can be seen from
Table 2, the new catalyst allows the smooth conver-
sion of aromatic and heteroaromatic carboxylic acids.
In particular, para-, meta- and ortho-substituted ben-
zoic acids reacted equally well. Five-ring heterocycles,
quinolines and terephthalic acid were also hydrogen-
ated in high yields. Many functional groups are toler-
ated: Carboxylic acids containing alkoxy, keto, cyano,
protected amino, and even ester groups were success-
fully converted without competing reduction of
double bonds or of functional groups.
Entry
Pd source
Pd(OAc)2
Phosphine
PPh3
PACHTUNGTRENNUNG
PACHTUNGTRENNUNG
SPhos
XPhos
DavePhos
JohnPhos
PCy3
CyJohnPhos
PPhCy2
PPhCy2
PPhCy2
PPhCy2
PPhCy2
PPhCy2
PPhCy2
PPhCy2
PPhCy2
PPhCy2
H2 [bar]
2a [%]
1
2
3
4
5
6
7
8
G
30
15
15
15
15
15
15
15
15
15
5
5
5
5
5
5
5
5
1
12
33
52
57
70
72
89
90
93
99
89
0
64
80
80
91
87
92
73
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(CN)2
Pd(dba)2
Pd(F6-acac)2
Pd(TFA)2
Pd(acac)2
Pd(acac)2
Pd(acac)2
Pd(acac)2
T
E
T
E
N
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
G
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
9
10
11
12
13
14
15
16
17[b]
18[c]
19[d]
N
N
R
E
G
R
A
In conclusion, the new catalyst system, generated in
situ from PdACTHUNRGTNEUNG(acac)2 and dicyclohexylphenylphosphine,
[a]
Reaction conditions: 1.00 mmol 1a, 3.00 mmol pivalic an-
hydride, 1 mol% Pd source, 5 mol% phosphine, 2 mL
THF, 808C, 14 h. Yields determined by HPLC analysis
using propiophenone as internal standard.
608C.
allows the selective hydrogenation of arenecarboxylic
acids to aryl aldehydes already at low hydrogen pres-
sures. The new protocol is convenient for applications
on both laboratory and industrial scales, and might
convince organic chemists to add the Yamamoto alde-
hyde synthesis to their chemical toolbox.
[b]
[c]
[d]
Acetone as solvent.
3 mol% of Pd
ACHTUNGTRENNUNG(acac)2, 15 mol% of dicyclohexylphenyl-
phosphine, DMF, 50 h.
Adv. Synth. Catal. 2010, 352, 2166 – 2170
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2167

Lukas J. Gooßen et al.
COMMUNICATIONS
Table 2. Hydrogenation of arenecarboxylic acids 1.[a]
Experimental Section
General Procedure for the Synthesis and
Characterization of the Aldehydes (1a–r)
An oven-dried, argon-flushed 10-mL glass vessel with
septum top was charged with benzoic acid 1a–r (1.00 mmol),
PdACTHNUGTRENNG(U acac)2 (3.05 mg, 0.01 mmol), and dicyclohexylphenyl-
phosphine (13.7 mg, 0.05 mmol). Degassed THF (2 mL) and
degassed pivalic anhydride (0.62 mL, 3.00 mmol) were
added. The vessel was placed in a steel autoclave, which was
then purged with hydrogen and then pressurized with 5 bar
of hydrogen. The reaction mixture was stirred at 808C for
20 h, then cooled to room temperature. The autoclave pres-
sure was released, the reaction mixture diluted with 10 mL
saturated NaHCO3 solution and extracted with ethyl acetate
(3ꢂ20 mL). The combined organic layers were washed with
water and brine and dried over MgSO4, filtered, and the vol-
atiles were removed under vacuum. The residue was puri-
fied by column chromatography (SiO2, ethyl acetate/hexane
gradient) yielding the corresponding aldehydes.
The particularly volatile aldehydes 2a, e, g, h, o, p, and q
were isolated as bisulfite adducts: after releasing the pressue
from the autoclaves, 38% sodium bisulfite solution
(1.50 mL, 8.10 mmol) was added via syringe to the reaction
vessel. The residue mixture was stirred at 508C for 4 h, then
cooled to room temperature. The white crystalline adducts
were filtered off and washed with chloroform (10 mL) and
water (0.50 mL), dried under vacuum, and weighed to deter-
mine the yield. For the spectroscopic characterization, part
of the aldehydes was then liberated by adding the adduct to
saturated aqueous NaHCO3 solution (2.0 mL), followed by
extraction with CDCL3 (2.0 mL). The NMR samples were
then washed with water and brine, dried over MgSO4 and
filtered.
Product
Yield [%] Product
Yield [%]
89
91
91
75
74
42
92
92
80
80
75
79
4-Methoxybenzaldehyde (2a): Compound 2a was pre-
pared from 4-methoxybenzoic acid (1a) (152 mg, 1.00 mmol)
affording 2a as a colorless oil; yield: 124.9 mg (90.8%). The
spectroscopic data matched those reported in the literature
for 4-methoxybenzaldehyde [CAS: 123-11-5].
4-Acetamidobenzaldehyde (2b): Compound 2b was pre-
pared from 4-acetamidobenzoic acid (1b) (179 mg,
1.00 mmol) affording 2b as a colorless crystalline solid;
yield: 149 mg (91%). The spectroscopic data (NMR)
matched those reported in the literature for 4-acetamido-
benzaldehyde [CAS: 122-85-0].
80
80
70
88
4-Cyanobenzaldehyde (2c): Compound 2c was prepared
from 4-cyanobenzoic acid (1c) (149 mg, 1.00 mmol) afford-
ing 2c as a colorless crystalline solid; yield: 98.6 mg (75%).
The spectroscopic data (NMR) matched those reported in
the literature for 4-cyanobenzaldehyde [CAS: 105-07-7].
4-Acetylbenzaldehyde (2d): Compound 2d was prepared
from 4-acetylbenzoic acid (1d) (164 mg, 1.00 mmol) afford-
ing 2d as a colorless crystalline solid; yield: 137 mg (92%).
The spectroscopic data (NMR) matched those reported in
the literature for 4-acetylbenzaldehyde [CAS: 3457-45-2].
4-tert-Butylbenzaldehyde (2e): Compound 2e was pre-
pared from 4-tert-butylbenzoic acid (1e) (180 mg,
1.00 mmol) affording 2e as a colorless oil; yield: 123 mg
(75.8%). The spectroscopic data (NMR) matched those re-
ported in the literature for 4-tert-butylbenzaldehyde [CAS:
939-97-9].
85
81[b]
[a]
Reaction conditions: 1.00 mmol of benzoic acid 1,
1 mol% Pd(acac)2, 5 mol% dicyclohexylphenylphosphine,
pivalic anhydride (3 equiv.), THF (2 mL), 808C, H2
ACHTUNGTRENNUNG
(5 bar), 20 h.
HPLC yields.
[b]
2168
asc.wiley-vch.de
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 2166 – 2170

Low-Pressure Hydrogenation of Arenecarboxylic Acids to Aryl Aldehydes
4-Methoxycarbonylbenzaldehyde (2f): Compound 2f was
prepared from 4-methoxycarbonylbenzoic acid (1f) (180 mg,
1.00 mmol) affording 2f as a colorless crystalline solid;
yield: 132 mg (80%). The spectroscopic data (NMR)
matched those reported in the literature for 4-methoxycar-
bonylbenzaldehyde [CAS: 1571-08-0].
4-Fluorobenzaldehyde (2g): Compound 2g was prepared
from 4-fluorobenzoic acid (1g) (143.0 mg, 1.00 mmol) afford-
ing 2g as a colorless oil; yield: 90 mg (80%). The spectro-
scopic data (NMR) matched those reported in the literature
for 4-fluorobenzaldehyde [CAS: 459-57-4].
4-(Trifluoromethyl)benzaldehyde (2h): Compound 2h was
prepared from 4-(trifluoromethyl)benzoic acid (1h) (190 mg,
1.00 mmol) affording 2h as a colorless oil; yield: 139 mg
(80%). The spectroscopic data (NMR) matched those re-
ported in the literature for 4-(trifluoromethyl)benzaldehyde
[CAS: 455-19-6].
1,4-Benzenedicarboxaldehyde (2i): Compound 2i was pre-
pared from 1,4-benzenedicarboxalic acid (1i) (166 mg,
1.00 mmol) affording 2i as a colorless crystalline solid; yield:
114 mg (85%). The spectroscopic data (NMR) matched
those reported in the literature for 1,4-benzenedicarboxalde-
hyde [CAS: 623-27-8].
3,4,5-Trimethoxybenzaldehyde (2j): Compound 2j was
prepared from 3,4,5-trimethoxybenzoic acid (1j) (212 mg,
1.00 mmol) affording 2j as a colorless crystalline solid; yield:
174 mg (75.8%). The spectroscopic data (NMR) matched
those reported in the literature for 3,4,5-trimethoxybenzal-
dehyde [CAS: 86-81-7].
3-Quinolinecarboxaldehyde (2k): Compound 2k was pre-
pared from 3-quinolinecarboxylic acid (1k) (177 mg,
1.00 mmol) affording 2k as a colorless crystalline solid;
yield: 102 mg (78%). The spectroscopic data (NMR)
matched those reported in the literature for 3-quinolinecar-
boxaldehyde [CAS: 13669-42-6].
1,3-Benzenedicarboxaldehyde (2l): Compound 2l was pre-
pared from 1,3-benzenedicarboxylic acid (1l) (166 mg,
1.00 mmol) affording 2l as colorless crystals; yield: 56.0 mg
(42%). The spectroscopic data (NMR) matched those re-
ported in the literature for 1,3-benzenedicarboxaldehyde
[CAS: 626-19-7].
ported in the literature for 3-thiophenecarboxaldehyde
[CAS: 498-62-4].
2-Methylbenzaldehyde (2q): Compound 2q was prepared
from 2-methylbenzoicacid (1q) (138 mg, 1.00 mmol) afford-
ing 2q as a colorless oil; yield: 106.0 mg (88%). The spectro-
scopic data (NMR) matched those reported in the literature
for 2-methylbenzaldehyde [CAS: 529-20-4].
(2E)-3-Phenylprop-2-enal (2r): Compound 2r [CAS:
16939-04-1] was prepared from (E)-3-phenylprop-2-enoic
acid (1r) (148 mg, 1.00 mmol). Unfortunately, the adduct
formation did not take place. Therefore, the identity of the
product 2r was confirmed by GC-MS and the yield deter-
mined by quantative HPLC to be 81% based on a response
factor obtained with commercial (2E)-3-phenylprop-2-enal
(2r) [CAS: 104-55-2] using propiophenone (25 mL) as an in-
ternal HPLC standard.
Preparative-Scale Synthesis of 4-Acetamidobenzaldehyde:
A 300-mL hydrogenation reactor was charged with 4-acet-
amidobenzoic acid (1a) (5.00 mmol, 896 mg), PdACHTUNGTRENNUNG(acac)2
(15.2 mg, 0.05 mmol) and dicyclohexylphenylphosphine
(72.2 mg, 0.25 mmol). THF (15.0 mL) and pivalic anhydride
(3.0 mL, 15.0 mmol) were added via syringe. The autoclave
was purged with hydrogen and then pressurized with 5 bar
of hydrogen. The reaction was stirred at 808C for 20 h, then
cooled to room temperature. The pressure was released, the
reaction mixture diluted with saturated NaHCO3 solution
(25 mL) and extracted with ethyl acetate (3ꢂ25.0 mL). The
combined organic layers were washed with water and brine,
dried over MgSO4, filtered, and the volatiles were removed
under vacuum. The residue was taken up in of diethyl ether
(10 mL) causing the product 4-acetamidobenzaldehyde (2b)
to precipitate in spectroscopically pure form; yield: 584 mg
(71%).
Acknowledgements
We thank HEC Pakistan for a scholarship to B.A.K.
References
3-Acetamidobenzaldehyde (2m): Compound 2m was pre-
pared from 3-acetamidobenzoic acid (1m) (179 mg,
1.00 mmol) affording 2m as a colorless crystalline solid;
yield: 150 mg (92%). The spectroscopic data (NMR)
matched those reported in the literature for 3-acetamido-
benzaldehyde [CAS: 59755-25-8].
3-Cyanobenzaldehyde (2n): Compound 2n was prepared
from 3-cyanobenzoic acid (1n) (150 mg, 1.00 mmol) afford-
ing 2n as colorless crystals; yield: 102 mg (78%). The spec-
troscopic data (NMR) matched those reported in the litera-
ture for 3-cyanobenzaldehyde [CAS: 24964-64-5].
Benzaldehyde (2o): Compound 2o was prepared from
benzoic acid (1o) (122 mg, 1.00 mmol) affording 2o as color-
less oil; yield: 84.0 mg (79%). The spectroscopic data
(NMR) matched those reported in the literature for benzal-
dehyde [CAS: 100-52-7].
3-Thiophenecarboxaldehyde (2p): Compound 2p was pre-
pared from 3-thiophenecarboxylic acid (1p) (128 mg,
1.00 mmol) affording 2p as a colorless oil; yield: 79.0 mg
(70%). The spectroscopic data (NMR) matched those re-
[1] B. M. Trost, Comprehensive Organic Synthesis, Perga-
mon, Oxford, 1991, Vol. 8, pp 259–305.
[2] Aluminium hydride derivatives: a) L. I. Zakharkin, V. V.
Gavrilenko, D. N. Maslin, I. M. Khorlina, Tetrahedron
Lett. 1963, 2087–2090; b) H. C. Brown, P. M. Weissman,
J. Org. Chem. 1966, 31, 283–287; c) S. Chandrasekhar,
M. S. Kumar, B. Muralidhar, Tetrahedron Lett. 1998, 39,
909–910; d) M. Muraki, T. Mukaiyama, Chem. Lett.
1974, 1447–1450.
[3] Boron hydride derivatives: a) thexylchloroborane-di-
methyl sulfide: H. C. Brown, J. S. Cha, N. M. Yoon, N.
Behrooz, J. Org. Chem. 1987, 52, 5400–5406; b) borane-
dimethyl sulfide/PCC: H. C. Brown, C. G. Rao, S. U.
Kulkarni, Synthesis 1979, 704–705.
[4] Other reducing agents: a) isobutylmagnesium bromide/
dichlorobis(p-cyclopentadienyl)titanium: F. Sato, T.
Jinbo, M. Sato, Synthesis 1981, 871; b) stannous chloride/
HCl: H. Stephen, J. Chem. Soc. 1925, 1874–1877; c) trie-
thylsilane: T. Fukuyama, S. C. Lin, L. Li, J. Am. Chem.
Soc. 1990, 112, 7050–7051; d) lithium/methylamine:
Adv. Synth. Catal. 2010, 352, 2166 – 2170
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2169

Lukas J. Gooßen et al.
COMMUNICATIONS
A. O. Bedenbaugh, J. H. Badenbaugh, W. A. Bergin,
J. D. Adkins, J. Am. Chem. Soc. 1970, 92, 5774–5775;
e) heterogeneous catalysts: T. Yokoyama, N. Fujita,
Appl. Catal. A: General 2004, 276, 179–185.
1143–1144; c) K. Nagayama, I. Shinizu, A. Yamamoto,
Bull. Chem. Soc. Jpn. 2001, 74, 1803–1815.
[7] For reactions based on similar concepts, see: a) L. J.
Gooßen, K. Ghosh, Angew. Chem. 2001, 113, 3566–
3568; Angew. Chem. Int. Ed. 2001, 40, 3458–3460; b) R
Kakino, H. Narahashi, I. Shimizu, A. Yamamoto, Chem.
Lett. 2001, 12, 1242–1243; c) L. J. Gooßen, J. Paetzold,
L. Winkel, Synlett 2002, 10, 1721–1723; d) L. J. Gooßen
and K. Ghosh, Chem. Commun. 2001, 2084–2085.
[8] L. J. Gooßen, K. Ghosh, Chem. Commun. 2002, 836–
837.
[5] a) K. W. Rosenmund, Ber. Dtsch. Chem. Ges. 1918, 51,
585–593; b) E. Mosettig, Org. React. 1948, 4, 362–377;
c) J. J. Li, Name Reactions, 2nd edn., Springer, Berlin
2003, p 343.
[6] a) K. Nagayama, F. Kawataka, M. Sakamoto, I. Shimizu,
A. Yamamoto, Chem. Lett. 1995, 367–368; b) K. Na-
gayama, I. Shinizu, A. Yamamoto, Chem. Lett. 1998,
2170
asc.wiley-vch.de
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 2166 – 2170