First Week of the Program From Left to Right: Jarad Mitchell, Olivia Del Guerico, Abigail Johnson, Eric Peterson, Andrew Davis, Mark Mathis, Agatha Ulibarri, Kierstin Sorensen
Eric Peterson: Dissociation of molecular hydrogen in a microwave-induced plasma
Whiteboard Video: This summer I worked at the University of Paris-Sud with Abby Johnson from the REU program, and Dr. Olivier Leroy of the Laboratoire de Physique des Gaz et des Plasmas, in collaboration with Dr. Jacques Robert of the Laboratoire Aimé Cotton. We used optical emission spectroscopy of the Hɑ line to investigate the dissociation of molecular hydrogen in a microwave-induced plasma confined to a capillary. Hydrogen gas is naturally found on earth in its molecular form H2; the goal of this project is to create an efficient source of atomic hydrogen for ultraviolet laser applications.
The plasma is generated in an argon/hydrogen gas mixture at low pressure (10^-5 to 10^-3 mbar) and using low powers (a few dozen Watts). Owing to its lack of chemistry, argon is amenable for sustaining a plasma, while addition of hydrogen to the mixture tends to reduce the plasma size. Due to this, there is a balancing game: pumping in more molecular hydrogen results in more atomic hydrogen production, but only up to a point, at which the amount of dissociated hydrogen saturates.To power the plasma, microwaves are sent from a generator through a novel cavity resonator based on a circular “stripline” after which they propagate as surface waves along the boundary between the capillary tube and plasma. Although the dynamics of the plasma are very complicated, one mechanism of hydrogen dissociation is through scattering with high-energy electrons.
We varied several experimental parameters, including pressure, gas mixture, and power input. For example, in one set of experiments, we observed the effect of changing the proportions of hydrogen and argon while keeping the pressure constant. In another set of experiments, we kept the gas flows constant and only changed the amount of power sent from the generator. We found two important results: first, that hydrogen production is maximized when hydrogen comprises 10-20% of the gas mixture, and second, that hydrogen production is proportional to the input power.
This research is part of a preliminary study for a project, coordinated with Dr. Jacques Robert, which aims to build a reliable continuous-wave Lyman-alpha laser. This laser would be useful for cold atom experiments, such as the spectroscopy and laser cooling of antihydrogen.
I would like to extend my thanks to Dr. Leroy for supervising me this summer, to LPGP for being friendly and welcoming, to the University of Michigan for coordinating this opportunity, and to the NSF for sponsoring it.
Eddy current testing is a common nondestructive procedure for detecting cracks in materials. An eddy current probe is made of two solenoids attached to two cores that are aligned parallel to one another. Essentially, an alternating electric field is passed through the solenoids. This generates a magnetic field. When placed close to the material in question, an electric current is generated that creates a magnetic field opposing the one generated by the probe. This electric current is known as an eddy current. The lag in the magnetic field through the material is due to an impedance that can be measured. Ferromagnetic materials are interesting because the stress on the material alters the magnetic properties, and for this reason these are known as magento-mechanical materials. Previously, the response of certain ferromagnetic materials to uniaxial stress has been modeled and observed experimentally. Now, the task has been to characterize this response in a biaxial stress state. A difficulty in measuring biaxial stress is that a single impedance measurement corresponds to multiple stress states. A method is being developed to describe the most likely stress configuration at a given impedance measurement. The two approaches to this method are known as the image processing method and the interpolation method. The image processing method takes as input a map of the measured impedance values at each stress configuration that have been mapped to a colormap. Using Open CV in Python, a binary mask is created for values that correspond to a given measurement. This is done for four different orientations of the eddy current probe. These orientations are superposed to give the most likely area of stress configurations. The measurements also fall within a defined lower and upper bound to take error into account. The second method, the interpolation method, is very similar to image processing except the binary mask is formed directly from the array of measured values before any processing. Values that fall between the given array of measured values are considered using an interpolation function. The two methods are compared and their limitations are explored. Important considerations in developing these methods were the delta-measure and bilinear interpolation. The delta-measure accounted for statistical and experimental error, to construct the lower and upper bound, we had to choose a plausible delta-measure. The delta-measure is designed to reflect this method when it is implemented in the real world. The interpolation technique provided more values upon which a stress identification could be done, thereby increasing precision. Although this did not add any information, it does provide more instances upon which to run the program. Next steps for this research are to extend to more complex stress configurations, apply this procedure to real world measurements, and to continue exploring the robustness and accuracy of the process. This summer was very engaging and challenging on many levels. In my experiences at the laboratory and in Paris I was able to interact with many perspectives. I learned the value of concise communication and of understanding different cultures. In the lab, I learned tactics to improve efficiency is delivering results, synthesize difficult information in a short time, and develop innovative solutions to problems. I also made strides in programming in Python and MATLAB. From this experience, I have developed a stronger interest in pursuing materials science in graduate school. This research was performed in the Group of Electrical Engineering Paris at the University Paris-Sud under the direction of Dr. Laurent Daniel, Dr. Yann le Bihan, and Laurent Santandrea. Top
Olivia Del Guercio: Direct Laser Writing of Magneto-photonic Microstructures for Biomedical Applications
Direct Laser Writing (DLW), a type of optical lithography, is a microfabrication tool which utilizes a laser to polymerize a photoresist thereby creating structures on the scale of nanometers. There are two main types of Direct Laser Writing: one photon absorption (OPA) or two-photon absorption (TPA). OPA works by placing a photoresist substance on a substrate and briefly shining a continuous laser of an absorptive frequency upon it. Due to strong linear absorption, photons cannot propagate more than a few micrometers through the photoresist, allowing for only 1D or 2D fabrication . Two-photon absorption works by stimulating local nonlinear absorption with a high power 800 nm femtosecond laser. This highly precise method enables 3D fabrication, but it does have the drawback of necessitating expensive equipment. Throughout this experiment we will utilize low one-photon absorption (LOPA), which not only uses a cost-effective low-power continuous 534 nm laser but also has the capability of producing high quality 3D structures. LOPA takes advantage of an oil-immersion high numerical aperture objective lens which can focus the few milliwatts of power emitted from the laser by up to a billion times at the focus in a volume smaller than a square micrometer . If applied to a low-absorption photoresist the photons will not have the energy to induce polymerization, with the exception of this highly concentrated focus point. As a result, temperatures can surpass the glass-transition temperature of SU-8 (210ºC) . This high temperature has the advantage of eliminating the need for a post-exposure heating that is necessary in OPA . The ability to create structures on the scale of nanometers has the potential for a wide variety of applications. In particular, when the magnetic nanoparticles are built into the structures they can be controlled remotely with a magnetic field. This technology would allow for vast improvements in targeted drug transport treatments . Additionally, Fe3O4 nanoparticles increase in temperature when a laser or magnetic field is applied, which can be useful in cancer hyperthermia treatment . SU-8 is toxic to humans, but in small doses with appropriate surface treatments cytocompatibility can be significantly increased . This summer, I was able to successfully fabricate magnetic microswimmers (approx. 10 µm in length) which could be controlled with a magnetic field. I also fabricated other structures including a propeller and a spring, but was unable to show a magnet induced response.
References: 1. Raja Das, Natalia Rinaldi-Montes, Javier Alonso, Zakariae Amghouz, Eneko Garaio, Jose Angel Garcia, Pedro Gorria, Jesus Angel Blanco, Manh-Huong Phan, and Hariharan Srikanth. Boosted hyperthermia therapy by combined ac magnetic and photothermal exposures in ag/fe3o4 nanoflowers. ACS Applied Materials & Interfaces, 8(38):25162–25169, 2016. 2. Mai Trang Do, Thi Thanh Ngan Nguyen, Qinggele Li, Henri Benisty, Isabelle Ledoux-Rak, and Ngoc Diep Lai. Submicrometer 3d structures fabrication enabled by one-photon absorption direct laser writing. Optics express, 21(18):20964–20973, 2013. 3. MicroChem. Su-8 2000 permanent epoxy negative photoresist. 2015. 4. Quang Cong Tong, Dam Thuy Trang Nguyen, Minh Thanh Do, Mai Hoang Luong, Bernard Journet, Isabelle Ledoux-Rak, and Ngoc Diep Lai. Direct laser writing of polymeric nanostructures via optically induced local thermal effect. Applied Physics Letters, 108(18): 183104, 2016. 5. Varadraj N. Vernekar, D. Kacy Cullen, Nick Fogleman, Yoonsu Choi, Andr ́es J. Garc ́ia, Mark G. Allen, Gregory J. Brewer, and Michelle C. LaPlaca. Su-8 2000 rendered cytocompatible for neuronal biomems applications. Journal of biomedical materials research. Part A, 89(1):138– 151, Apr 2009. 6. Xiaohui Yan, Qi Zhou, Jiangfan Yu, Tiantian Xu, Yan Deng, Tao Tang, Qian Feng, Liming Bian, Yan Zhang, Antoine Ferreira, and Li Zhang. Magnetite nanostructured porous hollow helical microswimmers for targeted delivery. Advanced Functional Materials, 25(33):5333– 5342, 2015.
Plasmonic nanoparticles are particles whose electron density can couple with electromagnetic radiation of wavelengths that are far larger than the particle due to the nature of the dielectric-metal interface between the medium and the particles. What is interesting about them is that at a macroscopic scale, they may cause the material they are made of to transmit a color different from what it usually transmits. They can be fabricated in several materials but my project being concerned with gold nanoparticles. These are formed after gold film is annealed to about 500 degrees Celsius, causing the entire film to morph into particles. However, Dr. Ngoc Diep Lai and his students developed a laser writing method that could control the exact position of which the nanoparticles are formed. This method, known as the Low One-Photon Absorption Direct Laser Writing method, involves a continuous wave laser at 532nm focused onto a glass substrate sputtered with gold. The point where the laser is focused reaches the 500 degrees C necessary to form nanoparticles in an area smaller than a micrometer. This point may be accurately moved around by procedurally programming a 3D Piezoelectric translator (PZT) which is included in the laser setup. By doing this, it is possible to expose specific areas of the substrate for controlled amounts of time, resulting in nanoparticles forming in the laser’s path.Quong Cong Tong, a former PhD student of Dr. Lai, wrote a series of programs that moved the PZT in such a way that the laser could print images onto the substrate. Two examples of this are a one euro coin image and the word "NANO" written at a micro scale made of gold nanoparticles. These images gave me an idea for my project this summer. First, I started with the "Nano Typewriter". Using an object-oriented style of programming, I expanded the program Cong used to write the individual letters so that it could write every letter, number, and ascii symbol. The main function allows the user to type a series of letters at a time, all of which will simultaneously be laser written. After adding more and more functionality, I had created what was essentially a micro-scale word processor. It is able to print letters even smaller than what Cong' claimed to be the smallest size that could be printed, less than 1.9μm. Shortly afterward, I expanded on the image processing program. I wrote it in such a way that it can take any jpeg or png image and break the picture down into its different hues, generating a trajectory map for the PZT of each color. Various parameters of the laser's power and the speed of the PZT could be adjusted so that additional colors could be seen in the final image. The colors in the resulting image are, in my opinion, not very vibrant and are quite green. However, this leaves the project open for future work. In addition to experimenting with additional parameters to achieve different colors, other materials may also be experimented with as well. Gold is not the only material that is able to produce plasmonic nanoparticles. There may be others that create nanoparticles that exhibit a far more vibrant spectrum of plasmonic resonances. The potential of LOPA DLW may be further reached if the variables contributing to this phenomenon are reached. One of these uses is replacing electron beam milling as the dominant technology behind producing chips, discs, and other data storage devices. Top
Mark Mathis: Gaussian Beam to Top-Hat Beam Shaping of an Ultrafast Laser Pulse
Whiteboard Video: This Summer, I had the pleasure of working under Rodrigo Lopez-Martens in the Physique du Cycle Optique (PCO) group at the Laboratoire d’Optiquee Appliquée. This group has created a laser system which produces pulses of only 3-5 femtoseconds. With this short, high-intensity pulse, the PCO group does experiments to measure plasma dynamics and generate soft X-rays, ion beams, and electron beams. My project related to high-intensity laser pulses hitting a solid target. The high-intensity creates a plasma at the interface of the solid which then acts as a mirror. As time progresses, the plasma expands outwards and changes the position of the mirror. By using one pulse to create the plasma mirror and another pulse to reflect off of the mirror with a specific time-delay, the velocity of the plasma expansion can be measured. The problem with this is that the first laser pulse has a Gaussian profile and the plasma expansion velocity is proportional to the square root of intensity. The plasma expansion velocity therefore changes as a function of position.
To fix this problem, I was given an optic which acts as a Gaussian to top-hat converter. The lab tried to use this optic once before, but did not do much characterization and hoped it would work out. It did not. I was tasked with simulating of what to expect from the device, characterizing it under ideal conditions, and characterizing it under the experimental conditions. I was given no information about the phase delay from the company (probably because they don’t want someone to steal their technology), so for the first week, I created an algorithm in Matlab to determine the phases at which a top-hat would form at focus. The algorithm never converged to something that gave a top-hat at focus. I then realized that the top-hat formed just a little before focus. I changed the algorithm to do far-field propagation to focus and then near-field propagation back to where the top-hat might form. There were two problems: having to propagate light twice within the program took a very long time and I was unsure of exactly where to expect the top-hat to form. I moved on to the experimental characterization.
The characterization of the device was the most fun task. I built my own optical systems based on what I thought was necessary, took data, and improved on my optical system. It seems like I made a new optical system every week. One week, I made a Michelson interferometer, another week, I aligned an off-axis paraboloid, another week, I made a spatial filter. It was a very fulfilling experience. I ended up characterizing where the top-hat formed, its size compared to focus, the average intensity relative to peak focal intensity, and how misalignment affected the top-hat. There were two other top-hats I characterized which weren’t perfect, but might’ve been of use for the experiment. There was not much time left for characterizing the top-hat converter under experimental conditions, but I did get a few measurements.
There are a lot of people to thank for allowing me to have such an amazing experience. Thank you Rodrigo for hosting me and helping me aid in your research, thank you to everyone in the PCO and APPLI group for helping me and for great lunch conversations, thank you to Steve Yalisove and John Nees for creating this opportunity, thank you to the National Science Foundation for sponsoring this, and thank you to my fellow REU students for making this a fun experience. Top
Abby Johnson: Generating Atomic Hydrogen in a Microwave Plasma
Whiteboard Video: This summer I worked at the Laboratoire de Physique des Gaz et des Plasmas (LPGP) with Olivier Leroy, my fellow American student Eric Peterson, and our collaborator Jacques Robert. The goal of our project was finding the best mixture of molecular hydrogen gas and argon gas that would not only support an ignited microwave plasma, but would produce as much atomic hydrogen as possible. Our other goal was to study how changes in the input microwave power, input gas flow, and the ratio of argon to hydrogen gas in our mixture influenced our microwave plasma.
The plasmas we ignited in a small diameter capillary tube are responsible for creating the desired atomic hydrogen. These generated plasmas and their intense heat cause the molecular hydrogen gas to disassociate. In other words, the energy from our plasma led to the break down of the molecular hydrogen, leaving behind two atomic hydrogen (2H).
Specifically, we created a 2.45 GHz microwave plasma in a small silica capillary tube with an inner diameter of 1 mm. The hydrogen and argon mass flows were small, so we were working in a low-pressure range of 10-5 to 10-3 mBar. A microwave generator provided an input power ranging from 20-50 Watts to our uniquely designed circular stripline excitator that actually ignited the plasma. This circular stripline device was specially created for this experiment. It utilizes the design of the commonly known stripline excitator, but wraps it in a circle and then sandwiches two of the circular stripline devices together. This round shape allows our small capillary tube to go through the device with ease. The circular stripline excitator is covered in an electro-magnetic shielding material to limit the EM radiation and maximize the energy going into the microwave plasma.
To determine how effective our mixtures were at creating atomic hydrogen, we used an Optical Emission Spectroscopy diagnostic that produced spectra. Specifically, we looked at the H-alpha, H-beta, and a few Argon lines in the spectra. The intensity of the H-alpha line has a positive correlation with the amount of atomic hydrogen created by the plasma. So, we wanted mixtures with large H-alpha intensities.
Our results will be important down the line for our collaborator Jacques Robert. His eventual goal is to have an experimental set up that generates a steady beam of atomic hydrogen from plasma. He will then cool this beam to 4 Kelvin or less. From there, Jacques Robert plans to study the laser applications of this cold hydrogen. Our work this summer is merely the first step towards making his theoretical research possible.
To find the ratio of argon to molecular hydrogen that produced the largest intensity in the H-alpha line, we first picked a set pressure value. Then we increased hydrogen and decreased argon flux, but continued to maintain our chosen pressure. We followed this procedure for three pressures: 1.7 x 10-4, 2.5 x 10-4, and 1.7 x 10-4 mBar. The results showed that all three pressures had a maximum H-alpha line intensity in mixtures containing 10-20% hydrogen. The lower the chosen pressure, the less hydrogen needed to generate the maximum intensity. While we examined other low pressure, gas ratio, and microwave power limits, this finding is most important for continuing the research.
In the future, we would like to alter our experiment to allow us to look at larger gas flows and pressures. That way, we can test mixtures with 10-20% hydrogen and see if they maintain their effectiveness in higher pressure. I would like to thank my advisor Olivier Leroy, Eric Peterson, LPGP, Jacques Robert, and U Michigan for this amazing, educational summer. It has been an incredible experience and I would highly recommend it to any rising senior who wants to not only learn what a research environment is like, but learn a great deal about a different culture. Top
Kierstin Sorensen: Steady-state anisotropy of lasing dyes
This summer, I’ve worked with Professor Ngoc Diep Lai’s research group at ENS Cachan, particularly on the project of This summer, during the Optics in the City of Light REU program, I worked with Prof. Mélanie Lebental at École Normale Supérieur (ENS) de Cachan in the Laboratoire de Photonique Quantique et Moléculaire (LPQM). In that time, I studied the steady-state anisotropy or stationary anisotropy for various lasing dyes at various concentrations. This data was necessary to calibrate highly time-resolved anisotropy experiments performed by Jean-Frédéric (Jeff) Audibert.
In our experiments, we passed linearly polarized laser light, polarized vertically, through a dye doped sample and analyzed the resulting light. After the laser light was passed through the sample, the resulting light was not only vertically polarized like the laser was, but some was also horizontally polarized. The portion that was vertically polarized like the laser is referred to as the parallel intensity (IP), and the portion that was horizontally polarized is referred to as the orthogonal intensity (IO). These different portions are used to find the anisotropy in a formula defined by: r = (IP – IO)/(IP + 2IO).
Our most substantial problems came from finding a solvent that suited our needs. Initially, we tried using poly(methyl methacrylate) (PMMA) beads dissolved using tetrahydrofuran (THF) in a glass test tube. When this failed, we tried to use Squalane in a 10-mm quartz cuvette. However, Squalane was not sufficiently viscous and was fluorescent itself. Both of these unsuccessful solvents were analyzed using a spectrofluorometer in a L-format. Finally, we settled on a previously successful method. Our sample was a spin-coated layer of PMMA and lasing dye on a glass slide. Our set-up involved a 532-nm pulse laser, a 343-nm pulse laser, a manual shutter, linear polarizer film, a spectrometer, and one of our samples with the layer side away from the laser.
To analyze the raw data from our many experiments, I created multiple computer programs in the Python editor Pyzo since Python is a user-friendly computer language widely used in the world of scientific research. One of the benefits to using the Pyzo editor was that it allowed me to create and easily adjust graphs of the raw data. This feature allowed me to align some of Jeff’s data that did not have a uniform time zero meaning that the peak in the parallel and orthogonal intensities was not always at time zero.
With the final set-up, we had some promising results. However, we ran out of time to use another solvent from which we expected to have better results. This may become follow-up work for another researcher.
Top Agatha Ulibarri: Optimizing Small Angle Polarization Rotation Measurements