A note on alignment and scattering of the TIR laser beam
74 views
Skip to first unread message
Kyle Douglass
Dec 3, 2015, 3:47:56 AM12/3/15
to pgFocus
Hi all,
I noticed something important recently about the alignment of the optics with regards to the pgFocus performance. I thought I would share what I learned on the forum.
In previous posts I noted that the pgFocus performance can be hindered if cells on the coverslip scatter the laser beam and create a messy beam profile. This essentially turns a nice, Gaussian beam into a beam with a lot of fringes, which is difficult for the pgFocus to lock onto. What I didn't realize at the time--though now it seems obvious--is that the strength of the scattering by the sample varies with the penetration depth of the evanescent field into the sample space. This penetration depth is essentially controlled by the location of the finite-sized, focused beam is in back exit pupil of the objective. I attached an illustration below to highlight this.
To get optimal alignment for pgFocus operation, I start with the IR beam in widefield, i.e. it goes straight up from the objective (be careful about looking into the objective when you do this). I then turn the screw on the mirror that translates the beam across the exit pupil of the objective. This translates the focused beam across the exit pupil, as seen in the left-most part of the figure. Once the beam enters the annular region where TIR occurs, and you will see a faint spot on the IR viewer if everything is well aligned. Translating slightly further, the beam goes fully into TIR, producing a bright spot on the IR viewer. This is not optimal, however, because the evanescent light is not strongly confined to the coverslip.
The ideal location is to have a very small amount of light from the focused beam passing near the edge of the aperture, with the rest of the light being cutoff from the finited-sized aperture. This ensures that most of the light is hitting the coverslip at the maximum angles allowed by the objective, which means the evanescent field penetration is small. This reduces scattering of the beam and should give a nice, clean profile. (Note that I also use a zero-aperture iris immediately in front of the laser to clean up the beam before it enters the microscope.)
By consciously placing most of the beam outside the aperture, I have managed to get ~2 to 3 nm std. deviation locking with an NA 1.4 objective and the sample in water.
I hope this helps everyone with their alignments!
Kyle
I noticed something important recently about the alignment of the optics with regards to the pgFocus performance. I thought I would share what I learned on the forum.
In previous posts I noted that the pgFocus performance can be hindered if cells on the coverslip scatter the laser beam and create a messy beam profile. This essentially turns a nice, Gaussian beam into a beam with a lot of fringes, which is difficult for the pgFocus to lock onto. What I didn't realize at the time--though now it seems obvious--is that the strength of the scattering by the sample varies with the penetration depth of the evanescent field into the sample space. This penetration depth is essentially controlled by the location of the finite-sized, focused beam is in back exit pupil of the objective. I attached an illustration below to highlight this.
To get optimal alignment for pgFocus operation, I start with the IR beam in widefield, i.e. it goes straight up from the objective (be careful about looking into the objective when you do this). I then turn the screw on the mirror that translates the beam across the exit pupil of the objective. This translates the focused beam across the exit pupil, as seen in the left-most part of the figure. Once the beam enters the annular region where TIR occurs, and you will see a faint spot on the IR viewer if everything is well aligned. Translating slightly further, the beam goes fully into TIR, producing a bright spot on the IR viewer. This is not optimal, however, because the evanescent light is not strongly confined to the coverslip.
The ideal location is to have a very small amount of light from the focused beam passing near the edge of the aperture, with the rest of the light being cutoff from the finited-sized aperture. This ensures that most of the light is hitting the coverslip at the maximum angles allowed by the objective, which means the evanescent field penetration is small. This reduces scattering of the beam and should give a nice, clean profile. (Note that I also use a zero-aperture iris immediately in front of the laser to clean up the beam before it enters the microscope.)
By consciously placing most of the beam outside the aperture, I have managed to get ~2 to 3 nm std. deviation locking with an NA 1.4 objective and the sample in water.
I hope this helps everyone with their alignments!
Kyle
Ken Jiiii
Dec 3, 2015, 9:17:19 AM12/3/15
to pgFocus
Hi Kyle,
Thank you for sharing your information. Actually that comes quite handy since I am about to align pgFocus in the next week. So far I just have my IR laser in Epi so that I can see a back reflexion with a webcam somwehere where the pgFocus should be. If I know turn on a knob of a mirror that reflection signal decreases as does the intensity in the specimen. At some point there is no signal left at all and if I turn the knob back I come to the point where I have started. In between should be TIR, shouldn't it? How do I recognize that specific position? By IR viewer do you mean an IR viewing card? Unfortunately I cannot see any back reflection with mine, seems the signal is too small that's why I am using the webcamdetecting the signal.
By the way, I haven't focused on the backfocalplane yet so I am just sending the collimated laser light into the aperture.
Thanks a lot!
Thank you for sharing your information. Actually that comes quite handy since I am about to align pgFocus in the next week. So far I just have my IR laser in Epi so that I can see a back reflexion with a webcam somwehere where the pgFocus should be. If I know turn on a knob of a mirror that reflection signal decreases as does the intensity in the specimen. At some point there is no signal left at all and if I turn the knob back I come to the point where I have started. In between should be TIR, shouldn't it? How do I recognize that specific position? By IR viewer do you mean an IR viewing card? Unfortunately I cannot see any back reflection with mine, seems the signal is too small that's why I am using the webcamdetecting the signal.
By the way, I haven't focused on the backfocalplane yet so I am just sending the collimated laser light into the aperture.
Thanks a lot!
Ken Jiiii
Dec 7, 2015, 9:33:26 AM12/7/15
to pgFocus
Hello Kyle,
I am sorry maybe I did not phrase my question throughly and in addition your uploaded image wasn't displaying properly last time. Now I had a nice glance at it.
With the image and text you wrote this time and which is written on your website and in another thread it seems clear to me what to do:
- Align IR laser beam so that it hits the middle of the back aperture of the objective on its optical axis
- Optional: place lense in optical path in order to focus the IR laser beam onto the backfocalplane of the objective. Without this lense the collimated laser beam is focused onto the coverslip surface.
- Displace dichroic mirror (in my case low pass filter) about 2.5mm (depending on the back aperture of the objective) to get to total internal reflection illumination. (this steps corresponds to moving the laser spot in the shaded area idicated in your image).
- Adjust displacement in order to have the ideal case for pgFocus
- place pgFocus in proper position
I hope I summarized your steps correctly so far.
At first I would like to try and set up pgFocus withouth focusing the IR beam on the backfocalplane of the objectiv. Therefore I moved the collimated beam into the middle of the back aperture of the objective (here I used two pinholes and two mirrors to have it on the optical axis). If I now place a webcam to the place where pgFocus is meant to be I can see a spot associated with the IR beam. This spot getts dimmer whe I hold a paper in front of the back aperture of the objective meaning that it is partially a backreflection of the objective and partially from the beamsplitter or low pass filter since it does not vanish completely.
Now I thought I could move the lowpass filter in a way that I see the spot in annother place on the webcam (TIR). And by moving the stage up and down I should see a lateral movement of the spot.
Unfortunately I am not able to find this position where I have this moving spot. I use a sample with some non fluorescent beads and focus on them and then try to get the right position. As I said, unfortunately my laser's intensity is not powerful enough to see it on a viewing card as you showed in your image. So I used the webcam where I also should see the same effect but which I actually don't.
Do you have any ideas what I am doing wrong? I think in addition to a further try I will order the lense and focus onto the backfocalplane. But in theory it should also work without it, shouldn't it?
I hope that I rephrased my problem for better understanding and maybe you can help me and thanks for the well explaining picture.
Thanks in advance!
I am sorry maybe I did not phrase my question throughly and in addition your uploaded image wasn't displaying properly last time. Now I had a nice glance at it.
With the image and text you wrote this time and which is written on your website and in another thread it seems clear to me what to do:
- Align IR laser beam so that it hits the middle of the back aperture of the objective on its optical axis
- Optional: place lense in optical path in order to focus the IR laser beam onto the backfocalplane of the objective. Without this lense the collimated laser beam is focused onto the coverslip surface.
- Displace dichroic mirror (in my case low pass filter) about 2.5mm (depending on the back aperture of the objective) to get to total internal reflection illumination. (this steps corresponds to moving the laser spot in the shaded area idicated in your image).
- Adjust displacement in order to have the ideal case for pgFocus
- place pgFocus in proper position
I hope I summarized your steps correctly so far.
At first I would like to try and set up pgFocus withouth focusing the IR beam on the backfocalplane of the objectiv. Therefore I moved the collimated beam into the middle of the back aperture of the objective (here I used two pinholes and two mirrors to have it on the optical axis). If I now place a webcam to the place where pgFocus is meant to be I can see a spot associated with the IR beam. This spot getts dimmer whe I hold a paper in front of the back aperture of the objective meaning that it is partially a backreflection of the objective and partially from the beamsplitter or low pass filter since it does not vanish completely.
Now I thought I could move the lowpass filter in a way that I see the spot in annother place on the webcam (TIR). And by moving the stage up and down I should see a lateral movement of the spot.
Unfortunately I am not able to find this position where I have this moving spot. I use a sample with some non fluorescent beads and focus on them and then try to get the right position. As I said, unfortunately my laser's intensity is not powerful enough to see it on a viewing card as you showed in your image. So I used the webcam where I also should see the same effect but which I actually don't.
Do you have any ideas what I am doing wrong? I think in addition to a further try I will order the lense and focus onto the backfocalplane. But in theory it should also work without it, shouldn't it?
I hope that I rephrased my problem for better understanding and maybe you can help me and thanks for the well explaining picture.
Thanks in advance!
Kyle Douglass
Dec 7, 2015, 10:15:25 AM12/7/15
to pgFocus
Hi Ken,
I'm sorry for my slow response to your previous e-mail. I have been a bit busy lately trying to finish some projects before Christmas :) See my responses to your question below:
I'm sorry for my slow response to your previous e-mail. I have been a bit busy lately trying to finish some projects before Christmas :) See my responses to your question below:
With the image and text you wrote this time and which is written on your website and in another thread it seems clear to me what to do:
- Align IR laser beam so that it hits the middle of the back aperture of the objective on its optical axis
Yes, I find it easiest that the laser beam starts on axis. This is because it is relatively easy to align the beam to the optical axis and it gives you a point of reference for when you begin to move the beam into TIR. Be sure to check that the beam is leaving the objective along the axis as well.
- Optional: place lense in optical path in order to focus the IR laser beam onto the backfocalplane of the objective. Without this lense the collimated laser beam is focused onto the coverslip surface.
I think this is where your problem is. Assuming the beam from your laser is collimated, it is not optional to place a lens that focuses the beam onto backfocal plane. You need to focus the beam onto the back focal plane so that the laser beam is collimated in the sample space. You can see this if you make a ray trace diagram of a thin, positive lens. All rays originating from the focal point of the lens are collimated on the opposite side of the lens; you want all the rays to totally internally reflect, and the way to achieve this is to make them all parallel to one-another in the sample space.
- Displace dichroic mirror (in my case low pass filter) about 2.5mm (depending on the back aperture of the objective) to get to total internal reflection illumination. (this steps corresponds to moving the laser spot in the shaded area idicated in your image).
This is correct. The spot I drew was intended to be a focused laser spot, not a collimated laser.
- Adjust displacement in order to have the ideal case for pgFocus
- place pgFocus in proper position
At first I would like to try and set up pgFocus withouth focusing the IR beam on the backfocalplane of the objectiv. Therefore I moved the collimated beam into the middle of the back aperture of the objective (here I used two pinholes and two mirrors to have it on the optical axis). If I now place a webcam to the place where pgFocus is meant to be I can see a spot associated with the IR beam. This spot getts dimmer whe I hold a paper in front of the back aperture of the objective meaning that it is partially a backreflection of the objective and partially from the beamsplitter or low pass filter since it does not vanish completely.
Remember that you should be focusing on the back focal plane.
I also made a mistake here. I should have mentioned in my diagram that I always see this back reflection from the objective as well. So in reality I see two spots: one that is always there--the back reflection from the objective--and one that is the TIR beam and depends on the position of the focused laser beam in the exit pupil.
Now I thought I could move the lowpass filter in a way that I see the spot in annother place on the webcam (TIR). And by moving the stage up and down I should see a lateral movement of the spot.
Unfortunately I am not able to find this position where I have this moving spot. I use a sample with some non fluorescent beads and focus on them and then try to get the right position. As I said, unfortunately my laser's intensity is not powerful enough to see it on a viewing card as you showed in your image. So I used the webcam where I also should see the same effect but which I actually don't.
I don't think you'll see a bright TIR spot unless you focus the beam onto the objective's back aperture. And remember that you will only see the beam in TIR when the focused spot is in a very small region of the exit pupil.
If you are still having trouble, try removing the liquid from your coverslip so it is an oil-glass-air interface. This makes it easier to bring the laser into TIR, which helps when troubleshooting your alignment.
Good luck! Let me know how it goes. I will update my diagram to include the back reflection for everyone's benefit.
Kyle
Kyle Douglass
Dec 8, 2015, 9:56:47 AM12/8/15
to pgFocus
When the laser is in TIR, you can tell which spot is which by translating the stage up and down. The direct reflection will not move, but the TIR reflection will translate up to a centimeter laterally for ~10 micron shifts as the objective-coverslip distance is varied.
Cheers,
Kyle
Ken Jiiii
Dec 9, 2015, 2:56:39 AM12/9/15
to pgFocus
Thanks again! I have ordered the lense so I should know more soon!
I thought by not using a lense in the IR beam path the collimated IR laser gets focused on a point of the coverslip while when I focus on the BFP I get a defocused illuminated area on the coverslip (as in epi just shifted for TIR). I have just attached a quick draw of what I have thought in an idelized way (objective lense missing and so on...).
All have a nice pre-Christmas time! :-)
Kyle Douglass
Dec 9, 2015, 3:52:33 AM12/9/15
to pgFocus
Hi Ken,
I think I see your problem. The rays in the figures aren't really correct in the "with lens" part. When the rays are focused in the back focal plane of the objective, they should all be parallel in the sample plane. The rays on the right-hand side of your figure (with lens) are not parallel in the sample plane like they should be. When they are all parallel to one another and they are incident on the coverslip beyond the critical angle, then they are all totally-internally reflected. The size of the illuminated area then depends on the beam's numerical aperture in the back focal plane.
If you have a focused beam, you have rays that span a number of different incidence angles on the coverslip. Some will be TIR, but most will not. The rays that are not in TIR will not be strongly reflected and you will therefore not get a very strong return beam. For this reason, you don't want a focused spot on the sample for efficient TIR.
I hope this better explains how the TIR mechanism is working in through-the-objective illumination. If you're still having trouble understanding it, then I suggest making a thin lens ray trace of a ray fan that is focused to an off-axis point in the back focal plane of the lens. All the rays on the other side of the lens should emerge parallel to one-another and at an angle to the axis that is proportional to the height of the focus above the axis.
Cheers,
Kyle
I think I see your problem. The rays in the figures aren't really correct in the "with lens" part. When the rays are focused in the back focal plane of the objective, they should all be parallel in the sample plane. The rays on the right-hand side of your figure (with lens) are not parallel in the sample plane like they should be. When they are all parallel to one another and they are incident on the coverslip beyond the critical angle, then they are all totally-internally reflected. The size of the illuminated area then depends on the beam's numerical aperture in the back focal plane.
If you have a focused beam, you have rays that span a number of different incidence angles on the coverslip. Some will be TIR, but most will not. The rays that are not in TIR will not be strongly reflected and you will therefore not get a very strong return beam. For this reason, you don't want a focused spot on the sample for efficient TIR.
I hope this better explains how the TIR mechanism is working in through-the-objective illumination. If you're still having trouble understanding it, then I suggest making a thin lens ray trace of a ray fan that is focused to an off-axis point in the back focal plane of the lens. All the rays on the other side of the lens should emerge parallel to one-another and at an angle to the axis that is proportional to the height of the focus above the axis.
Cheers,
Kyle
Ken Jiiii
Dec 10, 2015, 4:59:42 AM12/10/15
to pgFocus
Thanks Kyle, I think I got it now! I'll keep you posted. Best
Reply all
Reply to author
Forward
0 new messages