Photolysis of 1,2-naphthoquinone in oxygen-doped argon matrix at low temperature

Article abstract of DOI:10.1246/bcsj.82.737

Photolysis of 1,2-naphthoquinone (1) in argon matrix either with or without oxygen at 10K was investigated by IR spectroscopy in combination with DFT calculations. The results indicate that 1 gave bis(ketene) 2 as a main product along with a small amount

Full text of DOI:10.1246/bcsj.82.737

© 2009 The Chemical Society of Japan
Bull. Chem. Soc. Jpn. Vol. 82, No. 6, 737–742 (2009)
737
Photolysis of 1,2-Naphthoquinone in Oxygen-Doped
Argon Matrix at Low Temperature
Takayoshi Itoh,1 Jiro Tatsugi,*1 and Hideo Tomioka*1,2
1Department of Applied Chemistry, Aichi Institute of Technology, Toyota 470-0392
2Nagoya Industrial Science Research Institute, 1-13 Yotsuya-dori, Chikusa-ku, Nagoya 464-0819
Received February 2, 2009; E-mail: jtatsugi@aitech.ac.jp
Photolysis of 1,2-naphthoquinone (1) in argon matrix either with or without oxygen at 10 K was investigated by IR
spectroscopy in combination with DFT calculations. The results indicate that 1 gave bis(ketene) 2 as a main product along
with a small amount of inden-1-one (4), most probably as a result of Type I cleavage, followed by electron reorganization
and decarbonylation in the resulting diacyl radical, respectively. Bis(ketene) 2 was shown to transform to isomeric ketene
3 upon continued irradiation under these conditions. When the irradiation was carried out in argon matrix doped with 20%
oxygen, 1 decomposed much more efficiently than that in argon matrix and six- and eight-membered cyclic peroxides
presumably formed by trapping of initial diradical originating from ¡-cleavage by molecular oxygen were detected at
complete expense of 3 and 4. The formation of 2 even in the presence of oxygen suggested that 2 may partly be formed by
a concerted pathway in the singlet excited state. The present observation reveals that photolysis of ¡-diketones in oxygen-
doped matrix at low temperature provides useful information concerning the reaction pathway of the diketone upon
photoexcitation.
Photolysis of ¡-diketones in the presence of oxygen in
solution attracts considerable attention.1-5 Under these con-
ditions, diketones are converted apparently to intermediate
peroxy radicals, and observed final products are carboxylic
anhydrides or the corresponding acids. When alkenes are
present, the intermediate peroxy radicals react with the double
bond to give the corresponding epoxides, thus providing a mild
procedure for epoxidation of alkenes.
toring of the irradiation of 1 indicated that the peaks due to 1
decreased in intensity as new absorption bands showing peaks
in the region of 1700 to 2400 cm appeared. The absorption
bands due to products after 24 h irradiation are shown in
Figure 1b. A careful analysis of those product bands by plotting
their intensities as a function of irradiation time (Figure 2,
¹1
dotted lines) revealed that there were at least three products
¹1
designated as A, B, and C. The band due to A (at 2100 cm
)
It has been proposed that the most plausible peroxy radicals
are ketone-oxygen adduct diradical (I) formed as a result of
attack of triplet oxygen on the carbonyl carbon in the excited
triplet state of the diketones. Since the ¢-scission of an alkoxy
radical is a facile process, the diradical I then decays to radical
pair II. Formation of anhydride is explained in terms of attack
of acyl radical on the carbonyl oxygen of acyl peroxide in II
leading to III. Since acyl radical is known to efficiently react
with oxygen,6 acyl radical in II is trapped by a second oxygen
thus generating an acyl peroxy radical (IV), which eventually
produces acids (Scheme 1).
These intermediates have been proposed mostly based on
product analysis studies and it is only recently that the initial
ketone-oxygen adduct (I) has been detected by using time-
resolved UV spectroscopy.7 We have shown8 that IR studies of
photolysis of ketones in argon matrix in the presence of oxygen
at low temperature provide useful information on reaction
pathway of ketone upon photoexcitation. As an extension of
such studies, we carried out similar spectroscopic studies in the
photolysis of 1,2-naphthoquinone.
arose rapidly at the initial stage of the irradiation and started to
decrease as the irradiation was continued, while the band due
to B (2144 cm¹1) formed rather slowly at initial stage but
increased as the bands due to A decreased. The bands due to C
(1741 cm¹1) appeared to form and increase independently.
To characterize those products, DFT calculations were
carried out at the B3LYP/6-31G(d) level of theory. Thus, the
3
O
*
O
O
O2
R
hν
R
R
R
R
R
O
O
O
O
(I)
O
O
R
O
O
O
O
O
O
O
R
O
R
R
R
O
R
R
O
(III)
(II)
O
O
O2
H
O
O
R
Results
O
(IV)
1,2-Naphthoquinone (1) was deposited in an Ar matrix at
20 K and irradiated (- > 390 nm) at 10 K (Figure 1). IR moni-
Scheme 1.
Published on the web June 12, 2009; doi:10.1246/bcsj.82.737

738
Bull. Chem. Soc. Jpn. Vol. 82, No. 6 (2009)
Photolysis of Naphthoquinone in Ar-O2 Matrix
a)
B
2144
A
b)
A
B
C
2100
1741
1800
Product
Reactant
A
A
2080
2116
c)
d)
e)
B
2142
B
1798
1741
C
wavelength/nm
Figure 1. Photolysis of 1,2-naphthoquinone (1) in an Ar matrix at 10 K. (a) Calculated (B3LYP/6-31G(d)) spectra of 1. (b) IR
spectrum of the photoproduct formed after 24 h of irradiation (- > 390 nm) of 1. (c, d, and e) Calculated [B3LYP/6-31G(d)] spectra
of bis(ketene) 2 (c), keto-ketene 3 (d), and inden-1-one (4) (e), respectively.
geometries of plausible products expected to be formed in the
photolysis of 1 were optimized and the vibrational frequencies
were calculated. Since both A and B have intense and sharp
bands at around 2000-2200 cm¹1, they are most likely ketenes.
Thus, ketenes 2, expected to be formed by cleavage of CO-CO
bonds, and 3, presumably produced by cyclization of 2 were
optimized. Two isomers of 2 (2A and 2B) (Chart 1) are close in
energy (0.85 kcal mol¹1) and are shown to have almost
indistinguishable frequencies but with slight difference in the
intensities. Those bands match relatively well with that due to
vibrational frequencies calculated for inden-1-one (4) match
best with that due to C (Scheme 2).
In order to know a rough ratio of those three products,
calibrations were made by dividing experimentally observed
¹1
absorbance of each peak with that calculated (1426 km mol
¹1
¹1
¹1
at 2116 cm for 2, 685 km mol at 2142 cm for 3, and
231 km mol at 1741 cm for 4, respectively).9 The results
displayed in Figure 2 (solid lines) indicate that the main initial
pathway is the formation of ketene 2, which is converted to 3
upon photoexcitation and that the formation of 4 is a minor
process.
¹1
¹1
¹1
product A, assuming that the band at 2116 cm calculated for
2 is included in the hem of a rather broad band with maximum
at 2144 cm¹1. On the other hand, the bands calculated for 3
match well with that due to product B. The band at 1741 cm
due to C is obviously due to ketone carbonyl group. Among
several possible products including keto-carbenes to be formed
as a result of ring-contraction (see Supporting Information), the
When the irradiation was carried out in the presence of
oxygen, appreciably different spectral changes were obtained.
Thus, when 1 matrix-isolated in argon doped with 20% oxygen
was irradiated (- > 390 nm) at 10 K, the bands due to 1
appeared to diminish much more efficiently than that in an
undoped matrix. The absorption bands due to products after
¹1

T. Itoh et al.
Bull. Chem. Soc. Jpn. Vol. 82, No. 6 (2009)
739
2144 cm-1
0.02
0.018
0.016
0.014
0.012
0.00003
O
O
0.000025
0.00002
0.000015
0.00001
0.000005
0
O
O
2100 cm-1
0.01
0.008
0.006
0.004
0.002
0
1741cm-1
O
0
10
20
30
40
50
60
Irr.Time/h
Irradiation of NQ in Ar matrix at 10 K
Figure 2. Plot of peaks due to 2, 3, and 4 as a function of irradiation time. Dotted and solid lines are for the observed absorbance and
for the corrected absorbance obtained by dividing with the calculated intensities, respectively.
O
O
O
O
O
hν
+
O
+
Ar, 10 K
O
1
2
3
4
Scheme 2.
at 1763 cm
¹1
is in agreement with that calculated for 6
O
O
(Figure 3 and Scheme 3). Again calibrations were made to
estimated a rough ratio of the products by using calculated
¹1
intensities (560 km mol¹1 at 2145 cm¹1 for 5 and 461 km mol
¹1
at 1768 cm for 6, respectively) (Figure 4, solid lines), which
O
indicate that 5 was formed as major product while 6 was
produced in only approximately 20% of 5 and that unoxidized
product 2 was formed as a minor product.
O
2A
2B
Chart 1.
Discussion
Cyclic diketones are known to undergo Type I cleavage upon
photoexcitation to generate two acyl radicals that can lead to
decarbonylation. In a conjugated diketone, bis(ketene) forma-
tion is also possible.10 Irradiation of o-benzoquinone in matrix
at low temperature was monitored by IR spectroscopy, which
indicated the formation of bis(ketene) as a main product along
with the formation of cyclopentadienone.11 Although
the mechanism of the reaction has not been proposed, the
initial cleavage of the C-C bond between the two carbonyl
groups leading to diacyl radical, followed by either electronic
reorganization forming bis(ketene) or decarbonylation forming
cyclopentadienone is presumed to be the most likely pathway.
The formation of 2 and 4 in the photolysis of 1 in argon matrix
can be easily understood as the reaction of a benzo analogue of
o-quinone. On the other hand, ketene 3 is likely to be formed
from 2 upon photoexcitation, although 3 can also be produced
13 h irradiation are shown in Figure 3b. New bands formed
after photoirradiation were significantly different from that
observed in the irradiation of 1 in undoped matrix. The bands at
¹1
2100 and 2142 cm are very similar to those observed for A
(2) in the irradiation of 1 in undoped matrix. The other bands
were not observed in the undoped run and hence they are most
probably ascribable to oxidation products. Again analysis was
made on those product bands by plotting their intensities as a
function of irradiation time (Figure 4, dotted lines), which
revealed that there were at least three products designated as D
to F, all of which increased monotonously with the irradiation
time in this case. DFT calculations were carried out on
plausible oxidation products expected to be formed during the
photooxidation of 1. The bands due to D with a strong peak at
¹1
2143 cm match well with that calculated for 5, while a band

740
Bull. Chem. Soc. Jpn. Vol. 82, No. 6 (2009)
Photolysis of Naphthoquinone in Ar-O2 Matrix
a)
b)
2142
D
F
E
1778
D
1759
E
1763
F
2100
D
2145
D
E
c)
d)
1751
1768
E
1781
F
F
2080
2116
e)
wavelength/nm
Figure 3. Photolysis of 1,2-naphthoquinone (1) in 20% O2 doped Ar matrix at 10 K. (a) Calculated (B3LYP/6-31G(d)) spectra of 1.
(b) IR spectrum of the photoproduct formed after 13 h of irradiation (- > 390 nm) of 1. (c, d, and e) Calculated [B3LYP/6-31G(d)]
spectra of six-membered cyclic peroxide 5 (c), eight-membered cyclic peroxide 6 (d), and keto-ketene 3 (e), respectively.
O
O
O
O
It should be noted that 5 is formed predominantly over 6.
This is somewhat unexpected since ketene 2 presumably
derived from resonance form DRa is formed as major product
in an undoped matrix photolysis run. The profile of product-
irradiation time plot in oxygen-doped matrix indicates that two
oxidation products 5 and 6 are formed from the initial stage of
the irradiation and do not isomerizes each other upon further
irradiation and that ketene 2 is not likely to undergo photo-
oxidation under these conditions.12 The formation of oxidation
products then may be interpreted as indicating a stepwise
reaction between DR and triplet oxygen. Given the high
reactivity of acyl radical toward oxygen,6 it is reasonable to
assume that the benzoyl radical carbon of DR is trapped by
oxygen to form peroxy-acyl diradical (DR-O2). The oxygen
radical center of the diradical either undergoes coupling at the
acyl carbon in resonance form DRa forming 6 or at the benzyl
O
O
O
O
hν
+
20%-O2/Ar
10 K
O
1
5
6
O
Scheme 3.
directly from the diacyl radical (DR) in its resonance form
(DRb) (Scheme 4).
Irradiation of 1 in the presence of oxygen gave products as a
result of trapping of intermediates involved in photochemical
reaction of 1. The formation of 6, for instance, clearly supports
the intervention of the another resonance form (DRa) in the
photolysis of 1 leading to 2 and 4. The formation of 5 can also
be explained in terms of oxygen trapping of DR in its
resonance form DRb.

T. Itoh et al.
Bull. Chem. Soc. Jpn. Vol. 82, No. 6 (2009)
741
0.04
0.035
0.03
0.025
0.02
0.015
0.01
0.005
0
0.00008
O
0.00007
0.00006
0.00005
0.00004
0.00003
0.00002
0.00001
0.00000
O
O
O
2143 cm-1
O
O
O
2100 cm-1
O
O
1763 cm-1
O
0
5
10
15
Time/h
20
25
30
Figure 4. Plot of peaks due to 2, 5, and 6 as a function of irradiation time. Dotted and solid lines are for the observed absorbance and
for the corrected absorbance obtained by dividing with the calculated intensities, respectively.
O
O
O
O
O
O
O
O
O
hν
hν
O2
O
O
20%-O2/Ar
10 K
O
2
3
DR-O2
1
DR
O
O
O
O
O
O
O
O
hν
O
O
O
Ar, 10 K
O
O
1
DRa
- CO
DRb
6
5
O
O
Scheme 5.
2.5
2
: Ar Matrix
: 20%O2-Ar Matrix
4
Scheme 4.
1.5
1
carbon in the resonance form DRb leading to 5. The formation
of 5 is favored partly because the contribution of DRb is more
important than DRa due to benzyl stabilization and partly
because the formation of six-membered ring 5 is favored over
that of eight-membered ring 6 (Scheme 5).
The observation that 2 is formed even in the presence of
oxygen while 3 and 4 are almost completely quenched is also
noteworthy. If all three products 2, 3, and 4 are formed from a
common intermediate such as DR, they are expected to be
quenched equally in the presence of oxygen. The formation of
2 even in the presence of oxygen means that some of 2 should
be produced not by way of DR. A plausible pathway that is not
quenched by oxygen is a concerted reaction in the singlet
excited state of 1.
0.5
0
0
5
10
15
20
25
Time/h
Figure 5. Plot of decay of the band due to 1 as a function of
irradiation time in Ar (solid line) and in 20% O2 doped Ar
matrix (dotted line) at 10 K.
to that in the undoped matrix (Figure 5). This may be partly
due to the intervention of the quenching pathway of the
diradical by oxygen if one assumes that some of the initial
diradicals undergo recombination to return to starting quinine
1. On the other hand, oxygen is known to enhance the overall
Finally it is worth noting that the rate of photodecomposition
of 1 was markedly increased in the O2-doped matrix compared

742
Bull. Chem. Soc. Jpn. Vol. 82, No. 6 (2009)
Photolysis of Naphthoquinone in Ar-O2 Matrix
S1 ¼ T1 process.13 Thus, in the presence of oxygen, the
intersystem crossing efficiency of initially generated singlet
state to the triplet is accelerated, which results in the increase in
population of the triplet n, ³* state from which ¡-cleavage of
the ketones takes place.
References
See for reviews: a) J. Saltiel, H. C. Curtis, Mol. Photochem.
1969, 1, 239. b) K. Gollnick, Adv. Photochem 1968, 6, 1.
1
2
3
N. Shimizu, P. D. Bartlett, J. Am. Chem. Soc. 1976, 98, 4193.
a) Y. Sawaki, Y. Ogata, J. Am. Chem. Soc. 1981, 103, 2049.
b) Y. Sawaki, Bull. Chem. Soc. Jpn. 1983, 56, 3464. c) Y. Sawaki,
C. S. Foote, J. Org. Chem. 1983, 48, 4934. d) Y. Sawaki, Y. Ogata,
J. Org. Chem. 1984, 49, 3344. e) Y. Sawaki, Tetrahedron 1985, 41,
2199.
Conclusion
Present spectroscopic observations demonstrate that 1,2-
naphthoquinone undergoes mainly ¡-cleavage to give diacyl
diradical intermediate, which either leads to bis(ketene) or
undergoes decarbonylation. Photooxidation of the quinone on
the other hand, is completely different from that observed in
solution phase but provides useful information concerning the
reaction pathway of its reaction upon excitation.
4
P. D. Bartlett, A. A. M. Roof, N. Shimizu, J. Am. Chem.
Soc. 1982, 104, 3130.
5
A. P. Darmanyan, C. S. Foote, P. Jardon, J. Phys. Chem.
1995, 99, 11854.
6
C. E. Brown, A. G. Neville, D. M. Rayner, K. U. Ingold, J.
Lusztyk, Aust. J. Chem. 1995, 48, 363.
7
8
G. Cosa, J. C. Scaiano, J. Am. Chem. Soc. 2004, 126, 8636.
T. Itoh, J. Tatsugi, H. Tomioka, Bull. Chem. Soc. Jpn. 2009,
Experimental
Materials. 1,2-Naphthoquinone was purchased from Tokyo
Kasei Co., and used as provided.
82, 475.
9
It should be noted here that the reliability of calculated IR
Matrix-Isolation Spectroscopy.
Matrix experiments were
intensities are not established yet, in contrast with that of
frequencies. a) J. G. Radziszewski, M. R. Nimlos, P. R. Winter,
G. B. Ellison, J. Am. Chem. Soc. 1996, 118, 7400. b) A. V.
Friderichsen, J. G. Radziszewski, M. R. Nimlos, P. R. Winter,
D. C. Dayton, D. E. David, G. B. Ellison, J. Am. Chem. Soc. 2001,
123, 1977. c) Y. Yamaguchi, M. Frisch, J. Gaw, H. F. Schaefer, III,
J. S. Binkley, J. Chem. Phys. 1986, 84, 2262.
10 See for reviews: M. B. Rubin, in CRC Handbook of
Organic Photochemistry and Photobiology, ed. by W. M.
Horspool, P.-S. Song, CRC Press, Boca Raton, 1995, p. 437, and
references cited therein.
performed by means of standard techniques14,15 using a closed-
cycle helium cryostat. For IR experiments, a CsI window was
attached to the copper holder at the bottom of the cold head. Two
opposing ports of a vacuum shroud surrounding the cold head were
fitted with KBr with a quartz plate for UV irradiation and a
deposition plate for admitting the sample and matrix gas. For UV
experiments, a sapphire cold window and a quartz outer window
were used. The temperature of the matrix was maintained by a
controller (gold vs. chromel thermocouple).
All the samples used in this study were not easily vaporized and
hence were directly deposited on the window from a glass tube
equipped with a ceramic heater under a stream of argon. Therefore,
the concentration of the sample on the window was different at
each run.
11 a) H. Tomioka, H. Fukao, Y. Izawa, Bull. Chem. Soc. Jpn.
1978, 51, 540. b) G. Maier, L. H. Franz, H.-G. Hartan, K. Lanz,
H. P. Reisenauer, Chem. Ber. 1985, 118, 3196.
12 Ketenes are known to react with oxygen: a) N. Obata, T.
Takizawa, J. Chem. Soc. D 1971, 587. b) B. M. Trost, J. Am.
Chem. Soc. 1969, 91, 918. c) J. Meinwald, H. O. Klingele, J. Am.
Chem. Soc. 1966, 88, 2071. d) M. B. Rubin, R. G. LaBarge, J. Org.
Chem. 1966, 31, 3283. e) G. E. Gream, J. C. Paice, C. C. R.
Ramsey, Aust. J. Chem. 1969, 22, 1229; G. E. Gream, J. C. Paice,
C. C. R. Ramsey, Aust. J. Chem. 1967, 20, 1671. f) D. C. Zhao,
T. T. Tidwell, J. Am. Chem. Soc. 1992, 114, 10980.
13 N. J. Turro, Modern Molecular Photochemistry, Benjamin/
Cummings Publishing Co., Inc., New York, 1978.
14 H. Tomioka, N. Ichikawa, K. Komatsu, J. Am. Chem. Soc.
1992, 114, 8045.
Irradiations were carried out with a 500-W xenon high-pressure
arc lamp. For broad-band irradiation, cutoff filters were used (50%
transmittance at the specified wavelength). For monochromatic
light irradiation, a monochrometer was used. IR spectra were
measured on a Shimadzu FTIR-4800S spectrometer, and UV-vis
spectra were recorded on a Shimadzu UV-2450 spectrophotometer.
Computational Procedures. DFT calculations were carried
out using the Gaussian 94,16 programs. Optimized geometries were
obtained at the B3LYP/6-31G(d)17 levels of theory. Vibrational
frequencies obtained at the B3LYP level of theory were scaled by
0.961 and zero-point energies (ZPE) by 0.981.18 Transition states
were located using Gaussain program (Rational Function Opti-
mization-pseudo-Newton-Raphsonthe method).19 The nature of
each stationary point was confirmed with harmonic frequency
calculations, i.e., minima have exactly one imaginary frequency
related to the expected movement.
15 R. J. McMahon, O. L. Chapman, R. A. Hayes, T. C. Hess,
H. P. Krimmer, J. Am. Chem. Soc. 1985, 107, 7597.
16 M. J. Frish, G. W. Trucks, H. B. Schlegel, P. M. W. Gill,
B. G. Johnson, M. A. Robb, J. R. Cheeseman, T. Keith, G. A.
Petersson, J. A. Montgomery, K. Raghavachari, M. A. Al-Laham,
V. G. Zakrzewski, J. V. Ortiz, J. B. Foresman, J. Cioslowski, B. B.
Stefanov, A. Nanayakkara, M. Challacombe, C. Y. Peng, P. Y.
Ayala, W. Chen, M. W. Wong, J. L. Andres, E. S. Replogle, R.
Gomperts, R. L. Martin, D. J. Fox, J. S. Binkley, D. J. Defrees, J.
Baker, J. P. Stewart, M. Head-Gordon, C. Gonzalez, J. A. Pople,
Gaussian 94, Revision D.4, Gaussian Inc., Pittsburgh, 1995.
17 a) C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785.
b) B. Miehlich, A. Savin, H. Stoll, H. Preuss, Chem. Phys. Lett.
1989, 157, 200. c) A. D. Becke, J. Chem. Phys. 1993, 98, 5648.
18 A. P. Scott, L. Radom, J. Phys. Chem. 1996, 100, 16502.
19 J. Simons, P. Jørgensen, H. Taylor, J. Ozment, J. Phys.
Chem. 1983, 87, 2745.
The authors are grateful to the Ministry of Education,
Culture, Sports, Science and Technology of Japan for the
support of this work through a Grant-in-Aid for Scientific
Research (C, No. 18550048).
Supporting Information
Cartesian coordinates of compounds 1-6 and Calculated
[B3LYP/6-31G-(d)] IR spectra of other products expected to be
formed in the photolysis of 1. These materials are available free of
charge on the Web at: http://www.csj.jp/journals/bcsj/.