Photochromism of rotation-hindered furylfulgides influenced by steric modifications

Article abstract of DOI:10.1002/ejoc.201001649

The syntheses of a bicyclic furylfulgide 14 and a (benzofuryl)fulgide 15 with increased steric constraints are described. Their photochromic behaviors were analyzed by means of UV/Vis spectroscopic measurements, X-ray crystallography, and NMR experiments, and the results were compared to those of the furyl(methyl)fulgide 12 and the furyl(isopropyl)fulgide 13. Compounds 13E and 14E exhibit large quantum yields of 0.57 and 0.53 for the coloration reaction (E) → (C) compared with 12E and 15E (0.23 and 0.17). After irradiation with 350 nm light, 13E and 14E are transformed into the closed (C) forms almost quantitatively, whereas 12E and 15E result in a photostationary state with mixtures of the (E), (Z), and (C) forms. The crystal structures obtained for 13E, 14E, and 15E show that the fulgides adopt cyclizable helical (P)-E_α conformations with no significant differences in atomic distances in the hexatriene unit. 2D- and temperature-dependent NMR experiments showed that the enantio- and diastereotopomerization processes were suppressed in a fulgide for the first time. Compound 14E populates only the E _α conformational state. In contrast, 13E and 15E both exist in the cyclizable E_α and the non-cyclizable E_β conformations in solution. Due to the annulated benzene ring, 15E exhibits a higher thermodynamic barrier than 13E, so the "belly roll" process was reduced for 15, but the (E) → (Z) isomerization could not be suppressed. The structural modification of 14 successfully suppressed the (E) → (Z) isomerization as well as the belly roll process. The way in which the isomerization reaction is suppressed by steric hindrance could not be fully elucidated by using these methods. Copyright

Full text of DOI:10.1002/ejoc.201001649

FULL PAPER
DOI: 10.1002/ejoc.201001649
Photochromism of Rotation-Hindered Furylfulgides Influenced by Steric
Modifications
Frank Strübe,[a] Ron Siewertsen,[b] Frank D. Sönnichsen,[c] Falk Renth,[b] Friedrich Temps,[b]
and Jochen Mattay*[a]
Keywords: Aldol reactions / Atropisomerism / Conformation analysis / Photochemistry / Photochromism
The syntheses of a bicyclic furylfulgide 14 and a (benzofuryl)- distances in the hexatriene unit. 2D- and temperature-de-
fulgide 15 with increased steric constraints are described.
Their photochromic behaviors were analyzed by means of
UV/Vis spectroscopic measurements, X-ray crystallography,
and NMR experiments, and the results were compared to
those of the furyl(methyl)fulgide 12 and the furyl(isopropyl)-
fulgide 13. Compounds 13E and 14E exhibit large quantum
yields of 0.57 and 0.53 for the coloration reaction (E) Ǟ (C)
compared with 12E and 15E (0.23 and 0.17). After irradiation
with 350 nm light, 13E and 14E are transformed into the
closed (C) forms almost quantitatively, whereas 12E and 15E
result in a photostationary state with mixtures of the (E), (Z),
and (C) forms. The crystal structures obtained for 13E, 14E,
and 15E show that the fulgides adopt cyclizable helical (P)-
Eα conformations with no significant differences in atomic
pendent NMR experiments showed that the enantio- and
diastereotopomerization processes were suppressed in a fulg-
ide for the first time. Compound 14E populates only the Eα
conformational state. In contrast, 13E and 15E both exist in
the cyclizable Eα and the non-cyclizable Eβ conformations in
solution. Due to the annulated benzene ring, 15E exhibits a
higher thermodynamic barrier than 13E, so the “belly roll”
process was reduced for 15, but the (E) Ǟ (Z) isomerization
could not be suppressed. The structural modification of 14
successfully suppressed the (E) Ǟ (Z) isomerization as well
as the belly roll process. The way in which the isomerization
reaction is suppressed by steric hindrance could not be fully
elucidated by using these methods.
ents at the photochromic center causes a remarkable change
in the electronic state of the fulgides and therefore affects
the absorption spectra of the compounds. Electron-with-
drawing groups lead to a bathochromic shift of the absorp-
tion maximum, whereas electron-donating substituents
cause a hypsochromic shift.[6] Furthermore, heterocyclic
moieties have a large influence on the spectroscopic quality.
The open form of furylfulgides exhibits an absorption
maximum at about 340 nm, whereas electron-rich indolylf-
ulgides show a remarkable bathochromically shifted ab-
sorption maximum at about 400 nm.[7]
Introduction
Fulgides[1] are an important class of photoswitches that
complement diarylethenes[2] and spiropyrans.[3] Upon wave-
length-specific illumination, they undergo reversible color
changes and are thus of great interest for a range of applica-
tions.[4] The first fulgides were synthesized by Stobbe in the
early 20th century.[5] Like diarylethenes, they contain a
hexatriene system with at least one phenylic or heterocyclic
part as the central photochromic unit for the electrocyclic
reaction. There are three different forms of fulgides, two
mainly colorless, open forms, (E) and (Z), and a colored,
closed form, (C). The photochromic behavior is depicted in
Scheme 1. The photochromism can be monitored by UV/
Vis spectroscopic methods, because there is a considerable
change in the absorption spectra. Varyation of the substitu-
[a] Organische Chemie I, Fakultät für Chemie, Universität
Bielefeld
Postfach 100131, 33501 Bielefeld, Germany
Fax: +49-521-106-6417
Scheme 1. Photochromic reaction of furylfulgides with variable
substituents R at the photochromic core unit.
E-mail: mattay@uni-bielefeld.de
[b] Institut für Physikalische Chemie, Christian-Albrechts-
Universität zu Kiel
Ludewig-Meyn-Straße 8, 24118 Kiel, Germany
[c] Otto-Diels-Institut für Organische Chemie, Christian-
Albrechts-Universität zu Kiel
The fulgides of the first generation described by Stobbe
had limited thermal stability, and the photochromic mecha-
nism was intensively investigated only in the 1970s to
1980s.[8] A milestone in the history of fulgides was achieved
by Heller and co-workers in 1981 with the development of a
Otto-Hahn-Platz 4, 24098 Kiel, Germany
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201001649.
Eur. J. Org. Chem. 2011, 1947–1955
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1947

J. Mattay et al.
FULL PAPER
thermally stable fulgide.[9] The replacement of the hydrogen clizable Eα conformation, and the enantiotopomerization
atoms at the cyclohexatriene unit by methyl groups and the [(P)-Eα Ǟ (M)-Eα] and diastereotopomerization (Eα Ǟ Eβ)
introduction of a heterocyclic furyl moiety afforded fulgides processes would be blocked. Furthermore, the steric hin-
that are stable towards both oxidation and hydrogen trans- drance caused by the modification would be large enough
fer. This formed the basis for current applications of ful- to avoid the (E) Ǟ (Z) double bond isomerization. Herein,
gides, where high thermal and photochemical stability are we report the synthesis of new furylfulgides that implement
essential.
this concept. In (benzofuryl)fulgide 15 the formal benzo an-
Nevertheless, the efficiency of the photochromic reaction nulation of the furyl unit should limit its rotation to a mini-
is limited by the photoisomerization of the open forms. The mal degree. To entirely eliminate the belly roll, a fulgide
photochemical equilibrium between the (E) and the (Z) with a bicyclic framework 14 was designed and synthesized.
form is an unfavorable process that competes with the de- Here, the rotation of the furyl ring is restrained by an alkyl
sired ring-closing reaction from the (E) to the (C) form. chain connection between C-10 of the furyl moiety and C-
Therefore, the quantum yields for the coloration reaction 6 of the hexatriene unit.
are low, which is a significant disadvantage for potential
applications. New results derived from femtosecond time-
resolved measurements and DFT calculations show that,
Results and Discussion
besides the (E) Ǟ (Z) isomerization, there are other deacti-
vation processes that reduce the quantum efficiency.[10] In
The furylfulgides 12–14 were synthesized by starting
addition to the mentioned electronic substituent effects,
steric modifications have great influence on the photochro-
mism.[11] In 1988, Yokoyama et al. investigated the effects
of bulky substituents R at the cylohexatriene moiety on the
photochromic reaction of a furylfulgide.[12] The results
showed that the quantum yield for (E) Ǟ (Z) isomerization
becomes smaller, and the quantum yield for the (E) Ǟ (C)
coloration process increases with the bulkiness of the sub-
stituents. In the case of an isopropyl substituent, no (E) Ǟ
(Z) isomerization was observed, and a quantum yield for
the coloration reaction of 0.58 was found. In comparison,
the (E) Ǟ (C) quantum yield of the methyl compound 12
was determined to be only 0.19. Later, the influence of a
tert-butyl substituent was described, which led to a quan-
tum yield for the coloration process of 0.79.[13] For this
reason, sterically hindered fulgides should be even more vi-
able photochromic compounds for high-quality applica-
tions. Furthermore, Yokoyama indicated that, in addition
from commercially available 2,5-dimethylfuran and the ap-
propriate acyl chloride (Scheme 3). The acylated com-
pounds were obtained from a Friedel–Crafts reaction with
AlCl3 as catalyst. To minimize side reactions, the acylation
was carried out at 0 °C under argon. Although it is known
that furans polymerize in the presence of Lewis acids, the
reactions gave moderate yields (49–62%).[15]
Scheme 3. Friedel–Crafts acylation of 2,5-dimethylfuran {1 (R =
Me), 2 (R = iPr), 3 [R = (CH2)3CO2Et]}.
For the synthesis of the bicyclic furan 5, the intermediate
to (E) Ǟ (Z) isomerization, the open (E) forms of furylful- 3 was reduced in a Wolff–Kishner reduction under condi-
gides undergo a diastereotopomerization of the helical tions described by Huang–Minlon in diethylene glycol
structure between the (P)-Eα and (M)-Eβ conformations.[14] (Scheme 4). After formation of a hydrazone with hydrazine,
The conversion between these two states occurs by a rota- nitrogen was released by the action of potassium hydroxide
tion of the furyl subunit, which Yokoyama called a “belly at 200 °C. Simultaneously, the ester was saponified to the
roll” process (Scheme 2).
carboxylic acid[16] 4, which decomposed slowly at room
temperature, but was stable under argon at –20 °C. The cy-
clization process was catalyzed by polyphosphoric acid to
give 5. The Benzofuran compound 7 was synthesized in a
Nenitzescu-type reaction by starting from p-benzoquinone
and enamine ketone 6. The reaction was carried out in gla-
cial acetic acid, and the pure product precipitated from the
solution after a few minutes at room temperature.[17] The
hydroxy group was protected by methylation with methyl
iodide to give 8.
Scheme 2. Possible rotameric isomers of a furylfulgide. Due to the
geometric constitution, (M)-Eβ cannot undergo cyclization,
whereas (P)-Eα is the cyclizable isomer.
The limiting step in the fulgide synthesis is the Stobbe
condensation between the heterocyclic ketones 1, 2, 5, and
It is clear that only molecules in the Eα conformation, 8, and the diethyl isopropylidenesuccinate 9.[18] Scheme 5
wherein C-2 faces C-3, are in the correct geometric form to shows the synthesis of 12 as an example.
undergo the cyclization to the closed (C) isomer. If the rota-
Treatment of 9 with lithium diisopropylamide (LDA) at
tion can be suppressed by structural manipulation of the –78 °C and addition of the heterocyclic compound gave a
fulgide backbone, all molecules will be available in the cy- mixture of isomeric lactones 10. These intermediate prod-
1948
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 1947–1955

Photochromism of Rotation-Hindered Furylfulgides
Scheme 4. (a) Synthesis of bicyclic ketone 5 by reduction and cyclization (DEG = diethylene glycol; PPA = polyphosphoric acid). (b)
Nenitzescu-type reaction to form benzofuran 7 and subsequent methylation to give 8.
of the (Z) isomers are shifted by around 20 nm to longer
wavelengths. The closed (C) isomers have an absorption
maximum near 470–490 nm (Table 1).
Table 1. List of the synthesized fulgides 12–15 and precursors, to-
gether with their UV/Vis spectroscopic data {λmax [nm], (εmax
[Lmol–1 cm–1])}.
Scheme 5. Synthesis of 12 by Stobbe condensation. Saponification
of lactone mixture 10 gave diacid 11 as a mixture of (E)/(Z) iso-
mers, followed by dehydration to give a mixture of 12E and 12Z.
ucts were only detected by analytical methods (TLC, GC–
1
MS, and H NMR spectroscopy). After aqueous workup,
the reaction mixture was filtered through silica gel. The
crude product was dissolved in ethanol and treated with
saturated aqueous KOH solution at 70 °C to give the diacid
11 as a mixture of (E)/(Z) isomers. For anhydride forma-
tion, the crude diacid was dissolved in dichloromethane,
and N,NЈ-dicyclohexylcarbodiimide (DCC) was added. Af-
ter stirring for 24 h at room temperature, the fulgides were
obtained as a mixture of (E) and (Z) isomers. The isomers
were separated by column chromatography with different
mixtures of cyclohexane and ethyl acetate as eluent. A slight
excess of the (E) isomer over the (Z) isomer was obtained
(ratio of 1.2–1.5:1). The two isomers of 12–15 can easily be
1
distinguished by their H NMR spectra. Whereas the (E)
forms show an intense methyl group signal arising from the
isopropylidene group (CH3-3a; δ ≈ 1.3 ppm) in CDCl3, the
corresponding signal of the (Z) isomer is shifted downfield
(δ = 2.1–2.3 ppm). Furthermore, all (Z) forms exhibit lower
[a] The atom numbering of each fulgide is arbitrary. [b] 10–4 m solu-
tion in n-hexane at room temp.
Figure 1 shows the evolution of the absorption spectrum
Rf values than the corresponding (E) isomers by approxi- of 14E upon successive irradiation with 350 nm light. The
mately 10%. absorption band near 350 nm decreases to a minimal
Fulgides 12–15 exhibit the expected photochromic be- amount, while the strong band near 500 nm belonging to
havior of furylfulgides. The absorption maxima (λmax) of the (C) isomer increases. Irradiation with 500 nm light leads
the open (E) isomers show a strong absorption band at back to the initial spectrum of the (E) isomer. A similar
about 330–350 nm in the near-UV region. The λmax values behavior can be observed for 13E. However, solutions of
Eur. J. Org. Chem. 2011, 1947–1955
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1949

J. Mattay et al.
FULL PAPER
12E and 15E exhibit different characteristics upon irradia- Table 2. Isomerization quantum yields of fulgides 12–15.
tion with 350 nm light. The (C)/(Z)/(E) ratios in the photo-
[a]
[a]
[a]
[a]
Compound φEC
φEZ
φCE
φZE
stationary states were analyzed by NMR spectroscopy. A
solution of each (E) isomer in CDCl3 was successively irra-
diated with 350 nm light. The results show that, for 13 and
14, after an irradiation time of 90 min, the fulgides were 15
almost quantitatively transformed into the closed (C) forms.
12
13
14
0.23(335)
0.57(335)
0.53(350)
0.17(330)
0.13(335)
0.00
0.00
0.06(335), 0.10(500) 0.10(335)
[b]
0.12(335), 0.09(500)
0.13(350), 0.07(500)
–
–
[b]
0.15(330)
0.21(330), 0.16(500) 0.11(330)
[a] Irradiation wavelengths [nm] are given as subscripts in parenthe-
ses. [b] Not determined.
As for 13E, the (E) Ǟ (Z) isomerization is suppressed in
14E due to the steric hindrance at C-6. In the cases of 12
and 15, an equilibrium is established between all three iso-
mers in the photostationary states after an irradiation time
of 90 min. For 12, the conformation ratio (C)/(Z)/(E) is
1:0.15:0.15, and for 15 the ratio is 1:0.65:0.83. These results
show that the structural modification in 15 does not sup-
press the (E) Ǟ (Z) isomerization. An explanation for these
different photochemical properties caused by steric or elec-
tronic differences can be derived from femtosecond time-
resolved transient absorption spectroscopy.[10] In particular,
the time-resolved measurements revealed that the photo-in-
duced ring-closure reactions are considerably faster in the
cases of 13 and 14 (τ ≈ 50 fs) compared with 12 and 15 (τ
≈ 110 and 140 fs, respectively). Thus, the molecule-specific
dynamics in the excited electronic states appear to play de-
cisive roles.
side the hexatriene units of the fulgides show no significant
differences. Compared to the distance between the two ring-
closing atoms C(2) and C(3) of 12E [0.3445(4) nm],[14a] only
a small deviation is observed for 13E [0.3426(2) nm], 14E
[0.3559(2) nm], and 15E [0.3663(2) nm]. Therefore, the in-
teratomic distances do not provide an explanation for the
different photochemical properties. The dihedral angles
C(9)–C(5)–C(4)–C(3) are also comparable [148.82(11)° for
13E, 144.83(11)° for 14E, and 146.29(13)° for 15E]. Com-
pound 14E has a strained geometry due to the cycloheptyl
ring, which leads to an extension of the dihedral angle
C(5)–C(6)–C(1)–C(10) to 151.11(11)° compared with
138.14(10)° for 13E and 136.83(13)° for 15E. The complete
crystallographic data for 14 and 15 will be published else-
where.[22]
Table 3. Selected interatomic distances and dihedral angles of X-
ray structures of 13E, 14E, and 15E.
Compound Distance [nm] Dihedral angle [°]
Dihedral angle [°]
C(2)–C(3)
C(9)–C(5)–C(4)–C(3) C(5)–C(6)–C(1)–C(10)
13E
0.3426(2)
0.3559(2)
0.3663(2)
0.3540(2)
148.82(11)
144.83(11)
146.29(13)
152.36(13)
138.14(10)
151.11(11)
136.83(13)
129.36(13)
14E
15E[a]
[a] Two molecules are present in the unit cell, one of which showing
strained angles due to packing effects.
NMR experiments were carried out to assess the confor-
mational properties of the furyl fulgides 13E, 14E, and 15E.
The assignments were made by comparison with previously
assigned furyl fulgides[23] and on the basis of correlations in
two-dimensional gCOSY, HSQC, and HMBC experiments.
Figure 1. Irradiation of 14E with 350 nm light (I
=
1.10ϫ10–6 mol·s–1·L–1; time: 0, 0.25, 0.5, 0.75, 1, 1.5, 2, 2.5, 3, 4,
5, 6, 7.5, 9.5, 11.5, 14.5, 18.5, 23.5, 28.5, 33.5 min).
Specifically for 14E, C-7 and C-9 were distinguished in the
4
The quantum yields for the photoisomerization reactions HMBC experiments by the presence of a J coupling be-
of fulgides 12–15 in n-hexane were determined by using the tween H(Me-3a/b) and C-7, whereas C-4 and C-5 were es-
3
methods described by Uhlmann and Gauglitz[19] or tablished by strong J correlations to H(Me-3a/b) and 16a/
Maafi[20] and are given in Table 2. For the known com- b-H, respectively. The former correlations also distin-
pounds 12 and 13, the results are in good agreement with guished 13a/b-H from 16a/b-H, because the latter in turn
data reported in the literature. Some differences can be shows correlations to to C-11 of the furyl ring. These corre-
found for the value of φCE at 335/365 nm of 12. However, a lations distinguish C-11 from C-2, allowing for the assign-
2
range from 0.00 (φCE) up to 0.12 (φCE) is given in the litera- ments of C(Me-11) and C(Me-2) from J cross peaks. The
ture, which can be rationalized by the low absorption coeffi- 3J cross peaks of H(Me-11) and H(Me-2) then unambigu-
cients of the (C) forms at 365 nm.[19,21] In practice, only an ously identify C-10 and C-1, respectively.
irradiation wavelength around 500 nm appears to be suit-
able for the (C) Ǟ (E) isomerization and for the determi- to detect chemical-exchange processes such as rotational
nation of φCE barriers in the fulgides. At room temperature, the spectrum
NMR experiments at variable temperatures were utilized
.
For fulgides 13E, 14E, and 15E, X-ray crystallographic of 14E is characterized by one set of resonances for all
analyses gave structural information for the solid state methyl groups, indicating the presence of only one rotamer
(Table 3). All (E) isomers adopt the (P)-Eα conformation in (or fast conversion between the conformers). The prochiral
a helical structure in the crystal, and the bond lengths in- methylene ring protons in the cycloheptyl ring, however,
1950
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 1947–1955

Photochromism of Rotation-Hindered Furylfulgides
lead to sharp, well-separated resonances with clear coupling homogeneous magnetic field through shimming). No fur-
patterns that are very similar to those of the clearly distin- ther barrier was detected. For 14E, a kinetic barrier with
guishable prochiral methyl groups in furyl(isopropyl)fulgide a coalescence temperature of about 210 K (Figure 2c) was
13E.
observed. Below this transition, two sets of signals are pres-
Thus, NMR spectra of 14E were acquired at elevated ent in a ratio of 1:2.5. Most noticeable is the resonance
temperatures up to 318 K and compared to those of 13E doubling for 13a-H, with two resonances of similar multi-
(Figure 2). For 13E, it had been shown that the methyl plicities at δ = 4.20 and 3.82 ppm at 178 K. The signal of
groups exchange frequencies at elevated temperatures, with 16a-H also splits, but the upfield-shifted second resonance
a coalescence temperature of approximately 338 K (Fig- is superimposed by those of the methyl groups. For the
ure 2b).[14b,23] In contrast, the spectra of 14E were devoid methyl groups, the doubling is best seen for H(Me-3b) and
of any line broadening, even at the highest achievable tem- H(Me-2) owing to the larger shift difference between the
perature, suggesting the absence of any racemization. This peaks.
was confirmed by NOESY/EXSY experiments performed
at 323 K, which were void of any negative exchange cross- through a three-step process, because the (P)-Eα Ǟ (M)-Eα
peak contributions. enantiotopomerization/helical chirality inversion is pro-
Racemization of fulgides has been suggested to occur
To detect additional rotation barriers, the temperature of hibited by steric overlap.[14a] For 13E, the belly-roll process
the NMR probe was lowered from room temperature to (Eα Ǟ Eβ) converts the helical chirality in concert with the
178 K. In the data for 13E, subtle chemical-shift changes rotation around the C-1–C-6 bond. The enantiotopomeri-
were observed throughout the fulgide spectra (Figure 3a). zation [(P)-Eβ Ǟ (M)-Eβ] has the higher barrier of
All resonance linewidths changed almost imperceptibly and 53 kJmol–1 and is responsible for the coalescence observ-
uniformly, consistent with an increase due only to solvent- able at high temperature (Figure 2b).[14] The barrier for the
viscosity changes (and increasing difficulties in attaining a equivalent process in 14E can only be similar or higher, ow-
Figure 2. (a) 1D 1H NMR spectra of 14E at 298 and 318 K. Neither linewidth nor chemical shift change appreciably for any line,
including the strongly diastereotopic 13a-H or 16a-H (at δ = 4.10 and 2.60 ppm, respectively). (b) 1D 1H NMR spectra of 13E at elevated
1
temperatures from 298 to 338 K. (c) Expanded regions of the 1D H NMR spectra of 14E (200 MHz, CD2Cl2) at variable temperatures.
The vertical scale of the 13a-H and 16a-H protons (left panel) is increased four-fold versus the upfield region (right panel).
Figure 3. (a) Low-temperature 1D 1H NMR spectra of 13E (500 MHz, CD2Cl2). (b) 1D 1H NMR spectra of 15E at variable temperatures
(200 MHz, CD2Cl2).
Eur. J. Org. Chem. 2011, 1947–1955
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1951

J. Mattay et al.
FULL PAPER
ing to larger steric hindrance in 14E. Most importantly, the which is in excellent agreement with the calculated fraction
cyclic analogue 14E does not show this coalescence, indicat- of the Eα conformer of 82% and 18% of the Eβ con-
ing that the Eβ Ǟ Eβ conversion was successfully prevented former.[10] Based on the knowledge of rotational barriers in
(the respective coalescence temperature raised beyond de- the furylfulgide, it is likely that the Eα Ǟ Eβ conversion,
tection), thus indicating that a non-racemizing fulgide was which becomes visible in this furylfulgide analogue, is due
synthesized.
to increased steric hindrance.
Interestingly, a low-temperature transition was observed
for 14E, which, in principle, could originate either from the
belly roll or from an alternative rotational barrier. Energy
calculations of favorable ring conformations of 14E suggest
that the Eβ conformation is 35 kJ/mol less stable then the Eα
conformation. An interconversion between these rotamers
would thus be expected to have a rotational barrier much
higher than 35 kJ/mol, and thus lead to a chemical ex-
change process above room temperature. Furthermore, an
energy difference of 35 kJ/mol results in a population frac-
tion of 100% of the Eα conformer by using Boltzmann’s
distribution, whereas the ratio determined from NMR spec-
troscopic measurements was 1:2.5. It is far more likely in-
stead that the low-temperature interconversion stems from
two favorable, energetically nearly equal conformations of
the cycloheptyl moiety (Figure 4). This agrees well with the
observed chemical shift difference between the respective
rotamer resonances, with the largest shift differences being
observed for 13-H and 16-H, and with typical coalescence
temperatures for ring inversion.[24] Furthermore, the signals
of the isopropylidene methyl groups H(Me-3a/b) shift no-
ticeably, because their location beneath the furyl ring makes
them particularly sensitive to furyl ring rotation owing to
the ring current effect.
Conclusions
The synthesis of furylfulgides with different structural
modifications, which greatly influence the isomerization
processes of the photochromic compounds, has been devel-
oped. The modification at C-6/C-10 of the furyl backbone
strongly affects the belly roll process, or even eliminates it
completely. Whereas benzoannulation (15) reduced the belly
roll process (P)-Eα Ǟ (M)-Eβ, and introduction of an iso-
propyl group (13) reduced the (E) Ǟ (Z) isomerization,
both processes and enantiotopomerization were suppressed
for the bicyclic furylfulgide 14.
Experimental Section
General: All commercially available compounds, including 2,5-di-
methylfuran and LDA (2 m in THF/n-heptane/ethylbenzene) were
purchased from Acros, Alfa Aesar, or Sigma Aldrich and were used
without further purification. Solvents were dried according to stan-
dard procedures. Column chromatography was performed with sil-
ica gel (0.040–0.063 mm, Macherey–Nagel). All NMR spectro-
scopic data were acquired with Bruker Avance 600, DRX 500, or
Avance 200 spectrometers. 1D 1H NMR, 13C NMR, 13C-cpd,
DEPT135, gCOSY, HSQC, and HMBC were used for spectro-
scopic analysis. Samples were dissolved in CDCl3, [D6]DMSO, or
in CD2Cl2 with TMS added as internal standard. Spectra were ref-
erenced to TMS or solvent lines [CDCl3: δ = 7.24 ppm (1H),
77.0 ppm (13C); [D6]DMSO: δ = 2.49 ppm (1H), 39.5 ppm (13C);
CD2Cl2: δ = 5.30 ppm (1H), 53.52 ppm (13C)]. EI mass spectra were
recorded with a VG Autospec X (Micromass Co. UK Ltd.) and
HR mass spectra were performed with a Fourier transform ion
cyclotron resonance (FT-ICR) mass spectrometer APEX III
(Bruker Daltonik GmbH, Bremen, Germany). The quantum yields
of the photoisomerization reactions of fulgides 12–15 in n-hexane
were determined by monitoring the evolution of the absorption
spectra during irradiation of stirred solutions of the (E) isomers
using a 150 W xenon lamp as light source and a monochromator
for wavelength selection (typical spectral width Δλ = 5–10 nm). All
measurements were performed in a quartz cuvette with d = 1 cm
pathlength for irradiation and absorption measurements. The irra-
diation intensities were determined with a power meter (Coherent,
3Σ, PS19Q), and the accuracy of the intensity measurements was
checked by ferrioxalate actinometry.[26] Absorption spectra were
measured with a Shimadzu UV-2401 desktop spectrometer.
Figure 4. Calculated conformers of 14Eα. Ground-state geometry
optimizations were performed at the DFT/B3LYP/6-31+G(d,p)
level by using Gaussian 09.[25] The calculated energy difference is
approx. 6 kJ/mol, which results in a population fraction of approxi-
mately 0.9 of the more stable conformer (right) at room tempera-
ture by using Boltzmann’s distribution. The left conformer matches
the experimental X-ray crystallographic structure (carbon atoms
grey, oxygen atoms dark grey).[22]
Due to the absence of any diastereotopic groups in
(benzofuryl)fulgide 15E, the enantiotopomerization at high
temperatures cannot be detected. As expected, the com-
pound exhibits only one set of resonances at room tempera-
ture. At low temperatures, however, a rotational barrier is
detectable (Figure 3b). In dichloromethane the NMR
analysis at 500 MHz reveals that the coalescence is clearly
observable for 14-H (ca. 200 K) in the benzylic ring and
the furyl methyl group H(Me-2) (ca. 190 K), leading to an
estimate for the rotational energy barrier of 37 kJ/mol. The
corresponding proton chemical shifts for the two rotamers
at low temperatures were established by a 2D-EXSY experi-
General Preparation of 3-Acyl-2,5-dimethylfurans 1–3: The appro-
priate acyl chloride (47 mmol) was added to a suspension of AlCl3
(6.67 g, 50 mmol) in dichloromethane (150 mL) under argon at
0 °C. After stirring for 1 h, 2,5-dimethylfuran (5.00 mL, 47 mmol)
was added, and the solution immediately turned dark-red. After
30 min, the reaction mixture was cautiously hydrolyzed by addition
of aqueous HCl (2 m), and the organic layer was separated. The
aqueous layer was extracted with dichloromethane (3ϫ100 mL),
ment. The ratio between rotamers is approximately 6:1, and the combined organic layers were washed with saturated
1952
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 1947–1955

Photochromism of Rotation-Hindered Furylfulgides
aqueous NaCl solution, dried with MgSO4, and the solvent was
removed in vacuo. The residue was purified by column chromatog-
raphy.
0.10 mol) and stirred at room temperature for 2 h. After the exo-
thermic reaction had ended, the layers were separated, and the
aqueous layer was extracted with diethyl ether (2ϫ30 mL). The
combined organic layers were washed with water and with satu-
rated aqueous NaCl solution, and dried with MgSO4. All volatile
compounds were removed in vacuo, and the product was recrys-
tallized from diethyl ether. Compound 6 was obtained as colorless
needles (10.52 g, 0.09 mol, 90%), which sublimed in high vacuum.
1H NMR (500 MHz, CDCl3, 25 °C): δ = 10.67 (br. s, 1 H, NH),
1-(2,5-Dimethylfuran-3-yl)ethanone (1): Yield: 53%; Rf = 0.36 (cy-
1
clohexane/ethyl acetate, 9:1). H NMR (500 MHz, CDCl3, 25 °C):
δ = 6.13 (s, 1 H, Ar-H), 2.47 (s, 3 H, CH3), 2.30 (s, 3 H, CH3),
2.19 (s, 3 H, CH3) ppm. 13C NMR (126 MHz, CDCl3, 25 °C): δ =
194.2, 156.7, 149.8, 122.0, 106.0, 29.0, 14.2, 13.0 ppm. MS (70 eV):
m/z (%) = 138 (55) [M+], 123 (100) [M – CH3+], 81 (21), 43 (82).
3
4.95 (s, 1 H, CH), 2.89 (d, JH,H = 5.1 Hz, 3 H, NCH3), 1.96 (s, 3
1-(2,5-Dimethylfuran-3-yl)-2-methylpropan-1-one (2): Yield: 56%;
Rf = 0.61 (cyclohexane/ethyl acetate, 9:1). 1H NMR (500 MHz,
H, CH3), 1.88 (s, 3 H, CH3) ppm. 13C NMR (126 MHz, CDCl3,
25 °C): δ = 194.6, 164.0, 95.0, 29.3, 28.6, 18.5 ppm. MS (70 eV):
m/z (%) = 113 (50) [M+], 98 (100) [M – CH3+], 56 (65), 43 (16), 40
(14).
3
CDCl3, 25 °C): δ = 6.16 (s, 1 H, Ar-H), 3.01 (sept, JH,H = 6.6 Hz,
3
1 H, CH), 2.52 (s, 3 H, CH3), 2.22 (s, 3 H, CH3), 1.11 (d, JH,H
=
6.6 Hz, 6 H, CH3) ppm. 13C NMR (126 MHz, CDCl3, 25 °C): δ =
201.0, 157.5, 149.7, 120.5, 105.7, 38.1, 18.6, 14.3, 13.2 ppm. MS
(70 eV): m/z (%) = 166 (18) [M+], 123 (100), 43 (28).
1-(5-Hydroxy-2-methylbenzofuran-3-yl)ethanone (7):[17] To a solu-
tion of p-benzoquinone (2.13 g, 20 mmol) in glacial acetic acid
(80 mL), compound 6 (2.33 g, 21 mmol), dissolved in glacial acetic
acid (30 mL), was added. In an exothermic reaction, a colorless
precipitate formed after a few minutes. After stirring for 2 h, the
precipitate was filtered off, washed with water and a small amount
of diethyl ether, and dried in vacuo. The product 7 (2.19 g,
12 mmol, 58%) was used without further purification but can be
recrystallized from glacial acetic acid. 1H NMR (500 MHz, [D6]-
DMSO, 25 °C): δ = 9.31 (s, 1 H, OH), 7.35 (m, 2 H, Ar-H), 6.73
Ethyl 5-(2,5-Dimethylfuran-3-yl)-5-oxopentanoate (3): Yield: 62%;
Rf = 0.23 (cyclohexane/ethyl acetate, 7:3). 1H NMR (500 MHz,
3
CDCl3, 25 °C): δ = 6.14 (s, 1 H, Ar-H), 4.07 (q, JH,H = 6.9 Hz, 2
3
H, CH2), 2.67 (t, JH,H = 7.2 Hz, 2 H, CH2), 2.47 (s, 3 H, CH3),
3
2.32 (t, JH,H = 7.2 Hz, 2 H, CH2), 2.18 (s, 3 H, CH3), 1.92 (m, 2
3
H, CH2), 1.19 (t, JH,H = 6.9 Hz, 3 H, CH3) ppm. 13C NMR
(126 MHz, CDCl3, 25 °C): δ = 195.7, 173.2, 156.9, 149.8, 121.4,
105.5, 60.2, 39.9, 33.3, 19.1, 14.2, 14.1, 13.1 ppm. MS (70 eV): m/z
(%) = 238 (19) [M+], 193 (23) [M – OC2H5+], 192 (23), 123 (100),
43 (38).
3
4
(dd, JH,H = 8.8, JH,H = 2.5 Hz, 1 H, Ar-H), 2.72 (s, 3 H, CH3),
2.53 (s, 3 H, CH3) ppm. 13C NMR (126 MHz, [D6]DMSO, 25 °C):
δ = 193.8, 163.2, 154.3, 146.9, 126.7, 117.1, 112.9, 111.1, 106.4,
30.8, 15.4 ppm. MS (70 eV): m/z (%) = 190 (51) [M+], 176 (11), 175
(100), 147 (13), 43 (20).
5-(2,5-Dimethylfuran-3-yl)pentanoic Acid (4): Ester
3 (5.80 g,
24 mmol), KOH (6.86 g, 122 mmol), and hydrazine (3.54 mL,
73 mmol, 98% aq. sol.) were dissolved in diethylene glycol
(150 mL) and heated to reflux at 160 °C. After 1 h, the residual
hydrazine and water were distilled off, and the reaction mixture
was heated to 195 °C for 4 h. The reaction mixture was neutralized
with aqueous HCl (2 m) and extracted with ethyl acetate
(4ϫ75 mL). The combined organic layers were washed with satu-
rated aqueous NaCl solution, dried with MgSO4, and the solvent
was removed in vacuo. The product 4 (4.50 g, 23 mmol, 96%) was
used without further purification, but can be purified by column
chromatography. Rf = 0.59 (cyclohexane/ethyl acetate, 7:3). 1H
NMR (500 MHz, CDCl3, 25 °C): δ = 11.08 (br. s, 1 H, COOH),
1-(5-Methoxy-2-methylbenzofuran-3-yl)ethanone (8): Compound 7
(1.20 g, 6 mmol) was dissolved in dimethylformamide (40 mL), and
sodium hydride (0.30 g, 7 mmol, 60% susp.) was added at 0 °C un-
der argon. After stirring for 1 h, methyl iodide (0.47 mL, 7 mmol)
was added, and the reaction mixture was stirred at room tempera-
ture for an additional 2 h. The reaction was quenched by addition
of aqueous HCl (50 mL, 2 m), and the aqueous layer was extracted
with ethyl acetate (3ϫ75 mL). The combined organic layers were
washed with saturated aqueous NaCl solution, dried with MgSO4,
and the solvent was removed in vacuo. The product was purified
by flash column chromatography (cyclohexane/ethyl acetate, 7:3)
and recrystallization from cyclohexane. Compound 8 (1.00 g,
5 mmol, 82%) was obtained as yellow crystals. Rf = 0.64 (cyclohex-
ane/ethyl acetate, 7:3). 1H NMR (500 MHz, CDCl3, 25 °C): δ =
3
5.74 (s, 1 H, Ar-H), 2.34 (t, JH,H = 7.5 Hz, 2 H, CH2), 2.27 (t,
3JH,H = 7.4 Hz, 2 H, CH2), 2.19 (s, 3 H, CH3), 2.14 (s, 3 H, CH3),
1.60–1.66 (m, 2 H, CH2), 1.48–1.54 (m, 2 H, CH2) ppm. 13C NMR
(126 MHz, CDCl3, 25 °C): δ = 180.2, 149.1, 145.1, 118.9, 107.2,
33.9, 29.7, 24.4, 24.2, 13.4, 11.3 ppm. MS (70 eV): m/z (%) = 196
(34) [M+], 123 (12), 110 (37), 109 (100), 95 (18), 43 (82).
4
3
7.46 (d, JH,H = 2.5 Hz, 1 H, Ar-H), 7.31 (d, JH,H = 8.8 Hz, 1 H,
3
4
Ar-H), 6.86 (dd, JH,H = 8.8, JH,H = 2.5 Hz, 1 H, Ar-H), 3.86 (s,
3 H, CH3), 2.75 (s, 3 H, CH3), 2.60 (s, 3 H, CH3) ppm. 13C NMR
(126 MHz, CDCl3, 25 °C): δ = 194.0, 163.3, 156.8, 148.4, 126.8,
117.9, 112.7, 111.2, 104.5, 55.9, 30.9, 15.6 ppm. MS (70 eV): m/z
(%) = 205 (11), 204 (74) [M+], 190 (17), 189.0 (100) [M – CH3+],
174 (9) [M – 2 CH3+], 161 (10), 147 (7), 118 (7), 90 (6), 82 (6), 63
(8).
1,3-Dimethyl-5,6,7,8-tetrahydrocyclohepta[c]furan-4-one (5): Poly-
phosphoric acid (14 g) was warmed to 80 °C, and 4 (4.50 g,
23 mmol) was added. After stirring for 15 min, the reaction mixture
was hydrolyzed with water (120 mL) and extracted with ethyl acet-
ate (4ϫ75 mL). The combined organic layers were washed with
saturated aqueous NaCl solution, dried with MgSO4, and the sol-
vent was removed in vacuo. Purification by column chromatog-
raphy (cyclohexane/ethyl acetate, 9:1) gave the product 5 (3.74 g,
21 mmol, 93%) as a slightly yellow liquid. Rf = 0.31 (cyclohexane/
ethyl acetate, 9:1). 1H NMR (500 MHz, CDCl3, 25 °C): δ = 2.55
(m, 2 H, CH2), 2.51 (m, 2 H, CH2), 2.39 (s, 3 H, CH3), 2.11 (s, 3
H, CH3), 1.71–1.79 (m, 4 H, CH2) ppm. 13C NMR (126 MHz,
CDCl3, 25 °C): δ = 199.5, 155.8, 144.5, 122.5, 117.7, 42.3, 25.5,
22.4, 22.1, 13.6, 10.9 ppm. HRMS: m/z calcd. for C11H14O2 [M+]
178.09938; found 178.09900.
Diethyl 2-Isopropylidenesuccinate (9):[27] Diethyl succinate
(9.62 mL, 57.40 mmol) was added to a solution of potassium tert-
butoxide (6.78 g, 0.06 mmol) in tert-butyl alcohol (75 mL). After
stirring for 1 h, acetone (4.20 mL, 57.40 mmol) was added, and the
reaction mixture was heated to reflux for 20 h. The reaction mix-
ture was acidified with aqueous HCl (2 m) and extracted with di-
ethyl ether (4ϫ70 mL). The combined organic layers were washed
with saturated aqueous NaCl solution, dried with MgSO4, and all
volatile compounds were removed in vacuo. The dark residue was
(Z)-4-Methylaminopent-3-en-2-one (6): Methylamine (10 mL, dissolved in ethanol (140 mL) and acidified with concd. HCl
0.12 mol, 40% aq. sol.) was added to acetylacetone (10.30 mL, (7 mL). After stirring at room temperature for 48 h, the reaction
Eur. J. Org. Chem. 2011, 1947–1955
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1953

J. Mattay et al.
FULL PAPER
mixture was neutralized with saturated aqueous NaHCO3 solution
and extracted with diethyl ether (3ϫ100 mL). The combined or-
ganic layers were washed with water, saturated aqueous NaCl solu-
tion, dried with MgSO4, and all volatile compounds were removed
in vacuo. Distillation at 0.07 mbar gave 9 as a colorless liquid
(2.63 g, 8.94 mmol, 60%). B.p. 67–70 °C (0.07 mbar). 1H NMR
(s, 3 H, Me-3b), 1.87 (s, 3 H, Me-2), 1.34 (s, 3 H, Me-3a), 1.28 (d,
3
3JH,H = 6.9 Hz, 3 H, Me-13a), 0.85 [d, JH,H = 6.9 Hz, 3 H, Me-
13b] ppm. 13C NMR (126 MHz, CDCl3, 25 °C): δ = 163.2 (s, C-9),
163.1 (s, C-7), 157.9 (s, C-6), 154.1 (s, C-3), 150.6 (s, C-11), 147.0
(s, C-2), 120.7 (s, C-4), 120.1 (s, C-5), 119.2 (s, C-1), 105.6 (d, C-
10), 30.8 (d, C-13), 27.1 (q, Me-3a), 22.6 (q, Me-13b), 22.6 (q, Me-
3b), 20.5 (q, Me-13a), 13.3 (q, Me-11), 12.8 (q, Me-2) ppm. MS
(500 MHz, CDCl3, 25 °C): δ = 4.09 (q, 3JH,H = 6.91 Hz, 2 H, CH2),
3
4.05 (q, JH,H = 6.91 Hz, 2 H, CH2), 3.28 (s, 2 H, CH2), 2.06 (s, 3 (70 eV): m/z (%) = 289 (15), 288 (74) [M+], 274 (14), 273 (74) [M –
H, CH3), 1.78 (s, 3 H, CH3), 1.19 (t, JH,H = 6.91 Hz, 3 H, CH3),
CH3+], 245 (41), 217 (15), 201 (27), 199 (26), 173 (17), 128 (16),
3
3
1.16 (t, JH,H = 6.91 Hz, 3 H, CH3) ppm. 13C NMR (126 MHz, 115 (17), 96 (23), 91 (17), 77 (15), 43 (100). HRMS: m/z calcd. for
CDCl3, 25 °C): δ = 171.3, 167.7, 148.7, 120.6, 60.5, 60.1, 35.3, 23.1,
23.1 ppm. MS (70 eV): m/z (%) = 214 (2) [M+], 169 (55), 168 (76),
141 (14), 140 (28), 113 (23), 112 (100), 96 (13), 95 (52), 68 (19), 67
(58), 59 (14), 53 (16).
C17H20O4 [M+] 288.13616; found 288.13440.
Fulgide 13Z:[21] Yield: 5%; Rf = 0.32 (cyclohexane/ethyl acetate,
1
7:3). H NMR (500 MHz, CDCl3, 25 °C): δ = 5.86 (s, 1 H, Ar-H),
3
2.81 (sept, JH,H = 6.9 Hz, 1 H, CH), 2.38 (s, 3 H, CH3), 2.25 (s, 3
General Preparation of Fulgides 12–15:[18]
A
solution of
9
H, CH3), 2.12 (s, 3 H, CH3), 2.03 (s, 3 H, CH3), 1.16 (d, JH,H
=
3
3
(15 mmol) in THF (15 mL) was cooled to –78 °C, and LDA
(7.5 mL, 15 mmol, 2 m in THF/n-heptane/ethylbenzene) was added
under argon. After stirring for 1 h, the appropriate ketone
(10 mmol), dissolved in THF (30 mL), was added by using a sy-
ringe. The reaction mixture was warmed to room temperature over-
night and stirred for an additional 24 h. The progress of the reac-
tion was monitored by TLC. Upon completion, the reaction mix-
ture was acidified with aqueous HCl (2 m), and the aqueous layer
was extracted with ethyl acetate (3ϫ50 mL). The combined or-
ganic layers were washed with saturated aqueous NaCl solution,
dried with MgSO4, and the solvent was removed in vacuo. The
residue was dissolved in cyclohexane/ethyl acetate (7:3), filtered
through silica gel and the solvent removed in vacuo. The residue
was dissolved in ethanol (60 mL), and a saturated aqueous solution
of KOH (5 mL) was added. After stirring at 70 °C for 20 h, the
reaction mixture was poured onto ice and acidified with aqueous
HCl (2 m). The aqueous layer was extracted with ethyl acetate
(3ϫ50 mL), and the combined organic layers were washed with
saturated aqueous NaCl, dried with MgSO4, and the solvent was
removed in vacuo. The dark-brown residue was dissolved in dichlo-
romethane (50 mL), and DCC (4.13 g, 20 mmol) was added. After
stirring for 48 h, the reaction mixture was filtered through silica
gel, and the solvent was removed in vacuo. The products were puri-
fied by column chromatography and recrystallization from appro-
priate solvents.
6.9 Hz, 3 H, CH3), 0.97 (d, JH,H = 6.9 Hz, 3 H, CH3) ppm. 13C
NMR (126 MHz, CDCl3, 25 °C): δ = 163.4, 161.3, 155.7, 153.3,
151.8, 149.8, 121.1, 120.0, 115.2, 106.3, 34.2, 26.6, 22.1, 21.8, 19.1,
13.4, 12.6 ppm. MS (70 eV): m/z (%) = 289 (15), 288 (78) [M+], 273
(76) [M – CH3+], 245 (44), 227 (39), 217 (19), 201 (31), 199 (26),
173 (20), 128 (15), 115 (17), 96 (23), 91 (16), 43 (100).
Fulgide 14E: Yield: 6%; Rf = 0.75 (cyclohexane/ethyl acetate, 7:3).
3
1H NMR (500 MHz, CD2Cl2, 25 °C, TMS): δ = 4.10 (ddd, JH,H
3
2
3
= 13, JH,H = 6, JH,H = 3 Hz, 1 H, 13a-H), 2.60 (ddd, JH,H = 15,
3JH,H = 7, JH,H = 2 Hz, 1 H, 16a-H), 2.35 (m, 1 H, 16b-H), 2.32
2
(s, 3 H, Me-3b-H), 2.20–2.08 (m, 2 H, 13b-H, 14a-H), 2.16 (s, 3 H,
Me-11-H), 1.91–1.82 (m, 1 H, 15a-H), 1.84 (s, 3 H, Me-2-H), 1.70–
1.60 (m, 2 H, 15b-H, 14b-H), 1.35 (s, 3 H, Me-3a-H) ppm. 13C
NMR (125 MHz, CD2Cl2, 25 °C TMS): δ = 163.8 (s, C-9), 163.4
(s, C-7), 153.9 (s, C-3), 152.8 (s, C-6), 146.0 (s, C-11), 145.2 (s, C-
2), 125.3 (s, C-1), 121.1 (s, C-4), 118.5 (s, C-5), 117.6 (s, C-10), 34.0
(t, C-13), 29.2 (t, C-14), 28.0 (t, C-15), 26.7 (q, C-Me-3a), 24.2 (t,
C-16), 22.4 (t, C-Me-3b), 13.3 (q, C-Me-2), 11.0 (q, C-Me-11) ppm.
MS (70 eV): m/z (%) = 301 (20), 300 (100) [M+], 285 (14) [M –
CH3+], 257 (30), 243 (14), 241 (21), 213 (18), 128 (13), 115 (13),
43 (84). HRMS: m/z calcd. for C18H20O4 [M+] 300.13616; found
300.13550.
Fulgide 14Z: Yield: 5%; Rf = 0.65 (cyclohexane/ethyl acetate, 7:3).
1H NMR (500 MHz, CDCl3, 25 °C): δ = 2.55–2.62 (m, 2 H, CH2),
2.38 (s, 3 H, CH3), 2.24 (m, 1 H, CH2), 2.18 (m, 1 H, CH2), 2.16
(s, 3 H, CH3), 2.13 (s, 3 H, CH3), 2.04 (m, 1 H, CH2), 1.94 (s, 3 H,
CH3), 1.90 (m, 1 H, CH2), 1.41–1.55 (m, 2 H, CH2) ppm. 13C NMR
(126 MHz, CDCl3, 25 °C): δ = 163.8, 161.1, 153.4, 152.2, 151.2,
145.0, 121.8, 121.3, 118.2, 117.8, 37.6, 29.8, 28.4, 26.5, 25.2, 22.1,
13.3, 11.2 ppm. MS (70 eV): m/z (%) = 301 (21), 300 (100) [M+],
285 (17) [M – CH3+], 257 (35), 255 (16), 243 (16), 241 (23), 239
(18), 229 (15), 213 (25), 211 (18), 185 (16), 128 (16), 115 (16), 43
(92). HRMS: m/z calcd. for C18H20O4 [M+] 300.13616; found
300.13330.
Fulgide 12E:[9] Yield: 8%; Rf = 0.31 (cyclohexane/ethyl acetate, 9:1).
1H NMR (500 MHz, CDCl3, 25 °C): δ = 5.90 (s, 3 H, Ar-H), 2.55
(s, 3 H, CH3), 2.32 (s, 3 H, CH3), 2.22 (s, 3 H, CH3), 1.97 (s, 3 H,
CH3), 1.33 (s, 3 H, CH3) ppm. 13C NMR (126 MHz, CDCl3,
25 °C): δ = 163.8, 163.3, 153.7, 151.3, 148.3, 146.8, 124.2, 120.9,
119.1, 105.8, 26.8, 22.6, 22.2, 13.9, 13.3 ppm. MS (70 eV): m/z (%)
= 260 (49) [M+], 246 (16), 245 (100) [M – CH3+], 217 (21), 201 (17),
173 (24), 145 (16), 128 (14), 115 (12), 91 (12), 77 (12), 43 (66).
HRMS: m/z calcd. for C15H16O4 [M+] 260.10486; found 260.10260.
Fulgide 12Z:[9] Yield: 6%; Rf = 0.17 (cyclohexane/ethyl acetate, 9:1).
1H NMR (500 MHz, CDCl3, 25 °C): δ = 5.96 (s, 1 H, Ar-H), 2.40
(s, 3 H, CH3), 2.24 (s, 3 H, CH3), 2.18 (s, 3 H, CH3), 2.06 (s, 3 H,
CH3), 1.93 (s, 3 H, CH3) ppm. 13C NMR (126 MHz, CDCl3,
25 °C): δ = 163.7, 160.9, 153.8, 153.1, 150.3, 145.6, 121.8, 120.1,
119.4, 105.6, 26.8, 25.3, 22.3, 13.7, 13.3 ppm. MS (70 eV): m/z (%)
= 260 (45) [M+], 246 (16), 245 (100) [M – CH3+], 217 (23), 201 (21),
199 (16), 173 (29), 145 (19), 129 (15), 128 (17), 115 (15), 91 (14),
77 (15), 43 (79). HRMS: m/z calcd. for C15H16O4 [M+] 260.10486;
found 260.10310.
Fulgide 15E: Yield: 7%; Rf = 0.26 (cyclohexane/ethyl acetate, 9:1).
3
1H NMR (500 MHz, CDCl3, 25 °C, TMS): δ = 7.31 (d, JH,H
=
3
9 Hz, 1 H, 17-H), 6.88 (dd, JH,H = 9.3 Hz, 1 H, 16-H), 6.73 (d,
3JH,H = 3 Hz, 1 H, 14-H), 3.82 (s, 3 H, 19-H), 2.75 (s, 3 H, 13-H),
2.25 (s, 6 H, Me-3b-H, Me-2-H), 1.15 (s, 3 H, Me-3a-H) ppm. 13C
NMR (CDCl3, 125 MHz, 25 °C): δ = 163.6 (s, C-9), 163.0 (s, C-7),
156.2 (s, C-15), 155.5 (s, C-3), 153.5 (s, C-2), 148.8 (s, C-11), 145.0
(s, C-6), 126.5 (s, C-10), 121.5 (s, C-5), 120.8 (s, C-4), 119.6 (s, C-
1), 112.9 (d, C-16), 111.6 (d, C-17), 102.8 (d, C-14), 55.9 (q, C-19),
26.7 (q, C-Me-3a), 22.7 (q, C-Me-3b), 22.0 (q, C-13), 14.0 (q, C-
Me-2) ppm. MS (70 eV): m/z (%) = 327.2 (22), 326.2 (100) [M+],
Fulgide 13E:[21] Yield: 7%; Rf = 0.45 (cyclohexane/ethyl acetate,
1
7:3). H NMR (500 MHz, CDCl3, 25 °C): δ = 5.01 (s, 1 H, 10-H), 311.2 (43), 309.2 (12), 283.2 (20), 281.2 (12), 267.2 (41), 265.2 (15),
3
4.26 (sept, JH,H = 6.9 Hz, 1 H, 13-H), 2.23 (s, 3 H, Me-11), 2.25 253.2 (13), 239.2 (17), 223.2 (11), 204.1 (11), 165.1 (11), 162.1 (21),
1954
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 1947–1955

Photochromism of Rotation-Hindered Furylfulgides
S. Oliver, J. Chem. Soc. Perkin Trans. 1 1981, 197–201; d) A.
Santiago, R. S. Becker, J. Am. Chem. Soc. 1968, 90, 3654–3658;
e) M. Kaftory, Acta Crystallogr., Sect. C 1984, 40, 1015–1019.
[9] P. J. Darcy, H. G. Heller, P. J. Strydom, J. Whittal, J. Chem. Soc.
Perkin Trans. 1 1981, 202–205.
[10] R. Siewertsen, F. Strübe, J. Mattay, F. Renth, F. Temps, Phys.
Chem. Chem. Phys. 2011, 13, 3800–3808.
152.1 (12), 115.1 (13), 106.1 (12), 91.1 (12), 77.1 (10), 44.0 (12),
43.0 (16). HRMS: m/z calcd. for C19H18O5Na+ [M+] 349.10464;
found 349.10480; calcd. for (C19H18O5)2Na+ 675.22026; found
675.22007.
Fulgide 15Z: Yield: 5%; Rf = 0.11 (cyclohexane/ethyl acetate, 9:1).
3
1H NMR (500 MHz, CDCl3, 25 °C): δ = 7.31 (d, JH,H = 8.8 Hz,
[11] a) A. P. Glaze, H. G. Heller, J. Whittall, J. Chem. Soc. Perkin
Trans. 2 1992, 591–594; b) Y. Yokoyama, T. Iwai, N. Kera, I.
Hitomi, Y. Kurita, Chem. Lett. 1990, 263–264.
[12] Y. Yokoyama, T. Goto, T. Inoue, M. Yokoyama, Y. Kurita,
Chem. Lett. 1988, 1049–1052.
[13] J. Kiji, T. Okano, H. Kitamura, Y. Yokoyama, S. Kubota, Y.
Kurita, Bull. Chem. Soc. Jpn. 1995, 68, 616–619.
[14] a) Y. Yokoyama, K. Ogawa, T. Iwai, K. Shimazaki, Y. Kajih-
ara, T. Goto, Y. Yokoyama, Y. Kurita, Bull. Chem. Soc. Jpn.
1996, 69, 1605–1612; b) Y. Yokoyama, T. Iwai, Y. Yokoyama,
Y. Kurita, Chem. Lett. 1994, 225.
1 H, Ar-H), 6.83 (m, 2 H, Ar-H), 3.80 (s, 3 H, CH3), 2.46 (s, 3 H,
CH3), 2.36 (s, 3 H, CH3), 2.28 (s, 3 H, CH3), 2.04 (s, 3 H, CH3)
ppm. 13C NMR (126 MHz, CDCl3, 25 °C): δ = 163.2, 160.9, 157.2,
156.0, 154.3, 148.9, 143.8, 127.6, 122.8, 121.3, 114.9, 111.7, 111.6,
103.0, 66.0, 27.2, 24.0, 22.4, 13.6 ppm. MS (70 eV): m/z (%) = 327
(21), 326 (100) [M+], 311 (37) [M – CH3+], 309 (11), 283 (18), 281
(10), 267 (67), 265 (12), 253 (12), 239 (20), 227 (15), 204 (10), 162
(19), 152 (10), 115 (10). HRMS: m/z calcd. for C19H18O5 [M+]
326.11542; found 326.11390.
Supporting Information (see footnote on the first page of this arti-
cle): Detailed evaluation of the quantum yields and evolution of
1H NMR spectra of 12–15 upon successive irradiation.
[15] T. Reichstein, Helv. Chim. Acta 1930, 13, 356.
[16] A. Thurkauf, X. Chen, S. Zhang, Y. Gao, A. Kieltyka, J. W. F.
Wasley, R. Brodbeck, W. Greenlee, A. Ganguly, H. Zhao, Bi-
oorg. Med. Chem. Lett. 2003, 13, 2921–2924.
[17] T. I. Mukhanova, L. M. Alekseeva, E. F. Kuleshova, Y. N.
Sheinker, V. G. Granik, Pharm. Chem. J. 1993, 27, 136–142.
[18] a) C. J. Thomas, M. A. Wolak, R. R. Birge, W. J. Lees, J. Org.
Chem. 2001, 66, 1914–1918; b) M. L. Bossi, J. B. Rodriguez,
P. F. Aramendia, J. Photochem. Photobiol. A: Chem. 2006, 179,
35–41.
Acknowledgments
This work was supported by the Biophotonics Initiative of the Ger-
man Ministry of Research and Education (BMBF) (grant
13N9234) (F. S., J. M.) and funded by the Deutsche Forschungsge-
meinshaft (SFB 677, “Function by Switching” (R. S., F. R., and
F. T.).
[19] E. Uhlmann, G. Gauglitz, J. Photochem. Photobiol. A: Chem.
1996, 98, 45–49.
[20] M. Maafi, R. G. Brown, Photochem. Photobiol. Sci. 2008, 7,
1360–1372.
[1] a) Y. Yokoyama, Chem. Rev. 2000, 100, 1717–1739; b) Y.
Yokoyama, in Molecular Switches (Ed. B. L. Feringa), Wiley-
VCH, Weinheim, 2001, pp. 107–121.
[21] Y. Yokoyama, T. Inoue, M. Yokoyama, T. Goto, T. Iwai, N.
Kera, I. Hitomi, Y. Kurita, Bull. Chem. Soc. Jpn. 1994, 67,
3297–3303.
[22] F. Strübe, J. Mattay, B. Neumann, H. G. Stammler, Acta Crys-
tallogr., Sect. C 2011, 67, o33–o36.
[23] R. Siewertsen, F. Renth, F. Temps, F. Sönnichsen, Phys. Chem.
Chem. Phys. 2009, 11, 5952–5961.
[2] M. Irie, Chem. Rev. 2000, 100, 1685–1716.
[3] G. Berkovic, V. Krongauz, V. Weiss, Chem. Rev. 2000, 100,
1741–1753.
[4] a) S. Kawata, Y. Kawata, Chem. Rev. 2000, 100, 1777–1788; b)
B. Yao, Y. Wang, N. Menke, M. Lei, Y. Zheng, L. Ren, G.
Chen, Y. Chen, M. Fan, Mol. Cryst. Liq. Cryst. 2005, 430, 211–
219; c) S. Malkmus, F. Koller, S. Draxler, T. E. Schrader, W. J.
Schreier, T. Brust, J. A. DiGirolamo, W. J. Lees, W. Zinth, M.
Braun, Adv. Funct. Mater. 2007, 17, 3657–3662; d) A. S. Dvor-
nikov, E. P. Walker, P. M. Rentzepis, J. Phys. Chem. A 2009,
113, 13633–13644; e) F. M. Raymo, M. Tomasulo, Chem. Soc.
Rev. 2005, 34, 327–336; f) J. Cusido, E. Deniz, F. M. Raymo,
Eur. J. Org. Chem. 2009, 2031–2045.
[24] R. K. Harris, R. A. Spragg, J. Chem. Soc. B 1968, 6, 684–691.
[25] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B.
Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li,
H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Son-
nenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hase-
gawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai,
T. Vreven, J. A. Montgomery Jr., J. E. Peralta, F. Ogliaro, M.
Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Starov-
erov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell,
J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M.
Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Ad-
amo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev,
A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Mar-
tin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador,
J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B.
Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian 09, Re-
vision A.02, Gaussian, Inc., Wallingford, CT, 2009.
[26] C. G. Hatchard, C. A. Parker, Proc. R. Soc. London, Ser. A
1956, 235, 518–536.
[5] a) H. Stobbe, Ber. Dtsch. Chem. Ges. 1905, 38, 3673–3682; b)
H. Stobbe, Justus Liebigs Ann. Chem. 1911, 380, 1–129; c) H.
Stobbe, Ber. Dtsch. Chem. Ges. 1907, 40, 3372–3382.
[6] a) Y. Yokoyama, T. Tanaka, T. Yamane, Y. Kurita, Chem. Lett.
1991, 1125–1128; b) Y. Yokoyama, T. Sagisaka, Y. Mizuno, Y.
Yokoyama, Chem. Lett. 1996, 587–588; c) A. P. Glaze, H. G.
Heller, J. Whittal, J. Chem. Soc. Perkin Trans. 2 1992, 591–594.
[7] a) A. Kaneko, A. Tomoda, M. Ishizuka, H. Suzuki, R. Matsu-
shima, Bull. Chem. Soc. Jpn. 1988, 61, 3569–3573; b) W. Zhao,
Y. Ming, Z. Zhu, M. Fan, J. Photochem. Photobiol. A: Chem.
1992, 63, 235–240; c) L. Yu, Y. Ming, W. Zhao, M. Fan, J.
Photochem. Photobiol. A: Chem. 1992, 68, 309–317.
[27] C. G. Overberger, C. W. Roberts, J. Am. Chem. Soc. 1949, 71,
[8] a) O. Crescente, H. G. Heller, S. Oliver, J. Chem. Soc. Perkin
Trans. 1 1979, 150–153; b) H. G. Heller, S. Oliver, M. Shawe,
J. Chem. Soc. Perkin Trans. 1 1979, 154–157; c) H. G. Heller,
3618–3627.
Received: December 8, 2010
Published Online: February 23, 2011
Eur. J. Org. Chem. 2011, 1947–1955
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1955