Here I want to take a closer look at a recent development in the field of plasma or ionic loudspeakers. I feel it represents a wide range of issues with work in this field. I in no way want to detract from the hard work of the academics involved.
The design uses the speaker for noise reduction, specifically in confined spaces such as in gas ducts or possibly in areas requiring sight through it or better low frequency performance. It has been worked on for over 3 years in academia and now has a patent and has been sold to a start-up company.
Naming
There are three academic papers and a patent, all by the same author (leader of a team) about the same device from 2020 to 2023 with the following titles:
All of these are about the exact same transducer - what would be "commonly" and fairly accurately be known as a corona wind speaker, ionic loudspeaker or similar.
Is it an actuator? - not really, it's a transducer or loudspeaker.
Is it plasma based? - not really, there are corona discharges which are a low energy type of plasma but the active mechanism producing the vast majority, if not all, of the sound is the ion or electric wind.
Are they plasmacoustic - no, this is a made up word at most
Do they have metalayers - no. At most there are two, not many (as meta would suggest), and the corona discharge counts for little to none of the acoustic output so when considering the active areas the ion or electric wind "layer" is the active one.
"deep subwavelength" - you mean at best subwavelength, otherwise known as thin. Conventional loudspeakers are thinner.
If you read the content of the papers, even more ways of describing the same physical effects are used. Sometimes this may be accidental but there are certainly cases where this academic work just starts using marketing speak like "plasmacoustic" and "metalayer" deliberately. Things were relatively sane until the final paper - it certainly got more attention.
The patent uses none of the hyperbole and if you want to read a much better representation of the idea then I would recommend reading the patent.
Data Representation
In the third paper measurements of the ionic speaker are done to demonstrate its use with noise reduction. It is placed at one end of a tube and sound is played in the other end. A microphone is used to pick up the resulting sound waves and their attenuation.
Note that this is not in a free field. With no restrictions, noise cancelling is far harder without a massive speaker between you and the noise source. This is why active noise cancelling is generally applied to headphones and not speakers, and it is already very effective in headphones without this type of device. Unless your noise source is isolated and down a tube this system may not be very useful.
All data is represented in terms of the absorption level, 1 being total and 0 being none. This produces some excellent looking data with an almost flat response over a frequency range indicating almost total sound reduction from 20Hz to 2kHz.
From the abstract of the paper "perfect sound absorption".
Humans hear over an accepted range of 20Hz to 20kHz (not 2kHz) and we have analogue hearing that is broadly logarithmic in nature. Audio result data is commonly presented in sound level on the decibel scale to reflect this.
The 2kHz limit is apparently due to what the paper says is "theoretical assumption of an infinitely thin plasmacoustic layer is not valid at high frequencies". Make of that what you will, 2kHz is not considered a high frequency. The original raw test data is downloadable from the paper. The tests actually run from 10Hz to 3kHz and above 2kHz the results are bad with almost no absorption for some areas. It starts to look more like the inverse of a typical frequency response from an ionic speaker with no equalisation or feedback controls.
So picking the data range and plotting on a graph with a linear 0 to 1 y axis where almost all the data is near 1 makes for a convincing show of a flat response.
Media
Once the final paper was published in Nature Communications it was widely reported on by the science news community. The majority of reporting simply suggested that sound/noise can now be completely eliminated by the use of these devices. Done deal.
Reading the abstract and introduction would lead you to think this. As an example from the end of the introduction:
"In this paper, we propose to leverage the inherently non-inertial dynamics of ultrathin layers of air plasma to construct fundamentally broadband active plasmacoustic layers that can manipulate sound over >2 decades, with systems of unprecedentedly small sizes, down to λ/1000. Our method can manipulate an acoustic field by directly steering fluid particles, without resorting to any non-fluidic interface, simply by leveraging the partial ionization of air, controlled by an electrical field. Our theory of plasmacoustic metalayers unlocks the possibility to use the associated acoustic monopolar and dipolar sources to control the plasmacoustic surface impedance. We experimentally demonstrate practical applications of the concept to extremely thin and broadband perfect sound absorbers and tunable mirrors, with bandwidth/size ratios at least 3 orders of magnitude larger than the currently available solutions."
"air plasma" - is it air or a plasma, they are different things.
They now have a new theory of "plasmacoustic metalayers", something no one has done before because it is made up.
If you got through that it is mostly a combination of hyperbole mixed with obfuscation from scientific and pseudo-scientific words. In short, its an ionic loudspeaker which could reduce noise from 20Hz to 2KHz in a tube.
Reality
So in reality what we have is a basic ionic loudspeaker, a couple of papers written on that and a third paper demonstrating the use in a lab for noise cancelling down a tube.
Then we have a patent. This actually covers the design of the speaker in diagrams and the bulk of the writing, with only some attention paid to using the speaker for noise cancelling. Then all of the claims and titling are focused on it being a noise cancelling device.
This would be mainly because the speaker itself is not novel. So to get a patent you must obfuscate the design, or at least find some new angle. Now the patent can be used to try to stop anyone using their ionic speaker design for noise cancelling even though there is little novel there apart from the specific application in a tube.
However, now we have a patent and having had several years of funding at EPFL (tens of thousands of euros) for the research the whole thing has been sold to a start-up company, Sonexos, who will try to leverage the IP. The owner of that company has previously started up and sold audio related companies so he should know what he is doing, but the reality of this design and its real application means it is much more limited than the science-market-speak in a lot of its literature suggests. In its limited use case, especially in ducting, this may be very good with some more work - but why all the extra claims and fluff?
It is quite possible that people have been caught up in the word salad and pseudo-science and over estimated what the device is really capable of. Not to say the actual research isn't valid, it is interesting, just much more limited than implied by reports.
Conclusion
It seems that the scientific community is not immune to being caught up in the excitement that an ionic or plasma device can create. These devices are great, but their application is limited - explore the possibilities but resist the desire to add what we could call "unnecessary information" to what is an already fascinating subject.
Just my opinion.
