Illumination Captured In A Vacuum Jar

Experimentation with the unusual nature of things in the world is awesome… especially when the result is smokey glowing plasma. For this relatively simple project, [Peter Zotov] uses the purchase of his new vacuum pump as an excuse to build a mini vacuum chamber and demonstrate the effect his mosfet-based Gouriet-Clapp capacitive three-point oscillator has on it.

In this case, the illumination is caused due to the high-frequency electromagnetic field produced by the Gouriet-Clapp oscillator. [Peter] outlines a build for one of these, consisting of two different wound coils made from coated wire, some capacitors, a mosfet, potentiometer, and heat sink. When the oscillator is placed next to a gas discharge tube, it causes the space to emit light proportionate to the pressure conditions inside.


For his air tight and nearly air free enclosure, [Peter] uses a small glass jar with a latex glove as the fitting between it and a custom cut acrylic flange. With everything sandwiched snugly together, the vacuum hose inserted through the center of the flange should do its job in removing the air to less than 100 Pa. At this point, when the jar is placed next to the oscillator, it will work its physical magic…

[Peter] has his list of materials and schematics used for this project on his blog if you’re interested in taking a look at them yourself. Admittedly, it’d be helpful to hear a physicist chime in to explain with a bit more clarity how this trick is taking place and whether or not there are any risks involved. In any case, it’s quite the interesting experiment.

51 thoughts on “Illumination Captured In A Vacuum Jar

  1. This is interesting! I reminds me of an old post here on HaD about a suspended bubble of air in water blasted with ultrasonic frequencies of sound, and how it would sparkle and blink with tiny bits of light.

    On a second though, that soft ethereal glow is actually a fuckton of ultraviolet light. Might be a good idea to not leave it on next to your hand for too long, or stare at it for awhile without sunglassss for that matter….

          1. Normally this works great. Except for when I’m, say, on my phone, in a friend’s car, and my dyslexia decides to bug out. Words just stop making sense and it’s incredibly hard to edit all the errors away 100% of the time~

      1. Not Sonoluminescence. Sonolumiscence is produced by ultrasounds through a phenomenon called cavitation ( This circuit is a radio frequence oscillator. Nevertheless the way a gaz emit light is always a consequence of electrons falling at a lower level of energy. The electromagnic field transfert energy to electrons which step up to a higher level of energy. But higher levels of energy are not stable so electrons fallback to lower level emitting a photon. The color of light is determined by the energy of the photon according to Planck’s law. The energy of the photon is equal to energy released by the electron on is fallback.

    1. It is true that plasma can sometimes generate a nice amount of ultraviolet light. But luckily most normal glass jars are not transparent in the ultraviolet range… you would need special glass or quartz to be seriously exposed to ultraviolet radiation.

        1. The UV range spans from 380nm to 10 nm (380-200nm near UV and 200-10nm far UV), if you search for “transmission spectra glass” you will see that most glasses cut around 300nm.

          I’m a plasma physicist, so when analysing plasma with UV-visible spectrometers we use specific windows, quartz will cut around 170nm, Magnesium Fluoride around 100nm but with a strong attenuation. These specific optics cost much more than the glass optics we use otherwise.
          I’d be very surprised to find high quality quartz jars, so the risk of skin burn here is small.

          1. When I was doing research on Carbon Arc Lamps, I found that movie studios would put the arc behind a piece of glass for that same reason. The glass prevented the actors from getting sunburn, and probably made the lamps safer as well.(No exposed electrodes to get zapped on.)

    1. English is my second language. This is mostly an issue with household items, since I never learned their proper names while growing up; and googling them can be tricky.

      On a second thought, would it be called a “jar”?

      1. Yes, Jar is correct for non-metallic roughly cylindrical containers. Tin is appropriate for similarly sized containers made from metal, regardless of if it’s actually tin.
        I’m surprised online translators fail with simple things like jar. You can always read through the english version of the ikea catalog or something to find what you want to say :P

    1. citing
      “… our products are 98 percent efficient – offering less than two percent power dissipation of the energy that is actually converted into light.”

      I would be curious to know the real efficiency. The real efficiency is the ratio of visible light power emitted over the power consumed by the radio frequency emitter, which is not the same as what is stated here. This is a carefull phrasing for marketing purpose.

      1. Absolutely it’s advertising. There’s little to no hard data. I posted these to show that creating a plasma by electromagnetic coupling has been used.

        As I visit Tempe often, I’d really like to know which lights have been changed over, see how they start, color, etc.

  2. Very cool. This is halfway to a kind of DIY plasma cleaner. The colour will be related to the residual gas composition (in the writeup he said this colour change was permanent after a mini bake to 60dC). The light blue he ended up with looks like an oxygen plasma, while the pink/purple might be based on nitrogen … it’s kind of hard to say in this kind of rough vacuum regime.

    1. From my experience, pure oxygen plasma are yellow, the colour shown here looks more nitrogen to me. If the oil from the pump had a significant contribution I would expect the plasma to get lighter and the glass jar to show a coating tinting the glass (visible when the plasma is off).

      But the first colour could be attributed to residual coffee or cleaning products.

      As you say in rough vacuum virtually anything could contribute.

      1. There is no coating that I can see. And I don’t think there are any residues left–before taking these pictures, I used it, mostly with plasma inside, for a few hours–and not only it always had this color, but also anything left on the walls should have been long cleaned away.

        1. Plasma emission can be quite unstable, even extremely small contribution can influence the colour. The walls can “pollute” your plasma for several days, depending on the chemical compounds. But I was referring to wall contamination to explain the difference in colour induced by the baking.

          To tell more about the plasma we would need some Optical Emission Spectrometry.

          1. On a DVD you’ll have 1300 g/mm, which is more than enough resolution.
            It’s polycarbonate, so transmits in the visible range.
            I’m only concerned with the actual intensity from the plasma glow regarding the noise of the digital camera.

            But it’s definitely worth a try.

            I’ll have to think on a simple calibration for the DIY spectrometer.

  3. Not to be “that guy” but pulling a full vacuum in a glass vessel with flat sides is not a good idea. For those wanting to try this, use a canning (food preservation) type round jar as they are designed to hold a vacuum safely.

      1. It would most likely just crack–not that it necessarily will. Worst possible case, it shatters itself all around and I have to get out the dustpan. It’s not a pressure vessel, so the shards aren’t going to start moving fast.

        Safety glasses are a must, though.

        1. Again, I hate to be “that guy” but implosion is just as dangerous as a 1 atmosphere over pressure explosion. In a catastrophic event, the glass shards fly through the center of volume and out the other side at high speed.

  4. whitequark, could you please run an experiment for me please?

    I noticed that glass panes put one atop other tend to stick. Supposedly it is consequence of partial vacuum forming between the panes. This vacuum should get even lower in pressure if a frontal force is exerted upon on this glass “sandwich”. I even wondered if this effect could be used to easily (at home) form thin planar vacuum tubes (provided the attainable vacuum is very high and that we find way to stimulate electron emission without too much thermal stress put on the glass), but I digress.

    Now, my question is: will a glass plane sandwich contraption glow when put near your oscillator? That would confirm that there really is a vacuum between panes.

      1. Saying that a “vacuum” forms there is incorrect. There’s no empty space there. Rather, you merely squeeze out air, leaving nothing; and when you try to *separate* the panes, you would have to counteract atmospheric pressure. Same principle as a sucker, though glass is markedly less elastic.

        As per vacuum tubes, try putting a single hair between the panes. You will notice they no longer stick together.

        1. You are very likely right. However, I didn’t just pull that idea out of (no pun intended) thin air, and based it on trivial observation that “it sticks” (alright, I had to start from something). I believe It is reasonable to assume that when there is a gap in some nonmetallic material, or a contact between two physical objects of same material, there are repulsive Van Der Waals forces between electrons in outer shells of atoms residing on respective surfaces of the two objects. It means there is, however small, a standoff and because of it a minuscule volume of empty space in between. This space should be sufficient to allow free electrons, if there were any, to move through equipotential plane in the middle of the gap, where resultant force is zero. Now, if this gap glows in RF field, that would mean it is large enough to accept even larger particles, such as air molecules, which would probably reduce or exclude feasibility (if there ever was) of vacuum electronics in glass sandwiches.

          1. Van der Waals forces are weak and don’t create a vacuum between two glass plane. The pressure between the planes is very close to atmospheric pressure.

            This does not work.

  5. I’ve also heated the chamber to ~60°C while pumped down, which caused it to glow white rather than purple, even after it cools down: I’m not sure why this happens exactly.

    I guess you can get more vacuum, when you heat the vessel during evacuation because air expands. Maybe some residual air molecules are “evaporated” away.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.