Reporting the release of caged species by a combination of two sequential photoreactions, a molecular switch, and one color of light

Article abstract of DOI:10.1002/anie.201108336

In the right light: UV light triggers bond breaking, liberates a caged carboxylic acid, and generates the central C=C double bond in the photoresponsive hexatriene molecule of a dithienylethene molecular switch. Light of the same wavelength converts the colorless isomer into its colored counterpart (see picture) in a visually convenient method to report on the success of the release event. Copyright

Full text of DOI:10.1002/anie.201108336

Angewandte
Chemie
DOI: 10.1002/anie.201108336
Photochemistry
Reporting the Release of Caged Species by a Combination of Two
Sequential Photoreactions, a Molecular Switch, and One Color of
Light**
Tuoqi Wu, Hao Tang, Cornelia Bohne,* and Neil R. Branda*
The use of photolabile molecules to unmask biochemically
relevant compounds offers the spatial and temporal control
required to assess the effects of particular chemical species on
cells and organisms and to potentially release therapeutics on
command.[1] Because light can be tuned and focused, it
provides the on–off control many other stimuli lack and
explains the increased efforts to develop photoresponsive
small molecules capable of acting as molecular cages.
Representative members of photoresponsive families suc-
cessfully used for this application include coumarins,[2,3]
ketoprofens,[4] 2-nitrobenzyl derivatives,[5,6] o-alkylated aryl
ketones,[7,8] and molecules containing the benzoin[9,10] and p-
hydroxyphenacyl groups.[11,12] To date, there are numerous
reports describing how these compounds provide on-com-
mand control of bond-breaking processes in a variety of end
uses ranging from their acting as protecting groups,[2–14]
photolithographic agents,[15,16] molecular probes to studying
biological processes,[17] and molecular delivery vehicles.[18–21]
Despite their popularity and technological potential, one
major limitation of most release systems is their inability to
provide information about when and where the uncaging
event took place. Molecular delivery vehicles that can both
release their payloads and report back to the end user about
the event offer significant advantages to biochemical and
biomedical research and potentially phototherapy. Most
current systems that do report use analytical tools such as
UV/Vis absorption,[13,22] fluorescence[4,14,23–27] and 1H NMR
spectroscopy,[20,22,28] and HPLC.[29] These methods rely on
access to specific instruments and expertise that might limit
their application in many working environments where
convenience to nontechnical end users is a key requirement.
A convenient alternative method is to use a visual readout,
such as the changes in colors of solutions or films, which could
be a direct and easy way for detection even using the naked
eye. Another design feature that would be appealing when
developing a system for release and reporting is the ability to
tailor the molecular structure with functional groups that
allow the end user to decide and differentiate between read-
out signals, whether they have different colors or are
a combination of colors and more sophisticated signals such
as fluorescence, conductivity, and refractive index. The
molecular system we describe here perfectly fits this descrip-
tion and is less prone to any structural modifications resulting
in reduced photochemical performance.
Herein, we illustrate one of our strategies to develop
a relatively universal, visual release-and-report system based
on well-known photoreactions. The molecule highlighted
herein combines the appealing properties of the dialkoxy-
benzoin photocage and the diarylethene class of photo-
switches (Scheme 1), the former representing an effective
photorelease system, the latter being unrivaled as a versatile
molecular backbone that has predictable properties.
The hexatriene substructure found in diarylethene photo-
switches is responsible for their undergoing ring-closing
reactions when exposed to UV light (shown on the left-
hand side of Scheme 1).[30–34] This reaction produces a new
isomer having a unique set of optoelectronic properties due to
the creation of a conjugated p-electron system running
through the backbone of the molecule. Visible light triggers
the reverse reaction and regenerates the original isomer. The
significance of the diarylethene class of photoresponsive
compounds lies in the structural diversity and ease of
synthesis of systems that can be designed to undergo
predictable changes in many specific optoelectronic proper-
ties. All diarylethenes undergo changes in their color, which is
the focus of this report. They can also be tailored to change
the way they fluoresce and phosphoresce,[35,36] rotate polar-
ized light,[37] refract light,[38,39] and undergo electron-transfer
reactions.[40] Any one or a combination of these characteristics
can be used to provide spatial and temporal information
about the release process. In the present example, color is the
read-out signal, which is conveniently visible to the naked eye.
The other photoresponsive system we need to introduce is
based on dialkoxybenzoins, which undergo bond-breaking
reactions when stimulated with UV light and release a wide
range of masked compounds including carboxylic acids,
carbonates, and carbamates (shown on the right-hand side
of Scheme 1). The key feature of interest in this process is the
creation of a new carbon–carbon double bond in the
benzofuran product because this bond can be used as one of
the necessary alkenes in the photoresponsive hexatriene
central in the diarylethene photoswitches.
[*] T. Wu, Prof. N. R. Branda
Department of Chemistry and 4D LABS
Simon Fraser University
8888 University Drive, Burnaby, BC V5A 1S6 (Canada)
E-mail: nbranda@sfu.ca
H. Tang, Prof. C. Bohne
Department of Chemistry, University of Victoria
PO Box 3065 Victoria, BC V8W 3V6 (Canada)
E-mail: cornelia.bohne@gmail.com
[**] This research was supported by the Natural Sciences and Engi-
neering Research Council (NSERC) of Canada, the Canada Research
Chairs Program, University of Victoria, and Simon Fraser University.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201108336.
Angew. Chem. Int. Ed. 2012, 51, 2741 –2744
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2741

.
Angewandte
Communications
pendently for comparison
and undergoes similar
changes in its absorption
spectrum when exposed to
UV light (Figure 1b).[42] In
both cases, there is an imme-
diate appearance of a broad
absorption band in the visible
region of the spectrum (450–
550 nm). This band disap-
pears when either pink solu-
tion is irradiated with visible
light at wavelengths greater
Scheme 1. Diarylethenes undergo reversible ring-closing reactions when exposed to two different colors of
light (left). UV light triggers the release of carboxylic acids (and other functional groups) from
dimethoxybenzoins to produce benzofuran derivatives (right). The key structural elements for each
photoreaction are highlighted.
than 434 nm, which triggers the ring-opening reaction and
generates a solution of 2o, in both cases.
The differences between these two experiments are the
lack of an isosbestic point in the case of 1 (Figure 1a, top plot)
and the difference in the rate at which the absorbance at
510 nm increases (Figure 1c) when 1 is compared to 2o. While
the changes corresponding to the ring-closing reaction of the
pure photoswitch (2o!2c) is accompanied by a clear iso-
sbestic point at 285 nm as is expected for this reaction
(Figure 1b, top plot), the lack of one for the photolysis of
1 implies that there is more than one species involved in the
Scheme 2. UV light triggers both the release of acetic acid from the
2-acetoxy-1,2,2-tri(aryl)ethanone photocage and the ring closing of the
resulting diarylbenzofuran photoswitch. The structural components
required for each step are highlighted in compounds 1 and 2o.
Our design principle is shown by compound 1 in Scheme 2
and involves two sequential photoreactions triggered by the
same color of light. In the first reaction, the photorelease of
acetic acid (eventually other molecules) generates a tris(ben-
zofuran) derivative[41] 2o bearing all three C C double bonds
=
in the hexatriene required for subsequent ring-closing (prior
to the photorelease, only two of these double bonds are
present in 1). Responding to another photon of the same
color, compound 2o will be converted into its corresponding
ring-closed isomer 2c, which will possess unique, predictable
and tunable optoelectronic properties as explained previ-
ously. While the visible light-induced ring-opening reaction is
not formally involved in the release-and-report process, by
being able to toggle the system between two optically unique
isomers, it adds the component of time, which provides
a better and heightened level of spatial resolution and can
potentially distinguish false positive signals.
The synthesis of 2-acetoxy-1,2,2-tri(aryl)ethanone (1) is
described in the Supporting Information. When an aceto-
nitrile solution of this compound is irradiated with UV light
(313 nm),[42] there is an immediate change in its color, from
colorless to pink. This visual change in color is consistent with
the production of the ring-closed isomers of similar diaryl-
ethenes suggesting that the release of acetate followed by
ring-closing of the photoswitch occurred (1!2o!2c). This
two-step conversion is supported by comparing the changes in
the UV/Vis absorption spectra of 1 (Figure 1a) and the ring-
open isomer of photoswitch 2o, which was prepared inde-
Figure 1. Changes in the UV/Vis absorption spectra of CH3CN solu-
tions of a) photocage 1 (top: 1.3ꢀ10ꢀ5 m, bottom: 5.1ꢀ10ꢀ5 m) and
b) pure photoswitch 2o (top: 1.6ꢀ10ꢀ5 m, bottom: 1.6ꢀ10ꢀ5 m) as
they are irradiated with light of 313 nm. The concentrations for the
bottom two plots were chosen to ensure the same absorptivity at
313 nm. The top plot in (a) was measured at lower concentration to
show the lack of an isosbestic point. c) Growth of the normalized
absorbance bands at 510 nm during the irradiation experiments
described in (a) and (b).
2742
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 2741 –2744

Angewandte
Chemie
photoreaction and supports the conversion of 1 to 2c through
2o. This multi-step process is further supported by the fact
that a bulk photolysis of a solution of 1 using UV light,
followed by exposure to visible light, and subsequent removal
of solvent and purification of the residue by column
chromatography afforded pure 2o as characterized by
1H NMR and other spectroscopic techniques.[42]
we assign the off-set absorption for compound 1 to the
formation of 2c. The magnitude of the offset absorption for
2c is not significantly affected by the presence of oxygen.[46]
Photolysis of the ring-open isomer (2o) led to the
formation of a transient with a lifetime of 2.4 ms, which was
shortened in the presence of air. This reactivity is consistent
with the formation of the triplet excited state of compound 2o
as has been suggested in other studies.[47,48] The triplet excited
state of 2o is not involved in the formation of its ring-closed
counterpart since the offset absorption at long delays is the
same when the lifetime of the triplet is shorter in the presence
of air. The transient studies with 2o indicate that 2c is formed
within the laser pulse (15 ns), which is consistent with the
femtoseconds to a few picoseconds[47–50] conversion of diaryl-
ethene photoswitches from their ring-open to their ring-
closed isomers.
Figure 1c clearly shows that for the same light source and
the same optical density at 313 nm, the pure photoswitch 2o is
converted to its ring-closed counterpart much faster than
photocage 1 is converted to 2c, implying that the first reaction
(photorelease of acetate) is the rate-determining step in the
1
entire process.[43] This conclusion is supported by H NMR
spectroscopy of a CD2Cl2 solution of photocage 1 as it is
exposed to 313 nm light. When the irradiation is stopped
before both photoreactions are complete, the spectra show an
equilibrium consisting of all three compounds (1, 2o, and 2c)
as expected. What is important about this equilibrium is the
ratio of the integrated peaks corresponding to the ring-open
and ring-closed isomers of the photoswitch (66:34), which is
the same as for the photostationary state when a solution of
2o is exposed to the same light source. Given the fact that the
photoswitch in both of its isomeric forms (2o or 2c) shows no
spontaneous reactivity in the dark (the photoswitch has
bistability), the amount of each isomer in these experiments is
indicative of a slow photorelease followed by a much faster
ring-closing reaction. This is an important feature of the
system because it ensures that every release event provides
a report signal. If the reverse was true and 1 was converted to
2o faster than the ring-closing reaction (2o!2c), the report
signal would not accurately represent where and when the
release took place.
The transient decay of photocage 1 to the off-set
absorption cannot be fit with random residuals to a mono-
exponential function indicating the presence of more than
one transient. The transient absorption spectra in the
presence of nitrogen and air show the presence of a transient
that is not quenched by oxygen with a maximum absorption at
500 nm. In the nitrogen bubbled solutions a new transient
appears with broad absorption and a peak at 340 nm. The
decay at 500 nm fits well to the sum of two exponentials, with
recovered lifetimes of 0.7 and 2.4 ms. A previous report on the
photolysis of 3’,5’-dimethoxybenzoin esters showed the for-
mation of a triplet state with a lifetime of approximately 1 ms
and absorptions at 330 and 420 nm, and a cationic species with
an absorption maximum at 485 nm
and a lifetime of 1.0 ms.[51] Our results
for compound 1 are consistent with the
Laser flash photolysis studies[44,45] with excitation at
266 nm identified transient species in the UV-light photolysis
of a nitrogen-saturated CH3CN solution of compound 1,
having absorption peaks at approximately 340 and 500 nm
(Figure 2a). The decay does not return to the baseline
indicating the formation of a product with a lifetime longer
than 10 ms. Control experiments with 2o showed the forma-
tion of a species (Figure 2b) a long time after the laser pulse
having an absorption peak similar to the ring-closed isomer
2c formed in steady-state irradiation experiments. Therefore,
formation of its triplet excited state
and the comparable cationic species
A. A recent study on the 3’,5’-dime-
thoxybenzoin photocage suggested
that the photochemical pathway
involving this cyclohexadienyl cation
structure played only a minor role and
a faster process could be the predominant pathway.[52] Due to
the limitations of our equipment, we were not able to observe
any other intermediates prior to generating the proposed
cyclohexadienyl cation structure (A).
We are also assigning a triplet state to cage 1[51] when it is
exposed to UV light because the peak at approximately
340 nm in the transient absorption spectra (Figure 2a) does
not appear in an air-saturated CH3CN solution of the
compound. The fact that the rates at which the absorptions
corresponding to the ring-closed isomer (2c) increase under
steady state conditions are almost identical when carried out
on aerated and de-oxygenated solutions of 1 and 2o shows
that neither the triplet state effects the bulk photolysis of
either compound.[42]
The photocage described in this report is capable of
releasing and reporting using the same color of light. We
regard this system to be a universal type of release system
because a wide range of compounds should be readily
unmasked as has already been shown for simpler versions.
The diethienylethene architecture provides a high level of
Figure 2. Transient absorption spectra for deaerated CH3CN solutions
of a) 1 (7.2ꢀ10ꢀ5 m) at 0.12–0.17 ms ( ) and 7.54–8.52 ms ( ) delay
^
*
times, and b) 2o (3.8ꢀ10ꢀ5 m) at 0.12–0.17 ms ( ) and 7.58–8.56 ms
^
*
( ) delay times. The lines in both plots were included to guide the
eye. For delay times in between and decay traces see the Supporting
Information.
Angew. Chem. Int. Ed. 2012, 51, 2741 –2744
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2743

.
Angewandte
Communications
[24] S. Kitani, K. Sugawara, K. Tsutsumi, T. Morimoto, K. Kakiuchi,
Chem. Commun. 2008, 2103 – 2105.
[25] K. Hayashi, K. Hashimoto, N. Kusaka, A. Yamazoe, H. Fukaki,
M. Tasaka, H. Nozaki, Bioorg. Med. Chem. Lett. 2006, 16, 2470 –
2474.
[26] S. B. Cambridge, D. Geissler, S. Keller, B. Curten, Angew. Chem.
2006, 118, 2287 – 2289; Angew. Chem. Int. Ed. 2006, 45, 2229 –
2231.
versatility because the choice of read-out signal is not limited
to the color, as has been shown by derivatives, the photo-
isomers of which vary in their emission, refractive index, and
charge-transfer properties. Future work will focus on devel-
oping derivatives that are compatible with biological environ-
ments and release more biomedically relevant species.
Received: November 27, 2011
Published online: January 27, 2012
[27] H. Ando, T. Furuta, R. Y. Tsien, H. Okamoto, Nat. Genet. 2001,
28, 317 – 325.
[28] Y. P. Tsentalovich, A. A. Obynochnyi, M. V. Burlov, R. Z.
Sagdeev, P. Burkard, Theor. Exp. Chem. 1998, 24, 324 – 329.
[29] R. S. Rock, S. I. Chan, J. Am. Chem. Soc. 1998, 120, 10766 –
10767.
[30] H. Tian, S. Wang, Chem. Commun. 2007, 781 – 792.
[31] H. Tian, S. Yang, Chem. Soc. Rev. 2004, 33, 85 – 97.
[32] M. Irie, Chem. Rev. 2000, 100, 1685 – 1716.
[33] M. Irie in Molecular Switches (Ed.: B. L. Feringa), Wiley-VCH,
Weinheim, 2001, p. 37.
[34] M. Irie in Photochromic and Thermochromic Compounds, Vol. 1
(Eds.: J. C. Crano, R. J. Guglielmetti), Plenum, New York, 1999,
p. 207.
[35] Y. Jeonga, S. Yang, K. Ahn, E. Kim, Chem. Commun. 2005,
2503 – 2505.
[36] T. B. Norsten, N. R. Branda, J. Am. Chem. Soc. 2001, 123, 1784 –
1785.
[37] T. Yamaguchi, K. Uchida, M. Irie, J. Am. Chem. Soc. 1997, 119,
6066 – 6071.
[38] E. Kim, K. H. Choi, S. B. Rhee, Macromolecules 1998, 31, 5726 –
5729.
[39] T. Kawai, T. Koshido, K. Yoshino, Appl. Phys. Lett. 1995, 67,
795 – 797.
[40] M. N. Roberts, J. K. Nagle, M. B. Majewski, J. G. Finden, N. R.
Branda, M. O. Wolf, Inorg. Chem. 2011, 50, 4956 – 4966.
[41] The more commonly used dithienylethene version of the photo-
switch resulted in significant decomposition during the photol-
ysis and the creation of an uncharacterized product (unpublished
results).
[42] See the Supporting Information for details.
[43] Determining the photolysis quantum yield of compound
1 requires knowledge of the quantum yields for the photolysis
of 1, 2o, and 2c because all these species absorb at the excitation
wavelength. Such a detailed analysis was not pursued at this time
because the qualitative picture is clear—the rate of photolysis
for cage 1 is much slower than for photoswitch 2o.
[44] Y. Liao, C. Bohne, J. Phys. Chem. 1996, 100, 734 – 743.
[45] L. T. Okano, T. C. Barros, D. T. H. Chou, A. J. Bennet, C. Bohne,
J. Phys. Chem. B 2001, 105, 2122 – 2128.
Keywords: caged compounds · photochemistry ·
photochromism
.
[1] M. Goeldner, R. Givens, Dynamic Studies in Biology, Wiley-
VCH, Weinheim, 2005.
[2] T. Furuta, H. Torigai, M. Sugimoto, M. Iwamura, J. Org. Chem.
1995, 60, 3953 – 3956.
[3] V. Hagen, J. Bendig, S. Frings, T. Eckardt, S. Helm, D. Reuter,
U. B. Kaupp, Angew. Chem. 2001, 113, 1077 – 1080; Angew.
Chem. Int. Ed. 2001, 40, 1045 – 1048.
[4] G. Cosa, M. Lukeman, J. C. Scaiano, Acc. Chem. Res. 2009, 42,
599 – 607.
[5] J. H. Kaplan, B. Forbush, J. F. Hoffman, Biochemistry 1978, 17,
1929 – 1935.
[6] R. H. Pawle, V. Eastman, S. W. Thomas, J. Mater. Chem. 2011, 21,
14041 – 14047.
[7] N. C. Yang, C. Rivas, J. Am. Chem. Soc. 1961, 83, 2213.
[8] P. Klꢀn, A. P. Pelliccioli, T. Pospisil, J. Wirz, Photochem.
Photobiol. Sci. 2002, 1, 920 – 923.
[9] J. C. Sheehan, R. M. Wilson, A. W. Oxford, J. Am. Chem. Soc.
1971, 93, 7222 – 7228.
[10] N. Chumachenko, Y. Novikov, R. K. Shoemaker, S. D. Copley, J.
Org. Chem. 2011, 76, 9409 – 9416.
[11] C.-H. Park, R. S. Givens, J. Am. Chem. Soc. 1997, 119, 2453 –
2463.
[12] R. S. Givens, D. Heger, B. Hellrung, Y. Kamdzhilov, M. Mac,
P. G. Conrad, E. Cope, J. I. Lee, J. F. Mata-Segreda, R. L.
Schowen, J. Wirz, J. Am. Chem. Soc. 2008, 130, 3307 – 3309.
[13] V. Hagen, C. Dzeja, S. Frings, J. Bendig, E. Krause, U. B. Kaupp,
Biochemistry 1996, 35, 7762 – 7771.
[14] V. Hagen, S. Frings, J. Bendig, D. Lorenz, B. Wiesner, U. B.
Kaupp, Angew. Chem. 2002, 114, 3775 – 3777; Angew. Chem. Int.
Ed. 2002, 41, 3625 – 2628.
[15] L. Ionov, S. Diez, J. Am. Chem. Soc. 2009, 131, 13315 – 13319.
[16] P. Prompinit, A. S. Achalkumar, X. Han, R. J. Bushby, C. W. Iti,
S. D. Evans, J. Phys. Chem. C 2009, 113, 21642 – 21647.
[17] D. D. Young, A. Deiters, Org. Biomol. Chem. 2007, 5, 999 – 1005.
[18] S. Walbert, W. Pfleiderer, U. E. Steiner, Helv. Chim. Acta 2001,
84, 1601 – 1611.
[19] K. Zhang, J. E. T. Corrie, V. R. N. Munasinghe, P. Wan, J. Am.
Chem. Soc. 1999, 121, 6503 – 6504.
[20] C.-J. Carling, F. Nourmohammadian, J.-C. Boyer, N. R. Branda,
Angew. Chem. 2010, 122, 3870 – 3873; Angew. Chem. Int. Ed.
2010, 49, 3782 – 3785.
[46] These data can be found in Figure S4 and S5 in the Supporting
Information.
[47] Y. Ishibashi, M. Fujiwara, T. Umesato, H. Saito, S. Kobatake, M.
Irie, H. Miyasaka, J. Phys. Chem. C 2011, 115, 4265 – 4272.
[48] M. T. Indelli, S. Carli, M. Ghirotti, C. Chiorboli, M. Ravaglia, M.
Garavelli, F. Scandola, J. Am. Chem. Soc. 2008, 130, 7286 – 7299.
[49] K. Tani, Y. Ishibashi, H. Miyasaka, S. Kobatake, M. Irie, J. Phys.
Chem. C 2008, 112, 11150 – 11157.
[50] H. Miyasaka, T. Nobuto, M. Murakami, A. Itaya, N. Tamai, M.
Irie, J. Phys. Chem. A 2002, 106, 8096 – 8102.
[51] Y. Shi, J. E. T. Corrie, P. Wan, J. Org. Chem. 1997, 62, 8278 – 8279.
[52] C. Ma, W. M. Kwok, H.-Y. An, X. Guan, M. Y. Fu, P. H. Toy,
D. L. Phillips, Chem. Eur. J. 2010, 16, 5102 – 5118.
[21] B. Yan, J.-C. Boyer, N. R. Branda, Y. Zhao, J. Am. Chem. Soc.
2011, 133, 19714 – 19717.
[22] M. A. Ashraf, A. G. Russell, C. W. Wharton, J. S. Snaith,
Tetrahedron 2007, 63, 586 – 593.
[23] W. Lin, L. Long, W. Tan, B. Chen, L. Yuan, Chem. Eur. J. 2010,
16, 3914 – 3917.
2744 www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 2741 –2744