12

Optical microscopes are quoted as having a maximum magnification of 1500x to 2000x - what is this calculated from?

Rupert Maximum
  • 123

1 Answers1

13

If you stick 200x lens and 20x eyepiece - you theoretically can have 4000x magnification, but you would not be able to see more details compared to 100x lens and 20x eyepiece, because resolving the smallest visible details is limited to Rayleigh criterion (i.e. limited to diffraction).

enter image description here

Where λ is wavelength in nm, and NA is numerical aperture of the lens. So for violet light λ=405nm, and good lens with oil immersion (NA=1.25), you can have resolution 197nm.

So, in conclusion, optical microscopes are limited to ~x1500 because going any further does not resolve smaller details.

Community
  • 1
BarsMonster
  • 2,371
  • Thanks for your answer. Please could you clarify, how does 197nm relate to 1500x? – Rupert Maximum Sep 24 '12 at 05:31
  • 2
    No direct relation. What I say is that you can make microscope with 10'000x magnification, but you won't see more details. So, typically at 1500x magnification human eye can comfortably see 197nm details and higher magnification is not required nor helpful. – BarsMonster Sep 24 '12 at 11:16
  • 1
    To put some numbers in, 1500x magnification makes 197nm features look like they're about a third of a millimeter. You can zoom in on the image but it will get really grainy after that. – Emilio Pisanty Sep 24 '12 at 14:25
  • Thank you. How does this relate to digital image capture - presumably the magnification has to take into account the pixel size of the detector? So you might need a higher magnification to get the same resolution as an eye? – Rupert Maximum Sep 24 '12 at 16:22
  • With digital cameras usually noone talks about "magnification". Usually what is important is how many microns/nanometers is in each pixel. For example, on my microscope and my 5mp camera, on 10x lens 1 pixel=408nm. And this information is usually more important than magnification. – BarsMonster Sep 24 '12 at 19:57
  • 1
    @EmilioPisanty Actually (although we're not using precise arguments here) I would use the word "smeared" rather than grainy (although "grainy" will happen with a pixellated image). The key idea here is lowpass spatial filtering. So if you choose lenses for 4000x magnification, say, then the distance between two resolvable objects would truly seem to be magnified by 4000x if the imaging system's field of view is wide enough. However, at such extreme magnifications, the resolvable object size is a significant fraction of the field of view. – Selene Routley Sep 21 '13 at 00:16
  • As you imply, magnification is not really a meaningful number, aside from as an intermediate datum used to calculate the meaningful data of imaged field of view and numerical aperture. – Selene Routley Sep 21 '13 at 00:19