Carbon footprint of CO2-based polypropylene via methanol-to-olefins route

(1)

Sustainability Science and Solutions Master’s thesis 2020

Kaisa Kuusela

CARBON FOOTPRINT OF CO

2

-BASED POLYPROPYLENE VIA METHANOL-TO-OLEFINS ROUTE

Examiners: Assistant professor, D.Sc. (Tech.) Ville Uusitalo Assistant professor, D.Soc.Sc. Jarkko Levänen

(2)

ABSTRACT

Lappeenranta–Lahti University of Technology LUT LUT School of Energy Systems

Degree Programme in Environmental Technology Sustainability Science and Solutions

Kaisa Kuusela

Carbon footprint of CO2-based polypropylene via methanol-to-olefins route Master’s thesis

2020

73 pages, 13 figures, 3 tables and 1 appendix

Examiners: Assistant professor, D.Sc. (Tech.) Ville Uusitalo Assistant professor, D.Soc.Sc. Jarkko Levänen

Keywords: polypropylene, plastic, CO2, power-to-X, carbon footprint, life cycle assessment

Plastic is one of the most widely used materials today owing to its versatility. However, the rapidly growing plastics industry is largely dependent on fossil fuels and the source of ever- increasing greenhouse gas (GHG) emissions. Decoupling the feedstock from crude oil and natural gas and shifting to renewable sources of carbon is an impactful strategy to reduce these emissions. Power-to-X technology provides routes to convert carbon dioxide (CO2) into renewable plastics.

In this thesis, the carbon footprint of CO2-based polypropylene is calculated to provide insight to the following questions: i) does CO2-based polypropylene store more carbon than is emitted during its production; ii) how CO2-based and fossil-based polypropylene compare with respect to global warming potential; and iii) what are the GHG hotspots of CO2-based production? Life cycle assessment methodology is applied in the calculation, which is carried out using Gabi software.

CO2-based polypropylene is a net carbon sink at -0.64 kg CO2 equivalents (CO2e) per kg of the polymer and generates 2.27 kg CO2e less emissions than petrochemical polypropylene.

Electrolysis of water is identified as the GHG hotspot of this product system due to its high electricity consumption, which is currently a hurdle for power-to-X technology. The results show that the carbon negative production of polyolefins is possible with the current energy mix of Finland, and that CO2-based polypropylene can act as a carbon sink in long-term applications.

(3)

TIIVISTELMÄ

Lappeenrannan–Lahden teknillinen yliopisto LUT School of Energy Systems

Ympäristötekniikan koulutusohjelma Sustainability Science and Solutions Kaisa Kuusela

CO2-pohjaisen polypropyleenin hiilijalanjälki methanol-to-olefins -teknologialla

Diplomityö 2020

73 sivua, 13 kuvaa, 3 taulukkoa ja 1 liite

Tarkastajat: Apulaisprofessori, TkT Ville Uusitalo Apulaisprofessori, VTT Jarkko Levänen

Hakusanat: polypropyleeni, muovi, CO2, power-to-X, hiilijalanjälki, elinkaariarviointi

Muovi on monipuolisuutensa ansiosta yksi nykypäivän käytetyimmistä materiaaleista. Alati kiihtyvä muovintuotanto on kuitenkin riippuvainen fossiilisista polttoaineista sekä kasvava kasvihuonekaasu(khk)päästöjen lähde. Päästöjä voidaan vähentää korvaamalla raakaöljy ja maakaasu uusiutuvilla hiilen lähteillä muovin raaka-aineena. Power-to-X -teknologia mah- dollistaa uusiutuvan muovin valmistuksen hiilidioksidista (CO2).

Tässä työssä lasketaan hiilijalanjälki CO2-pohjaiselle polypropyleenille tavoitteena vastata seuraaviin kysymyksiin: i) sitooko CO2-pohjainen polypropyleeni enemmän hiiltä kuin syntyy khk-päästöinä sen valmistuksessa; ii) kuinka fossiili- ja CO2-pohjaisen polypropyleenin lämmityspotentiaalit vertautuvat toisiinsa; ja iii) missä CO2-pohjaisen valmistuksen vaiheissa syntyy eniten khk-päästöjä? Laskenta perustuu elinkaariarvioinnin metodologiaan Gabi-ohjelmistoa käyttäen.

CO2-pohjainen polypropyleeni on nettohiilinielu -0.64 kg CO2-ekvivalentilla (CO2e) kilo- grammaa polymeeriä kohden ja tuottaa 2.27 kg CO2e vähemmän khk-päästöjä kuin fossiili- nen polypropyleeni. Sähkönkulutuksensa vuoksi veden elektrolyysi on valmistusprosessin mahdollisesti merkittävin khk-päästöjen lähde. Tulokset osoittavat hiilinegatiivisen ja vähempipäästöisen polypropyleenin tuotannon olevan mahdollista Suomen nykyisillä energianlähteillä, ja että CO2-pohjainen polypropyleeni voi toimia hiilinieluna pitkäaikai- sissa sovelluksissa käytettyinä.

(4)

ALKUSANAT

Tämän työn uusiutuvista muoveista on mahdollistanut ja rahoittanut Päijät-Hämeen BIOSYKLI-hanke. Kiitokset apulaisprofessori Ville Uusitalolle mielenkiintoisesta aiheesta, innostavasta ohjauksesta sekä kommentoinnista ja palautteesta. Suuri kiitos myös apulaisprofessori Jarkko Leväselle hyödyllisistä kommenteista. Lisäksi kiitän Muoviyhdistyksen Vesa Taittoa hyvistä näkemyksistä muoveihin liittyen ja Transus-tutkimusryhmää mainiosta työporukasta.

Nämä LUTilla ja Skinnarilassa viettämäni kuusi vuotta tulen muistamaan kaiholla. Suuri kiitos siitä kuuluu ystävilleni, jotka ovat antaneet ja opettaneet huikean paljon sekä sopivasti potkineet minua eteenpäin – tai ihan muuten vain. Olen myös äärettömän kiitollinen perheelleni, joka on aina ollut vankka tuki ja turva.

Lahdessa 18.6.2020 Kaisa Kuusela

(5)

NOMENCLATURE

Abbreviations

DAC direct air capture DME dimethyl ether

DMTO dimethyl ether/methanol-to-olefins EU-28 28 member states of the European Union EUBP European Bioplastics

GaBi life cycle assessment software GHG greenhouse gas

IEA International Energy Agency LCA life cycle assessment

LCI life cycle inventory analysis LCIA life cycle impact assessment MEA monoethanolamine

MeOH methanol

MTO methanol-to-olefins MTP methanol-to-propylene

OECD Organisation for Economic Co-operation and Development PBS polybutylene succinate

PC polycarbonate

PCL polycaprolactone

PE polyethylene

PEM proton exchange membrane PET polyethylene terephthalate PHA polyhydroxyalkanoate PHB polyhydroxybutyrate

PLA poly(lactid acid), polylactide

PP polypropylene

PtX power-to-X

PU polyurethane

PVC polyvinyl chloride

(7)

SOE solid oxide electrolysis TPS thermoplastic starch

Chemical compounds

CH4 methane

C2H4 ethylene, ethene C3H6 propylene, propene CH3OH methanol

CH3OCH3 dimethyl ether

Cx hydrocarbon with x carbon atoms

CO carbon monoxide

CO2 carbon dioxide

H2 hydrogen

H2O water

Units

bar bar

CO2e carbon dioxide equivalent

J joule

kg kilogram

mol mole

t tonne

Wh watt-hour

Unit prefixes

G giga, 10⁹

k kilo, 10³

M mega, 10⁶

T tera, 10¹²

(8)

1 INTRODUCTION

Climate change is one of the greatest challenges the world is currently facing. Adverse impacts of climate change have long been observed around the world (IPCC 2014, 72;

Letcher (ed.) 2016). Rising global average temperatures affect the dynamics between the climate, oceans, and land, endangering the survival of many organisms and human populations. In general, climate change is a naturally occurring phenomenon, but large share of the ongoing global warming trend can be attributed to human activities with high confidence (Hansen and Stone 2015). Since the late 19^th century, when technological advancements enabled mankind to tap into fossil hydrocarbons to satisfy its growing demand for energy, burning of these fossil fuels has released significant amounts of carbon dioxide (CO2) into the atmosphere. CO2 is a prominent greenhouse gas (GHG) which contributes to global warming by capturing part of the heat radiated from Earth’s surface. The speed at which fossil carbon – otherwise stored into the earth’s crust – is introduced to the atmosphere as CO2 has been faster than the carbon sinks could absorb into natural storages. Feedback mechanisms, which could speed up or slow down global warming, are not fully understood (IPCC 2014, 64; Lenton 2011, 201–202). As a precautionary approach to avoid abrupt or irreversible changes to our biosphere, climate change mitigation is of high importance.

Fossil fuels – oil, coal and gas – are the base of our current energy system, and hence to all human activities requiring input of energy: they currently provide 80% of global primary energy demand (International Energy Agency [IEA] 2019a). Most of the annual fossil fuel production is utilized in transport and energy sectors as fuels, and a part are refined into chemicals and materials, of which plastics are the most ubiquitous. Durability, low price, ease of molding into shape, light weight, and other sought-after properties found in plastics have ensured their extensive application in almost every sector of industry. Plastics are used in packaging, construction sector, in textiles, vehicles, medicine, agriculture and consumer products, among others. The production volume of plastics has doubled in two decades (Geyer et al. 2017), the growth rate surpassing that of any other bulk material (IEA 2018, 18).

The demand for plastics will likely more than double (Material Economics 2018, 78) or even nearly quadruple (Ellen McArthur Foundation 2016) by 2050.

(9)

Despite the many conveniences of plastics, the material currently sits on unsustainable foundation due to dependency on limited fossil resources. Petrochemical industry is the most energy-consumptive, and third most emitting sector after steel and cement industries due to high oil and gas consumption for both energy and feedstock (IEA 2018, 12, 27). Zheng and Suh (2019) estimate that in 2015, GHG emissions of 1.8 Gt carbon dioxide equivalents attributable to plastics-related processes were released, corresponding to 3.8% of total GHG emissions that year. Carbon budget is a concept estimating the maximum amount of cumulative GHG emissions humans can emit while restricting the global warming to 2 °C (Ellen McArthur Foundation 2016, 17). With current trajectory, plastics alone could account for over tenth of the global carbon budget by 2050 (Kistler & Muffett 2019, 81) and more than a third by 2100 (Material Economics 2018, 79) compared to the 1 % of today (Ellen McArthur Foundation 2016, 29). The potential increase in GHG emissions is drastic if plastics are not decoupled from fossil fuels.

Thanks to their light weight and sturdiness, plastics typically reduce energy consumption during their service life when compared to other materials providing the same service. Some examples are reusable plastic crates for fruit and vegetable transport (Albrecht et al. 2013), single-use cutlery in aviation catering (Blanca-Alcubilla et al. 2020), plastics packaging items (Franklin Associates 2014), insulation (Pilz et al. 2010) and other benefits such as hygiene and food preservation (Millet et al. 2018, 19). These assessments, however, often omit many important considerations from environmental impacts of littering to health effects of possible toxicity (Schweizer et al. 2018; Crippa et al. 2019, 43, 45). Mismanagement of used plastics has led to global-scale plastic accumulation even in remote marine and terrestrial ecosystems (Barnes et al. 2009); plastics are abundant enough in global sediments that they are suggested to serve as the key indicator of the Anthropocene, an epoch marking the time of significant human impact on the Earth system (Zalasiewicz et al. 2016). Plastic debris has negative physical and ecotoxicological effects on wildlife (Li et al. 2016; Tyler et al. 2019), and via food chain and water supply, might pose some risks to humans as micro- and nanoplastics (Wright & Kelly 2017; Galloway et al. 2019). Systematic review of indirect impacts of plastic items would provide a more holistic view on their environmental effects (Molina-Besch et al. 2019).

(10)

It is evident that multiple measures should be taken in concert to diminish the negative impacts of plastic to the environment and associated CO2 emissions. The most impactful strategies proposed include: i) shifting from fossil feedstocks to sustainable ones; ii) decar- bonizing the energy supply and this way reducing the emissions from processing; iii) designing plastic products for reuse and recycling; iv) increasing recycling rates of plastics;

and v) reducing the overall demand of unessential plastics (Zheng & Suh 2019; Ellen McArthur Foundation 2019; Kistler & Muffett 2019). This thesis focuses on the feedstock perspective and presents renewable feedstocks and alternative production routes to produce plastics.

Biomass is a renewable source of carbon and typically a more sustainable option to crude oil as the raw material for plastics from climate perspective. Biomass-derived plastics often have reduced environmental impacts – especially lower GHG emissions and fossil resource use – compared to conventional plastics (Chen & Patel 2012; Zhu et al. 2016; Kabir et al.

2020). Other environmental advantages are not always clear (European Commission 2019, 14; Walker & Rothman 2020). Growing biomass requires suitable land area, which may compete with food production, and usage of fertilizers, pesticides and water resources may pose risks to biodiversity.

Power-to-X (PtX) technology is discussed as being a potential solution to global energy system decarbonization, as chemicals and fuels including plastics (the X) can be produced via chemical conversion processes (the to) from air and water using electricity (the power).

PtX is based on capturing CO2 and splitting hydrogen from water with renewable electricity, whose availability is not tied to the availability of biomass. The global technical potential of solar energy alone, for example, manifold exceeds the energy demand of humanity (Kabir et al. 2018) and the same applies for wind power (Lu et al. 2009; IEA 2019b). Furthermore, PtX has synergy with the intermittent wind and solar power generation: it offers flexible energy storage by converting excess electricity into energy carriers during times of overgeneration, helping avoid curtailment. Producing plastic from captured CO2 via PtX route could prove to decarbonize the feedstock, provide climate change mitigation, and create new business based on sustainability, all the while contributing to material demands.

(11)

The question is how much emissions could be avoided if plastic was made from CO2 rather than crude oil. CO2-based plastic could even act as a carbon sink if used in long-lived applications, such as in construction and building sector. Polypropylene (PP) is chosen as the focus of this work, as it has suitable properties to said applications, growing demand and a known chemical conversion route. The carbon footprint of CO2-based PP production is calculated to find the net global warming potential (GWP), or in other words, the GHG emissions, of CO2-based PP. The objectives are: i) to investigate if the PP stores more carbon than is emitted during its production; ii) to compare the net GWP of fossil and CO2-based PP and to estimate the potential reduction in GWP if fossil PP were substituted; and iii) to identify the GWP hotspots in the product system and evaluate their significance. Life cycle assessment methodology with a cradle-to-gate approach is used to quantify the carbon footprint of CO2-based PP, which is compared to that of petrochemical PP. The calculation is carried out using life cycle assessment software Gabi version 9.2. Sensitivity analysis with different scenarios is carried out to find the processes and flows having the most impact on the carbon footprint of CO2-based production.

(12)

2 CURRENT STATE OF THE PLASTICS INDUSTRY

Polymers are ubiquitous in both nature and human society. All macromolecules in nature are polymeric, including cellulose, proteins, and DNA. Man-made polymers include rubbers, synthetic fibers, surface finishes and coatings, adhesives, and plastics (Brazel & Rosen 2012, 3). Plastics as a collective term refers to synthetic or semi-synthetic organic polymers which, analogous to their name, are capable of being molded into shape. Polymers are large molecules composed of hundreds to thousands of monomers, smaller molecules capable of forming bonds with other monomers to form linear, branched and network structures.

Polymers are named after their ‘mer’ or repeat unit, which describe the whole polymer chain.

Polymers are characterized by great heterogeneity; they may differ in structure, size, form, and molecular composition. By modifying these molecular properties, it is possible to alter the thermal, mechanical, and electrical properties of the polymer. (Seppälä 2008, 8–9, 15.) In addition to the comprising polymer, plastics contain one or more additives which enhance select properties (Brazel & Rosen 2012, 363). Geyer et al. (2017) found that on average, additives compose 7% of all plastics by mass.

The plastics industry is growing, driven by developing economies and increasing wealth. In 2018, 359 Mt of plastics were produced globally, excluding some synthetic fibers. From 2017, production of plastics increased by 3%. China alone accounts for 30% of global plastics production, with the rest of Asia, North America and Europe contributing by 21%, 18% and 17%, respectively. In Europe, the key end-user sectors for plastics are packaging (39,9%), building and construction (19,8%), automotive industry (9,9%), and electrical and electronic applications (6,2%). (PlasticsEurope 2019.) Approximately 6% of global oil and gas production is used for plastics (Ellen McArthur Foundation 2016, 27).

There are two types of plastics, thermoplastics and thermosets. Thermoplastics can be melted and molded into shape repeatedly without changes in molecular structure, whereas thermosets deteriorate during heating. Plastics are typically grouped based on their market demand and production volumes: commodity plastics, engineering plastics and high- performance plastics. (Millet et al. 2018.) The most common plastic grades are presented in the section 2.2.

(13)

In 2019, bio-based plastics production was 3.6 Mt. For these plastics 5 Mt of biomass was consumed, most of which were biogenic by-products (46%). On average, bio-based plastics contain 43% actual bio-based material. (Nova-Institut GmbH 2020b.) European Bioplastics (EUBP) reports the global bioplastic production capacity being 2.1 Mt, not counting in thermosetting plastics (EUBP 2020a). The term ‘bioplastic’ refers to both bio-based and biodegradable plastics. These concepts are discussed in the section 2.1, as they are not unambiguous. The growth rate of bio-based plastic production is about 3%, similar to fossil- based plastics (Carus 2020, 4). Common bio-based plastics are presented in section 2.3.

Plastics contribute to fossil CO2 emissions both through the energy used in the production and the carbon comprising the material. The emissions from production differ by plastic grade and are generally lower for high-volume commodity plastics: the carbon footprint of 1 kg of plastic range from 1.6 kg CO2 equivalents (CO2e) of polypropylene to 7.2 kg CO2e of nylon 6 (PlasticsEurope 2020). On average, the carbon embedded in plastics corresponds to 2.7 kg of CO2, and the total carbon footprint of plastics is 5.1 kg CO2e (Material Econom- ics 2018, 79, 81). Significant amount of fossil carbon is stored in plastics; Geyer et al. (2017) estimate that of all plastics produced, 30%, or 2500 Mt, are in current use. The carbon is released during deterioration, which can occur slowly in the environment (section 2.1) or instantaneously via incineration.

Globally around 58% of plastics end up in landfills or the natural environment, while 24%

are incinerated and the rest enter recycling (Geyer et al. 2017). Ellen McArthur Foundation estimates that 95% of plastic packaging material value, about 80-100 billion USD, is lost annually through discarding, landfilling and other losses (Ellen McArthur Foundation 2016, 27). The aim of plastics recycling is to reduce the demand for virgin fossil material and lessen the environmental burdens of linear, use-and-dispose type of economy (OECD 2018, 23). The emissions from virgin plastic production could be reduced by over 70% even with low quality recycling (Material Economics 2018, 81). Recycling of plastics and its current challenges are discussed in section 2.4.

(14)

2.1 Terminology

The need to curb emissions and justified concern over global plastic pollution have driven the development of alternative plastics, which are not of fossil origin and degrade in the environment. The key environmental advantage of plastics from renewable feedstock is that the carbon composing them is part of a closed carbon cycle, which is not the case with fossil- based plastics. Greenhouse gas emissions from the material are thus deemed climate neutral.

Biodegradable plastics in turn could help in reducing the ecological harm caused by persistent plastic litter (Haider et al. 2018), as a fraction of plastics will always escape waste collection.

Bio-based plastics are wholly or partly derived from biomass. Several international and European standards describe the procedures to determine the bio-based carbon content (ISO 16620-2:2015, EN 16640:2017, ASTM D6866) or total bio-based content (ISO 16220-4, EN 16785-1:2015, EN 16785-2) of a product. Based on these standards there exist certificates and labels which communicate the bio-based raw material content in a product. For example, OK biobased label by TÜV Austria certifies a product one-star bio-based if at least 20% of its carbon content is of renewable origin (TÜV Austria 2020). Even a 100% bio-based content of a plastic itself does not, however, assure sustainability nor provide information on its environmental impacts, such as land and water use and soil degradation. Standard EN 16751, for example, aims to identify environmental, social, and economic aspects of sustainability of bio-based products.

Most plastics are resistant to biological activity but can be degraded into small fragments by abiotic environmental factors such as abrasion, radiation, and oxygen. All plastics are degradable. Plastics, which are capable of being ultimately broken down into water, methane or CO2 and new biomass by microorganisms, are biodegradable (ISO 18606:2013).

Biodegradability of a plastic derives from its polymer structure: both fossil-based and renewable plastics can be biodegradable (Kabasci 2014, 2). The rate of biodegradability depends on the exposure conditions such as temperature, pH, concentration of microorganisms, moisture and oxygen (Haider et al. 2018). However, the term biodegradable does

(15)

not contain information about the location, timeframe, or the completeness of the decomposition process. These are often specified with standards and test methods created for different habitats, but the vast range of physical and chemical conditions in uncontrolled environments may lead to poor predictability of actual biodegradation. (Harrison et al. 2018.)

Compostable plastics biodegrade under specific conditions achieved in industrial composting conditions with controlled temperature, moisture, and aeration. Several standards (EN 13432; ISO 18606; ISO 17088) specify the time frame, exposure conditions and completeness of biodegradation for compostable and biodegradable plastics.

‘Compostable’ plastics can be understood to be compostable at home, but the conditions may not be severe enough for biodegradation to occur. Moreover, the standardization for home compostable plastic packaging is currently lacking in European context (Crippa et al.

2018, 154–155). Compostability and biodegradability are separate concepts: compostable plastics are not necessarily biodegradable in the ambient environment and, conversely, biodegradable plastics do not necessarily fulfil the composting criteria (Rujnić-Sokele &

Pilipović 2017, 137). For example, ecotoxicological tests are mandatory for compostable plastics but are absent from standards for testing biodegradability in aquatic environments (Harrison et al. 2018, 9).

A widely adopted term to distinguish renewable plastics from petrochemical ones is bioplastics and sometimes biopolymers. According to EUBP, bioplastics are either bio- based, biodegradable, or both (EUBP 2018). EUBP mentions compostability as an alternative to biodegradability (EUBP 2020b). These definitions leave room for interpretations: a bioplastic can be fossil-based and compostable, but non-biodegradable; or (partly) bio-based and non-biodegradable. The origin of the plastic and its mechanism of degradation are effectively blurred with the term ‘bioplastic’. Other terms in use with a similar problem include, for example, green plastic, sustainable plastic, and eco-plastic. The prefix bio- itself encompasses meanings of naturality, non-toxicity and eco-friendliness.

Conveying these images could lead to choices which externalize the environmental responsibility from the consumer to the material. Discarding plastic waste labelled as biodegradable might be perceived as a viable option, leading to more littering (Napper &

Thompson 2019). Biopolymer, however, may indicate the biocompatibility of a fossil or bio-

(16)

based polymer in medical applications (Kabasci 2014, 2) and in biochemistry, refers to a naturally occurring polymer (Vert et al. 2012, 384). The terminology around novel plastics can thus be confused with established trade terminology. Ambiguous terms such as green plastic are used more in marketing, where mental images may outweigh exactness.

Even if waste collection and management systems are improved, there will always be some leakages of plastics into the environment. Shifting to biodegradable plastics would lessen the impacts of plastic debris on ecosystems, as they persist for shorter time than non- biodegradable plastics. However, the effects of degradation products and secondary plastic particles are not fully understood (da Costa 2019, 92–93). In addition, the emissions arising from the production of biodegradable plastics and their impact on the recycling chain needs consideration (Kistler & Muffett 2019, 84; Rujnić-Sokele & Pilipović 2017, 135–136).

Biodegradable and compostable plastics can create either benefits or lead to loss of value depending on the application – collectible bottles and fruit stickers as examples (Crippa et al. 2018, 153–156). Biodegrading plastics are especially useful in uncollectible products and agricultural mulch films which can directly recycle nutrients to soil (Carus 2017, 8). To avoid loss of value, scientific and precise communication is required with biodegradable and compostable plastics (Crippa et al. 2018, 153–156).

2.2 Conventional plastic grades

Polyethylene (PE) is the most common plastic, its production representing 36% of all plastics demand (Geyer et al. 2017). There are various grades of PE, but it is commonly categorized into three main groups according to demand, production method and properties: low-density PE (LDPE or PE-LD), linear low-density PE (LLDPE or PE-LLD) and high-density PE (HDPE or PE-HD). PE-LD and PE-LLD are used to make containers, films and plastic bags.

PE-HD is used in sturdier packaging and piping. PE is a simple-structured polymer as it comprises of ethylene (-C2H4-) repeat units. PE is synthesized via polymerization of ethylene sourced from cracking of crude oil and natural gas products, most commonly from naphtha and ethane (IEA 2007, 66). Ethylene is also used as a platform chemical for many other plastics, including polyvinyl chloride (PVC), polyethylene terephthalate (PET) and polystyrene (PS) (PlasticsEurope 2020).

(17)

Polypropylene (PP) is the second most used plastic, representing 21% of total plastics demand (Geyer et al. 2017). PE and PP are polyolefins, where olefin means a hydrocarbon with at least one double bond. Polyolefins thus represent over half of plastics demand, of which PE and PP almost wholly account for. PP is similar to PE-HD in many of its properties but is harder and has notably higher melting temperature. (Seppälä 2008, 163–166) PP is widely used in packaging, automotive industry, and consumer goods (PlasticsEurope 2019).

PP is polymerized from propylene (C3H6), a gaseous co-product from steam cracking or a by-product from refining processes (IEA 2018, 25).

The next most used grades of plastic are PVC and PET, which comprise about 9% and 8%

of total global plastics production (Geyer et al. 2017). PVC is very durable and sturdy plastic typically used in building and construction sector, where it has accounted for 69% of all plastics (Geyer et al. 2017). PVC is used to make pipes, floor coverings, window frames and pool items. PET is a strong polymer used for both packaging and synthetic fiber production.

In Europe, liquid bottles are virtually the sole application of PET (PlasticsEurope 2019). The main uses of common plastics by sector are visualized in Figure 1. Other common plastic grades include polyurethane (PU), (extended) polystyrene (PS/EPS) and acrylonitrile butadiene styrene (ABS).

Figure 1. Common plastics and their fields of application in Europe. PE-MD, medium-density polyethylene.

Modified from PlasticsEurope (2019).

(18)

Examples of fossil-based biodegradable polymers are polybutylene succinate (PBS), polycaprolactone (PCL) and polybutylene adipate terephthalate (PBAT). Together, their production capacity totaled almost 250 kt in 2016. PBAT and PBS capacity alone totaled 370 kt in 2019 (EUBP 2020a). PBS, PCL and PBAT are typically blended with bio-based plastics or starch to improve their mechanical properties (Rujnić-Sokele & Pilipović 2017, 134). PBAT and PBS can currently be up to 50% and 100% bio-based (Carus &

Aeschelmann 2017), which is why their production capacities are oftentimes reported along with other bio-based plastics.

2.3 Bio-based plastics

Polymers for bio-based plastics can be produced from natural polymers such as sugar, starch, cellulose and lipids via chemical modification, fermentation or directly by microorganisms.

Bio-based plastics include both new plastic grades such as starch blends, polylactide (PLA) and polyhydroxylalkanoate (PHA), and ‘drop-in’ plastics such as bio-PE, bio-PET and bio- PP. Drop-in plastics are chemically identical to their petrochemical counterparts, but the only distinguishing factor is the renewable feedstock. Reporting on bio-based plastics varies, but cellulose acetate, thermoplastic starch (TPS), PLA, bio-PE and bio-PET are the major non- fiber bio-based plastics by production capacity (EUBP 2020a; Institute for Bioplastics and Biocomposites 2020; Carus 2020; van den Oever 2017, 23).

TPS and cellulose derivatives are examples of modified natural polymers. As pure starch is not applicable for processing, it is converted to TPS by melt-extrusion (Bastioli et al.

2014, 12) with added plasticizers, such as glycerol (Greene 2014, 89–91). Starch is often blended with other polymers, which may reduce the biodegradability of TPS (Lambert &

Wagner 2017). Cellulose acetate is produced via esterification of cellulose and is used in synthetic fibers and cigarette filters (Lackner 2015, 21), of which the latter forms some of the most common type of marine debris (Greene 2014, 35).

PLA (polylactic acid or polylactide) is polymerized either from lactic acid or lactide, which are obtained via microbial fermentation of sugar. The properties of PLA can be modified during its forming, which makes PLA suitable for packaging, automotive materials, and

(19)

biomedical applications (Tsuji 2014, 171–173). Stereocomplexed PLA can compete with fossil-based plastics in engineering applications (Nakajima et al. 2017). Although PLA is often considered biodegradable, it generally degrades poorly in the environment (Lambert

& Wagner 2017), and rather is only compostable in high temperatures (Zhu et al. 2016).

PLA can be recycled back into lactide, but with low yields (Tsuji 2014).

Polyhydroxyalkanoates (PHAs) exist as pure polymer granulates in the cells of bacteria, which synthesize PHA under stressful conditions and use it as an energy storage. PHA is harvested from the cells by solvents, enzymes, and mechanical means. PHAs are used especially in packaging and medical applications, but the current production costs and material properties somewhat hinder the competitiveness of PHAs. As an advantage, carbon- rich waste streams can be used as a feedstock. (Koller et al. 2014, 144.)

Verified biodegradability of several polymers in different environments is visualized in Figure 2. Marine and freshwater environments are especially challenging for biodegradative processes to take place, which makes the accumulation of litter more probable. It is seen that various polyhydroxybutyrates (PHB), which are based on PHAs, can biodegrade under all the specified conditions. Starch and cellulose in the figure refer to natural polymers and they readily biodegrade under the specified conditions. Bio-based plastics made from starch or cellulose, however, may not be biodegradable due to their different polymer structure after processing and incorporation of non-biodegradable additives (Lambert & Wagner 2017).

(20)

Figure 2. Biodegradable polymers in various environments, for which the temperature, degree of biodegradation and timeframe are defined. PBSA, polybutylene succinate adipate. Modified from nova-Institut GmbH (2020a).

2.4 Recycled plastics

Even though recycling is the preferred treatment option for plastic waste from climate perspective (Kistler & Muffett 2019, 58; OECD 2018, 41), about 9% of plastics ever made have been recycled and only 10% of them more than once (Geyer et al. 2017). Most plastics that are recycled end up in lower-value applications whose further recycling is not viable (Ellen McArthur Foundation 2016, 47). The European recycling capacity for PE-HD, PE- LD, PP and PET is about 6 Mt (Plastics Recyclers Europe 2019), whereas the demand of these grades exceeds 20 Mt (PlasticsEurope 2019). PlasticsEurope reports that 32.5% of collected plastic waste in Europe is recycled (PlasticsEurope 2019), but the actual rate might be closer to 10% than 30% due to mixed flows and contamination (Material Economics 2018, 83).

A major challenge to recycling is the diversity of plastics – over 30 different types are in common use (Material Economics 2018, 83), and more types are emerging (Crippa et al.

2019, 113). The economics of recycling favor large volumes and uniform quality of the waste, so efficient collection and sorting of variable and novel post-consumer plastics is thus

(21)

difficult. Rigorous sorting provides high-purity polymer feed (Kosior et al. 2019, 164, 171), but even combining a recovered and a virgin polymer of the same type may yield lower quality polymer due to incompatibilities in physical properties (La Mantia & Scaffaro 2014, 1886). Biodegradable plastics in plastic waste, multi-layered packaging (Rujnić-Sokele &

Pilipović 2017) and polymer blends (La Mantia & Scaffaro 2014) further complicate recycling. Additives such as colorants, stabilizers and flame retardants in the polymers are difficult to remove and may restrict the use of recycled polymer (Material Economics 2018, 83). For example, polymers from non-food packaging applications are not eligible for food packaging after the recycling, creating problems for automatic sorting (Kosior et al. 2019, 166–168).

Currently, plastics are recycled mechanically by washing, shredding, and melting directly into new polymers (Rahimi & Garcia 2017). Chemical recycling, a technology still in development, instead converts polymers into purified polymers or monomers, yielding plastics of better quality (Crippa et al. 2018, 141). Chemical recycling could help with the recycling of contaminated and mixed plastic waste flows (OECD 2018, 68), as it is able to remove additives on the molecular level (Crippa et al. 2018, 141).

Standardization and transparency about plastic contents could alleviate incompatibility issues (Ellen McArthur Foundation 2016, 59–60; Deloitte 2017, 39). Drastic changes should take place simultaneously across the entire value chain for a successful plastics economy (Material Economics 2018, 94). Progress is being made in issue recognition, product design for reusability and recyclability, increasing recycling capacity and adopting policies such as extended producer responsibility to improve the recycling of plastics (Ellen McArthur Foundation 2019).

(22)

3 DESCRIPTION OF TECHNOLOGIES FOR LOWER OLEFINS PRODUCTION

This chapter explores the technologies for producing lower olefins ethylene and propylene, sometimes referred to as C2 and C3 according to the number of carbon atoms. Butylene (C4) is typically produced along with ethylene and propylene but is less used as the feedstock for polymers. With conventional methods, lower olefins can be produced via several process routes: steam or catalytic cracking, ethanol dehydration, propane dehydrogenation, Fischer- Tropsch, and methanol-to-olefins (Falcke et al. 2017, 126–127). The feedstocks for these processes can be sourced from both fossil and biomass sources and processed with the same technologies, as illustrated in Figure 3. Bio-based olefin production has been reviewed by, for example, Chieregato et al. (2016) and Zacharopoulou & Lemonidou (2017).

Figure 3. Production routes for ethylene and propylene. Propane is a co-product from hydrotreatment and cracking processes.

Using CO2 and renewable hydrogen as the feedstocks, power-to-X pathway offers additional olefin conversion technologies, of which two are shown in blue color in Figure 3. Power-to- X routes are introduced in the section 3.3, where the renewable methanol-to-olefins route is presented in-depth. Cracking of petroleum and bio-based feedstocks is briefly described in section 3.1. In section 3.2, an introduction to the biological production route is presented. As a common process step for all plastics, polymerization is briefly described in the section 3.4.

(23)

3.1 Cracking route

Thermal and catalytic cracking are basic refinery processes in petrochemical industry to produce olefins, aromatics, and other intermediates from larger hydrocarbons. Gaseous and liquid feedstocks for cracking are sourced from natural gas processing and oil refineries. The most common feedstocks are ethane and naphtha, a mixture of hydrocarbons: ethane is available in North America and Middle East and naphtha is commonly used in Europe and Asia (IEA 2018, 34). Liquefied petroleum gas, natural gas liquids, gas oil, propane, butane and other refinery co-products and off-gases are also used in cracking (Falcke et al. 2017, 139; IEA 2018, 30; PlasticsEurope 2016, 14). There are tradeoffs between yield and product variability (IEA 2018, 32): using ethane as the feedstock has the highest overall product yield, the product being mostly ethylene, whereas olefin and naphtha feedstocks yield more varied products, including propylene (Falcke et al. 2017, 132). Currently, 95% of ethylene and 60% of propylene in the market are produced by steam crackers (Sadrameli 2015). Due to large quantities of cheap ethane from shale gas extraction being available in the United States, interest in cracking heavier feedstock to obtain propylene and heavier olefins has increased (Amghizar et al. 2017), and the production of both ethylene and propylene from shale gas has been assessed (Yang & You 2017).

Steam cracking and fluid catalytic cracking are the two dominating cracking processes.

Steam cracking is based on high-temperature pyrolysis (‘cracking’) of the feedstock in the presence of steam (Falcke et al. 2017, 127). Steam cracking unit can be divided into four process steps: cracking furnaces, quench and fractionation, condensation and gas cleanup, and product fractionation (Sadrameli 2015; Falcke et al. 2017, 128). Depending on the feedstock used, the cracking temperature is 750–900 °C (Sadrameli 2015). Fluid catalytic cracking employs heat and a catalyst to crack the feed hydrocarbons in 500–540 °C temperature and 1.5–2 bar pressure (Barthe et al. 2015, 52).

Bio-based naphtha, sourced from biodiesel refining, can similarly be used as a feedstock for steam cracking to obtain olefins. Bio-based naphtha is co-produced from the hydrotreatment of oils and fats, in addition to liquid fuels (Pyl et al. 2011; Sadrameli & Green 2007). Two Finnish companies are producing renewable diesel with the technology. UPM uses crude tall

(24)

oil from pulping to produce renewable diesel and naphtha, which is converted to polyethylene by Dow (Mannonen 2018). Neste produces renewable diesel from vegetable and waste oils via their NExBTL process (Aatola et al. 2008). Neste and Borealis are collaborating to produce bio-based PP from propane, a co-product from the NExBTL process (Neste Corporation 2020). Moretti et al. (2020) assessed the environmental impacts of PP based on bio-based naphtha from the NExBTL process, the feedstock being used cooking oil. They found reductions of 40–62% in GHG emissions and 80–86% in fossil fuel resource use in comparison to conventional PP. The increasing demand for waste streams such as used cooking oil raises the question whether to consider them as co-products instead of waste (Moretti et al. 2020).

3.2 Fermentation route

Fermentation is a process where sugar, starch and cellulose are metabolized by microorganisms or converted with enzymes into simpler carbohydrates like alcohols and organic acids. Sugars are readily available for fermentation, but more complex polysaccharides starch and cellulose need to be pretreated and broken down into sugars via hydrolysis.

(Deloitte 2014, 11, 19.) Since lignocellulosic biomass is the most abundant renewable carbon feedstock and does not directly compete with food production, it attracts great interest in fermentation industry (Abo et al. 2019). However, processing of lignocellulose has not yet become economically viable (Chieregato et al. 2016, 5). Fermentation utilizing different microorganisms yields variety of products used as bio-based plastic feedstock: ethanol, lactic acid and succinic acid among others (Hill 2018, 8).

Ethanol is the main fermentation product of interest for bio-based olefin production (Hill 2018, 8). A large variety of bacteria and yeasts exist which ferment sugars into ethanol in mild temperatures of under 37 °C (Abo et al. 2019, 6–8). Ethanol is converted into ethylene via dehydration using solid acid catalysts and moderate temperatures of 180–500 °C (Chieregato et al. 2016, 5–6; Zacharopoulou & Lemonidou 2017, 5). The conversion of ethanol to propylene is more complex. Propylene is synthesized via metathesis of ethylene and butenes – sourced from ethylene dimerization (Chieregato 2016, 13; Zacharopoulou &

Lemonidou 2017, 8–9). One of the largest producers of bio-based polyethylene is Braskem,

(25)

a company based in Brazil, which uses sugar cane as its primary feedstock. Bio-based polypropylene via fermentation is produced by, for example, Mitsui Chemicals (Barrett 2019) and Global Bioenergies (Global Bioenergies 2018).

3.3 Power-to-X route

CO2-based polymers are produced from CO2 via multiple chemical conversion routes presented in Figure 4, all of which require a renewable energy supply. CO2 is obtained from ambient air or flue gas. Renewable hydrogen (H2) is extracted from water (H2O) with electric current. CO2 and H2 can be synthesized into renewable methane, synthetic natural gas (syngas) or methanol, which are both fuels and chemical precursors to olefins and other chemical products. CO2 can also be converted into plastic precursors more directly with electricity and catalysts.

Figure 4. Pathways to convert CO2 into polymers. PC, polycarbonate; PAC, aliphatic polycarbonate, polyalkylene carbonate; PU, polyurethane; PE, polyethylene; PP, polypropylene; PLA, polylactic acid; PHA, polyhydroxyalkanoate; PET, polyethylene terephthalate; PEF, polyethylene furanoate.

A large fraction of fossil carbon in polycarbonates (PCs) and polyurethane (PU) can be substituted with CO2 (Gridnard et al. 2019, 4467). CO2 is co-polymerized with an epoxide, commonly propylene oxide, to create polyether carbonates or CO2-polyols, the precursors for PU (Liu & Wang 2017). Commercial-scale production of CO2-based PU is being pursued

(26)

by companies such as Covestro and Aramco (Muthuraj & Mekonnen 2018) and Econic Technologies (Raschka et al. 2019). Polycarbonates are similarly produced via CO2 co- polymerization with epoxides.

Methane (CH4) is synthesized from CO2 and H2 via exothermic Sabatier reaction (Ericsson 2017, 21). (Non-)oxidative coupling of methane is a one-step route to produce ethylene, and methane halogenation yields ethylene and propylene (Kolesnichenko 2020). Syngas – which contains carbon monoxide (CO) – is produced via reverse water-gas shift reaction from CO2

and H2 (Ericsson 2017, 21) or via reforming of methane (Kolesnichenko 2020, 193). Syngas is subsequently converted to hydrocarbons via Fischer-Tropsch reaction, where the yield of lower olefins (C2 to C4) is limited to about 56.7% (Gao et al. 2017).

CO2 can be metabolized to hydrocarbons by certain microorganisms when sufficient energy is available. Gas fermentation is based on using H2 as energy carrier (Schievano et al. 2019).

Electro-fermentation – or microbial electrosynthesis – utilizes bacteria that can reduce CO2

to hydrocarbons with electric current. The system is based on oxidation and reduction reactions at two electrodes: microbes reduce the CO2 at cathode by accepting electrons from oxidation of H2O at the anode. Products include ethylene, methane, ethanol, acetate and acetic acid, of which the latter two are most extensively researched. The technology is at laboratory scale yet shows promise for more sustainable production of plastics. (Bajracharya et al. 2017.)

Electrochemical conversion of CO2 to C2+ hydrocarbons proceeds via hydrogenation of CO2

with H2. The reaction typically occurs with methanol or CO (Fischer-Tropsch route) as the intermediate C1 products, which are subsequently converted to C2+ products over a single bifunctional catalyst. The hydrogenation of CO2 to C2+ takes place at 200 to over 400 °C depending on the catalyst. The selectivity of C2-C4 can exceed 80% for catalysts with methanol intermediate, but a selectivity of about 50% is attained for Fischer-Tropsch conversion. (Ye et al. 2019.) Progress has been made in the conversion of olefins over bifunctional catalysts (Wang et al. 2020) yet understanding the catalyst structure requires more research (Ye et al. 2019).

(27)

The following sections describe the conversion of CO2 to olefins via methanol route, including methanol synthesis and methanol-to-olefins process. The principles of CO2 capture and water electrolysis for H2 production, however, hold relevance to the other CO2-based pathways as well.

3.3.1 CO2 capture

CO2 can be captured from ambient air or from flue gases originating from combustion or a conversion process. CO2 capture is typically separated into three types: post-combustion capture, in which CO2 is separated from flue gas; pre-combustion capture, where CO2 is separated from CO2/H2 mixture originating from water-gas shift reaction; and oxy-fuel combustion, which produces mainly CO2 and H2O vapor as flue gas. Available techniques for CO2 capture are absorption, physical or chemical adsorption, membrane separation and cryogenic distillation. Absorption using an amine solvent monoethanolamine (MEA) is the most mature technology and widely applied in industry due to cost-effectiveness and high absorption efficiency. (Ahmed et al. 2020.) However, MEA is corrosive, moderately toxic and its regeneration requires high thermal energy input, which is why sorbents with lower regeneration energy have been developed (Bui et al. 2018, 1080.) Adsorption of CO2 with highly porous solid or chemical adsorbents generally requires less energy for regeneration and avoids the corrosiveness and toxicity of amines (Yaumi et al. 2017). Solid adsorbents also work well with low CO2 partial pressure, which makes them a good option for direct air capture (DAC) (Bui et al. 2018, 1083). However, physical adsorbents have low tolerance to impurities, moisture, and lower CO2 affinity (Nie et al. 2018).

Amine scrubbing is presented here, as it has been somewhat of a benchmark process for CO2

capture. Figure 5 illustrates the process. Flue gas is pumped to the absorption column from the bottom, from where it flows upwards against a stream of lean amine flowing downwards.

Most of the CO2 is absorbed by the amine. Rich amine leaves the absorption column and is led through a heat exchanger, where it is heated by the hot lean amine coming from the regeneration column, or stripper. Then rich amine enters the top of the stripper, where it flows downwards against rising hot steam generated in the reboiler, which strips the amine solution of CO2. Water is condensed from the overhead stream and returned to the stripper

(28)

as reflux and CO2 is recovered. Lean amine exits from the bottom of the stripper to be cooled in the amine heat exchanger and fed back to the absorption column. (Liang et al. 2015.)

Figure 5. Simplified flowsheet of CO2 absorption process with amine solvent. Modified from Liang et al.

(2015).

Capturing CO2 is more effective from point sources where its volumetric concentration in the gas stream is in 3–20% range (Yaumi et al. 2017), in comparison to current 0.041% of ambient air. Nevertheless, a major fraction of CO2 emissions is generated from dispersed sources, and DAC offers the possibility to capture CO2 irrespective of location or point emissions. DAC systems typically comprise of contacting area and regeneration module, where CO2 sorption and desorption take place, respectively. Main technologies are high temperature process using aqueous basic solution and low temperature process using solid sorbents. (Fasihi et al. 2019.)

3.3.2 Electrolysis

In water electrolysis H2O molecules are split with electric current into hydrogen gas and oxygen as a by-product. The hydrogen production rate is directly proportional to the electric current passing through the electrodes. In standard ambient conditions, the energy required for water decomposition is 285.8 kJ/mol, split between electrical and thermal energy.

(Koponen 2020, 27.)

𝐻₂𝑂 ⇒ 𝐻₂+¹

2𝑂₂ (1)

(29)

Without added heat, the minimum amount of energy needed to produce 1 kg of hydrogen is 141.8 MJ (Vázquez et al. 2018). Due to heat losses, the actual energy demand is typically between 170–220 MJ/kgH2 (Buttler & Spliethoff 2018; Koponen 2020, 23). The energy for the reaction is provided entirely with electrical energy in commercial electrolysers, where the operating temperature is around 50–80 °C. With increasing temperature, however, the total energy demand for electrolysis decreases, as the demand for electrical energy decreases more than the demand for thermal energy increases. (Koponen 2020, 28.) High-temperature electrolysis is thus more energy efficient but is still in demonstration stage (Buttler &

Spliethoff 2018).

Currently, two commercialized hydrolysis technologies exist: alkaline and proton exchange membrane (PEM) electrolysers. Alkaline electrolysers have been utilized since 1920s and boast relatively low capital costs due to simple materials and mature components. However, intermittent power supply can negatively impact hydrogen production and costs with alkaline electrolysers. PEM electrolysers are more suited to dynamic operation with shorter cold-start time and more flexible hydrogen production rate. They are also more compact by design. (Schmidt et al 2017.) The produced hydrogen is also readily pressurized, as the typical operating pressure of PEM is in 30–80 bar range whereas it is 1–30 bar for alkaline electrolyser (Hosker (ed.) 2019, 44). Drawbacks with PEM electrolysers include shorter lifetime and expensive materials such as platinum group metals (Buttler & Spliethoff 2018).

In alkali electrolyser the electrodes are submerged into an alkaline aqueous solution, separated by a diaphragm through which hydroxide ions pass from cathode to anode (Figure 6). In PEM, a proton-conducting membrane made from a sulfonated fluoropolymer functions as the electrolyte. Water is fed to the anode, where electric current splits it into oxygen gas and hydrogen ions. The ions pass through the membrane and combine with electrons to form hydrogen gas at the cathode. (Koponen 2020, 21–22.)

(30)

Figure 6. Schematics of alkali (a) and PEM (b) electrolysers. Modified from Koponen (2020).

Whereas the operating temperature of alkaline and PEM electrolysers is under 100 °C, solid oxide electrolysis (SOE) cells operate at temperatures of 700–900 °C and have higher efficiency. Water is fed into SOE cell as high-temperature (700–1000 °C) steam, which offers a possibility for utilizing waste heat from subsequent exothermic synthesis reactions.

(Buttler & Spliethoff 2018.) SOE can also operate in reverse mode as a fuel cell and in co-electrolysis mode to produce syngas from CO2 and steam (Schmidt et al. 2017), which is advantageous for further power-to-X chemical conversions.

3.3.3 Methanol synthesis

Methanol is a versatile alcohol of significant importance in the chemical and energy industries, and it can be converted into range of chemicals, including light olefins (Jadhav et al. 2014). Methanol is typically produced from syngas commonly derived from natural gas or coal gasification. The synthesis reaction can proceed with CO or CO2 as reactants.

Methanol formation is described by the following reactions:

𝐶𝑂₂+ 𝐻₂⇔ 𝐶𝑂 + 𝐻₂𝑂 𝛥𝐻 = 41 ^𝑘𝐽

𝑚𝑜𝑙 (2)

𝐶𝑂₂+ 3𝐻₂⇔ 𝐶𝐻₃𝑂𝐻 + 𝐻₂𝑂 𝛥𝐻 = −49.5 ^𝑘𝐽

𝑚𝑜𝑙 (3)

𝐶𝑂 + 2𝐻₂⇔ CH₃OH 𝛥𝐻 = −90.5 ^𝑘𝐽

𝑚𝑜𝑙 (4)

As seen from reactions 3 and 4, the reaction is significantly less exothermic when using pure CO2 in comparison to syngas where CO is present. Using CO2 simplifies the process into

(31)

reactions 2 and 3, leading to milder conditions, simpler reactor design, safer process and smaller by-product concentrations (Marlin et al. 2018). Thermodynamically, methanol formation from CO2 is favorable at high pressure and low temperature (reaction 3).

However, as CO2 is an inert gas, its activation requires elevated temperatures, which in turn favors the unwanted endothermic reverse water-gas shift reaction (reaction 2). Reaction conditions and catalyst composition heavily affect the reaction pathways, temperature being the main factor affecting methanol selectivity and catalyst activity. (Din et al. 2018.)

Methanol synthesis is typically carried out in 220–300 °C temperature and 50–100 bar pressure using Cu/ZnO/Al2O3 catalyst. Carbon Recycling International, which has produced renewable methanol in Iceland since 2012 (Marlin et al. 2018), use 220–250 °C and 10–30 bar (Olah 2012, 106). Methanol synthesis is carried out via single- or two-step process, where CO2 is either directly converted into methanol or first converted to CO and then into methanol (Rivera-Tinoco et al. 2016). Single fixed-bed reactor is preferred when pure CO2

is used, since the operating conditions are milder and high equipment costs can be avoided (Marlin et al. 2018). Typical H2/CO2 feed molar ratio is 3, but increasing the ratio seems to improve CO2 conversion and methanol synthesis (Din et al. 2018). Simplified process flowsheet for methanol synthesis is provided in Figure 7. The process is described in more detail by Pérez-Fortes et al. (2016).

Figure 7. Methanol (MeOH) synthesis process. Feed gas H2/CO2 is compressed and heated for MeOH synthesis (1). Product gas mixture from the reactor is cooled to separate unreacted gas and MeOH in a flash vessel (2). MeOH/H2O mixture is heated for distillation (3), from where purified MeOH is acquired. Unreacted gas is mostly returned to the reactor and about 1% of it is combusted (4) for steam production to avoid inert gas accumulation. Modified from Wang et al. (2019).

(32)

3.3.4 Methanol-to-olefins

Methanol-to-olefins (MTO) and methanol-to-propylene (MTP) are two established routes to convert methanol into olefins. While MTO yields a mix of ethylene and propylene, the MTP process optimizes propylene production (Wang & Wei 2017). The chemical principle is the same in both technologies, but the main differences lie in used catalyst and reactor design (Hannula & Arpiainen 2014; Wang & Wei 2017, 271). The exact reaction mechanisms of methanol conversion to olefins are still under debate (Xu et al. 2017, 39, 53; Wang & Wei 2017, 273). Generally, methanol is converted to dimethyl ether (DME) and methanol and DME into olefins and other hydrocarbons (Figure 8). Some reaction product olefins and aromatics participate in the conversion process as well (Xu et al. 2017, 62). The MTO reactions are highly exothermic and produce substantial amounts of water.

Figure 8. Reaction pathway of methanol conversion to olefins. Modified from Xu et al. (2017).

Typical MTO process begins with methanol and water being fed into a fixed or fluidized bed reactor containing porous zeolite catalyst. Molecular sieves Zeolite Socony Mobil (ZSM)-5 and silicoaluminophosphate (SAPO)-34 are the main two catalysts used in commercial MTP and MTO processes, as in other olefin production processes (Zacharopoulou & Lemonidou 2017). The methanol conversion takes place in a temperature and pressure ranges of 350–

550 °C and 1–5 bar (Wang & Wei 2017, 291). To better manage the heat generation especially in the case of a fixed bed reactor, methanol may be partly dehydrated to DME in a pre-reactor. Due to coke formation, the catalyst requires oxidative regeneration. The product effluent is cooled down and some of the condensed water and unreacted methanol and DME are recycled back to the reactor. The product gas is dried, after which it enters the product recovery section comprising of multiple distillation columns to be fractioned into lighter and heavier streams, including ethylene and propylene. Heavier olefins (C4+) can be recycled into an olefin cracking unit to increase ethylene and propylene yields. The MTP process is otherwise similar but requires fewer distillation columns. Figure 9 presents a

(33)

simplified MTO process. More detailed process descriptions are presented, for example, by Dimian and Bildea (2017) and Yu and Chien (2016).

Figure 9. MTO process flowsheet. C2, ethane and ethylene; C3, propane and propylene.

Several commercialized MTO and MTP technologies exist: UOP/Hydro MTO developed by former Universal Oil Products and Norsk Hydro; dimethyl ether/methanol-to-olefins (DMTO-I and DMTO-II) by Dalian Institute of Chemical Physics; SMTO by SINOPEC;

Lurgi MTP by former Lurgi GmbH (Wei & Wang 2017); and SHMTO by Shenhua Group (Yu & Zheng 2019, 16). Differing from the other MTO technologies, Lurgi MTP follows the route originally proposed by Mobil and uses ZSM-5 and fixed reactors (Ye et al. 2015, 286). Another MTP process, FMTP developed in Tsinghua University, uses SAPO-34 and fluidized bed reactor, but this technology is yet to be commercialized. DMTO has the largest production capacity of MTO technologies in China. (Yu & Zheng 2019.)

3.4 Polymerization

Polymerization is a chemical reaction where the reacting molecules (monomers) are combined into polymer chains or three-dimensional structures. The process requires inputs of extremely pure monomer and other components, and thermal energy, even if the reaction is exothermic. Three distinctive polymerization reactions exist: chain-growth polymerization, step-growth polymerization (polycondensation) and polyaddition. Chain-growth polymerization is the most important process for polyolefin production, whereas

(34)

polycondensation and polyaddition are used to produce polyester fibers, PET, and PU, among others. Chain-growth polymerization is based on opening the double bond of the monomer and linking these molecules together. Polymer growth occurs in seconds or minutes, but it takes several hours for high conversion from monomer to polymer. Catalysts are used for initiating and accelerating the polymerization. (European Commission 2007.)

The reaction process can be operated continuously or in batches by suspension, bulk, emulsion, gas phase or solution polymerization. In polypropylene production, gas phase and

‘slurry’ or ‘bulk’ suspension polymerization are applied. The conditions for suspension process are 60–80 °C and 20–50 bar and for gas phase process 70–90 °C and 20–40 bar. The slurry process, where powdery polypropylene is formed suspended in a diluent, is mostly used to produce specialty products. Modern processes are based on bulk suspension in loop reactors, where polymerization takes place in liquid propylene. In gas phase processes, gaseous propylene is introduced to a solid catalyst typically in a fluidized bed reactor. Gas phase process is suitable to produce propylene co-polymers with ethylene. After the polymerization step, the polymer is purified from residual monomers and solvents by mechanical and chemical means, and additives needed for further processing are added to the polymer at this point. (European Commission 2007, 22, 52–60.)

(35)

4 LIFE CYCLE ASSESSMENT METHODOLOGY

The carbon footprint of CO2-based polypropylene is calculated by applying life cycle assessment (LCA) methodology. LCA is an established method for quantifying environmental impacts of products and systems based on material and energy flows and is standardized by the International Organization for Standardization (ISO). General principles and framework of LCA are described in ISO 14040. ISO 14044 defines the specific requirements for conducting LCA. LCA comprises of four phases: i) goal and scope definition, where the goal of the work, the boundaries of the examined system and the general approach and assumptions are defined; ii) life cycle inventory analysis (LCI), where the inputs and outputs of the product system are compiled and quantified; iii) life cycle impact assessment (LCIA), where the significance of the product system’s environmental impacts are evaluated; and iv) life cycle interpretation, where the LCI and/or LCIA results are interpreted and assessed in relation to the goal and scope to reach conclusions. (ISO 14040:2006.)

ISO 14067 describes the principles for quantifying the carbon footprint of products and follows the general layout of LCA. Carbon footprint of a product refers to the sum of GHG emissions and removals over the product’s life cycle or selected life cycle stages or processes, expressed as carbon dioxide equivalents (CO2e). The results are declared in mass of CO2e per declared or functional unit. GHG removal refers to withdrawal of GHGs from the atmosphere over the product’s life cycle or a life cycle phase. Fossil and biogenic emissions and removals are each expressed separately as a part of the carbon footprint.

Biogenic carbon content of the product, if calculated, is not included in the carbon footprint but shall be reported separately. (ISO 14067.)

Sometimes unit processes and product systems have multiple valuable output products, which arises the question of how the environmental impacts are divided between these products. Allocation means partitioning of inputs and outputs to the products and co- products according to a defined procedure. According to ISO 14044, allocation should be avoided in the first place by dividing the process into sub-processes or via system expansion, where the different functions of the co-products are considered. If not avoidable, allocation

(36)

should be based on relevant physical qualities of the products – such as mass or energy content. Where not applicable, economic value or some other relationship is used as the allocation basis. Allocation is not applied to waste flows (ISO 14044).

Modeling of the product system is carried out using GaBi version 9.2. GaBi is an LCA software which combines modeling, reporting and diagnostic tools to assess the environmental impacts of a product system. GaBi also offers databases and datasets based on primary industry data, which can be used for material and energy flow modeling. GaBi used here includes Professional database 2019 and Extension databases II, IV, V, VIII, IXa, XI and XII. Data for the unit processes is gathered from literary sources such as project reports and academic papers. Primary data sources from industry and pilots are favored, but secondary data such as from simulations are used wherever primary data is not available.

Carbon footprint of CO2-based polypropylene via methanol-to-olefins route