Lithium-based batteries are an essential component of modern society,and the separator is one of the components that most directly governs how such a cell behaves in service. This thesis examines the role of the separator in anode-free lithium metal batteries (AFLMBs), in which lithium is plated directly onto a bare copper current collector. Three commercial separators — Whatman GF/F borosilicate glass fibere, Freudenberg Viledon FS 3005-25 (PET/Al2O3 composite) and Celgard2320 (PP/PE / PP trilayer) —were assembled into lithium manganese oxide LiMn2O4‖ Cu cells and evaluated by galvano-static cycling with potential limitation (GCPL) at 1C rating and electrochemical impedance spectroscopy (EIS), together with a systematic variation of the number of separator layers. Increasing the number of layers consistently reduced cycling stability and shortened the useful cycle life, which is attributed to the additional transport path length, poorer rough-thickness wetting and the reduced compliance of a thicker separator stack. Among the tested materials, Celgard 2320 gave the best overall performance, reflect-ing a favourable compromise between ionic conductivity, mechanical stability and thermal-shutdown safety; scanning electron microscopy(SEM) showed that this material also pos-sesses the most uniform and best-interconnected pore network of the three. In addition, a separator-free anode-free configuration was studied and compared with the conventional assemblies. This configuration sustained the longest cycling among all the cells tested, consistent with its lower ohmic and interfacial resistances, and is used here as a model system rather than as a practical cell design. Because the separator-free geometry eliminates the membrane’s transport resistance, it provides a comparatively clean environment for studying lithium electrodeposition itself. The effect of the applied current density was therefore examined at constant deposited charge, over the range 0.5–10 mA cm−2. Quantitative image analysis showed that the mean deposit size decreases from 0.87 µm at 0.5 mA cm−2to 0.51 µm at 9 mA cm−2, following an approximate power law ¯d ∝J−0.16, where d is the mean deposit diameter and J is the applied current density, while the potential–time response retains a single characteristic shape.At 10 mA cm−2, both the potential response and the deposit morphology depart from this trend, and filamentary growth appears, indicating a transition from a nucleation-controlled to a transport-controlled deposition regime. Taken together, the results identify Celgard 2320 as the most suitable practical sepa-rator among those tested. They also show how excessive separator thickness reduces cell performance and define the range of current densities over which lithium can be deposited compactly on bare copper in this electrolyte.

The Effects of Charging Current and Separator Properties on Lithium Morphology and the Performance of Anode-Free Lithium-Metal Batteries (AFLMBs)

MOHAZABIEH, ELNAZ
2025/2026

Abstract

Lithium-based batteries are an essential component of modern society,and the separator is one of the components that most directly governs how such a cell behaves in service. This thesis examines the role of the separator in anode-free lithium metal batteries (AFLMBs), in which lithium is plated directly onto a bare copper current collector. Three commercial separators — Whatman GF/F borosilicate glass fibere, Freudenberg Viledon FS 3005-25 (PET/Al2O3 composite) and Celgard2320 (PP/PE / PP trilayer) —were assembled into lithium manganese oxide LiMn2O4‖ Cu cells and evaluated by galvano-static cycling with potential limitation (GCPL) at 1C rating and electrochemical impedance spectroscopy (EIS), together with a systematic variation of the number of separator layers. Increasing the number of layers consistently reduced cycling stability and shortened the useful cycle life, which is attributed to the additional transport path length, poorer rough-thickness wetting and the reduced compliance of a thicker separator stack. Among the tested materials, Celgard 2320 gave the best overall performance, reflect-ing a favourable compromise between ionic conductivity, mechanical stability and thermal-shutdown safety; scanning electron microscopy(SEM) showed that this material also pos-sesses the most uniform and best-interconnected pore network of the three. In addition, a separator-free anode-free configuration was studied and compared with the conventional assemblies. This configuration sustained the longest cycling among all the cells tested, consistent with its lower ohmic and interfacial resistances, and is used here as a model system rather than as a practical cell design. Because the separator-free geometry eliminates the membrane’s transport resistance, it provides a comparatively clean environment for studying lithium electrodeposition itself. The effect of the applied current density was therefore examined at constant deposited charge, over the range 0.5–10 mA cm−2. Quantitative image analysis showed that the mean deposit size decreases from 0.87 µm at 0.5 mA cm−2to 0.51 µm at 9 mA cm−2, following an approximate power law ¯d ∝J−0.16, where d is the mean deposit diameter and J is the applied current density, while the potential–time response retains a single characteristic shape.At 10 mA cm−2, both the potential response and the deposit morphology depart from this trend, and filamentary growth appears, indicating a transition from a nucleation-controlled to a transport-controlled deposition regime. Taken together, the results identify Celgard 2320 as the most suitable practical sepa-rator among those tested. They also show how excessive separator thickness reduces cell performance and define the range of current densities over which lithium can be deposited compactly on bare copper in this electrolyte.
File in questo prodotto:
File Dimensione Formato  
Mohazabieh_Elnaz_898542.pdf

non disponibili

Dimensione 11.42 MB
Formato Adobe PDF
11.42 MB Adobe PDF

I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.14247/30042