paper8-si-advmat-2023-cu-dendrites-plasmonic-her-cu2o

Raw source: doc_691593457211_8-adma202305742-sup-0001-suppmat.pdf

==> picture [222 x 63] intentionally omitted <==

Supporting Information

for Adv. Mater. , DOI 10.1002/adma.202305742

In-Situ-Grown Cu Dendrites Plasmonically Enhance Electrocatalytic Hydrogen Evolution on O Facet-Engineered Cu2

Hao Zhang*, Jiefeng Diao, Yonghui Liu, Han Zhao, Bryan K. Y. Ng, Zhiyuan Ding, Zhenyu Guo, Huanxin Li, Jun Jia, Chang Yu, Fang Xie, Graeme Henkelman, Maria-Magdalena Titirici, John Robertson, Peter Nellist, Chunying Duan, Yuzheng Guo*, D. Jason Riley* and Jieshan Qiu*

Supporting Information

In-Situ-Grown Cu Dendrites Plasmonically Enhance Electrocatalytic Hydrogen

Evolution on Facet-Engineered Cu2O

Hao Zhang[1,2,] , Jiefeng Diao[3] , Yonghui Liu[4] , Han Zhao[5] , Bryan K. Y. Ng[6] , Zhiyuan Ding[7] , Zhenyu Guo[8] , Huanxin Li[9] , Jun Jia[4] , Chang Yu[11] , Fang Xie[1] , Graeme Henkelman[3] , Maria-Magdalena Titirici[8] , John Robertson[4,9] , Peter Nellist[7] , Chunying Duan[10] , Yuzheng Guo[4,] , D. Jason Riley[1,* ] Jieshan Qiu[11,* ]

1Department of Materials and London Center for Nanotechnology, Imperial College London, London SW7 2AZ, UK. E-mail: jason.riley@imperial.ac.uk

2Chemistry Research Laboratory, Department of Chemistry, University of Oxford, Oxford OX1 3TA, UK. E-mail: hao.zhang@chem.ox.ac.uk

3Department of Chemistry and the Oden Institute for Computational Engineering and Sciences, The University of Texas at Austin, Austin TX 78712, USA.

4School of Electrical Engineering and Automation, Wuhan University, Wuhan 430072, P. R. China. E-mail: yguo@whu.edu.cn

5Department of Chemistry, University of Zurich, Winterthurerstrasse 190, CH-8057 Zurich, Switzerland.

6Wolfson Catalysis Centre, Department of Chemistry, University of Oxford, Oxford OX1 3QR, UK.

7Department of Materials, University of Oxford, Oxford OX1 3PH, UK.

8Department of Chemical Engineering, Imperial College London, London SW7 2AZ, UK.

9Department of Engineering, University of Cambridge, Cambridge CB2 1PZ, UK.

10School of Chemistry, Dalian University of Technology, Dalian, Liaoning 116024, P. R. China.

11State Key Lab of Fine Chemicals, School of Chemical Engineering, Liaoning Key Lab for Energy Materials and Chemical Engineering, Dalian University of Technology, Dalian 116024, P. R. China. E-mail: jqiu@dlut.edu.cn

1

Experimental Section

Materials and Methods

Materials: The following reagents were obtained and used as received without further processing: copper(II) chloride dihydrate (≥ 99%), sodium hydroxide (> 96%), hydroxylammonium chloride (>99.99%) and sodium dodecyl sulfate (SDS, 92.5-105%) were bought from Sigma-Aldrich. Reverse Osmosis water (> 18.2 MOhms cm[-1] ) was utilized in all experimental procedures. Preparation of C-Cu2O, T-Cu2O and O-Cu2O: 92.75, 92.25 and 91.75 mL of deionized water were added to a 100 mL beaker for the synthesis of C-Cu2O, T-Cu2O and O-Cu2O, respectively. 5 mL of 0.1 M CuCl2 solution and 0.8650 g SDS were added to each beaker and the prepared solution was sonicated for 10 min to form a homogeneous light blue solution. 2 mL of 2.0 M NaOH solution was subsequently added dropwise to the solution, and the prepared solution was sonicated for 10 min to obtain a blue flocculent suspension. 0.25, 0.75 and 1.25 mL of 1.0 M NH2OH HCl were added to each beaker for the synthesis of C-Cu2O, T-Cu2O and O-Cu2O, respectively. The mixture was sonicated for an additional 1 h, and the temperature was controlled in the range of 30~40[o] The resulting precipitates were collected by centrifugation at 3500 rpm for 7 min and washed 3 times with a mixture of deionized water and ethanol to remove residual chemicals and SDS. The obtained red powders were dried overnight at 60 °C.

Characterizations

Powder X-ray diffraction (XRD) patterns were collected on a Bruker D2 ADVANCE diffractometer with Cu Kα radiation (λ=1.5418Å). The structure and morphology of the samples were characterized by field-emission scanning electron microscopy (FESEM, Zeiss LEO 1525), transmission electron microscopy (TEM, JEOL 2100F) and scanning transmission electron microscopy (STEM, JEOL ARM300CF). Energy-dispersive X-ray spectroscopy (EDS) attached to the TEM was used to analyze the composition of the nanoscale samples. The pore structures of samples were characterized using the N2 adsorption/desorption isotherm tested on the Micromeritics 3 Flex Physisorption at 77 K. The specific surface area was determined by the multi-point BrunauerEmmett-Teller (BET) method and the pore-size distribution was calculated based on the BarrettJoyner-Halenda (BJH) method. X-ray photoelectron spectroscopy (XPS) analysis was conducted on a PHI-5000 VersaProbe X-ray photoelectron spectrometer using an Al Kα X-ray source. The powder

2

samples were stuck onto the specific sample holders using conductive double-sided carbon tapes for the test. X-ray absorption spectroscopy (XAS) analysis, including the X-ray absorption near edge structure (XANES) and extended X-ray absorption fine structure (EXAFS), was performed at the European Synchrotron Radiation Facility (ESRF), Swiss-Norwegian Beamline BM31, Grenoble, France. All the XAS data were collected via a transmission mode with a Si (111) double crystal monochromator, which was cooled with liquid nitrogen. All samples were characterized using Cu- K -edge XANES in transmission mode under ambient conditions with Cu foil and Cu2O as references. The EXAFS data analysis was performed using IFEFFIT with Horae packages. The fitting was done in R-space with k[3] -weighted data. The amplitude reducing parameter was obtained from fitting the Cu2O reference, which was used as a fixed input parameter in the data fitting to allow the refinement in the coordination number of the absorption element. The goodness of fit is evaluated by a low R- value. All EXAFS spectra are presented without phase correction. Operando Raman spectra was obtained with an inVia Renishaw confocal Raman microscope operated with an incident laser beam at 532 nm focused through a 50x objective (Leica). The laser intensity was set to < 1 mW and Raman spectra were collected in static mode, with an exposure time of a few seconds every 5 minutes to minimize sample heating. To monitor the evolution of catalyst samples during HER process, each Raman spectrum was collected after a constant potential was applied to the catalyst electrode for 5 min. Each Raman spectrum was obtained using an integration time of a few seconds, accumulating 5 times. The laser shutter remained closed between spectrum collections.

Electrochemical measurements

The (photo)electrochemical tests of the materials were performed using a Metrohm Autolab electrochemical workstation PGStat-12 (Utrecht, the Netherlands) connected to a three-electrode cell. The FTO glass of 1×2 cm[2] size coated with the Cu2O samples was served as the working electrode, and a graphite rod and a Hg/HgO/OH[-] electrode were employed as the counter electrode and reference electrode, respectively. 50 mg of the as-prepared Cu2O catalyst was dispersed in 4.5 mL of a water/isopropanol solution (1:3 in volume) containing 0.5 mL Nafion (5%). The resulting nanoobject suspension was sonicated for 0.5-1 h. The upper part (1×1 cm[2] ) of the FTO glass was covered with adhesive tape, leaving the well-defined lower part (1×1 cm[2] ) of the FTO glass to support the catalyst. The FTO substrate was first treated with UV-Ozone for 10 min to enhance the

3

substrate hydrophilicity, when the suspension was well dispersed, 60 μL of the above suspension was dropped onto the clean FTO glass evenly, the electrode was kept at ambient temperature for 12 h to evaporate the solvent and inhibit the formation of cracks.

For the comparison with the literature, a glassy carbon electrode (GCE) with a diameter of 3 mm drop-coated with Cu2O samples was used as the working electrode, and a graphite rod and a Hg/HgO/OH[-] electrode were employed as the counter electrode and reference electrode, respectively. 5 mg of the as-prepared Cu2O catalyst was dispersed in 4.5 mL of a water/isopropanol solution (1:3 in volume) containing 500 μL Nafion (5%). The resulting nanoobject suspension was sonicated for 0.5-1 h. When the suspension was well dispersed, 4 μL of the above suspension was dropped onto the clean GCE for electrochemical studies.

An aqueous solution of 1 M KOH was the electrolyte for HER experiments. Cyclic voltammetry (CV) curves were recorded at a sweep rate of 100 mV s[−1] for multiple cycles. Linear sweep voltammetry (LSV) was carried out at a scan rate of 5 mV s[−1] for polarization curves. LSV was performed several times until the signals were stabilized. EIS was measured by applying an AC voltage of 5 mV amplitude in a frequency range from 0.01 Hz to 100 kHz. CV curves with different scan rates (10-60 mV s[-1] ) were measured over a potential range in which redox processes were absent to calculate the electrochemical double-layer capacitance: Cdl = (ja – jc)/(2v) = (ja + |jc|)/(2v) = Δj/(2v) , where Cdl is the double-layer capacitance (F cm[-2] ) of the electroactive materials, ja and jc is the anodic and cathodic current density (mA cm[-2] ), respectively, recorded at the middle of the selected potential range, and v is the scan rate (mV s[-1] ). Mass activity (A g[-1] ) was calculated from the catalyst loading density m (mg) and the measured current density j (mA cm[-2] ) following the equation: mass activity = j / m . Turnover frequency (TOF, s[-1] ) was calculated using the equation: TOF = ( j × A ) / (4 × F × n ), where j is the current density, A is the geometric area of electrode, F is the Faraday constant (96485 C mol[-1] ), and n is the moles of the corresponding metal atom (mol) within the catalyst loading.

All results reported in this work were converted to the RHE scale according to the Nernst equation,

==> picture [384 x 12] intentionally omitted <==

where ERHE is the measured potential referred to RHE, EHg/HgO/OH[-] is the working potential, and

4

E[0] Hg/HgO/OH[-] equals to 0.098 V at 25[o] C.

The HER activity was obtained after the iR -correction to the LSV. Typically, the iR -correction is according to the equation, E = ERHE - 90%iR , where E is the potential after iR -correction, ERHE is the measured potential referred to RHE, i represents the measured current, and R is the uncompensated resistance which could be determined by electrochemical impedance spectroscopy (EIS). The uncompensated resistance is found as the real impedance where the imaginary part of the impedance is zero in a Nyquist plot. The samples after HER were scratched from the FTO glass and dispersed in a water/isopropanol solution (1:3 in volume) for post-HER characterization.

Photoelectrochemical measurements for the fabricated photoanodes were performed by using a solar simulator. The light source was a 150 W Xe lamp equipped with an AM 1.5 G filter (Newport), and the light intensity at the surface of the electrode was 100 mW cm[−2] . The signal was recorded by using a potentiostat and a lock-in amplifier by filtering the dark current. The Cu2Obased FTO samples were illuminated from the front-side (Cu2O samples) to back-side (FTO substrate). The active surface area 0.28 cm[-2] . Incident photon-to-current conversion efficiency (IPCE) measurements were carried out under potential-controlled condition of 0 V versus RHE as applied bias, and IPCE values were calculated using the equation (2),

IPCE=1240 J / Pλ ×100% (2)

where J is the photocurrent density (mA cm[-2] ), P is the power of monochromatic light irradiate on the electrode (mW cm[-2] ) and λ is the wavelength (nm).

Computational details

Density functional theory (DFT) calculation was performed using the generalized gradient approximation (GGA) Perdew-Burke-Ernzerhof (PBE) functional, and the projected augmented plane-wave method implemented in the Vienna ab initio simulation program (VASP) software code.1 The original Cu2O crystal structure is found from Material Project, with verification to experimental XRD data. The Cu2O (111) and Cu2O (100) slab structures are both cut from the original Cu2O crystal structure and with geometry optimization. Surface energy is calculated by Esurface = (Eslab-NEbulk)/(2Area of surface) , whereas Eslab is the energy of the slab system either for Cu2O (111) or Cu2O (100), Ebulk is the energy of one Cu2O in bulk, and N is the number of Cu2O in each slab system. The model of Cu2O (111) after cycles contains one layer of Cu (111) on Cu2O

5

(111) slab structure with fully geometry optimization.

According to the Volmer-Heyrovsky mechanism for HER in alkaline solution, the reaction happening on Cu2O (111) and Cu2O (100) slab structures can be written as:

slab + 2H2O + 2e[-] → slab-H2O[] + H2O + 2e[-] → slab-H[] + OH[-] + H2O + e[-]

==> picture [215 x 11] intentionally omitted <==

The initial system consists of a Cu2O slab, two free H2O molecules and two extra electrons. The intermediate system after the first H2O adsorption consists of a Cu2O slab bounded to one H2O molecule, one free H2O molecule and two extra electrons. The intermediate system after Volmer step consists of a Cu2O slab bounded to H[] , one free OH[-] , one free H2O molecule and an extra electron. The intermediate system after the second H2O adsorption consists of a Cu2O slab bounded to one H[] and one H2O molecule on different sites, one free OH[-] , and an extra electron. The final system after Heyrovsky step consists of a Cu2O slab, two free OH[-] and a H2 gas molecule.

The Gibbs free energy of each system in HER process can be calculated as:

∆G = Esys + ∆ZPE - T∆S

where Esys is the DFT potential energy of each system, ∆ZPE is the zero-point energy, T is the temperature (300 K), ∆S is the entropy (since the entropy of liquid and solid is minimum, we only consider the entropy of the H2 gas molecule). In DFT calculation, the plane wave basis set cutoffs of the wavefunctions were set at 400 eV and zero damping DFT-D3 method was used to investigate weak intermolecular interactions. All the energy data collected were after full relaxation of the system until it had forces on each atom less that 0.01 eV Å[-1] and with an electron step convergence criterion of 10[-6] eV. Structures and charge transfer were draw with VESTA and Bader charge was calculated with Bader charge analysis program.

The simulated UV-Vis absorption spectra of Cu (111) facets are modeled on the structure of a Cu (111) slab. The modeled structure is optimized, and the frequency-dependent dielectric matrix is calculated after the electronic ground state has been determined. The simulated adsorption spectrum was calculated based on the frequency-dependent dielectric matrix, of which the imaginary part is determined by a summation over empty states.

Ion-electron interactions were characterized by the projector-augmented-wave (PAW) approach.2-5 A vacuum zone of more than 20 Å was used to prevent interactions between the nearest

6

neighboring units. Because the band gaps of nanostructures are significantly underestimated by the PBE functional, the band gap values of Cu2O are obtained using the DFT + U method, wherein the U value of Cu atoms is obtained.6 Convergence tolerances lower than 10[-6] eV and 10[-2] eV Å[-1] were respectively used for the total energy and force on each atom. The kinetic energy cutoffs for wave functions were set to 500 eV. The Brillouin zones were sampled with a 1 × 1 × 1 Γ-centered MP grid in structure optimization calculations.

Ab initio nonadiabatic molecular dynamics (NAMD) simulations were conducted using HefeiNAMD code.7-9 The heterojunction was built with an orthogonal supercell containing 148 atoms and simulated their dynamic process at the Г point. The speed rescaling method was used in the NVT simulation to heat the system to 300 K in the 5 ps process. Afterward, the NVE ensemble was used to acquire 5 ps trajectories with a step time of 1 fs. Finally, 2000 structures were extracted from the 5 ps trajectories for self-consistent calculation. The carrier recombination calculation originated from 50 random primary structures and 2000 trajectories sampled for each primary system.

7

==> picture [396 x 104] intentionally omitted <==

Figure S1. STEM-EDS spectra of (a) C-Cu2O, (b) T-Cu2O and (c) O-Cu2O.

8

==> picture [435 x 115] intentionally omitted <==

Figure S2. k-space FT-EXAFS spectra of Cu K -edge of (a) C-Cu2O, (b) T-Cu2O and (c) O-Cu2O.

9

Table S1. FT-EXAFS spectra fitting parameters of K -edge of C-Cu2O.

==> picture [355 x 115] intentionally omitted <==

Notes: Scattering paths are generated from Cu2O (point group: Pn3̅m, mp-361) from materials project (REF: https://doi.org/10.1063/1.4812323). CN, Coordination number. R, Radial distance. D- W factor, Debye-Weller factor. The amplitude reduction factor of 0.69 for all samples was obtained from fitting the Cu2O reference measured along with the samples.

10

Table S2. FT-EXAFS spectra fitting parameters of K -edge of T-Cu2O.

==> picture [379 x 122] intentionally omitted <==

11

Table S3. FT-EXAFS spectra fitting parameters of K -edge of O-Cu2O.

==> picture [370 x 120] intentionally omitted <==

12

Table S4. The overpotential and Tafel slope comparisons of C-Cu2O, T-Cu2O and O-Cu2O for electrochemical HER in alkaline electrolyte.

==> picture [415 x 101] intentionally omitted <==

13

==> picture [429 x 156] intentionally omitted <==

Figure S3. (a) Polarization curves and (b) Tafel plots of O-Cu2O on an FTO and GCE substrate, respectively in alkaline electrolyte.

14

==> picture [252 x 177] intentionally omitted <==

Figure S4. Polarization curves of bare FTO, Nafion-modified FTO and bare GCE electrodes in alkaline electrolyte.

15

==> picture [428 x 109] intentionally omitted <==

Figure S5. CV curves of (a) C-Cu2O, (b) T-Cu2O and (c) O-Cu2O at different scan rates.

16

Table S5. HER activity comparison among O-Cu2O and other reported Cu-based catalysts in alkaline electrolyte.

==> picture [416 x 210] intentionally omitted <==

17

==> picture [281 x 198] intentionally omitted <==

Figure S6. Polarization curves of the O-Cu2O samples on FTO substrates normalized to the electrochemical active surface in alkaline electrolyte.

18

==> picture [223 x 157] intentionally omitted <==

Figure S7. Exchange current density of the O-Cu2O samples on FTO substrates in alkaline electrolyte.

19

==> picture [377 x 175] intentionally omitted <==

Figure S8. (a) TEM image, (b) HAADF-STEM and elemental mapping images of C-Cu2O-A. (c) TEM image, (d) HAADF-STEM and elemental mapping images of T-Cu2O-A.

20

==> picture [259 x 181] intentionally omitted <==

Figure S9. Pore size distribution of O-Cu2O and O-Cu2O-A.

21

==> picture [370 x 142] intentionally omitted <==

Figure S10. (a) UV-Vis spectra and (b) tauc plots of O-Cu2O and O-Cu2O-A.

22

==> picture [379 x 147] intentionally omitted <==

Figure S11. k-space FT-EXAFS spectra of Cu K -edge of (a) O-Cu2O-A and (b) Cu2O reference.

23

==> picture [297 x 208] intentionally omitted <==

Figure S12. R-space FT-EXAFS spectra of Cu K -edge of Cu2O reference.

24

Table S6. FT-EXAFS spectra fitting parameters of K -edge of O-Cu2O-A.

==> picture [385 x 158] intentionally omitted <==

25

Table S7. The overpotential and Tafel slope comparisons of O-Cu2O and O-Cu2O-A for electrochemical HER in the absence or presence of light irradiation in alkaline electrolyte.

==> picture [418 x 125] intentionally omitted <==

26

==> picture [411 x 150] intentionally omitted <==

Figure S13. (a) Polarization curves and (b) Tafel plots for O-Cu2O in alkaline electrolyte in the absence or presence of light irradiation.

27

==> picture [273 x 192] intentionally omitted <==

Figure S14. Detected hydrogen products from electrochemical and photoelectrochemical HER for O-Cu2O and O-Cu2O-A.

28

==> picture [264 x 185] intentionally omitted <==

Figure S15. Simulated UV-Vis spectrum of Cu (111) facets (inset: the model of Cu (111) facets).

29

==> picture [267 x 115] intentionally omitted <==

Figure S16. Atomic structure evolution of Cu76-(Cu2O)24 at 0 K (top) and 300 K (Bottom). Structural properties of Cu76-(Cu2O)24 system

Notes: Figure S16 shows the geometry of the Cu76-(Cu2O)24 system relaxed at 0 K (top) and a representative geometry from the molecular dynamics (MD) trajectory run at 300 K (bottom). Thermal effects have little influence on the Cu2O nanosheet on the scale of the simulation cell. In contrast, the Cu nanorod undergoes significant deformations. The extended system experiences long-wavelength fluctuations, and the Cu nanorod ends undergo structural reconstruction. The Cu nanorod ends reconstruct to minimize the free energy of the nanorod itself. It is quite common that surfaces, edges, and tips of nanoscale structures undergo reconstruction to minimize the free energy, which can be demonstrated by bond relaxation theory.[[10-12] ]

30

==> picture [397 x 144] intentionally omitted <==

Figure S17. (a) Current density of O-Cu2O-A with different light intensity. (b) HER activity of O- Cu2O-A with different light intensity.

31

References

1. W. Tang, E. Sanville, G. Henkelman, A grid-based Bader analysis algorithm without lattice bias. J. Phys. Condens. Matter. 2009 , 21 , 084204.

2. G. Kresse, J. Furthmüller, Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. Phys. Rev. B 1996 , 54 , 11169-11186.

3. G. Kresse, J. Furthmüller, Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set. Comput. Mater. Sci. 1996 , 6 , 15-50.

4. P. E. Blöchl, Projector augmented-wave method. Phys. Rev. B 1994 , 50 , 17953-17979.

5. J. P. Perdew, K. Burke, M. Ernzerhof, Generalized Gradient Approximation Made Simple. Phys. Rev. Lett. 1996 , 77 , 3865-3868.

6. M. Heinemann, B. Eifert, C. Heiliger, Band structure and phase stability of the copper oxides Cu2O, CuO, and Cu4O3. Phys. Rev. B 2013 , 87 , 115111.

7. W. Chu, Q. Zheng, A. V. Akimov, J. Zhao, W. A. Saidi, O. V. Prezhdo, Accurate Computation of Nonadiabatic Coupling with Projector Augmented-Wave Pseudopotentials. J. Phys. Chem. Lett. 2020 , 11 , 10073-10080.

8. L. Zhang, W. Chu, C. Zhao, Q. Zheng, O. V. Prezhdo, J. Zhao, Dynamics of Photoexcited Small Polarons in Transition-Metal Oxides. J. Phys. Chem. Lett. 2021 , 12 , 2191-2198.

9. S. Gumber, S. Agrawal, O. V. Prezhdo, Excited State Dynamics in Dual-Defects Modified Graphitic Carbon Nitride. J. Phys. Chem. Lett. 2022 , 13 , 1033-1041.

10. Y. Liu, C. Shao, W. Yu, Q. Gui, J. Robertson, Y. Guo, Atomic-size dependence of the cohesive energy, bandgap, Young’s modulus, and Raman frequency in different MA2Z4: A bond relaxation investigation. Appl. Phys. Lett. 2022 , 121 , 244105.

11. Y. Liu, M. Bo, X. Yang, P. Zhang, C. Q. Sun, Y. Huang, Size modulation electronic and optical properties of phosphorene nanoribbons: DFT-BOLS approximation. Phys. Chem. Chem. Phys. 2017 , 19 , 5304-5309.

12. Y. Liu, M. Bo, Y. Guo, X. Yang, X. Zhang, C. Q. Sun, Y. Huang, Number-of-layer resolved (Mo, W)-(S2, Se2) phonon relaxation. J. Raman Spectrosc. 2017 , 48 , 592-595.

32