Simulation Approach to Learning Polymer Science - Journal of

Article Cite This: J. Chem. Educ. XXXX, XXX, XXX−XXX

pubs.acs.org/jchemeduc

Simulation Approach to Learning Polymer Science Harith H. Al-Moameri,*,†,‡ Luay A. Jaf,† and Galen J. Suppes† †

Department of Chemical Engineering, University of MissouriColumbia, W2033 Lafferre Hall, Columbia, Missouri 65211, United States ‡ Department of Materials Engineering, Mostansiriyah University, Baghdad, Iraq

Downloaded via UNIV OF WOLLONGONG on August 3, 2018 at 16:28:35 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

S Supporting Information *

ABSTRACT: Traditional engineering textbooks are substantially limited by approaches relying on analytical solutions where most of the applications are described by differential equations without simple analytical solutions. An alternative is to design a textbook around computer-program-based simulations, where the simulations progressively evolve through the textbook with the complexity of the content. Simulations can solve more complex problems with immediate utility in practical applications, and these capabilities are able to achieve higher levels of Bloom’s Taxonomy with quiz-based methods that can be performed in a few minutes. By example, the polymer engineering topic of thermoset polymerization has a global market of tens of billions of dollars; yet, the textbook coverage of the subject is sparse due to the inability of simple analytical solutions to provide useful solutions to practical applications. This paper provides a foundation and approach for a simulation-based-learning textbook to cover the important topic of polymer engineering which will also bridge the gap between student and researcher. A questionnaire was passed to two groups of college students to identify their response to the simulation-based learning of the subject of polymer engineering. The majority of students recommended this method. KEYWORDS: Polymer Chemistry, Computer-Based Learning, Kinetics, Polymerization, Upper-Division Undergraduate

■

• applications of common polymers, (R) • methods of experimental investigation and manufacturing of polymers (e.g., cup foaming, small-box foaming, reaction injection molding, extrusion). (R) This paper provides methods that can attain higher levels of Bloom’s Taxonomy, engage visible and active learning modes, and provide commercially important results. These simulationbased-learning methods include activities of (a) writing computer programs, (b) evaluating problem statements to convert problems specifications to computer program input parameters, (c) using the computer programs to solve problems, and (d) adding incrementally to computer programming code to solve progressively more complex problems. Computer programming efficiently summarizes information, and the underlying algorithms of computer code are creative by nature. This results show learning achievement at “Analysis”, “Evaluation”, and “Creation” levels in exercises that can be completed in a matter of minutes. The following are specific examples of activities with an indication of the Blooms Taxonomy level of learning associated with each activity: 1. Using prewritten programs having less than 15 lines of code allows students to generate meaningful quantitative

TRADITIONAL VERSUS SIMULATION LEARNING Traditional approaches to learning polymer engineering have produced dozens of textbooks with emphases on the “Remember” (R) and “Understand” (U) levels of Bloom’s Taxonomy. The common subjects coved by traditional polymer textbooks including, but not limited to • common monomers and their structure, (R) • classifications of polymers (thermosets versus thermoplastics), (R) • definitions and identifications of key polymer properties (e.g., molecular weight, degree of polymerization, glass transition temperature, crystallinity, etc.), (R) • types of polymerization reactions (addition, condensation, etc.), (U) • common catalysts and typical catalyst mechanisms (chemistry perspective), (R) • additives and their impact on the final properties of the polymer (e.g., fillers, surfactants, chain extenders, crosslinkers, etc.), (R) • analytical methods for characterizing polymers (e.g., molecular weight analysis, molecular Structure, morphology, and physical properties), (R) • equations for calculating properties or characterizations (e.g., degree of polymerization, the extent of reaction, equivalent weight, hydroxyl number), (U) • mechanical properties of common polymers, (R) © XXXX American Chemical Society and Division of Chemical Education, Inc.

Received: April 5, 2018 Revised: July 6, 2018

A

DOI: 10.1021/acs.jchemed.8b00236 J. Chem. Educ. XXXX, XXX, XXX−XXX

Journal of Chemical Education

Article

level of learning at the end of the class. The results reveal a positive impact of using animations along with traditional learning as an effective instructional tool for enhancing chemistry learning. The research ends up with a conclusion that teachers nowadays need to use more than traditional techniques to get the attention of students. GeoGebra dynamic worksheets were created by Kostic et al.5 for solving quantitative composition problems. The software offers interactive connections with concrete and abstract visual representations of the problem. The outcomes of using this software on an experimental group of chemistry students were statistically significant. The results show better improvement of students’ ability to estimate quantitative relationships in a wellthought-out manner, as well as to have critical anticipation about the obtained results. Stereogame, an interactive computer game, was developed to engage undergraduate students on the topic of stereochemistry.6 The game assists students to answer 230 questions of three difficulty levels. A positive response was obtained from students who played the game. Python programming was used to assist the student to get a better understanding on topics of computational chemistry and in solving Schrödinger’s equation.7,8 The program saves the time of solving equations and provides processing, analyzing, and data visualizing which provide better engagement to students and better understanding of the topics Fasoula et al.9 developed a series of Microsoft Excel spreadsheets for simulating the process of separation optimization as part of advanced analytical chemistry course. Simple and easily implemented macros were used in performing the optimization process. This paper includes MatLab code for simulating thermoset polyurethane polymerization reactions; the initial code is based on Flory’s assumptions.10 MatLab is a programming platform that integrates computation, visualization, and programming in an easy-to-use environment. MatLab was adopted for this simulation because it is easy to learn, is normally available for students in universities, and provides a program that can be readily extended to greater complexity. While simulation reaction kinetics using MatLab have been published in many education papers, the current paper uniquely provides the following: 1. an alternative approach to learning polymer engineering, 2. programming code that provides immediate results and can be modified to simulate increasingly complex processes, 3. a visual presentation of how polymerization progresses in terms of a variety of meaningful properties, and 4. example problem statements that can be solved using simulation.

output in a matter of minutes with a little background on the subject (Apply). 2. Generating concentration, temperature, and viscosity profiles (as a function of time) provides a visible representation of reaction progress which helps students understand how polymer reactions proceed (Analysis). 3. Entering input parameters into the computer code to generate an output of immediate practical value (e.g., gel point time) engages students in an active learning process (Evaluate). 4. Entering of input parameters can include “trial and error” (“pattern recognition” or other) input to optimize formulations or meet performance specifications (Evaluate). 5. Adding or modifying computer code to simulate increasingly complex phenomena (e.g., catalysis) achieves creationlevel learning in an exercise that can be completed in a few minutes (Create). 6. Extending computer simulation to predict physical properties of polymer devices (e.g., thermal conductivity, compressive strength) extends the creation process to visible association with devices/products with which the students are familiar (Create). Active learning, visible learning, and achieving higher levels of learning all lead to improved retention by the students. The simulation approach to learning allows more variables and properties to be related to the progress of the polymerization process.

■

LITERATURE SURVEY ON SIMULATION LEARNING Chemistry and chemical engineering educators have been interested in incorporating software in student interactive learning since microcomputers were first available. A computer-assisted learning program was developed by Reid and Wheatley1 to teach molecular reaction dynamics as an advanced course in chemistry for undergraduate students. The goals of the program were similar to the program of the current paper which (a) developed student’s understanding by providing visual animation and (b) allowed students to develop a physical intuition of the impact of each initial condition on the outcome results. The program was presented in a simple way where students do not need advanced knowledge of reaction dynamics and was built with information familiar to the student. LUCID (learning and understanding through computerbased interactive discovery) software was incorporated for promoting student engagement in the learning process.2 LUCID’s interactive models, easy-to-use tools, multilevel feedback, network reporting, peer assessment, and performance distributions significantly enhance the learning ability of students. The software provides interactive models to increases students’ engagement, interest, and team working. Xie et al.3 described an open-source program called “Molecular Workbench” to simulate molecular dynamics and build a model capable of generating chemical reactions kinetics for classroom use. The software was limited to specific types of reactions with which undergraduate students are familiar and ignore some chemical reaction aspects toward simplicity. The software generates dynamic 2D visual representations to assist students to reach the better understanding level of the connections between chemical reaction equations and atomic interactions. The impact of associating animations in organic chemistry on students’ comprehension was studied.4 A designed test (Organic Chemistry Visualization Test) was utilized to measure student

■

DISCUSSION

Introduction to Thermoset Kinetics and Simulation

Polyurethane thermoset polymerization reaction has been chosen as an example for this polymer simulation. Polyurethane is a high-value thermoset polymer with a range of applications due to its high performance and versatility.11 Typically, polyurethane is produced from the exothermic stepgrowth polymeric reaction of an isocyanate and polyol monomers12 according to eq 1 where R and R′ are long organic chains that contain additional isocyanate and alcohol moieties. B



Article

ODEs. The solution of these ODEs is achieved through the initial value method of Matlab’s ODE45 function. ÉÑ ÑÑ dC A mon ÑÑ rA mon = = −r1 − r2 ÑÑ ÑÑ dt ÑÑ ÑÑ ÑÑ dC Bmon ÑÑÑ rBmon = = −r1 − r3 ÑÑ dt ÑÑ ÑÑ ÑÑ dC P ÑÑ rP = = r1 − r4 ÑÑ dt ÑÑ ÑÑ ÑÑ dC A moi Ñ rA moi = = −r1 − r2 − r3 − r4 ÑÑÑÑ ÑÑ dt ÑÑ ÑÑ dC Bmoi ÑÑ rBmoi = = −r1 − r2 − r3 − r4 ÑÑÑ ÑÑÖ dt (2)

RNCO + R′CH 2OH isocyanate

polyol

→ RNHCOOCH 2R

(1)

PU

For polymeric reactions of organic materials, the rate of reaction is not dependent upon the concentration of compounds, but, rather, depends on the concentration of moieties (functional groups) of compounds. The moieties are defined as the functional groups (e.g., an alcohol moiety) within the polymer molecule that are responsible for polymer-bond-forming chemical reactions. To handle polymerization reactions (with a reasonable set of reactions), elementary reaction mechanisms and resulting kinetic expressions should be based on moieties rather than monomers (compounds). Table 1 lists the basic Table 1. Polyurethane Basic Reactions Reaction Number

Reaction

Rate Expression

1 2 3 4

Amoi + Bmoi → P Amoi + PBmoi → P Bmoi + PAmoi → P PAmoi + PBmoi → P

r1 = k1 fACA f BCB r2 = k2 fACACPB r3 = k3CPA f BCB r4 = k4CPACPB

Here, r is the rate of change. The simulations of this paper are introduced in two phases. The first phase is a basic simulation of isothermal reactions. This simulation provides an output of concentration profiles of the monomers, polymer, and moieties; times at which the gel point occurs; impact of inter/intra polymer−polymer reactions on the gel point time; and profile of the degrees of polymerization. The second phase includes nearadiabatic reactions where the rate constants of reactions are calculated from the Arrhenius equation. To achieve near-adiabatic simulation, a rate of change of temperature was added. An empirical equation was introduced to approximate viscosity as a function of the degree of polymerization, and the impact of catalyst was also considered. The results of the two simulation phases are introduced for learning purposes and did not compare with experimental data. The following basis and heuristics were adopted for the simulation of this paper: 1. There is a 1-L basis at a constant density. 2. Monomer concentrations are followed where monomers are consumed when any moiety on the monomer reacts. 3. Polymers are indistinguishable once formed. 4. Polymer concentration is calculated but not used in any rate expression to avoid the propagating and compounding of any error in the polymer concentration.

reactions and rate expressions of the moieties of a multifunctional isocyanate and polyol monomers based on Flory’s assumption. Initial theories of gelation and cross-linking were developed by Paul Flory, a 1974 Nobel laureate on this subject. In an approach targeting the analytical solution of equations, rather than a computer simulation, Paul Flory and Walter Stockmayer based this theory on the following three assumptions: • All reactions occur between A and B moieties. • All functional groups on a branch unit are equally reactive. • There are no intramolecular reactions. The first of these assumptions is implicit by not including reactions beyond those provided in Table 1. The second of these assumptions is equivalent to the constraint that k1 = k2 = k3 = k4. The last assumption impacts the way reaction 4 impacts the rate of change of polymer molecules. Table 2 shows a minimal set of concentrations (and their initial values) that must be calculated to simulate polymerization

Simulation of Basic Isothermal Reactions

The computer programming code for the basic isothermal stepgrowth polymerization reaction was written in a Matlab’s Live Editor (“Isothermal”, available as Supporting Information). In this simulation example, the rate constants were assumed to be the same for all reactions and are not changing with temperature (isothermal conditions). A function (floryrxn) was written to calculate the rate of reactions and the rate of change of concentrations of A monomer, B monomer, polymer, A moieties, and B moieties, respectively. Step 1 of this example was generating a visual representation of the concentration profiles as shown in Figure 1. Polymer concentration becomes negative at a time of about 62 s. This is not realistic; it is an artifact of the numeric method. The simulation results after the polymer concentration become negative are all in error. One approach to prevent this error is to force a very small step size. An alternative approach is to set the ODE solver options so that the concentration variable cannot become negative by using the “odeset” function. This can be done by uncommenting lines 8 and 10 and commenting line 9. The resulting concentrations are shown in Figure 2.

Table 2. Concentration Profiles and Initial Conditions That Define a Simplified Liquid Polymer Reaction system Component

Initial Condition

Description

CA CB CP CAmoi CBmoi

CA0 CB0 0 fACA0 f BCMB0

Molarity of monomer A Molarity of monomer B Molarity of polymer Equivalence of moiety A Equivalence of moiety B

of polyurethane; these are calculated by solving the ordinary differential equations (ODEs) that describe their rates of change. Atomic balances allow addition concentrations of moieties on the monomers (CMA and CMB) and moieties on polymers (CPA and CPB) to be calculated as fACA, f BCB, CAmoi − fACA, and CBmoi − f BCB, respectively. It is necessary to calculate these latter four calculated quantities because they appear in the ODEs of eq 2. The rate expressions for the components of Table 2 are described in terms of the rate expressions of Table 1 by the eq 2 C



Article

unreacted monomers are in close proximity (on the same molecule) they will tend to react in intrapolymer reactions. The topic of intramolecular reactions is addressed by estimating the fraction of polymer−polymer reaction by introducing the fraction “intra” into the rate of change of the polymer concentration (step 2). To incorporate intramolecular reaction, uncomment lines 4 and 32 and comment line 31. Polymer self-reactions (intramolecular) reduce the rate at which moiety reactions reach the gel point. The inclusion of this in the model will lead to higher critical extents of reaction at the gel point. The sensitivity analysis of Figure 3 shows that increasing

Figure 1. Visual representation of concentrations profile of isothermal polymerization reactions.

Figure 3. Impact of fraction intrareactions on polymer concentration and gel point time. This figure was generated by drawing polymer concentration data of three runs using MS Excel.

the fraction of intra-polymer−polymer reactions increases the time of the gel point. The learning output of this figure is on how the fraction of intrapolymers reaction impacts polymer concentration and gel point time. This type of information cannot be obtained from traditional textbook approaches. The concentration profile of monomers and polymer allows the calculation of the average degree of polymerization (DP) during the reaction. This step can be done by uncommenting lines 16−20, and the results for DP are shown in Figure 4. Figure 2. Visual representation of concentrations profile of isothermal polymerization reactions.

Initially, the polymer concentration increases due to its formation. As more polymer is present and more of the reactive functional groups are attached to that polymer, polymer− polymer reactions occur leading to a decrease in polymer concentration. Eventually the cross-linking leads to a concentration that asymptotically approaches zero. The polymer concentration of zero represents the time at which polymer segments cross-link into one large molecule. This is the gel point, the time of which is an important specification in industrial processes. The results of Figures 1 and 2 provide immediate output in a visible format that can provide improved learning and retention. The results also provide an approximate to the gel point time which is not possible without simulation. Unfortunately, an isothermal reaction is not realistic for these systems, and gel point estimates of practical value are only possible with the nearadiabatic code provided later in this paper. Flory’s assumption of no intrapolymer reactions (third assumption) becomes increasingly inaccurate as the gel point is approached and with increasing functionalities of A and B monomers. As for functionality increases, the number of unreacted monomers on polymer increases, and since these

Figure 4. Visual representation of the degree of polymerization profile.

Virtual time estimates of the degree of polymerization provide visible results, with the profiles providing more meaningful learning than single point estimates possibly used in traditional textbook approaches. D



Article

Simulation of Near-Adiabatic Reactions

The learning outcome of simulating temperature profiles is a virtual-time realization of how values of thermodynamic and kinetic parameters impact the temperature. Temperature profiles contain valuable information, such as an approximation of extent of reaction, and can be readily verified in experimental studies. An important addition to “Adiabatic” is the calculation of viscosity. Studying viscosity change of thermoset polymerization reaction is highly important for a number of reasons (e.g., a measure of the progress of polymerization, flow dynamics in molds, practical realization of the gel point). From a simulation perspective, the viscosity is a key parameter that ultimately impacts diffusion rates. Diffusion rates impact reaction rate constants and mass transfer. Viscosity profiles can be added by uncommenting lines 21−27, and the results are shown in Figure 6.

“Adiabatic” simulation code (available as Supporting Information) is a straightforward application of Flory’s assumptions, and as such, it is a good starting point for a textbook. Near-adiabatic implies that only a small fraction of the heat is lost before the gel point is reached due to low surface-area to volume ratios of molds used to make thermoset devices. To simulate near-adiabatic polymerization, the Arrhenius equation (eq 3) is used to calculate the reactions rate constant (k) of the reactions in Table 1. k = k 0e−E0 / RT

(3)

Here, k0 is the pre-exponential factor, E0 is the activation energy, and T is the reaction temperature in kelvin. Also, the rate of change of the temperature must be added as summarized by eq 4 (based on the heat balance). The heat balance equation includes a term for the loss of heat to the mold. ∑(ΔHiri) + U Ar(TS − T ) dT = dt mC P

(4)

Here, ΔH is the heat of reaction, U is the overall heat transfer coefficient, Ar is the area, TS is the surrounding temperature, m is the mass, and CP is the heat capacity. To simplify the simulation, the heat of reaction of A and B moieties was assumed to be the same whether these moieties were attached to a monomer or a polymer. Thermodynamic properties in this simulation are assumed to be constants and are not changing with temperature. The overall heat transfer coefficient “U” is approximated as a constant. The area (Ar) for heat transfer can be incorporated into U or considered separately. The ambient temperature “TS” was added for heat loss calculation and separated from the initial condition (of monomers) temperature “T0”. The resulting output of temperature profile is shown in Figure 5. Temperature starts from the initial value of 295.15 K and increased Figure 6. Visual representation of viscosity profile of resin during the reaction.

The viscosity of the resin is increasing by several orders of magnitude during the reaction, eventually going toward infinity as DP goes to infinity. A variation of code would use polymer degree of polymerization (PDP) rather than DP with the choice of PDP versus DP being a topic of academic discussion. Students can use this figure to understand the dramatic change of thermoset viscosity during the reaction and to identify the time at which the gel point occurs. The results provide an approximate to the gel point time that approaches realistic conditions and is thus useful for practical applications. Catalysts are commonly used in polymerization reactions to increase the rates of reactions by reducing the activation energy required for the reactions. Catalytic reactions were reported to follow chain-growth mechanism as opposed to the step-growth mechanism of homogeneous reaction;13 however, for simplicity, a step-growth mechanism is easier to incorporate.14 To consider the impact of catalyst, uncomment lines 30−33 and line 36 and comment line 35. Figure 7 provides a comparison of reaction temperature at different catalyst loadings. The increase in the rate of reactions results in the faster temperature profile. The learning outcome of this example is an improved understanding of the role of catalysis and how catalysts impact the rate of reactions and gel point time. Possible advanced assignments on this topic would include considering a chain-growth mechanism for catalytic reactions.

Figure 5. Visual representation of temperature profile.

as heat generated from the exothermic reactions. Initial slope of the temperature profile is highly impacted by the assumed value of the pre-exponential factors of the Arrhenius equation; the curve hump is highly impacted by the assumed value of activation energies. The maximum reaction temperature depends on heat of reactions, and the slope of the last part of the temperature curve depends on the overall heat transfer coefficient. E



Article

current Matlab codes and add or modify some lines in the simulation to get the answer in a few seconds. The Supporting Information file “Example Textbook Problem Statements” shows two questions for isothermal and near-adiabatic simulations with question statements and the answer in the form of the polynomial equation. The second example provides example questions based on the use of simulation code including questions that target different levels of learning. All questions use a common starting sentence that introduces the application program code. Extended Simulation Capabilities

An important advantage of the simulation code presented in this paper is the ability to incrementally expand the code to simulate the most common industrial applications. Table 3 lists advanced topics that are within the capabilities of extended simulation code. As the complexity of the programming increases, the need to organize the program through the use of functions and overall algorithms becomes increasingly important. Figure 8 is a block diagram showing the “Adiabatic” programming code. For this example, the recipe and physical property estimates are provided in lines of code. The only function used in the program was ODE45 which solves the ordinary differential equation initial condition problem and returns concentrations and temperature profile solutions. As the recipes and database become more comprehensive it is advantageous to place them in separate functions or databases that are called upon by the Script code. Recipes and database include initial monomer concentration, monomer functionalities, and kinetic and thermodynamic properties. The “floryrxn” function calculates reaction rate constants, the mass balance of moieties, rates of reactions, and rate of change of components. Figure 9 provides a block diagram of a more advanced simulation. In this block diagram, ODE45 function calls on the derivatives of concentration profile, reaction temperature, and

Figure 7. Impact of catalyst loading on the reaction temperature. This figure was generated by drawing reaction temperature data of three runs using MS Excel.

Example Textbook Problem Statements

The key point of this paper is to identify how a textbook on polymer engineering can be designed around the use of simulation of polymer-forming reactions. Goals include attaining higher-level learning outcomes and capabilities that are more directly applicable to practical applications. The use of these simulations in a course on polymer engineering provides for the efficient specifying of a problem statement having an abundance of parameters and fundamental equations, and the program is able to transform these fundamental parameters and equations into meaningful visible output. This simulation approach far exceeds what is possible in traditional approaches putting forth equations which lack the tools needed to transform the equations into results to which the students can better relate. Instructors can give simulation-based questions to students as short question statements in common online learning platforms such as Canvas and Blackboard. Students are able to use the

Table 3. List of Advanced Topics, Approach Method, and Their Corresponding Learning and Operational Utility Advanced Topics (Method)

Learning Outcome (L) and Operational Utility (O)

Fractions of primary and secondary moieties15 (primary and secondary moieties reactions)

Identification of fractional content of primary vs secondary moieties (L/O) Comparison of different monomers (O) Identification of the mechanism of the catalytic reaction (L)

Chain-growth mechanism for catalytic reactions13 (catalyst coupling/ decoupling reactions)

Chain-growth mechanism resulting in higher viscosity (L) Identification catalyst-moiety attachment (O) Impact of viscosity increase on the rate of diffusion (L)

Mass transfer limitation16 (impact of diffusion as reversible reaction or on the frequency factor)17

Identification of reaction/diffusion limited regions (L) Evaluation of the effectiveness of similar reactive monomers of different viscosities (L/O) Calculation of rate of diffusion and diffusion (L)

Chemical/physical blowing agent (BA)18,19 (blowing agents fn − rate of BA evaporation)

Identification of fraction of BA diffuses vs retained (L/O) Prediction of cell diameter, the pressure inside cells, foam density, and foam shrinkage (L/O) Increase of viscosity during the reactions (L/O) Prediction of gel point time (L/O) Increase in heat of capacity during the reactions (L) Impact of monomers heat capacities (L/O) Impact of surfactant on blowing agent efficiency and cell diameter (L/O) Factors impacting final properties of the foam (L/O)

Viscosity profile of resin20 (viscosity fn − group contribution) Heat capacity profile (heat capacity fn − group contribution) Impact of surfactant (surface tension calculation) Prediction of physical properties21 (thermal conductivity and compressive strength fn)

Prediction of the final properties of the foam (O) F



Article

Figure 8. Block diagram of the “Adiabatic” simulation code.

Figure 9. Block diagram of an advanced simulation.

rate of blowing agent evaporation in the “floryrxn” function. In addition, ODE45 is exchanging concentrations and temperature of each iteration with the viscosity and heat capacity function of the resin and accounts for its impact on reactions. This block diagram is subject to future extension to account for the impact of surfactant and fire retardant on the reaction kinetics and/or on the final properties of the foam. A well-written code allows the functions and database to be modified without impacting the Script or other functions.

3. Does the visual representation of the variables provide a better understanding of the subject of polymer reaction? 4. Do you recommend using the simulation approach of learning as a learning method in future subjects (polymer science and other classes)? 5. Does simulation-based learning help you to learn complex topics that would be difficult to understand using traditional methods? 6. Do you recommend other students to take simulationbased classes? A response to these questions of “Strongly Agree”, “Agree”, and “Neutral” were obtained for an average of about 48 strongly agree, 9 agree, and 1 neutral student responses toward simulation-based learning as an effective learning method for education as shown in Figure 10.

Student’s Response

Two groups (31 and 27) of third-year students at the Materials Engineering Department/Mustansiriyah University studied the simulation-based-learning of thermoset polymer engineering as illustrated in the current study. Students were asked to fill a questionnaire to measure student response to the simulation-based method. The questionnaire had the following questions:

■

CONCLUSIONS Use of simulation code as a learning platform provides the ability to achieve higher-level learning outcomes in problem statements that can be solved in a few minutes as compared to days or weeks in alternative platforms. For polymer engineering, simulation code is able to achieve what is substantially not attainable in

1. Does the simulation approach of learning make the subject more interesting compared to the traditional learning? 2. Does the simulation approach of learning make the subject of polymer reaction easier compared to traditional learning? G



Article

representations in organic chemistry. Chem. Educ. Res. Pract. 2014, 15 (1), 47−58. (5) Kostić, V. D.; Stankov Jovanović, V. P.; Sekulić, T. M.; Takači, D. B. Visualization of problem solving related to the quantitative composition of solutions in the dynamic GeoGebra environment. Chem. Educ. Res. Pract. 2016, 17 (1), 120−138. (6) da Silva Júnior, J. N.; Sousa Lima, M. A.; Xerez Moreira, J. V.; Oliveira Alexandre, F. S.; de Almeida, D. M.; de Oliveira, M. d. C. F.; Melo Leite Junior, A. J. Stereogame: An Interactive Computer Game That Engages Students in Reviewing Stereochemistry Concepts. J. Chem. Educ. 2017, 94 (2), 248−250. (7) Weiss, C. J. Scientific Computing for Chemists: An Undergraduate Course in Simulations, Data Processing, and Visualization. J. Chem. Educ. 2017, 94 (5), 592−597. (8) Srnec, M. N.; Upadhyay, S.; Madura, J. D. A Python Program for Solving Schrödinger’s Equation in Undergraduate Physical Chemistry. J. Chem. Educ. 2017, 94 (6), 813−815. (9) Fasoula, S.; Nikitas, P.; Pappa-Louisi, A. Teaching Simulation and Computer-Aided Separation Optimization in Liquid Chromatography by Means of Illustrative Microsoft Excel Spreadsheets. J. Chem. Educ. 2017, 94 (8), 1167−1173. (10) Flory, P. J. Molecular size distribution in three dimensional polymers. I. Gelation. J. Am. Chem. Soc. 1941, 63 (11), 3083−3090. (11) Peacock, A. J.; Calhoun, A. Polyurethanes. In Polymer Science; Calhoun, A. J. P., Ed.; Hanser, 2006; pp 365−381. (12) Szycher, M. Szycher’s Handbook of Polyurethane; CRC Press, 2013. (13) Ghoreishi, R.; Suppes, G. J. Chain growth polymerization mechanism in polyurethane-forming reactions. RSC Adv. 2015, 5 (84), 68361−68368. (14) Zhao, Y.; Suppes, G. J. Simulation of Catalyzed Urethane Polymerization: An Approach to Expedite Commercialization of Biobased Materials. Catal. Surv. Asia 2014, 18 (2−3), 89−98. (15) Ghoreishi, R.; Zhao, Y.; Suppes, G. J. Electronic cigarette use among patients with cancer: Characteristics of electronic cigarette users and their smoking cessation outcomes. Cancer 2015, 121 (5), 800. (16) Al-Moameri, H.; Ghoreishi, R.; Suppes, G. Impact of inter- and intra-molecular movements on thermoset polymerization reactions. Chem. Eng. Sci. 2017, 161, 14−23. (17) Al-Moameri, H.; Jaf, L.; Suppes, G. J. Viscosity-dependent frequency factor for modeling polymerization kinetics. RSC Adv. 2017, 7 (43), 26583−26592. (18) Al-Moameri, H.; Zhao, Y.; Ghoreishi, R.; Suppes, G. J. Simulation of liquid physical blowing agents for forming rigid urethane foams. J. Appl. Polym. Sci. 2015, 132 (34), 42454 (1 of 7).. (19) Al-Moameri, H.; Zhao, Y.; Ghoreishi, R.; Suppes, G. J. Simulation Blowing Agent Performance, Cell Morphology, and Cell Pressure in Rigid Polyurethane Foams. Ind. Eng. Chem. Res. 2016, 55 (8), 2336− 2344. (20) Fu, Z.; Suppes, G. J. Group contribution modeling of viscosity during urethane reaction. J. Polym. Eng. 2015, 35 (1), 11−20. (21) Al-Moameri, H.; Ghoreishi, R.; Zhao, Y.; Suppes, G. J. Impact of the maximum foam reaction temperature on reducing foam shrinkage. RSC Adv. 2015, 5 (22), 17171−17178.

Figure 10. Student response toward simulation-based learning.

traditional textbook approaches. Outcome examples from simulation-based learning include estimating gel point times for nonisothermal polymer-forming processes, evaluating the impact of catalysts on gel point times, and evaluating how recipes may or may not reach critical maximum temperatures as needed to fully react the monomers in the recipe. The paper has provided example code, example simulation output, example problem statements, and a summary of how the capabilities of the simulation can be incrementally expanded. The capabilities of this learning methodology represent an important and needed advance, especially on the topic of polymer engineering.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available on the ACS Publications website at DOI: 10.1021/acs.jchemed.8b00236. Examples of texbook problem statements (PDF) “Isothermal” and “Adiabatic” simulation condes in MatLab Live Editor format (ZIP)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected] or almoamerih@gmail. com. ORCID

Harith H. Al-Moameri: 0000-0001-5985-0235 Luay A. Jaf: 0000-0001-8363-8365 Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS The authors thank the Higher Committee for Education Development (HCED) in Iraq for their financial support. REFERENCES

(1) Reid, K. L.; Wheatley, R. J.; Horton, J. C.; Brydges, S. W. Using Computer Assisted Learning to Teach Molecular Reaction Dynamics. J. Chem. Educ. 2000, 77 (3), 407−409. (2) Wolfskill, T.; Hanson, D. LUCID: A New Model for ComputerAssisted Learning. J. Chem. Educ. 2001, 78 (10), 1417−1424. (3) Xie, Q.; Tinker, R. Molecular Dynamics Simulations of Chemical Reactions for Use in Education. J. Chem. Educ. 2006, 83 (1), 77−83. (4) Al-Balushi, S. M.; Al-Hajri, S. H. Associating animations with concrete models to enhance students’ comprehension of different visual H


Simulation Approach to Learning Polymer Science - Journal of

Recommend Documents