Synthesis and photoisomerization of rhodopsin-based molecular switches

Article abstract of DOI:10.1002/ejoc.201200870

A new family of switches based on a rhodopsin chromophore is presented. These new photoswitches fulfil most of the required features for an efficient switch to be used in practical applications. The synthetic route is simple and versatile and enables substituents capable of acting as anchoring points to be incorporated. Also, substitution can be used to fine tune the switches' properties. The photostationary state can be reached easily by using low-energy light compatible with complex environments. Thermal reset allows these switches to operate through on/off cycles.

Full text of DOI:10.1002/ejoc.201200870

FULL PAPER
DOI: 10.1002/ejoc.201200870
Synthesis and Photoisomerization of Rhodopsin-Based Molecular Switches
Marina Blanco-Lomas,[a] Pedro J. Campos,[a] and Diego Sampedro*[a]
Keywords: Nitrogen heterocycles / Molecular devices / Photochemistry / Chromophores / Isomerization
A new family of switches based on a rhodopsin chromophore
is presented. These new photoswitches fulfil most of the re-
quired features for an efficient switch to be used in practical
applications. The synthetic route is simple and versatile and
enables substituents capable of acting as anchoring points to
be incorporated. Also, substitution can be used to fine tune
the switches’ properties. The photostationary state can be
reached easily by using low-energy light compatible with
complex environments. Thermal reset allows these switches
to operate through on/off cycles.
The discovery of new or alternative prototypes could ex-
pand the applicability of the switch concept to different and
increasingly complex molecular environments.
Introduction
The use of light to induce mechanical movement at the
molecular level has received considerable attention in recent
years.[1] Molecular switches based on E/Z photoisomeriza-
tion have been used in various contexts to control diverse
properties.[2] Among the different photochemical switches
developed, systems based on azobenzene have been used
extensively. For instance, azobenzenes have been employed
as photoswitches in biological applications,[3] for photo-
modification of polymers[4] and to induce photomechanical
movement in liquid crystals.[5] Azobenzene derivatives ab-
sorb light well and photoisomerize efficiently. Also, the
isomerization process induces a shape change that is pri-
marily responsible for the photoinduced transformations
triggered by this photochemical reaction. Lastly, an enor-
mous range of azobenzene derivatives is synthetically ac-
cessible and a detailed knowledge of the relationship be-
tween substitution and properties is available. A very dif-
ferent structure and different properties can be obtained by
the use of chiral diarylidenes.[6] In these compounds the chi-
rality of the molecule allows for a preferential direction of
isomerization. Thus, these compounds can act as molecular
rotors or motors. In spite of their unique properties as effec-
tive switches, azobenzene derivatives or chiral diarylidenes
are just examples of a whole family of compounds capable
of performing controlled E/Z isomerization. Consequently,
the preparation and characterization of new switches that
differ from these structures in properties such as size, wave-
length of absorption, reactivity, polarity and isomerization
mechanism represent an attractive research target.
Inspiration from nature can provide useful insights in the
search of new photoswitches. For instance, the retinal pro-
tonated Schiff base (PSB) chromophore of rhodopsin[7]
constitutes an example of a very efficient E/Z switch shaped
by biological evolution. In bovine rhodopsin (Rh), selective
photoisomerization of the 11-cis chromophore occurs
rapidly (150 fs) through evolution of a single πǞπ* excited
state (S1) providing the all-trans ground state (S0) product
with high efficiency (Φ = 67%).[7] These properties make
Rh an excellent reference for the design and synthesis of
artificial Rh-mimetic molecules. In fact, several families of
Rh-based photoswitches have already been synthesized and
their properties as photoswitches have been reported (Fig-
ure 1).
[a] Departamento de Química, Centro de Investigación en Síntesis
Química (CISQ), Universidad de La Rioja,
Madre de Dios, 51, 26006 Logroño, Spain
Fax: +34-941-299621
Figure 1. Rhodopsin-based molecular photoswitches.
E-mail: diego.sampedro@unirioja.es
First of all, a family of benzylidene-pyrroline derivatives
was synthesized and analyzed by means of CASSCF//
CASPT2 theoretical calculations.[8] Some promising results,
Homepage: http://cisq.unirioja.es/grufor.php
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201200870.
6328
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 6328–6334

Rhodopsin-Based Molecular Switches
such as photostability and photostationary state (PSS) de- substituent in that position could affect the photoisomer-
pendent on the irradiation wavelength, were overshadowed ization process. Also, a phenyl ring could substantially in-
by the low isomerization quantum yield reported (ca. 10–3). crease the UV-absorption and induce a bathochromic shift.
On the other hand, the synthetic route was very limited After retrosynthetic analysis, we decided to close the pyrrol-
and only three derivatives could be synthesized. Trying to ine ring in the last step. Previously, the central double bond
improve these results, a new generation of Rh-based molec- could be formed by a Wittig reaction between an aldehyde
ular photoswitches was developed, namely indanylidene- and a conveniently synthesized phosphonate derivative.
pyrrolines.[9] These switches feature a higher isomerization This would allow us to independently modify the substitu-
quantum yield (≈ 0.3), increased polarity with respect to ents on both sides of the switch. Also, combinations of dif-
azobenzene and a controlled and fast photoisomerization ferent aldehydes and phosphonates would easily yield a
path.[10] Further modification of the substructure allowed complete family of switches. We based our synthetic pro-
the preparation of a dipole-moment switch.[11] However, the cedure on a related approach that has been previously re-
synthetic route used was complex and hardly adaptable for ported.[15]
switches with different substitution groups. Finally, a dif-
ferent family of Rh-based switches was developed. Fluoren-
ylidene-pyrrolines can be easily synthesized with different
substituents within the structure[12] starting from fluor-
enone imines.[13] These compounds feature even higher
isomerization quantum yields (ca. 0.6) and can be operated
Figure 2. Retrosynthesis of retinal-based photoswitches.
by using visible light.[14] Unfortunately, it is difficult to vary
the composition of the PSS owing to the similar stability
The route (Scheme 1) requires the phosphonates to be
and UV absorption of the E and Z isomers. Despite the synthesized first. Phosphonate 2 with the desired R1 group
promising features of the aforementioned Rh-based photo- may be synthesized from corresponding 2-bromoketone 1.
switches and the detailed information accumulated by both Next, installation of an ethyl azide side-chain onto 2 by
theoretical calculations and experimental data in solution using 2-iodoethylazide by means of a phase-transfer pro-
for isolated molecules, little is known about the practical cedure[15] with tetrabutylammonium bisulfate and aqueous
ability of these systems to act as efficient switches in real sodium hydroxide in CH2Cl2 at reflux afforded azido phos-
applications. In most cases, the reason behind this lack of phonates 3 in 30–40% yield. Subsequent Horner–Wads-
practical application is the limited available structures that worth–Emmons reaction[15] between the azido-phos-
can be incorporated into a complex system.
phonate and the aldehyde with the corresponding R2, again
Any new attempt to build a prototype of an efficient by using a phase-transfer procedure with aqueous K2CO3
photoswitch should feature (i) an expedient and versatile and tetrabutylammonium bisulfate, led to the formation of
synthetic route; (ii) anchoring points adequate for linkage; azido enone in 70% yield. Although a 9:1 mixture of E and
(iii) high isomerization quantum yields and (iv) low-energy Z isomers of azido enones 4 was obtained, they could be
absorption bands compatible with complex environments. separated by chromatography on silica gel (hexanes/EtOAc,
As explained above, the three precedent Rh-based molecu- 10:1). However, only the E isomer was isolated as final
lar switches (Figure 1) were synthesized through different product 5 in 80% yield after carrying out the intramolecu-
synthetic routes but in all cases some experimental prob- lar aza-Wittig reaction with triphenylphosphane. It should
lems occurred and one or several of the aforementioned be noted that both isolated E and Z isomers of 4 yield the
requirements were not met. Thus, we developed a new and E isomer of 5 under the same reaction conditions. Thus, the
versatile entry route to a new family of molecular Rh-based mixture of isomers of 4 obtained in the reaction crude can
photoswitches that could improve some of the flaws of the be used in the last step without further purification. Moni-
previous compounds. It should be noted that the com- toring the course of the reaction with 1H NMR spec-
pounds reported here bear some structural differences with troscopy showed that both Z and E isomers of 5 were
the 11-cis-retinal. Specifically, these products lack the cis formed before consumption of the starting material but
configuration in the imine bond and present substituents at isomerization took place under the reaction conditions
the imine carbon. The compound was modified to include a leading to E as the only isolable isomer. Thus, switches 5a–
pyrroline ring which would avoid competitive isomerization o were obtained as pure E forms in all cases.
pathways as discussed previously.[8] This change would also
By using the methodology described above, a complete
help increase the quantum yield of previous benzylidene- library of molecular switches were obtained by changing
pyrrolines switches.[8]
R1 and R2 within the structure (see Table 1). It should be
highlighted that both alkyl and benzyl aldehydes with dif-
ferent electron withdrawing and donor groups may be used
in the synthesis, which constitutes a major advantage over
benzylidene- and indanylidenepyrrolines.
Results and Discussion
These molecular switches based on retinal are analogous
Hetero- and polyaromatic aldehydes could also be used
to benzylidene-pyrrolines, presenting a substituent R1 at the to extend the range of molecular switches showing this
C-5 position of pyrroline (Figure 2). We anticipated that a structure. We also checked the use of ketones with the aim
Eur. J. Org. Chem. 2012, 6328–6334
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
6329

M. Blanco-Lomas, P. J. Campos, D. Sampedro
FULL PAPER
uents R1 and R2. Also, the spectroscopic data correlates
well with the E isomer of similar compounds.[15] In order
to confirm the E configuration of final compounds 5, 2-D
NMR spectroscopy experiments were carried out (see Sup-
porting Information). From the NOESY spectrum of com-
pound 5d (Figure 3), it can be seen that there is an NOE
signal arising from the proton at the vynilic C-1Ј carbon
and the methyl at the C-5 of the pyrroline ring, and also
between the proton at the aromatic C-3Ј carbon and the
two protons at the C-3 of the pyrroline ring. Therefore, an
E configuration for this isomer could be assigned.
Scheme 1. Synthetic route to afford the retinal-based photo-
switches.
Figure 3. NOESY spectrum of the E isomer of switch 5d.
Further stereochemical information could be obtained
from the X-ray diffraction data of compound 5k (Figure 4).
This switch shows an E configuration of the central C=C
double bond in agreement with the NOESY experiment.
Also, X-ray diffraction data reveals that in the solid state
the central double bond is almost planar (torsion angle of
approximately 6°) whereas both phenyl groups are twisted
(approximately 35°) to avoid steric repulsion. Furthermore,
the o-Br substituent is placed opposite to the pyrroline ring,
as the NOESY spectrum showed in the case of the methoxy
substituent in 5d.
Table 1. Molecular switches synthesized.
R1
R2
Absorption[a]
Ratio at PSS[b]
E [%] Z [%]
Me
Me
Me
Me
Me
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
p-MeC6H4
p-MeOC6H4 5c 296 (26600)
o-MeOC6H4 5d 275 (8900)
CH3(CH2)4
Ph
5a 288 (8700)
5b 287 (7300)
56
43
47
30
85
76
72
75
44
40
41
30
80
60
76
44
57
53
70
15
24
28
25
56
60
59
70
20
40
24
5e 232 (15000)
5f 289 (22200)
5g 296 (24800)
p-MeC6H4
p-MeOC6H4 5h 304 (24900)
o-MeOC6H4 5i[c] 281 (15200); 317 (14200)
p-BrC6H4
o-BrC6H4
p-O2NC6H4
4-pyridyl
5j 282 (47700)
5k 282 (18200)
5l 335 (13700)
5m 284 (19300)
2-naphthyl
5n[c] 265 (25300); 311 (14200)
5o 294 (20100)
p-MePh p-MeC6H4
[a] λmax [nm] (ε [(m–1 cm–1]). [b] By using a Hg lamp after 1–3 h of
irradiation. [c] Two different absorption bands were observed.
of obtaining compounds where the number of torsional de-
grees of freedom of the switch backbone is decreased, espe-
cially the potential competitive rotation around the allylic/
benzylic single bonds. However, this idea was not possible
because it was impossible to carry out the Horner–Wads-
worth–Emmons reaction. However, selecting the proper R1
and R2 groups it would be possible to obtain a switch with
Figure 4. X-ray structure of the E isomer of switch 5k.
After the chemical characterization of the different com-
two anchoring points for linkage to complex structures in pounds was performed, their potential as molecular
order to explore the practical applications of PSB-retinal- switches was tested. A good molecular switch must be able
based molecular switches. Owing to the versatility of the to reversibly interconvert between two different states by
procedure, diverse substituents could be incorporated means of an external stimulus.[1a] In this particular case,
within the substructure to enable coupling to surfaces, the two states are the E and Z isomers. Therefore, these
nanoparticles or polymers. For example, the compound compounds need to isomerize from the E form to the Z
could be linked to a selected peptide to enable photocontrol form by irradiation and return from the PSS. First of all,
of its conformation and properties, as recently reported.[16] the UV/Vis spectra for all the compounds synthesized were
The E isomer of switches 5 is expected to be more stable recorded. In all cases, spectra featuring similar absorptions
than the Z isomer owing to steric repulsion between substit- were obtained. The values for the maximum wavelengths
6330
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 6328–6334

Rhodopsin-Based Molecular Switches
and the extinction coefficients of the bands found for each
Figure 5 also shows the UV-spectra for both E and Z
compound in acetonitrile are shown in Table 1. It should isomers of a typical switch, by using 5o as an example. The
be noted that the maximum of the absorption can be modi- Z isomer was obtained by separation from the PSS mixture.
fied by the substituents present within the structure. This In all cases, both isomers show only slight differences in the
could be useful in applications in which the wavelength used absorption bands; however, these differences could be used
to carry out the isomerization process has some restrictions. to tune the switch behavior under irradiation by using light
Thus, by modifying the substituents present, the photo- at an appropriate wavelength. In the two spectra shown in
physical properties of the switch could be fine-tuned. A fur- part b of Figure 5, the E isomer absorbs preferentially at
ther modification of the photophysical properties could be 295 nm whereas the Z isomer features stronger absorption
made by quaternization of the nitrogen atom as previously bands at 260 and 330 nm. Thus, varying the light wave-
shown.[8,12] By doing so, the photochemical and photophys- length from 260 to 295 nm the main isomer in the reaction
ical properties of the chromophore can be changed drasti- mixture would change. Unfortunately, thermal back-reac-
cally. These modifications are currently under study and tion with the current set-up prevented us from confirming
will be reported in due course.
the effect of the wavelength in the PSS. Instead, mixtures
A typical UV spectrum for switches 5a–o is shown in of different ratios of Z/E isomers were found when different
Figure 5. The UV spectrum remains almost the same when wavelengths were used and the PSS could not be reached
carried out in polar and non-polar solvents. Thus, photo- after several hours of irradiation. This suggests the influ-
physical properties, isomerization kinetics and PSS compo- ence of the wavelength in the PSS, as proven for a related
sition would be quite similar in both types of solvent. This compound that was cross-linked to a peptide and thermally
fact could allow applications of these switches in both polar more stable.[16] This fact has also some interesting implica-
and non-polar media, which provides a very wide field of tions in which the average position of the switch could be
action.
modified by adjusting only the wavelength of irradiation.
Then, a solution of the E isomer of each compound
(10–2 m) in acetonitrile was irradiated with a 125-W me-
dium-pressure Hg lamp and Pyrex filter (except for com-
pound 5e, in which quartz was used instead of Pyrex) until
the PSS was reached. Isomerization was followed by 1H
NMR spectroscopy because the E and Z isomers have dis-
tinctive signals. Depending on the absorption bands, the
mixture took 1–3 h to reach the PSS. The results of the
isomer ratio at the PSS for every compound are shown in
Table 1 and show some trends. When changing the R1
group from a methyl group (i.e. 5a) to a phenyl group (i.e.
5f) and R2 remains the same, the percentage of the Z isomer
obtained at the PSS decreases. This could be due only to
steric reasons. On the other hand, when the R1 and R2
groups are the same, but R2 is ortho (i.e. 5i) instead of para
(i.e. 5h), the percentage of isomer Z increases. Finally, main-
taining R1 but modifying R2 from H (i.e. 5f) to an electron-
withdrawing group in the para position, such as nitro (i.e.
5l), the percentage of the Z isomer increases significantly. As
explained previously, the different absorption of the isomers
under the reaction conditions given could explain this fact.
Moreover, the same methodology could be employed to
introduce the heterocycle 5m with little effect on both the
absorption properties and the PSS. However, the addition
of another aromatic ring 5n increases the percentage of Z
isomer in the PSS and also displaces the absorption-band
maximum. Finally, we checked if having an alkyl group as
R2 in 5e leads to a similar isomerization process as experi-
enced by the rest of compounds and the results confirmed
that an E/Z isomerization occurs despite the presence of
multiple single bonds capable of rotation. Therefore, we
conclude that the value of isomers ratio at the PSS can be
modified depending on the structure substitution, which is
a great advantage over fluorenylidene-pyrroline switches.[12]
We also checked the thermal stability for this family of
switches. In every case, the PSS is composed of a mixture
Figure 5. (a) UV/Vis spectrum of a 10–5 m solution of 5f in CH3CN
(black line) and CHCl3 (orange line). (b) UV/Vis of a 10–5 m solu-
tion of the E isomer (blue line) and Z isomer (pink line) of 5o in
CH3CN.
Eur. J. Org. Chem. 2012, 6328–6334
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
6331

M. Blanco-Lomas, P. J. Campos, D. Sampedro
FULL PAPER
of E/Z isomers and this rate is stable for some hours at
room temp. Also, the equilibrium between the E and Z iso-
mers can be displaced from the PSS after heating at 50 °C
for several hours, eventually reverting to the pure E form.
Thus, the main isomer can be easily selected by choosing
between light and heat. This could be important in practical
applications in which on/off switching cycles are necessary.
This thermal back-reaction towards the thermodynamically
more stable E isomer was studied. The rate of this thermal
reaction depends on the substituents R1 and R2 of the mo-
lecule. A few examples are shown in Table 2. The results
Experimental Section
General Details: All commercially available materials were used
without further purification. NMR spectra were recorded by using
a 300 or 400 MHz NMR Bruker spectrometer. UV/Vis spectra were
recorded by a Hewlett–Packard HP 8451A Diode Array UV/Vis
spectrophotometer. Quartz cuvettes (1.0 cm path length) were used
for the measurements.
Synthesis of Phosphonates 2: When phosphonate 2 was not com-
mercially available with the required R1 group, it was synthesized
by treatment of the corresponding 2-bromo ketone 1 with triethyl
phosphite at 160 °C for 3 h.[17]
1
were obtained after following by H NMR spectrum of the
Synthesis of Azido Phosphonates 3: To a mixture of phosphonate
2 (1 equiv.) and 2-iodoethyl azide (2.5 equiv.) in CH2Cl2 (5 mL) a
solution of tetrabutylammonium hydrogen sulfate (1 equiv.) in so-
dium hydroxide (2 equiv., 2 m) was added, and the resulting mixture
was heated to reflux for 36 h.[15] Later, it was cooled and treated
with water (50 mL) and CH2Cl2 (50 mL). The organic layer was
separated and concentrated under reduced pressure. The resulting
residue was dissolved in Et2O (100 mL) and the tetrabutylammo-
nium iodide precipitated. The salt was filtered off, and the filtrate
was dried (Na2SO4) and concentrated under reduced pressure to
give a colorless oil. The resulting product was purified by flash
thermal back-reaction to recover the E isomer of several
compounds.
Table 2. Thermal stability.
R1
R2
Compound
E isomer recovery[a]
Me
Me
Me
Ph
5a
5b
5d
12
60
156
p-MeC6H4
o-MeOC6H4
[a] Time in hours, in the dark at 25 °C.
From the data in Table 2, we infer that the compound chromatography on silica gel (1:1 hexanes/EtOAc) to afford 3 with
the required R1 group in 30–40% yield.
that reverts more readily to the E isomer is 5a, in which the
phenyl group in R2 does not have any substitution. But
Synthesis of Azido Enones 4: K2CO3 (28 equiv.), H2O (10 mL), tet-
when the phenyl group in R2 has a substituent, such as a
p-methyl (5b) or an o-methoxy (5d), the recovery of E is
rabutylammonium hydrogen sulfate (0.2 equiv.) and aldehyde with
the required R2 group (1.5 equiv.) were added to azido phos-
slower, taking a few days. The data in Table 2 shows that phonate 3 (1 equiv.) and the resulting mixture was stirred for 1
week at room temperature.[15] The reaction was followed by TLC
the more electron donating the substituent, the slower the
(5:1 hexanes/EtOAc) until azido phosphonate 3 was consumed.
return to the thermodynamically stable isomer (E). This
Then, the solvent was removed and water (50 mL) was added be-
change is a result of a higher energy barrier for the thermal
transition state.
fore extracting with CH2Cl2 (3ϫ 50 mL). The organic layers were
combined, dried (Na2SO4), and concentrated under reduced pres-
sure. The residual oil was purified by chromatography on silica gel
(5:1 hexanes/EtOAc) to give azido enone 4 as a 9:1 mixture of E
Conclusions
and Z isomers in 70% yield. The two isomers could be separated
by chromatography on silica gel (10:1 hexanes/EtOAc); however, it
The applicability of molecular switches to photocontrol
was possible to carry out the next step by using the 9:1 mixture of
complex systems requires the availability of a diverse range
E and Z isomers.
of compounds with different properties. The choice between
General Procedure for the Synthesis of Switches 5a–o: To a solution
of 4 (1 equiv.) in dry diethyl ether (20 mL) was added Ph3P
(3 equiv.). The resulting mixture was stirred for 24 h under nitrogen
at room temperature and concentrated under reduced pressure. The
resulting solid was purified by chromatography on silica gel (1:4
hexanes/EtOAc) to obtain the E isomer of (5a–o) in 60–85% yield.
different families of compounds could allow a good match
to be found for each practical application by taking into
account the photophysical properties and experimental per-
formance of the right switch. Fine-tuning of switch proper-
ties can only be carried out once a complete set of com-
pounds with different substitution patterns is synthesized.
Also, the synthetic route should be simple and versatile
enough to allow modifications and incorporation into com-
plex systems. In this paper we have shown how the rhodop-
sin chromophore can be used as inspiration for the design
and synthesis of a new family of molecular photoswitches
that fulfill most of the required features. The synthetic route
is expedient and versatile, and allows for the presence of
several substituents capable of acting as anchoring points
suitable for linkage. The PSS can be reached easily after
irradiation of the low-energy absorption bands compatible
with complex environments. Finally, the switch can be easily
reset after heating. Further efforts will be necessary to com-
pletely describe the photophysical properties and to test
these compounds in practical applications.
(E)-4-Benzylidene-5-methyl-3,4-dihydro-2H-pyrrole (5a): Brown oil,
1
yield 85%. H NMR (300 MHz, CDCl3): δ = 7.48 (d, J = 7.4 Hz,
2 H), 7.40 (t, J = 7.6 Hz, 2 H), 7.29 (t, J = 13.0 Hz, 1 H), 6.73 (s,
1 H), 4.02 (m, 2 H), 2.89 (m, 2 H), 2.24 (s, 3 H) ppm. 13C NMR
(300 MHz, CDCl3): δ = 173.06, 142.99, 137.05, 128.88, 128.74,
127.90, 124.90, 58.97, 29.81, 16.31 ppm. UV/Vis (CH3CN): λ [nm]
277 (ε
= = =
7640 m–1 cm–1), 288 (ε 868 m–1 cm–1), 303 (ε
585 m–1 cm–1). HRMS: calcd. for C12H13N [M + H]+ 172.1121;
found 172.1122.
5-Methyl-4-[(E)-4-methylbenzylidene]-3,4-dihydro-2H-pyrrole (5b):
1
Brown oil, yield 82%. H NMR (300 MHz, CDCl3): δ = 7.38 (d, J
= 8.1 Hz, 2 H), 7.21 (d, J = 8.0 Hz, 2 H), 6.71 (s, 1 H), 4.02 (m, 2
H), 2.88 (m, 2 H), 2.38 (s, 3 H), 2.23 (s, 3 H) ppm. 13C NMR
(300 MHz, CDCl3): δ = 173.30, 142.24, 138.15, 134.44, 129.66,
129.01, 124.99, 59.16, 29.29, 21.59, 16.49 ppm. UV/Vis (CH3CN):
6332
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 6328–6334

Rhodopsin-Based Molecular Switches
λ [nm] 287 (ε = 7345 m–1 cm–1), 300 (ε = 5715 m–1 cm–1), 305 (ε =
4790 m–1 cm–1). HRMS: calcd. for C13H15N [M + H]+ 186.1277;
found 186.1279.
H), 4.08 (m, 2 H), 3.61 (s, 3 H), 2.90 (m, 2 H) ppm. 13C NMR
(300 MHz, CDCl3): δ = 174.45, 156.98, 141.42, 134.23, 129.24,
128.91, 128.52, 128.01, 127.93, 125.59, 121.74, 119.92, 110.50,
58.95, 54.72, 30.77 ppm. UV/Vis (CH3CN): λ [nm] 237 (ε =
13840 m–1 cm–1), 281 (ε = 15225 m–1 cm–1), 294 (ε = 13840 m–1 cm–1),
317 (ε = 14189 m–1 cm–1). HRMS: calcd. for C18H17NO [M + H]+
264.1383; found 264.1394.
(E)-4-(4-Methoxybenzylidene)-5-methyl-3,4-dihydro-2H-pyrrole
1
(5c): Brown oil, yield 80%. H NMR (300 MHz, CDCl3): δ = 7.40
(d, J = 8.8 Hz, 2 H), 6.91 (d, J = 8.8 Hz, 2 H), 6.65 (s, 1 H), 3.98
(m, 2 H), 3.80 (s, 3 H), 2.80 (m, 2 H), 2.20 (s, 3 H) ppm. 13C NMR
(300 MHz, CDCl3): δ = 174.39, 159.50, 139.24, 130.45, 129.17,
126.35, 114.14, 57.39, 55.30, 29.65, 15.61 ppm. UV/Vis (CH3CN):
λ [nm] 296 (ε = 26636 m–1 cm–1). HRMS: calcd. for C13H15NO [M
+ H]+ 202.1226; found 202.1235.
4-[(E)-4-Bromobenzylidene]-5-phenyl-3,4-dihydro-2H-pyrrole
(5j):
White solid, yield 74%; m.p. 120–122 °C. 1H NMR (300 MHz,
CDCl3): δ = 7.63 (m, 2 H), 7.46 (m, 5 H), 7.26 (d, J = 8.6 Hz, 2
H), 6.74 (s, 1 H), 4.20 (m, 2 H), 2.99 (m, 2 H) ppm. 13C NMR
(300 MHz, CDCl3): δ = 174.74, 142.80, 135.94, 134.33, 131.74,
130.34, 129.71, 128.73, 128.53, 126.47, 121.80, 59.75, 30.82 ppm.
UV/Vis (CH3CN): λ [nm] 282 (ε = 47716 m–1 cm–1), 295 (ε =
43147 m–1 cm–1), 311 (ε = 23350 m–1 cm–1). HRMS: calcd. for
C17H14BrN [M + H]+ 312.0382; found 312.0385.
(E)-4-(2-Methoxybenzylidene)-5-methyl-3,4-dihydro-2H-pyrrole
1
(5d): Brown oil, yield 80%. H NMR (300 MHz, CDCl3): δ = 7.50
(m, 1 H), 7.28 (m, 1 H), 7.11 (s, 1 H), 6.96 (m, 2 H), 3.98 (m, 2
H), 3.88 (s, 3 H), 2.85 (m, 2 H), 2.25 (s, 3 H) ppm. 13C NMR
(300 MHz, CDCl3): δ = 173.52, 157.50, 143.02, 131.40, 129.26,
128.19, 120.59, 118.91, 110.74, 58.84, 55.64, 29.76, 16.50 ppm. UV/
4-[(E)-2-Bromobenzylidene]-5-phenyl-3,4-dihydro-2H-pyrrole (5k):
White solid, yield 77%; m.p. 106–108 °C. 1H NMR (300 MHz,
CDCl3): δ = 7.71 (m, 2 H), 7.57(t, J = 9.2 Hz, 2 H), 7.47 (m, 3 H),
7.33 (t, J = 7.3 Hz, 1 H), 7.14 (m, 2 H), 4.17 (m, 2 H), 2.95 (m, 2
H) ppm. 13CNMR (300 MHz, CDCl3): δ = 174.28, 143.75, 136.76,
134.10, 132.95, 129.75, 129.07, 128.78, 128.58, 128.44, 127.26,
126.40, 124.99, 59.28, 30.37 ppm. UV/Vis (CH3CN): λ [nm] 235 (ε
Vis (CH3CN):
λ [nm] 275 (ε = =
8938 m–1 cm–1), 291 (ε
6818 m–1 cm–1), 312 (ε
=
4713 m–1 cm–1). HRMS: calcd. for
C13H15NO [M + H]+ 202.1226; found 202.1228.
(E)-4-Hexylidene-5-methyl-3,4-dihydro-2H-pyrrole (5e): Brown oil,
yield 75%. 1H NMR (300 MHz, CDCl3): δ = 5.80 (s, 1 H), 3.90
(m, 2 H), 2.51 (m, 2 H), 2.14 (m, 2 H), 2.09 (s, 3 H), 1.47 (m, 2 H),
1.35 (m, 4 H), 0.93 (m, 3 H) ppm. 13C NMR (300 MHz, CDCl3): δ
= 171.93, 143.31, 126.62, 57.58, 31.71, 30.62, 28.78, 26.89, 22.68,
15.94, 14.18 ppm. UV/Vis (CH3CN): λ [nm] 232 (ε = 15000 m–1 cm–
1). HRMS: calcd. for C11H19N [M + H]+ 166.1590; found
166.1591.
=
15015 m–1 cm–1), 240 (ε
=
14114 m–1 cm–1), 282 (ε
=
18182 m–1 cm–1), 292 (ε = 17117 m–1 cm–1). HRMS: calcd. for
C17H14BrN [M + H]+ 312.0382; found 312.0388.
4-[(E)-4-Nitrobenzylidene]-5-phenyl-3,4-dihydro-2H-pyrrole
(5l):
Pale orange solid, yield 76%; m.p. 119–121 °C. 1H NMR
(300 MHz, CDCl3): δ = 8.23 (d, J = 8.8 Hz, 2 H), 7.66 (m, 2 H),
7.57 (d, J = 8.9 Hz, 2 H), 7.50 (m, 3 H), 6.90 (s, 1 H), 4.30 (m, 2
H), 3.10 (m, 2 H); 13CNMR (300 MHz, CDCl3) ppm. 13CNMR
(300 MHz, CDCl3): δ = 174.47, 146.61, 146.19, 143.64, 133.90,
129.96, 129.34, 128.67, 125.51, 125.43, 123.93, 123.87, 60.03,
31.30 ppm. UV/Vis (CH3CN): λ [nm] 233 (ε = 10150 m–1 cm–1), 262
(ε = 7026 m–1 cm–1), 335 (ε = 13684 m–1 cm–1). HRMS: calcd. for
C17H14N2O2 [M + H]+ 279.1128; found 279.1133.
(E)-4-Benzylidene-5-phenyl-3,4-dihydro-2H-pyrrole (5f): White so-
lid, yield 82%; m.p. 103–105 °C. 1H NMR (300 MHz, CDCl3): δ =
7.65 (m, 2 H), 7.40 (m, 7 H), 7.26 (m, 1 H), 6.83 (s, 1 H), 4.21 (m,
2 H), 3.06 (m, 2 H) ppm. 13C NMR (300 MHz, CDCl3): δ = 174.94,
142.12, 137.02, 134.55, 129.63, 128.92, 128.79, 128.60, 128.49,
127.92, 127.76, 59.68, 31.00 ppm. UV/Vis (CH3CN): λ [nm] 231 (ε
=
10059 m–1 cm–1), 248 (ε
=
10806 m–1 cm–1), 289 (ε
=
22232 m–1 cm–1), 303 (ε = 15717 m–1 cm–1). HRMS: calcd. for
C17H15N [M + H]+ 234.1277; found 234.1284.
4-[(E)-(2-Phenyl-4,5-dihydro-3H-pyrrol-3-ylidene)methyl]pyridine
(5m): Brown solid, yield 65 %; m.p. 99–101 °C. 1H NMR
(300 MHz, CDCl3): δ = 8.64 (m, 2 H), 7.65 (m, 2 H), 7.51 (m, 3
H), 7.30 (m, 2 H), 6.77 (s, 1 H), 4.29 (m, 2 H), 3.11 (m, 2 H) ppm.
13C NMR (300 MHz, CDCl3): δ = 174.48, 150.19, 146.69, 144.43,
133.93, 130.02, 128.76, 128.72, 125.10, 123.03, 59.98, 31.25 ppm.
UV/Vis (CH3CN): λ [nm] 245 (ε = 10028 m–1 cm–1), 275 (ε =
18384 m–1 cm–1), 284 (ε = 19304 m–1 cm–1), 300 (ε = 12368 m–1 cm–
1). HRMS: calcd. for C16H14N2 [M + H]+ 235.1230; found
235.1232.
4-[(E)-4-Methylbenzylidene]-5-phenyl-3,4-dihydro-2H-pyrrole (5g):
Orange solid, yield 70%; m.p. 64–66 °C. 1H NMR (300 MHz,
CDCl3): δ = 7.64 (m, 2 H), 7.45 (m, 3 H), 7.31 (d, J = 8.2 Hz, 2
H), 7.17 (d, J = 8.1 Hz, 2 H), 6.80 (s, 1 H), 4.19 (m, 2 H), 3.03 (m,
2 H), 2.35 (s, 3 H) ppm. 13C NMR (300 MHz, CDCl3): δ = 175.06,
141.17, 137.96, 134.63, 134.19, 129.56, 129.34, 128.90, 128.78,
128.45, 127.74, 59.62, 30.97, 21.36 ppm. UV/Vis (CH3CN): λ [nm]
283 (ε = 21154 m–1 cm–1), 296 (ε = 24808 m–1 cm–1), 309 (ε =
1923 m–1 cm–1). HRMS: calcd. for C18H17N [M + H]+ 248.1434;
found 248.1430.
4-[(E)-Naphthalen-1-ylmethylene]-5-phenyl-3,4-dihydro-2H-pyrrole
(5n): Yellow solid, yield 60%; m.p. 153–155 °C. 1H NMR
(300 MHz, CDCl3): δ = 7.88 (s, 1 H), 7.83 (m, 3 H), 7.68 (m, 2 H),
7.56 (m, 1 H), 7.49 (m, 5 H), 6.98 (s, 1 H), 4.26 (m, 2 H), 3.18 (m, 2
H) ppm. 13C NMR (300 MHz, CDCl3): δ = 175.16, 142.56, 134.61,
134.59, 133.40, 132.81, 129.64, 128.82, 128.51, 128.41, 128.23,
128.15, 127.88, 127.67, 126.52, 126.47, 126.46, 59.79, 31.12 ppm.
UV/Vis (CH3CN): λ [nm] 245 (ε = 26342 m–1 cm–1), 266 (ε =
34448 m–1 cm–1), 275 (ε = 34853 m–1 cm–1), 311 (ε = 28368 m–1 cm–1),
324 (ε = 24316 m–1 cm–1). HRMS: calcd. for C21H17N [M + H]+
284.1434; found 284.1439.
4-[(E)-4-Methoxybenzylidene]-5-phenyl-3,4-dihydro-2H-pyrrole (5h):
1
Brown oil, yield 83%. H NMR (300 MHz, CDCl3): δ = 7.62 (m,
2 H), 7.42 (m, 3 H), 7.32 (d, J = 8.7 Hz, 2 H), 6.87 (d, J = 8.8 Hz,
2 H), 6.76 (s, 1 H), 4.16 (m, 2 H), 3.75 (s, 3 H), 2.97 (m, 2 H) ppm.
13C NMR (300 MHz, CDCl3): δ = 174.87, 159.12, 139.66, 134.43,
130.19, 129.42, 129.35, 128.58, 128.25, 127.33, 127.24, 59.28, 55.07,
30.63 ppm. UV/Vis (CH3CN): λ [nm] 226 (ε = 12467 m–1 cm–1), 249
(ε
= = =
10345 m–1 cm–1), 304 (ε 24934 m–1 cm–1), 317 (ε
20689 m–1 cm–1). HRMS: calcd. for C18H17NO [M + H]+ 264.1383;
found 264.1389.
4-[(E)-2-Methoxybenzylidene]-5-phenyl-3,4-dihydro-2H-pyrrole (5i):
4-[(E)-4-Methylbenzylidene]-5-p-tolyl-3,4-dihydro-2H-pyrrole
5o): Orange solid, yield 80%; m.p. 83–85 °C. H NMR (400 MHz,
(E-
1
White solid, yield 85%; m.p. 111–113 °C. 1H NMR (300 MHz,
CDCl3): δ = 7.69 (m, 2 H), 7.46 (d, J = 7.7 Hz, 1 H), 7.38 (m, 3 CDCl3): δ = 7.55 (d, J = 8.0 Hz, 2 H), 7.32 (d, J = 8.1 Hz, 2 H),
H), 7.21 (m, 2 H), 6.94 (t, J = 7.5 Hz, 1 H), 6.77 (d, J = 8.3 Hz, 1 7.27 (d, J = 7.9 Hz, 2 H), 7.19 (d, J = 8.0 Hz, 2 H), 6.82 (s, 1 H),
Eur. J. Org. Chem. 2012, 6328–6334
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
6333

M. Blanco-Lomas, P. J. Campos, D. Sampedro
FULL PAPER
4.19 (m, 2 H), 3.05 (m, 2 H), 2.42 (s, 3 H), 2.36 (s, 3 H) ppm. 13C
NMR (400 MHz, CDCl3): δ = 174.91, 141.27, 139.56, 137.88,
134.29, 131.69, 129.32, 129.14, 128.89, 128.74, 127.68, 59.44, 31.01,
mendía, J. Photochem. Photobiol. C: Photochem. Rev. 2011, 12,
154–166; c) K.-E. Drexler, Nanosystems: Molecular Machinery,
Manufacturing and Computation, John Wiley & Sons, New
York, 1992; d) J.-P. Sauvage, Molecular Machines and Motors,
vol. 99, Springer, Heidelberg, Germany, 2001.
21.47, 21.35 ppm. UV/Vis (CH3CN):
λ
[nm] 254 (ε
=
12623 m–1 cm–1), 294 (ε = 20118 m–1 cm–1), 309 (ε = 15385 m–1 cm–1).
[2] a) R. Cacciapaglia, S. D. Stefano, L. Mandolini, J. Am. Chem.
Soc. 2003, 125, 2224–2227; b) B. Jousselme, P. Blanchard, N.
Gallego-Planas, J. Delaunay, M. Allain, P. Richomme, E. Levil-
lain, J. Roncali, J. Am. Chem. Soc. 2003, 125, 2888–2889; c) S.
Shinkai, T. Minami, Y. Kusano, O. Manabe, J. Am. Chem. Soc.
1983, 105, 1851–1856.
HRMS: calcd. for C19H19N [M + H]+ 262.1590; found 262.1598.
4-[(Z)-4-Methylbenzylidene]-5-p-tolyl-3,4-dihydro-2H-pyrrole
(Z-
5o): Orange solid, yield 15%. 1H NMR (400 MHz, CDCl3): δ =
7.13 (d, J = 8.0 Hz, 2 H), 6.92 (s, 1 H), 6.84 (d, J = 7.8 Hz, 2 H),
6.72 (q, J = 8.2 Hz, 4 H), 4.07 (m, 2 H), 2.96 (m, 2 H), 2.24 (s, 3
H), 2.19 (s, 3 H) ppm. 13C NMR (400 MHz, CDCl3): δ = 172.94,
140.04, 138.80, 136.58, 133.29, 132.68, 129.31, 128.29, 128.25,
127.95, 126.17, 57.71, 35.88, 21.44, 21.23 ppm. UV/Vis (CH3CN):
λ [nm] 260 (ε = 14600 m–1 cm–1), 295 (ε = 13400 m–1 cm–1). HRMS:
calcd. for C19H19N [M + H]+ 262.1590; found 262.1598.
[3] A. A. Beharry, G. A. Woolley, Chem. Soc. Rev. 2011, 40, 4422–
4437.
[4] A. Natansohn, P. Rochon, Chem. Rev. 2002, 102, 4139–4176.
[5] Y. Yu, M. Nakano, T. Ikeda, Nature 2003, 425, 145–145.
[6] a) B. L. Feringa, Acc. Chem. Res. 2001, 34, 504–513; b) B. L.
Feringa, J. Org. Chem. 2007, 72, 6635–6652.
[7] R. A. Mathies, J. Lugtenburg, in: Handbook of Biological Phys-
ics, vol. 3 (Eds.: D. G. Stavenga, W. J. de Grip, E. N. Pugh),
Elsevier, Amsterdam, The Netherlands, 2000, pp. 56–90.
[8] D. Sampedro, A. Migani, A. Pepi, E. Busi, R. Basosi, L. Lat-
terini, F. Elisei, S. Fusi, F. Ponticelli, V. Zanirato, M. Olivucci,
J. Am. Chem. Soc. 2004, 126, 9349–9359.
[9] F. Lumento, V. Zanirato, S. Fusi, E. Busi, L. Latterini, F. Elisei,
A. Sinicropi, T. Andruniów, N. Ferré, R. Basosi, M. Olivucci,
Angew. Chem. 2007, 119, 418–424; Angew. Chem. Int. Ed. 2007,
46, 414–420.
Typical Procedure for Irradiation: A solution of the E isomer of
each compound 5a–o (10–2 m) in CH3CN was prepared and irradi-
ated at room temperature through Pyrex glass with a 125-W me-
dium-pressure Hg lamp (except for compound 5e, where quartz was
used instead of Pyrex) until the PSS was reached. The isomeriza-
1
tion reaction was followed by H NMR because E and Z isomers
have distinctive 1H NMR signals. Depending on the absorption
bands, the mixture took 1–4 h to reach the PSS.
[10] A. Sinicropi, E. Martin, M. Ryasantsev, J. Helbing, J. Briand,
D. Sharma, J. Léonard, S. Haacke, A. Canizzo, M. Chergui, V.
Zanirato, S. Fusi, F. Santoro, R. Basosi, N. Ferré, M. Olivucci,
Proc. Natl. Acad. Sci. USA 2008, 105, 17642–17947.
[11] a) A. Melloni, R. Rossi Paccani, D. Donati, V. Zanirato, A.
Sinicropi, M. L. Parisi, E. Martin, M. Ryazantsev, W. J. Ding,
L. M. Frutos, R. Basosi, S. Fusi, L. Latterini, N. Ferré, M.
Olivucci, J. Am. Chem. Soc. 2010, 132, 9310–9319; b) R.
Rossi Paccani, D. Donati, S. Fusi, L. Latterini, G. Farina, V.
Zanirato, M. Olivucci, J. Org. Chem. 2012, 77, 1738–1748.
[12] L. Rivado-Casas, D. Sampedro, P. J. Campos, S. Fusi, V. Zanir-
ato, M. Olivucci, J. Org. Chem. 2009, 74, 4666–4674.
[13] L. Rivado-Casas, P. J. Campos, D. Sampedro, Organometallics
2010, 29, 3117–3124.
[14] L. Rivado-Casas, M. Blanco-Lomas, P. J. Campos, D. Sampe-
dro, Tetrahedron 2011, 67, 7570–7574.
[15] B. B. Snider, J. Zhou, J. Org. Chem. 2005, 70, 1087–1088.
[16] M. Blanco-Lomas, S. Samanta, P. J. Campos, G. A. Woolley,
D. Sampedro, J. Am. Chem. Soc. 2012, 134, 6960–6963.
[17] W. Nagata, T. Wakabayashi, Y. Hayase, Org. Synth. 1973, 53,
44–47.
Supporting Information (see footnote on the first page of this arti-
cle): Copies of the 1H NMR and 13C NMR spectra for compounds
5, 2D-NMR experiments of the E isomer of switch 5d.
CCDC-889062 contains the supplementary crystallographic data
for this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
Acknowledgments
Financial support from the Ministerio de Ciencia e Innovación
(MICINN) (CTQ2011-24800) is gratefully acknowledged. M. B.-L.
thanks the Spanish Ministerio de Educación y Ciencia (MEC) for
her grant. The authors thank Fernando Rodriguez for his assist-
ance in solving the X-ray diffraction structure.
[1] a) V. Balzani, A. Credi, M. Venturi, Molecular Devices and Ma-
chines: Concepts and Perspectives for the Nanoworld, Wiley-
VCH, Weinheim, Germany, 2008; b) M. L. Bossi, P. F. Ara-
Received: July 2, 2012
Published Online: September 28, 2012
6334
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 6328–6334