For many years, people have aimed to tap into the energy of stars to create electricity on Earth. And for almost as long, accomplishing this goal has always appeared to be about ten years away.

Currently, an array of startups are nearer than they have ever been and are hastening to construct fusion reactors capable of supplying power to the grid.

Fusion startups have attracted over $10 billion in funding, with more than a dozen securing upwards of $100 million. Numerous large funding rounds have concluded in the past year, as investors are drawn to the sector due to increasing energy demands from data centers and the progress being made by fusion startups.

At its essence, fusion energy aims to harness the energy emitted from merging atoms to produce electricity. Humans have understood how to combine atoms for decades, ranging from the hydrogen bomb—an instance of uncontrolled nuclear fusion—to various fusion devices constructed in laboratories worldwide. Experimental fusion machines have managed to control nuclear fusion, with one example generating more energy than was needed to initiate the process.

However, none have managed to yield a surplus sufficient to establish a viable power plant.

To address this challenge, fusion startups are exploring multiple methods. Experts hold differing views on which strategies have the best likelihood of success, though the sector remains in its early stages, meaning nothing is certain.

Here’s a concise summary of the primary methods for achieving fusion power.

Magnetic confinement

Magnetic confinement represents one of the most commonly used methods, employing strong magnetic fields to contain plasma, the mixture of superheated particles essential to a fusion device.

The magnets need to be exceedingly powerful. For instance, Commonwealth Fusion Systems (CFS) is constructing magnets capable of producing 20 tesla magnetic fields, which is roughly 13 times more robust than a conventional MRI machine. To cope with the enormous electricity needs, the magnets are crafted from high-temperature superconductors, which must still be cooled to –253˚ C (–423˚ F) utilizing liquid helium.

CFS is in the process of developing a demonstration device named Sparc on a significantly expedited schedule in Massachusetts. The company expects to power it up sometime in late 2026, and if successful, it plans to commence construction of Arc, its commercial power plant, in Virginia in 2027 or 2028. 

There are two primary categories of fusion devices that utilize magnetic confinement: tokamaks and stellarators.

Tokamaks were initially conceptualized by Soviet scientists in the 1950s, and they have been extensively researched since then. Tokamaks come in two fundamental forms: a doughnut shape with a D-profile and a spherical version featuring a small opening in the center. The Joint European Torus (JET) and ITER are two prominent experimental tokamaks; JET operated in the UK from 1983 to 2023, while ITER is slated to start functioning in France in the late 2030s.

Based in the UK, Tokamak Energy is innovating a spherical tokamak design. Its ST40 experimental apparatus is currently undergoing enhancements.

Stellarators form the other main variety of magnetic confinement devices. They are comparable to tokamaks in that they maintain plasma within a doughnut-like configuration. However, while tokamaks have geometric edges, stellarators are twisted and turned. The unique shape is established through modeling the plasma’s dynamics and optimizing the magnetic field to accommodate its characteristics rather than imposing a regular form.

Wendelstein 7-X, a substantial stellarator equipped with modular superconducting coils operated by the Max Planck Institute for Plasma Physics, has been functional in Germany since 2015. Several startups are also advancing their own stellarators, including Proxima Fusion, Renaissance Fusion, Thea Energy, and Type One Energy. 

Inertial confinement

The other primary method of fusion is referred to as inertial confinement, which compresses fuel pellets until the atoms within them fuse.

Most designs for inertial confinement utilize laser light pulses to compress fuel pellets. Multiple laser beams fire simultaneously, converging on the fuel pellet from every direction at the same time.

So far, inertial confinement is the only method that has achieved a landmark known as scientific breakeven, whereby the reaction yields more energy than it consumes. Such experiments have taken place at the National Ignition Facility (NIF) at the Lawrence Livermore National Laboratory in California. It’s important to note that measurements to determine scientific breakeven do not account for the electricity needed to power the experimental facility. 

Nevertheless, nearly a dozen startups are optimistic about inertial confinement and are creating reactors based on this concept. Notable examples utilizing lasers include Focused Energy, Inertia Enterprises, Marvel Fusion, and Xcimer.

There are two firms that are not employing lasers: First Light Fusion, which suggests using pistons, and Pacific Fusion, which intends to utilize electromagnetic pulses instead of lasers.

More to come

These are the two main strategies for achieving fusion power, though they are not the only options available. In the near future, we’ll provide additional information about alternative designs, including magnetized target fusion, magnetic-electrostatic confinement, and muon-catalyzed fusion.