Nuclear Instruments and Methods in Physics Research A252 (1986) 467-470 North-Holland, Amsterdam
467
Section I X . Silicon detectors POSITION SENSITIVE SILICON DETECTORS INSIDE THE TEVATRON COLLIDER * G. R. M. G.
A P O L L I N A R I , F. B E D E S C H I , G. B E L L E T T I N I , F. BOSI, L. BOSISIO, F. C E R V E L L I , D E L F A B B R O , M. D E L L ' O R S O , A. D I V I R G I L I O , E. F O C A R D I , P. G I A N N E T T I , G I O R G I , A. M E N Z I O N E , L. R I S T O R I , A. S C R I B A N O , P. S E S T I N I , A. S T E F A N I N I , T O N E L L I a n d F. Z E T T I
University, INFN and Scuola Normale Superiore of Pisa, Italy S. B E R T O L U C C I , M. C O R D E L L I , M. C U R A T O L O , B. D U L A C H , B. E S P O S I T O , P. G I R O M I N I , S. M I S C E T T I a n d A. S A N S O N I Laboratori Nazionali di Frascati of INFN, Italy S. B E L F O R T E , T. C H A P I N , G. C H I A R E L L I , N. G I O K A R I S , K. G O U L I A N O S , R. P L U N K E T T a n d S.N. W H I T E Rockefeller University, New York, USA
Four position sensitive silicon detectors have been tested inside the Tevatron beam pipe at Fermilab. The system is the prototype of the small angle silicon spectrometer designed to study primarily p-~ elastic and diffractive cross-sections at the Collider of Fermilab (CDF). Particles in the beam halo during p-~ storage tests were used to study the performance of the detectors. Efficiency, linearity of response and spatial resolution are shown. Measurements performed at different distances from the beam axis have shown that the detectors could be operated at 8.5 mm from the beam with low rates and no disturbance to the circulating beams. This distance corresponds to about 11 times the standard half-width of the local beam envelope. The behaviour of the detectors with the radiation dose has also been investigated.
I. Introduction The Collider Detector of Fermilab (CDF) will indude two spectrometers at extremely small angles (down to O - 0.2 mr) around the beam axis on the left and right side of the intersection. These spectrometers will exploit the machine dipoles and quadrupoles for bending and will employ seven detector stations at various distances from the beam-beam crossing, ranging from - 5 to - 55 m. The detectors will be multistrip silicon diodes. The physics addressed is small angle elastic scattering, total p - ~ cross section and inelastic diffraction, and, in general all interesting diffractive-like processes [1,2]. The CDF spectrometers represent a step forward compared to the "roman pot" system developed at CERN and based on conventional detectors in mobile frames outside the beam pipe [3]. Our silicon detectors operate under vacuum and can therefore reach smaller distances from the beams. Moreover, due to specially * Presented by Guido Tonelli. 0168-9002/86/$03.50 © Elsevier Science Publishers B.V. (North-Holland Physics Publishing Division)
designed mechanical supports, systems with complete azimuthal coverage can be implemented where useful for physics.
2. The prototype telescope During the Tevatron 1 test period (September-October '85) a prototype telescope of four silicon detectors was installed on the machine and tested for a number of important features. These detectors, with an active area of 13 cm2, were - 900 /,m thick and had a petal-like shape as illustrated in fig. 1. 100 /,m wide aluminum electrodes were obtained by means of a lithographic process on the surface of the detectors. The strips run along approximately constant polar or azimuthal angle, 1 mm apart from each other. The system was designed for operation in the high vacuum of the Tevatron and the detectors were mounted on a supporting rod held on a movable flange-bellow system which was remotely operated via the standard accelerator network. Depending on the beam conditions in the various runs (position of the beam axis and beam IX. SILICON DETECTORS
468
G. Apollinari et al. / Position sensitioe silicon detectors
could cause a local vacuum deterioration. The preamplifiers were located just outside the flange, on the feedthroughs pins themselves. The signals were transferred through the CDF "rabbit" system and then processed on an IBM-PC. A sketch of the test set-up is shown in fig. 2.
3. Test results
Fig. 1. The prototype telescope mounted on the supporting rod. size), the distance of the detectors from the beam could be accordingly varied with great accuracy. The detectors were connected to the outside world through multipin high vacuum feedthroughs. Besides the detectors themselves, all materials operating in vacuum, including signal cables, were carefully selected in order to avoid any virtual leakage which
As it was shown elsewhere [4], by exploiting the resistive layer of undepleted silicon at the ohmic contact it is possible to use the charge division between adjacent strips to determine the position of a crossing particle with an accuracy much better than the strip pitch; in principle, the accuracy is limited only by the signal-tonoise ratio and, in our case, should amount to less than 100/~m. The charge partition mechanism was accurately monitored using crossing particles from the beam halo. Minimum ionizing particles were selected by requesting a double coincidence between two out of four detectors. Under these conditions the collected events were subdivided in two main categories, single and double hit events, depending on the number of strips fired per event. The single hit events, corresponding to particles that hit directly the strips, were a small fraction of the total. Their pulse height distribution (fig. 3) shows a clear Landau peak. The pulse height distribution of the double hit events, corresponding to particles crossing the interstrip region,
40(
30(
20'
10(
~LLf SO0
PUL O E,6.T IADE channets]
Fig. 2. The mechanical assembly of the detectors inside the Tevatron beam pipe.
Fig. 3. Pulse height distribution for "single hit" events.
469
G. Apollinari et al. / Position sensitive silicon detectors
is a flat distribution with a tail (fig. 4a), but a simple sum of the pulse heights collected by two adjacent strips shows the m i n i m u m ionizing peak (fig. 4b.). The signal to noise ratio in this case was 13 to 1. Due to the large angular divergence of the particles in the beam halo and to the small n u m b e r of measurements (two points per coordinate), we can give only some indications on the detector resolution. In fig. 5a
%.
50
10
a.)
z. - j o
40 i
3°f
5
20
DETECTOR,S, STRIP~17
oq
z
300
10
O r
200 L ...z,5
100
I
1
1
0
f
250
l
500
n
I 1250 DETECTOR*/, STRIP~,18 / 1000
750
t/1
30c
1
POSITION[mm]
900um_l -,_
6 DETECTORe3
~
I
5
6
DETECTOR~I
Fig. 5. (a) Reconstructed impact points on strip 5 of detector 3. (b) Corresponding impact points as reconstructed on detector number 1.
o
20(
100
250
500
a
750
I
1250 PULSEHEIGHT [AOC channetsl 10tO0
i t/1 I.Z
we show a n u m b e r of impact points reconstructed on detector 3. The corresponding positions as reconstructed on detector 1 are shown in fig. 5b, where the influence of the large angular spread of the particles is quite clear (consider that the two detectors are only 7 m m apart). The plot of the difference posa-pos 1 is shown in fig. 6. Again the angular spread overcomes the spread due to the finite detector resolution, so, unfolding the particles divergence, we can only conclude that our data are consistent with the expected resolution of 80/~m.
O
t~
300
E =s o
200
20
15 100 10
5 I
S00
b
I
1000 PULSE HEIGHT IADC channelsl
Fig. 4. (a) Pulse height distribution for "double hit" events. (b) Sum of the pulse heights collected by two adjacent strips.
-1
0
1 POS3-POSI Imml
Fig. 6. Distribution of the difference pos 3-pos 1. IX. SILICON DETECTORS
470 E =. o
G. Apollinari et al. / Position sensitwe silicon detectors r
200/ I
I
t
,
t/1 I--z 0 l.d
N
~c
1
5
0
I
HIGH BETA SCAN
Ix L~ z l-z
~
o
I00 1 -
501~-
90
I
a
t
I
t
0
1
2
I/+
15
16
I
60
+
30
+
3 POSITION[mini
,
17
÷
÷
t 1/*
I
1
13
12
I
I
11
I
The relative detection efficiency for the four detectors was found to be greater than 99%. The linearity of response in the interstrip region has been checked plotting the impact positions, as reconstructed by the detectors, when the telescope was positioned at a distance from the beam at which the illumination was approximately uniform (fig. 7). All in all we conclude that the detectors behave according to expectations. A n u m b e r of scans of the beam halo was done both in high and low beta optics, by measuring the rate of m i n i m u m ionizing particles crossing the telescope when the detectors were moved toward the beam (in steps of 100/~m). The results are summarized in fig. 8. In both cases a distance smaller than 10 m m from the beam axis was easily reached without affecting the performance of the detectors or disturbing the stored beam. This distance corresponds to 11 times the standard beam half-width and this would indicate that 0 < 0.2 mr will be reachable in the actual experiment.
4. R a d i a t i o n r e s i s t a n c e
and conclusions
The radiation intensity on the detectors was continously monitored during the test. Measurements performed with thermoluminescence dosimeters positioned on the flange and all around the beam pipe showed an overall integrated dose of about 500 tad after more than one month of operation. Since previously performed tests with a high intensity beta source had shown, that for an irradiation up to 104 rad there was no practical effect on the electrode
f
9
DISTANCE [mm
/ DETECTOR /*
Fig. 7. The response of detector 4 to a uniform flux of particles.
lO
"NI -
~:
I
i
'
'
t
b 7.5
LOW BETASCAN
p,. co z pz-.~ C~ t,J
S.0
2.5
•
14
i
• i
• i
13
12
11
e e e e ~ i i
10 9 8 DISTANCE [mm)
Fig. 8. (a) Beam halo scan for high fl optics. (b) Beam halo scan for low fl optics.
response, we conclude that the planned cross-section measurements are feasible, since in similar future running conditions the detectors will be able to stand one full year of operation.
References
[1] S. Bertolucci et al, The Small Angle Silicon Detector System of CDF (March 1985) CDF 285. [2] G. Bellettini, Physics at the Tevatron Collider, Proc. of the Aspen Winter Conference Series (January 6-12, 1985). [3] R. Battiston et al., Nucl. Instr. and Meth. A238 (1985) 35. [4] S.R. Amendolia et al., Nucl. Instr. and Meth. 226 (1984) 82.