Someone asked me a question. But I could not find the question in the comment area. So I'd like to copy and paste his/her question here and respond. Hopefully he/she will see my response. Here is the question: Thanks for the video. I have a question : Does Olex2 refine charged atoms ? I have a structure where a nitrogen atom should have a charge of +1 but when I check it with a program like Mercury it appears without charge. Is there any method to put the charge in Olex2 or is it automatic and my refinement is wrong ? Thanks for your answer. My answer: Olex2 is a GUI for the "brain" program "SHELXL". When SHELXL is called by Olex2 to do the refinement, it only refines on the electron densities per se. According to the amount of electron densities, it attempts to assign the atom types. For example, in the structural solving stage ( by SHELXT for example) , if the program find at a certain location there is about 6 electrons (after proper scaling), then it will tentatively assign it as carbon atom. During the refinement stage, the electron amount could vary, then you can adjust its atom type. More often, you assign the atom types based on your prior knowledge of the crystal. Only high quality data can differentiate between atoms with similar atomic weight (such as C/N/O). In other word, program won't be able to tell if the atom is C or N solely based on the electron density if the data quality is not high enough. In your particular case, neutral nitrogen atom has 7 electrons. If it bears +1 charge, it will have the same amount of electrons as carbon. So the program potentially could miss-assign it as carbon. However, you should keep it as nitrogen. On the other hand, its charge is reflected by its binding environment. For example, neutral nitrogen will have 3 bond. If it has 4 bonds, then it bear +1 charge. You can adjust the amount of Hydrogen atoms on it to show its correct valence. Hope this helps.
A quick question, when I try to refine the res file by shelxl, it always shows "can't open lst file" and the file can't be refined. I can't find a lst file from my data folder, how should I do?
Very helpful video Frank. I have an issue though where after I click solve, no structure is generated. I'm not sure why this might be, would you have any ideas?
It can be one of the following reasons: wrong space group, unaccounted twinning, very low resolution data, etc. There is no easy answer for this question. You will have to examine the data set carefully to eliminate the above possibilities one by one. Good luck.
If the bond is not too much out of the normal range, you can use a restraint called DFIX to make it behave better. However, if it is very much shorter, then you need to investigate the real cause of it. It could indicate some thing more serious, such as wrong atom type, even twinning sometime.
very usefull video! could you show us how to model a structure whit high disorder? I have to solve a structure in OLEX but the results table shows a lot of problems
I am glad to hear that. Do you mind sharing your problematic data set with me? You can send your data to afrankproduction@gmail.com. I will see what I can offer.
@@valentinaberlin5581 I took a look at your data. I believe the problem is due to unaccounted twinning. If you click the "%" icon right after the R1 value to the top right corner of the Olex2 window, you will see that the Fobs are systematically larger than Fcal. I.e., the data are all above the diagonal. Another way to see this quantitatively is, click Info, click "Bad reflections", you will see all the errors are of negative values. That again means Fcalc are systematically lower than Fobs. This strongly suggest there is/are other unaccounted twin component(s) contributing to the Fobs. To confirm this, you can go back to visit the original data, examine carefully in the reciprocal lattice viewer to see if you see twinning there. If you see twinning there, there are two ways to proceed. One is to try to process the twin data properly (identify the twin domains & twin law, perform proper reduction and scaling). The other one, which is what I recommend, is to find another single crystal and recollect the data. Summary, when everything apparently look normal, but somehow the R1 is high, it is time to check twinning. One of the strong indication is the most disagreeable reflections between Fobs and Fcalc. And when twin happens, the first thing should be to find another single crystal specimen. Only after you tried and could not find a better/single crystal, you try to work with the twin data. Let me know if you will find twinning in reciprocal lattice viewer. Good luck.
Thanks. I will make more soon. As for reducing R1: 1) make sure you have a good data set, high resolution, low Rint, high redundancy. 2) make sure the space group is correct. 3) make sure your model is correct. No missing atoms, incorrectly assigned atoms. Model the disorder correctly and sufficiently. 4) Make sure you account for any twinning if exist. For beginners, you can take advantage of checkcif report. It tends to give very good feedback about your data, your model, etc. Be careful though as checkcif sometimes give "inaccurate/false" alarms. It is meant to bring your attentions to those details. You make the final decision based on the suggestions.
@@frankknowitall6430 Dear sir, Thanking you for a valuable suggestion. can you suggest me, You have any other platform, I am highly interested to learn a regular way of crystallography with you. Please help me. My email id- anujqt@gmail.com
@Frank Know It All , I have started seeing your tutorial and found it very usuful. I have too a dataset which have some positional disorder in between 5 membered (thiophene) and 6 membered (fluorobenzene). These two different ring are displaying somehow both the character. May I know how to tackle such dataset. I heard about the commands SUMP, SADI, SIMU and RIGU but dont know actually how to apply these commands over this diorder. I have also the corrected dataset too of the same file using these commands received from someone else. It would be really great to me if you make a video on that dataset and for other researchers too.
Platon is a third party independent program which Olex2 can recall from within as a plug-in. To do so you need to take 2 steps. One, install the platon. Please google "Platon" for its downloading and installation. Second, after installation, open Olex2, click "Home", "settings", find the line "PATH", add the installation path to that line. Exit Olex2. Open Olex2 again. Now the Platon icon will be shown on top of the tool lane. Good luck.
Yes you can. The "hadd" add the hydrogens "theoretically", but does not consider the Q peaks. If you have Q peaks at ideal positions for Hydrogens, you may assign them to Hydrogens directly. If you do not see (all) the Q peaks for your expected hydrogens, you may use "hadd".
I am a beginner and inspired by your video. I need a help to solve a problem . Space group is P21/m and it is a pure aromatic compound ( C, H, N, F). I can send you the file.
Hi, I downloaded the sample data but when I tried to open it, it says that it's something wrong with so I cant access to the data to practice. Could you help me?
I guess it is due to the compression format. Try to use a different way to unzip it. If still does not work, just try to get a different data set. For example, you can download from CCDC. Any data set will do. Good luck
@@frankknowitall6430 Thank you. I would like to ask you if you have videos explaining how to model disorder of an aromatic compounds? or something that could help me with that. I looked for in your videos but I didnt find. Thanks!
@@stephannyvillamizar4637 I don't think my current videos have that. It is one of my future plan. If you have a data set you want me to demo on, please share with me the data set.
@@santamaria1363 Yes. Pretty much. It has a predefined range for detecting these interaction. If you need to look further beyond the "normal" range, you need to manually change some parameters. But usually the defaults are good.
thank you Frank for making your knowledge free. do upload more videos in future
Someone asked me a question. But I could not find the question in the comment area. So I'd like to copy and paste his/her question here and respond. Hopefully he/she will see my response. Here is the question:
Thanks for the video. I have a question : Does Olex2 refine charged atoms ? I have a structure where a nitrogen atom should have a charge of +1 but when I check it with a program like Mercury it appears without charge. Is there any method to put the charge in Olex2 or is it automatic and my refinement is wrong ? Thanks for your answer.
My answer:
Olex2 is a GUI for the "brain" program "SHELXL". When SHELXL is called by Olex2 to do the refinement, it only refines on the electron densities per se. According to the amount of electron densities, it attempts to assign the atom types. For example, in the structural solving stage ( by SHELXT for example) , if the program find at a certain location there is about 6 electrons (after proper scaling), then it will tentatively assign it as carbon atom. During the refinement stage, the electron amount could vary, then you can adjust its atom type. More often, you assign the atom types based on your prior knowledge of the crystal. Only high quality data can differentiate between atoms with similar atomic weight (such as C/N/O). In other word, program won't be able to tell if the atom is C or N solely based on the electron density if the data quality is not high enough.
In your particular case, neutral nitrogen atom has 7 electrons. If it bears +1 charge, it will have the same amount of electrons as carbon. So the program potentially could miss-assign it as carbon. However, you should keep it as nitrogen. On the other hand, its charge is reflected by its binding environment. For example, neutral nitrogen will have 3 bond. If it has 4 bonds, then it bear +1 charge. You can adjust the amount of Hydrogen atoms on it to show its correct valence. Hope this helps.
Thank you. I am waiting for your future video!!
A quick question, when I try to refine the res file by shelxl, it always shows "can't open lst file" and the file can't be refined. I can't find a lst file from my data folder, how should I do?
Very helpful video Frank. I have an issue though where after I click solve, no structure is generated. I'm not sure why this might be, would you have any ideas?
It can be one of the following reasons: wrong space group, unaccounted twinning, very low resolution data, etc. There is no easy answer for this question. You will have to examine the data set carefully to eliminate the above possibilities one by one. Good luck.
Great video! Could you show how to manipulate bond lengths? I get an error in CheckCIF indicating one of my bonds is too short.
If the bond is not too much out of the normal range, you can use a restraint called DFIX to make it behave better. However, if it is very much shorter, then you need to investigate the real cause of it. It could indicate some thing more serious, such as wrong atom type, even twinning sometime.
Great video. Thank you very much.
Thank you sir, this video is helpful for me.
Amazing! Thank you!
very usefull video! could you show us how to model a structure whit high disorder? I have to solve a structure in OLEX but the results table shows a lot of problems
I am glad to hear that. Do you mind sharing your problematic data set with me? You can send your data to afrankproduction@gmail.com. I will see what I can offer.
@@valentinaberlin5581 I took a look at your data. I believe the problem is due to unaccounted twinning. If you click the "%" icon right after the R1 value to the top right corner of the Olex2 window, you will see that the Fobs are systematically larger than Fcal. I.e., the data are all above the diagonal. Another way to see this quantitatively is, click Info, click "Bad reflections", you will see all the errors are of negative values. That again means Fcalc are systematically lower than Fobs. This strongly suggest there is/are other unaccounted twin component(s) contributing to the Fobs. To confirm this, you can go back to visit the original data, examine carefully in the reciprocal lattice viewer to see if you see twinning there. If you see twinning there, there are two ways to proceed. One is to try to process the twin data properly (identify the twin domains & twin law, perform proper reduction and scaling). The other one, which is what I recommend, is to find another single crystal and recollect the data. Summary, when everything apparently look normal, but somehow the R1 is high, it is time to check twinning. One of the strong indication is the most disagreeable reflections between Fobs and Fcalc. And when twin happens, the first thing should be to find another single crystal specimen. Only after you tried and could not find a better/single crystal, you try to work with the twin data. Let me know if you will find twinning in reciprocal lattice viewer. Good luck.
Waiting for videos oriented to how to make cif file publishable, tips on olex2 and using of platron to start with.
Great video. please where do I get torsion angles from olex2?
You can click on "Report", Olex will generate a report containing tables of all kinds of values.
@@frankknowitall6430 Thank you
thank you, very helpful video.
very helpful video. thank you frank
视频讲解很细致，太感谢了！
oh what a video, thanks you very much
Dear sir, it's a great video for students. can you upload some more videos for students? How to reduce R factor.?
Thanks. I will make more soon. As for reducing R1: 1) make sure you have a good data set, high resolution, low Rint, high redundancy. 2) make sure the space group is correct. 3) make sure your model is correct. No missing atoms, incorrectly assigned atoms. Model the disorder correctly and sufficiently. 4) Make sure you account for any twinning if exist.
For beginners, you can take advantage of checkcif report. It tends to give very good feedback about your data, your model, etc. Be careful though as checkcif sometimes give "inaccurate/false" alarms. It is meant to bring your attentions to those details. You make the final decision based on the suggestions.
@@frankknowitall6430 Dear sir, Thanking you for a valuable suggestion. can you suggest me, You have any other platform, I am highly interested to learn a regular way of crystallography with you. Please help me.
My email id- anujqt@gmail.com
@@ANUJKUMAR-db8fl Unfortunately I don't have other platforms. You can find some textbook to read. Even some literatures will be beneficial. Good luck.
@Frank Know It All , I have started seeing your tutorial and found it very usuful. I have too a dataset which have some positional disorder in between 5 membered (thiophene) and 6 membered (fluorobenzene). These two different ring are displaying somehow both the character. May I know how to tackle such dataset. I heard about the commands SUMP, SADI, SIMU and RIGU but dont know actually how to apply these commands over this diorder. I have also the corrected dataset too of the same file using these commands received from someone else. It would be really great to me if you make a video on that dataset and for other researchers too.
very useful video for me
Thanks for the video! Could you show us other example with disorder please?
Hi, sure I can do that. However, I do not have a good data set to do so. Do you happen to have one data set I can use in the demo?
@@frankknowitall6430 sure! actually I have problems with the disorder in one ring of my recent compounds. I can send it to you
Ok. You can send it to me via afrankproduction@gmail.com. I will take a look for you. Btw, you are ok for me to use your data in the demo?
@@frankknowitall6430 I don´t have any problem. Thank you I'll send you the data.
@@frankknowitall6430 I shared you a folder via Dropbox
Thank you so much, very informative and extremely useful!
I have to lern lot thanks for creation. I try to lern it require the help.
In your olex installation video.. platon was not there. Bt in this video hw did platon come?
Platon is a third party independent program which Olex2 can recall from within as a plug-in. To do so you need to take 2 steps. One, install the platon. Please google "Platon" for its downloading and installation. Second, after installation, open Olex2, click "Home", "settings", find the line "PATH", add the installation path to that line. Exit Olex2. Open Olex2 again. Now the Platon icon will be shown on top of the tool lane. Good luck.
Can you assign those Q peaks to H atoms? What's the difference with 'hadd'?
Yes you can. The "hadd" add the hydrogens "theoretically", but does not consider the Q peaks. If you have Q peaks at ideal positions for Hydrogens, you may assign them to Hydrogens directly. If you do not see (all) the Q peaks for your expected hydrogens, you may use "hadd".
What is the command for splitting a atom in Olex2?
"Split" will split the selected atom(s) along the longest axis.
I am a beginner and inspired by your video. I need a help to solve a problem . Space group is P21/m and it is a pure aromatic compound ( C, H, N, F). I can send you the file.
Ok. You may send the data to me via afrankproduction@gmail.com.
How to vet data completeness to 100%? I get only 92%
Hi, I downloaded the sample data but when I tried to open it, it says that it's something wrong with so I cant access to the data to practice. Could you help me?
I guess it is due to the compression format. Try to use a different way to unzip it. If still does not work, just try to get a different data set. For example, you can download from CCDC. Any data set will do. Good luck
@@frankknowitall6430 Thank you. I would like to ask you if you have videos explaining how to model disorder of an aromatic compounds? or something that could help me with that. I looked for in your videos but I didnt find. Thanks!
@@stephannyvillamizar4637 I don't think my current videos have that. It is one of my future plan. If you have a data set you want me to demo on, please share with me the data set.
@@frankknowitall6430 Ok, how can I share the data to you? Thank you.
@@stephannyvillamizar4637 Send it to afrankproduction@gmail.com. Thanks.
great
Hello Sir, am here again. Please how do i use Olex2 to determine pi-pi interaction in a crystal structure?
Just type pipi, it will give you a list of the distances.
@@frankknowitall6430 Thank you, so when it says no interaction. Does it mean no pipi interaction?
@@santamaria1363 Yes. Pretty much. It has a predefined range for detecting these interaction. If you need to look further beyond the "normal" range, you need to manually change some parameters. But usually the defaults are good.
Thank you so much am grateful