Are my samples suitable for GPC/SEC?
I’ve written a decent amount about OMNISEC and GPC/SEC characterization over the years, describing what makes GPC/SEC data trustworthy, how multi-detection GPC/SEC works, what the post-column filter does, and how compositional and branching analyses work, among other topics. In this post, I want to take a step back and discuss a topic that frequent OMNISEC users might take for granted: what samples are suitable for GPC/SEC analysis.
With the range of characterization data OMNISEC can offer, it’s understandable if you want to inject all your samples! However, before you get ahead of yourself, there are certain aspects of your sample you will need to consider. In the sections below, I’ll describe why sample solubility is critical, the molecular weight and size range appropriate for GPC/SEC samples, and how functional groups present on your sample can potentially affect the observed data.
If your sample meets all the criteria below, then you are welcome to inject away!
Solubility
The most important feature of a GPC/SEC sample: the sample must be completely soluble! GPC/SEC is a solution-based technique that utilizes a liquid mobile phase to carry the sample through the stationary phase and the detectors. If the sample does not create a homogenous solution when dissolved, then it likely doesn’t even make it into the flow path and has little chance of reaching the detectors.
If the sample is only partially soluble, then you can only observe and characterize the portion that is in solution. This can still provide useful data in some cases, especially if you have an established dn/dc value and using GPC/SEC to determine sample concentration.
Emulsions and dispersions/suspensions are not ideal samples because, by definition, they are not homogenous. In some instances, using a solvent that solvates the sample and is miscible with the liquid medium can overcome this challenge. For example: using THF to dissolve a latex dispersion; the THF can potentially dissolve the organic polymer, and the THF is also miscible with the water. In my experience, this only works on rare occasions, but is worth trying.
Some samples might require pH adjustment to maintain their solubility. In these cases, it’s important to check the specifications of the system and columns to ensure you stay within acceptable ranges for both.
Temperature can also be a tool to facilitate sample dissolution. OMNISEC has temperature-controlled zones (autosampler, column set, and detector compartments) to provide temperature control where the sample spends the majority of its time within the system. While using temperature to maintain sample solubility can work in some situations, I recommend trying to find an alternative solvent that doesn’t rely on temperature.
Molecular weight
Since molecular weight is the primary reason to seek GPC/SEC analysis, even though the separation occurs based on molecular size, I’ll start here and then look at size. The upper and lower molecular weight limits will differ slightly with each system, set of conditions, and sample type, but in general, the molecular weight range of a GPC/SEC instrument is from hundreds of Da on the low end to millions of Da on the high end. Our polystyrene column calibration standards range from 1,200 Da to more than 4 MDa, as shown below.
Pushing the low end of the molecular weight range are discrete molecules with molecular weights in the hundreds of Da, such as the examples in this post and the analysis of sucrose, with a molecular weight of 342 Da, described in this post.
Molecular size
From a size perspective, samples with a diameter of 200 nm or smaller tend to work best. This is not an absolute maximum, rather one with a practical basis: we strongly recommend all OMNISEC users filter their samples through a 0.2 µm syringe filter during preparation (shown below). This protects all parts of the instrument, including the detectors, which require costly repairs if they become dirty or plugged. Additionally, if you have a post-column filter in your system, which we also strongly recommend, its filter membrane has 0.2 µm pores.
For polymer samples, the 200 nm threshold allows almost every sample I’ve attempted to analyze. The 4.2 MDa polystyrene in the image above has a diameter of about 150 nm. For bio-related samples, some viruses and nanoparticles may be too large and may require system customization and specialty columns. It helps me to remember that GPC/SEC is designed to analyze molecules. If you describe your sample as particles, that may still work, but you may also want to explore some of the alternative techniques listed below.
On the small end, some of the oligomers and the sucrose examples linked above have diameters of about 1 nm – that’s pretty small!
Functional groups
There’s nothing wrong with having functional groups on your GPC/SEC sample – that’s what makes them interesting! However, if the functional groups impact a sample’s solubility, interact with the column, or affect a detector response then they need to be considered so the analysis / data can be evaluated with the proper context.
Functional groups can facilitate the solubility of organic polymers in aqueous media (e.g. the sulfonate groups of polystyrene sulfonate), or simply increase the solubility of a sample to allow for GPC/SEC analysis (e.g. the hexyl groups of poly-3-hexylthiophene). That’s perfect…as long as those functional groups don’t also interact with the stationary phase of the column.
For example, samples with pendant amine groups can run well in a slightly acidic aqueous media, since the amines become protonated ammonium groups. In these situations, a cationic column must also be used, otherwise the newly formed positively charged ammonium groups can interact with the gel of a neutral GPC/SEC column and negatively affect the chromatography.
Last, how the functionality of the samples affects the detectors can be critical in some situations. If you are using a UV-Vis detector to measure concentration, then your samples need to have a chromophore. And you need to be using a mobile phase that does not also contain that chromophore.
As I detailed in an example in my post last month, if your sample absorbs at 640 nm, which is the wavelength of the light source of the light scattering detector, then the light scattering response for your sample will likely be compromised and the calculated data unreliable. Being aware of these potential issues can save a lot of time and headaches when trying to make sense of the data after the fact!
Alternative characterization techniques
If your samples aren’t suitable for GPC/SEC analysis, there are alternative characterization techniques available. Several of them are offered by Malvern Panalytical!
If your sample is too small or low molecular weight for GPC/SEC, then you can try using nuclear magnetic resonance (NMR), gas chromatography (GC), mass spectrometry (MS), other separation mechanisms (e.g. HPLC, CE), infrared spectroscopy (IR), or even a stand-alone UV-Vis spectrophotometer.
If your sample is not completely soluble, or is too large, or has a molecular weight too high for GPC/SEC, then you can try dynamic and electrophoretic light scattering (DLS & ELS; with the Zetasizer!), laser diffraction (with the Mastersizer 3000+!), particle size and shape analysis (using the Morphologi 4!), and nanoparticle tracking analysis (NTA, with the NanoSight Pro!), with the Malvern Panalytical technologies shown below. Other alternatives include electron microscopy (EM), field flow fractionation (FFF), and rheology.
I’m sure I’ve accidently left some techniques off the list – if you think of one, please email and let me know!
Final thoughts
I hope this post has increased your understanding of GPC/SEC as a technique by focusing on the features of suitable samples. When working with a sophisticated multi-detector GPC/SEC system like OMNISEC, it is important to think about your sample in the context of these requirements and how it will (or will not) behave in your instrument. If you have any questions, please don’t hesitate to contact us or email me directly at kyle.williams@malvernpanalytical.com.
Related content