Why France Build's SSNs only for itself and Only Exports Diesel-Electric & AIP Subs

Partly utilising the wonders of Google AI Pete has created:

Question: Why does France only use SSNs for itself while only making diesel electric submarines for export?

Answer: Based on the sweat of millions of writers (and I don't get paid - just patriotism :) comes:

France uses nuclear-powered attack submarines (SSNs) for its own navy because it requires global reach and continuous strategic deterrence, but it exports diesel-electric submarines because strict non-proliferation laws, high costs, and low buyer demand make nuclear exports impractical. [1, 2, 3, 4]

1 is https://www.nti.org/analysis/articles/france-submarine-capabilities/

Why France Uses SSNs

Global Operations: France has overseas territories and global defense commitments. Nuclear subs can travel fast and far without ever needing to surface.

Carrier Strike Group Support: French SSNs protect the Charles de Gaulle nuclear aircraft carrier, matching its high sustained speeds. It can do at least 27 knots - see right sidebar.

Strategic Doctrine: France treats nuclear propulsion as a vital, sovereign military secret tied directly to national safety. [1, 2, 3, 4]
On top of SSNs France cannot afford diesel-electric+AIP subs for itself. In any case, there are sufficient supportive NATO allies (Germany, Italy now Sweden and eventually Norway and Poland) with AIP for shorter range sea missions - mainly against Russia.

• [Pete comment: Only SSNs can reliably defend France's SSBNs from Russian SSNs and Russian fast surface ships - particularly when French SSBNs are leaving-entering France's SSBN base at Île Longue (Long Island), Brest, Brittany, western French coast.] 

An excellent Indian In Our Defence video (here and above). Interviewee explains the differences between Diesel-Electric+AIP and Nuclear powered submarines. (Accents not too heavy for me - but please also use the "Subtitles/Closed captions" thingy second to the left on the right-lower corner of the video image).

---

Why France Only Exports Diesel-Electric + Rarely AIP add-on Submarines

• Non-Proliferation Rules: Exporting naval nuclear reactors risks spreading sensitive technology that can be weaponized. [And France would get blamed as the original reactor source. Hence France claims Brazil is only developing an indigenous Brazilian submarine reactor. A fine line.]
  
  
Market Demand: Most nations only need coastal or regional defence submarines. Diesel-electric boats (like the Scorpène-class) are quieter in shallow water, much cheaper to buy, and cheaper to run. 
Buyer Constraints: Most countries lack the specialized nuclear infrastructure, legal frameworks, and trained personnel required to safely operate and refuel nuclear reactors. [1]

 Pete Comment

After France developed first generation MESMA AIP (with Pakistan the only customer) France has not fully developed second or third generation AIP at sea operationally. Part of this may be development cost, India resenting any more French AIP for Pakistan. competition too high from Sweden’s Stirling and German PEM AIP. Also France’s Scorpene customers (Chile, Malaysia, Brazil and eventually Indonesia) get more benefit from batteries. Whether India’s Kalvari Scorpenes will ever be retrofitted with AIP is an open question.  

About larger French designed subs for the terminated Australian Shortfin/Attack class and the Netherlands' continuing Orka Program - they did not want AIP. This is because their long range/high speed transit mission profiles gained/gain little or nothing from heavy liquid oxygen (LOX) AIP technology after the LOX is used up. 

Unsurprisingly there is no standard power mix solution for submarine exporters' own needs and their customers' needs. It depends on their mission needs, geography, cost and many other factors. 

Novel (non-PWR) Reactors for Submarine Problematic: SK's Molten Salt Reactor Program

Arpit Kanodia at 7/31/2026 6:41 PM made some very interesting comments regarding problematic non-PWR novel reactors:

"Reply to Shawn C [Pete will also reply to Shawn in a future article] from previous post: The marine reactor South Korea (SK) is trying is an MSR (Molten Salt Reactor). [Pete AI confirmation is HERE] I don't know what their goals are or what they are trying to achieve, but not a single land-based commercial reactor based on an MSR exists; they are all test reactors. And the (SKs) want this for marine ships and submarines? This is nothing but hubris." [Pete comment - Very true Arpit] 

[Pete's has used Google AI, yielding ]

 "Neither the US nor the Soviet Union ever used Molten Salt Reactors (MSRs) in operational submarines. However, both nations experimented with liquid-metal cooled reactors during the Cold War, which are sometimes confused with molten salt systems. 

Soviet Liquid-Metal Submarines Project 645 (K-27): The Soviet Navy built one experimental submarine powered by twin lead-bismuth eutectic (liquid metal) cooled VT-1 reactors. It suffered severe reactor problems and a core partial meltdown in 1968.

Alfa Class (Project 705 "Lira"): The Soviets later built a series of high-speed attack submarines powered by lead-bismuth liquid-metal reactors (OK-550 and BM-40A). While fast and deep-diving, their coolant systems had to be kept externally heated at all times in port to prevent the metal alloy from solidifying. [Pete: The Alfas docked in very cold Arctic ports.]

See WaldenFont's comments 6 years ago: 

"Those reactors are something else. The coolant is a molten lead alloy. Once the reactor is running, it has to keep running all the time. If it ever stopped, and the alloy cooled, the entire cooling system seizes up and can’t be started again. Maintenance had to be done in port, with the coolant rerouted through an external furnace to keep it molten."

US Liquid-Metal Submarine [a very early "Seawolf" SSN-575 - nothing to do with the 1990 onwards seawolf class]: The US Navy tested a liquid sodium-cooled reactor on this single-hull attack boat. Sodium proved hazardous and difficult to manage due to its violent reaction with water, so the Navy decommissioned the plant and rebuilt the submarine with a standard PWR." [AI portion ends]

The US Atomic Energy Commission (AEC) historians' [according to Wikipedia] account of the sodium-cooled reactor experience was:

"Although makeshift repairs permitted the Seawolf to complete her initial sea trials on reduced power in February 1957, RICKOVER had already decided to abandon the sodium-cooled reactor. Early in November 1956, he informed the AEC that he would take steps toward replacing the reactor in Seawolf with a water-cooled plant similar to that in the Nautilus. 

The leaks in the Seawolf steam plant were an important factor in the decision but even more persuasive were the inherent limitations in sodium-cooled systems. In Rickover's words they were "expensive to build, complex to operate, susceptible to prolonged shutdown as a result of even minor malfunctions, and difficult and time-consuming to repair."[6]

The S2G reactor was replaced with a pressurized water reactor similar to Nautilus and designated S2Wa, the replacement process lasting from 12 December 1958 to 30 September 1960.[7]"

[Arpit continues] 

"This was the same thing tried by India in the 1980s. In 1980, DRDO proposed a [pressurised HEAVY water reactor] PHWR reactor for submarines, which [Indian] Naval Headquarters refused to sanction, and rightly so, it was madness."

SubMatts will resume with Arpit's very interesting comments on the dangers and complexities (adding up to extreme costs) of refuelling submarine reactors later in August.

Submarine Reactors from Rickover to Gorshkov and Beyond

My responses to parts of the late 2026 comment string include.

Even the US may have been unable to build a submarine reactor without the political connections, budgeting and engineering brilliance of Admiral Rickover. This Wikipedia entry has seemingly intentionally deleted almost all of Rickover's huge contribution to nuclear power - giving "Mamie Eisenhower" equal opportunity billing. Rickover successfully project micro-managed construction, land and at-sea testing of submarine and surface ship reactors and dictated the regulatory environments for them. He did the same for the land based civilian reactor sector and most things radioactive safety.

Britain totally relied on the US for submarine reactor development from 1959. Britain then refused to admit it benefitted from continuing US assistance and inspiration for PWR1. British nationalistic pride, Rolls Royce commercial objectives and US military industrial reluctance may have amounted to insufficient US assistance for PWR2. This partly explains PWR2's technical shortcomings. I admit I talk too often about PWR2 - yet it has largely incapacitated the 10s of $Billions UK Astute class, while also limiting the UK Vanguard class. Whether the UK can garner sufficient US assistance for PWR3 and PWR3+/4 remains an Anglophone anxiety.

Russia had/has a large enough industrial base, inventive talent, Admiral Gorshkov, set of national priorities (eg. nuclear weapon strategy) and skilled espionage to largely parallel develop submarine reactors. 

China, France and India benefitted from similar strengths as Russia.

Arpit is right in questioning whether middle power, South Korea, can develop submarine reactors like the great-sometimes super powers above. China, a superior innovator and manufacturer to SK, has taken decades longer than China hoped to develop quiet, efficient, submarine reactors. 

Also SK suffers from poor geography. However good the nuclear submarines SK builds they are hemmed in by the Japanese island chain and close proximity to NK and more especially to Chinese and Russian fixed undersea sensors, SSNs, SSKs and other nearby ASW platforms. 

SK and China may be strongly opposed to Japanese naval/military resurgence in the shape of Japanese developed nuclear submarines.

It is only the US that has the benign vast ocean space geography to fully benefit from the speed and other characteristics of nuclear submarines. The US advantage includes "owning" the Western Hemisphere for thousands kms until it gets to adversaries - Bering Strait excepted.

There is much more in the late July 2026 comment string that I'll respond to in future weeks.

SSN Proliferation & Australia Fears English Reactor Disease

Jun's, on July 19, 2026's comments and links of July 19, 2026 spawned a wide variety of interesting topics and issues. South Korea (SK's) bid for US help not only relates to SSN construction (KSS-N) but an SK desire to build SSBNs (SK-SSBNs) to deter nuclear armed North Korea.

SK's program is also a trigger "justifying" Japan building its own Mutsu-class nuclear  submarines. AUKUS (from good US, UK etc intelligence) can be seen as anticipating the above - justifying Australia's wants - AUKUS Pillar 1.

Australia is relying on UK and US assistance, at great cost. SK is relying on US assistance and international legal permission to use reactors for military/naval purposes. SK would also have liaised with France for 40+ years due to France's long experience with lower proliferation sensitivity-safer-cheaper LEU submarine reactors.

Of major value is Jun’s provision of an unofficial translation of Seong-Chang CHEONG's, Vice President, Sejong Institute's softpower@sejong.org Strategy and Roadmap for ROK-France Cooperation on Nuclear-Powered Submarines: EU-Based Naval Integration and a Mutually Beneficial Partnership Model at SEJONG FOCUS, February 10, 2026 at https://sejong.org/web/boad/1/egofiledn.php?conf_seq=22&bd_seq=12692&file_seq=41163

The Strategy and Roadmap is important for what it says and eventually says. Particularly on page 4: 

"France is precisely the only Western nation to have accumulated more than 40 years of experience in all three of these areas [including LEU reactors] since the commissioning of its first Rubis-class attack submarine in 1983. France has designed, built, and operated a total of 17 nuclear-powered vessels: 6 Rubis-class SSNs, 6 Suffren-class SSNs (including those under construction), 4 Triomphant-class SSBNs, and the aircraft carrier Charles de Gaulle…” 

The above statement is so far incorrect, as it omits the not usual 4 SSBNs but 6 French Redoutable-class SSBNs that each used up to 90% HEU reactors. 

Only further down on page 4 of Strategy and Roadmap is France's use of HEU remembered: 

"France also has a history of using highly enriched uranium (HEU) in its early nuclear submarine development. However, it has now fully transitioned to LEU. The early variants of the Le Redoutable-class SSBNs, France’s first independently developed strategic nuclear submarines in the 1970s, were equipped with reactors using weapons-grade HEU. In the 1960s, while enrichment capabilities were limited, highly enriched fuel was essential to increase the power output of compact submarine reactors. Weapons-grade uranium produced at the Pierrelatte military enrichment facility was used for this purpose."

Some Background and Issues on France's Wise Choice to Stick with LEU

Noting Submarine Matter's articles France’s Future SSBN “SNLE 3G” Starts Production of March 28, 2024; France's New K22 reactor for Carrier & 3G SSBN of December 27, 2023; and French LEU Reactor & SSN Implications of March 24, 2023: 

The USN, and partly the UK RN, rely on the US' massive HEU naval reactor industrial base to develop new submarine reactor technology and to create the 94+% HEU fuel.

If HEU submarine reactors and piping go drastically wrong - as with the UK's deeply flawed PWR2 reactors - HEU problems seem much harder and more expensive to rectify than regularly fixed LEU reactors.

Perhaps the engineering and potentially toxic health problems of accessing the PWR2's HEU assembly is the the problem that has "beached" all of the UK RN's Astute SSNs simultaneously and, I suspect, HMS Vanguard SSBN (withdrawn under inspection and maintenance for 10 years 2012 to 2022).

If the UK's future HEU PWR3 and/or PWR3+/PWR4 reactors fail to resolve the UK's PWR2 reactor technical disaster one hopes it doesn't impact Australia's SSN-AUKUS program. This is a hugely expensive issue people worry about in Canberra.

To avoid the English reactor disease should Australia plead with the US to sell us (at even greater expense) not 3, not 5, but all 8 SSNs Australia is seeking under AUKUS? Unlike the UK the US apparently makes reliable reactors.

I think the UK and US would prevent Australia turning to French LEU technology. One reason is the RAN officers/crew and shipyard workers currently receiving Top Secret nuclear training in the US and UK would not be trusted by the US and UK NOT to pass this Anglosphere SSN and reactor knowledge onto France.

Also how many countries can we throw Australian taxpayer's money at in our SSN quest? Not only to the US and UK, but to France and maybe even to South Korea's (world's most efficient) submarine building sector?

Pete is in Canberra for a week.

 Hi all,

Pete has emailed me that he's in Canberra for a week and will be slow to respond to his email notifications on comments on SubMatts.

In the meantime, my recommended submarine movies and a TV documentary to watch:

Das Boot 1981

Hunt for Red October 1990

Down Periscope 1996

Submarine Patrol 2011

UK TV documentary series shot during HMS Turbulent's last patrol in 2010.

South Korean Nuclear Powered Subs by the 2030s: Predicted 2015-2021

Pete's earliest detailed comments on South Korea and its SSN ambitions were April 16, 2015's "South Korean Submarines, 3,000+ ton KSS-III, Nuclear Potential" at https://gentleseas.blogspot.com/2015/04/south-korean-submarines-3000-ton-kss.html 15 comments, 10,155 pageviews.

Pete wrote "S Korean nuke sub Reactor likely to be French K15 Variant" on December 16 2021 at https://gentleseas.blogspot.com/2021/12/s-korean-built-nuke-sub-reactor-likely.html 4 comments, 2,220 pageviews.
Please see text below: 

"...SUBMARINE MATTERS BACKGROUND AND COMMENT [written in 2021]

South Korea is leaning toward a [Low Enriched Uranium (LEU)] reactor solution perhaps initially on a South Korea SSBN - which will be based around the ballistic missile submarine KSS-III SSB. It is not surprising South Korea has long been in talks with France (since 2017 if not earlier) about South Korea building a variant of France's K15 LEU reactor. The K15 is on France's Triomphant class SSBNs and, in slightly smaller form, on the France's Barracuda-Suffren class SSNs. 

In Submarine Matters’ “South Korea seeks Submarine Reactors from US and RUSSIA” of October 22, 2019 South Korea may have concluded it is better to buy an existing submarine reactor or at least a ship reactor than totally reinvent a submarine reactor. 

Since 2017 (if not earlier) South Korea has been considering France’s new Barracuda SSN with its K15 (aka K 15 aka K-15) reactor. With North Korea's buildup of nuclear weapon and missile capabilities South Korea, in October 2019, has been testing any increased US willingness for South Korea to explore nuclear propulsion options. 

The Triomphant-class's 2(?) x K15s and Barracuda's single K15 reactor stands for 150 MWt  power which translates to a total of 30 MWe electrical power (for the French Barracuda’s non-propulsive electrical services hotel load + propulsion).

"In October 2017, the [South Korean] Navy commissioned the Seoul-based Korea Defense Network to conduct a five month study on the feasibility of developing an indigenous nuclear-powered attack submarine. The think tank reported in March 2018 the results to the Navy, suggesting the service build a nuclear attack submarine along the lines of the French 5,300-ton Barracuda-class sub. The French sub is fuelled by low-enriched uranium."

Now that the US and UK have given the green light to supplying Australia with a reactor under the AUKUS deal it is conceivable that France may want [to] supply just the Barracuda’s K15 reactor or transfer technology (including a reactor design) for South Korea to incorporate in its 4,000+ tonne KSS III Batch III building program. 

The Barracuda’s K15 reactor has the:

-  political/regulatory advantage of using LEU ie. less than 20% U-235 which South Korea sees as not being restricted by the NPT or nuclear Safeguards Agreements. This is compared to the political sensitivities and anti-proliferation regulatory restrictions (up until September 2021's AUKUS deal) placed on exporting/importing US and UK HEU submarine reactors. The UK/US reactors use weapons’ grade HEU of 93-97 percent (see p. 20). 

-  technical advantage of being built for a submarine of just over 5,000 tons, ie. in South Korea's KSS III Bach III weight bracket. While the KSS III Batch II with 10 x VLS silos  has diesel-electric (also AIP) propulsion the proposed KSS III Batch III may eventuall have 12 to 16 VLS silos requiring an unusually large propulsion solution. Nuclear has been the solution for SSBNs built by the US, Russia, UK, France, China and India. 

-  Also Naval Group's assistance to Brazil to create a SSN, known as Alvaro Alberto [~5,900 tonnes submerged?], is highly relevant to the South Korean SSN (aka KSS-N) project, Alvaro Alberto's projected launch year (2029) and commissioning (2032-2034) may occur in the same timeframe as the South Korean SSN project.  At Submarine Matter’s 2015 article see the subheading "Brazilian Nuclear Submarine (SN-BR)” “There currently appears to be expectations that SN-BR will be around 100m long and 9m wide. This coincides with France's Barracuda SSN dimensions of: 99.5m long and beam: 8.8m.” 

-  So a A K15 variant first mounted on a South Korean SSBN by the early 2030s may be followed by mounting a K15 reactor variant on a South Korean SSN by the mid 2030s. In terms of vertically launched cruise or ballistic missiles the KSS Batch I features 6 VLS tubes, each with one missile. The KSS III Batch II [now features 10] and the KSS III Batch III perhaps 12 to 16. 

South Korea's neighbour Japan (a part strategic competitor of South Korea) is observing South Korea's interest in nuclear propulsion and ever larger submarines.

India, already having SSBNs, is interested in building 6 x Project 75 Alpha class SSNs with specialised SSN reactors. India is naturally talking to France's Naval Group. Naval Group is already  helping India build the 6 Kalvari-class Scorpenes, and bidding for India's 6 x AIP diesel-electric Project-75I SSK competition. Naval Group produces all of France's SSNs and SSBNs. So India can have far ranging discussions, on many topics, with Naval Group."

SubMatts article ends.

------------------

Very recently, in July 2026, Bill Seney and I have been discussing the above topic:

"Bill Seney [said]

I have seen news reports:
https://www.japantimes.co.jp/news/2026/05/26/asia-pacific/south-korea-nuclear-submarine-2030s/

that South Korea [SK] is looking at a nuclear powered variant of the KSS-III. Given that US nuclear subs seem to be getting further and further away and that UK subs are at least 2 decades away is it worth Australia's while to look at a SK option?

Given that [SK] has yet to build a nuclear submarine I could see the project slipping or even facing cancelation, so it is high risk. On the other hand, we know SK can build diesel boats. Perhaps a mixed fleet, with initial deliveries of conventional KSS-III followed by a nuclear variant when available? If the nuclear option is cancelled Australia would still end up with a powerful fleet of conventional submarines.

7/15/2026 10:41 AM
---

Pete2 [said]

Thanks Bill [I agree with your 7/15/2026 10:41 AM comment]]

I'll develop your idea as an article probably tomorrow.

I've long theorised here at SubMatts that SK would look to France's K15 LEU reactor for technology transfer. Also that the last KSS-III Batch 2 or Batch 3s would be nuclear powered.

Cheers Pete

7/15/2026 10:53 AM

---

Bill Seney [said]

Thanks Pete

Something I forgot to mention in my earlier comment is that the KSS-III is about the same size as the American Skipjack class SSN, so a nuclear KSS-III variant should be technically feasible.
[Pete comment - Yes the Skipjack-class SSN (1956-1990) was indeed KSS-III sized, at ~ 3,075t (surfaced) & ~3,513t (submerged) and the
French K15 using, Suffren/Barracuda SSN is ~4,765t-5,300t] 

Also, if Australia wanted to partner with South Korea on SSNs we can provide not just uranium [Australia has the 2nd largest Uranium reserves worldwide] but enrichment technology [Australian invented SILEX] as well, something the Koreans may be interested in.

7/15/2026 10:53 PM
---------------------

Pete Comment

Whether the US and its changeable Trump card can play a useful part in SK's (at least 11 year old) nuclear powered submarine program follows in a week. 

Certainly the US has an international law "123 Agreement" hold over SK that transactional Trump will maintain unless SK pays the US the price Trump seeks, that may vary from week to week.