Access Official Technical PDF

Share Summary:LinkedInEmail

MEMS technology extends silicon manufacturing beyond electronic circuits. By shaping silicon and building structures on its surface, engineers can create microscopic mechanical components alongside electronic functions. This shift helped establish the More-than-Moore approach: using semiconductor processes to add capabilities beyond simply making transistors smaller and packing more of them onto a chip. For industrial automation professionals, the history of MEMS offers a useful view of how materials, fabrication methods and coordinated engineering efforts can turn a promising device concept into a broader technology platform.

From isolated experiments to a field

Early work on silicon micromachining began well before “MEMS” became a familiar term. In the 1950s and 1960s, researchers investigated silicon etching and the ways it could be used to form small structures. Early devices included the Resonistor and the resonant gate transistor. The latter demonstrated a key idea behind surface micromachining: standard planar wafer processes could be used to create shallow, three-dimensional mechanical structures above the silicon substrate.

Further development of bulk micromachining helped bring practical devices such as pressure sensors and inkjet nozzles into view during the 1970s. Bulk micromachining shapes the silicon wafer itself, while surface micromachining builds structures in layers on top of the wafer. These approaches gave engineers different ways to translate mechanical functions into processes compatible with semiconductor manufacturing.

At first, this work was dispersed among individual research groups, without a unified MEMS community or sustained industrial focus. A turning point came with Kurt Petersen’s 1982 paper, “Silicon as a Mechanical Material.” It helped bring attention to silicon’s potential for mechanical applications and connected researchers working on related problems. The resulting growth in technical community and industrial interest helped MEMS become part of the wider semiconductor conversation.

Fabrication advances that expanded MEMS technology

Silicon fusion bonding

Silicon fusion bonding, or SFB, joins silicon wafers to create bonded structures. The process was developed in the context of silicon-on-insulator circuits and later used to build MEMS devices. Its importance lay in making wafer bonding a repeatable fabrication method rather than an accidental outcome.

The development illustrates how process understanding matters. Early experiments showed that wafers in contact could bond during high-temperature processing, but subsequent work investigated the conditions behind that result. Research into wafer cleanliness, surface contact and the role of surface chemistry helped make the process more controlled. In MEMS, wafer bonding provided a way to combine silicon layers and form structures that could not be made as effectively using a single wafer alone.

Deep reactive ion etching

Deep reactive ion etching (DRIE), also known as the Bosch process, enabled deep, near-vertical etching in silicon. Unlike anisotropic wet etching, which follows the crystal structure of silicon, DRIE can form vertical profiles independent of the wafer’s crystal orientation.

The process alternates between polymer deposition to protect the sidewalls and etching steps that remove material from the exposed areas. This sequence allows engineers to create deep structures with high aspect ratios. DRIE was developed for MEMS, but its uses also extended to deep trench capacitors in DRAM and through-silicon vias (TSVs) used in three-dimensional circuit integration. The process therefore illustrates how a fabrication method developed for one technology area can influence other parts of semiconductor manufacturing.

Polysilicon surface micromachining

Surface micromachining using polysilicon and silicon dioxide provided another route to three-dimensional structures. In a typical approach, polysilicon forms the active mechanical layers and silicon dioxide acts as a sacrificial material. Once the structure is built, removing the sacrificial layer releases the mechanical components.

This layer-by-layer method made complex moving structures feasible above the wafer surface. Examples described in the history of MEMS include digital micromirror devices, inertial sensors, micromechanical resonators, electrostatically actuated microgears and implantable neural probes. These applications demonstrate the range of functions that can emerge from shared materials and fabrication techniques.

Why the development model mattered

MEMS did not advance through a single invention. Its growth depended on early device concepts, improvements in fabrication and people who connected research to potential applications. Petersen’s role as a technical advocate and entrepreneur helped bring attention to the field, while contributions from researchers and engineers across universities and industry supplied the processes and devices that made it practical.

That pattern is relevant to industrial automation. A sensor or actuator is not defined only by its operating principle; manufacturing methods determine whether its small mechanical features can be made consistently and integrated with electronics. The MEMS story shows why evaluating a technology requires attention to both device design and the fabrication platform that supports it.

Silicon micromachining also broadened what semiconductor manufacturing could deliver. Instead of focusing exclusively on denser electronic circuits, engineers used established wafer processes to produce mechanical, optical and sensing structures. That wider capability is the essence of More-than-Moore: adding distinct functions to silicon technology rather than relying solely on continued transistor scaling.

Conclusion: a platform built from process innovation

MEMS technology matured through the combination of bulk and surface micromachining, wafer bonding, deep reactive ion etching and sustained collaboration across the research and industrial communities. These advances transformed silicon from a material primarily associated with electronics into a platform for making microscopic mechanical structures as well.

For automation engineers and technical decision-makers, the practical lesson is that manufacturing processes can be as consequential as device concepts. MEMS grew when fabrication methods became repeatable, complementary techniques enabled more complex structures, and a broader community could develop applications around them. That combination continues to define the value of MEMS within the More-than-Moore landscape.

Access Official Technical PDF

About The Author