1 Assessment of the Čerenkov light produced in a PbWO 4 crystal by means of the study of the time structure of the signal L. Cavallini, C. Mancini, A. Mecca and D. Pinci Università La Sapienza and INFN Sezione di Roma, Italy E-mail:davide.pinci@roma1.infn.it N. Akchurin, L. Berntzon, H. Kim, Y. Roh and R. Wigmans Texas Tech University, Lubbock, USA A. Cardini INFN Sezione di Cagliari, Italy R. Ferrari, S. Franchino, G. Gaudio and M. Livan Università di Pavia and INFN Sezione di Pavia, Italy J. Hauptman Iowa State University, Ames, USA L. La Rotonda, E. Meoni, A. Policicchio and G. Susinno Università della Calabria and INFN Sezione di Cosenza, Italy H. Paar University of California at San Diego, La Jolla, USA A. Penzo INFN Sezione di Trieste, Italy S. Popescu 1 and W. Vandelli CERN, Geneva, Switzerland 1 on leave from IFIN-HH, Bucarest, Romania On beam tests were carried out on PbWO 4 crystals. One of the aims of this work was to evaluate the contribution of the Čerenkov component to the total light yield. The difference in the timing characteristics of the fast Čerenkov
2 signals with respect to the scintillation ones, which are emitted with a decay time of about 10 ns, can be exploited in order to separate the two proportions. In this paper we present the results of an analysis performed on the time structure of signals, showing how it is possible to detect and assess the presence and the amount of Čerenkov light. Since Čerenkov light is emitted only by the electromagnetic component of a hadronic shower, its precise measurement would allow to account for one of the dominant sources of fluctuations in hadronic showers and to achieve an improvement in the energy resolution of a hadronic calorimeter. Keywords: Čerenkov light, PbWO 4 crystal, high energy particle calorimetry. 1. Introduction The possibility of evaluating the amount of Čerenkov light produced by fast particles in an hadronic calorimeter represents a promising method to detect the electromagnetic fraction of an hadronic shower. 1,2 Long and detailed on-beam tests were performed on PbWO 4 crystals, which represents an attractive material for detecting electromagnetic showers on high energy colliders because of its high density and fast response. 3 5 One of the main purposes of the test was to evaluate the contribution of the Čerenkov component to the total light yield. In order to assess the presence of Čerenkov light and to evaluate its ratio to the scintillation one, two of their main differences were exploited: (1) While the scintillating light is emitted isotropically, the Čerenkov light is produced in a cone with an opening angle cosθ = 1/βn as shown in Fig. 1. (2) The emission process of the scintillating light, produced by the molecular de-exitation, has a characteristic decay time (about 10 ns for PbWO 4 crystal) while the Čerenkov light is produced prompt after the particle crossing. 2. Study of the signal time structure In order to evaluate the effects of Čerenkov photons to time structure of the PMT signals, during the on-beam tests the PMT signal pulse shapes were acquired with a very fast Flash ADC (effective sampling frequency of 800 MHz) or with a 10 Gigasample/s oscilloscope. 2.1. Signal timing properties The timing properties of the signals have been analysed by two simple methods shown in Fig. 2. In a first method the leading edge of the signal
3 Fig. 1. Directionality of the Čerenkov light and sketch of the setup used to study the difference of the responses on the two crystal sides. is fitted to a Fermi-Dirac function: [ ] 1 V (t) = A e (t tl)/τl + 1 1 An increase in the Čerenkov content of the signal will manifest itself as a decrease in the value of the leading constant τ L, since the leading edge is becoming steeper. In a second method we evaluated crossing time, i.e. the time at which the pulse height crosses a certain threshold level (e.g. 30 mv in Fig. 2 right). An increase of the Čerenkov component will shift that point to an earlier moment. Fig. 2. Methods to determine the time structure of the signal. On the left the leading edge is fitted with a Fermi-Dirac. On the right the time at which the pulse height exceeds a certain threshold is used to this purpose. In Fig. 3 the leading constant and the difference between the R and L crossing times are shown as a function of the crystal angle respectively on the left and on the right. The effects of Čerenkov photons on the signal timing are visible on both the variables. The precision shown for the data at 0 o, is completely dominated by
4 Fig. 3. Left: angular dependence of the leading constant τ L. Right: difference of the times needed by the signals in L and R to reach a fixed threshold as a function of the angle. photo-electron statistics and are such that the used variables do not provide statistically significant information about the contribution of Čerenkov light to the signal in question on a event by event basis. The difference between the crossing times on L and R was also measured by means of a standard CAMAC discriminator and a TDC without any offline pulse shape analysis (Fig. 4). Although a maximum time difference Crossing Time Difference L-R (ns) 0.8 0.6 0.4 0.2-0 -0.2-0.4-0.6-0.8-50 -40-30 -20-10 0 10 20 30 40 50 Angle (degrees) Fig. 4. Results of the measurement of the crossing time difference by means of standard CAMAC electronics. of only about 1 ns was found, the effect is clearly visible. 2.2. The qratio method A third method to evaluate the Čerenkov contribution to the total light yield is to calculate the ratio between the light collected in the first few instants of the signal and the total one. This charge ratio (qratio) is expected to increase when the Čerenkov light is collected because the signal becomes faster and higher. In Fig. 5 very preliminery results on the behaviour of qratio as a function of the angle are shown for the R side signals. The value of qratio is almost constant over the whole range of angles in which
5 Fig. 5. Results of the qratio analysis (see text) for the crystal R side. no Čerenkov light is collected by the PMT R. For θ < 0 it starts to increase reaching its maximum for θ 30 o. The maximum value is about 10% (once offset-subtracted) which is equal to the ratio between the Čerenkov and the total light calculate with signal shape analysis. 3 No effects due the increase of the effective particle path are found for large angles. 3. Conclusion Several variables sensitive to the effects of the Čerenkov light correlated with the signal time structure were studied. The prompt Čerenkov photons give rise to a fast signal whose time characteristics can give information about their presence and their amount. The use of the qratio method, which adds to the effects on the signal timing the effects on the increase of total charge can represent a promising way to assess the Čerenkov contribution to the total light yield also in homogeneous PbW0 4 calorimeters. References 1. N. Akchurin et al., Nucl. Instr. and Meth. A 537 (2005) 537. 2. N. Akchurin et al., (DREAM Collaboration), Dual-readout calorimetry with lead tungstate crystals, Nucl. Instr. and Meth. 2007, submitted for publication. 3. N. Akchurin et al.. Contributions of Čerenkov light to the signals from lead tungstate crystals in press on Nucl. Instr. and Meth. A. (arxiv:0707.4013). 4. N. Akchurin et al. Dual-Readout Calorimetry with Lead Tungstate Crystals in press on Nucl. Instr. and Meth. A. (arxiv:0707.4021). 5. N. Akchurin et al. Measurement of the contribution of neutrons to hadron calorimeter signals in press on Nucl. Instr. and Meth. A. (arxiv:0707.4019).