A photoswitchable bis-azo derivative with a high temporal resolution

Article abstract of DOI:10.1039/c4cc05331a

The novel photoswitchable bis-azo derivative reported herein shows a high temporal resolution of 2 × 10⁸ times between the thermal relaxation rates of its two constituting photochromes. Moreover, the slow and fast azo building blocks of this molecular construct can be triggered by using UV and visible light, respectively.

Full text of DOI:10.1039/c4cc05331a

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
A photoswitchable bis-azo derivative with a high
temporal resolution†
Cite this: Chem. Commun., 2014,
50, 11462
Jaume Garcia-Amoros,a Marta Reig,a Alba Cuadrado,a Mario Ortega,a Santi Nonellb
´
and Dolores Velasco*a
Received 10th July 2014,
Accepted 11th August 2014
DOI: 10.1039/c4cc05331a
www.rsc.org/chemcomm
The novel photoswitchable bis-azo derivative reported herein shows blockers,8 optogenetics9 and even for restoring visual function to
a high temporal resolution of 2 ꢀ 108 times between the thermal blind mice.10
relaxation rates of its two constituting photochromes. Moreover, the
Light-addressable materials integrating multiple azo chromo-
slow and fast azo building blocks of this molecular construct can be phores on their covalent skeleton are currently of considerable
triggered by using UV and visible light, respectively.
interest since they allow easy access to applications that would,
otherwise, be difficult to achieve with individual azo derivatives,
Molecular photoswitches use light, an unlimited, ubiquitous, including photoactive dendrons for drug delivery,11 foldamers,12
free, and sustainable source of energy, as a fuel to swap and molecular accordions13 and hinges.14 However, in these
between two or more different states of the molecule with instances, the very same azo compound is used as a repetitive
markedly different properties.1 The photochemical events that template to create or decorate the basic skeleton of the final
occur upon illumination induce structural variations at the material. As a result, the different azo moieties work synergisti-
molecular scale, which manifest in the modulation of the cally, that is, they activate at once upon illumination and recover
macroscopic features of the material, for instance, its shape, their initial trans form within the same timescale in the dark.
ability to absorb light, electronic behaviour and fluorescent Consequently, the final material can only perform one task at a
emission. On this basis, the development of new organic photo- time precluding multifunctionality.
chromes that respond to light in a desired way has become a
demanding challenge in the overall design of photo-responsive bis-azo derivative (1 in Fig. 1) by combining into the very same
smart materials.
molecular entity a 4,40-dialkoxy-substituted azobenzene (S1 or
With this in mind, we have engaged in the design of a new
Among all organic photochromes, azobenzene and its substituted ‘‘slow’’ part in Fig. 1), which requires several hours to undergo its
derivatives are doubtless the most frequently used for light-triggered thermal back-reaction, and a 40-alkoxyazopyridinium salt (F1 or
materials. Azobenzene displays a reversible and clean photo- ‘‘fast’’ part in Fig. 1), which needs several hundreds of micro-
isomerisation process between its trans and cis isomers. More- seconds instead. Such a design principle endows the resulting
over, the cis-to-trans conversion also takes place thermally in the dye with an unprecedented temporal resolution of more than
dark.2 In fact, the incorporation of azo dyes into different matrices eight orders of magnitude between the thermal isomerisation
has led to the preparation of a great variety of photoresponsive rates of its two constituting azo moieties. In addition, and owing
all-organic and hybrid materials with unique properties and
abilities. Potential applications not only include artificial muscle-
like actuators,3,4 molecular scissors5 and photo-oscillators,6,7 but
also lie in the biomedical field, e.g. as photoswitchable ion-channel
a
2
`
`
Grup de Materials Organics, Institut de Nanociencia i Nanotecnologia (IN UB),
´
`
´
`
Departament de Quımica Organica, Universitat de Barcelona, Martı i Franques 1,
E-08028, Barcelona, Spain. E-mail: dvelasco@ub.edu; Fax: +34 93 339 78 78;
Tel: +34 93 403 92 60
b
´
`
´
Grup d’Enginyeria Molecular, Institut Quımic de Sarria, Universitat Ramon Llull,
Via Augusta 390, E-08017, Barcelona, Spain
† Electronic supplementary information (ESI) available: Synthesis of the different
dyes and the experimental part of the kinetic analyses. See DOI: 10.1039/
c4cc05331a
Fig. 1 Chemical structures of azo dyes 1–3.
11462 | Chem. Commun., 2014, 50, 11462--11464
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
to the different lateral substitution of both azo building blocks, p–p* transition of F1. In fact, similar absorption bands peaking
there is a gap of 50 nm between their wavelengths of maximum at 354 and 403 nm are also present in the absorption spectrum
absorption. This feature enables the activation of the two azo of the corresponding model compounds 2 and 3, respectively
units by using UV and green light, respectively. Thus, the design (b and c in Fig. 2). Indeed, the absorption spectrum of 1
principle reported herein paves the way to new multifunctional resembles the one obtained by adding up those of the two
azo dyes for photonic applications where simultaneous switching separated model compounds, evidencing that the flexible ethylene
in two different timescales is required.
spacer keeps both azo dyes close in space but electronically
Bis-azo dye 1 was prepared in an overall yield of 25% through a uncoupled.
convergent synthetic pathway involving two key precursors (4 and
Upon irradiation with UV light (a high pressure Hg lamp,
5 in Fig. S1, ESI†). 4-(2-Bromoethoxy)-40-methoxyazobenzene (4) 500 W, lirrad = 320–390 nm), the steady-state absorption spec-
was obtained by alkylating 4-hydroxy-40-methoxyazobenzene with trum of an ethanol solution of 1 exhibited a decrease in the
2-bromoethanol in THF. Methylation of 4-(40-hydroxyphenylazo)- intensity of the band peaking at 370 nm, owing to the trans-to-
pyridine with Me2SO4 in DMF furnished 4-(40-methoxyphenylazo)- cis photo-isomerisation of S1. Contrarily, the band placed at
pyridine (5). The nucleophilic substitution between 4 and 5 413 nm remained unaltered, indicating thereby the short life of
in MeCN yielded the target bis-azo dye 1. The characterization cis-F1. The thermal cis-to-trans isomerisation kinetics for both
of both the spectroscopic and kinetic behaviour of the two azo photoactive fragments of 1 was studied in ethanol. In all
units of 1 also requires the synthesis of appropriate models. instances, the corresponding relaxation time (t in Table 1)
4,40-Dimethoxyazobenzene (2 in Fig. 1) was prepared according was determined by fitting a monoexponential function to the
to the literature procedure.15 N-Methyl-4-(40-methoxyphenylazo)- corresponding absorbance vs. time trace (b and c in Fig. 3). The
pyridinium iodide (3 in Fig. 1) was obtained by reacting 5 with thermal back-reaction of cis-S1 was monitored by conventional
MeI in MeCN in a yield of 87%.
time-resolved UV-vis spectroscopy (a in Fig. 3). cis-S1 exhibited a
Before irradiation, the absorption spectrum of 1 (a in Fig. 2) very slow thermal isomerisation with a relaxation time of 15 h
in ethanol at 25 1C shows two main absorption bands peaking at 25 1C. On the other hand, the thermal relaxation of cis-F1 was
at 370 and 413 nm (lMax in Table 1) with molar absorption analysed by means of the nanosecond laser flash-photolysis
coefficients of 33.0 and 26.9 mMꢁ1 cmꢁ1, respectively (e in technique (c in Fig. 3) prior to activation of 1 with visible light
Table 1). The former band, which falls within the UV region (a Continuum Surelite I-10 Q-switched Nd-YAG laser, 5 ns pulse
of the electromagnetic spectrum, arises from the symmetry- width, 10 mJ per pulse, lirrad = 532 nm) registering thereby a
allowed p–p* transition of S1. Otherwise, the latter one, which relaxation time of only 239 ms. In this way, bis-azo dye 1 exhibits
extends up to the visible region, can be attributed mainly to the an excellent temporal resolution of 2 ꢀ 108 times between
the thermal isomerisation rates of its two constituting azo
fragments. Model compounds 2 and 3 showed a relaxation
time similar to those recorded for S1 and F1, respectively (see
Table 1), evidencing, once again, the independent role of the
two photoactive platforms of 1.
The variation of the rate constants with temperature allowed
for the determination of the associated enthalpies and entropies
of activation (DHa and DSa in Table 1 respectively), using Eyring
plots (d and e in Fig. 3). The differential thermal activation
parameters registered for the isomerisation process of S1 and F1
indicate clearly that these two azo fragments thermally isomerise
through different intimate mechanisms. The lower enthalpy of
activation (50 vs. 91 kJ molꢁ1, Table 1) and the lesser negative
entropy of activation (ꢁ8 vs. ꢁ31 J Kꢁ1 molꢁ1, Table 1) measured
for the thermal back-reaction of cis-F1 point out to a less
energetic and less organized transition state for the process.
Such a feature evidences the operation of the rotational mecha-
nism. On the other hand, the thermal isomerisation of cis-S1
Fig. 2 Absorption spectra of ethanol solutions (10 mM, 25 1C) of the
bis-azo dye 1 (a) and its model compounds 2 (b) and 3 (c).
proceeds through an inversional pathway, as demonstrated
previously for its model compound 2.16 In fact, and to the best
of our knowledge, bis-azo dye 1 is the first azo compound
isomerisation process of azo dyes 1–3 in ethanol
Table 1 Spectroscopic and kinetic data for the cis-to-trans thermal
reported so far where both isomerisation mechanisms coexist
in the very same molecule.
As mentioned before, for multifunctionality to be achieved,
the target azo dye must not only exhibit a suitable temporal
resolution but also enable an orthogonal activation of its
constituting photochromes under optical control. In order to
lMax
(nm)
e
DHa
DSa
(mMꢁ1 cmꢁ1
)
t298 K
(kJ molꢁ1
)
(J Kꢁ1 molꢁ1
)
1
370, 413
33.0/26.9
15 h
239 ms
15 h
91 ꢂ 1
50 ꢂ 1
92 ꢂ 1
53 ꢂ 1
ꢁ31 ꢂ 4
ꢁ8 ꢂ 4
ꢁ29 ꢂ 3
ꢁ9 ꢂ 3
2
3
354
403
21.8
31.0
752 ms
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 11462--11464 | 11463

View Article Online
ChemComm
Communication
Fig. 3 Changes in the electronic spectrum (a) and first-order kinetic trace at lobs = 370 nm (b) of a 1 cis-to-trans thermally isomerising ethanol solution
at 25 1C in the dark after its photo-isomerisation with UV light (a high pressure Hg lamp, 500 W, lirrad = 320–390 nm). Transient absorption change
(c) photo-induced by laser pulsed irradiation with green light (an Nd-YAG laser, 5 ns pulse width, 10 mJ per pulse, lirrad = 532 nm, lobs = 420 nm) for azo
compound 1 in ethanol at 25 1C. Eyring plot for the thermal back-reaction of cis-S1 (d) and cis-F1 (e) in ethanol. In all instances [1] = 10 mM.
J.G.-A. is grateful for a post-doctoral grant from the Generalitat
de Catalunya (Spain, 2011 BP-A-00270). M. R. acknowledges the
award of an ADR pre-doctoral grant by the Universitat de
Barcelona (Spain, 2012).
Notes and references
1 G. W. Gray, Handbook of Liquid Crystals, Wiley-VCH, Weinheim,
1998.
2 H. Rau, Photochemistry and Photophysics, CRC Press, Boca Raton, Fl,
1990.
3 (a) T. Ikeda and T. Ube, Mater. Today, 2011, 14, 480; (b) A. Priimagi,
A. Shimamura, M. Kondo, T. Hiraoka, S. Kubo, J.-I. Mamiya,
M. Kinoshita, T. Ikeda and A. Shishido, ACS Macro Lett., 2012, 1, 96.
Fig. 4 Absorption spectrum of an ethanol solution (10 mM, 25 1C) of 1 (a)
and absorption change after 5 ns of pulsed illumination at 355 nm (b) and
532 nm (c) respectively.
´
4 (a) J. Garcia-Amoros, H. Finkelmann and D. Velasco, J. Mater. Chem.,
´
´
2011, 21, 1094; (b) J. Garcia-Amoros, M. Martınez, H. Finkelmann
and D. Velasco, Phys. Chem. Chem. Phys., 2014, 16, 8448.
5 M. Zaremba and V. Siksnys, Proc. Natl. Acad. Sci. U. S. A., 2010,
107, 1259.
test the latter feature, the transient absorption spectrum of an
ethanol solution of 1 at 25 1C was recorded 5 ns after the pulse
of a laser operating at 355 nm and 532 nm, respectively (b and c
in Fig. 4). Upon irradiation at 355 nm, the system showed
response mainly from 320 nm to 400 nm, indicating that the S1
fragment is activated. On the other hand, after irradiation at
532 nm, the system modified its optical density from 370 nm to
480 nm, owing to the selective activation of the F1 platform. In
the wake of these results, and even though there is some
spectral overlap between the main absorption bands of the S1
and F1 moieties (see a in Fig. 1), both photochromes can be
photoactivated upon suitable illumination.
In summary, the excellent temporal resolution of more than
eight orders of magnitude together with the possibility to
activate its slow and fast constituting azo photochromes by
using UV and green light, respectively, makes this photoactive
bis-azo dye a highly valuable chromophore for designing new
multifunctional materials for photonic applications where
simultaneous switching in two very different timescales might
be essential.
6 K. M. Lee and T. J. White, Polymers, 2011, 3, 1447.
´
7 (a) J. Garcia-Amoros, S. Nonell and D. Velasco, Chem. Commun.,
´
2011, 47, 4022; (b) J. Garcia-Amoros, S. Nonell and D. Velasco, Chem.
´
Commun., 2012, 48, 3421; (c) J. Garcia-Amoros, M. Reig,
M. C. R. Castro, A. Cuadrado, M. M. M. Raposo and D. Velasco,
Chem. Commun., 2014, 50, 6704.
¨
8 M. Schonberger, A. Damijonaitis, Z. Zhang, D. Nagel and
D. Trauner, ACS Chem. Neurosci., 2014, 5, 514.
¨
9 T. Fehrentz, M. Schonberger and D. Trauner, Angew. Chem., Int. Ed.,
2011, 50, 12156.
10 I. Tochitsky, A. Polosukhina, V. E. Degtyar, N. Gallerani,
C. M. Smith, A. Friedman, R. N. Van Gelder, D. Trauner, D. Kaufer
and R. H. Kramer, Neuron, 2014, 81, 800.
11 A.-M. Caminade, C.-O. Turrin, R. Laurent, A. Ouali and B. Delavaux-
Nicot, Dendrimers: Towards Catalytic, Material and Biomedical Uses,
Wiley-VCH, Weinheim, 2011.
12 (a) Z. Yu and S. Hecht, Angew. Chem., Int. Ed., 2013, 52, 13740;
(b) Z. Yu, S. Weidner, T. Risse and S. Hecht, Chem. Sci., 2013,
4, 4156.
´
13 D. Bleger, T. Liebig, R. Thiermann, M. Maskos, J. P. Rabe and
S. Hecht, Angew. Chem., Int. Ed., 2011, 50, 12559.
14 (a) Y. Norikane and N. Tamaoki, Org. Lett., 2004, 6, 2595;
(b) S. A. Nagamani, Y. Norikane and N. Tamaoki, J. Org. Chem.,
2005, 70, 9304.
´
15 J. Garcia-Amoros, A. Szymczyk and D. Velasco, Phys. Chem. Chem.
Phys., 2009, 11, 4244.
Financial support for this research was obtained from the
´
´
16 J. Garcia-Amoros, M. Martınez, H. Finkelmann and D. Velasco,
´
Ministerio de Ciencia e Innovacion (Spain, CTQ2012-36074).
J. Phys. Chem. B, 2010, 114, 1287.
11464 | Chem. Commun., 2014, 50, 11462--11464
This journal is ©The Royal Society of Chemistry 2014