Recursive generation of one-loop amplitudes in the Standard Model

Abstract

We introduce the computer code Recola for the recursive generation of tree-level and one-loop amplitudes in the Standard Model. Tree-level amplitudes are constructed using off-shell currents instead of Feynman diagrams as basic building blocks. One-loop amplitudes are represented as linear combinations of tensor integrals whose coefficients are calculated similarly to the tree-level amplitudes by recursive construction of loop off-shell currents. We introduce a novel algorithm for the treatment of colour, assigning a colour structure to each off-shell current which enables us to recursively construct the colour structure of the amplitude efficiently. Recola is interfaced with a tensor-integral library and provides complete one-loop Standard Model amplitudes including rational terms and counterterms. As a first application we consider Z + 2 jets production at the LHC and calculate with Recola the next-to-leading-order electroweak corrections to the dominant partonic channels.

Published for SISSA by Springer

Received: December 3, 2012 Revised: March 12, 2013 Accepted: March 13, 2013

Published: April 5, 2013

Recursive generation of one-loop amplitudes in the

Standard Model

S. Actis,a A. Denner,b L. Hofer,b A. Scharfb S. Ucciratic

aPaul Scherrer Institut, Wrenlingen und Villigen,

CH-5232 Villigen PSI, Switzerland

bUniversitat Wrzburg, Institut fr Theoretische Physik und Astrophysik, Emil-Hilb-Weg 22, D-97074 Wrzburg, Germany

cUniversit di Torino, Dipartimento di Fisica, and INFN, Sezione di Torino, via Giuria 1, Torino, Italy

E-mail: mailto:[email protected]

Web End [email protected] , mailto:[email protected]

Web End [email protected] , mailto:[email protected]

Web End [email protected] , mailto:[email protected]

Web End [email protected] , [email protected]

Abstract: We introduce the computer code Recola for the recursive generation of tree-level and one-loop amplitudes in the Standard Model. Tree-level amplitudes are constructed using o -shell currents instead of Feynman diagrams as basic building blocks. One-loop amplitudes are represented as linear combinations of tensor integrals whose coe cients are calculated similarly to the tree-level amplitudes by recursive construction of loop o -shell currents. We introduce a novel algorithm for the treatment of colour, assigning a colour structure to each o -shell current which enables us to recursively construct the colour structure of the amplitude e ciently. Recola is interfaced with a tensor-integral library and provides complete one-loop Standard Model amplitudes including rational terms and counterterms. As a rst application we consider Z + 2 jets production at the LHC and calculate with Recola the next-to-leading-order electroweak corrections to the dominant partonic channels.

Keywords: NLO Computations, Hadronic Colliders

ArXiv ePrint: 1211.6316

c

JHEP04(2013)037

SISSA 2013 doi:http://dx.doi.org/10.1007/JHEP04(2013)037

Web End =10.1007/JHEP04(2013)037

Contents

1 Introduction 1

2 Recola: REcursive Computation of One-Loop Amplitudes 32.1 Tree-level recursion relations 32.2 One-loop recursion relations 62.3 Rational terms and renormalisation 122.4 Treatment of colour 12

3 Electroweak corrections to Z + 2 jets production at the LHC 153.1 Details of the calculation 153.1.1 General setup 163.1.2 Virtual corrections 173.1.3 Real corrections 183.1.4 Implementation 183.1.5 Accuracy and e ciency of Recola 193.2 Numerical results 203.2.1 Input parameters and selection cuts 203.2.2 Results 21

4 Conclusions 24

1 Introduction

The study of the mechanism of electroweak (EW) symmetry breaking and the search for physics beyond the Standard Model (SM) is the primary goal of the Large Hadron Collider (LHC) and the corresponding experiments. With the discovery of a bosonic resonance with a mass of around 125 GeV important progress has been achieved. Still it remains an open question if this resonance is the SM Higgs boson and if there are phenomena of new physics at the TeV scale. Evidence for a discovery of new particles and the precise determination of their masses and couplings on the one hand as well as the establishment of exclusion limits on the other hand are achieved by sophisticated experimental analyses, capable of highlighting a small signal on a huge background. For the interpretation of the data a precise knowledge of the background is essential. This often relies on theoretical descriptions, sometimes also on data-driven estimations where the extrapolation to the signal region is based on theoretical distributions. A sound comparison of experimental signals with theoretical predictions allowing precise tests of the SM (or of theories beyond) requires high precision from both experiment and theory.

1

JHEP04(2013)037

Theoretical predictions at leading order (LO) in perturbation theory are usually insu cient to match the experimental precision. At a hadron collider QCD corrections are indispensable, but also EW corrections can have an important impact. For instance, for Higgs-boson production in vector-boson fusion, EW and QCD corrections are of the same order of magnitude [1]. Moreover, the high energies attained by the LHC allow to collect data in phase-space regions where the e ects of logarithms of EW origin become sizeable. The high centre-of-mass energies available at the LHC generate a lot of events with many particles in the nal state. Therefore a proper theoretical description of LHC physics requires next-to-leading-order (NLO) computations of multiparticle processes (with ve, six, or more external legs) in the full SM (including EW corrections).

In the past years many groups have concentrated their e orts to make such calculations feasible. New techniques have been proposed, mainly for the computation of one-loop virtual corrections which are considered the bottleneck of NLO calculations. In the standard approach based on Feynman diagrams the major problem is caused by huge algebraic expressions appearing in the computation of the virtual amplitudes. The development of techniques based on Generalised Unitarity [29] allowed a change of perspective. The formal starting point of these methods is the general decomposition of one-loop amplitudes as linear combinations of scalar integrals, as obtained from the standard Passarino-Veltman reduction [10]. The computation of the coe cients of the scalar functions is then reduced to the calculation of tree-level amplitudes by means of cutting equations. The simplicity of these relations allowed the automation of NLO QCD computations leading to the development of computer programs [1118] which enabled the calculations of many QCD processes [1936] of the Les Houches priority list [37].

More recently, based on ideas of Soper [38] purely numerical methods for the calculation of one-loop QCD amplitudes have been put forward [39]. They are based on subtracting the soft, collinear, and ultraviolet divergences of one-loop amplitudes and performing the loop integration of the remaining nite integrals numerically after suitably deforming the integration contours in the complex space. These methods do not rely on Feynman graphs and have been proven to work for multi-parton amplitudes [40].

The success of the new methods did, however, not supersede the traditional diagrammatic approach. The Generalised Unitarity methods still su er from the numerical instabilities characteristic of the Passarino-Veltman reduction. These are overcome by computing the amplitude in quadruple or multiple precision at critical phase-space points, reducing however the CPU e ciency [41, 42]. While rescue solutions in this context have been proposed in refs. [43, 44], it is well-known that numerical instabilities can be avoided by constructing the amplitude as a linear combination of tensor integrals, which is the natural representation in the Feynman-diagrammatic approach. Various groups [18, 4548] have developed e cient computational techniques for the calculation of the tensor integrals that avoid numerical instabilities. Using these improved methods in the Feynman-diagrammatic approach yields more than competitive numerical codes. In particular the methods of refs. [46, 48] have been successfully applied to the calculation of both EW corrections [1, 4952] and QCD corrections [5356].

2

JHEP04(2013)037

Recently the diagrammatic approach has been further boosted by the advent of the OpenLoops algorithm [57]. Organising the diagrams into cut-opened topologies, the coe cients of the tensor integrals are recursively built with tree-level-like techniques; the e ciency is increased by pinch identities that relate higher-point loop diagrams to precomputed lower-point diagrams. Since the method is based on individual topologies, colour factorises and is treated algebraically. The present implementation of OpenLoops can handle NLO QCD corrections to any Standard Model process.

An interesting hybrid method, proposed by Andreas van Hameren in ref. [58] for evaluating one-loop gluonic amplitudes, combines stable and universal tensor-reduction methods for loop integrals with a basic result of Generalised Unitarity, namely the reduction of the computation of a one-loop amplitude to that of a set of tree-level amplitudes. The technique relies on the representation of the amplitude in terms of tensor integrals, whose coe cients are computed recursively [59] without resorting to Feynman diagrams at any stage.

In this paper we present Recola, a generator of SM one-loop (and tree) amplitudes. It is based on an algorithm which implements recursion relations for the computation of the coe cients of the tensor integrals. The goal is to combine the e ciency of the numerically stable tensor-integral reduction with the automation made possible by a completely recursive non-diagrammatic approach.

As a rst application of Recola we have calculated EW corrections to pp Z +

2 jets. Due to its large cross section and similar signatures this process provides a major background for Higgs-boson production in vector-boson fusion kinematics. The dominant NLO QCD corrections of O(3s) have been investigated in refs. [60, 61], while a subset of

EW Z+2 jets production has been studied at NLO QCD in ref. [62]. Here, we calculate the EW corrections of O(2s2) to the dominant partonic channels for Z + 2 jets production.

This paper is organised as follows: in section 2 we present the algorithm for the construction of amplitudes, introducing rst our implementation of recursion relations at tree level (section 2.1) and discussing in section 2.2 its generalisation to one-loop amplitudes. After some remarks on the computation of the rational terms and counterterms (section 2.3), we describe in section 2.4 a new algorithm for the treatment of colour. The setup of our calculation for Z + 2 jets production is detailed in section 3.1, and numerical results are discussed in section 3.2. Finally, section 4 contains our conclusions.

2 Recola: REcursive Computation of One-Loop Amplitudes

Recola is a code written in Fortran90 for the computation of tree and one-loop scattering amplitudes in the SM, based on recursion relations.

2.1 Tree-level recursion relations

The tree-level recursive algorithm is inspired by the Dyson-Schwinger equations [6365] and follows closely the strategy of Helac [6668], using o -shell currents as basic building blocks.

3

JHEP04(2013)037

Let us consider a process with E external legs, and select a particle P of the model1 together with a sub-set of n external legs. The o -shell current w(P, {n}) is then dened

as the sum of all Feynman sub-graphs which generate P combining the selected n external particles:

w(P, {n}) = n

P . (2.1)

Here the shaded bubble pictorially represents all possible sub-graphs, and the dot indicates the o -shell part of the current. If according to the Feynman rules of the theory a sub-set of n external particles cannot generate P , the corresponding current vanishes. For n > 1 the propagator of P is included in the denition and the current is called internal. A current generated by only one (n = 1) external particle P is called external; if P = P , it coincides with the wave function of the particle, otherwise it vanishes.

In a theory with tri- and quadri-linear couplings only, the internal o -shell currents can be constructed using recursively the Dyson-Schwinger equations:

n P =

i+j=n

{i},{j}

XPi,PjPi

i Pi

j Pj

k Pk

. P .

JHEP04(2013)037

i

P +

i+j+k=n

{i},{j},{k}

(2.2)

Each term of the sums represents a branch, which is obtained by multiplication of the generating currents w(Pi, {i}), w(Pj, {j}) [and w(Pk, {k}) for the second term of (2.2)] with

the interaction vertex, marked by the small box, and the propagator of P , marked by the thick line. Branches have to be built for all possible generating currents formed by sub-sets

{i}, {j} (and {k}) of the n external particles such that i + j (+k) = n. The recursive con

struction of the amplitude can be e ciently implemented using lower-multiplicity o -shell currents as seeds for the numerical evaluation of higher-multiplicity ones. First, all possible currents with two external legs (n = 2) are calculated combining through a tri-linear coupling pairs of external currents only. Next the currents with three external particles (n = 3) are generated summing the branches with three external currents linked through a quadri-linear coupling and those with one external current and one of the calculated internal currents with two external legs, linked through a tri-linear coupling. Analogously the currents with n = 4 are then computed combining two or three currents with n = 1, n = 2 and n = 3, and so on. As a consequence of the summation in (2.2), the current w(P, {n}) depends on the particle P and on the set of generating external particles {n} but

not on the particular way these particles have been combined in order to get w(P, {n}). In

this respect, working with o -shell currents rather than Feynman diagrams allows to avoid recomputing identical sub-graphs contributing to di erent diagrams, since each current is computed just once; furthermore, the summation in (2.2) reduces the number of generated objects which are passed to the next step of the recursion.

1Here all particles of the model have to be taken into account, also unphysical ones like would-be Goldstone bosons.

4

j

Pj

XP

i,Pj,Pk

In order to calculate the amplitude for a tree-level process A B we rst use crossing

symmetry to switch to incoming particles and consider the corresponding process A+B

0, where B is charge conjugate to B. Next we choose one of the external legs, for deniteness the Eth leg, to close the construction of the amplitude. Then all currents resulting from the other E 1 external legs are recursively constructed. In the last step of the procedure,

i.e. for n = E 1, we require the generated particle P to be the selected Eth external

particle. As a consequence this last current is unique, and the amplitude M is obtained

multiplying with the inverse of its propagator and with the wave function of the selected Eth particle (which coincides with the Eth external current):

M = E1 (propagator)1 E . (2.3)

The recursive evaluation of the currents begins with the external currents (n = 1), which, for colourless particles of a given polarisation and momentum p, are given by the corresponding wave functions:

p

= u (p),

p

JHEP04(2013)037

= v (p),

p

= (p),

p

= 1 . (2.4)

The explicit expressions for the spinors u (p), (p), v (p), v (p) of fermions and for the polarisation vectors (p), (p) of vector bosons have been coded using ref. [69]. For coloured external particles also the information on colour has to be kept, as we explain in section 2.4.

Once the external currents have been numerically evaluated, the internal ones are built using the Feynman rules of the theory. For example, given a pair of external e

and e+ with momenta p1 and p2 and currents u 1(p1) and v 2(p2), one can generate the internal current of a photon contracting the two external currents with the QED vertex

i e and the photon propagator i g /(p1 + p2)2. Once the particle P of the o -shell

propagator is xed, in general, several branches contribute to the same internal current since di erent combinations of particles with appropriate interaction vertices can generate the same particle P . In these cases the contributions can be simply summed up, according to (2.2).

The recursive algorithm is implemented in the code Recola through two steps.

In the rst part, the initialisation phase, the currents are identied by integer numbers which contain all the relevant non-dynamical information: the particle content (e, e+, ,

etc.), the colour information (see section 2.4) and an integer tag number. The tag number is assigned according to a binary notation [66, 70]. In practice, the external currents get a label 2i1, where i is the ith external leg, i.e. 1 1, 2 2, 3 4, . . . , E 2E1.

The tag number of an internal current is obtained by a summation of the integer tags of the external currents contributing to it. The binary notation ensures that, given the tag number of any internal current, the contributing external legs can be uniquely identied. Moreover, it reects the basic property that a current depends on the generating external particles {n}, but not on the particular way these particles have been combined in order

to obtain the current.

5

The initialisation part of the code builds a skeleton of the amplitude: all needed o -shell currents are enumerated and, for each branch, all generating o -shell currents and the generated one are identied. This part is run once for all, before giving explicit values to the momenta of the external particles, i.e. before performing the phase-space integration in a Monte Carlo program.

The second part of the code, the dynamical production phase, uses the results of the rst part to actually compute the amplitude for each point of the phase-space. First, the external currents are numerically computed; then, branch after branch, all internal currents are recursively evaluated according to the skeleton generated in the initialisation phase. Here the code has to be as e cient as possible in terms of CPU time because it must run on a large grid of points in phase-space; the computation of all non-dynamical quantities in the initialisation phase allows to avoid a repetition of those operations which can be performed once independently of the particular values of the external momenta.

2.2 One-loop recursion relations

Let us now move to the one-loop case. After summing all contributing Feynman graphs Gi

every one-loop amplitude can be written as a linear combination of tensor integrals:

M = X

i

Here the tensor coe cients c(j,rj,Nj)1rj do not depend on the loop momentum q, which is

present only in the tensor integrals

T

6

Gi =

Xj

Xrjc(j,rj,Nj)1rj T1rj(j,rj,Nj) . (2.5)

JHEP04(2013)037

1rj

(j,rj,Nj) =

(2)4D

i2

Z dDq

q1 qrj Dj,0 Dj,Nj1

, Dj,a = (q + pj,a)2 m2j,a. (2.6)

The index j classies the di erent tensor integrals needed for the process, the integer Nj equals the number of loop propagators, and rj ( Nj in the t Hooft-Feynman gauge) is

the rank of the tensor integral.

Leaving aside the computation of the tensor integrals, to be performed with the preferred technique, we focus here on the tensor coe cients c(j,rj,Nj)1rj, which for multi-leg processes, due to the complexity of the SM, result in long algebraic expressions in the standard Feynman-diagram approach. An interesting idea has been proposed in ref. [58], where recursive relations for the tensor coe cients of gluon amplitudes have been derived for colour-ordered amplitudes of purely gluonic processes. We have further developed this approach to deal with the full SM. There is a clear topological correspondence between a one-loop diagram with E external legs and a tree diagram with E + 2 external legs, obtained after cutting one of the loop lines. After uniquely xing this correspondence, one can compute the tensor coe cients c(j,rj,Nj)1rj with recursion relations similar to those used

for tree amplitudes.

Given a process A B at one loop, we consider rst the set of all tree processes

A + B + P + P 0 for each particle P of the SM.2 However, the set {A + B + P +

P

0, P SM} contains more diagrams than the original one-loop process A B. This

2Here all unphysical particles, in particular also Faddeev-Popov ghosts, must be included.

is due to the fact that we can cut the loop diagram at any of its loop lines and that we can run along the loop clockwise and counterclockwise. Therefore, we have to x some rules to discard the redundant diagrams. To explain these rules, we can work without loss of generality in a theory with a single scalar particle with a tri-linear coupling (the generalisation to the presence of a quadri-linear coupling is straightforward). In such a theory the set {A+B+P +

P 0, P SM} reduces to one process, i.e. A+B++

0.

We use tag numbers for external and internal currents as explained in section 2.1 and assign tag numbers 2E and 2E+1 to the currents corresponding to the two additional external legs of P and P. These currents are called external loop currents, while the legs of P and P are called external loop legs.

Let us rst consider the sets of diagrams with three and four external legs where only external particles enter the loop. Marking with a cross the two external loop legs of the trees, we get:

1

JHEP04(2013)037

4

8 16

+

8 16

+

8 16

+

8 16

+

8 16

+

8 16

,

1

4

1

2

2

4

2

1

4

2

4

1

2

2

4

1

4

1

2

1

2 4

8

+

1

2 8

4

+

1

8 2

4

16 32

+

16 32

+

16 32

+

16 32

1

2 4

8 2 4 8

1 4 8 1

2 8 1 2

4

16 32

+

16 32

+

16 32

+

16 32

+

8 4 2

1 1 8 4

2 2 1 8

4 4 2 1

8

16 32

+

16 32

+

16 32

+

16 32

+

1

2 8

4 2 8 4

1 8 4 1

2 4 1 2

8

16 32

+

16 32

+

16 32

+

16 32

+

4 8 2

1 1 4 8

2 2 1 4

8 8 2 1

4

16 32

+

16 32

+

16 32

+

16 32

+

1 8 2

4 8 2 4

1 2 4 1

8 4 1 8

2

16 32

+

16 32

+

16 32

+

16 32

4 . (2.7)

+

4 2 8

1 1 4 2

8 8 1 4

2 2 8 1

The diagrams on the right-hand side are obtained by cutting in all possible ways one of the loop lines of the diagrams on the left-hand side. The tree diagrams have been drawn in such a way, that one can easily identify the original sequence of the loop lines (called loop ow), starting from the external loop leg with tag number 2E and ending with the external loop leg with tag number 2E+1. Two simple rules can avoid the redundant tree diagrams and x properly the correspondence with the loop diagrams:

1) The external current 1 must be attached to the external loop current with tag number

2E. This rule xes the starting point of the loop ow and thus reduces the redundancy already up to a factor of two, the direction of the loop ow.

7

2) The external currents 1, 2 and 4 must be attached to the loop ow in ascending order (the other external currents can enter the loop ow everywhere, also between 1, 2 and 4). This rule uniquely xes the direction of the loop ow.

With these rules the way of cutting each loop diagram of (2.7) becomes unique:

8 16

16 32

1

2 4

4

4

1

1

,

2

2

1

8

1

4

1

4

4

16 32

+

JHEP04(2013)037

+

+

+

. (2.8)

The rules have to be generalised to other classes of diagrams where external legs can combine in tree sub-graphs before entering the loop ow. To this end we dene an identier number for each current, given by the smallest external tag among those forming its tag number. For example, a current with tag number 13 = 1 + 4 + 8, which has been created combining the external legs 1, 4 and 8, has identier 1; a current with tag number 6 = 2+4 has identier 2. For external currents the identier coincides with the tag number. In case of a quadri-linear coupling, the two currents entering the loop are represented by a common identier, the minimum of the two identiers for the individual currents.

Now the generalisation of the two rules is straightforward:

1) The current with identier 1 must be attached to the external loop current with tag number 2E.

2) The currents with the three smallest identiers must be attached to the loop ow following the ascending order of their identiers.

The proper treatment of self-energy insertions deserves particular care. For two identical particles owing in the self-energy loop, the selection rules (actually the rst rule alone) reduce the number of cut diagrams to one. On the other hand in this case the self-energy diagram gets a symmetry factor 1/2 because of Wicks theorem. If the two particles in the loop are di erent, we get two cut diagrams, which however give the same contribution. Therefore, the cutting procedure for self-energies reproduces the loop diagrams correctly if the corresponding tree contributions are multiplied by a factor 1/2 in all cases.

In addition to rules 1) and 2), in most renormalisation schemes some classes of diagrams have to be discarded, namely tadpoles and self-energy insertions on external legs. Therefore, being S = 1 + 2 + 4 + + 2E1= 2E 1 the sum of all external tags of the

process, we use the following additional rules:

3) A current with tag number equal to S or equal to S 2n, n = 0, 1, . . . E 1, cannot enter the loop ow in a branch with a tri-linear vertex. This eliminates tadpole diagrams and those self-energy contributions made of tri-linear vertices which are inserted on external legs.

4) If in a branch with a quadri-linear vertex one of the two currents entering the loop ow is external, the sum of their tag numbers cannot be equal to S. This eliminates the self-energy contributions involving one quadri-linear vertex inserted on external legs.

8

2 4

2 8

8 2

8 1 2 8

16 32

4 1 8 2

Applying these four rules to the generation of the skeleton for the o -shell currents of the tree processes of the set {A B + P +

P, P SM}, we obtain the proper skeleton for

the one-loop process A B.

Having reduced the formal generation of the one-loop amplitude to the generation of a set of tree-level processes, we can build the loop o -shell currents in a similar way as the tree-level currents in section 2.1, in order to obtain the tensor coe cients c(j,rj,Nj)1rj of (2.5).

At one-loop level, the particle which closes the recursive construction is the external loop leg with tag number 2E+1. Since all loop lines are virtual lines and retain their propagator, the last step of (2.3), where the last generated current is multiplied with the wave function of the particle closing the recursion, 2E+1, is performed without multiplication by

the inverse propagator resulting in

M =

1 2E1

X

P

2E

P

P 2E+1 . (2.9)

The external currents for the rst E external legs are dened as in section 2.1; the external currents of the two external loop legs are dened such that the contraction originally contained in the loop can be easily reproduced. To this end, we introduce a suitable set of spinors i = ui, vi and polarisation vectors i for the cut fermions and vector bosons,

(i) = (

i) = i , with

4

JHEP04(2013)037

1 2E1

Xi=1( i) (i) = ,

i = i, with

4

Xi=1i i = , (2.10)

where i denotes the polarisation, and , and , are spinor and Lorentz indices, respectively. The loops are glued together as:

scalars:

1

1 ,

i

vector bosons:

4

Xi=1

i ,

i

fermions:

4

Xi=1

i ,

i

4

Xi=1

i . (2.11)

Except for the scalar case, the cutting procedure associates to the one-loop amplitude the sum of four tree-level amplitudes with particular spinors/polarisation vectors for the cut particle.

9

Having xed the external currents, we describe how to compute the internal ones. As explained in section 2.1, for tree amplitudes these are computed summing up the currents generated in branches where the generating currents are multiplied with the Feynman rules for the vertex and the propagator of the generated particle. This is valid also at one-loop level for pure tree currents built by combining the original external legs 1, . . . , 2E1.

The new features of the loop case are connected to the loop o -shell currents involving the external loop leg with tag number 2E carrying a loop momentum q. The external loop current with tag number 2E denes the beginning of the loop ow; all currents with tag number 2E belong to the loop ow and are called loop currents, while the branches gener

ating them are called loop branches. Every internal loop current contains a q-dependence, generated by the Feynman rules for the vertex and the propagator. Working in the t Hooft-Feynman gauge in the SM, the q-dependence of (vertex)(propagator) takes the form

(vertex) (propagator) =

aq + b

(q + p)2 m2

, (2.12)

where the linear q-dependence in the numerator results from a fermion propagator, from a coupling of three vector bosons, or from a coupling between one vector boson and two scalar/ghost particles (in all other cases a = 0). If the interacting particles are fermions and/or vector bosons, the coe cients a and b have an additional Dirac and/or Lorentz structure which is not made explicit here for simplicity. Denoting by w1(q) the rst internal loop current, we have

w1(q) =

d11,1q1 + d1,0

(q + p1)2 m21

JHEP04(2013)037

, (2.13)

where p1 is the sum of the external momenta entering the rst loop branch while d11,1 and d1,0 result from a product of the generating currents with the constants a and b in (2.12). If the current w1 corresponds to a fermion/vector boson, w1 as well as d11,1 and d1,0 carry an additional spinor/Lorentz index, suppressed in (2.13). Proceeding along the loop ow, the second internal loop current is built combining w1(q) with tree currents and with a product (vertex)(propagator) of the form (2.12) and reads

w2(q) =

d122,2q1q2 + d12,1q1 + d2,0

[(q + p1)2 m21][(q + p2)2 m22]

. (2.14)

The recursively constructed lth loop current of the loop ow is of the form

wl(q) =

l

Xk=0d1kl,kq1 qk

Qlh=1[(q + ph)2 m2h]

. (2.15)

For the last loop current (with l = Nj = number of loop lines) the momentum pNj is equal to the sum of all external momenta and thus vanishes.

The recursion relations (2.2) are valid also for the loop currents, but cannot be used to compute them numerically (unless we give an explicit value to the loop momentum q). However, one can dene similar relations to compute the set of coe cients {dl,0, d1l,1, . . . , d1ll,l}, using the general form of (2.12) for the q-dependence of loop branches. In fact in a

10

loop branch with a tri-linear vertex, knowing the generating tree current wt and the coe cients {dl1,0, d1l1,1, . . . , d1l1l1,l1} of the generating loop current, the coe cients {dl,0, d1l,1, . . . , d1ll,l} of the generated loop current are given by

{dl,0, d1l,1, . . . , d1ll,l} = wt {0, dl1,0, d2l1,1, . . . , d2ll1,l1} a1

+ {dl1,0, d1l1,1, . . . , d1l1l1,l1, 0} b , (2.16)

where we have again omitted the Dirac/Lorentz indices associated with fermionic or vectorial currents as in (2.13). For quadri-linear vertices, the situation is even simpler because in this case a = 0 and (2.16) simplies to

{dl,0, d1l,1, . . . , d1ll,l} = wt1wt2 {dl1,0, d1l1,1, . . . , d1l1l1,l1, 0} b, (2.17)

where wt1 and wt2 are the two generating tree-level currents. The expressions (2.16) and (2.17) have to be used at each loop branch. The generated coe cients d1kl,k, k = 0, . . . , l are in general not symmetric under the exchange of their indices 1 k, but, being implicitly multiplied by the symmetric product q1 qk, they can be symmetrised

at each step. In this way the number of independent coe cients d1kl,k is decreased, leading to a reduction of operations in subsequent steps of the recursion.

The recursion relations for loop branches allow us to compute the coe cients of the loop currents, but we are not allowed to sum them unless their denominators are equal. From (2.15) one can see that the denominators are products of propagators and are determined by a sequence of o -set momenta {p1, . . . , pl} and masses {m1, . . . , ml}. Therefore,

while tree currents are dened by the tag number, the particle content, and the colour information, the loop currents need an additional parameter, called sequence number, containing the information on {p1, . . . , pl} and {m1, . . . , ml}. In this way, loop currents

with di erent q-dependent denominators are distinguished, and contributions from loop branches with the same denominators can be summed as for tree branches.

The introduction of the sequence number spoils the uniqueness of the last current. Given the cut particle and its polarisation [the index i in (2.10)], the coe cients

{dNj,0, d1Nj,1, . . . , d

1Nj

Nj,Nj

} of the last current with sequence number ns give a contribution

to the tensor coe cients of (2.5) with j = ns. The index j, describing the class of the tensor integrals, is then identied with the sequence number of the last currents. At this level, contributions from di erent polarisations of the cut particle can be summed up. For di erent cut particles one gets in general contributions to di erent classes of tensor integrals. However, since not the particles but only their masses and momenta enter the sequence number, also contributions to the same tensor integral classes appear and are combined.

Also at one-loop level the code is divided into an initialisation and a production phase. In the initialisation phase, the skeleton of the branches is generated and all quantities which do not depend on the momenta are xed. In particular, the sequence numbers of the last currents allow already at this step to determine the list of needed tensor integrals. Therefore the computation of the tensor integrals can be done independently of the one of the tensor coe cients (in the production phase of the code).

11

JHEP04(2013)037

2.3 Rational terms and renormalisation

In dimensional regularisation the calculation of Feynman amplitudes is performed in D = 4 2 space-time dimensions, and the result is arranged as a power series in . In this way,

UV divergences of the loop integrals manifest themselves as 1/ poles to be subtracted upon renormalisation. For consistency, all space-time related objects entering the amplitude have to be promoted to their D-dimensional generalisations and all manipulations have to be performed in D dimensions. Otherwise terms of order O() are missed which, combined

with the 1/ pole of the loop integral, give nite contributions to the amplitude. Because these are rational functions of the kinematical invariants, they are conventionally called rational terms. A rational term is dubbed R1-term if it results from the -dependence of the denominators of the loop integrals and it is called R2-term if it is generated by a O()

term in the numerator of the Feynman amplitude [71].We assume that the tensor integrals, taken as input by Recola, contain the R1-

terms, either by keeping the denominators of the loop propagators D-dimensional in the calculation or by explicitly adding these terms. Note, however, that even if the calculation of the tensor integrals is performed in D dimensions, they enter Recola as numerical four-dimensional tensors. The numerical construction of the tensor coe cients, on the other hand, works strictly in four space-time dimensions, so that the R2-terms are not taken into account automatically. These terms can, however, be easily computed using e ective Feynman rules which have been implemented in our code using the results of refs. [7174] .3 The insertion of the e ective Feynman rules for vertices and propagators is performed in the tree-level amplitude generator, taking care that only one of the vertices results from a rational term.

Renormalisation is performed via counterterms based on the conventions of ref. [75]. In analogy with the e ective Feynman rules for the rational terms, the insertion of counterterms takes place in the tree-level amplitude generator. Presently, counterterms are xed following the complex-mass scheme of refs. [76, 77] and the results of ref. [75] for all parameters of the SM. The strong coupling constant is renormalised in the MS-scheme at a general scale Q for contributions coming from gluons and light quarks, while the top-quark contribution is subtracted at zero momentum.

2.4 Treatment of colour

In the computation of the currents an important aspect is the treatment of colour, which does not factorise in the recursive construction (contrary to the diagrammatic approach). One could obtain the factorisation of colour by splitting the amplitude in a sum of colour-ordered amplitudes, as done in ref. [78]. This, however, would increase the number of amplitudes to compute and would become complicated in the full SM. Alternatively, one could compute colour-dressed amplitudes, as for instance in ref. [66], where o -shell currents would carry explicit colour indices and would have to be computed for each index separately, slowing down the calculation considerably. Although the number of colour-dressed amplitudes to be computed can be decreased by a Monte Carlo sampling over

3We thank R. Pittau for clarications concerning refs. [73, 74].

12

JHEP04(2013)037

colour congurations [67], the number of operations at intermediate steps remains large. In order to optimise the colour treatment further, we developed an alternative approach based on structure-dressed amplitudes, where each current gets an explicit colour structure.

This is easily achieved working in the colour-ow representation of the 1/Nc expansion [79], introduced in refs. [80, 81] for perturbative QCD computations, where the conventional 8 gluon elds Aa are replaced by a 3 3 matrix (A)ij =

12 Aa(a)ij with the trace

Pi (A)ii = 0. Quarks and antiquarks maintain the usual colour index i = 1, 2, 3,

while gluons get a pair of indices i, j = 1, 2, 3; the Gell-Mann matrices a and the structure constants in the Feynman rules are then substituted by combinations of Kronecker s. The propagators read

j i

p = ij

i(p/ + m) p2 m2

condition

JHEP04(2013)037

,

i1

j1 i2

j2

j2

i g

p =

i1

j1 i2

p2 = i1j2i2j1

i g

p2 ,

i1

j1 i2

j2p =

i1

j1 i2

j2

ip2 = i1j2i2j1

ip2 , (2.18)

while the vertices become i1

j3 =

i1

j2

j3

igs2 = i1j3i3j2

1Nc i1j2i3j3

i3

j3

i3

1 Nc

i1

j2

i3

igs 2,

j2

i1

j1

i2j2

p1

p2

j3

i3

=

i1

j1

i2j2

j3

j1

i1

j2i2

i3

i gs 2h

g (p1p2)+g (p2p3)+g(p3p1) i

i3

j3

p3

= i1j3i2j1i3j2 i1j2i2j3i3j1 i gs2 h g (p1p2)+g (p2p3)+g(p3p1) i ,

i1

j1

j4

i4

= i g2s

h

i1j4i2j1i3j2i4j3 + i1j2i2j3i3j4i4j1

2gg gg g g

i2j2 i3j3

2 + i1j3i2j4i3j2i4j1 + i1j4i2j3i3j1i4j2 2g g gg gg

+ i1j3i2j1i3j4i4j2 + i1j2i2j4i3j1i4j3 2gg g g gg i ,

i1

j1

i2j2

p1

j3

i3

=

j1

i1

j2i2

i3

j3

j3

i1

j1

i2j2

i gs2 p1 =

i1j2i2j3i3j1 i1j3i2j1i3j2

i3

i gs 2 p1 .

(2.19)

In all Feynman rules the colour part is described by products of Kronecker s, and therefore in the colour-ow representation the colour structure of the amplitude can be simply ob-

13

tained as a linear combination of all possible products of Kronecker s carrying the colour indices of the external particles. For a process with k external gluons and m external quark-antiquark pairs the amplitude takes the simple form:

A 1,..., n 1,..., n = X

P (1,...,n)

1 1 n n A1,...,n, n = k + m, (2.20)

where in general all n! permutations P (1, . . . , n) of the indices 1, . . . , n have to be considered.

In a framework based on colour-dressed amplitudes, the colour indices of the external particles would be xed and at each branch, given the colour of the generating currents, all colour congurations (3 for quarks or antiquarks, 9 for gluons) for the generated current would be computed. Many of them are zero, and the others di er just by simple factors. Since in this approach one would dene and compute unnecessarily many currents, we decided to follow a di erent strategy.

Instead of assigning an explicit colour to the currents, we assign them a colour structure, which is a product of Kronecker s. In order to understand how these structures look like, let us rst consider the external currents for quarks, antiquarks, and gluons:

i = u (p) i , j = v (p) j,

i

j = (p) i j , (2.21)

where and are the colour indices of the external particles while i and j are open colour indices, which, during the recursive construction of internal currents, are contracted with the indices of the Feynman rules of (2.19). In the recursive procedure these contractions generate products of s: some of them carry indices of external particles only, as in (2.20), and some others involve the open indices of the generated current. For example, the combination of an external quark with colour structure i1 1 with an external gluon with colour structure i2 2 2j2 produces, according to the Feynman rules of (2.19), a quark with two possible colour structures: 2 1i 2 and 2 2i 1. In both structures the rst carries just external indices 1, 1, 2, while the second one contains also the open index i of the generated quark current.

The resulting colour structures for the o -shell currents are in complete correspondence to the colour structure in (2.20) for the full amplitude. The indices of the -structures of a particular o -shell current are given by the colour indices i, j of the external particles generating the current and by potential open indices for the generated particle: no open indices for a colour-neutral particle, one for a quark/antiquark, two for a gluon. Therefore, in general, the colour structure of a gluon current, obtained from k external gluons and n k external quark-antiquark pairs, takes the form

1 1

k k

k+1

n

k+1

n

JHEP04(2013)037

i

j 1 1 n1 n1 i n nj , (2.22)

14

where permutations P (1, . . . , n + 1) of the indices 1, . . . , n, j on the right-hand side correspond to di erent currents. For the colour structure of a quark (antiquark) current, obtained from k external gluons, n k 1 external quark-antiquark pairs and an addi

tional quark (antiquark) we have

1 1

k k

k+1

n1

k+1

n

j 1 1 n1 n1 nj,

(2.23)

where again permutations P (1, . . . , n) of 1, . . . , n in the rst case and of 1, . . . , n1, j

in the second correspond to di erent currents.

We can easily distinguish two parts in the colour structure: the open part, which is always present in coloured currents, contains one (for quarks and antiquarks) or two (for gluons) s with open colour indices i and/or j, while the saturated part, which is absent in external currents, is a product of s with only external colour indices. Only the open parts of the structures play an active role in the combination of currents, while the saturated parts of the generating currents simply multiply to give the saturated part of the generated current. In our code these two parts are represented by integer numbers, based on a binary notation.

Moving from colour-dressed to structure-dressed currents reduces already the number of currents. For example, there are 9 colour-dressed currents for a gluon generated from the currents of a quark and an antiquark (although many of them vanish), while we have just one structure-dressed current. A further optimisation can be achieved introducing a colour label and giving the same colour label to currents di ering just by a colour factor (due to subsequent multiplication of di erent colour coe cients). In the example considered above of an external quark combined with an external gluon, the coe cients of the two possible colour structures 2 1i 2 and 2 2i 1 di er only by a factor 1/Nc. For structure-dressed

currents this label is easily introduced, together with the corresponding colour factors, already in the initialisation phase and can be used in the production phase of the code to compute just one of the currents with the same colour label.

3 Electroweak corrections to Z + 2 jets production at the LHC

As a rst example for the application of the code Recola, we consider the EW corrections to the dominant partonic channels contributing to the process pp Z + 2 jets.

3.1 Details of the calculation

At leading order (LO) in perturbation theory, the production of a Z boson at the LHC in association with a pair of hard jets is governed by the partonic subprocesses

qi g qi g Z , (3.1) qi qi qj qj Z, qi, qj = u, c, d, s, b, (3.2)

15

i 1 1 n1 n1 i n,

1 1

k k

k+1

n

k+1

n1

JHEP04(2013)037

qi

g

Figure 1. From left to right: sample tree diagrams for the QCD contributions to qi g qi g Z and to qi qi qj qj Z, and the EW contributions to qi qj qi qj Z.

and their crossing-related counterparts. Since we neglect avour mixing as well as nite-mass e ects for the light quarks, the LO amplitudes do not depend on the quark generation, and the contributions of the various generations to the cross section di er only by their parton luminosities. While the mixed quark-gluon (gluonic) channels (3.1) contribute to the cross section exclusively at order O(2s), the four-quark channels (3.2) develop LO

diagrams of strong as well as of EW nature leading to contributions of order O(2s),

O(2s), and O(3) to the cross section. Representative Feynman diagrams are shown

in gure 1. If standard experimental acceptance cuts are applied (see section 3.2.1 for the specication of our cuts), the mixed quark-gluon channels clearly dominate over the four-quark channels, with the subprocesses ug ugZ and dg dgZ contributing 70%

and the complete class (3.1) of partonic subprocesses contributing 80% to the total cross

section.4 Therefore as a rst step towards a complete NLO calculation of EW e ects in Z + 2 jets we here calculate EW corrections to the gluonic channels (3.1).

3.1.1 General setup

In our calculation we describe potentially resonant Z-boson propagators appearing in loop diagrams (see gure 2 left for a sample diagram) by attributing a complex mass

2Z = MZ2 iMZ Z (3.3)

to internal Z bosons. To this end, we consistently use the complex-mass scheme [76, 77, 82] where 2W and 2Z are dened as the poles of the W- and Z-boson propagators in the complex plane. On the other hand, the external Z boson is treated as a stable nal-state particle with its invariant mass being xed to MZ, where MZ2 is the real part of the complex pole of the Z-boson propagator. The pole values MV and V (V = W, Z) for the mass and width of the W and Z boson are related to the on-shell results MOSV and OSV obtained from the

LEP and Tevatron experiments by [83]

MV = MOSV/ q1 + ( OSV/MOSV)2 , V = OSV/ q1 + ( OSV/MOSV)2 . (3.4)

4In a scenario where vector-boson fusion kinematics is imposed, the dominance of the gluonic channels (3.1) shrinks. Requiring, for instance, two tagging jets in opposite hemispheres with rapidity di erence |yj1 yj2| > 4, the gluonic channels contribute only 65%.

16

g

Z

Z

qi

qj

Z

qi

qi

W

W

g

qi

qi

qj

qj

qj

JHEP04(2013)037

Z/

g

g

Z

qi

qi

qi

Z

Z

qi

g

g

Figure 2. Left: box diagram involving a potentially resonant Z-boson propagator. Right: pentagon diagram involving a 5-point tensor integral of rank 4.

For the denition of the electromagnetic coupling constant we adopt the G scheme,i.e. we x the value of via its tree-level relation with the Fermi constant G:

G =

2GMW2

1

MW2

MZ2

. (3.5)

Compared to the Thomson-limit denition of , the denition of G in the G-scheme incorporates e ects of the renormalisation-group running from the scale Q2 = 0 to the

scale Q2 = M2W. In addition NLO corrections involving logarithms of light quark masses are avoided as such contributions do not enter the muon decay.

3.1.2 Virtual corrections

The virtual corrections involve O(300) diagrams per partonic channel, including 20 pen

tagon and 71 box contributions.5 The most complicated topologies are given by pentagons involving 5-point functions of rank up to r = 4 (see gure 2 right for a sample diagram). The virtual amplitude is calculated using the t Hooft-Feynman gauge. The calculation of the tensor integrals is performed employing recursive numerical reduction to scalar integrals based on refs. [46, 48, 8486]. Numerical instabilities from small Gram determinants are avoided by resorting to various expansion algorithms for the problematic momentum congurations [48]. Both, in the case of UV divergences as well as in the case of infrared (IR) divergences, dimensional regularisation is applied to extract the corresponding singularities. The EW sector of the SM is renormalised using an on-shell prescription for the W-and Z-boson masses in the framework of the complex-mass scheme [77]. As the coupling G is derived from MW, MZ and G, its counterterm inherits a correction term r from the weak corrections to muon decay [87].

For virtual NLO contributions the nite top-quark mass a ects partonic channels involving external bottom quarks in a di erent way than channels with external quarks of the rst two generations. While the top-quark mass is properly taken into account in closed fermion loops, nite top-quark-mass e ects constrained to diagrams with external bottom quarks are neglected (

()

bg

JHEP04(2013)037

bgZ and bb ggZ are suppressed by the bottom PDFs,

gg b

()

bZ contributes about 1% at LO).

5While Recola does not use Feynman diagrams, we give these numbers as a measure of the complexity of the process.

17

3.1.3 Real corrections

The EW real corrections to the subprocess (3.1) are induced by photon Bremsstrahlung and given by

qi g qi g Z . (3.6)

Emission of a soft or a collinear photon from an external quark leads to IR divergences which are regularised dimensionally. If an IR-safe event denition is used, the nal-state singularities cancel with corresponding IR poles from the virtual corrections. For the initial-state singularities this cancellation is incomplete but the remnant can be absorbed into a redenition of the quark distribution function. Technically we make use of the Catani-Seymour dipole formalism as formulated in ref. [88], which we transferred in a straightforward way to the case of dimensionally regularised photon emission.

In addition to the singularities from soft and collinear photon emission we face a further source of IR divergences originating from a soft nal-state gluon (see ref. [50]). Isolated soft gluons do not pose any problem as they do not pass our selection cuts because the requirement of two hard jets is not fullled. However, in IR-safe observables quarks, and thus all QCD partons, have to be recombined with photons if they are su ciently collinear. Thus, if a soft gluon is collinear to a photon, it still passes the selection cuts if recombined with the collinear photon, giving rise to a soft-gluon divergence that would be cancelled by the virtual QCD corrections to Z + 1 jet + production. Following refs. [50, 89] we eliminate this singularity by discarding events which contain a jet consisting of a hard photon recombined with a soft parton a (a = qi, qi, g) taking the photon-jet energy fraction z = E /(E + Ea) as a discriminator. Photonic jets with z above a critical value zcut are attributed to the process pp Z + 1 jet + and therefore they are excluded.

However, this event denition is still not IR-safe because the application of the z -cut to recombined quark-photon jets spoils the cancellation of nal-state collinear singularities with the IR divergences from the virtual corrections. This is cured by absorbing the leftover singularities into the measured quark-photon fragmentation function [90, 91].

3.1.4 Implementation

In order to ensure correctness of our results, and in particular of the calculation with Recola, we have performed two independent calculations which we nd to be in mutual agreement. While the rst one applies the technique of recursive amplitude generation as described in section 2, the second one relies on the conventional Feynman-diagrammatic approach.

In the rst calculation the amplitude generator Recola provides the Feynman amplitudes. For the evaluation of the tensor integrals Recola is interfaced with the Fortran library Collier [92]. To this end Collier has been extended by an e cient algorithm for building up the tensor integrals from the recursively calculated Lorentz-invariant coefcient functions. The phase-space integration is performed by means of a generic in-house Monte-Carlo generator [93] following the multi-channel sampling approach.

The second code uses FeynArts 3.2 [94, 95] and FormCalc 3.1 [96] for the generation and simplication of the Feynman amplitudes. For the numerical evaluation the

18

JHEP04(2013)037

Process class Virtual [ fb] |RP 1|[%] real [ fb] |RP 1|[%] NLOEW [ fb] |RP 1|[%]

qg qgZ, 14463 10 0.3 0.2

825 9 2 3

15288 13 0.3 0.2

qg qgZ 14499 27 841 22 15340 35

qq ggZ

1395 2 0.8 0.5

118 1 0.01 1

1277 2 0.9 0.6

1406 7 118 1 1288 7

gg qqZ

1024 2 0.5 0.4

186 1 0.7 0.9

1210 2 0.3 0.3

1018 3 187 1 1206 3Table 1. Comparison of numerical results from Recola (upper numbers) and Pole (lower numbers) for the NLO contribution NLOEW to the total cross section of various partonic process classes (summed over q = u, d). In addition to the complete NLO correction we separately give the nite virtual and real corrections and the relative di erences between Recola and Pole.

amplitudes are translated into the Weyl-van der Waerden formalism [97] using the program Pole [98]. The tensor integrals are again evaluated by Collier which by itself provides two independent implementations of all its building blocks. Finally, the phase-space integration is performed with the multi-channel generator Lusifer [99].

3.1.5 Accuracy and e ciency of Recola

In this section we estimate the accuracy and e ciency of the purely numerical algorithm Recola by comparing with the code Pole which is based on algebraically generated analytical expressions.

In table 1 we compare results obtained with Recola (upper numbers) and Pole (lower numbers) for the NLO contribution NLOEW to the total cross section for various classes of partonic channels. We further display separate results for the nite virtual corrections including the integrated dipoles (virtual) and for the nite real corrections including the dipole-subtraction terms (real). In addition we also present the relative deviation |R/P1|

between the results of Recola (R) and Pole (P). For the results obtained by Recola we have requested 5 106 accepted events in the case of the virtual corrections and 108

accepted events in the case of the real corrections, while (roughly by a factor of 10) lower statistics has been used for the calculation with Pole. The results in table 1 demonstrate that our two independent calculations agree with each other within the Monte Carlo errors in the per-mille range.

A more detailed comparison can be obtained by comparing the weights at individual phase-space points. For 106 Monte Carlo-generated phase-space points we have compared the pure virtual contribution 2 Re(MLOMNLO) to the squared matrix element (with di

vergences omitted) calculated by Recola and Pole for the partonic process ug ugZ.

In gure 3 we show the integrated fraction of the 106 phase-space points for which the agreement |R/P1| between Recola and Pole is worse than . We nd a typical agree

ment of 1011 1014, with less than 1% of the phase-space points showing an agreement

worse than 107 and less than 0.02% showing an agreement worse than 105. The result of this test of the precision of the code is similar to the one performed by the OpenLoops

19

JHEP04(2013)037

100

101

102

103

104

105

106

maximum accuracy

Figure 3. Level of agreement of 2 Re(MLOMNLO) between Recola and Pole for ug ugZ.

The plot shows the probability for an agreement worse than for 106 phase-space points generated by the Monte Carlo.

collaboration [57] which also uses the tensor-integral library Collier. Note, however, that we compare two independent codes for the calculation of the tensor coe cients which are based on two entirely di erent algorithms. Furthermore the distribution of the phase-space points is in our case determined from the multi-channel Monte Carlo generator adapted to the peaking structure of the underlying process.

Finally, we give some details on timing and the amount of memory required. The evaluation of the spin- and colour-summed one-loop matrix elements takes about 30 ms per phase-space point for ug ugZ (or any other partonic process considered in this paper) on

a single Intel i7-2720QM core with gfortran 4.6.1. The size of the executable on disk is about 3 MB. The matrix elements are constructed during the run and the required memory depends strongly on the size of the dynamically generated internal arrays and thus on the considered process. For pp Z + 2 jets memory is no issue. More complicated processes

will be considered in the future.

3.2 Numerical results

3.2.1 Input parameters and selection cuts

We use the following set of input parameters [100],

G = 1.1663787 105 GeV2,

MWOS = 80.385 GeV, OSW = 2.085 GeV, MZOS = 91.1876 GeV, OSZ = 2.4952 GeV,

MH = 125 GeV, mt = 173.2 GeV (3.7)

with the value for the top-quark mass taken from ref. [101]. The pole masses MW,Z and

widths W,Z entering our calculation are obtained from the stated on-shell values MOSW,Z

20

fractionofevents

JHEP04(2013)037

1016

1012

108

100

104

according to (3.4). The electromagnetic coupling constant G is determined from G via (3.5). The CKM matrix only appears in loop amplitudes and is set to unity.

For the prediction of the hadronic pp Z + 2 jets cross section the partonic cross sec

tions have to be convoluted with the corresponding parton distribution functions (PDFs). Since our calculation does not take into account NLO QCD e ects, we consistently resort to LO PDFs, using the LHAPDF implementation of the central MSTW2008LO PDF set [102]. From there we infer the value

LOs(MZ) = 0.1394 (3.8)

for the strong coupling constant. We identify the QCD factorisation scale F and the renormalisation scale R choosing

F = R = MZ . (3.9)

Note that the choice of the scales F,R as well as the actual value for the strong coupling s plays a minor role for our numerical analysis of EW radiative corrections in section 3.2.2. We focus on the relative importance of the NLO EW corrections considering the ratio NLOEW/LO from which the s and the scale dependence drop out.

For the jet-reconstruction we use the anti-kT clustering algorithm [103] with separation parameter R = 0.4. For our scenario with exactly two partons and one potential photon in the nal state this simply amounts to recombining the photon with a parton a if Ra =

q(ya y )2 + 2a < R. Here y = 12 ln[(E + pL)/(E pL)] is the particles rapidity with E

denoting its energy and pL its three-momentum component along the beam axis, and a is the azimuthal angle between the the photon and the parton a in the plane transverse to the beam axis. In case of recombination, the resulting photon-parton jet is subjected to the cut z = E /(E +Ea) < 0.7 in order to distinguish between Z+2 jets and Z+1 jet+

production as explained in section 3.1.3. After a possible recombination, we require two hard jets with

pT,jet > 25 GeV, |yjet| < 4.5 (3.10)

for the nal event.

3.2.2 Results

In this section we present results for the total cross section and various di erential distributions using the numerical input parameters and acceptance cuts introduced above. The total cross section and its composition at LO for the 8 TeV LHC is shown in table 2 where the absolute and relative contributions of the partonic channels are listed. In the lower part of table 2 we provide the contribution to the total cross section of partonic processes with external gluons (gluonic) and of the four-quark processes (four-quark). We nd the total cross section dominated by processes with external gluons, in particular by the quark-gluon induced processes (69%). We also provide the NLO cross section and the relative EW corrections for the gluonic channels in the last two columns of table 2. For our set of cuts, they range between 1.0% and 1.4% for the di erent gluonic channels.

21

JHEP04(2013)037

Process class LO [pb] LO/LOtot [%] NLOEW [pb]

NLOEW LO

1 [%]

ug ugZ, dg dgZ, 1324.1(2) 68.79 1308.8(2) 1.16

g gZ,

dg

dgZ u ggZ, d

d ggZ 128.84(2) 6.69 127.56(2) 0.99

gg uZ, gg d

dZ 87.40(2) 4.54 86.18(2) 1.37

uu uuZ, dd ddZ, 88.41(2) 4.59

Z,

dd

ddZ

JHEP04(2013)037

u uZ, d

d d

dZ, 87.98(2) 4.57

u uZ, d

d d

dZ

u d

dZ, dd uZ, 16.566(3) 0.86

u d

dZ, dd uZ

ud udZ,

d

dZ, 111.74(3) 5.81

ud udZ,

d

dZ

dZ,d dZ, 79.70(2) 4.14

ud u

dZ,d dZgluonic 1540.4(2) 80.02 1522.5(2) 1.16

four-quark 384.41(4) 19.98 sum 1924.8(2) 100.00

Table 2. Composition of the LO cross section for pp Z + 2 jets at the 8 TeV LHC. In the rst

column the partonic processes are listed, where u, u denote the up-type quarks u, c and d, d the down-type quarks d, s, b. The second column provides the corresponding cross section where the numbers in parenthesis give the integration error on the last digit. The third column contains the relative contribution to the total cross section in percent. In the fourth column we list the NLO EW cross section for the gluonic channels and in the last column the relative EW corrections.

In the following we present results for distributions at LO and NLO for the gluonic channels only. Although these channels dominate the total cross section we emphasise that there are certain phase-space regions where the relative importance of the four-quark processes is enhanced and these channels and the corresponding EW corrections should not be neglected to describe pp Z + 2 jets properly.

For each distribution we provide two plots: the upper panels show the LO and NLO prediction for the di erential cross section while the lower panels show the NLO result normalised to the LO result. In gure 4 we present results for the di erential cross section as a function of the transverse momentum and the rapidity for the harder jet j1 and softer jet j2, respectively. Both transverse momentum distributions show steep slopes over six orders of magnitude in the displayed pT range. The EW corrections lower the LO prediction, and their relative size grows in absolute value with increasing transverse momentum due to

22

ud u

105

105

104

LO

NLO

LO

NLO

104

103

103

d

dp T , j[fb/GeV]

d

dpT,j [fb / GeV]

102

102

101

101

100

100

101

101

102

1

1

0.8

0.8

250 500 750 1000

pT,j [GeV]

250 500 750 1000

JHEP04(2013)037

pT,j [GeV]

4 105

4 105

LO

NLO

LO

NLO

3 105

3 105

d

dy j[fb]

2 105

2 105

d d

yj [fb]

1 105

1 105

0 100

0 100

1

1

0.99

0.99

0.98

4 3 2 1 0 1 2 3 4 yj

4 3 2 1 0 1 2 3 4

0.98

yj

Figure 4. Distributions of the transverse momentum and the rapidity of the harder jet j1 and the softer jet j2 at the 8 TeV LHC at LO (blue, dashed) and NLO (red, solid). The lower panels show the ratio of the NLO distribution over the LO distribution.

the well-known EW Sudakov logarithms [104107]. For transverse momenta of the softer jet pT,j2 1 TeV the impact of EW corrections amounts up to 10% relative to the Born

approximation, while for transverse momenta of the harder jet pT,j1 1 TeV the EW e ects

are of the order of 15%. Since, the rapidity distributions are not sensitive to Sudakov

logarithms, the corresponding EW corrections are at and around 1% as for the total

cross section.

The di erential cross section as a function of the di-jet invariant mass and as a function of the transverse momentum of the Z boson is shown on the left-hand side in gure 5. For both distributions we nd the expected dependence on the Sudakov logarithms, although the sensitivity in the di-jet invariant-mass distribution is less pronounced than in the pT,Z-

distribution. For Mjj 500 GeV the EW corrections are of the order of 2%; they amount

up to 4% for Mjj 2 TeV. The transverse momentum distribution of the Z boson receives

large corrections, from 15% for pT,Z 500 GeV to 25% for pT,Z 1 TeV. On the upper

right-hand side in gure 5 we present the di erential distribution of the relative azimuthal angle jj between the two jets. The jj-distribution shows that the two jets are preferably back-to-back in the transverse plane and that the EW corrections lower the di erential cross section by 11.5%. They induce a shape change at the permille level relative to

23

104

1 106

LO

NLO

LO

NLO

103

1 106

d

dM jj[fb/GeV]

102

8 105

d d

jj [fb]

101

6 105

100

4 105

101

2 105

1

1

0.99

0.9

0

4

2

3

4

0.98

0 500 1000 1500 2000

Mjj[GeV]

jj

JHEP04(2013)037

104

3 105

LO

NLO

103

d

dp T , Z[fb/GeV]

102

2 105

101

LO

NLO

d d

yZ [fb]

100

1 105

101

102

0 100

1

1

0.8

0.99

250 500 750 1000

pT,Z[GeV]

4 3 2 1 0 1 2 3 4

0.98

yZ

Figure 5. Distributions of the di-jet invariant mass, the relative azimuthal angle between the two jets, the transverse momentum and the rapidity of the Z boson at the 8 TeV LHC at LO (blue, dashed) and NLO (red, solid). The lower panels show the ratio of the NLO distribution over the LO distribution.

the LO approximation. The rapidity distribution of the Z boson is depicted in the lower right-hand side of gure 5. In the central region |yZ| < 2, where most of the Z bosons are

produced, the EW corrections lower the LO cross section by 11.5% while for |yZ| > 2 their e ect drops to the permille level.

4 Conclusions

The full exploitation of the Large Hadron Collider relies on precise theoretical predictions. To this end QCD and electroweak next-to-leading order corrections have to be calculated for many processes involving many particles in the nal state. This requires e cient and reliable automatic tools.

In this paper we have presented Recola, a Fortran90 code for the REcursive Computation of One-Loop Amplitudes. It uses methods based on Dyson-Schwinger equations to calculate the coe cients of all tensor integrals appearing in a one-loop amplitude recursively. The tensor integrals can then be evaluated with e cient numerically stable techniques. The algorithm has been implemented for the full electroweak Standard Model, including counterterms and rational terms, but could be generalised to more complicated

24

theories in a straightforward way. The implementation supports the complex-mass scheme and is thus applicable to processes involving intermediate unstable particles. For the treatment of colour we have developed a new recursive algorithm based on colour structures that naturally appear in the colour-ow representation.

As a rst application of Recola, we have calculated the electroweak corrections to the dominant partonic channels in Z+2 jets production at the LHC. The results have been veried with an independent calculation based on Feynman-diagrammatic methods. For a typical set of cuts, the electroweak corrections are negative at the level of one percent, but become sizeable where large energy scales are relevant. However, in general and in particular for large energies of the jets, a meaningful prediction requires the inclusion of next-to-leading-order corrections to all partonic channels. This will be pursued in a forthcoming publication.

Acknowledgments

We are grateful to S. Pozzorini for many useful discussions and to A. Mck for help concerning Pole. This work was supported in part by the Swiss National Science Foundation (SNF) under contract 200021-126364, by the Deutsche Forschungsgemeinschaft (DFG) under reference number DE 623/2-1, and by the Helmholtz alliance Physics at the Terascale.

References

[1] M. Ciccolini, A. Denner and S. Dittmaier, Electroweak and QCD corrections to Higgs production via vector-boson fusion at the LHC, http://dx.doi.org/10.1103/PhysRevD.77.013002

Web End =Phys. Rev. D 77 (2008) 013002 [arXiv:0710.4749] [http://inspirehep.net/search?p=find+EPRINT+arXiv:0710.4749

Web End =INSPIRE ].

[2] Z. Bern, L.J. Dixon, D.C. Dunbar and D.A. Kosower, One loop N point gauge theory amplitudes, unitarity and collinear limits, http://dx.doi.org/10.1016/0550-3213(94)90179-1

Web End =Nucl. Phys. B 425 (1994) 217 [http://arxiv.org/abs/hep-ph/9403226

Web End =hep-ph/9403226 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/9403226

Web End =INSPIRE ].

[3] Z. Bern, L.J. Dixon, D.C. Dunbar and D.A. Kosower, Fusing gauge theory tree amplitudes into loop amplitudes, http://dx.doi.org/10.1016/0550-3213(94)00488-Z

Web End =Nucl. Phys. B 435 (1995) 59 [http://arxiv.org/abs/hep-ph/9409265

Web End =hep-ph/9409265 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/9409265

Web End =INSPIRE ].

[4] R. Britto, F. Cachazo and B. Feng, Generalized unitarity and one-loop amplitudes in N = 4 super-Yang-Mills, http://dx.doi.org/10.1016/j.nuclphysb.2005.07.014

Web End =Nucl. Phys. B 725 (2005) 275 [http://arxiv.org/abs/hep-th/0412103

Web End =hep-th/0412103 ] [http://inspirehep.net/search?p=find+EPRINT+hep-th/0412103

Web End =INSPIRE ].

[5] G. Ossola, C.G. Papadopoulos and R. Pittau, Reducing full one-loop amplitudes to scalar integrals at the integrand level, http://dx.doi.org/10.1016/j.nuclphysb.2006.11.012

Web End =Nucl. Phys. B 763 (2007) 147 [http://arxiv.org/abs/hep-ph/0609007

Web End =hep-ph/0609007 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/0609007

Web End =INSPIRE ].

[6] G. Ossola, C.G. Papadopoulos and R. Pittau, Numerical evaluation of six-photon amplitudes, http://dx.doi.org/10.1088/1126-6708/2007/07/085

Web End =JHEP 07 (2007) 085 [arXiv:0704.1271] [http://inspirehep.net/search?p=find+EPRINT+arXiv:0704.1271

Web End =INSPIRE ].

[7] R.K. Ellis, W. Giele and Z. Kunszt, A numerical unitarity formalism for evaluating one-loop amplitudes, http://dx.doi.org/10.1088/1126-6708/2008/03/003

Web End =JHEP 03 (2008) 003 [arXiv:0708.2398] [http://inspirehep.net/search?p=find+EPRINT+arXiv:0708.2398

Web End =INSPIRE ].

[8] W.T. Giele, Z. Kunszt and K. Melnikov, Full one-loop amplitudes from tree amplitudes, http://dx.doi.org/10.1088/1126-6708/2008/04/049

Web End =JHEP 04 (2008) 049 [arXiv:0801.2237] [http://inspirehep.net/search?p=find+EPRINT+arXiv:0801.2237

Web End =INSPIRE ].

[9] R.K. Ellis, W.T. Giele, Z. Kunszt and K. Melnikov, Masses, fermions and generalized d-dimensional unitarity, http://dx.doi.org/10.1016/j.nuclphysb.2009.07.023

Web End =Nucl. Phys. B 822 (2009) 270 [arXiv:0806.3467] [http://inspirehep.net/search?p=find+EPRINT+arXiv:0806.3467

Web End =INSPIRE ].

25

JHEP04(2013)037

[10] G. Passarino and M. Veltman, One loop corrections for e+e annihilation into + in the Weinberg model, http://dx.doi.org/10.1016/0550-3213(79)90234-7

Web End =Nucl. Phys. B 160 (1979) 151 [http://inspirehep.net/search?p=find+J+Nucl.Phys.,B160,151

Web End =INSPIRE ].

[11] C. Berger, Z. Bern, L. Dixon, F. Febres Cordero, D. Forde et al., An automated implementation of on-shell methods for one-loop amplitudes, http://dx.doi.org/10.1103/PhysRevD.78.036003

Web End =Phys. Rev. D 78 (2008) 036003 [arXiv:0803.4180] [http://inspirehep.net/search?p=find+EPRINT+arXiv:0803.4180

Web End =INSPIRE ].

[12] W. Giele and G. Zanderighi, On the numerical evaluation of one-loop amplitudes: the gluonic case, http://dx.doi.org/10.1088/1126-6708/2008/06/038

Web End =JHEP 06 (2008) 038 [arXiv:0805.2152] [http://inspirehep.net/search?p=find+EPRINT+arXiv:0805.2152

Web End =INSPIRE ].

[13] A. Lazopoulos, Multi-gluon one-loop amplitudes numerically, arXiv:0812.2998 [http://inspirehep.net/search?p=find+EPRINT+arXiv:0812.2998

Web End =INSPIRE ].

[14] W. Giele, Z. Kunszt and J. Winter, E cient color-dressed calculation of virtual corrections, http://dx.doi.org/10.1016/j.nuclphysb.2010.07.007

Web End =Nucl. Phys. B 840 (2010) 214 [arXiv:0911.1962] [http://inspirehep.net/search?p=find+EPRINT+arXiv:0911.1962

Web End =INSPIRE ].

[15] S. Badger, B. Biedermann and P. Uwer, NGluon: a package to calculate one-loop multi-gluon amplitudes, http://dx.doi.org/10.1016/j.cpc.2011.04.008

Web End =Comput. Phys. Commun. 182 (2011) 1674 [arXiv:1011.2900] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1011.2900

Web End =INSPIRE ].

[16] V. Hirschi, R. Frederix, S. Frixione, M.V. Garzelli, F. Maltoni et al., Automation of one-loop QCD corrections, http://dx.doi.org/10.1007/JHEP05(2011)044

Web End =JHEP 05 (2011) 044 [arXiv:1103.0621] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1103.0621

Web End =INSPIRE ].

[17] G. Bevilacqua, M. Czakon, M. Garzelli, A. van Hameren, A. Kardos et al., HELAC-NLO, http://dx.doi.org/10.1016/j.cpc.2012.10.033

Web End =Comput. Phys. Commun. 184 (2013) 986 [arXiv:1110.1499] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1110.1499

Web End =INSPIRE ].

[18] G. Cullen, N. Greiner, G. Heinrich, G. Luisoni, P. Mastrolia et al., Automated one-loop calculations with GoSam, http://dx.doi.org/10.1140/epjc/s10052-012-1889-1

Web End =Eur. Phys. J. C 72 (2012) 1889 [arXiv:1111.2034] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1111.2034

Web End =INSPIRE ].

[19] C. Berger, Z. Bern, L.J. Dixon, F. Febres Cordero, D. Forde et al., Next-to-leading order QCD predictions for W + 3-jet distributions at hadron colliders,http://dx.doi.org/10.1103/PhysRevD.80.074036

Web End =Phys. Rev. D 80 (2009) 074036 [arXiv:0907.1984] [http://inspirehep.net/search?p=find+EPRINT+arXiv:0907.1984

Web End =INSPIRE ].

[20] R.K. Ellis, K. Melnikov and G. Zanderighi, W + 3 jet production at the Tevatron, http://dx.doi.org/10.1103/PhysRevD.80.094002

Web End =Phys. Rev. D 80 (2009) 094002 [arXiv:0906.1445] [http://inspirehep.net/search?p=find+EPRINT+arXiv:0906.1445

Web End =INSPIRE ].

[21] C. Berger, Z. Bern, L.J. Dixon, F. Febres Cordero, D. Forde et al., Precise predictions for W + 3 jet production at hadron colliders, http://dx.doi.org/10.1103/PhysRevLett.102.222001

Web End =Phys. Rev. Lett. 102 (2009) 222001 [arXiv:0902.2760] [http://inspirehep.net/search?p=find+EPRINT+arXiv:0902.2760

Web End =INSPIRE ].

[22] G. Bevilacqua, M. Czakon, C. Papadopoulos, R. Pittau and M. Worek, Assault on the NLO wishlist: pp ttb

b, http://dx.doi.org/10.1088/1126-6708/2009/09/109

Web End =JHEP 09 (2009) 109 [arXiv:0907.4723] [http://inspirehep.net/search?p=find+EPRINT+arXiv:0907.4723

Web End =INSPIRE ].

[23] K. Melnikov and G. Zanderighi, W + 3 jet production at the LHC as a signal or background, http://dx.doi.org/10.1103/PhysRevD.81.074025

Web End =Phys. Rev. D 81 (2010) 074025 [arXiv:0910.3671] [http://inspirehep.net/search?p=find+EPRINT+arXiv:0910.3671

Web End =INSPIRE ].

[24] G. Bevilacqua, M. Czakon, C. Papadopoulos and M. Worek, Dominant QCD backgrounds in Higgs boson analyses at the LHC: a study of pp tt+ 2 jets at next-to-leading order,

http://dx.doi.org/10.1103/PhysRevLett.104.162002

Web End =Phys. Rev. Lett. 104 (2010) 162002 [arXiv:1002.4009] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1002.4009

Web End =INSPIRE ].

[25] C. Berger, Z. Bern, L.J. Dixon, F. Febres Cordero, D. Forde et al., Next-to-leading order QCD predictions for Z, + 3-jet distributions at the Tevatron,http://dx.doi.org/10.1103/PhysRevD.82.074002

Web End =Phys. Rev. D 82 (2010) 074002 [arXiv:1004.1659] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1004.1659

Web End =INSPIRE ].

[26] T. Melia, K. Melnikov, R. Rontsch and G. Zanderighi, Next-to-leading order QCD predictions for W +W +jj production at the LHC, http://dx.doi.org/10.1007/JHEP12(2010)053

Web End =JHEP 12 (2010) 053 [arXiv:1007.5313] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1007.5313

Web End =INSPIRE ].

[27] C. Berger, Z. Bern, L.J. Dixon, F. Febres Cordero, D. Forde et al., Precise predictions for W + 4 jet production at the large hadron collider, http://dx.doi.org/10.1103/PhysRevLett.106.092001

Web End =Phys. Rev. Lett. 106 (2011) 092001 [arXiv:1009.2338] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1009.2338

Web End =INSPIRE ].

26

JHEP04(2013)037

[28] G. Bevilacqua, M. Czakon, A. van Hameren, C.G. Papadopoulos and M. Worek, Complete o -shell e ects in top quark pair hadroproduction with leptonic decay at next-to-leading order, http://dx.doi.org/10.1007/JHEP02(2011)083

Web End =JHEP 02 (2011) 083 [arXiv:1012.4230] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1012.4230

Web End =INSPIRE ].

[29] T. Melia, K. Melnikov, R. Rontsch and G. Zanderighi, NLO QCD corrections for W +W

pair production in association with two jets at hadron colliders,

http://dx.doi.org/10.1103/PhysRevD.83.114043

Web End =Phys. Rev. D 83 (2011) 114043 [arXiv:1104.2327] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1104.2327

Web End =INSPIRE ].

[30] R. Frederix, S. Frixione, V. Hirschi, F. Maltoni, R. Pittau et al., W and Z/ boson

production in association with a bottom-antibottom pair, http://dx.doi.org/10.1007/JHEP09(2011)061

Web End =JHEP 09 (2011) 061 [arXiv:1106.6019] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1106.6019

Web End =INSPIRE ].

[31] H. Ita, Z. Bern, L. Dixon, F. Febres Cordero, D. Kosower et al., Precise predictions for Z + 4 jets at hadron colliders, http://dx.doi.org/10.1103/PhysRevD.85.031501

Web End =Phys. Rev. D 85 (2012) 031501 [arXiv:1108.2229] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1108.2229

Web End =INSPIRE ].

[32] G. Bevilacqua, M. Czakon, C. Papadopoulos and M. Worek, Hadronic top-quark pair production in association with two jets at next-to-leading order QCD,http://dx.doi.org/10.1103/PhysRevD.84.114017

Web End =Phys. Rev. D 84 (2011) 114017 [arXiv:1108.2851] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1108.2851

Web End =INSPIRE ].

[33] Z. Bern, G. Diana, L. Dixon, F. Febres Cordero, S. Hche et al., Four-jet production at the large hadron collider at next-to-leading order in QCD, http://dx.doi.org/10.1103/PhysRevLett.109.042001

Web End =Phys. Rev. Lett. 109 (2012) 042001 [arXiv:1112.3940] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1112.3940

Web End =INSPIRE ].

[34] N. Greiner, G. Heinrich, P. Mastrolia, G. Ossola, T. Reiter et al., NLO QCD corrections to the production of W +W plus two jets at the LHC, http://dx.doi.org/10.1016/j.physletb.2012.06.027

Web End =Phys. Lett. B 713 (2012) 277 [arXiv:1202.6004] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1202.6004

Web End =INSPIRE ].

[35] G. Bevilacqua and M. Worek, Constraining BSM physics at the LHC: four top nal states with NLO accuracy in perturbative QCD, http://dx.doi.org/10.1007/JHEP07(2012)111

Web End =JHEP 07 (2012) 111 [arXiv:1206.3064] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1206.3064

Web End =INSPIRE ].

[36] S. Badger, B. Biedermann, P. Uwer and V. Yundin, NLO QCD corrections to multi-jet production at the LHC with a centre-of-mass energy of s = 8 TeV,http://dx.doi.org/10.1016/j.physletb.2012.11.029

Web End =Phys. Lett. B 718 (2013) 965 [arXiv:1209.0098] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1209.0098

Web End =INSPIRE ].

[37] SM AND NLO MULTILEG and SM MC Working Groups collaboration,J. Alcaraz Maestre et al., The SM and NLO multileg and SM MC working groups: summary report, arXiv:1203.6803 [http://inspirehep.net/search?p=find+EPRINT+arXiv:1203.6803

Web End =INSPIRE ].

[38] D.E. Soper, Techniques for QCD calculations by numerical integration, http://dx.doi.org/10.1103/PhysRevD.62.014009

Web End =Phys. Rev. D 62 (2000) 014009 [http://arxiv.org/abs/hep-ph/9910292

Web End =hep-ph/9910292 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/9910292

Web End =INSPIRE ].

[39] S. Becker, C. Reuschle and S. Weinzierl, Numerical NLO QCD calculations, http://dx.doi.org/10.1007/JHEP12(2010)013

Web End =JHEP 12 (2010) 013 [arXiv:1010.4187] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1010.4187

Web End =INSPIRE ].

[40] S. Becker, D. Goetz, C. Reuschle, C. Schwan and S. Weinzierl, NLO results for ve, six and seven jets in electron-positron annihilation, http://dx.doi.org/10.1103/PhysRevLett.108.032005

Web End =Phys. Rev. Lett. 108 (2012) 032005 [arXiv:1111.1733] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1111.1733

Web End =INSPIRE ].

[41] G. Ossola, C.G. Papadopoulos and R. Pittau, CutTools: a program implementing the OPP reduction method to compute one-loop amplitudes, http://dx.doi.org/10.1088/1126-6708/2008/03/042

Web End =JHEP 03 (2008) 042 [arXiv:0711.3596] [http://inspirehep.net/search?p=find+EPRINT+arXiv:0711.3596

Web End =INSPIRE ].

[42] P. Mastrolia, G. Ossola, T. Reiter and F. Tramontano, Scattering amplitudes from unitarity-based reduction algorithm at the integrand-level, http://dx.doi.org/10.1007/JHEP08(2010)080

Web End =JHEP 08 (2010) 080 [arXiv:1006.0710] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1006.0710

Web End =INSPIRE ].

[43] G. Heinrich, G. Ossola, T. Reiter and F. Tramontano, Tensorial reconstruction at the integrand level, http://dx.doi.org/10.1007/JHEP10(2010)105

Web End =JHEP 10 (2010) 105 [arXiv:1008.2441] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1008.2441

Web End =INSPIRE ].

27

JHEP04(2013)037

[44] R. Pittau, Testing and improving the numerical accuracy of the NLO predictions, http://dx.doi.org/10.1016/j.cpc.2010.09.013

Web End =Comput. Phys. Commun. 181 (2010) 1941 [arXiv:1006.3773] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1006.3773

Web End =INSPIRE ].

[45] A. Ferroglia, M. Passera, G. Passarino and S. Uccirati, All purpose numerical evaluation of one loop multileg Feynman diagrams, http://dx.doi.org/10.1016/S0550-3213(02)01070-2

Web End =Nucl. Phys. B 650 (2003) 162 [http://arxiv.org/abs/hep-ph/0209219

Web End =hep-ph/0209219 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/0209219

Web End =INSPIRE ].

[46] A. Denner and S. Dittmaier, Reduction of one loop tensor ve point integrals, http://dx.doi.org/10.1016/S0550-3213(03)00184-6

Web End =Nucl. Phys. B 658 (2003) 175 [http://arxiv.org/abs/hep-ph/0212259

Web End =hep-ph/0212259 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/0212259

Web End =INSPIRE ].

[47] T. Binoth, J.P. Guillet, G. Heinrich, E. Pilon and C. Schubert, An algebraic/numerical formalism for one-loop multi-leg amplitudes, http://dx.doi.org/10.1088/1126-6708/2005/10/015

Web End =JHEP 10 (2005) 015 [http://arxiv.org/abs/hep-ph/0504267

Web End =hep-ph/0504267 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/0504267

Web End =INSPIRE ].

[48] A. Denner and S. Dittmaier, Reduction schemes for one-loop tensor integrals, http://dx.doi.org/10.1016/j.nuclphysb.2005.11.007

Web End =Nucl. Phys. B 734 (2006) 62 [http://arxiv.org/abs/hep-ph/0509141

Web End =hep-ph/0509141 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/0509141

Web End =INSPIRE ].

[49] A. Bredenstein, A. Denner, S. Dittmaier and M. Weber, Radiative corrections to the semileptonic and hadronic Higgs-boson decays H W W/ZZ 4 fermions,

http://dx.doi.org/10.1088/1126-6708/2007/02/080

Web End =JHEP 02 (2007) 080 [http://arxiv.org/abs/hep-ph/0611234

Web End =hep-ph/0611234 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/0611234

Web End =INSPIRE ].

[50] A. Denner, S. Dittmaier, T. Kasprzik and A. Mck, Electroweak corrections to W + jet hadroproduction including leptonic W-boson decays, http://dx.doi.org/10.1088/1126-6708/2009/08/075

Web End =JHEP 08 (2009) 075 [arXiv:0906.1656] [http://inspirehep.net/search?p=find+EPRINT+arXiv:0906.1656

Web End =INSPIRE ].

[51] A. Denner, S. Dittmaier, T. Kasprzik and A. Mck, Electroweak corrections to dilepton + jet production at hadron colliders, http://dx.doi.org/10.1007/JHEP06(2011)069

Web End =JHEP 06 (2011) 069 [arXiv:1103.0914] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1103.0914

Web End =INSPIRE ].

[52] A. Denner, S. Dittmaier, S. Kallweit and A. Mck, Electroweak corrections to Higgs-strahlung o W/Z bosons at the Tevatron and the LHC with HAWK, http://dx.doi.org/10.1007/JHEP03(2012)075

Web End =JHEP 03 (2012) 075 [arXiv:1112.5142] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1112.5142

Web End =INSPIRE ].

[53] A. Bredenstein, A. Denner, S. Dittmaier and S. Pozzorini, NLO QCD corrections to pp ttb

b+ at the LHC, http://dx.doi.org/10.1103/PhysRevLett.103.012002

Web End =Phys. Rev. Lett. 103 (2009) 012002 [arXiv:0905.0110] [

[54] A. Denner, S. Dittmaier, S. Kallweit and S. Pozzorini, NLO QCD corrections to W W bb production at hadron colliders, http://dx.doi.org/10.1103/PhysRevLett.106.052001

Web End =Phys. Rev. Lett. 106 (2011) 052001 [arXiv:1012.3975] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1012.3975

Web End =INSPIRE ].

[55] F. Campanario, C. Englert, M. Rauch and D. Zeppenfeld, Precise predictions for W +jet production at hadron colliders, http://dx.doi.org/10.1016/j.physletb.2011.09.072

Web End =Phys. Lett. B 704 (2011) 515 [arXiv:1106.4009] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1106.4009

Web End =INSPIRE ].

[56] L. Reina and T. Schutzmeier, Towards W bb + j at NLO with an automatized approach to one-loop computations, http://dx.doi.org/10.1007/JHEP09(2012)119

Web End =JHEP 09 (2012) 119 [arXiv:1110.4438] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1110.4438

Web End =INSPIRE ].

[57] F. Cascioli, P. Maierhofer and S. Pozzorini, Scattering amplitudes with open loops, http://dx.doi.org/10.1103/PhysRevLett.108.111601

Web End =Phys. Rev. Lett. 108 (2012) 111601 [arXiv:1111.5206] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1111.5206

Web End =INSPIRE ].

[58] A. van Hameren, Multi-gluon one-loop amplitudes using tensor integrals, http://dx.doi.org/10.1088/1126-6708/2009/07/088

Web End =JHEP 07 (2009) 088 [arXiv:0905.1005] [http://inspirehep.net/search?p=find+EPRINT+arXiv:0905.1005

Web End =INSPIRE ].

[59] F.A. Berends and W. Giele, Recursive calculations for processes with N gluons, http://dx.doi.org/10.1016/0550-3213(88)90442-7

Web End =Nucl. Phys. B 306 (1988) 759 [http://inspirehep.net/search?p=find+J+Nucl.Phys.,B306,759

Web End =INSPIRE ].

[60] J.M. Campbell and R.K. Ellis, Next-to-leading order corrections to W + 2 jet and Z + 2 jet production at hadron colliders, http://dx.doi.org/10.1103/PhysRevD.65.113007

Web End =Phys. Rev. D 65 (2002) 113007 [http://arxiv.org/abs/hep-ph/0202176

Web End =hep-ph/0202176 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/0202176

Web End =INSPIRE ].

[61] J.M. Campbell, R.K. Ellis and D.L. Rainwater, Next-to-leading order QCD predictions for W + 2 jet and Z + 2 jet production at the CERN LHC, http://dx.doi.org/10.1103/PhysRevD.68.094021

Web End =Phys. Rev. D 68 (2003) 094021 [http://arxiv.org/abs/hep-ph/0308195

Web End =hep-ph/0308195 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/0308195

Web End =INSPIRE ].

28

JHEP04(2013)037

http://inspirehep.net/search?p=find+EPRINT+arXiv:0905.0110

Web End =INSPIRE ].

[62] C. Oleari and D. Zeppenfeld, QCD corrections to electroweak ljj and +jj production, http://dx.doi.org/10.1103/PhysRevD.69.093004

Web End =Phys. Rev. D 69 (2004) 093004 [http://arxiv.org/abs/hep-ph/0310156

Web End =hep-ph/0310156 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/0310156

Web End =INSPIRE ].

[63] F. Dyson, The S matrix in quantum electrodynamics, http://dx.doi.org/10.1103/PhysRev.75.1736

Web End =Phys. Rev. 75 (1949) 1736 [http://inspirehep.net/search?p=find+J+Phys.Rev.,75,1736

Web End =INSPIRE ].

[64] J.S. Schwinger, On the Greens functions of quantized elds. 1., Proc. Nat. Acad. Sci. 37 (1951) 452 [http://inspirehep.net/search?p=find+J+Proc.Nat.Acad.Sci.,37,452

Web End =INSPIRE ].

[65] J.S. Schwinger, On the Greens functions of quantized elds. 2., Proc. Nat. Acad. Sci. 37 (1951) 455 [http://inspirehep.net/search?p=find+J+Proc.Nat.Acad.Sci.,37,455

Web End =INSPIRE ].

[66] A. Kanaki and C.G. Papadopoulos, HELAC: a package to compute electroweak helicity amplitudes, http://dx.doi.org/10.1016/S0010-4655(00)00151-X

Web End =Comput. Phys. Commun. 132 (2000) 306 [http://arxiv.org/abs/hep-ph/0002082

Web End =hep-ph/0002082 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/0002082

Web End =INSPIRE ].

[67] C.G. Papadopoulos and M. Worek, Multi-parton cross sections at hadron colliders, http://dx.doi.org/10.1140/epjc/s10052-007-0246-2

Web End =Eur. Phys. J. C 50 (2007) 843 [http://arxiv.org/abs/hep-ph/0512150

Web End =hep-ph/0512150 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/0512150

Web End =INSPIRE ].

[68] A. Cafarella, C.G. Papadopoulos and M. Worek, Helac-Phegas: a generator for all parton level processes, http://dx.doi.org/10.1016/j.cpc.2009.04.023

Web End =Comput. Phys. Commun. 180 (2009) 1941 [arXiv:0710.2427] [http://inspirehep.net/search?p=find+EPRINT+arXiv:0710.2427

Web End =INSPIRE ].

[69] K. Hagiwara and D. Zeppenfeld, Amplitudes for multiparton processes involving a current at e+e, ep and hadron colliders, http://dx.doi.org/10.1016/0550-3213(89)90397-0

Web End =Nucl. Phys. B 313 (1989) 560 [http://inspirehep.net/search?p=find+J+Nucl.Phys.,B313,560

Web End =INSPIRE ].

[70] F. Caravaglios and M. Moretti, An algorithm to compute Born scattering amplitudes without Feynman graphs, http://dx.doi.org/10.1016/0370-2693(95)00971-M

Web End =Phys. Lett. B 358 (1995) 332 [http://arxiv.org/abs/hep-ph/9507237

Web End =hep-ph/9507237 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/9507237

Web End =INSPIRE ].

[71] G. Ossola, C.G. Papadopoulos and R. Pittau, On the rational terms of the one-loop amplitudes, http://dx.doi.org/10.1088/1126-6708/2008/05/004

Web End =JHEP 05 (2008) 004 [arXiv:0802.1876] [http://inspirehep.net/search?p=find+EPRINT+arXiv:0802.1876

Web End =INSPIRE ].

[72] P. Draggiotis, M. Garzelli, C. Papadopoulos and R. Pittau, Feynman rules for the rational part of the QCD 1-loop amplitudes, http://dx.doi.org/10.1088/1126-6708/2009/04/072

Web End =JHEP 04 (2009) 072 [arXiv:0903.0356] [http://inspirehep.net/search?p=find+EPRINT+arXiv:0903.0356

Web End =INSPIRE ].

[73] M. Garzelli, I. Malamos and R. Pittau, Feynman rules for the rational part of the electroweak 1-loop amplitudes, http://dx.doi.org/10.1007/JHEP01(2010)040

Web End =JHEP 01 (2010) 040 [Erratum ibid. 1010 (2010) 097] [arXiv:0910.3130] [http://inspirehep.net/search?p=find+EPRINT+arXiv:0910.3130

Web End =INSPIRE ].

[74] H.-S. Shao, Y.-J. Zhang and K.-T. Chao, Feynman rules for the rational part of the standard model one-loop amplitudes in the t Hooft-Veltman 5 scheme, http://dx.doi.org/10.1007/JHEP09(2011)048

Web End =JHEP 09 (2011) 048 [arXiv:1106.5030] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1106.5030

Web End =INSPIRE ].

[75] A. Denner, Techniques for calculation of electroweak radiative corrections at the one loop level and results for W physics at LEP-200, Fortsch. Phys. 41 (1993) 307 [arXiv:0709.1075] [http://inspirehep.net/search?p=find+EPRINT+arXiv:0709.1075

Web End =INSPIRE ].

[76] A. Denner, S. Dittmaier, M. Roth and D. Wackeroth, Predictions for all processes e+e 4 fermions +, http://dx.doi.org/10.1016/S0550-3213(99)00437-X

Web End =Nucl. Phys. B 560 (1999) 33 [http://arxiv.org/abs/hep-ph/9904472

Web End =hep-ph/9904472 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/9904472

Web End =INSPIRE ].

[77] A. Denner, S. Dittmaier, M. Roth and L. Wieders, Electroweak corrections to charged-current e+e 4 fermion processes: technical details and further results,

http://dx.doi.org/10.1016/j.nuclphysb.2011.09.001

Web End =Nucl. Phys. B 724 (2005) 247 [Erratum ibid. B 854 (2012) 504-507] [http://arxiv.org/abs/hep-ph/0505042

Web End =hep-ph/0505042 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/0505042

Web End =INSPIRE ].

[78] Z. Bern and D.A. Kosower, Color decomposition of one loop amplitudes in gauge theories, http://dx.doi.org/10.1016/0550-3213(91)90567-H

Web End =Nucl. Phys. B 362 (1991) 389 [http://inspirehep.net/search?p=find+J+Nucl.Phys.,B362,389

Web End =INSPIRE ].

[79] G. t Hooft, A planar diagram theory for strong interactions, http://dx.doi.org/10.1016/0550-3213(74)90154-0

Web End =Nucl. Phys. B 72 (1974) 461 [ http://inspirehep.net/search?p=find+J+Nucl.Phys.,B72,461

Web End =INSPIRE ].

[80] A. Kanaki and C.G. Papadopoulos, HELAC-PHEGAS: automatic computation of helicity amplitudes and cross-sections, http://arxiv.org/abs/hep-ph/0012004

Web End =hep-ph/0012004 [http://inspirehep.net/search?p=find+EPRINT+hep-ph/0012004

Web End =INSPIRE ].

29

JHEP04(2013)037

[81] F. Maltoni, K. Paul, T. Stelzer and S. Willenbrock, Color ow decomposition of QCD amplitudes, http://dx.doi.org/10.1103/PhysRevD.67.014026

Web End =Phys. Rev. D 67 (2003) 014026 [http://arxiv.org/abs/hep-ph/0209271

Web End =hep-ph/0209271 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/0209271

Web End =INSPIRE ].

[82] A. Denner and S. Dittmaier, The complex-mass scheme for perturbative calculations with unstable particles, http://dx.doi.org/10.1016/j.nuclphysBPS.2006.09.025

Web End =Nucl. Phys. Proc. Suppl. 160 (2006) 22 [http://arxiv.org/abs/hep-ph/0605312

Web End =hep-ph/0605312 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/0605312

Web End =INSPIRE ].

[83] D.Y. Bardin, A. Leike, T. Riemann and M. Sachwitz, Energy dependent width e ects in e+e annihilation near the Z boson pole, http://dx.doi.org/10.1016/0370-2693(88)91625-5

Web End =Phys. Lett. B 206 (1988) 539 [http://inspirehep.net/search?p=find+J+Phys.Lett.,B206,539

Web End =INSPIRE ].

[84] W. Beenakker and A. Denner, Infrared divergent scalar box integrals with applications in the electroweak standard model, http://dx.doi.org/10.1016/0550-3213(90)90636-R

Web End =Nucl. Phys. B 338 (1990) 349 [http://inspirehep.net/search?p=find+J+Nucl.Phys.,B338,349

Web End =INSPIRE ].

[85] A. Denner, U. Nierste and R. Scharf, A compact expression for the scalar one loop four point function, http://dx.doi.org/10.1016/0550-3213(91)90011-L

Web End =Nucl. Phys. B 367 (1991) 637 [http://inspirehep.net/search?p=find+J+Nucl.Phys.,B367,637

Web End =INSPIRE ].

[86] A. Denner and S. Dittmaier, Scalar one-loop 4-point integrals, http://dx.doi.org/10.1016/j.nuclphysb.2010.11.002

Web End =Nucl. Phys. B 844 (2011) 199 [arXiv:1005.2076] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1005.2076

Web End =INSPIRE ].

[87] S. Dittmaier and M. Kramer, Electroweak radiative corrections to W boson production at hadron colliders, http://dx.doi.org/10.1103/PhysRevD.65.073007

Web End =Phys. Rev. D 65 (2002) 073007 [http://arxiv.org/abs/hep-ph/0109062

Web End =hep-ph/0109062 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/0109062

Web End =INSPIRE ].

[88] S. Catani and M. Seymour, A general algorithm for calculating jet cross-sections in NLO QCD, http://dx.doi.org/10.1016/S0550-3213(96)00589-5

Web End =Nucl. Phys. B 485 (1997) 291 [Erratum ibid. B 510 (1998) 503-504] [http://arxiv.org/abs/hep-ph/9605323

Web End =hep-ph/9605323 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/9605323

Web End =INSPIRE ].

[89] A. Denner, S. Dittmaier, T. Gehrmann and C. Kurz, Electroweak corrections to hadronic event shapes and jet production in e+e annihilation, http://dx.doi.org/10.1016/j.nuclphysb.2010.04.009

Web End =Nucl. Phys. B 836 (2010) 37 [arXiv:1003.0986] [http://inspirehep.net/search?p=find+EPRINT+arXiv:1003.0986

Web End =INSPIRE ].

[90] E.N. Glover and A. Morgan, Measuring the photon fragmentation function at LEP,http://dx.doi.org/10.1007/BF01560245

Web End =Z. Phys. C 62 (1994) 311 [http://inspirehep.net/search?p=find+J+Z.Physik,C62,311

Web End =INSPIRE ].

[91] ALEPH collaboration, D. Buskulic et al., First measurement of the quark to photon fragmentation function, http://dx.doi.org/10.1007/s002880050037

Web End =Z. Phys. C 69 (1996) 365 [http://inspirehep.net/search?p=find+J+Z.Physik,C69,365

Web End =INSPIRE ].

[92] A. Denner, S. Dittmaier and L. Hofer, COLLIER, a Complex One-Loop Library In Extended Regularizations, in preparation.

[93] T. Motz, Generic Monte Carlo event generator for LHC processes, PhD Thesis, ETH, Zrich (2011).

[94] T. Hahn, Generating Feynman diagrams and amplitudes with FeynArts 3, http://dx.doi.org/10.1016/S0010-4655(01)00290-9

Web End =Comput. Phys. Commun. 140 (2001) 418 [http://arxiv.org/abs/hep-ph/0012260

Web End =hep-ph/0012260 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/0012260

Web End =INSPIRE ].

[95] T. Hahn and C. Schappacher, The implementation of the minimal supersymmetric standard model in FeynArts and FormCalc, http://dx.doi.org/10.1016/S0010-4655(01)00436-2

Web End =Comput. Phys. Commun. 143 (2002) 54 [http://arxiv.org/abs/hep-ph/0105349

Web End =hep-ph/0105349 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/0105349

Web End =INSPIRE ].

[96] T. Hahn and M. Prez-Victoria, Automatized one loop calculations in four-dimensions and d-dimensions, http://dx.doi.org/10.1016/S0010-4655(98)00173-8

Web End =Comput. Phys. Commun. 118 (1999) 153 [http://arxiv.org/abs/hep-ph/9807565

Web End =hep-ph/9807565 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/9807565

Web End =INSPIRE ].

[97] S. Dittmaier, Weyl-van der Waerden formalism for helicity amplitudes of massive particles, http://dx.doi.org/10.1103/PhysRevD.59.016007

Web End =Phys. Rev. D 59 (1998) 016007 [http://arxiv.org/abs/hep-ph/9805445

Web End =hep-ph/9805445 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/9805445

Web End =INSPIRE ].

[98] E. Accomando, A. Denner and C. Meier, Electroweak corrections to W and Z production at the LHC, http://dx.doi.org/10.1140/epjc/s2006-02521-y

Web End =Eur. Phys. J. C 47 (2006) 125 [http://arxiv.org/abs/hep-ph/0509234

Web End =hep-ph/0509234 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/0509234

Web End =INSPIRE ].

[99] S. Dittmaier and M. Roth, LUSIFER: a LUcid approach to SIx FERmion production, http://dx.doi.org/10.1016/S0550-3213(02)00640-5

Web End =Nucl. Phys. B 642 (2002) 307 [http://arxiv.org/abs/hep-ph/0206070

Web End =hep-ph/0206070 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/0206070

Web End =INSPIRE ].

30

JHEP04(2013)037

[100] Particle Data Group collaboration, J. Beringer et al., Review of particle physics (RPP),http://dx.doi.org/10.1103/PhysRevD.86.010001

Web End =Phys. Rev. D 86 (2012) 010001 [http://inspirehep.net/search?p=find+J+Phys.Rev.,D86,010001

Web End =INSPIRE ].

[101] Tevatron Electroweak Working Group, CDF Collaboration, D0 collaboration,Combination of CDF and D0 results on the mass of the top quark using up to 5.8 fb1 of

data, arXiv:1107.5255 [http://inspirehep.net/search?p=find+EPRINT+arXiv:1107.5255

Web End =INSPIRE ].

[102] A. Martin, W. Stirling, R. Thorne and G. Watt, Parton distributions for the LHC,http://dx.doi.org/10.1140/epjc/s10052-009-1072-5

Web End =Eur. Phys. J. C 63 (2009) 189 [arXiv:0901.0002] [http://inspirehep.net/search?p=find+EPRINT+arXiv:0901.0002

Web End =INSPIRE ].

[103] M. Cacciari, G.P. Salam and G. Soyez, The kt jet clustering algorithm, http://dx.doi.org/10.1088/1126-6708/2008/04/063

Web End =JHEP 04 (2008) 063 [arXiv:0802.1189] [http://inspirehep.net/search?p=find+EPRINT+arXiv:0802.1189

Web End =INSPIRE ].

[104] P. Ciafaloni and D. Comelli, Sudakov enhancement of electroweak corrections,http://dx.doi.org/10.1016/S0370-2693(98)01541-X

Web End =Phys. Lett. B 446 (1999) 278 [http://arxiv.org/abs/hep-ph/9809321

Web End =hep-ph/9809321 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/9809321

Web End =INSPIRE ].

[105] J.H. Khn and A. Penin, Sudakov logarithms in electroweak processes, http://arxiv.org/abs/hep-ph/9906545

Web End =hep-ph/9906545[ http://inspirehep.net/search?p=find+EPRINT+hep-ph/9906545

Web End =INSPIRE ].

[106] V.S. Fadin, L. Lipatov, A.D. Martin and M. Melles, Resummation of double logarithms in electroweak high-energy processes, http://dx.doi.org/10.1103/PhysRevD.61.094002

Web End =Phys. Rev. D 61 (2000) 094002 [http://arxiv.org/abs/hep-ph/9910338

Web End =hep-ph/9910338 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/9910338

Web End =INSPIRE ].

[107] A. Denner and S. Pozzorini, One loop leading logarithms in electroweak radiative corrections. 1. Results, http://dx.doi.org/10.1007/s100520100551

Web End =Eur. Phys. J. C 18 (2001) 461 [http://arxiv.org/abs/hep-ph/0010201

Web End =hep-ph/0010201 ] [http://inspirehep.net/search?p=find+EPRINT+hep-ph/0010201

Web End =INSPIRE ].

JHEP04(2013)037

31