複合型固態電解質因兼具高分子材料之良好加工性與界面接觸性，以及無機陶瓷填料之優異熱穩定性與抑制鋰枝晶能力，被視為提升鋰金屬電池安全性與循環穩定性之重要材料系統。然而，填料粒徑、添加量與分散狀態會影響電解質膜之微觀形貌與電化學穩定性。特別是在高陶瓷添加量下，填料團聚與漿料凝膠化等問題將限制其實際應用。因此，本研究以異氰酸丙基三乙氧基矽烷 (IPTS) 修飾不同粒徑 (100 nm、300 nm、700 nm) 之鋰鑭鋯鋁氧化物 (LLZAO) 陶瓷填料，並將其導入聚偏二氟乙烯-六氟丙烯共聚物 (Poly(vinylidene fluoride-co-hexafluoropropylene)， PVDF-HFP) 與聚碳酸丙烯酯 (Poly(propylene carbonate)， PPC) 之高分子基材中，依序探討表面修飾、陶瓷添加量及粒徑對複合型固態電解質結構與電化學性能之影響。 材料鑑定結果顯示，IPTS 修飾後 LLZAO 仍維持原有晶相，且表面官能基與分散狀態產生改變，顯示修飾可在不破壞晶體結構下改善其與高分子基材之相容性。進一步以 300 nm 之 M3LLZAO 比較不同添加量，M3LLZAO-20 仍以高分子相為主要連續結構；提高至 50 wt% 後，可增加陶瓷/高分子界面並維持均勻分布與膜體完整性；達 70 wt% 時，則出現表面起伏、裂紋與局部不均。M3LLZAO-50 之離子導電率達 1.52 × 10-3 S cm-1與臨界電流密度為 0.80 mA cm-2，並維持 1.18 MPa 之拉伸強度，顯示 50 wt% 添加量在離子傳輸、電化學穩定性與機械性能間呈現較佳平衡。 固定陶瓷添加量為 50 wt% 後，比較不同 LLZAO 粒徑之影響。其中，推論M7LLZAO因填料尺寸較大而導致膜體不平整，造成與電極界面接觸不良，使整體電性表現較為遜色。反觀 M1LLZAO-50 與 M3LLZAO-50，其氧化電位分別為 4.95 與 4.93 V、鋰離子遷移數為 0.54 與 0.63，並且透過離子導電率與遷移數之乘積 (鋰離子導電率) 評估鋰離子貢獻度，分別為 5.78 × 10-4 S cm-1 與 9.58 × 10-4 S cm-1，顯示 M3LLZAO-50 之有效界面利用率與鋰離子連續傳輸表現較佳。另一方面，M1LLZAO-50 於 1.5 C 下具有最高放電比電容 85.5 mAh g-1，顯示小粒徑填料有利於高倍率容量輸出；然而，其 1.5 C 庫倫效率僅為 57.1%；相較之下，M3LLZAO-50 於 1.5 C 下約為 70.3 mAh g-1，庫倫效率達 95% 以上，說明其界面穩定性可能較佳。 在長期循環穩定性方面，M1LLZAO-50 與 M3LLZAO-50 均可於 0.1 mA cm-2 下維持約 1000 h 之鋰剝離/鍍覆循環，M3LLZAO-50 於循環過程呈現較穩定的電壓響應。應用於高壓鋰金屬電池後，M1LLZAO-50 雖具有較高的初始容量，但隨循環快速衰退；相較之下，M3LLZAO-50 經 100 圈 1 C 循環後之電容維持率達 80.5%。綜合結果，M1LLZAO-50 在高倍率容量輸出與電性表現上皆具有潛力，但仍需改善填料分散與電極界面穩定性；M3LLZAO-50 則在膜體結構、鋰離子傳輸及長期循環穩定性之間呈現較均衡的綜合表現，顯示其作為高壓鋰金屬電池複合型固態電解質之應用潛力。;Composite solid electrolytes (CSEs)， which combine the good processability and interfacial contact of polymer with the excellent thermal stability and lithium dendrite suppression capability of inorganic ceramic fillers， have been regarded as promising electrolyte systems for improving the safety and cycling stability of lithium metal batteries (LMBs). However， the particle size， loading amount， and dispersion state of ceramic fillers significantly affect the microstructure and electrochemical performance of CSEs， especially at high ceramic loadings where filler aggregation and slurry gelation may limit practical application. In this study， LLZAO ceramic fillers with different particle sizes (100 nm， 300 nm， and 700 nm) were surface-modified using 3-isocyanatopropyltriethoxysilane (IPTS) and incorporated into a Poly(vinylidene fluoride-co-hexafluoropropylene) (PVDF-HFP)/ Poly(propylene carbonate) (PPC) polymer matrix. The effects of particle size and loading amount of IPTS-modified LLZAO on the structure， electrochemical properties， and coin cell performance of CSEs were systematically investigated. Material characterization showed that IPTS-modified LLZAO retained its original crystalline phase， while changes in the surface functional groups and dispersion behavior confirmed successful surface modification without disrupting the ceramic structure. The effect of ceramic loading was subsequently investigated using 300 nm M3LLZAO. M3LLZAO-20 remained dominated by a continuous polymer phase， whereas increasing the loading to 50 wt% increased the ceramic/polymer interfacial area while maintaining relatively uniform filler distribution and membrane integrity. At 70 wt%， pronounced surface roughness， cracks， and local structural nonuniformity were observed. M3LLZAO-50 exhibited an ionic conductivity of 1.52 × 10-3 S cm-1， the highest critical current density of 0.80 mA cm-2， and a tensile strength of 1.18 MPa， indicating that a ceramic loading of 50 wt% provided a favorable balance among ion transport， electrochemical stability， and mechanical performance. After fixing the ceramic loading at 50 wt%， the influence of LLZAO particle size was further evaluated. The relatively inferior electrochemical performance of M7LLZAO-50 was attributed to its larger filler size， which resulted in an uneven membrane structure and poorer electrode/electrolyte interfacial contact. In contrast， M1LLZAO-50 and M3LLZAO-50 exhibited oxidation onset potentials of 4.95 and 4.93 V， respectively， and Li⁺ transference numbers of 0.54 and 0.63. Their Li+ conductivities， calculated from the product of ionic conductivity and Li+ transference number， were 5.78 × 10-4 and 9.58 × 10-4 S cm-1， respectively， indicating that M3LLZAO-50 provided more effective interfacial utilization and more continuous Li+ transport. M1LLZAO-50 delivered the highest discharge specific capacity of 85.5 mAh g-1 at 1.5 C， demonstrating the potential of smaller fillers for high-rate capacity output. However， its coulombic efficiency (CE) at 1.5 C was only 57.1%. By comparison， M3LLZAO-50 delivered approximately 70.3 mAh g-1 while maintaining CE above 95%， suggesting more stable electrochemical reactions at high rates. In terms of long-term cycling stability， both M1LLZAO-50 and M3LLZAO-50 sustained Li stripping/plating cycling for approximately 1000 h at 0.1 mA cm-2， while M3LLZAO-50 exhibited more stable voltage response during cycling. When applied in room-temperature NCM811 || CSE || Li cells， M1LLZAO-50 showed a higher initial capacity but underwent rapid capacity fading. In contrast， M3LLZAO-50 achieved a capacity retention of 80.5% after 100 cycles at 1 C. Overall， M1LLZAO-50 demonstrated considerable potential for high-rate capacity output， although further improvements in filler dispersion and electrode interfacial stability are required. M3LLZAO-50 exhibited a more balanced overall performance in terms of membrane structure， Li⁺ transport， interfacial stability， and long-term cycling performance， demonstrating its potential as a composite solid electrolyte for high-voltage lithium metal batteries.
