It is a novelty you pay for. UV leds have been limited to 395 and the odd 375 nm ones for quite some time, with shorter wavelength at astronomical prices.
Surely the output power of these is very small, but they are relatively affordable. I'm not really sure what you would need them for, but if there is some application that works much better on 351 nm compared to 375 i'm sure the market will mature eventually.
I can imagine they'd be useful for curing resins or something like that when output power picks up a bit.
Indeed, the photoelectric effect can be very important for some applications.
351 nm could be very good for applications like uv-curing resins and such. 395 nm may not cut it there no matter how intense, the energy per photon is just too low to initiate the reaction.
I'm sure sure application whould be niche for now, but once these things roll off the production lines with more power and affordable prices i can see them used in applications like sterilization, ozonation and such.
The proof is in the measurements ofcourse, but i'm assuming they are what is advertised unless proven wrong.
I'm not really sure what the limit of wavelength is where UV becomes totally invisible. Perhaps even 350 nm may be somewhat visible if looking directly into the source, although that would not be a wise thing to attempt.
There is a big difference between lasers and leds when in comes to monochromaticity. A 351 nm led will probably have a gaussian wavelength profile that perhaps drops to 1% at 375 nm or something like that.
The 355 is a dpss with an extremely narrow output bandwidth, and would be under 1% within a few nm of its operational output.
Hmmm...we could take the gaussian function describing intensity as a function of wavelength and multiply it by the curve describing relative human eye visibility as a function of wavelength to get a weighted curve of the visibility of LEDs vs. lasers in these wavelengths.
Too bad the visibility curve probably has a lot of variation from person to person at these wavelengths...but this would be fun to investigate.
It works the other way, too. A 1W 365nm LED still has a bit of power at 351nm. Probably more than the 5mm LED mentioned at the beginning. (my spectrometer's response drops off rapidly under 370nm or so.)
The problem with near-UV sources is that they just make everything fluoresce, so it's very hard to tell true invisibility without looking directly into them. And even then, if it fluoresces a bit within your eye, does it could as invisible?