2014 I N T U T E Supervisor:D .S F (BMENTI)B R Ph.D.ThesisBookletI D

(1)

BLANKETS OF A FUSION REACTOR

Ph.D. Thesis Booklet

I STVÁN R OVNI

Supervisor: D

^R

. S

^ÁNDOR

F

^EHÉR

(BME NTI)

B

UDAPEST

U

NIVERSITY OF

T

ECHNOLOGY AND

E

CONOMICS

I

NSTITUTE OF

N

UCLEAR

T

ECHNIQUES

2014

(2)

Preliminaries and objectives of research 1

Goals 2

Investigation methods 3

New scientific results 4

References 6

Publications related to the PhD work 9

Other publications of the author 10

(3)

Preliminaries and objectives of research

ITER [Ayma 02, ITER 14] marks significant progress towards future power plants that use the energy produced by the fusion of deuterium and tritium. It is a large-scale scientific experiment, which is aimed at demonstrating that the artificial fusion power will be available for mankind in the near future. Because reasonably extractable tritium resources are not available in nature, tritium has to be produced by the fusion reactor itself in so-called breeding blankets, where neutron induced reactions generate tritium from Li. In case of the ITER the so-called Test Blanket Modules (TBM) [Bocc 09] are designed into some selected equatorial ports (special region of the vacuum vessel) in order to demonstrate the tritium production and handling techniques. The diagnostics developed to be built into the TBMs need to stand the extreme conditions inside the TBM, such as the 800^◦C temperature, high magnetic field, high γ- and neutron dose.

Therefore, the passive measurement techniques are preferred to be integrated. In my thesis some of the passive neutron diagnostic methods were under the scope. Thus, the following outline concerns the neutron activation techniques and the passive tritium production rate (TPR) measurement techniques.

The activation foil is a general denomination, which refers to a small piece of material containing primarily one component. In such a sample, the activity of the nuclides produced via one principal nuclear reaction is measured in order to determine the neutron flux. Sometimes this principal reaction is named as an activation foil. The value of the cross section of a nuclear reaction changes to the extent of several orders of magnitudes versus the energy. So, the measured activity of a foil provides information corresponding only to a part of the neutron spectrum, which is usually described by the concept of the response range (see the definition in Ref. [Czif 95]). Therefore, even more foils having different cross sections are irradiated, the more detailed picture can be obtained about the neutron spectrum. Nevertheless, the properties of the multiple foil activation method [11] are the best to cope with the harsh environment inside the TBM among the other neutron flux monitoring solutions.

The half-life of the product of the nuclear reaction applied to monitor the neutron flux has significant influence on the uncertainties of derived quantities of an investigated sample, such as the reaction rates, isotopic composition etc. Sima [Sima 93] described the effects of the reactor noise on neutron activation analysis (NAA) [Gira 64,Gira 65]

results by applying the Markovian type autocorrelation function of the neutron flux density. This theory is useful to investigate the processes having a time scale similar to that of the flux fluctuations, which usually take place on a short time scale. Flux variations having a longer time scale were investigated by Jacimovi´c et al. [Jaci 12], who introduced correction factors of NAA results to consider the linear decrease of the flux. Therefore, during the selection of the monitor foils this effect needs to be also taken into account.

Passive detectors dedicated to measuring another important property of the TBMs, the TPR are Li-pellets, thermoluminescence detectors (TLD) and activation detectors

(4)

with cross sections similar to the ones of the tritium producing reactions. The irradiated Li-pellets [Verz 96] are usually evaluated using the Liquid Scintillation (LSC) technique [Haye 53,Dier 73]. LSC measurement performed in ideal conditions can be always considered as a reference method because it directly measures the amount of tritium through itsβ⁻ decay. Usually LiF compound is used as a material for TLDs (see Refs. [Shar 82, Poho 11, Poho 10]). In such a material the neutron dose is recorded by the electrons trapped in luminescence centres, which are excited by the high energy charged particles originated to the⁶Li(n,T)α and⁷Li(n,n’T)α reactions [Fure 03].

Another TPR measurement technique is that, when not directly the tritium amount is measured but the activity of an activation foil having a cross section (σ(E)) with similar energy dependency to the one of the tritium producing reactions [Verz 07]. At the domain of the thermal neutrons the ³¹P(n,γ)³²P reaction can be used to monitor the dominating ⁶Li(n,T)α reaction, while in the fast energy range the ⁷Li(n,n’T)α reaction gets more contribution to the tritium production, which can be monitored by the

32S(n,p)³²P reaction.

I have developed a new passive technique for TPR measurement, which is based on the secondary charged particle activation method (SCPA) [Sher 56]. For sake of clarity the termtritonis used to distinguish the ionized, accelerated ³H particle from a tritium atom. In this method a sample containing two main components is irradiated: a target (in our case this is ⁶Li or⁷Li), which produces tritons and a so-calledindicator nuclide, which has a relatively high cross section for an incoming triton. During the neutron irradiation of such a sample, ⁶Li and⁷Li produce tritium (primer activation).

The energy of the emitted triton is high enough to cause charged particle activation reactions on the indicator nuclei (secondary activation), what results radioactive nuclei (monitor nuclei). After being irradiated, the activity of the produced monitor nuclei, which is proportional to the amount of generated tritium, can be determined usingγ- spectroscopy.

Goals

The scope of my PhD work was to develop passive measurement techniques to monitor the neutron field and the neutron induced interactions in breeding blankets of a fusion reactor. The elaboration of a new characteristic of an activation foil concerning the relation between its cross section and the neutron spectrum was necessary for aiming at selecting foils for an irradiation experiment. Another question to be answered was, that if the neutron flux varying in time is monitored by an activation foil, then how high is the influence of the half-life of the product of the nuclear reaction taking place in the foil on the uncertainties of the quantities need to be determined in an investigated sample, such as the reaction rates, mass of the components etc. Formulae had to be worked out both for estimating the extent of such an uncertainty and to choose a monitor foil or irradiation time matching to a specific experimental condition to perform a

(5)

more accurate activation analysis. For TPR measurement a new passive technique needs to be developed, which is a competitive alternative to the existing ones. Therefore, an indirect method using the SCPA had to be tested both experimentally and with simulations. The outcome of the research was requested to be a reliable, verified technique, which can be integrated into the TBMs.

Investigation methods

The equations belonging to the subject of neutron activation techniques are applied for activity calculations. The foil selection algorithm has been tested using the SULSA unfolding code [Suda 89], which is based on the generalised least squares method [Pere 77,Muir 86].

For investigating the TPR measurement potential of the SCPA method a series of irradiations have been carried out in the Training Reactor of BME NTI. Two types of samples were prepared: a mixture of powders which were compressed with a spindle press for making pellets, and an aqueous solution of Li₂CO₃ which has been sealed into quartz ampoules. The irradiations have been performed at the thermal irradiation position of the rabbit (sample transport) system, and in the vertical irradiation channel (G6/right/2). After the irradiations the samples were analysed byγ-spectroscopy using a HPGe detector having a carbon-epoxy window. The tritium amount of the liquid samples has been measured in the Laboratory of Environmental Studies of ATOMKI in Debrecen using the³H−³He ingrowth method, where the³He produced during the storage time is measured by a static noble gas mass spectrometer (VG-5400).

The measured quantities of the SCPA related experiments were also calculated using the MCNPX [Pelo 05] Monte Carlo particle transport code, which is capable of simulating the transport of both neutrons and charged particles, such as tritons. Simula- tions have been performed for both the neutron field according to the irradiation experiments in the Training Reactor, and the expected neutron field in the TBM. During the modelling the ENDF/B-VI.1 [McLa 96], the ENDF/B-VII.0 [Chad 06], the FENDL- 3.0/SLIB (Release 4) [Trko 09], the JEFF-3.1.2 [Stat 11], the TENDL-2010 [Koni 10]

and the TENDL-2013 [Koni 12] data libraries were used.

(6)

New scientific results

Thesis 1

I have defined the sensitivity range, which is such a domain of energy, where the flux can be estimated with the lowest uncertainty by considering a specific nuclear reaction and the actual measurement conditions of a chosen activation foil.

Using this new characteristic I have prepared an algorithm which can suggest a selection of foils, containing a predefined number of foils already amongst an unlimited set, for spectrum unfolding. I demonstrated the usefulness of this algorithm in practice using a spectrum unfolding technique based on the generalized least squares method [1] & [6].

Thesis 2

I have set up a theoretical model to provide an upper-bound estimate for the uncertainty of the reaction rates as integral quantities in NAA. This type of uncertainty is originated to the variation in time of neutron flux, when the flux value is determined by an activation (monitor) foil during the irradiation. Based on the elaborated formulae one can determine a domain of the possible half-lives, which has to contain the half-life of the chosen monitor foil to obtain a flux variation related uncertainty less than a required value. If the flux takes its values only within its uncertainty range determined by the direct measurement of the flux, then the uncertainty predicted by the model will be the same, which proves the self-consistency of the model [4].

Thesis 3

I have set up and experimentally proved a passive measurement technique aiming at measuring the neutron induced tritium production on lithium target, which is a new course of action in the experimental determination of the tritium breeding ratio, a characteristic of the breeding blanket of a fusion reactor. The developed procedure is based on the secondary charged particle activation(SCPA), where the tritons produced in the primary⁶Li(n,T)αreaction induce secondary reactions on indicator nuclei, and the activity of such radioactive products is measured. I prepared the samples of Li2CO3, MgO, S and LiAlO2 powders, and I designed and planned the standard procedure for handling the materials and pelletising. I irradiated samples made of both Li₂CO₃-MgO and Li₂CO₃-S powders in different mixing ratios and LiAlO2powders in the Training Reactor of BME. I observed the

26Mg(T,p)²⁸Mg, ²⁶Mg(T,n)²⁸Al, ³²S(T,n)^34mCl, ¹⁶O(T,n)¹⁸F, ²⁷Al(T,p)²⁹Al and

18O(T,α)¹⁷N reactions in the activated samples usingγ-spectroscopy and neutron detectors. By the first four reactions I investigated the effect of the indicator/⁶Li atomic ratio on the sensitivity of the determination of the tritium breeding [2] & [7].

(7)

Thesis 4

I have verified the adequacy of modelling in the case of the ²⁶Mg(T,p)²⁸Mg,

26Mg(T,n)²⁸Al,³²S(T,n)^34mCl and¹⁶O(T,n)¹⁸F reactions by the simulation of the transport phenomena in pellets using the MCNPX Monte Carlo particle transport code. The tendency of the curves fitted to the experimentally determined, normalized net peak areas, drawn in the function of indicator/⁶Li atomic ratios, corresponded with the ones originated to simulations in the case of each nuclear reaction. This finding is solid evidence that the difference between the measured and simulated values originates only from the uncertainty of the cross sections of the triton induced reactions. According to my simulations using the neutron field calculated for the volume of the HCPB-TBM, I could give an estimate concerning the expected reaction rates in the case of the operation conditions of ITER in pellets already used within the experiments [2] & [7].

Thesis 5

Based on the analysis of the simulation results of the prepared pellets, I came to a conclusion that the asymptotic peak area normalized by the pellet weight has a maximum in the function of the indicator/⁶Li atomic ratio in the case of indicators

26Mg and³²S, while the tendency of the asymptotic peak area normalized by the amount of the produced tritium is strictly increasing. In the case of¹⁸F coming from the¹⁶O(T,n)¹⁸F reaction in Li₂CO₃-MgO powder mixture, the tendency of the asymptotic peak area normalized by the amount of the produced tritium is different, as it is monotonically decreasing. I could explain this effect with my simplified model set up for describing the charged particle induced reactions.

This model points to the relation that an activity induced by a secondary particle taking part in SCPA is proportional to the product of the density of the indicator nuclei and the free mean path of the charged particle [2] & [7].

Thesis 6

I set up and executed a measurement procedure to precisely verify the calculation of the reaction rate of ⁶Li(n,T)α reaction using MCNPX. I applied the combi- nation of already known experimental techniques. During the experiment I irradiated an aqueous solution of Li₂CO₃in hermetically closed quartz ampoules in the thermal neutron field of the Training Reactor of BME. The amount of tritium produced in the ampoules is measured with the T-³He ingrowth method at the Laboratory of Environmental Studies of ATOMKI in Debrecen. I performed the simulations in the case of the ENDF/B-VI.1 , the ENDF/B-VII.0 , the FENDL- 3.0/SLIB and the JEFF-3.1.2 data libraries. Thanks to both the high accuracy measurement results and simulation outcomes with good statistics, I could determine the C/E ratios of these integral quantities in higher precision –with uncertainties below 2%– than that found in literature in the case of the afore-mentioned data libraries. The C/E ratios in turn obtained by myself are: 0.9242±0.0181, 0.9462±0.0184, 0.9201±0.0188 and 0.9224±0.0179 [3], [8] & [5].

(8)

References

[Ayma 02] R. Aymar, P. Barabaschi, and Y. Shimomura. “The ITER design”. Plasma- Phys. Control Fusion, Vol. 44, No. 5, pp. 519 – 565, 2002.

[Bocc 09] L. Boccaccini, J.-F. Salavy, O. Bede, H. Neuberger, I. Ricapito, P. Sardain, L. Sedano, and K. Splichal. “The EU TBM systems: Design and development programme”. Fusion Engineering and Design, Vol. 84, No. 2-6, pp. 333 – 337, 2009. Proceeding of the 25th Symposium on Fusion Tech- nology (SOFT-25).

[Chad 06] M. Chadwicket al. “ENDF/B-VII.0: Next Generation Evaluated Nuclear Data Library for Nuclear Science and Technology”. Nuclear Data Sheets, Vol. 107, No. 12, pp. 2931 – 3060, 2006. Evaluated Nuclear Data File ENDF/B-VII.0.

[Czif 95] S. Czifrus. “Processing of ENDF-6 format resonance region covariance data using a new algorithm”. Kerntechnik, Vol. 60, No. 4, pp. 152–156, 1995.

[Dier 73] R. Dierckx. “Direct tritium production measurement in irradiated lithium”.

Nuclear Instruments and Methods, Vol. 107, No. 2, pp. 397 – 398, 1973.

[Fure 03] C. Furetta. Handbook of Thermoluminescence. Word Scientific, 2003.

[Gira 64] F. Girardi, G. Guzzi, and J. Pauly. “Activation Analysis by Absolute Gamma Ray Counting and Direct Calculation of Weights from Nuclear Constants.”.

Analytical Chemistry, Vol. 36, No. 8, pp. 1588–1594, 1964.

[Gira 65] F. Girardi, G. Guzzi, and J. Pauly. “Reactor Neutron Activation Analysis by the Single Comparator Method.”. Analytical Chemistry, Vol. 37, No. 9, pp. 1085–1092, 1965.

[Haye 53] F. Hayes and R. Gould. “Liquid scintillation counting of tritium-labeled water and organic compounds”. Science, Vol. 117, No. , pp. 480 – 482, 1953.

[ITER 14] ITER Organization. 2014. www.iter.org.

[Jaci 12] R. Jacimovic, A. Trkov, and P. Stengar. “Error in k₀-NAA measurement due to temporal variation in the neutron flux in TRIGA Mark II reactor”.

Journal of Radioanalytical and Nuclear Chemistry, Vol. 294, No. 1, pp. 155 – 161, 2012.

(9)

[Koni 10] A. J. Koning and D. Rochman. “TENDL-2010: Talys- based Evaluated Nuclear Data Library”. Tech. Rep., Nu- clear Research and Consultancy Group (NRG), dec 2010.

ftp://ftp.nrg.eu/pub/www/talys/tendl2010/tendl2010.html.

[Koni 12] A. Koning and D. Rochman. “Modern Nuclear Data Evaluation with the TALYS Code System”. Nuclear Data Sheets, Vol. 113, No. 12, pp. 2841 – 2934, 2012. Special Issue on Nuclear Reaction Data.

[McLa 96] V. McLane. “ENDF/B-VI Summary Documentation Supplement 1”. Tech.

Rep., National Nuclear Data Center, Brookhaven National Laboratory, dec 1996. BNL-NCS-17541.

[Muir 86] D. Muir. “A Non-restrictive Derivation of the Generalized Method of Least Squares”. In: Proc. of IAEA Specialists’ Meeting on Covariance Methods and Particles in the field of the Nuclear Data, p. 117, Rome, Italy, 1986.

[Pelo 05] D. B. Pelowitzet al. MCNPX User’s Manual. Oak Ridge National Labora- tory, 2.5.0 Ed., April 2005.

[Pere 77] F. Perey. “Spectrum unfolding by the least squares method”. In: IAEA Technical Committee Meeting on the Current Status of Neutron Spectrum Unfolding, Oak Ridge, Tennessee, 1977.

[Poho 10] W. Pohorecki, P. Bilski, T. Kuc, and B. Ostachowicz. “Thermoluminescent method for the measurements of tritium production in neutronics experiments”. Radiation Measurements, Vol. 45, No. 3-6, pp. 736 – 738, 2010.

Proceedings of the 7th European Conference on Luminescent Detectors and Transformers of Ionizing Radiation (LUMDETR 2009 ).

[Poho 11] W. Pohorecki, T. Kuc, B. Ostachowicz, and P. Bilski. “Novel methods of tritium production rate measurements in HCLL TBM mock-up experiment with liquid scintillation technique”. Fusion Engineering and Design, Vol. 86, No. 9-11, pp. 2429 – 2432, 2011. Proceedings of the 26th Sympo- sium of Fusion Technology (SOFT-26).

[Shar 82] H. Sharabati, R. Hecker, and O. Joneja. “Tritium breeding measurements in a lithium aluminate blanket assembly using thermoluminescent dosime- ters”. Nuclear Instruments and Methods in Physics Research, Vol. 201, No. 2–3, pp. 445 – 449, 1982.

[Sher 56] R. Sher and J. J. Floyd. “Triton-Induced Reactions”. Phys. Rev., Vol. 102, pp. 242–242, Apr 1956.

[Sima 93] O. Sima. “Uncertainties in neutron activation analysis resulting from reactor noise”. Journal of Radioanalytical and Nuclear Chemistry, Vol. 174, No. 1, pp. 65–72, 1993.

(10)

[Stat 11] “Status of the JEFF Nuclear Data Library”. Journal of the Korean Physical Society, Vol. 59, No. 2, pp. 1057 – 1062, August 2011.

[Suda 89] S. Sudar. “A solution for the neutron spectrum unfolding problem without using input spectrum”. Report INDC(HUN)-026/L, IAEA, Vienna, Austria, 1989.

[Trko 09] A. Trkov, R. Forrest, and A. Mengoni. “Summary Report from First Re- search Coordination Meeting on Nuclear Data Libraries for Advance Sys- tems - Fusion Devices (FENDL - 3), International Atomic Energy Agency (IAEA) Vienna, Austria 2 - 5 December 2008”. Report INDC(NDS)-0547, International Atomic Energy Agency, Vienna, Austria, March 2009.

[Verz 07] Y. Verzilov, S. Sato, K. Ochiai, M. Wada, A. Klix, and T. Nishitani. “The integral experiment on beryllium with D-T neutrons for verification of tritium breeding”. Fusion Engineering and Design, Vol. 82, No. 1, pp. 1 – 9, 2007.

[Verz 96] Y. Verzilov, F. Maekawa, and Y. Oyama. “A Novel Method for Solving Lithium Carbonate Pellet by Binary-Acid for Tritium Production Rate Mea- surement by Liquid Scintillation Counting Technique”. Journal of Nuclear Science and Technology, Vol. 33, No. 5, pp. 390–395, 1996.

(11)

Publications related to the PhD work

Journal papers

[1] I. Rovni, M. Szieberth, S. Fehér, and A. Klix, “A proposed method for foil set qualification for multiple foil activation measurements in the TBMs,”Fusion Engi- neering and Design, vol. 86, no. 9–11, pp. 2330 – 2333, 2011. Proceedings of the 26th Symposium on Fusion Technology (SOFT-26).

[2] I. Rovni, M. Szieberth, and S. Fehér, “Secondary charged particle activation method for measuring the tritium production rate in the breeding blankets of a fusion reactor,” Nuclear Instruments and Methods in Physics Research Section A: Accelera- tors, Spectrometers, Detectors and Associated Equipment, vol. 690, pp. 85 – 95, 2012.

[3] I. Rovni, M. Szieberth, L. Palcsu, Z. Major, and S. Fehér, “High accuracy tritium measurement for the verification of the tritium production rate calculations with MCNPX,” Nuclear Instruments and Methods in Physics Research Section A: Ac- celerators, Spectrometers, Detectors and Associated Equipment, vol. 714, pp. 141 – 146, 2013.

[4] I. Rovni, “Flux variation related uncertainty in neutron activation analysis,”Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrome- ters, Detectors and Associated Equipment, vol. 729, pp. 360 – 364, 2013.

Conferences

[5] I. Rovni, M. Szieberth, S. Fehér, A. Klix, D. Gehre, and L. Palcsu, “Experimental Tests of the Secondary Charged Particle Activation Method for Tritium Production Rate Determination in ITER TBMs,” in 27th Symposium on Fusion Technology, SOFT2012, (Liége, Belgium), 2012.

Internal reports

[6] I. Rovni, M. Szieberth, and S. Fehér, “Measurement Techniques Development for Breeder Blankets I.,” Tech. Rep. BME-NTI-547/2010, BME, Institute of Nuclear Techniques, Budapest, December 2010.

[7] I. Rovni, M. Szieberth, and S. Fehér, “Measurement Techniques Development for Breeder Blankets II.,” Tech. Rep. BME-NTI-579/2011, BME, Institute of Nuclear Techniques, Budapest, December 2011.

(12)

[8] I. Rovni, G. Kleizer, M. Szieberth, and S. Fehér, “Measurement Techniques Development for Breeder Blankets III.,” Tech. Rep. BME-NTI-605/2012, BME, Institute of Nuclear Techniques, Budapest, December 2012.

Other publications of the author

[9] I. Rovni, M. Szieberth, and S. Fehér, “Investigation of direct transmutation of ac- tinides by spallation neutrons,”Proceedings of the International Conference on the Physics of Reactors 2010, PHYSOR 2010, vol. 3, pp. 2016–2027, 2010.

[10] A. Klix, A. Domula, U. Fischer, D. Gehre, P. Pereslavtsev, and I. Rovni, “Test facility for a neutron flux spectrometer system based on the foil activation technique for neutronics experiments with the ITER TBM,”Fusion Engineering and Design, vol. 86, no. 9–11, pp. 2322 – 2325, 2011. Proceedings of the 26th Symposium of Fusion Technology (SOFT-26).

[11] A. Klix, A. Domula, U. Fischer, D. Gehre, P. Pereslavtsev, and I. Rovni, “Neutron- ics diagnostics for European ITER TBMs: Activation foil spectrometer for short measurement cycles,”Fusion Engineering and Design, vol. 87, no. 7–8, pp. 1301 – 1306, 2012. Tenth International Symposium on Fusion Nuclear Technology (ISFNT-10).

[12] I. Rovni and M. Szieberth, “WP12-DTM04-T04: Nuclear analysis of the Tri- tium breeding ratio in DEMO1 (2CXCYY v1.2),” Tech. Rep. EFDA_D_2CXCYY v1.2, Budapest University of Technology and Economics, Institute of Nuclear Tech- niques, 1111 Budapest, M˝uegyetem rakpart 9., December 2012.

[13] F. Ogando, J. Catalán, J. Sanz, P. Sauvan, I. Rovni, and M. Szieberth, “WP12- DTM04-T08: DEMO1 blanket activation calculation (2LNNCQ v1.3),” Tech. Rep.

EFDA_D_2LNNCQ v1.3, UNED/CIEMAT(Spain) and BME-NTI(Hungary), De- cember 2012.

[14] U. Fischer, C. Bachmann, B. Bienkowska, J. Catalan, K. Drozdowicz, D. Dworak, D. Leichtle, I. Lengar, J.-C. Jaboulay, L. Lu, F. Moro, F. Mota, J. Sanz, M. Szieberth, I. Palermo, R. Pampin, M. Porton, P. Pereslavtsev, F. Ogando, I. Rovni, G. Tracz, R. Villari, and S. Zheng, “Neutronic analyses and tools development efforts in the european DEMO programme,” Fusion Engineering and Design, 2014. In Press, Corrected Proof.