Downloaded by UNIV ILLINOIS URBANA-CHAMPAIGN on November 10, 2016 | http://pubs.acs.org Publication Date (Web): August 16, 2016 | doi: 10.1021/bk-2016-1217.ch003
Chapter 3
Generalized Organic Chemistry: Teaching Chemistry Using a Framework Approach for a MOOC Audience Michael J. Evans* School of Chemistry and Biochemistry, Georgia Institute of Technology, Atlanta, Georgia 30313, United States *E-mail:
[email protected]
Organic chemistry pedagogy is changing as educators are coming to understand and better appreciate the conceptual organization of the field and how to teach so that students’ knowledge becomes transferrable. As part of a second-semester organic chemistry course on campus, we have developed a generalized approach to teaching organic chemistry that makes use of the flipped classroom model, computational tools, and novel conceptual frameworks. In translating this course to a MOOC environment, we have found that MOOC students in particular respond well to the generalized approach, which helps students learn in a context-independent manner and transfer their knowledge to other fields.
Shifting Organic Chemistry Pedagogy Organic chemistry occupies a critical position in the curricula of college students with majors in the sciences, engineering, and health-related fields. As the last chemistry course that many of these students will take, organic chemistry can define a student’s view of chemistry for years after graduation. Furthermore, organic chemistry is directly relevant to health science professionals who must grapple either directly or indirectly with biochemical substances and reactions. Despite the importance of organic chemistry to the health sciences, little de facto progress has been made in adapting the pedagogy of organic chemistry to the population of students taking it today. Students often rely on rote learning and
© 2016 American Chemical Society Sörensen; Online Course Development and the Effect on the On-Campus Classroom ACS Symposium Series; American Chemical Society: Washington, DC, 2016.
Downloaded by UNIV ILLINOIS URBANA-CHAMPAIGN on November 10, 2016 | http://pubs.acs.org Publication Date (Web): August 16, 2016 | doi: 10.1021/bk-2016-1217.ch003
recognition of the surface features of organic structures and reactions to get by in their organic chemistry courses (1). Traditionally, organic chemistry is taught using an approach that focuses on the structure and reactivity of various functional groups. Textbooks that use this approach follow predictable patterns in organization. After a brief introduction to organic structure, nomenclature, and reactivity, alkyl halides are covered in the context of substitution and elimination reactions and alkenes and alkynes are used to illustrate addition reactions. Alcohols, amines, and aromatic compounds may be associated with dedicated units, and the chemistry of carbonyl compounds forms a large section of the second-semester sophomore organic chemistry course. The effectivess of this approach, which we will refer to as the functional-group approach, has been called into question in recent years (2). For a particular functional group, a large amount of information is typically presented: the structure of the group, physical properties of archetypal compounds, reactions of the group, and mechanisms. Although the expertise of organic chemistry instructors often allows them to draw this information from memory, whether the functional-group approach is an effective way for students to learn organic chemistry for the first time remains unclear. This is particularly true for students of the health sciences, who may rarely come across the names of functional groups in their careers and for whom a mental schema based on functional-group organization is unlikely to be useful. Emerging alternatives eschew the functional-group approach in favor of approaches that respect the organization and development of knowledge in novices and experts. An emerging respect for constructivism among organic chemistry educators has been an important driver behind these new pedagogies (3). Studies of the ways in which experts organize organic chemistry knowledge in their minds point to systematic deficiencies in the functional-group approach. For example, expert chemists typically associate physical properties such as melting and boiling point with static structural features such as dipole moments, while linking reaction mechanisms with the dynamic behavior of electrons. Thus, teaching physical properties alongside reaction mechanisms in a functional-group unit could be misleading to the student, since this approach conflates static and dynamic aspects of structure (4). One appealing approach to teaching organic chemistry uses an organization focused on the mechanistic steps that related reactions have in common. Flynn and Ogilvie have recently described such an approach (5). Foundational principles of organic reaction mechanisms are taught before any specific reactions are introduced, so that students are primed to recognize deep mechanistic similarities between reactions that appear quite different at a superficial level. Students immediately begin developing facility with the curved-arrow formalism and the principles that govern reasonable reaction mechanisms, which are applied extremely often over the course of the remaining curriculum. Ideally, this approach encourages students to apply general principles repeatedly such that they will transfer their knowledge or skills in one context of organic chemistry to another (say, from reactions of ketones and aldehydes to reactions of imines). Teaching and learning for transfer are becoming increasingly important as the demands of the modern workplace shift from rote tasks and well-worn problems 22 Sörensen; Online Course Development and the Effect on the On-Campus Classroom ACS Symposium Series; American Chemical Society: Washington, DC, 2016.
Downloaded by UNIV ILLINOIS URBANA-CHAMPAIGN on November 10, 2016 | http://pubs.acs.org Publication Date (Web): August 16, 2016 | doi: 10.1021/bk-2016-1217.ch003
to complex cognition in new scenarios (6). Evidence suggests that highly general transfer across disciplinary lines is very difficult for both novices and experts in a domain. However, the transfer of general principles within a domain has been demonstrated with appropriate instruction. In the context of organic chemistry, teaching for transfer looks very different from traditional approaches. The exposition of general principles governing structure and reactivity at the outset of instruction—as described by Flynn and co-workers—is necessary to promote repeated application and transfer of these principles. Historically, the distillation of general ideas from specific examples has been left up to the student; a wealth of empirical evidence indicates that transfer will not occur under such myopic instructional conditions. An appreciation of the structure of the learning mind as it is presently understood by cognitive psychologists can help organic chemistry educators better design their courses for transfer. In particular, cognitive psychologists draw a distinction between working memory and long-term memory. Working memory is a limited store of information in which the essential elements of a problem or task (as well as cognitive interpretations of sensory input) are held. Long-term memory, on the other hand, is a much larger “vault” of information that remains available for use over long periods of time and in many different contexts. In general, concepts from long-term memory are readily transferred to other contexts, but not concepts from short-term memory. However, the transition of general principles from working to long-term memory requires a great deal of repeated practice and feedback, which cannot be achieved without early exposition of such principles (7). Part of our aim in creating a massive open online course (MOOC) for organic chemistry was to showcase a generalized approach that focuses on the early exposition of general principles and concepts. In the next section, we consider the typical student in our MOOC and argue that a generalized approach designed to promote transfer may be even more valuable in the context of a MOOC than in a traditional on-campus context.
The Student Population of Our Organic Chemistry MOOC Studies of the student populations of MOOCs across many disciplines have revealed some surprising demographic results. In general, MOOC participants are older, international, and current members of the workforce in some capacity. MOOCs with a clear disciplinary focus often attract students with some previous exposure to the domain, sometimes in the distant past (8). The educational level of MOOC participants tends to be higher than that of the general population, leading some experts to suggest that the stated goal of many MOOC platforms to “bring education to the masses” will be difficult to achieve in practice. Studies of the digital activity of MOOC participants have revealed striking contrasts between the behaviors of traditional students and MOOC participants. Perhaps unsurprisingly, MOOC participants generally act like disengaged consumers of educational content, while traditional students are motivated to engage with course materials and instructors to achieve their desired outcomes on 23 Sörensen; Online Course Development and the Effect on the On-Campus Classroom ACS Symposium Series; American Chemical Society: Washington, DC, 2016.
Downloaded by UNIV ILLINOIS URBANA-CHAMPAIGN on November 10, 2016 | http://pubs.acs.org Publication Date (Web): August 16, 2016 | doi: 10.1021/bk-2016-1217.ch003
assessments. One study reported an average completion rate of only 6.5% for a set of 279 MOOCs offered on Coursera, Udacity, and EdX and found that while enrollment numbers are positively correlated with course length, completion rate is negatively correlated with course length (9). Although longer courses may suggest greater depth of learning and thereby attract more potential participants, completion seems to be reserved for a core group of dedicated participants whose size is independent of course length. Research to date has illuminated many of the demographic and behavioral characteristics of MOOC students, but MOOCs have by and large not responded to these studies with approaches designed to promote engagement. This is particularly true for MOOCs in the physical sciences and mathematics, which have been slower to harness the social dimension of learning (via peer grading and discussion forums, for example) than courses in the humanities and social sciences (10). Studies of the demographics of our organic chemistry MOOC are generally consistent with prior research in similar MOOCs. Most survey respondents indicated that organic chemistry was relevant in some way to their academic fields of study, with a plurality responding that relevance to their field was an “extremely important” part of their decision to enroll. A similar fraction of respondents indicated that the course teaches skills that will help in future careers, and that this was a factor in their decision to enroll. On the other hand, simply learning more about organic chemistry was an extremely important factor to nearly 50% of survey respondents, and the perception that learning organic chemistry would be fun was also a factor of considerable importance. As has been documented elsewhere, we observed that enrollees derived little motivation from the particular university and professors offering the course. Furthermore, earning a tangible credential was important to a very small fraction of enrollees. Demographically, the typical MOOC student is very different from the traditional college student. Survey respondents in our course were for the most part between the ages of 18 and 40, with some prior experience in organic chemistry. Very few students were younger than 18 years old. Our MOOC student body contained a roughly 3:1 ratio of international to American students, 60% of which were male and 40% of which were female. These survey results paint a picture in stark contrast to the traditional student of sophomore organic chemistry at an American university. Most traditional on-campus students entering sophomore organic chemistry have recently completed a course in general chemistry but have had no prior exposure to organic chemistry. Furthermore, most on-campus students are undecided with respect to their future careers, while MOOC students appear to be generally established in a career (at least on the basis of age data). External motivators such as grades evidently represent an important driving force for traditional students, while the participation of MOOC students is based almost exclusively on self-regulation and internal motivators. Our profile of the typical MOOC student suggested that a fundamental shift was needed in the way we taught organic chemistry. In fact, such a shift had already been occurring in our on-campus course, which had recently undergone significant reorganization and changes in topics covered. We concluded that a student body composed primarily of aspiring physicians, nurses, pharmacists, 24 Sörensen; Online Course Development and the Effect on the On-Campus Classroom ACS Symposium Series; American Chemical Society: Washington, DC, 2016.
Downloaded by UNIV ILLINOIS URBANA-CHAMPAIGN on November 10, 2016 | http://pubs.acs.org Publication Date (Web): August 16, 2016 | doi: 10.1021/bk-2016-1217.ch003
and dentists would be better served by an organic chemistry course sequence that exposes patterns in structure and reactivity and focuses ultimately on bio-organic chemistry, rather than more traditional organic reactions. From an educational psychology perspective, the highly contextualized functional-group approach was unlikely to provide any assets in long-term memory to a body of students whose working memories are likely already taxed by the daily demands of the workplace. Because of its focus on the exposition of general patterns in the structure and reactivity of organic compounds, our method for teaching organic chemistry represents a generalized approach. The overarching goal of the approach is to supply the student with fundamental problem-solving skills by laying out organic chemistry as a framework of concepts that build on one another with essential patterns as their common basis. The following sections describe the details of the generalized approach and explain how technology was used both on campus and in the MOOC environment to facilitate learning and explore contexts of interest to students, particularly in the second-semester course.
Generalized Organic Chemistry: Principles and Organization At the University of Illinois at Urbana-Champaign, students with a pre-health focus are required to take CHEM 232 and 332, Elementary Organic Chemistry I and Elementary Organic Chemistry II. The first course, CHEM 232, begins with an introduction to organic structure and the language of organic chemistry and proceeds in an accelerated but fairly traditional manner through substitution, elimination, and addition reactions. The course culminates in a discussion of aromatic structure and reactivity. The second course, CHEM 332, begins by revisiting the fundamentals of organic structure and reactivity, exposing the patterns in structures and reaction mechanisms that form the basis for heuristic problem solving. Roughly the second quarter of the course addresses the fundamentals of organic reaction mechanisms both in the “bookkeeping” context of the curved-arrow formalism and in the deeper context of physical organic chemistry. The second half of CHEM 332 involves the application of these fundamental ideas to bio-organic contexts: proteins and enzymes, carbohydrates, nucleic acids, and the machinery of protein biosynthesis. Both courses were taught on-campus in a flipped (also called “blended” or “hybrid”) format prior to the introduction of the corresponding MOOC courses (11). Before class, students watched a series of videos that served as an exposition of important empirical results, concepts, and conventions for the upcoming class session. In essence, the video series replaced a traditional lecture. Each class session was associated with a set of problems—“Problems of the Day” or “POTDs”—that students would work through with the help of the course instructor. The problems were presented through an online homework system for organic chemistry called ACE Organic (12), such that students received feedback both from the instructor in class and from the online homework platform both inside and outside of class. Follow-up quizzes in ACE Organic, due twice weekly 25 Sörensen; Online Course Development and the Effect on the On-Campus Classroom ACS Symposium Series; American Chemical Society: Washington, DC, 2016.
Downloaded by UNIV ILLINOIS URBANA-CHAMPAIGN on November 10, 2016 | http://pubs.acs.org Publication Date (Web): August 16, 2016 | doi: 10.1021/bk-2016-1217.ch003
on Tuesdays and Thursdays, served as formative assessments and encouraged students to continually practice and deepen their knowledge. We had two primary motivations for teaching organic chemistry in this format. On the one hand, the flipped approach is associated with general pedagogical benefits (13). Instructors can shift their effort from presenting lecture material, which is largely the same from semester to semester, to training students to become better problem solvers, which involves rich student-faculty interaction and dynamic approaches. Students can pause a video or alter its speed to ensure that they have fully grasped a concept before moving forward. On the other hand, the flipped approach provides chemical educators specifically with unique opportunities to integrate computational chemistry into their courses. We embraced a very broad definition of computational chemistry and began to explore how access to quantum chemistry software and databases of chemical information are likely to change how the workforce of the future practices and applies organic chemistry. Computers are best known for enabling organic chemists to perform quantum-chemical calculations on large organic molecular systems in reasonable amounts of time. Open-source quantum chemistry packages and web-based tools for computational chemistry such as WebMO (14) have facilitated the teaching of quantum chemistry in educational settings that have historically encountered barriers to implementation. Examples from the literature suggest an increasing interest in incorporating computational quantum chemistry into laboratory curricula in general (15) and physical chemistry (16). Organic chemists make regular use of molecular orbital theories in certain contexts, even in educational settings. Frontier molecular orbital (FMO) theory has been particularly important in rationalizing outcomes in SN2 reactions and pericyclic processes (17). Hückel molecular orbital theory is regularly used to describe the most important orbitals of conjugated π systems. Natural bond orbital (NBO) analysis may be applied to describe the electronic structures of molecules in localized terms without sacrificing rigor, and to make connections between structure and reactivity (18). These theories are typically applied by organic chemistry instructors and students without the help of computers, via heuristics associated with patterns in electronic structure. For example, the general rule that nonbonding (n) NBOs are higher in energy than π NBOs can be applied to predict that the nitrogen of an allylic amine is more basic than the alkenic carbons. The contextualized nature of organic applications of molecular orbital theories may suggest to students a lack of importance or real-world applications. Computational chemistry software, on the other hand, helps students see the broad utility of orbital theories in the context of rigorously calculated results. Such software expands the purview of these theories: all of a sudden, molecular systems that were inaccessible to organic chemists restricted to paper-and-pencil “calculations” are now amenable to study. One example from our own course concerns heteroatomic π systems, which can be studied by simple Hückel molecular orbital theory albeit with some complications not relevant to hydrocarbon systems. To students familiar with the Hückel treatment of ethylene, butadiene, et cetera, the conceptual leap to heteroatomic π systems is not difficult; however, the shift makes paper-and-pencil 26 Sörensen; Online Course Development and the Effect on the On-Campus Classroom ACS Symposium Series; American Chemical Society: Washington, DC, 2016.
Downloaded by UNIV ILLINOIS URBANA-CHAMPAIGN on November 10, 2016 | http://pubs.acs.org Publication Date (Web): August 16, 2016 | doi: 10.1021/bk-2016-1217.ch003
and heuristic computations difficult (asymmetry and polar bonds are introduced). To circumvent these issues, we teach students to build molecules in and interpret results from a simple Hückel MO theory calculator (SHMO) (19). Using SHMO, students are able to calculate Hückel orbitals for aromatic heterocycles with biochemical relevance, such as purine and pyrimidine. Teaching students to use SHMO has expanded the set of contexts in which we can teach π molecular orbital theory. Although the introduction of this tool is not without problems—issues of interpretation of results and inappropriate application must be considered—we have found that the payoff in student engagement for our on-campus course has been well worth these issues. Ultimately, we have observed a marked increase in the accessibility of orbital theories to students who have taken up the use of quantum chemistry software. Computers can also be applied in organic chemistry courses as unprecedented stores of data and information. At the broadest level, resources such as ChemSpider (20) and PubChem (21) serve as easily accessible databases of general chemical information. More specific databases can provide information relevant to more specific contexts. For example, the Protein Data Bank contains crystallographic data on proteins and relevant biochemical information, such as secondary and tertiary structure (22). The FooDB database is a comprehensive collection of chemicals in food and can be searched by food type as well as physical and chemical properties. Similarly, the SuperScent database includes a collection of 1200 scent and flavor molecules, categorized and searchable according to their scents (23). All of these databases contain chemical information in a strategically organized form, allowing it to be accessed and used easily in a particular context. Computational databases can thus serve as a starting point for the contextualized study of organic chemistry, which is limited only by the availability of a database geared toward a particular application. For example, focusing on the Protein Data Bank and related databases, we implemented a semester-long project for students in Organic Chemistry II on an enzyme-catalyzed mechanism. Using resources pulled from these databases and primary literature articles, students prepared an interactive, web-based wiki article in which they described and justified the mechanism. Throughout this process, students were required to integrate fundamental concepts from the course with the information found in computational databases (24). This approach in our on-campus course helped students transfer their fundamental organic chemistry knowledge to a new context. Application of a similar approach in our Organic Chemistry MOOC seemed natural, as we expected MOOC students to be involved in a variety of careers or fields of study, and MOOC participants would be expected to have easy access to open databases. Our Intermediate Organic Chemistry MOOC was designed to deliver fundamental principles of organic structure and reactivity in the first eight-week course, followed by a second eight-week course focusing mostly on applications (particularly biochemical applications). To construct our Organic Chemistry MOOC, we began with a set of video lectures used in our on-campus course to introduce concepts and other content. The videos had been produced gradually over several semesters and were based 27 Sörensen; Online Course Development and the Effect on the On-Campus Classroom ACS Symposium Series; American Chemical Society: Washington, DC, 2016.
Downloaded by UNIV ILLINOIS URBANA-CHAMPAIGN on November 10, 2016 | http://pubs.acs.org Publication Date (Web): August 16, 2016 | doi: 10.1021/bk-2016-1217.ch003
primarily on previously used PowerPoint slides. Each set of videos corresponding to a single class session constituted a lesson, and two to three lessons per week were delivered through a single “Week n” course page. The first section of each weekly course page (Overview) included a few paragraphs describing the content of that week as well as broad learning goals. The next section (Time) listed the approximate length of time required to work through all of the videos and problems for that week. Weekly workloads were estimated to some extent by the amount of time our on-campus students reported working for their organic chemistry courses. In the next section, titled Goals and Objectives, more specific learning objectives culled from the videos for that week were listed. We used these specific skills and concepts to develop new assessments on the Coursera platform for MOOC participants (primarily because use of our online homework system for on-campus students by MOOC participants was impractical). A related section called Key Terminology & Concepts was used to introduce technical terms from the videos that were important for students to learn and use in conversations on the discussion forums. The following Instructional Activities section listed the videos and readings to be watched for that week, while Tips for Success and Giving and Receiving Help provided students with general pointers and resources to improve their learning.
Generalized Organic Chemistry in Action: Building Blocks of Organic Structure This section and the next describe practical examples of the generalized approach taken in our MOOC. In this section, we show how general principles of organic structure and bonding are introduced first, followed by practice and feedback with specific examples in quizzes. Because the “general principles” of organic structure are really structural patterns that appear across different molecules, we refer to this method of teaching organic structure as the building-block approach or building-block formalism. Prior to the preparation of the MOOC, we found that the building-block formalism helped students in our on-campus Organic Chemistry II course develop pattern recognition skills. To a certain extent, this approach leverages the previous exposure that most MOOC participants have had to organic structure and bonding. Rules and processes for drawing Lewis structures are not covered in detail. Fundamental bonding concepts such as molecular orbital theory and hybridization are not addressed until after the introduction of the building blocks because particular structures that share the same general building block are often associated with isolobal orbitals (particularly localized molecular orbitals), similar geometries, and similar trends in reactivity. We reasoned that students armed with the ability to recognize and apply the building-block formalism will be able to apply more fundamental concepts across a wider variety of structures. We distinguish between two types of building blocks: general and particular. The general building blocks represent “template” structures with a placeholder atom X at the center and an arrangement of bonds and lone pairs around the central atom. Either single lines (single bonds or lone pairs) or multiple lines (multiple 28 Sörensen; Online Course Development and the Effect on the On-Campus Classroom ACS Symposium Series; American Chemical Society: Washington, DC, 2016.
Downloaded by UNIV ILLINOIS URBANA-CHAMPAIGN on November 10, 2016 | http://pubs.acs.org Publication Date (Web): August 16, 2016 | doi: 10.1021/bk-2016-1217.ch003
bonds) may emanate from the central X atom. Thus, in addition to accounting for the number of regions of electron density around the central atom as in VSEPR theory, the general building blocks include information about the number and positions of multiple bonds about an atom. They are constrained by the octet and duet rules but represent the most general level of organic structure. The particular building blocks are specific instances of the general building blocks and include three additional features. Firstly, the template X atom is replaced by an atomic symbol. Secondly, lone pairs are positioned at one or more of the regions of electron density that are not multiple bonds (i.e., single lines in the general building block). Finally, formal charge is added if the number of valence electrons formally belonging to the central atom in the building block differs from the number of valence electrons in the neutral atom. The typical electron-counting system for formal charge is used, with the central atom owning half of the electrons in bonds and both electrons in all lone pairs. Naturally, the particular building blocks are linked together via covalent bonds in molecules. The particular building blocks can be systematically produced from the general building blocks by adding atomic symbols, lone pairs, and formal charges. In the videos and reading materials for the first lesson of the MOOC, we do this for main-group elements in the second period (B, C, N, O, and F). Notably, a student or practitioner actually solving a problem relevant to organic chemistry would almost never use this kind of general-to-particular thought process. However, the process running in the opposite direction—from an instance of a particular building block to its associated general—is critical for the recognition of patterns in multiple structures. Figure 1 shows the particular building blocks that are associated with each general building block. Particular building blocks that share the same general building block have a number of important features in common. Ignoring for the moment the complications of resonance delocalization, analogous particular building blocks are isolobal—they have the same hybridization at the central atom and the same kinds of π bonds and localized σ orbitals. They often react in similar ways; for example, an atom bearing a double bond and one or two lone pairs may undergo electrophilic or nucleophilic addition processes (via the π bond) or act as a Lewis or Brønsted base via the lone pair(s). Although changing the type of atom at the center may affect the rates of different types of reactions, the possibility of a particular reaction type is for the most part built into the structural features (lone pairs, π bonds, and polarized σ bonds) of the building block. While the videos and reading materials of this lesson proceed in a general-to-particular direction, the associated quiz requires students to reason in the opposite direction: from specific examples to the general building blocks. Quiz questions are designed very deliberately to test students’ mastery of the building block formalism. For example, in one question structures of a carbocation and an organoborane are presented alongside one another. The student is asked to select, from five choices, the general building block that the cationic carbon and boron atom have in common. In a second example, structures of carbon dioxide and a carbodiimide are shown with the same prompt. These problems are very straightforward to solve provided the student is familiar with the building-block formalism, and students did perform very well on these 29 Sörensen; Online Course Development and the Effect on the On-Campus Classroom ACS Symposium Series; American Chemical Society: Washington, DC, 2016.
Downloaded by UNIV ILLINOIS URBANA-CHAMPAIGN on November 10, 2016 | http://pubs.acs.org Publication Date (Web): August 16, 2016 | doi: 10.1021/bk-2016-1217.ch003
problems. Despite their ease, we view problems of this type as fundamental in activating students’ pattern-recognition apparatus early in the course. Pattern recognition enables and is a prerequisite for reasoning by analogy and transfer, and MOOC participants will likely have to transfer their learning from our course to a variety of different contexts.
Figure 1. The general building blocks (gray, center) and their associated particular building blocks.
Although the building blocks themselves represent structural patterns, the formalism also helps students make connections between structure and reactivity. After introducing the building blocks and how they are connected in organic structures, we turn to a discussion of the dynamics of the structural elements: how lone pairs, π bonds, and even σ bonds can act as electron donors or sources, and how empty atomic orbitals, π* orbitals, and σ* orbitals can act as electron acceptors or sinks. In particular, we introduce the structural elements of resonance and identify general structural patterns that point to the relevance of resonance within a structure. By using the symbology of molecular orbital theory to do this, we introduce the student to a simplified variant of molecular orbital theory that will be relevant throughout the MOOC. This localized molecular orbital framework provides a physically realistic basis for rules governing electron flow in both resonance and reaction mechanisms. 30 Sörensen; Online Course Development and the Effect on the On-Campus Classroom ACS Symposium Series; American Chemical Society: Washington, DC, 2016.
Downloaded by UNIV ILLINOIS URBANA-CHAMPAIGN on November 10, 2016 | http://pubs.acs.org Publication Date (Web): August 16, 2016 | doi: 10.1021/bk-2016-1217.ch003
Generalized Organic Chemistry in Action: The Curved-Arrow Formalism and Reaction Mechanisms The previous section demonstrated the utility of teaching from general principles using structure as a context; this section highlights an application in the context of reactivity and mechanism. Students of organic chemistry often have trouble drawing reasonable organic reaction mechanisms. Unfamiliar with the physical rules governing electron flow between or within molecules, students tend to use the curved-arrow formalism in an arbitrary manner, pushing electrons until they “get to the product.” Beyond a vague sense that a mechanism “feels right,” novices are often unable to argue for the reasonableness of a particular mechanism. It is unsurprising therefore that students also experience difficulties using organic reaction mechanisms to make predictions, propose experiments, or optimize reactions (even though experts rely on mechanisms for exactly these kinds of inferences). Treating mechanisms as isolated “works of art” makes them difficult to learn because steps within a mechanism are not viewed as useful outside of the context of a single reaction. Evidently, on-campus students who view mechanisms in this light are nonetheless motivated to learn them, at least to a point where they can regurgitate them on exams. There are good reasons to believe that MOOC students will not be so driven: they typically have much less time to devote to studies, are older and mistrustful of esoteric or highly specific knowledge, and may already have a bad taste in their mouths from previous experience with organic chemistry. As in the study of organic structure, in learning about organic reactivity the MOOC student is best served by a generalized framework for reaction mechanisms, which includes the possibilities available for elementary mechanistic steps and guidelines for choosing among them in a rational way. Such a framework enumerates elementary steps in a systematic way and treats them as interchangeable parts that can be bolted together to produce a reasonable mechanism. In the history of organic chemistry education, there is a rich tradition of such mechanistic frameworks being used to teach undergraduates. Starting from the dynamic building blocks and resonance theory as previously described, we introduce localized molecular orbitals as a framework for understanding general patterns in polar (two-electron) organic mechanisms. This framework is based on natural bond orbital (NBO) theory and frontier molecular orbital (FMO) theory, but uses these theories as means to the end of predicting organic reactivity. NBO theory has been promoted in recent years by Landis, Weinhold, and others (17). Natural bond orbitals are localized and map well onto bonds and lone pairs in Lewis structures, at least in molecules for which resonance is irrelevant or minimally important. Single bonds are associated with σ and σ* NBOs, multiple bonds are associated with π and π* NBOs, lone pairs are associated with filled nonbonding n orbitals (typically hybrids), and electron-deficient carbocations are associated with empty nonbonding a orbitals. The empirical implications of NBO theory are equivalent to those of canonical MO theory, but natural bond orbitals are much easier to interpret than canonical orbitals due to the localized nature of the former and their grounding in Lewis structures. The energy of an occupied NBO, for example, corresponds in 31 Sörensen; Online Course Development and the Effect on the On-Campus Classroom ACS Symposium Series; American Chemical Society: Washington, DC, 2016.
Downloaded by UNIV ILLINOIS URBANA-CHAMPAIGN on November 10, 2016 | http://pubs.acs.org Publication Date (Web): August 16, 2016 | doi: 10.1021/bk-2016-1217.ch003
a rough way to the reactivity of the atom(s) on which that NBO is located. NBO calculations can thus provide useful (albeit not entirely foolproof) heuristics for predicting organic reactivity from Lewis structures alone. Both NBOs and the curved-arrow formalism of organic reaction mechanisms map gracefully onto the features of Lewis structures. As a consequence, NBO theory appears attractive as a physical basis for the curved-arrow formalism—it can provide insights (such as subtle stereoelectronic effects) that are much more difficult to see using canonical MO theory or the curved-arrow formalism alone. Most important is the idea that every “reasonable” elementary step in a polar reaction mechanism represents one or more interactions between a donor orbital and an acceptor orbital. As shown in Figure 2, multiplying the three classes of localized donor orbitals (σ, π, n) by the three classes of localized acceptor orbitals (a, π*, σ*) gives nine possible elementary electron flows in polar organic reaction mechanisms. These fully general steps, based on a prior framework developed by Lewis (25), represent the core of our localized molecular orbital framework. Principles of frontier molecular orbital theory are necessary to complete the physical basis of the framework. FMO theory was developed by Fukui and distilled into an educational tool for teaching organic reactivity by Lewis. Because the nine-step framework just described relies on reactant NBOs, it is critical that reactant orbitals correlate well with reaction outcomes. This is exactly what Fukui’s FMO theory shows: from the highest-energy occupied molecular orbital (HOMO) and lowest-energy unoccupied molecular orbital (LUMO) of nucleophilic and electrophilic reactants respectively, one can often predict the progress of a reaction. When interpreting NBOs, therefore, we often focus on the highest-energy filled NBO and the lowest-energy unfilled NBO in the reactants.
Figure 2. Matrix of the nine elementary steps according to the localized molecular orbital theory framework. Just as the building-block formalism helps students make connections between superficially unrelated structures, the localized MO framework helps them make connections between elementary steps in different reaction mechanisms that are manifestations of the same localized orbital interaction. In Week 5 of the MOOC, we present more specific versions of the general steps that we call the elementary steps, which differ in both the donor and acceptor orbitals involved and the types of bonds formed or broken. For example, a distinction is made between proton transfer and SN2 (both of which involve n-to-σ* orbital interactions) and dissociation of a leaving group is defined as an internal σ-to-a interaction involving no bond formation. 32 Sörensen; Online Course Development and the Effect on the On-Campus Classroom ACS Symposium Series; American Chemical Society: Washington, DC, 2016.
Downloaded by UNIV ILLINOIS URBANA-CHAMPAIGN on November 10, 2016 | http://pubs.acs.org Publication Date (Web): August 16, 2016 | doi: 10.1021/bk-2016-1217.ch003
Figure 3 enumerates the elementary steps and provides prototypical examples for each. The symbol assigned to each step serves three purposes. First, the symbol classifies the step as proton transfer (pt), addition (A), elimination (E), substitution (S), dissociation (D), or rearrangement (R). Second, the subscript clarifies the electrophilic (E) or nucleophilic (N) nature of the step. Third, additional numbers or letters address the positioning of the electron source and sink (as in the 1,2R and Eβ steps) or the molecularity of the step (as in SN2 and E2). Although these symbols have historical significance, we also make heavy use of the donor-to-acceptor nomenclature listed in Figure 2 above.
Figure 3. The elementary steps used as the fundamental elements of polar organic reaction mechanisms. These elementary steps then become the basic building blocks of organic reaction mechanisms, which we showcase in a survey of different reaction types in the same week. In the on-campus course, the elementary steps are applied extensively to the bio-organic context of enzymatic reaction mechanisms. A planned Intermediate Organic Chemistry II MOOC will incorporate this aspect of the on-campus course in the future.
Reach of the Course and Student Engagement Was our MOOC successful? To a large degree the answer depends on our definition of “success” and the types of data collected. Assessing the impact of a MOOC on student learning is typically very difficult, but engagement and reach statistics based on web analytics can provide qualified measures of success. Although the amount of data we have is limited, analytics suggest that our Intermediate Organic Chemistry MOOC reached an unprecedented number of students worldwide (approximately 26,000 learners), including a substantial portion (30%) from developing economies. On average learners viewed approximately 16 videos (411,000 total views), while rates of submitting exercises 33 Sörensen; Online Course Development and the Effect on the On-Campus Classroom ACS Symposium Series; American Chemical Society: Washington, DC, 2016.
were much lower. While the impact of the course on student learning remains uncertain, it is clear that the course was successful in at least disseminating a new approach to teaching organic chemistry to a broad base of learners.
Downloaded by UNIV ILLINOIS URBANA-CHAMPAIGN on November 10, 2016 | http://pubs.acs.org Publication Date (Web): August 16, 2016 | doi: 10.1021/bk-2016-1217.ch003
Conclusion In conclusion, we have developed an Intermediate Organic Chemistry MOOC based on our on-campus flipped course, which makes use of online resources and tools as well as a set of generalized conceptual frameworks. As educators have reexamined how organic chemistry is organized, presented, and taught, a number of innovative approaches have emerged. More focus is being placed on learning the general rules governing organic structure and reactivity prior to the introduction of specific examples. Although the general principles themselves are not new, the hope is that “slowing down” the teaching of organic chemistry and being explicit about connections between related structures and reactions will improve students’ performance and increase the transferability of their knowledge. In our courses, these ideas take the form of a set of frameworks for understanding organic structure and reactivity. This chapter has outlined our two most important frameworks: the building-block formalism and the elementary steps of organic chemistry. The former is a set of general atomic structures found in organic Lewis structures, which aids students in identifying isomorphic features within different organic molecules. The latter is a set of electron flows grounded physically in NBO and FMO theory that constitute the allowed elementary steps of organic reaction mechanisms. When we were first approached by university administrators about developing a MOOC for organic chemistry, our on-campus course already included a generalized approach and heavy use of online tools for problem solving and chemical information searching. Ultimately, taking a similar approach in the MOOC was advantageous due to the prevailing demographics and prior experience of the MOOC audience. Our MOOC students generally were older, had some prior experience with organic chemistry, and were already working or studying in a field to which organic chemistry was applicable. Student feedback has indicated that learning via a generalized approach, MOOC participants could readily transfer their knowledge to their own careers or studies. We have found that integrating computational tools into a MOOC (beyond the capabilities of the MOOC platform itself) is important for several reasons. Most generally, using software or online resources in an online course justifies its existence—a MOOC can be built entirely around a set of software in a way that is more difficult to achieve in an on-campus course. In addition, using large online stores of information in a MOOC allows students to learn in context from examples that are relevant or interesting to them. Students are able to interact with chemical models (such as protein crystal structures) in entirely new ways on a computer. Finally, computational tools for problem solving provide a mechanism for students to practice and receive feedback efficiently.
34 Sörensen; Online Course Development and the Effect on the On-Campus Classroom ACS Symposium Series; American Chemical Society: Washington, DC, 2016.
References 1.
2.
Downloaded by UNIV ILLINOIS URBANA-CHAMPAIGN on November 10, 2016 | http://pubs.acs.org Publication Date (Web): August 16, 2016 | doi: 10.1021/bk-2016-1217.ch003
3.
4.
5.
6. 7. 8.
9.
10.
11. 12. 13. 14. 15.
16.
Grove, N. P.; Lowery Bretz, S. A continuum of learning: from rote memorization to meaningful learning in organic chemistry. Chem. Educ. Res. Pract. 2012, 13, 201. Advances in Teaching Organic Chemistry; Duffy-Matzner, J. L., Pacheco, K. A. O., Eds.; ACS Symposium Series 1108; American Chemical Society: Washington, DC, 2012. Tien, L. T.; Roth, V.; Kampmeier, J. A. Implementation of a peer-led team learning instructional approach in an undergraduate organic chemistry course. J. Res. Sci. Teach. 2002, 39, 606–632. Cooper, M. M.; Underwood, S. M.; Hilley, C. Z. Development and validation of the implicit information from Lewis structures instrument (IILSI): Do students connect structures with properties? Chem. Educ. Res. Pract. 2012, 13, 195. Flynn, A. B.; Ogilvie, W. W. Mechanisms before reactions: A mechanistic approach to the organic chemistry curriculum based on patterns of electron flow. J. Chem. Educ. 2015, 92, 803–810. Education for Life and Work; Pelligrino, J. W., Hilton, M. L., Eds.; National Academies Press: Washington, DC, 2012. Ambrose, S. A.; Bridges, M. W.; DiPietro, M.; Lovett, M. C.; Norman, M. K.; Mayer, R. E. How Learning Works; Jossey-Bass: San Francisco, 2010. Christensen, G.; Steinmetz, A.; Alcorn, B.; Bennett, A.; Woods, D.; Emanuel, E. J. The MOOC Phenomenon: Who Takes Massive Open Online Courses and Why? SSRN Journal, 2013. http://dx.doi.org/10.2139/ ssrn.2350964 (accessed February 2016). Jordan, K. Initial trends in enrolment and completion of massive open online courses. International Review of Research in Open and Distributed Learning 2014, 15, 133–159. Kulkarni, C.; Wei, K. P.; Le, H.; Chia, D.; Papadopoulos, K.; Cheng, J.; Koller, D.; Klemmer, S. R. In Design Thinking Research; Understanding Innovation; Springer International Publishing: Cham, 2015; pp 131–168. Moore, J. S.; Janowicz, P. A. Chemistry goes global in the virtual world. Nature Chem. 2009, 1, 2–4. Chamala, R. R.; Ciochina, R.; Grossman, R. B. J. Chem. Educ. 2006, 83, 164. Seery, M. Chem. Educ. Res. Pract. 2015, 16, 758–768. Schmidt, J. R.; Polik, W. F. WebMO. http://www.webmo.net (accessed October 29, 2015). Ruddick, K. R.; Parrill, A. L.; Petersen, R. L. Introductory molecular orbital theory: An honors general chemistry computational lab as implemented using three-dimensional modeling software. J. Chem. Educ. 2012, 89, 1358–1363. Johnson, L. E.; Engel, T. Integrating computational chemistry into the physical chemistry curriculum. J. Chem. Educ. 2011, 88, 569–573.
35 Sörensen; Online Course Development and the Effect on the On-Campus Classroom ACS Symposium Series; American Chemical Society: Washington, DC, 2016.
Downloaded by UNIV ILLINOIS URBANA-CHAMPAIGN on November 10, 2016 | http://pubs.acs.org Publication Date (Web): August 16, 2016 | doi: 10.1021/bk-2016-1217.ch003
17. Fukui, K.; Yonezawa, T.; Nagata, C.; Shingu, H. Molecular orbital theory of orientation in aromatic, heteroaromatic, and other conjugated molecules. J. Chem. Phys. 1954, 22, 1433–1442. 18. Weinhold, F.; Landis, C. R. Natural bond orbitals and extensions of localized bonding concepts. Chem. Educ. Res. Pract. 2001, 2, 91. 19. Rauk, A.; Cannings, R. Simple Huckel Molecular Orbital Theory Calculator. http://www.ucalgary.ca/rauk/shmo (accessed October 29, 2015). 20. Pence, H. E.; Williams, A. ChemSpider: An Online Chemical Information Resource. J. Chem. Educ. 2010, 87, 1123–1124. 21. Wang, Y.; Xiao, J.; Suzek, T. O.; Zhang, J.; Wang, J.; Bryant, S. H. PubChem: A public information system for analyzing bioactivities of small molecules. Nuc. Acids Res. 2009, 37 (Web Server issue), W623–W633. 22. Berman, H. M. The Protein Data Bank. Nuc. Acids Res. 2000, 28, 235–242. 23. Dunkel, M.; Schmidt, U.; Struck, S.; Berger, L.; Gruening, B.; Hossbach, J.; Jaeger, I. S.; Effmert, U.; Piechulla, B.; Eriksson, R.; Knudsen, J.; Preissner, R. SuperScent—A database of flavors and scents. Nuc. Acids Res. 2009, 37, D291–D294. 24. Evans, M. J.; Moore, J. S. A collaborative, wiki-based organic chemistry project incorporating free chemistry software on the web. J. Chem. Educ. 2011, 88, 764–768. 25. Lewis, D. E. Organizing organic reactions: The importance of antibonding orbitals. J. Chem. Educ. 1999, 76, 1718.
36 Sörensen; Online Course Development and the Effect on the On-Campus Classroom ACS Symposium Series; American Chemical Society: Washington, DC, 2016.
