PUBLICATION Constraints on the FCC

CERN-ACC-TEST–2015-002
Future Circular Collider
PUBLICATION
Constraints on the FCC-ee Lattice from the
Compatibility with the FCC Hadron
Collider
Haerer, B (CERN) et al.
28 January 2015
The research leading to this document is part of the Future Circular Collider Study
The electronic version of this FCC Publication is available
on the CERN Document Server at the following URL :
<http://cds.cern.ch/record/1982444
CERN-ACC-TEST–2015-002
CONSTRAINTS ON THE FCC-EE LATTICE FROM THE
COMPATIBILITY WITH THE FCC HADRON COLLIDER
B. Haerer∗ , CERN, Geneva, Switzerland, KIT, Karlsruhe, Germany,
W. Bartmann, M. Benedikt, B. J. Holzer, J. A. Osborne, D. Schulte, R. Tomas,
J. Wenninger, F. Zimmermann, CERN, Geneva, Switzerland
M. J. Syphers, MSU, East Lansing, Michigan, USA
U. Wienands, SLAC, Menlo Park, California, USA
Abstract
Following the recommendations of the European Strategy
Group for High Energy Physics, CERN launched the Future
Circular Collider Study (FCC), a design study for possible
future circular collider projects to investigate their feasibility
for high energy physics research. The FCC Study covers
three different machines with a circumference of 100 km:
an electron positron collider with collision energies in the
range of 90 GeV to 350 GeV (FCC- ee), a proton proton collider with a maximum energy of 100 TeV (FCC-hh) and an
electron proton option (FCC-he). This paper will present
the constraints on the design of the FCC-ee lattice and optics from geometry and lattice considerations of the hadron
machine.
INTRODUCTION
With the discovery of a Higgs boson all particles of the
standard model of particle physics have been found. In order
to discover new physics CERN started to study a future discovery machine called FCC-hh with proton proton collisions
at 100 TeV center of mass energy. Considerations presented
in this paper will show that such a machine will need to have
a circumference of 80 km-100 km given by the achievable
technology. Having this tunnel available it is obvious to think
about an electron positron collider for precision measurements as well [1]. The large circumference allows operation
with an acceptable amount of synchrotron radiation losses
and the costs for a second machine decrease drastically, as
no extra tunnel has to be built. However straight sections
for RF installation have to be provided to deal with the synchrotron radiation loss in such a storage ring. This part of
the design study, earlier known as TLEP, is called FCC-ee.
The third part of the study, FCC-he, covers the investigation
of future electron proton collisions in order to study deep
inelastic scattering. This comprises two options: a LHeC
like linac-ring option and, in case FCC-hh and FCC-ee can
be hosted and operated in the tunnel at the same time, a
ring-ring option. Each machine of the FCC study has special requirements, that have to be considered in the design
phase. This paper focuses on the constraints on the FCC-ee
lattice design from the compatibility with FCC-hh. Contrary
to FCC-ee, for a beam energy of 50 TeV in the hadron machine a new magnet technology has to be developed. The
maximum bending radius in the arcs and consequently the
∗
bastian.harer@cern.ch
circumference of the machine directly depends on the achievable magnetic field. The length of the long straight sections
needed for insertions also contributes to the circumference.
They must provide enough space to house RF installation,
collimators, kickers for injection and beam dump and the
detectors. If LHC is used as an injector, the circumference
and harmonic number of FCC should be rational multiples
of the LHC’s to allow bunch to bucket transfer. Furthermore
the FCC-hh and LHC tunnels should be close to each other
to guarantee a reasonable length of the transfer lines. For
locating a 100 km circular collider also geologic aspects play
a major role. The constraints arising from the requirement
of hosting both machines in the tunnel at the same time and
from the compatibility with FCC-he are not covered in this
paper.
BENDING RADIUS AND
CIRCUMFERENCE
The beam rigidity of a 50 TeV proton beam is
B ρ = p/e ≈ 1.67 × 105 Tm.
(1)
To bend such a stiff beam in a reasonable radius a new technology of superconducting magnets needs to be developed.
A prototype dipole based on Nb3 Sn technology could reach
a magnetic field of B =16 T [2]. Such a magnet would
define a bending radius of ρ = 10.7 km. If even higher magnetic fields of B =20 T could be achieved, the bending radius
could be reduced to ρ =8.5 km. Assuming 16 T magnets and
67 % of the whole circumference including long straight section being occupied by bending magnets the circumference
C would approximately be 100 km. As mentioned before,
if LHC is used as an injector, the circumference of FCC
should be a multiple of the LHC circumference, which is
26.66 km [3]. For 16 T magnets approximately 106.64 km
should be taken as circumference and 79.98 km for the 20 T
version. Both possibilities are studied, the final choice will
depend on the technical progress in magnet technology.
Table 1: Circumference and bending radius for different
magnetic fields of the bending magents
B in T
16
20
ρ in km
C in km
10.7
8.5
106.64
79.98
LAYOUT OBJECTIVES AND SHAPE OF
THE MACHINE
The maximum momentum of the particle beam in a
hadron machine is limited by the bending radius of the
dipoles. Therefore the design aims to maximize the integral
magnetic dipole field around the machine
Laboratoire Européen pour la Physique de Particules
European Laboratory for Particle Physics
GS/SE-DOP CH-1211 GENEVE 23 -Tél: central: +41 (22) 767 6111 - direct 41 (22) 767 3414
FCC
Définition de l'Emprise d'Etude Géologique
1
Emprise CERN
2
(2)
100
k
ra
ce
km
10
0
Practically that means highest possible dipole fill factor and
avoiding sections without full strength bending magnets as
far as possible. This includes straight sections but missing
bend or half-bend dispersion suppressors as well. The energy
in a lepton machine on the contrary is limited by synchrotron
radiation. The energy loss per turn is given by [4]
γ4
4π
α c
U0 =
3
ρ
/
Etabli à partir du
Système d'Information
Géographique
du CERN
Emprise GADZ
tra
ck
B(s) ds.
ORGANISATION EUROPEENNE POUR LA RECHERCHE NUCLEAIRE
EUROPEAN ORGANISATION FOR NUCLEAR RESEARCH
83
km
ci r
cul
LOCATION OF THE FCC TUNNEL
A variety of boundary conditions has to be considered in
context of a possible location of FCC.
m ra
cetr
ack
1 0 0k
m circular
6
k
83
83km racetrack
(3)
The lost energy needs to be fed back to the beam in straight
sections with RF installation. Those cavities must be positioned in dispersion free sections to avoid coupling between
the longitudinal and transversal planes. The design for a
lepton machine not only aims for maximum dipole fill factor
in order to maximize the bending radius and minimize the
radiation power, but contrary to a hadron machine also for
a high number of straight RF sections to narrow down the
energy sawtooth of the orbit. This is important because particles with large energy deviations will move on dispersion
orbits with large amplitude and cross strong non-linear fields
or even get lost while hitting the geometric aperture. Those
layout requests end up in different shapes of the machine:
in case of a hadron machine, the minimal number of dispersion suppressors is given by a racetrack design, where
all infrastructure like injection, ejection, RF and interaction
regions is concentrated in two very long straight sections
which are connected by two arcs. In case of a lepton machine a circular design with equally distributed straight RF
sections is preferred. The more dispersion free RF sections
are provided the smaller is the energy sawtooth.
In general it is useful to ensure a phase advance of
(n + 1/2)π with n being an integer between two interaction points to correct higher order chromaticity. In a racetrack like geometry with clustered interaction regions the
preservation of the phase advance is easier than in a circular
layout because of the smaller distance between those points
(e. g. SSC design [5]). Still measurements at LHC have
shown, that the phase advance can be kept stable between the
two high-luminosity experiments ATLAS and CMS, which
have a distance of ca. 13.3 km. So clustering the interaction
points is not a compulsory requirement from an optics point
of view.
a rL H C
m
rac
a
etr
5
100km circular
100km racetrack
Dessiné par : ETL
83km circular
LHC
83km racetrack
Date: 17.04.2014
3
4
Légende
Machine Scenarios :
ck
GADZ
CERN
Points
1
2
3
4
5
6
0
Emprise GADZ
Long
Lat
6.041411
46.333368
6.355130
46.060480
6.306633
6.193496
6.008318
5.883819
5'000
46.287224
45.952495
45.968384
46.157249
10'000
Points
1
2
3
4
5
6
Emprise CERN
Long
6.049043
6.345873
6.355130
6.193496
6.008318
5.866667
Lat
46.354992
46.313785
46.060480
45.952495
45.968384
46.156926
20'000
Meters
Figure 1: The geological study area with some proposals for
the location for both circular and racetrack shaped machines
with the circumference 80 km and 100 km (Defined by the
FCC Civil Engineering Group chaired by J. Osborne)
Geology
Building a tunnel with the length of 100 km is a civil
engineering challenge. The geology of the area plays a
major role and must be studied carefully before deciding the
exact location. The geological study area defined by the FCC
Civil Engineering Group is shown in figure 1. Its borders
are defined by the surrounding area: the Pre-Alps in the
East and South-East, the Jura Mountains in the West, Mount
Vuache in the South and Lake Geneva to the North. The
mountains consist mostly of Limestone, with some Karstic
features, making the risk of encountering water very high.
Tunnel excavation in this type of rock can be complicated
and expensive due to stored water in the rocks. The depth
of the access shafts needs to be minimized to facilitate the
installation of services and optimize the safety paths. As
far as possible, the tunnel should be housed in the Molasse
Rock, a sedimentary rock made up of sandstone and marls.
Tilting the tunnel
For finding an appropriate location for the future FCC
tunnel it is considered to tilt the tunnel plane. This was
already done when the LHC tunnel was built for LEP. The
median plane was tilted with respect to the horizontal by
1.42 % [6] to maximize the tunnel extent in Molasse Rock
environment and to minimize the depth of the access shafts.
In case of FCC such a tilt could allow passing Lake Geneva
further in the North without increasing the depth of the
access shafts to the Southern part of the machine.
Location relative to LHC
With the LHC, CERN already has a high energy hadron
machine available that could be used as injector for FCC-hh.
Below 3.5 TeV the power converters and the cooling system
allow to ramp the LHC bending magnets with up to 50 A/s.
Combined with its double ring layout LHC would be a very
efficient injector delivering up to 2800 bunches per beam.
To keep the transfer lines reasonably short the FCC tunnel
should overlap or cross one LHC straight section. Nevertheless there is a minimum horizontal distance between the
LHC extraction point and a possible FCC injection point,
which is necessary to overcome the difference in depth. Due
to maintenance and engineering reasons the slope of the
transfer line tunnel should not be larger than 5 %. Thus the
required distance depends on the beam energy, the magnet
technology used and the difference in depth. Figure 2 shows
this dependency for a beam energy 3.5 TeV and a dipole filling factor of 0.75, which leaves enough space for machine
protection elements. The different lines correspond to different vertical distance between the LHC and the FCC tunnels,
which is labeled on the right. Obviously the transfer line
must be longer, if the vertical distance of the tunnels is larger
or the magnetic field of the dipoles is smaller. So choosing normal conducting magnets will elongate the transfer
line. Assuming a difference of depth of less than 200 m and
reasonable dipole fields the LHC extraction point and FCC
injection point need to have a distance between about 0.5 km
and 1.5 km.
Beam energy: 3.5 TeV
Filling factor: 0.75
1.8 T TI 2/8 dipoles
2.6 T CF JPARC magnets
3.9 T Tevatron dipoles
8.3 T LHC dipoles
Figure 2: The plot shows the required horizontal distance
between the extraction point at LHC and a future injection
point at FCC-hh ring as a function of the magnetic field of
the used bending magnets. Each line stands for a certain
vertical distance of the two tunnels, which is labeled on the
bottom right of the plot.
LENGTH OF LONG STRAIGHT
SECTIONS
Long straight sections are used for various purposes: RF
installation, injection, beam dump, collimation and mini-
beta insertions for the experiments. For the injection of preaccelerated beams from the arriving transfer line to the main
ring, septum magnets are used followed by kicker magnets.
Additionally machine protection elements will be needed
to prevent damage in case of failure. For the injection of a
3.3 TeV proton beam into FCC-hh these group of elements
will require space with the length of 600 m, corresponding to
about three FODO cells, that will have to be modified. The
extraction to the beam dump line uses the same scheme, but
it has to be laid out for up to 50 TeV beam energy. Though
longer kicker rise times can be accepted, more space will be
necessary. First estimates predict a length between 800 m
and 1000 m. For the collimation system two 2.8 km long
straight sections are foreseen.
INTERACTION REGION
The biggest experiment at LHC is the ATLAS detector
with about 45 m length and 25 m height. However for particle detection at 100 TeV center of mass energy even larger detectors are needed, because the particle jets penetrate deeper
before the particles are absorbed in the calorimeter. The current FCC-hh interaction region design therefore assumes an
L ∗ of 46 m, which corresponds to a 92 m long drift space for
the detector. The length of the complete FCC-hh interaction
region including matching sections and dispersion suppressors is approximately 1100 m long. The interaction region
layout for FCC-ee is completely different: In the FCC-hh
final focus system a very tiny crossing angle of ca. 70 µrad
is used. In the FCC-ee design the ambitious request for the
vertical beta function at the interaction point β ∗ = 1 mm
needs to be combined with a small L ∗ of 2 m to prevent
the beta function from growing to unreasonable size. To
minimize synchrotron radiation the two beams must be focused by separated quadrupoles, thus the crossing angle is
defined by the quadrupole’s aperture and the size of the coils.
The CERN interaction region design is based on a crossing
angle of 11 mrad, while the final focus system designed by
the Budker Institute of Nuclear Physics (BINP) is based
upon 30 mrad [7]. In combination with a local chromaticity
correction scheme this leads to different geometries of the
interaction regions, that are shown in figure 3. While in
the FCC-hh design the two beams are separated by 40 cm
in the matching sections, the ones for FCC-ee have a distance up to 2 m. Including matching sections and dispersion
suppressors the FCC-ee interaction region will probably be
longer than the one for FCC-hh. So the dimensions of the
straight sections next to the experiments will be defined by
the requests of FCC-ee.
CONCLUSION
The compatibility of the FCC-ee lattice with FCC-hh requires the same layout and geometry given by the tunnel.
While the magnetic fields for FCC-ee can easily be obtained
(see Table 2), the circumference and the bending angle for
the FCC-hh machine are defined by the achievable magnetic
fields. The length of the straight sections depends on the re-
0.2
Beam 1
Beam 2
x [m]
0.1
0
FCC-hh
-0.1
-0.2
2
0
100
200
300
400
500
600
700
1
x [m]
800
e+e
0
FCC-ee, BINP design
-1
-2
2
0
1
x [m]
100
200
300
400
500
600
700
800
e+e
0
FCC-ee, CERN design
-1
-2
0
100
200
300
400
s [m]
500
600
700
800
Figure 3: Comparison of the geometry of the interaction region designs for FCC-hh and FCC-ee [7]
Table 2: List of baseline parameters for FCC-hh and the different physics programs of FCC-ee ( [8], [9])
Beam energy (GeV)
Circumference (km)
Dipole field
Arc filling factor
Number of IPs
Peak luminosity (1034 cm−2 s−1 )
Betatron function at IP β ∗ (m)
- Horizontal
- Vertical
FCC-hh
FCC-ee Z
FCC-ee W
FCC-ee H
FCC-ee tt
50000
100 (80)
16 (20)
0.79
2+2
5
45.5
100
0.014
0.84
4
28.0
80
100
0.024
0.84
4
12.0
120
100
0.036
0.84
4
5.9
175
100
0.053
0.84
4
1.2
1.1
1.1
0.5
0.001
0.5
0.001
0.5
0.001
1
0.001
quirements for the injection, extraction, collimation regions.
While the size of the caverns housing the experiments is
defined by the FCC-hh detectors, the straight sections for
the final focus system must provide enough space for the
geometry of the FCC-ee design. The possible location of a
future FCC tunnel depends on geological aspects but also
on the already existing infrastructure. The LHC would be
a very effective pre-accelerator for FCC-hh, so a possible
injection point at FCC-hh should be located close to a LHC
straight section but keep a minimum distance to overcome
the difference in depth.
Especially in this early stage of the FCC study it is important to consider the compatibility of the sub-studies with
each other to ensure a successful development of the project.
ACKNOWLEDGMENT
This work is supported by the German Bundesministerium
für Bildung und Forschung (BMBF).
REFERENCES
[1] TLEP Design Study group: "Referee report: Answers and
actions taken", December, 2013
[2] LBNL Superconducting Magnet Program Newsletter Issue
No. 2, "HD-1 Sets New Dipole Field Record", LBNL, Berkeley,
CA, USA, October 2003
[3] LHC Design Report, (Geneva: CERN, 2004)
[4] M. Sands, The Physics of Electron Storage Rings. An Introduction, SLAC-121, 1970
[5] SSCL, "Conceptual Design of the Superconducting Super Collider", SSC-SR-2020, 1986
[6] LEP Design Report, (Geneva: CERN, 1984)
[7] R. Martin et al., "Status of the FCC-ee Interaction Region
Design", These Proceedings, HF2014, Beijing, China (2014)
[8] Future Circular Collider Study, "Hadron Collider Parameters",
FCC-1401101315, CERN, 2014
[9] Future Circular Collider Study, "Lepton Collider Paramters",
FCC-1401201640, CERN, 2014