Deep Silicon Etching and TSV Integration Framework Streamlines Multi-Modal Sensor Fabrication for Intelligent Systems

Atomica, a U.S.-based microfabrication foundry, today announced the launch of its Physical AI Sensor Platform for customers developing sensors for robotics, autonomy, industrial automation, medical systems, defense applications, smart infrastructure, and edge intelligence.

Artificial intelligence is moving beyond software and centralized compute into robots, drones, vehicles, industrial equipment, medical devices, and other intelligent systems. These systems increasingly depend on accurate information about motion, force, pressure, vibration, gases, magnetic fields, light, depth, flow, and surrounding conditions, creating demand for more sensing capability across a wider range of products and environments.

"AI models may provide the reasoning layer, but sensors provide the real-world data layer," said Eldon Klaassen, CEO of Atomica. "As AI becomes embedded in more physical systems, the need for compact, reliable, integrated sensors will grow with it. The opportunity is not just to invent new sensing technologies, but to make them manufacturable at the scale these markets will require."

The challenge becomes harder as products move beyond laboratory demonstrations. Sensors must be fabricated repeatably, integrated, qualified, and produced cost-effectively. Atomica's platform is designed to address those requirements earlier in development.

The platform spans five major sensor families: motion and inertial; pressure, force and acoustic; environmental and chemical; magnetic and position; and optical and spectral sensing. Devices can include accelerometers, gyroscopes, pressure and flow sensors, microphones, gas and chemical sensors, magnetic and position sensors, microlenses, micromirrors, spectral sensors, and LiDAR-related structures.

What connects these devices is not a single process flow, but recurring microfabrication needs. Pressure sensors, microphones, flow sensors, and microfluidic valves may rely on diaphragm structures. Accelerometers, gyroscopes, resonators, and optical scanners may use released MEMS structures and controlled mechanical motion. Gas sensors, biosensors, and environmental sensors may require membranes, heaters, electrodes, and functional materials. Optical devices may incorporate microlenses, micromirrors, optical benches, and wafer-level packaging, while compact sensor modules may also require TSVs, TGVs, interposers, and other package structures.

A platform approach allows these elements to be reused, modified, and combined where appropriate. This reduces reinvention while preserving room for customer-specific innovation in sensing materials, device geometry, functional layers, optical features, electronics interfaces, and packaging.

"Physical AI will not be built around one sensor," Klaassen said. "Different applications will need different combinations of motion, pressure, chemical, magnetic, optical, and other sensing functions. Atomica's role is to provide the microfabrication foundation that allows customers to turn those ideas into manufacturable devices."

The platform draws on Atomica's capabilities including deep silicon etching, thin-film deposition, membrane and cavity formation, wafer bonding, metallization, optical coatings, microfluidics, through-wafer interconnects, interposers, and wafer-level packaging. Different devices use different combinations of these capabilities, allowing customers to build on established process knowledge rather than starting from zero.

This becomes increasingly important as products combine multiple forms of sensing. A robot may use inertial, force, position, and optical inputs. A drone may combine motion, pressure, magnetic, optical, and environmental sensing. Industrial systems may monitor vibration, gas, flow, pressure, and optical signals, while medical devices may integrate pressure sensing, microfluidics, sampling, and optical detection.

Integration into the larger product is also critical. Sensors may require vias, redistribution layers, optical windows, fluidic ports, cavities, seals, interposers, and wafer-level packaging to connect the sensing element with readout electronics, local processing, and system control. These interfaces can determine whether a device that works in the lab can perform reliably in a commercial product.

Customers can begin with a sensing concept, material, device architecture, or proof of principle and progress through feasibility structures, engineering samples, prototype hardware, qualification, pilot production, and scalable manufacturing. By reusing established process knowledge where appropriate, Atomica can help customers focus more of their development effort on what differentiates their product.

The Physical AI Sensor Platform extends Atomica's broader AI strategy. Existing platforms address optical connectivity, optical sources, photonic integration, and thermal management. The sensor platform adds the devices that allow systems operating in the physical world to collect information from their surroundings.

Source:

Tell Us What You Think

Do you have a review, update or anything you would like to add to this news story?

Leave your feedback
Your comment type
Submit

Sign in to keep reading

We're committed to providing free access to quality science. By registering and providing insight into your preferences you're joining a community of over 1m science interested individuals and help us to provide you with insightful content whilst keeping our service free.

or

While we only use edited and approved content for Azthena answers, it may on occasions provide incorrect responses. Please confirm any data provided with the related suppliers or authors. We do not provide medical advice, if you search for medical information you must always consult a medical professional before acting on any information provided.

Your questions, but not your email details will be shared with OpenAI and retained for 30 days in accordance with their privacy principles.

Please do not ask questions that use sensitive or confidential information.

Read the full Terms & Conditions.