應用光碟製程製作微流體晶片於均一粒徑微膠囊之生成與最佳化之研究

Using Optical Disc Process to Fabricate Microfluidic Chips for Optimization of Uniform Microcapsules Generation

吳俊翰(Chun-Han Wu)

指導教授 ： 徐文祥 ； 林裕城

國立交通大學/工學院/機械工程系所/博士(2011年)

https://doi.org/10.6842/NCTU.2011.00014

摘要

本研究利用光碟製程製作微流體平台以應用於微膠囊之生成，在先前以微流體方式生成乳化球之研究，皆以灌模製程製作的PDMS微流體晶片，或是以CO2雷射在塑膠基材上雕刻製作的微流體晶片。然而，PDMS微流體晶片材質較軟，容易造成流道阻塞，且每成形一片晶片需花費3~4小時以上。而CO2雷射雕刻製作的微流體晶片，晶片較粗糙度、不平整及容易變形，且流道寬度常常受限於雷射雕刻機台之極限，每生產一片晶片依圖案設計而不同約10分鐘以上，因此這些方法並不適於大量的生產。本論文提出以新型式的光碟製造技術製作微流體模仁，並結合光碟微成形技術製作塑膠微流體晶片，確保晶片內流道平滑、平整及不變形，並以光碟射出成形技術大幅縮短塑膠晶片成形時間。
本論文分為三大部分：(1)微流體晶片設計；(2)微流道模仁及晶片製作；(3)微乳化球生成。在微流體晶片設計部分，微流道深度設計為50/200 µm兩種深度的結構，並利用CFD-RC模擬軟體模擬微流道中流體及乳化球運動現象。在微流體模仁製作方面，採用一種新型式光碟模仁製作方法以製作兩層結構之微流體模仁。可避免直接採用傳統光碟模仁製作方法在射出成形時，造成模具鏡面損壞的問題。在模具系統方面，採用模仁固定治具及真空吸附方式，將模仁固定在模具內；使得在更換模仁時，只需花費幾分鐘的時間，比傳統方法需花費數個小時縮短許多，大大地提高成形機台的稼動率。在微流體晶片成形方面，本研究以實驗計畫法探討射出參數（如模溫、料溫、鎖模力及射速）對塑膠微流體晶片品質（如深度複製率、寬度變異、雙曲折射、翹曲以及表面粗糙度）之影響作一探討。在微乳化球生成部分，本研究以十字型微流道實驗，證明藉由操控連續相/分離相的流量比值，可輕易地控制去離子水及 1.5% (w/v) 褐藻酸鈉微乳化球之粒徑。並採用田口方法探討分離相流量、連續相/分離相流量比值、分離相黏滯係數以及界面活性劑 Span 80添加濃度等控制因子，對微乳化球的生成粒徑尺寸影響作一探討。
在微流道乳化球生成模擬結果顯示，當連續相流速與非連續相流比值越高時，生成的微乳化球粒徑越小，且微乳化球生成速度越快。在微流體模仁製作方面，成功地製成厚度為300 µm的雙層微結構之模仁，且模仁背面平坦無任何圖案，在射出成形時不會造成模具鏡面之損壞。在微流體晶片製作部分，以實驗計畫法探討射出參數對塑膠微流體晶片品質。實驗結果顯示以2nd段鎖模力及射速對晶片深度複製率、寬度變異以及表面粗度影響較大；料溫和射速對晶片雙曲折射影響較大；模溫對晶片翹曲影響較大。在實驗中每一片微流體晶片的成形時間皆為 4秒鐘，比傳統的成形的時間縮短十幾倍，可大幅降低每一片塑膠晶片生產成本。在微乳化球生成方面，以田口法探討微乳化球均一粒徑生成最佳化實驗。在最佳化參數組合下，可生成最小粒徑為 19.58 µm 及粒徑變異係數（Coefficient of variation, CV）值為 1.95%之微乳化球，較原始設計條件下產出的乳化球尺寸要小 23.46%，尺寸變異值也小20.73%。並利用相同的乳化球生成方法及結合外部凝膠法，成功的將奈米金粒子及免疫蛋白充填在褐藻酸鈣微膠囊內。
本論文成功地採用新型式的光碟製程製造微流體晶片，每片晶片生產時間只需4 seconds, 大幅地降低微流體晶片生產時間。並利用此晶片製成微流體平台可生成均一粒徑的微乳化球，且具有粒徑可操控及製程簡易、成本低、產能高及非常適合大量生產等優點。

關鍵字

微流體 ； 微乳化球 ； 射出成型

並列摘要

This study presents a microfluidic platform fabricated by optical disc process for microcapsules generation. In the previous reports, the generation of microemulsion adopted microfluidic methods. The microfluidic chips were fabricated by casting process or by adopting CO2 laser engraving on plastic sheet. However, microfluidic chip of PDMS material is too soft, and can easily induce microchannel clogged. Furthermore, the cycle time for manufacturing the microfluidic substrate is over 3~4 hours. As for CO2 laser engraving, the surface of the microfluidic channels was too rough, uneven, deformed and different from the original designs. Besides, the size of the channels was limited by laser machine, and the cycle time for manufacturing the microfluidic substrate is over 10 minutes. The cycle time for both casting process and CO2 laser engraving process is too long. The long cycle time will be a drawback for mass production. Therefore, we present a new process of optical disc to fabricate microchannels of mold insert and adopt the micromolding technique of optical disc to prevent roughness, unevenness, deformation, and clog of microchannels.
This paper is divided into three sections. (1) a design of microfluidic chip; (2) the fabrication of microfluidic mold insert and microfluidic chip; (3) generation of microemulsion droplet. In the design of microfluidic chip section, the depth of microchannels were 50/200 μm respectively and CFD-RC simulation software was adopted to simulate the motion of fluids and emulsions in microfluidic channels. In the fabrication of microfluidic mold insert, we propose a new process of optical disc to manufacture microfluidic mold insert. This new process can prevent the damage on the mirror plate of the mold caused by the traditional process of optical disc. The mold system is composed of a mold insert (stamper) holder and a vacuum system, which is used to join the mold insert with the mold. In this way, the time to change the stamper is drastically decreased. Therefore, it can improve the utility rate of injection molding. In the microfluidic chips of injection molding, we investigate the influence of controlled factors on the quality properties (depth of replication rate, width deviation, birefringence, tilt and surface roughness) of microfluidic chips. Those controlled factors include mold temperature, cylinder temperature, clamping force and injection speed. In the microemulsion generation, we use cross-junction microchannel to form uniform water-in-oil (w/o) emulsions. We prove that the size of these emulsion drops can be easily controlled by adjusting the ratio of disperser phase flow / continue phase flow. These emulsion drops, consisting of 1.5% (w/v) sodium alginate (Na-alginate), are then dripped into a solution containing 20% (w/v) calcium chloride (CaCl2) to create Ca-alginate microcapsules in an efficient manner. We apply Taguchi method to investigate the influence of controlled factors on the size of microemulsion drops which include the flow rate of dispersed phase flow, flow rate ratio, viscosity and surfactant concentration.
The simulation results demonstrate that the bigger the flow rate ratio (continue phase flow /dispense phase flow) is, the smaller the size of microemulsion drops is, and the faster the speed of the generation microemulsion will become. In the fabrication mold insert, we successfully fabricate two layer of microfluidic mold insert, 300 um in thickness and the back side of mold insert is smooth and even. The proposed method will not damage the mirror of the mold.
In the fabrication of microfluidic chips, we adopt DOE method to investigate the influence of controlled factors on the quality properties. The results indicated 2nd clamping force and injection speed have a greater effect on depth replication rate, width deviation and surface roughness. The injection speed and cylinder temperature have a greater effect on birefringence and the mold temperature has a greater effect on vertical deviation. The cycle time of injection molding is 4 seconds. It can be reduced to more than tenfold compared with the traditional ones. In the generation of microemulsion drops, we apply Taguchi method to investigate the influence of controlled factors on the size of microemulsion drops. The results show that the smallest size is 19.58 μm and coefficient of deviation of emulsion drop is 1.95%, the size of emulsion drop is 23.46% lower and the size deviation of drop is 20.73% lower than original design. We demonstrate that the Au nanoparticles and Immunoglobulins are encapsulated into Ca-alginate microcapsules.
In this paper, we successfully devise the new method of optical disc process to fabricate microfluidic chips. The cycle time of microfluidic chips is 4 seconds. It is faster than traditional methods. The proposed microfluidic platform is capable of generating relatively uniform emulsions and has the advantages of active control of the emulsions diameter, a simple and low cost process, and a high throughput, highly efficient and suitable for mass production.

並列關鍵字

microfluidic ； microemulsion drop ； injection molding

參考文獻

2. Martinsen, G. Skjakbraek, and O. Smidsrod, “Alginate as immobilization material: I. Correlation between chemical and physical-properties of alginate gel beads,” Biotechnology and Bioengineering, 33, 79-89, 1989.

3. P. DeVos, B. DeHaan, and R. VanSchilfgaarde, “Effect of the alginate composition on the biocompatibility of alginate-polylysine microcapsules,” Biomaterials, 18, 273-278, 1997.

4. J.H. Cui, J.S. Goh, P.H. Kim, S.H. Choi, and B.J. Lee, “Survival and stability of bifidobacteria loaded in alginate poly-l-lysine microparticles,” International Journal of Pharmaceutics, 210, 51-59, 2000.

5. W. Sabra, A.P. Zeng, and W.D. Deckwer, “Bacterial alginate: Physiology, product quality and process aspects,” Applied Microbiology and Biotechnology, 56, 315-325, 2001.

6. N. Gerbsch and R. Buchholz, “New processes and actual trends in biotechnology,” Fems Microbiology Reviews, 16, 259-269, 1995.