金屬粉末射出成型能夠製造形狀複雜、機械性質優良之淨尺寸工件,其製造程序中以脫脂為最關鍵製程,各種脫脂方法中,毛細吸附脫脂方法可明顯縮短製程時間,因此成為重要的產業技術。事實上,胚體/吸附材的空孔度是隨著控制體積改變而改變,但是文獻上很少人研究,「如何量測這些空孔度分佈情形」以及這些「隨著控制體積改變而改變的空孔度」,如何影響整個脫脂過程。本研究將引用「局部空孔度分佈」概念,並從空孔結構圖像來量測空孔度分佈情形。利用最大熵方法來決定「最適宜空孔度分佈」及「特徵長度」,應用適合度檢定方法,找出最適宜「理論分佈函數」。根據此「理論分佈函數」,藉由「亂數產生器」產生大量空孔度數據,代入數值程式中模擬,其中「特徵長度」將是控制體積邊長依據。數值模擬結果發現,吸附材之潤濕邊界會隨機化的移動擴張,其外輪廓線亦相當不規則,整個脫脂時間與胚體半徑成線性關係。此外,本研究也成功完成一系列實驗,觀察到潤濕邊界移動擴張的情形,並量測脫脂時間、黏結劑在胚體殘留量和在吸附材內之填充量。 Metal powder injection molding (MIM) can manufacture parts of net-shape with intricate contour as well as mechanical properties. The debinding stage is a critical process. The wick debinding can reduce the debinding cycle time and is a crucial technique. The porosity of the compact/wick varies with control volume, but there are few literatures in determining the porosity distribution of the compact/wick and how the porosity varying with control volume affects the wick debinding process. A local porosity distribution (LPD) is quoted and measured from the digital image of pore structure. Applicable LPDs and typical length scales are then calculated by maximum entropy method (MEM). Proper theoretical porosity distribution functions are adopted to fit the applicable LPDs. According to the theoretical distribution function, a random number generator is used to generate data of porosity with quantitative randomness for numerical simulations. The typical length scale is an important basis for determining the size of the control volume (or grid). The porosity distributions are used in simulations and show the walking flow edges behave randomly, the contours of wetting wick are irregular and the total debinding time is linearly dependent on the radius of the compact. Besides, an experiment has successfully been developed that can observe the phenomenon of the walking flow edges, measure the total debinding time and calculate the percentages of residual binder and the pore space filled. The numerical results agree well with the experimental results.