(263e) Integrated Molecular Modeling Education in the Chemical Engineering Curriculum

Given the critical importance of thermodynamic and process modeling, these concepts have been successfully integrated into the chemical engineering curriculum. However, as in-silico design gains credibility in engineering, molecular modeling is becoming more useful in chemical engineering. Molecular modeling with never replace experimental synthesis and characterization, just as process modeling will never completely replace pilot plant experiments. However, molecular modeling can provide insight into the design of optimal chemical and material systems, just as process modeling provides a basic design before final optimization of the process. It is important for chemical engineering students to see the value of molecular modeling in their field as its use continues to grow. Unfortunately, the integration of molecular modeling into the chemical engineering curriculum lags behind that of process simulation despite its increasing importance. We have studied molecular modeling education in various classes with chemical engineers including: a specialty elective in applied molecular modeling, molecular modeling modules in a chemical product design course and a chemical engineering numerical methods course, and molecular modeling modules in a polymer science course and laboratory. Our experience has taught us that the most effective approach to teaching this subject employs active learning in which the students carry out applied molecular modeling exercises in class and for assigned projects. The biggest challenge using this approach is a lack of scaffolded computational infrastructure. Process simulation software like ASPEN, or numerical simulation platforms like MATLAB or COMSOL possess an ease of entry for students due the their advanced user interfaces. Chemical engineers find it more challenging to use academic molecular modeling programs that have more primitive interfaces. This is analogous to the command line interface of the original academic version of ASPEN. Some commercial platforms such as Biovia’s Materials Studio have efficient easy-to-learn user interfaces, but do not allow sufficient access to the underlying algorithms and models to facilitate fundamental learning by the student.

We have found that the Molecular Operating Environment (MOE) from Chemical Computing Systems addresses both these challenges. MOE allows students access to fundamentals while simultaneously providing a high level vector-based coding language to write their own simulation and analysis applications without having to code low-level geometric and energetic routines. These qualities allow us to carryout active learning in the classroom where students can run, code, and analyze molecular simulations in class to learn both fundamentals and relevant engineering applications. Also, because MOE integrates across the relevant scales of molecular modeling from quantum mechanics, through molecular modeling, and up to meso-scale models, it can be used to show students the importance of multi-scale modeling in this field. We will present examples of how this integrated approach works in practice using our current classroom model that employs both active learning and a flipped classroom mode. Our flipped classroom displaces the typical lecture on modeling fundamentals from the classroom to pre-class videos. This allows us to carry out actual molecular simulations and analysis in class to illustrate how molecular modeling works and why it is valuable. Without this optimal platform such an approach would not be possible. Further, we utilize some of our recent work on cognitive load theory to formulate optimal modeling examples in this class. Previously, we showed that the optimal choice of a programming examples in a numerical methods course can maximize the germane cognitive load (mental effort used in learning), while minimizing extraneous cognitive load (mental effort that distracts from learning). We illustrate how this optimal choice can potentially improve learning outcomes in a numerical modeling classes.