relative SEC and hydrodynamic volume


I have been analyzing some polymers with a more polar solvent on SEC - ELSD detection. (relative to Polystyrene)

Where I normally measure a MW of 30K for PLA (relative to Polystyrene) in chloroform, I now measure a MW of 140K. 

In other words, in this polar solvent, the PLA has a bigger hydrodynamic volume, right?

If the solvent surrounds the polymer more, and this causes a bigger hydrodynamic volume, would the analysis then be incorrect?

Let me know what you think,


  • That conclusion certainly makes sense; GPC separates things based on size in solution (rather than molecular weight), and so an earlier elution time usually indicates a larger size in solution.

    A couple of things to bear in mind ...

    Firstly, if the solvent is not ideal for the PLA polymer, is it actually separating the polymer into individual chains; does it take longer to dissolve or could there be some intermolecular interaction leading to the polymer forming dimers or higher order species ?  This is perhaps not very likely.  Another explanation might be ...

    Secondly, if the original solvent is not ideal for the polymer there may be secondary interactions between the sample and the stationary phase causing the polymer to elute later than it should and so appear to be smaller in size (and lower in molecular weight).

    PLA and polystyrene are very different polymers and so finding a column stationary phase and mobile phase solvent that works equally well for both and ensures that there are no secondary interactions affecting the retention time is much more difficult.  It is better to choose a calibration standard that is more similar to the sample you are analyzing.

    Please check if the retention times of the polystyrene standards have shifted at all in the different solvents.  Are the polystyrene standards eluting later than expected in the polar solvent making the PLA appear to be larger than before ?

  • Hi Andy,

    First off, thank you for your reply!

    Chloroform does dissolve my polymers easier than the polar solvent I am testing against.
    However, I do not expect that it would form dimers or higher order species.

    After taking up a chromatogram with the same flowrate, I compared Polystyrene in chloroform and the polar solvent.
    The Polystyrene does shift to a later point in the chromatogram.
    Because PLA and Polystyrene are very different indeed, I also calibrated the system on PMMA standards.
    This gave a quite similar shift, and therefore more logical relative Mw's. (Thanks!)

    Polystyrene: Elutes later in polar solvent
    PLA: Elutes earlier in polar solvent
    PMMA: Elutes earlier in polar solvent (similar to PLA)

    To prove that the shifts happen because of their hydrodynamic volumes, I want make a Mark Houwink plot in different solvents.
    This way I should be able to prove how "good" the solvent is, and the correlation to how the polymer unfolds in the solvents. Correct?


    "The Mark–Houwink equation, between intrinsic viscosity  and molecular weight.

    From this equation the molecular weight of a polymer can be determined from data on the intrinsic viscosity and vice versa.

    The values of the Mark–Houwink parameters depend on the particular polymer-solvent system.


    For solvents, a value of a = 0,5  is indicative of a theta solvent. ( = ideal chains in solvent)

    A value of a = 0,8  is typical for good solvents."

  • Hi Roy,

    Glad to be of some assistance.

    It seems like the polystyrene standards are eluting later than expected, so it is likely they are interacting with the stationary phase when using the polar solvent.

    If you have a viscometer then you can certainly gain more insights into the behavior of the polymer in the solvent, but it also helps in situations like yours where the standards you are using are somewhat different to the sample you are analyzing.

    If you haven't already read this, I recommend you take a look at this primer on multi-detector GPC: 5990-7196EN-BJM.indd (

    Good luck !


  • Hi Andy,

    Many thanks again!
    I will add the use of a multi-detector in my recommendation, and hopefully I can further develop the method with the multi-detector as well. 


Reply Children
No Data
Was this helpful?