DOE-HDBK-1129-99, Tritium Handling and Safe Storage

TS NOT MEASUREMENT SENSITIVE DOE-HDBK-1129-99 March 1999 DOE HANDBOOK TRITIUM HANDLING AND SAFE STORAGE U.S. Department of Energy AREA SAFT Washington, D.C. 20585 DISTRIBUTION STATEMENT A. Approved for public release; distribution is unlimited. This document has been reproduced from the best available copy. Available to DOE and DOE contractors from ES&H Technical Information Services, U.S. Department of Energy, (800) 473-4375, fax: (301) 903-9823. Available to the public from the U.S. Department of Commerce, Technology Administration, National Technical Information Service, Springfield, VA 22161; (703) 605-6000. DOE-HDBK-1129-99 iii TABLE OF CONTENTS SECTION PAGE FOREWORD.............................................................................................................................vii ACRONYMS ..............................................................................................................................ix 1.0 INTRODUCTION.................................................................................................................. 1 1.1 Purpose ............................................................................................................................ 1 1.2 Scope ............................................................................................................................... 1 1.3 Applicability....................................................................................................................... 1 1.4 Referenced Material for Further Information...................................................................... 2 2.0 TRITIUM............................................................................................................................... 3 2.1 Radioactive Properties...................................................................................................... 3 2.2 Physical Properties ........................................................................................................... 4 2.3 Chemical Properties.......................................................................................................... 5 2.4 Biological Properties ......................................................................................................... 5 2.5 Preferred Forms................................................................................................................ 7 3.0 BASIC TRITIUM REGULATORY INFORMATION.............................................................. 17 3.1 Tritium Accountability...................................................................................................... 17 3.2 Tritium Safeguards and Security ..................................................................................... 21 3.3 Tritium Facility Safety Analysis and Regulatory Quantity Limits ...................................... 22 3.4 Radiological Materials Quantity Limits............................................................................. 27 3.5 Tritium Unpackaging, Handling, and Packaging Areas, Quantity Limits .......................... 28 3.6 Tritium Waste Collection and Waste Packaging Area, Quantity Limits ............................ 28 3.7 Tritium Focus Group (TFG)............................................................................................. 28

4.0 FACILITY DESIGN............................................................................................................. 28 4.1 Tritium System Philosophy.............................................................................................. 28 4.2 Building Ventilation System............................................................................................. 38 4.3 Chilled Water System ..................................................................................................... 40 4.4 Seismic and Wind Design and Evaluation of Structures and Facilities ............................ 40 4.5 Other Design Considerations .......................................................................................... 42 4.6 Lessons Learned ............................................................................................................ 43 5.0 DESIGN OF EQUIPMENT.................................................................................................. 46 5.1 Material Compatibility...................................................................................................... 47 5.2 First Wall Design............................................................................................................. 52 5.3 Secondary Wall Design................................................................................................... 53 5.4 Cleanup System Design ................................................................................................. 53 5.5 Storage System Design .................................................................................................. 53 5.6 Surveillance and Maintenance ........................................................................................ 55 5.7 Seismic Considerations .................................................................................................. 55 5.8 Fire Scenarios................................................................................................................. 60 5.9 Instrumentation ............................................................................................................... 60 6.0 TRITIUM PURCHASING AND RECEIVING ....................................................................... 65 6.1 Shipping Packages ......................................................................................................... 65 6.2 Product Containers ......................................................................................................... 66 6.3 Valve Container Operations ............................................................................................ 68 DOE-HDBK-1129-99 iv 6.4 Receiving Tritium ............................................................................................................ 72 6.5 Storage of Packaged Nuclear Materials .......................................................................... 72 7.0 PACKAGING AND TRANSPORTATION............................................................................ 73 7.1 General Administrative Packaging and Transport Requirements .................................... 73 7.2 Selection of Proper Packaging........................................................................................ 73 7.3 Package Loading and Preparation for Shipment ............................................................. 79 7.4 Documentation and Records........................................................................................... 80 7.5 Quality Assurance/Control Requirements ....................................................................... 84

8.0 TRITIUM WASTE MANAGEMENT..................................................................................... 85 8.1 Approved Limits for the Release of Contaminated Materials and Property Containing Residual Radioactive...................................................................................................... 86 8.2 Waste Characterization................................................................................................... 93 8.3 Waste Packaging............................................................................................................ 98 8.4 Waste Shipping............................................................................................................... 99 FIGURES: FIGURE 2-1. Rate of tritium decay of one mole of tritium ........................................................... 4 FIGURE 2-2. Pressure versus time in a container of tritium........................................................ 4 FIGURE 2-3. Comparison of aqueous tritium levels found in the nuclear industry ...................... 9 FIGURE 2-4. Dissociation pressure for uranium, hydride, deuteride, and tritide ....................... 11 FIGURE 2-5. Plot of a good fit curve for the dissociation pressure of uranium hydride, deuteride, and tritide ........................................................................................... 11 FIGURE 2-6. Dissociation pressure of palladium hydride and deuteride................................... 13 FIGURE 4-1. Tritium facility single-pass ventilation system...................................................... 30 FIGURE 4-2. Secondary containment ...................................................................................... 32 FIGURE 4-3. Secondary confinement ...................................................................................... 32 FIGURE 4-4. Building confinement system .............................................................................. 33 FIGURE 4-5. Typical gas-to-water tritium removal system flow schematic ............................... 34 FIGURE 4-6. Confinement volume cleanup rate as a function of system time constant, F/V, assuming an exponential dilution rate................................................................. 36 FIGURE 5-1. Development of the seismic equipment list ......................................................... 58 FIGURE 5-2. Comparison of seismic capacity spectra to seismic demand spectra .................. 59 FIGURE 6-1. Use of double valve container ............................................................................. 69 FIGURE 6-2. Purge ports and isolation valves ......................................................................... 70 FIGURE 8-1. Ultimate disposition of tritiated material.............................................................. 97 TABLES: TABLE 2-1. Derived air concentrations for tritium and tritiated water.......................................... 6 TABLE 2-2. ICRP-71 dose conversion factors for inhalation of tritiated particulates ................... 7 TABLE 2-3. Dissociation pressure equation parameters for uranium hydride, deuteride,

and tritide ............................................................................................................. 10 TABLE 2-4. Dissociation pressure equation parameters for palladium hydride and deuteride .. 13 TABLE 3-1. EPA maximum contaminant level for tritium .......................................................... 20 TABLE 3-2. Worst-case doses in rem resulting from release of 1 Ci of tritium.......................... 24 TABLE 5-1. Representative equipment found in tritium facilities............................................... 57 TABLE 5-2. SAM technical specifications................................................................................. 65 TABLE 7-1. Allowable quantities of tritium per 49 CFR 173...................................................... 76 DOE-HDBK-1129-99 v APPENDICES: APPENDIX A: USEFUL NUMERICAL VALUES APPENDIX B: DEFINITIONS APPENDIX C: ASSAY METHODS APPENDIX D: CONTAMINATION AND SURFACE ACTIVITY THRESHOLDS APPENDIX E: RESOURCES USED IN WRITING THIS HANDBOOK APPENDIX F: TRITIUM FOCUS GROUP CHARTER DOE-HDBK-1129-99 vi This page is intentionally blank. DOE-HDBK-1129-99 vii FOREWORD Tritium handling practices have evolved over several decades at Department of Energy tritium facilities. The objective has been to accomplish required tritium work while minimizing and controlling the exposure of workers, the public, and the environment from tritium. This document provides guidance for the handling, storing and shipping of tritium. Literature references furnish information current through mid-1998. This Department of Energy Handbook is approved for use by all DOE components and their contractors. There are no requirements generated by this document, only requirements referenced from other sources. The principal authors, Bill Weaver of DOE-EH and William R. Wall of LLNL and SNLL, wish to acknowledge the contributions of Jim Bachmaier and Bill Fortune of DOE-EH; Elliot Clark and Bob Rabun of WSRC; Ray Hahn of Envirocare; Ron Hafner, Gary Mansfield, Mark Mintz, Robert C. Murray, and Stanley C. Sommer of LLNL; Tobin Oruch and Diana West of LANL; Mike Rogers and Paul Lamberger of Mound; Nazir Kherani of Ontario Hydro; Keith Rule of PPPL; Jeff Paynter of E2 Consulting Engineers; Phil Grant and Barbara Kneece of Wastren; and Jimmy Myers and Elaine Merchant of Parallax, Inc. DOE-HDBK-1129-99 viii This page is intentionally blank. DOE-HDBK-1129-99 ix ACRONYMS AEA Atomic Energy Act of 1954 ALARA As Low As Reasonably Achievable ALI Annual Level of Intake APT Accelerator Production of Tritium ASCE American Society of Civil Engineers ASME American Society of Mechanical Engineers ANSI American National Standards Institute CEDE Committed Effective Dose Equivalent CERCLA Comprehensive Environmental Response, Compensation, and Liability Act CLWR Commercial Light Water Reactor CoC Certificate of Compliance CWA Clean Water Act D&D Decontamination and Decommissioning DAC Derived Air Concentration DBA Design Basis Accident DBE Design Basis Earthquake DCF Dose Conversion Factor DCG Derived Concentration Guide DNFSB Defense Nuclear Facilities Safety Board DOE U.S. Department of Energy DOT U.S. Department of Transportation EDL Economic Discard Limit EH Office of Environment, Safety and Health EPCRA Emergency Planning and Community Right-to-Know Act EPDM Ethylene Propylene Diene Monomer EPA Environmental Protection Agency EIS Environmental Impact Statement FCA Fire Control Area FDTAS Field Deployable Tritium Analysis System FY Fiscal Year HDPE High Density Polyethylene HIVES Highly Invulnerable Encased Safe HMR Hazardous Material Regulations HSV Hydride Storage Vessel HSWA Hazardous and Solid Waste Amendments HTP Hydride Transport Vessel HTV Hydride Transport Vessel HVAC Heating, Ventilation, and Air Conditioning IAEA International Atomic Energy Agency ICRP International Commission on Radiological Protection ISM Integrated Safety Management keV Kilo-Electron-Volts LANL Los Alamos National Laboratory LLNL Lawrence Livermore National Laboratory LDR Land Disposal Restriction LSA Low Specific Activity LDPE Low-Density Polyethylene

DOE-HDBK-1129-99 x LLD Lower Limit of Detection LLW Low-Level Waste mCi Millicurie MCL Maximum Contaminant Level MEI Maximally Exposed Individual mm Millimeters mrem Millirem NFPA National Fire Protection Association NPDWR National Primary Drinking Water Regulation NPH Natural Phenomena Hazard NRC U.S. Nuclear Regulatory Commission PC Performance Category PCB Polychlorinated Biphenyl PMR Palladium Membrane Reactor PPPL Princeton Plasma Physics Laboratory psia pounds per square inch absolute psig pounds per square inch gauge PTFE Polytetrafluoroethylene PVC Polyvinyl Chloride PV Product Vessel RCRA Resource Conservation and Recovery Act RMA Radioactive Materials Area RTF Replacement Tritium Facility SAES Societá Apparecchi Elettrici e Scientifici SAM Surface Activity Monitor SAR Safety Analysis Report SCO Surface Contaminated Object SEL Seismic Equipment List SEP Seismic Evaluation Procedure SMT Stable Metal Tritide SNL Sandia National Laboratory SNLL Sandia National Laboratory, Livermore SNM Special Nuclear Material SRS Savannah River Site SSCs Structures, Systems, and Components TRL Tritium Research Laboratory, Sandia National Laboratory TSD Treatment, Storage, and Disposal TSR Technical Safety Requirement UBC Uniform Building Code UHMWPE Ultra-High-Molecular-Weight Polyethylene WETF Weapons Engineering Tritium Facility, Los Alamos National Laboratory WSRC Westinghouse Savannah River Company DOE-HDBK-1129-99 1 1.0 INTRODUCTION There are several tritium-handling publications, including International Atomic Energy Agency (IAEA) Technical Report Series, Number 324, “Safe Handling of Tritium,” published in 1991; and U.S. Department of Energy (DOE) publications, most notably DOE Handbook, DOE-HDBK-1079- 94, “Primer on Tritium Safe Handling Practices,” published in 1994. The DOE Handbook was developed as an educational supplement and reference for operations and maintenance personnel. Most of the tritium publications are written from a radiological protection perspective. This handbook provides more extensive guidance and advice on the full range of tritium operations. 1.1 Purpose This handbook can be used by personnel involved in the full range of tritium handling from receipt to ultimate disposal. Compliance issues are addressed at each stage of handling. This handbook can also be used as a reference for those individuals involved in real time determination of bounding doses resulting from inadvertent tritium releases. 1.2 Scope This handbook provides useful information for establishing processes and procedures for the receipt, storage, assay, handling, packaging, and shipping of tritium and tritiated wastes. It includes discussions and advice on compliance-based issues and adds insight to those areas that currently possess unclear DOE guidance. It is intended to be a “living document,” being revised periodically. For example, planning for and implementing contamination control as part of normal operation and maintenance activities is an important function in any tritium facility. The best practices from around the complex are presently being accumulated for inclusion in the next revision of this Handbook. Likewise, it is planned that the next revision will include a section on training issues for tritium operators and maintenance personnel. 1.3 Applicability

DOE facilities range from small radiological facilities engaged in operations using a few millicuries (mCi) up to 16,000 Ci of tritium; to large-scale facilities referred to as Non-Reactor Nuclear Facilities, using greater than 16,000 Ci (1.6 grams) of tritium. Guidance in this handbook is applicable to any scale of operations. Some sections of this handbook resulted from consensus agreement between DOE’s Tritium Focus Group members and the authors, representing the views of the Office of Nuclear and Facility Safety, EH-3, an organization within the Office of Environment, Safety and Health at DOE Headquarters. Other sections are strictly the viewpoint of the Office of Nuclear and Facility Safety. This office solicits comments and active discussion of the handbook content in order to improve the handbook on its next revision. DOE-HDBK-1129-99 2 1.4 Referenced Material for Further Information • DOE O 420.1, “Facility Safety” • DOE G 420.1/B-0, “Implementation Guide for use with DOE Orders 420.1 and 440.1, Fire Safety Program” • DOE G 420.1-X, “Implementation Guide for Non-Reactor Nuclear Safety Design Criteria and Explosives” • DOE G 420.1-Y, “Implementation Guide for the Mitigation of Natural Phenomena Hazards for Non-Nuclear Facilities” • DOE O 435.1, “Radioactive Waste Management” • DOE P 450.4, “Safety Management System Policy” • DOE G 450.4-1, Volumes 1 and 2, “Integrated Safety Management System Guide, Revision 0” • DOE O 460.1A, “Packaging and Transportation Safety” • DOE G 460.1-1, “Implementation Guide for Use with DOE O 460.1A” • DOE O 460.2, “Departmental Materials Transportation and Packaging Management” • DOE M 474.1-2, “Manual for Nuclear Materials Management and Safeguards System Reporting and Data Submission” • DOE Order 5400.5, “Radiation Protection of the Public and Environment” • DOE Order 5480.23, “Nuclear Safety Analysis Reports” • DOE-EM-STD-5502-94, “Hazard Baseline Documentation” • DOE Order 5610.12, “Packaging and Transportation of Nuclear Components” • DOE Order 5633.3B, “Control and Accountability of Nuclear Materials” • DOE Order 5820.2A, “Radioactive Waste Management” • DOE-STD-1020-94, “Natural Phenomena Hazards Design and Evaluation Criteria for Department of Energy Facilities,” including Change Notice 1, January 1996 • DOE-STD-1021-93, “Natural Phenomena Hazards Performance Categorization Guidelines for Systems, Structures, and Components,” including Change Notice 1, January 1996 • DOE-STD-1022-94, “Natural Phenomena Hazards Characterization Criteria,” including Change Notice 1, January 1996 • DOE-STD-1023-95, “Natural Phenomena Hazards Assessment Criteria,” including Change Notice 1, January 1996 • DOE-STD-1027-92, “Hazard Categorization and Accident Analysis Techniques for Compliance with DOE Order 5480.23, Nuclear Safety Analysis Reports,” including Change Notice 1, September 1997 • DOE-STD-1120-98, “Integration of Environment, Safety, and Health into Facility Disposition Activities” • DOE-STD-3009-94, “Preparation Guide for U.S. DOE Nonreactor Nuclear Facility Safety Analysis Reports” • DOE-HDBK-1105-96, “Radiological Training for Tritium Facilities” • DOE-HDBK-1079-94, “Primer on Tritium Safe Handling Practices” • DOE-HDBK-3010-94, “Airborne Release Fraction/Rates and Respirable Fractions for

Nonreactor Nuclear Facilities” • DOE/TIC-11268, “A Manual for the Prediction of Blast and Fragment Loading of Structures” • ANS 14-1994, “Internal Dosimetry Standards for Tritium” • ASCE 7-95, “Minimum Design Loads for Buildings and Other Structures” • 10 CFR 20, “Standards for Protection Against Radiation” • 10 CFR 71, “Packaging and Transportation of Radioactive Material” • 10 CFR 830, “Nuclear Safety Management” DOE-HDBK-1129-99 3 • 10 CFR 835, “Occupational Radiation Protection” • 40 CFR 261, “Identification and Listing of Hazardous Waste” • 40 CFR 262, “Standards Applicable to Generators of Hazardous Waste” • 40 CFR 302.4, “Designation of Hazardous Substances” • 49 CFR 172, “Hazardous Materials Table, Special Provisions, Hazardous Materials Communications, Emergency Response Information, and Training Requirements” • 49 CFR 173, “Shippers – General Requirements for Shipments and Packagings” • 49 CFR 177, “Carriage by Public Highway” • 49 CFR 178, “Specifications for Packaging” • 62 FR 62079, Joint NRC/EPA Guidance on Testing Requirements for Mixed Radioactive and Hazardous Waste, November 20, 1997 • IAEA Technical Report Series #324, “Safe Handling of Tritium” • IAEA-SM-181/19, “Estimates of Dry Deposition and Plume Depletion over Forests and Grassland” • ISO 7503-2, “Evaluation of Surface Contamination – Part 2: Tritium Surface Contamination” • Office of Nuclear and Facility Safety Technical Notice 94-01, “Guidelines for Valves in Tritium Service” • OSWER 9928.4-03, Waste Analysis at Facilities that Generate, Treat, Store, and Dispose of Hazardous Wastes: A Guidance Manual, April 1994 • Westinghouse Savannah River Company, WSRC-TR-94-0596, Titanium for Long Term Tritium Storage (U), December 1994. 2.0 TRITIUM Isotopes are elements that have the same atomic number (same number of protons in the nucleus) but of different atomic mass (total number of protons plus neutrons in the nucleus). There are three isotopes of hydrogen. Ordinary hydrogen, referred to as protium (1H 1, atomic mass of 1), is the most abundant element in the universe and has one proton in the nucleus. Heavy hydrogen, referred to as deuterium (1H 2, or D, atomic mass of 2), makes up about 0.015 percent of the hydrogen, and has one proton and one neutron in the nucleus. Radioactive hydrogen, referred to as tritium (1H 3, or T, atomic mass of 3), has one proton and two neutrons in the nucleus. Refer to DOE-HDBK-1079-94 and the IAEA Guide on Safe Handling of Tritium for basic information on tritium, its properties, and compounds. 2.1 Radioactive Properties Tritium is a beta emitter. It decays to 3He by emitting a beta particle (electron) and a neutrino from one of the neutrons in the nucleus. The energy of the beta particle varies from 0 to 18.6 kilo- electron-volts (keV) with an average energy of 5.69 keV. For scientific purposes, the generally accepted value for the half-life of tritium, as measured by Mound Laboratories, is 12.323 ± 0.004 years (4500.88 ± 1.46 days). For DOE accountability purposes, the half-life of tritium, as stated in DOE M 474.1-2, Figure IV-2, is 12.33 +/- 0.06 years. Figure 2-1 shows the rate of decay of one mole of tritium over six half-lives. DOE-HDBK-1129-99 4 FIGURE 2-1. Rate of tritium decay of one mole of tritium 2.2 Physical Properties Tritium gas is colorless, odorless, tasteless, and radioactive. It decays to 3He, a monatomic gas, by emitting an electron and neutrino from the nucleus. Tritium has a high coefficient of diffusion. It readily diffuses through porous substances such as rubber and can also diffuse through metals.

As tritium decays in ca container of constant volume at a constant temperature, the tritium partial pressure decreases and the partial pressure of 3He increases. The pressure in the container approaches twice that of the original container pressure. The rate of pressure change over time is shown in Figure 2-2. 0.0 0.5 1.0 1.5 2.0 0.000 12.323 24.646 36.969 49.292 61.615 73.938 Time in Years Time Period Shown = 6 Half-Lifes P re ss u re A n y U n it s T2 Partial Pressure He3 Partial Pressure T2 Partial Pressure + He3 Partial Pressure P(T2 at t years) =P(T2 initial) e {[t(years) x ln 0.5]/12.323} P(He3 at t years) =2P (T2 initial) (1-P(T2 initial) e {[t (years) x ln 0.5]/12.323} ) P(Total at t years) =P(T2 initial) (2 - e{[t (years) x ln 0.5]/12.323} ) FIGURE 2-2. Pressure versus time in a container of tritium Moles of T 2 and 3He Versus Time, Per Mole of Tritium At the Start, In A Container Starting With Pure Tritium At t=0 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00 0.0 00 3.0 81 6.1 62 9.2 42 12. 32 3 15. 40 4 18. 48 5 21. 56 5 24. 64 6 27. 72 7 30. 80 8 33. 88 8 36. 96 9 40. 05 0 43. 13 1 46. 21 1 49. 29 2 52. 37 3 55. 45 4 58. 53 4 61. 61 5 64. 69 6 67. 77 7 70. 85 7 73. 93 8 Elapsed Time In Years In Increments Of 1/4 Half Life Mo les Per Mo le Of Trit iu m At t=0 mHe3At-t mT2At-t mTotAt-t m=moles m (T2 at t years) = m (T2 initial) e {[t(years) x ln 0.5]/12.323} m (He3 at t years) = 2 m (T2 initial) {1 - e{[t(years) x ln 0.5]/12.323} } m (T2 + He3 at t years) = m (T2 initial) {2 - e {[t(years) x ln 0.5]/12.323} } DOE-HDBK-1129-99 5 Other properties of tritium are listed below. Additional characteristics are given in Appendix A. • Atomic Weight = 3.01605 • Gram Molecular Weight = 6.03210 • Diameter of a tritium atom (approximate) = 1.1 Angstroms • Dissociation energy, T2 to 2T = 4.59 eV • Ionization energy, T to T+ e- = 13.55 eV • Half Life = 12.323 +/- 0.004 years In this handbook, tritium in the form of the oxide (HTO, DTO, and T2O), unless otherwise specified, is HTO. Likewise, tritium in its elemental form (HT, DT, and T2) is HT. 2.3 Chemical Properties The electronic configuration of tritium is the same as protium and deuterium. The chemical properties of the isotopes are also the same. The rates of reaction vary for the different isotopes due to the difference in the atomic masses. Additionally, the energy provided by the radioactive decay of tritium provides the activation energy required so that some reactions will occur with tritium that will not occur with deuterium or hydrogen. Hydrogen is present in almost all materials. If tritium is present in a material containing hydrogen, the tritium atoms will exchange with hydrogen atoms to form a tritiated molecule of the material. 2.4 Biological Properties The body has no need for elemental hydrogen, deuterium, or tritium and does not readily absorb H2, HT, HD, D2, DT, or T2 from inhaled gases or through the skin. A small fraction of the inhaled hydrogen isotopes, in gaseous form, is not exhaled, but is dissolved in the blood stream and then exhaled after a few minutes. Tritium in the form of water (HTO, DTO, and T2O) is adsorbed through the skin and in the lungs from inhaled gases. Tritium in water form is readily retained in the body and remains with a biological half-life of approximately 10 days. Due to the body’s ready adsorption of tritium in the form of tritiated water, exposure to tritiated water in air is up to 25,000 times more hazardous than exposure to gaseous tritium (HT, DT, and T2).

The Derived Air Concentration (DAC) for tritium is the airborne concentration that, if inhaled over a one-year period, would produce approximately a 5-rem dose to the “average” worker. The DAC is derived by the formula: DAC = ALI/2400 where DAC = derived air concentration (µCi/m3) ALI = annual limit of intake (µCi) 2400 = breathing volume for the average worker over 1 year in m3 = .02 m3/min x 60 min/hr x 40 hr/wk x 50 wk/yr The DACs for elemental tritium and tritiated water [1] are listed in Table 2-1. DOE-HDBK-1129-99 6 TABLE 2-1. Derived air concentrations for tritium and tritiated water µCi/ml Bq/m 3 HT HTO 5E-01 2E-05 2E10 8E05 Unlike the situation for oxide and elemental, DACs for tritides (e.g., titanium tritide or hafnium tritide) are currently not defined in relevant regulations and, therefore, the process for radiation posting at 10 percent of the DAC level cannot be finalized for these tritiated compounds. The Office of Worker Protection Programs and Hazard Management within the Office of Worker Health and Safety (EH-5) and Environmental Management’s Office of Safety and Health (EM-4) at DOE Headquarters are currently pursuing this issue, and a white paper, entitled Workplace Indicators and Bioassay Limitations When Dealing With Stable Metal Tritides (SMTs), was authored at Mound in September 1998. The Mound paper uses the terms stable and unstable to refer to the measure of the degree of tritium releasability from the metal. Easily releasable tritium is unstable, and is not a workplace monitoring or bioassay problem, as the tritium is more reasily released as elemental or oxide. The tritides in which the tritium is difficult to release, or stable, are the topic of interest here. The biological effect resulting from these tritides is also not as well understood or modeled as those from oxides and elemental. The biological half-life of some tritides is at least an order of magnitude higher than oxide (see Appendix A-3). Tritium that is adsorbed or contained within respirable particulates (e.g., TiT2, or tritium gas in glass microspheres) presents unique dosimetry problems. The physical configuration of such material will affect the uptake, distribution, retention of the tritium, and will also affect the amount of energy deposited from each transformation. Other complexities (e.g., production of bremsstrahlung radiation) may have to be considered. The possible dose pathways from inhalation of tritiated particulates are direct irradiation of the surrounding tissue; irradiation of the surrounding tissue by bremsstrahlung from the particle; uptake of any tritium oxide (HTO) contamination from the surface of the particle; and absorption (leaching) of tritium from the particle into the bloodstream. The distribution and retention of tritiated particulates is largely dictated by the chemical and physical characteristics of the particles. For example, to gain a perspective on this issue, distribution and retention of inhaled TiT2 has been modeled using a modification of the ICRP-30 Respiratory Tract Model with modified deposition fractions and compartmental half-lives of several hundred days. [2] The factors, which affect the distribution and retention of such particulates, can act to both reduce and enhance the dose received per unit intake. For example, a very large fraction of inhaled activity is quickly eliminated from the body if the particle size distribution is relatively large. The small residence time, coupled with the effective encapsulation of the tritium activity and the short range of the beta particle emitted, all lead to a significant reduction in dose per unit intake (not uptake). Conversely, a postulated or observed long residence time in the lungs may lead to significant lung doses.

Due to the high degree of self-absorption of beta energy that takes place within the particle, only a very small fraction of the beta energy from each transformation is expected to escape the particle. It has been estimated [3] that this self-absorption may reduce the effective dose by as much as a factor of 2,000. Additionally, macrophages are expected to envelop particles in the lungs and add tissue shielding around the particle. DOE-HDBK-1129-99 7 The contribution to dose from bremsstrahlung has been estimated. [2-3] Although this contribution is probably relatively small, it should be considered in the evaluation of lung doses from significant intakes of tritiated particulates. Recently, estimates of effective doses from inhalation of various types of tritiated particulates were made [4]. These estimates are based on in-vitro and animal (rat) studies with titanium tritide. The assumption is made that the tritium, once dissolved, has the same distribution and retention as HTO in the body. The dose conversion factors (DCF) calculated for such intakes in adults are listed in Table 2-2. TABLE 2-2. ICRP-71 dose conversion factors for inhalation of tritiated particulates Lung Absorption Type* DCF (Sv/Bq inhaled) DCF (rem/µCi inhaled) Fast Moderate Slow 6.2 E-12 4.5 E-11 2.6 E-10 2.3 E-05 1.7 E-04 9.6 E-04 *Based on ICRP-66 Respiratory Tract Model and ICRP 60 Tissue Weighting Factors The dose from inhalation of tritiated particulates must be evaluated on a case-by-case basis. In general, as with any suspected intake of tritium, urine samples should be collected. In the event of a suspected significant exposure of tritiated particulates, collection of early (first few days) fecal samples should be considered. Inhaled particulates are expected to be eliminated via this route. Cases involving inhalation of metal tritides have occurred in which significant tritium was seen in the feces, but not in the urine [5]. Collection and analysis of fecal samples (analysis based on fecal sampling must be cognizant of particle matrix absorption effects) allows confirmation of intake and may give information regarding the fractional uptake. 2.5 Preferred Forms Most tritium in the DOE complex exists as a gas, in the form of tritiated water, or as a metal tritide. The preferred form of tritium is dependent upon its use in a process, length of storage, or its classification as a waste. 2.5.1 Characterization of Tritium Forms 2.5.1.a Gaseous Tritium The use, transfer, storage, and shipment of gaseous tritium at or near atmospheric pressure has a long history in the DOE complex and has been safely used for over thirty years. Gaseous tritium at or near atmospheric pressure occupies 22.414 L/mole at 0° C, and approximately 24.2 L/mole at room temperature, and requires approved packages for shipment in either Type A or B quantities. If the containers are not properly designed or if they are damaged, the gas can leak from the container into the environment. Gaseous tritium at ambient pressure is easily handled by most gas handling systems and is a good source for general-purpose use. At low pressure and temperature, the tritium does not penetrate deeply into the container wall. Helium and tritium embrittlement of the container wall is not a significant issue at low pressures even after several years of exposure. As tritium decays, the pressure in the container increases (see Figure 2-2) due to the generation of the monatomic gas

DOE-HDBK-1129-99 8 3He. This pressure increase, at most, would only be double the initial pressure. This factor should be considered during the initial design of the vessel so it does not become an issue. Gaseous tritium at high pressure takes up less space but is more difficult to contain in part due to the potential for tritium and helium embrittlement of the vessel materials. This embrittlement increases the probability of a tritium leak or catastrophic container failure. Unloading high- pressure gas requires specifically designed systems and experienced, skilled operators. 2.5.1.b Metal Tritides Metal tritides reduce the overall volume of the stored tritium, but some of the finely divided metals used are pyrophoric. Some metals form low melting point alloys with the materials used in the construction of the metal tritide containers. Others require extremely high temperatures in order to recover tritium from the material. 2.5.1.c Tritiated Water Tritium in the form of T2O may be difficult to store for long periods due to its corrosive properties. Classified experiments with T2O indicate that pure T2O is corrosive. This corrosiveness is likely due to tritium oxide generating free radicals (OH-) from radiolytic decomposition of water in addition to extra energy from beta decay impinging on surrounding molecules. Additionally, pure T2O, like distilled H2O, will dissolve many materials. No data currently exist that quantify the degree of corrosiveness; therefore, there is no basis to definitively state that the U. S. Environmental Protection Agency (EPA) threshold of corrosivity (i.e., a characteristic of hazardous waste), defined in Code of Federal Regulations 40 CFR 261.22, is not exceeded. The authors believe, however, that only a high-purity product, and not waste, would have a reasonable chance of exceeding this threshold. A broader discussion of the relationship between tritiated water and hazardous wastes is contained in Section 3.1.3. Dilute tritiated water recovered from tritium removal systems has also proven to be corrosive and difficult to contain. In a severe case, storage of tritiated water recovered from tritium removal systems in liquid form at concentrations as low as a few curies per milliliter has corroded through the weld area of stainless steel vessels after only a few days of exposure. In this specific example, it is probable that the extreme corrosive nature of this dilute tritiated water was due, in large measure, to chlorine contamination of the catalyst in the tritium removal system. This corrosion is evidently inhibited by absorption of the tritiated water on clay or in molecular sieve material. Figure 2-3 provides a comparison of the various concentrations of tritiated water found throughout the nuclear industry. DOE-HDBK-1129-99 9 106 105 104 103 102 101 100 10-1 Heavy Water Region (Ci/L) Typical DOE HTO Aqueous Glovebox Waste Canadian Heavy Water Reactor DOE Heavy Water Reactors Light Water Region (Ci/L) 10-1 10-2 10-3 10-5 10-6 10-7 10-8 10-4 Boiling Water Reactor Coolant Pressurized Water Reactor Coolant Three Mile Island Accident-Generated Water NRC Part 20 Release Limit Canadian Drinking Water Limit U.S. Drinking Water Limit Environmental Levels (Ci/L) 10-7 10-8 10-9 10-10 10-11 10-12 10-13 10-14 Sea Water River Water Laboratory LLD Portable System Lower Limit of Detection (LLD) DOE Site Releases U.S. Drinking Water Limit

DOE Derived Concentration Guide Limit FIGURE 2-3. Comparison of aqueous tritium levels found in the nuclear industry 2.5.2 Identity of Common Forms Tritium is usually supplied in gaseous or uranium tritide form. Other forms are also available but are not in common use for bulk shipment. 2.5.2.a Gas In gaseous form, tritium is usually supplied at a purity of 90 to 95 percent tritium (99 percent in research applications) with deuterium and protium as the primary impurities. 2.5.2.b Metal Tritides The use of metal tritide storage beds is one of the most convenient ways of handling tritium. The metal tritide beds have different operating parameters and characteristics, and there are advantages and disadvantages in use of the different materials. 2.5.2.b(1) Uranium Uranium is currently the most useful material for general-purpose tritium storage beds. The equation form for the dissociation pressure of uranium tritide, deuteride, and hydride for the pressure in millimeters of mercury is, Pmm = 10 (-A/T +B) , where A and B are parameters listed in Table 2-3. Westinghouse Savannah River Company (WSRC) has determined that in the hydride transport vessel (HTV), for a uranium to tritium ratio of 1:2.9 in the HTV vessel, the equation is Patm = 10 (-4038.2/T + 6.074) . At room temperature, tritium in the presence of uranium powder forms uranium tritide. The tritium partial pressure in the bed is very low. As a result, at room temperature the bed acts as a vacuum pump that getters all of the hydrogen isotopes. The impurity gases that may be present, such as DOE-HDBK-1129-99 10 3He, N2, O2, or Ar, either remain in the overpressure gas in the bed or react with uranium to form stable compounds. Inert gases, such as Ar, will remain in the overpressure gas, and can be removed by pumping off with a vacuum pump after the pressure has stabilized; however, helium-3 cannot be pumped off without first heating the bed. N2 and O2 will react chemically with the uranium to form stable uranium compounds in the bed, and, therefore, cannot be pumped off at all. As the temperature of the bed is increased, the tritium partial pressure increases as a function of temperature. Depending upon the U:T ratio, it can reach a pressure of around 500 pounds per square inch absolute (psia) at 600oC. The tritium may be transferred into and out of manifolds, containers, etc., by heating the bed and then cooling it to room temperature. The general form of the equation for the dissociation pressure, P, in millimeters of mercury (mm) for uranium hydride, deuteride, and tritide is: log Pmm = - A/T + B or Pmm = 10 –(A/T(K)) + B where T = temperature (K) The values for A and B for hydrogen, deuterium, and tritium, determined by several different investigators, are listed in the following paragraphs, and the results are shown in Table 2-3 and plotted in Figure 2-4. Figure 2-5 is a plot of the general characteristics of uranium hydride, deuteride, and tritide. TABLE 2-3. Dissociation pressure equation parameters for uranium hydride, deuteride, and tritide Metal Tritide Reference Temperature Range (oC) Investigated A (/Kelvin) B Temperature (°C) required to generate a pressure of 1 atmosphere UH3 Spedding, et al. Destriau & Seriot Wicke & Otto Mogard & Cabane Libowitz & Gibb 260 to 430 243 to 412 200 to 430 500 to 650 450 to 650 4500 4255 4450 4480 4410 9.28 9.08 9.20 9.20 9.14 430 415 434 435 432 UD3 Spedding, et al. Destriau & Seriot Wicke & Otto

unspecified unspecified unspecified 4500 4401 4500 9.43 9.01 9.40 414 445 417 UT3 Flotow & Abraham WSRC unspecified unspecified 4471 4038.2 9.461 6.074 408 unspecified DOE-HDBK-1129-99 11 FIGURE 2-4. Dissociation pressure for uranium, hydride, deuteride, and tritide FIGURE 2-5. Plot of a good fit curve for the dissociation pressure of uranium hydride, deuteride, and tritide Each time the tritium is cycled into the system manifolds it picks up impurity gases. These impurities collect in the bed overpressure gas and may be pumped off to remove them after each 0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360 370 380 390 400 410 420 430 440 Temperature Degrees Centigrade D is so ci at io n P re ss u re in t o rr UT3-FlAb UD3-Spd UD3-DeSe UD3-WiOt UH3-Spd UH3-DeSe UH3-WiOt UH3-LiGi) UH3-MoCa UD3 UT3 UH3 Dissociation Pressure of Uranium Tritide, Deuteride, and Hydride 0 10 20 30 40 50 60 20 0 22 0 24 0 26 0 28 0 30 0 32 0 34 0 36 0 38 0 40 0 42 0 44 0 46 0 48 0 50 0 52 0 54 0 56 0 58 0 60 0 62 0 64 0 66 0 Temperature in Degrees Centigrade D is so ci at io n P re ss u re in A tm o sp h er es UD3 UT3 UH3 Good fit equations for the dissociation pressure of uranium tritide, deuteride, and hydride in units of atmospheres. The data from various experimenters was used to extend the range from 200 degrees Centigrade to 650 degrees Centigrade. PatmUD3={10-(4500/T deg K) +9.4 }/760 PatmUT3={10-(4471/T deg K) +9.461 }/760 PatmUH3={10-(4525/T deg K) +9.27}/760 DOE-HDBK-1129-99 12 heating/cooling cycle. Active impurity gases, such as oxygen and nitrogen, are irreversibly removed by reaction with the uranium. Disadvantages to using uranium tritide beds are: 1) uranium powder is pyrophoric; 2) the generation of significant tritium pressure requires a high temperature that results in permeation of tritium through the vessel wall; and 3) the capacity is also permanently reduced by exposure to active impurity gases. 2.5.2.b(2) Palladium Palladium is a metallic element of group 8 in the Periodic Table. The symbol for palladium is Pd, the atomic number is 46, the atomic weight is 106.42, and the melting point is 1554.9°C. At room temperature, palladium absorbs up to 900 times its own volume in hydrogen. It diffuses easily through heated palladium; this is one means of purifying the gas. Finely divided Pd is a good catalyst, and is used for hydrogenation and dehydrogenation reactions. Palladium metal is used in dentistry, as an alloy in making jewelry, and in making surgical instruments, watches, and electrical contacts. Palladium powder is currently the second most used material for general-purpose tritium storage beds. Palladium can be obtained in powdered form and loaded directly into the container used for the metal tritide bed. Palladium was used extensively at both LLNL and SNLL in the tritium storage beds. When the tritium is exposed to the powder, it dissolves in the palladium powder with a maximum Pd:T ratio of approximately 0.7. Palladium powder is not pyrophoric, but it has a higher tritium partial pressure than uranium at room temperature.

At room temperature, tritium, deuterium, and protium dissolve in the palladium powder and the tritium partial pressure in the gas over the powder is approximately 50 torr. The over pressure increases as a function of temperature. As the temperature of the palladium is increased by heating the bed, the tritium partial pressure increases as a function of the temperature and reaches a pressure of around 750 psia at 350oC. The general form of the equation for the dissociation pressure, P, in millimeters of mercury (mm) for palladium hydride, deuteride, and tritide is: log Pmm = (-A/T + B) or Pmm = 10 (-A/T + B) Where T= temperature (K) The values for A and B for hydrogen and deuterium determined by different investigators are given in Table 2-4, and the equations developed by the different experimenters over the temperature range they investigated are plotted in Figure 2-6. DOE-HDBK-1129-99 13 TABLE 2-4. Dissociation pressure equation parameters for palladium hydride and deuteride Metal Tritide Reference Temperature Range (oC) Investigated A B PdHx Gillespe & Hall Gillespe & Hall Ratchford & Castellan Wicke & Nernst 0 to 180 200 to 300 unspecified -78 to 175 1835.4 1877.82 2028.2 2039 7.3278 7.483 7.9776 7.65 PdDx Gillespe & Downs Wicke & Nernst to 300 unspecified 1696.11 1940 7.5138 8.00 The 3He generated as a result of decay of the tritium absorbed in the palladium is trapped in the palladium and is not released until the bed is heated or until the T:3He ratio reaches a particular value. 3He generated as a result of decay in the overpressure gas is not absorbed in the palladium and remains in the overpressure gas. Most impurities do not react with, and are not gettered by, the palladium powder. These impurities accumulate in the overpressure gas as the bed is used to support operations. The generation of significant pressure at low temperature (750 psia at 350oC) is the primary advantage of palladium. The primary disadvantage of palladium is the high partial pressure of tritium over the powder at room temperature (50 torr at room temperature). FIGURE 2-6. Dissociation pressure of palladium hydride and deuteride 0 1000 2000 3000 4000 5000 6000 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 Temperature (C) D is so ci at io n P re ss u re ( to rr ) log Pmm=7.3278-1835.4/T log P m=7.483-1877.82/T log Pmm=7.9776-2028.2/T log Pmm=7.65-2039/T log Pmm=7.5138-1696.11/T log Pmm=8-1940/T log Pmm=7.6-1900/T log Pmm=7.75-1810/T Good Fit Line PdD Good Fit Line PdH DOE-HDBK-1129-99 14 2.5.2.b(3) Titanium Titanium is a metallic element in group 4 of the Periodic Table. The symbol is Ti, the atomic number is 22, the atomic weight is 47.90, and the melting point is 1660oC. It is a low-cost metal, and can absorb and store tritium in a compact solid form at a tritium pressure of approximately 1E- 7 torr. Titanium hydride, TiH2, in powder form, is a black metallic dust that is less prone to spontaneous ignition in air than the parent metal. Finely divided titanium hydride is reported to ignite at 440oC and its dust is an explosion hazard, which dissociates above 288oC. Titanium hydride is used in powder metallurgy, hydrogen production, foamed metals, glass solder, and refractories, and as a gas getter in the electronics industry. Titanium tritide in solid or massive form is stable in air for extended periods of time. Titanium retains the decay helium up to a concentration of 0.3 He atom per Ti atom. Both SRS [6] and Ontario Hydro [7-11] have selected titanium as their long-term storage medium. The SRS titanium beds have an expected useful life of less than 10 years, while the Ontario Hydro beds are expected to be in operation for well over 20 years. The Canadians are not space-restricted, and therefore do not load their beds to the degree that SRS does. Additionally, longer times to maximum helium retention ratio (0.3 He/Ti) can be achieved by diluting the tritium concentration with deuterium or protium.

2.5.2.b(4) Zirconium Zirconium is a metallic element of group 4 in the Periodic Table. The symbol for zirconium is Zr, the atomic number is 40, and the atomic weight is 91.22. Zirconium is flammable as a powder and melts at 1850oC. Zirconium is a hard, lustrous, grayish metal that is strong and ductile, and is used in alloys, pyrotechnics, welding fluxes, and explosives. Zirconium hydride ZrH2 is a flammable gray-black powder and is used in powder metallurgy, nuclear moderators, and as a reducing agent. Finely divided zirconium hydride suspended in air will ignite at 430oC. Zirconium hydride contains about twice as many hydrogen atoms per unit of volume as liquid hydrogen. Massive zirconium hydride is stable in air for extended periods of time at temperatures below 600oC. Certain DOE radioactive zirconium fines (which are destined for disposal) are managed as D001 mixed ignitable wastes. These radioactive zirconium fines are pyrophoric under 40 CFR 261.21(a)(2) – i.e., they are capable of causing fire through friction. If zirconium used for tritium storage beds is destined for disposal (i.e., constitutes a waste), the RCRA hazardous waste characteristic of ignitability must be analyzed. A broader discussion of RCRA hazardous waste is contained in Section 3.1.3. 2.5.2.b(5) Societá Apparecchi Elettrici e Scientifici (SAES) Getters During the past several years, investigations have been conducted concerning the use of SAES Getters to remove tritium from tritium-contaminated gaseous waste streams. These investigations have concentrated on getters that cracked the gases containing tritium and removed the resulting free tritium from the gas stream. The primary advantage of these getter systems is that tritium is not converted by the tritium removal system to the more radiotoxic tritiated water. Additionally, the tritium can be recovered in gaseous form from the getter, purified, and reused. The materials being tested include those manufactured in the form of pressed pellets that can be used in low pressure drop packed bed reactors designed for the size required by the application DOE-HDBK-1129-99 15 flow rate and lifetime requirements. The basic strategy currently being implemented in prototype systems is to crack the molecules on a hot getter and remove the non-tritiated reactive impurities that interfere with the performance of the hydrogen gettering alloys. Following purification, the gas is passed through a hydrogen gettering bed to remove the hydrogen isotopes from the gas stream. 2.5.2.b(6) LaNi5 Based Alloys The use of lanthanum-nickel hydrides has been a continuing topic of interest, and as recently as 1997, promising results for hydrogen storage have been reported (“LaNi5 Intermetallic Hydride,” extracted from State-of-the-Art Review of Hydrogen Storage in Reversible Metal Hydrides for Military Fuel Cell Applications, Gary Sandrock, Ph.D., for the Department of the Navy, Office of Naval Research, N00014-97-M-0001, July 24, 1997). Promising research results were also reported in the literature in 1988 (“Hydrogen Isotope Sorption Properties of LaNi3Mn2 Alloy as a Candidate for the Tritium Storage Material,” T. Ide et al., Sumitomo Heavy Industries, Ltd. and H. Yoshida et al., Japan Atomic Research Institute, published in Fusion Technology, September 1988). Earlier research at Mound, however, in the 1970s and early 1980s indicated that lanthanum-nickel-based alloys were not appropriate for tritium service due, in part, to disproportionation. When cycled, the LaNi5 had a tendency to separate to form the parent metals (La or Ni) or different alloys. The disproportionation tended to change the pressure, concentration, and temperature properties of the metal/alloy mix and increase the quantity of tritium bound in the heel that was not easily recoverable.

Further investigations will be needed to better ascertain the capabilities of LaNi alloys for tritium service. 2.5.2.c Absorbed Water Molecular sieve material is used in tritium removal systems for removal of water contaminated with tritium. Systems such as tritium removal systems, effluent recovery systems, and cleanup systems remove tritium from a gas by cracking the tritium-containing components on a heated precious metal catalyst. The free tritium then combines with oxygen in the gas stream to form tritiated water. The gas stream is then cooled to room temperature, and the water contained in the gas stream, including the tritiated water, is removed by a molecular sieve trap. A molecular sieve will hold about 18 percent water by weight, and the sieve may be regenerated to remove the water so it can be reused. The issue of flammability limits of these containers is discussed in Section 3.1.2. Tritiated water absorbed on molecular sieve is not corrosive and may be stored in this way for long periods without damage to the container wall. Water contaminated with HTO is also stored on clay. The common method of solidification of tritium-contaminated wastewater for disposal is to solidify the water on clay so that it can be classified as solid waste. Clay will hold approximately 60 percent water by volume. Waste disposal sites generally require the use of 100 percent more clay than required to solidify the water, and, as a result, the water is generally limited to 30 percent of the volume of the clay for waste solidification purposes. Water absorbed on clay is not corrosive and may be stored for long periods without damage to the container wall. DOE-HDBK-1129-99 16 2.5.3 Summary 2.5.3.a Best for Storage Conditions The decision on storage media is a function of the storage length and frequency of unloadings. Media range from gas (short time frame, many movements) to titanium (long time frame, few or no movements) with other media (e.g., uranium) in between. Although the preferred form for storage is a metal tritide and the least desirable form is a liquid, there are always exceptions to this rule. Factors to be considered include: Solid (Metal Tritides): Unless already in solid form, tritium is not readily available as a solid metal tritide and requires conversion before storage. Metal tritides can store large quantities of tritium without occupying large volumes, but the storage containers are more complex than gaseous storage containers. Depending upon the metallic tritide chosen for storage, there are both advantages and disadvantages. Titanium tritide is very stable, even when exposed to air, but it is more difficult to recover tritium from the titanium than from other metals. Stored as uranium tritide, the tritium can be easily and quickly recovered and provides for the removal of most impurities that might accumulate during storage. However, uranium powder is also pyrophoric, and starts releasing 3He after a few months. Liquid (T2O): Tritium is not readily available in water form and requires conversion before storage. It is up to 25,000 times more hazardous in oxide form than in elemental form. It takes very little space, but is difficult to store due to the corrosivity of the water. The tritiated liquid can be solidified on clay, molecular sieves, or polymers prior to disposal. It may also be converted to its gaseous form for storage purposes. In either case, the final decision should be based on the quantity of tritium and the tritium concentration of the water to be stored.

Gas (T2): Tritium is readily available in gaseous form. A great deal of experience already exists on the design of gaseous tritium storage systems. As a gas it takes up more volume than as a liquid or solid, and can be easily released to the environment if the tritium container is breached. Gas also presents a flammability vulnerability. 2.5.3.b Best for Operations/Process Solid (Metal Tritides): When used as a gas in research where tritium is issued and returned as a gas, there are advantages to the use of uranium beds for storage. The heating/cooling cycle used to store and recover tritium from the bed results in routine removal of the 3He and other impurities from the tritium supply. It can be reused in other processes at reasonably high purity. Additionally, storage as a metal tritide allows the bed to be used as a pressure generator and, in some cases, eliminates the need for mechanical pumps. Liquid (T2O): Unless the process itself uses tritium in the form of water, there are no advantages to storage of tritium in liquid form for operations. Gas (T2): Tritium is primarily used in gaseous form, purified in gaseous form, assayed in gaseous form, and is more useful in this form than any other form. As a result, storage in gaseous form for operations is appropriate. DOE-HDBK-1129-99 17 2.5.3.c Best for Disposal Conditions Disposing of tritium in the form of a liquid waste or gaseous waste is difficult. Generally speaking, waste tritium is converted to solid form so that the material can be disposed of as a solid low-level (radioactive) waste, assuming there is no RCRA hazardous component. Solid (Metal Tritide): It is possible to dispose of gaseous tritium by converting it to a solid metal tritide. However, the disposal site requires that the metal tritide not be pyrophoric. If it is in particulate form, the metal tritide must be contained to meet disposal site requirements. Liquid (T2O): If the waste is in gaseous form, the tritium is normally removed from the gas mixture and is reused. If the concentration of tritium in the waste gas is too low to make recovery of the tritium economically worthwhile, the waste gas is sent to an effluent processing system where the tritium is removed before the gases are released to the environment. The current effluent processing systems remove tritium from the waste gas by converting it to water. The water is then solidified on molecular sieve, clay, mixtures of clay and cement, or Stergo superabsorbent (discussed in Section 8.1.4.b(2)), and is then packaged as solid waste and shipped to the disposal site. Gas (T2): Waste disposal sites generally will not accept packages containing pressurized gases or those in which a potential exists for generating 1.5 atmospheres (absolute) of pressure over time. 3.0 BASIC TRITIUM REGULATORY INFORMATION Due to its more hazardous profile, most of the regulatory interest in tritium is concerned with the oxide form. Figure 2-3 pictorially illustrates various concentrations and regulatory set points. The radiological materials inventory for tritium accounting purposes may not coincide with the radiological materials inventory for safety analysis report (SAR) purposes, which may not coincide with the radiological materials inventory for Environmental Impact Statement (EIS) purposes. This is due to prescribed allowances for excluding various portions of the inventory, as discussed in Sections 3.1 and 3.3. 3.1 Tritium Accountability

3.1.1 Radiological Materials Inventory The Atomic Energy Act describes three categories of materials: Byproduct, Source and Special Nuclear Material (SNM). There are also three categories of nuclear materials described in DOE Order 5633.3B, “Control and Accountability of Nuclear Materials”: SNM, Source and Other. The order designates that tritium is an “Other” category material and is accountable nuclear material. Tritium is accountable for DOE down to 0.01 grams. Items that contain tritium quantities of 0.005 grams or greater are rounded to the nearest 0.01 grams and become accountable items. Items that contain less than 0.005 grams round to zero and are not accountable items. For the purposes of accountability, the radiological materials inventory of tritium is the sum of the tritium quantities contained in the accountable items. This sum does not include any items that contain less than DOE-HDBK-1129-99 18 0.005 grams of tritium or other items that are not part of the facility accountable nuclear materials inventory. Quantities of tritium contained in waste may be part of the facility accountable nuclear materials inventory until it is removed from the facility to a waste accumulation area for storage or to a waste packaging area for packaging. Waste can also be put in a prescribed form, as specified in DOE Order 5633.3B. Individual waste items that contain at least 0.005 grams of tritium are accountable nuclear materials items and become part of the facility accountable nuclear materials inventory. Tritium contained in water (H2O or D2O) used as a moderator in a nuclear reactor is not an accountable material. DOE requires that a tritium facility establish material control and accountability systems to provide accurate nuclear materials inventory information. The facility tritium inventory and scrap levels of tritium must be minimized consistent with the operational needs and safeguards practices of the facility. For each facility, a materials control and accountability program must be established for all nuclear materials on inventory, including those designated as uneconomical to recover. The facility materials control and accountability system should include the following: • Accounting System Database • Account Structure • Records and Reports • Physical Inventories • Periodic Physical Inventories • Special Inventories • Inventory Verification/Confirmation Measurements • Measurements and Measurement Control • Organization • Selection and Qualification of Measurement Methods • Training and Qualification of Measurement Personnel • Measurement Systems • Measurement Control • Material Transfers • External Transfers • Internal Transfers • Material Control Indicators • Shipper/Receiver Difference Assessment • Inventory Difference Evaluation • Evaluation of Other Inventory Adjustments • Documentation and Reporting Forms • Procedures and Requirements 3.1.2 EPA Maximum Contaminant Level for Tritium The current National Primary Drinking Water Regulation (NPDWR) for beta- and photon-emitting radionuclides is 4 millirem (mrem)/year. The regulatory compliance level for tritium, corresponding to the 4 mrem/year level determined by EPA (40 CFR Part 141) is 20,000 pCi/L. By comparison, DOE-HDBK-1129-99 19 the World Health Organization and Canadian levels are approximately 200,000 pCi/L (189,000 pCi/L for Canada and 210,000 pCi/L for the World Health Organization. These are based on a value of 10 percent of the public dose limits as calculated in accordance with ICRP-60 recommendations. [12]

EPA calculated a maximum contaminant level (MCL) for tritium of 20,000 pCi/L in 1976 based on values for worker exposure contained in Handbook 69, “Maximum Permissible Concentrations of Radionuclides in Air and Water for Occupational Exposure,” published by the National Bureau of Standards in 1959 and amended in 1963. This calculation assumes that an extra dose resulting from organically bound tritium, equal to a 20 percent increase over that determined for worker exposure dose in Handbook 69, should be factored into the MCL. The resulting MCL (20,000 pCi/L) is based on not exceeding 4 mrem/year on a total body basis, and assumes a daily rate of ingestion of 2L of water. In July 1991, EPA proposed revised MCLs for radionuclides based on its dosimetric model RADRISK, which included many of the concepts and assumptions included in the effective dose equivalent method (ICRP 30). In the proposed regulation, EPA determined that the MCL for tritium was 60,900 pCi/L. DOE provided comments to EPA suggesting that the appropriate value for the tritium MCL should be 80,000 pCi/L, based on dose to soft body tissues. This value is consistent with the Derived Concentration Guide (DCG) value in DOE Order 5400.5, “Radiation Protection of the Public and the Environment.” Apparently, EPA used its RADRISK model to calculate the 60,900 pCi/L value, which differed from the models and parameters recommended by ICRP 30. The proposed regulation has not been finalized. Projections from EPA are that a re-proposed rule is expected no sooner than December 2000. In early 1997, EPA prepared a Direct Final Rule that would have updated the methodology used in the 1976 final NPDWR. This methodology would be consistent with the method contained in ICRP 30 and Federal Guidance Report 11; i.e., the effective dose equivalent concept–rather than the “total body or organ dose equivalent” method used in 1976. This direct final rule would have retained the 4 mrem/year standard, which is based on daily ingestion of 2L of water and corresponds to a lifetime cancer risk of 10-4. In April 1997, however, EPA determined that the Safe Drinking Water Act Amendments of 1996 contained a provision that would preclude EPA from promulgating a standard that results in increased risk, in the context of revising an existing standard. Although the direct final rule was described as a technical modification to the methodology used to calculate a regulatory compliance level, and that the basic standard (4 mrem/year) was unchanged, the concentration levels for many of the radionuclides would increase, thereby increasing risk. The level for tritium is the most dramatic example, since the change in methodology would increase the regulatory compliance level from 20,000 pCi/L to 86,000 pCi/L. To date, EPA has not promulgated this direct final rule. It is the current understanding of the Office of Environment Safety and Health (EH’s) that EPA is considering a risk-based approach to setting NPDWRs for radionuclides, rather than the dose- based approach currently followed. No decisions have been made. EH is not aware of any plan for providing an advance or draft rule for review and comment on this approach. If a risk-based approach were to be adopted, it is likely that the MCL for tritium would be on the order of 30,000 pCi/L rather than the 60,900 pCi/L proposed in 1991. A summary of the rule and proposals is given in Table 3-1. DOE-HDBK-1129-99 20 TABLE 3-1. EPA maximum contaminant level for tritium

Rule DOE Effects MCL Source 1976 Final Rule (41 FR 28404) 4 mrem/year Total Body or Organ Dose Equivalent 20,000 pCi/L NBS Handbook 69 1991 Proposed Rule (56 FR 33050) 4 mrem/ year Effective Dose Equivalent 60,900 pCi/L RADRISK (ICRP 30) 1997 Direct Final Rule (Never Issued) 4 mrem/year Effective Dose Equivalent 86,000 pCi/L Federal Guidance Report #11 3.1.3 Reactor- Versus Accelerator-Produced Tritium Under the implementing regulations of the Resource Conservation and Recovery Act (RCRA) [specifically at 40 CFR 261.4(a)(4)], source, special nuclear and byproduct material as defined by the Atomic Energy Act of 1954 (AEA) is excluded from the definition of solid waste (and thus, from the RCRA hazardous waste management requirements). The AEA definition of byproduct material includes “any radioactive material (except special nuclear material) yielded in or made radioactive by exposure to the radiation incident to the process of producing or utilizing special nuclear material.” Tritium produced in a reactor (i.e., through the use of special nuclear or source materials) meets the definition of “by-product” material; and, therefore, the waste streams derived from reactor-produced tritium are excluded from RCRA regulation (provided such waste streams do not also contain a RCRA hazardous waste component in addition to the by-product material component). Accelerator-produced tritium, on the other hand, does not qualify for this exclusion (since the tritium is produced by a linear accelerator, and does not involve the production or utilization of special nuclear materials or the extraction or concentration of source material). Thus, for reactor-generated tritium waste to be considered hazardous waste, the waste stream also would have to contain a RCRA listed or non-tritium derived characteristic hazardous waste component. For accelerator-produced tritium waste streams (with only tritium, or tritium and other non-hazardous waste components), the waste stream would not be excluded from RCRA. However, unless such tritium wastes exhibit one of the characteristics of RCRA hazardous waste [ignitability, corrosivity, reactivity, or toxicity (40 CFR 261.21 - .24)], the waste streams would not need to be managed as a RCRA hazardous/mixed waste. Pursuant to the RCRA regulations, it is the responsibility of the generator of a waste to determine if that waste is subject to the hazardous waste requirements [40 CFR 262.11]. Categories of characteristic hazardous waste (and associated properties) that appear to have some potential to apply to certain accelerator-produced tritium wastes are as follows: • Ignitability [40 CFR 261.21] – Ignitable wastes are solid wastes that exhibit any of the following properties: liquids with a flash point of less than 60°C (140°F); solids that are capable of causing fires through friction, absorption of moisture, or spontaneous chemical changes; ignitable compressed gases as defined in 49 CFR 173.300; or oxidizers as defined in 49 CFR 173.151. • Corrosivity [40 CFR 261.22] – Corrosive wastes are solid wastes that exhibit any of the following properties: an aqueous material with pH <2 or >12.5; or a liquid that corrodes steel at a rate greater than ¼ inch per year at a temperature of 55°C (130°F). DOE-HDBK-1129-99 21

• Reactivity [40 CFR 261.23] – Reactive wastes are solid wastes that exhibit any of the following properties: (1) they are normally unstable and readily undergo violent change without detonating; (2) they react violently with water; (3) they form potentially explosive mixtures with water; (4) when mixed with water, they generate toxic gases, vapors, or fumes in a quantity sufficient to present a danger to human health or the environment; (5) they are a cyanide or sulfide bearing waste which, when exposed to pH conditions between 2 and 12.5, can generate toxic gases, vapors, or fumes in a quantity sufficient to present a danger to human health or the environment; (6) they are capable of detonation or explosive reaction if subjected to a strong initiating source or if heated under confinement; (7) they are readily capable of detonation or explosive decomposition or reaction at standard temperature and pressure; (8) they are a forbidden explosive as defined in 49 CFR 173.51, or a Class A explosive as defined in 49 CFR 173.53 or a Class B explosive as defined in 49 CFR 173.88. The discussion in Section 2.5.1.c provides a qualitative argument for the determination that for waste, the characteristic of corrosivity typically are not exhibited. However, little data are currently available to confirm whether or not the vapor space of some tritium containers (e.g. tritium oxide adsorbed on molecular sieves) would exhibit the hazardous characteristic of ignitability or reactivity over time due to radiolytic decay. As explained above, reactor-generated tritium waste streams that do not contain a hazardous waste component may be excluded from the RCRA hazardous waste regulations pursuant to 40 CFR 261.4(a)(4). This may be the case even if sufficient quantities of both hydrogen and oxygen are present to exhibit characteristics of ignitability or reactivity. This is based on a regulatory policy that EPA has applied in certain cases whereby residuals derived from the management of exempt or excluded waste retain the exemption or exclusion. [13-17] However (as indicated above), if a tritium waste (reactor- or accelerator- produced) also contains a distinct hazardous waste component, the waste stream should be managed as a radioactive mixed waste under the AEA and RCRA. The application of RCRA to certain tritium waste streams may be subject to regulatory interpretation and enforcement discretion. With this in mind, it is recommended that determinations as to whether or not certain tritium wastes constitute RCRA hazardous waste be discussed and validated with the appropriate regulatory agency (i.e., the EPA Region or RCRA- authorized State agency). Section 8.2.2 provides a flow diagram and expanded discussion on this issue, in addition to the definitions and options for tritium recovery and disposal. 3.2 Tritium Safeguards and Security Tritium is a nuclear material of strategic importance and must be safeguarded from theft or diversion. Safeguard requirements are based on the category of the nuclear material as specified in Figure I-2, “Nuclear Material Safeguards Categories,” of DOE Order 5633.3B; i.e., Category I through Category IV, and the attractiveness level; i.e., Attractiveness Level A through E. Tritium is either a Category III or Category IV material depending upon the following: • Category III – Weapons or test components containing reportable quantities of tritium. Deuterium-tritium mixtures or metal tritides that can be easily decomposed to tritium gas, containing greater than 50 grams of tritium (isotope) with a tritium isotopic fraction of 20 percent or greater.

• Category IV – All other reportable quantities, isotopic fractions, types, and forms of tritium. DOE-HDBK-1129-99 22 3.3 Tritium Facility Safety Analysis and Regulatory Quantity Limits A safety analysis is required by DOE Order 5480.23, “Nuclear Safety Analysis Reports,” for all nuclear facilities. Irrespective of these requirements, the good practices associated with the implementation of integrated safety management principles necessitate that hazards be identified and controlled, which is a major step in the safety analysis process. 3.3.1 Safety Analysis 3.3.1.a Facility Requirements There are a few fundamental assumptions normally made when performing safety analyses on tritium facilities, which if not satisfied will require more detailed analysis or corrective actions. These include: • The integrity of the primary container should be ensured for all normal operations, anticipated operational occurrences, and for the design basis accidents (DBA) it is required to withstand. • If the facility structure is not part of the secondary barrier , its failure as a result of severe natural phenomena or other postulated DBAs should not prevent the primary container or the secondary containment/confinement systems from performing their necessary safety functions. • When secondary containers (secondaries) are used, a tritium effluent removal system to handle tritium leakage from primary containers is recommended by this Handbook but not required by regulations. 3.3.1.b Radiological Materials Inventory Attachment 1 of DOE-STD-1027-92, “Hazard Categorization and Accident Analysis Techniques for Compliance with DOE Order 5480.23, Nuclear Safety Analysis Reports,” Change Notice 1 states, “Additionally, material contained in Department of Transportation (DOT) Type B shipping containers (with or without overpack) may be excluded from summation of a facility’s radioactive inventory if the Certificates of Compliance are kept current and the materials stored are authorized by the Certificate. However, Type B containers (see Section 6.1) without an overpack should have heat protection provided by the facility’s fire suppression system.” Further guidance from this handbook expands this clarification to mean that if Type B containers are currently certified to withstand the credible facility accidents in which they are located, then their inventories could be excluded for SAR purposes. If the containers are not currently qualified (qualifications expire on a fixed schedule), then their contents must be included in the summation of facility inventory and be included in accident analyses, irrespective of accident conditions. If, however, the container is currently certified, then a comparison of the conditions resulting from transportation and facility accidents is performed. For example, consider a currently certified Type B container that has been qualified to withstand fire and crush loads (i.e., Type B Hypothetical Accident Crush Test loads) associated with a transportation accident. If these transportation accident conditions are more severe than the associated credible facility accident conditions (e.g., fire and seismic crush loads), then the inventory in these containers can be excluded (irrespective of fire suppression system coverage) in the safety analysis. If the facility fire conditions exceed the transportation fire conditions, the inventory is included. Note that credit for safety-grade or safety- DOE-HDBK-1129-99 23

significant class fire suppression will affect the selection of the credible facility fire scenario and therefore the fire conditions for comparison to the transportation fire conditions. If the transportation accident conditions are less severe than the facility accident conditions, then the contents are included in the accident scenarios. For example, if the seismic crush loads exceed the transportation crush loads, the contents cannot be excluded from the facility seismic accident scenarios. They could, however, still be excluded from the facility fire scenarios if the transportation fire conditions bound the facility fire conditions. DOE-STD-1027-92 Change Notice 1 does not provide explicit exemption criteria for other than Type B containers. Future guidance for other than Type B containers, in addition to the guidance concerning the degree and amount of residual tritium inventory to include in safety analyses, will be included in the next revision of this handbook. The threshold quantities of radiological material inventory [18-20] are as follows: • Hazard Category I: (Nuclear Facility) Generally Limited to Nuclear Reactors: Regardless of the quantity of tritium in the inventory, a facility that handles only tritium is not classified, by tritium quantity alone, as a Hazard Category I. The PSO may designate a tritium facility as Category I if the potential for significant offsite consequences exists. [18] • Hazard Category II: (Nuclear Facility): To be classified as a Category II nuclear facility, the facility tritium inventory must be > 30 grams. [18] • Hazard Category III: (Nuclear Facility): To be classified as a Category III nuclear facility, the facility tritium inventory must be > 1.6 and < 30 grams. [18] • Radiological Facilities: (Not Nuclear Facilities): Facilities that have less than 1.6 grams of tritium in the facility radiological material inventory are not classified as “nuclear facilities,” unless they contain higher than threshold values of other radionuclides. These facilities are classified as radiological facilities. [18] • Non-Radiological Facilities: Facilities that have less than 100 Ci of tritium are classified by the DOE Offices of Defense Programs (DP) and Environmental Management (EM) as non- radiological facilities, unless they contain other radionuclides above reportable quantities. [19, 20] 3.3.1.c Material Release Assumptions Once the total inventory available for release is known, the appropriate source term can be calculated. The components of the source term, as described in DOE-HDBK-3010-94, “Airborne Release Fraction/Rates and Respirable Fractions for Nonreactor Nuclear Facilities,” are material at risk, damage ratio, airborne release fraction, respirable fraction, and leakpath factor. The factors for airborne release and respirable fractions are normally assumed to be 1.0 for elemental tritium and oxide. The other factors are facility specific. For fire scenarios, the fraction of the release assumed to be oxide is normally 100 percent. Typical scenarios to review for inclusion in safety analyses are area fires (single and multiple Fire Control Areas (FCA)), full facility fire, leaks/spills, hydrogen explosions, and Natural Phenomena Hazards (NPH) events (the design basis earthquake (DBE) may have a seismic induced fire, which usually results in the bounding DOE-HDBK-1129-99 24 accident). DOE-STD-3009-94, “Preparation Guide for U.S. DOE Nonreactor Nuclear Facility Safety Analysis Reports,” provides detailed guidance for performing accident analyses.

One method of estimating dose is through the use of the HOTSPOT health physics codes developed by Steven C. Homan of LLNL. These codes operate on a PC, and provide a fast calculational tool for evaluating accidental releases. They are a first-order approximation of the radiation effects associated with short-term (less than 24 hours) atmospheric release of radioactive materials. Reference [21] is a detailed analysis by Bill Wall of LLNL for estimating maximum dose from a tritium release that uses HOTSPOT methodology as its base. Fifteen DOE tritium release locations (note that ground release locations will differ slightly from stack locations) were analyzed, the results of which should be considered as preliminary, and are presented in Table 3-2. A set of consistent conservative assumptions was used for all locations. Dose results are presented in current SARs, and are not directly intercomparable due to differing assumption sets among sites. These were calculated, in part, to help provide personnel involved in a real-time determination of bounding doses from tritium releases. These tabled values are worst-case doses for the associated curie releases at these locations, and cannot be compared to the dose values calculated in the SARs, which are normally estimated by more detailed computational methods. Additionally, the use of actual meteorological conditions and site topography would mitigate the tabled values here to smaller doses. An estimation of the impact of topography on the deposition velocity can be found in the document, “Estimates of Dry Deposition and Plume Depletion over Forests and Grassland” (IAEA-SM-181/19), Ray Hosker, Jr. In fact, in many cases, the atmospheric conditions needed for the worst-case doses from ground and stack releases presented here are mutually exclusive. Indeed, one of the objectives of reference [21] was to determine which combinations of conditions and parameters result in worst-case doses. Readers are encouraged to review and comment on the detailed analysis presented in [21]. TABLE 3-2. Worst-case doses in rem resulting from release of 1 Ci of tritium SAVANNAH RIVER SITE OXIDE ELEMENTAL Facility Ground Stack Ground Stack Site Boundary 6.0E-7 1.8E-7 2.4E-11 7.2E-12233-H 100 meters 1.6E-3 2.1E-5 6.4E-8 8.4E-10 Site Boundary 6.0E-7 1.5E-7 2.4E-11 6.0E-12238-H 100 meters 1.6E-3 9.8E-6 6.4E-8 3.9E-10 Site Boundary 6.0E-7 3.7E-8 2.4E-11 1.5E-12234-H 100 meters 1.6E-3 2.1E-7 6.4E-8 8.4E-12 Site Boundary 6.0E-7 3.7E-8 2.4E-11 1.5E-12232-H-line 1&2 100 meters 1.6E-3 2.1E-7 6.4E-8 8.4E-12 Site Boundary 6.0E-7 3.7E-8 2.4E-11 1.5E-12232-H-line 3 100 meters 1.6E-3 2.1E-7 6.4E-8 8.4E-12 MOUND OXIDE ELEMENTAL Facility Ground Stack Ground Stack Site Boundary 1.6E-3 3.2E-6* 6.4E-8 1.3E-10*NCDPF 100 meters 1.6E-3 2.3E-6 6.4E-8 9.2E-11 Site Boundary 1.6E-3 2.5E-6* 6.4E-8 1.0E-10*HEFS 100 meters 1.6E-3 1.4E-6 6.4E-8 5.6E-11 Site Boundary 8.0E-4 2.5E-6* 3.2E-8 1.0E-10*SWIC 100 meters 1.6E-3 1.4E-6 6.4E-8 5.6E-11 DOE-HDBK-1129-99 25 OXIDE ELEMENTAL Facility Ground Stack Ground Stack Site Boundary** 7.0E-4 1.5E-6* 2.8E-8 6.0E-11*T West 100 meters 1.6E-3 2.1E-7 6.4E-8 8.4E-12 Site Boundary** 2.8E-4 1.4E-6 1.1E-8 5.6E-11T East 100 meters 1.6E-3 2.1E-7 6.4E-8 8.4E-12 LOS ALAMOS NATIONAL LABORATORY OXIDE ELEMENTAL Facility Ground Stack Ground Stack WETF #1/#2 Site Boundary 2.0E-5 8.5E-7* 8.0E-10 3.4E-11* 100 meters 1.6E-3 8.7E-6 6.4E-8 3.5E-10 TSFF Site Boundary 3.0E-5 8.5E-7* 1.2E-9 3.4E-11* 100 meters 1.6E-3 8.7E-6 6.4E-8 3.5E-10

TSTA Site Boundary 1.2E-4 2.3E-6 4.8E-9 9.2E-11 100 meters 1.6E-3 6.1E-6 6.4E-8 2.4E-10 LAWRENCE LIVERMORE NATIONAL LABORATORY OXIDE ELEMENTAL Facility Ground Stack Ground Stack Site Boundary 1.5E-4 2.8E-6 6.0E-9 1.1E-10Bldg. 331, #1/#2 100 meters 1.6E-3 6.1E-6 6.4E-8 2.4E-10 PANTEX Facility OXIDE ELEMENTAL Ground Stack Ground Stack Site Boundary 1.2E-5 NA 4.8E-10 NASNM Facility Building 12-116 100 meters 1.6E-3 NA 6.4E-8 NA * This value is the worst-case dose to the public, which does not occur on site boundary but at a larger distance from the facility, as estimated in reference [22]. The individual is designated to be the maximally exposed member of the public. ** Mound Central Operations Support Building is not considered a public boundary by DOE Miamisburg interpretations. [22] In some cases, primarily at the DOE Mound facility, the worst-case offsite dose does not occur at the site boundary, but can occur some distance away (due, in part, to high stacks and short distance to site boundaries). Additionally, the DOE Mound Office defines the personnel using the buildings onsite that have been turned over to the community (e.g., the Central Operations Support building) as collocated workers and not as members of the public [22]. If these individuals were instead reclassified as members of the public, then use of this calculational method would result in an increase in some of the tabled site boundary doses from ground releases at Mound. It should be noted that several federal environmental laws (e.g., Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA), Emergency Planning and Community Right-to-Know Act (EPCRA), and the Clean Water Act (CWA) require that releases of hazardous substances above reportable quantities (for tritium, this is 100 curies in 24 hours) be reported to the National Response Center. In many cases, the curie levels are back-calculated from offsite dose receptor requirements (see Section 8.1.3). Current guidance from “Hazardous Substance Release Reporting under CERCLA, EPCRA §304, and DOE Emergency Management System/Occurrence Reporting Requirements,” DOE/EH-0383, makes a distinction between “normal” or “routine” releases and “abnormal” or “accidental” releases, and suggests reporting abnormal or accidental releases, even if they are below federally permitted levels. DOE-HDBK-1129-99 26 3.3.2 Integrated Safety Management All tritium-related operations and activities, including design, construction, system acceptance and turnover, operations, shutdown, deactivation, and decommission for Category 2 and 3 Nuclear Facilities (and Category 1, if ever applicable), should follow commitments identified in DOE Integrated Safety Management (ISM) directives of the 450.4 series, which were an outgrowth from the Defense Nuclear Facilities Safety Board (DNFSB) 95-2 Implementation Plan. The commitments for ISM follow the seven Guiding Principles: 1. Line Management Responsibility for Safety 2. Clearly defined Roles and Responsibilities 3. Personnel Competence Commensurate with Responsibilities 4. Balance of Priorities 5. Identification of Safety Standards and Requirements 6. Hazard Controls Tailored to the Work Being Performed 7. Operations Authorization.

The process for planning and conducting pre-work hazardous analysis for all work operations at tritium facilities should be consistent with the Guiding Principles listed and should follow a structured approach commensurate with the risks and hazards involved. Several methods of enhanced work planning and related work planning strategies used throughout the DOE complex follow five steps: 1) Define Scope of Work, 2) Determine and Analyze Hazards, 3) Develop and Implement Controls, 4) Perform Work, and 5) Monitor Work Performed and Provide Feedback/Improvements. This sequence is one of many ways to ensure a structured approach in work planning with a focus on worker safety and reduced environmental risks. 3.3.3 Facility Segmentation Facility segmentation is discussed in DOE-STD-1027-92. Activities conducted in the same facility may be separated into different areas for analysis purposes, but only if independence (in accident space) can be shown. Potential areas for segmentation include: • Receiving area, • Receiving storage area, • Shipping area, • Shipping storage area, • Tritium unpackaging, handling, and packaging operations, • Low-level waste accumulation area, • Low-level waste packaging area, and • Packaged low-level waste storage area. The independence of the segments, however, is not easily demonstrated. Common piping and heating, ventilation, and air conditioning (HVAC) cannot exist to use this segmentation process. Additionally, common cause initiating events (e.g., fire, seismic) affect multiple segments, thereby placing an additional burden of proof on the analyst to demonstrate independence. A successful segmentation analysis, however, would not affect the applicability of Price-Anderson enforcement of the Nuclear Safety Rules to the facility. DOE-HDBK-1129-99 27 3.4 Radiological Materials Quantity Limits 3.4.1 Tritium Shipping, Radioactive Material Inventory, Quantity Limits Once an item has met the DOE and U.S. Department of Transportation (DOT) requirements for shipment (properly packaged, radioactively surveyed, properly marked, properly filled out shipping papers, etc.) then the item inside the approved shipping package can be shipped to a new location. During shipment, the item inside the approved shipping package is expected to be subject to the normal activities associated with its movement from one location to another; for example, loading and unloading the package from vehicles, transport to a shipping area, storage in the shipping area prior to transport, loading and unloading the package onto trucks/trains/airplanes, and storage in the receiving area after arrival at the new destination. The packaging required by DOT/NRC regulations is designed to protect the workers, the public, and the environment from the radioactive material during normal package handling, transport, and shipping/receiving storage. The applicable requirements for various quantities of tritium for transportation and storage are roughly as follows: • Limited Quantity (< 21.6 Ci for gas; see Table 7-1 for solids, liquids, and applicable 49 CFR 173 requirements): Limited Quantities of tritium can be packaged and shipped in strong, tight containers (paper boxes, paint cans) with proper markings. • Type A Quantity (21.6 to < 1080 Ci): Type A quantities of tritium use DOT Specification 7A containers, properly marked and surveyed prior to shipment. A number of different packages are available from small cans, 55 and 85 gallon drums, 4 x 4 x 7-foot steel boxes, up to and including oversized, specially designed containers. These containers are relatively inexpensive.

• Type B Quantity (> 1080 Ci): Type B quantities of tritium must be shipped in a certified DOT Type B package. There is only a limited number of these expensive Type B packages available for tritium shipment (e.g., UC-609, TRUPACT-II), and special routing (i.e., prescribed routes employing highway route control such as using major highways, bypassing cities, etc.) is required. • Type B Quantity, Low-Level Radioactive Waste: For the purposes of storage at the waste site, Type B quantity solid waste can be stored in Type A containers. At the waste generation location, items containing greater than 1080 Ci quantities of tritium are normally stored in Type A containers. These Type A containers containing over 1080 Ci must then be placed into Type B containers for shipping to DOE waste sites. At the DOE waste site, the Type A containers can be removed from the Type B package and stored in the Type A package. This allows the expensive Type B package to be freed up to be returned to the shipper and reused for another Type B shipment. A radiological materials inventory must still meet the requirements as discussed in Section 3.3.1.b. DOE-HDBK-1129-99 28 3.4.2 Tritium Receiving Area, Shipping Area, Quantity Limits There are no limits (other than local administrative limits) on the inventory of approved containers that can be stored in a properly marked shipping area, receiving area, shipping storage area, and receiving storage area. There are, however, site-wide limits that may be imposed in the EIS that could restrict the total inventory quantities. These areas must be inside an area posted as a “Controlled Area.” The controlled area encloses an area posted as a Radioactive Materials Area (RMA). The Controlled Area and the Radioactive Materials Area can be the same area. The RMA must be periodically surveyed and appropriately marked to indicate the nuclear hazard associated with the area. These areas may have more than one incoming or outgoing certified package containing Type A or B packages waiting for transport. 3.5 Tritium Unpackaging, Handling, and Packaging Areas, Quantity Limits The tritium is removed from the specification package/certified package in unpackaging, handling and packaging areas, and, as such, these areas should be designated as radiological areas or Hazard Category II or III nuclear facilities, depending upon the quantity of tritium in the inventory. 3.6 Tritium Waste Collection and Waste Packaging Area, Quantity Limits The waste collection/packaging area will have, in process, the collection and packaging of low- level waste. The waste collection/packaging area should be designated as a radiological facility or nuclear facility as defined by the quantity of tritium. 3.7 Tritium Focus Group (TFG) The TFG is comprised of both DOE federal and contractor personnel associated with tritium operations. It was formed in 1991 in response to the Secretary of Energy’s Task Force on tritium operations, and has expanded in scope and stature since that time. Its charter is contained in Appendix F. The TFG maintains a Web site (www.er.doe.gov/production/er-80/tfg), which is a comprehensive vehicle containing information on current tritium issues. 4.0 FACILITY DESIGN DOE tritium facilities, as a subset of DOE nonreactor nuclear facilities, must conform to the design requirements contained in DOE O 420.1, “Facility Safety.” Implementation guidance of this Order can be found in DOE G 420.1/B-0, and in two interim-use Guides, DOE G 420.1-X and DOE G 420.1-Y.

4.1 Tritium System Philosophy Tritium-related construction projects in the past have been designed and managed by personnel who are qualified to design standard industrial buildings and equipment but lack experience with tritium operations. During the design and construction phase, limited use had been made of tritium expertise, and, after building completion, the staff must adapt the facility for tritium operations. It is more desirable to have tritium construction projects managed by a team that includes tritium DOE-HDBK-1129-99 29 expertise on staff. The building/systems would be designed to meet the needs of the user, and costly retrofits after completion could be avoided. During the first 25 years of tritium technology, the handling techniques in use were designed to protect the worker from exposure to tritium. Worker protection was provided primarily by the use of single-pass ventilation systems designed to rapidly remove any tritium released in the breathing space from the area occupied by the workers. The ventilation gases were released through an elevated stack at high velocity to massively dilute the gases before they could reach ground level. Single-pass ventilation systems and high-velocity hoods were used extensively and quite successfully for worker protection during these early years. These high-velocity ventilation, high- velocity air hood, and elevated release techniques are still used for worker protection, but generally as a supplement to improved barriers that better protect the environment. In those early years, the room or building enclosing the tritium activity was equipped with a single- pass ventilation system that did not recirculate the air back into the facility. Outside air was brought in by the ventilation fans, conditioned for comfort, passed through the building spaces one time, and was then released to the environment through an elevated stack. The room air exchange rate generally accepted to be adequate for worker protection was 6 to 10 room air changes per hour. The tritium apparatus was enclosed in a high-velocity air hood, and the worker worked through gloves in the doors or reached in through hood openings to operate the equipment. The high- velocity air hoods were maintained at a pressure negative to the room spaces, and the natural air flow was from the room through the hood opening and then out the ventilation duct work and up the stack to the environment. Tritium releases that occurred due to normal operations, component failure, and worker error were inside the hood. The high-velocity air flowing through the hood swept any released tritium away from the worker and out the stack to the environment where the tritium was massively diluted in concentration. These techniques protected the worker; however, tritium was released to the environment. As concern for the environment increased, tritium technologists first attempted to control releases by increasing design and material selection requirements, adding and enforcing regulations, and increasing worker training and awareness. Time would prove that these techniques, although helpful, were not completely successful. The tritium workers were already operating at a high performance level and the tritium releases associated with equipment design and material selection were not measurably decreasing as a result of the more stringent design requirements and reviews. Tritium releases continued to occur. The expectation of faultless materials and errorless workers was unrealistic. By the late 1960s and early 1970s, it became obvious that a more realistic tritium culture was needed that better protected the worker, the public, and the environment. The new culture was one of capture, containment, and cleanup.

4.1.1 Tritium Capture, Contain, and Cleanup Process The capture, contain, and cleanup process encloses the primary or first wall tritium container inside a secondary container such as a double-walled container, glovebox, room, or building so that any tritium escaping from the primary container is captured in the secondary container. A tritium removal system associated with the secondary container then removes the tritium from the secondary by circulating the captured gases through a cleanup system. DOE-HDBK-1129-99 30 FIGURE 4-1. Tritium facility single-pass ventilation system 4.1.1.a Containment and Confinement Systems There are many different technical descriptions of the terms containment and confinement. The simple dictionary definitions are: containment – being contained, which in turn is to hold or enclose; and confinement – being confined, which in turn is to restrict, to keep within limits. It is beneficial to define how these words are used in the tritium field. In some facilities and applications, the terms are used interchangeably. The Tritium Focus Group definitions are defined in Appendix B, Definitions. This Handbook, as noted in the Definitions, makes a distinction between these terms as follows: Containment system – A collection of passive barriers that can satisfy a specified leak criterion without operation of any ancillary equipment. An example of a containment system is a series of piping and vessels enclosing tritium gas operations. Confinement system – A collection of barriers that can satisfy a specified leak criterion contingent upon operation of its ancillary (active) system. An example of a confinement system is a glovebox and its associated cleanup system. Note that in the context of these definitions, a glovebox with an associated glovebox cleanup system is a confinement system. A glovebox structure itself is a containment system if, and only if, the specified leak criterion can be met by the structure itself. Air Intake Hood Exhaust Pressure zone control used to move air from lower contamination areas to higher contamination areas, and then out the stackHigh-velocity air hood Typical face velocity of 150 lineal ft/min DOE-HDBK-1129-99 31 4.1.1.a(1) Primary Containers Operations are conducted with tritium enclosed inside leak tight primary containers consisting of parts such as valves, tubing, pipe, components, transducers, pumps, and vessels. The leak rate of these primary containers, at operating pressure, is generally certified to be less than 10-6 to 10-7 cm3 of helium per second (cm3 He/sec). The quantity of tritium released during all normal operations is extremely small and can be estimated from the engineering specifications. Many primary tritium systems are designed with pressure relief protection. These devices should not relieve directly into the environment, but rather into holding tanks designed with sufficient capacity to retain the entire contents of the primary system. 4.1.1.a(2) Secondary Containers In modern tritium operations, the primary container is enclosed inside a secondary barrier such as a glovebox. The secondary system is only exposed to the tritium, which is released from the primary barrier. Secondary containers in the DOE complex vary in size, shape, leak rate, and quality depending upon the age, projected use, and the quantity of tritium at risk. 4.1.1.a(2)(a) High-Quality Secondary Containers

Some operations are equipped with high quality, double-walled pressure vessels. The outer pressure vessel is the secondary containment system, and the inner container is the primary container. This type of secondary containment system is generally used for storage of large quantities of tritium of very high quality, and is certified at operating pressure to leak rates of less than 10-6 to 10-7 cm3 He/s. These high-quality secondary containment systems safely store any tritium released from the primary container for several days or weeks without a significant release to the environment. The maximum quantity of tritium released from these high-quality systems during a significant primary container leak can be accurately estimated from the secondary containment system engineering specifications. The space between the primary vessel and the secondary container in these systems is usually evacuated during service. If tritium is released into these spaces, there are no dilution gases present, and the gas leaking from the secondary container is in the same form as the gas in the primary container. If the gas released into this high-quality secondary is 90 percent tritium, less than 1 Ci of tritium will be released to the surrounding area for each four to 40 storage days. Following a release into a high-quality secondary container, the tritium can be recovered in almost the original purity by pumping it from the secondary container into another primary container. Several days can elapse during the recovery process without a significant release of tritium to the environment. 4.1.1.a(2)(b) Medium-Quality Secondary Container It is not practical or possible to enclose all primary tritium containers inside high-quality, non- diluting, evacuated secondary containers. Gloveboxes, as discussed in 4.1.1.a, can be utilized in both secondary containment and secondary confinement systems. Most primary containers can DOE-HDBK-1129-99 32 FIGURE 4-2. Secondary containment be enclosed in gloveboxes. The box gives access to the primary containers for ease of operation and maintenance. The leak rate of a glovebox can generally be certified to be no more than 10-3 to 10-4 cm3 He/s. Tritium permeation and diffusion through the elastomeric seals and gloves is reasonably well known, and the tritium released from the glovebox during a primary container leak can be accurately estimated. If tritium is released into the glovebox, the gases in the glovebox are mixed with and dilute the tritium. The gases leaking from the glovebox are then in the form of tritium mixed with the glovebox gases. FIGURE 4-3. Secondary confinement DOE-HDBK-1129-99 33 Assuming a glovebox with a volume of 1 m3 and a release of 10 grams of tritium, the glovebox concentration following the release will be 0.1 Ci/cm3. Even at the relatively high leak rates of 10-3 to 10-4 cm3 He/s, approximately 0.1 Ci of tritium will be released for every 1,000 to 10,000 seconds of elapsed time. This is a release of approximately 1 Ci from the facility stack for every 3 to 30 storage hours, which generally does not pose a significant health risk. Following a release into a glovebox, the tritium is recovered in a tritium removal system. This low- level waste is in the form of water contaminated with tritium. Several hours can elapse during the recovery process without a significant release of tritium to the environment. 4.1.1.a(2)(c) Low-Quality Secondary Containers

Other facilities are equipped with lower quality systems such as rooms or buildings. It is difficult to determine the leak rate of these low-quality systems and, therefore, it is difficult to estimate the quantity of tritium that will be released to the environment during a primary container leak. If the secondary container is a room or building, the ventilation system is shut off to prevent release of the tritium through the building stack following the release. The tritium then finds its way to the environment at ground level through the building walls, ceiling, doors, and windows. The tritium released contaminates the building and the area adjacent to the building. If a tritium cleanup system is employed, (in a room or a building), the ventilation system is switched over to the recirculation mode to prevent release of tritium through the building stack. Even in this mode, tritium is released to the environment as described above, but in a lesser quantity. For example, assuming a 100 m3 (20 x 18 x 10 feet) building and a 1-gram release to the building, the tritium concentration in the building gases will be 0.0001 Ci/cm3. If the building is the containment barrier, it will leak approximately 5 percent of its volume to the environment per hour; therefore, an environmental tritium release of 500 Ci/hr will result. A significant fraction of the tritium released from the primary containment system will be released to the environment before recovery is accomplished due to the high leak rate of these low quality secondary systems. However, these low-quality systems still have a place in the tritium containment strategy. These room- or building-type systems, equipped with high flow rate tritium removal systems, are used in facilities where there are no other feasible alternatives due to the large size of the equipment being enclosed. FIGURE 4-4. Building confinement system DOE-HDBK-1129-99 34 4.1.2 Tritium Cleanup and Removal Systems Current designs employ tritium removal systems. When tritium is released into a secondary, the associated cleanup system starts, and the gases containing tritium are circulated through the cleanup system, and the tritium is removed. See Figure 4-5 for a diagram of a typical gas-to-water conversion tritium removal system. FIGURE 4-5. Typical gas-to-water tritium removal system flow schematic If the cleanup system is associated with a high quality barrier (leak rate of less than 1 Ci of tritium in a period of 4 to 40 days) then tritium transfer to another container can take several days to complete without a significant release of tritium to the environment. If the cleanup system is associated with a medium quality barrier (leak rate of less than 1 Ci in 3 to 30 hours), then the cleanup system flow rate needs to be high enough to remove tritium within a few hours in order to prevent a significant release of tritium to the environment. If the cleanup system is associated with a low quality barrier, (leak rate of 8 Ci or more per minute), then the cleanup system flow rate needs to be very high so that tritium is removed from the gas before it is released to the environment. The captured tritium is generally removed by circulating the gas through a system that removes tritium down to the part-per-billion level. Typical present-day cleanup systems remove tritium by cracking the molecules on a hot catalyst. The free hydrogen atoms combine in the catalytic reactor with the oxygen present to form water. The tritiated water is then removed from the gas by molecular sieve traps. Depending upon the tritium species, the concentration reduction of these systems can be from one million to one hundred million in a single pass.

The tritium released to the environment during this process is a function of the quantity of tritium released, the volume, and leak rate of the container, and the cleanup rate of the tritium removal system. DOE-HDBK-1129-99 35 There are several considerations in determining the adequacy of the tritium removal system. First is the volume: the larger the volume, the longer it will take to remove tritium from the gases. Second is the tritium removal system flow rate. This rate affects the time required to remove tritium from the gases. The lower the flow rate, the longer it will take to remove tritium from the volume. Third is the cleanup rate of the tritium removal system. This system is typically operated in a recirculating mode, but may also be operated in a single-pass mode. The tritium-contaminated gases are pumped through the tritium removal system, and the cleaned gases are either returned to the volume or released to the environment. In a large complex system filled with equipment, it is difficult to know exactly how the returned or make-up gases will mix with gas in the system. The gas exit port should be spaced several feet away from the return port. It could be assumed that gas flows through in a slug, like a piston, and that only a single pass would be required to remove all of the tritium from the gases. Slug or piston displacement flow, however, is unlikely, and a more common assumption is to assume that the incoming tritium free gases exponentially dilute the gases. Employing the standard assumption of exponential dilution (which has been experimentally verified in the DOE complex [23]), the quantity of tritium in the system Qt is expressed by Qt = Qi e -t(F/V) where Qi = initial quantity of tritium released t = time after starting the tritium removal system V = volume of the system F = flow rate of the tritium removal system A simple way of using the equation is to calculate the time it will take to reduce the system tritium concentration by a factor of 10. That is: Qt/Qi = 1/10 = e -t(F/V) This implies: -t(F/V) = ln 1/10 = -2.3 or t = 2.3 x (F/V) Therefore, assuming exponential dilution, the tritium concentration of the gas will be reduced by a factor of 10 for every 2.3 time constants determined by F/V. Similar calculations show that the concentration will be reduced by a factor of 100 and 1000 for every 4.6 and 6.9 time constants, respectively. Figure 4-6 is the plot of the relationship between the percent of tritium remaining in the volume with time. DOE-HDBK-1129-99 36 0 10 20 30 40 50 60 70 80 90 100 0.00 0.30 0.60 0.90 1.20 1.50 1.80 2.10 2.40 2.70 3.00 3.30 3.60 3.90 4.20 4.50 4.80 5.10 5.40 5.70 6.00 t/(V/F) = Number of System Time Constants (F/V) Q (i )/Q (t ) x 10 0 = P er ce nt o f Q (i ) re m in in g in t h e co n fi n em en t vo lu m e Note that the quantity of tritium remaining in the confinement volume decreases by a factor of 10 for each 2.3 time constants, i.e. when t=2.3 x F/V Q(t)=Q(i) e-t/(F/V) FIGURE 4-6. Confinement volume cleanup rate as a function of system time constant, F/V, assuming an exponential dilution rate The fourth consideration is the system leak rate. While the tritium removal system is in operation, some of the gases will leak out of the system to the environment. The barrier must be leak tight enough to prevent a significant release of tritium to the environment while the tritium is being removed. A high flow rate tritium removal system can be used to help offset a leaky system, but will significantly increase the cost of the removal system.

The quantity of tritium released to the environment due to the leak rate can be calculated. Assuming exponential dilution, the concentration of tritium in the gas C T2(t) is CT2(t) = CT2(i)e -t/(F/V) where CT2(i) = initial concentration of tritium in the system t = time V = volume of the system F = flow rate of the tritium removal system Note, CT2(i) = QT2/V where QT2 is the quantity of tritium released. DOE-HDBK-1129-99 37 Therefore, the rate of tritium released from the system, d C d t T 2 , at any time, t is d C d t T 2 = LCT2(t) = LCT2(i)e -t(F/V) where L = leak rate of the system The tritium release from the system due to leakage T2(rel) over any time period from 0 to t, is T2(rel) = ∫LCT2(i)e -t(F/V) dt = (V/F) LCT2(i)e -t(F/V) |t0 = (V/F) LCT2(i)e -t(F/V) The calculated leak rate that will result in a tritium release, T2(rel), of 1 Ci due to an initial tritium release, CT2(i), of 1 g (10,000 Ci) into a glovebox with a volume, V, of 1m3 and a flow rate, F, of 1m3/min is L = T 2(rel)/(V/F) CT2(i)e -t(F/V) = 1/ (1 x 10000 x 1) = .0001 m3/min = 1.667 cm3/s Therefore, a glovebox leak rate of 1.667 cm3/s will result in the release of 1 Ci of tritium from a 1 m3 volume glovebox during the time it takes a tritium removal system operating at a flow rate of 1 m3/min to clean up the glovebox following a tritium release. 4.1.3 Future Directions in Tritium Removal and Cleanup Future tritium facilities should analyze the applicability of confinement systems in their facilities. For example, DOE O 420.1, Change 2, states, “For a specific nuclear facility, the number and arrangement of confinement barriers and their required characteristics shall be determined on a case-by-case basis. Factors that shall be considered in confinement system design shall include type, quantity, form, and conditions for dispersing the material. Engineering evaluations, trade- offs, and experience shall be used to develop practical designs that achieve confinement system objectives. The adequacy of confinement systems to effectively perform the required functions shall be documented and accepted through the Safety Analysis Report.” As regulatory release criteria and ALARA concerns are strengthened, the desirability of barriers increases. Most current confinement systems are of the glovebox and of the room or building cleanup types. One disadvantage of these current designed systems is that tritium is converted to oxide, which eventually must be handled and disposed (with the attendant risks). Room or building type confinement systems are used in some tritium facilities such as in the T-building at Mound, in TFTR at Princeton Plasma Physics Laboratory, and in TSTA and WETF at Los Alamos. If the system is not very close to 100% efficient, the released oxide (which is ~ 25,000 times more toxic than the gas) could give an overall detriment (e.g., if 25,000 Ci of gas are collected, but only one Ci of oxide is released by the system, the whole operation is just a draw). Some current designs, notably the cleanup systems at RTF, make use of gettering without oxidation. Additionally, research efforts for other systems that replace the oxidation step are active in the DOE complex. The primary advantage of these getter systems is that tritium is removed and stored in elemental DOE-HDBK-1129-99 38

form and is not converted by the tritium removal system to the more radiotoxic tritiated water. The application for glovebox atmospheres is good; however, severe challenges for removing tritium from room atmospheres with the required flow rates, while not fouling the getters, appear less promising. Research into other than metal (e.g., fullerenes) gettering material has been sponsored by EH; however, results have not been promising to date. It is possible that advances in design will make room or building-type confinement systems highly desirable, or it may turn out that these cleanup systems in general are found not to be cost-effective, and that a better use of resources to decrease environmental releases could be made by upgrading the existing primary and secondary systems. Additionally, there may be future glovebox decisions in which the oxidation process is still chosen over the gettering process due to programmatic reasons. 4.1.4 Inspection and Surveillance Requirements Instrumentation should be provided to monitor the leak-tight integrity of process piping, tanks, and other equipment. The surveillance measurements may include those of pressure differentials or flow rates and should relate to the design leak rate. Radiation monitoring instrumentation may be used to qualitatively assess changes in leak rate. 4.2 Building Ventilation System If tritium does penetrate its barriers, it can be released into the worker breathing space. In this scenario, the ventilation system should be designed to meet the following objectives. • Move released tritium from the worker breathing space as soon as possible. • Minimize the contamination of other areas while moving released tritium. • Release the tritium-contaminated gases at an elevation and velocity that will result in massive dilution and mixing with outside air before the tritium reaches ground level. The ventilation system should be designed using the following guidelines: • The ventilation system should be a single-pass ventilation system. Outside air is brought in through a supply fan, conditioned for the comfort of the workers, passes through the ductwork to the ventilated spaces one time, goes through the exhaust ductwork to an exhaust fan, and is released to the environment through the facility stack. • The air supply and exhaust systems should be designed to eliminate dead air spaces where tritium-contaminated gases may accumulate. • The ventilation system ductwork should not be shared with non-tritium operations. • To minimize cross-contamination from one room to another, the exhaust gas from each room should dump into a central exhaust duct. Exhaust gases from several rooms should not be combined before being dumped into the central exhaust duct. • To minimize cross-contamination from one ventilation function to another, the gases for each type of function, such as room ventilation, high velocity air hood ventilation, and glovebox ventilation, from a single room should not be combined until they reach the central exhaust duct. DOE-HDBK-1129-99 39 • To ensure good mixing and dilution by outside air, the exhaust gases should be released to the environment through an elevated stack at high velocity.

• The ventilation system should be designed to use pressure zone control to minimize cross- contamination. The ventilation control system is designed to hold the spaces occupied by the tritium operations at a negative pressure relative to the spaces surrounding the facility. If the whole building is a tritium facility, then the building is at a slightly negative pressure relative to the environment. If the tritium is in a single room in a building, the room is at a negative pressure relative to the adjacent rooms. Air continuously leaks into the tritium operating areas from the surrounding environment. The ventilation zones of tritium processing areas should be maintained under controlled temperature and humidity conditions at all times to reduce the in- leakage of moisture to inert atmosphere gloveboxes. (Further reductions in inleakage of moisture into inert gloveboxes can be achieved by selecting a glove material that has a low permeation rate for moisture and by reducing the exposure time (e.g., glove port covers) and/or total glove surface area). • The walls separating adjacent rooms must be reasonably sealed to minimize tritium released from contaminating an adjacent room. Administrative controls need to be in place to require caulking and sealing around wall penetrations such as conduit and piping. • For the ventilation system to work as designed, the inside and outside doors and airlocks need to be used properly. Propping an outside or inside door open will upset the pressure zone control system. An outside door bypasses the stack and provides a path for a ground level tritium release. The doors should be equipped with automatic door closures, and all personnel should be instructed on the proper use of doors. • For facilities handling gram quantities of tritium, a rule of thumb is 6 to 10 air changes per hour as the standard of performance. Depending upon the operating conditions associated with the ventilated area, ventilation rates of 3 to 20 air changes per hour have been used. The ventilation rate should be based on analysis of the hazards of the operations. A facility that has the potential to release only a few curies of gaseous tritium into the breathing space does not need to operate at the same ventilation rate as a facility that has the potential to release tritium into the breathing space. • Single-pass ventilation systems are expensive to operate because all air must be conditioned to make its single pass through the facility, and this conditioned air is not reused. Additionally, a high flow rate is desired in order to remove any released tritium from the facility as soon as possible to protect the workers. • Special precautions such as Personnel Protective Equipment, including respiratory protection and passing exhaust through particulate filters, are needed when working with SMTs. D&D work with SMTs should occur within a confined airspace, if possible. It is feasible to design future ventilation systems to operate at a variable flow rate that is a function of the time of day and the measured tritium concentration in the rooms. This would entail a higher initial cost, but would decrease the long-term operating costs without significant impact on the facility safety. DOE-HDBK-1129-99 40 4.3 Chilled Water System

The chilled water system that is used to cool the tritium-related activities in a facility must be carefully designed to minimize the volume and tritium concentration of the contaminated water generated. The use of single-wall, water-to-gas heat exchangers in tritium removal systems and vacuum furnaces for example, will result in tritium contamination of the chilled water system. In some facilities, the same chilled water system is used to cool non-tritium activities in the same building, and, at some sites, the same chilled water system is also used in non-tritium related activities in adjacent buildings. As a result, tritium-contaminated wastewater is generated, and tritium contamination is spread from one piece of equipment to another, from one room to another, and from one building to another through the chilled water supply. The chilled water system should be designed to minimize the volume of water that can be contaminated with tritium. One technique is to use water-to-water heat exchangers or double- walled, gas-flushed water-to-gas heat exchangers to isolate the high volume central system from the tritium-related activities. The volume of water in the secondary loop is much smaller than that in the primary loop. The primary reason for using chilled water for cooling equipment is cost. The systems are reliable, low in cost, small in size, readily available in many different sizes from many different manufacturers, and easy to maintain. Air-to-air heat exchangers can, in some cases, be substituted for chilled water cooling, but are larger in size and will not work in some applications. Refrigerated cooling systems are more expensive, but operationally eliminate the need for disposing of tritium-contaminated water generated by the chilled water systems. 4.4 Seismic and Wind Design and Evaluation of Structures and Facilities Savannah River Site (SRS) has been the lead site for interfaces with the DNFSB and DOE EH concerning NPH design issues associated with tritium. A summary of DOE Natural Phenomena Hazards Policy is included in Section 5.7.1. This policy is also applicable to structures and the complete facility. Section 5.7.2 and its references should be reviewed as part of the overall seismic and wind design for evaluation of a tritium facility. For wind loading provisions, American Society of Civil Engineers (ASCE) 7-95, “Minimum Design Loads for Buildings and Other Structures,” has recently been completed and issued and is changed greatly from past versions. A “3-second peak wind gust” is used rather than “fastest mile wind speed” as a design measure. Change Notice of DOE-STD-1020, “Natural Phenomena Hazards Design and Evaluation Criteria for Department of Energy Facilities,” is in error in referencing ASCE 7-95, but keeping “fastest mile wind speed” as a measure. Wind speeds must be converted to “3-second peak wind gust” for ASCE 7-95 to be correctly used. An understanding of the types of loading produced by earthquakes and windstorms is useful in planning mitigating approaches. An earthquake produces shaking, which will affect the entire facility and its contents. Attention to anchorage and connection details to all structures, systems, and components is essential. Earthquakes may also cause ground rupture if the facility is near a fault. Loss of ground stability may also occur due to settlement or liquefaction and will depend on soil types and location of the ground water table.

Windstorms generally affect the structural shell and systems and components outside of the structure. Windstorms will produce direct pressure and suction on walls and roofs with increased DOE-HDBK-1129-99 41 loading at corners, eaves, and ridges. Extreme windstorms can produce missiles from debris near the facility or from nearby facilities. These missiles can strike walls and roofs and their effects should be evaluated. During tornadoes, in addition to pressure effects and windborne missiles, there will be an atmospheric pressure change or a pressure drop below ambient pressure as the tornado passes over a facility. This can affect wall and roof openings and ventilation system filters. An approach used at SRS is to store resources in a Highly Invulnerable Encased Safe (HIVES). HIVES are metal cabinets designed to store tritium reservoirs and Hydride Storage Vessels (HSV). The HIVES are designed to be capable of protecting the pressure boundary integrity of stored reservoirs or HSV against impact of structural elements free falling from the collapse of Building 234-H and the adjacent 296-H stack caused by credible NPH Design Bases Accidents. The HIVES are constructed of hardened T-1 steel, primarily from 1-inch plate. The overall dimensions are approximately 22”Wx39”Dx68”H, which includes both a cabinet and a matching bonnet assembly bolted to the cabinet top plate. This bonnet contains an aluminum hexcall honeycomb material designed to absorb the impact loading stated above. The following should be considered when designing or evaluating structures or facilities for earthquakes and windstorms: • The Performance Category (PC) of the structure or facility should be determined. It is not necessary for the entire facility to have the same PC; that is, different parts of a facility could be in different performance categories. For example the office portion of a Tritium Handling Facility may be PC1 while the laboratory portion may be PC3. • Judgment should be exercised to ensure that various parts of the facility have been categorized in a rational manner. For example, a PC1 facility does not physically support a PC3 facility. • In addition to the vertical load carrying system, a lateral force resisting system should carry the seismic and wind loads. This may be a frame or shear wall system or separate bracing system. Attention to connection details is very important in ensuring adequate structural integrity to resist the limited energy input motion produced by earthquakes and the longer duration wind loading. • An understanding of the type of loading that earthquakes and extreme winds produce on the overall structure is essential during the design and evaluation process. • An adequate load path must be ensured throughout the structure — Roof and wall connection details to structural system — Overall roof-to-wall connections — Wall-to-foundation connections — Foundation adequacy • Innovative technologies, such as base isolation and passive energy dissipation devices, should be considered during design of new facilities and for upgrading of existing facilities to reduce seismic loads to systems and equipment within the structure. These approaches have been used in many projects throughout the world during the last 10 years. • Avoid the type of details and structural features that have performed poorly during past earthquakes. These are as follows: DOE-HDBK-1129-99 42

Structural Systems — Unreinforced masonry construction — Non-ductile concrete construction — “Soft” story construction — Incomplete lateral force resisting systems — Welded steel connections — Irregularities, eccentricities, and discontinuities in both weight and geometry — Pounding or impact with nearby facilities Details — Inadequate hoop reinforcement in concrete columns — Reinforcing steel connections — Connection details — Accommodation of differential motion • Ensure foundation adequacy by avoiding the following: — Ground failures — Landslides — Liquefaction — Excessive or differential settlement • Ensure that lifelines are not damaged. — Provide flexibility in lifeline connections to the facility. — Identify important lifeline functions such as power, water, sewer, gas, and communications that will be needed immediately after an earthquake or windstorm. — Examine the vulnerability of lifelines onsite as well as offsite. — Plan for alternative lifeline support. • Examine and prepare for the possibility of a fire following the earthquake. Employing mitigation efforts before earthquakes or windstorms occur is a very cost-effective means to provide life safety, to minimize damage and losses, and to reduce the impact on the facility and operations. It is extremely important to pay attention to all details because natural phenomena will find the weak links and cause damage. 4.5 Other Design Considerations The following items should be considered in the design of a facility with tritium operations. • Designated safety-class systems should employ the concepts of redundancy, separation, and diversity. Some designs in the past, notably at Building 233-H at SRS (i.e., RTF), employed redundant signals to a single actuation device. An improvement to this design concept would be to provide these redundant signals to two independent, separate (by distance) actuation devices. • A design requirement should be included to prevent the formation of explosive mixtures during the handling, processing, and storage of tritium gas and other hydrogen isotopes. DOE-HDBK-1129-99 43 • Calibrated tanks and associated piping that are used for pressure-volume-temperature measurements should have surface treatment of their interiors (e.g., electropolished and passivated) to allow accurate volume measurements. • Tritium process and handling systems should use, wherever possible, nonflammable hydraulic, lubricating, and cooling fluids. • The designer should consider not using hydrogenous fluids where they might become contaminated with tritium. • The designer should consider providing for the retention of firewater with subsequent monitoring prior to disposal for all buildings where there is tritium. • Barriers should be provided to prevent damage to equipment and injury to personnel while performing testing operations that could produce missiles or blast pressures. These barriers should be designed using conservative and proven design principles, such as those of DOE/TIC 11268, “A Manual for the Prediction of Blast and Fragment Loading of Structures.” • An independent air system should be provided for breathing air. It should have dedicated, oil- free compressors or pressurized cylinders of breathing air, and provide breathing air in all areas of the tritium facility where it may be needed for maintenance operations and/or personnel safety. Contamination of the air supply (e.g., from refrigerant leaks or air intakes) should be detectable at levels low enough so as not to pose health concerns.

• The design considerations of the Fire Protection System should include the following: — determining safety classification of SSCs of both fire detection and fire suppression systems — consideration of unique fire sources (e.g., uranium beds used for tritium storage) — compatibility of fire extinguishing agents with the fire sources in a tritium facility — containing and handling requirements for expended fire fighting agents (including water inventory) that may become contaminated with tritium 4.6 Lessons Learned Buildings 232-H, 233-H, and 234-H at SRS, the Weapons Engineering Tritium Facility (WETF) located at Los Alamos National Laboratory (LANL) and the Tritium Research Laboratory (TRL) at Sandia National Laboratory, Livermore (SNLL) are examples of facilities that were initially designed to perform tritium handling. Most other facilities in use were originally designed to do other work and have been retrofitted to perform tritium handling. One of the major problems with the facility retrofit process is that the existing utilities such as ventilation, floor drains, gas supplies, and chilled water systems are shared with the adjacent non- tritium areas. The same duct work, which is used to sweep released tritium from tritium operating areas, is used to provide ventilation for offices and non-tritium areas, and, as a result, tritium back diffuses into the non-tritium areas through the shared duct work. The same chilled water system used to cool the tritium-related equipment is used to cool the non-tritiated office spaces, and leaks of tritium-contaminated chilled water result in contamination of clean areas. The floor drains from the non-contaminated areas drain into the same system as the floor drains from the tritiated areas, and, through the drains, gases flow from one area to another. DOE-HDBK-1129-99 44 Tritium facility utilities such as chilled water, compressed air, gas supplies, ventilation, floor drains, sink drains, storm drains, and stacks should not be shared with other non-tritium areas. Sharing these systems spreads tritium contamination to other areas, complicates day-to-day management of the facility, and impacts the transition process if the facility is transitioned to other use. The use of hazardous materials should be reduced or eliminated in the initial design stages of the facility, as it will likely lead to the generation of mixed wastes and increased D&D costs. 4.6.1 SNLL Tritium Research Laboratory Tritium operations have been terminated at the TRL, and the facility has been transitioned to other use. The problems encountered during transition of the TRL should serve as an example for facility designers of the future. The TRL was designed beginning in 1972 specifically to handle tritium, and a few changes to the initial design would have resulted in significant cost and timesaving during the transition process. • The tritium removal system in the central glovebox had a supply and return manifold fabricated of six-inch diameter stainless steel pipe. The associated vacuum pump effluent manifold consisted of an all welded two-inch diameter pipe. These manifolds were approximately 200 feet long and extended down the central corridor of the building. The manifolds extended into each room from the corridor and were of all welded construction. During the transition process, special tooling had to be purchased to cut the six-inch and two-inch diameter pipe into sections small enough to be disposed of as solid waste.

Designing the system to include flanges and isolation valves at specific locations would have resulted in some cost savings during facility dismantlement and transition. However, since too many valves and flanges can increase the potential for leaks during operation, the installation points should be placed only at strategic locations. • The floor covering installed in the TRL consisted of 12-inch tiles glued to the concrete floor. Both the floor tile and the adhesive contained asbestos that had to be removed during transition of the facility. Additionally, tritium-contaminated liquids were spilled on the floor during operation and leaked through the tile into the concrete below. The use of adhesives made of non-hazardous materials would have resulted in significant cost saving. Princeton University Plasma Physics Laboratory has suggested that sealing the concrete with a thick, hard finish epoxy paint prior to installation of any floor covering would mitigate the impact of tritiated liquid spills, although even epoxy will experience tritium permeation. • The TRL was equipped with a recirculating chilled water system. The water-to-gas heat exchangers used the chilled water to cool the glovebox and vacuum effluent tritium removal system gases, vacuum furnace heat exchangers, glovebox temperature control systems, and in a variety of other tritium-related tasks. After a few months of operation, tritium contamination was found in the chilled water. In order to control the buildup of tritium in this system, the contaminated water was periodically drained from the system and replaced with uncontaminated water. If the initial design had used water-to-water heat exchangers as barriers, the volume of contaminated water would have been minimized. If double-walled, gas-flushed, water-to-gas heat exchangers had been used, the contamination would have been significantly lower. DOE-HDBK-1129-99 45 • The TRL used approximately 100 oil-type vacuum pumps. Oil-free vacuum pumps were not as common when the TRL was designed. The vacuum pump oil becomes contaminated with tritium during use. At some DOE locations, tritium-contaminated oil is regulated as a mixed waste. Handling the tritium-contaminated oil is a significant safety hazard to operating personnel and should be eliminated where practical. If oil-free vacuum pumps had been used, the generation of mixed waste in the form of tritium- contaminated vacuum pump oil would have been eliminated as would the hazard associated with changing vacuum pump oil. • The TRL was equipped with a ten-air-change-per-hour, single-pass, pressure-zone-controlled ventilation system. Some of the ductwork became contaminated and was removed and disposed of as solid low-level (radioactive) waste during the transition process. Experience from Sandia has suggested that if the ventilation ductwork for room air ventilation had been separated from the ductwork used for high velocity air hood and glovebox ventilation, the contamination would have been easier to control during the transition process. • The presence of hazardous materials in the form of asbestos and oil complicated the transition process. The materials used in fabrication and during facility operation should be reviewed, and, if possible, all hazardous materials should be eliminated.

• The initial design of the TRL included collection of wastewater from floor drains and sinks in two underground holding tanks so that the water could be checked for contamination before it was sent for disposal. The holding tanks were buried and could not be inspected. The wastewater drain system consisted of several hundred feet of buried drainpipe, which drained into the holding tanks. After a few years of operation, the underground tanks were replaced with holding tanks enclosed in a below-ground-level open concrete pit. If the tanks leaked, the concrete pit would contain the leak. This design also provided for inspection of the tanks for leaks. The buried, underground drain lines remained in place throughout the life of the facility. Wastewater holding tanks and collection systems should be designed so that potential leaks can be contained and the holding tanks and drain system can be inspected periodically. 4.6.2 SRS Old Tritium Extraction Facility Between 1995 and 1997, a team of specialists decontaminated and decommissioned (D&D) the Old Tritium Extraction Facility (232-F) at the SRS. [24] The work was conducted in six phases. Each phase presented challenges regarding the behavior of tritium, particularly tritium in equipment and structures of metal and concrete. A description of each phase is as follows: • Phase I, pre-Characterization/Isolation, was designed to identify the specific buildings, ancillary equipment, structures and premises to be cleaned, and isolate them in a way that avoided inadvertent contamination elsewhere in the complex. • Phase II, Detailed Facility Characterization, included determination of the type and amount of actual contaminants and determined the location and type of disposal operations required. This phase was the most complex in terms of characterizing tritium contamination. DOE-HDBK-1129-99 46 • Phase III, Decontamination and Dismantlement, involved removing hazardous materials such as PCBs, mercury, and lead, as well as radioactive constituents such as tritium and fission products. It also involved dismantling equipment and some structures. • Phase IV, Demolition with Explosives, involved the removal and destruction of the facility and stack. • Phase V, Waste Management, was a continuing process throughout all six phases. It involved the determination of how much of each contaminant was present at the site and where the various site contaminants would be disposed. Phase VI, Green Grass Restoration, restored the site to a usable, visually aesthetic entity. Detailed characterization was one of the most challenging phases in dealing with tritium, and required an expert understanding of the behavior of tritium in porous and non-porous materials. For example, because tritium migrates to the subsurface of porous materials, volumetric characterization required core boring. This was not well understood initially and early characterization activities, which involved smear and scabbling techniques or wiping and scraping the surface of concrete, gave false (low) tritium contamination levels. In some instances, areas that were smeared clean, began to reveal elevated readings following rain inleakage. In addition, destructive testing for nonporous material, such as metals, was required to fully characterize volumetric tritium contamination. DOE-STD-1120-98, “Integration of Environment, Safety, and Health into Facility Disposition Activities,” provides guidance for integrating and enhancing worker, public and environmental protection during facility disposition activities.

5.0 DESIGN OF EQUIPMENT The design requirements for tritium are a function of the tritium form, quantity, concentration, pressure, and period of storage. High concentrations of tritium gas stored at high pressure (> 2,000 psia) are difficult to contain due to tritium and helium embrittlement of the container materials. Design of these systems requires careful selection of the materials of construction and must be designed using expertise in high-pressure tritium containment. Low concentrations of tritium in gaseous form mixed with other gases at low (< 600 psia) to medium (600 to 2,000 psia) pressure, regardless of the quantity, do not significantly impact the strength of the materials they are stored in. As a result, standard designs can be used. Tritiated water in the form of T2O is somewhat corrosive unless properly stored with an overpressure of T2 gas. This is due to the suppression of the formation of oxygen in the cover gas and peroxides in solution. [25] Tritium systems should be designed by persons with tritium experience. Low concentrations of HTO (mCi/mL to Ci/mL) recovered from tritium removal systems have proven to be corrosive when stored in liquid form in metallic containers and have resulted in the development of significant leaks in containers within days or weeks. Storage of this same water DOE-HDBK-1129-99 47 solidified on clay or on molecular sieve material, regardless of the quantity, is stable and non- corrosive and may be stored for many years in the container. 5.1 Material Compatibility Proper materials selection and rigorous design have led to tritium-handling systems that are extremely safe for long periods of time. Materials exposed to tritium under certain conditions can be susceptible to hydrogen embrittlement. The chances of embrittlement are significantly reduced by proper material selection. No additional thickness of components, such as the “corrosion allowance” used to mitigate uniform corrosion, is added to components to reduce or eliminate the chance of hydrogen-induced cracking and subsequent failure as material degradation effects are not uniform. In the past, tritium systems were located in hood systems having a single-pass air flow to protect workers should a component fail and tritium be released; the released tritium would be removed from the facility into the atmosphere by the hood exhaust system. Current system design normally consists of a closed glovebox, with a stripper or getter system to trap the released tritium. This arrangement protects workers and prevents releases to the environment. Tritium can permeate vessel barriers, especially in components operating at elevated temperature. Currently available tritium permeation data is normally sufficient to estimate the order of magnitude of tritium permeation through barriers. These estimates can be used during system design or to determine whether additional purging and stripping systems are required to “clean up” permeated tritium or whether other design changes (such as wall thickness, material, or coating) are required to reduce permeation. Tritium permeation can lead not only to contamination outside the barrier, but it can also result in significant quantities of tritium being dissolved in parts, which can lead to hydrogen embrittlement. Over time, this tritium decays to 3He, which has been found to accentuate hydrogen embrittlement and to cause weld cracking (see below). 5.1.1 System Design

Tritium primary containers must have an exceptionally low leak rate (10-6 to 10-7 cm3 4He/second) and probability of failure or leaking. The consequences of tritium leaks can include personnel uptake, release to the environment, ignition (if mixed with oxygen), and violation of operating permits. Pressure and vacuum vessels used in tritium systems are generally designed and constructed using codes and standards applicable to boiler and pressure vessels. In the United States, use of the American Society of Mechanical Engineers (ASME) Boiler and Pressure Vessel Code is recommended but not required. Pressure and vacuum vessels constructed using this code have an accepted rigor in design, construction, and inspection that facilitates approval and acceptance by regulators. The ASME Boiler and Pressure Vessel Code primarily covers the design of vessels, and does not cover all of the aspects of a tritium system design. Other design standards, such as resistance to seismic events, also must be followed, depending on the location and regulators of the facility. See Sections 4.4 and 5.7 for more discussion of seismic design. 5.1.1.a Leak Testing After fabrication, thoroughly leak testing tritium systems is extremely important. Normally, leak testing employs commercial helium-mass-spectrometer-based systems and is performed after any other required proof, pressure, or vacuum performance tests. Other leak detection methods, such as rate-of-rise, can also be employed in addition to helium mass spectrometry. A dilute solution of tritium in an inert gas can also be used to detect small leaks. Tritium is a highly effective leak DOE-HDBK-1129-99 48 detection species, since it travels rapidly through cracks and can be easily detected at very low levels. 5.1.1.b Joining Components of tritium systems are commonly joined by welding. Welds are normally designed so they can be non-destructively inspected by a suitable method such as radiography or ultrasonic testing. Also, the weld design should minimize so-called “virtual leaks” on the interior of tritium containing volumes. Examples of weld practices to minimize virtual leaks include 1) using full penetration welds where possible, and 2) welding feed-throughs to the interior wall surface (not on the outside, which would leave a gap on the inside around the feed-through that is difficult to outgas). Standard weld rod filler materials are chosen, depending on the base alloy. Every effort should be made to reduce or eliminate welding residual stresses. There is less experience using other joining methods such as brazing or high temperature soldering in tritium systems. Dissimilar materials may have to be joined by a transition junction, using an intermediate material to enable proper welding and accommodate differences in thermal expansion and other properties. Tritium gas permeates austenitic stainless steels, and, over time, 3He is created by beta decay of tritium in solution in the material. Welding stainless steel containing solute helium is difficult because intergranular cracking can occur. During welding, the solute helium agglomerates at grain boundaries and forms both intergranular cracks in the heat-affected zone and pores in the fusion zone. Low-heat-input weld techniques have been shown to mitigate this problem to some degree; however, weld repair of helium-containing stainless steel is normally difficult to perform without some cracking.

All-metal mechanical joints are also a sound way to join components in tritium systems. Typically, copper, silver-plated nickel, or silver-plated stainless steel have been used as gaskets. Commercial high- and ultrahigh- vacuum fittings are normally compatible with tritium. 5.1.1.c Surface Coatings and Treatments Aluminum and aluminide coatings have been successfully employed on stainless steel to reduce permeation into and through the steel. These coatings can be applied on large items using a proprietary fluidized bed furnace, having a controlled atmosphere (in the so-called “calorization” process). Gold has also been used as a permeation barrier in some applications and is often applied over a thin nickel buffer layer (“strike”) that has been applied to the bulk metal (e.g., stainless steel) after proper surface preparation. Several companies treat stainless steel surfaces using various proprietary electrochemical processes to “passivate” the surface. These processes probably enrich the chrome content of the surface oxide, polish the surface (thereby reducing the effective surface area), and remove carbon and hydrogen from near the surface. All of these microstructural changes may be desirable for tritium systems in which the process gas must remain at high purity. Capillary lines that route gas to mass spectrometers are commonly passivated, to reduce changes of gas composition by isotope exchange, while the gas flows from the sample location to the mass spectrometer. Passivated surfaces reduce the rate of isotope exchange in hydrogen isotope mixtures, probably by reducing the catalytic effect of the surface decomposing hydrogen isotope molecules to atoms, which enables isotope exchange. Vacuum systems having surfaces treated in this way evacuate DOE-HDBK-1129-99 49 faster. This type of surface passivation can be expensive, so many parts of tritium systems are not passivated; normally only parts requiring the special properties of passivated surfaces are treated. 5.1.2 Structural Metals Exposure of metals to high pressure (> 2,000 psia) hydrogen, deuterium, or tritium will result in hydrogen embrittlement of the material. This could eventually result in material failure. The time until failure is a function of the container material, the pressure, and temperature. Additionally, materials exposed to high-pressure tritium are also subject to helium embrittlement. Tritium at high pressure enters the metal and decays to 3He. The buildup of helium in the metal results in helium embrittlement, which, depending upon the pressure, temperature, and type of material, will eventually result in failure of the material. Exposure of metals to low to medium pressure tritium at normal temperatures does not generally result in material failure within any reasonable period of time. Some metals are more resistant to embrittlement than others and, therefore, are more compatible with tritium. Depending upon the specific application, 304L and 316L stainless steel are generally considered to be the most hydrogen compatible and readily available stainless steels for tritium service. High-pressure vessels, valves, and tubing designed of these materials when used at their rated pressure and temperature will provide many years of service without material failure. When equipment is designed for tritium operations, a materials expert should be consulted to ensure that the materials selected are compatible with their intended service.

5.1.2.a Austenitic Stainless Steels The recommended materials of construction for tritium-handling systems are from the class of wrought 3XX series of austenitic (face-centered-cubic) stainless steels, including Types 304L, 316L, and 347. Types 304L and 316L are most often used in tritium processing systems. These steels provide good strength, weldability and resistance to hydrogen embrittlement. Components fabricated from these materials are procured routinely. Many commercially available vacuum system components that are used in tritium systems, such as valves, piping, pumps, and analytical instrument sensors, are fabricated from these types of austenitic stainless steel. Wrought materials are preferred to cast because wrought materials normally have a more homogenous microstructure. In the past, tritium has leaked through parts having poorly oriented stringers and inclusions. The forging direction of some wrought components has been specified so that the orientation of inclusions is not in a direction that could result in a tritium leak path. Low carbon grades (such as 304L and 316L) are preferred to avoid weld sensitization and to reduce the number of inclusions (impurity particles such as oxides and carbides). Modern vacuum-arc- remelted steels are a good choice because they have lower impurity levels, thereby resulting in fewer inclusions that could aid hydrogen-induced cracking or provide leak paths. Typically, tritium system components employ seamless pipe and tube where practical. Stabilized grades, such as Type 347, have been employed in applications where post-weld heat treatment is not possible. This usually occurs when a process vessel contains a working material (such as hydride or getter) that will degrade when exposed to the post-weld heat treatment, which is typically performed at about 1,100°C for austenitic stainless steels. High carbon grades, such as Type 347H or Type 316H (having 0.04 percent carbon minimum), have been successfully employed for tritium service if high temperature strength is required. Type 347H is employed in the Hydride Transport Vessel (see section 6.2.2), and Type 316H was DOE-HDBK-1129-99 50 employed for the SRS Building 232-H Extraction Furnace retorts. Type 310 stainless steel has good oxidation resistance and can be considered for elevated temperature applications if oxidation is a concern. Type 316 stainless has superior creep resistance but inferior oxidation resistance to Type 310. Some types of higher-strength austenitic stainless steels not generally employed for tritium service may be required for fasteners (such as nuts and bolts) in mechanical joints in high-temperature regions. This may be acceptable if the bolts are exposed to only residual amounts of tritium. These materials also may be used to contact 3XX stainless steels to avoid galling of mating screw contact surfaces. Typical materials used in these applications include Nitronic 60, Nitronic 50 (also called 21-13-9) and Nitronic 40 (also called 21-6-9); these are all nitrogen-strengthened austenitic stainless steels. 5.1.2.b Copper and Copper Alloys

In principle, copper should be a suitable material in tritium systems. Copper has several tritium- compatible properties. Tritium has a low permeability in copper, and copper is a ductile, stable, face-centered-cubic metal and so is resistant to hydrogen embrittlement. The high thermal conductivity of copper is a desirable property for process vessels requiring heat flow or constant temperature. Copper can be easily joined in a number of ways (e.g. soldering, brazing, welding). In spite of these advantageous properties, copper and copper alloys are not commonly used in tritium systems. Several factors may account for this. The ASME allowable design strength of copper falls rapidly at temperatures above 200 °C, making it difficult to use copper for process beds that operate at elevated temperature. Also, hydrogen isotopes can react at elevated temperature with oxygen in copper, whether the oxygen is in solid solution or in copper oxide precipitates. In either case, water is formed, and over time water vapor agglomerates at grain boundaries, which eventually results in intergranular cavitation, cracking, and failure. This failure mechanism is sometimes termed “steam embrittlement.” Also, a transition junction (normally nickel) is required to join copper and the stainless steel components of the remainder of the system. 5.1.2.c Aluminum and Aluminum Alloys Aluminum has properties making it potentially desirable for tritium systems. It has a low density and a high thermal conductivity. Hydrogen isotope permeability is very low in aluminum compared to stainless steel. Aluminum is used in applications where light weight is important, such as in containers that must be lifted by personnel in gloveboxes. However, aluminum is not commonly used in tritium systems. Stainless steel has much higher strength, at room and elevated temperature. Welding aluminum requires more precautions because aluminum reacts with atmospheric water vapor, which can cause porosity due to hydrogen in the weld fusion zone. 5.1.2.d Materials to Avoid Plain carbon steels and alloy steels must not be used for tritium service. These steels have high strength and (normally) a body-centered-cubic crystal structure, both of which make the material less ductile and much more susceptible to hydrogen embrittlement. Ferritic stainless steels (such as Type 430), martensitic stainless steels (both quench-and-tempered (such as Type 410) and precipitation hardening (such as 17-4 PH and PH 13-8 MO)) and precipitation hardened austenitic stainless steels (such as AM-350) should not be used for general tritium service; they are all more DOE-HDBK-1129-99 51 susceptible to hydrogen embrittlement than the austenitic stainless steels. Additionally, free- machining grades of austenitic stainless steel (such as Type 303) should not be used. Other materials that must not be used for tritium service are any material that forms hydride near room temperature and atmospheric pressure. Examples include zirconium, tantalum, niobium, and many alloys of these materials. 5.1.3 Polymers

All polymers degrade when exposed to radiation. Both tritium and tritiated water permeate all polymers, and permeated tritium deposits the beta decay energy throughout the polymer bulk. (Although the tritium beta energy is very low and has a small penetration depth in matter, permeation allows tritium atoms to be near enough to polymer chains throughout the bulk to cause changes in the polymer by radiation.) Types of radiation-induced changes in polymer properties include either softening (degradation) or hardening, ductility loss, color change, dimensional change, and gas evolution. Because of these effects, polymers should only be used in tritium systems where no metal alternatives exist. Normally, only polymers that harden during radiation exposure are employed, and are replaced before they begin to deteriorate. In addition to polymer breakdown itself, products of degradation can form corrosive liquids or acids such as HF and HCl. Polymer parts must be easily replaceable as a part of normal operations, and a program of regular inspection and replacement should be established. The system should be designed to expose any polymers to as little tritium as possible. Typical uses of polymers in gas-handling systems include gaskets, O-rings, electrical cable insulation and valve parts, including seats, stem tips, and packing. Polymers relatively resistant to radiation can typically withstand up to about 1 million rad (1 rad = 100 erg energy deposited per gram of material). By knowing the solubility of tritium in a polymer at a given temperature and tritium partial pressure and the decay rate of tritium, the approximate dose can be calculated, assuming the tritium concentration has reached equilibrium. Many effects of radiation on polymers are accentuated by oxygen. Protecting polymers from oxygen or air will likely lengthen the lifetime of polymers exposed to tritium. Also, temperatures above about 120°C accelerate radiation effects in polymers, so the temperature of any polymer parts should be kept as low as possible. Inert additives such as glass or graphite generally enhance the resistance of polymers to radiation. Addition of antioxidants may also enhance radiation resistance. 5.1.3.a Plastics VespelTM, a polyamide, has been successfully used for valve stem tips in some tritium laboratories. Ultra-High-Molecular-Weight polyethylene (UHMWPE) and High Density Polyethylene (HDPE) have been used for valve stem tips in automatic valves. Quantitative data still needs to be accumulated on the effectiveness of all of these materials. Success as a stem tip material, particularly UHMWPE and HDPE, have been overstated in the past as described in EH Technical Notice 94-01, “Guidelines for Valves in Tritium Service.” Low-Density polyethylene (LDPE) is very permeable by tritium and tritiated water and should not be considered for use in tritium systems. Polytetrafluoroethylene (PTFE, a trade name is TeflonTM) degrades and decomposes in tritium, thus resulting in HF and HCl. In humid air, hydrochloric and hydrofluoric acid are then formed, which are highly corrosive. Generally, chlorofluorocarbon DOE-HDBK-1129-99 52 polymers should not be used in tritium service. Polyvinyl chloride (PVC) and vinylidene chloride (Saran) are among several polymers used in tritium protective clothing, but should not be used for process equipment because they contain chlorine. 5.1.3.b Elastomers

Tritium readily permeates into and diffuses through elastomeric materials and, depending upon thickness, begins appearing on the outside of the elastomeric seal within hours after exposure to tritium. Elastomers are subject to radiation damage due to exposure to tritium and harden and lose their sealing ability due to exposure to high concentrations of tritium. Ethylene propylene diene monomer (EPDM) elastomers are employed for low-pressure process flange gaskets because of EPDM is relatively good performance in tritium service. In some cases, Buna-N process flange gaskets are being replaced by EPDM when the gaskets are changed; however, in other applications Buna-N remains in use. VitonTM , a common O-ring material, is used, but can embrittle in months in tritium service. Butyl rubber has low permeation for both tritium and tritiated water, but is not as resistant to radiation damage as EPDM. Butyl rubber is used for glovebox gloves. Water vapor in the air outside gloveboxes permeates gloves and can lead to a significant portion of the residual tritiated water vapor in tritium gloveboxes. Kel-FTM is a common chlorofluorocarbon polymer and is incompatible with tritium. It, like TeflonTM, degrades in tritium gas and should not be used. 5.2 First Wall Design 5.2.1 High-Pressure Tritium For high-pressure tritium, it is generally recommended that the first wall be of all metal construction and hydrogen compatible materials including valves, valve seats, and tubing. The use of non- hydrogen-compatible materials will result in material failure and the release of the contained tritium. Elastomers are not tritium compatible, and, as a result, elastomeric seals and valve seats are not recommended for use in the containment of high-purity tritium. There are exceptions to this general case, and other criteria may be used when justified by analysis. 5.2.2 Low- and Medium-Pressure Tritium For the containment of low- and medium-pressure, high-purity tritium, it is generally recommended that the first containment wall be of all metal construction of hydrogen-compatible materials where possible, including valves, valve seats, and tubing. Hydrogen and helium embrittlement of the materials of construction is not usually significant at low and medium pressures. As a result, non- hydrogen-compatible materials may be used if required by the design or if the required component is not available in other materials. Again, elastomers are not tritium-compatible, and, as a result, elastomeric seals and valve seats are not recommended for use in the containment of high purity tritium. There are also exceptions to this general case, and other criteria may be used when justified by analysis. It is difficult to design a vacuum system, which in some cases is the first wall, without including some non-hydrogen-compatible materials and elastomers. However, embrittlement of the materials is not an issue because the tritium exposure is transient and the pressure is low. Under these conditions, the elastomers are not exposed to tritium continuously, and most can be used in DOE-HDBK-1129-99 53 tritium operations under this condition. Surveillance and/or preventive maintenance schedules should be selected in order to maintain elastomer functionality. 5.3 Secondary Wall Design 5.3.1 High-Quality Secondary The design requirements for a high-quality secondary wall are the same as the primary wall. If the secondary wall is required to provide long-term containment of high concentrations of tritium, it should meet the same requirements as the primary or first wall container.

5.3.2 Medium-Quality Secondary If the secondary wall is a glovebox and only contains tritium that has been diluted by the glovebox gases for a short duration (a few hours), e.g., while the glovebox is cleaned up by the tritium removal system, then the requirements can be relaxed. Although the quantity of tritium contained may be quite large, the low pressure and concentration of tritium will not result in material failure due to tritium exposure. 5.3.3 Low-Quality Secondary If the secondary wall is a room or building and only contains tritium that has been diluted by the air contained in the room or building while the room or building is cleaned up by the tritium removal system, then the construction requirements can be relaxed. Although the quantity of tritium contained may be quite large, the low pressure and concentration of tritium will not result in material failure. 5.4 Cleanup System Design Most of the components of the tritium removal and cleanup system are only exposed to tritium at low concentrations and pressure for short time periods. The only long-term exposure is in the water collection system where the water is collected at low pressure on a molecular sieve. All metal construction is recommended, but the materials of construction are not required to be hydrogen-compatible materials. Where appropriate, elastomeric sealing materials have been used successfully in these systems for many years without significant problems. When possible, metal seals should be used because they are more durable and reliable and require less maintenance than elastomeric seals. To minimize the potential for the generation of mixed waste and to decrease radiation exposure of the workers, oil-free pumps should be used where possible. 5.5 Storage System Design Storage systems must consider the total cost of the storage cycle and the purpose for the storage. Storage techniques that increase the complexity of the handling process without adding beneficial features should not be used. The barrier concept discussed in Sections 4.1.1 through 4.1.3, in addition to the wall design considerations discussed above, should be incorporated into all storage system designs. DOE-HDBK-1129-99 54 5.5.1 Short-Term Storage Tritium used to support the day-to-day activities in a facility must be readily available to the facility customers. If the facility uses tritium in gaseous form and its decay to helium does not impact the process, then, to simplify the operation and the equipment, the tritium can be stored in gaseous form. The storage container should be fabricated of all metal, hydrogen-compatible materials including valves, valve seats, and seals. 5.5.2 Medium-Term Storage If tritium is only used in periods of two years or less, the requirements do not change significantly from those of short-term storage. Experience has shown that tritium can be stored safely at near atmospheric pressure for long periods of time. If the buildup of helium in the supply does not impact the use, then storage as a gas is an acceptable alternative. There are, however, advantages of tritide bed storage for medium-term use. Impurities such as nitrogen and oxygen form uranium nitride and uranium oxide and are removed from the gas stream as the bed is heated and cooled. Helium, which accumulates due to the decay of tritium, and any other impurities remain in the overpressure gas above the bed and may be pumped off after the bed had been cooled down and the tritium has been gettered by the uranium. As a result, the uranium bed not only provides for tritium storage but also provides a means of maintaining a reasonably pure and stable tritium supply.

5.5.3 Long-Term Storage Due to the half-life, storage of tritium for several years implies that it is not readily needed. It should be placed in a safe and stable condition while the tritium decays. 5.5.3.a Storage as a Gas Compared to the fabrication and preparation of metallic storage beds, regardless of what metal is used, the cost of storage of tritium as a gas at near atmospheric pressure is economical. ASME- code-designed stainless steel tanks are available or can be designed and fabricated at a reasonable cost. A tank at atmospheric pressure would end with a final of 15 pounds per square inch gauge (psig) pressure after all of the tritium had decayed, so embrittlement is not an issue. The long-term storage of hydrogen and tritium in containers is well understood in comparison to the understanding of long-term storage of metal tritides. 5.5.3.b Storage as a Metal Tritide Uranium beds designed at Sandia in the late 1970s for laboratory use were about the size of two one-gallon paint cans. This included the secondary containment system and electric heaters used to drive the tritium off during tritium removal. These beds were designed to store 50 grams of tritium as uranium tritide that was easily recoverable in a matter of less than an hour. Long-term storage in this type of container is expensive, but the tritium can be easily and quickly recovered for use. Additionally, large uranium beds capable of storing kilogram quantities are possible, and would significantly decrease the volume required to store large quantities of tritium. Also, impurities such as nitrogen and oxygen form uranium tritide and uranium oxide, and are removed DOE-HDBK-1129-99 55 from the gas stream as the bed is heated and cooled. Helium, which accumulates due to the decay of tritium, and other non-reacting impurities remain in the overpressure gas above the bed. To help resolve unknowns regarding consequences of air-ingress accidents in uranium beds, a series of air-ingress experiments was conducted at Ontario Hydro Research Division, with the participation of Princeton Plasma Physics Laboratory (PPPL) and the Idaho National Environmental and Engineering Laboratory. The experiments indicated that the resulting reaction was restrained with only modest temperature excursions. This leads to the conclusion that the hazards associated with an air-ingress accident involving a uranium bed is smaller than previously anticipated. Additionally, tests conducted by WSRC indicate that, except for catastrophic container failure, the tritium release due to air inleakage into a uranium tritide bed is limited by diffusion. Titanium hydride is not pyrophoric at room temperatures, is a stable material, and has been studied for use in the long-term storage of tritium. It is reported to be less prone to spontaneous ignition in air than the parent metal. Following the hydriding process, if the titanium hydride is exposed to air under controlled conditions, a small quantity of hydrogen is released from the material as the oxide layer forms on the surface of the material. Following formation of the oxide layer, titanium hydride is stable in air. Hydrogen will not be released unless the material temperature is significantly increased. Palladium tritide is not pyrophoric at room temperature, is a stable material, and the overpressure of tritium over the bed at room temperature is approximately 50 torr.

Metal tritides have the advantage of significantly decreasing the volume required to store tritium without increasing the pressure of the gas during storage. 5.6 Surveillance and Maintenance The level of equipment surveillance (including radiological monitoring) and maintenance required is based on the hazard class of a facility; i.e., Hazard Category I through III or Radiological. The specific requirements for the different classes of facility equipment are a function of the safety issues associated with the equipment. These are specified in the facility safety analysis report or other facility safety documentation and in the facility maintenance plan. 5.7 Seismic Considerations This section describes 1) DOE Natural Phenomena Hazards Policy, which defines the requirements for protection against natural phenomena such as earthquakes, and 2) Seismic Design and Evaluation of Equipment and Distribution Systems. Seismic and Wind Design and Evaluation of Structures and Facilities are discussed in Section 4.4. 5.7.1 DOE Natural Phenomena Hazards Policy The DOE has developed a policy for the mitigation of natural phenomena (such as earthquakes, extreme winds, and floods) on its facilities. This policy is in the form of an Order, an Implementation Guide, and a series of Standards. DOE Order 420.1 and its Implementation Guide provide the overall requirements for mitigation of the effects of natural phenomena. The Standards lay out the basic performance requirements for structures, systems, and components (SSCs) subjected to loads caused by earthquakes, extreme winds, and floods. DOE-HDBK-1129-99 56 The approach in the DOE policy consists of placing each SSC into a PC based on its function, its importance to safety, and the quantity and type of material involved, if any. DOE-STD-1021-93, Change Notice 1 provides guidance for categorizing SSCs. Background material used in the development of DOE-STD-1021-93 can be found in UCRL-ID-112612. The Standard provides useful dose information for radiological materials and toxic chemicals. Additional information that is useful in determining categorization results from accident analyses, information from a safety analysis report, and information in support of DOE-STD-3009. An equipment list will be developed with the PC for each SSC identified. DOE policy invokes DOE-STD-1022-94, Change Notice 1, “Natural Phenomena Hazards Characterization Guide,” and DOE-STD-1023-95, Change Notice 1, “Natural Phenomena Hazards Assessment Criteria,” if any SSC is categorized as a PC3 or PC4. These Standards define acceptable procedures to determine levels of the natural phenomena hazards at the site for use in design or evaluation. If only PC2 or lower SSCs exist at the site, then Uniform Building Code (UBC) design values are acceptable. However, site-specific values are always preferable if they are available. Design and evaluation criteria are provided in DOE-STD-1020-94, Change Notice 1. This handbook provides an acceptable approach for the design and evaluation of SSCs in PCs 1, 2, 3, and 4. The design basis earthquake and wind can be developed from the available natural phenomena hazard description, either site specific or from the UBC. A table of representative seismic and wind design values for most DOE sites is contained in DOE-STD-1020-94, Change Notice 1. Use of the values in DOE-STD-1020-94, Change Notice 1 must be justified for the site. DOE-STD-1020-94, Change Notice 1 also discusses the 5 percent damped in-structure response spectra, which is used for the design and evaluation of equipment. These spectra must be developed from the description of the ground motion and the facility characteristics.

The user should review these standards, the appendices in the standards, and the references listed below and be aware of the type of information provided therein. DOE has conducted training on the use of the design and evaluation standard since 1989, and copies of the training material are also available. 5.7.2 Seismic Design and Evaluation of Equipment and Distribution Systems In the event of an earthquake, DOE facilities need to have adequate measures for the protection of the public, workers, environment, and investment. Due to the evolutionary nature of design and operating requirements as well as developments in engineering technology, existing DOE facilities embody a broad spectrum of design features for earthquake resistance. These features depend on factors such as vintage of the facility design and construction and hardware supplier practices at the time of design and construction. The earliest vintage facilities often have the least design consideration for seismic-induced forces and displacements and exhibit the greatest difference between their design basis and current requirements for seismic design criteria for new facilities. DOE has developed a Seismic Evaluation Procedure (SEP) to summarize a technical approach and provide generic procedures and documentation requirements, which can be used at DOE facilities to evaluate the seismic adequacy of equipment and distribution systems. The SEP is intended to provide DOE facility managers, safety professionals, and engineers with a practical procedure for evaluating the seismic adequacy of equipment. Often the approach used to review the seismic capacity of equipment in facilities is to conduct sophisticated evaluations, which can be very time consuming, complex, and costly. Much of the available funding and time can be spent on analysis rather than on the real objective of increasing the seismic capacity of the DOE-HDBK-1129-99 57 equipment. The SEP is designed to be an extremely cost-effective method of enhancing the seismic safety of facilities and reducing the potential for major economic loss that can result from equipment damaged by an earthquake. The following is a suggested list of topics to consider in the design of new systems and equipment or the evaluation of existing systems and equipment. Representative equipment that may be found in tritium facilities is listed in Table 5-1. The PC for each system or component to be reviewed must be defined, (PC 1, 2, 3, or 4) in the Seismic Equipment List (SEL). The methodology and procedures for evaluating the seismic adequacy of equipment described in the SEP are based on the observed performance, failure, and response of various types of SSCs during and after they were subjected to either actual or simulated earthquake motion. An SSC in a DOE facility can be evaluated for seismic adequacy provided that the associated guidelines, limitations, requirements, and caveats described in the SEP are satisfied. TABLE 5-1. Representative equipment found in tritium facilities ELECTRICAL EQUIPMENT MECHANICAL EQUIPMENT Batteries on Racks Fluid-Operated / Air-Operated Valves Motor Control Centers Motor/Solenoid-Operated Valves Low- and Medium-Voltage Switchgear Horizontal Pumps Distribution Panels Vertical Pumps Transformers Chillers Battery Chargers and Inverters Air Compressors Instrumentation and Control Panels Motor-Generators Instruments on Racks Engine-Generators Temperature Sensors Air Handlers Computer Data / Storage Systems Fans Alarm Instrumentation HEPA Filters Communications Equipment Gloveboxes

PIPING AND RACEWAY SYSTEMS Miscellaneous Machinery Cable and Conduit Raceway Systems ARCHITECTURAL FEATURES Piping Unreinforced Masonry (URM) Walls HVAC Ducts Hollow Clay Tile Walls Underground Piping Suspended Ceilings Underground Raceways Raised Floors Stacks Storage Racks Conveyors of Material Cranes TANKS Elevators Vertical Tanks SWITCHYARD AND SUBSTATION EQUIPMENT Horizontal Tanks and Heat Exchangers Power Transformers Underground Tanks Miscellaneous Equipment Canisters and Gas Cylinders Miscellaneous Tanks The general approach for the development of the SEL is envisioned to be a three-step process as depicted in Figure 5-1. After a SEL Team is selected, the first step of the process is the development of the preliminary SEL from a list of the facility SSCs. The SEL Team consists primarily of safety professionals and systems engineers with assistance from seismic engineers and facility operators. Only a portion of the facility SSCs will be contained in the SEL and, in many cases, the SEL will contain only safety-related SSCs that must function during or after a seismic event. The selection of the SSCs belonging on the SEL should be based on the results of accident DOE-HDBK-1129-99 58 analyses. These accident analyses should consider the appropriate facility hazards as required by DOE Orders and Standards. For the DOE facility being seismically evaluated, accident analyses and their results are typically provided in a SAR. The preliminary SEL should be based on information provided in this SAR. For a nonreactor nuclear facility, DOE-STD-3009 provides guidance on the preparation of a SAR. Using the guidance in DOE-STD-3009 and the appropriate accident analyses in the SAR, SSCs can be differentiated into Safety Class or Safety Significant, and the preliminary SEL can focus on those facility SSCs. For facilities without a SAR, accident analyses comparable to those required for a SAR should be performed. FIGURE 5-1. Development of the seismic equipment list Additional guidance for the development of the preliminary SEL is provided in DOE-STD-1021. This Standard considers the results of facility hazard classification, SSC safety classification, and performance categorization. With these considerations, the facility SSCs are assigned to the appropriate performance category. The preliminary SEL focuses on those SSCs that are classified above a specified performance category. In addition to selecting SSCs based on a SAR or DOE- STD-1021, there are system safety considerations and seismic vulnerability considerations that should be addressed when developing the preliminary SEL. Next, the location of equipment items in the facility and in-structure response spectra for this level or location are determined. The in-structure response spectra represent the modification of the ground response spectra by the facility. There may be amplification or attenuation of the motion at various frequencies at different locations within the facility. As discussed in DOE-STD-1020-94 (CH-1), the 5 percent damped in-structure response spectra is used for design or evaluation of equipment at a specific location. Guidance for determining in-structure response spectra is provided in the Standard. To evaluate equipment and distribution systems, the following items should be considered: DOE-HDBK-1129-99 59 • Functionality Requirements Does the item need to function during the earthquake? Does it need to function only after the earthquake? Can it be placed back into service by operator actions?

• Does the item have sufficient capacity to meet the earthquake demand specified for its PC? An example of a comparison of capacity to demand is shown in Figure 5-2. • Is the item adequately anchored or braced? There are special considerations for base-isolated or vibration-isolated items since they have behaved poorly during earthquakes. • Does an adequate “Load Path” exist to transfer all seismic loads in addition to operating loads? A load path is a continuous path, which carries load from the top of a structure or component down through an element and finally into the foundation. Loads produced by the earthquake must be combined with normal operating loads such as weight, pressure, temperature, and any temporary loads that may exist. • Are there system interaction issues? Can something fall and cause the item not to function? Can the item impact with nearby items? Is there differential motion between the item and the structure? Is there a water spray or flood issue? Is there a fire issue? FIGURE 5-2. Comparison of seismic capacity spectra to seismic demand spectra Each new system or new component must be designed to meet the DOE seismic criteria specified in DOE Natural Phenomena Policy as specified in Section 4.4. Adequate documentation to support its design must be maintained by the facility management. Adequate inspection should be DOE-HDBK-1129-99 60 conducted to make sure the system or component was fabricated and installed as specified in the design documents. Documentation must be maintained current throughout the life of the item. Evaluation of existing systems or components must be conducted based on the “Actual” condition of the item. This may be different than the design or “As-Built” condition due to field modifications or deterioration of the item during its service. An examination of the item, its installation, and current condition should be made during a walkdown by seismic engineers as defined in the SEP. Existing systems and components may also be evaluated by use of experience data. This is an alternative approach that can be used if the systems or components are installed in an acceptable manner and meet the rules specified in the SEP to ensure that the item is similar to items in the experience database. DOE has developed resources for the evaluation of existing systems and components, has conducted training on their application, and has been implementing the development of seismic evaluation guidelines for systems and components at DOE facilities. 5.8 Fire Scenarios There have been recent investigations that point to the conclusion that fire scenarios are the dominant risk at most tritium facilities. These fire scenarios are not restricted to seismic-induced fires, but include fires from all sources. These recent analyses show that the slower burning fires are limiting, and that in almost all cases, these fires result in oxidation rates that are high (in excess of 90%). Additionally, compliance with National Fire Protection Association (NFPA) requirements, while limiting the potential for fire spread, does not ensure that a full-facility fire scenario need not be analyzed. Reference [26] describes the latest findings in this area.

It is possible that fires of sufficient magnitude and frequency exist at some tritium facilities so as to warrant safety system classifications of both the fire suppression and fire detection systems. It has long been the position of EH-3 that tritium fire suppression and detection systems should be classified as safety systems (at a minimum as safety-significant, and in some cases as safety- class (e.g., where other safety-class equipment operation is contingent upon fire system operation or where fire scenarios result in unacceptable public dose)), unless documented analyses such as the FHA or the SAR justify non-classification. This importance that EH-3 places on fire protection at tritium facilities was highlighted as early as September 1991 during a review of the then-under- construction RTF. The exemption request from some requirements of DOE Orders 5480.7 and 6430.1A for RTF were not approved by EH/NS until physical modifications were made to the tritium facility. [27]. The definition of tritium that was Material at Risk in selected fire (and seismic) scenarios was also developed prior to startup at RTF, which eventually led to decisions on upgraded selected storage containers and inventory limits. [28] 5.9 Instrumentation This section provides information on several instruments used to detect and monitor tritium. The references are provided for information only. DOE does not certify that a particular product is superior to any other product from another vendor. References in this Handbook do not imply endorsement by DOE. 5.9.1 Tritium Monitoring Systems Several different types of instruments may be used to detect and measure tritium in the operation of a facility. Examples and a discussion of such instrumentation follow: DOE-HDBK-1129-99 61 • Ionization Chambers — Tritium decays to 3He by the ejection of a beta particle. The beta particle generated by the decay of tritium ionizes the surrounding gas. The number of ions produced due to the loss of energy of the beta particle is a function of the type of gas. A sample of gas is collected in the ionization chamber and the ionization current is measured. The resulting chamber ionization current is proportional to the quantity of tritium in the gas. The larger the measuring chamber volume the higher the output current and the easier it is to measure. However, as the volume of the chamber increases, the longer it will take to get an accurate measurement. Modern electronic systems have solved most of the problems associated with measuring small ionization currents in small volumes and as a result, the volume of the ionization chambers has been reduced over the years from 50L down to 1 or 2L. Most tritium measuring instruments have an ionization chamber. • Proportional Counters — Gas proportional counters are also used to measure the amount of tritium contained in a gas. A sample of the gas to be monitored is mixed with a counting gas and passed through a proportional counter tube where the pulses caused by the decay of tritium are counted. Proportional counter monitors can be used for most gas monitoring applications and are also available to measure surface contamination.

• Scintillation Crystal Detectors — Scintillation detector systems are used to measure the total mole percent of tritium in a sample of gas independent of the chemical composition of the tritium in the gas (HT, DT, T2, and CHxTy). A sample of the gas is introduced into a measurement chamber at low pressure, generally less than a few torr. The chamber contains a scintillation crystal, which is exposed to the tritium as it decays. The light pulse produced in the scintillation crystal is either counted or is used to produce a current, which is proportional to the mole percent tritium contained in the gas sample. Crystal scintillation detection is generally used to measure the mole percent of tritium in gases containing high concentrations of tritium. • Mass Spectrometer — Magnetic sector, quadrupole, and drift tube mass spectrometers are used as analytical tools to measure the individual components that make up the gas being measured. Mass spectrometers are generally used for the purposes of assay and accountability or for scientific purposes. A sample of the gas to be measured is introduced at low pressure (a few microns) into a chamber and ionized. The ions produced are then measured by a means that discriminates on mass. The number of ions produced at each mass is measured and is proportional to the partial pressure of the component in the gas sample. The sophistication of the measurement systems varies greatly from facility to facility through out DOE. Light isotope, drift tube, mass spectrometers require a large capital investment and require a skilled staff to operate, and, in some cases, may not be cost-effective. All DOE tritium facilities do not require a light isotope drift tube mass spectrometer. Quadrupole mass spectrometers and crystal scintillation detectors are much less expensive, but still require operation by knowledgeable well-trained personnel. The DOE assay and accountability requirements and regulations do not currently reflect this difference in sophistication and cost and currently place the same requirements on small as well as large-scale operations. • Liquid Scintillation Counters — Due to the need to measure the removable tritium on surfaces and in the body water of workers, almost all tritium facilities are equipped with or have access to a liquid scintillation counter. If the scintillation counter is not available on site the service can generally be purchased from a local firm. Liquid scintillation counters are used to measure the quantity of tritium on surfaces, in liquids, and in dissolved samples. For removable surface contamination measurements, a wipe of the surface to be measured is taken using dry paper DOE-HDBK-1129-99 62 or a Q-Tip. The filter paper or Q-Tip is then placed in a scintillation cocktail, and the quantity of tritium is measured by counting the light flashes that occur in the scintillation cocktail as the tritium decays. The surface contamination is then calculated in units of dpm/100 cm2. For liquid measurement, a sample of the liquid to be measured is placed in the liquid scintillation cocktail and measured. The tritium concentration of the liquid is calculated in Ci/mL. For solid measurement, a known weight of a material is dissolved to produce a liquid and then the liquid is sampled and measured in the scintillation counter. The quantity of tritium is then calculated in units of Ci/g of the original solid.

• Gas Samplers — Many different types of gas samplers have been developed and used for measuring very small quantities of tritium in very large volumes of gas. These samplers are used to measure quantities of tritium released through a facility stack and for environmental monitoring at a site. Stack exhaust gas monitoring systems generally use an ionization chamber to measure the tritium in the stack gases and a gas sampler to measure the extremely low levels of tritium that cannot be measured by an ionization chamber. Most of the stack gas samplers are patterned after the ethylene glycol sampling system developed at Mound Laboratories. A commercial version of this system is now available. In this system, a sample of gas from the stack is circulated through six ethylene glycol bubblers in series. The first three bubblers remove tritium in the form of HTO, DTO, and T2O. The gas stream is passed through a heated catalytic reactor where tritium in the form of HT, DT, T2, and CHxTy, is cracked and oxidized to form water. This sample is then passed through three more ethylene glycol bubblers to remove the tritium gas, which is now in the form of water. After a period from a few hours to days, a sample of the ethylene glycol from each bubbler is removed and counted using a scintillation counter to determine the quantity of tritium in each bubbler. Tritium recovered from the first three bubblers is proportional to the tritium in liquid form contained in the stack gases and the tritium recovered in the last three bubblers is proportional to the quantity of tritium in gaseous form contained in the exhaust gases. Due to the extremely small quantity of tritium contained in the atmospheric gases surrounding a tritium facility, the environmental gas samplers use higher flow rate sampler systems than those required for stack monitoring, and, in general, collect the water on molecular sieve traps. The water collected on the molecular sieve traps is then recovered from the trap and the tritium concentration of the gas passing through the trap is calculated from the tritium concentration of the collected water, the gas flow rate through the trap, and the sampling time. • Portable Room Air Monitors — There are several hand held portable room air monitors on the market and their capabilities and ranges vary as a function of the different manufacturer and the purpose for which they were designed. It is convenient in some activities to have the capability to connect a small hose to the monitor so that it may be used to detect tritium leaks around equipment. • Fixed Station Room Air Monitors — Fixed station monitors are designed to be installed in fixed locations and to be used to monitor the room air tritium concentrations. Depending upon the manufacturer they may have several ranges and are equipped with one or two alarm set points and audible as well as visual alarms. DOE-HDBK-1129-99 63 • Glovebox Atmosphere Monitors — Glovebox monitors may be open mesh or closed ionization chambers and are designed to monitor the higher levels of tritium inside the glovebox containment systems. • Hood and Exhaust Duct, Air Monitors — Hood and exhaust duct air monitors are similar to fixed station