DIII-D's role as a national user facility in enabling the commercialization of fusion energy

The United States is developing a strong vision for commercial fusion energy. The fusion community has established a consensus plan for a fusion energy path that has been unanimously endorsed by Department of Energy (DOE's) Fusion Energy Science Advisory Committee (FESAC) (FESAC, 2020). The Biden-Harris Administration has announced “a bold decadal vision to accelerate fusion” (WH, 2022) through commercialization. An agency-wide DOE initiative is planned, with near-term DOE budgetary proposals targeting this path and ITER (FES, 2023).

This consensus plan is based on accelerating science and technology development so that a viable conceptual design for a compact, cost-effective, first-of-a-kind Fusion “Pilot” Plant (FPP) can be developed. This device (or multiple such devices) will demonstrate net electricity and prove the remaining technological solutions sufficiently that the private sector can confidently embark on widespread deployment of commercial fusion reactors to meet the nation's rapidly evolving energy needs. However, reaching this point requires new and better technological solutions than are known now, to deliver the containment, control, heat extraction, and breeding of a fusion plasma, as well as improved plasma confinement solutions that reach the higher levels of normalized and absolute performance necessary in a more compact device.

An important element of this strategy is ITER, which will become operational around the end of this decade. ITER is providing unique data and training on the design, construction, and licensing of fusion reactors. Its scientific exploitation will produce vital insight into the behavior of a reactor-scale plasma, exploring power and particle interactions in opaque, coupled, thermalized regimes, obtaining key tests of behavior to validate trends and projections toward an FPP, and demonstrating the self-heated plasma state.

Confronting these goals poses a formidable challenge, requiring a wide range of research into fusion science, technology, and engineering. A critical aspect is the resolution of a plasma core and power handling solution, and its compatibility with surrounding technologies and control, where despite substantial progress in recent years, the approaches necessary and their integration remain to be proven for the reactor. In particular:

• A compact FPP must achieve higher fusion performance in a smaller volume than ITER. It must therefore operate with substantially higher energy confinement efficiency and pressure, and higher heat fluxes. This must be sustained with components under substantial plasma and neutron fluences, for timespans that are orders of magnitude longer than ITER. This requires solutions that go beyond techniques developed so far—for the core, for the plasma exhaust, and for the various plasma-interacting materials and technologies. These solutions must each be pioneered and understood in reactor-relevant conditions, so they can be projected with confidence to an FPP. The mutual interaction and compatibility between techniques must be addressed as they trade-off against each other. This is referred to as the “Integrated Tokamak Exhaust and Performance” (ITEP) gap in the FESAC plan (FESAC, 2020).

• ITER needs to resolve how to use its considerable toolset to achieve high fusion gain plasmas quickly. The application of systems to navigate the path to required performance and provide stability has yet to be demonstrated in ITER-relevant conditions (low torque and collisionality, high opacity, equal electron and ion temperatures, and radiative mantle). This must be established and explored in present devices, so that solutions can be rapidly realized in ITER and state-of-the-art simulations and approaches and models developed by U.S. scientists can be validated for the FPP.

The DIII-D program seeks to confront these challenges directly, by providing an integrated test facility in reactor relevant conditions. While other facilities around the world target long-pulse operation, particular wall choices, divertor designs, D-T operation, and scaling to lower ρ*, DIII-D brings the capability to explore new technologies and heat flux management with a high-performance core. This will be achieved by expanding DIII-D's unique testbed facilities, flexibility, and diagnostic set to pioneer the new approaches needed, resolving the techniques and their scientific foundations to project them with confidence. This will support the development of new fusion materials and technologies as well as the plasma techniques and interactions common to range a fusion concept, while testing the tokamak approach. The essential argument is that DIII-D has the flexibility to rapidly develop the solutions necessary, can access the relevant physics, and has the measurement and investigative capabilities needed.

To provide the breadth of research necessary, the DIII-D user model is being broadened. As a DOE Office of Science User Facility, DIII-D already provides a proven platform for the community to pioneer the development of new solutions, train researchers, and resolve the scientific basis to project reactors with confidence:

• As a multi-institutionally led facility, DIII-D provides the capacity to support a diverse range of research lines integrated into a goal-oriented approach for one hundred collaborating institutions, with up to 20 weeks of plasma operation supporting ∼100 experiments each year, and analysis of many further phenomena.

• The facility's active collaborative engagement model provides development and leadership opportunities for students, scientists, and engineers from across the United States.

• As a hands-on facility, students and researchers are able to rapidly implement and innovate groundbreaking techniques, install new systems, and even make changes between discharges.

• The comprehensive diagnostic set provides a world-leading platform to develop and validate theoretical models, with many theory groups testing these against key data from DIII-D.

• DIII-D's outstanding engineering team delivers substantial upgrades and capability improvements alongside run time every year, enabling the facility to support its rapidly advancing research agenda.

This approach is being expanded to meet the challenges of the commercialization agenda, with the user model rewritten to engage private sector users and to provide capabilities for prototyping new fusion technologies and concepts. A new Plasma Interacting Technology program has been launched and is outreaching to private sector and national laboratory programs, with a testbed platform approach. New engagements with high-performance supercomputer facilities are underway, combined with the embedding of new codes and control emulators to accelerate validation of theoretical models and control technologies. This is overseen by a new independent User Board as well as an international Program Advisory Committee, comprising of public and private sector members to oversee and guide developments toward these new goals. This work will help DIII-D pioneer technological approaches, serving as a testbed for the United States, and a vehicle to test underlying concepts and theoretical models developed in the U.S. program.

Expansion of workforce development is also being pursued, with a career mentorship program now launched, apprenticeship schemes planned, and training programs underway. This is part of a broader effort to diversify pathways into the field, providing and enabling personnel from all backgrounds through an inclusive approach, with equitable open access to opportunities at the facility, monitoring to avoid subconscious bias, and workforce feedback to identify opportunities for improvement. The goal is to ensure strong engagement of people from all backgrounds and grow the workforce for the U.S. program as a whole.

DIII-D can support the high scientific throughput needed to meet this mission, with high participation, publication, and citations rates of the facility's work, and over 700 users actively engaged each year in research. Investments are being made in facility reliability, protection, and automation to further speed program delivery. With its rapid facility development cycle, operating and upgrading every year, its talented team, and high flexibility, DIII-D is a proven tool to rapidly advance the next generation agenda.

A key step is to build on the formidable flexibilities inherent already in the facility while deploying modest upgrades to close the remaining gaps on reactor-relevant physics regimes. A series of further tool developments through additional shaping, divertor geometry, 3D fields, technology testbeds, and materials flexibilities then enable various aspects of the solution and technologies to be developed. Here, various technology platforms are envisaged, to test reactor control, diagnostics, materials, components, RF techniques, and particle injectors.

An important further step is a heating and current drive upgrade. While present capabilities provide the scope to test limits and principles of individual solutions in the core and the edge, increases in heating power will transform access to reactor-relevant regimes at higher density, where plasmas become opaque and thermalized, with more electron–ion coupling, and low rotation and collisionality. This will resolve compatibility between plasma interacting technologies, power handling solutions, and high-performance cores in reactor-relevant regimes, to project future reactors with confidence.

DIII-D has a proven track record of such delivery. Recent developments include the doubling of electron cyclotron (EC) current drive efficiency through top launch technology (Chen , 2022), the pioneering of helicon ultrahigh harmonic fast wave current drive with an innovative traveling wave antenna (Pinsker , 2023), resolution of safe quench technologies (Shiraki , 2021), and developing advanced high performing plasma scenarios such as super-H mode (Solomon , 2014), quiescent H (QH) mode (Chen , 2020) and Burrell , 2020), and the high q_min advanced tokamak (AT) (Thome, 2022), as well as innovative power handling solutions such as the Small Angle Slot divertor (Guo , 2019) and techniques for core-edge integration and control (Wang , 2021 and Wilks , 2021), including the exciting potential of negative triangularity, yielding high density, current and confinement without Edge Localized Modes (ELMs) or H-mode (Marinoni , 2021 and Thome , 2023). This is reflected in the high number of talks at international meetings such as the IAEA Fusion Energy Conference, where oral selection is based on fusion energy significance.

DIII-D is thus poised to address the fusion commercialization mission, and an ambitious program is proposed. Commensurate with its role as a national user facility, rather than focusing a priori on one solution, the program targets a broad range of concepts and approaches to discover the path. It then seeks to develop and narrow choices as it progresses, guided by simulation and theory toward holistic solutions. A goal-oriented approach is pursued with Technical Readiness Levels (TRL) schemes generalized for plasma research and used to guide and prioritize which research to undertake to retire risks. This will resolve a range of technology issues beyond the tokamak approach. It also addresses fundamentals of plasma science to underpin projection and wider fusion concepts. For the tokamak, both pulsed and steady-state tokamak approaches are explored. Theory is embedded at all stages to guide development and interpretation of new concepts. Where topics are better addressed by other facilities, they are ceded. Where aspects need to be tested in conditions explored elsewhere, a partnership approach is employed.

Thus, DIII-D, as an Office of Science User Facility, provides a vehicle to rapidly and cost-effectively deliver the transformational understanding and solutions needed for the U.S. mission to fusion energy, resolving the conceptual approach, workforce, and solutions necessary.

In this article, we set out the needs and role of a national facility to support these ambitions, and how DIII-D as a user facility can meets this goal (Sec. II). We then look at the technical needs and challenges of an FPP (Sec. III) and of ITER (Sec. IV), before setting out the developments necessary for DIII-D to close research gaps (Sec. V). Finally, we consider the role in national and international contexts, explaining how DIII-D complements and fits alongside with other facilities to deliver insight needed (Secs. VI and VII).

The goal of magnetically confined fusion energy sets challenging requirements on plasma science and technology. A fusion reactor operates at extremes in parameters that place enormous loads on components and demands higher levels of performance from plasmas and technologies than are typically sustained in present devices. It will also operate in physics regimes that are substantially different to those accessed so far in most present devices.

While there are many challenges associated with the technology development, power extraction, breeding, and nuclear handling of a fusion reactor, key aspects arise from the plasma solution and plasma interacting technology. This is illustrated in Fig. 1, where confinement quality and effective power handling capability (divertor heat flux) serve as critical levers for a cost-effective reactor design, and thus, components and plasma solutions that manage the loads and improve the efficiency are crucial. Furthermore, key plasma interacting technologies, such as diagnostics, materials, components, particle injectors, and RF must be tested in relevant plasma environments, both to qualify the technology and its operation, but also to understand the back-reaction on the plasma and the constraints that applies. Core fusion performance (H factor) plays a key role in driving down the challenge and cost.

A key further need arises in finding a compatible solution between the fusion core, the power handling, and surrounding wall and plasma facing technologies. This is the so-called “Integrated Tokamak Exhaust and Performance” (ITEP) gap, coined in the 2020 FESAC report (FESAC, 2020), a solution for which is vital to make challenges placed on reactor technologies tractable. Lack of resolution will increase device risk, cost and/or size, or even make the approach intractable. Developing such solutions provides a relevant basis to test various plasma interacting technologies.

These are the challenges we seek to confront with DIII-D—to pioneer plasma core solutions and power handling approaches that are compatible with each other and the required plasma facing technologies and control tools. This must be undertaken in reactor-relevant physics regimes and levels of challenge, requiring upgrades to the facility. The research plan targets these critical gaps with distinctive contributions in an international context, and deep investigative capability in order to resolve approaches and project them with confidence.

We start by considering the properties necessary of a national research facility to support this agenda. First, it requires deep investigative capabilities and expertise. Development of solutions for fusion reactors needs considerable innovation and discovery to identify the new techniques necessary to resolve viable approaches. These solutions cannot simply be demonstrated in a present device and assumed to work in reactors; they must be tested for integration and an in-depth physics understanding must be established to project them with confidence. Expertise must be developed to develop a working knowledge of how to manipulate the technique, pitfalls, challenges, and interactions in order to adapt them to reactor conditions. Thus, flexibility and diagnosis are key.

A second requirement is programmatic breadth. Each technique deployed requires engagement and understanding from multiple disciplines that govern its implementation and impact on the plasma, in order to establish its viability. For instance, the required high-performance core solutions must be stable, sufficiently energy confining, and avoid exciting waves that can eject fusion born alpha particles. Similar considerations apply in power handling, materials and control, which each integrate multiple research questions and disciplines, and must themselves be made compatible with the core performance solution and surrounding components. Thus, a topically broad approach is an essential feature in the development of required fusion techniques, in order to understand how a solution will behave and project.

Third, it is vital to resolve how different aspects of the solution trade-off against and compromise each other, to resolve their mutual compatibility. Thus, it is necessary that the different elements of core, exhaust, wall, and technologies be explored together, in relevant physics regimes. For example, a dissipative divertor solution rests on achieving high local densities and radiative conditions; the interaction of this with the plasma core through scrape-off layer (SOL) and pedestal must be understood in regimes with suitable short mean free paths and high neutral opacity to correctly capture the dynamic. The interaction with surrounding technologies and materials is key, as there is a load on these technologies and back-reaction on the core. With various facilities around the world focused on addressing key issues such as long-pulse operation, interaction with metal walls, and D-T performance, a gap emerges for a facility that has the flexibility to explore this integration. This is a critical role that DIII-D aims to fulfill, to explore a range of plasma and technology solutions, learn how to marry them together.

Thus, to develop projectable solutions, and the depth of expertise necessary, places significant requirements on investigative capability, community engagement, model validation, and programmatic breadth. This is only possible with a high capacity and well-engaged national user facility such as DIII-D, which can support the volume of users, run time, and research lines necessary. We summarize these requirements on three fronts:

• Technical

  • To be able to rapidly implement new techniques to explore candidate solutions or critical measurements, to meet the FPP/ITER timeframes.

  • To have high flexibility to pioneer the new solutions necessary.

  • To access reactor-relevant physics regimes.

  • To be able to test how elements of the solutions trade-off against each other.

  • To provide a broad investigative capability across the multiple research fields that are required to underpin the development and projection of viable solutions.

• Capacity

  • To support a large user group from multiple perspectives, engaging the requisite breadth of expertise and innovative perspectives, including public and private sector.

  • To provide substantial run time to support the multipronged research lines necessary to resolve questions across disciplines.

  • To deploy multiple high precision diagnostics to characterize behavior, to resolve theory models for predictive capability.

• Engagement

  • To embed theory and simulation to close out critical physics questions, validates models, and is guided to viable solutions.

  • To develop and enable early career researchers and engineers from a diversity of perspectives, so they can implement their understanding in future research and reactors.

  • To consider the gamut of techniques, and not a priori favor one approach over another, until the data independently discover the viable path(s).

These are demanding but vitally necessary requirements to be placed on a national facility if the United States is to be able to develop and validate the required techniques to succeed in its fusion mission. The DIII-D program has a unique track record at operating at these levels, and thus has unique potential to meet these broader requirements.

The DIII-D program meets U.S. plasma research needs with a highly integrated multi-disciplinary and multi-institutional team of users, working on site and remotely on a highly flexible facility. The device presently delivers up to 20 run weeks of research (800 run hours) each year supporting around 100 experiments. Each of these typically incorporates more than one research proposal and engages multiple scientists with a number of scans, tests, and analysis approaches. With over 800 participants, of which over 650 users were audited as directly contributing to research products in 2022, and over 200 connected students, postdocs, and interns, the device serves as a facility for the entire U.S. magnetic confinement fusion research community, with a program that is jointly led by scientists from the host, national laboratories, and leading universities in roughly equal proportions. The facility thus plays a strong role in developing the next generation of fusion scientists, with many examples of students, postdoctoral researchers, and early career scientists leading studies and securing high profile invited talks and journal papers in their training.

DIII-D provides a strong research capability that is in high demand by the fusion community. Run time is open to the whole fusion research community, with community-wide solicitations generating hundreds of proposals for run time (623 for the 2022–2023 campaign). A goal-oriented approach is defined in consultation with users. The experimental program is formulated with critical physics and technology explorations worked into studies that pioneer new techniques and improved solutions through heavily participative team planning activities. Multiple theory programs and projects are interfaced to the facility (26 theory groups, 72 codes), with advanced simulation tools embedded even into control room analysis, and used to plan, predict, and interpret experiments, as well as guide the facility development. The facility undertakes careful monitoring of its scientific opportunities to ensure balanced representation across user groups, and, for instance, has been at the forefront of fostering opportunities for gender minorities.

The facility provides an unparalleled and collaborative scientific tool, with enormous capability to support research. In addition to delivering run time each year, it also undergoes regular extensive openings in which multiple new systems are implemented to drive research forward. As a result of this, and the foresight in its original design, the facility has developed exceptional flexibility as summarized in Table I. DIII-D's basic vessel geometry and its flexible 18 poloidal field coil set enable it to dial up shapes ranging from single null ITER-like configurations to double null, with variable elongation, and low, high, or even negative triangularity. This also permits the facility to draw out and vary the flux expansion for the divertor in a range of ways to study configurations that may isolate or contain a dissipative “detachment” zone.

This is complemented by heating and current drive systems that provide ample power to reach high normalized pressure (β_N values up to 5 achieved) with peaked or broad current profiles, to pioneer plasma operating scenarios suitable for continuous “steady-state” operation in an FPP, or higher current inductive scenarios for ITER or a pulsed FPP. The key is flexibility, with neutral beams configured to provide bulk heating, current and/or torque, either on or off axis, and with full co-injection, or 50% balanced torque power. Electron cyclotron systems provide additional directable torque-free electron heating or current drive, with precisely tunable deposition for profile and transport investigations, and stability control. A new helicon ultrahigh harmonic fast wave technique with innovative traveling wave antenna was just pioneered and is delivering substantial power. A high field side lower hybrid current drive system is also being installed. Both offer the prospect of bulk on or off-axis current drive with high efficiency at higher plasma densities.

The facility also has extremely good particle control, thanks to its cryopumps, high temperature bake system, and regular deposition of boron layers (“boronization”). A range of gases can be introduced into different regions of the plasma through multiple valves, as well as via fueling pellets and granule injectors from various locations, which can further modify machine conditions. At higher throughputs, particle injectors serve as quench systems to explore innovative methods to safely terminate fusion plasmas. Three arrays of six magnetic perturbation coils provide means to control a range of instabilities in the plasma, and to tune and study “error” fields and plasma rotation through resonant braking and neoclassical toroidal viscosity (NTV) effects, which also influence density through particle pump out phenomena.

With the engagement of its many user groups, the facility has also developed a world-leading, comprehensive, state-of-the-art diagnostic set (Fig. 2). Systems measure nearly every plasma property conceivable (though some key gaps remain). Measurements include single parameter, line-averaged, 2D, and even 3D diagnosis of key characteristics, including temperature, density, current, field, impurities, radiative power, velocity, and energetic particles with techniques ranging from spectroscopy, infrared, visible, UV, EUV, soft and hard x-ray detector arrays and imaging, exploring both natural emissions and also reflection of light and interactions of particles in the plasma. These capabilities are vital to enabling the facility to not only quantify many main equilibrium parameters, including crucial aspects such as current profile, but also explore dynamic events in high time and spatial resolution, such as fluctuations, turbulence, instabilities, fast ion redistribution, and rapid termination or reconnection events. This provides unparalleled capability in the world program to explore the science of plasma interaction processes to understand and guide the development of solutions—a key reason why the facility has been so effective in generating research advances.

Thus, DIII-D provides a strong platform to develop robust, controlled, and precisely defined plasma scenarios in which discovery of new techniques and scientific investigation go hand in hand. The investigative range is large, with the facility able to explore basic foundational science, reactor scenarios, and a range of innovative techniques where the main plasma shape, divertor, and internal magnetic and kinetic configurations are widely controllable. This provides the ability to pioneer a wide gamut of physics regimes and investigations, with approaches ranging from inductive to steady-state, broad to peaked profiles, low to high β, high to low q₉₅, externally driven to self-driven, and positive to negative triangularity. DIII-D is small enough to innovate rapidly, while being large enough and powerful enough to access the key physics regimes of a reactor. It also provides extensive hands-on access that will not be available in future devices.

These capabilities have enabled DIII-D to confront more research issues to address the breadth of topics demanded by a national program, and it now provides the principle experimental facility to support magnetic confinement research in the United States. It also supports many international teams and collaborations, including ITER. In recent years, users of the facility have also made key discoveries in foundational plasma science, from magnetosphere, solar and astrophysics perspectives, and even space craft reentry, informing processes common to fusion and general plasma science such as reconnections, particle–wave interactions, erosion, and fundamental turbulence.

The move to the reactor scale poses new challenges to develop effective solutions that can be readily implemented in ITER and FPP. The nature of this challenge cannot be met by any one facility (unless that facility were the reactor itself). It thus requires a multifaceted approach encompassing facilities, theory programs, and non-confinement capabilities across nations. Thus, the DIII-D program does not seek to address all aspects of the problem, but instead targets the gap questions that leverage its strengths to make unique and needed contributions in an international context.

With DIII-D's high level of flexibility, the key role of the facility is to innovate to identify potential solutions and investigate them to understand their limits and projection. The model is that DIII-D uses its rapid innovation cycle and investigative capabilities to scope solutions, while other high parameter facilities, which may be less flexible or diagnosable, offer key tests of trends in physics to validate the scaling of projected solutions from DIII-D. The approach is illustrated in Fig. 3. For some phenomena, DIII-D can already access reactor-relevant regimes right up to potential power plant parameters (shown in the blue oval). Examples include high β_N access with broad current profiles for steady-state, or low safety factor, rotation, and collisionality for ITER-like regimes. In other cases (green and cyan ovals), there will need to be some level of projection which requires not only innovation to develop the required solution but also development of scientific understanding to identify the correct governing parameters and validate models to project behavior. An example here is turbulence broadening in the scrape off layer, governed by parallel heat flux; DIII-D can presently access this phenomenon at a minimal level, while proposed upgrades will extend this capability to enable testing of divertor solutions and exploration of behavior at a much wider range of parameters and with different techniques tested.

Inherent in this approach is strong collaboration with theory and simulation to guide the development of the innovative solutions needed and to provide validated models to project solutions to the reactor. It also requires relevant diagnostic measurements to obtain rigorous validation of those models. This motivates some of the breadth of engagements that DIII-D employs with multiple institutions leading innovative systems to provide those measurements and codes with sufficient scope, resolution, and range. Ongoing strong collaboration with partner facilities in the United States and around the world is also key.

Thus, rather than determining a priori a solution simply to be validated (which often can lead to unpleasant surprises in research), the DIII-D approach seeks to discover the solutions to meet these goals. DIII-D has the range and scope to vary approaches to find the viable path. For ITER and FPP, a spectrum of scenarios is being explored from the baseline to elevated safety factor and β for ITER, and from non-inductive “steady-states” with high bootstrap to efficiently driven inductive or pulsed solutions for FPP. Thus, any upgrades to the facility are posed not only to provide access to relevant regimes but to also flexibility to innovate and pioneer the solutions.

A key point is that it is important to take data in the relevant physics regime, the right side of “physics phase transitions” that affect behavior and projection (the vertical dashed line in Fig. 3). Such phase transitions can be seen as the plasma becomes collisionless, or opaque, or low torque-shear, or thermalized, or when electrons couple strongly to ions. In some cases, modest upgrades are necessary for DIII-D to close remaining gaps in physics regimes (blue arrow). Principally these gaps arise from the need for higher density operation in DIII-D to drive more coupling between species, thermalization, opacity, and radiation. This requires more RF power to maintain reactor plasma configurations, and thus, a major RF upgrade is a key theme in the facility development proposed in this plan (see Sec. V).

Examples of the parameter access enabled by these developments are provided in Fig. 4. The target FPP design shown here is that of the Compact AT (“CAT”) concept (Buttery , 2021), with parameter values and simulations for DIII-D and CAT described in the  Appendixes of this paper. For the core (upper panel), modest upgrades enable a full range of reactor-relevant behaviors to be explored (blue cases), though not always simultaneously (two integrated examples are shown in green overlays). For the exhaust solution (lower panels), the DIII-D divertor already accesses many of the physics parameters of a pilot plant (left panel, light green)—though not all; upgrades (dark green) will deepen access and multiply the range in dimensional variables (right panel) to test scalings toward the reactor, and underlying physics assumptions behind those scalings.

Thus, with modest performance upgrades, DIII-D will provide the scope to develop reactor solutions, in relevant regimes, and to resolve the scientific basis sufficiently to project them to reactor parameters with confidence, when taken in concert with other funded facilities around the world (see Sec. V).

Finally, collaboration plays a key role. International facilities bring many different strengths. DIII-D engages with many of these, so that the scientific insight and innovation developed at DIII-D can be tested in these broader contexts, and vice versa. Examples include long-pulse operation at superconducting facilities in Asia, reoptimizing scenarios for walls with reactor-like materials, most notably on JET and ASDEX Upgrade, exploring innovative divertor configurations (MAST-Upgrade, Tokamak à Configuration Variable (TCV)), exploiting non-confinement materials testbeds, and D-T operation in JET, while looking to the future, the JT-60SA device offers important potential to assess the scaling of advanced tokamak regimes.

Nevertheless, critical questions remain for DIII-D. For example, whether an FPP should be steady-state or pulsed? How to achieve sufficient confinement and stability? How to make a high-performance core compatible with an exhaust solution and device walls? How to integrate required control techniques for transient events? To pioneer the development and plasma compatibility of required plasma interacting technology solutions for issues such as disruption mitigation, reactor heating and current drive, and new materials with fusion plasmas. These set the themes of the DIII-D approach in Secs. III–V.

To identify the critical steps needed, a technology or technical readiness scale has been implemented to define current status and draw clear paths for FPP development. This is based on adapting established technology readiness approaches for the situation of plasma physics research, where the requirements also extend to the development of predictive understanding of plasma behaviors and configurations. The concept of Technology Readiness Levels (TRLs) was developed at NASA during the 1970s. TRLs estimate the maturity of technologies under development, and each stage of TRL is determined by a Technology Readiness Assessment (TRA) that examines concepts, requirements, and demonstrated technology capabilities. TRLs scale from 1 to 9 with 9 being the most mature and demonstrated technology (NASA, 2019). The European Association of Research and Technology Organizations (EARTO) published a comprehensive approach and discussion of TRLs (EARTO, 2014).

The TRA for fusion technology is an essential part of the preparation of a fusion power plant. The fusion community started to adopt the concept of TRLs for fusion technology development and applied it to the Advanced Reactor Innovation and Evaluation Studies (ARIES) project in 2010, for instance (Tillack , 2010). A TRL comparison for helical and tokamak DEMOs was made (Sagara , 2015). Identifying the TRL of each fusion technology is important as the focus moves from physics to engineering, and the importance of advancing low TRL for closing of key technology gaps toward fusion reactors has been recognized in the United States (FESAC, 2018 and FESAC, 2020). However, work also needs to deal with the realization and maturation of physics concepts within the plasma.

This TRL approach has been adapted for applicability to plasma physics research based on work in the Korean DEMO program (Ghim, 2019), where the Korean fusion community assessed levels of physics readiness in plasma techniques, as well as in technology gaps, for realization of fusion. Following this assessment, plasma research programs to advance approaches with low TRLs were identified and selected. This led to a plan for a Korean DEMO program submitted to the Korea ministry of science and technology. As this approach led to key insight on the applicability to plasma physics research, the governing metrics, and other requirements, it has been further adapted to provide a generalized framework for DIII-D. The key point is that this is used to drive decision making. The approach is presented in Table II. It draws on underlying TRL methodology where the needs for developing a projectable solution for an FPP are:

1. To demonstrate the relevant solution or technique.

2. Demonstrate it in relevant physics regimes.

3. Resolve a validated scientific basis to project it with confidence.

This starts by identifying problems and potential for an approach, before proceeding with proofs of concept and simple models of process, then advancing toward reactor-relevant physics regimes and confronting the full challenge in a relevant way. The goal for present devices would ideally be to reach a TRL of 7; lower levels may be acceptable but represent risks that will need management and mitigation. TRLs of 8 and 9 represent application and robust operation in the reactor itself. The red path of Table II indicates key elements ideally targeted by the DIII-D program.

On this basis, TRLs of DIII-D research program elements have been analyzed and strategies to advance the technologies and physics solution with low TRLs devised. While this has strong parallels with traditional TRL approaches, ambiguities emerge between qualitatively different activities such as plasma configuration development vs proving hardware technologies. Indeed, it is quite easy to get into not entirely helpful technical debates about exactly where a particular technique sits on a TRL scale. However, the point is to provide a focus to drive research decisions and investments to close out critical questions for an FPP. For the topics DIII-D engages on, this typically means hitting values above 5 and up to 7, dependent on international capabilities and complementarities that collectively get close to 7 overall. This will retire risks for future reactors, enabling confident projections and validated, viable conceptual solutions so that decisions can be taken on FPP approach and design.

In developing the research approach, it is important to develop skillsets for national efforts toward an FPP and ITER, and to address historical gaps in access to national programs such as DIII-D. The user model needs not only to draw in personnel with the relevant technical capabilities, but also provide training and opportunities for development. It has been recently recognized that an FPP, and likely major facilities leading up to it, will have to be built in the private sector, but with public sector assistance. Thus, the approach also needs to interface with an even wider cross section of the community of universities, national laboratories, and private industry.

Considering the range of technology research needed to get to an FPP, a significant portion of which must interface to the plasma solution, it is clear that an integrated approach is needed. Thus, the DIII-D User Facility model is being extended to enable proprietary and nonproprietary private sector engagements, and key aspects of the program are being posed as flexible testbeds, where new ideas or key components may flow into the program for testing in the plasma environment.

A vital aspect of enabling a broader national team lies in facilitating workforce development. The most critical element in the new national context is enabling diversity to achieve equity, inclusion, and access, which remains a major issue for the field nationally, with underserved communities still seeing a dearth of opportunities in the United States. The approach necessary involves entry points, but also must address opportunity for those within the program. A foundational element lies in providing an environment to foster inclusivity. As set out in the framework by Young-McLear et al., this must be founded on psychological safety, moral courage, and cultural competence (Fig. 5) in order to ensure a workforce engagement that is enabling to all, and where challenges can be called out and acted upon where needed.

The DIII-D program is taking proactive steps to address these issues. An independent DEI user group has identified many issues and opportunities for improvement. The program also commissioned an American Physical Society Climate Review with a panel of leading national experts visiting the facility in August 2022 to identify opportunities for improvement. A focus on education about, and awareness of, workplace environment is being propagated throughout the team (for example, with community agreements developed, Fig. 6). The facility is also embarking on a program to facilitate access through increased training, monitoring, and opportunity development as well as double-anonymized deconflicted peer review for experiment proposals to ensure they are assessed on quality rather than name. Expansion of engagement of under-served communities and institutions is being pursued, with local colleges and schools. The program management would nevertheless recognize this is a journey in which not all the tools and approaches have yet been satisfactorily defined, and further initiatives need to be implemented.

The program also naturally creates a range of opportunities for early career scientists to develop high profile scientific roles (keynote studies, invited talks, journal papers) and leadership with projects and scientific field leaderships. The facility is building on this with specific foci on students and postdoctoral researchers, with dedicated run time set aside for PhD theses, alongside an internship program now hosting around 20 students a year, and an education team formed for pastoral support, tracking, and knowledge development via regular student/postdoc-only mentored seminar events. A career mentorship scheme is being piloted for early career staff, open to all collaborative partners on site, to help provide pathways, professional development, and knowledge expansion during this next five-year period. This is extending to the technical and engineering sides where the facility provides opportunities to learn a breadth of disciplines across collaborating institutions.

Finally, the facility has a long-standing tradition of strong theory engagement, benefiting from the direct on-site participation of theory groups, extending to 26 groups nationally and internationally. New tools and approaches are deepening such collaboration, with the “predict first” initiative now encouraging analysis to precisely predict and design experiments before execution. Furthermore, a range of computational tools are implemented that provide readily available interfaces of state-of-the-art tools to experimental data and even control room analysis. This is augmented by access to high-performance computation amongst collaborative partners, with ESnet access now upgraded to higher speeds to permit more intershot and potentially real time computation, and big data analysis with U.S. supercomputer facilities. Artificial intelligence and machine learning techniques are also being extended to operational approaches and to provide rapid analysis, based on offline major computation. Synthetic diagnostics are an increasing feature of code and diagnostic collaborations to effectively test models. These techniques integrate well with DIII-D's focus on scientific understanding and innovation in its experiments.

The facility also engages multiple international teams to lever its research. The ITER team are deeply involved in planning and prioritizing for research. Collaborations with the Korean and Chinese superconducting facilities, KSTAR and EAST, have been used to understand how plasma scenarios developed on DIII-D can be extended to long-pulse. Collaboration with other key facilities is testing critical underlying physics such as pedestal, core transport, and disruption mitigation on JET, divertor science on MAST Upgrade, and energetic particle and 3D physics on ASDEX Upgrade. These foci will continue, with a goal to extend further on plasma wall interaction issues with U.S. domestic testbed facilities and internationally on WEST in France.

These approaches and techniques are open to, and will naturally extend to, the new public private partnership agenda, with commercial companies able to sign up easily and have their team members gain access to the DIII-D team, meetings and all internal data. Intellectual property of government funded research is protected in this model by a new user agreement agreed with DOE, which requires institutions engaged in “non-proprietary usage” of the facility to commit to ensuring all DIII-D-derived data to remain open and in the public domain, and that it cannot form the basis of patents by the company. Under this approach, engagement is based on an alignment between the company's goals and DIII-D DOE approved goals, and thus, the company's users engage in DIII-D and are supported (and trained) in the usual way of all DIII-D users to collective discuss plans and progress work. Nevertheless, private companies are able to bring proprietary technologies to DIII-D for testing—the program is not obligated to disclose or learn about these, and the IP ownership of the particular technology stays with the company that brought it. Furthermore, if companies do wish to own DIII-D-derived data from such an engagement, a “proprietary usage” of the facility is possible in which all costs of the facility usage will be recovered from the private company. However, it is anticipated that the majority of private company engagement needs are likely to be through the nonproprietary approach. Furthermore, in the national user facility model of the United States, the facility program and operational teams will support such engagements and users—not least because the goal of the government funded program is to support and enable the commonly determined research agenda. A service philosophy is inherent amongst the DIII-D program as all research is the product of team efforts involving multiple personnel from multiple institutions providing mutual support to each other's goals, based on the many different elements and expertise they each bring. This is the only way by which research in such a complex facility can be performed. The facility is moving to increase its training and information systems, so that new users can more rapidly onboard and gain familiarity with the techniques they need to learn, as pioneered for Frontiers science engagement successfully. Finally, fully remote, and now hybrid remote scientific operation has been developed at the facility, further enabling remote researchers to directly engage more quickly and cheaply.

These elements provide the building blocks of a highly versatile and available user facility.

We now turn to the specific research challenges for the U.S. path, analyzing needs for FPP. Some of these aspects are specific to tokamak or toroidal confinement concepts, which represents the majority of the private sector by dollar investment, and much of the public sector. Others generalize to wider fusion concepts, particularly on the technology side. With approaches in the private sector relatively narrowly focused, broad investigative capabilities are needed to guide and solve problems for the FPP, both to provide critically needed foundational data and insight on the techniques needed, and to adapt for surprises that may emerge in the private or public paths. We start with the tokamak aspects.

The U.S. goal of a low capital cost FPP places exacting challenges on the plasma physics and plasma interacting technologies to sustain the performance required and handle the demanding loads that result at the compact scale. Traditional fusion power plant concepts already pose scientific and technological challenges that go well beyond the techniques developed for ITER. As well as technology and plant challenges associated with the heavily nuclear environment, breeding, and power extraction, the device must sustain high levels of performance and efficiency for net electricity generation, while avoiding degradation of components from heat, particle, and neutron fluxes over extended periods.

A range of tokamak power plant concepts have been proposed (e.g., Table III) based on varying levels of optimism about what can be achieved in terms energy confinement, heating, current drive, and thermodynamic efficiencies, and plasma performance, plasma exhaust mitigation, and materials approaches, as well as assumptions about what is needed in terms of net electric power and sustainment of the device. The low capital cost approach advocated in the United States seeks to a make a smaller and lower cost device than the GWe level demonstration devices proposed elsewhere. It aims to prove the principle of a power plant and close remaining research gaps, to enable the subsequent commercialization of fusion energy.

The essence of the U.S. approach is to pursue scientific and technological advances in the near term to enable a compact “Fusion Pilot Plant” (FPP) reactor concept to be developed in the medium term. This will be easier and faster to resource than the traditional GWe scale demonstration approach, and leads to better long-term prospects, as answers come sooner and at lower cost. Nevertheless, the compact scale makes some aspects of the challenge harder, and a number of risks must be retired and new solutions developed if a viable overall concept is to be identified.

An FPP must attain sufficient core confinement. Without this, the plasma will not sustain its heat, and auxiliary heating systems will need to consume more of the electricity produced just to keep the plasma fusing. Fundamentally, a compact FPP has less volume to fuse in than a conventional demonstration power plant such as EU-DEMO or ITER. It must therefore fuse at higher power density and thus pressure. The smaller volume will also tend to lose heat more rapidly. This requires means to sustain higher levels of energy confinement time (the ratio of plasma energy over heating power). To achieve this requires either to increase the plasma current (the primary driver of tokamak confinement, but also of tokamak instabilities), or to improve the insulating properties of the plasma through its geometric and internal magnetic structure in order to operate at high pressure relative to current (β_P) and make a more efficient magnetic confinement configuration. [An example of improving magnetic confinement efficiency is the routinely utilized “H mode” operating regime, discovered on ASDEX (Wagner , 1984), where shears in the plasma magnetic structure and flow lead to a transport “barrier” at its edge. Other examples include transport improvements originating from shaping, broad current profiles or configurations with high pressure normalized to magnetic field (Taylor , 1994; Kinsey , 2006); and Staebler, 2018).]

This leads to a divergence of approaches. High current would require high levels of power-hungry radio frequency power if it were to be sustained non-inductively, and so optimizes instead to an inductively driven pulsed configuration. In contrast, in the high-pressure approach, collisional effects at high density and temperature gradients translate net poloidal currents into a toroidal “bootstrap” current [Galeev and Sagdeev (1968)] that can drive most or all of the required plasma current. This then needs only modest radio frequency power to sustain the plasma non-inductively in a “steady-state” configuration. However, modifications to geometry and internal plasma structure are needed to raise the energy confinement and stability to sustain high plasma temperatures at the high pressures needed to arrive at a self-consistent solution. The issues for these two approaches are thus somewhat different:

The steady-state approach is favored in many reactor studies (e.g., Table III), because it is stationary, and thus stable, and because it has less current and avoids thermal cycling, thereby lowering stresses and heat loads, and permitting leaner builds. The steady-state approach is based on a confluence of strong shaping, broad current and pressure profiles, and high β (pressure normalized to field) operation—although other more peaked profile configurations may also be possible. This leads to a virtuous cycle of improved transport and increased stability. The resulting high pressure drives bootstrap currents that provide most or all of the current, and are naturally broadly distributed, and so aligned to the distribution that is favorable for stability and transport—a self-consistent solution.

Self-consistent steady state solutions have been predicted by theory-based simulations for compact FPPs (Buttery , 2021 and Holland , 2023). Many of the mechanisms predicted behind this approach have been not only explained theoretically but also validated experimentally [see Buttery (2019), and many references therein]. For example, broad current and pressure profiles, high normalized pressure, and shaping have been demonstrated to improve transport (Staebler, 2018), reduce energetic ion driven Alfvénic instabilities (Kramer, 2017 and Collins , 2021), and raise stability limits (Holcomb , 2014 and Hanson , 2017).

However, full development and validations in reactor-relevant physics regimes remains the challenge: to prove that a suitable self-consistent confinement solution with sufficient energy confinement can exist, stably, and meet performance and confinement goals. A key need is to prove models and test proposed solutions with sufficiently broad current profile at high β. Also important is to test behavior in reactor-like physics regimes, which for DIII-D means operating at higher densities, to access regimes that are more thermalized, with lower rotation, coupled species, opacity to neutrals, and higher bootstrap currents. An important aspect of this is to assess and validate models of Resistive Wall Mode (RWM) stability, their potential for dissipation by kinetic damping, and what 3D field control techniques may be needed. These scenarios will require increased current drive on DIII-D via increases in RF capability, now commenced, which will equip the facility with unique capability to resolve the steady-state path.

The pulsed (inductive) approach is more challenging to operate due to virulent tearing resonances that are typically removed in steady-state configurations. These are more pervasive at high current because of lower order resonances that are more readily excited. However, higher current provides a direct way to raise confinement, and for this reason, it was selected for the high performance Q = 10 scenario in ITER. For a pilot plant, the challenges are greater, as the facility must operate without significant wear for much longer periods, requiring elimination of unmitigated plasma termination events and other transients. The higher current and thermal cycling inherent in the approach will also increase device stresses and heat loads, potentially requiring a stronger and thus larger device.

These issues are being extensively studied for ITER, focused on how ITER can use its flexible tool set to reach and maintain satisfactory behavior. As with the steady-state approach, many of the key techniques have been developed, and their scientific foundations are becoming increasing well understood, such as neoclassical tearing mode (NTM) control (Zohm, 1999 and Lahaye , 2002), ELM control (Evans , 2008), disruption mitigation (Shiraki , 2021), and the basic scenario design (Sips , 2018), while more recently, JET has shown how to adapt D-T scenarios for metal walls (in preparation).

The challenge for an FPP lies both in integrating the above techniques and demonstrating them at the levels of performance required. Tearing modes are a particular concern, because their thresholds fall with ρ^* (La Haye , 2000), as well as rotation and q₉₅ at low collisionality (Buttery, 2015). Radiative heat and ELM mitigation must also be addressed, to develop scenario solutions for FPP. For DIII-D, a proposed ECH upgrade (Sec. V) plays a critical role, to provide reactor-like electron heating and access to low torque and collisionality regimes, and higher density to resolve solutions with thermalized, coupled and opaque plasmas.

In an FPP, heat and particle exhaust that are typical of present device operating regimes, will likely lead to erosion, and physically damage plasma facing structures over the long periods of operation needed. Techniques to mitigate this potential outcome, such as extensive use of radiators in the plasma edge regions, will likely quench or degrade the core. Solutions are thus needed that go beyond the techniques and levels planned for ITER. Specifically, a pilot plant must undertake high duty cycle operation for years, and this means erosion or damage, even at very low rates, must be virtually eliminated. The challenge is further increased by the compact scale of the device that could lead to very narrow heat fluxes in the divertor power handling region.

More innovative geometric and magnetic configurations must therefore be developed to contain a cold dense radiative region in the divertor. This must also be resolved for more reactor-relevant regimes than are typically accessible now, with high heat fluxes and short particle mean free paths compared to distance scales. The latter is crucial, as it governs both the dynamics of the processes within the divertor region itself, and its interaction with the fusion core. Planned power rises on DIII-D will support plasmas at the higher densities, opacities, and heat fluxes needed, with short mean free paths and neutral penetration depths. A shaping, volume, and current rise is projected to expand density and pressure limits to reach these conditions, while a stage divertor program will explore the physical and magnetic configurations necessary to contain detachment, and resolve the physics basis to project solutions (see Sec. V).

Another promising approach is negative triangularity operation, which is observed to sustain good energy confinement, but without the transition to H mode. This eliminates ELMs and the need for power into the separatrix usually required to sustain H-mode. It therefore has the potential to greatly decrease the particle and heat flux challenge. A campaign to explore in DIII-D (Thome , 2023) has validated key aspects of potential, with high confinement, density, current, and β solutions obtained, but testing with high power pumped divertors and exploring core profile configurations remain important.

Transient events pose a serious concern for tokamak based FPPs, and likely other concept. These typically arise from plasma instabilities or control failures and lead to termination of the plasma and/or substantial losses of heat and particles that can damage or progressively erode plasma facing components.

The most serious concern comes from violent, abrupt plasma termination events, termed “disruptions.” Here, macroscopic magnetohydrodynamic (MHD) instabilities grow and control is lost, leading to large scale interaction of the plasma with the wall of the tokamaks, and thus rapid impurity influx and quenching of the core. This event itself can lead to localized thermal loads and large induced forces on surrounding components. Even if the plasma is successfully quenched, flux conservation can drive the formation of an energetic electron tail and consequent “runaway” electron beams.

The first line of defense is to avoid disruptions by anticipating and controlling incipient events in real time, to either remain in stable operation windows, or to actively manage back to a safe state. The predominant concern for tokamaks comes from tearing instabilities that split low order rational surfaces with large growing magnetic islands that drive macroscopic distortions of the plasma. These are more prevalent in pulsed and peaked current profile concepts for an FPP but remains a concern close to ideal MHD limits for steady-state regimes as well. ECH upgrades at DIII-D will enable it to tailor the current profile to optimize tearing stability, or locally drive current to stabilize such modes.

A further concern comes from ideal MHD modes. Though more predictable and detectable, it is vital to understand how plasma configurations drive these instabilities and how to control the configuration or act on the modes directly to manage them. DIII-D has unique flexibility to confront this challenge over the range of current and pressure profiles expected for a reactor, to develop a basis for stability projection and real time control.

Nevertheless, fallback techniques to mitigate disruptions are vital. Here, DIII-D has been at the forefront of the field, resolving the complex physics and of this non-linear multi-scale event, and pioneering safe quenching technologies, such as shattered pellet injection (SPI), now adopted in ITER. However, techniques need to go further for an FPP to mitigate the heat, forces, and runaway electrons comprehensively. Robust solutions require quenching from the inside out, and thus need to get particles right to the core of the plasma in massive quantities, with faster injection. It may help to use or drive MHD (with perturbative coils) to ergodize the plasma and thus dissipate the energies evenly. These investigations are planned for DIII-D in the near term, which, thanks to its forgiving carbon wall and excellent diagnostics, is well placed to meet this challenge.

Finally, the rapid bursts of heat and particles arising from ELMs, must be eliminated or strongly mitigated. For an FPP, these are a particular concern as non-linear modeling with the JOREK code (Huijsmans , 2019) indicates even small ELMs will punch through a dissipative divertor solution and erode the device wall. They can also trigger other instabilities such as tearing modes. Techniques are challenged in reactor regimes, particularly as low rotation, q₉₅, and collisionality are approached and tearing modes emerge. Approaches focus on use of perturbative 3D fields and manipulation of the plasma edge region to changes its stability (e.g., QH mode, grassy ELM regimes, I mode). It is critical to resolve these techniques in reactor regimes—on the peeling branch of the ELM instability at low collisionality, and at low torque and high opacity which govern these dynamics. Here, upgrades to ECH on DIII-D will play a key role.

Innovations are required in plasma interacting technologies to provide the required measurement, control, heating, and structural integrity in the harsh plasma environment of an FPP. These technologies themselves have demanding roles to heat, contain, measure, and control the plasma that already require considerable innovation to reach the required performance level. Development programs for these often go beyond plasma confinement research, although testing the interaction of such technologies with the plasma is a vital step. However, exposure to the plasma environment of a fusion reactor places further challenges on the design and survivability of the technologies, and crucially, as high impact on the core solution. This drives three key goals:

1. To benchmark approaches directly in the plasma environment.

2. To assess the plasma interaction physics of that technology.

3. To find an integrated solution with the plasma core.

DIII-D is offered as a testbed facility to meet these challenges, with a test platform approach across a range of fields, to serve the tokamak, and in many cases, the wider fusion community on the plasma interaction of a range of essential fusion technologies, with rapid implementation and testing of new techniques possible. The programmatic approach is not only to anticipate some well-known technology development needs (where the program may already be active), but to seek out new challenges and see the need emerging from the public and private sector, as FPP plans develop.

The most critical aspect comes from the need for a viable materials solution that can withstand plasma exposure and whose backreaction on the plasma still enables high-performance. Even with radiative and divertor mitigation techniques discussed in challenge 2, the remaining challenges from heat flux, particle energies, and neutrons will be formidable, requiring innovation in plasma facing components and materials. Materials and component testbeds in the reactor environment, under relevant fluxes and physics regimes, are thus vital. This forms a major plank of the DIII-D program already, with sample exposure facilities, refined control, targeted diagnostics, and a carbon wall that permits perturbative studies of reactor materials. However, larger scale structures must be tested, and net erosion effects explored, and thus, a new “Wall Interaction Test Station” (WITS) component test facility is proposed. Furthermore, it is important to understand how these issues interact with the design of a reactor wall itself, and thus toroidally symmetrized limited are proposed, to assess the erosion, migration, transport, and redeposition pathways about and through the plasma, and the wall design itself.

An FPP also needs robust heating and/or current drive schemes to start up and, in most cases, maintain the plasma through the discharge. Critical physics challenges include coupling these to the plasma, antenna/mirror survivability, and core deposition of heat or current drive. Technology challenges are also posed to develop and validate systems at the parameters and specifications needed for a reactor—for example, higher frequency ECH must be invented, and thus need test sockets and associated infrastructure. This requires access to key plasma regimes and conditions that are essential to test these techniques (for example, high β_e plasmas for current drive schemes). This flexible antenna and RF techniques are key, with the ability to explore and vary novel launching techniques such as the traveling wave antenna recently pioneered for helicon ultrahigh harmonic fast wave on DIII-D (Pinsker , 2023), or the high field side launch for lower hybrid current drive, now being installed for 2024 testing.

Furthermore, development of FPP compatible diagnostic and control techniques poses a key challenge. The harsh nuclear environment eliminates many in vessel and direct line of sight diagnostic techniques, requiring new approaches to be developed and proven against existing techniques, as well as tested for survivability in the plasma environment. Here, a dedicated diagnostic development platform is vitally needed to streamline capabilities for rapid installation and testing, and this is an ideal role for DIII-D, thanks to its high flexibility and plethora of existing measurement systems and relevant physics regimes. Similarly, dedicated control infrastructure is needed to enable the development of reactor control schemes, and test them in real world plasmas under reactor-like limitations and constraints. DIII-D's analog + digital twin is already being used to pioneer new schemes. Combined with other developments and proposed improvements, this will enable DIII-D to mimic reactor constraints and modes of behavior to resolve control for FPP.

It is also vital to develop and validate the required simulation models. This is a strong role DIII-D fulfills already, and infrastructure has recently been upgraded with 100 GB/s connectivity for supercomputer simulation engagement directly in experiments. The adoption of standardized data standards such as IMAS (Integrated Modelling & Analysis Suite), coupled with the strong engagement of theory groups and broad diagnostic capabilities of DIII-D, and its high flexible and well controlled plasmas, make it an ideal proving ground for plasma simulation.

Finally, there are a range of further components and technologies that need to be tested, and indeed more are likely to emerge. As well as safe quenching approaches mentioned under challenge 3, systems like DIII-D's particle injection technologies, sample, and component testing abilities provide broad flexibility to assess new components, for example, to fuel plasmas, explore liquid surface components, or indeed spin polarized fusion techniques.

Thus, adapting a facility such as DIII-D, with its high flexibility, diagnosis, and proven technical team, is vital to U.S. goals to rapidly pioneer and validate the fusion technologies needed for an FPP, including qualifying them in the plasma environment. DIII-D is ideally placed to engage with wider programs in the United States in the private and public sectors to facilitate this mission.

The U.S. fusion path benefits enormously from its ITER engagements, which is already pioneering the approaches, design, engineering, safety, licensing, and many other aspects of an FPP class reactor. Overcoming the challenges of bringing ITER to full performance is providing, and will continue to provide, vital opportunities to confront and resolve many of the challenges for an FPP (especially those that an FPP concept shares with ITER). The successful operation of ITER itself will provide unique data that is close to FPP parameters, and thus can validate key projections.

In southern France, 35 nations are working together to build the world's largest tokamak, ITER. ITER represents the biggest construction project within the DOE Office of Science and when operational will demonstrate and study unprecedented pulsed and steady-state high-gain burning plasma regimes at the power-plant scale. During construction and assembly, ITER is contributing to both the knowledge base and workforce that will in turn feed further advances toward U.S. fusion power plants.

ITER's design is based on the results from several of the world's tokamaks, with DIII-D having played major roles in the design and preparations for operation of ITER. DIII-D's contributions span a wide range of areas, including plasma shape, vertical stability, transient control, plasma control, and operating scenarios. The DIII-D team has a history of responsiveness to ITER needs, for example, the rapid realization of a Test Blanket Module (TBM) mockup in DIII-D port to specify limits to ferromagnetic materials in the ITER (Schaffer , 2011). Today, DIII-D continues to execute vital preparation for ITER of the operating techniques, the scientific understanding and the models for validation in ITER, in order to ensure ITER can move rapidly to achieve its research goals, and the United States can extract maximum learning from ITER.

ITER construction is well along, having reached approximately 80% of completion. However, there are still numerous scientific issues that need to be addressed to ensure ITER is a success. DIII-D's role in this mission is set out in Fig. 7, where new capabilities will enable DIII-D to confront remaining challenges for rapid exploitation of ITER. Increasing ECH power, and shape and current increases will replicate ITER conditions for electron-heated, high density, low collisionality plasmas with little or no torque input. This will enable new studies in key areas, including H-mode access, ELM and Neoclassical Tearing Mode (NTM) control (Zohm, 1999), and scenario development and qualification, as also discussed for FPP.

Upgrades to the plasma control system (PCS) will enable DIII-D to mimic the ITER PCS with consideration of ITER's control and actuator set. Routine implementation of disruption-free protocols, coupled with investigations of the physics of the NTM, will play a key role in developing robust stable operation in ITER. New disruption mitigation system (DMS) tools will allow DIII-D to close out issues for Shattered Pellet Injection (SPI, ITER's day-1 DMS) and develop alternative DMS techniques as a backup.

A successful ITER research program will also rely on a talented workforce with the knowledge and experience only a major facility like DIII-D can provide. As DIII-D prepares the scientific basis for ITER's success, it will also serve as a training ground for many of the scientists and engineers who will participate in ITER research. DIII-D will be the facility where they can obtain direct experience using ITER-like tools to prepare them to address both expected and unexpected challenges in ITER. The scope of required training is broad, including everything from Physics Operator to preparing and using simulation tools and workflows that can directly interface to ITER through systems like IMAS. Scientists will learn the techniques and pitfalls of techniques ITER will need and be able to understand how to adapt to challenges in ITER when issues are manifest. Implementation of ITER-relevant control constraints and enhancements to control system, coupled with ITER-relevant physics access and new hardware tools, will enable scientists to work directly on ITER challenges in ITER-relevant ways, resolving control approaches for the range of issues expected in ITER.

These benefits are enhanced by simulation engagement that will directly test and help develop, simulation codes on DIII-D to be used in ITER. Here, the adoption of the IMAS framework on DIII-D will enable development of workflows and simulation suites that translate directly to ITER, while the rich research focus will resolve models to be able to anticipate behaviors in ITER, to both maximize ITER's performance and validate those models in ITER for application to the U.S. reactor path.

To gain the insight necessary, and ultimately attain the required performance in ITER, requires us to confront four critical challenges:

ITER needs to rapidly move to operating scenario solutions, with little time to locally solve challenges. ITER's two high-level goals aim to achieve fusion gain Q = 10 in a pulsed operating scenario for hundreds of seconds, and Q = 5 in a steady-state operating scenario, sustained for thousands of seconds. Most of the focus, including in studies at DIII-D, has been on the 15 MA pulsed “Baseline” Scenario and the high q_min steady-state scenario. Low-current alternative scenarios may enable ITER to achieve its goals in more robust (less disruptive) conditions. ITER's capabilities will be built up during its Pre Fusion-Power Operation phases, and it will explore scenarios at lower current, field, and heating power with non-nuclear hydrogen and deuterium fuel. Each of these steps will present challenges in accessing, maintaining, and safely exiting operating scenarios. In many cases, these challenges are exacerbated over those encountered in present devices, because of the low rotation and collisionality in fusion reactors, challenging controllability, and the need to integrate required ELM and radiative heat dissipation approaches.

With increased ECH power, it is anticipated DIII-D will be able to match most of ITER's dimensionless parameters (except ρ^*), and thus its physics challenges, in ITER-like conditions: electron-heated, little to no applied torque, high density, and low collisionality. These all make behavior more challenging; integrated scenario solutions have yet to be demonstrated. This will provide unique capabilities to study ITER conditions with unprecedented fidelity in order to resolve viable scenarios, and key questions of performance and stability for ITER and fusion reactors more generally.

Scenario development has traditionally focused on separating the core and the boundary, and also separating the initiation, flattop, and termination phases. However, the proposed performance upgrades on DIII-D will be able to explore this “core-edge” interaction in reactor-relevant conditions of short mean free path and high opacity to help understand how to reach a mutually compatible core-edge solution for performance, power handling, and wall material. In particular, this will help resolve questions of fueling and the resulting plasma profiles in regimes with high opacity to neutrals. Overall, this work will use uniquely relevant parameter access for ITER to help resolve the path to viable operating scenarios.

Confinement poses several challenges to achieving ITER's mission, both to reach the H-mode regime, and attain sufficient confinement at low rotation with electron heating for ITER's high-Q goals (a challenge for many reactor concepts). Predictions of the L–H transition threshold based on an empirical scaling relation indicate that access to H-mode may be marginal in ITER. Techniques to improve this prediction, and to reduce the needed power, are already under study at DIII-D, and may be essential during early research operation of ITER. Alternatively, the results of studies on DIII-D may provide needed evidence to support increasing the heating power in ITER (and other reactors), for example, through additional microwave power, presently being considered for ITER.

Reactors will operate in more challenging regimes that present devices, with equilibrated electron–ion populations and lower rotation leading to increases and changes in the manifestation of turbulent transport. Investigating these issues is highly demanding, requiring large amounts of electron heating power to access the relevant physics and conduct perturbative transport experiments such as heat pulse propagation and gradient modulation. DIII-D will be well equipped to study this physics and adapt operating scenarios to permit early adjustments of plans for ITER. This will extend to opaque regimes, where neutrals no longer penetrate into the core and density gradients become governed by combined pinch-turbulent transport effects. High normalized density operation at lower collisionality is required to address this question and to achieve mission goals without degradation in confinement that could result from impurity dilution and excessive core radiation.

Avoidance of transient events becomes more important as ITER approaches reactor conditions, with mitigation more challenging, and requirements more stringent. The challenges are similar to FPP, and here, we consider the ITER specific issues for its success, and how this further levers other concepts through validation in ITER.

For the ELM, ITER has options for ELM suppression by 3D fields, ELM avoidance with naturally ELM-free scenarios such as the QH-mode, and to some degree, pacing of ELMs with pellets. Each of these has been demonstrated on DIII-D over a range of parameters, but as for FPP research, the ECH upgrades proposed will be key to extending this to the low collisionality peeling branch and low rotation conditions for ITER Pre-Fusion Power Operation (PFPO). Beyond simple demonstration, each of these techniques will be optimized to control ELMs with a minimal impact to the core performance. Success in ITER will provide a key validation of concepts and needs for an FPP.

Disruptions are a bigger threat to ITER, though unlike an FPP, some are tolerable in ITER. If unmitigated, the thermal energy release could melt in-vessel components and the magnetic energy could impose damaging forces on components, while runaway electrons have the potential to drill through plasma facing components. DIII-D is using its advanced plasma control system to develop the capability to safely operate near stability boundaries without crossing them. This requires advances in prediction, measurement, and control actuators that are included in this plan and will directly translate to tools that can be used for the same purpose in ITER. Sophisticated machine learning, active sensing, and real time stability projection are areas the program is actively pursuing in this plan. Here, the facility serves as a testbed for multiple groups and techniques to predict, sense or safely recover from such events, including installing hardware from commercial providers with state-of-the-art machine learning technology. The ITER goal provides a sharp focus on the development of an integrated solution with the DIII-D program.

To mitigate those disruptions that do occur, ITER's primary tool is Shattered Pellet Injection (SPI), developed on DIII-D. Work will continue in concert with international collaborators to optimize SPI design. However, more advanced techniques planned for FPP provide further options for fallbacks on ITER.

Thus, ITER's key role, and the work leading to it, is to provide the opportunity to take, test, and finalize ELM control, disruption avoidance, and disruption mitigation at the reactor scale, in the last facility that will be permitted to undergo such events.

A further element in preparing for ITER is the vital need to resolve validated predictive models of the complex, non-linear, multiscale processes involved. Work on DIII-D will connect to U.S. SCIDAC initiatives and theory groups across the world program, also benefiting from unique diagnostics to characterize plasma phenomena and develop predictive models to interpret and achieve goals in ITER.

ITER represents a huge investment by the United States; ensuring its success is essential. DIII-D is well-placed and necessary to make critical and unique scientific contributions to its preparation. The return on this investment will come in the form of experienced scientists and engineers understanding the issues from work in the U.S. program and through participation at ITER, to deliver the results and validated models for the FPP path. The DIII-D program will thus act as a springboard and resource for this engagement. Combined with the innovative hardware capabilities proposed, DIII-D is uniquely placed to meet this challenge.

Sections II–IV identified the qualities DIII-D brings to support fusion energy path, and how DIII-D works with the community, and can support the private sector and public-private partnership to rapidly yield the answers necessary. The research challenges themselves, and DIII-D's role in supporting them, were set out in Secs. III and IV. This require modest investments to meet the goals. In this section, we summarize the nature of the changes needed, and the integrated modeling and state-of-the-art simulations that indicate how key plasma research and technology gaps are closed by these developments. Details of the simulations behind this work are provided in  Appendixes A and  B, setting out the projections for core and divertor, and their integration.

The critical requirement to develop a viable plasma confinement concept for an FPP is to resolve a highly dissipative divertor solution and a high-performance core that are mutually compatible, and to do so also with compatible technology resolutions, stress tested in the relevant ways. Thus, an upgrade is proposed to close this gap, equipping DIII-D to address each aspect and pioneer solutions at relevant parameters, and crucially, test their integration. The key is to raise pressure in order to sustain high density dissipative regimes with hot, low collisionality core plasmas and high power density. This is achieved through power upgrades combined with increased plasma shaping, volume, and current. Key flexibilities are introduced in core and divertor configurations, and in plasma interacting technologies, in order to resolve innovative solutions from the wall to the core, and how to integrate them together.

A performance upgrade will be achieved by combining a shape and volume rise with increased heating and current drive power in DIII-D. EPED (Snyder , 2011) simulations show the benefits of the shape and volume rise by removing the upper inner cryopump (Fig. 8), while retaining two outer cryopumps that are calculated to provide sufficient capacity for density control. This also increases the plasma current accessible for a given q₉₅. A significant rise in pedestal pressure is possible even with conventional pedestals (dark blue arrow), while a valley of improved stability, termed “super-H mode,” which was predicted by theory-based models (Snyder , 2011) and discovered experimentally on DIII-D in 2014 (Solomon , 2014), offers exciting further potential to raise conventional tokamak limits in density and pressure (light blue arrow).

The increasing pressure projected from this shape, volume, and current rise enables reactor-relevant low collisionality regimes (middle plot of Fig. 8, green to yellow shading) to be sustained and taken to higher densities and opacities. Calculations with EIRENE and SOLPS show neutral penetration depths falling to a fraction of pedestal width (Fig. 8, right plot color shading), thus capturing reactor-like regimes where particle dynamics and density profiles are governed by transport and pinch processes, rather than deep neutral particle deposition that is typical of most present tokamaks.

The detailed analysis and capabilities enabled by this configuration are set out in  Appendix A. The change will enable DIII-D to scope the limits of tokamak pedestals, understand stability effects at high β, and the impact of low neutral penetration on turbulence, and density and temperature profiles. The consequent access to regimes with higher density and opacity, and lower mean free paths would provide the first clear understanding of how pedestal and near edge regions behave in a reactor. They will also be critical in studying interaction with dissipative divertor techniques (the ITEP gap) as we discuss further in Sec. V B. The new divertor to achieve this shape rise is now being installed and will be operational in 2024.

To support such plasmas, an increase in heating and current drive is underway. Increased RF power provides reactor-relevant torque-free electron heating, as well as current drive with controllable deposition. This will be delivered with new triple frequency gyrotrons, tunable between 105, 138, and 170 GHz. 105 GHz is similar to the present system frequency of 110 GHz, typically used at 1.6–1.7 T for steady state scenarios. 138 and 170 GHz extend ECH to higher density and current regimes, and full toroidal field operation (2.15 T). It also provides higher current drive efficiency, especially with the top launch technique recently pioneered on DIII-D (Chen , 2022). These higher gyrotron frequencies are compatible with a potential upgrade to higher field operation of DIII-D, for which ongoing refurbishments and machine protections are being implemented. The field rise step is presently reserved for when greater risks can be accepted, noting uncertainties in hardware condition, and that the physics studies summarized here indicate that present fields are sufficient to access the relevant FPP and ITER physics questions.

The rises in ECH are levered by other heating and current drive improvements. Increases in neutral beam power through improved RF sources and increased voltage operation are underway and will efficiently raise bulk heating power and current drive. Furthermore, two new RF technologies are in development: “helicon” ultrahigh harmonic fast wave and high-field-side lower hybrid current drive (HFS-LHCD). These offer the potential for additional high efficiency off-axis current drive at high density, that may reduce total gyrotron power needs. A helicon prototype has just been commissioned successfully to 650 kW (Pinsker , 2023), while a 1 MW class HFS-LHCD system is being installed for 2024 testing. If effective, these can be upgraded to higher power, and could substantially augment and thereby reduce total power needs from ECH.

Full physics integrated transport simulations using IPS-FASTRAN (Integrated Plasma Simulation–Fast Transport solver) (Park , (2017) project these upgrades will deliver the transformation in the parameter space needed. A detailed mapping of the parameters obtained is attached in  Appendix B. Here, we provide some key data points and excerpts to illustrate the main conclusions:

Starting with the core-edge integration mission, Fig. 9 indicates some typical parameter values accessed and associated RF trajectories. Further data in Fig. 4 indicate ranges in individual parameters more broadly (upper panel, blue shading), and what is possible simultaneously (green shading indicates two cases with different emphases). It is found that these inductive scenarios are expected to reach low collisionality peeling-limited pedestals at high pressure and with low neutral penetration depths. Generally, the cores are highly thermalized, with low rotation and equilibrated electron and ion temperatures. They also reach substantial pressure, triple fusion product, and Greenwald fraction, while parallel heat fluxes reach 10 GW/m² to stress test and explore physics of relevant divertor solutions. These plasmas are thus well suited for testing core-edge integration issues and burning plasma relevant core conditions. The regimes are also suitable as candidate pulsed FPP configurations, and well suited to help resolve the challenges of that approach.

High β steady state solutions are also projected (Fig. 10) with safety factor maintained above 2 across the plasma to avoid tearing modes. β_N approaches 5 and bootstrap current reaches up to 75% of total plasma current, exceeding many power plant configurations set out in Table III. These plasmas are well-thermalized, operating at approximately double the density of present steady state regimes on DIII-D, with high levels of opacity and heat flux, making them well suited to assess how to integrate divertor solutions. While this example represents a so-called “high q_min” steady state solution, the flexible heating and current drive systems discussed above enable exploration of the full range of steady state and pulsed reactor solutions, ranging from peaked to broad current profiles, with varying challenges in ideal and Alfvénic stability, tearing mode and transport, in order to discover pathways to a reactor solution.

These simulations used 14 MW of ECH and significant helicon and/or LHCD power. A careful analysis of physics projections in  Appendix B suggests that reactor relevant physics integration regimes (low collisionality with short neutral penetration depths and good levels of thermalization) may be reached at lower EC powers (∼11–12 MW) if helicon or LHCD prove effective, and are operated at the ∼2 MW level. This suggests the key entry point for core-edge integration studies and the ITEP gap on DIII-D is around 16 lines of ECH—about six more than present funding is expected to support. This is a modest investment for the transformational change in mission accomplished, which otherwise requires a far more expensive new facility.

The increase in parallel heat flux and density from the performance upgrade raises plasma opacity and pressure gradients and shortens mean free paths to access reactor relevant physics in the divertor. The detailed analysis of divertor and core-edge metrics is provided in  Appendix A and was summarized in the lower panels of Fig. 4. In the left lower panel of Fig. 4, a range of divertor physics metrics that characterize and determine plasma behavior are shown (as discussed in  Appendix A, Question 3). These were compared to projected values for a typical compact pilot plant (Buttery , 2021), shown in the gray outline of Fig. 4 plots. It is noted that DIII-D already accesses many of the physics governing parameter values of an FPP, and even where it does not (such as with the turbulence broadening parameter, α/α_crit), it nevertheless already encounters the pertinent phenomena. Nevertheless, as noted in  Appendix A, Question 3, the performance upgrade deepens access in several normalized parameter regimes where a gap to FPP remains. It also deepen access in dimensional parameters (lower right panel of Fig. 4), extending scaling tests by a factor ∼3 to qualify projections toward fusion power plants, and test underlying physics assumptions inherent in the physics parameter plots. This thus provides a strong basis to develop and test divertor solutions at relevant parameters.

To pioneer the improved power handling solutions and physics projection capabilities needed for a power plant, a series of well-diagnosed divertor upgrades are planned (Fig. 11). These benefit from the considerable poloidal field flexibilities of DIII-D, which not only allow for different plasma geometries within a given limiter, but also considerable variation in flux expansion, leg length, and even secondary X points. Starting in 2023, Divertor 1 implements the shape and volume rise to test predictions of improved pedestal and core performance. This divertor is modeled on principles explored in the earlier DIII-D Small Angle Slot divertor (Guo , 2019) which directs neutrals to optimize detachment behavior. Crucially, however, this is tested with high power handling and pumping capability. This divertor will enable an early exploration of core-edge integration physics tests and pedestal performance, as well as some of the underlying principles of a slot divertor at high power.

Divertor 2, proposed for 2025–2026 installation, temporarily implements a much deeper baffled configuration leveraging the cooling capability of recycling flux with long-leg baffling. This is defined by the need to isolate and constrain key aspects of the divertor physics and their models (radiation, ionization, recombination, drifts, flows, friction), and to understand how to keep the radiative zone from the X point, which typically collapses performance when this occurs. In a third stage in 2027–2028, Divertor 3, will be implemented, based on the learning achieved in stages 1 and 2, to test core-edge optimized configurations, with high divertor dissipation that is compatible with high core performance in integrated, reactor-like configurations. This will explore use of impurity radiators (sometimes in conjunction with novel wall materials, as set out in Sec. V C) such as nitrogen, neon, argon, krypton, or xenon, depending on needs to mimic various aspects and regions of reactor scenarios, and reduce carbon induced erosion (Casali , 2022).

These developments benefit from flexible copper pedestal technology which enables rapid changes to physical structure, and further variations within each of the main stages discussed above to be relatively easily implemented. At each stage, tiles can be coated for a period to explore core-edge integration challenges regarding high-Z impurity sourcing and core contamination. Combined with DIII-D's flexible magnetic geometry, this will enable resolution of innovative divertor configurations and validated physics model for the design of an FPP divertor. It will also resolve a path compatible with the core. As discussed in Question 4 of  Appendix A, the performance upgrades will ensure metrics can be reached that simultaneously access relevant regimes in the divertor and core, and in the regions connecting them, while the long leg length of divertor #2 will isolate key physics governing this interaction and the stability of candidate solutions to provide confidence in projection models.

The upgrades in performance and relevance outlined above also facilitate key tests of plasma interacting technologies in terms of their survivability and their plasma interactions. A testbed platform approach is adopted in which new techniques can be rapidly implemented and assessed. The divertor series above is one example. To address the wider set of technology challenges in Sec. III, a series of further testbeds are planned, where collaborative partners can propose components or approaches:

• A Divertor Materials Evaluation System (DiMES) (Rudakov , 2017) sample test facility is already used to evaluate materials interaction and gross erosion for small samples, servicing a wide range of programs, including tests of wall materials for upcoming public and private sector devices such as ITER and SPARC [a compact fusion device being built by Commonwealth Fusion Systems (Creely ., 2020)].

• A Wall Interaction Test Station (Fig. 12, upper) will enable larger components to be exposed to the main chamber plasma environment, and exchanged between discharges, with analysis possible after exposure while still under vacuum. This can assess heated samples, and do so on a scale in which net erosion effects can be isolated from gross erosion. This tool could also be used to introduce and assess other components and structures to explore technologies under main plasma exposure more broadly.

• Additional testing of materials is planned in the divertor test series mentioned above, where it is crucial to understand not only impurity erosion issues with suitable exhaust management, but also leakage through the divertor into the fusion core. Coatings of tile arrays across the divertor zone will explore this issue.

• New toroidally symmetric limiters (Fig. 12, lower) will more closely mimic the geometry of fusion reactor walls in order to test impurity transport models in the far SOL and the optimization of FPP wall designs. These will also be used to explore new plasma facing component (PFC) materials such as SiC or other advanced composites or alloys, as precursors to a potential full wall change out.

• An FPP diagnostic platform is proposed to assess new techniques that are needed for reactor compatibility, noting the challenges for optical diagnostics, long pulses, and plasma facing in vessel components in a fusion power plant.

• A gyrotron test socket platform will provide opportunities to test new microwave technologies, including the high frequencies required for an FPP, benefiting from DIII-D infrastructure.

• Ongoing developments of heating and current drive systems will provide opportunities for testing of novel ideas for antennae for helicon and HFS-LHCD techniques possible as these systems are expanded.

• Control infrastructure is already being used to assess a range of novel control schemes, including advanced machine learning techniques. The present digital–analog twin will be expanded to enable more integrated testing of reactor algorithms, while the plasma control system will also be developed to mimic the control and diagnostic limitations in a fusion reactor.

• Disruption mitigation technologies will focus on approaches that can inject particles more rapidly, to reach the core of the plasma and thus quench it from the inside out, thereby minimizing induced forces and runaway electrons.

• A runaway electron mitigation coil (Fig. 13) is planned to test concepts of passive mitigation through induced voltages during a disruption driving currents in the coil. Simulations (Weisberg , 2021) indicate the associated MHD driven will dissipate the electrons, making it a highly promising tool close the loop on this final most critical part of the disruption problem.

• Novel fueling approaches will be tested, including an exciting proposal to test the principles of “spin polarized fusion” whereby fuel ingredients are spin polarizes which increases fusion cross sections by 50% and consequent fusion gains by a larger factor.

These testbed platforms and tools are modest in cost, benefiting from DIII-D infrastructure and hardware already required for the wider program. Their implementation will create space for new ideas to come and be tested, as work in the fusion community progresses through FPP candidate solutions. To this end, a new Plasma Interacting Technology program has been launched to oversee this field, outreach to the wider community and support their engagement.

These developments transform DIII-D capabilities to pioneer integrated core and edge solutions for fusion power plants and provide crucial assessments of fusion power plant technologies, and their compatibility with plasma solutions. This therefore provides equivalent-to-new-facility capabilities, at a fraction of the cost and lead time of a new facility, thereby removing the need for additional device steps and reducing time before the FPP. Thus, this represents the most effective, timely, and cost-competitive path to resolve the plasma research challenges for an FPP and ITER.

To provide a rigorous basis for an FPP and ITER, understanding must be combined across facilities and programs worldwide, in order to develop not only promising solutions but also robust validation and understanding for confident projection to the reactor. In this section, we discuss how the DIII-D program intersects with the rest of the world program, summarizing the distinctive elements DIII-D brings, key complementarities, and the collaborations and engagements that are particularly important to pursue to close gaps to FPP and ITER.

The capabilities and developments discussed in earlier sections provide a powerful basis to support the fusion commercialization agenda, public–private partnership, and particularly the private sector. As an Office of Science national user facility, DIII-D exists to support both public and private sector partners in the research objectives of the Unites States. We have set out how the DIII-D user model is well suited to engaging and supporting a very wide diversity of users already, and how the user model and approach has been further broadened to enable such engagement, more broadly and more easily. Earlier developments, such as in the Frontiers Science field, have shown that how the facility can rapidly incorporate and support new perspectives to deliver outstanding scientific insight. Already, private sector partners have begun testing key components and concepts for future devices at DIII-D, including not only for the tokamak path but also more novel and innovative concepts.

Under this approach, institutions will be free to sign up for DIII-D engagement, immediately gaining access to internal discussions, expertise, and program development processes for the personnel they sponsor. As individuals join the program, they will be able to participate in discussions to define the research program and the various steps and tools to be pursued. This applies not only to run time; a lot is possible through installation of systems, samples and components, whose testing may require little to no dedicated run time (and aligns with programmatic goals to support the public–private path and fusion technology development). As a user facility, such access is free and supported, with expert advice, computer and data access, information systems, office space, and personnel support. This includes access to DIII-D personnel development and education mechanisms ranging from student and intern support to postdoc, early career and support to experienced personnel.

The testbed platform approach and new Plasma Interacting Technology program will provide particular opportunities for the private sector to engage. Here, the program is presently in a major phase of consulting the community and working with them to develop the research priorities and identify the needs that the private sector and the wider U.S. fusion path needs to pursue. The testbed platform approach is intended as a way to make such engagements easier, by providing expedited means and standardized interfaces for bring new pieces of equipment to the facility and exposing them to relevant conditions and tests. Furthermore, DIII-D oversight and guidance bodies have been reformed to involve private sector perspectives.

Many engagements are also possible in the wider research program that focuses on the plasma physics side. Here, DIII-D fulfills an important role of being able to use its flexibility both to try out ideas and paths for other fusion players, and to develop fallback techniques, if a particular approach is not able to resolve a particular challenge. This can thus answers questions quickly, rather than require expensive and time-consuming additions to the private sector facility. For example, should tearing modes emerge as a problem for FPP concepts, DIII-D can use its highly flexible RF and perturbative coil capabilities to resolve strategies for their control. DIII-D can also use its physics investigative capabilities to resolve models in order to enhance confidence in projection of the relevant approach.

Thus, DIII-D, as an already-resourced highly configurable and diagnosed facility, with extensive capabilities that are ready to be used now, provides a key opportunity to close gaps and accelerate the path for the private sector.

There are many aspects to qualifying a reactor solution that go beyond DIII-D capabilities. Tokamaks around the world bring many strengths to close these gaps. A summary is provided in Fig. 14 arranged by role, with more details in Table IV. Some key elements include interaction with reactor-relevant walls (where ASDEX Upgrade, JET, and EAST bring key strengths), long-pulse behavior (EAST, KSTAR, and ITER), with super-Alfvénic fast ions (MAST-U, JET, ASDEX Upgrade, and NSTX-U) and with tritium [ITER, SPARC, and BEST (Slendebroek, 2022)]. In some cases, it will also be important to test key projections of physics mechanisms closer to reactor parameters where JET, JT-60SA, SPARC, Divertor Test Tokamak (DTT), and ITER may provide key tests.

It is vital to explore these issues by bringing together the unique elements of each facility to develop holistic solutions. The areas where DIII-D brings distinctive capabilities are outlined in Table V. These provide a key basis for collaboration between the United States and other partners. In some cases, DIII-D may do pioneering work, with checks and extensions made elsewhere. In others, DIII-D may be needed to provide tests to resolve physics understanding and projections, or assess integration with other techniques. For example, on divertor configuration (MAST-U, TCV, and later DTT), 3D fields (KSTAR, ASDEX Upgrade, MAST-U), heating technology (ASDEX Upgrade), and materials (testbeds, WEST, NSTX-U, ASDEX Upgrade, and JET). We discuss these aspects below.

A key issue is the extension to long-pulse, where DIII-D's collaboration with the EAST facility, with a joint task force operating on both facilities for several years, has helped to qualify plasma solutions for an FPP and ITER. DIII-D has focused on exploiting its flexibility to develop so-called high β_P core scenarios where core transport properties are substantially improved (Fig. 15, left panel). This has significantly alleviated the challenge from edge localized modes, while more recent work established use of core-radiators to mitigate divertor heat flux challenge. On EAST, these techniques were extended to long-pulse (Fig. 15, right panel), demonstrating their stability and wall compatibility. This led to key insight in the interaction with heating and current drive systems (here LHCD), as density profiles broadened and reached a naturally self-consistent state with broad current profile (Garofalo , 2017). This engagement provided demonstrated key principles of high density and confinement for FPP and provides potential for ITER, showing how discharges can be sustained and made compatible with long pulse wall load requirements.

Divertor research is another example. The MAST-U facility provides unprecedented ability to manipulate divertor configuration, while TCV brings other important geometric flexibilities. However, DIII-D provides unparalleled diagnosis to resolve underlying physics questions, while the upgrades and higher power loads on DIII-D will close key gaps on reactor physics regimes, particularly in terms of neutral path lengths, ionization-recombination ratios, turbulence broadening, and fluidity. DIII-D team members are thus jointly engaged with MAST-U colleagues to support the field and have played key roles in enabling configuration development on MAST-U. Working together offers the potential to resolve the physics of divertor detachment and optimize physical geometries for an FPP.

These arguments generalize when considering DIII-D's relationship with higher field or larger scale facilities. In particular, the SPARC facility, now under construction, provides an exciting opportunity to access parameters closer to a fusion reactor. Though lacking the range of investigative capabilities of DIII-D, SPARC will thus deliver critical tests of physics scaling for a range of phenomena for the pulsed FPP approach, including divertor science and core profiles at high opacity, as well as tearing stability, transport, and H mode access. However, the SPARC facility has made particular choices and has narrow investigate range in areas such as the planned divertor, choice of 3D field flexibility, or absence of tearing mode control. Thus, an approach based on SPARC alone risks delaying the path unless these particular solutions break the right way. In contrast, DIII-D, working alongside SPARC, provides the flexibility and power to resolve the path and project it with confidence. Similarly, JT-60SA, which has higher current and field, longer pulses than DIII-D, and significant high β and off-axis current drive capabilities, will provide important tests for advanced tokamak steady state approach, complementing flexibilities and physics research capabilities on DIII-D. Finally, the Divertor Test Tokamak (DTT) (Zagórski et al., 2017), proposed for operation in the 2030s in Italy will provide complementarity in analyzing novel divertor tests nearer FPP parameters. This is particularly relevant for the larger EU-DEMO path, with more conservative confinement scenario, but comes rather late for the U.S. path. It may nevertheless help refine or verify designs of FPP, or take other adjustments depending on stage in the FPP cycle.

Complementarity on wall choice is another key aspect. While DIII-D's carbon wall is not a reactor candidate solution, it permits access to key physics regimes such as low collisionality pedestals, steady-state regimes, and the study of key phenomena that may damage walls. These are difficult to access in the current generation of tungsten devices because the tungsten radiates in the core, requiring gas puff to help flush it out, raising collisionality. This permits DIII-D to address critical challenges such as the study of stability, disruption mitigation, and ELM control at low collisionality, and to push further to higher energy confinement regimes where tungsten accumulation may limit performance in the present generation of metal-walled devices (but not future reactors, where temperatures are higher and radiation further out). It also enables DIII-D to study material interactions and transport perturbatively, to get at the physics behind projections of metal wall behavior. Thus, DIII-D executes complementary model validation on plasma physics and material interactions to those being explored in the ASDEX Upgrade, WEST, and JET facilities, and there is value in coordinating such studies to provide more robust and validated projection capabilities for future reactors overall.

A final example is ELM control. Highly collaborative research between DIII-D, MAST-U, ASDEX Upgrade, and KSTAR facilities is utilizing complementary parameter access and coil design features to address critical questions as to how 3D fields can manipulate the pedestal to suppress ELMs, and which governing parameters matter (e.g., density or collisionality). This goes to the heart of whether ELM suppression techniques will project to future devices with different rotation, β, or shape. The differently structured 3D field coil sets providing complementary model validations on the plasma response part of the process. Work on DIII-D will confront key questions by studying ELM suppression on the reactor-relevant low collisionality peeling branch, to assess behavior at low rotation and high β, and explore compatibility with various shape configurations.

More broadly, JET is performing critical work on scenarios and control with a metal wall and divertor. It is also validating key aspects of models initially tested on DIII-D such as peeling-ballooning (PB) pedestals and gyrokinetic turbulence, and the scaling of techniques such as disruption mitigation with massive particle injection. Collaboration with spherical tokamaks such as MAST-U and NSTX-U is helping assess how physics changes with aspect ratio, and the role super-Alfvénic ion populations. In the next decade, NSTX-U plans to provide an alternative liquid wall approach to power handling, for which preliminary component tests could be undertaken in the proposed WITS on DIII-D. Innovative work in the yet more compact Pegasus facility is exploring solenoid-less startup which would provide a key further simplification of the tokamak concept, with later possible testing in DIII-D scenarios.

Thus, DIII-D occupies a distinctive role, with its capabilities and approaches complementing those elsewhere. The key qualities DIII-D brings are flexibility and integration—to probe potential solutions amongst a wide range of possibilities, and to resolve how different techniques, that can often trade-off against each other, can be combined. With DIII-D's strong physics investigative capabilities and diagnostics, and access to relevant reactor physics regimes, this make DIII-D a strong facility to identify key techniques of fusion energy and develop the physics understanding and validated models to project them to future reactors.

Recent strategic studies in the United States have proposed a further intermediate scale facility on the path to fusion energy (FESAC, 2020). This proposal was essentially for a substantially higher field version of DIII-D, with short pulses and deuterium operation in order to avoid activation and provide a flexible hands-on facility, a la DIII-D. This has been variously termed Sustained High Power Density (SHPD), Next Tokamak User Facility (NTUF), and EXhaust and Confinement Integration Tokamak Experiment (EXCITE). The recent FESAC report recommended its construction but stated that “mission need” needed to be established, and compared with the alternative path through upgrades to existing facilities and international collaboration. We argue this latter path.

Analysis summarized in this paper has identified that an upgraded DIII-D can access the relevant reactor physics regimes to address this “EXCITE mission.” DIII-D has the flexibility and relevance to pioneer key parts of the solutions. Complementarity and collaboration with other funded facilities will ensure the remaining gaps can be closed, including providing checks closer to FPP parameters, and at longer pulse with relevant walls.

While a high field, flexible science device would indeed validate models closer to reactor parameters, it is not critically needed, and would in any case not fully resolve reactor questions itself, due to the short pulse nature and lack of tritium proposed for it. Furthermore, a build from scratch new facility will come too late for the decadal vision to answer questions in time for design decisions on the FPP, adding 1–2 decades and at least $2Bn to the U.S. path. These resources are better directed into technology programs, while we exploit existing capabilities to close plasma research gaps.

Thus, the case for a build-from-scratch new facility to conduct an EXCITE mission has not been made. The model of exploiting the full potential of DIII-D, alongside engagement with other complementary facilities in the U.S. and internationally, represents a better, more cost-effective and faster path to prepare for an FPP and ITER. It should be noted, however, that should results emerge that do motivate such a higher field intermediate facility, the most cost-effective path to achieve this would be as a rebuild of DIII-D with a new toroidal field set.

DIII-D provides a vital resource to the United States for the development and commercialization of fusion energy. It provides breadth, depth, and capacity to support the required plasma research, and the range of pathways and technologies that must be tested. As a national user facility, it exists to serve this national agenda, sharing its expertise and capabilities in an open and transparent way across public and private sector. As a high-capacity, flexible research tool with outstanding technical team, it is able to rapidly implement and test new concepts to pioneer the path and technologies for fusion energy. DIII-D thus represents a crucial opportunity to enable the public–private path to fusion energy, and, in particular, support the burgeoning range of developments on the private sector.

The facility is thus being reinvented to meet this new national agenda. A new user model has been developed to facilitate the engagement of new groups and privately funded fusion companies and organizations, while protecting intellectual property both sides. This enables a rapid and highly collaborative engagement with low bureaucracy and cost, provided DIII-D-derived data remain in the public domain. A new Plasma Interacting Technology program has been launched and is outreaching the program to these groups on the technology side, with a new platform approach to open up DIII-D capabilities and flexibilities for rapid prototyping of fusion techniques. A new User Board has been established to provide independent oversight of this program, with a Program Advisory Committee also embedding national and international perspectives from public and private sector participants.

The research community has identified a range of critical plasma research and technology challenges that must be addressed if a viable and compact Fusion Pilot Plant, with the necessary higher levels of performance and power handling needed for a compact device, is to be realized. DIII-D already has the capabilities needed to resolve many of these challenges, such as disruption mitigation, control, performance limits, divertor configurations, and advanced pedestals. It also has deep investigative capabilities to explore the underlying science of a wide gamut of plasma physics phenomena in order to develop validated physics models for the phenomena it accesses. Modest further investments in testbed facilities and additional flexibilities will thus exploit this foundation to rapidly resolve many of the key challenges, rapidly and cost-effectively delivering critical answers for the FPP and ITER, and turbocharging the private sector approach to fusion energy.

Further DIII-D will close the Integrated Tokamak Exhaust and Performance gap. To address this requires DIII-D to operate at higher densities while maintaining reactor-relevant hot collisionless cores, in order pioneer dissipative divertor solutions and resolve compatibility with the core in relevant opaque integration conditions that govern the interaction of the two regions. This requires an RF upgrade to support such plasmas, which when combined with higher shape, volume, current, and field operation, is projected to access the relevant integration regimes. This is readily achievable with off-the-shelf purchases and implementation of proven technologies, providing equivalent-to-new-facility capabilities, at a fraction of the cost and timescale. DIII-D's proven team, flexibilities and diagnostics will thus be able to rapidly close this gap, meeting the timescale of White House decadal vision.

Thus, DIII-D occupies a lynchpin role in the world fusion program, being able to prove out and marry together core, edge, and technology solutions with unique insight and explorations possible, that complement the distinctive capabilities of other facilities around the world. DIII-D will thus provide a basis for U.S. leadership on these questions and for engagement with the international community, where its proven collaborative models will enable the international community can move to holistic answers.

The expanded agenda and needs of the new U.S. mission also requires additional development of personnel. Here, DIII-D is providing broad support, with many opportunities for early career leadership and development, a pilot career mentorship program and over 200 students, interns and postdoctoral fellows, with dedicated run time for Ph.D. students. With over 700 audited users engaged, over 30 DOE funded PIs, and around 100 collaborative institutions, DIII-D provides a strong platform to engage and enable the development of scientists and engineers to meet U.S. workforce needs. However, critical in the pursuit of this mission is to enable engagement and progression of personnel from all backgrounds. Plasma physics in the United States remains dominated by the traditional majority, with exceptionally low participation of gender, ethnic, and other minorities. It is vital to address this from a social justice perspective. However, if a more calculated justification is needed, the growth of the workforce required needs the national program to draw from a greater diversity of sources; one simply cannot do it with the present majority and R1 institutions alone. Here, DIII-D seeks to a beacon of good practice in enabling opportunities for all, by creating a more empowering and fostering workplace environment, and by diversifying opportunities and pathways, with support through training, mentorship, surveys, professional advice, implicit bias monitoring, open opportunities policies, underserved community engagement, communication of management expectations, and a code of conduct with consequences. There remains much to do to achieve the cultural change needed. However, with continued progress, and the expansion of engagements and opportunities underway, DIII-D has the potential to strongly support the development and diversification of the U.S. workforce for fusion energy.

Thus, DIII-D has unique potential and is, in fact, vitally needed to meet the U.S. fusion commercialization agenda and the decadal vision to fusion energy. The DIII-D team has a proven track record of outstanding delivery of technical projects and science, routinely achieving world leading status in the quality and quantity of magnetic confinement fusion research. The proposed research will deliver the critical insight needed to take fusion forward, to test key plasma interacting fusion technologies, and resolve critical answers for the FPP, the private sector, and ITER. The user model and workforce development approach is being transformed to meet this new agenda, expanding opportunities and enabling diversity. The facility is thus ready to meet the challenge, with proven competence and capabilities, to rapidly deliver answers that would otherwise need a new facility, at a fraction of the cost and timescale, and with better scientific insight, in order to enable the decadal vision for fusion energy in the United States of America.

This material was based upon work supported by the U.S. Department of Energy, Office of Science, Office of Fusion Energy Sciences, using the DIII-D National Fusion Facility, a DOE Office of Science User Facility, under Award Nos. DE-FC02-04ER54698, DE-AC05-00OR22725, and DE-AC52-07NA27344.

This report 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, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness 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. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

The authors have no conflicts to disclose.

Richard John Buttery: Conceptualization (lead); Formal analysis (equal); Funding acquisition (lead); Investigation (lead); Methodology (equal); Project administration (lead); Supervision (lead); Visualization (equal); Writing – original draft (lead); Writing – review & editing (equal). Tyler Abrams: Conceptualization (supporting); Visualization (supporting); Writing – review & editing (supporting). Livia Casali: Conceptualization (supporting); Formal analysis (supporting); Investigation (supporting); Methodology (supporting); Validation (supporting). C. M. Greenfield: Writing – original draft (supporting); Writing – review & editing (supporting). Richard Groebner: Conceptualization (supporting); Formal analysis (supporting); Investigation (supporting); Writing – original draft (supporting); Writing – review & editing (supporting). Christopher Thomas Holcomb: Conceptualization (supporting); Investigation (supporting); Methodology (supporting); Supervision (supporting); Writing – review & editing (supporting). Suk-Ho Hong: Writing – review & editing (supporting). Aaro E. Järvinen: Conceptualization (supporting); Formal analysis (supporting); Investigation (supporting); Methodology (supporting); Validation (supporting); Writing – review & editing (supporting). Anthony Leonard: Conceptualization (supporting); Formal analysis (supporting); Investigation (supporting); Methodology (supporting); Writing – original draft (supporting); Writing – review & editing (supporting). Adam G. McLean: Conceptualization (supporting); Writing – review & editing (supporting). Thomas H. Osborne: Conceptualization (supporting); Formal analysis (supporting); Investigation (supporting); Writing – review & editing (supporting). David C. Pace: Conceptualization (supporting); Formal analysis (supporting); Methodology (supporting); Project administration (supporting); Resources (supporting); Supervision (supporting). Jin Myung Park: Formal analysis (supporting); Investigation (supporting); Methodology (supporting); Software (supporting); Validation (supporting); Visualization (supporting). Craig Petty: Conceptualization (supporting); Investigation (supporting); Project administration (supporting); Supervision (supporting). M. W. Shafer: Conceptualization (supporting); Formal analysis (supporting); Validation (supporting); Writing – review & editing (supporting). Adrianus Sips: Funding acquisition (supporting); Project administration (supporting); Resources (supporting).

The data that support the findings of this study are available from the corresponding author upon reasonable request.

The 2020 FESAC report, “Power the Future—Fusion and Plasmas” (FESAC, 2020), provides a U.S. vision for a strategic path to fusion energy which focuses on building an FPP by the 2040s. To help ensure the success of this effort, it is highly desirable or even essential to reduce a number of scientific gaps between existing machines and an FPP. The developments underway for DIII-D can significantly reduce, although not completely eliminate, several important gaps.

The planned upgrade to DIII-D will increase the volume of plasmas in the existing vacuum vessel through increases of elongation and triangularity. The higher volume will allow plasmas with higher plasma current I_p than can be achieved now. Modeling with the EPED code predicts that these conditions would provide access to plasmas with much higher pedestal pressure with higher pedestal densities at low collisionality than can be currently achieved.

These upgrades will enable the study of several important questions related to issues of core-edge integration for a fusion power plant. This appendix lists several of these questions, physics processes inherent in them, metrics for the processes, and evaluates how far DIII-D developments can advance these metrics toward a possible FPP. For this study, the metrics for an FPP are obtained from a study of fusion power plants, based on the compact advanced-tokamak (CAT) concept (Buttery , 2021). In particular, optimizations for a toroidal field B_T of 7 T, major radius R of 4 m, and aspect ratio of 3.1 are used.

Q1: Can DIII-D Achieve Reactor-Relevant High Pedestal Pressure at High Density?

An FPP, operating at high pedestal pressure and high density, will have a combination of lower pedestal collisionality and much higher opacity to neutrals than can be obtained in existing devices. These characteristics raise a number of important questions for the performance of the FPP. Some of these questions are:

• How will low collisionality at high beta affect the MHD stability and transport of the pedestal?

• How will lack of neutral penetration affect the pedestal? Will the pedestal density profile have a weak gradient, will the gradient region of the pedestal shift outward at high density, will large values of η (ratio of density to temperature scale length) be achieved, and if so, will they lead to increased transport?

• Will a flattened density profile with a strong temperature gradient occur and lead to pedestal neoclassical screening of impurities?

• How does a low collisionality, high-density pedestal affect attainment of a dissipative divertor?

Design studies using the EPED code predict that a DIII-D upgrade could enable dramatic increases in the pedestal pressure (and therefore pedestal density) with pedestal (evaluated as pedestal-top electron-ion collisionality) below unity. These studies predict DIII-D could reach a regime with very shallow penetration of neutrals into the pedestal and thus enable the study of pedestals with high opacity to neutral fueling. The modeling to date considers the proposed volume and shape increases with the increase in plasma current I_p.

This upgrade would enable validation of the EPED model (Snyder , 2009 and Snyder , 2011) under extreme, reactor-relevant conditions. EPED predicts that at the high current and shaping of this upgrade, operation along the peeling branch of the pedestal MHD limit will occur at very high densities. The pedestal is predicted to be operating in the super-H regime (Solomon , 2014 and Snyder , 2015), and modeling studies indicate that the region of access will be much larger than is now available in DIII-D. Except where noted, the studies presented here do not incorporate an increase in B_T to 2.5 T, which would provide additional significant increases in the ability of the machine to advance pedestal metrics toward an FPP.

The projected pedestals of this shape upgrade would enable a number of reactor-relevant studies to:

• explore high pressure and density limits in the pedestal,

• study relevant transport and pedestal instabilities at low collisionality,

• study pedestal transport in a regime not driven by neutral deposition, and

• integrate high-performance pedestal with dissipative divertor and reactor-like recycling.

For purpose of discussing whether the proposed DIII-D upgrade could achieve a high-pressure, high-density pedestal that would be useful for closing important gaps to an FPP, Table VI lists physics processes predicted by the EPED model to control the pedestal pressure profile, along with useful metrics for assessing the performance characteristics of the pedestal. The physics processes listed in Table VI, MHD stability due to peeling-ballooning (PB) modes and limits on the pedestal pressure gradient due to kinetic ballooning modes (KBMs), are combined by EPED and used to predict the height and width of the pedestal pressure profile. Based on extrapolation from an experimental super-H discharge, EPED calculations have been used to predict the metrics that could be obtained with a volume and shape upgrade of DIII-D. The metrics of interest are the total pedestal pressure p^ped, the pedestal electron density n_e^ped, the ratio of the pedestal density to the Greenwald density f_GW, and the pedestal top electron–ion collisionality

The discharge of the reference DIII-D case has upper and lower triangularity of 0.45 and 0.66, respectively, I_p = 1.04 MA, B_T = 2.12 T, and q₉₅ = 4.0. Based on EPED modeling, a range of DIII-D upgrade possibilities was examined. The case considered here had the same toroidal field as the reference, triangularity of 0.90, I_p = 2.2 MA, and q₉₅ = 4. In Table VII, the pedestal performance metrics for the DIII-D reference and upgrade cases are shown along with the metrics from an FPP design point, which has machine parameters of B_T = 7 T, R = 4 m, I_p = 9.5 MA, aspect ratio of 3.1, and f_GW (fraction of Greenwald density) of 0.9. In these computations, the measured Z_eff value of 2.7 is used for the DIII-D reference case, and an estimate of 1.9 is used for DIII-D upgrade, based on projections of carbon density from the reference case. The FPP design is assumed to have an argon impurity fraction of 0.05% (Buttery , 2021), which implies Z_eff ∼ 1.1. Two study cases are shown for DIII-D upgrade, one at f_GW of 0.8 and one at f_GW of 1.0.

As shown in Table VII, the predicted metrics of the DIII-D upgrade reduce most gaps by a significant amount: the pedestal pressure is predicted to increase by a factor ∼3, and values of the pedestal density and Greenwald fraction are predicted to be very near those of the FPP. DIII-D upgrade cases 1 and 2 are predicted to have less than or equal to unity with n_e^ped equal to or greater than the FPP.

The DIII-D upgrade would thus test whether high pedestal pressures predicted by the EPED model can be achieved under conditions of low pedestal fueling, which may produce flat density profiles that are very different from the EPED density profile assumptions. If the predictions are verified in experiment, the metrics predict that the upgrade would provide a high-performance pedestal which would substantially advance pedestal metrics toward reactor-relevant conditions, thereby enabling reactor-relevant studies, such as those listed above.

Q2: What are the Characteristics of the Pedestal if Neutrals Only Fuel it to a Small Depth?

Predictions and expectations for ITER and fusion power reactors are that the density in the divertor, scrape-off layer, and pedestal will be so dense that very few neutrals will cross the separatrix far enough to sufficiently fuel the pedestal (Romanelli , 2015). As a result, the pedestal density is predicted to have a low gradient compared to existing machines. As a consequence, the E × B shear in the pedestal may be low and thus enable the growth of long wavelength modes which would have a deleterious effect on pedestal performance. Values of η for both ions and electrons may be large with associated increased turbulence.

A potential positive benefit is that a low-density gradient might enable neoclassical transport to cause an outward flow of impurities and thus provide impurity screening for the core. Another potential benefit is that modestly increased turbulence might flatten the pedestal pressure gradient, allowing the pedestal width to increase and ultimately leading to higher pedestal pressure such as for wide-pedestal QH-mode. Uncertainties in these issues result in significant uncertainties about whether or not future large machines will be able to operate in H-mode as desired.

If DIII-D could achieve a pedestal that only had fueling to a small pedestal depth, it could address the issues of pedestal structure and transport in the absence of a source, as noted above. This capability would enable tests of transport models in much more reactor-relevant conditions than can be done now. Table VIII lists physics processes and metrics, related to fueling, which are used in Table IX to evaluate if the proposed DIII-D developments could achieve sufficiently shallow fueling to better test these reactor-relevant questions. The primary physics processes of interest are ionization and charge exchange of neutrals, as listed in Table VIII. A simple metric for comparing the opaqueness to neutrals between machines and scenarios is a/λ_N, the ratio of the minor radius a to the neutral penetration depth λ_N for ionization by electrons. Larger values of this metric imply increased opacity (reduced transparency) to neutral penetration. With the relatively weak dependence of cross sections on temperature, the scaling λ_N ∼ 1/n is assumed and thus the metric a/λ_N ∼ an_e^ped, shown in Table VIII.

A metric designed to look more closely at the ability of neutrals to fuel the plasma and affect the pedestal structure is the ratio of the neutral depth λ_N to the pedestal width w_ped. The utility of this metric is that if a pedestal structure is observed and the metric λ_N/w_ped ≪ 1, we would conclude that non-diffusive transport, i.e., a particle pinch, was playing a dominant role in determining the pedestal structure.

For the purpose of assessing the ability of an upgraded DIII-D to perform meaningful studies of this regime, neutral fueling calculations have been performed for a wide range of parameters and profiles that are expected to be accessible in the upgrade, as predicted by the EPED model. These fueling calculations were performed with the ONETWO transport code (St. John , 1994), which uses the kinetic neutrals 1D transport model NEUCG (Burrell, 1978). SOLPS/EIRENE (Schneider , 2006) modeling was used to adjust the flux expansion parameter (for modeling a 2D neutral distribution) and neutral energy in the ONETWO 1D model to match SOLPS. This modeling includes the effects of both ionization and charge exchange on the penetration of the neutrals (Casali , 2020). For the purpose of this study, the neutral penetration depth is evaluated as the distance from the separatrix at which the neutral density is reduced by 80% from its value at the separatrix and is designated as λ_N80 in Tables VIII and IX. Collisionality is also added as a metric since the regime of interest for a reactor has both low collisionality and high opacity to neutrals.

Table IX shows opacity metrics for the DIII-D reference and upgrade cases, evaluated at = 1, as well as from the same FPP design case used for Table VII. The predicted opaqueness for the DIII-D upgrade significantly reduces the gap to an FPP, being about 60% of the FPP design point. The values of neutral depth/pedestal width, λ_N80/w_ped, achievable in the DIII-D upgrade are lower by a factor of 2–4 as compared to the reference DIII-D case. More importantly, the modeling indicates that the neutral penetration depth would be 10%–20% of the predicted pedestal width. Modeling of neutral penetration for an FPP has not been performed, but the λ_N80/w_ped metric is expected to be much less than unity. If the predicted conditions were achieved in DIII-D upgrade, we would conclude that neutral fueling was not the primary determinant of the pedestal structure and would be able to study physics of the density pedestal in reactor-relevant conditions, albeit at higher collisionality than in the FPP. Densities up to 2.3 × 10²⁰ m⁻³ are predicted to be supportable with the new triple frequency gyrotrons now being procured when deployed at 170 GHz with top launch, O mode polarization, and present toroidal field limits, or with outside launch if higher fields are permitted. Use of the 137 GHz extends the range in fields downward, though at a slightly reduced density of 1.6 × 10²⁰ m⁻³. This will be supported by a new pellet injection system, which is planned to inject at up to 160 Torr L/s at 20 Hz, provides up to 10²² electrons/s throughout the discharge. Additional systems can be added if needed.

The calculations above use the B_T of the existing DIII-D. For an upgrade to B_T = 2.5 T, significant further increases of pedestal pressure and therefore increases of neutral opacity and reductions of collisionality would be expected. Two assumptions on the scaling of pressure can be used to provide estimates of what might be expected. Under the assumption that p^ped ∼ B_T, an increase in p^ped of ∼18% would be expected. Assuming that I_p could be increased with B_T and that p^ped ∼ B_TI_p, an increase in p^ped of ∼40% would be expected. At fixed pedestal density, these increases would result in reductions of as 1/(p^ped)² and could reduce the DIII-D upgrade projection of collisionality from ∼1 to ∼0.5–0.7 at these very high densities.

Q3: Can DIII-D Achieve Regimes that Test the Key FPP Dissipative Divertor Physics?

Reactor designs use two approaches to exhaust the heat to ensure plasma facing components are not destroyed. These are: (1) divertors that use line radiation from seeded impurities to disperse the heat and expand flux surfaces to reduce incident heat fluxes, and (2) a radiative mantle to disperse the heat over a large surface area. Below, the ability of an upgrade to close gaps for a radiative divertor is examined. Our studies predict that DIII-D upgrade would have several important metrics near reactor values for an FPP. These metrics indicate DIII-D will be able to study the underlying physics processes that are expected to play an important role in heat exhaust control for FPP. Confidence in FPP divertor design will be enhanced using well-diagnosed discharges to test and validate codes used in the design of reactors.

A dissipative divertor in an FPP will use the same physics processes that occur in current divertors. Relevant divertor processes that might scale from existing tokamak divertors to a reactor include (1) the ionization distance of recycled neutrals compared to the physical size of the divertor, (2) the level of plasma recombination compared to ionization, (3) the absorption distance of Ly-α emission from recombining plasmas compared to device size, (4) the neutral–neutral collision distance compared to device size (i.e., how fluid-like the neutrals are), and (5) the plasma pressure along open field lines that may drive significant radial transport. These physics processes with metrics are shown in Table X.

Table XI shows the evaluation of the above metrics for the current DIII-D, a DIII-D upgrade, and the FPP design. Parameters for the reference DIII-D case are different than those for the previous work in this note. Divertor conditions for the reference DIII-D case are Ip of 1.4 MA and exhaust power into the scrape-off layer (P_SOL) of 10 MW. These parameters resulted divertor densities and temperatures measured by our divertor Thomson Scattering (DTS) diagnostic to be approximately 2 × 10²⁰ at 10 eV to 6 × 10²⁰ at 1 eV. These are roughly the temperature regimes at which the processes described above dominate, ionization and impurity radiation ∼10 eV and recombination associated processes at ∼1 eV.

An upgraded DIII-D is assumed to have B_T = 2.5 T, I_p = 2.5 MA, and P_SOL of 40 MW. The FPP is assumed to have R = 4 m, B_T = 7 T, I_p = 9 MA, and P_SOL of 90 MW. For projecting the metrics from DIII-D to the upgrade and to an FPP, it is necessary to project the divertor density and temperature. For this purpose, the divertors of all machines are assumed to be operating at the same T_e as in DIII-D since divertor dissipation due to atomic physics processes maximizes at fixed T_e. The divertor density is assumed to be proportional to the square root of the parallel power (heat flux density) since atomic radiation is proportional to the square of n_e at fixed impurity fraction. In turn, the heat flux density width is assumed to scale as 1/B_p,mid, the ITPA (International Tokamak Physics Activity) scaling, where B_p,mid is the poloidal magnetic field at the midplane. We also note that the divertor width estimate for DIII-D is a combination of density and temperature widths and is also assumed to scale as 1/B_p,mid. The divertor length is assumed to scale with major radius.

Table XI (first six rows) shows several of the parameters used to calculate the metrics as well as the actual metrics. The values of ionization length, Ly-α length, and neutral length all decrease from DIII-D to DIII-D upgrade to the FPP due to increased plasma density and resulting neutral density. However, when normalized to the divertor width which decreases with magnetic field, these metrics are all relatively similar between the machines. The important factor is that these characteristic lengths are all smaller than the divertor width, whether in DIII-D upgrade or FPP. This implies the important processes of radial transport of neutrals, radiation, and ionization can be studied in similar regimes in DIII-D upgrade as in the FPP. For the divertor poloidal length, the ratios vary nearly an order of magnitude from DIII-D to the FPP with DIII-D upgrade partially covering that gap. However, the ratios are very large, suggesting the relevant physics processes can be studied in DIII-D. In contrast, the ratios of rates for recombination to ionization are very similar in all cases.

The metric α_MHD/α_CRIT, the normalized pressure gradient in the midplane SOL, is considered an important parameter for radial transport. The value of α_MHD has been measured in DIII-D at the outboard midplane to be near the ideal ballooning MHD limit, α_CRIT, for the high-density reference case of Table XI. The value of α_MHD/α_CRIT is scaled to DIII-D upgrade and FPP by assuming the ITPA heat flux width scaling continues to hold and the separatrix density scaling with the square root of the parallel heat flux. This metric being above unity in DIII-D upgrade suggests that the machine would push into the region where turbulence broadening might set in and provide increase in the SOL heat width. It is important to test this physics process in DIII-D upgrade as it increases the heat flux width in FPP and will significantly affect FPP's divertor design and operation.

This ability to achieve several normalized parameters comparable to an FPP and to close gaps for other metrics, combined with comprehensive diagnostic measurements of the divertor, would enable testing and improvement of models of the physics processes expected to dominate the dissipative divertor regime in an FPP. In addition, the ability to achieve these metrics for a dissipative divertor in combination with the pedestal pressures, densities, and collisionalities discussed above would enable studies of the coupling of a dissipative divertor to a much more reactor-relevant pedestal than is now possible.

Q4: Will the Pedestal of an FPP Suffer Degradation from Divertor Detachment?

Contemporary machines often observe a substantial loss of pedestal pressure and temperature when the divertors are operated near or at detachment. Such pedestal degradation would be a major detriment to a reactor. Table XII shows show two physics processes that may be related to this phenomenon. By its nature, a detached divertor requires high density to achieve sufficient radiation to provide the needed dissipation. Useful metrics for high density are n_e^ped and f_GW. The high density may be providing high radiation inside the separatrix which reduces confinement by reducing the power flowing through the pedestal so that the pedestal P_SOL is too near the L–H power threshold to maintain a robust pedestal. Another metric is collisionality, since increased collisionality associated with the increased pedestal density and associated reduction of pedestal temperature can lead to significant reductions of bootstrap current. One characteristic of this phenomenology is that the pedestal is pushed further into the ballooning limited regime with reduced pedestal pressure at increased density.

However, the EPED model predicts that pedestals of DIII-D upgrade would be operating along the peeling branch rather than the ballooning branch and that increased density would be associated with an increase rather than a decrease in pedestal pressure. The collisionality is a useful metric to characterize and predict operation in the peeling (low) vs ballooning limited (high) regimes. An additional metric, p^ped, is useful for charactering the pedestal performance.

DIII-D upgrade is projected to achieve high pedestal density at low collisionality. This regime would allow the study of pedestal degradation at more reactor-relevant conditions than is possible in current machines. The relevant projected metrics are those of Table VII. These indicate that DIII-D upgrade (cases 1 and 2) could achieve densities comparable to FPP at collisionalities of unity or less. The corresponding pedestal pressures are very high and would partially close the gap between DIII-D and an FPP. The EPED model uses well-tested physics for the bootstrap current and peeling-ballooning physics in making its predictions and shows no evidence of pedestal degradation. Thus, operation of a dissipative divertor in DIII-D upgrade would provide useful information for projecting integrated operation of a dissipative divertor with a high-performance pedestal in an FPP.

A potential cause of pedestal and core confinement degradation is cooling of the X-point plasma due to divertor detachment. Propagation of the detachment front into the closed field line X-point region could potentially result in excessive radiative losses inside the separatrix and/or damping of the H-mode transport barrier due to neutral fluxes with associated ionization and charge-exchange losses. Recent analysis of DIII-D DTS data finds a maximum poloidal gradient of 200 eV/m across a range of detached divertor plasmas. These gradients are consistent with 2D fluid models of poloidal energy transport in the divertor dominated by both parallel and ExB convection. The gradients are too shallow for electron conduction to play a significant role for T_e ≤ 40 eV. This transport analysis further implies a minimum divertor leg length ≥40 cm to achieve at detached target plasma with T_e ∼ 1 eV while maintain a hot X-point plasma with T_e ≥ 80 eV. The staged divertor for DIII-D upgrade will allow for divertor leg lengths of ≥50 cm which will allow study of stability of the divertor detachment front and its role in X-point cooling and pedestal degradation. While the FPP will have a divertor leg length well beyond the poloidal T_e gradient scale length, DIII-D upgrade will have sufficient divertor leg length to examine these important physics issues.

Summary on Pedestal and Divertor Questions

Modeling studies have been performed to predict achievable metrics of the pedestal and divertor for upgrades to DIII-D of volume, current, and toroidal field. The metrics are chosen to study important physics questions for a fusion power plant. These metrics predict that a DIII-D upgrade could provide substantial advances in the metrics between DIII-D and an FPP and that the Upgrade could provide important information to reduce gaps in knowledge between current machines and an FPP.

This appendix documents the results of IPS-FASTRAN (Park , 2017) integrated modeling runs to predict achievable DIII-D fully non-inductive operation with parameters relevant to an FPP like the CAT-DEMO (Buttery , 2021). The simulations use theory-based integrated modeling with TGLF (Trapped ion Gyro-Landau-Fluid model) + EPED and are self-consistent with heating and current drive and the MHD equilibrium. A large set of full IPS-FASTRAN simulations was generated with random sampling in the multi-dimensional operation space. Then key output performance metrics such as β_N and others shown in the figures were parameterized as a function of the operational input parameters, allowing a systematic optimization of the target scenarios under given constraints.

Figure 16 shows the highest density cases as a function of q₉₅ at the fully non-inductive f_NI = 1 condition for a range of the power upgrade level, B_T, and shape/volume increase. The modeling indicates that the full upgrade option (Blue: includes up to 3 MW absorbed Helicon, 3 MW absorbed Lower Hybrid, 14 MW delivered ECH (20 multiple-frequency (∼110/140/170 GHz) gyrotrons, and up to 6 gyrotrons injected via top launch), 25 MW NBI (93 keV), high volume and shaping with Stage 1 and 3 Divertors, and 2.5 T operation) provides scope to test steady-state operation while coming close to or matching multiple FPP normalized parameters simultaneously, e.g., ν^*_ped near 0.1, f_BS = 0.5–0.8, β_N > 3.6, T_e/T_i ∼ 1, etc. There is flexibility to relax certain quantities to push more on others, for example, ν^* and f_GW.

Assessment of core physics dependence on normalized parameters is obtained simultaneously with achieving FPP-relevant high pedestal pressure and significantly reduced neutral penetration depth (estimated by Δ_cx/a) and energetic ion fraction (f_b).

For all upgrade options, the resistive diffusion time of the global current profile is greater than at present. This means sustained high-power plasmas can be close to, but not at the final relaxed state by the end of the flattop. This motivates using sophisticated profile controls to match the final predicted state as much as possible, as early as possible, in the discharge.

Figure 17 indicates the loss of capabilities compared to the full potential cases if some upgrades are not performed. Without Helicon and Lower Hybrid (dotted lines), steady-state operation is possible only at the upper range of q₉₅, and obtainable core performance (β_T ∼ β_N/q₉₅, p_ped, and Q ∼ β_NH/q₉₅²) is generally lower, but still relevant. Access to FPP (i.e., CAT-DEMO) normalized targets is more restricted, e.g., fast ion fraction and normalized neutral penetration depth only approach target values of ∼0.1 at the very highest q₉₅. This indicates the value of higher power Helicon and Lower Hybrid for electron heating and current drive, without injecting torque or creating fast ions. ∼30% lower ECH power (black compared to red) results in a shift in the wrong direction (i.e., away from FPP expected parameters) for all quantities, generally on the order of 30% or less.

If the shape is limited to the present lower κ, δ, and volume double-null shape (i.e., absent the expected new first diverter stage or third divertor stage—dashed lines), then steady-state operation with normalized core parameters similar to those with the full upgrade are possible, although over a much narrower range in q₉₅ and much reduced pedestal pressure. A less strongly shaped, smaller volume plasma does result in almost a factor of 2 lower obtainable pedestal pressure compared to the full upgrade case.

Figure 17 also shows the same intermediate upgrade cases as Fig. 16, but with an additional pair of solid-line curves that show the effects of running large highly shaped plasmas at lower B_T, 2.2 T instead of 2.5 T. At any particular q₉₅, lower B_T enables extension of relatively high pedestal pressures to higher β_N, f_BS, f_GW,ped, and lower fast ion fraction, at the cost of increased collisionality. Not upgrading NBI to be 93 keV capable (i.e., using 83 keV) results in ∼10% lower obtainable β_N, and ∼5% lower p_ped, when using B_T = 2.2 T and the high shaping/high volume (not shown).

Figure 18 shows that with the full upgrade, including 93 keV NBI, it is possible to test low-rotation, steady-state scenarios using the existing NBI angle capabilities. Many normalized parameters approach those in the CAT-DEMO study with high p_ped, T_e/T_i > 1, ν*_ped < 0.1, and low f_b < 0.1. β_N is limited to values near 3 with RF-dominated heating (20 MW RF and 12.5 MW NBI), but this is similar to or mimics the electron–ion heat mix from fusion alphas for an FPP. For all cases, the rotation, ω(0)-ω(pedestal) is less than 10 krad/s.