A rice-grain microcomb makes many millimetre-wave tones at once, but the stability lives in a fibre loop

A team led by Loughborough University physicists, working with international collaborators, has shown that a microchip the size of a grain of rice can produce a spectrum of precisely spaced optical frequencies, and that those frequencies can be converted into several millimetre-wave signals at once ^[1]. That matters because the bottleneck in millimetre-wave systems has not been available bandwidth but the difficulty of generating those frequencies with the precision and stability advanced applications need ^[2].

The comb-to-millimetre-wave route itself is established: a microcomb is a dense set of optical frequencies, and a specialised antenna converts them into millimetre waves ^[3]. Earlier work used microcombs to generate a single precise millimetre-wave frequency ^[4]. Producing many simultaneously is what would open multiple parallel data channels, and it demands an exceptionally clean and stable comb ^[5]. The Loughborough-led paper, published in Nature Communications, reports that combination ^[6].

The mechanism is the interesting part. Microcombs are normally made by shining laser light into a microresonator on a chip ^[7]. This design pairs the chip resonator with a larger loop of optical fibre, so the light circulates continuously through both ^[8]. According to Dr Luke Peters of Loughborough's Emergent Photonics Research Center, the loop keeps feeding light back through the chip, which lets the comb states build up efficiently, start on their own, and stay stable when the system is disturbed ^[9]. Peters says the setup survived people jumping up and down next to it with the comb still stable ^[10], which is an anecdote rather than a measurement, but it is the right class of anecdote for anyone contemplating a mast or a satellite bus.

Two results carry the practical weight. The team showed it could shape the comb, making individual lines stronger or weaker, which Peters frames as necessary because different applications want different frequency combinations ^[11]. And it showed the comb's precision and stability were inherited by the millimetre-wave signals it produced ^[12] rather than lost in conversion.

The honest caveat is in the geometry. The chip is rice-sized, but the complete system is a tabletop laboratory setup ^[13], and the element doing the stabilising is an external fibre loop ^[8], so this is stability achieved on a bench that includes a chip, not stability achieved on a chip. Peters says the applications are still some way off and that challenges remain before the technology can be used in real systems ^[14][15]. The stated ambition is a more compact, more energy-efficient version, potentially shoebox-sized ^[16], with satellites named as a target because size, weight and power dominate there ^[17]. The applications the group points at are faster, higher-capacity 6G, radar, spectroscopy and astronomical instruments ^[15], plus precision timing, which Peters ties to emerging quantum technologies ^[18].

Watch for numbers, because this account carries none ^[19]: phase noise at defined offsets, comb line spacing, and how many millimetre-wave tones were actually generated and at what power. Watch how much of that fibre loop can be shortened or integrated without losing the self-starting behaviour, since that ratio decides whether a shoebox is plausible. And watch the timing work, which the group says it is still pursuing ^[18], because clock-grade accuracy is a much harder test than a comb that shrugs off someone jumping next to it.