DOI: 10.3303/CET2292103 Paper Received: 13 February 2022; Revised: 25 March 2022; Accepted: 6 May 2022 Please cite this article as: Umeda R.E., Schiavon M.I.R.B., Munhoz A.L., Maciel Filho R., Wolf Maciel M.R., 2022, Simulation of Hydrolytic Ring Opening Polymerization of Nylon 12 from Laurolactam, Chemical Engineering Transactions, 92, 613-618 DOI:10.3303/CET2292103 CHEMICAL ENGINEERING TRANSACTIONS VOL. 92, 2022 A publication of The Italian Association of Chemical Engineering Online at www.cetjournal.it Guest Editors: Rubens Maciel Filho, Eliseo Ranzi, Leonardo Tognotti Copyright © 2022, AIDIC Servizi S.r.l. ISBN 978-88-95608-90-7; ISSN 2283-9216 Simulation of Hydrolytic Ring Opening Polymerization of Nylon 12 from Laurolactam Rafael E. Umeda*, Maria I. R. B. Schiavon, André L. J. Munhoz, Rubens M. Filho, Maria R. W. Maciel School of Chemical Engineering, State University of Campinas – Unicamp, 13083-852, Campinas/SP, Brazil r264217@dac.unicamp.br The pandemic caused by COVID-19 resulted in a significant shortage of personal protective equipment (PPE) and additive manufacturing (AM) has been used to produce masks and face shields quickly and with quality. A widely used input is Nylon 12, a polyamide with excellent mechanical properties, low toxicity, better adapting to PPE by AM compared to other polyamides. It is frequently obtained from ring opening polymerization (ROP) route from laurolactam and can be derived from both petroleum and renewable sources. There is a great challenge toward few studies related to the production of Nylon 12 and mainly laurolactam monomer derived from renewable sources. This work aims to simulate the production of Nylon 12 by hydrolytic ROP in the Aspen Polymer Plus® simulator. A sensitivity analysis was carried out investigating the variation of the conversion, number average molecular weight, and number average degree polymerization, providing a better understanding of the process. Based on the results, it is concluded that the simulation is viable, since it was observed a formation of a polymer with high conversion and molecular weight according to experimental data. 1. Introduction The pandemic caused by COVID-19 presents itself as one of the greatest global health challenges of this century. Scientists are studying the characteristics of the virus as speed of spread and the need to mitigate exposure (Advincula et al., 2020). The protection of health professionals is extremely important, requiring the use of personal protective equipment (PPE) such as masks and face shields, however, the global emergency together with an unforeseen situation resulted in the scarcity of these PPE's. Additive manufacturing (AM) emerged as an alternative for the production of PPEs, enabling the production of complex objects with great speed and quality (Amin et al., 2020). The AM process produces 3D objects with different types of materials such as metals, ceramics and plastics. The mechanical properties of objects are heavily affected primarily by the fill design. The greater the fill, the better the object's strength, however, the fill is reduced in less printing time, energy and cost (Patti et al., 2021). According to Mostafa et al. (2018), polyamides are widely used in AM as engineering polymers, in particular Nylon 12, the most used polyamide due to its excellent mechanical properties such as high resistance to solvents and corrosion, great thermal resistance allowing the PPE to be sterilized by autoclave. The crystallinity and melting temperature are lower than other polyamides, requiring less power, therefore energy is spent on the construction of objects. In addition, Nylon 12 has biocompatibility and does not generate toxicity on skin contact (Rahim et al., 2017). Nylon 12 is a 12 carbon polyamide used as engineering thermoplastic with a wide range of applications, synthesized from ω-aminolauric acid or laurolactam (Griehl and Ruesteivi, 1970). It has excellent mechanical properties, low moisture absorption, durability and good chemical and thermal resistance. It is the most suitable Nylon for use in AM as it exhibits lower crystallinity and melting temperature compared to Nylon 6, which results in the need for less power to melt the material and consequently less energy expenditure to produce three-dimensional objects (Mostafa et al., 2018). Nylon 12 is synthesized from different technological routes, however the most used and industrially efficient is the hydrolytic ring opening polymerization (ROP). This route uses laurolactam monomer, a raw material usually produced from petroleum-derived sources such as butadiene, but there is the possibility of being obtained from renewable sources (Kyulavska et al.,2019). The polymerization is 613 divided into three main reactions, first the ring opening reaction (ROR) which consists of the reaction between the cyclic monomer and water which acts as initiator with the ring opening of the monomer forming the amino acid. Then, polycondensation occurs, which consists of the reaction between two or more amino acid molecules, in which the amine functional group reacts with the carboxylic acid functional group, forming a dimer and water. Simultaneously with polycondensation, polyaddition reactions also occur, which are reactions between the amino acid and the laurolactam monomer that cause the ring opening of the monomer and the addition of this molecule to the structure (Russo and Casazza, 2012). Thereby, there is a great challenge due to the scarcity of works and articles related to the production of Nylon 12 and especially the laurolactam monomer derived from renewable sources. Therefore, it is extremely important to study the polymerization process of Nylon 12, with process simulation being a very effective means for analyzing several variables that influence the process. Therefore, this work aims to reproduce the hydrolytic ROP process of Nylon 12, using the Aspen Polymer Plus® process simulator, allowing the identification of the relationship between polymer properties and process operating conditions. 2. Methodology Firstly, the kinetics of the hydrolytic ROP of Nylon 12 is shown, then the operational conditions of the process are described. 2.1 Kinetics The kinetic mechanism of the polymerization of Nylon 12 proposed for this process is done by the functional groups method and shown in Table 1. The reaction scheme was obtained from the work of Seavey and Liu (2008), who performed the simulation of the hydrolytic ROP of Nylon 6 and adapted for the polymerization of Nylon 12. This can be done because the reactions involving both the polymerization of Nylon 6 and Nylon 12 are similar. Eight reactions are considered in the process, three main ones involving laurolactama monomer, two cyclic dimer reactions, and three reactions involving acid and amine monofunctional. All reactions take place under uncatalyzed and catalyzed conditions by acid groups, with both being considered in the reaction rate expression. Table 1: Kinetic mechanism for Nylon 12 hydrolytic ring opening polymerization Reactions 1. Ring opening of laurolactam 𝑊 +𝐿𝐿 𝑘1 ′= 𝑘1 𝐾1 ⇔ 𝑘1 𝐴𝐷𝐴 2. Polycondensation 𝑃𝑚 +𝑃𝑛 𝑘2 ′= 𝑘2 𝐾2 ⇔ 𝑘2 𝑃𝑚+𝑛 +𝑊 3. Polyaddition of laurolactama 𝑃𝑛 +𝐿𝐿 𝑘3 ′= 𝑘3 𝐾3 ⇔ 𝑘3 𝑃𝑛+1 4. Ring opening of cyclic dimer 𝑊 +𝐶𝐷 𝑘4 ′= 𝑘4 𝐾4 ⇔ 𝑘4 𝑃2 5. Polyaddition of cyclic dimer 𝐶𝐷 +𝑃𝑛 𝑘5 ′= 𝑘5 𝐾5 ⇔ 𝑘5 𝑃𝑛+2 6. Acid end-group termination 𝐻𝐴𝐶 +𝑃𝑛 𝑘2 ′= 𝑘2 𝐾2 ⇔ 𝑘2 𝑃𝑛,𝑥 + 𝑊 7. Amine terminator laurolactama addition 𝐶𝐻𝐴+𝐿𝐿 𝑘3 ′= 𝑘3 𝐾3 ⇔ 𝑘3 𝑃𝑛,𝑦 8. Amine end-group termination 𝐶𝐻𝐴+𝑃𝑛 𝑘2 ′= 𝑘2 𝐾2 ⇔ 𝑘2 𝑃𝑛,𝑦  +  𝑊 W: Water, LL: Laurolactama, ADA: 12-Aminododecanoic acid, CD: Cyclic dimer, HAC: Acetic acid, CHA: Cyclohexylamine, 𝑃𝑛: Polymer chain with n repeat units, 𝑃𝑚: Polymer chain with m repeat units, 𝑃𝑛,𝑥: Polymer chain with n repeat units and acetic acid terminated, 𝑃𝑛,𝑦: Polymer chain with n repeat units and cyclohexylamine terminated. 614 2.2 Process description The polymerization process of Nylon 12 in Aspen Polymer Plus® was simulated using the experimental plant from the work of Kawakami et al. (1994). It consists of a polymerization column that has three zones (upper, intermediate, lower). Initially, the feed is at 533.15 K and 1 atm, flow rate of 7.35 kg/h and composition of 98.9 % laurolactam, 1.0 % water initiator and 0.1 % monofunctional acid catalyst acetic acid which are fed into the upper zone of the column. In the upper zone most ROR occur which must take place at high temperatures and pressures. The resulting product is transferred to the intermediate zone, which is at lower temperature, where the ROR are completed and the polycondensation and polyaddition reactions in the system start. Finally, in the lower zone at lower temperatures, polycondensation and polyaddition reactions finish. 3. Results and discussion Figure 1 shows the flowchart of the polymerization process for Nylon 12. The process is fed (FEED) in the simulated upper zone by a continuous stirred tank reactor (CSTR1) which must operate in a vapor-liquid phase, temperature 573.15 K, pressure 5 atm, condensed phase with a volume of 30 m3. The product vapor stream (R1VAP) from the CSTR1 reactor is transferred to a condenser (DISTIL) which recycles the unconverted monomer and water to the feed of the process and discards the excess water (COND) from the process. The product liquid stream (R1OLIGO) is transferred to the intermediate zone that was simulated with another continuous stirred tank reactor (CSTR2) operating in a vapor-liquid phase, temperature 548.15 K, pressure 2 atm, and a condensed phase with a volume of 15 m3. The vapor stream (R2VAP) from the CSTR2 is recycled directly to the FEED. The product liquid stream of the CSTR2 (R2OLIGO) is transferred to the lower zone that was simulated with a plug flow reactor (PFR1) with a thermal fluid of constant temperature of 573.15 K and heat transfer of 0.7 J/s.m2.K. The dimensions of the plug flow reactor are 3 meters long and 0.125 meters in diameter. The product stream of the lower zone (POLY1) is the final product that contains the Nylon 12. The conditions of the main streams of the polymerization column are shown in Table 2. Figure 1: Hydrolytic ROP process flowsheet Table 2: Main streams of flowsheet Properties FEED R1OLIGO R1VAP COND R2OLIGO R2VAP POLY1 Temperature [K] 523.00 573.15 573.15 373.17 548.15 548.15 547.24 Pressure [atm] 1 5 5 1 2 2 2 LL [kg/h] 7.2692 0.851 0.0284 - 0.5652 0.0017 0.4033 W [kg/h] 0.0735 0.0217 0.1934 0.0591 0.0111 0.0126 0.0112 ADA [kg/h] - 0.0008 - - 0.0003 - 0.0003 NYLON12 [kg/h] - 6.3813 - - 6.6596 - 6.8127 HAC [kg/h] 0.0074 0.0003 0.0011 8.93x10-7 0.0001 - 0.0001 DIMER [kg/h] - 0.0502 - - 0.0546 - 0.0634 MWN [g/mol] - 15,057 - - 21,880 - 22,866 615 In the upper section, the conversion of laurolactama (XLL) to Nylon 12 was 88.23 %. This occurs mainly because this section has, for the most part, hydrolytic ROR and, in parallel, polycondensation and polyaddition reactions. It is observed in Table 2, the R1VAP and R1OLIGO currents are at 573.15 K and 5 atm, indicating that they passed through the CSTR1 reactor at high temperatures and pressures for the hydrolytic ROR to occur. Polycondensation and polyaddition reactions also occur, responsible for the growth of the polymer chain, observed by the high number average molecular weight (�̅�n) of 15.057 g/mol, indicating that there were polycondensation and polyaddition reactions parallel to the hydrolytic ROR in the upper section of the polymerization column. Note that there are two output streams from the CSTR1 reactor, with the R1VAP steam stream having the function of separating water and laurolactam. This separation of the vapor and liquid streams is very important to be done mainly to separate the process water, to eliminate the excess and to recycle it to the feed stream, as this way the water does not react in the reverse polycondensation reaction that reduces the conversion of Nylon 12 and is reused in feed as an initiator of the hydrolytic ROR. In the intermediate section, there is an increase in XLL, number average molecular weight (�̅�n) and number- average degree of polymerization (DPn). This conversion increase to 92.18% indicates that the hydrolytic ROR are finishing and there is a greater occurrence of polycondensation and polyaddition reactions in the process. These reactions, mainly polycondensation, are observed by the large increase in the �̅�n to 21.880 g/mol and the DPn to 111.11. As in this section mainly polycondensation reactions occur, it is necessary to separate the water to later use it in the process feed. For this reason, the R2VAP steam stream which is composed of water and laurolactam is recycled to the feed stream. The lower section had a XLL of 94.41 %. According to the experimental work by Kawakami et al. (1994), at the end of the polymerization column XLL to Nylon 12 is 99.5 %. This discrepancy between conversions is due to factors such as the kinetic parameters used in the simulation. Due to the scarcity of information on Nylon 12 in the literature, such as kinetic parameters, for the simulation the kinetic parameters of the hydrolytic ROP of Nylon 6 from the work of Seavey and Liu (2008) were used. Another factor that may have influenced the result is the limitation of the simulator's operating unit, as the polymerization column was not available in the simulator, it was necessary to perform the simulation using CSTR and PFR reactors to simulate similar column conditions. However, the simulation conversion was satisfactory despite the polymerization column conversion being lower than reported in the literature. The polymer has a low extractable percentage of 6.51%, which represents a low number of unreacted components, and a high �̅�n of 22.866 g/mol, higher than values reported in the literature from 5,000 g/mol to 9,500 g/mol. This high value indicates the formation of a high polymer influencing the qualitative properties of the polymer. The term high polymer refers to the minimum molar mass of a polymer chain that has the physical properties of a polymer, as opposed to a monomer or oligomer. Using the sensitivity analysis tool, several process variables were analysed, varying the mass flow rate of the feed stream from 0 to 100 kg/h to determine their influence on the process parameters, as shown in Figure 2. Analyses were also carried out in the plug flow reactor to determine the relationship of the variation of the reactor length with the �̅�n and the DPn. In Figure 2 (A), the effects of the variation in the mass flow rate of the feed stream on the DPn and on the �̅�n are observed. The R1-MWN and R1-DP curves correspond to the �̅�n and the DPn of the R1OLIGO stream. Similarly, R2-MWN and R2-DP are from the R2OLIGO stream and R3-MWN and R3-DP are from the POLY1 stream. The profile of the �̅�n and DPn curves are superimposed due to the �̅�n being dependent on the DPn, which results in a similar behaviour. It is noted that these properties decrease with the increase of feed mass flow rate, because these are related to the conversion of Nylon 12. As conversion decreases with increasing feed mass flow rate, the �̅�n and DPn of polymers produced in the reactors also decrease. Figure 2 (B) presents the effects of the variation of the feed mass flow rate on the conversion. The conversion of laurolactam in all reactors decreases with increasing feed mass flow rate. With more feed mass flow rate, lesser the residence time of the material in the reactors, causing a decrease in the laurolactam conversion. The effects of the variation of the feed mass flow stream on the extraction value are presented in Figure 2 (C). This polymer property corresponds to the unreacted components of laurolactam, 12- aminododecanoic acid, dimer and water. There is an increase in the extraction value with feed mass flow rate due to a decrease in conversion with feed mass flow rate. 616 Figure 2: Effect varying the feed mass flow rate in (A) �̅�n and DPn, (B) XLL, (C) extraction value and (D) water weight fraction and varying the length in (E) �̅�n and DPn In Figure 2 (D), the effect of the variation of the feed stream mass flow rate on the water weight fraction in the reactors input streams is shown. With the increase in feed mass flow rate, there is an increase in the water weight fraction in R1OLIGO in the CSTR2 reactor and in R2OLIGO in the PLUG reactor. These increase in the streams R1OLIGO and R2OLIGO occurs because the XLL is decreasing, consequently, more water will be in the process. This occurs because water acts as an initiator for the polymerization of Nylon 12 in hydrolytic ring opening reactions, however, in polycondensation reactions it acts in the reverse direction of the reactions, decreasing the conversion of laurolactam. The main hydrolytic ring opening reactions take place in the upper section (CSTR1) with the unconverted water being recycled to the feed stream. In the middle and lower section, few hydrolytic ROR occur and polycondensation and polyaddition reactions start, therefore, the presence of water to increase the conversion is undesirable. In the intermediate section, unconverted water is recycled into the FEED, so a greater variation in the R1OLIGO can be seen. 8,200 10,200 12,200 14,200 16,200 18,200 20,200 22,200 24,200 40 50 60 70 80 90 100 110 120 5 10 20 30 40 50 60 70 80 90 100 𝑀 ̅ND P N Feed flow rate (kg/h) (A) R1-DP R2-DP R3-DP R1-MWN R2-MWN R3-MWN 0.4 0.5 0.6 0.7 0.8 0.9 1 0 10 20 30 40 50 60 70 80 90 100 X LL Feed flow rate (kg/h) (B) R1-OLIGO R2-OLIGO POLY1 0 5 10 15 20 25 30 35 40 0 10 20 30 40 50 60 70 80 90 100 E x tr a c ti o n v a lu e ( % ) Feed flow rate (kg/h) (C) 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0 10 20 30 40 50 60 70 80 90 100 W a te r w e ig h t fr a c ti o n Feed flow rate (kg/h) (D) R1OLIGO R2OLIGO 21,850 22,050 22,250 22,450 22,650 22,850 23,050 111 112 113 114 115 116 117 0 0.3 0.6 0.9 1.2 1.5 1.8 2.1 2.4 2.7 3 𝑀 ̅nD P n Lenght of PFR (m) (E) R3-DPN R3-MWN 617 A sensitivity analysis was carried out in the lower section to evaluate the profile of the �̅�n and DPn as the flow passes through the plug flow reactor, that is, the variation in the length of the reactor as shown in Figure 2 (E). It is observed in the graph that both the R3-DPN and R3-MWN curves overlap, as the �̅�n is dependent on the DPn and increases as the length of the plug flow reactor increases, reaching an equilibrium value at the end of the reactor. Both curves have an ascending profile in the reactor from 0 to 1.8 meters, indicating that the reactions and conversion to Nylon 12 are taking place. However, it is observed that at the end of the reactor from 1.8 to 3 meters there is a decrease in �̅�n and DPn, due to the presence of water in the reaction shifting the reaction direction to the reverse direction. 4. Conclusions The results presented from the simulation of hydrolytic ROP of Nylon 12 performed in the Aspen Polymer Plus® simulator demonstrated that the polymerization simulation is viable to be carried out, since it was possible to observe the formation of a polymer with high conversion of 94.41 % and high molar mass of 22.866 g/mol. In the literature, it is reported in the experimental work the formation of Nylon 12 with 99.5 % conversion and molar mass between 5000 g/mol to 9500 g/mol. Evaluating polymer properties such as conversion and molar mass is extremely important, since such properties have a significant impact on the characteristics of the final product and its application. The analysis of the mass flow rate variation profiles with several variables allowed a better understanding of the process. The simulation allowed to identify the relationship between the various polymer variables such as the conversion that decreases with increasing feed mass flow rate, consequently decreasing the �̅�n and DPn and increasing the extraction value. The simulation proved to be important because it allows to carry out studies related to the hydrolytic ring opening polymerization process of Nylon 12 with greater agility. The results allowed a better understanding of the process and can be applied in the optimization of the experimental plant. Acknowledgments The authors would like to thank grant #2015/20630-4 São Paulo Research Foundation (FAPESP), CAPES and CNPq. References Advincula R. C., Dizon J. R. C., Chen Q., Niu I., Chung J., Kilpatrick L., Newman R., 2020, Additive manufacturing for COVID-19: Devices, materials, prospects, and challenges. Mrs Communications, Cambridge, UK, 10, 3, 413–427. Amin D., Nguyen N., Roser S. M. Abramowicz S., 2020, 3D Printing of Face Shields During COVID-19 Pandemic: A Technical Note. Journal of Oral and Maxillofacial Surgery, 78, 8, 1275–1278. Griehl W., Ruesteivei D., 1970, Nylon-12-Preparation, Properties, and Applications. Industrial and Engineering Chemistry, 62, 3, 16–22. Kawakami A., Shimada T., Tsumiyama T., Yamaguchi H., 1994, Continuous Polymerization Method of Laurolactam and Apparatus Therefor. US n. 5362448, 12 oct. 1993, 8 nov. 1994. 179–191. Kyulavska M., Toncheva-Moncheva N., Rydz J., 2019, Biobased Polyamide Ecomaterials and Their Susceptibility to Biodegradation. In: Martínez L., Kharissova O., Kharisov B. (eds) Handbook of Ecomaterials. Springer, 2019. Cap. 118. p. 2902-2927. Mostafa K. G., Montemagno C., Qureshi A. J., 2018, Strength to cost ratio analysis of FDM Nylon 12 3D Printed Parts. Procedia Manufacturing, 26, 753–762 accessed 06.12.21. Patti A., Cicala G., Tosto C., Saitta L., Acierno D., 2021, Characterization of 3D printed highly filled composite: Structure, thermal diffusivity and dynamic-mechanical analysis. Chemical Engineering Transactions, 86, 1537-1542 <10.3303/CET2186257>. Rahim T. N. A. T., Abdullah A. M., Akil H. M., Mohamad D., Rajion Z. A., 2017, The improvement of mechanical and thermal properties of polyamide 12 3D printed parts by fused deposition modelling. Express Polymer Letters , 11, 12, 963–982. Russo S., Casazza E., 2012, Ring-Opening Polymerization of Cyclic Amides (Lactams). In: MATYJASZEWSKI, Krzysztof. Polymer Science: A Comprehensive Reference. Elsevier Science. Chap. 4.14. p. 331-396. Seavey K. C., Liu Y. A.,2008, Step‐Growth Polymerization Process Modeling and Product Design. John Wiley & Sons, 393-461. 618 227umeda.pdf Simulation of Hydrolytic Ring Opening Polymerization of Nylon 12 from Laurolactam