Reversible Spatial Control in Aqueous Media by Visible Light: A Thioindigo Photoswitch that is Soluble and Operates Efficiently in Water

Article abstract of DOI:10.1002/chem.201803675

Aiming to extend the scope of indigoid photoswitches to polar protic environments, we have synthesized a sulfonated thioindigo derivative highly soluble in water. Studies by UV/Vis absorption, fluorescence, and NMR spectroscopy indicate that, despite aggregation effects at micromolar concentrations, the novel dye offers satisfactory performance in aqueous solution in the absence of solvation aides. Enrichment of the metastable cis isomer by irradiation may exceed 65 % and its half-life at room temperature may exceed several hours. Degradation after 20 yellow and blue light irradiation cycles within 2 hours is less than 2 %. Performance can be expected to further improve in future molecular designs if the tendency towards self-association is further reduced. Crucially, photoisomerization of indigoids is not necessarily inhibited by water and, thus, the superior spatial control offered by this class of molecular switches may be of great benefit also in biological systems.

Full text of DOI:10.1002/chem.201803675

A Journal of
Accepted Article
Title: Reversible Spatial Control in Aqueous Media by Visible Light: a
Thioindigo Photoswitch Soluble and Operating Efficiently in Plain
Water
Authors: Benjamin Peter Koeppe and Florian Römpp
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201803675
Link to VoR: http://dx.doi.org/10.1002/chem.201803675
Supported by

10.1002/chem.201803675
Chemistry - A European Journal
COMMUNICATION
Reversible Spatial Control in Aqueous Media by Visible Light: a
Thioindigo Photoswitch Soluble and Operating Efficiently in Plain
Water
Benjamin Koeppe,* and Florian Römpp[a]
Abstract: Aiming to extend the scope of indigoid photoswitches to
polar protic environments, we have synthesized sulfonated
structures (Scheme 1b)[19] in which the two hemispheres have a
well defined mutual orientation. Unlike most other trans↔cis
isomerizing photochromic compounds, the blue shift in the cis
isomer of thioindigos is not caused by sterically induced
distortion from planarity and impaired conjugation through the
central double bond, but much rather by non-bonding interaction
between the sulfur and oxygen atoms.[20] These features which
a
thioindigo derivative highly soluble in water. Studies by UV/Vis
absorption, fluorescence and NMR spectroscopy indicate that,
despite aggregation effects at micromolar concentrations, the novel
dye offers satisfactory performance in aqueous solution in the
absence of solvation aides. Enrichment of the metastable cis isomer
by irradiation and its half life at room temperature may exceed 65 %
and several hours, resp. Degradation after 20 yellow and blue light
irradiation cycles within 2 hours is less than 2 %. Performance can
be expected to further improve in future molecular designs if the
tendency towards self association is further reduced. Crucially,
photoisomerization of indigoids is not necessarily inhibited by water
and thus the superior spatial control offered by this class of
molecular switches may be of great benefit also in biological
systems.
thioindigos share with only
a
few other photochromic
compounds (particularly N-functionalized indigos[21] and certain
hemithioindigos[22]) make for an excellent molecular machinery
for spatial control: depending on the flexibility of the linkage,
attached molecular payload can be allowed or even forced to
come close or separate on rotation around the central double
bond of the chromophore (Scheme 1b). As a result of trans→cis
isomerization, some substituents in opposite hemispheres (such
as 4 and 7’) depart from each other and get separated by the
central dye atoms, while others (especially 4 and 4’) approach
and get to face each other. Moreover, bond orientations change:
e.g. the bonds between dye and the 7-substituents are anti-
parallel in trans and become (almost) parallel in cis. Such spatial
changes enabled for example the development of an isomer
specific cryptand for metal cations by the reversible folding and
At present, the most intensively studied classes of photochromic
compounds[ 1 ] include azobenzenes[ 2 ], diarylethenes,[ 3 ] and
spiropyrans.[4] In the context of biological systems, azobenzenes
are one of the most widely used classes of photochromic
compounds,[ 5 ] owing, at least in part, to the pronounced
geometric change that these molecules undergo during
trans↔cis isomerization. Geometric changes of photoactive
molecules may impart further conformational changes in their
surroundings, e.g. in proteins, and to this end water soluble
azobenzenes[ 6 ] including a protein cross-linker[ 7 ] have been
developed. The spatial control exerted by azobenzenes,
however, is limited by the low barriers of rotation around the C-N
unfolding of a crown ether.[23]
a)
single bonds which contribute to
a wide distribution of
b)
geometries particularly for the cis isomer[8] and similar limitations
apply to most other available water soluble photochromic
compounds conveniently referenced for example in the work
about a fulgimide.[9] In contrast, superior spatial control is offered
by thioindigos (Scheme 1a) whose photochemistry and
photophysics have been studied in great depth particularly in the
1970’s by groups around Wyman,[10, 11,12,13] Lüttke,[14, 15] and
Paetzold,[16,17] but which have received less attention in the last
decades.[18] Thioindigos interconvert between the trans and the
metastable cis isomer by a near 180° rotation around the central
double bond.[17] Both isomers generally have rigid, almost planar
4
4
4‘
7‘
5
5
5‘
6‘
6
6‘
6
5‘
7
7
4‘
7‘
cis-1
trans-1
Scheme 1. a) Isomeric structures of the thioindigos addressed. b) Schematic
representation of the geometric features of the isomers of
symmetrically modified with symbolic substituents.
a thioindigo
The trans→cis isomerization of typical thioindigos is
conveniently triggered by green or yellow light while the reverse
process can be initiated by blue light (negative
photochromism[24]). Quantum yields for both processes (φtc and
φct) reach values close to 50 % in favorably substituted
derivatives such as 1b,[13] while higher values are usually
precluded by stochastic aspects of the isomerization
[a]
Dr. Benjamin Koeppe, M. Sc. Florian Römpp
Institut für Chemie
Humboldt-Universität zu Berlin
Brook-Taylor-Str. 2, 12489 Berlin, Germany
E-mail: koeppebe@chemie.hu-berlin.de
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/chem.201803675
Chemistry - A European Journal
COMMUNICATION
mechanisms.[25] The degree of light triggered conversion in each
direction may exceed 80 % depending on the choice of
irradiation wavelength. The thermal half-life of the metastable
isomer is typically in the order of a few hours for room
temperature solutions.[17]
O
2O
O
SO2
a
b
c
HO
S
O
2
O
N(iPr)2
3
2
5
6
4
d
It is probably the inherently low solubility of most thioindigos in
all common solvents
–
usually below routine 1H NMR
e
concentrations[ 26 ] – that poses the major obstacle to their
broader use as molecular switches. While aromatic and
halogenated hydrocarbons allow for the micromolar
concentrations required in spectroscopic studies in the UV and
visible range, solubility in more polar solvents, protic and aprotic,
is typically even lower. Thioindigo derivatives soluble in very low
polarity solvents have been specifically synthesized,[15] but
systematic efforts towards solubility in polar and protic and
eventually aqueous media – as would be the prerequisite for
application in biological systems – have, to the best of our
knowledge, not been reported and success in such an
undertaking was far from certain. The crucial question loomed
whether thioindigo moieties - even if appropriately solubilized -
would be appreciatively photoactive in polar protic media. The
fact that alcohols quench trans→cis isomerization of thioindigos
in benzene solution,[11, 27 ] and literature statements such as
“more polar solvents inhibit [the] trans→cis isomerization [of
thioindigos]”[18] which aligned with our own initial observations
seemed to predict a negative outcome.
Our early attempts at the design of water soluble thioindigo dyes
relied on ether side chains of the type (CH2CH2O)nMe, or
charged sulfate or ammonium groups indirectly attached to the
chromophore through flexible multi atom linkers. The ethers do
display appreciable solubility in many organic solvents including
certain water mixtures, but photoactivity does significantly
decrease in the presence of protic medium constituents and
solubility in water in the absence of additives is negligible. The
above charged species did not display satisfactory solubility in
water or even other solvents.[28]
The breakthrough to be reported here was finally achieved with
derivative 1d (Figure 1). This thioindigo derivative contains
sulfonic acid residues directly attached to the 5-positions (for
solubility enhancement) and additional alkoxy side chains in the
6-positions that serve several purposes among which are the
electronic activation of the aromatic system for substitution and
the concurrent tuning of photochemical properties. Key steps of
the synthesis, previously reported in a similar fashion,[18,29] are
the introduction of the methylsulfanyl residue after ortho
metalation of the amide 4 (Figure 1 step b), and the ring closure
of the resulting thioether by intramolecular substitution followed
by oxidative dimerization to a thioindigo derivative (step d). We
added to the concept a chlorosulfonation of 5 and subsequent
reaction with neo-pentanol that yields the masked sulfonic acid 6
from which step d produces the thioindigo 1c which, after
straightforward purification, may be cleanly deprotected (step
e)[30] to give the di-anion 1d. Notably, both aromatic substitution
reactions take place with perfect regioselectivity. This synthesis
route was chosen with simplicity of individual steps and literature
precedence - not brevity - as main criteria; it should be possible
to devise significantly shorter paths to 1d or similar derivatives.
However, attempts at the direct (chloro)sulfonation of thioindigo
were not successful.
Na21d
1c
Figure 1. Synthesis of water soluble thioindigo 1d: a) four steps[31] b) 1. tert-
BuLi, THF, -78°C, 2. MeSSMe; c) 1. HSO3Cl (neat), r.t., 2. after workup: tert-
Bu-CH2OH, DCM, DABCO, reflux; d) 1. LDA, THF, -50°C, 2. aq. KHCO3,
K4Fe(CN)6, r.t.; e) 1. DMF, TEABr, 100°C; 2. ion exchange to Na+.
The initially produced tetraethylammonium salt (NEt4)21d is
highly soluble in polar organic solvents as well as in water, while
the sodium salt Na21d obtained after ion exchange is soluble
almost exclusively in water or its mixtures (e.g. neither in
absolute methanol nor acetonitrile). This product Na21d, readily
characterized by mass spectrometry as well as 1H and 13C NMR
spectroscopy in D2O solution,[31] was used exclusively in all of
the following and will simply be referred to as 1d without
specification of the counter ion.
Concentrated aq. solutions (> 1 mM) of the dye prepared under
ambient light do not contain amounts of the cis isomer
detectable in routine NMR experiments. After further dilution to
typical concentrations of UV/Vis spectroscopy, however,
samples need to be handled in the dark or under red light if
generation of the cis isomer is undesired. Taking this precaution,
optical absorption spectra of the trans isomer are readily
collected (Figure 2a). A concentration series performed in the
range of 0.1 to 5 µM revealed a small but significant broadening
of absorption bands and a reduction in fluorescence quantum
yields with increasing concentration.[31] Both results are
indicative of solute-solute interactions present at those low
concentrations and we note that deviations from Beer‘s law
caused by such interactions inevitably limit the precision of
spectrophotometric results presented herein.
The rates of thermal cis→trans isomerization of 1d in aq.
solution are generally similar (see below for exceptions) to those
of other thioindigos in apolar solvents: in a 15 µM aq. solution
with a volume fraction of 80 % acetonitrile the half life of the cis
isomer is ca. 170 min, while in the absence of the cosolvent the
process showed complex and slower kinetics, presumably as a
result of aggregation effects.[31]
The absorption spectrum of the cis isomer of the dye shown in
Figure 2a (dotted trace; λmax = 476 nm) was determined by a
procedure which is an extension of the method of Fischer.[31,32]
By excitation at 580 nm in (cosolvent free) water, we obtained
the displayed spectrum containing cis-1d enriched to 65 %. The
dashed trace in Figure 2a is the isomer difference spectrum
indicating that the regions of maximum isomerization related
spectral changes are at 460 and 530 nm. The wavelengths of
highest molar absorbance coefficient ratios
tc,opt = 564 nm and εcis/εtrans at λct,opt = 460 nm (see Figure 2b)
are those where excitation should be aimed when optimum
εtrans/εcis at
λ
This article is protected by copyright. All rights reserved.

10.1002/chem.201803675
Chemistry - A European Journal
COMMUNICATION
isomer selectivity for maximum photochemical isomer
enrichment is desired as the photochemical equilibrium is
generally governed by the relationship
with increasing cis content. This shift clearly indicates that the
isomer equilibrium involves aggregates. Furthermore, the finding
that mainly the 7-H signal (in a relatively hydrophobic region of
the molecule) is subject to a composition dependant shift
suggests (but does not prove) that this is the region where
aggregating molecules approach, while the polar proximities of
the sulfonate groups remain exposed to the polar solvent. If the
latter region included the carbonyls that are among the main
constituents of the chromophore system, this picture would
serve to explain why the electronic spectra of 1d, unlike those of
other thioindigos, show so little aggregation related perturbation.
To assess the performance of our molecular switch, we have
determined quantum yields of trans→cis and reverse
isomerization (φtc and φct) for aqueous solutions containing
variable acetonitrile fractions: yields range from 4.9 to 21 % and
19 to 41 %, for φtc and φct, resp., growing monotonically with
increasing fractions of organic cosolvent (see Figure 3 and
Table S1). These trends are probably related to a decrease in
aggregation tendency in organic solvent rich mixtures. Event
though there seems to be no critical cosolvent fraction above
which parameters converge to any “full” (unperturbed) values (φct
barely increases to the typical value of ≈ 45 %[13] even at lowest
solvent polarity), changes are most pronounced in the region of
high water content. The latter suggests it is the increase in
cosolvent concentration (as solvation aid) rather than the
reduction of the concentration of water (as inhibitor) that governs
isomerization efficiency.
c
cis /ctrans= [εtrans/εcis]λ [φtc/φct]
[Eq. (1)]
if quantum yields are independent of wavelength.
a)
c)
ε / mM-1 cm-1
4-H
7-H
15
cis-1d
trans-1d
PSS
10
10 % cis
30 % cis
60 % cis
(65 % cis)
5
0
ε
trans - εcis
-5
ε
trans / εcis
ratio
b)
10
5
ε
cis / εtrans
0
400
500
λ / nm
600
700
8.5
8.0
7.5
7.0
δ(1H) / ppm
Figure 2. a) Solid traces give UV/Vis absorption spectra of 1d at 15 µM
concentration in water (containing 1 mM pH 7 phosphate buffer): dark sample
(trans-1d) and photostationary state (PSS) under irradiation at 578 nm. Dotted
and dashed traces give the spectrum of cis-1d determined as described in the
text and the corresponding difference in isomer spectra, respectively. b) Ratios
of molar absorbance coefficients of the isomers. c) 1H NMR spectra of 1 mM
1d in aqueous solution (20 % MeCN as solubilization aid, 1 % TFA to stabilize
cis isomer, see text) at various isomer ratios.
quantum yields
50
and ratios, resp.
(φct / φtc ) x 10
Attempts at an NMR based determination of isomer ratios and
the electronic spectrum of the cis isomer were thwarted
unexpectedly by a substantial increase in the rate of thermal cis
trans isomerization upon increase of the dye concentration: at
1 mM, the thermal process deviates clearly from first order
(see trace labels)
40
30
behavior, and
a
tentative bi-exponential treatment (rate
20
φct x 102
constants k1 and k2) resulted in characteristic times (k-1 ln 2) of
just 10 and 46 min.[31] The existence of a concentration
dependence again points to an effect of aggregation. We were
relieved to find that acidification reduces the concentration
related acceleration sufficiently to make cis-1d accessible to
routine NMR experiments (see below).[33] But this latter finding
also shows that the influence of aggregation on thermal
isomerization is an intricate matter in which also pH plays a role
besides both absolute and relative concentrations of the isomers
as well as the concentration of the cosolvent. Depending on
these factors, it seems, the net effect of aggregation on thermal
isomerization rate can be an increase or a decrease relative to
the case of the monomeric dye. Even at minimum solute
concentrations, however, the monomeric state can be realized at
best in the presence of a large fraction of cosolvent. Being thus
effectively obscured by complex aggregation effects, the intrinsic
behavior of the (monomeric) dye in water as our primary focus of
interest is hardly accessible to systematic studies.
10
φtc x 102
0
0
20
40
60
80
100
(VMeCN / Vtotal) x 102
Figure 3. Absolute and relative quantum yields φtc and φct for trans→cis and
cis→trans isomerization of 1d in aqueous solution as a function of volume
fractions of acetonitrile additive. Lines connecting data points are guides for
the eye.
Overall, results are very satisfactory: relatively high values for φtc
and accordingly low values for φct / φtc mean, according to Eq. (1),
that photostationary states with high cis content can be reached.
By featuring a trans→cis quantum yield of φtc ≈ 5 % in the
absence of organic solvent, our switch 1d displays an efficiency
at least equaling that of parent thioindigo 1a in benzene.[11] At
Coming back to NMR results, Figure 2b shows spectra obtained
from samples of various isomer ratios of 1d in an aq. medium
containing 1 % trifluoroacetic acid. In all cases, integrals of both
sets of signals of 4-H and 7-H (the latter identified by a NOESY
contact[31]) unanimously reflect the isomer fractions indicated.
Interestingly, the 7-H signal of the trans isomer shifts downfield
higher acetonitrile contents, xcis greater than 80
% are
observed – and similar values are to be expected in the absence
of solvation aides if in future molecular designs aggregation
tendency is further reduced.
To demonstrate the reversible operation of the photoswitch, we
performed an experiment subjecting a 20 µM aq. solution of 1d
This article is protected by copyright. All rights reserved.

10.1002/chem.201803675
Chemistry - A European Journal
COMMUNICATION
(no cosolvent) to alternating irradiation with yellow and blue light
under spectral monitoring (see Figure 4).
[1]
a) H. Dürr, H. Bouas-Laurent, Photochromism: Molecules and Systems,
Elsevier Science, 2003; b) Organic Photochromic and Thermochromic
Compounds (Eds.: J.C. Crano, R.J. Guglielmetti), Plenum Press, New
York, 1999.
0.65
E
100 % trans
13 % cis
0.55
[2]
[3]
H. M. Bandara, S. C. Burdette, Chem. Soc. Rev. 2012, 41, 1809-1825.
M. Irie, T. Fukaminato, K. Matsuda, S. Kobatake, Chem. Rev. 2014,
114, 12174-12277.
0.45
0.35
0.25
0.15
58 % cis
[4]
[5]
V. I. Minkin, Chem. Rev. 2004, 104, 2751–2776.
W. Szymański, J. M. Beierle, H. A. V. Kistemaker, W. A. Velema, B. L.
Feringa, Chem. Rev. 2013, 113, 6114-6178.
520 nm
493 nm
100 % cis
[6]
[7]
A. A. Beharry, O. Sadovski, G. A. Woolley, J. Am. Chem. Soc. 2011,
133, 19684-19687.
0
20
40
60
t / min
80
100
120
Z. Zhang, D. C. Burns, J. R. Kumita, O. S. Smart, G. A. Woolley,
Bioconjugate Chem. 2003, 14, 824-829.
[8]
[9]
R. L. Klug, R. Burcl, J. Phys. Chem. A 2010, 114, 6401-6407.
C. Slavov, C. Boumrifak, C. A. Hammer, P. Trojanowski, X. Chen, W. J.
Lees, J. Wachtveitl, M. Braun, Phys. Chem. Chem. Phys. 2016, 18,
10289-10296.
Figure 4. Time evolution of absorbance at two selected wavelengths of 20 µM
1d in aq. pH 7 buffer under alternating irradiation with yellow (4 min) and blue
light (2 min).[31] Solid lines connecting experimental points are guides for the
eyes. Horizontal dashed lines indicate absorbance at 520 nm for given isomer
compositions at initial dye concentration.
[10] a) G. M. Wyman, W. R. Brode, J. Am. Chem. Soc. 1951, 73, 1487-
1493; b) G. M. Wyman, B. M. Zarnegar, J. Phys. Chem. 1973, 77,
1204-1207; c) G. M. Wyman, B. M. Zarnegar, D. G. Whitten, J. Phys.
Chem. 1973, 77, 2584-2586.
Plotted absorbance at 520 and 493 nm (the latter being the
isosbestic point) serve to monitor isomer composition and total
dye concentration, resp. During the first yellow light irradiation of
4 min, isomer composition changed from the initial 100 % trans
to about 58 % cis, before blue light (2 min) brought the cis
content back to 13 %. These processes were repeated nineteen
times during which absorbance at 493 nm underwent a shallow
decline with a fitted slope of -6.1 x 10-5 min-1, corresponding to
1.6 % dye degradation within 2 hours and 20 cycles.
[11] G. M. Wyman, B. M. Zarnegar, J. Phys. Chem. 1973, 77, 831-837.
[12] G. M. Wyman, J. Chem. Soc. D 1971, 1332-1334.
[13] A. D. Kirsch, G. M. Wyman, J. Phys. Chem. 1977, 81, 413-420.
[14] W. Lüttke, M. Klessinger, Chem. Ber. 1964, 97, 2342-2357.
[15] H. Meier; W. Lüttke, Liebigs Ann. Chem. 1981, 1303-1333.
[16] a) M. Erler, G. Haucke, R. Paetzold, Z. phys. Chem. (Leipzig) 1977,
258, 315-320; b) G. Haucke, R. Paetzold, J. Prakt. Chem. (Leipzig)
1979, 321, 978-986.
[17] G. Haucke, R. Paetzold, Nova Acta Leopoldina, Supplement 11 1978.
[18] a) M. Dittmann, F. F. Graupner, B. Maerz, S. Oesterling, R. deVivie-
Riedle, W. Zinth, M. Engelhard, W. Lüttke, Angew. Chem. Int. Ed. 2014,
53, 591-594; b) M. Dittmann, F. F. Graupner, B. Maerz, S. Oesterling,
R. deVivie-Riedle, W. Zinth, M. Engelhard, W. Lüttke, Angew. Chem.
2014, 126, 602-605.
Summarizing, we have developed
a prototype thioindigo
photoswitch remarkably soluble and efficiently switching in water.
Its crucial design element seems to be the presence of charge in
direct proximity of the chromophore scaffold whose stiffness and
apolar nature otherwise imparts high tendency towards self
association and very low solubility. The rigidly attached charge
not only increases affinity to a polar medium, but by its spatial
proximity to the conjugated system quite effectively repels
equally charged chromophore moieties. Even though some
uncertainty still remains about residual aggregation, thermal
isomerization and pH effects, we are convinced that this
approach can be applied in the conception of diverse functional
(thio)indigo photoswitches with unique properties and that this
opens new possibilities which should draw considerable
attention to the employment of a classic – yet presently much
underused – class of photochromic compounds.
[19] D. Jacquemin, J. Preat, V. Wathelet, M. Fontaine, E. A. Perpète, J. Am.
Chem. Soc. 2006, 128, 2072-2083.
[20] J. Sühnel, K. Gustav, J. Prakt. Chem. (Leipzig) 1978, 320, 917-921.
[21] C.-Y. Huang, A. Bonasera, L. Hristov, Y. Garmshausen, B. M. Schmidt,
D. Jacquemin, S. Hecht, J. Am. Chem. Soc. 2017, 139, 15205-15211.
[22] M. Guentner, M. Schildhauer, S. Thumser, P. Mayer, D. Stephenson, P.
J. Mayer, H. Dube, Nat. Commun. 2015, 6, 8406.
[23] a) M. Irie, M. Kato, J. Am. Chem. Soc. 1985, 107, 1024-1028; b) S. M.
Fatah-ur Rahman, K. Fukunishi, J. Chem. Soc. Chem. Commun. 1994,
917-918.
[24] V. A. Barachevsky, Rev. J. Chem. 2017, 7, 334-371.
[25] a) T. Karstens, K. Kobs, R. Memming, Ber. Bunsen-Ges. 1979, 83,
504-510; b) R. Memming, K. Kobs, Ber. Bunsen-Ges. 1981, 85, 238-
242; c) C. P. Klages, K. Kobs, R. Memming, Chem. Phys. Lett. 1982,
90, 51-54.
[26] G. Voss, Color. Technol. 2006, 122, 317-323.
[27] T. Karstens, Ber. Bunsen-Ges. 1982, 86, 311-315.
Acknowledgements
[28] a) F. Römpp, Bachelor thesis, Humboldt-Universität zu Berlin, 2014; b)
F. Römpp, Master thesis, Humboldt-Universität zu Berlin, 2017.
[29] C. Mukherjee, S. Kamila, A. De, Tetrahedron 2003, 59, 4767-4774.
[30] J. C. Roberts, H. Gao, A. Gopalsamy, A. Kongsjahju, R. J. Patch,
Tetrahedron Lett. 1997, 38, 355-358.
The authors would like to thank Prof. Stefan Hecht, Ph.D. for
hosting this project in his lab and all general support and we also
acknowledge the valuable help of M.Sc. Sebastian Fredrich.
This work was supported by the DFG (project KO 5131).
[31] See the Supporting information for details or spectra, respectively.
[32] E. Fischer, J. Phys. Chem. 1967, 71, 3704-3706.
[33] On the other hand, basic conditions (pH 10) accelerate thermal
isomerization considerably, even in low concentration samples.
Keywords: kinetics • photochromism • solubility • thioindigo •
water chemistry
This article is protected by copyright. All rights reserved.