Correlations of $\\mu\\mu$, $e\\mu$, and $ee$ pairs in $p$+$p$ collisions\n at $\\sqrt{s}=200$ GeV and implications for $c\\bar{c}$ and $b\\bar{b}$\n production mechanisms

C. Aidala, Y. Akiba, Alfred, M., V. Andrieux, N. Apadula, Asano, H., B. Azmoun, Babintsev, V., A. Bagoly, N. S. Bandara, K. N. Barish, Bathe, S., A. Bazilevsky, M. Beaumier, R. Belmont, A. Berdnikov, Y. Berdnikov, D. Blau, Boer, M., J. Bok, M. L. Brooks, S. Chattopadhyay, Bumazhnov, V., S. Campbell, P. Camerini, R. Cervantes, W. C. Chang, M. Chiu, T. Chujo, J. Choi, Z. H. Citron, M. Connors, A. Mehta, M. Csanád, Tamás Ferenc Csörgő, T. W. Danley, M. Daugherity, D. d’Enterria, K. Dehmelt, A. Denisov, W. H. Smith, A. Durum, Dion, A., Dhruv Utpalkumar Dixit, Uliana Dmitrieva, A. Drees, Drees, K. A., J. M. Durham, A. Enokizono, En'yo, H., O. Evdokimov, O. Eyser, W. Fan, A. Hinzmann, D. E. Fields, Y. Assran, S. Fokin, Y. Goto, Franz, A., M. Große Perdekamp, C. Gal, Petr Gallus, P. Garg, H. Ge, A. Di Mattia, H. M. Gray, T. Gunji, H. Guragain, T. Hachiya, K. I. Hahn, H. Hamagaki, E. D. Hall, Hamilton, H. F., J. Hanks, B. Jayatilaka, X. He, M. Heffner, Hemmick, T. K., K. K. Hill, A. M. Hodges, B. Holzman, B. Hong, S. Choi, J. Huang, S. Huang, J. Murata, M. Inaba, D. IvanishchevarXiv 2018 · DOI · arXiv

Abstract

PHENIX has measured the azimuthal correlations of muon pairs from charm and bottom semi-leptonic decays in $p$+$p$ collisions at $\sqrt{s}=200$ GeV, using a novel analysis technique utilizing both unlike- and like-sign muon pairs to separate charm, bottom and Drell-Yan contributions. The dimuon measurements combined with the previous electron-muon and dielectron measurements span a wide range in rapidity, and are well described by PYTHIA Tune A. Through a Bayesian analysis based on PYTHIA Tune A, we show that leading order pair creation is the dominant ($76\%\pm^{14}_{19}\%$) contribution for $b\bar{b}$ production, whereas the data favor the scenario in which next-to-leading-order processes dominate $c\bar{c}$ production. The small contribution of next-to-leading-order processes in $b\bar{b}$ production at the collision energies of the Relativistic Heavy Ion Collider contrasts with the case at Large-Hadron-Collider energies, where next-to-leading-order processes are expected to dominate.

Introduction: The "How" of Heavy Quarks

In the realm of high-energy particle physics, heavy quarks—specifically charm ($c$) and bottom ($b$)—serve as precision probes for testing Perturbative Quantum Chromodynamics (pQCD). While counting the total number of these quarks (inclusive measurements) provides some insight, it often fails to resolve discrepancies between theoretical predictions and experimental data. The core of the issue lies in the production mechanism: are these quarks created in a simple, back-to-back interaction (Leading Order, or LO), or through more complex, multi-step processes like "gluon splitting" or "flavor excitation" (Next-to-Leading Order, or NLO)?

Methodology: Disentangling the Signal

The PHENIX Collaboration at the Relativistic Heavy Ion Collider (RHIC) addressed this by analyzing the azimuthal correlations—the angular "handshakes"—between pairs of leptons (electrons and muons) resulting from heavy quark decays. By examining $p+p$ collisions at $\sqrt{s} = 200$ GeV, the team combined new dimuon ($\mu\mu$) data with previous electron-muon ($e\mu$) and dielectron ($ee$) measurements to cover a wide range of rapidity.

A key innovation in this study was the use of like-sign muon pairs ($\mu^+\mu^+$ or $\mu^-\mu^-$). Unlike charm and Drell-Yan processes, which primarily produce unlike-sign pairs, bottom quarks can produce like-sign pairs through $B^0\bar{B}^0$ oscillations or specific decay chains. This allowed for a cleaner isolation of the bottom quark signal from the dominant backgrounds.

Azimuthal correlations of muon pairs from $c\bar{c}$ (top) and $b\bar{b}$ (bottom) decays. The data are compared with PYTHIA Tune A and POWHEG models, showing the decomposition into Pair Creation (PC), Flavor Excitation (FE), and Gluon Splitting (GS).

The researchers then employed Bayesian inference to determine the relative contributions of three primary production mechanisms:

  1. Pair Creation (PC): The LO process resulting in back-to-back quarks.
  2. Flavor Excitation (FE): An NLO process where a heavy quark from the parton distribution is "knocked out."
  3. Gluon Splitting (GS): An NLO process where a high-energy gluon splits into a $Q\bar{Q}$ pair.

Results: A Mass-Dependent Shift

The analysis revealed a striking reversal in the production hierarchy between charm and bottom quarks. For the heavier bottom quark, the energy required for NLO processes is less accessible at RHIC energies, making simple LO pair creation the dominant pathway. Specifically, PC accounts for $76\% \pm^{14}_{19}\%$ of $b\bar{b}$ production.

In contrast, charm production is "messier," dominated by Flavor Excitation. This explains why simple LO calculations often struggle to predict charm yields; the NLO "background" is, in fact, the primary driver of production at these energies. The study also set 95% confidence upper limits for the contribution of gluon splitting at 52% for charm and 31% for bottom.

Bayesian analysis results showing the extracted fractions for the three production mechanisms ($F_{PC}$, $F_{FE}$, $F_{GS}$) for charm (top row) and bottom (bottom row). The contrast between the PC-dominated bottom production and the FE-dominated charm production is evident.

Implications: RHIC vs. LHC

The study also compared the data against major event generators. Interestingly, PYTHIA Tune A (using a parton-shower approach) described the angular correlations more accurately than POWHEG, which predicted broader distributions than observed in the data.

These findings establish RHIC as a unique regime. At the much higher energies of the Large Hadron Collider (LHC), NLO processes are expected to dominate even for bottom quarks. By identifying LO dominance for $b\bar{b}$ at RHIC, this work provides a crucial baseline for future studies in proton-nucleus and nucleus-nucleus collisions, helping researchers understand how the Quark-Gluon Plasma (QGP) or cold nuclear matter might differently affect various production mechanisms.

Summary updated

Abstract

PHENIX has measured the azimuthal correlations of muon pairs from charm and\nbottom semi-leptonic decays in $p$+$p$ collisions at $\\sqrt{s}=200$ GeV, using\na novel analysis technique utilizing both unlike- and like-sign muon pairs to\nseparate charm, bottom and Drell-Yan contributions. The dimuon measurements\ncombined with the previous electron-muon and dielectron measurements span a\nwide range in rapidity, and are well described by PYTHIA Tune A. Through a\nBayesian analysis based on PYTHIA Tune A, we show that leading order pair\ncreation is the dominant $(76\\%\\pm^{14}_{19}\\%)$ contribution for $b\\bar{b}$\nproduction, whereas the data favor the scenario in which next-to-leading-order\nprocesses dominate $c\\bar{c}$ production. The small contribution of\nnext-to-leading-order processes in $b\\bar{b}$ production at the collision\nenergies of the Relativistic Heavy Ion Collider contrasts with the case at\nLarge-Hadron-Collider energies, where next-to-leading-order processes are\nexpected to dominate.\n
BibTeX
@misc{aidala2018correlations,
  author = {C. Aidala and Y. Akiba and Alfred, M. and V. Andrieux and N. Apadula and Asano, H. and B. Azmoun and Babintsev, V. and A. Bagoly and N. S. Bandara and K. N. Barish and Bathe, S. and A. Bazilevsky and M. Beaumier and R. Belmont and A. Berdnikov and Y. Berdnikov and D. Blau and Boer, M. and J. Bok and M. L. Brooks and S. Chattopadhyay and Bumazhnov, V. and S. Campbell and P. Camerini and R. Cervantes and W. C. Chang and M. Chiu and T. Chujo and J. Choi and Z. H. Citron and M. Connors and A. Mehta and M. Csanád and Tamás Ferenc Csörgő and T. W. Danley and M. Daugherity and D. d’Enterria and K. Dehmelt and K. Dehmelt and A. Denisov and W. H. Smith and A. Durum and Dion, A. and Dhruv Utpalkumar Dixit and Uliana Dmitrieva and A. Drees and Drees, K. A. and J. M. Durham and A. Durum and A. Enokizono and En'yo, H. and O. Evdokimov and O. Eyser and W. Fan and A. Hinzmann and D. E. Fields and Y. Assran and S. Fokin and S. Fokin and Y. Goto and Franz, A. and M. Große Perdekamp and C. Gal and Petr Gallus and P. Garg and H. Ge and A. Di Mattia and Y. Goto and H. M. Gray and M. Große Perdekamp and T. Gunji and T. Gunji and H. Guragain and T. Hachiya and M. Große Perdekamp and K. I. Hahn and H. Hamagaki and E. D. Hall and Hamilton, H. F. and J. Hanks and B. Jayatilaka and X. He and X. He and M. Heffner and Hemmick, T. K. and K. K. Hill and A. M. Hodges and B. Holzman and B. Hong and S. Choi and J. Huang and J. Huang and J. Huang and S. Huang and J. Murata and M. Inaba and B. Holzman and D. Ivanishchev and D. Ivanishchev},
  title = {{Correlations of $\textbackslash\{\}\textbackslash\{\}mu\textbackslash\{\}\textbackslash\{\}mu$, $e\textbackslash\{\}\textbackslash\{\}mu$, and $ee$ pairs in $p$+$p$ collisions\textbackslash\{\}n at $\textbackslash\{\}\textbackslash\{\}sqrt\{s\}=200$ GeV and implications for $c\textbackslash\{\}\textbackslash\{\}bar\{c\}$ and $b\textbackslash\{\}\textbackslash\{\}bar\{b\}$\textbackslash\{\}n production mechanisms}},
  year = {2018},
  eprint = {1805.04075},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/1805.04075},
}
Figure