Re: PSF measurement using Au beads

Posted by Kyle Michael Douglass on
URL: http://confocal-microscopy-list.275.s1.nabble.com/PSF-measurement-using-Au-beads-tp7581962p7581966.html

        From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of Yan, Lu
        Sent: vendredi 2 mai 2014 03:24
        To: [hidden email]
        Subject: Re: PSF measurement using Au beads

                On May 1, 2014 6:35 PM, "MODEL, MICHAEL" <[hidden email]> wrote:

                        From: Confocal Microscopy List <[hidden email]> on
                         behalf of Lu <[hidden email]>
  Sent: Thursday, May 01, 2014 5:49 PM
  To: [hidden email]
                  Subject: PSF measurement using Au beads

                        Problems:
                  1. Two reflective layers showed up as we do axially scanning,
  separated by about 8 um, and the bead turned out to be attached to one
                         of them. The axial PSF looks terribly distorted. It is very much like
                         a four lobes pattern, i.e. an intensity null surrounded by 4 lobes on
  top/bottom and right/left.
                         You could imagine that at some z positions, the lateral intensity
                         pattern has a donut-shape. We do have a good explanation why this
  happened. Does any one ever have similar problem? Is my sample
                         preparation wrong?

  2. If I put a iris before the back aperture of the objective, and
                  closed it a little bit to truncated my collimated beam to half of its
                         original size, then the axial PSF suddenly got cleaned up, i.e. a
                         single nice vertical lobe appeared. But 2 reflective layers were still
                         there observable. Any idea why?
                         We thought the achromatic double for collimation might induce some
                         higher order free space mode other than pure Gaussian mode, such that
                         when we close the iris we effectively cut off some high k vectors of
  those 'other modes', leaving nicer Gaussian going into the objective
                         to produce nicer axial PSF.
                         Does this make sense to you guys?

                It seems to me that reflecting beads would be a tricky object to get
  PSF from because you have to deal with the angular dependence of
  scattering and reflection. 8 um might be the distance between a slide
  and a coverslip, both surfaces should reflect.

        Thanks for your reply. Yes you are right those two layers are cover glass and glass slide. I have seen in many papers people using gold beads to probe the focal intensity distribution which somewhat related to the PSF of the system, so I just figured it might be easier to measure in this way, and i wanted to know if they had similar problems. But this distorted PSF seems to be related to the fact that the incident beam has non perpect Gaussian profile, which confuses me most.

Hi Lu,

I think that Mike is right about having to consider the angular dependence of scattering from the beads. (I'm assuming that your beam is collimated at the back focal plane). In confocal setups you illuminate your bead with a number of plane waves (i.e. k-vectors) that span a solid cone.  Each plane wave within this cone is going to independently scatter light into a pattern that you can determine using Mie theory. Because the scattering is coherent, the total scattered field is just the sum of the different field scattering profiles from all the plane waves traveling in different directions within the cone, and the intensity that you measure is a projection of the squared-field distribution that fits within your system's numerical aperture onto a plane.

All that being said, if you reduce the size of your beam in the back focal plane of the objective, then this is equivalent to illuminating your beads with a cone of light with a smaller apex angle. Equivalently, you have fewer plane waves that scatter and the total scattered field is summed over a smaller number of Mie field profiles. I suspect that this reduces any interference effects in the total scattered field and is what eliminates the lobes you're observing.

To summarize, the messy axial profile might not originate from a dirty excitation beam, but simply because you're exciting the sphere with a number of plane waves traveling in different directions and the scattered fields from each of these plane waves is are interfering. If the spheres are touching the glass, this could also introduce asymmetries in the profile but I doubt it since you mentioned the spheres are mounted in glycerol, which has a refractive index close to glass.

You could check the Mie profiles with this handy webapp: http://omlc.ogi.edu/cgi-bin/mie_angles.cgi?diameter=0.15&lambda_vac=0.650&n_medium=1.47&nr_sphere=0.18&ni_sphere=-3.42&n_angles=100&density=0.1

I already entered the material parameters based on what you mentioned, but you should double check them anyway.

Good luck!

Kyle

Dr. Kyle M. Douglass
Postdoctoral Fellow
The Laboratory of Experimental Biophysics
EPFL, Lausanne, Switzerland
+41 21 69 30556 (Office)