Beam Collimation and Shielding in the Fermilab Proton
Driver
A!. Drozhdin, N.V. Mokhov
FNAL, BataviaJL 60510, USA
Abstract.
A high beam power in the proposed Fermilab Proton Drivers - 1.2 MW in 16-GeV PD-I and 0.48 MW in
8-GeV PD-II - implies serious constraints on beam losses in these machines. Only with a very efficient beam
collimation system can one reduce uncontrolled beam losses in the machine to an allowable level. The entire
complex must be well shielded to allow acceptable hands-on maintenance conditions in the tunnel and a noncontrolled access to the outside shielding at normal operation and accidental beam loss. Collimation and shielding
performances are calculated and compared for both Proton Drivers.
BEAM LOSS AND SHIELDING
STRATEGY
The Proton Driver design strategy is that the beam losses
are localized and controlled as much as possible via the
dedicated beam collimation system. This way, the source
term for the radiation analysis is a derivative of the collimation system performance. A high loss rate is localized in the collimation section with components locally
shielded to equalize prompt and residual radiation levels in the tunnel and drastically lower uncontrolled beam
loss rates in the rest of the lattice [2, 3]. The radiation
transport analysis is fundamentally important because of
the impact on machine performance, conventional facility design, maintenance operations, and related costs. Results of this paper are based on detailed Monte Carlo simulations with the STRUCT [4] and MARS [5] codes.
The Fermilab regulatory requirements imply that: 1)
prompt dose rate in non-controlled areas on accessible
outside surfaces of the shield is <0.05 mrem/hr at normal
operation and <1 mrem/hr for the worst case due to
accidents; radionuclide concentration of 20 pCi/ml for
3
H and 0.4 pCi/ml for 22Na in any nearby drinking water
supplies are not exceeded; and anywhere in the machine,
residual dose rate Py<100 mrem/hr=l mSv/hr at 30 cm
from the component surface, after 100 day irradiation at
4 hrs after shutdown (averaged over all the components,
P7 < 10-20 mrem/hr=0.1-0.2 mSv/hr.
The radiation analysis for the arcs and long straight
sections is performed both for normal operation and for
accidental beam loss. The maximum shielding thickness
from the both cases is put into the design as the tunnel
shielding in that part of the machine. In normal operation, the source term is based on the beam loss distributions calculated with a beam collimation system described in the next section. For accidental beam loss, a
credible accident is considered: a point-like loss of 0.1%
of the full beam intensity during one hour. This is about
1015 protons. Once such an accident happens, the machine is shut down within 1 second to analyze the cause
and undertake appropriate measures.
COLLIMATION SYSTEM
Assuming 1% of beam loss at the top energy, one gets
total beam power of 11.5 kW and 4.8 kW lost in PDI and PD-II, respectively. Calculations show that the
peaks (at some quadrupoles) in the beam loss distribution can reach several kW/m that is a few thousand times
higher than the tolerable levels [2, 3]. Therefore, multicomponent beam collimation systems are designed for
the projects. In PD-I, the system is located in a dedicated
long straight section, while in the 8 GeV machine, due to
space constrains, it is placed in the drifts available in the
beginning of the arc. The systems consist of horizontal
and vertical primary collimators, and several secondary
and supplementary collimators as shown in Fig. 1. Secondary collimators generate out-scattered particles lost
later in the lattice. This component is reduced with a 3stage collimation system positioning three (PD-I) or four
(PD-II) secondary collimators close to the beam to deal
with protons scattered in the primary collimator and five
(PD-I) or two (PD-II) supplementary collimators farther
from the beam to catch particles out-scattered from the
main secondary collimators.
Secondary collimators need to be placed at phase advances which are optimal to intercept most of particles out-scattered from the primary collimators during
the first turns after the halo interaction with the primary collimator. The optimal phase advances are around
k - n ± 30°. The horizontal and vertical primary collimators are placed at the edge of the beam after painting, with secondary collimators father from this position
by an offset d. Beam loss distributions at injection and
CP642, High Intensity and High Brightness Hadron Beams: 20th ICFA Advanced Beam Dynamics Workshop on
High Intensity and High Brightness Hadron Beams, edited by W. Chou, Y. Mori, D. Neuffer, and J.-F. Ostiguy
© 2002 American Institute of Physics 0-7354-0097-0/02/$ 19.00
180
secondary collimators
ofiiifflrs
50
100 150 200 250 300 350 400 450
10
20
30
40
-H+f^l^
tifMbwJM^
0
100
200
300
400
500
600
EMJ=UnJ J L
.IJ.J . I. ub.[|
700
FIGURE2.
2. Beam
Beamloss
loss atatinjection
injectionin
inPD-n
PD-IIatat0.6
0.6GeV
GeV(top)
(top)
FIGURE
and PD-I
PD-I at
at0.4
0.4 GeV
GeV (bottom).
(bottom).
and
prH/" VI
1e+06
100000
\
10
20
30
40
Path length, m
50
1e+06
100000
10000
10000
60
10
100 150 200 250 300 350 400 450
20
30
40
50
60
70
Path length, m
Path length, m
n lUrzyrzj
1e+06
10000
160
180
200
220
240
Path length, m
tl [jnijnj 'j L jmiij n'. i^ i.ri
FIGURE 3.
3. Beam
Beam loss
loss in
in PD-II
PD-II atat 88 GeV
GeV (top)
(top)and
andPD-I
PD-Iatat
FIGURE
16
16 GeV
GeV (bottom).
(bottom).
With
With the
the proposed
proposed system,
system, ~99%
∼99% of
of the
the beam
beam halo
halo
isis intercepted
intercepted in
in the
the collimation
collimation section.
section. About
About 1%
1% isis
lost
lost in
in the
the rest
rest of
of the
the machine
machine with
with mean
mean rate
rate of
of 0.2
0.2
and
and 0.12
0.12W/m
W/min
inPD-I
PD-Iand
andPD-II,
PD-II,respectively.
respectively.At
Atseveral
several
locations
locations the
the beam
beam loss
loss isis noticeably
noticeably higher
higher (~2
(∼2 W/m),
W/m),
exceeding
exceeding the
the tolerable
tolerablerates.
rates. Such
Such"hot"
“hot”locations
locationsneed
need
special
special care.
care. Beam
Beam loss
loss rates
rates in
in the
the collimation
collimation system
system
section
section itself
itself are
are very
very high
high requiring
requiring aa special
special shielding
shielding
design.
design.
A
A practicality
practicality of
of aa rapid
rapid cycling
cycling proton
proton synchrotron
synchrotron
dictates
dictates aa stationary
stationarycollimator
collimatorapproach
approachwith
withcollimator
collimator
jaws
jaws in
in aa fixed
fixed position
position with
with respect
respect to
to the
the beam
beam orbit
orbit
during
during the
the entire
entire cycle.
cycle. In
In an
an ideal
ideal case,
case, the
the circulating
circulating
beam
beam should
should be
be kept
kept close
close to
to the
the collimators
collimatorsedge
edgedurduring
the
cycle.
This
requires
rather
complicated
ing the cycle. This requires rather complicatedhorizonhorizontal
tal and
and vertical
vertical bumps,
bumps, created
created by
by several
several fast
fastmagnets
magnets
for
foreach
eachdirection.
direction.To
Tosimplify
simplifythe
thesystem,
system,we
wepropose
proposetoto
keep
keepthe
thebeam
beamat
atthe
theedge
edgeof
ofthe
theprimary
primarycollimators
collimatorsand
and
close
close to
to the
the first
first secondary
secondary collimators
collimatorsusing
using only
onlythree
three
fast
fast magnets
magnets for
for each
each direction.
direction. Most
Most of
of the
the particles
particles
scattered
scattered out
out of
of the
the primary
primary collimators
collimators are
are intercepted
intercepted
then
then by
by these
these secondary
secondarycollimators,
collimators,with
withother
othercollimacollima-
FIGURE 1. Beam collimation system layout,
layout, beta
beta functions
functions
and dispersions in the
the 88 GeV
GeV (top)
(top) and
and 16
16 GeV
GeV (bottom)
(bottom)
synchrotrons.
top energies are shown in Figs. 2 and 3 for the
the systems
systems
with 0.3-mm thick tungsten
tungsten primary
primary collimators,
collimators, four
four
secondary collimators (0.5-m
(0.5-m long
long stainless
stainless steel
steel or
or copcopper) positioned at d = 2 mm and two 0.3-m
0.3-m long
long supplesupplementary collimators at d = 4 mm.
mm. The
The right
right sides
sides of
of the
the
Figures show details of beam
beam loss
loss in
in the
the collimation
collimation reregions. It is assumed
assumed in
in calculations
calculations that
that 10% of
of the
the beam
beam
is lost at injection
injection and
and 1%
1% at
at the
the top
top energy,
energy, and
and 2/3
2/3 of
of
these amounts interact the horizontal
primary
collimator
horizontal primary collimator
(a half
half for off-momentum
off-momentumprotons
protons with
with A/?//?
∆p/p == ±0.002
± 0.002
and a half
half for on-momentum
on-momentum protons)
protons) and
and 1/3
1/3 the
the vertivertical primary collimator. The /3-function
β -function varies
varies along
along the
the
secondary collimators, therefore
therefore the
the collimator
collimator apertures
apertures
are tapered to follow
follow the beam
beam envelope
envelope after
after painting.
painting.
181
tors intercepting the larger amplitude and off-momentum
tors intercepting
the larger
amplitude
and off-momentum
protons.
Such a scheme
allows
to localize
a majority of
protons.
Such
a
scheme
allows
to
localize
a majority of
the beam loss in a short region.
the
beam
loss
in
a
short
region.
A detailed sensitivity analyses were performed for
A detailed systems
sensitivityin analyses
performed
for
the collimation
the two were
machines.
Closed
the
collimation
systems
in
the
two
machines.
Closed
orbit deviations during the cycle and from cycle to cycle
orbit adeviations
during
the cycleoffset
and from
to cycle
change
secondary
collimator
withcycle
respect
to
change
a
secondary
collimator
offset
with
respect
the primary one. In the worst case, the beam can hitto
the primary
one. Incollimator.
the worst case,
initially
a secondary
This the
willbeam
resultcan
in hit
a
initially
a
secondary
collimator.
This
will
result
in a
lower collimation efficiency and can cause damage of the
lower collimation efficiency and can cause damage of the
collimator. Positioning secondary collimators 1 to 3 mm
collimator. Positioning secondary collimators 1 to 3 mm
farther from the beam increases slightly beam loss rates
farther from the beam increases slightly beam loss rates
in the
ring,
butbut
allows
larger
closed
in the
ring,
allows
larger
closedorbit
orbitdeviations
deviations(up
(up
to ±3
mm)
at
these
locations.
The
tune
causes
to ±3 mm) at these locations. The tune causeschange
changeofof
phase
advances
between
thethe
collimators
phase
advances
between
collimatorsand
anddistance
distancetoto
the the
resonances.
As
betatron
amplitudes
of
protons
resonances. As betatron amplitudes of protonsafter
after
interaction
with
primary
collimators
areare
large,
interaction
with
primary
collimators
large,the
thesecond
second
factor
cancan
cause
collimation
efficiency
factor
cause
collimation
efficiencydegradation.
degradation.We
We
found
thatthat
tune
deviations
affect
found
tune
deviations
affectmostly
mostlythe
thebeam
beamloss
loss
peaks
increasing
them
byby
upup
to to
a factor
peaks
increasing
them
a factorfive.
five.
TheThe
mechanical
design
of of
thethe
secondary
mechanical
design
secondarycollimators
collimatorsisis
similar
to that
of of
those
already
built
similar
to that
those
already
builtand
andinstalled
installedininthe
the
Tevatron
forfor
Collider
Run
II.II.The
Tevatron
Collider
Run
Thecollimator
collimatorjaws
jawsconconof two
pieces
30-40
mmwide
widewelded
weldedtogether
togetherinin
sist sist
of two
pieces
30-40
mm
an 130-mm
configuration.Primary
Primarycollimators
collimatorsare
are
an 130-mm
“L”"L"
configuration.
made
tungsten
1 mm
thick.Secondary
Secondaryand
andsupplesupplemade
of of
tungsten
1 mm
thick.
mentary
collimators
made
stainlesssteel
steelororcopcopmentary
collimators
areare
made
ofofstainless
(choice
subjectofoffurther
furtherthermal
thermalanalyanalyperper
(choice
willwill
be be
thethe
subject
m (secondary)
(supplementary)long.
long.
ses)ses)
0.50.5
m (secondary)
andand
0.30.3
mm
(supplementary)
These
dimensions
will
accommodate
the
full
beam
size,
These dimensions will accommodate the full beam size,
after
painting,
as
well
as
maximum
impact
parameters.
after painting, as well as maximum impact parameters.
Machining
assembly
tolerances
areeasily
easily
Machining
andand
assembly
tolerances
ofof2525µ jum
m are
met
for
the
collimator
jaws.
All
collimators
will
be
met for the collimator jaws. All collimators will be ininaa
fixed
position
during
machine
cycle,but
butmotion
motionconconfixed
position
during
thethe
machine
cycle,
trol is required in order to adjust collimators to their optitrol is required in order to adjust collimators to their optimum position. The collimator assembly is welded inside
mum position. The collimator assembly is welded inside
a stainless steel box with bellows on each end.
a stainless steel box with bellows on each end.
RADIATION ANALYSIS
RADIATION ANALYSIS
MARS calculations show that residual dose rates on
calculationsand
show
that of
residual
dose rates
on
collimators
magnets
the collimation
system
the significantly
collimators and
magnets
of the collimation
system
exceed
the hands-on
maintenance
limits
significantly
maintenance
limits
(Fig. 4). Toexceed
reduce the
thesehands-on
levels and
protect ground
wa(Fig.
To reduce
these walls,
levels the
andentire
protect
ground
wa-to
ter4).
outside
the tunnel
region
needs
ter be
outside
the The
tunnel
walls, the found
entire for
region
needs
to
shielded.
configuration
PD-II,
consists
be shielded.
The configuration
found
PD-II,ofconsists
of steel shielding
uniform in
two for
sections
the 58of steel
shielding
uniform
two 0.5
sections
of theof58m region:
first, 5-m
long,instarts
m upstream
the
secondary
collimator
HIstarts
and second
is in the remainm region:
first,
5-m long,
0.5 m upstream
of the
ing downstream
region.
Thesecond
first section
m (vertisecondary
collimator
H1 and
is in is
the1 remainand 1.3 region.
m (horizontally)
on each
side(vertiof the
ing cally)
downstream
The first thick
section
is 1 m
secondary
and 0.6-m
The
cally)
and 1.3 collimators
m (horizontally)
thickaround
on eachmagnets.
side of the
second
section
is
0.65-m
(vertically)
and
0.95-m
(horisecondary collimators and 0.6-m around magnets. The
zontally)
thick
on each (vertically)
side of the and
collimators,
second
section
is 0.65-m
0.95-m 0.25-m
(horizontally) thick on each side of the collimators, 0.25-m
MARS
the
around dipoles, and 0.4 m (vertically) and 0.7 m (horiaround dipoles,
0.4 m (vertically)
and 0.7
m (horizontally)
around and
quadrupoles.
This reduces
residual
dose
zontally)
around
quadrupoles.
This
reduces
residual
doseof
rates below the limits, provides adequate protection
rates below
the limits,
provides
protection
of
cables
and other
components
in adequate
the tunnel
and ground
cables
and
other
components
in
the
tunnel
and
ground
water around the tunnel, and equalizes (to some exwaterthearound
the tunnel,
andaround
equalizes
(to some
extent)
dirt shielding
needed
the entire
machine.
tent)
the
dirt
shielding
needed
around
the
entire
machine.
Therefore, the same external shielding design in the arcs
Therefore,
same external
shielding
in theof
arcs
and
straightthe
sections
is applied.
Takingdesign
maximum
the
and
straight
sections
is
applied.
Taking
maximum
of
the
normal operation and beam accident cases, the thickness
operationabove
and beam
accident
cases,
the thickness
ofnormal
dirt shielding
the tunnel
(with
a safety
factor of
of dirt shielding above the tunnel (with a safety factor of
three) is 5.8 m or 19 feet. The maximum dose accumuthree) is 5.8 m or 19 feet. The maximum dose accumulated in the collimators and hottest spots of the magnet
lated in the collimators and hottest spots of the magnet
coils
The maximum
maximumyearly
yearlydose
doseatat
coilsreaches
reaches 200
200 Mrad/yr.
Mrad/yr. The
cable
locations
is
about
150
krad
per
year.
cable locations is about 150 krad per year.
.
** / v.....
o
0
-g
'co
Q)
30
40
50
Z(m)
FIGURE 4. Maximum contact residual dose on site surfaces
FIGURE 4. Maximum contact residual dose on site surfaces
of the PD-II collimation section components (diamonds) and
ofshielding
the PD-II
collimation section components (diamonds) and
(circles).
shielding (circles).
REFERENCES
REFERENCES
1. 'The Proton Driver Design Study" FERMILAB-TM-2136,
1. “The
Proton2000.
Driver Design Study” FERMILAB-TM-2136,
December
2000.O.E. Krivosheev, N.V. Mokhov, 'Beam Loss
2. December
A.I. Drozhdin,
2. A.I.
O.E.
Krivosheev,
Mokhov,
Loss
and Drozhdin,
Collimation
in the
Fermilab N.V.
16-GeV
Proton“Beam
Driver",
and
Collimation
in the
FermilabConference,
16-GeV Proton
Driver”,
Proc.
2001 Particle
Accelerator
Chicago,
Proc.
2001
Particle
Accelerator Conference,
p. 2572
(2001);
Fermilab-Conf-01/128
(2001). Chicago,
3. p.N.V.
Mokhov,
Drozhdin, O.E. Krivosheev,
2572
(2001);A.I.
Fermilab-Conf-01/128
(2001). 'Radiation
Shielding
of the
Fermilab
Proton
Proc. “Radiation
2001
3. N.V.
Mokhov,
A.I.
Drozhdin,
O.E.Driver",
Krivosheev,
Particle Accelerator
Conference,
25782001
(2001);
Shielding
of the Fermilab
ProtonChicago,
Driver”,p.Proc.
Fermilab-Conf-01/132
(2001). Chicago, p. 2578 (2001);
Particle
Accelerator Conference,
4. Fermilab-Conf-01/132
I.S. Baishev, A.I. Drozhdin,
N.V. Mokhov, 'STRUCT
(2001).
Program
User's
Manual",
SSCL-MAN-0034
4. I.S.
Baishev,
A.I.Reference
Drozhdin,
N.V. Mokhov,
“STRUCT
(1994); http://www-ap.fnal.gov/~drozhdin/.
Program
User’s Reference Manual”, SSCL–MAN–0034
5. (1994);
N. V Mokhov,
'The MARS Code System User's
http://www-ap.fnal.gov/∼drozhdin/.
Guide",
Fermilab-FN-628
(1995);
V Mokhov
and
5. N.
V. Mokhov,
“The MARS
CodeN.System
User’s
O.
E.
Krivosheev,
'MARS
Code
Status",
Fermilab-ConfGuide”, Fermilab-FN-628 (1995); N. V. Mokhov and
00/181
(2000); http://www-ap.fnal.gov/MARS/.
O.
E. Krivosheev,
“MARS Code Status”, Fermilab-Conf00/181 (2000); http://www-ap.fnal.gov/MARS/.
182