Record of Decision: Tritium Supply and Recycling Programmatic Environmental Impact Statement

SUMMARY: The Department of Energy (DOE) is issuing this Record of

Decision regarding DOE's proposal for Tritium Supply and Recycling

Facilities. The Department is making three simultaneous decisions.

First, the Department will pursue a dual track on the two most

promising tritium supply alternatives: to initiate purchase of an

existing commercial reactor (operating or partially complete) or

irradiation services with an option to purchase the reactor for

conversion to a defense facility; and to design, build, and test

critical components of an accelerator system for tritium production.

Within a three-year period, the Department would select one of the

tracks to serve as the primary source of tritium. The other

alternative, if feasible, would be developed as a back-up tritium

source. Second, the Savannah River Site is selected as the location for

an accelerator, should one be built. Third, the tritium recycling

facilities at the Savannah River Site will be upgraded and consolidated

to support both of the dual track options. If the commercial reactor

alternative is selected as the primary source, a tritium extraction

facility will also be constructed at the Savannah River Site. The

environmental analysis to support this decision was issued by the

Department in the Final Programmatic Environmental Impact Statement for

Tritium Supply and Recycling (PEIS) DOE/EIS-0161 (October 1995). The

PEIS identified the dual-track strategy described above as the

preferred technology alternative. The Savannah River Site was

identified as the preferred site for an accelerator, and the site for

the upgrade and consolidation of existing recycling facilities. The

Department has decided to implement the preferred alternatives.

FOR FURTHER INFORMATION CONTACT: Further information on the Final

Programmatic Environmental Impact Statement can be obtained by calling

800-776-2765, or writing to: Stephen M. Sohinki, Director, Office of

Reconfiguration, DP-25, U.S. Department of Energy, P.O. Box 3417,

Alexandria, VA 22302.

Information on the Department of Energy National Environmental

Policy Act process can be obtained by contacting: Carol M. Borgstrom,

Director, Office of NEPA Policy and Assistance, EH-42, U.S. Department

of Energy, 1000 Independence Ave. SW., Washington DC 20585, Telephone:

(202) 586-4600 or (800) 472-2756.

SUPPLEMENTARY INFORMATION: The Department of Energy has prepared this

Record of Decision pursuant to the Council on Environmental Quality

(CEQ) Regulations for implementing the procedural provisions of the

National Environmental Policy Act (NEPA) (40 CFR Parts 1500-1508) and

the Department of Energy regulations implementing the National

Environmental Policy Act (10 CFR Part 1021). This Record of Decision is

based on the Final Programmatic Environmental Impact Statement for

Tritium Supply and Recycling (DOE/EIS-0161, October 1995) and the

Technical Reference Report for Tritium Supply and Recycling (DOE/DP-

0134, October 1995). The Technical Reference Report summarizes

schedule, production assurance and cost data and presents the results

of the uncertainty analysis. Several comments and a report from

Congress were received after the documents listed above were published.

This additional information was taken into consideration in preparing

this Record of Decision.

In January 1991, the Department announced it would prepare a

Programmatic Environmental Impact Statement (PEIS) examining

alternatives for the reconfiguration of the Department's nuclear

weapons complex. The framework for the Reconfiguration PEIS was

described in the Nuclear Weapons Complex Reconfiguration Study (DOE/DP-

0083), issued in January 1991. A Notice of Intent to prepare the PEIS

was published in the Federal Register on February 11, 1991 (56 FR

5590). The purpose of the PEIS was to establish the locations for

future weapons complex missions. The missions to be analyzed included

plutonium and uranium component fabrication and processing, weapons

assembly and disassembly, high explosive production, tritium recycling,

and nonnuclear component fabrication.

At the time the Reconfiguration PEIS was begun, technology and

siting alternatives for a new tritium supply facility were being

examined in a separate New Production Reactor Capacity Environmental

Impact Statement. On September 27, 1991, President Bush announced an

initiative to reduce the Nation's nuclear weapons stockpile. In

response to this initiative, the need for new facilities was delayed

and the Department announced, on November 1, 1991, that it would delay

decisions on the new production reactor technology and siting and

include the environmental analysis for a new tritium production source

in the Reconfiguration PEIS. The Department's intent to incorporate the

New Production Reactor capacity analysis into the Reconfiguration PEIS

was published in the Federal Register on November 29, 1991 (56 FR

60985).

In June 1992, the United States and Russia announced an arms

reduction agreement which was signed in January 1993 as the START II

Protocol. This agreement caused the most significant reductions to date

in planned future weapons stockpiles of both nations. It also provided

the Department with the opportunity to consider a much smaller weapons

complex than previously envisioned. Therefore, the Department

determined that it was necessary to reevaluate the Reconfiguration

Program to insure that alternatives which reflected requirements of a

greatly downsized nuclear weapons stockpile would be assessed in the

PEIS. On July 23, 1993, a revised Notice of Intent was published in the

Federal Register (58 FR 39528) which described a smaller, more

integrated nuclear weapons complex. Additionally, long-term storage

alternatives for plutonium and uranium were added to the analysis. As a

result of this reevaluation and public comment, the Department

published a notice in the October 28, 1994, Federal Register (59 FR

54175), that would separate the Reconfiguration PEIS into two separate

analyses: the Tritium Supply and Recycling PEIS and Stockpile

Stewardship and Management PEIS.

On March 1, 1995 the Department issued a Draft Programmatic

Environmental Impact Statement for Tritium Supply and Recycling (DOE/

EIS-0161) which presented an analysis of the environmental impacts of

the proposed action and alternatives. In the Draft PEIS, the Department

indicated that the use of a commercial reactor was not a reasonable

long-term tritium supply alternative due to concerns about the use of

civilian reactors for military purposes. However, the Draft PEIS

evaluated the impacts associated with the use of a commercial reactor

to make tritium, whether such a reactor were used as a contingency

source of tritium in the event of a national emergency, or the

Department chose to purchase an existing reactor and convert it to a

defense facility for long-term tritium supply. Comments received during

the agency and public review of the Draft PEIS asserted that the use of

[[Page 63879]]

an existing commercial reactor had the potential to be the lowest cost

option and indicated confusion as to whether purchase of a commercial

reactor or irradiation services from a privately owned reactor were

treated as reasonable alternatives capable of meeting long-term tritium

requirements. These comments and concerns prompted the Department to

issue a Federal Register announcement on August 25, 1995 (60 FR 44327)

in which the Department reopened the comment period for 21 days

regarding its intention to treat both the purchase of irradiation

services and the purchase of an existing or partially completed reactor

as reasonable alternatives for long-term tritium supply. The Department

summarized all comments received from both comment periods, prepared

responses to the summaries, made revisions to the PEIS based on the

comments, and identified its preferred alternative. The Notice of

Availability of the Final Programmatic Environmental Impact Statement

was published in the Federal Register on October 27, 1995 (60 FR

55021).

Comments have been received since the Notice of Availability was

published asserting that there are errors in the analysis of cost,

schedule and production assurance, especially regarding a new large

Advanced Light Water Reactor. Comments were also received regarding the

multipurpose reactor concept, and the use of the Fast Flux Test

Facility at the Department's Hanford site to produce tritium. These

comments are addressed in a subsequent section of this Record of

Decision.

Alternatives Considered

Proposed Action: The Department of Energy proposes to provide

tritium supply and recycling facilities for the Nation's Nuclear

Weapons Complex. Tritium, a radioactive isotope of hydrogen, is

produced in nature, but in very small amounts. Therefore, since it is

an essential component of every warhead in the current and projected

U.S. nuclear weapons stockpile, the amounts required must be man-made.

Tritium decays at a rate of approximately 5.5 percent per year and must

be replaced periodically as long as the Nation relies on a nuclear

deterrent. Currently, the Department does not have the capability to

produce the required amounts of tritium. The Department needs a

capability that can produce tritium to meet the requirements set forth

in the 1994 Nuclear Weapons Stockpile Plan, the latest official

guidance. These requirements have been defined as a steady-state mode

of 3/16 of the goal amount previously established for a nuclear reactor

under the Department's New Production Reactors (NPR) program. The

tritium supply source should also be capable of producing 3/8 of NPR

goal amount if necessary either to eliminate inventory shortfalls or to

support a larger stockpile size. The Department is currently meeting

tritium requirements for the stockpile by utilizing tritium recycled

from dismantled weapons. Ratification of the START II Protocol would

mean that new tritium would be required by approximately 2011. The

ability to meet an earlier date, if stockpile requirements should

change, was also analyzed.

New tritium would be supplied, in either a reactor or accelerator,

by irradiating target materials with neutrons and subsequently

extracting the tritium in pure form for its use in nuclear weapons. The

tritium recycling process consists of recovering residual tritium from

weapons components, purifying it, and refilling weapons components with

pure tritium. The Department's tritium recycle facilities are located

at the Savannah River Site (SRS) near Aiken, SC.

Four technology alternatives were evaluated for a new supply

facility--a heavy water reactor, an advanced light water reactor--both

large (1,300 MWe) and small (600 MWe); a modular high temperature gas-

cooled reactor; and a linear accelerator. Emerging design options for

the heavy water reactor and the modular high temperature gas-cooled

reactor were also reviewed. The advanced light water reactor and

modular high temperature gas-cooled reactor alternatives were also

evaluated as to the potential use of fuel fabricated from plutonium

excess to weapons program requirements while simultaneously producing

tritium and electricity (the so-called ``multipurpose reactor''). Five

sites were evaluated for a new facility--the Idaho National Engineering

Laboratory (INEL), near Idaho Falls, ID; the Nevada Test Site (NTS),

near Las Vegas, NV; the Oak Ridge Reservation (ORR), Oak Ridge, TN; the

Pantex Plant (Pantex), Amarillo, TX; and SRS. The Department also

evaluated the use of existing commercial light water reactors, either

through purchase of an existing or partially completed reactor that

would be converted for the production of tritium or through purchase of

irradiation services from a privately owned reactor. The purchase of an

existing or partially completed reactor would allow the Department,

should it choose to do so, to implement the multipurpose reactor

concept. Such use is evaluated in the Final PEIS and the Technical

Reference Report. Additionally, in accordance with CEQ regulations, the

no action alternative (not providing a new supply of tritium) was

evaluated.

Tritium recycling alternatives evaluated included no action

(utilizing existing facilities at the Savannah River Site with no

upgrades or consolidation), consolidation and upgrading of the existing

facilities, or construction of new recycling facilities to be

collocated with a new tritium supply facility if the Savannah River

Site were not chosen as the site for a new tritium supply facility. The

consolidation and upgrading of the Savannah River Site recycling

facilities would support either a new tritium supply facility (if

constructed at the Savannah River Site) or the use of an existing

commercial reactor (if a commercial reactor were ultimately selected as

a long-term tritium supply source or became necessary as a contingency

source of tritium). In addition, a new tritium extraction facility

would be constructed at the Savannah River Site.

Tritium Supply Technology Alternatives

This section describes each of the alternatives. The size of the

facilities, land area requirements, and construction and operation

workforces are presented.

1. No Action: No Action is presented for comparison with the action

alternatives. Under No Action, the Department would not establish a new

tritium supply capability, the current inventory of tritium would

decay, and the Department would eventually not meet stockpile

requirements for tritium.

Construct and Operate New Facilities

2. Accelerator Production of Tritium (APT): An APT would accelerate

a proton beam in a long tunnel toward one of two target/blanket

assemblies located in separate target stations. Such an accelerator

would be approximately 4,000 feet in length and would be housed in a

concrete tunnel buried 40 to 50 feet underground. It would require

approximately 550 MWe of electricity during peak production periods (to

meet the 3/8 requirement) and 355 MWe to produce the steady-state

requirement (to meet the 3/16 requirement) of tritium. In addition to

the accelerator, the facility would include a klystron manufacturing

and remanufacturing building as well as waste treatment, maintenance,

operation, and administrative buildings, and a security infrastructure.

Two target types are being analyzed, a helium-3 target which uses

helium-3 gas to make tritium or a spallation-induced lithium

[[Page 63880]]

conversion (SILC) target which uses lithium-6 to make tritium. The

facilities required for the helium-3 target include target fabrication

and target processing (including extraction) buildings. Facilities for

the SILC target include target fabrication, target processing, and

tritium extraction buildings. The APT complex would cover approximately

173 acres. Construction would take approximately 5 years and require

approximately 2,760 workers during the peak construction year.

Operation of the APT would require approximately 624 workers.

3. Advanced Light Water Reactor (ALWR): The ALWR would be a high

temperature, high pressure reactor whose primary purpose would be to

produce tritium, but which would also generate substantial amounts of

electricity. There are two options for the ALWR technology: A large

ALWR (1,300 MWe) and a small ALWR (600 MWe). Both options use light

(regular) water as the reactor coolant and moderator, and include a

power conversion facility as an integral part of the design. The design

of the ALWR complex would include an interim spent fuel storage

building, a waste treatment facility, a tritium target processing

facility, warehouses, and security infrastructure. Fuel rods would be

purchased from commercial suppliers.

Large ALWR: The large ALWR complex would require approximately 350

acres. Construction would take approximately 6 years and approximately

3,500 workers during the peak construction year. Operation would

require approximately 830 workers.

Small ALWR: The small ALWR complex would also require approximately

350 acres. Construction would take approximately 5 years and require

approximately 2,200 workers during the peak construction year.

Operation would require approximately 500 workers.

4. Heavy Water Reactor (HWR): The HWR would be a low pressure, low

temperature reactor whose sole purpose would be to produce tritium. The

HWR uses heavy water (i.e. deuterium oxide) as the reactor coolant and

moderator. Because of the low temperature of the exit coolant, a power

conversion system designed to produce electrical power as an option

would not be feasible. The conceptual design of the HWR complex

includes a fuel and target fabrication facility, a tritium target

processing building, an interim spent fuel storage building, a general

services building, and security infrastructure. The HWR complex would

cover approximately 260 acres. Construction would take somewhat less

than 8 years and require approximately 2,320 workers during the peak

construction year. Operation would require approximately 930 workers.

Small Advanced HWR: The small advanced HWR is an emerging design

variation of the HWR. The design output of the small advanced HWR would

be 470 MWt compared to 990 MWt for the HWR. It would have the same

configuration of support buildings although they would be somewhat

smaller. The design could be developed to produce tritium to meet

steady-state tritium requirements, or modified to meet peak capacity

requirements. The total area required for the complex would be 150 to

170 acres. Construction would take approximately 5 years and require

approximately 1,800 workers during the peak year of construction. An

operational workforce has not been estimated.

5. Modular High Temperature Gas-Cooled Reactor (MHTGR): The MHTGR

would be a high temperature, moderate pressure reactor whose primary

purpose would be to produce tritium, but which would also generate

substantial amounts of electricity. The MHTGR would use helium gas as a

core coolant and graphite as a moderator. A steam cycle MHTGR would use

a heat converter to transfer the heat from the helium coolant to

feedwater producing super-heated steam which is then used to drive a

turbine in the production of electricity.

The steam cycle MHTGR requires three 350 MWt reactors to produce

the maximum (3/8) requirement of tritium. Because of the high

temperature of the exit coolant, a power conversion facility designed

to produce electricity is an integral part of the design. The design of

the MHTGR complex, in addition to the three reactors, includes a fuel

and target fabrication facility, a tritium target processing facility,

helium storage buildings, waste treatment facilities, interim spent

fuel storage facility, general services building, security

infrastructure, and power conversion facility. The MHTGR complex would

cover approximately 360 acres. Construction of the MHTGR would take

about 9 years and require approximately 2,210 workers during the peak

construction period. Operation would require approximately 910 workers.

Direct Cycle MHTGR: A direct cycle MHTGR is an emerging design

variation of the steam cycle MHTGR. In this design the primary helium

coolant drives a turbine generator through a gas-compression/gas-

expansion, heating/cooling cycle. Two 600 MWt direct cycle reactors

would be needed to produce the maximum (3/8) requirement of tritium.

The support facilities, resource requirements, and environmental

impacts of the direct cycle MHTGR are similar to the steam cycle MHTGR.

A two reactor direct cycle MHTGR would require fewer operating

personnel than the three module steam cycle MHTGR.

Use Existing Reactors

6. Existing Commercial Reactors: The purchase by the Department of

an existing operating reactor, the purchase of a partially completed

reactor, or the purchase of irradiation services from a commercial

power reactor(s)(with an option to purchase the reactor) are the three

options evaluated which utilize existing facilities. Commercial light

water reactors use both pressurized water and boiling water

technologies. The Department has conducted significant development work

on tritium targets for pressurized water reactors. Significant

additional development work would likely be required to develop a

target for a boiling water reactor. The Department plans to proceed

with development of the target for the pressurized water reactor, but

has not ruled out the use of boiling water reactors if industry

demonstrates an advantage to the Department in developing such a

target.

Commercial pressurized water reactors are high-temperature, high

pressure reactors that use ordinary light water as the coolant and

moderator and are capable of generating large amounts of electricity

through a steam turbine generator. A typical commercial light water

reactor facility includes the reactor building, turbine generator

building, auxiliary buildings, interim spent fuel storage facilities,

cooling towers, a switchyard for the transmission of electricity,

maintenance buildings, administrative buildings, and security

facilities.

Purchase of an Operating Commercial Light Water Reactor or Purchase

of Irradiation Services: Approximately 72 to 127 workers (depending

upon the number of reactors utilized) would be added to the work force

because of the tritium activities. New fencing and security buildings

may be required to support additional security requirements. Road

access restrictions or construction of new roads may also be required.

Purchase of a Partially Constructed Commercial Light Water Reactor:

The number of construction workers and the length of the construction

period would vary depending on the percentage of completion of the

plant. Data were available for a two-unit reactor plant

[[Page 63881]]

with one unit 45 percent complete and the second unit 85 percent

complete. The schedule data estimated completing the 45 percent

complete unit in 5 years or both units simultaneously in 7 years. Since

the Department is only interested in one unit, the 5 year estimate was

selected. It is possible that the 85 percent unit could be completed in

a shorter time. For the 45 percent complete unit, peak year workers

were estimated to be approximately 2,065. The 85 percent complete unit

would require a peak work force of approximately 1,525. Operations

would require approximately 830 workers.

Other Missions Beyond Tritium Production

Multi-Purpose Reactor Concept: The ALWR, MHTGR, and the purchase

options of the commercial reactor alternative would also be capable of

utilizing fuel fabricated from excess plutonium to make tritium and

generate electricity. To ``burn'' plutonium in an ALWR or a commercial

light water reactor, a plutonium Pit Disassembly, Conversion, and Fuel

Fabrication Facility would be needed to fabricate the plutonium and

uranium (mixed oxide) fuel rods. For the MHTGR, only a plutonium Pit

Disassembly and Conversion Facility would be needed, because the MHTGR

design already includes a fuel fabrication facility. The MHTGR, if used

to ``burn'' plutonium, would utilize fuel fabricated solely from

plutonium without blending it with uranium. However, because tritium

production declines significantly in a plutonium-fueled MHTGR, twice as

many reactors would be necessary in order to produce the steady-state

(3/16) tritium requirements. The need to include a plutonium Pit

Disassembly, Conversion, and Fuel Fabrication facility for the ALWR and

commercial reactor options, and the need for plutonium Pit Disassembly

and Conversion Facility and more reactors for the MHTGR, would be major

contributors to potential direct environmental impacts.

If an ALWR or commercial light water reactor were used as multi-

purpose facilities, the new plutonium Pit Disassembly, Conversion, and

Fuel Fabrication Facility would cover up to 129 acres and require a

peak construction work force of approximately 745 during the 6-year

construction period. Operation would require approximately 650 workers.

If an MHTGR were used as a multi-purpose reactor, the new plutonium Pit

Disassembly and Conversion Facility would cover up to 30 acres and

require a peak construction work force of approximately 125 during the

6-year construction period. Operation would require approximately 520

workers.

Recycling Facilities

The tritium recycling facility processes and recycles tritium for

use in nuclear weapons. This includes emptying reservoirs returned from

weapons in the stockpile, recovering and purifying the tritium,

reclaiming reusable reservoirs, providing new gas mixtures, and

refilling reservoirs. The facility also tests reservoirs and provides

appropriate waste management activities.

1. No Action: The Department currently operates tritium recycling

facilities at the Savannah River Site. These facilities would continue

to operate without modifications or consolidation to meet

environmental, health, and safety requirements, or to maximize

efficiencies. Environmental impacts would not change from those

experienced today.

2. Construct New Facilities: If the tritium supply and recycling

facilities were to be located at any site other than the Savannah River

Site, new recycling facilities could be collocated with the supply

facilities. The tritium recycling activities would be housed in two

buildings for operations and several support facilities. All tritium

handling activities would be completed in the tritium processing

building, which would be designed to contain tritium releases should

they occur. An auxiliary building would house non-tritium activities

and extremely small amounts of working tritium. The recycling

facilities would cover approximately 196 acres. Construction would take

approximately 4 years and require approximately 335 workers during the

peak year of construction. Operation of the recycling facilities would

require approximately 910 workers.

3. Upgrade Existing Facilities at Savannah River Site: There are

two options for the upgrade of recycling facilities at the Savannah

River Site. The first, the unconsolidated upgrade, would result in the

continued use of all existing facilities and thus no consolidation of

tritium handling activities. Five buildings would be upgraded in order

to meet environmental, health, and safety requirements. No additional

land area would be required. Construction of the upgrades would take

approximately 3 years and require approximately 26 workers during the

peak construction year. Operations would require approximately 970

workers.

The second option, the consolidated upgrade, would result in

closing one building and transferring its functions to two existing

buildings. Four buildings would be upgraded to meet environmental,

health, and safety requirements and one to accept the transferred

activities. The land area required for the facilities would not change.

Construction would take approximately 3 years and require approximately

36 workers during the peak construction year. Operations would require

approximately 910 workers.

Siting of New Tritium Supply Facilities

New tritium supply facilities, if constructed, would be located at

one of five sites currently owned by the Department. These five sites

are:

1. Idaho National Engineering Laboratory: The INEL is situated on

approximately 570,000 acres, approximately 50 miles west of Idaho Falls

and presently employs approximately 10,100 workers. The site has been

used to test, build, and operate nuclear facilities. Research and

development activities include reactor performance studies, materials

testing, environmental monitoring, waste processing, breeder reactor

development, and naval reactor operator training. Currently, there are

four operational reactors. In addition to nuclear research, INEL

supports processing and/or storage of high-level, low-level, and

transuranic radioactive wastes.

2. Nevada Test Site: The NTS is situated on approximately 854,000

acres, 65 miles northwest of Las Vegas. Approximately 6,850 workers are

presently employed at the site. The site is a remote secure facility

for conducting underground testing of nuclear weapons and evaluating

the effects of nuclear detonations on military communications,

electronics, satellites, sensors, and other materials. NTS is also the

location of a low level radioactive waste management facility.

3. Oak Ridge Reservation: The ORR is located on approximately

35,000 acres, 20 miles west of Knoxville, TN. Approximately 15,000

workers are presently employed at the site. It includes three major

facilities: the Oak Ridge National Laboratory; Y-12 Plant, and the K-25

site. The Oak Ridge National Laboratory conducts basic and applied

scientific research and technology development. The K-25 site is the

location of the former Oak Ridge Gaseous Diffusion Plant. It currently

serves as an operations center for environmental restoration and waste

management programs. Y-12 is the primary location for nuclear weapons

activities at Oak Ridge. These include

[[Page 63882]]

the dismantling of nuclear weapons components, maintaining uranium and

lithium component fabrication capabilities, and storing special nuclear

materials.

4. Pantex Plant: The Pantex Plant is located on 10,000 acres, 17

miles northeast of Amarillo, TX. Approximately 3,400 workers are

presently employed at the site. Activities at Pantex include

fabrication of chemical explosives, nuclear weapons assembly and

disassembly, testing, repair and disposal of nonnuclear components, and

development activities in support of the national laboratories. Pantex

also is the interim storage site for sealed plutonium components from

dismantled weapons.

5. Savannah River Site: The SRS is situated on approximately

198,000 acres, 12 miles south of Aiken, SC. Approximately 20,300

workers are presently employed at the site. Currently, tritium

recycling operations to support nuclear weapons activities are

conducted at the SRS. Other activities include interim storage of

plutonium, waste management, and environmental monitoring and

restoration. Past activities at SRS have included nuclear fuel and

tritium target fabrication, operation of reactors for nuclear material

production, chemical separation for recovery of plutonium and plutonium

isotopes, tritium extraction, and uranium fuel reprocessing. The

facilities that supported these past activities are currently

supporting waste management and environmental cleanup activities and

will ultimately be decommissioned and decontaminated.

Commercial Reactor Site: The commercial light water analysis does

not evaluate a specific site. Currently, commercial light water

reactors are operating on 59 sites in 32 states. Approximately one-half

of these sites contain two or three nuclear units. The sites range in

size from 84 to 30,000 acres. The largest use of the sites is for

cooling systems, including reservoirs and artificial lakes, and safety

buffer areas. Analysis of specific candidate reactors would be

conducted in a separate NEPA document.

Preferred Alternative

Based on the analysis presented in the PEIS and Technical Reference

Report, the Department announced a preferred alternative in the FINAL

PEIS. The preferred alternative is a acquisition strategy that assures

tritium production for the nuclear weapons stockpile rapidly, cost

effectively, and safely. The preferred strategy is to begin work on the

most promising production alternatives of purchasing an existing

commercial light water reactor or irradiation services with an option

to purchase the reactor for conversion to a defense facility, and to

design, build, and test critical components of an accelerator system

for tritium production.

The Savannah River Site was designated as the preferred site for an

accelerator, should one be built. The preferred alternative for tritium

recycling and extraction activities was to remain at the Savannah River

Site with appropriate consolidation and upgrading of current

facilities, and construction of a new extraction facility.

Tritium Supply Evaluation

This section describes the results of the Department's evaluation

of each of the alternatives. It summarizes their environmental impacts,

costs, and schedule and production assurance risks. The evaluation of

schedule, production assurance and costs were completed by developing

base estimates and then conducting a formal assessment by experts to

determine the risk. The risk is presented as the probability of

achieving a specific objective. Base cases were developed for six

schedule components, production capacity and availability, and five

cost components. The estimates were normalized to insure consistency

across all tritium supply alternatives. Technical experts (different

groups for schedule, production assurance, and cost) were asked to

provide judgments of the probability of success of the base estimates

for each of the schedule components, capacity and availability, and

each of the cost components. In addition, potential technical,

regulatory, or institutional problems were identified for each tritium

supply alternative and their probability for causing schedule delay,

production assurance uncertainty or cost uncertainty were assessed. The

impacts of the problems on schedule, capacity and availability, and

cost were assessed. This information was combined through multiple

simulations to develop probabilities of meeting various schedule,

production assurance and cost objectives. The environmental impacts

reported in the PEIS were evaluated for discriminators among tritium

supply technologies and among sites.

The schedule, production assurance, and waste factors which

discriminate among tritium supply technology alternatives are

summarized in Table 1. These are: (1) The capability of meeting a

schedule supporting a START II Protocol stockpile size; (2) the

likelihood of producing the amount of tritium necessary to meet maximum

(3/8) tritium requirements; (3) amount of additional spent fuel

generated; and (4) amount of additional solid low level radioactive

waste generated. Costs are presented in Table 2. They are divided into:

(a) Total life cycle cost with revenue; (b) total life cycle cost

without revenue; (c) total project cost; (d) operations and maintenance

cost; and (e) revenue.

Additional environmental discriminators are the need for or

generation of electricity, and cancer risk from a severe accident. The

APT and HWR are users of electricity while the ALWR(s), MHTGR(s), and

purchase of a partially completed or existing commercial reactor will

result in the generation of additional electricity. The range between

the potential amount of electricity used (550 MWe for the APT) and the

potential amount of electricity generated (1,300 MWe for the large

ALWR) is 1,850 MWe. The amount of electricity used was evaluated for

each candidate site against the capability of the power pool to supply

electricity. No significant impacts on the pool or the ability to

supply the required amounts were identified. A separate evaluation of

the option of the construction and operation of a dedicated 550 MWe

coal or gas-fired electrical generating plant was completed for the

APT. The potential impacts of a gas-fired electrical generating plant

were incorporated into the environmental analysis for each of the

sites. The cancer risks attributable to a severe accident are, in

absolute terms, very low for each alternative. However, in comparative

terms, the APT clearly has a significantly lower cancer risk than any

of the new facility reactor alternatives. Therefore, cancer risk is

considered a discriminator between the APT and new reactor alternatives

for the purposes of this decision. The results of the evaluations are

described below.

[[Page 63883]]

Table 1.--Schedule, Production Assurance, and Waste Discriminators

--------------------------------------------------------------------------------------------------------------------------------------------------------

Probability of

Probability of producing START I Additional spent fuel generated per Additional solid low level waste

Alternatives delivering first amounts in any year (yd \3\/yr) generated (yd \3\/yr)

gas in 2011 a one year

--------------------------------------------------------------------------------------------------------------------------------------------------------

No Action......................... 0 0 0..................................... 0

APT............................... 0.76 0.77 0..................................... 57 e

Large ALWR........................ 0.78 0.96 55.................................... 710

Small ALWR........................ 0.78 0.89 36.................................... 660

HWR............................... 0.40 0.93 7..................................... 5,200

Small Advanced HWR................ b 0.79 f.................................. f

Steam Cycle MHTGR................. 0.22 0.86 80.................................... 1,300

Direct Cycle MHTGR................ c 0.49 82.................................... 1,300 g

Purchase Existing CLWR............ >0.99 d >0.96 40.................................... 160

Purchase Partially Complete CLWR.. >0.99 d >0.96 Similar to Large ALWR................. Similar to Large ALWR

Purchase Irradiation Services..... >0.99 d >0.96 0 to 40 depending on number of 160

reactors used.

--------------------------------------------------------------------------------------------------------------------------------------------------------

a Includes technical, regulatory, and institutional delays.

b Due to emerging state of technology longer delays than HWR assumed.

c Probability without any delays is 0.14. Delay would reduce this probability.

d Assumes institutional questions are resolved.

e For Helium-3 target; 544 yd3/yr for SILC target.

f No analysis completed, however, expected to be the same or less than the HWR.

g No analysis completed, however, expected to be approximately the same as the steam cycle MHTGR.

1. Ability to meet required schedules. To meet projected stockpile

requirements for tritium, new tritium gas is required by 2011. This

date is based on a stockpile consistent with the START II Protocol.

Maintaining a stockpile consistent with the START I Treaty would

require new tritium gas by 2005. The schedule analyses assumed a

requirement to deliver tritium in 2011. A sensitivity analysis assessed

the ability of the alternatives to deliver new tritium gas in 2005.

[[Page 63884]]

Table 2.--Cost Evaluation Data at Savannah River Site

[In Billions of 1995 Dollars]

--------------------------------------------------------------------------------------------------------------------------------------------------------

Total life cycle cost w/o Total life cycle cost w/o Total project cost a Operations and maintenance b

revenue a revenue a ------------------------------ costs a

Alternatives ------------------------------------------------------------ -----------------------------

Low Mean High Low Mean High Low Mean High Low Mean High

--------------------------------------------------------------------------------------------------------------------------------------------------------

No Action....................... 0 0 0 0 0 0 0 0 0 0 0 0

APT............................. 3.6 5.1 7.8 3.6 5.1 7.8 2.0 3.0 5.5 1.4 2.1 2.9

Large ALWR...................... 1.3 3.7 7.6 4.8 6.5 10.2 2.4 4.0 7.2 1.9 2.5 3.3

Small ALWR...................... 1.5 2.7 4.5 3.2 4.2 5.8 1.6 2.3 3.9 1.4 1.8 2.5

HWR............................. 4.3 5.8 8.1 4.3 5.8 8.1 2.6 3.8 6.4 1.4 2.0 2.9

Small Advanced HWR.............. 2.9 4.2 5.7 2.9 4.2 5.7 1.8 2.7 4.1 0.9 1.5 2.1

Steam Cycle MHTGR............... 4.1 6.3 9.9 5.0 7.1 10.7 3.0 4.5 7.8 1.8 2.6 3.7

Direct Cycle MHTGR.............. 3.0 5.0 8.3 4.2 6.0 9.5 2.5 4.1 7.4 1.4 1.9 2.6

Purchase existing CLWR.......... 0.8 1.4 3.8 2.8 4.1 5.2 1.0 1.7 2.8 1.7 2.4 3.2

Purchase partially complete CLWR 0.1 2.0 4.4 2.9 4.4 6.6 1.1 1.9 3.4 1.7 2.5 3.8

Purchase Irradiation Services... 0.8 1.2 1.7 0.8 1.2 1.7 0.3 0.5 0.7 0.5 0.7 1.1

--------------------------------------------------------------------------------------------------------------------------------------------------------

a The costs are for steady-state production, discounted at 4.9% per year at SRS for new facilities. The low estimate corresponds to the 5th percentile

of the cost probability distribution. The high estimate corresponds to the 95th percentile of the cost probability distribution.

b Operations and maintenance costs include decontamination and decommissioning costs.

[[Page 63885]]

The potential for technical or regulatory delays in the baseline

schedule was also considered in assessing schedule uncertainties for

each of the technologies. Technical delays relate to issues such as the

maturity of the facility design, operational experience associated with

the technology and maturity of the target design. Regulatory delays

relate to the potential that independent reviews by organizations

external to the Department could take longer than anticipated, either

due to administrative licensing proceedings or to resolution of

technical issues that delays design acceptance by the reviewing

organization. By the end of 1995, a Task Force on External Regulation

established by the Department is scheduled to present its

recommendations whether the Department's nuclear facilities should be

externally regulated, and if so, by what entity. While a number of

different outcomes are possible as a result of the Task Force efforts,

the Nation's commercial nuclear reactors are now regulated by the

Nuclear Regulatory Commission (NRC). Therefore, in considering

scenarios that involved regulatory delay, the Department used the NRC

regulatory process and structure as the basis for this consideration,

and assumed that an NRC license would be obtained for construction and

operation of the reactor technologies.

Since the NRC has the greatest amount of experience with regulation

of light water reactors, the potential regulatory delays associated

with the light water options, either the new ALWR designs or the

existing commercial reactor options, were assumed to be the shortest

among the reactor technologies. Potential regulatory delays associated

with the MHTGR and the HWR would be greater than for the light water

candidates because changes to the NRC's regulatory structure would be

required to license these technologies. While there will be technical

and potential regulatory reviews associated with the APT design, the

safety issues associated with this technology are not nearly as complex

as those associated with any of the reactor technologies. Therefore,

the potential for regulatory delays was assessed to be minimal. The

purchase of an existing or partially complete commercial reactor would

also require the transfer of a license to the Department, which would

require a change to the Atomic Energy Act and corresponding changes to

the NRC regulations.

While issues related to the new facility technologies are primarily

technical and regulatory, existing commercial reactors are subject to

an additional set of institutional issues that must be resolved before

this option could be implemented to meet long-term tritium

requirements. These center around concerns about the use of civilian

commercial reactors for purposes which support military requirements.

Such issues have been raised in the past predominantly in conjunction

with the use of civilian reactors to produce special nuclear materials

(highly enriched uranium and plutonium) which would, in turn, be used

to make nuclear weapons. Any concerns will have to be addressed and

resolved over the course of the next several years if the commercial

reactor alternative options are to be utilized as the primary long-term

source of tritium.

The no action alternative would not be able to produce new tritium.

Therefore, it could not meet the schedule requirements.

Of the action alternatives, the commercial reactor options have the

highest probability of meeting the 2011 start date, if there are no

technical or institutional delays. However, as noted above, there are

institutional issues related to their implementation. If these issues

cannot be resolved, the commercial reactor alternative would remain

only as a contingency source of tritium in the event of an emergency.

Even when delays or major issues are taken into account, the ALWRs,

among the new facility alternatives, have a high probability of meeting

the required 2011 start date. The base case construction schedule of

the small ALWR is one year shorter than that of the large ALWR.

However, the small ALWR has a higher risk of technical delays due to

the uncertainties surrounding its passive safety system and potential

regulatory delays, due to the fact that it has not yet received NRC

design certification. The APT has only a slightly smaller probability

of meeting the 2011 date compared to the ALWRs, and it is expected to

have very few technical or regulatory delay problems. The HWR and the

MHTGR would have difficulty in meeting the 2011 date.

The sensitivity analysis on producing tritium as early as 2005

assumed that the base schedules could be compressed by 2 years, and

that no technical or regulatory delays would occur. It showed that the

commercial options have a high probability (0.80 to 0.99) of meeting

the 2005 date. The APT and the small ALWR have a small (0.20)

probability of producing tritium by 2005 if no delays are experienced.

None of the other alternatives could produce tritium by 2005.

The assessment also showed that the schedule for completing all

activities to develop a multipurpose reactor would be similar or

identical to that of the MHTGR, ALWRs, and purchase of a commercial

reactor options if they are used for tritium production alone, as long

as the tritium mission is given priority over the plutonium burning and

electricity production missions.

In summary, the no action alternative is not able to meet tritium

schedule requirements. The HWR and MHTGR have the potential for major

technical or institutional delays; thus, there is a low probability of

their making tritium by the 2011 start date. The ALWRs and the APT have

a very high probability of delivering tritium by 2011. The commercial

options have the highest potential for delivering tritium by 2011, if

the institutional issues associated with the defense use of such

facilities can be resolved. Only the commercial options have a high

probability of delivering tritium by 2005, if that becomes a

requirement.

2. Ability to produce the required amounts of tritium. Production

assurance refers to the ability of the tritium supply alternatives to

meet the annual production requirements for maintaining the tritium

inventory. The steady-state (3/16) and maximum (3/8) production rates

were used in the production assurance analysis.

The second column of Table 1 summarizes the results of the

production assurance analysis in terms of the probability that a

tritium supply option can meet the maximum rate in any given year.

Since the facility is designed to operate for 40 years, a technology

that produced at more than the maximum rate in any given year would

produce excess tritium. If such a year is followed by a year that the

technology produced at less than the maximum rate, the combination of

years would still produce roughly the desired overall quantity of

tritium over the 40-year lifetime of the facility. Thus, a production

rate with a 0.50 probability of a rate meeting or exceeding the maximum

rate in any given year provides a reasonable degree of production

assurance. A 0.75 probability of meeting or exceeding the maximum rate

every year is a high degree of production assurance, since it means

that roughly during 30 years of the 40 years of production the maximum

rate will be exceeded.

For all tritium supply alternatives, with the exception of the

direct cycle MHTGR, there is a high probability of producing the

required amounts of tritium (0.77 or higher). The direct cycle

[[Page 63886]]

MHTGR has a moderate probability of production assurance (0.49).

The production assurance of a multipurpose reactor would not change

from that of the MHTGR, ALWR, and commercial reactor purchase options,

as long as tritium production is the primary mission of the facility.

National security requirements mandate that tritium supply remain the

primary mission of a multipurpose reactor.

In summary, the no action alternative has no chance of meeting the

tritium production requirements. With the exception of the direct cycle

MHTGR, all other alternatives have very high probabilities of meeting

the steady-state and maximum production requirements.

3. Environmental Impacts. The Final PEIS presents numerous

environmental impacts for a variety of resource areas for each of the

new tritium supply facility alternatives at each of the five sites, and

generic impacts for the commercial reactor options. The analysis was

completed for meeting the maximum (3/8) goal requirement of tritium.

Many of these impacts are very small. For example, the air quality

impacts of all technological alternatives at all sites are very low.

Most other impacts show little or no differentiation among

alternatives. The evaluation of the tritium supply alternatives

focuses, therefore, on the three environmental impacts that

differentiate among the tritium supply alternatives: spent fuel

generation, low level radioactive waste generation and risks from

severe accidents. For all three of these area of environmental impact,

the no action alternative would not change the status quo, i.e., no

tritium would be produced. Therefore, it has the lowest environmental

impact. This section presents the evaluation of tritium supply

technology alternatives which are not site dependent. The following

section presents the evaluation of the sites.

3.1 Spent fuel. Spent fuel is measured by the cubic yards of

radioactive spent fuel rods produced during reactor operations in one

year. The third column of Table 1 shows the annual amounts of spent

fuel generated by the reactor supply alternatives. The new reactors

generate spent fuel amounts ranging from 7 cubic yards to 80 cubic

yards. The options to purchase an operating reactor or to purchase

irradiation services would create up to 40 cubic yards of additional

spent fuel (if only one reactor were utilized) due to shorter refueling

cycles. If there were no change to the refueling cycles, no additional

spent fuel would be generated. The option to purchase an incomplete

reactor would create amounts of spent fuel comparable to those of the

large ALWR. The APT does not generate any spent fuel. No additional

spent fuel would be produced by virtue of the use of fuel fabricated

from excess plutonium for the ALWR, MHTGR, or purchase commercial

reactors options.

3.2 Low level radioactive waste. The fourth column of Table 1

shows the annual amounts of low level radioactive waste produced by the

supply alternatives. For the new facility alternatives the HWR creates

by far the most low level radioactive waste (5,200 cubic yards),

followed by the other new rectors. The APT generates the least amount

of low level radioactive waste (57 cubic yards) when using the helium-3

target, and 544 cubic yards when using the SILC target. The options to

purchase an operating commercial reactor or to purchase irradiation

services would create 160 cubic yards of additional low level

radioactive waste due to the use of additional fuel rods and to

handling additional radioactive materials. The option of purchasing an

incomplete reactor would produce amounts of low level radioactive

wastes that are similar to those of the large ALWR. A multipurpose

reactor would generate about the same amount of low level radioactive

waste as the reactor when used for tritium production alone. However,

the plutonium Pit Disassembly and Conversion and Mixed-Oxide Fuel

Fabrication Facility for the ALWR and commercial reactor options would

generate approximately 540 cubic yards of low level radioactive waste

annually. The plutonium Pit Disassembly and Conversion Facility for the

MHTGR would generate approximately 10 cubic yards of low level

radioactive waste per year.

3.3 Severe accidents. Risk is the probability of an accident

occurring times the consequences of the accident if it occurred. Cancer

risk to a population within a 50-mile radius of a facility is

influenced by the size of the population within the radius. However,

technologies can be compared if the same 50 mile radius is used for the

analysis. For the purposes of comparison the SRS is used. The annual

cancer risk from a severe accident to the population within 50 miles of

the facility for the new reactor technologies is very low, ranging from

5.1x10-5 to 2.6x10-7 at the SRS. The APT would have the

lowest annual cancer risk (2.8x10-11) for all the new facility

alternatives. The options to purchase an operating reactor or to

purchase irradiation services would pose no significant additional

severe accident risks because of adding tritium production. The option

to purchase an incomplete commercial reactor would have severe accident

risks that are comparable to that of a large ALWR.

The use of plutonium as mixed oxide fuel in an ALWR or the purchase

of commercial reactor options would not significantly affect the

consequences of radioactivity releases from severe accidents though

there would be some small changes in the source term release spectrum

and frequency. The MHTGR would have twice as many reactors when

operated in the multipurpose mode, and therefore, while extremely

small, the accident risk for the MHTGR would double if used in this

mode compared to the risk if used for tritium production alone.

An accident at a plutonium Pit Disassembly and Conversion and

Mixed-Oxide Fuel Fabrication Facility for the ALWR and purchase of

commercial reactor options would result in a small additional cancer

risk from a severe accident if located at the SRS. A severe accident at

the plutonium Pit Disassembly and Conversion facility for the MHTGR

would also result in a small additional cancer risk.

In summary, the no action alternative has no additional

environmental impacts. The APT and the commercial options to purchase

an operating reactor or to purchase irradiation services, if the fuel

cycle is not changed, generate no additional spent fuel, and have the

lowest amounts of additional low level radioactive waste and cancer

risks from a severe accident. The new reactor alternatives and the

completion of a partially complete commercial reactor produce spent

fuel and low level radioactive waste, and they present a very small

additional cancer risk from a severe accident.

4. Affordability (Cost). For each action alternative, a range of

costs, and the probability distributions over the range, were developed

for Total Life Cycle Cost (TLCC), Total Project Cost (TPC), and

Operation and Maintenance (O&M). The O&M costs included decontamination

and decommissioning. No costs were developed for the no action

alternative. For the action alternatives, results were calculated for

both undiscounted and discounted cost. The discount rate used was 4.9%

per year in accordance with Office of Management and Budget guidance.

The ALWR, MHTGR, and purchase commercial reactor options can produce

revenues through electricity generation. The TLCC was calculated with

and without revenues for these alternatives. Costs were estimated both

for steady-state and maximum production rates.

[[Page 63887]]

The results of the cost ranges for steady-state production using

discounted 1995 dollars are shown in Table 2. For each alternative a

low, mean and high cost estimate is presented for TLCC with revenue,

TLCC without revenue, TPC and O&M. The low estimate is the 5th

percentile of the cost probability distribution, i.e., there is a 5%

chance that the true cost will fall below the low estimate. The mean

estimate is the average of the cost probability distribution. The high

estimate is the 95th percentile of the cost probability distribution,

i.e., there is a 95% chance that the true cost will fall below it.

The TLCC with revenue represents the estimated cumulative

discounted net cost to the government or the taxpayers for each of the

alternatives, since revenues from electricity sales would come to the

government, not the Department. The Department must budget for all

costs; therefore, the TLCC without revenue shows the estimated

cumulative discounted cost to the Department. TPC represents the

discounted capital cost estimates to develop, construct and make

operational each alternative. The O&M costs are the discounted costs

after the facility would become operational.

For TLCC with revenues (first column of Table 2), the option to

purchase irradiation services has the lowest mean estimated cost (1.2

billion dollars) with uncertainty adding approximately 500 million

dollars (95th percentile above the mean). The option to purchase an

existing reactor has a mean cost of 1.4 billion dollars (17 percent

higher than purchasing irradiation services) with uncertainty adding

approximately 2.4 billion dollars. The option to purchase a partially

complete commercial reactor has a mean cost of 2.0 billion dollars (67

percent higher than purchasing irradiation services) with uncertainty

adding 2.4 billion dollars. The new reactor technology alternatives

have mean costs that range from 2.7 billion dollars for the small ALWR

(125 percent higher than purchasing irradiation services) to 6.3

billion dollars for the steam cycle MHTGR (425 percent higher than

purchasing irradiation services). All new reactor alternatives have

significant cost uncertainties, which add from 1.5 billion dollars

(small advanced HWR) to 3.9 billion dollars (large ALWR). The APT has a

mean cost of 5.1 billion dollars (325 percent higher than purchasing

irradiation services) with uncertainty adding approximately 2.7 billion

dollars. The large uncertainties create a substantial overlap in the

cost distributions of the alternatives, except for the purchase of

irradiation services.

For TLCC without revenues (second column of Table 2), the option to

purchase irradiation services has the lowest mean estimated cost (1.2

billion dollars) with uncertainty adding approximately 500 million

dollars (95th percentile above the mean). The option to purchase an

existing reactor has a mean cost of 4.1 billion dollars (242 percent

higher than purchasing irradiation services) with uncertainty adding

approximately 1.1 billion dollars. The option to purchase a partially

complete commercial reactor has a mean cost of 4.4 billion dollars (267

percent higher than purchasing irradiation services) with uncertainty

adding approximately 2.2 billion dollars. The new reactor technology

alternatives have mean costs that range from 4.2 billion dollars for

the small ALWR and small advanced HWR (250 percent higher than

purchasing irradiation services) to 7.1 billion dollars for the steam

cycle MHTGR (492 percent higher than purchasing irradiation services).

All new reactor alternatives have significant cost uncertainties, which

add from 1.5 billion dollars (small advanced HWR) to 3.7 billion

dollars (large ALWR). The APT has a mean cost of 5.1 billion dollars

(325 percent higher than purchasing irradiation services) with

uncertainty adding approximately 2.7 billion dollars. The large

uncertainties create a substantial overlap in the cost distributions of

the alternatives, except for the purchase of irradiation services.

For TPC (third column of Table 2), the option to purchase

irradiation services has the lowest mean estimated TPC (0.5 billion

dollars) with uncertainty adding approximately 200 million dollars

(95th percentile above the mean). The option to purchase an existing

reactor has a mean TPC of 1.7 billion dollars (240 percent higher than

purchasing irradiation services) with uncertainty adding approximately

1.1 billion dollars. The option to purchase a partially complete

commercial reactor has a mean TPC of 1.9 billion dollars (280 percent

higher than purchasing irradiation services) with uncertainty adding

1.5 billion dollars. The new reactor technology alternatives have mean

TPCs that range from 2.3 billion dollars for the small ALWR (360

percent higher than purchasing irradiation services) to 4.5 billion

dollars for the steam cycle MHTGR (800 percent higher than purchasing

irradiation services). All new reactor alternatives have significant

cost uncertainties, that add from 1.4 billion dollars (small advanced

HWR) to 3.3 billion dollars (Direct Cycle MHTGR). The APT has a mean

TPC of 3.0 billion dollars (500 percent higher than purchasing

irradiation services) with uncertainty adding approximately 2.5 billion

dollars. The large uncertainties create a substantial overlap in the

TPC distributions of the alternatives, except for the purchase of

irradiation services.

The O&M costs make up the fourth cost item (fourth column of Table

2). The option to purchase irradiation services has the lowest mean

estimated O&M cost (700 million dollars) with uncertainty adding

approximately 400 million dollars (95th percentile above the mean). The

option to purchase an existing reactor has a mean O&M cost of 2.4

billion dollars (243 percent higher than purchasing irradiation

services) with uncertainty adding approximately 800 million dollars.

The option to purchase a partially complete commercial reactor has a

mean O&M cost of 2.5 billion dollars (257 percent higher than

purchasing irradiation services) with uncertainty adding 1.3 billion

dollars. The new reactor technology alternatives have mean O&M costs

that range from 1.5 billion dollars for the small advanced HWR (114

percent higher than purchasing irradiation services) to 2.6 billion

dollars for the steam cycle MHTGR (271 percent higher than purchasing

irradiation services). All new reactor alternatives have significant

O&M cost uncertainties, that add from 600 million dollars (small

Advance HWR) to 1.1 billion dollars (steam cycle MHTGR). The APT has a

mean O&M cost of 2.1 billion dollars (200 percent higher than

purchasing irradiation services) with uncertainty adding approximately

800 million dollars. The large uncertainties create a substantial

overlap in the cost distributions of the alternatives, except for the

purchase of irradiation services.

The costs of a multipurpose reactor were analyzed separately from

the tritium supply alternatives. The Department's Fissile Materials

Disposition Office and an independent contractor prepared separate

estimates. Different discount rates were used in the reports, which

also only identified the minimum and maximum cost range. The results of

the independent analysis, in discounted 1995 dollars are: (1) $4.5

billion to $14 billion for a government-owned large ALWR, $2.9 billion

to 8.6 billion for a small ALWR, and $2.7 billion to $9.9 billion for a

commercial reactor option; (2) $5.2 billion to $25.4 billion for a

privatized large ALWR, $3.1 billion to $14 billion for a small ALWR,

and $1.9 billion to $11.3 billion for a commercial reactor option. The

result of the Department's analysis, in discounted 1993 dollars, is:

(1) For a

[[Page 63888]]

government-owned large ALWR costs would range from $1.5 billion to $3.5

billion, and 2) for a privately financed large ALWR costs would range

from $0.7 billion to $5.0 billion. These amounts include revenue from

electricity sales.

In summary, the purchase of irradiation services is the lowest cost

in all categories and has the lowest uncertainty. The other commercial

options have the lowest cost estimates for TLCC both with and without

revenues, and for TPC but with a higher degree of uncertainty. The APT,

small ALWR, and small advanced HWR make up a middle group with

approximately similar discounted mean costs for TLCC without revenue,

and TPC. The small ALWR and small Advanced HWR have smaller

uncertainties than the APT in both these categories. TLCC with revenue

shows the small ALWR to have a lower mean cost than the APT or the

small advanced HWR and adds the large ALWR to this middle group. The

large ALWR is in the higher mean cost group for TLCC without revenue

and for TPC, along with the MHTGRs and HWR, which also have higher

uncertainties. The O&M analysis shows that the purchase of irradiation

services has clearly the lowest mean cost, with all other alternatives

grouped together. The uncertainties for all the alternatives generally

have a substantial overlap in their cost distributions.

Evaluation of Site Alternatives

The five sites for new tritium supply and recycling facilities were

evaluated with respect to environmental impacts and cost. Two criteria

emerged as discriminators: (1) Ability to handle low-level radioactive

waste; and (2) cost. No siting analysis was needed for the commercial

reactor options, since they all currently exist, and any reactor

ultimately selected would have to undergo a separate NEPA review.

Numerous environmental impacts were examined in the Final PEIS. The

analysis either showed very small or no impacts, or the impacts did not

differentiate among sites including cancer risks from a severe

accident. Impact differences are primarily due to the differences in

the size of the population within 50 miles of the site. Because cancer

risk is low for all sites, it is not a discriminator between sites. The

cost estimates for site alternatives are published in the Technical

Reference Report.

The results of the evaluations are summarized in Table 3 and

described below.

Table 3.--Site Evaluation

----------------------------------------------------------------------------------------------------------------

Percent adjustment to

base cost site (INEL) due

Cost of adding non- to site differences

Criterion site Ability to dispose of evaporative cooling --------------------------

wastes on site (reactors only)a Operation &

Construction maintenance

(percent) (percent)

----------------------------------------------------------------------------------------------------------------

INEL............................... Yes.................... $86 to $208............ 0 0

NTS................................ Yes.................... 99 to 239.............. 5 15

ORR................................ Yes.................... 0...................... 5 0

PANTEX............................. No..................... 98 to 239.............. -10 15

SRS................................ Yes.................... 0...................... 0 10

----------------------------------------------------------------------------------------------------------------

a Mean discounted cost in millions of 1995 dollars, using a 4.9% annual discount rate.

1. Ability to Handle Wastes. As shown in column 2 of Table 3, with

the exception of Pantex, all sites can dispose of low level radioactive

waste on site. The wastes from Pantex would be shipped to an approved

off site low level radioactive waste disposal facility.

2. Cost. The results of the cost comparisons are shown in Table 3.

Cost differences among sites are determined by three major factors:

(1) The cost for the non-evaporative cooling system needed at sites

which do not have ample water availability (this does not apply to the

APT, which is not designed to use non-evaporative cooling),

(2) The percentage differential in construction costs (primarily

because of labor rates), and

(3) The percentage differential in operation and maintenance costs

(primarily because of labor and electricity rates).

The third column of Table 3 shows the range of additional costs due

to the need for non-evaporative dry cooling for reactors at INEL, NTS,

and Pantex. The high end of these costs would occur for the large ALWR.

The fourth and fifth columns of Table 3 show the percent increases

in cost of construction, and operation and maintenance over the least

expensive site (INEL). For construction, Pantex shows a decrease, SRS

shows no change, and NTS and ORR show small increases. Operation and

maintenance costs are higher at NTS and Pantex than INEL, with SRS

higher than INEL but less than NTS and Pantex. ORR shows the same cost

to INEL. These differences are fairly small compared to the large

uncertainties in the actual costs of the facilities.

Evaluation of Tritium Recycling Alternatives

If a new supply facility is chosen at INEL, NTS, ORR, or Pantex,

the alternatives are to build a new recycling facility collocated with

the supply facility or to upgrade the SRS facility. Constructing a new

tritium recycling facility (1.9 to 2.1 billion dollars) is more

expensive (between $500 million and $750 million) than upgrading

existing tritium recycling facilities (1.3 billion) at SRS. The

operational environmental impacts would be similar.

If a new supply facility is chosen at SRS or if a commercial

reactor option is chosen, upgrading the existing tritium recycling

facility is the only option considered, since building a new recycling

facility at another site is more expensive and has no other advantages.

Cumulative Impacts

Impacts from the siting, construction, and operation of new tritium

supply and recycling facilities would be cumulative with impacts from

existing and planned facilities and actions at the five candidate

sites. The consequences of each new tritium supply alternative and

recycling alternatives include the cumulative effect of tritium supply

and recycling impacts and impacts from existing, planned, and

reasonably foreseeable operations. Other more long-term impacts

associated with the Department's proposed Environmental

[[Page 63889]]

Management Program and the Storage and Disposition of Weapons-Usable

Fissile Materials Program are speculative at this time but could

increase or decrease cumulative impacts, depending on the decisions

resulting from the PEISs being prepared for these programs and the time

frame of site-specific projects. Information on potential waste

management activities at the candidate sites was included as

appropriate in the assessment of waste management impacts in the

Tritium Supply and Recycling PEIS.

The Storage and Disposition PEIS alternative of burning plutonium

in a reactor could result in increased cumulative impacts at the

candidate sites if this Record of Decision selected a new facility, and

the Record of Decision for the Storage and Disposition PEIS selected a

separate new reactor. The impacts of combining tritium production and

plutonium disposition in a single reactor, the multipurpose reactor,

were evaluated in the Tritium Supply and Recycle PEIS. Cumulative

impacts from constructing two separate reactors would approximately

double those presented for a single reactor in the Tritium Supply and

Recycling PEIS. Cumulative impacts from construction of a APT for

tritium production and a new reactor for plutonium disposition would be

represented by adding together the APT and ALWR or MHTGR impacts

evaluated in the Tritium Supply and Recycling PEIS. Cumulative impacts

would be minimized if tritium production and plutonium disposition were

to take place in a single reactor.

The Environmentally Preferable Alternative

The environmentally preferable alternative is the alternative that

would cause the least impact to the physical environment, and best

protect worker and public health.

With respect to all three decisions, the no action alternative is

the environmentally preferable alternative. Under the no action

alternative, tritium requirements to support the nuclear weapons

stockpile would continue to be met by recovering residual tritium from

weapons components, purifying it, and refilling weapons components.

These activities would be performed at the Savannah River Site, the

current location of this function. However, under the no action

alternative, the Department would not establish a new tritium supply

capability and the Department would not meet future stockpile

requirements of tritium. This would be contrary to the Department's

mission as specified by the Atomic Energy Act of 1954, as amended.

Thus, no action is not a reasonable alternative.

Of the alternatives that would satisfy the Department's mission,

the potential environmental impacts are generally small and, except for

the commercial reactor options to purchase an existing reactor or

irradiation services, the impacts are within the same range. The

Department considers the commercial reactor options of purchasing an

existing reactor or irradiation services to be the environmentally

preferred alternative.

Implementation of either of these options would result in certain

environmental impacts. The environmental impacts of construction

activities would be limited to any support facilities that would be

required. Operation of the commercial reactor options would have few

potential environmental impacts. No additional spent fuel over and

above what the reactor(s) would otherwise generate during their planned

lifetime would be generated, assuming that operating scenarios do not

change fuel cycles. If fuel cycles were changed, additional spent fuel

would be generated.

There are no environmental grounds for discrimination among sites

for the tritium supply alternatives. Therefore, the SRS is the

environmentally preferred site since impacts from upgrading tritium

recycling facilities are less than building new facilities at any of

the other sites. Resource areas where no major differences exist, or

where potential environmental impacts are small are: land resources,

air quality, water resources, geology and soils, biotic resources,

socioeconomics, and site infrastructure.

Comments on the PEIS and Related Documents

Several comments were received on the Final PEIS during the 30-day

period following the filing of the Final PEIS with the Environmental

Protection Agency (EPA). The EPA stated that all of its specific

comments on the Draft PEIS had been adequately addressed in the Final

PEIS. A vendor for one of the ALWRs commented that on the Final PEIS

did not adequately reflect the fact that the electricity-producing

reactor options have an environmental benefit. That is, construction of

such a reactor would offset the need to build and operate an equivalent

capacity of fossil-fueled power plants, whereas the accelerator would

have an additional environmental impact from a power plant needed to

provide electricity for operating the accelerator.

The Final PEIS assessed the environmental impacts associated with

providing power to the APT. Two methods were assessed: (1) Purchasing

electricity from regional power pool grids; and (2) building and

operating a dedicated power plant. If a new dedicated power supply were

constructed, impacts would occur to air resources, land use, soils,

biotic resources, and socioeconomics at the construction site.

Operation of a dedicated power supply, or increased electrical demand

on the power pool would result in increased impacts to air resources,

water resources, waste management systems, and local traffic. Impacts

to land use, soils, waste management systems, and biotic resources

could occur at the plant location and along the transportation system

supplying the coal or gas to the power plant. While these environmental

impacts were assessed, no decision regarding a preferred source of

power is appropriate at this time. If an accelerator were eventually

built, the site-specific NEPA review would more fully explore the

options of providing power to the accelerator, and the appropriate

decision would be made at that time. The environmental impacts that

could be avoided through the use of a multipurpose reactor are

discussed qualitatively in the Final PEIS for both the ALWR, MHTGR, and

commercial reactor alternatives. These impacts are presented as part of

the cumulative impacts discussion in the previous section.

Additional comments on the Technical Reference Report and cost

analysis were also received from the vendor for one of the ALWRs. The

vendor questioned the basis of the cost estimate and the judgments used

in developing the uncertainties related to schedule, production

assurance, and cost as presented in the Technical Reference Report. The

commentor presented a revised set of assumptions resulting in

modifications to the cost ranges for the large and small ALWRs, APT and

commercial reactor options. The Department does not agree with these

assumptions. However, if these assumptions were accepted

hypothetically, and applied consistently and appropriately to each of

the ALWR, APT, and commercial reactor options, the result would be to

increase the cost range of the purchase of irradiation services and

lower the cost ranges of all other light water alternatives. Thus,

there still would be significant overlap in the cost of these

alternatives, and there would be no effect on the decisions presented

in this Record of

[[Page 63890]]

Decision. The Department selected experts knowledgeable in schedule,

cost or production assurance for the assessment panels who did not

stand to gain from the results of the assessment. In addition, each

panel included experts knowledgeable in the different technologies and

the mean results of their combined judgments were used in the

uncertainty analysis.

The Department received on October 11, 1995, a Congressional

report: ``Getting On With Tritium Production: A Report to Speaker Newt

Gingrich''. The primary recommendation of the report is that the

Department base its selection of a tritium production source on two

objectives: Maximizing the assurance that tritium sources will be

available when needed and minimizing costs to the taxpayers. The

Department's acquisition strategy described in this Record of Decision

implements this recommendation of the Congressional report. Additional

recommendations related to insuring that the plutonium disposition

mission and the tritium production mission be reviewed for combining

efforts to save money, and the new reactor option must be evaluated to

the same level of detail as the commercial reactor options. The

responsibility for tritium production and fissile material disposition

rests with two separate offices in the Department, the preparation of

the Tritium Supply and Recycle PEIS and Technical Reference Report was

closely coordinated with the Office of Fissile Materials Disposition.

Therefore, the option of using a reactor in a multipurpose mode is

analyzed in these two documents and the factors relevant to decision

making are presented in this Record of Decision. Due to the rapid decay

of tritium, and the long lead time required to bring a new tritium

source on line, even supplies of tritium from retired weapons are not

sufficient to postpone the need for a tritium supply facility to the

point where decisions concerning technology and site selection can be

deferred. With regard to equal evaluation, the Department believes that

the analysis completed to date accomplishes this recommendation. Cost

considerations associated with the reactor alternatives point the

decision toward existing commercial reactors. Moreover, a new reactor

has no major schedule or production advantage over an existing reactor

that would justify incurring the additional cost and environmental

impacts associated with a new reactor.

A private group has recently suggested that it purchase the Fast

Flux Test Facility (FFTF) from the Department and that the Department

then contract with the private group to make tritium at that facility.

In the PEIS, the use of FFTF was considered and dismissed as a long-

term tritium supply option because the amount of tritium that it could

produce would only meet a percentage of the steady state tritium

requirements, and it was not reasonable to rely on operating the

facility far beyond the end of its design life. However, the Department

will evaluate the presentation made by the private group to determine

whether the operation of the FFTF might be able to play any role in

meeting future tritium requirements. If any changes are warranted to

this Record of Decision following that review, or further NEPA

documentation is required, the Department will take appropriate action.

Decision

The Department is making three simultaneous decisions regarding

tritium supply and recycling. First, the Department will pursue a dual

track on the two most promising tritium supply alternatives: (1) To

initiate the purchase of an existing commercial reactor (operating or

partially complete) or irradiation services with an option to purchase

the reactor for conversion to a defense facility; and (2) to design,

build, and test critical components of an accelerator system for

tritium production. Within a three-year period, the Department would

select one of the tracks to serve as the primary source of tritium. The

other alternative, if feasible, would be developed as a back-up tritium

source. Second, the Savannah River Site is selected as the location for

an accelerator, should one be built. Third, the tritium recycling

facilities at Savannah River Site will be upgraded and consolidated to

support both of the dual track options. A tritium extraction facility

will be constructed at Savannah River Site. The basis for these

decisions is as follows.

Tritium Supply Decision: The options of the commercial reactor

alternative are the best in terms of schedule, production assurance and

cost. However, there are institutional issues with these options that

must be resolved, or else the alternative can only be used as a

contingency.

Institutional issues regarding the use of a commercial reactor(s)

must be resolved. Since commercial reactors are already constructed and

operating, adding the tritium mission to an existing reactor does not

significantly increase any existing environmental impact. Using

existing commercial reactors offers the least expensive approach. The

purchase of irradiation services presents the lowest cost and has the

lowest uncertainty. The purchase of an existing or partially completed

commercial reactor has the lowest capital and life cycle costs but a

greater degree of uncertainty than the purchase of irradiation

services.

Among the new facility alternatives, the accelerator has the

highest probability to meet earlier production requirements because of

less regulatory uncertainty. Among the new facility alternatives, the

accelerator also has the least environmental impact because it does not

use fissile material, generates no high-level wastes, and while the

risk from a severe accident is very small for all of the alternatives,

the risk for the accelerator is the smallest. While all of the

components of the accelerator have been proven, the entire system needs

to be demonstrated to assure the components work together as a complete

system. From a cost perspective, the APT is grouped with the small ALWR

and small advanced HWR in a middle range of costs if revenue is not

taken into consideration. There is significant overlap among the

alternatives, however. The two reactor alternatives have a smaller

uncertainty than the APT. If revenue is included, the small ALWR has a

lower mean cost than the APT and small advanced HWR. Also the large

ALWR is added to this middle group. The Department has confidence that

as we optimize the accelerator design over the next several years, the

resulting costs will fall within the lower end of the cost range

presented in the Technical Reference Report.

Based on the these considerations, the Department will implement a

dual acquisition strategy that assures tritium production for the

nuclear stockpile rapidly, cost-effectively, and safely. This dual-

track strategy for meeting tritium supply requirements provides the

following advantages:

Resolves major uncertainties over the next three years,

before selection of the primary alternative;

Selects the new facility that has the lowest estimated

environmental impacts, an accelerator, and the environmentally

preferred alternative, purchase of a existing commercial reactor or

irradiation services;

Lessens programmatic risk because it: 1) pursues two

technically different and independent approaches which provide fall

back in the event either approach develops significant problems; 2)

provides proven independent capability to increase production; 3)

develops and protects contingency capability to support requirements in

the event of a national emergency; 4) selects a strategy that has the

greatest

[[Page 63891]]

flexibility to meet production requirements earlier than 2011, if

necessary, and 5) includes the least cost option (irradiation

services); and

Preserves an option for simultaneous reactor ``burning''

of excess weapons plutonium, if the Storage and Disposition of

Weapons--Usable Fissile Materials Record of Decision selects reactor

burning of that material.

Site Decision: For the commercial options, the potential sites are

where existing facilities are located. Selection will be subject to a

separate NEPA analysis. For the APT, environmental impacts and costs

are not significant discriminators. The Savannah River Site will be the

site for the APT, if one is constructed, because it has the only

existing tritium recycling capability and infrastructure of the

candidate sites.

Tritium Recycling Decision: Upgrading and consolidating the tritium

recycling facilities at the Savannah River Site is the least expensive

option and avoids additional transportation of tritium between sites if

the APT is constructed. Therefore, if the APT is the primary source of

tritium, the existing tritium recycling facilities at Savannah River

Site will be consolidated and upgraded. If one of the commercial

reactor options becomes the primary source of tritium, the existing

recycling facilities at Savannah River Site will be consolidated and

upgraded, and a new extraction facility will be constructed.

Issued in Washington D.C., December 5, 1995.

Hazel R. O'Leary,

Secretary.

[FR Doc. 95-30238 Filed 12-11-95; 8:45 am]

BILLING CODE 6450-01-P