Explain Different Types of Light Microscopy

[patrickbotros]

Okay so first I would like someone to add detail to my descriptions of different types of light microscopy. Here's what I know:
Brightfield (unstained): standard view of partially opaque, live cells.
Brightfield (stained): standard view of colored, dead cells.
Phase Contrast: Not sure how it works but it gives a view of a stained, live cell.
Differential Interference Contrast/Nomarski: Not sure how this works but it gives 3D view of live cells.
Fluorescence: Uses proteins that on one end bind to specific organelles and on the other end glow under certain radiation light
Deconvolution: Not really sure but is colored
Confocal: Basically fluorescence only using a computer to only look at light from different depths so no out of focus light is included. Good for dead, stained cells.
Superresolution: Uses different fluorescent molecules that light up under different types of light and then combines all the images.
If I'm wrong about any of this please let me know.
Also why are electron microscopes so much more accurate than light microscopes?
Thanks :D

 

[atyy]

Also why are electron microscopes so much more accurate than light microscopes?

Light travels in straight lines. But can light bend around the corner? Yes, if it's wavelength is long enough. Bending around the corner will make images fuzzy.

Electrons have shorter wave lengths, so they bend less and cast sharper shadows, so resolution is better.

https://en.wikipedia.org/wiki/Diffraction
http://hyperphysics.phy-astr.gsu.edu/hbase/phyopt/raylei.html (this is a rule of thumb - one can do better if one has additional information - this is the basis of some forms of super-resolution http://advanced-microscopy.utah.edu/education/super-res/)

 

[patrickbotros]

Light travels in straight lines. But can light bend around the corner? Yes, if it's wavelength is long enough. Bending around the corner will make images fuzzy.

Electrons have shorter wave lengths, so they bend less and cast sharper shadows, so resolution is better.

https://en.wikipedia.org/wiki/Diffraction
http://hyperphysics.phy-astr.gsu.edu/hbase/phyopt/raylei.html (this is a rule of thumb - one can do better if one has additional information - this is the basis of some forms of super-resolution http://advanced-microscopy.utah.edu/education/super-res/)

Wait what? Why does wavelength affect scattering/bending? Is this just a fact or is there an intuition behind it?

 

[Andy Resnick]

Okay so first I would like someone to add detail to my descriptions of different types of light microscopy. Here's what I know:
Brightfield (unstained): standard view of partially opaque, live cells.
Brightfield (stained): standard view of colored, dead cells.

Yeesh.. you have a long list. So far, so good... basically- I assume you mean 'stain' like an H&E stain (as opposed to fluorescent stains...) And many of the techniques here can be done either with transillumination or epi-illumination.

Phase Contrast: Not sure how it works but it gives a view of a stained, live cell.

No- phase contrast is generally used on unstained cells. Contrast is generated by interfering scattered and unscattered light; phase objectives often have a mostly-opaque soot ring with a diameter matched to the condenser phase mask- this is called 'matched filters'.

Differential Interference Contrast/Nomarski: Not sure how this works but it gives 3D view of live cells.

No, it only appears 3-D. If contrast in phase contrast imaging is due to (spatial) differences in the refractive index, DIC creates contrast according to the spatial gradient of the refractive index. And it is possible to color-code the images with a spectacularly poor quarter-wave plate called a 'deSenarmont compensator'. Another term for DIC is 'wavefront shear interferometry', and these are variations of DIC including Hoffman modulation.

Fluorescence: Uses proteins that on one end bind to specific organelles and on the other end glow under certain radiation light

Sure, basically- there are primary and secondary antibodies involved, and the idea is to image where specific proteins are located.

Deconvolution: Not really sure but is colored

Deconvolution is conceptually like division. If the image is the object convolved with the point spread function, deconvolution attempts to reconstruct the original object. It's not a trivial computational exercise.

Confocal: Basically fluorescence only using a computer to only look at light from different depths so no out of focus light is included. Good for dead, stained cells.

Close enough- an illumination point is imaged onto the sample, and so is 'confocal' to the object. Depth sectioning is the main rationale for using a confocal system.

Superresolution: Uses different fluorescent molecules that light up under different types of light and then combines all the images.

No- superresolution is a constellation of imaging techniques that usually combine many images to produce a single image with resolution in excess of the diffraction limit. There are many techniques, my favorite is STORM, but to each their own.

If I'm wrong about any of this please let me know.
Also why are electron microscopes so much more accurate than light microscopes?

Thanks :D[/QUOTE]

I don't know about 'accurate', but the lateral resolution of an electron microscope is in excess of a light microscope, since the wavelength of the electrons is less than that of visible light. I don't know about longitudinal resolution; electron microscopes usually have a lot of spherical aberration.

You forgot about darkfield, oblique imaging, two-photon, conoscopic observation, Rheinberg illumination (a personal favorite), dispersion staining,...

Get thee to https://www.microscopyu.com/, post-haste.

 

[atyy]

Wait what? Why does wavelength affect scattering/bending? Is this just a fact or is there an intuition behind it?

It is just a fact that is described mathematically by the wave equation. This phenomenon is called diffraction.

But there is an intuition to it (it's a bit better for sound waves, and for light waves this is quite misleading, so discard this if you don't like it, better to learn the equations). If light were fundamentally a ray, then it makes sense that it would travel in straight lines. But if you think light is a wave, then it makes sense that it would not travel in straight lines. After all, if you put an obstacle in the path of waves in the sea, the waves can go around the obstacle. The bigger the obstacle, the less the waves can go around it. On the other hand, the bigger the wavelength, the smaller the obstacle is relative to the wave, so this gives some intuition as to why wavelength affects bending around an obstacle.

Actually, I don't know if this answer is misleading, so I will post your question in general physics, and see what answers we get.

Edit: Here is the thread I started in general physics with your question: https://www.physicsforums.com/threads/why-does-wavelength-affect-diffraction.834361/#post-5238693.