From: Jacob Zabicky (zabicky$##$bgumail.bgu.ac.il)
Date: Wed Feb 16 2000 - 07:51:53 EST
Dear Jonas,
All the advice given below is very wise, and you should heed it. IMHO
carrying out a true analytical task may be too much for your or anybody's
abilities, unless one is a bona fide analyst dealing with organic elemental
analysis, using calibrated contraptions, in a dedicated lab. So, besides
whatever you do, prepare an authentic sample, and send it to a service
laboratory asking for CHNO etc., even if that costs a few crowns (its
cheaper than if you DIY). You may even ask them to dry the stuff again at
whatever temperature you like and warn them of hygroscopy or other
troublesome properties, the'll take care.
All the best,
Jacob
>Thankyou for your ideas. I will post them here below. However, some
>clarifying of my problem seem to be in place.
>
>1. Yes. I have analysed for oxygen, as an effort to isolate the problem.
>2. There is no known impurity that would make up for the large error. I've
>tried H2O/MeOH/CO2/CH3COOH. I have a program which checks for this.
>3. I do have the acetate salt to some extent. This however will not make up
>for the impurities. 16% of the content should be something else but C,H,N,O.
>
>4. I have eluated 700 ml of solvent from the HPLC-system, evaporated it and
>it gives no weight.
>5. The sample is combusted at 1800 degrees centigrade. This is not an
>ajustable parameter. Maybe the oxygen flux can be raised. I will check.
>
>Here is my first letter:
>_______________________________________
>Jonas Nilsson wrote:
>
>> Hello
>>
>> I'm having some trouble with my elemental analysis of some of my
>> compounds. The ones which gives me trouble are all aromatic amidines
>> Ar-C(=NH)-NH2. This is an example of how it works out:
>>
>> Theory:Given
>> C:65.3%:54.0%
>> H:6.4%:5.6%
>> N:14.2%:12.0%
>> O:14.2%:12.4%
>>
>> Total:100%:84.0%
>>
>> The compound is purified on HPLC (C18-column, MeOH/H2O/CH3COOH), there
>> should be no other elements than C,H,N and O. The isolated compound is
>> an acetate salt of the amidine. The NMR looks very pure. The sample is
>> dried overnight at 80 deg. centigrade, 5 mTorr (0,005 mm Hg) before
>> elemental analysis is run.
>>
>> The other functionalities in the compound are another aromatic system,
>> and a diamide:
>> Ar-NH-CO-C-C-CO-NH-CH2-Ar
>>
>> As you all can see all percentages are much to low.
>> Now, what is the 16% which is not C,H,N,O? And how can the C analysis be
>> 11% low?! I have 8 other samples all containing the amidine which gives
>> me similar results.
>>
>> Please advice. I cannot publish this until i get it right, and I know
>> the compounds are pure. (There are at least not 10-20% impurities).
>>
>
>_______________________________________
>_______________________________________
>_______________________________________
>Here are the replys:
>_______________________________________
>From: "Dr. Ulrich Jordis" <ujordis$##$pop.tuwien.ac.at>
>
>Did you really analyze for oxygen? Oxygen is not analyzed routinely but
>calculated as the difference of all others to 100%
>
>Furthermore you can try out my JASPER program to find solvents, impurities:
>see http://perdix.tuwien.ac.at/ANALYSE2.HTM.
>_______________________________________
>From: "Yuehui Zhou" <yuehuizhou$##$hotmail.com>
>
>Hi, the problem may lie in the acetic acid....amidine is very strong
>base...it may form salt with acetic acid....try to recalculate your result
>as salt....good luck!
>_______________________________________
>From: "Juan F. Miravet Celades" <miravet$##$qio.uji.es>
>
>Regarding the problem with your elemental analysis of amidines posted in
>orglist, I find that the found relations C/H, C/N and C/O are quite good.
>It seems that you have your compound and something that lowers all the
>percentages, probably inorganic. Could it be possible that some derivative
>of silica is coming from the column (if I am right C18 is silica-gel
>modified with C18-alkyl chains) (methanol-H20 mixture is able to solubilize
>to some extent silica-gel). You could try to take the same amount of
>eluent (that has passed through the column) that you have when you collect
>your sample and add a quantity (similar to that isolated of your stuff) of
>pure, reference compound and evaporate and process the sample as you use
>to. Check if it gives you wrong numbers again.
>
>_______________________________________
>From: Hormuzd Mulla <mulla$##$caribe.chem.uky.edu>
>
>Hi There
>I think the culprit is water or methanol. Try including integral
>numbers of water molecules and try calculating again. I'm sure you will
>come up with a requisite number of water/methanol molecules that might
>have been adsorbed on the surface of your compound. I had a similar
>problem with some pyridinium salts I'd made but we circumvented that when
>we considered water in the crystal. I am pretty sure you will iron out the
>kinks by doing this.
>_______________________________________
>From: Kotohiro Nomura <nomurak$##$ms.aist-nara.ac.jp>
>
>I 'm afraid that this might be the problem of combustion of your sample in
>the analyzer if your sample is pure enough. You may ask to raise the
>temperature of combustion tube or add oxygen more to complete the reaction.
>If not, take the analysis with smaller amount.
>
>I hope this would be fine for you.
>
>Kotohiro Nomura
>_______________________________________
>_______________________________________
>_______________________________________
>Thats all!
>/jN
>
>_____________________ _____________________
>| Jonas Nilsson | | |
>|Linkoping University | | Telephone |
>| IFM | | --------- |
>| Dept. of Chemistry | | work: +46-13-285690 |
>| 581 83 Linkoping | | fax: +46-13-281399 |
>| Sweden | | home: +46-13-130294 |
>|_____________________| |_____________________|
>
>
>__________________
>
>ORGLIST - Organic Chemistry Mailing List
>Website and Archive: http://www.orglist.net/
>List coordinator: Joao Aires de Sousa (jas$##$mail.fct.unl.pt)
__________________
ORGLIST - Organic Chemistry Mailing List
Website and Archive: http://www.orglist.net/
List coordinator: Joao Aires de Sousa (jas$##$mail.fct.unl.pt)
This archive was generated by hypermail 2.1.4 : Fri Sep 19 2003 - 12:15:34 EDT