Top Page > Browsing

How to apply periodic boundary conditions Date: 2019/01/17 18:23 Name: sanjeev I am new to openmx. I am trying to reproduse some papers on graphene published by "openmx". I want to know how to apply periodic boundary conditions in different direction, which keywords will control this. Now suppose in a paper published by Lou &Lee, (J. phys.chem, c, 2009,113, 21213). He stated that "we adopted a supercell geometry where the length of a vacuum region along the nonperiodic direction (y- direction) was 20 A, and the lattice constant along the periodic direction (x direction) wa 3.11 A. I can control the dimension of the supercell by the keywords like given below. <Atoms.UnitVectors 0.000 0.000 4.88 20.000 0.000 0.000 0.000 27.000 0.000 Atoms.UnitVectors> But how can we say that our calculations have periodicity in x direction. thanking in advance sanjeev

Page: [1]

Re: How to apply periodic boundary conditions ( No.1 ) Date: 2019/01/18 01:25 Name: Naoya Yamaguchi Hi, OpenMX always considers periodic boundary conditions along a-, b-, and c-axes except some methods, such as NEGF and ESM. For example, slab calculations are usually performed by inserting a vacuum region with an enough width to remove influence of image cells through "Atoms.UnitVectors" as you said. Regards, Naoya Yamaguchi Re: How to apply periodic boundary conditions ( No.2 ) Date: 2019/01/18 13:53 Name: sanjeev Dear Yamaguchi, Thank you very much for your reply. But still my doubt is not clear. In the above mentioned paper the author has mentioned that he made Y & Z directions nonperiodic. My question is that, How he made Y & Z nonperiodic and X as periodic. while, openmx consider all x,y,z directions as periodic. Now suppose I want to simulate graphene fragments with a supercell dimensions x= 10, y=50, z=50, for this I set. <Atoms.UnitVectors 10.000 0.000 0.000 0.000 50.000 0.000 0.000 0.000 50.000 Atoms.UnitVectors> it means I have created a vacuum in Y & Z directions, but how it mean that my calculations is periodic in x direction only. It is right if I say that my calculations are periodic in all x,y,z directions Thanking in advance. Sanjeev Re: How to apply periodic boundary conditions ( No.3 ) Date: 2019/01/19 02:01 Name: Naoya Yamaguchi Dear Sanjeev, >My question is that, How he made Y & Z nonperiodic and X as periodic. while, openmx consider all x,y,z >directions as periodic. Now I read the paper and I guess that they did band calculations using OpenMX 3.4 (or 3.5) and inserted a vacuum region, for example: Atoms.UnitVectors.Unit Ang <Atoms.UnitVectors 3.11 0.0 0.0 0.0 L 0.0 0.0 0.0 L Atoms.UnitVectors> , where L>20. >Now suppose I want to simulate graphene fragments with a supercell dimensions >x= 10, y=50, z=50, for this I set. I think it is right for calculations of one-dimensional materials such as nanoribbons. Regards, Naoya Yamaguchi Re: How to apply periodic boundary conditions ( No.4 ) Date: 2019/02/08 23:03 Name: Kylin  <Kylincaster@foxmail.com> Dear all, The periodicity in OpenMX calculation is automatically determined and straightforwardly shown in the screen log files as "The system is molecule|chain|slab|bulk" will be displayed as in the screen log file, representing 0D|1D|2D and 3D system respectively. So be careful to check periodicity in our calculation which was determined by PAOs cut-off (Rc) and dimensions of box sizes. It becomes non-periodic only with 2Rc > box_size. Moreover, the calculation of Uele was based on FFT which is only valid for 3D periodicity box which may induce some problems with the low dimensional material studies. So a large interlayer vacuum spacing is necessary or taking the effective screening medium (ESM) method Cheers, Kylin

Page: [1]