6. ADVANCED PACKAGING, MICROASSEMBLY, AND TESTING TECHNIQUES
(Stephen C. Jacobsen)

International Priorities -- the Role of MEMS Technology

Together with other business and governmental activities, technology can contribute significantly to the solution of international problems. The means by which technology contributes is through the application of products and processes such as those listed for MEMS in Figure D.12. Such new products typically evolve by including emerging component-level subsystems developed by projects that utilize the new technology. The new subsystems are typically aimed at providing discriminating product features that improve performance, reliability and/or cost, and enhance other factors such as recyclability, safety, and use of strategic materials.

New Products

Increasing Complexity with Better Performance and Economy

. Modern machines are evolving toward increasing complexity levels. The trend, which is unavoidable, is driven by populations seeking new products and services, and a simultaneous imperative to address environmental and safety objectives. The architecture of the new machines is evolving away from collections of barely-compatible, unrelated components toward systems planned from the outset as integrated machines destined for a complete life cycle compatible with the goals mentioned above. The machines are information intensive and involve different computational and moving parts as outlined in Part a of Figure D.13 (upper left). The machines, which will be called here Information-Driven Machines that Move (IDMTMs), serve five basic functions: as models, sensors, controllers, effectors, and interfaces. The design and manufacture of IDMTMs require development of subcomponents of many types and from different disciplines, as depicted in Part b of Figure D.13 (upper right).

Product complexity is increasing across-the-board in many areas, ranging from consumer products to manufacturing machinery, as shown in Figure D.12. Part c of Figure D.13 (lower left) shows a progression in complexity in robotic systems, with sensors and actuators increasing from just a few to hundreds in only a decade. Part d of Figure D.13 shows that even in a subarea of the robot, the effectors (muscles), complexity has progressed substantially from just an actuator and sensor and some wire. Sensors, actuators, controllers, and structures must now be considered as not just randomly installed elements but as integral machine subsystems connected to computers and other peripherals over digital multiplexed buses.


Figure D.12. Application areas for microelectromechanical systems.


Figure D.13 (Parts a-d). Information-driven machines that move. Parts a, b, c, and d, are located in the upper left, upper right, lower left, and lower right, respectively.

As noted in the previous section, the architecture that defines the arrangement of the subsystems is also changing. Larger numbers of sensors and actuators are being interconnected to control higher modes of machine operation. Complex control requires a model-based approach, which necessitates an expansion of local chip-based control circuits. New methods for the arrangement and interpretation of sensor information are being developed, as shown in Part b of Figure D.14. Sensor and actuator design must now address economic constraints and target additional properties, as reviewed in Parts c and d of Figure D.14.

Very Small Machines

The new technology area of MEMS is focused on developing the design and fabrication resources necessary to produce small electromechanical structures that can be combined to manufacture Very Small Machines (VSM). The VSMs will function in a variety of physical regimes, including: mechanical, electrical, optical, thermal, chemical and others. In fact, VSMs have been designed and manufactured for years. The goal of VSMs has been to use size advantages (scaling laws) to achieve enhanced performance in areas such as speed, complex function, low weight, and packaging advantages. In most cases, however, small size has produced desired performance levels but has not achieved the cost and reliability levels necessary for widespread use in lower-level products.

Over the last few years, MEMS technology has been brewing in various laboratories. This approach should allow the economic manufacture of microsystems with the simultaneous achievement of reliability and performance goals.

MEMS (Microelectromechanical Systems)

MEMS can contribute significantly to the economic production of VSMs and other components for conventional products. As described in the earlier sections of this report, MEMS are very small systems fabricated using primarily silicon micromachining and VLSI processes. Such machines consist of mechanically and electrically active subelements that range in size from 0.5 to 500 mm. MEMS represents a new field that will revolutionize many products for military and commercial applications and should permit the development of better subsystems based on mechanical, electrical, fluid, thermal, chemical, optical, and other phenomena.

The generation of concepts, design tools, and fabrication approaches is already moving ahead. It should be noted here that MEMS and VSMs are not specific devices or processes, but approaches that include new design tools, fabrication processes, control approaches, and integration strategies. The appeal of the approach is soundly based in potential near-term economic advantages and associated competitive advantages.


Figure D.14. Issues in the development of sensors, actuators, and MEMS.

The basic advantages of MEMS are briefly listed in Figure D.15 along with its status and barriers. Advantages include not only cost and performance but also the prospect of practical networked systems. As discussed above in section 5 of this appendix, nets allow distributed computation for local interpretation and multiplexing, which reduces wiring and permits substantial improvements in design flexibility (i.e., components can be placed where convenient, do what is required, be commanded strategically, and be easier to maintain).

MEMS: Targets, Status, Barriers, and Approach

Development targets in laboratories around the world include: sensors, actuators, systems, design tools, packaging approaches, and integration strategies.

Sensor development projects under way include:

Actuator developments are focused on almost any material that will alter size or shape in response to an energetic input. For local self-management, many actuation schemes also include the placement of sensors for local actuator control. Approaches include:

o magnetic field o electrostatic field o electrostrictive
o magnetostrictive o piezoelectric bimorph o thermal expansion
o shape memory alloys o phase transition o impact
o chemical o polymeric o and others

Systems are being developed that include all elements of Figure D.13. These include sensors, actuators, structures, controllers, buses, and computers. Examples include:

o inertial platforms o chemical processors
o virtual reality systems o robots
o optical arrays for light control o fluid control systems
o vibration compensated machine tools o color displays
o chemical analytic instruments o printing systems
o embedded sensor-actuator structures o deformable RF antennas
o networked sensor arrays o medical devices
o smart materials - deformable o and others


Figure D.15. The advantages of MEMS along with its current status and remaining barriers.

Design tools and fabrication systems are also in development. Design tools that already exist are being modified for use in MEMS. New systems are being developed at MIT, the University of Michigan, MCNC, and other locations, to permit integrated design of systems with simultaneous consideration of mechanical, EM field, fluid, and other physical effects.

Developments in MEMS have taken the first steps. Products are on the market and many more commercial successes await only additional effort. Part b of Figure D.15 indicates the present status as well as barriers to future progress, which are in five development categories: design tools, fabrication resources, packaging and integration, systems architecture, and business.

The approach for future work must include a balanced view of the following:

o technology o projects
o components o products
o markets o businesses

Manufacturing and Packaging

Manufacturing and packaging issues now loom as substantial barriers to the real development of a large body of the concepts generated by the interested R&D and business communities. It is now being recognized that the approach must become broad-based with fabrication processes (forming, machining, coating, etc.) developed in anticipation of broader issues such as assembly procedures, packaging, and other product-related factors. Many processes (in addition to the classical VLSI-related ones) such as those listed in Figure D.16 are becoming available as they are in larger systems, and nontraditional job-shop resources, such as the effort at MCNC, are emerging.

Packaging of MEMS devices is more complicated than for their electronic or mechanical ancestors. New sensors and actuators must physically interact with the environment and thus require passthroughs for the transmission of fields, photons, fluids, moving shafts, chemicals, and so forth. In fact, packaging and assembly processes will dominate the economics of the final product as well as substantially determine factors related to ruggedness, reliability, and maintenance. For applications and devices requiring direct contact with the environment, recent developments in chip-level encapsulation appear promising in terms of collapsing the package to the chip itself and allowing access to specific transducers via photolithographically-formed openings in the passivating films. Future systems will also require systems with substantially greater levels of three dimensionality. Assembly and packaging methods for generating such systems are in development.


Figure D.16. Processes used in MEMS fabrication along with the various levels of environmental access required by different types of devices.


Published: September 1994; WTEC Hyper-Librarian