玻璃基板切割之應力分佈及破裂研究

Scoring Stress Distribution and Fracture Research on Glass Substrate

陳坤智(Kun-Chih Chen)

指導教授 ： 許源泉

國立虎尾科技大學/工程學院/機械與電腦輔助工程系碩士班在職專班/碩士(2015年)

https://doi.org/10.6827/NFU.2015.00191

摘要

平板玻璃基板為TFT-LCD面板重要的原料之一，相關的後續製程均建立在玻璃基板上，切割為一重要的製程，一般使用高硬度的金屬或鑽石切割刀輪。在刀輪選用上，大都先透過經驗法測及使用實驗的方式，直接在生產機台上測試影響性及驗證其效能，較少事先透過數值分析的方式來做初步的選用。
本論文運用維氏微型硬度試驗機，針對玻璃做壓痕及裂痕測試來得到塑性變形的機械性質。再透過相關學者研究的經驗公式來得到破裂韌性數值及產生裂痕極限應力值。在數值分析方法上，運用DEFORM 3D有限元素分析軟體，建立模型、設定分析參數及材料性質，來比對模擬與實驗結果的差異。
將硬度試驗壓痕器及玻璃切割刀輪的作用機構建立模型，分別模擬壓痕器及切割刀輪下壓的動作，在玻璃表面及內部形成裂痕。比較不同切割負荷及不同刀輪角度形成中間裂痕的差異性，並與相關研究的經驗公式來比較。
維氏硬度實驗與有限元素模擬結果比對，實際壓痕長度與有限元素模擬結果有較小的差異性8.9%，可有效的模擬硬度量測值。實際壓痕深度與有限元素模擬結果有較大的差異21%，由壓痕器尺寸計算的深度及考慮玻璃有回彈量無法得到確切的深度。實際表面裂痕與模擬表面及內部裂痕平均值較為接近差異為15%。
不同切割刀輪負荷及刀輪角度對中間裂痕的影響，比對經驗公式與有限元素模擬結果，玻璃脆性材料的特性及裂痕對作用應力的敏感性，在中間裂痕預測上，也有較大差異13~17%的結果及呈現兩段曲線趨勢。分析判斷為殘留應力的影響需建立彈塑性分析模型，實際切割刀輪滾動方式與有限元素模型直接下壓動作的差異。
但在模型機構建立及相對性的數值分析資料上，可幫助了解切割應力分佈及裂痕的延伸判斷。在實際測試前，事前的模擬規劃及比對現行的製程，仍可得到有效的參考性資料。

關鍵字

有限元素分析 ； 維氏硬度試驗 ； 破裂韌性 ； 玻璃切割

並列摘要

Glass substrate is one of the important materials of TFT-LCD panel. The following LCD manufacturing processes are built on glass substrate and the scoring is one of the key processes. Normally, the scoring wheel material is made by using high hardness metal or diamond. The evaluation of score wheel type selection is decided by experience and experiment in the production machine to test the effect and verify the performance. Not usually to use numerical analysis method to predict the effect for score wheel type selection.
This study used Vickers micro hardness tester to do glass indentation and crack test and obtain mechanical property of plastic deformation. Also used empirical formula from related scholars studied to acquire limit fracture toughness and limit stress to cause crack propagation. In numerical analysis method, applied DEFORM 3D Finite Element Analysis (FEA) software to build modeling and set analysis parameters, material property to compare simulation and experiment result.
There are two mechanisms modeling for both Vickers hardness tester indenter and glass score wheel. Simulated indenter and score wheel move down motion and contact with glass to create glass surface crack and inner crack. Compared the difference between different score load and different score wheel angle, and compared with related research of empirical formula.
Compared Vickers hardness tester experiment result with FEA result. There was 8.9% error between indentation lengths with FEA result. It could predict material hardness by simulation method. There was 21% error between indentation depths with FEA result. The bigger error for indentation depth was indenter real dimension and glass restore couldn’t build accuracy depth calculation. There was 15% error between real surface crack with simulation surface and inner crack average result.
For effect of different score load and different score wheel angle, compared empirical formula and FEA result. The characteristic of glass brittle material and sensitivity of crack by stress applied were contributed bigger error 13~17% and two different curves be found on median crack prediction. Analyzed and judged the effect of residual stress need to use FEA Plastic-Elastic model to get more accuracy prediction. The difference between real score wheel rolling motion on glass with FEA modeling single axis move down motion also effect the result.
At modeling mechanism building and relative numerical analysis data, it could help to realize the stress distribution and crack propagation judgment. Before the actual test, the advance planning and simulation compared to existing process can be still effective information material.

並列關鍵字

Finite element analysis ； Vickers Hardness test ； Fracture Toughness ； glass score

參考文獻

[7] Toshihiko Ono & R.A. Allaire, Scoring and Separating Recommendation, Corning Technical Information Paper, TIP 305, November 2004, pp.1

[8] Toshihiko Ono & R.A. Allaire, Fracture Analysis, a Basic Tool to Solve Breakage Issue, Corning Technical Information Paper, November 2004, pp.1-2，pp.7-8

[10] Toshihiko Ono and Kohichi Tanaka, Effective of Scribing Wheel Dimensions on the Cutting of AMLCD Glass Substrates, Corning Technical Information Paper, TIP 306, November 2004

[15] Griffith A. A. , The Phenomena of Rupture and Flow in Solids, Phil. Trans. R. Soc. Lond. January 1921, pp.163-198.

[20] Jianghong Gong, “Determining Indentation Toughness by Incorporating True Hardness Into Fracture Mechanics Equations”, Journal of the European Ceramic Society 19, 1999, pp.1585-1592