Generation of fluorinated m-benzyne derivatives in neon matrices

Article abstract of DOI:10.1002/1099-0690(200212)2002:23<3927::AID-EJOC3927>3.0.CO;2-R

The fluorinated m-didehydrobenzenes 5-H-, 5-deuterio-, 5-(trifluoromethyl)-, 5-(trimethylsilyl)-, and 5-iodo-2,4,6-trifluoro-1,3-didehydrobenzene 6b-6f were generated by UV photolysis of the corresponding 1,3-diiodobenzene derivatives 4b-4f in solid neon

Full text of DOI:10.1002/1099-0690(200212)2002:23<3927::AID-EJOC3927>3.0.CO;2-R

FULL PAPER
Generation of Fluorinated m-Benzyne Derivatives in Neon Matrices
Hans Henning Wenk[a] and Wolfram Sander*[a]
Keywords: Benzynes / Density functional calculations / IR spectroscopy / Matrix isolation / Radicals
The fluorinated m-didehydrobenzenes 5-H-, 5-deuterio-, 5-
(trifluoromethyl)-, 5-(trimethylsilyl)-, and 5-iodo-2,4,6-tri-
fluoro-1,3-didehydrobenzene 6b−6f were generated by UV
photolysis of the corresponding 1,3-diiodobenzene derivat-
the matrix at 7.5 K results in a decrease in all IR absorptions
assigned to the radical products and in back-formation of the
diiodobenzenes. The influence of the substituents in the 5-
position on the geometries and the electronic structures of
ives 4b−4f in solid neon at 3 K. The photolysis of 1 proceeds the diradicals 6 is discussed. The calculations indicate that
stepwise via the phenyl radical intermediates 5b−5f. The rad- the singlet-triplet splittings of the substituted m-benzynes
icals 5 and m-benzynes 6 were identified by comparison of
their experimentally measured IR spectra with those pro-
duced by DFT calculations. The formation of mono- and di-
6b−6f are larger and the C1−C3 distances shorter than those
in the parent m-benzyne (C6H4).
( Wiley-VCH Verlag GmbH, 69451 Weinheim, Germany,
radicals 5 and 6, respectively, is reversible, and annealing of 2002)
Introduction
The three isomeric didehydrobenzenes (benzynes) are in-
teresting reactive species that have been discussed as inter-
mediates in a number of reaction mechanisms. o-Benzyne
(1) and its derivatives have found many applications in or-
ganic synthesis, and its chemical and spectroscopic proper-
ties are well understood. m-Benzyne (2) is much less well
understood, and substituent effects on the electronic[1] and
spectroscopic[2,3] properties and on the reactivity[4Ϫ6] of m-
benzynes were reported only recently.
So far, the application of matrix isolation spectroscopy
for the investigation of derivatives of m- and p-benzyne has
been hampered by a lack of suitable photochemical pre-
cursors. Photolysis of matrix-isolated radical precursors al-
ways produces radical pairs in matrix cages. If the radical
pairs are not separated by inert molecules such as photo-
chemically formed fragments (CO2 in the photolysis of di-
acyl peroxides, for example) or the matrix gas, the radical
pair will rapidly recombine. To circumvent this problem,
meta-para-cyclophanediones[3] have been used as precursors
of m-benzyne (2) and several of its derivatives. In this case
the photochemical fragmentation produces the benzyne and
the relatively stable p-xylylene instead of radicals. A disad-
vantage of this precursor is that only very limited substitu-
tion patterns of the cyclophane are synthetically available.
Substituted diiodobenzenes are readily available and
should be suitable precursors of substituted m- and p-
benzynes. However, the recombination of iodine atoms with
the radical fragments in solid argon is rapid and prevents
the formation of detectable amounts of diradicals. In con-
trast, irradiation in solid neon at very low temperatures
(3 K) produced the monoradical and benzyne derivatives in
reasonable yields.[2,7] The formation of the benzynes de-
pends not only on the matrix material and temperature, but
also on the substituents on the diiodobenzene. Thus, fluor-
inated diiodobenzenes produce benzynes in moderate to
high yields, while the parent 1,3- and 1,4-diiodobenzene
yield only traces of the corresponding benzynes. The reason
for this striking difference in product formation is not yet
clear. Here we describe the photochemical synthesis and
spectroscopic characterization of the substituted phenyl
radicals 5bϪf and the m-benzynes 6bϪf.
[a]
Lehrstuhl für Organische Chemie II der Ruhr-Universität,
44780 Bochum, Germany
Fax: (internat.) ϩ 49-(0)234/321-8593
E-Mail: wolfram.sander@ruhr-uni-bochum.de
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Eur. J. Org. Chem. 2002, 3927Ϫ3935  2002 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim 1434Ϫ193X/02/1223Ϫ3927 $ 20.00ϩ.50/0
3927

H. H. Wenk, W. Sander
FULL PAPER
is supported by DFT calculations at the BLYP/6-311G(d,p)
level of theory (Figure 1, Table 2).
Results and Discussion
1,3-Didehydro-2,4,6-trifluorobenzene (6b)
Irradiation (5 min, 254 nm) of 2,4,6-trifluoro-1,3-diiodo-
benzene (4b) in a neon matrix at 3 K results in a reduction
of intensity in the IR bands of 4b and concurrent formation
of a new compound A which was identified as the 2,4,6-
trifluoro-3-iodophenyl radical 5b, formed by the photo-
chemical cleavage of one of the CϪI bonds. This assign-
ment is confirmed by comparison with the spectrum calcu-
lated at the B3LYP/6-311G(d,p) level of theory, which
nicely reproduces the observed IR spectrum (Table 1). The
most intense absorption at 1556.7 cmϪ1 (calculated at
1585.2 cmϪ1) is due to a CϪC stretching vibration of the
aromatic ring system.
Table 1. IR spectroscopic data of the 2,4,6-trifluoro-3-iodophenyl
radical 5b
Figure 1. IR spectra illustrating the photochemistry of 2,4,6-tri-
fluoro-1,3-diiodobenzene (4b) in solid neon at 3 K; (a) difference
spectrum; bands pointing upward increase, bands pointing down-
ward decrease on irradiation (10 min, 254 nm); increasing absorp-
tions are assigned to monoradical 5b; (b) difference spectrum;
bands pointing upward increase, bands pointing downward de-
crease on annealing the neon matrix (7.5 K, 10 min) after 16 h of
irradiation with 254 nm; bands marked with ° are assigned to di-
radical 6b; (c) calculated [BLYP/6-311G(d,p), unscaled] spectrum
of 1,3-didehydro-2,4,6-trifluorobenzene (6b)
[b]
[b,c]
Mode Sym. ν˜exp. [cmϪ1 [a]
]
Irel,exp.
ν˜calcd. [cmϪ1 [c]
]
Irel,calcd.
9
10
11
12
13
14
15
aЈ
aЈ
Ϫ
Ϫ
aЈ
a"
a"
501.4
561.0
Ϫ
Ϫ
Ϫ
0.06
0.05
Ϫ
Ϫ
Ϫ
509.6
565.1
582.7
583.6
0.02
0.08
0.03
0.02
0.00
0.01
0.03
Ϫ
Ϫ
Ϫ
Ϫ
621.3
680.2
Table 2. IR spectroscopic data of 1,3-didehydro-2,4,6-trifluoroben-
zene S-6b
831.2
836.1
985.3
1039.1
1046.9
1147.0
1272.6
1365.9
1408.7
1425.4
1556.7
1619.7
1624.6
Ϫ
0.10
0.10
0.93
0.56
0.22
0.26
0.10
0.07
0.48
0.67
1.00
0.30
0.20
Ϫ
16
a"
829.2
0.20
17
18
19
20
21
22
23
24
25
aЈ
aЈ
aЈ
aЈ
aЈ
aЈ
aЈ
aЈ
aЈ
992.8
1047.3
1058.6
1162.3
1314.3
1383.1
1423.5
1443.2
1585.3
0.44
0.23
0.51
0.43
0.13
0.22
0.58
0.42
1.00
[b]
[b,c]
Mode Sym. ν˜exp. [cmϪ1 [a]
]
Irel,exp.
ν˜calcd. [cmϪ1 [c]
]
Irel,calcd.
7
8
9
b1
b2
a1
b2
a2
b1
a1
b1
b2
a1
b2
b2
a1
b2
a1
b2
a1
a1
488.1
492.7
Ϫ
Ϫ
Ϫ
0.07
0.07
Ϫ
Ϫ
Ϫ
470.1
498.6
542.9
547.2
569.1
607.4
750.8
842.0
874.8
1014.9
1110.2
1242.7
1331.3
1401.3
1414.8
1545.6
1788.3
3123.7
0.04
0.05
0.00
0.00
0.00
0.00
0.20
0.09
0.60
0.55
0.33
0.11
0.04
0.07
0.10
0.85
1.00
0.01
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Ϫ
Ϫ
26
27
aЈ
aЈ
1650.3
3209.0
0.73
0.02
768.5
872.9
916.3
1061.4
1145.8
1299.4
1382.0
Ϫ
1461.4
1565.0
1770.7
Ϫ
0.04
0.05
0.53
0.41
0.38
0.15
0.12
Ϫ
0.19
1.00
0.48
Ϫ
[a] Neon, 3 K. [b] Relative intensities based on the strongest absorp-
tion. [c] Calculated at the B3LYP/6-311G(d,p) level of theory. Band
positions are unscaled. The assignment is tentative and is based on
band positions and intensities.
Prolonged irradiation results in a further decrease in 4b,
the yield of the 2,4,6-trifluoro-3-iodophenyl radical 5b in-
creases, and a further species B appears in the IR spectrum.
The formation of both phenyl radical 5b and the secondary
photoproduct B is thermally reversible. Several minutes an-
nealing of the matrix at 7.5 K reforms the 2,4,6-trifluoro-
1,3-diiodobenzene (4b). The IR spectrum of B exhibits a
broad absorption at 1769 cmϪ1, similar to the a1 symmet-
rical ring stretching vibration at 1823 cmϪ1 observed for
[a] Neon, 3 K. [b] Relative intensities based on the strongest absorp-
[c]
tion. Calculated at the BLYP/6-311G(d,p) level of theory. Band
positions are unscaled. The assignment is tentative and is based on
band positions and intensities.
Comparison of the experimentally measured IR spectrum
1,3-didehydro-2,4,5,6-tetrafluorobenzene (6a). From the with the calculated one confirms that the band at
stepwise formation, the thermal reversibility, and the ana- 1769 cmϪ1 is indeed due to an a1-symmetrical vibration pre-
logous reaction of the perfluorinated derivative 6a, the sec- dicted at 1788.3 cmϪ1. This absorption is characteristic of
ondary product is assigned the structure of the diradical m-benzynes with a fluorine substituent in the 2-position.
1,3-didehydro-2,4,6-trifluorobenzene (6b). This assignment The corresponding b2-symmetrical combination at
3928
Eur. J. Org. Chem. 2002, 3927Ϫ3935

Generation of Fluorinated m-Benzyne Derivatives in Neon Matrices
FULL PAPER
1565 cmϪ1 (calculated at 1545.6 cmϪ1) is the dominant ab-
sorption in the IR spectrum of 6b. A b2-symmetrical, in-
plane ring deformation with strong contribution of the rad-
ical centers is found at 492.7 cmϪ1 (calculated at
498.6 cmϪ1). This band is not much affected by substitution
and has been observed in several m-benzyne derivatives: at
547 cmϪ1 in 1,3-didehydrobenzene,[8] at 543 cmϪ1 in 1,3-di-
dehydro-5-fluorobenzene,[3] and at 511 cmϪ1 in 1,3-didehy-
dro-2,4,5,6-tetrafluorobenzene.[2] The intensity of this ab-
sorption in the fluorinated derivatives is rather low com-
pared to that in the parent m-benzyne. The other intense
bands of 6b are due to in-plane ring deformation vibrations
and combinations of these with ν(CϪF) or ν(CϪH) vibra-
tions.
The monodeuterated benzyne 6c and phenyl radical 5c
were generated analogously, by photolysis of 4c (Table 3).
The most pronounced isotopic shift in 5c is observed for
the δoop(CϪH) vibration [ν˜D/ν˜H ϭ 0.88, ν˜ ϭ 836.1 cmϪ1
Figure 2. IR spectra illustrating the photochemistry of 5-deuterio-
2,4,6-trifluoro-1,3-diiodobenzene (4c) in solid neon at 3 K; (a) dif-
ference spectrum; bands pointing upward increase, bands pointing
downward decrease on irradiation (10 min, 254 nm); increasing ab-
sorptions are assigned to monoradical 5c; (b) difference spectrum;
bands pointing upward increase, bands pointing downward de-
crease on annealing the neon matrix (7.5 K, 10 min) after 16 h of
irradiation with 254 nm; bands marked with ° are assigned to di-
radical 6c; (c) calculated [BLYP/6-311G(d,p), unscaled] spectrum
of 1,3-didehydro-5-deuterio-2,4,6-trifluorobenzene (6c)
for RϪH, ν ϭ 737.2 cmϪ1 for RϪD]. The CϪH and CϪD
˜
stretching vibrations of 5b and 5c, respectively, are of low
intensity and could not be unequivocally assigned.
Table 3. IR spectroscopic data of the 5-deuterio-2,4,6-trifluoro-3-
iodophenyl radical 5c
[b]
[b,c]
Mode Sym. ν˜exp. [cmϪ1 [a]
]
Irel,exp.
ν˜calcd. [cmϪ1 [c]
]
Irel,calcd.
Table 4. IR spectroscopic data of 1,3-didehydro-5-deuterio-2,4,6-
trifluorobenezne S-6c
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
aЈ
aЈ
aЈ
a"
a"
a"
aЈ
aЈ
aЈ
aЈ
aЈ
aЈ
aЈ
aЈ
aЈ
aЈ
aЈ
544.9
Ϫ
Ϫ
Ϫ
Ϫ
0.14
Ϫ
Ϫ
Ϫ
Ϫ
548.0
580.4
597.0
620.9
667.4
727.7
867.5
997.0
1050.5
1076.6
1311.9
1379.7
1414.5
1424.9
1574.0
1644.8
2373.6
0.08
0.01
0.02
0.02
0.01
0.13
0.01
0.36
0.70
0.47
0.14
0.24
0.57
0.63
1.00
0.77
0.02
[b]
[b,c]
Mode Sym. ν˜exp. [cmϪ1 [a]
]
Irel,exp.
ν˜calcd. [cmϪ1 [c]
]
Irel,calcd.
13
14
15
16
17
18
19
20
21
22
b1
a1
b2
b2
a1
b2
a1
b2
a1
b2
734.5
761.1
796.7
1021.2
1052.6
1285.0
1377.1
Ϫ
1449.0
1556.8
1766.3
1769.8
Ϫ
0.07
0.10
0.18
0.43
0.97
0.24
0.21
Ϫ
0.20
0.23
0.51
0.49
Ϫ
703.6
741.7
776.2
0.04
0.18
0.24
0.55
0.56
0.24
0.06
0.11
0.09
0.78
737.2
Ϫ
0.12
Ϫ
973.8
992.2
1039.5
1066.3
1270.7
1366.1
1388.6
1410.4
1544.4
1614.7
Ϫ
0.20
0.66
0.33
0.11
0.17
0.14
0.68
1.00
0.41
Ϫ
1004.2
1223.4
1323.2
1382.7
1402.9
1536.0
23
24
a1
a1
1787.8
2313.6
1.00
0.00
[a] Neon, 3 K. [b] Relative intensities based on the strongest absorp-
tion. Calculated at the BLYP/6-311G(d,p) level of theory. Band
positions are unscaled. The assignment is tentative and is based on
band positions and intensities.
[a] Neon, 3 K. [b] Relative intensities based on the strongest absorp-
tion. [c] Calculated at the B3LYP/6-311G(d,p) level of theory. Band
positions are unscaled. The assignment is tentative and is based on
band positions and intensities.
[c]
3123.7 and 2313.6 cmϪ1, respectively) could not be identi-
fied in the spectra due to their low intensity.
The IR spectrum of 6c agrees well with the calculated
spectrum (Figure 2, Table 4), and the experimental isotopic
˜
˜
shifts νD/νH are accurately reproduced by these calcula-
1,3-Didehydro-2,4,6-trifluoro-5-trifluoromethylbenzene (6d)
tions (Table 5).
The largest isotopic shifts are observed for a b1-symmet-
As a possible explanation for the formation of benzenoid
rical δoop(CϪH) vibration (mode 14 in 6b, 13 in 6c), the diradicals in solid neon, rotation of the photochemically
corresponding b2-symmetrical δip(CϪH) vibration (mode generated radical fragments inside the matrix cage, which
15), and a combination of the latter with a ring deformation prevents recombination, has been devised by Radziszew-
(mode 17 in 6b, 16 in 6c). The a1-symmetrical ring stretch- ski.[9] According to this theory, bulky substituents should
ing mode 23 is hardly affected by isotopic labelling, while hinder the rotation of the mono- and diradicals, thereby
the b2-symmetrical mode is shifted by 8 cmϪ1 to 1557 cmϪ1
Again, the ν(CϪH) and ν(CϪD) vibrations (calculated at troduction of a trifluoromethyl group in the 5-position is
.
lowering the yield of didehydrobenzene derivatives. The in-
Eur. J. Org. Chem. 2002, 3927Ϫ3935
3929

H. H. Wenk, W. Sander
FULL PAPER
Table 5. Calculated and observed ν˜D/ν˜H values for diradicals 6b
and 6c
Symm.
Mode RϪH
Mode RϪD
(ν˜D/ν˜H)calcd.
(ν˜D/ν˜H)exp.
Assignment
a1
b1
b2
a1
b2
b2
a1
b2
a1
b2
a1
13
14
15
16
17
18
19
20
21
22
23
14
13
15
17
16
18
19
20
21
22
23
0.99
0.84
0.89
0.99
0.88
0.98
0.99
0.99
0.99
0.99
1.00
0.99
0.84
0.87
0.99
0.89
0.99
1.00
Ϫ
ν(CϪF) ϩ ring def.
δoop(CϪH)
δip(CϪH) ϩ ring def.
ν(CϪF) ϩ ring def.
δip(CϪH)
δip(CϪH) ϩ ring def.
ν(CϪF) ϩ ring def.
δip(CϪH) ϩ ν(CϪC)
ν(CϪF) ϩ ν(CϪC)
δip(CϪH) ϩ ν(CϪC)
ν(CϪF) ϩ ν(CϪC)
0.99
0.99
1.00
suitable for study of the steric effects in comparison with
Continued irradiation generates another species, consist-
those affecting the perfluoro derivative. The radical sites are ently with the stepwise cleavage of the two CϪI bonds of
still surrounded by fluorine atoms (which facilitate the diiodobenzene derivative 4d. Annealing at 7 K reforms
formation of the diradical), while the electronic properties starting material 4d at the expense of monoradical 5d and
of the molecule are relatively unaffected by the replacement diradical 6d, which confirms the assignment of the second
of a fluorine atom by the electron-withdrawing CF3 group. product as 1,3-didehydro-2,4,6-trifluoro-5-(trifluoromethyl)-
Again, short-term photolysis (λ ϭ 254 nm) of 2,4,6-tri- benzene. The agreement of the observed and the calculated
fluoro-1,3-diiodo-5-(trifluoromethyl)benzene (4d), matrix- IR spectra is good, and, as in the case of the m-benzyne
isolated in neon at 3 K, results in a decrease in the pre- derivatives described above, the characteristic a1-symmet-
cursor concentration and the formation of a new species. rical ring stretching vibration (observed: 1795.1 and
The product is assigned as 2,4,6-trifluoro-3-iodo-5-(trifluo- 1805.3 cmϪ1, calculated: 1816.8 cmϪ1) is identified in the
romethyl)phenyl radical 5d by comparison with the calcu- spectrum, as well as the corresponding b2-symmetrical vi-
lated IR spectrum (Table 6). The spectrum is dominated by bration (observed: 1569.5 cmϪ1, calculated: 1548.4 cmϪ1),
the ν(CϪCF3) vibration, which is split into two compon- which is the most intense band of 6d (Figure 3, Table 7).
ents: at 1273.7 and 1282.8 cmϪ1 [calculated at 1273.6 cmϪ1
B3LYP/6-311G(d,p)].
,
Table 6. IR spectroscopic data of the 2,4,6-trifluoro-3-iodo-5-(tri-
fluoromethyl)phenyl radical 5d
Mode Sym. ν˜exp. [cmϪ1 [a]
]
Irel,exp.
ν˜calcd. [cmϪ1 [c]
]
Irel,calcd.
[b]
[b,c]
19
20
21
22
23
24
25
26
27
28
29
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
617.6
Ϫ
Ϫ
699.9
Ϫ
Ϫ
1025.2
1070.5
1077.3
1164.7
1168.2
1273.7
1282.8
Ϫ
Ϫ
1416.0
Ϫ
1543.2
1548.7
1630.6
0.06
Ϫ
Ϫ
0.04
Ϫ
Ϫ
0.30
0.03
0.14
0.19
0.22
0.69
0.31
Ϫ
Ϫ
0.32
Ϫ
0.05
0.07
0.15
618.6
620.3
659.1
701.8
746.8
0.07
0.00
0.01
0.02
0.01
0.01
0.23
0.04
0.08
0.48
0.16
777.0
1031.1
1079.4
1083.0
1156.9
1175.7
Figure 3. IR spectra illustrating the photochemistry of 2,4,6-tri-
fluoro-1,3-diiodo-5-(trifluoromethyl)benzene (4d) in solid neon at
3 K; (a) difference spectrum; bands pointing upward increase,
bands pointing downward decrease on irradiation (10 min,
254 nm); increasing absorptions are assigned to monoradical 5d;
(b) difference spectrum; bands pointing upward increase, bands
pointing downward decrease on annealing the neon matrix (7.5 K,
10 min) after 16 h of irradiation with 254 nm; bands marked with
° are assigned to diradical 6d; (c) calculated [BLYP/6-311G(d,p),
unscaled] spectrum of 1,3-didehydro-2,4,6-trifluoro-5-trifluorome-
thylbenzene (6d)
30
Ϫ
1273.6
1.00
31
32
33
34
Ϫ
Ϫ
Ϫ
Ϫ
1314.1
1389.8
1427.0
1459.9
0.00
0.01
0.27
0.04
35
36
Ϫ
Ϫ
1570.7
1657.7
0.31
0.31
[a] Neon, 3 K. [b] Relative intensities based on the strongest absorp-
tion. [c] Calculated at the B3LYP/6-311G(d,p) level of theory. Band
positions are unscaled. The assignment is tentative and is based on
band positions and intensities.
The formation of the sterically more demanding diradical
6d shows that the mere size of the molecule apparently has
3930
Eur. J. Org. Chem. 2002, 3927Ϫ3935

Generation of Fluorinated m-Benzyne Derivatives in Neon Matrices
FULL PAPER
Table 7. IR spectroscopic data of 1,3-didehydro-2,4,6-trifluoro-5-
trifluoromethylbenzene S-6d
Table 8. IR spectroscopic data of the 2,4,6-trifluoro-3-iodo-5-(tri-
methylsilyl)phenyl radical 5e
[b]
[b,c]
[b]
[b,c]
Mode Sym. ν˜exp. [cmϪ1 [a]
]
Irel,exp.
ν˜calcd. [cmϪ1 [c]
]
Irel,calcd.
Mode Sym. ν˜exp. [cmϪ1 [a]
]
Irel,exp.
ν˜calcd. [cmϪ1 [c]
]
Irel,calcd.
20
21
22
23
24
25
26
27
28
29
30
31
32
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
727.0
Ϫ
861.3
990.9
Ϫ
1174.9
Ϫ
1270.2
1287.9
Ϫ
Ϫ
Ϫ
1569.5
1795.1
1805.3
0.32
Ϫ
0.33
0.51
Ϫ
0.63
Ϫ
0.62
0.71
Ϫ
Ϫ
Ϫ
1.00
0.17
0.41
687.3
704.3
825.2
0.10
0.00
0.38
0.63
0.01
0.69
0.18
1.00
0.51
0.03
0.15
0.01
0.69
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
a"
aЈ
aЈ
a"
aЈ
aЈ
aЈ
aЈ
a"
aЈ
aЈ
aЈ
aЈ
aЈ
aЈ
a"
aЈ
a"
aЈ
a"
aЈ
aЈ
aЈ
769.5
Ϫ
0.12
Ϫ
789.0
789.9
878.4
882.9
886.4
0.11
0.04
0.22
0.32
0.84
0.34
0.04
1.00
0.15
0.11
0.14
0.08
0.01
0.55
0.14
0.00
0.00
0.00
0.03
0.02
0.03
0.51
0.70
849.7
856.0
862.8
1008.7
1046.5
1082.6
1258.1
Ϫ
1260.9
1268.7
Ϫ
1380.1
1403.6
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
1514.2
1597.6
0.12
0.16
0.11
0.40
0.16
1.00
0.16
Ϫ
0.08
0.08
Ϫ
0.31
0.09
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
0.38
0.21
945.8
1034.3
1071.7
1094.6
1211.3
1227.9
1355.5
1379.2
1463.5
1548.4
1011.9
1051.4
1083.7
1296.9
1297.3
1300.0
1311.0
1342.8
1395.6
1426.8
1453.2
1458.1
1462.5
1468.1
1469.8
1480.2
1539.1
1638.9
33
Ϫ
1816.8
0.87
[a] Neon, 3 K. [b] Relative intensities based on the strongest absorp-
tion. Calculated at the BLYP/6-311G(d,p) level of theory. Band
positions are unscaled. The assignment is tentative and is based on
band positions and intensities.
[c]
less influence than other substituent effects on the photo-
chemistry of diiodobenzene derivatives in neon matrices.
[a] Neon, 3 K. [b] Relative intensities based on the strongest absorp-
tion. [c] Calculated at the B3LYP/6-311G(d,p) level of theory. Band
positions are unscaled. The assignment is tentative and is based on
band positions and intensities.
1,3-Didehydro-2,4,6-trifluoro-5-(trimethylsilyl)benzene (6e)
The successive cleavage of two CϪI bonds could also be
demonstrated in trimethylsilyl derivative 4e, although with
lower efficiency than with the other derivatives so far re-
ported. The IR spectra of the 2,4,6-trifluoro-3-iodo-5-(tri-
methylsilyl)phenyl radical 5e (formed by 5 min of irradi-
ation with 254 nm, Table 8) and 1,3-didehydro-2,4,6-tri-
fluoro-5-(trimethylsilyl)benzene (6e) (formed by long-term
irradiation with 254 nm, together with additional amounts
of monoradical 5e, Figure 4, Table 9) could be measured
and assigned by comparison with calculated spectra. The
reverse reaction was observed on annealing, diiodide 4e
once more being formed. The a1-symmetrical ring stretch-
ing vibration described above appears as two broad bands
at 1765.3 and 1770.1 cmϪ1 (calculated at 1789.3 cmϪ1), the
corresponding b2 mode at 1550.8 cmϪ1 (calculated at
1511.9 cmϪ1).
Photochemistry of 1,3,5-Trifluoro-2,4,6-triiodobenzene (4f)
Figure 4. IR spectra illustrating the photochemistry of 2,4,6-tri-
fluoro-1,3-diiodo-5-(trimethylsilyl)benzene (4e) in solid neon at
3 K; (a) difference spectrum; bands pointing upward increase,
bands pointing downward decrease on irradiation (10 min,
254 nm); increasing absorptions are assigned to monoradical 5e;
(b) difference spectrum; bands pointing upward increase, bands
pointing downward decrease on annealing the neon matrix (7.5 K,
10 min) after 16 h of irradiation with 254 nm; bands marked with
° are assigned to diradical 6e; (c) calculated [BLYP/6-311G(d,p),
unscaled] spectrum of 1,3-didehydro-2,4,6-trifluoro-5-(trifluorome-
thyl)benzene (6e)
The formation of mono- and diradicals from diiodoben-
zene derivatives suggested that it might be possible to gener-
ate a triradical from the corresponding triiodobenzene.
Since fluorine substitution apparently facilitates the isola-
tion of didehydrobenzenes originating from aromatic diiod-
ides, 1,3,5-trifluoro-2,4,6-triiodobenzene (4f) was chosen to
test this possibility. Thanks to its high symmetry, the IR
spectrum of matrix-isolated 4f exhibits only a few intense
absorptions. Irradiation at 254 nm for 10 min results in the
loss of intensity of these bands, and simultaneous formation
As anticipated, another compound is formed on con-
of the 2,4,6-trifluoro-3,5-diiodophenyl radical 5f, as con- tinued irradiation, while the intensity of all bands assigned
firmed by the calculated IR spectrum (Table 10).
to monoradical 5f stagnates after approximately 6 h and
Eur. J. Org. Chem. 2002, 3927Ϫ3935
3931

H. H. Wenk, W. Sander
FULL PAPER
ated absorptions decrease on annealing at 7.5 K, while the
bands of trifluorotriiodobenzene 4f increase concurrently.
Table 9. IR spectroscopic data of 1,3-didehydro-2,4,6-trifluoro-5-
(trimethylsilyl)benzene [(S)-6e]
[b]
[b,c]
Mode Sym. ν˜exp. [cmϪ1 [a]
]
Irel,exp.
ν˜calcd. [cmϪ1 [c]
]
Irel,calcd.
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
823.1
Ϫ
0.30
Ϫ
801.5
859.7
861.3
874.6
903.6
0.37
0.16
0.20
0.23
0.40
0.69
0.25
0.07
0.05
0.01
0.03
0.21
0.03
0.00
0.00
0.00
0.02
0.01
0.02
0.48
916.3
951.9
968.4
1100.7
1254.8
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
0.14
0.22
0.26
1.00
0.20
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
1048.7
1204.2
1262.8
1266.6
1277.9
1301.1
1323.7
1420.7
1422.0
1426.9
1430.6
1434.9
1436.6
1448.3
1511.9
Figure 5. IR spectra illustrating the photochemistry of 1,3,5-tri-
fluoro-2,4,6-triiodobenzene (4f) in solid neon at 3 K; (a) difference
spectrum; bands pointing upward increase, bands pointing down-
ward decrease on irradiation (10 min, 254 nm); increasing absorp-
tions are assigned to monoradical 5f; (b) difference spectrum;
bands pointing upward increase, bands pointing downward de-
crease on annealing the neon matrix (7.5 K, 10 min) after 16 h of
irradiation with 254 nm; bands marked with ° are assigned to di-
radical 6f; (c) calculated [BLYP/6-311G(d,p), unscaled] spectrum of
1,3-didehydro-5-iodo-2,4,6-trifluorobenzene 6f
1550.8
1765.3
1770.1
0.41
0.43
0.24
51
Ϫ
1789.3
1.00
[a] Neon, 3 K. [b] Relative intensities based on the strongest absorp-
tion. Calculated at the BLYP/6-311G(d,p) level of theory. Band
positions are unscaled. The assignment is tentative and is based on
band positions and intensities.
[c]
Table 10. IR spectroscopic data of the 2,4,6-trifluoro-3,5-diiodo-
phenyl radical 5f
Table 11. IR spectroscopic data of 1,3-didehydro-2,4,6-trifluoro-5-
iodobenzene [(S)-6f]
[b]
[b,c]
Mode Sym. ν˜exp. [cmϪ1 [a]
]
Irel,exp.
ν˜calcd. [cmϪ1 [c]
]
Irel,calcd.
[b]
[b,c]
Mode Sym. ν˜exp. [cmϪ1 [a]
]
Irel,exp.
ν˜calcd. [cmϪ1 [c]
]
Irel,calcd.
14
15
16
17
18
19
20
21
22
23
24
25
26
27
b2
b1
a2
b2
b1
a1
b2
a1
b2
a1
b2
a1
a1
b2
612.0
Ϫ
Ϫ
691.4
712.2
1007.2
1058.4
0.20
Ϫ
Ϫ
0.04
0.03
0.83
0.43
608.2
613.1
637.4
667.6
696.6
1009.0
1068.5
1069.5
1297.8
1386.5
1413.6
1418.8
1552.6
1641.2
0.18
0.02
0.00
0.10
0.10
0.61
0.30
0.20
0.17
0.00
1.00
0.81
0.82
0.45
814.3
823.8
968.2
1074.7
1274.5
Ϫ
0.28
0.24
0.72
0.21
0.23
Ϫ
16
a1
785.5
0.75
17
18
19
20
21
22
23
b2
a1
b2
b2
a1
a1
b2
927.1
1019.2
1225.3
1339.8
1352.4
1391.9
1538.4
0.59
0.09
0.24
0.09
0.18
0.63
0.73
Ϫ
Ϫ
1447.4
1550.6
1773.5
1782.8
0.87
1.00
0.50
0.27
1309.2
Ϫ
1400.7
1403.4
1526.4
1588.1
0.03
Ϫ
0.88
1.00
0.65
0.16
24
a1
1780.1
1.00
[a] Neon, 3 K. [b] Relative intensities based on the strongest absorp-
tion. Calculated at the BLYP/6-311G(d,p) level of theory. Band
[c]
positions are unscaled. The assignment is tentative and is based on
band positions and intensities.
[a] Neon, 3 K. [b] Relative intensities based on the strongest absorp-
tion. [c] Calculated at the B3LYP/6-311G(d,p) level of theory. Band
positions are unscaled. The assignment is tentative and is based on
band positions and intensities.
Under no conditions was a third photoproduct poten-
tially attributable to the desired 1,3,5-tridehydro-2,4,6-tri-
fluorobenzene detected.
slightly decreases afterwards. The second product shows the
broad, split absorption attributable to the characteristic a1-
symmetrical vibration at 1774 and 1783 cmϪ1, as well as the
b2-symmetrical combination at 1550.6 cmϪ1, and is there-
Comparison of the m-Benzyne Derivatives
Two IR absorptions are characteristic of all 2-fluoro de-
fore assigned as 1,3-didehydro-2,4,6-trifluoro-5-iodoben- rivatives of m-benzyne (2) so far investigated: an a1-sym-
zene (6f) (Figure 5, Table 11). All photochemically gener- metrical mode in the 1765 cmϪ1 to 1824 cmϪ1 range and a
3932
Eur. J. Org. Chem. 2002, 3927Ϫ3935

Generation of Fluorinated m-Benzyne Derivatives in Neon Matrices
FULL PAPER
b2-symmetrical ring stretching vibration between 1551 and Winkler et al. tested the performance of various ab initio
1605 cmϪ1 (Figure 6).
and DFT methods, finding that the BLYP functional repro-
duces the vibrational spectra of m-benzynes quite accur-
ately.[13] We thus used this level of theory to calculate the
C1ϪC3 distances.
The singlet-triplet splitting ∆ESϪT values of all fluorin-
ated m-benzynes so far investigated are larger than that of
the parent m-benzyne (2). In contrast, the ∆ESϪT values of
perfluorinated p-benzynes are smaller than that of the par-
ent p-benzyne (3), which is attributable to stabilization of
the symmetrical combination ϕS of the singly occupied
lone-pair orbitals by the σ*(CϪF) orbitals (Figure 7).[7]
Through-bond coupling places ϕS energetically above the
antisymmetrical combination ϕA,[12,14,15] which gives rise to
a lower HOMOϪLUMO gap and to a decreased ∆ESϪT
value (Figure 8).
Figure 6. Observed IR absorptions assigned to the m-didehy-
drobenzene derivatives 6aϪ6f; the characteristic a1- and b2-sym-
metrical vibrations (or the corresponding vibrations for molecules
of lower than C2v symmetry) are marked by dotted lines
The substituent at C5 is hardly involved in these vibra-
tions (very small isotopic shifts in 6b/6c) and so changes in
the band positions of these vibrations are mainly the result
of electronic effects of the substituent on the ring system.
Calculations also predict these vibrations to be present in
the parent m-benzyne (2), but the intensity is much lower
than in the fluorinated derivatives. As a result, the a1-sym-
metrical mode [calculated at 1681.7 cmϪ1
, BLYP/6-
311G(d,p)] cannot be observed in 2, while the b2-symmet-
rical vibration corresponds to a weak absorption at
1486 cmϪ1
[calculated
at
1471.1 cmϪ1
,
BLYP/6-
311G(d,p)].[8]
There is a correlation between the frequency of the a1-
symmetrical mode and the calculated C1ϪC3 distance. For
6a and 6d, with the shortest calculated C1ϪC3 distances
Figure 7. Symmetric (S) and antisymmetric (A) combination of the
singly occupied lone-pair orbitals at the radical centers for parent
m-benzyne (2) (left) and perfluorinated 6a (right)
˚
(1.895 and 1.871 A, respectively), the band is shifted to
higher frequencies than seen in other derivatives with larger
C1ϪC3 distances (Table 12). It should be noted, however,
As in the para derivative, fluorine substitution stabilizes
that the energy well for change in the C1ϪC3 distance is the symmetrical combinations ϕS in o- and m-benzyne. Be-
very flat, and calculations give qualitatively different results cause of the increased through-space interaction in these
depending on the applied method. In an exhaustive study, molecules, however, this combination lies below the anti-
Table 12. Selected calculated and experimental data for m-benzyne derivatives
[a]
[b]
Singlet-triplet splitting [kcal molϪ1], calculated at the BLYP/6-311G(d,p) level of theory.
Calculated [BLYP/6-311G(d,p)] distance
[A] of the radical centers for singlet diradicals. Observed position of the characteristic CϪC/CϪF stretching vibration [cmϪ1].
˚
[c]
Eur. J. Org. Chem. 2002, 3927Ϫ3935
3933

H. H. Wenk, W. Sander
FULL PAPER
1,3,5-Trifluoro-2,4,6-triiodobenzene (4f): A modification of the lit-
erature procedure for the synthesis of tetrafluoro-1,3-diiodoben-
zene[19] by Neenan et al. was applied for the preparation of 4f.
Periodic acid (3.0 g, 13.2 mmol) was suspended in concentrated
H2SO4 (20 mL) at 0 °C. Finely ground KI (6.56 g, 40.0 mmol) was
added in small portions over 5 min. The dark mixture was stirred
and cooled with an ice bath, while 1,3,5-trifluorobenzene (Aldrich,
97%, 1.17 g, 8.86 mmol) was added over 15 min by syringe. The ice
bath was removed and the solution was heated to 70 °C for 4 h.
After cooling to room temperature, the reaction mixture was
poured onto ice (200 g) and extracted with diethyl ether. The or-
ganic phase was washed with aqueous sodium thiosulfate and water
and dried (MgSO4), and the solvents were evaporated to dryness.
The diiodide was sublimed in vacuo to give 4f as white crystals
(89% yield, 4.01 g, 7.86 mmol). The analytical data complies with
ref.[20] 13C NMR (CDCl3, 100 MHz): δ ϭ 63.7 (dt), 162.3 (dt) ppm.
EI-MS: m/z (%) ϭ 510 (100) [Mϩ], 383 (15), 256 (30), 129 (40), 79
(30). IR (Ne, 3 K): ν˜ (%) ϭ 1573 (41), 1440 (2), 1432 (2), 1417
(100), 1345 (5), 1062 (35), 1057 (16), 716 (2), 664 (61) cmϪ1
.
Figure 8. Effect of fluorinesubstitution on the HOMO-LUMO gap
of benzynes; stabilization of the antisymmetric (A) combination of
the singly occupied lone-pair orbitals at the radical centers results
in an increased HOMOϪLUMO gap for ortho- and meta-benzynes,
while for para-benzynes the energy difference is reduced; for p-
benzynes the order of the A and S combination is reversed by
through-bond interaction[14,15]
2,4,6-Trifluoro-1,3-diiodobenzene (4b): nBuLi (1.6  solution in
hexane, 1.1 mL, 1.7 mmol) was added at Ϫ78 °C, under argon, to
a
solution of 1,3,5-trifluoro-2,4,6-triiodobenzene (4f, 850 mg,
1.67 mmol) in dry diethyl ether (10 mL). The mixture was stirred
for 30 min, quenched with H2O (5 mL), and allowed to warm to
room temperature. Diethyl ether (80 mL) was added, the organic
phase was washed with dilute HCl and water and dried (MgSO4),
and the solvents were evaporated to dryness. The crude product
was sublimed in vacuo to give 4b as white needles. The analytical
data are identical to those in ref.[20] EI-MS: m/z (%) ϭ 384 (50)
[Mϩ]. IR (Ne, 3 K): ν˜ (%) ϭ 1597 (15), 1593 (34), 1586 (100), 1454
(81), 1449 (29), 1424 (7), 1419 (80), 1411 (37), 1369 (28), 1169 (10),
1149 (47), 1048 (32), 1043 (91), 1041 (15), 840 (16), 671 (56), 604
symmetrical ϕA, and so the HOMOϪLUMO gaps and the
∆ESϪT values are increased. This is in agreement with a
recent theoretical investigation of substituent effects on
benzyne electronic structures reported by Cramer et al.,
who found that the singlet states are selectively stabilized
over the triplets by appropriate substituents.[1]
In summary, we have demonstrated that photolysis of
suitably substituted 1,3-diiodobenzenes 4 results in the
formation of m-benzynes 6 in a stepwise reaction via the
phenyl radicals 5. This allows the substituents in 6 to be
varied systematically and the influence of these variations
on the spectroscopic properties to be studied.
(47), 584 (7), 562 (8) cmϪ1
.
5-Deuterio-2,4,6-trifluoro-1,3-diiodobenzene (4c): The synthesis was
carried out as described for the non-deuterated derivative 4b, but
the phenyllithium derivative was quenched with D2O instead of
H2O. 13C NMR (CDCl3, 50 MHz): δ ϭ 64.8 (ttd), 100.2 (ttd),
161.8 (dtt), 163.1 (dtt) ppm. EI-MS: m/z (%) ϭ 385 (100) [Mϩ], 258
(20), 220 (5), 205 (15), 192 (10), 131 (60), 100 (10), 81 (20), 62 (10).
IR (Ne, 3 K): ν˜ (%) ϭ 1595 (6), 1587 (71), 1577 (44), 1455 (8), 1446
(5), 1437 (40), 1434 (9), 1431 (50), 1424 (15), 1419 (7), 1406 (100),
1386 (6), 1368 (32), 1077 (8), 1068 (67), 1050 (14), 1043 (6), 1032
Experimental Section
Matrix Isolation: Matrix isolation experiments were performed by
standard techniques[16] with a Sumitomo Heavy Industries RDK-
408D closed-cycle refrigerator. Matrices were produced by co-de-
position of a large excess of neon (MesserϪGriesheim, 99.9999%)
and the trapped species on a cold CsI window (3 K). Infrared spec-
tra were recorded with a Bruker Equinox 55 FTIR spectrometer
(81), 883 (6), 749 (10), 671 (64), 562 (26), 552 (10), 550 (7) cmϪ1
.
2,4,6-Trifluoro-1,3-diiodo-5-(trifluoromethyl)benzene (4d): Periodic
acid (3.0 g, 13.2 mmol) was suspended at 0 °C in concentrated
H2SO4 (20 mL). Finely ground KI (6.56 g, 40.0 mmol) was added
in small portions over 5 min. The dark mixture was stirred and
cooled with an ice bath, while 2,4,6-trifluoro-1-(trifluoromethyl)-
benzene (Acros, 98%, 1.66 g, 8.3 mmol) was added over 15 min by
syringe. The ice bath was removed and the solution was heated to
70 °C for 4 h. After cooling to room temperature, the reaction mix-
ture was poured onto ice (200 g) and extracted with ether. The or-
ganic phase was washed with aqueous sodium thiosulfate and water
and dried (MgSO4), and the solvents were evaporated to dryness.
The diiodide was sublimed twice in vacuo to give 4d as white crys-
with
a
standard resolution of 0.5 cmϪ1 in the range of
400Ϫ4000 cmϪ1. Irradiation was carried out with a Gräntzel low-
pressure mercury lamp.
Theoretical Methods: Calculations were performed with the
GAUSSIAN98 suite of programs.[17] Geometries and vibrational
spectra of diradicals 6bϪ6f were calculated by use of the RBLYP
functional, while the B3LYP functional was applied for phenyl rad-
icals 5bϪ25 for consistency with previous publications. The 6- tals (3.0 g, 6.64 mmol, 80% yield) melting at 41 °C. 13C NMR
311G(d,p) basis set was used for all molecules.[18] Tight cutoffs were
used for the geometry optimizations. Vibrational frequency calcula-
tions were carried out in all cases in order to obtain the IR spectra
and to determine the nature of the stationary points. For discus-
sions of the performance of different theoretical methods for dide-
hydrobenzenes, cf. refs.[10Ϫ13]
(100 MHz, CDCl3): δ ϭ 67.0 (dt), 105.3 (t), 120.8 (q), 161.3 (dt);
164.0 (dt) EI-MS: m/z (%) ϭ 452 (100) [Mϩ], 433 (10), 325 (15),
226 (5), 198 (45), 179 (15), 148 (20), 129 (10), 110 (5), 79 (15), 69
(15). IR (Ne, 3 K): ν˜ (%) ϭ 1606 (35), 1587 (19), 1438 (64), 1381
(4), 1296 (100), 1286 (12), 1268 (7), 1191 (28), 1166 (20), 1094 (34),
1085 (15), 1073 (7), 725 (3), 671 (20), 654 (23) cmϪ1
.
3934
Eur. J. Org. Chem. 2002, 3927Ϫ3935

Generation of Fluorinated m-Benzyne Derivatives in Neon Matrices
FULL PAPER
G. Radziszewski, Angew. Chem. 1998, 110, 1001Ϫ1005; Angew.
Chem. Int. Ed. 1998, 37, 955Ϫ958.
E. Kraka, J. Anglada, A. Hjerpe, M. Filatov, D. Cremer, Chem.
Phys. Lett. 2001, 348, 115Ϫ125.
V. Polo, J. Gräfenstein, E. Kraka, D. Cremer, Chem. Phys. Lett.
2002, 352, 469Ϫ478.
T. D. Crawford, E. Kraka, J. F. Stanton, D. Cremer, J. Chem.
Phys. 2001, 114, 10638Ϫ10650.
2,4,6-Trifluoro-1,3-diiodo-5-(trimethylsilyl)benzene (4e): The syn-
thesis was carried out as described for 2,4,6-trifluoro-1,3-diiodo-
benzene (4b), but the phenyllithium derivative was treated with an
excess of ClSi(CH3)3 instead of H2O. Sublimation in vacuo gave 4e
as a white solid melting at 73 °C. 1H NMR (200 MHz, CDCl3):
δ ϭ 0.36 (t). 13C NMR (50 MHz, CDCl3): δ ϭ 65.1 (ddd), 110.4
(dt), 162.3 (td), 166.4 (ddd) ppm. EI-MS: m/z (%) ϭ 456 (28), 441
(13), 364 (6), 248 (32), 237 (13), 281 (10), 187 (38), 167 (18), 127
(22), 110 (18), 87 (26), 81 (36), 104 (100), 73 (73). IR (Ne, 3 K): ν˜
(%) ϭ 2971 (3), 2914 (2), 1582 (100), 1573 (9), 1556 (19), 1424 (5),
1417 (31), 1393 (52), 1366 (13), 1339 (6), 1270 (13), 1260 (23), 1112
(90), 1061 (39), 1058 (8), 1040 (26), 859 (13), 852 (35), 772 (11),
[10]
[11]
[12]
[13]
[14]
M. Winkler, W. Sander, J. Phys. Chem.
10422Ϫ10433.
A 2001, 105,
R. Hoffmann, A. Imamura, W. J. Hehre, J. Am. Chem. Soc.
1968, 90, 1499Ϫ1509.
K. D. Jordan, Theor. Chem. Acc. 2000, 103, 286Ϫ288.
I. Dunkin, Matrix Isolation Techniques: A Practical Approach,
Oxford University Press, Oxford, UK, 1998.
[15]
[16]
733 (5), 663 (56), 621 (13) cmϪ1
.
[17]
M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M.
A. Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A. Montgom-
ery, Jr., R. E. Stratmann, J. C. Burant, S. Dapprich, J. M. Mil-
lam, A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J.
Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Po-
melli, C. Adamo, S. Clifford, J. Ochterski, G. A. Petersson, P.
Y. Ayala, Q. Cui, K. Morokuma, D. K. Malick, A. D. Rabuck,
K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Ortiz, B.
B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.
Gomperts, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham,
C. Y. Peng, A. Nanayakkara, C. Gonzalez, M. Challacombe,
P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, J. L. Andres,
M. Head-Gordon, E. S. Replogle, J. A. Pople, Gaussian 98,
Revision A.3, Pittsburgh PA, 1998.
Supporting Information Available: Matrix infrared spectra (neon,
3 K) of diiodobenzenes 4bϪ4f and difference spectra showing the
formation of phenyl radicals 5bϪ5f on short-term irradiation at
254 nm. Calculated geometries (Cartesian coordinates in A) and
vibrational spectra of phenyl radicals 5bϪ5f and didehydroben-
zenes 6bϪ6f (see also footnote on the first page of this article).
˚
Acknowledgments
This work was financially supported by the Deutsche Forschungs-
gemeinschaft and the Fonds der Chemischen Industrie.
[18]
This basis set was obtained from the Extensible Computational
Chemistry Environment Basis Set Database, as developed and
distributed by the Molecular Science Computing Facility, En-
vironmental and Molecular Sciences Laboratory, which is part
of the Pacific Northwest Laboratory, P. O. Box 999, Richland,
Washington 99352, USA, and funded by the U.S. Department
of Energy. The Pacific Northwest Laboratory is a multi-pro-
gram laboratory operated by the Battelle Memorial Institute
for the U.S. Department of Energy under contract DE-AC06-
76RLO 1830. Contact David Feller or Karen Schuchardt for
further information.
[1]
W. T. G. Johnson, C. J. Cramer, J. Phys. Org. Chem. 2001, 14,
597Ϫ603.
[2]
H. H. Wenk, W. Sander, Chem. Eur. J. 2001, 7, 1837Ϫ1844.
[3]
W. Sander, M. Exner, J. Chem. Soc., Perkin Trans. 2 1999, 2285.
[4]
K. K. Thoen, H. I. Kenttamaa, J. Am. Chem. Soc. 1997, 119,
3832Ϫ3833.
[5]
K. K. Thoen, H. I. Kenttamaa, J. Am. Chem. Soc. 1999, 121,
800Ϫ805.
E. D. Nelson, A. Artau, J. M. Price, H. I. Kenttamaa, J. Am.
Chem. Soc. 2000, 122, 8781Ϫ8782.
H. H. Wenk, A. Balster, W. Sander, D. A. Hrovat, W. T.
[6]
[7]
[19]
[20]
T. X. Neenan, G. M. Whitesides, J. Org. Chem. 1988, 53,
2489Ϫ2496.
G. B. Deacon, R. N. M. Smith, Aust. J. Chem. 1982, 35,
1587Ϫ1597.
Borden, Angew. Chem. 2001, 113, 2356Ϫ2359; Angew. Chem.
Int. Ed. 2001, 40, 2295Ϫ2298.
[8]
R. Marquardt, W. Sander, E. Kraka, Angew. Chem. 1996, 108,
825Ϫ827; Angew. Chem. Int. Ed. Engl. 1996, 35, 746Ϫ748.
Received May 16, 2002
[9]
R. Marquardt, A. Balster, W. Sander, E. Kraka, D. Cremer, J.
[O02256]
Eur. J. Org. Chem. 2002, 3927Ϫ3935
3935