Order Code RL30540
CRS Report for Congress
Received through the CRS Web
The National Ignition Facility:
Management, Technical, and Other Issues
Updated November 8, 2001
Richard Rowberg
Senior Specialist in Science and Technology
Resources, Science, and Industry Division
Congressional Research Service ˜
The Library of Congress
The National Ignition Facility: Management, Technical,
and Other Issues
Summary
The U.S. Department of Energy (DOE) is building a very large laser facility to
provide scientific support for its program to maintain the safety, reliability, and
performance levels of the nation’s stockpile of nuclear weapons in the absence of
nuclear testing. This National Ignition Facility (NIF), which is designed to simulate
the behavior of nuclear explosions, is under construction at DOE’s Lawrence
Livermore National Laboratory (LLNL).
In August 1999, however, DOE announced that the project had run into some
serious problems that would likely result in a significant increase in its cost estimate.
The root cause of this problem was determined to be inadequate experience and
capabilities in managing large, complex projects on the part of the LLNL NIF project
team. Correcting these problems would require obtaining outside expertise and
management that would significantly increase the project’s cost. In addition, a
review by the NIF Council of LLNL reported that substantial technical issues
remained unresolved, although LLNL is confident that they can be resolved. In some
cases, however, solutions do not appear to be obvious.
In its revised project baseline and schedule, DOE now estimates the project cost
to be $3.45 billion, 60% above the original estimate, and its completion date as 2006.
A GAO report issued in August 2000 estimated the cost of the project at $3.89
billion. In response to congressional direction, DOE submitted a letter to Congress
on April 5, 2001, certifying that NIF should be built to the full 192 beams and that the
September cost estimate and schedule would be met.
On June 1, 2001, GAO released a second report stating that the three DOE
weapons labs do not agree on the value of NIF, as currently configured, for the
stockpile stewardship program. GAO also noted that DOE has not set up an external,
independent review of the project and does not intend to do so.
For FY2002, DOE requested $245 million for NIF construction. Congress
(H.R.2311 and S.1171, H.Rept.107-258) appropriated the full FY2002 request for
NIF, adding $7 million to the DOE inertial confinement fusion (ICF) program for
NIF-related activities. The House (H.R.2586) and Senate (S.1438) each authorized
$245 million for NIF construction, adding $10 million to the DOE ICF program for
NIF-related activities.
In addition to the management and technology concerns, the NIF project raises
other issues, including: the effectiveness of NIF in meeting the goals of the DOE
program to maintain the nuclear weapons stockpile; the likelihood that NIF will reach
its principal scientific goals; the potential effect NIF may have on nuclear
nonproliferation; and uncertainty about NIF’s contribution to the advancement of
inertial fusion energy research. The management and technical issues, and the role of
NIF in the stockpile stewardship program appear to be of particular concern.
Contents
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Stockpile Stewardship . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Technical Description and Applications of NIF . . . . . . . . . . . . . . . . . . . . . . 2
NIF Project Schedule and Cost Overruns – Chronology . . . . . . . . . . . . . . . 3
Congressional Actions – 106th Congress . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Congressional Actions – 107th Congress . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Project Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Technical Concerns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Contribution to Stockpile Stewardship . . . . . . . . . . . . . . . . . . . . . . . 19
NIF and Ignition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
NIF and Nonproliferation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
NIF and Inertial Fusion Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Outlook for the 107th Congress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
The National Ignition Facility: Management,
Technical, and Other Issues
Introduction
As part of the U.S. Department of Energy’s (DOE) program to maintain the
nation’s nuclear weapon stockpile, the DOE is building the National Ignition Facility
(NIF), which is planned to be the largest laser ever built. DOE anticipates that NIF
could achieve conditions of temperature and pressure in a laboratory setting that have
only been reached in explosions of thermonuclear weapons and in the stars. It is
hoped that NIF would be capable of achieving ignition of thermonuclear fusion of
deuterium and tritium — isotopes or forms of hydrogen — for the first time in the
laboratory.
NIF is to be a principal tool of the nation’s Stockpile Stewardship Program
(SSP) whose goal is to maintain the safety, reliability, and performance level of the
nation’s nuclear weapons in the absence of nuclear testing. It is one of several
experimental facilities being built for that program and is expected to provide
important contributions to the goals of stockpile stewardship and to contribute to the
advancement of inertial fusion energy (IFE) and other civilian scientific research
efforts. Recent problems with the construction of the facility, however, are likely to
result in a significant increase in the project’s cost. Also, technical problems remain
unsolved that might compromise the ability of NIF to reach its goals. In addition to
having implications for DOE appropriations and authorization, problems with NIF
also could increase the difficulty of the Administration’s continuing efforts to gain
ratification of the Comprehensive Test Ban Treaty.
Stockpile Stewardship
Largely as a result of the end of the Cold War, the United States has initiated
several steps designed to reduce the threat of nuclear weapons. On August 11, 1995,
the President announced that the United States would pursue a zero-yield
Comprehensive Test Ban Treaty (CTBT), which would institute a worldwide ban of
nuclear weapons tests of any yield.1 The treaty was signed by the United States at the
United Nations in September 1996. At the same time, the nation is not abandoning
I For a discussion of the CTBT see, Congressional Research Service.
Nuclear Weapons
Testing and Negotiation of a Comprehensive Test Ban Treaty. CRS Issue Brief 92099.
Regularly updated. On October 13, 1999, the U.S. Senate voted against ratification of the
CTBT. While the Russian Duma recently (April 2000) ratified START II, other related
actions must be taken by the U.S. Senate before it can enter into force. See, Congressional
Research Service,
START II Debate in the Russian Duma: Issues and Prospects, by Amy
Woolf, CRS Report RL97-359, updated April 17, 2000.
CRS-2
its reliance on a nuclear deterrent as a key element of its national security policy.
Under the terms of the Strategic Arms Reduction Treaty II (START II), which has
been ratified by both the United States and Russia, the United States would retain a
stockpile of about 3,500 weapons if and when the treaty is fully implemented.2 The
Administration and Congress have stated that a capability must be in place to maintain
the safety and reliability of the existing stockpile.3
A program to carry out this mandate, the Stockpile Stewardship and Program,
has been instituted by DOE.4 Because the SSP cannot rely on nuclear testing, a
surrogate testing process based on laboratory experiments and computer simulation
is being developed by the SSP, the research portion of the SSP. The goals of the SSP
include enhanced surveillance of the aging weapons stockpile to detect aging-related
defects, an improved capability to predict when defects will occur, increased
understanding of the physics of nuclear weapons, and better computer simulation of
nuclear explosions to assess the potential effects of defects. In addition, the program
will explore means to simulate the testing of nuclear explosion effects. Finally, the
SSP will attempt to help preserve the core expertise in weapons science and
technology now in place at the DOE weapons laboratories.
Technical Description and Applications of NIF
An important component of the SSP is to be the National Ignition Facility (NIF),
which will house the largest laser ever built. If completed, NIF will be capable of
delivering light from an assembly of 192 laser beams at an energy level of 1.8 million
joules (MJ) to a very small target.5 The energy will be delivered in a few billionths of
a second, and the total power delivered on target will be about 500 trillion watts. The
ultimate target is a tiny pellet of deuterium and tritium, two isotopes or forms of the
element hydrogen. The laser beams, called drivers, are configured to compress this
pellet to very high temperatures and pressures, forcing the deuterium and tritium to
undergo nuclear fusion, releasing large amounts of energy. If all goes as planned,
fusion ignition will occur, whereby the nuclear fusion reactions would release more
energy than provided by the laser to cause the reactions in the first place.
NIF is designed to achieve, on a laboratory scale, conditions similar to those in
a nuclear weapon explosion. It will be the only experimental facility that would be
capable of attaining fusion ignition conditions. The pressure and temperature reached
in an NIF laser-driven capsule that reaches ignition is projected to be similar to that
2 U.S. Department of Energy,
Draft Programmatic Environmental Impact Statement for
Stockpile Stewardship and Management, DOE/EIS-0236, February 1996, p. S-10.
3 President William Clinton, “Statement on a Comprehensive Nuclear Weapons Test Ban,”
August 11, 1995,
Weekly Compilation of Presidential Documents, August 14, 1995.
Washington, U.S. Govt. Print. Off., p. 1433; and P.L. 103-160, Sec. 3138.
4 U.S. Department of Energy, Office of Defense Programs,
The Stockpile Stewardship and
Management Program: Maintaining Confidence in the Safety and Reliability of the
Enduring U.S. Nuclear Weapon Stockpile, May 1995, p. 3.
5 The largest laser was the NOVA facility at Lawrence Livermore National Laboratory, which
was capable of delivering 40 thousand joules. It was recently decommissioned.
CRS-3
in a nuclear weapon, although the total energy output would be at least a million times
less than that of a weapon. The principal goals of NIF are to improve understanding
of the physics of specific aspects of weapon behavior, to validate improved computer
simulation codes,6 to evaluate the impact of defects discovered in the stockpile, to
simulate the effects of nuclear explosions on systems, and to help maintain the core
of “intellectual and technical competency” of the U.S. nuclear weapons program. NIF
is also expected to have several civilian scientific applications. DOE hopes that NIF
will be able to demonstrate fusion ignition as an important step toward harnessing
fusion energy for the production of electric power (inertial fusion energy or IFE).
NIF Project Schedule and Cost Overruns – Chronology
When the NIF project first appeared in the DOE budget request for FY1996, its
total project cost (TPC) — construction plus associated project costs — was reported
as $1.074 billion including construction and design costs (total estimated costs or
TEC) of $842.6 million. A completion date of the third quarter, 2002 was also given.
In the FY1998 budget request, DOE reported the total project cost of NIF to be an
estimated $1.196 billion including construction and design costs of $1.05 billion and
$146 million in associated project costs. The increase was a result of DOE’s decision
to stretch the project out by one year to a completion date of the third quarter, 2003.
These figures remained unchanged through the FY2000 budget request, and no
mention was made of any possible difficulties in meeting those targets. A hint of
some problems, however, could be seen in the detailed project description
accompanying the FY2000 request. DOE reported that total construction and design
cost estimates increased over the previous DOE estimate, but those changes were
compensated by a decrease in contingency funds from 12.6% of the total estimated
costs (the original contingency factor was 15%) to 7.6% of the total estimate. At this
point, the project was less than 50% complete, and the size of the remaining
contingency funds appeared to be quite low for a project this complex.
In March 1999, a preliminary assessment by NIF project officials at Lawrence
Livermore National Laboratory (LLNL) determined that there were major potential
problems with NIF construction that threatened to increase substantially project
costs.7 LLNL staff attempted over the next several months to see if the additional
cost could be accommodated within total project costs already approved, but did not
notify DOE until June 1999, when it reported these problems to NIF project staff at
DOE Headquarters. It was not until August 1999, however, that the Secretary of
Energy was informed. At that point, LLNL reported that the project cost would
increase by about 20–30% and completion would be delayed by 12 to 18 months.
LLNL stated that the cause of the cost increase was that assembly of the laser system
6 As weapon codes are changed, they need to be validated to ensure that they are accurately
modeling weapon processes. In the past, nuclear weapons tests were the primary means of
accomplishing this validation. Without tests, the codes will have to be checked by seeing how
well they describe the phenomena that occur during operation of the stockpile stewardship
experimental facilities such as NIF.
7U.S. Department of Energy,
Energy Department Announces Major Changes to Improve
National Ignition Facility Management and Accountability, News Release, March 24, 2000,
[http://www.doe.gov/news/releases00/marpr/pr00084.htm]
CRS-4
infrastructure was considerably more complicated than anticipated.8 An important
contributor to this added complexity was the difficulty in maintaining NIF’s stringent
cleanliness requirements during assembly. Officials at LLNL determined that the lab
did not have the capability to meet these more complex assembly requirements and
must contract for them. According to LLNL, securing those services through the
contract process was the primary source of the added cost and time to completion.
Upon learning of the potential cost overruns, the Secretary of Energy requested
an internal review of the project to determine the source of the problems. A special
task force of the Secretary of Energy Advisory Board (SEAB) was set up to
investigate management and technical problems at the project. In addition, the
University of California (UC), which manages LLNL under contract to DOE, also
initiated a review by a special committee of the UC President’s Council. The Council
was set up to advise the UC President on matters about the three DOE labs managed
by UC — LLNL, Los Alamos National Laboratory, and Lawrence Berkeley National
Laboratory. A third study focusing on technical issues was carried out by the
Technology Resource Group of the NIF Council, an advisory group internal to
LLNL.
The SEAB9 and UC10 reviews both confirmed that the assembly and installation
of the NIF laser system would be more difficult and rigorous, and would require a
cleaner environment than originally anticipated. They both concluded that the original
approach of having the assembly done by an operating contractor and LLNL
personnel while relying extensively on LLNL management would not succeed.
Rather, an outside contractor with experience in clean assembly of large, complex
systems and management of large engineering projects would be needed.11 As a
consequence, a significant increase in cost and time-to-completion would result.12
The NIF technical review found a number of potential technical issues that if not
resolved could seriously compromise the operation of NIF.13 Foremost is the
possibility that the optics facing the laser target — the final optics assembly — would
not be able to handle the laser energy intensities needed to achieve ignition. Both the
UC and SEAB reviews agreed with that assessment.
8Personal communication, David Crandall, United States Department of Energy, Oct. 20,
1999.
9Secretary of Energy Advisory Board, U.S. Department of Energy,
Interim Report of the
National Ignition Facility Laser system Task Force, January 10, 2000. (Hereafter called the
SEAB Laser Task Force.)
10University of California,
Report of the University of California President’s Council
National Ignition Facility (NIF) Review Committee, November 18, 1999,
[http://labs.ucop.edu/internet/nr/nr112399.html]. (Hereafter called the UC President’s
Council.)
11UC President’s Council, 3.
12SEAB Laser System Task Force, 3.
13Lawrence Livermore National Laboratory, Technology Resource Group of the NIF Council,
NIF Technology Review, Nov. 4, 1999. [http://www.nrdc.org/fpprog.html]
CRS-5
In response to the management problems uncovered by the SEAB and UC
reviews, the Secretary of Energy in March 2000, announced several changes in the
project structure: an Associate Director of NIF reporting to the LLNL director was
established; UC established a special panel of the President’s Council charged with
oversight of the NIF project; a headquarters NIF project office was established; the
NIF project was made part of the DOE Project Management and Oversight function
— mandated by Congress — that will provide an on-site DOE contractor with
expertise in large, complex project management; and important decisions about NIF
had to be approved by the DOE Deputy Secretary before funding is provided.14 In
making this announcement, the Secretary reiterated his belief that while the NIF
project has had serious problems with its management, the science underlying NIF
“remains sound.”15
On May 3, 2000, the Secretary of Energy announced a new budget and schedule
for NIF.16 He stated that DOE would increase funding for NIF by $95 million for
FY2001 to a total of $169.1 million for construction. In addition, the NIF funding
requirements would increase from $100 million to $150 million for FY2002
depending on the results of the new baseline, then scheduled for completion in August
2000. DOE did not ask for additional funding for FY2001, but proposed shifting
funds from other sources within the department. It was further stated at the press
conference accompanying the DOE announcement that most of the funds would come
from LLNL.17 The new schedule announced by the Secretary would have the first
batch of laser beams in operation by 2004 and the entire facility completed by 2008,
nearly five years past the original date. That completion date would be well beyond
the first estimates of the delay cited above.
On June 1, 2000, DOE delivered an interim report to Congress containing a new
plan for NIF with a revised cost estimate and schedule.18 In this plan, NIF would have
a total project cost (TPC) of $2.12 billion, would provide first laser light at the end
of FY2004, and would reach the full 192 -beam facility late in FY2008. Cost
increments would be $95 million for FY2001 and $150 million for FY2002 over the
original plan ($74.1 million for FY2001 and $65 million for FY2002). In addition to
the TPC, DOE also included costs that are directly related to NIF that had not
previously been counted as part of the NIF project but had been included in the
Inertial Confinement Fusion (ICF) program. These costs, estimated to be $833.1
million at the time of the FY1998 baseline, were now estimated to be $1.137 billion.
Therefore, the total NIF cost estimate in the June 1, 2000 document were $3.26
billion compared to the original baseline estimate of $2.03 billion, a cost overrun of
$1.23 billion or about 60%.
14U.S. DOE March 24, 2000, News Release.
15U.S. DOE March 24, 2000, News Release.
16U.S. Department of Energy,
Secretary Richardson Settles on a New Budget Schedule for
National Ignition Facility, News Release, May 3, 2000.
17Tarun Reddy, “New NIF plan calls for delay in facility’s opening, more funding,”
Inside
Energy/ with Federal Lands, McGraw-Hill, May 8, 2000, 3.
18Letters to Senator Pete V. Domenici and Representative Ron Packard from Bill Richardson,
Secretary of Energy, June 1, 2000. [http://www.dp.doe.gov/dp_web/news_f.htm]
CRS-6
As the basis for this interim report, on May 30, 2000, the Director of LLNL
delivered a letter to the Secretary of Energy providing a certified estimate for a
revised cost and schedule baseline for NIF.19 The letter noted that LLNL as directed
by DOE had developed a detailed plan for completing the project as rapidly as
possible. Under this plan, the first laser light ( 8 beams) would be obtained at the end
of FY2003 and the full 192 beams would be operational at the end of FY2006. In
each case, two years later than originally planned. The total project cost of this new
plan was estimated at $1.95 billion. Funding increases of $150 million for FY2001
and $240 million for FY2002 over the original plan would be required.
The LLNL Director noted, however, that such increases would likely have a
“negative impact” on the rest of the stockpile stewardship program. As a result, he
provided other estimates that would not require such large initial funding increases.
Those estimates, however, were not developed in a detailed, bottoms-up manner, as
was the plan noted above, but were determined by estimating the effects of stretching
out that plan. When this stretched-out plan was presented to the NIF Program
Review Committee (NPRC), a group formed as part of the management reforms
instituted by DOE, the Committee expressed its dissatisfaction with the LLNL cost-
estimate methodology. Further it was concerned about how well the stretched-out
plan would meet the requirements of the stockpile stewardship program (SSP) to
certify the readiness of the nation’s nuclear stockpile. As a result, LLNL created a
new plan that meets the funding increase limits imposed by DOE for the next two
years (as reported by the Secretary of Energy in March 2000) and would meet the
SSP certification requirements. This plan was the basis of the June 1, 2000 interim
report.
Because the June 1, 2000 estimate had not undergone a detailed, bottoms-up
review, DOE did not consider that it was adequate to fulfill the congressional
requirement for a revised baseline. This review was completed in August 2000, and
DOE delivered the revised baseline to Congress on September 14, 2000.20 The new
cost estimate is $2.248 billion for the TPC and $1.2 billion for associated program
activities giving a total of $3.448 billion or $1.42 billion above the original baseline.
Of this amount, $1.46 billion will have been spent by the end of FY2000. According
to the new baseline, first light (eight of the planned 192 beams) would be achieved in
FY2004 and the project would be completed by FY2008.
DOE also submitted to Congress on June 27, 2000, an FY2001 budget
amendment requesting budget authority for FY2001 for an additional $95 million for
NIF to come from other Stockpile Stewardship programs and $40 million to be
reallocated from the FY2001 Readiness in Technical Base and Facilities originally
planned for NIF operational support. This amendment, which would have added
$135 million in budget authority to the original FY2001 request of $74 million, was
rejected by the House and Senate authorization and appropriations committees.
DOE, however, reiterated the request in the September 14, 2000 revised baseline. As
19Letter to the Secretary of Energy from the Dr. Bruce Tarter, Director, Lawrence Livermore
National Laboratory, May 30, 2000.[http://www.dp.doe.gov/dp_web/news_f.htm]
20Letters to Senator Pete V. Domenici and Representative Ron Packard from Bill Richardson,
Secretary of Energy, September 14, 2000. [http://www.dp.doe.gov/dp_web/news_f.htm]
CRS-7
noted below, most of this request was approved by the House-Senate conference on
the FY2001 DOE appropriations bill, although with significant provisos.
On April 6, 2001, the NNSA of DOE delivered a report to Congress responding
to the certification request contained in the FY2001 Energy and Water Development
Appropriations Act (see below for details).21 DOE was required to submit this report
in order to receive the remainder of the FY2001 appropriation for NIF. In the letter
accompanying the report, the Administrator of the NNSA recommended that the NIF
proceed with the full complement of 192 beams. He also certified that the project was
on schedule and budget as set forth in the revised baseline. The Administrator stated
that a Workshop had been held to consider alternatives to a 192-beam facility and had
concluded that full-scale NIF was the best course for meeting the Stockpile
Stewardship program requirements. The Workshop concluded that any construction
path that paused the project at fewer than 192 beams and then completed the project
would cost more than proceeding directly to the full 192 beams. The Administrator
also noted that a five-year plan for the SSP is currently under development and will
be submitted to Congress after completion of the President’s national security
strategic review.
The Administrator also concluded that the High-Energy-Density Physics (HEDP)
program within the SSP should not only include a 192-beam NIF, but also the Omega
laser at the University of Rochester and the Z-machine at Sandia National Laboratory.
As for the latter, the Administrator stated that the proposed refurbishment of that
facility was promising but that it would not be a replacement for NIF. Finally, the
Administrator emphasized the importance of NIF in attracting and training new
scientific and technical personnel to the HEDP program.
In a related action, the Secretary of Energy announced on April 5, 2001, that
DOE would not make any changes in the design, construction, or operation of NIF
as a result of a supplemental environmental impact statement (SEIS).22 The SEIS was
carried out as a result of the uncovering of capacitors containing PCBs during
excavation for the NIF.
Congressional Actions – 106th Congress
In September 1999, the House Science Committee requested an investigation of
the NIF project by the General Accounting Office (GAO). The GAO review, released
in August 2000, concluded that
“NIF’s cost increases and schedule delays were caused by poor Laboratory
management, which included weaknesses in planning, budgeting, and project
21Letter from John A. Gordon, Administrator of NNSA to The Honorable Dennis J. Hastart,
Speaker of the House of Representatives, April 6, 2001. This letter and the accompanying
documents were obtained at [http://www.dp.doe.gov/dp_web/news_f.htm]. At this time,
however, the document is no longer on that website.
22U.S. Department of Energy, National Nuclear Security Administration, “Record of Decision
for the Final Supplemental Environmental Impact Statement for the National Ignition
Facility,”
Federal Register, 60, No. 66, (April 5, 2001) 18078.
CRS-8
control. DOE’s oversight weaknesses contributed to cost and schedule
problems, and coupled with no effective independent review of NIF, allowed
NIF problems to go undetected.”23
The GAO report also noted that other factors contributed to the cost overruns and
delays, including initial budgets that did not reflect project complexity and
performance of major project construction in parallel with critical R&D.
In its report, GAO estimated total costs for the NIF project to be $4.08 billion.24
In addition to the TPC and costs for the associated program activities included by
DOE in its revised baseline, GAO includes $627 million for target-physics R&D and
NIF support supplied by other DOE labs that are not included in the DOE revision.
DOE disputes the inclusion of these costs arguing that they are part of the general
inertial confinement fusion research program supported by DOE and support
experiments in all of its ICF facilities. GAO claims that the target-physics work is
essential to NIF and should be counted. No estimate of these disputed costs for the
original baseline is given either by DOE or GAO. Assuming they would have been
the same, the original baseline estimate would have been $2.66 billion, leading to a
cost overrun, as accounted for by GAO, of 53.4%. GAO did not attempt to make an
independent assessment of the construction and supporting R&D cost estimates that
DOE did acknowledge.
On October 30, 2000, the President signed the Floyd D. Spence National
Defense Authorization Act for FY2001 (P.L. 106-398, H.Rept. 106-945). The Act
authorizes $209.1 million for NIF, $135 million above the original request. Included
in the funds are $40 million to be transferred from the Inertial Confinement Fusion
program. The Congress also included a provision in the Act requiring a report from
the NNSA with a revised baseline cost and schedule estimate. The Act prohibited
more than 50% of the authorized funds from being spent until this report was
delivered. The Act also required the General Accounting Office to carry out a
detailed review of the NIF program focusing on NIF’s role in ensuring stockpile
reliability and safety, its relationship with the rest of the stockpile stewardship
program, and the possible effects further delays in the NIF project could have on the
stockpile stewardship program. In addition, the GAO was to review the revised NIF
baseline report. Congress expressed its continuing disappointment in the management
and technical difficulties of the project, but also stated that it continues to believe in
the importance of the project and that recent management improvements justify the
funding increase.
On September 28 and October 3, 2000, the House and Senate respectively
approved the Energy and Water Development Appropriations Act conference report
23United States General Accounting Office, Report to the Committee on Science, House of
Representatives,
National Ignition Facility: Management and Oversight Failures Caused
Major Cost Overruns and Schedule Delays, GAO/RCED-00-141, August 2000, p.13.
24GAO,
National Ignition Facility, p. 10. The total reported by GAO was $3.89 billion.
That figure was based on an earlier DOE estimate for the TPC and associated costs. If the
DOE estimate for these costs given in the revised baseline are used, the GAO estimate would
increase to $4.08 million.
CRS-9
(H.Rept. 106-907). The bill was signed by the President on October 27, 2000 (P.L.
106-377, H.Rept. 106-988). The Act provided an appropriations of $199.1 million
for NIF for FY2001.25 Of this amount, $25 million was to come from a reduction in
other weapons activities funds allocated to LLNL, $40 million from transfer of NIF
operating funds to NIF construction, and $60 million in new appropriations.
Congress also included language in the bill stating that only $130 million of the
appropriation would be available at the start of FY2001 with the remaining funds
becoming available after March 1, 2001, upon certification by NNSA that several
requirements had been met. (See above for a discussion of that certification report.)
Among those requirements are a recommendation from NNSA on the path the project
should take based on an assessment of options including a smaller facility (48 or 96
beams); certification that the project is meeting its construction milestones and is on
budget and schedule; completion of a study as to whether a full-scale NIF is needed
to meet SSP goals; and a five-year budget plan for the SSP including how NIF will
be paid for in the out years. Congress expressed its belief that NNSA had not
adequately reviewed all options for NIF, particularly an assessment of a smaller
facility. Congress also directed the NNSA to “fully reflect a balanced set of programs
and investments within the stockpile stewardship program,” in its FY2002 budget
request and to ensure that the $3.4 billion NIF can be accommodated within these
programs.
Congressional Actions – 107th Congress
The DOE budget submission for FY2002 requested $245.0 million for NIF
construction, 24.2% above the FY2001 level. In addition, DOE requested $222.9
million for the remainder of the inertial confinement fusion program, much of which
is directly tied to the NIF project. According to the budget submission, the request
is consistent with the revised baseline described above. No changes were reported in
the FY2002 budget submission in total cost, scope, or schedule of the project from
the revised baseline.
On June 1, 2001, the GAO released its follow-up report as called for by the
FY2001 defense authorization act (see above).26 GAO concluded that the role of NIF
in the SSP was uncertain because the three weapons laboratories were not in
agreement on the contribution NIF would be able to make. In addition, GAO noted
that DOE has not yet certified that completion of NIF at its full design level will have
no negative consequences for the rest of the SSP. GAO further stated that any more
delays in completing NIF could affect the weapons’ research program primarily by
adversely affecting the programs ability to attract high quality scientific talent. GAO
also noted that DOE has not set up an independent external review process for NIF
and does not intend to do so.
25For a discussion of the entire DOE R&D budget request for FY2001, see, Congressional
Research Service,
Department of Energy Research and Development Budget for FY2001:
Description and Analysis, by Richard Rowberg, CRS Report RL30445, Jan. 5, 2001.
26U.S. General Accounting Office,
Department of Energy: Follow-Up Review of the National
Ignition Facility, GAO-01-677R, June 1, 2001.
CRS-10
On June 28, 2001, the House approved funding the full request for NIF
construction for FY2002 (H.R. 2311, H.Rept. 107-112). In addition, the House
appropriated $492.9 million for the ICF program, $25 million above the request. The
increase is to be used for R&D on high average power lasers. The House noted the
recent GAO report about NIF and cited some of the findings including the lack of an
outside, independent review group. The House stated its expectation that DOE
would address the GAO concerns. On July 19, 2001, the Senate approved $492.4
million for the DOE ICF program for FY2002 (S. 1171, S.Rept. 107-39). Included
was the full request for NIF construction, and $59.7 million for ICF/NIF experimental
support activities. The latter is $24.5 million above the request. Included in that
increase are $10 million to enhance NIF diagnostics and target development, and $7
million for the base program. In approving the request for NIF construction, the
Senate stated that despite the certification provided by DOE about the NIF cost and
schedule, that a “high level of vigilance and extraordinary management attention is
warranted” to ensure that the program remains on course. The Senate also noted its
agreement with the recent GAO study. In particular, the Senate expressed its concern
about the continuation of project oversight by the same people in that role before
project costs began to grow in 1999.
On November 1, 2001, both the House and Senate approved the conference
report (H.Rept.107-258) for the Energy and Water Development Appropriation Act,
2002. In that act, Congress approved the full request for NIF for FY2002. Congress
also provided and additional $7 million to the ICF program to enhance NIF
diagnostics and target development. There was no language in the conference report
about the cost overruns or other potential concerns about the project.
On September 25, 2001, the House passed its version of the National Defense
Authorization Act for Fiscal Year 2002 (H.R. 2586, H.Rept. 107-194). The bill
authorizes the entire request for NIF of $245.0 million for FY2002. In addition, the
House authorized an increase of $10 million for the remainder of the ICF program to
be used for NIF efforts to reduce risk and develop needed technology. The House
argued that there was a funding shortfall in that part of the NIF program. The House
also noted the improvements in NIF management and project oversight but expressed
concern about the remaining technical problems facing the project. It directed DOE
to inform it “expeditiously” of any future delays or cost escalation. The House noted
that NIF offers many basic and applied research opportunities beyond its weapons
mission, but directed DOE to ensure that the Stockpile Stewardship program be the
facility’s primary focus.
The Senate passed its version of the defense authorization act (S.1438,
S.Rept.107-62) on October 2, 2001. In that bill, the Senate authorized an
appropriation of $245.0 million for NIF construction, the requested amount. The
Senate also authorized an additional $10 million for the ICF program to support
enhancements of NIF diagnostics and target development. In addition, the Senate
directed the NNSA to report to the two Armed Services Committees when the NIF
project has reached its level 1 and level 2 milestones. Noting the importance of the
project to the stockpile stewardship program, the Senate directed these reports to
ensure that Congress is kept apprised of the project’s progress in meeting the revised
baseline. While expressing confidence that NIF’s budgetary and scheduling problems
have been solved, the Senate still intends that Congress keep close tab on the project.
CRS-11
Issues
Project Management. Until the reports of cost overruns in early September
1999, DOE officials were quite confident of cost and schedule estimates for NIF. As
recently as June 11, 1999, the Secretary of Energy publicly noted that the project was
on schedule and budget. As noted above, the project’s cost estimates were increased
from the original estimates by about 11.6% during the Title I design phase, but no
other changes were reported. The project had undergone several technical and
management reviews and confidence was high that the estimates would hold. Clearly,
that confidence was misplaced.
The SEAB, UC, and GAO reviews found the NIF project to suffer from severe
management problems. The principal source of the problems was LLNL’s
inexperience with managing large, complex engineering projects.27 The SEAB Task
Force concluded that the management structure in which the DOE weapons labs
(LANL, LLNL, and Sandia National Lab) operate encourages a self-reliant attitude
that inhibited LLNL’s willingness to seek help from outside for activities in which it
did not have adequate competence, in particular management of large engineering
systems.28 This conclusion was also arrived at by the UC review. A particularly
egregious example of this inexperience, according to all three reviews, was the setting
of the original contingency factor at 15%. They all argued that this was much too low
for a project this complex. The fact that the contingency level had dropped to less
than 8% with the project less than half done appears to corroborate this conclusion.
GAO noted that LLNL failed at the outset to appoint a project manager with the
“management and technical expertise” required of such a large and complex project.
It also stated that LLNL did not install a reporting system that could have effectively
noted technical and other issues that would put the project at risk.29
The SEAB and GAO reviews also concluded that all three participants in the
process — LLNL, DOE, and UC — must share in the blame for allowing this cost
overrun to occur. The UC report stated that the project was far more complex than
any ever attempted by the LLNL and required management capabilities it did not
possess. LLNL was not capable of identifying those management weaknesses, and
neither were DOE nor UC.30 The three assessments also noted that the project was
not adequately reviewed and monitored by either DOE or UC. The SEAB Task
Force stated that independent reviews are a necessity, a view concurred with by UC.
All three reviews pointed out the absence of effective communications throughout the
entire NIF project. The UC review suggested that success in solving earlier problems
on the project led to a reluctance to report problems to anyone outside the immediate
project staff. GAO concluded that DOE’s historic reliance on contractors was at the
27SEAB Laser System Task Force, 9.
28SEAB Laser System Task Force, 5.
29GAO,
National Ignition Facility, p. 15.
30UC President’s Council, 5.
CRS-12
“core” of its oversight weakness. GAO stated that DOE did not have staff with
“adequate management and technical skills” to provide effective oversight.31
The GAO report elaborated on the absence of effective DOE oversight.32 GAO
concluded that the monthly and quarterly reports prepared by LLNL did not indicate
the growth and extent of cost and schedule problems as they appeared. Further, the
chain of command between DOE and LLNL was confusing leaving ambiguous at best
questions of to whom laboratory officials should report. GAO also argued that DOE
did not act expeditiously when problems were first suspected.
DOE has taken many steps to implement the recommendations of the SEAB and
UC reviews. The announcements — noted above — that were made by the Secretary
of Energy indicate that substantial management changes have been made at both
LLNL and DOE. The May 30 LLNL letter also notes substantial changes in project
management.33 In particular, the letter says that the external review process has been
“completely revamped” including the establishment of the NIF Programs Review
Committee. The LLNL director also noted that every attempt is being made to
coordinate the activities of all of the new review groups. The committee established
to validate the revised baseline of the NIF project concluded that DOE has made the
management changes necessary to ensure successful completion of the project.34 It
stated that the “management approach is appropriate for the project with the size and
complexity of NIF” and “that the proposed NIF scope, cost, and schedule baselines
are reasonable.” The committee also agreed that the new contingency level of 25%
of remaining costs was reasonable. The Chairman of the SEAB review team also
concluded that the management concerns raised in its interim report have been
addressed and that he was “confident” that the responsible parties could perform their
tasks “properly.”35 It is apparent that changes have been made that suggest a shift
from science to engineering expertise has taken place within the project management
hierarchy.
It remains to be determined, of course, whether these changes will guide the
project to a successful completion within the new baseline. A particularly crucial
issue is the degree of outside, independent review. GAO noted that the “absence of
effective independent reviews” was a major reason that problems were not recognized
and dealt with in a more timely manner.36 As noted above, the latest GAO report
31GAO,
National Ignition Facility, p.16.
32GAO,
National Ignition Facility, p.16.
33LLNL, May 30, 2000, letter, 7.
34U.S. Department of Energy,
Department of Energy Rebaseline Validation Review of the
N a t i o n a l I g n i t i o n F a c i l i t y P r o j e c t , A u g u s t , 2 0 0 0 ,
[http://www.dp.doe.gov/dp_web/news_f.htm], ii.
35Letters to William B. Richardson, Secretary of Energy, from John P. McTague, Chairman,
S E A B N I F L a s e r S y s t e m T a s k F o r c e , S e p t e m b e r 7 , 2 0 0 0 .
[http://www.dp.doe.gov/dp_web/news_f.htm]
36GAO,
National Ignition Facility, p.18.
CRS-13
repeated the observation that DOE has not yet formed an external independent
review, and further noted that DOE does not intend to do so.
While new oversight panels have been established by UC, DOE, and LLNL it is
not clear how effective these reviews will be. In addition, while coordination among
these review groups is important to avoid duplication and minimize the burden on the
NIF project team, care must be taken to ensure the independence of the review
groups. Without an effective outside review mechanism, there is increased risk that
the project will continue to run into significant problems even with the new
management structure in place.
The revised baseline validation committee also pointed out a number of potential
problem areas.37 It concluded that a single NIF review process that satisfies NNSA,
UC, and LLNL was necessary to avoid excessive demands on the NIF project. The
committee also recommended strengthening the NIF project office within NNSA
headquarters. In addition, it argued for strong support from LLNL in terms of getting
and retaining the right personnel for the project. The committee also found that a
system for measuring progress and performance using earned-value techniques was
critical for the project and should be put in place. The committee also noted that the
separation of most R&D activities from project control was a significant contributor
to the problems encountered by NIF. While recognizing the steps taken to integrate
R&D into the project, the committee stated that the scope and cost of R&D necessary
to complete NIF needs to be identified and incorporated into the project’s schedule.
In connection with this last point, the June 2001 GAO report noted that about
$700 million of R&D that is supporting the NIF project is not managed by the project
and is not included in the monthly reports issued by NIF.38 While the report notes that
DOE is currently reviewing how the integrate those activities within the NIF project.
A critical aspect of these management issues is the degree of involvement of
industry in the project. As noted, one of the major contributors to the cost overrun
was the substantial underestimate of how difficult it would be to assemble the 192
beamlines to the required degree of cleanliness. According to the May 30 LLNL
letter, the project, after consultation with the semiconductor industry, has taken two
steps to correct this problem.39 It is asking industry to provide a higher level of
component integration than originally planned. Component integration by laboratory
personnel will be greatly reduced. Also, installation of the mechanical and electrical
infrastructure for the beampath will be done by one contractor. These actions will
place much of the installation of the beam assembly in the hands of industry with more
experience in large, complex project management. In August 2000, DOE completed
these actions by awarding a $230 million contract to Jacobs Engineering for
integration management and installation.
37DOE, Rebaseline Validation, p.39.
38GAO,
Department of Energy, 4.
39LLNL, May 30, 2000, letter, 8.
CRS-14
Because of the unique nature of NIF, however, it is not clear whether these steps
will ensure successful construction of NIF within the revised budget. In particular,
care will be needed to ensure a close and cooperative working relationship between
industry with the project management expertise and LLNL staff with expertise in large
laser systems. The revised baseline evaluation committee also argued for close
management, communications, and monitoring by LLNL of the industrial contractors
obtained to perform beam hardware integration.40
Another important management issue is the rate the NIF should be completed
given the extent of the cost overruns. A number of options were considered by
LLNL at the direction of DOE.41 Two options were described in a review carried out
by the Target Physics Program Review Committee at the request of the NIF Program
Review Committee on the impact of NIF construction delays on the DOE SSP.42 The
first option called for the first laser bundles to come on line by the end of FY2003 and
final project completion by the end of FY2006. The other option (labeled option 3)
called for the first bundles by the end of FY2004 with final project completion by the
end of FY2009.
The May 30, 2000, letter from LLNL to DOE suggests that LLNL favored an
aggressive completion path described by the first option. As noted, that path would
have required an increase in FY2001 funding of $150 million and an increase of $240
million for FY2002.43 This option was also the one apparently favored by the Target
Physics Review Committee (TPRC).44 It argued that the first option would provide
the “earliest benefit” from NIF because it would allow operation of a partially
completed facility — 48 beams — well before the other options (the end of FY2004)
while the facility was being completed. The TPRC stated that operation of a 48-beam
cluster would be very beneficial for DOE’s SSP. The TPRC also expressed its belief
that if deployment of NIF took too long, the project might not be completed.
DOE has chosen a path about mid-way between the two options. As argued in
the May 30, 2000, LLNL letter to DOE, the large increases required by the first
option could have serious consequences for the rest of the SSP. Therefore, DOE has
chosen to limit those increases. Presumably, DOE hopes that its path will balance
budget concerns with the technical and programmatic concerns of the Target Physics
Review Committee. The DOE-selected path would reach the 48-beams bundle in
FY2006, about one and one-half years later than the LLNL choice, and result in a
completed NIF about one to 1.5 years after the LLNL choice. This path was
40DOE Rebaseline Validation, p. 20.
41U.S. Department of Energy, Memorandum for Dr. Bruce Tarter, Director, Lawrence
Livermore National Laboratory, Options for National Ignition Facility, March 9, 2000.
42U.S. Department of Energy, Lawrence Livermore National Laboratory,
Report of the
National Ignition Facility Target Physics Program Review Committee, Draft, April 13,
2000, 3. The NIF Program Review Committee is the successor to the NIF Council. The
Committee along with the Target Physics Program Review Committee are made up of
individuals outside the NIF program and, for the most part, outside of the DOE labs.
43LLNL, May 30, 2000, letter, 2.
44Target Physics Review Committee, 4.
CRS-15
confirmed in the revised baseline presented on September 15, 2000 and in the March
2001 certification report to Congress.
A final point concerns project milestones. One result of the management
problems has been the establishment of detailed milestones for the project including
a comprehensive work schedule. The June 2001 GAO report, however, noted that
none of the major milestones take place until after 2004.45 As a consequence, GAO
recommended that earlier milestones be set up in order to track more carefully the
progress of the project.
Technical Concerns. In addition to the management issues uncovered by the
reviews of the NIF program, several technical concerns were highlighted. These
included the high degree of cleanliness required during assembly of the laser system
infrastructure and optics, and some potentially serious limitations of key optical
components.
The NIF design has always incorporated the need to maintain stringent
cleanliness requirements during construction and operation. As noted above,
however, the degree of cleanliness required was greater than anticipated. As pointed
out by the SEAB Task Force, maintaining the necessary level of cleanliness while
assembling the laser infrastructure and optics is a major challenge. In addition, budget
constraints early in the project constrained the size of the laser building, limiting clean
air flow below the levels needed.46 Further, because the infrastructure is to be
completed before the first optical elements are installed, any cleanliness-related
lessons learned during installation cannot be used to modify the infrastructure.
As noted by LLNL and the SEAB Task Force, the size and cleanliness
requirements of NIF are similar to those of semiconductor fabrication, and the lab
plans to make use of that expertise. There are, however, some differences.47 Unlike
a semiconductor manufacturing facility, the NIF infrastructure will lack adequate air
flow, which, combined with the constrained height of the building, will make it
impossible for the laser bay to be a truly clean room. Also, unlike semiconductor
manufacturing, there can be no period of operation at less than optimal cleanliness
conditions because an NIF laser bundle cannot be operated with any particulate
contaminations. Also, nonvolatile organics — used to clean optical components prior
to installation — must be at a much lower concentration for NIF than for a
semiconductor plant. Therefore, it is not clear that using industrial expertise to
manage the installation of the laser assembly will by itself ensure that cleanliness
requirements will be met. If NIF had constructed an engineering prototype beam,
some of these problems might have been addressed earlier. Construction of such a
beam does not appear feasible at this point and it is likely that the first bundle will
have to serve in that capacity. The SEAB Task Force, however, stated that if
adequate care is taken and it is understood by NIF project management that assembly
45GAO,
Department of Energy, 4.
46SEAB Laser System Task Force, 15.
47SEAB Laser System Task Force, 16.
CRS-16
of that first bundle will necessarily require a steep learning rate, the cleanliness issue
should not stop NIF from achieving its goals.
The review by the revised baseline validation committee did not explicitly
address the cleanliness issue. It did, however, comment on beamline assembly and
installation, and on the beampath infrastructure systems, all of which are critically
coupled to the cleanliness issue. The committee found the personnel involved in these
activities to be of high quality and experienced.48 It further noted that the Integration
Management and Installation (IMI) contract was an “innovative solution” to the
problem of installing the beamline in the NIF building with adequate cleanliness. The
committee cautioned, however, that the contract must be well-managed and
monitored closely to ensure success. This is particularly true because the contract
does not contain allowances to correct problems that might arise during installation.
It is clear that DOE is depending critically on the IMI contractor to resolve the
cleanliness issue.
Also problematic are the issues about the laser optics. The report of the
Technology Resource Group (TRG) of the NIF Council noted that major technical
advances in eight areas were needed from the start of the project to reach NIF’s goals.
In other words, substantial research and development (R&D) to develop key project
components remained even while NIF was under construction. Furthermore, the
original NIF budget and program did not adequately account for that R&D, and the
technology development budget associated with the NIF project was exhausted by the
end of FY1998.49 The TRG noted that the deployment goal, which assumed that the
required technology and associated manufacturing processes were complete and
available, was unrealistic. It went on to say that the original plan, to begin operation
as soon as the facility was complete, was not possible because of the technical
uncertainties and required technology development.
While the TRG reports that significant progress has been made in these eight
areas, substantial technical challenges remain.50 The TRG identified four specific
technical areas of special concern. The most important of these, as noted above,
appears to be the potential for severe damage to the final optics assembly (FOA)
because of NIF energy intensity requirements.51 To reach the design level of 1.8
million joules (MJ), a fluence — beam energy per unit of area — of 8 to 9 joules per
square centimeter (J/cm2) is required. This level is a result of an original design
compromise that reduced the number of beams from 240 to 192. At that fluence
level, it was discovered, unexpectedly, that the fused silica optics when operated in
a vacuum with light at the wavelength at which NIF would operate — conditions that
48DOE Rebaseline Validation, p. 20, 25.
49Technology Resource Group, 6.
50Technology Resource Group, 5.
51The final optics assembly is that portion of the beam path that fronts on the target chamber.
That portion of the beam path operates in a vacuum.
CRS-17
would occur in the FOA — are subject to “catastrophic” damage in that the damage
grew rapidly with each pass of the beam.52
At this point, there is no obvious technical solution to this problem. The TRG
stated that design alternatives for the FOA are likely to be needed. It proposed that
the NIF project “get out of the box” in addressing the FOA problems.53 Further, the
TRG noted, it is possible that fused silica may never be able to permit full-fluence
operation of NIF, and that such operation is likely to need a superior optical material.
Without a solution, NIF would be forced to operate at either fluence levels of about
one-half the design level, reducing laser output below one MJ, precluding achieving
ignition, or the optics would have to be changed at a rate 10 times greater than now
planned, significantly increasing NIF operating costs.54 Indeed, GAO argues in its
June 2001 report that DOE may have significantly underestimated the annual
operating cost of the completed NIF because those cost estimate assume solution to
the FOA problem. DOE claims that it has developed a theoretical solution, but it has
not been tested in the laboratory.55
Another of the four areas concerns the quality of laser glass produced by the
continuous melt process. That process was selected by the NIF project team because
of the large quantity of glass required for the facility. The first batches produced in
this manner showed higher levels of water vapor trapped in the glass and greater
variations in the optical quality (measured by the index of refraction) than could be
tolerated. The laser glass is where the physical phenomena that produces the laser
beam takes place, and the two problems would result in beam energy intensity below
the design level. LLNL believes that it has found solutions for these problems. It
intends to use special furnaces to heat the laser glass material prior to its introduction
into the continuous melt process to reduce its water content. Recently, LLNL
announced that continuous-melt laser glass produced by one it its principal vendors
using this preheating technique met specifications for water content.56 As a result, it
appears that water content, at least, will not prevent the laser glass from being able
to support the design fluence of 8-9 J/cm2.57 The laser glass from this run, however,
still did not meet optical quality specifications. For this purpose, LLNL proposes
using special polishing techniques called “small tool” polishing. Such methods, while
appearing straightforward, have not yet been tested for NIF laser glass design
parameters.
52Technology Resource Group, 17. This problem is also referred to as the 3T problem
because the frequency of the laser light is increased by a factor of three as it enters the FOA
region. It is this higher frequency that creates the problems with the FOA optics.
53Technology Resource Group, 18.
54SEAB Laser System Task Force, 21.
55GAO,
Department of Energy, 4.
56Lawrence Livermore National Laboratory, “National Ignition Facility and Glass Vendors
Achieve Continuous Melt Milestone,” NR-00-03-13, March 31, 2000.
57Beam conditions for the laser glass and the laser glass itself are different than the FOA glass,
and the ability of the former to support the design fluence will not affect the problems
described about the FOA glass.
CRS-18
The TRG offered a number of recommendations to address the concerns it
raised. It estimated the cost of implementing these recommendations to be about $50-
80 million over the next eight years. The Group concluded that by following these
recommendations, “NIF can reach its original performance goals, although with some
delay and at higher cost than projected.”58
The MAY 30 LLNL letter also addressed the technical issues. The LLNL
Director reported that all of the external reviews concluded that NIF is “technically
sound and based on good engineering design.”59 Addressing the FOA issue explicitly,
the LLNL Director noted that it was not yet completely resolved. He stated that the
current plan is to operate NIF at fluence levels below the damage threshold for the
initial experiments. For those experiments that require higher levels, such as for
achieving ignition, it would be possible to replace the optics more frequently although
operating costs would increase. The NIF project team, however, expressed its belief
that the optics current limitations with the FOA will be overcome and that NIF will
be able to operate economically at the higher fluence levels.60
The revised baseline evaluation committee also addressed the FOA damage issue.
It concluded that existing technology would allow NIF to meet its goals but at a very
high damage rate that would adversely affect operating costs.61 At the same time, the
committee, expressed confidence that the problem would be solved provided
sufficient resources are made available. It stated, however, that while progress
towards a solution appears to be promising, the relevant R&D program described in
the NIF project, now estimated to cost $27 million, appears to be underfunded and
the committee recommended an increase of $30 million.62 The DOE certification
report of April 2001 did not address the FOA problem. On a related issue, the
committee recommended that LLNL contract with a second vendor for the critical
optics to ensure that a single-point production failure situation would not occur. Such
an occurrence is possible with just one vendor producing all of the critical optics.
While the cost estimate of the TRG and the revised baseline evaluation
committee are relatively modest compared to estimates of $3.5 to 4.1 billion for the
total NIF project cost, it should be emphasized that the issue with these technology
concerns is not cost but whether they can be solved at all. Solutions will likely require
basic research for which the outcomes cannot be predicted and success is only weakly
dependent on funds available. This is particularly true of the FOA concerns.
Therefore, while the TRG, the NIF Laser Task Force, and the revised baseline
evaluation committee all appear confident that the technical problems can be solved,
there remains considerable uncertainty, and success is not assured. Underscoring this
concern is a finding of the GAO study that criticized DOE for starting construction
of the project in parallel with critical R&D. GAO noted that results of that R&D have
58Technology Resource Group, 5.
59LLNL, May 30, 2000, letter, 4.
60LLNL, May 30, 2000, letter, 6.
61DOE Rebaseline Validation, p. 19.
62DOE Rebaseline Validation, p. 18-19.
CRS-19
forced changes in the project that were difficult to make and contributed to the cost
overrun because construction of the infrastructure had advanced so far.63
The SEAB Task Force also argued that LLNL should consider a different
approach to installing the full 192 beams. Currently, plans call for installation of two
bundles of 96 beams, the second to follow immediately upon completion of the first.
The Task Force suggested that an extended pause between the two bundles would be
preferable. While adding costs to the facility, such a pause would allow the NIF team
to commission the first 96 beams and determine if any modifications need to be made
in the second bundle before it is installed.64 Such a process could save a significant
amount of time in installing the second bundle. In essence, the first bundle could act
as the prototype which was not constructed earlier in the project. The Target Physics
Review Committee, however, did not favor a pause. It stated that there did not
appear to be any “programmatic reason” as far as the SSP was concerned for any
pause.65 The Committee noted, however, that if technical or budget reasons required
a pause, it should be done at a point where enough beams were available to do
important stockpile stewardship physics. It recommended that 120 beams would
provide the best opportunity for this situation.
The LLNL director in his May 30 letter argued that the path proposed by DOE
would have the same effect as the pause recommended by the SEAB Task Force.66
He stated that while there is no “distinct” pause in the preferred schedule, the fact that
it is stretched out would give the project team time to carry out the testing
recommended by SEAB. He noted that the completion dates in the preferred
schedule are the same as those recommended by SEAB including the pause. The
revised baseline evaluation committee did not make an explicit statement about
phased beam installation. Rather, the report appeared to assume a full 192-beam
facility.
Contribution to Stockpile Stewardship. The potential contribution of
NIF research to the Stockpile Stewardship program (SSP) is substantial and diverse.
It could be a valuable platform for investigating a range of scientific and technical
problems about nuclear weapons performance and reliability that could be significant
for the long-term security of the nation. All outside review groups setup by DOE to
date agree that these capabilities are critical for the success of the DOE SSP.67
Some other groups, however, believe that nuclear weapons safety should be the
primary concern of the SSP, and oppose NIF because its contribution to verifying
63GAO,
National Ignition Facility, p.14.
64SEAB Laser System Task Force, 11.
65Target Physics Review Committee, 8.
66LLNL, May 30, 2000, letter, 6.
67 See, for example, JASON, The Mitre Corporation,
Science Based Stockpile Stewardship,
JSR-94-345, McLean, VA, November 1994. This study was performed under DOE contract.
CRS-20
weapon safety would be relatively small.68 In addition, a few critics argue that the
level of reliability and performance that DOE is trying to maintain for the stockpile
through NIF and other elements in the program is excessive and not needed for a
credible deterrent. Others, however, believe that the level of confidence desired by
the DOE program cannot be achieved without nuclear testing.
Recent advances in pulsed-power research at Sandia National Laboratories
(SNL) have also raised questions about whether other, lower cost technologies might
be able to achieve many of the goals of NIF. Based on successful experiments
completed in 1998 on the SNL’s Z accelerator, a new machine, called the X-1, is
being proposed at a cost of about $300 million. That machine might be able to
explore important conditions needed for the SSP.69 As currently envisioned, NIF
would still be able to reach temperatures greater than those likely for the X-1 allowing
NIF to investigate conditions not otherwise attainable. It is possible, however, that
the X-1 could also achieve ignition, although that is uncertain at this time. If so, it
might be able to achieve a large fraction of NIF objectives. Currently, no decision has
been made whether to proceed with the X-1. For FY2001, DOE is requesting
funding to continue experiments on existing pulsed-power devices in order to assess
such devices for a high yield. DOE, however, did not request any funding for
development of advanced pulsed-power devices, suggesting that any decision about
the X-1 remains on hold. The Target Physics Program Review Committee
recommended that the Z accelerator, along with the Omega laser at the University of
Rochester, be used extensively to explore important SSP scientific issues as both a
supplement to NIF and to mitigate some of the consequences of the delay in
completing NIF.70 The Committee, however, said nothing about the X-1.
The May 30, 2000, LLNL letter repeats the arguments made at the projects
inception in support of the need for NIF.71 It states that the completed NIF (with the
full 192-beam assembly) is the only instrument capable of reaching the conditions of
temperature and density necessary for greater understanding of the physics of nuclear
explosions. Further, it is stated in the letter that NIF will be necessary to help validate
the codes being developed for the Accelerated Strategic Computer Initiative whose
goal is to simulate the nuclear explosion as a substitute for nuclear weapons testing.
In both cases, the letter’s author notes that ignition is essential, and the potential of
NIF to reach ignition is one of the most important features of the project.
The delay in construction and cost overruns have raised the question about
whether NIF should be completed to its original design level of 192 laser beams. In
particular, could NIF still make a significant contribution to the SSP if it were built
68 See the National Resources Defense Council web site for an extended critique of NIF and
SSP [http://www.nrdc.org/nrdcpro/fpprog.html]; and Unpublished memorandum to Marylia
Kelly, Tri-Valley Citizens Against a Radioactive Environment, from Greg Mello, May 13,
1994.
69James Glanz, “Will NIF Put the Squeeze on Sandia’s Z Pinch?”
Science, v. 277, July 18,
1997.
70Target Physics Review Committee, 8-9.
71LLNL, May 30, 2000, letter, 4.
CRS-21
with only a fraction of the beams, say 48 or 96? The Target Physics Program Review
Committee argues that the facility should be completed to the full 192 beams.72 In
particular, the Committee states that ignition is very important if NIF is to be of
maximum utility, and it very likely will require all 192 beams to reach ignition. At the
same time, however, the committee noted that important contributions to the SSP can
be made with a 48-beam laser, and it recommended that experimentation begin as
soon as that level was reached. The committee also argued that operation of 48
beams could help resolve several of the operational issues and ease attainment of the
192-beam goal.73
A report issued by directors of the three weapons laboratories also argued for
completing the laser to the full 192 beams.74 In order to study critical physical
phenomena in the laboratory in regimes that permit extrapolation or the results to the
behavior of actual nuclear weapons, minimum laser energy levels are needed. The
report notes that the energy threshold for some of these phenomena will be crossed
at the 48 to 96 beam level. Others will require 96 beams or more, while still others
will require the full 192 beams.75 Furthermore, the report’s authors argue, the full 192
beams are likely to be required to reach ignition and to provide enough energy to test
the computer models being developed to simulate nuclear weapons behavior.76 The
April 2001 certification report reiterated the need to develop the full complement of
beams. As noted above, the report cited a study by the High-Energy-Density Physics
Workshop that examined alternatives to a 192-beam facility and concluded the SSP
requirements were best met with the full 192 beams.
It appears that a critical factor in this debate is whether all 192 beams are needed
to reach ignition. While some scientific issues of stockpile stewardship can be
addressed if NIF did not reach ignition, many observers believe it is critical to do so.
In particular, the authors of the Foster report concluded that it “is very important that
the NIF produce ignition in order to address a new range of stockpile issues ... .”
They argued that “a half-power NIF without ignition is not worth the investment for
stockpile stewardship.”77 If ignition could be reached with a smaller number of beams,
it is possible that a smaller facility could be built or that NIF can make a greater
contribution to the SSP before it is complete than now seems to be the case. The
studies requested in the conference report along with the FY2001 DOE
appropriations appear to ask for a DOE justification of building the full-scale NIF.
Another aspect of this issue is how the delays in completing NIF and paying for
its cost overruns will affect the remainder of the SSP. GAO concluded that the
72Target Physics Review Committee, 4.
73Target Physics Review Committee, 7.
74U.S. Department of Energy,
The National Ignition Facility and Stockpile Stewardship
White Paper, April 24, 2000, [http://www.dp.doe.gov/dp_web/news_f.htm].
75White Paper, 6.
76White Paper, 7.
77John S. Foster, Jr., et.al.,
FY2000 Report to Congress of the Panel to Assess the Reliability,
Safety, and Security of the United States Nuclear Stockpile, February 1, 2001.
CRS-22
consequences for the program could be substantial.78 In particular, GAO reported
that construction delays will mean that some of the experiments needed to reach the
SSP milestones for 2010 could be delayed by three years. The budget consequences
could be much more serious. GAO argues that if the cost overruns are taken from
existing programs as is now DOE’s plan, “significant portions of the weapons
program could be affected.”79 Further, GAO reports that the three weapons labs —
LLNL, LANL, and SNL — are not in agreement about how to proceed with NIF to
minimize the potential budgetary impact. At this point, according to GAO, DOE has
not sought to resolve these differences and the new baseline decision was made
without adequate consideration of its effects on the rest of the SSP.80
The revised baseline evaluation committee also raised this issue, although
somewhat obliquely.81 It noted that NNSA and DOE Defense Programs should take
steps to ensure a balance between the long-term NIF operating costs and the
resources needed for other facilities and activities “essential” for the stockpile
stewardship program. The committee stated that a plan for governing NIF as a user
facility must be in place before research begins on the facility.
The June 2001 GAO report noted that there appears to be difference of opinion
among the three weapons laboratories about the potential usefulness of NIF to the
SSP.82 In particular, it stated that Los Alamos National Laboratory believes that NIF
will be limited in its contribution to the SSP unless it can use plutonium and can
achieve ignition. The former is currently not planned and the latter, as discussed
below, is uncertain. Sandia National Laboratory argues that NIF’s contribution will
be limited because it will not be able to certify whether weapons can exist in a hostile
(radiation) environment. GAO notes that the two labs reportedly have recommended
that NIF be limited to a smaller number of beams until its performance can be assured.
In any case, the potential contribution of NIF is a matter of dispute. The
increased cost of the project has revived these concerns. Although Congress at this
point appears to have agreed that a full-scale NIF is necessary for the program, it has
asked DOE, through the NNSA, to certify this conclusion in light of the cost
increases. It may be difficult, however, to verify whether the several contributions of
NIF, combined with those of the other elements of the SSP, are effective in helping
to maintain performance and reliability without some kind of nuclear test validation.
NIF and Ignition. NIF is being designed to reach thermonuclear ignition —
the condition where energy released by the fusion reactions is sufficient to sustain the
reaction. Nearly all scientists who have reviewed the project believe that ignition will
78GAO,
National Ignition Facility, p.25.
79GAO,
National Ignition Facility, p.26.
80GAO,
National Ignition Facility, p.27.
81DOE Rebaseline Evaluation, p. 48.
82GAO,
Department of Energy, 2-3.
CRS-23
be reached at the design level of 192 laser beams.83 Some critics have argued that
certain physical conditions concerning the interaction of the laser beam and the plasma
that is generated when the target is vaporized will prevent attainment of ignition.
They have said that more analysis is needed before a decision to begin physical
construction is made.84 Recent experiments on the NOVA laser at LLNL, however,
showed that some of those concerns might not be warranted. Other improvements
in target design have also raised confidence that ignition can be achieved. Indeed,
some have suggested that combining all of the advances could result in ignition being
reached with as few as 96 beams, although there would be virtually no margin in
terms of laser performance.85 Nevertheless, if these improvements are possible, then
the margin at 192 beams should be substantial. Furthermore, it is possible that higher
fusion energy yields (gain)86 than were envisioned in the original NIF design would
be possible, leading to the ability to study physical phenomena still closer to actual
conditions in nuclear weapons.
While confidence is high that ignition will be achieved, it is still not certain,
however, and confirmation probably will have to await until NIF is operational. If NIF
cannot achieve ignition for any reason, its value to the SSP would be reduced
significantly as noted above. It would still be able to perform physics useful for the
program, although it is not clear whether its contribution in that case would be
commensurate with its cost. Because of the uncertainty and the importance of
ignition to the program, the lab director’s and the Target Physics Program Review
Committee recommend the full 192 beams.
NIF and Nonproliferation. Because NIF would explore certain physical
regimes (temperature and density) similar to that of a thermonuclear weapon, some
concern has been expressed that results from NIF studies could contribute to nuclear
proliferation.87 Accessibility to NIF for investigations of civilian applications of
inertial confinement fusion would likely enhance the expertise of individuals who are
or could become weapons designers. Such knowledge likely is not sufficient by itself
for proliferation, however, and is irrelevant for the design of
fission weapons likely
to be the objective of first-time nuclear weapons states or organizations. It should be
noted that the basic scientific principles of operation of thermonuclear weapons are
widely known. Also, a DOE study, coordinated with the Arms Control and
Disarmament Agency, found that potential proliferation concern can be managed.88
83 JASON,
Science Based Stockpile Stewardship; Target Physics Review Committee, 11.
84 Stephen E. Bodner, “Time-Dependent Asymmetries in Laser-Fusion Hohlraums,”
Comments on Plasma Physics and Controlled Fusion, v. 16, 1995, pp. 351-374.
85Target Physics Review Committee, 10.
86Fusion yield or gain is the ratio of energy produced by nuclear fusion reactions to the energy
required to drive those reactions. Ignition occurs when gain just exceeds one.
87 Jacqueline Cabasso and John Burroughs, “End Run Around the NPT,”
The Bulletin of the
Atomic Scientists, September/October 1995, pp. 27-29.
88 U.S. Department of Energy, Office of Arms Control and Nonproliferation,
The National
Ignition Facility (NIF) and the Issue of Nonproliferation, NN-40, December 19, 1995.
CRS-24
NIF and Inertial Fusion Energy. While NIF is being designed to achieve
ignition, an important milestone for inertial fusion energy (IFE) research, the specific
type of laser used by NIF is unlikely to be a candidate for a practical fusion power
plant driver, the device that supplies energy to initiate the fusion reaction. Although
DOE is currently funding research to develop alternate driver technology, there is no
certainty one will emerge that can meet conditions of a practical power plant.
Without such assurance, therefore, even successful demonstration of ignition on NIF,
prior to alternate driver development, may be for naught as far as development of
inertial fusion energy is concerned. On the other hand, proceeding with driver
development before demonstration of ignition could also result in misallocated
resources. If ignition cannot be achieved at conditions that would allow a practical
fusion reactor, then a feasible driver candidate could be of little value.
A potential added complication is the emergence of the National Nuclear
Security Administration (NNSA). This organization, which was mandated by
Congress in the FY2000 defense authorization act (P.L. 106-65, H.Rept. 106-301),
might create added organizational barriers between the defense and civilian activities
within DOE.89 If so, coordination between the civilian and defense inertial
confinement fusion research efforts might be made more difficult. Recently, however,
DOE announced the formation of a “Heavy Ion Fusion Virtual National Laboratory
(VNL)” involving three of the DOE labs including LLNL.90 The purpose of the
collaboration is to conduct research on heavy-ion drivers for inertial fusion energy
applications. Such an agreement might serve as a means to smooth the passage for
future inertial fusion energy research involving NIF.
Outlook for the 107th Congress. The limitations on spending for the NIF
project imposed by the FY2001 DOE appropriations and Defense authorization acts
(see above) ensure that Congress will continue to monitor this project closely during
the 107th Congress. The primary focus will probably be how DOE is performing in
meeting the changes in the project baseline announced on September 15, 2000. In
addition, there is likely to be close oversight of DOE’s efforts to resolve the remaining
technical problems with the project. Of particular interest are the consequences of
any actions on the eventual cost and effectiveness of NIF. Besides the management
and technical issues, the role of NIF in the stockpile stewardship program may be
raised in the 107th Congress. As noted above, questions to this effect were raised
during the 106th Congresses review of the difficulties encountered by the NIF project.
Any assessment of the stockpile stewardship program as a whole is also likely to
consider the need for NIF.
Conclusions
The importance of NIF to the Stockpile Stewardship and Management Program
makes it difficult to separate decisions about NIF from decisions about the entire
program. If one agrees that the SSP as defined by DOE is both feasible and
89Congressional Research Service,
Department of Energy: programs and reorganization
proposals, by Carl Behrens and Richard Rowberg, RL30307, Oct. 28, 1999.
90Fusion Power Associates, “Heavy Ion Fusion VNL,” FPN00-20, April 5, 2000,
[http://aries.ucsd.edu/fpa/fpn00-20.shtml].
CRS-25
necessary, then NIF now appears necessary to achieve the program’s goals,
particularly if it achieves ignition. Without NIF or nuclear testing, there currently
does not appear to be any way to examine important aspects of nuclear weapon
physics critical for the computer codes designed to simulate nuclear weapons or to
carry out complete validation of those codes. If one believes that the SSP is either not
feasible or not needed, however, then NIF could be called into question.91 If NIF
were not needed to assure a credible nuclear weapons stockpile, then it would have
to be evaluated on the merits of its civilian scientific applications. While those areas
appear to be important, the project’s high cost would make it difficult to justify solely
as a civilian scientific research instrument.
The recent problems with NIF have added considerably to the controversy
surrounding the project. With a cost overrun of nearly 60%, it is difficult to see how
completion of NIF could be funded from existing stockpile stewardship funds.
Indeed, actions by Congress for FY2001 are providing additional funds although not
the entire increase requested, however, do not suggest that additional funds will be
easily forthcoming. While the current sentiment is to build NIF to its original design
specifications, it is possible that the laser will be scaled back. If the cost overruns are
even greater than now estimated, if the technical challenges with the laser optics are
more daunting than now appears, and/or confidence in ignition with a lower number
of beams grows, pressure could mount to reduce the scope of the project.
At this point, it is unclear how extensive any action on NIF by the 107th Congress
will be. The 106th Congress did agree that the project was necessary and approved
most of the adjustments requested by DOE. There has been no indication thus far that
the 107th Congress may have a different view of NIF. If the new project management
demonstrates that it has control of the project and meets the revised schedule, and if
good progress is made in resolving the technical issues, then it is likely that oversight
will be confined to the review of the budget request. If new problems emerge,
however, it is likely that the entire project will be subject to extensive review and
possible modification.
91 For a description of other options for stockpile stewardship see Congressional Research
Service,
Nuclear Weapons Stockpile Stewardship: Alternatives for Congress, by Jonathan
Medalia, CRS Report 96-11 F, December 14, 1995.