Teaching advanced science concepts through Freshman Research Immersion

The first semester of the FRI programme is the same for all students regardless of research stream. Students from all streams are combined into a single large lecture hall, where more general topics are initially introduced. This includes topics such as laboratory safety, ethics in research, and effective communication of research results. After introducing a topic, students are then separated into research streams for more in-depth discussions and interactive activities. Volunteer undergraduate peer mentors are utilised during this semester to provide guidance and support to the new students. These peer mentors help to provide the students with more direct, one-on-one interactions than would typically be possible in a large lecture course. Peer mentorship also provides benefits to the peer mentors themselves, helping to develop their interpersonal communication skills and also to improve their understanding of the topics introduced throughout the programme (Kopp 2000, Close et al 2016).

During the first semester, the students are tasked with working in teams of 4–5 to create and present a scientific poster reporting on the current state of research for a smart energy related topic. Students must first generate specific research questions and formulate hypotheses relating to their chosen topic, requiring the students to find, compile, and thoroughly read the current scientific literature. The students then generate a data set by compiling relevant data from multiple literature sources, and use the compiled data set to either support or refute their original hypotheses. For example, one group of students created a poster comparing the pros and cons of different types of electric/hybrid vehicle batteries currently being researched (e.g., lithium ion, aluminum-air), requiring them to compile and compare reported battery costs, capacities, and cyclabilities from many different sources. Students present and defend their team poster at a public poster session attended by their peers, university faculty, and administrators.

The second semester of the FRI smart energy stream has been designed to rapidly introduce students to advanced condensed matter physics and chemistry concepts in a logical progression; reinforcing their learning by simultaneously introducing hands-on experimental techniques and on-campus research that demonstrates these concepts. New topics are introduced each week during a short, informal lecture period incorporating group discussions and visual demonstrations. The majority of class time is then spent in a dedicated laboratory space learning to use laboratory equipment and performing hands-on activities, with students typically working in groups of 4–5. Peer mentors are also utilised during this semester, and are essential for supervising large groups of undergraduate students in the laboratory space.

The dedicated FRI smart energy laboratory space has been set up specifically for the needs of the stream and contains several useful but costly pieces of equipment, including a spin coater ($2000), probe station ($16 000), Ultraviolet–visible spectroscopy (UV–vis) and ATR-IR spectrometers ($14 000 and $20 000), and Hall measurement system ($33 000). While the initial startup costs for this FRI stream may seem high, the dedicated laboratory space is necessary to allow the students to perform hands-on, authentic smart energy research activities. In addition, a value-added benefit of the laboratory space is the possibility for non-FRI students and faculty to make use of the equipment outside of the designated class time. Unlike similarly priced teaching laboratory equipment (TeachSpin, Inc. 2016), the pieces of equipment purchased for the FRI laboratory are all fairly common in real physics and chemistry research laboratories. In this way, the dedicated laboratory space can be an asset to the entire university.

The basic schedule of research activities and topics is given in table 1. The order of activities was chosen not only to introduce the corresponding topics in a logical progression, but also to provide the students with as much hands-on, in-laboratory experience as possible. The first experimental techniques introduced are also those that can be performed in our dedicated FRI laboratory space by the students with minimal supervision. This allows the students to master these techniques by the end of the semester through continual repetition and allows for the collection of large amounts of data, facilitating discussions of experimental error and statistics. Later techniques, such as x-ray photoelectron spectroscopy, are not only more complicated and difficult to perform, but also require the use of expensive, shared laboratory equipment located in other buildings on campus. As such, these later techniques have been introduced through interactive demonstrations rather than hands-on activities.

By the end of the second semester, students are expected to present a proposal for a semester-long research project that complements and parallels on-campus faculty research. Each proposal must contain a coherent summary of the current state of the literature pertaining to the project, a legitimate and well-thought-out research goal, a detailed and realistic budget, and a schedule of activities to reach said goal. Students are expected to develop these proposals and perform any necessary preliminary experiments during the last 5 weeks of the semester. While the proposals must be developed and written by the students, the RE and research faculty act similarly to PhD Thesis advisors throughout the process, helping to generate ideas and guide the development.

The third and final semester of the FRI smart energy stream is far less rigid in structure than the previous semesters. At this point in the programme, the students are expected to work independently in the laboratory space with minimal guidance. The goal of this semester is to increase students' independence and self-motivation in the lab. Students must work together to undertake the research projects proposed at the end of the previous semester and are expected to adhere to their proposed research schedule and budget. However, students must also be able to adapt to changing circumstances and overcome the inevitable obstacles and setbacks. This involves gathering, managing, and analysing experimental data throughout the semester to inform the ongoing direction of their research.

During this semester, lectures act as 'group meetings' where students report on their weekly research progress and solicit advice from their peers and mentors. Students must present their current research results to the class throughout the semester and continually improve upon their presentations following peer and instructor feedback. By the end of the semester, students are expected to write up a formal research report and present their final research results to students and faculty. These research reports must be written in the style of a scientific manuscript, complete with an abstract, introduction, relevant background information, description of research methods and techniques, results, discussion, and conclusion. The report content must be adequately supported by experimental results and corroborated by relevant supporting literature. Students complete this final semester by presenting their research findings at a public poster session.

At the beginning of the second semester of FRI, smart energy students are taught how to fabricate copper iodide (CuI) and poly(3,4-ethylenedioxythiophene) (PEDOT), shown in figures 1(a) and (b). Both materials are transparent p-type semiconductors with numerous applications in smart energy related devices, such as solar cells and super-capacitors (Elschner et al 2010, Madl et al 2011, Grundmann et al 2013, Schein 2013). Importantly, both materials can be created through simple, low-temperature fabrication processes that the students can easily perform themselves. This enables the students to quickly master the processes and perform repeat syntheses throughout the semester with minimal supervision. Other model materials that were initially considered for this course (e.g., layered-oxide battery cathode materials and tin oxide semiconductor thin films) took too long to synthesise, used costly deposition equipment, and required far too much instructor involvement. Simple, quick, and hands-on laboratory work was found to be critical for encouraging student engagement in the research.

Zoom In Zoom Out Reset image size

Shown in figure 1(c), copper can be transformed into CuI at low temperatures using a simple laboratory setup. The necessary reaction chamber can be created using only two glass laboratory dishes of different sizes, a hot plate, and some double-sided tape. A similar setup can be used to create PEDOT from an oxidant solution and 3,4-ethylenedioxythiophene (EDOT). As chemical vapours are created during both of these processes, they should be performed within a fume hood for safety.

The main differences between the two synthesis processes are the required precursor materials (i.e., copper films and iodine or oxidant films and EDOT monomer). While fabrication of the precursor films is not particularly difficult or time consuming, it requires additional equipment which many educators might not have immediately available in their teaching laboratories. For example, while our students were able to fabricate the oxidant films in-lab, the copper films required multi-user deposition equipment residing in another building.

Metallic copper precursor films of a few hundred angstroms in thickness were used for the CuI synthesis. These films were deposited using electron beam evaporation in the Nanofabrication Laboratory at Binghamton University. Unfortunately, the freshmen students were not permitted to use this deposition equipment on their own. A graduate teaching assistant deposited the copper films throughout the semester, while the students could only observe the process. However, there are other metal deposition methods, such as electroplating, that could allow for far more hands-on student participation. In addition, these alternate methods could potentially be performed by the students in the FRI laboratory space. These possibilities are currently being investigated.

The oxidant precursor films used for PEDOT synthesis were fabricated using a solution spin coating method. First, the students created an oxidant solution by dissolving iron(III) p-toluenesulfonate hexahydrate (1.0 g) in a solution of pyridine (60 μl) and 1-butanol (10 ml) (Madl et al 2011). A thin coating of this oxidant solution was then applied to substrates via spin coating at a speed of ∼ 750–1500 rpm. The faster the spin coating speed, the thinner and more uniform the resulting PEDOT film. Varying the composition of the oxidant solution will also produce very different film properties. For example, the pyridine is added to suppress unwanted side polymerisation reactions and thus the amount of pyridine greatly affects the resulting thickness and uniformity of the PEDOT film. The students then dried these films on a hotplate for 10 min to evaporate any remaining butanol solvent.

Once precursor films and additional materials are obtained, the final films are synthesised as follows.

• (i)  
  A small amount of iodine/EDOT is placed in the smaller dish.
• (ii)  
  The precursor film is attached to the larger dish using double-sided tape.
• (iii)  
  The larger dish is flipped upside down and placed over the smaller dish.
• (iv)  
  This reaction chamber is placed on a preheated hot plate in a fume hood.
• (v)  
  The iodine/EDOT then vaporises within the reaction chamber.
• (vi)  
  Once the process is complete, the chamber is opened releasing residual vapour.
• (vii)  
  The transformed thin film can then be detached from the larger dish.

Shown in figure 2(a), during the CuI transformation a small amount of iodine is vaporised by heating it to $\sim 120^\circ$ C. The copper film will quickly incorporate the purple iodine vapour, transforming into polycrystalline CuI. This transformation is readily observed when using transparent substrates, as the copper changes from highly reflective to transparent. The entire process should take no more than a few minutes if using copper thin films of ∼100 nm in thickness.

Zoom In Zoom Out Reset image size

Similarly, liquid EDOT monomer is vaporised during the fabrication of PEDOT by heating it to ∼50 °C–100 °C, shown in figure 2(b). This allows the EDOT to react with the oxidant film on the substrate and polymerise, changing the film from yellow to a faint blueish-green in approximately 15–30 min. Higher temperatures speed up the process, creating thicker and often less smooth and uniform films. This process is referred to as vapour phase polymerisation (Madl et al 2011). After polymerisation, the PEDOT films must be soaked in ethanol or another alcohol for several minutes to remove the remaining oxidant, turning the films their final light blue colour.

All of these synthesis parameters are varied throughout the semester by the students to produce films with slightly different properties. These properties are then measured by the students using the aforementioned experimental techniques. The students can then interpret the data and attempt to quantitatively explain how the film properties are affected by the synthesis parameters.

UV–vis is the first experimental technique covered in the second semester of FRI. This technique is an excellent introduction to the research process as it is simple for the students to perform and often simple to interpret the resulting data. Students can qualitatively observe the optical transmittance and reflectance of a sample with their own eyes, and then compare their observations to the more quantitative UV–vis measurements.

Shown in figure 3(a), two example CuI films appear both transparent yet very different in colour; one almost completely colourless and the other a vivid yellow. The colourless sample was created from freshly deposited Cu, while the yellow sample precursor was annealed prior to iodisation. Figure 3(b) shows two example PEDOT films that also appear transparent but different in colour; one a light blue and the other a darker blue-green. The difference between the two PEDOT films is the temperature of the hotplate during the polymerisation reaction. The lighter PEDOT was polymerised at 50 °C and the darker PEDOT at 80 °C.

Zoom In Zoom Out Reset image size

While the visual differences between the films are readily apparent to the naked eye, UV–vis measurements can provide the students with much more quantitative information. The transmittance spectra in figure 3(c) show the bad CuI film as having much lower transmittance in the blue-green wavelength range than the good CuI film, which directly correlates to the colour difference. In addition, the bad CuI film does not possess the two characteristic photon absorptions of crystalline CuI, suggesting it is an entirely different material. The transmittance spectra for the PEDOT films can also be correlated to the colour differences. The bad PEDOT shows much greater absorption of light on the blue end of the visible spectrum and lower transmittance overall, resulting in its darker greenish tint.

Throughout the stream, additional experimental techniques can be used by the students to further probe and explain these differences and characterise the films. Example data for samples of CuI and PEDOT taken using other experimental techniques are shown in figure 4. Some of the additional techniques introduced throughout the stream are useful for probing crystalline systems like CuI, while others are better suited for molecular solutions or polymers like PEDOT. As such, investigating both CuI and PEDOT creates many opportunities for critical thinking and group discussion.

Zoom In Zoom Out Reset image size

Example x-ray diffraction (XRD) spectra for both the good and bad CuI samples are shown in figure 4(a). The bad CuI displays a diffraction peak attributable to the Cu₂O crystal structure, suggesting a large amount of segregated copper oxide in the film. Shown in figure 4(b), large differences in film morphology are apparent in scanning electron microscopy images of the film surfaces. While the good film is composed of small crystallites, the bad film appears more amorphous with triangular crystallites dispersed across the surface. Taken together, all of these results suggest that the bad CuI film is in fact Cu_xI_yO_z with some phase segregation. From this it can be inferred that the annealing process resulted in a substantial amount of copper oxidation.

Shown in figure 4(c), Raman spectra for the PEDOT samples display characteristic peaks. The positions and relative intensities of these peaks can provide the students with valuable information relating to the polymer structure, type of dopant, and level of doping (Chiu et al 2005, 2006, Han et al 2011). As PEDOT films grown following the prescribed recipe rely on residual oxidant as a dopant, differences between the spectra of two films must then be due to changes in the polymer structure or total amount of doping. The peaks at ∼1500 and 1560 cm⁻¹, attributed to asymmetrical stretching of the ${{\rm{C}}}_{\alpha }={{\rm{C}}}_{\beta }$ bond, are observed to have a greater relative intensityfor the bad PEDOT. Additionally, a small peak around 1100 cm⁻¹, attributed to ${{\rm{C}}}_{\alpha }$–S–${{\rm{C}}}_{\alpha }$ bending, has slightly lower intensity (Zhou et al 2011). This experimental evidence points towards a change in the sulfur–carbon bonding constituting the backbone of the PEDOT polymer, potentially caused by differing levels of residual oxidant doping.

While these conclusions are not groundbreaking research results, the sample preparation, data compilation, and data analysis necessary to scientifically support these conclusions is an excellent introduction to the world of authentic science research for undergraduate students.