ML22353A101

Scale/Melcor non-LWR Source Term Demonstration Project - Molten Salt Reactor (MSR)

ML22353A101
Person / Time
Issue date:	09/13/2022
From:	NRC/NRR/DANU, Oak Ridge, Sandia
To:
References
Download: ML22353A101 (1)

Text

SAND2022-12146 PE SCALE/MELCOR Non-LWR Source Term Demonstration Project - Molten Salt Reactor (MSR)

September 13, 2022 Sandia National Laboratories is a multimission laboratory managed and operated by National Technology and Engineering Solutions of Sandia LLC, a wholly owned subsidiary of Honeywell International Inc. for the U.S. Department of Energys National Nuclear Security Administration under contract DE-NA0003525. SAND20XX-XXXX P

Outline NRC strategy for non-LWR source term analysis Project scope Overview of Molten Salt Reactor (MSR)

MSR reactor fission product inventory/decay heat methods & results MELCOR molten salt models MSR plant model and source term analysis Summary 2

Integrated Action Plan (IAP) for Advanced Reactors Near-Term Implementation Action Plan Strategy 1 Strategy 4 Knowledge, Skills, Industry Codes and Capacity and Standards Strategy 5 Strategy 2 Technology Analytical Tools Inclusive Issues ML17165A069 Strategy 3 Strategy 6 Flexible Review Communication Process 3

IAP Strategy 2 Volumes These Volumes outline the specific analytical tools to enable independent analysis of non-LWRs, gaps in code capabilities and data, V&V needs and code development tasks.

Introduction Volume 1 ML20030A174 ML20030A176 Volume 3 Volume 2 Volume 4 Volume 5 ML20030A177 ML21085A484 ML21088A047 ML20030A178 4

NRC strategy for non-LWR analysis (Volume 3) 5

Role of NRC severe accident codes 6

Project Scope Project objectives Understand severe accident behavior

• Provide insights for regulatory guidance Facilitate dialogue on staffs approach for source term Demonstrate use of SCALE and MELCOR

• Identify accident characteristics and uncertainties affecting source term

• Develop publicly available input models for representative designs 8

Project scope Full-plant models and sample calculations for representative non-LWRs 2021

• Heat pipe reactor - INL Design A

• Pebble-bed gas-cooled reactor - PBMR-400

• Pebble-bed molten-salt-cooled - UCB Mark 1

• Public workshop videos, slides, reports at advanced reactor source term webpage 2022

• Molten-salt-fueled reactor - MSRE - public workshop 9/13/2022

• Sodium-cooled fast reactor - ABTR - public workshop 9/20/2022 2023

• Additional code enhancements and sample calculations 9

Project approach

1. Build SCALE core model and MELCOR full-plant model

2. Select scenarios that demonstrate code capabilities

3. Perform simulations

• Use SCALE to model decay heat, core radionuclide inventory, and reactivity feedback

• Use MELCOR to model accident progression and source term

• Perform sensitivity cases 10

Molten Salt Reactor (MSR)

Molten-salt reactor history (1/2)

Aircraft Nuclear Propulsion Program (ANP) - 1946-1961

• Long-term strategic bomber operation using nuclear power

• ORNL developed the nuclear concept with the Aircraft Reactor Experiment (ARE)

Originally sodium cooled, but shifted to molten salt 2.5 MW molten salt-cooled reactor operated for 96-MW-hours in November 1954

• Three Heat Transfer Reactor Experiments at Idaho National Laboratory to demonstrate the jet engine propulsion

• Aircraft Shield Test (AFT) - B-36 with an operating reactor flew 47 times over West Texas and New Mexico to study shielding (i.e., the reactor was operating but not part of the propulsion system)

• Terminated due to inventing ballistic missile and supersonic aviation Heat Transfer Reactor Experiment #3

ia/File:HTRE-3.jpg

The B-36 Aircraft Shield Test

[1] 12

Molten-salt reactors history (2/2)

ORNL Molten Salt Reactor Experiment (MSRE)

• AEC funded

• Operated from 1965 to 1969

• 10 MWth

• Used for SCALE MELCOR source term demonstration calculations MSRE Graphite Core Structure

[2]

MSRE

[ORNL-TM-0728]

13

MSRE (1/5)

Reactor

• 10 MWth

• Reactor consists of a graphite core structure (see photo on previous slide)

• Fuel dissolved in the molten salt coolant fissions when it passes through the graphite core structure

• Graphite provides moderation

• 0.075 m3/s (1200 gpm) core flowrate

• 635 core inlet

• 668 core outlet

• Near atmospheric pressure in the helium above the salt

• Coolant included variations of lithium, beryllium, and zirconium fluoride salts that contain uranium, or uranium and thorium fluorides

• INOR-8 nickel-based alloy vessel MSRE vessel

[ORNL-TM-0728]

14

MSRE (2/5)

Coolant salt circulation

• Primary loop with pump and heat exchanger

• Intermediate loop with pump and air-cooled radiator

• No fuel in intermediate loop Air-cooled radiator rejects heat to the plant stack MSRE primary heat exchanger MSRE schematic

[3] [ORNL-TM-0728] 15

MSRE (3/5)

Reactor Cell acts as containment

• Contains the reactor vessel, the primary circulating fuel loop, and most of the coolant salt loop

• Circulating salt to air-cooled radiators located outside of the reactor cell

• 95% N2

• 0.875 bar absolute

• 320 m3

• Leak rate = 0.42 standard Reactor Building Reactor Cell liters per hour at 0.875 bar (12.7 psia) (0.23 mm dia.)

• Attached by a tunnel to the drain tank cell MSRE Reactor Cell

[ORNL-TM-0728] Drain Tank Cell 16

MSRE (4/5)

Vapor-condensing system

• Connects to the reactor cell via a 30 pipe

• Normally isolated from the stack with 2 rupture disks 30.5 cm (12) line with 1.38 bar (20 psig) rupture disk 10 cm (4) line with 1.03 bar (15 psig) rupture disk

• Condensing tank with 34 m3 (1200 ft3) of water

• Gas retention tank 93 m3 (3300 ft3)

• 5 cm (2) line to the filters and the stack From the Reactor Cell MSRE vapor-condensing system

[ORNL-TM-0728]

17

MSRE (5/5)

Off-gas filtration system

• Large network that includes 6000 liter/day helium flow through the primary and secondary pump bowls

• Pump bowl helium effluent connects to a series of holdup volumes (large volume &

low flow) inside and outside the reactor cell

• 2 filter trains with 0.623 m3 (22 ft3) of charcoal Pump Bowl One train typically isolated Auxiliary charcoal filter for reactor cell venting

• 3x32 m2 (3x350 ft2) fiberglass roughing filters 90-95% efficiency for dust

• 3x2.23 m2 (3x24 ft2) HEPA absolute filters with 99.7%

MSRE off-gas system efficiency for 0.3 micron particles [ORNL-TM-0732]

• Filtered flow merges with 9.9 m3/s (21,000 cfm) building HVAC out the plant stack for dilution 18

SCALE Molten Salt Reactor Inventory, Decay Heat, Power, and Reactivity Methods and Results Sandia National Laboratories is a multimission laboratory managed and operated by National Technology and Engineering Solutions of Sandia LLC, a wholly owned subsidiary of Honeywell International Inc. for the U.S. Department of Energys National Nuclear Security Administration under contract DE-NA0003525. SAND20XX-XXXX P

NRC SCALE/MELCOR Non-LWR Demonstration Project Objectives:

• Develop approach and models for SCALE analysis to obtain:

• Radionuclide inventory

• System decay heat

• Power profiles

• Reactivity coefficients Key differences to LWR analysis:

• Continuous circulation of the fuel

• Consideration of both core and loop

• Nuclide removal in loop Approach:

• Generate system fuel salt composition considering continuous circulation of the fuel salt and nuclide removal in the loop

• Investigate location-dependent fuel salt inventory in the system SCALE MSRE core model

• Evaluate neutronic characteristics at specific point in time 20

Workflow SCALE specific Generic End-user specific SCALE Inventory Other Binary Output Interface File MACCS Input SCALE Power SCALE Text Distributions MELCOR Input Output Kinetics Data

• SCALE capabilities used:

• Codes:

• Sequences:

ORIGEN for depletion CSAS for criticality/reactivity KENO-VI 3D Monte Carlo neutron transport TRITON for reactor physics & depletion

• Data: ENDF/B-VII.1 nuclear data library*

• The recently published NUREG/CR-7289 Nuclear Data Assessment for Advanced Reactors details the impact of the nuclear data library on non-LWR reactor physics calculations. 21

MSRE Model Description Description Value Power 10 MWth (initial criticality) / 8 MWth (during operation)

Fuel/coolant LiF-BeF2-ZrF2-UF2 Enrichment 34.5 wt.% 235U Moderator Graphite Structure Nickel-based alloys Core volume 0.7 m3 System volume 2 m3 Heavy metal loading 0.233 tHM Loop transit time 25.2 seconds

• Noble gases via Off-Gas System (OGS)

Nuclide removal

• Noble metal plate-out at heat exchanger (HX)

Re-fueling Irregular re-fueling by capsules with HEU fuel salt Operating time ~375 equivalent full-power days with 235U fuel Basis for core model development: Zero-power first critical experiment with 235U from the OECD/NEA International Handbook of Reactor Physics Experiments [2]

[1] R. C. Robertson (1965), MSRE Design and Operations Report Part I: Description of Reactor Design, ORNL-TM-0728, ORNL.

[2] M. Fratoni, et al. (2020), Molten Salt Reactor Experiment Benchmark Evaluation, DOE-UCB-8542, 16-10240, UC Berkeley, doi:10. 2172/1617123 MSRE reactor vessel [1]

22

SCALE analysis approach Time-dependent inventory Core

+

loop

• Considers core + loop +

off-gas + plating-out Xe, Kr Se, Nb, etc.

• Predicts system-average inventory over time OGS HX System-average System-average inventory at point in time inventory at point in time Location-dependent Core power/flux inventory in loop distribution

• Considers power profile and

• Predicts neutron flux and off-gas Power/flux power profiles at point in time

• Predicts inventory in each profile region of the loop 23

SCALE analysis approach Time-dependent inventory Core

+

loop

• Considers core + loop +

off-gas + plating-out Xe, Kr Se, Nb, etc.

• Predicts system-average inventory over time OGS HX System-average System-average inventory at point in time inventory at point in time Sensitivity study:

Region-dependent nuclide inventory Location-dependent Core power/flux inventory in loop distribution

• Considers power profile and

• Predicts neutron flux and off-gas Power/flux power profiles at point in time

• Predicts inventory in each profile region of the loop 24

SCALE analysis approach Time-dependent inventory Core

+

loop

• Considers core + loop +

off-gas + plating-out Xe, Kr Se, Nb, etc.

• Predicts system-average inventory over time OGS HX Consistency System-average System-average assessment on inventory at point in time inventory at point in time removal rates Location-dependent Core power/flux inventory in loop distribution

• Considers power profile and

• Predicts neutron flux and off-gas Power/flux power profiles at point in time

• Predicts inventory in each profile region of the loop 25

SCALE analysis approach System-average + OGS Time-dependent inventory Core

+

inventory and decay loop heat, and removal rates

• Considers core + loop +

for MELCOR off-gas + plating-out Xe, Kr Se, Nb, etc.

Power profiles and

• Predicts system-average temperature-dependent inventory over time OGS HX Region-dependent reactivity coefficients inventory and decay heat for MELCOR System-average System-average inventory at point in time inventory at point in time Location-dependent Core power/flux inventory in loop distribution

• Considers power profile and

• Predicts neutron flux and off-gas Power/flux power profiles at point in time

• Predicts inventory in each profile region of the loop 26

SCALE MSRE full core model Sandia National Laboratories is a multimission laboratory managed and operated by National Technology and Engineering Solutions of Sandia LLC, a wholly owned subsidiary of Honeywell International Inc. for the U.S. Department of Energys National Nuclear Security Administration under contract DE-NA0003525. SAND20XX-XXXX P

MSRE full core model TRITON-KENO model based on IRPhEP benchmark specifications fuel fuel graphite fuel fuel Cross section of YZ-cut through SCALE 3D model XY-cut through SCALE 3D model graphite stringer 28

Time-dependent inventory Sandia National Laboratories is a multimission laboratory managed and operated by National Technology and Engineering Solutions of Sandia LLC, a wholly owned subsidiary of Honeywell International Inc. for the U.S. Department of Energys National Nuclear Security Administration under contract DE-NA0003525. SAND20XX-XXXX P

Time-dependent inventory - model development Goal: Generate system-average (fuel salt in core+loop) nuclide inventory at end of operation Model: TRITON-KENO core slice model

• Representative spectral conditions through radial leakage and representative moderator-to-fuel ratio, while allowing shorter runtimes compared to full core

• Depletion up to 375 days, the total operation time of MSRE with 235U fuel

• Representation of system (core+loop) through adjusted power level:

Core power 8 MWth, total mass of 0.218 tHM in the system Specific power of 36.697 MW/tHM

• Consideration of nuclide removal through TRITON-MSR 1,2 (next slide)

[1] B. R. Betzler, al., Molten salt reactor fuel depletion tools in SCALE, Proc. Global/Top Fuel, Seattle, WA, September 22-27, 2019.

SCALE 2D slice model

[2] P. J. V. Valdez, et al., Modeling Molten Salt Reactor Fission Product Removal with SCALE, ORNL/TM-2019/1418, 2020.

30

Time-dependent inventory - nuclide removal tank/

OGS bed Nuclide removal via TRITON-MSR plate-

• Time-dependent removal of nuclides from one out mixture into another

• User-specified removal constant i,rem as used by core + loop stack ORIGEN to solve ODE:

Production of nuclide i from Loss rate of nuclide i due Source of decay and/or irradiation of to decay, irradiation, or + nuclide i nuclide j other means (flow) 31

Time-dependent inventory - nuclide removal

• Noble gas removal in the off-gas system:

• Main experimental basis is the xenon poison fraction (ratio of absorption by 135Xe to absorption by 235U), reported as 0.3-0.4%

• Noble gas removal fraction was set at 0.03 to match xenon poison fraction

• Noble metal plating-out at the heat exchanger:

• After operation, plated-out noble metals found, with 40% of noble metals plated out in heat exchanger, 50% on all other surfaces in the loop

• Noble metal plate-out removal rate determined from region-wise removal rates, as determined from mass transfer rate, surface area, and fuel salt volume

• Total removal rate calculated as sum of component-wise removal rates 32

Time-dependent inventory

- Depletion at low power level of 8 MWth, with flux level 1.881013 n/cm2-s 10-5 1.17 Xe Nuclide density [atoms/b- cm]

Kr

- No re-fueling in this depletion calculation 10-6 k

1.16

- At 375 days: 10-7 1.15

- 5.627% 235U consumed, 10-8 1.14 k

- 0.455% 238U consumed,

- 13.76 GWd/tHM burnup achieved 10-9 1.13 10-10 1.12 Amount removed w/o Xe and Kr removal after 375 days w/ Xe and Kr removal 10-11 1.11 Noble gas 0 50 100 150 200 250 300 350 400 0.170 kg / 30.6 L (Xe + Kr) Days Insoluble metals 0.611 kg Comparison of Xe and Kr nuclide (Mo + Tc + Ru + Rh + Pd + Ag + Sb) densities with and without Xe/Kr removal Sometimes soluble metals 0.057 kg (Se + Nb + Te) 33

Decay heat after shutdown at 375 days System decay heat [% operating power] Top contributors 34

Decay heat after shutdown OGS at 375 days HX MSRE operating power: 8 MWth 35

Demonstration of continuous feed / refueling 9.6x 10- 5 1.17

• Side calculation with 235U feed through TRITON-MSR Nuclide density [atoms/b- cm]

9.3x 10- 5

• Continuous feed rate of 1.16 1.49x10-3 g/s to yield 9.0x 10- 5 approximately constant eigenvalue 8.7x 10- 5 235 U 1.15 k k

• Increasing 235U fuel 8.4x 10- 5 concentration compensates 1.14 for fission product buildup 8.1x 10- 5 w/o 235 U feed 235 w/ U feed

• Consider low burnup, and 7.8x 10- 5 1.13 hardly any 239Pu buildup 0 50 100 150 200 250 300 350 400 Days Comparison of 235U nuclide densities and eigenvalue with and without 235U feed 36

Core power/flux distribution Sandia National Laboratories is a multimission laboratory managed and operated by National Technology and Engineering Solutions of Sandia LLC, a wholly owned subsidiary of Honeywell International Inc. for the U.S. Department of Energys National Nuclear Security Administration under contract DE-NA0003525. SAND20XX-XXXX P

Core power/flux distribution - model development Used TRITON-KENO 3D full core model based on IRPhEP benchmark specifications as basis Analyzed 3D flux profiles and 3D fission rate via mesh tally capability informed discretization of core region Discretized model uses 34 axial and 8 radial zones Discretized model 38

Core power/flux distribution

- fission rate and flux Upper end of graphite structure Sample basket Lower end of Fission rate distributions graphite structure 39

Core power/flux distribution - power Normalized radial power 40

Core power/flux distribution - power Top of graphite stringers Different fuel-to-moderator ratios in upper and lower region cause small peaks in axial power Bottom of graphite stringers Normalized axial power 41

Core power/flux distribution - power Impact of temperature distribution on the power profile

• Nominal case: 911 K in the fuel salt and graphite structure

• Temperature distribution from MELCOR: 910.5 -937.7 K for the fuel salt, 912.3-937.7 K for the graphite structure Normalized radial power profile Normalized axial power profile 42

Core power/flux distribution - reactivity coefficients Determined reactivity coefficients by temperature/density perturbation:

• Calculated reactivity at multiple temperature/density points

• Fitted reactivity

• Determined reactivity coefficient as derivative of fitted curve Component Fresh core 375 days eff [pcm] 704 +/- 14 697 +/- 22 Graphite temperature reactivity

-5.13 +/- 0.05 -4.83 +/- 0.07 coefficient [pcm/K]

Fuel salt temperature and density

-8.27 +/- 0.12 -8.28 +/- 0.12 reactivity coefficient [pcm/K]

43

Location-dependent inventory in loop Sandia National Laboratories is a multimission laboratory managed and operated by National Technology and Engineering Solutions of Sandia LLC, a wholly owned subsidiary of Honeywell International Inc. for the U.S. Department of Energys National Nuclear Security Administration under contract DE-NA0003525. SAND20XX-XXXX P

Location dependent inventory - model development Developed ORIGEN model to predict nuclide inventory in each region of the loop at ~375 days

• Divided MSRE system into 9 general regions, with the core region subdivided into 30 axial zones

• Used fuel salt composition from 2D TRITON-MSR calculation at 375 days as the start

• Developed chain of ORIGEN inputs that use residence time and flux of the fuel 30 axial zones salt in each region and removes noble gases (Kr, Xe) in off-gas system

• 1 ORIGEN input corresponds to the salt traveling 1 time through the whole loop Regions in MSRE system for ORIGEN model 45

Location dependent inventory - model development As fuel salt travels the loop

• Long-lived* nuclides will slowly accumulate/be removed

• Short-lived* nuclides will oscillate around an equilibrium

• Equilibrium established after a few loops (resulting in inventory at just a few minutes after 375 days) 30 axial zones

• relative to the loop transit time (~25 s for MSRE)

Regions in MSRE system for ORIGEN model 46

Location dependent inventory analysis example

• Observed constant densities of long-lived nuclides for several loops

• Observed convergence of short-lived nuclides after ~6 loops Example: Short-lived I-137 at the bottom of the core

- 13 1.32x 10 nuclide (I-137, t1/2=24.5s)

Nuclide Density [atoms/b- cm]

as a function of time at the bottom of the core 1.28x 10- 13 1.24x 10- 13 1.20x 10- 13 1.16x 10- 13 0 50 100 150 200 250 300 Time [s]

47

Location dependent inventory analysis example

• Compared short-lived nuclide densities between different regions

• Found that inventory/decay heat does not significantly differ between regions when summed up into element classes due to short loop transit time in MSRE 1.6x 10- 13 1D loop Nuclide density [atoms/b- cm]

System average 1.5x 10- 13 Example: Short-lived 2 1. Core nuclide (I-137, t1/2=24.5s) 3 4

2. Upper head as a function of location 1.4x 10- 13 5 3. Piping to Pump

4. Pump/OGS in the loop 6 5. Piping to HX 1.3x 10- 13 6. HX 7

7. Piping to RX 8

8. Inlet 1.2x 10- 13 Core (1) 9. Lower head 9

1.1x 10- 13 0 5 10 15 20 25 Time [s]

48

Delayed neutron precursor drift

• Delayed neutrons are important for 240.0 reactivity control Upper head

• Fission products that emit delayed 200.0 87 Br Core neutrons are called delayed neutron 137 I

precursors (DNP) 160.0 88 Br Height [cm]

93

• In flowing fuel systems, delayed neutrons Rb 120.0 138 may be born outside of the core, I 94 commonly called DNP drift Rb 89 80.0 Br 139

• For example I 90

• MSRE eff ~ 700 pcm without drift 40.0 Br

• eff decreases as flow speed (drift) increases Lower head 0.0

• A DNP drift model has not yet been 10- 15 10- 14 10- 13 incorporated in this work Nuclide density [atoms/b- cm]

• Sensitivity studies show using detailed axial-dependent nuclide density versus Selected delayed neutron precursors system-average has negligible effect on calculated by ORIGEN the core power shape 49

Delayed neutron precursor drift (cont.)

240.0

• DNP drift is most relevant for Upper head transient calculations 200.0 87 Br Core 137 I

• Two approaches will be pursued 160.0 88 Br Height [cm]

• MELCOR DNP drift model based 93 Rb on standard 6-group delayed 120.0 138 I

neutron precursors 94 Rb 89

• New higher-fidelity model in 80.0 Br 139 MELCOR based on explicit delayed I 90 neutron precursor nuclides, as 40.0 Br available through ORIGEN Lower head 0.0 10- 15 10- 14 10- 13 Nuclide density [atoms/b- cm]

Selected delayed neutron precursors calculated by ORIGEN 50

Summary of SCALE methods and results Sandia National Laboratories is a multimission laboratory managed and operated by National Technology and Engineering Solutions of Sandia LLC, a wholly owned subsidiary of Honeywell International Inc. for the U.S. Department of Energys National Nuclear Security Administration under contract DE-NA0003525. SAND20XX-XXXX P

SCALE MSR Summary SCALEs capabilities were demonstrated:

• 3D modeling with TRITON-KENO for time snapshots of power profiles and reactivity coefficients

• TRITON-MSR for time-dependent system-average inventory considering noble gas and noble metal removal through off-gas system and plating out, respectively

• ORIGEN for region-dependent inventory considering noble gas removal Planned enhancements:

• TRITON-MSR with continuous feed

• Tracking of removed nuclides in ORIGEN

• ORIGAMI for MSRs

• Integration of ORIGEN into MELCOR 52

MELCOR Molten Salt Reactor Models

Molten Salt Reactor Modeling Molten Salt Reactor modeling in MELCOR

• Accident progression

• Source term Control Fluid point kinetics, transmutation Cool Generalized EOS, CVH, Contain HS Radionuclide species, transport, and retention 54

Modeling MSR Accidents with MELCOR -

Hydrodynamics and Heat Transport CV WW CV XX MELCOR remains a general purpose, multi-physics code to FL WX model integral plant response under accident conditions H

• Serves as an effective foundation to support NRC readiness to S FL WY FL XZ license advanced nuclear energy technologies X Fluid fuel thermal hydraulics Z

• Leverage existing thermal hydraulics modeling in MELCOR FL YZ

• Utilize fundamental two-phase thermal hydraulic equations CV YY CV ZZ

• Introduce new thermo-physical properties and phase diagram HS ZZ of fluid specific to FLiBe

• Generalized EOS - Equations of state for multiple working fluids are presently available in MELCOR including water, sodium, and FLiBe Thermal hydraulics - CVH/FL Model Packages

• Control Volume Hydrodynamics (CVH) package defines control volumes (CV)

• Flow path modeling package defines flow paths (FL)

Heat Transfer - HS/CVH/COR Packages

• The HS package defines heat structures (HS) that model radiative and conductive heat losses

• CVH package manages convective heat losses 55

Modeling MSR Accidents with MELCOR -

Reactivity Control Fluid fuel point kinetics enables simplified, but appropriate treatment of neutronic transients Fuel point kinetics - derived from standard point kinetic equations and solved similarly Range of feedback models available for flexible modeling of transients

• User-specified external input

• Other implementations in the code (e.g., Doppler, fuel and moderator density) generally not used for MSR applications because they were derived for other types of reactor cores

• Flow reactivity feedback effects integrated into the equation set Control volume fluid core with power distribution

• Neutronics model provides power in core-region, distribution of precursor radionuclides in the core and around the loop

• Radionuclides advected with the flowing salt contribute to decay heat in different regions of the reactor Fission product transmutation enhancement

• Coupling with SCALE/Oak Ridge Isotope GENerator (ORIGEN) ongoing 56

Modeling MSR Accidents with MELCOR - Fission Product Transport and Release Molten salt serves as a potential means of fission product retention Generalized Radionuclide Species

• Users can redefine/add RN classes

• RN classes exhibit similar transport and retention behavior

• Approach taken for molten salt systems - unique fission product chemistry relative to water-moderated systems

• See Slide 75 for example grouping chosen for MSRs Fluid fuel radionuclide transport

• Generalized Radionuclide Transport and Retention (GRTR) modeling framework

• Molten salt chemistry and physics pertaining to radionuclide transport

• GRTR for MSRs but generalized and applicable to other systems (e.g., liquid metal) 57

FLiBe Equation of State Generic working fluid EOS capability facilitates FLiBe as hydrodynamic material

• MELCOR employs fluid property files - INL fusion safety program

• Chens soft sphere model used for FLiBe (INL/EXT-17-44148)

• Property database from ORNL data (ORNL-TM-2316)

• Verified MELCOR EOS library and properties for FLiBe Initial validation activity against ORNL MSRE 58

FLiBe Equation of State - Implications of Salt Freezing Freezing of molten salt an important consideration for a range of accident conditions

• Address fluids that freeze in an accident such as a salt spill

• Freezing in cooling systems (e.g., DRACS)

Adding capabilities to explicitly treat freezing of fluids

• Currently an approximation is used to handle conditions where fluids reach temperatures at or below their freezing point

• Generalized capability for other fluids (e.g., sodium) 59

Fluid Core and Power Distribution Fluid fuel core defined within the graphite stringers

• The fluid volume within the graphite stringers comprise the active Core

• Loop volumes comprise a portion of the primary fuel flow loop OUTSIDE the active core

• Allows specification of the axial and radial power distribution from SCALE Feedbacks and power governed by flowing fluid fuel point reactor kinetics model Fission power generation in core and loop control volumes

• Fission power and feedbacks are calculated for the core volumes

• No fission power energy generation in loop volumes

• Decay heat (due to radionuclide class mass carried in pool) for both volume types

• Graphite heating due to neutron absorption

• Provisions for shutdown in a spill accident 60

Fluid Fuel Neutronic Transients - Modified Point Kinetics Fission inside core

• Neutrons generated and moderated

• DNPs generated DNPs that do not decay in core-region flow into loop

• Decay in loop or advect back into core-region A B C A- In-Vessel DNP gain by fission D E B- In-Vessel DNP loss by decay and flow C- In-Vessel DNP gain by Ex-Vessel DNP flow D- Ex-Vessel DNP gain by In-Vessel DNP flow E- Ex-Vessel DNP loss by decay, flow

• • DNP = Delayed Neutron Precursor 61

Fluid Fuel Point Kinetics - Initial Validation MELCOR non-LWR validation is leveraging available data

• Validation basis will continually expand with evolution of tests and deployments Initial validation has been performed against zero-power MSRE pump flow coast-down test 62

GRTR - Generalized Radionuclide Transport and Retention Track where fission products are and how much is released from liquid to atmosphere Characterizes evolution of fission products between different physico-chemical forms

• Fission product evolution from a liquid pool to an atmosphere

• Influenced by solubility and vapor pressure ,

• Insoluble fission product deposition on structures GRTR mass transport modeling characterizes

• Concentration of radionuclide forms

• Concentration gradients between radionuclide forms

• Resistance to mass transfer between radionuclide forms using standard correlation-based interfacial mass transport theory 63

GRTR and Integral MELCOR Simulations GRTR Physico- Advective and Inputs to GRTR Chemical Transport Fission/Transmutation Model Dynamics Dynamics Radionuclide mass in (or released to) liquid pool Soluble radionuclide form mass Advection of radionuclides in liquid pool or atmosphere Chemical speciation Colloidal radionuclide form mass Pressure in hydrodynamic volume Deposited radionuclide form mass Temperature in regions of hydrodynamic volume (e.g., liquid and Decay of radionuclides in atmosphere) hydrodynamic control volume Coupling with ORIGEN Advective flows of liquid and atmosphere between hydrodynamic Gaseous radionuclide mass volumes For Each Timestep 64

GRTR - Range of Mass Transport Processes Evolution of fission products from molten salts primarily focused on vaporization

• Provides ability to perform best estimate evaluations of release from molten salts

• Demonstration calculations have focused on direct comparison to MSRE for the maximum hypothetical accident

• Exercise of model will be performed next year Mass transfer interfaces

• Liquid-gas atmosphere interfaces

• Liquid-solid structure interfaces

• Gas atmosphere-solid structure interfaces

• Model allows new interfaces to be defined Sparging gas flows (i.e., helium gas injection) will result in fission products entrained in the gas bubble formed by injection Jet breakup when contaminated fluids are released into a gas atmosphere (e.g., due to a pipe break) 65

Illustrating Components of Vaporization Mass Transfer Example CsF vapor pressure - subject to change as thermochemistry evolves

=

Mass transfer coefficient captures effectiveness of species diffusion into atmosphere from liquid-gas interface as well as convective flows carrying vapor away from interface 66

Evolution of GRTR Modeling Focus of modeling efforts evolving based on insights from current demonstration calculations For salt spills, MELCOR GRTR model predicts very small vaporization releases of CsF, Cs and CsI from salt

• Relatively low temperature molten salt temperature leads to a very low vaporization (<<10-6)

• Contribution of the vaporization term in a spill scenario is negligible Ongoing model development utilizes flexibility to explore different ways to characterize other release mechanisms

• Jet breakup and splashing models

• Aerosol release from bubble bursting 67

Molten Salt Reactor Plant Model and Source Term Analysis

MELCOR nodalization - core and reactor vessel Vessel nodalization 130 - Outlet Plenum

• Assumes azimuthal symmetry

• The graphite core structure is subdivided into v 219 v 229 v 239 v 249 v 259 100 10 axial levels and 5 radial rings v

218 v 228 v 238 v 248 v 258 Next slide shows mapping from SCALE 217 227 237 247 257

• Molten fuel salt enters through an annular distributor (cv-100) that directs the flow into the annular 216 226 236 246 256 105 Annulus Downcomer downcomer (cv-105) and the core inlet plenum 215 225 235 245 255 (cv-110) v

• The core is formed by graphite stringers that include v 214 v

224 v

234 v

224 v

254 flow channels v 213 v

223 v

233 v

243 v

253

• The molten fuel salt flows through the stringers 212 222 232 242 252 (CV-210 through CV-259), where the fuel fissions 211 221 231 241 251 Core region 210 220 230 240 250 110 - Inlet Plenum 69

MELCOR core region mapping to SCALE 35.93cm 34 33 600 stringers MELCOR axial mapping is 3 SCALE levels per 159.24cm 3-32 1 MELCOR level MELCOR radial mapping to SCALE ORNL Radial Zone (r) 1&2 3 4 5 6&7 2 MELCOR Radial Zone (r) 1 2 3 4 5 38.55cm 1 2 3 4 5 6 1

Percent 3.5% 19.3% 35.3% 36.7% 5.2%

7 70

MELCOR nodalization - primary recirculation loop Recirculation flows

• Pump bowl spray = 50 gpm

• Pump shaft = 15 gpm 306 402 401 400 320 310 305 403 321 300 398 Helium off-gas flows Total core power

• Pump shaft = 1279 l/d 330

• Fission = 8.8 MW

• Pump bowl = 3456 l/d

• Graphite heating = 0.7 MW

• Overflow tank = 1279 l/d

• Decay heat = 0.4 MW Flows

• Primary loop = 1200 gpm

• Intermediate loop = 850 gpm 306 150 71

MELCOR nodalization - reactor cell, condensing tank, and reactor building Leakages FL-599

• Reactor cell = 0.42 scfh at 12.7 psia FL-520 Bldg leakage

• Reactor bldg = 10% per day at 0.25 psig Reactor Building HVAC supply CV-520 FL-525 FL-560 To filters & stack Reactor cell leakage HVAC exhaust Bldg leakage FL-515 FL-598 Gas retention tank CV-540 FL-525 - Vacuum pump Reactor Cell Rupture disks FL-535 = 15 psig (4 line) FL-555 CV-510 FL-540 = 20 psig (12 line)

CV-530 FL-550 FL-540 FL-535 Vacuum brkr CV-525 To the stack 30 vent line Closed valves Water Condensing tank Drain Tank Room CV-535 CV-515 FL-510 CV-399 FL-545 Pump furnace 150 72

MELCOR nodalization - offgas system Building HVAC Plant stack cv-699 Water-cooled flow Charcoal beds Water-cooled flow cv-601 cv-605 cv-610 From the pump bowl cv-615 cv-620 cv-600 Roughing filter Absolute filter Aux. Charcoal beds cv-625 Filter pit cv-635 73

MELCOR model inputs (1/2)

Equilibrium inventory and decay heat by region from SCALE Radial and axial power profiles from SCALE Reactivity and Xe feedbacks from SCALE Radionuclide distribution from SCALE Collaborative redefinition of radionuclide classes with ORNL

• Re-grouping from LWR definition based on solubility estimations from MSRE and suggestions by Britt (ORNL)

• Insoluble elements isolated into two groups (next slide)

Antoine coefficient estimates for few species (Cs, CsF)

• Cesium and cesium fluoride estimated using MSTDB-TC

[Phillip Britt, Future Research Directions, DOE-NE Molten Salt Chemistry Workshop, April 10-12, 2017, ORNL] 74

MELCOR model inputs (2/2)

MELCOR Elemental Grouping Xe : He, Ne, Ar, Kr, Xe, Rn, H, N Cs : Li, Na, K, Rb, Cs, Fr, Cu Ba : Be, Mg, Ca, Sr, Ba, Ra, Es I : F, Cl, Br, I, At S : S, Po Re : Re, Os, Ir, Pt, Au, Ni V : V, Cr, Fe, Co, M, Ta, W Mo : Mo, Tc, Ru, Rh, Pd, Ag, Ge, As, Sn, Sb Nb : Nb, Zn, Cd, Se, Te Ce : Ti, Zr, Hf, Ce, Th, Pa, Np, Pu, C La : Al, Sc, Y, La, Ac, Pr, Nd, Pm, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu, Am, Cm, Bk, Cf U:U Cd : Hg, Ga, In Ag : Pb, Tl, Bi B : B, Si, P Mo class assumed to be insoluble

[Phillip Britt, Future Research Directions, DOE-NE Molten Salt Chemistry Workshop, April 10-12, 2017, ORNL] 75

Scenario Spill of molten salt into the reactor cell (containment)

• Full reactor spill - maximum credible accident in the MSRE safety analysis Spill onto the floor without coincident water leak (MCA1-MCA5)

Spill with coincident water leak (MCA6-MCA9)

Exploratory radionuclide source term due to limited information from the molten salt thermophysical databases

• ORNL-TM-0732 MSRE safety analysis source term Integral calculation with aerosol physics

• GRTR vaporization model without splashing Cs, CsI, and Xe releases Sensitivities

• HVAC operating or off

• Auxiliary filter operation

• Aerosol size 76

Salt spill cases Walk-through MCA1 (base case)

Spill creates aerosols with 1 µm mass median diameter (MMD) Case Aerosol size Stack Fans Aux. Filters Water Spill with a 1.5 geometric standard MCA1 1 µm Yes No No deviation (GSD) MCA2 10 µm Yes No No The HVAC remains running and MCA3 1 µm No No No ventilating the reactor building MCA4 1 µm Yes Yes No The auxiliary filters are not used MCA5 1 µm No Yes No to filter the reactor cell MCA6 1 µm Yes No Yes There is no coincident water spill MCA7 1 µm Yes Yes Yes onto the molten salt MCA8 1 µm No Yes Yes MCA9 1 µm No No Yes 77

MCA1 salt spill base case - Primary System Response 78

MCA1 salt spill base case - Reactor Cell Response 79

MCA1 reactor cell thermal-hydraulic response Mass of molten salt spilled and its temperature 9000 1400

• The primary loop salt inventory spills to the 8000 Mass 1300 reactor cell in 10 minutes 7000 Temperature 1200 Temperature of the molten salt is relatively constant 6000 1100 Mass (lbm) Temperature (F) with a slow cooling trend 5000 1000 4000 900

• There is an immediate pressurization of the 3000 800 gas space from subatmospheric to ~20 psia 2000 700 Heating due to hot molten salt (~1100F) 1000 600 Heating due to the released radionuclides 0 500 0 10 20 30 40 50 60 Time (min)

Reactor cell pressure and temperature

• Reactor cell gas temperature initially rises to 25 1000 over 900F and then slowly cools 900 20 800 Pressure 700 Temperature (F)

Temperature Pressure (psia) 15 600 Normal pressure (12.7 psia) 500 10 400 300 5 200 100 0 0 0 10 20 30 40 50 60 Time (min) 80

MCA1 reactor cell radionuclide releases Radionuclide airborne release 1.2

• Airborne release assumptions 1.0

[ORNL-TM-0732] Xe Fraction of the initial inventory (-)

Cs Xe (noble gases)

Iodine (gas) 0.8 100% of the noble gases Iodine (aerosol)

Ce (Pu) 10% of the iodine 0.6 10% of all other volatile and non-volatile radionuclides 0.4 MCA from MSRE safety analysis 0.2 All others

• Radionuclides phase (aerosol or gas) depends 0.0 1 10 100 1000 on temperature and chemical form Time (sec)

Vapor pressure in the atmosphere Analysis used some modification for likely MSR 1.E+03 chemical forms 1.E+02 Some gaseous iodine (5%)

Cesium and iodine combine (CsI) 1.E+01 Pressure (psi)

I2 Cs metal Remaining cesium as CsF 1.E+00 CsOH CsI Ce MELCOR allows exploration of alternate chemical CsF forms for the MSR 1.E-01 1.E-02 0 400 800 1200 1600 2000 2400 2800 3200 3600 4000 Temperature (°F) 81

MCA1 gaseous radionuclide distributions Xe distribution 1.E+00 Gaseous releases (xenon and iodine gas) respond 1.E-01 Reactor cell Reactor bldg similarly Reactor cell Environment Fraction of the initial inventory (-)

1.E-02

• Most of the gases retained in the reactor cell 1.E-03 Environment

• Reactor cell slowly leaks to the reactor building 1.E-04 Reactor bldg 1.E-05 The reactor building HVAC is operating in MCA1, 1.E-06 which exhausts gases from the reactor building 1.E-07 through the absolute filters to the plant stack 1.E-08 1 10 100 1000 10000 100000

• 0.2% of the xenon reaches the environment Time (sec)

Iodine gas distribution

• 0.02% of the gaseous iodine reaches the environment 1.E+00 Reactor cell 1.E-01 Reactor bldg Environment Reactor cell Fraction of the initial inventory (-)

1.E-02 Building HVAC Plant stack cv-699 Water-cooled flow 1.E-03 Charcoal beds Water-cooled flow cv-601 cv-605 cv-610 From the pump bowl Environment cv-620 cv-615 1.E-04 cv-600 Roughing filter Absolute filter Aux. Charcoal beds cv-625 Filter pit cv-635 1.E-05 Reactor bldg 1.E-06 1.E-07 1.E-08 1 10 100 1000 10000 100000 Time (sec) 82

MCA1 aerosol radionuclide distributions CsF distribution Release fractions of radionuclides that form 1.E+00 aerosols in the reactor building 1.E-01 Reactor cell Fraction of the initial inventory (-)

• CsF and CsI illustrate radionuclide chemical forms that 1.E-02 include both vapor and aerosol forms in the reactor cell 1.E-03 Reactor cell Reactor cell airborne Higher leakage of vapors formation of small aerosols in 1.E-04 Reactor bldg Filters Environment the reactor building more to environment & the filters 1.E-05 Environment Reactor bldg

• Ce is an aerosol that primarily settles in the reactor cell 1.E-06 Filters and captured by the absolute filters 1.E-07 1.E-08 1 10 100 1000 10000 100000 Time (sec)

Ce distribution CsI distribution 1.E+00 1.E+00 1.E-01 1.E-01 Reactor cell Fraction of the initial inventory (-) Fraction of the initial inventory (-)

1.E-02 Reactor cell 1.E-02 Reactor cell 1.E-03 1.E-03 Reactor cell airborne Reactor cell Reactor bldg Reactor cell airborne Environment Filters 1.E-04 Reactor bldg 1.E-04 Environment Filters 1.E-05 Environment 1.E-05 Reactor bldg Reactor bldg Filters 1.E-06 1.E-06 Filters 1.E-07 Environment 1.E-07

(<10-8) 1.E-08 1.E-08 1 10 100 1000 10000 100000 1 10 100 1000 10000 100000 Time (sec) Time (sec) 83

Salt spill with water base case Case Aerosol size Stack Fans Aux. Filters Water Spill Walk-through MCA6 MCA1 1 µm Yes No No Spill creates aerosols with 1 µm mass median MCA2 MCA3 10 µm 1 µm Yes No No No No No diameter (MMD) with a 1.5 geometric standard MCA4 1 µm Yes Yes No deviation (GSD) MCA5 1 µm No Yes No MCA6 1 µm Yes No Yes The HVAC remains running and ventilating the MCA7 1 µm Yes Yes Yes MCA8 1 µm No Yes Yes reactor building MCA9 1 µm No No Yes The auxiliary filters are not used to filter the reactor cell Water spill onto the molten salt Equilibration of all the fuel salt with the cell atmosphere and just enough water to form the maximum amount of saturated steam would result in the maximum pressure in the secondary container.

With no relief device, pressures as high as 110 psig could result.

[ORNL-TM-0732]

[ORNL-TM-0732] 84

MCA6 reactor cell thermal-hydraulic response

• Molten salt is assumed to mix with coincidentally spilled water Rapid pressurization of the reactor cell as it fills with steam Reactor cell and gas retention tank pressure and temperature

• Reactor cell pressure rises to 46 psia 50 500 15 psi rupture disk opens at 41 sec 45 20 psi rupture disk opens at 115 sec 40 400

• Reactor cell temperature initially rises to 35 330F but falls after the 20 psi rupture disk Pressure (psia) Temperature (F) 30 300 opens 25 20 200 15 Reactor cell pressure Gas retention tank pressure 10 Reactor cell temperature 100 Gas retention tank temperature 5 15 psi rupture disk 20 psi rupture disk 0 0 0 100 200 300 400 500 600 Time (sec)

MSRE vapor-condensing system

[ORNL-TM-0728] 85

MCA6 gaseous radionuclide distributions Xe distribution 1.E+00 Same MCA airborne releases into reactor 1.E-01 cell Fraction of the initial inventory (-)

1.E-02

• Release assumptions 1.E-03 100% of the noble gases Reactor cell 1.E-04 Cond and gas retention tanks Reactor bldg 10% of the iodine 1.E-05 Environment 10% of all other volatile and non-volatile radionuclides 1.E-06

• Strong flows to the condensing and gas 1.E-07 retention tanks capture most of the 1.E-08 1 10 100 Time (sec) 1000 10000 100000 radionuclides released from the spilled salt 1.E+00 Iodine gas distribution Condensing tank retains most of the aerosols and the 1.E-01 gas retention tank captures any radionuclides that Fraction of the initial inventory (-)

pass through the pool 1.E-02 All noble gases and most of the gaseous iodine 1.E-03 passes through the condensing pool 1.E-04 1.E-05 Reactor cell Cond & gas retention tanks Reactor bldg 1.E-06 Environment 1.E-07 1.E-08 1 10 100 1000 10000 100000 Time (sec) 86

MCA6 aerosol radionuclide distributions CsF distribution 1.E+00 Most of the aerosol releases are retained in the 1.E-01 condensing tank Fraction of the initial inventory (-)

1.E-02

• CsF and CsI form aerosols in a water spill accident and 1.E-03 behave similarly to the cerium aerosols 1.E-04 The large steam source contributes to aerosol 1.E-05 Reactor cell Reactor cell airborne Condensing tank agglomeration and more rapid settling in the 1.E-06 Reactor bldg Filters 1.E-07 reactor cell than the dry case 1.E-08 Environment 1 10 100 1000 10000 100000 Time (sec)

Ce distribution CsI distribution 1.E+00 1.E+00 1.E-01 1.E-01 Fraction of the initial inventory (-)

1.E-02 Reactor cell Fraction of the initial inventory (-)

Reactor cell airborne 1.E-02 Reactor cell Condensing tank Reactor cell airborne 1.E-03 Reactor bldg 1.E-03 Condensing tank Filters Reactor bldg 1.E-04 Environment Filters 1.E-04 Environment 1.E-05 1.E-05 1.E-06 1.E-06 1.E-07 1.E-07 1.E-08 1 10 100 1000 10000 100000 1.E-08 1 10 100 1000 10000 100000 Time (sec)

Time (sec) 87

Overall insights (1/4)

The xenon release to the environment spanned many All results of xenon release to the environment orders of magnitude depending on scenario Xe release to the environment assumptions 1.E+00 HVAC + Aux filter

• Lowest releases with no HVAC and no Aux filter flow 1.E-01

• Auxiliary filter operation increases the release of xenon 1.E-02 to the environment while it provides filtering of airborne 1.E-03 aerosols Release Fraction (-)

1.E-04 1.E-05 MCA1 Case Aerosol size Stack Fans Aux. Filters Water Spill 1.E-06 MCA2 MCA1 1 µm Yes No No 1.E-07 MCA3 MCA4 MCA2 10 µm Yes No No No HVAC + no Aux filter cases MCA5 MCA3 1 µm No No No 1.E-08 MCA6 MCA4 1 µm Yes Yes No 1.E-09 MCA7 MCA5 1 µm No Yes No Aux filter cases MCA8 MCA6 1 µm Yes No Yes 1.E-10 MCA9 MCA7 1 µm Yes Yes Yes 1.E-11 MCA8 1 µm No Yes Yes 0 6 12 18 24 MCA9 1 µm No No Yes Time (hr)

Note: Results assume no xenon retention in the charcoal filters.

88

Overall insights (2/4)

Ce release to the environment 1.E-03 The aerosol releases to the environment were small 1.E-04 All results of cerium release MCA1 MCA2 MCA3 to the environment due to: 1.E-05 MCA4 MCA5 MCA6 MCA7

• Gravitational settling Release Fraction (-)

1.E-06 MCA8 MCA9 in the reactor cell (all cases), 1.E-07 the reactor building, filter pit, and stack (without HVAC flow) 1.E-08

• Capture in the filter 1.E-09

• Capture in the condensing tank in the water spill cases 1.E-10 1.E-11 0 6 12 18 24 Time (hr)

Ce in the Condensing Tank Ce airborne in the Reactor Cell 1.E+00 MCA1 1.E+00 MCA2 All results of cerium capture in 1.E-01 All results of cerium airborne MCA3 the condensing tank fraction in the reactor cell MCA4 1.E-02 MCA5 MCA6 1.E-03 MCA7 1.E-01 Water spill cases = ~65% No water spill cases MCA8 Release Fraction (-)

Release Fraction (-)

1.E-04 MCA9 MCA1 1.E-05 MCA2 MCA3 1.E-06 MCA4 1.E-02 1.E-07 MCA5 MCA6 1.E-08 MCA7 Water spill cases MCA8 1.E-09 No spill cases = 0% MCA9 1.E-10 1.E-03 1.E-11 0 6 12 18 24 0 6 12 18 24 Time (hr) 89 Time (hr)

Overall insights (3/4)

Cs release to the environment Due to the high temperatures in the reactor cell in the cases 1.E-01 without a water spill (~900F), the two chemical compounds of 1.E-02 No water spill results of CsF release to the environment cesium were in aerosol & vapor form 1.E-03 1.E-04

• Released CsI and CsF vapors subsequently condensed in the reactor Release Fraction (-)

1.E-05 building and the offgas system to form very small aerosols 1.E-06

• CsF and CsI vapors remained airborne in the reactor cell 1.E-07 This contributed to higher cesium environmental releases than 1.E-08 MCA1 the lower temperature cases with a water spill 1.E-09 MCA2 MCA4 1.E-10 MCA5 1.E-11 0 6 12 18 24 Time (hr)

Cs release to the environment 1.E+03 1.E+00 MCA6 1.E-01 MCA7 1.E-02 Spill results of CsF release to MCA8 1.E+02 the environment MCA9 1.E-03 Release Fraction (-)

1.E-04 1.E+01 Pressure (psi)

I2 1.E-05 Cs metal CsOH 1.E-06 CsI 1.E+00 Ce 1.E-07 CsF 1.E-08 1.E-01 1.E-09 1.E-10 1.E-02 1.E-11 0 400 800 1200 1600 2000 2400 2800 3200 3600 4000 0 6 12 18 24 Time (hr)

Temperature (°F) 90

Overall insights (4/4)

The aerosol mass in the reactor building also spanned many orders of magnitude depending All results of cerium in the reactor building on scenario assumptions Ce in the Reactor Building 1.E-01

• Lowest amounts occurred when aerosols were MCA1 captured in the condensing tank and any leaked 1.E-02 MCA2 MCA3 aerosols were filtered via the reactor building HVAC MCA4 MCA5 flow 1.E-03 MCA6 MCA7

• The highest amounts occurred when there was no Release Fraction (-)

MCA8 No water spill & no HVAC MCA9 water spill and no HVAC 1.E-04

• Aerosols leaked into the reactor building without 1.E-05 No water spill & HVAC HVAC and a water spill primarily settled (flat line)

Led to a small amount of leakage to the environment 1.E-06 Water spill & no HVAC

• Finally, the HVAC swept long-term releases into the 1.E-07 reactor building in the no spill cases Water spill cases & HVAC 1.E-08 0 6 12 18 24 Time (hr) 91

Sensitivity to increased reactor cell leakage Xe in the Reactor Building and Environment 1.E+00

• Green line shows impact of reactor cell leakage to the reactor 1.E-01 Green - Xe in the reactor building versus reactor cell leak rate building for the MCA3 scenario 1.E-02 Gaseous Xe leak to the reactor building continues while the Ce Fraction of initial inventory (-)

1.E-03 aerosol leakage stops early due to aerosol settling 1.E-04 Blue - Xe in the environment 1.E-05 for 1X reactor cell leak rate

• Blue line shows impact of reactor cell leakage on 1.E-06 environmental releases for the MCA3 scenario 1.E-07 The leakage to the reactor cell has an approximately linear impact 1.E-08 100X reactor cell leak 10X reactor cell leak 1X reactor cell leak on the reactor cell leak rate versus a slightly larger than linear 1.E-09 100X reactor cell leak 10X reactor cell leak effect on the environment leakage 1.E-10 1X reactor cell leak

• The impacts are expected to be smaller with HVAC operation 1.E-11 0 6 12 18 24 Time (hr)

Ce in the Reactor Building and Environment 1.E+00 1.E-01 100X reactor cell leak 10X reactor cell leak 1.E-02 1X reactor cell leak 100X reactor cell leak 10X reactor cell leak Fraction of initial inventory (-)

1.E-03 Green - Ce in the reactor building versus reactor cell leak rate 1X reactor cell leak 1.E-04 Cases without a water spill 1.E-05 Case Aerosol size Stack Fans Aux. Filters Water Spill MCA1 1 µm Yes No No 1.E-06 MCA2 10 µm Yes No No 1.E-07 Blue - Ce in the environment for MCA3 1 µm No No No 1.E-08 1X reactor cell leak rate MCA4 1 µm Yes Yes No 1.E-09 MCA5 1 µm No Yes No 1.E-10 1.E-11 0 6 12 18 24 Time (hr) 92

Sensitivity to increased reactor building leakage Xe in the Environment 1.E-06

• The reactor building surrounds the reactor cell and provides 0 mph 100X RB leak Xe in the environment versus the final barrier for leakage when the filters are not operating 0 mph 10X RB leak 0 mph 1X RB leak RB leakage and wind speed 5 mph 100X RB leak

• Impact of reactor building leakage as a function of wind speed 5 mph 10X RB leak Fraction of initial inventory (-)

5 mph 1X RB leak 10 mph 100X RB leak shows a very small impact on the release to the environment 10 mph 10X RB leak 10 mph 1X RB leak for the MCA3 scenario 1.E-07 Similar to increased building leakage, a higher wind speed increases the building infiltration and exfiltration rate The impact is slightly larger for gas leakage (i.e., aerosols also settle) 1.E-08

• The nominal (1X) reactor building leakage is very low 0 6 12 Time (hr) 18 24 Only 10% per day at 0.25 psig, 0.002 in2 1.E-09 Ce in the Environment The very large building (480,000 ft3) has no appreciable Ce in the environment versus pressurization (i.e., <<0.25 psig) RB leakage and wind speed Fraction of initial inventory (-)

1.E-10 Cases without a water spill 0 mph 100X RB leak 0 mph 10X RB leak Case Aerosol size Stack Fans Aux. Filters Water Spill 0 mph 1X RB leak 5 mph 100X RB leak MCA1 1 µm Yes No No 5 mph 10X RB leak 5 mph 1X RB leak MCA2 10 µm Yes No No 1.E-11 10 mph 100X RB leak 10 mph 10X RB leak MCA3 1 µm No No No 10 mph 1X RB leak MCA4 1 µm Yes Yes No MCA5 1 µm No Yes No 1.E-12 0 6 12 18 24 Time (hr) 93

Summary Conclusions

• Demonstrated use of SCALE and MELCOR for MSRE safety analysis

• Simulated the entire accident starting with the initiating event

• system thermal hydraulic response

• fuel heat-up

• heat transfer through the reactor to the surroundings

• radiological release

• Evaluated effectiveness of passive mitigation features 95

=

Background===

Slides

Further SCALE analysis details

Time-dependent inventory - nuclide removal Nuclide removal from tank fuel salt in core+loop: bed

• Plating-out of noble metals (Se, Nb, Mo, Tc, Ru, etc.) plate-out at at heat exchanger heat exchanger

• Removal of halogens (I, Br) from plated-out material

• Removal of noble gases from plated-out materials core + loop stack

• Removal of noble gases (Xe, Kr) from fuel into off-gas system

• Removal of gas into charcoal bed

• Removal of gas into stack Figure modified from: R. C. Robertson (1965), MSRE Design and Operations Report Part I: Description of Reactor Design, ORNL-TM-0728, ORNL. 98

Core power/flux distribution - flux Thermal flux Fast flux

(< 0.625 eV) (> 0.625 eV) 99

Region-wise data 100

MELCOR for Accident Progression and Source Term Analysis

MELCOR Development for Regulatory Applications What Is It?

MELCOR is an engineering-level code that simulates the response of the reactor core, primary coolant system, containment, and surrounding buildings to a severe accident.

Who Uses It?

MELCOR is used by domestic universities and national laboratories, and international organizations in around 30 countries. It is distributed as part of NRCs Cooperative Severe Accident Research Program (CSARP).

How Is It Used?

MELCOR is used to support severe accident and source term activities at NRC, including the development of regulatory source terms for LWRs, analysis of success criteria for probabilistic risk assessment models, site risk studies, and forensic analysis of the Fukushima accident.

How Has It Been Assessed?

MELCOR has been validated against numerous international standard problems, benchmarks, separate effects (e.g., VERCORS) and integral experiments (e.g., Phebus FPT), and reactor accidents (e.g., TMI-2, Fukushima).

102

Source Term Development Process Experimental Basis PIRT process Oxidation/Gas Generation Melt Progression Fission Product Release Fission Product Transport Accident Analysis Design Synthesize MELCOR Scenario # 1 Scenario # 2 timings and Basis

. . release Source fractions Term Scenario # n-1 Scenario # n Cs Diffusivity 103

SCALE/MELCOR/MACCS Neutronics Integrated Severe Radiological MELCOR SCALE MACCS

• Criticality Accident Progression Consequences

• Shielding

• Hydrodynamics for range

• Near- and far-field

• Radionuclide inventory of working fluids atmospheric transport

• Burnup credit

• Accident response of and deposition

• Decay heat plant structures, systems

• Assessment of health and components and economic impacts

• Fission product transport Nuclear Reactor System Applications Non-Reactor Applications Design/Operational Safety/Risk Assessment Regulatory Fusion Spent Fuel Facility Safety Support

• Technology-neutral

• License amendments

• Design analysis scoping

• Neutron beam injectors

• Risk studies

• Leak path factor o Experimental

• Risk-informed regulation calculations

• Li loop LOFA transient

• Multi-unit accidents calculations o Naval

• Design certification (e.g.,

• Training simulators analysis

• Dry storage

• DOE safety toolbox codes o Advanced LWRs NuScale)

• ITER cryostat modeling

• Spent fuel

• DOE nuclear facilities o Advanced Non-LWRs

• Vulnerability studies

• He-cooled pebble test transport/package (Pantex, Hanford, Los

• Emergency preparedness blanket (H3) applications Alamos, Savannah River

• Accident forensics Site)

(Fukushima, TMI)

• Emergency Planning Zone

• Probabilistic risk Analysis assessment 104

MELCOR Attributes Foundations of MELCOR Development Fully integrated, engineering-level code Phenomena modeled

• Thermal-hydraulic response of reactor coolant system, reactor cavity, rector enclosures, and auxiliary buildings

• Core heat-up, degradation and relocation

• Core-concrete interaction

• Flammable gas production, transport and combustion

• Fission product release and transport behavior Level of physics modeling consistent with

• State-of-knowledge

• Necessity to capture global plant response

• Reduced-order and correlation-based modeling often most valuable to link plant physical conditions to evolution of severe accident and fission product release/transport Traditional application

• Models constructed by user from basic components (control volumes, flow paths and heat structures)

• Demonstrated adaptability to new reactor designs - HPR, HTGR, SMR, MSR, ATR, Naval Reactors, VVER, SFP, 105

MELCOR Attributes MELCOR Pedigree International Collaboration Cooperative Severe Accident Research Program (CSARP) - June/U.S.A Validated physical models MELCOR Code Assessment Program (MCAP) - June/U.S.A European MELCOR User Group (EMUG) Meeting - Spring/Europe

• International Standard Problems, benchmarks, experiments, and reactor Asian MELCOR User Group (AMUG) Meeting - Fall/Asia accidents

• Beyond design basis validation will always be limited by model uncertainty that arises when extrapolated to reactor-scale Cooperative Severe Accident Research Program (CSARP) is an NRC-sponsored international, collaborative community supporting the validation of MELCOR International LWR fleet relies on safety assessments performed with the MELCOR code 106

Common Phenomenology 107

MELCOR Modeling Approach Modeling is mechanistic consistent with level of knowledge of phenomena supported by experiments Parametric models enable uncertainties to be characterized

• Majority of modeling parameters can be varied

• Properties of materials, correlation coefficients, numerical controls/tolerances, etc.

Code models are general and flexible

• Relatively easy to model novel designs

• All-purpose thermal hydraulic and aerosol transport code 108

MELCOR State-of-the-Art MELCOR Code Development Version Date 2.2.18180 M2x Official Code Releases December 2020 2.2.14959 October 2019 2.2.11932 November 2018 2.2.9541 February 2017 2.1.6342 October 2014 2.1.4803 September 2012 2.1.3649 November 2011 2.1.3096 August 2011 2.1.YT August 2008 2.0 (beta) Sept 2006

MELCOR Software Quality Assurance - Best Practices MELCOR SQA Standards Emphasis is on Automation SNL Corporate procedure IM100.3.5 Affordable solutions CMMI-4+

NRC NUREG/BR-0167 Consistent solutions MELCOR Wiki Bug tracking and reporting

• Archiving information

• Bugzilla online

• Sharing resources (policies, conventions, information, progress) Code Validation among the development team.

• Assessment calculations

• Code cross walks for complex phenomena where Code Configuration Management (CM) data does not exist.

• Subversion

• TortoiseSVN Documentation

• Available on Subversion repository with links from

• VisualSVN integrates with Visual Studio wiki (IDE)

• Latest PDF with bookmarks automatically generated from word documents under Subversion Reviews control

• Code Reviews: Code Collaborator

• Links on MELCOR wiki

• Internal SQA reviews Project Management Continuous builds & testing

• Jira for tracking progress/issues

• DEF application used to launch multiple

• Can be viewable externally by stakeholders jobs and collect results

• Regression test report Sharing of information with users

• External web page

• More thorough testing for code release

• MELCOR workshops

• Target bug fixes and new models for

• MELCOR User Groups (EMUG & AMUG) testing 110

MELCOR Verification & Validation Basis LWR & non-LWR applications Volume 1: Primer & User Guide Volume 2: Reference Manual Volume 3: MELCOR Assessment Problems

[SAND2015-6693 R]

Specific to non-Analytical Problems Saturated Liquid Depressurization Adiabatic Expansion of Hydrogen AB-1 LOF,LOHS,TOP MSRE Air-Ingress Transient Heat Flow in a Semi-Infinite Heat Slab AB-5 TREAT M-Series LWR application experiments Helical SG HT T-3 ANL-ART-38 Cooling of Heat Structures in a Fluid Radial Heat Conduction in Annular Structures Establishment of Flow Sodium Fires Molten Salt Sodium Reactors HTGR (Completed) (planned) (planned) (planned) 111

Sample Validation Cases TRISO Diffusion Release Turbulent LACE LA1 and LA3 IAEA CRP-6 Benchmark tests experimentally Deposition Fractional Release examined the Case 1a 1b 2a 2b 3a 3b transport and US/INL 0.467 1.0 0.026 0.996 1.32E-4 0.208 retention of US/GA 0.453 0.97 0.006 0.968 7.33E-3 1.00 aerosols through US/SNL 0.465 1.0 0.026 0.995 1.00E-4 0.208 pipes with high US/NRC 0.463 1.0 0.026 0.989 1.25E-4 0.207 speed flow France 0.472 1.0 0.028 0.995 6.59E-5 0.207 Korea 0.473 1.0 0.029 0.995 4.72E-4 0.210 Germany 0.456 1.0 0.026 0.991 1.15E-3 0.218 Resuspension (1a): Bare kernel (1200 oC for 200 hours0.00231 days <br />0.0556 hours <br />3.306878e-4 weeks <br />7.61e-5 months <br />) A sensitivity study to examine STORM (Simplified Test of Resuspension (1b): Bare kernel (1600 oC for 200 hours0.00231 days <br />0.0556 hours <br />3.306878e-4 weeks <br />7.61e-5 months <br />) fission product release from Mechanism) test facility (2a): kernel+buffer+iPyC (1200 oC for 200 hours0.00231 days <br />0.0556 hours <br />3.306878e-4 weeks <br />7.61e-5 months <br />) a fuel particle starting with a (2b): kernel+buffer+iPyC (1600 oC for 200 hours0.00231 days <br />0.0556 hours <br />3.306878e-4 weeks <br />7.61e-5 months <br />)

(3a): Intact (1600 oC for 200 hours0.00231 days <br />0.0556 hours <br />3.306878e-4 weeks <br />7.61e-5 months <br />) bare kernel and ending with (3b): Intact (1800 oC for 200 hours0.00231 days <br />0.0556 hours <br />3.306878e-4 weeks <br />7.61e-5 months <br />) an irradiated TRISO particle; Aerosol Physics

• Agglomeration

• Deposition

• Condensation and Evaporation at surfaces Validation Cases

• Simple geometry: AHMED, ABCOVE (AB5 & AB6), LACE(LA4),

• Multi-compartment geometry: VANAM (M3), DEMONA(B3)

• Deposition: STORM, LACE(LA1, LA3) 112

MELCOR Modernization Generalized numerical solution engine Hydrodynamics In-vessel damage progression Ex-vessel damage progression Fission product release and transport 113

Cs vapor pressures in GRTR calculations GRTR - Generalized Radionuclide Transport and Retention Enhancement to MELCOR radionuclide transport modeling

• Incorporate unique chemistry of fission products in new fluids potentially mitigating release to atmospheres of reactor vessel, off-gas systems, and confinement/containment Retention in fluids influenced by physico-chemical form of fission products - strong influence of thermochemistry

• Is the fission product compound soluble?

• Is the fission product compound insoluble (i.e., colloidal)?

• Is the fission product compound a gaseous vapor?

• Has the fission product compound deposited on a structural surface?

• Is the fission product located at a liquid-atmosphere interface?

• Interface between liquid pool and overlying gas atmosphere

• Interface between liquid and gas bubbles (e.g., generated by sparging helium gas)

Introduce new physico-chemical forms that supplement existing MELCOR representation of distinct radionuclide classes

• Soluble fission products

• Insoluble/colloidal fission products

• Deposited fission products

• Gaseous fission products Each tracked form is identified with either a liquid pool, an atmosphere, or deposited on a structure 115

Results of the sensitivity studies Cases without a water Spill Xe release to the environment Gaseous xenon release to the environment 1.E+00 1.E-01 without a water spill 1.E-02

• MCA4 had the highest release due to auxiliary filter 1.E-03 Release Fraction (-)

1.E-04 venting of the reactor cell after 1-hr and enhanced 1.E-05 leakage due the HVAC flow through the reactor building 1.E-06 MCA1 MCA2

• MCA3 had the lowest release with no HVAC flow in the 1.E-07 MCA3 MCA4 1.E-08 reactor building (stack fans) and no auxiliary flow 1.E-09 MCA5

• MCA1 and MCA2 were identical because xenon is not 1.E-10 an aerosol 1.E-11 0 6 12 Time (hr) 18 24

• MCA5 did not have enhanced releases due to HVAC Cases without a water spill venting the reactor building leaks but did include the Case Aerosol size Stack Fans Aux. Filters Water Spill auxiliary filter flow after 1-hr MCA1 1 µm Yes No No MCA2 10 µm Yes No No MCA3 1 µm No No No MCA4 1 µm Yes Yes No MCA5 1 µm No Yes No Note: Results assume no xenon retention in the charcoal filters.

116

Results of the sensitivity studies Cases with a water Spill Xe release to the environment Gaseous xenon release to the environment with a 1.E+00 1.E-01 water spill 1.E-02

• MCA7 had the highest release due to auxiliary filter 1.E-03 Release Fraction (-)

1.E-04 venting of the reactor cell after 1-hr and enhanced 1.E-05 leakage due the HVAC flow through the reactor building 1.E-06

• MCA7 had a lower release than the corresponding dry 1.E-07 1.E-08 MCA6 MCA7 case due to xenon capture in the gas retention tank 1.E-09 MCA8 MCA9

• MCA9 had the lowest release due to no HVAC flow in 1.E-10 the reactor building (stack fans) and no auxiliary filter 1.E-11 0 6 12 Time (hr) 18 24 flow Cases with a water spill

• Leaks into the reactor building from MCA6 were vented Case Aerosol size Stack Fans Aux. Filters Water Spill to the environment due to the HVAC operation MCA6 1 µm Yes No Yes MCA7 1 µm Yes Yes Yes

• MCA8 did not have enhanced releases due to HVAC MCA8 1 µm No Yes Yes venting any reactor building leaks but did include MCA9 1 µm No No Yes venting to the environment from the auxiliary filter flow after 1-hr Note: Results assume no xenon retention in the charcoal filters.

117

Results of the sensitivity studies Cases without a water Spill Ce release to the environment 1.E+00 The cerium aerosol release to the environment 1.E-01 MCA1 MCA2 without a water spill were very low 1.E-02 MCA3 MCA4 1.E-03 MCA5

• MCA3, MCA4, and MCA5 releases to the environment Release Fraction (-)

1.E-04 were approximately the same and larger than MCA1 1.E-05 MCA4 and MCA5 included continuous venting of very small 1.E-06 1.E-07 aerosols from the reactor cell through the auxiliary filter 1.E-08 MCA3 results show impact of nominal leakage from the 1.E-09 reactor building (i.e., no filtration) 1.E-10 1.E-11 MCA1 included filtration of the reactor building but no 0 6 12 Time (hr) 18 24 auxiliary filter flow 1 versus 10 µm aerosol behavior Ce settling and filter behavior

• MCA2 had larger aerosols, which settled faster and the 1.E-01 MCA1 airborne in the reactor cell 1.E-02 smallest amount released to the environment 1.E-03 MCA1 captured by filters MCA1 environment MCA2 airborne in the reactor cell MCA2 captured by filters Cases without a water spill 1.E-04 MCA2 environment Release Fraction (-)

1.E-05 Case Aerosol size Stack Fans Aux. Filters Water Spill Airborne in the reactor cell MCA1 1 µm Yes No No 1.E-06 MCA2 10 µm Yes No No 1.E-07 Filters MCA3 1 µm No No No 1.E-08 MCA4 1 µm Yes Yes No Environment 1.E-09 MCA5 1 µm No Yes No 1.E-10 1.E-11 Note: The capture efficiency of the absolute filters for aerosols below <0.3 µm was assumed to be zero. 0 6 12 18 24 Time (hr) 118

Results of the sensitivity studies Cases with a water Spill Cerium aerosol release to the environment 1.E+00 Ce release to the environment with a water spill 1.E-01 MCA6 MCA7

• MCA6 and MCA7 had the higher releases release 1.E-02 MCA8 MCA9 1.E-03 due the HVAC flow through the reactor building Release Fraction (-)

1.E-04 The auxiliary filter flow increased releases to the 1.E-05 environment due to non-perfect capture by the absolute 1.E-06 filters 1.E-07 1.E-08 MCA6 and MCA7 were higher than the corresponding 1.E-09 dry cases (MCA1 and MCA4) due to higher leakage 1.E-10 from the reactor cell 1.E-11 0 6 12 18 24 Time (hr)

• MCA8 and MCA9 are essentially identical releases to the environment (explained on next slide) Cases with a water spill MCA8 and MCA9 did not have the building HVAC flow Case Aerosol size Stack Fans Aux. Filters Water Spill MCA6 1 µm Yes No Yes MCA7 1 µm Yes Yes Yes MCA8 1 µm No Yes Yes MCA9 1 µm No No Yes 119

Results of the sensitivity studies Impact of HVAC flow with auxiliary filter flow Ce reactor building and filter behavior Comparison of MCA7 and MCA8 shows the HVAC 1.E-01 Condensing tank MCA7 condensing tank flow sweeps a portion of the small aerosols through 1.E-02 1.E-03 MCA7 offgas the filters and out the stack 1.E-04 MCA7 & MCA8 MCA7 environment MCA8 condensing tank MCA8 offgas

• Most aerosols in the condensing tank Release Fraction (-)

MCA8 environment 1.E-05

• Stack flow capture and aerosol pass-through is more 1.E-06 MCA7 offgas important than only the auxiliary filter flow 1.E-07 Environment 1.E-08 MCA8 offgas Comparison of MCA8 and MCA9 shows the auxiliary 1.E-09 filter has a negligible impact on the environmental 1.E-10 release 1.E-11 0 6 12 18 24

• Capture in the condensing tank, rapid settling in the reactor Time (hr)

Impact of auxiliary filter without HVAC flow cell, and retention in the offgas system (filter pit and stack) Ce settling and filter behavior overwhelms the importance of the auxiliary flow when the 1.E-01 Condensing tank HVAC is not operating 1.E-02 MCA8 condensing tank 1.E-03 MCA8 airborne in the reactor cell MCA8 & MCA9 MCA8 offgas 1.E-04 MCA8 environment Release Fraction (-)

MCA9 condensing tank 1.E-05 Airborne in the MCA9 airborne in the reactor cell reactor cell MCA9 offgas 1.E-06 MCA9 environment Cases with a water spill 1.E-07 Case Aerosol size Stack Fans Aux. Filters Water Spill 1.E-08 MCA8 offgas MCA6 1 µm Yes No Yes 1.E-09 MCA7 1 µm Yes Yes Yes 1.E-10 Environment MCA9 offgas

(<10-11)

MCA8 1 µm No Yes Yes MCA9 1 µm No No Yes 1.E-11 0 6 12 18 24 Time (hr) 120