SILICON PIN DIODE HYBRID ARRiYS FOR CHARGED PARTICLE DETECTION: -BUILDING BLOCKS FOR VERTEX DETECTORS AT THE SSC”

Two-dimensional arrays of solid state detectors have long been used in visible and infrared systems. Hybrid arrays with separately optimized detector and readout substrates have been extensively developed for infrared sensors. The characteristics and use of these infrared readout chips with silicon PIN diode arrays produced by MICRON SEMICONDUCTOR for detecting high-energy particles are reported. Some of these arrays have been produced in formats as large as 512 x 512 pixels; others have been radiation . . :. hardened to total dose levels beyond 1 Mrad. Data generation rates of 380 megasampies/second have been achieved. Analog and digital signal transmission and processing techniques have also been developed to accept and reduce these high data rates. INTRODUCTION -High-resolution vertex and tracking devices at the SSC will be important for the reconstruction of complex events, including secondary vertices close to the primary interaction point. The main properties required of these devices are: fast response time, fine spatial resolution, multiple particle resolution, and radiation hardness. A resolution of better than 10 pm and the ability to distinguish the many particles of a jet can be achieved with two-dimensional pixel devices. Due to the three-dimensional nature of the coordinate information provided, these provide efficient track finding with a minimum number of layers in the high-multiplicity environment of the SSC. The SSC beam crossing period will be 15 ns, and interesting events will occur every LOO-to 10,000 crossings. Hit times must be recorded to an accuracy of one crossing period, and the readout time of the detectors must be less than the mean interval between interesting events to avoid large dead time losses. *Work supported by the Department of Energy, contract DE-AC03-76SF00515. **Permanent address: Electra-Optical & Data Systems Group, Hughes Aircraft Company, El Segundo, CA 90245. Presented at the 1989 International Industrial Symposium on the Super Collider, New Orleans, LA, February 8-10, 1989. The expected combination of high luminosity and large particle multiplicity will produce a high-radiation environment in the vicinity of the collision point. For a proton. proton total cross section of approximately 100 mbarn, the SSC will yield approximately lo* interactions per second when operating at its design luminosity of 1033/cm2/sec. At a distance of 5 cm from the beam, the absorbed radiation dose will be approximately 1 Mrad per year. New electronic detector systems usually require at least several years for development. Examples are CCDs, infrared detector arrays and x-ray bolometers. The schedule for the SSC construction is commensurate with the schedule for new vertex detector development if the latest existing technology is used as a base. DEVICE CANDIDATES Candidates for vertex/tracking devices at the SSC include: wire chambers, scintillating fiber detectors, silicon microstrip detectors, double-sided microstrip detectors, silicon drift chambers; and pixel devices in the form of CCD arrays, monolithic silicon arrays and hybrid microdiode arrays. It is generally recognized that in terms of space, momentum and two-track resolution, wire chambers at present are inferior to silicon de.-.---vices. It is also generally recognized that pixel devices are preferable to strip devices or drift devices, as far as two-track resolution is concerned. Pixel devices-in particular, silicon diode arrays-are a natural choice for vertex detectors. These devices provide three-dimensional coordinate information with a spatial resolution of a few microns. However, as we move farther from the collision point to the tracking detector, many of the other devices mentioned above become attractive candidates for a variety of reasons. . . A vertex detector based on the use of a pixel device would provide efficient trackfinding with-a minimum number of layers in the high-multiplicity environment of the --. SSC due to the three-dimensional nature of the coordinate information provided. The absence of ambiguities in coordinate matching which are always present in nonpixel devices allows the number of detector layers to be minimized, thereby reducing the size and _ cost of the-vertex detector. The only pixel devices presently in use in high energy physics experiments today are CCBs.‘~‘These devices are inappropriate to the SSC environment for a variety of reasons, including: (i) a small signal size of about 1000 electrons, which necessitates their being cooled, renders them extremely susceptible to degradation caused by radiation damage, and requires costly and delicate readout electronics; (ii) a varying transfer efficiency across their faces; (iii) no off gate during readout, resulting in a time/space ambiguity; and (iv) a serial readout resulting in a readout time of about 10 msec for a device with lo5 pixels. A number of researchers are presently working on the problem of fully integrated 2’3 pixel detectors in monolithic arrays. The goal is to fabricate the readout electronics on the same high-resistivity silicon as the detector diodes. Success in these efforts will result ie detectors having the minimum thickness possible. However, there will be a loss of flexibility. A change in either the detector specification or the readout specification will result in not only a circuit change, but also a re-analysis of the production process, since they are now intimately coupled. There will be difficulty in production, since both the detector and the readout electronics must be fabricated in high-resistivity facilities which can handle this new process. There are far fewer of these facilities worldwide than there are those which can produce more conventional circuits.