NRC's Part 53 Framework Opens New Licensing Pathway for Advanced Reactors

The first new reactor licensing regulations since 1989 take effect April 29, and for a sodium-cooled SMR developer preparing its inaugural NRC submission, the practical implications are more specific than the rule's policy branding suggests.

Part 53, which the NRC voted to approve on March 25, 2026 and published in the Federal Register five days later, is the agency's "Risk-Informed, Technology-Inclusive Regulatory Framework for Advanced Reactors." A K&L Gates analysis published April 7 parsed the mechanics and identified exactly where the licensing experience changes and where it does not.

Under Part 50, which dates to 1956, and Part 52, created in 1989, a sodium-cooled reactor developer faced a framework tailored for light-water technology. Reaching compliance meant costly exemptions wherever the design diverged from criteria written for water-cooled reactors. Part 53 eliminates that friction: applicants no longer need to seek exemptions from LWR assumptions. Molten salt, gas-cooled, and sodium-cooled designs are explicitly accommodated. The rule also opens multiple licensing pathways, including staged licensing and standardized designs, giving developers structural flexibility neither prior rule offered.

What remains hard is the PRA burden. Probabilistic Risk Assessment becomes the principal element of safety case development under Part 53, with expected coverage spanning internal and external hazards, multi-module interaction effects, and risks from novel fuels and coolants. This is not lighter documentation than Part 50 imposed. A first-time applicant must define risk performance objectives, justify them through PRA, and link physical design features to quantified risk through cross-cutting evidence, rather than checking compliance against externally set criteria. The NRC has stated those objectives do not impose continuous real-time demonstration requirements, which matters for multi-module operators, but the front-end investment in PRA capability is considerable.

The radiation protection treatment also shifts. Part 53 departs from traditional prescriptive ALARA language, with the NRC indicating that reference to Part 20 requirements may be sufficient in many contexts. If your regulatory workflows were built around a standalone ALARA demonstration, plan for a deliberate review of that assumption.

If you are weighing Part 53 against the traditional pathway, the K&L Gates analysis reduces to a sequence of yes/no questions that reveal whether the rule actually compresses schedule risk for your project. Does your design fall outside the LWR envelope Parts 50 and 52 assume? Can your team build and defend a PRA covering multi-module effects, external hazards, and novel material risks? Are you prepared to define your own safety metrics rather than inherit fixed criteria? Have you budgeted for pre-application engagement with NRC staff to shape PRA acceptance criteria before filing? And critically: have you reviewed not just the Part 53 rule text but the accompanying NRC guidance documents that define what "acceptable" means in practice?

That last question carries the non-obvious implication. Part 53's flexibility exists at the level of regulatory text. NRC guidance documents accompany the rule, and staff PRA acceptance criteria will effectively constrain what applicants can propose regardless of what the regulation itself permits. K&L Gates flagged early NRC engagement as necessary, not optional, precisely because guidance and precedent can narrow the design space considerably, negating much of the schedule benefit the rule is meant to provide.

Parts 50 and 52 remain available. But the NRC's vote on March 25 set a new baseline for what a modern reactor licensing submission looks like, and the guidance documents materializing around Part 53 will define its real-world boundaries more precisely than the rule text alone.