Inorganic Anion using Indirect UV LC

[oscarBAL]

Hi All; I will have to implement a metohd for analysis of Anion (Cl-, NO3, PO4, Br, SO4) by LC/UV.
I have done NO2/NO3 in the past by RP-HPLC UV using IP with C18 columns and worked great for me.
Now my problem is little bit dirent.

I have to be able of quantitate until 0.1 ppm; I was reading and Indirect UV Phtalic Acid or 4-amino-2 hidroxibenzoic acid is very comun methodology for these kind of separations, but I am concern about sencitivity .
Someone knows if using any of these mobil phases might I reach thos deteccion levels I need?

I would like to use C18 again but this time with 3 or 2.5um columns in order to get faster resultus; I won't expect any mayor isues with an Ion Pair machanism; but it would be good to read the opinion from some one with previos Experience or knowledge.

Thanks in advance and best regards.

Oscar

[Markus Laeubli, Metrohm]

To run anions with indirect UV Iwould not recommend to use an RP column. There would be no separation on a bare RP column. Ion pairing on such a column could work. That is not really used.

I would use any of the existing IC columns for non-suppressed IC. Metrosep Dual 3 or Super-Sep Anion would be Metrohm column examples. But there are many more also from other suppliers.

Phthalic acid eluent (around pH=4) is quite often used. But here phophate elutes very early in the chromatogram (as H2PO4-) and may interfere with e.g. fluoride. In alkaline eluents (around pH= 8-9) phosphate is eluted between nitrate and sulfate.

A major problem with indirect UV is that only very little light is reaching the detector. Therefore it has to be extremely stable and low in electronic noise.
Anyway under suitable conditions 0.1 ppm of the respective ions could be reached.

[oscarBAL]

thanks Markus for your replay; is very clear fo me.

BR

Oscar

[carls]

Markus Laeubli -
What buffer/indirect UV agent do you recommend for alkaline pH? Phthalic acid adjusted to pH 9 with ammonium hydroxide?

oscarBAL-
Due to the low %transmittance mentioned by Markus Laeubli indirect UV has limited "linear" range. Getting the concentration of the indirect UV agent right is a bit tricky but should be optimized for best results.

[Vlad Orlovsky]

with indirect detection you will also experience poor baseline and unfavorable S/N ratio.

[Markus Laeubli, Metrohm]

Dear carls

I would not use phthalic acid at pH = 9 as there is no buffer capacity at that pH. Also at pH = 9 phthalic acid is totally deprotonated an you will have a very strong eluent.
You need to use an organic acid that is single charged e.g. the mentioned aminohydroxibenzoic acid or hydroxibenzoic acid.

About linearity: as long as the detector is linear in this low transmittance, the method should have perfect linearity as ion exchange takes place. This means the ionic strength of the solution is absolutely constant. You just replace x absorbing charges (phthalate) bei x non-absorbing charges (e.g. chloride).

@ Vlad: that is exactly what I said. With a low quality detector baseline noise is high. But with a high quality detector you might get quite good results. But as we are using the preferred conductivity detection, we did not much of tests with different detectors. Therefore I cannot give recommendations (other than "use conductivity detection"):

[carls]

At pH 9 ammonium will be the buffering agent.

I see now that I was incorrect in thinking that a divalent background anion would give HIGHER sensitivity with indirect UV detection but this is backwards.

Do you recommend aromatic acids as buffers when using conductivity detection due to lower background conductivity (reduced mobility relative to smaller inorganics)?

[Markus Laeubli, Metrohm]

Dear Carl

In direct conductivity detection the measured signal is proportional to the difference of the equivalent conductivities of eluent and sample ion.

k = Conc × [lambda(el.) - lambda(samp.)]

To get response you need either low conducting eluents (e.g. aromatic organic acids) or very high conducting eluents (e.g. OH-). In the latter case you get negative but pretty sensitive peaks. It also allows to see weak acid anions like silicate or borate (which definitely is not possible with suppressed conductivity.

[oscarBAL]

Thanks again to all of you for your replays; I will use a 2487 UV detector; I think this is a good one; Linearity is a good point and it has to be optimized; as long as I can reach 0.1ppm that's by the moment my main concern.

Thanks again.

Oscar

[DJ]

Curious to hear what IP agent you're using, and how much?