Article

A novel approach for an integrated straw tube-microstrip detector

INFN, Frascati, Italy
IEEE Transactions on Nuclear Science (Impact Factor: 1.28). 07/2006; 53(3):1375 - 1379. DOI: 10.1109/TNS.2006.872865
Source: IEEE Xplore
ABSTRACT
We report on a novel concept of silicon microstrips and straw tubes detector, where integration is accomplished by a straw module with straws not subjected to mechanical tension in a Rohacell® lattice and carbon fiber reinforced plastic shell. Results on mechanical and test beam performances are reported as well.

Full-text

Available from: M. Caponero, Oct 24, 2012
arXiv:physics/0512257v1 [physics.ins-det] 28 Dec 2005
LNF-05/32 (P)
27 dicembre 2005
A NOVEL APPROACH FOR AN INTEGRATED
STRAW TUBE-MICROSTRIP DETECTOR
E.Basile(*), F.Bellucci (***), L.Benussi, M. Bertani, S. Bianco, M.A. Caponero (**), D. Colonna (*),
F. Di Falco (*), F.L. Fabbri, F. Felli (*), M.Giardoni, A. La Monaca, G. Mensitieri (***), B. Ortenzi,
M. Pallotta, A. Paolozzi (*), L.Passamonti, D.Pierluigi, C. Pucci (*), A. Russo, G. Saviano (*), F. Massa(****).
Laboratori Nazionali di Frascati dell’INFN
F.Casali, M.Bettuzzi, D.Bianconi
University of Bologna and INFN, Bologna, Italy
F. Baruffaldi, E. Petrilli
Laboratorio di Tecnologia Medica, Ist. Ortop. Rizzoli and University of Bologna, Bologna, Italy
Abstract
We report on a novel concept of silicon microstrips and straw tubes detector, where integration is
accomplished by a straw module with straws not subjected to mechanical tension in a Rohacell
r
lattice
and carbon fiber reinforced plastic shell. Results on mechanical and test beam performances are reported
on as well.
Index Term: Elementary Particles, Detectors, Tracking.
Submitted to Transactions on Nuclear Science
Permanent address: “La Sapienza University - Rome.
∗∗
Permanent address: ENEA Frascati.
∗∗∗
Permanent address: “Federico II” University - Naples.
∗∗∗∗
Permanent address: INFN Rome 1.
Page 1
1 CONCEPT
Modern physics detectors are based on tracking subcomponents, such as silicon pixels and strips, straw
tubes and drift chambers, which require high space resolution, large geometrical acceptance and extremely
large-scale integration. Detectorsare oftenrequested with demandingrequirementsofhermeticity and com-
pactness that must satisfy the minimization of materials. We have developed for the BTeV experiment[1]
an integrated solution that accommodates straw tubes and silicon strips in a common structure. Our novel
design utilizes glued straw tubes mechanically un-tensioned and embedded in a Rohacell
r
lattice. The
straw-Rohacell
r
composite is enclosed in a carbon fiber reinforced plastic (CFRP) shell and supports the
microstrip detector. Un-tensioned straw tubes have been used in the past [2]. The untensioned straws in the
ATLAS TRT detector [3] at LHC need support dividers every 25cm of their lengths, and four sets of thin
carbon fiber filaments each to provide required stiffness. The TOF detector [4] at COSY uses a very large
gas overpressure (2bar) for straw stiffness. Our design avoids supports and filaments, operates straws in
standard conditions of very small gas overpressure, and allows integration between straws and microstrip
detectors.
2 BTEV DETECTOR
The BTeV experiment[1] at the Fermilab proton-antiproton collider (the Tevatron) will produce and study
particles containing the beauty heavy quark, in order to investigate the phenomenon called CP violation,
and understand if the Standard Model of particles and interactions is sufficient to describe the world we live
in. BTeV is composed of tracking detectors (pixel, strips, straws) for detection of charged particles, a RICH
Cerenkov detector for identification of pions, kaons and protons, a crystal EM calorimeter for detection of
neutral particles (photons, π
0
) and identification of electrons, and a muon detector. The experimental setup
is shown in Fig.1.
3 M0X CONCEPT
The M0Xmodule is a special moduleto be placed closestto the beam, measuring the x-coordinateof tracks.
It houses straw tubes and supports silicon strip detector planes. M0X is made of straw tubes embedded in
a Rohacell
r
IG50 foam, inside a CFRP shell. Straws are glued together and onto the foam, the foam is
glued to the CFRP shell. CFRP is chosen to allow the fabrication of a rigid mechanical structure with
high transparency to incoming particles. CFRP is also used for the M1 modules, conventional straw tubes
sub-detectorsthat act as struts sustaining the mechanicaltension of the remaining straw modules. Six straw-
microstrip stations are deployed in BTeV, each station made of three views (X, U, V), each view made of
two half-views. The X view (vertical straws) measures the X coordinate, while the two stereo views (U,V)
2
Page 2
are at ±11.3
around the Y bend coordinate. Straw inner diameter is 4mm, straw lengths vary from 54cm in
the first station to 231cm in the sixth station. Fig.2 shows a conceptual design of M0X and M1X assembly
(left), and a 60-cm-long prototype with straws embedded in Rohacell
r
(right).
4 TOMOGRAPHY
A checkof the eccentricity of the straws and of their positions in the groovescan be done with a tomography
method. The tomography uses X-rays and can reconstruct sections of the scanned region. The technique
determines location and geometry of straws by reconstructing images of their cross sections. Computed
images are reconstructed from a large number of measurements of X-ray transmission. Reconstruction
provides 2-dimensional and 3-dimensional images of straws. 2-dimensional images of 6-channel M0X pro-
totype, 3-dimensional reconstruction, and 2-dimensionalimage of final assembly technique M0X prototype
are shown in Figure 3 left, center, right, respectively. Preliminary results show that a precision of about
20µm can be reached on the measurement of straw radii. The maximum variation from circularity allowed
is 100µm, in order not to change the straw gain by more than 10%.
5 FINITE ELEMENT ANALYSIS
A finite element analysis (FEA) [5],[6] allows to estimate the displacements of the M0X module under the
loads of the microstrip detectors and straw tubes. Time stability and maximum displacements of the order
of 10 µm are requested, in order not to spoil the space resolution of the microstrip detectors. The FEA
analysis has been carried on the M0X of the sixth station, the longest straw length. A straw load of 12N
in each corner of M0X has been simulated to reproduce the mechanical tension of wires. A straw load of
12N and a torque of 2Nm have been applied to simulate the weight of the micro-strip. The geometry and
mechanical properties of materials used are reported in Tab.1. Fig.4 shows the graphicaloutput of FEA with
isocurves of deformation under simulated load of microstrip detector and associated electronics. Maximum
deformation is pointed to by arrow in Fig.4. FEA shows a maximum displacement of about 15µm (4µm
in the axial direction, 9µm x direction, 11µm y direction), close to the required specification. We have
used shell elements for the simulation of the carbon fiber reinforced polyester structure and bricks for the
Rohacell
r
simulation. Beam elements were used for introducing glue between the CRFP module and the
cylindrical plate where microstrips were placed. The geometry shown in Tab.1 and Fig.4 corresponds to a
0.007X
0
material thickness in radiation lengths units, which represents a 60% reduction in material with
respect to a design with independent supports.
3
Page 3
M0 AND M1 STRUCTURE MICRO STRIP CYLINDER ROHACELL FOAM
THICKNESS 0.07 EACH PLY WITH 0.07 EACH PLY (0/90/0)
[MM] FIBRES DISPOSITION WITH A ROHACELL 2
OF 0/90/0 FOAM OF 5CM
E
11
[GP a] 260 590 0.019
E
22
[GP a] 10 10 -
E
12
[GP a] 7.2 7.2 -
ν
12
0.3 0.3 0.3
Table 1: Geometrical arrangement (thicknesses), Young modules (E
11
, E
12
, E
22
), and Poisson coefficient
(ν
12
) used in FEA simulation of M0X.
6 FBG SENSORS
Fiber Bragg Grating (FBG) sensors have been used so far as telecommunicationfilters, and as optical strain
gauges in civil and aerospace engineering [7], and, only recently, in HEP detectors [8]. The BTeV detectors
utilize Fiber Bragg Grating (FBG) sensors to monitor online the positions of the straws and microstrips.
The optical fiber is used for monitoring displacements and strains in mechanical structures such as the
straw tube-microstrip support presented here. A modulated refractive index along the FBG sensor produces
Bragg reflection at a wavelength dependent on the strain in the fiber, permitting real-time monitoring of the
support. According to these properties, an FBG sensor is going to be placed in the M0X structure between
the Rohacell
r
foam and the CFRP shell. Sensors will be located in spots of maximal deformation, as pre-
dicted by FEA simulation. Fig.5 shows long-term behaviour of FBG sensors while monitoring micron-size
displacements, compared to monitoring via photographic methods.
7 PROTOTYPE
M0X prototypes have been fabricated in order to study the construction procedures, mechanical properties,
material characterization, and physical behaviour for detection of particles in test beam set-ups. Straw ma-
terials such as mylar and kapton have been studied and characterized for tensile properties under exposure
to Ar-CO
2
mixtures [9]. The most demanding design requirement is the assembly of straws in a close pack,
with no mechanical tension applied. Several gluing techniques have been examined and tested to deter-
mine the optimal technique. Straw tubes are glued together in three layers, and the upper and lower layer
are glued to the Rohacell
r
foam. Glues with different viscosity, and several gluing techniques, have been
used. Glues tested range from cyanoacrylate (Loctite 401) to epoxy (Eccobond series). Gluing techniques
ranged from brush, to injection, to spray gluing. The most promising results have been obtained by using
an Eccobond 45W and catalyst mixture (1:1 by weight), diluted with dimethylcheton solvent. For each 40g
of glue-catalyst mixture, 40cm
3
of solvent was used.
The assembly process proceeds as follows. Stainless steel rods (4-mm-diameter) are inserted in each straw
4
Page 4
tube. A straw layer is formed by locating 16 straws on a machined grooved plate. The glue-solvent mix-
ture described is sprayed, with 2bar air pressure, and 20cm distance between spray gun and straw layer.
Mechanical pressure is applied to layers during curing. After curing at room temperature, the straw layers
are sprayed again and more layers are superimposed. After additional curing, the stainless steel rods are
removed from the straws. The mechanical pressure applied during curing allows loose (0.1%) requirements
on rod straightness. Conductive contact between straw cathodes and aluminum endplate, is accomplished
via spraying of Eccobond 57C. Fig.6 shows a complete prototype after gluing, with endplate for wiring on
one side. The glued layers of straws provide excellent stiffness (see Sect. Tomography) for operation, with-
out need of Rohacell
r
foam which, instead, contributes to mechanical properties when used in integration
with microstrip detector. Fig.7 shows detail of the CFRP shell near the beam pipe region (left), and the
grooved foam (right).
8 COSMIC RAY AND TEST BEAM RESULTS
Preliminary results with cosmic rays show very clean pulses (Fig.8) in gas mixtures of interest for BTeV
(Ar-CO
2
80/20), with shape parameters typical of operation with such mixtures. High-voltage applied
is +1400V, a trans-impedance preamplifier [10] provides a 2V/mA gain, followed by a low-walk double-
threshold discriminator [11]. Prototypes have been exposed to beam particles in the Frascati Test Beam
Facility [12]. Preliminary results show the expected response of prototype to minimum ionizing particles.
The distribution of drift time of the ionization electrons to the sense wire (Fig.9 left) over the straw 2mm
radius is compatible with the drift velocity in the Ar-CO
2
(80/20) gas mixture used. A 10mV threshold is
applied. Beam particles tracks are reconstructed by the M0X prototype, tracks residuals are shown in Fig.9
right. The distribution of residuals is well fitted by a Gaussian with a rms width of about 130µm.
9 CONCLUSIONS
We have developed a novel concept for integration of straw tubes tracking detectors and silicon microstrip
detectors, for use in HEP experiments at hadron colliders. In our design, silicon microstrips are integrated
to a special straw tube module M0X via a CFRP mechanical structure. M0X is realized via glued straws
embeddedin a Rohacell
r
lattice with no need of mechanicaltension. Detailed finite element analysis shows
that deformations affect negligibly the tracking performance of the system. A complete system based on
Fiber Bragg Grating sensors - acting as optical strain gauges - monitors the position of each sub detector
with micron resolution. Test beam studies are underway to verify that M0X can provide the 200µm resolu-
tion needed by the BTeV tracking detector requirements.
5
Page 5
ACKNOWLEDGEMENTS
We thank G.Mazzitelli (LNF INFN, Italy) and all the DAΦNE team for smooth running of the Beam Test
Facility.
References
[1] Fermilab Experiment E-0897/E-0918,J.Butler, S. Stone co-spokespersonsw; see www-btev.fnal.com.
[2] S.H.Oh et al., Nucl. Instr. Meth. A309 (1991) 368-376.
[3] T. Akesson et al., Nucl. Instr. Meth. A522 (2004) 131-145.
[4] K. Nuenighoff et al., Nucl. Instr. Meth. A477 (2002) 410-413.
[5] E. Basile, ”Scelta dei materiali ed analisi strutturale per supporti di rivelatori di particelle
dell’esperimento BTeV a Fermilab (U.S.A.)”, degree thesis, University ”La Sapienza”, Rome, 2003
(in Italian). Also available at http://www-btev.fnal.gov/cgi-bin/public/DocDB/ShowDocument
[6] C. Pucci, ”Analisi strutturale del supporto per microstrip straw tubes per l’esperimento di fisica delle
particelle BTeV”, degree thesis, University ”La Sapienza”, Rome, 2004 (in Italian). Also available at
http://www-btev.fnal.gov/cgi-bin/public/DocDB/ShowDocument .
[7] S. Berardis et al., ”Fiber optic sensors for space missions” 2003 IEEE Aerospace Conference Proceed-
ing, Big Sky Montana, March 8-15, 2003, pp. 1661-1668
[8] L. Benussi et al., ”Results of Long-TermPosition Monitoring by Means of Fiber Bragg Grating Sensors
for the BTeV Detector”, Frascati preprint LNF - 03 / 15(IR)
[9] E.Basile et al., ”Study of Tensile Response of Kapton, and Mylar Strips to Ar and CO2 Mixtures for the
BTeV Straw Tube Detector”, presented by F. Di Falco at 10th Vienna Conference On Instrumentation
16-21 Feb 2004, Vienna, Austria, LNF - 04 / 5(P).
[10] L.Benussi et al., Nucl. Instr. Meth. A361 (1995) 180-191
[11] A.Balla et al., Nucl. Instr. Meth. A461 (2001) 524-525
[12] G. Mazzitelli et al., Nucl. Instr. Meth. A515 (2003) 524-542
6
Page 6
Figure 1: BTeV detector layout; the straw tube chamber are in yellow, the silicon strips in red.
Figure 2: Exploded view of BTeV microstrip and straws tubes integration (left); M0X prototype with straw
tubes embedded in Rohacell
r
(right).
7
Page 7
Figure 3: 2-dimensional images of 6-channel M0X prototype (left); 3-dimensional reconstruction (center);
2-dimensional image of final assembly technique M0X prototype (right). Tomographs are for straws at the
foam-supported stage.
Figure 4: FEA results for the simulation of M0X (straws-Rohacell
r
-CFRP) straw-microstrip detector. Ge-
ometry and materials are shown in Tab.1. Colour levels show curves of equal deformation under simulated
load of microstrip detector and associated electronics. Units are micrometers. Maximum deformation is
14.7 µm (arrow).
8
Page 8
Figure 5: FBG long-term monitoring stability results. FBG output (crosses) is validated by TV camera
(bars). The bar size indicates the best resolution of TV camera.
Figure 6: M0X module prototype during assembly. Straw tubes are glued together and positioned between
end-plates (one shown) without mechanical tension. Rohacell
r
foam and CFRP shell not shown.
9
Page 9
Figure 7: CFRP shell prototype immediately after fabrication (left); grooved Rohacell
r
foam (right).
Figure 8: Cosmic rays signals in M0X prototype with (Ar-CO
2
80/20) gas mixture. High-voltage applied
is 1400V, a transimpedance preamplifier provides a 2V/mA gain.
10
Page 10
Figure 9: Distribution of drift times from beam particles. Times are expressed in time-to-digital-converter
counts (300ps/count), Ar-CO
2
(80/20) gas mixture is used, with 10mV threshold (left); tracks residuals of
beam particles tracks reconstructed by the M0X prototype, the distribution of residuals is well fitted by a
Gaussian with a rms width of about 130µm (right).
11
Page 11