Visible‐Light‐Driven Hydrogen Evolution Using Planarized Conjugated Polymer Photocatalysts

Abstract Linear poly(p‐phenylene)s are modestly active UV photocatalysts for hydrogen production in the presence of a sacrificial electron donor. Introduction of planarized fluorene, carbazole, dibenzo[b,d]thiophene or dibenzo[b,d]thiophene sulfone units greatly enhances the H2 evolution rate. The most active dibenzo[b,d]thiophene sulfone co‐polymer has a UV photocatalytic activity that rivals TiO2, but is much more active under visible light. The dibenzo[b,d]thiophene sulfone co‐polymer has an apparent quantum yield of 2.3 % at 420 nm, as compared to 0.1 % for platinized commercial pristine carbon nitride.

General methods: All reagents were obtained from Sigma-Aldrich or from TCI Europe and used as received, except for 2,7-dibromo-N-methylcarbazole, 2,7-dibromo-9,9dimethylfluorene, and 3,7-dibromodibenzo[b,d]thiophene 5,5-dioxide which were synthesized using literature procedures. 1,2 Graphitic carbon nitride (Nicanite) was obtained from Carbodeon Ltd Oy and used as received. Water for the hydrogen evolution experiments was purified using an ELGA LabWater system with a Purelab Option S filtration and ion exchange column (ρ = 15 MΩ cm) without pH level adjustment. Reactions were carried out under nitrogen atmosphere using standard Schlenk techniques. 1 H and 13 C NMR spectra were recorded on Bruker Avance 400MHz NMR in CHCl 3 at 25 ºC. 13   were obtained with a 13 C rf field of 60 kHz, while the 1 H rf field amplitude was ramped to obtain maximum signal at a 1 H rf field of approximately 60 kHz, and a contact time of 2 ms. For 13 C spectra, 2048-4096 scans were accumulated with a 3 s recycle delay. The 13 C chemical shifts were referenced to the CH carbon of adamantane at 29.45 ppm. 4 Samples were packed in a zirconia rotor with a KelF cap (samples for NMR spectra recorded at ν r = 6 and 7 kHz spectra were centre packed with a PTFE plug), and NMR data were obtained and analysed using TopSpin 3.2.

Hydrogen evolution experiments:
A flask was charged with the polymer powder (25 mg), water (7.5 mL), triethylamine (7.5 mL), methanol (7.5 mL), and sealed with a septum. The resulting suspension was ultrasonicated until the photocatalyst was dispersed before degassing by N 2 bubbling for 30 minutes. The reaction mixture was illuminated in a 90° angle with a 300 W Newport Xe light-source (Model: 6258, Ozone free) for the time specified. The lamp was cooled by water circulating through a metal jacket. Gas samples were taken with a gas-tight syringe, and run on a Bruker 450-GC gas chromatograph equipped with a Molecular Sieve 13X 60-80 mesh 1.5 m × ⅛" × 2 mm ss column at 50 °C with an argon flow of 40.0 mL min -1 . Hydrogen was detected with a thermal conductivity S3 detector referencing against standard gas with a known concentration of hydrogen. Hydrogen dissolved in the reaction mixture was not measured and the pressure increase generated by the evolved hydrogen was neglected in the calculations. The rates were determined from a linear regression fit and the error is given as the standard deviation of the amount of hydrogen  terminated; minor series bearing either phenyl or the arene co-monomer as end-groups were also present, and debromination seems to occur either during the reaction or during work-up.
No fragments corresponding to homocoupling were observed. The MALDI-TOF data confirm that the polymerizations all occurred as expected, but do not provide a quantitative measure of molecular weight for these insoluble polymers. 5

Apparent Quantum Yields
For determining the apparent quantum yield, the photocatalyst (P1K, P6, P7 and CP-CMP10; 40 mg) was suspended in water (12 mL), methanol (12 mL) and TEA (12 mL). In the case of g-C 3 N 4 , the photocatalyst was suspended in water (35 mL) and triethanolamine (5 mL) and loaded with Pt nanoparticles (3 wt.% from H 2 PtCl 6 solution). Hydrogen evolution was studied using a focused λ = 420 nm LED (67.7 mW cm -2 ) controlled by an IsoTech IPS303DD Power Supply. An area of 5.73 cm 2 was illuminated and the light intensity was measured with a ThorLabs S120VC photodiode power sensor controlled by a ThorLabs PM100D Power and Energy Meter Console. The apparent quantum yields were estimated using Equation (1):

Absorption on-set
We calculated the lowest vertical singlet-singlet excitation energy (LVEE), for conjugated para- The lowest energy conformer for each of the oligomers was subsequently optimized using ground state density functional theory (DFT) and their vertical absorption spectra calculated using TD-DFT.

Thermodynamic ability of polymers to drive water splitting half-reactions
For the different oligomers to be able to act as overall water splitting photocatalysts, they need to be able to thermodynamically drive both the reduction of protons and the oxidation of water. Following our previous work on poly(p-phenylene) 16 and carbon nitride, 17 we calculated for the different oligomers the standard reduction potentials of half-reactions in which the exciton, formed upon absorption of light, donates either an electron or a hole (half reactions 1 and 2). We also calculated similar standard reduction potentials for free electrons and holes that can be formed through dissociation of the exciton (half reactions 3 and 4). The relevant half-reactions, written following convention in the form of reductions, are: S44 P + e ! ⇄ P ! (3) Here P stands for the neutral oligomer in its electronic ground state, P * the oligomer in its lowest electronically excited state, and P + and Pfor the oligomer with a free hole and free electron respectively. In half reactions 1 and 3, the polymer donates an electron, i.e. act as a reductant, and the reaction in practice will run in the opposite direction. The potentials for the oligomer half-reactions are than compared with the potentials of the proton reduction and water oxidation half-reactions: Again written as reductions. When the standard reduction potentials of half-reactions 1 (IP * ) and/or 3 (EA) are more negative than that of half-reaction 5, proton reduction is thermodynamically feasible.
Similarly, when the standard reduction potentials of half-reactions 2 (EA * ) and/or 4 (IP) are more positive than that of half-reaction 6, water oxidation is thermodynamically feasible.
We also considered the half-reaction corresponding to the 2-electron oxidation of triethylamine, the sacrificial hole-acceptor used experimentally, which, when written as a reduction, is: Oxidation of triethylamine is thermodynamically feasible when the standard reduction potentials of half-reactions 2 and/or 4 are more positive than that of half-reaction 7.
The standard reduction potentials for half-reactions 1-7 were calculated using a similar set-up as the absorption on-set calculations (see above) with one important difference. We perform our calculations using the relative dielectric permittivity of water (80.1) rather than the chloroform value in the COSMO solvation model. Use of the dielectric permittivity of methanol or triethylamine instead would only make a small difference as shown in our previous work on poly(p-phenylene). 16 The potentials for the half-reactions 5-7 by necessity include nuclear relaxation and even a vibrational free-energy contribution. As found in our previous work, 16,17 the magnitude of the contribution of vibrational free-energy to the potentials of half-reactions 1-4 is small, because of the structural similarity between the compounds on either side of the reaction arrow, and large for half-reactions 5-

7.
Finally, just as in our previous work, 16,17 for computational reasons we used for the potential of half-reaction 5, the experimentally measured absolute potential of the standard hydrogen electrode. 18 predicted to be considerably more negative than the proton reduction potential and hence all oligomers should be able to reduce protons to hydrogen. In contrast, water oxidation is predicted to be generally endothermic, where for most oligomers studied, the EA * and IP potentials are predicted to lie below (i.e., be more negative than) the water oxidation potential. Finally, in line with the successful use of triethylamine as sacrificial electron donor, triethylamine oxidation is overall predicted to be exothermic, where the calculated EA * and IP potentials are considerably more positive than the triethylamine oxidation potential (the same also holds for methanol oxidation, not shown, though there the overpotential is smaller as the methanol oxidation reaction is predicted to have a potential of +0.29 V vs. +0.03 V in the case of triethylamine).