Matrix isolation and spectroscopic characterization of the phenylperoxy radical and its rearranged products

Article abstract of DOI:10.1002/chem.200801546

The phenylperoxy radical 1 has been synthesized by the reaction of the phenyl radical 2 with ³O₂. Radical 1 could be either generated in the gas phase and subsequently trapped in solid argon at 10 K, or directly synthesized in argon

Full text of DOI:10.1002/chem.200801546

DOI: 10.1002/chem.200801546
Matrix Isolation and Spectroscopic Characterization of the Phenylperoxy
Radical and Its Rearranged Products
Artur Mardyukov and Wolfram Sander*[a]
Abstract: The phenylperoxy radical 1
[D5]-16O2-1, [H5]-18O2-1, and [D5]-18O2-1
were matrix-isolated and characterized
by IR spectroscopy. The experimental
IR spectra are in excellent agreement
with results from DFT calculations. Ir-
radiation of 1 with visible light produ-
ces the 2-oxepinoxy radical 5 in a clean
reaction. Subsequent irradiation results
in ring-opening and formation of sever-
al conformers of ketoketene 6. The
radicals 1, 5, and 6 play an important
role in the combustion of aromatic hy-
drocarbons and could now be isolated
and spectroscopically characterized for
the first time.
has been synthesized by the reaction of
3
the phenyl radical 2 with O2. Radical 1
could be either generated in the gas
phase and subsequently trapped in
solid argon at 10 K, or directly synthe-
sized in argon matrices. By reacting 2
as well as its perdeuterated isotopomer
[D5]-2 with 16O2 and with 18O2, respec-
tively, the four isotopomers [H5]-16O2-1,
Keywords:
IR spectroscopy
·
matrix isolation · phenoxyl radical ·
phenyl radical · radicals
Introduction
reaction rate should be at the diffusion limit. At very high
temperatures the main fate of 1 is to lose an oxygen atom
O(3P) and form the phenoxyl radical 3, which is still an exo-
thermic reaction. The phenoxyl radical subsequently elimi-
nates CO to yield the cyclopentadienyl radical C5H5. In
The phenylperoxy radical 1 is the primary product of the re-
action of the phenyl radical 2 with molecular oxygen. This
reaction is considered to be of key importance in the com-
bustion of aromatic hydrocarbons[1,2] and in the degradation
of benzene in the troposphere.[3,4] Due to the high CH bond
dissociation energy of 113 kcalmolÀ1 in benzene,[5] the
phenyl radical 2 is one of the most reactive organic radicals,
and the reaction of 2 with molecular oxygen to give the
peroxy radical 1 is exothermic by 46.3 kcalmolÀ1 according
to ab initio G2M calculations.[6] This highly exothermic pri-
mary step of the oxidation of 2 is followed by a complex se-
quence of secondary reactions.[6,7] The kinetics of the 2 +
3O2 reaction has been studied with the cavity-ring-down
method by measuring the appearance of 1,[8] and from these
data a small negative activation barrier of À0.32 kcalmolÀ1
was determined. Variational transition state theory reveals
3
cross-beam reactions between 2 and O2 an oxygen atom is
abstracted from molecular oxygen to produce 3 and O(3P)
atoms.[10] Under these conditions the phenyl peroxy radical
1 is an extremely short-lived intermediate with lifetimes
below 10 fs.
Despite its importance in various processes the phenyl
peroxy radical 1 has not yet been isolated or characterized
spectroscopically. By cavity-ring-down spectroscopy a very
weak and broad, structureless absorption in the visible
region between 495 and 525 nm was found which could be
used for kinetic measurements but which is not suitable for
the spectroscopic identification of 1.[4,9] In aqueous solution,
a broad peak with a maximum at 490 nm has been assigned
to 1.[11,12] According to DFT calculations the transitions of 1
in the visible region of the spectrum are due to intramolecu-
lar charge transfer and thus very sensitive to the solvent po-
larity.[12]
In contrast to 1, the phenyl radical 2 has been trapped in
inert gas matrices and characterized spectroscopically in
great detail.[13–16] Basically, two methods can be used for the
matrix isolation of 2: the photodissociation of a matrix-iso-
lated precursor or flash vacuum pyrolysis (FVP) of a precur-
sor with subsequent trapping of the products in low temper-
ature matrices. To investigate the thermal reaction of 2 with
3
that the reaction of 2 with O2 is barrierless,[9] and thus the
[a] A. Mardyukov, Prof. Dr. W. Sander
Lehrstuhl fꢀr Organische Chemie II
Ruhr-Universitꢁt Bochum
44780 Bochum (Germany)
Fax : (+49)234-321-4353
E-mail: wolfram.sander@rub.de
Supporting information (IR spectra of deuterated 1 and its photo-
products and tables with Z-matrices and energies of 1, 5, and three
conformers of 6) for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.200801546.
1462
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 1462 – 1467

FULL PAPER
3O2 in low temperature matrices, the matrix has to be an-
nealed at approximately two-thirds of its melting point to
allow the diffusion of the trapped oxygen molecules. Since
the matrix photodissociation necessarily produces radical
pairs trapped in the same matrix cage, the subsequent an-
nealing of these matrices results in unwanted radical recom-
bination instead of reactions with other trapped molecules
(e. g. oxygen) in most cases. FVP with subsequent trapping
in matrices, on the other hand, produces radicals in individu-
al matrix cages. Now the reaction with oxygen molecules
can efficiently compete with radical recombination.
weaker band at 1122.9 cmÀ1 shows a huge redshift of
À67.2 cmÀ1 which clearly indicates a OO stretching vibra-
tion. Other large redshifts are found for a weak absorption
at 793.9 cmÀ1 (À15.6 cmÀ1) assigned to a CO stretching vi-
bration and at 607.2 cmÀ1 (À11.8 cmÀ1) assigned to a COO
deformation mode. A comparison of the newly formed
bands in O2-doped matrices with the results of DFT calcula-
tions (UB3LYP/cc-pVTZ) of the phenylperoxy radical 1
shows an excellent agreement. The two perdeuterated iso-
topomers [D5]-1 and [D5]-18O2-1 were also synthesized and a
careful analysis of their spectra confirms the assignment of
the phenylperoxy radical (Figure 2, Table 1). These experi-
ments clearly show that the phenyl radical 2 reacts with mo-
lecular oxygen in the gas phase to produce 1. The phenoxyl
radical 3 was not observed under these conditions.
Here, we report the matrix isolation and IR spectroscopic
characterization of the phenyl peroxy radical 1 and several
of its isotopomers. In addition, a second important isomer,
the 2-oxepinoxy radical 5, could also be characterized.
The matrix produced after FVP of 4 in 2% O2-doped
argon followed by trapping of the products at 10 K contains
both the phenylperoxy radical 1 and the phenyl radical 2. At
10 K the matrix is very rigid and does not permit the diffu-
sion of oxygen. However, annealing at 30–35 K results in
the rapid diffusion of oxygen, which allows one to investi-
Results and Discussion
Various precursors, such as nitrosobenzene, iodobenzene,
benzoyl peroxide, or benzoic anhydride, have been used as
precursors for the matrix isolation of 2.[13,17] In our experi-
ments we use azobenzene (4) as a new thermal precursor of
2. FVP of 4 at temperatures between 600 and 7008C with
subsequent trapping with a large excess of argon at 10 K
produces 2 in good yields (Scheme 1). By-products found in
these matrices are nitrogen (invisible in the IR), benzene
(formed via hydrogen abstractions from 2), and traces of
acetylene (product of the thermal fragmentation of 2). The
phenyl radical 2 was characterized by comparison of its
matrix infrared spectrum with literature data.[13,17] FVP of
perdeuterated azobenzene [D10]-4 results in the formation of
[D5]-2, again its IR spectrum is in good agreement with the
data reported in literature.
To synthesize the oxygen trapping products of 2, the
argon was doped with 2% O2. FVP of 4 in O2-doped argon
again resulted in the formation of 2, benzene, and acetylene.
In addition, new strong to medium IR absorptions are found
at 1481.2, 1463.9, 905.1, 751.9, and 679.2 cmÀ1 (Figure 1,
Table 1). If 18O2 is used in the FVP, these bands show only
very small isotopic shifts of less than 1 cmÀ1. However, a
3
gate the thermal reaction of 2 with O2 under the conditions
of matrix isolation (see Figure 1S in the Supporting Infor-
3
mation). The reaction of triplet carbenes with O2 had been
investigated previously in a similar way.[18–20] Theses reac-
tions proceed within several minutes and can be directly
monitored by IR spectroscopy. Warming a matrix containing
3
2 and excess O2 from 10 K to 35 K results in a rapid de-
crease of 2 and the concurrent formation of 1. The forma-
tion of 1 in a thermal reaction at temperatures as low as
30 K indicates a very small or no activation barrier for this
reaction, in accordance with theoretical predictions.[9]
Matrices containing 1 show a slight orange color caused
by a broad absorption with a maximum around 480 nm. This
finding is in accordance with the broad band in the visible
region observed in cavity-ring-down experiments.[8] Irradia-
tion of this band (argon, 10 K, l > 400 nm) rapidly results
in the disappearance of all bands assigned to 1 and forma-
tion of a new set of bands (Figure 3). The strongest IR band
of the new compound at 1726.9 cmÀ1 shows a strong 18O iso-
tope shift of À30.7 cmÀ1 and is thus assigned to a C=O
stretching vibration (Table 2,
see Figure 2S in the Supporting
Information). Other strong
bands are found at 1307.9,
1095.8, and 724.9 cmÀ1. The IR
spectra of a number of isomers
of 1 were calculated by using
DFT, and an excellent agree-
ment was found for the 2-oxe-
AHCTUNGTERGpNNUN inoxy radical 5. The IR fre-
quencies, intensities, and isotop-
ic shifts of all four isotopomers
closely match the calculated
data.
The rearrangement of 1 to 5
is obviously a multistep reac-
Scheme 1. Reaction of the phenyl radical 2 with molecular oxygen.
Chem. Eur. J. 2009, 15, 1462 – 1467
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1463

W. Sander and A. Mardyukov
Figure 1. IR spectra showing the products of the FVP of 4 in 2% O2-doped argon with subsequent trapping at 10 K. a) Spectrum obtained after FVP
with 16O2. b) Spectrum of 1 calculated at the UB3LYP/cc-pVTZ level of theory. c) Spectrum obtained after FVP with 18O2. d) Spectrum of 18O2-1 calcu-
lated at the UB3LYP/cc-pVTZ level of theory. Bands of remaining precursor 4 are marked *, bands of the phenyl radical 2 are marked & and bands of
*
benzene are marked
.
Table 1. IR spectroscopic data of four isotopomers of the phenyl peroxy radical 1.
Mode[a]
C6H516O16O (1)
C6H518O18O (18O2-1)
C6D516O16O ([D5]-1)
C6D518O18O ([D5]-18O2-1)
Sym. Assignment
argon[b]
DFT[c]
argon[b]
DFTc
argon[b]
DFT[c]
argon[b]
DFT[c]
33
32
31
30
28
26
25
23
20
19
18
17
13
11
10
9
3118.7
3189.1
3089.3
3070.8
–
3227.7 (3)
3201.9 (8)
3193.8 (18) 3086.9
3183.4 (15) 3070.5
3117.9
3187.2
3227.7 (3)
3201.9 (8)
3193.8 (18)
3183.4 (15) 2285.4
1643.4 (5) 1583.4 (9)
–
2387.7 (3)
2372.8 (15)
2362.5 (19)
2351.2 (19)
1612.2 (3)
–
–
–
–
2387.7 (3)
2372.8 (15)
2362.5 (19)
2351.2 (19)
1611.7 (3)
1382.4 (15)
1375.7 (23)
1336.6 (2)
1114.6 (7)
1106.2 (7)
829.6 (11)
822.5 (7)
783.6 (15)
748.6 (3)
642.6 (23)
562.1 (100)
586.6 (19)
423.1 (3)
A’
A’
A’
A’
A’
A’
A’
A’
A’
A’
A’
A’
A’’
A’
A’’
A’’
A’
A’
A’’
C2–H str.
2318.9
–
C4,5,6–H str.
C3,4,6–H str.
C3,5–H str.
1643.9 (5)
–
1582.4 (9)
CCC str. as.
1481.2 (52) 1514.2 (27) 1480.8 (55) 1513.9 (27) 1355.5 (17) 1384.1 (11) 1353.2 (30)
1463.9 (12) 1498.6 (8) 1462.7 (10) 1498.1 (8) 1347.6 (32) 1376.0 (23) 1346.5 (37)
1313.1 (3)
1122.9 (8)
-
1067.7 (4)
1018.5 (9)
905.1 (12)
793.9 (2)
C2,3,4,5–H def.
C4,5–H def.
1345.6 (1)
1175.1 (8)
1129.2 (0)
1098.1 (8)
1040.9 (5)
940.5 (10)
807.9 (1)
1312.4 (2)
1055.7 (8)
1109.6 (3)
1071.1 (2)
1017.3 (4)
904.9 (14)
778.3 (2)
751.3 (100) 777.1 (98)
679.2 (24)
–
1344.6 (1)
1108.1 (5)
1130.2 (5)
1096.1 (3)
1040.5 (5)
940.4 (10)
792.3 (1)
1297.7 (7)
1126.0 (8)
-
814.5 (9)
809.4 (9)
754.3 (21)
–
624.3 (20)
547.3 (100) 562.4 (96)
595.0 (28)
582.1 (11)
429.3 (37)
1336.8 (3)
1178.3 (11) 1068.1 (14)
1297.0 (7)
C3,5–H def./n str.
O–O str.
1112.8 (3)
830.6 (11)
824.0 (11)
783.8 (15)
761.6 (3)
1007.6 (3)
813.1 (18)
807.6 (14)
753.8 (26)
–
624.4 (21)
547.0 (100)
572.3 (20)
–
C1–O7 str/C2,6–H def.
C2,4,6–H def.
C2,3,5,6–H def.
C2,4,6–H wag.
C1–O7 str/C3C4C5 def.
C2,3,4,5,6–H wag.
C2,4,6–H wag.
ring def. in plane
C1O7O8 def.
751.9 (100) 777.3 (98)
642.7 (26)
679.2 (22)
615.1 (9)
607.2 (5)
480.8 (7)
702.9 (38)
628.5 (3)
621.8 (8)
499.3 (10)
702.9 (38)
626.7 (0)
604.8 (10)
498.4 (10)
8
7
6
608.0 (19)
597.9 (7)
433.8 (42)
595.4 (3)
479.7 (8)
429.5 (30)
433.2 (42)
ring def. out of plane
[a] Mode numbers based on the C6H516O16O isotopomer. [b] Argon matrix at 10 K. Wavenumbers in cmÀ1, relative intensities in parenthesis. [c] Calculat-
ed at the UB3LYP/cc-pVTZ level of theory.
tion. The formation of a spirodioxiranyl radical from 1 as an
intermediate which rearranges to 5 via a dioxy triradical has
been shown by computations to be a feasible mecha-
nism.[2,21,22] These and other proposed intermediates of the
rearrangement are expected to be highly labile and are not
observed during the photolysis of 1. According to DFT cal-
culations by Merle and Hadad the addition of a further mol-
ecule of oxygen at one of the three possible positions is en-
dothermic by 9–13 kcalmolÀ1 with activation barriers of 17–
18 kcalmolÀ1. Thus, a further addition of molecular oxygen
to 5 is not expected under the conditions of matrix isolation.
1464
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 1462 – 1467

Phenylperoxy Radical
FULL PAPER
Figure 2. IR spectra showing the products of the FVP of [D5]-4 in 2% O2-doped argon with subsequent trapping at 10 K. a) Spectrum obtained after
FVP with 16O2. b) Spectrum of [d5]-1 calculated at the UB3LYP/cc-pVTZ level of theory. c) Spectrum obtained after FVP with 18O2. d) Spectrum of
[D5]-18O2-1 calculated at the UB3LYP/cc-pVTZ level of theory. Bands of remaining precursor 4 are marked *, bands of the phenyl radical 2 are marked
*
& and bands of benzene are marked
.
siderable 18O-isotopic shifts, which suggests ketene groups
for the bands between 2150–2050 cmÀ1 and enones for the
bands between 1650–1600 cmÀ1. Since some of these bands
grow at different rates during the photolysis, we assume that
a mixture of at least two compounds is formed. It is there-
fore tempting to assign ketoketene 6 and its conformers and
configurational isomers to the photoproduct of 5. However,
since the IR spectra calculated for various geometrical iso-
mers of 6 are very similar, an assignment to one of the iso-
mers was not possible. In a study by Fadden and Hadad for
the ring-opening of 5 to ketoketene 6 a thermal activation
barrier of 23.4 kcalmolÀ1 was calculated,[23] while two other
rearrangements of 5 were found to proceed via considerably
higher activation barriers. Our experiments do not allow us
to determine if the ring-opening is a photochemical or a hot
ground state reaction. Regardless of this, the product ob-
served is that which is formed with the lowest predicted acti-
vation barrier.
Figure 3. Difference IR spectra showing the photochemistry (l>400 nm)
of 1, matrix-isolated in argon at 10 K. Bands pointing downwards are dis-
appearing during irradiation and assigned to 1. Bands pointing upwards
are appearing and assigned to 5. a) IR spectrum of 1 calculated at the
UB3LYP/cc-pVTZ level of theory. b) Difference IR spectrum of the 16O2
isotopomer. c) IR spectrum of 5 calculated at the UB3LYP/cc-pVTZ
level of theory.
Conclusion
Prolonged irradiation of matrix-isolated 5 at 10 K with
visible light (l > 400 nm) results in the complete photolysis
of 5 and formation of a new compound with intense IR ab-
sorptions in the region between 2150–2050 cmÀ1 and 1650–
1600 cmÀ1 (see Figure 4S and 5S, and Table 7S in the Sup-
porting Information). The bands in both regions show con-
The phenyl radical 2 rapidly reacts with molecular oxygen
to produce the phenoxylperoxy radical 1 both in the gas
phase and in argon matrices. Phenoxyl radical 3 is not
formed under these conditions. This indicates that the acti-
vation barrier for the oxygen addition to 2 is very low and
that, despite the high exothermicity of this reaction, the pri-
Chem. Eur. J. 2009, 15, 1462 – 1467
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1465

W. Sander and A. Mardyukov
Table 2. IR spectroscopic data of four isotopomers of the 2-oxepinoxy radical 5.
Mode[a]
C6H516O16O (5)
argon[b] DFTc
C6H518O18O (18O2-5)
argon[b] DFTc
C6D516O16O ([D5]-5)
argon[b] DFT[c]
C6D518O18O ([D5]-18O2-5) Sym. Assignment
argon[b] DFT[c]
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
1726.9 (100) 1790.0 (99) 1696.2 (100) 1754.4 (99) 1724.8 (100) 1787.7 (99) 1691.7 (100) 1751.4 (99)
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
C=O str.
1575.8 (13)
1515.8 (5)
1429.7 (1)
1415.8 (1)
1375.8 (9)
1307.9 (18)
1235.5 (2)
1177.4 (2)
1095.8 (25)
1029.6 (7)
–
1609.2 (14) 1574.8 (16)
1609.2 (14) 1533.3 (16)
1564.5 (16) 1531.3 (23)
1472.9 (15) 1438.1 (26)
1563.0 (17)
1467.0 (13)
1407.9 (4)
1249.9 (3)
1186.3 (6)
1057.5 (10)
1014.4 (5)
905.1 (0)
944.2 (0)
862.4 (0)
809.5 (0)
848.4 (2)
768.3 (1)
813.1 (0)
762.1 (4)
670.1 (0)
755.9 (1)
599.8 (1)
643.0 (4)
505.4 (2)
468.2 (2)
483.4 (0)
C5C6 str
1553.6 (3)
1472.6 (2)
1447.4 (1)
1412.8 (8)
1514.6 (1)
1428.3 (2)
–
1552.1 (2)
1471.3 (3)
1446.6 (0)
1410.5 (6)
1443.8 (12)
1361.6 (6)
1228.1 (10)
1177.2 (9)
C2C3 str
1408.1 (4)
1255.5 (4)
1193.6 (7)
1360.7 (10)
1222.2 (8)
1167.4 (13)
C2,3–H def./C3–C4 str.
C5,6–H def./ring str.
C2,3,4–H def./C1C2 str.
C5,6–H def./C1O8 str.
C2,3–H def.
1375.4 (5)
1339.6 (13) 1297.2 (15)
1329.8 (12) 1092.6 (33)
1079.2 (10) 1070.2 (52)
1259.2 (1)
1208.5 (1)
1234.2 (2)
–
1259.2 (1)
1206.1 (1)
1072.5 (13)
1039.1 (5)
1001.9 (0)
952.4 (0)
950.0 (0)
879.0 (3)
860.1 (3)
804.1 (1)
789.7 (2)
745.1 (12)
671.6 (5)
642.1 (3)
549.4 (0)
497.7 (0)
1010.1 (6)
–
–
–
1015.8 (5)
909.9 (0)
944.7 (0)
867.9 (0)
809.6 (0)
849.9 (2)
769.5 (2)
815.8 (0)
764.2 (3)
671.1 (0)
768.4 (3)
600.3 (1)
657.7 (4)
507.4 (2)
468.5 (2)
498.5 (0)
1009.7 (15)
–
–
–
C4,5–H def.
1083.9 (14) 1083.4 (10)
C1O8 str./C4,5–H def.
ring str.
1048.7 (4)
1002.0 (0)
953.6 (0)
950.0 (0)
887.9 (2)
860.5 (3)
805.7 (1)
805.6 (2)
746.2 (11)
687.5 (5)
644.3 (3)
550.6 (1)
513.4 (0)
1016.7 (3)
–
–
–
C2,3,4–H wag.
ring str.
–
–
837.2 (4)
836.8 (7)
–
–
760.2 (3)
–
C5,6–H wag.
888.5 (6)
833.4 (3)
797.9(2)
–
724.9 (19)
694.2 (4)
623.9 (5)
527.8 (2)
–
881.3 (8)
832.5 (6)
796.3 (1)
784.0 (4)
723.9 (22)
–
621.7 (5)
–
–
–
–
ring str.
748.5 (6)
–
–
584.6 (3)
669.8 (5)
–
746.4 (6)
–
–
584.4 (6)
659.2 (7)
–
ring str./C2,4,6–H wag.
C2,3,4,5,6–H wag
C2C1O8 def
C4,5,6–H wag
C1O8 str
8
7
6
C2,3,4,5–H wag
ring def. out of plane
C1O7 def.
450.9 (5)
–
450.6 (7)
–
[a] Mode numbers based on the C6H516O16O isotopomer. [b] Argon matrix at 10 K. Wavenumbers in cmÀ1, relative intensities in parenthesis. [c] Calculat-
ed at the UB3LYP/cc-pVTZ level of theory.
13C NMR (50.33 MHz, CDCI3, 258C, TMS): dC =152.99, 78.06, 77.43,
76.79 ppm; MS (RI, 70 eV): (m/z,%): 192 [M+], 110, 82, 54, 45.
mary product 1 can efficiently dissipate the excess energy
without fragmentation. Loss of oxygen from 1 obviously re-
quires a higher thermal excitation in high-temperature pro-
cesses or in cross-beam experiments. This is of relevance to
tropospheric chemistry, since the phenyl radical 2 can be
produced in the troposphere from benzene via H-abstrac-
tion by OH radicals. The phenyl radical 2 will be efficiently
quenched by oxygen to form peroxy radical 1. Under these
conditions it is not expected that 1 cleaves to phenoxyl radi-
cal 3 and O(3P). Formation of triplet oxygen atoms would
regenerate ozone and subsequently OH radicals. The photo-
chemical rearrangement of 1 by visible light produces anoth-
er intermediate of importance in the oxidation of benzene:
the 2-oxepinoxy radical 5. This radical was also character-
ized by IR spectroscopy. Further irradiation finally leads to
the cleavage of the ring and formation of a mixture of sever-
al conformers of ketoketene 6. Our experiments do not
allow us to determine if the photochemical steps which lead
to 5 and 6 proceed on an excited surface or if these reac-
tions are hot ground state reactions. Nevertheless, the prod-
ucts found in the matrix are the products that have been
predicted by ab initio and DFT calculations to be formed
with the smallest activation barriers. These products are
thus also relevant for the thermal combustion of benzene.
Matrix isolation: Matrix isolation experiments were performed by stan-
dard techniques[25] using a closed-cycle helium cryostat and a CsI spectro-
scopic window cooled to 10 K. FTIR spectra were recorded with a stan-
dard resolution of 0.5 cmÀ1, using a N2(l)-cooled MCT detector in the
range 400–4000 cmÀ1. UV spectra were recorded in the spectroscopic
range 800 to 200 nm with a standard resolution of 0.02 nm, using a
Varian UV/Vis NIR spectrophotometer from a sample deposited on a
sapphire window cooled to 10 K by a closed cycle cryostat. Flash vacuum
pyrolysis (FVP) was carried out by slowly subliming azobenzene
4
through a 7 cm quartz tube heated electrically with a tantalum wire.
Broadband irradiation was carried out with mercury high-pressure arc
lamps in housings equipped with quartz optics and dichroic mirrors in
combination with cutoff filters (50% transmission at the wavelength
specified).
Computational methods: Optimized geometries and vibrational frequen-
cies of all species were calculated at the UB3LYP[26–28] level of theory
using the 6–311+GACTHNUTRGNEUNG
(d,p) polarized valence-triple- basis set[29,30] and Dun-
ningꢃs cc-pVTZ[31–33] basis sets. All DFT calculations were carried out
with Gaussian 03.[34]
Acknowledgement
This work was financially supported by the Deutsche Forschungsgemein-
schaft and the Fond der Chemischen Industrie.
[1] C. Barckholtz, M. J. Fadden, C. M. Hadad, J. Phys. Chem. A 1999,
103, 8108–8117.
[2] M. J. Fadden, C. Barckholtz, C. M. Hadad, J. Phys. Chem. A 2000,
104, 3004–3011.
Experimental Section
Azobenzene and [D10]azobenzene: Azobenzene
99.8%) was used without further purification. The perdeuterated azoben-
zene [D10]-4 was synthesized according to
literature procedure.[24]
4
(Acros Organics
[3] G. D. Edwards, C. A. Cantrell, S. Stephens, B. Hill, O. Goyea, R. E.
Shetter, R. L. Mauldin, III, E. Kosciuch, D. J. Tanner, F. L. Eisele,
Anal. Chem. 2003, 75, 5317–5327.
a
1466
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 1462 – 1467

Phenylperoxy Radical
FULL PAPER
[4] K. Tonokura, Y. Norikane, M. Koshi, Y. Nakano, S. Nakamichi, M.
Goto, S. Hashimoto, M. Kawasaki, M. P. S. Andersen, M. D. Hurley,
T. J. Wallington, J. Phys. Chem. A 2002, 106, 5908–5917.
[5] S. J. Blanksby, G. B. Ellison, Acc. Chem. Res. 2003, 36, 255–263.
[6] I. V. Tokmakov, G.-S. Kim, V. V. Kislov, A. M. Mebel, M. C. Lin, J.
Phys. Chem. A 2005, 109, 6114–6127.
[7] J. K. Merle, C. M. Hadad, J. Phys. Chem. A 2004, 108, 8419–8433.
[8] T. Yu, M. C. Lin, J. Am. Chem. Soc. 1994, 116, 9571–9576.
[9] G. da Silva, J. W. Bozzelli, J. Phys. Chem. A 2008, 112, 3566–3575.
[10] X. Gu, F. Zhang, R. I. Kaiser, Chem. Phys. Lett. 2007, 448, 7–10.
[11] X. Fang, R. Mertens, C. von Sonntag, J. Chem. Soc. Perkin Trans. 2
1995, 1033–1036.
[20] W. Sander, G. Bucher, S. Wierlacher, Chem. Rev. 1993, 93, 1583–
1621.
[21] B. K. Carpenter, J. Am. Chem. Soc. 1993, 115, 9806–9807.
[22] C. Barckholtz, M. J. Fadden, C. M. Hadad, J. Phys. Chem. A 1999,
103, 8108–8117.
[23] M. J. Fadden, C. M. Hadad, J. Phys. Chem. A 2000, 104, 8121–8130.
[24] H. E. Bigelow, D. B. Robinson, Org. Synth. 1942, 22, 28–29.
[25] I. Dunkin, Matrix Isolation Techniques: A Practical Approach, 1998,
pp. 256.
[26] A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652.
[27] C. Lee, W. Yang and R. G. Parr, Phys. Rev. B 1988, 37, 785–789.
[28] B. Miehlich, A. Savin, H. Stoll, H. Preuss, Chem. Phys. Lett. 1989,
157, 200–206.
[12] S. Naumov, C. von Sonntag, J. Phys. Org. Chem. 2005, 18, 586–594.
[13] A. V. Friderichsen, J. G. Radziszewski, M. R. Nimlos, P. R. Winter,
D. C. Dayton, D. E. David and G. B. Ellison, J. Am. Chem. Soc.
2001, 123, 1977–1988.
[29] A. D. McLean, G. S. Chandler, J. Chem. Phys. 1980, 72, 5639–5648.
[30] R. Krishnan, J. S. Binkley, R. Seeger, J. A. Pople, J. Chem. Phys.
1980, 72, 650–654.
[14] A. Lapinski, J. Spanget-Larsen, M. Langgard, J. Waluk, J. G. Radzis-
zewski, J. Phys. Chem. A 2001, 105, 10520–10524.
[15] J. G. Radziszewski, Chem. Phys. Lett. 1999, 301, 565–570.
[16] J. Pacansky, J. Bargon, J. Am. Chem. Soc. 1975, 97, 6896–6897.
[17] J. G. Radziszewski, M. R. Nimlos, P. R. Winter, G. B. Ellison, J. Am.
Chem. Soc. 1996, 118, 7400–7401.
[31] D. E. Woon, T. H. Dunning, Jr., J. Chem. Phys. 1993, 98, 1358–1371.
[32] R. A. Kendall, T. H. Dunning, Jr., R. J. Harrison, J. Chem. Phys.
1992, 96, 6796–6806.
[33] T. H. Dunning, Jr., J. Chem. Phys. 1989, 90, 1007–1023.
[34] M. J. e. a. G. G. Frisch, Inc., Pittsburgh, PA, 2003.
[18] W. Sander, Angew. Chem. 1986, 98, 255–256; Angew. Chem. Int. Ed.
Engl. 1986, 25, 255–257.
Received: July 29, 2008
Revised: October 16, 2008
[19] W. W. Sander, J. Org. Chem. 1989, 54, 333–339.
Published online: January 8, 2009
Chem. Eur. J. 2009, 15, 1462 – 1467
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1467