The Boric acid does not neutralize the HF per se, it rather complexes free F- ions to form
Fluoroboric acid (HBF4). The boric acid was initially added to prevent the formation of
insoluble metal fluorides, such as CaF2. Adding boric acid will result in a mixture of fluoroboric acid and
some hydrofluoric acid, both of which will still attack glass ware, and still require similar safety precautions.
In addition there will be a significant increase of Boron backgrounds in the ICP instrument. I've also seen that excess of Boric acid will
also precipitate in the spray-chamber and even torch.
Based on my experience I'd rather omit the use of boric acid and invest in HF-resistant sample introduction
components, such as teflon spray-chamber and nebuliser and torch with sapphire injector - maybe also Pt cones.
Even dilute HF will still slowly corrode the glass ware, I've noticed significant increases in Si backgrounds with as
little as 10ppm HF.
Or if you are considering silicate analyses, try fusing the silicate sample with an alkaline salt, such as NaOH etc.
The fusion product can then simply be dissolved in a weak HNO3, ready for ICP analyses with conventional
glass ware introduction.
Dr. R. Bastian Georg
Trent University Water Quality Centre
1600 West Bank Dr.
Peterborough, Ontario, Canada
office: +1.705.748.1011 ext: 7318
On 2011-09-15, at 3:32 AM, Ph. De Raeve wrote:
> Dear all,
> Very often, Silicon containing samples are digested in a micro-wave oven with acid mixtures containing HF.
> In order to allow injection in a ICP with quartz nebulizer and torch, remaining HF may be eliminated by addition of Boric acid.
> In most papers, a second micro-wave cycle at about 200°C is required to complete this HF "neutralization".
> Is high temperature necessary for this neutralization ?
> Some propose a simple mixing with an excess of a saturated Boric acid solution...
> Is there a way to evaluate the remaining HF after H3BO3 treatment ?
> Philippe De Raeve
> Quality Assistance SA
> Technoparc de thudinie, 1
> 6536 Donstiennes
> [log in to unmask]