Nanoplasma-enabled picosecond switches for ultrafast electronics

Wonder how they turn it off. Plasma recombination is schlooowwwww.

Cheers

Phil Hobbs

And does it wear out?

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

Almost certainly. Produce a plasma in air and you've got ozone and nitric oxide, which rapidly oxidises to nitrogen dioxide, which reacts with water vapour to make nitric acid.

For a while some place in Norway made nitrate fertiliser by blowing air through an electric arc and capturing nitrogen dioxide produced.

Neither the polyimide substrate nor regular metal tracks would to last too long.

A teflon substrate and noble metal tracks might do better, but an air plasma is very aggressive.

If the gap is narrow enough and the electric field is high enough, the electrons might only be being generated by cold field emission, and might get across the gap without ionising too many oxygen or nitrogen molecules, but then you wouldn't have a plasma.

You would be looking at the high-field low-gap end of the Paschen curve

The authors of the Nature paper should know all about that, but might not bother telling their target audience.

--
Bill Sloman, Sydney

Less than 20 ns, apparently. The paper is downloadable from sci-hub.tw with DOI 10.1038/s41586-020-2118-y .

It's amusing how carefully the authors avoid using the term "spark gap." I wonder if there's any way to create a sub-100 nm gap without resorting to ion beam etching? J. C. Bose probably did it in 1895, somehow...

-- john, KE5FX

Useful link.

"Townsend avalanche is the dominant plasma formation mechanism for g > 5 ?m, while field-effect emission and tunnelling are dominant for

5 nm < g < 5 ?m, and g < 5 nm, respectively18. d, The higher electric field in a shorter gap distance results in a much faster electron transport for nanoplasma devices."

I suspect that their "nanoplasma" is just free electrons.

The mean free path of an electron in air seems to be about 500nm, which is the upper limit of the gaps being talked about, and the electron has to pic k up 15.6 eV of energy to ionise a nitrogen molecule, and the fast switchin g still seems to be there at voltages below that.

--
Bill Sloman, Sydney

Sure. You build it vertically and then cleave the wafer.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC / Hobbs ElectroOptics 
Optics, Electro-optics, Photonics, Analog Electronics 
Briarcliff Manor NY 10510 

http://electrooptical.net 
http://hobbs-eo.com

Doesn't seem to be what they've done.

The pictures are of something planar.

--
Bill Sloman, Sydney

Sparks gaps can dump megawatts in picoseconds. There's nothing else like that.

NLTLs, shock lines, can do around 10 volts with rise times of a few ps, and won't wear out.

--

John Larkin         Highland Technology, Inc 

Science teaches us to doubt. 

  Claude Bernard

Yeah. When I saw the circuit, I had to laugh. When I was a teenager in the 1950s, I found the following book in a curio store:

"Wireless Telegraphy", by Dr. J. Zenneck, translated fron the German by A.E.Seelig, E.E., First Edition, McGraw-Hill 1915, 443 pages.

Still have it. All about Spark-gap transmitters, Poulsen Arcs, Alexanderson Generators, and the like. Vacuum tube generators just becoming practical, for small stuff. Alexanderson Generators are still used in industry, to generate RF power for induction heating.

Joe Gwinn

Oh yeah.

As for wear, if one were to do this for real, I'd guess that one would build the devices of ceramic and a refractory metals, like hydrogen thyratrons.

The spark gap can do 100 volts, no problem.

In frequency-multiplication service,NLTLs are 10 to 30 dB quieter than SRDs, but expensive.

Joe Gwinn

There have been some high-voltage NLTLs. The McEwan guy made them from discrete inductors and diodes, nanosecond and kilovolt stuff. There have also been some planar structures, metal strips on a slab of nonlinear ceramic.

--

John Larkin         Highland Technology, Inc 

Science teaches us to doubt. 

  Claude Bernard

I recall those from McEwan. I was tempted to make one, but it was too much work. If I recall, one needed 50 or 100 sections for it to work well. A Blumelin line was easier:

Joe Gwinn

Now free on archive.org:

(Not that anybody actually reads all the way through scanned e-books.) :(

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC / Hobbs ElectroOptics 
Optics, Electro-optics, Photonics, Analog Electronics 
Briarcliff Manor NY 10510 

http://electrooptical.net 
http://hobbs-eo.com

Thus, the vacuum-sealed mercury wetted relay is a popular variant, if the small duty cycle isn't an issue. Ozone, too, means that one doesn't want an airgapped set of points in the breathing air.

Having trouble visualizing this. Is there a diagram somewhere?

I tried cutting a slot in the top layer of a copper PCB strip using a scalpel, the idea being to pinch it shut manually until it fires somewhere near the bottom of the Paschen curve:

(closeup) (test setup for HV edge) (test setup for 50 ohms)

Driving it at -350V through a 100K resistor makes the gap fire somewhat randomly at about 100 V/ns, limited by the probe and various strays. The recovery time is also swamped by the scope probe RC:

A 50-ohm series tap is faster, but likely still limited by strays:

(MSO6054A, 500 MHz BW) (TDS 694C, 3000 MHz BW)

I did see some edges closer to .35/3000 = 117 ps, but 164 ps was the fastest one that I saved before the PCB strip finally broke.

-- john, KE5FX

I haven't read the paper, because very small spark gaps aren't that interesting--you can turn them on pretty fast, but (as in this case) turning them off takes thousands of times longer, which makes them of little interest in applications. (Not to mention their probably very short lifetime.)

You can get edges nearly that fast from a properly designed mercury-wetted relay. One of Jim Williams's book chapters talks about an ancient Tek pulser that worked that way. It's all about minimizing inductance.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC / Hobbs ElectroOptics 
Optics, Electro-optics, Photonics, Analog Electronics 
Briarcliff Manor NY 10510 

http://electrooptical.net 
http://hobbs-eo.com

Barefoot thyratrons and krytrons aren't that fast--you need a pulse-forming network such as a shock line.

Interesting. My first task in my first engineering job (mid-1981) was to rejigger an SRD multiplier to use a different diode. (The old diode was self-biased with a bit of conductive ink, so I scribbled on the new one with a soft pencil to find the right value--fun.)

However, I've never used a shock line multiplier. Do you have a reference handy?

Thanks

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC / Hobbs ElectroOptics 
Optics, Electro-optics, Photonics, Analog Electronics 
Briarcliff Manor NY 10510 

http://electrooptical.net 
http://hobbs-eo.com

is a huge collection of old Electr onics magazines and books that were aimed at Hobbyists and professionals. A ll can be downloaded for free. They even have the four early Morgan books ' Radio book for Boys' My school had the first three, back in the '60s. I did n't know that there was a fourth book. Archive has a lot of old Amateur rad io magazines, as well. These items were what fired my interest in Electroni cs as a kid. Instead of spending my allowance on candy or toys, I spent it on magazines. I still have a few of them, but most were lost or damaged whi le in storage over 30 years ago.

suggests that these gaps - which are very narrow - 1um down to 200nm - do turn off pretty quickly. The rise time is less than 1nsec (which is to say at least as fast as the 1GHz scope can measure), but this is shown with a 1

0MHz pulse train, so the nanoplasma has gone away in less than 50nsec.

It does look as if the authors "nanoplasma" is more like a burst of field-e mission electrons. If they do produce any positive ions as they move throug h the air gap, those positive ions may capture electrons and remain electro nically excited very briefly but they are in a very small volume and can lo se energy pretty quickly.

I'd be interested in a high resolution image of edges of the gap. Field emi ssion is enhanced at sharp points, and the edges of a metal gap are going t o show their crystalline microstructure at this scale. Repeated operation m ay sharpen the points - arc discharges are sustained by field emission from a surface which has got hot enough to deform into a two dimensional array of spark spikes under the influence of a high electric field.

Arc lamps are physically a lot bigger and do tend to have tungsten electrod es - the centre of the arc is rather hotter than the surface of the sun.

The nano-gaps aren't going to get anything like as hot, but the metal at th e surface of the gap may get warm enough to move around a bit (one atom at at time) when the electric field is high.

--
Bill Sloman, Sydney