Reactions of Microwave-Generated O(3P) Atoms with Unsaturated Hydrocarbons

Article abstract of DOI:10.1021/jo972083g

The reactions of neat olefins or solutions of olefins in acetone at low temperature with oxygen atoms were examined. O(³P) atoms were produced by microwave irradiation of He/O₂ mixtures, followed by contact of the plasma with the fluid at low pressure and temperature. Addition of oxygen atoms to olefins results in skeletal rearrangements involving hydrogen and alkyl migration reactions and ring rearrangements of the intermediate oxygen adducts in competition with epoxide formation. While epoxide formation predominates for simple olefins such as 1- and 4-octene with minor yields of rearrangement products, for highly substituted or strained olefins, such as norbornadiene, skeletal rearrangement dominates following oxygen atom addition. When oxidation of norbornadiene is carried out in the presence of a radical inhibitor to suppress secondary oxidation leading to benzene, the novel ring-rearrangement product, bicyclo[3.^2,31.0]hex- 3-ene-endo-6-carboxaldehyde, is produced from norbornadiene in significant yields.

Full text of DOI:10.1021/jo972083g

J . Org. Chem. 1998, 63, 4587-4593
4587
Rea ction s of Micr ow a ve-Gen er a ted O(3P ) Atom s w ith Un sa tu r a ted
Hyd r oca r bon s
D. D. Tanner,*,1 P. Kandanarachchi,2 N. C. Das,1 M. Brausen,3 C. T. Vo,3
D. M. Camaioni,2 and J . A. Franz*,2
Department of Chemistry, University of Alberta, Edmonton, Alberta T6G 2G2, Canada, and
Pacific Northwest National Laboratory, Richland, Washington 99352
Received November 13, 1997
The reactions of neat olefins or solutions of olefins in acetone at low temperature with oxygen
atoms were examined. O(3P) atoms were produced by microwave irradiation of He/O2 mixtures,
followed by contact of the plasma with the fluid at low pressure and temperature. Addition of
oxygen atoms to olefins results in skeletal rearrangements involving hydrogen and alkyl migration
reactions and ring rearrangements of the intermediate oxygen adducts in competition with epoxide
formation. While epoxide formation predominates for simple olefins such as 1- and 4-octene with
minor yields of rearrangement products, for highly substituted or strained olefins, such as
norbornadiene, skeletal rearrangement dominates following oxygen atom addition. When oxidation
of norbornadiene is carried out in the presence of a radical inhibitor to suppress secondary oxidation
leading to benzene, the novel ring-rearrangement product, bicyclo[3.2,31.0]hex-3-ene-endo-6-
carboxaldehyde, is produced from norbornadiene in significant yields.
In tr od u ction
show deep-seated fragmentation and rearrangements of
substrates, indicative of energetically enriched interme-
diates.
It is clear from our previous work on the solution-phase
reactions of atomic (H•)4 and more complex radical species
(CH3 ) formed using a microwave discharge that the
A number of studies that generate ground-state oxy-
gen, O(3P), have reported reactions carried out in the
condensed phase or on solid surfaces.9-11 The liquid-phase
reactions have been carried out using γ-irradiation of
substrates dissolved in liquid CO2,8 by the microwave
discharge of CO2,10 with oxygen atoms generated by an
O2/He plasma,11 or with a plasma formed from a mixture
of N2 and N2O.12
The mechanism proposed for the addition of O(3P)
atoms to an olefin was shown to proceed via a triplet
diradical since cis- or trans-stilbene yielded both cis- and
trans-stilbene oxide as well as benzyl phenyl ketone.9a,b
The use of a O2/He plasma to generate O(3P) atoms has
been questioned since the reactive species produced by
the plasma were possibly contaminated with O2, O2(1Σg),
or O3.9i When O(3P) atoms were generated using a
microwave discharge of O2/He, the product ratios formed
• 5
conditions produce thermalized reactants whose products
are the result of the reactions of the primary intermedi-
ates. Kinetic studies show that addition of oxygen to
olefins at room temperature is nearly diffusion-controlled,
whereas hydrogen abstraction is more than a factor of
103 slower per hydrogen than addition to an olefin.6
Thus, low temperatures will further enhance selectivity
for addition, in addition to producing thermalized inter-
mediates. A number of publications have reported the
vapor- phase reactions of atomic oxygen, O(3P) generated
by the mercury-photosensitized decomposition of nitrous
oxide7 or by a microwave plasma promoted by O2/He at
low pressures.8 The vapor-phase reactions in general
(1) Department of Chemistry, University of Alberta
(2) Pacific Northwest National Laboratory, Richland, WA 99352.
(3) NRC Summer Student Research Fellow, 1993.
(8) (a) Bell, A. T. In Techniques and Applications of Plasma
Chemistry; Hollahan, J . R., Bell, A. T., Eds.; Wiley-Interscience: New
York, 1974; Chapter 1, pp 1-55. (b) Sato, S.; Cvetanovic, R. J . Can. J .
Chem. 1959, 37, 953. (c) Cvetanovic, R. J . Can. J . Chem. 1960, 38,
1678. (d) Huie, R. E.; Herron, J . T. Int. J . Chem. Kinet. 1972, 4, 521.
(9) (a) Sakurai, H.; Akimoto, K.; Toki, S.; Takamuku, S. Chem. Lett.
1975, 469. (b) Goto, S.; Hori, A.; Takamuku, S.; Sakurai, H. Bull. Chem.
Soc. J pn. 1976, 49, 2965. (c) Hirokami, S.; Wojnarovits, L.; Sato, S.
Bull Chem. Soc. J pn. 1979, 52, 299. (d) Karawawa, H.; Samamoto, T.;
Yugeta, R.; Sato, S. Bull. Chem. Soc. J pn. 1979, 52, 902.
(10) (a) Zadok, E.; Amar, D.; Mazur, Y. J . Am. Chem. Soc. 1980,
102, 6369. (b) Zadok, E.; Sialom, B.; Mazur, Y. Angew. Chem., Int. Ed.
Engl. 1980, 19, 1004. (c) Zadok, E.; Mazur, Y. Tetrahedron Lett. 1980,
21, 4955. (d) Zadok, E.; Mazur, Y. J . Org. Chem. 1982, 47, 2233. (e)
Zadok, E.; Rubinraut, S.; Mazur, Y. Isr. J . Chem. 1983, 23, 457. (f)
Zadok, E.; Aronovitch, C.; Mazur, Y. Nouv. J . Chim. 1982, 6, 699. (g)
Zadok, E.; Rubinraut, S.; Frolow, F.; Mazur, Y. J . Am. Chem. Soc. 1985,
107, 2489. (h) Zadok, E.; Rubinraut, S.; Frolow, F.; Mazur, Y. J . Org.
Chem. 1985, 50, 2647. (i) Zadok, E.; Rubinraut, S., Mazur, Y. J . Org.
Chem. 1987, 52, 385.
(4) (a) Tanner, D. D.; Zhang, L. J . Am. Chem. Soc. 1994, 116, 6683.
(b) Tanner, D. D.; Zhang, L.; Vigneswaran, M.; Kandanarachchi, P. J .
Org. Chem. 1995, 60, 4481. (c) Tanner, D. D.; Zhang, L.; Kandan-
arachchi, P. J . Phys. Chem. 1996, 100, 11319. (d) Tanner, D. D.; Zhang,
L.; Hu, L. Q.; Kandanarachchi, P. J . Org. Chem. 1996, 61, 6818. (e)
Tanner, D. D.; Koppula, S.; Kandanarachchi, P. J . Org. Chem. 1997,
62, 4210.
(5) Final Report CANMET Contract No. 23440-4-1053/01-SQ.
(6) Bucher, G.; Scaiano, J . C. J . Phys. Chem. 1994, 98, 12471-1.
For example, cyclopentene reacts with O(3P) at 1.6 × 1010 M-1 s-1 at
room temperature in acetonitrile, compared to cyclopentane, 3.8 × 108
M-1 s-1, both rate constants on a per molecule basis.
(7) (a) Cvetanovic, R. J . Adv. Photochem. 1963, 1, 115. (b) Cvet-
anovic, R. J .; Ring, D. F.; Doyle, L. C. J . Phys. Chem. 1971, 75, 3056.
(c) Atkinson, R.; Cvetanovic, R. J . J . Chem. Phys. 1971, 55, 659. (d)
Hirokami, S.; Cvetanovic, R. J . J . Am. Chem. Soc. 1974, 96, 3738. (e)
Singleton, D. L.; Cvetanovic, R. J . J . Am. Chem. Soc. 1974, 96, 6812.
(f) Gaffney, J . S.; Atkinson, R.; Pitts, J . N., J r. J . Am. Chem. Soc. 1976,
98, 1828. (g) Havel, J . J . J . Am. Chem. Soc. 1974, 96, 530. (h) Havel,
J . J .; Chamberlain, W. T.; Krautter, P. M. J . Am. Chem. Soc. 1974,
96, 632. (i) Havel, J . J .; Chan, K. H. J . Org. Chem. 1974, 39, 2439; J .
Am. Chem. Soc. 1975, 97, 5800. (j) Havel, J . J . J . Org. Chem. 1978,
43, 762.
(11) (a) Sato, S.; Cvetanovic, R. J . Can. J . Chem. 1959, 37, 529. (b)
Cvetanovic, R. J . Can. J . Chem. 1960, 38, 1678. (c) Huie, R. E.; Herron,
J . T. Int. J . Chem. Kinet. 1972, 4, 521.
(12) Ung, A. Y. M. Chem. Phys. Lett. 1975, 32, 351.
S0022-3263(97)02083-5 CCC: $15.00 © 1998 American Chemical Society
Published on Web 06/17/1998

4588 J . Org. Chem., Vol. 63, No. 14, 1998
Ta ble 1. Rea ction P r od u cts of Cycloh exen e w ith Atom ic O(3P ), -78˚C
Tanner et al.
a
b
Area % of products. Acetone solutions.
Sch em e 1
from the oxidation of styrene9i or cis-stilbene9a were not
appreciably different. However, with all three of the
microwave-promoted methods for the generation of O(3P)
atoms, an appreciable amount of benzaldehyde (5-13%)
was formed from the oxidation reactions of styrene.9i The
formation of a cleavage product, benzaldehyde, suggested
that under the reaction conditions in all three cases ozone
is formed.
To establish the identity of the reactive reagent and
to compare some of the more striking differences between
the vapor-phase and solution-phase reactions of O(3P)
atoms, the reaction of a number of olefins was investi-
gated. The model substrates chosen are examples of
vapor-phase or condensed-phase chemistry that involved
extensive molecular rearrangement or fragmentation.
Resu lts a n d Discu ssion
Cycloh exen e. The vapor-phase oxidation of cyclo-
hexene (1) by O(3P) atoms was reported6 to give epoxy-
cyclohexane (2), cyclohexanone (3), 5-hexenal (4), 2-hex-
enal (5), and cyclopentanecarboxaldehyde (6). In the
present solution-phase study, neat and acetone solutions
of cyclohexene (-78 °C) formed 2, 3, 6, and several
cyclohexyl and cyclopentyl products, but did not form
open-chain materials in appreciable yields; see Table 1.
A mechanism that accounts for the products is consis-
tent with the intermediacy of a triplet diradical (a ). The
reactive intermediate b formed after intersystem crossing
(ISC) can rationalize the formation of the other products;
see Scheme 1. Cyclohexenol (7) may be formed by
deprotonation/reprotonation of b; cyclohexanone is formed
by a 1,2-hydrogen shift of b. Products 3 and 6 are formed
by hydrogen migration and skeletal rearrangement of b.
We prefer to attribute some zwitterionic character to b
in the polar solvent, acetone, because the skeletal rear-
rangement leading to the production of cyclopentanecar-
boxaldehyde, 6, and in particular, the formation of 1,2-
cyclohexanediol, 8, by hydration, are both suggestive of
a dipolar character for b in Scheme 1. The possibility
that 3, 6, and 8 could form by acid-catalyzed hydrolysis
of 2 was also considered. Titration of a reaction mixture
following 60% conversion of cyclohexene revealed 0.01-
0.04 M acid content following exposure to the plasma.
However, when the reaction mixture resulting from
exposure of cyclohexene to oxygen atoms was analyzed
by GC and NMR over a period of 18 h, a negligible
decrease in the concentration of cyclohexene oxide or
change in product ratios was observed. Finally, the
possibility that the products were formed from reactions
of ozone was ruled out by exposing cyclohexene to ozone
under identical reaction conditions. The formation of
ketones and aldehydes in the gas-phase addition of
oxygen atom to olefins has been suggested to involve 1,2-
hydrogen, -alkyl, and -aryl migrations.13 The photolysis
of epoxides, presumably leading to the same diradical
intermediates, may involve 1,2-alkyl and -hydrogen
migration reactions for simple alkene oxides.14 The
present results are consistent with rapid 1,2-hydrogen
and 1,2-alkyl (ring contraction) shifts involving the
diradical intermediate.
1-Octen e. The reaction of 1-octene with the O2/He
plasma yields five major products accounting for 99% of
the substrate that reacts. Several (approximately five)
minor products could also be detected; see Table 2. A
mechanism that rationalizes the production of 96-98%
of the products formed proceeds by a single intermediate,
a diradical, formed from the addition of O(3P) to the
olefin, Scheme 2. Products reflecting the participation
of O2 or ozone account for several percent of the products.
The hydrogen shift leading to RCH2CHO may occur by
hydride shift to carbon, proton shift to oxygen, or depro-
tonation/reprotonation involving solvent or equivalently
may be described as a direct 1,2-hydrogen atom migration
(13) Suhr, H. Plasma Chem. Process. 1982, 3, 1.
(14) (a) Nastasi, M.; Streith, J . In Rearrangements in Ground and
Excited States, De Mayo, P., Ed.; Academic Press: New York, 1980;
Vol. 3, p 445. (b) Gritter, R. J .; Sabatino, E. C. J . Org. Chem. 1964,
29, 1965.

Reactions of Microwave-Generated O(3P) Atoms
J . Org. Chem., Vol. 63, No. 14, 1998 4589
Ta ble 2. Products from the Reaction of 1-Octene with a O2/He Plasma, ∼78˚Ca
a
b
He/O2 ratio 4.5/1. Five other products, area (%).
Ta ble 3. P r od u cts of th e Rea ction s of cis-4-Octen e w ith
Atom ic Oxygen (-78 °C)
Sch em e 2
product yield (%)
[cis-4-octene]a time conversion
cis-4,5-
trans-4,5- 4-octa-
(mol L-1
)
(min)
(%)
epoxyoctane epoxyoctane none
0.71
0.71
0.18
2
10
5
35.9
70.1
85.0
59.1
42.2
35.7
39.8
27.6
24.1
1.09
30.2
40.2
a
Acetone solutions (oxygen flow rate ) 6 mL/min).
Sch em e 3
of the singlet diradical, reminiscent of photochemical
ring-opening reactions of epoxides14,15 in solution that
produce ketones and aldehydes. The singlet diradical
may be expected to exhibit a greater degree of zwitteri-
onic character in the polar solution phase than in the
gas phase. As with cyclohexene, polar diradical charac-
teristics will effect alkyl and hydrogen migration reac-
tions and reactions with nucleophiles such as water.
Activation barriers for 1,2-alkyl and 1,2-hydrogen migra-
tion on the singlet surface must be very nearly barrier-
less, since they must compete with ring closure of the
diradical to the epoxide on the singlet surface. In one
theoretical study of the biradical from addition of oxygen
atom to CH2dCH2, attempts to optimize the geometry of
the singlet diradical led only to collapse to acetaldehyde,
suggestive of the relative facility of the 1,2-hydrogen shift
compared to oxirane formation.15 Finally, we note that
the major product observed from 1-octene, octene oxide,
was not formed by reaction of ozone with the olefin, as
demonstrated in control ozonolysis experiments.
cis-4-Octen e. To confirm the observations previously
reported that cis- or trans 1,2-disubstituted ethylenes
yielded, among other products, both cis- and trans-
epoxides, in the condensed phase,9i or in the vapor
phase,16 the reaction of cis-4-octene was carried out using
an O2/He plasma; see Table 3. The yield of epoxides
(99%) formed at moderate (36%) conversion is fully
consistent with the proposed mechanism, formation of
the triplet diradical adduct, which undergoes internal
rotation on a comparable or shorter time scale than
intersystem crossing and epoxide formation. With ex-
tensive conversion (70-85%), lower yields of epoxides
(60-70%) and higher yields of 4-octanone are observed.
The latter product may reflect secondary attack on the
epoxide at extensive conversion.
2,3-Dim eth yl-2-bu ten e. The products from the reac-
tion of a tetrasubstituted ethylene include both addition,
rearrangement, and fragmentation as well as further
oxidation; see Table 4. The fragmentation to acetone
accounted for approximately 22% of the products at -42
°C and can be tentatively proposed as arising from the
further reaction of the diradical with undissociated O2;
see Scheme 3. While the participation of O2 or ozone
provides a pathway for acetone production, the majority
of the products can be rationalized by rearrangements
of the biradical intermediate resulting from oxygen atom
addition. The production of methyl tert-butyl ketone
suggests that 1,2-alkyl migration competes with oxirane
formation in the biradical system.
Nor bor n en e. The reaction of norbornene with an O2/
He plasma yields three major products (95-98% yield);
see Table 5. The product ratios do not appear to change
as a function of the percentage reaction, indicating that
the three products, epoxide, ketone, and alcohol, are
primary products. The formation of the three oxidation
products can be rationalized as occurring from a common
intermediate see Scheme 4. The triplet oxygen atom
adduct undergoes intersystem crossing to the singlet
adduct followed by ring closure to the epoxide or hydro-
gen migration to form the ketone, path c. The formation
of the ketone can be rationalized as occurring via c or by
formation and disproportionation (path d) of hydroxynor-
bornyl radical. However, formation of hydroxynorbornyl
(15) Belbruno, J . J . J . Phys. Org. Chem. 1997, 10, 113-120.
(16) Cvetanovic, R. J . Can. J . Chem. 1958, 36, 623.

4590 J . Org. Chem., Vol. 63, No. 14, 1998
Ta ble 4. P r od u cts fr om th e Rea ction of 2,3-Dim eth yl-2-bu ten e w ith O(3P ) Atom s
Tanner et al.
Sch em e 4
Sch em e 5
radical from disproportionation of the initial oxygen
adduct would require oxygen adduct diradicals to exhibit
unrealistically long triplet lifetimes and to exist at
suitably high adduct concentrations, for the triplet bi-
radicals to undergo bimolecular disproportionation, an
unlikely situation.17 Thus, the participation of hydroxyl
radical is depicted in the production of 2-hydroxynorbor-
nane in Scheme 4, although the source of hydroxyl radical
is uncertain. The formation of hydroxyl radical by O(3P)
abstraction of hydrogen is unlikely in the presence of
olefins.6
Nor bor n a d ien e. The reaction of norbornadiene with
O(3P) gave benzene as the primary product, see Table 6.
At higher conversion (2-90%), a number (∼10) of other
minor products can be detected. Seven aldehydes with
the molecular formula C7H8O were detected, along with
traces of phenol. All of the aldehyde protons exhibited
1
coupling to an adjacent proton in the H NMR spectra;
A mechanism that is consistent with the formation of
benzene, but that forms bicyclic aldehyde, 9, as a major
isolable product, is given in Scheme 5. Addition of O(3P)
to norbornadiene produces the triplet diradical adduct.
After intersystem crossing, the singlet diradical under-
goes typical cationic rearrangements involving the two
intermediates, A20 and B.21 Because of the similarity of
the products to cationic transformations, we depict a
zwitterionic characteristic of the biradical in Scheme 5.
Structure 11 is a possible candidate for one of the seven
product aldehydes but was not positively identified. The
inhibitor suppresses the disappearance of 9, which then
accumulates as a major product at the expense of
benzene. Oxidation at the 7-carbon of norbornadiene is
not the source of benzene. While it is known that
reactions leading to the formation of norbornadien-7-one
result in the efficient formation of benzene,22 the much
higher rate constants for reaction of oxygen atoms with
none of the cyclic aldehydes was conjugated with an
olefinic system. No benzaldehyde was present in the
reactions before workup, although benzaldehyde was
detected following chromatographic workup of the reac-
tion. The aldehyde formed in dominant yield (up to 63%
of the total aldehydes produced) was bicyclo[3.2,31.0]-
hexene-6-endo-carboxaldehyde, 9. An aldehyde compris-
ing 8-23% of the total aldehydes present was tentatively
identified by GC/IR18 and GC/MS19 as 1,2-dihydroben-
zaldehyde, 10. Bicyclo[3.2,31.0]hexene-6-endo-carboxal-
dehyde, 9, was always detected in low absolute yields,
but when a small amount of a radical inhibitor, 2,6-di-
tert-butyl-4-methyl phenol, was added, 9 became the
major product at the expense of benzene; see Table 6.
(17) The 254-nm photolysis of propylene oxide gave acetone, iso-
propyl alcohol, and hexane-2,5-dione as major products (ref 14b). The
authors propose a mechanism in which the diradical from the epoxide
abstracts hydrogen from the epoxide ring carbon, leading to acetonyl
radicals and the observed products. Thus, the biradicals are assumed
to have a sufficient lifetime for abstraction of hydrogen from the solvent
(neat epoxide).
(20) (a) Winstein, S.; Shatavsky, M. J . Am. Chem. Soc. 1956, 78,
592. (b) Winstein, S.; Walborsky, H. M.; Schreiber, K. 1950, 72, 5795.
(21) Story, P. R. Tetrahedron Lett. 1962, 401.
(18) Aldrich/HP Vapor-Phase IR Database.
(19) MIST/EPA/MSDS Mass Spectral Data Base, 1990, PC Version
3.0.
(22) (a) Landsberg J . M.; Sieczkowski J . J . Am. Chem. Soc. 1971,
93, 972. (b) Hunt D. F.; Lillya C. P.; Rauch M. D. J . Am. Chem. Soc.
1968, 90, 2561.

Reactions of Microwave-Generated O(3P) Atoms
J . Org. Chem., Vol. 63, No. 14, 1998 4591
Ta ble 5. P r od u ct Distr ibu tion fr om th e Rea ction of n or bor n en ea w ith Atom ic Oxygen (-72 °C)
a
1 M solution in acetone.
Ta ble 6. P r od u cts fr om th e Rea ction of Atom ic Oxygen w ith Nor bor n a d ien e (-72 °C)a
a
b
An acetone solution of 1 M norbornadiene Reactions 1-5 were performed using norbornadiene fractionally distilled before use.
Reactions 6-9 were carried out with an added radical inhibitor, 2,6-di-tert-butyl-2-methylphenol. c The GC/IR spectra indicated the presence
of an unsaturated nonconjugated aldehyde, tentatively assigned structure of 10. The molecular formula was assigned on the basis of the
GC/MS spectra. Several minor and trace compounds were detected, including phenol and several carbonyl compounds with the molecular
d
formulas C7H8O and C7H8O2.
olefins relative to hydrocarbon CH abstraction6 make a
pathway involving the norbornadien-7-yl or norborna-
dien-7-oxyl radical unlikely. The absence of norborna-
dien-7-ol argues against a Russell-type autoxidative
mechanism23 leading to benzene (via a tetraoxide precur-
sor of norbornadien-7-one and norbornadien-7-ol). When
a radical inhibitor is added to the oxidation reaction,
aldehyde 9 and benzene account for approximately 90%
of the total reaction. Formaldehyde accompanies the
formation of benzene. These results suggest that some
of the aldehyde products 9 and probably 10 (and other
minor aldehydes) are converted in an oxidative radical
chain decomposition to give benzene. In an attempt to
reproduce the conditions for the chain decomposition, 9
was exposed to atomic oxygen. A number of dioxygenated
and polymeric products were produced, but benzene was
produced in only trace quantities. However, when 9 was
heated at 60 °C in acetone in the presence of AIBN (10%
of concentration of 9) for ∼1 half-life, 24 h, benzene was
formed as the only product. For each equivalent of AIBN,
approximately 20 molecules of benzene are formed,
corresponding to a chain length of ∼10. When an
identical solution was irradiated at -78 °C with a 200-W
incandescent lamp, benzene was produced as the only
product. In Scheme 5, both 9 and 10 are depicted as
benzene precursors involving a common intermediate.
However, 9 is clearly the primary source of benzene, since
it is the major aldehydic product.
Con clu sion s
This study has shown that olefins react with atomic
oxygen in solution at reduced temperature to give
predominately epoxides. Unlike gas-phase oxygen-atom
reactions, reactions in the liquid phase at low tempera-
ture produce synthetically useful quantities and distribu-
tions of products. Epoxide formation was shown to be
accompanied by apparent 1,2-hydrogen and 1,2-alkyl
migrations and ring contraction reactions in competition
with ring closure of the diradical to oxiranes. The results
suggest that both hydrogen and alkyl migration must be
nearly barrierless processes in order for them to compete
with oxirane formation on the singlet surface.
Exp er im en ta l Section
Gen er a l Meth od for th e O(3P ) Atom Rea ction s. The
substrate was placed in the Pyrex reactor and maintained at
-42 to -78 °C. Helium and oxygen were introduced into the
system at the desired flow rate, and the pressure of the system
was maintained at 2-4 Torr. The O2/He plasma was gener-
ated in the microwave cavity, and the oxygen atoms were
swept over the stirred solution; see Figure 1.
1
In str u m en ta tion . The H, 2H, and 13C NMR spectra were
obtained using either a Bruker AM-400 (400 MHz), Bruker
AM-300 (300 MHz), or Bruker WH-200 (200 MHz) NMR
spectrometer. Unless otherwise noted, the 1H NMR spectra
are referenced to TMS as an internal standard at 0.00 ppm or
CHCl3 as an internal standard at 7.26 ppm. The 13C NMR
chemical shifts are reported in δ (ppm) relative to chloroform
(δ CDCl3 ) 77.0). The 13C spectra were determined using APT
(23) Russell, G. A. J . Am. Chem. Soc. 1957, 79, 3871.

4592 J . Org. Chem., Vol. 63, No. 14, 1998
Tanner et al.
F igu r e 1. Diagram of apparatus for O(3P) atom reactions.
(attached proton test) to determine the number of protons
attached to each carbon.
reaction with the oxygen plasma did not yield benzene (0.6 M
in acetone, oxygen flow rate 6 mL/min, -72 °C for 8-30 min).
Nor bor n en e. A acetone solution of norbornene (1 mol/L,
10 mL) was subjected to the oxygen/helium plasma (oxygen
flow rate 2-40 mL/min, total pressure 2.0-2.5 Torr, -72 °C).
Samples were withdrawn at intervals and analyzed by gas
chromatography. The products were identified by comparison
of their GC retention times and GC/IR and GC/MS spectra
with those of authentic compounds.
GC/IR data were obtained using a HP 5965A IRD GC/FTIR
interfaced to a HP 5890 gas chromatograph fitted with a DB-5
(30 m × 0.25 mm) glass capillary column.
GC/MS data were obtained using a VG-70E EI+ spectrom-
eter fitted with a Varian Vista 6000 gas chromatograph having
a glass capillary column (DB-5, 30 m × 0.25 mm, J . & W.
Scientific) and interfaced to a 1125 data system.
GC analyses were carried out using a Varian Vista 6000,
FID, gas chromatograph fitted with a glass capillary column
(PONA, 30 m × 0.25 mm, Hewlett-Packard) and interfaced to
a Varian Vista CDS 401 chromatographic data system. The
yield of the reaction products was determined by GC analysis
using a standard calibration solution of known concentrations
of the authentic materials and an added internal standard (1-
chlorooctane or o-dichlorobenzene).7 The purity of the reac-
tants (1-octene, cyclohexene, norbornene, and 2,3-dimethyl-2-
butene, Aldrich Chemical Co.) was checked by GC and found
to be >99% pure. Norbornadiene (Aldrich Chemical Co.) was
washed several times with a solution of sodium hydroxide
(10%), dried over magnesium sulfate, and fractionally distilled
(bp 89 °C) before use. The product was found to be >99% pure
by GC. cis-4-Octene (TCI America) was used as received.
The products were identified by a comparison of their GC
retention times, GC/IR spectra, and GC/MS spectra with those
of authentic samples.
Cycloh exen e. Acetone solutions (1 M, 10 mL) or neat
liquid were subjected to the oxygen/helium plasma (oxygen
flow rate 6-30 mL/min, -78 °C total pressure 2.0-2.5 Torr).
Samples were withdrawn and analyzed by gas chromatogra-
phy. The total acid content of the reaction mixtures was
determined by titration with a standardized sodium hydroxide
solution (0.104 mol/L) and found to be 0.007 and 0.03 M for
neat cyclohexene and 1 M solutions in acetone, respectively.
The product distributions of the acidic reaction mixtures were
monitored for 18 h at room temperature by gas chromatogra-
phy. A reaction mixture was neutralized by adding NaHCO3
and analyzed by gas chromatography. A reaction performed
in deuterated acetone was analyzed by both 1H NMR and gas
chromatography, and the product ratios between cyclohexene
oxide, cyclohexanone, and cyclopentanecarboxaldehyde were
compared in both analysis methods. The concentration of
cyclohexene oxide decreased by 4% in 18 h, and the ratio of
the products remained constant within experimental error over
that period of time.
Nor bor n a d ien e. An acetone solution of norbornadiene (1
M, 10 mL) was exposed to a He/O2 plasma in the same manner
as was norbornene. Several oxidations were carried out with
an added inhibitor, 2,5-di-tert-butyl-4-methyl phenol; see Table
6. At low conversion (2-7%), the major product was benzene
(95.7%), with traces of other higher boiling compounds de-
tected. At higher conversion (>20%), benzene was still the
major product (82%), and several other compounds (>8) were
formed (18%). Phenol was identified as one of the minor (<1%)
products. Formaldehyde was detected by trapping in acetone
at -72 °C by GCIR and identified by its reaction with
resorcinol.25 The aldehyde bicyclo[3.2,31.0]hexa-2-en-6-carbox-
aldehyde (9) was detected and isolated (see below). With
added inhibitor, 9 became the major product at the expense
of benzene; see Table 5. Norbornadien-7-ol was not detected
among the products. Six other aldehydes were also detected.
Aldehyde peaks for the seven products were detected by 1H
NMR at δ 9.1 (d, 5 Hz), 9.15 (10, d, 5 Hz), 9.25 (9,d, 5 Hz),
9.44 (d, 5 Hz), 9.84 (d, 7.5 Hz), 9.96 (d, 7.5 Hz), and 10.19 (d,
7.5 Hz). In a reaction carried out to 25% conversion of
norbornadiene, the percent composition of the aldehydes
(relative to the total aldehyde concentration) was 7.3, 8.4, 62.5,
8.8, 4.4, 2.2, and 6.6%, respectively. At 90% conversion, the
composition of the aldehydes was 11.4, 23.0, 44.5, 2.84, 1.6,
trace, 16.6%, respectively. In both of the latter reactions, the
yield of benzene was greater than 90% and the sum of other
products less than 10%.
Bicyclo[32,3.2.0]h ex-2-en e-6-car boxaldeh yde (9) was sepa-
rated from the reaction mixture by concentration under
reduced pressure to remove benzene and acetone, followed by
chromatography of the remaining residue on a silica gel
column using methylene chloride/hexane (3:1) as eluent. The
product was identified by NMR (13C (APT), proton-coupled and
-decoupled spectra), GC/IR, and GC/MS: 1H NMR (400 MHz,
CDCl3) δ 9.32 (d, 1H, J ) 5.0 Hz), 6.45 (m, 1H), 5.60 (m, 1H),
2.62 (m, 2H), 2.60 (m, 1H), 2.38 (m, 1H), 1.26 (m, 1H); 13C (300
MHz, CDCl3) δ 200.6 (1H), 131.8 (1H), 130.9 (1H), 40.4 (1H),
36.2 (2H), 35.4 (1H), 26.9 (1H); IR (vapor phase ν 3070, 2921,
2819, 2722, 1720, 1392, 1170, 1006, 922, 834, 708 cm-1; E+
(GC/MS, VG 70) m/z 108.1 (M+), 79.0, 77.1, 63.0, 51.0; HRMS
calcd for C7H8O 108.057 52, found 108.058 09.
Bicyclo[2.2.1]h ep ta -2,5-d ien -7-ol was synthesized accord-
ing to the literature procedure24 by the reaction of 7-acetonyl-
norbornadiene with methylmagnesium bromide in THF: 1H
NMR (CDCl3) δ 6.50-6.60 (m, 4H), 3.40-3.50 (m, 2H), 3.05
(d, 1H), 2.15 (s, 1H); IR (vapor phase) ν 2586, 3079, 2997, 1545,
1398, 1320, 1202, 1088, 807, 738, 663 cm-1; MS m/z 107.0 (M+),
91.0, 79.1, 77.0, 64.4, 44.2. Bp 169-170 (750 mmHg). Its
Rea ction of Bicyclo[32,3.1.0]h ex-2-en e-6-ca r boxa ld e-
h yd e (9) w ith Atom ic Oxygen . A solution of 9 in acetone
(25) Feigl, F. Spot Tests in Organic Analysis, 5th Engl. ed.;
Elsevier: Amsterdam, 1956; p 331.
(24) Story, P. R. J . Org. Chem. 1961, 26, 287.

Reactions of Microwave-Generated O(3P) Atoms
J . Org. Chem., Vol. 63, No. 14, 1998 4593
(0.2 M, 2 mL) was subjected to a O2/He plasma at -72 °C for
10 min (oxygen flow rate ) 6 mL/min). Benzene was only
formed in trace amounts, along with dioxygenated products.
Rea ction of Bicyclo[32,3.1.0]h ex-2-en e-6-ca r boxa ld e-
h yd e (9) w ith AIBN. A solution of 9 was heated at 60 °C in
acetone in the presence of AIBN (10% of concentration of 9)
for ∼1 half-life, 24 h. Benzene was formed as the only product.
When an identical solution was irradiated at -72 °C with a
200-W incandescent lamp, benzene was produced in quantita-
tive yield.
Ozon olysis Exp er im en ts. Ozonolysis control experiments
were performed using an ozone generator (Welsbach Corp.,
Pittsburgh, PA) at -78 °C (0.06-0.07 SCFM of oxygen flow,
14-16 mg of O3/min at 8 lbs per in.2 O2 pressure) using 5 g of
neat substrate or 10 mL of a 1 M solution of substrate in
to 90.5% conversion by exposure to O3. A cloudy precipitate
formed that dissolved. No cyclohexene oxide, cyclopentanone,
or cyclopen-tanecarboxaldehyde was observed. Octen e. Neat
1-octene (5 g) was exposed to ozone until 43.4% reaction had
occurred. A white precipitate was formed that dissolved as
the solution was brought to room temperature. No epoxide of
1-octene was observed. Heptanaldehyde was detected as the
major product (∼58%). A 1 M solution of 1-octene in acetone
(10 mL) was exposed to ozone to 96.3% conversion of 1-octene.
No epoxide of 1-octene was observed. Heptanaldehyde was
detected as a major product (∼49%).
Ack n ow led gm en t. This work was supported in part
by the U.S. Department of Energy, Office of Energy
Research, Office of Basic Energy Sciences, under Con-
tract No. DE-ACO6-76RLO 1830 with Battelle Memorial
Institute, which operates the Pacific Northwest National
Laboratory for the U.S. Department of Energy.
acetone over
a 20-minute exposure. The products were
analyzed by gas chromatography. Cycloh exen e. Neat cy-
clohexene (5 g) was taken to 1.5% conversion in the liquid
phase. A white precipitate formed. No cyclohexene oxide,
cyclopentanone, or cyclopentane carboxaldehyde was observed.
A 1 M solution of cyclohexene in acetone (10 mL) was taken
J O972083G