Wavelength-dependent reactivity of a quinolinone: Toward a photochromic three-state system

Article abstract of DOI:10.1021/ol801420d

(Chemical Equation Presented) The photochemical reactivity of the quinolinone 3 was investigated using NMR by monitoring its reactions under appropriate irradiation wavelengths. Besides the irreversible formation of degradation products which were structurally identified, the reversible formation of the enol 4 and cyclobutenol 5 was also observed. The enol and cyclobutenol can be switched or reversed back to the quinolinone 3, resulting in a photochromic three-state system in which the relative ratio of the three components largely depends on the irradiation wavelength used.

Full text of DOI:10.1021/ol801420d

ORGANIC
LETTERS
2008
Vol. 10, No. 17
3773-3776
Wavelength-Dependent Reactivity of a
Quinolinone: Toward a Photochromic
Three-State System
Je´roˆme Berthet, Jean-Claude Micheau,† Vladimir Lokshin,‡ Magali Vales,‡
Gaston Vermeersch, and Ste´phanie Delbaere*
CNRS UMR 8009, Lille 2 UniVersity, NMR and Photochemistry team,
F-59006 Lille Cedex, France
stephanie.delbaere@uniV-lille2.fr
Received June 23, 2008
ABSTRACT
The photochemical reactivity of the quinolinone 3 was investigated using NMR by monitoring its reactions under appropriate irradiation
wavelengths. Besides the irreversible formation of degradation products which were structurally identified, the reversible formation of the
enol 4 and cyclobutenol 5 was also observed. The enol and cyclobutenol can be switched or reversed back to the quinolinone 3, resulting in
a photochromic three-state system in which the relative ratio of the three components largely depends on the irradiation wavelength used.
The photochemistry of ortho-alkylphenyl ketones has been
thoroughly explored, and the reaction mechanism is now
well-established.1 γ-Hydrogen abstraction produces a bi-
radical triplet o-xylylenol.2 It decays to isomeric ground-
state o-xylylenols with both Z and E configurations of the
OH group with regards to the ortho-alkyl group.3,4 The
Z-enol is very short-lived, reverting to starting ketone by a
rapid 1,5-sigmatropic H-shift, while the E-isomer is long-
lived enough to undergo a thermal cyclization into cy-
clobutenol derivatives.5,6 Besides the reversibility of the enol
into the ketone, the control by light and/or heat of its
reversible cyclization into a four-membered ring represents
a challenge as it enables us to envisage a photochromic fully
switchable three-species system.7 Quinolinone 1 (Figure 1)
has been recently investigated. The expected enols were
observed, but the formation of undesirable cyclization and
oxidation side products gave rise to a rapid loss of photo-
chromic properties.8
On the other hand, irradiation of the derivative 2 did not
lead to any enol but generated two stable diastereomeric
cyclized structures. These photoisomers returned to the initial
state under visible light irradiation, thus resulting in a two-
† CNRS UMR 5623, IMRCP, University Sabatier, Toulouse, France.
‡ CNRS UPR 3118, CINaM, University Me´diterrane´e, Marseille, France
(1) Sammes, P. G. Tetrahedron 1976, 32, 405–422.
(6) (a) Wagner, P. J.; Subrahmanyam, D.; Park, B.-S. J. Am. Chem.
Soc. 1991, 113, 709–710. (b) Wagner, P. J.; Sobczak, M.; Park, B.-S. J. Am.
(2) Das, P. K.; Scaiano, J. C. J. Photochem. 1980, 12, 85–90.
(3) Haag, R.; Wirz, J.; Wagner, P. J. HelV. Chim. Acta 1977, 60, 2595–
Chem. Soc. 1998, 120, 2488–2489
.
2607
.
(7) (a) Du¨rr, H.; Bouas-Laurent, H. Photochromism: Molecules and
Systems; Elsevier: Amsterdam, 1990. (b) Crano, J. C.; Guglielmetti, R. J.
Organic Photochromic and Thermochromic Compounds; Plenum Press:
New York, 1999. (c) Feringa, B. L. Molecular Switches; Wiley-VCH: New
York, 2001. (d) Irie, M. Photo-ReactiVe Materials for Ultrahigh-Density
Optical Memory; Elsevier: Amsterdam, 1994.
(4) Das, P. K.; Encinas, M. V.; Small, R. D., Jr.; Scaiano, J. C. J. Am.
Chem. Soc. 1979, 101, 6965–6970
.
(5) (a) Scaiano, J. C.; Encinas, M. V.; George, M. V. J. Chem. Soc.,
Perkin Trans. 2 1980, 724–730. (b) Sobczak, M.; Wagner, P. J. Tetrahedron
Lett. 1998, 39, 2523–2526. (c) Koner, A. L.; Singhal, N.; Nau, W. M.;
Moorthy, J. N. J. Org. Chem. 2005, 70, 7439–7442. (d) Konosonoks, A.;
Wright, P. J.; Tsao, M. L.; Pika, J.; Novak, K.; Mandel, S. M.; Bauer,
J. A. K.; Bohne, C.; Gudmundsdottir, A. D. J. Org. Chem. 2005, 70, 2763–
(8) (a) Berthet, J.; Lokshin, V.; Vale`s, M.; Samat, A.; Vermeersch, G.;
Delbaere, S. Tetrahedron Lett. 2005, 46, 6319–6324. (b) Berthet, J.;
Micheau, J.-C.; Vermeersch, G.; Delbaere, S. Tetrahedron Lett. 2006, 47,
2485–2488.
2770
.
10.1021/ol801420d CCC: $40.75
Published on Web 07/26/2008
2008 American Chemical Society

Figure 1. Structures of quinolinones.
state fully photoswitchable photochromic system.9,10 These
results paved the way for the design of multistate photo-
switchable molecules within the quinolinone series.
Quinolinone 3 was therefore synthesized and investigated
by UV-visible spectroscopy (Figure 2). An increase of the
Figure 3. Structure of the various photoproducts of 3.
was unequivocally identified using long-range correlations from
the quaternary C6′ at 145.4 ppm to the dimethyl protons at 1.53
and 1.73 ppm and the aromatic H2′ at 6.94 ppm, evidencing
the Ca-C6′ cyclization. The dimeric entity was proved from a
parallel experiment, carried out using a mixture of 3 and its
difluoro derivative (3′), leading, under the same conditions of
irradiation, to three dimers: two were symmetrical (3-3 and 3′-
3′), and the third one was mixed (3-3′). The dimer 6 evolved
thermally but very slowly to a new structure 9. The reaction
was accelerated upon aeration of the sample. However, although
6 and 9 are considered as side products, it must be noted that
it was possible to photochemically regenerate the dimeric
structure from compound 9 in toluene, which is a hydrogen-
donor solvent. Indeed, the reaction was not observed in benzene,
except when benzhydrol was added.
In the photoreduced alcohol derivative 7, a scalar coupling
between the methyl protons at 0.84 and 1.17 ppm and a
septuplet at 2.52 ppm indicated the presence of the isopropyl
group. In the COSY map, a cross peak exhibited a correlation
between Hb at 6.25 ppm and OH at 6.80 ppm, thus
underlining the reduction of the carbonyl function.
Finally, the rearranged five-membered ring 8 was char-
acterized via the scalar couplings from the proton Ha with
the single methyl group and with the diastereotopic Hc and
Hd. The formation of 8 indicates the involvement of a 1,5
biradical as the result of hydrogen abstraction on the δ carbon
in its alkyl side chain.12,13
Figure 2. Absorption spectra of 3 in acetonitrile before irradiation
(black), after irradiation with 313 nm light (red), after irradiation
with visible light (green), and after reirradiation with 313 nm light
(pink).
colorability has been observed after the following cycle:
irradiation at 313 nm, bleaching with visible light, then
reirradiation at 313 nm. To explore this unexpected result,
structural and kinetic analyses of the photochemistry of 3 in
degassed apolar toluene and degassed polar acetonitrile have
been carried out using NMR spectroscopy.11
Wide-band UV irradiation of 3 in toluene led to the
formation of the enol 4, the dimer 6, and the alcohol 7, while
in acetonitrile, two new photoproducts were additionally
detected: the cyclobutenol 5 and the cyclopentenol 8. All
these compounds were structurally identified and character-
ized by NMR spectroscopy (Figure 3).
The isomer of the enol 4 was proved to be Z (with regards
to CdO) as the high deshielding of OH (δ ) 17 ppm)
indicated the H-bonding between OH and CdO.
For the kinetic analyses, a sample was irradiated, and 1H
NMR spectra were recorded periodically. By measuring the
peak intensities of each photoproduct, the time evolution of
their concentrations can be plotted.
Among all the previously assigned structures, the dimer
6, the alcohol 7, and the cyclopentenol 8 are irreversible
degradation products. As displayed in Figure 4, the monitor-
The four-membered ring in the cyclobutenol 5 was proved
from 1H-13C NMR long-range correlations (HMBC). The two
methyl groups became diastereotopic thus resulting in two
separated lines at 0.54 and 1.25 ppm, coupled through three
bonds with C2 at 163 ppm and with Cb at 83 ppm. Dimer 6
(9) Berthet, J.; Micheau, J.-C.; Lokshin, V.; Vale`s, M.; Samat, A.;
Vermeersch, G.; Delbaere, S. J. Photochem. Photobiol. A, Chem. 2007,
187, 269–274
.
(10) Lokshin, V.; Vales, M.; Samat, A.; Pepe, G.; Metelitsa, A.;
Khodorkovsky, V. Chem. Commun. 2003, 16, 2080–2081
(12) (a) Wagner, P. J. Acc. Chem. Res. 1989, 22, 83–91. (b) Wagner,
.
P. J.; Wang, L. Org. Lett. 2006, 8, 645–647.
(11) Delbaere, S.; Vermeersch, G. J. Photochem. Photobiol. C, Photo-
chem. ReV. 2008, 9, 61–80.
(13) Wagner, P. J.; Zepp, R. G. J. Am. Chem. Soc. 1971, 93, 4958–
4959
.
3774
Org. Lett., Vol. 10, No. 17, 2008

Figure 4. Time evolution of concentrations of 3, 4, 5, and merged
6 + 7 + 8 during wide-band UV irradiation (a) in acetonitrile and
(b) in toluene. Solid lines have been plotted as a visual aid.
ing of the photochemical reaction shows that photodegra-
dation is the main reaction process when the system is
irradiated with wide-band UV light. To reduce this, irradia-
tion has been carried out using monochromatic UV light at
313 and 365 nm.
In toluene, irradiation with 313 nm light results in the
accumulation of the enol 4 up to 50%, while the concentra-
tion in side products does not go over 12% (see Supporting
Information). The cyclobutenol 5 is not detected, nor with
365 nm light, even at low temperature. This indicates that
whatever the irradiation conditions (wide UV band, 313 or
365 nm) the cyclobutenol is not observed in a toluene
solution.
Figure 5. Time evolution of concentrations of enol 4, cyclobutenol
5, and side products (6 + 7 + 8) upon irradiation of the quinolinone
3 in acetonitrile at (a) 313 nm and (c) 365 nm. Part b (405 after
313 nm) and part d (313 after 365 nm) show the effect of a change
in the irradiation wavelength on the kinetics. Data points are the
experimental results. Solid lines are the numerical simulations of
the kinetic model using the list of photochemical processes and
parameters in Table 1.
In acetonitrile, irradiation at 313 nm also favors a notable
formation of the enol 4, while the cyclobutenol 5 remains
the minor product. Results displayed in Figure 5a and Figure
5d show a similar product distribution whatever the initial
conditions, irradiating at 313 nm, either pure quinolinone 3
or the mixture obtained after 365 nm irradiation. In contrast,
irradiation at 365 nm of pure quinolinone 3 leads mainly to
the cyclobutenol 5, while the enol 4 becomes the minor
product (Figure 5c). On the other hand, photobleaching with
405 nm of the mixture obtained after 313 nm irradiation is
shown in Figure 5b. Enol 4 is transformed into the expected
quinolinone 3 but also into the cyclobutenol 5.
Table 1. Apparent Photochemical Rate Constants for Processes
Occurring under Irradiation of 3 in Acetonitrile at Various
Wavelengths
Figure 2a
Figure 2d
Figure 2b
Figure 2c
processes λ ) 313 nm λ ) 313 nm λ ) 405 nm λ ) 365 nm
3f4: h3-4
4f3: h4-3
4f5: h4-5
5f4: h5-4
3fdeg:
2.06E-04
5.05E-04
2.57E-04
15.9E-04
0.76E-04
1.27E-04
0.65E-04
7.20E-04
0.80E-04
13.8E-04
6.40E-04
7.29E-04
3.64E-04
h3-deg
0.25E-04
1.96
0.37E-04
1.95
0.06E-04
2.00
0.17E-04
2.16
h4-3/ h4-5
All these kinetic curves recorded from 1H NMR data under
continuous monochromatic irradiation have been submitted
to kinetic analysis and numerical fitting to estimate the
apparent rate constants of the various photoisomerization and
photodegradation processes. Results are collated in Table 1.
The apparent photochemical rate constants displayed in
Table 1 are proportional to the quantum yield of the
photochemical process and to the molar extinction coefficient
of the photosensitive compound (hij/Φij·εi). Hence, the ratio
of the parameters h4-3 and h4-5 (enol 4 back to quinolinone
3 vs enol 4 to cyclobutenol 5) can be directly related to the
ratio of quantum yields of the formation of 3 and 5 from 4.
In our conditions, this ratio, which does not depend on the
irradiation wavelength, lies around 2 indicating that from
enol 4 the back-reaction to quinolinone 3 is twice as efficient
as the cyclization to cyclobutenol 5.
rise to the enol 4. The latter undergoes a cyclization to
cyclobutenol 5. This reaction is here photochemical as
already observed by Griesbeck et al.14 and absolutely not
thermal as usually reported in the literature for o-alkylphenyl
ketones.5b,6 The only observed thermal reaction is the back-
reversion of enol 4 toward the quinolinone 3 in acetonitrile
(
295k∆ ) 2·10-4 s-1).
After 313 nm irradiation, the ratio of enol 4/cyclobutenol
5 is higher than 1 but lower than 1 after 365 or 405 nm
irradiation. This effect can be correlated with UV spectra
displayed in Figure 2. The red spectrum obtained after
irradiation with 313 nm light exhibits a remarkable absor-
bance at 365 and 405 nm, thus characterizing the enol
absorption bands. It is then expected that the enol 4 would
It is likely that upon irradiation of quinolinone 3 with 313
and 365 nm light intramolecular γ-hydrogen abstraction
occurs, leading to a 1,4 biradical whose rearrangement gives
(14) Griesbeck, A. G.; Stadtmuller, S. Chem. Ber. 1993, 126, 2149–
2150.
Org. Lett., Vol. 10, No. 17, 2008
3775

be easily photobleached under 365 or 405 nm irradiation as
shown in Figure 3b and 3c. This is corroborated by the green
UV spectrum recorded after visible irradiation, displaying
the absorption bands of the cyclobutenol. Finally, reirradia-
tion with 313 nm light transforms the cyclobutenol into enol,
but more efficiently than the transformation of the initial
quinolinone, and hence in complete agreement with the rate
constants calculated and reported in Table 1. The ratio of
about 9 between the parameters h5-4 and h3-4 (enol from
cyclobutenol vs enol from quinolinone) further supports the
greater efficiency of the process from cyclobutenol 4
compared with that from the initial quinolinone 3.
is negligible and it accumulates. As quinolinone 3 does not
absorb 405 nm irradiation light and is therefore not reactive
under these conditions, irradiation of enol 4 leads to a mixture
of 3 and 5. Thermal relaxation also enables the reversion of
the enol 4 to the initial quinolinone 3.
In toluene, this three-state system is rendered into a simpler
photochemical reversion between 3 and 4, i.e., to a two-
state system.
In conclusion, the photochemistry of quinolinone 3 in two
solvents, toluene and acetonitrile, has been investigated.
Upon wide-band UV irradiation, the major part of the
reactivity corresponds to degradation processes with the
formation of photo-oxidized, photorearranged, and photore-
duced products. Under monochromatic irradiation, the pho-
todegradation has been drastically reduced. The relative
distribution of the photoreversible compounds 3, 4, and 5
depends on the irradiation wavelength. After 313 nm
irradiation, the system reaches a situation where quinolinone
3 > enol 4 > cyclobutenol 5. After 365 nm irradiation, the
situation is clearly different with quinolinone 3 ≈ cy-
clobutenol 5 . enol 4. Different again is the situation
observed after 405 nm irradiation: quinolinone 3 > cy-
clobutenol 5 > enol 4. The present results underline a
potential three-state photochromic system. It enables us to
envisage some structural modifications to improve the
resistance to degradation as well as to transform it into a
fully photoswitchable three-state system.
Therefore, we propose the reaction pathways shown in
Figure 6 to suggest the various processes connecting the
Acknowledgment. The 300 MHz NMR facilities were
funded by the Re´gion Nord-Pas de Calais (France), the
Ministe`re de la Jeunesse de l’Education Nationale et de la
Recherche (MJENR), and the Fonds Europe´ens de De´vel-
oppement Re´gional (FEDER). Part of this collaborative work
was realized within the framework GDRI CNRS 93 “Phen-
ics” (Photoswitchable Organic Molecular Systems & De-
vices).
Figure 6. Summary of the various isomerization processes. The
size of the arrows is a visual indication of the relative importance
of the apparent rate constants. Their relative ratio can be slightly
affected by a change in the initial concentrations of 3, 4, and 5.
three-state system. Irradiation in acetonitrile of quinolinone
3 with 313 nm light generates a small amount of cyclobutenol
5, the enol 4 playing the role of a photoreversible intermedi-
ate. The main process is the transformation of the cy-
clobutenol 5 into enol 4. The situation is somewhat different
under 365 nm irradiation where the main process is the back-
isomerization of the enol 4 into the quinolinone 3. Under
these conditions, the photoreactivity of the cyclobutenol 5
Supporting Information Available: Details of irradiation
techniques and NMR experiments, with characterization of
all cited compounds. This material is available free of charge
via the Internet at http://pubs.acs.org.
OL801420D
3776
Org. Lett., Vol. 10, No. 17, 2008