Now you see me and now you don’t: GPC analysis of polysiloxanes

Introduction

Gel permeation chromatography (GPC) is at the heart of polymer characterization. This characterization technique is implemented throughout the life of the material.This can include after initial polymerization, before and after purification, after post-polymerization modification, in the final formulation, after aging, failure and/or degradation. The primary focus of analyzing polymers using GPC is to characterize their molecular weights and molar mass distributions. There are many motivations behind determining the molecular weight including predicting how the products' properties will be affected and determining how well a reaction went. In addition to molecular weight, scientists also want to know about changes to the sample’s structure, branching, size, solution viscosity and compositional changes; the OMNISEC multi-detection GPC system provides this information with the highest level of sensitivity. The OMNISEC uses a combination of refractive index, light scattering, UV/vis spectrometer and viscometer detectors to provide a far more complete characterization than the relative molecular weight results provided by using only RI detection. 

Polysiloxanes are an interesting material to analyze by GPC, as if one tries to analyze the most common polysiloxane, polydimethylsiloxane (PDMS), in the standard organic solvent for GPC, THF, the peaks in the RI are extremely small and none existent in the light scattering detectors. However, when analyzed in toluene the PDMS signals in both of these detectors are intense. This is due to the refractive index increment (dn/dc) of the material (for further details on the dn/dc value please read this interesting blog post)

Polysiloxanes are widely used for many applications as they are very chemically stable, unreactive, hydrophobic, have good electrical insulation and can form a variety of material types, such as elastomers, gels, lubricants, foams, and adhesives. The chemical functionality incorporated into a polysiloxane can be chosen to convey specific properties to the polysiloxane. 

The polysiloxanes analyzed in this study were provided as a part of collaboration with Michal Grabka of the Military University of Technology in Warsaw, Poland who is investigating the use of polysiloxanes in military chemical warfare agent detection devices. Polysiloxanes are used in surface acoustic waves-based (SAW) sensors and quartz crystal microbalance (QCM) comprised of thin sorbent layers of a chemically selective material to collect and concentrate the analyte of interest. The collection of analyte increases the mass of the sensor, resulting in a change in resonant frequency. Chemoselective polymers for nerve agent detection should be both strongly dipolar/ polarizable and hydrogen-bond acidic to complement the dipolar and hydrogen-bond basic properties of nerve agents. Thus, polysiloxane functionalized with difluorophenol groups could be a promising sensitive material for nerve agent detection, and it may also exhibit a significantly lower response to water molecules because of the effect of polarizable aromatic groups. 

Michal Grabka is investigating the synthesis of substituted polysiloxanes with the goal of testing their suitability for chemical warfare agent detection. The starting material (Figure 1 – sample 1) is a copolymer of PDMS-co-poly(methylhydrosilane), purchased from a supplier, that was reacted in a hydrosilylation reaction to functionalize the polysiloxane with pendant fluorinated and non-fluorinated alkyloxyphenol groups (Figure 1 – samples 2 and 3). The main focus of Michal Grabka was to investigate how the polymer was altered during the reaction in terms of molecular weight, molar mass distribution, size and structure using GPC. 

Figure 1. Chemical structures of the 3 samples analyzed: starting material polysiloxanes copolymer (sample 1) and functionalized polysiloxanes copolymers (samples 2 and 3).

The chemical structures of the three samples analyzed and discussed in this Application Note are shown in Figure 1. PDMS is the main component of all three samples as it is 7:3 with the hydro and alkoxyphenol substituted monomeric units. 

Method

The three samples were measured by multi-detection GPC on a Malvern Panalytical OMNISEC system including refractive index (RI), UV-Vis, light scattering (right-angle light scattering (RALS) and low-angle light scattering (LALS)), and viscometer (IV) detectors. The samples were dissolved to concentrations of approximately 3 mg/mL. The samples were prepared and analyzed using the OMNISEC in two different solvents: 1) toluene, 2) THF. For analysis in toluene, the samples were separated using three linear mixed bed columns.

Results 

Analysis in toluene

Sample 1 

Figure 2. Multi-detection chromatogram for polysiloxane copolymer sample 1 analyzed in toluene. Each color represents the responses from a different detector; Red: Refractive index detector, Blue: Viscometer, Green: Right angle light scattering detector, and Black: Low angle light scattering detector.

The data acquired from the multi-detection system provides a detailed characterization of the sample in terms of absolute molecular weight, IV and size calculated at every data slice. This information can be combined to produce the graph that is presented in Figure 3. The sample’s characteristics are graphed in Figure 3 as a function of Log(molecular weight); the pink graph is the molar mass distribution, the light blue plot is Log(IV), and finally, the sample’s hydrodynamic radius is plotted logarithmically in olive green. The IV is a directly measured parameter that is an inherent property of the molecule under study. Since IV is a function of the molecule’s structure, it can be used in combination with molecular weight to comment on the structure of the molecule. It is inversely proportional to density and is, therefore, an indication of the compactness of a molecule in the solvent. The hydrodynamic radius (Rh) of the samples is calculated using the light scattering and viscometer detectors.

Figure 3. Overlay of molecular weight distribution (purple), hydrodynamic size (olive green) and intrinsic viscosity (light blue) distributions for the polysiloxane copolymer sample 1.

Samples 2 and 3

Multi-detection chromatograms produced as a result of analyzing polysiloxanes sample 2 and sample 3 in toluene contrasted with that of sample 1. Figure 4 presents the overlays of RI (left) and RALS (right) chromatograms for samples 1, 2 and 3 analyzed in toluene. From these overlays, it is clear that samples 2 and 3 neither refracted nor scattered light. This is indicative of the change in the chemical composition of the polymers after the hydrosilylation reaction and the new functional groups attached to the polysiloxane backbone leads to it being iso-refractive in toluene, i.e. the dn/dc of samples 2 and 3 in toluene is approximately zero. 

Figure 4. Left: An overlay of the RI chromatograms produced from the analysis of samples 1 (red), 2 (green), and 3 (purple) in toluene, the peaks after ca. 20 mL represent the permeation peaks. Right: An overlay of the RALS chromatograms produced from the analysis of samples 1 (red), 2 (purple), and 3 (green) in toluene.   

Analysis in THF

All three samples were prepared and analyzed in THF. As hypothesized the newly functionalized materials gave clear, strong signals in each of the detectors. Sample 1 was not detected in the refractive index or light scattering detectors due to it having a dn/dc value of approximately zero in THF. The multi-detection chromatogram for sample 3 is shown below in Figure 5, and it was noted that both samples 2 and 3 had an excellent signal to noise ratios. The potential presence of large materials (impurities) in the samples as a side-product of the functionalization reaction, can also be detected thanks to the light scattering detector. In sample 3 the impurities were in fact identified from the peaks in the RALS and LALS from 7 to 10 mL (Figure 5). The absence of these peaks in the RI chromatogram indicates that they are low in concentration. These large materials are not present in sample 2 and this indication of high molecular weight impurities will likely help in the selection of the most suitable functionalization conditions. 

Figure 5. The multi-detection chromatogram of sample 3 analyzed in THF.

The additional functional groups added to samples 2 and 3 contain aromatic ring which absorbs UV light. The PDA UV/vis detector of the OMNISEC Reveal provides an additional method to determine the presence of these new functional groups. Figure 6 illustrates an overlay of the UV absorbance chromatograms at 250 nm produced by the three samples analyzed in THF. The unfunctionalized sample 1 does not show any absorbance of UV light at 250 nm whereas the samples that were functionalized have strong absorbance. This indicates the presence of the desired functional groups attached to the polysiloxane backbone. 

Figure 6. An overlay of UV absorbance at 250 nm chromatograms produced from the analysis of samples 1 (green), 2 (red), and 3 (purple) in THF. 

Quantitative results 

Having found the appropriate solvents to analyze the three different polysiloxanes, as determined by their optical properties, the samples were processed using the OMNISECTM software. Thanks to the multi-detection calculation method the samples weight-average molecular weight (Mw, g/mol), number-average molecular weight (Mn, g/mol), dispersity (Ð or Mw/Mn), intrinsic viscosity (IV, dL/g), hydrodynamic radius (Rh, nm), and dn/dc (mL/g) were determined. The average values of duplicate injections are presented in Table 1. The results in Table 1 show that the samples greatly differ from one another and provided a detailed picture of how the functionalization reaction impacted the polymers molecular characteristics. 

Table 1. Quantitative results from the analysis of samples 1, 2, and 3. The data was processed using a triple detection calculation method. Sample 1 was analyzed in Toluene, Sample 2 and 3 were analyzed in THF

Conclusions

This application note has shown that a material’s chemical composition plays a very important role in determining the conditions for GPC analysis. The different functionalities present on a polymer chain can deeply affect the detectors’ response to the sample. It was shown that a copolymer that was primarily consisted of PDMS (sample 1), when functionalized with alkylphenol groups (samples 2 and 3), had a dramatic change to the polymer solutions’ optical properties and thus required the use of THF for the GPC analysis. Whereas, the parent polymer (sample 1) required the analysis to be conducted in toluene due to its isorefractive nature in THF. The OMNISEC system revealed a change in solution optical properties and UV absorbance in samples 2 and 3 and provided a good indication of the successful functionalization of sample 1 to form samples 2 and 3. In addition, thanks to the light scattering detector it was possible to identify high molecular weight impurities within a sample after functionalization reaction, helping the researcher to identify the best conditions for the modification reaction of the parent polymer (sample 1). The complete characterization of the three samples illustrated that the samples greatly differ from each other in a multitude of ways and provided a key insight into how the polymer has been impacted by the postpolymerization modifications.