Bulgaria’s plan to add two new nuclear units at Kozloduy is often presented as a question of timing. In reality, it is a question of structure, cost discipline, and system value. The upcoming state aid notification to the European Commission, expected to be submitted by Kozloduy NPP – New Build in the coming weeks, confirms political momentum. It does not yet confirm economic closure.
The final investment decision remains scheduled for late 2026 or early 2027. That alone places the project firmly in the category of long-horizon infrastructure, where early assumptions tend to be revised, sometimes substantially.
Capacity is easy to state. System value is harder to prove.
Each of the planned new units is expected to add roughly 1,100 MW of installed capacity. On paper, that places the combined output of Units 7 and 8 above 2 GW, enough to cover a significant share of Bulgaria’s baseload demand.
But capacity figures, by themselves, are misleading.
Bulgaria’s annual electricity consumption fluctuates around 35–38 TWh, while nuclear generation already accounts for roughly one third of domestic production. The real question is not whether the country needs more megawatts, but whether it needs firm, dispatchable capacity capable of stabilising a grid increasingly influenced by renewables.
This is where the choice of AP1000 becomes relevant.
Flexibility in numbers, not slogans
According to design parameters, AP1000 reactors can operate between 30% and 100% of nominal power, with ramp rates of approximately 3–5% per minute. In practical terms, this allows daily load-following and participation in primary and secondary frequency control.
That matters.
In 2024 alone, Bulgaria added several hundred megawatts of solar capacity. On days with high PV output, midday prices collapse, while evening ramps place stress on thermal units. Nuclear plants that cannot adjust output quickly become liabilities in such conditions. Those that can, become anchors.
Still, flexibility on paper does not automatically translate into flexibility in operation. Grid codes, market design, and operational culture will ultimately determine whether these capabilities are fully used.
Availability and economics over decades
Another frequently cited figure is availability. AP1000 reactors are designed for capacity factors above 90%, supported by fuel cycles of around 18 months. Over a 60-year lifetime, even small differences in availability translate into massive differences in total energy produced.
At 90% capacity factor, a single 1,100 MW unit produces roughly 8.7 TWh per year. Over six decades, that approaches 520 TWh—before any lifetime extensions.
These numbers explain why nuclear economics are unforgiving. A one-year delay or prolonged outage does not simply cost millions. It reshapes the entire financial logic of the project.
Safety statistics are not abstract anymore
Post-Fukushima, safety is no longer discussed in generic terms. The AP1000’s passive safety systems are designed to maintain core cooling and containment integrity for up to 72 hours without external power or operator action, relying instead on gravity and natural circulation.
This design philosophy aligns with updated international safety expectations promoted by the International Atomic Energy Agency. It also reflects a broader industry shift: reducing dependence on human intervention during the first critical hours of an accident.
Whether this translates into lower insurance costs, faster licensing, or higher public acceptance is less clear—and will vary by jurisdiction.
Environmental permitting: where timelines stretch
Environmental Impact Assessments rarely dominate headlines, but they dominate schedules.
Unit 7’s EIA process began in 2012 and reached legal finality only in 2019 after prolonged court challenges. Unit 8 faces an even broader process, involving five neighbouring countries and more than 60 institutions and NGOs.
Cross-border consultations add legitimacy, but also uncertainty. Even without formal rejection, procedural delays alone can push milestones back by years. Current expectations point to a final EIA decision for Unit 8 around 2027, already close to the planned investment decision window.
Financing ratios are simple. Conditions are not.
The project’s indicative financing structure around 30% equity and 70% debt is typical for large nuclear builds. What matters is not the ratio, but the cost of capital and risk allocation.
Export credit financing, supported by state guarantees, can significantly reduce borrowing costs. However, EPC contract terms, escalation clauses, and localisation requirements will ultimately determine whether the project remains bankable under stress scenarios.
So far, approximately BGN 500 million has been spent on preparation. That figure is manageable. What comes next will not be.
What the timeline actually tells us?
If first concrete for Unit 7 is poured around 2030, as currently envisioned, Bulgaria would be commissioning new nuclear capacity in the mid-to-late 2030s—precisely when coal exits accelerate and electrification deepens.
That alignment makes strategic sense.
Whether the economics align as cleanly will depend on decisions taken well before construction begins. For now, the Kozloduy expansion is less a construction project than a test of Bulgaria’s ability to manage complexity over time.
The numbers are promising. The gaps between them are where the real risk lives.
Why this matters for Power Loop readers?
For Power Loop readers, the Kozloduy new build project is not just another long-term infrastructure plan. It sits at the intersection of three trends that are already reshaping Europe’s energy landscape.
First, it is a stress test for how large, capital-intensive energy projects are financed in an era of higher interest rates and tighter public budgets. The structure of state aid approval, export credit financing, and risk allocation will set a precedent, not only for nuclear, but for other strategic energy investments that require long payback periods and political backing.
Second, the project highlights a shift in how system stability is valued. As variable renewables continue to grow, the debate is moving away from installed capacity toward controllability, flexibility, and availability. Nuclear units that can load-follow and support grid balancing challenge the outdated view that nuclear is incompatible with high shares of wind and solar. Whether markets and regulation catch up to this reality remains an open question.
Third, Kozloduy is a real-world case study in timing risk. Environmental permitting, cross-border consultations, and regulatory sequencing are already pushing critical decisions close to the end of the decade. For analysts, investors, and system planners, this raises an uncomfortable but necessary question: can Europe deliver complex energy projects fast enough to match its decarbonisation timelines?
For Power Loop readers tracking energy policy, grid development, and capital flows, the importance of this project lies less in its final capacity figure and more in what it reveals about how Europe plans, finances, and executes its energy transition under pressure.
