These look like simple coherent artifacts, caused by interference between your sample and illumination light. What is the illumination NA? This looks to me like your illumination coherence is rather high (like you're illuminating with a monochromatic, plane-wave laser?).

You probably want to decrease the source coherence by broadening the bandwidth (maybe using a spectrally broader 405nm LED rather than a laser?) and the apparent source size (by placing a diffuser in the beam path and imaging the Fourier transform of the diffuser onto the sample in a Kohler geometry).

Zachary Smith : Zachary, thanks for your prompt reply. We actually use a 405nm laser with coherence length longer than 50 meters (you are right, our illumination coherence is rather high). As you mentioned, those artifacts can be coherent artifacts, but we saw several repeating shadows with sort of strong intensity around. The sample is like a USAF target, where we put aluminum on a silica substrate (the thickness is 170 μm).

We initially doubt that there may be multiple reflections between the subtrate and the light comes out a few times and thus introduce the shadows around. Do you mean the light between our sample and illumination light will interfere and cause these artifacts? Could you please provide more details about this? Thanks so much.

BTW, Merry Christmas!

It's possible there could be multiple reflections, but I guess it's simply an artifact of you using a highly coherent imaging system. Regarding details, I would check some classic texts like Goodman's Fourier Optics which would give more detail about the coherent image formation process. To solve it, you will need to reduce your illumination coherence drastically by increasing the wavelength diversity (temporal coherence) or angle diversity (spatial coherence). The second can be achieved by increasing the apparent size of the source using a thick diffuser (which might need to be rotating if your coherence is really 50m).

Zachary Smith Hi Zachary, we have tried another light source using white light. Unfortunately, the artifacts are still exsiting though the imaing quality looks improved as attached. The sample is put properly on the focus plane and we can see the surrounding artifacts.

The coherence lenght of the whitelight should be very less at micrometer level, but we are still not sure of the reasons causing these artifacts.

Hava a happy new year!

What is the spatial coherence of your white-light source? Is it being focused on to the sample with a high angular diversity (e.g. using a condenser), or is the light simply placed a distance away from the sample? If it's far from the sample (more than ~10 focal lengths?) and isn't being focused or diffused, than only roughly parallel rays are passing through the optical system from the source to the image plane, meaning that the image still has a high degree of spatial coherence.

If that's the case, you probably need to focus or diffuse the light. Ideally the illumination would be coming from the source being placed in a Kohler configuration, or some reasonable approximation of Kohler. If you're already doing that and still getting these artifacts, then maybe it is something to do with multiple reflections (although using a low temporal coherence source would argue against that...). Could you throw the sample on a standard commercial microscope and see if the images look similar? That would at least help you understand if your microscope's design can be improved, or if it's something specific about this sample.