Silver triflate-catalysed synthesis of γ-lactones from fatty acids

Article abstract of DOI:10.1039/b916853b

A silver(i)-based catalytic system was found to efficiently promote the double-bond isomerization of unsaturated fatty acids and at the same time to mediate the intramolecular addition of the carboxylate group to the isomerized double bond under selective formation of five-ring lactones. The resulting one-step isomerization-γ-lactonization process allows the direct conversion of unsaturated fatty acids into the corresponding γ-lactones in good yields.

Full text of DOI:10.1039/b916853b

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/greenchem | Green Chemistry
Silver triflate-catalysed synthesis of c-lactones from fatty acids†
a
a
b
Lukas J. Gooßen,* Dominik M. Ohlmann and Markus Dierker
Received 17th August 2009, Accepted 21st November 2009
First published as an Advance Article on the web 16th December 2009
DOI: 10.1039/b916853b
A silver(I)-based catalytic system was found to efficiently
promote the double-bond isomerization of unsaturated
fatty acids and at the same time to mediate the intramolec-
ular addition of the carboxylate group to the isomerized
double bond under selective formation of five-ring lactones.
The resulting one-step isomerization–c-lactonization pro-
cess allows the direct conversion of unsaturated fatty acids
into the corresponding c-lactones in good yields.
This and related strategies are useful for the synthesis of short-
chain lactones, where the aldehydes are readily available in large
quantities. However, those aldehydes required to access longer-
chain derivatives such as g-stearolactone are rare and expensive.
For the C18-lactone, an expedient synthesis from 1-hexadecene
derived from petrochemical sources by a free radical reaction
mechanism has been described by Burger et al. Unfortunately,
this process involves superstoichiometric amounts of the heavy
metal manganese.
13
Arguably the most sustainable synthetic entry to aliphatic
g-lactones would instead start from long-chain unsaturated
carboxylic acids derived directly from renewable fats and oils,
which have the complete carbon skeleton in place as well as the
correct oxidation state.
As shown in Scheme 2, one can imagine the possibility
of shifting the double bond of such compounds toward the
carboxylate group by several consecutive catalytic double-bond
migration steps.
Aliphatic g-lactones are widely present in plants, vegetables and
1
fruits. Due to their pleasant odor, they are also industrially
2
important as ingredients in perfumes and as food additives.
Usually, g-lactones are synthesized in multi-step sequences, e.g.
via g-hydroxycarboxylates, or from aldehydes and acrylates by
3
reductive coupling using superstoichiometric amounts of SmI .
2
In a particularly expedient approach, aliphatic aldehydes are
reacted with malonic acid in a special variant of the Knoevenagel
condensation that leads to the formation of the 3-alkenoic
4
acids instead of the expected a,b-unsaturated isomers. The 3-
alkenoic acids are then cyclized to the corresponding lactones
1,5,6
(
Scheme 1).
A variety of mediators for such intramolecular
5
-exo-trig additions of carboxylate groups to C–C double bonds
7
8
9
are known, among them iodine, salts of silver, thallium, and
palladium, Brønsted acids, and solid acid catalysts.
10
11
12
Scheme 1 Traditional and new synthesis of g-lactones.
a
Fachbereich Chemie, Technische Universit a¨ t Kaiserslautern,
Erwin-Schr o¨ dinger-Straße 54, 67663 Kaiserslautern, Germany.
E-mail: goossen@chemie.uni-kl.de; Fax: +49-631-3921
Scheme
lactonization.
2
Proposed mechanism of the tandem isomerization–
b
Care Chemicals Technology, Cognis GmbH, Henkelstraße 67, 40551
This should take place via a Lewis or Brønsted acid-promoted
carbocation formation (I), followed by a series of 1,2-H-shifts
(II–III). The 4-alkenoic acid isomer (V), activated by catalyst
D u¨ sseldorf, Germany
14
†
Electronic supplementary information (ESI) available: Synthetic meth-
11b
ods. See DOI: 10.1039/b916853b
This journal is © The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 197–200 | 197

View Article Online
a
Table 1 Optimization of catalyst and reaction conditions
◦
b
Entry
Catalyst
TsOH
Mol%
Solvent
T ( C)
Yield (%)
1
2
3
4
5
6
7
8
9
5
5
5
5
5
5
5
5
5
5
5
2.5
5
5
5
5
5
—
—
Toluene
—
—
—
—
—
—
—
—
—
—
NMP
DMSO
160
160
110
160
160
160
160
160
160
160
160
160
160
160
160
160
85
0
0
0
0
0
0
0
4
11
0
21
34
50
0
0
0
0
(PPh
(PPh
3
)
)
3
H
2
(CO)Ruc
3
3
(H)(Cl)Ru
AuCl
AgBF
PdCl
Pd(dba)
Bi(OTf)
3
4
2
2
3
Cu(OTf)
LiOTf
2
1
1
1
1
1
1
1
1
1
1
2
0
1
2
3
4
5
6
7
8
9
0
In(OTf)
Cu (OTf)
2
2
2
·C
6
H
6
AgOTf
AgOTf
AgOTf
AgOTf
AgOTf
AgOTf
AgOTf
AgOTf
Tetraglyme
1,2-Dichloroethane
Dowtherm A
PhCl
5
5
10
160
130
130
50
50
80 (71)
d
PhCl
a
b
c
Reaction conditions: 1.0 mmol of acid 1b, 2 mL of solvent (if used), 24 h. Determined by GC using n-tetradecane as internal standard. Generated
d
3 3 2 4
in situ from (PPh ) RuCl and excess NaBH . Isolated yield.
coordination to the C–C double bond (VI), would be removed
from the resulting dynamic equilibrium of isomers by a fast
in situ 5-exo-trig cyclization of the carboxylate group, giving
the stable lactone ring of product 2. Ideally, the overall process
would lead to a quantitative formation of the desired g-lactone
in one step.
tandem isomerization–lactonization as a modern synthetic tool,
a new process is vital in which only catalytic quantities of a non-
toxic and easy-to-handle mediators are necessary. Our strategy
to develop such a process was to focus the catalyst search
on Lewis-acidic metal cations that have a marked affinity for
carbon–carbon double bonds and a strong potential to effect
isomerizations of olefinic carboxylic acid substrates, and at the
same time to promote the addition of nucleophiles to multiple
bonds. This way, we hoped to establish a rapid equilibration
between all possible positional isomers, so that the selective
trapping of one species by ring closure would lead selectively
to a single product.
However, it has not been possible so far to selectively achieve
isomerizing lactonizations in the absence of stoichiometric
mediators: Previous attempts were plagued by side reactions,
mainly di- and oligomerizations at the double bonds, or
intermolecular additions of carboxylate groups to double bonds.
Angelici et al. reported that when using Amberlyst-15 resin,
these side reactions predominate for substrates that need to
We chose the terminally unsaturated 10-undecenoic acid (1b)
as a first model substrate and performed some preliminary
12
undergo a double-bond shift over five or more carbon atoms.
◦
Isomerizing lactonizations of 9- or 10-alkenoic acids would
require an even larger number of double-bond migration steps
along the alkyl chain before the desired cyclization can take
place, with the potential for side reactions over the entire
course of the isomerization process. As a result, most known
protocols starting from fatty acids involve the use of large
quantities of corrosive and hazardous mediators, and their
yields are low. For example, superstoichiometric amounts of
experiments with neat reaction mixtures at 160 C. We screened
a broad range of Lewis and Brønsted acids as well as metal
hydrides, with a focus on easily handled, commercially avail-
able catalyst precursors. Selected results are summarized in
Table 1.
As expected, simple Brønsted acids (TsOH, HBF
4
,
F
3
CCOOH) were inactive when only used in catalytic amounts
(entry 1). Known isomerization catalysts, i.e. either preformed
(entry 2) or in situ-generated ruthenium hydride complexes
(entry 3), were similarly disappointing. Furthermore, the results
15
16
17
sulfuric, methanesulfonic or even explosive perchloric acid
are required to achieve the isomerizing cyclization of oleic
acid (1a) to the g-lactone 2a. While only perchloric acid gave
satisfactory yields, the use of this unstable, strongly oxidizing
material at high temperatures on large scales represents a major
health and safety hazard, precluding its application in chemical
manufacturing.
obtained with simple Lewis acids (AuCl
3
, AuI, CuF
2
, AgBF ,
4
19
FeCl , FeCl ) (entries 4 and 5) or Pd catalysts (entries 6
2
3
and 7) were unsatisfactory. In contrast, several metal triflates
immediately afforded visible catalytic turnover in the absence of
solvent (entries 8–13). Amongst them, the soft silver(I) triflate
proved to be most effective and gave almost full conversion.
However, only 50% yield was obtained due to oligomerization
reactions.
We became interested in this transformation in the context
of our research on the use of carboxylic acids as substrates
18
in sustainable catalytic transformations. In order to establish
1
98 | Green Chem., 2010, 12, 197–200
This journal is © The Royal Society of Chemistry 2010

View Article Online
a
Table 2 Substrate scope for the catalytic isomerization–lactonization
Entry
Acid
m
n
Mol% cat.
g-Lactone
x
Yield (%)
1
2
3
4
1a
1b
1c
1d
8
0
6
6
7
9
7
3
10
15
15
15
2a
2b
2c
2d
12
5
10
6
51
72
57
66
a
◦
b
Reaction conditions: 1.0 mmol of acid 1, given amount of AgOTf, 2 mL of PhCl, 130 C, 24 h. Isolated yields. Products were characterized by
1
13
means of H-NMR and C-NMR.
In order to drive back these intermolecular processes in favor
of the desired intramolecular lactonization, we sought for a
suitable solvent (entries 14–19). Moderate yields were achieved
yields of 57% and 66%, respectively (Table 2, entries 3
and 4).
With such a viable synthetic entry into g-stearolactone in
place, we are currently exploring the properties of the material
itself along with its use as a synthetic intermediate. Scheme 3
details successful derivatization reactions, giving access e.g. to
◦
when using the non-chlorinated solvent Dowtherm A at 160 C
(
entry 18), but the best chemoselectivity was achieved with
◦
chlorobenzene at the lower temperature of 130 C (its boiling
point). Under these conditions and when stopping the reaction
at incomplete conversion, the formation of unwanted and hard-
to-separate oligomers was suppressed (entry 19).
2
1,22
long-chain g-hydroxycarboxylates (A, B),
g-hydroxyalcohols
23
24
(C), or g-hydroxyamides (D) useful as detergents, as well
25
as sphingolipid-like structures. Saponification of 2a with
potassium hydroxide led to the formation of g-hydroxystearic
acid (3a) in nearly quantitative yield. The g-hydroxyester 3b was
obtained in 69% yield from a transesterification with n-butanol.
The best results (80% yield determined by GC, entry 20) were
finally obtained with a catalyst loading of 10 mol%, and g-
lactone 2b was isolated in 71% yield after column chromatogra-
15b
phy (Table 2, entry 2). In contrast to other known procedures,
Reduction of 2a with ZnCl
2
/NaBH furnished the long-chain
4
the catalytic process was highlyregioselective: We never observed
more than trace amounts of the thermodynamically only slightly
disfavored six-ring lactone.
1,4-diol 3c in 84% yield, and the stearamide 3d was isolated in
71% yield after treating 2a with ethanolamine.
14
After having thus developed an effective protocol that allowed
a double-bond isomerization over five chain positions, we
next approached the even more challenging transformation of
oleic acid (1a) to g-stearolactone (2a). With its 33 possible
double-bond isomers, this substrate can be seen as an ulti-
mate benchmark for tandem isomerization–functionalization
20
processes. Under the optimized conditions we directly obtained
a in 33% yield. With a slightly higher catalyst loading of 15
2
mol%, 2a was isolated in a remarkable 51% yield after column
chromatography (Table 2, entry 1). Again, only traces of the six-
ring lactone were detected, and the main side products resulted
from oligomerization.
The product samples of this non-commercial long-chain
lactone obtained with this protocol displayed interesting prop-
erties for applications in cosmetics. For the preparation of
larger quantities of the g-stearolactone required for further
testing, we reduced the catalyst loading to 10% and increased
Scheme 3 Derivatization reactions of g-stearolactone (2a).
◦
the temperature to 160 C by working at reflux in a 1:6
mixture of chlorobenzene and 1,2-dichlorobenzene. Moreover,
we attempted to substitute the chromatographic purification by
a more scalable process. Distillation proved not to be feasible
as the lactone underwent thermal oligomerization. Instead, we
found that the product could be obtained in high purity by
fractional crystallization from ethanol (see ESI†). This way, we
obtained g-stearolactone on 180 g scale in a yield of 51%.
Using the same procedure, commercially available palmitoleic
acid (1c) and cis-5-dodecenoic acid (1d) were also smoothly
converted into the corresponding long-chain g-lactones in
Conclusions
Overall, an efficient catalytic tandem isomerization–
lactonization protocol has been developed for the preparation
of g-lactones from renewable fatty acids. The products were
obtained in good yields even in the rather extreme test cases
of oleic and palmitoleic acid, where only one of the numerous
possible double bond isomers could give rise to the desired
g-lactone. The products were shown to be useful starting points
for further derivatizations.
This journal is © The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 197–200 | 199

View Article Online
Acknowledgements
13 B. V. Burger, F.-C. Tien, M. Le Roux and W.-P. Mo, J. Chem. Ecol.,
1
996, 22, 739.
We thank Cognis GmbH and NanoKat for financial support.
D.M.O. thanks the Stiftung der Deutschen Wirtschaft for a
scholarship.
14 T. A. Isbell and B. A. Plattner, J. Am. Oil Chem. Soc., 1997, 74, 153.
1
5 (a) A. Granata, F. Sauriol and A. S. Perlin, Can. J. Chem., 1994, 72,
1
684; (b) S. C. Cermak and T. A. Isbell, J. Am. Oil Chem. Soc., 2000,
7, 243.
7
1
1
6 R. Stern and G. Hillion, Fr. Pat. 1987/2623499.
7 J. S. Showell, D. Swern and W. R. Noble, J. Org. Chem., 1968, 33,
2
697.
Notes and references
1
8 (a) L. J. Gooßen, N. Rodr ´ı guez and K. Gooßen, Angew. Chem., Int.
Ed., 2008, 47, 3100; (b) L. J. Gooßen and J. Paetzold, Angew. Chem.,
Int. Ed., 2004, 43, 1095; (c) L. J. Gooßen and A. D o¨ hring, Adv. Synth.
Catal., 2003, 345, 943; (d) L. J. Gooßen, J. Paetzold and L. Winkel,
Synlett, 2002, 1721.
1
D. D. Zope, S. G. Patnekar and V. R. Kanetkar, Flavour Fragrance J.,
006, 21, 395 and references therein.
2
2
(a) J.-A. Hislop, M. B. Hunt, S. Fielder and D. D. Rowan, J. Agric.
Food Chem., 2004, 52, 7075 and references therein; (b) B. Schlutt, N.
Moran, P. Schieberle and T. Hofmann, J. Agric. Food Chem., 2007,
19 for Pd-catalyzed isomerizations, see(a) Y. Wang, X. Dong and R. C.
Larock, J. Org. Chem., 2003, 68, 3090; (b) C. Jim e´ nez-Rodriguez,
G. R. Eastham and D. J. Cole-Hamilton, Inorg. Chem. Commun.,
2005, 8, 878.
5
5, 9634.
3
S. Fukuzawa, A. Nakanishi, T. Fujinami and S. Sakai, J. Chem. Soc.,
Perkin Trans. 1, 1988, 1669.
4
5
6
7
8
9
S. E. Boxer and R. P. Linstead, J. Chem. Soc., 1931, 740.
R. Fittig, Chem. Ber., 1883, 373.
20 In catalytic isomerizing hydroborations, the formation of side
products was substantially higher when using oleates compared
to short-chain substrates: K. Y. Ghebreyessus and R. J. Angelici,
Organometallics, 2006, 25, 3040.
21 P. W. Clutterbuck, J. Chem. Soc., 1924, 125, 2330.
22 T. A. Isbell and B. A. Steiner, J. Am. Oil Chem. Soc., 1998, 75, 63.
23 B. V. S. K. Rao and R. Subbarao, J. Lipid Sci. Technol., 2006, 38,
185.
24 (a) J. K. Well, F. D. Smith and W. M. Linfield, J. Am. Oil Chem. Soc.,
1972, 49, 383; (b) R. Lagerman, S. Clancy, D. Tanner, N. Johnston,
B. Callian and F. Friedli, J. Am. Oil Chem. Soc., 1994, 71, 97.
25 Y. Cai, C.-C. Ling and D. R. Bundle, Org. Biomol. Chem., 2006, 4,
1140.
R. P. Linstead, J. Chem. Soc., 1932, 115.
P. A. Bartlett and J. Myerson, J. Am. Chem. Soc., 1978, 100, 3950.
C.-G. Yang, N. W. Reich, Z. Shi and C. He, Org. Lett., 2005, 7, 4553.
H. M. C. Ferraz and C. M. R. Ribeiro, Synth. Commun., 1992, 22,
3
99.
0 (a) R. C. Larock and T. R. Hightower, J. Org. Chem., 1993, 58, 5298;
b) U. Annby and M. Stenkula, Tetrahedron Lett., 1993, 34, 8545.
1 (a) R. S. Mali and K. N. Babu, Helv. Chim. Acta, 2002, 85, 3525;
1
1
(
(
b) M. F. Ansell and M. H. Palmer, J. Chem. Soc., 1963, 2640; (c) L.
Coulombel and E. Du n˜ ach, Synth. Commun., 2005, 35, 153.
1
2 Y. Zhou, K. Woo and R. Angelici, Appl. Catal., A, 2007, 333, 238.
2
00 | Green Chem., 2010, 12, 197–200
This journal is © The Royal Society of Chemistry 2010