【作者】 孔祥东；

【导师】 张玉林；

【作者基本信息】 山东大学 ， 控制理论与控制工程， 2005， 博士

【摘要】 随着微机电系统的深入研究和快速发展,需要能够精确产生复杂曲面和各种形状微结构的加工方法与之相适应。当前能够制作三维微结构的技术主要有体硅微加工技术、LIGA(Lithographie,Galvanoformung andAbformung)技术和IH(Integrated Harden Polymer Stereo Lithography)三维光刻技术。体硅微加工技术和LIGA技术能够制作高精度、高深宽比的陡直微细结构,却难于加工含有各种微曲面和结构较为复杂的器件;IH三维光刻技术从理论上能够加工出含有任意曲面和任意高深宽比的复杂微结构,但因其工艺中x、y向的扫描是靠X/Y工作台的机械移动来完成的,所以加工精度较低,分辨率目前仅能达到亚微米级。因此这些微三维加工技术不能很好的适应将来微机电系统的高速发展,需要寻求更好的加工手段。 电子束曝光技术是目前公认的最好的高分辨率图形制作技术,主要用于0.1～0.5μm的超微细加工。在实验室条件下,已能将电子束聚焦成尺寸小于2nm的束斑,实现了纳米级曝光。电子束具有波长短、易于控制、精度高、灵活性大等优点,目前主要用于精密二维掩膜制作,但还不能制作三维立体结构。因此本论文率先提出了一种国内外均未见报道的、能够将电子束曝光的高精度性能和IH工艺能加工任意复杂器件的性能结合起来,既能加工出任意复杂的微结构又能保证其亚微米级甚至更高精度的三维掩膜/模板加工新方法——电子束液态曝光技术(Electron Beam LiquidLithography,EBLL)。作为一项开创性的工作,本论文的主要工作是首次提出了基于电子束曝光机(Electron Beams Exposure System,EBES)的电子束固化曝光微三维成型技术,从理论和实验上对其可行性以及对电子束的能量和曝光剂量与固化厚度之间的关系进行了论证研究,并在SDS-II型电子束曝光机上完成了以液态TMPTA和Epoxy618为固化对象的电子束固化实验,得到了固化的十字型微结构,充分证明了电子束液态曝光技术是切实可行的。本论文的主要工作和创新点如下:更多还原

【Abstract】 With the rapid progress and deep research of MEMS (Micro Electro Mechanical Systems), the technique to produce exactly complicated surface and arbitrary microstructure is needed. Now the methods of three-dimensional microfabrication are mainly bulk-silicon micromachining, LIGA (Lithographie, Galvanoformung and Abformung) and IH (Integrated Harden Polymer Stereo Lithography), etc. LIGA and bulk-silicon micromachining can produce steep microstructure with high-precision and high-aspect-ratio, but it is difficult for them to create micro devices which have complex curved surface and structure; theoretically, IH can produce devices with arbitrary curved-surface and high-aspect-ratio, but its scan in X-Y is finished by the moving of X-Y stations mechanically, the precision is low and only submicron resolution can be achieved . So these methods can not satisfy the real 3-D structures of MEMS, novel micro-fabrication methods should be developed.Electron beam lithography(EBL) is considered as the best technique to produce patterns with high resolution, and it is generally used to super microfabrication with resolution 0.1~0.5μm. In lab, Electron beam (EB) can be focused to less than 2 nm, and EBL has been used to exposure nanometer figures. EB has many advantages, such as short wavelength, easily be controlled, high precision and sensitivity, etc. Now it is mainly used to make precise masks and can’t produce real 3-D microstructures. Combining the high-precision of EBL and arbitrary high aspect ratio of IH, we can develop a novel method of electron beam liquid lithography (EBLL). This method is liquid lithography of EB, which can not only fabricate arbitrary micro structures, but also can keep sub-micron precision ,or even more precise. For an initiate research, the main study works of this dissertation are that the EBLL is presented for the first time in the world, the feasibility of EBLL andthe relations between curing depth and electron energy, exposure dose are investigated in detail. The curing experiments of liquid oligomer TMPTA and epoxy 618 are finished in SDS-II electron beam lithography system for the first time, and the cured structures, crosses, are got. All of above show that EBLL is feasible. The main study works and achievements are summed up as follows.1. Electron beam liquid lithography is presented. This technique combines the high-precision of EBL and arbitrary high aspect ratio of IH. It uses electron beams as the radiation source to cure liquid resist to create a complicated 3D structure of polymer by piling up the cured two-dimensional sliced structures. Its resist is liquid oligomer. Its main advantages are high-precision, high-resolution, arbitrary high aspect ratio and easily to be controlled. The disadvantage is low productivity. So it is optimum to produce 3-D microstructure templates.2. The lithography modes of EBLL are studied. The moving tracks of focused electron beams that come across metal film windows and the scattering process of the high-speed electrons in air are simulated by Monte Carlo method. The exposure method that educes the electron beam into the air to expose liquid resist is denied, and the method that exposes liquid resists in the electron beam lithography systems is chosen so that the focused thin beam-spot can be kept.3. The selective method of liquid resists suitable for liquid lithography is proposed. The step of the first method is as follows. First, we draw therelation curve between lnP and — through published reference data, thencalculate saturated vapor pressure values at different temperatures. Second, we consider these saturated vapor pressure values as references, and validate the valuse through experiments. At last we are able to decide whether the liquid ologomer is chosen to be used as the resist by the experiment results. The second method is different. We perform an experiment on vacuumvolatilization in SDV-II directly, without reference data to look up. Then make an decision whether the liquid ologomer is suitable to EBLL. We can get the suitable liquid resist easily by the two methods.4. The relationships among mean free path, energy loss ratio of high-speed electron, and vacuum degree is studied. The lowest vacuum degree of the SDS-II lithography system is calculated firstly, and the lower limit lxlO"4 Torr (about 0.01 Pa) of the vacuum degree that SDS-II system can work well is got. Under above degree, the mean free path of the electron is 283cm which is 3 times longer than the length of the drawtube of the lithography system, and about 80% electrons from cathode can get to the center of the object lens without any collision, 98% electrons from object lens can get to the substrate without any collision. So these electrons can be focused into fine spot on the substrate.5. The gasification of kinds of resists at 0.005Pa (vacuum degree is a little higher than the lower limit lxlO"4 Torr of SDS-II) are experimented through SDV-II vacuum chamber made by ourselves. The results show that the volatilization ratios of TMPTA and Epoxy618 are lower than 0.5% and 0.1% respectively, and these two liquid resists do not destroy vacuum degree of the vacuum chamber. So they both can suit for the exposure and cure in the lithography system.6. Whether the energy and dose of the electron beam provided by the SDS-II can meet with the radiant and curing demand of the liquid resist is investigated. The result is that the electron beams provided by the SDS-II can fully meet with the curing demand, and it is validated by experiments.7. The applicability of the Monte Carlo simulation method and Grun formula to liquid resists is researched, and it is validated by experiment on TMPTA and Epoxy618. The result shows that we can forecast the valid penetrating depth of high energy electrons in liquid resists by using these two methods.8. The mechanism of liquid resist’s spin-coating is studied. The relationamong the resist thickness, viscosity and the rotating speed of motor is got qualitatively. Also we quantificationally studied the above relation through experiment on special liquid resists. The rule is as follows: the lower of the viscosity and the higher of the rotating speed, the thinner of the resist coating, and vice versa. The result provides the valid reference basis to obtain the coating thickness exactly.9. The impact of the electron’s energy and radiant dose on curing thickness is researched experimentally and theoretically. The results show that either increasing initial energy of the electron beams (increasing accelerating voltage of the electron beams) or increasing the incident dose of the electron beams can increase the curing thickness. According to the results, we got the experimental relation curve of special liquid resists among electrons’ energy, radiant dose and curing depth. It supplies the reference to cure different thick coating. Based on these, the exposure experiments on special liquid resists are finished firstly through the method of EBLL. The cured crisscross microstructure is got, which fully proves that the technique of curing and exposure with electron beam is feasible.At last, the main work of the paper is summarized, and the further research is proposed.更多还原