Hypercrosslinked Polymers as a Photocatalytic Platform for Visible‐Light‐Driven CO2 Photoreduction Using H2O

Abstract The design of robust, high‐performance photocatalysts is key for the success of solar fuel production by CO2 conversion. In this study, hypercrosslinked polymer (HCP) photocatalysts have been developed for the selective reduction of CO2 to CO, combining excellent CO2 sorption capacities, good general stabilities, and low production costs. HCPs are active photocatalysts in the visible light range, significantly outperforming the benchmark material, TiO2 P25, using only sacrificial H2O. It is hypothesized that superior H2O adsorption capacities facilitate access to photoactive sites, improving photocatalytic conversion rates when compared to sacrificial H2. These polymers are an intriguing set of organic photocatalysts, displaying no long‐range order or extended π‐conjugation. The as‐synthesized networks are the sole photocatalytic component, requiring no added cocatalyst doping or photosensitizer, representing a highly versatile and exciting platform for solar‐energy conversion.


Characterisation Chemical and structural properties
Solid-state NMR was carried out on a Bruker Avance NEO 500 wide bore system (Bruker BioSpin, Rheinstetten, Germany) using a 4 mm triple resonance magic angle spinning probe.
Around 15 to 25 mg of material was packed into a 4 mm zirconia CRAMPS rotor. The resonance frequency for 13 C NMR was 125.78 MHz, the MAS rotor spinning was set to 14 kHz. Cross polarization was achieved by a ramped contact pulse with a contact time of 3 ms.
During acquisition 1 H was high power decoupled using SPINAL with 64 phase permutations.
The 1 H π/2 pulse was 2.5 µs, the relaxation delay was set to 4 s, and with roughly 2000 scans a sufficient signal to noise could be achieved. The chemical shifts for 13 C are reported in ppm and are referenced external to adamantane by setting the low field signal to 38.48 ppm.
Fourier-transform infrared (FTIR) spectroscopy was performed in the range of 500 -4000 cm -1 on finely ground samples using a PerkinElmer Spectrum 100 FT-IR spectrometer equipped with an attenuated total reflectance (ATR) accessory.
Thermal analyses were performed using a Netzsch TG209 F1 Libra thermogravimetric analyser. At least 10 mg of sample was heated from room temperature to 900 °C at a rate of 10 °C min −1 under either air or N2 gas flow (flow rate 100 mL min −1 ). An initial isothermal step of 1 h was included at 120 °C to ensure removal of adsorbates before heating continued.
X-ray photoelectron spectroscopy (XPS) measurements were carried out on a Thermo Scientific K-Alpha+ X-ray photoelectron spectrometer equipped with a MXR3 Al Kα monochromated X-ray source (hν = 1486.6 eV). Samples were ground and mounted on the XPS holder using a conductive carbon tape. The X-ray gun power was set to 72 W (6 mA and S2 12 kV). Survey scans were acquired using 200 eV pass energy, 0.5 eV step size, and 100 ms (50 ms x 2 scans) dwell times. Data analysis was performed using the Thermo Avantage data analysis program.
Nitrogen isotherms were measured using a porosity analyser (Micromeritics 3Flex) at -196 o C.
Prior to measurement, all samples were degassed overnight at 393 K at around 0.2 mbar pressure. Samples underwent a further degas step at 393 K in-situ on the porosity analyser for 4 h, this time at around 0.003 mbar. Surface areas were calculated using the Brunauer−Emmett−Teller (BET) method. 1 The total volume of pores (VTOT) was calculated from the volume of N2 adsorbed at P/P0 = 0.97 and micropore volume (VMICRO) was determined using the Dubinin−Astakhov method. 2 The pore size distribution was derived from the adsorption isotherms by using an built-in software from Micromeritics and selecting the DFT model for carbon slit shape pores (N2@77 on Carbon Slit Pores by NLDFT).
Powder X-ray diffraction (PXRD) measurements were recorded at room temperature on a BRUKER 2D PHASE diffractometer operating at 30 kV and 10 mA with monochromatised Cu Kα radiation (λ = 0.15418 nm).
The morphology of the samples was imaged using a scanning electron microscope (SEM, Leo Gemini 1525, Zeiss) in secondary electron mode (InLens detector) at 5 kV. The samples were ground, deposited on carbon tape, and coated with 20 nm of chromium to reduce charging.
For determination of iron content in HCP-1,-2 and -3, 30 mg of sample were added to 8 mL of sulfuric acid. The resulting mixture was heated on a hot plate until the boiling point of sulfuric acid boiling was reached. Then, 4 mL of HClO4 was added dropwise to oxidise the sample to complete digestion. Digested samples were analysed using a Varian 720 ES with simultaneous ICP-OES.

S3
Elemental analysis was performed using a Eurovector EA 3000 CHNS-O Elemental Analyser.
Between 0.75 and 3.0 mg of each sample was weighed into tin vials (4×6 mm) for each individual run and each sample was ran at least in duplicate. Sample weighing is done using a micro balance (Sartorius, ME 5 OCE) to ensure accuracy. The operating temperatures for the combustion and reduction were 1000 °C (1480 °C for O analysis) and 750 °C, respectively, with high purity helium (99.999+) used as a carrier gas.

Optoelectronic properties
Valence band X-ray photoelectron spectroscopy (XPS) and work function measurements were carried out on a Thermo Scientific K-Alpha + X-ray photoelectron spectrometer equipped with a MXR3 Al Kα monochromated X-ray source (hν = 1486.6 eV). Samples were ground and mounted on the XPS holder using a conductive carbon tape. The X-ray gun power was set to 72 W (6 mA and 12 kV). Valence band spectra were obtained using 15 eV pass energy and 0.05 eV step size. Data analysis was performed using the software Thermo Avantage. The work functions of the polymers were determined by measuring the secondary electron cut-off in the low kinetic energy region. The sample holder contained a clean gold standard sample, which was used as a reference material to ensure correct calibration. A sample bias of -29.47 V was applied to the samples using an ion gun and the cut-off spectra were obtained using a pass energy of 10 eV. To account for potential variations across the surface of the material, the work function was measured at three different locations and the average was taken. A total standard deviation of ± 0.04 eV is associated to the band edge positions. To convert the valence band position and the work function to the absolute energy scale vs. vacuum with the redox potential scale vs. SHE, a factor of 4.44 was required, as 4.44 eV on the former corresponds to 0.00 V on the latter, at 25 °C.

S4
Diffuse reflectance ultraviolet-visible (DR-UV-Vis) spectra were obtained using a Perkin-Elmer Spectrum 100 Spectrometer equipped with an integrating sphere. Spectral band width was set to 2 nm, with Spectralon as a standard.
Time-correlated single photon counting (TCSPC) experiments were carried out using a commercial TCSPC setup (Horiba DeltaFlex) equipped with a pulsed LED excitation source (Horiba NanoLED series) and a fast rise-time photomultiplier detector (Horiba PPD-650 and PPD-900). The instrument response function (IRF) was measured at the wavelength of the excitation source (282 nm). During all other measurements, a suitable long pass filter was inserted between sample and detector to block off scattered excitation light.

Gas and water sorption
Water vapor, CO2 and H2 adsorption isotherms were collected at 25 °C using a Micromeritics isotherms. For H2O isotherms, miliQ water with a resistance > 18.2 micro-ohms was purified by 4 freeze pump thraw cycles. Water isotherms were collected up to a relative pressure of 0.8 to avoid condensation. For "wet" CO2 uptake, i.e. investigating CO2 uptake on HCPs preexposed to H2O, HCPs were exposed to humid air (>99 % humidity) by placing HCPs in a sealed vessel containing liquid water and a hygrometer for at least 48 hours at room temperature. HCPs were not in contact with the liquid water during this process. After removal, CO2 adsorption isotherms were performed at 25 °C up to 1 bar, skipping all prior degas steps.
The first pressure CO2 adsorption point was collected at around 10 mbar (~2 orders of S5 magnitude higher than a standard 'dry' measurement) to minimise water desorption. Resulting isotherms for wet polymers gave negative adsorption values at low absolute pressures due to some water desorption in the initial stages of measurement. Therefore, to allow comparison to dry samples, a factor was applied to the isotherm, raising the lowest absolute pressure measurement to 0 mmol/g adsorbed CO2. It is worth noting that some subsequent uptake may be due to re-adsorption of desorbed water.

Photocatalytic properties
H2O2 was detected using a colorimetric peroxide test stick from sigma Aldrich. 20 mg of HCP-1 was disperse into 12 mL of H2O. The photoreactor was vacuum 3 times and backfilled with Research grade (99.999%) CO2 for 1h to reach the adsorption/desorption equilibrium of CO2 on the catalyst surface. The photoreactor was then irradiated for 3 h under UV-vis light. After irradiation, HCP-3 was filtered and removed from the solution. A peroxide test stick was then dipped 1 seconds into the aqueous solution. In presence of H2O2 a colour change from white to blue can be observed (see Figures S16).
Formic acid and methanol were quantified using HPLC. After 3 h of UV-vis irradiation, the sample was dispersed in 2 mL of water for 10 min. After immersion, HCP-3 was filtered and removed from the aqueous solution. 20 µl of the filtrate was then analysed by HPLC    S14 Figure S8. PXRD patterns of a) HCP-1, b) HCP-2, and c) HCP-3. S15 Figure S9. CO2 adsorption isotherms at 298 K of both dry and wet (i.e. exposed to humid air for at least 48 h before measurement, 99% humidity) networks. a) HCP-1, b) HCP-2, and c) HCP-3. Filled symbols represent adsorption, empty symbols represent desorption.
S24 Figure S15. HPLC calibration curves to estimate the concentrations of formic acid and methanol produced during CO2 photoreduction. The data points obtained when using HCP-1 are indicated on the graphs.
S25 Figure S16. Results of the colorimetric test carried out to detect the presence of H2O2 as a result of the photocatalytic process. H2O2 was detected for both HCP-1 (top) and HCP-3 (bottom). A concentration of around 3 ppm was detected after 3 h of UV-vis irradiation for HCP-1 and of around 7 ppm for HCP-3.

Quantum efficiency (QE) calculations
The quantum efficiency at a given wavelength is defined as: . ℎ 100 (Equation S1) Based on the stoichiometry of the CO evolution redox reaction: . ℎ

(Equation S2)
For a polychromatic light source, we must consider the total number of absorbed photons across the wavelength range (270 -900 nm for UV-vis and 400 -900 nm for visible irradiation) and the corresponding total moles of CO evolved during the irradiation time: The number of photons absorbed by the photocatalyst at a given wavelength is given by: where (λ) denotes the intensity emitted from the irradiation source at a given wavelength and (λ) denotes the intensity after passing through the photocatalyst at the same wavelength .
The intensity of irradiation absorbed by the photocatalyst at a given wavelength is given by: which, using equation (S6), can be written as: Substituting equations (S11) and (S12) to (S3) gives us the final expression for the quantum efficiency for CO evolution: