Use of the anti-oxidant butylated hydroxytoluene in situ for the synthesis of readily oxidized compounds: Application to the synthesis of the moth pheromone (Z,Z,Z)-3,6,9-nonadecatriene

Article abstract of DOI:10.1071/CH07223

The triene (Z,Z,Z)-3,6,9-nonadecatriene was synthesized in three steps from methyl linolenate. The key to the synthesis was the use of the anti-oxidant butylated hydroxytoluene in situ to provide protection of the unstable triene from autoxidation during reaction workup. This simple modification resulted in an increase in the yield from 20 to 85% over three steps. CSIRO 2007.

Full text of DOI:10.1071/CH07223

Rapid Communication
CSIRO PUBLISHING
Aust. J. Chem. 2007, 60, 848–849
www.publish.csiro.au/journals/ajc
Use of the Anti-Oxidant Butylated Hydroxytoluene
in situ for the Synthesis of Readily Oxidized Compounds:
Application to the Synthesis of the Moth Pheromone
(Z,Z,Z)-3,6,9-Nonadecatriene
Noel W. Davies,A Graham Meredith,B Peter P. Molesworth,B
and Jason A. SmithB,C
ACentral Science Laboratory, Private Bag 74, University of Tasmania, Hobart TAS 7001, Australia.
BSchool of Chemistry, Private Bag 75, University of Tasmania, Hobart TAS 7001, Australia.
CCorresponding author. Email: jason.smith@utas.edu.au
The triene (Z,Z,Z)-3,6,9-nonadecatriene was synthesized in three steps from methyl linolenate.The key to the synthesis was
the use of the anti-oxidant butylated hydroxytoluene in situ to provide protection of the unstable triene from autoxidation
during reaction workup. This simple modification resulted in an increase in the yield from 20 to 85% over three steps.
Manuscript received: 27 June 2007.
Final version: 15 August 2007.
O
(i)
The C19 triene (Z,Z,Z)-3,6,9-nonadecatriene 1 has been iden-
tified as a pheromone for the autumn gum moth Mnesampela
OH
OCH3
2
1
3
4
privata, which is a threat to blue gum plantations.[1] To collect
moths to study their migration pattern, baiting traps with a suit-
able attractant, such as the appropriate chemical sex attractant,
have proved to be a valuable tool.[1] They have also been used to
control moth populations by disruption of the mating cycle.[2]
However, the high cost of synthesis and low yield of triene 1
obtained in the most recent synthesis,[1] 18%, make large-scale
trapping expensive and impractical.[3] The cost of using pure
linolenic acid could be alleviated by using less expensive ∼70%
methyl linolenate obtained from the transesterification of linseed
oil. The problem with the latter is that it is contaminated by the
diene methyl linoleate, which may interfere with the pheromone
response of the triene.To overcome these problems, we set about
investigating the use of an organocuprate that has been reported
previously, but for the synthesis of pheromone cocktails[3] from
mixtures of linolenic, linoleic, and other long-chain fatty esters.
To test if the method was suitable when using a pure or enriched
triene, we started with a sample of 95% pure methyl linolenate
(as determined by gas chromatography/mass spectrometry (GC/
MS) analysis), which was obtained by the transesterification of
linseed oil and then purified by a previously reported method,[4]
before using the more expensive ≥99% pure material.
(ii)
(iii)
OTs
Scheme 1. (i) LiAlH4, THF, 0◦C to room temp.; (ii) p-toluenesulfonyl
chloride, pyridine, CH2Cl2, 0◦C; (iii) (CH3)2CuLi, THF, 0◦C.
It was obvious that the low yield and subsequent decrease of the
triene to diene ratio was due to autoxidation of the compounds
by adventitious oxygen. As the samples were stored refriger-
ated under an atmosphere of nitrogen, it was assumed that the
decomposition was occurring during reaction workup and purifi-
cation as silica chromatography was required to remove polar
decomposition products. The use of anti-oxidants, such as buty-
lated hydroxytoluene (BHT) or 2,6-di-t-butyl-4-methylphenol,
to inhibit autoxidation of readily oxidized compounds, such
as methyl linolenate, is well established[6] but its use in situ
during a reaction has been typically associated with inhibiting
polymerization of alkenes under thermal conditions such as in
the Diels–Alder reaction.[7] We hypothesized that, provided the
anti-oxidant is not consumed or transformed under the reac-
tion conditions employed in a synthesis, it could be added to
the methyl linolenate and carried through the synthesis, provid-
ing the required protection from undesired autoxidation. Owing
to the synthetic sequence employed for the transformation of
methyl linolenate, the anti-oxidant BHT was chosen as it would
be regenerated on workup after reaction with lithium aluminium
hydride and would not affect the tosylation or cuprate reactions.
Thesyntheticplanwasthentotakethe95%enrichedlinolenic
ester 2, which was reduced with lithium aluminium hydride to
give the alcohol 3 and which was then transformed to the tosy-
late 4 (Scheme 1). The final step involved the displacement of
the tosylate with dimethyl cuprate,[5] which gave the C19 triene.
However, whereas each step occurred cleanly to give the desired
compound, the overall yield was low, ∼10–20%, and the purity
of the triene had decreased significantly to ∼80% by GC/MS
analysis. This was disappointing as there was no improvement
of the overall yield compared with the method of Steinbauer.[1]
© CSIRO 2007
10.1071/CH07223
0004-9425/07/110848

Synthesis of (Z,Z,Z)-3,6,9-Nonadecatriene
849
Therefore, the modified synthesis started with a small amount
(1%) of BHT (2,6-di-t-butyl-4-hydroxytoluene) added to the
95% methyl linolenate, which was then subjected to the reaction
sequence discussed above. Reaction of the ester with lithium
aluminium hydride gave the alcohol, still containing the BHT,
in ∼80% while tosylation proceed similarly. As predicted, the
removal of BHT before reaction was not required and therefore
simplified the method while maintaining the required protective
properties. The final cuprate step also proceeded in the presence
of BHT to yield the desired triene in quantitative yield with-
out need for chromatography at any step. This was surprising
as the tosylate was contaminated by traces of tosyl chloride, yet
no traces of it or any by-products were observed on workup.
The addition of BHT increased the product yield significantly
from 10–20 to 65% over the three steps while maintaining the
purity of the triene at ∼95% (GC). This was a pleasing result
and now meant that beginning with the more expensive starting
material would not result in unnecessary loss due to decompo-
sition during workup. As such, the reaction of the ≥99% methyl
linolenate[8] containing 1% of BHT gave 0.947 g of the triene
1 with no detectable C19 contaminants in 85% overall yield for
the three steps.
an atmosphere of nitrogen with protection from light, and the
reaction mixture stirred at ambient temperature for 24 h. The
reaction mixture was diluted with dichloromethane (50 mL) and
quenched with 2 M HCl (30 mL). The organic phase was sepa-
rated, washed with 2 M Na2CO3 (30 mL), dried over magnesium
sulfate, filtered, and the solvent removed under reduced pressure
to yield the tosylate 4 (3.43 g, 95%) as a pale yellow oil. δH 7.79
(2H, d, J 8.7), 7.34 (2H, d, J 8.7), 5.35 (6H, m), 4.01 (2H, t, J
6.6), 2.80 (4H, m), 2.45 (3H, s), 2.05 (4H, m), 1.83 (2H, m), 1.25
(10H, m), 0.97(3H, t, J7.5). δC 132.2, 130.5, 130.0, 128.5, 128.4,
128.2, 128.1, 127.9, 127.32, 127.29, 70.9, 30.5, 29.8, 29.5, 29.3,
29.1, 29.0, 27.4, 25.8, 25.7, 25.5, 21.8, 14.5.
(Z,Z,Z)-3,6,9-Nonadecatriene 1
Me2CuLi (12.7 mmol) was prepared in situ by dropwise addition
of methyllithium (18.2 mL, 1.4 M in diethyl ether) to a stirred
solution of cuprous iodide (2.41 g, 12.7 mmol) in ether (55 mL)
at 0◦C under an atmosphere of nitrogen, yielding a clear near-
colourless solution that was stirred for 30 min. A solution of the
tosylate 4 containing BHT (1.60 g, 3.84 mmol in ether, 77 mL)
was added dropwise to the cuprate and the resultant solution
stirred for 3.5 h. The yellow solution was quenched with satu-
rated ammonium chloride and the product extracted with ether
(3 × 80 mL). The combined organic layers were washed with
brine, dried with magnesium sulfate, filtered, and the solvent
removed under reduced pressure, yielding the desired product 1
as a pale oil (0.947 g) in 94% yield. The product still contained
∼1% BHT by NMR spectroscopy. δH 5.38 (6H, m), 2.80 (4H,
m), 2.15 (4H, m), 1.27 (14H, m), 0.98 (3H, t, J 7.5), 0.88 (3H,
m). δC 132.2, 130.7, 128.5, 128.4, 127.8, 127.3, 32.1, 30.5, 29.8,
29.8, 29.5, 27.4, 25.8, 25.7, 22.9, 21.4, 20.7, 14.5, 14.3.
In conclusion, not only have we developed a reliable and
efficient synthesis of these unstable long-chain unsaturated
hydrocarbons, but we have also shown that anti-oxidants can
be used in situ to provide unstable compounds protection from
autoxidation during workup.
Experimental
General Experimental
1H and 13C NMR spectra were recorded at 300 MHz and
75 MHz, respectively, on a Varian Mercury 2000 spectrometer.
Spectra were run in CDCl3 unless otherwise stated. Chemical
shifts are measured in ppm and referenced internally to residual
CHCl3.
Acknowledgements
P. P. M. is thankful to the University of Tasmania and the Thomas Crawford
Foundation for a postgraduate scholarship.
References
Linolenyl Alcohol 3
[1] M. J. Steinbauer, F. Östrand, T. E. Bellas, A. Nilsson, F. Andersson,
E. Hedenström, M. J. Lacey, F. P. Schiestl, Chemoecology 2004, 14,
217. doi:10.1007/S00049-004-0281-5
[2] R. T. Carde, A. K. Minks, Annu. Rev. Entomol. 1995, 40, 559.
doi:10.1146/ANNUREV.EN.40.010195.003015
[3] (a) H. Yu, H. K. Mangold, Chem. Ind. 1988, 787.
(b) H. Yu, H. Becker, H. K. Mangold, Chem. Ind. 1989, 836.
[4] J. D. Nadenicek, O. S. Privett, Chem. Phys. Lipids 1968, 2, 409.
doi:10.1016/0009-3084(68)90014-5
[5] (a) C. R. Johnson, G. A. Dutra, J. Am. Chem. Soc. 1973, 95, 777.
(b) W. Huang, S. P. Pulaski, J. Meinwald, J. Org. Chem. 1983, 48, 2270.
doi:10.1021/JO00161A025
A solution of the ≥99% methyl ester 2 (2.50 g, 8.56 mmol)
containing 1% BHT in anhydrousTHF (60 mL) was added drop-
wise to a suspension of lithium aluminium hydride (430 mg,
11.3 mmol) in anhydrousTHF (30 mL) at −5◦C under a nitrogen
atmosphere.Afteradditionwascomplete, themixturewasstirred
at room temperature before two further portions of lithium alu-
minium hydride (425 mg, 11.2 mmol; and 435 mg, 11.5 mmol)
were added after 2 h and 4 h.After 6 h, the reaction was quenched
by the addition of methanol and the lithium salts were dissolved
by the addition of saturated KHSO4 (100 mL) before extraction
with diethyl ether (3 × 80 mL). The combined organic extracts
were washed with brine, then dried with sodium sulfate, filtered,
and the solvent removed under reduced pressure yielding the
alcohol 3 (2.171 g, 96%) as a pale oil. δH 5.38 (6H, m), 3.64 (2H,
t, J 6.9), 2.80 (4H, m), 2.08 (4H, m), 1.25–1.60 (12H, m), 0.97
(3H, t, J 7.5). δC 132.2, 130.6, 128.5, 127.9, 127.3, 125.8, 63.3,
33.0, 30.5, 29.9, 29.7, 29.6, 29.5, 27.5, 25.9, 25.8, 25.7, 14.5.
[6] K. E. Peers, D. T. Coxon, H. W. S. Chan, J. Sci. Food Agric. 1984, 35,
813. doi:10.1002/JSFA.2740350720
[7] (a) L. Evanno, A. Deville, L. Dubost, A. Chiaroni, B. Bodo, B. Nay,
Tetrahedron Lett. 2007, 48, 2893. doi:10.1016/J.TETLET.2007.02.089
(b) E. L. Pearson, L. C. H. Kwan, C. I. Turner, G. A. Jones, A. C. Willis,
M. N. Paddon-Row, M. S. Sherburn, J. Org. Chem. 2006, 71, 6099.
doi:10.1021/JO0607818
[8] At the time of the present study, ≥99% methyl linolenate from Sigma-
Aldrich Australia was A$151 per gram.
Linolenyl Tosylate 4
p-Toluenesulfonyl chloride (2.04 g, 10.7 mmol) was added to
a solution of the alcohol 2 (2.171 g, 8.21 mmol), and pyri-
dine (1.00 mL) in anhydrous dichloromethane (3 mL) under