Angewandte
Communications
Chemie
Table 1: Catalyst screening for the [2+2] photocycloaddition of 3 and 2 in
Table 2: Influence of the catalyst loading on the intermolecular [2+2]
photocycloaddition of 3 and 2 in CH Cl for various thiourea catalysts.
[
a]
[a]
CH Cl .
2
2
2
2
[
b]
[c]
[c]
[c]
[b]
Entry
Catalyst
2/4
MB [%]
Conv. [%]
Yield [%]
Entry
Catalyst
mol%)
2/4
(
12 h
24 h
36 h
40 h
1
2
3
4
5
6
–
89:11
32:68
38:62
76:24
38:62
22:78
86
96
96
88
93
96
11
65
56
22
58
76
9
62
54
19
54
73
[
c]
1a
1b
1c
1d
1e
1
2
3
4
5
6
7
–
95:05
23:77
56:44
55:45
31:69
75:25
87:13
85:15
01:99
23:77
22:78
01:99
57:43
77:23
77:23
66:34
[d]
[d]
1a (10)
1b (10)
1d (10)
1e (10)
1e (5)
1e (1)
–
–
10:90
10:90
7:93
[
d]
–
–
[d]
[d]
–
[d]
46:54
65:35
31:69
50:50
[
a] Catalyst loading: 100 mol%; irradiation time: 19 h. Averages of
[d]
a minimum of three runs (error: Æ5%) are given. [b] Determined by GC
analysis with a ChirasilDex-CB column. [c] The conversion of 2, the mass
[a] Averages of a minimum of three runs are given (error: Æ5%).
1
balance (MB), and the yield of 4 were determined by H NMR
spectroscopy using triphenylmethane as an internal standard in CDCl or
[b] Determined by GC analysis with a ChirasilDex-CB column. [c] Irradi-
ated for 48 h. [d]>90% conversion observed with shorter irradiation
times. Hence, longer irradiation times were not studied. For further
studies on the catalyst loading, see the Supporting Information.
3
CD CN.
3
transmission using
lamp).
a 450 W medium-pressure mercury
[
7]
was 50:50 (entry 7) after 40 h irradiation. As expected, longer
irradiation times were required for lower catalyst loadings to
achieve a similar level of conversion. Evidently, the thiourea
catalyst 1e efficiently facilitated the reaction with good
turnover even at low catalyst loading. To demonstrate the
versatility of the thiourea catalyst in promoting large-scale
photoreactions, we performed the reaction on gram scale. We
achieved 80% conversion with 30 mol% of 1b, and the
Table 1 reveals that the thiourea-based organocatalysts
are effective in promoting the [2+2] photocycloaddition of
coumarin with tetramethylethylene with moderate to high
conversion and with excellent mass balance. The conversion
was higher with the thiourea than with the corresponding urea
[
11]
catalyst. For example, 65% conversion was observed with
thiourea 1a compared to 30% conversion with the corre-
[
11]
[11]
sponding urea 1 f.
Other symmetric and unsymmetric
desired photoproduct was isolated in 75% yield.
catalysts gave modest conversions (Table 1; see also the
Detailed photophysical investigations were carried out to
understand the role of the thiourea catalysts in promoting the
[2+2] photocycloaddition of 2 with 3. Thioureas 1d and 1e
were selected to understand the role of both the thiourea and
naphthalene chromophores in promoting the photoreaction.
Absorption spectra of the thioureas and 2 were recorded to
determine which of the chromophores is predominantly
absorbing light. Coumarin shows no absorbance above
350 nm, whereas 1d absorbs weakly up to 400 nm (Fig-
ure 1A). It should be pointed out that the 350 nm lamp that
was used for the photocatalytic reactions has a spectral width
of Æ 30 nm. No appreciable fluorescence was observed for 2 at
room temperature in toluene. We thus recorded the lumines-
cence of 2 at 77 K in MCH glass (Figures 1B). We observed
strong fluorescence centered around 408 nm and weak
phosphorescence (overlapping with the fluorescence
[
11]
Supporting Information, Table S1). The most striking result
was achieved with the naphthyl-substituted catalyst 1e, which
gave 76% conversion with excellent (96%) mass balance
(
Table 1, entry 6). As thiourea-based catalysts were effective
in promoting the intermolecular photoreaction of 2 with 3, it
became critical to ascertain the best solvent(s) for this
photocatalytic process. We investigated the efficiency of the
photocycloaddition using 1a, 1b, 1d, and 1e in four different
solvents, namely methylcyclohexane (MCH), toluene, tetra-
[11]
hydrofuran (THF), and dichloromethane. In dichlorome-
thane and toluene, the reaction proceeded with excellent
[11]
conversion, yield, and mass balance (Table S2). In THF, the
conversion varied between 26 to 99%, and the mass balance
[
11]
was moderate. Low conversions and product yields were
observed in MCH.
[11]
Based on the results shown in Table S2, we selected
dichloromethane as the solvent of choice to investigate the
efficiency of the reaction with respect to the thiourea catalyst
loading (Table 2). The catalyst loading was varied from
signal).
The emission characteristics matched the previ-
[
12]
ously reported profile, which was rationalized to be due to
the monomer and the dimeric aggregate of 2 in MCH glass at
77 K. Kinetic decay analysis of the luminescence at 412 nm
showed two lifetimes of 1.2 ns and 4.0 ns. On the other
hand, luminescence measurements with a 1:1 mixture of 1 f
and 2 showed a significant reduction of the coumarin
fluorescence in addition to reduced lifetimes at 412 nm
[11]
7
0 mol% to 10 mol% for thioureas 1a, 1b, and 1d
[11]
(
Table S3, entries 2–13) and from 70 mol% to 1 mol% in
[11]
the case of 1e (Table S3, entries 14–19). To compare the
efficiency of the thioureas at different loadings, the reactant/
photoproduct ratio (2/4) was investigated at various irradi-
ation times (12–48 h) for a given amount of catalyst. Table 2
shows that at a catalyst loading of 10 mol%, the 2/4 ratio for
[11]
(0.6 ns and 3.2 ns). More importantly, for a 1:1 mixture of
2 and 1d, the emission intensity depended on the excitation
wavelength (Figure 1B,C). For the 1:1 mixture of 2 and 1d,
2
4 h irradiation was 1:99 for 1a (entry 2), 23:77 for 1b
with l = 340 nm, weak emission from the coumarin and
ex
(
entry 3), 22:78 for 1d (entry 4), and 1:99 for 1e (entry 5). As
a new emission centered at approximately 523 nm were
observed. Upon changing the excitation wavelength to
360 nm, where 2 does not have any significant absorption,
the new emission band was predominantly observed, which is
likely due to the selective excitation of the coumarin–catalyst
catalyst 1e was found to be optimal for the [2+2] photo-
cycloaddition of 2 with 3, we investigated its efficiency at
loadings of 5 and 1 mol%. With 5 mol% of 1e, the 2/4 ratio
was 31:69 (entry 6), and at 1 mol% loading, the ratio of 2/4
5
448
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 5446 –5451