Gravity measurements, atomic spectroscopy and trapped antiprotons test different foundations of physics. Recent results show how control of antimatter is turning ambitious questions into increasingly exact comparisons.
Antimatter research invites unusually large questions. Why does the observable Universe contain so much more matter than antimatter? Does an anti-atom fall like an ordinary atom? Could its spectral lines expose a flaw in the laws that describe particles? These questions share a scientific setting, but an experiment that answers one does not automatically resolve the others.
At CERN, the growing ability to produce, trap, cool and manipulate antiparticles is making the distinctions experimentally useful. A gravitational acceleration, an atomic transition frequency and a charge-to-mass ratio each test a different physical relationship. Progress now depends on selecting the right comparison and controlling the apparatus well enough to make a possible discrepancy meaningful. Results published in 2026 show that both the measurements and the infrastructure supporting them are advancing.
Falling antihydrogen and the meaning of a gravity test
In 2023, ALPHA-g observed how trapped antihydrogen escaped as the magnetic confinement was reduced. Antihydrogen consists of an antiproton and a positron and has no net electric charge. The experiment inferred a downward acceleration of 0.75 times ordinary gravitational acceleration, with uncertainties of 0.13 from statistical and systematic contributions and 0.16 from simulation. Within those uncertainties, the result was consistent with normal downward acceleration.
The central value did not demonstrate that antimatter falls 25% more slowly. The uncertainties were essential to the interpretation. The result ruled out the specific alternative of upward acceleration with the same magnitude as ordinary gravity and established attractive gravitational behaviour for the antihydrogen studied.
A precision test of universality requires going further. Magnetic fields are needed to hold the anti-atoms but can also influence their escape. The analysis must therefore connect observed annihilation locations to particle trajectories and the changing trap. This is a controlled measurement of motion, not a simple filmed drop of an antimatter object.
The same atom offers other tests
Gravity is only one part of the programme. Spectroscopy measures the frequencies needed to change an atom’s internal state. In 2018, ALPHA characterised the antihydrogen 1S–2S transition and found agreement with the hydrogen expectation to about 5 kilohertz out of roughly 2.5 quadrillion hertz. That corresponds to a relative precision of two parts per trillion for this comparison.
Such measurements test predictions associated with the combined symmetry of charge conjugation, parity and time reversal, usually abbreviated CPT. Within the relevant theoretical framework, matter and antimatter partners must have corresponding properties. Comparing atomic spectra tests that relationship through energy levels. It does not measure the gravitational acceleration of the atom, nor does agreement establish why the early Universe evolved towards its present matter abundance.
The practical gain from multiple tests is that a hypothetical new interaction need not affect every observable equally. A spectral transition sensitive to one combination of particle properties can complement a magnetic moment or a free-fall measurement sensitive to another.
A smaller energy gap brings new sensitivity
ALPHA’s 2026 measurement examined the ground-state hyperfine splitting, an energy difference associated with the relative spin configurations of the positron and antiproton. Microwave spectroscopy using roughly 24,000 anti-atoms reached four parts per million precision, improving the previous measurement by two orders of magnitude. The result agreed with expectations for hydrogen.
Although four parts per million sounds less impressive than parts per trillion, it describes a different transition. The researchers note that antiproton internal structure contributes at about 40 parts per million to this quantity. Reaching the reported precision therefore makes that structure relevant to the comparison. Scientific sensitivity cannot be ranked solely by the number of decimal places in unlike measurements.
A separate 2025 study demonstrated simultaneous measurements of both accessible hyperfine components of the 1S–2S transition. Its protocol characterised the relevant lines in one day, a 70-fold increase in data-taking rate. Faster measurements create opportunities to repeat observations under different conditions and examine systematic effects, as well as collect more statistics.
Single antiprotons provide a different reference
The BASE collaboration compares individual trapped particles using electromagnetic fields. Its 2022 charge-to-mass comparison achieved a fractional uncertainty of 16 parts per trillion, combining four long-term studies recorded over approximately a year and a half. The relevant comparison concerns the magnitudes of the proton and antiproton charge-to-mass ratios; their electric charges have opposite signs.
Measuring a ratio is different from separately determining charge and mass. The experiment provides an exceptionally stringent test of their relationship, while interpretation as a bound on one property requires assumptions or additional information about the other. That distinction prevents an accurate experimental result from acquiring a broader meaning than the apparatus can support.
BASE also measures the antiproton’s magnetic moment, which describes its magnetic response associated with spin. A 2017 measurement achieved 1.5 parts per billion precision at 68% confidence and was consistent with the corresponding proton result. This probes another aspect of particle–antiparticle symmetry through a different frequency comparison.
These single-particle experiments and antihydrogen spectroscopy are complementary because building an anti-atom introduces additional relationships between its constituents. Agreement at both levels tests a more connected account than either programme can establish alone.
Cooling and transport change what can be measured
Some advances concern the quality of the sample rather than an immediate new symmetry bound. In 2021, ALPHA demonstrated laser cooling of antihydrogen, reducing its median transverse energy by more than an order of magnitude. The observed 1S–2S spectral line was approximately four times narrower after cooling. Slower anti-atoms make their motion easier to control and reduce one source of broadening in spectroscopy.
In 2026, BASE-STEP demonstrated another capability: road transport of 92 trapped antiprotons over a 7.5-kilometre route without particle loss or measurable deterioration in trap vacuum. The transportable cryogenic trap is intended to let researchers move antiprotons to laboratories with lower magnetic noise than their production environment.
That achievement establishes transport, not the eventual precision improvement anticipated from relocation. Its significance is practical: the best place to produce antiprotons need not be the best place to make the most delicate frequency measurements. Separating those activities could allow a research programme to optimise both, provided transport and subsequent measurement conditions remain under control.
The common engineering requirement is stability over long periods. A trapped antiparticle is useful only if researchers understand its environment well enough to connect the measured signal to a property of nature. Vacuum, temperature, magnetic fields and timing therefore become part of the scientific argument.
This also changes the value of experimental time. Repeating a measurement reduces random uncertainty, but a stable bias can survive every repetition. Teams therefore need opportunities to vary the conditions: changing a field setting, preparing a different atomic state or checking the same signal through another detection route. Improved production and control make such comparisons less costly in scarce antiparticles. The engineering gain is consequently broader than a higher count rate. It can strengthen the ability to distinguish an instrumental shift from a physical difference between matter and antimatter.
Different routes to the next comparison
AEgIS is developing a beam-based route to antihydrogen gravity measurements. Its 2025 detector study describes a scintillator system intended to measure antihydrogen time of flight through a moiré deflectometer. Knowing the flight time helps interpret the vertical displacement accumulated during motion. The paper concerns detector design and commissioning, rather than a completed gravitational result.
A beam and a trap-release experiment offer different ways to connect motion with gravity. Their eventual comparison could therefore be valuable even if one initially has larger uncertainties. An apparent difference between matter and antimatter would become much more convincing if it survived changes in preparation, confinement and detection method.
For now, the clearest outcome is a growing set of well-defined comparisons consistent with established expectations. That is not a resolution of the cosmic matter–antimatter imbalance. It is a progressively more demanding test of the ingredients available to explain nature. The next decisive advance may be a discrepancy, or an improved agreement that removes a proposed explanation. Either requires knowing exactly which property has been tested and which experimental capability made the test possible.
