PETpla.net Insider 06 / 2018

MATERIALS / RECYCLING PET planet Insider Vol. 19 No. 06/18 www.petpla.net 21 useful as “drop-in” replacements for the usual market commodity quali- ties. But this question is not as simple as it seems; it took the polyester industry decades of time and lots of money during its development phase in the last century to establish raw material qualities that are taken for granted today. And now, added to the word “Quality” comes the demand for “Quality consistency”. Polyester continuous production plant capaci- ties are within a range of 500-1,500t/d today; lines of smaller capacities have been and are being shut down one by one. As a consequence, consist- ent quality of PTA is crucial to making the running of plants of such capacity stable and successful. When we look for publicly-avail- able PTA product specifications it is striking that there is no or very little information on basic details, like grain size, grain size distribution, flowabil- ity or shelf life in big bags. The right grain size of PTA, which is within the 40-240μm range, at an average of 150- 170μm is a basic property that governs reactivity, flowability and, together with moisture of less than 0.1%, the guar- antee of flowability in silos and pneu- matic conveying systems, as well as shelf life in big bags, over many years. This means of achieving this feature is, for example the recrystallisation of crude TA under pressure in water or diluted acetic acid, centrifugation and subsequent drying. Because of the high corrosive internal environment of recrystallisation and drying, this equip- ment is made of titanium or titanium lined stainless steel. Another basic question appears to be: how to sepa- rate TA from IPA. Recrystallisation in diluted acetic acid or organic solvents is required for this operation. Such essential chemical operations are, for any investor and in terms of process economy and profitability, meaningful only at capacities commonly used in basic material chemistry. Using the PTA quality specifica- tion in Figure 1 as example, and looking at the control of metal con- tent (<10ppm for Mg, Ca, Na, K, and Al together) it becomes obvious that recycling processes that use as hydrolysis aids large quantities of alkaline like NaOH or KOH, will need highly sophisticated purification and recrystallisation technology to achieve <10ppm metal content. One way of avoiding high expenses for purification and finishing of PTA, IPA and MEG is described by an EU-registered patent (EP2838941 A1) in which PET waste glycolysis is done by MEG, fine filtration of the solved PET waste and direct reuse of the glycolate blend as 20-100% raw material replacement in a running PET filament plant. Similar approaches are also described in Chinese patent literature. 2.2 What shall the PET waste input quality be? While PET bottle waste which is the most common input material for polyester recycling, there is a broad range of quality in the market. All kinds of material are available, from baled bottles of >80 <95% PET, or PET flakes up to 99,9% purity with variation in colour and transparency. The vital question is: what level of purity is feasible for the chemical recycling process? It becomes obvious immediately that high impurity waste streams, such as unsorted household packaging waste, presorted polyester textiles or polyester-cotton blend textiles, which would otherwise be used as waste incineration power plants fuel, are useful in terms of economy and CO 2 footprint. As soon as waste of purity levels >80% PET is used for chemi- cal recycling, this input is compet- ing with mechanical recycling where a large number of well-established, optimised and low cost processes are available. When considering the CO 2 foot- print of PET chemical recycling, it is frequently forgotten that the rPTA, rIPA and rMEG produced are con- verted to polyester again, with known conversion costs (energy, investment, labour). In simple terms, chemical recycling to produce PTA, IPA and MEG must use crude waste instead of crude oil. 2.3 What is the minimum fea- sible and economical plant capacity of such a chemical recycling project? This is the most crucial ques- tion. Current PTA and MEG plant production capacities are within the 0.6-1.2mt/a or 1.6-3.3kt/d range, which is beyond all current recy- cling capacity. Let us consider a very simple example; producing PTA and MEG of bottle or fibre grade market quality and assuming a PET waste input of 200t/d and purity of 70% PET. Provided the PET fraction could be converted with 90% yield to crude PTA and 90% yield to MEG, final products are estimated as fol- lowing: Example: input waste should contain 70% PET The 140t PET contained within the 200t waste is theoretically con- verted to 119.8TA + 44.6t MEG. Rectification and recrystallisa- tion losses are estimated at 10%; products resulting from 200t waste are therefore: 107.8t PTA and 40.2t MEG. Non-PET waste is 60t, plus 16.4t production waste; 76.4t waste in total. This example project would produce ~110t PTA, 40t MEG and 76t non-PET wastes per day. The equipment these amounts of PTA and MEG will cover the size of a small pilot facility; it will cleave away from common polyester raw mate- rial production capacities in eco- nomical size. The simple conclusion is – ignoring imponderables – the production of market standard PTA and MEG from PET waste based on a green field investment is not commercially viable within an input capacity range of 50-200t/d. Common production capacity of PTA from PX is 1.2mt/a. “Small” plants have a capacity of 600,000t/a. Old-fashioned PTA plants have a capacity range of 200,000-400,000t/ a. Another interesting figure: down- sizing capacity from 100 to 50% still requires 70% of the investment! 2.4 Are there existing chemical PET recycling processes on an industrial scale? It is surprising that, while newer processes of chemical recycling are invented, there are a number of well- functioning and commercially-proven recycling processes on an industrial scale already available. RECYCLING S P E C I A L