DEEP DIVE // PUBLISHED MARCH 2026 | UPDATED MARCH 2026
The Complete Buyer's Guide to Small Modular Reactors
PWR, MSR, SFR, HTGR, LFR, and Microreactor: a comprehensive comparison of every major small modular reactor type. Technical specifications, leading companies, advantages, challenges, fuel requirements, and investment implications for each reactor modality.
Published: March 2026 | Updated: March 2026 | Source: smrintel.com
SECTION 01 // INTRODUCTION
Why This Guide Matters
The small modular reactor market is fragmenting into six or more distinct reactor types, each with fundamentally different engineering approaches, coolants, fuels, safety profiles, and deployment timelines. Unlike the conventional nuclear fleet where pressurized water reactors dominate (accounting for ~70% of global capacity), the next generation of nuclear plants will feature genuine technological diversity. For utilities, data center operators, industrial buyers, and investors, understanding these differences is essential for making informed decisions.
The choice of reactor type determines nearly everything: what fuel is needed (standard LEU vs HALEU), how quickly it can be licensed, what applications it serves (grid power vs process heat vs remote deployment), and when it will be available. Each type carries different supply chain risks, regulatory timelines, and cost trajectories.
This guide covers all six major reactor types with technical specifications, key players, advantages, and challenges. For deployment timelines, funding data, and market analysis, see our State of SMR 2026 Annual Report.
SECTION 02 // PRESSURIZED WATER REACTORS (PWR)
Pressurized Water Reactors (PWR)
PWR-based SMRs are the most proven and regulatory-ready reactor type, based on the same fundamental technology that powers ~70% of the world's existing nuclear fleet. They use ordinary (light) water as both coolant and moderator, with the primary cooling loop pressurized to prevent boiling. Modern SMR designs like NuScale's VOYGR incorporate integral designs where the steam generator sits inside the reactor vessel, eliminating large-bore piping and reducing the potential for loss-of-coolant accidents.
The critical advantage of PWR SMRs is fuel supply: they use standard LEU enriched below 5%, the same fuel produced at scale by the global nuclear fuel supply chain. This means no dependency on the constrained HALEU supply chain. PWR designs also benefit from decades of operating experience, established regulatory frameworks, and an existing workforce trained on light-water reactor technology.
NRC Readiness
Most advanced
KEY PLAYERS
NYSE: SMR. VOYGR 77 MWe/module (up to 12-module 924 MWe). ONLY NRC-certified SMR design. Romania RoPower 462 MWe in development. ~$5.3B market cap.
470 MWe per unit. UK GDA Step 3 (expected Aug 2026). Wylfa site selected. First concrete as early as 2027. UK's technology of choice.
300 MWe BWR variant. First North American SMR under construction at Darlington (CNSC LTC Apr 2025). GE Vernova (NYSE: GEV).
SMR-300 (300 MWe). DOE $400M First Mover Award. NRC CP app filed Dec 2025. IPO filed Feb 2026 targeting $10B+ valuation. Hyundai E&C partner.
ADVANTAGES
+ Most proven technology — based on 70% of global operating fleet
+ Standard LEU fuel — no HALEU supply chain dependency
+ Most advanced NRC licensing (NuScale design certified)
+ Existing trained workforce and supply chain
+ Established safety track record spanning decades
CHALLENGES
- Water coolant limits outlet temperature (~300 C) — less suitable for process heat
- Pressurized primary system requires robust containment
- Higher $/kW for FOAK units vs conventional large reactors
- NuScale CFPP cancellation raised cost credibility concerns
- Less thermally efficient than high-temperature designs
SECTION 03 // MOLTEN SALT REACTORS (MSR/FHR)
Molten Salt Reactors (MSR/FHR)
Molten salt reactors use liquid salts — typically fluoride-based compounds like FLiBe (lithium fluoride-beryllium fluoride) — as the primary coolant, and in some designs, as both coolant and fuel carrier. The Kairos Power KP-FHR is a fluoride salt-cooled high-temperature reactor that uses solid TRISO pebble fuel with molten salt coolant, while Terrestrial Energy's IMSR dissolves the fuel directly in the salt.
MSR designs have inherent safety advantages: the salt operates at atmospheric pressure (eliminating pressurization risks), and a freeze plug at the bottom of the reactor vessel passively drains fuel to a dump tank if temperatures exceed design limits. Kairos Power's Hermes demonstration reactor at Oak Ridge received the first NRC construction permit for an advanced reactor in December 2023.
Coolant
FLiBe / Fluoride salt
Fuel
TRISO pebble / Dissolved
Enrichment
HALEU (Kairos) / LEU (IMSR)
KEY PLAYERS
KP-FHR 75 MWe. Hermes demo under construction (first advanced reactor NRC CP). Google 500 MW deal. FLiBe coolant with TRISO pebbles. $300M+ raised.
IMSR 195 MWe (twin-unit 390 MWe). Fuel dissolved in fluoride salt. CNSC Phase 2 VDR complete. First NRC MSR safety evaluation.
SSR-W Stable Salt Reactor. Molten salt fuel in static tubes — reduces engineering complexity. NB Power partnership in New Brunswick.
Thorium MSR in 40-foot shipping container. 100 MWth. Vision: one reactor per day per assembly line. Two years continuous pump operation validated.
ADVANTAGES
+ Atmospheric pressure operation — no pressurization risk
+ Passive safety via freeze plug drain mechanism
+ High outlet temperatures (600-700 C) enable process heat applications
+ Kairos has first advanced reactor NRC construction permit
+ Some designs can use thorium fuel cycle (Copenhagen Atomics)
CHALLENGES
- Corrosive salt environment requires specialized materials
- Limited operating history — no commercial MSR has operated
- FHR designs require HALEU fuel (supply chain risk)
- Tritium management in FLiBe coolant is an unsolved challenge
- Salt chemistry monitoring and maintenance complexity
SECTION 04 // SODIUM-COOLED FAST REACTORS (SFR)
Sodium-Cooled Fast Reactors (SFR)
Sodium-cooled fast reactors use liquid sodium metal as the primary coolant and operate with a fast neutron spectrum (no moderator). The "fast" designation means neutrons are not slowed down, enabling more efficient use of fuel and the ability to "breed" new fissile material from fertile isotopes. TerraPower's Natrium design adds an innovative molten salt thermal energy storage system, enabling the plant to ramp from 345 MWe baseload to 500 MWe peak output — a load-following capability that makes it particularly attractive for grids with high renewable penetration.
SFR technology has decades of operational precedent: sodium-cooled reactors have operated in the US (EBR-II), France (Phenix, Superphenix), Russia (BN-600, BN-800), Japan (Monju), and India (PFBR). TerraPower's Natrium plant at Kemmerer, Wyoming secured its NRC construction permit and is the furthest-along US advanced reactor project with a 2030 target completion.
KEY PLAYERS
Natrium 345 MWe + 500 MWe peak. Kemmerer WY NRC CP issued. Meta deal: 8 plants (2.8 GW). DOE ARDP $2B. Bill Gates-founded. KHNP investment.
NYSE: OKLO. Aurora 15-75 MWe. ~$12.9B market cap. Broke ground at INL Sep 2025. Meta 1.2 GW campus deal. COLA Phase 1 submission planned 2026.
ARC-100 (100 MWe). First SFR to complete CNSC VDR Phase 2 (Jul 2025). Point Lepreau deployment with NB Power. Series B closed Dec 2025.
ADVANTAGES
+ Decades of operational precedent (EBR-II, BN-600/800, Phenix)
+ Natrium thermal storage enables 345-500 MWe load-following
+ Fast spectrum enables more efficient fuel utilization
+ Atmospheric pressure operation (sodium boils at 883 C)
+ TerraPower has NRC CP — furthest US advanced reactor project
CHALLENGES
- Sodium reacts violently with water and air (fire risk)
- Requires HALEU fuel — supply chain bottleneck
- Opaque coolant complicates in-service inspection
- Historical SFR projects had cost/schedule overruns (Superphenix, Monju)
- Intermediate sodium loop adds complexity and cost
SECTION 05 // LEAD-COOLED FAST REACTORS (LFR)
Lead-Cooled Fast Reactors (LFR)
Lead-cooled fast reactors use liquid lead (or lead-bismuth eutectic) as the primary coolant. Lead's extremely high boiling point (1,749 C) means the reactor operates at atmospheric pressure with enormous thermal margins — the coolant cannot boil under any credible accident scenario. Lead is also chemically inert with water and air, eliminating the fire risk associated with sodium coolant.
Newcleo is the leading LFR developer, with a distinctive approach: using MOX fuel fabricated from reprocessed spent nuclear fuel, addressing both waste management and fuel supply simultaneously. The company moved its headquarters from the UK to Paris in 2025 and is building a non-nuclear PRECURSOR prototype in Italy. Blykalla's SEALER reactor in Sweden is another promising LFR design.
KEY PLAYERS
LFR-AS-200 (200 MWe). MOX fuel from reprocessed spent fuel. HQ Paris. $755M+ total raised. Italy PRECURSOR prototype by end 2026. UK GDA accepted Jun 2025.
SEALER-55 (55 MWe). Lead-cooled. $50M raised. Test facility at Oskarshamn NPP — Phase 1 construction complete. First criticality target 2029.
~450 MWe optimized. Lead coolant, atmospheric pressure, air-cooled power conversion. 8 UK test facilities operational since 2023.
ADVANTAGES
+ Extremely high boiling point (1,749 C) — no pressurization needed
+ Chemically inert with water and air — no fire risk
+ MOX fuel from reprocessed waste (Newcleo) — addresses spent fuel problem
+ Fast spectrum enables fuel breeding and waste transmutation
+ Massive thermal margins for passive safety
CHALLENGES
- Lead is extremely heavy and corrosive at high temperatures
- Lead freezing point (327 C) means careful thermal management required
- No commercial operating history for power-generating LFRs
- Materials science challenges for long-term lead corrosion
- Regulatory novelty — limited licensing precedent
SECTION 06 // HIGH TEMPERATURE GAS REACTORS (HTGR)
High Temperature Gas Reactors (HTGR)
HTGRs use helium gas as the coolant and TRISO (Tri-structural Isotropic) fuel particles — tiny uranium kernels coated in multiple layers of ceramic that can withstand temperatures above 1,600 C without releasing fission products. This makes fuel meltdown physically impossible: the fuel itself is the containment. HTGRs achieve the highest outlet temperatures of any reactor type (700-950 C), making them uniquely suitable for industrial process heat applications like hydrogen production, petrochemical refining, and desalination.
China's HTR-PM at Shidaowan became the world's first Gen IV reactor in commercial operation in 2023, providing real-world proof of the HTGR concept. X-energy's Xe-100 design is the leading Western HTGR, backed by Amazon's investment and a DOE ARDP award of up to $1.2 billion. X-energy also operates TRISO-X, the first NRC-licensed commercial TRISO fuel fabrication facility.
Fuel
TRISO pebble / prismatic
Enrichment
HALEU (most) / LEU (HTR-PM)
Key Safety
Meltdown impossible (TRISO)
KEY PLAYERS
Xe-100 80 MWe/module (4-pack 320 MWe). $700M Series D (Amazon, Citadel). DOE ARDP $1.2B. TRISO-X fuel facility NRC-licensed Feb 2026. Dow Seadrift + Amazon/Energy NW deployments.
MMR 5 MWe/15 MWth. Fully Ceramic Microencapsulated (FCM) TRISO fuel. Chalk River, Ontario deployment target. Pre-licensing with CNSC.
210 MWe (twin reactor, single turbine). World's first Gen IV in commercial operation (2023). TRISO pebble bed. Shidaowan, Shandong Province.
ADVANTAGES
+ TRISO fuel makes meltdown physically impossible
+ Highest outlet temperatures (700-950 C) — ideal for process heat
+ HTR-PM provides operational proof of concept (China, since 2023)
+ X-energy has integrated fuel supply (TRISO-X facility)
+ Strong DOE backing ($1.2B ARDP) and Amazon investment
CHALLENGES
- Most Western HTGR designs require HALEU fuel
- Helium leak management is a known engineering challenge
- Graphite dust accumulation in pebble bed designs
- Large reactor vessel relative to electrical output
- Limited load-following capability compared to SFR designs
SECTION 07 // MICROREACTORS
Microreactors
Microreactors are the smallest category of nuclear reactors, typically producing 1-20 MWe. They are designed for factory fabrication, truck transportability, rapid deployment, and autonomous operation with minimal on-site staffing. Use cases include remote communities, military bases, mining operations, disaster relief, edge data centers, and industrial facilities. The key value proposition is replacing diesel generators with decades-long, carbon-free power sources.
Radiant Industries is targeting the first US civilian microreactor criticality test on July 4, 2026 at the INL DOME facility, with commercial production planned for 2028 at its R-50 factory in Oak Ridge (50 units/year capacity). Last Energy has taken a different approach with a 20 MWe PWR-based microreactor and 80+ commercial agreements across Europe, with half targeting data center customers.
Coolant
Varies (He, heat pipe, water)
Transportable
Yes (truck/container)
Refueling
10-20+ year core life
Staffing
Minimal / autonomous
KEY PLAYERS
Kaleidos 1 MWe portable. $300M+ Series D. INL DOME criticality Jul 4, 2026. R-50 factory (50 units/year). Equinix 20-reactor deal. US Air Force/DIU FOAK 2028.
5 MWe/13+ MWth heat pipe reactor. Solid core design. Pre-licensing with NRC. 1/5 scale DOME test planned 2026. Penn State partnership.
NYSE: NNE. KRONOS (HTGR) + ZEUS (solid core) + LOKI + ODIN designs. ~$1.2B market cap. $577.5M cash. UAE MoU. First public US microreactor company.
PWR-20 (20 MWe) + PWR-5 (5 MWe pilot). 80+ European agreements, half for data centers. 30 Texas microreactors. $100M Series C. RELLIS pilot at Texas A&M.
ADVANTAGES
+ Factory-built and truck-transportable — fastest deployment
+ Replaces diesel generators with decades-long clean power
+ Minimal staffing / autonomous operation
+ Ideal for remote, military, and edge data center applications
+ Multiple designs competing drives innovation and cost reduction
CHALLENGES
- Most designs require HALEU fuel (supply chain risk)
- Higher $/kW than larger SMRs (but lower absolute cost)
- No commercial microreactor has operated in the US civilian sector
- NRC licensing pathway for transportable reactors is undeveloped
- Security and safeguards for distributed deployment TBD
SECTION 08 // HEAD-TO-HEAD COMPARISON
Head-to-Head Comparison
The following table compares all six reactor types across the key metrics that determine commercial viability: output range, coolant, fuel type, HALEU dependency, regulatory readiness, construction timeline, best application, and passive safety characteristics.
| Type | MWe | Coolant | Fuel | HALEU? | NRC Readiness | Timeline | Best For | Load-Follow | Passive Safety |
|---|
| PWR | 77-470 | Water | UO2 LEU | No | Certified | 2029-33 | Grid, data centers | Limited | Natural circulation |
| MSR/FHR | 75-195 | FLiBe salt | TRISO/Dissolved | Varies | CP Issued (Hermes) | 2027-33 | Process heat, grid | Moderate | Freeze plug drain |
| SFR | 15-500 | Sodium | Metallic HALEU | Yes | CP Issued (Natrium) | 2028-32 | Grid + storage | Excellent | Passive decay heat |
| LFR | 55-450 | Lead | MOX/HALEU | Varies | Early stage | 2029-35 | Grid, waste mgmt | Moderate | High boiling point |
| HTGR | 5-320 | Helium | TRISO HALEU | Most | CP App (Xe-100) | 2027-33 | Process heat, H2 | Limited | TRISO containment |
| Micro | 1-20 | Various | TRISO/HALEU/LEU | Most | Pre-App/DOE | 2026-30 | Remote, edge DC | N/A | Design-specific |
Data as of March 2026. NRC readiness reflects most advanced design per type. Timelines represent nearest projected deployment.
SECTION 09 // WHICH REACTOR TYPE IS BEST FOR...
Which Reactor Type Is Best For...
Large-Scale Data Centers (100+ MW)
PWR (NuScale VOYGR, BWRX-300, Holtec SMR-300) or SFR (TerraPower Natrium) for scale. HTGR (X-energy 4-pack, 320 MW) also viable. Standard LEU fuel avoids HALEU supply risk for PWR options.
Edge / Small Data Centers (<20 MW)
Microreactors: Radiant Kaleidos (1 MWe), Last Energy PWR-20 (20 MWe), Nano Nuclear KRONOS. Fastest deployment timeline and smallest footprint. Last Energy uses standard LEU.
Industrial Process Heat (>500 C)
HTGR (X-energy Xe-100, Ultra Safe MMR) — highest outlet temperatures (700-950 C). Ideal for hydrogen production, petrochemical refining, desalination. MSR also viable at 600-700 C.
Grid Replacement / Baseload Power
PWR (largest proven designs: RR SMR 470 MWe, BWRX-300 300 MWe) for near-term deployment. SFR (Natrium 345-500 MWe with storage) for grids with high renewable penetration requiring load-following.
Remote / Off-Grid Communities
Microreactors (Radiant, eVinci, Last Energy) for communities, mining operations, and military installations. Long core life (10-20+ years) with minimal staffing eliminates refueling logistics.
Developing Nations
PWR-based SMRs — proven technology, simpler licensing (precedent from NuScale NRC certification), standard LEU fuel available globally. KHNP i-SMR and SMART100 designed for export markets.
SECTION 10 // FUEL SUPPLY IMPLICATIONS
Fuel Supply Implications
The choice of reactor type has direct implications for fuel supply chain risk. Designs requiring HALEU face a significant bottleneck: US production is currently limited to ~900 kg/year from Centrus Energy's demonstration cascade, while planned reactors will need multi-ton annual supplies. Designs using standard LEU avoid this constraint entirely.
FUEL REQUIREMENTS BY REACTOR TYPE
Standard LEU (<5%)
NuScale VOYGR, BWRX-300, RR SMR, Holtec SMR-300, AP300, Linglong One, i-SMR, Last Energy PWR-20
LOW RISK
LEU+ (5-10%)
Some MSR designs, Terrestrial IMSR
MEDIUM RISK
HALEU (5-20%)
TerraPower Natrium, Oklo Aurora, X-energy Xe-100, Kairos Hermes, Radiant Kaleidos, eVinci, most microreactors
HIGH RISK
KEY INSIGHT
PWR-based SMRs using standard LEU fuel face zero HALEU supply chain risk and can leverage the existing global enrichment infrastructure. This is a significant near-term advantage for NuScale, BWRX-300, Rolls-Royce SMR, and Holtec. Advanced designs (SFR, HTGR, most microreactors) requiring HALEU face a 3-5 year supply gap until Centrus full-scale production comes online. See our
fuel supply chain tracker for live data.
SECTION 11 // INVESTOR IMPLICATIONS
Investor Implications
Each reactor modality carries a different risk/reward profile for investors. The fundamental tension is between proven technology (PWR) with lower returns and advanced technology (SFR, HTGR, MSR) with higher upside but greater execution risk.
MODALITY RISK/REWARD ANALYSIS
PWR: Lowest risk, nearest deployment
NuScale (SMR) is the only NRC-certified SMR. BWRX-300 is under construction. Holtec targeting IPO at $10B+. Near-term revenue generation most likely from this category. Lower upside ceiling but highest probability of commercial success.
SFR: Highest valuation, big tech backing
Oklo (OKLO, $12.9B) is the most valued SMR pure-play. TerraPower has Meta's largest nuclear deal. Both require HALEU and are pre-revenue. High reward if they deliver on schedule; significant downside if HALEU supply or construction delays materialize.
HTGR: Amazon-backed, dual fuel/reactor play
X-energy is privately held but may IPO. Only company with both reactor design and licensed fuel facility (TRISO-X). Amazon backing provides demand certainty. Process heat applications expand addressable market beyond electricity.
Microreactor: Fastest to first revenue
Nano Nuclear (NNE) is publicly traded. Radiant, Last Energy, and others are private with strong commercial traction (80+ agreements for Last Energy). Smaller absolute market but faster deployment timelines.
Fuel supply: Pick-and-shovel play
Centrus Energy (LEU) has a near-monopoly on US HALEU production. Cameco (CCJ) dominates uranium mining. These companies benefit regardless of which reactor type wins. Lower volatility, more predictable revenue.
PORTFOLIO STRATEGY
For diversified nuclear exposure, consider a barbell approach: fuel supply chain companies (LEU, CCJ) for steady growth, plus pure-play SMR developers (OKLO, SMR, NNE) for high-growth potential. Nuclear utilities (CEG, GEV) offer nuclear exposure with revenue diversification. See our
stock watchlist and
funding dashboard for live data.
SECTION 12 // FAQ
Frequently Asked Questions
Which SMR reactor type is the safest?
All modern SMR designs incorporate passive safety systems that rely on natural forces (gravity, convection, thermal expansion) rather than active intervention. PWR-based SMRs like NuScale have the longest safety track record since they are based on proven light-water reactor technology. HTGR designs like the Xe-100 use TRISO fuel with ceramic coatings that can withstand temperatures above 1,600 C, making fuel meltdown physically impossible. Molten salt reactors have inherent safety from the fuel being already liquid (no meltdown possible) and a freeze plug that drains fuel to a passively cooled dump tank. Lead-cooled reactors benefit from lead's extremely high boiling point (1,749 C), eliminating pressurization concerns.
Which reactor type is best for data centers?
For large-scale data center campuses (100+ MW), PWR-based SMRs (NuScale VOYGR, BWRX-300, Holtec SMR-300) and SFR designs (TerraPower Natrium) offer the most capacity per unit. HTGR designs like the Xe-100 can be deployed in 4-packs (320 MW). For edge data centers or smaller facilities, microreactors (Radiant Kaleidos at 1 MWe, Last Energy PWR-20 at 20 MWe) offer rapid deployment and smaller footprints. The key consideration is timeline: PWR designs using standard LEU fuel face fewer fuel supply constraints than advanced designs requiring HALEU.
What does HALEU mean for SMR fuel supply?
HALEU (High-Assay Low-Enriched Uranium) is uranium enriched to 5-20% U-235, required by most advanced reactor types including SFR (TerraPower, Oklo), HTGR (X-energy, Ultra Safe), FHR (Kairos), and most microreactors (Radiant, eVinci). PWR-based SMRs (NuScale, BWRX-300, Holtec, Rolls-Royce) use standard LEU (<5%), avoiding the HALEU bottleneck entirely. As of 2026, Centrus Energy is the only US HALEU producer at demonstration scale (~900 kg/year), far below multi-ton demand. This supply constraint is a significant risk factor for HALEU-dependent designs.
How much does a small modular reactor cost?
SMR costs vary widely by design and are still largely theoretical for FOAK (First of a Kind) units. TerraPower's Natrium plant at Kemmerer has a total project cost of approximately $4 billion with $2B DOE cost-share. NuScale's cancelled CFPP project was estimated at $9.3 billion for 462 MWe (about $20,100/kW), though the company projects NOAK costs significantly lower. The industry targets LCOE (Levelized Cost of Energy) in the range of $50-100/MWh for NOAK units, competitive with natural gas combined cycle. Microreactors target higher $/kW but lower absolute cost ($50-200M per unit).
Which SMR companies are the best investment opportunities?
smrintel.com does not provide investment advice. However, publicly traded SMR-exposed companies span several risk profiles: Oklo (OKLO, ~$12.9B market cap) and NuScale (SMR, ~$5.3B) are pure-play SMR developers with high growth potential but pre-revenue risk. Centrus Energy (LEU, ~$3B) benefits from HALEU monopoly position. GE Vernova (GEV, ~$90B) and Constellation Energy (CEG, ~$80B+) offer nuclear exposure with diversified revenue. Cameco (CCJ, ~$25B+) provides uranium supply-chain exposure. Holtec's pending IPO at $10B+ valuation will add another major name. For detailed analysis, see our stock watchlist at smrintel.com/stocks.
Published: March 2026 | Updated: March 2026 | smrintel.com