Proceedings of a Workshop—in Brief

Convened March 12–13, 2025

U.S. Army Next Generation Armor
Proceedings of a Workshop—in Brief

On March 12–13, 2025, the National Academies of Sciences, Engineering, and Medicine’s Board on Army Research and Development (BOARD) convened a workshop to explore barriers and opportunities for developing the U.S. Army’s next generation of armor. This meeting was the primary element of a two-part workshop. The second, supplemental part was held at a later date (June 3–4, 2025) due to the restricted nature of the information presented.

Both meetings included experts from industry, academia, government, and national laboratories. The information summarized in this Proceedings of a Workshop—in Brief reflects the knowledge and opinions of individual workshop participants and should not be viewed as a consensus of the workshop participants, the BOARD, or the National Academies.

OPENING REMARKS

Workshop chair George “Rusty” Gray III, laboratory fellow and staff member (emeritus), Los Alamos National Laboratory, welcomed workshop participants and speakers. Gray led with his concern that the Army may not be taking advantage of the latest opportunities to incorporate lighter weight armor into its combat vehicles. Therefore, part of the planning committee’s challenge was to uncover what barriers may exist to the introduction of a new generation of materials that could provide somewhat comparable kinetic weapons protection with significant weight savings. He noted that a number of other factors can also influence an armored vehicle’s weight, but the focus of this workshop was on opportunities for lighter materials. He also noted that while workshop discussions were centered largely on steel, other materials such as aluminum, titanium, and ceramics were similarly crucial to armor design and could be suitable topics for additional workshops.

HISTORY OF THE MINE-RESISTANT, AMBUSH-PROTECTED VEHICLE

Paul Mann, program executive, Sea-Based Weapons, at the Missile Defense Agency and former program manager for the Mine Resistant Ambush Protected Vehicle (MRAP) program from 2006 to 2010, explained how leadership changes after 2006, including Secretary Gates’s appointment, led the Department of Defense (DoD) to embrace new technologies and partnerships. This shift allowed the U.S. Army and Marine Corps to develop an acquisition strategy for up to 25,000 MRAPs, which were critical to counter-insurgency operations in Iraq and Afghanistan. Ultimately, the program produced 27,000 vehicles in less than 5 years, with funding surging from nearly $1 billion to $18 billion in 3 years.

Mann attributed the focus and speed of the MRAPs development to the urgent need to counter improvised explosive devices (IEDs) and adapt to threats from “dynamic adversaries.” Key government officials played crucial roles in navigating defense industrial policy and ensuring access to critical materials, particularly steel. His team secured priority when acquiring the resources that were in high demand. This was in part due to the aid of international deals that were brokered with ambassadorial support as the U.S. supplies were falling short. Industry partnerships also enabled development by assuming initial material costs and managing the supply chain. The leadership at Aberdeen Proving Ground (APG) played an important role by turning it into a “24/7 armor testing camp for 4 years.” These testing and development efforts were a rapidly iterative experimental process that ultimately allowed 1,000 vehicles per month to be delivered to Iraq and Afghanistan.

Mann noted that the mission to protect the warfighter united everyone. The MRAPs rapid execution was successful because of trust and transparency between government and industry, as he warned against proprietary silos. He also stressed the need for supply chain sensitivity analysis, stockpiling widely used materials, and monitoring skilled workforce (e.g., welders) capacity. He advocated for physics-based models in design—while acknowledging their limits—and real-world battlefield testing for effective armor development.

During a question and answer (Q&A) session, Mann reflected that the relative lack of specific requirements became a strength during development, as it allowed him to focus solely on vehicle survivability and speed of production. He commented that overwhelming desire to prevent deaths on the battlefield led to a deeply supportive leadership, unusually tolerant of missteps. He noted that requirements were added as the technology was developed and tested, although he lamented at the lack of useful modeling and simulation data. Gray noted that testing is important because there are limits to what modeling and simulation can do. The physics and damage mechanism behind different weaponry (shape charges, explosively formed penetrators, rigid projectiles, long rods) are all different, and above a certain threat kinetic energy, “it comes down to strength and density.” He emphasized that vehicles with aluminum or magnesium armor will not stop a long-rod penetrating weapon.

STATE OF STEEL MANUFACTURING AND MATERIALS STOCKPILE

Raymond Monroe, vice president of the Steel Founders’ Society of America, a 120-year-old association of steel foundries, spoke on the state of the U.S. steel industry and its materials stockpile. Monroe opened by describing an acquisition panel he formed after Executive Order 14017 identified castings and forgings as a DoD supply chain concern. He found the acquisition process to be a key issue, as well as 40 years of policy that progressively degraded capital-intensive industries from being profitable enough for innovation.

Monroe explained that historically, U.S. capital investment followed a 30-year cycle, peaking in 1928, 1958, and 1980. However, by 2003, the pattern changed due in part to leaving the gold standard in 1972, which made the price of steel more volatile. Additionally, public policy changes first imposed in the late 1980s began to discourage domestic producers from reinvesting in equipment as the price of that equipment had surged. In contrast, before 1988, U.S. industry tended to direct much of its profits to capital investment and modernization, which created a reliable supply chain for both DoD and private industry. After 1988, buyers gained the upper hand as excess capacity drove down prices—leading some, such as General Motors, to refuse to pay above actual production costs.

Today, Monroe warned, Americans find themselves in the opposite situation, one where a limited supply will cause challenges as the lack of capacity will drive higher prices. He explained that this creates another threat because as capacity is constrained, the United States is forced to turn to international allies for production, whose plants are sold to or owned by Chinese companies. Another key issue is that the companies who make the equipment needed to modernize the U.S. steel industry infrastructure are not domestic.

Monroe noted that U.S. steel prices are always the highest, whereas European prices are lower, and Chinese prices are the lowest. In 2019, the United States was selling the same steel product as China at more than $200.00 per ton more. The traditional economic assump-

tion that U.S. steel makers are inefficient, thus making the product more expensive due to its low quantity, is incorrect. He said that China and the United States have the same associated costs with steel production, but China is selling it at lower prices. The difference, he argued, is that the United States has been burdened by public policy requirements that add extra costs to steel manufacturers, leaving them at an approximately 20 percent cost disadvantage to the Europeans and a 40 percent cost disadvantage to the rest of the world. This is, in part, because other countries subsidize the cost of their steel and pay for electricity costs. The U.S. steel industry lacks enough incentive to modernize much, to the U.S. government’s disappointment. This lack of motivation stems from the inability to produce enough profit to sustain upgrade costs while maintaining operations.

One of these cost barriers is how U.S. steel manufacturers are unable to fully deduct leasing costs in their taxes, which forces them to make cheaper, incremental upgrades rather than modernizing, even during profitable years. For example, despite a continuing domestic workforce shortage in the steel sector, U.S. companies cannot adequately invest in robotics due to high costs.

Monroe then opened the floor to Q&A. Gray asked what feed materials are needed for steel production. Monroe answered that chrome, molybdenum, and cobalt are the most problematic to acquire. William “Bruno” Millonig, board director at the National Academies, asked what Monroe’s recommendation to the Industrial Base Analysis and Sustainment (IBAS) program would be. Monroe suggested that using long-term contracts to maintain a strategic reserve of critical armor materials could stabilize the situation and also noted that early DoD purchases could prevent bidding wars during demand surges. He also noted that annual Office of Enterprise Management contracts often fail to adjust for inflation, making DoD orders the least profitable. One participant, Al Sciarretta, a retired senior test engineer at CNS Technologies and a member of the National Academies’ BOARD, commented about rising electricity costs for automating steel foundries, which Monroe acknowledged.

STATE OF THE STEEL INDUSTRY WORKFORCE AND TRAINING

In her presentation, Monica Pfarr, executive director of the American Welding Society, discussed recruitment strategies to address manufacturing trends and the shortage of skilled welders. She cited a 2024 Deloitte survey, which found that the need for high-skill manufacturing roles requiring technical, digital, and soft skills will grow the fastest through 2032.¹ Pfarr also noted that 55 percent of industrial manufacturers already use generative artificial intelligence (AI), with another 40 percent planning to increase investment in it over the next 3 years. By 2028, AI-driven product innovation will push half of all large manufacturers into searching for new opportunities within their legacy products.

Pfarr noted that the current U.S. welding workforce has 771,000 workers. However, 150,000 welding professionals are approaching retirement and will need to be replaced within the next couple of years. In addition, welding jobs are concentrated in Texas, California, and the Great Lakes region (which accounts for 25 percent of demand). Due to industry growth and worker attrition, 320,000 welders will be needed by 2029. Annually, this would amount to 78,000 new hires across six occupations (boilermakers, sheet metal workers, structural iron and steel workers, structural metal fabricators and fitters, welders and cutters, and solderers and brazers), with welders, cutters, solderers, and brazers making up the majority of the deficit. Also, despite a dearth of workers, compensation has not drastically increased over the past decade, only exacerbating the shortage even more.

Pfarr noted that across the United States, there are 2,400 institutions that offer programs in welding-related training—56 percent of those are high school programs and 43 percent are at the post-secondary level, including the Army Ordinance School, which has a welding program that requires national standardized testing. Program lengths range from 9 to 12 months for a welding certificate to 2-year community college degrees. Apprenticeships are common in most of the sector as well. Despite the fact that in 10 years, an individual’s total return on investment to become a skilled welder is almost double that of a 4-year college student, persistent shortages remain. The positive differential for the skilled welder is due to their earlier entry into the job market and relative lack of school debt.

__________________

The American Welding Society Foundation, which is the philanthropic arm of the nonprofit American Welding Society, awards more than $3 million annually to support welding education, with $1 million in grants to schools for welding program improvements and expansion, and the rest for scholarships. For example, in 2020, it awarded 16,188 scholarships, benefiting more than 400 academic institutions. Fifteen percent of scholarships went to associate degree students and 19 percent went to bachelor’s degree students, mainly for welding training but also in metallurgical and materials sciences. The foundation also grants $50,000 annually to 121 institutions nationwide. To anticipate the growth of automation in the industry, it also focuses on apprenticeship programs focused on robot automation.

During a Q&A, Pfarr responded to Millonig’s question about scholastic welding programs and capabilities that the U.S. Army should be aware of, noting the relevant research undertaken at the EWI² in Columbus, Ohio. She noted, however, that the U.S. education system has not caught up to DoD needs. Timothy Snyder, technical fellow at Oshkosh Defense, asked if Pfarr was familiar with Army welding techniques, which are very different (e.g., cold welding) than traditional techniques. Pfarr noted that she worked with the technical committees that focused on these capacities.

PANEL DISCUSSION: STATE OF THE STEEL INDUSTRY

Panelist Aaron Stebner, a professor at Georgia Institute of Technology (Georgia Tech), led with a discussion on armor development, mentioning his experience with nickel-titanium alloys and molybdenum, as some U.S. companies have exclusive agreements for C-103 niobium alloy, limiting access to DoD. He suggested alternatives to C-103 niobium alloy, like titanium alloys, given their ability to be used at a larger scale for a lower cost. For example, titanium alloys can be made at 40 tons per week. Stebner also highlighted his work with television personality Adam Savage showcasing a 3D-printed titanium “Iron Man” armor that withstood bullets from various calibers.

Stebner also discussed barriers to the development of additively manufactured alloys, noting that the performance of powder-based alloys was unpredictable and that manufacturers of the powders could not recreate the same material properties as alloys created by more traditional means. In part, Georgia Tech opened the Advanced Manufacturing Pilot Facility (AMPF), in which Stebner was recruited to lead its expansion. The facility’s $50 million expansion will make it the country’s first high-risk proving ground for developing AI manufacturing technologies. The 65,000-square-foot facility will open next year, with more than 25 faculty and 60 companies, and it has more than 350 agreements in review, with plans to partner with government laboratories and researchers.

The AMPF mission will include metal and polymer synthesis, manufacturing, recycling, and composite production, along with characterization, heat treatment, and mechanical testing. A wet laboratory will be available for elemental analysis, rheology, and metallography. Additionally, it will have extensive thermomechanical processing capabilities. The facility will also feature X-ray inspection, acoustic non-destructive evaluation, and magnetic metrology. Automated material handling, autonomous mobile robots, and mobile mobility systems with collaborative robotics for payload transport will also be integrated.

During Q&A, Mark Rupersburg, vice president of engineering at Rafael Systems Global Sustainment, asked Stebner to expand on the facility’s mission. Stebner answered that the work within the facility informs industry of early high-risk iterations that allow them path planning to speed technology maturation. Additionally, the role of the university pilot facility is to test, find problems, and reduce risk. Snyder asked if Stebner was looking at product quality or process efficiency and what would be the parameters with the processes. Stebner answered that the goal is to build quality models and collect technical data that can be utilized and applied to different scientific needs.

Panelist Greg Olson, co-founder and chief scientist at QuesTek Innovations, spoke next and noted that the integrated computational materials engineering technology that his team built, which was grounded in the Computer Coupling of Phase Diagrams and Thermochemistry (CALPHAD) system of fundamental databases

__________________

(the materials genome), has allowed him to develop high-performance aerospace materials. His first example was next-generation aircraft landing gear steels, which were high-quality, double vacuum-melted steels. He highlighted that this was the first application of the methodology developed during the Defense Advanced Research Projects Agency Accelerated Insertion of Materials program for accelerated qualification. A key part of the process was to design for large-scale manufacturing, which involved empirical development. He worked with a likely supplier to use process simulators, providing the graphene nanoribbon solidification for 15-ton, vacuum-refined virtual reality electrodes, and then constrained the new alloy design. The alloys were homogenized under practical conditions, and his team addressed the alloy’s solidification micro segregation and back diffusion. As the team scaled up production, it validated its predictions with progressively larger crops. He noted that designing for scale was critical and emphasized that designing materials, rather than discovering them, was necessary to reduce downstream problems. He also noted that designing materials also reduces the need for pilot facilities, as virtual scaling minimizes the need for physical piloting.

Olson created different high-performance carburizing steels that are being utilized in helicopter components and in Formula One racing, which he believes to be relevant to armor systems as well. He mentioned a collaboration between his Office of Naval Research–sponsored research at the Massachusetts Institute of Technology and activities at QuesTek Innovations, focusing on next-generation, CALPHAD-based, screen-induced transformation models to enhance alloy performance. By removing the quench-hardening constraint and utilizing deformation-induced martensite alloy, a broader composition space can be explored for improved martensitic steels. Additionally, QuesTek Innovations is currently working on developing affordable, lightweight, multistep structures for personal armor to address emerging threats. Olson also noted the potential to design alloy in response to higher friction surface treatment and that he is looking at carbon nitriding steels that go up to Rockwell C70 for personnel armor sheet-based systems.

During the Q&A, LTG Sean MacFarland, senior mentor at the U.S. Army War College, and Gray debated cast versus welded metals—with MacFarland highlighting weld vulnerabilities and Gray noting that castings lack the high-hard properties of plate. Stebner highlighted the work of George Spanos, director of New Initiatives, Science, and Engineering at The Minerals, Metals & Materials Society, and Allen Liu, professor at the University of Michigan, on metamorphic manufacturing, which merges casting and robotic forging to enhance hybrid processing capabilities. Suveen Mathaudhu, professor of metallurgical and materials engineering at the Colorado School of Mines, acknowledged gradient structured armor as innovative but emphasized the need for repairability and weldability in combat, adding that such metals could be used in bolted armor plates instead. Jeff Lloyd, scientist at the Army Research Laboratory (ARL), noted that optimal materials design needs to be done in conjunction with armor design.

CASE STUDY: DESIGN AND PRODUCTION FOR SPACEX STARSHIP

Jim Wright, principal engineer and senior manager of materials engineering at SpaceX, discussed designing complex systems and concurrent engineering in alloy development for space applications. He emphasized that the process begins by identifying design variables to meet objectives like blast and penetration resistance with low mass, while considering manufacturing requirements (e.g., cost) to ensure a fast cycle time to facilitate production at scale. These objectives are then broken into subsystems, where design trade-offs are evaluated to achieve global optimization and avoid bottlenecks. He also noted that suboptimal designs often mirror the organizational structures that create them, in part due to excessive requirements.

Wright explained how the process starts at SpaceX with concept development, followed by setting system requirements for subsystems and components. Engineers then focus on implementing, integrating hardware and software, and validating components in order to refine the design. This phase involves successive iterations to speed up the build–test–design process and move it into operational validation, as engineers often discover system-wide issues only after full testing.

Expanding on his comment on how organizational structure influences design, Wright emphasized the importance of vertical integration when manufacturing. Collaboration

is key in vertically integrated systems, as engineers and coworkers all share the same end goal and must work together to achieve it. In this model, design engineers also oversee the manufacturing of their parts, which encourages them to design with manufacturing challenges in mind. Wright noted the importance of fostering a “chief engineer” mindset, where everyone focuses on both system and global optimization. While optimizing specific subsystems, the engineering team must work with others within the organization to achieve global goals.

Wright recognized that armored vehicle designs do not lend themselves as easily to vertical integration. He noted the importance of supplier relationships, emphasizing that building collaborative relationships with key suppliers in steel is essential to facilitate timely innovation. In addition, echoing Mann’s earlier points, transparency between companies is important, because non-disclosure agreements and trade secrets can create stovepipes that limit collaboration toward mutually beneficial objectives. Wright also highlighted the difficulty of incorporating new alloy designs into concurrent engineering due to the slow pace of advancements in material engineering—something Elon Musk, chief executive officer and chief technology officer at SpaceX, calls an “impedance mismatch.” Wright advised that the best approach to prevent “impedance mismatch” is to design, build, and test to the requirements and then assess new materials. If a design cycle is only 1 or 2 years, it is crucial to quickly establish requirements, develop new materials, and integrate them into the design of the system.

Wright described SpaceX as an organization that designs, manufactures, and launches advanced rockets in space with the end goal of enabling human life on other plants. SpaceX is now one of the world’s fastest growing providers of launch and Internet services. Wright explained development at SpaceX relies on what Musk calls “The Algorithm,” which is a five-step process used by the company’s engineers:

1. Question every requirement.
2. Delete any part or process you can.
3. Simplify and optimize.
4. Accelerate.
5. Automate.

The algorithm begins by questioning every requirement, assigning each to a different engineer. If a requirement is removed but later needed, it can be reinstated. Wright suggested, for example, considering if armor could be integrated into the vehicle structure, potentially removing any distinction between them. While unsure of its feasibility, he believed some components could be removed. Once requirements are questioned and parts simplified, alloy development follows. With clearer requirements, alloys can be optimized for both manufacturing and performance. After these steps, production can be accelerated, but not before the process is well understood. Finally, automation through robotics comes after these steps to ensure clarity on what can be automated.

Wright discussed the evolution of SpaceX’s Starship, formerly the Big Falcon Rocket, highlighting how questioning requirements can lead to success. In 2018, SpaceX initially tried using carbon fiber composites but faced manufacturing challenges. Musk then proposed using stainless steel, inspired by water tower construction. This led to the development of the “Hopper,” a prototype made from 304 steel. While not orbit-capable, it proved the material’s viability. The switch to steel simplified manufacturing, which contributed to a stronger, more reusable rocket. Additionally, steel’s strength and toughness both increase from room temperature to near 0 K, offering yet another advantage over carbon fiber composites. As a second example, the Starship rocket is innovative due to its massive structure and advanced Raptor engines. Engineers streamlined the Raptor engine design by eliminating unnecessary parts, improving packaging, and simplifying components, which made the engine and rocket more efficient.

Wright discussed the development of a new cryogenic-tough material for aerospace combustion chambers and nozzles. Initially, the company explored high-strength, cryogenic-capable materials like Inconel alloy 718 and to improve copper’s thermal stability but found a better solution in a material similar to 15-5 martensitic stainless steel with enhanced cryogenic toughness. The new material is highly castable, easy to weld, and being optimized as a precipitation-toughened alloy. He also discussed thermal simulations on steel solidification, highlighting how nickel promotes austenite, chromium

favors ferrite, and that managing phase formation minimizes segregation and improves material properties. He also noted that a small amount of ferrite at the end of the process helps absorb impurities like sulfur.

Wright then discussed optimizing steel for structural applications. His team initially used metal inert gas–welded 304 steel for rapid manufacturing and testing but switched to full-hard 301 for better strength and elongation. However, weld failures led them back to 304, which, while weaker, was easier to manage as it failed by tensile overload rather than brittle fracture. Further iterations led to testing with 30X steel, showcasing SpaceX’s continuous material optimization for structural performance.

Wright described how in 2020, SpaceX began a rapid increase in facility capacity to advance its infrastructure. He reminded the workshop participants that the last phase of the SpaceX algorithm focuses on automation and stated that while there are many robots doing laser welding within the facility, all welds started off manually. Through successive designs, they were able to better determine what could and could not be automated with robots.

Wright opened a Q&A discussion. MacFarland asked how SpaceX learns so quickly from its failures. Wright explained that they conduct failure analyses to find the root cause and develop a remediation plan, with rapid design, build, and test cycles. Monroe stressed that failure is essential for optimizing products, as iteration remains the key. Wright agreed, adding that a culture embracing failure starts at the top and can thrive when talented, passionate people drive it.

Millonig asked Wright to describe Tesla’s approach to non-steel materials incorporation. Wright answered that Tesla uses a lot of nickel-based superalloys for high-temperature properties and incorporates copper where they need extremely high thermal conductivity properties because those cannot be replaced in steel. Nickel-based parts also help to reduce costs.

Valerie Wiesner, senior research materials engineer in the Advanced Materials Processing Branch at the NASA Langley Research Center, asked how computational tools inform Wright’s decisions. Wright noted that he uses various tools, including Abacus and ANSYS software, with Abacus used primarily for studying static stress in thermo-protection. He also develops models for casting simulations and physical processes and uses thermal CALPHAD for alloy development. He noted that while modeling has its limits, certain focal areas have more efficacy (e.g., strain), and coupling it with rapid testing could be a powerful approach.

PANEL DISCUSSION: TESTING AND PRODUCTION PIPELINE FOR NEW MATERIALS AND STEEL ALLOYS

Panelist Bryan Webler, co-director of the Center for Iron and Steelmaking Research at Carnegie Mellon University (CMU), opened the discussion with a breakdown of trends in U.S. steel production over the past 20 years, noting that the United States produces roughly 80 million tons of steel per year. The share of basic oxygen furnace steel has decreased to 30 percent, while the share of electric arc furnace (EAF) steel has increased—with the United States producing 70 percent of this steel globally. Webler noted that the United States relies heavily on scrap, most of which is produced by major companies using EAF, nearly all of it is continuously cast. He emphasized that steel does not need to start as an ingot, as there are many production options with EAF and in weights ranging from 50–300 tons, which offers flexibility. A growing trend is to reduce direct iron feeding into these furnaces, and companies like Nucor have built facilities to produce an alternative way of turning iron ore into metallic iron. This is a low-impurity iron source, and while production is still low, the demand is increasing. Webler highlighted a 35-year shift in U.S. steel production, with growth in the Southeast due to Nucor and Steel Dynamics, while output has declined in the Rust Belt and Northeast. The mills are becoming more geographically distributed as they move closer to the demand centers. Another trend is industry consolidation—in 2003 the top six steel producers held 39 percent of U.S. casting capacity and in 2025 they control more than 60 percent, as smaller firms have been acquired.

Webler noted that while Cleveland Cliffs, Coatesville, and Conshohocken have rolling plate capacity, there are new facilities like Brandenburg, which have melting, casting, and rolling operations. Webler remarked that rolling plate capacity in the United States can go up to 18-inch ingots, while castings can go up to 12 inches thick. However, developing new alloys requires pilot-scale melting and rolling, which few facilities offer. He offered the

examples of CMU with a small-scale operation, Missouri Rolla possesses a pilot-scale rolling facility, and the Navy is working on a facility in Memphis. He noted that pilot testing is important because materials behavior can differ between the development and production scales. In addition, as alloy concentrations in steel change, simulations can become less accurate and steel structure and properties can vary in unexpected ways. Finally, introducing new materials into large-scale production is often difficult, as steel plants’ margins are small and they tend to prioritize speed and efficiency.

During a Q&A session, Gray asked what made simulations so challenging. Webler explained that some simulations are less accurate because there are less data on new materials, which limits their efficacy of the simulations. An ARL participant noted that IBAS is funding that work at production scales and is in the second year of those efforts.

Panelist David Alman, technical director, Advanced Alloys Signature Center at the National Energy Technology Laboratory (NETL), began his presentation with a brief overview of NETL, explaining that it is one of 17 national laboratories within the Department of Energy (DOE) complex. Alman highlighted NETL’s long history in alloy development, particularly in commercializing alloys. He mentioned how the Albany, Oregon, site, which was originally a Bureau of Mines facility from 1943, played a key role in zirconium and titanium commercialization. That plant, known as the Wah Chang plant (now owned by ATI) in Albany is currently the major producer of titanium, zirconium, niobium, and refractory metals, and the largest employer in the city. Now working for the DOE Office of Fossil Energy and Carbon Management (FECM), the plant provides critical materials for fossil energy applications.

Alman described the plant’s unique onsite capabilities for making alloys, which include air induction melting up to 300 pounds; vacuum induction melting up to 500 pounds; vacuum, rock remelting, and electro slag remelting equipment with 3-inch diameter capabilities; and two 50-pound EAFs. The plant also has thermomechanical processing capabilities designed for super alloys and can heat up to 1650°C. It has controlled cooling to quench from a vacuum and quench the material for heat treatment. The facility also has a 500-ton press forge, along with two high-roll mills and wire drawing equipment. In addition, it is currently installing a 900-ton extrusion press and is acquiring a 1,500-ton forge.

Alman explained that NTEL’s fossil or energy-related research focuses on integrated materials engineering approaches that leverage computational materials coupled with large-scale manufacturing that can easily be translated to industry. This is because the plant tries to replicate industry feedstocks when developing its capabilities. NTEL has also developed three alloys, including a new Iron 9 Chrome steel (Fe-9Cr) that is a workhorse power plant steel with a 50-degree increase in temperature working range. It is also increasing the temperature capabilities of gamma prime strengthened nickel through either increasing gamma prime concentration or increasing grain boundary strengthening of the material. In addition, it has a project to enable thick wall castings of nickel-base super alloys and is developing pipeline steels to increase fracture toughness.

Alman also shared that through the DOE FECM and NETL-led eXtremeMAT National Laboratory Collaboration, NETL is partnering with Los Alamos National Laboratory and other DOE national laboratories to develop high creep models that accurately predict creep behavior in steels, particularly those used in fossil energy applications. NETL can now simulate 10 years of creep behavior in roughly 5 hours. Models have also been developed to look at stress on materials and how temperature cycles can impact behavior and the life of those materials. The next panelist, Mathaudhu, noted that scientists are not maximizing their use of computational materials design for materials manufacturing applications or the analysis of ballistic and high strain events and are not leveraging AI and machine learning techniques as quickly as needed. They are also not leveraging novel material configurations to achieve the hardness and toughness needed for armor applications.

Mathaudhu suggested that in 2040, the next generation of tanks or heavy armored vehicles will be challenged by weight considerations and the inability to specify requirements for their armor. He challenged assumptions about new, exotic materials, “unobtaniums” being ready for the U.S. soldier by 2040–2060, suggesting a focus on

optimizing existing steel through new processing or configurations. Additionally, he questioned if current testing is being used enough and if it aligns with military specifications (MilSpec), arguing that such specifications are cumbersome and delay military deployment. Lastly, he explored which simulations could yield reliable results.

Mathaudhu emphasized that the biggest contributing factor to protection in advanced steels is hardness. At low hardnesses, there is plastic flow and the target absorbs some of the impact’s energy, and at higher hardnesses, there is adiabatic shear of the penetrator material. At harder hardnesses, the material breaks up and for ultra-hard material, it shatters. Across this range of hardness are multiple mechanisms for threat defeat, which makes consideration of this range the most important focus for next-generation armor development.

Mathaudhu described armor steels as materials with Brinell hardness values (HBW) in the 340–390 range as defined by U.S. MilSpec MIL-DTL-12560K (Class 1). MilSpec MIL-DTL-46100E applies for hardness of 477–534 HBW and MIL-DTL-32332 (Class 1 and 2) applies for hardness values greater than 570 HBW. These MilSpecs point to hardness as the main threat defeat mechanism. Different classes of steel protect warfighters in combat vehicles. The first type are blast protection steels, which are energy-absorbent, bendable, shapeable, and weldable, with relatively low hardness (370–460 HBW). These are at the bottom of vehicles and are often rolled into single 6 meter-long plates. They offer a good combination of hardness and toughness against IEDs and similar threats. Next are ballistic protection steels, these “construction steels” are high-hard and are bendable, weldable, cuttable, and fatigue-resistant.³ They are divided into two categories—high hard (477–534 HBW), where a thickness of 6.5 mm is needed to stop an SS109 munition (also known as the M855 5.56 × 45 mm NATO ammunition) and very high-hard (530–570 HBW), which also stops SS109 munition, but at a reduced thickness of 5.5 mm. Outside of these are steels that are not easily weldable or cuttable but offer great protection with ultra-high hardness of 580–640 HBW. Ultra-high hardness steels are able to stop the same SS109 munition at only 5 mm thickness. These steels can be prone to cracking without too much bending; however, this is being addressed with variations in compositions within the MilSpec (MTL-DTL-46100) across vendors. He mentioned there are also some exotic, extreme-high hard steels (650–700 HBW) that resemble ceramics, and therefore they are brittle, hard to weld, and unsuitable for vehicle repair.

Mathaudhu discussed configurations of these different steels, noting that they could be spaced. Spaced armor involves two plates separated by a 10 mm (half inch) air gap, which are typically bolted together rather than welded or fused. To stop a SS109 bullet threat, one could take about 6.5 mm of high-hard steel and then add another 4 mm of plate and put an air gap in between. This combination now enhances the protection against a 7.62 × 51 armor piercing (AP) rifle round. The total steel thickness needed for protection goes from 15 mm of stand-alone, high hard steel, to roughly 11 mm combined, which would give a cost reduction and a weight savings of 25 percent just by creating a space between the two plates. Mathaudhu concluded these thoughts by stating that the Army could optimize its supply chain by using steels that, while not the hardest or strongest, allow physics (and the air gap) to reduce both weight and cost by using less materials.

Mathaudhu noted another approach could be to use perforated steels where perforations smaller than the size of the munition could be implemented to reduce weight. While somewhat expensive and time-consuming—because perforations would be laser cut or punched out either during pre- or post-hardening—it would lighten the armor. Mathaudhu emphasized that to protect against Dragunov 7.62 mm × 54 mm AP rounds, instead of using a 16 mm thick solid-steel plate, one could use 6.5 mm of standard high hard (500 HBW-type) and add on 4 mm of perforated 600 HBW-type ultra-high hard steel, which would give a weight savings of 40 percent and a thickness reduction of a few millimeters. These solutions could use conventional stockpiles of plates to make very high-performing armor.

The last solution, alluded to by Olson, would utilize dual hardness steel where a high-carbon hard steel plate is roll bonded to a lower-carbon steel plate. This would provide a combination of strength and toughness, with a harder outer layer and a softer inner layer that has a gra-

__________________

dient of properties versus monolithic plates. Mathaudhu said that these novel configurations have been tested and work, so it becomes a question of whether they can be manufactured effectively and learning the trade-offs between cost and weight savings to determine the best approach.

Mathaudhu pivoted to presenting findings from a 7-year research effort on Ferrium® M54®, a computationally designed alloy developed for hookshanks and, potentially, arrestor cables. Mathaudhu noted that Ferrium M54 demonstrates significantly superior mechanical properties over conventional alloys such as Aermet and Maraging steels. It was also the first DoD-fielded computationally designed material to reach full production and required extensive time and effort.

While originally designed for hookshanks, a production of ausformed M54 from the 2022 study was successfully scaled for use in protective applications. This process involved multiple steps, including rolling and quenching. A critical, yet unpublished step—skin pass rolling—enables modifications to plate thickness by a few millimeters while significantly increasing surface hardness in a scalable manner. The core hardness of M54 is already high at 58 HRC (570 HBW), but this cold processing technique enhances surface hardness to approximately 64 HRC (800 HBW) without additional tempering or hardening. It was demonstrated that even small reductions (1–3 percent per pass) have substantial benefits in weight, cost, and protection. For example, reducing thickness from 5.5 mm to 5 mm provided considerable advantages. Stress-strain data confirmed that this process yields extremely high strength and toughness, which surpasses traditional steel alloys.

Production was successfully scaled at the Pacific Northwest National Laboratory using a decommissioned uranium rolling mill from the Manhattan Project, producing 1-inch thick, 12 × 12-inch plates for ballistic testing. The results indicate outstanding properties, with yield strength approaching two gigapascals and toughness significantly exceeding that of conventional alloys such as Aermet and AF 1410. Mathaudhu noted that this could be made of gradient materials. He also emphasized that it was an exemplar that shows “we haven’t exhausted our ability to use the capabilities that we have to roll materials in the United States or to process them in different ways, or in different in configurations to find the best low-weight solutions for protecting our soldiers.” Mathaudhu then began speaking about density noting that in addition to hardness, density is a key factor in production against advanced threats, with higher density materials generally resulting in a higher level of protection. Mathaudhu further remarked that despite this trade-off, there were still opportunities to lower the density of steel armor, such as the efforts of Ryan Howell, project manager Abrams at Program Executive Office Ground Combat Systems, and Richard Gerth, scientist at the U.S. Army’s Tank Automotive Research, Development and Engineering Center (now called the Combat Capabilities Development Command Ground Vehicle Systems Center [GVSC]). They have made progress in designing a high-manganese, high-aluminum steel alloy with a hardness of 350 HBW where the density goes from roughly 7.2 g/cc to 6.5 g/cc. He noted that these materials could be a suitable substitution for large steel plate applications, such as the ballistic under protection in armored vehicles.

Mathaudhu also emphasized that lightweight materials like aluminum, titanium, and magnesium will not stand up to the threats expected on the 2040 battlefield, and it is unrealistic to expect a fully armored vehicle with only 10 tons of armor. He urged moving beyond thinking only about lighter components and instead thinking about adaptive or modular protection or remote systems. He suggested one solution would be to replace soldiers with drones or uncrewed vehicles to maintain vehicle protection and reduce steel usage. New, more significant high-energy threats have emerged, and were seen during the Ukraine war. Moving to uncrewed, remote systems would reduce the need for heavy steel protection and save lives. Combining soldiers with robot-driven vehicles could allow for the same battlefield lethality without the need to invest in new armor technologies. In terms of the material combinations previously described, Mathaudhu noted that MilSpecs would need to be reconsidered to define the minimum qualifications and material combinations for the most accelerated testing to ensure reliability for soldiers. By doing so, production could be accelerated, thereby creating a strong supply chain. Researchers also need to identify what can be reliably

simulated, which will require a unified effort to work with the AI and computational tools available.

During a Q&A, Mathaudhu noted that the benefits of AI have not fully been explored or exploited. Sciarretta noted that AI may help with the physics involved with analyzing threat–armor interactions, specifically in helping to determine the computational fluid dynamics of the velocity of munitions rounds. Mathaudhu agreed this could be helpful.

MacFarland and others expressed that they liked the idea of modularity. MacFarland agreed with the idea of using AI and robotics to replace part of the crew to reduce the need for armor. Wright explored an idea to put aluminum in the air gaps that Mathaudhu had suggested placing between steel. Mathaudhu answered that adding aluminum instead of air would need further analysis, and he was similarly unsure if replacing air with a composite material would be helpful.

CLOSING DISCUSSION AND THEMES OF THE WORKSHOP

Throughout the presentations and during discussions, Gray, MacFarland, and other workshop participants observed several recurring themes. During the closing discussion of the workshop, each participant and members of the workshop planning committee provided their concluding remarks, and the themes are summarized in brief below.

considerations like evaluating stockpiling effectiveness (e.g., the MRAP case study) by analyzing its impact on delivery times, and which materials should be stockpiled (e.g., what thickness), and explored the implications of stockpiling the wrong materials, or those that become outdated, which may cause significant waste. Workshop participants emphasized that understanding material requirements would be the first step in determining which materials to stockpile and noted rare materials could be stockpiled for future uses.

PREVIEW SUMMARY OF WORKSHOP: PART 2

The supplement to the workshop, due to the restricted nature of its topic areas, was held on separate dates, June 3–4, 2025, and focused on vehicle design, requirements generation, and protection as a system. Below is a summary of the meeting’s agenda with some additional details on the focus areas of the discussion. For a full summary of those workshop discussions, please contact the workshop sponsor.

Day 1 opened with a series of briefings on the current state of threats facing armored formations and the outlook on future active protection systems. During this session, the committee and participants were briefed by the Army G2, National Ground Intelligence Center, and GVSC. The briefings were followed by a panel discussion focused on the challenges and opportunities in requirements generation and contracting for future combat vehicles. Jim Schirmer, senior vice president and deputy managing director for American Vehicles at Rheinmetall, and Snyder served as panelists for the discussion.

Day 2 opened with a panel discussion on vehicle design opportunities and trade-offs. Geoff Norman, director of U.S. Strategy and Growth at General Dynamics Land Systems; Erick Sagebiel, director of Vehicle Survivability at Integris Composites; and Jeff Carie, associate director, GVSC Advanced Concepts Team, served as panelists for the discussion. The first panel session was followed by a presentation from the Army’s The Research and Analysis Center (TRAC), headed by Theodore “T.W.” Jackson, director of studies and analysis at TRAC-White Sands Missile Range, on the role of modeling and simulation in future combat vehicle design. The day was concluded with a panel discussion on “Protection as a System” focusing on the trade-offs between individual vehicle and formation-wide protection. Lou DiMarco, a professor at the United States Army Command and General Staff College; Stephen Rapp, senior scientist and decision analysis process leader at GVSC; and Marc Taylor, program engineering manager at Integris Composites, served as panelists for the discussion.

DISCLAIMER This Proceedings of a Workshop—in Brief was prepared by Liza Hamilton as a factual summary of what occurred at the workshop. The statements made are those of the rapporteur or individual workshop participants and do not necessarily represent the views of all workshop participants; the planning committee; or the National Academies of Sciences, Engineering, and Medicine.

PLANNING COMMITTEE George T. (Rusty) Gray (NAE) (Chair), Los Alamos National Laboratory; Julie A. Christodoulou, Materials and Manufacturing Innovations, LLC; Thomas Kurfess (NAE), Georgia Institute of Technology; LTG Sean MacFarland, U.S. Army War College; Phillip Miller, Los Alamos National Laboratory; Mark Rupersburg, Rafael Systems Global Sustainment, LLC; Timothy Snyder, Oshkosh Defense, LLC; Valerie Wiesner, NASA Langley Research Center. The National Academies’ planning committees are solely responsible for organizing the workshop, identifying topics, and choosing speakers. Responsibility for the final content rests entirely with the rapporteur and the National Academies.

REVIEWERS To ensure that it meets institutional standards for quality and objectivity, this Proceedings of a Workshop—in Brief was reviewed by Suveen Mathaudhu, Colorado School of Mines; Cameron Oskvig, National Academies of Sciences, Engineering, and Medicine; Jim Schirmir, American Rheinmetall; and Timothy Snyder, Oshkosh Defense, LLC. Katiria Ortiz, National Academies of Sciences, Engineering, and Medicine, served as the review coordinator.

SPONSOR This Proceedings of a Workshop—in Brief is based on work supported by the Office of Deputy Assistant Secretary of the Army for Research and Technology and the Army Research Office under Contract No. W911NF-23-D-0002. Any opinions, findings, conclusions, or recommendations expressed in this publication do not necessarily reflect the views of any organization or agency that provided support for the project.

STAFF William (Bruno) Millonig, Steven Darbes, Micaela Pacheco, Debrah Adedeji, and Tina Latimer

SUGGESTED CITATION National Academies of Sciences, Engineering, and Medicine. 2025. U.S. Army Next Generation Armor: Proceedings of a Workshop—in Brief. Washington, DC: National Academies Press. https://doi.org/10.17226/29220.

For additional information regarding the workshop, visit https://www.nationalacademies.org/our-work/army-next-generation-armor-workshop.

Division on Engineering and Physical Sciences Copyright 2025 by the National Academy of Sciences. All rights reserved.