Fluid-driven ‘gears’ could out-perform mechanical devices
14 January 2026 
US researchers have created a gear mechanism that uses fluids, rather than mechanical teeth, to generate rotation. They believe their invention could lead to a new generation of devices that offer greater flexibility and durability than existing gears.
“We have invented new types of gears that engage by spinning up fluid rather than interlocking teeth, and we have discovered new capabilities for controlling the rotation speed, and even direction,” says Jun Zhang, a professor of mathematics and physics at New York University (NYU), where the research was performed.
Gears are among the oldest machine parts, dating back to 3,000 BC in China, where they were used in chariots. However, their teeth – whether wood, metal or plastic – are inflexible and vulnerable to breaking, and must interlock perfectly to work.
“Regular gears have to be carefully designed so their teeth mesh just right, and any defect, incorrect spacing, or bit of grit causes them to jam,” explains Leif Ristroph, an associate professor of mathematics at NYU, who was a member of the team. “Fluid gears are free of all these problems, and the speed and even direction can be changed in ways not possible with mechanical gears.”
The researchers, from NYU’s Courant Institute School of Mathematics, Computing, and Data Science, wanted to know if it was possible to make devices that work like gears, but that don’t need teeth to be in contact with each other to function.
Recognising that air and water flows are used to rotate structures such as turbines, the team, led by Zhang, hypothesized that fluids could also serve as gears’ “teeth”, if their flows could be directed accurately.
They conducted an series of experiments that included immersing cylinders or rotors in a glycerol-water solution whose properties, such as viscosity and density, they could manipulate.
One cylinder was powered actively to rotate, while a second was unpowered or passive. The researchers hypothesized that the active cylinder could generate fluid flows that would cause the passive one to rotate. To monitor this, they added tiny bubbles to the mixture, allowing them to track the movement of the flows and to see how the fluids functioned as gears. They ran experiments with the cylinders at various separations, and with the active cylinder rotating at different speeds.
They found that the active cylinders, combined with fluid flows, could cause the passive cylinders to move in ways that were sometimes similar to gears, and sometimes more like pulleys connected by a belt. When the cylinders were close, the flows functioned in a similar way to teeth that engage on the facing sides of two gears, causing them to rotate. The swirling flows effectively gripped the passive rotor and made it spin in the opposite direction to the active one.
However, when the cylinders were further apart, and the active one was spun faster, the flows looped around the outside of the passive one, like a belt around a pulley that then rotates in the same direction as the active one.
Zhang, Ristroph and Jesse Etan Smith, an NYU doctoral candidate, have described their work in the journal, Physical Review Letters. They suggest possible uses for their discovery in gears and pulley systems – and point to potential advantages over current mechanisms.
