stive Plate Chambers as Active Medium for the HCAL
5/27/03
Task A Supplemental Request: Study of Resistive Plate
Chambers as Active Medium for the HCAL

Faculty:
Associate Professor John M. Butler


Assistant Professor Meenakshi Narain

Engineer: David
Osborne


Project Overview

The optimal application of Energy Flow Algorithms for the measurement of hadronic jets
requires a finely segmented electro-magnetic and hadronic calorimeter (HCAL). The
latter is envisaged to be a sandwich type calorimeter and to contain cells of the order of
1cm
2
, read out separately for each layer. The resulting number of readout channels is
approximately 5x10
7
. Detector simulation studies have demonstrated that, given the fine
segmentation, the energy resolution is preserved with only a digital readout of the HCAL.

Our aim is to develop an active medium for the HCAL, which is reliable, simple to build,
comparatively thin (under 10 mm), with the ability to be segmented laterally into cells of
1 cm
2
,

and affordable. Resistive Plate Chambers (RPCs) have been utilized in a number
of HEP experiments over the past decade and appear to satisfy all of the above
requirements. We propose to study the suitability of RPCs as active medium for the
HCAL.


Description of Resistive Plate Chambers

A sketch of a generic Resistive Plate Chamber is shown in Figure 1. Two parallel plates
of high resistivity, = 10
10
to 10
12
cm generate a uniform, intense electric field, about
4kV/mm, in a typically 2 mm wide gas gap. The plates are coated, on the external sides,
with thin graphite layers connected to high voltage and ground, respectively. Due to their
high surface resistivity of about 100 k/ , these graphite electrodes are transparent to the
transients of electrical discharges generated in the gas. Capacitive signal readout is
therefore possible through pads which are insulated from the graphite carrying the high
voltage
by
a
layer
of
mylar.





Page 1 of 6
5/27/03


Resistive Plate
Pick- up pads
Mylar

HV
Gas
Graphite

Figure 1: Sketch of a generic Resistive Plate Chamber.


The simplicity of the concept of these chambers allows for a large variety of design
choices. Two types of resistive plates have been used for the construction of most
chambers: glass and bakelite. The advantage of bakelite is its somewhat faster recharging
capability; however the optimal performance requires the application of a coat of linseed
oil to the inner surface of the plates - a somewhat delicate operation. Chambers have been
built with one single gap, some as wide as 8 mm, or multiple and smaller gaps for better
timing resolution at uncompromised signal efficiency. The thickness of the glass plates,
typically 2 mm, can be varied; however thinner plates will be distorted by the forces
resulting from the high electric field between the plates. The resistivity of the graphite
layer can be varied and will affect both the rate performance and the amount of cross-talk
between adjacent readout pads. Finally, the chambers can be operated either in the
avalanche mode (at a lower high voltage setting) or in streamer mode. The collected
charge in the streamer mode is approximately a factor of 50 larger than in the avalanche
mode. Different gas mixtures have been explored, some with the ability of efficiently
suppressing streamers.

In general, the physics of RPCs is well understood and Monte Carlo programs exist
which simulate the various physics processes occurring when the chambers are traversed
by a particle. More details on the current status of research related to RPCs and of recent
applications of RPCs in HEP can be found in the contributions to last years workshop
dedicated to these chambers [1].




Rate considerations

RPCs are slow to recharge: typically recharging times of 1 (streamer mode) and 0.1 ms
(avalanche mode) have been observed. The times depend on several factors, such as the
resistivity of the graphite layer, the material and the resistivity of the plates and the
applied high voltage.

Page 2 of 6
5/27/03
Assuming a linear collider operating at s=500 GeV with a luminosity of 0.5·10
34
cm
-2
s
-1
,
the rate of annihilation events is approximately 1 event every 50 s and, therefore, the
recharging rate of the chambers is not a concern. More worrisome is the rate of 2 events
leading to pairs of muons or to hadrons traversing or entering the hadron calorimeter.

The total cross section for e
+
e
-
e
+
e
-
µ
+
µ
-
(e
+
e
-
e
+
e
-
h) is estimated [2] to be 420 nb
(162 nb), leading to a rate of 2.1·10
3
muon pairs (810 hadron events) per second. These
rates are comparable to the recharging rate of the chambers when operated in streamer
mode and, given the fine segmentation of the hadron calorimeter, should not pose a
problem.

The calculation for the process e
+
e
-
e
+
e
-
h has also been verified using the PYTHIA
Monte Carlo Program. The resulting cross section is somewhat smaller, 34 nb, leading to
a rate of 171 events per second [3]. Figure 2 shows the rate of particles/100 s as a
function of the polar angle at the production vertex.



Figure 2: Rate of particles per 100 seconds from the process
e
+
e
-
e
+
e
-
h as a function
of
the polar angle, as predicted by PYTHIA.



Assuming the geometry of the TESLA detector as publicized in the TESLA Technical
Design Report, these events generate a negligible rate of particles in the barrel hadron
calorimeter. However, each endcap calorimeter sees a rate of approximately 613 particles
(mostly pions) per second with an average energy of 1.5 GeV. A cut at 1 GeV, as an
estimate of the amount of energy required to traverse the electromagnetic calorimeter
located in front of the hadron calorimeter, reduces the rate to 283 Hz. Again, given the
fine segmentation of the calorimeter, this rate should not be a problem for the RPCs.

Page 3 of 6
5/27/03
The rates for all 2 processes increase logarithmically with energy. For instance, the cross
section for e
+
e
-
e
+
e
-
h increases from 162 nb at s = 500 GeV to 189 nb at s = 800
GeV.


Description of project activities

We will initiate a detailed R&D program to evaluate the merits of RPCs as active
medium of the HCAL:

1) We will complete the evaluation of a small number of RPCs which we obtained
from other experiments.
2) We will construct a small number of test chambers with various
- glass and gas gap thicknesses
- resistivity of the layers of ink (distributing the high voltage onto the glass)
- geometries of the readout pads.
3) We will develop a readout system based on a one-level discriminator. This system
will be used to evaluate the different chamber designs and pad geometries.
4) We will test these chambers in a cosmic ray test stand and evaluate their:
- noise characteristics
- signal strength versus applied high voltage and for different gas mixtures
- efficiency for the detection of minimum ionizing particles
- cross talk between adjacent channels
- long term stability
5) Following the completion of the above tests, we will design and build a small test
section of an (electro-magnetic) calorimeter, approximately 25 cm in all three
dimensions. This test section will feature of the order of 10,000 readout channels.
The electronic readout system will be based on a custom chip. The mechanical
set-up will be designed such as to allow for easy implementation of other active
media, as they might become available.
6) We will test this calorimeter in particle beams which are available at the major
particle physics laboratories, such as DESY and CERN or elsewhere. These tests
will be important in verifying the functionality of the chambers and their
electronic readout system.
7) We will initiate long-term tests of our prototype chambers to assess their stability
over long periods of time and to detect any possible aging effects.
8) Contingent on the successful tests of our small (electro-magnetic) calorimeter, we
will design and build a test section of the hadronic calorimeter, sized
approximately 1 m
3
, which is sufficient to contain hadronic showers both laterally
and longitudinally. This calorimeter section will again be subjected to extensive
tests in particle beams.

We expect to complete items 1) 4) in FY 2003, items 5) 6) in FY 2004, initiate item
7) in FY 2004 and item 8) in FY 2005.


Engineering and technical effort during FY2003
The following engineering and technical activities are planned for FY2003:
Page 4 of 6
5/27/03

1) Construction of a small number of test chambers with different dimensions (glass
and gas gap thicknesses.) This involves the cutting and gluing of glass.
2) Development of a technique to apply layers of resistive ink (graphite) with
different, but uniform thicknesses, leading to specific values of the surface
resistivity.
3) Design and production of spacers and rims (out of plastic) for the construction of
the chambers needed for the assembly of the electro-magnetic size calorimeter.
4) Design and building of a readout system for the test chambers and possibly for the
electro-magnetic size calorimeter. The readout scheme will include only the digital
information. The