以實驗方法針對在基材中添加0.25wt.%碳奈米管及石墨烯微片的碳纖維/環氧樹脂複合材料進行層間破裂韌性及脫層疲勞行為特性研究。在試片的製備上，使用手積層方式搭配溶劑澆注法，最後進行熱壓，將考慮六種不同的碳奈米加強材配置比例，包含：無添加、碳奈米管：石墨烯微片比例為0:10、1:9、5:5、9:1、10:0，來探討有無添加碳奈米加強材、添加碳奈米加強材的種類及兩種碳奈米加強材的含量比例等三項變因對碳纖維/環氧樹脂複合材料mode I層間破裂韌性及疲勞破裂性質的影響。實驗中將探討之破裂特性包含：mode I層間破裂韌性、mode I脫層起始壽命以及mode I脫層成長速率，論文中將利用協同效應指數針對添加兩種碳奈米加強材在這些破裂相關性質的協同效應進行探討。實驗結束後，將利用電子顯微鏡檢視試片之破斷面，以確認碳奈米加強材對所探討之破裂特性的加強機制。實驗結果顯示：層間破裂韌性方面，添加單一碳奈米加強材於基材中時，添加石墨烯微片比添加碳奈米管對於積層板的mode I層間破裂韌性有較好的加強效果。同時添加兩種碳奈米加強材於基材中時，添加比例為9:1之試片具有最高的起始破裂韌性。脫層瞬時阻抗方面，添加碳奈米加強材比例為9:1的試片具有最佳的表現；而碳奈米加強材添加比例為5:5的積層板試片在脫層起始破裂韌性GIC及脫層瞬時阻抗GIR均無加強效果。脫層起始壽命方面，在低負荷階情況下，添加比例9:1有最明顯的加強效果，其次，單獨添加單一種碳奈米加強材，對於脫層起始壽命也有加強效果，隨著負荷階提高，相較於未添加之碳纖維/環氧樹脂積層板，碳奈米管/石墨烯微片添加比例0:10、9:1、10:0均有加強效果，反之，添加比例1:9、5:5均無明顯加強效果。脫層成長速率方面，選擇負荷階50%進行測試，可明顯比較出單獨添加石墨烯微片能夠最有效的維持脫層成長速率。

奈米碳管(Carbon nanotubes, CNT)是一種新型碳結構，由碳原子形成的石墨烯片層捲成的無縫且中空的管體。本研究利用簡易的分散技術來製備節肢型奈米碳管補強環氧樹脂奈米複合材料，利用田口方法來針對製程進行優化，並以奈米複材之機械性質為評估指標來瞭解奈米碳材混入環氧樹脂中之分散情形、補強行為。
研究結果顯示，奈米碳材的最佳的添加量為0.5 wt.%。本研究使用實驗分析法當中的田口方法。首先使用L16直交表並且搭配四種控制因子(分散的轉速、分散的時間、節肢型奈米碳管的添加比例、熱壓成形的壓力)及三種不同的水準參數，再以三種機械性質為品質特性的衡量標準，接著以變異數分析探討各控制因子對於該項機械性質的貢獻度，最後透過形態學分析去比較各製程的補強行為之差異，來找出各種機械特性之最佳製程。

The research aims to investigate the fracture toughness of graphene nanocomposites using single edge notch bending (SENB) specimens. Two different methods were employed to create the sharp crack tip in the SENB samples. One is the tapping method and the other is the sawing method. For the tapping method, a crack instantly propagated around 2 mm under a single tap such that a naturally sharp tip was generated. On the other hand, for the sawing method, a sharp notch was generated by sawing the sample first. The razor blade slightly cut into the samples so that the crack extended along with the sharp blade edge. The samples were then subjected to the three point tests, from which the fracture toughness were obtained. It is noted the graphene loading in the tests are 0.15 and 0.3wt%. For each case, at least five samples were tested. Results shown in Table 1 indicated that associated with the same graphene loading, the tapped samples demonstrated lower fracture toughness than the sawed samples. In addition, for the tapped samples, the fracture toughness of the samples basically are not affected dramatically by the graphene loadings. However, for the sawed samples, the fracture toughness decreases with the increment of graphene loading. When the crack tip is sharped enough, the fracture process zone is relatively small (brittle fracture behavior) and thus the influence of graphene in the onset of the crack extension is minor. On the other hand, when the crack tip is blunt, the fracture process zone is lager and the graphene loading would influence the tip behavior.

The research aims to predict the deformation of laminated plate subjected to impact loading using neutral network (NN) method. The laminated plate containing one layer of cover plate and one layer of Ag was placed on an acrylic substrate and then impacted by a steel ball at speed of 5.05m/s. In addition, the oblique impacts at angles of 10, 20, 30 and 40 with respective to the surface normal were considered in the analysis. The elastic modulus of the cover plate ranged from 0.5GPa to 20GPa and the Poisson ratio varied from 0.1 to 0.49. These two material constants were created randomly using LS-OPT software. During the impact process, the maximum deformation of the Ag layer was of concern. The impact simulations were conducted using LS-DYNA commercial software in conjunction with LS-OPT, from which total 800 datasets were generated. In other words, for each cover plate and impact angle, the maximum deformation of Ag layer was evaluated. The whole datasets were divided into two parts, 90% of which was used for NN model training and the remaining was employed for the model validation. The comparisons of the NN model predictions with the label data obtained from LS-DYNA are shown in Figures 1 and 2, respectively for the displacement and the stress sate. Results reveal that the maximum errors were within 4.9%. Thus, the trained NN model can be employed for the prediction of the deformation and stress of laminated plate with accuracy.

The research aims to investigate the deformed behaviors of a frame structure under a rigid body impact. The frame was made of mild steel the nonlinear behaviors of which were described using an elastic-viscoplastic model. In addition, the joints between the frame members were welded and the effects of the welding strength in the deformed and failure behaviors of the structure after impact were of focus. The impactor is a wedge-type rigid body with either sharp angle or blunt angle. The impact velocity is 40 km/h. The impact simulations were performed using LS-DYNA commercial software. In order to reduce the computational cost, the frame was modeled using shell elements. The welded joints were modeled using the nodal rigid body approach provided by LS-DYNA where the nodes in the web and those in the flange were individually tied together, the failure of which was determined by the failure criterion. Figure 1 illustrates the deformed configuration and the maximum displacement of the frame structure after the impact in terms of three different joint strengths. It is revealed that when the strength of the joint is less than that of the mild steel (326MPa), the weak joint strength may result in the frame structure suffering severe damage. Nevertheless, when the joint strength is greater than that of the members, for the blunt angle case, the deformed configuration is not dramatically affected by the joint strength. However, for the sharp angle case, the joint strength can affect the deformed configuration. In addition, based on our current simulations, the blunt angle impactor may have the maximum deformation greater than the sharp angle impactor. This could be due to the larger contact area generated during the impact for the blunt angle impactor, resulting in the larger deformation in the frame structure. The remaining space of the frame structure after the impact will also be investigated in terms of the joint strength and the impact shapes. In addition, the possibility of replacing the mild steel with the composite materials for building the frame structures will also be examined in the future.

高分子複合材料具有質輕、抗腐蝕、高強度、耐熱性佳及產品設計製造彈性大等等優點，近年來已經漸漸取代金屬，廣泛應用於民用航空器、船舶、汽車、建築、電子電機等等。拉擠成型法（Pultrusion）為一自動化連續式加工製程，利用連續補強纖維經樹脂充分浸潤後，刮除多餘樹脂，由拉引機拉入一成型模具內，加熱固化定型後再行切割，得到最終產品。其優點是線性等截面輪廓、連續自動化生產、外型與強度設計彈性大、較低的設備與製造成本。
本研探討使用蒙脫土或合成土利用化學改質法，將其備製成奈米級黏土。再參混入熱固性樹脂中進行物理性分散(5、10、15、20 wt %)，製備成奈米複合材料基材。將玻璃纖維、奈米複合材料基材兩者利用拉擠成型法製備成奈米複合材料，探討拉擠成型複合材料製程、機械性質等之影響。
研究結果顯示，耐衝擊強度由22.14ft-lb/in (未添加奈米黏土)提升到38.2ft-lb/in (添加奈米黏土)，提升的57.8%。抗折強度由336.05 MPa (未添加奈米黏土)提升到428.2MPa (添加奈米黏土) 提升的78.4%。抗拉強度由331.13Mpa (未添加奈米黏土)提升到346.45MPa (添加奈米黏土) 提升的95.5%。
在本研究裡，奈米黏土是以插層的方式存在於樹脂基材與玻離纖維之間。真正影響拉擠成型複合材料強度的是玻璃纖維束與樹脂基材間的介面結合。由於蒙脫土的離子交換當量較合成土的離子交換當為高，蒙脫土的層與層之間較易撐開，易與樹脂基材、玻璃纖維結合，進而增加其強度，加上蒙脫土模數大於玻璃纖維。且蒙脫土的徑厚比遠大於合成土，與玻璃纖維的徑長比直接近，可提升複合材料的強度。較少的填充量(10wt%)即可大幅提升複合材料的機械強度。可應用在需要較高的機械強度、耐酸鹼、耐候等環境中，如汽-機車外觀件或機能件、橋梁防撞保護基座、碼頭防撞保護設等。

軟性電子產品有著輕、薄、可撓與耐衝擊的特性，其中軟性基板是軟性電子產品的關鍵元件。基於不同軟性電子產品應用對基板材料有不同的要求，基板材料研究成果將決定軟性電子產品發展方向。近年來在軟性電子或感測器相關應用研究上，常見以添加石墨烯材料於透明光學薄膜以提高導電性，機械性能、電磁效應、熱傳導效應和光學透射性質。有鑒於軟性顯示器或電子的薄膜介面黏著力不足或殘留應力問題導致光學薄膜特性、水氧穿透率與機械保護膜層製程參數和產品壽命都還未達到於可靠度測試下之真正商品化的條件，如何檢測這些薄膜應力問題，以回饋給製程端來改善，是一個很重要的議題。甚至軟性顯示器還會在彎曲狀態下顯示影像，所以軟性電子在外界施力下之彎曲的薄膜特性，將成為此軟性電子是否具備足夠的產品壽命之判斷依據。基於目前熟知的應力量測技術中，在薄膜彎曲狀態量測是不容易的。因此，本研究提出了一種在彎曲狀態下之機械特性的測量方法。本研究使用彎曲光學干涉技術於透明導電石墨烯薄膜高分子聚合物基板的光學相位延遲量測，並再透過試片在全面性摺疊測試前與後來分析試片之鍍膜成效之彎曲機械特性。本研究利用兩片單層石墨烯薄膜鍍膜在40 × 40 mm的 250 μm厚的聚對苯二甲酸乙二酯(PET)基材上，進行11,000次之摺疊半徑為10 mm的全面性摺疊測試，並針對測試前後在50、30、20和10 mm半徑的條件下之彎曲應力特性做量測與分析。經過 11,000 次全面性摺疊測試後，第二片石墨烯/PET中的壓縮殘餘應力有被釋放出來。並由彎曲機械特性量測結果可知，第二片石墨烯/PET中之殘餘應力較小由於第一片石墨烯/PET。因此第二片石墨烯/PET試片的薄膜機械性能較優於第一片石墨烯/PET試片，而這原因可以推測為製造過程的不定參數因子所造成。所以基於軟性透明導電基板的彎曲機械性質量測與分析研究之結果，可以為製程設計人員提供產品開發的改進資訊，更可以幫助製造商改進製造設計，以增強薄膜鍍膜技術的提升。