仿生膜的結構與機械特性研究

Abstract

高分子薄膜是由雙親性高分子所構成。它的高分子量使得機械、輸送性質強過於脂質薄膜。儘管許多實驗技術已投入薄膜特性的研究，微觀的結構樣貌與性質的量測仍飽受限制。介觀尺度的模擬方法可以補足實驗 在微觀尺度上所達不到的限制，因此能提供一套可行的方法，幫助了解聚合體薄膜結構型態與其物化性質間的關聯性。此外，模擬方法的預測性也將有助於仿生膜發展與應用性的開發。 本論文利用耗散粒子動力學法(DPD)來對三種高分子系統進行研究。我們特別針對高分子薄膜進行微觀性質與其生物性行為的研究。此論文分為三個部分，論文的第一部分(第三章)，利用耗散粒子動力學模擬探討支撐性高分子雙層膜(SPB)的物理性質。在選擇性溶劑裡，吸附在表面修飾基材上的雙嵌段共聚物囊胞之型態會受到親水段鏈和基材之間以及疏水段鏈和溶劑之間的作用力影響，導致破裂發生進而形成SPB或者是保持完整的囊胞型態。針對SPB的研究，發現其幾何和機械性質與親水段鏈和基材之間以及疏水段鏈和溶劑之間的作用力有關。當親水段鏈和基材之間有很強的吸附性時，會容易發生囊胞破裂。此外，當親水段鏈和基材之間有很強的吸附性以及疏水段鏈和溶劑之間缺少良好親和力時，形成SPB的薄膜高度會受到阻礙。然而，SPB的面積震盪會提升。 論文的第二部分(第四章)，利用耗散粒子動力學模擬探討三嵌段ABA共聚物裡共存I型和U型分子鏈。平衡過程中，鬆弛時間(relaxation time)會受到初始I型和U型鏈之間的比例以及親水(A)和疏水(B)段鏈之間作用力不同而影響。平衡時，無論初始I型和U型鏈的比例或是親水和疏水段鏈之間作用力如何改變，I型和U型鏈都會到達固定比例。此外，薄膜的幾何、機械、和輸送性質隨著 A 和 B 段鏈之間不相容性不同之變化。當 A 和 B 段鏈之間不相容性變強時，薄膜的厚度以及疏水厚度 (h)都會上升，因為分子間排列地更加緊密。拉伸係數(KA)和彎曲係數(KB)都隨著 A 和 B 段鏈間不相容性上升也呈現上升的趨勢，並且 KA、KB、h 間算出 KB/KAh2 大 約是 2×10-3，符合薄膜間這三者性質應是常數的關係式。雖然 A 和 B 段鏈間不相容性對擴散係數的影響不明顯，但隨著 A 和 B 段鏈之間作用力的上升，滲透度有相當明顯地下降。 論文的第三部分(第五章)，利用耗散粒子動力學模擬分別探討雙面樹狀高分子和脂質與雙面樹狀高分子共組裝所形成的薄膜和囊胞之性質。根據雙面樹狀高分子裡的疏水碳氟鏈(RF)在苯甲酸取代基位置可分為3,4-，3,5-，和3,4,5-三種樹狀形式。這三種樹狀形式的疏水層厚度大小順序為3,5-RF < 3,4-RF < 3,4,5-RF，而疏水的交錯(interdigitation)程度為3,5-RF > 3,4-RF > 3,4,5-RF。3,4,5-樹狀形式具有最高的拉伸係數(KA)和最小的擴散係數(D)，而3,5-RF比3,4-RF具有較高的KA和較小的D。另一方面，脂質和雙面樹狀高分子共組裝所形成的混合薄膜性質會受到脂質濃度(∅_l)以及不同樹狀形式影響。當混合薄膜含有3,5-RF會比含有3,4,5-RF較容易產生形變。在混合薄膜裡的脂質區域，其疏水厚度會隨著∅_l增加而上升。反之，交錯程度會隨∅_l減少而下降。定性上，雙面樹狀高分子共組裝形成的薄膜和囊胞這兩種結構，其定性上沒有太大差異。然而，囊胞的疏水厚度會比其薄膜的厚度薄，進而讓囊胞的交錯程度變得比其薄膜來的嚴重。此外，3,5-RF混合囊胞的形變會比3,5-RF混合囊胞嚴重。在高比例的脂質濃度情況下，3,5-RF混合囊胞會發生出芽(budding)構造。

Self-assembled polymer membranes have attracted a growing attention due to their multifunctionality and stability. Compared to lipid membranes, polymer membranes have enhanced mechanical and transport properties with high molecular weight. Numerous experimental techniques have been developed to explore membrane characteristics; however, experimental microscopic observations and knowledge of vesicles are limited. Mesoscale simulations can complement experimental studies of the membrane features at the microscopic level and thus provide a feasible method to better understand the relationship between the fundamental structures and physicochemical properties of a membrane. Moreover, the predictive ability of the simulation approaches may greatly assist developments and future applications of biomimetic membranes. This dissertation uses dissipative particle dynamics (DPD) to explore the self- assembly of three polymeric systems. We have paid particular attention to the fundamental properties of polymer membranes and their biological behaviors. There are three parts in this dissertation. In the first part (Chapter 3), the formation and physical properties of solid-supported polymer bilayer (SPB) on an adhesive substrate have been explored. SPB is developed by the adsorption of vesicles formed by diblock copolymers in a selective solvent. The adsorbed vesicle can remain intact or become ruptured into SPB, depending on the interaction between solvophobic block and solvent and the interaction between solvophilic block and substrate. The morphological phase diagram of adsorbed vesicles is acquired. The influences of polymer adhesion strength and solvophobicity on the geometrical and mechanical properties of SPB are systematically studied as well. It is found that vesicular disruption is easily triggered for strong adhesion strength Moreover, for strong adhesion strength and weak solvophobicity, the fluctuation of membrane height is impeded while the area fluctuation is enhanced. In the second part (Chapter 4), Instead of forming typical bilayer or monolayer membrane, both the bridge (I-shape) and loop (U-shape) conformations are coexistent in the planar membranes formed by ABA triblock copolymers in a selective solvent. The non-equilibrium and equilibrium relaxation dynamics of polymer conformations are monitored. The non-equilibrium relaxation time depends on the initial composition and grows (increases) with (an increase in) the immiscibility between A and B blocks. The equilibrium composition of the loop-shape polymer is independent of the initial composition and A-B immisibility. However, the extent of equilibrium composition fluctuations subsides as A and B blocks become highly incompatible. The influences of the A-B immiscibility on the geometrical, mechanical, and transport properties of the membrane are also investigated. As immiscibility increases, the overall membrane thickness and the B block layer thickness (h) rise (increase) because of the increment of (in) the molecular packing (density). As a result, both the stretching (K_A) and bending (K_B) moduli grow significantly with increasing A-B immiscibility. Consistent with typical membranes, the ratio K_B/K_A h^2=2×〖10〗^(-3) is a constant. Although the lateral diffusivity of polymers is insensitive to the immiscibility, the membrane permeability decreases substantially as A-B immiscibility is increased. In the third part (Chapter 5), The influences of the branching patterns on the membrane properties of Janus dendrimers in water have been investigated. The hydrophobic fluorinated dendron (RF) contains three types of branching patterns, including 3,4-, 3,5-, and 3,4,5-RF. Consistent with experimental results, the hydrophobic layer thickness (H_B) follows the order: 3,5-RF < 3,4-RF < 3,4,5-RF, which can be explained by the extent of interdigitation (∆h) : 3,5-RF > 3,4-RF > 3,4,5-RF. Moreover, the 3,4,5-RF membrane shows the highest stretching modulus (KA) and the lowest lateral diffusivity (D). The 3,5-RF membrane is similar to the 3,4-RF membrane but exhibits higher KA and smaller D. For the nano-sized dendrimersome, its bilayer thickness is less than that of the planar membrane due to its larger extent of interdigitation. The coassembly of dendrimersomes with lipids has been studied as well. The thickness and the extent of interdigitation of the lipid-rich domain for the hybrid membrane is significantly affected by the lipid concentrations (∅_l) and the branching patterns. As ∅_l increases, the thickness of the lipid-rich domain grows corresponding to the decrease of interdigitation of the lipid-rich domain.