Surfactant Removal

Discussions about sample preparation: extraction, cleanup, derivatization, etc.

11 posts Page 1 of 1
I figured this would be a good place to bounce some ideas off of everyone and get an idea if my sample preparation ideas might work.

I have a formulation that is a combination of methyl soyate and ethyl lactate, and contains either 1 or 2 non-ionic surfactants along with the active ingredients.

The surfactants are definitely causing some interference in the chromatography. One idea I have is to utilize flocculation to remove the surfactants. I would use a solution of ferric chloride, mix it with the formulation which should hopefully precipitate the surfactants, then possibly centrifuge and proceed as normal with sample preparation.

I have never worked with removing a surfactant or doing any kind of flocculation and all of the papers I have read seem to reference waste-water clean up.

So I guess my question(s) to everyone is - has anyone performed any kind of precipitation reaction for surfactants and what effect did it have on your active ingredients or recovery? Did you notice it was difficult if the formulation itself that contained the surfactants was not completely aqueous? If anyone has any insight before I start my trial and error, that would be a big help! Thanks!
I don't have experience with this precipitation reaction, but it's used to precipitate surfactants from aqueous solutions... If I understand correctly, your analytes are dissolved in an organic, rather apolar solvent mixture. If you add a ferric chloride solution, will it even mix?

To help brainstorm, I'd like to ask about the nature of your analytes, and what kind of chromatography you're doing. What is the current procedure in a nutshell?
Nonionic surfactants are not easy to remove from aqueous samples. I assume - or hope - that you are using gas chromatography to analyze for your two stated analytes. I know: everyone wants a single assay for stuff, not always possible.

We assayed fatty acid methyl esters routinely by GC. Methyl soyate - from Google Search - appears to be a mixture of the methyl esters of the fatty acids from soybean oil, and if one needs the total amount of methyl soyate or a breakdown of the various components would determine how to proceed with that. So for methyl soyate, one could add some methanol and water and some NaCl to your aqueous sample, then extract the methyl soyate into a smaller volume of hexane or similar non-miscible organic solvent for GC analysis.

For ethyl lactate, much smaller molecule, I don't know if that's readily extracted into hexane. You couldn't do as above because you could get transesterification from methanol, you'd have to try using pure ethanol (that's ethanol not denatured with any methanol). You may want to consider headspace analysis for ethyl lactate (that would eliminate any nonionic surfactants obviously) or even saponifying the sample and assaying subsequent lactic acid by ion chromatography.
Rndirk wrote:
I don't have experience with this precipitation reaction, but it's used to precipitate surfactants from aqueous solutions... If I understand correctly, your analytes are dissolved in an organic, rather apolar solvent mixture. If you add a ferric chloride solution, will it even mix?

To help brainstorm, I'd like to ask about the nature of your analytes, and what kind of chromatography you're doing. What is the current procedure in a nutshell?


Our analyte has two isomers that elute separately and contain multiple rings (5 and 6 carbon rings) as well as multiple esters, and the formulation is not aqeueous, it is primarily methyl soyate and ethyl lactate mixed together with some surfactants. I am currently doing HPLC. I cannot really talk too much about the exact analyte of interest too much.

Normally our procedure is very simple - weigh X amount of formulation directly into a flask, dilute to volume with methanol, mix very well, and inject. The problem is that with our exploration into new surfactants we are unable to quantify the actives because both the surfactants as well as the new formulation itself, contains interfering peaks.

You are correct in that the solution won't really mix, but I am hoping that if I add it and give it a really good mix, I might be able to get the surfactants into the aqueous phase of the ferric chloride solution...maybe I can use a phase transfer catalyst in the attempt to precipitate to promote the surfactants from moving into the newly formed aqueous layer after heavy mixing.

I realize there are many other ways to solve the problem such as altering the formulation, but let's not get into the politics of what we are told to do :).
Consumer Products Guy wrote:
Nonionic surfactants are not easy to remove from aqueous samples. I assume - or hope - that you are using gas chromatography to analyze for your two stated analytes. I know: everyone wants a single assay for stuff, not always possible.

We assayed fatty acid methyl esters routinely by GC. Methyl soyate - from Google Search - appears to be a mixture of the methyl esters of the fatty acids from soybean oil, and if one needs the total amount of methyl soyate or a breakdown of the various components would determine how to proceed with that. So for methyl soyate, one could add some methanol and water and some NaCl to your aqueous sample, then extract the methyl soyate into a smaller volume of hexane or similar non-miscible organic solvent for GC analysis.

For ethyl lactate, much smaller molecule, I don't know if that's readily extracted into hexane. You couldn't do as above because you could get transesterification from methanol, you'd have to try using pure ethanol (that's ethanol not denatured with any methanol). You may want to consider headspace analysis for ethyl lactate (that would eliminate any nonionic surfactants obviously) or even saponifying the sample and assaying subsequent lactic acid by ion chromatography.


Unfortunately I am dealing with a solution that isn't aqueous because, as you said, it does contain methyl soyate. I cannot analyze my formulation with GC, my active ingredients are not GC amenable. So, it's not a possibility for me to play around with any GC at the moment to analyze my formulation.

Unfortunately my formulation from both the solvent system itself (methyl soyate/ethyl lactate) as well as the surfactants used, both have separate slightly or completely interfering peaks with my active ingredients.

So right now my goal is to start with the simpler of the two tasks - removing the interference from the surfactants. Ideally I can precipitate them from the final formulation without affecting the active ingredients, then do my normal sample preparation and see what happens.

As you mentioned, I was in fact looking at saponification as a way to converty the FAME of the methyl soyate into the carboxylic acid, and thus changing the chemistry and possibly shifting the interfering peaks away from the active ingredients. Thoughts on that?
Saponification is worth a try, provided your compounds of interest can survive it. This works best if you can extract your analytes from the obtained (m)ethanolic alkaline phase to an immiscible solvent like hexane.

It seems you have a rather apolar analyte in an apolar matrix, this is usually the most challenging combination in terms of sample prep. Another traditional approach to separate this combination is GPC, but I think the molecular sizes of your analytes and matrix are too close to eachother.

In this situation I would try to change the chemical environment of your sample in a sort of screening test. Prepare a series of extracts, where A is acidified, B is alkaline, C is with acetonitrile added, ... All within the limits of what your analyte can survive, and what is reasonable to inject. It's very trial, error and fun.
Rndirk wrote:
Saponification is worth a try, provided your compounds of interest can survive it. This works best if you can extract your analytes from the obtained (m)ethanolic alkaline phase to an immiscible solvent like hexane.

It seems you have a rather apolar analyte in an apolar matrix, this is usually the most challenging combination in terms of sample prep. Another traditional approach to separate this combination is GPC, but I think the molecular sizes of your analytes and matrix are too close to eachother.

In this situation I would try to change the chemical environment of your sample in a sort of screening test. Prepare a series of extracts, where A is acidified, B is alkaline, C is with acetonitrile added, ... All within the limits of what your analyte can survive, and what is reasonable to inject. It's very trial, error and fun.


Yes, in general the active ingredient of interest is quite difficult to work with in our prototype formulations. We recently had a granular formulation that took me about 6 months just to be able to come up with a way to extract it off the granule , and develop a quantification method because it has such a strong affinity for organic substrates (which our granule is).

As you mentioned, I figured a lot of what I will need to do will be trial and error, which of course is my favorite because it's the best time to try new things and learn something new...it's just quite time intensive and we have a fairly unreasonable timeline given to us for this project...so I figured I'd take my extraction to the forum here and try to extract some information from everyone's brain before I moved forward :). I appreciate your input and I'll be sure to report back if I can get something to work!
I just wanted to provide an update so far. I was able to reduce significantly the surfactant interference. I did a screen of multiple ferric chloride concentrations in a set amount of sample solution, mixed, centrifuged, and sample the top later (the methyl soyate layer).

Granted, I first started with only a blank solution containing only the surfactants so I have no idea how it will affect my active ingredient...but I wanted to start somewhere.

Here is a comparison of the full blank, with and each individual surfacant. The individual surfactant solutions did not undergo flocculation (they had been previously prepared to help me pinpoint which peaks were coming from which part of the formulation). We can see that the peaks circled in red have been reduced fairly significantly (the black chromatogram is the full blank after the flocculation reaction, and the second colored chromatogram in the overlay is the individual surfactant chromatogram, one for each surfactant). All sample solutions had an identical final theoretical concentration of surfactant.

Image

Image

There is a lot of optimization to go, but it looks like the theory of flocculation works when my solution isn't aqueous. Next I'll be looking into how it impacts the recovery of my analyte as well as looking to reduce the methyl soyate interference. Thanks for everyone's input, and if I do have some great success I'll post my results in case it can ever help anyone in the future.
Zoraku wrote:
Thanks for everyone's input, and if I do have some great success I'll post my results in case it can ever help anyone in the future.


That'd be interesting!

This topic reminds me of a new (?) sample preparation technique for LC that recently sparkled my interest. Surfactants are deliberately added to samples (!), to (selectively) extract analytes in their micellar structures that form depending on solvent composition, pH,.. I don't have any practical experience with it, just been reading a bit. I link this to your problem because I was wondering how to avoid having chromatographic issues related to the surfactants themselves. Perhaps the answer lies in those protocols?

Here's an introduction if anyone would be interested.
Rndirk wrote:
Zoraku wrote:
Thanks for everyone's input, and if I do have some great success I'll post my results in case it can ever help anyone in the future.


That'd be interesting!

This topic reminds me of a new (?) sample preparation technique for LC that recently sparkled my interest. Surfactants are deliberately added to samples (!), to (selectively) extract analytes in their micellar structures that form depending on solvent composition, pH,.. I don't have any practical experience with it, just been reading a bit. I link this to your problem because I was wondering how to avoid having chromatographic issues related to the surfactants themselves. Perhaps the answer lies in those protocols?

Here's an introduction if anyone would be interested.


Thank you for that link, I'll take some time today to read it.

So, what happens to be very interesting is that when I perform the flocculation and centrifuge, I end up with a pellet and two different layers - aqueous and non-aqueous layer.

The upper layer is significantly reduced of surfactant content (although not completely) and contains almost solely methyl soyate. This is the initial layer, and only layer I had analyzed the other day because even though it was a blank...I assumed my active ingredients would stay in this layer due to the apolar nature of my active ingredient and wanting to avoid the polarity of the aqueous layer.

The bottom layer is aqueous (from the ferric chloride solution) and possibly ethyl lactate. Unfortunately I do not know the proprietary ratio of Methyl Soyate:Ethyl Lactate as the supplier just calls it Steposol.

After speaking with our formulation scientists, she mentioned our active is soluble equally in both methyl soyate and ethyl lactate...but dissolves and goes into ethyl lactate at a significantly quicker rate.

Yesterday I prepared an actual sample and analyzed both the top and bottom layer and my active ingredient was essentially all contained in the bottom aqueous layer (very interesting because my active also has extremely low solubility in water) - I believe the peaks in the upper layer are just the combination of methyl soyate and surfactant interferences. The lower layer is also completely devoid of any methyl soyate peaks as well as surfactant peaks...I'm going to analyze the blank lower layer today just to verify there's no hidden interference but I think this flocculation type of reaction, in an acidic environment, separated the methyl soyate out as well as the surfactants while retaining my peaks of interest in the aqueous phase. I will post some chromatograms after I do a few more injections.
Just to close this topic out, as I said I'd post more chromatograms, we ended up completely moving away from that formulation as we found a better more compatible formulation system for our active ingredients for this product.

So, unfortunately, I didn't have time to really do more work with this extraction even though I found it very interesting. There is definitely some promise using either a flocculation type reaction or saponification reaction to separate out surfactants or even specific formulation components (such as fatty acids) that might interfere on HPLC.
11 posts Page 1 of 1

Who is online

In total there are 6 users online :: 0 registered, 0 hidden and 6 guests (based on users active over the past 5 minutes)
Most users ever online was 599 on Tue Sep 18, 2018 9:27 am

Users browsing this forum: No registered users and 6 guests

Latest Blog Posts from Separation Science

Separation Science offers free learning from the experts covering methods, applications, webinars, eSeminars, videos, tutorials for users of liquid chromatography, gas chromatography, mass spectrometry, sample preparation and related analytical techniques.

Subscribe to our eNewsletter with daily, weekly or monthly updates: Food, Environmental, (Bio)Pharmaceutical, Bioclinical, Liquid Chromatography, Gas Chromatography and Mass Spectrometry.

Liquid Chromatography

Gas Chromatography

Mass Spectrometry