http://confocal-microscopy-list.275.s1.nabble.com/Condenser-lens-choice-for-a-given-objective-tp7584901p7584922.html
Resolution in transmitted light bright field microscopy depends just as much on the condenser NA as on the objective's NA.
NO, NO, NO!!!!!!!! I refuse to repeat myself, just go back and reread this thread.
Thanks everyone for your input. Here's the summary of what I learned (please correct me if I am wrong):
Resolution in transmitted light bright field microscopy depends just as much on the condenser NA as on the objective's NA. The light that is scattered from the sample into the objective by light waves traveling off-axis encode the fine details in the specimen. Illuminating the sample at large angles results in scattered waves entering the objective that otherwise would not have been within the objective's NA had the sample been illuminated with on-axis, parallel light.
On the other hand, to improve contrast, there are numerous digital means offering background subtraction, etc. Optically, contrast can be improved by illuminating the sample with on-axis, collimated light.
I would conclude based on the discussion that, if one performs no post-processing, there is a trade-off between resolution and contrast in transmitted light bright field microscopy. One must decide what is best for his or her experiment and set the illumination accordingly.
> *****
> 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 Michael,
>
> Contrast limited? ... only if using eyes.
>
> Video (1970s+, Reynolds, Shinya Inoue, Robert & Nina Allen, later
> "Nanovid" folks, Dodt NIR-VEC-DIC), computer frame grabber background
> subtraction + offset (late 1980s, ex. Matrox MVP-AT board and
> Image-1/AT) and more recently (very late 1980s, early 1990s)
> scientific digital CCD cameras (ex. Photometrics Star-1, Hamamatsu
> digital CCDs [C4742?]) and background subtraction in computer memory.
>
> Reference (which can correct any of my dates): Shinya Inoue, 1986
> Video Microscopy (or more recently Inoue & Spring).
>
> //
>
> An early review on video (brightfield is not mentioned in the
> abstract) - Reynolds 1972 PubMed 4404351
>
> George T. Reynolds
>
>
> Image intersification applied to biological problems
>
>
> Quarterly Reviews of Biophysics
> <
http://journals.cambridge.org/action/displayJournal?jid=QRB> /Volume
> 5
> <
http://journals.cambridge.org/action/displayBackIssues?jid=QRB&volume> Id=5>
> /Issue 03
> <
http://journals.cambridge.org/action/displayIssue?jid=QRB&volumeId=5&> seriesId=0&issueId=03>
> / August 1972, pp 295 - 347
>
> DOI:
http://dx.doi.org/10.1017/S0033583500000974(About DOI
> <
http://journals.cambridge.org/action/stream?pageId=3624&level=2&sessi> onId=F32BC6B4AAE6DB50F8D0214BB5DF4743.journals#30>),
>
>
> In many important types of observations in biological research, the
> information provided by the specimens is in the form of photons-quanta
> of visible light, u.v., or X-rays. The process of observation becomes
> one of recording this information in useful form, with as high an
> efficiency as possible. The problem becomes particularly important
> when for some reason or other the total number or rate of quanta
> provided by the specimen is small. Examples of such limitations are
> included in the following: (i) Processes permitting only low-intensity
> illumination in order not to interfere with the biological processes
> under observation. (ii) Processes changing very rapidly and requiring
> rapid sequence recording. (iii) Processes providing only a limited
> number of photons per event, such as bioluminescence. (iv) Processes
> in which radioactive tracers are utilized, and observation of
> radioactivity is desirable at low specific activity or within short
> time intervals. (v) X-ray diffraction processes where the specimen is
> weakly diffracting or where the X-ray intensity must be kept low in
> order not to damage the specimen. (vi) Processes involving the
> observation of fluorescence, where the intensity is low because of
> limitations on the amount of tagging material.
>
> A
> George
>
> On 3/20/2016 7:33 AM, Michael Model 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.
>> *****
>>
>> In my opinion, visibility in bright field is more limited by contrast
>> than by theoretical resolution...
>>
>> As for brightness, "in the absence of strong scattering, brightness
>> in transmitted illumination depends mostly on direct, and not on
>> diffracted, light (as the popular formula (NAob/M)2 assumes) and thus
>> on the smallest NA between the objective and condenser; this can be
>> easily verified by using an objective with a variable numerical
>> aperture. "
>>
>> Mike Model
>>
>> On Sat, Mar 19, 2016 at 1:07 PM, George McNamara
>> <
[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.
>>> *****
>>>
>>> Hi Kyle,
>>>
>>> Your post indicates that for now you just want to get a brightfield
>>> image through your 100x objective lens. The purpose of an objective
>>> lens is two
>>> fold:
>>> 1. resolution ... emphasized in the replies below ... high NA
>>> condenser results in higher spatial resolution.
>>> 2. brightness ... get enough photon flux so you can do your experiment.
>>>
>>> NA: at Guy's limit of zero NA on the illumination side, you will few
>>> photons reaching the objective lens and detector(s). You will also
>>> have
>>> (practically) infinite depth of focus with respect to imaging dust
>>> and other things on every optical surface. Many of these could be
>>> cleaned up by background subtraction (plus constant).
>>>
>>> Brightness is proportional to:
>>>
>>> NA^4
>>> B ~ ----------
>>> M^2
>>>
>>> The numerator is assuming equal NA (ex. epi-illumination with a
>>> single objective lens). I assume this could be re-written as
>>>
>>> (NAcond^2)*(NAobj^2)
>>>
>>> which is ok until Guy's limit of NAcond = 0. I'll also mention that
>>> you do not even need a transmitted light LED, condenser, or
>>> condenser arm if you have a sensitive enough detector(s): room
>>> lights, desk lamps, computer monitors can provide enough light (I
>>> first encountered this problem/feature working with James Sabry in
>>> Jim Spudich's lab using a back illuminated CCD on an inverted
>>> microscope, no recalling what objective lens, but was 19 years ago
>>> and available in their published papers).
>>>
>>> More importantly, M^2 means that your 100x lens is putting 1% as
>>> much photon flux onto a pixel as a 10x lens would.
>>>
>>> My advice: go find a long working distance objective lens that gets
>>> enough light onto your specimen to get you a useful brightfield
>>> image. You could later figure out if you need a phase contrast/DIC
>>> turret, what NA and working distance you need etc.
>>> I also suggest instead of hardware contrast (DIC requires polarizers
>>> and prisms ... can avoid illumination side polarizer if using a
>>> laser as in confocal microscope stand DIC using the transmitted
>>> light pathway in reverse; phase contrast requires a phase ring in
>>> the objective lens - usually with a lower NA for a given price
>>> point), you start looking into synthetic contrast options. The
>>> simplest is to just go with digital contrast by background
>>> subtraction.
>>> Software to get quantitative phase microscopy data of wet mass
>>> (leading to dry mass) - a couple of links and comments:
>>>
>>>
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3555125/ ... McCarty ...
>>> uses 0.1 NA on condenser side.
>>>
>>>
http://www.jove.com/video/50988/quantitative-optical-microscopy-meas>>> urement-cellular-biophysical
>>>
>>> ... McCarty, see downloads (.M files).
>>>
>>> Nugent / IATIA (now Ultima Capital) developed the first software
>>> only solution
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4624349/.
>>>
http://www.ncbi.nlm.nih.gov/pubmed/15800856 ... Nugent / IATIA on
>>> confocal
>>>
>>>
http://onlinelibrary.wiley.com/doi/10.1111/j.1440-1681.2004.04100.x/>>> abstract
>>>
>>>
http://aups.org.au/Proceedings/34/121-127/121-127.pdf ... fig 3
>>> illustrates synthetic phase and DIC; fig 4 shows improves segmentation.
>>>
>>>
http://www.ultimacapital.net/iatiaimaging/Publications/Iatia%20Imagi>>> ng/applicationNotes/comparisonWithOpticalPhaseContrastModalities.pdf
>>>
>>>
>>>
http://www.ultimacapital.net/iatiaimaging/Publications/Iatia%20Imagi>>> ng/applicationNotes/measurementOfAreaChanges.pdf
>>>
>>> (I think you can ignore the "Confidence-Publication Pending" at top
>>> - this
>>> appnote was posted by the manufacturer and has been online for years).
>>>
>>>
>>> Hardware assisted (not a complete list - some use holography, others
>>> interferometry):
>>> Ovizio
>>> Gabriel Popescu
>>> Graham Dunn
>>>
>>> PubMed has more - a simple search is: "quantitative phase microscopy"
>>>
>>> McCarty's JoVE article now has downloadable .M (MatLab) files. If
>>> Anne Carpenter or anyone on the Cellprofiler team is reading this
>>> (or someone send is it their way), I encourage Anne to work with
>>> McCarty and their University to get "MaCarty QPm" into Cellprofiler.
>>>
>>> not QPm, this may still be of interest to listservites:
>>> Direct imaging of phase objects enables conventional deconvolution
>>> in bright field light microscopy
>>>
http://www.ncbi.nlm.nih.gov/pubmed/24558478>>>
>>> //
>>>
>>> Getting more out of high NA objective lens ... confocal or widefield
>>> interference reflection microscopy (IRM) provides data on
>>> cell-substratum adhesion ... including contact area. In reflection
>>> confocal (ok, for Jim Pawley and Guy Cox: scattered confocal), you
>>> can get optical sections of the cell, "label free". Some IRM data I
>>> posted online:
>>>
>>>
http://works.bepress.com/gmcnamara/10/>>>
http://works.bepress.com/gmcnamara/7/>>>
>>> Widefield IRM is very simple if your filter cube does not have an
>>> exciter filter (best to do this with a wavelength and intensity
>>> selectable LED illuminator than a broad spectrum arc lamp): just
>>> turn on a
>>> wavelength(s)
>>> that enable some light to bounce from the cells/coverglass through
>>> the dichroic and emission filter (could do even better with a
>>> dichroic only, and even better with a 50/50 beamsplitter only). My
>>> thanks to Tom DiMatteo, Epi Technology, for telling me about single
>>> LED control on my early gen SOLA. I had a long chat with Tom at his
>>> ABRF booth a couple of years ago.
>>> IRM can be quantitative
>>>
http://www.ncbi.nlm.nih.gov/pubmed/23024911>>>
http://www.ncbi.nlm.nih.gov/pubmed/20013754>>>
http://www.ncbi.nlm.nih.gov/pubmed/3900106 ... Verschueren
>>>
>>>
>>>
>>> George
>>>
>>>
>>>
>>> On 3/18/2016 9:00 PM, Guy Cox 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.
>>>> *****
>>>>
>>>> This is a little bit oversimplified. The Rayleigh criterion does not
>>>> apply to a widefield image, but does apply in fluorescence. The
>>>> condenser NA normally IS the objective NA, since they are usually
>>>> one and the same thing, but condenser NA only affects brightness,
>>>> not resolution.
>>>>
>>>> The Abbe criterion r = 0.5 lambda / NA applies in transmitted
>>>> light, but ONLY if the condenser aperture equals or exceeds the
>>>> objective NA.
>>>> Reducing the condenser NA does not have the same effect as reducing
>>>> the objective NA. Reducing the condenser NA to 0 (parallel
>>>> illumination)
>>>> worsens the resolution to r = lambda/NA - ie 50% of what the
>>>> objective should give.
>>>>
>>>> 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 Aryeh Weiss
>>>> Sent: Friday, 18 March 2016 11:38 PM
>>>> To:
[hidden email]
>>>> Subject: Re: Condenser lens choice for a given objective
>>>>
>>>> *****
>>>> 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.
>>>> *****
>>>>
>>>> In a transmitted-light brightfield image, the Rayleigh criterion
>>>> includes both the objective NA and the condenser NA
>>>> (1.22 lambda/(NA_obj +NA_cond)) . The makes sense because even a
>>>> very small NA objective can receive light scattered at a large
>>>> angle if the NA of the condenser is large. (This is how dark-field
>>>> works).
>>>> So it would appear that in principle, you benefit from having a
>>>> condenser with as large as NA as possible (although you may not
>>>> have much contrast on that brightfield image).
>>>>
>>>> BTW, you can have a "poor" man's dark field scope by using a low-NA
>>>> objective with a phase ring made for a higher NA objective.
>>>> For example, in my teaching lab, the students get very nice
>>>> darkfield images using our 4x/NA=0.1 objective with the ph2 phase ring.
>>>>
>>>> --aryeh
>>>>
>>>>
>>>>
>>>> On 18/03/2016 10:26 AM, Kyle Douglass 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.
>>>>> *****
>>>>>
>>>>> Thanks for the feedback, Barbara. It is very helpful.
>>>>>
>>>>> I have heard the advice before about the condenser NA needing to
>>>>> be greater than or equal to the objective NA. Can you offer some
>>>>> physical explanation or intuition for why this is?
>>>>>
>>>>> One admittedly incomplete explanation I can think of for the
>>>>> recommendation goes like this: the light collected by the
>>>>> objective consists of two parts. One part is the transmitted light
>>>>> that is not scattered by the sample. The other part is the light
>>>>> scattered by the sample. If the condenser's working NA is smaller
>>>>> than the objective's, then the unscattered, transmitted light
>>>>> fills only a portion of the objective's back focal plane. However,
>>>>> the light scattered by the sample will probably be dispersed
>>>>> across the entire back focal plane because it will encode all the
>>>>> spatial frequencies of the sample.
>>>>>
>>>>> I wonder if it's the inhomogeneous distribution of light from the
>>>>> two components in the objective's back focal plane that leads to
>>>>> the matched NA requirements of the condenser and objective. Does
>>>>> this make sense?
>>>>>
>>>>> Thanks!
>>>>> Kyle
>>>>>
>>>>> On 03/17/2016 08:22 PM, Barbara Foster 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.
>>>>>> *****
>>>>>>
>>>>>> Hi, Kyle
>>>>>>
>>>>>> The general rule of thumb is that the NA on the condenser should
>>>>>> meet or exceed that of the objective.
>>>>>>
>>>>>> If you are using oil immersion objectives, ideally, to achieve
>>>>>> that goal, you should use an oil immersion condenser, otherwise
>>>>>> you are limited to an NA of 0.9.
>>>>>>
>>>>>> Also, remember that the aperture iris in the condenser adjusts
>>>>>> the condenser's WORKING numerical aperture. Just because the
>>>>>> condenser is marked 1.4 NA doesn't mean that, in a practical
>>>>>> experiment, it will be operating at 1.4. I follow the guidelines
>>>>>> set down by Frits Zernicke (inventor of Phase contrast): gently
>>>>>> close the aperture iris to the "Oomph" position: that delicate
>>>>>> balance between sufficient edge definition and optimum resolution
>>>>>> (Yes, the condenser does contribute to resolution).
>>>>>>
>>>>>> As for planning for growth:
>>>>>> You might want to invest in a turret condenser early on. That
>>>>>> will give you the option to add those other contrast techniques
>>>>>> as you grow into them.
>>>>>>
>>>>>> And just one more reminder, specifically regarding DIC:
>>>>>> If you are going to use plastic vessels (petri dishes, multi-well
>>>>>> plates, growth flasks), use Hoffman Modulation Contrast instead
>>>>>> of DIC. DIC uses polarized light. The plastic will affectt the
>>>>>> shear and cause effects that will be hard to interpret. Some HMC
>>>>>> set-ups do use pol to control the width of the slit in the
>>>>>> condenser, but all of that is on the incoming side of the sample
>>>>>> and will not be affected by plastic containers.
>>>>>>
>>>>>> Good hunting!
>>>>>> Barbara Foster, President & Chief Consultant
>>>>>> Microscopy/Microscopy Education ... "Education, not Training"
>>>>>> 7101 Royal Glen Trail, Suite A - McKinney, TX 75070 - P:
>>>>>> 972-924-5310 www.MicroscopyEducation.com
>>>>>>
>>>>>> Microscopy/Microscopy Education is a division of The Microscopy &
>>>>>> Imaging Place, Inc.
>>>>>>
>>>>>>
>>>>>> NEW! Getting involved in Raman or FTIR?
>>>>>> MME is now offering courses in these areas specifically for
>>>>>> microscopists!
>>>>>> Now scheduling courses through the mid 2016. We can customize a
>>>>>> course on nearly any topic, from fluorescence to confocal to
>>>>>> image analysis to SEM/TEM.
>>>>>> Call today for a free training evaluation.
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> At 08:36 AM 3/17/2016, Kyle Michael Douglass 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.
>>>>>>> *****
>>>>>>>
>>>>>>> Hello listers,
>>>>>>>
>>>>>>>
>>>>>>> I have a couple of questions about condensers for you. I'd like
>>>>>>> to do some transmitted light imaging in an inverted microscope
>>>>>>> using high magnification, oil-immersion objectives. For the
>>>>>>> moment, I don't need to do anything other than brightfield with
>>>>>>> a high power LED light source. It might be nice to do phase
>>>>>>> contrast or DIC in the future, but I don't need it now.
>>>>>>>
>>>>>>>
>>>>>>> My questions are:
>>>>>>>
>>>>>>>
>>>>>>> 1) What are the rules of thumb for matching a brightfield
>>>>>>> condenser to an objective? I won't be using anything but oil
>>>>>>> immersion objectives with NA's greater than 1.4.
>>>>>>>
>>>>>>>
>>>>>>> 2) If I do want to do phase contrast or DIC in the future,
>>>>>>> should I put special consideration into the condenser lens
>>>>>>> selection now? I imagine the condenser NA will determine what
>>>>>>> phase contrast rings I can use, but does it impact DIC?
>>>>>>>
>>>>>>>
>>>>>>> Thanks!
>>>>>>>
>>>>>>> Kyle
>>>>>>>
>>>>>>>
>>>>>>> Dr. Kyle M. Douglass
>>>>>>> Post-doctoral Researcher
>>>>>>> EPFL - The Laboratory of Experimental Biophysics
>>>>>>>
http://leb.epfl.ch/ http://kmdouglass.github.io>>>>>>>
>>>> --
>>>> Aryeh Weiss
>>>> Faculty of Engineering
>>>> Bar Ilan University
>>>> Ramat Gan 52900 Israel
>>>>
>>>> Ph: 972-3-5317638
>>>> FAX: 972-3-7384051
>>>>
>>>>
>>> --
>>>
>>>
>>>
>>> George McNamara, Ph.D.
>>> Single Cells Analyst, T-Cell Therapy Lab (Cooper Lab) University of
>>> Texas M.D. Anderson Cancer Center Houston, TX 77054 Tattletales
>>>
http://works.bepress.com/gmcnamara/42>>>
http://works.bepress.com/gmcnamara/75>>>
https://www.linkedin.com/in/georgemcnamara>>>
>
>
Kyle M. Douglass, PhD