Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal

Just a little more clarification:

The formula given by Jim is correct for x-y resolution (and assumes a
1.4 NA lens) but not for calculating the Nyquist sampling distance
that is appropriate along the z-axis. Actually, the latter is very
complicated when considering the thickness of the sample since
spherical aberration and RI mismatch can cause severe broadening of
the illumination intensity in the z-axis the deeper one looks into
the sample. This causes the z-axis resolution to decrease
significantly with depth, which also is inversely a function of the
square of the numerical aperture and not linear as given in Jim's
formula.

If you are using a biofilm-water/coverslip/ oil objective system, at
10 um from the coverslip the FWHM z-axis beam spread will give you a
number from about 0.4 um to 0.8 um using a 1.4 NA oil lens. ~0.4 um
would be the ideal theoretical number that may or not be achieved in
practice. Best to measure it with 100 nm fluorescent spheres and know
for sure.

However, past 10 um viewing depth, things start to get progressively
worse such that you end up rapidly approaching and exceeding 1 um
FWHM. The greater the RI mismatch and the worse the spherical
aberration, the greater the loss of resolution with depth. For this
reason water objectives often give the best performance possible
because the RI mismatch is considerably less.

Bottom line is that the z-axis delta need not usually be less than
0.2 um and when imaging deeper into the sample, the z- steps can be
considerably greater. Ideally, a water objective combined with
deconvolution applied to your stacks should give you the best quality
images. Deconvolution can correct for much of spread along the z-axis
and greatly improve resolution and signal to noise if calibrated and
used correctly with your confocal stacks.

Regards,

Mario


>Search the CONFOCAL archive at
>http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
>
>Yes, that will work.  I also have a chart I can e-mail (the listserv
>does not allow attachments) you can e-mail separately at [log in to unmask]
>
>-----Original Message-----
>From: Confocal Microscopy List [mailto:[log in to unmask]] On
>Behalf Of Andrew Phillips
>Sent: Wednesday, December 17, 2003 3:38 AM
>To: [log in to unmask]
>Subject: Z-Steps
>
>To clarify Nyquist, Z sampling should be done at half objective
>resolution. A very crude but near enough calculation for this is;
>
>Emissive Wavelength / 2 x NA of the objective.
>
>i.e. 530nM / (2 x 1.4) = 189nm
>
>Divide this by 2 = 94.5
>
>In truth, if you acquire at 0.1 micron per step (thisis often the finest
>step a Z motor will do) you will be acquiring according to Nyquist
>
>
>-----Original Message-----
>From: Confocal Microscopy List [mailto:[log in to unmask]] On
>Behalf Of Wijnholds, Anita
>Sent: Wednesday, December 17, 2003 2:47 AM
>To: [log in to unmask]
>Subject: resolution vs. analysis time
>
>Search the CONFOCAL archive at
>http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
>
>Dear all,
>
>We are measuring the development of biofilms. The z-step is calculated
>according to the resolution of the lens. In first instance, we get only
>a
>few slices as the film is not so thick. But the film grows to more than
>50
>micron and so we get a lot of slices with that resolution.The analysis
>of
>the stacks is going to take too much time.
>Can we reduce the amount of slices? What are criteria to do so?
>I hope you can give me some good thougts and ideas about this.
>
>Kind regards,
>
>Anita Wijnholds
>NIOO-CEME


--
_________________________________________________________
Mario M. Moronne, Ph.D.
NanoMed Technologies LLC
President and CTO
ph (510) 528-2400
FAX (510) 528-8076
1561 Posen Ave
Berkeley, CA
94706

[log in to unmask]
[log in to unmask]