arXiv subjects

Philip Chang

Publications and source records attributed to Philip Chang.

At least 19 recordsLinked to original sources

arXiv2026

Moving-Mesh Simulations of Mini-Common Envelope Ejection in Classical Novae

Although well studied, our understanding of the mass ejection mechanisms of cataclysmic variables remains incomplete. Recent work suggests that binary interaction plays an important role in driving and shaping this mass ejection and may affect the long-term evolution of the system. In this paper, we perform a three-dimensional moving-mesh hydrodynamic simulation of a cataclysmic variable system to study the effect of binary interaction on mass ejection. We find that once the flow crosses the ${\rm L}_1$ Lagrange point, the material is ejected roughly isotropically. This can be seen in a roughly spherical distribution of the ejecta at large radii. We also show that the ${\rm L}_2$ Lagrange point is not important in the ejection of mass, contrary to the assumption in some previous work in this area. Finally, we find that the specific angular momentum of the ejected material is larger than its initial specific angular momentum. This enhanced angular momentum ejection likely affects the long-term evolution of the binary system.

astro-ph.HE↗

arXiv2026

Stars with Plumbing Issues: The Formation of Collimated Outflows on Common-Envelope Simulations and Comparison to Water Fountains Observations

Common-envelope evolution (CEE) is one of the biggest open questions in binary stellar evolution, despite being the main channel for the formation of close binaries. One of the main reasons CEE is difficult to model is the lack of direct observations that could constrain numerical simulations. One exception is luminous red novae, which are thought to represent CEEs that end in mergers. Unfortunately, there are no confirmed direct detections of ongoing events that result in the survival of a close binary, and we must rely on observations of post-CEE systems. Among these, planetary nebulae (PNe) are particularly important because their morphologies can probe how the envelope is ejected. However, post-CEE PNe do not reflect the ejected envelope in its pristine form, as winds from the central core also affect their morphology. In this context, Water Fountains (WFs), a class of objects proposed to form during CEE, provide an ideal comparison. They are identified by their collimated water masers, and most are still in the post-AGB phase. As such, WFs provide some of the best observational constraints for simulations, since they likely capture a snapshot of the envelope ejection while it is still happening. In this paper, we show that the formation of a circumbinary disk with collimated outflows surrounding the central binary arises naturally from hydrodynamical simulations of CEE, and that their morphology and kinematics are consistent with observations of WFs. We also present insights into how the properties of WFs may provide clues to understanding how CEE proceeds and help guide future simulations.

astro-ph.SR↗

arXiv2026

Building an AI-native Research Ecosystem for Experimental Particle Physics: A Community Vision

Experimental particle physics seeks to understand the universe by probing its fundamental particles and forces and exploring how they govern the large-scale processes that shape cosmic evolution. This whitepaper presents a vision for how Artificial Intelligence (AI) can accelerate discovery in this field. We outline grand challenges that must be addressed to enable transformative breakthroughs and describe how current and planned experimental facilities can implement this vision to advance our understanding of the vast and complex physical world from the smallest to the largest scales. We show how facilities currently under construction, such as the HL-LHC, DUNE and soon EIC, can both benefit from and serve as proving grounds for this vision, while also enabling a longer-term goal for how future experiments -- like FCC-ee at CERN, IceCube-Gen2, a Muon Collider in the U.S., and smaller to mid-scale projects -- can be fully AI-native. We describe how a truly national-scale collaboration, jointly managed across large funding partners, and involving both DOE laboratories and universities, can make this happen.

hep-ex↗

arXiv2025

Neutron Star-Main Sequence Collisions Robustly Form Dynamically Stable Thorne-\.Zytkow Objects

Thorne-\.Zytkow Objects (T\.ZOs) are hypothetical hybrid stars with a neutron star at the core of a large, diffuse envelope. (T\.ZOs) may be formed when a newly formed neutron star that is kicked by its supernova collides with its main-sequence companion. Using a moving-mesh hydrodynamics solver integrated into the parallel-code Charm N-body GrAvity solver, we demonstrate that these ``impact scenario'' formation processes robustly form (T\.ZOs) for periastron distances less than one stellar radius. These (T\.ZOs) are dynamically stable and they can serve as initial models for further evolutionary studies.

astro-ph.SR↗

arXiv2025

Constraints on a dark matter sub-halo near the Sun from pulsar timing

Using pulsar accelerations, we identify and constrain the properties of a dark matter sub-halo in the Galaxy for the first time from analyzing the acceleration field of binary and solitary pulsars. Our MCMC calculations show that this sub-halo has a mass of $2.45^{+1.07}_{-0.96} \times 10^{7}~M_{\odot}$ and is located at Galactocentric coordinates $X = 7.43^{+0.2}_{-0.12}~\rm$ kpc, $Y = 0.38^{+0.11}_{-0.16} ~\rm kpc$, $Z = 0.21^{+0.06}_{-0.11} ~\rm kpc$, using flat, uninformative priors, where we have modeled the sub-halo as a compact object. The Bayes factors for the models are in the range of $\sim$ 20-40, which indicates tentative evidence (though not yet decisive) for the sub-halo. Modeling the sub-halo with a NFW profile gives a sub-halo mass within the scale radius (0.1 kpc) of $0.48^{0.15}_{-0.16} \times 10^{7} M_{\odot}$, located at $X = 7.47^{+0.21}_{-0.14}$, $Y=0.38^{+0.11}_{-0.16}$, $Z=0.21^{+0.06}_{-0.11}$. We examine \textit{Gaia} data and the atomic and molecular hydrogen data of our Galaxy and show that the measured deviation from a smooth potential cannot arise from the gas or the stars in our Galaxy. By analyzing the full sample of binary pulsars with available acceleration measurements, we find that massive (with mass $>10^{8}~M_{\odot}~$) sub-halos are disfavored for the Milky Way within several kiloparsec of the Sun. Smaller sub-halos are beyond the reach of current direct acceleration measurements. The presence of a $\sim 10^{7}~M_{\odot}$ sub-halo within a few kpc of the Sun is potentially consistent with the expected number counts of sub-halos in the prevailing $\Lambda$CDM paradigm, for a substantial sub-halo mass fraction. This work now provides a proof of principle for probing nearby, low-mass sub-halos, and has implications across many fields of astrophysics - from understanding the nature of dark matter to galaxy formation.

astro-ph.GA↗

arXiv2025

SkimROOT: Accelerating LHC Data Filtering with Near-Storage Processing

Data analysis in high-energy physics (HEP) begins with data reduction, where vast datasets are filtered to extract relevant events. At the Large Hadron Collider (LHC), this process is bottlenecked by slow data transfers between storage and compute nodes. To address this, we introduce SkimROOT, a near-data filtering system leveraging Data Processing Units (DPUs) to accelerate LHC data analysis. By performing filtering directly on storage servers and returning only the relevant data, SkimROOT minimizes data movement and reduces processing delays. Our prototype demonstrates significant efficiency gains, achieving a 44.3$\times$ performance improvement, paving the way for faster physics discoveries.

cs.DC↗

arXiv2024

Line Segment Tracking: Improving the Phase 2 CMS High Level Trigger Tracking with a Novel, Hardware-Agnostic Pattern Recognition Algorithm

Charged particle reconstruction is one the most computationally heavy components of the full event reconstruction of Large Hadron Collider (LHC) experiments. Looking to the future, projections for the High Luminosity LHC (HL-LHC) indicate a superlinear growth for required computing resources for single-threaded CPU algorithms that surpass the computing resources that are expected to be available. The combination of these facts creates the need for efficient and computationally performant pattern recognition algorithms that will be able to run in parallel and possibly on other hardware, such as GPUs, given that these become more and more available in LHC experiments and high-performance computing centres. Line Segment Tracking (LST) is a novel such algorithm which has been developed to be fully parallelizable and hardware agnostic. The latter is achieved through the usage of the Alpaka library. The LST algorithm has been tested with the CMS central software as an external package and has been used in the context of the CMS HL-LHC High Level Trigger (HLT). When employing LST for pattern recognition in the HLT tracking, the physics and timing performances are shown to improve with respect to the ones utilizing the current pattern recognition algorithms. The latest results on the usage of the LST algorithm within the CMS HL-LHC HLT are presented, along with prospects for further improvements of the algorithm and its CMS central software integration.

hep-ex↗

arXiv2024

Galactic Structure From Binary Pulsar Accelerations: Beyond Smooth Models

We measure the line-of-sight accelerations of 26 binary pulsars due to the Milky Way's gravitational potential, and produce a 3-dimensional map of the acceleration field of the Galaxy. Acceleration measurements directly give us the change in the line-of-sight velocity at present day, without requiring any assumptions inherent to kinematic modeling. We measure the Oort limit ($ρ_0=0.062\pm0.017$ \msun/pc$^3$) and the dark matter density in the midplane ($ρ_{0,\textrm{DM}}=-0.010\pm0.018$ \msun/pc$^3$); these values are similar to, but have smaller uncertainties than previous pulsar timing measurements of these quantities. Here, we provide for the first time, values for the Oort constants and the slope of the rotation curve from direct acceleration measurements. We find that $A=15.4\pm2.6$ km/s/kpc and $B=-13.1\pm2.6$ km/s/kpc (consistent with results from \textit{Gaia}), and the slope of the rotation curve near the Sun is $-2\pm5$ km/s/kpc. We show that the Galactic acceleration field is clearly asymmetric, but due to data limitations it is not yet clear which physical processes drive this asymmetry. We provide updated models of the Galactic potential that account for various sources of disequilibrium; these models are incompatible with commonly used kinematic potentials. This indicates that use of kinematically derived Galactic potentials in precision tests (e.g., in tests of general relativity with pulsar timing) may be subject to larger uncertainties than reported. The acceleration data indicates that the mass of the Galaxy within the Solar circle is $2.3 \times 10^{11}$ M$_\odot$, roughly twice as large as currently accepted models. Additionally, the residuals of the acceleration data compared to existing Galactic models have a dependence on radial position; this trend can be explained if the Sun has an additional acceleration away from the Galactic center.

astro-ph.GA↗

arXiv2024

Improving tracking algorithms with machine learning: a case for line-segment tracking at the High Luminosity LHC

In this work, we present a study on ways that tracking algorithms can be improved with machine learning (ML). We base this study on the line segment tracking (LST) algorithm that we have designed to be naturally parallelized and vectorized in order to efficiently run on modern processors. LST has been developed specifically for the CMS Experiment at the LHC, towards the High Luminosity LHC (HL-LHC) upgrade. Moreover, we have already shown excellent efficiency and performance results as we iteratively improve LST, leveraging a full simulation of the CMS detector. At the same time, promising deep-learning-based tracking algorithms, such as Graph Neural Networks (GNNs), are being pioneered on the simplified TrackML dataset. These results suggest that parts of LST could be improved or replaced by ML. Thus, a thorough, step-by-step investigation of exactly how and where ML can be utilized, while still meeting realistic HL-LHC performance and efficiency constraints, is implemented as follows. First, a lightweight neural network is used to replace and improve upon explicitly defined track quality selections. This neural network is shown to be highly efficient and robust to displaced tracks while having little-to-no impact on the runtime of LST. These results clearly establish that ML can be used to improve LST without penalty. Next, exploratory studies of GNN track-building algorithms are described. In particular, low-level track objects from LST are considered as nodes in a graph, where edges represent higher-level objects or even entire track candidates. Then, an edge-classifier GNN is trained, and the efficiency of the resultant edge scores is compared with that of the existing LST track quality selections. These GNN studies provide insights into the practicality and performance of using more ambitious and complex ML algorithms for HL-LHC tracking at the CMS Experiment.

physics.ins-det↗

arXiv2023

Envelope Ejection and the Transition to Homologous Expansion in Common-Envelope Events

We conduct a long-timescale ($5000\,$d) 3-D simulation of a common-envelope event with a $2\,M_{\odot}$ red giant and a $1\,M_{\odot}$ main sequence companion, using the moving-mesh hydrodynamic solver MANGA. Starting with an orbital radius of $52\,R_{\odot}$, our binary shrinks to an orbital radius of $5\,R_{\odot}$ in $200\,$d. We show that over a timescale of about $1500\,$d, the envelope is completely ejected while $80$ per cent is ejected in about $400\,$d. The complete ejection of the envelope is solely powered by the orbital energy of the binary, without the need for late-time reheating from recombination or jets. Motivated by recent theoretical and observational results, we also find that the envelope enters a phase of homologous expansion about $550\,\rm d$ after the start of our simulation. We also run a simplified 1-D model to show that heating from the central binary in the envelope at late times does not influence the ejection. This homologous expansion of the envelope would likely simplify calculations of the observational implications such as light curves.

astro-ph.SR↗

arXiv2023

A non-interacting Galactic black hole candidate in a binary system with a main-sequence star

We describe the discovery of a solar neighborhood (d=468 pc) binary system with a main-sequence sunlike star and a massive non-interacting black hole candidate. The spectral energy distribution (SED) of the visible star is described by a single stellar model. We derive stellar parameters from a high signal-to-noise Magellan/MIKE spectrum, classifying the star as a main-sequence star with $T_{\rm eff} = 5972 \rm K$, $\log{g} = 4.54$, and $M = 0.91$ \msun. The spectrum shows no indication of a second luminous component. To determine the spectroscopic orbit of the binary, we measured radial velocities of this system with the Automated Planet Finder, Magellan, and Keck over four months. We show that the velocity data are consistent with the \textit{Gaia} astrometric orbit and provide independent evidence for a massive dark companion. From a combined fit of our spectroscopic data and the astrometry, we derive a companion mass of $11.39^{+1.51}_{-1.31}$\msun. We conclude that this binary system harbors a massive black hole on an eccentric $(e =0.46 \pm 0.02)$, $185.4 \pm 0.1$ d orbit. These conclusions are independent of \cite{ElBadry2022Disc}, who recently reported the discovery of the same system. A joint fit to all available data (including \cite{ElBadry2022Disc}'s) yields a comparable period solution, but a lower companion mass of $9.32^{+0.22}_{-0.21} M_{\odot}$. Radial velocity fits to all available data produce a unimodal solution for the period that is not possible with either data set alone. The combination of both data sets yields the most accurate orbit currently available.

astro-ph.GA↗

arXiv2022

Line Segment Tracking in the HL-LHC

The major challenge posed by the high instantaneous luminosity in the High Luminosity LHC (HL-LHC) motivates efficient and fast reconstruction of charged particle tracks in a high pile-up environment. While there have been efforts to use modern techniques like vectorization to improve the existing classic Kalman Filter based reconstruction algorithms, Line Segment Tracking takes a fundamentally different approach by doing a bottom-up reconstruction of tracks. Small track stubs from adjoining detector regions are constructed, and then these track stubs that are consistent with typical track trajectories are successively linked. Since the production of these track stubs is localized, they can be made in parallel, which lends way into using architectures like GPUs and multi-CPUs to take advantage of the parallelism. The algorithm is implemented in the context of the CMS Phase-2 Tracker and runs on NVIDIA Tesla V100 GPUs. Good physics and timing performance has been obtained, and stepping stones for the future are elaborated.

physics.ins-det↗

arXiv2022

Segment Linking: A Highly Parallelizable Track Reconstruction Algorithm for HL-LHC

The High Luminosity upgrade of the Large Hadron Collider (HL-LHC) will produce particle collisions with up to 200 simultaneous proton-proton interactions. These unprecedented conditions will create a combinatorial complexity for charged-particle track reconstruction that demands a computational cost that is expected to surpass the projected computing budget using conventional CPUs. Motivated by this and taking into account the prevalence of heterogeneous computing in cutting-edge High Performance Computing centers, we propose an efficient, fast and highly parallelizable bottom-up approach to track reconstruction for the HL-LHC, along with an associated implementation on GPUs, in the context of the Phase 2 CMS outer tracker. Our algorithm, called Segment Linking (or Line Segment Tracking), takes advantage of localized track stub creation, combining individual stubs to progressively form higher level objects that are subject to kinematical and geometrical requirements compatible with genuine physics tracks. The local nature of the algorithm makes it ideal for parallelization under the Single Instruction, Multiple Data paradigm, as hundreds of objects can be built simultaneously. The computing and physics performance of the algorithm has been tested on an NVIDIA Tesla V100 GPU, already yielding efficiency and timing measurements that are on par with the latest, multi-CPU versions of existing CMS tracking algorithms.

physics.ins-det↗

arXiv2022

Snowmass2021 Cosmic Frontier White Paper: Observational Facilities to Study Dark Matter

We present an overview of future observational facilities that will significantly enhance our understanding of the fundamental nature of dark matter. These facilities span a range of observational techniques including optical/near-infrared imaging and spectroscopy, measurements of the cosmic microwave background, pulsar timing, 21-cm observations of neutral hydrogen at high redshift, and the measurement of gravitational waves. Such facilities are a critical component of a multi-pronged experimental program to uncover the nature of dark matter, while often providing complementary measurements of dark energy, neutrino physics, and inflation.

astro-ph.CO↗

arXiv2021

Eclipse timing the Milky Way's gravitational potential

We show that a small, but \textit{measurable} shift in the eclipse mid-point time of eclipsing binary (EBs) stars of $\sim$ 0.1 seconds over a decade baseline can be used to directly measure the Galactic acceleration of stars in the Milky Way at $\sim$ kpc distances from the Sun. We consider contributions to the period drift rate from dynamical mechanisms other than the Galaxy's gravitational field, and show that the Galactic acceleration can be reliably measured using a sample of $\textit{Kepler}$ EBs with orbital and stellar parameters from the literature. Given the uncertainties on the formulation of tidal decay, our approach here is necessarily approximate, and the contribution from tidal decay is an upper limit assuming the stars are not tidally synchronized. We also use simple analytic relations to search for well-timed sources in the \textit{Kepler} field, and find $\sim$ 70 additional detached EBs with low eccentricities that have estimated timing precision better than 1 second. We illustrate the method with a prototypical, precisely timed EB using an archival \textit{Kepler} light curve and a modern synthetic \textit{HST} light curve (which provides a decade baseline). This novel method establishes a realistic possibility for obtaining fundamental Galactic parameters using eclipse timing to measure Galactic accelerations, along with other emerging new methods, including pulsar timing and extreme precision radial velocity observations. This acceleration signal grows quadratically with time. Therefore, given baselines established in the near-future for distant EBs, we can expect to measure the period drift in the future with space missions like \textit{JWST} and the \textit{Roman Space Telescope}.

astro-ph.GA↗

arXiv2021

A measurement of the Galactic plane mass density from binary pulsar accelerations

We use compiled high-precision pulsar timing measurements to directly measure the Galactic acceleration of binary pulsars relative to the Solar System barycenter. Given the vertical accelerations, we use the Poisson equation to derive the Oort limit, i.e., the total volume mass density in the Galactic mid-plane. Our best-fitting model gives an Oort limit of $0.08^{0.05}_{-0.02} M_{\odot}/\rm pc^{3}$, which is close to estimates from recent Jeans analyses. Given the accounting of the baryon budget from McKee et al. (2015), we obtain a local dark matter density of $-0.004^{0.05}_{-0.02}~M_{\odot}/\rm pc^{3}$, which is slightly below other modern estimates but consistent within the current uncertainties of our method. While this first measurement of the Oort limit (and other Galactic parameters) has error bars that are currently several times larger than kinematical estimates, they should improve in the future. We also constrain the oblateness of the potential, finding it consistent with that expected from the disk and inconsistent with a potential dominated by a spherical halo, as is appropriate for our sample which is within a $\sim$ kpc of the Sun. We find that the slope of the rotation curve is not constrained by current measurements of binary pulsar accelerations. We give a fitting function for the vertical acceleration $a_{z}$: $a_{z} = -α_{1}z$; $\log_{10} (α_{1}/{\rm Gyr}^{-2})=3.69^{0.19}_{-0.12}$. By analyzing interacting simulations of the Milky Way, we find that large asymmetric variations in $da_{z}/dz$ as a function of vertical height may be a signature of sub-structure. We end by discussing the power of combining constraints from pulsar timing and high-precision radial velocity (RV) measurements towards lines-of-sight near pulsars, to test theories of gravity and constrain dark matter sub-structure.

astro-ph.GA↗

arXiv2020

The Effect of Impact Parameter on Tidal Disruption Events

Stars that pass too close to a supermassive black hole are disrupted by the black hole's tidal gravity and leads to some debris being ejected while the remainder accretes into the black hole. To better study the physics of this debris, we use the moving mesh code MANGA to follow the evolution of the star from its initial encounter to its complete destruction. By varying the impact parameter ($β$) of the star, we studied the energy distribution of the remaining material and the fallback rate of the material into the black hole as a function of time. We show that the spread of energy in the debris and peak luminosity time ($t_{\rm peak}$) are both directly related to the impact parameter. In particular, we find a $β^{1/2}$ scaling for the energy spread for $β=2-10$ and a frozen evolution for $β\gtrsim 10$. We discuss implication of this scaling for the rise time of the light curve and broadness of the luminosity peak for these lower $β$'s. These relationships provide a possible means of inferring the impact parameters for observed TDEs.

astro-ph.HE↗

arXiv2020

Towards a direct measure of the Galactic acceleration

High precision spectrographs can enable not only the discovery of exoplanets, but can also provide a fundamental measurement in Galactic dynamics. Over about ten year baselines, the expected change in the line-of-sight velocity due to the Galaxy's gravitational field for stars at $\sim$ kpc scale distances above the Galactic mid-plane is $\sim$ few - 10 cm/s, and may be detectable by the current generation of high precision spectrographs. Here, we provide theoretical expectations for this measurement based on both static models of the Milky Way and isolated Milky Way simulations, as well from controlled dynamical simulations of the Milky Way interacting with dwarf galaxies. We simulate a population synthesis model to analyze the contribution of planets and binaries to the Galactic acceleration signal. We find that while low-mass, long-period planetary companions are a contaminant to the Galactic acceleration signal, their contribution is very small. Our analysis of $\sim$ ten years of data from the LCES HIRES/Keck precision radial velocity (RV) survey shows that slopes of the RV curves of standard RV stars agree with expectations of the local Galactic acceleration near the Sun within the errors, and that the error in the slope scales inversely as the square root of the number of observations. Thus, we demonstrate that a survey of stars with low intrinsic stellar jitter at kpc distances above the Galactic mid-plane for realistic sample sizes can enable a direct determination of the dark matter density.

astro-ph.GA↗