Read the title before the marketing: US12650115B2, "Hall-effect thruster system with applied counter-torque," granted June 9, 2026 to the University of Michigan, is about a side effect most coverage of electric propulsion never mentions. A Hall-effect thruster accelerates ionized propellant through crossed electric and magnetic fields. That discharge is efficient and gentle — but it is not perfectly balanced, and the residual angular momentum it deposits has to go somewhere. It goes into the spacecraft, as an unwanted slow spin.

The independent claim already tells you the mechanism, and it is more mechanical than the abstract lets on. Claim 1 recites a thruster body plus "a diffuser configured to eject ionizable propellant into an annular channel formed in the thruster body," with "curvilinear propellant channels formed through a body of the diffuser." Each of those channels has an inlet end and an outlet end that are "axially spaced apart in the direction of a central axis of the diffuser." The claim then pins down the physics it is fighting: "a Hall current provided in the annular channel induces a torque on the thruster body in a first rotational direction, and ejection of the propellant from the diffuser applies a counter-torque to the thruster body in a second rotational direction opposite the first rotational direction." In other words, the propellant is not simply pushed straight back — it is fed through curved passages so that it leaves with a swirl, and that swirl reacts against the body to cancel the spin the plasma current would otherwise impart.

"A Hall effect thruster system includes a thruster body and a diffuser configured to apply a torque to the thruster body during operation."— U.S. Patent No. 12,650,115 source

The dependent claims are where the moat sits, and they describe how the geometry is realized. Claim 2 specifies that the diffuser ejects propellant "with a tangential velocity"; claim 3 narrows the curvilinear channels to "helical channels"; claim 4 has the diffuser "formed as a single continuous piece with the thruster body," i.e., integrated rather than bolted on. Claim 5 captures the elegance of the approach — the counter-torque is applied "without reversing a polarity of a magnetic field," so the fix costs no extra power supply or magnet reconfiguration. Claims 7 through 9 tie the swirl direction to the orientation of the radial magnetic field (inward- versus outward-pointing), and a parallel independent claim (claim 10), plus a method claim (claim 16), extend the same idea to the act of operating a spacecraft so as to "counteract the torque induced on the thruster system body by the Hall current." The CPC classification — F03H 1/0062 for the ion/plasma engine, with B64G 1/413 tying it to spacecraft electric propulsion — confirms this is a spacecraft-propulsion claim, not a lab plasma curiosity.

Why does this matter for the people who actually fly these things? Because momentum management is a hidden tax on every long-duration electric-propulsion mission. Reaction wheels saturate; desaturating them costs propellant or power; and a thruster that quietly torques the bus makes the whole problem worse. The abstract is explicit that the swirl torque comes from "the ionizing Hall current flowing in an annular channel," and that the counter-torque is "useful in deep space applications outside the Earth's magnetic field" — precisely the regime where you cannot lean on magnetic torque rods to bleed off momentum. Building the counter-torque into the thruster is an attempt to shrink that tax: fewer desaturation cycles, longer wheel life, less coupling between the propulsion and attitude subsystems.

The prosecution-history caveat applies, as always on this desk: a granted claim covers what its language covers, and the protected configuration is specifically a diffuser with axially-offset curvilinear (or helical) channels imparting a tangential exit velocity — not a monopoly on the idea of balanced thrusters. The fact that this came out of a university, not a prime, also matters; it is the kind of foundational mechanism that gets licensed into commercial thruster lines rather than flown by its inventor.

One detail in the claim set rewards a second look: the swirl is geometry, not active control. Because claim 4 allows the diffuser to be "formed as a single continuous piece with the thruster body" and claim 12 (on the parallel independent claim) fixes it so it "does not rotate with respect to the thruster body," the counter-torque is produced by a static, shaped part with no moving elements, no second propellant feed, and no electrical penalty. The propellant simply takes a curved path on its way into the discharge channel and leaves with the tangential velocity claim 2 requires. That is a meaningfully different design philosophy from canting the whole thruster or adding a dedicated reaction-control jet: the disturbance is removed inside the same hardware that creates it, at essentially zero added system cost.

The field-orientation claims (7 through 9) are what make the geometry general. Because a Hall thruster's swirl torque direction depends on whether its radial magnetic field points inward or outward, a fixed swirl would only cancel the disturbance for one field configuration. The claims instead tie the outlet offset direction to the field direction — radially-inward field paired with one handedness, radially-outward with the other — so the same inventive principle covers thrusters of either magnetic polarity. For an operator running electric propulsion for years at a stretch, the practical reading is simple: the patent describes a way to make the thruster's own exhaust do the momentum bookkeeping that reaction wheels and desaturation burns would otherwise have to do.

But the direction is unmistakable, and it is the direction the whole smallsat-electric-propulsion sector is moving: integrate the control problem into the propulsion hardware instead of bolting on more attitude actuators. When a constellation operator has to keep hundreds of platforms pointed for years on a power budget, a thruster that doesn't fight the bus — and does so with nothing more than the shape of its propellant passages — is worth more than a thruster with a slightly higher specific impulse. This grant is a bet on exactly that trade.