没有合适的资源?快使用搜索试试~ 我知道了~
温馨提示
试读
24页
通过使用微分方程,我们分析性地计算了出现在CP偶数\重的夸克的排他性和衰变的NNLO QCD校正中的两环主积分的椭圆扇形,这证明是相关计算中的最后也是最困难的部分 。 发现这些积分可以表示为Goncharov多对数和椭圆函数上的迭代积分。 主积分可以应用于其他一些关于重夸克排他性排他生产的NNLO QCD计算,例如γ*γ→QQ¯$$ {\ gamma} ^ {\ ast} \ gamma \ to Q \ overline {Q} $$,e + e−→γ+QQ¯$$ {e} ^ {+} {e} ^ {-} \至\ gamma + Q \ overline {Q} $$,和H / Z0→γ+QQ¯$$ H / {Z} ^ 0 \至\ gamma + Q \ overline {Q} $$,则重质石英专用包衰变,以及CP甚至重质石英包含的包容衰变。
资源推荐
资源详情
资源评论
JHEP04(2018)080
Published for SISSA by Springer
Received: January 14, 2018
Revised: April 8, 2018
Accepted: April 9, 2018
Published: April 13, 2018
Two-loop integrals for CP-even heavy quarkonium
production and decays: elliptic sectors
Long-Bin Chen,
a
Jun Jiang
b
and Cong-Feng Qiao
b,c,1
a
School of Physics & Electronic Engineering, Guangzhou University,
Guangzhou 510006, China
b
School of Physics, University of Chinese Academy of Sciences,
YuQuan Road 19A, Beijing 100049, China
c
CAS Center for Excellence in Particle Physics,
Beijing 100049, China
Abstract: By employing the differential equations, we compute analytically the elliptic
sectors of two-loop master integrals appearing in the NNLO QCD corrections to CP-even
heavy quarkonium exclusive production and decays, which turns out to be the last and
toughest part in the relevant calculation. The integrals are found can be expressed as Gon-
charov polylogarithms and iterative integrals over elliptic functions. The master integrals
may be applied to some other NNLO QCD calculations about heavy quarkonium exclusive
production, like γ
∗
γ → Q
¯
Q, e
+
e
−
→ γ + Q
¯
Q, and H/Z
0
→ γ + Q
¯
Q, heavy quarkonium
exclusive decays, and also the CP-even heavy quarkonium inclusive production and decays.
Keywords: Heavy Quark Physics, Perturbative QCD
ArXiv ePrint: 1712.03516
1
Corresp onding author.
Open Access,
c
The Authors.
Article funded by SCOAP
3
.
https://doi.org/10.1007/JHEP04(2018)080
JHEP04(2018)080
Contents
1 Introduction 1
2 Notation and kinematics 2
3 Iterated integrals and complete elliptic integrals 3
4 Elliptic integral sectors 4
4.1 Sector I: integrals with massive sunrise integrals as subtopology 5
4.2 Sector II: non-planar two-loop three-point integrals 12
4.3 Analytic continuation and discussions 14
5 Conclusions and outlooks 14
A The definition for integrals 15
B The typical analytical results 16
1 Introduction
Precision physics in colliders requires more higher-order corrections in perturbation theory.
Unravelling the mathematical structure of Feynman integrals in multiloop calculation is
somehow critical to handle the complexity of higher order calculations and may help us to
obtain a better control of the perturbative expansion. In recent years, the corresponding
research achieved some breakthroughs and becomes now one of the hot topics in physics
and mathematics.
One of the powerful methods to evaluate the master integrals analytically attributes to
the differential equation [1–5]. With recent developments [6–10], this method becomes now
a prevailing one in tackling those integrals unsolvable before. It was noticed by Henn that
generically in multi-loop calculation, choosing a set of suitable basis for master integrals can
greatly simplify the corresponding differential equations [6], which can be calculated iter-
atively in dimensional regularization scheme. In light of this proposal, many of multi-loop
Feynman integrals for various phenomenological processes have been calculated [11–23].
Note, some Feynman integrals in two-loop or higher order possess new mathematical struc-
tures [24–31], which cannot be expressed as multiple polylogarithms and ask for different
technique to deal with. A typical example is the massive two-loop sunrise integral, which
has been studied intensively [32–40].
The heavy quarkonium production and decay are one of the hot topics in particle
physics ever since the first discovery in 1974, especially with the advent of Nonrelativistic
Quantum Chromodynamics (NRQCD) factorization formalism [41]. Up to date there still
– 1 –
JHEP04(2018)080
γ
∗
γ
γ
∗
γ
Figure 1. Typical two-loop Feynman diagrams for CP-even heavy quarkonium production.
exist some discrepancies between experimental data and theoretical expectations [42–45],
which appeal for precision calculations. In one of our previous works [46] we gave out a set
of 86 two-loop master integrals about heavy quarkonium production and decay, which can
be cast into the canonical form and expressed in terms of multiple polylogarithms. However,
for those Feynman integrals with functions beyond the realm of multiple polylogarithms the
calculation is not done yet. In fact, to date, only a limited number of similar calculations
have been performed in the literature.
In this work, we calculate analytically all remaining integrals with different mathemat-
ical structures from multiple polylogarithms in CP-even heavy quarkonium production and
decays. The master integrals will be classified into two sectors, one with integrals contain-
ing sub-topologies related to the two-loop massive sunrise integrals and the other involving
non-planar two-loop three-point integrals. Following the strategy suggested in ref. [39] and
with properly chosen basis, we cast the differential equations of those integrals in the first
sector into a proper form that can be solved recursively. Of the second sector, the key
point is to find the homogeneous solutions for the second-order differential equations of the
two-loop non-planar three-point massive integrals, with that the full solutions can then be
obtained by constant variation.
The paper is organized as follows. In section 2, the kinematics is discussed and the
derivatives with respect to kinematic variables will be given. In section 3, the iterative
integrals and complete elliptic integrals are introduced. In section 4, the elliptic type
integrals will be separated into two sectors, and the calculation procedure for them will
be elucidated respectively. For illustration, specific examples will be given. Section 5
is remained for conclusions and outlooks. The definition of master integrals is given in
appendix A, and several simple but typical analytical results are presented in appendix B.
2 Notation and kinematics
The heavy quarkonium exclusive production in electron-positron collision has a relatively
low background, and has played an important role in the study of quarkonium production
mechanism. Here we calculate the CP-even quarkonium production in two correlated
processes, that is in γ
∗
γ collision and in electron-position annihilation associated with
a photon,
γ
∗
(k
1
) + γ(k
2
) → Q(k
q
)
¯
Q(k
¯q
) , (2.1)
γ
∗
(k
1
) → Q(k
q
)
¯
Q(k
¯q
) + γ(k
2
) , (2.2)
– 2 –
JHEP04(2018)080
where k
2
1
= 2ss, k
2
2
= 0 and k
2
q
= k
2
¯q
= m
2
q
. The typical Feynman diagrams are showed in
figure 1. The process (2.1) is in Euclidean region with ss < 0, and the momenta satisfy
the following relations
(k
1
+ k
2
)
2
= (k
q
+ k
¯q
)
2
= 4m
2
q
. (2.3)
Whereas, the process (2.2) is in Minkowski region with 2ss > 4m
2
q
, and
(k
1
− k
2
)
2
= (k
q
+ k
¯q
)
2
= 4m
2
q
. (2.4)
Note, in the threshold expansion approach, quark and anti-quark momenta are taken to
be equal, i.e. k
q
= k
¯q
.
In order to express the results compactly, here we introduce three dimensionless vari-
ables x, y and z as follows:
ss
m
2
q
= −
(1 − x)
2
2x
= (y + 2) = (z + 1) . (2.5)
The NNLO QCD corrections to processes (2.1) and (2.2) are calculated in light of
Feynman diagrams. As a routine, with some algebraic manipulations, the amplitudes can
be reduced to a set of scalar integrals. We use the Mathematica package FIRE [47–49]
to reduce the scalar integrals to a minimum set of independent master integrals. The
calculation of these master integrals is the central issue, and normally turns out to be
a nontrivial work. In our calculation, we apply the method of differential equations to
calculate the master integrals.
The first step of deriving differential equations is taking derivatives of the Lorentz in-
variant kinematic variables, and expressing them as linear combinations of master integrals.
The FIRE is also employed in the derivation of differential equations. The derivatives of
the external momenta can be expressed as the derivatives of ss and m
2
q
, like
k
i
·
∂
∂k
j
= k
i
·
∂ss
∂k
j
∂
∂ss
+ k
i
·
∂m
2
q
∂k
j
∂
∂m
2
q
(2.6)
with i(j) = 1 or 2. And in reverse, the derivative
∂
∂ss
can be expressed as a linear combi-
nation of derivatives k
i
·
∂
∂k
j
, i.e.,
2ss
∂
∂ss
= k
1
·
∂
∂k
1
+
ss + 2m
2
q
ss − 2m
2
q
!
k
2
·
∂
∂k
2
. (2.7)
The derivative transform can be readily obtained according to equation (2.5). With the
variables chosen in above, analytical results of the integrals can then be formulated in a
compact form, in terms of iterative integrals and elliptic integrals.
3 Iterated integrals and complete elliptic integrals
The Goncharov polylogarithms (GPLs) [50] are defined as
G
a
1
,a
2
,...,a
n
(x) ≡
Z
x
0
dt
t − a
1
G
a
2
,...,a
n
(x) , (3.1)
G
−→
0
n
(x) ≡
1
n!
log
n
x , (3.2)
– 3 –
JHEP04(2018)080
which in fact are special cases of a more general type of integrals, named Chen-iterated
integrals [51]. If all indices a
i
belong to set {0, ±1}, the Goncharov polylogarithms can
then be transformed into the well-known Harmonic polylogarithms (HPLs) [52]
H
−→
0
n
(x) = G
−→
0
n
(x) , (3.3)
H
a
1
,a
2
,...,a
n
(x) = (−1)
k
G
a
1
,a
2
,...,a
n
(x) , (3.4)
where k equals to the number of times the element (+1) appearing in (a
1
, a
2
, . . . , a
n
) .
The GPLs satisfy the following shuffle rules:
G
a
1
,...,a
m
(x)G
b
1
,...,b
n
(x) =
X
c∈aXb
G
c
1
,c
2
,...,c
m+n
(x) . (3.5)
In above equation, aXb is composed of the shuffle products of a
i
(i = 1, 2 . . . m) and b
i
(i =
1, 2 . . . n), which is defined as the set of lists containing all elements of a
i
and b
i
, with the
order of elements a
i
and b
i
preserved. The GPLs and HPLs can be numerically evaluated
by implementing the GINAC [53, 54], and the Mathematica package HPL [55, 56] is
applicable to the HPLs reduction and evaluation. Both GPLs and HPLs can be transformed
into functions ln, Li
n
and Li
22
up to weight four in light of the method described in ref. [57].
In our calculation, the complete elliptic integrals are necessary to express the integrals
encountered. The first and second kinds of complete elliptic integrals are defined as
K(x) =
Z
1
0
dt
p
(1 − t
2
)(1 − x t
2
)
(3.6)
and
E(x) =
Z
1
0
√
1 − x t
2
√
1 − t
2
dt . (3.7)
They satisfy the following derivative relations:
dK(x)
dx
=
E(x) − (1 − x)K(x)
2(1 − x)x
,
dE(x)
dx
=
E(x) − K(x)
2x
. (3.8)
The Legendre relation is useful in simplifying the complete elliptic integrals, i.e.,
K(x)K(1 − x) − K(x)E(1 − x) − E(x)K(1 − x) = −
π
2
. (3.9)
4 Elliptic integral sectors
The symbols and canonical basis in the calculation of elliptic integrals keep the same as
in the preceding work [46], where the linear differential equations can be expressed, via a
suitable basis choice of master integrals, as canonical form [6]
d F = (d A) F (4.1)
– 4 –
剩余23页未读,继续阅读
资源评论
NEDL003
- 粉丝: 160
- 资源: 978
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功