why does high NA excitation illumination give better resolution in fluorescence microscopy?

[Daniel White-2]

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

Hi again all,

So from the informative answers so far...
it seems that the idea of filling the back focal plane of the objective
(acting as the condenser lens) in epi fluorescence, both widefield and
confocal, is more about getting lots of light into the sample in order to
get enough signal to then be able to see the resolution that is there,
since effective resolution is limited by signal to nose (sorry, noise)

In other words, is there enough light going in (and consequently coming
back out) that we can see the Rayleigh criterion dip between the two peaks
of the object images,  despite any noise that's present?

Is this following statement then true?:

In the case of infinite signal to noise, where contrast is optimal and not
limited by noise, where illumination power is close to saturation of the
fluorescence excited state, exposure time is long enough, etc....
the effective NA of illumination (back aperture filling) has no effect on
achievable lateral or axial resolution - only signal:noise does.

Are there cases where this is false? Anisotropy?

cheers

Dan

[mmodel]

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

I think, in theory, illumination aperture doesn't affect resolution in widefield fluorescence (as it does in confocal). But since the same objective is also used for detection, it does.

Mike

-----Original Message-----
From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of Daniel White
Sent: Monday, February 03, 2014 11:42 AM
To: [hidden email]
Subject: why does high NA excitation illumination give better resolution in fluorescence microscopy?

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

Hi again all,

So from the informative answers so far...
it seems that the idea of filling the back focal plane of the objective (acting as the condenser lens) in epi fluorescence, both widefield and confocal, is more about getting lots of light into the sample in order to get enough signal to then be able to see the resolution that is there, since effective resolution is limited by signal to nose (sorry, noise)

In other words, is there enough light going in (and consequently coming back out) that we can see the Rayleigh criterion dip between the two peaks of the object images,  despite any noise that's present?

Is this following statement then true?:

In the case of infinite signal to noise, where contrast is optimal and not limited by noise, where illumination power is close to saturation of the fluorescence excited state, exposure time is long enough, etc....
the effective NA of illumination (back aperture filling) has no effect on achievable lateral or axial resolution - only signal:noise does.

Are there cases where this is false? Anisotropy?

cheers

Dan

[Arne Seitz]

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

Hi Dan,

the discussions so far are strictly true for wide field microscopy. There the NA of the condenser should not play a role (given enough SNR). And to my understanding light-sheet microscopy is just a prominent example of such a scenario.
For confocal microscopy things are to my understanding a bit more complicated. The gain in lateral resolution with an infinitesimal small pinhole is due to the fact that the excitation point-spread function is convolved with the emission point spread function. Thus in that case the NA of the condensor should play a role.  But as the lateral resolution improvement is hardly seen in practice this is just a theoretical limitation.

Cheers
Arne

> Hi again all,
>
> So from the informative answers so far...
> it seems that the idea of filling the back focal plane of the objective (acting as
> the condenser lens) in epi fluorescence, both widefield and confocal, is more
> about getting lots of light into the sample in order to get enough signal to
> then be able to see the resolution that is there, since effective resolution is
> limited by signal to nose (sorry, noise)
>
> In other words, is there enough light going in (and consequently coming back
> out) that we can see the Rayleigh criterion dip between the two peaks of the
> object images,  despite any noise that's present?
>
> Is this following statement then true?:
>
> In the case of infinite signal to noise, where contrast is optimal and not
> limited by noise, where illumination power is close to saturation of the
> fluorescence excited state, exposure time is long enough, etc....
> the effective NA of illumination (back aperture filling) has no effect on
> achievable lateral or axial resolution - only signal:noise does.
>
> Are there cases where this is false? Anisotropy?
>
> cheers
>
> Dan

[Reto Fiolka]

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

Dear all

Besides confocal microscopy, SIM very much benefits from high NA on the
illumination side.

The higher the NA, the smaller the line spacing can be. The resolution gain in
SIM is reciprocal to the line spacing.
To give numbers, with 488nm excitation, I have achieved 120nm lateral
resolution with SIM with a NA 1.2 objective, whereas with a NA 1.45 TIRF
objective, 80nm lateral resolution is possible.

For the latter case, the corresponding line spacings, as low as 170nm, only
exist in the near field at 488nm.

This is of importance (also for confocal imaging) when imaging watery samples.
An NA above ~1.33 will not improve resolution in the far field, as anything
above is lost in the nearfield.

I have seen many papers where people employed high NA oil objectives for
confocal imaging in biological samples. Besides not being indexed matched, they
will not provide more resolution, as any illumination above the critical angle is
lost in the near field (critical angle for n=1.33 is reached at NA 1.33).

Best,
Reto

[Guy Cox-2]

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

I agree with this statement but this applies only to widefield fluorescence, absolutely NOT to confocal.  See another post (which I am about to write).

                                             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 Daniel White
Sent: Tuesday, 4 February 2014 3:42 AM
To: [hidden email]
Subject: why does high NA excitation illumination give better resolution in fluorescence microscopy?

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

Hi again all,

So from the informative answers so far...
it seems that the idea of filling the back focal plane of the objective (acting as the condenser lens) in epi fluorescence, both widefield and confocal, is more about getting lots of light into the sample in order to get enough signal to then be able to see the resolution that is there, since effective resolution is limited by signal to nose (sorry, noise)

In other words, is there enough light going in (and consequently coming back out) that we can see the Rayleigh criterion dip between the two peaks of the object images,  despite any noise that's present?

Is this following statement then true?:

In the case of infinite signal to noise, where contrast is optimal and not limited by noise, where illumination power is close to saturation of the fluorescence excited state, exposure time is long enough, etc....
the effective NA of illumination (back aperture filling) has no effect on achievable lateral or axial resolution - only signal:noise does.

Are there cases where this is false? Anisotropy?

cheers

Dan

[Guy Cox-2]

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

Arne,

This is not quite true.  In terms of confocal imaging with a pinhole of one Airy diameter or larger, the lateral resolution is determined only by the excitation NA, not the detection NA.   They are or course the same, but you must expand the illumination beam to fill the pupil of the objective or you will lose resolution.  

In confocal we are also concerned with axial (Z) resolution, and here both NAs are important, very important since the axial resolution depends on the square of the NA.  

                        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 Seitz Arne
Sent: Tuesday, 4 February 2014 4:25 AM
To: [hidden email]
Subject: Re: why does high NA excitation illumination give better resolution in fluorescence microscopy?

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

Hi Dan,

the discussions so far are strictly true for wide field microscopy. There the NA of the condenser should not play a role (given enough SNR). And to my understanding light-sheet microscopy is just a prominent example of such a scenario.
For confocal microscopy things are to my understanding a bit more complicated. The gain in lateral resolution with an infinitesimal small pinhole is due to the fact that the excitation point-spread function is convolved with the emission point spread function. Thus in that case the NA of the condensor should play a role.  But as the lateral resolution improvement is hardly seen in practice this is just a theoretical limitation.

Cheers
Arne

> Hi again all,
>
> So from the informative answers so far...
> it seems that the idea of filling the back focal plane of the
> objective (acting as the condenser lens) in epi fluorescence, both
> widefield and confocal, is more about getting lots of light into the
> sample in order to get enough signal to then be able to see the
> resolution that is there, since effective resolution is limited by
> signal to nose (sorry, noise)
>
> In other words, is there enough light going in (and consequently
> coming back
> out) that we can see the Rayleigh criterion dip between the two peaks
> of the object images,  despite any noise that's present?
>
> Is this following statement then true?:
>
> In the case of infinite signal to noise, where contrast is optimal and
> not limited by noise, where illumination power is close to saturation
> of the fluorescence excited state, exposure time is long enough, etc....
> the effective NA of illumination (back aperture filling) has no effect
> on achievable lateral or axial resolution - only signal:noise does.
>
> Are there cases where this is false? Anisotropy?
>
> cheers
>
> Dan

[Steffen Dietzel]

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

Am 03.02.2014 18:39, schrieb Reto Fiolka:

> I have seen many papers where people employed high NA oil objectives for
> confocal imaging in biological samples. Besides not being indexed matched, they
> will not provide more resolution, as any illumination above the critical angle is
> lost in the near field (critical angle for n=1.33 is reached at NA 1.33).

Well, NA 1.33 does sound like better resolution than the NA 1.2 of the
best Water-Coverslip-Objectives, does it not?

And just for the fun of it, if cells grow directly on the coverslip, n
is somewhat larger than 1.33 since you stay inside the cell for the
important parts of the image. (not that I would expect this to make a
big difference though)

Steffen


--
------------------------------------------------------------
Steffen Dietzel, PD Dr. rer. nat
Ludwig-Maximilians-Universität München
Walter-Brendel-Zentrum für experimentelle Medizin (WBex)
Head of light microscopy

Mail room:
Marchioninistr. 15, D-81377 München

Building location:
Marchioninistr. 27,  München-Großhadern

[Guy Cox-2]

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

I think SA would negate any resolution improvement.  But what many people have suggested is that these people are getting 'accidental TIRF' - in other words, TIRF-enhanced fluorescence of structures very close to the coverslip. I don't know of any serious studies investigating this.  

                                Guy

-----Original Message-----
From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of Steffen Dietzel
Sent: Wednesday, 5 February 2014 4:56 AM
To: [hidden email]
Subject: Re: why does high NA excitation illumination give better resolution in fluorescence microscopy?

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

Am 03.02.2014 18:39, schrieb Reto Fiolka:

> I have seen many papers where people employed high NA oil objectives for
> confocal imaging in biological samples. Besides not being indexed matched, they
> will not provide more resolution, as any illumination above the critical angle is
> lost in the near field (critical angle for n=1.33 is reached at NA 1.33).

Well, NA 1.33 does sound like better resolution than the NA 1.2 of the
best Water-Coverslip-Objectives, does it not?

And just for the fun of it, if cells grow directly on the coverslip, n
is somewhat larger than 1.33 since you stay inside the cell for the
important parts of the image. (not that I would expect this to make a
big difference though)

Steffen


--
------------------------------------------------------------
Steffen Dietzel, PD Dr. rer. nat
Ludwig-Maximilians-Universität München
Walter-Brendel-Zentrum für experimentelle Medizin (WBex)
Head of light microscopy

Mail room:
Marchioninistr. 15, D-81377 München

Building location:
Marchioninistr. 27,  München-Großhadern

[John Oreopoulos]

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

In the context of widefield epifluorescence, evidence of "accidental TIRF" is presented in this article:

Burghardt, T.P., Evanescent field shapes excitation profile under axial epi-illumination. Journal of Biomedical Optics, 2012. 17(6).

I would think that the effect described in this paper could be extended to confocal imaging as well. Quotes from the paper:

"Generally, TIRF or epi- illumination excitations pertain to evanescent or propagating field microscopies that are appropriate for different applications. I show here that the TIRF objective under common axial epi-illumination conditions produces an evanescent field that favorably remodels the excitation volume for samples near the coverslip."

"Curve fitting of the axial emission profile (Fig. 10) showed that for the fluorescent sphere at the lower coverslip, exciting light axial polarization from the evanescent field contributes substantially to the observed fluorescence. This is additional evidence for the significance of the evanescent field under axial epi-illumination excitation. It shows that evanescent excitation contributes to observed fluorescence whenever a TIRF objective is used and suggests that the sample material nearest the coverslip disproportionally contributes to the observed fluorescence signal."



John Oreopoulos
Staff Scientist
Spectral Applied Research Inc.
A Division of Andor Technology
Richmond Hill, Ontario
Canada
www.spectral.ca



On 2014-02-05, at 6:42 AM, Guy Cox wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> I think SA would negate any resolution improvement.  But what many people have suggested is that these people are getting 'accidental TIRF' - in other words, TIRF-enhanced fluorescence of structures very close to the coverslip. I don't know of any serious studies investigating this.  
>
>                               Guy
>
> -----Original Message-----
> From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of Steffen Dietzel
> Sent: Wednesday, 5 February 2014 4:56 AM
> To: [hidden email]
> Subject: Re: why does high NA excitation illumination give better resolution in fluorescence microscopy?
>
> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Am 03.02.2014 18:39, schrieb Reto Fiolka:
>
>> I have seen many papers where people employed high NA oil objectives for
>> confocal imaging in biological samples. Besides not being indexed matched, they
>> will not provide more resolution, as any illumination above the critical angle is
>> lost in the near field (critical angle for n=1.33 is reached at NA 1.33).
>
> Well, NA 1.33 does sound like better resolution than the NA 1.2 of the
> best Water-Coverslip-Objectives, does it not?
>
> And just for the fun of it, if cells grow directly on the coverslip, n
> is somewhat larger than 1.33 since you stay inside the cell for the
> important parts of the image. (not that I would expect this to make a
> big difference though)
>
> Steffen
>
>
> --
> ------------------------------------------------------------
> Steffen Dietzel, PD Dr. rer. nat
> Ludwig-Maximilians-Universität München
> Walter-Brendel-Zentrum für experimentelle Medizin (WBex)
> Head of light microscopy
>
> Mail room:
> Marchioninistr. 15, D-81377 München
>
> Building location:
> Marchioninistr. 27,  München-Großhadern

[Reto Fiolka]

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

"Well, NA 1.33 does sound like better resolution than the NA 1.2 of the
best Water-Coverslip-Objectives, does it not?"

For 3D imaging I would always prefer an index-matched objective, even though it
may have lower NA. The higher NA (the part that actually can make it into the far
field) of the oil objective will not be of much use in a watery sample, as spherical
and higher order abberations will quickly eat up any resolution gains when going
into the sample.

The NA 1.2 water instead will achieve a much higher Strehl ratio within the
sample and thus will come closer to diffraction limited performance.

The situation would change if all microscopes were equipped with adaptive optics,
then one would need just one high NA objective that could do it all.

Best,
Reto