# Christian Reuschle

Researcher, PhD (Doktor Rer. Nat.), Master of Science (Diplom), Master of Science### Research

**General Interests**

My research interests are in theoretical particle physics and high-energy collider phenomenology. In particular in Monte Carlo event generation, the automation of precision calculations, and fundamental aspects of gauge theories and scattering amplitudes. Phenomenologically I currently focus on the physics of higher-order strong (QCD) and electroweak (EW) corrections to the production of Higgs bosons and EW vector bosons, in association with QCD jets and massive quark flavours at high-energy colliders.

Earlier research of mine comprises the calculation of gravitational one-loop corrections to Standard Model processes (particularly to the muon g-2), as well as of the experimental analysis of very-high-energy cosmic gamma-ray data (utilizing machine-learning techniques).

**Specific Projects**

I am the co-developer of a new method for the fully numerical evaluation of complex many-particle processes at next-to-leading-order (NLO) QCD accuracy, dubbed the virtual numerical method (VNM). Furthermore I am an active author of the general-purpose event generator Herwig 7, one of the major Monte Carlo event generators for high-energy collider physics. Furthermore I co-supervise the development of a new automated tool for onep-loop QCD+EW corrections in the Standard Model of particle physics, NLOX, of which I am a founding author.

*VNM (aka Numerical Loop Calculations)*

Overview: In the virtual numerical method (VNM) the idea is to evaluate loop integrals fully numerical (thereby avoiding many problems of analytic approaches to loop integrals), by Monte Carlo integration, possible through subtraction terms for IR and UV singularities on the level of the loop integrand, and an efficient numerical deformation of the loop-integration contour into complex space. The integration kernels are constructed by factorization into SU(Nc) group-theoretical factors and kinematic integrands (cyclic ordered primitive amplitudes in our case), and efficient recursive formulae for the kinematic integrands. So far we have developed the VNM for NLO QCD corrections.

Particular contributions: I developed and implemented a method for local UV renormalization (a precursor to four-dimensional renormalization), as well as efficient recursive formulae for the kinematic integrands, and developed an original analytical method to systematize the factorization of SU(Nc) amplitudes into group-theory factors at NLO QCD in full generality (see next point on Color Decomposition).

Studies to which I contribute(d) in the framework of the VNM: We performed a first-time prediction of jet-rates for 7-jet production in electron-positron annihilation at NLO QCD accuracy. We investigated the application of the VNM to many-jet production in association with a Z boson at the LHC.

*Color Decomposition*

The description of QCD amplitudes in terms of primitive amplitudes is in general non-trivial. For pure gluon amplitudes in the large-Nc limit there is a direct correspondence to planar amplitudes of n=4 super Yang-Mills theory. In the realistic case of arbitrarily many quark pairs we have (for the first time) derived closed formulae, based on shuffle relations, which yield a systematic expansion in Nc at the tree and one-loop level (see above point on the VNM).

The automation and efficient numerical evaluation of higher-order corrections is tightly connected with the construction of the underlying scattering amplitudes, and methods like color decomposition, more generally referred to as quantum number management, are an efficient way to organize the computation of scattering amplitudes, utilizing the fundamental properties of the underlying gauge theories. Together with utilizing so-called spinor-helicity methods, this is used successfully in automated NLO QCD calculations. Further application of these techniques in NLO EW and NNLO QCD calculations, as well as for other algebraic groups in the Standard Model of particle physics and beyond, are conceivable.

*Herwig 7*

Overview: Herwig 7 is a general-purpose Monte Carlo event generator for high-energy colliders, written in C++. A large part of the new developments in Herwig 7 is about automated NLO QCD and parton-shower matched calculations. We have extended the interfaces of Herwig 7 to various one-loop providers, which enables fully automated NLO calculations with Herwig 7. We implemented a massive dipole subtraction method based on Catani-Seymour dipoles in Herwig 7, which enables it to include hard-scattering processes with massive final-state partons automatically at NLO QCD accuracy and offers the possibility to study mass effects in parton-shower matched calculations, in particular in top-quark production and the production of heavy-flavour jets.

Particular contributions: I implemented the infrastructure for NLO QCD calculations with massive final-state particles in Herwig 7, enabling one of the major general-purpose event generators to study state-of-the-art heavy-jet production at high-energy colliders. I implented parts of the interfaces of Herwig 7 to various one-loop providers, in particular to the one-loop program GoSam.

Studies to which I contribute(d) in the framework of Herwig 7: W+W- production at the LHC, including the loop-induced process, as well as effects of anomalous couplings, top-mass effects and matching with a parton shower. Z+b-jets production at the LHC and bottom-mass effects in 4- and 5-flavour schemes. tTH production at the LHC, comparing various ME+PS approaches. NLO matching systematics in top-quark pair production. Further interests, and current work in progress, are in loop-induced Higgs-pair production (including the two-loop virtual corrections by interfacing to dedicated two-loop providers), and mixed NLO QCD+EW corrections in the context of parton-shower matching. A comparison of 4- vs. 5-flavour schemes in NLO+PS simulations is an ongoing effort in the tTH subgroup of the Higgs Cross secction Working Group, where the background process of tTbB production is of current interest.

*NLOX*

Overview: NLOX is a program for the automated calculation of fully renormalized one-loop QCD, EW and mixed QCD+EW corrections in high-energy particle collisions in the Standard Model. NLOX relies on traditional diagrammatic methods and tensor reduction, and has a sophisticated, fast and stable, C++ tensor-reduction library, TRED, which uses recursion and caching through C++ pointers.

Particular contributions: A precursor of the new NLOX was used in the past for one-loop QCD corrections to selected processes, but an effort to promote NLOX to a fully automated matrix-element provider for one-loop QCD and EW corrections had been initiated, for which I have eventually been hired, in order to co-supervise the project and bring it to fruition. I oversee and contribute to the entire project, but in particular I re-wrote the generation of the matrix-element code for mixed QCD+EW corrections, and I added a coherent renormalization programme for mixed QCD+EW corrections.

Studies to which I contribute(d) in the framework of NLOX: We studied mixed NLO QCD+EW corrections in Z+b-jets production, including the effects of initial- and final-state masses, important for further precision studies of Higgs-boson properties and the precise determination of b-quark PDF. We partook in a comparison of programs for the automation of EW corrections in the framework of the Standar Model Electroweak Working Group. Further studies of V+b-jets production are current work in progress.

## Recent research outputs

Research output: Contribution to journal › Article

Research output: Contribution to journal › Article

Research output: Contribution to journal › Article