Angewandte
Communications
Chemie
Supporting Information). X-ray diffraction analysis of pure 1r
Table 1: First-order rate constants (k) determined for the rearrangement
reaction of DAEs 1c and 3c with various amines.
revealed that the DAE core rearranges into a benzothiophene
[18]
[
a]
[b]
[b]
scaffold, causing the decomposition of the thiophene ring
opposite to the aldehyde, generating a photochemically
unreactive compound (Figure 2b).
Amine
Structure
k(1c)/min
k(3c)/min
SPER
1.67
2.66
The presence of N-benzylformamide as a by-product in
the HNMR experiments provides further insight into
CAD
DAP
OA
0.30
0.49
0.12
0.033
0.50
1.58
0.13
0.025
1
a reasonable mechanism for the observed rearrangement
reaction. Initially, the nucleophilic attack of the amine at the
highly reactive, aliphatic aldehyde of the closed DAE isomer
produces the corresponding hemiaminal intermediate. Sub-
sequently, this intermediate rearranges to the more stable
compound 1r with elimination of N-benzylformamide
BA
MBA
BMA
0.002
0.0012
0.0010
n.r.
0.0007
n.r.
DMBA
(
Figure 1, top). To confirm the proposed mechanism for the
À2
À5
formation of 1r, we synthesized the imine derivative 1i
starting from the open DAE 1o. As expected, 1i can be
switched between the open and the closed isomers by
alternating UV- and visible-light irradiation in a reversible
fashion and without generation of 1r (Scheme 1, see Fig-
ure S35 in the Supporting Information). This is in agreement
with the proposed mechanism for the formation of 1r, which
requires the intermediacy of the hemiaminal derivative
[
a] 1ꢂ10 m in CH CN. [b] 2ꢂ10 m in CH CN.
3
3
Supporting Information); afterwards a specific amount of
amine was added. We monitored the variation of optical
density at the maximum of the visible absorption band
belonging to the closed isomer. This region was chosen since
the absorption of the rearrangement products 1r–4r is
minimal and the absorbance change thereby maximized.
The rate constants were determined by fitting the UV/Vis
spectral changes with a first-order reaction. As expected from
the proposed mechanism, tertiary amines are not reactive.
Secondary amines react much slower than primary ones as
indicated by comparing the rates of BMA with BA. This
discrimination is presumably due to the increased steric
hindrance of the secondary amines. For similar reasons BA
reacts faster than MBA. Purely aliphatic amines are about
one order of magnitude more reactive than BA. Finally,
aliphatic oligoamines, such as DAP, CAD, and SPER react
faster than what would be expected by simply correcting for
the number of amino groups present in the corresponding
molecules. Clearly, neighboring group effects play an impor-
tant role in these cases.
(
Figure 1, top). Similar experiments were performed using
1
switches 2–4. UV/Vis absorption and HNMR spectra con-
firmed the transformation of the investigated switches into
products that show the same central scaffold as 1r, and differ
only in the substitution pattern (for further details see the
Supporting Information). The main difference in the amine-
induced rearrangements of DAEs 1c–4c is the reaction time,
which reflects the reactivity of the carbonyl moiety. As
expected, the reactivity follows the order 3c > 1c ꢀ 2c > 4c
(
for the UV/Vis spectra see the Supporting Information). This
order can be rationalized considering that the less reactive
ketone 4c forms the hemiaminal intermediate more slowly
than the aldehydes, of which 3c is the fastest as it contains no
electron-donating methoxy group. Note that the effect of
substitution on the thiophene ring is smaller than that
resulting from a slight change in the amine structure, high-
lighting the degree of chemoselectivity for sensing amines.
After serendipitously discovering this unique rearrange-
ment reaction of the closed, carbonyl-substituted DAE
isomers, we engaged in a systematic spectroscopic investiga-
tion in order to 1) quantify the rate constant for the reaction
of all four switches with a variety of amines and to
When considering these reactivity data, particular atten-
tion should be paid to 3c, which displays better photoswitch-
ing performance (high PSS), excellent thermal stability (see
Figure S18 in the Supporting Information), and also higher
reactivity. According to the data obtained for CAD, the
rearrangement of 3c occurs 1.6 times faster than that of 1c
À5
and a limiting concentration of 2.0 ꢁ 10 m was determined,
2
) determine the lowest concentration range for which an
with a 50% decrease of the absorption at 580 nm within
110 min (see Figure S57 in the Supporting Information).
To optimize the conditions for sensing of the amines we
studied the influence of the pH on the kinetics of the reaction
(see Section 6.1. in the Supporting Information). The optimal
rate was obtained upon addition of 10 equiv of p-toluenesul-
evolution of the spectral bands takes place over a reasonable
time frame (up to 2 h).
A collection of eight different amines was studied in
combination with the synthesized DAE switches (Table 1 and
the Supporting Information). Amines were chosen to repre-
sent different classes, such as tertiary amines (N,N-dimethyl-
benzylamine (DMBA)), secondary amines (N-benzylmethyl-
amine (BMA)), bulky primary amines (benzylamine and (S)-
fonic acid to a solution of 3c in CH CN followed by the
3
addition of 10 equiv of CAD. This resulted in a 40-fold rate
increase as compared to the rate in the absence of the acid
catalyst. A larger excess of acid led to decreasing rates. This
observation points to the classic scenario of an optimal,
(
À)-a-methylbenzylamine (MBA)), primary linear aliphatic
amines (octylamine (OA)), primary diamines (1,3-diamino-
propane (DAP) and cadaverine (CAD)), and aliphatic tris-
[
19]
slightly acidic pH window, where a small amount of acid
increases the electrophilicity of the carbonyl group yet too
large amounts lead to protonation of the amine, thereby
reducing its nucleophilicity.
amines (spermidine (SPER)). In a typical experiment, a
2
its PSS (for the PSS values of the DAEs see Table S2 in the
À5
ꢁ 10 m solution of the DAE was irradiated until it reached
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3
These are not the final page numbers!