Nanobubble
A nanobubble is a bubble of gas in a liquid that is less than a micrometer across. The word covers three different things: bubbles that sit on a solid surface, called surface nanobubbles, whose existence is established; bubbles free in the liquid, whose existence is still argued; and bubbles trapped in a pore, which concern anyone who builds a solid-state nanopore because one of them can cut the current through the pore or fill the record with noise. All three are puzzling for the same reason. The surface tension of water presses so hard on a bubble this small that the gas inside should be forced into solution almost at once.
The pressure inside
A curved water surface presses on the gas it encloses with the Laplace pressure, 2γ/r, where r is the radius of the bubble and γ the surface tension of water, about 0.072 N/m at room temperature. The pressure is inversely proportional to the radius, so it climbs steeply at the scale of a pore.
| Radius | Excess pressure | In atmospheres |
|---|---|---|
| 5 µm | 0.029 MPa | 0.28 |
| 500 nm | 0.29 MPa | 2.8 |
| 50 nm | 2.9 MPa | 28 |
| 5 nm | 29 MPa | 284 |
A bubble 5 nm in radius has a volume of 520 nm3, about half a zeptoliter, and at 29 MPa the ideal-gas law puts about 3,000 gas molecules in it. A gas at that pressure dissolves into the liquid around it. A classical estimate of the lifetime of a bubble of this size is microseconds.[1]
Surface nanobubbles
Atomic force microscopes began to image bubbles about 10 to 100 nm high and several times wider than that on water-immersed hydrophobic surfaces around 2000, and they lasted for hours or days. The lifetime that theory gave and the lifetime that was observed differ by nine orders of magnitude or more. The explanation that the review of Lohse and Zhang favors is pinning: the line where the bubble, the liquid and the solid meet sticks to chemical or geometric irregularities of the surface, and a bubble held in place by its edge cannot shrink.[1]
In nanopores
In a nanopore the surface is the wall, and the wall of a freshly made pore is often hydrophobic. A bubble that lodges in it takes up room that the electrolyte should occupy and the pore conducts less than the formula in Pore conductance predicts, or not at all. Smeets and colleagues measured the same solid-state pore repeatedly while moving it through the focus of an infrared laser. The conductance and the low-frequency noise changed greatly from run to run, with noise power differing by up to five orders of magnitude, and they concluded from a model of a cylinder holding a bubble that nanobubbles were the dominant source of noise in solid-state nanopores.[2] A bubble that grows or detaches during a run shows up as a step or a burst in the current, which can be taken for a molecule.
Bubbles come from three places. They are trapped when the pore is first filled, they leave solution as dissolved gas when the liquid warms or the pressure drops, and they are made at the electrodes if the voltage is high enough to split water. The last is avoided by using Ag/AgCl electrodes, whose reaction makes no gas.
The remedies are the ones in Nanopore wetting: cleaning the pore in a plasma to make its walls hydrophilic, filling it first with a liquid of lower surface tension, such as ethanol, that is then exchanged for water, and degassing the solutions. The open design of the MR1 Nanopore Reader specifies, for the interposer that carries the chip, ten minutes of ultrasonic cleaning in isopropanol, ten minutes in ultrapure water and three minutes of plasma at 30 W "for bubble-free priming". Its firmware includes a pulse of 800 mV for 250 ms "for pore wetting / unclogging".[3]
At Molecular Reality
Kent Kemmish named wetting the "biggest, ugliest, dumbest" problem of the company's early work with focused-ion-beam pores, and most of those pores were broken in the first week while the team worked out a protocol (see Dumb problems).[4] The company's flow cell is made partly of polytetrafluoroethylene, which repels water and is a place where air bubbles stay.
The longest recording in the public Streams archive, 23 minutes and 51 seconds, is titled "airbubbletest-tap-wter". The 200-micron laser-drilled pore is large enough that a bubble in it is a visible object and not a matter of surface tension: at a radius of 100 µm, the Laplace pressure is 1.4 kPa, or 1.4 percent of an atmosphere. In its critique of its own plans for machine learning, the company's essay on a world model tells itself to count a signal as a discovery only when many devices see it: "a lone anomaly is a bubble".[5]
See also
References
- ^ a b Lohse, D.; Zhang, X. (2015). "Surface nanobubbles and nanodroplets". Reviews of Modern Physics 87 (3): 981-1035. doi:10.1103/RevModPhys.87.981.
- ^ Smeets, R. M. M.; Keyser, U. F.; Wu, M.-Y.; Dekker, N. H.; Dekker, C. (2006). "Nanobubbles in solid-state nanopores". Physical Review Letters 97 (8): 088101. doi:10.1103/PhysRevLett.97.088101.
- ^ "MR1 Nanopore Reader". GitHub. October 16, 2026.
- ^ Kemmishtree (March 28, 2026). "Focused Ion Beam pore". EQBio-OS.
- ^ "Keltar's Critique v2: The Nanopore World Model (Hardened)". Molecular Reality Corporation. June 2026.