CERN’s Future Circular Collider: A Huge Mistake?

Updated October 2026 with the figures from the 2025 FCC Feasibility Study and the European Strategy update adopted by the CERN Council in May 2026.

CERN’s plan for a Future Circular Collider (FCC), a ring of about 91 km that would first collide electrons with positrons and later protons at about 85 TeV, has generated both enthusiasm and scepticism in the particle physics community [1, 2].

Proponents point to the FCC’s ability to push into uncharted territory in energy, luminosity and precision, opening new windows on phenomena beyond the Standard Model (SM) such as dark matter, and on questions out of reach of the Large Hadron Collider (LHC) [3, 6]. The price gives pause. The first stage alone, the tunnel plus the electron-positron collider and its four detectors, is estimated at CHF 15.3 billion [2], and the proton collider would come on top of that. The question is whether such an investment is the best strategy for advancing particle physics.

This ScienceShot examines the main motivations put forward for the FCC and weighs them against its economic, environmental and opportunity costs.

The FCC and the New Physics It Could Reach

A central motivation for larger accelerators is to push the energy frontier and probe particles and interactions that no machine has reached. The Standard Model itself predicts none of them, but many of its proposed extensions do: supersymmetric partners, new heavy Z’ bosons or dark matter particles could sit in the 10 to 100 TeV range, which would need a collider far more powerful than the LHC to produce them [6, 8].

The 2019 conceptual design aimed for 100 TeV collisions in a ring of about 100 km [3]. The 2025 feasibility study shortened the ring to 90.7 km and set a baseline of 14-tesla niobium-tin magnets, giving about 85 TeV, six times the LHC’s 14 TeV design energy; high-temperature superconducting magnets could push that towards 120 TeV if they mature [1, 2]. The 2019 design also targeted a peak luminosity of 5 × 10³⁴ cm^-2 s^-1 at first, rising to 3 × 10³⁵ in a later phase, thirty times the LHC’s design value, which would give far larger samples of rare processes [3]. Ring size also sets how much energy the beam radiates as it bends, and the accelerator power calculator works that out for a given beam energy and bending radius.

In addition to producing new particles directly, the FCC would make high-precision measurements of known ones. Deviations from SM predictions in precision tests could offer indirect evidence of new particles and interactions [6]. The electron-positron stage, FCC-ee, would run at energies from the Z pole at about 91 GeV up to 365 GeV, above the top-quark pair threshold. Over fifteen years it would produce about 6 × 10¹² Z bosons, 2.4 × 10⁸ W pairs, almost 3 million Higgs bosons and 2 million top-quark pairs, improving many electroweak, QCD and flavour measurements by one to two orders of magnitude over LEP [4].

Heavy-ion collisions in the hadron stage would also create quark-gluon plasma (QGP) at higher temperatures and in larger volumes than the LHC can, with collective effects visible even in high-multiplicity proton collisions [7]. That would probe the QCD phase diagram under more extreme conditions and shed light on colour confinement.

These capabilities make a compelling physics case. The question is whether the scale and format of the FCC are the best investment.

Costs and Tradeoffs of a 91 km Collider

When CERN first laid out its options in 2019, the full 100 TeV hadron collider was priced at about €21 billion [5]. The 2025 study puts the first stage, FCC-ee, at CHF 15.3 billion spread over about 15 years from the early 2030s, with the tunnel’s civil engineering about a third of it [2, 26]. In May 2026 the CERN Council named FCC-ee its preferred option and asked for a financially feasible funding plan, to be worked out in talks with Member, Associate and non-Member States and the European Union, before a decision targeted for 2028 [23]. Karl Jakobs, the strategy secretary, has said the project as envisioned needs resources beyond CERN’s present budget [27].

Building and operating a ring of this size takes vast spending on civil engineering, cryogenics, magnets and detectors, which limits the money available for other science. There is also a risk of cost overruns of the kind that sank earlier colliders. The Superconducting Super Collider in Texas was cancelled in 1993 after its estimated cost grew from about $4.4 billion to roughly $11 billion or more [9]. Any FCC plan needs generous contingency, which raises the upfront cost.

Power consumption is another cost. The study estimates that FCC-ee would use 1.1 to 1.8 TWh of electricity a year, depending on the operating energy, against about 1.3 TWh a year for CERN today. The hadron stage would draw about 360 MW, or 2.3 TWh a year in its reference scenario, possibly below 300 MW if the magnets run at 4.5 K [2]. Over decades of operation, the electricity bill would run into billions.

Digging a 91 km tunnel, installing cryogenic and ventilation plants and handling activated materials also has an environmental cost, through excavated spoil, disturbance to local habitats and the carbon embedded in concrete and steel [25].

For scale, CERN’s own direct emissions were about 170,000 tonnes of CO2 equivalent in 2024, mostly fluorinated gases leaking from detector gas systems, plus about 67,000 tonnes attributed to its electricity use, which reached 1.29 TWh that year [10]. A machine that uses more electricity than CERN does today will add to that unless its power comes from low-carbon sources.

The funds and resources consumed by the FCC also carry an opportunity cost, limiting the budgets available for astroparticle physics, laser and plasma accelerators, or large-scale computing.

Investing billions in a single megaproject leaves less room for these complementary approaches than a balanced portfolio would.

Alternative Strategies for Exploring the Energy Frontier

New Ways to Accelerate Particles

Rather than scaling up existing collider designs, some concepts accelerate particles in entirely new ways.

Laser-driven acceleration focuses intense, ultrashort laser pulses into a plasma to create electric fields strong enough to accelerate electrons to high energies over very short distances [11, 13]. As petawatt lasers with femtosecond pulses become more widely available, they could become compact drivers of energetic particle beams. However, despite success in the lab, laser-driven accelerators do not yet have the efficiency, beam quality and repetition rate a collider needs.

Plasma wakefield acceleration is another promising technique [12, 13]. Sending an intense laser pulse or particle bunch through an ionized gas sets up a wake with electric fields that can reach around 100 GV/m. Trailing particles that ride this wake gain energy over centimetres rather than kilometres, with gradients orders of magnitude higher than in conventional radiofrequency cavities. At CERN, the AWAKE experiment has already used proton bunches from the SPS to drive such a wake and accelerate electrons [14].

If these techniques can control beam emittance and energy spread while scaling up beam power and energy, plasma accelerators may one day drive colliders far more compact than an FCC-scale ring. Test facilities such as FACET-II at SLAC are working towards the gradients, efficiency, stability and beam brightness a future collider would need [15].

Studying Cosmic Messengers and Astroparticles

Rather than rely only on accelerators built on Earth, physicists can also study particles produced naturally across the cosmos. Experiments such as IceCube, ANITA and the Pierre Auger Observatory probe the origins of cosmic rays, neutrino properties and possible dark matter signals by detecting particles from energetic astrophysical events [16]. Instruments in space add more: the Alpha Magnetic Spectrometer on the International Space Station has measured cosmic-ray antimatter with high precision [17], and the CALET telescope, on the ISS since 2015, has measured the cosmic-ray electron and positron spectrum into the TeV range [18]. As detectors grow, using radio or acoustic sensors spread over vast natural volumes, their sensitivity to rare processes keeps improving.

Underground experiments such as LZ, XENONnT and SuperCDMS search directly for dark matter and constrain its properties in ways that complement colliders [19]. Axions and axion-like particles, and violations of fundamental symmetries, can also be probed efficiently with table-top experiments [20]. Compared with large accelerators, these experiments can achieve substantial physics impact on modest budgets.

Harnessing High-Performance Computing

Supercomputers reached the exascale in 2022, and computing opens other avenues for particle physics. Lattice QCD calculations derive quantities such as quark masses and the strong coupling from first principles, and large Monte Carlo productions model the background processes the LHC experiments must understand to interpret their data.

In my view, cheaper computing can stretch budgets further than single experiments, by guiding accelerator design and theory before anything is built. Simulation cannot replace measurement, though: it can only predict what a known theory would produce. Machine learning is already improving how experiments analyse their huge datasets [21].

Colliders Beyond Earth

Some physicists have looked even further. One proposal sketches a hadron collider of about 11,000 km circling the Moon, which with 20-tesla magnets could reach about 14 PeV, a thousand times the LHC [22]. The authors present it as a next-next-generation machine, and the practical hurdles in launch, construction, power and maintenance are enormous. Such ideas are thought experiments for now, useful mainly for asking what the energy frontier could look like after the FCC.

Our Opinion

A 91 km collider like CERN’s Future Circular Collider sets its sights on precision far beyond LEP and, in its second stage, on energies well beyond the LHC.

Such a facility could expand our knowledge of fundamental particles and early-universe conditions, from the Higgs boson to dark matter and quark-gluon plasma. But the costs give pause on whether other approaches might yield a higher return on investment.

A CHF 15 billion first stage also limits the money available for a diversified programme of novel acceleration methods, astroparticle physics, large-scale computing and other complementary techniques. We therefore argue for a balanced portfolio that pursues several collider concepts and detector technologies in parallel, instead of a narrow bet on one facility. Laser and plasma wakefield accelerators, as they mature, could eventually make colliders more compact and economical than a larger copy of today’s ring designs.

Meanwhile, astroparticle experiments use nature’s own beams, and supercomputing makes theory and design cheaper to test. All of these alternatives have their own challenges today, but investing in them seems prudent to hedge the risk of committing almost everything to a single megaproject. The CERN Council has now named the FCC-ee its preferred option [23], and China has put its competing CEPC on hold until at least 2030 [24]. That makes the next two years of debate, before the 2028 decision, all the more important.

References

• [1] FCC Collaboration (2025). Future Circular Collider Feasibility Study Report, Volume 2: Accelerators, Technical Infrastructure and Safety. arXiv:2505.00274.
• [2] CERN Courier (2025). FCC feasibility study complete.
• [3] FCC Collaboration (2019). FCC-hh: The Hadron Collider. European Physical Journal Special Topics, 228, 755-1107.
• [4] FCC Collaboration (2019). FCC-ee: The Lepton Collider. European Physical Journal Special Topics, 228, 261-623.
• [5] Castelvecchi, D. (2019). Next-generation LHC: CERN lays out plans for €21-billion supercollider. Nature, 15 January 2019.
• [6] Golling, T., et al. (2017). Physics at a 100 TeV pp collider: beyond the Standard Model phenomena. CERN Yellow Reports, 3, 441-634.
• [7] Dainese, A., et al. (2017). Heavy ions at the Future Circular Collider. CERN Yellow Reports, 3, 635-692.
• [8] Giudice, G. F. (2008). Naturally speaking: the naturalness criterion and physics at the LHC. In Perspectives on LHC Physics. World Scientific.
• [9] Riordan, M., Hoddeson, L., & Kolb, A. W. (2015). Tunnel Visions: The Rise and Fall of the Superconducting Super Collider. University of Chicago Press.
• [10] CERN (2025). CERN Environment Report 2023-2024: emissions and energy.
• [11] Hooker, S. M. (2013). Developments in laser-driven plasma accelerators. Nature Photonics, 7, 775-782.
• [12] Joshi, C., & Malka, V. (2010). Focus on laser- and beam-driven plasma accelerators. New Journal of Physics, 12, 045003.
• [13] Leemans, W. P., et al. (2014). Multi-GeV electron beams from capillary-discharge-guided subpetawatt laser pulses in the self-trapping regime. Physical Review Letters, 113, 245002.
• [14] Adli, E., et al. (AWAKE Collaboration) (2018). Acceleration of electrons in the plasma wakefield of a proton bunch. Nature, 561, 363-367.
• [15] Yakimenko, V., et al. (2019). FACET-II facility for advanced accelerator experimental tests. Physical Review Accelerators and Beams, 22, 101301.
• [16] Ahlers, M., & Halzen, F. (2018). Opening a new window onto the universe with IceCube. Progress in Particle and Nuclear Physics, 102, 73-88.
• [17] Aguilar, M., et al. (AMS Collaboration) (2021). The Alpha Magnetic Spectrometer (AMS) on the International Space Station: Part II, results from the first seven years. Physics Reports, 894, 1-116.
• [18] Adriani, O., et al. (CALET Collaboration) (2018). Extended measurement of the cosmic-ray electron and positron spectrum from 11 GeV to 4.8 TeV with the Calorimetric Electron Telescope on the International Space Station. Physical Review Letters, 120, 261102.
• [19] Mount, B. J., et al. (2017). LUX-ZEPLIN (LZ) Technical Design Report. arXiv:1703.09144.
• [20] Graham, P. W., Irastorza, I. G., Lamoreaux, S. K., Lindner, A., & van Bibber, K. A. (2015). Experimental searches for the axion and axion-like particles. Annual Review of Nuclear and Particle Science, 65, 485-514.
• [21] Guest, D., Cranmer, K., & Whiteson, D. (2018). Deep learning and its application to LHC physics. Annual Review of Nuclear and Particle Science, 68, 161-181.
• [22] Beacham, J., & Zimmermann, F. (2022). A very high energy hadron collider on the Moon. New Journal of Physics, 24, 023029.
• [23] CERN Courier (2026). CERN Council updates the European Strategy.
• [24] CERN Courier (2025). CEPC matures, but approval is on hold.
• [25] FCC Collaboration (2025). Future Circular Collider Feasibility Study Report, Volume 3: Civil Engineering, Implementation and Sustainability. European Physical Journal Special Topics.
• [26] CERN. The Future Circular Collider.
• [27] Physics Today (2026). If CERN can come up with the requisite billions, the Future Circular Collider is a go. 18 June 2026.

@online{scienceshot-the-perils-of-pursuing-2023,
  author       = {Turner, Alex},
  title        = {{CERN's Future Circular Collider: A Huge Mistake?}},
  organization = {ScienceShot},
  date         = {2023-08-13},
  year         = {2023},
  url          = {https://scienceshot.com/post/the-perils-of-pursuing-larger-accelerators-do-we-need-the-fcc}
}