Enhancing H2 evolution performance of an immobilised cobalt catalyst by rational ligand design† †Electronic supplementary information (ESI) available: Additional figures and tables, synthetic procedures, experimental details for NMR and UV-vis spectroscopy, electrochemistry and photocatalytic experiments. See DOI: 10.1039/c4sc03946g

Rational ligand design was employed to improve the proton reduction activity of an immobilised cobalt diimine–dioxime catalyst.


Immobilisation of Molecular Catalysts on ITO|mesoITO Electrodes
ITO|mesoITO electrodes were prepared as previously reported. 1a 2 µL of an ITO nanoparticle suspension (Sigma Aldrich, diameter < 40 nm, BET = 27 m 2 g -1 , 90% In 2 O 3 , 20wt% in 5 M acetic acid in ethanol) were drop-casted onto ITO-coated glass slides (Vision Tek System Ltd. 30 Ω sq -1 , 1 x 2 cm 2 ) with Scotch ® tape (3M) as spacers (0.5 x 0.5 cm 2 ). After drying in air, the slides were annealed at 350 ºC for 20 min (heating rate to 350 ºC: 4 ºC min -1 ). The geometrical surface area of the mesoporous ITO coating was 0.25 cm 2 (thickness: 13 µm) 1a and the electrodes were cleaned with 2-propanol and acetone and dried under a stream of N 2 prior use. The ITO|mesoITO electrodes were then immersed into a 6 mM solution of the catalyst (CoP 1 , CoP 2 or CoP 3 ) in dry DMF for 15 h to allow for adsorption of the catalyst.
The hybrid electrodes were gently rinsed with dry DMF and dried under a stream of N 2 .

Photocatalytic Experiments
Photocatalytic experiments were performed using a Newport Oriel solar light simulator (100 mW cm -2 , AM 1.5G). The light source was equipped with a water filter to remove IR irradiation and a 420 nm cut-off filter to eliminate UV irradiation if required. Samples were prepared by sonicating 5 mg of nanoparticles (TiO 2 or ZrO 2 ) in an appropriate volume of buffered solution (AA, 0.1 M, pH 4.5 or TEOA buffer, 0.1 M, pH 7) for 10 min followed by addition of the catalyst (CoP 1 , CoP 2 or CoP 3 , 1 mM in water). After stirring the resulting suspension for 10 min, the RuP dye (1 mM in water) was added, the sample vial sealed and purged with N 2 containing 2% CH 4 as internal GC standard for 10 min. Solution samples were prepared as described above without addition of any particles. The total volume of the suspension/solution was 2.25 mL and the temperature of the photoreactor was kept at 25 ºC with a water-jacketed and temperature-controlled water bath during the experiment. The remaining headspace (5.59 mL) of the photoreactor was analysed by gas chromatography S4 (GC, Agilent 7890A Series) in regular time intervals. The GC was equipped with a 5 Å molecular sieve column (45 °C) and a thermal conductivity detector. N 2 was used as carrier gas (flow rate: 3 mL min -1 ). All experiments were repeated at least three times. The mean value and standard deviation σ were calculated. A minimum σ of 10 % was assumed.

Quantification of Attachment of CoP 3 and RuP to TiO 2 Nanoparticles
TiO 2 nanoparticles (5 mg) were sonicated in 2.15 mL of TEOA buffer for 10 min, followed by addition of CoP 3 (0.1 µmol, 1 mM in water). After stirring the mixture for 10 min, the suspension was centrifuged (8000 rpm, 10 min), the supernatant separated from the nanoparticles and centrifuged again (8000 rpm, 10 min). The UV-vis spectrum of the supernatant was recorded and compared to the UV-vis spectrum recorded prior to addition of TiO 2 . The loading of RuP onto TiO 2 particles was studied in the presence of CoP 3 (0.1 µmol each). Samples were prepared as described above and the suspension was stirred for further 10 min after the addition of RuP (0.1 µmol, 1 mM in water). If the centrifuged particles were used in photocatalysis, they were re-dispersed in 2.25 mL of a fresh TEOA buffer.

Quantum efficiency measurement
The external quantum efficiency (EQE) was determined for RuP|TiO 2 |CoP 3 in pH 7 TEOA solution and the sample was prepared following the standard procedure as described above.
The photoreactor was purged with N 2 (2% CH 4 as internal standard) for 10 min, followed by irradiation with blue light (λ = 465 nm) from an Ivium modulight LED light source. The light intensity was I = 22 mW cm -2 and the irradiated area was A = 3.6 cm 2 . The headspace gas in the reactor was analysed by GC. The following equation was used to determine the EQE from the amount of H 2 produced after 1 h irradiation: where n(H 2 ) = moles of H 2 produced, N A = Avogadro constant, h = Planck constant; c = speed of light and t = irradiation time. Note that the obtained EQE is a lower limit of quantum efficiency of the system, as it was assumed that all incident light was absorbed by the suspension.   and diethyl phosphite (1.4 mL, 11.7 mmol) were added and the reaction mixture was refluxed under N 2 for 48 h. After cooling to r.t., the reaction mixture was filtered and concentrated to dryness under reduced pressure. The crude product was purified by column chromatography (silica, chloroform followed by a chloroform/methanol gradient of 0 to 3% methanol).      Tables   Table S1.    The I cat /I P ratio was determined for the peak current I p of the non-catalytic Co III /Co II reduction at a scan rate of 20 mV s -1 . For I cat the peak current of the reduction wave following the Co II /Co I reduction was determined. b At pH values above 4, the Co III /Co II couple in CoP 2 becomes broad and irreversible preventing the reliable determination of I p .    Figure S1. 1 H NMR spectrum of compound 1 in CDCl 3 . Figure S2. 1 H NMR spectrum of compound 2 in CDCl 3 . S17 Figure S3. 13 C and 31 P NMR spectra of compound 2 in CDCl 3 . Figure S4. 1 H NMR spectrum of compound 3 in CDCl 3 . S18 Figure S5. 13 C and 31 P NMR spectra of compound 3 in CDCl 3 .