otto had some 50 ml of anise oil in the fridge(1). encouraged by the recent posts about PMMA he decided to give it a try.
GC of the anise oil showed three peaks. one of these is about 90area% of all peaks. otto assumes that this is anethole. on TLC there was only one spot visible under UV-
light as well as with permanganate (RF 0.8 - 0.9 with Hexane/EtOAc 4:1).
the oil was subjected as is to a buffered performic using the following proportions(2):
50 ml anise oil
100 ml DCM
15 g NaHCO3
this was put together with a stirring bar into a 500 ml erlenmeyer flask. then
45 ml 30% H2O2 (~ 50 g)
70 ml 85% HCOOH
where mixed and set aside for 1 hour. the performic acid was added then slowly and under stirring to the mixture in the erlenmeyer flask so that the DCM doesn't boil. the stirring was continued 17 hours at r.t. (flask losely stoppered).
the DCM layer was separated, the aqueous layer extracted twice with 50 ml DCM and again twice with 50 ml EtOAc. the evaporated extracts show 3 major peaks (very close to each other)(3) in GC. also there are minor peaks of anethole, PMP2P (already!) and of one of the other constituents of the oil (one of the three peaks from first GC), which apparently remained unchanged(4).
the residue was transferred into a 1 l RBF, 100 ml methanol and 300 ml H2SO4 were added and the flask was put aside for 1 week (otto had no time). the organics formed a bottom layer. GC of this layer showed 2 major peaks totally different from those of the raw performic product(5). and: the area of the PMP2P - peak was increased (comparison
to anethole and the other oil constituent)!
otto thought, that the rearrangement can take place at lower temp than 80°C or reflux, so he heated the mixture to 65°C for 30 minutes under stirring to see the resulting GC.
it worked. area of PMP2P-peak is now about 1/3 of area of the two diol-peaks. some peaks of heavier products occur (6).
2 hours at 80°C converted almost completely to the ketone. almost means, there are still 5 - 10% diol unchanged (7).
the mixture was cooled, the organic phase separated, the aqueous layer extracted 3 times with 50 ml EtOAc and the combined organic phases were dried with MgSO4 and evaporated in vacuo. of the product obtained, about 70% is PMP2P (
.
vacuum destillation yielded two fractions. first fraction (pale yellow) was collected at 63-63°C at 1.7 mbar (9), total 5.5 ml. second fraction (pale yellow) was collected at 70°C at 0.8 mbar, total 35.5 g.
remainder was a thick black oil, total 5-10 ml.
GC of each fraction tells:
fraction 1: about 50% unchanged constituent of the essential oil, 20% anethole, 20% PMP2P and some minor peaks.
fraction 2: 98% PMP2P.
residue: about 5% PMP2P and a lot of other compounds but no diol(10).
isolated product: 35.5 g (71%) of a pale yellow liquid. bp 70°C at 0.8 mbar. TLC RF = 0.3 eluent Hexane/ethyl acetate 4:1 (11).
marks(1) according to literature anethole melts somewhere around r.t. it would not surprise otto, if this can be used for its isolation from the essential oil. see also the thread
can't get safrole?-FREEZE IT in the newbee forum.
(2) the proportions were taken from rhodiums site and only scaled down.
(3) none of which was the epoxide, as confirmed by an external sample.
(4) this other constituent of the anise oil has a lower bp than anethole. on some websites it is stated that it would be methyl chavicol (the p-MeOallylbenzene).
however, it is not affected by the performic.
(5) apparently either of these peaks is from diol since the only rearrangement product is PMP2P. in the literature there are described 2 different diols (the alpha and beta - isomer) having different m.p.
(6) these peaks of heavy molecules are then found in the residue of vac-dest. the PMP2P peak has increased
to 1/3 area of the diol peaks. no quantitative conclusion can be drawn from here since PMP2P was present before.
(7) in the attached reflux condenser there was little condensate. otto put it into GC, too.
it was mainly methylformate.
(
as in most cases, otto assumes, that area ratios in GC correspond to weight ratios, which is not always the case. in other words: otto hopes that 1 g of PMP2P gives the same peak area as 1 g of any other compound. so the 70 % are not a super exact value.
(9) at such low pressures destillation takes a much longer time than normal.
(10) the flask could be easily cleaned using ethanol.
(11) anethole is very active in UV. TLC is thus a good proof of proper separation.
RF: anethole 0.8 - 0.9, PMP2P 0.3 in Hexane/EtOAc 4:1.
conclusions about 70% of
isolated yield have been obtained by otto which is in good correlation to other reports at the hive. it seems that a 70 - 80% yield range is general for this technique as stated many times before.
from the remaining 30% a few yield% of PMP2P are in the other fractions and the rest formes heavier products. these products resulted mainly from the rearrangement.
so, IOO (in otto's opinion) rearrangement is the yield limiting step in buffered performic.
this procedure works fine with raw anise oil. the non reacting constituent has to be removed at some stage though.
17 hours were not sufficient for full consumption of all anethole used.
two diol isomers have been formed.
the rearrangement
can be run at lower temp.
2 hours @ 80°C left some 10% of diol unchanged.
some of the methanol added is converted to methylformate.
questions / outlookare there higher yielding rearrangement proc's? what is the methanol added for and is it necessary at all?
the anethole should be isolated by freezing prior to reacting it. TLC should be used in the performic (disappearance of anethole) step to check for full conversion. time for rearrangement should be increased a little.
otto