A PMMA-based heterogeneous photocatalyst for visible light-promoted [4 + 2] cycloaddition

Macromolecular organic photocatalysts consisting of a completely conjugated network have broad and promising applications in visible light-promoted photoredox catalysis. Precise reproduction and control of exact conjugation length remains an important challenge for fully conjugated and macromolecular photocatalysts. Here, we introduce a new photocatalytic material based on classical PMMA copolymerised with defined electron donor and acceptor units with precisely controllable redox potential and conjugation length, creating a promising class of metal-free, stable and low-cost heterogeneous photocatalysts. Furthermore, swelling of the PMMA copolymer matrix in organic solvents led to enhanced substrate diffusion and thereby increased catalytic efficiency. High efficiency and selectivity was achieved for photocatalytic [4 + 2] cycloaddition reactions in air with low effective photocatalyst loading. The photocatalytic efficiency of the PMMA photocatalyst was comparable with the state-of-the-art metal or non-metal catalysts while facilitating easy recyclability.

The classical polymer chemistry toolbox provides access to tuneable polymeric materials with variable solubility, biological compatibility, mechanical properties, and responses to external triggers. However, the role of classical polymers as a platform for photocatalytic systems has barely been explored and studies are limited to attaching photoactive metal complexes 27,28 or organic dyes 29 to linear polymers. Recently, our group published temperatureresponsive photocatalytic nanogels based on poly-Nisopropylacrylamide, demonstrating the versatility of combining photoactive units with well-known polymeric systems. 30 In general, the incorporation of photocatalytic properties into classical polymer networks via copolymerisation of photoactive monomers could be an effective strategy towards a novel class of efficient, cheap and metal-free photocatalysts. [31][32][33] Moreover, this combination may yield photocatalytic materials with considerable advantages in terms of stability and long-term usage for visible light-promoted redox catalysis.
In this work, we designed a cross-linked copolymer based on polyĲmethyl methacrylate) (PMMA) with integrated photocatalytic units consisting of 4,7diphenylbenzothiadiazole (cPMMA-BTPh 2 ). The resulting polymer is an efficient metal-free and redox-active heterogeneous photocatalyst for visible light-promoted chemical transformations. This material must exhibit optical and redox properties similar to the molecular photocatalytic analogue unit, whilst demonstrating advantageous chemical robustness and recyclability attributed to polymer materials. [4 + 2] cycloaddition reactions (the Diels-Alder reaction) could be efficiently catalysed by the PMMA photocatalytic material with low catalyst loading, in particular, 0.06 mol% of the effective photocatalytic unit with respect to the substrate. The copolymer photocatalyst exhibited comparable photocatalytic efficiency and selectivity to the state-of-the-art molecular and transition metal-containing photocatalysts. Specifically, the photocatalytic copolymer material exhibited comparable efficiency when compared to its small molecule analogue. No photobleaching effect of the polymer photocatalyst was observed, as confirmed by recycling experiments. Furthermore, a detailed mechanistic study employing advanced photophysical methods, such as timeresolved photoluminescence and transient absorption spectroscopy, was conducted. Further to the detailed studies on the benchmark reaction, the scope of the reactivity of the copolymer photocatalyst was expanded to a broader range of substrates, proving its excellent efficiencies and functional group tolerance (Scheme 1).

Results and discussion
The photocatalytic copolymer (cPMMA-BTPh 2 ) in this study was synthesized via free-radical polymerisation of methyl methacrylate (MMA), polyethylene glycol dimethacrylate (PEGDMA) and 4-phenyl-7-(4-vinylphenyl)benzothiadiazole (BTPh 2 ) in low concentrations. Details on synthetic procedures and characterization are given in the Experimental section and ESI. † In order to confirm the incorporation of the photoactive BTPh 2 unit into the cross-linked PMMA, solid-state nuclear magnetic resonance (ssNMR), UV/vis, photoluminescence (PL) and cyclic voltammetry (CV) analyses were performed. The ssNMR spectra of cPMMA-BTPh 2 and a photoinactive cPMMA analogue show clear signals of the photoactive unit ( Fig. S1 and S2 †). The UV/vis absorption spectrum of cPMMA-BTPh 2 showed an absorption range up to 475 nm with a maximum at 386 nm, while the emission area between 425 nm and 650 nm with a maximum at 512 nm was determined (Fig. 1a). The result is consistent with the optical properties of the corresponding small molecule BTPh 2 and previous reports. 34,35 The slight red shift of cPMMA-BTPh 2 compared to BTPh 2 can be attributed to hyperconjugation effects of the photoactive unit in cPMMA-BTPh 2 , which is one-sidedly attached to the polymer backbone. The optical properties of cPMMA-BTPh 2 in terms of emission and absorption behaviors are similar to those of fac-[IrĲppy) 3 ], a wellestablished transition metal photocatalyst. 36 Cyclic voltammetry (CV) measurements revealed that the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) of cPMMA-BTPh 2 are at +1.62 eV and −1.33 eV vs. SCE ( Fig. 1b and S3 †), which matches with the potentials of the analogous small molecular organic photocatalyst BTPh 2 . Theoretical calculations for cPMMA-BTPh 2 using the density functional theory (DFT) at B3LYP/6-31G(d) level (Table S1 †) revealed that the electron densities of HOMO and LUMO are mainly located on the diphenylbenzothiadiazole unit. This indicates that the electronic structure is determined with a side-chain consisting of the D-A-type unit BTPh 2 . The small deviations between the UV/vis, PL and CV spectra of BTPh 2 and cPMMA-BTPh 2 were reproducible and we attributed them to the hyperconjugation effects in cPMMA-BTPh 2 .
The swelling behavior of cPMMA-BTPh 2 was investigated by live-imaging the photocatalytic material after exposure to Scheme 1 Illustration of the design concept of the PMMA-based photocatalyst copolymer and its application in the photocatalytic [4 + 2] cycloaddition. CH 3 NO 2 with a bright-field/fluorescence microscope. The photocatalytic gel was observed to swell to about 4 times its original size within 2 s. The ability of the gel to swell is controlled by the cross-linking density and the cross-linker length. The long ethylene glycol chains in the cross-linker PEGDMA allow the gel to expand, enabling easy diffusion of substrates and products in and out of the photocatalytic material. Further swelling experiments showed the strong dependency of the swelling behaviour on the solvent choice ( Fig. S4 †).
Fourier transform infrared (FTIR) spectroscopy and thermogravimetric analysis (TGA) characterization were performed with the cPMMA-BTPh 2 photocatalyst and compared to a non-photoactive cPMMA analogue. The FTIR spectrum (Fig. 1d) showed typical signals, for example, at 2992 and 2950 cm −1 , which are characteristic of the C-H bond stretching vibrations of the -CH 3 and -CH 2groups, respectively. The signal at 1724 cm −1 indicates the presence of an acrylate carboxyl group. The broad signals at 1239 and 1143 cm −1 can be assigned to the C-O-C stretching vibrations. The two bands at 1387 cm −1 and 754 cm −1 can be attributed to the α-methyl group vibrations. The band at 1441 cm −1 can be attributed to the bending vibrations of the C-H bonds of the -CH 3 group. Interestingly, the influence of the photoactive unit on the FTIR spectrum is minimal (Fig. S5 †), showing that the chemical constitution and connectivity of the catalyst is mainly determined by the PMMA network. Thermogravimetric analysis (TGA) showed that the photocatalyst remained intact up to 295°C under nitrogen, comparable to the blank cPMMA sample. FTIR and TGA measurements prove the robustness of the cPMMA-BTPh 2 photocatalyst, which shows that the properties of the cross-linked PMMA network are directly transferred to the polymeric cPMMA-BTPh 2 system.
A rough estimation of the material cost of the cPMMA-BTPh 2 photocatalyst gave a value of approximately 7€ per g based on the current commercial prices of the chemicals used in the synthesis (see Fig. S6 †). The low photoactive unit loading in the catalyst, as well as using PMMA as a commodity polymer, gives the system a cost advantage.
To demonstrate the photocatalytic activity of the cPMMA-BTPh 2 system, the stereochemically highly challenging [4 + 2] reaction was selected. The screening and control experiments of the model reaction between trans-anethole and isoprene are listed in Table 1. Significantly, the [4 + 2] cycloaddition of trans-anethole and isoprene was achieved with high selectivity (>99%) in a quantitative manner within 4 h at room temperature in air (entry 1). An apparent quantum yield of 2.27% could be observed, with 0.6 μmol mL −1 photoactive unit present in the model reaction, which is 0.06 mol% with respect to the substrate. A first order kinetic reaction was observed (Fig. S7 †). A control experiment conducted under light, then in the dark, and then under light again yielded only a trace amount of the product in the dark period. This indicated that the generated radical cation of trans-anethole was only active during light irradiation ( Fig. S8 †), suggesting temporal and spatial limitations of the radical-based reaction. 37 Nitromethane outperformed both acetonitrile and dichloromethane for the model reaction (entries 2 and 3). The solvent dependency could be ascribed to the polaritydependent stabilization of the excited state. Additionally, the solvent-dependent swelling and accessibility of the photocatalytic units could play a major role. The reaction did not proceed without light irradiation or the use of cPMMA-BTPh 2 (entries 4 and 5). In the absence of oxygen, no product formation could be observed after 4 h (entry 6). When using CuCl 2 as an electron scavenger under a nitrogen atmosphere, an increased reaction conversion (71%) was obtained, indicating that the copper salt can take the role of oxygen as the catalyst regenerator (entry 7). Addition of KI as a photogenerated hole scavenger only led to traces of the desired product (entry 8). The small molecular photocatalyst BTPh 2 (applied with a similar loading) shows full conversion of the [4 + 2] adduct within 2 h of reaction time, demonstrating the efficiency of homogeneous catalysis. The PMMA photocatalyst in this study works efficiently on a similar timescale to its small molecular analogue, which hints at the good accessibility of photocatalytically active moieties and diffusion of reactants throughout the network.

View Article Online
In general, the benchmark reaction demonstrates the high efficiency of cPMMA-BTPh 2 in catalysing organic reactions and the successful application of the concept. The cPMMA-BTPh 2 photocatalyst could be repeatedly used for several cycles without losing its photocatalytic efficiency (Fig. S9 †), outperforming the molecular analogue BTPh 2 (Fig. S10 †). No photobleaching was observed in the UV/vis absorption spectra of cPMMA-BTPh 2 after the reaction (Fig. S11 †). Solubility tests with nitromethane, chloroform and THF underlined the heterogeneity of the photocatalyst (Fig. S12 †). These findings demonstrate the robustness, high stability and reusability of the photocatalytic material designed.
We propose a reaction mechanism based on our observations and previous reports. [38][39][40] Three reaction pathways are possible, as illustrated in Fig. 2. Under light irradiation, the substrate trans-anethole was oxidised by the photogenerated hole of cPMMA-BTPh 2 to its cationic radical intermediate, which reacts with isoprene to form a cyclic intermediate. The cyclic intermediate could be reduced by cPMMA-BTPh 2 formed via reductive quenching, regenerating the photocatalyst back to its ground state and forming the final product to complete the catalytic cycle (pathway i). In addition, the final product may also be obtained via either oxidation of another neutral trans-anethole by the cyclic intermediate (pathway iii), or oxidation of O 2˙− , 5 generated by electron transfer from cPMMA-BTPh 2 to oxygen (pathway ii). The formation of the electron-activated superoxide radical (O 2˙− ) was confirmed by electron paramagnetic resonance (EPR) experiments. In a control experiment using 5,5dimethyl-1-pyrroline N-oxide (DMPO) as a superoxide radical trapping agent, typical EPR patterns for DMPO-O 2˙− adducts were obtained (Fig. 3a).
Time-resolved photoluminescence (TRPL) was used to gain a deeper insight and elucidate the electron transfer between BTPh 2 as the photocatalytic unit and substrates. As shown in Fig. 3b, the fluorescence lifetime of BTPh 2 in CH 3 CN after purging with N 2 was determined to be 12.9 ns, which was comparable to that of the well-applied organic photocatalysts, such as Acr + -Mes (15 ns) 41 and eosin Y (6 ns). 42 In the presence of O 2 , the excited state of BTPh 2 could be quenched to a decreased fluorescence lifetime of 10.7 ns. A similar phenomenon was observed when using trans-anethole as a quencher, leading to a reduced fluorescence lifetime of 11.5 ns. This indicated a hole filling process by electron transfer between BTPh 2 and trans-anethole. 5 By simulating the conditions of the model reaction with trans-anethole in air, a similar decay lifetime of BTPh 2 (11.2 ns) was observed, further confirming that the [4 + 2] cycloaddition is hole initiated. Transient absorption (TA) spectra were recorded and are consistent with the TRPL data shown (Fig. S13-S18 †).
The versatility of the photocatalytic material was investigated by employing different dienes and dienophiles in the [4 + 2] cycloaddition reaction (Fig. 4). High reaction yields from 76 to 99% were obtained, confirming the general applicability of cPMMA-BTPh 2 for organic photoredox reactions. The conversions were found to be selective towards the [4 + 2] cycloaddition. The formation of (2 + 2) cycloaddition products could not be observed.

Materials and physical methods
All chemicals and solvents were purchased from commercial sources and used as received unless otherwise noted. For the photocatalytic Diels-Alder reaction, all dienes and reaction solvents, i.e. nitromethane, were purified by elution through neutral aluminium oxide (50-200 um) and anhydrous CaCl 2 (w/w, 95/5). Flash column chromatography was conducted over silica 60 (0.063-0.2 mm). Reaction yields refer to the pure compounds after being purified via column chromatography. Solid-state diffuse reflectance UV-vis absorption and fluorescence spectra were recorded on a PerkinElmer Lambda 100 spectrophotometer and J&M TIDAS spectrofluorometer at ambient temperature, respectively. EPR (electron paramagnetic resonance) was conducted on a Magnettech MiniScope MS200 spectrometer at room temperature. DMPO (0.1 M) and the catalyst (1 mg mL −1 ) in acetonitrile were used for the spin trap experiment. Irradiation was conducted with an integrated blue light source. Cyclic voltammetry measurements were carried out on a Metrohm Autolab PGSTAT204 potentiostat/galvanostat with a three-electrode-cell system, glassy carbon electrode as the working electrode, Hg/HgCl 2 electrode as the reference electrode, platinum wire as the counter electrode, and Bu 4 NPF 6 (0.1 M in acetonitrile) as the supporting electrolyte, with a scan rate of 100 mV s −1 in the range of −2 eV to 2 eV. GC-MS measurements were performed on a Shimadzu GC-2010 plus gas chromatograph and QP2010 ultra mass spectrometer with a fused silica column (Phenomenex, Zebron-5ms nonpolar) and flame ionization detector. 1 H and 13 C NMR spectra of all the compounds were measured using a Bruker Avance 300 MHz. Solid State 13 C CP MAS NMR measurements were carried out using a Bruker Avance II solid state NMR spectrometer operating at 300 MHz Larmor frequency equipped with a standard 4 mm magic angle spinning (MAS) double resonance probe head. FT-IR measurements were conducted with a Bruker Tensor II FTIR spectrometer. Bright field and fluorescence images were acquired on a Leica DMi8 inverted light microscope. The morphology was recorded with a scanning electron microscope (SEM) (LEO Gemini 1530, Germany) with an inlens SE detector. Thermogravimetric analysis (TGA) was conducted in a nitrogen atmosphere with temperature increasing from room temperature to 1000°C at a rate of 10 K min −1 . All DFT calculations were carried out using the Gaussian 09 package. 43 The structures were optimized at the B3LYP level of theory, 44 with the basis set of 6-31G*. 45,46 TD-DFT results were obtained from excited state calculations at the same level of theory. Time-resolved photoluminescence (TR-PL) spectra were obtained with a C4742 Hamamatsu streak camera system in slow sweep mode. Excitation pulses at 400 nm were provided by frequency doubling the output of a commercial femtosecond amplifier laser system (Coherent LIBRA-HE).

General procedures and synthesis
Synthesis of BTPh 2 .
A 250 mL three-necked flask was loaded with 4,7-dibromo-2,1,3-benzothiadiazole (1.47 g 5 mmol), phenylboronic acid (2.68 g, 22 mmol) and toluene (25 mL). Then, a solution of potassium carbonate (2.76 g, 20 mmol) in H 2 O (10 mL) was added into the flask. After degassing for 30 min, PdĲPPh 3 ) 4 (0.047 g, 0.04 mmol) was added, and the reaction mixture was heated to 90°C and reacted overnight under a N 2 atmosphere. After cooling to room temperature, the mixture was poured into water and extracted with dichloromethane. The organic layer was washed with water and dried over anhydrous MgSO 4 . After concentrating with a rotary evaporator, the crude product was purified via column chromatography on silica with dichloromethane as an eluent. For further purification, the crude product was recrystallized with methanol to afford BTPh 2 as yellow needles. Yield: 1.17 g (82%). 1

View Article Online
Synthesis of monomers.

Conflicts of interest
The authors declare no conflict of interest.