2D Polymer LC as a High-Speed, High-Throughout Application

Martina Adler

,
Peter Kilz

LCGC Europe

LCGC Europe, LCGC Europe-10-01-2006, Volume 19, Issue 10
Pages: 552–556

2D polymer liquid chromatography is a powerful tool for the deformulation of complex samples; however, it is considered to be very specialized and time-consuming. This article shows how recent hardware and software improvements have led to the technique becoming a method for routine analysis.

2D polymer liquid chromatography is a powerful tool for the deformulation of complex samples; however, it is considered to be very specialized and time-consuming. This article shows how recent hardware and software improvements have led to the technique becoming a method for routine analysis.

The demands on chromatography, regarding the information density of the results, increase with the complexity of the samples under investigation. Every single separation method suffers from limited chromatographic resolution (peak purity, limited peak capacity, determination of multiple property distributions etc.). Many attempts have been made to overcome the intrinsic limitations of the chromatographic process.

A powerful approach is to combine different separation techniques into a single experiment (multidimensional chromatography; also known as 2D chromatography, orthogonal chromatography and cross-fractionation).

The combination of independent separation techniques improves the chromatographic resolution because of increased peak capacity, which is not the sum but the product of the peak capacities of each technique:1

The separation "angle" between two techniques that separate strictly according to different properties is considered as 90° and the separation is referred to as "orthogonal".

An advantage of chromatographic hyphenation over performing two isolated separation methods is the ability to correlate and identify the properties of different sample components.

The mapping of four 2D experiments is illustrated in Figure 1 — the separation according to one property is plotted versus the separation according to a second property. As can be seen, projections to the axes of each "contour plot" are the same for all examples. The traces of those projections would also result from isolated, one-dimensional separations. Therefore, any traditional separation optimization strategies would fail in all of these situations. Nevertheless, the direct combination of both separation techniques can easily show the sample differences. The 2D separation is achieved by a consecutive separation of fractions eluting from the 'first dimension' using the 'second dimension' technique. The resulting contour plots can be read in a similar way to a geographical map that shows the first dimension on the y-axis and the second dimension on the x-axis.

Figure 1

This type of separation is extremely useful for any kind of complex samples, which are difficult to separate, identify comprehensively or fully deformulate. This is most often the situation for synthetic, natural and biopolymers (monodisperse proteins are one well-known exception), which possess coexisting multiple property distributions [such as molar mass distribution, functionality type distribution, chemical composition distribution, molecular architecture (shape) distribution, to name just a few].

This article will focus on comprehensive 2D polymer liquid chromatography (LC) coupling a liquid adsorption mode technique (LAC) with size-exclusion chromatography/gel permeation chromatography (SEC/GPC), and on the options to use them as a high-throughput application.

Since the 1980s, several groups have tried to implement this technique for polymer analysis. However, it was seen as a very time-consuming technique that could take days for a single experiment — depending on the degree of automation (off-line or stopped flow). In the last decade of the twentieth century, the first fully automated 2D LC experiments were reported that could further reduce the time consumption to less than one day. This led to an increased use of 2D techniques in R&D laboratories.

More information on other aspects of 2D chromatography may be taken from a review about 2D chromatography of synthetic polymers.2 Multidimensional separations are also performed in other areas, such as in a current review that focuses on biomedical and pharmaceutical applications.3

Experimental

The implementation of 2D chromatography is straightforward if robust separation methods are available for each dimension. Two existing chromatography systems (consisting of a minimum of two pumps, two columns, one injector and one detector) can be used for the separation. A transfer valve for automated fractionation and appropriate software are also required.

Comprehensive 2D work employs complete transfer of injected mass from the first to the second dimension. This ensures that the complete sample is analysed and detected in the second separation method. The advantages and limitations of different approaches will be discussed.2

Full automation is a basic requirement for short analysis times. However, in the last couple of years it has become possible to reduce the total analysis time for a 2D experiment, including data evaluation, to one or two hours and in some situations to even shorter times. The following considerations are based on the most commonly used set-up for 2D chromatography in polymer analysis with a dual loop transfer injection valve (8-port, 2-position valve), an HPLC separation in the first dimension and a SEC/GPC (size-exclusion chromatography) separation in the second.4 A schematic of this set-up is shown in Figure 2.

Figure 2

The effluent of the first dimension (HPLC column) is collected in a loop of the transfer valve and then injected automatically into the second dimension system (GPC column). The two loops are filled and emptied into the second dimension alternately. The interval between two valve switchings is determined by the time needed for one second-dimension analysis. To keep the cycle times as short as possible, the second dimension flow-rate must be high. The first-dimension flow-rate is usually very low and can easily be calculated, dividing the loop volume through the above mentioned time interval (for 100% sample transfer). Using this procedure, the sample eluting from the first dimension will consecutively be fractionated and analysed completely in the second dimension.

There are three important technical improvements that can be used to "speed up" 2D chromatography: column technology, detection options and chromatography data systems.

Column technology: It is important to keep the volume that is going to be fractionated as low as possible, so this should be considered during the development of the first dimension separation method. Recent introductions of short HPLC columns with particles smaller than 2 μm might help to transfer existing methods with long columns and bigger particles to those that will reduce the volume consumption but retain the resolution.

The approach to speed up GPC separations is a different one. The size-exclusion separation is based on the accessible pore volume and if the porosity of the stationary phase is increased above a critical value, the matrix will no longer be stable. Therefore, the column volume should not be changed if a comparable resolution is to be achieved. In 2D chromatography, SEC is usually used as the second dimension technique that must be very fast to have a short interval between the transfer of two fractions. Normal analytical SEC columns (300 mm × 8 mm) are used with an eluent flow-rate of 1 mL/min, which leads to an analysis time of about 12 min. This requires a very low flow-rate in the first dimension (and, therefore, a very long analysis time for the complete experiment), since the storage loop of the transfer valve must not be overfilled before the next transfer to the second dimension takes place. This interval could be shortened by increasing the flow-rate of the SEC, but normal analytical columns have an optimum flow-rate of about 1 to 1.5 mL/min. Newer HighSpeed SEC columns can allow the flow-rate to be increased up to 10 mL/min, resulting in associated decreases in analysis time, but without a loss in resolution. This could be achieved by changing column dimensions to 50 mm × 20 mm and further optimization of the stationary phase. As the total column volume is similar, the quality of the separation is maintained.5

Detection options: Even if columns allow high-speed analysis, the detector must be able to detect the sample under the given conditions. For several reasons evaporative light-scattering detectors (ELSD) are often used for 2D applications. During detection, the eluent will be evaporated, which is a challenge at high flow-rates. This was often a limitation for high-speed measurements because the ELSDs were not powerful enough to evaporate many solvents with flow-rates higher than 3–5 mL/min. Most manufacturers of such detectors have optimized their instruments in recent years allowing the HighSpeed columns to be used at their optimum flow-rates.

Chromatography data system: Equally as important as the separation and detection is control of the transfer valve and automated data acquisition. It should also be possible to overlap the transfer injections, thus reducing the transfer interval and, hence, overall analysis time. Overlapping means that a sample is injected before the previous sample separation is complete. The separation mechanism in SEC causes samples to elute in a given region and allows approximately 1/3 of the elugram to overlap without any negative consequences.

Table 1: Comparison between "normal" and "high speed" 2D experiments.

Data reduction as the final step should not be underestimated. Comprehensive quantification of 2D raw data with all the calculations needed to determine the property distributions without adequate software can take several hours. Even if experimental time can be reduced, the data evaluation will be a bottleneck. Software that automatically constructs the contour plot, quantifies the results and allows analysis and reporting templates to be created is the last step needed to implement 2D chromatography as a tool for the routine analysis of complex polymers, such as in research or quality control.

An example of how a 2D experiment can be speeded up in the described way is given in Table 1.

The loop volume for this example is kept constant at 100 μL. By reducing the loop volume, resolution can be further increased. This will also increase the analysis time (half the loop volume results in double analysis time) and so in this instance a balance between good resolution and optimized analysis time should be chosen.

Figure 3

Results and Discussions

It is possible to decrease the time of a 2D experiment, including data evaluation, to approximately one hour. It can, therefore, be an attractive technique for quality control, for example. Figure 3 shows an example separation of two thermoplastic elastomers from different vendors. The routine analysis in the QC laboratory showed similar main and byproducts and comparable molecular weight results (Table 2) but one of the materials gave a poor final product when used.

Table 2: Molar mass results from conventional SEC, obtained with narrow PS calibration.

A 2D experiment with an HPLC method in the first dimension and SEC columns (HighSpeed, PSS GmbH, Mainz, Germany) in the second dimension could show differences in composition and molar mass (Figure 4). The molar masses are similar so separation could not be achieved with SEC alone. A narrow chemical composition distribution (CCD) of one product opposes a tetramodal distribution of the other product and explains the different behaviour. The expenditure of time for the optimized 2D experiments was just a third higher than for the conventional SEC and the results clearly justify the use of 2D for such purposes.

Figure 4

Conclusion

In 2D LC two established chromatographic techniques are hyphenated to allow the in-depth analysis of complex samples. In the past, the technique has had the reputation of being a time-consuming method for special applications. However, recent developments in column technology, detection and software have overcome these traditional limitations.

Martina Adler graduated in analytical chemistry from the University of Technoloy in Darmstadt (Germany) and she did her PhD thesis at the German Institute for Polymers (DKI). She has worked for PSS in the software development and support team since 2004.

Peter Kilz is leader of the PSS software development team. He received chemistry degrees at the University of Mainz (Germany) and the University of Liverpool (UK).

References

1. P. Kilz and H. Pasch, Coupled LC techniques in Molecular Characterization; in: Encyclopedia of Analytical Chemistry; R.A. Myers, Ed. (Wiley, Chichester, UK, pp. 7495–7543, 2000).

2. P. Kilz, Chromatographia, 59, 3 (2004).

3. S.P. Dixon, Biomed. Chromatogr., 20, 508–529 (2006).

4. H. Pasch and B. Trathnigg, "HPLC of Polymers", (Springer, Berlin, Heidelberg, New York, USA, 1997).

5. H. Pasch and P. Kilz, Macromolec. Rapid Commun., 24, 104 (2003).