Selective preparation of terminal alkenes from aliphatic carboxylic acids by a palladium-catalysed decarbonylation-elimination reaction

Article abstract of DOI:10.1016/j.tetlet.2010.05.018

Trialkylamines were used as additives in the decarbonylation-elimination reaction catalysed by the combination of palladium(II) chloride and DPE-Phos. Aliphatic carboxylic acids were transformed at relatively low temperature into terminal alkenes in high yield and high selectivity, without the need for distillation, thereby avoiding isomerisation.

Full text of DOI:10.1016/j.tetlet.2010.05.018

Tetrahedron Letters 51 (2010) 3712–3715
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Selective preparation of terminal alkenes from aliphatic carboxylic acids
by a palladium-catalysed decarbonylation–elimination reaction
b
a
Jérôme Le Nôtre a, Elinor L. Scott a, , Maurice C. R. Franssen , Johan P. M. Sanders
*
a Valorisation of Plant Production Chains, Wageningen University and Research Centre, PO Box 17, 6700 AA Wageningen, The Netherlands
b Laboratory of Organic Chemistry, Wageningen University and Research Centre, Dreijenplein 8, 6703 HB Wageningen, The Netherlands
a r t i c l e i n f o
a b s t r a c t
Article history:
Trialkylamines were used as additives in the decarbonylation–elimination reaction catalysed by the com-
bination of palladium(II) chloride and DPE-Phos. Aliphatic carboxylic acids were transformed at relatively
low temperature into terminal alkenes in high yield and high selectivity, without the need for distillation,
thereby avoiding isomerisation.
Received 18 March 2010
Revised 21 April 2010
Accepted 7 May 2010
Available online 12 May 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Decarbonylation
Fatty acids
Palladium
Terminal alkenes
In recent years, biofuels have received significant interest due to
the depletion of fossil feedstocks and the need for a reduction of
carbon dioxide emission.1 While bioethanol is the most important,
fatty esters derived from renewable resources can also be used
effectively as a transportation fuel. Fatty esters are easily prepared
from vegetable oils by transesterification with alcohols.2 However,
a drawback is their reduced compatibility with conventional en-
gines. Therefore, a more suitable approach consists of their trans-
formation into high quality hydrocarbons.
Elimination of oxygen is required to convert fatty acids into
hydrocarbons, for which several options are available. Catalytic
deoxygenation reactions such as decarboxylation reactions can
be performed using heterogeneous catalysts at high tempera-
tures.3–7 Conventional hydrotreating catalysts such as CoMo or
NiMo,4 as well as palladium on carbon,5 platinum on alumina,6
or zeolite7 have been used to convert fatty acids or esters into al-
kanes. In addition, homogeneous catalysts based on palladium
and rhodium can also be employed in deoxygenation reactions, fol-
lowing a different mechanistic pathway which allows the conver-
sion of carboxylic acids into alkenes with the production of
carbon monoxide and water.8
ficult to achieve, especially with homogeneous catalysts, because
of the high temperature needed for the reaction (200–280 °C).8a,b
Heterogeneous catalysts also give modest selectivities (70–85%).
In the latter case the alkene is distilled from the reaction mixture
directly upon its formation, thereby reducing the risk of isomerisa-
tion.8c However, combining homogeneous catalysts such as palla-
dium and rhodium salts with continuous distillation of the
products during the reaction afforded selectivities above 95%.8d
Alternatively, the reaction is stopped before completion. With a
palladium-catalysed system (Scheme 1), high selectivity towards
the 1-alkene 2 can only be achieved by stopping the reaction at a
conversion of 85%; the product was isolated in a yield of 66%.8e
In our current research on converting biomass into bulk chem-
icals, we are concentrating our efforts on the transformation of
amino acids, obtained from protein waste streams of biofuels pro-
duction, into products for the chemical and plastics industries.9 In
particular, we focused our work on decarboxylation reactions, and
showed for instance that glutamic acid, which can be obtained
from the waste stream of bioethanol production, can be converted
Decarbonylation/dehydration reactions have been scarcely
studied, even though they would constitute a useful tool in organic
synthesis. Performing the transformation selectively is a challenge
in itself, since isomerisation into internal alkenes is thermodynam-
ically favoured. High selectivity towards the 1-alkene is indeed dif-
* Corresponding author. Tel./fax: +31 317 483011.
E-mail address: elinor.scott@wur.nl (E.L. Scott).
Scheme 1. Decarbonylation/dehydration reaction of 4-phenylbutyric acid (1).
0040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.05.018

J. Le Nôtre et al. / Tetrahedron Letters 51 (2010) 3712–3715
3713
enzymatically into 4-aminobutyric acid in an economically viable
process.9b
A variety of trialkylamines were tested as additives in catalytic
amounts, and for all reactions, the selectivity towards the 1-alkene
remained high, between 92% and 98% (Table 1, entries 8–11).
A closer investigation of the catalytic system was then per-
formed. A decrease in the palladium loading to 1 mol % had a dra-
matic effect on the reaction with a drop in conversion down to 35%
after 18 h. In the same way, a decrease of the phosphine/palladium
ratio had a similar consequence. Concerning the reaction condi-
tions, the use of triethylamine also positively influenced the kinet-
ics of the reaction. A conversion of 85% was obtained after 16 h
reaction at 110 °C without additive (Table 1, entry 1), the presence
of triethylamine gave the same level of conversion and selectivity
after only 8 h at the same temperature (Table 2, entry 2). The reac-
tion temperature could also be decreased to 90 °C without affect-
ing the conversion and the selectivity (Table 2, entry 3), but a
lower temperature appeared to be detrimental (Table 2, entry 4).
In their report, Gooßen and Rodriguez used pivalic anhydride as
a reactant to form in situ mixed anhydride 4 (R0 = t-Bu) with the
substrate.8e They explained that the new anhydride then adds oxi-
datively to the palladium(0) catalyst to form the less sterically
demanding acyl complex 5, which then undergoes a decarbonyla-
tion step in the catalytic cycle (Scheme 2, Pathway A).
However, acetic anhydride has been used previously in decarb-
onylation/dehydration reactions of carboxylic acids, and Miller
et al. showed that it led to high yields and good selectivity at high
temperature (250 °C).8d These results indicated that the steric ef-
fect of the mixed anhydride formed with the substrate (Scheme 2,
Pathway A when R0 = Me) probably had no influence on the mech-
anistic pathway. These authors also explained that at high temper-
ature, a new anhydride 6 could be formed by the substrate only.
From this, one can conclude that oxidative addition to palla-
dium(0) to obtain the acyl complex 5 is not determined by steric
effects (Scheme 2, Pathway B). An additional advantage of using
acetic anhydride is its ready availability and lower price compared
to pivalic anhydride.
Herein we report a catalytic method to fully convert aliphatic
carboxylic acids into terminal alkenes in high yields and with high
selectivities without the need for continuous distillation of the
products. Trialkylamines are used as additives to stabilise the cat-
alytic species and the reaction can proceed without isomerisation
of the terminal double bond at relatively low temperatures (ca.
100 °C).
In the article of Gooßen and Rodriguez,8e palladium(II) chloride
was used as the catalyst source with a bidentate phosphine ligand,
bis(2-diphenylphosphinophenyl)ether (DPE-Phos) (Scheme 1). An
excess of pivalic anhydride led to the in situ formation of a mixed
anhydride with the substrate, which underwent a decarbonylation/
dehydration reaction to form the corresponding alkene, carbon
monoxide and pivalic acid. No distillation set-up was used, but
the reaction had to be stopped before completion to avoid any
isomerisation of the product (Table 1, entry 1). Indeed, when we
repeated these experiments and allowed the reaction to reach
complete conversion, the selectivity of 97:3 in favour of 1-alkene
2 decreased to 85:15 (Table 1, entry 2). In their research, Gooßen
et al. tried to improve the selectivity by using different additives,
but the addition of bases such as potassium carbonate and pyridine
did not enhance the selectivity towards terminal alkenes (Table 1,
entry 3).
It is known that alkylamines can have a positive effect on palla-
dium-catalysed reactions.10 They can play a role in reduction of the
palladium(II) species and facilitate the formation of a palladium(0)
active species.11 In addition, they are suspected to play a role in the
stabilisation of key palladium(II) intermediate species by coordina-
tion.12 Therefore, we decided to use triethylamine as an additive in
the catalytic system for the decarbonylation/dehydration reac-
tion.13 By using 1 equiv of triethylamine, full conversion was ob-
tained after 18 h at 110 °C, the 1-alkene selectivity reached 99:1,
and allylbenzene was isolated in 97% yield (Table 1, entry 4). The
effect of trialkylamines on the catalytic system was then studied
by changing the nature and the quantity of the base (Table 1, en-
tries 5–11). In all the experiments the conversion was complete
after 18 h at 110 °C. Triethylamine was used in excess compared
to the substrate and it did not affect the 1-alkene:2-alkene selec-
tivity, which remained above 95% (Table 1, entries 5 and 6). The
use of a catalytic amount of base (9 mol % triethylamine) had the
same positive effect as the use of 1 equiv and 98% selectivity was
obtained (Table 1, entry 7). Thus, it appears that triethylamine only
has an influence on the catalyst species, probably by activation and
stabilisation, and is not involved in the catalytic cycle as a reactant.
Hence, we decided to perform the reaction by substituting piva-
lic anhydride with acetic anhydride and we obtained comparable
results as previously with full conversion and a selectivity of 97%
for the 1-alkene 2 (Scheme 3). This result indeed proves that the
presence of a bulky group on the anhydride reagent is not
necessary.
According to Gooßen and Rodriguez,8e a highly polar solvent
such as DMPU is needed to obtain the best turnovers and the high-
est selectivities. This was confirmed in our studies, since no solvent
better than DMPU was found. Reactions using acetic anhydride as
solvent only gave poor conversions. However, interesting results
were obtained with acetonitrile. This solvent not only allowed a
good conversion at reflux temperature but also provided a switch
of selectivity, which reached 92% in favour of the 2-alkene 3, and
the configuration of the olefin was mainly (E) [(E)/(Z) = 92:8]
(Scheme 4). It is possible that a highly efficient isomerisation cat-
alyst, such as bis(acetonitrile)palladium(II) chloride, was formed
to some extent in the reaction mixture.14
Table 1
Effect of base additive on the decarbonylation/dehydration reactiona
Entry
Base
Quantity
Conversionb (%)
Selectivityb 2/3
1c
2
—
—
—
—
85
100
80
97:3
85:15
70:30
99:1
98:2
99:1
99:1
99:1
96:4
97:3
99:1
3c
4
Pyridine
Et3N
Et3N
Et3N
Et3N
nPr3N
nBu3N
TMEDAd
DBUe
1.0 mmol
1.0 mmol
2.0 mmol
3.0 mmol
9 mol %
9 mol %
9 mol %
9 mol %
9 mol %
100
100
100
100
100
100
100
100
5
6
7
8
9
10
11
Table 2
Reaction conditions for the decarbonylation/dehydration reactiona
Entry
T (°C)
Time (h)
Conversionb (%)
Selectivityb 2/3
1
2
3
4
110
110
90
6
8
18
18
74
81
100
20
98:2
96:4
95:5
99:1
a
Reactions conditions: 1.0 mmol 4-phenylbutyric acid (1), 2.0 mmol pivalic
anhydride, 3 mol % PdCl2, 9 mol % DPE-Phos, DMPU, 110 °C, 18 h.
70
b
Determined by 1H NMR spectroscopy.
c
a
Literature result.8e
Reactions conditions: 1.0 mmol 4-phenylbutyric acid (1), 2.0 mmol pivalic
d
N,N,N0,N0-Tetramethylethylenediamine.
1,8-Diazabicyclo[5.4.0]undec-7-ene.
anhydride, 3 mol % PdCl2, 9 mol % DPE-Phos, 9 mol % Et3N, DMPU.
e
b
Determined by 1H NMR spectroscopy.

3714
J. Le Nôtre et al. / Tetrahedron Letters 51 (2010) 3712–3715
Scheme 2. Anhydride formation and insertion of palladium(0) in the first step of the two possible mechanisms.
kylamines as additives had a drastic effect on the stability of the
catalytic species avoiding olefin isomerisation during the reaction.
We showed that the less expensive acetic anhydride can be use as a
reactant for the decarbonylation–elimination reaction at relatively
low temperature, and that the reaction can be applied to the deox-
ygenation of fatty acids into unsaturated hydrocarbons. Moreover,
the side products of the reaction (carbon monoxide and acetic acid)
can be eventually recycled, which shows the advantages of this ap-
proach compared to standard deoxygenation where carbon dioxide
is produced. We are continuing to extend the scope of this reaction
towards the conversion of biomass into bulk chemicals.
Scheme 3. Decarbonylation–elimination reaction using acetic anhydride.
Acknowledgement
We are grateful to the NWO-ACTS-ASPECT program for financial
support.
References and notes
1. (a) Huber, G. W.; Iborra, S.; Corma, A. Chem. Rev. 2006, 106, 4044; (b) Huber, G.
W.; Corma, A. Angew. Chem., Int. Ed. 2007, 46, 7184.
2. (a) Ma, F.; Hanna, M. H. Bioresour. Technol. 1999, 70, 1; (b) Xie, W.; Peng, H.;
Chen, L. Appl. Catal.,
A 2006, 300, 67; (c) Ramadhas, A. S.; Jayaraj, S.;
Muraleedharan, C. Fuel 2005, 84, 335.
Scheme 4. Decarbonylation–elimination reaction in acetonitrile.
3. (a) Maier, W.; Roth, W.; Thies, I.; van Ragué Schleyer, P. Chem. Ber. 1982, 115,
808; (b) Snåre, M.; Kubickova, I.; Mäki-Arvela, P.; Eränen, K.; Murzin, D. Yu. Ind.
Eng. Chem. Res. 2006, 45, 5708; (c) Kubickova, I.; Snåre, M.; Eränen, K.; Mäki-
Arvela, P.; Murzin, D. Yu. Catal. Today 2005, 106, 197.
We next focused our efforts on the application of the new cat-
alytic system to the transformation of fatty acids. Using the opti-
mum reaction conditions the transformation of stearic acid 7 was
performed (Scheme 5). Full conversion was obtained with 1 equiv
of triethylamine as an additive and the corresponding olefins 8 and
9 were isolated in 97% yield with a selectivity of 96% for 1-hepta-
decene 8. The reaction was repeated in acetonitrile and an inver-
sion of selectivity, albeit modest, was observed with a ratio 1-
heptadecene/2-heptadecene of 41:59.
ˇ
ˇ
4. (a) Laurent, E.; Delmon, B. Appl. Catal., A 1994, 109, 77; (b) Kubicka, D.; Šimácek,
ˇ
P.; Zilková, N. Top. Catal. 2009, 52, 161; (c) Liu, Y.; Sotelo-Boyás, R.; Murata, K.;
Minowa, T.; Sakanishi, K. Chem. Lett. 2009, 38, 552.
5. (a) Lestari, S.; Mäki-Arvela, P.; Simakova, I.; Beltramini, J.; Lu Max, G. Q.;
Murzin, D. Yu. Catal. Lett. 2009, 130, 48; (b) Mäki-Arvela, P.; Kubickova, I.;
Snåre, M.; Eränen, K.; Murzin, D. Yu. Energy Fuels 2007, 21, 30; (c) Mäki-Arvela,
P.; Snåre, M.; Eränen, K.; Myllyoja, J.; Murzin, D. Yu. Fuel 2008, 87, 3543.
6. Do, P. T.; Chiappero, M.; Lobban, L. L.; Resasco, D. E. Catal. Lett. 2009, 130, 9.
7. (a) Danuthai, T.; Jongpatiwut, S.; Rirksomboon, T.; Osuwan, S.; Resasco, D. E.
Appl. Catal., A 2009, 361, 99; (b) Scheibel, J. J.; Reilman, R. T. PCT Int. Appl.
WO2007136873, 2007; Chem. Abstr. 2008, 148, 35396.
In conclusion, we have developed a palladium-catalysed reac-
tion allowing the transformation of aliphatic carboxylic acids into
terminal alkenes in high yield and high selectivity. The use of trial-
8. (a) Fenton, D. M. U.S. Patent 3,530,198, 1970; Chem. Abstr. 1970, 73, 130601.;
(b) Foglia, T. A.; Barr, P. A. J. Am. Oil Chem. Soc. 1976, 53, 737; (c) Stern, R.;
Hillion, G. U.S. Patent 4,554,397, 1985; Chem. Abstr. 1985, 103, 70934.; (d)
Miller, J. A.; Nelson, J. A.; Byrne, M. P. J. Org. Chem. 1993, 58, 18; (e) Gooßen, L. J.;
Rodriguez, N. Chem. Commun. 2004, 724.
9. (a) Könst, P. M.; Franssen, M. C. R.; Scott, E. L.; Sanders, J. P. M. Green Chem.
2009, 11, 1646; (b) Lammens, T. M.; De Biase, D.; Franssen, M. C. R.; Scott, E. L.;
Sanders, J. P. M. Green Chem. 2009, 11, 1562.
10. Muzart, J. J. Mol. Catal. A 2009, 308, 15.
11. Vollmüller, F.; Mägerlein, W.; Klein, S.; Krause, J.; Beller, M. Adv. Synth. Catal.
2001, 343, 29.
12. Damez, C.; Estrine, B.; Bessmertnykh, A.; Bouquillon, S.; Hénin, F.; Muzart, J. J.
Mol. Catal. A 2006, 244, 93.
13. General procedure for the decarbonylation/elimination reaction of carboxylic acids
in the presence of trialkylamines: In a Schlenk tube under a nitrogen atmosphere
were introduced palladium(II) chloride (3 mol %), bis(2-diphenylphosphino-
phenyl)ether (DPE-Phos, 9 mol %) and carboxylic acid (1 equiv). Anhydrous
DMPU (4 mL) was then added followed by Ac2O (2 equiv) and the trialkylamine
(1 equiv or 9 mol %). The reaction mixture was heated at 110 °C for 18 h. After
completion of the reaction, Et2O (ca. 10 mL) was added and the organic layer
Scheme 5. Decarbonylation–elimination reaction of stearic acid (7).

J. Le Nôtre et al. / Tetrahedron Letters 51 (2010) 3712–3715
3715
was washed with saturated NH4Cl solution followed by H2O and brine.
Evaporation of the solvent afforded a crude mixture, which was analysed by 1H
NMR spectroscopy to determine the conversion. The crude was purified by
PhCH@CHCH3), 6.22 (dq, 1H, J = 15.7 Hz, J = 6.5 Hz, PhCH@CHCH3), 1.87 (dd,
3H, J = 6.5 Hz, J = 1.6 Hz, PhCH@CHCH3); 13C NMR (100 MHz, CDCl3): d = 137.2,
130.8, 128.1, 125.8, 125.6, 125.4, 18.1; MS (EI): m/z (%) = 118 (65) [M+], 117
(100), 115 (35), 91 (34), 65 (12), 63 (15), 51 (34), 50 (26), 39 (36).
filtration through
a plug of silica gel using hexane as an eluent. After
concentration to dryness the product was obtained as a colourless oil.
3-Phenyl-1-propene (2): Colourless oil; 1H NMR (400 MHz, CDCl3): d = 7.24–
7.18 (m, 2H, aryl), 7.14–7.08 (m, 3H, aryl), 5.95–5.84 (m, 1H, CH2CH@CH2),
5.03–4.96 (m, 2H, CH2CH@CH2), 3.31 (d, 2H, J = 6.7 Hz, CH2CH@CH2); 13C NMR
(100 MHz, CDCl3): d = 140.0, 137.4, 128.5, 128.4, 126.0, 115.7, 40.2; MS (EI): m/
z (%) = 118 (60) [M+], 117 (100), 91 (60), 65 (52), 63 (45), 62 (14), 51 (64), 50
(39), 39 (90).
1-Heptadecene (8): Colourless oil; 1H NMR (400 MHz, CDCl3): d = 5.86–5.76 (m,
1H, CH2CH@CH2), 4.98 (br d, 1H, J = 17.1 Hz, trans-CH2CH@CH2), 4.92 (br d, 1H,
J = 10.2 Hz, cis-CH2CH@CH2), 2.07–2.00 (m, 2H, CH2CH@CH2), 1.40–1.22 (m,
26H), 0.88 (t, 3H, J = 6.8 Hz); 13C NMR (100 MHz, CDCl3): d = 139.2, 114.1, 33.8,
32.0, 29.72, 29.70, 29.68, 29.65, 29.5, 29.4, 29.2, 29.0, 22.7, 14.1; MS (EI): m/z
(%) = 238 (6) [M+], 125 (10), 111 (18), 98 (10), 97 (38), 85 (12), 84 (22), 83 (50),
82 (20), 71 (31), 70 (42), 69 (58), 68 (16), 67 (15), 57 (81), 56 (57), 55 (89), 54
(22), 43 (100), 42 (34), 41 (92), 40 (15).
Trans-1-phenyl-1-propene (3): Colourless oil; 1H NMR (400 MHz, CDCl3):
d = 7.34–7.24 (m, 4H, aryl), 7.22–7.14 (m, 1H, aryl), 6.45–6.36 (m, 1H,
14. Yu, J.; Gaunt, M. J.; Spencer, J. B. J. Org. Chem. 2002, 67, 4627.