The authoritative public forum
  for Metal Finishing since 1989
-----

Analysis of aluminum in alum by XRF

2003

Dear Sir,

We are receiving a huge amount of alum for water purification to remove the turbidity. The time for analysis of aluminum takes long time for estimation by classical method by EDTA. We are trying to do the analysis of aluminum in alum by XRF method. Due to hygroscopic nature the alum we could not make pellet for analysis by XRF we are also trying to use cellulose power for dilution purpose but it gives the total aluminum content as aluminum sulphate as well as insoluble content of the raw material used for manufacturing purpose.

Please suggest,

Dr Balakrushna Padhi
NATIONAL ALUMINIUM COMPANY LIMITED - DAMANJODI, ORISSA, INDIA

Sounds like an application for a graphite furnace AA or an ICAP-MS analytical unit.

James Watts
- Navarre, Florida
2003

---

In what way is your alum likely to deviate from the ideal formula? I think the probable issue is the amount of water of hydration. If this is the case, you can give XRF a rest and get yourself a good pycnometer. Do a few accurate calibration determinations on an alum batch using the EDTA method, then use the same batch to make a concentrated solution of known apparent concentration and measure its density accurately. The data of density of alum solutions are available in the literature, and an easily measured 0.01% change in density corresponds roughly to a 1/4000 change in concentration. Try to get this accuracy with XRF!

Alternatively, again if water content is the issue, you might try a Karl Fisher titration or a variation on it (there are similar methods that do not use objectionable chemicals and are quicker than classical Karl Fisher).

Emanuel Cooper
IBM - Yorktown Heights, New York
2003

It may be possible to use XRF for this analysis. It sounds like you have a problem introducing the sample into the instrument. I would suggest the use of a XRF cup with a thin mylar film on the bottom. Selection of the proper film will be critical since it might absorb Al K-alpha lines. The test chamber will have to be flushed with helium to exclude the atmospheric absorbers of Al K-alpha lines. You will have to work some sort of standardization. I would suggest the old EDTA method. Once you get a good solid standardization curve built, you should be able to get a pretty good analysis. You could also dissolve the alum and try to look at the Al in the ionized form in the XRF. This might be more accurate that analyzing the powder.

Edward R. Durham
Cessna Aircraft - Columbus, GA, USA
2003