Remote substituent effects on the photooxygenation of 9,10- diarylanthracenes: Strong evidence for polar intermediates

Article abstract of DOI:10.1039/b719637g

Two different reaction pathways in the photooxygenation of 9,10-diarylanthracenes are identified, with strong evidence for polar (forward, singlet oxygen addition) and radical (backward, thermolysis) intermediates. The Royal Society of Chemistry.

Full text of DOI:10.1039/b719637g

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Remote substituent effects on the photooxygenation of
9,10-diarylanthracenes: strong evidence for polar intermediatesw
Werner Fudickar and Torsten Linker*
Received (in Cambridge, UK) 20th December 2007, Accepted 21st January 2008
First published as an Advance Article on the web 14th February 2008
DOI: 10.1039/b719637g
Two different reaction pathways in the photooxygenation of
9,10-diarylanthracenes are identified, with strong evidence for
polar (forward, singlet oxygen addition) and radical (backward,
thermolysis) intermediates.
groups should shift the absorption bands. However, for all
systems almost same UV-vis spectra were obtained, which
indicates the orthogonal orientation of the benzene rings in the
ground state. These observations are also in accordance with
11,12
X-ray analyses.z
In contrast, remote effects should be
Anthracenes are important aromatic compounds,1 which react
with singlet oxygen (1O2) to form the corresponding endoper-
oxides.2 For 9,10-diphenylanthracene (1a), the addition is fully
reversible and the parent anthracene is regenerated by the
release of oxygen, to a certain extent in the excited state.2,3 The
reversibility of the reaction enabled applications of anthra-
cenes as part of photochromic systems,4 as photoresists5 and
carriers of singlet oxygen.6
operative for the photoionized radical cations, supported by
strong spectral shifts in the transient absorption spectra and
their trapping with nucleophilic solvents.13
To prove the coplanar alignment of the anthracenes 1a–h in
the radical cations, we determined the redox potentials by
cyclic voltammetry (Table 1). Indeed, the oxidation potentials
vary strongly with the electronic nature of the substituents and
the electron-withdrawing cyano group (1d,g) shifts the poten-
tial by about 200 mV compared to the methoxy group (1b,e).
The different propensities to form the radical cations should
The mechanism of the addition of 1O2 to polyacenes is still a
subject of intensive theoretical and experimental research.7 In
principle both a concerted pathway via the formation of
exciplexes and a stepwise process are evident.8 Recent calcula-
tions based on the UB3LYP method indicated that biradical
intermediates were formed by the photooxygenation of
anthracenes.9 However, a stepwise mechanism characterized
by polar or zwitterionic intermediates has not been considered
for the reaction of polyacenes. We observed rotations around
the aryl–aryl bond during the photooxygenation of ortho
substituted 9,10-diarylanthracenes for the first time, which
resulted in an efficient molecular switch between cis and
trans isomers.10 This process required a planarization of the
biaryl moiety in the transition state or intermediates, but no
mechanistic rational could be provided. To elucidate the
reaction mechanism, we synthesized eight different 9,10-di-
arylanthracenes, 1, by Suzuki coupling from commercially
available starting materials in only one step.11 Thus, the
electronic nature of the two appending benzene rings could
be altered by the substitution pattern. The photooxygenation
afforded the corresponding endoperoxides, 2, in high yields.
Remote substituent effects on the reactivity give strong
evidence for zwitterionic intermediates, 3, during the reaction
with singlet oxygen (Scheme 1).
1
also be reflected in the reactivity of the anthracenes with O2,
but only if a zwitterionic intermediate, 3, and not a radical, 4,
would be formed.
Therefore, anthracenes 1a–h were photooxygenated to the
corresponding endoperoxides 2a–h by irradiation in the presence
First, the ground state geometries of the anthracenes 1a–h
were compared by UV-vis spectroscopy (Table 1). Assuming a
coplanar alignment of the arene rings, the acceptor and donor
Department of Chemistry, University of Potsdam, Karl-Liebknecht
Str. 24-25, D-14476 Potsdam/Golm, Germany.
E-mail: linker@chem.uni-potsdam.de; Fax: +49 331 9775056;
Tel: +49 331 9775212
w Electronic supplementary information (ESI) available: Detailed
experimental procedures, 1H and 13C NMR and analytical data for
all compounds. See DOI: 10.1039/b719637g
Scheme 1 Reaction of the diarylanthracenes, 1, with singlet oxygen
to form the endoperoxides, 2, via zwitterions, 3, and reversible
thermolysis via biradicals, 4.
ꢀc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 1771–1773 | 1771

View Article Online
Table 1 Spectroscopic and voltammetric data of anthracenes 1a–h
and their rate constants with singlet oxygen
be 120 kJ molꢁ1 17
,
but the high energy brought by oxygen in
the excited state seems to overpower this barrier in a polar
intermediate. Additionally, this result explains the function of
the anthracenes 1e–h as molecular switches.10 The lower
reactivity of ortho (1e–h) compared to para (1b–d) substituents
indicates that steric interactions are indeed operative, since the
coplanar alignment is more difficult to achieve. Finally, the
Anthracene
lmax/nm
EOx/Va
ERed/Va kd/106 Mꢁ1 sꢁ1
1a
1b
1c
1d
1e
1f
1g
1h
a
357, 376, 396 1.29
359, 377, 398 1.19 (1.4)b
357, 376, 397 1.26
356, 375, 395 1.45
357, 376, 396 1.25
357, 375, 396 1.32
355, 373, 394 1.55
1.21
1.11
1.15
1.36
1.18
1.24
1.45
3.0 ꢂ 0.15
5.44 ꢂ 0.26
4.1 ꢂ 0.15
0.74 ꢂ 0.03
2.09 ꢂ 0.1
0.46 ꢂ 0.02
0.11 ꢂ 0.01
3.69 ꢂ 0.16
1
reactivities of the 9,10-diarylanthracenes 1e–h with O2 again
nicely correlate with their oxidation potentials (Table 1),
which gives further evidence for zwitterions 3e–h, even for
ortho substituted systems.
357, 376, 396 1.10 (1.41)b 1.04
In dichloromethane vs. standard calomel electrode at 100 mV sꢁ1
,
To investigate the remote effects on the reverse reaction, we
thermolyzed the endoperoxides 2a–h at temperatures ranging
from 60–110 1C, and obtained thermodynamic data from
Arrhenius and Eyring plots.11 The parent anthracenes 1a–h
were reisolated quantitatively and singlet oxygen yields were
determined by trapping with tetracyclone (Table 2). The
positive activation entropies and low yields of 1O2 affirm a
strong participation of a stepwise pathway, which is in accor-
dance to studies of Turro et al. for the cycloreversion of
unsubstituted 9,10-diphenylanthracene (1a).3
supporting electrolyte was Bu4NClO4 (0.1 M), measured vs. ferrocene
b
oxidation (Ea = 0.51 V vs. SCE). The value in parentheses gives the
potential of the second oxidation peak.
of methylene blue (MB, Scheme 1) and their bimolecular rate
constants (kd) were determined spectroscopically (Table 1).11,14
Indeed, the reactivity of the 9,10-diarylanthracenes 1a–d
with 1O2 strongly depends on the substituent in the para
position. Thus, the rate constant increases, when switching
from the electron-withdrawing cyano group to the electron-
donating methoxy group, nicely correlating with the oxidation
potentials (Table 1). A simple electron transfer mechanism in
the first step, resulting in the formation of a superoxide radical
anion as the sole reactive oxygen species, was disproved by
quenching experiments using DABCO as singlet oxygen
trap.15 Therefore, substituents affect exclusively the intermedi-
ate peroxo adduct electronically. Such a remote substituent
effect was hitherto unknown and gives strong evidence for
zwitterionic intermediates 3 with a coplanar alignment of the
benzene moiety (Scheme 1). A radical intermediate 4, which
was previously postulated by theoretical studies,8,9 is therefore
unlikely, since donor and acceptor groups stabilize radicals in
a similar way.16
Interestingly, ortho substituted systems react faster than the
para derivatives (Fig. 1, Table 2). Steric and not electronic
factors may play the dominant role, since the concave geo-
metry of the endoperoxides 2 confines the space near the ortho
substituents. On the other hand, the electronic nature of a
substituent in the para position, in definite contrast to the
addition of singlet oxygen, plays no role for the reactivity of
endoperoxides 2a–d, which is demonstrated by almost iden-
tical rates of thermolysis of the endoperoxides 2a–d (Fig. 1,
Table 2). Electronic effects are therefore not present in the
family of para substituted endoperoxides. Consequently, cy-
cloreversion proceeds via a different pathway as the cycloaddi-
tion and the principle of microreversibility is not fulfilled.3
The fact that para substituents have no influence on the rate
of the thermolysis of endoperoxides 2a–d can be best rationa-
lized by radical intermediates 4 (Scheme 1). Zwitterions 3
would be more stabilized by a methoxy group (2b), which is
not the case. On the other hand, radicals are stabilized by
electron donors as well as acceptors,16 and we found clear
evidence for such intermediates during the thermolysis of
endoperoxides.
Interestingly, the ortho substituted 9,10-diarylanthracenes
1e–h exhibit similar remote substituent effects as the para
isomers 1b–d (Table 1). Thus, the electron-withdrawing cyano
group (1g) reduces the rate of the photooxygenation by a
factor of 20 compared to the methoxy derivative 1e. This
remarkable result can again only be rationalized by polar
intermediates 3 and a coplanar alignment of the aryl ring with
the sp2 hybridized carbenium ion, despite the unfavourable
steric interaction with peri hydrogen atoms in the 1 position
(Scheme 1). Indeed, the rotation barrier around the aryl–aryl
bond of anthracene 1f in the ground state was determined to
In conclusion, the cycloreversion is strongly facilitated by
introduction of substituents in ortho position and the kinetic
data point at a rather non-concerted mechanism via radical
intermediates. Although such intermediates were postulated
Table 2 Activation parameters and singlet oxygen yields from the thermolysis of the endoperoxides 2a–h
1
Yield O2 (%)
Peroxide
T1/2b/min
DEA/kJ molꢁ1
logA
DHA/kJ molꢁ1
DSA/J Kꢁ1 molꢁ1
2aa
2b
2c
2d
2e
2f
121 ꢂ 2
88 ꢂ 1
119 ꢂ 2
78 ꢂ 1
21 ꢂ 1
1 ꢂ 0.5
3 ꢂ 1
132 ꢂ 2
127 ꢂ 3
127 ꢂ 6
134 ꢂ 6
117 ꢂ 1
108 ꢂ 2
115 ꢂ 2
117 ꢂ 1
b
14.4 ꢂ 0.3
13.9 ꢂ 0.4
13.8 ꢂ 1.0
15.0 ꢂ 0.9
13.2 ꢂ 0.1
13.4 ꢂ 0.4
13.8 ꢂ 0.4
13.1 ꢂ 0.1
129 ꢂ 2
124 ꢂ 3
124 ꢂ 4
131 ꢂ 2
114 ꢂ 2
105 ꢂ 2
113 ꢂ 1
114 ꢂ 2
+21 ꢂ 4
+12 ꢂ 8
+9 ꢂ 8
+33 ꢂ 4
ꢁ1 ꢂ 5
32
25
34
46
23
28
35
31
+4 ꢂ 1
+14 ꢂ 4
ꢁ2 ꢂ 5
2g
2h
a
23 ꢂ 1
Literature data as reported in ref. 3. At 373 K in xylene.
ꢀc
This journal is The Royal Society of Chemistry 2008
1772 | Chem. Commun., 2008, 1771–1773

View Article Online
This work was generously supported by the Deutsche
Forschungsgemeinschaft (Li 556/9-1).
Notes and references
z CCDC 674791 and 674792. For crystallographic data in CIF or
other electronic format see DOI: 10.1039/b719637g
1 (a) R. G. Harvey, Polycyclic Aromatic Hydrocarbons, Wiley, New
York, 1997; (b) M. Bendikov, F. Wudl and D. F. Perepichka,
Chem. Rev., 2004, 104, 4891.
2 J.-M. Aubry, C. Pierlot, J. Rigaudy and R. Schmidt, Acc. Chem.
Res., 2003, 36, 668.
3 (a) N. J. Turro and M.-F. Chow, J. Am. Chem. Soc., 1979, 101,
1300; (b) N. J. Turro, M.-F. Chow and J. Rigaudy, J. Am. Chem.
Soc., 1981, 103, 7218.
4 (a) R. Schmidt, W. Drews and H.-D. Brauer, J. Photochem., 1982,
18, 365; (b) R. Schmidt, K. Schaffner, W. Trost and H.-D. Brauer,
J. Phys. Chem., 1984, 88, 956.
5 (a) W. Fudickar, A. Fery and T. Linker, J. Am. Chem. Soc., 2005,
127, 9386; (b) W. Fudickar and T. Linker, Chem.–Eur. J., 2006, 12,
9276.
Fig. 1 Thermolyses of the endoperoxides 2a–h at 100 1C.
6 G. R. Martinet, J.-L. Ravant, M. H. Medeiros, J. Cadet and P. D.
Mascio, J. Am. Chem. Soc., 2000, 122, 10212.
7 E. L. Clennan, Tetrahedron, 1991, 47, 1343.
8 (a) M. Bobrowski, A. Liwo, S. Oldziej, D. Jeziorek and T.
Ossowski, J. Am. Chem. Soc., 2000, 122, 8112; (b) A. G. Leach
and K. N. Houk, Chem. Commun., 2002, 1243.
9 A. R. Reddy and M. Bendikov, Chem. Commun., 2006, 1179.
10 D. Zehm, W. Fudickar and T. Linker, Angew. Chem., Int. Ed.,
2007, 40, 7689.
11 Experimental details are included in the electronic supplementary
informationw.
12 H.-D. Becker, V. Langer, J. Sieler and H.-C. Becker, J. Org.
for the thermolysis of endoperoxides previously,3 we found
new sterical remote effects on the thermolysis of anthracene
endoperoxides. Thus, the diarylanthracenes 1a–h represent
interesting mechanistic probes, which allow to distinguish
between two different reaction pathways in the oxidation
and cleavage step.
The reaction of 1O2 with the methyl derivative 1f merits
special attention since, compared to the other anthracenes,
subsequent singlet oxygen addition and thermolysis are fast-
est. It is therefore an interesting candidate for a photochromic
system with tolerable fast write and excellent fast erasing
characteristics.5
Chem., 1992, 57, 1883.
13 (a) U. Svanholm and V. D. Parker, J. Am. Chem. Soc., 1976, 98,
2942; (b) M. S. Workentin, V. D. Parker, T. L. Morkin and D. D.
M. Wayner, J. Phys. Chem. A, 1998, 102, 6503; (c) M. Oyama, M.
Yamanuki, T. Sasaki and S. Okazaki, J. Chem. Soc., Perkin Trans.
2, 2000, 1745.
14 (a) F. Wilkinson, W. P. Helman and A. B. Ross, J. Phys. Chem.
Ref. Data, 1995, 24, 663; (b) J.-M. Aubry, B. Mandard-Cazin, M.
Rougee and R. V. Bensasson, J. Am. Chem. Soc., 1995, 117, 9159.
15 (a) J. Erikson, C. S. Foote and T. L. Parker, J. Am. Chem. Soc.,
1977, 99, 6455; (b) R. L. Clough, J. Am. Chem. Soc., 1980, 102,
5242.
16 (a) J. Fossey, D. Lefort and J. Sorba, Free Radicals in Organic
Chemistry, John Wiley & Sons, Chichester, 1995; (b) D. A. Pratt, J.
S. Wright and K. U. Ingold, J. Am. Chem. Soc., 1999, 121, 4877; (c)
H. Zipse, Top. Curr. Chem., 2006, 263, 163.
17 (a) A. Caspar, S. Altenburger-Combrisson and F. Gobert, Org.
Magn. Reson., 1978, 11, 603; (b) P. J. Marriott and Y.-H. Lai, J.
Chromatogr., A, 1988, 447, 29.
In summary, we have shown that the reactivity of 9,10-
diarylanthracenes towards singlet oxygen and the rate of
cycloreversion of the corresponding endoperoxides is strongly
controlled by stereoelectronic remote substituent effects. The
results support zwitterionic intermediates in a stepwise addi-
tion of singlet oxygen and biradical intermediates for the
thermolysis. This change in the pathways of singlet oxygen
reactions was hitherto unknown and should be of interest for
future mechanistic discussions. Finally, the anthracenes are
easily accessible and mark important examples of photochro-
mic materials, whose performance can be tuned by a targeted
choice of the substitution pattern.
ꢀc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 1771–1773 | 1773