Diffraction
Abbe imaging and spatial filtering
Ernst Abbe explained in 1873 how a microscope forms an image. The object diffracts the light into orders, beams that leave at set angles; the objective lens collects them, and the image is their interference. Detail whose orders miss the lens vanishes from the image. A 4f system, two lenses with each part one focal length from the next, shows this on the bench, with an iris where the orders come to a focus.
What you are looking at
A laser lights a Ronchi ruling with bars 0.1 mm apart. The first lens, 200 mm on, focuses each diffraction order to its own point in its focal plane, where an iris stands. The second lens turns the orders back into an image of the ruling on the screen, upside down and the same size.
The iris is open to 2 mm: it passes the straight-through light and the two first orders, 1.27 mm either side, and stops the rest. The screen shows the bars softened into stripes.
How the bench is set up
HeNe laser with a 1 mm waist, a Ronchi ruling of period 0.1 mm, two 200 mm lenses 400 mm apart with an iris midway between them, and a screen 200 mm beyond the second lens: each part one focal length from the next.
The theory
A lens of focal length makes the Fourier transform of the light in its front focal plane in its back focal plane. A ruling of period becomes a row of spots, one for each order, at
1.27 mm apart here, and the second lens transforms them back. A 50% ruling transmits , with and no even orders. With only orders 0 and ±1 through the iris, the image is
stripes at the ruling’s period. Close the iris inside the first orders and only the uniform zero order is left: the bars are gone however good the lenses are. That is Abbe’s limit: a microscope resolves a period only if its objective catches the first order, , where NA, the numerical aperture, is the sine of the widest angle the objective collects in air.
What to measure
Orders λf/d apart; no first orders, no bars
- Take out the iris and the second lens and move the screen to x = 600 mm, where the iris was: measure the spacing of the spots, = 1.27 mm.
- Put them back and close the iris to 1 mm: the bars vanish from the image.
What the model leaves out
- The lenses are ideal thin lenses, free of aberrations and wide enough for every order.
- The iris stands exactly in the focal plane, where each order is a point.
Things to try
- Open the iris to 5 mm: the third orders, 3.8 mm out, pass too and the stripes sharpen towards bars.
- Swap the iris for an opaque disc of 0.3 mm radius, which blocks the zero order alone (dark-field): the image goes bright everywhere except the edges of the bars, which show as dark lines, two to a period.
- Halve the ruling’s period: the first orders move out to 2.5 mm, beyond the iris, and the image is blank.
References
- E. Abbe (1873), Beiträge zur Theorie des Mikroskops und der mikroskopischen Wahrnehmung, Archiv für mikroskopische Anatomie 9, 413.
- J. W. Goodman (2017), Introduction to Fourier Optics, 4th edition, W. H. Freeman.
More experiments, and how the lab works, are in the laser lab’s overview.
