Discrimination of Plastic Waste Pyrolysis Oil Feedstocks Using Supercritical Fluid Chromatography: An Interview with Kevin A. Schug

The advancement of chemical recycling techniques has led to new ways to handle and recycle mixed plastic waste. Among these processes is pyrolysis, which involves the heating of plastic waste in an oxygen-free environment to create pyrolysis oils. Kevin A. Schug of The University of Texas at Arlington and his team partnered with The Lummus Technology (Pasadena, TX) to develop a method based on supercritical fluid chromatography (SFC) with ultraviolet (UV) detection. The goal was to use this method to further advanced chemical recycling techniques and to provide new avenues for handling and recycling mixed plastic waste. Schug spoke to LCGC International about the process and the paper that resulted from its development.

Your paper (1,2) details a method based on supercritical fluid chromatography (SFC) with ultraviolet (UV) detection which developed to analyze plastic waste pyrolysis oils. Briefly discuss what pyrolysis oils are and why do they merit analysis.

Pyrolysis oils are created from mixed waste plastic (MWP) by heating them to a high temperature in an oxygen free environment. This causes the polymer chains in the plastic to decompose into smaller hydrocarbon segments, which can be further processed. The creation of pyrolysis oils is the most promising approach currently for advanced chemical recycling of MWP, because it is not limited to a particular plastic type. Traditional mechanical recycling can only be applied to a limited number of plastics (Type 1 and 2) and only a limited number of cycles can be performed before plastic quality degrades beyond reuse. Pyrolysis of MWP avoids the down-cycling of plastic quality, which accompanies mechanical recycling, and the resultant oils can be further refined into fuels and chemical feedstocks, which can be used to create new plastic materials. Thus, a big advantage of pyrolysis of MWP is to reduce reliance on hydrocarbon fossil fuel streams for creating new plastics. A challenge comes in the complexity of generated pyrolysis oils from MWP. Because the feedstock is highly variable, so is the composition of the generated oil. The oil must be further refined to be used, and some catalytic processes are susceptible to degradation from different oil components, such as those containing sulfur. Also, compared to a typical fossil fuel hydrocarbon stream, pyrolysis oils can have a much higher olefinic character. The variable composition and content of pyrolysis oils requires the development of new analytical methods, which can help guide their subsequent processing into usable products.

In the paper’s introduction, you state that gas chromatography (GC) has been the primary tool utilized when trying to identify the composition of the different types of this oil. Why did you feel a need to improve on what was already being used, and why specifically use SFC?

Gas chromatography has been, and still is, the workhorse for hydrocarbon stream characterization. Constant development of multidimensional gas chromatography technologies has expanded what is possible. However, pyrolysis oils can contain some significant high molecular weight and heteroatom functionalized species that are challenging for GC, because of their limited volatility. SFC has more versatility in handling higher molecular weight species, and it can be used to effectively pre-fractionate pyrolysis oils to reduce sample complexity for subsequent analysis. SFC has a multitude of column chemistries, which can be used to affect separation of different polar and semipolar species. It can be used to create profiles and fractions that highlight different components of the pyrolysis oil mixtures for more efficient characterization.

The research and the resulting paper are a joint venture between The University of Texas in Arlington and The Lummus Technology, LLC in Pasadena, Texas. What does Lummus bring to the table for this project?

Lummus Technology, LLC was really the initiator of the project. They have been working with other companies to help develop variable scale industrial processes, including pyrolysis systems. They have developed and use many different methods for the characterization of pyrolysis oils, but they have still found it to be a formidable challenge. Since we have both SFC and GC-VUV technology, they decided to partner with us to see if we could help develop some new methods. They have been instrumental in providing us with a variety of samples, as well as in guiding our thinking for what are the most important challenges to address with the topic. Working closely with an industrial partner is important for the project and for the students. Lummus has provided important direction in our work and a source of collaboration to make sure what we are doing is meaningful and useful.

Working with SFC rather than GC, were there any specific changes in the analysis involved, and were there any challenges associated with making these changes?

SFC can handle a wide range of compound types, from nonpolar to highly polar compounds. There are a wide variety of stationary phase chemistries that can be utilized to create different separation profiles. Our initial intent was to use SFC to fractionate pyrolysis oils for subsequent GC-VUV analysis, to provide better characterization of PIONA classes and heteroatom content in the samples. We are still working on that. Along the way, we found we could fingerprint oils simply using SFC-UV. To be able to create a useful method with something as simple as SFC-UV makes it more universally available. There are still major challenges with the variability in composition and the physical properties exhibited by the oils. Some oils can look as clean as castor oil, while others resemble solid tar. To create a universal means to prepare a sample for injection is a big challenge and is one that we are still working to develop. Some pyrolysis oils appear to contain large asphaltene type compounds which can be difficult to handle. These need to be precipitated prior to analysis.

Other than the technique and equipment being used, does your work differ from what has been previously done by yourself or others?

I think the notion of using SFC as a sample preparation technique, to reduce the complexity of oils for subsequent analysis, is a concept that can be more broadly applied to complex mixtures. Most of the previous work on separating MWP pyrolysis oils has focused on GC and multidimensional GC. Bringing SFC to the table makes sense, because of its rich history for use in characterizing hydrocarbon streams. When you add to this the capabilities of more modern instrumentation, including fraction collection and mass spectrometry, you get a very versatile tool, which should find greater use in pyrolysis oil analysis.

Briefly state your findings.

In this work, we were able to show that traces of separations using SFC-UV could be used to differentiate pyrolysis oils generated from polyethylene and polypropylene feedstocks. This was consistent across several different samples of each, and we could even create mixtures that we could also well differentiate from the pure oils. The potential to couple multiple SFC columns in series allowed us to increase our plate capacity, while taking advantage of the beneficial properties of SFC separations, relative to GC and HPLC. This is an important first step in developing technology for better determination of feedstock types for MWP pyrolysis oils.

Do your findings correlate with what you had hypothesized?

It became quickly apparent that we could get different SFC-UV profiles for different pyrolysis oils, especially as we investigated different column chemistries. We believed that if we could obtain oils from different pure plastics, it would be possible to create a database from which we would eventually be able to perform mixture analysis. The ability to differentiate PE and PP using this technique confirmed our initial thoughts, but we will need to obtain other pure oils to expand the capabilities and make it a more broadly usable strategy.

Was there anything particularly unexpected or surprising that stands out from your perspective?

Because SFC-UV is blind to the saturated hydrocarbons, which make up a large portion of the sample, we could focus our oil fingerprinting on those species that were more unique to the oils. It is still quite difficult to determine what these species are, with UV detection alone, but the simplicity of the approach is certainly a benefit. We did not expect to see such a possibility when we first started the work.

What best practices can you recommend in this type of analysis for both instrument parameters and data analysis?

Be careful with what you inject in your instrument, be it SFC, GC, or HPLC! With complex mixtures there are always components with a range of solubilities. The complex mixture somehow exists in a semi-homogeneous state because of the large variability of its constituents. When you start injecting and separating components, you can find that many do not behave well, and they can contaminate the instrument. It is important to survey a range of solvents to mix with the oils to see what kinds of changes in solubility you might observe before performing instrumental separations. This can ultimately lead to an effective and important pretreatment step for the samples before they are injected into your high efficiency separation systems.

Can you please summarize the feedback that you have received from others regarding this work?

There is an enormous abundance of plastics that enter our environment and have deleterious consequences. We have received universally positive attention for contributing one small part to a possible solution – that is, the potential to help characterize MWP pyrolysis oils. Eventually, advanced chemical recycling will enable much better handling of waste plastics, as well as reduce reliance on the creation of virgin plastics from fossil fuel sources. These are both good things to work toward, since our society has an insatiable desire for more and more plastic products.

What are the next steps in this research?

It would be most interesting to be able to get more pure oils from specific polymer feedstocks. This is challenging, since most industry producers of pyrolysis oils are trying their best to handle MWP, not single polymer types. Still, if a database of fingerprints for pure oils could be established, then it would not be so difficult to back-calculate the compositions of a mixed feedstock. This would further aid choices in downstream refining of the oils. Additionally, we would like to advance more on-line and off-line multidimensional separation techniques, especially the utilization of different detector technologies to allow better characterization of components.

Kevin A. Schug of The University of Texas at Arlington.

References

1. Kaplitz, A. S. et al, Discrimination of Plastic Waste Pyrolysis Oil Feedstocks Using Supercritical Fluid Chromatography, J. Chromatogr. A 2024, 1720, 464804. DOI: 10.1016/j.chroma.2024.464804

2. Science X staff page. MSN.com : https://www.msn.com/en-us/news/technology/a-research-team-is-developing-a-method-to-recycle-more-plastics/ar-BB1mShRp?ocid=BingNewsSearch&apiversion=v2&noservercache=1&domshim=1&renderwebcomponents=1&wcseo=1&batchservertelemetry=1&noservertelemetry=1 (accessed 2014-06-24)