There's something very funny to me about building an enormous machine so we can ask questions about incredibly small things, and China's High Energy Photon Source is an excellent example. Its storage ring is roughly 1.36 kilometers around, while the research it supports can involve the structure of a tiny sample or a change happening inside a material. I like that mismatch in scale, partly because the building looks like it should be doing something much larger and louder. Instead, the interesting part might be what a battery electrode is doing, or where a small defect has formed in a piece of metal.
The facility is usually called HEPS, and it's in Beijing's Huairou District, where the Institute of High Energy Physics has developed a new synchrotron light source. That means it produces intense light for experiments, especially X-rays, rather than looking out into space like a telescope. I love space, so I could easily get distracted by a machine described as seeing the universe, but this story brings us much closer to the objects around us. There's plenty of interesting science inside an ordinary material, provided we can build the equipment, prepare the sample, and work out how to ask it a useful question.
A Very Large Ring With Electrons Going Around It
In a synchrotron light source, electrons are accelerated to very high energies and circulate in a storage ring, guided and focused by magnets. When their direction changes, they emit electromagnetic radiation, which can be collected and sent toward experimental stations. Special arrays of magnets called undulators make the electrons follow a series of small sideways deviations, helping produce especially bright X-rays. The storage ring and the experiments therefore have different jobs, with the big accelerator supplying the light and the stations using it to investigate samples.
HEPS uses electrons at an energy of 6 gigaelectronvolts, or 6 GeV, which is the six-billion-electronvolt figure that appears in descriptions of the project. That's an energy measurement, rather than a speed or a number of pictures per second. The electrons travel close to the speed of light, while the equipment keeps their path and beam properties under careful control. I find the control almost as interesting as the speed, because getting particles moving very quickly is only useful here if the light they produce can reliably reach the experiment in a form somebody can use.
The Experiments Happen Along the Beamlines
A beamline takes light from the source through equipment that selects and shapes it for an experiment, with the sample and detectors farther along the setup. Different beamlines can use different energies, focusing arrangements, and measurement techniques, which is why one facility can support a range of research at the same time. HEPS's first general proposal call listed 14 user beamlines, including protein crystallography, engineering materials, hard X-ray imaging, structural dynamics, and high-pressure research. That's a much better way to understand its range than thinking of HEPS as one enormous camera with one setting.
The measurements can also look quite different from the pictures we might expect. X-rays passing through a sample can help form an image, while scattering and diffraction patterns give information about structure, and absorption measurements can reveal chemical details. Computers and models help turn those measurements into something we can interpret, so the final result may be a reconstructed view, a structural model, or a spectrum. I tend to want a picture immediately, because I am apparantly very easily persuaded by a good visual, but some of the most useful information arrives in a form that needs a little more explanation.
Brighter Light Has a Fairly Specific Meaning Here
When researchers talk about the brilliance of an X-ray source, they're considering how much useful light comes from a small area, within a narrow spread of directions and a defined energy range. Having lots of photons is helpful, but so is being able to deliver them where the experiment needs them. HEPS belongs to a newer generation of synchrotrons designed around very compact, well-controlled electron beams, using arrangements of many magnets to reduce a property called emittance. Lower emittance helps make the source more useful for experiments that need concentrated or coherent X-rays, rather than simply making the laboratory look more brightly lit.
That distinction gives the engineering a practical purpose, especially when a sample is small or the signal being measured is difficult to collect. Researchers may want to focus on a tiny region, distinguish a subtle feature, or follow a change while the sample is working under selected conditions. The instrument has to supply the right kind of information for that task, which is why the optics and detectors matter alongside the source. I like the amount of cooperation hidden inside the phrase “a better light source,” because a great many pieces have to behave themselves before a scientist gets a useful result.
A Picosecond Is Tiny, but the Number Still Matters
A picosecond is one trillionth of a second, and it really is difficult to get an intuitive sense of how short that is. The important point for HEPS is that an experiment's time resolution depends on the light pulses, detectors, timing, and technique being used. A 2025 overview written by the HEPS project team described temporal resolution on the order of 100 picoseconds, which is about one hundred trillionths of a second. That's remarkably fast, while being a different claim from saying that every experiment can film events one picosecond apart.
Spatial resolution and time resolution answer different questions, too, since locating a tiny feature and distinguishing two closely spaced moments aren't the same measurement. An experiment designed to show fine structural detail may make different choices from one designed to follow a fast change. I find those trade-offs useful to know because they make the subject much more interesting than a single astonishing number. The challenge becomes deciding what we want to learn and arranging the measurement to capture that information, which is a very satisfying sort of problem to give an enormous machine.
What We'd Actually Want to Put in Front of the X-Rays
Battery research is one application I can appreciate fairly quickly, because I'd be very happy for the batteries in everyday things to last longer and behave more reliably. Synchrotron X-rays can help researchers investigate changes in battery materials and how those changes relate to performance and degradation. The useful questions concern what happens inside the material during use, with the experimental setup designed around the kind of cell or component being studied. A test cell prepared for a beamline can give access to information that the sealed battery in a finished product would make much harder to obtain.
Engineering materials provide another set of questions, including where defects form and how a structure changes during a manufacturing process. The HEPS team reported pilot work involving defect detection in engineering materials and high-speed imaging and diffraction of metal 3D printing during commissioning. I like that example because “a piece of metal” can sound like a fairly settled object until you start asking about its internal structure and the way it was made. A part may look entirely ordinary from the outside while containing details that matter a great deal to how it performs.
Protein crystallography brings us to a different scale and a different kind of sample, using diffraction measurements from crystals to help determine three-dimensional molecular structures. That information can contribute to work on enzymes and to drug design, including understanding how a candidate compound interacts with a target. The work still depends on preparing suitable material, interpreting the data, and testing ideas beyond the structure itself. I find it wonderful that the same broad light-source technology can support questions about a metal part and a biological molecule, even though the people using it need very different expertise and experimental arrangements.
Getting a New Facility Ready Takes Several Milestones
HEPS began construction in June 2019, and the institute's chronology records a sequence of installation and commissioning milestones rather than one instant when the entire project became finished. The first stored beam came in August 2024, performance acceptance followed in October 2025, and pilot operation was announced on December 3, 2025. Those stages involve making different parts of the facility work and checking what they can do together. I think that sequence gives us a much better sense of the project than a description of a giant building that suddenly switches on and starts answering every question at once.
In March 2026, HEPS invited its first round of general research proposals from users worldwide, covering all 14 user beamlines and stating a plan for official operation in June. The notice set out how researchers could apply and directed them toward beamline information and the user manual. That's an actual invitation with a process behind it, rather than a guarantee that any proposed experiment can happen immediately. As of this October 2026 update, the dated milestones and notices show the transition toward user research, without needing to turn an earlier planned opening date into an independently confirmed announcement.
There's a Whole World of Synchrotrons
HEPS is a major addition to a field with important facilities in several countries, including the European Synchrotron Radiation Facility in France, Diamond Light Source in the United Kingdom, and the Advanced Photon Source in the United States. The upgraded Advanced Photon Source delivered its first X-rays for science in June 2024, which is a useful reminder that this is an active international field of engineering and research. A facility's electron energy alone doesn't rank every experiment it can perform, because the beam properties, instruments, techniques, and scientific question all play a part.
The collaborations are interesting in their own right, too, with HEPS and Spain's ALBA signing a cooperation agreement in 2025 and IHEP working with Brazil's CNPEM on a joint synchrotron laboratory. Those examples involve accelerator technology, beamlines, and scientific exchange, giving the idea of cooperation something specific to stand on. I'd much rather hear what teams can learn from one another than treat every new instrument as the end of science everywhere else. There are enough difficult questions about materials and biology to keep considerably more than one country occupied, and I suspect the samples won't mind whose passport is sitting in the office.
The People Still Have to Work Out What the Data Means
One of the most interesting parts of a large facility is everything that happens around the beam itself, including sample preparation, choosing the measurement, and deciding whether the result answers the original question. The HEPS team described commissioning work on optics, positioning, control systems, and data acquisition alongside the accelerator work, which helps show how much has to come together. I tend to get distracted by the large hardware first, but the experiment might depend just as much on a tiny sample being in the right place. A very impressive source can't rescue a question that the chosen measurement doesn't address.
I'd want to hear how researchers decide which question is worth their beam time, especially when several techniques could offer different pieces of an answer. The interesting result might be an unexpected pattern, a defect they can finally locate, or evidence that an explanation needs changing. That leaves room for the less tidy side of research, where a better instrument gives us something new to puzzle over before it gives us something new to make. I like that possibility, even though it means the story continues well beyond the moment somebody produces a striking image.
A Giant Machine for Questions That Keep Getting Smaller
What draws me to HEPS is the amount of effort behind that ability to ask a more specific question, from the accelerator and magnets to the person preparing a sample for a beamline. There's something appealing about bringing all that equipment together to investigate a tiny detail that might otherwise stay hidden. The detail could help explain a familiar problem, or it could start a line of research we hadn't thought to follow. Either way, I'd like to know what somebody saw, how they measured it, and what they think it means, preferably with enough explanation that the rest of us can enjoy the interesting part too.
I still love the scale of the place, though, because a ring more than a kilometer around is a fairly extraordinary thing to build in pursuit of small answers. I can admire the machine and remain curious about what researchers actually do with it, including the experiments that turn out differently from what they expected. A battery material, a protein crystal, and a metal component might arrive with completely different problems, while all benefiting from a carefully controlled source of X-rays. That's the part of this story I'd like to keep following, the variety of things we can learn when somebody gives us a better way to look.
A USB digital microscope would give my own tiny questions somewhere to start. Take a leaf or a grain of salt, look more closely and see how quickly an ordinary bit of the kitchen becomes the afternoon’s main event.
Related Reads from the Archive
Rewinding Time at the Cellular Level: How Scientists Made Human Skin Cells 30 Years Younger
Will Nanorobots Make Us Immortal by 2030? The Future of Forever
Sources
Chinese Academy of Sciences: HEPS background and ring circumference
HEPS project team: Commissioning overview in the APS 2025 newsletter
Advanced Photon Source: First X-ray light after its upgrade, June 2024
