Hidden Valley Processes
This Hidden Valley (HV) scenario has been developed specifically
to allow the study of visible consequences of radiation in a
hidden sector, by recoil effect. A key aspect is therefore that
the normal timelike showering machinery has been expanded with a
third kind of radiation, in addition to the QCD and QED ones.
These three kinds of radiation are fully interleaved, i.e.
evolution occurs in a common pT-ordered sequence.
The scenario is described in [Car10]. Furthermore
hadronization in the hidden sector has been implemented.
Three main scenarios for production into and decay out of the
hidden sector can be compared, in each case either for an
Abelian or a non-Abelian gauge group in the HV. For further details
see [Car11].
Warning: do not use the alternative hadronization
scenarios StringPT:thermalModel = on
or
StringPT:mT2suppression = on
in conjunction with
Hidden Valley processes, since they have not been constructed to
work together.
Particle content and properties
For simplicity we assume that the HV contains an unbroken SU(N)
gauge symmetry. This is used in the calculation of production cross
sections. These could be rescaled by hand for other gauge groups.
mode
HiddenValley:Ngauge
(default = 3
; minimum = 1
)
is U(1) for Ngauge = 1
, is SU(N) if
Ngauge > 1
. Note that pair production cross sections
contains a factor of Ngauge
for new particles
in the fundamental representation of this group.
A minimal HV particle content has been introduced. Firstly, there is
a set of 12 particles that mirrors the Standard Model flavour
structure, and is charged under both the SM and the HV symmetry groups.
Each new particle couples flavour-diagonally to a corresponding SM
state, and has the same SM charge and colour, but in addition is in
the fundamental representation of the HV colour, as follows:
Dv
, identity 4900001, partner to the normal
d
quark;
Uv
, identity 4900002, partner to the normal
u
quark;
Sv
, identity 4900003, partner to the normal
s
quark;
Cv
, identity 4900004, partner to the normal
c
quark;
Bv
, identity 4900005, partner to the normal
b
quark;
Tv
, identity 4900006, partner to the normal
t
quark;
Ev
, identity 4900011, partner to the normal
e
lepton;
nuEv
, identity 4900012, partner to the normal
nue
neutrino;
MUv
, identity 4900013, partner to the normal
mu
lepton;
nuMUv
, identity 4900014, partner to the normal
numu
neutrino;
TAUv
, identity 4900015, partner to the normal
tau
lepton;
nuTAUv
, identity 4900016, partner to the normal
nutau
neutrino.
Collectively we will refer to these states as Fv
;
note, however, that they need not be fermions themselves.
In addition the model contains the HV gauge particle, either
a HV-gluon or a HV-photon, but not both; see Ngauge
above:
gv
, identity 4900021, is the massless
gauge boson of the HV SU(N) group;
gammav
, identity 4900022, is the massless
gauge boson of the HV U(1) group.
Finally, for the basic HV scenario, there is a new massive particle
with only HV charge sitting in the fundamental representation of the
HV gauge group:
qv
, identity 4900101.
The typical scenario would be for pair production of one of the
states presented first above, e.g. g g → Dv Dvbar.
Such a Dv can radiate gluons and photons like an SM quark,
but in addition HV-gluons or HV-photons in a similar fashion.
Eventually the Dv will decay like Dv → d + qv.
The strength of this decay is not set as such, but is implicit in
your choice of width for the Dv state. Thereafter the
d and qv can radiate further within their
respective sectors. The qv, gv or gammav
are invisible, so their fate need not be considered further.
While not part of the standard scenario, as an alternative there is
also a kind of Z' resonance:
Zv
, identity 4900023, a boson that can couple
both to pairs of Standard Model fermions and to qv qvbar
pairs. Mass, total width and branching ratios can be set as convenient.
This opens up for alternative processes
l^+l^-, q qbar → Zv → qv qvbar.
The possibility of a leakage back from the hidden sector will be
considered in the Hadronization section below. For the U(1)
case the gammav acquires a mass and can decay back to a
Standard-Model fermion pair, while the qv remains invisible.
The SU(N) alternative remains unbroken, so confinement holds
and the gv is massless. A string like
qv - gv - ... - gv - qvbar can break by the production of
new qv - qvbar pairs, which will produce qv-qvbar
mesons. It would be possible to build a rather sophisticated hidden
sector by trivial extensions of the HV flavour content. For now,
however, the qv can be duplicated in up to eight copies
with the same properties except for the flavour charge. These are
assigned codes 4900101 - 4900108. This gives a total of 64 possible
lowest-lying mesons. We also include a duplication of that, into two
multiplets, corresponding to the pseudoscalar and vector mesons of
QCD. For now, again, these are assumed to have the same mass and
other properties. By default, only the flavour-diagonal ones can
decay back into the Standard-Model sector, however, while the rest
remains in the hidden sector. It is therefore only necessary to
distinguish a few states:
pivDiag
, identity 4900111, a flavour-diagonal
HV-meson with spin 0 that can decay back into the Standard-Model sector;
rhovDiag
, identity 4900113, a flavour-diagonal
HV-meson with spin 1 that can decay back into the Standard-Model sector;
pivUp
, identity 4900211, an off-diagonal
HV-meson with spin 0 that is stable and invisible, with an antiparticle
pivDn
with identity -4900211; the particle is
the one where the code of the flavour is larger than that of the
antiflavour;
rhovUp
, identity 4900213, an off-diagonal
HV-meson with spin 1 that is stable and invisible, with an antiparticle
rhovDn
with identity -4900213; again the particle is
the one where the code of the flavour is larger than that of the
antiflavour;
ggv
, identity 4900991, is only rarely used,
to handle cases where it is kinematically impossible to produce an
HV-meson on shell, and it therefore is assumed to de-excite by the
emission of invisible gv-gv v-glueball bound states.
By changing ParticleData, it is possible to allow decays even for the
off-diagonal Up/Dn mesons.
Only the spin of the HV-gluon or HV-photon is determined unambiguously
to be unity, for the others you can make your choice.
mode
HiddenValley:spinFv
(default = 1
; minimum = 0
; maximum = 2
)
The spin of the HV partners of the SM fermions, e.g.
Dv, Uv, Ev and nuEv.
option
0 : spin 0.
option
1 : spin 1/2.
option
2 : spin 1.
mode
HiddenValley:spinqv
(default = 0
; minimum = 0
; maximum = 1
)
The spin of qv when the Fv (the HV partners of
the SM fermions) have spin 1/2. (While, if they have spin 0 or 1,
the qv spin is fixed at 1/2.)
option
0 : spin 0.
option
1 : spin 1.
parm
HiddenValley:kappa
(default = 1.
)
If the Fv have spin 1 then their production
cross section depends on the presence of anomalous magnetic dipole
moment, i.e. of a kappa different from unity. For other spins
this parameter is not used.
flag
HiddenValley:doKinMix
(default = off
)
allow kinematic mixing or not.
parm
HiddenValley:kinMix
(default = 1.
)
strength of kinetic mixing.
You should set the Fv and qv masses appropriately,
with the latter smaller than the former two to allow decays.
When U(1) hadronization is switched on, you need to set the
gammav mass and decay modes. For SU(N) hadronization
the HV-meson masses should be set to match the qv ones.
The simplest is to assume that m_qv defines a constituent
mass, so that m_HVmeson = 2 m_qv. The hvMesonDiag
decay modes also need to be set.
Production processes
flag
HiddenValley:all
(default = off
)
Common switch for the group of all hard Hidden Valley processes,
as listed separately in the following.
flag
HiddenValley:gg2DvDvbar
(default = off
)
Pair production g g → Dv Dvbar.
Code 4901.
flag
HiddenValley:gg2UvUvbar
(default = off
)
Pair production g g → Uv Uvbar.
Code 4902.
flag
HiddenValley:gg2SvSvbar
(default = off
)
Pair production g g → Sv Svbar.
Code 4903.
flag
HiddenValley:gg2CvCvbar
(default = off
)
Pair production g g → Cv Cvbar.
Code 4904.
flag
HiddenValley:gg2BvBvbar
(default = off
)
Pair production g g → Bv Bvbar.
Code 4905.
flag
HiddenValley:gg2TvTvbar
(default = off
)
Pair production g g → Tv Tvbar.
Code 4906.
flag
HiddenValley:qqbar2DvDvbar
(default = off
)
Pair production q qbar → Dv Dvbar
via intermediate gluon.
Code 4911.
flag
HiddenValley:qqbar2UvUvbar
(default = off
)
Pair production q qbar → Uv Uvbar
via intermediate gluon.
Code 4912.
flag
HiddenValley:qqbar2SvSvbar
(default = off
)
Pair production q qbar → Sv Svbar
via intermediate gluon.
Code 4913.
flag
HiddenValley:qqbar2CvCvbar
(default = off
)
Pair production q qbar → Cv Cvbar
via intermediate gluon.
Code 4914.
flag
HiddenValley:qqbar2BvBvbar
(default = off
)
Pair production q qbar → Bv Bvbar
via intermediate gluon.
Code 4915.
flag
HiddenValley:qqbar2TvTvbar
(default = off
)
Pair production q qbar → Tv Tvbar
via intermediate gluon.
Code 4916.
flag
HiddenValley:ffbar2DvDvbar
(default = off
)
Pair production f fbar → Dv Dvbar
via intermediate gamma*/Z^*.
Code 4921.
flag
HiddenValley:ffbar2UvUvbar
(default = off
)
Pair production f fbar → Uv Uvbar
via intermediate gamma*/Z^*.
Code 4922.
flag
HiddenValley:ffbar2SvSvbar
(default = off
)
Pair production f fbar → Sv Svbar
via intermediate gamma*/Z^*.
Code 4923.
flag
HiddenValley:ffbar2CvCvbar
(default = off
)
Pair production f fbar → Cv Cvbar
via intermediate gamma*/Z^*.
Code 4924.
flag
HiddenValley:ffbar2BvBvbar
(default = off
)
Pair production f fbar → Bv Bvbar
via intermediate gamma*/Z^*.
Code 4925.
flag
HiddenValley:ffbar2TvTvbar
(default = off
)
Pair production f fbar → Tv Tvbar
via intermediate gamma*/Z^*.
Code 4926.
flag
HiddenValley:ffbar2EvEvbar
(default = off
)
Pair production f fbar → Ev Evbar
via intermediate gamma*/Z^*.
Code 4931.
flag
HiddenValley:ffbar2nuEvnuEvbar
(default = off
)
Pair production f fbar → nuEv nuEvbar
via intermediate gamma*/Z^*.
Code 4932.
flag
HiddenValley:ffbar2MUvMUvbar
(default = off
)
Pair production f fbar → MUv MUvbar
via intermediate gamma*/Z^*.
Code 4933.
flag
HiddenValley:ffbar2nuMUvnuMUvbar
(default = off
)
Pair production f fbar → nuMUv nuMUvbar
via intermediate gamma*/Z^*.
Code 4934.
flag
HiddenValley:ffbar2TAUvTAUvbar
(default = off
)
Pair production f fbar → TAUv TAUvbar
via intermediate gamma*/Z^*.
Code 4935.
flag
HiddenValley:ffbar2nuTAUvnuTAUvbar
(default = off
)
Pair production f fbar → nuTAUv nuTAUvbar
via intermediate gamma*/Z^*.
Code 4936.
flag
HiddenValley:ffbar2Zv
(default = off
)
Production f fbar → Zv where Zv is a generic
resonance that couples both SM fermion pairs and a qv qvbar
pair. Not part of the framework of the above processes, but as an
alternative. Code 4941.
Timelike showers
One key point of this HV scenario is that radiation off the
HV-charged particles is allowed. This is done by the standard
final-state showering machinery. (HV particles are not produced
in initial-state radiation.) All the (anti)particles Fv
and qv have one (negative) unit of HV charge. That is,
radiation closely mimics the one in QCD. Both QCD, QED and HV
radiation are interleaved in one common sequence of decreasing
emission pT scales. Each radiation kind defines a set of
dipoles, usually spanned between a radiating parton and its recoil
partner, such that the invariant mass of the pair is not changed
when a radiation occurs. This need not follow from trivial colour
assignments, but is often obvious. For instance, in a decay
Qv → q + qv the QCD dipole is between the q and
the hole after Qv, but qv becomes the recoiler
should a radiation occur, while the role of q and qv
is reversed for HV radiation.
This also includes matrix-element corrections for a number
of decay processes, with colour, spin and mass effects included
[Nor01]. They were calculated within the context of the
particle content of the MSSM, however, which does not include spin 1
particles with unit colour charge. In such cases spin 0 is assumed
instead. By experience, the main effects come from mass and colour
flow anyway, so this is not a bad approximation. (Furthermore the
MSSM formulae allow for gamma_5 factors from wave
functions or vertices; these are even less important.)
An emitted gv can branch in its turn,
gv → gv + gv. This radiation may affect momenta
in the visible sector by recoil effect, but this is a minor
effect relative to the primary emission of the gv.
While the default model has a fixed Hidden Valley coupling
alpha_HV, some further work [Scw15] has considered
the impact of a running coupling. This is included as an option.
flag
HiddenValley:FSR
(default = off
)
switch on final-state shower of gv or gammav
in a HV production process.
mode
HiddenValley:alphaOrder
(default = 0
; minimum = 0
; maximum = 1
)
Order at which alpha_HV runs,
option
0 : zeroth order, i.e. alpha_HV is kept
fixed at the value alphaFSR.
option
1 : first order, with the beta function based
on Ngauge and Nflav.
parm
HiddenValley:alphaFSR
(default = 0.1
; minimum = 0.0
)
fixed alpha scale of gv/gammav emission; corresponds to
alpha_strong of QCD or alpha_em of QED. For
shower branchings such as Dv → Dv + gv the coupling is
multiplied by C_F = (N^2 - 1) / (2 * N) for an
SU(N) group and for gv → gv + gv by N.
parm
HiddenValley:Lambda
(default = 0.4
; minimum = 0.1
)
the parameter used for the case of a running (first order) alpha_HV.
parm
HiddenValley:pTminFSR
(default = 0.44
; minimum = 0.1
)
lowest allowed pT of emission. Should be greater than
or equal to 1.1 times Lambda, or it will be reset
automatically.
Hadronization
By default the HV particles with no Standard Model couplings
are not visible. Their presence can only be deduced by the
observation of missing (transverse) momentum in the event as a
whole. In the current implementation it is possible to simulate
two different scenarios where activity can leak back from the
hidden sector.
The first possibility is relevant for the U(1) scenario.
The U(1) group may be broken, so that the gammav
acquires a mass. Furthermore, the gammav may have a
small mixing angle with the normal photon, or with some Z'
state or other mediator, and may thus decay back into Standard
Model particles. The qv still escapes undetected;
recall that there is no confinement in the U(1) option.
In order to enable this machinery two commands are necessary,
4900022:m0 = ...
to set the gammav mass
to the desired value, and 4900022:onMode = on
to enable
gammav decays. The default gammav decay
table contains all Standard Model fermion-antifermion pairs,
except top, with branching ratios in proportion to their coupling
to the photon, whenever the production channel is allowed by
kinematics. This table could easily be tailored to more specific
models and needs. For instance, for a mass below 1 - 2 GeV, it
would make sense to construct a table of exclusive hadronic decay
channels rather than go the way via a hadronizing quark pair.
The gammav are expected to decay so rapidly that no
secondary vertex will be detectable. However, it is possible to
set 4900022:tau0
to a finite lifetime (in mm) that
will be used to create separated secondary vertices.
The second, more interesting, possibility is relevant for the
SU(N) scenarios. Here the gauge group remains unbroken, i.e.
gv is massless, and the partons are confined. Like in
QCD, the HV-partons can therefore be arranged in one single
HV-colour-ordered chain, with a qv in one end, a
qvbar in the other, and a varying number of
gv in between. Each event will only contain (at most)
one such string, (i) since perturbative branchings
gv → qv qvbar have been neglected, as is a reasonable
approximation for QCD, and (ii) since HV-colours are assigned in the
N_C → infinity limit, just like in the handling of
string fragmentation in QCD. The HV-string can then fragment by the
nonperturbative creation of qv qvbar pairs, leading to
the formation of HV-mesons along the string, each with its
qv from one vertex and its qvbar from
the neighbouring one.
Since, so far, we have only assumed there to be one qv
species, all produced qv qvbar HV-mesons are of the
same flavour-diagonal species. Such an HV-meson can decay back to
the normal sector, typically by whatever mediator particle allowed
production in the first place. In this framework the full energy put
into the HV sector will leak back to the normal one. To allow more
flexibility, an ad hoc possibility of n_Flav different
qv species is introduced. For now they are all assumed
to have the same mass and other properties, but distinguished by
some flavour-like property. Only the flavour-diagonal ones can decay,
meaning that only a fraction (approximately) 1/n_Flav of the
HV-energy leaks back, while the rest remains in the hidden sector.
This scenario contains more parameters than the first one, for the
U(1) group. They can be subdivided into two sets. One is
related to particle properties, both for qv and for the
two different kinds of HV-mesons, here labeled 4900111 and 4900113
for the diagonal ones, and +-4900211 and +-4900213 for the
off-diagonal ones. It makes sense to set the HV-meson masses to be
twice the qv one, as in a simple constituent mass context.
Furthermore the hvMesonDiag decay modes need to be set up.
Like with the
gammav in the U(1) option, the default decay table
is based on the branching ratios of an off-shell photon.
The second set are fragmentation parameters that extend or replace
the ones used in normal string fragmentation. Some of them are not
encoded in the same way as normally, however, but rather scale as
the qv mass is changed, so as to keep a sensible default
behaviour. This does not mean that deviations from this set should
not be explored, or that other scaling rules could be preferred
within alternative scenarios. These parameters are as follows.
flag
HiddenValley:fragment
(default = off
)
switch on string fragmentation of the HV partonic system.
Only relevant for SU(N) scenarios.
mode
HiddenValley:nFlav
(default = 1
; minimum = 1
; maximum = 8
)
number of different flavours assumed to exist in the hadronization
description, leading to approximately 1/n_Flav of the
produced HV-mesons being flavour-diagonal and capable to decay back
to Standard Model particles.
parm
HiddenValley:probVector
(default = 0.75
; minimum = 0.
; maximum = 1.
)
fraction of HV-mesons that are assigned spin 1 (vector), with the
remainder spin 0 (pseudoscalar). Assuming the qv have
spin 1/2 and the mass splitting is small, spin counting
predicts that 3/4 of the mesons should have spin 1.
parm
HiddenValley:aLund
(default = 0.3
; minimum = 0.0
; maximum = 2.0
)
The a parameter of the Lund symmetric fragmentation function.
See the normal fragmentation
function description for the shape of this function.
parm
HiddenValley:bmqv2
(default = 0.8
; minimum = 0.2
; maximum = 2.0
)
The b parameter of the Lund symmetric fragmentation function,
multiplied by the square of the qv mass. This scaling ensures
that the fragmentation function keeps the same shape when the
qv mass is changed (neglecting transverse momenta).
parm
HiddenValley:rFactqv
(default = 1.0
; minimum = 0.0
; maximum = 2.0
)
r_qv, i.e. the Bowler correction factor to the Lund symmetric
fragmentation function, which could be made weaker or stronger than
its natural value.
parm
HiddenValley:sigmamqv
(default = 0.5
; minimum = 0.0
)
the width sigma of transverse momenta in the HV fragmentation
process, normalized to the qv mass. This ensures that
sigma scales proportionately to m_qv.
See the normal fragmentation
pT description for conventions for factors of 2.