Bubble trouble in electric vehicle fluids

By Dr. Nancy McGuire, Senior Feature Writer | TLT Feature Article July 2024

Air bubbles within and on the surface of electric vehicle fluids are unseen culprits behind a variety of thermal management and lubrication problems.

Foaming and aeration cause changes in electric vehicle fluids that can contribute to a variety of vehicle performance problems.
Complex interactions among base fluids and additives—including defoamers—can increase foaming and fluid aeration. 
Evolving vehicle designs and a lack of standardized test methods that replicate electric vehicle operating conditions complicate the task of optimizing fluid formulations.

Recent news reports have focused on the woes of electric vehicle (EV) owners stranded at charging stations during power outages and other such problems. Although EV fluid foaming and aeration haven’t posed any headline-making issues, engineers and chemists are hard at work behind the scenes making sure that, even as EVs make increasing demands on their fluids, these fluids do their part to ensure safe and effective vehicle operation. 

How big of an issue is foaming and aeration in EV fluids? “I think, in general, the problem is it’s an unknown in terms of severity and exact impacts to performance,” says STLE member Rebecca Warden, senior research engineer, driveline and hydraulics at Chevron Oronite. “The models some OEMs use to look at parameters such as heat transfer and gear protection, for example, normally assume the best case: little to no aeration or foam.” She adds that foaming and aeration definitely occur in the field, but they are more difficult to define as the direct source of a vehicle’s performance problem, and the extent of their impact isn’t always well known. “Unless you were in that gearbox, watching it, it’s hard to know just how much aeration is happening and the areas it’s impacting the most.” Warden adds that it might not be practical or possible to eliminate foaming and aeration completely, but acceptable levels for a given application need to be defined by the OEM. 

Even though the extent of these effects isn’t fully characterized, the issues that foaming and aeration can cause are well known. These include pump cavitation, oil degradation due to oxidation and thermal effects, and a decrease in the fluid’s ability to transfer heat, provide hydraulic pressure or maintain an effective lubricating film. 

A complex problem 
Like their internal combustion engine (ICE) vehicle counterparts, EVs rely on hardware designs and materials that work together with fluid formulations to minimize problems and maximize performance. The main difference between EV motors and ICEs is their operating speeds, says STLE member Jared Nelson, manager of Emery Oleochemicals’ product and applications development lab. Higher rotational speeds improve the efficiency of EV motors, he explains, enabling the motors to generate higher levels of torque. EVs currently on the market typically operate in the range of 10,000 to 20,000 rpms, Nelson says, and some manufacturers are aiming for 30,000 rpms or more in future models. (Typical gasoline-powered ICE engines operate in the 3,000 to 5,000 rpm range.) EV lubrication and cooling fluids are subjected to substantially higher shear forces, which can increase air entrainment that leads to foam formation. Under some conditions, entrained air can micronize, which stabilizes and thickens the foam, making the situation worse. 

Foam and entrained air can reduce the ability of a fluid to lubricate as well as to conduct heat away from bearings, gears and batteries. Entrained air and foam are poor heat conductors, and they resist flow, further hindering heat transfer. The resulting localized hot spots can cause air to dissolve in the fluid. Further, entrained air and foam occupy more space than fluid, and in some cases, they can cause vent expulsions in the reservoir or motor housing, resulting in fluid leakage. Reduced fluid volumes, and the resulting buildup of heat in the system, can cause mechanical and electrical malfunctions, and can further increase foam formation and air entrainment, which can accelerate fluid oxidation.1 

Several factors, including foaming, prevent the use of one fluid for both lubrication and thermal management, says Nelson. Safe, efficient battery performance depends on maintaining an ideal temperature, so thermal management is vital. Cabin heating and cooling systems rely on heat transfer from the motor compartment into the cabin for heating and away from the AC compressor for cooling. “If you have foaming buildup here in the motor at this location, and the fluid is not circulating as it should, then that one weak point in the system affects everything,” he adds. “In addition, the viscosity of fluid that is ideal for one part of the vehicle may not be ideal for another part. There is a multitude of technical challenges to overcome.” 

For a conventional ICE vehicle, you typically see 5% to about 10% fluid aeration, says STLE member Troy Muransky, lead organic materials engineer for American Axle & Manufacturing (AAM), a global Tier 1 automotive and mobility supplier. The fastest gears spin in the differential at around 5,000 rpm, “so it’s not normally a big issue, even in engines or transmissions today,” he adds. However, today’s EV motors have shafts and gears that can rotate at speeds up to 24,000 rpm, “so it’s a totally different realm,” he adds. Because none of the currently available industry testing methods measure conditions that directly correlate back to EV hardware, the impacts of foaming and aeration in EV-specific hardware are poorly understood, he says. This forces hardware engineers to evaluate fluid performance directly on their assembled designs, rather than testing at a component level or simulating performance in computer models. 

“Today with ICE axles, we are able to validate components at a component level,” Muransky explains, “so we have a high level of confidence when we start testing at an assembly level that we will pass that testing. Assembly level testing happens typically much closer to production launch, so if we end up with a problem, we have to go back and redesign the component, which is very expensive and can cause launch delays. With the EVs, I do not have that option because we don’t have a test method or bench test to be able to evaluate the component ahead of the assembly testing—this provides a major challenge for hardware providers.” 

Because the oil comes into contact with various internal components, Muransky says, “if we find we have a problem, we have to go back and find whether it’s [because of] a hardware design or an oil design, and it can be a very expensive change.” He adds that as a result, companies have to take a conservative approach to hardware design. “A lot of times that adds over-engineering or extra cost to the system just to make sure we’re going to be okay. We don’t want the consumers, our customers, to end up on the side of the road with a vehicle not working.” 

Muransky is working with various industry colleagues to come up with a system that evaluates aeration in EV fluids under actual driving conditions. This data could provide a much better understanding of aeration and fluid performance under realistic application conditions and be the key for hardware and fluid design and manufacturing. It also will be critical for computational modeling, as it’s difficult to construct accurate models of density-dependent fluid properties and their effects on heat transfer and bearing and gear life. “I don’t think it’s a straight linear relationship,” Muransky says, adding that “we don’t know if it’s ever reaching a steady state or not in the vehicle.” 

Gearing is one of the main locations where aeration can initiate, Muransky says. “It’s like a blender—you drive air into the fluid. It’s a big problem for the overall system,” he says, “because you’re adding oxygen into the fluid, driving faster oxidation of the fluid and any materials that it’s touching.” Oxidized fluid is more prone to generating sludge and forming insulating deposits, which stabilizes entrained air and foam and reduces the heat transfer within the system. Aeration can reduce a fluid’s performance, even in fluids that are not oxidized, by reducing its ability to provide a lubricating film. He says that, under highspeed conditions, lubricating films can be only about half as thick as models predict. Thinner films provide less separation between metal surfaces, increasing the likelihood of frictional heat building up. 

Aeration also can pose problems for fluid pumps, Muransky says. Dissolved and entrained air goes through a pump, and if that pump cavitates, it drives more air into the fluid. The air bubbles are condensed as they pass through the pump, but they expand again when they enter an open chamber with a vent. The fluid then releases the expanded bubbles, forming foam on the surface. 

Decreasing fluid viscosity also tends to decrease a fluid’s flash point—the temperature at which the fluid emits enough flammable vapor to flash (but not necessarily continue to burn) when it is exposed to a flame. This holds true for any type of vehicle, Nelson says, but is an especially important factor for EVs. 

Aeration also changes the dielectric properties of the fluid, Muransky says. Because EV fluids are exposed to high voltages and currents, they are expected to provide a sufficient level of insulation to prevent electrical shorts, but not cause a buildup of static charge. He notes that fluids that circulate through the gearbox are physically isolated from battery coolant fluids, but they still encounter various electrical components such as motor windings and high voltage bus bars. If an electrical current finds water or air in the fluid, he says, it could find a path to ground, which could short out the whole system. 

Foaming also can present problems in heat transfer fluids such as water-glycol- based solutions. Contaminants, supplemental coolant additives, insoluble surfactants and cracks or leaks that introduce air into the system can cause and stabilize foam. Foaming can occur due to turbulence as the system is being filled, and antifoaming agents can become depleted as a fluid ages during operation. Proactive system maintenance can help prevent problems caused by fluid conditions or mechanical problems.2 

Formulations: a balance of properties 
Until a few years ago, EVs used automatic transmission fluids for their electric driveline units because they were the best products available on the market, Muransky says. But as electric drive units are becoming smaller, lighter and more power-dense, they are driving the demand for new types of fluids, especially to address gear durability and scuffing issues observed with some transmission fluids. 

Unlike fluids for ICE vehicles, where conventions and standards have been evolving for more than 100 years, “If you look at EVs,” Muransky says, “everybody’s designs are very, very different. There are some commonalities between them. But there’s a lot of differences. So, depending on how you design the system, it’s almost going to have to be a bespoke fluid for every different one right now. There’s not going to be an off-theshelf product where one size fits all.” 

Photo courtesy of Cschirp, CC BY-SA 3.0, via Wikimedia Commons.

Foaming and aeration are different but related problems. Nelson notes that EV fluids are especially prone to incorporating bubbles within the fluid, not just on the surface. The goal is to find ways to dissipate foam and aeration as quickly as possible, or even better, prevent them from forming in the first place. He explains that foaming tendency relies not only on the inherent properties of the fluid, but also on the interactions among the various components in the formulation, and mixing various fluids complicates things still further.3 “If you have a base oil that’s a Group III petroleum lubricant and you start adding esters to it, that changes the polarity and the overall average molecular size. Good antifoam performance depends on choosing the appropriate ester (or other additives) to minimize foam formation or aid in its dissipation.” 

Conventional antifoam components that you find in automatic transmission fluids or gear oils might not be up to the job for the extreme conditions that you might find in some EV drive units, says STLE member Dr. Travis Holbrook, Chevron Oronite research scientist and formulator, who joined in on the interview with Warden. However, understanding and developing an effective solution in terms of componentry and fluid design is complicated by the fact that vehicle component designs are a moving target. “OEMs don’t have a harmonized design to set performance around,” he says. 

EV base oils tend to have low viscosity, which makes them less susceptible to foaming, but the higher gear speeds and lower oil volumes in EV drive units can counteract this advantage. Warden adds that the complex interactions between additives in a formulation make it difficult to quantify how effectively individual additive components are performing. “Just because an antifoam chemistry works well in one formulation doesn’t mean it will work well with a different additive system or base oil.”

The industry is working on understanding how fluid formulations perform not only for fresh oils, where most bench testing is done, says Holbrook, but also how these performances change as the oil ages. How do thermal, dielectric, oxidative and foaming properties change over time? What parameters are important for performance tests? And how do foaming and aeration affect aging and property changes? 

The trend toward lower-viscosity fluids, which reduce drag and increase a motor’s efficiency, also helps to counteract the foaming tendency from high motor speeds. Lower-viscosity fluids and lower lubricant film thicknesses help bubbles to break down faster, but this cannot come at the expense of the fluid’s ability to reduce friction and wear. Even though esters, fatty acids and their derivatives have been in use and have been studied for centuries, they are still the topic of a lot of cutting-edge research, Nelson says. Various combinations of fatty acids and alcohols provide a high degree of flexibility to produce esters with specific properties, he adds. 

Base oil selection can have a significant effect on performance, says Holbrook. However, new base oils are gradually making their way into the marketplace. Esters and re-refined base stocks are attracting attention because of sustainability considerations, but price and performance tradeoffs have slowed their entry into the marketplace. 

“Looking at many conventional applications, performance specifications haven’t necessarily changed significantly, but because of regulations, [some] additive components that were used historically can no longer be used,” Warden says. She cites health, safety and environmental concerns, as well as sustainability and renewable sourcing, as drivers of these changes. “Your list of available chemistries starts to shrink, but it forces companies like us to become more creative and develop replacements that in some cases are even better than what we originally had.” Cost and availability issues are driving formulation changes as well. Warden notes that a company might have multiple vendors for a given material, but if all of these vendors are relying on a single supply source for a raw material, the company still has a supply risk. 

Group III and IV polyalphaolefin (PAO) oils are popular base stocks for EV fluids, Muransky says. He adds that Group V oils offer a wide variety of compositions and properties, but their high cost is a barrier to widespread adoption. Blending a low-viscosity PAO base fluid with a higher-viscosity ester having good film formation properties, for example, can help a formulator achieve a better balance of properties, Nelson says. He adds that even though esters can be used as base oils, it’s more likely for them to be used as a component in a formulation that blends esters with PAO, Group III or gas-to-liquid (GTL) base fluids. These blends balance the properties of each component to provide the needed viscosity, lubricity, sustainability and seal swell characteristics. For example, many base fluids can shrink seals by extracting plasticizer, but some pure esters can cause seals to swell. Blending the two in the right proportions helps to maintain seal size. Biobased esters also add a sustainable element to an otherwise petroleum-based formulation. Equally important, esters provide a high level of solvency, providing homogeneous fluids and cleaner operations, he says. 

Nelson adds that corrosion inhibitors, antiwear agents and antifoaming agents all have effects on their own. “But now you’re in a multidimensional system where you’ve got all the interactions between the different components. If you add a defoamer that interacts with another additive and it causes a foam to build, then it’s obviously not suitable for that system.” Even if an antifoaming agent is effective at collapsing bubbles, if the foam forms faster than it can dissipate, it could cause problems under the high-rpm conditions of an EV motor. 

Conventional fluid antifoaming agents used under the high-speed conditions in today’s EV motors have been shown to actually increase foaming, Muransky says. He foresees new chemistries developing to address this issue, “but we don’t have a full understanding of what’s going on in the vehicle level today, to correlate back to a bench test, which would then help us understand and differentiate fluids. It’s going to take time to get there.” He estimates that it will take another three to five years to develop solid, reliable lab tests and correlating bench tests for measuring foaming and aeration. However, he is optimistic that these tests, currently under development, will be standardized and adopted, given the impetus from recent legislation mandating deadlines for phasing out sales of new ICE vehicles. 

Other additives add even more complexity to the mix. For example, driveline fluids often contain viscosity modifiers, long-chain molecule additives that undergo shearing during vehicle operation. Although they do not play as prominent a role in low-viscosity EV fluids as they do in ICE vehicle fluids, viscosity modifiers can cause these fluids to entrain more air, which can lead to higher foam volumes. 

Polar additives in an electromagnetic environment pose an additional concern. When these molecules reorient themselves in an electromagnetic field, it can alter their ability to lubricate gears and bearings. The practical implications of this effect are not completely understood, Muransky says, but “once we get aeration better understood and some of the EM effects understood, then we’ll really be able to start looking specifically if they have an issue together.” 

Muransky notes that in his role as a hardware provider, he is in frequent contact with representatives from oil and additive companies, reviewing their portfolios to help his company identify the best candidates. “It’s not really a huge change of pace for us,” he says, adding that his company already uses proprietary gear oils and other fluids for the OEM first fill. “It’s just more challenging now because the hardware testing that we need to do doesn’t match the component bench test, and that isn’t available today.” 

Sustainability is increasingly a factor in fluid formulations. This can mean longer intervals between fluid changes, biobased feedstock sources or less carbon-intensive production and delivery processes. However, Nelson foresees that petroleum will be in the picture for the foreseeable future, if for no other reason than that so many large-scale processes have been established and developed around petroleum formulations. Customer demand is a decisive factor, he adds, and this depends in a large part on price. 

Nelson notes that small increases in efficiency can have a bigger impact on EVs than for ICE vehicles, so improving the tribology of a lubricating fluid can make a significant contribution. An ICE vehicle typically runs at about 12% efficiency,4 so a 0.1% increase will not have a significant impact on performance, he says. “But if you have an EV that goes, for example, from 90% to 90.1% efficiency, you’re inching closer to that perfect efficiency, and it’s making some real impact,” he says. (EVs powered from renewable energy sources currently show energy efficiencies of about 40% to 70%.4

Better fluids by design 
Test methods under development are attempting to simulate a truer representation of real-world conditions in terms of motor speeds, shear rates, operating temperatures, tolerances between surfaces and quantifying the amount of entrained air and foam in a fluid (see Ref. 5 for a discussion of standard fluid test methods and their application to EV fluid testing). Although electric currents might not affect foaming and aeration directly, they do affect a fluid’s chemistry, which could affect the fluid’s ability to resist or expel entrained air and foam. “One of key focus areas right now in testing is running traditional tests, but with electrification,” Nelson says. 

Nelson’s group is working on understanding how changes they make in their company’s esters affect EV fluid performance properties. They are especially interested in quantifying how changes in the backbone of an ester molecule could affect viscosity, specific heat, thermal properties and lubrication and tradeoffs between these properties and other properties that must be managed. In a collaboration with the formulation and data science groups at The Lubrizol Corp., they are examining factors like molecular chain length and differences between mono- and diacids. “Lab testing is the first effort,” he says, but these tests are backed up with rig testing and computational methods. 

Photo courtesy of Bill Abbott, CC BY-SA 2.0, via Wikimedia Commons.

This collaborative effort also has delved into statistical modeling and design-of-experiment approaches. “It’s definitely a combination in a team effort to comprehensively approach this as a problem with a solution and see what parameters we can change, and when quantifying those multivariate changes, how they’re going to affect the physical properties,” Nelson says. “It’s getting to be more like the pharmaceutical industry problems where you cannot possibly test every combination of ester raw materials in a reasonable time duration. One thing affects another—you change one parameter, it throws off the whole system,” he says, explaining that “we’re trying to deconvolute the variables more strategically and find the corners of what parameters are valuable to test. Then we can build that model that helps us predict what’s going to be the better target molecule to achieve the specific performance requirements.” 

Artificial intelligence (AI) is on the horizon as well, but Nelson notes that AI models need a well-organized library of data to train the algorithms in order to produce meaningful results. He explains that various companies along the supply chain maintain their own closely held databases. “Building a model that’s really comprehensive would be impossible without some kind of openness in sharing that data,” he says. Also, it’s typical for companies to develop specific components and formulations to suit a specific purpose and test the parameters related to that application, rather than taking a comprehensive approach to define a parameter space. “Filling in the gaps to make a consistent model would be the most difficult challenge,” he says. 

Warden notes that current test methods used to evaluate EV fluids were developed for conventional applications (e.g., axles and transmissions), and that they do not necessarily reflect field conditions for EVs.5 She adds that developing EV-specific test methods is an active area of discussion. To date, most of the work has been done to investigate aeration in specific vehicle designs. She says that some designs are more prone to foam and aeration than others. No matter the hardware design, the fluid additive system can help control the foam and aeration, but the key is understanding the need for a given application and having a way to effectively evaluate performance. 

Holbrook says that more understanding will come during the next several years as various types of EVs are in the field and the fluids they use begin to age. “The electrical drive fluid (EDF) space is still pretty new,” he says. Many of these new performance requirements like extreme aeration resistance, dielectric properties and thermal properties are new and unique to these fluids, and we’re continually developing a database of components and fluid designs to be used toward predictive modeling of performance testing. When you are developing and implementing any predictive model, you want to be able to see that it correlates not just to bench tests, but more importantly, to field performance.” He adds that results from modeling and bench testing should always be validated by field test performance if they are to be useful. “Predictive modeling can be an extremely powerful tool to a formulator to alleviate product development cycle times and confounding componentry effects; however, it should always be recognized as a guiding tool rather than an absolute.” 

Although customers want a test that allows them to separate oils that perform well for an application from oils that perform poorly, the conditions and limits imposed by a particular test method might not reflect how a particular oil performs in real life. Any given test could miss important limitations of an oil formulation, or it could rule out formulations that actually perform well in the field. Being able to show a difference in a test method doesn’t mean anything if that difference doesn’t correlate to the field. Warden notes that one formulation might perform worse (or equally as well) as others, but it could offer advantages in other areas—price, for example. 

Looking ahead 
Given the early stages that the EV industry is in, many companies are joining precompetitive consortiums and working groups to work on issues that benefit the industry as a whole. Groups working on these include the Advanced Fluids for Electrified Vehicles (AFEV) Consortium, SAE, ASTM and various European groups, among others. These consortiums and working groups are developing and standardizing test methods and making recommendations to the industry. 

The pressing need for better understanding the interactions between EV hardware, fluid formulations and real-world performance has opened the door to interactive efforts among various suppliers and precompetitive collaborations between competing companies, Muransky says. As with anything else, cost constrains the types of solutions that are viable in the marketplace, he adds. “If cost wasn’t an issue, we could probably solve things fairly quickly. But we can’t make a $500,000 car for a daily driver.” 

Most EV lubricant manufacturers are still feeling their way, said STLE Fellow Jack Zakarian, owner and principal consultant of JAZTech Consulting, in a 2020 article.6 He noted a lack of industry standards, adding that OEMs typically do not publish specifications “beyond their circle.” He foresaw, at least for the near future, that fluid formulators would rely on proven additives from their existing portfolios that have demonstrated good thermal and electrical properties. He added that, in general, current materials are suitable for EV applications, providing that they are combined well. However, in the same article, Adam Banks, e-mobility marketing manager for Afton Chemical, stated that using existing formulations themselves is an interim solution at best. As EV designs are refined, new fluid types and different balances of fluid properties will be required, he said, driving the emergence of new categories of additives. He cited the example of electric transmission fluid for direct-cooled e-axles, a completely new product type. 

Warden cites the importance of focusing not only on individual product development, but also technology development on a broader scale. Because hardware is changing so rapidly, a formulation that works well with today’s EV models might not perform well with newer models a few years down the road. Product development must build in enough adaptability to allow the product to change with their customers’ needs, and that is empowered through technology development. 

Electric passenger vehicles have higher market penetrations than heavy-duty and off-road vehicles, although heavy-duty and off-road production EVs are currently on the market. Knowledge gained from passenger vehicles’ performance on the road can benefit development of their larger counterparts, but there also are important differences to consider. “Like what we’ve seen with conventional powertrains, there’s going to be a point where everyone determines what is the most efficient and cost-effective design,” says Warden. “At that point, fluid requirements may standardize as well, but we’re not there yet.” 

EVs and the fluids that keep them running remain a work in progress, but progress is happening. Although EVs pose a variety of new problems, other industries have already faced novel problems with foaming and aeration, and they have come up with innovative solutions. Today’s EV designers and fluid formulators are hard at work coming up with their own innovations to ensure that EV “bubble trouble” never presents a big enough problem to make it into the headlines.

1. Dow, C. (Sept. 5, 2022), “What is an electrically chargeable lubricant?” Automotive World. Available here.
2. Canter, N. (Feb. 2022), “Heat-transfer fluids: growing in visibility and importance,” Thermal Processing, p. 27. Available here.
3. Van Rensselar, J. (2024), “Synthetic esters: Beyond base oils,” TLT, 80 (2), pp. 32-39. Available here.
4. Albatayneh, A., et al. (2020), “Comparison of the overall energy efficiency for internal combustion engine vehicles and electric vehicles,” Environmental and Climate Technologies, 24 (1). Available here.
5. McGuire, N. (2023), “Test methods for evaluation of electric vehicle drivetrain fluids” TLT, 79 (10), pp. 44-52. Available here.
6. “Electric vehicle fluids: Off-the-shelf or brand new additive chemistries?” F+L Magazine, Nov. 10, 2020. Available here.

Nancy McGuire is a freelance writer based in Albuquerque, N.M. You can contact her at nmcguire@wordchemist.com.