New Methods in Raman Spectroscopy Advance the Analysis of Polymers and Electrospun Fibers

Christian Pellerin, a professor in the Department of Chemistry at the University of Montreal in Canada, is using Raman spectroscopy and the most probable distribution method to quantify the orientation of polymers and electrospun fibers. Spectroscopy recently spoke with him about this work.

At the RamanFest conference this year, you discussed the concept and practical applications of a new procedure proposed by your group to quantify orientation of polymers by Raman spectroscopy based on the most probable distribution (MPD) method. Can you tell us about the MPD method and how your group started working on it? How is it different from the depol constant (DC) method, which was used by previous researchers for orientation quantification (1)?

Pellerin: My group studies a wide variety of materials by confocal Raman spectroscopy. In particular, we want to understand how to control the structure and the molecular orientation in electrospun fibers of various polymers. My PhD student Marie Richard-Lacroix realized that the range of samples that we could study and the orientation information that we could extract from the polarized Raman spectra were limited by the orientation quantification procedure itself. This is because the standard DC method, which gives access to the

 and

 orientation parameters, first requires determining the depolarization ratio of the band of interest from an isotropic sample. Unfortunately, it is impossible in many cases to prepare an isotropic sample, at the submicrometer level, with exactly the same chemical and phase composition as the oriented samples of interest. Examples of problematic samples include polymers containing stress-induced crystalline polymorphs or mesomorphous phases, and biological materials (such as silks) that are naturally oriented. The DC method also assumes that the depolarization ratio does not change with orientation, but most groups who have had the perseverance to use the "complete method" (including Lesko [2], Pe´zolet [3], and Michielsen [4]) have shown that this approximation is rarely accurate. Our own simulations (5) have shown that a small change of the depolarization ratio upon orientation can lead to severe errors on the orientation parameters.

Our main goal was to develop an accurate method that would be experimentally more convenient (no need to measure the depolarization ratio) and more flexible (no need for an isotropic sample). The MPD method is in fact quite similar to the DC method, except that it eliminates the assumption about the depolarization ratio and replaces it by the assumption that the orientation in the sample follows its "most probable distribution" — hence the MPD acronym. In practice, this means that we assume that the

 parameter is a dependant variable that takes the most probable value associated with the

 parameter. We think that sacrificing the knowledge about

 is usually acceptable since, most of the time, it is not required to get a good description and comparison orientation in samples.

What kind of results have you seen using the MPD method combined with confocal Raman spectroscopy?

Pellerin: We first simulated all possible combinations of

 and

 (each corresponding to a different orientation distribution of the molecules in a sample) and found that the calculated

 values were very close to the expected ones, even if the simulated samples had a distribution of orientation different from their MPD (5). This is good because it tells that results of the MPD method are reliable for most conditions. We then validated it experimentally using samples of polyethylene (PE), poly(ethylene terephthalate) (PET), and polystyrene (PS) that cover the largest possible range of orientation values (from isotropic to highly oriented), showing both parallel and perpendicular orientations, and with bands with a wide range of depolarization ratios. In each case, we quantified the orientation by confocal Raman spectroscopy using both the new MPD and standard DC methods and, finally, compared it with the reference values obtained with the help of another technique. This comparison clearly demonstrated that the MPD method could provide accurate

 values even when the DC method failed to properly describe the orientation (6).

Has this method been used to characterize any other polymers?

Pellerin: The MPD method expands the range of materials for which we can easily characterize orientation by Raman spectroscopy while concurrently improving the accuracy of the results. We have chosen to demonstrate its experimental validity with the help of PE, PET, and PS because we had access to published orientation values that were directly comparable to our own results. Also, we had a specific interest in determining the orientation in PS and PET samples for another project on electrospun fibers. Since its development, we have used the MPD method to characterize orientation in electrospun nanofibers composed of poly(ethylene oxide) (a highly crystalline polymer) and poly(methyl methacrylate). We also studied other types of materials such as self-assembled fibrils of a polyferrocenylsilane-polyisoprene copolymer, byssal threads from the blue mussel, and more. In general, I believe the MPD method should be applicable to most systems composed of polymers and small molecules that show partial order. On the other hand, it would not be applicable to samples such as single crystals for which the orientation is anything but the most probable one.

In another paper (7), you discuss how confocal Raman spectroscopy can be used to characterize the molecular orientation and structural characteristics of individual electrospun nanofibers. These fibers are usually characterized by techniques such as X-ray diffraction and IR spectroscopy. What advantages did the Raman spectroscopy approach provide compared with the traditional techniques?

Pellerin: Raman spectroscopy is a very powerful technique thanks to the rich molecular information in the spectra combined with its capability of studying individual fibers. Because of their small size (diameters typically range from a few hundred nanometers to a few micrometers), electrospun fibers are usually characterized as bundles of thousands of fibers when using conventional techniques such as X-ray diffraction or infrared (IR) spectroscopy. The problem is that in a typical bundle, the fibers are not perfectly aligned, there is often a wide diameter distribution, and there may be several defects. From the point of view of orientation, the imperfect alignment means that the

 values are systematically underestimated. As an extreme example, a random mat always appears to be isotropic when studied as a bundle even if the polymer orientation is in fact very high.

More generally, it is often hard to disentangle the impact of various parameters when studying bundles. For instance, it is difficult to understand the impact of the processing conditions on the orientation (or any other structural parameter of interest such as crystallinity) if it simultaneously changes the fibers diameter, alignment, and defect content. Atomic force microscopy (AFM) is one technique that allows separating such effects by measuring mechanical properties of individual fibers. It has revealed a strong diameter-dependence of the elastic modulus that may be due to increased orientation. Confocal Raman spectroscopy has the great advantage of providing structural information at the molecular scale for individual fibers. In contrast to electron diffraction, which also enables studying single fibers, it gives access to information on both the crystalline and amorphous phases of materials, on the individual components of blends or composites, and so on. Confocal Raman spectroscopy currently enables us to rapidly study large quantities of individual electrospun fibers to improve our understanding of these materials.

What are the next steps in your work?

Pellerin: We seek to improve our fundamental understanding of the structure of electrospun fibers and to better control their properties. Confocal Raman microscopy and the MPD method now enable us to quantify the orientation of individual fibers with diameters down to 500 nm. We are currently studying the impact of the fiber diameter on orientation and trying to establish a correlation with the mechanical properties determined by AFM on individual fibers. The following steps will be to determine how the orientation is influenced by the nature of the polymer, in particular its propensity to crystallize, and by various processing conditions.

References
(1) S. Frisk, R.M. Ikeda, D.B. Chase, and J.F. Rabolt, Appl. Spectrosc.58, 279−286 (2004).
(2) C.C.C. Lesko, J.F. Rabolt, R.M. Ikeda, B. Chase, and A. Kennedy, J. Mol. Struct.521, 127−136 (2000).
(3) M. Pigeon, R.E. Prud’homme, and M. Pézolet, Macromolecules24, 5687−5694 (1991).
(4) S. Yang and S. Michielsen, Macromolecules36, 6484−6492 (2003).
(5) M. Richard-Lacroix and C. Pellerin, Appl. Spectrosc.67, 409­–419 (2013).
(6) M. Richard-Lacroix and C. Pellerin, Macromolecules46, 5561­–5569 (2013).
(7) M. Richard-Lacroix and C. Pellerin, Macromolecules45, 1946­–1953 (2012).