METAL FOIL DETECTORS AND THEIR APPLICATIONS

1
METAL FOIL DETECTORS AND THEIR APPLICATIONS
V. Pugatch a,b, V. Aushev a,c, C. Bauer b, K.-T.
Knöpfle b, M. Schmelling b, M.Tkatch a, Yu.
Vassiliev a,c aKiev Institute for Nuclear
Research, bMax-Plank-Institut für Kernphysik,
cDeutsche Elektronen-Synchrotron DESY
VCI 2004
ABSTRACT To monitor charge particle fluence
distribution, various methods have been
developed. We report here on approach based on
the Metal Foil Detectors (MFD). The MFD is a
device which measures a flux of charged
particles by integration of the charge created in
a thin metal foil due to the Secondary Electron
Emission (SEE) initiated by the incident
particles. The radiation profile is reconstructed
from the charge distribution in a set of metal
foils partitioning the area under study. Their
size has to be optimized to provide appropriate
profiling and, at the same time, provide charges
which are big enough to trigger read-out Charge
Integrators (CIs) which perform a current to
frequency conversion. The foils have to be thin
to minimize multiple scattering of the tracks,
and, on the other hand, thick enough for
mechanical stability.
Introduction The first monitor based on
Secondary Electron Emission (SEE) initiated by
the incident particles has been build already in
1955. Since then various types of the SEE beam
profile monitors were designed, yet, to our
knowledge the principle has not been explored for
radiation monitoring using comparatively large
area sensors placed in the atmosphere. It was
established that the yield of the SEE exceeds by
factor of 10 the one of directly produced
?-electrons. Metal Foil Detectors (MFD) were
built and successfully tested at KINR (Kiev),
DESY (Hamburg), MPIfK (Heidelberg) and CERN
(Geneva) during the last years for charge
particle flux-monitoring. MFD BPIT at 21 Mev
proton beam For the purpose of silicon strip
detector irradiation study we have built and
tested the MFD Beam-Profile-Imaging-Target (MFD
BPIT) at MPIfK (Heidelberg). To simulate a
non-uniform radiation load 21 MeV protons
scattered by the 50 ?m thick Au-target were
used. The test experiment with a 12.5?m thick
Ni-foil (2x15 mm2) positioned at very forward
scattering angles has confirmed that the rate
output of a charge integrator (100 Hz per 1 pA
calibrating current) connected to the foil
perfectly followed the expected Rutherford
scattering behaviour. The obtained values of the
proton fluences agreed well with calculations.
The BPIT with sizes large enough to cover nearly
the whole silicon detector area has been made out
of 10 Ni-foils (12.5 ?m thick , 4 mm wide , 50 mm
long) placed on a PCB with a pitch of 5mm.
Figure 1. Left part BPIT (50 x 50 mm2). Proton
beam axis is perpendicular to the BPIT plane.
Right part 21 MeV proton beam profile measured
by the BPIT. Beam axis is between strips 5 and 6
Beam Profile Monitor for the LHCb Inner Tracker
at CERN
MFD at 920 GeV proton beam (HERA-B
experiment) Another way to reconstruct the beam
profile was realized by us for the HERA-B
experiment, which operates a set of 8 target
ribbons (wires) surrounding the 920 GeV proton
beam at two locations along the beam. The typical
dimensions of a wire are 125 -500 ?m along the
beam and 50 ?m in the radial direction (Fig. 3),
to provide an Interaction Rate (IR) 5-40 MHz. In
addition to the data from Cis connected to the
individual target wires we employed the Silicon
Vertex Detector (SVD) data to measure the beam
profile in the plane perpendicular to the
beam. At the nominal target position of 4
standard deviations from the beam centre (? 350
?m) the IR is very sensitive to the beam shape
and its fluctuations (Fig.4), and the knowledge
of the beam profile is of a paramount importance
for the target steering. The feedback data for
steering are derived from Scintillator Hodoscopes
for the overall IR and CIs for partial
contributions of the individual wires.
At X7 test beam area of the CERN-SPS 16 X- and 16
Y-sensors (Al, 50 ?m thick, 5 mm wide, 6 mm
pitch) covering the area of 96 x 96 mm2 were
connected to 32 Charge Integrators (CIs) via 3 m
long cables. The X- and Y- sensors were
separated (3 mm distance) by accelerating and
shielding Al-foils. A data acquisition system
periodically read out the rates through a 32
channel VME Scaler, which were then translated
into a graphical presentation of the beam profile
in the control room. The MFD BPM data have shown
perfect correspondence with a similar data
measured by a regular BPM devices (multi-wire
proportional cham-bers) provided by CERN. A MFD
monitor of much smaller size (32 Al strips, 10 ?m
wide with a 32 ?m pitch deposited onto a 20 ?m
thick Si-wafer) has been designed and tested for
the online control, positioning and focusing of
32 MeV alpha-particles for Single-Event-Upset
studies of the BEETLE readout chip at the Tandem
generator of the MPI für Kernphysik (Heidelberg).
HERA-B Target Setup
Y (cm)
Al
Pd
W
C
P - beam
Ti
C
C
Figure 2. On-line MFD BPM performance at X7 test
beam facility at CERN 120 GeV ?-beam with 5.2
105 particle per 2.4 s spill.
Ti
Z (cm)
X (cm)
HERA-B Rate sharing at multi wire operating mode
Figure 4. Correlations IR, CI, beam and target
position, Primary vertices.
Figure 3. Reconstructed vertices from 8 targets
As an example Fig.6 shows the distribution of the
vertices over 4 inserted Titanium targets. The
numbers of vertices at each target are nearly
equally distributed over the targets and are in a
good agreement with the IR distribution provided
by MFD and CIs.
Another essential feature of the Luminosity
monitoring by the MFD is that this is a nearly
background free method. Indeed, due to small size
of the targets the response of the corresponding
CI to any background flux of particles is small
in comparison with a signal obtained by the
beam-target interactions. The Fig. 4 illustrates
the evolution of the rates in Hodoscopes (black
line), CI-Inner2 (pink) and background counters
(red line). The upper plot shows red Inner2
target position, blue X beam position and green
dots Primary vertices reconstructed off-line in
the Inner2 target. The unusual behavior of beam
movement was perfectly compensated by the Target
steering program and nicely seen online by the
MFD.
Luminosity Monitoring for the HERA-B experiment
An application of the MFD Luminosity Monitor for
the HERA-B Minimum Bias and Physics Trigger runs
are shown on the Fig. 5. All 8 HERA-B Targets
were operated for few minutes at 0.5, 1, 5 and
51, 0.5 MHz interaction rate (rate scan). The
interaction rate measurement by the target
hodoscopes (scintillation counters) is shown on
the upper plot, while the same measurement by the
MFD is shown on the lower one. Very good
agreement between MFD and hodoscopes results
allowed us to conclude that MFD could be used as
Luminosity Monitors. Since MFDs have orders of
magnitude higher acceptance then the hodoscopes
(50 vs 1) and do not show saturation effects,
they are more sensitive to any changes of the
beam position and/or hanges in the distribution
of detector material close to the beam, like e.g.
the positions of the VDS roman pots. Most of the
HERA-B Minimum Bias runs 2002-2003 show perfect
(within 2) stability of the delivered luminosity
measured by the MFD, Hodoscopes, Lumi-counters
and another parts of the HERA-B detector, like
the ECAL and the RICH.
Al C Ti W Ti Pd C C
Al C Ti W Ti Pd C C
Figure 6. Vertices distribution over 4 targets
surrounding the 920 GeV proton beam (IR 30
MHz). X and Y are the coordinates of the targets
in the SVD coordinate system the beam axis is
shifted down by 1 mm and closer by 3.2 mm to the
inner target from the SVD axis.
Table 1. Relative rate sharing among wires
obtained with the Charge Integrators and primary
vertices counting
Figure 5. Luminosity Monitoring of MB runs by MFD
Wire Charge Integrators Vertices Above
26.06 ? 0.08 26.6 ? 0.7 Below
24.26 ? 0.10 25.9 ? 0.7 Inner
23.49 ? 0.06 21.4 ?
0.7 Outer 26.20 ? 0.07
26.1 ? 0.7
Conclusions Reliable results on the beam
profiling were obtained with the help of the MFD
BPIT at proton beam energies of 21 MeV and 920
GeV and with a 120 GeV ?--beam . The achieved
sensitivity of the BPIT allows for an on-line
beam-profile measurements at beam currents
exceeding 1 pA. The developed MFD proved to be a
reliable new tool for the charged particle
radiation monitoring in a wide range of
applications. The advantages of the MFD are
extremely low mass of the detecting material, low
cost, simple structure, low operating voltage (20
V), simple read-out electronics (charge
integrators and scalers) and very high radiation
tolerance.
The relative sharing of the IR among the inserted
targets is illustrated by Table 1 for the case of
the overall IR 30 MHz. There is a perfect
linear relation between the number of the
reconstructed vertices and CI-rates for IRs per
individual target up to 10 MHz (till multiple
interactions per bunch crossing start to affect
the vertex reconstruction efficiency.