## 2D networks

Rudi
Synchrotron
Posts: 708
Joined: 12 Dec 2008, 15:56
Location: Saxony

### 2D networks

Hi,

is anyone familiar with the Wells notation of networks? The attached scan is from the textbook 'Structural Inorganic Chemistry'.

I know that in the middle in the AX case is the usual (4,4) square grid type. But what is the notation of the AX2 one at the right hand?
You do not have the required permissions to view the files attached to this post.

johnewarren
Synchrotron
Posts: 1841
Joined: 13 May 2006, 14:25
Location: UK

### Re: 2D networks

Hi Rudi

Do you mean how to convert the $\small AX_2$ in to an alternative notation?

Rudi
Synchrotron
Posts: 708
Joined: 12 Dec 2008, 15:56
Location: Saxony

### Re: 2D networks

Hi John,

I mean the Wells notation of nets, something like (4,4) as depicted in the middle. Or SiO2 has a (4,2) net for example. There should be also a Wells notation for the net depicted at the right hand?

johnewarren
Synchrotron
Posts: 1841
Joined: 13 May 2006, 14:25
Location: UK

### Re: 2D networks

Cheers Rudi

I'd missed the words Wells I think I need some new glasses.

oleg
Olex2 Boffin
Posts: 694
Joined: 24 Nov 2008, 21:34

### Re: 2D networks

Hi Rudi,

the Wells notation for the net on the right is the same - (4,4) or 44 or 4462 (???$4^4 or 4^46^2$???). The spacers are normally excluded from the notation,

Cheers,
o
Last edited by johnewarren on 15 Jan 2010, 16:33, edited 2 times in total.
Reason: Tex with <sup> = superscript added. Forum now supports native sup/sub with correct braces - changed <> braces []

Rudi
Synchrotron
Posts: 708
Joined: 12 Dec 2008, 15:56
Location: Saxony

### Re: 2D networks

oleg wrote: the Wells notation for the net on the right is the same - (4,4)
Really?
or 4<sup>4</sup> or 4<sup>4</sup>6<sup>2</sup>.
I guess this is Schläfli notation, isn't it?
The spacers are normally excluded from the notation,
And in the case of a (4,2) net in SiO2? Does the 2 not describe the oxygen spacers between the tetrahedral silicium?

oleg
Olex2 Boffin
Posts: 694
Joined: 24 Nov 2008, 21:34

### Re: 2D networks

Hmm, I though that quartz has 6<sup>4</sup>8<sup>2</sup> network (it is the Schläfli notation indeed).
Considering the symbol you provided - (4,2), I would think of 1D array, though in this case without the connectivity it would make either a 'ladder' of rectangles sharing edges for three connected node or a chain of rectangles sharing vertexes for 4-connected node...
May be I am confusing something regarding inorganic and organo-metallic frameworks... But if 2-connected atoms (nodes) were included into the topological symbols for organo-metallic frameworks they would become useless (because they would describe chemical connectivity, preventing comparing and classifying the topologies) and unmanageable.
But in the case you really wanted to include the 2-connected entities into the network description, the symbol would be
(8,2)<sub>2</sub>(8,4) - for the 2 and 4 connected nodes respectively

Rudi
Synchrotron
Posts: 708
Joined: 12 Dec 2008, 15:56
Location: Saxony

### Re: 2D networks

For certain reasons I'm dealing with terminology of networks again. So, I would like to bring this thread into the focus.
oleg wrote:Hmm, I though that quartz has 6<sup>4</sup>8<sup>2</sup> network (it is the Schläfli notation indeed).
Considering the symbol you provided - (4,2), I would think of 1D array, though in this case without the connectivity it would make either a 'ladder' of rectangles sharing edges for three connected node or a chain of rectangles sharing vertexes for 4-connected node...
Yes, you're right, Oleg. Two-connected points are not included, since they do not alter the basic system. Therefore, a Wells notation (4,2) should not exist. My fault.

In the attached figure (taken from Steed/Atwood - Supramolecular Chemistry), the corresponding Wells notation of the plane net with the Schläfli-Symbol (4.82) is (4,82), according to the figure caption. This is not clear to me.

In the (n,p) notation of Wells, n is the size of the smallest circuit (i. e. number of nodes in the shortest path) and p is the connectivity of the nodes.

For the mentioned net, the nodes are certainly not 8-connected but rather 3-connected. The size of the smallest circuit should be 4. Is it a mistake in the figure caption?
You do not have the required permissions to view the files attached to this post.

oleg
Olex2 Boffin
Posts: 694
Joined: 24 Nov 2008, 21:34

### Re: 2D networks

Hi Rudi,

yes it must be a typo... or misunderstanding , please refer to this recent paper to find out why :
CrystEngComm, 2010, 12, 44 - 48
This paper explains the difference between different notations and names/symbols and their missuses.
So for non-uniform nets the point symbols should be used (as by Wells) and that would be 4.82 (page 46 of that paper),

Cheers,
Oleg

Rudi
Synchrotron
Posts: 708
Joined: 12 Dec 2008, 15:56
Location: Saxony

### Re: 2D networks

Hi Oleg,
oleg wrote: yes it must be a typo... or misunderstanding , please refer to this recent paper to find out why :
CrystEngComm, 2010, 12, 44 - 48
This paper explains the difference between different notations and names/symbols and their missuses.
I know this recent paper by Blatov et al. but it's rather confusing in my opinion.
So for non-uniform nets the point symbols should be used (as by Wells) and that would be 4.82 (page 46 of that paper),
What is the typo now? Is 4.82 not a Schläfli but a point symbol? What about 63? Schläfli or point symbol? Blatov et al. stated that point symbol have often been misused as Schläfli symbols.

oleg
Olex2 Boffin
Posts: 694
Joined: 24 Nov 2008, 21:34

### Re: 2D networks

According to the paper the (p,q) notations can only be used for the uniform nets and for non-uniform nets a combination of point symbols should be used, so 4.82 is the Wells (point???) symbol for this non-regular net and (4,82) makes no sense... I do also find the paper a bit confusing but it makes the point - many people call same things differently and different things the same... For example Olex evaluates and prints short and long topological terms, which are as I suspect called point symbols and vertex symbols in this paper...

Rudi
Synchrotron
Posts: 708
Joined: 12 Dec 2008, 15:56
Location: Saxony

### Re: 2D networks

oleg wrote:According to the paper the (p,q) notations can only be used for the uniform nets and for non-uniform nets a combination of point symbols should be used, so 4.82 is the Wells (point???) symbol for this non-regular net
I guess uniform or regular net means that all nodes exhibit the same connectivity (uninodal) and all circuits have the same size. Correct?

According to Blatov's paper Schläfli symbols are those in braces, e.g. {p,q} for a 2D net in which q p's meet at each knot. For instances in the common honeycomb pattern, three hexagons meet at each knot and the Schläfli symbol is hence {6,3}. However, I always thought that Schläfli symbols are noted with the supscript (e. g. 63) according to the figure from the textbook. In Blatov's paper those symbols are called point symbols and the symbols in the textbook would be another example for misuse of the term Schläfli symbol? Quite confusing.
and (4,82) makes no sense...
Ok, maybe I should drop Jon Steed a short line. If you meet him in the hall, you may ask him, too.
many people call same things differently and different things the same...
This phenomenon seems to be quite universal in chemistry but this is another thing.
For example Olex evaluates and prints short and long topological terms, which are as I suspect called point symbols and vertex symbols in this paper...
Interesting. I was not aware that Olex2 is able to evaluate network topologies.

oleg
Olex2 Boffin
Posts: 694
Joined: 24 Nov 2008, 21:34

### Re: 2D networks

I guess uniform or regular net means that all nodes exhibit the same connectivity (uninodal) and all circuits have the same size. Correct?
exactly

In the stack order... Olex is not Olex2 and is actually developed for that particular case to get those topological terms (I must have been very clever those days not call them anything - but just topological terms ). And if anybody still thinks that Olex is a predecessor to Olex2 (http://en.wikipedia.org/wiki/Olex2), that is wrong... The only seams are the OpenGL and me... Also please take a look at this newsletter: http://iucrcomputing.ccp14.ac.uk/iucr-t ... index.html, where I do describe how those various things are calculated or rather enumerated.

Then whatever you call a misuse could be a simple misunderstanding... And most things are really confusing
Last edited by oleg on 18 Jan 2010, 10:31, edited 1 time in total.

Rudi
Synchrotron
Posts: 708
Joined: 12 Dec 2008, 15:56
Location: Saxony

### Re: 2D networks

oleg wrote: And if anybody still thinks that Olex is a predecessor to Olex2 (http://en.wikipedia.org/wiki/Olex2), that is wrong, wrong aim, wrong audience...
Oh, I see. But you have to admit that the name Olex likely implies that it's a predecessor to Olex2. Anyway, is Olex still an active project?
Then whatever you call a misuse could be a simple misunderstanding... And most things are really confusing
It's not grown on my dung. Blatov et al. speak of misuse in more than 300 RSC and ACS publications.

oleg
Olex2 Boffin
Posts: 694
Joined: 24 Nov 2008, 21:34

### Re: 2D networks

No Olex is not being actively developed, however if any bugs are found or assistance is required - I would be glad to help.