ALD/ALE 2025 Session AA2-WeA: Battery Applications II

Adopting lithium metal (Li) as anodes and nickel (Ni)-rich lithium nickel manganese cobalt oxides (LiNi_xMn_yCo_zO₂, NMCs, x ≥ 0.6, x + y + z = 1) as cathodes, the resultant Li||NMC lithium metal batteries (LMBs) could be twice higher in energy (up to 500 Wh/kg) but 50% lower in cost ($100/kWh) than that of LIBs, holding great promise to replace LIBs for the applications of portable electronics, electric vehicles, and aircrafts. Unfortunately, such a compelling technology has been hindered from commercialization due to some serious interfacial issues related to the Li anodes and NMC cathodes. Aimed at addressing these challenges, we recently have developed a series of novel coatings via atomic and molecular layer deposition (ALD and MLD). ALD and MLD share several unique merits but are complementary in their target materials. They have emerged as two new techniques of interface engineering of rechargeable batteries in the past decade.^1-5 They both could deposit conformal and uniform coatings over complex shapes of different substrates, operate at low process temperature, and accurately control coating thickness. Through adopting different precursors, ALD exclusively deposits inorganic films while MLD specially grows organic or hybrid films. For the issues of Li anodes and NMC cathodes, we particularly designed function-oriented coatings via ALD and MLD. Using our ALD and MLD processes, very encouragingly, both the surface-coated Li anodes and NMC cathodes have exhibited remarkable improvements in their electrochemical performance. Our studies have further shown that the combination of these coatings can synergistically maximize their benefits to achieve higher performance of Li||NMC LMBs, enabling a cell capacity fading 10 times slower than that of bare Li||NMC cells and a capacityretention improvement over 60%after 500 charge/discharge cycles. In this talk, we will introduce these novel coatings and their compelling effects. Particularly, we would like to explain the underlying mechanisms related to their benefits. Thus, our studies have not only opened new areas of surface coatings but also demonstrated their technical feasibility for developing high-performance LMBs.

References:

1. Adv. Mater. 2012, 24, (27), 3589-3615.

1. Energy Storage Materials 2020, 30, 296-328.

1. J. Mater. Chem. A 2017, 5, 10127-10149.

1. J. Mater. Chem. A 2017, 5, (35), 18326-18378.

1. J. Mater. Res. 2021, 36, 2-25.

Lithium-based layers play key roles in developing nanostructured energy storage systems. As such, ultrathin lithium phosphorous oxynitride LiPON deposited by Atomic Layers Deposition is incorporated as solid-electrolyte for on-chip microsupercapacitors [1-2]. In this way, fundamental understanding of precursors chemistry and stability could be beneficial to control thermal ALD process. In this work, we will focus on the use of Lithium hexamethyldisilazide (LiHMDS) and Diethylphosphoramidate (DEPA) precursors. Both precursors are maintained in canisters at 90°C (DEPA) and 70°C (LiHMDS). Their ageing time in the canisters was considered. New and aged precursors were characterized by Thermogravimetry (TGA), infrared spectroscopy (FTIR), Powder X-Ray Diffraction (PXRD), Nuclear Magnetic Resonance (NMR) and Pyrolysis coupled with Gas Chromatography Mass Spectrometry (PY-GCMS). The growth per cycle, stoichiometry and ionic conductivity of LiPON films were followed.

It was found that new and aged LiHMDS kept the same thermal behavior and the same structure, till 200 days of use. Hence, there was no proof of significate degradation of LiHMDS with storage duration. On the contrary, new and aged DEPA showed differences. The TGA curves progressively changed from one steep mass loss at 220°C to two partial mass losses occurring between 200°C and 320°C. FTIR spectra showed that the amine group of the aged DEPA disappeared after 60 days of storage. NMR data confirmed a deep modification of the P-N-H₂ chain. A possible polymerization of DEPA monomers might take place. Furthermore, yellow spots were observed in the inner bottom of the DEPA’s storage canister. A SEM/EDX analysis revealed deposits enriched with phosphorous. These first measurements pointed out that DEPA has degraded in the canister. The PY-GCMS data confirmed a congruent total evaporation for new DEPA, contrary to new LiHMDS. Its vapor was made of two third of gaseous LiHMDS and one third by a lighter unknown compound.

Shortly, a mass spectrometer will be plugged to the reactor to complete the study by the understanding of the LiPON growth mechanism.

[1] Gölert et al, 2017, https://doi.org/10.1016/j.nanoen.2017.01.054

[2] Sallaz et al, 2024, https://doi.org/10.1021/acselectrochem.4c00022

Solid-state reactions are a foundational and widely used method for synthesizing inorganic solid materials, especially metal oxide ceramics. In typical processes, solid precursors are mixed and heated to high temperatures to induce heterogeneous reactions forming new phases. However, solid-state diffusion-driven phase transitions at elevated temperatures often introduce structural inhomogeneity. For example, Li-ion battery cathodes such as LiTMO₂ (TM = Ni, Mn, Co) are commonly produced via high-temperature reactions involving TM(OH)₂ precursors and lithium sources (e.g., LiOH) in oxidative atmospheres. This complicated non-equilibrium reaction also suffers inherent heterogeneity arises from insufficient solid state lithium diffusion.

While previous studies emphasize optimizing lithium diffusion and particle growth, the intrinsic heterogeneity in solid-state calcination calls for more advanced control strategies. In this work, we found that early-stage formation and coalescence of primary layered particles on polycrystalline Ni_0.9Co_0.05Mn_0.05(OH)₂ (NCM(OH)₂) hinder lithium diffusion, resulting in structural non-uniformity and reduced electrochemical performance. To overcome this, we developed a conformal WO₃ coating via atomic layer deposition (ALD), which effectively regulated lithium transport and suppressed particle coarsening at grain boundaries during calcination. In situ high-temperature X-ray diffraction (XRD) revealed that the WO₃ layer shifted the early-stage reaction from growth-dominated to nucleation-dominated, thus maintained the grain boundary integrity and improved reaction homogeneity. Scanning transmission electron microscopy (STEM) analysis confirmed that rocksalt phases and voids formed in the uncoated product which is signatures of poor lithium diffusion, and they were absent in the WO₃-coated product. As a result, the modified cathode delivered significantly improved performance (92.9% capacity retention after 200 cycles, vs. 78.7% for the pristine sample) under 2.8V-4.4V charging/discharging cycles.

This work advances the understanding of early-stage solid-state reactions and provides a pathway to achieve homogeneity in high temperature solid-state reactions for next-generation cathode materials through grain-boundaries engineering by ALD technique.