GPC/SEC standards: How often do I need to run them?

calendar for scheduling when to run GPC/SEC standards

In a previous post I described how to extend the lifetime of your GPC/SEC standards from Malvern Panalytical.  A natural follow-up question, and one that has been asked of me many times, is “how often do I need to run my GPC/SEC standards?”  Like most things related to GPC/SEC, the short answer is “it depends.”  In this post, I’ll explain what factors affect the frequency with which you should run standards and suggest an example schedule.  After reading this post, you should have the information needed to make the appropriate choice for you and your instrument.

Malvern Dextran

Why do I need to run standards, anyway?

Great question!  If you are using more than one detector to calculate data then you need to run a single narrow standard to determine detector response factors, detector offset volumes, and to determine band broadening corrections.  This is the case if you are using a refractive index (RI) detector and a light scattering detector to calculate absolute molecular weight, or a concentration detector and a viscometer detector to perform a universal calibration analysis.  We also strongly recommend running a second standard to confirm that the calibrated method is accurate.  That’s the reason for the broad verification standard.

peak alignment
Example of raw detector signals before and after a calibrated method with detector offsets and band broadening corrections has been applied

Do all light scattering detectors need calibration?

Yes, all light scattering detectors, regardless of the manufacturer, require calibration! There are different approaches which can be taken as outlined in our white paper Static Light Scattering Technologies for GPC/SEC explained. Importantly, all detectors require calibration with a known molecular weight standard before they can be used in the analysis of samples. 

If you are calculating molecular characterization data without using a light scattering detector, then you will need to run a series of standards to generate a calibration curve.  This is true if you are using a single RI or UV detector to obtain relative molecular weight from a conventional calibration or adding a viscometer to perform a universal calibration.  You might have noticed that for a universal calibration you need both a narrow standard and a series of standards!

If I have a light scattering detector…

Whether you’re working with our right angle + low angle light scattering detector in OMNISEC REVEAL or our SEC-MALS 20 unit, you will need to use a narrow standard to calibrate your calculation method (which involves calculating the detector response factors, detector offset volumes, and band broadening corrections, as mentioned above).  Once you have completed that process, those calculated values are saved in the method to be applied to the analysis of unknown samples.  This method will provide correct data as long as those detector constants accurately reflect the state of the detectors. 

However, nothing lasts forever, even the impressive performance of the OMNISEC and SEC-MALS 20 detectors.  Over time, systems change even if all the conditions remain identical, and this requires that the previously calculated detector constants, offsets and band broadening parameters to be determined again.  When that occurs, you’ll want to make an injection of a freshly prepared narrow standard and repeat the calibration process.

How do I know when to do this?

Now we’re getting closer to the main question of this post.  The best way to ensure you’re always working with an appropriate calculation method is to run a verification standard periodically.  This standard can be the broad verification standard from Malvern Panalytical, or it can be your own standard/sample of which you have a large supply and established characterization data.  If your calculated results for this standard are differ by more than + 5% of the expected values, for example, that can indicate it is time to re-run a narrow standard.

So how often do I need to run my GPC/SEC standards?

This will depend on how frequently you use your instrument.  If you are using your instrument daily, then I suggest running a verification standard at least once a week.  If something looks different or you change a system component/setting (e.g. column set, mobile phase, flow rate, temperature, etc.) that’s a good opportunity to run verification standard or go ahead a run a fresh narrow standard and re-calibrate.  The calculated band broadening corrections depend on mobile phase, and all other run conditions, so re-calibrating is recommended if anything is changed. 

If that’s not an option, such as a situation in which there are no available standards compatible with your mobile phase, then calibrating in one mobile phase and analyzing samples in another is possible, as long as you understand what assumptions you’re making.

Ultimately, it’s up to you.  You can’t run a standard, narrow or broad, too often.  It’s a matter of balancing the cost of standards with ensuring your detector constants are up to date and accurately reflect your instrument in its current state.

Can you suggest a schedule for running standards?

Sure!  Let’s say you use your instrument 2-3 days a week to run 20-25 samples a month.  You can set up a three month cycle where at the beginning of the first month you can run both a narrow calibration and broad verification standard with your first sequence.  Then you can run a broad verification standard with one of your sequences two weeks later (or even at the beginning of the next month).  After three months of running a broad verification standard every 2 or 4 weeks, you can then re-run the narrow calibration and broad verification standard to start the cycle over again.

What if I don’t have a light scattering detector?

Standards used to generate conventional calibration curve

Don’t worry, I didn’t forget about you guys!  For the column calibration methods, any time you change your column set, mobile phase, flow rate, temperature, tubing, anything that will affect a sample’s retention volume, you need to generate a new calibration curve. 

If you haven’t changed a system component/setting, then I recommend using the procedure described above to periodically run a verification standard to confirm your calibration is still valid.

Final thoughts

In conclusion, I hope this helps you understand the calibration process, why it’s important, and how often you should calibrate.  If you have any questions, please don’t hesitate to contact us or email me directly at

Related content

View all articles