Radiochemical Determination of Neutron Capture Cross Sections of 241Am

Transcription

1 Journal of NUCLEAR SCIENCE and TECHNOLOGY, Vol. 34, No. 7, p (July 1997) Radiochemical Determination of Neutron Capture Cross Sections of 241Am Nobuo SHINOHARAt, Yuichi HATSUKAWA, Kentaro HATA and Nobuaki KOHNO Japan Atomic Energy Research Institute* (Received October 21, 1996) The thermal neutron capture cross sections and the neutron capture resonance integrals of 241Am leading to the production of the isomer 242mAm and the ground-state 242gAm were measured radiochemically by the Cd-ratio technique with neutron flux monitors of Co/Al and Au/Al alloy. Highly-purified 241Am targets were irradiated in an aluminum capsule by using JMTR. The neutron fluxes and their epithermal neutron fractions were determined by measuring g-rays of 60Co and 198Au. The yields of 242mAm and 242gAm were decided by analyzing growth and decay curves of the a-ray activity ratios 242Cm/241Am. The resultant thermal neutron capture cross sections are b and b for 242mAm and 242gAm, and the resonance integrals b and 1, b, respectively. The differences between the present results and the evaluated values by Mughabghab are 38-59%. The isomeric ratios, g/(m+g), of for thermal neutrons and for epithermal neutrons are, however, almost consistent with evaluated values. KEYWORDS: neutron capture cross sections, americium 241, thermal neutrons, resonance integrals, epithermal neutrons, resonance integral, americium 242m, americium 242g, curium 242, isomeric ratio, chemical separation, alpha spectrometry, evaluations I. INTRODUCTION Minor actinides are produced from successive neutron capture reactions starting from 238U in nuclear fuel and accumulated together with fission products in a high burnup reactor. The minor actinides cause severe problems in waste management. One of the important reactions in the buildup of the minor actinides is the neutron capture of 241Am. Both isomer-state 242mAm (48.63keV) and ground-state 242gAm are produced by the 241Am (n,-g) reaction. Isomeric 242mAm is very significant because the isomer has a very large capture cross section at thermal energies(1) and consequently can be transmuted into higher americium isotopes. The decay heat from the relatively short-lived isotope of 242Cm, which is the daughter isotope of 242gAm, gives a consequential problem in spent-fuel handling. However, there is still discrepancy among the data on the absorption cross sections of 241Am leading to the production of 242mAm and 242gAm(2). The accurate data with accuracy of 5% on the capture cross section of 241Am have been required as registered in the World Request List for Nuclear Data WRENDA 91/92(3). In this study, highly-purified 241Am targets were irradiated in Japan Material Testing Reactor (JMTR). The thermal neutron capture cross sections, the resonance integrals of 241Am and the isomeric ratios have been radiochemically measured. The obtained data are compared with evaluated and previously measured values. * Tokai-mura, Naka-gun, Ibaraki-ken Corresponding author, Tel , t Pax , E -mail: shino@popsvr.tokai.jaeri.go.jp II. EXPERIMENTAL 1. Principle for Determination of the Capture Cross Sections and the Resonance Integrals For the purpose of measuring the capture cross section for 2,200m/s neutrons and the resonance integral, this experiment was carried out by the Cd-ratio method(4): a Cd-covered target and another without Cd-cover are irradiated together with neutron flux monitors. In the convention by Westcott et al.(5), the effective cross section for Maxwellian neutrons, s, is defined by R=nv0s, (1) where R is the reaction rate per atom in the case of the irradiation without Cd-cover, n the neutron density including both thermal and epithermal neutrons, and v0=2,200m/s. flux". The cross section s is also given by The quantity nv0 plays the role of the " where s0 is the reaction cross section for 2,200m/s neutrons, and g the measure of the cross section deviation from the 1/v law in the thermal energy region. The r(t/t0)1/2 gives the fraction of epithermal neutrons in the neutron spectrum, and s0 is defined by where I'0 is the resonance integral after subtracting the 1/v component. The Gth and Gepi denote self-shielding coefficients for thermal and epithermal neutrons, respec- (2) (3) 613

2 614 N. SHINOHARA et al. tively. The values of g and Gth are almost unity in this experiment(6). From Eqs. (1) and (2), the next equation is deduced, In the case of the irradiation with Cd-cover, similar equations are derived as follows: (4) (5) From a Maxwellian distribution of the neutron field in JMTR(7), the epithermal neutron distribution can be cut off at 0.5eV (a Cd cut-off energy) as assumed in Refs. (5) and (8). The values of p1, p2, p'1, and p'2 can be obtained from the irradiation of flux monitors (two sets of Co and Au wires in this study) put near the target. From Eqs. (5) and (6), (6) and then are obtained, where the quantity s0 is the value to be measured first in this study. Because the thicknesses of the 241Am targets prepared in this work are thin enough (1.5 and 3.8mg/cm2) to reasonably assume the Gepi to be unity, the quantity I'0 can be obtained using Eq. (3). The 1/v contribution to the resonance integral for a Cd cut-off energy of 0.5eV is given by 0.45s0, so (7) (8) I0=I'0+0.45s0, (9) where I0 is the resonance integral to be also determined(8). Fig. 1 Procedure for the chemical separation of Am by an anion-exchange method 2. Radiochemical Procedures Before target preparation, the americium samples were purified chromatographically from other actinides (U, Np, Pu, and Cm), which disturb the cross section measurement of 241Am, by an anion exchange method(9). Figure 1 shows the scheme of the procedure for the ion-exchange separation. The sample was dissolved in a mixture of 0.1ml of conc. nitric acid and 1ml of ethyl alcohol. Americium was isolated by eluting from a 4mm diam. x40mm column of MCI GEL CA06Y anion exchange resin with a mixture of 0.5M nitric acid and 80% of methyl alcohol at room temperature. The ion-exchange separation procedure was carried out twice in order to perfectly isolate the Am from Cm. Chemical recovery of the americium after the ion exchange was more than 99.9%. The purity of the Am was checked by a- and g-ray measurements; no a-rays from 243Am and other actinides and no g-ray from 239Np (daughter of 243Am) were not detected. Therefore, the isotopic composition of the 241Am samples was confirmed to be 99.99%. > The 241Am was electrodeposited(10) in an active area of 5mm in diameter on aluminum disks of 12mm in diam. and 0.5mm in thickness. The amounts of 241Am deposited were determined by a -ray spectrometry; the atom number of the target without Cd-cover was ( )x1014 (36kBq) and that with Cd-cover ( )x1015 (94kBq). Weighed 0.504%-Co/Al and 0.061%-Au/Al alloy wires were used for monitoring the neutron fluxes at the target positions. Gamma-ray activities from 60Co and 198Au of neutron flux monitors were measured with a coaxial HPGe detector and a multi-channel analyzer. The detection efficiency was calibrated with a mixed standard source (Laboratoire de Merologie des Rayonnements Ionisants). Absolute activities of the monitors were determined by measuring the g rays of 1,173 and 1,333keV for 60Co and 412, 676, and 1,088keV for 198Au. Alpha-rays emitted from the 241Am targets before and after the irradiation were measured with a silicon surface barrier detector and a multi-channel analyzer. Figure 2 gives a-ray spectra of the target measured in this study. The a-ray activity ratios of 242Cm/241Am the in irradiated targets were computed from the spectra by correcting for the background counting and the tailing of the peaks by a GP method(11), which is based on the geometric progression for the far tail of the spectrum. In order to measure the activity of 242Cm growing from the 242mAm (for the determination of the amount of 242mAm, mentioned later in detail), the same chemical procedures of ion exchange and electrodeposition were JOURNAL OF NUCLEAR SCIENCE AND TECHNOLOGY

3 Radiochemical Determination of Neutron Capture Cross Sections of 241Am Irradiation Two sets of the 241Am target with the Co- and Auflux monitor wires were irradiated in JMTR for 4 days by using a hydraulic type capsule shown in Fig. 3. One set was covered with an Al container lined with Cd and another with an unlined container, in order to measure the cross sections of thermal neutron and the resonance integrals. The JMTR is a high flux reactor with a rated power of 50MW. The maximum neutron flux is about 4x1014n/cm2.s, and the typical thermal and fast neutron flux distributions and the detailed neutron spectra in the core were given in the literature(7). Fig. 2 a-ray spectra of the 241Am target again applied to the irradiated targets after 180 days elapsed from EOI (end of the irradiation). The radiometric sources prepared from the irradiated targets were measured by the a-ray spectrometry. 4. Analysis of the Decay and Growth Curves of the Activity Ratio 242Cm/241Am After the neutron capture reaction of 241Am, the 242Cm nuclide is formed through the decays of the produced nuclides, i.e. 242mAm and 242gAm. In this study, the growth and decay curves on the a-ray activity ratios of 242Cm/241Am have been precisely observed by the a-spectrometric method. By analyzing the curves in detail, the atom number ratios of 242Cm/241Am and 242mAm/241Am can be computed as follows: (1) Determination of the Atom Number Ratios of 242Cm/241Am from the Decay Curves We define the numbers of atoms as N0, N1, N2, N3 and N4 for 241Am, 242mAm, 242gAm, 242Cm and 238Pu, respectively. We assume that only the 241Am target and the decay product 242Cm from the 242gAm which had been directly produced by the reaction existed at EOI, because the contribution of the 242Cm supplied from 242mAm can be neglected as mentioned later. The amount of the 238Pu accumulated from the a-decay of 242Cm during the four-day irradiation is negligibly small (the activity of 238Pu was estimated to be <0.01% of that of 242Cm). Therefore, the activity ratio, Ames, of Fig. 3 Hydraulic type capsule irradiated in JMTR, (1), outer tube, (2) sample container lined with Cd, ) spacer, (4) unlined sample container, (5) inner tube, (6) Al-foil absorber, (7) plug, and (8) 241Am (3 target and Co/Al and Au/Al alloy wires (Unit: in mm) VOL. 34, NO. 7, JULY 1997

4 616 N. SHINOHARA et al. 242Cm/(241Am+238Pu) measured in this study, where a- rays with almost the same energy of 5.5MeV are released both from the 241Am target and from the decay product 238Pu of 242Cm, is expressed by (10) (11) (17) where N0 means the number of atoms as of the chemical separation (180 days after EOI) and N03c=0 because of the Am purification. The amount of 238Pu produced from 242Cm, N4c, is negligibly small as compared with that of 241Am (target) in this growth analysis. The values k1 and k2 are the correction factors for branching ratios of IT of 242mAm (0.995) and b-decay of 242gAm (0.827), respectively. Therefore, the growth of the activity ratio of 242Cm/241Am can be expressed by (12) (13) where N0 means the number of atoms at EOI. By least squares fitting of Eq. (13) to the decay curves measured in this study (see Fig. 5), the atom number ratios of 242Cm/241Am at EOI, N03/N00, can be determined. Since the 242gAm that was directly produced by the 241Am(n, g)242gam reaction existed at EOI and decayed completely to 242Cm at the activity measurement, the value of N03/N00 to be obtained in the decay analysis is actually (N03+N02)/N00 under this assumption. Therefore, the net atom number of N03 must be obtained by correction for the contribution of 242gAm with the calculated value of N02 as mentioned later. (2) Determination of the Atom Number Ratios of 242mAm/241Am from the Growth Curves When the atom numbers after the chemical separation of the irradiated targets are indicated as N0c, N1c, N2c, N3c and N4c for 241Am, 242mAm 242gAm, 242Cm and 238Pu, respectively, those are expressed by N0c=N00ce-l0t, (14) N1c=N01e-l1t, (15) (16) (18) By least squares fitting of this formula to the growth curve (see Fig. 5), the atom number ratio of 242mAm/241Am at the chemical separation, N01c/N00c, can be determined, where the N02c/N00c is the atom number ratio of the decay product 242gAm of 242mAm to the 241Am target at the separation and is a useless value in this study. III. RESULTS AND DISCUSSION 1. Neutron Flux The reaction rates, R and R', of 59Co and 197Au (neutron flux monitors) were obtained from the absolute activities measured by the g-ray spectrometry. The values of R/s0 and R'/s0 in Eqs. (5) and (6) were computed by using the data of s0(6)(8) listed in Table 1. The obtained values of R/s0 and R'/s0 were plotted as a function of the parameter S0Gepi as shown in Fig. 4, where the other parameters used here are also given in Table 1. By least squares fitting of Eq. (5) or (6) to the data (see Fig. 4), the neutron fluxes p1, p2, p'1 and p'2 were determined as shown in Table 2. The fitting errors are given in the table. The ratios of p'1 to p1 and p'2 to p2 are and , respectively. 2. Atom Number Ratios of 242Cm/241Am and 242mAm/241Am at EOI Figure 5 shows the decay and growth curves of the activity ratios of 242Cm/241Am measured both for the irradiated 241Am targets and for the radiometric sources JOURNAL OF NUCLEAR SCIENCE AND TECHNOLOGY

5 Radiochemical Determination of Neutron Capture Cross Sections of 241Am 617 Table 1 Nuclear data used for the neutron-flux monitoring and the cross-section determination(6)(8) s0: Cross section for 2,200m/s neutrons I0: Resonance integral s0=2/rp,i'0/s0, where is the I'0 resonance integral after subtracting the 1/v component Gepi: Self-shielding coefficient for epithermal neutrons Absorption cross section for 2,200m/s neutrons sabs: Absorption cross section for epithermal aabs': neutrons Table 2 Neutron flux determined by the flux monitors R/s0=p1+p2s0Gepi, p1=nv0, p2=nv0rrt/t0 Fig. 4 Plot of R/s0 against S0Gepi for neutron flux monitors irradiated with and without Cd-cover Statistical errors of the R/s0 values are depicted in the figure. 3. Cross Sections of 241Am(n, g)242mam and Am(n, g)242gam 241 During the irradiation, the atom number of the target 241Am decreases because of nuclear reactions such as capture and fission reactions. Moreover, not only the nuclides of 242mAm and 242gAm produced by the neutron capture reaction but also their decay product 242Cm need to be corrected for the decrease of the atom number caused by the nuclear reactions and the decays during irradiation. The procedure to calculate the time-derivative of atom number is as follows; in this calculation, the sum of the two capture cross sections, i.e. s242mam+s242gam, is supposed to be the absorption cross section of 241Am because the fission cross section of 241Am (for example, about 3b in the JENDL evaluation(1)) is smaller than the absorption cross section (about 600b). We define that Ninit0 is the initial atom number of the target 241Am before irradiation, N1(t), N2(t) and N3(t) are the atom numbers of 242mAm, 242gAm and 242Cm during irradiation at time t, respectively, sabs0, s0m and s0g, are the absorption cross section of 241Am (_??_s0m+s0g), the capture cross section of 241Am leading to 242mAm and prepared chemically from the irradiated targets. Statistical errors of the activity ratios measured in this work were %, because the high-resolution sources for -ray spectrometry could be prepared by the electrode- a position(10). The atom number ratios of 242Cm/241Am the capture cross section of 241Am leading to 242gAm, respectively, and s1, s2 and s3 are the absorption cross (actually, (242Cm+242gAm)/241Am) and 242mAm/241Am at EOI, obtained by the least squares fitting and decay sections of 242mAm, 242gAm and 242Cm, respectively. The correction, are shown in Table 3, where the errors of time-derivative of atom numbers of 242mAm and 242gAm the ratios result from the fitting. Using these values, during irradiation are approximately given by the cross sections of the reactions of 241Am(n,g)242mAm and 241Am(n, g)242gam were determined by an iteration calculation as described below. (19) and VOL. 34, NO. 7, JULY 1997

6 618 N. SHINOHARA et al. Fig. 5 Decay and growth of 242Cm in the irradiated 241Am target The open and closed circles denote the data observed with and without Cd-cover, respectively. The solid lines are obtained by the least squares fitting of Eqs. (13) and (18). Table 3 Measured atom ratios of 242Cm/241Am and 242mAm/241Am at EOI and the calculated results by the iteration method t The measured atom numbers of 241Am (target) before the irradiation are ( )x1014 and ( )x1015 for the sample without Cd-cover and for the sample with Cd-cover, respectively. respectively, where the amount of 242gAm supplied from (20) the IT decay of 242mAm is negligibly small in this irradiation because of the large difference of the half-lives of these nuclides. The time-derivative of atom number of 242Cm is expressed by (21) and (22) (23) The solutions of these differential equations are respectively (24) JOURNAL OF NUCLEAR SCIENCE AND TECHNOLOGY

7 Radiochemical Determination of Neutron Capture Cross Sections of 241Am 619 where Aj and Cj are the constants and the integral constants, i.e., (25) (26) (27) (28) (29) (30) (31) Algorithm to obtain the atom numbers of 242mAm, 242gAm and 242Cm at EOI by an iteration method is as where ti is the irradiation time (4d). (iii) Getting the temporary value of s0m by using Eq. (22). (iv) From the measured atom ratio of 242Cm/241Am, (242Cm/241An)measured, calculating the atom number N3(tI) of 242Cm at time ti by the relation of (33) (ix) Calculating sabs0 by summing s0m and s0g. (x) Iterating the procedures (ii)-(ix) until the value of sabs0 converges. (xi) Calculating N1(tI) and N2(tI) by using the converged value sabs0 which is the actual cross section in the neutron spectrum of JMTR. The same iteration calculation as above was carried out for the Cd-covered sample by temporarily setting an initial value of sabs'0 (i.e., 1, b), then N1(tI)' and N2(tI)' were also obtained by using the converged value s abs'0. The final results calculated by the iteration are given in Table 3. On the basis of the obtained values, the nuclear transmutation of 241Am by the neutron-induced reaction can be understood quantitatively; 3.5% of the atom number of the 241Am target without Cd-cover is transmuted to other nuclides after 4-day irradiation in JMTR. The transmuted amount goes to 242mAm (8.0%), 242gAm (19.4%) and 242Cm (63.9%), while the remaining percentage (8.7%) may be attributed to the nuclides follows: of 243Am and 243Cm produced by two-neutron cap- (i) Setting an initial value of sabs0 of 241Am temporarily (i.e., b). The absorption cross sections of 239Pu, etc., and fission products. The radioactivities of ture, other minor actinides as decay products like 238Pu, 242mAm, 242gAm and 242Cm evaluated in JENDL-3.2(1) 243Am and 243Cm could not be detected in this a-ray were used as constant values in this calculation. The nuclear data and their uncertainties are shown in Table 1. spectrometry. On the other hand, in the case of the irradiation of the Cd-coverd target, 0.4% of the 241Am target is lost by the irradiation and the transmuted amount (ii) From the measured atom ratio of 242mAm/241Am, (242mAm/241Am)measured, calculating the atom number changes to 242mAm (5.2%), 242gAm (21.7%) and 242Cm 1(tI) of 242mAm at ti by the relation of N (69.5%). Remaining part of the products (3.6%) may be occupied by 243Am, 243Cm, other actinides, and fission (32) products. From the values of sabs0, sabs'0 N1(tI), N1(tI)', N2(tI) and N2(tI)' obtained by the iteration calculation, the ratios of R/R' for 242mAm and 242gAm can be calculated by using the relations of (35) The obtained N3(tI) value contains the component of the decay product 242Cm of 242gAm that was directly produced by the 241Am(n, g)242gam reaction. Such 242gAm existed at EOI but decayed out thoroughly at the activity measurement in this study. (v) Obtaining the temporary value of s0g (i.e., ) by using the s0m value calculated in _??_sabs0-s0m (iii). (vi) Getting the atom number N2(tI) of 242gAm by using the s0g value and Eq. (23). (vii) Calculating the net atom number N3(tI)net of 242Cm, which does not contain the component of the decay product 242Cm from the directly produced 242gAm, by using N3(tI)net=N3(tI)-k2N2(tI) (34) (viii) Getting the temporary value of s0g by using the N3(tI)net value and Eq. (24). and (36) (37) VOL. 34, NO. 7, JULY 1997

8 620 N. SHINOHARA et al. (38) where A'1=l+s'1p' and A'2=l2+s'2p f. The ultimate cross sections of 241Am(n, g)242mam and 241Am(n, g)242gam for 2,200m/s neutrons, s0, were obtained from Eqs. (7) and (8) by using the measured values of p1, p2, p'1, p'2, the reaction rate by thermal neutrons (R) and that by epithermal neutrons (R') obtained by the iteration calculation. The resonance integrals, I0, were computed from Eqs. (3) and (9). Table 4 shows the cross sections obtained by this method. Considering the propagation of the errors both in the measured atom number ratios (0.2 to 1.1%) and the measured neutron fluxes (1.4 to 5.7%) and in the evaluated half-lives (0.1 to 1.4%) and the evaluated cross sections (4 to 30%) used for the iteration calculation, the overall uncertainties of the cross sections measured in this study are estimated to be 6.2 to 8.6%. Figure 6 shows the formation process of 242mAm, 242gAm and 242Cm in the 241Am target irradiated by ther - mal neutrons, calculated from Eqs. (22)-(24) using the cross sections measured in this study. The formation of 242mAm, 242gAm and 242Cm can be quantitatively understood from this figure. In order to quantitatively consider the further transmutation of the products starting from 241Am, it is essential to obtain the precise nuclear data of cross section, half-life and branching ratio of decay on 242mAm, 242gAm, 242Cm, 243Am and 243Cm. Fig. 6 Formation of 242mAm, 242gAm and 242Cm in the 241Am target irradiated by thermal neutrons 4. Comparison with Experimental and Evaluated Values The thermal cross sections and the resonance integrals evaluated by Mughabghab(12) and by JENDL-3.2(1) are listed in Table 4. The differences between the present results and the evaluated values by Mughabghab are 38-59%. In our method, the cross sections can be derived from the amount of 241Am deposited on the targets determined with precision of % and the relative activ- 242Cm to 241Am measured with precision ity of ratios of %. Moreover, because the highly-purified 241Am targets (>99.99% of 241Am for isotopic abundance and no other actinides contained) were used in this work, there is no influence from impurity. Table 4 Cross sections and the isomeric yield ratios of 242Am produced by the 241Am(n, g) reaction t value at meV neutrons, tt at-30-kev neutrons JOURNAL OF NUCLEAR SCIENCE AND TECHNOLOGY

9 Radiochemical Determination of Neutron Capture Cross Sections of 241Am 621 The isomeric ratios of the ground state to the total capture cross sections, g/(m+g), are also given in Table 4. The ratio of 0.90 measured at thermal neutron energy agrees excellently with the evaluated value by Mughabghab(12) and the previous data measured by Wisshak et al.(13) However, the isomeric ratio of 0.94 in the epithermal region is not consistent with the value of 0.65 by Wisshak et al. It seems that this discrepancy is due to the difference between our measurement performed with a reactor neutron spectrum and their quasimonoenergetic experiment with -30-keV neutrons. ACKNOWLEDGMENT We would like to appreciate many persons of Japan Atomic Energy Research Institute (JAERI) for their helps to make this study possible in many ways, namely, the staff of Mechanical Engineering Division for fabricating the capsule, the JMTR crew for irradiation, the Hot Lab. crew, Mr. S. Motoishi and Mr. K. Kobayashi for cutting the irradiated outer and inner capsules, and Mr. T. Sukegawa and Mr. A. Sato for sample transportation. Thanks are also due to Dr. Y. Nakagawa for comments on the nuclear data of minor actinides and Dr. M. Tanase for encouragement in this study. -REFERENCES- (1) Nuclear Data Center, Japan Atomic Energy Research Institute: "Tables of Nuclear Data", (Feb. 1996). (2) Zhou Delin, Gu Fuhua, Yu Baosheng, Zhuang Youxiang, Lui Tong, Shi Xiangjun, Yan Shiwei, Wang Cuilan, Zhang Jingshang: CNIC-00690, 97 (1992). (3) Kocherov, N., McLaughlin, P. K. (eds.): INDC(SEC)- 104, (1993). (4) Sekine, T., Hatsukawa, Y., Kobayashi, K., Harada, H., Watanabe, H., Katoh, T.: J. Nucl. Sci. Technol., 30, 1099 (1993). (5) Westcott, C. H., Walker, W. H., Alexander, T.K.: 2nd Int. Conf on the Peaceful Uses of Atomic Energy, Vol. 16, p. 70, IAEA, Geneva, (1958). (6) Matsuoka, H., Sekine, T.: JAERI-M 9552, (1981), [in Japanese]. (7) Kondo, I., Sakurai, K.: J. Nucl. Sci. Technol., 18, 461 (1981). (8) Mughabghab, S. F., Garber, D. I.: BNL 325, (3rd ed.), Vol. 1, (1973). (9) Usuda, S., Kohno, N: Sep. Sci. Technol., 23, 1119 (1988). (10) Shinohara, N., Kohno, N.: Appl. Radiat. Isot., 40, 41 (1989). (11) Umemoto, S.: Radiochim. Acta, 8, 107 (1967). (12) Mughabghab, S. F.: BNL 325, (4th ed.), Vol. 1, Part B, (1984). (13) Wisshak, K., Wickenhauser, J., Kappeler, F., Reffo, G., Fabbri, F.: Nucl. Sci. Eng., 81, 396 (1982). VOL. 34, NO. 7, JULY 1997