Teaching Experimental Design Using Virtual Laboratories: Development, Implementation And Assessment Of The Virtual Bioreactor Laboratory

Presently there is a need to develop more effective ways to integrate experimental design into the engineering curriculum. To address this need, we are developing virtual laboratories that provide students a capstone experience in which they can apply experimental design in a context similar to that of a practicing engineer in industry. In a virtual laboratory, simulations based on mathematical models implemented on a computer are used to replace the physical laboratory. However, as opposed to being constructed as a direct one-to-one replacement, the virtual laboratory is intended to complement the physical laboratories in the curriculum so that certain specific elements of the experimental design process are addressed. We have previously reported on the Virtual CVD Laboratory, a simulation of an industrial-scale chemical vapor deposition (CVD) reactor. Analogously to the Virtual CVD laboratory, a Virtual Bioreactor laboratory has been developed based on an industrial scale bioreactor process. The development, implementation and assessment of the Virtual Bioreactor in the senior laboratory in Chemical, Biological and Environmental Engineering are discussed. Analysis of student surveys was undertaken to exam student metacognition of the virtual laboratory and compare their ideas of learning to the physical laboratories in the same course. Analysis shows that the experimental design, critical thinking and higher order cognition that are promoted in the Virtual CVD laboratory are manifest in the metacognitive statements of students in the Virtual BioR Laboratory. Both virtual laboratories are available for use upon request. Introduction In a typical laboratory class, students are tasked with taking a set of experimental measurements, analyzing the data, often in the context of underlying theory in the curriculum, and reporting the findings. This work is performed using dedicated equipment physically located in the laboratory. The pedagogical value of the hands-on experience that a laboratory provides is ubiquitously endorsed by educators; however, in practice the engineering laboratory has limitations as well. The traditional mode of delivery requires large amounts of resources for a high quality student experience since students must be supervised and equipment is expensive to purchase and maintain. Moreover, versatile laboratory experiences are needed that can accommodate students enrolled via distance education. Virtual laboratories can overcome these limitations. In a virtual laboratory, students do not interact with real equipment, but rather with use computer simulations of laboratory equipment to obtain data. The virtual laboratory allows future engineers to practice the skills they will need in industry, in much the same way a flight simulator is used for training pilots. Various uses of virtual laboratories in the engineering curricula have been reported. The most extensive deployment of virtual laboratories of chemical processes is an impressive set of modules developed at Purdue University. Seven different laboratories based on traditional chemical engineering processes such as styrene-butadiene copolymerization or hydrogen liquefaction have been used. However, assessment of student learning from these modules has been sparse. We have previously reported on the implementation of a Virtual CVD laboratory, a simulation of an industrial-scale chemical vapor deposition (CVD) reactor. The focus of its instructional design is to complement, not replace, existing physical laboratories. The Virtual CVD laboratory provides a capstone experience in which students apply experimental design in a context similar to that of a practicing engineer with a wider design space than is typically seen in the university laboratory. Specifically, it is designed to allow students to engage more fully in certain aspects of the experimental design process such as: the experimental strategy, the analysis and interpretation of data, and the iterative process of redesign. For clarity, the following distinction of terms is made; the term “experimental design” is used to describe the more general, usually iterative, approach of addressing an open ended problem through experiment while the term “design of experiments” is reserved for that specific statistical methodology . Task analysis of “think-aloud” sessions has verified that students are engaged in the intended, iterative experimental design approach of practicing engineers. Additionally, this laboratory experience was demonstrated to promote higher level cognition in students. This paper addresses the development, implementation and assessment of a second virtual laboratory in the curriculum, the Virtual Bioreactor (BioR) Laboratory, which was offered in addition to the Virtual CVD laboratory. The virtual laboratories studied were delivered as part of the first quarter of the capstone laboratory sequence in the School of Chemical, Biological, and Environmental Engineering. Students completed three laboratories in this course: Ion Exchange (IX), Virtual Laboratory (VL) and Heat Exchange (HX). In the VL unit, students choose either the Virtual CVD Laboratory or The Virtual BioR Laboratory. The hypothesis of this study is that deployment of virtual laboratories in different content areas can be similarly effective, as long as the key elements of instructional design are incorporated. In particular, we wish to determine if the area of experimental design, critical thinking and higher order cognition that are promoted in the Virtual CVD laboratory are also promoted in the Virtual BioR Laboratory. The preliminary assessment is based on analysis of a student survey. While the overall goal in assessment of this project is to determine the ways that students learn key cognitive processes and specific domain content in a virtual environment, the preliminary assessment reported in this paper does not compare and contrast the different amount of learning achieved by the students in a virtual laboratory experience with that learned in two typical hands-on laboratory experiences. Rather, the intent of this preliminary analysis is to describe the differing student perceptions of the learning that they were to take away from the three different laboratory experiences. Students’ perceptions of the learning intentions of three different laboratory experiences provide a lens into their metacognitive processes. Metacognition as a regulatory activity involves students thinking about their thinking in a way that externalizes their perceived knowledge gain and knowledge awareness. Research in metacognition in engineering education has demonstrated the efficacy of providing students with learning environments that enhance students’ regulation of their own learning. This research sought to identify the ways that student knowledge and awareness of their own learning might evolve as they move through three structured laboratory experiences. The intent of the research is to demonstrate that the virtual laboratory provided a context in which the students’ perception of the laboratory experience would move away from acquisition of technical skills and application of bounded knowledge to using conceptual systems to generalize problem solving beyond the immediate context of the laboratory problem. The perspective of formative assessment processes indicates that student self-assessment defines what students understand about the goals and objectives of their learning experiences. Student understanding of the goals of learning experiences is a critical element in student acquisition of the content understanding and deep cognitive and procedural skill development in higher education. Metacognition as the process of students monitoring their own learning is an important element of student learning in the engineering context. Simulation and Software Design The Virtual BioR Laboratory is based on an industrial stirred-tank fed-batch bioreactor, as shown in Figure 1. The bioreactor can be used for different functions, such as production of a product or degradation of waste. The sequence of events that occur in the bioreactor include cleaning and sterilizing the bioreactor, loading with sterile medium, inoculation with the desired cell line, batch-growth on substrate, followed by fed-batch growth where new medium is fed to the bioreactor and the volume increases with time. Finally, the run is stopped, and the contents are emptied from the bioreactor. The simulation of the Virtual BioR is based on a mathematical model that accounts for the kinetics of the different processes that occur. Since real systems do not deterministically adhere to fundamental models, random process and measurement variation is added to the output. The mathematical model and the software architecture used in the bioreactor simulation are presented in Appendix A. The instructor interface allows the instructor to access the Virtual BioR through the web via an instructor login. The instructor can create and monitor students’ accounts (username, password, simulation). By modifying the MatLab simulation files, each account can use a different set of instructor specified parameters, such as temperature optimum, degree of substrate inhibition, maximum specific growth rate, etc. The parameters that characterize the simulated cultivation are stored in the MatLab files, so each student could be running a bioreactor with a different