CONFERENCE PRE-PRINT DEVELOPMENT OF HIGH POLOIDAL BETA SCENARIO FOR LONG-PULSE OPERATION IN COLLABORATION BETWEEN DIII-D AND KSTAR Y.M. JEON Korea Institute of Fusion Energy Daejeon, Republic of Korea Email: [email protected] S. DING, H.Q. WANG, A.M. GAROFALO, D. ELDON General Atomics San Diego, USA Q.M. HU, S.K. KIM Princeton Plasma Physics Laboratory Princeton, USA J.M. PARK Oak Ridge National Laboratory Oak Ridge, USA K. KWON Oak Ridge Associated Laboratory Oak Ridge, USA J.M. LEE, Y.H. Lee, H.S. Kim Korea Institute of Fusion Energy Daejeon, Republic of Korea The development of high poloidal beta (βₚ) scenarios is crucial for steady-state tokamak operation due to their large bootstrap current fraction and favourable confinement. Building on recent progress in DIII-D, a joint DIII-D/KSTAR effort has established a new high-βₚ scenario under KSTAR-like operational constraints. In DIII-D, stable operation with H98 ~ 1.5, βₚ ≥ 3.0, and fBS ≥ 0.5 was achieved through the formation of a large-radius internal transport barrier (ITB), even with limited heating power, slow current ramping, and delayed shaping. These results demonstrate the feasibility of high-density operation near the Greenwald limit (fGW ≥ 0.9) with full divertor detachment. Initial implementation of this scenario in KSTAR confirmed large-radius ITB formation (ρ ~ 0.5), providing ~30% performance improvement despite tungsten impurity accumulation and restricted neutral beam power. Although overall confinement remained limited, these findings underscore both the opportunities and challenges of applying high-βₚ scenarios to superconducting long-pulse devices with tungsten divertors. Future work will focus on optimizing current profile evolution, mitigating impurity effects, and extending high-βₚ operation toward reactor-relevant conditions for DEMO and pilot plants. Steady-state, long-pulse tokamak operation remains a central challenge in magnetic confinement fusion research. High-βₚ operation, characterized by high bootstrap fraction and favourable confinement, is considered one of the most attractive scenarios for next-generation devices such as DEMO and fusion pilot plants (FPPs). KSTAR has recently demonstrated a 100-second high-performance long-pulse operation with a tungsten divertor, marking a milestone in advanced tokamak research. Parallel efforts at DIII-D have shown that high-βₚ plasmas with large-radius ITBs can simultaneously achieve high confinement (H98y2 ≥ 1.5), density above the Greenwald limit, and improved core-edge integration. Bridging these achievements between KSTAR and DIII-D is critical for developing reactor-relevant, reproducible scenarios. This paper provides a detailed account of these developments, highlighting the technical challenges, physics insights, and potential implications for future fusion devices. By elaborating on both the KSTAR and DIII-D perspectives, we aim to provide a comprehensive reference for the research community. Since the first high-βₚ discharge in 2015, KSTAR has pursued long-pulse operation based on this scenario. These efforts, combined with the recent tungsten divertor upgrade, culminated in the first successful demonstration of a high-performance plasma sustained for ~100 s (see Fig. 1b). KSTAR high-βₚ discharges exhibit distinct features compared with the well-known DIII-D high-βₚ regime. In particular, operation at high q95 (≥7.0) yields high βₚ (≥3.0), resulting in large bootstrap current fractions favourable for long-pulse operation. Uniquely, however, central ECH plays a critical role. A representative discharge is shown in Fig. 1a. At q95 ~11, with 5.1 MW of NBI heating and 0.5 MW of central ECH, the plasma was stably maintained in a fully non-inductive state with βₚ > 3.0. Figure 1. (a) A representative KSTAR high-βₚ discharge; (b) ~100-s long-pulse high-βₚ discharge. The critical role of central ECH is highlighted in Fig. 2. Figure 2a compares discharges with and without ECH, showing performance enhancements of ~67% in total stored energy and ~75% in βₚ when ECH is applied. Figure 2b demonstrates the plasma response when ECH is switched off: a substantial performance degradation occurs, though on a surprisingly long-time scale (~2.0 s), suggesting underlying physics mechanisms that require further investigation. Figure 2. The important role of central ECH in high-βₚ discharges Another major difference from DIII-D is that ITBs have not been observed in KSTAR high-βₚ plasmas. The observed performance improvements therefore arise from mechanisms other than ITB formation. Indirect evidence points to enhanced fast-ion confinement and beam absorption. Figure 3 (left) shows kinetic profiles of ion temperature, electron temperature, electron density, and toroidal rotation, indicating no ITB structure. However, the right-hand plots reveal a dramatic increase in absorbed NBI power and corresponding thermal energy in high-βₚ discharges. TRANSP analysis suggests that, while the anomalous diffusion coefficient of fast ions in standard H-mode plasmas is ~0.4, it approaches zero in high-βₚ plasmas, consistent with neoclassical transport. These results indicate that central ECH significantly improves NBI power absorption and thereby boosts plasma performance. Figure 3. Kinetic profiles of high-βₚ discharges and TRANSP analysis In summary, the KSTAR high-βₚ scenario exhibits distinct characteristics from conventional high-βₚ operation. By achieving high βₚ and sustaining large bootstrap currents, it enables long-pulse operation, culminating in the successful demonstration of ~100-s discharges. The next step for KSTAR is to advance this high-βₚ scenario toward more reactor-relevant conditions. Specifically, efforts are directed toward reproducing the DIII-D high-βₚ regime that couples large-radius ITB formation with higher density operation, enabling improved confinement (H98y2 ≥ 1.5) under conditions applicable to future reactors. The most significant feature of DIII-D high-βₚ discharges is the formation of large-radius internal transport barriers (ITBs), which enable superior confinement performance. Importantly, such high-performance plasmas can be sustained at densities exceeding the Greenwald limit, making this scenario highly relevant for DEMO and future fusion pilot plants (FPPs). These advantages are illustrated in Figure 4. Figure 4a shows the correlation between plasma density and thermal energy confinement using DIII-D data. For non-high-βₚ discharges (shown in yellow), confinement clearly decreases with increasing density, falling sharply as fGW approaches unity. In contrast, high-βₚ discharges maintain strong confinement (H98y2 ~1.5) even at high density. Figure 4b further highlights this distinction: low-q95 H-mode plasmas exhibit severe degradation at high density, while high-βₚ plasmas display the opposite trend, with confinement improving as density increases. These results demonstrate that achieving higher density is a key requirement for accessing the high-βₚ regime and naturally facilitates divertor detachment. Figure 4. (a) Correlation between plasma density and H98y2; (b) contrasting trends by operational scenario Understanding the key differences between the KSTAR and DIII-D high-βₚ scenarios is essential for designing experiments and addressing implementation challenges. First, with respect to ITB formation, KSTAR relies on central electron cyclotron heating (ECH) under standard H-mode conditions. This produces a peaked current density profile early in the discharge, characterized by high li and low qmin (~1.0). In contrast, DIII-D requires a broad current profile for ITB formation, with low li and qmin > 2.0. Achieving such conditions necessitates fast Ip ramp-up, early shaping, and early L/H transition. Second, KSTAR typically accesses the high-βₚ regime at relatively low plasma density (fGW < 0.5), whereas DIII-D favours high-density operation (fGW > 0.8), which not only improves confinement but also supports divertor detachment. In addition, differences in plasma shaping, aspect ratio (R/a), and available heating power must be considered. To bridge these differences, strategic experiments were first carried out in DIII-D under KSTAR-like constraints. The optimized scenario was then transferred to KSTAR for validation. The high-βₚ scenario has been developed and validated extensively in DIII-D, and successful implementation was also achieved under KSTAR-like constraints. Figure 5a shows the time evolution of a representative discharge, where a broad current profile with qmin > 2.0 was established and maintained. This configuration enabled stable high performance with βN ~ 3.0, βₚ > 3.0, and H98y2 ~ 1.5 at high plasma density (fGW ~ 0.9). As shown in Figure 5b, strong large-radius ITBs (ρ = 0.6–0.7) were formed in all profiles. Figure 5. (a) A new high-betap discharge on DIII-D with KSTAR-like constraints. (b) Strong ITBs with large-radius are formed in all profiles. In addition, Figure 6 demonstrates that full divertor detachment was successfully achieved through neon seeding with only a modest reduction in performance (10–20%). The right-hand plots clearly indicate detachment in divertor current, temperature, and pressure following neon injection. Figure 6. Full divertor detachments by neon seeding A new high-βₚ scenario, optimized in DIII-D, was tested in KSTAR. Initial experiments, though limited, yielded encouraging results. Figure 7 shows a representative case. Up to 2 s, only pre-heating was applied to form a broad current profile (high qmin and low li), followed by main heating to increase density and βN. Plasma density was brought near the Greenwald limit, facilitating ITB formation. qmin reached ~3.0, but li remained high (~0.9), leading to performance degradation typical of non-high-βₚ discharges. Moreover, performance was sensitive to gas puffing, underscoring the need for optimized density control. An important physics insight from DIII-D high-βₚ experiments is the role of magnetic shear and the normalized pressure gradient. Access requires bypassing the unstable hill and reaching a secondary stable region. As shown in Figure 8, transiently reducing magnetic shear at low pressure gradient opens this pathway. In KSTAR, a second Ip ramp-up was introduced to alter plasma evolution, enabling marginal ITB conditions. Figure 7. Initial result of the new high-βP scenario on KSTAR Figure 8. Importance of magnetic shear in high-βP access Figure 9. (a) Weak ITB formation triggered by perturbations; (b) two-step performance enhancement through ITB and pedestal improvement Figure 9a shows that perturbations such as NBI blips can trigger ITB formation even under marginal conditions, leading to performance gains associated with ion ITB formation. Figure 9b illustrates that these gains occurred in two stages: first, ITB formation in ion temperature, followed by pedestal enhancement. Together, these produced ~25% βN improvement at constant Ip, BT, and heating power. The results show that while DIII-D successfully developed high-βₚ scenarios with large-radius ITBs under KSTAR-like constraints, implementation in KSTAR remains challenging. Key issues include establishing broad current profiles in superconducting operation, controlling tungsten accumulation, and achieving high-density operation while sustaining ITBs. Nevertheless, weak ITB formation and pedestal improvement in KSTAR suggest a viable path forward. Further experiments should focus on optimizing magnetic shear evolution, employing real-time profile control, and testing advanced impurity mitigation. KSTAR successfully demonstrated ~100 s long-pulse operation under tungsten environment using high-βₚ scenario. DIII-D established large-radius ITB-coupled high-βₚ regimes under KSTAR-like constraints, with strong confinement and density performance. Initial KSTAR experiments implementing this scenario showed encouraging, albeit challenging, results: weak ITB triggered by NBI perturbation and performance improvements via pedestal enhancement. Continued joint optimization is required to resolve access issues and extend reactor-relevant high-βₚ operation toward DEMO/FPP needs. This study underscores the necessity of international collaboration and the cross-validation of advanced tokamak scenarios. The pathway toward steady-state operation in future reactors will rely on iterative advances across multiple devices, supported by coordinated theory, modelling, and experimental campaigns. ACKNOWLEDGEMENTS This research was supported by the R&D Program “High Performance Tokamak Plasma Research & Development (code No. EN2501)” and “Korea–US Collaboration Research for High Performance Plasma on Tungsten Divertor (code No. EN2503)” through the Korea Institute of Fusion Energy (KFE), funded by the Government of the Republic of Korea. Additional support was provided by the U.S. Department of Energy under Award Numbers DE-FC02-04ER54698, DE-AC02-09CH11466, DE-SC0010685, DE-SC0023399, and DE-AC05-00OR22725. REFERENCES [1] H.S. Kim, et al., Proc. 2nd IAEA-TM on Long-Pulse Operation of Fusion Devices, Vienna, October 2024. [2] S. Ding, et al., Nature, 629, pp. 555–560 (2024). [3] S. Ding, et al., Phys. Plasmas, 32, 022502 (2025). [4] Y. Wang, et al., Nat. Commun., 2021.