Hello.

Short answer:

- Pullulan standards (can be ordered in variety of molecular weights)

- Dextrans (can be ordered in variety of molecular weights)

- Polyethylene oxide (PEO, can be ordered in variety of molecular weights)

- Polystyrene or Polystyrene Sulfonate (PS, PSS, can be ordered in variety of molecular weights)

- Bovine Serum Albumin (BSA, 66 kDa)

- Immunoglobulin G (IgG, 150 kDa)

- Thryoglobulin (330 kDa)

The dn/dc (or extinction coefficients) of all those is either supplied by the manufacturer or it's easily available online upon search. Same for the viscosity. One you know one, you can measure the other automatically using the multi-detector mode you have (knowing dn/dc allows you to measure extinction coefficient and vice versa). Mark–Houwink–Sakurada plot (LogM vs n) can be used if you know the theoretical Mw and your inline detector measures the experimental viscosity (n).

Detailed answer:

For calibration curves you need polymers that can be synthetized or isolated in controlled fashion (low polydispersity) and analyzed reliably. Those standards will depend on your target analyte. Retention time curves are only valid if the spatial geometry of the standards and the analyte are similar. Tight compact globular polymer and extended linear polymer will not at the same time even if they have the same molecular weight. If you would be using a calibration curve then the conformation (sphere, random coil, linear) of those standards should be similar to your analyte; ideally the charge and viscosity should be similar as well to minimize effects due to ion-exclusion and peak shape alteration due to viscosity. If you are working with proteins and polypeptides, use protein curve (BSA, IgG, etc), and so on.

The dn/dc and viscosities are best determined experimentally because they will depend on temperature, laser wavelength, solvent, conformation and oligomeric state, and on the state (free acid/base vs salt). However, it usually okay to use approximate values even if the exact conditions from literature differ from yours. For example, globular proteins are typically used with dn/dc of 0.185.

Side note. Just to clarify the last of your sentences: You cannot get "absolute" molecular weight using a calibration curve. By absolute molecular weight it's understood that the methodology used is absolute, i.e., does not require any external standards or calibration. Such technique, for example, is MALS (multi-angle light scattering) coupled with either refractive index or UV detection.

Look online for "Refractive Increment Data-Book for Polymer and Biomolecular Scientists" and see if you can get your hands on it.

See some referenced dn/dc coefficients here:

https://www.wyatt.com/support/ask-the-expert-i-found-the-perfect-mobile-phase-for-my-experiments.-now-how-do-i-find-the-right-validation-standard.html

Some standards to purchase:

https://www.agilent.com/en/product/gpc-sec-columns-standards/gpc-sec-standards/pullulan-standards

https://www.agilent.com/en/product/gpc-sec-columns-standards/gpc-sec-standards/dextran-standards#:~:text=Dextran%20standards%20are%20the%20most,wide%20variety%20of%20application%20needs.

https://www.sigmaaldrich.com/US/en/product/sigma/a1900

https://www.sigmaaldrich.com/US/en/technical-documents/technical-article/analytical-chemistry/large-molecule-hplc/size-exclusion-chrom-protein

Hi Petroslav Chipinski ,

Thank you so much for the help. I have been using pullulan to perform system calibration but my eluent is phosphate buffer saline. Would you recommend this eluent or water for polysaccharide analysis?

Secondly, I obtained the dn/dc value from the textbook you have recommeded but our software is still requiring me to put extinction coeffiicent, refractive index of the eluent and intrinsic viscosity. I have attached an images. May you kindly assist?

Best regards,

Tina

PBS is fine with Pullulans and polysaccharides. The dn/dc of Pullulans in PBS is about 0.14 - 0.16. Last time I measured it for 22 kDa standard (aqueous buffer pH 4.0, 658 nm on the refractive index detector) it came to 0.147.

The extinction coefficient for Pullulans is near zero at 280 nm (0.03 is what I have in my notes) but will depend the wavelength. It will always be marginal aside from 190-205 nm range. You can measure it by preparing 5-6 linear solutions of it and running it on UV reader at the desire wavelength you want. Build a curve with concentration (x-axis) and absorbance (y-axis); the slope of the linear equitation will give you the extinction coefficient.

The refractive index of the PBS buffer is about 1.33 at 25oC (~1.32 at 40oC). The viscosity of PBS is around 0.89 cP at 25oC (around 0.65 at 40oC I believe). The viscosity of the solvent is used by the software to adjust the expected optical properties of the medium.

Usually GPC, MALS, and DLS software contain all this data for variety of common solvents and can make auto-adjustment for temperature, so check your manual help button.

Hi Petroslav Chipinski

Thank you so much for your help. I have found the literature values and just when I wanted to try it, I had a new issue with the software. The first issue, the traces are not lining up and chromatograms do not start at 0 mins. Secondly, every integration parameter is greyed out and does not allow me to perfomr any integration and select the peak of interest. Might you have any ideas how to resolve this?

I have attached some photos.

I appreciate your help.

Best regards,

Tina