From DOE Authorization to NRC Licensing
- sarahgibboney
- Jul 21
- 8 min read
Intended audience: For advanced reactor developers transitioning from DOE or Department of War authorization toward an NRC commercial license application, and the investors evaluating whether that transition is being handled correctly.
Executive summary
The Nuclear Regulatory Commission's (NRC) April 2026 draft interim staff guidance (ISG) on reactor designs previously authorized by U.S. Department of Energy (DOE) or Department of War (DOW) is best read as a roadmap, not a waiver. It signals that the NRC is willing to leverage prior DOE/DOW-reviewed information, including material qualification testing, analytical tool validation data, partial-design criticality work, scaled prototypes, and even major portions of previously authorized designs, but only if the applicant still submits a complete NRC application and can show that the prior work is applicable to the proposed commercial design, applicant, and site. The practical gate is Quality Assurance (QA) pedigree: for Part 50 and Part 52 applications, the guidance still points applicants back to 10 CFR Part 50 Appendix B, RG 1.28 Rev. 6, and the NRC-endorsed use of ASME NQA-1 code versions for design and construction. The draft is also notably non-prescriptive on the exact sufficiency threshold for DOE pedigree, which means applicants should expect a case-by-case review.
What DOE work can actually carry forward
The strongest candidates for reuse are DOE/DOW work products that already look like NRC licensing artifacts: controlled design analyses, test results tied to a defined configuration, validation of analytical methods, and documented safety analyses from facilities that were authorized, tested, and shown to function safely. The ISG is paired with a proposed rule to revise 10 CFR 50.43(e) so that prior DOE/DOW authorization and successful testing can become an explicit route for demonstrating safety functions. But the guidance does not create a blanket transfer of approval; it says the NRC’s leveraging will depend on how well the applicant demonstrates overlap with NRC requirements and on early pre-application engagement so the staff understands what DOE/DOW actually reviewed.
The real QA bar for Part 50 and Part 52 applicants
For startups moving from DOE-authorized work into an NRC application, the most important message in Appendix A is that the QAP/QAPD remains foundational. The ISG says Quality Assurance Programs (QAPs) for these applications must meet Appendix B to 10 CFR Part 50 unless an exemption is justified; it recommends the QAPD be a standalone topical report that can be incorporated by reference into the Safety Analysis Report (SAR); and it states that a construction permit Quality Assurance Program Description (QAPD) should cover design, fabrication, construction, post-construction testing, and preoperational testing. RG 1.28 Rev. 6, in turn, endorses Part I and Part II of NQA-1-2017, NQA-1-2019, and NQA-1-2022, with NRC clarifications and regulatory positions, as acceptable methods for meeting Appendix B during design and construction.
Recent NRC practice shows what reviewers will likely look for in that QAPD. In the staff’s acceptance of Blue Energy’s QAPD, the NRC found acceptable controls for design inputs, outputs, changes, interfaces, records, and organizational interfaces; document control over preparation, review, approval, issuance, and changes; software verification under NQA-1-2022 Subpart 2.7; records administration covering receipt, preservation, retention, storage, safekeeping, retrieval, and classification; supplier qualification through triennial audits plus annual evaluations; and commercial-grade dedication and procurement controls that explicitly invoke 10 CFR Part 21. In other words, “configuration management” is not treated as a separate buzzword so much as the combined output of design control, change control, document control, supplier control, and retrievable QA records. If those threads are weak, NRC reviewers will notice quickly.
Software, testing, and the Part 57 fitness-for-duty wrinkle
The draft guidance matters just as much for how DOE work is documented as for what can be reused. For software, the NRC has recently accepted the principle that design-analysis programs must either be verified under the applicable NQA-1 requirements before use or have their results independently verified for each application. For pre-construction and test work, the message is similar: testing that supports licensing should sit inside controlled procedures and a retrievable records system, not in ad hoc engineering folders. If a startup intends to dedicate commercial software or design-analysis tools as basic components, NRC guidance under RG 1.231 and 10 CFR Part 21 becomes relevant as well.
Separately, the proposed Part 57 rule is worth watching because it may change the staffing and drug/alcohol-testing burden for some microreactor applicants. The proposal would allow a licensee to use a Fitness for Duty (FFD) program of its own specification if operator action is not required to keep the plant within the § 57.25(a) 1 rem TEDE entry criterion and if a credible operator or maintenance error could not cause that criterion to be exceeded. Otherwise, proposed Part 26 Subpart P would provide a more flexible framework than today’s prescriptive model (though, in my opinion, more expensive and invasive), including evaluating alternative biological specimens, innovative screening/testing approaches, remote assessments in some cases, and core controls such as behavioral observation and written procedures. For startups, that could be a meaningful, abet expensive, future simplification — but it is a proposed Part 57 pathway, not a present-day substitute for Part 50/52 QA expectations.
A Notable Omission: Limited Integration with 10 CFR Part 53
One of the more striking aspects of the NRC’s draft guidance is what it does not address in detail: the agency’s newly finalized risk-informed licensing framework under 10 CFR Part 53.
The guidance is largely written through the lens of traditional licensing paradigms — emphasizing reconstruction of licensing bases, verification of quality assurance pedigree, and reconciliation of historical DOE-authorized activities with existing NRC expectations. While those topics are essential, the document provides little discussion of how these transition issues may differ under a Part 53 application strategy.
That omission matters for several reasons.
First, many advanced reactor developers considering transition from DOE authorization are simultaneously evaluating whether Part 50, Part 52, or Part 53 provides the most practical licensing pathway. Part 53 was specifically developed to accommodate advanced technologies, modern analytical methods, performance-based requirements, and staged development approaches. Yet the draft guidance largely treats the transition problem as though legacy NRC licensing structures remain the only relevant framework.
Second, Part 53’s emphasis on risk-informed and performance-based methodologies could materially affect how the NRC evaluates historical DOE-generated data, testing programs, and operational experience. The degree of rigor required to establish acceptability may differ substantially from traditional deterministic expectations. However, the draft guidance does not explore whether previously authorized DOE activities could be leveraged more flexibly within a Part 53 framework.
Third, the guidance misses an opportunity to clarify how existing DOE quality programs map into Part 53’s graded approaches to quality assurance and safety significance. For developers that intentionally built their programs around DOE nuclear quality standards rather than Appendix B-style frameworks, this distinction is not academic — it directly impacts cost, schedule, and the usability of years of engineering and operational work.
The absence of meaningful Part 53 discussion suggests the NRC may still be operationally orienting this transition guidance around legacy regulatory structures despite simultaneously promoting modernization initiatives elsewhere. For industry stakeholders, that creates uncertainty: developers are left to infer how the agency intends to reconcile DOE-origin activities with the flexible licensing philosophy that Part 53 was intended to embody.
As the guidance evolves, one of the most valuable improvements the NRC could make would be explicit discussion of how transition expectations differ — or should differ — under a Part 53 application. Without that clarification, applicants may continue to over-prepare for legacy licensing assumptions that Part 53 was designed to move beyond.
What small teams should do now
For a small team designing a multi-unit small Small Modular Reactor (SMR) pack, the near-term strategy is straightforward: build one controlled reference design, one leverage matrix, and one minimum viable nuclear-grade management system. My read is that reviewers will focus first on five questions:
Is the DOE work product tied to the same design you are now licensing?
Is the QA pedigree traceable to Appendix B/NQA-1 expectations?
Were changes after DOE authorization controlled?
Are supplier and commercial-grade dedication boundaries clear?
Can records be retrieved and defended quickly?
The draft ISG does not specify a universal evidence package for those answers, so assume the burden is on the applicant to make the package obvious.
Immediate next steps
For those programs intending to leverage progress with DOE authorization into an NRC commercial application, my recommendations are as follows:
Run a pedigree triage on every DOE-generated calculation, test, software tool, and supplier artifact: reusable as-is, reusable with gap closure, or reperform under Appendix B/NQA-1 controls.
Issue a lean QAPD and core procedures now for design control, document control, records, software QA, supplier oversight, nonconformance/corrective action, Part 21 evaluation, and commercial-grade dedication.
Use pre-application engagement aggressively and walk the NRC through a DOE→NRC crosswalk showing exactly what DOE reviewed, what changed, and what NRC should be able to leverage.
Frequently Asked Questions
Q: Does DOE or DOW authorization guarantee the NRC will accept my prior work in a commercial license application?
A: No. The NRC's April 2026 draft interim staff guidance (ADAMS ML25363A192) treats prior DOE/DOW authorization as a roadmap, not a waiver. The agency may leverage prior material qualification testing, analytical tool validation, partial-design criticality work, or scaled prototypes, but only if the applicant still submits a complete NRC application and demonstrates the prior work applies to the proposed commercial design, applicant, and site.
Q: What is the real bar for Quality Assurance when moving from DOE authorization to an NRC Part 50 or Part 52 application?
A: The QAP/QAPD remains foundational. The draft guidance still points applicants to 10 CFR Part 50 Appendix B, RG 1.28 Rev. 6, and NRC-endorsed versions of ASME NQA-1 (2017, 2019, and 2022) for design and construction. The QAPD should be a standalone topical report incorporated by reference into the Safety Analysis Report, and for a construction permit it must cover design, fabrication, construction, post-construction testing, and preoperational testing.
Q: What does the NRC actually look for in a QAPD during review?
A: Recent practice, based on the NRC's acceptance of Blue Energy's QAPD, shows reviewers focus on design control (inputs, outputs, changes, interfaces), document control, software verification under NQA-1-2022 Subpart 2.7, records administration, supplier qualification through triennial audits and annual evaluations, and commercial-grade dedication tied to 10 CFR Part 21. Configuration management isn't a standalone checkbox — it's the combined output of these controls working together, and reviewers notice quickly if any one of them is weak.
Q: Does the draft ISG address how DOE-to-NRC transition works under the new 10 CFR Part 53 risk-informed framework?
A: Not in meaningful detail. The guidance is written largely through the lens of traditional licensing paradigms and doesn't explore how Part 53's risk-informed, performance-based methodology might affect the evaluation of historical DOE data, testing programs, or graded QA approaches. For developers weighing Part 50, 52, or 53, this is a real gap — the guidance leaves open how existing DOE quality programs map onto Part 53's graded approach to safety significance.
Q: How might the proposed Part 57 rule change fitness-for-duty requirements for microreactor applicants?
A: Proposed Part 57 would allow a licensee to use its own fitness-for-duty program specification if operator action isn't required to stay within the 1 rem TEDE entry criterion under § 57.25(a), and if a credible operator or maintenance error couldn't cause that criterion to be exceeded. Otherwise, proposed Part 26 Subpart P would apply — a more flexible but, in my assessment, more expensive and invasive framework than today's model. Either way, this is a proposed pathway, not a present-day substitute for Part 50/52 QA expectations.
Q: What's the difference between "reusable as-is," "reusable with gap closure," and "reperform" when triaging DOE work products for an NRC application?
A: This is the pedigree triage every DOE-generated calculation, test, software tool, and supplier artifact should go through before an NRC submission: reusable as-is means the work already meets Appendix B/NQA-1 expectations and is traceable to the current design; reusable with gap closure means the underlying work is sound but needs additional documentation or verification to close a specific gap; reperform means the work needs to be redone under Appendix B/NQA-1 controls because it can't be sufficiently traced or verified. Sorting artifacts into these three buckets early avoids discovering the gap during NRC review instead of before it.
Q: What should a small team do now if they're transitioning from DOE authorization toward an NRC commercial license?
A: Run the pedigree triage described above on every DOE-generated artifact; issue a lean QAPD and core procedures now covering design control, document control, records, software QA, supplier oversight, nonconformance/corrective action, Part 21 evaluation, and commercial-grade dedication; and use pre-application engagement aggressively to walk NRC staff through a DOE-to-NRC crosswalk showing exactly what DOE reviewed, what changed since, and what NRC should be able to leverage.
Sarah Gibboney, P.E. is the Founder of Gibboney Nuclear, PLLC, a nuclear licensing consultancy serving advanced reactor developers. She has 17 years of nuclear energy experience, including co-authoring Construction Permit Applications for both ARDP awardees, TerraPower Natrium and X-energy Xe-100.
Originally published by Sarah Gibboney on LinkedIn, May 14, 2026.
Comments