'Light leak' due to transmission through probe shank?

[volke...@gmail.com]

We performed some imaging in fluorescein solution to Show light Emission. This way we discovered light 'leaking' from the tip (even while using the most distant LED). Please check the attached Images. There you can also see light mmission from the sides of the shank. We wonder whether this is normal and if you are planning to eliminate this in future versions of the device. 

At the moment we still lack Information on the dimensions of the LED in our particlar case. This lets us struggle to describe the light power at the LED surface and therefore we can't declare whether the leaked amount is severe and overcomes the neuron threshold.

Your documentation is well done, but we lack knowledge which of the presented probe versions is the one we have.

Please Support us with the requested Information.

Thank you very much!

[John Seymour]

Hallo Volker,

Thanks for running that test and sharing. We have seen it in all of our images (especially in air where even more light gets trapped in the SiO2 insulation layer). The short answer is that Khan Kim, senior graduate student working on uLEDs, has offered to re-run his model on this but the first time he ran it the magnitude was too low to cause stimulation. But I'll let him reply when he has time give specific values. Still, I would agree this needs to be considered. Maybe it is a good thing in some situations, or are you afraid you cannot record from neurons on the edge?

We also had that question in the Mendrela 2018 publication. Our response to this reviewer question was...

A consistent light leakage seems to arise at the side of µLED illumination sites (Figure7, inset panel, top right). Have the authors measured the optical power that is leaked? It would be interesting to know the authors thoughts on this.

Reply: We noticed the light leakage and believe that the phenomenon is due to waveguide effect that thin and transparent SiO2 passivation layers on the optoelectrode shanks provide. Unfortunately, we were not able to accurately measure the amount of the leaked light (both the radiant flux and the irradiance) with equipment we have. However, we believe that the intensities, especially the irradiance, of the light leaking from the sides are significantly lower than those exiting the µLED top surface.

First, it should be noted that the amount of light that seems to be leaking from the sides of the optoelectrode in the figure is rather exaggerated since high forward current was used to make the illumination from the µLED stand out in the middle of the illumination from the microscope. Figure S2 shows images of the illumination from a µLED with low forward current values with no environmental illumination. It can be noted that the amount of light (indicated by both the intensity of the light and the area of the illuminated region) leaking from the sides is significantly lower than that of light exiting the front.

Second, the actual amount of the light leaking from the sides of the optoelectrodes in vivo should be considerably lower than that observed in the air, the environment in which the microscopic images were obtained, because the high refractive index of the brain tissue (n ≅ 1.36) makes light extraction from the front side of the optoelectrode approximately twice more efficient than that in the air (n = 1) by making the critical angle much larger (θc,air ≅ 42.7° , θc,brain ≅ 67.4°). We modified the caption in Fig. 7 to convey this point:

“Photographs of the assembled headstages. Insets show (top) microphotographies of the tips of the optoelectrodes, and (bottom) the schematic diagram of the polyimide-based flexible cable interposer. The light leakage from the sides of the optoelectrode shank, shown in the top right inset, is an artifact due to the combination of poor light coupling efficiency in the air and high optical output power. Modified from (Mendrela et al., 2017).”

If accurate measurement of the light intensity from all the directions is required, a (hemi-)spherical photodector array (Zhang et al., Nature Communications 2017) might be utilized for detection of the intensity of the light from all the directions around the µLED on the optoelectrode, while both the optoelectrode and the photodector array are submerged inside a liquid medium whose refractive index is similar to that of the brain tissue (e.g. ethanol).