Re: Coverslips

Posted by James Pawley on
URL: http://confocal-microscopy-list.275.s1.nabble.com/Coverslips-tp6865241p6870787.html

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

>*****
>To join, leave or search the confocal microscopy listserv, go to:
>http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>*****
>
>A couple of things I have noticed about
>coverslips.  Firstly, if you ever do measure
>commercial coverslips, you will notice that
>within a batch, most are within a 5 um tolerance
>with a few outliers.  From batch to batch you
>will notice far more differences (I had one
>batch of #1.5's that was all almost 190 um).
>Secondly, the comment about oil vs. water
>objectives is a very important one.  Oil
>objectives are designed to image just above the
>coverslip.  Since the oil, coverslip, and front
>lens are all of the same refractive index, it
>makes no difference how thick the coverslip
>is--that ends up being compensated by the
>increased oil thickness necessary to achieve the
>same focal depth into the sample.  For water, I
>have never noticed a 2 um difference by eye, but
>fluorescence correlation spectroscopy is
>uniquely sensitive to these things and I don't
>see a significant difference as long as I stay
>within a 5 um tolerance.  I used to think that
>the solution was to carefully adjust the
>correction collar on water objectives and soon
>discovered that the lag in the correction collar
>is as much as 100 um!  As a result, the numbers
>on the correction collar are essentially
>meaningless unless you always adjust from the
>same direction (note that I have only tested
>this for Zeiss objectives).  The best practice
>is to adjust the collar with a sensitive sample
>and then leave it and use measured or high
>precision coverslips for the remainder of the
>experiment.
>
>Jay

To see it by eye, you need only to have a bright,
small object, a sensitive detector and lots of
magnification. Adjust the collar, focus up and
down. Look for symmetry (same "blobby-ring" image
1 µm above best focus as 1µm below). If it isn't
z-symmetrical, adjust the collar and try again
until it is z-symmetrical (Forget trying to guess
if the spot is smaller. You can't, and z-symmetry
is what you want.). After having done this a
number of times, you will find you have adjusted
the collar to the exactly same setting +- 2 µm
(the small spacings on the collar of the Zeiss
1.2 cAPO).

Such images are seen in FIg. 20.3 of the
Handbook. The collar setting for series D was 2µm
of glass-replaced-by-water different from that
used in series A. The image is of 3 small holes
in a mirror film.

Cheers,

JP

>-----Original Message-----
>From: Confocal Microscopy List
>[mailto:[hidden email]] On
>Behalf Of James Pawley
>Sent: Friday, October 07, 2011 10:32 AM
>To: [hidden email]
>Subject: Re: Coverslips
>
>*****
>To join, leave or search the confocal microscopy listserv, go to:
>http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>*****
>
>>*****
>>To join, leave or search the confocal microscopy listserv, go to:
>>http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>>*****
>>
>>Thanks for everyone's input on this, it's really appreciated.  Julien -
>>although it doesn't mention it specifically I think the Zeiss link
>>makes the point about the requirement for precise cover glass thickness
>>with SIM.
>>If you're to achieve the best resolution possible you have to use glass
>>of a thickness that most closely matches that for which the objective
>>is designed (usually 170 um).  Any deviation from this will introduce
>>sperical aberration.  My guess is that because the resolution of a
>  >confocal system is lower, you won't notice much of a difference between
>>an image acquired using 170 and a 180 um glass, whereas on a SIM system
>>you will.
>>Simon
>>
>
>
>Just one more point: It makes a lot of
>difference whether we are talking about oil or
>water lenses.
>
>As immersion oil is almost the same RI as the
>BK-7 glass that I think is used for coverslips,
>changes in thickness from the optimum are much
>less important than if you are using a water
>lens. In the latter case, (RI = 1.515 vs Ri
>1.334?) even a 2µm error in thickness is easily
>noticeable by eye with an NA 1.2 objective.
>Other factors that affect the PSF (such as the
>temperature of the immersion oil!) are well
>covered in Chapter 11 of the Handbook.
>
>SIM is indeed curcially dependent of having a
>known and unaberrated PSF and this is only
>possible if you re free from SA. To check this,
>always have some sub-resolution beads in your
>preparation and ensure that they "go out of
>focus" in a manner that looks the same whether
>you focus up or focus down.
>
>JP
>
>
>>
>>-----Original Message-----
>>From: Confocal Microscopy List
>>[mailto:[hidden email]] On Behalf Of Julien Cau
>>Sent: 07 October 2011 08:26
>>To: [hidden email]
>>Subject: Re: Coverslips
>>
>>*****
>>To join, leave or search the confocal microscopy listserv, go to:
>>http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>>*****
>>
>>Hi Martin,
>>
>>You are absolutely right and SIM does not need better coverslips than
>>any other 0.17 objective-based microscopy technique.
>>Our users use these coverslips for any experiment (widefield, confocal,
>>SIM) and they appreciated the difference.
>>The SIM point is more "don't use a fancy device that promiss you
>>superesolution if you waste this potential with crap coverslips". If
>>you buy a wonderfull coffee machine, will you put in it moldy coffee beans?
>>Wrong thickness coverslips induce decreased light collection, spherical
>>aberrations and then reduced resolution.
>>
>>It is worth using them for any type of experiment using a 0.17 lens.
>>Alternatively, you can buy an expensive lens with a correction collar
>>that can counteract the effects of the coverslip thickness error.
>>
>>Best regards
>>
>>PS : see for instance
>>http://www.meditec.zeiss.com/4125681F004CA025/Contents-Frame/C66A2521E5
>>24C891852575A200721A7C for a direct comparison of imaging with #1.5 vs
>>#2 coverglass.
>>
>>Le 07/10/2011 03:26, Martin Wessendorf a écrit :
>>>   *****
>>>   To join, leave or search the confocal microscopy listserv, go to:
>>>   http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>>>   *****
>>>
>>>   On 10/6/2011 3:43 AM, Simon Walker wrote:
>>>
>>>>   Not strictly a confocal question, but I'm sure someone out there
>>>>  will  be able to  help.  We are about to start using a SIM super res
>>>>  system and have been  advised that one of the critical factors in
>>>>  acquiring optimal images  is the cover  glass.  Specifcally, the
>>>>  thickness of the glass needs to be  consistant across  the whole
>>>>  coverslip (e.g. 170 um +/- 2 um), and reproducible between
>>>>  coverslips.  One option is to measure each individual coverslip
>>>>  before use, but  this seems rather impractical to me.  Has anyone
>>>>  looked into this,  and if so,  are there any manufacturers out there
>>>>  who can provide cover glass  with this  high specification?
>>>
>>>   My ignorance is showing here.  Why are higher-quality coverslips
>>   > needed for SIM?  Or is this for live-cell imaging?
>>>
>>>   Best wishes--
>>>
>>>   Martin Wessendorf
>>
>>--
>>
>>____________________________________________
>>
>>*/Julien Cau, PhD./*
>>
>>/Montpellier RIO Imaging Facility manager/Responsable technique MRI/
>>
>>Montpellier RIO Imaging
>>
>>Montpellier BIOCAMPUS, UMS3426
>>
>>Arnaud de Villeneuve Campus Imaging Facility
>>
>>Institut de Génétique Humaine-CNRS
>>
>>141, rue de la Cardonille
>>
>>F-34396 Montpellier(France)
>>
>>e-mail: [hidden email].fr_ <mailto:[hidden email]>
>>
>>phone: [phone]
>>
>>mobile: [phone]
>>
>>fax: [phone]
>>
>>URL: _http://www.mri.cnrs.fr/_
>>
>>____________________________________________
>
>
>--
>James and Christine Pawley, 21 N. Prospect Ave.
>Madison, WI, 53726   Phone: [phone]

--
James and Christine Pawley, 21 N. Prospect Ave.
Madison, WI, 53726   Phone: [phone]