KU Physics & Astronomy LOcally Organized Assembly (PALOOZA) 2023

Saturday 25 Mar 2023, 09:00 → 17:20 America/Chicago

G192 (Slawson Hall)

Welcome to the third annual PALOOZA! Having been organized by and for Physics & Astronomy students for the past two years, we are excited to bring in our interdisciplinary friends in the Chemistry department.

PALOOZA is a one day event wherein students are able to give 20 minute talks or present posters about their current research so we can learn what our colleagues are doing, learn something new while doing so and get to know each other a little better. Undergraduates are welcome too!

The event takes place in a hybrid format this year on Saturday, March 25th, starting at 9 am. Those interested in attending in person can meet in Slawson Hall, G192 and Ritchie Hall 165. Zoom coordinates are below. Food and drink will be offered both in the morning as a breakfast and afternoon as a lunch.

This event is supported by the University of Kansas Department of Physics & Astronomy. 

 

Zoom:
https://kansas.zoom.us/j/97272166431

 

Meeting ID: 972 7216 6431
Passcode: 100539

09:00 → 09:20 Welcome

I know it's 9am, but get hyped or else

09:20 → 09:40 GUNS at Earth

Neutrinos are the fundamental particles which has promising potent to unveil the mysteries of our universe. Besides their importance in understanding fundamental Physics, a new era of multimessenger astronomy with neutrinos is outspreading which is evident from the large scale neutrino detectors around the world sensitive from meV to PeV energy scales. In this presentation, we discuss the grand unified neutrino spectrum (GUNS) at Earth from various sources. The original work was done by Vitagliano, Tamborra, and Raffelt in their review paper, DOI: 10.1103/RevModPhys.92.045006.

Speaker: Mohammad Ful Hossain Seikh (University of Kansas)

GUNS_At_Earth_PALOOZA2023_Mohammad_Seikh.pdf

09:40 → 10:00 Virgo WISESize: Investigating Environmental Processes of Galaxies in the Virgo Cluster

The cosmic web of the Universe consists of a patchwork of dense clusters and empty voids, with strands of filamentary networks connecting and feeding these central clusters. Observations have demonstrated that these clusters harbor a lower fraction of star-forming galaxies relative to the isolated galaxies of the field, but astronomers have largely “washed out” filaments in their descriptions of cosmic environments, considering only the distance from the cluster center. However, recent large-scale hydrodynamic simulations have revealed filaments to be non-negligible sites of preprocessing as galaxies funnel through these networks into the high density clusters. The aim of the WISESize project is to directly probe galaxies in a more nuanced range of environments, using a combination of infrared and optical wavelength bands to quantify extrinsic galaxy quenching in the local Universe. In particular, we compare the spatial extent of obscured star formation to that of the stars, using the ratio to measure the processing of a galaxy’s gas reservoir in a given environment. With our results, we ultimately will identify the physical processes contributing to this quenching, and in so doing help improve prevailing models of galaxy evolution.

Speaker: Kim Conger (University of Kansas)

PALOOZA Presentation March 2023-1.pdf

10:00 → 10:20 HARPSpec: Using Machine Learning to Find Chemical Compositions of Stars from Their Spectra

Machine Learning is finding its way into an increasingly large number of
scientific fields. In the field of stellar compositions, a new tool called
the Cannon can analyze spectra using a regressive machine learning model to
extract various properties of the stars, most notably their compositions. This
tool is being used to construct HARPSpec, which analyzes publicly available
spectra from the HARPS instrument located in the ESO. This talk covers the
various steps that have been performed to make the Cannon work effectively with
the spectra along with an analysis of its current performance. It will
additionally cover possible future courses of action for further improvements.

Speaker: Joseph Hand

presentation.pdf

10:20 → 10:40 Exploring Quantum Algorithms for Combinatorial Problems at Colliders

Collisions in colliders usually only create a pair of particles. But these particles can decay and create jets leading to many final state particles. It is a binary classification problem to determine which of the two mother particles these final state particles belong to. This becomes exponentially more complex as the number of final particles increases. Variational Quantum Algorithms (VQAs) are hybrid quantum algorithms in which a quantum circuit is parameterized and a classical optimizer is used to minimize the energy of the final state of the quantum circuit. Encoding a cost function as an ansatz for the circuit allows for the solving of this minimization problem. I explore a specific type of VQA called Quantum Approximation Optimization Algorithm (QAOA) applied to the decay of a top and antitop quark pair, in a simple example, and show how a cost function can be transformed into a Hamiltonian and applied as a time-evolution operator as a circuit.

Speaker: Jacob Scott

palooza_2023.pdf

10:40 → 11:00 Break

You're Doing Great!

11:00 → 11:20 Methods to Improving Monte Carlo

Brief presentation of methods to improving your Monte Carlo programming.
To be covered:
0.) "Do you need that precision?"
1.) Sorting assisted random sampling
2.) Fit assisted random sampling
3.) A derivation of an analytic form of random sampling, its benefits
4.) Fits for verification
5.) Statistical measures for verification

Speaker: Brendon Madison (KU HEP , ILC)

BCM_PALOOZA23.pdf

11:20 → 11:40 Modeling the Variability of the Sun's Total Solar Irradiance through Supervised Machine Learning Techniques

With the Sun as the Earth’s main source of energy and heat, it is critical to understand how and why it varies. The Sun’s irradiance changes over its 11-year solar cycle driven by variabilities in the magnetic field. The primary objective of this project was to create a machine learning model that predicts the total solar irradiance (TSI) and then analyze the contributions of each feature within the model in driving change of the Sun’s TSI. This improves upon the previous model by incorporating machine learning techniques on a wider array of data and features. Initially, we utilized the supervised machine learning technique of multiple linear regression to have the model predict an irradiance value for a given day based on given solar features. This model uses data of solar features visible in intensitygrams and line-of-sight magnetograms from the Helioseismic and Magnetic Imager (HMI) instrument aboard the Solar Dynamics Observatory (SDO). Additionally, this model utilized information about the active regions (AR) on the Sun’s surface and several space weather parameters. To ascertain the sunspot area, we performed image processing on the intensitygrams from the SDO HMI. Additionally, seventeen space weather parameters from the SHARP data were added to the model. Due to the number of features, a correlation matrix was created to eliminate features that were highly correlated and determine how features correlated with TSI. In order to quantify the success of this model, the TSI composite data produce was used as it is a historical TSI data record that can be compared to the model’s predictions. Preliminary results show that training the model on these features yields a root mean squared error (RMSE) of 0.0491 and a mean absolute error (MAE) of 0.1600 showing that the model is performing as expected in this early stage. A residual plot was also created to showcase the relation between the actual TSI values and the predicted TSI values. This work serves as the initial step towards continuing to understand the variabilities in the Sun’s irradiance along its solar cycle which will, in turn, aid in understanding how the Sun’s variabilities affect Earth.

Speaker: Ashley Lieber (University of Kansas)

PALOOZA 2023 Slides - Ashley Lieber

11:40 → 12:00 Interlayer Electric Field Screening in Thin Films

An overview of excitonic and eletcric field screening is given and then applied to a heterostructure of Graphene/WS$_2$. An optical setup is devised to quantify this effect within the heterostructure and preliminary results are shared. It is found that the excitonic screening effect can be quantified easily through this optical process through the change of the dielectric of the material. Though challenges will be faced in the future, the goal of quantifying this result continues currently with this project.

Speaker: Neema Rafizadeh

Rafizadeh_Neema.pdf

12:00 → 13:20 Lunch Break

Eat!

12:00 → 13:20 Poster Session

Beautiful, beautiful rectangles of words and colors

12:00 Elucidating the Role of Autotaxin and Lysophosphatidic Acid in Myelination

Multiple sclerosis (MS) is a chronic neurodegenerative disease that affects 2.8 million people worldwide. While several treatments are approved to manage the symptoms of MS, there is no cure and there are no treatments that address the root cause of the disease. The symptoms are caused by the degradation of myelin in the central nervous system (CNS), leaving portions of axons damaged and uninsulated. In healthy patients, oligodendrocytes, the myelinating cells of the CNS, would normally travel to sites of myelin damage and repair these demyelinating lesions. However, in MS patients, oligodendrocytes do not undergo healthy development and therefore do not always appreciably repair the myelin damage. A specific lipid and enzyme pair, lysophosphatidic acid (LPA) and autotaxin, has been established to play a key role in oligodendrocyte development, but the signaling pathway of LPA has not been fully characterized. Through our research, we show that inhibiting autotaxin with orthosteric small molecule inhibitors impedes oligodendrocyte development and that this effect can be rescued by the addition of exogenous LPA. We further plan to investigate the lipid profiles of oligodendrocytes at various stages of development to identify signaling lipids that potentially impact oligodendrocyte maturation. Lastly, we are working to synthesize a bifunctional LPA probe to identify novel binding partners of LPA. By elucidating the mechanisms of LPA action, we could develop better therapeutic strategies to target this pathway and treat MS.

Speaker: Matt Zupan

12:00 Fluorescence-free EV-mRNA expression assay for breast cancer subtyping

Breast cancer prognosis and treatment is generally dependent on a patient’s molecular subtype, those being luminal A, luminal B, HER2-enriched, or basal-like (triple-negative). Current subtyping tests rely on fluorescence-based reporting with barcoded probes, but have low sampling efficiency (>100 ng RNA) and as such, may not accommodate liquid biopsy samples. The goal of this project is to develop a single-molecule mRNA expression assay for breast cancer molecular subtyping using a solid-phase isolation and hybridization approach followed by resistive pulse sensing (RPS). Tumor-derived exosomes are first enriched from plasma samples using a specially designed microfluidic chip with the appropriate affinity markers followed by exosomal mRNA isolated using immobilized gene-specific capture probes. The enriched exosomal mRNA is then hybridized to gene-specific avidin-containing reporter probes. A photocleavable linker allows for release of labeled mRNA-probe complexes with subsequent RPS using a dual in-plane nanopore sensor to identify specific mRNA molecules with single-molecule sensitivity.

Speaker: Mattie Gordon (University of Kansas, Chemistry Department)

12:00 MULTI-CRITERIA OPTIMIZATION

One of the main cancer treatments is radiotherapy
(radiation therapy). In order to reduce the risk of
radiation-related problems, it is important to pro-
vide the tumor with the specified radiation dose
while limiting the unavoidable dose to the nearby
healthy organs. The optimization of radiotherapy
treatment plans looks for machine settings that
produce desirable treatment schemes like pareto
surface or weighted-sum methods. Large-scale
nonconvex multi-criteria optimization is the chal-
lenge at hand

Speakers: Johnpaul Mbagwu (University of Kansas), Michael Chukwuka

MCO_Poster.pdf

12:00 Real Time Imaging of Autotaxin Activity in Oligodendrocyte Precursor Cells

Multiple sclerosis (MS) is a common neurodegenerative disease that has limited treatment options and involves the demyelination of axons within the central nervous system. Myelin sheaths are formed by oligodendrocytes, and this process is regulated by various lipid signaling molecules. One known signaling lipid is lysophosphatidic acid (LPA), which is made from the enzymatic action of autotaxin on lysophosphatidylcholine (LPC). However, previous studies have not fully defined the mechanism by which LPA production promotes myelination. Fluorescence resonance energy transfer (FRET) LPC probes will be developed to monitor autotaxin activity and LPA production in real time during oligodendrocyte differentiation. Two probes will be synthesized, a fluorescence-on and a fluorescence-off probe, which will allow for tracking of autotaxin activity and location within cells. Once these probes have been characterized by 1H nuclear magnetic spectroscopy and mass spectroscopy, they will be evaluated for their performance in vitro to determine which probe provides enhanced fluorescent imaging ability and enzyme activity. The identified probe will then be applied to primary cultures of rodent oligodendrocyte precursor cells (OPCs) and two-photon microscopy will be used to image the probe and autotaxin activity within cells. These studies will provide further insight into the location and activity of LPA production and its involvement in OPC differentiation and myelination, which may reveal new possible targets for future treatment of neurological diseases involving myelination.

Speaker: Esther Holt

13:20 → 13:40 Thermal Noise Reduction for CMS Instrumentation Testing

Discussion of cooling equipment built here in the High Energy Physics (HEP) lab to cool particle detectors for experimentation. The cooling is necessary to reduce thermal noise, an ever present antagonist to a good room-temp experiment.

Speaker: Derek Grove (Physics Grad Student)

Cooling_Block_DWGppt2.pptx

13:40 → 14:00 Vertex Reconstruction of a neutrino event In ARA

The Askaryan Radio Array (ARA) is a large neutrino detector, aimed to detect ultra-high energy neutrinos. The installation of the detector has been started in 2011 and is still going on on the geographic south pole. One of the main motivations behind this experiment is to know the sources of the ultra-high energy neutrinos in the cosmos. It can be pointed out using the vertex reconstruction of the neutrino interaction in Ice. This work contains some results of the vertex reconstruction of the pulses that have been sent through the calibration antennas and received by the RF antennas of the ARA02.
The reconstruction is also used for calibration purposes of the antennas of ARA.

Speaker: Shoukat Ali

PaloozaTalk.pdf

14:00 → 14:20 How nanopatterns in organic molecules affect the interlayer exciton dynamics in organic-MoS2 heterostructures.

We investigated two organic molecules, PTCDI and PTCDA, which stack by making different molecular patterns on the MoS2 surface. Using photoemission spectroscopy, we observed a HOMO and LUMO splitting near the interface in PTCDI/MoS2 but not in PTCDA/MoS2. Density functional theory calculation showed that the splitting of the energy levels originates from the difference in the molecular patterns. Due to the splitting of energy levels, exciton dynamics in PTCDI/MoS2 is very different from that of PTCDA/MoS2. Time-resolved photoemission measurements show that electrons within the IX excitons are spatially localized near the interface in PTCDI/MoS2, whereas in PTCDA/MoS2, they are spatially delocalized across the molecular films. These results demonstrate that nano patterns of organic molecules can be used to control the exciton dynamics in organic/TMD heterostructures.

Speaker: Kushal Rijal (Chan lab KU)

PALOOZA 2023-today.pptx

14:20 → 14:40 Break

Never Give Up!

14:40 → 15:00 Quantum gates implementation for Reinforcement learning decision- making process

This study introduces a method to improve the decision-making process of reinforcement learning agents via quantum information technology. In this approach, states |├ s⟩ of the system are replaced by quantum states (eigenfunctions) of the system |├ ψ⟩, in which a system can be in a superposition of states, and rewards of each step are calculated based on the calculated eigenvalues of the previous step. The agent decides the next step of the system based on the result of the quantum gate’s effect on the available options. The agent uses a probability-based approach to maximize the reward. Due to the high-speed, escalated performance of quantum algorithms, it is hoped that this method will improve the performance of reinforcement learning agents in unknown environments and complicated quantum systems. Finally, a theoretical basis of an experimental setup has been suggested to implement quantum gates for AI agents, using optical methods.

Speaker: Aarash Maroufian (university of kansas)

PLOOZA-2023.odp

PLOOZA2023.pptx

15:00 → 15:20 Multiwavelength Comparison of X-ray and Mid-Infrared Selected Active Galactic Nuclei

Active galactic nuclei (AGN) have different emission spectra depending on their physical characteristics. Using only one wavelength region to identify an AGN may miss sources with unique characteristics. In this work, we compare the multiwavelength properties of X-ray and mid-infrared (MIR) selected AGN in the COSMOS field. Our X-ray-selected AGN sample includes all sources with a $0.5-10$ keV X-ray luminosity greater than $10^{43}$ erg s$^{-1}$. We define our MIR-selected AGN sample using the four Spitzer/IRAC channels. We find that 70% of X-ray selected AGNs are not identified as AGNs using their IR colors, and 40% of IR-selected AGNs do not meet the X-ray luminosity threshold. On average, X-ray-detected MIR-selected AGNs are more powerful than weakly MIR-emitting sources. X-ray selection captures the full range of emission properties, but misses a significant fraction of heavily obscured AGN. MIR-selected AGNs are more luminous in the MIR than X-ray selected AGNs, but have a similar range of ultraviolet (UV) emissions. AGNs that meet both X-ray and MIR selection criteria have the highest median relative energy emission in the UV; they are more luminous across their entire spectra, in general.

Speaker: Thresa Kelly (University of Kansas)

PALOOZA_AGNselection_ThresaKelly_2023.pptx

15:20 → 15:40 Memristor Tuning Through Layer Selection Using ALD

Memristor fabrication techniques have been able to create memristors in the ultrathin regime (sub-2 nm). The fundamental paradox of memristors at this scale is that the number of defects in the active layer must be minimized to prevent undesirable leakage current but needs to include a sufficient and well distributed quantity of oxygen vacancies to promote conductive filament formation. Defects such as pin hole defects and defective metal-insulator interfacial layers can be eliminated through atomic layer deposition (ALD) of a conformal dielectric layer without breaking vacuum following the electrode deposition. ALD can also promote oxygen vacancy by mixing the primary dielectric with a material with a material that acts as an oxygen vacancy reservoir. Predicted by the density function theory simulation, MgO doping in Al2O3 can promote formation of oxygen vacancies while increases resistivity of Al2O3. Motivated by this, this work reports on ultrathin Al2O3 memristors with MgO atomic layers inserted at selected positions in the Al2O3 atomic layer stacks using an in vacuo ALD method to promote oxygen vacancy formation in a controlled manner. Memristors with different numbers of MgO atomic layers and MgO/Al2O3 atomic layer tacking sequences were evaluated to extract the correlation between the atomic structure and oxygen vacancy doping and the memristor performance. It was found that for our memristor thickness (1.8 nm) the number of MgO layers included in the memristor affected the memristor switching behavior and filament formation characteristics. Memristors with fewer MgO layers were found to have lower yield and more likely to undergo dielectric breakdown while memristors with more MgO layers were more likely to exhibit multiple conductive filament formation and variation in the conductive filament size.

Speaker: Berg Dodson (University of Kansas)

2022.03.25_PALOOZA_Berg.pptx

15:40 → 16:00 Break

It's okay to slow down :)

16:00 → 16:20 Bremsstrahlung radiation

Relativistic heavy ions can produce tremendously strong can produce tremendously strong electro-magnetic fields. For Pb-208 nuclei at the LHC, the Electro-magnetic fields have a total energy of approximately 20% of the kinetic energy of the nuclei (1 nucleus ~ 100TEV and 1GEV ~ 1 proton)

To consider a real-life example, let’s take a fly of mass 5 grams moving with a velocity of 10cm/sec. Then its kinetic energy would be 500 ergs. The K.E of the photon field collision would be comparable to one- third of a fly (~ 160 ergs).

When charged fields decelerate the E-M fields must change and the object emits bremsstrahlung (Breaking radiation) photons. Collisions of Pb nuclei produce the most rapid decelerations in the universe (in the order of 10^22 g’s).

We can measure the deceleration “a” using low energy data with a factor of ¾. This will help us figure out the number of gluons in a Pb nuclei.

Speaker: Abishai Mathai

Palooza talk Bremstralung completed.pptx

16:20 → 16:40 Review of Mathematics, Numerical Factors, and Corrections for Dark Matter Experiments Based on Elastic Nuclear Recoil

Many theories are postulated for what Dark Matter may be, one of them is a simple particle. A summary and derivation of nuclear recoil interactions are presented for weakly interacting dark matter particles. Modifications for the such interactions include Earth's motion, possibility of spin for the particles and nuclear form factors. Different elements are discussed and how it may impact on detectors.

Speaker: Miguel Soto Alcaraz

DMPalooza.pdf

16:40 → 17:00 Clouds and Clarity: Revisiting Atmospheric Feature Trends in Neptune-size Exoplanets

The last decade has seen the maturation of precise exoplanet transmission spectroscopy techniques, allowing for the identification of individual opacity sources in exoplanetary atmospheres. As more and more of these planets have been observed, tentative trends have been identified relating the strength of observed atmospheric features to other fundamental planetary and system parameters. However, small sample sizes and limited spectral sensitivity (primarily from the 1-1.8 micron G141 grism on Hubble's WFC3) make these identified trends uncertain. Previous work has, out of necessity, grouped dissimilar planets (massive Hot Jupiters as well as smaller, cooler gaseous planets) together in order to compile samples large enough to draw inferences from, but it is unclear that these planets exhibit all the same atmospheric physics which could affect their observed atmospheric spectra. We present a re-analysis of a smaller, yet more consistent sample of Neptune-size exoplanet atmospheric spectra, across a range of temperatures, and discuss whether these previously identified trends hold at the dawn of the JWST era.

Speaker: Yoni Brande

Brande_PALOOZA.pptx

17:00 → 17:20 Conclusion

Good job, now for beers or beer-adjacents