Ultrasound Tech Offers Sophisticated Battery-Testing Intelligence

Ultrasound is showing promise as a cost-efficient method for testing and analyzing lithium-ion batteries.

Cylindrical-cell battery production line. (Panasonic)

The production and supply of lithium-ion batteries for the global EV and stationary energy storage systems (ESS) markets is growing at an exponential rate. At the same time, the need for safer and more-efficient solutions for the testing, management and re-use of batteries has also become an important component of reaching a “net-zero” carbon-dioxide future.

Capacity Micromachined Ultrasonic Transducers (CMUTs) used to generate ultrasound signals. (Ultrasound BMS)

EV vehicle design and manufacturing represents a major paradigm shift for the automotive industry such as new drive systems, technologies and test plans – This presents new challenges in parallel with the expanding electronic and software content of vehicles.

One of the major validation and safety challenges to be tackled for EVs concerns the effective testing of the battery pack itself, as well as battery management systems (BMS) – the complex electronic system that supervises the performance and safety of the battery pack and the high levels of electrical energy stored inside. A vital concern impacting EVs in particular is the possibility of a battery/system failure that potentially could lead to a fire or explosion.

Thermal-runaway events are the primary cause of catastrophic lithium-ion battery failure, with many hardware and software factors affecting lithium-ion battery safety. Some of the main contributors are lithium plating, stranded energy, sensor and control accuracy and resiliency issues and unique aging characteristics.

Ultrasound technology is demonstrating promise as a cost-efficient test method that provides greater accuracy, safety and performance for testing, management and re-use of lithium-ion batteries. Ultrasound – sound waves at frequencies above 20kHz that humans cannot detect – has been used for some time as a proven and safe, non-destructive technology best-known for bringing joy to parents after “seeing” their unborn babies while still in the womb. Beyond medical applications, ultrasound also is employed to measure distances and process materials and chemicals, as well as for various construction and military applications. and is now also being applied to battery production and testing.

Combining ultrasound and batteries

Manufacturers of battery cells and EVs, as well as consumer-electronics OEMs all are asking the same question: can the quality, performance and safety limitations of legacy testing systems be outperformed by looking “inside the cell?”

Ultrasound signal traveling through lithium-ion battery. (Ultrasound BMS)

An ideal battery lifecycle is a closed-loop system, where the utility of batteries is maximized before they are sent for recycling. Although trends indicate that battery architectures are becoming increasingly more considerate of end-of-life practices, the majority of batteries continue to be designed to best satisfy the needs of first-life applications, as those uses represent the longest period of use.

Battery-management systems are the primary control units that monitor and manage the batteries to ensure safety and performance in their first-life applications. Traditional BMS use a combination of voltage (V), current (I), and temperature (T) inputs to determine how to safely supervise lithium-ion batteries’ output, state-of-charge (SoC) and state-of-health (SoH). However, incumbent BMS technology does not actually measure SoC and SoH – it estimates those metrics. Furthermore, incumbent BMS cannot measure the SoC and SoH of a specific cell inside an EV or ESS as it is in operation, and this challenge cannot be solved with software alone.

To measure the SoC and SoH of a batteries, technology capable of sensing internal changes in the battery is required, rather than deriving estimates based on variables external to the battery. This is exactly how ultrasound is used to diagnose humans – and now the same methods can be applied to batteries. Using capacity micromachined ultrasonic transducers (CMUTs), an ultrasonic signal is either reflected or transmitted through the cell.

The signal then is analyzed and characterized using machine learning (ML) algorithms that aggregate thousands of hours of battery cycling at different temperatures and degradation rates to build a computational model of each battery.

Table 1: Ultrasound performance during various overcharge conditions inducing four types of failure. (Ultrasound BMS)

Once batteries are characterized, ultrasound is used to measure SoC and SoH in real-time, resulting in superior accuracy. Meanwhile, dynamic measurements generate the capacity and lifetime of batteries, eliminating the need for the artificial buffers used in traditional BMS to compensate for a lack of real-time data.

Unlike traditional BMS, ultrasound technology also can detect and prevent thermal runaway, alerting end-users and adjusting battery behavior prior to catastrophic events. Table 1, extracted from the Journal of Power Sources, summarizes ultrasound performance during various overcharge conditions, delineating two types of pre-failure notifications:

  • Warning time, when ultrasound starts to deviate from baseline
  • E-Stop, when significant changes in ultrasound show catastrophic failure is imminent

Repeatedly inducing four distinct failure types, Titan Advanced Energy Solutions has consistently identified a warning condition, allowing time to alter battery control, and an “E-stop” condition to disconnect the battery entirely to avoid catastrophic failure. For comparison, the Global Technical Regulation (GTR) No. 20 requires a 5-minute warning of battery failure to prevent human injuries (GTR is global regulatory organization under the United Nations Economic Commission for Europe, or UNECE).

Figure 4: Experimental data evaluating the effects of fast charging on lithium-ion batteries. (Ultrasound BMS)

Early experimentation using ultrasound to evaluate effects (such as lithium plating) related to fast-charging also showed promising results. Figure 4 demonstrates how ultrasound signal features are drastically different when fast charging occurs (1.67C vs C/4).

Streamlining quality control in cell manufacturing at gigafactories

CT and ultrasound scan of new batteries used to detect anomalies; a CT scan (left) and an ultrasound scan (right) of a battery cell, respectively. Ultrasound detected a tear inside one of the cell layers, which was confirmed using CT scan. Ultrasound also was able to clearly detect other physical anomalies (purple shadowing) that were undetected using CT technology. (Ultrasound BMS)

Batteries fail for a variety of reasons: mechanical abuse, electrical abuse, thermal abuse, poor management. But issues often are traced back to cell-manufacturing defects. Not all batteries are made equal – and the holy grail in battery manufacturing is the production of perfectly homogenous batteries. “Production costs account for a significant share, 22% of the cell cost, thus represent the most promising lever for progressive cost reduction and it is here above all that equipment suppliers can make the difference,” said a Roland Berger report in 2020.

The majority of battery quality and warranty issues can be traced to the electrode-manufacturing stage, and as such, streamlining quality control is key to reducing battery cost. Due to the lack of affordable sensing technology, there are significant gaps in cell insights from the electrolyte filling stage to the end of the manufacturing line.

Up to this point, costly testing technology has prevented the implementation of quality-control systems that can inspect every cell. Ultrasound presents an opportunity to provide highly valuable insights into anomalies in each cell at a fraction of the cost of existing computerized tomography (CT) systems.

Ultrasound technology can also be used on the production line to detect battery cell issues quickly and accurately, avoiding field issues and reducing manufacturing scrap costs. Gigafactories in Europe and the United States are investing this research to meet the rapidly-growing demand for batteries to support EVs.

Looking to the future

The application of ultrasonic waves to battery testing and management still is in its early days. However, its ability to view inside batteries and create models based on real-time molecular observations make the technology highly promising. Ultrasound’s numerous potential implementations throughout the battery lifecycle have made it attractive to industry players and investors who are funding early technological development.

To that point, in June 2023, Titan Advanced Energy Solutions was announced as a Phase III winner of the U.S. Dept. of Energy (DOE) Lithium-Ion Battery Recycling Prize. DOE set up the Battery Recycling prize as a multiphase competition to incentivize American entrepreneurs “to develop and demonstrate processes that, when scaled, have the potential to profitably capture 90% of all discarded or spent lithium-based batteries in the United States for eventual recovery and reintroduction of key materials into the U.S. supply chain.”

Although the DOE’s competition was seeking to maximize the recycling rates for lithium-ion batteries, Titan’s submission focused on extending the lifetime use of the batteries before recycling them. By enabling batteries to have a second life in energy stationary-storage applications, Titan reduced the need for new batteries in these markets where technical requirements could be satisfied with used EV batteries.

– Srdjan Mutabdzija is head of Product Development for Titan Advanced Energy Solutions Ultrasound BMS.