- PART I. A chemical examination of the materials toxic to fish in
kraft pulp mill wastes has been made as part of a water pollution
study. Thirty gallons of "foul condensate" from a typical kraft mill
gave 14 g of organic material when extracted with methylene chloride,
and an additional 10 g was obtained when the residue was extracted
with diethyl ether, Each extract was separated into approximately
20 fractions by column chromatography over silica gel. The fractions
were tested for toxicity to guppies, and the most toxic fractions
were then studied in some detail.
Four compounds were identified with certainty: guaiacol (13.2% of the total organic material), cis-tenpin hydrate (1.4%), α-terpineol
(1.1%), and 4-(p-tolyl)-1-pentanol (8.0%). Two other compounds were
tentatively identified as ethyl guaiacol (2.9%) and trans-sylveterpin
(0.2%). The remainder of the organic material contained two major
components and an extensive series of minor constituents. Both
major components were relatively low in toxicity but were nevertheless
examined in some detail.
A ketone, = 1720 cm⁻¹, constituted about 26.6% of the isolated
material. It was converted to two derivatives, both oils. The suspicion
that this ketone was not pure was confirmed by mass
spectrometry since early scans on a sample passed to the ionizing
chamber by a molecular leak differed significantly from later scans.
All attempts to separate this mixture failed.
The most interesting substance was a keto-alcohol that was
present to the extent of 8.7%. While not highly toxic itself, this keto-alcohol
could be reduced to a diol which appears to be the most toxic
compound to fish we have studied. The diol seems to be present in
the waste in only minor amounts, but could account for a considerable
fraction of the total toxicity. The keto-alcohol does not produce a
single solid derivative. It appears to be uniform, according to all
tests we could apply. It is optically active and has three CHCH₃
groups present; the composition C₁₅H₂₈O₂ appears most probable.
Apparently it is derived from a monocyclic sesquiterpene, but to date
we have not been able to degrade it to a known substance. PART II. Thermal reaction of 1, 2-divinylcyclohexane-1,2-diol was
found to produce 1(7)-bicyclo[5.3.0] decen-2-one in good yield. This
is the aldol product derived from 1, 6-cyclodecanedione, the product
expected from direct ketonization of the initial Cope rearrangement
process. Since such 1, 3-hydrogen shifts are forbidden in the
Woodward-Hoffmann theory it seemed of some importance to learn
more about the mechanism of this reaction.
Thus 1, 2-dideuteroxy-1, 2-divinylcyclohexane was vaporized
into a nitrogen stream at low pressure and passed through a reactor
at 350° C. The 1(7)-bicyclo[5.3.0] decen-2-one was trapped and
purified and its deuterium content was determined by mass
spectrometry. The unsaturated ketone was reduced by lithium in
liquid ammonia and the saturated ketone was treated with methoxide
ion in methanol. Analysis for deuterium content was again made by
The unsaturated ketone contained 29 ± 6% D₀, 39 ± 2% D₁,
23 ± 3% D₂, and 7 ± 1% D₃ species. Each of these had the following
percent of alpha deuterium: D₁ 3 ± 3%, D₂ 14 ± 1%, and D₃
75 ± 12%.
We assume that the aldol condensation involves a mono-enol monoketo
form as the reactant. To account for the data obtained in this
study it is necessary to postulate that a series of rapid enol-keto interconversions
follow the initial Cope rearrangement. The series, is
terminated either by the intramolecular aldol condensation or the
dehydration to the unsaturated ketone.