Mass spectrometry and electron paramagnetic resonance study of free radicals spontaneously formed in nitrone-peracid reactions

Article abstract of DOI:10.1021/jo952031w

Reactions of spin traps (C-phenyl N-tert-butyl nitrone (PBN) and 5,5-dimethyl-2-phenyl-1-pyrroline N-oxide (2-Ph-DMPO)) with peracids have been investigated by both mass spectrometry (MS) and electron paramagnetic resonance (EPR). The peracids m-chloroperbenzoic acid, perbenzoic acid, and perpropionic acid, which can be considered models of biological peracids produced during lipid peroxidation, were found to react with spin traps to spontaneously produce significant amount of aminoxyl radicals. The radical products, as well as the nonradical products were detected and their structures identified by EPR and/or MS. Mechanisms for the formation of these products are proposed.

Full text of DOI:10.1021/jo952031w

2358
J . Org. Chem. 1996, 61, 2358-2363
Ma ss Sp ectr om etr y a n d Electr on P a r a m a gn etic Reson a n ce
Stu d y of F r ee Ra d ica ls Sp on ta n eou sly F or m ed in
Nitr on e-P er a cid Rea ction s
Hong Sang, Edward G. J anzen,* and Brian H. Lewis†
National Biomedical Center for Spin Trapping and Free Radicals, Free Radical Biology and Aging
Research Program, Oklahoma Medical Research Foundation, 825 N.E. 13th Street,
Oklahoma City, Oklahoma 73104
Received November 16, 1995X
Reactions of spin traps (C-phenyl N-tert-butyl nitrone (PBN) and 5,5-dimethyl-2-phenyl-1-pyrroline
N-oxide (2-Ph-DMPO)) with peracids have been investigated by both mass spectrometry (MS) and
electron paramagnetic resonance (EPR). The peracids m-chloroperbenzoic acid, perbenzoic acid,
and perpropionic acid, which can be considered models of biological peracids produced during lipid
peroxidation, were found to react with spin traps to spontaneously produce significant amount of
aminoxyl radicals. The radical products, as well as the nonradical products were detected and
their structures identified by EPR and/or MS. Mechanisms for the formation of these products
are proposed.
Sch em e 1
In tr od u ction
C-Phenyl N-tert-butyl nitrone (PBN) is known to
produce aminoxyl (nitroxide) spin adducts from the
addition reaction with reactive free radicals.1-3 Since
almost all reactive radicals are trapped by PBN it has
been assumed that this compound should be a good
inhibitor for lipid peroxidation. The complete absence
in the literature of reports on this topic must mean
that this is not true. In fact in a homogeneous chemi-
cal lipid peroxidation system, PBN has no inhibitory
effect whatsoever.4 And yet we know that the peroxyl
radical intermediates in this system are trapped by
PBN.5 How can inhibition be lacking if radical trapping
occurs?
One explanation is that the peroxyl radical spin adduct
(e.g. first molecule in Scheme 1) is itself unstable and
decomposes to give more free reactive radicals. We have
provided evidence for this intrinsic instability of peroxyl
adducts by studying the products of 2-cyanoprop-2-yl
radicals in the presence of oxygen.6 Indeed the 2-cyano-
prop-2-ylperoxyl radical spin adduct of PBN produces
2-cyanoprop-2-yloxyl radicals presumably by two-bond
cleavage. Some of these radicals may be quenched in the
solvent cage by reacting with the benzaldehyde formed
in the decomposition presumably by cleavage of two
bonds, since the combination of the benzoyl radical with
2-methyl-2-nitrosopropane (MNP) produces the well-
known benzoyl tert-butyl aminoxyl or PBNOX. Also some
2-cyanoprop-2-yloxyl radicals are trapped by MNP and
the resulting alkoxyl tert-butyl aminoxyl is detected by
EPR (Scheme 1).
In rat liver microsomal dispersions, lipid peroxidation
is inhibited by PBN but not very efficiently. In a
preliminary study we have reported that millimolar
amounts of PBN are necessary for inhibition to be
detected.7 Clearly a very poor understanding exists for
the inhibitory effect, or lack thereof, of PBN and other
nitrones of this type. Further studies are underway in
these laboratories on this topic.
In this connection we have considered the possibility
that peracids themselves may react with nitrones spon-
taneously to produce reactive free radicals. Since per-
acids are not free radicals this reaction would be a case
of nonradical molecular combination of peracid with a
nitrone to produce radicals: a kind of molecule-induced
free radical formation.8 Since aldehydes produce peracids
in the presence of oxygen or with some other oxidizing
agents, any lipid peroxidation process producing alde-
hydes is also unavoidably producing peracids. The
question we wanted to answer was, do these peracids in
biological systems spontaneously produce reactive free
radicals with PBN?
* To whom correspondence should be addressed. Tel. 405-271-7570;
Fax 405-271-3980. Alternate address: Departments of Clinical Studies
and Biomedical Sciences, Ontario Veterinary College, University of
Guelph, Guelph, Ontario, N1G2W1, Canada.
† OMRF Fleming Scholar, 1995 summer.
X Abstract published in Advance ACS Abstracts, March 1, 1996.
(1) J anzen, E. G. Acc. Chem. Res. 1971, 4, 31.
We have reported a preliminary study on this point
using commercially available m-chloroperbenzoic acid
(CPBA) as a model peracid.9 Based on EPR results alone,
(2) Perkins, M. J . Adv. Phys. Org. Chem. 1980, 17, 1.
(3) J anzen, E. G. In Free Radicals in Biology; Pryor, W. A., Ed.;
Academic Press, Inc.: New York, 1980; pp 115-154.
(4) Unpublished results with Prof. Ross Barclay, Mt. Allison Uni-
versity.
(5) McCay, P. B.; Lia, E. K.; Poyer, J . L.; DuBose, C. M.; J anzen, E.
G. J . Biol. Chem. 1984, 259, 2135-2143.
(6) J anzen, E. G.; Krygsman, P. H.; Lindsay, D. A.; Haire, H. D. J .
Am. Chem. Soc. 1990, 112, 8279.
(7) J anzen, E. G.; West, M. S.; Poyer J . L. In Frontiers of Reactive
Oxygen Species in Biology and Medicine; Asada, K., Yashikawa, T.,
Eds.; Amsterdam: Elservier Science Publishers, 1994; pp 431-434.
(8) Pryor, W. A. Free Radicals; McGraw-Hill: New York, 1966; pp
113-126.
0022-3263/96/1961-2358$12.00/0 © 1996 American Chemical Society

Nitrone-Peracid Radical Reactions
J . Org. Chem., Vol. 61, No. 7, 1996 2359
it was shown that free radicals are formed when CPBA
is mixed with PBN in benzene or acetonitrile.
This paper describes a more detailed study of this
system utilizing both mass spectrometry (MS) and EPR
so that nonradical products as well as radical products
can be detected and the structures identified. Also in
addition to CPBA, perbenzoic and perpropionic acids
produced in situ from their aldehyde precursors are
investigated. Since perpropionic acid is an aliphatic
peracid, the study of this system is relevant to biological
peracids formed from lipid aldehydes produced in lipid
peroxidation.
Exp er im en ta l Section
Ch em ica ls. m-Chloroperbenzoic acid (CPBA) (purity 57-
86%, remainder m-chlorobenzoic acid and water), benzalde-
hyde, benzophenone (99%), and benzoyl peroxide were pur-
chased from Aldrich. Propanal was obtained from Eastman
Organic Chemicals. TEMPO was obtained from Sigma. All
solvents used were HPLC grade. PBN, p-cyano-PBN, p-
methoxy-PBN, PBN-d14, 2-phenyl-DMPO, nitronyl-13C labeled
2-Ph-DMPO, and MNP were obtained from OMRF Spin Trap
Source (Oklahoma Medical Research Foundation, 825 N.E.
13th Street, Oklahoma City, OK 73104).
P er a cid s. Three peracids were used in the nitrone-peracid
reactions: m-chloroperbenzoic acid (ClC6H4COOOH), perben-
zoic acid (C6H5COOOH), and perpropionic acid (C2H5COOOH).
Perbenzoic acid was obtained but not isolated from a reaction
between O2 and benzaldehyde in the presence of UV light.
After UV irradiation, the remaining benzaldehyde was washed
away by hexane; then the residue was dissolved in benzene
or acetonitrile. Because of the poor UV light absorption,
perpropionic acid was obtained, but not isolated, by UV
irradiation of propanal in the presence of benzophenone and
O2. Concentrations of the peracids were determined by
NaS2O3/NaI titration,10 using a 1.5 g of NaI, 50 mL of H2O, 5
mL of glacial acetic acid, and 5 mL of chloroform mixture with
a known volume of peracid solution. The mixture was then
titrated with 0.1 M NaS2O3, where 1 mL ) 0.0086 g of CPBA,
0.0069 g of perbenzoic acid, or 0.0044 g of perpropionic acid.
By titration, the purity of CPBA used in this work was shown
to be 80%. Calculated concentrations are listed below:
perbenzoic acid
perpropionic acid
F igu r e 1. EPR spectrum obtained from 20 mM PBN and 1.74
mM CPBA in the absence of air at room temperature (a) in
CH3CN; (b) in benzene; (c) in benzene 3 days after the
spectrum a was recorded; (d) in benzene 5 days after the
spectrum a was recorded.
a (mM) 6.16 ( 0.49
b (mM) 17.50 ( 0.35
c (mM)
1.28 ( 0.03
3.83 ( 0.77
21.60 ( 0.60
4.32 ( 0.12
d (mM) 3.50 ( 0.07
MS/MS Mea su r em en ts. Argon gas was used for the
collision-induced dissociation (CID). The collision energy was
100 eV.
EP R Mea su r em en t. EPR spectra were measured in a
Bruker ER300 spectrometer with 100 KHz field modulation.
Round quartz sample cells with 3.5 mm i.d. were employed
for benzene solutions, and flat quartz cells were used to record
acetonitrile solutions.
where a is the initial concentration present in the aldehyde, b
is the concentration after 15 min of UV irradiation in the
presence of O2, c is the concentration after 15 min of UV
irradiation in the presence of benzophenone and O2, and d is
the concentration used in PBN reaction.
P r od u ct Yield s. Product yields were estimated from
integrating EPR spectra which were calibrated by standard
solutions of TEMPO ranging from 5 × 10-7 mM to 1 × 10-5
mM. Results presented throughout are an average of three
independent experiments.
Resu lts
Electr on Im p a ct Ion iza tion (EI) Mea su r em en ts. EI
mass spectra were obtained with a VG-Fisons Quattro triple
stage quadrupole mass spectrometer. A direct insertion probe
was used. Source temperature was 180 °C. The electron
energy employed was 70 eV. The probe temperature was
about 60 °C.
P BN-CP BA System s. In an acetonitrile solution,
EPR spectra of PBN-CPBA products consist of a triplet
of doublets as the main component, as shown in Figure
1a. The N- and â-H hyperfine splitting constants (hfsc’s)
are indicative of m-chlorobenzoyloxyl adduct of PBN, 1,
as the major product; aN ) 13.43 G, aâH ) 1.76 G.11
A
minor product in this solvent has only three lines due to
(9) J anzen, E. G.; Lin, C. R.; Hinton, R. D. J . Org. Chem. 1992, 57,
1633.
(10) Vogel, A. Vogel’s Textbook of Practical Organic Chemistry, 4th
ed.; Longman: London and New York, 1978; pp 307-308.
(11) J anzen, E. G.; Evans, C. A.; Nishi, Y. J . Am. Chem. Soc. 1972,
94, 8236.

2360 J . Org. Chem., Vol. 61, No. 7, 1996
Sang et al.
the well-known benzoyl tert-butyl aminoxyl radical (PB-
NOX), 2, N-hfsc ) 8.01 G.
In benzene the initial major product is 2 (Figure 1b).
Evidence of further radical formation is found with time
(Figure 1, parts c and d). As shown in Figure 1c, another
set of three lines is clearly evident after 3 days in benzene
with a N-hfsc of 15.44 G. This triplet is assigned to di-
tert-butylaminoxyl, 3, because this hyperfine splitting
pattern is not changed by using the R-13C-PBN instead
of PBN. The spectrum recorded after 5 days (Figure 1d)
consists solely of a triplet of doublets with larger values
for the N- and â-H hfsc’s (aN ) 14.24 G, aâH ) 2.04 G)
than for the comparable 1. This spectrum is tentatively
assigned to the m-chlorophenyl adduct of PBN, 4, pro-
duced by decarboxylation of m-chlorobenzoyloxyl radicals.
MS presents further evidence for the formation of the
above radical products. Because of the existence of the
isotopes of 35Cl (75.8%) and 37Cl (24.2%), ions containing
the chlorine atom can be recognized in the mass spectrum
by observation of the isotope peaks. An ion containing
one chlorine atom gives two peaks with relative intensity
100:32.5.12 Therefore the presence of peaks at m/ z 332
and 334 in Figure 2a indicates the formation of spin
adduct 1. The assignment of the peaks at m/ z 331 and
m/ z 333 will be discussed later (see below). Peaks at
m/ z 276 and 278 in Figure 2a are the result of the loss
of isobutylene from the molecular ions (M•+) by internal
hydrogen transfer (McLafferty-like rearrangement).13
Peaks at m/ z 260 and 262 are the result of loss of both
isobutylene and oxygen from M•+.14
F igu r e 2. (a) The mass spectrum recorded from the reaction
mixture of 20 mM PBN and 1.74 mM CPBA in CH3CN in the
absence of air at room temperature. (b) The precursor ion
spectrum of m/ z 57 recorded from the same system as in a.
(c) The CID spectrum of m/ z 193 recorded from the same
system as in a.
MS also provides evidence for another previously
undetected product of the PBN-CPBA system, namely,
m-chlorobenzoyloxyl PBN, 6, which has only one mass
dalton less than 1. The nonradical product, 6, is not
identified by EPR. MS, however, confirms its structure
in Figure 2a, where peaks at m/ z 331 and 333 are
appropriate molecular ions for 6. The nitrone 6 must
come from oxidation of the initially formed spin
adduct 1.
Because PBN spin adducts usually produce m/ z 57 as
an abundant fragment, a precursor ion spectrum of m/ z
57 (Figure 2b) is recorded during tandem mass spectrom-
etry (MS/MS) analysis to find ions that are potentially
due to PBN spin adducts from a mixture sample. Peaks
at m/ z 332 and 334 are due to 1. The presence of the
peak at m/ z 193 is an indication of 2 plus a proton, 5.
Other peaks are probably due to fragments or peaks from
nonradical species. The assignment of m/ z 193 is further
confirmed by observation of fragments at m/ z 137 [M -
C4H8]+, m/ z 105 [C6H5CO]+ and m/ z 57 [C4H9+], which
are produced by the collision-induced dissociation (CID)
of ion at m/ z 193 (Figure 2c).
To aid in the assignment of structure 6, deuterated
PBN (phenyl and tert-butyl perdeuteratedsnot the ni-
tronyl hydrogen, PBN-d14) was used. After using PBN-
d
14, the molecular ions at m/ z 332 and 334 of 1 in Figure
(12) MaLafferty, F. W. Interpretation of Mass Spectra; 3rd ed.;
University Science Books: Mill Valley, CA, 1980; p 16.
(13) J anzen, E. G.; DuBose, C. M. Anal. Lett. 1993, 26, 2661.
(14) J anzen, E. G.; Sang, H.; Kotake, Y.; Arimura, M.; DuBose, C.
M.; Poyer, J . L. J . Am. Soc. Mass Spectrom. 1995, 6, 847.
2a shift 14 daltons to m/ z 346 and 348, respectively;
molecular ions at m/ z 331 and 333 of 6 also shift 14
daltons to m/ z 345 and 347, respectively, which indicates

Nitrone-Peracid Radical Reactions
J . Org. Chem., Vol. 61, No. 7, 1996 2361
F igu r e 3. EPR spectrum obtained from 20 mM (a) 2-Ph-
DMPO and (b) R-13C-2-Ph-DMPO and 1.74 mM CPBA in the
absence of air at room temperature.
F igu r e 4. The mass spectrum recorded from the reaction
mixture of 20 mM (a) 2-Ph-DMPO; (b) R-13C-2-Ph-DMPO and
1.74 mM CPBA in benzene in the absence of air at room
temperature.
that the â-H must be lost from structure 1 to form
structure 6. If this was not the case, a 13 daltons shift
should be observed after using PBN-d14 because the loss
of deuterium atom causes 2 mass dalton change.
by the observation of 2 mass daltons shift after using of
R-13C-2-Ph-DMPO instead of 2-Ph-DMPO (Figure 4b).
In addition to the identification of the above free
radical adducts, another nonradical product is also
identified by MS, much like the PBN-CPBA system. A
significant relative abundance of m/ z 343 and m/ z 345
ions which have 1 mass dalton less than 7 was observed
in Figure 3a. Unlike PBN, here no â-H exists in spin
adducts of 2-Ph-DMPO which can be lost. This product
might be due to structure 9a or 9b.
2-P h -DMP O-CP BA System . 2-Ph-DMPO reacts
with CPBA in a similar way to that of PBN. The EPR
spectrum recorded from this system shows two triplets
in benzene (Figure 3a) with hfsc’s aN ) 14.35 G and aN
) 12.80 G, respectively. The signal with hfsc aN ) 12.80
G is assigned to the m-chlorobenzoyloxyl adduct of 2-Ph-
DMPO, 7, based on its hfsc value.15 The assignment of
the second triplet was made based on the use of R-13C-
2-Ph-DMPO15 instead of 2-Ph-DMPO. The R-13C-2-Ph-
DMPO gives more hyperfine splitting information due
to the existence of the R-13C.16 When R-13C-2-Ph-DMPO
was used, the EPR spectrum (Figure 3b) consists of one
13C
triplet of doublets with hfsc’s aN ) 12.80 G and aâ
)
3.76 G and one triplet with aN ) 14.35 G. Because no
13C hyperfine splitting was observed for the second
nitroxide, this radical is assigned to the structure 8. The
mechanism of formation of this radical is probably similar
to that of 3 in the PBN-CPBA system (see Discussion).
The peaks at m/ z 328 and m/ z 330 are considered to
be fragments of m/ z 343 and m/ z 345 probably produced
from 7 by loss of oxygen. All these assignments are
supported by the experiment using R-13C-2-Ph-DMPO
(see Figure 4b).
P BN-P er ben zoic Acid System (in Situ ). Perbenzoic
acid behaves in much the same way as CPBA when
mixed with PBN. Analysis of the EPR spectrum indi-
cates the formation of the benzoyloxyl adduct and PB-
NOX. The hfcs’s of the benzoloxyl adduct are aN ) 13.39
G, aâH ) 1.50 G in benzene;11 the hfsc of PBNOX is aN
)
The assignments of structure 7 and 8 are also con-
firmed by MS experiments. Figure 4a is the mass
spectrum recorded from the system of 2-Ph-DMPO and
CPBA in benzene. The isotope peaks at m/ z 344 and
346 are due to 7. The ion at m/z 379 is considered to be
the product 8 minus a proton, which is further confirmed
8.01 G.
P BN-P er p r op ion ic Acid System (in Situ ). Perpro-
pionic acid produces the EPR spectrum shown in Figure
5, thus confirming that aliphatic peracids also react with
PBN to form free radical products. When benzene or
CH3CN is used as a solvent, the EPR spectrum shows
two triplets of doublets. The main spin adduct formed
(15) J anzen, E. G.; Zhang, Y.-K.; Haire, D. L. J . Am. Chem. Soc.
1994, 116, 3738.
(16) Haire, D. L.; Oehler, U. M.; Krygsman, P. H.; J anzen, E. G. J .
Org. Chem. 1988, 53, 4535.
in this system has hfcs’s as follows: aN ) 13.95 G, aâH
)
2.08 G in benzene. These values are consistent with an

2362 J . Org. Chem., Vol. 61, No. 7, 1996
Sang et al.
alkoxyl adduct of PBN.17 The minor component has hfcs’s
radicals since the spin adduct is not stable, producing
typical of alkyl radical adducts: aN ) 14.60 G and aâH
)
tert-butyl radical as one of the cleavage products:
3.25 G.18 We suggest that these two species are the
ethoxyl and ethyl spin adducts, respectively, produced
from the decarboxylation of the propionoyloxyl radicals
and production of alkoxyl radicals from alkylperoxyl
radical intermediates. These results are preliminary,
and further work is planned to understand systems
involving aliphatic peracid and PBN.
P r od u ct Yield s. Acetonitrile and benzene produced
significant differences in the initial yields of product 1
and 2, as shown in Table 1. Product 2 is the major initial
product formed in the benzene system. The concentra-
tion of 1 is increased with the use of acetonitrile, while
oxidized PBN yield is drastically lowered.
Para-substituted PBN’s display different effects in the
reaction with CPBA. With electron-withdrawing sub-
stituents, e.g. CN-PBN, very small amounts of either
radical are produced. When an electron-donating sub-
stituent is used, e.g. CH3O-PBN, the highest total yield
of 1 and 2 is found compared to CN-PBN or PBN.
With an increase of CPBA concentration in the region
of 0.5 mM to 3 mM, the product yields also increase. The
presence of air changes product amounts. In the pres-
ence of air, a larger amount of PBNOX is formed in
benzene. Thus in benzene, 2-Ph-DMPO yields more
m-chlorobenzoyloxyl adduct than PBN (results shown in
Table 1).
Indeed when benzoyl peroxide in benzene was mixed
with MNP at near room temperature where benzoyloxyl
radicals are known to be formed,19 strong EPR signals
due to 3 were detected without any trace of spectra
attributed to the benzoyloxyl radical adduct of MNP.
However, the detection of 1 and 4 indicates that free
m-chlorobenzoyloxyl radicals are also formed when CPBA
is used. Perhaps a precursor molecular addition product,
12, is formed. This product has the structure of a
hydroxylamine of the peroxyl adduct of PBN, which
might be expected to produce radicals spontaneously by
an internal redox reaction:
Discu ssion
The mechanism of reaction between a peracid and a
nitronyl function should first of all include an “epoxida-
tion”. Since imines react with peracids to produce
oxaziranes, nitrones might be expected to produce ox-
azirane N-oxides:
However, the hydroxyl adduct of PBN has not been
detected in any systems reported in this investigation.
It is possible that 12 could also be oxidized by another
CPBA molecule to form 15 before internal disproportion-
ation occurs:
The chemistry of 10 is unknown, but hydride rear-
rangement or cleavage could account for some of the
products observed:
The production of large amounts of di-tert-butylami-
noxyl, 3, can come from reactions of MNP with acyloxyl
(17) Bluhm, A. L.; Weinstein, J .; Sousa, J . A. Nature 1971, 229, 500.
(18) Bluhm, A. L.; Weinstein, J . J . Org. Chem. 1972, 37, 1748.
(19) J anzen, E. G.; Blackburn, B. J . J . Am. Chem. Soc. 1968, 90,
5909-5910.

Nitrone-Peracid Radical Reactions
J . Org. Chem., Vol. 61, No. 7, 1996 2363
Ta ble 1. Yield s of Sp in Ad d u cts P r od u ced fr om th e CP BA System in th e Absen ce of Air
benzene yield (M) × 107
CH3CN yield (M) × 107
1
1a
2
3
2
CN-PBN
PBN
CH3O-PBN
<1.0
4.9 ( 1.6
3.2 ( 3.2
<1.0
22.5 ( 9.5
152.0 ( 46.2
0
0
3.7 ( 1.1
0
44.5 ( 5.5
2.9 ( 0.9
2.6 ( 1.0
24.5 ( 0.5
113.5 ( 16.5
7
8
7
8
2-Ph-DMPO
19.9 ( 5.6
9.7 ( 2.4
36.8 ( 0.4
<1.0
a
1, 2, 3, 7, 8 are product 1, 2, 3, 7, 8 respectively, which are described in text.
adduct of PBN, 15, is not expected to be stable,6 decom-
position will spontaneously produce benzaldehyde, m-
chlorobenzoyloxyl 14, and MNP. Some of 14 will be
quenched in the solvent cage by reacting with PhCHO;
the produced benzoyl radical will then react with MNP
to yield 2.
As shown from yield analysis, 2 is the major product
in benzene; however, 1 becomes the predominant product
in CH3CN. A possible explanation for this might be that
the molecular addition product, 12, is more likely to be
produced in polar media.
Con clu sion s
Spin traps PBN and 2-Ph-DMPO react with peracid
to spontaneously produce free radicals. PBN and CPBA
initially produce the m-chlorobenzoyloxyl adduct and
PBNOX. The product yields depend on the solvent
polarity, peracid concentration, and substituent group on
PBN. The radical products and nonradical products can
be identified by EPR and/or MS. Perbenzoic acid pre-
pared in situ behaves in much the same way as CPBA.
The aliphatic acid, perpropionic acid, also prepared in
situ reacts with PBN to produce C2H5OPBN as a main
radical spin adduct.
F igu r e 5. (a) EPR spectrum obtained from 20 mM PBN and
4.32 mM perpropionic acid (in situ) in benzene in the absence
of air at room temperature. (b) Computer simulation of
spectrum a.
Ack n ow led gm en t. This work was supported by the
National Institute for Research Resources, NIH Grant
No. RR05517. Grateful acknowldgement is hereby
made.
If these reactions described above occur faster than the
disproportionation reaction indicated earlier, spin adduct
2 will be detected as a main product. Since the peroxyl
J O952031W