Building the adjacency matrix for the vab net. Producing subnets for the nbo net.
Home / FAQ / Building the adjacency matrix for the vab net. Producing subnets for the nbo net.
1. Open the database art_net. There is one entry named "Unknown distribution of points".
2. Open the window of the program AutoCN and set the following options: Method: Ranges; Dist.
Ranges: C-C 0:1.01 to store all carbon-carbon contacts of length 0–1.01 Å. In general, for the range you
need to use the format A-B x.x:y.y; several different bonds could be searched simultaneously. For all
other types of contacts, not mentioned in the Dist. Ranges window AutoCN will use the standard
distance criterion: a contact A–B is stored in adjacency matrix if its distance is less than the sum of
Slater's radii of the atoms A and B multiplied by (1+ Extra Dist.). If you want to ignore some A–B
contacts you have to type A-B 0:0.
3. Run AutoCN, then close the window.
4. Open the Crystal Data window and go to Adjacency matrix tab. You have got a 4-coordinated net (You
will learn later how to identify a net, this one is with the nbo topology.)
5. Expand the list of contacts of the carbon atom and right-click on the contact. Choose Mark Equivalent. If two contacts are selected, the two contacts are equivalent by symmetry.
Repeat the procedure for the next unselected contact until the message here below appears.
6. Right-click again and change the type of the contact to No bond.
To summarize the 4 bonds are grouped in three independent sets 1+1+2. Two are related by symmetry.
7. Save the changes.
8. Duplicate the record and open Adjacency matrix tab again. Restore the type for the broken contact to
Valence. Assign the type No bond to another contact. Save changes. You have got two 3-coordinated
nets. Call them e.g. "3-c A derived from 4-c" and "3-c B derived from 4-c", later you will be able to
identify the nets and see that one is single vab and the other is an array of two interpenetrated nets with
the bmn topology. Pay attention that you cannot get the two derived 3-c nets by an analysis of
interatomic distances: the nearest neighbor are always 4 at the same distances of 1.00Å
With this last Task, you found a possible relation between 3-c net vab or 2-fold interpenetrated
bmn nets to 4-c nbo. This may be related to some transformations in solid-state. Keep the database to be
Could you derive other subnets form the original 4-c one? 2D or 1D (rod packings)? Is the 4-c
net found a sphere packing? Are the derived 3-c nets also sphere packings?