reflected light profilometry trrough plastic - mystery bands

[Stanislav Vitha-2]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hallo,
I have a little mystery.
I am trying to measure depth of a microchannel (empty, air filled) using
confocal microscopy in reflected mode.
The microchannel is created by sandwiching a ~20 um PET (Mylar) spacer
between two ~170 um thick Mylar sheets. The spacer has a thin layer of
cyanoacrylate glue on both the top and bottom side.
I am performing line-Z scan, with 0.25 um z-step, using a 20x/0.75 dry
objective.
I expected to get two reflections, one from the top of the channel, one
from the bottom of the channel, just like it worked on microchannels made
out of glass that I imaged previously.
However, I am getting additional reflections and am not sure why and also
do not know which one I should take as representing the bottom of the
channel.
See the image here: http://i.imgur.com/Scj4i6y.jpg
Reractive index of mylar is about 1.5, but I do not think I need to correct
for z-axis compression since the channel is fille with air.
The reflections are at 20, 40, and 60 um from the first reflection (top of the
channel)
The lineZ-scan through the spacer shows two reflection peaks, nominally
~40um apart, after z-axis correction it is ~60 um.   Even if this is true and
the spacer is really 60 um thick, I am still puzzled by the extra reflections
in the air channel.
Any hints/input is appreciated.

With regards,

Stan Vitha
Microcopy and Imaging Cneter
Texas A&M University

[Craig Brideau]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Looks like a lot of scatter off something in the sample. When you scan and
look at the sample sitting on the stage do you see a lot of light off the
sample? I mean by eyeball, not with the microscope optics. It's possible
your glue is scattering and confusing your interfaces, also possible that
the mylar is very scattering. You can also have lots of multiple
reflections between surfaces, although the intensity of such 'ghost'
reflections tend to be much lower than the main reflection. Finally, what
size confocal pinhole are you using, and have you tried other wavelengths?

Craig Brideau

On Fri, Jan 29, 2016 at 11:55 AM, Stanislav Vitha <[hidden email]> wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> Post images on http://www.imgur.com and include the link in your posting.
> *****
>
> Hallo,
> I have a little mystery.
> I am trying to measure depth of a microchannel (empty, air filled) using
> confocal microscopy in reflected mode.
> The microchannel is created by sandwiching a ~20 um PET (Mylar) spacer
> between two ~170 um thick Mylar sheets. The spacer has a thin layer of
> cyanoacrylate glue on both the top and bottom side.
> I am performing line-Z scan, with 0.25 um z-step, using a 20x/0.75 dry
> objective.
> I expected to get two reflections, one from the top of the channel, one
> from the bottom of the channel, just like it worked on microchannels made
> out of glass that I imaged previously.
> However, I am getting additional reflections and am not sure why and also
> do not know which one I should take as representing the bottom of the
> channel.
> See the image here: http://i.imgur.com/Scj4i6y.jpg
> Reractive index of mylar is about 1.5, but I do not think I need to correct
> for z-axis compression since the channel is fille with air.
> The reflections are at 20, 40, and 60 um from the first reflection (top of
> the
> channel)
> The lineZ-scan through the spacer shows two reflection peaks, nominally
> ~40um apart, after z-axis correction it is ~60 um.   Even if this is true
> and
> the spacer is really 60 um thick, I am still puzzled by the extra
> reflections
> in the air channel.
> Any hints/input is appreciated.
>
> With regards,
>
> Stan Vitha
> Microcopy and Imaging Cneter
> Texas A&M University
>

[Stanislav Vitha-2]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Craig,
the scattering is not that much.
The glue is applied to the spacer only, so there should be no glue over the
clear channel.
The Mylar top and bottom sheets look optically clear.
The spacer is also mylar but is opaque/black.

Perhaps it has something to do with birefringence of the plastics, but I did
not quite think it through.

On Fri, 29 Jan 2016 14:02:31 -0700, Craig Brideau
<[hidden email]> wrote:

>Looks like a lot of scatter off something in the sample. When you scan and
>look at the sample sitting on the stage do you see a lot of light off the
>sample? I mean by eyeball, not with the microscope optics. It's possible
>your glue is scattering and confusing your interfaces, also possible that
>the mylar is very scattering. You can also have lots of multiple
>reflections between surfaces, although the intensity of such 'ghost'
>reflections tend to be much lower than the main reflection. Finally, what
>size confocal pinhole are you using, and have you tried other
wavelengths?
>
>Craig Brideau



 With regards,
>>
>> Stan Vitha
>> Microcopy and Imaging Cneter
>> Texas A&M University
>>

[Guy Cox-2]

This is truly an optically horrible setup.  I'm assuming that you are using a coverslip-corrected objective and therefore that your top 170µm of mylar won't be too far out.  But this is not designed for then imaging into an air cavity, so you will have major problems with spherical aberration.  To make matters worse, mylar is birefringent so your polarized beam of laser light will be split into two beams, each perceiving a different refractive index and therefore aberrated differently.  For an example of how SA can screw up depth measurements see G.C. Cox and C.J.R. Sheppard, 2001.  Measurement of thin coatings in the confocal microscope.  Micron 32, 701-705.  

There is another possible explanation, and that is that you are seeing multiple passes through the cavity.  So at 20µm you are seeing the true reflection, then 40µm down you are not seeing light from there at all, but light that has been reflected twice in the cavity, etc.  This would explain why all your peaks are multiples of the expected cavity depth.  

Paradoxically, I'd suggest using a worse optical system, such as a 10x NA 0.3 objective.  This will reduce the effects of SA.  Your reflection lines will now be much broader and your scans will look terrible BUT the true surface will still be at the brightest point (neglecting any residual SA).  So a line profile across your scan will show you the true depth.  

                                                Guy


Guy Cox, Honorary Associate Professor
School of Medical Sciences

Australian Centre for Microscopy and Microanalysis,
Madsen, F09, University of Sydney, NSW 2006

-----Original Message-----
From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of Craig Brideau
Sent: Saturday, 30 January 2016 8:03 AM
To: [hidden email]
Subject: Re: reflected light profilometry trrough plastic - mystery bands

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Looks like a lot of scatter off something in the sample. When you scan a nd look at the sample sitting on the stage do you see a lot of light off the sample? I mean by eyeball, not with the microscope optics. It's possible your glue is scattering and confusing your interfaces, also possible that the mylar is very scattering. You can also have lots of multiple reflections between surfaces, although the intensity of such 'ghost'
reflections tend to be much lower than the main reflection. Finally, what size confocal pinhole are you using, and have you tried other wavelengths?

Craig Brideau

On Fri, Jan 29, 2016 at 11:55 AM, Stanislav Vitha <[hidden email]> wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> Post images on http://www.imgur.com and include the link in your posting.
> *****
>
> Hallo,
> I have a little mystery.
> I am trying to measure depth of a microchannel (empty, air filled)
> using confocal microscopy in reflected mode.
> The microchannel is created by sandwiching a ~20 um PET (Mylar) spacer
> between two ~170 um thick Mylar sheets. The spacer has a thin layer of
> cyanoacrylate glue on both the top and bottom side.
> I am performing line-Z scan, with 0.25 um z-step, using a 20x/0.75 dry
> objective.
> I expected to get two reflections, one from the top of the channel,
> one from the bottom of the channel, just like it worked on
> microchannels made out of glass that I imaged previously.
> However, I am getting additional reflections and am not sure why and
> also do not know which one I should take as representing the bottom of
> the channel.
> See the image here: http://i.imgur.com/Scj4i6y.jpg Reractive index of
> mylar is about 1.5, but I do not think I need to correct for z-axis
> compression since the channel is fille with air.
> The reflections are at 20, 40, and 60 um from the first reflection
> (top of the
> channel)
> The lineZ-scan through the spacer shows two reflection peaks, nominally
> ~40um apart, after z-axis correction it is ~60 um.   Even if this is true
> and
> the spacer is really 60 um thick, I am still puzzled by the extra
> reflections in the air channel.
> Any hints/input is appreciated.
>
> With regards,
>
> Stan Vitha
> Microcopy and Imaging Cneter
> Texas A&M University
>

[Zdenek Svindrych-2]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi Stanislav,
I would expect two reflections outside the channel ([air - first mylar] and
[last mylar - air]) and four reflections across the channel (every air -
mylar inerface), unless your whole sandwich is glued to some glass support
or something...

This does not quite fit your measured profile, as the two reflection
(outside the channel) should be 360 um apart (without correction).
Furthermore, the [channel air - mylar] interface further from the objective
should not correspond (in depth) to any reflection outside the channel, the
air channel should appear as having different thickness than the mylar
spacer, due to different depth correction.

I can't see any of those in your images... Is your channel really air-
filled? Can you fill it with fluorescent solution and take a profile of the
fluorescence alongside with the reflection?
Best, zdenek
--
Zdenek Svindrych, Ph.D.
W.M. Keck Center for Cellular Imaging (PLSB 003)
University of Virginia, Charlottesville, VA
http://www.kcci.virginia.edu/
tel: [phone]


---------- Původní zpráva ----------
Od: Stanislav Vitha <[hidden email]>
Komu: [hidden email]
Datum: 29. 1. 2016 15:51:19
Předmět: reflected light profilometry trrough plastic - mystery bands

"*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hallo,
I have a little mystery.
I am trying to measure depth of a microchannel (empty, air filled) using
confocal microscopy in reflected mode.
The microchannel is created by sandwiching a ~20 um PET (Mylar) spacer
between two ~170 um thick Mylar sheets. The spacer has a thin layer of
cyanoacrylate glue on both the top and bottom side.
I am performing line-Z scan, with 0.25 um z-step, using a 20x/0.75 dry
objective.
I expected to get two reflections, one from the top of the channel, one
from the bottom of the channel, just like it worked on microchannels made
out of glass that I imaged previously.
However, I am getting additional reflections and am not sure why and also
do not know which one I should take as representing the bottom of the
channel.
See the image here: http://i.imgur.com/Scj4i6y.jpg
Reractive index of mylar is about 1.5, but I do not think I need to correct
for z-axis compression since the channel is fille with air.
The reflections are at 20, 40, and 60 um from the first reflection (top of
the
channel)
The lineZ-scan through the spacer shows two reflection peaks, nominally
~40um apart, after z-axis correction it is ~60 um. Even if this is true and
the spacer is really 60 um thick, I am still puzzled by the extra
reflections
in the air channel.
Any hints/input is appreciated.

With regards,

Stan Vitha
Microcopy and Imaging Cneter
Texas A&M University"

[Steffen Dietzel]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

 From your image it seems much easier to estimate the thickness of the
spacer rather than the one of the air filled chamber. Not surprising
considering the mess in optical properties you have in the air chamber
light path in combination with the high NA air objective (see Guy's
post). Is spacer thickness not good enough?

Also, if the low NA objective approach Guy suggested should not work
out, maybe you can align the mylar with the polarization of the laser
beam - what happens if you turn the sample?

Steffen

--
------------------------------------------------------------
Steffen Dietzel, PD Dr. rer. nat
Ludwig-Maximilians-Universität München
Biomedical Center (BMC)
Head of the Core Facility Bioimaging

Großhaderner Straße 9
D-82152 Planegg-Martinsried
Germany

http://www.bioimaging.bmc.med.uni-muenchen.de

[Stanislav Vitha-2]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Guy,
I agree the setup has problems. Yes, the objective is coverslip corrected.
Thanks for the reference to your Micron paper.

I think the multiple reflections are a good possibility, or the strange effects
of birefringence.
regarding the birefringence, are there any good tricks ti minimize the
problem in this setup? Like putting a 1/4 lambda plate somewhere in the
lightpath?  
 
I do not quite understand your comment regarding spherical aberration
when imaging with dry objective into an air cavity. I thought we have a
perferct RI match there (air objective, air in the specimen) so no SA would
be expected providing the coverglass thickness is correct.
I also tried an objective with coverglass adjustment (0 to 2 mm range) and
optimized the setting for maximum signal (= minimum SA) for the
reflection from the first surface of the channel. The results were similar.

I will test my 10x/0.4 objective and see what I get, but I will probably end
up filling the channel with fluorescencent due solution and measure it this
way. It will be interesting to compare the two methods.

Stan Vitha
Microscopy and Imaging Center
Texas A&M University

 

=====
On Sat, 30 Jan 2016 01:08:23 +0000, Guy Cox wrote:
This is truly an optically horrible setup.  I'm assuming that you are using a
coverslip-corrected objective and therefore that your top 170µm of mylar
won't be too far out.  But this is not designed for then imaging into an air
cavity, so you will have major problems with spherical aberration.  To make
matters worse, mylar is birefringent so your polarized beam of laser light
will be split into two beams, each perceiving a different refractive index and
therefore aberrated differently.  For an example of how SA can screw up
depth measurements see G.C. Cox and C.J.R. Sheppard, 2001.  
Measurement of thin coatings in the confocal microscope.  Micron 32, 701-
705.  
 
There is another possible explanation, and that is that you are seeing
multiple passes through the cavity.  So at 20µm you are seeing the true
reflection, then 40µm down you are not seeing light from there at all, but
light that has been reflected twice in the cavity, etc.  This would explain
why all your peaks are multiples of the expected cavity depth.  
 
Paradoxically, I'd suggest using a worse optical system, such as a 10x NA
0.3 objective.  This will reduce the effects of SA.  Your reflection lines will
now be much broader and your scans will look terrible BUT the true surface
will still be at the brightest point (neglecting any residual SA).  So a line
profile across your scan will show you the true depth.  
 
                                                Guy
===========

[Stanislav Vitha-2]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi Zdenek,
the start and end of the z-scan was within the top and biottom mylar
sheets, so you do not see the reflections from the outside surface of the
sandwich.
I was only interested in the channel itself, not the exact thickness if the
Mylar sheets - I checked and they are each about 170 um thick.

The reflections off the spacer show about 40 um apparent thickness, so
with correction for RI it is abour 60 um.
The last reflection off the  air channel (4th peak in the profile) corresponds
to 60 um depth from the first reflection, so that seems to make sense and
agree with the spacer thickness, but the two reflections in between are the
puzzle.  The image shows the third reflection in the air channel (at 40 um)
that happens to be ain the same position as the bottom reflection of  
spacer.  

I will try few more tricks, probably wwill end up filling the channel with
fluorescent dye solution. having a less confusing images for analysis will
be  a plus.

Stan
.

On Sun, 31 Jan 2016 04:35:26 +0100, Zdenek Svindrych  wrote:

>Hi Stanislav,
I would expect two reflections outside the channel ([air - first mylar] and
[last mylar - air]) and four reflections across the channel (every air -
mylar inerface), unless your whole sandwich is glued to some glass support
or something...

This does not quite fit your measured profile, as the two reflection
(outside the channel) should be 360 um apart (without correction).
Furthermore, the [channel air - mylar] interface further from the objective
should not correspond (in depth) to any reflection outside the channel, the
air channel should appear as having different thickness than the mylar
spacer, due to different depth correction.

I can't see any of those in your images... Is your channel really air-
filled? Can you fill it with fluorescent solution and take a profile of the
fluorescence alongside with the reflection?
Best, zdenek
--
Zdenek Svindrych, Ph.D.
W.M. Keck Center for Cellular Imaging (PLSB 003)
University of Virginia, Charlottesville, VA
http://www.kcci.virginia.edu/
tel: [phone]


---------- Původní zpráva ----------
Od: Stanislav Vitha <[hidden email]>
Komu: [hidden email]
Datum: 29. 1. 2016 15:51:19
Předmět: reflected light profilometry trrough plastic - mystery bands

"*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hallo,
I have a little mystery.
I am trying to measure depth of a microchannel (empty, air filled) using
confocal microscopy in reflected mode.
The microchannel is created by sandwiching a ~20 um PET (Mylar) spacer
between two ~170 um thick Mylar sheets. The spacer has a thin layer of
cyanoacrylate glue on both the top and bottom side.
I am performing line-Z scan, with 0.25 um z-step, using a 20x/0.75 dry
objective.
I expected to get two reflections, one from the top of the channel, one
from the bottom of the channel, just like it worked on microchannels made
out of glass that I imaged previously.
However, I am getting additional reflections and am not sure why and also
do not know which one I should take as representing the bottom of the
channel.
See the image here: http://i.imgur.com/Scj4i6y.jpg
Reractive index of mylar is about 1.5, but I do not think I need to correct
for z-axis compression since the channel is fille with air.
The reflections are at 20, 40, and 60 um from the first reflection (top of
the
channel)
The lineZ-scan through the spacer shows two reflection peaks, nominally
~40um apart, after z-axis correction it is ~60 um. Even if this is true and
the spacer is really 60 um thick, I am still puzzled by the extra
reflections
in the air channel.
Any hints/input is appreciated.

With regards,

Stan Vitha
Microcopy and Imaging Cneter
Texas A&M University"

[Guy Cox-2]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Stanislav,

          Coverslip correction refers to imaging a sample in the same refractive index as the coverslip.  That is what you are doing in the conventional biological imaging of a fixed specimen.  Going back into air effectively negates the correction, and you would be better off with a non-coverslip objective.  (But of course you'd have to apply a depth correction).  

  I cannot offer  a solution to imaging through mylar.  40+ years ago I was struggling with this problem and couldn't solve it.  (It would have made a really good paper if I could have!)

                                Guy




Guy Cox, Honorary Associate Professor
School of Medical Sciences

Australian Centre for Microscopy and Microanalysis,
Madsen, F09, University of Sydney, NSW 2006

-----Original Message-----
From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of Stanislav Vitha
Sent: Tuesday, 2 February 2016 3:08 AM
To: [hidden email]
Subject: Re: reflected light profilometry trrough plastic - mystery bands

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Guy,
I agree the setup has problems. Yes, the objective is coverslip corrected.
Thanks for the reference to your Micron paper.

I think the multiple reflections are a good possibility, or the strange effects of birefringence.
regarding the birefringence, are there any good tricks ti minimize the problem in this setup? Like putting a 1/4 lambda plate somewhere in the lightpath?  
 
I do not quite understand your comment regarding spherical aberration when imaging with dry objective into an air cavity. I thought we have a perferct RI match there (air objective, air in the specimen) so no SA would be expected providing the coverglass thickness is correct.
I also tried an objective with coverglass adjustment (0 to 2 mm range) and optimized the setting for maximum signal (= minimum SA) for the reflection from the first surface of the channel. The results were similar.

I will test my 10x/0.4 objective and see what I get, but I will probably end up filling the channel with fluorescencent due solution and measure it this way. It will be interesting to compare the two methods.

Stan Vitha
Microscopy and Imaging Center
Texas A&M University

 

=====
On Sat, 30 Jan 2016 01:08:23 +0000, Guy Cox wrote:
This is truly an optically horrible setup.  I'm assuming that you are using a coverslip-corrected objective and therefore that your top 170µm of mylar won't be too far out.  But this is not designed for then imaging into an air cavity, so you will have major problems with spherical aberration.  To make matters worse, mylar is birefringent so your polarized beam of laser light will be split into two beams, each perceiving a different refractive index and therefore aberrated differently.  For an example of how SA can screw up depth measurements see G.C. Cox and C.J.R. Sheppard, 2001.  
Measurement of thin coatings in the confocal microscope.  Micron 32, 701-
705.  
 
There is another possible explanation, and that is that you are seeing multiple passes through the cavity.  So at 20µm you are seeing the true reflection, then 40µm down you are not seeing light from there at all, but light that has been reflected twice in the cavity, etc.  This would explain
why all your peaks are multiples of the expected cavity depth.  
 
Paradoxically, I'd suggest using a worse optical system, such as a 10x NA
0.3 objective.  This will reduce the effects of SA.  Your reflection lines will now be much broader and your scans will look terrible BUT the true surface will still be at the brightest point (neglecting any residual SA).  So a line
profile across your scan will show you the true depth.  
 
                                                Guy
===========

[Stanislav Vitha-2]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Guy,
I must be a bit dense, but your explanation does not make sense for e.g.
water immersion objectives.
When coverslip correction on these lenses is applied, as long as the
refractive index of the sample (water) is exactly the same as the design
immersion medium (water), spherical aberration is corrected throughout
the imaging depth (whatever the working distance permits).
So essentially, a coverslip of proper thickness (corresponding to the
settings of the correction collar, or to the fixed design thickness of the
objective) can be positioned anywhere between the objective front element
and the specimen.

I believe the same concept applies to dry lenses.
I have seen a paper that I need to dig out, where this was explained (with
math) and experimentally shown for dry lenses.
So if you have a dry objective designed for 0.17mm coverslip and need to
look at the first surface of a naked specimen, you could attach the coverslip
to the front of the objective and achieve SA correction. I can confirm that
this works when I need to do reflected brightfield or reflected DIC of
surfaces with 20x/0.5 and 40x/0.75 lenses on my old Zeiss Axiophot.  A
careful smear of silicone grease on the metal surrounding the (concave)
front lens will hold the coverslip nicely.  

Bit I am getting a bit off topic.
I am about to fill the microchannel with rhodamine dye solution and
measure the depth by fluorescence.
I will report back how it went.

Stan Vitha  
 
On Tue, 2 Feb 2016 11:53:35 +0000, Guy Cox
<[hidden email]> wrote:

>Stanislav,
>
>           Coverslip correction refers to imaging a sample in the same
refractive index as the coverslip.  That is what you are doing in the
conventional biological imaging of a fixed specimen.  Going back into air
effectively negates the correction, and you would be better off with a non-
coverslip objective.  (But of course you'd have to apply a depth
correction).  
>
> I cannot offer  a solution to imaging through mylar.  40+ years ago I
was struggling with this problem and couldn't solve it.  (It would have
made a really good paper if I could have!)

>
> Guy
>
>
>
>
>Guy Cox, Honorary Associate Professor
>School of Medical Sciences
>
>Australian Centre for Microscopy and Microanalysis,
>Madsen, F09, University of Sydney, NSW 2006
>
>-----Original Message-----
>From: Confocal Microscopy List

[mailto:[hidden email]] On Behalf Of Stanislav
Vitha
>Sent: Tuesday, 2 February 2016 3:08 AM
>To: [hidden email]
>Subject: Re: reflected light profilometry trrough plastic - mystery bands
>
>*****
>To join, leave or search the confocal microscopy listserv, go to:
>http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>Post images on http://www.imgur.com and include the link in your
posting.
>*****
>
>Guy,
>I agree the setup has problems. Yes, the objective is coverslip corrected.
>Thanks for the reference to your Micron paper.
>
>I think the multiple reflections are a good possibility, or the strange
effects of birefringence.
>regarding the birefringence, are there any good tricks ti minimize the
problem in this setup? Like putting a 1/4 lambda plate somewhere in the
lightpath?  
>
>I do not quite understand your comment regarding spherical aberration
when imaging with dry objective into an air cavity. I thought we have a
perferct RI match there (air objective, air in the specimen) so no SA would
be expected providing the coverglass thickness is correct.
>I also tried an objective with coverglass adjustment (0 to 2 mm range)
and optimized the setting for maximum signal (= minimum SA) for the
reflection from the first surface of the channel. The results were similar.
>
>I will test my 10x/0.4 objective and see what I get, but I will probably
end up filling the channel with fluorescencent due solution and measure it
this way. It will be interesting to compare the two methods.

>
>Stan Vitha
>Microscopy and Imaging Center
>Texas A&M University
>
>
>
>=====
>On Sat, 30 Jan 2016 01:08:23 +0000, Guy Cox wrote:
>This is truly an optically horrible setup.  I'm assuming that you are using

a coverslip-corrected objective and therefore that your top 170µm of mylar
won't be too far out.  But this is not designed for then imaging into an air
cavity, so you will have major problems with spherical aberration.  To make
matters worse, mylar is birefringent so your polarized beam of laser light
will be split into two beams, each perceiving a different refractive index and
therefore aberrated differently.  For an example of how SA can screw up
depth measurements see G.C. Cox and C.J.R. Sheppard, 2001.  
>Measurement of thin coatings in the confocal microscope.  Micron 32, 701-
>705.  
>
>There is another possible explanation, and that is that you are seeing
multiple passes through the cavity.  So at 20µm you are seeing the true
reflection, then 40µm down you are not seeing light from there at all, but
light that has been reflected twice in the cavity, etc.  This would explain
>why all your peaks are multiples of the expected cavity depth.  
>
>Paradoxically, I'd suggest using a worse optical system, such as a 10x NA
>0.3 objective.  This will reduce the effects of SA.  Your reflection lines will
now be much broader and your scans will look terrible BUT the true surface
will still be at the brightest point (neglecting any residual SA).  So a line
>profile across your scan will show you the true depth.  
>
> Guy
>===========

[Stanislav Vitha-2]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi Steffen,
the spacer thickness would probably be good enough but the spacer is very
opaque (black) and the reflections not very strong, given that it is the
same plastic  with just a thin layer of glue holding it together. That is why I
went initially for the brighter signal in the air channel.

I will try the 10x/0.4 objective and also try rotating the sample.

Stan Vitha
Microscopy and Imaging Center
Texas A&M University

 so the signal was
On Mon, 1 Feb 2016 14:49:38 +0100, Steffen Dietzel
<[hidden email]> wrote:

> From your image it seems much easier to estimate the thickness of the
>spacer rather than the one of the air filled chamber. Not surprising
>considering the mess in optical properties you have in the air chamber
>light path in combination with the high NA air objective (see Guy's
>post). Is spacer thickness not good enough?
>
>Also, if the low NA objective approach Guy suggested should not work
>out, maybe you can align the mylar with the polarization of the laser
>beam - what happens if you turn the sample?
>
>Steffen
>
>--
>------------------------------------------------------------
>Steffen Dietzel, PD Dr. rer. nat
>Ludwig-Maximilians-Universität München
>Biomedical Center (BMC)
>Head of the Core Facility Bioimaging
>
>Großhaderner Straße 9
>D-82152 Planegg-Martinsried
>Germany
>
>http://www.bioimaging.bmc.med.uni-muenchen.de