About chromatographic separation for an exotic polymer

[Alfred88]

Dear Chromatographers:
I was assigned to develop an assay for a polymer (let's just called it NewX). The manufacturer provided no assay to demonstrate purity. The C of A has info on solid content (yy%), but not potency. Because I don't want to affect the marketing strategy of our company, I prefer to leave out the chemical name of 'NewX.'

I have had difficulties with the method development, because of the followings: lack of expertise, lack proper detector (we have UVD, RID, and ELSD, but no MS), no official monograph, no reference standard, and the low level of the NewX in the final product.
This NewX is not retained by a traditional C18 or C8 column. When I tried on a short macropore (5um, 300A) C18 column used for peptides, I got a nice peak but it was not reproducible! When a longer macropore C18 column is used with gradient elution, I got 3 clusters of peaks in ELSD (this NewX has weak UV abs). I realized that the portions of NewX eluted at approx. 25%, 60%, and 90% organic (I recall these # from my memory). I have reported this 'discovery' to my upper people, but they only want to know when a test is developed.

I also tried to check the MW distribution (actually, the dynamic volume) of diluted NewX, using one GPC 7.8x300mm column. From that run, it appeared that NewX had at least three groups of significant different MW ave (could be more than one order of magnitude).

My questions are:
1. How to quantify NewX if it contains at least three distinct groups of polymers (not just differ in chain length).
2. How to create an in-house standard, when we can't buy one.
3. Should we replace NewX by something else more uniform/homogeneous.

Thank you.

[Uwe Neue]

Well, the NewX probably works for what it was designed for, and this is the reason why you are supposed to analyze it.

Please tell us more about your SEC. You got three peaks, which in SEC commonly means that you are dealing with three different molecular weights. But before coming to this conclusion, it would be nice to know the GPC column and the GPC conditions, together with the elution times of your different peaks.

Considering that you are dealing with potentially three components, the conclusion that you get no retention on a standard C18 may not be valid. What if ONE of these three things is not retained on your C18, and the other two do not show up?

[Alfred88]

Dear Uwe Neue:
Thank you for your response!

Conditions of GPC/SEC:
Column (one column): Waters Ultrahydrogel Linear 7.8 x 300 mm (according to Waters, max organic is 50%; however, I had to use 90% organic to ensure elution).
MP: Isocratic, 90% Methanol, and 10% of Formic Acid (0.1% in PW). MP flow rate: 0.6mL/min. Column temp: 40C. For ELSD: Nebulizer: 30C; Evap.: 90C. Gas flow: 1.4 SLM.
Sampling rate: 1Hz. Run time: 45 min.
Elution time of clusters in GPC/SEC: 8.5, 10.5, and 13.5 min. Also, one extra cluster at 17.5 min (the last group consisted of small bumps).

This material showed poor (weak) response by RID.
We also learned that the diluted material had negative abs. in UV, so we switched to check the transmittance instead. New ‘discovery’: increasing concentration will shift the max. trans. of the NewX solution from 234 to 231 nm; and NewX solutions had high trans. values (>300).

I desperately need to solve this puzzle. Thank you for your help.

[Vlad Orlovsky]

Hello Alfred,

I can offer you free method development for your compound. If you send us few mg of your polymer we will try to develop a method for you (I think that we already done this once for you).

If you are interested please contact me through email.

[Uwe Neue]

Some of the material has a very large MW, essentially excluded from your SEC column. If may be tough to get good and reproducible RP retention from something like this. You definitely hsould use a large-pore RP packing.

You can collect the fractions from the SEC, and then concentrate a bit and inject into the RP mode. If the three SEC peaks correspond to the three RP peaks, I would concentrate on developing the SEC method, potentially by adding a Ultrahydrogel 2000 in front of the linear column to enhance the separation for the first peak. For something like this, SEC is likely to be better than RP.

[Alfred88]

Dear Dr. Vlad:
Thank you for your free offer. I would love to take it, but my direct boss disagreed. Could you advise us on how to proceed? Great thanks.

Dear Dr. Uwe Neue:
Currently I got weak response by ELSD with one 7.8x300 mm column. If two 300mm columns are used (coupled), I am afraid that I may get no peak at all.
I am thinking about using the Strata C18-T (wide-pore SPE, Phenomenex) to pre-concentrate the solutions before injecting to GPC/SEC. Should we explore this first? The Ultrahydrogel 2000 will cost ~$1300 (we don’t have this currently).

[Uwe Neue]

You will get some dilution from the second column, but if your injector allows, you can inject double the volume. People commonly inject a much larger volume in SEC that one does in RP.

Your first peak is nearly or completely excluded from the SEC column, which is why I suggested to use the largest pore size packing to get some additional definition.

You can also do this the other way round and collect the peaks from the RP and inject them into the SEC: the point is, if the three peaks in SEC correspond to the three peaks in RP, SEC will be much simpler and less problematic for future analytical work.

[Ken Tseng]

Conditions of GPC/SEC:
Column (one column): Waters Ultrahydrogel Linear 7.8 x 300 mm (according to Waters, max organic is 50%; however, I had to use 90% organic to ensure elution).

I agree that in this case, SEC is probably better than RP. However, when you use more organic solvent than the spec, you run into the risk of shortening your column life. Worse, you may be unable to conclusively rule out that those three MW distributions are all from your sample.

Here is a link to Shodex webpage with some basic SEC information and solvent properties. http://www.shodex.com/english/dc0616.html