From owner-chemistry@ccl.net Fri Dec 24 23:27:00 2010 From: "Stephen Bowlus chezbowlus#,#comcast.net" To: CCL Subject: CCL: calculation of logP using atom contribution Message-Id: <-43455-101224162546-10700-SrDyitp2sii3Q6yep9sacA*_*server.ccl.net> X-Original-From: Stephen Bowlus Content-Type: multipart/alternative; boundary=Apple-Mail-43--761367522 Date: Fri, 24 Dec 2010 13:25:28 -0800 Mime-Version: 1.0 (Apple Message framework v936) Sent to CCL by: Stephen Bowlus [chezbowlus##comcast.net] --Apple-Mail-43--761367522 Content-Type: text/plain; charset=WINDOWS-1252; format=flowed; delsp=yes Content-Transfer-Encoding: quoted-printable "Formal charge" and "oxidation state" are different animals. The =20 oxidation state of nitrogen in an ammonium ion is -3; its formal =20 charge is +1. I don't see how it is possible for a carbon bearing only 1 hydrogen =20 (all other connections to carbon) to have an oxidation number =3D 0; but = =20 it is very likely that the carbon would have formal charge =3D 0. In =20 quinidine, all the carbons have a formal charge of 0. But their =20 oxidation states vary: for example: -2 (sp3 C in methyl group in =20 methoxy - -3 to balance 3 H and +1 to balance 1/2 an oxygen); +1 =20 (aromatic "sp2" C with methoxy substituent); 0 (!, sp3 bridge bearing =20= hydroxyl group - 0 for carbon links, -1 to balance H and +1 to balance =20= 1/2 O); and -1 (aromatic "sp2" C bearing hydrogens); -2 (sp3 =20 methylenes in quinuclidine rings). So do you need oxidation states or formal charges? It's been too long =20= for me to remember any details of the Ghosh and Crippen model. -sb On Dec 23, 2010, at 9:07 AM, Loan Huynh huynhkl2000 : yahoo.ca wrote: > Hi Steve, > > > Thank you very much for your help. As you suggest, I did try to =20 > calculate the oxidation number base on general rule = (http://en.wikipedia.org/wiki/Oxidation_state=20 > ). > > > However, when I apply general rule for calculating formal charge of =20= > carbon, I can=92t get the formal charge reported in Grosh and Crippen =20= > ( J. Computational Chemistry, 1988, 9, p.80-90) paper. > > > According to Mannhold (J. Computer-Aided Molecular Design, 2001, 15, =20= > 337), Quinidine has 1 H that attached to Csp3 with formal charge of =20= > 0. It seem like the general rule is not applicable to this case. > > > Then I try another method for calculating formal oxidation number. =20 > According to Viswanadhan (J. Chem. Inf Comput. Sci. 1989, 29, 163), =20= > =93the formal oxidation number of a carbon atom =3D sum of formal bond = =20 > orders with electronegative atoms=94, so H that attached to Csp3 with =20= > carbon has formal charge of 3. However, Mannhold didn't report any =20 > functional group that have 1H attached to Csp3 with carbon has =20 > formal charge of 3. > > > Any other suggestion is greatly appreciate. > > > Loan > > > --- On Wed, 12/22/10, Stephen Bowlus chezbowlus.:.comcast.net chemistry^ccl.net> wrote: > > From: Stephen Bowlus chezbowlus.:.comcast.net chemistry^ccl.net> > Subject: CCL: calculation of logP using atom contribution > To: "Huynh, Loan " > Received: Wednesday, December 22, 2010, 7:56 PM > > What seems to work is the usual rules, supplemented by the rule: =20 > bonds to the same element don't count. In the usual formulation of =20 > the rules, this addition is a generalization of "elements have an =20 > oxidation state of 0" and is why the "exception" of oxygen =3D -1 in =20= > peroxides works. The only place this is addressed, as far as I =20 > know, is in general chemistry texts. The expanded rule can be =20 > justified on the basis that the electrons in a bond between the same =20= > elements are shared equally. > > One does get some wierdnesses depending on how the carbon is =20 > substituted, but the _change_ in oxidation states in the course of =20 > reaction seems reasonable. So the concept is successful as a =20 > bookkeeping method. One has to remember that it is the change in =20 > most cases that is physically relevant; otherwise, oxidation state =20 > is calculated on a completely non-physical basis. > > There is actually a nice Wikipedia article "Oxidation State" that =20 > shows situations as carbon's oxidation state varies from +4 (carbon =20= > tetrachloride) to -4 (methane). For an acetylene that you describe, =20= > I would calculate the oxidation state as -1. C-1 of propyne would =20 > be -1, C-2 would be 0 and C-3 would be -3. The four H's (a +1 each) =20= > make the molecule neutral. > > Steve > > On Dec 22, 2010, at 12:31 PM, Loan Huynh huynhkl2000 _ yahoo.ca wrote: > >> Dear CCL, >> >> I am currently calculating the logP values using the atom =20 >> contribution by Grosh and Crippen ( J. Computational Chemistry, =20 >> 1988, 9, p.80-90). >> >> >> I have done quite a lot of searching on calculating the oxidation =20 >> state of carbon. However, I have trouble calculating the formal =20 >> oxidation number of various hybridization carbons. For example, =20 >> for H attached to sp carbon, I cannot obtain the formal charge of 3 =20= >> for sp carbon. >> >> >> Is there any document that shows the calculation of oxidation =20 >> number for sp carbon? >> >> >> Thank you very much for your help, >> >> >> Loan >> >> >> > > --Apple-Mail-43--761367522 Content-Type: text/html; charset=WINDOWS-1252 Content-Transfer-Encoding: quoted-printable "Formal charge" and "oxidation = state" are different animals.  The oxidation state of nitrogen in = an ammonium ion is -3; its formal charge is +1.

I = don't see how it is possible for a carbon bearing only 1 hydrogen (all = other connections to carbon) to have an oxidation number =3D 0; but it = is very likely that the carbon would have formal charge =3D 0. In = quinidine, all the carbons have a formal charge of 0.  But their = oxidation states vary: for example: -2 (sp3 C in methyl group in methoxy =  - -3 to balance 3 H and +1 to balance 1/2 an oxygen); +1 (aromatic = "sp2" C with methoxy substituent); 0 (!, sp3 bridge bearing hydroxyl = group - 0 for carbon links, -1 to balance H and +1 to balance 1/2 O); = and -1 (aromatic "sp2" C bearing hydrogens); -2 (sp3 methylenes in = quinuclidine rings).

So do you need oxidation = states or formal charges?  It's been too long for me to remember = any details of the Ghosh and Crippen = model.

-sb

On Dec = 23, 2010, at 9:07 AM, Loan Huynh huynhkl2000 : yahoo.ca wrote:

Hi Steve,

 

Thank = you very much for your help. As you suggest, I did try to calculate the = oxidation number base on general rule (http://en.wikipedia.= org/wiki/Oxidation_state).

 

However, = when I apply general rule for calculating formal charge of carbon, I = can=92t get the formal charge reported in Grosh and Crippen ( J. = Computational Chemistry, 1988, 9, p.80-90) paper.

 

According = to Mannhold (J. Computer-Aided Molecular Design, 2001, 15, 337), = Quinidine has 1 H that attached to Csp3 with formal charge of 0. It seem = like the general rule is not applicable to this case.  

 

Then I = try another method for calculating formal oxidation number. According to = Viswanadhan (J. Chem. Inf Comput. Sci. 1989, 29, 163), =93the formal = oxidation number of a carbon atom =3D sum of formal bond orders with = electronegative atoms=94, so H that attached to Csp3 with carbon has = formal charge of 3. However, Mannhold didn't report any functional group = that have 1H attached to Csp3 with carbon has formal charge of 3. =

 

Any other suggestion is greatly appreciate. =

 

Loan

 
--- On Wed, 12/22/10, = Stephen Bowlus chezbowlus.:.comcast.net = <owner-chemistry^ccl.net> wrote:

From: Stephen Bowlus chezbowlus.:.comcast.net = <owner-chemistry^ccl.net>
Subject: CCL: calculation of logP = using atom contribution
To: "Huynh, Loan " = <huynhkl2000^yahoo.ca>
Received: Wednesday, December 22, 2010, = 7:56 PM

What seems to work is the usual = rules, supplemented by the rule: bonds to the same element don't count. = In the usual formulation of the rules, this addition is a generalization = of "elements have an oxidation state of 0" and is why the "exception" of = oxygen =3D -1 in peroxides works.  The only place this is = addressed, as far as I know, is in general chemistry texts. The expanded = rule can be justified on the basis that the electrons in a bond between = the same elements are shared equally.

One does get = some wierdnesses depending on how the carbon is substituted, but the = _change_ in oxidation states in the course of reaction seems reasonable. = So the concept is successful as a bookkeeping method. One has to = remember that it is the change in most cases that is physically = relevant; otherwise, oxidation state is calculated on a completely = non-physical basis.

There is actually a nice = Wikipedia article "Oxidation State" that shows situations as carbon's = oxidation state varies from +4 (carbon tetrachloride) to -4 (methane). =  For an acetylene that you describe, I would calculate the = oxidation state as -1.  C-1 of propyne would be -1, C-2 would be 0 = and C-3 would be -3. The four H's (a +1 each) make the molecule = neutral.

Steve

On Dec = 22, 2010, at 12:31 PM, Loan Huynh huynhkl2000 _ yahoo.ca wrote:

Dear CCL,

I = am currently calculating the logP values using the atom contribution by = Grosh and Crippen ( J. Computational Chemistry, 1988, 9, p.80-90). =

 
I have done quite a lot of searching on = calculating the oxidation state of carbon. However, I have trouble = calculating the formal oxidation number of various hybridization = carbons.  For example, for H attached to sp = carbon, I cannot obtain the formal charge of 3 for sp carbon. 

 

Is there any document that shows the = calculation of oxidation number for sp carbon?  =

 

Thank you very much for your = help,

 

Loan





= --Apple-Mail-43--761367522--