Spin selectivity in the oxygenation of singlet phenylhalocarbenes with oxygen

Article abstract of DOI:10.1016/S0040-4039(02)02538-8

Singlet phenylhalocarbenes are shown to react with triplet oxygen and the apparent spin-forbidden oxygenation rates are strongly dependent on substituents, i.e. in the decreasing order of Br>Cl?F for halogen and of p-NO₂>H?p-MeO for p-substituents. These results suggest the oxygenation of triplet halocarbene equilibrated with ground-state singlet, resulting in the first estimation of energy difference between the singlet and triplet states.

Full text of DOI:10.1016/S0040-4039(02)02538-8

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 865–868
Spin selectivity in the oxygenation of singlet phenylhalocarbenes
with oxygen
Taiki Makihara, Takayuki Nojima, Katsuya Ishiguro† and Yasuhiko Sawaki*
Department of Applied Chemistry, Graduate School of Engineering, Nagoya University, Chikusa-ku, Nagoya 464-8603, Japan
Received 16 October 2002; revised 7 November 2002; accepted 8 November 2002
Abstract—Singlet phenylhalocarbenes are shown to react with triplet oxygen and the apparent spin-forbidden oxygenation rates
are strongly dependent on substituents, i.e. in the decreasing order of Br>ClꢀF for halogen and of p-NO2>Hꢀp-MeO for
p-substituents. These results suggest the oxygenation of triplet halocarbene equilibrated with ground-state singlet, resulting in the
first estimation of energy difference between the singlet and triplet states. © 2003 Elsevier Science Ltd. All rights reserved.
The reaction of oxygen with various substrates are
known to be relatively slow in spite of their high
exothermicity. The apparent discrepancy is sometimes
explained by the spin-forbidden step in the reaction of
Phenylhalocarbenes (ArCX, 1), conveniently produced
by photolyzing the corresponding diazirines (2),8 are
known to be singlet9 and reactive for various substrates
as studied by laser-flash spectroscopy.10 Since the decay
of phenylchlorocarbene (1a) was unchanged by the
presence of oxygen, the parent carbene has been
assumed not to be reactive toward oxygen.6b–d We
could reveal the moderate reactivity of 1a with oxygen
by using Freon 113 as a non-reactive solvent,11 i.e. the
major product changed from dimer olefin under argon
to benzoyl chloride under oxygen.12 A laser-flash spec-
troscopy (YAG laser, 355 nm, ꢁ5 mJ)13 indicated that
the decay of 1a at 300 nm14a is accelerated by oxygen,
3
triplet oxygen, O2, and singlet substrates. In fact, it is
well known that excited singlet oxygen, 1O2, reacts
facilely with various olefins and sulfides.1 In relation to
the spin selectivity on oxygenations, we became inter-
ested in the reaction of oxygen with carbenes of various
spin-states.
Arylcarbenes of triplet ground-state are known to react
3
smoothly with O2 to yield carbonyl oxides2 as estab-
lished by laser flash spectroscopy.3 On the contrary,
only a little is known on the oxygenation of singlet
carbenes. Imidazolylidenes are a quite stable singlet
carbene and could be treated under air.4 Recently we
have reported that the singlet carbene reacts fast with
1O2 to form a characteristic carbonyl oxide.5 Likewise,
phenylhalocarbenes, a more reactive singlet intermedi-
ate, are reported not to react with oxygen6 except for
the case of the p-nitrophenyl one.7 It is an open ques-
tion whether the apparent spin-forbidden process for
the reactive singlet carbenes does occur or not. Herein,
we report an interesting aspect on the spin selectivity in
resulting in the oxygenation rate constant: kO =1.2×106
2
M−1 s−1 (in Freon 113 at room temperature). A similar
result was obtained for the case of phenylbromocarbene
(1b); the decay of 1b at 347 nm14b (lifetime t=5.8 ms)
was decreased by oxygen (t=1.1 ms) affording kO =
2
1.7×106 M−1 s−1. Now it is apparent that the reported
no reaction of these carbenes with oxygen is simply due
to the fast reaction toward solvent hydrocarbons.
Liu et al. reported on the exceptional reactivity of
p-nitrophenylchlorocarbene (1c) with oxygen; its
absorption at 320 nm in isooctane was quenched by
oxygen producing carbonyl oxide absorption at 400
nm.7 We could reproduce similar results; the decay rate
(1.5×106 s−1) of 1c in Freon 113 under oxygen was
practically identical with the growth of carbonyl oxide
(1.6×106 s−1). The rate constant for the reaction with
3
the oxygenation of phenylhalocarbenes with O2.
Keywords: carbenes and carbenoids; oxygenation; halogens and com-
pounds; mechanisms.
* Corresponding author. Present address: Department of Applied
Chemistry, Aichi Institute of Technology, Toyota 470-0392, Japan.
† Present address: Faculty of Science, Yamaguchi University,
Yamaguchi 753-8512, Japan.
oxygen, kO =2.0×107 M−1 s−1, was ca. 8-fold faster
2
than that of 1a.15 Products from 2c (irradiated at >400
nm) were azine 5c under argon and aroyl chloride 6c,
suggesting an efficient oxygenation of carbene 1c. Quite
similar data were obtained for bromocarbene (1d); the
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)02538-8

866
T. Makihara et al. / Tetrahedron Letters 44 (2003) 865–868
Products and oxygenation rates for six phenylhalocar-
benes are summarized in Table 1. It is clear that the
oxygenation efficiencies are profoundly dependent on
aryl substituents and a-halogens. The oxygenation was
accelerated by the order of p-NO2>Hꢀp-MeO on
phenyl and of Br>ClꢀF for halogens. These experi-
mental results are discussed in relation to two possible
mechanisms: (a) spin-forbidden oxygenation of halocar-
bene 1 and (b) spin-allowed oxygenation via 31 as
shown in Scheme 2. Mechanism (a) is the spin-forbid-
1
3
den oxygenation between singlet carbene 1 and O2.
Oxidations are classified as the charge or electron trans-
fer from substrates to oxidants and so it is generally
accepted that electron-donating substituents on sub-
strates facilitate the reaction.1a,17 Such a trend was not
observed for the present case and the reactivity order of
p-NO2>Hꢀp-MeO is not explicable on the basis of
charge or electron transfer from halocarbenes to
oxygen.
Scheme 1.
absorption of 1d was observed at 335 nm and the
oxygenation rate, kO =5.3×107 M−1 s−1, was slightly
faster than that of 1c2.
The product study for phenylhalocarbenes is summa-
rized in Scheme 1. Photolysis of diazirines (2) produces
halocarbenes (1), which are converted to dimer olefins
(4) or azines (5). Under oxygen carbenes (1) are trapped
to form carbonyl oxides (3) yielding finally benzoyl
halides (6). Facile conversions of carbonyl oxides to the
corresponding carbonyl compounds are well established
as a reaction between two molecules of the oxide.2,3
Mechanism (b) is the spin-allowed oxygenation of
3
3
triplet 1 with O2 by assuming an equilibrium between
3
11 and 1. Here, the equilibrium should be established
because the oxygenation rate was proportional to oxy-
gen concentration.18 Assumption of the allowed reac-
tion of 31 with 3O2 (Mechanism b) explains our
experimental results reasonably as discussed in the fol-
Intermediates and products for the photolysis of
diazirines 2 under argon and oxygen are summarized in
Table 1. For the case of 2a–d, the corresponding carbe-
nes 1a–d were observed with umax at 300–350 nm and
their decay was accelerated by oxygen, the major path-
way being the oxygenation to yield 3 and finally 6. On
the other hand, the oxygenation of carbene 1e (X=F)
and 1f (Ar=p-MeOC6H4) was inefficient. The major
product from fluorocarbene 1e was dimer olefin 4e even
under oxygen, oxygenated product being only <1%.
Here the oxygenation rate may be estimated to be
<1×105 M−1 s−1 in comparison to the efficient reaction
for carbenes 1c and 1d under the same conditions. For
the case of p-MeO isomer (1f) its absorption maximum
was observed at 360 nm,16 but the decay was not
altered by oxygen and 6f was not detected. From the
decay rate of 1f (i.e. 5×105 s−1) and detection limit
3
lowing. The spin-allowed reaction of triplet Ph2C: and
3O2 is known to be as fast as kO =(1–5)×109 M−1 s−1;19
2
these values are acceptable since the formation of sin-
glet products from the two triplet materials is 1/9,20 i.e.
1/9 of the diffusion rate constant. Thus, the equilibrium
1
3
constant K for 1 and 1 could be calculated from the
(0.5%), the kO value for 1f may be estimated to be
2
<1×105 M−1 s−1.
Scheme 2.
Table 1. Products and oxygenation rates in the photolysis of diazirines (2) in Freon 113
ArXCN2 (2)
ArXC: (1)a
(umax, nm)
Productsb
O2
Oxygen. of 1c
X
Ar
Argon
(kO , M−1 s−1
)
2
a
b
c
d
e
f
Cl
Br
Cl
Br
F
C6H5
C6H5
p-NO2C6H4
p-NO2C6H4
p-NO2C6H4
p-MeOC6H4
300
347
315
4
4
5
5
4
5
6
6
6
6
1.2×106
1.7×106
2.0×107
5.3×107
B1×105,e
B1×105,e
335
d
4+6 (B1%)
5
Br
360
a Transient spectra of carbenes (1) 250 ns after laser flash at 355 nm at room temperature.
b Products from the photolysis of 2 under argon or oxygen. See Ref. 12 for details.
c Second-order rate constants, kO (M−1 s−1), obtained from the dependence of decay of 1 on [O2].
2
d Could not be determined due to the strong absorption of unknown by-product.
e Estimated from product selectivity for 6e and 6f. See text for details.

T. Makihara et al. / Tetrahedron Letters 44 (2003) 865–868
867
1
3
selectivity.25 The present oxygenation of halocarbenes is
shown to be one of typical examples governed by the
spin-selection rule.26
Table 2. Estimated K and DGST values for 1 and 1
Carbene
Ka
DGST
DG
b
calcd,c
ST
1a
1b
1c
1d
1e
1f
1.2×10−3
1.7×10−3
2.0×10−2
5.3×10−2
B1×10−4
B1×10−4
−17
−16
−10
−7.5
B−23
B−23
−40; −15d
−28
−8.8
In conclusion, the apparent spin-forbidden oxygenation
of singlet halocarbenes (1) with 3O2 is shown to proceed
3
3
via the spin-allowed reaction between O2 and 1 equili-
−5.4
1
−40; −27d
−38d
brated with 1. Only by this mechanism are explicable
the substituent effects of p-NO2>Hꢀp-MeO and Br>
ClꢀF. This is the first successful study in determining
singlet–triplet energy gap of ‘singlet’ phenylhalo-
carbenes.
a Equilibrium constants (K) between 11 and 31 by assuming the kT
value (Scheme 2) as 1×109 M−1 s−1
.
b Singlet–triplet energy gap (kJ mol−1) from −RT ln K.
c Energy gap (kJ mol−1) by DFT of BLYP/6-311G*.
d DFT calculations of pBP/DN** (Spartan).
References
observed value by assuming the oxygenation rate of
kT=1×109 M−1 s−1 via pathway (b) of Scheme 2. The
free energy differences DGST between the S and T states
may be obtained from DGST=−RT ln K and the result-
ing values are listed in Table 2. It is clear that halocar-
benes suffering with more efficient oxygenation possess
a smaller negative DGST value, i.e. the smaller energy
1. (a) Matsuura, T. Oxygen Oxygenation Reaction;
Maruzen: Tokyo, 1977; (b) Singlet Oxygen; Wasserman,
H. H.; Murray, R. W., Eds.; Academic Press: New York,
1979; (c) Singlet O2; Frimer, A. A., Ed.; CRC Press: Boca
Raton, FL, 1985; Vols. 1–4.
2. (a) Kirmse, W.; Horner, H.; Hoffmann, H. Justus Liebigs
Ann. Chem. 1958, 614, 19; (b) Murray, R. W.; Suzui, A.
J. Am. Chem. Soc. 1971, 93, 4963; (c) Sawaki, Y.; Kato,
H.; Ogata, Y. J. Am. Chem. Soc. 1981, 103, 3832; (d)
Sander, W. Angew. Chem., Int. Ed. Engl. 1990, 29, 344;
(e) Bunnelle, W. H. Chem. Rev. 1991, 91, 335; (f) McCul-
lough, K. J.; Nojima, M. In Organic Peroxides; Ando,
W., Ed.; Wiley: New York, 1992; p. 661; (g) Ishiguro, K.;
Nojima, T.; Sawaki, Y. J. Phys. Org. Chem. 1997, 10,
787; (h) Ishiguro, K.; Sawaki, Y. Bull. Chem. Soc. Jpn.
2000, 73, 535.
3. (a) Sugawara, T.; Iwamura, H.; Hayashi, H.; Sekiguchi,
A.; Ando, W.; Liu, M. T. H. Chem. Lett. 1983, 1261; (b)
Casal, M. L.; Sugamori, S. E.; Scaiano, J. C. J. Am.
Chem. Soc. 1984, 106, 7623; (c) Casal, H. L.; Tanner, M.;
Werstiuk, N. H.; Scaiano, J. C. J. Am. Chem. Soc. 1985,
107, 4616; (d) Barcus, R. L.; Hadel, L. M.; Johnston, L.
J.; Platz, M. L.; Savino, T. G.; Scaiano, J. C. J. Am.
Chem. Soc. 1986, 108, 3928; (e) Fujiwara, Y.; Tanimoto,
Y.; Itoh, M.; Hirai, K.; Tomioka, H. J. Am. Chem. Soc.
1987, 109, 1942.
1
3
differences between 1 and 1.
calcdd
ST
In the last column of Table 2 are listed DG
values
calculated theoretically by the DFT method; mostly
with BLYP/6-311G*21 and partially with pBP/DN**22
calcd
ST
(Spartan). It is interesting to note that these DG
values are roughly in line with those of experimental
DGST ones. In other words, the theoretical prediction
reproduces well the observed substituent effect of p-
NO2C6H4>C6H5ꢀp-MeOC6H5 and of Br>ClꢀF for
the present oxygenation via the triplet sate. For the
simple cases of halocarbenes their singlet state has been
calculated to be stabilized in the order of FꢀCl>Br for
halogens.23 Similarly, theoretical calculations predicted
that for the case of aryl carbenes their triplet state is
stabilized by a p-nitro group and destabilized by a
p-methoxy substituent.24
Temperature effect on the oxygenation has been studied
for the case of carbene 1c. The kO values in Freon 113
2
were 1.01, 1.20. 1.23, 1.43, and 1.83×107 M−1 s−1 at
4. (a) Arduengo, A. J., III; Harlow, R. L.; Kline, M. J. Am.
Chem. Soc. 1991, 113, 361; (b) Arduengo, A. J., III Acc.
Chem. Res. 1999, 32, 913.
263.0, 273.0, 283.0, 293.0, and 303.0 K, respectively.
From the plot of ln kO versus 1/T were calculated
2
Ea=9.0 kJ mol−1 and A=6.13×108. The activation
energy of 9.0 kJ mol−1 is close to the DGST value of 10.0
kJ mol−1, supporting Mechanism (b) involving the equi-
5. Ishiguro, K.; Hirabayashi, K.; Nojima, T.; Sawaki, Y.
Chem. Lett. 2002, 796.
6. (a) Padwa, A.; Eastman, D. J. Org. Chem. 1964, 34, 2728;
(b) Turro, N. J.; Butcher, J. A., Jr.; Moss, R. A.; Guo,
W.; Munjal, R.; Fedorynski, M. J. Am. Chem. Soc. 1980,
102, 7576; (c) Griller, D.; Liu, M. T. H.; Scaiano, J. C. J.
Am. Chem. Soc. 1982, 104, 5594; (d) Moss, R. A.;
Lawrynowicz, W.; Turro, N. J.; Gould, I. R.; Cha, Y. J.
Am. Chem. Soc. 1986, 108, 7028.
1
3
librium between 1 and 1.
All of the results are explained reasonably by assuming
the spin-allowed oxygenation of triplet halocarbenes
with 3O2 (Mechanism (b)). The observed substituent
effects on phenyl and halogens are only understood by
the stabilizing or destabilizing effect of these groups.
For example, the p-NO2C6H4 group interacts strongly
with the p-orbital of carbene (cf. Scheme 2), resulting in
the stabilization, compared to phenyl one, of the T
state. In contrast, the fluoro group interacts with the
s-orbital, stabilizing the S state significantly. In the field
of carbene reactions, singlet–triplet energy gaps are an
important factor for controlling their reactivity and
7. Liu, M. T. H.; Bonneau, R.; Jefford, C. W. J. Chem.
Soc., Chem. Commun. 1990, 1482.
8. Graham, W. H. J. Am. Chem. Soc. 1965, 87, 4396.
9. Moss, R. A. In Carbenes; Jones, M., Jr.; Moss, R. A.,
Eds.; Wiley: New York, 1973; p. 153.
10. Moss, R. A.; Turro, N. J. In Kinetics and Spectroscopy of
Carbenes and Biradicals; Platz, M. S., Ed.; Plenum: New
York, 1990; p. 213.

868
T. Makihara et al. / Tetrahedron Letters 44 (2003) 865–868
the oxygenation of 1 should not be, contrary to the
experimental result, proportional to O2 concentrations.
19. (a) Closs, G. L.; Rabinow, B. E. J. Am. Chem. Soc. 1976,
98, 8190; (b) Werstiuk, N. H.; Casal, H. L.; Scaiano, J. C.
Can. J. Chem. 1984, 62, 2391.
20. (a) Gijzeman, O. L. J.; Kaufman, F.; Porter, G. J. Chem.
Soc., Faraday Trans. 2 1973, 69, 708; (b) Koizumi, M.;
Kato, S.; Mataga, N.; Matuura, T.; Usui, Y. Photosensi-
tized Reactions; Kagakudojin: Kyoto, 1978; p. 210.
21. (a) Becke, A. D. Phys. Rev. A 1988, 38, 3098; (b) Lee, C.;
Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785.
22. Hehre, W. J.; Shusterman, A. J.; Huang, W. W. A
Laboratory Book of Computational Organic Chemistry;
Wavefunction Inc: Irvine, CA, 1995.
23. (a) Bagas, B. J. J. Am. Chem. Soc. 1977, 99, 7106; (b)
Vargas, R.; Galvan, M.; Velga, A. J. Phys. Chem. A
1998, 102, 3134.
24. (a) Li, Y.-Z.; Schuster, G. B. J. Org. Chem. 1988, 53,
1273; (b) Geise, C. M.; Hadad, C. M. J. Org. Chem.
2000, 65, 8348.
25. (a) Platz, M. S.; Malony, V. M. In Kinetics and Spec-
troscopy of Carbenes and Biradicals; Platz, M., Ed.;
Plenum: New York, 1990; p. 320; (b) Cremer, C. J.;
Truhlar, D. G.; Falvey, D. E. J. Am. Chem. Soc. 1997,
119, 12338; (c) Tomioka, H. Bull. Chem. Soc. Jpn. 1998,
71, 1501.
26. (a) Chlorocarbene 1a was reported to react slowly with
oxygen under low temperature matrix conditions.26b The
reaction was explained by a special interaction between
1a and oxygen; (b) Ganzer, G. A.; Sheridan, R. S.; Liu,
M. T. H. J. Am. Chem. Soc. 1986, 108, 1517.
11. Porter, N. A.; Roe, A. N.; McPhail, A. T. J. Am. Chem.
Soc. 1980, 102, 7576.
12. Irradiations were carried out at >400 nm with a medium
pressure 300 W Hg lamp. Products were determined by
1H NMR and GC–MS analyses in comparison to authen-
tic samples. Aroyl halides were analyzed after conversion
to the corresponding methyl esters with sodium methox-
ide.
13. Laser kinetic spectroscopy was carried out as described
previously. See: (a) Nojima, T.; Ishiguro, K.; Sawaki, Y.
J. Org. Chem. 1997, 67, 6911; (b) Yokoi, H.; Nakano, T.;
Fijita, W.; Ishiguro, K.; Sawaki, Y. J. Am. Chem. Soc.
1998, 120, 12453.
14. (a) Reported umax values for 1a and 1b are 310 and 330
nm in 3-methylpentane;14b (b) Gould, I. R.; Turro, N. J.;
Butcher, J., Jr.; Doubleday, C., Jr.; Hacker, N. P.; Lehr,
G. F.; Moss, R. A.; Cox, D. P.; Guo, W.; Munjal, R. C.;
Perez, L. A.; Fedorynski, M. Tetrahedron 1985, 1587.
15. Reported rate constant kO =2.24×107 M−1 d−1 for 1c in
2
isooctane is close to the present one, but the products
(e.g. ArCHO, ArCOCl and ArCHCl2) are different from
the present results in Freon 113. The contrast is due to
the no reaction with the solvent Freon and to the irradia-
tion with the filtered light at >400 nm.
16. Absorption maximum for p-MeOC6H4CCl is 340 nm in
3-methylpentane,14b and the red shift is noted by bromo
substitution.
17. Organic Peroxides; Ando, W., Ed.; Wiley: Chichester,
1992.
18. If the equilibrium between 11 and 31 were not established,