g g → Z H gg\to ZH at NLO matched to parton showers with ggxy and POWHEG Preprint: CERN-TH-2026-051, P3H-26-019, TTP26-008
Joshua Davies, Kay Schönwald, Matthias Steinhauser, Daniel Stremmer · 2026
Read g g → Z H gg\to ZH at NLO matched to parton showers with ggxy and POWHEG Preprint: CERN-TH-2026-051, P3H-26-019, TTP26-008 by Joshua Davies, Kay Schönwald, Matthias Steinhauser, Daniel Stremmer (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-15762.txt.
Introduction
Davies and co-authors add a rare Higgs channel to shower-ready simulation tools. Gluon fusion into a Z boson plus a Higgs boson needs next-to-leading order care. They code the new analytic QCD results into the ggxy program. The setup supports several top quark mass schemes. Leptonic Z decays join the run, with spin links and off-shell effects kept. An interface to POWHEG hands events to Pythia showers. Partonic and hadronic cross sections both come out of the framework.
Catalog record
| CREATOR | Joshua Davies, Kay Schönwald, Matthias Steinhauser, Daniel Stremmer |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 42801 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.15762
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
g g → Z H gg\to ZH at NLO matched to parton showers with ggxy and POWHEG Preprint: CERN-TH-2026-051, P3H-26-019, TTP26-008 Joshua Davies, Kay Schönwald, Matthias Steinhauser, Daniel Stremmer Abstract We implement the recently-calculated analytic expressions for the next-to-leading order QCD corrections to g g → Z H gg\to ZH in ggxy . This provides a flexible framework for investigating partonic and hadronic cross sections for various top quark mass renormalization schemes. We augment the Z Z boson with leptonic decays, including spin correlations and off-shell effects, and furthermore provide an interface to POWHEG . This enables simulations with parton showers, performed using Pythia . 1 Introduction The associated production of a Higgs boson with a gauge boson is an important process for investigating the properties of the Higgs boson. The experimental precision obtained at the Large Hadron Collider (LHC) is steadily increasing (see, e.g., Refs. 1 ; 48 ) which demands the continuous increase in precision of theory predictions. Whereas the associated WH production is only possible through a Drell-Yan type process, for ZH production there is in addition a gluon-induced channel, which is numerically important. In recent years many higher-order corrections have been computed both for the total cross section and for differential distributions 43 ; 33 ; 22 ; 34 ; 17 ; 41 ; 27 ; 50 ; 4 ; 21 ; 20 ; 24 ; 25 including the resummation of soft gluon effects 44 ; 39 ; 23 . Most of them are implemented in the publicly available computer codes vh@nnlo 14 ; 38 and HAWK 31 ; 32 . In this work we concentrate on the process gg\to ZH and implement it in the library ggxy 28 including next-to-leading order (NLO) QCD corrections which are, to date, not available in public computer codes. ggxy is written in C++ , and due to its modular structure it provides a fast and flexible framework to compute partonic and hadronic cross sections. It is possible to choose either the pole or the \overline{\rm MS} scheme for the top quark mass. The implementation closely follows that of gg\to HH at NLO QCD, which is described in Ref. 28 . The results discussed in this paper are based on the analytic expressions obtained in the large top quark mass limit 25 , around the forward limit, and for high energies 24 . In Section 2 we describe the implementation in ggxy 28 and report on the validation against known results. We also describe the implementation of leptonic Z boson decays, including off-shell effects and spin correlations. In Section 3 we show how hadronic cross sections are computed for both stable and decaying Z bosons. Example files show how ggxy can be used to reproduce results from the literature available for stable Z bosons. For stable Z bosons we can combine our results with the Drell-Yan contributions from vh@nnlo , which is also discussed in Section 3 . Section 4 is dedicated to the interface of ggxy to POWHEG , again with the option for stable or unstable Z bosons. We validate against ggxy and then move on to simulations including parton showers with PYTHIA 47 . Note that within the POWHEG framework the Higgs boson is stable (though may be off-shell). Higgs boson decays can be performed together with the parton showers. We conclude in Section 5 . The example files shipped together with ggxy should allow anyone to implement the process gg\to ZH , including NLO QCD corrections, into their own analysis code. The ggxy library can be downloaded from https://gitlab.com/ggxy/ggxy-release . For the installation of ggxy we refer to Section 4 of the original paper 28 . Since the analytic expressions for the various expansions are quite large, compilation time is approximately an hour. It can be reduced by selecting only the required processes ( ggHH or ggZH ) in the file example-build.sh . 2 Implementing gg\to ZH to NLO QCD in ggxy For our calculation and implementation in ggxy we consider the two-loop amplitudes, required for the NLO QCD corrections, that have been calculated in Ref. 24 using two complementary expansion methods; the forward and high-energy expansions. In the forward expansion we consider a hierarchy in which the Mandelstam variable t and the external virtualities 1 1 1 Note that both the Higgs boson and Z boson can be off-shell. q_{Z}^{2} and q_{H}^{2} are the smallest scales of this process: |t|,q_{Z}^{2},q_{H}^{2}\ll m_{t}^{2},s . In the high-energy expansion we consider q_{Z}^{2},q_{H}^{2}\ll m_{t}^{2}\ll s,|t| . In particular, it was also shown in Ref. 24 that these two expansions methods cover the full phase space with high precision. The amplitude for gg\to ZH is expressed in terms of six independent form factors, \displaystyle F_{12}^{+}(t,u),\>F_{12}^{-}(t,u),\>F_{2}^{-}(t,u),\>F_{3}^{+}(t,u),\>F_{3}^{-}(t,u),\>F_{4}(t,u)\,, (1) see also Eqs. (34) and (36) of Ref. 24 for more details. We have implemented the analytic expressions of the two expansion methods for all form factors, and provide functions which automatically use the appropriate expansion following a similar logic as for gg\to HH , which is described in detail in Ref. 28 . For gg\to ZH there are a few differences which we describe in the following. First, we use different switching points between the two expansions for different form factors, which range from 200~{\rm GeV}<p_{T}<260~{\rm GeV} for any value of \sqrt{s} . In addition, we construct Padé approximations following Ref. 26 not only for the high-energy but also for the forward expansions if p_{T}>180~{\rm GeV} . This improves the convergence of the series such that both expansions have a similar accuracy at the switching points. In order to improve the numerical stability of the high-energy expansion we perform an additional expansion in -t/s if this ratio is smaller than 0.05 . For small center-of-mass energies below \sqrt{s}<200 GeV we switch from the forward expansion to the large top-quark mass expansion to improve the numerical stability and evaluation time. The latter expansion has been implemented in ggxy up to three loops in Ref. 25 . We note that at LO we use the exact expressions that have been implemented in combination with OneLOop 49 for the evaluation of the one-loop scalar integrals and for the double triangle contribution we use the expressions from Ref. 27 . The core of the implementation consists of the six form factors in Eq. ( 1 ) at one- and two-loop order from which the (partonic) Born cross section and virtual corrections used for the hadronic cross sections are computed. In the following we present the prototypes of the C++ functions, which defines the input and output values in a unique way. The one- and two-loop form factors can be computed with ⬇ complex < double > ggzhFF ( int loops , string / int ff , double s , double t , double mzs , double mhs , double mts , double murs = ggzhFFmursDefault , double muts = 0.0, unsigned scheme = 0, complex < double > kappa_hzz = 1.0, double dTriCoeff = 1.0), where the input parameters are defined as, loops: QCD loop order, 1 or 2 ff: choice of form factor, either as a string "F12p","F12m","F2m","F3p","F3m","F4" or as an integer between 1,…,6 following the same ordering s,t: Mandelstam variables mzs: