Widefield laser epi-illumination

[stephenx_86]

Hi,

I'm trying to set up a microscope to have wide-field laser epi-illumination. This is to be used with an infinity-focused objective lens. My current plan has the laser expanded 4x using a Galilean beam expander (from 1mm to 4mm) and coupled into the back of the objective using a dichroic. However, from reading around online I'm a bit confused about a couple of things.

1. I want to have homogeneous illumination across my sample; however, if my logic serves, by shooting the parallel laser light into the back of the objective it will focus down to a spot at the sample. Is this correct? (i.e. will I just get a spot of illumination at the centre of my image?)

2. If I expand the beam sufficiently to fill the back aperture of the objective, will this still be an issue?

2. Alternatively, should I be focussing my laser to the back aperture of the objective? I've seen things that indicate this is the case, but I can't work out if this approach only applies to finite-focus objectives.

Any advice is greatly appreciated.

Thanks
Stephen

 

[UltrafastPED]

If the beam goes through the objective lens it will focus onto the slide.

Since your objective is infinity focused the focus will be strongest if the beam is parallel, which is what the laser and the beam expander provide.

You could try making the beam expand or converge prior to entering the objective; if you hit the right combination you may get a larger spot, but most of the light will be clipped by the aperture stops.

Check the manufacturer's specifications for your lens to discover any provision for lighting; better yet - contact the manufacturer's sales support team and pose your question to them. I'm sure that they have an answer.

 

[Andy Resnick]

Hi,

I'm trying to set up a microscope to have wide-field laser epi-illumination. <snip>

You are correct in that illuminating the back pupil of the objective with a plane wave will produce a focused spot (Airy disc) at the sample plane. Recall that off-axis illumination points at the sample produce a tilted plane wave at the back pupil plane. If you want to produce a uniformly illuminated 'image' circle at the sample plane, you must illuminate the back pupil plane with something like a diverging spherical wave. A point source located at the back pupil plane will produce a plane wave at the front focal plane.

I can't be too much more specific without knowing any details about the microscope objective- especially the numerical aperture.

What is your application? laser liight is generally poorly suited for wide field illumination

 

[stephenx_86]

Thank you both for your replies.

The objective is a 60x Olympus UPlanSApo lens, which has a numerical aperture of 1.35. We wish to excite relatively small numbers of fluorophores within the imaged region. Just to clarify, would focussing the laser at the back pupil plane achieve a diverging spherical wave?

Would you recommend we implement a mercury lamp instead?

Thanks again,
Stephen

 

[Claude Bile]

Hi Stephen,

1. To get even illumination, microscopists use Kohler illumination;

http://en.wikipedia.org/wiki/Köhler_illumination

This requires that the source be located in the conjugate Fourier plane of the sample. For an infinity-focused objective, this means focusing the source on the back focal plane of the objective.

2. Expanding the beam will "fill" the objective, increasing the bandwidth of the plane wave spectrum incident upon the sample. In your case, filling the objective will only result in reduced coherence; usually the benefit is increased power, but since you are telescoping your source (rather than aperturing it which is normally done), this is moot.

To address your second post, it sounds like you actually want your illumination to be focused (rather than even over the focal plane). Check out confocal imaging; see if it fits your needs.

Just to clarify, would focussing the laser at the back pupil plane achieve a diverging spherical wave?

Any incident wave will result in a spherical wave - the question is where the spherical wave is focused. For an infinity-corrected objective, a plane wave will result in a spherical wave centered on the focal plane. Tilted plane waves will shift the focal point over the focal plane. For an aberration-free objective, this focal plane will be a true (flat) plane.

Incident spherical waves will result in spherical waves centered on planes away from the focal plane (as per the standard lens equation).

Claude.

 

[Andy Resnick]

Thank you both for your replies.

The objective is a 60x Olympus UPlanSApo lens, which has a numerical aperture of 1.35. We wish to excite relatively small numbers of fluorophores within the imaged region. Just to clarify, would focussing the laser at the back pupil plane achieve a diverging spherical wave?

Would you recommend we implement a mercury lamp instead?

Thanks again,
Stephen

I'm still a little unclear on the application- do you have a low concentration of fluorophore, or are you trying to selectively illuminate/bleach a sub-population (e.g. blinking, photoactivation...)?

Is your current setup unable to meet your needs, or are you building an entirely new setup?

 

[stephenx_86]

Hi,

Thank you all once again for these useful responses.

We've got a low concentration of fluorophore and intend to see small populations (each protein will be tagged with a single fluorophore, but may get tens of proteins bind together). We're not intending to photobleach the sample, although I realize that this may be an unavoidable problem with illuminating a large region. The reason lasers were being used were the well defined excitation peak.

I'm adapting an existing setup, which was being built by someone else. They were using laser epi-illumination, but never got round to completing it before leaving. From what was left, it looks like they were sending in a parallel beam; however, talking to others who are still in the lab it sounds like they had been working on focussing the laser at the back pupil.

Thanks
Stephen