T12


MORE ON THE INTERACTION REGION OF PEP-N*

M. Sullivan**, SLAC, Stanford, CA 94025, USA

Abstract magnetic elements needed for the VLER would take up
The PEP-N project [1,2] consists of a small, very most of the small angle acceptance of the detector. This
low-energy e- storage ring (VLER) located in one of is important since the energy asymmetry in PEP-N is
the interaction-straight regions of PEP-II. The small very large over most of the E range of interest. The
cm
ring is brought into collision with the low-energy (3.1 LER energy is held constant at 3.1 GeV while the
GeV) e+ beam (LER). The center-of-mass energies VLER has an energy range of about 100-800 MeV. The
detector angular acceptance in the forward boost
from this collision are between the and J/ direction is 100 mrad along the beam direction. In
resonances. We achieve a head-on collision through the addition, a crossing angle collision means that the beam
use of a central magnetic dipole field that generates a must be brought back over the LER beam in order to
large horizontal bending field. This field is also the keep the small storage ring on one side of the LER
central field of the detector. The large energy range of beam line. This is especially difficult because it has to
the VLER, in order to maximize the center-of-mass be done very soon after the collision. There is not much
energy range, complicates the collision point geometry.
In order to maintain the beam orbits near the collision space in the 10m long interaction region hall to get the
point two techniques are used. The first is to scale the beam back to the other side of the LER. This left the
central dipole field up and down with the energy of the head-on solution as the best choice for the collision.
VLER and the second is to use passive shielding to The beams are brought into collision and separated by a
large horizontal dipole field located at the interaction
decrease the integral Bdl of the dipole field seen by the point that insures that the VLER stays on the same side
VLER. Changes in the orbit of the LER are corrected of the LER beam line. This same magnetic field serves
with local bending magnets. Further details of the as the detector field.
interaction region geometry as well as design issues The present working design uses a field model that
that include synchrotron radiation from the high-current has a maximum field strength of 3 kG. Figure 1 is a
positron beam are discussed. plot of the field from this magnet along the z axis.

1 INTRODUCTION

The SLAC, LBNL, LLNL, PEP-II B-factory[3]
consists of two storage rings located one above the
other in the PEP tunnel at SLAC. The low-energy ring 3
(LER) of positrons is 89 cm above the high-energy ring
(HER) of electrons. At design operation, the LER beam
current is 2.14 A in 1658 bunches that are 1.26 m apart.
The HER beam current is 0.75 mA. The PEP-N 2
proposal consists of a small storage ring of electrons
that collide with the LER to produce a center-of-mass kG
energy (E ) between the and the J/ (approximately
cm
1 GeV to 3 GeV). The very low-energy (VLER) 1
electron storage ring would be located in one of the old
interaction region halls of PEP-I (interaction region 12
is presently being considered).

-2 -1.2 -0.4 0.4 1.2 2
2 INTERACTION REGION DESIGN m
Figure 1. Plot of the magnetic field from the central
Three major collision designs were considered: 1) dipole magnet. The field extends out to at least 2 m
head-on, 2) a small angle collision, similar to KEKB, from the center. We will shield the accelerator beam
3) and a very large angle collision (> 100 mrads). The pipes as much as possible to minimize the integrated
very large angle collision design would be an strength of this magnet.
interesting accelerator to build but it was considered
very risky and had a high probability of not producing The IP is located -25 cm from the center of the field.
the required luminosity as well as introduce The 3 kG field corresponds to an accelerator design
perturbations on the LER that might adversely affect B- that has a 557 MeV beam energy for the VLER.
factory running. The smaller angle collision (~20 mrad Placing the collision point -25 cm from the center of
total angle) presents two difficulties. One is that the


T12
the magnet increases the amount of detector field in the m) that both beams travel through. This magnet can be
boost direction and minimizes the amount of upstream used to add or subtract to the central dipole field and is
magnetic field. This lowers the amount of upstream used to maintain the VLER orbit when the VLER
bending (and hence synchrotron radiation) in the LER. energy is changed. The next element is QDI1 (1.5-1.8
m). However, on this side of the IP this large aperture
2.1 Downstream beam lines magnet is seen by both beams and is a normal steel
magnet. The center of this horizontally defocusing
The downstream side of the collision point is in the magnet is positioned close to the LER beam
direction of the LER. This is the side where most of the minimizing the bend for this beam and maximizing the
physics particles go. On this side, the VLER is bend for the VLER. The extra horizontal kick from this
deflected horizontally 192 mrads while the LER is magnet separates the beams enough so that the next
deflected 34 mrads. This results in a separation of 26.7 element (QFI2, 2.5-2.8 m) can be a septum quadrupole
mm between the two beams at the first parasitic with a field-free drift region for the LER. The rest of
crossing, 0.63 m from the IP, which translates into 38
the magnets in the VLER are essentially the same as
for the VLER and 68 for the LER. This large
x x the downstream side with the VLER beam parallel to
separation makes any beam-beam effect from the and offset from the LER design trajectory by 40 cm 4.1
parasitic crossing negligibly small. The beams are m from the IP. The LER beam line includes four
separated enough to allow each beam to enter a horizontal dipole magnets to steer the LER back to the
separate beam pipe about 1.3 m from the IP. nominal orbit and to close dispersion. Figures 2 and 3
The first accelerator element after the central dipole are layout pictures of the interaction region.
field is a vertically focusing quadrupole (QDI1) for the
VLER located 1.5-1.7 m from the IP. QDI1 is
constructed from permanent magnet material. The
compactness of the permanent magnet design permits
this magnet to be 1.5 m from the IP and yet not have QDI3
B1VL QDI3 QFI2 B1VL
QFI4
any effect on the nearby LER beam. The small design QFI4
QFI2 B0VL
QDI1
also maximizes the solid angle acceptance of the 0.5 VLER QDI1 VLER
detector. Following QDI1 is a horizontal bending
magnet (B0VL). This magnet starts the reverse bend on
the VLER that brings the VLER back parallel to the rs
LER. The next VLER element is a horizontally te 0
LER LER
focusing quadrupole (QFI2) located 2.5-2.8 m from the me B1L
IP. This magnet is far enough away to no longer B0L
B1L QD1L
QF2L Shielding QF2L
interfere with the LER and it has only a minor impact B2L QD1L B2L
B0VL
on the detector solid angle. Following QFI2 is another -0.5 Central Dipole
quad QDI3 from 3.3-3.6 m. A reverse bend horizontal
dipole (B1VL) at 3.7-4.1 m straightens out the VLER
orbit to be parallel to and 40 cm from the LER -10 -5 0 5 10
followed by one more matching quad (QFI4) at 4.2-4.5 meters
m. Figure 2. Layout of the interaction region. Please note
The downstream LER beam line includes 3 the exaggerated left-hand scale.
horizontal dipole bend magnets to correct the orbit back
to the nominal trajectory and to match dispersion.

2.2 Upstream beam lines

The upstream side of the IP has very similar magnet
placement as the downstream side, however the beams
are not separated as quickly so the implementation is
different. On this side the beam separation at the first
parasitic crossing is 22.8 mm which is 32 for the
x
VLER and 58 for the LER, still large enough to make
x
tune shifts from this parasitic crossing negligible.
Moving out from the IP we find that the first
accelerator element is a horizontal dipole magnet (1-1.3


T12
0.3 -25 cm upstream from the center of the main field and
QDI1
that we use the offset QDI1 magnet to further separate
QFI4 B1VL
B1VL QDI3 QFI4 the beams means that the upstream LER has relatively
QDI3 QFI2 weak bending magnets. The main source of
B0VL synchrotron radiation power comes from the two
QFI2 QDI1 closest of the four dipole magnets on the LER beam
rs line. The strength of these magnets for the 780 MeV
te 0 VLER is 2.1 and 2.0 kG and they generate 465 and
me 1054 W respectively with a 2.14 A LER beam. The
Shielding B1L critical energies of these bend magnets are 1.36 and 1.3
keV. The power levels are low enough to not pose a
Central Dipole
B0L problem for beam pipe cooling. More complete studies
B0VL need to be made, but synchrotron radiation power does
not seem to be an issue and the very low critical
-0.3 - energies argue that detector background levels from
4 -2 0 2 4
meters synchrotron radiation will be low. Figure 4 shows the
Figure 3. Close up of the interaction region. fan of radiation coming from these magnets.


0.3 QDI1
3 CHANGING THE CENTER-OF-MASS QFI4 B1VL
ENERGY OF PEP-N B1VL QDI3 QFI4
QDI3 QFI2
The LER energy is fixed by PEP-II at 3.1 GeV. In B0VL
order to change the Ecm we must change the energy of QFI2 QDI1
the VLER. The interaction region baseline design of rs
PEP-N has a 557 MeV VLER and a 3 kG central dipole te 0 SR fan
field. In order to reach the J/ resonance the VLER me from B0L
energy must be increased to 780 MeV. In order to and B1L
Shielding B1L
maintain the beam orbits and get the beams to separate Central Dipole
properly we proportionally increase the field of the B0L
central dipole to 4.2 kG. Decreasing the VLER energy B0VL
from the 557 MeV deign point is done differently. The
detector collaboration prefers a higher central magnetic -0.3 -4 -2 0 2 4
field while the accelerator designers prefer a lower meters
central field. With this in mind the present design tries Figure 4. Interaction region layout with the synchrotron
to maintain 3 kG as a minimum value for the central radiation fan from the B0L and B1L magnets of the LER.
field. Therefore, in order to lower the VLER energy The fan strikes some of the upstream beam pipes. The low
and maintain the central field at 3 kG, passive shielding critical energies of this bending radiation argues that
is added to the beam pipe to subtract some of the synchrotron radiation backgrounds will not be an issue but
central field from the VLER beam. With this technique, a more complete study needs to be made to insure that
we achieve a VLER energy of 347 MeV. backgrounds are acceptable.
In order to go still lower in VLER energy, the
present strategy would be to rebuild the beam pipes in
the interaction region allowing for a much larger angle 5 SUMMARY
of separation at 347 MeV through the use of the The interaction region design of PEP-N employs a
unshielded central field of 3 kG. Once again, we would central horizontal dipole field to bring the beams into
add passive shielding to the VLER beam pipe as the collision and serve as the central detector field. The
VLER energy is lowered from the 347 MeV starting collision point is shifted -25 cm upstream in the LER
point. We note here that this strategy is still preliminary in order to maximize the detector field in the forward
and further refinements will no doubt be forth coming. direction where most of the physics particles travel as
well as minimize the upstream bending magnet
4 SYNCHROTRON RADIATION strengths that contribute to detector background issues.
The 780 MeV VLER design has the highest levels of A baseline design with a VLER energy of about 500
synchrotron radiation. However, the fact that the IP is MeV has a 3 kG central field. In order to move the Ecm


T12
up to the J/ we increase the VLER energy up to 780 beams can be moved rapidly out of collision by a
MeV and scale up the central field to 4.2 kG. This change in the VLER RF phase.
maintains the orbit separation geometry. Lowering the
VLER energy below the 557 MeV baseline energy is
accomplished by adding passive shielding around the REFERENCES
beam pipe. This allows the central field to remain at 3
kG down to a VLER energy of 347 MeV. Lowering the [1] PEP-N project LOI, PEP-II AP Note 2000.05, Sep.
VLER energy further involves modifying the orbit 2000,
separation geometry. http://pepn.pd.infn.it/LoI.loi_accelerator.html.
The central dipole field separates the beams so [2] Physics LOI in http://pep-n.pd.infn.it.
efficiently (>50 ) that there are virtually no beam- [3] "PEP-II an Asymmetric B Factory", Conceptual
x Design Report, CALT-68-1869, LBL-PUB-5379,
beam effects from parasitic crossings 63 cm from the SLAC-418, UCRL-ID-114055, UC-IIRPA-93-01,
IP. In fact, there is even enough separation (>14 ) at
x June 1993.
the halfway point (31.5 cm) to allow the VLER
bunches to be positioned there which means that the



