2025 EIC User Group Early Career Workshop

Jul 11, 2025, 1:00 PM → Jul 13, 2025, 9:30 PM US/Eastern

Henry Klest (Argonne National Laboratory), Jennifer Rittenhouse West (jennifer.rittenhouse@gmail.com) (UC Berkeley), Wenliang (Bill) Li (College of William and Mary), Douglas Higinbotham (Jefferson Lab)

The 2025 EIC Early Career (EC) Workshop will take place from July 11-13, 2025 at Jefferson Lab, Newport News, VA, US.

This event is dedicated to students, postdocs & early career researchers but is open to everyone. The meeting will be run in a hybrid mode, both in-person and remotely on Zoom.
 

Important Registration Information for this hybrid workshop:

For those attending in person, we would like to highly encourage you to present your research to your peers.  As something new, we are asking those giving talks to include one or two opening slides at a TED-talk level (i.e., how would you explain what you are doing to a general audience and/or how it fits in to the bigger EIC picture). These TED slides will go at the beginning of your talk, and then you will continue with a "standard" talk for the rest of your time slot.

While no payment is needed to complete the EIC Early Career Workshop registration we do need to make sure everyone completes two steps:

1. Apply for Jefferson Lab site access ( for your total time at JLab): https://misportal.jlab.org/jlabAccess/guests/42840/visits/new
2. Register for free for the early career meeting and let us know if you would like to be considered for an invited talk: https://misportal.jlab.org/ul/conferences/generic_conference/registration.cfm?conference_id=EarlyCareerWorkshop25

 

To participate in the EICUG main event, please complete a separate registration page: https://www.jlab.org/conference/eicugepic/registration. ( there is no fee for students to attend the main meeting ) 

 

 

               

EICUG-EarlyCareer-2025v9.pdf

Fri, July 11

1:00 PM → 5:30 PM Friday Afternoon

1:00 PM What's so epic about ePIC?

Experiment at the EIC

Speaker: Maria Zurek (Argonne National Laboratory)

1:40 PM EIC Theory

2:20 PM EIC Software Discussion

Speakers: Holly Szumila-Vance (FIU), Stephen Kay (University of York)

2:40 PM Coffee Break

3:00 PM Machine Learning for Event Reconstruction at the Electron-Ion Collider (EIC)

We present a novel two‐stage particle‐identification (PID) workflow for the ePIC Barrel Imaging Calorimeter (BIC) at the future Electron–Ion Collider. In the first stage, we exploit the classical calorimeter‐to‐track energy ratio E/p (optimally summing energy across the first eight SciFi layers) to achieve a 97 % electron efficiency and a pion rejection factor R_π≈23.5. In the second stage, we reshape high‐granularity per‐hit data into a pseudo-image (layers × top-hits) × feature-channels, encoding normalized hit energy, radial coordinate, angular separations, and sub-detector flags. A VGG-style convolutional neural network trained on these five channels learns residual shower‐shape differences between electrons and pions. By selecting the CNN output threshold to yield an additional 97 % efficiency on the pre-cut sample, we demonstrate an overall electron efficiency of 95 % with a combined pion rejection R_π≈174, representing nearly an order-of-magnitude improvement over E/p alone. Our work delivers a production-ready ML‐augmented PID module, poised for integration into the EICrecon framework and deployment in upcoming ePIC physics analyses.

Speaker: Tomas Sosa Giraldo (University of Manitoba)

3:20 PM Status of the ePIC EEEMCal Test Beam Analysis*

In February 2025, a prototype for the ePIC Electron Endcap Electromagnetic Calorimeter (EEEMCal) was tested at the DESY II Test Beam Facility in Hamburg, Germany. The prototype consisted of 25 lead-tungstate ($\hbox{PbWO}_4$) crystals with an SiPM readout. Each crystal was instrumented with 16 SiPMs, read out either individually, with four in parallel, or with all 16 in parallel by the H2GROC ASIC, which was developed for the CMS High Granularity Calorimeter upgrade and is being adapted into the CALOROC readout for the ePIC detector. The prototype was exposed to beams of electrons with an energy from $1$ to $6$ $\rm GeV$ to characterize primarily the energy resolution of the detector, as well as position scans which can be used to study the position resolution. This talk will summarize the test beam campaign up through the current state of the analysis.

Speaker: Tristan Protzman (Lehigh University)

3:40 PM Energy Resolution Analysis for EIC's Barrel Imaging Calorimeter*

The Electron-Ion Collider is an in-development facility that will enable further experimental study of the properties and behavior of fundamental and hadron-scale particles. In order for this project to proceed optimally, considerable efforts are required to design detectors and other equipment suitable to the physics specifications set for the Electron-Ion Collider. One of many planned detectors for this facility is an electromagnetic barrel calorimeter called the Barrel Imaging Calorimeter. The design for this calorimeter will be unique worldwide, as it will involve energy-summing layers alongside imaging layers, allowing for state-of-the-art particle detection. In order to construct this detector, many design elements require finalization. This work’s main focus is on energy resolution studies for the tracking layers of the Barrel Imaging Calorimeter to verify that charged particles operating on a kinematic range comparable to that of the Electron-Ion Collier will be detected with sufficient resolution. Components of this CERN ROOT based analysis include gain calibrations, quality checks on data, and analysis of energy leakage in the test calorimeter. Data collection took place on a stand-in singular calorimeter segment at Jefferson Lab.

Speaker: Maggie Kerr (Massachusetts Institute of Technology)

4:30 PM Early Career Panel Discussion

Sat, July 12

8:00 AM → 12:15 PM Saturday Morning

8:00 AM High Crimes and Misdemeanors in Data Analysis

Speaker: Dr Douglas Higinbotham (Jefferson Lab)

9:00 AM Gravitational form factors in holographic QCD

Gravitational form factors (GFFs) offer profound insights into the spatial distribution of the strong force within hadrons and are intimately linked to the trace anomaly -- a key feature of QCD that underpins the proton’s finite mass and the pion’s near-zero mass. These fundamental observables will be accessible with high precision in upcoming electron-ion collider experiments. On the theoretical front, holographic QCD provides a powerful semiclassical framework for modeling hadron structure, offering intuitive and physically compelling descriptions. However, extracting the GFF D-term -- crucial for understanding mechanical properties such as internal pressure and shear forces -- has remained a persistent challenge in holographic approaches. This difficulty arises because the gravitational fluctuation in five-dimensional anti-de Sitter space must satisfy transversality and tracelessness conditions.

In this talk, I present a novel approach that overcomes this limitation by exploiting the deep correspondence between light-front QCD and semiclassical gravity in a warped five-dimensional spacetime. The method leverages light-front holography, which elegantly bridges the perturbative regime at large momentum transfers ($Q^2$) with the nonperturbative domain at low $Q^2$, where interactions with scalar and tensor glueballs dominate. As concrete applications, we compute the gravitational form factors of both the pion and the nucleon, and compare our results with recent lattice QCD simulations, demonstrating promising agreement.

Speaker: Yang Li (University of Science and Technology of China)

9:20 AM Affinity potentiality in EIC data exploration*

AFFINITY is a numerical tool that allows experimental data to be connected to the corresponding theoretical framework: collinear or TMD factorization, as appropriate. Recently the affinity tool has been modified and upgraded in such a way that event-by-event analyses can be performed, improving its reliability and precision. I will present our most recent results for JLab12 phase space coverage and the projections for JLab22. Probing the same approach on EIC data might bring us more information about the “Matching region” where there is no clear separation between Collinear and TMD factorization. EIC phase space will also be analyzed.

Speaker: Tetiana Yushkevych (University of Turin, Physics Department)

9:40 AM Pushing Boundaries in Accelerator Physics: A Decade of R&D at Jefferson Lab

During a ten-year tenure that began with a summer internship, and comes to an end as scientific staff, I had the opportunity to contribute directly and indirectly to nearly every aspect of the development of photoguns and Wien spin rotators at Jefferson Lab: from GaAs photocathode activation to high-voltage testing and beam delivery, and from conceptual sketches to a patented design. This talk presents a summary of the technical challenges, project milestones, and fortunate breakthroughs that, combined with the guidance of dedicated, empathetic mentors, and the support of a diverse and collaborative scientific team, culminated in the delivery of the 200 keV polarized photoelectron gun and Wien spin rotators developed for the 12 GeV Continuous Electron Beam Accelerator Facility energy upgrade.

Speaker: Gabriel Palacios Serrano (Thomas Jefferson National Accelerator Facility)

10:00 AM Coffee Break

10:20 AM Evaporative Coating R&D at SBU for EIC PID*

An overview of electron beam deposition techniques and experiments conducted at Stony Brook University for PID Detectors. Namely, an emphasis on pfRICH (proximity focusing RICH detector) and the coating of high reflectivity mirrors for ePIC, as well as a novel R&D effort to redefine multi-wire drift chambers (MWDC) technology using coated carbon fiber wires.

Speaker: Preet Mann (Stony Brook University (CFNS))

10:40 AM Mirror testing for pfRICH detector

The Proximity Focusing RICH (pfRICH) detector is a key sub-system fur particle identification (PID) at ePIC. It sits about 1.2 m from the interaction point in the electron-going direction, covering $-3.5 < \eta < -1.5$ in pseudorapidity and full $2\pi$ in azimuth. It will be primarily used for identification of scattered electrons from deep inelastic scattering (DIS) events, as well as for identification of charged hadrons, namely $\pi^\pm$, $K^\pm$, with $p$ and $\bar{p}$, which will be vital e.g. for semi-inclusive DIS analyses. In order to achieve the aforementioned acceptance in pseudorapidity, the pfRICH will be equipped with conical mirrors that will reflect the Cherenkov light, that would otherwise escape its volume, back onto the sensor plane. For optimal performance, these mirrors need to have reflectivity at least 90% in wavelength range from 200 nm to 600 nm. In this presentation, I will summarize current efforts in the pfRICH mirror reflectivity measurement at BNL which is a key part of the mirror development process.

Speaker: Jan Vanek (Brookhaven National Laboratory)

11:00 AM TMDPDFs extractions with DNNs

Transverse Momentum Dependent Parton Distribution Functions (TMDPDFs) provide crucial insights into the three-dimensional structure of hadrons and can be extracted from processes involving multiple kinematic scales, including Drell-Yan (DY), Semi-Inclusive Deep Inelastic Scattering (SIDIS), and $e^+e^-$ annihilation. Deep Neural Networks (DNNs) have emerged as powerful tools for information extraction and modeling based on data with multi-dimensional kinematics and offer new possibilities for TMDPDFs extractions. This talk will detail our flavor-dependent extraction of Sivers functions within the $SU(3)_{\text{flavor}}$ framework through fits to SIDIS data and projections to DY kinematics. I will also present preliminary results for unpolarized TMDPDFs.

Speaker: Ishara Fernando (University of Virginia)

11:20 AM Prototype Testing of AstroPix Sensors and System Integration for BIC at ePIC

AstroPix is a high-voltage CMOS (HV-CMOS) monolithic silicon sensor and a key component of the Barrel Imaging Calorimeter (BIC) for the ePIC experiment, alongside the lead/scintillating-fiber (Pb/SciFi) sampling calorimeter. Interleaved within the calorimeter layers, AstroPix provides fine-grained shower imaging, enabling critical performance capabilities such as electron/pion or gamma/pion separation.
Based on the ATLASPix design, the AstroPix sensor was originally developed for NASA’s All-sky Medium-Energy Gamma-ray Observatory eXplorer (AMEGO-X) and has been tested in parallel at Argonne National Laboratory (ANL) in a bench environment. Its precise energy and position resolutions, low noise, low power consumption, and minimal dead material satisfy the requirements of both the space mission and the BIC system in the future Electron-Ion Collider (EIC) detector.
As part of the ongoing detector R&D effort, I have been testing various AstroPix version 3 configurations: the single chip, a quad-chip assembly, a three-layer stack of quad chips, and a 9-chip PCB module that represents the smallest prototype unit of the imaging layer.
This presentation will highlight recent performance test results from these AstroPix configurations and provide an update on system-level testing status and synchronization efforts between AstroPix and the Pb/SciFi calorimeter at ANL.

Speaker: Bobae Kim (Argonne National Laboratory)

11:40 AM The dRICH Interaction Tagger – An Idea to Reduce the dRICH Data Throughput

The ePIC dual-radiator Ring Imaging CHerenkov detector (dRICH) will serve as a particle identification (PID) system in the hadronic endcap of the experiment. Its main goals are to distinguish charged hadrons in the momentum range from 3 to 50 GeV/c and to assist in electron identification up to 15 GeV/c. To achieve these goals, it exploits Cherenkov radiation produced in two radiators—an aerogel and a C₂F₆ gas volume—to measure the particle velocity. Combined with momentum information, this allows the reconstruction of the particle mass and, consequently, its identification.

Cherenkov photons will be detected by approximately 320,000 silicon photomultipliers (SiPMs) operating at the single-photon threshold. Despite a dedicated cooling system maintaining the temperature at -40 °C, the dark count rate (DCR) remains non-negligible. Furthermore, radiation damage accumulated over the ePIC operational lifetime is expected to increase the DCR to as much as 300 kHz per channel, corresponding to a total of 96 GHz. This is significantly higher than the expected rate of physically relevant interactions, estimated to be around 500 kHz. The dRICH could reach a total data throughput of up to 1.4 Tbps, mostly composed of background signals.

Two strategies are under development to reduce this data volume: a machine learning-based data reduction algorithm implemented on the DAQ FPGA, and an Interaction Tagger, which flags the occurrence of actual particle passages through the detector. In this talk, I will present the latter approach. I will discuss how the addition of two layers of scintillating fibers (SciFi) in front of the dRICH can significantly reduce the data throughput by providing a fast interaction signal. The talk will cover the concept, the simulation studies, and the preliminary tests performed to assess the feasibility of the Interaction Tagger.

Speaker: Simone Vallarino

1:00 PM → 5:00 PM Saturday Afternoon

1:00 PM Strong Coupling Measurement at the Future EIC*

The Electron-Ion Collider (EIC) will be the next major facility to study the smallest building blocks of matter and their interactions. This facility will collide spin-polarized electrons and nuclei, thus providing new opportunities to study the spin structure of nucleons. Given the expected performance of the facility, we simulated polarized electron-proton and electron-Helium 3 collisions using double-tagging, an innovative approach resulting in improved precision for neutron data. These projections provide the expected precision for measurements of the inclusive spin-structure functions of the proton and neutron. To increase the kinematic coverage and improve the accuracy of this study, we use a parameterization from existing experimental data. We integrate over quark momentum to obtain the proton and neutron spin structure function moments with their respective experimental uncertainties. Then employ a Monte-Carlo approach to determine the systematic uncertainty. Forming Bjorken sums from the moments, we fit the sums using a series representation of the Bjorken Sum Rule to obtain the value and uncertainty for the coupling of the strong nuclear force, α_s. Improved experimental methodologies paired with improved theoretical calculations permit the extraction of α_s with a relative precision of 1.3 percent, competitive with the most precise global analyses on world data for Deep-Inelastic Scattering.

Speaker: Darren Upton (Old Dominion University)

1:20 PM Precision Event Shape Analysis for DIS at HERA and EIC

This talk presents a cutting-edge analysis of the 1-jettiness event shape in deep inelastic scattering (DIS), utilizing Soft-Collinear Effective Theory. The study achieves high precision at N$^3$LL + O($\alpha_s^2$) accuracy, incorporating full fixed-order matching along with nonperturbative corrections. It represents one of the most accurate event-shape predictions in DIS to date and supports precise determinations of the strong coupling constant $\alpha_s$ and the universal nonperturbative parameter $\Omega_1$. The approach is directly relevant to data from both HERA and the future Electron-Ion Collider (EIC).

Speaker: June-Haak Ee (Los Alamos National Laboratory)

1:40 PM Polarization effects on 3D Nucleon structure

The structure of quarks inside a nucleon is affected by its polarization. This effects can be investigated throught the extraction of the TMD Sivers distribution from DY, SIDIS measurements.

Speaker: Filippo Delcarro (University of Pavia)

2:00 PM Exploring nuclear structure and its fluctuations using Sartre

Sartre is an event-generator based on the color dipole model of deep inelastic scattering (DIS). It employs the Good-Walker mechanism to simulate event-by-event fluctuations, which are key to describing the incoherent cross section, where the target breaks up following the interaction. Sartre has been extensively used to describe photon–nucleus interactions at the Electron-Ion Collider (EIC), as well as in ultra-peripheral collisions (UPCs) at the LHC and RHIC. Accurately reproducing recent LHC and RHIC data requires incorporating different length scale fluctuations. We demonstrate that introducing sub-nucleonic fluctuations in the nuclear structure is sufficient to describe most current measurements. However, at larger momentum transfers, where the gluon distribution is probed at finer spatial resolution, the present level of model complexity may not fully capture the observed features.

We also present a novel approach to probe the gluonic structure of the pion through tagged-measurements. Future experiments at the LHC and especially EIC, offer exciting opportunities to investigate the nuclear structure, its fluctuations and correlations.

Speaker: Arjun Kumar (CFNS, Stony Brook University)

2:20 PM QCD First Inverse Problem using Maximum Likelihood Method from Exclusive Experiments.

Two processes contribute to the $ep \rightarrow e' p' \gamma'$ reaction: Deeply Virtual Compton Scattering, where the photon is produced at the proton vertex, and the Bether-Heitler process, where the photon is radiated from the electron. A major hurdle in the extraction of Generalised Form Factors (GPDs) from experiment arises due to the presence of two Inverse problems: a first one for the extraction of the various Compton form factors (CFFs) from data, and a subsequent one concerning the extraction of GPDs from CFFs. Here we aim to address the first inverse problem. A maximum likelihood analysis for deeply virtual exclusive photoproduction off a proton target is presented based on MCMC sampling. We provide and use a method which derives a joint likelihood of the CFFs, parameterizing the deeply virtual Compton scattering amplitude in QCD, for each observed combination of the kinematic variables defining the reaction. The likelihood for the unpolarized cross section at twist-two fully constrains only three of the CFFs.

Speaker: Dr Saraswati Pandey (University of Virginia)

2:40 PM Coffee Break

3:00 PM Feasibility Study of Measuring Λ0->nπ0 Using a High-Granularity Zero-Degree Calorimeter at the Future Electron-Ion Collider

Key measurements at the future Electron-Ion Collider (EIC), including first-of-their-kind studies of kaon structure, require the detection of $\Lambda^0$ at forward angles. We present a feasibility study of $\Lambda^0 \rightarrow n\pi^0$ measurements using a high-granularity Zero Degree Calorimeter to be located about 35 m from the interaction point. We introduce a method to address the unprecedented challenge of identifying $\Lambda^0$s with energy $O(100)$ GeV that produce displaced vertices of $O(10)$ m. In addition, we present a reconstruction approach using graph neural networks. We find that the energy and angle resolution for $\Lambda^0$ is similar to that for neutrons, both of which meet the requirements outlined in the EIC Yellow Report. Furthermore, we estimate performance for measuring the neutron’s direction in the $\Lambda^0$ rest frame, which reflects the $\Lambda^0$ spin polarization. We estimate that the neutral-decay channel $\Lambda^0 \rightarrow n\pi^0$ will greatly extend the measurable energy range for the charged-decay channel $\Lambda^0 \rightarrow p\pi^-$, which is limited by the location of small-angle trackers and the accelerator magnets. This work paves the way for EIC studies of kaon structure and spin phenomena.

Speaker: Sebouh Paul (UC Riverside)

3:20 PM Beam Test Results for a SiPM-on-Tile ZDC Prototype

The SiPM-on-Tile technology was adopted by several calorimetry systems
of ePIC, including the forward and backward hadronic calorimeters (1.5 < η <
3.0), the high granularity insert (3.0 < η < 4.0) and the Zero-Degree Calorime-
ter (ZDC, η > 6.0). We developed a ZDC prototype to validate the design and
evaluate its performance with low-energy particles. Beam tests conducted at
Jefferson Lab (JLab) and ongoing evaluations at the NASA Space Radiation
Laboratory (NSRL) have characterized the prototype’s capabilities. This pre-
sentation will highlight the JLab beam test results and provide an update on
the current status of testing at NSRL.

Speaker: Weibin Zhang (UC Riverside)

3:40 PM Testing HRPPDs at Yale for the pfRICH*

The High Rate Picosecond Photodetector (HRPPD) is a vital part of the Proximity-Focusing Ring Imaging Cherenkov Detector (pfRICH) that will be included in the ePIC experiment. HRPPDs will be used to measure the photons emitted in the Cherenkov cone emitted by particles traveling thorough the aerogel wall of the pfRICH. Reconstructing the ring of photons will allow for particle species identification based on the radius from the radiating particle. Therefore, understanding performance properties of HRPPDs, such as the uniformity of the quantum efficiency and gain as well as the dark rate, are important for the optimal operation of the detector. In this talk, I will present the current status of a test stand at Yale to measure these properties of HRPPDs, alongside preliminary results.

Speaker: Andrew Tamis

4:00 PM EIC Detector-2 Study: Diffraction and Tagging Physics Program and Muon Identification

The Electron-Ion Collider (EIC) is a next-generation facility designed to investigate the gluonic structure of nucleons and nuclei via Deep Inelastic Scattering. The EIC physics program, outlined in the EIC White Paper and Yellow Report, led to the development of the general-purpose ePIC detector. To enhance the scientific reach of the EIC, the community supports the addition of a second general-purpose detector with complementary technologies and capabilities. This second detector would enable crucial cross-checks, improved control of systematics, and specific measurements that are less well addressed in ePIC. The second interaction region offers enhanced forward acceptance at low transverse momentum and a secondary beam focus, making it particularly suitable for exclusive, tagging, and diffractive physics studies. In this talk, I present a simulation study focused on incoherent diffractive vector meson production, $e + Pb \rightarrow e' + J/\Psi + X$, using a proposed layout for the second detector. I also discuss muon identification in ePIC and its relevance for developing complementary capabilities in Detector-2.

Speaker: Jihee Kim (Brookhaven National Laboratory)

4:20 PM Why the EIC community should care about MPGD-based Transition Radiation Detectors*

TBA

Speaker: Lauren Kasper (Vanderbilt University (US))

Sun, July 13

9:00 AM → 9:30 PM Activity: Special Activity