HPC Enables New Approaches to Eliminate Waste Plastics
Scientists seek to improve on a natural enzyme used by plastic-eating bacteria
By Rob Johnson
According to recent research, a million plastic bottles are bought around the world every minute, and the number will jump another 20 percent by 2021. Most of these bottles are composed of polyethylene terephthalate (PET), which is a highly recyclable material. Unfortunately, many of the empty bottles never make it to the recycling plant. Instead, they end up in landfills or oceans and take many decades to degrade. One possible solution to this environmental problem originated in an unlikely place — a landfill in Japan.
Scientists there noticed the organic breakdown of waste bottles and explored further. As it turns out, Mother Nature started attacking the discarded bottles on her own. A spontaneously-occurring enzyme made it possible for bacteria to feed on the plastic. In turn, the “digestion” process breaks PET down into its constituent molecules. While this natural process offers incredible inspiration, much work remains to make it a viable way to reduce plastic waste on an industrial scale. The enzyme, while effective, still requires years to break down PET. A practical solution for commercial-scale use must accomplish the same task in a few days.
Enter Gregg Beckham, a Group Leader at the US National Renewable Energy Laboratory (NREL), H. Lee Woodcock, Associate Professor of Chemistry, at the University of South Florida, and John McGeehan, a Professor at the University of Portsmouth. With the help from their colleagues around the globe, Beckham, Woodcock, and McGeehan are tapping the power of high-performance computing (HPC) to seek a more effective enzyme which can decompose plastic bottles even faster. If successful, the leftovers from the bacteria’s PET dinner will offer the building blocks for new types of up-cycled materials which can serve a variety of uses.
While the concept seems straightforward, the process of identifying an improved enzyme is not. Woodcock describes the complexity of the process. “First, we need to understand the original enzyme, and how plastics interact with it. Access to HPC resources, to carry out molecular simulations, will aid our understanding and ultimately allow us to develop strategies for improving upon the naturally-occurring enzyme. A new enzyme must facilitate PET decomposition quickly to have practical application in real-world situations. We have a long way to go, but the research is promising.”
Currently, three extremely powerful HPC systems utilizing some of the latest Intel Xeon processors contribute their muscle to the task. The Texas Advanced Computing Center’s (TACC) “Stampede2”, NREL’s “Eagle,” and San Diego Supercomputer Center’s (SDSC) “Comet,” and other HPC systems each have a role to play in this monumental endeavor. “We could never accomplish a scientific project of this scope without the aid of these HPC systems,” Beckham said. “We are extremely thankful to these research facilities who support our efforts in finding new approaches for reclaiming renewable resources.”
Stampede2 ranks #17 on Top500.org’s November 2018 list of the most powerful HPC system worldwide. It achieves peak speeds of 18,309 teraflops thanks to 367,024 Intel Xeon Scalable (and Intel Xeon Phi) processor cores and Intel Omni-Path architecture in a system supplied by Dell. Eagle, NREL’s flagship HPC system designed by Hewlett-Packard Enterprise (HPE), achieves roughly eight-petaflop performance with the aid 4,228 Intel Xeon Scalable processors which put 76,104 cores on tap. SDSC’s Comet system architected with Dell is capable of 2.76 petaflops. Each of Comet’s 1,944 nodes contains a pair of Intel Xeon processors with 12 cores each, 128Gb or DRAM, and 320Gb of local flash memory.
Woodcock describes how the various HPC systems work together to help solve different pieces of the puzzle, “Comet has specialized hardware and management for running our shorter enzyme modeling simulations very rapidly. That information is passed on to multiple compute nodes on Stampede2 and Eagle. Each system is configured to use those data sets and perform very large-scale simulations over a longer timeframe.” The combined effort evaluates millions of possible enzyme configurations to identify a modification — or mutation — of the original enzyme which will make it far more efficient in breaking down PET. Once found, such an enzyme could be reproduced artificially to assist large-scale recycling.
Once the team’s effort enables the engineering of an optimal enzyme, the next hurdle is effective and large-scale implementation. Woodcock, McGeehan, and Beckham envision large PET processing reactors created at dedicated facilities or attached to existing recycling centers. There, collected and shredded plastic bottles can undergo the enzymatic degradation process using the new enzyme. Temperatures above 70°C will likely be optimal to help accelerate the PET breakdown process further and melt the plastics into a physical state that is more easily accessible to the enzymes.
The process must also be financially viable to encourage additional recycling. Because the PET components have the potential for upcycling into other materials, the researchers have already seen interest in their work from companies seeking the opportunity to reclaim the PET byproducts for other uses. Woodcock also notes the importance of collaboration with industry partners. “We focus on the research to make this approach possible, but we are also pragmatic. If the PET reclamation process can create a new business opportunity from discarded bottles, more and more plastic waste will find new uses. Everyone wins.”
Added Woodcock, “Ultimately, this is a problem no single person or team can solve. Fortunately, many other scientific teams around the world are pursuing this endeavor, bringing new ideas to the table, and offering different approaches to help address the plastic problem.”
Beckham echoes Woodcock’s optimistic and holistic view of their scientific endeavor. “We have barely scratched the surface at this point, although this work is quite promising because it tells us that there is room for improvement. In the Herculean global effort to reduce plastic waste, our research is by no means a silver bullet. The support of the global community will be required to tackle this problem — not just researchers, but also companies like Intel who are advancing the HPC technology which facilitates our work.” Continuing, he notes, “We must also reduce global production of single-use plastics and recycle those we do use. By reducing the amount of plastic packaging that winds up in landfills, every person on earth can contribute to the bigger solution.”
Rob Johnson spent much of his professional career consulting for a Fortune 25 technology company. Currently, Rob owns Fine Tuning, LLC, a strategic marketing and communications consulting company based in Portland, Oregon. As a technology, audio, and gadget enthusiast his entire life, Rob also writes for TONEAudio Magazine, reviewing high-end home audio equipment.
# # #
This article was produced as part of Intel’s HPC editorial program, with the goal of highlighting cutting-edge science, research and innovation driven by the HPC community through advanced technology. The publisher of the content has final editing rights and determines what articles are published.
