Developing highly reversible Li-CO 2 battery: from on-chip exploration to practical application

Li-CO2 batteries (LCBs) hold significant potential for meeting the energy transition requirements and mitigating global CO2 emissions. However, the development of efficient LCBs is still in its early stages, necessitating...

wavelength laser and 1800 grating with a 50x long working distance lens.All Raman spectra were intensity normalized to the maximum peak height (maximum normalization).FTIR measurement was conducted on ThermoFisher Scientific, Nicolet iS50.The in situ atomic force microscopy (AFM) observations were conducted inside an argon atmosphere glovebox by a Bruker Multimode 8 system under Scansyst-fluid mode with an insulating silicon nitride tip.The on-chip Li-CO2 battery is cut off from the silicon wafer and situated at the bottom of the AFM fluid cell.0.1 mL electrolyte (1 M LiTFSI in TEGDME) is injected into the in-situ cell for each experiment.The electrodes are connected to a GAMRY potentiostat (Gamma 1000) for battery discharge/charge and CV test (Fig S10).The transmission electron microscopy (TEM) images and selected area electron diffractions (SAED) were collected by Talos™ F200i TEM (Thermo Scientific) using a 200 keV electron beam.

Electrochemical measurements
The galvanostatic discharge and charge test, which was used to measure the battery-specific capacity, cyclic stability, energy efficiency and rate performance, was conducted on a NEWARE battery tester (CT-4008Q-5 V100 mA).
Cyclic voltammetry (CV) was measured with a GAMRY potentiostat (Gamma 1000).The energy density of the Li-CO2 coin cells was calculated by the using the recorded discharging energy (final discharging state) dividing the catalyst weight of per electrode.

Computational methods
The first-principle calculations were conducted within the density functional theory (DFT) through the Vienna abinitio Simulation Package (VASP).The Perdew-Burke-Ernzerhof (PBE) exchange-correlation functional within the generalized gradient approximation was employed.The cut-off energy for the kinetic energy was set to 520 eV and a 3×3×1 Monkhorst-Pack k-mesh was applied for sampling in the Brillouin zone.The energy convergence tolerance and the force certification were set as 1 × 10 -5 eV/atom and 0.02 eV/Å, respectively.The DFT-D3 method with Becke-Jonson (BJ) damping was used for the Van der Waals (vdW) corrections.The absorption calculations were performed in the (111) plane of Pt.The general adsorption energies (Eads) were obtained through the following equation: Eads = Esubstrate-adsorbate -Esubstrate -Eadsorbate, where Esubstrate-adsorbate, Esubstrate, Eadsorbate represent the total energy of the substrate with adsorbed species, the substrate, and the molecule/atom (e.g., CO2, Li, Li2CO3), respectively.
The Gibbs free energy change (ΔG) was calculated based on the following equation: ΔG = ΔE + ΔZPE -TΔS, where ΔE is the reaction energy change, T is the temperature (298.15K), ΔZPE and ΔS denote the change in zero-pointenergy and entropy, respectively.For the demonstration of practical high loading of catalyst and carbon material as current collector, we used the spray coating techniques to load activated carbon (AC).The porous activated carbon (AC, specific surface ~ 2000 m 2 g -1 ) was mixed with carbon black and polyvinyl pyrrolidone (PVP) in isopropanol.A uniformly dispersed slurry can be obtained after ultrasonic dispersion.The slurry can be further spray-coated onto the on-chip devices with tailored mask and the electrode thickness can be controlled.After vacuum drying, the working electrode can be uniformly covered by the carbon material as current collectors.The areal mass loading of the AC-based on-chip LCB is around 0.5 mg cm -2 which is at the common level compared with literature.By physically mixing different catalyst with the carbon material slurry or chemically synthesizing catalyst onto the AC and then spray coating onto the devices, a configuration of catalyst on carbon material substrate can be achieved.In summary, the demonstrated on-chip platform can be also used for a wide range of catalyst with high loading and deliver integrated test and analysis functions.

Fig
Fig S1 Schematic of on-chip Li-CO2 batteries fabrication process.The (a) current collector, (b) cathode material, and (c) anode material are deposited on a 4-inch wafer in sequence.

Fig
Fig S2 Images of on-chip Li-CO2 devices at different fabrication steps: (i) after current collector and cathode material deposition, (ii) after anode deposition, (iii) after device encapsulation.

Fig
Fig S3 Scanning electron microscopy (SEM) images of deposited six different cathode films with the thickness of 100 nm on SiO2/Si wafer: (a) Pt, (b) Cu, (c) Ag, (d) Au, (e) Fe, and (f) Ni.In order to improve the film quality of the cathode material, a 10-nm-thick adhesion layer was first deposited before the cathode material was evaporated.

Fig
Fig S4 The X-ray diffraction (XRD) pattern of deposited six different cathode films with the thickness of 100 nm on SiO2/Si wafer: (a) Pt, (b) Cu, (c) Ag, (d) Au, (e) Fe, and (f) Ni.In order to improve the film quality of the cathode material, a 10-nm-thick adhesion layer was first deposited before the cathode material was evaporated.

Fig
Fig S5 On-chip Li-CO2 batteries electrochemical performances comparison.Discharge-charge curves of different cathode materials with a limit capacity of 5 μAh at the constant current of 1 μA: (a) Pt, (b) Cu, (c) Ag, (d) Au, (e) Fe, and (f) Ni.

Fig
Fig S6 In situ Raman spectra of Pt cathode recorded during the corresponding Galvanostatic discharge-charge process.Black, red, and blue curves represent the battery in open circuit potential state (OCP), discharge state, and charge state, respectively.

Fig
Fig S7 In situ FTIR measurement of the Pt-based on-chip LCBs.

Fig
Fig S8 In situ FTIR measurement of the Cu-based on-chip LCBs.

Fig
Fig S9 In situ FTIR measurement of the Ni-based on-chip LCBs.

Fig
Fig S11 In situ atomic force microscopy (AFM) images of Pt cathode in Ar atmosphere obtained at (a) OCP, (b) discharge to 2.0 V, (c) charge to 2.8 V, and (d) charge to 3.1 V.The white arrows represent the scanning direction.

Fig
Fig S12 Three possible reaction pathways for the formation of C and Li2CO3.The * represents the basal plane of Pt(111).

Fig
Fig S13 The schematic of three reaction pathways and the optimized structures of reactants and intermediates.* represents the Pt(111) substrate.

Fig
Fig S16 Images of commercial Li-CO2 battery test box: (a) front view, (b) top view.

Fig
Fig S18SEM images of pure CP, pristine Pt@CP, Pt@CP after discharging and Pt@CP after recharging.

Fig
Fig S19Raman spectra of Pt@CP at pristine state, after discharging state and after recharging state.

Fig
Fig S21 Spray coating porous activated carbon on the on-chip devices: optical images (left) and SEM images (right).
Fig S22 which confirms its functionality under high mass loading of catalyst.Fig S23 displays the Raman measurement of the AC-based on-chip devices before cycling, typical Raman spectra of activated carbon can be clearly observed.

Fig S22 .
Fig S22.Voltage curve of the AC-based on-chip LCB.

Fig S23 .
Fig S23.Raman measurement of the AC-based on-chip LCB (before cycling).

Fig S24 .
Fig S24.Recycling test of the on-chip Li-CO2 platform.(a) Digital image of on-chip LCB with spray coating AC as cathode electrode.(b) Digital image of the washed platform after testing.(c) Repeated coating the platform with activated carbon.(d) Voltage-curves of the first test and recycle test on the same platform.

Table S1
Electrochemical performance comparison of Li-CO2 batteries with different cathode catalysts.

Table S2
The Gibbs free energy change (ΔG) of each reaction step on the basal plane of Pt(111).

Table S3 .
Summary of suitable catalysts loading techniques for the demonstrated platform.