Machine · materials

Materials, neutronics, safety & environment.

What the machine is made of, the neutron environment those materials live in, and the safety and environmental case that follows — stated with conditions, not slogans.

Solve the blanket, live

neutronics · live

Change Li-6 enrichment, thickness, multiplier and coverage and watch net TBR recompute on the deposited engine — the surrogate OpenMC retires. Request the full Monte-Carlo run to HPC.

Open the blanket instrument →

Materials & Neutronics · Breeder

Materials & Neutronics

The materials, neutronics, activation and exhaust engineering behind Hyperion — with the hard problems (first-wall dpa, divertor radiative fraction) shown as plainly as the strengths.

Superconducting Magnets · REBCO

Peak field

16.84 T on-conductor

≤ 20.1 T limit

3.3 T margin to the [HAR23] conductor limit.

REBCO neutron lifetime

OPEN

the lab-free risk

Jc degrades above a fast-neutron fluence — the one risk lab magnets never face.

Advanced Structural & Conductor Materials

Conductor substrate

MP35N S_y ~1.7 GPa

Ni-Co superalloy

High-strength substrate carrying the winding-pack Lorentz load at 4.5 K.

Preload system

~2 GPa (ice-piston)

strain 0.233%

Water→20 K expansion preloads the pack, offsetting Lorentz stress; keeps REBCO strain under 0.4%.

Vessel candidate

CrMoNbV HEA

reduced-activation

High-entropy-alloy vessel candidate — strength plus reduced activation (development-stage).

High-temp armour

HfC / TaC ~4200 K

ultra-refractory

Ultra-high-temperature ceramics for the highest-flux armour zones.

Plasma-Facing Materials & First Wall

First-wall load

1.97 MW/m²

damage-life limited

14 MeV neutron load → 19.7 dpa/fpy → ~2.5 fpy component life.

Divertor PFC

W monoblocks

tungsten

Tungsten monoblock divertor; high melt point, high sputtering threshold.

Blanket & Tritium Breeding

Breeding ratio

1.8 TBR target

Li / Be scanned

TBR scanned over Li / Li₂O / FLiBe × enrichment × thickness; demonstrated Be 1.34.

Tritium output

4.0 kg/yr

~2× world flow

Net tritium at TBR 1.8; He-3 co-product ~1.97 kg/yr.

Neutronics

Neutron power

70.7 MW

f_n 0.797

14 MeV neutron power — the irradiation product that breeds tritium.

Activation & Waste

Waste class

≤ C with RAFM

Nb-controlled

RAFM Nb<10 ppm → SoF 0.464 (2.2× under Class C). Generic 316SS → GTCC (no US pathway).

Waste volume

264–661 t / 40 yr

34–85 m³

Activated steel, low(50 dpa) → high(20 dpa) range over plant life.

Exhaust · Divertor

Parallel heat flux

14.3 GW/m²

~14× ITER

Driven by the small major radius + high poloidal field — the same compactness that makes it cheap.

Required radiative frac

90 %

THE BINDING GATE

At the extreme edge of demonstrated detachment; NT narrowing pushes it to 92–94%.

ELM elimination

✓ NT edge

DEMONSTRATED

Negative-triangularity removes Type-I ELMs entirely (Thome 2024) — a real benefit; trades a transient problem for a steady one.

Magnet & tape deep-dive · REBCO deposit 10.5281/zenodo.21842514

Grounded in the 16-track REBCO magnet study + tape-conductor program. Every value carries its basis tag: SOURCED / PUBLIC-DATA, DESIGN-ASSUMPTION, REQUIREMENT, or REFUTED / INFEASIBLE.

Field ladder — plug requirement vs. demonstrated coils

REQUIREMENT + PUBLIC

Plug throat — REQUIREMENT (lower)

26.49 T

Plug throat — REQUIREMENT (upper)

39.74 T

WHAM HTS mirror (CFS, as-built)

17 T

SPARC TFMC (coil test, no plasma)

20.1 T

Frank/Realta end-plug (design study)

25 T

NHMFL all-SC DC record

32.35 T

Hahn "Little Big Coil" (conductor bound)

45.5 T

The plug field is a requirement pair, 26.49–39.74 T (never quote the lower alone). It sits above today's as-built fusion magnets (WHAM 17 T, SPARC TFMC 20.1 T) but within the envelope of record all-superconducting coils (32.35 T) and demonstrated conductor capability (45.5 T). Bore/radial build for the plug is a design assumption, not sourced.

REBCO tape — critical current J_c(B) at 4.2 K

PUBLIC-DATA

B = 5 T

17.59 MA/cm²

B = 10 T

10.83 MA/cm²

B = 15 T

8.15 MA/cm²

B = 20 T

6.66 MA/cm²

Superconductor-layer J_c from public tape data (power-law B-scaling from a 20 T anchor, 50 µm Cu / 1.5 µm REBCO in 100 µm tape). Current density falls with field — the plug operates deep on this curve, which is why the winding-pack cross-section (below) matters so much.

Winding-pack cross-section (λ breakdown)

SOURCED

REBCO (superconductor)

1.34 % carries all J_c

Silver overlayer

1.34 % cap / current-share

Copper stabilizer (2×20 µm)

26.81 % quench protection

Hastelloy C-276 substrate

33.51 % tape backbone (σ_y~1.2 GPa)

Buffer stack

0.13 % ~negligible area

MP35N co-wind

33.51 % structural reinforcement (σ_y~2 GPa)

Turn insulation

3.35 % thin oxide / polyimide

The superconductor is only 1.34% of the cross-section; two-thirds is structure (Hastelloy + MP35N co-wind) carrying the hoop load. This is why the magnet is a structural problem, not just a superconductor one.

Structural materials

SOURCED-CRYO + DESIGN

Material	Property	Value	T
MP35N (cold-rolled+aged)	0.2% yield	2500 MPa	77 K
MP35N	design allowable (this study)	2000 MPa	20 K
MP35N	Young's modulus	230 GPa	77 K
Hastelloy C-276	substrate yield	~1200 MPa	cryo
Carbon-fiber overband	tensile (fiber dir.)	2000 MPa	77 K
Carbon-fiber overband	modulus (fiber dir.)	150 GPa	77 K

Bore-resolved static stress @ 26.49 T — refutes the field-only scaling

COMPUTED vs REFUTED

bore 0.02 m

87.4 MPa

bore 0.03 m

131.1 MPa

bore 0.05 m

218.5 MPa

bore 0.07 m

306 MPa

bore 0.08 m

349.7 MPa

Bore-resolved hoop stress (87–350 MPa across the swept bore) sits well under the reinforced allowable (1091 MPa) and the MP35N single-material allowable (2000 MPa). This refutes the naïve field-only scaling that predicted ~1047 MPa and looked marginal. Static structure is feasible; the binding problem is cyclic (next).

Cyclic fatigue at operating field — the binding gate

INFEASIBLE

Case	σ_max	N to failure	Design cycles	Verdict
26.49 T, no preload	3289 MPa	2.3×10⁻³	10,000	INFEASIBLE
26.49 T, engineered preload	2916 MPa	9.3×10⁻²	10,000	INFEASIBLE
39.74 T, engineered preload	3881 MPa	3.9×10⁻⁵	10,000	INFEASIBLE

Layer-resolved FEA hoop stress at operating field exceeds the fatigue allowable — the coil would not survive its cycle count. This is stated as a gate, not a solved result: it needs a full bore-resolved winding-pack FEA and a fatigue-qualified reinforcement/preload scheme. It is the #1 magnet open item on the Open Risks ↗ page.

Magnet neutron lifetime vs. tungsten shield

DESIGN-STUDY

no shield

0.023 FPY

5 cm W

0.067 FPY

10 cm W

0.194 FPY

12.5 cm W

0.33 FPY

15 cm W

0.562 FPY

Even at 15 cm of tungsten the magnet reaches only ~0.56 full-power-years — below the 2 FPY replacement target (gold line = target). Shield optimisation and advanced-pinning (APC) conductor are needed; confirm with neutron-transport (MCNP) runs. An open lifetime item.

Quench protection

SELF-PROTECTING

Case	Stored E	T_max	V_terminal	Self-protecting
26.49 T	2.35 MJ	145 K	405 V	YES
39.74 T	39.4 MJ	109 K	455 V	YES

The VOx metal-insulation approach is self-protecting (peak temperatures well below damage limits), benchmarked against Suetomi 2021 (model reproduces the 330 K reference). Good news among the harder magnet items.

Materials qualification — where each stands

HONEST STATUS

Item	Status	Note
REBCO tape	Commercial	J_c well-characterised; fusion-fluence lifetime open
MP35N reinforcement	Qualified (cryo)	Cryogenic yield sourced
Hastelloy C-276	Commercial	Standard tape substrate
Tungsten PFC / shield	Established	PFC-grade W; shield thickness to be optimised
Plug winding-pack (fatigue)	NOT qualified — GATE	Needs bore-resolved FEA + fatigue design
Neutron shield adequacy	Design study	MCNP confirmation pending

Breeder deposit · DOI 10.5281/zenodo.21746157 · reproduced from the open Zenodo/GitHub deposit (CC BY 4.0). Canonical freeze: FROZEN_2026-08_PUBLICATION.csv. · Materials/neutronics screening; UNCOMPUTED terms named, not estimated.

🔒 Materials & Neutronics is Kronos-only. The overview above is public — the full data, tables, live calculations and engineering detail on this page require a Kronos team sign-in.

Safety case · Breeder · Hyperion

Safety Case

Preliminary hazard analysis for the frozen design — failure modes (FMEA), process deviations (HAZOP) and the radiological picture (activation, decay heat, waste). Fusion's inherent safety is the backdrop: no chain reaction, no criticality, no meltdown path, and decay heat orders of magnitude below fission.

⚠ Preliminary — screening level. This is a design-stage hazard analysis, not a licensed safety case. FMEA/HAZOP entries are representative and await a full facilitated study; the activation inventory awaits a FISPACT-II run. It states the hazards honestly rather than asserting compliance.

Criticality risk

NONE

no chain reaction · no meltdown path

FMEA items

0 at RPN ≥ 100

Decay heat vs fission

≪ 1%

vs ~7% · passively removable

Waste class

Class C LLW

no HLW · no actinides

FMEA — failure modes & effects

RPN = Severity × Occurrence × Detection (1–10 each) · higher = more risk · click a row for detail

Item	Failure mode	Effect	S	O	D	RPN	Requirement
TF magnet	Quench	Field loss → plasma benignly terminates; possible coil damage	7	3	4	84	REQ-B09 ↗
Plasma	Major disruption	EM + thermal transient on first wall & vessel	6	4	4	96	REQ-B08 ↗
Divertor coolant	Loss of flow	Target overheating → PFC erosion	6	3	3	54	REQ-B07 ↗
Tritium plant	Containment breach	Tritium release — radiological	8	2	5	80	REQ-B02 ↗
Vacuum vessel	Air ingress	Plasma disrupts (benign) + oxidation risk	5	2	3	30	REQ-B08 ↗
Cryoplant	Helium release	Oxygen-deficiency (ODH) in the hall	6	2	4	48	—

HAZOP — guideword deviations

Guideword	Parameter	Cause	Consequence	Safeguard
NO	Coolant flow	Pump trip / blockage	PFC & first-wall overheat	Interlock trip + passive cooling loop
MORE	Coolant pressure	Blockage / thermal transient	Pipe rupture	Relief valves + design pressure rating
LESS	Chamber vacuum	Leak / seal failure	Plasma disruption (benign)	Fast shutdown + isolation valves
MORE	Magnet temperature	Cooling loss	Quench	Detection + stored-energy dump
AS WELL AS	Tritium in coolant	Permeation	Secondary contamination	Detritiation loop + monitoring
MORE	Plasma density	Control fault	Disruption	Density-limit control + mitigation valves

Radiological — activation, decay heat & waste

Dominant hazard

Tritium

on-site inventory · multi-barrier

Activated structure

RAFM steel

Class C LLW · recyclable ~100 yr

Long-lived actinides

NONE

not a fission fuel cycle

Neutron wall load

1.97 MW/m²

14 MeV D–T dominated

Decay heat after shutdown representative

Radiological picture

Criticality / meltdownnot physically possible

Decay heat at shutdown~1% of P_fus (fission ~7%)

Passive heat removalsufficient — no active cooling needed to stay safe

Waste classificationClass C low-level

Stored magnet energyconventional industrial hazard — quench-protected

FISPACT-II nuclide inventoryUNCOMPUTED — needs the activation run

The breeder concentrates the radiological burden at one isolatable, remotely-sited machine — tritium handling is the design-driving hazard, addressed by multiple confinement barriers.

Preliminary hazard analysis over the frozen design · decay-heat curve is representative (needs a FISPACT-II / decay-heat run) · FMEA & HAZOP await a facilitated study. PRELIMINARY · NOT A LICENSED SAFETY CASE

🔒 Safety Case is Kronos-only. The overview above is public — the full data, tables, live calculations and engineering detail on this page require a Kronos team sign-in.

Environmental · screening-level LCA (ISO 14040/44)

Environmental Profile

A comparative, honest life-cycle assessment of the two-machine fleet against the technologies it displaces and against other fusion cycles. Screening-grade where noted; terms not yet computed are named, not estimated.

Lifecycle carbon · gCO₂eq/kWh (IPCC AR5 WG3)

Coal820 g

Natural gas490 g

Biomass230 g

Solar PV48 g

Hydro24 g

Nuclear12 g

Wind onshore11 g

Kronos burner12 g

— operational only~0 op

IPCC AR5 WG3 median lifecycle emissions. Fusion has no combustion → operational ~0; embodied is nuclear/wind class.

Neutron cleanliness · % of fusion energy carried by neutrons

D–T tokamak (ITER/DEMO)80 %

Kronos breeder (product)79.7 %

Kronos burner · x 0.305.44 %

Kronos burner · x 0.432.77 %

p–B¹¹ (ideal aneutronic)0.5 %

The burner is low-neutron (5.44%), not aneutronic — ~10–15× cleaner than a mainstream D–T tokamak, not as clean as ideal p–B¹¹. A free clean-shift to x 0.35 reaches 4.18% and stays net-positive. The breeder is deliberately neutron-rich — its 14 MeV neutrons are the product.

Safety · deaths per TWh (Our World in Data)

Coal

24.6

deaths/TWh

Natural gas

2.8

deaths/TWh

Nuclear

0.03

deaths/TWh

Wind / solar

~0.03

deaths/TWh

Radioactive waste

Breeder (Hyperion) ≤ Class C

with low-activation RAFM steel

RAFM Nb<10ppmSoF 0.464 (2.2× under)

RAFM Nb<1ppmSoF 0.087 (12× under)

generic 316SSGTCC — no US pathway

volume264–661 t · 34–85 m³ / 40 yr

Burner (Aegis/MetroVolt) ~ Class A

~0 scheduled activated waste

first-wall changes0.035–0.144 / 30 yr

scheduled stream~0 (negligible)

wall loading41–168× below breeder

REBCO plug coilsUNCOMPUTED

Fleet strategy · concentrate & isolate vs distribute clean

The fleet concentrates its unavoidable radiological cost at one isolatable, remotely-sited breeder, while distributing clean, low-neutron burners to the point of use. Radiological burden is centralized and contained; generation is clean and distributed — the fleet is ~10–15× cleaner in neutron terms than a mainstream D–T tokamak, and net-favourable versus the incumbents it displaces on carbon, land, waste longevity and safety.

Embodied carbon & materials

Burner

12 gCO₂eq/kWh (5–30)

nuclear/wind class; construction 60–70% of emissions

Breeder

~13 tCO₂ / kg-T

embodied screening; operational is grid-dependent (25.9 MWe draw)

Driver

steel + concrete + magnets

same finding across fusion LCA studies

Stated honestly · not yet computed

We name what we have not computed rather than estimate beyond the evidence:

Absolute D–D tritium production at commercial scaleUNCOMPUTED

Full activation curie inventory & decay heat (needs FISPACT on a certified heat)UNCOMPUTED

Embodied carbon at bill-of-materials fidelityUNCOMPUTED

Tritiated waste volume (HTO) — needs detritiation designUNCOMPUTED

Burner REBCO plug-coil activated-magnet stream (plug neutron lifetime)UNCOMPUTED

Screening-level LCA (ISO 14040/44). Comparators: IPCC AR5 WG3 Annex III · Our World in Data · fusion LCA (Tokimatsu et al.). Every burner figure carries the plug-density requirement caveat.

🔒 Environmental Profile is Kronos-only. The overview above is public — the full data, tables, live calculations and engineering detail on this page require a Kronos team sign-in.

KRONOS FUSION ENERGY · Kronos Model Conceptual design and simulation study; no machine has been built