Models to determine the process parameters required to sculpt desired micro-feature topographies on flat and curved surfaces using abrasive jet technology

由空气或水推动的小磨料颗粒射流多年来一直用于改变工程表面的地形。磨料射流技术（AJT）最重要的应用之一是作为微流体、微机电系统（MEMS）和光电元件的低成本和快速微加工平台。为此，本文将重点研究两种AJT：空气驱动磨料射流微加工（AJM）和磨料水射流微加工（AWJM）。这些技术的优势有很多，但也许最重要的是它们可以在不产生热影响区的情况下进行加工，并且它们具有大多数竞争技术所不具备的独特的定向蚀刻能力。例如，传统的各向同性湿法蚀刻通道会产生一个基本的u形微通道横截面。然而，AJT的定向蚀刻能力允许通过改变工艺参数（例如喷射扫描速度和倾斜角，粒度等）雕刻许多不同的形状。拟议的研究将利用这一独特的能力，允许该技术用于制造新型设备。我们之前已经为各种材料的AJM开发了“表面演化”模型，可以预测在各种工艺参数组合下最初平坦表面上的加工形貌的发展。然而，下一代微流体和MEMS器件将需要微加工3D（即非平面）组件，这是AJM尚未探索的领域，尽管其潜力巨大。同样，目前也没有表面演化模型来预测AWJM的加工形貌。AWJM是一种较新的工艺，由于磨料浆回流效应和缺乏掩膜，它与AJM有着根本的不同。通过与工业合作伙伴的独家协议，我们拥有一个带有独特微喷嘴的AWJM装置，这将使我们能够在这一领域进行开创性的研究。AJT的一个复杂因素是颗粒嵌入表面的趋势，从而影响表面质量，侵蚀速率，粗糙度等。目前，还没有模型来预测使用AJT加工金属时，什么颗粒和工艺参数控制这种嵌入的程度。提议的研究的最初部分将集中于解决AJT过程建模中的这些重要缺点。表面演化模型很重要，因为它们可以预测加工后的形貌作为输入工艺参数的函数；然而，目前还没有解决反问题的技术，即预测雕刻特定所需地形所需的工艺参数。该研究的最后一部分将解决这个重要的问题，这将允许使用AJT雕刻特定所需的特征形状。换句话说，我们将开发一种方法，允许表面演化方程的输入（工艺参数）在未来某个时间从方程的期望解（期望的横截面轮廓）中确定。由于曲面演化偏微分方程是非线性的，不能用封闭形式求解，因此该问题具有挑战性。最初，优化程序将用于确定最接近所需地形的参数集。后来，新的技术雕刻所需的形状的表面使用倾斜和垂直入射喷嘴的组合将发展。这些技术将为3D MEMS和微流体器件的设计开辟大量新的器件设计机会，从而支持加拿大不断发展的微技术领域。

Jets of small abrasive particles propelled by air or water have been used for many years to modify the topography of engineered surfaces. One of the most important applications of abrasive jet technology (AJT) is as a low cost and rapid micro-fabrication platform for microfluidics, micro-electomechanical systems (MEMS), and opto electronics components. The proposed research will focus on two AJT's for this purpose: air driven abrasive jet micro-machining (AJM) and abrasive waterjet micro-machining (AWJM). The advantages of these technologies are many, but perhaps the most important are that they can machine without creating a heat affected zone, and that they have a unique directional etch capability that most competing technologies do not. For example, traditional isotropic wet etching of channels results in a single basic U-shaped micro-channel cross-section. The directional etch capability of AJT, however, allows the sculpting of many different shapes by changing the process parameters (e.g. jet scan speed and inclination angle, particle size, etc). The proposed research will exploit this unique capability, allowing the technology to be used in the manufacture of novel devices. We have previously developed "surface evolution" models for the AJM of a wide variety of materials that can predict the development of machined topography on an initially flat surface for various combinations of process parameters. The next generation of microfluidic and MEMS devices, however, will require micro-machining 3D (i.e. non-planar) components, an area that has not yet been explored for AJM, despite its great potential. Similarly, there is currently no surface evolution model to predict machined topography using AWJM, a newer process that is fundamentally different than AJM because of abrasive slurry backflow effects and the lack of a mask. Through an exclusive agreement with an industrial partner, we have an AWJM setup with a unique micro-nozzle that will allow us to do ground-breaking research in this area. A complicating factor for AJT is the tendency for particles to embed into the surface and thus affect the surface quality, erosion rate, roughness, etc. Currently, no model exists for predicting what particle and process parameters control the extent of this embedding when machining metals using AJT. The initial portion of the proposed research will focus on addressing these important shortcomings in the modeling of AJT processes. Surface evolution models are important because they can predict machined topography as a function of input process parameters; however, there are currently no techniques to solve the inverse problem, i.e. predicting the process parameters necessary for sculpting particular desired topographies. The final portion of the proposed research will tackle this important problem that would allow the sculpting of particular desired feature shapes using AJT. In other words, we will develop methodologies that allow the inputs (the process parameters) to the surface evolution equation to be determined from a desired solution of the equation at some future time (the desired cross sectional profile). The problem is challenging because the surface evolution partial differential equation is nonlinear and cannot be solved in closed form. Initially, optimization routines will be used to determine the set of parameters that comes closest to a desired topography. Later, novel techniques for sculpting surfaces of desired shapes using combinations of inclined and perpendicular incidence nozzles will developed. These techniques will open up a host of new device design opportunities for the design of 3D MEMS and microfluidics devices, and thus support Canada's growing micro-technology sector.