In-depth Explanation of MEMS Technology and Its Applications

What are MEMS Devices and Their Applications?

MEMS (Micro-Electro-Mechanical Systems) devices are primarily used in the automotive and consumer electronics industries. In the future, they are expected to become more prevalent in medical, industrial, and aerospace markets as well. What exactly is MEMS? What are its characteristics? What challenges are there in designing and manufacturing MEMS devices? This article will answer these questions.

What is MEMS?

Micro-Electro-Mechanical Systems (MEMS), also known as Microsystems in Europe or Micro-Machinery in Japan, are a class of devices characterized by their tiny size and unique manufacturing process. The characteristic length of MEMS devices ranges from 1 millimeter to 1 micrometer—1 micrometer is much smaller than the diameter of a human hair.

MEMS often use microscopic counterparts to common mechanical parts and tools. These may include channels, holes, cantilevers, membranes, cavities, and other structures. However, MEMS manufacturing is not mechanical in nature. Instead, MEMS devices are produced using batch-processing micro-manufacturing technologies similar to those used in integrated circuits (ICs).

Today, many products utilize MEMS technology, including micro heat exchangers, inkjet printheads, micro-mirror arrays for high-definition projectors, pressure sensors, and infrared detectors.

Why Do We Need MEMS?

“They told me about a tiny motor the size of a fingernail. They told me that there is a device on the market through which you can write on the tip of a pin. But that’s nothing—it’s the smallest step in a direction I am going to pause. Beneath this is an astonishingly small world. When they look back from the year 2000, they will wonder why it took until 1960 for anyone to seriously start working in this direction.”
— Richard Feynman, “There’s Plenty of Room at the Bottom,” presented at the American Physical Society meeting at Caltech, December 29, 1959.

In his famous talk, “There’s Plenty of Room at the Bottom,” Richard Feynman described how we could write the entire Encyclopedia Britannica on the tip of a needle. But we might ask: why do we need to create these objects at such a tiny scale?

MEMS devices can perform many of the same tasks as their larger counterparts while offering several unique advantages. The most obvious of these is miniaturization. As mentioned earlier, MEMS devices are small enough to be mass-produced using batch manufacturing processes similar to those used in the integrated circuit industry. This mass manufacturing significantly reduces the cost of large-scale production. Additionally, MEMS devices typically require very small amounts of material for manufacturing, further lowering costs.

Besides being more cost-effective, MEMS devices can also be applied in areas where larger alternatives cannot. For instance, in smartphones, cameras, airbag control units, or similar compact devices, it would be impossible to design an accelerometer using traditional metal balls and springs. However, by shrinking the scale by several orders of magnitude, MEMS devices can be used in applications where traditional sensors would not fit.

Another advantage of MEMS technology is its ease of integration. Since MEMS use manufacturing processes similar to those used for ASICs (Application-Specific Integrated Circuits), MEMS structures can be more easily integrated with microelectronics. While integrating MEMS with CMOS (Complementary Metal-Oxide-Semiconductor) structures in a truly integrated device is challenging, it is increasingly being realized. Many manufacturers are using hybrid methods to create cost-effective and commercially successful MEMS products.

An example of this is Texas Instruments’ Digital Micromirror Device (DMD), which is core to their DLP® (Digital Light Processing) technology and widely used in commercial and educational projectors and digital cinemas. Each 16-micrometer micromirror is electrostatically actuated by the potential difference between it and its underlying CMOS storage unit. Grayscale images are created by the on/off states of the mirrors, and color is added through either a three-chip system (one for each primary color) or a single-chip system with a color wheel or RGB LED light source.

Perhaps one of the most fascinating features of MEMS technology is how designers can exploit the unique physics of such small scales. This will be explored further later.

Applications of MEMS

For various reasons, many MEMS products have achieved significant commercial success, and numerous devices are now widely used. The automotive industry is one of the main drivers of MEMS technology. For example, MEMS-based vibration structure gyroscopes, a new and affordable device, are currently used in anti-lock braking systems or electronic stability control systems. Murata’s SCX series MEMS accelerometers, gyroscopes, and inclinometer sensors, as well as those that integrate these functions into a single chip, support high-precision automotive applications. MEMS-based airbag sensors have replaced mechanical collision sensors in nearly all cars since the 1990s. Figure 2 shows a simplified MEMS accelerometer, similar to those used in collision sensors. A cantilever with a mass block attached to one or more fixed points acts as a spring. When the sensor accelerates along the beam’s axis, the beam moves a certain distance, which can be measured through changes in capacitance between the beam’s “teeth” and an external fixed conductor.

Many commercial and industrial inkjet printers use MEMS-based printheads, which store ink droplets and precisely dispense them when needed—this technology is known as drop-on-demand (DoD). The ink droplets are placed across piezoelectric materials (such as lead zirconate titanate), and a voltage is applied to squeeze the ink. This increases pressure in the printhead’s ink chamber, forcing out a small amount of ink from the nozzle.

Meanwhile, some MEMS technologies are just beginning to enter the market at large scale. Micro-mechanical relays (MMRs), like those developed by Omron, are faster, more efficient, and have unprecedented levels of integration. Omron has leveraged its expertise in MEMS to bring to market new temperature sensors: the D6T non-contact MEMS temperature sensor. The D6T MEMS sensor integrates an ASIC and thermopile elements into a compact non-contact temperature sensor, measuring just 18×14×8.8 mm (for a 4×4 element type).

MEMS Design and Manufacturing

“It’s interesting what problems small machines encounter. First, if the pressures in the various parts remain the same, the force decreases as the area shrinks, so weight and inertia become negligible. In other words, the strength of the material becomes more important. For example, as we reduce the size, unless the rotational speed increases proportionally, the centrifugal force caused by the flywheel will need to maintain the same proportion of pressure and expansion.”
— Richard Feynman, “There’s Plenty of Room at the Bottom”

Scaling and Miniaturization

An introduction to MEMS design and manufacturing often begins with a review of scaling and miniaturization. For example, if we ask why we cannot simply shrink an air compressor or ceiling fan down to flea size, the answer lies in the laws of compression. A flea-sized fan operates differently than a fan 1,000 times its normal size because the forces involved change. The scaling factor, S, helps us understand what happens in such a case.

Consider a rectangle with an area equal to the product of its length and width. If the rectangle is scaled down by a factor of 100 (i.e., both the length and width are divided by 100), its area will shrink by a factor of (1/100)^2 = 1/10,000. Therefore, the scaling factor for the area is S². Similarly, the scaling factor for volume is S³—meaning that as the scale shrinks, the impact of volume becomes much greater than the impact of surface area.

On a given scale, carefully considering the scaling factors of different forces can reveal the most relevant physical phenomena. Surface tension scales with S¹, pressure and electrostatic forces with S², magnetic forces with S³, and gravity with S⁴. This explains why a water strider can walk on water, and why the performance of ball bearings is different from that of a binary star system. While complete mathematical models are necessary for any design, scaling factors guide us in designing MEMS-sized devices.

Subsystem Modeling

Since submillimeter devices are often unintuitive, modeling is essential in MEMS design. Generally speaking, a complete MEMS system is too complex to be modeled as a whole, so the system is typically divided into multiple subsystems.

One approach to subsystem modeling is classification by function, such as sensors, actuators, microelectronic components, and mechanical structures. Lumped-element modeling follows this approach, representing the system’s physical parts as idealized separated components. Electrical circuits are modeled in the same way, using idealized resistors, capacitors, diodes, and various complex elements. In many cases, electrical engineers simplify circuit modeling using Kirchhoff’s laws instead of Maxwell’s equations.

Similar to electronics, systems can be modeled abstractly using block diagrams. At this level, the physical properties of each component are set aside, and only transfer functions are used to describe the system. This type of MEMS model is more suited to control theory, a vital tool for high-performance design.

Design Integration

While standard IC design typically follows a series of steps, MEMS design is different; the design, layout, materials, and packaging are intrinsically intertwined. As a result, MEMS designers must be especially adept at ensuring that a complete system functions well, from the microscale all the way up to the finished device.

MEMS design often involves a mix of mechanical and electrical engineers. This cross-disciplinary team approaches the task as they would an integrated system, which includes the coupling of mechanical, electrical, thermal, optical, fluidic, and biological subsystems. Therefore, MEMS design is sometimes seen as a system integration challenge rather than just a mechanical or electronic one.

Conclusion

MEMS technology has revolutionized many industries by providing miniaturized, cost-effective solutions for various applications. From sensors and actuators to MEMS-based cameras, these devices are enabling innovation across automotive, electronics, healthcare, and beyond. As MEMS technology continues to advance, we can expect further breakthroughs in design, efficiency, and functionality across industries.