Min-Bias and the Underlying Event at the LHC

TL;DR

This work analyzes minimum-bias (MB) and underlying-event (UE) phenomena at the LHC by comparing CMS, ATLAS, and ALICE data against PYTHIA tunes tuned to Tevatron results. It assesses how well PYTHIA Tune DW describes the UE and how Tune Z1/Z2 perform for MB and UE, highlighting the need for diffraction modeling to describe MB and the importance of PDFs and energy-dependent pT0 in tune extrapolations. The study finds that UE data at the LHC broadly align with expectations, with Tune Z1 delivering a particularly good description of UE, while MB remains imperfect without explicit diffraction modeling; no single tune perfectly describes both MB and UE. The results underscore the ongoing need for refined tuning and cross-validation with newer generators to achieve consistent descriptions across MB and UE at multiple energies.

Abstract

In a very short time the experiments at the LHC have collected a large amount of data that can be used to study minimum bias (MB) collisions and the underlying event (UE) in great detail. The CDF PYTHIA 6.2 Tune DW predictions for the LHC UE data at 900 GeV and 7 TeV are examined in detail. The behavior of the UE at the LHC is roughly what we expected. The LHC PYTHIA 6.4 Tune Z1 does an excellent job describing the LHC UE data. The modeling of MB (i.e. the overall inelastic cross section) is more complicated because one must include a model of diffraction. The ability of PYTHIA Tune DW and Tune Z1 to simultaneously describe both the UE in a hard scattering process and MB collisions are studied. No model describes perfectly all the features of MB collisions at the LHC.

Figures (25)

• Figure 1: Illustration of correlations in azimuthal angle $\Delta\phi$ relative to (left) the direction of the leading charged particle, PTmax, or to (right) the leading charged particle jet, chgjet#$1$. The relative angle $\Delta\phi=\phi-\phi_1$, where $\phi_1$ is the azimuthal angle of PTmax (or chgjet#$1$) and $\phi$ is the azimuthal angle of a charged particle. There are two "transverse" regions $60^\circ<\Delta\phi< 120^\circ$, $|\eta|<\eta_{cut}$ and $60^\circ<-\Delta\phi< 120^\circ$, $|\eta|<\eta_{cut}$. The overall "transverse" region of $\eta$-$\phi$ space is defined by $60^\circ<|\Delta\phi|< 120^\circ$, $|\eta|<\eta_{cut}$. The "transverse" charged particle density is the number of charged particles in the "transverse" region divided by the area in $\eta$-$\phi$ space. Similarly, the "transverse" charged PTsum density is the scalar PTsum of charged particles in the "transverse" region divided by the area in $\eta$-$\phi$ space.
• Figure 2: CDF Run 1 data from Ref. cdfue1 at $1.8\,\textrm{TeV}$ on the density of charged particles ($p_T\!>\!0.5\,{\rm GeV/c}$, $|\eta|\!<\!1$) in the "transverse" region as defined by the leading charged particle jet, chgjet#$1$, as a function of $P_T$(chgjet#$1)$. The data are compared with ISAJET $7.32$ without MPI (top) and HERWIG $6.4$ without MPI (bottom) using the ISAJET and HERWIG default parameters with $p_T({\rm hard})$$>3\,\textrm{GeV/c}$. The Monte-Carlo predictions are divided into two categories: charged particles that arise from the break-up of the beam and target (beam-beam remnants); and charged particles that arise from the outgoing jets plus initial and final-state radiation (hard scattering component).
• Figure 3: CDF Run 1 data from Ref. cdfue1 at $1.8\,\textrm{TeV}$ on the density of charged particles ($p_T\!>\!0.5\,{\rm GeV/c}$, $|\eta|\!<\!1$) in the "transverse" region as defined by the leading charged particle jet, chgjet#$1$, as a function of $P_T$(chgjet#$1)$. The data are compared with PYTHIA $6.206$ with MPI (top) using the PYTHIA default parameters with $p_T({\rm hard})$$\ge0$ with the CTEQ3L, CTEQ4L, and CTEQ5L parton distribution functions. (bottom) Two CDF PYTHIA $6.2$ tunes, Tune A and Tune B. Tune A was adjusted to fit the CDF Run 1 data with PARP($67$) $=4.0$ and Tune B was adjusted to fit the same data but with PARP($67$) $=1.0$.
• Figure 4: (top) PYTHIA Tune A predictions at $630\,\textrm{GeV}$ for the charged PTsum density ($p_T\!>\!0.4\,{\rm GeV/c}$, $|\eta|\!<\!1$) in the "transverse" region as defined by the leading charged particle jet, chgjet#$1$, as a function of $P_T$(chgjet#$1)$ with $\epsilon=$ PARP($90$) $=0.0$, $0.16$ (default), and $0.25$. The CDF Run 1 data at $630\,\textrm{GeV}$ from Ref. cdf630 indicated a value of the PTsum density of around $0.54\,\textrm{GeV/c}$ at $P_T$(chgjet#$1)$$\approx50\,\textrm{GeV/c}$ (red line) which favors the PARP($90$) $=0.25$ curve. (bottom) Shows the $2$-to-$2$ hard scattering cut-off, $p_{T0}$, versus center-of-mass energy from PYTHIA Tune A with the default value PARP($90$) $=0.16$ and the Tune A value of PARP($90$) $=0.25$.
• Figure 5: The CDF Run 1 data from Ref. cdfzpt on the Z-boson $p_T$ distribution ($<\!p_T(Z)\!>\approx11.5\,\textrm{GeV/c}$) compared with (top) PYTHIA Tune A ($<\!p_T(Z)\!>=9.7\,\textrm{GeV/c}$) and PYTHIA Tune AW ($<\!p_T(Z)\!>=11.7\,\textrm{GeV/c}$) and compared with (bottom) PYTHIA Tune DW and HERWIG $6.4$ (without MPI).