Can my GPC/SEC chromatogram be trusted?
You’ve done the hard parts to get your GPC/SEC instrument ready: prepared the mobile phase, stabilized the detector baselines, and prepared your sample. After injection and completion of your sample sequence, you have acquired data. Congrats! Now, before you even analyze it, how do you know if your sample chromatogram is trustworthy?
Fortunately, there are an handful regions you can easily identify in your chromatogram with particular features that will offer a good indication of the quality of your data. These include the Starting Baselines, Exclusion Volume, Sample Peak, Solvent Peak, Viscometer Inverse Peak, and Finishing Baselines, which are color-coded sequentially on the chromatogram in Figure 1 below. I will step through each of these, describe what is expected, what a different result might mean, and highlight some examples. While your sample data may not have the profile you anticipated, that doesn’t mean there’s anything wrong. In fact, those chromatograms tend to be the most interesting!
Starting Baselines (I)
This one is pretty simple: the detector baselines should be flat, stable, and quiet immediately after injection. The appropriate baseline levels and acceptable noise ranges for each detector are listed in the system’s manual.
If there is a large dip in viscometer signals (DP and/or IP) that could indicate air in the sample loop that produces a decrease in pressure upon injection. The chromatogram in Figure 1 shows a slight dip in the DP viscometer signal. The presence of air in the injection loop means that the actual sample concentration does not correspond with the prepared and input sample concentration.
Exclusion Volume (II)
The exclusion volume represents the lowest retention volume at which anything will elute from the column set. Therefore, nothing should elute before this point. Each column in a set contributes about 5 mL to the exclusion volume, so a two-column set would have an exclusion volume at approximately 10 mL. The specific exclusion volume of each column will vary based on the particle and pore size of its stationary phase. If something does elute early, extend your run time; something from the previous injection maybe just now eluting.
Sample Peak (III)
The sample should elute between the exclusion limit and the solvent peak. If that’s not the case, the column set in use needs to be adjusted and optimized for that particular sample/application. All three detectors (potentially four, if the sample has a chromophore that stimulates a UV detector) should show a response. It is important to note that the detector responses probably won’t, and probably shouldn’t look the same since each detector responds to a different aspect of the sample. However, there should be some apparent agreement between the responses of the detectors. The sample chromatogram in Figure 2 highlights this; the main sample peak is present in the refractive index (RI, red), right-angle light scattering (RALS, green), and viscometer (blue) signals.
However, there is a small shoulder on the earlier eluting front of the RI peak, indicating there is a low concentration of a slightly larger material present (as the RI detector responds to a sample’s concentration). Not only is this confirmed by the RALS signal, which responds most strongly to a sample’s molecular weight, but the presence of another larger (because it elutes earlier) species is apparent at 11-13 mL. Since this peak is visible only with the RALS detector we know that it exists in very low concentration (no RI response), has a dense molecular structure (no viscometer response), and has a high molecular weight (even at low concentration it still produces a RALS response). This type of profile is most indicative of aggregates, which is what is shown here.
Solvent Peak (IV)
Since solvent molecules are small and discrete, they are the last to emerge from the column set and represent the end of eluting material from a sample injection. Other small molecule components of the sample are included in this region, such as monomeric or oligomeric materials, salts, reaction byproducts, and residual reaction solvents. Because of this, the solvent peak region often shows multiple peaks and may be positive, negative, or a combination based on the individual components’ dn/dc value.
The RI detector responds most strongly in this region, as its response is directly related to the eluting materials’ concentration. The light scattering response will be minimal since the molecular weight of these compounds is so low. If a light scattering response is present, it is often because there is a large concentration of a component. This is illustrated in both Figures 1 and 2, where the dissolution solvent is different from the mobile phase.
Viscometer Inverse Peak (V)
When the sample enters the viscometer detector the flow path splits, which allows half of the sample to continue through unimpeded while the other half is held in a delay column. When the delayed half of the sample finally elutes, it produces an inverse response in the viscometer signal. This inverse peak will generally elute 7-20 mL after the sample peak. If the solve peak produces a viscometer response, e.g. the dissolution solvent is different than the mobile phase, it will also have a corresponding inverse peak. The chromatogram in Figure 1 shows both the sample inverse peak (25-32 mL) and solvent inverse peak (32-34 mL). The chromatogram in Figure 2 presents a more extreme example of this, as the solvent peak at 25-28 mL is so strong it produces an inverse peak at about 40 mL that sends the viscometer response plummeting.
If you don’t observe an inverse peak for your sample (or solvent) it could be that the chromatogram ends prior to its elution and/or that the sample (or solvent) viscometer response is too small. It is important to extend the analysis run time to include these inverse peaks so they don’t run the risk of eluting during the next injection and interfering with the subsequent sample peak.
Finishing Baselines (VI)
All detector baselines should return to their starting levels by the end of the chromatogram. If they don’t, there’s a good chance the sample is interacting with the column set in an unexpected manner and slowly eluting. Optimizing the chromatography by changing the mobile phase, adding salts or cosolvents, or even changing the column set can eliminate this issue.
Next time one of your samples provides a chromatogram that doesn’t look exactly like you thought it would take a few minutes to break it down by each region and understand what the data is telling you. If there are recognizable problems, please don’t hesitate to contact us and ask for assistance. And if everything checks out, you may have produced groundbreaking data!
Previous blogs: