Dumb problems
Dumb problems is the name Kent Kemmish, the founder of Molecular Reality Corporation, gives to the routine obstacles that stand between a solid-state nanopore experiment and its first useful data: getting liquid into the pore, keeping bubbles out, not breaking the chip, stopping leaks. In a 2026 post he defined them as "the simple sounding things that are actually what make the lab bench hell when you're trying to get these things to work", and named the worst: "The biggest, ugliest, dumbest one is: how do you get water through the pore?"[1]
The post, written in 2026 about experiments of around 2018, ends: "These 'dumb problems' of frontier solid-state nanopore research are precisely why making the MR devices as easy to use as toys is everything."[1]
Definition
The post that introduced the term, "Focused Ion Beam pore", continues:
It takes insane levels of persistent, quiet focus to overcome such "dumb problems", none of which are really interesting, they're mostly established, known protocols/situations that people have to solve in a bespoke fashion in almost every situation so they can then do the exact same thing 100 times.
Kent Kemmish, "Focused Ion Beam pore", 2026[1]
Solving one, in his account, is not a discovery: "When you 'solve' these problems, you don't make any breakthrough, you've just established a place to start for actual frontier exploration."[1]
The case history
The post tells the story of the company's best early experiment, at its office on Edison Way in Menlo Park, California, around 2018. The pores were silicon chips drilled with a focused ion beam by a researcher "3,000 miles away", at "about $100 each". Two colleagues spent "an entire week, maybe two" on dumb problems before the run. Of the chips, Kemmish wrote: "in the first week or so of messing around with these, trying to determine the right wetting protocol, we broke most of them."[1]
Then one afternoon worked. The team had "a beautiful, fully wet, stable 200 nm underpore", visible in the company's video of the run as "the tiny white nipple you see in the middle of the dark circle", with the current rising and falling as a blade slid across it. Kemmish called it "the best single afternoon of collecting data we ever achieved" and noted that he was not in the room. The experiment was not repeated: "The tragedy is that we didn't have hundreds of these pores, or, with a bank account approaching zero, any time to try it hundreds of times."[1]
Wetting
A solid-state pore conducts only when it is completely filled with electrolyte, and a hole a few nanometers wide does not always fill. Gas trapped in it is held by surface tension: the Laplace pressure across a curved surface of radius is , so for water (surface tension about 0.072 N/m) a bubble of radius 5 nm is held in by about 29 megapascals, roughly 280 atmospheres. A 2006 study from Cees Dekker's laboratory at Delft linked such nanobubbles inside solid-state pores to excess electrical noise.[2]
The company's open MR1 Nanopore Reader design specifies ten minutes of ultrasonic cleaning in isopropanol, ten minutes in ultrapure water and three minutes of air or oxygen plasma at 30 watts "for bubble-free priming". Its firmware includes a voltage pulse "for pore wetting / unclogging", by default 800 millivolts for 250 milliseconds.[3] The company's guide describes the experience from the user's side: "wetting a new pore, coaxing water into a hole it would rather reject, can take hours of patience."[4] See nanopore wetting and nanobubble.
Other examples
- Clogging
- Particles and molecules stick in pores and stop the current. The company's 2022 game console had a "zapper stick" that applies a brief spike of maximum voltage, listed among other uses for "clearing clogged smaller pores".[5] The company's sliding pores can instead be opened to flush debris. See clogging (nanopore).
- Leaks
- Salt water that escapes the intended path carries current that the amplifier cannot tell apart from the signal. The casting guide in the MR1 repository puts it in one sentence: "a physical leak is an electrical leak". The same design surrounds the wet zone with a weep moat, a trace meant to detect stray saline so that the firmware can stop the pumps and set the bias to zero.[3]
- Air
- The first recordings in the company's public archive of live streams are of air, with names from air-test-10 to air-test-99. A later one is titled "airbubbletest-tap-wter".[6]
Influence on the company
The company's 2022 overview gave two reasons that solid-state nanopore research had been slow, the first being that "hands-on ss-nanopore research is extremely tedious and expensive", and proposed a console that "replaces the frustration of failed experiments with the fun and challenge involved in game play".[5] The idea is described in Experigaming. Asked why the company's live stream is often quiet, its guide answers: "the experiments take hours of preparation for minutes of signal, and much of the work is unglamorous: cleaning, wetting pores, swapping flowcells, staring at baselines."[4]
In the same post Kemmish asked whether the 2018 setup could have caught and read a DNA molecule, and answered: "Maybe, maybe not." The underpore that afternoon was 200 nm wide, about a hundred times the width of a DNA double helix, though the sliding blade could narrow the opening.[1]
See also
References
- ^ a b c d e f g Kemmishtree (March 29, 2026). "Focused Ion Beam pore". EQBio-OS.
- ^ Smeets, R. M. M. (2006). "Nanobubbles in solid-state nanopores". Physical Review Letters 97 (8): 088101. doi:10.1103/PhysRevLett.97.088101.
- ^ a b "MR1 Nanopore Reader". GitHub. October 16, 2026.
- ^ a b "Keltar's Field Guide to Molecular Reality". Molecular Reality Corporation. August 2026.
- ^ a b Kemmishtree (March 28, 2026). "2022 sci-tech overview". EQBio-OS.
- ^ Lenora Ashgrove (July 29, 2026). "Nineteen Billion Ions a Second, Live on the Internet". Signal & Noise.