Ultrathin Phosphate‐Modulated Co Phthalocyanine/g‐C3N4 Heterojunction Photocatalysts with Single Co–N4 (II) Sites for Efficient O2 Activation

Abstract Realization of solar‐driven aerobic organic transformation under atmospheric pressure raises the great challenge for efficiently activating O2 by tailored photocatalysts. Guided by theoretical calculation, phosphate groups are used to induce the construction of ultrathin Co phthalocyanine/g‐C3N4 heterojunctions (CoPc/P‐CN, ≈4 nm) via strengthened H‐bonding interfacial connection, achieving an unprecedented 14‐time photoactivity improvement for UV–vis aerobic 2,4‐dichlorophenol degradation compared to bulk CN by promoted activation of O2. It is validated that more •O2 − radicals are produced through the improved photoreduction of O2 by accelerated photoelectron transfer from CN to the ligand of CoPc and then to the abundant single Co–N4 (II) catalytic sites, as endowed by the matched dimension, intimate interface even at the molecular level, and high CoPc dispersion of resulted heterojunctions. Interestingly, CoPc/P‐CN also exhibits outstanding photoactivities in the aerobic oxidation of aromatic alcohols. This work showcases a feasible route to realize efficient photocatalytic O2 activation by exploiting the potential of ultrathin metal phthalocyanine (MPc) assemblies with abundant single‐atom sites. More importantly, a universal facile strategy of H‐bonding‐dominating construction of MPc‐involved heterojunctions is successfully established.


Computation details
The equilibrium geometrical optimizations were completed using DFT/WB97XD method implemented in the Gaussian 09 package. [1,2] Gradient optimizations were carried out using the 6-31G* basis set for C, O, P, N, and H and LANL2DZ for metal atom(Co, Fe, Ni, Cu, Zn).
The stationary frequency calculations at 298.15 K and 1 atm were performed at the same level for each of the optimized structures to examine any imaginary frequency or the corresponding vibrational modes. TDDFT (time-dependent density functional theory) was employed to predict excited state energies and properties to obtain the information of charge transfer. [1] For giving more reliable interaction energies (Adsorption Energies, E ad, or Bonding energies, E b ), the Basis Set Superposition Error (BSSE) corrections are carried out. E ad or E b can be obtained according to the formula E ad /E b = E_corrected-(E(1) + E(2) +... E(j) ... +E(n) ), in which E_corrected = E_complex + E_BSSE and E(j) means the energy doing monomer centered basis set (MCBS) calculation for fragment j.

Chemicals
All phthalocyanines CoPc (II), FePc (II), CuPc (II), ZnPc (II), NiPc (II) and H 2 Pc were purchased from Sigma-Aldrich and other chemicals (analytical grade) were purchased from Aladdin. All reagents were directly used without further purification. Deionized water was used throughout.

Synthesis of bulk CN:
The bulk CN was simply obtained by the calcination of melamine. [3] In a typical experiment, a proper amount of melamine was taken in the ceramic crucible and heated in the muffle furnace at 500 °C for 4 h at a temperature ramp of 5 °C min −1 .

Synthesis of ultrathin CN nanosheets:
Firstly, melamine (10 g) and cyanuric acid (4 g) were dispersed in (500 mL) deionized water. The mixed suspension of cyanuric acid and melamine was stirred for 3 h at 80 °C, then cooled to room temperature, centrifuged at 4000 rpm for 15 min, washed with deionized water and ethanol several times, then dried overnight at 80 °C.
The obtained dried powder was transferred into a closed quartz boat and heated from room temperature to 550 °C with a heating rate of 5 °C min −1 under air atmosphere for 4 h, which product then undergone a second-time calcination treatment in the semi-closed porcelain boat at 500 °C for 2 h under air atmosphere. After natural cooling to room temperature, asobtained fluffy product was treated by HNO 3 solution then dried overnight at 80 °C to obtain final CN nanosheets.

Sample characterization
The X-ray powder diffraction (XRD) patterns of the samples were measured with a Bruker D8 Advance diffractometer, using Cu Ka radiation. Thermal gravimetric (TG, Netzsch, STA

Analysis of hydroxyl radicals
The sample (0.05 g) was dispersed in (50 mL) coumarin aqueous solution (0.001 mol L -1 ) in a beaker. Prior to irradiation, the reactor was magnetically stirred for 10 min to attain an adsorption-desorption equilibrium in dark. After irradiation for 1 h with a 150 W Xenon lamp (GYZ220 made in China), the sample was centrifuged, and then a certain volume of solution was transferred into a Pyrex glass cell for the fluorescence measurement of 7-hydroxycoumarin at 332 nm excitation wavelength with an emission peak wavelength at 460 nm through the above mentioned spectrofluorometer.

Evaluation for O 2 temperature-programmed desorption (O 2 -TPD)
The O 2 -TPD was measured with a home-built flow apparatus. The typical method is as follows: the sample (50 mg) was pre-heated to 300 °C maintaining for 0.5 h to remove any moisture and then cooled to room temperature under an ultra-high-pure He stream with a flow rate of 30 mL min −1 . After that, the system was cooled to room temperature and then the sample was blown continuously with O 2 for 60 min at 30 °C. The excess weakly adsorbed O 2 was removed by exposure to ultra-high pure He. Then the temperature was increased to 500 °C with a heating rate of 10 °C min -1 under pure He.

Photoelectrochemical (PEC) and electrochemical (EC) experiments
Photoelectrochemical (PEC) and electrochemical (EC) reduction measurements were carried out in a traditional three-electrode system. The prepared sample was used as working electrode, a platinum plate (99.9%) was used as the counter electrode, a saturated KCl Ag/AgCl electrode was used as the reference electrode, and (0.5 mol L -1 ) Na 2 SO 4 solution as the electrolyte. High purity nitrogen gas (99.999%) was bubbled through the electrolyte before and during the experiments. PEC experiments were performed in a quartz cell using a 500 W xenon lamp with a cut-off filter (λ > 420 nm) as the illumination source. An IVIUM V13806 electrochemical workstation was employed to test photoelectrochemical and electrochemical performance of a series of catalysts. All the experiments were performed at room temperature (about 25 °C).

EPR measurements
The •O 2 − radicals usually can be trapped by 5,5-dimethyl-l-pyrroline N-oxide (DMPO), producing the EPR signals of their adducts. The sample (5 mg) was dissolved in DMPO solution with CH 3 OH as solvent to obtain the liquid mixture. After illumination, the mixture was characterized using a Bruker EMX plus model spectrometer operating at room temperature.

Photocatalytic degradation of organic contaminants:
The photocatalytic activities of the as- isopropyl alcohol (IPA) was added in the photocatalytic system, respectively, to find out the effects of the corresponding active species on the photocatalytic reaction. Additionally, the EDTA (for aquaous system) and TEA (for organic system) are applied as h + scavenger, BQ as •O 2 − radical scavenger and IPA as •OH radical scavenger, respectively.

Statistical analysis
All the data were presented as means ± standard deviation (SD). In order to test the significance of the observed differences between the study groups, analysis by variance (ANOVA) statistics was applied and a value of P < 0.05 was considered to be statistically significant.      The XPS survey spectra of CN, CoPc, 9P-CN, 0.5CoPc/CN and 0.5CoPc/9P-CN were displayed in Figure S6a. For CN, the XPS spectra of C 1s exhibit two main peaks at 288.0 and 284.6 eV, respectively, as normally assigned to N-C=N and C-C ( Figure S6b). [4] By contrast, both the C 1s peaks for 9P-CN appear shift to lower binding energies, while the N 1s peaks show no obvious change compared with those for CN. This indicates the phosphate modification mainly affect carbon in CN, which is consistent with the theoretical model for P-CN since the hydroxyl groups locate on the C atoms. Furthermore, the loading of CoPc does not obviously change the C 1s and N 1s peaks of CN or 9P-CN, which reflects that no strong covalent interaction appear between CoPc and support. In our work, Raman and FT-IR along with DFT calculations have validated the interaction between CoPc with support is H-bonding as described in the main text. The Co 2p spectra for pristine CoPc, 0.5CoPc/CN and 0.5CoPc/9P-CN have been refined and illustrated, respectively, in Figure S6e. For pristine CoPc, two Co2p peaks at 795.6 eV and 780.6 eV along with the satellite peaks are attributed to Co 2p 1/2 and Co 2p 3/2 transitions, respectively. Compared with pristine CoPc, the intensities of Co 2p peaks for 0.5CoPc/CN and 0.5CoPc/9P-CN obviously decrease, which is due to the tiny CoPc as-loaded on CN and P-CN. Moreover, negligible Co 2p peak shifts could be observed, indicating the interfacial interaction between CoPc and CN or P-CN did not significantly affect the coordination environment of cobalt atoms. This is in agreement with the H-bonding-dominated model for CoPc/CN and CoPc/P-CN.       The optical band gap energy (E g ) of a semiconductor material could be evaluated by the following formula [5] : αhν = A(hν -E g ) n/2 where α, h, ν, A, and E g represent the absorption coefficient, planck constant, light frequency, proportionality and band gap energy, respectively.
From Figure S13a inset, the E g value for CoPc is calculated to be 2.44 eV. The Mott-Schottky curve for CoPc was analyzed by the impedance-potential measurements, [6][7][8] as shown in Figure S13b, which was determined to be approximately +0.85 eV for the HOMO level of CoPc. By the following empirical equations [9,10] : E VB =E CB +E g The LUMO level of CoPc is calculated to be -1.39 eV. Furthermore, the work functions of 9P-CN and CoPc were shown in Figure S13c.         Note: a) Photoelectrons from MPc to support versus from support to MPc. b) CBZ is short for carbamazepine. c) RhB is short for rhodamine B. d) AR1 is short for acid red 1.
Five indexes (first row and column 3−7 of Table S2) have been set for evaluating the scientific depth of individual work. In terms of the interfacial interaction between MPc and support material, most works have not clarified or evidenced the specific interactions and residue works have adopted the covalent-bonding to load MPc (entry 7−10), which synthetic processes are relatively complex. Noteworthily, our work is the first one to clearly validate the H-bonding interfacial interaction, especially which has been achieved by a facile phosphate-induction resulting in a high dispersion of CoPc. Secondly, in terms of the investigation on the photocatalytic mechanism, among all works our one has firstly clearly clarified the dual charge separation modes with the support of experimental evidences like EPR results, etc as well as DFT calulations. Furthermore, the photoactivity contribution corresponding to the charge separation mode has been individually illustrated. More important, few works focus on utilizing the single metal centers in MPc as the catalytic sites and only in our work the single-atom Co-N 4 (II) sites were highlighted and well studied by EPR and DFT calculations. Finally, what's the most unique aspect of our work is the different application, which could be found the photocatalytic O 2 activation especially the successful aerobic selective alcohol oxidation has been realized for the first time among similar photocatalytic systems. Based above, it's concluded that among similar photocatalytic systems our work is against the grain and has achieved deeper scientific recognization.  Note: a) Selectivity towards the main product benzylaldehyde.
For further highlighting the advancement of our work, we have compared the photocatalytic performance with reported CN-based ones for the aerobic oxidation of benzyl alcohol as the benchmarked reaction and shown in Table S3. Noteworthily, without additives like H 2 O 2 as entry 1, or high temperature and high O 2 pressure like entry 2, 1.8CoPc/12P-CN still showed impressive benzyl aldehyde yield, which also has higher selectivity over entry 3. Therefore, for CN based photocatalysts for aerobic alcohol oxidation, CoPc/P-CN photocatalysts have exhibited great potential to achieve highly active and selective conversion under room temperature as well as atmospheric pressure.