70 ¡C. After a warm-up period of exactly 3 min, recording of
the gaseous product evolved was started with the aid of a
syringe gas buret. The gas measuring device automatically
maintained the inside pressure at the atmospheric value. After
a reaction time of 120 min, the stopcock was closed and the
reaction vessel was cooled Ðrst to 6È8 ¡C and then to liquid
nitrogen temperature. When the reaction mixture was frozen,
a 5.00 cm3 gas sample was taken with a pressure-lok syringe;
after thawing, the concentration of t-BHP remaining in the
solution was determined by iodimetry.
products which remained dissolved in the chlorobenzene solu-
tion were determined by gas chromatography.
In the presence of 0.120 mmol of Hex NCl during 280 min,
4
the decomposition of 2.108 mmol of t-BHP in 8.00 cm3 of
chlorobenzene produced 0.687 mmol of O , which corre-
2
sponds to 1.374 mmol of t-BHP; *[t-BHP] was found to be
1.376 mmol. Additionally, 1.329 mmol of tert-butanol (96.6%)
and 0.0406 mmol di-tert-butyl peroxide (5.9%) were found by
gas chromatography. These data indicate that only O , tert-
2
butanol and some di-tert-butyl peroxide are formed during
The quantity of O evolved was estimated by a modiÐed
the decomposition.
In the case of Hex NBr as catalyst under similar conditions,
0.851 mmol of t-BHP was decomposed and produced 0.424
2
Winkler method.9 After O traces has been removed by thor-
2
4
ough Ñushing with N , 3.0 cm3 of concentrated NaOH (20
2
mmol) and 8.0 cm3 of manganese(II) chloride (9 mmol) solu-
tions (both previously freed from O by an N stream) were
introduced into the reaction vessel whilst the O -free environ-
mmol of O (which is equivalent to 0.848 mmol of t-BHP).
2
Moreover, 0.795 mmol of tert-butanol (93.4%) and 0.0275
2
2
2
mmol of di-tert-butyl peroxide (6.4%) were found, which is in
good harmony with the stoichiometric factors given in Table
2.
ment was secured with a stream of N . A white precipitate of
2
Mn(OH) was formed in the reaction vessel. The inside pres-
2
sure was next reduced to 5È10 Torr, the previously taken 5.00
cm3 gas sample was introduced through the septum and O
E†ects of onium iodides
2
was absorbed quantitatively by thorough stirring and agita-
tion for 30 min. A 15 cm3 volume of 20% hydrochloric acid
was then added to dissolve the hydroxide precipitate, about 1
g of KI was added and the liberated iodine was titrated with
thiosulfate reagent solution.
The other products of decomposition were analysed by gas
chromatography on a column (2 m ] 4 mm id) Ðlled with
Chromosorb W-AW-DMCS coated with Carbowax 20M with
When the decomposition of t-BHP was carried out as
described in the Experimental section with Hex NI as cata-
4
lyst, di†erent results were obtained (Table 3). Rapid decompo-
sition was observed, but only a little gaseous product was
obtained, and the value of the stoichiometric factor was conse-
quently well in excess of 2. The reason for this can be under-
stood by considering the data from those measurements in
which the quantity of catalyst was altered. These indicate that
the decomposition is too fast relative to the resolving ability
of the gas measuring device used. When the catalyst was
added to the closed reactor without waiting 3 min for
warming up, during which the stopcock of the reactor was
open to the atmosphere, the decomposition took place with a
stoichiometric factor of 2. The volume increments caused by
the temperature increase were taken into consideration.
N as carrier gas at 40 cm3 min~1, Ñame ionization detection
2
and identiÐcation with the aid of authentic samples. For
control, MS-10 and QMS 200 mass spectrometers were
applied.
Results and discussion
E†ect of water
In the presence of 0.120 mmol of Hex NI as catalyst during
4
120 min, 2.045 mmol of t-BHP were decomposed (out of 2.108
It was observed earlier6,8 that water exerts a catalytic inÑu-
ence on the oxidation of hydrocarbons in the presence of
onium salts. It seemed reasonable, therefore, to investigate
whether the decomposition of t-BHP catalysed by onium salts
is similarly inÑuenced by the presence of water. The data in
Table 1 show that the decomposition of t-BHP in the presence
of a cationic phase-transfer reagent is enhanced by the addi-
tion of water up to a given quantity; but at higher water con-
centrations the rate of the decomposition decreases. This
observation allows the conclusion that water contributes to
the catalyzed decomposition of the initiator.
mmol) and this furnished 1.022 mmol of O (equivalent to
2
2.044 mmol of t-BHP, 99.9%) and 1.831 mmol of tert-butanol
(89.5%) and 0.108 mmol of di-tert-butyl peroxide (10.6%). It
may therefore be concluded that the decomposition of t-BHP
is catalyzed analogously by all cationic PTCs, independently
of the nature of the counteranions, although the decomposi-
tions rates di†er considerably. All these interactions produce
O , tert-butanol and some di-tert-butyl peroxide as the main
2
products. Mass spectrometry furnished similar results, but
also indicated the formation of traces of CO .
2
Limited decomposition of t-BHP can be induced by the
presence of NaI as well as by Me NI, again with a stoichio-
E†ects of counteranions on the decomposition of t-BHP
4
metric factor of 2. Interestingly, Me NI proved to be a very
4
Table 2 presents data relating to the decomposition of t-BHP.
In the absence of onium salts, only a very slight decomposi-
tion can be observed at 70 ¡C during 120 min, and a small
amount of gaseous product (gp) is formed with the stoichiom-
etry *[t-BHP]/*[gp] \ 2. The same stoichiometry was found
when di†erent cationic phase-transfer reagents were applied
with di†erent counteranions, with the exception of iodide. The
catalytic activities of the onium salts di†ered and depended on
the nature of the counteranion. For tetrabutylammonium
cation, for instance, the sequence of efficiency was Br~ [
SCN~ [ Cl~ [ NO ~ [ ClO ~ [ F~, whereas for the tetra-
poor catalyst in comparison with the other quaternary
ammonium ions containing longer aliphatic chains.
Since a small quantity of iodine was formed on mixing
during the application of onium iodides (except for the spar-
ingly soluble Me NI!), the inÑuence of iodine was also investi-
4
gated separately. In the lower half of Table 3, in addition to
0.120 mmol of Hex NX (X \ Cl~, Br~ or I~), 0.019 or 0.118
4
mmol of iodine was also applied in the samples. The iodine
was homogeneously dissolved with a reddish violet color in
chlorobenzene. In the presence of iodine, the rate of t-BHP
decomposition catalyzed by chloride and bromide salts was
enhanced, and larger amounts of gaseous products were
formed, but the ratio *[t-BHP]/*[gaseous product] remained
unchanged at 2.
3
4
hexylammonium cation it was Br~ [ Cl~ [ HSO ~ [ OH~.
4
The sequence of catalytic efficiencies of di†erent quaternary
ammonium ions, all with chloride counteranion, was
Hex N` [ BzBu N` [ BzEt N` [ MeOct N` [ Me N` [
In the presence of the iodide salt, however, di†erent behav-
4
3
3
3
4
(CetPy`).
ior was observed. At 0.008 mmol Hex NI, the decomposition
4
The gaseous product of the catalyzed decomposition of
of t-BHP took place almost quantitatively in 10È15 min, with
t-BHP overwhelmingly contained O in each case, indepen-
dently of the nature of the counteranion. The decomposition
a stoichiometric factor of 2. When both iodide salt and iodine
were applied, the rate of decomposition of t-BHP was reduced
2
3802
Phys. Chem. Chem. Phys., 2000, 2, 3801È3805