Saturday, October 13, 2012

Running my first ion exchange column

So, we finally achieved complete dissolution and homogenization of my samples after 2 weeks in high-temperature, high-pressure bombs.  Now, the actual chemistry to separate the elements can begin.  Recall the scheme I showed earlier:

In this post, I'll describe a bit about running the first column (Column "0", named so because it was added into an older scheme which didn't originally deal with Mg-rich samples like pyroxenites and komatiites - my type of sample).  This column separates Hf along with the other HFSE (high field-strength elements: Zr, Ti, Nb, Ta) from the REE (rare earth elements, including Lu, Sm, and Nd).  Note that if your sample is not Mg-rich (defined here as >15% MgO), you wouldn't have to go through with this step and could simply add conc. HF to your attacked sample residue and heat in a closed beaker.  In this case, REE would be bound up into insoluble fluorides (which are centrifuged out), whereas Hf (and other HFSE) would remain in solution.  

First, a very introductory bit about ion-exchange chromatography.  Chromatography is a separation method that utilizes a stationary phase (the ion-exchange resin) and a mobile phase (an aqueous solution containing the sample).  Different ions are separated from each other based on their varying affinities for the solid phase, in this case an ion-exchange resin.  The strength of this affinity dictates the strength (normality) of the acid needed to elute that particular ion.  There are cation-exchange and anion-exchange resins; a cation exchange resin takes up cations from the sample solution and in turn releases an equal amount of resin cation into solution, vice versa for anion exchange.  A cation (or anion) exchange column must be calibrated (I didn't do that) before samples are run on it.  Calibrating involves collecting and analyzing (on an ICPMS) each successive eluate (what drips out of the column), so that one knows when an element of interest is eluted (washed through the column) during a particular analytical scheme.

Shown below are the steps involved in running Column "0".  For simplicity I omit some analytical details like eluent volume, resin bed volume, etc. The point is to show overall what goes on:
 An interesting thing about this column is that the resin bed contracts with increasing acid normality (we start with 1 M HCl and end with 6 M).  This has to do with the chemistry of the resin and different solvents.  A resin is a hydrocarbon skeleton which has functional groups added to it to give it particular ion exchange properties.  The functional group is composed of a fixed functional group and mobile counter ion, which reacts with the ions in the sample solution.  Water, because it is polar, is attracted to the charged functional groups in the resin.  Thus, when you make a resin from initially dry resin beads and mix with water, it is swelled up to the max because all the functional groups are hydrated.  Thus, as you sequentially elute stronger and stronger acids through the resin, the resin will shrink.  Because of this, if you were to use the resin again, you have to backwash it with water so that it can swell up to its original size.

Here is a photo of Step 3, when the sample is going through and we are collecting (in nice clean teflon beakers) Hf, Zr, and Ti.  The darker band in the middle of the resin is Fe from the sample going through.  The reason it appears "smeared" out throughout the column is because of turbulence within the column (due to the large diameter of this particular column).

Here is a photo of Step 4.  Here, we've finished collecting our elements of interest (Hf, Zr, Ti), and so we now elute to get rid of all the other elements except REE.  This includes Fe, which as you can see has made it all the way through the column (no more dark band).  It's now in the trash beaker below.  Fe in solution with HCl has a Gatorade-like yellow color.

Next time, I'll show you how Column "1" goes! :-)

No comments: