Hi Dr Ruudawski, I recently started to watch your videos . Thanks for the padagogic approach gettinf down to concepts, and then applying it in practive. In this respect, and related to SAED patterns, would you please explain the concept of camera length, which is important for diffractions. Basically, my intutition says this goes back to the formation of the diff pattern but it will be nice to get the concept right. Many thanks Beatriz
Hey, so very sorry for the very late response. Glad to hear my videos are helping you. Camera length is an archaic term that is used to describe the magnification of the diffraction pattern. It doesn't really apply physically in an S/TEM because the physical distance between the diffraction pattern and the camera is fixed, but we still use the term because of historical reasons. In S/TEMs, we can effectively change it by simply changing the strength of lenses in the projection system. It would be a good video to do though, going over the origins of "camera length", so thanks for the suggestion.
Thanks for this video. So when you make an exposure test, you start with 0.01 in camera view and 0.1 in acquire view and check the count in the acquired image? am I right?. you said that The max count for the acquired image is 650000 counts. is there a maximum count for live images? Another question is when you change the time from 0.1 to o,5 in the acquired camera is there an equation to calculate or expect the count before we acquire the image and test it because the count at 0.1 was between 6000 to 8000 and at 0,5 was between 25000 and 30000 counts. so what happened when we increased the value 5 times in the counts?. Thanks
Thanks for the questions. The max counts in both the live and acquired images is both 2^16 = 65,000 as the dynamic range is 16 bit; the difference between live and acquire modes here is the pixel resolution, with live mode having 512x512 resolution and acquire having 2048x2048 resolution (this is how I have these modes specifically set, but these could be set differently if desired, you could set both modes to be 2048x2048, if you wanted). As far as changing the exposure time and the effect on the resulting counts, the response should be very linear (assuming the same pixel resolution), so there is no special equation needed. In other words, if you multiply the acquire exposure time by 5x, you should get (approximately) 5x more counts; same thing would apply for live exposure. This is basically consistent with what you pointed out, though I agree that the math does not work out perfectly, but that is not necessarily unexpected for very high dynamic range situations (very large difference between darkest and brightest features in the image) like diffraction patterns. I hope this helps and thanks for checking out my video!
Hi Eli: glad you like the videos; yes, it is very feasible to perform nano-area diffraction on a Tecnai; please see this video I did covering CBED: ruclips.net/video/DofabqzZcG8/видео.html
Hi Nicholas. I have found the CBED patterns. My question is: if I am looking at the zone axis of an fcc nanostructure and I wanna determine if it is along [1-10] or [110]. I already know if I tilt the sample to let the [1,-1,-9] and [1,-1,11] reflections fulfilled, I should see the 00-2 disk is dark or bright. But I don't know how to tell where is the place conditions fulfilled. I suppose when the reflections fulfilled, I can only see the disks from [1,-1,-9] to [1,-1,11] (or [1,-1,9] to [1,-1,-11] ). But it seems to me I can always see a lot of disks. Could you please give me some suggestions? Cheers.
Thanks for the question.! If your structure is FCC, then it has inversion symmetry and is also cubic; without going into a detailed explanation, this means that [1-10] and [110] are crystallographically equivalent, which means that either one would be valid, so there effectively isn't any difference; when I get around to filming the video for CBED, I will compare structures with and without inversion symmetry so you can see how this manifests in the patterns.
Thanks, Nicholas. I am currently trying to use F20 to get CBED patterns but I have troubles to see big disks. From my understanding, I should use smaller apertures and short camera length. Could you please give me some suggestions on the C1/C2 number and camera length? AND looking forward to your video about CBED.
Hey, I'm so sorry I never replied to your question. I did do a CBED video, so I hope you found it over on my channel. As far as disk size in CBED, you are usually trying to avoid overlap. The main key to avoiding disk overlap is to keep the condenser system in "microprobe" mode rather than "nanoprobe" mode. In microprobe mode with the beam at crossover the convergence semi-angle is about 5x smaller compared to nanoprobe mode. If there is still too much overlap, then you could also try reducing the size of the C2 aperture. As far as camera length, this is simply adjusted as needed; a higher camera length will give more magnification, but will reduce the field of view in the CBED pattern. Regarding C1 (spot size), I would opt for something on the high side (7 or higher) to reduce the probe size as well as the current; a lower current will make it easier to safety record the CBED pattern with your digital camera.
Hi Dr Ruudawski,
I recently started to watch your videos . Thanks for the padagogic approach gettinf down to concepts, and then applying it in practive. In this respect, and related to SAED patterns, would you please explain the concept of camera length, which is important for diffractions. Basically, my intutition says this goes back to the formation of the diff pattern but it will be nice to get the concept right. Many thanks
Beatriz
Hey, so very sorry for the very late response. Glad to hear my videos are helping you. Camera length is an archaic term that is used to describe the magnification of the diffraction pattern. It doesn't really apply physically in an S/TEM because the physical distance between the diffraction pattern and the camera is fixed, but we still use the term because of historical reasons. In S/TEMs, we can effectively change it by simply changing the strength of lenses in the projection system. It would be a good video to do though, going over the origins of "camera length", so thanks for the suggestion.
Thanks for this video. So when you make an exposure test, you start with 0.01 in camera view and 0.1 in acquire view and check the count in the acquired image? am I right?. you said that The max count for the acquired image is 650000 counts. is there a maximum count for live images? Another question is when you change the time from 0.1 to o,5 in the acquired camera is there an equation to calculate or expect the count before we acquire the image and test it because the count at 0.1 was between 6000 to 8000 and at 0,5 was between 25000 and 30000 counts. so what happened when we increased the value 5 times in the counts?. Thanks
Thanks for the questions. The max counts in both the live and acquired images is both 2^16 = 65,000 as the dynamic range is 16 bit; the difference between live and acquire modes here is the pixel resolution, with live mode having 512x512 resolution and acquire having 2048x2048 resolution (this is how I have these modes specifically set, but these could be set differently if desired, you could set both modes to be 2048x2048, if you wanted). As far as changing the exposure time and the effect on the resulting counts, the response should be very linear (assuming the same pixel resolution), so there is no special equation needed. In other words, if you multiply the acquire exposure time by 5x, you should get (approximately) 5x more counts; same thing would apply for live exposure. This is basically consistent with what you pointed out, though I agree that the math does not work out perfectly, but that is not necessarily unexpected for very high dynamic range situations (very large difference between darkest and brightest features in the image) like diffraction patterns. I hope this helps and thanks for checking out my video!
These videos are fantastic! Have you tried nano area diffraction on this instrument? Do you think it's feasible?
Hi Eli: glad you like the videos; yes, it is very feasible to perform nano-area diffraction on a Tecnai; please see this video I did covering CBED:
ruclips.net/video/DofabqzZcG8/видео.html
Hi Nicholas. I have found the CBED patterns. My question is: if I am looking at the zone axis of an fcc nanostructure and I wanna determine if it is along [1-10] or [110]. I already know if I tilt the sample to let the [1,-1,-9] and [1,-1,11] reflections fulfilled, I should see the 00-2 disk is dark or bright. But I don't know how to tell where is the place conditions fulfilled. I suppose when the reflections fulfilled, I can only see the disks from [1,-1,-9] to [1,-1,11] (or [1,-1,9] to [1,-1,-11] ). But it seems to me I can always see a lot of disks. Could you please give me some suggestions? Cheers.
Thanks for the question.! If your structure is FCC, then it has inversion symmetry and is also cubic; without going into a detailed explanation, this means that [1-10] and [110] are crystallographically equivalent, which means that either one would be valid, so there effectively isn't any difference; when I get around to filming the video for CBED, I will compare structures with and without inversion symmetry so you can see how this manifests in the patterns.
Hi Dr Ruudawski,
fantastic videos! Is there any video tutorial for dark-field TEM?
Thanks, Mingquan: please see here for tutorial video on dark-field TEM:
ruclips.net/video/uTA-OWV2Shc/видео.html
@@NicholasRudawski thanks for your reply and sharing
Thanks, Nicholas. I am currently trying to use F20 to get CBED patterns but I have troubles to see big disks. From my understanding, I should use smaller apertures and short camera length. Could you please give me some suggestions on the C1/C2 number and camera length? AND looking forward to your video about CBED.
Hey, I'm so sorry I never replied to your question. I did do a CBED video, so I hope you found it over on my channel. As far as disk size in CBED, you are usually trying to avoid overlap. The main key to avoiding disk overlap is to keep the condenser system in "microprobe" mode rather than "nanoprobe" mode. In microprobe mode with the beam at crossover the convergence semi-angle is about 5x smaller compared to nanoprobe mode. If there is still too much overlap, then you could also try reducing the size of the C2 aperture. As far as camera length, this is simply adjusted as needed; a higher camera length will give more magnification, but will reduce the field of view in the CBED pattern. Regarding C1 (spot size), I would opt for something on the high side (7 or higher) to reduce the probe size as well as the current; a lower current will make it easier to safety record the CBED pattern with your digital camera.
Thank you very much. Could you please explain how you determine acceptable exposure time by Histogram?
Great question! What I am going to do is film a separate video specifically covering use of the digital camera, so please be on the lookout for that.