European defence-policy commentary in 2026 is mostly a conversation about platforms. How many Type 26 frigates Norway has just bought. Whether the FCAS or GCAP sixth-generation fighter programmes will merge. The state of Polish K2 production. The British carrier readiness debate. These are all real questions and they all matter, but they are not the binding constraint on whether the bloc can sustain the war it is preparing to deter.

The actual binding constraint is chemistry. A 155mm artillery shell is a steel forging filled with explosive compound, propelled by a charge of nitrocellulose-based powder, ignited by a primer using sensitive compounds, and timed by a fuze. The steel forging is straightforward. The bloc has plenty of steel and plenty of forging capacity. The chemistry is not.

The European chemistry that fills shells, drives propellants and produces the high explosives in modern munitions sits in a small number of facilities operated by an even smaller number of companies. The most important of those is Eurenco, a state-controlled French company few people outside the industry have heard of, operating principally from two sites: Bergerac in the Dordogne and Sorgues in Vaucluse. Bergerac restarted nitrocellulose and propellant powder production in May 2023, after France shut down domestic powder production in 2007 and switched to imports. Sorgues produces hexogen and NTO, the high explosives that fill modern shells. Eurenco's combined investment programme 2024 to 2026 is €650 million, of which €500 million is the Bergerac restart and capacity expansion.

If Eurenco fails, the bloc's munitions scaling story fails with it. The platforms still exist, the steel still flows, and Rheinmetall's new shell-forging plant at Unterlüß still hits its 350,000-shell-per-year target by 2027. None of that is useful if the propellant powders, the high explosives and the primer compounds that turn forgings into weapons are not there in the matching quantity. Steel without chemistry is paperweight. The European defence debate has not yet caught up to this, mostly because the people who write about defence are not chemists, the people who are chemists do not write about defence, and the people doing both at scale work for Eurenco.

What Eurenco actually does

Eurenco operates four production sites: Bergerac (Dordogne), Sorgues (Vaucluse), Karlskoga (Sweden), and a smaller facility in Belgium (Bofors-historical). The two French sites are the strategic core.

Bergerac restarted nitrocellulose production in May 2023, sixteen years after France stopped making the stuff and went to imports. Nitrocellulose is the base material for modern propellant powders: the substance that, in correctly formulated grades, drives a 155mm shell out of an artillery tube at 850 metres per second. Nitrocellulose itself is made by treating cotton linters or wood pulp with nitric and sulphuric acids under controlled conditions. The chemistry is reasonably well understood and has been since the late nineteenth century. What is not reasonably available is the industrial capacity to make it at modern military scale, particularly outside China and Russia.

The Bergerac target is 1,200 tonnes per year of explosive powder, equivalent to about 500,000 modular propellant charges. That figure is the answer to the question "how many shells can the European bloc fire per year if French propellant is the only source." It is not the only source, but it is the largest single non-Polish source inside the bloc, and the figure is roughly an order of magnitude below what would be needed for a sustained continental war at Ukraine intensity.

Sorgues is the high-explosives site. It produces hexogen, also known as RDX, which is the principal main-charge explosive in modern artillery shells, missile warheads and demolition compounds. It also produces NTO (3-nitro-1,2,4-triazol-5-one), used in insensitive munitions where survivability against accidental ignition is critical, and other high-energetics. A €100 million expansion was approved in 2017 to add hexogen capacity. The site occupies 200 hectares and employs 435 people. Sorgues is the closest thing the bloc has to the United States Army's Holston facility in Tennessee, which now produces 20 million pounds of RDX a month, except smaller and not American.

Karlskoga in Sweden produces propellants and energetic chemistry as part of the same group, including supply for Saab Dynamics munitions.

The combined picture is that the bloc's energetic chemistry capacity at scale sits in three sites in three countries, principally in two French sites, all operated by one state-controlled company. A single industrial accident at Bergerac or Sorgues, or a single sabotage incident, would cripple the bloc's munitions scaling story for as long as the site took to recover. The recovery time would be measured in months, possibly years.

Why this is the bottleneck

European defence ministries have for two years now spoken in public about the goal of producing two million 155mm shells per year. The European Union committed in March 2024 to achieving that capacity by the end of 2025. The deadline slipped. The Commission's October 2025 review confirmed that capacity equipment was installed but full ramp would not be reached until 2026. The current expectation is that the bloc passes 2 million shells per year in 2026 and reaches around 2.4 million by 2027, with Rheinmetall the largest single producer at a target of 1.1 million per year by 2027 across all of its facilities.

That is the platform side of the story. The chemistry side is that for every shell forged you need a propellant charge, a high-explosive fill, a fuze and a primer. The propellant charge requires nitrocellulose. The fill requires hexogen, NTO or comp B (a hexogen-TNT mixture). The fuze requires energetic compounds for delay and fire. The primer uses lead styphnate or similar sensitive compounds.

If Bergerac is producing 1,200 tonnes of explosive powder per year, that is about 500,000 modular charges. The shell production target of 2.4 million per year requires several million modular charges, depending on the firing tables and zone selection. The arithmetic is uncomfortable: Bergerac alone supplies roughly a fifth to a quarter of the modular charge requirement for the bloc's 2027 shell-production target. The remainder has to come from Nitro-Chem in Poland, from imports, from the Karlskoga line, and from emergency expansion programmes that have not yet been fully funded.

For high explosives the situation is similar. Sorgues is the largest European hexogen producer. Nitro-Chem produces TNT and some hexogen. Chemring in the UK does some explosives manufacturing. The bloc's combined hexogen output is somewhere between five and ten thousand tonnes per year. Russian production is estimated by Estonian intelligence at significantly higher than that, and Russia's overall artillery output reached seven million shells in 2025, supplemented by another five to seven million from Iran and North Korea. The bloc is producing roughly a third of what Russia and its supply partners are producing, despite outspending Russia on defence by a factor of three.

The reason is not budget. It is industrial chemistry capacity, and that capacity sits in two French sites operated by one company.

How Europe got here

France stopped producing nitrocellulose at Bergerac in 2007 because the strategic logic of the time, dominated by the post-Cold-War assumption that large-scale conventional war was no longer plausible, made domestic powder production look like an unjustifiable industrial overhead. Imports were cheaper and the demand had collapsed. The same logic was applied across most of NATO Europe in the 2000s and 2010s. Munitions stockpiles were drawn down, production lines were closed, and the chemistry workforce aged out without replacement. The hollowing out was uniform and intentional. By 2014, when the Russian annexation of Crimea began the slow recognition that the assumption had been wrong, the bloc no longer had the chemistry it needed to fight a sustained conventional war.

Eurenco's restart at Bergerac in 2023 is the first material reversal of that decision. The €500 million investment, partly funded by €76 million in EU subsidies through the ASAP (Act in Support of Ammunition Production) programme, takes Bergerac from a closed site to a 1,200-tonne-per-year facility employing 450 people. That is real. It is also small relative to the requirement.

The harder problem is that the workforce, the engineering culture and the supplier chains that supported European energetic chemistry at Cold War scale no longer exist. Building 1,200 tonnes per year is a different problem from building 12,000 tonnes per year. The latter requires not just capacity but a functioning industrial ecosystem of cotton-linter suppliers, acid suppliers, plant operators, chemical engineers and process specialists at numbers the bloc does not currently have. That ecosystem takes a generation to rebuild.

What needs to happen

If the bloc wants to be able to sustain 12-month high-intensity continental war by 2032, the chemistry capacity needs to expand by an order of magnitude from current trajectory. The cheapest single highest-impact investment available in European defence right now is hardening and expanding Eurenco's Bergerac and Sorgues capacity, alongside parallel expansion at Nitro-Chem in Poland and Chemring in the United Kingdom.

The numbers are not large by defence-procurement standards. Doubling Bergerac is approximately €1 billion. Tripling Sorgues approximately €1.5 billion. New facilities at Eurenco's Belgian and Swedish sites approximately €1 billion. Comparable expansion at Nitro-Chem approximately €800 million. The total programme is in the order of €5 billion over five years, against a bloc defence budget that runs to several hundred billion. It is one F-35 squadron's worth of money, deployed where it would provide significantly more strategic value.

The reasons this has not happened are bureaucratic and structural. EU defence funding has been routed through ASAP, EDIRPA and the broader EDIP programmes, all of which are useful but small. The €150 billion SAFE facility is large, but the energetics expansion has not been prioritised within it. National procurement programmes have predominantly funded the platform end of the supply chain, not the chemistry end, partly because the political optics of writing a cheque for a French chemicals plant are different from the optics of writing a cheque for a Polish artillery battery. The political incentives are not aligned with the strategic need.

The structural answer is to designate Eurenco, Nitro-Chem and equivalent energetics producers as European critical-infrastructure providers, route SAFE funding into them at scale, and harden the production sites against sabotage and air attack. None of this is technically difficult. It is politically unsexy and therefore not getting done.

What it would mean if Eurenco failed

The single-point-of-failure risk at Bergerac and Sorgues is acute. A sustained kinetic strike on either site, a sabotage incident causing a multi-month outage, or even a serious industrial accident, would constrain bloc-wide artillery output for as long as recovery took. There is no spare capacity inside the bloc to substitute. Imports from the United States are politically unreliable. Imports from Asia are operationally unreliable. The result of an Eurenco outage would be a hard ceiling on bloc shell production for as long as the recovery took, which would be measured in months at minimum.

Russian and Chinese intelligence services are aware of this and have been mapping the European energetics supply chain in detail. The Bergerac site sits inside French territory and within range of long-range cruise missiles launched from Russian submarines. The Sorgues site sits in southern France, more difficult to reach but not impossible. Neither facility has dedicated air defence as of early 2026.

Hardening these facilities should be a 2026 priority. The cheapest single line item available for high strategic effect is dedicated air defence and physical-security hardening at Bergerac and Sorgues. Together that programme would cost less than half a billion euros. It is not in any current public defence procurement. It should be.

Closing

The European defence debate spends its time arguing about platforms because platforms are visible, photogenic, and easy to count. The chemistry is invisible, unphotogenic, and difficult to count. That asymmetry of attention has produced an asymmetry of investment. The bloc has agreed to spend several hundred billion euros on platforms and on the SAFE facility that funds them. It has not yet agreed to spend the relatively small amount required to make sure those platforms have anything to fire.

Eurenco's Bergerac and Sorgues sites are the most strategically important small facilities in European defence. The strategic policy attention they deserve is roughly the attention currently being given to the FCAS programme, the GCAP programme, or the British carrier readiness debate. They are not getting that attention. Until they do, the bloc's munitions ceiling is a chemical limit, not an industrial choice.

The relevant sentence is short. European defence policy is bottlenecked by chemistry, the chemistry has an address, and the address needs more money than it is currently getting. None of those three facts is in dispute among people who work in the field. None of them has yet been said in serious English-language defence-policy commentary in those terms.

That is the gap.

This piece reflects independent analysis. The author writes in a personal capacity.