Proceedings of VII International School-Seminar on Heavy ion Physics, Dubna, Russia, May 27 - June 1, 2002 Submitted to Ядерная Физика at

Transcription

1 NEW INDICATIONS OF COLLINEAR TRIPARTITION IN 252 CF (SF) STUDIED AT THE MODIFIED FOBOS SETUP Yu.V. Pyatkov 1,2), D.V. Kamanin 1)*, A.A. Alexandrov 1), I.A. Alexandrova 1), S.V. Khlebnikov 3), S.V. Mitrofanov 1), V.V. Pashkevich 1), Yu.E. Penionzhkevich 1), Yu.V.Ryabov 4), E.A. Sokol 1), V.G. Tishchenko 1), A.N. Tjukavkin 2), A.V. Unzhakova 1), S.R. Yamaletdinov 1) 1) Joint Institute for Nuclear Research, Dubna, Russia 2) Moscow Engineering Physics Institute, Moscow, Russia 3) V.G. Khlopin Radium Institute, St. Petersburg, Russia 4) Institute for Nuclear Research RAN, Moscow, Russia An attempt has been made to search for collinear tripartition in spontaneous fission of 252 Cf by using the FOBOS setup coupled with the neutron detector belt. The group of rare events were detected which characterized by reduced both the total kinetic energy and total nuclear charge gated by the large neutron multiplicity measured. This fact is considered to be an experimental indication of collinear tripartition in 252 Cf. The theoretical indication of the possible existence of the collinear cluster tripartition valley was obtained for the first time. The present paper is devoted to the search for the particular case of true ternary fission at the lowest excitation, i.e. during spontaneous decay. In the early experiments [1] at the FOBOS detector [2] aimed at the study of rare fission modes in spontaneous decays of 248 Cm and 252 Cf a group of the precisely collinear pairs of heavy fragments with a large deficit both of the total mass and of the total kinetic energy has been detected. This group lies rather far from the locus connected to conventional binary fission in the correlation plot of the fragment masses (selected area, Fig.1) and its yield in the lowest limit amounts to ~ of the whole data body. The mass-energy correlations for these rare events allow one to assign them to the system fission via an elongated three-body chain like configuration. In this case two outside fragments fly apart along the chain axis and can be confused with ordinary fission fragments (FF), while the central fragment can stay almost in rest and, as a result, can hardly be registered. Such an experiment especially suffers from the background of false events due * Corresponding author: kamanin@fobos.jinr.ru 1

2 to the partial loss of the FF energy in the construction elements of detector modules. The results obtained have been treated as an indication of collinear cluster tripartition (CCT) of the heavy nuclei under study. The advanced experiment aimed specially at the investigation of CCT of the 252 Cf nucleus has been performed at the modified 4π spectrometer FOBOS at the Flerov Laboratory of the JINR [3]. In order to overcome the above mentioned method obstacles and to improve the quality of the data some modifications have been introduced into the experimental scheme. These modifications concerned the configuration of the detectors including the start detector, the electronics and also the data acquisition system. Some preliminary results of the recent experiment are discussed below. Fig. 1. Fission fragments mass-yields matrix Y(Ma,Mb) for 252 Cf(sf) decay. The FF trigger consisted of two groups each containing six FOBOS modules used as a double-armed spectrometer for measuring the FF velocities (V) by means of measuring their time-of-flight and energies (E). Thus both the «2-V» and the «V-E» methods are accessible for the calculation of the FF masses. In using the latter method one does not need to use any a-priori assumptions concerning the process to be exclusively binary. 2

3 The double-armed configuration of the FOBOS spectrometer consisted of the specially designed wide aperture start-detector with an internal FF source [3]. The full symmetrization of the spectrometer arms achieved due to such a start detector improves the quality of the data substantially. According to the model of the CCT process proposed in Ref. [2] the central fragment of the three-body pre-scission chain gains almost the entire deformation energy of the system. Being presumably in rest it would be an isotropic source of post-scission neutrons of a high multiplicity (~10) in the lab system. On the contrary, the neutrons emitted from the moving fission fragments are focused along the fission axis. In order to exploit this phenomenon for revealing the CCT events the neutron belt [4] consisting of 140 separate hexagonal modules based on 3 He-filled proportional counters was assembled in a plane being perpendicular to the symmetry axis of the spectrometer, which serves as the mean fission axis at the same time. The center of this belt coincides with the location of the FF source. The typical spectrum of frequency versus the number of the tripped neutron counters for the registered fission events is presented in Fig. 2. The function obtained agrees well with the theoretical calculations based on the known probabilities for emitting a certain number of neutrons per fission meaning the total registration efficiency of about 3.8% [4]. The overall registration efficiency for an isotropic source and, hence, for the neutrons from the CCT events amounts to ~11%. A simple calculation accounting for this difference in the registration efficiencies and in the primary multiplicity spectra ascribed to the processes considered (see Ref. [4]) reveals that the registration probability for more than three neutrons from ordinary spontaneous fission in this geometry is two orders of magnitude lower than the same probability for the CCT events. Our empirical model described in [4] reproduces the shape of the multiplicity distribution from 0 to 3 precisely. In particular, the best fit presented in Fig. 2 gives the average multiplicity differing from the experimental one only by the value of and such a good reproducibility of the distribution shape is found to be stable against reasonable variations of the parameters of the model. The latter fact is extremely important for the tail of the distribution and, therefore, we checked the most important constants additionally by using the simulations with the well-known MCNP code. 3

4 1 0,1 0,01 <ν sim >= <ν exp >= Experiment Simulations total true Probability 1E-3 1E-4 1E-5 1E-6 2'812' '479 33'108 1' E Measured neutron multiplicity Fig. 2. The spectrum of frequency versus number of tripped neutron counters for the recorded fission events. The actual numbers of counts are labeled. True contribution is conditioned only by the number of emitted neutrons being not less than the measured multiplicity. One should note that the lowest estimate of ~ of the yield of CCT with respect to ordinary fission formally assumes the contribution of at least 3-30 CCT events to the multiplicity spectrum in Fig. 2 and these events should contribute mainly to high-multiplicity values. On the other hand, we did not include deliberately the contribution of CCT into our simulations. Although there is no experimental data on multiplicity 6 and higher the best fit misses surprisingly ~20 events for the measured values of 4 and 5. Based on the statistics of 3 million events presented in Fig. 2 we would not insist on the evidence of some additional high-multiplicity process. However, the observed difference between the experiment and simulations we treat as an additional argument for searching the CCT events under the condition of ν exp >3. In addition, according to our simulations the contribution of the true high-multiplicity events with ν emitted >3 to the measured multiplicity greater than 3 amounts to ~80% (Fig. 2) 4

5 while the rest is mostly due to random coincidences with ν emitted =3. This mean the reliability of the experimental data selected by such a large number of fired neutron counters More than 3 netrons detected 120 Ab/ channels E b / MeV selected events E a / MeV A a /channels Fig.3. Amplitude distribution Y(A a,a b ) of the complimentary fragments under the condition that at least 4 neutron counters were fired. The first interesting result can already be discussed based on a part of the whole data body counted, 1.4*10 7 events. Fig.3 shows the amplitude distribution Y(Aa,Ab) of the fragments detected in coincidence in two arms of the spectrometer (labeled a and b) gated by the large neutron multiplicity ν exp >3, as is discussed above. The set of points which looks like a parabola (denoted as selected events in Fig. 3) attracts one s attention. For comparison, a similar plot obtained for the events falling inside the contour shown in Fig.1 is given in Fig.4, although this plot is not gated by the number of the fired neutron detectors. The events forming an angle-like structure in Fig.1, i.e. linked with tripartition, are connected by the parabolic curves. 5

6 Fig.4. Energy distribution Y(Ea,Eb) of the complimentary fragments obtained for the events falling inside the contour marked in the Fig At least one neutron detected 140 Aa/ channels E a / MeV ,0 2,2 2,4 2,6 2,8 3,0 3,2 3,4 3,6 3,8 4,0 t a / ms Fig. 5. The distribution of the FF amplitude versus drift time measured in a BIC when at least one neutron detector is fired. 6

7 The serious test of the reliability of the CCT events is the low total charge of the both detected fragments, less than Z C /2, where Z C is the charge number of the fissioning nucleus. Unfortunately the Bragg-spectroscopy is out of rule for the heavy ions whose energy is typically lower than 1 AMeV for the FF in spontaneous fission. In order to perform the test for the total charge an additional parameter was recorded for each fission fragment registered. This alternative method was proposed in Ref. [5]. It is based on measuring the delay between the time the fragments enter the Bragg ionization chamber (BIC) and the time the anode pulse crosses a given level, i.e. the parameter connected with the drift time of a charge in a BIC larger charge More than 3 netrons detected 120 Aa/ channels E a / MeV smaller charge 60 <t a > selected events 40 2,5 2,6 2,7 2,8 2,9 3,0 3,1 3,2 3,3 3,4 t a / ms Fig. 6. The same distribution as in Fig. 5 when at least four neutron detectors fired. The selected events are the same as in Fig. 3. The distribution of the FF amplitude A a versus drift time t a obtained in our experiment for one of the detectors is shown in Fig. 5. The FF fall to the light and the heavy mass peaks which are easily distinguished. An individual charge at a given mass should look like a steep power function. Of course, numerical simulations and a direct calibration are needed for an accurate analysis of such a complicated parameter as t a. Indeed, the drift time t a is the function of the nuclear charge of the FF, its energy and, less pronounced, its mass simultaneously. 7

8 However a preliminary conclusion can be drawn from the drift time vs amplitude plot shown in Fig. 6, which is accumulated under the same condition of the neutron multiplicity ν exp >3 as Fig. 3. For a usual fission event one should expect that the measured total charge of complimentary FF is, on the average, equal to the charge of the Cf nucleus, i.e. <Z a >+<Z b >=98. This should mean that the complimentary FF has to be found, on the average, on different sides of the mean charge <t a > in Fig. 6. However, all the selected events are located on the left and downward from <t a > thus revealing a lower mean fragment charge whereas no events are found on the other side. Hence, despite of a poor charge resolution the average total charge for the selected events is at least notably lower than 98 with a high probability. In order to make an ultimate conclusion about the total charge one should consider the correlated FF pairs and trace down the lines of individual charges. In addition to the experimental work we continue the theoretical study of CCT. A significant support for the CCT hypothesis was obtained recently in our more detailed calculations of the potential energy surface (PES) of the Cf nucleus carried out in the framework of the procedure presented in [6,7] based on the Strutinsky method. Fig. 7a depicts the shape of the fissioning nucleus at the bottom of the symmetry valley) with the quadrupole moment Q 2 =7.52a.u. (see fig.8 in ref. [7]). As was already pointed out in our previous works [7-10], the system which fissions in the vicinity of the bottom of the potential valley constitutes two magic nuclei (clusters) connected by a neck. In fig.6a these clusters are the deformed magic nuclei of 106 Mo 64 (β 2 ~ 0.58). In the calculations the shape of the system was varied in such a way that a value of Q 2 parameter remained constant while a massasymmetry η changed starting from the value corresponding to the valley bottom. By definition η= (M 1 -M 2 )/M c, where M 1,2 - is the mass of the system concentrated respectively, on the left and on the right sides of the varied boundary, which divides the nuclear body onto two parts (marked by vertical lines in fig.7); M c is the mass of the fissioning nucleus. As a result, the new shape of the system shown in fig.7b was revealed for the first time. The energy of the system is only slightly higher (by ~2 MeV) than the corresponding value at the bottom. The distinguishing 8

9 feature of the shape observed is the double belt which vividly divides the system into three parts comparable of sizes. It would appear reasonable to identify the double rupture of such a configuration as the true ternary fission so long looked for. The fact that has not been discovered before may be due to the special kinematics of the process being collinear in nature. Fig. 7. The shape of the nucleus at the bottom of the symmetric valley (Q 2 =7.52, η=0.074) - (a); the same system at the point Q 2 =7.52, η= (b) As mentioned above, two outside fragments of the three-body chain like configuration should fly apart along the chain axis while the central one could stay almost in rest. In addition, one might expect that the scission probability achieves its maximum value when the length of the system is close to the sum of big axes of the constituted clusters. The clusters of Mo, Sr and, 9

10 presumably, 48 Ar 30 (according to ref.[11] there is a strong shell correction minimum for N=30 at β 2 ~ 0.52 ) could be the products of the CCT in this case. For the moment we have obtained only the first theoretical indication of the possible existence of the Mo - Ar - Sr CCT valley in the 252 Cf nucleus. It seems to us that the possible existence of akin valleys (modes) built on other clusters is not ruled out. To summarize, we conclude that we observe the structure linked presumably with collinear tripartition just in the raw neutron gated data presented in Fig.3. This conclusion is confirmed by the preliminary analysis of the FF charge and neutron multiplicity measured as well as by the recent calculations of the PES. That is the most promising result for the moment while processing of the data obtained is still in progress. The authors are grateful to L.V. Pashkevich for her help in preparing the English version of the manuscript. This work is supported in part by the Russian Foundation of Basic Research, grant and by the CRDF, grant Mo References: 1. H.-G. Ortlepp et al., Nucl. Instr. Meth. A 403, 65 (1998) 2. Yu. V. Pyatkov et al., Proc. Int. Conf. on Nuclear Spectroscopy Nuclear shells - 50 Years, Dubna, 1999, edited by Yu.Ts. Oganessian and R.Kalpakchieva (World Sci., Singapore, 2000), p A. A. Alexandrov et al., Heavy Ion Physics FLNR JINR scientific report , Dubna, 2001, Ed. By A.G.Popeko, E , p D. V. Kamanin et al., Neutron channel of the FOBOS spectrometer for the study of spontaneous fission, in this book 5. A. Oed et al., Nucl. Instr. Meth. A 205, 451 (1983) 6. V. V. Pashkevich, Nucl.Phys. A 169 (1971) Yu. V. Pyatkov et al., Nucl.Phys. A 624 (1997) Ю.В.Пятков, Р.А.Шехмаметьев, Яд.Физ., т. 57, (1994) Ю.В.Пятков и др., Изв. РАН, сер. физ., т. 60, (1996) Yu.V.Pyatkov et al., Nucl. Phys. A 611, (1996) H. Märten (private communication) 10