OAK RIDGE NATIONAL LABORATORY operated by UNION CARBIDE CORPORATION NUCLEAR DIVISION T for the U.S. ATOMIC ENERGY COMMISSION ORNL- TM-2685 COPY NO. - RECEIVED BY DTIE OCT 9 196¢ DATE - August 10, 1969 RECEIVED «. INHERENT NEUTRON SOURCE IN MSRE WITH CLEAN 233U FUEL R. C, Steffy, Jr. ABSTRACT After about three years of nuclear operation, the MSRE fuel, enriched ’ 235U, was replaced with a 233y fuel mixture. In this new mixture there - are quantities of 232y, 233U, and 2%%U., Each of these, along with the 2329 decay chain, is a strong alpha emitter and interacts with fluorimne, beryllium, and lithium to produce neutrons. This neutron source is time-dependent be- cause of the buildup of 2329 daughters, and at the time of reaching critieality with the 2%%U fuel, the neutron source in the MSRE core was about 4 x 108 neutron/sec, primarily from the reactions ’Be(a,n)!'?C and !°F(a,n)?2Na. Alpha-n reactions with lithium will produce <3 X 10° neutrons/sec. Spontaneous fission will produce <102 neutromns/sec. , Keywords: Inherent neutron source, 233y fuel, (a,n) reactions, alpha particles, particle sources, 232y decay chain, fluorine, beryllium, lithium. NOTICE This document contains information of a preliminary nature and was prepared primarily for interncl use at the Oak Ridge National Laboratory. It is subject to revision or correction and therefore does not represent o final report. QETESUSESS U5 Do MOCUAISE i UBLALE i — LEGAL NOTICE — —— ————— This report was prepared as an account of Government sponsored work., Neither the United States, nor the Commission, nor any person acting on behalf of the Commission: A. Makes any warranty or representation, expressed or implied, with respect to the accuracy, completeness, or usefulness of the information contained in this report, or that the use of any information, apparetus, method, or process disclosed in this report may not infringe privately owned rights; or B. Assumes any liabilities with respect to the use of, or for damages resulting from the use of any information, apparatus, method, or process disclosed in this report, As used in the above, “‘person acting on behalf of the Commission'’ includes any employee or contractor of the Commission, or employee of such contractor, to the extent that such employee or contractor of the Commission, or employee of such contractor prepares, disseminates, or provides access to, any information pursuant to his employment or contract with the Commission, or his employment with such contractor, Introduction . . . Fuel Composition . Alpha Source . . . Neutron Source ., . "Li(o,n)1°B . » Spontaneous Fissions. CONTENTS “Be(a,n) 2 and °F(upn)>7Na. . Discussion . . . . Appendix . . . . . . - Calculation of Inherent Neutron for 23 Fuel Mixture . . . . . + v v v v v v v e e e .. 1D Source from °Be(q,n)!*C and ®F(a,n)®®Ne . . . . . . . . 12 Source from "Li(a,n)?°B . o o 0 @CO*]*Q-\]O\L"J:‘}EQ e e e e e e e e e e e 12 Source in MSRE * . . . L e - - . ° - » - lT LEGAL NOTICE Thia report was prepared as an account of Goverament sponsored work. Neither the United States, nor the Commission, neT ALy person acting on bshalf of the Commdssion: A. Makes any warranty or representation, expressed or implied, with respect to the accu- racy, completeness, or usefulneas of the information contained in this report, o¢ that the use of any information, apparatus, method, or process disclosed {n this report may not infringe privately owned rights; or B. Assumes any liabilities with respect to the use of, or for gamages resulting from the use of any information, apparatus, method, or process disclosed in this report. As used in the above, ‘‘person acting on behalf of the Commiasion’ includes any em- ployee or of the C jnsl or empioyee of such , to the extent that auch ployee or of the Ci izal or empl of such contractor prepares, disseminates, or provides access to, any information pursuant to hia employment or contract with the Commiasion, or his employment with auch contractor, ENIPEDNCTS £ T3 TR e RITmeTR | ® i 2 \ INTRODUCTION In the MSRE fuel mixtures, alpha-emitters are dispersed in large quantities of fluorine, beryllium, and lithium, All three of these ele- ments undergo alpha-n reactions to produce neutrons. Haubenreichl calcu- lated the inherent neutron source for the initial fuel loading (~ 35% U and ~ 65% ©8U), but the different concentrations of uranium isotopes for the 233 fuel loading, all of which are strong alpha emitters, required that another calculation be performed. The calculations reported herein were performed before the ©° was loaded into the reactor. This memorandum is divided into two parts. The first part presents final results and general discussions; the second part (appendix), intermediate results and details of the calculations. FUEL COMPOSITION The strength of the inherent neutron source is, of course, determined by the composition of the fuel salt. Except for the change in uranium, the major constituents of the fuel salt were essentially unchanged by the chemical processing which removed the 235y~-rich uranium and replaced it with & 223 fuel load. Thus, changes in the strength of the inherent neutron source were basically a function of the change in the alpha pro- duction rate and the associated alpha energies. Table 1 lists the approximate weight percentages of the components in the £33 fuel loading. These were estimated based on the predicted critical uranium concentration of 15.8 grams of uranium per liter of salt (~ 1 1b/ft’). It was assumed that all of the uranium in the fuel was loaded as part of the 233 fuel loading, and that the small amount present from other sources was negli- gible. The concentrations of the uranium isotopes which are present in the 277U fuel loading are listed in Table 2 with a comparable listing for the €U fuel loading. Table 1 MSRE Fuel Salt Composition _ Wt 9% in Element 233 Mixture F 684 * Li 11.0 Be 8.8 r 11.0 U 0.73 *99.99% 7Li Table 2 Uranium Isotopie Concentration in MSRE Fuels - Atom % in Atom % in Nuclide 233 Mixture Initial Loading 235y 0.022 --- =33y 91.4g --- =347 7.6 0.3 235U O-T ' 35 236y 0.05 0.3 23877 O.]_Lj. 6L . L aData from Reference 2. I “Data from Reference 1. ALPHA SOURCE Each of the uranium isotopes in the 3-fuel mixture decays by alpha emission, These all have long half-lives and emit alpha particles at essentially a constant rate for the lifetime of the MSRE. With one exception, the daughters of the uranium isotopes also have long half-lives for alpha emission and never reach high enough concentrations to emit alpha particles at a significant rate. The very important exception is 228Th, the daughter of 232U, which decays with a 1.91 year half-life by alpha emission. This is followed by an entire string of alpha emitters with very short half-lives; so short, in fact, that they are essentially in equilibrium with the 223Th, An alpha decay by ©28Th is quickly fol- lowed by a cascade of alpha emissions until the parent has finally decayed to 2°8pp, which is stable, Immediately after the 24 was purified (~ June 196L), the only alpha emissions were from the decaying uranium atoms, but as the concentration of ##8Th increased, the alpha activity obviously increased. So the alpha activity in the fuel salt is composed of a time-independent contribution from the decaying uranium atoms and a time-dependent contribution from the 272y decay chain. The time-dependent contribution is controlled by the 228Th half-life of 1.91 years and reaches 75% of the saturation value about L years after the 27U purification. The energies at which the alpha particles are emitted are also im- portant., All of the significant alpha emissions from the uranium atoms are between 4.1 and 5.3 Mev. The alphas from the lower members of the 233y decay chain are emitted with energies between 5.2 and 8.77 Mev. Since the neutron yield increases sharply with incident alpha energy, the higher~energy alphas in the 233j decay chain add weight to the importance of the chain, Tabulated information pertaining to the properties of the decay of the uranium isotopes and the 233y decay-chain are given in the appendix in addition to tables of the numbers of alphas/sec emitted by the various isotopes. NEUTRON SOURCE Calculation of the neutron source is necessary primarily to insure that, even when subcritical, there are enough neutrons available to meet detection criteria and guard against a startup accident.® From this standpoint, the graphite region in the core is the region of concern, for only in this region of the MSRE is criticality feasible. The calcu- lations presented herein are for 25 ft° of salt, the approximate volume of salt in the effective core, whereas the total volume of salt is about 75 £t2. So multiplication of the source strength in the core by ~ 3 will give the approximate source in a fuel drain tank when it contains the entire fuel loading. 7Li{a,n) 198 Haubenreich? found lithium to be an insignificant neutron producer when compared to beryllium and fluorine in the 35U fuel mixture. A cal- culation (see appendix) of the neutron yield from the most energetic alpha (8.77 Mev) in the 33U fuel showed that lithium produced less than one neutron for every 110 produced by beryllium and fluorine. Alphas starting at lower energies produce proportionately fewer neutrons from lithium because of the rafiid decrease with decreasing alpha energy in the 4 (Threshold energy for cross section of lithium for (a,n) reactions. “Li{a,n)*®B is ~ 4.3 Mev.) The total neutron source from all (a,n) re- actions with lithium is <3 x 10° n/sec. Spontaneous Fissions All of the uranium isotopes which are present in the £33 fuel loading undergo spontaneous fission, with half-lives for this process ranging from 8 x 1012 yr for 233 to 3 x 107 yr for 23F. The total rate of spon- taneous fissions in'fihe MSRE is simply the sum of the products of the inventory of each isotope and its spontaneous fission time constant. Less than 100 n/sec will be produced in the MSRE core by spontaneous fission. “Be(o, n) 120 and 19F(q, n) ©“Na High neutron production rates result from alpha reactions with beryllium and fluorine., For the alpha particles emitted from the uranium isotopes (energies between k.1 and 5.3 Mev) the neutron production rates from the beryllium and fluorine are approximately equal. The neutron yield of fluorine® increases more rapidly with increasing alpha energy than does the yield of beryllium (see Fig. 1);%,7 therefore, for the higher-energy alpha particles produced by the €3&j decay chain, most of the neutrons result from (a,n) reactions with fluorine, Figure 2 shows the total calculated neutron source for the MSRE core region and the individual production rates for the most significant alpha- emitters. The 233 decay chain obviously dominates the neutron production with #12Po being the alpha emitter (alpha energy of 8.77 Mev) which causes the most prolific neutron production, DISCUSSION When the #7%U mixture was placed in the MSRE fuel salt, the reactor already had about three years of operating history. The long runs at high powver insured that a large photoneutron source would be present for some time after shutdown of the reactor. However, if sufficient time were al- lowed to elapse, this source would weaken., More important as a reliable neutron source in the MSRE is the inherent neutron production from (q,n) reactions with the salt constituents. The (aq,n) source is essentially independent of power history and actually increases asymptolically to a limiting value. At the time the MSRE achieved criticality with 2°% as the fissile material (the 23 mixture was the equivalent of ~ 4 years 01d),® the inherent neutron source in the core region was ~ & x 10%® n/sec, about a factor of 1000 over the inherent source which was calculated for the #75U fuel mixture.? The accuracy of these calculations is dependent on the accuracy to which the fuel isotopic composition and the various yield data are known. The fuel composition is thought to be known to well within #* 5% and is ORNL—-DWG 69— 40028A 10 Be (DATA FROM REF 6) N=0152 £3-85 n — == EXTRAPOLATED NEUTRONS PER MILLION ALPHAS N’ F (DATA FROM REF 5) S N=1.02x10"*F683 0 0.5 1 2 5 10 £, ALPHA ENERGY (MeV) 10 Fig. 1. Neutron Yield from Thick Beryllium and Flucrine Targeis as a Function of Incident Alpha-Particle Energy. INHERENT NEUTRON SOURCE IN THE MSRE CORE 10 ORNL—DWG 69-100304A 10 5 TOTAL NEUTRON SOURCE 2 212p, (877 Mev) \ALPHA ENERGY 108 216py(6.77) TOTAL URANIUM 5 220Rn (6.27) 233y (4.7-4.8) 224pq (5.2~ 5.7) 232\ (5.1 -5.3) 2 2281 (5 2-5.4) (OCT 1968 - MSRE CRITICAL ; WITH 233) 10 0 { 2 3 4 5 6 7 TIME AFTER PURIFICATION OF THE 233U FUEL (years) Fig. 2. Total Inherent Neutron Source and the Neutron Source Resulting from Individual Alpha Emitters in the 233U-Fueled-MSRE Core. 11 probably not an appreciable source of error. The area of this calculation which contains the most possibility of error is the yield data for the reaction, °F(q,n)®?Na. Due to insufficient data, we had to extrapolate the known yield data (see Fig. 1), which only went to alpha energies of 5.3 Mev, to get the yield from the higher-energied alphas from the Z22U decay chain, We made the assumption that the yield continued increasing with the same functional relationship to the higher energies. While it is reasonable to expect the neutron production rate to increase with alpha energy, it is not known whether the yield will increase according to the same power function which could be used to describe its behavior for lower energied alphas. Yield data for beryllium (which is known from experiment for alphas to >6 Mev and is extrapolated to ~ 8 Mev by therexperimenters)6 is also shown in Fig. 1 and does indeed increase as a power function with increasing alpha energy. Since the data for beryllium increases uniformly to higher alpha energies, one would tend to expect the assumption about fluorine to be good, but even if the assumption were not good and the yield data for the 8.77-Mev alpha were lower by a factor of 2 than the extrapolated curve, the calculated total neutron source would be in error by only ~ 25%. We feel that it is fair to assign a probable error band of + 25% to these calculations with the expectation that the error is much smaller but with the realization that more experimental data on high-energy alpha interactions with fluorine is necessary before stronger confidence in the calculated neutron yield is warranted, 12 APPENDIX Calculation of Inherent Neutron Source in MSRE for 233 Fuel Salt Mixture Source from ZBe(a,n)13C and !°F(q,n)Z=Na For both reactions %Be(q,n)'2C and '°F(q,n)%®Na, the general equation used to determine the neutron source was N = (Ax/10°)(N__ Y(N/N__) (1) where Ax = Alpha activity (a/sec) in MSRE core (25 £t of salt) N = Neutron source (n/sec) N = Maximum number of neutrons produced if target were max . . composed completely of neutron-producing nuclei (°Be or °F as the case may be). N,,, is given in units of neutrons per million alphas. (N/Nmax)= Fractional yield of neutrons produced in fuel mixture to number that would be produced in pure target material. Each of the three terms enclosed in parentheses in this equation is found independently. Each term is discussed in the following paragraphs. For the life of the MSRE, each of the uranium isotopes present in the fuel will be a near-constant alpha source due to its long half-life (see Table 3). Each of the isotopes, except 74U, decays to a nuclide which alsc has a long decay half-life, The second, and subsequent nuclides in each of these decay chains are thus eliminated as significant alphs sources., While 273y produces a constant supply of alphas, it also produces the same number of “*8Th atoms. Thorium-228 has a 1.91 year half-life and starts a chain of short half-lived elements (see Table 4). Since the fuel mixture was about four years old when loaded into the reactor, we may assume that ®28Th and its decay chain are in equilibrium (i.e., the activity (alphas/sec) from #25Th and each member of its decay chain will be the same) . 13 Table 3 Alpha Source in MSRE Core from Uranium Isotopes Decay o Half Life Energy Emission a Source Isotope (yr) (Mev) Percentage (a/sec) 233y 7l 5.31 68 1.29 x 1012 - 5.26 31 0.59 x 1012 5,13 0.3 0.06 x 1012 23 1.62 x 10° 4,82 83 3.01 x 1012 4.78 15 0.5k x 1012 h.73 2 0.07 x 101® 234y 2,48 x 10° 4,76 Th 1.45 x 10t? %.70 23 0.45 x 101t 4.60 3 0.06 x 101% 238y 7.13 x 108 4,39 86 5.30 x 10° h,57 10 0.63 x 10° 4,18 b 0.25 x 10° 2386y 2.39 x 107 4,50 T3 9.80 x 108 b, L5 o7 3.61 x 10° 238y 4,51 x 10° 4,19 T 1.53 x 10° 4,15 23 0.46 x 10° ik Table & Information Related to Members of :®5 neutrons per million alphas, (3) This relation is expected to be good over the range of alpha energies encountered in the MSRE fuel. Apparently little work has been performed on the neutron source from (a,n) reactions with fluorine. The only neutron yield data we could lo- cate was in an article by Segre and Wiegand.® The yield data was only given for alpha energies up to 5.3 Mev., Up to this energy, the yield was increasing as a power'function of energy. Assuming that this held true up to 8.77 Mev (as was the apparent case with beryllium), the data was extrapolated, The analytic expression for the neutron production of fluorine as a function of alpha energy was found to be N ooy = 1.02 % 10™% E®-83 neutrons per million alphas. (4) Equations (3) and (4) analytically define the expressions to be used for the N term in Equation (1) The term (N/Nfiax) in the general equation converts the neutron yield from the yield in a pure medium to yield from a mixture. An expression, which agrees well with observed data,6 for calculating this term is 16 (—) - R (5) where n, is the number density of a nuclide and Si is the 'relative atomic stopping power". Subscript "p" refers to the particular nuclide for which the calculation is being performed., Number values for "S" were obtained from an article by Livingstone and Bethe® where data is given for a variety of elements as a function of incident alpha energy. Table G gives values for "S" for MSRE fuel salt, obtained by interpolation of the Livingstone- Bethe data. | Table 9 Relative Atomic Stopping Powers, Sj b 5 6 9 Li 0.55 0.55 0.55 0.55 Be 0.63 0.63 0.63 0.63 F 1.1 1.1 1.1 1.2 Zr 2.8 2.9 3.1 3.2 U 4.0 L.5 4.8 5.0 Shown in Table 10 are the calculated values for the fractional yield for beryllium and fluorine as a function of incident alpha energy. In brief, to calculate the neutron source from a particular alpha emitter, Equation (1) is used. Alpha activity (Ax) may be obtained from Table 3 or 5 or calcu- lated using Equation (2). Using Equations (3) and (L), N, . may be calcu- ax lated for both beryllium and fluorine., The appropriate (N/N, ,,) may be found in Table 10. Multiplying these quantities, one gets the source from beryllium and fluorine. Addition of these two gives the total source from a particular alpha. 17 Table 10 Fractional Yield Data for Be and F (N/N ..) ¢ Energy Mev) h > 6 9 Element Be 0.079 0.079 0.078 O.QTh F 0.695 0.692 0.688 0.64T Source from ‘Li(a,n)*°B Haubenreich® found that for lower energy alphas (<5 Mev) lithium was not a significant neutron producer in the MSRE. To be sure that it did not become significant at higher alpha energies, the neutron source from the 8.77-Mev alphas was calculated. Neutron yield from lithium may be calculated using the relation® ‘ 0 N = Aozf N, o(E) (-dx/dE) 4E, (6) 4,3 Where N = Neutron Yield (n/sec) Ao = alpha activity (alphas/sec of energy E) Eq = Energy of incident alpha (Mev) Np = Number density of lithium o(E) = Cross section for (m,n) reaction 4.3 = Threshold energy for "Li (a,n)?°B reaction. The term (-dx/dR) was obtained from an article by Harris'l in which he plots (-dx/AdE) as a function of alpha energy for a number of elements, For the MSRE fuel mixture this was found by the relation 18 (-dx/dE)mix = }4 wi(-dx/dE)i (7} i where w, corresponds to the weight fraction of component i and (-dx/dE), is the value taken from Harris for the 1°0 component, Units on (-dx/dE) are gm.cm~Z:Mev™!, Table 11 gives the values of (-dx/dE) for the components of the MSRE fuel mixture. Table 11 (-dx/dE) for MSRE Fuel Components {-dx/dE) x 10° (gm-cm”™2-Mev~1) Wy Incident Alpha Energy (Mev) Element (vt %) L 5 6 9 Li 11 1.1 1.3 1.k 1.9 Be 8.8 . 1.2 1.4 1.5 1.9 F 68.L 1.3 1.5 1.8 2.4 Zr 11 2.5 2.8 3.1 3.9 U 0.73 L. L 5.0 5.5 7.3 (ndx/dE)mix = Z wi(-dx/dE) 4 1.4 1.6 1.9 2.5 19 Lithium's microscopic cross section for (a,n) reaction* reaches a maximum of ~ 250 mb for T.l-Mev alphas, and drops exponentially with de- creasing energy. Between 7.5 and 8.8 Mev the cross section is constant at ~ 150 mb. | By approximating the cross section by straight lines and carrying out the integration, the neutron scurce from 8.77-Mev alpha reactions with lithium is found to be <1.1 x 10° n/sec. This is seen to be negli- gible when compared with the ~ 108 n/sec from fluorine and beryllium, The lower energy alphas will produce proportionately less due to the exponential decay of the cross section at lower energies., 10. 11. 20 REFERENCES P. N. Haubenreich, Inherent Neuiron Source in Clean MSRE Fuel Salt, USAEC Report ORNL-TM-611, Oak Ridge National Laboratory, August 27, 1963. ' Oak Ridge National Iaboratory, MSRP Semiann., Progr. Rept. Aug. 31, 1967, USAEC Report ORNL-4191, p. 50. J. R. Engel, P. N. Haubenreich, and B. E. Prince, MSRE Neutron Source Requirements, USAEC Report ORNL-TM-935, Oak Ridge National Laboratory, September 11, 1964, J. H. Gibbons and R. L. Macklin, Charged Particle Cross Sections, Phys. Rev., 114, p. 571, 1959. E. Segre and C. Wiegand, Thick-Target Excitation Functions for Alpha Particles, USAEC Report LA-136, Los Alamos Scientific Laboratory, September 1944 (also issued as USAEC Report MDDC-185). O. J. C. Runnals and R. R. Boucher, Neutron Yields from Actinide- Beryllium Alloys, Can. J. Phys., 3k:94g (1956). J. H. Gabbons and R. L. Macklin, Total Cross Section for ZBe(a,n), The Physical Review, 137, No. 6B. B1508 - B1509, 22 March 1965. J. M. Chandler, personal communication to P. N. Haubenreich, Oak Ridge National Laboratory, October 1967. M. S. Livingstone and H., A. Bethe, Nuclear Dynamics, Experimental, Revs. Mod. Phys., 9:272 (1937) W. N. Hess, Nuetrons from (o,n) Sources, Annals of Phys., 2: 115-133, (1959). D. R, Harris, Calculation of the Background Neutron Source in New, Uranium-Fueled Reactors, USAEC Report WAPD-TM-220, Bettis Atomic Power Laboratory, March 1960. o . - O o= 0w EFw o+ - - MomHoOHEAPa-gT Q0GOS s s sEHGyDdOHDTENLDNOQ QT . G. Affel I.. Anderson F. Baes . J. Ball F. Bauman E. Beall S. Bettis Blumberg G. Bohlmann Boyd Briggs Chandler Compere Cook Cottrell Crowley Culler Ditto Eatherly Engel Ferguson Ferris Fraas Franzreb Fry Gabbard Gallaher Grimes Grindell Guymon Harley Haubenreich . Hightower outzeel Hudson . Kasten Kedl Kerlin Kerr Kirslis - - . - gnf-az*.q:n.r!:r::uzmmg‘);owmzm*uzw:ufum_tflpw;np.zwm 21 Internal Distribution b1, 42, L3, Ly, 45, L6, L. 48, 49, 50-51. 52, 53. 5k, 55. 56. 5T 58. 60. * >z.zc4§)w>mm>c+bmraommwwz g . - . - < B W . R. Weir E D UQEURYNGYIHO >R ORNL-TM-2685 A. I. Krakoviak T. S. Kress R. B. Lindauer Tundin Lyon MacPherson . McCoy McCurdy McGlothlan . McIntosh, AEC Wash, McIlain McNeese McWherter Miller Moore Nicholson Perry Prince Ragan . Redford Richardson - - PFHEHERECOESOTEHES OQOEHEZ . W. Rosenthal . W, Savolainen nlap Scott J. Skinner N. Smith L. Smith Spiewak C. Steffy, Jr. A, Sundberg . R. Tallackson E. Thoma . Trauger . West Whatley . Whitman Q 5 s d. ® Gale Young 22 Internal Distribution (continued) 91-92, Central Research Library (CRL) 93-94. Y-12 Document Reference Section (DRS) 94-97. lLaboratory Records Department (LRD) 98. Laboratory Records Department -Record Copy (LRD-RC) 99, ORNL Patent Office. 100. Nuclear Safety Information Center (NSIC) External Distribution 101-115. Division of Technical Information Extension (DTIE) 116. lLaboratory and University Division, ORO