Massachusetts Institute of Technology
Doctorate, Electrical Engineering June 2010 to June 2015
We proposed, designed, and demonstrated a first-of-a-kind millimeter-scale thermophotovoltaic (TPV) system using a metallic microburner, photonic crystal (PhC) emitter, and low bandgap TPV cells. Many technologies (fuel cells, Stirling, thermoelectric, etc.) that potentially enable a portable millimeter-scale hydrocarbon microgenerator are under active investigation because conventional fuels offer energy densities fifty times that of batteries. In a TPV system, combustion heats an emitter to incandescence and the resulting thermal radiation is converted to electricity by photovoltaic cells. Our approach uses a moderate temperature (1000-1200 C) microburner coupled to a high emissivity, high selectivity PhC selective emitter and low bandgap TPV cells. The PhC emitter and low bandgap cells minimize total microgenerator mass by enabling simultaneous high efficiency and high power density, even at moderate temperatures which allow efficient coupling to the combustion process by reducing undesired heat loss mechanisms. This approach is predicted to be capable of up to 30% efficient fuel-to-electricity conversion within a millimeter-scale form factor.
Although considerable technological barriers need to be overcome to reach full performance, we have performed a robust experimental demonstration that validates the theoretical framework and the key system components. We first demonstrated a first-of-a-kind TPV system built from a 10x10 mm catalytic silicon MEMS microburner with a Si/SiO2 1D PhC matched to the InGaAsSb (Eg = 0.55 eV) cells which achieved 2.7% fuel-to-electricity efficiency, a millimeter-scale record, at a power of 344 mWe. We then proposed, designed, and demonstrated a highly robust metallic platform comprised of a 20x20 mm Inconel microburner and a higher performance 2D tantalum PhC emitter. With the new system, we experimentally demonstrated a similar efficiency but can achieve 5% with simple mechanical improvements. These two experimental demonstrations will pave the way for a lightweight, high energy density TPV microgenerator. We modeled a complete microgenerator based on the experimental system and found an energy density of 850 Wh/kg and power density of 40 W/kg are achievable.
Massachusetts Institute of Technology
Master of Engineering, Electrical Engineering June 2008 to June 2010
Hydrocarbon fuels have such a high energy density that even a relatively inefficient converter of chemical energy into electrical can significantly exceed the energy density of state- of-the-art batteries. This work attempts to do exactly this on a millimeter scale by means of thermophotovoltaic (TPV) power conversion approach. We demonstrated the first-of-a-kind propane-oxygen fueled catalytic silicon based MEMS microreactor integrated with low-bandgap GaInAsSb (0.53 eV bandgap) photovoltaic cells to create a fully operational millimeter scale TPV system. The initial fuel to electricity system efficiency was measured at 0.8%. A cell area of 2 cm2 produced 200 mW of electricity from a chemical input of 28 W. These results match well with developed system models. Additionally, we predict the efficiency can be doubled by improving the view factor, vacuum packaging, and eliminating parasitic radiation from the edges of the reactor. By integrating simple one-dimensional silicon/silicon dioxide photonic crystal on the micro-reactor as spectral shaping device efficiency can reach 5%.
Massachusetts Institute of Technology
Bachelor of Science, Electrical Engineering and Physics Sept. 2004 to June 2008
Selected coursework, including graduate: Quantum (6.730, 8.04), Solid State Physics (6.731), Nanoelectronics (6.975), Semiconductor Optoelectronics (6.732), Optics (6.637), Mathematical Methods in Nanophotonics (18.369), MEMS (6.777), Microfabrication Lab (6.152), Analog Design (6.301) and Lab (6.101), Digital Lab (6.101), Classical Feedback Control (6.302). GPA 4.5/5.0.
D. M. Bierman, A. Lenert, W. R. Chan, B. Bhatia, I. Celanovic, M. Soljacic, E. N. Wang, "Enhanced photovoltaic energy conversion using thermally based spectral shaping," Nature Energy, vol. 1, 23 May 2016.
V. Stelmakh, W. R. Chan, M. Ghebrebrhan, J. Senkevich, J. D. Joannopoulos, M. Soljacic, I. Celanovic, "Sputtered Tantalum Photonic Crystal Coatings for High-Temperature Energy Conversion Applications," IEEE Transactions on Nanotechnology, vol. 15, no. 2, pp. 303-309, March 2016.
Y. Yeng, W. Chan, V. Rinnerbauer, V. Stelmakh, J. Senkevich, J. Joannopoulos, M. Soljacic, and I. Celanovic, "Photonic crystal enhanced silicon cell based thermophotovoltaic systems," Opt. Express, vol. 23, no. 3, pp. A157-A168, 9 Feb. 2015.
A. Lenert, D. M. Bierman, Y. Nam, W. R. Chan, I. Celanovic, M. Soljacic, E. N. Wang, "A nanophotonic solar thermophotovoltaic device," Nature Nanotechnology, vol. 9, no. 2, pp. 126-130, Jan. 2014
Y. Yeng, W. Chan, V. Rinnerbauer, J. Joannopoulos, M. Soljacic, and I. Celanovic, "Performance analysis of experimentally viable photonic crystal enhanced thermophotovoltaic systems," Opt. Express, vol. 21, no. S6, pp. A1035-A1051, Nov. 2013.
W. R. Chan, P. Bermel, R. C. N. Pilawa-Podgurski, C. H. Marton, K. F. Jensen, J. J. Senkevich, J. D. Joannopoulos, M. Soljacic, I. Celanovic, "Toward high-energy-density, high-efficiency, and moderate-temperature chip-scale thermophotovoltaics," Proceedings of the National Academy of Sciences, vol. 110, no. 14, pp. 5309-5314, Jan. 2013.
Y. X. Yeng, M. Ghebrebrhan, P. Bermel, W. R. Chan, J. D. Joannopoulos, M. Soljacic, and I. Celanovic, "Enabling high-temperature nanophotonics for energy applications," Proceedings of the National Academy of Sciences, vol. 109, no. 7, pp. 2280-2285, Oct. 2012.
W. Chan, R. Huang, C. Wang, J. Kassakian, J. Joannopoulos, I. Celanovic, "Modeling low-bandgap thermophotovoltaic diodes for high-efficiency portable power generators," Solar Energy Materials and Solar Cells, vol. 94, no. 3, pp. 509-514, Mar. 2010.
P. Bermel, M. Ghebrebrhan, W. Chan, Y. X. Yeng, M. Araghchini, R. Hamam, C. H. Marton, K. F. Jensen, M. Soljacic, J. D. Joannopoulos, S. G. Johnson, I. Celanovic, "Design and global optimization of high-efficiency thermophotovoltaic systems," Opt. Express, vol. 18, no. 103, pp. A314-A334, Sept. 2010.
W. R. Chan, V. Stelmakh, C. M. Waits, M. Soljacic, J. Joannopoulos, I. Celanovic, "A Thermophotovoltaic System Using a Photonic Crystal Emitter," ASME Micro Nano Heat Mass Transfer, MNHMT2016-6695, Jan. 2016.
W. R. Chan, V. Stelmakh, C. M. Waits, M. Soljacic, J. D. Joannopoulos, and I. Celanovic, "Photonic Crystal Enabled Thermophotovoltaics for a Portable Microgenerator," J. Phys. Conf. Ser., vol. 660, p. 012069, Dec. 2015. (PowerMEMS 2015)
W. R. Chan, V. Stelmakh, B. Wilhite, J. J. Senkevich, M. Soljacic, J. D. Joannopoulos, and I. Celanovic, "Millimeter scale, all metallic, high-energy-density thermophotovoltaic generator," presented at 11th Conference on Theromophotovoltaics, Amsterdam, Sept. 2014. (Invited)
W. R. Chan, V. Stelmakh, V. Rinnerbauer, J. J. Senkevich, M. Soljacic, J. D. Joannopoulos, I. Celanovic, "Thermophotovoltaic fuel-to-electricity conversion enabled by high-temperature photonic crystals," presented at XII International Conference on Nanostructured Materials, Moscow, July 2014.
W. R. Chan, C. M. Waits, J. D. Joannopoulos, and I. Celanovic, "Thermophotovoltaic and thermoelectric portable power generators," Proc. SPIE 9083, Micro- and Nanotechnology Sensors, Systems, and Applications VI, vol. 9083, pp. 90831W-90831W-8, 2014. (Invited)
W. R. Chan, B. A. Wilhite, J. J. Senkevich, M. Soljacic, J. Joannopoulos, and I. Celanovic, "An all-metallic microburner for a millimeter-scale thermophotovoltaic generator," J. of Phys. Conf. Ser., vol. 476, no. 1, pp. 012017, Dec. 2013. (PowerMEMS 2013)
W. R. Chan, V. Rinnerbauer, Y. X. Yeng, J. Senkevich, M. Soljacic, J. D. Joannopoulos, I. Celanovic, "Millimeter scale metal based catalytic reactor TPV system," presented at 10th Conference on Thermophotovoltaics, Frankfurt, Sept. 2012.
W. Chan, P. Bermel, R. Pilawa-Podgurski, C. Marton, K. Jensen, M. Soljacic, J. Joannopoulos, I. Celanovic, "A high-efficiency millimeter-scale thermophotovoltaic generator," presented at 9th Conference on Thermophotovoltaics, Valencia, Sept. 2010.
W. Chan, I. Celanovic, N. Pallo, J. Kassakian, J. Joannopoulos, R. Huang, C. Wang, "Modeling InGaAsSb TPV Cells for Portable Power System Design," presented at 8th Conference on Thermophotovoltaics, Palm Springs, Nov. 2008.
I. Celanovic, W. Chan, P. Bermel, Y. X. Yeng, C. Marton, M. Ghebrebrhan, M. Araghchini, K. Jensen, M. Soljacic, J. Joannopoulos, S. G. Johnson, R. Pilawa-Podgurski, P. Fisher. Thermophotovoltaic energy generation, November 24 2011. US Patent App. 13/112,831.
A Nixie tube is a neon-filled "vacuum tube" display typically seen in home- or kit-built clocks. Nixie Rex was an entirely custom designed and built from scratch. The Nixie tubes were fabricated from Pyrex glass and wire. Custom electronics were also needed to supply the higher voltages and currents required. Each Nixie tube is nearly two feet tall, making it not only the only Nixie clock built from scratch but also the largest.
New York Maker Faire 2013
Homemade Light Bulb
Inspired by Claude Paillard's triodes, we decided to build our own vacuum tubes. We would need to start with something a little simpler: a Thomas Edison style light bulb. The bulb was constructed from two pieces. The stem contained two dumet glass-to-metal seals, stainless steel support wires, and a tungsten filament. The metal components were assembled with a homebuilt mini spot welder. The bulb was sealed onto the completed stem and contained a tubulation for evacuating the bulb. The finished bulbs had limited lifetime due to the envelope blackening, possibly caused by residual water vapor.
Partridge, John Henry. Glass-to-metal Seals. Sheffield: Society of Glass Technology, 1949.
The Flammenwald is a two dimensional Rubin's tube. While one dimensional Rubin's tubes are commonplace, 2D Rubin's tubes are quite a rarity in the literature. Like its 1D analog, the Flammenwald visualizes cavity modes. The cavity is excited acoustically and the flame height is correlated with RMS pressure, e.g. nodes have low flame height and anti-nodes have high flame height. In our system, the modes did not match pure square modes or pure circular modes because of the speakers in the corners.
Harold A. Daw. Art on a Two-Dimensional Flame Table. Leonardo, Vol. 24, No. 1 (1991), pp. 63-65.
Jasia Reichardt. Art at Large. New Scientist, Vol. 52 (1972), p. 525.
MIT has a limitless supply of decommissioned CRT monitors and the best thing to do with 60 monitors was build a CRT wall 8 feet tall and 24 feet wide. The Lead Curtain was built for East Campus rush and lit up the dance floor at the East Side Party. The heart of the system was a Xilinx FPGA (field programable gate array) which generated the video signals. The signals were distributed to multiple DAC (digital to analog converter) boards. One DAC board generated the analog VGA signal to drive four monitors using shift registers and resistor networks. Each DAC was connected to the FPGA by an RS485 serial link. This bandwidth limited the resolution of the individual monitors to 16 pixels, giving the overall display a resolution of 16 by 60. This resolution was chosen because it is the highest resolution that could be easily achieved without a more sophisticated architecture that could handle an order of magnitude higher data rates. Regardless, standing at the center when 10 kilowatts worth of monitors come out of simultaneously is quite an experience, assuming the circuit breakers don't trip.
Portable Plasma Globe
A small circuit was built to turn an ordinary light bulb into a miniature plasma globe for the East Campus Soldering Seminar. A plasma globe was the only way to top the Pirate Radio soldering seminar from 2007, both technically and in coolness. The goal was to convert an ordinary light bulb into a handheld, AA-powered plasma globe using the magic of power electronics. After several forays into resonant and class E converters, we settled on a simple flyback topology. The flyback used a CCFL (cold cathode fluorescent light) transformer from Coilcraft. These are ideal because they're small, cheap, rated for high voltage and have a ridiculous turns ratio. The only downside is that the transformer must be operated just on the verge of dielectric breakdown to get plasma. Some kind of snubber, clamp or protection circuit is essential or the transformer will blow the plasma even strikes. Our major breakthrough was realizing that only certain brands, styles and wattages of light bulbs work. We ended up testing dozens and buying a stockpile of the one that worked best.
Hydra Cornelius Laser Show
One of the major problems with traditional mechanically scanned laser shows is that inertia limits the spatial bandwidth of the image. Circles are fine but try drawing a star with a homemade galvo scanner. Scanning offers complete control over the beam's trajectory but this is unnecessary for music visualization. The Hydra Cornelius is a novel approach to laser music visualization using a dynamic lens to continuously draw an abstract pattern.
The Hydra was based on a dynamic lens formed by a small puddle of water. An expanded laser beam was shot through the bottom of the puddle and projected onto the ceiling. A small speaker excites the water, creating patterns much like those made by a sunlight in a swimming pool. An audio input modulates the amplitudes of two oscillators tuned to excite a different resonant modes in the water. The end result is a slowly varying hypnotizing pattern that jumps and moves with the music.
Jearl Walker. Caustics: Mathematical Curves Generated By Light Shined Through Rippled Plastic. The Amateur Scientist. Scientific American. Sept 1983.
Jearl Walker. A Drop of Water Becomes a Gateway into the World of Catastrophe Optics. The Amateur Scientist. Scientific American. Sept 1989.
Don't mock that cheap AM/FM radio—at least until you have some idea of what is going on inside, and all for under $5 in components. MIT students couldn't reproduce the performance of even the cheapest radio for the East Campus Soldering Seminar. The pirate radio station was an attempt to introduce incoming freshmen to real electrical engineering, in an event long dominated by blinkey LED thingies.
A pirate radio station broadcasting at 1.4MHz was constructed from a function generator and a re-purposed cable TV distribution amplifier. The antenna was a T roughly 50 feet high and 100 feet long. Because it was impractical to construct a full sized antenna, we had to make due with a matching network to get any power out of the power amplifier. Even so, the transmitter was quite weak and was unable to cover the campus.
For the receiver, we tried super heterodyne and super regenerative, AM and FM, 1.4MHz and 10.7MHz, and just about everything in between. Definitely a learning experience. Out of desperation, we used a simple resonant circuit for tuning into an unused AM station, a simple four stage RF amplifier, a diode demodulator, and an op amp audio amplifier. As a result the radio has poor sound quality and awful tuning, but the frosh loved it.
Fluorescent VU Meter
What is cheap, fast, efficient, and bright? Not LEDs—but the lowly fluorescent light. In the VU meter, the fluorescent lights were not operated in the traditional manor. Instead they were driven similarly to a plasma globe, with a high voltage, high frequency source from a single electrode. By modulating the voltage, the length of the tube that was lit could be controlled. The voltage was changed by modulating the drive frequency to a high-Q series resonant circuit. The VU meter was controlled by an XMMS plugin which outputs the frequency data to a PIC microcontroller via the serial port.
Analog Laser Light Show
For 6.101 (Analog Circuit Lab) I built an analog laser light show consisting of two galvos and a controller. In each galvo, a magnet is held in a vertical orientation because it is attracted to an iron housing. Coils create a horizontal magnetic field that slightly deflects the magnet. An analog feedback loop controls the coil current to maintain a position command. Position is measured by an LED-photodiode. The two galvos are arranged in an x-y configuration to steer a laser beam. The system could draw circles, Lissajous figures, and other interesting patterns. Given how poorly everything was constructed, it is surprising it worked at all. In fact, it worked for several months after the class until a cactus fell onto it.
Bemix Integrated Music System
Bemix eludes definition. Nominally, Bemix is a system for playing music in multiple locations. It began as a one-day project to inconspicuously pipe music into a bathroom and grew into an umbrella name that encompassed everything that combined to software, electronics, or music with a bit of engineering or creativity. The name is a combination of the name of our section of an MIT dorm, Bemis, and mix.
Bemix went through several iterations. Version one was a set of $2 PC speakers hidden above the ceiling tiles of the Bemis bathroom. The audio cable was run out the window of the bathroom and into my room where it was connected to the guts of a Pentium II computer carefully sprawled out on my bookcase. The PII box ran Glirnath which allowed residents to queue up music for their bathroom needs over the web. A crude touch switch was added to allow users to mute the speakers directly from the bathroom. Even before the first iteration was finished, the next one was in the works. Glirnath was not scalable to multiple bathrooms. The touch switch was unreliable and did not communicate with the computer.
In the latest iteration, playlists could be constructed through a web interface and loaded into any player. Each player could be controlled remotely over the web interface or locally with specialized hardware such as capacitive touch switches disguised as murals. At its pinnacle, Bemix included nearly half a mile of wiring running through hundreds of feet of fire code-approved cable trays, a LAN with sound card servers, web servers, file servers, serial port servers, and public terminals, a Christmas light visualization in a lounge, and about 23 software layers to run the beast.
Autonomous Robotic Soccer
MIT-Harvard Robocup Team as electrical engineering team leaderFall 2005 to Summer 2009
Architected system level design of a team of five autonomous soccer playing robots for international competition. Led a team that designed and implemented electronics. Designed closed loop brushless DC motor driver that controls the wheels using an optical shaft encoder, PIC microcontroller and three phase mosfet bridge. Designed high voltage capacitor kicking system that kicks the ball using a solenoid. Capacitors are charged to 450 volts using a flyback converter and discharged through the solenoid with an SCR.
High Speed Amplifier Design
Linear Technology as design intern June 2007 to Sept. 2007
Characterized open and closed loop AC and DC performance of a differential, high speed operational amplifier using spectrum analyzer, network analyzer, oscilloscope, and function generator. Discovered, debugged, and corrected problem where ESD protection cell would short circuit when the IC was hot socketed.
Nanotube and Graphene Devices
Massachusetts Institute of Technology as undergraduate researcher Jan. 2006 to June 2007
Worked under Prof. Jing Kong on carbon based transistors. Developed a procedure to grow long aligned nanotubes using chemical vapor deposition (CVD). Built a low pressure CVD system for nanotube growth. Characterized devices by scanning electron microscopy, atomic force microscopy, and electrically.
Mars Gravity Biosatellite
Massachusetts Institute of Technology as undergraduate researcherSept. 2004 to Dec. 2005
Designed high efficiency LED lighting system using a constant current switching supply for a mouse cage to be used aboard a satellite. Developed algorithm to process a live video feed of mouse movement.
MIT EECS Best Masterworks Oral Presentation 2009
Massachusetts Institute of Technology
77 Massachusettes Ave.
Cambridge, MA 02138