New Reactor Concepts and New Nuclear Data Needed to Develop Them

S. Ganesan

Reactor Physics Division Bhabha Atomic Research Centre, Trombay, Mumbai-400085 Email: [email protected]

Abstract. Developments of new reactor designs for the utilization of , such as the Advanced Heavy Water Reactor, especially demand creation of new nuclear data for all the isotopes of the thorium fuel cycle. Improved nuclear data are essential to support new initiatives such as the international project on innovative nuclear reactors and fuel cycles (INPRO), which aims to support the safe, sustainable, economic, and proliferation-resistant use of nuclear technology to meet the global energy needs of the 21st century. The detailed pursuit of development of Generation IV nuclear energy systems that offer advantages in the areas of economics, safety, reliability, and sustainability require significantly improved nuclear data. The development of Accelerator Driven Sub-critical Systems proposed by Carlo Rubbia and others require a significant amount of new nuclear data in extended energy regions and improvement of the presently available nuclear data. The quality assurance in design and safety studies in nuclear energy in the next few decades and centuries require new and improved nuclear data with high accuracy and energy resolution. The paper presents, from the perspective of the author, an overview of some of the improvements in nuclear data required for a sound scientific basis of advanced nuclear systems. Also, from the perspective of benchmarking and integral validation of nuclear data, presented briefly is the status of thorium irradiations performed in a Pressurized Heavy Water Reactor (PHWR) in India and new results of post-irradiation analyses available thus far.

INTRODUCTION THE INDIAN ATOMIC ENERGY PROGRAMME AND NUCLEAR DATA Presented in this paper are some thoughts on REQUIREMENTS nuclear data needed for new reactor systems with special emphasis on nuclear data needed for thorium The thorium fuel cycle is of great importance for utilization. Nuclear power is an inevitable option for our country, as we possess one third of world’s India. India has a national policy to implement a thorium reserves. The Indian Atomic Energy closed fuel cycle programme involving multiple Programme [1] follows a carefully planned strategy fuels. As multiple fuel cycles (e.g., U-Pu, Th-U), comprising a three-stage programme, bearing in with the option of closing the fuel cycle are mind the limited uranium resources and vast thorium envisaged, the nuclear data requirements that are reserves in India. needed to develop the new systems with high burnup are demanding and include all the range of actinides and fission products for multiple fuels. There is considerable overlap between the Indian programme THE NEW NUCLEAR DATA NEEDED with respect to thorium as a fuel and the ongoing international efforts to develop innovative, inherently As a remark, the existing data of the U-Pu cycle safe, proliferation-resistant, and long-life-cores, with also needs improvement for long burnup cores. features using thorium as in INPRO and Generation Reliable design and an operator’s manual for each IV systems. stage of the cycle based upon accurate

1411 knowledge of nuclear data will help in the safe use of thorium fuelled thermal core somewhat similar to nuclear energy by providing proper guidance on that present in AHWR. safety precautions and behaviour under all system conditions. For multiple recycled fuels, the quality of An AHWR project has been taken up [1,6] in nuclear data of higher isotopes of plutonium, minor BARC for large-scale commercial utilization of the actinides (e.g., isotopes of Am and Cm), and fission thorium fuel cycle. The nuclear design of the lattice products need to be brought up to par with that of and of the core has been finalized. The design main fissile and fertile nuclei. The demands on envisages the use of a novel 54-pin MOX cluster accurate nuclear data in the resonance region that with different enrichments of 233U and Pu in thoria affect plant safety-related feedback coefficients, such fuel pins and displacer rods with a dysprosium as Doppler and coolant void reactivity effects as a cluster and central water hole at the centre. It is function of burnup for advanced systems, are high. designed to have a negative void coefficient of reactivity. The 300-MWe AHWR is foreseen with 233 232 239 232 Remarks on Nuclear Data of Individual Isotopes the fuel: U/ Th MOX + Pu/ Th MOX. Fuel assembly characteristics are evaluated by using The rough equivalence of the isotopes in the two transport theory codes with a 69- or 172-group fuel cycles are as follows: 232Th <=> 238U (main WIMS-D library and are compared with MCNP- fertile); 233U <=> 239Pu (main fissile); 231Pa <=> based calculations. The core calculations are made 237Np; 232Pa <=> 238Np; 233Pa <=> 239Np with few-group diffusion theory methods. (Intermediate isotope in conversion); 232U <=> 238Pu; 234U <=> 240Pu; 235U <=> 241Pu; 236U <=> 242Pu; and 230Th <=> 236U. The status of nuclear data of the A Multi-Purpose Critical Facility (CF) major and minor isotopes 230Th, 232Th, 231Pa, 233Pa, 232U, 233U, and 234U in the thorium fuel cycle needs to As a general rule, all designs of innovative be brought at least to the present level of quality that reactor systems need integral validation studies. The exists for the isotopes correspondingly, as mentioned new concepts involving thorium systems require above [2,3]. Note that the data of isotopes in the specially detailed basic nuclear data measurements U-Pu cycle themselves also need further and integral validation studies, as thorium has not improvement for advanced reactor systems [4]. received the required attention in the past. As in other new concepts, the AHWR simulations have many assumptions and modeling approximations and An Advanced Heavy Water Reactor are sensitive to nuclear data uncertainties, especially Project (AHWR) because of the thorium fuel cycle. The evaluation of the lattice characteristics requires experimental The detailed pursuits of development of validation to freeze the design and obtain regulatory Generation IV and INPRO nuclear energy systems clearance before we embark on fuel fabrication. A that offer advantages in the areas of economics, multi-purpose Critical Facility (CF) has been safety, reliability, and sustainability require designed [6] and is in the advanced stage of significantly improved nuclear data. The construction. Our sensitivity calculations illustrate International Atomic Energy Agency [5] has selected that the critical height of the CF with the AHWR the AHWR concept as a case study under INPRO. representative core increases by 5 cm and 7 cm, respectively, when the “iaea.lib” and “jendl3.lib” Indian nuclear data activities generically libraries replace the “endfb6.lib” WIMS-D library. For the natural uranium core this is about 2 cm. The encompass the user-oriented approach starting from 232 data files distributed by the IAEA. The Indian replacement of multi-group data of Th alone in nuclear programme envisages extensive utilization of “jendl3.lib” by “endfb6.lib” changes the k-infinity by thorium for power production in the coming decades. 10.24 mk, “jendl3.lib” yielding a higher calculated The road map [1] for the third stage, keeping in mind value of k-infinity. These results are similar to the the current international trends in nuclear sensitivity results reported by the Kyoto team [7]. technology, is as follows [1]: Advanced Heavy These results are consistent with the differences seen Water Reactor (AHWR), which may run in parallel in the comparisons of multi-group data of the two sets. The self-shielded capture resonance integrals with our Second Stage; high-temperature reactor 232 based power packs; accelerator-driven fertile for Th are higher in “jendl3.lib” by several tens of converters; an accelerator-driven system with a fast percent as compared to 0.1% target accuracy [2,3]. reactor sub-critical core together with a mainly

1412 A Compact High Temperature Reactor FBTR (CHTR) In the 40-MWth Fast Breeder Test Reactor (FBTR) already operating at , 54 thoria A Compact High Temperature Reactor (CHTR) th 233 subassemblies (717 Kg) have been loaded in the 9 using UC2 (2.3 kg) + ThC2 (5.7 kg) is under development [8]. The fuel is in the form of compacts ring in the radial blanket after the nickel reflector located in 6-8 rings. Plans are underway to consider of TRISO-coated fuel particles. A Pb/ Pb-Bi eutectic th alloy has been proposed as the CHTR coolant. The loading an additional 100 thoria assemblies in 7 and 8th rings as well. The 233U produced in this reactor CHTR has been designed with the objective of 232 development and demonstration of technologies for will be of low content (5 ppm) of U as compared very high-temperature reactors for producing high- to several hundreds of ppm in other situations such temperature process heat for hydrogen generation. as a normal fast reactor cores, an ADSS core with thorium and thoria in PHWR. The reason for the The basic design and technology development work 232 is in progress. The CHTRs will be a compact and expected low ppm of U in FBTR is understood as portable, low-power, long-life core (more than being due to three factors influenced by nuclear data 5 years) for producing non-grid-based electricity in and physics considerations: the nickel reflector brings the neutrons below the threshold of the (n, 2n) remote regions. Achieving a very high degree of 232 231 passive safety in compact, high-temperature, liquid reaction in Th; the effective Pa (n, γ) cross metal cooled designs is needed for practically section is much lower in a fast spectrum as the eliminating the need for highly-skilled operators. capture cross section falls rapidly with increasing energy. Thirdly, the accumulation potential of 233U produced is more in saturation in a fast spectrum, The design of CHTR, which exhibits an 232 233 intermediate neutron spectrum, was strongly making the ppm content of U in U much influenced by considerations of nuclear data and smaller. associated uncertainties during its evolution. The cross sections for several new materials, such as Er, KAMINI Bi, and Ga that were considered for CHTR, show 233 large discrepancies in different cross-section A 30-kWh U reactor at Kalpakkam, known as libraries. Another interesting issue, for instance, was KAMINI, is the only reactor in the world operating 233 related to the fact that it is mandatory to have a with U fuel. It is a low-power (30 kW) research negative Doppler feedback effect in the design. The reactor designed and built by a joint venture of initial choice of pure 233U as fuel was abandoned, as BARC and the Indira Gandhi Centre for Atomic its use resulted in a calculated positive Doppler Research (IGCAR). effect. Further, the calculated Doppler effect of 233U has a large uncertainty because the nuclear data of In KAMINI, the measured Moderator resolved and unresolved resonance regions are highly Temperature Coefficient (the average MTC) is found uncertain. As the spectrum covers regions above the to be −5.62 ± 0.45 pcm/ºC. Thus the inherent safety thermal range, accurate knowledge of various of KAMINI is established. However the calculated transport and inelastic cross sections of various value of MTC reported by the Kalpakkam team using constituents, such as 233U, 232Th, Be, Er, Th, which the available WIMS1986 library is 6.2 pcm/ºC. affect the design significantly at high temperatures, Simulations of the KAMINI reactor using the latest are required. Experimental work to demonstrate available ACE files from ENDF/B-VI.8, JENDL-3.2, these systems is planned. and new IAEA WIMS-D libraries are in progress to perform further analyses.

Thorium Fuel Cycle Nuclear Data Considerations in KAMINI and FBTR USE OF THE NEW IAEA WIMS-D Reactors LIBRARIES

In this Section, we briefly discuss the use of We illustrate, by the following two striking thorium and the thorium fuel-related nuclear data examples of practical importance, the need to issues that arise in KAMINI and FBTR reactors [9]. continuously update nuclear data in reactor analyses even when the reactors are already operating. The results discussed in this section were obtained using the new WIMS-D libraries of the IAEA Co-ordinated

1413 Research Programme entitled, “Final Stage of the fuel re-thermalization effect. Recent calculations of WIMS-D Library Project,” [10] which ended in FTC of PHWR lattices, performed independently by 2001. The discussions, though limited in physics by several researchers [13], illustrate the following: The WIMS-D conventions, are illustrative to throw light 27-group wims1981 library has a crossover point, for on the use of different nuclear data sets. FTC at about 12000MWD/Te burnup; at about 9400MWD/Te with the same but 69-group library, at In the early sixties and later, corresponding to the about 6000MWD/Te for a 19-rod cluster with the 1971-81 WIMSD libraries, the effective neutron- new “iaea.lib” library and at about 4500MWD/Te for nuclear interaction cross sections of major fissile and a 37-rod cluster of PHWR with the “iaea.lib” library. fertile isotopes (235U, 238U, and 239Pu) were adjusted Actually the crossover point of the FTC is not just to fit the results of integral experiments. Not all the the issue but how negative it should be in order to adjustments in the basic data were justified when overcome positive reactivity that includes the xenon improved differential measurements were conducted, kill feedback that is positive whenever power and results of new basic data became available transient occurs. We have also observed that the several years later. That means there are considerable calculated coolant void reactivity using the new cancellations of errors in using old WIMSD data “iaea.lib” library is lower than the earlier results sets. These procedures in the sixties were applicable obtained using the 1971 library. The KAPS-1 in a restricted way to only specific systems for the overpower transient could be explained only with the limited burnup and operating parameters that were use of new WLUP libraries. studied. Today, after more than three decades since the 1971-81 libraries, many of the same experiments Analyses of Irradiation of Thorium Bundles in can be simulated using the new WLUP IAEA PHWRs libraries without any adjustment of any of the cross sections. Identical loading of thorium bundles was used in KAPP-1 and 2, KAIGA-1 and 2, and RAPS-3 and 4 The predictions of fissile materials and isotopes to attain flux flattening in the initial core. The produced in the operating research and power thorium oxide used is about 400 kg in all 35 bundles reactors (PHWRs) in India are much better with the put together in a reactor. The bundles loaded in use of new IAEA WLUP libraries as compared to the KAPP-1 and 2, KAIGA-1 and 2, and RAPS-3 and 4 use of 1971-81 WIMSD libraries used originally in have already been discharged from the core. Samples the design. were obtained from one of the irradiated ThO2 bundles and have been analyzed experimentally by For Safe Operation of Existing Reactors: alpha spectrometry for 232U and by thermal A Practical Example ionization mass spectrometry for 233U, 234U, 235U, and 236U by two different groups in BARC. The previous Recently, an incident involving power rise took analyses (see [2]) by two teams in BARC gave a place [11] in KAPS, Unit 1. A recent public release factor of 6 to 8 under-predictions in the production of [12] dated April 22, 2004 by the Atomic Energy 232U. The discrepancy was traced to be due to the fact Regulatory Board provides the details of this that the effective one-group values of cross sections incident. On March 10, 2004, KAPS-1 experienced for isotopes of the thorium fuel cycle and the use of an incident involving incapacitation of the reactor assumptions in the ORIGEN code are not applicable regulating system, leading to an unintended rise in to the irradiation of thorium in our PHWRs. reactor power from 73% Full Power (FP) to near 100% FP, with a trip occurring on Steam Generator We have successfully attempted [2] to rigorously DELTA T High Level 2 on the INES Scale. simulate the thorium experiment using the new WIMS-D libraries. New results of sensitivity of Now, we find that the fuel temperature coefficient different modeling approaches such as single-cell (FTC) calculated by the new 69-group “iaea.lib” versus super-cell model and treatment of the (n, 2n) library gives significantly different results at higher process (pseudo-fission versus explicit) to prediction burnups and explains as a preliminary observation of isotopic contents of urania were performed. Also, the unexpected power rise that occurred in the generation of integral data by gamma spectrometric KAPS-1 unit. In a PHWR, the precise crossover analysis of the irradiated thorium fuel is part of this point in burnup where the FTC becomes positive activity. depends on many parameters such as the temperature range and a 19- versus 37-rod cluster. The FTC is A direct consequence of 232U concentration in due to the combined effect of the Doppler effect and bred 233U from PHWRs is its effect on radiation

1414 shielding modification in the AHWR critical facility. Preliminary research for ADSS uses existing We are planning to use the separated 233U fuel from nuclear data developed for thermal, fast, and fusion 210 bundles irradiated in 6 PHWR units in the 9 test reactors and those generated for fundamental physics fuel clusters of AHWR. It is planned to store these in understanding as in astrophysics. The existing the critical facility complex in the middle of the strength of currently available, state-of-art nuclear concrete, which is the biological shield of the Critical databases in use for various applications is highly Facility. We had to increase the outer thickness of commendable but inadequate to meet the nuclear the concrete in our design by nearly 10% to data needs of new reactor concepts because different compensate for the additional gamma dose neutron energy spectra and materials and emanating from the fuel clusters based upon compositions are involved. For instance, the FENDL experimental results of our own PIE analyses and libraries created through successful international co- using the new basic evaluated nuclear data files. ordination by the IAEA and tailored to meet the design needs of the International Thermonuclear The simulation of thorium irradiation would not Experimental Reactor do not cover, by design, have been possible with the 1971 WIMSD libraries actinides. The FENDL experience revealed that a as the thorium chain was incomplete and several data large amount of work would also be essential in of the reaction paths contributing to the formation of order to create ENDF/B formatted evaluated nuclear 232Th such as 232Th (n, 2n) or 231Pa (n, γ) were absent data files using new experimental data. These tasks in the old data set. As a better scientific foundation, included follow-up with compilation, critical use of improved differential data is recommended as evaluation, production of new ENDF/B formatted a rule. libraries extending to higher energies, and quality assured nuclear data-processing activities to provide the designers/users of innovative systems with CREATION OF EXPERIMENTAL “ready to plug-in” processed data that are integrally INDIAN BENCHMARKS validated for use in applications. The ADSS concepts have provided a fresh look at India is actively participating in the IAEA-CRP the use of the thorium fuel cycle in a lead-bismuth [2] on “Evaluated Nuclear Data for Thorium- coolant environment. The need for precise, Uranium Fuel Cycle,” to share our information and experimental, neutron-induced, and charged particle- to benefit from the developments related to the use of induced nuclear data remains indeed very strong to thorium around the world. The two immediate tasks enable potential economies related, for instance, to on hand are the following: new fuel designs such as using thorium and higher burnup (few hundreds of GWD/Te) to be made. The 1. KAMINI experimental benchmark: experimental validation efforts in critical facilities Considerable data have already been released to can never exactly verify the simulated states of the IAEA. Further work is underway to prepare higher burnup. Improved nuclear data are therefore an international quality description of the essential for fission products and minor actinides in benchmark, with quality such as in the state-of- developing advanced reactor systems, such as art NEA Nuclear Science Committee document actinide burner systems. The interesting results of [14]. large discrepancies obtained by the author [16] in the calculated criticality properties of minor actinides, 2. Thorium Irradiation experiments and burnup 241 243 231 232 233 measurements in PHWRs: A benchmark is being such as Am, Am, Pa, U, and Pa, shed prepared on this as well. light on the inadequacy of nuclear data of these minor actinides in the fast-energy region. The nuclear data of minor actinides and fission products are also crucial in international formulation of SOME ASPECTS OF NUCLEAR DATA radioactive transport regulations. All these THAT INFLUENCE EVOLUTION OF requirements demonstrate the immediate need for the NEW REACTOR CONCEPTS research carried out in experiments. The new measurements of nuclear data reported at this The AHWR, CHTR studies [6,8] and other ND2004 International Conference help to reduce the research studies of the new concept [15] in BARC existing uncertainties in simulation studies of new use available “ready-to-plug-in,” lattice code- reactor concepts. The knowledge management and compatible, multi-group libraries. critical evaluation of basic nuclear data and

1415 associated uncertainties should be rigorously INDC(NDS)-447 (Dec. 2003), pp. 29-52; See sustained, well supported, and pursued further. electronic document at: www-nds.iaea.or.at/reports/indc-nds-447.pdf. 3. S. Ganesan, “A review of the current status of nuclear data for major and minor isotopes of thorium fuel CONCLUDING REMARKS cycle,” Paper published in the CDROM Proceedings of the International Topical Meeting on Advances in In this paper, we dealt with some aspects of Reactor Physics and Mathematics and Computation nuclear data requirements with respect to into the Next Millennium, May 7-11, 2000, American development of advanced and new reactor concepts Nuclear Society, Pittsburgh, Pennsylvania, USA. with special emphasis on using thorium in the 4. P. Finck, H. S. Khalil and M. Salvatores, “Developments in Nuclear Energy technologies and coming decades. India recognizes the need for Nuclear data needs,” in International Conference on reliable nuclear data for all evaluations for several Nuclear Data for Science and Technology 2004, AIP hundreds of isotopes/elements in all stages of the Conference Proceedings (Melville, New York, 2005). nuclear fuel cycle. We are interacting with Nuclear 5. See the IAEA website: http://www.iaea.org. Data Section of the IAEA closely in this regard. 6. R. Srivenkatesan, Umasankari Kannan, Arvind Kumar, India has committed to sponsor a regional NDS S. Ganesan and S. B. Degwekar, “Indian Advanced nuclear data Mirror Site for the Asian region at Heavy Water Reactor for Thorium Utilization and BARC. India is also working on creation of Nuclear Data Requirements and Status,” Paper experimental benchmarks for international distribu- presented at the AGM on Long Term Needs for Nuclear Data Development, 28 Nov. - 1 Dec. 2000 at tion and possible inclusion subject to peer-review IAEA, Vienna. and feedback on the quality of description of the 7. H. Unesaki, K. Kobayashi And S. Shiroya, experimental benchmarks as part of Indian “Assessment of 232Th nuclear data through analyses of participation in this ongoing IAEA-CRP [2]. thorium loaded critical experiments in thermal neutron systems Using the Kyoto University Critical assembly,” J. Nucl. Sci. Technol. 38, 370-378 (2001). 8. R. K. Sinha, “Compact High Temperature Reactor,” ACKNOWLEDGMENTS IAEA Consultancy Meeting on “Small reactors without on-site fuelling,” March 15-17, 2004. The author thanks Shri. R. K. Sinha, Dr. R. 9. See the website: http://www.igcar.ernet.in. Srivenkatesan, BARC for many useful discussions 10. See the website: www-nds.iaea.org/wimsd. on AHWR and CHTR and for their keen interest in 11. S. S. Bajaj, H.P. Rammohan and A.N. Kumar, this work. A large number of colleagues from “KAPS-1 Overpower Incident,” Private various Divisions in BARC and other units of DAE Communication. Unpublished. Nuclear Power have actively participated in the discussions. Thanks Corporation of India (July 2004). are due to Drs. P. Mohanakrishnan, C. P. Reddy, 12. See website: www.aerb.gov.in/prsrel/prsrel.asp 13. Private communications from R. Srivenkatesan, P.D. D. K. Mohopatra, IGCAR, Kalpakkam for Krishnani, R. Karthikeyan M. V. Parikh and discussions on KAMINI and FBTR. Umasankari Kannan (July 2004). Unpublished. 14. NEA, International Handbook of Evaluated Criticality Safety Benchmark Experiments, CDROM September REFERENCES 2004 Edition, NEA/NSDC/DOC(95)03. 15. V. Jagannathan, Usha Pal, R. Karthikeyan, S. Ganesan, R. P. Jain, and S. U. Kamat, “ATBR – A Thorium 1. See Websites: www.dae.gov.in and www.barc.ernet.in. Breeder Reactor Concept for An Early Induction of 2. S. Ganesan, “Integral Benchmarks With Reference To Thorium in an Enriched Uranium Reactor,” Nucl. Thorium Fuel Cycle,” IAEA Co-ordinated Research Technol. 133, 1-32 (2001). Project on “Evaluated Data for the Thorium-Uranium 16. S. Ganesan and H. Wienke, “A detailed re-assessment fuel cycle,” First Research Co-ordination Meeting, of the criticality property of pure 241Am,” J. Nucl. Sci. August 25-29, 2003, Vienna, Austria, Proceedings Technol., Suppl. 2, 951-954 (2002). edited by A. Trkov and available as INDC report:

1416