Re: Kohler illumination question...

GTO wrote:

Did we reach the limit of what we can accomplish with optical microscopy? I do not think so. An interesting step into the right direction was recently commercialized by Olympus (followed by others, of course) that is known as TIRF [1]. But more exciting ways to improve optics is the theory of surface plasmon polaritons, a rather new field in optics [2]. Surface plasmon polaritons offer a resolution beyond the diffraction limit and may open new gateways to future developments in optical microscopy while still be considered non near-field applications. Unfortunately, theoretical work on SPP's just started. More about SPP's is in [3].

We are essentally at the limit for far-field imaging. There are some trade-offs that can be exploited to produce subwavelength images (the lithography industry controls the phase in their mask to project extremely fine details, for example), but analysis is clear on this point- schemes to increase resolution (super-resolution filters, for example) decrease the signal-to-noise ratio of the image, which imposes an additional limit.

TIRF, SPP, etc. etc. are near-field methods, and the physics is different than far-field methods. There, the resolution is given by the size of the source (or aperture, or probe tip) used.

Again, to push optical microscopy into the future, much Physics will have to be learned and perfected. But to create a commercial product that outperforms the competition, many experiments must be conducted that cannot completely be described with Physical models. Either because we just don't know exactly the Physics involved or it is just computationally not feasible to quantitatively cover the qualitative output of a complex experiment.

Optics, and electromagnetism, are mature fields. The governing theory is complete. There is some wiggle room with respect to constitutive relationships- index of refraction, permittviity, suseptibility, etc. etc., and this is where useful engineering occurs (left-handed materials, for example). To be sure, as technology improves, we can develop new materials, or grind lenses to new shapes, etc. etc. but the theory is done. This is good, because be can understand the new things we produce.

Most likely, the chair you are sitting in cannot exactly be described with any Physical model and all the computational power available. But for your chair, what we can calculate might be good enough. Or not? - Wait until your chair breaks.

This is a straw argument- what do you mean "exactly"? Chairs can be constructed out of many materials, in many shapes, and can last for arbitrarily long periods of time. What is intersting about chairs is not if we understand how a chair "exactly works", but if we can make new kinds of chairs, out of new materials. If we can, we understand chairs.


If now the good old Köhler illumination has reached its limit regarding today's implementation is a question I cannot objectively answer. But it is always good to look at the details, although some of them may not fit into the equation. But again, since the perfect microscope does not exist (e.g. Apochromats are not corrected for all wavelengths but just for a few!), we may already screw up our implementation so much that "perfect Köhler" will not come to the rescue.

Koehler illumination itself is a limit! So is critical illumination. You are missing the point- it is not possible to correct for all wavelengths for any real material that we know of. Theory helps us deal with this deficiency. Critical illumination and Koehler illumination cannot strictly be acheived in practice. Theory guides how to get as close as possible while still having a system that can be manufactured cheaply.

You may be claiming "perfect imaging does not exist". This is true. Perfect imaging, which is the mapping of a point in object space to a point in image space, cwill never exist. The only way it can exist is to have an optical system that captures all of the light scattered by the object, and this is clearly impossible in practice.
--
Andrew Resnick, Ph.D.
Department of Physiology and Biophysics
Case Western Reserve University
.

Relevant Pages

  • Re: Nanotechnology applications
    ... Drastically improved video monitor resolution? ... Artifical brain cells?< ... > super-strong materials, ...
    (rec.arts.sf.science)
  • Re: 3D Printing - which process?
    ... The Eden is the best I've ever seen for resolution. ... materials is the big question. ... Rapid Prototyping of the model parts. ... I have had the best success with Objet prototypes. ...
    (comp.cad.solidworks)
  • Looking for clarity / accuracy not megapixel, any recommendations?
    ... will be used for taking pictures of textures and materials. ... MegaPixel / ... resolution isn't as important but I would say 3MP is a minimum. ...
    (rec.photo.digital)