I'm trying to do a large scale search to try and determine all experimentally observed crystal structures that have occurred in ternary phase diagrams from combinations of one of each HCP, BCC, FCC, inorganic metals. Currently I've been relying on the ICSD for these searches, however, in order to be exhaustive of even a fraction of this list it would theoretically take 10^19 individual search queries to do that.
Rather than try to (and dying long before completing) these 10^19 searches, all I would require is a one-line style query to get the same thing done. A search like [(FCC1 OR FCC2 OR FCC3...)AND(BCC1 OR BCC2 OR BCC3...)AND((HCP1 OR HCP2 OR HCP3...)AND(3 COMPOUNDS AND NOT (already known crystal structures))]
Thanks to the structuring of the database I haven't found a way of combining my queries with their limited logic operations (e.g.: chemical species limited to AND or NOTs with no ORs. and several of the other ways I've tried to get their website operating for this task have all been a bust.).
This is really frustrating since they even let you review the formatted query you submit but they don't let you edit it after the fact.
For Example:
CHEMISTRY ( pse: [unit of coefficients=Moles, restrictEl=false], composition: Li Co O, elCount: 3, )
Is there another database where I could submit my own queries instead of wasting time trying to hack the GUI apart to do a non-trivial query. I would not be opposed to having to download a large online repository and searching those, but if you suggest one could you also present me with a pointer on how to get started on searching that database.
Thank you very much.
Skyler,
Please try this: http://oqmd.org
It has a huge list of materials.
Mahesh
Try this as well:
http://rruff.geo.arizona.edu/AMS/amcsd.php
Cheers
Mahesh
Hi Skyler,
you can try this: http://database.iem.ac.ru/mincryst/index.php
here you can search both with material parameters and the elements in the material. Good luck!
Wouldn't a simple query with 3 elements+the relevant space group in their "basic" interface do the trick? Or do I miss what you are aiming at here.
ICSD in principle won't let you do anything very advanced, you are restricted to their interface (i.e. their imagination). Ordering it on a disk doesn't help, because the database itself is encrypted and you only get some awful program that is way worse than their web interface to work with. They are, however, cooperative if you are interested in doing weird things that they haven't thought of. I got permission to export the database to a bunch of CIF files (which was a pain) that I could then search and handle any way I like (although, obviously, not redistribute).
Another option might be COD:
http://www.crystallography.net/
which you can check out as a collection of CIF files from an SVN repository. It is open and somewhat larger than ICSD, but ICSD has a better coverage of inorganic structures.
I tend to work with these CIF collections directly, looping all files with python scripts, unless I can do what I like with some trivial query.
Hi Torbjorn,
I think there was a small miscommunication. Basically I'm trying to make a list of the structures that has been observed to have occurred experimentally from 1 element that when pure is an FCC species combined with 1 originally BCC AND 1 originally HCP species. There is not direct way to do this kind of sorting since the final chemical species can't be sorted for from the constraint (1 of each of the 3 pure structures). Instead I have a list of about 45 elements that is split into the 3 structures and then you would need to to search all possible combinations of 1 of each type.
Can you provide an example of your python script, I'm still a bit new to Python and seeing those scripts also might shed some light on how COD is organized.
This is a very simple task for the MaterialsProject (www.materialsproject.org), which can be asked somewhat complicated queries like yours by the MaterialsProject REST API (https://www.materialsproject.org/open, https://www.materialsproject.org/docs/api). You would need to install python, and then pymatgen, but these are very painless tasks and can be done quickly, and would further enable you to ask much more additionally complicated queries on the MaterialsProject. There is a lot of documentation, and you should be able to find what you need for your exact query.
It turns out all you have to do is put your family of chemical species you would like to do an OR search for in parenthesis for the ICSD. to get my search I did Chemical species "(List 1) (List 2) (List 3)" and Number of elements "3". I don't know if when more than 3 elements are selected
- Badges
- Science method
- Similar topics
- Methodology
- Crystallography