|
 |
|
|
 |
 |
 |
 |
 |  |
 |
 |
 |
 |
 |
MANCEF Products:
Roadmap
First Edition of International Micro-Nano Roadmap Executive Summary Download the order form (.pdf) Structure of the International Micro-Nano Roadmap The roadmap is organized into three sections with fifteen chapters. The sections and chapters are structured as follows: Section One:
Section Two:
Section Three:
Major Findings Our contributors have provided many salient conclusions throughout our chapters. Some of the roadmap information supports the current Microsystems technologies as the potential next "Big Commercial Opportunity." Others identify bottlenecks and roadblocks. The nature of a roadmap is that all chapters are interdependent and many of the major findings are voiced in differing ways across the body of the work. We specifically highlight fifteen major conclusions our contributors have made in the IMR. Roadmap Chapters Section I Chapter 2: Commercialization of Microsystems Industries based on novel technologies are bereft with impediments delaying their creation. Many times emerging technologies commercial development may be buried in a single firm's pursuit of a singular solution for a specific customer. The creation of an initial product is not necessarily the harbinger of a dynamic industrial or market formation or even for that matter the birth of a new firm. In order for a technology to form the basis of a new industrial revolution, it must answer the question of an initial commercial application and follow that by demonstrating its ability to solve similar problems in a uniquely commercially valuable manner. Cadres of technology product platforms are being developed that are commercially effective across multiple technology product domains. This chapter utilizes the historical data resident in the Microsystems revolution modifying it through state-of-the-art technology commercialization methods to provide knowledge and wisdom to the reader while shedding light on the various aspects of the current and future state of Microsystems, the Second Micro-Manufacturing Revolution. A complete, succinct overview of the current state of commercialization in the industry is provided through the use of a MEMS report card. Lastly, we provide models that illustrate the current state and the future promise of Microsystems, as well as indicate the scope and future of Small Tech. Chapter 3: Optical MEMS Optical Microsystems hold much of the same promise as all of the MEMS-based Microsystems. Optical MEMS have made great strides in telecommunication, sensing, and projection applications. Contributors to the roadmap early on described an optical switch as a "Holy Grail" application with huge market potential. Our contributors envisioned Optical MEMS (also referred to as MOEMS) as a "Market Driver" for the optics and telecommunications industries. However, in recent years it has been observed that the first MEMS device to achieve widespread acceptance in telecom has been the optical attenuator. While these devices do not share the complexity of the high port-count switch, they have enabled the acceptance of MEMS as a "Real" technology. Telecom MEMS have evolved from an esoteric university research topic to real packaged products in the space of approximately 5 years. Outside of the technological achievement that this represents, it is a welcome sign that the industry is accepting micro-mechanical technology as an enabling technology for telecom. TI and others are famous for having initiated Optical MEMS with the Digital Micro-Mirror Device (DMD) on which TI's DLS technology is based. The potential of an all-optical switching device for the telecommunications industry has greatly increased the interest in Optical MEMS and is poised to be one of Microsystems' first "Killer Applications." Unfortunately, the equivalent of an economic winter has blanketed the telecommunications industry since 2000, lessening its desire for all optical switching solutions over the very near term. Most analysts feel that the demand for all optical switches is not over, but simply has been delayed by a few years. As well, it is important to note that even in the telecom downturn (and the associated cutbacks in new product development), MEMS are emerging in many important, but less "sexy", roles such as optical attenuation (Nortel], JDSU, LightConnect and others), small port switches (JDSU and Kymata) protection networks, tunable RF (multiple firms such as Raytheon.), and integrated optics (Silex). However, telecommunication companies are failing daily and the negative impact on the Optical MEMS market is severe and profound. Chapter 4: BioMEMS MEMS in the medical-biochemical fields are presently poised to surpass MEMS applications in other areas, in terms of market revenue. Until the mid 1990's, MEMS techniques in the medical-biochemical fields were usually associated with blood pressure sensors. In the past 5 years, a host of other less-publicized BioMEMS based devices are being used in medical equipment or have been prototyped and are nearing entry into the market. Thus, estimates place the percentage occupied by BioMEMS in the total MEMS market at 40-45% of the total MEMS market in 2000 and see the total BioMEMS market rising to $445 billion in the 2004 - 2005 timeframe. The BioMEMS market can be divided into in-vivo and in-vitro segments. The in-vivo or "inside-the human body" includes BioMEMS devices like micro motors, retinal implants, micro catheters, etc., and are significantly small compared to the in-vitro segment or "outside-the human body". In-vitro devices range from body fluid microanalysis devices to microsurgical equipment and account for the bulk of the present BioMEMS market. The only BioMEMS device that can be expected to have a billion dollar revenue is the biochip, an in-vitro device in which bio-molecules are chemically analyzed on a credit card sized chip as if it were a full-fledged laboratory. The value of the bio-chip market in 2000 was $500 million, but is expected to be $3 billion by 2004 or 2005. However, the subset of BioMEMS applications is large with several applications expected to surpass the $500 million dollar mark by 2005 - 2006. The BioMEMS market is relatively new and has not reached maturity and there is still enough room for new players. The only BioMEMS product that has a significant number of competitive companies is DNA Array. The majority of most other BioMEMS device based companies have few market entries. Niche BioMEMS based markets that were not visualized before are being developed continuously. BioMEMS devices are increasingly acting as technological, as well as, market drivers for Microsystems. The in-vivo and in-vitro applications that make up the BioMEMS family of products are currently utilizing considerably different front-end and packaging technologies. Material requirements suggest that this dynamically growing Microsystems market area will use many differing technologies in the future. Unlike other MEMS markets that use silicon as the primary material, BioMEMS devices encompass both IC and Non-IC compatible technologies more broadly. Non-IC based technologies in BioMEMS have been developed in Europe, North America and Asia. LIGA and plastic extrusion technologies are led in Europe by institutes like Germany's Forschung Zentrum Karlsruhe (FZK) and in North American by leaders such as Sandia National Laboratories and the Center for Advanced Structures and Devices (CAMD) in Louisiana. Chapter 5: An Improved Method and Forecast for the World-Wide Market Growth of MEMS The roadmap provides a forecast based on expert opinions for figures representing the sum of the minima and maxima of the twenty-six major markets identified for MEMS. Current studies show a tremendous variation in future sales and so do our experts opinions. This chapter provides a Monte Carlo simulation model to help determine ranges for the studies. In the short run, maximum forecasts can diverge by a few hundred percent in each market. Whereas by 2025, the divergence between minimum and maximum forecasts are between one and two orders of magnitude for all applications considered. The forecast for current global sales ranges between $600 million and $31 billion. By 2025, the forecast for global sales of MEMS ranges between $9 billion and $360 billion! Section II There are at least seven front-end manufacturing technologies falling into three classes: traditional bulk micromachining, sacrificial surface micromachining, and high-aspect ratio micromachining (HARM). The latter includes deep ultraviolet (DUV) lithography techniques and x-ray-based methods such as LIGA (from the German Lithographie, Galvanoformung, Aformung, meaning lithography, electroforming or plating, and molding). The MEMS roadmap divides these three classes of manufacturing technologies into two categories: IC and potentially IC like" vs. non-IC like technologies. Non-IC like includes spark erosion, LIGA, silicon sculpting, and processes incorporating certain incompatible materials. These manufacturing processes are complex, and the trend is toward increased complexity. By introducing substantially greater demands for fabrication with structural features in the third dimension, MEMS manufacturing technology will incorporate new capabilities, such as deep reactive-ion etching (DRIE), double-side aligners, stiction-abatement equipment, back-end flipchip equipment, and bonded-wafer technology. MEMS manufacturing process development will borrow from advances in IC fabrication, such as increased aspect ratios and reduced line widths, added metallization layer(s), chemical mechanical polishing (CMP), more sacrificial and structural layers, use of exotic materials and non-silicon based technologies, vastly more process steps, and increased wafer diameter. Although the traditional approach in MEMS has been to develop application specific front-end manufacturing processes, the trend is reversing and contributors to the roadmap see a few dominant front-end process technologies emerging. Chapter 6: IC Compatible and Potentially IC-Compatible Microsystems Manufacturing Approximately fifty years ago, what is now known as Microsystems, MEMS or Micro machining was born of the use of semiconductor micro-fabrication techniques for mechanical applications. The first micro-manufacturing industrial revolution gave rise to the second while the first was still in its infancy. And thus was born the use of IC and IC-like manufacturing processes for Microsystems. This has given rise to the axiom "if it isn't going to be first manufactured using IC-related technologies in Microsystems, it is likely not ever to be produced." The majority of current MEMS based products are produced in this manner. The flagship of this technology segment is sacrificial surface micro machining which is directly discussed in the Integration chapter. The Analog Automotive Accelerometer, and the Digital Light Machine technology from TI are examples of IC like manufactured devices. Chapter 7: Non-IC Compatible Microsystems Manufacturing Non-IC like processes do not suffer the same barriers that IC like MEMS processes endure, but then neither do they benefit from the man-years of experience inherent in utilizing a semiconductor micro-fabrication based process. Nevertheless, these processes can be bifurcated between those, which are revolutionary in nature, and those, which are more evolutionary. Non-IC like processes can be split into four groups. They are those that utilize; a current process, evolutionarily (incrementally) improves an existing process, a new process centered on existing materials and a new process with new materials and tools required. Non-IC compatible Microsystems manufacturing systems cannot leverage or piggyback the semiconductor Microsystems marketplace to the same extent that IC like MEMS technologies do. They can and do leverage knowledge in the plating arenas and plastic extrusion technologies, but in general they have a steeper learning curve than their IC -like MEMS manufacturing cousins. They do have some materials and structural advantages in the MEMS manufacturing world. Non-IC like MEMS have the potential to build some of the most cost effective MEMS devices. Non-IC like MEMS processes make both the "Biggest Mass" small devices and can utilize the widest range of materials such as bio-compatible materials suitable for in-vivo manufacturing. A small but rapidly growing group of companies, provide, utilize or base their search for competitive advantage on these Non IC compatible and very promising processing steps. They include; Microparts (EU), QSI (NA) and Twente Microproducts (EU). Foundries such as Axsun (see photo) and technology factor developers such as FZK are based on Non-IC like technologies. The Non IC compatible industry is developing its own support infrastructure with firms such as Jenoptik (EU). The health industry is the emergent industry for these technologies with firms like Mannesmann (EU), ISTAT (NA) and many other medical systems integrators purchasing and investigating products based on this technology Chapter 8: Design, Simulation, and Modeling Microsystems engineers are increasingly turning to commercial tools designed specifically for MEMS devices. Unlike microelectronic devices, MEMS are not laid out in street-grid geometry, so MEMS based tools allow the designers to work in either 2D or 3D and incorporate process information to convert between them. These tools provide a seamless path to 3D analysis tools and contain a robust library of standard parts and components. These tools build a visual model of the final product using process information and system definitions. The trend in tool development will increasingly be to provide both 3D representation and arbitrary cross-sectional capability performance. These tools will be instrumental in shortening MEMS product development cycles and in improving yield during the manufacturing phase. Design Rule Checkers (DRCs) will ensure that MEMS designs do not violate the foundry's fabrication rules. These tools will provide feedback about aspect ratios, etch release systems, side wall angles, process variability, mechanical tolerances, and material properties. Chapter 9: Microsystems Reliability, Testing, and Metrology MEMS reliability and device commercialization go hand-in-hand. MEMS devices, which have been successfully tested for functionality and reliability, provide the basis for Microsystems commercial successes. Pressure sensors, optical switches, and other devices are examples of such commercial success. In-process reliability, in robust manufacturing processes, provides the basis for critical infrastructure in the future. Process quality and the manufacturing practice itself influence reliability. New standards and methods for long term reliability testing must be developed to bring MEMS process yields and reliability to the same levels enjoyed by IC product manufacturers. The "In-use" reliability of current state-of-the-art MEMS devices demonstrates the clear need for defining such standard procedures for measurements of various parameters such as fracture strength and wear resistance. MEMS devices currently are able to sense, think, act, communicate, navigate and/or self-power many different energy domains and environmental conditions. Their ability to continue functioning in their designed manner for the designed lifetime when subjected to varying degrees of mechanical, thermal, chemical, electromagnetic, frictional, tribological, or other stress is the ultimate measure of reliability. The smaller dimensionality and the multiple physical and energy domains of MEMS operations make reliability issues far more complex than for products like integrated circuits, which operate primarily in a single (electrical) domain. Furthermore, materials in the micro domain behave differently than their macroscopic counterparts. Material properties such strength, stiffness, and wear are significantly impacted by scaling effects resulting in increased reliability issues and complexities of analysis. Chapter 10: MEMS Packaging and Assembly Packaging constitutes the single largest element of cost and a major limitation to the miniaturization potential of MEMS devices. Effective packaging solutions remain a major obstacle to MEMS commercialization. Materials, design, assembly methods, encapsulation processes, environmental concerns, and functionality are the major challenges facing effective packaging solutions. MEMS applications frequently challenge packaging engineers to accommodate a variety of increasingly complex operating environments. Major limitations of current MEMS packaging technology include a lack of standards or standardized packaging methods. Currently, few MEMS devices incorporate concurrent designs of the device and package, typically resulting in the use of retrofitted IC packaging solutions. The trend is towards increased use of a concurrent design. The trend among MEMS packaging engineers is moving beyond reliance on IC standards. MEMS packaging still needs to address the traditional IC packaging problems, as well as the need to accommodate the mechanical extension of the MEMS device. MEMS packaging often plays the dominant role in determining the cost, reliability, and accuracy of the completed MEMS. Section III Chapter 11: Status and Future of Microsystems/MEMS Foundries The challenge for MEMS manufacturing lies in cost-effective fabrication of small volumes, and the preparation for ramp-up to high volumes. MEMS is still an emerging and disruptive process technology that is not yet standardized to any great breadth or depth. Dominant MEMS manufacturing technologies in many application spaces are still to be determined. Therefore, MEMS manufacturers must be prepared to deal with multiple wafer materials, various shapes and sizes, as well as supporting multiple process technologies. Furthermore, the cost of any new MEMS foundry infrastructure is very high. Fortunately, last-generation IC foundries can be used for fabricating present-generation MEMS. However, as MEMS devices mature, becoming more complex, many contributors foresee more MEMS specific foundries offering unique process capabilities. The roadmap contributors suggest a model similar to the fabled semiconductor concept. Two scenarios will potentially evolve:
International (SEMI) has long been a residual source for semiconductor-based standards reference, but is also emerging as an important standard-setting body for the MEMS industry. The multitude of foundries demonstrates the need for standards. Standards unofficial or official already are being set at the process step level. Yet, there are still many ways to manufacture a MEMS device. Sacrificial surface micro machining has at least five different processes noted as initially MEMS in the middle, MEMS up front and MEMS at the end, with bonded wafer and silicon germanium alternatives. The CMOS standard for sacrificial surface micro-machining is still far off. However, efforts at Sandia National Laboratories and DARPA have initiated efforts along this path in IC like arenas with the SUMMIT and MUMPS processes and programs. Chapter 12: MEMS/MST Cost Model The MEMS/MST Cost Model chapter contributors also contributed greatly to the foundry and the manufacturing technologies chapters. One of our contributors highlights the differences between capital and operating costs for different selections of front-end MEMS manufacturing technologies. The contributors of this chapter discuss both actual and option-based costs. The chapter points out the cost of choosing a non-dominate technological path or a pathway that may not be optimal for the given product technology paradigm or application space. In that vein, commercial foundries are likely to play an even more important role in the future because it is the foundry, and not the firm, that has to bear the burden of the technological pathway choice. Bulk, High Aspect Silicon Etching (HARSE) and SSM (Surface Sacrificial Micro-Machining) as shown in Figure 14 are all forms of MEMS front-end manufacturing technologies. Chapter 13: Standards, or Lack Thereof There are more norms than standards in the industry because it is still an emerging technology. The trend towards integrated MEMS systems becomes more important. Standardization in the MEMS field will accelerate as growth propels more applications into the billion-dollar orbit. Many user societies such as SAE have form, power, fit and function standards to which MEMS devices already conform. Establishment of standards in all aspects of MEMS will accelerate development of new applications. Standards efforts face industry resistance as industry leaders hesitate to share their hard won knowledge that may constitute competitive advantage. Active standardization committees already exist for materials, equipment, design, modeling, test, interconnections, processes, and packaging. Standards for materials and equipment have already started to appear; with design, modeling and test standards imminent; to be followed by interconnections, process and packaging standards. Semiconductor Equipment Manufacturing International (SEMI) has long been a residual source for semiconductor-based standards reference, but it is also emerging as an important standard-setting body for the MEMS industry. The multitude of foundries demonstrates the need for standards. Standards unofficial or official already are being set at the process step level. Yet, there are still many ways to manufacture a MEMS device. Sacrificial surface micro machining has at least five different processes noted as initially MEMS in the middle, MEMS up front, and MEMS at the end, with bonded wafer and silicon germanium alternatives. The CMOS standard for sacrificial surface micro-machining is still far off. However, efforts at Sandia National Laboratories and DARPA have initiated efforts along this path in IC-like arenas with the SUMMIT and MUMPS processes and programs. Chapter 14: Integration Trends point toward more highly integrated system solutions. This trend is toward developing more complex devices that sense, think, act, communicate, and/or navigate, rather than simple sensing or actuating. The choice of whether to integrate the control and signal conditioning electronics on the same chip, as the MEMS elements do (monolithic integration), or to use hybrid packaging technology is now one of the most contentious issues in MEMS technology. The ultimate decision will result from effective Microsystems design balancing tradeoffs between device performance, manufacturing complexity, and yield. Placing both the MEMS and the IC solutions on one chip approximately doubles the cost of silicon and limits manufacturing flexibility. But, integration reduces packaging costs while providing performance advantages. Sometimes monolithic integration is physically impossible due to the incompatibilities of IC fabrication technology and MEMS fabrication technology. When this occurs, hybrid solutions satisfy the requirements and/or provide superior design. Monolithic integration must overcome yield concerns and must be cheaper to produce or perform significantly better than hybrid solutions. Sometimes the application itself is not conducive to any integrated solution Chapter 15: Glossary One of the most difficult tasks in micro and nano technologies is to agree on a consistent use of terms. Indeed, a common vocabulary is often the first step in developing norms or standards in an industry. This is even true for the terms describing the field: MEMS, Microsystems, MicroMachining, M3, and Top-Down Nanotech. Since they are defined differently depending on which continent one works, whether your emphasis is on micro versus nano. This is further complicated in the Microsystems arena since devices function in many differing energy and environmental domains. Some symbols are used differently and mean different things in those domains, and a higher level of sophisticated understanding becomes involved in the field. These differences at basic levels reflect, in turn, differing industry scopes. Difference in Microsystems nomenclature can cause wide variation in the comprehension of the topic. The IMR Glossary is designed to provide a comprehensive, definitional effort for terms used in this emerging industry. We have provided operational terms, definitions used in differing application spaces, processing jargon, regional terms, management terminology, statistical terms, test, packaging and reliability terminology and many others. Roadmap Sponsors
Download the order form (.pdf) © MANCEF - Micro And Nanotechnology
Commercialization Education Foundation |
