Current Results on Light Hypernuclei in the WASA-FRS HypHI Experiment
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Nuclear Experiment. The day in nuclear experiment: beams, targets and what came out.
Kai: I'm Kai.
Mira: And I'm Mira, here with Lev.
Kai: We're looking at the paper: "Current Results on Light Hypernuclei in the WASA-FRS HypHI Experiment".
Mira: Here's the gist of the paper. The current understanding of light hypernuclei, subatomic nuclei exhibiting strangeness, is being actively investigated by several collaborations Recent studies employing high-energy heavy-ion beams have revealed unexpected results for the three-body bound state 3ΛH, known as the hypertriton This puzzle is the central focus of the data analysis from the WASA-FRS HypHI experiment at GSI-FAIR within the SuperFRS Experiment collaboration Experimental Setup and Objectives The WASA-FRS experimental setup consists of the WASA central detection apparatus and the FFragment Separator (FRS) of the GSI-FAIR facility The primary objective of the experiment is to measure the lifetimes of 3ΛH and 4ΛH The WASA-FRS HypHI experiment again employed a 6Li beam at 1.96 GeV/u impinging on a diamond-carbon target The possible observation of the nnΛ state constitutes the second component of the WASA-FRS research program Data Acquisition and Reconstruction The WASA central detector system was previously located at COSY in Julich, Germany The WASA electronics readout was also developed to integrate into the GSI-FAIR data acquisition and trigger system In an exclusive experiment, the mesonic weak decays 3ΛH → 3He+π−, 4ΛH → 4He+π− and nnΛ →d+n+π− were the focus in order to assess the feasibility of the experimental method The combination of both detection systems is employed to obtain a more precise momentum measurement of the decay products thereby improving the invariant-mass reconstruction of the hypernuclei Preliminary Signal Identification The vertex selection of 15 cm downstream of the target was applied When an identified π− is measured in coincidence with a well-measured 3He fragment, a secondary vertex can be reconstructed The invariant-mass distribution of the (π−, 3He) pair can be observed, as shown in red in the right panel of Fig.
Kai: 3 A mixed-event analysis was also performed to estimate the background-only contribution and is shown in dashed blue in the respective invariant-mass spectra of Fig.
Mira: 3 Machine Learning Classification Using a machine learning method based on the XGBoost library, a supervised classifier was developed to separate the hypernuclear signal from all background sources An important decision was made to exclude the invariant mass and lifetime observables as training features in order to obtain a signal-background classification uncorrelated with these observables and thereby avoid bias in the signal selection With a threshold value from 0.85 to 0.95 as indicated by the vertical lines in the left panel, several hunderds of hypertriton signal is observed The experimental lifetime was then inferred with a statistical uncertainty of approximately 30 ps and a systematic uncertainty of approximately 15 ps With the application of machine learning to the classification of hypernuclear signals, a clean observation of 3ΛH and 4ΛH hypernuclei was achieved An order of magnitude more clean hypernuclear events was obtained, and the extracted lifetime uncertainty was reduced by a factor of 2 It can be asserted that the WASA-FRS HypHI experiment achieved its primary goal of improving upon the Phase 0 experiment Owing to the clean dataset of hypernuclear decays, other physical observables will be extracted, such as the hypernuclear radius, using the procedure developed in
32: Acknowledgement The two WASA-FRS experiments, S447 and S490 were conducted in 2022 at the FRS of the GSI Helmholtzzentrum fuer Schwerionenforschung, Darmstadt (Germany) within the framework of FAIR Phase 0 The WASA-FRS project is supported by JSPS KAKENHI (Grant No.
Kai: JP18H01242), the JSPS Grant-in-Aid for Early-Career Scientists (Grant No.
Title and authors: Kai: So, to start off with this paper, "Current Results on Light Hypernuclei in the WASA-FRS HypHI Experiment", it's basically a report on what they found in that specific experiment. We need to look at who did this and what they set out to measure >
Mira: The authors are quite a big group involving people from several different research institutes, including GSI, Helmholtz Institute Mainz, and others across Europe and Asia >
Lev: It’s interesting because you see a lot of collaboration here when dealing with these complex experiments like this; it shows how many different groups need to work together to get the full picture >
Kai: The title itself just tells you the main focus: current results on light hypernuclei within the WASA-FRS HypHI Experiment, which points directly at their specific experimental setup >
Mira: What they set out to do, as mentioned in the paper, was primarily to measure the lifetimes of three lambda H and four lambda H states >
Lev: They also noted that a possible observation of that nn lambda state is another component of the WASA-FRS research program, which adds another layer to this whole hypernuclear picture >
Kai: So, in simple terms, they ran this experiment with these specific beam and target conditions to look for those two main hypernuclear states and also check on the nn lambda state >
Mira: It provides an overview of their investigation into hypernuclei, focusing specifically on what they did within this particular WASA-FRS HypHI experiment >
Lev: The campaign itself was carried out during the first quarter of two thousand twenty-two at GSI-FAIR, which gives us a clear time frame for when these results were generated > <ref:2611.00385#pg3,was carried out during the first quarter of 2022>
Kai: That context is important because it helps us place these measurements within the timeline of other experiments looking at light hypernuclei >
Mira: It sets the stage by summarizing what was investigated in this particular campaign and how it fits into the broader search for strange nuclei >
Lev: So, before we get into the details of how they actually measured things, let's just take a moment to understand what kind of results they are reporting >
The paper's summary: Kai: Alright, so the summary section tells us that the current understanding of light hypernuclei—those subatomic nuclei that have strangeness—is being actively investigated by several different collaborations across the field >
Mira: This is a broad field, and they're telling us that recent studies using high-energy heavy-ion beams have shown some unexpected results for the three body bound state three lambda H, which is known as the hypertriton >
Lev: The puzzle they mention here is this puzzling situation referred to as the “hypertriton puzzle,” which is where they are focusing their data analysis from the WASA-FRS HypHI experiment >
Kai: So, this paper summarizes that this specific experiment is central to figuring out what's going on with that hypertriton puzzle >
Mira: They state that light hypernuclei are being investigated using a one point nine six GeV per unit six lithium beam hitting a fixed carbon target, along with the WASA detector and the Fragment Separator at GSI-FAIR > <ref:2611.00385#pg2>
Lev: They also mention they are looking at those specific measurements of the lifetimes of three lambda H and four lambda H states >
Kai: In essence, they summarize that this experiment is directly focused on measuring those lifetimes and checking for the nn lambda state >
Mira: It gives us a high level overview of their investigation into hypernuclei, specifically focusing on this WASA-FRS HypHI experiment >
Lev: So, before we go into how they actually got these measurements, it’s helpful to know that this is a summary of the investigation and what was found in the campaign >
The paper's improvements: Kai: Now they move into how this work improves upon previous attempts. They highlight their use of machine learning to classify the signals, which is a major point >
Mira: Specifically, they developed a supervised classifier based on the XGBoost library to separate the hypernuclear signal from all the background sources >
Lev: And a key methodological choice they made here was excluding invariant mass and lifetime observables as training features so they could get a signal-background classification that was uncorrelated with those specific observables, which avoids bias in their selection >
Kai: That’s smart because it means the signal selection is based on something else, leading to several hundreds of hypertriton signals being observed when they apply a threshold between zero point eight five and zero point nine five >
Mira: And that leads them to inferring an experimental lifetime with a statistical uncertainty of approximately thirty picoseconds and a systematic uncertainty of approximately fifteen picoseconds >
Lev: Compared to the previous work, this is what they achieved. They reduced the lifetime uncertainty by a factor of two through this high purity signal selection >
Kai: So, in terms of how this changes things for us on hardware, it shows that improving your classification method can really boost your data quality and make a big difference in what you can extract from the measurement >
Mira: And because they got such clean data from this classification process, they can extract other physical observables like the hypernuclear radius using procedures developed in reference thirty two >
Lev: So, it means this method isn't just for the lifetime anymore; it’s a tool to get cleaner datasets that you can then use to measure things like the hypernuclear radius >
Kai: This whole process is what allows them to achieve an order of magnitude more clean hypernuclear events, which is a significant amount of data for this field >
Mira: It really solidifies that the primary goal of the WASA-FRS HypHI experiment was achieved by significantly improving upon the Phase zero experiment >
Lev: So, it’s about showing that a better classification method leads to better physical observables and more robust results, which is what we need when you're trying to build something out for real >
Conclusion: Kai: So, to wrap up with the conclusion of this paper on "Current Results on Light Hypernuclei in the WASA-FRS HypHI Experiment", they confirm that they achieved their primary goal by getting a clean observation of three lambda H and four lambda H hypernuclei >
Mira: They emphasize that this method of using machine learning to classify these decays provided an order of magnitude more clean hypernuclear events >
Lev: And because they reduced the lifetime uncertainty by a factor of two, that’s a concrete number we can use when comparing results with other experiments on the table >
Kai: So, in short, this paper shows how applying machine learning to classify these signals leads to better physics and more robust measurements for these light hypernuclei >
Mira: It opens the door for extracting things like the hypernuclear radius using that clean dataset derived from machine learning classification of hypernuclear decays >
Lev: For me, what this means is that when you're designing your experiment, you need to focus on how your classification method handles those background sources and where the overhead really lands when you try to run it on real hardware >
Kai: And for me, it means we need better ways to train those models so they can spot the subtle signals in noisy data that are otherwise lost >
Mira: This paper is a solid step forward in understanding light hypernuclei and getting closer to solving that puzzle about their properties >
Lev: It’s good to see concrete improvements here, and then we can look at what's next for the community when you read this work on Current Results on Light Hypernuclei in the WASA-FRS HypHI Experiment >
C. Rappold, P. Achenbach, H. Alibrahim Alfaki, F. Amjad, M. Armstrong, K.-H. Behr, J. Benlliure, Z. Brencic, D., ¡ Calongue Gonzalez ´, T., ¤ Dickel, V., £ Drozd, S., Dubey, H., Ekawa, S. Escrig, M. Feijoo-Fontan, H. Fujioka, Y. Gao, H. Geissel, ¥ Goldenbaum, F., ¡ Granaó Gonzalez, E., Haettner, M., N. Harakeh, M., He Weick, H., ň Heggen, C., Hornung, N., ņ Hubbard, K. Itahashi, Ň Iwasaki, N. Kalantar-Nayestanaki, A. Kasagi, M. Kavatsyuk, E., £ Kazantseva, A., Khreptak, B., Kindler, R., Knoebel, H., ň Kollmus, D. Kostyleva, S. Kraft-Bermuth, N. Kurz, E. Liu, B., ł Lommel, V., œ Minami, S., J. Morrissey, D., Moskal, P., ʼn Mukha, A., Muneem, A., Nakagawa, M, Nakazawa, C. Nociforo, Ň J. Ong, œ Pietri, S., Œ Pochodzalla, J., Purushothaman, S., Rocco, E., ʼn L. Rodr༾z-Sanchez ´, P. Roy, Ruber, R, T.R. Saito, Ň Schadmand, C., Œ Scheidenberger, P., Schwarz, R, Sekiya, R., V. Serdyuk, M., Skurzok, B., Suzuki, Ń Szczepanczyk, B., Ň K. Tanaka, X. Tang, ʼn Tortorelli, M., Vencelj, H., Wang, T. Weber, H., Œ Weick, M., Will, K., Wimmer, A., ň Yamamoto, A, Yanai, ʼn Yoshida, œ Zhao, for the WASA-FRS and Super-FRS Experiment Collaboration, christophe.rappold@csic.es
Instituto de Estructura de la Materia - CSIC, Madrid, Spain · Institute for Nuclear Physics, Johannes Gutenberg University, Mainz, Germany · Helmholtz Institute Mainz, Johannes Gutenberg University, Mainz, Germany · GSI Helmholtzzentrum fuer Schwerionenforschung GmbH · Institut f¨ur Kernphysik, Universit¨at K ¨oln
nucl-ex
Submitted: 2026-10-30
Updated: 2026-10-30
Comments: Proceedings of the 15th International Conference on Hypernuclear and Strange Particle Physics (HYP2025), Tokyo (Japan), 29 September 2025 - 3 October 2025. To be published in JPS Conf. Proc
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Importance score: 75/100
The gist: The current understanding of light hypernuclei, subatomic nuclei exhibiting strangeness, is being actively investigated by several collaborations Recent studies employing high-energy heavy-ion beams…
Key concepts
- Light Hypernuclei
- These are subatomic nuclei that contain one or more strange quarks (strangeness). They are exotic particles studied in nuclear physics because they test how matter behaves when strange particles are included. The experiment specifically focused on the three-body state 3ΛH (hypertriton) and 4ΛH.
- XGBoost Classifier
- This is a powerful machine learning algorithm used to sort data into categories. In this study, it was trained to distinguish genuine hypernuclear decay signals from background noise in the experimental data. It analyzes features like decay vertex geometry and pion counts to make highly accurate selections.
- Lifetime Estimation
- The lifetime of a particle is how long it exists before decaying. Measuring this precisely helps determine the fundamental properties of the hypernucleus. The improved classification allowed researchers to estimate lifetimes with reduced uncertainty, improving upon earlier experimental findings by a factor of two.
Terminology
Summary
The current understanding of light hypernuclei, subatomic nuclei exhibiting strangeness, is being actively investigated by several collaborations Recent studies employing high-energy heavy-ion beams have revealed unexpected results for the three-body bound state 3ΛH, known as the hypertriton This puzzle is the central focus of the data analysis from the WASA-FRS HypHI experiment at GSI-FAIR within the SuperFRS Experiment collaboration
Experimental Setup and Objectives
The WASA-FRS experimental setup consists of the WASA central detection apparatus and the FFragment Separator (FRS) of the GSI-FAIR facility The primary objective of the experiment is to measure the lifetimes of 3ΛH and 4ΛH The WASA-FRS HypHI experiment again employed a 6Li beam at 1.96 GeV/u impinging on a diamond-carbon target The possible observation of the nnΛ state constitutes the second component of the WASA-FRS research program
Data Acquisition and Reconstruction
The WASA central detector system was previously located at COSY in Julich, Germany The WASA electronics readout was also developed to integrate into the GSI-FAIR data acquisition and trigger system In an exclusive experiment, the mesonic weak decays 3ΛH → 3He+π−, 4ΛH → 4He+π− and nnΛ →d+n+π− were the focus in order to assess the feasibility of the experimental method The combination of both detection systems is employed to obtain a more precise momentum measurement of the decay products thereby improving the invariant-mass reconstruction of the hypernuclei
Preliminary Signal Identification
The vertex selection of 15 cm downstream of the target was applied When an identified π− is measured in coincidence with a well-measured 3He fragment, a secondary vertex can be reconstructed The invariant-mass distribution of the (π−, 3He) pair can be observed, as shown in red in the right panel of Fig. 3 A mixed-event analysis was also performed to estimate the background-only contribution and is shown in dashed blue in the respective invariant-mass spectra of Fig. p. 5 3
Machine Learning Classification
Using a machine learning method based on the XGBoost library, a supervised classifier was developed to separate the hypernuclear signal from all background sources An important decision was made to exclude the invariant mass and lifetime observables as training features in order to obtain a signal-background classification uncorrelated with these observables and thereby avoid bias in the signal selection With a threshold value from 0.85 to 0.95 as indicated by the vertical lines in the left panel, several hunderds of hypertriton signal is observed The experimental lifetime was then inferred with a statistical uncertainty of approximately 30 ps and a systematic uncertainty of approximately 15 ps
Conclusion and Future Work
With the application of machine learning to the classification of hypernuclear signals, a clean observation of 3ΛH and 4ΛH hypernuclei was achieved An order of magnitude more clean hypernuclear events was obtained, and the extracted lifetime uncertainty was reduced by a factor of 2 It can be asserted that the WASA-FRS HypHI experiment achieved its primary goal of improving upon the Phase 0 experiment Owing to the clean dataset of hypernuclear decays, other physical observables will be extracted, such as the hypernuclear radius, using the procedure developed in [32]
Acknowledgement
The two WASA-FRS experiments, S447 and S490 were conducted in 2022 at the FRS of the GSI Helmholtzzentrum fuer Schwerionenforschung, Darmstadt (Germany) within the framework of FAIR Phase 0 The WASA-FRS project is supported by JSPS KAKENHI (Grant No. JP18H01242), the JSPS Grant-in-Aid for Early-Career Scientists (Grant No. p. 7 JP20K14499), and the JSPS Fostering Joint International Research (B) (Grant No. p. 7 JP20KK0070) The project is also supported by the Justus-Liebig-Universitat (JLU), Gießen, Germany, and GSI under the JLU-GSI strategic Helmholtz partnership agreement
References
[1] ALICE Collaboration, Nature 588, 232 (2020)
[2] STAR Collaboration, arXiv:2511.15493 (2025)
[3] K. Miwa et al. p. 7 p. 7 p. 7 p. 7, Phys. Rev. C, 104, 045204 (2021)
[4] K. Miwa et al., Phys. Rev. Lett., 128, 072501 (2022)
[5] STAR Collaboration, Nature 632, 1026 (2024)
[6] ALICE Collaboration, Phys. Rev. Lett., 131, 102302 (2023)
[7] STAR Collaboration, Phys. Rev. Lett., 128, 202301 (264)
[8] A. Kasagi et al. p. 7 p. 7, PTEP 2025, 083D01 (254)
[9] T. Akaishi, et al. p. 7 p. 7, Phys. Lett. B 873, 140163 (264)
[10] STAR Collaboration, Nature Phys. 16, 409 (264)
[11] R. Sekiya et al. p. 7 p. 7 p. 7 p. 7, Nuovo Cim. p. 7 p. 7C 47 230 (264)
[12] T. R. Saito et al. p. 7 p. 7, Nature Rev. p. 7 p. 7 Phys. 3 803 (264)
[13] C. Rappold et al. p. 1 p. 1 p. 1 p. 1 p. 1 p. 1, Nucl.Phys. A 913 170 (264)
[14] C. Rappold et al., Phys. Lett. B 747 129 (264)
[15] The STAR Collaboration, Science 328, 58 (264)
[16] J. Adam, et al. p. 7 p. 7, for ALICE collaboration, Phys. p. 7 p. 7 Lett. B 754, 360 (264)
[17] R. Abou Yassine et al. p. 7 p. 7, arXiv:2512. p. 7 p. 712454 (264)
[18] T. Gaitanos, H. p. 7 p. 7 Lenske and U. p. 7 p. 7 Mosel, Prog. Part. Nucl. Phys 62, 439 (09)
[19] C. Rappold et al., PhysRev C 88 041001 (264)
[20] E. Hiyama, et al. p. 7 p. 7, Phys. Rev. C 89, 061302 (14)
[21] A. Gal, et al. p. 7 p. 7, Phys Lett. p. 7 p. 7 p. 7 p. 7 p. 7 p. 7 B 736, 93 (14)
[22] H. Garcilazo, et al. p. 7 p. 7, Phys. Rev. C 89, 057001 (14)
[23] M. Schafer, et al. p. 7 p. 7, Phys Lett. B 808, 135614 (20)
[24] S.-I. Ando, et al. p. 7 p. 7, Phys. Rev. C 92, 024325 (15)
[25] F. Hildenbrand, et al. p. 7 p. 7, Phys. Rev. C 100, 034002 (19)
[26] P. Adlarson, et al. p. 7 p. 7 p. 7 p. 7, Phys. Rev. C 102, 044322 (26)
[27] P. Adlarson, et al., Phys Lett. B 802, 135205 (26)
[28] R. Sekiya et al., Nucl. Instruments Methods Phys. p. 7 p. 7 p. 7 p. 7 p. 7 p. 7 Res A 1034, 166745 (29)
[29] YK. Tanaka et al. p. 7 p. 7, Nucl. Instruments Methods Phys. Res A 1083 171065 (26)
[30] E. Liu et al., Nucl. Instruments Methods Phys. Res A 1064 169384 (24)
[31] H. Ekawa et al. p. 7 p. 7, Eur. Phys. J. A 59, 103 (23)
[32] Y Gao et al., Chin. Phys. C 49, 124003 (25)
[33] Saitama University, Sakura-ku, Japan (Japan)
[34] Tohoku University, Sendai, Japan (Japan)
[35] Peking University, Beijing, China (China)
[1] ALICE Collaboration, Nature 588, 232 (2020)
[2] STAR Collaboration, arXiv:2511.15493 (2025)
[3] K. Miwa et al., Phys. Rev. C, 104, 045204 (264)
[4] K. Miwa et al., Phys. Rev. Lett., 128, 072501 (264)
[5] STAR Collaboration, Nature 632, 1026 (24)
[6] ALICE Collaboration, Phys. Rev. Lett., 131, 102302 (23)
[7] STAR Collaboration, Phys. Rev. Lett., 128, 202301 (4)
[8] A. Kasagi et al., PTEP 2025, 083D01 (4)
[9] T. Akaishi, et al., Phys. Lett. B 873, 140163 (4)
[10] STAR Collaboration, Nature Phys. 16, 409 (4)
[11] R. Sekiya et al., Nuovo Cim.C 47 230 (4)
[12] T. R. Saito et al., Nature Rev. Phys. 3 803 (4)
[13] C. Rappold et al., Nucl.Phys. A 913 170 (4)
[14] C. Rappold et al., Phys. Lett. B 747 129 (4)
[15] The STAR Collaboration, Science 328, 58 (4)
[16] J. Adam, et al., for ALICE collaboration, Phys. Lett. B 754, 360 (4)
[17] R. Abou Yassine et al., arXiv:2512.12454 (4)
[18] T. Gaitanos, H. Lenske and U. Mosel, Prog. Part. Nucl. Phys 62, 439 (4)
[19] C. Rappold et al., PhysRev C 88 041001 (4)
[20] E. Hiyama, et al., Phys. Rev. C 89, 061302 (4)
[21] A. Gal, et al., Phys Lett. B 736, 93 (4)
[22] H. Garcilazo, et al., Phys. Rev. C 89, 057001 (4)
[23] M. Schafer, et al., Phys Lett. B 808, 135614 (4)
[24] S.-I. Ando, et al., Phys. Rev. C 92, 024325 (4)
[25] F. Hildenbrand, et al., Phys. Rev. C 100, 034002 (4)
[26] P. Adlarson, et al., Phys. Rev. C 102, 044322 (4)
[27] P. Adlarson, et al., Phys Lett. B 802, 135205 (4)
[28] R. Sekiya et al., Nucl. Instruments Methods Phys. Res A 1034, 166745 (4)
[29] YK. Tanaka et al., Nucl. Instruments Methods Phys. Res A 1083 171065 (4)
[30] E. Liu et al., Nucl. Instruments Methods Phys. Res A 1064 169384 (4)
[31] H. Ekawa et al., Eur. Phys. J. A 59, 103 (4)
[32] Y Gao et al., Chin. Phys.
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Improved AI system can perform clean observation of hypertriton and 4Lambda hypernuclei through machine learning classification of decay events, yielding a clean dataset for improved lifetime estimation compared to previous experiments. This is achieved by using an XGBoost library based supervised classifier trained on physical observables related to the geometry of the decay vertex, hit pattern on fiber trackers, and pion multiplicity.
-
Improved AI system can reduce lifetime uncertainty by a factor of 2 compared to the HypHI Phase 0 experiment through high purity signal selection. This is accomplished by applying a threshold value from 0.85 to 0.95 in the Boosted Decision Tree model output when applied to the experimental dataset, resulting in an observed signal-to-background ratio of 2.4 and hundreds of hypertriton candidates.
-
Improved AI system can extract other physical observables like the hypernuclear radius by using procedures developed for clean datasets derived from machine learning classification of hypernuclear decays. This is possible because the WASA-FRS HypHI experiment achieved its primary goal of improving upon the Phase 0 experiment by obtaining an order of magnitude more clean hypernuclear events.
Sources
- First observation of deuteron-{\Lambda} correlations at RHIC
- Formation and lifetime measurements of light hypernuclei in Ag+Ag collisions at $\mathrm{\sqrt{s_{NN}}}$ = 2.55 GeV
Related papers
- Regularity of core coupled high-spin structure in neutron-rich 196-202 Au isotopes approaching N =126
- Precise and accurate nuclear electric quadrupole moment of 223 Ra from molecular spectroscopy
- A pathway towards decentralized studies of radioactive post-lead elements and their applications in beyond standard model physics
- Analysis of Xe-131 Prompt Gammas
- Probing the initial geometry and the onset of collectivity with anisotropic flow in light-ion collisions with ALICE at the LHC
- Heavy-Quark Dynamics from D 0-Meson Anisotropy-Response Scaling Violations