Laser flash photolysis of 3-noradamantyl(phenyl)diazomethane: Generation, detection and kinetics of 2-phenyladamantene

Article abstract of DOI:10.1002/(SICI)1099-1395(199902)12:2<165::AID-POC128>3.0.CO;2-N

Laser flash photolysis of 3-noradamantyl(phenyl)diazomethane in degassed benzene at room temperature generates 2-phenyladamantene, which decays with second-order kinetics (2k/εl = 1.5 × 10² s^-1) to give a dimer and is shown to react with oxygen and tri(n-butyl)tin hydride much faster than with methanol, thus revealing profound radical character of the twisted double bond. Copyright

Full text of DOI:10.1002/(SICI)1099-1395(199902)12:2<165::AID-POC128>3.0.CO;2-N

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 12, 165–169 (1999)
Short Communication
Laser ¯ash photolysis of 3-noradamantyl(phenyl)diazo-
methane: generation, detection and kinetics of
2-phenyladamantene
Katsuyuki Hirai,1 Hideo Tomioka,1* Takao Okazaki,2 Kazuhiko Tokunaga,2 Toshikazu Kitagawa2 and
Ken'ichi Takeuchi2*
1Chemistry Department for Materials, Faculty of Engineering, Mie University, Tsu, Mie 514 Japan
2Department of Energy and Hydrocarbon Chemistry, Graduate School of Engineering, Kyoto University, Sakyo-ku, Kyoto 606-01, Japan
Received 29 June 1998; revised 14 August 1998; accepted 14 August 1998
ABSTRACT: Laser flash photolysis of 3-noradamantyl(phenyl)diazomethane in degassed benzene at room
temperature generates 2-phenyladamantene, which decays with second-order kinetics (2k/el = 1.5 Â 102 s 1) to give a
dimer and is shown to react with oxygen and tri(n-butyl)tin hydride much faster than with methanol, thus revealing
profound radical character of the twisted double bond. Copyright 1999 John Wiley & Sons, Ltd.
KEYWORDS: bridgehead alkenes; alkene strain; dimerization; polar vs radical characters
Bridgehead alkenes, often referred to as Bredt’s rule
violators, have had a long and glorious history1 since the
work of Bredt et al.2 in the early part of this century and
to date many bridgehead alkenes of varying stability have
been studied. Naturally, most studies have been devoted
to clarifying the relationship between chemical behaviors
and twisting distortion of p bonds. Since bridgehead
double bonds are readily accommodated in larger ring
systems, bridgehead alkenes are classified into three
categories: isolable, observable and unstable. Adaman-
tene, solidly categorized as an unstable species, based on
a heat of formation [32.36 kcal mol 1 (1 kcal = 4.184 kJ)]
and olefinic strain (39.5 kcal mol 1)3 and also the twist
angle of the p portion of the double bond (64°),4 has been
a subject of considerable interest, especially owing to its
contradictory history.5,6 The complete picture of the
nature of adamantene has begun to be unfolded recently
by painstaking chemical analysis. More direct informa-
tion obtained using time-resolved spectroscopic techni-
ques available for the study of transient species chemistry
is, however, still scant in this field. We therefore studied
the chemistry of 3-noradamantyl(phenyl)diazomethane
(1), a potential precursor of phenyladamantene (2), by
using laser flash photolysis (LFP) techniques.
temperature with a 308 nm pulse from a XeCl excimer
laser (10 ns, 70–90 mJ) produced a transient species
showing a fairly strong broad absorption at 434 nm,
which appeared coincident with the pulse. A shoulder
appeared at 323 nm in the transient optical density
spectrum 20 ms after the laser flash. At 100 ms after the
laser flash, the absorption maximum had shifted to
323 nm. Figure 1 shows the transient spectrum obtained
100 ms after the excitation where decay of the initial
species (A) was not complete; the inset in Fig. 1 shows
the decay of the 434 nm species and the formation of the
323 nm species (B), indicating that the decay is
kinetically correlated with the growth of the new species.
The decay of A was found to be second order (2 k/
el = 1.5 Â 102 s 1) and the lifetime is estimated in the
form of the half-life, t1/2, to be roughly 5.6 ms. On the
other hand, the secondary formed species (B) is too long-
lived to be monitored by our system.
What are these species? It is well documented that the
photolysis of diazomethanes generates carbenes as initial
intermediates which then decay to form final products8
and that some arylcarbenes are often observed in solution
at room temperature by LFP.9 However, judging from the
absorption and lifetime features, it is highly unlikely that
the initially formed transient species (A) can be assigned
to phenyl(noradamantyl)carbene (3). For instance, the
optical absorption spectra of triplet diarylcarbenes have
been assigned and typically they consist of an intense UV
band around 300–340 nm along with a very weak visible
transition around. 400–500 nm.10 The lifetime of triplet
Laser flash photolysis7 of 3-noradamantyl(phenyl)dia-
zomethane (1) (prepared by oxidation of 3-noradamantyl
phenyl ketone) in a degassed benzene solution at room
*Correspondence to: H. Tomioka, Chemistry Department for
Materials, Faculty of Engineering, Mie University, Tsu, Mie 514,
Japan.
Copyright 1999 John Wiley & Sons, Ltd.
CCC 0894–3230/99/020165–05 $17.50

166
K. HIRAI ET AL.
Figure 1. Absorption spectrum of the transient products formed during the irradiation of 3-noradamantyl(phenyl)diazomethane
(1) in degassed benzene recorded 100 ms after excitation. Inset shows oscillogram traces monitored at 323 and 434 nm
diphenylcarbene in degassed benzene is 4 ms (H.
Tomioka and K. Hirai, unpublished work). The absorp-
tion bands and lifetime for monoarylcarbenes should be
shifted to shorter wavelength and shorter time regions
compared with the diarylcarbenes.
stant, kobs, of the transient band is expressed as
kobs ˆ k0 ‡ kMeOH[MeOH]
where k0 represents the rate of decay of A in the absence
of methanol and kMeOH is the quenching rate constant of
A by methanol. A plot of the observed pseudo-first-order
rate of the decay versus methanol concentration in the
range 0.5–2.0 M was linear and the slope of this plot
ꢀ1†
In order to gain more insight into the nature of the
transient species, the following trapping experiments
were carried out. First, LFP of 1 in degassed benzene in
the presence of pyridine gave essentially the same
transient absorption in terms of intensity and kinetics,
no new absorption bands being generated. It is well
known that carbenes are readily trapped by pyridine to
form ylides, which usually show intense absorption in the
visible region.11 Second, LFP of 1 in the presence of
methanol also generated the transient band, although the
optical yield of A was slightly decreased. However, the
decay rate was found to increase as the methanol
concentration increased. The apparent decay rate con-
yields the rate constant for the reaction of A with MeOH,
1
kMeOH = 8.7 Â 10 l mol
s
1, and the intercept yields
1
k0 = 7 Â 10 s (r = 0.999). Methanol is one of the most
efficient trapping reagents for carbenes usually in the
singlet state.8 The bimolecular quenching rate constant
for triplet diphenylcarbene, for instance, is 3 Â 107 l
1
mol
s
1.9 Even sterically hindered monophenylcar-
benes, e.g. triptycyl(phenyl)carbene, is trapped by
1
1 12
methanol at a rate of 3.2 Â 105 l mol
s
.
Thus,
kMeOH observed with A is at least three orders of
Figure 2. Absorption spectrum of the transient products formed during the irradiation of 1 in degassed benzene containing
2
Bu3SnH (3.72 Â 10 M) recorded 20 ms after excitation. Inset shows oscillogram traces monitored at 320 and 434 nm
Copyright 1999 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 12, 165–169 (1999)

LASER FLASH PHOTOLYSIS OF 3-NORADAMANTYL(PHENYL)DIAZOMETHANE
167
Scheme 1
magnitude smaller than that expected for carbene.
Finally, LFP in the presence of oxygen also resulted in
a decrease in the optical yield of A and an increase in the
decay rate. Again a plot of kobs against the O2
concentration is linear, and the slope gives a bimolecular
rate constant of 7.0 Â 106 l mol 1 s 1 for the reaction of
NMR (CDCl3, 400 MHz), ꢀ 7.49 (br d, 1H, J = 8.0 Hz),
7.36 (br d, 1H, J = 7.6 Hz), 7.31–7.13 (m, 4H), 7.09–6.98
(m, 2H), 6.92–6.81 (m, 1H), 2.94 (br s, 1H), 2.85 (br d, 1H,
J = 12 Hz), 2.68–2.39 (m, 5H), 2.32 (br d, 1H, J = 12 Hz),
2.10–1.45 (m, 16H), 1.38 (br t, 2H, J = 12 Hz), 1.31–1.21
(m,1H);13CNMR(CDCl3,100.4 MHz),ꢀ146.9(C),144.8
(C), 139.3 (C), 132.3 (CH), 131.6 (CH), 126.0 (CH), 125.6
(CH), 125.4 (CH), 125.24 (CH), 125.21 (CH), 125.1 (CH),
125.0 (CH), 54.4 (C), 48.0 (CH2), 46.1 (CH), 41.4 (CH2),
40.8 (C), 40.6 (CH2), 39.8 (CH2), 39.0 (C), 38.1 (CH2),
37.22 (CH2), 37.21 (CH2), 36.9 (CH2), 34.8 (CH2), 31.5
(CH), 31.3 (CH2), 29.8 (CH), 29.5 (CH), 29.2 (CH), 28.5
A with oxygen (kO ). It is well documented that most
2
triplet arylcarbenes are trapped by oxygen to generate the
corresponding ketone oxides very efficiently, mostly with
a nearly diffusion-controlled rate constant.9 For instance,
even triptycyl(phenyl)carbene is trapped by oxygen with
1
1 12
a bimolecular rate constant of 1.6 Â 109 l mol
s .
1
Moreover, the carbonyl oxides usually exhibit a broad
band around 390–450 nm13 and are long-lived enough to
be easily observed to grow as the carbene signals decay.9
No new transient absorption appeared in LFP of 1 in the
presence of O2 as transient bands due to A decayed.
All of these observations point to the same conclusion
that transient absorption at 434 nm is obviously not
ascribable to carbene 3. A plausible structure is then 2-
phenyladamantene (2), since 1-noradamantyldiazo-
methane is known to undergo ring expansion upon
decomposition.5,14 This assignment is supported by a
product analysis study. Thus, the main product isolated
from the spent solution in benzene was assigned to a dimer
of the adamantene based on the following data: HRMS,
calculated for C32H36 420.2817, found 420.2829; 1H
(CH), 28.4 (CH). The structure which explains these H
NMR data best is proposed to be 5, which is produced by
dimerizationin4 ‡ 2fashionincludingthearomaticringto
form the initial dimer (4), followed by H migration. These
product analysis data suggest that the phenyladamantene
mainly decays by undergoing dimerization in the absence
oftrapping reagent, whichisin accordance with the kinetic
behavior of the initial transient species A observed in LFP
of 1. On the other hand, the UV absorption spectrum of the
dimer(5)didnotcoincidewiththatobservedfortheproduct
from 2. However, we were able to isolate the initial dimer
(4) by generating the desired adamantene (2) by treating
1,2-diiodo-2-phenyladamantane with tert-butyllithium at
78°C. Dimer 4: colorless crystals; m.p. 242.0–243.5°C;
HRMS, calculated for C32H36 420.2817, found 420.2829;
Copyright 1999 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 12, 165–169 (1999)

168
K. HIRAI ET AL.
1H NMR (CDCl3, 400 MHz), ꢀ 7.36 (d, 1H, J = 8.3 Hz),
7.28–7.04 (m, 4H), 6.57 (d, 1H, J = 9.5 Hz), 6.11–5.97 (m,
2H),5.88(m,1H),4.04(brs,1H),3.28(brs,1H),2.85(brd,
1H, J = 13 Hz), 2.75–2.58 (m, 3H), 2.31 (br d, 1H,
J = 14 Hz), 2.16 (br d, 1H, J = 12 Hz), 2.05 (br s, 2H),
1.89–1.41(m,14H),1.34(brd,1H,J = 12 Hz),1.17(dt,1H,
J = 13, 2.7 Hz), 0.27 (dt, 1H, J = 13, 3.4 Hz); 13C NMR
(CDCl3, 67.8 MHz), ꢀ 147.7 (C), 146.9 (C), 132.1 (CH),
130.6 (CH), 130.4 (CH), 127.7 (CH), 126.1 (CH), 124.7
(CH), 123.8 (CH), 122.3(CH), 122.1(CH), 121.5(C), 53.5
(C), 50.3 (CH2), 46.7 (CH), 45.4 (C), 43.4 (CH2), 39.92
(CH2), 39.87 (CH2), 39.1 (CH2), 38.1 (CH2), 37.6 (CH2),
37.5 (C), 37.2 (CH2), 36.2 (CH2), 35.5 (CH2), 32.2 (CH),
31.9 (CH), 30.0 (CH), 29.33 (CH), 29.27 (CH), 28.0 (CH).
ThisdimerisnotonlyshowntoundergoHmigrationtogive
the final dimer 5 upon irradiation but also exhibits an
absorptionmaximum at 323 nm. HenceAismostprobably
phenyladamantene (2) which decays with second-order
kinetics to form a Diels–Alder dimer (4 = B), showing an
absorption maximum at 323 nm. Analysis of the photo-
products in methanol, on the other hand, showed the
presence of 2-phenyl-2-methoxyadamantane (6) as the
main product, obviously formed as a result of proton-
ation of 2 followed by nucleophilic attack of the solvent.
Phenyl(noradamantyl)methyl methyl ether (6'), expected
to be formed from carbene (3), was also detected, albeit
in a minor amount.
The present observations revealed several interesting
features of the reactivities of adamantene. It is generally
accepted that the stability of bridgehead alkenes is greatly
affected by substituents. For instance, in the case of
homoadamant-3-ene, while the parent compound can be
observed only at very low temperature,15 the 4-phenyl
derivative has been shown to have a half-life of over 12 h
in solution at room temperature, presumably owing to a
conjugative effect.16 Adamantene is also stabilized by
phenyl substitution to the extent that it is observable in
solution at room temperature, but its half-life is still seven
orders of magnitude smaller than that of the correspond-
ing homoadamantene. This is the first quantitative
experimental evidence confirming the difference in the
that of the decay of the peak due to 2, showing that 2 is
quenched by the hydride (Fig. 2). Since the excellent
hydrogen donor properties of the tin hydride have been
well recognized18 and since most benzyl radicals show
characteristic UV absorption peaks around 310–320
nm,19 the spectral changes are interpreted as indicating
that phenyladamantene abstracts hydrogen from Bu3SnH
to generate benzyl-type radicals (7). The absolute rate
constants for the abstraction reaction was estimated to be
1
1
6.3 Â 104 l mol
s .
Twisted p bonds are usually believed to have a
polarized character which reduces the strain energy.
Therefore, frequently employed trapping reagents are
polar, e.g. alcohols. However, when radical reaction
channels are available, radical-type reactions are some-
times observed. For instance, in the reactions with simple
alkenes and dienes, radical mechanisms are proposed
mostly based on product analysis data.5,17,20 However,
these radical reactions are usually observed at much
higher temperature in these cases. The present results are
therefore noteworthy in that the radical nature of the
twisted p bonds is verified at a much lower temperature
by direct observation of the intermediates in a quantita-
tive manner, and revealed that the twisted p bond in 2
shows profound radical over ionic character.
Finally, it is important to comment on the mechanism
of the formation of 2 in the photochemical reaction of 1.
LFP data in the presence of typical trapping reagents for
carbenes, e.g. pyridine, methanol, oxygen and Bu3SnH,
clearly suggested that carbene does not intervene at least
as the main intermediate leading to 2. What is a ‘real’
intermediate, then? It is now well documented21,22 that
excited states of diazomethanes often mimic the reactions
of carbenes. Intramolecular chemistry such as 1,2-
migrations is particularly bedeviled by such reactions,
in which diazomethanes play the roles traditionally
assigned to carbenes. The present LFP data are also
compatible with a similar assumption that 2 is produced
directly from the excited state of 1.
measure of ‘olefinic stability’ between adamantene
REFERENCES
1 3
(39.5 kcal mol
)
and
homoadamantene
(20.2
kcal mol 1).15
1. P. M. Warner, Chem. Rev. 89, 1067 (1989); W. Luef and R. Keese,
Top. Stereochem. 20, 231 (1991); G. Szeimes, in Advances in
Strain in Organic Chemistry, edited by B. Halton. JAI Press,
Greenwich, CT (1992).
2. J. Bredt, H. Thouet and L. Schmitz, Justus Liebigs Ann. Chem. 437,
1 (1924).
3. W. F. Maier and P. v. R. Schleyer, J. Am. Chem. Soc. 103, 1891
(1981).
4. J. Michl, J. G. Radziszewski, J. W. Downing, J. Kopecky, P.
Kaszynski and R. D. Miller, Pure Appl. Chem. 59, 1613 (1987).
5. For a brief history of adamantene, see N. Bian and M. Jones Jr, J.
Am. Chem. Soc. 117, 8957 (1995).
6. T. Okazaki, H. Isobe, T. Kitagawa and K. Takeuchi, Bull. Chem.
Soc. Jpn. 69, 2053 (1996).
7. For a description of the LFP system, see H. Tomioka, K. Mizutani,
K. Matsumoto and K. Hirai, J. Org. Chem. 58, 7128 (1993).
8. R. A. Moss and M. Jones Jr, Carbenes, Wiley, New York, Vol. 1
(1973) and Vol. 2 (1975); W. Kirmse, Carbene Chemistry,
The quenching rate constants of 2 with O2 and MeOH
are also noteworthy. The data suggest that phenylada-
mantene is five orders of magnitude more reactive
towards O2 than MeOH. This is unexpected in terms of
both trends and magnitude since bridgehead alkenes are
known to undergo polar addition of alcohols smoothly
whereas their reactions with O2 are rare.16,17 More
interestingly, the transient absorption band was found to
be quenched even by hydrogen donors. Thus, LFP of 1 in
the presence of tri(n-butyl)tin hydride again resulted in a
dramatic decrease in the lifetime of 2 and a concurrent
appearance of a new absorption band at 320 nm. The rate
of increase in the band at 320 nm is almost the same as
Copyright 1999 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 12, 165–169 (1999)

LASER FLASH PHOTOLYSIS OF 3-NORADAMANTYL(PHENYL)DIAZOMETHANE
169
Academic Press, New York (1971); M. Regitz, Methoden der
Organischen Chemie (Houben Weyl), Vol. E19b. Georg Thieme,
Stuttgart (1989).
9. M. S. Platz, Kinetics and Spectroscopy of Carbenes and
Biradicals, Plenum Press, New York (1990).
10. A. M. Trozzolo, Acc. Chem. Res. 1, 329 (1968); W. Sander, G.
Bucher and S. Wierlacher, Chem. Rev. 93, 1583 (1993).
11. J. E. Jackson and M. S. Platz, in Advances in Carbene Chemistry,
edited by U. H. Brinker, Vol. 1, p. 89. JAI Press, Greenwich, CT
(1994).
12. H. Tomioka, J. Nakajima, H. Mizuno, T. Sone and K. Hirai, J. Am.
Chem. Soc. 117, 11355 (1995).
13. W. W. Sander, Angew. Chem., Int. Ed. Engl. 29, 344 (1990).
14. D. J. Martella, M. Jones Jr and P. v. R. Schleyer, J. Am. Chem. Soc.
100, 2896 (1978).
15. D. J. Martella, M. Jones Jr, P. v. R. Schleyer and W. F. Maier, J.
Am. Chem. Soc. 110, 7634 (1979).
16. M. Ohno, M. Itoh, M. Umeda, R. Furuta, K. Kondo and S. Eguchi,
J. Am. Chem. Soc. 118, 7075 (1996).
17. H. O. House, R. J. Outcalt, J. L. Haack and D. Van Derveer, J. Org.
Chem. 48, 1654 (1983).
18. See, for instance, G. S. Hammond and P. A. Leermakers, J. Am.
Chem. Soc. 84, 207 (1962).
19. K. Tokumura, T. Ozaki, H. Nosaka, Y. Saigusa and M. Itoh, J. Am.
Chem. Soc. 113, 4974 (1991).
20. K. B. Becker and M. K. Hohermuth, Helv. Chim. Acta 65, 229
(1982); K. B. Becker, M. K. Hohermuch and G. Rihs, Helv. Chim.
Acta 65, 235 (1982).
21. S. Celebi, S. Leyva, D. A. Modarelli and M. S. Platz, J. Am. Chem.
Soc. 115, 8613 (1993), and references cited therein.
22. H. Tomioka, H. Kitagawa and Y. Izawa, J. Org. Chem. 44, 3072
(1979); H. Tomioka, H. Okuno and Y. Izawa, J. Org. Chem. 45,
5278 (1980); H. Tomioka, M. Kondo and Y. Izawa, J. Org. Chem.
46, 1090 (1981).
Copyright 1999 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 12, 165–169 (1999)