Optimizing profitability through maximizing materials efficiency

Designing with data creates value - at a rate critical to inform technical and business innovations in materials recovery

Written by Kezi Cheng & Peter Christensen

• Harvard School of Engineering and Applied Sciences, kezi_cheng@g.harvard.edu

• Co-Founder | FLO.materials, pete@flomaterials.com

03/06/2020

Abstract 

Since the industrial revolution, process and resource optimization have increased manufacturing profitability. Automation has lowered costs by creating and moving more product per unit time. Additive manufacturing techniques such as 3D printing have optimized profitability by reducing time to market, while also enabling hyper-localized, low-volume (customizable) production that reduces both distance and time between producers and consumers. As we continue to advance towards producing optimal quantities at lower costs, higher speeds, with faster delivery times, we should also consider waste output (including end-of-life products) as a critical area for optimizing profitability. Here, we discuss several key factors surrounding material circularity that make it difficult to optimize for profitability. We suggest that in order to close the gap between profitability and maximum materials efficiency, it is necessary to create a careful balance between material demand, material recovery, and product lifetime.

While our discussion generally applies to many materials and products, we will commonly use plastics as a particularly problematic and broad archetypal example. We discuss factors including inferior and inconsistent quality of recycled material, low cost of virgin materials, low and inconsistent supply of recycled material, complicated and costly recovery process and transportation as well as a lack of shared information between stakeholders. Looking to the future, we highlight select recent advances ranging from fundamental academic research to promising startups and commercial developments paving the way to profitability through material circularity. Our discussion is focused around the major components of a circularity ecosystem: economics, research, industry, and policy.

Introduction

In the last century we have seen an unprecedented increase in the use of natural resources and materials. While feedback in nature is continual, industrial systems are not. Industries ingest energy, metals and minerals, water, forest, fisheries, and farm products. It excretes liquid and solid waste - variously degradable or persistent toxic pollutants - and exhales gases, which are a form of molecular garbage.[1] Materials efficiency, simply put, is the output of waste over the input of resources. The OECD finds that global raw material usage rose at almost twice the rate of population growth.[2] As material demands have soared, the consumption and depletion of raw resources stock has increased, as well as the generation of by-products, including waste, wastewater, air pollution and greenhouse gases per unit of output. The need for increasing materials efficiency, reducing the quantity of inputs needed to produce a unit of output, by doing more with less, becomes more urgent than ever before.

The basis in which materials become restorative and resource efficient in its relation to society moves the economy from a linear to a circular-based approach. The circular model is a comprehensive framework where the design of the product, the production process, the distribution channel and every aspect of a firm's life is around a minimal use of new resources and reuse of existing ones.[3] This requires making intentional choices of what the resource inputs in consumption and production systems are, what exchange of waste material will become feedstock for another, and which materials to recover at the end of their life.[4]  But, is this circular model economically and environmentally sustainable? How can we extract value from a process that inherently loses value over time and use? Who benefits most from the implementation of circular systems? These questions tap into potential solutions around efficiency, productivity, and innovation, which can quickly make a growing concern into a rich opportunity.

Recycling innovations do not fully solve the circularity challenge

In order to extract greater value out of a materials system, it is important to clearly identify the stakeholders involved, and think of them as stationary points in which materials flow through from cradle to grave. These hubs that represent firms and individuals include: mining, extraction, design, production, manufacturing, transportation, collection, use, disposal, recycling. What do these individual hubs gain or lose by carrying out business as usual, and what incentives are needed for each of them to make adjustments to their businesses to reach for greater circularity?

In the case of plastic waste, Beckman discussed various reasons for the low rate of recycling.[5] One is an emphasis on customization where the design of materials and products for particular applications or customers inhibits collection, separation, and reuse of the materials. The second is contamination of recyclable polymers with other substances. The third is complexity where products such as electronics that are often made with multiple polymers, additives, colors, and other materials, make recycling challenging or impossible. Tom Graedel, a notable industrial ecologist, has shown that a typical mobile phone now uses more than two thirds of the periodic table of elements.[6] Even when recycling is feasible, it is difficult to increase demand for recycled materials when the cost of virgin resources is significantly lower and more stable. In addition, collection and processing infrastructures aren’t designed to take into account the customization, contamination, and complexity of the waste streams. Moreover, the post-consumer and post-industrial worlds have very different recycling needs where the former requires more sorting and separation of complex streams, and the latter requires higher standards in integrity, and transportation associated with larger volume and mass. This makes circularity an extremely difficult problem to solve holistically.

Although plastic recycling is difficult and current rates are low, different technological innovations in mechanical, chemical, gasification, incineration, solvent-mediated additive removal dissolve in solvent, and pyrolysis (feedstock recycling) offer promising ways to “close the loop” by recovering materials and thereby reducing the need to produce new materials. Brian Riise, a project manager at the REMADE Institute, stated that mechanical recycling is particularly attractive because it uses only 10–20% of the energy required to make virgin plastics.[5] However, several requirements such as purity in appearance and additives, as well as the ability to mix recycled pellets with virgin materials makes reprocessing nonetheless difficult.

Chemical depolymerization, which converts polymers to monomers, is another potential way to improve recycling capabilities and appears to be practical for some polymers, such as polyethylene terephthalate (PET), polystyrene, and recently developed poly(diketoenamines) (PDKs). Researchers are pursuing various approaches such as chemical strategies to “attack” the bonds that hold the polymer together (Fukushima et al. 2011, 2013), enzymatic approaches to transform a polymer into useful building blocks (Yoshida et al. 2016), thermolytic techniques to convert polyolefins into liquid fuels (Wong et al. 2016), chemical methods for depolymerizing polyolefins (Williamson et al. 2018).[5] Living organisms can also be used to biologically degrade materials (Yang et al. 2015). In addition, biocatalysts such as microbial/fungal enzymes have natural hydrolase that breaks down the outer layer of leaves and has been shown to degrade PET (Ribitsch et al. 2015). According to Geyer et al., of the 6300 Mt of plastic waste generated to date, only 9% was recycled by these methods, 12% was incinerated, and 79% accumulated in landfills and the natural environment.[7]

Figure 2: Cumulative plastic waste generation and disposal (in million metric tons).[7]

How do we intentionally design materials of tomorrow to increase efficiency?

While we can continue to innovate to increase materials efficiency through different methods of recycling, we are innovating around solutions at a slower rate than we are creating new problems. This issue is readily understood by the analogy to a leaky faucet. If all it took to stop the water was to turn off the faucet, then the logical thing would be to do that first and then mop up the floor. But if the leakage from the faucet required time and tools to fix, then which step you choose to do first would depend on how long it takes to fix the faucet versus how fast the water is leaking. But both are equally important and must be approached in combination to resolve the problem. Technological innovations to decompose materials sustainably take almost as much time to develop as designing new material systems from scratch. Therefore, to tackle the current problems while preventing future problems, we need to reinvent with design and reinform with data for both technological and business model innovations.

The commercialization of new material discoveries can take anywhere from 10 to 20+ years.[8] With such protracted development cycles, if we are not taking the appropriate steps to combine designing for desired properties and applications for the first life, as well as for the second or third life, then we will never catch up with the problems that are created. But in order to design intentionally, we need to know where the material will be at the end of its first life and where it will be most valuable at the beginning of its second life. By collecting data on the time, place, and rate of material degradation, we can use this data to reinform design considerations. To promote industrial symbiosis whereby byproducts of one process become inputs of another, material flows should be monitored throughout the supply chain and data analytics should be made available to all stakeholders, including consumers. While the material value is determined by quality, appearance, purity, and integrity; design and data around the material can be valuable, intangible assets. However, even if we can accurately track material flows throughout the supply chain, inherent design flaws in products and the materials that go into them (especially for plastics) result in downcycled materials with inferior aesthetic and mechanical performance.

Plastics often contain multiple types of additives such as colorants, flame-retardants, stabilizers, plasticizers, etc. These complex mixtures of additives persist in recycled plastics, contaminating the material and limiting reuse opportunities. Furthermore, the incumbent method of mechanically grinding and re-melting plastics physically degrades the material, resulting in a lower-value product with unpredictable properties. To address these inherent material design flaws, new solutions are being researched. Among the more promising candidates for recovering the highest value in recycled materials is a process known as “chemical depolymerization” (aka. monomer-to-monomer recycling), where the recycling process turns the plastic (polymer) back into the building blocks (monomers) to rebuild the plastic in its next life. By recovering an upstream precursor to the material in question, the same processes that make virgin materials can be used, allowing recycled materials to retain the highest aesthetic and engineering standards throughout multiple lifecycles.

Figure 3: General scheme showing monomer to monomer recycling.

Towards monomer-to-monomer recycling, advances in new chemistry (Garcia et al. 2014;[9] Zhu et al. 2018;[10] Christensen et al. 2019[11]), as well as commercial efforts to depolymerize existing polymer waste (BioCellection, Loop Technologies) are paving the way towards a true, circular economy of plastics involving high throughput and efficient recovery. Monomer-to-monomer recycling allows recycled plastics to achieve virgin-quality properties, incentivizing circular life cycles by maximizing the value of discarded material. However, a circular economy of anything is only as efficient and reliable as the ability of the ecosystem to support and incentivize circular flows of material. In other words, without an ecosystem of companies that can achieve profitability in this space, striving for greater material efficiency will not economically succeed.

How can we track and monitor material flows?

By tracking materials as they move through the supply chain from raw unrefined ingredients, to finished goods, and back to raw materials, this valuable data will inform decision making that enables both material efficiency and optimization for profitability. Furthermore, if material flows can be shared in an open platform, available to all stakeholders in the value-chain, critical inefficiencies in material circularity can be optimized.  While this data will be useful for optimizing the status quo materials ecosystem, it may also be possible to use this data to design new materials by highlighting inefficiencies. For example, the quality of a product could be enhanced through the addition of the right additives or compatibilizers to create a blend of miscible plastics if information about the composition of the waste stream is well understood. 

A pressing concern for municipal recycling facilities (MRF) in the U.S. is the separation and recycling of black plastics. Today, 15% of the plastic in recycling centers is black, and the majority of that is single-use food containers.[12] Plastics that are black in color are the hardest to recycle due to inefficient sorting technology. Black plastics are commonly colored using carbon black, an additive that can pose significant health hazards. Carbon black is used as a colorant in food packaging for a number of reasons because of a contrasting background, and it allows the colors in the food to stand out. Carbon black is low cost, has good dispersion and masking properties, which allows other colors to be mixed together and manufactured into black items. The reason that black plastics cannot be easily sorted in the current MRF is because the near infrared radiation (NIR, 800-2500 nm) used by optical sorters gets absorbed by carbon black, making it difficult to identify and separate streams. Black plastic either gets dirtier or ends up in unsorted residue that goes to a landfill. Unfortunately, our current methods of recycling them is like a polluted stream that cycles back into itself, causing a quasi-circular economy.

I saw the black plastic problem firsthand at one of the newest MRF that opened in 2013 in Brooklyn, New York and serves 8.6 million people, cost $120 million to build, and has 16 advanced optical sorters. Innovations in finding compatibilizers to separate out carbon black and adding another colorant[13] that is easily detected by NIR could be one solution to this problem. Another could be incorporating technological innovations using different optical techniques to separate out all of the black plastics, such as long-distance Raman (Pendar) or photon up conversion for mid-infrared radiation (MIR) detection.[14] In addition to detection, deep learning technology and chemometrics[15] combined with robotic sorters enabled with machine vision could be a streamlined process in separation and sorting over high cost spectroscopy techniques. Examples of these technologies include AMP Robotics and Zen Robotics. RoCycle, a robotic system with tactile sensors to detect if an object is paper, metal, or plastic came out of MIT CSAIL and is able to determine at 85% accuracy when a material is stationary, and at 63% accuracy on an actual simulated conveyer belt.[16]

However, while all of these approaches are efficient at sorting and generating useful data, a combination of several of the above approaches could be potentially transformative. An example to tackle black plastic could be a 4-step process to allow contaminated materials to be recycled in high purity. First, given the current MRF sorting processes that use NIR are very efficient, an add-on of a simple vision system (camera) to current optical sorters that separates black and all other colors could be the right solution. Second, black plastics could be collected and brought to a facility for MIR or Raman detection into different streams to identify impurities or composites. Third, segregation and removal of the identified carbon black would allow the majority of the remaining material to be recycled in high purity. Finally, the addition of detectable markers or other black colorants at the manufacturing step would allow for easier detection by current optical sorters. This would allow black plastics in the current stream to get cleaner and cleaner as opposed to more and more contaminated. Finding simple ways to retrofit existing technologies to collect data in production, manufacturing, and recycling facilities regarding material composition, integrity, and properties could lead to invaluable informatics that become the materials’ intangible assets.

Figure 4: An ecosystem throughout the plastics value chain to monitor and optimize circular flows of material. The ecosystem is a network of partnerships, consisting of several “hubs” (materials, design, manufacturing, logistics, brand, infrastructure) working together, with an interconnected open data stream.

One of the most widely used approaches to understanding the environmental impact of an industrial process is called a life-cycle-analysis (LCA). LCAs are particularly useful tools for understanding and minimizing CO2 emissions, water usage, waste generation, and even capital costs if properly defined. At a very high-level, LCAs account for physical (e.g., materials and energy) inputs and outputs (products and CO2 emissions) around an arbitrarily defined “system boundary”. The granularity of the analysis depends on what is included in the system boundaries. For example, an LCA could highlight low emissions and water use for a new manufacturing process, but it may fail to capture the emissions and water cost of handling the end-of-life products produced using this process. To this end, LCAs will provide a very useful framework for optimizing circular material systems if material waste and end-of-life products are also captured as material inputs. Only if the inefficiencies of circular systems are revealed will we be able to understand how to efficiently optimize them. If we can accurately track material flows throughout the supply chain, this data could pair very well with an LCA-driven machine-learning or generative design platform that self optimizes for profitable circular material flows.

Integrating new technologies with tracking capabilities to solve the bottlenecks around distribution and collection infrastructures is an active, nascent research area. The design of a ‘molecular barcoder’ is underway in different forms such as using chipless RFID[17] to tag plastics at the source, or digital IoT tags to follow trash as it moves through a city. Carlo Ratti, at the MIT Senseable City Lab, designed Trash Track to make what his team calls the removal chain more transparent.[18] The goal of the project is to promote behavioral changes and encourage people to make more sustainable decisions about what they consume and how it affects the world around them. Firms such as Amazon, Uber, or Chevron can play an important role in the collection and transportation bottlenecks. In one case, if Chevron builds stationary recycling or collection facilities near its existing gas stations, and charges a discount on gas for cars/trucks that bring mono-streamed waste to those points, then the value, purity, and quantity of the waste could be prime for recycling. This could increase the value of the recycled materials and make collection attractive from an economic standpoint, which has been demonstrated with ferrous and non-ferrous scrap metals. This could be similarly incorporated into the likes of Uber and Amazon, as the transport mechanism is built for on demand collection and disposal.

The economics of a circular system - Is profitability necessary and possible?

How can we create value for all stakeholders in the ‘materials’ ecosystem? Can circular businesses thrive without reaching profitability? If not, how long can they survive before giving an irreversible lead to their non-circular competitors? The discussion below use the following terminology to describe key aspects of a circular system that require consideration:

Virgin Material (market) Value -  $V

Recycled Material (market) Value - $R

Material Demand Rate - volume per unit time - (x/t)

Average Product Lifetime - L(t)

Material Recovery Rate - volume per unit time - (y/t)

Material (aesthetic & mechanical) Quality - Ratio (P) of virgin (PV) to recycled (PR) performance, P = PR / PV, where P ≤ 1.

While several scenarios will affect the economics of material circularity, let us consider the optimal condition where the supply and quality of recycled material matches the demand quantity (x = y), (P = 1), and the cost of recycled material is equal to virgin material ($R = $V). This optimal condition is, however, difficult to achieve due to several factors, but mainly because the volume of available recovered material (y), with high enough quality (P), does not meet the demand (x).

Figure 5. (A) Key relationships to consider when using materials efficiency to push the limits of economic efficiency. (B) Relationship between material recovery (available volume to be recovered for reuse) and product lifetime (L(t)). Here, the y-intercept (L(t)=0) reflects the amount of waste generated during manufacturing (e.g., subtractive vs. additive) also called post-industrial waste. Conversely, when L(t) > 0 material quantities reflect various product lifetimes coming from post-consumer waste.

The following section uses simple relationships to discuss factors that place upper and lower limits, or tolerable boundaries on quality, and supply of recycled material, ultimately identifying product types with the best chance of profitability optimization. 

●      Given a sufficiently high rate of material return (y/t → x/t), what is a tolerable quality ratio (P)?

●      Given a tolerable quality ratio (P), what is a tolerable rate of demand (x/t)?

●      Given a tolerable quality ratio (P), what product types are within reach? 

●      Given a tolerable rate of material return (y/t), what product types should be focused on?

●      Where does material return come from: Post-industrial (fast), post-consumer (variable)?

●      Given a sufficiently high rate of material return (y/t → x/t), what product types should be focused on? (Figure 5B)

Economic efficiency thrives in a free market, a market that adjusts automatically to new information about scarcity, demand, and processes of production to allocate resources in the most efficient way possible. Economic output is proportional to materials output in developed countries. In developed economies, annual consumption of materials is (~4x) global average.[6] From “Financing the circular economic model” published in 2020, “Circular business models are based on collaboration and cultivating relationships between clients and suppliers. The most important asset in a circular company is its processes, the trust that it has with customers and the reliability of its business model. There is an urgent need to understand how to value these new types of assets in order to enhance the allocation of funds within a circular economy”.[3] For plastics, value creation can reach $40-70 billion with a 30-50% material recovery rate.[19] Several solutions include optimizing sorting, acting in local communities, educating stakeholders, and stabilizing prices of recycled materials, which can be achieved through price insurance on recyclables and cost-plus contracts. While there is high upfront investment for R&D, the bottom line is there are strong economic and social incentives for reducing waste and making sustainable materials. The upsides include creation of ‘green collared’ jobs and minimization of health hazards (microplastics, pollution, etc.) leading to lower burdens on healthcare costs.

While the theoretical materials efficiency differs for different products and systems, our current materials restoration process is far from the maximum achievable. The need to increase materials efficiency opens up opportunities for increases in system productivity and innovation which creates value. In order for sustainable models to compete with virgin material costs and current linear models, we need to implement innovations to increase materials efficiency in areas that will drive the most profit. If this cannot be done, then the market will push out the non-profitable companies. To do this efficiently, we need data and design innovations in addition to technologies to increase the rate and precision of the development of new circular businesses. These intangible assets are required to close the gap in economic efficiency in the context of materials. That is to say, it is extremely important to monitor material data faster and more accurately in order to make informed decisions on how to allocate and restore resources efficiently.

What roles do empathy, education, and policy play in a circular economy?

How did I get so caught up on the plastic waste problem? Well, like most people, I heard about the statistics that by 2050, the mass of the polymers in the ocean will outweigh the mass of fish.[5] To add on to that, I saw a beautifully tragic documentary called “Albatross” showing the plastic-full stomachs of birds on the Midway Island in the Pacific. For an avid diver and ocean lover, this broke my heart to hear and see. But it didn’t immediately click for me that what I had been working on for the past 8-9 years – materials discovery, particularly polymer chemistries for consumer applications - somewhat contributed to this problem. While plastics in the stomachs of Albatross’ on the Midway Islands will bring public empathy and attention, decisions need to be made through sensible estimates of scale and impact. So, how does empathy, education, and policy tie in with increasing materials and economic efficiency?

In Allwood and Cullen’s Sustainable Materials: with Both Eyes Open, a striking example is made regarding the effect of a ban on plastic bags. In the UK, carrier bags account for less than 1% of the total use of plastics, and plastics account for 1-4% of CO2 emissions. Hence, stopping the use of all plastic bags would reduce the UK's national emissions by <0.01%. This is equivalent to avoiding driving 4 miles per year each, or turning off one 60W light bulb for one day, once a year.[6] Legislations banning plastic bags and straws are becoming widespread in many places, and it incorrectly leads consumers to think these policies are enacted at the scale necessary to solve major problems such as ocean plastics.    

Figure 6: left) weekly shopping requires 13 carrier bags weighing around 100g, right) weight of the other plastics brought home in the carrier bags was 1kg. [6]

In a detailed study published by Lebreton et al., it was estimated that fishing (17.9%), aquaculture (1.3%) and shipping (8.3%) are collectively responsible for 28.1% of the global plastic inputs into the oceans, based on coastal clean-up data[20]. Yet, not enough legislation is targeted towards fishing and shipping industries in reducing waste. To compensate, small and large companies are beginning to innovate creatively around these key issues as education and advocacy becomes more factual and accurate. Interface, a commercial flooring company, has been committed to becoming the world’s first environmentally sustainable and restorative company since 1994. They have a portfolio of products from 100% carbon neutral flooring to bio-plastic carpeting[21]. Interface also founded Net-Works, which collects and sells discarded fishing nets and brings it back to a global supply chain to make recycled nylon. This allows broken nets to gain a second life as beautiful, long-lasting carpet tiles[22].

When policy makers jump to pass legislation around public (mis)conceptions, more problems can arise than the laws are set forth to fix. One example is the California state policy “TB117” which required furniture manufacturers to treat their products with flame retardant chemicals, mainly to protect against fires started by neglected cigarettes. Since California contributes to 1/8th of the U.S. GDP, manufacturers started making and selling flame retardant sofas in many states throughout the country. It was later realized that flame retardants posed significant health hazards which led California to then change the law to no longer requiring flame retardants for sofas. The old sofas were donated to charity or low-income households or sent to landfills. This example aligns with what the Nobel Prize-winning economist F.A. Hayek warned against, which is crafting public policy through a "pretense of knowledge." Lots of levers need to be pulled through policy to increase circularity, and although there will always be unforeseen and unintended consequences, actions should be taken with regards to mitigation of risks. This is why it is crucial for scientists and economists to be consulted to guide design, delivery, and policy decisions to increase materials and economic efficiency.

Conclusion

While strategies for increasing materials efficiency towards a circular economy are becoming more competitive, they do not guarantee economically efficient systems. It is critical to close the profitability gap for current circular businesses. Intangible assets around material flow can promote industrial symbiosis where the byproducts of one process become inputs of another to the benefit of different stakeholders. A domino effect will likely come into play regarding a circular adoption. The question is when and how the critical mass will be obtained. Circular-forward companies will reach profitability at a faster rate with the help of data and design informed solutions. Once these trailblazers become profitable, their aligned values to go circular can expand the culture to others. By absorbing the initial risk and buying time for other players - those with more challenging technologies and higher costs - successful companies offer space to further innovate. This in turn can support a circular ecosystem that invests back into itself.  Furthermore, they will encourage more marketable and innovative solutions to accelerate the current recovery progress towards theoretical maximum materials efficiency.

When I took my first materials science class, we learned about the relationship between structure, processing, and properties. What I didn’t appreciate until recently is materials’ broader connection to data, sustainability, and design. It quickly became my mission to use sustainability as a critical element and informatics as a tool, to design material systems of the future that minimize the use of raw resources, waste, and emissions - aligned with efforts we should all be undertaking as a society.

“Hopelessness is arrogant because it assumes we know everything and precludes us from the opportunities to fix problems.”

— Jeremy Jackson, at a talk in 2019 to honor the late Harvard Professor Jim McCarthy, who was a champion of the environment.

---

References

[1] Hawken, P., Lovins, A. B., & Lovins, L. H. (1999). Natural capitalism: Creating the next industrial revolution. Boston: Little, Brown and Co.

[2] OECD WORK ON ENVIRONMENT 2019-2020; https://www.oecd.org/environment/brochure-oecd-work-on-environment-2019-2020.pdf

[3] Aboulamer A, Soufani K, Esposito M. Financing the circular economic model. Thunderbird Int. Bus. Rev. 2020;1–6. https://doi.org/10.1002/tie.22123

[4] CDC ESG Toolkit. https://toolkit.cdcgroup.com/es-topics/resource-efficiency/

[5] National Academies of Sciences, Engineering, and Medicine. 2020. Closing the Loop on the Plastics Dilemma: Proceedings of a Workshop–in Brief. Washington, DC. https://doi.org/10.17226/25647.

[6] Allwood, Julian M. & Cullen. (2012).  Sustainable materials : with both eyes open.  Cambridge :  UIT Cambridge

[7] Geyer, Roland & Jambeck, Jenna & Law, Kara. (2017). Production, use, and fate of all plastics ever made. Science Advances. 3. e1700782. 10.1126/sciadv.1700782.

[8] Yue Liu, Materials discovery and design using machine learning, Journal of Materiomics, Volume 3, Issue 3, 2017, Pages 159-177. https://doi.org/10.1016/j.jmat.2017.08.002.

[9] https://www.ncbi.nlm.nih.gov/pubmed/24833389; Recyclable, strong thermosets and organogels via paraformaldehyde condensation with diamines. Science. 2014 May 16;344(6185):732-5. doi: 10.1126/science.1251484.

[10] https://science.sciencemag.org/content/360/6387/398.editor-summary; A synthetic polymer system with repeatable chemical recyclability. J ZHU, SCIENCE27 APR 2018 : 398-403

[11] https://www.nature.com/articles/s41557-019-0249-2; Christensen, et al. Closed-loop recycling of plastics enabled by dynamic covalent diketoenamine bonds. Nat. Chem. 11, 442–448 (2019).

[12] https://www.fastcompany.com/90174404/black-plastic-is-killing-the-planet-its-time-to-stop-using-it; Black Plastic is Killing the Planet. It’s Time to Stop Using It. M Wilson. Fast Company. June 2018

[13] http://www.wrap.org.uk/sites/files/wrap/Recyclability%20of%20black%20plastic%20packaging.pdf; Development of NIT Detectable Black Plastic Packaging. R Dvorak. Wrap. September 2011

[14] https://www.mdpi.com/2073-4360/9/9/435/htm; Detection of Black Plastics in the Middle Infrared Spectrum (MIR) Using Photon Up-Conversion Technique for Polymer Recycling Purposes. W Becker. Polymers. 2017

[15] https://www.oceaninsight.com/solutions/case-studies/; Ocean Insight

[16] http://news.mit.edu/2019/mit-robots-can-sort-recycling-0416

[17] https://www.wired.com/2005/08/rfid-to-tag-or-not-to-tag/

[18]https://360.here.com/2017/05/23/making-rubbish-smarter-trash-track-initiative/

[19] McKinsey KIT, Plastic Waste Talk, Shannon Bouton; https://www.mckinsey.com/industries/chemicals/our-insights/how-plastics-waste-recycling-could-transform-the-chemical-industry

[20] https://www.nature.com/articles/s41598-018-22939-w#ref-CR62

[21] https://www.interface.com/US/en-US/sustainability/carbon-neutral-floors-en_US

[22] https://www.interface.com/EU/en-GB/about/mission/Net-Works-en_GB