NRC and Nuclear Industry Generation of False Records and Tolerance of Inaccurate Technical Information On Zirconium Nuclear Fuel Rod Cladding and Other Systems Critical To Public Safety

NRC and Nuclear Industry Generation of False Records and Tolerance of Inaccurate Technical
Information on Zirconium Nuclear Fuel Rod Cladding and Other Systems Critical to Public Safety
M. J. Kelly. Sept. 8, 2013 Draft*
Key Points Summary
Nuclear Industry Technical Inaccurate information was found in reports describing zirconium alloys that were
Information Includes sent to several companies in the nuclear industry. Zirconium alloys are used in
Significant Errors. nuclear fuel rod cladding, guide tubes, and in other reactor core components.
An NRC Report Confirmed The supplier responsible for the errors refused to allow reviews of past reports for
that Industry Reports errors or industry notifications. The NRC was informed. The NRC confirmed that
Included Inaccurate two types of zirconium alloy texture errors were in reports. GE Nuclear had
Information. notified suppliers that these errors could create radiological hazards.
Scientific Studies Confirm GE Numerous technical publications support GE Nuclear concerns that
Nuclear Safety Concerns. misunderstanding zirconium alloy textures creates hazards. Fuel rod cladding and
other zirconium alloy components can leak, deform, and break due to bad texture.
The NRC stated that the NRC Relying on two limited consultations, the NRC dismissed safety concerns. Intended
would not require uses of the erroneous information were unknown to the NRC and its consultants.
corrections of known errors They could not know potential uses of the inaccurate information. To date, some
or reviews of suspect known inaccurate information identified after the initial NRC report remains
reports. uncorrected. Suspect information remains unreviewed. Clients remain unnotified.
The NRC disregarded GE The NRC dismissed written safety notifications and disregarded publications
Nuclear and scientific studies describing safety concerns. An NRC report indicated that the NRC would not
indicating safety concerns. pursue dissemination the erroneous information. The NRC reaffirmed this in a
2011 letter to Sen. R. Lugar. (Both the NRC report and 2011 NRC letter are
reproduced in the appendix below.)
NRC Audit Records NRC personnel later participated in two audits at the facility responsible for the
Document Recurring Nuclear erroneous reports. These audits identified recurring procedural, training,
Industry Deficiencies. software, and other QA deficiencies that an earlier NRC report and industry
records confirmed were the sources of errors in earlier nuclear industry reports.
An NRC Audit Record One NRC audit record inaccurately indicates that no previous NRC inspections had
Contains Significant False been done at the facility. This NRC audit noted many of the same deficiencies that
Information. were identified in earlier NRC records of an audit at the same company.
The False NRC Information The inaccurate NRC record decouples documented errors from federal regulations
Has the Effect of Concealing and explicit GE Nuclear safety notifications applicable to earlier reports. These
Recurring Nuclear Industry reports relied upon the same defective procedures, software, training, and QA for
Deficiencies. accuracy. Thus, this inaccurate NRC record conceals defects, could misdirect
reviews for inaccurate technical information, and could undermine failure analysis
investigations. (The NRC audit records are reproduced in the appendix below.)
What are the Bases of Given NRC tolerance of inaccurate information and generation of false records,
Current Nuclear Regulations? are nuclear industry regulations always grounded in facts, reality, and science?
NRC and Nuclear Industry Generation of False Records and Tolerance of Inaccurate Technical Information
NRC Tolerance of Bad Spent nuclear fuel is one of the most hazardous materials created by man. Nuclear
Zirconium Alloy Data isotopes are contained by cladding composed of zirconium alloys. Failures of fuel rod
Creates Specific Nuclear cladding have allowed escape and spread of hazardous radioisotopes. Numerous scientific
Hazards studies indicate that bad zirconium texture can lead to cladding failures. Understanding
cladding texture and failure modes is critical to safe use of nuclear fuel rods and safe
storage and handling of spent nuclear fuel rods. Integrity of cladding is essential to
radiological containment in spent fuel cooling pools and during subsequent handling and
storage. Bad texture properties contribute to failure modes of other zirconium alloy core
components. Dissemination of bad zirconium texture information in industry is hazardous.
Poor NRC Information Technical information of known reliability is required to make good decisions concerning
Practices Are Systemic designs and uses of alloys, to guide actions taken in event of accidents, and to develop safer
Problems Undermining nuclear systems. NRC failures to correct known bad information and to require reviews of
Nuclear Safety suspect data undermine safety. Some potential uses of the inaccurate and suspect
information cannot be known, so their impact on nuclear safety cannot be assessed.
NRC Tolerance of Bad Documents demonstrate that the industrial scientist that the NRC consulted for its safety
Data Undermines assessment worked for a company that had received and failed to recognize erroneous
Industry Credibility zirconium alloy information. Therefore, information from this consultant is suspect.
Courts Rely on NRC Legal decisions have relied on the NRC report to dismiss safety concerns about the spread
Safety Assessments of inaccurate technical information in the nuclear industry. Citing the NRC report, courts
have disregarded specific industry warnings, studies published by experienced scientists
and engineers, and testimony describing hazards of inaccurate information.
NRC Toleration of Relying on the NRC report, one court found that the nuclear supplier violated no laws in
Inaccurate Information requiring employees to sign false QA documents. Evidence demonstrated that the
Led Courts to Rule that falsifications concealed the errors, that the QA documents were reviewed by outside
Generation of False auditors, and that accident investigations relied on similar internal records. NRC failures to
Records by the Nuclear require reviews for and corrections of inaccurate technical information and fraudulent
Industry is Legal and documents creates opportunities for companies to avoid liabilities in event of nuclear
Reasonable accidents by requiring employees to aid in generating false records that conceal problems.
The NRC Should Require The NRC should recognize that tolerance of inaccurate data and false records hinders
Correction of the nuclear technology and put public safety at risk. Known inaccurate zirconium information
Erroneous Reports, must be corrected. Accurate information about cladding and all other nuclear materials will
Related Inaccurate be required for safe waste handling and storage. Suspect information must be reviewed.
Records, and the NRC's Such information is relied upon for decades. Permanent solutions to nuclear waste disposal
Own Flawed remain as chimerical as nuclear energy too cheap to meter promised at the dawn of the
Information Practices in nuclear industry-- no doubt these inaccurate projections arose from faulty understanding of
Order to Reduce the realities of nuclear technology due to reliance on bad information. If unanticipated
Nuclear Hazards nuclear fuel rods degradation causes radiation leakage and, later, accidents, deficient NRC
and industry information practices are responsible. However, tolerance of any deficient
information practices creates systemic hazards. 2012 NRC Indian Point license renewal
hearing transcripts contain examples of ongoing NRC problems with information.
Are accidents the time to ask Can the data be trusted?
Page 2
Table of Contents
Key Points and Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
An Overview: Some Quotes from NRC and Court Records . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1. Materials Information and Zirconium Alloys in Nuclear Applications . . . . . . . . . . . . . . . . . . . . . . . . . 6
2. Errors in Reports on Zirconium Texture Sent to the Nuclear Industry . . . . . . . . . . . . . . . . . . . . . . . . .17
3. An NRC Report Confirmed Errors in Nuclear Industry Zirconium Alloy Reports. . . . . . . . . . . . . . . . . 23
4. Published Scientific Studies Describe Nuclear Safety Implications of Zirconium Texture. . . . . . . . . .24
5. NRC Responses to the Texture Analysis Errors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
6. Recurrences of Zirconium Texture Analysis Failures and Falsification of Records. . . . . . . . . . . . . . . 31
7. Courts Rely on Uninformed NRC Safety Assessments to Dismiss Safety Concerns. . . . . . . . . . . . . . .37
8. Opaque and Incomplete NRC Investigations and Audits. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .49
9. Inaccurate NRC Audit Records Conceal Recurring Nuclear Industry Problems. . . . . . . . . . . . . . . . . .51
10. Accurate Records and Good QA Are Essential to Failure Analysis and Error Identification. . . . . . . 56
11. Conclusions and Recommended NRC Actions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Appendix I. Four NRC Records Documenting Flawed NRC Information Practices . . . . . . . . . . . . . . . 84

Appendix II. Inaccurate Information Impeded UA 232 Accident Responses and Failure Analysis. . . . 87
Appendix III. Short ARB Decision and Related Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Authors Experience and Credentials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .104
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
________________
(Draft Ver. Sept 8, 2013. No proprietary company information is included in this document. Note: This
reports primary purpose is to describe NRC and industry information problems in order to get the
specific inaccurate information and unsound NRC information policies corrected. Further details and
analysis information are available on request. mark.kellyzr2013@gmail.com. ProDoc:Protect8-20-13 )
Page 3
An Overview: Some Quotes and Excerpts from NRC and Court Records
(The following excerpts can be read in the full context of their source documents. The NRC report is Appendix I. The others are posted on-line.)
The NRCs summary of errors appearing in reports of zirconium alloy cladding materials properties:
(Source: NRC Allegation NRR-1999-A-0057. 1 Texture Analysis of Zirconium Alloy. Reproduced in Appendix I Part A below.)
The NRCs lack of action regarding known errors in reports:
(Source: NRC ReportNRR-1999-A-00571.)
GE Nuclear Notification to Lambda Regarding Zirconium Alloy Texture Work:
(Source: Order. Case No. C-1-00-661. United States District Court for the Southern District of Ohio Western Division. 12-3-01. 19 The GE
Purchase order notification reproduced on page 19 in the decision.)
Zirconium in the Nuclear Industry. 14th International Symposium. Paper on Texture:

Manufacturing technology used nowadays was to a great extent determined by the relation of texture
with physical and mechanical properties of zirconium tubes, their dimensional stability under irradiation,
and sensitivity towards stress corrosion cracking . . 12 (Grytsyna et al. Destruction of Crystallographic Texture in Zirconium
Alloy Tubes. Zirconium in the Nuclear Industry. 14th International Symposium, p 305. Journal of ASTM International, Sept. 2005 Vol 2 No. 8.
12
)
Sworn DOL Hearing Testimony from a Physicist Concerning Zirconium Texture Report Errors:
(Source: Case No. 2000-ERA-0035. ALJ Hearing. testimony. Department of Labor, Office of Administrative Law Judges (ALJ). 31 )
NRC Conclusion Regarding Safety Implications of Known Report Errors Conflicts with GE Nuclear and
Others:
(Source: NRC ReportNRR-1999-A-00571.)
Page 4
Sworn DOL Hearing Testimony from a Physicist Concerning Zirconium Texture Analysis Report Errors:
(Source: Case No. 2000-ERA-0035. ALJ Hearing. testimony. Department of Labor, Office of Administrative Law Judges (ALJ). 31 )
Federal Courts Relied Upon the NRC Report to Make Broad Findings on Nuclear Safety:
( Source: United States Court of Appeals for the Sixth Circuit. Case No. 02-3035. Decision On Appeal from the United States District Court for
the Southern District of Ohio. 15 )
A Federal Court Ruling on Falsified Records Concerning Records of Errors in Nuclear Reports:
(Order. Case No. C-1-00-661. United States District Court for the Southern District of Ohio Western Division. 12-3-01. 19 )
Flawed NRC information practices over the decades: Uncertainty and lost records, but not great
science in 2012 Indian Point Relicensing Board Hearing transcripts excerpts.
But what this does is leave us with a degree of uncertainty as to exactly what the source for the
contaminants that are referred to. It may have been plutoniumbut at this point theres no way to
really be sure of that.
38.
United States of America Nuclear Regulatory Commission. Atomic Safety and Licensing Board Panel Hearing. Docket Nos. 50-247-LR and 50-
286-LR. ASLBP 07-858-03-LR-BD01. In the Matter of: Entergy Nuclear Operations, Inc. (Indian Point Generating Units 2 and 3). Judge McDade.
Page 2026.
We have a gaping hole in the record, and we have no way of knowing what went into these costs.
38.
286-LR. ASLBP 07-858-03-LR-BD01. In the Matter of: Entergy Nuclear Operations, Inc. (Indian Point Generating Units 2 and 3). Lemay. Page.
2043.)
So, okay, we lost the reference to 1984. Someone convinced themselves, and then the Staff viewed it
as reasonableand, again, I know were not looking at great science, but what Im struggling with is,
okay, we lost the reference, but it is 30 years later, and were trying to deal with a plant-specific analysis
for this facility.
38.
286-LR. ASLBP 07-858-03-LR-BD01. In the Matter of: Entergy Nuclear Operations, Inc. (Indian Point Generating Units 2 and 3). The Staff is
the NRC staff. Judge Kennedy. Page 2016.
Page 5
1. Materials Information and Zirconium Alloys in Nuclear Applications.
Reliable technical information of known quality is critical to safe nuclear technology. Good
technical information is required to control the chemistry and physics and safely operate nuclear
reactors. Accurate records of events and the understanding that has guided past operations are critical
to assessing current operations and improving future procedures and designs. Accidents are not the
time to ask Can the data be trusted? Unlike sales and legal professions, where advocates routinely
and purposely manipulate and distort information selectively to persuade others to facilitate targeted
outcomes, engineering and science rely on consideration of all of the best available information of
known quality to achieve societal goals. Actual performances of engineered systems like nuclear
reactors are ultimately determined by the unbending laws of chemistry and physics, not the persuasive
powers of advocates from the legal and sales professions.
Inaccurate and distorted information corrupts nuclear technology in several ways. Bad
information can mislead engineers about the underlying physics and chemistry of nuclear reactors. As a
result, they can make poor decisions and miss opportunities to improve safety and avoid hazardous
practices. Poor decisions sometimes lead to disastrous accidents. When reliance on bad information is
exposed, public confidence in the technology and those controlling it is undermined. At Fukushima,
reliance on bad information on tsunamis led to decisions to construct sea walls which proved to be too
low. These flawed decisions led to the nuclear accidents that the spread radioactivity that have
rendered areas of Japan uninhabitable. Reactions of the public and politicians have initiated actions
that have curtailed and may eventually end commercial nuclear energy in Japan and Germany. As a
direct result of the accidents and poor responses, the Japanese Nuclear and Industrial Safety
Administration, NISA, has been abolished. Due to the high actual and potential costs of such
unnecessary accidents, the future of nuclear power faces uncertainties.
The United States Nuclear Regulatory Commission and nuclear industry information practices
are not adequate to guarantee the integrity of technical data and records necessary for a safe nuclear
industry. To demonstrate the NRCs information problems, NRC documents and records wherein the
NRC clearly states that the NRC would take no action to correct known inaccurate and suspect nuclear
industry information on zirconium alloys used in nuclear fuel rod cladding and other nuclear
components critical to safety are reproduced and described in the following. Additional NRC
documents, court records, published technical information, and other information are included to
demonstrate why NRC tolerance of inaccurate zirconium information is scientifically unsound and
inherently unsafe. 39
The focus is on the zirconium information problems because these information problems are
well documented. However, the zirconium information problems include just a few types of examples
of broader NRC information problems. While in-depth understanding of the technical details may be
difficult, the NRCs generic problem of tolerance of inaccurate technical information is simple to
Page 6
understand. Accurate information is required to understand and control most processes, including
nuclear power generation.
NRC officials have stated that they would not take any action to correct or limit the
distribution of inaccurate technical information in the nuclear industry. When confronted with records
demonstrating that inaccurate technical information purporting to describe properties of zirconium
alloys intended for use in nuclear fuel rod cladding and other reactor components was distributed in the
nuclear industry, NRC officials wrote:
(Source: NRC ReportNRR-1999-A-00571. The full NRC report is discussed in more detail below. The NRC reaffirmed this report in a 20112. The
full NRC report and letter are described in detail the following and are included in the appendix below.)
Correction of information now known to be inaccurate, review of suspect information, and

notification of those clients who had received the inaccurate information and NRC actions to that end
would indicate that the NRC now takes a scientific approach to technical information and records.
Review and correction could be simple and inexpensive. However, the NRC appears to continue to take
an irresponsible approach to technical information. The NRC report quoted above been a legal barrier
to corrections of known errors, reviews, and notifications. Furthermore, in 2011, the NRC has reacted to
attempts to get the information corrected by reaffirming that they would take no action and would not
require correction of inaccurate information.2 In 2012, NRC records containing false information4 that
have the effect of concealing the extent of suspect information have been identified. (Note that the
NRC records reproduced in this document are sufficient to demonstrate that these problems have
occurred. The remainder of this document was written to provide background information and context
of the NRC informationproblems.39)
Unfortunately, NRC information practices remain deficient and do not appear to have improved.
NRC tolerance and generation of inaccurate information creates systemic hazards in the nuclear
industry. Therefore, responsible parties should work towards pressuring the NRC to improve its
information practices.
Immediate goals might be to have the NRC or other responsible party take the following actions:
1) Correct the known inaccurate zirconium information in records and reports sent to nuclear industry
clients. 2) Review zirconium fuel rod cladding information that is suspect. 3) Notify the affected parties
(which available records indicate are unaware of the known bad information). 4) Correct the false
information in NRC records. 5) Document these activities so that they serve to improve the
understanding of recent NRC and industry practices and their influence on the integrity of technical
information. These goals could be accomplished simply, quickly, and with expenditure of few resources.
Broader goals might include improvement of flawed NRC information practices evident in this
and other situations. As described below, the NRC and nuclear industry appear to tolerate inaccurate
information and inadequately documented procedures that appear to be necessary to manage materials
degradation responsibly. Unfortunately, with todays automated instrumentation, inadequately trained
Page 7
or overburdened analysts can now generate and disseminate vast quantities of inaccurate information
quickly in automated laboratories if not properly supervised and reviewed. The NRC must require
industry to maintain accurate technical information and records. Implementation of a zero tolerance
policy for bad technical information similar to those policies practiced by most reputable academic
institutions and by most reputable technical publishing companies would be a significant step towards
nuclear regulation that is based on the best available science and is grounded in facts and reality.
NRC officials generated records containing false information that conceal repeated
occurrences of industry problems that caused inaccuracies in technical information.3,4 NRC officials
have also generated records that decouple false information from known industry problems.3 (The NRC
documents are described below in detail. Some relevant parts are reproduced in the appendix.)
Court records demonstrate that the NRC has tolerated irresponsible industry behavior that
includes making false statements and generating fraudulent records that conceal the nature and extent
of distribution of inaccurate technical information within the nuclear industry.
Hearing transcripts from the 2012 NRC Atomic Safety and Licensing Board Panel in the matter
of Indian Point Generating Units 2 and 3 indicate other NRC information problems arising from flawed
NRC practices over the decades. Important documents, including those used to develop nuclear
regulations, appear to be lost or poorly understood. The origins and appropriate applications of
information in NUREGs and source references appeared to be open to question.
Why does the accuracy of technical information and related NRC and nuclear industry records
matter?
Accurate technical information is required to make the correct decisions necessary for safe
operations, for appropriate accident responses, and for waste handling and storage procedures.
Inaccurate and suspect materials information can undermine nuclear safety in many ways.
A) Design and manufacturing of zirconium cladding and other components are impaired by bad
information.
B) Bad decisions can be made when those decisions rely on bad information about the state,
behavior, and degradation of zirconium nuclear fuel rod cladding in reactor cores during
operations and when making the urgent decisions required to respond to accidents.
C) Safe cooling, handling, transportation, and storage of spent nuclear fuel rods and other nuclear
waste require an accurate understanding of spent fuel rod cladding and other materials
conditions, properties, and degradation in order to prevent radiation releases and other
hazards. 5
D) Flawed failure analyses of incidents and accidents arising from inaccurate technical information
and inaccurate industry records leads to faulty Lessons Learned. Faulty Lessons Learned
Page 8
could lead to bad decisions that make subsequent accidents worse. Origins of accidents may
never be understood, which could result in recurring mistakes and accidents.
E) Bad information misdirects limited resources and thereby impairs technological advances.
Inaccurate technical information could lead to development of inherently unsafe reactor designs
and procedures.
Reliable materials information can be particularly important in event of nuclear accidents, when
good decisions must be made quickly.6 Those decisions often have significant consequences.
Misunderstanding the condition of reactor materials and nuclear waste stored around reactors can lead
to decisions that make accidents worse.
In past incidents and accidents, many important questions concerned the states of the
zirconium nuclear fuel rod cladding in cores and in spent fuel pools. Have elevated temperatures from
loss-of-coolant accidents compromised cladding integrity and allowed release of radioisotopes? 6 Has
exposure of fuel rod cladding to air compromised cladding integrity? Have zirconium fuel rod cladding
failures led to loss of radioactive materials during normal operations? Have cladding failures
increased temperatures during ongoing accidents and caused releases of radioistotopes into the
environment?
According to reports from the US General Accounting Office7 and Organization for Economic Co-
operation and Development6 , in many incidents, the answers to these questions have too often been
Yes . Cladding has failed, and those failures have created significant hazards. Furthermore, in
response to the incidents in the reports, the then-current understanding of what had happened to the
fuel rods before and during incidents and accidents guided urgent decisions that determined how
limited resources were spent to prevent ongoing accidents from getting worse. These decisions have
themselves have had significant influences on the courses and consequences of accidents, both good
and bad. As the Fukushima and Chernobyl accidents demonstrated, nuclear facilities sometimes
explode before plants can be brought under control and shut down after an incident. Inaccurate and
suspect information is always an unwelcome and pernicious factor when critical decisions must be made
quickly.
Why Is Understanding the Condition and Behavior of Zirconium Alloys Important?
Accurate technical information on zirconium alloy properties is required to assess reactor core
conditions and to anticipate how zirconium components will continue to degrade as a nuclear accident
develop. Inaccurate materials information can undermine the sophisticated procedures and computer
modeling that nuclear operators might implement in their attempts to understand reactor conditions
and take appropriate actions that will reduce, not increase, the possibility and consequences of
accidents. 6, 8, 39
Nuclear designs rely on zirconium alloys because their properties make them highly suitable
for use in critical components. 9 Zirconium alloys can be manufactured to retain high strength, melting
Page 9
points, radiation resistance, and corrosion resistance under the demanding conditions within a nuclear
reactor. Nuclear reactor components made of zirconium alloys include pressure tubes, fuel channel
tubes, and, most notably, fuel rod cladding. When manufactured and used appropriately, components
made of zirconium alloys resist degradation in nuclear reactors and can be relied on to contain
radioisotopes in nuclear waste for long times.
Zirconium components do degrade within reactors. With time, the cumulative effects of
radiation exposure in the high temperature and chemically-corrosive environments of nuclear reactors
during normal operations cause zirconium properties to deteriorate. Such deterioration creates weak
links that can initiate failures or can contribute to the cascades of materials and systems failures that
often make ongoing nuclear accidents worse.6
Zirconium components can deteriorate and fail quickly if exposed to extremes and fluctuations
in radiation, temperature, pressure, and chemical conditions6,8,9 that occur during excursions from
normal operating conditions (such as those that occur during nuclear incidents or accidents).
Zirconium degradation contributions to failure cascades occurring in the courses of nuclear accidents
often make accidents far more difficult to understand and control. As a result, accidents and their
consequences can become far worse. Zirconium component failures have led to increased spread of
radionuclides during nuclear accidents.6
Coolant reactions with zirconium alloys generate hydrogen. 5,6,8 Rates of these hydrogen
generation reactions increase as temperatures increase. Much of this hydrogen reacts to form
hydrogen gas. Hydrogen gas can accumulate in reactor buildings and can explode, as occurred multiple
times at Fukushima. A significant portion of the generated hydrogen also diffuses into zirconium alloys,
forming zirconium hydrides. Zirconium hydrides weaken components and are highly reactive. Zirconium
hydrides within the alloy act as energy reservoirs. When exposed to oxidants, hydrogenated
zirconium alloys can undergo runaway reactions or burn. Burning zirconium can release large
amounts of energy in short periods of time. This release of energy can not only destroy reactor
components (like cladding, whose destruction releases radioisotopes into cooling water), but can also
raise core temperatures. Loss of system control can occur as a result. Runaway zirconium hydride
reactions in spent fuel pools can initiate containment failures and explosions that allow the uncontrolled
spread of radioisotopes in the environment.
Maintaining the integrity of the zirconium cladding of spent fuel rods is necessary to
containing radioisotopes. Spent nuclear fuel contains radioisotopes that make it one of the most
hazardous substances made by man.7 Intact zirconium fuel rod cladding is necessary to prevent release
of radionuclides into the environment during the years they are cooled in spent fuel pools. Therefore,
degradation of zirconium fuel rod cladding continues to be a concern long after spent fuel rods are
removed from nuclear reactors. For economic reasons, zirconium-cladded fuel rods are increasingly
taken to higher burnups in cores at higher temperatures for longer periods of time than in the past,
increasing the deterioration of zirconium cladding and the frequency with which fuel rods leak and
Page 10
break. 10 Deteriorated zirconium cladding is more likely to crack and release radiation, especially in
spent fuel pools and when spent fuel is handled and transported.7
Accidents have released excessive amount of radionuclides whose spread cannot be
controlled by current containment technologies. Limitations of currently available practical cleanup
methods allow some contamination to persist and spread indefinitely.11 Once released from fuel rods,
gaseous and soluble radioisotopes can spread steadily and quickly by winds and water. Fissionable
materials have been irretrievably lost due to zirconium cladding failures. 7
Use of zirconium alloys in reactor components is necessary for a combination of technical and
economic reasons. The performance of zirconium alloys under some of the harsh conditions
encountered in nuclear applications and accidents is unparalleled. When manufactured properly, cost-
effective zirconium alloy components can be fabricated and safely used in many nuclear applications for
limited periods of time.9 For technical and economic reasons, there are no currently-available
substitutes for zirconium alloy in some nuclear applications.
However, the safe use of zirconium alloys and storage of spent components requires an
accurate understanding of zirconium alloys properties and performance under anticipated conditions.
Numerous studies have demonstrated that understanding the deterioration of zirconium alloys in
nuclear applications requires accurate information concerning the original properties of zirconium alloy
components and how those properties change in nuclear reactors and during storage as waste. 5-13
Accurate technical information on zirconium alloy properties is essential. Designs depend

upon accurate alloy information. 9 Cladding condition is often the limiting factor that determines how
long nuclear fuel rods can be safely used in reactors. 5-13 In event of accidents, good decisions that
determine how limited resources will be allocated will must be made quickly. Computer and lab
simulations required to assess component reactor conditions and to guide operator decisions, especially
under accident conditions, rely on the accuracy of information on alloy properties for accurate
descriptions and projections of alloy behaviors. 9 Although they did not initially involve zirconium
directly, bad operator decisions and misunderstandings were responsible for the severe nature of the
Chernobyl accident.
Inaccurate alloy properties information can undermine and compromise the accuracy
of results obtained from computer simulations of reactor systems. If computer simulations that are
compromised by bad data are utilized to help guide reactor maintenance or responses to a nuclear
incident or accident6, bad decisions might be made that could initiate or exacerbate the situation.
Because of the unpredictable nature of incidents during normal operations and during accidents, and
due to the long periods of time s of nuclear operations, computer modeling is commonly used to quickly
predict the long-term degradation and condition of reactor systems and components for both practical
and economic reasons.6,38 Current technology makes useful (although not perfect) simulations of the full
operational lives of components, projections of deterioration that might occur during hundreds of years
of storage of nuclear waste, and prediction of reactor conditions that might be created by various
Page 11
possible responses to unanticipated accidents within short periods practical. However, such computer
simulations rely on accurate input information.
Designs of nuclear power plants that can be decommissioned safely require accurate projections
of anticipated component condition and degradation. Nuclear waste generated over the last half
century will have to be stored for thousands of years. Those making who are now making the decisions
that will determine if these activities can actually be done safely for the indefinite future will be long
gone before the consequences of their decisions are apparent. Those who will be required to deal with
the consequences of future unanticipated incidents and accidents will require accurate technical
information concerning the properties of zirconium alloys and other materials used today in order to
make appropriate decisions and responses.
Why is Zirconium Texture Important?
The distribution of crystallographic textures within zirconium alloy components is one of

many crucial factors known to contribute to changes in zirconium alloy properties that occur during
manufacturing processes and during exposure to high temperatures, radiation, or corrosive
environments. 5-13 Under some conditions, some textures of zirconium alloy components can be the
weak links that cause components to break or fail. Materials scientists have demonstrated that
zirconium components with some textures perform well and that components with other textures
perform poorly. It is known that component textures and other properties change with time in nuclear
reactors in significant ways. Many of those changes depend on the influences of reactor conditions with
initial alloy textures. 5-13 Zirconium fuel rod cladding is welded shut; zirconium alloy texture influences
weld strength, corrosion resistance, and weld embrittlement and how they change with exposure.14
(A technical note: Some pertinent information for those with technical backgrounds that is not
necessary to understand the NRCs information problems. Careful consideration of texture is sometimes
more important in designs and applications employing zirconium alloys than for those using steel alloys.
Some of the importance of zirconium alloy texture originates in the hexagonal zirconium atomic
arrangements (ie, crystal structure) that tend to form during component processing and that can change
under the high energy conditions within nuclear reactors. Because of its crystal structure and chemistry,
zirconium is fundamentally different than steel. The hexagonal atomic arrangement has a limited
number of directions that zirconium atoms within a given crystallite can readily slip to relieve stresses.
In highly textured polycrystalline hexagonal zirconium alloy, the ordering of these crystallites can
amount to alignments of weak links within a component that can lead to failure. Simplistically, in steel
and many other common materials, atoms can be presumed to assume higher symmetry cubic
crystalline arrangements. The cubic arrangements allow atoms to more readily relieve stresses by
slipping in more directions than is possible in a hexagonal atomic arrangement, thereby reducing
likelihood of failures. In steels, alloy textures are sometimes not important factors in failures.
Therefore, texture is sometimes neglected in steel component design. As a consequence, texture is less
familiar to some engineers. In some nuclear applications, zirconium alloys are used because zirconiums
corrosion and radiation resistances (and other properties) are greater than those of steel alloys.
However, unlike steel, zirconium alloy components can more readily fail to perform as expected if
Page 12
texture is not taken into account in component design, manufacture, and operation.5, 9 Studies have
shown that zirconium alloy textures existing within a component can change during operation, which
further complicates understanding texture effects on alloy behavior, particularly in cases where the
available texture data is inaccurate. 5-13 )
Bad materials properties information can impact nuclear safety for decades. Nuclear reactors
operate for decades. Materials information is used to produce designs and simulation software that are
relied upon for decades. Spent nuclear fuel is sometimes handled decades after being removed from
reactors. Permanent solutions to nuclear waste disposal remain as chimerical as nuclear energy too
cheap to meter. (Those projections of clean and cheap nuclear energy were no doubt based on
inaccurate technical information.) As long as nuclear waste might be handled, accurate understanding
of the materials properties of components will be required to avoid creation of unnecessary nuclear
hazards.
Some details of the influences of crystallographic textures on alloy states and properties are still
not completely understood6, 9 . Improvements in the understanding of zirconium alloy texture are
necessary in order to improve component performance, to anticipate component deterioration, and to
mitigate consequences of nuclear accidents.6,9 Unreliable texture information undermines efforts to
improve alloy understanding.
Developers of some computer models recognize the importance of the crystallographic texture
in computer simulations and designs, but must simplify or neglect texture due to limitations in
understanding how to incorporate complex but fundamental texture properties into modeling
programs6. Inaccurate texture information undermines efforts to improve this situation. In any event,
inaccurate data can render any computer modeling results inaccurate. (Texture is often neglected in
designs incorporating steel alloys with crystal structures of higher symmetry because texture is often
not a limiting property; this differs from the situation driving neglect of zirconium texture in some
nuclear designs, where complexities and lack of necessary understanding limit efforts to address texture
in detail. 6 )
The NRCs toleration of zirconium texture technical information known to be inaccurate is a

serious specific problem that has not yet been corrected. However, NRC toleration of any inaccurate
technical information and generation of false records by NRC officials are serious systemic problems that
undermine NRC and industry credibility and, more importantly, create serious but unnecessary
radiological hazards.
Technical Notes.
1) What is "crystallographic texture"? Crystallographic texture is a term used to describe the

relative order and disorder of crystallographic orientations of crystallites (or grains) that make up alloys
(Figure 1). Specimens composed of alloy crystallites whose orientations are more aligned with each
other are described as having "higher" or "more" texture. Groups of crystallites which are more
Page 13
randomly are said to have "lower" or "less" texture. Although some groups of crystallites are disordered
to the degree where they are described as having "no texture", texture can develop in zirconium
components manufactured to have no texture when the components are exposed to the high energy
conditions found in nuclear reactors.
FIGURE 1. Texture in Polycrystalline Zirconium Alloys.
Less Texture (Crystallites alignments shown by arrows are more random. Weak directions of
crystallites, indicated by arrows, are randomly oriented. Weak links are not frequently well aligned.)
A. B.
____________
More Texture (Crystallites alignments shown by arrows are less random. Weak directions of
crystallites tend to be oriented sideways. Forces aligned with weak directions might be more likely to
cause failures.)
C. D.
Figure 1 (caption). Texture in Polycrystalline Zirconium Alloy. Atomic arrangements within grains and
relative orientations of those grains can determine alloy strengths and weaknesses. More technical
description. Simplified 2-dimensional representations of a five-grain alloy sample with low texture (A
and B) and an alloy sample with much higher texture (C and D) are shown. Arrows indicate
crystallographic directions. O's were used to represent the atoms in A and C in order to indicate
possible arrangements of atoms lying in alternating "basil planes" of individual hexagonal crystallites.
The figures are not drawn to scale-- grains sizes are typically on the order of microns, while atoms are
Page 14
much smaller. The O's are removed in B and D to simplify the diagrams- their textures are the same as
those indicated in A and C respectively . Individual grains in each of the two groups of five adjacent
irregular crystallites were drawn with the same shapes and sizes in the figures to emphasize that texture
is independent of crystallite shapes and sizes. In real alloys, three dimensional grain shapes, sizes, and
orientations exhibit great variations and rarely appear to be identical. Note that in the highly textured
groups of grains shown in C and D, the arrows (or "weaker" crystallographic directions) roughly line up
horizontally in the group of grains, indicating a high degree of texture. Susceptible sites in textured
samples like that show in C and D are often more likely to degrade because weaknesses can fail
synergistically or to link up to initiate or propagate cracks or other flaws. Distortion of part dimensions
(which can impede insertion of control rods) and formation of brittle hydride growth (which lead to
cracks that allow radioisotopes to leak into cooling water) are examples of alloy failures that can be
caused by inappropriate component textures. Note that while sorting out the details of texture can be
complex and that a single specimen can have multiple textures, in the final analysis, texture simply
amounts to how the atoms and crystallites are arranged and oriented. In components with strong
texture, "weak links" can be more likely to line up and act in concert to cause failures. High energies in
systems like nuclear cores can cause changes in texture.5 Inaccurate texture information can produce
inaccurate understanding of deterioration of alloy properties and can hinder optimization of materials
properties. Appropriate texture can improve component performance under some circumstances. Note
that, for alloys composed of hexagonal zirconium phases, alloy texture is often far more important for
component designs than for cubic alloys. (Alloy microstructure is somewhat more complex than
explained here. Zirconium alloys can be processed to be composed of cubic phases. Transitions
between the low temperature hexagonal zirconium phases and the higher temperature cubic zirconium
phases can occur at significant rates and thereby alter texture within the temperature regime of a
nuclear core, especially in event of an accident. See Tenckhoff 9 for a more sophisticated description of
texture in zirconium alloys.)
----------------------
2) Why is crystallographic texture important? Crystallographic texture can determine how

zirconium nuclear fuel rod cladding and other reactor components fail during normal operations, under
the extreme conditions generated in the courses of many types of accidents, and during transportation
and storage of spent fuel rods. Weak links are sometimes more likely to line up in components with a
high degree of texture. For example, some cracking will be more likely to grow, spread, and eventually
cause failure.
A simplified diagram indicating how cladding failures involving formation and cracking of
zirconium hydrides occurs is shown in Figure 2. Brittle zirconium hydrides form in cladding due to alloy
reactions with some of the hydrogen that is generated when fuel rods are in contact with cooling
waters. Orientations and extents of zirconium hydride formation depend strongly on alloy texture.
Some alloy textures lead to the precipitation of larger hydrides which are oriented in ways that make
the cladding far more susceptible to cracking when subjected to stresses.
Page 15
Figure 2. Highly Textured Alloys Can Be More Susceptible To Deterioration and Failure. Hydrogen
generated by oxidation of surface zirconium from coolant water can diffuse into alloys and react with
zirconium to form weak and brittle zirconium hydrides . Grey areas indicate zirconium hydrides that can
grow within cladding in contact with coolant water. Cracks like those shown in D can be initiated by
horizontal tensile stresses (ie, stresses with components in the direction indicated by the arrows). At
high hydride levels, alloy failures might be initiated by stresses in other directions.
A. As- Manufactured. B. Hydrides Precipitate C. Hydrides Grow D. Hydrides Crack,

Components Fail
Figure 2 (caption). Highly Textured Alloys Can Be More Susceptible To Deterioration and Failures.
Formation and growth of brittle hydride precipitates lead to formation of cracks in components
manufactured with textured zirconium alloys. Simplified 2-dimensional diagram showing five grains of
a highly textured component. Arrows indicate the "c -crystallographic direction" of the three
dimensional hexagonal arrangements of zirconium atoms in each of the five grains composing the
sample). The levels of zirconium hydrides (irregular grey areas) forming at some types of grain
boundaries increase with time of exposure to coolant water (A to D).
A. As- manufactured zirconium has no or extremely low levels of zirconium hydrides. No exposure to
coolant.
B. Coolant water reacts with surface zirconium to form hydrogen. The hydrogen can diffuse into the
alloy and react with zirconium at higher rates at certain surfaces to form localized hydride precipitates
(shown in grey). Certain unfavorable textures facilitate hydrogen uptake and transport through the
alloy, leading to hydride accumulations in certain areas. Hydride growth rates can increase with
temperature.
C. Hydride precipitates grow and sometimes link up, forming larger brittle regions in the alloy. 5 Taking
fuel rods to higher burn-ups (ie, operating reactors at higher temperatures and using fuel rods for longer
periods) increases the rate and extent of hydride formation. Accident conditions can increase rates of
hydride formation and levels. Alloys with certain unfavorable textures form large plates that are more
susceptible to cracking.5 Hydrides formed in some highly textured alloys are more likely to link up.
Zirconium hydride volumes can be about 16% larger than parent alloy volumes and are unevenly
distributed, so hydride formation increases mechanical stresses within components.
D. Brittle hydrides can crack due to thermal, pressure, or mechanical stresses. Cracks can spread and
cause components to fail catastrophically. Cracks in fuel rod cladding can allow radionuclides to escape
into cooling water and the environment. Zirconium hydrides are highly reactive. When cracked fuel
rods containing hydrides are exposed to oxidizing conditions (eg, exposed to air), hydrides can react
energetically. Large energy releases and temperature increases can occur very rapidly, initiating
Page 16
explosions. Cracks increase the amounts of alloy and hydrides exposed to oxidizers, thereby increasing
rates of energy release and producing higher temperatures in shorter times. Components with
unfavorable textures are more likely to form larger hydrides with orientations that make them
susceptible to breaking completely when subjected to stresses. (See Chung et. al. 5 for a more
sophisticated description of hydride formation within zirconium alloys.) Reactor operation and control
depends on maintaining appropriate distributions of radioisotopes in the core. Fuel rod failures and
releases of radioisotopes can change distributions of radioisotopes within the core and other critical
reactor conditions. Failure to understand these processes can lead to poor decisions in response to
incidents or accidents, making bad situations worse. 6
-----------------------------------------
Cracked cladding has led to cladding failures and releases of radionuclides into the environment.
For economic and practical reasons, large releases of radionuclides sometimes cannot be cleaned up.11
Large areas around the Chernobyl and the Fukushima remain uninhabitable years after accidents
involving releases of radioisotopes.
Under other chemically oxidizing conditions, accumulation of hydrides and cracks in cladding
can contribute to release and spread of radiation in other ways.6 Hydrides can react vigorously when
exposed to water or air, releasing large amounts of energy very quickly. Cladding cracks increase the
surface area of hydrides exposed to the environment; reactions of cracked cladding and oxidants can
release more energy in a shorter period of time, leading to significantly higher temperatures. Some
reports indicated that these types of "runaway reactions" occurred during the Three Miles Island and
other nuclear accidents.6 Understanding cladding texture is critical to understanding these types of
reactions, to designing parts that are less susceptible to runaway reactions, and to making the right
decisions in response to accidents involving operational and spent fuel rods.
Although unfamiliar to many scientists, important aspects of crystallographic texture are

relatively simple to understand. As shown in the simplified diagrams in Figure 1, texture indicates how
atoms in individual alloy grains (or crystallites) are oriented relative to those in other grains. The
importance of texture lies in how those orientations impart strengths and weaknesses to components
made from an alloy. Texture measurements describe various aspects of alloy texture quantitatively.
While detailed descriptions of the distributions of orientations within alloy components made of
multitudes of microscopic grains can become complex9, the importance of understanding the texture
within a component often simply amounts to understanding how weak links in an alloy interact with
each other and with external and internal stresses on that component (Figure 2).
2. Errors in Reports on Zirconium Texture Sent to the Nuclear Industry.
Errors were found in zirconium alloy (Zr) properties data reported to the nuclear industry by
Lambda Research, Inc., a private laboratory that specialized in materials analysis. The US Nuclear
Regulatory Commission verified that Lambda reports contained two types of errors that led to
inaccurate descriptions of zirconium alloy component properties. 1
Page 17
The first error was a 90 degree rotation error.
An NRC report stated the following:
( Source: NRC Allegation NRR-1999-A-0057 Concerns 1 thru 3.1 Texture Analysis of Zirconium Alloy. A copy is reproduced in Appendix I Part A
at the end of this document. Note that court records demonstrate that report results, not specimens, were incorrectly rotated 90 degrees in
orientations. The NRC was provided with information indicating that this was a data problem, not a specimen rotation problem. This
misleading NRC statement is significant when considered in the context of fraudulent activity in industry. )
Reports with these 90 degree data rotation errors indicated inaccurate textures in the samples
that were perpendicular to actual measured textures, as shown in Figure 3.
Figure 3. Effects of a Ninety Degree Rotation Error on Reported Texture Results.
A. Arrows Show Crystallographic B. Accurate Indication C. A 90 Degree Rotation Error:

Orientations of Individual Grains. of Average Texture in A. Inaccurate Indication of
Average Texture in A.
Figure 3. Effects of a Ninety Degree Rotation Error on Reported Texture Results.

A. Texture of individual grains of actual specimen is indicates by arrows.
B. Large arrow indicates the actual average texture within a specimen.
C. In a report with a 90 degree rotation error, the specimen average texture is inaccurately indicated as
being oriented in the wrong direction.
Technical notes of interest to those with technical backgrounds, but not necessary to understand the
error: Texture in real specimens containing a multitude of microscopic crystallites is often more
complex than that in the simple hypothetical five grain specimen shown. For example, a single
specimens can appear to have two or more significant populations of crystallites with different
orientations that could each be considered as having its own distinct texture. Using multiple arrows
sometimes would not be a clear and effective way to indicate the crystallographic textures found in
more complicated specimens. Therefore, materials scientists often rely on pole figures, orientation
distribution functions, and other representations to indicate texture. As texture can be considered as
the degree of randomness of crystallographic orientations, representations that indicate texture in
terms statistical properties of crystallite populations such as mean orientations and calculated
Page 18
distributions of orientations about that mean are often used. However, as complex as a detailed
picture of texture might be, a 90 degree rotation error simply flips that picture on its side.
For any engineer using texture information to make decisions, 90 degree rotation errors are as
serious as not knowing which way is up. For example, in nuclear fuel rod cladding tubes, 90 degree
rotation errors can be the difference between fabricating tubes with alloy orientations so that through
wall stress corrosion cracking rates are maximized (making tubes that are most likely to fail) instead of
minimized (making tubes that are less likely to fail). Since there is a 90 degree orientation difference
between the tubes thinnest direction and the tubes thickest direction, the 90 degree rotation errors
could mislead a designer or engineer to select the worst alloy orientations (which could lead to faster
cladding failures) instead of the best orientations (which could maximize cladding performance).
Identification of errors in one report led to reviews of past reports for data rotation errors.
These reviews identified several past zirconium analysis reports containing rotation errors that were
sent to companies in the nuclear industry, including Westinghouse, for some time. Some clients were
notified that their past reports contained errors and were given revised reports. The rotation errors
were described in a Lambda QA Incident Report (QAR).
These data rotation problems appeared to be corrected to most parties involved. The NRC
was not notified of these errors at the time that the errors were found. (Actually, the problems were
not completely corrected at that time. Significant false records purporting to describe the nature of the
errors were generated by Lambda management, unbeknownst to most Lambda employees and affected
clients (including nuclear industry clients and at least one US military lab). These false records appear to
have deceived affected parties, may have deceived the NRC if they reviewed them, and, if an accident
were to occur, could have the effect of misleading accident investigators, as described below.)
Lambda procedures were reviewed and changed in an attempt to prevent recurrences of data
rotation errors. Unfortunately, these procedure revisions automated some of the data analysis;
analysis and review responsibilities were spilt between analysts in ways that made understanding and
recognition of other different Lambda texture analysis errors less likely when they occurred.
A second independent error led to distortions of data. The second type of error involved
software failures, defects in written procedures, inadequate training, and other quality assurance (QA)
deficiencies led to data distortions in reports sent to GE Nuclear just months later.
An NRC report stated the following.
(Source: NRC Allegation NRR-1999-A-0057 Concerns 1 thru 3. 1 Texture Analysis of Zirconium Alloy. Reproduced in Appendix I Part A below.)
Page 19
A QA report (QAR) describing these distortion errors and QA deficiencies was written and signed
by the technician and lab manager. This QAR recommended that past reports be reviewed for this type
of error, that procedures (some of which had been provided to clients at their request) be revised, and
that affected clients be notified of the problems. The QAR objectively described software failures,
procedure deficiencies, employee training failures, and other QA deficiencies that contributed to the
errors. The QAR was appropriately critical of Lambda QA and the technician and manager.16
Lambda Research management refused to allow reviews of past reports for these distortion
errors. Management demanded that the QAR be edited to include false and misleading information.
Edits included removal of statements indicating the need for reviews of past reports for distortion
errors, changes indicating that nuclear clients had already been notified of distortion errors, and
removal of recommendations that procedures and software be revised to include measurements and
checks to prevent recurrences of distorted results. A decision from the United States Court of Appeals
for the Sixth Circuit decision summarized some of the events.
Page 20
Page 21
(United States Court of Appeals for the Sixth Circuit. Case No. 02-3035. Decision. On Appeal from the United States District Court
for the Southern District of Ohio. 15 Note that the decision inaccurately describes the earlier 90 degree rotation error as an unrelated
procedural data-collection flaw, not a data rotation during post-data collection analysis by the previous lab manager, apparently following the
NRCs inaccurate description of this error1. This propagation of NRC errors in court decisions is significant, as described below. Note that
evidence in the court record indicated zirconium nuclear fuel rod cladding specimens, not zirconium insulation specimens, were analyzed
incorrectly. Also note that evidence demonstrating other problems such as software deficiencies, procedure problems, calibration deficiencies,
QA, and GE Nuclear audit problems were in the court record, but are not addressed in this description. However, despite these issues, this
portion of the decision does give an overview of some of the events demonstrated by the evidence available in the court record. )
The handwritten QAR edits described in the court decision inaccurately indicate that the client
was notified data were reviewed. See attached list and that procedure revisions were Already
present.
(Source: NRC web site document ML011430469-1.pdf. 16 Portion of QAR edited by Lambda Owner Paul Prevey. Note that edits removed
recommendations that past data be reviewed, that the client be notified, that procedures be revised to include measurements, and that
directions to cover software failures be addressed. The see attached list edit referenced an email (reproduced in Section 10 below) listing
reports containing the unrelated 90 degree rotation errors written months prior to the GE Nuclear distortion errors. Note that these three
issues were the points of dispute about the incident. The technician and manager did not dispute the other statements in this and the owners
versions of the QAR.)
Note that the decisions Footnote 3 in the above excerpt from the decision states that At
deposition, Prevey conceded in that, contrary to his revision to Kellys QA Report, Lambda had not
informed GEND of the procedural defects in Lambdas mounting process-- the decision confirms that
Preveys revision rendered the QAR inaccurate. (Preveys explanation for his demand that the QAR be
edited to inaccurately indicate that the client had already been notified is a separate issue. The decision
indicates that Prevey explained that he had confused that issue, the distortion error, for the ninety
Page 22
degree rotation error. Basically, he claimed he made a mistake about a mistake when confronted with
evidence that his QAR edits were not accurate and forbade discussion under penalty of firing because he
didnt want to waste time on the issues. The QAR was a QA record concerning nuclear work and was
audited by the nuclear industry- even if Preveys explanation is accepted, what does compounding
mistakes say about industry practices regarding information? Evidence appeared to indicate that there
were other incidents where management took actions that concealed errors and defective QA. )
Note also that the procedures were revised after procedures failed again later, as described in
Section 6 below.
3. An NRC Report Confirmed Errors in Nuclear Industry Zirconium Alloy Reports
Court records included documents from GE Nuclear audits of Lambda zirconium texture analysis
QA work and procedures. GE audit documents included explicit notifications that Lambda texture
analysis was required to comply with 10CFR21 and 10CFR50b (which addresses nuclear supplier QA ).
Some past GE Nuclear purchase orders clearly stated that 10CFR21 applied to the texture analysis
because errors could lead to significant risk of radiological hazards. (One such 10CFR21 notification is
described in some detail in Footnote 2 in the above excerpt of the United States Court of Appeals for
the Sixth Circuit, Case No. 02-3035.15 )
Safe operations require good information about components materials properties. In event of
failures or an accident, failure analysis relies upon good information. Lambda employees knew that bad
information could lead to bad engineering decisions affecting safety, ineffective corrective actions,
repeats of accidents, and uncertainties about the effectiveness of corrective actions intended to correct
safety problems.
When it became obvious that Lambdas owner would not allow correction of the problems or
notifications of clients, the NRC was contacted about the errors and attempts to conceal them. The NRC
was asked to correct the technical problems. The NRC verified the problems. An NRC report stated the
following.
Note the types of problems found to cause distortions in information in zirconium alloy reports
sent to the nuclear industry were fairly wide-ranging. Problems included procedure deficiencies,
training, and software deficiencies.
Page 23
4. Published Scientific Studies Describe Nuclear Safety Implications of

Zirconium Texture
Nuclear industry and materials science studies demonstrate the importance of understanding
the crystallographic texture of zirconium alloy in nuclear applications. The following quotes from
published studies indicate that misunderstanding zirconium texture can create serious nuclear hazards.
( Note that crystallographic texture is referred to as preferential or preferred orientation in some
quotes. Questions and some comments were included in order to organize the information and focus
attention on the significance of the quotes. Comments were added in parenthesis to further explain
some of the quotes. Note that these parenthetical comments were not from the publications cited.)
How do some studies describe the importance of zirconium alloy texture?

It would be difficult to identify another problem that has aroused so much permanent interest and
been the focus of such continuous attention, investigation, and scientific debate as zirconium reactor
core component texture. (Emphasis added.) (Grytsyna et al. Destruction of Crystallographic Texture
in Zirconium Alloy Tubes. Zirconium in the Nuclear Industry. 14th International Symposium, p 305.
Journal of ASTM International, Sept. 2005 Vol 2 No. 8. 12)
(Several references published before 1999 were cited in support of this statement, so the NRC should
have been aware of the attention, investigation, and scientific debate prior to issuing their report.)
Does texture influence how zirconium reactor core components are fabricated?
Manufacturing technology used nowadays was to a great extent determined by the relation of texture
with physical and mechanical properties of zirconium tubes, their dimensional stability under irradiation,
and sensitivity towards stress corrosion cracking . . 12 (Emphasis added). (In support of this statement,
the authors cite ten published references describing zirconium alloy applications, degradation modes,
and performance failures in nuclear reactors related to texture. Stress corrosion cracking is a
degradation process that can occur in reactor cores and in spent fuel pools.)
What did an Argonne National Laboratory study funded by the NRCs Office of Nuclear Regulatory
Research state about the general importance of zirconium hydrides crystallographic properties
(which include texture)?
The morphology, orientation, distribution, and crystallographic aspects of Zr hydrides in fuel cladding
fabricated from Zr-base alloys play important roles in fuel performance during all phases before and
after discharge from the reactor, i. e., normal operation, transient and accident situations in reactor,
temporary storage in a dry cask, and permanent storage in a waste repository. 5 (Emphasis added.)
(Source: H. M. Chung, R. S. Daum, J. M. Hiller, and M. C. Billone, Characteristics of Hydride
Precipitation and Reorientation on Spent-Fuel Cladding. Zirconium in the Nuclear Industry. 13th
International Symposium, p. 561. 5) (Note that zirconium hydrides are mechanically brittle and,
chemically, highly reactive species that form and accumulate in zirconium cladding while in use in the
reactor. When stressed or exposed to coolant water, hydrides are weak links in cladding that can
Page 24
form pinholes and initiate cracks that leak radioisotopes. Cracking can cause fuel rods to break and
release fuel pellets and fission products. )
What did this same group state about zirconium alloy texture and precipitation of deleterious
zirconium hydrides within cladding?
the susceptibility to hydride reorientation will be more pronounced in a cladding with a 0 degree, 30
degree, or 90 degree basal-pole texture, because easy precipitation is more likely to occur on either
prism, pyramidal, or basal planes. 5 (Emphasis added.)
(This quote describes textures effects on formation of Zr hydrides, which are weak links that can
determine how cladding will fail. See Figure 2 in Section 1 above for a simplistic picture of this process.)
DIMENSIONAL CHANGES.
Does reactor core radiation rearrange and destabilize zirconium alloy atoms and thereby influence
their crystal structures?
"Incident neutrons cause displacement of zirconium atoms from their stable positions by collision." 13
(Source: M. Dahlback, M. Limback, L. Hallstadiu, P. Barberis, G. Bunel, C. Simonot, T. Anderson, P.
Askeljung, J. Flygare, B. Lehtinen, A. Massish. The Effect of Beta-Quenching in Final Dimension on the
Irradiation Growth of Tubes and Channels. Zirconium in the Nuclear Industry. 14th International
Symposium, P. 276. Journal of ASTM International, June. 2005 Vol 2 No. 6 . 13 )
How do these radiation-induced displacements of atoms in crystallites deform zirconium nuclear

components during irradiation growth?
For zircaloy guide tubes, prism hcp planes are preferentially aligned perpendicular to axial
(longitudinal) direction. This means that irradiation growth causes axial length to increase and tube
diameter and wall thickness to contract." 13 (Emphasis added. Note that prism hcp planes are
preferentially aligned means that crystallographic texture is present and that lengths, thicknesses, and
shapes of guide tubes will change or warp in ways determined in part by texture. Significant changes
in component dimensions could cause serious reactor malfunctions, such as impairment of control rod
insertion. Such problems create hazards.16,17)
Do component distortions create measurable bows and dimensional changes in zirconium

components due to texture effects?
Irradiation-induced lateral displacements (bowing) of Zircaloy-2 channels irradiated in various BWRs to
different exposures have been measured using Westinghouse standard channel deformation
measurement procedure. 13 ( It is of interest to note that Westinghouse failed to recognize that it had
received inaccurate results from Lambda Research. This leads to questions: If some Westinghouse
technical data is inaccurate, do some Westinghouse measurement procedures and some Westinghouse
computer models produce inaccurate and misleading gibberish? )
What problems can these dimensional changes of zirconium components cause in operating nuclear
reactors?
Page 25
Excessive or unexpected dimensional changes of guide tubes or spacer grids of a nuclear fuel assembly
can result in operational issues such as incomplete (control) rod insertion (IRI) or potential fuel assembly
interactions and handling concerns due to increases in the fuel assembly envelope resulting from the
lateral growth of the grids. 17 (Source: King, S. J., Kesterson, R. L., Yueh, K. H., Comstock, R. J., Herwig,
W. M. and Ferguson, S. D., Impact of Hydrogen on Dimensionaly Stability of ZIRLO Fuel Assemblies.
Zirconium in the Nuclear Industry: Thirteenth International Symposium. ASTM SPT 1423, G. D. Moan
and P. Rudling, Eds. ASTM. pp. 471-489. 17 Note that incomplete control rod insertion can prevent
interfere with and possibly prevent shutdown (eg, scrams), which can lead to loss of reactor control and
meltdown.)
CORROSION:
What did a corrosion study state about the importance of in-reactor degradation of zirconium alloy
cladding?
"The extended fuel burnup, longer fuel cycles, power uprates, higher temperatures, and increased
lithium in PWR primary water put a premium on the ability of the cladding to withstand in-reactor
degradation." 18
( Source: Arthur T. Motta, Aylin Yilmazbayham, Robert Comstock, Jonna Partezanna, George P. Sabol,
Barry Lai, and Zonghou Cai. Microstructure and Growth Mechanism of Oxide Layers Formed on Zr
Alloys Studied with Micro-Beam Synchrotron Radiation. Zirconium in the Nuclear Industry. 14th
International Symposium18, P. 205. Journal of ASTM International, May. 2005 Vol 2 No. 5. 18 )
What did the same study state about the importance of texture for understanding of zirconium alloy
cladding performance and degradation?
"We believe that, in addition, the local texture and grain-to-grain alignment are crucial for
understanding differences between the alloys. 18 (Emphasis added.)
HYDRIDE PRECIPITATION : Zirconium corrosion due to formation of zirconium hydrides within alloys,
such as the processes diagramed in Figure 2 above. Note that hydride levels increase with temperature
due to increased hydrogen generation at water-alloy interfaces.5
Does zirconium alloy texture determine the formation, distribution, and crystallographic
characteristics (eg, texture) of deleterious zirconium hydride precipitates?
Face-centered-cubic delta hydrides frequently precipitate on a certain habit plane of the unit cell of
the hcp alpha-phase Zircaloy metal. 5
(Source: H. M. Chung et al. Characteristics of Hydride Precipitation and Reorientation on Spent-Fuel
Cladding. Zirconium in the Nuclear Industry. 13th International Symposium, P. 562. 5 This statement
means that precipitated hydrides will accumulate in ways determined by texture. The implications of
this include the fact that that cladding and other components can accumulate hydrides and fail in ways
determined by texture. Certain types of zirconium hydrides are known to accumulate in certain types of
zirconium cladding alloy crystallite planes. If Zircaloy develops certain types of textures, then zirconium
hydrides formed in the core could make cladding susceptible cracking, leaking, or breaking when
Page 26
stressed. The study stated Hydride platelets that are distributed perpendicular to the direction of
primary stress (i. e., radial hydrides) are particularly deleterious to the claddings mechanical
properties. 5 Essentially, bad texture can promote growth of zirconium hydrides that creates weak
links in cladding. See Figure 2 above. Also note that 90 degree rotation errors in texture analysis would
thwart accurate understanding of primary stresses and radial hydrides. In fact, 90 degree rotation
errors could mislead designers to select the worst rather than the best orientations. See Figure 3
above.)
Does zirconium texture determine how zirconium hydride reorientation occurs?

the susceptibility to hydride reorientation will be more pronounced in a cladding with a 0 degree, 30
degree, or 90 degree basal-pole texture, because easy precipitation is more likely to occur on either
prism, pyramidal, or basal plane. 5 (Emphasis added.)
How well is hydride precipitation in nuclear fuel rod cladding understood?

the fundamental crystallographic and microstructural characteristics of hydride precipitation in
irradiated cladding remain unknown. 5
(Among other things, this statement indicates that texture is an area of active research and
development. In general, inaccurate information undermines research and development.)
Why is the texture of hydrides important to consider in cladding design?

Hydride platelets that are distributed perpendicular to the direction of primary stress (i. e., radial
hydrides) are particularly deleterious to the claddings mechanical properties. 5
(In other words, bad texture promotes growth of zirconium hydrides that are weak links in cladding.)
Is hydride formation and orientation important to zirconium degradation and waste handling?
One example is the hydride reorientation that can occur during the initial stage of vacuum drying and
transfer of spent- fuel assembly to a dry-storage cask. Radial hydrides in such a situation could lead to
significant degradation of creep and other mechanical properties under long-term dry storage. 5
(This means that cladding continues to degrade and eventually could fail when handled and after being
put into dry storage long after fuel rods are removed from the core.)
These studies clearly indicate that misunderstanding zirconium texture can create serious
hazards. Very hazardous situations can develop when nuclear reactors are operated at higher
temperatures, cooling pools are packed with ever increasing numbers of spent and deteriorating fuel
rods, and when spent fuel rods are handled if unfavorable zirconium alloy textures are present or
develop in components.
The Fukushima and Chernobyl nuclear plants were technological wonders right up until
accidents destroyed them. Multiple systems were in place to ensure safety. Prevention of each
accident was well within technological capabilities. The causes of the accidents were understood before
the accidents. The accidents were caused by wrong decisions based on wrong technical information.
(Decisions leading to inappropriate heights of sea wall and locations of backup generators caused the
Page 27
Fukushima accidents.) Few recognized the dangers created by the bad decisions until it was too late;
the accidents progressed to the point where the effects of poor decisions could not be reversed. Once
the accidents started, many safety systems were overwhelmed and defeated in cascades of systems
failures, making the accidents worse. Cleanup of these accidents is well beyond current technological
and economic capabilities.
Requiring industry to maintain accurate technical information amounts to cost-effective and

responsible accident prevention. Once cladding breaks, radioisotopes can spread uncontrollably and
cleanup often becomes impractical. NRC failures to require correction of and to block the spread of
known bad and suspect technical information that could undermine attempts to prevent such incidents
is irresponsible. Checking data is cheap. Nuclear accidents are unacceptably expensive and hazardous.
What did GE Nuclear state about the importance of accurate zirconium texture analysis data in some
of its purchase orders?
(Source: Order. Case No. C-1-00-661. United States District Court for the Southern District of Ohio Western Division. 12-3-01. 19 The GE
Purchase order notification reproduced on page 19 in the decision.)
5. NRC Responses to the Texture Analysis Errors.
The NRC confirmed that both types of distortion errors were included in reports sent to the nuclear
industry. An NRC report1 stated the following.
The NRC confirmed the origins of the distortion error. An NRC report1 stated the following.
The NRC assessment of the errors was superficial and based upon uninformed sources.
The NRC report1 states that the NRC did not understand texture, so the NRC relied upon the unreviewed
and unpublished opinions from two selected consultants. These consultants also did not know intended
and potential uses of the data by GE Nuclear and other nuclear industry clients.
Page 28
(Source: NRC Allegation NRR-1999-A-0057 Concerns 1 thru 3. 1 Texture Analysis of Zirconium Alloy. Reproduced in Appendix I Part A below.
Documents posted on the NRC web site appeared to indicate that the major nuclear fuel fabrication company was Westinghouse, one
company which court records demonstrate had received and failed to recognize bad zirconium texture data from Lambda. Does the situation
where the NRC relies on consultant from a company who had failed to recognize bad information in order to determine effects of that bad
information amount to an absurd case of the blind leading the blind?)
The NRC report1 indicates that neither the NRC nor its two selected consultants knew
intended or potential uses of texture information by GE Nuclear and Lambdas other clients.
The NRCs expert consultants qualified their limited conclusions, which appear to be
somewhat vague and open to interpretation. Both experts believed that texture analysis, alone,
cannot be used to determine the intrinsic mechanical properties of zirconium-based tubing material.
The experts covered themselves by qualifying their statements, as intrinsic mechanical properties
can be considered as different from other properties such as stress corrosion cracking resistance. Did
the NRC inappropriately disregard these qualifications to make broad, general conclusions in their
report?
The NRC report indicated that their industrial experts company used different methods to
evaluate properties of zirconium alloys routinely. The report indicated that the experts believed that
the errors could indicate that good material was unacceptable, which could lead to bad decisions
about zirconium uses. (Note that evaluations of materials properties and performance testing often rely
on unacceptable and good materials as benchmarks and standards; accurate understanding of both
unacceptable and good materials is essential. Misidentification of good material as bad can
create serious misunderstandings about how alloys perform and degrade when exposed to harsh
conditions like those in nuclear applications. More importantly, the distortion errors could make a
zirconium alloy specimen texture look more random than it actually is. Some publications indicate
that more random textures perform better in some nuclear applications than textured specimens.12 In
this respect, both the NRC and their selected experts appear to have been technically wrong about
potential hazards of distorted zirconium texture information. Also, the key problem is generation and
spread of inaccurate information, not the particular test methods used to qualify zirconium. The NRC
consultants responses dont excuse uncontrolled dissemination of known inaccurate technical
information in the nuclear industry.)
The NRC report did not address published studies indicating that zirconium alloy texture was
an important property that determined, to a large extent, how zirconium alloys would perform and
Page 29
fail in nuclear applications.5-14, 17,18 As summarized in Section 4 above, many studies and performance
testing demonstrate that texture influences zirconium failures. (Although some of the studies quoted in
Section 4 were published after the initial NRC report, these studies cited earlier publications which
described the importance of texture in zirconium degradation. Some earlier studies are quoted in
Section 7 below.)
An internal NRC document (downloaded from the NRC web site in 2011, but apparently
written around the time of the NRC report1) indicates the importance of understanding uses of the
data in determining concerns over bad data. Internal NRC documents downloaded from the NRC web
site indicate that the NRC had doubts based on uncertainty of how GE used the information. However,
the NRC report1 did not include the questions and concerns noted in their own documents. The
following is an excerpt from the document from the NRC web site that indicates that some at the NRC
knew that texture could be important to nuclear safety:
(Source: NRC web site document ML011430491.pdf. )
Available records do not indicate what was meant by some minimal properties of the
cladding. If texture is not accurately measured, how can it be controlled? Despite its ignorance of the
uses of the data and a great amount of information indicating that understanding texture is critical to
understanding zirconium alloy component performance and failures, what did the NRC conclude about
safety?
Publications indicate that distortion errors like those in the Lambda reports could make material look
more suitable for some nuclear applications than it actually was.
The distortion errors had the effect of making pole figures from some of the samples indicate
lower texture than actually was present in the sample. Grytsyna et al stated in their paper that There is
no doubt that lack of texture would be optimal in terms of radiation stability.12 The findings in this and
other papers indicate that the NRCs conclusions that texture errors are not safety concerns are
inaccurate because, as this and other papers note,13,17 radiation stability is a significant concern for
zirconium fuel rods and other core components. Simply put, even after consultations with two
experts, the NRC got it wrong.
Did the NRC take any action concerning distorted or suspect results in past reports?
Page 30
The NRC report indicates that the NRC relied upon limited statements from two selected
expert consultants (one of whos company had received but failed to recognize inaccurate texture
information) as the basis for the NRCs broad conclusions and statements that the distribution of
inaccurate technical data was not safety concern and to decide that the NRC would not take action to
eliminate or prevent the spread of suspect technical information about critical reactor core components
in the nuclear industry.
6. Recurrences of Zirconium Texture Analysis Failures and

Falsification of Records
As described above, after the NRC failed to take any action, technical issues concerning review
of past reports for distortion errors, the notifications to past clients, and Lambda procedural and related
QA problems remained unresolved and errors remain uncorrected and unresolved.
Lambdas owner (Mr. Prevey) had forbidden Lambda Phase and Texture Analysis Laboratory
Manager (Kelly) from discussing them under penalty of dismissal.
(Source: Administrative Review Board (ARB) Case No. 02-075. Final Decision and Order. 20 The 3 page decision is reproduced in Appendix III
below.)
Preveys own deposition testimony showed that Preveys edits included false information and
that their inclusion in the QAR would have made the QAR inaccurate.
(Source: United States Court of Appeals for the Sixth Circuit. Case No. 02-3035. 15 )
Prevey had edited the QAR to remove recommendations to improve defective procedures which
had been repeatedly provided to GE Nuclear in the course of audits of the texture analysis procedures
for compliance with 10CFR50b and 10CFR21.
Page 31
(Source: United States Court of Appeals for the Sixth Circuit. Case No. 02-3035. Decision. On Appeal from the United States District Court for
15
the Southern District of Ohio. )
Prevey removed recommendations to address failures of the procedure to reliably produce flat
samples, measurement of sample flatness, and software failures were removed from the QAR. He
wrote that they were Already present on the margin of the QAR.
( NRC web site document ML011430469-1.pdf. QAR edited by Lambda Owner Paul Prevey. 16 )
However, after the NRC was notified of the problems, the sample mounting procedure again
failed during analysis of another GE Nuclear zirconium fuel rod cladding specimen. In an unlikely
coincidence, Lambda management passed a version of Preveys edited QAR covering the past GE
Nuclear failure to the lab manage for signature on the same day as the GE Nuclear procedure failed
again.
21
(Source: Case No. 2000-ERA-0035. Recommended Decision and Order. Department of Labor, Office of Administrative Law Judges (ALJ). )
Confirmation of the necessity to revise the procedures and to notify GE Nuclear (and other
clients) that procedures used on past samples were deficient became apparent the next day.
21
(Source: Case No. 2000-ERA-0035. Recommended Decision and Order. Department of Labor, Office of Administrative Law Judges (ALJ).
Note that Prevey had demanded that this same procedure not be changed. He had backed his
demands with firing threats.20 (Evidence indicated that Prevey fired people frequently. In one instance,
he fired the whole staff of a lab. Firing threats at Lambda were to be taken seriously.) Preveys actions
appear to confirm that procedure deficiencies contributed to the problems, that procedures needed
revision, and that his demanded edits to the QAR would have rendered it inaccurate.
Page 32
Despite the recurrence of a procedure failure and subsequent procedure revisions, the QAR
remained inaccurate in ways that could mislead a future accident investigation and in ways that could
lead to bad decisions about review and use of inaccurate technical information. The original QARs
descriptions of past procedure failures that led to GE Nuclear analysis errors, descriptions of required
procedure revisions, recommendations to notify GE Nuclear (and other clients) of procedure
deficiencies, and recommendations to review past work done with the defective procedures for errors
were kept out of the QAR prepared by Lambda management. Therefore, this QAR would have misled
any subsequent failure analysis investigation about the history and extent of zirconium nuclear fuel rod
cladding analysis errors.
Furthermore, the QAR and managements new texture procedure still did not address the
procedures lack of sample flatness measurement and analysis software failures. (Procedure revisions to
require flatness measurements were made much later after court actions were initiated. Effective NRC
action could have forced correction of these problems years earlier. Note that although the NRC was
notified that inaccurate information in reports and records were found during legal proceedings, the
NRC appears to have done nothing. As of 2013, there are no indications that effective reviews of past
work done with defective procedures were made and there are no indications that results known to be
inaccurate were corrected and that affected nuclear industry clients were notified of the errors. To the
contrary (as unbelievable as it may seem), in 2011, NRC officials wrote a letter to Senator Lugar
indicating that the problems were not pursued.2)
The failures of the Lambda procedures could have and should have been resolved months
earlier, when the original QAR was written. However, Lambdas owners removal of this procedure
revision from the QAR and his firing threats prohibiting discussion of the QAR and its problems
prohibited correction of these deficiencies.20,21 (Note that several judicial decisions findings of facts
describe how the procedure deficiencies, demands to prepare and sign inaccurate QARs, failures to
allow reviews of past reports for errors and notifications to clients, and the firing threats were
intertwined. )15,19,20,21
Prevey continued to pressure lab analysts to sign the QAR whose inaccuracy was confirmed by
procedure failures and his revision of the procedure.
Page 33
(United States Court of Appeals for the Sixth Circuit. Case No. 02-3035. Decision. On Appeal from the United States District Court for the
Southern District of Ohio.15 )
Coincident with the recurrence of the GE Nuclear cladding analysis problems, the owner
changed lab conditions to make any type of work impractical, made several angry outbursts, and made
various threats. (Court and deposition testimony demonstrate that lab workers did not argue or
confront Prevey showing anger.) Since the threats were more serious and the owner created
conditions under which responsible completion of technically difficult work was becoming increasingly
impractical, the NRC was again contacted and asked to investigate the testing problems, firing threats, a
physical threat, and retaliation surrounding concealment of the nature of the GE Nuclear errors and
defective procedures. The NRC did not respond during this critical period.
Prevey demanded signatures on an inaccurate QAR that concealed the full effect of the
zirconium alloy texture analysis problems. He initially demanded that the QAR be signed within five
minutes, which meant signing an unread QAR. The technician signed. Kelly asked for time to read it.
Prevey demanded that the QAR be signed before Kelly left his office. He then stated that there would
be no discussion or changes to the QAR. Lambda managements version of events recorded in the QAR
was to remain unchanged. Lambdas inaccurate QAR was to be signed unedited.
The ALJ decision described these events.
Page 34
21
Although Kelly had attempted to contact the NRC in the week prior to the demands to sign the
false QAR, the NRC did not respond until well after events were over. (Oddly, one NRC investigator later
asked if Prevey had smiled or laughed during his angry demands and exchanges. In another odd
exchange, Prevey had made the cryptic statement These people are not your friends shortly before he
demanded that the QAR be signed, but he never indicated who these people were. The NRC
investigator later participated in the audit described by NRC audit records containing false information.4)
What purposes do signed false document have when those documents have not been read by
the signers? Do they have any use other than to serve as possible cover for errors or misconduct? Do
false documents muddy the water so that, in event of an accident, misconduct by culpable parties
might go undetected? Do inaccurate and conflicting records foam the runway to limit liabilities for
accidents caused or exacerbated by companies who have engaged in unethically risky but profitable
practices? False information certainly has no role in good science. False records have no legitimate
place in engineering failure analysis investigations or in the nuclear industry.
Months later, other documents containing false statements that had been prepared by Lambda
management and signed by Prevey were found in legal discovery. The false statements concealed the
extents of analysis errors, including zirconium texture analysis errors. These fraudulent documents
indicate why signatures of the technician and Lambda manager on the inaccurate QAR might have been
useful to parties attempting to avoid being held responsible in event of an accident.
Historically, inaccurate documentation has impeded accident investigations, concealed root

causes of accidents, could have led to inappropriate accident responses that made consequences of
accidents worse, and left doubts about effectiveness of operational decisions intended to prevent
recurrence of accidents. Notably, in the aviation industrys UA Flight 232 accident investigation 22 , the
FAA stated that corrective action intended to prevent a repeat of an engine explosion due to a materials
failure that caused an accident that killed over 110 passengers may fail to have its intended effect
Page 35
because suspect disks could remain in service due to inaccuracies in decades-old manufacturer records.
22
(Source: National Transportation Safety Board, PB90-910406 NTSB/AAR-90/06, Aircraft Accident Report United Airlines Flight 232. 22)
In Nov. of 2011, the NRC reaffirmed its conclusions that the NRC would take no action on
known inaccurate and suspect zirconium materials information in a letter to US Senator Richard Lugar.
To date, the NRC has not reported any correction of known inaccurate and suspect zirconium
alloy data in industry. After being told of inaccurate information identified after their initial report, the
NRC did not follow-up or request details required to retrieve records containing false information for
review. To the contrary, the NRC reaffirmed its 1999 report (ignoring subsequent errors and misconduct
noted above) in a letter to Senator Richard Lugar in 2011.
(Source: NRC Correspondence. R. W. Borchardt letter to Senator Richard Lugar, dated Nov. 23, 2011.2 This letter is reproduced in the
appendix below.)
The NRC had new information between the date of their original report and the time that the
NRCs 2011 letter to Senator Lugar was written. The NRC was notified of specific bad zirconium data
found in discovery in court actions. The NRC was on distribution lists of legal proceedings which
included and involved generation of fraudulent industry records which were unavailable at the time of
the original NRC report. The NRC was provided with published references describing how zirconium
texture influenced the deterioration of nuclear components. The NRC was told of bad and suspect data
identified after their initial report. The NRC itself had funded studies indicating that zirconium texture
helped determine component deterioration in nuclear applications years earlier.5 (Does the NRC pay
any attention when it is notified of bad information? Does the NRC read and understand the studies
that it funds at taxpayer expense? )
In 2011, problems involving control rod insertion systems were identified at several plants.23
Were industry and NRC-directed corrective actions intended to fix these problems based on accurate
technical information?
Page 36
7. Courts Rely on Uninformed NRC Safety Assessments to Dismiss Safety Concerns.
Judicial decisions have significant impact on nuclear safety 24, 25. Courts proceedings will help
determining if existing nuclear industry practices are adequate to allow continued safe operation at
nuclear power plants.38 Courts decide whether or not frauds and bad behavior that could create
unacceptable hazards will be corrected. Courts set binding precedents concerning nuclear safety that
are followed by other courts.25
Some current legal precedents have been interpreted to require courts to rely on NRC
information. In one recent case, the decision supported NRC positions over opposing views and
evidence that lower courts found to be more credible and responsible. Citing legal precedent, the
appeals court ruled that the lower courts overstep their legal bounds by preempting Congress role in
regulating nuclear material under the Atomic Energy Act.24,25
Does the NRCs stated toleration of bad information in industry and reliance on industry for
technical guidance amount to "institutionalizing ignorance"? Does this NRC approach lead to legal
decisions and laws that require practices known to be unsafe by the scientific community? Do such
rulings set precedents that allow pro-industry and politicized elements within the NRC to trump science
and technology with authoritarian pronouncements based on legal interpretations of technically-flawed
NRC reports?
The following section focuses on how the NRC safety assessment accepting inaccurate technical
data and records while disregarding explicit industry warnings and the scientific literature impacted legal
proceedings arising from the zirconium errors. A few broader information issues arising during 2012
Indian Point nuclear plant licensing hearings are also described. The purpose is to show that legalistic
approaches to technical information can be inadequate due to authoritarian (as opposed to scientific)
approaches to technical information. Courts inappropriately relying on the NRC as an authority can lead
to decisions that increase nuclear hazards. Flawed NRC information practices are root causes of these
types of problems.
NRC Dismissals of Safety Concerns Lead to Judicial Findings that Refusal to Falsify
Documentation of Zirconium Analysis Errors Was Unwillingness to Obey a Valid Order.
The problems with the zirconium analysis errors led to several legal actions. The NRC
recommended making a complaint with the US Department of Labor based on discrimination and filing
for unemployment. Cases arose from the Energy Reorganization Act, the Lanham Act, and various other
laws involving industry frauds and demands by industry that employees participate in frauds.
Courts repeatedly cited the NRC report in dismissing safety concerns. Multiple actions and
multiple appeals led to multiple decisions in multiple courts which also relied on the NRC assessment.
Courts disregarded evidence from more-technically informed parties and publications that refuted the
Page 37
NRC conclusions. Was the NRC channeling complaints into legal venues that would bury industry
problems?
The 6th District Court of Appeals relied on the NRCs Report to determine that erroneous
texture analysis of the zirconium encasement on fuel rods will not have adverse safety consequences.
( Source: United States Court of Appeals for the Sixth Circuit. Case No. 02-3035. Decision On Appeal from the United States District Court for
the Southern District of Ohio. 15 Note that the NRCs letter is the cover letter to the NRC report which is part of Reference 1, reproduced in
Appendix 1 .)
In light of other NRC information and other technical information not addressed in the NRCs
letter, this courts statements are technically incorrect and far too broad. What hazardous practices
could be ruled legal and thus, practically unstoppable, if this ruling was regarded as a legal precedent?
The courts broad conclusions conflict with internal NRC records which indicate that there are
uses where texture results may be very important.
(Source: NRC web site document ML011430491.pdf. Note that this document was not made available by the NRC during court
proceedings. The NRC kept the investigation open until legal proceedings are concluded. Ongoing investigations can be barriers to discovery of
internal NRC documents.)
The courts broad conclusions also conflict with a later Argonne National Laboratory study
funded by the Nuclear Regulatory Commissions Office of Nuclear Regulatory Research state about the
general importance of zirconium hydrides crystallographic properties (which encompass texture).
The morphology, orientation, distribution, and crystallographic aspects of Zr hydrides in fuel cladding
fabricated from Zr-base alloys play important roles in fuel performance during all phases before and
after discharge from the reactor, i. e., normal operation, transient and accident situations in reactor,
temporary storage in a dry cask, and permanent storage in a waste repository. 5(Emphasis added.)
(Source: H. M. Chung et al. Characteristics of Hydride Precipitation and Reorientation on Spent-Fuel Cladding. Zirconium in the Nuclear
Industry. 13th International Symposium, p. 561.) (Note that zirconium hydrides are mechanically brittle and, chemically, highly reactive
species that form and accumulate in zirconium cladding while in use in the reactor. When stressed or exposed to coolant water, hydrides are
weak links in cladding that can form pinholes and initiate cracks that leak radioisotopes. Cracking can cause fuel rods to leak, break, and
release fuel radioactive fission products.)
The Federal Court relied on the NRC report to find that the inaccurate data and generation of
fraudulent documents were not safety concerns or violations of law.
Page 38
( Source: Order. Case No. C-1-00-661. United States District Court for the Southern District of Ohio Western Division. 12-3-01.19)
Once courts discard safety concerns, courts appeared to be free to find that companies can lawfully
demand preparation of and signatures on documents containing false information.
(Order. Case No. C-1-00-661. United States District Court for the Southern District of Ohio Western Division. 12-3-01. 19 Note that the NRC
documents reproduced in the appendix below and GE Nuclear documents in the court records indicate that the NRC and GE Nuclear each
reviewed these internal quality assurance incident reports whose falsification the court deemed legal. Its also interesting to note that Lambda
had actually introduced copies of the Quality Assurance Incident Report (QAR) in a hearing along with an ex parte communication to the
hearing officer explaining the QAR. Practically, this court decision appears to amount to legalizing deception and fraud. Technically, this is a
recipe for disaster and cover-up by the nuclear industry at the expense of the general public. Acceptance of and deference to corporate
misconduct is not unusual in the greater Cincinnati legal community where the legal proceedings took place. 29 )
The Sixth District Court ignored the unsure statement in the NRC report, characterized the few
NRC phone calls to its two select consultants as a specific investigation into the potential dangers posed
by zirconium-cladded fuel rods which had been inaccurately tested, and regarded the admittedly
ignorant NRC as the overriding authority in legally determining what constituted a safety concern (even
though the NRC safety assessment appears to have relied upon an employee of one of the major fuel
fabrication companies which received bad texture information and failed to recognize it as inaccurate).
(United States Court of Appeals for the Sixth Circuit. Case No. 02-3035. Decision. On Appeal from the United States District Court for the
Southern District of Ohio. 15 The statement dismissing Glavicic is inaccurate: Glavicic clearly testified that he was a physicist, not a chemist, and
one of his publications concerning his work on nuclear reactor materials in nuclear engineering applications was in evidence. Also in evidence
was Kellys experience in metallurgical analysis, including work in support of the UA 232 accident investigation described below. The courts
statements are good examples of the judiciarys inappropriate authoritarian approaches to science, technology, and information.)
Page 39
In counterpoint to the last sentence in the above excerpt from a federal courts decision, it
should be noted that materials scientists have indicated that materials must be understood in terms of
fundamental chemical processes. Professor Donald Sadoway of MIT succinctly commented on the
importance of understanding the chemical origins of materials At the end, everything is chemistry.
The rest is stamp collecting. 30
The Department of Labor Administrative Law Judge (ALJ) relied on the NRC report dismissing safety
concerns to dismiss allegations that the errors and defective procedures were safety concerns.
(Source: Case No. 2000-ERA-0035. Recommended Decision and Order. Department of Labor, Office of Administrative Law Judges (ALJ). 21
(Its interesting to note that no evidence indicated that Prevey had performed any meaningful assessment of the safety situation. Preveys
testimony indicates that he did not know how GE Nuclear or other nuclear industry clients could use or misuse inaccurate information and did
not have the knowledge or resources to perform a safety assessment.)
No evidence indicated that Prevey ever performed or was qualified to make a credible safety
assessment. While the excerpted statement from the decision appears to be creative invention by ALJ,
reliance on the NRC led the ALJ to ignore the very real GE Nuclear 10CFR21 and 10CCFR50b safety
notices, testimony, and published scientific literature describing texture effects on cladding
deterioration in evidence that accurately described the safety situation.
GE Nuclear notifications, testimony, and publications in evidence demonstrating safety concerns were
disregarded by courts due to the NRC report.
The GE Nuclear purchase orders (PO) warnings and audit documentation for zirconium texture
analysis indicated that 10CFR21 and 10CFR50b applied to the work and described nuclear hazards that
could arise from errors.
(Source: Order. Case No. C-1-00-661. United States District Court for the Southern District of Ohio
Western Division. 12-3-01. GE Purchase order reproduced on page 19 in the decision.)19
Testimony from a former Lambda physicist hostile to the legal actions (one of whose
publications in the nuclear field was in evidence) indicated that texture was Most definitely a safety
concern:
Page 40
(Source: Case No. 2000-ERA-0035. ALJ Hearing. testimony. Department of Labor, Office of Administrative Law Judges (ALJ).
This physicist also testified about the importance of texture in more technical terms (which
may be less clear to those without appropriate technical backgrounds):
(Source: (Source: Case No. 2000-ERA-0035. ALJ Hearing. testimony. Department of Labor, Office of Administrative Law Judges (ALJ). 31)
Page 41
(Source: Case No. 2000-ERA-0035. ALJ Hearing. testimony. Department of Labor, Office of Administrative Law Judges (ALJ). 31)
Published scientific studies linking texture to nuclear safety concerns were in evidence.
One paper in evidence, The influence of Crystallographic Texture and Test Temperature on
Initiation and Propagation of Stress-Corrosion Cracks in Zircaloy, D. Knorr et. al. Zirconium in the
Nuclear Industry. 6th International Symposium (1982), P. 627 32 concludes Crystallographic texture
exerts a strong influence on both the initiation and propagation of iodine stress-corrosion cracks in
unirradiated Zircaloy. (p. 649.)
The published discussion of this paper includes this comment by A. B. Johnson of Battelle
Northwest that indicates that failures occur suddenly without warning: Having inspected many cross
sections of irradiation Zircaloy fuel cladding, I am impressed that evidence of incipient cracks is generally
absent. I am aware of a view that a crack will propagate rapidly to failure if it initiates under reactor
operating conditions. This could explain why we generally observed either failures or absence of
cracks.32
The paper and discussion include comments on texture effects on methods commonly applied
to materials failure analysis. As the papers title The influence of Crystallographic Texture and Test
Temperature on Initiation and Propagation of Stress-Corrosion Cracks in Zircaloy indicates,
crystallographic texture influences zirconium failures in nuclear reactors and in spent fuel storage pools,
especially when overheated (eg, during loss of coolant accidents ).
The abstract of another paper in evidence Variation in the Strain Anisotropy of Zircaloy with
Temperature and Strain, by Hindle and Worswick, published in Zirconium in the Nuclear Industry, 6th
International Symposium (1982) p. 133, 33 begins The strong crystallographic texture which is
developed during the fabrication of zirconium-based alloys cause pronounced anisotropy in their
mechanical properties, particularly deformation. The abstract ends The effects of the strain
anisotropy observed are discussed in relation to the effects of temperatures on the ductility of
Zircaloy fuel cladding tubes during postulated large-break loss-of-coolant accidents.
Lacking technical expertise to evaluate technical issues and relying on the NRC, courts did not
appear to understand or appreciate the fact that that inaccurate technical information undermines such
studies and the understanding of the very same experts who the NRC and others would likely call upon
Page 42
for guidance in event of nuclear accidents. (Papers published as late as 2009 indicate that the roles of
texture in zirconium component failures are still not fully understood6, indicating that the 2011 NRC
response to Senator Lugar2 was irresponsibly inaccurate.)
Dismissing safety concerns, courts ruled that refusals to prepare and sign an inaccurate record
concerning zirconium alloy errors in reports distributed to the nuclear industry were failures to obey
a valid order.
Despite finding that Preveys edits would have rendered the QAR inaccurate and despite the
later history of actually changing the procedure as requested in the original accurate QARs, the ALJ court
still found that the QAR should have been changed as requested by Lambda management and that
refusal amounted to unwillingness to obey a valid order .
21
The 6th District court had noted that Preveys own deposition indicated that one ordered edit of
the QAR was inaccurate:
`(Source: United States Court of Appeals for the Sixth Circuit. Case No. 02-3035. Decision. On Appeal from the United States
District Court for the Southern District of Ohio.15 Note that the DOL ALJ hearing included the evidence that Preveys QAR edits contained false
informationthe ALJ ignored this evidence in the decision21.)
Prevey actually participated in changing the procedure at a later date, demonstrating that
other ordered QAR edits removing mention of the procedure deficiencies and the need for procedure
revisions were inaccurate.
21
Yet a court found that QAR recommendations to change the procedure were unreasonable.
(Source: Administrative Review Board (ARB) Case No. 02-075. Final Decision and Order.20)
Page 43
Three courts appear to have ruled that preparation of a QAR record containing false information
about notification of defects to a client is unwillingness to obey a valid order21, is unreasonable20,
and does not violate any federal, state, or local laws19.
Refusal to obey a valid order is a firing offense. Unreasonable behavior is often considered
reasonable grounds for dismissal. Employers can require employees perform actions that dont
violate any federal, state, or local laws. Courts into which the NRC directed legal complaints appear to
rely on NRC guidance to render decisions that support nuclear industry demands on employees to
generate false records that conceal nuclear industry defects.
Hearing rules limited introduction of evidence into the record. Technical evaluations are made
by judges unqualified to make technical assessments. Different courts made different and sometimes
incompatible findings of facts. But in nearly every case, the courts reliance on bad information from
the NRC was the greatest barrier to correction of known errors in nuclear technical information.
Legal precedents and flat-earther authoritarian pronouncements led the judges to rely on the NRC and
to disregard more credible sources. (The NRC report1 appears to have been effectively employed by
industry and their lawyers and relied upon by the courts to block reviews and corrections of errors in
reports sent to the aviation and other industries, including US military organizations.34,35 )
In 2012, flawed NRC information practices became apparent during the NRC Atomic Safety
and Licensing Board Panel Hearing. (Docket Nos. 50-247-LR and 50-286-LR. ASLBP 07-858-03-LR-
BD01. In the Matter of: Entergy Nuclear Operations, Inc., Indian Point Generating Units 2 and 3. The
following quotes should be reviewed in the full context of the official hearing transcripts, exhibits, and
other evidence when the NRC makes them available in order to understand how NRC information
problems impacted the course of this hearing.)
In 2012, an NRC Atomic Safety and Licensing Board panel of three judges held evidentiary
hearings involving contentions and concerns raised by New York State, Riverkeeper, Hudson River Sloop
Clearwater, and other parties opposing relicensing of the applicant Entergys Indian Point Generating
Station. The transcripts of this hearing38 provide several examples of the NRC information problems
which could adversely influence decision processes. (In the following, the location of the discussion in
the hearing transcript is indicated by a T. followed by the page number in the hearing transcript.)
Discussion of differences in results obtained from computer models included testimony about
NUREG-1150 (Nuclear Regulation 1150), which concerns consequences of nuclear reactor accidents.
Origins of certain numbers, input parameters, and other information important to accident cleanup
calculations were questioned. (T. page 2006). This led a judge to ask if a witness if it was his position
that a Sandia study referenced in a NUREG actually looked into a dispersion of primarily plutonium
contamination? (T. page 2007.) A witness identified the referenced Sandia study (also called Os84):
Its Ostmeyer and Runkle, An assessment of decontamination costs and effectiveness for accident
radiological releases, Sandia National Labs. And, again, its stated as To Be Published. (T. page 2008.)
Page 44
However, the Sandia study referenced by the NUREG appeared to have remained unpublished.
An attorney for the state of New York said we searched libraries. I also contacted Brian Harris,
counsel for NRC Staff and requested a copy of Os84. Mr. Harris represented to me that he searched
internal NRC libraries, as well as Sandia libraries, and that no copy of Os84 in any form, draft, or
otherwise could be located, so we do notnone of the parties have a copy of this document. (T. page
2010.)
One hearing judge questioned the situation. Let me ask, if the NRC doesnt have a copy of it,
how can the NRC -- how did the NRC rely on it in developing this guidance on NUREG/CR-3673? How
were you able to reference it if you never had it? (T. page 2010.)
A copy could not be located. Mr. Joseph Jones of the NRC responded to a question about
whether Sandia Lab could provide the study. No, I checked with our librarians, as well, and we were
unable to locate it. (T. page 2011.)
There were guesses about assumptions about the kinds of contamination described in the lost
Sandia study. Dr. Anita Ghosh stated And it does talk about severe accidents, so I guess we would
have no reason to assume that they were looking at a different composition than what a severe reactor
accident would produce. (T. page 2011.)
Judge Michael Kennedy commented on the lost Sandia study. So, okay, we lost the reference
to 1984. Someone convinced themselves, and then the Staff viewed it as reasonableand, again, I
know were not looking at great science, but what Im struggling with is, okay, we lost the reference, but
it is 30 years later, and were trying to deal with a plant-specific analysis for this facility. (T. page
2016.)
Judge Lawrence McDade commented. But what this does is leave us with a degree of
uncertainty as to exactly what the source for the contaminants that are referred to. It may have been
plutoniumbut at this point theres no way to really be sure of that. (T. page 2026.) (Note that
plutonium contamination can have serious implications.)
Its unclear if the lost and apparently unpublished Sandia study was located. Its difficult to
determine from the available transcripts if the questions about values were resolved. Was acceptance
of the values rationalized based on a guess about what the NRC would have done at the time (T. page
2011), on the observation that, so far, the NUREGs have stood the test of time (T. 2251), or on the
basis of responsible and reviewable great science that many would reasonably expect from the NRC
(T. page 2010)? Even if the issues were all resolved in a technically responsible manner, the course of
the hearing did seem to indicate that the NRC had significant problems with information that they
should have maintained. The NRCs failures may have undermined the credibility of the NUREGs that
rely upon the lost information, may have cast doubts on the credibility of hearings like these, and did
lead to the waste of hearing time and the waste of the limited resources of the parties involved.
Page 45
The judges approach to the information problems in the hearing seemed appropriate for a
legal exercise. Would the same approach be appropriate (or even be possible) if answers to question
questions about the origins and the reliabilities of values used to derive or derived from NUREG
information were needed by engineers and scientists charged with making urgent assessments and
decisions that would determine responses to an ongoing nuclear accident? In legal proceedings, there is
time to question NUREG references (T. page 2010), guess about assumptions (T. page 2011), and
consult with librarians (T. page 2011). Lawyers and judges dont need great science for their legal
semantics (T. Page 2016). In event of a nuclear accident, decisions would probably have to be made
quickly. Guesses and assumptions would not be welcome. There might not be time to ask question
librarians about document locations. Decisions would be difficult if information sources did not answer
question like Were these estimates of cleanup effectiveness of a certain set of available methods based
on an accident involving plutonium or just other radioisotopes? Would information whose credibility
was largely based on the fact that it stood the test of time for all of three decades be regarded as
reliable? Decisions determining the heights of the Fukushima sea walls and locations of backup
generators stood the test of time for four decades before those errors were exposed.
Accidents are the times when great science is required. However, the NRC, the courts, and
industry lawyers appear to work to ensure that the best science available in event of a nuclear accident
will fall far short of great science. The resulting doubts and confusion about information will serve
the purposes of parties attempting to evade being held responsible for poor operations, designs,
decisions, and responses quite well.
Different types of NRC information problems were evident earlier in the hearing, when
contentions concerning metal fatigue and flow-accellerated corrosion of pipes were addressed.
Questionable technical information appeared to be the basis for use (abuse?) of hearing rules to exclude
important technical information that was not from the nuclear industry from consideration (T. page
1543). The significance of this problem is that the judges were controlling the course of the hearing,
including what evidence would be allowed into the record that would be relied upon in the licensing
renewal decision.
One contention in the hearing concerned materials degradation due to flow-accelerated

corrosion (FAC) of reactor coolant pipes. Parties disagreed about the effectiveness of the aging
management program. The effectiveness of software, scheduling, and inspection procedures employed
to address FAC was questioned. The NRC, industry, and the parties raising the concerns seemed to
agree that FAC was responsible for significant amounts of degradation at the plant. Testimony and
other evidence seemed to hinge on definitions of what constituted FAC and what degradation was not
FAC. (T. page 1395.) The industrial applicant and the NRC appeared to narrowly define FAC as involving
a relatively simple degradation process that thinned pipes at rates linear with time. Industry pipe
inspection programs included some scheduling that was to be initiated by a relatively simple computer
program that assumed that pipe thinning with time was a relatively simple linear process. In their
procedures, pipe thinning inspections for pipes believed susceptible to nonlinear modes of erosion that
are known to occur synergistically in pipe elbows and other flow geometries involving turbulent flow
Page 46
were to be managed by application of operating experience and engineering judgment. (T. page
1439.) This vague criteria appears to be more appropriate for limiting legal liabilities than for
preventing pipe failures.
Other testimony and evidence raised questions about whether or not, in the real world, the
more complex nonlinear thinning processes could be separated or defined out of FAC. Corrosion and
erosion due to cavitation and other degradation appear to occur synergistically in many systems;
considering them as distinct, separate processes would appear to be, technically, the wrong approach.
Inspection procedures and frequencies under the applicants aging management program did not
appear to be adequate to detect wall thinning due to synergistic corrosion and erosion before failure.
Both parties agreed that the software used for scheduling of inspections for many types of pipes did not
address the nonlinear modes of corrosion involving synergistic erosion and cavitation processes (T. page
1438).
In support of their view, the NRC and industry appeared to rely on vague procedures and
descriptions of how to manage inspections and degradation. Judges questioned statements like
procedures would be consistent with legal requirements for an aging management program were
adequate. (T. pages 1344- 1346, 1368-1371.) Application of operating experience and engineering
judgment appeared to mean that certain things are kind of within the skill of the personnel that are
implementing the program that may be covered by the procedures... (T. page 1372). Kind of...--what
does that mean? NRC staff had a highly questionable response to questions about this procedure and a
four page long Flow-accelerated corrosion program: But I would expect that the FAC engineer
probably has a notebook thats probably about that thick that would have all of the procedures that
exist to implement the program. (T. page 1417.) In event of an accident, would one expect that a
judge probably might not find industry liable if the NRC knew of and approved of such apparently
undocumented and uncontrolled procedures and allowed industry to operate? (Note that a federal
judge in one zirconium case described above ruled that industry did nothing illegal by filing internally
false documentation.19 Does this type of ruling meant that the notebook of procedures thats probably
about that thick could legally contain false information generated to reduce legal liabilities in event of
accidents? ) After the Fukushima accidents, TEPCO blamed NISA for some of the circumstances leading
to the accident. TEPCO was bailed out. NISA was abolished.
This type of testimony may have been effective in supporting legal arguments advanced at this
and other licensing renewal hearings. However, many responsible engineers and scientists would
probably find aging reactor management plans that include statements like operating experience and
engineering judgment and that actions would be consistent with legal requirements for an aging
management program would be too nebulous and vague to be considered reliable procedures.
Objective procedures and criteria are a far better basis for establishing timely inspection schedules and
procedures. Accurate understanding of materials are the basis these types of technical decisions, which
are required to safely operate reactors and to respond to accidents in ways that will take into account
the actual behaviors and conditions of decades-old systems and components. Vague procedures and
company records may limit legal liabilities in event of failures, but they also increase the probability that
Page 47
engineered systems will fail and accidents will actually occur as a result of misunderstandings of the
chemistry and physics of plant degradation.
At some points in the hearing, the parties and the judges appeared engaged in absurd-sounding
semantics and word games. The arguments might sound like great law to a lawyer, but they read like
terrible science to an electrochemist or a materials scientist.
The license hearings are also fundamentally flawed because they limit participation to a few
parties on either side of the issue opposing or supporting the license renewal. Much of the scientific
information is shielded from public review on the grounds that it is proprietary. Unlike a scientific
discussion, which benefits from open input from many parties, the NRC hearing process gives voice to
and, as importantly, limits access to much of the industry information to only a few parties. (That might
be fine if the NRC could limit the health and economic risks to those few parties who profit from them.)
Some parties appear to be follow narrow risk-averse legal strategies to advance positions most likely to
succeed rather than advocate for a decision based on a comprehensive review of the technical issues.
The proceedings sometimes appear to amount to arguments for justifiable caution against arguments
backed by dubiously optimistic projections from industry and the NRC that will maximize industry
profits.
The NRC and the nuclear industry appear to be excluding significant technical information from
the record and appear to be concealing technical information that is in the record from public scrutiny
with claims of proprietary confidential information. The legal decision might well be rendered based
on the legal resources of the opposing parties and the judges that control the course of the
proceedings. They might not be based on open, objective, and complete evaluations of the chemistry
and physics of the materials in the aging reactors.
(Going beyond the specific information issues, are the NRC ALJ hearings appropriate venues for
resolving relicensing disputes involving complex technical issues? While courts and science both employ
logic as a tool, logic is only as good as the facts or premises to which it is applied. Courts have rules of
evidence which are sometimes arbitrarily interpreted by judges to control which facts that the court will
decision in rendering decisions. Courts also rely on precedents for decisions in attempts to maintain
consistencies in application of the law across disparate sets of cases. Courts often apply the legal
principle of res judicata (or a thing settled) and consider Findings of Facts and decisions as final). In
a technical review, such practices would be regarded as scientific misconduct. Although they are
imperfect, scientific reviews attempt to consider facts based on their reliability and do not disregard
facts or remove from consideration due to exclusionary rules or other rules of the process. Issues are
reconsidered when warranted by improvements in understanding and other circumstances -- there is no
equivalent of res judicata in science. Science does not defer to precedents; each review stands on its
own merits. Valid scientific reviews question and critically evaluate the facts used as the basis of past
conclusions when revisiting a dispute. While scientific reviews may include some elements that could
be considered advocacy, scientific reviews are fundamentally different from legal proceedings in that
good lawyers are expected to put clients interests above all else. Lawyers win-at-all costs ethics
Page 48
often leads them to seek exclusion of reliable facts and evidence contrary to their position. Scientists
would regard such behavior as scientific misconduct. Is the NRC committing a fundamental error by
relying on courts that operate under hearing rules that have developed in civil law, not science?)
Based on the number of times that lawyers were able to elicit admissions of uncertainty and
error from the other side, the superficial projection of technical competence and confidence by the
applicant, and the legal logic of the arguments in the transcripts, the applicant appears to have
presented evidence and a set of facts more favorable to relicensing the plant, from a legal viewpoint.
However, the NRC staffs consistent support of the industry that they regulate appears to be one
of the most influential factors on judges controlling hearings like these. Courts have ruled that lower
courts overstep their authority when their ruling contradict NRC decisions. 24, 25 Regardless of the merits
of any particular cases,, the NRC disregard of the need for accurate and reviewable records combined
with the courts deference to the NRC amount to institutionalization of ignorance and are a recipe for
disaster.
What effects do court rulings that refusals to falsify documentation of nuclear errors are
unreasonable, violate no laws, and amount to unwillingness to obey a valid order have on the
integrity of nuclear industry information that will be relied on to mitigate effects of nuclear accidents
and to derive the lessons learned from post accident analysis of industry information? Are
inaccurate, suspect, or lost technical information and false records more likely to cause or prevent
nuclear accidents?
8. Opaque and Incomplete NRC Investigations
Although impaired by the judges lack of technical expertise and reliance on the uninformed
NRC report, court proceedings did lead to the generation of court records that contained the evidence
reviewed. Judicial proceedings and decisions did serve to describe judges perceptions and reasoning,
right or wrong.
In contrast, the NRC investigation of the inaccurate and suspect zirconium cladding information
was opaque. The NRC investigation appears to have consisted of limited interviews, followed by a
whitewash audit supplemented by inaccurate NRC records4 that conflicted with other NRC records3.
NRC officials had indicated that they would conclude their investigation after legal proceedings
were finished. Practically, keeping their investigation open and ongoing appears to kept significant
NRC information concealed from and unavailable to parties in court actions. Some examples of
unavailable information include inaccurate and conflicting NRC audit records 3,4 which demonstrated
that the same problems that caused distribution of inaccurate zirconium cladding alloy information
recurred in analysis of materials used for high level nuclear waste containers3. (These inaccurate NRC
records were found on the NRC web site in 2011. Parts are reproduced in the appendix below.) This
Page 49
raises questions. What other information and activities does the NRC conceal in the course of their
investigations that could be viewed as adverse to nuclear industry agendas?
During discovery, the NRC was informed of inaccuracies in reports sent to the nuclear industry
that were found during discovery, as well as other misconduct. The NRC was on the distribution lists of
legal filings which included exhibits documenting misconduct. Available records indicate that the NRC
took no action to correct bad information or review suspect information. The NRC did not even request
information that would allow retrieval of specific inaccurate and suspect documents for review. The
NRC 2011 letter to Senator Lugar and other documents indicate that some NRC officials responsible for
investigating these problems were not even aware of many of the court records and other information
provided to or by the NRC.
Eventually, the NRC dismissed all allegations against Lambda. When asked about these results
and misconduct, one NRC official stated that the NRC applied a no harm-no foul standard to these
situations. He then indicated that the issue was over. His statements indicated that the entire NRC
process was not intended to find and fix nuclear industry problems, but a delaying tactic and a farce
intended to fix employees and critics troubling the nuclear industry. In the NRCs view, what
constitutes a foul which would motivate them to take action to correct known inaccurate
information? Excessive damage arising from a nuclear accident?
Related NRC audit records indicate that there were gaps in the NRC investigation and
considerable tolerance of inaccurate industry information and faulty QA practices. NRC audit records
also contain false information apparently generated by NRC investigators. However, the NRC did not
provide details of their investigation or audit records to parties who could evaluate these problems and
who recognize these problems. Some relevant NRC records that were generated during legal
proceedings were provided to the nuclear industry, but not to the courts or plaintiffs involved.
Eventually, some relevant NRC records were found on-line.
(Note that excerpts of court decisions, NRC records, and related records are reproduced in order
to describe the circumstances surrounding the attempts to correct the inaccurate and questionable zr
information distributed in the nuclear industry. Complete court decisions 15, 20, 21 and related NRC
records 3,4 are posted on the internet. Different judges arbitrarily relied on different evidence to arrive
at different findings. Prior to the zirconium case ALJ decision, a settlement offer of $60,000 with
restrictions on plaintiffs speech regarding the issues was countered with $90,000 (approximate value of
lost salary and benefits from the entire situation) to open a lab (ie, no restrictions due to anti-
competition) and no other restrictions on plaintiff. A day later, a second alternative counter offer of
$50,000 but requiring an ALJ decision and no restrictions on plaintiff speech was submitted through the
complainants lawyer. Being on the distribution list, the NRC and others should have had access to the
court records at that point. The ALJ issued a decision against the complainant before a response was
received. Appeals over the next five years cited the NRC report to find that it was not reasonable to
view inaccuracies in reports on zirconium nuclear components as safety concerns. Regardless of the
decisions and viewed in the context of the published literature and NRC statements, judicial records
Page 50
indicate that the NRC and the nuclear industry still have significant information problems that could
create serious hazards in event of a nuclear accident. Like the NRC, the judiciary appears to have a
consistent pro-industry bias, although the details in findings of facts that judges arbitrary select from
evidence in the courts records are inconsistent with each other. In every nuclear accident, physics and
chemistry will creates one set of facts. The consequences of many of those facts will be adverse to
all concerned, and no judicial decision will undo the real-world consequences of those facts. Reliance
on bad information in deciding what actions should be taken in response to accidents could very well
make accidents worse. The NRC should pursue inaccurate and suspect information in the nuclear
industry and require that industry review for the errors and maintain only accurate information that will
be of value in event of a nuclear accident and flag suspect information as unreliable. Reliance on
inaccurate technical information can be far more hazardous than recognizing that reliable technical
information is not available.)
9. Inaccurate NRC Audit Records Conceal Recurring Nuclear Industry Problems.
The NRC itself generated inaccurate information that conceals the recurring nature and
significance of QA problems that led to the release of erroneous reports. The NRCs tolerance of
inaccurate information, faulty procedures, deficient Quality Assurance (QA) practices, unreliable
software, and false records is irresponsible behavior for a regulatory agency. However, NRC audit
records found in 2012 in the course of preparing this document indicate that NRC itself has actually
generated inaccurate records that could conceal industry problems and has generated records that
intentionally and disingenuously decouple industry deficiencies from regulatory requirements. These
significant regulatory failures demonstrate that the NRC has systemic information problems.
A. The NRCs 1999 report clearly states that errors in Lambda reports were due to QA
problems, yet the NRC took no effective actions at that time.
(Source: NRC Allegation NRR-1999-A-0057 Concerns 1 thru 3. Texture Analysis of Zirconium Alloy.1 )
B. Later, NRC observation of an audit of Lambda high level nuclear waste container materials
analysis noted several of the same types of procedure, training, software, record, and QA problems
that led to the earlier-identified dissemination of inaccurate Zr data in the nuclear industry.
In 2001, Lambda was audited to evaluate its implementation of QA in work unrelated to their
error-prone Zr texture analysis.3 The NRC, DOE, and Bechtel SAIC participated in this 2001 audit. The
purpose of the audit was evaluation of QA for the residual stress measurements of welded mockups
supporting the design of the high-level waste containers for nuclear waste repositories. The work was
related to Yucca Mountain.
Page 51
The 2001 NRC Observation Audit Report No. OAR-02-01 Observation Audit of the Bechtel
SAIC LLC Audit No. BSC-SA-01-30 of Lambda Research Inc. states that NRC auditors were at Lambda in
2001:
(Source: US NRC Observation Audit Report No. OAR-02-01, NRC web site file ML013330072.pdf. 3 Parts of this audit record are reproduced in
the appendix below. Note that Lambda occupied only one small building located in Cincinnati, Ohio.)
This 2001 audit identified deficiencies in training, procedures, software, and records while
observing testing activities at Lambda:
(Source: US NRC Observation Audit Report No. OAR-02-01, NRC web site file ML013330072.pdf. 3)
These deficiencies in the nuclear waste repository work are the same types of problems that
were noted in the earlier NRC in Texture Analysis of Zirconium Alloy report1, quoted above. Residual
stress analysis is different than zirconium texture analyses, but stress analysis relied on some of the
same procedures. The same Lambda QA system covered residual stress analysis, zirconium analysis, and
other analysis done at Lambda.
As a result of this audit, some corrective actions were recommended. The QA deficiencies that
contributed to the generation of inaccurate zirconium texture analysis information and the distribution
of inaccurate reports were not mentioned in this NRC audit report. Lambda appears to have been
allowed to continue QA work on waste containers.
An incredible irony apparently lost on the NRC was that Lambdas QA-deficient work for Bechtel
was itself QA support. What good is QA support if it is based on bad QA? Is this an example of the type
of defense-in-depth safety that the NRC and nuclear industry claim that they practice? Does bad QA
guard against bad QA? Or is it closer to intentional incompetence that would be very convenient and
useful in the finger-pointing and blame shifting that often follow accidents as companies and their
lawyers exploit every inconsistency in records in legal maneuvers intended to avoid liabilities. From a
standpoint of public safety, which is worse -- incompetence or corruption? (It may also be interesting
to note that Lambda marketing literature claimed that Lambda was certified by all major corporations in
the nuclear industries for meeting their specific criteria for laboratory testing.)
(Its also significant to note that, as described in Section 7, a federal judge ruled that generation
of an internal false QA record did not violate any federal, state, or local laws.19 The NRC, GE Nuclear,
and other nuclear industry audits rely in large part on review of internal QA records of the company
being audited. Is the NRC audits tolerance of poor QA setting the NRC up to be held responsible in
Page 52
event of an accident? After Fukushima, some TEPCO officials offered the defense that TEPCO only did
what Japans nuclear and industrial safety agency allowed TEPCO to do. Do demonstrably inaccurate
records of a subcontractor limit legal liabilities of larger client companies in event of a nuclear accident?
Its interesting to note that this judge relied on the earlier NRC report indicating that the NRC would
not pursue whether distorted results were reported to Lambda clients1 for some findings in the
decision indicating that filing false internal QA records was not illegal19.)
C. Misleading information and inaccuracies in NRC records of a later audit of Lambda zirconium
analysis have the effect of concealing a history of recurring QA problems that caused dissemination of
inaccurate information about nuclear fuel rod materials in the nuclear industry.
On September 27-28, 2004, the Nuclear Regulatory Commission (NRC) audited portions of
Lambdas Quality Assurance (QA) program and effectiveness of 10CFR Part 21 controls.4 The2004 audit
examined a Quality Assurance Incident Report of problems with crystallographic texture analysis of
zirconium alloys for GE Nuclear.
Comparisons of 2004 NRC audit records of Lambda Research texture work with 2001 NRC audit
records of Lambda Research high level nuclear waste container stress work indicate that the NRC
produced audit records containing serious false and misleading information. This NRC production of
false information in NRC audit records has the effect of concealing the recurring nature of nuclear
industry QA problems. Concealing such problems amounts to tolerating their recurrence. Industry
technical and QA information becomes increasingly unreliable under such conditions, sometimes
apparently being corrected only when problems are discovered by external parties due to a failure and
industry is forced to fix problems. In event of an accident or incident, follow-up investigations intended
to prevent recurrence of the accident or incident could be compromised by conflicting information, as
occurred in the UA 232 accident investigation22.
This 2004 NRC audit record contains false information that conceals the recurring nature of
QA problems.
The NRC 2004 audit reports part 2.0 Status of Previous Inspection Findings is inaccurate.
(Source: NRC Inspection Report 99901345/2004-201. NRC web site file ML042750057.pdf. 4 Portions of this document are reproduced in the
appendix below.)
The statement There were no NRC inspections or technical reviews performed at the Lambda
facility prior to this inspection is both false and misleading. The NRC observation audit at Lambda
Research in 2001 was both an NRC inspection and an NRC technical review of some of that same
Page 53
procedures and QA at the same Lambda facility in Cincinnati, Ohio. The statement is misleading in that
it conceals deficiencies in Lambda QA found in the earlier Bechtel audit of the same QA system. The
Bechtel audit records clearly indicate that NRC observers observed Lambda employees execute
procedures at Lambda, and that this earlier audit of the same Lambda QA system identified deficiencies
that the NRC also noted in the earlier NRC report on the zirconium analysis.
D. This 2004 Audit Record Disingenuously Decoupled Deficiencies in Procedures, Training, Records,
and other QA Problems Described in the QA Incident Reports Describing Errors in Zr Texture Analysis
From 10CFR21 and 10CFR50b Requirements that Applied to Safety-Related Services.
(Source: NRC Inspection Report 99901345/2004-201. 4 NRC web site file ML042750057.pdf. Some relevant portions of this audit record are
reproduced below in Appendix Part D.)
(Source: NRC Inspection Report 99901345/2004-201. 4 NRC web site file ML042750057.pdf.)
Note that this QAR is the same QAR that Lambda's owner demanded be edited to include false
information about the GE Nuclear texture analysis errors. This NRC audit record explicitly decoupled
safety-related activities from Lambdas history of repeatedly releasing bad nuclear fuel rod cladding
alloy information by defining safety-related activities as those reports whose POs explicitly cited
10CFR21 or 10CFR50b requirements. This NRC record appears to support the industry claim that the
inaccurate information is not safety -related, despite the fact that the review of the errors, past analysis
work, GE Nuclear audits, and the scientific literature indicate that the inaccurate information is safety-
related. This NRC audit record could shield industry from liability.
The NRC audit records fail to note the fact that court records and Lambda Research records
demonstrate that the same defective procedures and software described in this QAR were used by the
Page 54
same employees and other employees with the same training to do analysis for other GE Nuclear
projects that did explicitly cite 10CFR21.
The NRC audit fails to note that Lambdas owner prevented reviews of that previous GE Nuclear
and other work by making firing threats.20,21 Some POs from that previous work did explicitly state that
10CFR21 applied. The same defective procedures, software, training , and QA that produced errors
were used for that previous work.
The NRC report also fails to mention that GE Nuclear NIAC audits of the same deficient
procedures and QA practices that led to the errors did explicitly state that 10CFR50b and 10CFR21
applied to the work and were included in those audit scopes.
To summarize, the NRC audit report4 fails to note the following: 1) 10CFR21 applied to past GE
nuclear work done with the same procedures, software, training, and other QA elements found to be
defective. 2) Lambdas owner changed versions of the QAR so that contained false statements that GE
Nuclear had been notified of the defects. 3) Lambdas owner directed firing threats at employees in
order to prevent reviews of reports for errors and to prevent discussion of defects in procedures,
software, training, and QA. 4) Using the NIAC standard, GE Nuclear audited the same defective
procedures, software, training, records, and related QA that contributed to errors. The GE Nuclear NIAC
audit scopes did explicitly indicate that 10CFR21 and 10CFR50b applied to the Zr texture analysis. These
GE NIAC audits failed to detect deficiencies, failed to note inaccuracies, failed to include contacts with
analysts actually doing Zr texture work, but did include contacts with technicians who were not doing
the texture analysis.
The 2004 NRC audit records did not reference or mention the NRCs 1999 report1 describing the
inaccurate zirconium texture analysis information sent to industry and QA problems. (It is also
interesting to note that the 1999 report was not posted on the NRC web site. In fact, the NRC requested
a copy of this report from those involved in the case in mid-2011, just months before they wrote their
letter to Senator Lugar citing that report. Did the NRC lose or destroy their copies of the 1999 report on
Lambda Zr analysis errors? The NRCs 1999 report appears to undermine conclusions in the later NRC
audit records. What other records documenting nuclear problems has the NRC lost or destroyed? )
A major NRC regulatory failure is evident in the failure of the NRC audit to address the failures
of multiple GE Nuclear NIAC audits to detect the deficiencies in those same procedures, software,
records, and training that the NRC found to have contributed to the fuel rod cladding analysis failures
which caused distribution of inaccurate zirconium information to the nuclear industry. Available NRC
audit records do not mention the Lambda records indicating that the GE Nuclear/Global Nuclear Fuels/
Nuclear Industry Assessment Committee (NIAC) audits of Lambdas Zr texture analysis procedures,
software , and QA system did explicitly cite 10CFR21 and 10CFR50b. NRC records indicate and court
records confirm that these QA deficiencies actually did contribute to distribution of inaccurate zirconium
information within the nuclear industry. Court records demonstrate that these NIAC-audited
procedures were used for analyses work having POs that did explicitly state that 10CFR21 applied to the
Page 55
work. The GE Nuclear NIAC records from several audits were provided to the NRC investigators
conducting the audit prior to their audit. The NRC failed to explain how or why GE NIAC audits missed
the deficiencies that led to repeated errors in reports before and after GE NIAC audits.
NRC audit records, the NRC reports, and the 2011 NRC letter to Senator Lugar ignored literature
explicitly describing bad zirconium texture effects on serious safety issues, including alloy degradation,
and component failures involving corrosion, radiation damage, and welding quality. 5-14, 17,18, 32, 33 (Some
notable studies were published prior to any of these NRC documents.32,33 While some of the scientific
papers cited were published after some of the NRC documents, all were published prior to the 2011
letter to Senator Lugar.) The NRC audit records do not indicate if the problems with Lambda software,
training, and procedures that caused release of distorted texture information to nuclear industry clients
which were described in the audited QARs were ever corrected. While GE Nuclear was provided with
copies of the defective texture procedures, no records were produced that indicated that GE Nuclear
had ever been notified of defects in the procedures. No records of objective reviews of past zirconium
texture reports for distorted results were ever produced.
Despite finding 10CFR21 violations (and apparently dismissing audit records demonstrating that
the examined QAR records indicated that the same problems could have led to previous releases of
inaccurate reports to the nuclear industry and that previous NIAC audits missed the QA problems prior
to generation of some of the inaccurate report, as well as threats that included firing and other
retaliation for refusing to participate in falsification of records), the NRC minimized the concerns and
found that Lambda QA was "generally acceptable with some exceptions".
(Source: NRC Inspection Report 99901345/2004-201.4 NRC web site file ML042750057.pdf.)
To summarize, the 2004 NRC audit generated records that : 1) contained false
information, 2) concealed previous audit history describing the recurring Lambda QA problems related
to nuclear work and 3) decoupled Lambdas release of faulty nuclear fuel rod information from safety
concerns, 4) excused blatant 10CFR21 and 10CFR50b violations and related firing threats.
10. Accurate Records and Good QA Are Essential to Failure Analysis and
Error Identification.
The NRC conducted or participated in audits of Lambda Research which indentified their many
QA deficiencies, including inaccurate and inconsistent record keeping. Most labs and auditing bodies
have zero tolerance policies towards known inaccurate information and known bad QA. However, the
Page 56
NRC not only repeatedly did not require review or correction of recurring QA problems and record
inaccuracies that led to actual errors in reports, but the NRC itself also generated records containing
false statements that appear to intentionally have the effect of concealing inaccurate information and
concealing the recurring nature of the QA problems. NRC audit records also decoupled explicit
industry radiological safety warnings from identified deficiencies and errors in reports sent to clients,
including GE Nuclear.
This type of behavior in the nuclear industry is both dangerous and scientifically irresponsible.
Any one of these irresponsible NRC actions could have the effect of concealing nuclear information
inaccuracies and defects. Inaccurate information could lead to bad decisions that could cause accidents,
that could make ongoing accidents worse, that could compromise safety, and that could undermine
improvements in designs or operator procedures intended to reduce the possibility of accidents and
mitigate accident effects.
Uses of information determine quality requirements for that information. When the
information will be relied upon for design, manufacturing, or operations, information quality should be
known and match the intended uses. However, the potential uses of information to make decisions
during nuclear accidents and during subsequent nuclear failure analysis investigations cannot be
anticipated. Therefore, the quality and reliability of technical information in the nuclear industry must
be understood. Limitations of data must be known since all potential uses of data are unknown.
Information known or suspected of being inaccurate should not be tolerated.
Failure analysis investigations often serve as sources for lessons learned that could, if
accurate, prevent or improve the effectiveness of responses to future accidents. Flawed investigations
or lessons learned can lead to poor decisions and unintended creation of additional hazards.
A. The UA 232 Accident Investigation: Inaccurate Records, Conflicting Documents, and

Materials Defects.
Very clear and notable examples of inaccurate records impairing corrective actions intended to
reduce the possibility recurring accidents and impeding accident investigations are found in the aviation
industry. The FAA accident investigation report of United Airlines Flight 232 crash at Sioux City describes
how inaccurate and conflicting records impeded urgent responses required to avoid recurrence of the
accident and undermined failure analysis investigations needed to determine responsibilities. 22
In this accident, a GE engine disk exploded in flight due to a materials defect. In an event
characterized as one in a billion, the explosion severed critical hydraulic lines, rending critical aircraft
control systems useless. Extraordinary efforts by the airliners crew enabled them to save many lives by
crash landing the crippled jet in Sioux City Iowa, but over 110 passengers were killed and many were
injured.
Page 57
Examination of the wreckage indicated that rotational forces caused the explosion of the
defective disk, initiating the accident. Defects were present in the disk when it was manufactured about
two decades earlier. Conflicting and missing records of that defective disk impeded the post-accident
failure analysis investigation.
The FAA report described doubts created by conflicting manufacturing records that might have
led to inappropriate prioritization of grounding and inspection of other airliners likely to have defective
disks. If airliners with defective disks continued to fly, engine explosions could have caused repeats of
the accident.
It is important to note that the inaccurate records and the defective disk they purport to
describe appear to have been produced about two decades before the accident occurred: the defective
disk and flawed information literally amounted to two decades of an accident waiting to happen.
The crash disk had the serial number of a disk which some manufacturing records indicated had
an anomaly. Records on the disk with the anomaly are inconsistent and incomplete.
(Source: National Transportation Safety Board Aircraft Accident Report United Airlines Flight 232 McDonnell Douglas DC-10-10
Sioux City Gateway Airport Sioux City Iowa July 19 1989 PB90-910406Y "NTSB/AAR-SO/06"22. The No. 2 engine in the accident airplane
exploded and severed aircraft control lines, initiating the UA 232 accident.)
Doubts about the origin of the disk and conflicting records led to doubts about whether the
accident disk and other airliner engine disks made at the same time were composed of titanium that
was suitable for use in rotating parts like fan disks.
(Source: National Transportation Safety Board Aircraft Accident Report United Airlines Flight 232 McDonnell Douglas DC-10-10 Sioux City
Gateway Airport Sioux City Iowa July 19 1989 PB90-910406Y "NTSB/AAR-SO/06"22.)
Page 58
The investigation could not determine if all disks suspected of having defects similar to those
that caused the crash had been removed from operating aircraft.
(Source: National Transportation Safety Board Aircraft Accident Report United Airlines Flight 232 McDonnell Douglas DC-10-10 Sioux City
Gateway Airport Sioux City Iowa July 19 1989 PB90-910406Y "NTSB/AAR-SO/06" 22.)
(Note that the probability of an accident involving loss of all flight controls due to severing of the
redundant hydraulic lines was characterized as one in a billion. Japan Airlines Filght 123 had crashed
just a few years earlier when its hydraulic lines were severed, also rendering critical aircraft control
systems useless.26 The estimation of loss of all flight controls due to the severing of all hydraulic lines as
one in a billion appears to have been inaccurate; this underestimation of the risk appears to be an
example of risk management based on bad technical information. Accidents whose frequencies were
described as one in a billion have occurred far too frequently in the aviation industry. Accidents
whose projected probabilities are inaccurate due to reliance on bad technical information probably
could also occur repeatedly in the nuclear industry.)
Eventually, several disks containing the same defects as the UA 232 disk were identified and
removed from service prior to explosive failure. If they had not been identified prior to failure, what
impact could their failures have had on the risk analysis estimates for this type of accident?
B. Accurate Records and Responsible QA Practices Have Led to Identification and Correction
of Inaccurate Information in Nuclear Industry Reports.
Several reports containing 90 degree rotation texture analysis errors were identified through
QA records. A Lab and QA manager was rotating results by 90 degrees to make them look right.
Results in some reports were literally tipped on their sides.
Lambdas owner testified about 90 degree rotation errors in reports:
Page 59
( Source: Case No. 2000-ERA-0035. ALJ Hearing. testimony. Department of Labor, Office of Administrative Law Judges (ALJ). 31 ALJ Testimony
of Lambda Owner. Perry was Lambdas QA manager and had been the manager of the lab where zirconium texture analysis was done. Note
that court records document that Perry was the previous manager of the laboratory. He trained analysts in texture analysis, popLA, and
other XRD analysis work. He also was the QA administrator. The rotation described was the 90 degree rotation error described in the NRC
report1.)
Note that changing data so that a result looks right to correct for making some mistake is
generally considered to be fudging data in a way not acceptable in responsible laboratories. Any
suspect result is usually checked so that it is understood, not arbitrarily changed so that it looks right.
Changing results in this way amounts to falsification and is not regarded as acceptable behavior in
science.
Initially, rotation errors were found in one report.
When asked about this inaccurate report, the manager making the errors recognized that all of
his past reports were suspect.
(Source: Lambda Internal Email. NRC web site document ML011430464.pdf. 34)
Following the record trail led to identification of other suspect reports. Specifically, past reports
prepared by this manager were reviewed.
Emails describe a limited review of past reports conducted by Lambdas texture expert as best
he can. This review identified several reports with the 90 degree rotation errors, referred to and
described as having an extra 90 degree rotation, rotated, skewed, incorrect, the mistake,
definitely wrong, and difenitely wrong (sic) in this Lambda email.
Page 60
(Source: Lambda Internal Email. NRC web site document ML011430469-1.pdf. 35 Note that inaccurate results were not limited to nuclear
materials. For example, this email shows that Lambda results didnt accurately indicate which way is up in reports describing analysis of
aircraft materials in Wright Patterson Air Force Base (WPAFB) reports.)
Accurate records and following good QA practices led directly to identification of all past reports
done by the manager as suspect. As his email states, then all others are also wrong. Suspect reports
were reviewed for errors. Nuclear industry clients, including Westinghouse, were issued corrected
reports for zirconium texture analysis. (Note that the above email does not describe reviews of GE
Nuclear reports.)
Westinghouse had not recognized the errors in their reports and had not requested that
zirconium alloy texture results be checked. It is interesting to note that, around the same time,
Westinghouse did request a revised zirconium report to correct a misspelling. (What do these practices
and failures involving Westinghouse say about Westinghouse competence? What do these failures say
about Westinghouse information and software? Spelling is important, but does Westinghouse prioritize
expenditure of attention and resources towards such superficial appearances or prioritize the accuracy
of technical information that could influence decisions and understanding of nuclear materials?) NRC
records appear to indicate that the senior engineer of one of the major nuclear fuel fabrication
companies who the NRC investigators consulted was a Westinghouse engineer. Other records indicate
that the Westinghouse client who received and apparently repeatedly failed to recognize that some
information was definitely wrong was a Westinghouse scientist who worked with that NRC consultant.
Page 61
In summary, previous inaccurate reports were identified because 1) it was known that the
Lambda QA and lab manager was rotating results by 90 degrees and 2) unlike the situation with GE
records related to the UA 232 crash, accurate internal records indicating which work was done by that
Lambda manager enabled identification, review, and in many cases, correction of suspect reports. (This
situation should not be taken to imply that Lambda QA did not have serious deficiencies. Note that
other information indicates that errors remained uncorrected in some Lambda reports and that related
communications with clients contain fraudulent information concealing the nature (and thereby the
extent) of errors.)
Records inaccurately indicating that the 90 degree rotation errors were due to technician errors
would mislead an investigator attempting to trace origins of inaccurate information into an
inappropriately limited review of that technician's work that could miss the source of the errors entirely.
Some reports containing errors made by the manager might not be reviewed if investigators were
looking only at reports involving that technician. Misleading records of this type would impair
prioritization of information review and could delay urgent actions required to respond appropriately to
a nuclear incident or accident.
Note that the NRC report1 and at least one court decision15 inaccurately stated that the sample
(not the data) was rotated 90 degrees; sample rotation inaccurately indicates that a technician, not a
manager, would have made the errors, per Lambdas QA system and records. Also, rotating data is a
fundamentally different process than physically rotating a sample. Note also that, after the above
emails were sent, management prepared and sent false information about these errors to nuclear
industry clients stating that the 90 degree rotations were due to technician errors, which makes the
problems seem less severe than the systemic errors due to the QA and lab manger arbitrarily changing
results to make them look right. During an accident investigation, would flawed analytical work that
had been altered to look right be detected?
In the case of the rotation errors, some results were recalculated or regenerated. Corrected
reports were issued to clients. (Records do not indicate that errors were found in GE Nuclear zirconium
texture analysis reports prepared by this manager. The inconsistencies between the GE Nuclear and
Westinghouse texture analyses results that were obtained using the same procedures are not clear.) It
is also interesting to note that records from a GE Nuclear/ NIAC audit of Lambdas zirconium nuclear fuel
rod cladding texture analysis procedures conducted between the date that the first rotation error was
found and the dates on corrected reports do not mention the rotation errors, but audit documents in
court records do appear to indicate that the rotation error should have been disclosed to GE and noted
in NIAC audit records. Also note that the distorted GE Nuclear work discovered soon afterward involved
the same procedures and the same technician as the work containing rotation errors.
Later, texture analysis records sent to nuclear industry clients were found that contained both
false information that conflicted with the earlier email reproduced above35 and additional errors.
Page 62
Its also significant to note that Lambda documents indicate that Lambda management knew
that this manager had caused problems with this analysis years prior to the discovery of the 90 degree
rotation error. Lambda QA records indicate that this manager also trained other analysts and
technicians in the analysis. However, the managers approach to arbitrarily changing data to match
expectations appears to have arisen from a misunderstanding about appropriate laboratory conduct
rather than an attempt to intentionally falsify information to order to conceal errors. His email 34
indicates that he was helpful in identifying suspect reports. Intentional record falsification to conceal
defects involved other management personnel.
C. Zirconium Texture Analysis Distortion Errors: Management Refusals to Allow Reviews Led to
Persistence of Inaccurate and Suspect Information in the Nuclear Industry.
The second type of zirconium nuclear fuel rod cladding texture analysis error Involving data
distortions was identified shortly after the rotation errors were corrected. One report containing this
type of distorted data was sent to GE Nuclear. These distortions and the QA deficiencies which caused
the distortions were discovered about a month after the 90 degree rotation errors were corrected and
just months after the GE Nuclear NIAC audit of the Lambda texture analysis procedures and QA system
for compliance with 10CFR50b and 10CFR21. The owners hostile reaction to investigation of these
problems impeded, delayed, and, in critical aspects, prevented review and disclosure of these distortion
problems and corrective actions required to correct recurring errors and related QA problems. 15,19-21
Note that results with distortion errors can make samples appear to have less texture.12 Note
that some published work indicates that some zirconium components having less texture are better
suited for some nuclear applications than components with greater texture. Thus, texture distortion
errors could make some alloys appear to be more suitable for nuclear applications than they actually
are. Technically, the NRC reports statement that texture procedures producing the errors cannot be
used to qualify substandard material1 appears to be incorrect in some applications.
In an attempt to prevent recurrence of the 90 degree rotation error, management had Lambdas
texture expert write software to automate some of the data acquisition and data analysis tasks.
Unfortunately, this automation split up responsibilities and oversight of sample preparation, data
acquisition, data analysis, and report writing tasks; previously, these tasks were supposed to be done in
one lab under one fully- trained supervisor, which made QA more reliable. The net effect of the
changes intended to prevent recurrence of rotation errors prevented recognition that atypical results
were actually inaccurate results.
Lambdas texture expert described some distorted results as indicating that one specimen was
atypical and exhibited a texture that is more complex than texture found in other specimens.
Page 63
(Source: Case No. 2000-ERA-0035. ALJ Hearing. testimony. Department of Labor, Office of Administrative Law Judges (ALJ). Testimony of
Lambda Texture Expert31).
(Note that writing in a report the texture is more complex and a specimen was atypical from
what is generally seen in zircolloid tube specimens clearly indicate that the report intended to describe
the sample --ie, not the data. The objection seems to have been intended to prevent the witness from
confirming that his report specifically stated that a physical specimen, not data, was being described in
the report. The witness did confirm this fact; the judge was too slow to stop the witnesss response
confirming his statements in his report whose content he described in his testimony. Whats more
importantlegal gamesmanship or finding out what happened? Does the NRC direct those who refuse
to cooperate with industry misconduct into legal venues where judges can arbitrarily interpret rules and
complex technical issues that they dont understand, thereby preventing disclosures by trumping
evidence and common sense with preemptive applications of hearing rules? )
Problems causing texture distortions in Lambda report were identified after GE Nuclear
requested that texture analysis of the zirconium nuclear fuel rod cladding sample be repeated in order
to determine if reported atypical results described in the report as indicating more complex
zirconium nuclear fuel rod cladding texture were correct.
The atypical results were repeatable if Lambdas analysis procedures were used on the original
specimen prepared by the technician (who was trained to prepare samples by the same Lambda
Page 64
manager who had made the 90 degree rotation errors). However, since the original specimen was not
flat, and since sample preparation was part of repeating the analysis, the lengthy process of preparing
a new analysis specimen from the same sample for analysis was also started. GE was kept informed of
these activities.
Texture analysis results from this flatter new specimen were typical and did not match the
original distorted results, demonstrating that the original distorted results first reported to GE Nuclear
on this zirconium alloy nuclear fuel cladding material specimen were incorrect. (Note that no tube
specimens prepared by this Lambda procedure were ever perfectly flat; they were visibly wavy and
bent. Also, the procedure did not include any measurement to determine if specimen flatness were
adequate for accurate analysis.)
Court testimony from the owner of Lambda indicated that the texture results originally
described in the report as being from an "atypical specimen" having more complex texture were
"physically impossible":
(Case No. 2000-ERA-0035. ALJ Hearing. testimony. Department of Labor, Office of Administrative Law Judges (ALJ).31 Testimony of Lambda
Owner. Note that his testimony about This report refers to the same report written by and described in the previously quoted testimony of
Mike, Lambdas texture expert. ).
GE Nuclear was issued a revised report correcting the one particular distortion error.
Reviews of Lambda procedures for the origins of the distortion errors indicated that failures in
training software and procedures had contributed to the distortion error. Since the same technician
used the same procedures and software to produce earlier specimens and reports, other distortion
errors could have occurred.
The original prepared specimen was not flat. Preparing flat specimens from tubing was difficult,
expensive, and time consuming. The procedures did not always work. The technician who prepared the
specimens was asked how he checked to see if specimens were flat within 0.002 inches.
Page 65
The technician testified that he used visual inspection to determine if specimens were sufficiently flat
for texture analysis:
(Case No. 2000-ERA-0035. ALJ Hearing. testimony. Department of Labor, Office of Administrative Law Judges (ALJ). 31 Testimony of Lambda
Technician. It after the first Q (for question) refers the edited QAR16 excerpted above.)
The technicians testimony reproduced above indicates that his method to determine if a
sample was flat enough was Look at it and decide whether or not you think if its good enough. The
owners QAR edits indicating that flatness measurements were already present16 indicates that he
approved of eyeballing flatness measurements. However, good QA system procedures include both a
method and specific criteria to determine if a measurement is good enough; specimens prepared with
a procedure that included adequate flatness measurements would be easy to evaluate. Later, the
technician confirmed that the procedure indicated that specimen surfaces had to be within a 0.002
inches (or 0.002 mils) range to be considered flat enough for texture analysis:
(Source: Case No. 2000-ERA-0035. ALJ Hearing. testimony. Department of Labor, Office of Administrative Law Judges (ALJ). 31 Testimony of
Lambda Technician. Here refers to the procedure.)
Page 66
When pressed, Lambdas technician admitted that his look at it and decide if samples were
good enough method was not that good in determining if samples met the 0.002 flatness criteria:
Lambda Technician. )
This technician had previously done other analyses on zirconium for the nuclear industry.
Reviews of the procedures actually used indicated that tube specimens were very difficult or impossible
to flatten to within 0.002 inches using Lambdas procedure. The procedure described no measurement
procedure to measure specimen flatness. It is also significant that Lambda records showed that this
technician had been trained in the sample preparation procedure by the same Lambda manager who
made the 90 degree rotation errors. Apparently, the trained technician was as comfortable with
eyeballing 0.002 inch measurements as the Lambda QA manager was with rotating results 90 degrees
to match expected results.
These facts indicate that previous analysis involving this technician should be reviewed for the
same distortion errors, just as previous analysis by the QA manager were reviewed for 90 degree
rotation errors. As described previously, Lambda management refused to allow reviews of previous
reports for distortion errors or allow procedures to be changed to include actual flatness measurements.
In, fact Lambda management demanded preparation of QA records falsely indicating that suspect
reports had been reviewed and inaccurately describing procedure changes as already present.15, 16
Zr Texture Analysis Procedure and Software Problems Contributed to Production of Distorted Results.
Lambda records demonstrated deficiencies in texture analysis software and other procedural
problems. Lambdas owner testified:
Lambda Owner.)
Page 67
Lambdas owner was asked about other texture procedure and software QA issues:
Lambda Owner. Note that Embat was software.)
Lambdas owner described the data collected by the automated software for GE Nuclear as just
nonsense because data was not collected properly:
Lambda Owner.)
Later, distorted texture analysis results were found in another report sent to the nuclear
industry. The NRC was notified, but apparently took no action.2 Circumstances indicated that the
results were used for alloy design work and may have been published. There is no available evidence
that some distorted results were ever corrected.
Note that the owner himself reviewed and released reports containing distorted results.
Lambdas owner 1) demanded removal of any mention of software and written procedure problems
from the QA Incident report. 2) demanded removal of mention of the lack of any measurement
procedure or objective criteria to determine if the specimen was flat enough for analysis. 3) refused to
allow reviews of past reports for errors arising from these deficiencies or notification to GE Nuclear that
the Lambda procedures and software audited under the NIAC standard were deficient.15, 16
The NRC report lists the QA problems that led to the errors:
Page 68
(NRC Allegation NRR-1999-A-0057 Concerns 1 thru 3. Texture Analysis of Zirconium Alloy1. )
These are serious problems. In todays labs dominated by computer-controlled

instrumentation, one poorly trained technician can generate and widely disseminate large amounts of
inaccurate information very quickly, especially if the technician relies on faulty software and procedures
whose output he does not understand. Errors escape when technicians cannot recognize nonsense
and physically impossible results. Personnel records and other evidence21 indicated that
management regarded data collection instruments idled because technicians were busy checking and
analyzing data as wasted resources. The lab manager was told that this situation amounted to stealing
from the owner. Detection and retrieval of massive amounts of erroneous results can be difficult under
some circumstances. Repeating analysis could cost as much as $1000 per sample. However, costs of
correcting these problems would pale in comparison to the cost of an accident.
Objective Reviews of Past Work Were Never Done.
These problems indicate that inaccurate results and information about zirconium nuclear fuel
rod cladding and other materials were being distributed to the nuclear industry due to systemic QA
problems. Later audits conducted by or involving the NRC found the same types of problems with
Lambda and its procedures again, well after the zirconium texture analysis problems were initially
identified.
Yet, no reviews or corrections of past reports were done.
Attempts at accurately documenting the problems with recommendations to review work and
notify clients led to firing threats. 15
Courts did find that the reports were inaccurate. 15 Courts found that intentional preparation of
false reports constituted no violations of law. 19
Courts indicated that refusals to falsify reports or sign inaccurate reports were unreasonable 20
and amounted to unwillingness to obey valid orders21.
D. Poor QA and inaccurate records undermine failure analysis and error identification. The NRC
audits not only repeatedly excused recurring QA and records problems that led to actual errors in
reports, but the NRC itself also generated records containing false statements that could have the effect
of concealing the recurring nature of the records and QA problems. NRC audit records also decouple
explicit industry radiological safety warnings from identified deficiencies.
Page 69
The NRCs behavior is both dangerous and scientifically irresponsible. Any one of these NRC
actions could have the effect of concealing nuclear information inaccuracies and defects. Inaccurate
information could lead to bad decisions that could cause accidents, make accidents worse, compromise
design efforts, and undermine improvements in designs or operator procedures intended to reduce the
possibility of accidents and mitigate their effects. Conflicting and missing records impede failure
analysis investigation and leave doubts about effectiveness of corrective actions and the value of
lessons learned, as the FAA report of the UA 232 accident investigation stated.
(Source: National Transportation Safety Board, PB90-910406 NTSB/AAR-90/06, Aircraft Accident Report United Airlines Flight 232. 22 )
Technical information and company records known or suspected of being inaccurate should be
reviewed and corrected when they are recognized as suspect or inaccurate, not after an accident.
Accidents are not the time to ask Can we trust the data? Post- accident investigations, during which
companies sometimes attempt to protect profits and evade responsibilities, are not the time to discover
ambiguous technical information and records. After accidents have occurred, ambiguities about
information only serve to help parties whose irresponsible actions have contributed to accidents avoid
being held liable. Advocates, such as lawyers, are motivated to exploit such questions in their attempts
to serve irresponsible parties. Requiring industry to keep accurate records is cheap compared to the
serious consequences and expenses of nuclear accidents.
The NRC has been informed of known bad technical information and company records that are
still present in the nuclear industry. The NRC has not only decided not to require industry to correct this
information, but has actually generated false information that will muddy the water in ways that could
mislead anyone attempting to find out what actually happened and correct the problems. NRC
generation of false records makes it complicit in industry misconduct.
Like dubious records, irresponsible actions by regulatory agencies can shield industry being held
accountable for their actions. Industry can risk highly profitable optimistic behavior if the public pays
when the nuclear industrys optimistic projections prove to be unrealistic.
After Fukushima, some TEPCO officials blamed lack of preparations and failures during their
accident on Japans Nuclear and Industrial Safety Agency (NISA), stating that TEPCO only did what that
Page 70
agency allowed them to do. In the aftermath of the accident, large areas of Japan remain abandoned
due to high levels of radioactivity. TEPCO was bailed out. NISA was abolished. The public is paying for
massive cleanup efforts. Many residents will not be able to return home for decades, if at all.
Fukushima shows that positioning regulators to serve as scapegoats reduces industry liabilities.
Are NRC employees and industry positioning the NRC to serve as a scapegoat in the event of a serious
nuclear accident?
11. Conclusions and Recommended NRC Actions.
1. The NRC must require industry to correct the specific zirconium nuclear fuel rod cladding alloy
information known to be inaccurate.
Inaccurate technical information about zirconium nuclear fuel rod cladding alloys can create
significant hazards, as described above. Correction would be very inexpensive and fast. A main barrier
to correction appears to be NRC and industry unwillingness to admit that inaccurate technical
information has been tolerated. However inaccurate technical information is hazardous because it
remains in circulation for decades.6 (The NRC should also recognize that their report was relied upon by
industry and the courts to block correction of analysis problems that involved aviation and other
industries, including military organizations.34,35)
Spent nuclear fuel rods and nuclear waste will probably be transported for great distances when
a permanent repository is identified. Until then, spent fuel rods will continue to be stored on-site at
many nuclear plants. During storage and transportation, containment of radioactive materials will
depend on the integrity of the zirconium nuclear fuel rod cladding.5 Storing and handling fuel rods in
ways that will not compromise cladding integrity rely on good cladding alloy information. The history of
waste handling and transportation indicates that accidents are inevitable. One serious mishandled
nuclear accident will cost more and do far more damage to the nuclear industry than correcting known
inaccurate data.
Accurate understanding of nuclear fuel rod cladding will remain important as long as nuclear
fuel waste is generated, handled, transported, and stored. Release of radioactive species due to
cladding and other containment failures can render large areas of land uninhabitable, as occurred at
Fukushima and Chernobyl. Technical information will be relied upon to minimize the chances of
cladding failures for decades, if not centuries.
2. The NRC must require industry to review the zirconium alloy information that appears suspect.
Suspect information is identifiable by the circumstances under which it was created. If reviews identify
errors in suspect reports, flagging or correction of the inaccurate materials information should be
required.
Page 71
3. The NRC and industry must be required to adopt zero tolerance policies for known inaccurate
information. This is only common sense. The idea that a regulatory agency with responsibilities like
those of the NRC would tolerate known inaccurate information would seem to be seem incredible if the
NRC report1 and 2011 NRC letter to Senator Lugar2 did not both indicate that the NRC would not pursue
known inaccurate technical information. NRC tolerance of known inaccurate information has effect of
institutionalizing ignorance. Accurate technical information is essential to prevention of nuclear
accidents, making correct decisions to limit effects of ongoing nuclear accidents, and safe handling and
storage of nuclear waste.5,7 NRC regulations and practices are regarded as models for the nuclear
industry worldwide.6
4. The NRC must require industry to keep accurate records. Accurate records are essential to quick
responses in event of nuclear accidents and to post-accident failure analysis investigations relied upon
for lessons-learned intended to prevent recurrences of accidents. Inaccurate records serve only to
confuse investigations and allow parties who have profited from irresponsible behavior to avoid being
held liable for consequences of damage caused by their actions. Accidents are not the time to ask Can
records be trusted?
5. The NRC must provide open venues for investigations of industrys technical performance.
Irresponsible and incompetent industry and NRC actions must be openly documented as lessons
learned about bad information practices. Hearings for plant licensing and other actions that impact
public safety are being conducted under scientifically-dubious legal procedures where only parties
following narrow legal strategies have access to all of pertinent technical information. Public safety
relies on openness; neither should not be compromised by secrecy imposed by presumptions that all
technical information and company records are proprietary. Due to the fact that public safety and
public assets are put at risk when a nuclear plant operates, public disclosure of company technical
information can be considered a reasonable cost of doing business.
6. Individual companies that have inappropriately concealed information and engaged in other
misconduct should be presumed to have forfeited any right to conceal information by claiming it to
be proprietary by their history of misconduct. Companies must be prevented from concealing evidence
that their practices put public safety at risk for the sake of their short term profits. In the long run,
disclosure of the information will improve the safety of the technology for all involved and improve
profits of competent companies who do not profit by putting public assets at risk. All parties who could
lose their health, home, peace of mind, or livelihood as a consequence of risks taken by industry benefit
from open reviews of industry information. The likelihood that fraud, information problems, and
unnecessary risks will be exposed should increase when many eyes with different backgrounds and
interests are able to examine inherently risky activities. Technology advances (and is usually made
safer) by the free flow of information.
7. The NRC itself must not produce false information. Manipulation of public perceptions by
whitewash audits, promotion of a nuclear renaissance, and pro-industry biases are counter to the
intent of the Energy Reorganization Act (ERA) which led to the creation of the NRC.
Page 72
8. The NRC should not "lose" or destroy information. The NRC emailed parties requesting some of the
NRC information used to produce the 2011 response to Senator Lugar's inquiries. Why? The Indian
Point hearing exposed gaps in NRC records. Did the NRC lose or destroy information? What confidence
can the public have that the NRC will be able to respond quickly to incidents when the NRC cant find
basic information? Why wasnt the missing information posted on their web site?
From "cradle-to-grave", nuclear activities involve long time periods. Plants have operated and
spent fuel has accumulated for over a half of a century. The NRC allows companies to take up to 60
years to complete plant decommissioning after power generation operations cease. Permanent
solutions to nuclear waste disposal remain as chimerical as nuclear energy too cheap to meter
promised at the dawn of the nuclear industry. (No doubt these inaccurate projections arose from faulty
understanding of the realities of nuclear technology due to reliance on bad information. ) Spent nuclear
waste, which a GAO report describes as containing the most hazardous substances on earth7, continues
to accumulate at many nuclear plants in the US. Safe storage of spent fuel rods in cooling pools (some
of which are leaking into groundwater) rely largely on intact zirconium cladding to prevent releases of
radionuclides into coolant and into the environment.
Information about cladding and other nuclear components will be relied upon for an indefinite
period of time. Some uses of the information, like some types of accidents, cannot be foreseen.
Therefore, the nuclear industry and regulatory agencies must maintain all records indefinitely. As the
UA 232 FAA accident report demonstrated, accidents are not the time to ask "Can the records be
trusted?"
Page 73
Appendix I. Four NRC Records Documenting Flawed NRC Information Practices
Part A. The Initial NRC Report Describing Zirconium Texture Analysis Errors
NRC correspondence and a report indicating that the NRC did not consider dissemination of
known bad texture information in industry to be a safety concern and indicating that the NRC took no
action to curtail the distribution of inaccurate information on zirconium fuel rod cladding material
properties.
At the NRCs request, a copy of this report to the NRC in mid-2011. This is the report referred to
in the NRCs November 2011 letter to Senator Lugar included in Appendix I Part B.
Note that the NRC cites the unpublished and unreviewed opinions of two experts to support its
conclusions. One consulted expert, described as an engineer of one of the major nuclear fuel
fabrication companies who has experience in texture analysis from a fuel cladding design perspective,
appears to be from Westinghouse. Westinghouse failed to recognize texture analysis errors in multiple
reports and also received fraudulent information which inaccurately describes causes of errors.
Note that the NRC report does not explicitly address information in published papers indicating
that texture influences the deterioration and failure of zirconium alloys in nuclear application. For
example, see The Influence of Crystallographic Texture and Test Temperature on Initiation and
Propagation of Stress-Corrosion Cracks in Zircaloy, Zirconium in the Nuclear Industry. 6th International
Symposium (1982) 32 and citations therein. Note that numerous studies have since confirmed that
understanding zirconium texture is necessary for some nuclear applications. 5-10, 12-14, 17,18
Page 74
Page 75
(The NRC report sent with the attached letter is reproduced below under Allegation NRR-A-0057.)
Page 76
Page 77
Page 78
Page 79
Appendix I
Part B. 2011 NRC Correspondence to Senator Lugar
2011 NRC Correspondence to Senator Lugar reaffirming the initial NRC report. The NRC indicates that
they believe problems were remediated. However, the NRC did not appear to understand what data
and records were later found to contain errors, so remediation was unlikely to have been effective.
No record of remediation is available. The NRC appears to have taken no effective action to curtail the
distribution of the bad data, to require review of suspect data, or correction of inaccurate records
identified after their earlier report.
Page 80
Appendix I
Part C. NRC Observation Audit Record of Lambda High-Level Waste Container Material Analysis
2001 NRC Audit Records of Lambda Research work on high-level waste container design for
Yucca Mountains describing Lambda QA deficiencies. Note that the report states The audit team
identified potential deficiencies in training, calibrations, document control, software verification and
validation, and corrective action. The same types of deficiencies led to distribution of reports with
inaccurate texture information to the nuclear industry. Note also that this Yucca Mountain work was
itself QA work. This record was posted on the NRC web site in ML013330072.pdf.3
Page 81
Page 82
Page 83
Appendix I. Part D. NRC Audit Record Containing False and Misleading

Information
Excerpts from 2004 NRC Audit Records of Lambda Research. (The full audit record was available on
the NRC web site in the document posted as ML042750057.)
Note that part 2.0 STATUS OF PREVIOUS INSPECTION FINDINGS inaccurately states There
were no NRC inspections or technical reviews performed at the Lambda facility prior to this inspection.
Note that this statement can be shown to be inaccurate because it conflicts with the records from the
2001 NRC audit of Lambda Research (reproduced in Appendix I Part C above), which demonstrates that
NRC auditors did perform a technical review at Lambda of the same QA system and some of the same
procedures. This inaccuracy is significant in that it could conceal earlier audit records that describe QA
deficiencies related to Lambda analysis of high level nuclear waste container material. Accident
investigators could be misled into believing that the zirconium audit was the first NRC audit of Lambda;
investigators might not look for earlier records.
Also note that Lambda records, GE Nuclear records, and court records indicate that the same
deficient Lambda procedures, software, and training that created errors in the GE Nuclear work
described in the QAIR (Quality Assurance Incident Report) were employed in work whose purchase
orders explicitly stated that 10CFR21 was applicable. Records show that these NRC auditors knew the
following: 1) Defective texture analysis procedures, software, and QA had been used zirconium analysis
work which did explicitly state that 10CFR21 was applicable (eg, see ML011430464.pdf). 2) Defective
texture analysis procedures, software, and QA had supposedly been audited by GE Nuclear and Global
Nuclear Fuels under the NIAC (Nuclear Industry Assessment Committee) audit standard for compliance
with 10CFR21 and 10CFR50b (eg, see ML011430464.pdf). However, the following NRC audit record
states that The inspectors were not able to identify any QAIR records relating to a 10CFR Part 50,
Appendix B purchase order and Part 21 was not applicable to the Lambda activities performed for
the specific testing. Two of the NRC auditors had been provided with copies of the GE Nuclear and
NIAC audit records and information describing violations of 10CFR50b and 10CFR21 requirements. The
following NRC audit records do not mention the initial NRC report describing errors in the audited
zirconium texture analysis (reproduced in Part A above). The NRC audit records do not mention the
problems getting information corrected, which included threats that employees reviewing or discussing
the errors and deficiencies would be fired. At the time of the NRC audit, the ALJ had sent the decision
wherein the judge found that Kelly engaged in protected activity when he prepared the QAR and that
Preveys threat to fire Kelly was an adverse employment action. 20,21 Furthermore, the NRC had
information that Prevey had demanded that Kelly edit the QAR to inaccurately indicate that
Westinghouse and other nuclear industry clients listed in an email describing rotation errors had been
appropriately notified of the distortion errors. (The Court of Appeals decision notes that, contrary to
his revision to Kellys QA report, Lambda had not informed GEND of the procedural defects alleged by
Kelly.15 Decision. Page 5 Footnote 3, reproduced above in Section 2.)
Page 84
This 2004 NRC Audit Report contains false statements that are significant in that they conceal
the history and extent of deficiencies in Lambda analysis for GE Nuclear, Global Nuclear Fuels,
Westinghouse, and other nuclear industry clients.
Page 85
(The full NRC audit record is posted on the NRC web site as ML013330072.pdf.)
Page 86
Appendix II. How Inaccurate Information Impeded UA 232 Accident Responses

and Failure Analysis
A. NTSB Accident Report on UA Flight 232. A Case Study of Post-Accident Decisions Impaired
by Bad and Suspect Information about a Component Designed and Manufactured Decades
Before the Accident.
The history of the effects of bad information on nuclear accidents and steps required to prevent
their recurrence is limited. However, NTSB Aircraft Accident Report United Airlines Flight 232 22
describes how inaccurate and suspect information can create hazards and thwart efficient failure
analysis efforts and corrective actions.
The Accident. A titanium disks uncontained separation, fragmentation, and forceful discharge from a
GE gas turbine engine destroyed the hydraulic systems of United Airlines Flight 232 while in flight.22
Overcoming extreme difficulties, the flight crew crash-landed the plane in Sioux City, Iowa. Basically, a
bad engine disk exploded and ruined critical aircraft controls. Over 110 fatalities and numerous injuries
occurred during the crash landing.
1. The Accident Investigation: The NTSB accident report states that the accident was initiated by a
defect in an engine disk. The defect in the crash disk was present when this disk was fabricated decades
earlier. The disk was in service for almost two decades before deterioration in the disk materials
condition caused the catastrophic failure. See: National Transportation Safety Board Aircraft Accident
Report United Airlines Flight 232 McDonnell Douglas DC-10-10 Sioux City Gateway Airport Sioux City
Iowa July 19 1989 PB90-910406Y "NTSB/AAR-SO/06". Section 1.17.3.22 Excerpts of this NTSB report are
reproduced below.
Bad GE materials and bad GE information appeared to have been major factors in this crash.
Inaccurate records impeded identification and removal of other suspect disks from service. If left in
service, other bad disks could also have failed and caused more crashes at any time.
The NTSB report described inaccurate engine manufacturing records that should have contained
reliable information about the origins and history of the crash disk. Accurate and self-consistent records
should have enabled fast identification of similar (therefore, suspect) disks in service so that they could
be immediately removed from service and inspected before another accident occurred due to the same
type of manufacturing defect. Examination of records of the crash disk and suspect disks along with
inspections of suspect disks should also have enabled fast and efficient determination of the root causes
of the accident.
However, inaccurate, conflicting, and suspect records impeded the accident response and
investigation. The report indicates that two disks from two different sources had the same serial
Page 87
number as the crash disk:
Since conflicting GE and supplier records indicated that there were at least two possible sources
of the bad disk (and some records indicated that the disk was from materials cut after the engine was
shipped, which is physically impossible unless something really unusual occurred), investigators
attempted to identify the source of the crash disk by chemical analysis.
Results of the chemical analysis could not identify the source of the crash disk. Reasons for the
disk failure remained speculative.
GE started inspections and the NTSB recommended examination of over 100 engine disks. Disks
manufactured with and like the crash disk were considered the riskiest disks. They should have been
removed from service for inspections first. The failure to understand the source of the bad material
made immediate identification of riskiest disks for prioritized inspections impossible. During
inspections, flaws like those found in the crash disk were found in some disks were removed from
Page 88
service and inspected after the crash. However, suspect manufacturing information left doubts.
The phrase intended effect because suspect disks could remain in service indicates other
engines with bad disks could have remained in service because of bad information.
GE records contained anomalies, some of which appeared to obscure the disk history at GE.
Several record anomalies call into question all disk records from the same period.
The most interesting anomalies are records and evidence indicating that the crash disk was
made from titanium material that other records show: 1) was not supplied to GEAE 2) was not cut until
two months after GE shipped the engine with the crash disk 3) was not qualified for rotating parts like
disks (note that parts that rotate fast tend to explode if they have certain types of defects, so special
materials and processes are used to manufacture them.) 4) a disk with the same serial number as the
crash disk had an unsatisfactory ultrasonic indication.
About the time that the crash disk was manufactured, GE records indicate that inspection
showed that a disk with the serial number of the crash disk contained an unacceptable flaw.
Page 89
Note that the No. 2 engine in the accident airplane is the engine that exploded.
Other GE records indicated that GE reassigned this serial number to the crash disk using
titanium from another supplier. However, this suppliers record indicate that this titanium was not ever
shipped to GE. GE had records indicating that GE shipped the flawed disk to another location to verify
the flaw. The NTSB report appears to indicate that GE fouled up their records by recycling an essentially
free serial number (rather than using a flawed titanium disk costing thousands in materials whose
replacement might cost much more in manufacturing delays).
The NTSB report indicates that no record of warranty claims or credit for the disk with defects
were found (p. 55). This might seem trivial to some who dont work at a company like GE, which is
obsessed with supplier performance, costs, and budgets. Those disks cost thousands.
One crash, one failed disk, records of two disks with the same serial number, two disk suppliers,
and several sets of conflicting records. Chemical analysis failed to resolve the conflicts in information.
Did the NTSB ever sorted out the record anomalies described in their report?
2. Chemical Analysis Questions.
Some of the blue-etch anodization work that helped confirm that the crash disk flaw described
in the NTSB report was due to the presence of a hard alpha phase in the titanium was done by Kelly
prior to his employment at Lambda research. A record describing this work is the following GE
Page 90
achievement recognition from GEs Engines Materials Technologies Laboratories (EMTL) where the
work was done.
Page 91
Kellys blue etch work on this investigation was also mentioned in his GE performance review.
Page 92
Page 93
Kelly was later asked to do analysis of soil from the farm field where the crash disk was found as
part of a post-accident investigation. GEs position appeared to be that United Airlines missed the flaw
during fluorescent penetrant inspection of the disk before the crash. Some phosphorous analysis done
by another materials analyst using a method for metals analysis seemed to support this position. Since
the lab did not have appropriate methods for soil analysis, a sample was submitted to an agricultural
lab. Their results indicated that some initial GE phosphorous analyses results were questionable.
The previous analysis for phosphorous in soil from the crash disk appeared to have been done
using a method designed for metals. The blanks done as part of this analysis appeared to have been
prepared from deionized water reagent blanks. The results were questioned because the crash disk was
found in a farm field. Analysis indicated that the dirt contained quartz and possibly other silicates.
Silicates can interfere with the colorimetric phosphorous analysis method under certain conditions.
Therefore, a soil blank would have been appropriate for analysis of the crash disk sample in question.
Other uncertainties about the results arose from the fact that typical agricultural fields contain
phosphate fertilizers. Analysis of more samples might have helped determine levels and variability of
phosphorous in the soil. Analysts involved agreed that these were issues prior to report preparation.
The report did not describe any conclusive results. The report just raised some questions about
a previous analysis. Previous analysis results should have been checked (ie, repeated) or discarded.
The FAAs investigation made it seem that flagging data as questionable might be in the best
interests of all concerned, especially GEAE. Credibility is always an important consideration in a failure
analysis investigation.
The same manager who had signed Kellys evaluation and achievement recognition records
strongly disagreed with releasing these technical findings and indicated that Kellys report did not
present a just picture of the colorimetric P analysis done here at GE (see his note at the bottom of the
soil analysis report reproduced below).
Page 94
The manager and Kelly discussed the report, as requested. The technical issues were clear.
Some were certain of the uncertainties in the phosphorous results. There were concerns that bad
decisions might be made. Also, if the matter were questioned by the FAA, NTSB, or United Airlines,
these problems with the phosphorous analysis would seem obvious. If not addressed, the problems
could undermine the credibility of GE and the client requesting the work. Changing the report to
remove uncertainties did not appear to be appropriate.
Page 95
The reaction of the manager was surprising. He expressed anger over the fact that this and
subsequent revisions included findings that he wanted out. The discussion lasted about a minute. The
customer got the information, but did not react as expected.
It is not clear if the questions about the validity of some of the results indicating that
phosphorous was found on the crash disk surface were ever investigated. If phosphorous from
fluorescent penetrant was found on the crash disk surface, failure analysis questions would focus on
maintenance inspection failures to detect that fluorescence. However, if the phosphorous found came
from the farm field or fertilizer or the high results were due to silica or other analysis interferents , the
results of that phosphorous analysis would be irrelevant in determining the accidents origins.
There were other measurements which stand independently of the labs phosphorous analysis.
The FAA and NTSB did appear to find that maintenance inspection failures caused the crash. However,
several questions arise. Why not check the validity of suspect results? Why risk having suspect results
undermine a failure analysis investigation? Why risk a labs credibility on information that may appear
suspect to experienced analysts? While GEAE EMTL was technically an outstanding organization and
most GEAE scientists in the labs were very capable professionals with high integrity, some management
actions raised serious questions.
In summary, this accident investigation report and related documents describe how a materials
failure severed critical controls in-flight decades after the flawed material was manufactured. The
failure of the airliner engines titanium disk initiated at a metallurgical flaw that led to disk failure,
causing a crash landing with over 100 fatalities and additional injuries. Basically, a bad engine disk
exploded during United Airlines Flight 232. Problems with records generated decades earlier and other
information impeded fast identification and removal of other suspect engines from service and raised
doubts during the subsequent failure analysis investigation.
3. Zirconium Information Problems and Court Proceedings
Based on experience, Kelly and others at Lambda were aware that these types of problems and
hazards could arise from inaccurate data, records, and other information problems in the nuclear
industry.
Lambda audit records show that GE NIAC audits included examination of other QARs as
evidence of compliance with NIAC requirements. NRC audit records indicate that QARs were examined.
Lambdas owner later described QARs to an OSHA/DOL investigator:
Page 96
Lambdas owner later wrote an affidavit submitted to federal court describing the QARs
differently.
Does comparison of the affidavit with statements made to the OSHA investigator and audit
records indicate that the owner either submitted a false affidavit to federal court, gave false statements
to a federal investigator, or both?
It is interesting to note that, months prior to introducing this affidavit, Lambda sent copies of
the QAR along with an ex parte communication to a hearing officer explaining their position. Prior to
preparing the above affidavit, Lambda had itself used a QAR for a purpose other than internal quality
control.
The judge cited the above affidavit several times in a decision dismissing the complaint19.
Motions for reversal citing the fraudulent nature of the affidavit were not acted on by the court.
Page 97
The NRC did not act on the information. The NRC did say that they would conclude their
investigation of Lambda when court actions (some of which they recommended) were finished. After
some court actions were finished and the, the NRC said it applied a no harm-no foul approach to these
situations.
The ALJ decision characterized the QAR as a document available for review by outside parties.
(Administrative Review Board (ARB) Case 02-075 Final Decision and Order. Appeal for reconsideration of decision of Administrative Law Judge
Case 2000-ERA-035.)
A Federal Court decision stated the following.
(Order. Case No. C-1-00-661. United States District Court for the Southern District of Ohio Western Division. 12-3-01. )
The NIAC audit records, the NRCs audit records, and other records indicate that the affidavit is
false and misleading. Is producing a false affidavit to federal court acceptable to the NRC? How serious
was the NRC investigation if they ignored a false affidavit to federal court and/ or false statements made
to a federal investigator? In event of a nuclear accident, will the NRC allow industry to make false
statements and alter industrys versions of events recorded in statements made to federal investigators
documented in sworn affidavits to suit their shifting positions as situations develop?37 Are companies
free to generate records containing false information that would be useful to conceal defects and
reduce liabilities in event of costly incidents or accidents?
Page 98
Appendix III. Short ARB Decision and Related Information

The three page Administrative Review Board Final Decision and Order. (Note that the order
does not mention evidence and findings in other decisions demonstrating that management demanded
signatures on the edited QAR on the last day of employment. Note also that factual errors which judges
arbitrarily included in this and other decisions are uncorrectable. For example, this decision incorrectly
states that Lambda was established in 1997. However, earlier errors might exist. Lambda was in
business over a decade earlier.)
Page 99
Page 100
Page 101
Its interesting to note that this decision states that the ALJ determined that Kelly did not
demonstrate by a preponderance of evidence that submitting the QAR contributed to his termination of
employment from Lambda. Statements in the ALJ decision do seem to indicate that the QAR
contributed to the termination of employment.
Page 102
21
Page 103
Authors Experience and Credentials.
Mark Kelly holds a Ph. D. in chemistry from the University of Cincinnati, where he studied
electroanalytical chemistry. As a chemist in GEs Engine Materials and Technologies Laboratories, he
preformed materials analysis in support of engine development, operations, and failure analysis, which
included work on the UA 232 Sioux City accident investigation described above. He has also worked as
an analyst in the chemical, materials, and water treatment industries. He has experience implementing
auditable laboratory quality systems and laboratory information systems that complied with ISO 9002
standards. He has written numerous internal and external analysis reports concerning materials
properties, chemical analyses, and regulated contaminants for the aviation, materials, chemical, and
other industries. He co-authored the following three publications concerning materials,
electrochemistry, and radiation effects on materials.
S. T. Wlodek, M. Kelly, and D. Alden. "The Structure of Rene '88 DT". Superalloys 1996. Edited by R. D.
Kissinger et al. The Minerals, Metals, and Materials Society. Warrendale, PA, p. 129 (1996).
S. T. Wlodek, M. Kelly, and D. Alden. "The Structure of N18". Superalloys 1992. Edited by S. D. Antolovich
et al. The Minerals, Metals, and Materials Society. Warrendale, PA, p. 467 (1992).
Mark J. Kelly and William R. Heineman. "Gamma-Irradiated Polymer-Modified Mercury Film Electrodes".
J. Electroanal. Chem. 222. pp. 243-256 (1986).
Page 104
References:
1
NRC Allegation NRR-1999-A-0057 Concerns 1 thru 3. Texture Analysis of Zirconium Alloy. NRC
correspondence and report stating that the NRC does not regard known bad texture information to be a
safety concern and will not pursue bad data distributed in industry. (A copy is reproduced above in the
appendix and has been posted on the Internet at Scribd.com as NRCAllegationNRR-1999-A-0057. )
2
Correspondence. R. W. Borchardt, NRC Executive Director for Operations, to Senator Richard Lugar,
United States Senate. November 21,2011. NRC correspondence reaffirming NRC Report NRR-1999-A-
0057. (A copy is reproduced in appendix and has been posted on the Internet at Scribd.com as
NRCBorchardtToSenateLugarNov212011.)
3
NRC Reading Room File ML013330072.pdf. NRC Observation Audit Report No. OAR-02-01 and related
DOE correspondence. (Portions of this document are reproduced in appendix.)
4
Source: NRC Inspection Report 99901345/2004-201. NRC web site file ML042750057.pdf. (Portions
of this document are reproduced in the appendix.)
5
H. M. Chung, R. S. Daum, J. M. Hiller, and M. C. Billone , Characteristics of Hydride Precipitation and
Reorientation on Spent-Fuel Cladding. Zirconium in the Nuclear Industry. 13th International
Symposium, 2002, P. 449.
6
Nuclear Fuel Behavior in Loss-of-Coolant Accident (LOCA) Conditions. State of the Art Report. Nuclear
Energy Agency (NEA No. 6846). Organization for Economic Co-operation and Development (OECD)
Report. 2009.
7
NRC Needs to Do More to Ensure that Power Plants Are Effectively Controlling Spent Nuclear Fuel.
Unites States Government Accountability Office. GAO-05-339. April, 2005.
8
Delayed Hydride Cracking in Zirconium Alloys in Pressure Tube Nuclear Reactors. Final Report of a
Coordinated Research Project 1998-2002. International Atomic Energy Agency. IAEA TECDOC-141-.
ISBN 92-0-110504-5. October, 2004.
9
E. Tenckhoff, Review of Deformation Mechanisms, Texture, and Mechanical Anisotropy in Zirconium
Alloys. Zirconium in the Nuclear Industry. 14th International Symposium, P. 25. Journal of ASTM
International, April. 2005 Vol 2 No. 4.
10
Arthur T. Motta, Aylin Yilmazbayham, Robert Comstock, Jonna Partezanna, George P. Sabol, Barry Lai,
and Zonghou Cai, Microstructure and Growth Mechanism of Oxide Layers Formed on Zr Alloys Studied
with Micro-Beam Synchrotron Radiation. Zirconium in the Nuclear Industry. 14th International
Symposium18, P. 205. Journal of ASTM International, May. 2005 Vol 2 No. 5.
11
Timble, D., Hanford Waste Treatment Plant. DOE Needs to Take Action to Resolve Technical and
Management Challenges. GAO -13-38. December, 2012.
Page 105
12
Grytsyna et al. Destruction of Crystallographic Texture in Zirconium Alloy Tubes. Zirconium in the
Nuclear Industry. 14th International Symposium, p 305. Journal of ASTM International, Sept. 2005 Vol 2
No. 8.
13
Dahlback M. Dahlback, M. Limback, L. Hallstadiu, P. Barberis, G. Bunel, C. Simonot, T. Anderson, P.
Askeljung, J. Flygare, B. Lehtinen, A. Massish, The Effect of Beta-Quenching in Final Dimension on the
Irradiation Growth of Tubes and Channels. Zirconium in the Nuclear Industry. 14th International
Symposium, P. 276. Journal of ASTM International, June. 2005 Vol 2 No. 6 .
14
P. Rudling et. al. Welding of Zirconium Alloys. IZNA7 Special Topic Report Welding of Zirconium
Alloys 2007. Advanced Nuclear Technology International. 4-3(4-13).
15
United States Court of Appeals for the Sixth Circuit. Case No. 02-3035. Decision. On Appeal from the
United States District Court for the Southern District of Ohio. A copy has been posted on the Internet at
Scribd.com as USCourtAppeals6thDistrictCase02No3035.
16
NRC web site document ML011430469-1.pdf. QAR edited by Lambda Owner Paul Prevey.
17
King, S. J., Kesterson, R. L., Yueh, K. H., Comstock, R. J., Herwig, W. M. and Ferguson, S. D., Impact of
Hydrogen on Dimensionaly Stability of ZIRLO Fuel Assemblies. Zirconium in the Nuclear Industry:
Thirteenth International Symposium. ASTM SPT 1423, G. D. Moan and P. Rudling, Eds. ASTM. pp. 471-
489.
18
Arthur T. Motta, Aylin Yilmazbayham, Robert Comstock, Jonna Partezanna, George P. Sabol, Barry Lai,
and Zonghou Cai, Microstructure and Growth Mechanism of Oxide Layers Formed on Zr Alloys Studied
with Micro-Beam Synchrotron Radiation. Zirconium in the Nuclear Industry. 14th International
Symposium18, P. 205. Journal of ASTM International, May. 2005 Vol 2 No. 5.
19
Order. Case No. C-1-00-661. United States District Court for the Southern District of Ohio Western
Division. 12-3-01. A copy has been posted on the Internet at Scribd.com as USDistCtCaseC1No00661.
20
Administrative Review Board (ARB) Case No. 02-075. Final Decision and Order. A copy is reproduced
in the Appendix III above and has been posted on the Internet at Scribd.com as
DOLAdminRevBoardCaseNo02075.
21
Case No. 2000-ERA-0035. Recommended Decision and Order. Department of Labor, Office of
Administrative Law Judges (ALJ). Reproduced in Appendix III above.
22
National Transportation Safety Board, PB90-910406 NTSB/AAR-90/06, Aircraft Accident Report
United Airlines Flight 232. Retrieved April 2012.
23
Possible Fuel Rod Hazard Seen at Some Nuclear Plants. The Wall Street Journal. Feb. 16, 2011.
24
Mathew Walds. Judge Rules Vermont Cant Shut Nuclear Plant. NY Times. Jan. 19 2012.
Page 106
25
Supreme Court Declines Nuclear Waste Case, Government and Policy Concentrates, Chemical and
Engineering News, Dec. 17, 2005. P. 23.
26
Japan Airlines Flight 123, Boeing 747-SR100, JA8119. FAA Web Site.
lessonslearned.faa.govll_main.cfm?TabID=1&LLID=16 . Retrieved July 12, 2013.
27
R. C. McMaster. Nondestructive Testing Handbook. The Ronald Press Co. NY. 1963. P. 32-7.
28
NRC web site document ML011430491.pdf. Note that the NRC keeps the investigation open until
legal proceedings are concluded. Ongoing investigations can be barriers to discovery of internal NRC
documents.
29
The history of the Cincinnati area legal community and law enforcement in business-worker litigation
is documented. While the following examples might seem off-topic, they offer insights into seemingly
outrageous behavior by the legal communities and courts in which the zirconium actions played out.
First example: A series of 1998 Cincinnati Enquirer Chiquitas Secrets articles were critical of
Cincinnati-based Chiquita business practices in Latin America involving mistreatment of workers, forcible
prevention of workers from unionizing, and other misconduct. Chiquita filed federal lawsuits against a
reporter and several Chiquita employees believed to be responsible for leaking the information. A
special prosecutor obtained a guilty pleas from the reporter for felonies involving his illegal access of
Chiquita voice mails that he had used to obtain or corroborate information about Chiquita activities
(Dan Horn, Former Enquirer Reporter Guilty, in The Cincinnati Enquirer, Sept . 25, 1998, retrieved
2/20/1013, Douglas Frantz, Mysteries Behind Storys Publication, in The New York Times, July 17,
1998). In Chiquitas case, despite the fact that some of the descriptions of illegal and unethical activities
in the stories did not rely on any illegal activities by the reporter or were primarily corroboration of
other evidence of illegal activities, Chiquitas president was quote in the 9/25/98 Enquire article as
saying Todays guilty plea by the lead reporter clearly supports the fact that these stories were false,
misleading, and lacking credibility. The decision in this Chiquita court case indicated that the reporter
committed illegal acts while gathering information, but that does not mean that this or other
information used as sources for the stories were false or lacking credibility. While the FBI and local law
enforcement had aggressively investigated the employee and reporter misconduct, available reports do
not indicate that Chiquitas illegal business practices involving their Cincinnati offices were investigated
in 1998. The 2009 Al Jazeera English program People and Power- Chiquita: Between Life and Law 10
June 09 (www. Youtube.com/watch?v=PMCTPoLT42U retrieved Feb. 24, 2013) reported that, between
1997 and 2004, Chiquita gave millions of dollars to Columbian a paramilitary groups who employed
violence and murder to intimidate workers and union leaders. Did the FBI and local law enforcements
failures to investigate Cincinnati Chiquita offices roles in illegal activities in Latin America, along with
law enforcements aggressive pursuit of employees and reporters who brought some of those types of
activities to light, allow Chiquita to continue to pay paramilitary groups that committed violence and
murder against banana workers for years after information about these types of payments were leaked
by Chiquita employees and reported in the Cincinnati Enquirer? In 2001, one group paid by Chiquita
was designated a Specially-Designated Global Terrorist, making it a federal crime for Chiquita, as a US
Corporation, to provide money to this group. In 2002, senior Chiquita executives established
Page 107
procedures to pay directly and in cash; a senior Chiquita officer described these procedures to Chiquitas
Audit Committee. Against advice of legal counsel, Chiquita continued to pay until 2004. (See the
Department of Justices 2007 document #07-161:03-19-07 Chiquita Brands International Pleads Guilty
to Making Payments to a Designated Terrorist Organization And Agrees to Pay $25 Million Fine
retrieved 2-20-2013.) Instead of stopping the misconduct in 1998, inaction of law enforcement and the
courts appear to have allowed Chiquitas funding of organizations responsible for intimidation, violence,
and murder to continue until 2004; in the meantime, The Cincinnati Enquirer paid a million dollar
settlement to Chiquita and subjected Enquirer readers to a long series of positive front-page stories
about Chiquita and its directors. It is significant to note that the original Enquirer series of articles
"Chiquita's Secrets Revealed" and some subsequent articles describing Chiquita's behavior, such as
Salon's article, cannot be found readily on the internet (if at all). Does silencing critics by prosecution
under pretenses that disclosing evidence of criminal activity amounts to violation of laws protecting
companies proprietary information amount to establishing a corporate code of omerta that runs
counter to public policy? Second Example: Cincinnati-based P&G contacted local law enforcement
about suspicions that some of its employees were leaking information to the press. Reports indicate
that P&Gs CEO directed eager-to-please law officers, hip-pocket prosecutors, and a city detective
who gets a second salary from the company to investigate. Subpoenas were issued for thousands of
home and business phones in actions described as intended to intimidate its critics and conceal its
embarrassments. (William Safire, Essay; At P&G: It Sinks NY Times, Sept 5, 1991). One reporter
was accused of breaking the law. An attorney called the Cincinnati Police chief to stop one of their
officers with ties to P&G from continuing to harass the reporter. An Ohio State professor is quoted as
saying The people who govern Cincinnati treat the town like a plantation. Local officials protect the
good old boys. The Ohio Legal Director for the American Civil Liberties Union is quoted as saying P&G
has law enforcement to its beck and call (Alecia Swasy, Soap Opera The Inside Story of P&G Times
Books, 1993, p. 297-305). The reporter wrote about various P&G problems, including toxic shock
syndrome deaths that appeared to be caused by a P&G product, prostitution and drug scandals at the
company, and retaliation against employees for raising concerns about illegalities. Notably, despite the
fact that prostitution and drug trafficking were illegal, law enforcement appeared in indicate that they
performed no investigation of the evidence that these activities occurred at P&G. Third Example: A
Cincinnati-area chemical companys CEO informed a state attorney general that an employees
disclosure of information demonstrating that regulated chemical products marketed as approved for
food contact applications and cosmetic ingredients were contaminated with heavy metals was a felony,
as the company regarded records of the contamination as proprietary information. The highly profitable
cost-cutting practices included "test and switch" of raw materials and processing conditions.
Documentation indicates that, after testing indicated that products passed regulatory tests, substitution
of cheaper raw materials with higher levels of regulated contaminants led to production savings of 20%
or more. Processing lines of regulated and unregulated contaminated products were also shared,
leading to more savings. Company sales literature included claims that these products were FDA
approved for food contact. A different set of company sales literature did not indicate that one of the
products was even suitable to be thrown in the garbage, per some states' "Toxics in Packaging"
regulations. In response to Freedom of Information Act (FOIA) requests regarding the food contact
approval claims and related regulations, the FDA indicated that it had no information about the
Page 108
situation, despite having been provided with the information requested in the FOIA over a year earlier
and information indicating that the product met FDA requirements appearing the Code of Federal
Regulations. (The FDA lab that might do testing for this type of contamination is located in Cincinnati.)
In addition to appearing to have sufficient influence to manipulate the CPD and an alphabet soup of
other governmental groups to harass and possibly prosecute those calling attention to corporate
misconduct, companies appear to have sufficient resources to employ private firms, medical entities,
and other consultants to harass and undermine their critics credibility. Some of the Greater
Cincinnati governmental offices involved in these activities appear to have been consolidated into a
"fusion center", making their activities less visible to the public.
As one former DOL investigator explained, the Cincinnati DOL office takes the employers side in
complaints unless they are certain that the employer will lose in order to avoid sending the wrong
message. Some Cincinnati employments lawyers have stated that they dont do DOL ALJ hearings.
Corporate influence over governmental actions appears to create a situation where law enforcement is
co-opted in actions that leverage the law to conceal (rather than investigate, expose, and limit)
corporate misconduct and that target and silence those who call attention to problems. However,
information demonstrating record falsification and inaccurate nuclear industry technical information
must be made available to those who could act on it (particularly in view of events at Fukushima and the
increased urgency to address the spent nuclear waste accumulating at US reactors). Unfortunately,
circumstances indicate that it is likely that such efforts could create undesirable outcomes that prevent
corrections, as occurred in the case of Chiquita.
30
Donald Sadoway. MIT 3.091SC Introduction to Solid State Chemistry, Lec 20, Fall 2010.
www.youtube .com/watch?v=malCa9k17Ag&list=PL1F2179A79AEC920D. At 18:37 min. Information
retrieved 3/12/2012.
31
Case No. 2000-ERA-0035. ALJ Hearing. testimony. Department of Labor, Office of Administrative Law
Judges (ALJ).
32
D. Knorr et. al., The Influence of Crystallographic Texture and Test Temperature on Initiation and
Propagation of Stress-Corrosion Cracks in Zircaloy, Zirconium in the Nuclear Industry. 6th International
Symposium (1982), P. 627.
33
Hindle and Worswick , Variation in the Strain Anisotropy of Zircaloy with Temperature and Strain,
Zirconium in the Nuclear Industry, 6th International Symposium (1982), p. 133.
34
Lambda Internal Email. NRC web site document ML011430464.pdf
35
Lambda Internal Email. NRC web site document ML011430469-1.pdf.
36
Some parts of this document are redundant in some respects. Redundancy appeared to be necessary
in order to allow some sections to "stand alone" and in order to improve clarity and to reduce the
effects of misunderstandings arising from recurring uncontrollable instabilities in the word processing
Page 109
program that have changed, swapped, or removed words and phrases in the document as it was being
revised.
37
The technical problems with fuel rod cladding properties information were confirmed in
court decisions 15,19,20,21 and by the NRC.1,2 The false statements in the QA Incident Report were
confirmed in court decisions. Employees actions intended to correct problems and notify clients were
blocked by adverse actions by management prior to notification of the NRC. Demands to sign the QAR
backed by firing threats and other adverse actions was confirmed in court decisions.20,21 GE Nuclears
safety warnings were reproduced in at least two decisions.15, 21 The fact that erroneous technical
information and related records have not yet been corrected was confirmed in court decisions and has
been confirmed (or reaffirmed) by the NRC.2 Note that these findings were based on evidence from
Lambda records backed by testimony of Lambda employees hostile to the action. (The DOL disregarded
the testimony of the alleger or plaintiff based on its interpretation of responses about testimony about a
partially inaudible statement in a partially inaudible telephone unemployment hearing (that was part of
an unemployment process that the NRC told the alleger to file) and based on flawed interpretation of
some highly technical testimony that the DOL judge did not appear to comprehend. These DOL findings
and their impact on company records used in the decision were not apparent until published in the
decision. Neither these findings nor their impact on the decision could be reversed on appeal.
However, it is important to reiterate that the nature of the errors, the threats over document
generation, the false nature of the document, and GEs notification of safety issues, along with other
elements, were established with other evidence despite the DOL judges selective use testimony and
arbitrary discarding of other records and evidence.) The judge in the federal case also relied on the
owners false affidavit and disregarded GE Nuclear audit documents and other records to dismiss the
case covering broader issues and evidence indicating that inaccurate materials analysis was more
widespread than the zirconium analysis. 19 The NRCs report documents the NRCs ignorance of the
nature and uses of texture information. The NRC and courts that rely upon the NRC have blocked
correction of correction of erroneous technical information about nuclear fuel rod cladding properties
and removal of related fraudulent records from company archives. The NRC has also generated false
audit records and generated audit records that decouple safety concerns from errors and deficiencies
that the NRC repeatedly found in industry procedures.4,3,1 Calling attention to these events and
problems, which is in none of the parties direct interests, appears necessary to establish that
responsible parties must correct the specific and systemic information problems. As the Fukushima
accidents and ongoing leakage of radioactivity indicate, current technology does not have the
capabilities necessary to stop the spread of radioactive materials once a serious accident has occurred.
Understanding the quality of information that will be relied upon when incidents and accidents occur is
required to limit the impact of nuclear technology on the public health, the environment, and the
economy. Concerns about the problems described above were described in a Limited Appearance
Statement on the Renewal of the Licenses for the Indian Point Energy Center, which is posted on the
NRC ADAMS database with Accession Number ML12270A373. This statement and other earlier
communications to the NRC got no meaningful response. Steve Schulin at power plant news from
nuclear.com made a comment indicating that the NRC problems with hiding and mishandling
information continue, writing It is downright ironic that ADAMS database entry for this document
Page 110
shows author name of Mary Kelly rather than Mark Kelly in the document title, and Kelly, M in the
author name field. (Steve Schulin. #nuclear #Zr Lax and flawed NRC information practices. Thats Dr.
Mark Kellys polite description. Posted at nuclear.com. 10-15-12. Concerns directed at congressmen
seemed to get more response, but every indication is that bad information remains uncorrected.
Correcting inaccurate data and reviewing suspect data is incredibly simple. Accidents and incidents are
incredibly expensive. Lack of responsible NRC action appears to indicate that the NRC is more interested
in managing nuclear perceptions rather than nuclear safety.
38.
United States of America Nuclear Regulatory Commission. Atomic Safety and Licensing Board Panel
Hearing. Docket Nos. 50-247-LR and 50-286-LR. ASLBP 07-858-03-LR-BD01. In the Matter of: Entergy
Nuclear Operations, Inc. (Indian Point Generating Units 2 and 3).
39.
Much of the content of this report serves to provide context for described NRC and industry
information problems. During preparation of this report, considerable efforts were made to describe
the issues accurately and the relevant details completely and understandably within the limitations of
the available space and time. However, many of the issues are complex and interrelated and many lack
background and experience in some of the specialized technical areas. Therefore, it is recommended
that those interested in particular issues consult referenced materials and other sources in order to
reduce the possibilities and effects of errors in understanding. While reproduction of this entire report
should reduce problems of concern, in order to reduce the possibility of misunderstandings, please do
not reproduce isolate excerpts of this document without consulting original sources. When possible, the
source materials should be referenced directly. Questions (and suggestions) concerning the technical
issues and other content can be directed to: mark.kellyzr2011@gmail.com.
Page 111

NRC and Nuclear Industry Generation of False Records and Tolerance of Inaccurate Technical Information On Zirconium Nuclear Fuel Rod Cladding and Other Systems Critical To Public Safety

Uploaded by

Copyright:

Available Formats

You might also like

NRC and Nuclear Industry Generation of False Records and Tolerance of Inaccurate Technical Information On Zirconium Nuclear Fuel Rod Cladding and Other Systems Critical To Public Safety

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

NRC and Nuclear Industry Generation of False Records and Tolerance of Inaccurate Technical Information On Zirconium Nuclear Fuel Rod Cladding and Other Systems Critical To Public Safety

Uploaded by

Copyright:

Available Formats

NRC and Nuclear Industry Generation of False Records and Tolerance of Inaccurate Technical

Key Points Summary

Are accidents the time to ask Can the data be trusted?

Key Points and Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

An Overview: Some Quotes from NRC and Court Records . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

1. Materials Information and Zirconium Alloys in Nuclear Applications . . . . . . . . . . . . . . . . . . . . . . . . . 6

2. Errors in Reports on Zirconium Texture Sent to the Nuclear Industry . . . . . . . . . . . . . . . . . . . . . . . . .17

3. An NRC Report Confirmed Errors in Nuclear Industry Zirconium Alloy Reports. . . . . . . . . . . . . . . . . 23

5. NRC Responses to the Texture Analysis Errors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

6. Recurrences of Zirconium Texture Analysis Failures and Falsification of Records. . . . . . . . . . . . . . . 31

8. Opaque and Incomplete NRC Investigations and Audits. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .49

11. Conclusions and Recommended NRC Actions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

Appendix I. Four NRC Records Documenting Flawed NRC Information Practices . . . . . . . . . . . . . . . 84

Authors Experience and Credentials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .104

The NRCs lack of action regarding known errors in reports:

(Source: NRC ReportNRR-1999-A-00571.)

GE Nuclear Notification to Lambda Regarding Zirconium Alloy Texture Work:

Zirconium in the Nuclear Industry. 14th International Symposium. Paper on Texture:

(Source: NRC ReportNRR-1999-A-00571.)

1. Materials Information and Zirconium Alloys in Nuclear Applications.

Correction of information now known to be inaccurate, review of suspect information, and

Why Is Understanding the Condition and Behavior of Zirconium Alloys Important?

Accurate technical information on zirconium alloy properties is essential. Designs depend

Why is Zirconium Texture Important?

The distribution of crystallographic textures within zirconium alloy components is one of

The NRCs toleration of zirconium texture technical information known to be inaccurate is a

1) What is "crystallographic texture"? Crystallographic texture is a term used to describe the

FIGURE 1. Texture in Polycrystalline Zirconium Alloys.

2) Why is crystallographic texture important? Crystallographic texture can determine how

A. As- Manufactured. B. Hydrides Precipitate C. Hydrides Grow D. Hydrides Crack,

Although unfamiliar to many scientists, important aspects of crystallographic texture are

2. Errors in Reports on Zirconium Texture Sent to the Nuclear Industry.

The first error was a 90 degree rotation error.

An NRC report stated the following:

Figure 3. Effects of a Ninety Degree Rotation Error on Reported Texture Results.

A. Arrows Show Crystallographic B. Accurate Indication C. A 90 Degree Rotation Error:

Figure 3. Effects of a Ninety Degree Rotation Error on Reported Texture Results.

An NRC report stated the following.

3. An NRC Report Confirmed Errors in Nuclear Industry Zirconium Alloy Reports

4. Published Scientific Studies Describe Nuclear Safety Implications of

How do some studies describe the importance of zirconium alloy texture?

How do these radiation-induced displacements of atoms in crystallites deform zirconium nuclear

Do component distortions create measurable bows and dimensional changes in zirconium

Does zirconium texture determine how zirconium hydride reorientation occurs?

How well is hydride precipitation in nuclear fuel rod cladding understood?

Why is the texture of hydrides important to consider in cladding design?

Requiring industry to maintain accurate technical information amounts to cost-effective and

5. NRC Responses to the Texture Analysis Errors.

(Source: NRC web site document ML011430491.pdf. )

6. Recurrences of Zirconium Texture Analysis Failures and

The ALJ decision described these events.

Historically, inaccurate documentation has impeded accident investigations, concealed root

7. Courts Rely on Uninformed NRC Safety Assessments to Dismiss Safety Concerns.

One contention in the hearing concerned materials degradation due to flow-accelerated

8. Opaque and Incomplete NRC Investigations

9. Inaccurate NRC Audit Records Conceal Recurring Nuclear Industry Problems.

A. The UA 232 Accident Investigation: Inaccurate Records, Conflicting Documents, and

Lambdas owner testified about 90 degree rotation errors in reports: