.
Angewandte
Communications
1
0.42 ppm and the shift of the signals corresponding to the
II to a temperature of 608C (see Table 1, entries 3 and 9)
CH groups of the respective NHC moieties (see Figure 2b).
MALDI-TOF mass spectra of the mechanocatalyst I also
proves the prospected structure, fitting well with its calculated
induces only a poor activity (less than 3% conversion as
determined by H NMR spectroscopy). However, triggering
1
the reaction by ultrasound, which is a generally accepted tool
+
[18,22]
isotope patterns (e.g. Ia, C130H242Cu N O , [M] : simulated
for applying mechanochemical force in solution,
leads to
1
4
À1
2
À1
1
956.829 gmol , found: 1956.846 gmol ; for detailed infor-
a significant catalytic activity (Table 1, entries 5—7, and 10. In
[18]
mation see the Supporting Information). Moreover, the
increase in the molecular weight upon formation of the
accordance with different mechanochemical systems,
an
increase in the chain length of the attached polymers, acting
À1
biscarbene [Cu(PIB-NHC) ]X (I) complex was proven by
as a mechanical handle, from M (GPC) = 4750 gmol (Ia) to
2
n
À1
comparing the molecular weights by gel-permeation chroma-
tography (GPC; see Figure 2d as well as the Supporting
Information). A doubling of the molecular weight in the case
17200 gmol (Ic)) leads to an increase in the catalytic activity
from 10 to 27% in the conversion of the “click” reaction,
congruent to an increased mechanical activation through the
À1
of the biscarbene complex Ia (4750 gmol ) compared to the
applied ultrasound. Moreover, a model [Cu(NHC) ]X cata-
2
À1
starting polymer 4a (2200 gmol ) and to the monocarbene
lyst (IV; see Figure 2c) without attached polymer chains
shows only a poor catalytic activity for the “click” reaction
after 24 h at room temperature (Table 1 entry 11). As
a consequence of the expected higher rigidity of the poly-
(styrene)-based catalyst II, its catalytic activity after activa-
tion by ultrasound is 44% (Table 1 entry 10) and thus higher
than those of catalyst Ic, which contains PIB with an even
higher molecular weight (Table 1 entry 7). Furthermore, the
PIB-based monocarbene complex (IIIa) is active, even with-
out applied ultrasound, as no shielding second ligand blocks
the catalytic active site of the copper center.
À1
complex IIIa (2900 gmol ) clearly indicated the desired
chemical structures (see Table S1 in the Supporting Informa-
tion). Similar observations were made for the [Cu(PS-
NHC) ]X (II) complex, thereby proving the final biscarbene
2
complex structure (see the Supporting Information).
The reactivity of the catalysts towards “click” reactions
was probed first in solution (see Table 1) by using a conven-
tional “click” reaction of phenylaceylene (2) and benzylazide
1
(
8), monitored by H NMR spectroscopy. The native catalysts
I and II are inactive at room temperature (see Table 1
entries 2 and 8), irrespective of the molecular weight of the
attached polymer chain, thereby proving that the initial
Finally, and most importantly, the mechanochemical
activation was probed within a real solid polymer matrix,
devoid of any solvent. As it is known that an optimal
transmittance of mechanochemical force is achieved when the
structural polymer contains crystalline regions, high-molec-
biscarbene [Cu(polymer-NHC) ]X complexes I and II are
2
inactive in their “latent” state. Even heating the catalysts Ia or
[
23]
ular-weight poly(tetrahydrofurane) (PTHF)
1
(M (GPC) =
n
À1
12 000 gmol ; crystallinity 68%) was chosen as the struc-
Table 1: Ultrasound triggered “click” reaction of phenylacetylene (2) and
tural polymer, into which the mechanocatalyst I was embed-
ded to check the pressure-induced activation. Critical for the
detection of the so-generated, mechanically induced catalytic
activity is a reliable detection system for monitoring the
progress of the “click” reaction directly within a solid polymer
matrix. Based on previous knowledge of the fluorogenic
benzylazide (8) using mechanocatalysts [Cu(PIB-NHC) ]X (I) or [Cu(PS-
2
[
a]
NHC) ]X (II) in solution.
2
[
24]
“
click” reaction,
the fluorogenic 7-hydroxy-3-(4-phenyl-
I
[c]
[d]
[25]
Ultra-
sound
T
t
[h]
Cu
M (GPC)
Conversion
[%]
1H-[1,2,3]triazole-1-yl)-coumarin dye (3)
(lem = 427 nm,
n
[
b]
À1
catalyst [gmol
]
lex = 260 nm as well as 360 nm) was used to probe and
quantify the catalyst activity in the solid state.
1
2
3
4
5
6
7
8
9
1
1
on
off
off
off
on
on
on
off
off
on
off
RT
RT
72
168 Ia
without
–
0
0
4750
4750
2900
4750
8900
17200
13100
13100
13100
–
The biscarbene [Cu(PIB-NHC) ]X complex Ic (concen-
2
À6
À1
60 8C 72
Ia
IIIa
Ia
<2
11
10
17
27
0
<3
44
0
tration 5.17 10 mmolCu mgsample ) was embedded into the
PTHF matrix (65% crystallinity after embedding; deter-
mined by DSC), together with the two reactive components
phenylacetylene (2) and the initial nonfluorescent 3-azido-7-
[
e]
RT
RT
RT
RT
RT
24
24
24
24
Ib
Ic
À4
À1
hydroxycoumarin dye (1; 1.56 10 mmoldye mgsample ). Ap-
plication of 10 tons pressure by a hydraulic press (3 30 min)
causes a clear induction of the catalytic activity as detected by
144 II
60 8C 72
RT
RT
II
II
IV
0
1
24
24
the increased fluorescence at 427 nm (l = 360 nm) of the
ex
transformed fluorogenic dye 3, thus proving the mechano-
chemical activation of the latent copper catalyst by compres-
sion (see Figure 3b, B and C). The measured fluorescence
after compression corresponds to a dye concentration of
[
a] For all reactions 1 equiv azide 8, 1 equiv alkyne 2, and 0.01 equiv of
I
the Cu catalyst (0.75 mm; I, II, III) were used in 30:1 THF/MeOH. All
experiments were repeated at least twice and display the same results
with maximum deviation of Æ1%. [b] 90 min pulsing at 20 kHz with
3
1
0% of maximal amplitude of 125 mm with a pulse sequence of 5 s pulse,
0 s break, 60 min without pulsing, repeated three times. [c] For I and III
À6
À1
about 6.66 10 mmoldye mgsample and correlates to a con-
version of approximately (4.3 Æ 1)% of the “click” reaction
a PIB standard and for II a PS standard were used. [d] In [D ]THF,
detected through the increasing triazole resonance at d=8.12 ppm as
8
(see the Supporting Information), thereby proving that no
significant fluorescence is observed before compression (Fig-
ure 3a). Control experiments were conducted to exclude the
well as the shift of the CH resonance from d=4.34 to 5.58 ppm.
2
[
e] 1.1 equiv Na/ascorbate were additionally used for reaction of IIIa.
1
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 13918 –13922