Polluted White Dwarfs (2022 - Present)

About 25-50% of observed white dwarfs atmospheres are polluted by elements heavier than Helium, even though these elements are expected to sink in quickly. Previous work show that accretion of materials over the course of several billions years is needed to match observations. Current consensus is that the white dwarf accretes from the remnant planets and planetesimals, which are tidally disrupted as they get close to the star. Two important questions arise:

1. What are potential reservoirs in an evolved planetary system?
2. Which mechanisms excite these bodies' eccentricity so that they may be able to get close to the white dwarf?

For my PhD, I am considering several alternative reservoirs and proposing novel mechanisms to deliver these materials close to the star. These projects are mostly numerical (and some analytical), and are focused on the planetary dynamics of evolved stellar systems.

D. Pham, H. Rein, “Polluting White Dwarfs with Oort Cloud Comets”, 2024, MNRAS, accepted.
D. Pham, H. Rein, D.S. Spiegel; “A new timestep criterion for N-body simulations”, 2024, OJAp, 7, 1.

Photometric Characterization of Terrestrial Exoplanets (2017 - Present)

Upcoming missions like HabEx or LUVOIR are designed to be able characterize rocky exoplanets spectroscopically. However, spectroscopy is time-intensive and in the (hopefully) not too far future, we will need initial characterization to prioritze targets for spectroscopy follow-up. In this project, we focus on using photometry as a tool for initial characterization using machine learning and Bayesian inferencing techniques.

J. Li, L. Kaltenegger, D. Pham, D. Ruppert “Characterization of extrasolar giant planets with machine learning ”, 2021, MNRAS Letters, 527, L137.
D. Pham, L. Kaltenegger; “Follow the water: finding water, snow, and clouds on terrestrial exoplanets with photometry and machine learning”, 2022, MNRAS Letters, 513, L72.
D. Pham, L. Kaltenegger; “Color Classification of Earth-like Planets with Machine Learning”, 2021, MNRAS, 504, 6106.

Simulation & Atmosphere Removal for the TIME Collaboration (2021 - Present)

TIME (Tomographic Ionized Carbon Intensity Mapping Experiment) is an [CII]/CO line-intensity mapping instrument seeking to understand the epoch of reionization. Having an understanding of the epoch of reionization remains a central task in building a complete picture of the Universe's evolution, early galaxies formation, and star formation history. In this project, I am mostly involved in building a simulation pipeline and perform atmosphere removal for TIME observations.

Clustering and Predicting Planet Types from Protoplanetary Disk Properties in Simulations (2019 - 2021)

State-of-the-art planet formation models are becoming increasingly capable of accounting for a wide spectrum of planet types, at the cost of increasing complexity. A natural question arises: are there links between initial conditions and the simulated planets' observables? We perform a clustering analysis using unsupervised machine learning methods on state-of-the-art planet formation models. From the planet mass, radius and orbital parameter space, Gaussian Mixture Model was able to cluster and identify the same planet types as those defined by observations, without any a priori knowledge of the model or planetary architecture. A Random Forest Classifier is then trained to predict which planet type would result, given some disk properties. Through this, we found important features of the protoplanetary disks in core accretion models.

M. Schlecker, D. Pham, R. Burn, Y. Alibert, C. Mordasini, A. Emsenhuber H. Klahr, Th. Henning, L. Mishra; “The Determinism of Global Planet Formation Models”, 2021, A&A, 656, A73.

Bounded cosmic string backreaction and connecting Green's functions for the wave equation (2019-2020)

Senior thesis work. Cosmic superstrings are one dimensional strings from string theory. They are stretched to macroscopic scale during the early inflationary phase of the universe. There have been work to place experimental constraints on these strings using various astrophysical techniques ranging from lensing to gravitational wave background. First, we present a brief analysis on the cosmic string equation of motion and backreaction, along with simplifications through physical setup and gauge choices, for the one dimensional bounded cosmic string. Then, we change our focus to present a complete analysis of the wave equation under various boundary conditions: infinite string, semi-infinite string, periodic, Dirichlet, and Neumann. Through that, potential problems are considered and relationships between these boundary conditions are connected.

Undegraduate Senior Thesis

Prediction of Lensing Events by Compact Objects (2015 - 2017)

Gravitational lensing allows mass measurement, without actually detecting light from the lens itself. In this aspect, compact objects are ideal lenses, since they are massive and are dim. In this work, we study and predict lensing events for nearby, known compact objects (black holes, neutron stars, white dwarfs). Here, we focus on the centroid shift caused by the lens onto background stars. For white dwarfs, we find that 30-50 events are expected per decade. In addition, monitoring microlensing events will allow us to detect binary companions and orbiting exoplanets.

A. Harding, R. Di Stefano, S. Lépine, J. Urama, D. Pham, C. Baker; “Predicting Gravitational Lensing by Stellar Remnants”, 2018, MNRAS, 475, 79.