Chromatography Forum

Hi Forum,

I am trying to publish a paper that reports a new application of LC instruments to study adsorption at liquid/solid interfaces. I can't say much about it - because I am trying to publish a paper that describes the new technique (I think its new, anyway) and highlights its applications in geochemistry.

However, I have been getting pushback from reviewers saying that what I am doing is the same as what Mikael Tsvet did. I have yet to find a good source of information that describes Tsvet's work in the level of detail that I seek. What I understand is that his work on plant pigments was based on weak interactions b/w the plant pigments and the stationary phase. Although this is described as 'adsorption' chromatography - my understanding is that his worked involved weak reversible adsorption. By the end of his experiments everything he injected came back off the column, i.e. with ultimately total recovery. Resolved into bands of course, which was an amazing finding and a great development!

If someone uses a column experiment to study adsorption - irreversible adsorption, if they inject a mixture - some will bind strongly and irreversibly, and some will pass through the column and be detected. The recovery for an injection in which binding occurs will be less than 1. And so in my mind there is a real difference.

Am I misunderstanding Tsvet's work? To me it was taught in terms of equilibrium.

I don't know about the Tswett reference. He did chromatography, you are doing the same thing. We are all doing the same thing. So we should not publish any more?
What do you think is new about your technique? Is it the partial breakthough that you described in the other post? If you think that this is new, I disagree. This phenomenon has been known for a long time to everybody doing sample prep. If you flow too fast, you get partial breakthrough. This is a question of particle size, diffusion coefficent and residence time.
You can e-mail me and I will explain it further.

A problem is that there are a nearly infinite number of ways to get a "recovery" of less than unity, all of which you would need to eliminate before you could be sure that irreversible binding was really happening.

Peter

Would this be one way to prove it that would convince you? After saturating the column with our adsorbate, we then injected a material that is known to have very strong competitive adsorption interactions; phosphate. Phosphate ion is not seen in our detectors, but we can see the organic material that is desorbed when we inject phosphate. With repeated injections of phosphate we get decreasing amounts of organic material desorbing.

How else might it be proven that there is irreversible adsorption? to say that there are an infinite number of ways you can get partial breakthrough, I find a bit nebulous really. I can't find it very useful as a scientific argument unless you are more specific.

David

Hi David

By injecting phosphate, and detecting the absorbate, you are recovering the adsorbate, not so ?. So what you are demonstrating is not irreversible adsorption, detected as recovery being less than one, but that under one set of conditions the adsorbate does not elute from the column, and under another set of conditions it does. This, I'm afraid, is nothng new.

I didn't say that there were an infinite number of ways that you can get partial breakthrough. I said that there were an infinite number of ways that you can get recoveries of less than 1, which in your first post was your criterion for irreversible binding. Partial breakthrough (with e.g. an SPE cartridge) in only one of them. To measure the recovery you have to know what peak area to expect from a given quantity of analyte, and then compare that to the peak area that you actually get when you analyse a (usually spiked) sample containing that quantity of analyte. If you have "irreversible" binding how are you going to measure either of these peak areas ?

You need not worry about saying too much about your findings in advance of publication - you only ned to keep things secret if you want to patent them, and you will blow that as soon as you publish anyway.

Peter

In this sort of thing one has to decide whether the best results are obtained with a flow system (like LC) or in a bulk (here slurry) system. If you are packing geological material in a column your data may reflect packing skill. .... My guess would be that equilibrium studies in bulk are easier to control, thus more accurate.

Hi,

Mr. Mueller, thank you for your post. That is an interesting way to put it - in terms of how well controlled dynamic column investigations of adsorption could be relative to batch experiments. I wonder what suite of information would be required so that one could be ensured that column adsorption experiments were well controlled. As an expert chromatographer, what type of information would you be convinced by?

At this time I am performing a set of experiments to characterize the columns that I use in my work, including meaurements of column packing reproducibility by studies of NaNO3 elution, i.e. dead time measurements.

There has been over the years, numerous instances in which people have studied adsorption at liquid/solid interfaces using continuous-flow 'breakthrough' column experiments. Why does any bother with those?

I have a couple of ideas on why one would even study adsorption this way. First off, have you ever carried out batch experiments? They are notoriously tedious and time consuming. Researchers attempt to model isotherms with less than 10 data points, and the error bars are +/- 10% at least. LC experiments, if one has sensitive detectors, and an auto-sampler, are extremely efficient in comparison and promise more precision. However, if a person is interested in equilibrium systems, batch tests are indeed the way to go.

Studies with dynamic columns can replicate non-equilibrium systems in which there is flow - such as in the vadose zone above aquifers - and in technical systems that purify water based on filtration/adsorption.

I have really learned that I need to be careful in the way that I describe the methodology I am using. It is not new - in the sense of our awareness of all the things that can happen on a column - but it may represent a method that is new and useful in geochemistry.
Thanks again, and have a good weekend..

David

I'm not sure if this is related, but many years ago, Csaba Horvath published a series of papers on "displacement chromatography." If I recall correctly (it's been a long time), he would deposit components on the column under "weak" eluting conditions (complete retention, no movement, etc.), and then begin the eluting process with either displacer molecules or just stronger mobile phase (does anyone remember details?). The resulting profile was did not consist of individual peaks, but rather rectangular bands, stacked in order of increasing affinity for the column.

Is this in any way similar?

dkreller, based on your info it is not possible to ascertain which method is less tedious, better, etc. I can only relate own experience like the one of a few years ago when I helped to synthesize radioactively (C-14) labeled nonanephenols. Derivatives of these were used to study their attachment to soils.. The goal was to get accurate distributions of the detergents, via radioactivity measurements and determination of their integrity (stability) during the experiment I don´t even have the beginning of a thought as to why such experiments should be transferred to a column. The science was paramount, not the time it took to do it.
Also, obviously there is not enough information to tell whether yours is a new technique, nor whether it is a very interesting new application.

Just to reiterate what Peter Apps has stated above: partial breakthrough (with e.g. an SPE cartridge) is a very well known phenomenon.