W. G. Jary et al.
an active catalyst for the conversion of hydrogen peroxide into singlet oxygen. This
method was also developed into an industrial scale application [8, 9].
In 1961, Thompson reported the observation of labile intermediates in the low-
temperature ozonation of triphenylphosphite, which has oxidizing powers and
ꢂ
decomposes above ꢁ30 C with evolution of oxygen and formation of triphenyl-
phosphate [10]. Later on Murray et al. have described phosphites as possible
auxiliaries for the chemical generation of singlet oxygen [11]. It has been found
that phosphite ozonides, formed by low-temperature reaction of phosphites with
ozone, can be cleaved into phosphate and singlet oxygen.
Bartlett et al. [12, 13] have first derived a dual mechanism for the oxidation
reactions of the triphenylphosphite-ozone adducts, one at temperatures below
which it decomposes and the other at its decomposition temperature [14, 15].
According to this work and subsequent publications, a direct reaction between
the triphenylphosphite-ozone adduct and substrate singlet oxygen seems possible.
It seemed to us of interest to follow these general lines to develop a novel
catalyst system for chemically generated singlet oxygen from ozone, which by
means of its physical properties allows an immediate process control detection.
Results and Discussions
It is known from the literature that the reaction of ferrocene with ozone results in
an oxidative decomposition of the Fe–C bond [16]. In contrast to these results
0
preliminary experiments proved that substituted ferrocenes, such as 1,1 -bis(diphe-
nylphosphino)ferrocene (1), colored yellow to orange, are stable towards ozone and
can even generate singlet oxygen [17]. Therefore, the singlet oxygen activity of 1
was determined by luminescence and trapping experiments. According to the lumi-
nescence signal and the ozone uptake observed in this experiment, a stoichometric
conversion of 1 with ozone gave singlet oxygen. The reference used was the 100%
singlet oxygen yield of ozone=triphenylphosphite. The luminescence experiments
proved that generation of singlet oxygen by the reaction of ozone with 1 was
possible in principle. Based on the total ozone uptake of two equivalents ozone
1
the efficiency of singlet oxygen formation (mol O =mol O ) amounted to 84%.
2
3
This high efficiency of singlet oxygen formation can be attributed on the one hand
to the reaction of 1 with ozone to form the ozonide 2, and on the other to a
decomposition of 2 or a peroxidic intermediate.
ꢂ
Therefore an experiment was carried out, ozonising 1 at ꢁ20 C while adding
continuously the trap ꢀ-terpinene. The slow addition of the trap was necessary in
order to favor the reaction of ozone with 1 and suppress a direct oxidation of the
organic substrate. Comparative experiments, ozonizing equal amounts of 1 and ꢀ-
terpinene at the same time in a ‘‘one-pot reaction’’ proved that a considerable
amount of ozone reacted directly with the trap by cleaving it according to a Criegee
mechanism. The reaction proceeds at first with the formation of the phosphite
ozonide 2, which then decomposes, even at low temperature, into singlet oxygen
and the phosphate 3.
For the preparative use of the ozone=1 system one has to take into account that
most substrates for singlet oxygen are easily attacked by ozone. To avoid this side
reaction of ozone with the substrate one way is using the primary reaction of 1 with
ozone at low temperatures, stripping off surplus ozone, treating the solution with the