![Die shot of an integrated circuit controlling a microbattery matrix designed to explore a 2-in-1 storage and voltage boost method. [Image courtesy of Patrick Mercier/UC San Diego]](https://www.rdworldonline.com/wp-content/uploads/2025/02/ucsd-300x188.webp)
Die shot of an integrated circuit controlling a microbattery matrix designed to explore a 2-in-1 storage and voltage boost method. [Image courtesy of Patrick Mercier/UC San Diego]
“In addition to the specific case experimentally targeted in the paper, this study more broadly explores the key trade-offs between electrical performance (such as achievable maximum voltage and driving frequency), autonomy, and the capability to drive various loads.”
Probing industrialization opportunities
This focus on fundamental design principles and industrialization is a strength of the research. “CEA-Leti collaborates closely with the startup InjectPower to industrialize these storage devices, though for different applications, specifically in the medical field,” Pillonnet said. “This paper paves the way for leveraging this unique technology in broader applications, such as serving as a compact building block for actuating micro-actuators in various domains, including drones, medical systems, and micro or collaborative micro-actions.”
Design advantages
Patrick Mercier, UCSD’s expert in energy-efficient microsystems, provided more context on themicroactuator technology and its implications for microrobotics and beyond:
Q: How feasible is it to scale this solid-state battery matrix approach for mass production?
A: Solid-state batteries are already available commercially, so it would be relatively straightforward to scale up to mass production once the designs are well optimized. Of course, getting the highest possible energy density out of solid-state batteries in well-yielding manufacturing is a continuous challenge to battery makers, especially with the more intricate packaging needed here, but it’s a challenge the semiconductor industry is well equipped to handle.
Q: What are the long-term reliability and cycling characteristics of these microbattery matrices?
A: This is a good question and requires further study. With most battery technologies, microcycling is generally more favorable than deep discharging and charging, so we anticipate good long-term reliability, but this definitely requires more research.
Q: Can you clarify the partnership between CEA-Leta and UCSD?
A: CEA-Leti are the battery designers. We at UCSD are focused on developing the chip that runs the microactuation system.
Q: What design constraints would you imagine that engineers should consider when integrating this 2-in-1 storage and voltage boost concept into existing systems?
A: The proposed solution helps address a key challenge in microrobotic systems: most systems want to use a Li-ion-type of battery, as they offer high energy density, and thus can maximize operation time of the microrobot, and yet, their relatively low voltage of around 4V is not compatible with microactuators that requires MUCH higher voltages (10s to 100s of volts). The key challenge is how to generate this much higher voltage. Most conventional approaches use a boost converter, which requires large inductors or many capacitors, which add size and weight to the system, ultimately reducing operation time as a result. The key outcome of the flying battery strategy is that it eliminates the need for inductors or flying capacitors. Thus, we can either save that weight to increase flight time, or allocate proportionally more weight to the batteries, again increasing flight time.
Q: Is the team planning to collaborate with industry partners on commercialization or licensing?
A: We are open to this, yes. There’s still more research to be done, but we do believe this is technology that can eventually translate into commercial use for these exciting new microactuation applications.
The paper describes how the system divides a solid-state battery—which is required in microdrone or microrobotic designs—into smaller units and dynamically reconfigures those units to create high-voltage outputs. Traditional bulky components like inductors and large capacitors are not needed. Instead, lightweight blocks of the battery matrix provide the higher voltages required for electromechanical actuators. Early testing demonstrated up to 56.1V generation at up to a few hertz of operating frequency — a range suitable for a range of microactuation applications.
“The batteries operate in a micro-cycle mode (1 ppm depth of charge), effectively minimizing cycling issues,” Pillonnet said. “Unlike traditional batteries that use liquid electrolytes, this technology employs a solid electrolyte based on a silicon wafer substrate, providing enhanced mechanical properties to this emerging technology. Like all batteries, energy is stored electrochemically, making performance highly dependent on temperature. Ongoing research aims to extend the operational temperature range for extreme use cases.”
Building on prior advances
While the current prototype uses commercially available components, the potential for further advances is significant. “Recently, TDK has commercialized solid-state batteries,” Pillonnet said. “The specific reference used in this paper is explicitly cited within the text.” The ability to build on existing commercial solid-state batteries underscores the system’s viability, while ongoing work with advanced prototypes from CEA-Leti suggests further scalability. The research team’s simulations indicate the approach can be reduced to as low as 14 mg in total weight without sacrificing efficiency.
“Microdrones and microrobotic systems already require a battery, and so it costs us next to nothing to use a solid-state battery, split it up into smaller pieces, and dynamically rearrange the small pieces to generate the voltages we need,” said Patrick Mercier, professor of electrical and computer engineering at UC San Diego who is also a co-author of the paper. “This is the smallest and lightest way we could think of to generate the high voltages needed to run these sorts of systems.”
The researchers are exploring commercialization pathways for this technology. Pillonnet reports that there are current discussions with partners but he and his colleagues “remain open to engaging with other interested investors.”
