Power Electronics Strategic Framework - Draft Report..pdf

Wide Bandgap Power Electronics Strategic Framework

January 2025

DRAFT REPORT

DRAFT – Wide Bandgap Power Electronics Strategic Framework ii

U.S. Department of Energy | Office of Energy Efficiency & Renewable Energy iii

Disclaimer This work was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, nor any of their contractors, subcontractors or their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or any third party’s use or the results of such use of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise, does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof or its contractors or subcontractors. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof, its contractors or subcontractors.

iv

Authors The U.S. Department of Energy’s Advanced Materials and Manufacturing Technologies Office coordinated the development of this strategy. The lead authors are:

Syers, Paul, U.S. Department of Energy (DOE)

Soloman, Nebiat, U.S. Department of Energy (DOE)

Joel Varley, Lawrence Livermore National Laboratory

Ted Laurence, Lawrence Livermore National Laboratory

Lars Voss, Lawrence Livermore National Laboratory

Andriy Zakutayev, National Renewable Energy Laboratory

Mowafak Al-Jassim, National Renewable Energy Laboratory

Andrew Norman, National Renewable Energy Laboratory

Brooks Tellekamp, National Renewable Energy Laboratory

Stephen Schaefer, National Renewable Energy Laboratory

Kingsley Egbo, National Renewable Energy Laboratory

Syed Hasan, National Renewable Energy Laboratory

Faisal Khan, National Renewable Energy Laboratory

Shuofeng Zhao, National Renewable Energy Laboratory

Gilbert Moreno, National Renewable Energy Laboratory

Paul Paret, National Renewable Energy Laboratory

Sreekant Narumanchi, National Renewable Energy Laboratory

Barry Mather, National Renewable Energy Laboratory

Ram Thiagarajan, National Renewable Energy Laboratory

Sam Reese, National Renewable Energy Laboratory

Sarwar Islam, National Renewable Energy Laboratory

Xuhui Feng, National Renewable Energy Laboratory

Bidzina Kekelia, National Renewable Energy Laboratory

Ross Kerner, National Renewable Energy Laboratory

Doug DeVoto, National Renewable Energy Laboratory v

Andrew Binder, Sandia National Laboratory

Jack Flicker, Sandia National Laboratory

Jake Mueller, Sandia National Laboratory

Felipe Palacios, Sandia National Laboratory

Michael Rios, Sandia National Laboratory

Andrew Dow, Sandia National Laboratory

Richard Floyd, Sandia National Laboratory

Kevin Reilly, Sandia National Laboratory

Brian Rummel, Sandia National Laboratory

Peter Dickens, Sandia National Laboratory

Tony Rice, Sandia National Laboratory

Todd Monson, Sandia National Laboratory

Luke Yates, Sandia National Laboratory

Leah Appelhans, Sandia National Laboratory

Acknowledgments This document was developed under the guidance of Paul Syers, Nebiat Soloman, Tina Kaarsberg, and Diana Bauer of the U.S. Department of Energy (DOE) Office of Energy Efficiency and Renewable Energy, Advanced Materials and Manufacturing Technologies Office (AMMTO). The members of the working groups who made significant contributions through facilitated workshop sessions are identified below. ICF International and Nexight Group supported the road mapping process and report development.

Working Group Members

Special thanks to Paul Syers, DOE, who provided DOE leadership to all working groups, and to Doug DeVoto, NREL and Jonathan Bock, SNL for their crosscutting support.

Materials and Active Components

Andrew Binder, SNL (Working Group Chair)

Andriy Zakutayev, NREL

Mowafak Al-Jassim, NREL vi

Andrew Norman, NREL

Richard Floyd, SNL

Kevin Reilly, SNL

Brian Rummel, SNL

Peter Dickens, SNL

Tony Rice, SNL

Brooks Tellekamp, NREL

Stephen Schaefer, NREL

Kingsley Egbo, NREL

Syed Hasan, NREL

Reactive & Passive Components and Packaging

Faisal Khan, NREL (Working Group Chair)

Shuofeng Zhao, NREL

Gilbert Moreno, NREL

Paul Paret, NREL

Sreekant Narumanchi, NREL

Barry Mather, NREL

Ram Thiagarajan, NREL

Sam Reese, NREL

Sarwar Islam, NREL

Todd Monson, SNL

Luke Yates, SNL

Leah Appelhans, SNL

System-level Design and Application-Specific Opportunities

Jack Flicker, SNL (Working Group Chair)

Barry Mather, NREL

Faisal Khan, NREL

Shuofeng Zhao, NREL

Xuhui Feng, NREL

Bidzina Kekelia, NREL

Sreekant Narumanchi, NREL

Jake Mueller, SNL

Felipe Palacios, SNL vii

Michael Rios, SNL

Andrew Dow, SNL

Life Cycle Assessment (LCA) for Power Electronics Supply Chain

Samantha Reese, NREL (Working Group Chair)

Ross Kerner, NREL

Andriy Zakutayev, NREL

Barry Mather, NREL

Doug DeVoto, NREL

Analysis Measurement & Characterization

Joel Varley, LLNL (Working Group Chair)

Ted Laurence, LLNL

Lars Voss, LLNL

Strategic Framework Draft Development and Editing Support Staff

ICF International staff members who supported the development of the document, including working group management, stakeholder engagement, authorship, and editing include Ridah Sabouni, Thalib Razi, Wayne Johnson, Andy Ritenour, Emmanuel Taylor, Whitney Bell, Christopher Costello, and Corey Fenwick.

Nexight Group staff members who contributed to document authorship, editing, and review include Jackie Bowen, Lindsay Pack, Jared Kosters, and Karsten Daponte.

This report was prepared by ICF and the Lawrence Livermore, National Renewable Energy, and Sandia National Laboratories of or the U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, Advanced Materials and Manufacturing Technologies Office.

viii

List of Acronyms 2DEG Two-dimensional electron gas

AC Alternating current

Ag Silver

AI Artificial intelligence

Al Aluminum

Al2O3 Aluminum oxide

AlGaN Aluminum gallium nitride

AIN Aluminum nitride

ALD Atomic layer deposited

AM Additive Manufacturing

AMB Active metal brazing

AMMTO Advanced Materials and Manufacturing Technologies Office

ARPA-E Advanced Research Projects Agency – Energy

BaTiO3 Barium Titanate

BESS Battery Energy Storage Systems

BEV Battery electric vehicles

BiGaN Bidirectional Gallium Nitride

Bi(M)O3-BaTiO3 Bismuth modified Barium Titenate, including an additional element, represented as “M”

BFOM Baliga figure-of-merit

C Celsius cm Centimeters

CMOS Complementary metal-oxide-semiconductors

CO2 Carbon dioxide ix

CPU Central Processing Unit

CTE Coefficient of thermal expansion

Cu Copper

CZ Czochralski

DBA Direct bond aluminum

DBC Direct bond copper

DC Direct current

DER Distributed energy resources

DMOSFET “Depletion Mode” MOSFET

DOE U.S. Department of Energy

DSP Digital signal processing

EFG Edge-defined film-fed growth

EIM Electromagnetic interference

EM Electromagnetic

E-mode Enhancement mode

EoL End-of-life

EPA U.S. Environmental Protection Agency

ESL Equivalent series inductance

ESR Equivalent series resistance

EV Electric vehicle

EWD Education and workforce development

FinFET A type of field effect transistor with taller, “fin-like” structure

FOM Figure of Merit

FQS Four quadrant switches

Ga Gallium x

GaN Gallium Nitride

Ga2O3 Gallium Oxide

GDP Gross domestic product

GHG Greenhouse gas

GPU Graphics processing unit

GREET Greenhouse gases, Regulated Emissions, and Energy use in Transportation

GW Gigawatt

GW-hr Gigawatt hours

HEMT High electron mobility transistor

HFTO Hydrogen and Fuel Cell Technology Office

HVAC Heating ventilation and cooling

HVDC High voltage direct current

HVPE Hydride vapor-phase epitaxy

ICE Internal combustion engine

IGBT Insulated-gate bipolar transistor

IMOSFET Trench-form MOSFET. The deep trench shape looks like an “I”

InAIN Indum Aluminum Nitride

Ir Iridium

JFET Junction field-effect transistor

Kg Kilogram kHz Kilohertz km Kilometers kW Kilowatt kW-hr Kilowatt-hour

LCA Lifecycle assessments xi

LCOE Levelized cost of electricity m Meters

MBE Molecular beam epitaxy

MC-HEMT Multi-channel HEMT

MC2-HEMT Multi-channel monolithic-cascode HEMT

MECS Manufacturing Energy Consumption Survey

MESFET Metal-semiconductor field-effect transistors

MeV Mega electron volts

MHz Megahertz

ML Machine learning

MME Metal-modulated epitaxy

MMT Million metric tons

MOCVD Metal-organic chemical vapor deposition

MOSFET Metal-oxide-semiconductor field-effect transistor

MTDC Multi-terminal HVDC systems

MVDC Medium voltage DC

MW Megawatt

O&M Operations and maintenance

P2H Power-to-hydrogen

PCB Printed circuit boards

PCR Primary control reserve

PE Power electronics

POLFETS Polarization-doped field effect transitors

Pt Platinum

PTFE Polytetraflouroethylene xii

PV Photovoltaic

R&D Research and development

RCH Channel resistance

RF Radio frequency

Rh Rhodium

RSUB Substrate resistance

SCR Secondary control reserve

Si Silicon

SiC Silicon carbide

SIPO Series input, parallel, output

SMC Soft magnetic composites

STEM Science, technology, engineering, and mathematics

SWaP Size, weight, and power

SWaP-C Size, weight, power, cost

T&D Transmission and distribution

TCO Total cost of ownership

TECHTEST Techno-economic, Energy, & Carbon Heuristic Tool for Early-State Technologies

TEA Techno-economic analysis

TWh Terawatt hours

UFOM Unipolar figure-of-merit

USGS U.S. Geological Survey

UWBG Ultra-wide bandgap

V Volt

VSD Variable speed drives

WBG Wide bandgap xiii

WTG Wind turbine generators xiv

Executive Summary This U.S. Department of Energy (DOE) Office of Energy Efficiency and Renewable Energy, Advanced Materials and Manufacturing Technologies Office (AMMTO) strategic framework identifies and elucidates the necessary technological innovations for wide bandgap (WBG) and ultra-wide bandgap (UWBG) power conversion technologies to advance America’s energy independence and secure our power grid. WBG and UWBG materials offer performance and efficiency advantages over their silicon-based counterparts, and the opportunities for design and manufacturing improvements are significant that the future of the power electronics industry is mainly through leveraging WBG and UWBG materials. This framework is divided into specific areas of focus, considering the needs of 1) the various applications that use PE and the infrastructures they make up, 2) the common materials and devices shared across many PE technologies, 3) the packaging and passive elements that support PE devices and build full systems, and 4) nontechnical needs for the PE industry to grow and thrive. In order to meet these needs, it encourages those working to advance these areas to follow the recommendations and goals laid out in this framework, including sharing information and partnering with each other and the DOE on research, development, demonstration and integration activities where appropriate. AMMTO also encourages readers of this strategic framework to work with the DOE to build detailed roadmaps that industry can collectively follow to achieve the ambitious goals set forth in this framework.

Background

PE are essential to the conversion and control of electrical power in U.S. critical infrastructure. As this infrastructure—including grid-integrated power, industrial manufacturing, transportation, and digital networking and communications—brings more innovative technologies online to meet changing market demands and improve grid resiliency, it also places significantly higher performance demands on PE technologies.

Silicon (Si) PE have been relied on for decades to meet infrastructure’s power conversion and control needs, but these intensifying performance requirements have surpassed what traditional Si PE can offer.

Wide bandgap semiconductor (WBG) PE offer a replacement that has the potential to address U.S. infrastructure’s changing needs, advancing America’s energy independence and securing our power grid. To achieve this potential, technological innovation is needed across the entire PE system—at the material level, in the devices that allow these high-performance materials to function, and in the packaging that builds these devices into end-use applications.

In addition, the size, weight, power, cost (SWaP-C), and reliability requirements differ across PE applications as well as the infrastructure sector in which they are implemented. These “levers” need to be carefully considered and adjusted appropriately in the design phase to ensure that end products meet their sector-specific needs. The ability to source materials and manufacture advanced PE technologies at scale and at the differing volumes needed by each sector are also critical considerations in the innovation stage.

xv

AMMTO supports the enhancement of economic competitiveness of U.S. manufacturing through increased energy efficiency and improved energy conversion technologies, and improving PE technologies is a crosscutting way to accomplish these goals. Improving PE technologies means not just increasing the efficiency and performance of PE in multiple application areas to meet future industry demands, such as supporting the reliability of the grid under expected future conditions, improving the efficiency and performance of the transportation infrastructure, enabling continued electrification of industrial processes to improve industrial flexibility and efficiency, and improving the efficiency of data centers and communication networks to handle ever increasing demand. It also means strengthening the PE innovation ecosystem through increased collaboration and information sharing, to shorten the time from idea to impact in the industry. To best focus the office’s efforts to support the U.S. power electronics industry and to provide a vision for the future that can inspire coordinated action, AMMTO set out in creating a document that provides a strategic overview of the PE industry, its current trends, challenges, and pathways of opportunity to further growth and competitiveness.

The AMMTO Wide Bandgap Power Electronics Strategic Framework aims to address the challenges and considerations above by examining them in relation to different aspects of the PE landscape (e.g., application, materials and devices, passives and packaging, and nontechnical) and identifying the pathways and recommendations necessary for high-performance PE advance America’s energy independence and secure our power grid.

Framework Overview and Summary of Recommendations

This framework begins by providing necessary context of the importance of PE to many major modern infrastructures as well as an overview of the ecosystem that drives PE innovation and manufacturing, and trends driving the growing shift to WBG PE, which also signal possible future demand for UWBG PE. Following this, the framework is divided into four main sections, each addressing a major focus area of the PE industry.

Each focus area highlights some current challenges and innovation pathways to address them. Focus Area 1 takes a market needs perspective, looking at the variety of applications, grouped by major infrastructure type, for power electronics. Focus Area 2 looks at the core devices that make up PE and the growing and emerging materials systems upon which they are built. Focus Area 3 expands the technical scope from active devices to the supporting passive devices and the packaging materials & methods that work together with the active PE devices to build complete, fully functioning PE modules and systems. Finally, Focus Area 4 looks at non-technical aspects that the PE industry needs to thrive, such as a well-trained workforce and a robust understanding of full lifecycle and techno-economic drivers of the PE supply chain.

Each section also provides targeted recommendations, both technical and nontechnical in nature, to accelerate the innovations and impacts mentioned in that section. Here the recommendations from each section are listed.

xvi xvii xviii

Strategic Goals for WBG PE

The collected recommendations summarized above were also analyzed together to identify common themes and connections among recommendations. In alignment with the different aspects of the PE landscape and AMMTO’s vision for a competitive U.S.

manufacturing sector supporting American energy independence, the Office has identified six specific strategic goals for advancing WBG PE, which are outlined below.

Three of these strategic goals are technical in nature, and three are focused more broadly on supporting the growth of the U.S. power electronics manufacturing ecosystem. AMMTO and the authors of this framework hope that readers follow individual recommendations most relevant to their work in the PE ecosystem, described in the sections of this framework and summarized above, but we especially encourage the industry to work towards accomplishing the six goals described below. AMMTO welcomes input and engagement from industry to help develop a commonly supported roadmap to lay out timelines and estimated levels of coordination and effort to help achieve these goals.

xix

Table ES-1. PE Strategic Framework Goals and Areas of Focus

Technical Goals Ecosystem Support Goals

T1. Increase system power density E1. Collaborate on pre-competitive R&D and innovation

Increasing system power density is the most impactful single metric for most PE applications and will in most cases lead to cost reductions and other performance improvements. Nearly all major applications for PE will benefit from the impacts of improving system power density.

Higher frequency operations are the most promising pathway to achieve this, as higher frequency operation results in more compact and efficient PE systems. Significant effort is needed to optimize materials, manufacturing processes, and system designs (including device, passives, and even packaging) to support high-frequency performance while maintaining system reliability and stability.

Achieving the full potential of WBG PE demands a unified effort across the entire innovation ecosystem. Unlocking the transformative benefits of WBG technologies requires collaboration on pre-competitive R&D that brings together industry leaders, academic researchers, government agencies, and research institutions. Leveraging the strengths and resources of all stakeholders will accelerate advancements in WBG PE technologies, drive down costs, overcome critical technical barriers, and encourage broader adoption.

T2. Enable higher voltage and power operation for grid applications

E2. Quantify and validate the full lifecycle costs & benefits of WBG PE

Achieving higher voltage and power operation will facilitate the integration of renewable energy sources, advanced grid management systems, and emerging high-power applications in areas like energy storage and transportation electrification. The need for PE systems that can operate at higher voltages and power delivery is acute.

Leverage techno-economic analysis (TEA) methodologies as tools for strategic planning and resource allocation in PE innovation. By systematically analyzing costs, benefits, and potential barriers, TEA can pinpoint specific areas where investment is needed to optimize performance, reduce costs, improve efficiency, overcome technical challenges, and expand market adoption for WBG-PE systems.

Conduct comprehensive lifecycle assessments (LCAs) to demonstrate the performance, environmental, and economic benefits of WBG technologies over traditional Si-based PE. Utilize LCAs to baseline environmental impacts, promote sustainable practices, and target advancements that foster circularity and waste reduction.

xx

Technical Goals Ecosystem Support Goals

T3. Integrate and commercialize materials advancements to improve PE devices & systems

E3. Increase educational and workforce training activities across the country to support PE industry growth

Identification, development, and improvement of the fundamental materials in PE devices and packages are the most impactful pathway to improving the performance of the overall systems themselves. Complete integration and commercialization also require improvements in manufacturing processes and scale-up of the technologies that incorporate these materials, as well as supply chain optimization to ensure reliable access to them. Driving the largest impacts in WBG PE technologies will happen through materials innovation.

Expand educational programs and workforce training initiatives at all levels— ranging from technical training for manufacturing roles to advanced research programs for engineering and design positions. Establish partnerships with academic institutions, industry, and government to ensure a steady pipeline of skilled workers to meet the growing demands of the domestic PE industry.

xxi

Table of Contents Working Group Members ............................................................................................. v

Materials and Active Components ............................................................................ v

Reactive & Passive Components and Packaging ....................................................vi

System-level Design and Application-Specific Opportunities ...................................vi

Life Cycle Assessment (LCA) for Power Electronics Supply Chain......................... vii

Analysis Measurement & Characterization ............................................................. vii

Strategic Framework Draft Development and Editing Support Staff ........................... vii

Executive Summary ...................................................................................................... xiv

Background ............................................................................................................... xiv

Recommendations Overview ......................................................................................xv

Strategic Goals for WBG PE .................................................................................... xviii

1 Introduction

1.1 What are PE?

1.2 Hierarchical Nature of PE

1.3 Trends Driving WBG PE Adoption

1.4 Advantages of WBG and UWBG Semiconductors

1.5 Strategic Framework Structure and Organization

Focus Area 1: WBG PE Applications to Infrastructure

2.1 Grid Infrastructure

2.2 Transportation Infrastructure

2.3 Digital Infrastructure

2.4 Industrial Infrastructure

3 Focus Area 2: Materials and Devices for WBG PE

3.1 Silicon Carbide

2.1 Gallium Nitride

3.2 Ultrawide Bandgap Materials: Aluminum Gallium Nitride and Aluminum Nitride

3.3 Ultrawide Bandgap Materials: Gallium Oxide

4 Focus Area 3: Passive Components and Packaging for WBG PE xxii

4.1 Passive Components

4.2 Packaging

5 Focus Area 4: Non-Technical WBG Development Support Areas

5.1 Education and Workforce Development for the U.S. PE Industry

5.2 Techno-Economic Development and Lifecycle Assessment for the U.S. PE Industry

5.3 Recommendations to Address Non-Technical WBG Development Challenges

6 Path Forward

6.1 Call to Action

End Notes xxiii

List of Figures Figure ES-1. Need caption .............................................. Error! Bookmark not defined.

Figure ES-2. Need caption .............................................. Error! Bookmark not defined.

Figure ES-3. Need caption .............................................. Error! Bookmark not defined.

Figure 1. The “atoms to applications” hierarchical nature of PE, in which the capabilities of lower levels of the pyramid determine the capabilities at the apex, or application-level Figure 2. The “atoms to applications” hierarchical pyramid only differentiates at an end-use application Figure 3. Energy bandgap comparison of conventional semiconductors (e.g., Si), WBG (e.g., SiC), and UWBG (e.g., gallium oxide) Figure 4. Contours of constant Baliga figure-of-merit (BFOM) showing relationship between on-resistance and breakdown voltage for various conventional, WBG, and UWBG semiconductors Figure 5. Growth in Energy Storage Deployments Figure 6. Losses in Battery Energy Storage Systems (BESSs) Figure 7. Share of transportation sector emissions by source Figure 8. PE components in a BEV Figure 9. Power vs frequency comparison of different semiconductor technologies for EV power converters Figure 10. Heavy duty vehicle charging technologies overview Figure 11. Architecture of power-to-hydrogen(P2H) with renewable energy input for industrial utilization Figure 12. Technology maturity and application space chart for various material systems showing select highlighted research areas considered to be impactful Figure 13. Innoscience BiGaN can replace two back-to-back Si MOSFETs Figure 14. Comparison of various unipolar figure-of-merit (UFOM) plots: (a) “classic” plot, which shows an incredibly favorable outlook for GaN, (b) an updated UFOM, and

(c) direct comparison between the classic and updated UFOMs Figure 15. Calculated unipolar figure of merit lines for Si and several WBG and UWBG semiconductors Figure 16. Schematic of vertical β-Ga2O3 based transistor employing an integrated epitaxial p-type layer (JFET) or an extreme k-dielectric oxide (MOSFET) towards > 5kV breakdown voltages Figure 17. Advances in Ga2O3 epitaxy and bulk growth are both necessary for large-scale deployment Figure 18. Artist’s rendition of an inorganic/organic soft magnetic nanocomposite Figure 19. Power device package with associated requirements Figure 20. Comparison of heat flux capacity and thermal resistance of different cooling designs xxiv

Figure 21. Advanced power module designs for SiC [16]

List of Tables Table ES-1. PE Strategic Framework Goals and Areas of Focus ................................. xix Table 1. Priority Metrics for PE Application Areas Table 2. Comparison of Single-Sided and Double-Sided Cooling Designs

1 Introduction Power electronics (PE) are essential to the conversion and control of electrical power in U.S. critical infrastructure. Innovations in PE will help the United States meet rising energy demands and improve grid resiliency goals through more efficient energy conversion, grid stability and reliability, and integration of renewable energy sources.

Modern society has been rapidly changing over the past 50 years, which has also impacted how energy is produced, transmitted, and used. These changes have led to an overall increase in conversion, transmission, and use of energy in the form of electricity, as well as the complexity and volatility of demand on all levels of the nation’s electric grid. Ever higher demand is placed on the PE that enable and control the conversion, transmission, and use of electricity throughout our nation. This section provides context of what PE are as a technology category, current trends driving the PE industry, and relevant details of how wide-bandgap (WBG) PE compare to the legacy Silicon (Si)-based PE in wide use today.

1.1 What are PE?

PE enable the manipulation of electricity, allowing it to be delivered and used when and how it is wanted and needed. They control power flow at both the point of generation and usage, convert electrical energy from one form to another (e.g., from alternating current [AC] to direct current [DC]), and change voltage levels as needed. Roughly 70% of all electrical energy today is handled by PE devices at some point between generation, transmission, and usage, with that percentage expected to reach 100% in the future.1

Compared with microelectronics, which process and control electricity for the purpose of storing, transforming, and transmitting data and therefore only need to operate at a narrow range of power and frequency, PE operate at the milliwatt (mW) or megawatt (MW) level2 with up to a few megahertz (MHz) switching frequency.3 This makes PE essential to the larger scale conversion and control of electrical power across a range of U.S. critical infrastructures:

Grid infrastructure – Enables electric grids to process and deliver power with efficiency and reliability. Example: Advanced PE—including solid-state transformers, fault current limiters, high-voltage direct current systems, and power flow controllers—can reduce transmission and distribution (T&D) losses, optimize power delivery, protect critical assets, and enhance resilience.4

Industrial infrastructure – Helps power large-scale manufacturing, including use within equipment for enhanced control. Example: Variable speed drives (VSDs) improve productivity and energy efficiency by allowing operators to optimize motor speed based on loading conditions for industrial equipment including pumps, fans, and compressors.5

Transportation infrastructure – Allows for faster and more reliable charging of electric vehicles (EVs)6 and use of power within vehicles. Example: PE circuits including battery packs, power actuators, and traction motors help to facilitate the efficient conversion, transformation, and transfer of electrical power through the powertrain.7

Digital infrastructure – Facilitates seamless, modern communication networks including enabling signal processing and transmission through circuits to enhance signals, improving system performance through digital signal processing (DSP) and power-efficient amplifiers, and enabling wireless power transfer technologies.

While both microelectronics and PE are essential to critical infrastructure, microelectronics are essential in the same way to all infrastructures (i.e., electricity for data) while PE are essential in different ways to different infrastructures. This means that microelectronics have traditionally relied on fewer fundamental system components (e.g., CPU, GPU, memory, and storage, all built on Si complementary metal-oxide-semiconductors [CMOS]), while PE require different core technologies for different application areas (e.g., high voltage direct current [HVDC] vs micro-inverters vs electric motors, etc.).

1.2 Hierarchical Nature of PE

In addition to requiring different core technologies for different applications, PE innovation must also occur at many levels—including at the material, device, and circuit (packaging and thermal management) level—to have the desired effect at the application level. Similarly, the needs at the application level determine the necessary parameters at lower hierarchical levels. This tight coupling between the needs of the application area and the capabilities of subsequent technologies is known as the “atoms to applications” concept, which is demonstrated in the DOE/AMMTO-developed figure, shown below (see Error! Reference source not found.).

Figure 1. The “atoms to applications” hierarchical nature of PE, in which the capabilities of lower levels of the pyramid determine the capabilities at the apex, or application-level

While each application has its own hierarchical structure, there is significant overlap at lower levels of the hierarchy with differentiation occurring mainly at the end-use application level. As a result, fundamental research in materials, devices, circuits, and subsystems can impact multiple end-use applications levels. For example, power converters that interface solar energy resources to the grid and power converters that interface wind turbines to the grid have significantly different application areas, while the underlying technologies (e.g., the power conversion circuits, switching and passive elements, and materials systems) may be identical. Fundamental innovations in, for example, silicon carbide (SiC) switching elements, may impact multiple application levels simultaneously.

Because the innovations and usages at lower hierarchical levels are closely linked, there can be significant commonality between circuit architectures, devices, and materials systems supporting similar applications. As a result, the isolated hierarchy shown in Figure 1 actually becomes nested pyramidal hierarchies with significant areas of overlap at the lower levels, so that some innovations at lower hierarchies can impact multiple application areas, as shown in the DOE/AMMTO-developed figure, shown below (see Figure 2). For example, both photovoltaics and energy storage require medium voltage interconnects moderated through PE converters. Doing this efficiently and economically requires the advent of modular, medium voltage PE in certain cases that can be sized for specific system voltages and power levels arbitrarily.

Figure 2. The “atoms to applications” hierarchical pyramid only differentiates at an end-use application

1.3 Trends Driving WBG PE Adoption

To date, silicon (Si) has been the primary semiconductor used in PE devices across all industries—in 2022, Si-based PE accounted for 96% of the market share by value.8 But the following trends have been driving a search for alternatives:

Increasing demand for higher performing PE to ease the adoption of distributed energy resources (DERs). The future resilient and reliable grid of 2035 requires the addition of 100s of GW of solar, wind, and other renewable resources. Wind energy specifically requires higher voltage PE to enable the operation of larger wind turbines (e.g., 34.5 kV, AC is considered a medium voltage rating for 100+ MW wind plants). Wind energy also requires simpler, more efficient PE to connect to the electric grid, since wind energy inherently relies more on the transmission system to deliver power due to the best wind resource areas often being far from load centers. In addition, solar and energy storage need better performing PE to achieve better SWaP and longer lifetimes.

The modernization of the electrical grid is projected to require 20-50% of all electricity generation to come from photovoltaic (PV) sources. To achieve this, DOE has set aggressive cost goals to lower the levelized cost of electricity (LCOE) of utility-scale solar energy (< 0.03 $/kW-hr for utility-scale installations by 2030), which requires significant reductions in inverter costs and improvements in service lifetimes.

Increasing need for responsiveness of the electric grid due to faster changing and less predictable electric loads. The electrification of daily life in the United States—such as household electronics, electric buildings systems (e.g., HVAC, heat pumps), data centers, and the growth of EV charging and distributed solar—has resulted in a shift from more stable electric loads and sources to faster switching, less predictable loads and sources. Advanced PE technologies that can handle higher frequencies and have increased reliability are essential to enabling this transition.

Increasing priority of efficiency for power solutions in growing markets.

Demands for higher efficiency technology in data and transportation markets are increasing pressure to develop advanced PE technologies. The advent of cloud computing has resulted in an explosion of systems that store, process, and network data, all of which are inherently electrical processes—estimated global data center electricity consumption in 2022 was 240-340 TWh, or around 1-1.3% of global electricity demand. New methods of power routing and usage enabled by advanced PE are essential to ensuring that the grid can continue meet these rapidly expanding electricity usage needs. In the EV market, there is a push to make electric powertrains even more efficient to reduce component footprint and improve overall vehicle performance. Increasing PE switching frequency can help reduce the size and weight of PE components but also relies on greater improvements and reliability in PE technologies.

Demand for conventional and wide band semiconductors is projected to increase by more than an order of magnitude by 2050.9 The $20.9B 2022 power electronic market (including discrete and modules) is expected to reach $33.3B by 2028, with the 1200V voltage range expected to have the largest increase.10 Yole further projects that WBG materials such as SiC and GaN will go from an 8% market share in 2022 to 28% in 2028. Additional details concerning market dynamics are described in the application areas chapter of this document.

1.4 Advantages of WBG and UWBG Semiconductors

PE technologies rely on semiconductor-based devices, which directly influence the overall size, weight, cost, and efficiency of subsequent power systems. While Si has been the predominant semiconductor in PE systems for many decades, application requirements have exceeded the material limits of Si and demand a more suitable alternative. WBG and UWBG material systems have a significantly wider energy gap between the valence and conduction bands (Figure 3), which results in significantly higher breakdown strength and operating temperature for these materials. Thus, for a given breakdown voltage rating, these materials provide lower specific on-resistance (Figure 4).

Figure 3. Energy bandgap comparison of conventional semiconductors (e.g., Si), WBG (e.g., SiC), and

UWBG (e.g., gallium oxide)

Figure 4. Contours of constant Baliga figure-of-merit (BFOM) showing relationship between on-resistance and breakdown voltage for various conventional, WBG, and UWBG semiconductors11

Achieving lower specific on-resistance for a given voltage rating ultimately enables reduced chip area (die size) and correspondingly lower device capacitance, which increases switching frequency and reduces losses. In total, WBG and UWBG materials offer significant improvements in power handling, switching speed, and temperature robustness in a smaller form factor. These benefits allow the use of smaller passive components and reduce the external cooling requirements, which are both major factors in the overall size, weight, and cost of the system. The increased frequency of operation and power density of WBG-based devices can advance efficiency and reduce chip-size requirements beyond what is possible with Si-based devices. Thus, overall system performance can be increased while costs can be reduced by transitioning to WBG-based devices. Furthermore, in some cases (U)WGB devices enable operation in environments that are not supported with legacy Si-based devices. For example, SiC devices have been demonstrated to operate up to 500°C,12 and (U)WBG devices can offer significantly more resilience in radiation environments.13 Additionally, from a systems perspective, some applications are simply not viable without the benefit derived from (U)WBG devices (e.g., pulsed-power and solid-state transformers). As the push toward modernization and the demand for reliable, efficient electricity continue to ramp up across U.S. critical infrastructure, this strategic framework aims to identify recommended actions that will enable the replacement of Si in PE at scale.

1.5 Strategic Framework Structure and Organization

The PE industry draws from a small number of core technologies to ultimately serve a large number of end-use markets with an extremely wide range of performance and cost requirements. Therefore, this framework aims to examine the technical landscape from both the “market pull” and “technology push” perspectives. Furthermore, this framework also considers a few nontechnical but nevertheless critical needs the industry will have over the next decade. The document is thus broken down into the focus areas that follow.

These four focus areas create a comprehensive framework for understanding both the technical and nontechnical needs required to drive the adoption and advancement of WBG PE in critical sectors. In cases where these impactful technologies overlap, this framework intends to:

Introduce the research needs for areas with overlapping lower-level hierarchies only once where they first occur and then refer back to that section in specific application writeups.

Identify targets and goals specific to the given application area

Focus Area 1: WBG PE Applications to Infrastructures

This section explores how WBG PE are transforming key sectors of infrastructure. WBG devices, which offer superior performance in terms of efficiency, power density, and high-temperature operation compared to traditional silicon-based technologies, are now being applied to a wide range of end-use markets:

Grid infrastructure: As power grids evolve to accommodate renewable energy and electric vehicles, WBG PE play a critical role in enhancing grid efficiency, enabling HVDC interconnections, and integrating distributed generation and energy storage systems.

Transportation infrastructure: Electrification of vehicles, from light-duty to heavy-duty, relies heavily on WBG PE for efficient power conversion, fast charging, and managing high-power drive systems. These technologies help overcome key challenges such as electricity demand management and energy efficiency in electric vehicles.

Digital infrastructure: With the rising demand for data centers and digital communication systems, WBG devices are pivotal in reducing energy consumption and improving efficiency in powering data centers, telecommunication systems, and AI-driven infrastructure.

Industrial infrastructure: WBG technologies improve energy efficiency and process control across manufacturing and industrial sectors, reducing energy consumption and improving the performance of industrial drives, motors, and heavy equipment. WBG technologies also serve as a key enabler for hydrogen generation technologies, which are key in modernizing hard-to-electrify industries.

Focus Area 2: Materials and Devices for WBG PE

The foundation of WBG PE lies in the materials and semiconductor devices used to create high-performance power components. This section outlines the different materials that drive advancements in WBG technologies:

Silicon carbide (SiC): Known for its high breakdown voltage, thermal conductivity, and efficiency, SiC is widely used in power converters, especially in high-voltage and high-power applications such as electric vehicles and grid systems.

Gallium nitride (GaN): GaN is ideal for high-frequency and high-efficiency applications, making it a key enabler in power supplies, RF amplifiers, and consumer electronics.

Aluminum gallium nitride (AlGaN) and aluminum nitride (AlN): These materials are emerging for applications that demand extreme power density, thermal performance, and reliability, particularly in aerospace, military, and other high-stress environments.

Gallium oxide: This newer WBG material offers extremely high breakdown voltages, which could enable future breakthroughs in ultra-high voltage applications like solid-state transformers and HVDC power transmission.

Focus Area 3: Passive Devices and Packaging for WBG PE

While WBG semiconductors are critical, the devices’ ultimate performance depends heavily on the passive components and packaging technologies that surround them. This section focuses on:

Established packaging and passives: Traditional packaging solutions and passive components like inductors, capacitors, and transformers are being adapted to handle the higher frequencies, power densities, and temperatures associated with WBG devices.

Emerging packaging and passives: Novel approaches to packaging, such as advanced thermal management techniques and 3D packaging, along with passives designed for ultra-high-speed operation, are being developed to further maximize the performance of WBG technologies.

Focus Area 4: Nontechnical WBG Development Support Areas

In addition to technical innovations, there are crucial nontechnical factors that will drive the successful adoption of WBG technologies. This section examines these broader support areas:

Education and workforce development: As WBG technologies proliferate, there is an urgent need for training programs to develop a skilled workforce proficient in the unique properties and applications of WBG semiconductors. This will ensure that companies can scale and implement WBG technologies effectively.

Lifecycle assessment (LCA): Assessing the environmental impact of WBG PE throughout their lifecycle is essential for promoting sustainable development and meeting global energy efficiency and energy security goals.

Technoeconomic analysis: The widespread adoption of WBG technologies hinges on their cost-competitiveness with established technologies. This area focuses on cost reduction strategies, manufacturing scale-up, and economic models to drive the market penetration of WBG PE.

Focus Area 1: WBG PE Applications to Infrastructures

This chapter is split into four broad Infrastructure Areas: Grid Infrastructure, Transportation Infrastructure, Digital Infrastructure, and Industrial Infrastructure. In each area, this chapter discusses the current state of PE, the technical challenges for WBG PE that need to be overcome, and technological innovation pathways available for addressing those challenges.

The following tables summarizes the key metrics driving innovation in each infrastructure where PE has application. While it can be argued that SWaP-C, reliability, and scalability are critical in all PE applications, some metrics matter more in certain contexts. Table 1 below summarizes the priority metrics for each application area discussed in this chapter.

Table 1. Priority Metrics for PE Used in Various Infrastructures and Applications

S iz e, W e ig h t, P o w e r D e n si ty

H ig

V lt ag e a n d h

P o w e r O p e ra ti

L w C o s t C o m p o n e n t s

R e lia b il it y C ir cu it s

S c al a b il it y h

E ff ic ie n c y P o w e r

C o n ve rs io

Grid Infrastructure

Photovoltaics X X X X

Wind Energy X X X X

Energy Storage X X X X X

Solid-State Transformers X X X

Transportation Infrastructure

Light-Duty Vehicles X X X

Heavy-Duty Vehicles X X X

Electric Rail X X X

Digital Infrastructure

Datacenters X X X X

Industrial Infrastructure

Hydrogen Production X X X

2.1 Grid Infrastructure

Since its formulation in the late 1800s, the electrical grid has been dominated by generation, transmission, and conversion equipment that is electromechanical in nature rather than PE-based. This has yielded the grid that we know today, which is frequently called the largest and most complex machine that mankind has ever devised.14

The existential threat of climate change and the subsequent need for a modern and secure electrical grid require a more efficient, resilient, and cost-efficient electrical grid that enables flexibility at both load and generation.

The future grid of 2035 requires the addition of hundreds of GW of solar, wind, and other renewable resources. Much of this energy will be generated locally, near load centers, and will be interfaced into the electrical grid through PE converters. Significant portions of this new renewable generation will be focused in high-value areas with abundant resources, such as the desert Southwest (for solar), Midwest (for on-shore wind), and Atlantic seaboard (for off-shore wind). Efficient transmission of electricity from these areas to load centers will require multi-terminal HVDC interconnections, which are enabled by PE conversion. To mitigate short-term and long-term variability, generation resources will need hundreds of GW and GW-hrs of both short-term and long-duration (> 4 hour) energy storage. Each generation and storage asset will be interfaced into the electrical grid via PE.

The PE uptake in generation and transmission will be matched by PE penetration at the load side. Increased efficiency has driven PE-mediated load over the past 10 years, and this trend will continue until all loads are PE-based. Additionally, as transportation (light-and heavy-duty EVs, aviation, etc.) and heavy industry become increasingly electrified, the amount of load on the system will significantly increase over the next several decades. The growth of high-power loads—through charging stations to support electrified transportation, proliferation of data centers, and increased industrial electrification—combined with more extreme weather events will also mean that the level of flexibility and resiliency in the grid will need to significantly increase.

2.1.1 Grid Infrastructure Application Areas

Application Area 1: Photovoltaics

The modernization of the electrical grid is projected to require 20-50%15 of all electricity generation to come from PV sources for a total power level of 1 TWAC. While 15 GW of PV was installed in 2020,16 an increase in rate of 2-5 times is required to achieve 1 TW.

To achieve this sort of year-over-year increase in PV installation economically requires an extreme reduction in the LCOE of PV implementations through the reduction of Day 1 costs for PV installations, reduction in long-term operations and maintenance costs, and extension of lifetime for PV system. The DOE Solar Energy Technology Office (SETO) targets an LCOE of < 0.03 $/kW-hr for utility-scale installations by 2030.17 This requires a PV Balance of System cost of 0.54 to 0.62$/W with a fielded service lifetime of 30-50 years.

Application Area 2: Wind Energy

Wind energy currently supplies more than 10% of U.S. electricity and is now the largest single technology contributor to U.S. renewable electricity after wind generation surpassed hydroelectric generation in 2019.18 While land-based wind generation is predominant, offshore wind development is growing as the ability to generate wind energy near coastal population centers and site plants in areas with excellent capacity factors promises future economical deployment. Significant additional capacity of both land-based and offshore wind is expected to meet clean energy targets for the U.S.

electrical grid by 2035 and beyond.

Application Area 3: Energy Storage

The deployment of energy storage in the U.S. electric grid is experiencing a near-exponential rise, as measured and reported by the U.S. Energy Information Administration and shown in Figure 5.

Figure 1. Growth in Energy Storage Deployments19

Figure 5. Growth in Energy Storage Deployments20

Prior analyses have indicated that the PE used to interconnect batteries to the grid constitute the primary energy loss mechanism in grid-scale battery energy storage systems.21 This conclusion is represented in Figure 6. In the figure referenced, PCR (primary control reserve), SCR (secondary control reserve), and PV-B (storage of surplus photovoltaic power) represent different applications for energy storage that were tested as a part of the published analysis.

Figure 6. Losses in…