聚焦離子束運用於生物醫學研究

Abstract

聚焦離子束 (focused ion beam, FIB) 是半導體界和材料科學界的重要強大工具。積體電路常常需要觀察修補金氧半導體元件，以及連接線路的組成與配置。而具備適當能量的聚焦離子束，能切割和銑削積體電路，同時能以機器內建之掃描式電子顯微鏡 (scanning electron microscopy, SEM) 觀察超微結構。 而FIB可以使用在生物與醫學的範疇嗎？這是一個新興的研究發展項目，相關的研究仍然不多。生物材料的配置，前處理，與切削的參數，都仍然是必須開發的研究項目。使用傳統的透射電子顯微鏡（transmission electron microscopy, TEM），生物組織必須先經過固定，脫水，染色，塑膠包埋 (plastic embedding)，切片等一系列複雜的流程。常常必須先做粗切片，先於光學顯微鏡下觀察，再開始做超細微切片，繼而上機尋找想看的關鍵處。但若上機後看不到想看的關鍵處，就必須重復粗切與細切的過程，再重新上機觀察。整個步驟非常的繁瑣耗時。 而FIB/SEM 可以做原位特定點 (in-situ site-specific) 切片並直接觀察。亦即FIB可以在SEM下，直接找尋關鍵特徵，進行切削，切削後也可以直接觀察切削的表面。所以可以大幅減低傳統TEM必需消耗的時間與人力。FIB/SEM也適合做序列切片，而重建三維空間的影像。FIB/SEM也可製備用於TEM分析的薄片剝離技術 (lift-out)。 本研究聚焦於發展簡單而有效的實際方法，以運用FIB/SEM切削並觀察生物醫學標本。­本研究一開始的時候只採用冰凍法，直接處理標本。極為耗時耗力耗財，並不是很實用的方法。而且實際冰凍FIB切片結果，影像解析度並不佳。我們便開始思考是否有其他較具實用性的切片方法。 我們於是從三個方向著手：第一個方向，是標本的處理。第二個方向，是標本載台的選擇與製備。第三個方向，是切削方法與參數的嘗試。而本研究實際時用的研究標本，包括酵母菌，新鮮植物葉片的葉綠體，人類網膜色素上皮細胞胞株，人類白內障水晶體的前囊，人類的紅血球與白血球細胞，與電子顯微鏡量子點免疫染色等。 標本處理方向，我們研究了液態氮或戊二醛樣品固定、四氧化鋨後固定和硫代碳酰肼增強的各種不同組合與順序。標本載台選擇製備方向，研發包括金屬和半導體的各種不同基板。我們進行光蝕刻，在半導體板上產生圖案，以提高圖像分辨率。我們研究了不同類型的半導體基板，例如 N、N+、P、P+、GaN、GaN+ 等。更進一步轉印石墨烯薄膜在各種相應的半導體板基上。切削方法參數方向，我們嘗試運用不同的切削角度與照相角度，以及各種電壓電流切削標本。 結果我們發現，傳統TEM必須先使用環氧樹脂，做細胞組織樣本包埋才能切片，而在我們的系統上可以避免。傳統TEM必須使用鑽石刀，切片環氧樹脂包埋的樣本，也可以避免。因此我們的系統能大幅削減所需的人力物力與時間。標本的前處理，我們設計的新T-O-T-O-T方法，效果勝於傳統O-T-O-T-O方法。切削的標本載台，我們發現有幾種半導體基質的效果較佳，如N+, GaN+。我們發現具有轉印石墨烯薄膜的半導體基質，可以減低白點背景雜訊，也可以更加提高解析度。我們也發現使用低角度切削，效果比傳統的垂直切削更加優越。我們成功切削了酵母菌，葉綠體，網膜色素上皮細胞胞株，水晶體的前囊，紅血球與白血球細胞。我們完成了序列切片，也完成電子顯微鏡量子點免疫染色。 與傳統TEM相比，我們的FIB/SEM系統獲取的圖像，能顯示出相似的分辨率和對比度。我們從頭開始，一步步建立了可運用於生物醫學領域的 FIB/SEM 方法系統。這樣的系統，可以減輕TEM樣品製備與切割的沉重負擔，並實現了直接可視化下的切削需求，具有運用在臨床生物與醫學上的巨大潛力。

The focused ion beam (FIB) is an important and powerful tool in the semiconductor and materials science communities. Integrated circuits often require observation and repair of metal oxide semiconductor components, as well as the composition and configuration of connecting lines. A focused ion beam with appropriate energy can cut and mill integrated circuits, and at the same time, it can observe the ultrastructure with the built-in scanning electron microscope (SEM). Can FIB be used in the field of biology and medicine? This is an emerging research development project, and there are still not many related studies. Biomaterial configuration, pretreatment, and cutting parameters are still the research topics that must be explored. Using traditional transmission electron microscopy (TEM), biological tissues must first undergo a series of complex procedures such as fixation, dehydration, staining, plastic embedding, and sectioning. It is often necessary to do rough sections first, observe under an optical microscope, and then start to do ultra-fine sections, and then go on the machine to find the key points to see. But if you can't see the key points after getting on the machine, you must repeat the process of rough cutting and fine cutting, and then go on the machine again to observe. The whole step is very cumbersome and time-consuming. FIB/SEM can do in-situ site-specific sectioning and direct observation. That is to say, FIB can directly find key features under SEM, perform cutting, and directly observe the cut surface after cutting. Therefore, the time and manpower that the traditional TEM must consume can be greatly reduced. FIB/SEM is also suitable for serial sectioning and reconstruction of three-dimensional images. FIB/SEM can also prepare a lift-out technique for TEM analysis. This study focuses on developing simple and effective practical methods to cut and visualize biomedical specimens using FIB/SEM. In the beginning of this study, only the freezing method was used, and the specimens were directly processed. It is extremely time-consuming and labor-intensive, and it is not a very practical method. The actual frozen FIB section results in poor image resolution. We then began to think about whether there are other more practical slicing methods. So we started from three directions: The first direction is the processing of specimens. The second direction is the selection and preparation of the specimen stage. The third direction is the attempt of cutting methods and parameters. The actual research specimens used in this study include yeast, chloroplasts of fresh plant leaves, human retinal pigment epithelial cell line, human cataract lens anterior capsule, human erythrocytes and leukocytes, and electron microscopy quantum dot immunostaining, etc. . For specimen processing, we investigated various combinations and sequences of liquid nitrogen or glutaraldehyde sample fixation, osmium tetroxide postfixation, and thiocarbazide enhancement. The preparation direction of the specimen stage is selected, and various substrates including metals and semiconductors are developed. We perform photolithography to create patterns on semiconductor plates to increase image resolution. We study different types of semiconductor substrates such as N, N+, P, P+, GaN, GaN+, etc. Further transfer the graphene film on various corresponding semiconductor substrates. In the direction of cutting method parameters, we try to use different cutting angles and camera angles, as well as various voltages and currents to cut specimens. As a result, we found that traditional TEM must first use epoxy resin to embed cells and tissue samples before sectioning, which can be avoided on our system. Conventional TEM must use a diamond knife, which can also be avoided for slicing epoxy-embedded samples. Therefore, our system can greatly reduce the required manpower, material resources and time. For the pretreatment of specimens, the new T-O-T-O-T method we designed is more effective than the traditional O-T-O-T-O method. For the cut specimen stage, we found that several semiconductor substrates work better, such as N+, GaN+. We found that semiconductor substrates with transferred graphene films can reduce white spot background noise and further improve resolution. We have also found that using a low angle cut is more effective than a traditional vertical cut. We successfully dissected yeast, chloroplasts, retinal pigment epithelium cell lines, the anterior capsule of the lens, and red and white blood cells. We performed serial sections and also performed electron microscopy quantum dot immunostaining. Compared to conventional TEM, images acquired by our FIB/SEM system show similar resolution and contrast. We started from scratch and built step by step a FIB/SEM method system that can be applied in the biomedical field. Such a system can reduce the heavy burden of TEM sample preparation and cutting, and realize the cutting requirements under direct visualization, which has great potential for application in clinical biology and medicine.