How to measure objective transmission curves?
Locked
12
12
 
|
Shalin Mehta |
Re: How to measure the actual numerical aperture of a microscope objective?
|
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
Hello Guy,
Interesting! Some clarification though... With equispaced lines on the board and a string, we are essentially measuring (working distance)*tan(aperture) on the board, right? Cheers Shalin On 9/11/07, Guy Cox <[hidden email]> wrote: Search the CONFOCAL archive at -- My co-ordinates: Shalin Mehta, Graduate student Graduate Programme in Bioengineering, NUS, Singapore Email: shalin {dot} mehta {at} gmail {dot} com Blog: electricsbm.blogspot.com Mobile: +65 90694182 |
|
|
John Oreopoulos |
Re: How to measure the actual numerical aperture of a microscope objective?
|
|
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
Actually, I think what he's saying is that you can measure the half angle of the cone by holding the board with distance markings a known distance away from the objective and counting the number of divisions visible on the board. Then, the half angle theta is given by:
tan(theta) = (horizontal distance visible on the board/2) / distance board is held away from the from the objective you take the inverse tangent to get the value of theta, and then: NA = n sin (theta) But one problem (maybe). Guy, do you not have to take into account that this measurement is done in air but the objective is meant to operate with oil? I certainly can't fill my room with oil... But I suppose I could rig a glass slide with divisions and hold it a much closer distance to the objective and put a small glass container of oil in between them. Is this what you meant, Guy? John On 10-Sep-07, at 9:35 PM, Shalin Mehta wrote: Search the CONFOCAL archive at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal Hello Guy, |
|
|
In reply to this post by John Oreopoulos
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
The answers to this, so
far, have not been quite accurate.
n should be the refractive
index of both the oil and the
mounting medium, and if this is not so
aberrations will be
bad and resolution reduced.
The reason n is in the formula at all is
that the wavelength
of the light depends on the refractive
index of the medium
through which it is passing. It is
the shorter wavelength
at higher refractive index that gives us
the improved
resolution, so the resolution will be
determined by the
RI of the medium in which the sample is
located.
Why then does the RI of the oil
matter? Because if it
doesn't match the coverslip and the sample
then light will
be refracted away from the optic axis, thus
reducing the
effective NA. So BOTH have to match
what the lens
is designed for.
This is all explained (with diagrams, which
may make it
easier) on pp15-17 of my book - sadly
these pages aren't in
the online sample pages, so you'll have to
buy or borrow a
copy!
Guy
Optical Imaging Techniques in Cell
Biology
by Guy Cox CRC Press / Taylor & Francis http://www.guycox.com/optical.htm ______________________________________________ Associate Professor Guy Cox, MA, DPhil(Oxon) Electron Microscope Unit, Madsen Building F09, University of Sydney, NSW 2006 ______________________________________________ Phone +61 2 9351 3176 Fax +61 2 9351 7682 Mobile 0413 281 861 ______________________________________________ http://www.guycox.net From: Confocal Microscopy List on behalf of John Oreopoulos Sent: Tue 11/09/2007 3:33 AM To: [hidden email] Subject: Confusion about the equation for numerical perture Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal Hi, I'm a little
confused about the traditional equation for a microscope's numerical
aperture:
NA = n x sin(theta)
where theta is the maximum half angle subtended by the lens's light
collection cone. It's the "n", the refractive index I'm confused about. I've now
read several sources and books that say n is the refractive index of the medium
that the sample is embedded in (above the coverslip), and in other places, I
read that n is the refractive index of the immersion fluid, say oil (below the
coverslip). I understand that in the ideal cases, both of these refractive
indexes should be matched for the best possible imaging, but in reality they
never really are, right? So which medium does the "n" refer too? And what about
in the case for TIRF microscopy, where it is required that you have a oil
immersion objective to image a sample in water for example? Which n should I use
to calculate my objective NA? John Oreopoulos, BSc, PhD Candidate University of Toronto Institute For Biomaterials and Biomedical Engineering Centre For Studies in Molecular Imaging Tel: W:416-946-5022 |
|
|
John Oreopoulos |
Re: Confusion about the equation for numerical perture
|
|
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
So then if I am to understand you correctly, in the non-ideal case, if the refractive index above and below the coverslip are not the same, then the effective NA of the objective is less than what is listed on the barrel of the objective, and so my resolution must therefore be decreased?
And again, what about the case of TIRF microscopy then? This imaging mode depends on there being a refractive index mismatch in order to achieve low penetration depths. Did the designers of this type of objective account for this when they calculated the NA of the objective before it was built? John On 10-Sep-07, at 11:29 PM, Guy Cox wrote: Search the CONFOCAL archive at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal |
|
|
In reply to this post by Shalin Mehta
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
So you're saying you can't
convert from tan to sin? :)
Actually our class setup has the scope
horizontal on
a board with radiating lines at the actual
angles and they
use a vertical rod to find the
'disappearing point'.
But it's intended as a demo not a
scientific instrument!
The actual implementation (and whether to
have the
lines equi-spaced or spaced to give NA
directly is
really up to what bits and pieces you have
at your disposal
and whether you're going to use it once or
often.
Guy
Optical Imaging Techniques in Cell
Biology
by Guy Cox CRC Press / Taylor & Francis http://www.guycox.com/optical.htm ______________________________________________ Associate Professor Guy Cox, MA, DPhil(Oxon) Electron Microscope Unit, Madsen Building F09, University of Sydney, NSW 2006 ______________________________________________ Phone +61 2 9351 3176 Fax +61 2 9351 7682 Mobile 0413 281 861 ______________________________________________ http://www.guycox.net From: Confocal Microscopy List on behalf of Shalin Mehta Sent: Tue 11/09/2007 11:35 AM To: [hidden email] Subject: Re: How to measure the actual numerical aperture of a microscope objective? Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal Hello
Guy,
Interesting! Some clarification though... With equispaced lines on the board and a string, we are essentially measuring (working distance)*tan(aperture) on the board, right? Cheers Shalin On 9/11/07, Guy Cox
<[hidden email]>
wrote:
Search the CONFOCAL archive at -- My co-ordinates: Shalin Mehta, Graduate student Graduate Programme in Bioengineering, NUS, Singapore Email: shalin {dot} mehta {at} gmail {dot} com Blog: electricsbm.blogspot.com Mobile: +65 90694182 |
|
|
In reply to this post by John Oreopoulos
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
So then if I am to understand you correctly, in the non-ideal case,
if the refractive index above and below the coverslip are not the same, then the
effective NA of the objective is less than what is listed on the barrel of the
objective, and so my resolution must therefore be decreased?
That is absolutely correct. Furthermore, unless you are
imaging very close to
the coverslip, spherical aberration will degrade your resolution
far more than the
lower RI will. If you want to look at samples in water
(except in TIRF) then
a water-immersion coverslip-corrected lens will ALWAYS give you the
best resolution.
And again, what about the case of TIRF microscopy then? This imaging mode
depends on there being a refractive index mismatch in order to achieve low
penetration depths. Did the designers of this type of objective account for this
when they calculated the NA of the objective before it was built?
TIRF is a special case, and the NA is calculated so that total internal
reflection can be achieved at a glass-water interface. So it is the
refractive
index of the oil that counts. And you can use a TIRF lens as a
conventional
oil lens with a sample in high RI medium. (In fact it's really good
as such since
it usually has a correction collar so you can tune out SA caused by
tiny
index mismatches - your average 1.4 planApo can't do that).
But when it's used in TIRF we are imaging in the evanescent wave (as in
NSOM)
so that diffraction limitation doesn't apply. And the distance is so
short that
SA won't cause problems (you have a collar, anyway, but it's more to adjust
for oil temperature).
Guy
On 10-Sep-07, at 11:29 PM, Guy Cox wrote: Search the CONFOCAL archive at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal |
|
|
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal Apologies for dual posting - email meltdown lost messages over last few days... To the person interested in analysing starch structure by confocal, check out Blennow et al. (2003) J Struct Biol 143: 229-241, and more recently Chanzy et al. (2006) J Struct Biol 154: 100-110 for a nice method to do this. cheers, Rosemary Rosemary White CSIRO Plant Industry GPO Box 1600 Canberra, ACT 2601 Australia 61 2 6246 5475 |
|
|
Steffen Dietzel |
Re: Confusion about the equation for numerical perture
|
|
In reply to this post by Guy Cox
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal At 05:29 11.09.2007, you wrote: >Why then does the RI of the oil matter? Because if it >doesn't match the coverslip and the sample then light will >be refracted away from the optic axis, thus reducing the >effective NA. So BOTH have to match what the lens >is designed for. However, unfortunately, RI of the oil and glass usually don't match! Typical RI for oil is n at 546.1 nm = n(e) = 1.518. Most coverslip producers don't tell you the RI of their glass. The one that does that I found gives n(e) = 1.5255 +/- 0.0015 source: http://www.hecht-assistent.de/download/Englisch/Microscopic.pdf So the coverslip has a higher RI than the oil. Lens designers know that of course and I am under the impression that they correct for this mismatch. To do that, they have to assume a specific thickness of the coverslip: 170 um. Which is why for high resolution microscopy it is mandatory to use such coverslips. One would think that coverslip producers have matching thickness as their standard. However this is far from true. What you usually get is "Thickness 1" which is defined as 130 - 160 nm. I never found out why they to this to the microscopic community. Maybe somebody can enlighten me. In any case, it should be worth to spend some time to find a source for coverslips of the right thickness in your neighborhood - err your continent. Steffen -- --------------------------------------------------------------------------------------------------- Steffen Dietzel, PD Dr. rer. nat Ludwig-Maximilians-Universität München Walter-Brendel-Zentrum (WBZ) Head of light microscopy Building location and address for courier, parcel services etc: Marchioninistr. 27, D-81366 München (Großhadern) Mail room (for letters etc.): Marchioninistr. 15, D-81366 München Phone: +49/89/2180-76509 Fax: +49/89/2180-76503 (please anounce incoming fax by e-mail) skype: steffendietzel e-mail: [hidden email] (for everything university related) or [hidden email] |
|
|
In reply to this post by John Oreopoulos
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal And apart from all this the RI of the oil is only specified for one temperature and if your room temperature is even a few degrees different it will be wrong .... All of which is precisely why for serious oil immersion microscopy a TIRF lens (which has a correction collar) is a very good lens to use even for non-TIRF applications. Guy Optical Imaging Techniques in Cell Biology by Guy Cox CRC Press / Taylor & Francis http://www.guycox.com/optical.htm ______________________________________________ Associate Professor Guy Cox, MA, DPhil(Oxon) Electron Microscope Unit, Madsen Building F09, University of Sydney, NSW 2006 ______________________________________________ Phone +61 2 9351 3176 Fax +61 2 9351 7682 Mobile 0413 281 861 ______________________________________________ http://www.guycox.net -----Original Message----- From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of Steffen Dietzel Sent: Thursday, 13 September 2007 7:17 PM To: [hidden email] Subject: Re: Confusion about the equation for numerical perture Search the CONFOCAL archive at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal At 05:29 11.09.2007, you wrote: >Why then does the RI of the oil matter? Because if it doesn't match >the coverslip and the sample then light will be refracted away from the >optic axis, thus reducing the effective NA. So BOTH have to match what >the lens is designed for. However, unfortunately, RI of the oil and glass usually don't match! Typical RI for oil is n at 546.1 nm = n(e) = 1.518. Most coverslip producers don't tell you the RI of their glass. The one that does that I found gives n(e) = 1.5255 +/- 0.0015 source: http://www.hecht-assistent.de/download/Englisch/Microscopic.pdf So the coverslip has a higher RI than the oil. Lens designers know that of course and I am under the impression that they correct for this mismatch. To do that, they have to assume a specific thickness of the coverslip: 170 um. Which is why for high resolution microscopy it is mandatory to use such coverslips. One would think that coverslip producers have matching thickness as their standard. However this is far from true. What you usually get is "Thickness 1" which is defined as 130 - 160 nm. I never found out why they to this to the microscopic community. Maybe somebody can enlighten me. In any case, it should be worth to spend some time to find a source for coverslips of the right thickness in your neighborhood - err your continent. Steffen -- --------------------------------------------------------------------------------------------------- Steffen Dietzel, PD Dr. rer. nat Ludwig-Maximilians-Universität München Walter-Brendel-Zentrum (WBZ) Head of light microscopy Building location and address for courier, parcel services etc: Marchioninistr. 27, D-81366 München (Großhadern) Mail room (for letters etc.): Marchioninistr. 15, D-81366 München Phone: +49/89/2180-76509 Fax: +49/89/2180-76503 (please anounce incoming fax by e-mail) skype: steffendietzel e-mail: [hidden email] (for everything university related) or [hidden email] No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.5.485 / Virus Database: 269.13.16/1004 - Release Date: 12/09/2007 5:22 PM No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.5.485 / Virus Database: 269.13.16/1004 - Release Date: 12/09/2007 5:22 PM |
|
|
Craig Brideau |
Re: Confusion about the equation for numerical perture
|
|
In reply to this post by Steffen Dietzel
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal It's hard to mass-produce 'flat' glass of consistant thickness from rolled glass. Anything precise has to be custom ground with interferometric measurement as a check during the process! Craig On 9/13/07, Steffen Dietzel <[hidden email]> wrote: > Search the CONFOCAL archive at > http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal > > At 05:29 11.09.2007, you wrote: > >Why then does the RI of the oil matter? Because if it > >doesn't match the coverslip and the sample then light will > >be refracted away from the optic axis, thus reducing the > >effective NA. So BOTH have to match what the lens > >is designed for. > > However, unfortunately, RI of the oil and glass usually don't match! > > Typical RI for oil is n at 546.1 nm = n(e) = 1.518. > > Most coverslip producers don't tell you the RI of their glass. > The one that does that I found gives n(e) = 1.5255 +/- 0.0015 > source: http://www.hecht-assistent.de/download/Englisch/Microscopic.pdf > > So the coverslip has a higher RI than the oil. > Lens designers know that of course and I am under > the impression that they correct for this > mismatch. To do that, they have to assume a > specific thickness of the coverslip: 170 um. > Which is why for high resolution microscopy it is > mandatory to use such coverslips. > > One would think that coverslip producers have > matching thickness as their standard. However > this is far from true. What you usually get is > "Thickness 1" which is defined as 130 - 160 nm. I > never found out why they to this to the > microscopic community. Maybe somebody can enlighten me. > |
|
|
In reply to this post by Guy Cox
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
Hello All,
We are imaging fluorescent virus particles immobilized on the glass
coverslip - on average the signal is restricted to 4 - 9 pixels
depending on mag.
In this case we are imaging right on the surface of a cover glass.
When imaging with the 60x NA 1.4 lens a half of the image, top
and bottom (or corners of the image) are out of focus. The RI of the
mounting media is >1.43 (varies from 1.43 to 1.47, maturation time
dependent). When imaging with the 100x NA 1.45 TIRF lens, the out of focus
portion of the image is less pronounced (higher mag. and/or higher
NA).
In addition, the background fluorescence is much lower with another,
lower RI mounting media (RI 1.42), than with the higher RI media mentioned above
(RI 1.43-1.47). As expected the background fluorescence is reproducible
with the RI=1.42 media, and varies up to 2-fold with the other. SAs are often
seen at the corners of the image.
What is the contribution of the flatness of a coverslip to optical
artefacts observed, as standard coverslip thickness is 160-190 nm? How rough is
the surface of a standard cover glass?
Another issue is inclusions, dust, air bubbles that likely to accumulate at
the surface of a cover glass, and which could be very much similar in
size to virus particles - ca. 120 nm.
To what extent the difference in RI between the glass and the mounting
media contribute to the level of light dispersion and reflection form the cover
glass, especially at high NA?
Thus, what is the most appropriate mounting media for the imaging of
fluorescent molecules immobilized on the surface of a cover glass (in wide-field
microscopy).
Or evanescent wave excitation with a thicker glass (#2, to
reduce photobleaching) would be of great advantage under the TIR
conditions, when even illumination could be achieved?
Vitaly
NCI-Frederick
301-846-6575
|
«
Return to Confocal Microscopy List
|
1 view|%1 views
| Free forum by Nabble | Edit this page |