lable at ScienceDirect

Progress in Nuclear Energy 76 (2014) 183e190

Contents lists avai

Progress in Nuclear Energy

journal homepage: www.elsevier .com/locate/pnucene

Review

Thermal-hydraulic and neutronic experimental research in the TRIGAreactor of Brazil

Amir Zacarias Mesquita*, Rose Mary Gomes do Prado SouzaNuclear Technology Development Center/Brazilian Nuclear Energy Commission (CDTN/CNEN), Campus of UFMG e Pampulha, 31270-901 Belo Horizonte,MG, Brazil

a r t i c l e i n f o

Article history:Received 1 April 2014Received in revised form15 May 2014Accepted 24 May 2014Available online

Keywords:Thermal-hydraulicNeutronicTRIGA reactorTemperatureReactivity

* Corresponding author.E-mail addresses: amir@cdtn.br, amir.mesquita@h

souzarm@cdtn.br (R.M. Gomes do Prado Souza).

http://dx.doi.org/10.1016/j.pnucene.2014.05.0220149-1970/© 2014 Elsevier Ltd. All rights reserved.

a b s t r a c t

A 100 kW TRIGA Mark I research reactor has been functioning since 1960 at the Nuclear TechnologyDevelopment Center (CDTN), Belo Horizonte, Brazil. The reactor was installed for research, manpowertraining and production of radioisotopes to be used in industry, agriculture and medical treatment. TheIPR-R1 TRIGA is a pool type reactor cooled by natural circulation of light water and an open surface.TRIGA reactors, developed by General Atomics (GA), are the most widely used research reactors in theworld and they are characterized by inherent safety. The IPR-R1 was designed to operate at 30 kWsteady-state power and 100 kW power for limited periods, namely four hours a day, two days a week.The reactor could not meet the increasing demand for operation with natural circulation alone. Theforced cooling system was built in the 70s and the power was upgraded to 100 kW. Recently the coreconfiguration and instrumentation were upgraded again to 250 kW at steady state. This paper describessome results of experiments that have been carried out in the IPR-R1 reactor in recent years with the aimof evaluating the behavior of some thermal-hydraulic and neutronic operational parameters. Theresearch project meets the recommendations of the IAEA for safety, modernization and development ofstrategic plan for research reactors utilization. This work is in line with the strategic objectives of Brazil,which aims to design and construct the Brazilian Multipurpose research Reactor (RMB).

© 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Brazil has four research reactors in operation: the MB-01, a0.1 kW critical facility; the IEA-R1, a 5 MW pool type reactor; theArgonauta, a 500 W an Argonaut type reactor, and the IPR-R1, a100 kW TRIGA Mark I type reactor. They were constructed mainlyfor using in nuclear research, education and radioisotope produc-tion. Of these reactors, the IPR-R1 TRIGA is the only one that dis-plays coupled neutronic and thermal-hydraulic behavior. Thecoupling between these two areas is an innovative field of study.

The experience acquired in the project can be applied in thedevelopment of the Brazilian research Multipurpose Reactor (RMB)as well as for staff training in nuclear reactor technology. Thisresearch project meets the recommendations of the InternationalAtomic Energy Agency (IAEA), which has encouraged its membersto develop strategic plans for the use of its research reactors (IAEA,

otmail.com (A.Z. Mesquita),

2001). The IAEA also recommends the modernization of control formonitoring the operational limits of nuclear reactors (IAEA, 1998,2008).

The IPR-R1 TRIGA reactor at CDTN in Belo Horizonte is atypical TRIGA Mark I light-water reactor cooled by assisted nat-ural convection with an annular graphite reflector. It is a researchpool reactor and the core is placed at the bottom of an open tankof about 6 m height and 2 m diameter, able to assure an adequateshielding of radiation from the core. The cylindrical fuel elementsare a homogeneous mixture of zirconium hydride and uranium20% enriched in 235U with a cylinder of metallic zirconium insideand aluminum or stainless steel cladding. The moderating effectsare entrusted both to the light water coolant and to the zirco-nium hydride in the mixture. The hydride fuel possesses a veryhigh negative prompt temperature coefficient that is the mainreason for the high inherent safety behavior of the TRIGA re-actors. This temperature coefficient allows great freedom insteady state and transient operations. Fig. 1 shows two photo-graphs of the reactor room and its pool shown the core with thereactor in operation.

Fig. 1. IPR-R1 TRIGA Mark 1 reactor room and pool.

A.Z. Mesquita, R.M. Gomes do Prado Souza / Progress in Nuclear Energy 76 (2014) 183e190184

2. Material and methods

Fig. 2 shows the diagram of the pool and the core of the IPR-R1TRIGA reactor. Fig. 3 shows the forced cooling system, whichtransfers the heat generated in the reactor core to a water-to-waterheat exchanger. The secondary cooling system transfers the reactorcore heat from the heat exchanger to a cooling tower. In the dia-gram it is also shown the instrumentation distribution and theforced and natural circulation paths in the pool.

A data acquisition system was developed in order to facilitatethe data acquisition in experiments performed in the reactor(Fig. 4) (Mesquita et al., 2011a). The information of reactor oper-ation is displayed on the computer screen. The color graphicmonitors display real-time operation data in concise, accurate,and easily understood formats. Besides showing the real-timeperformance of the plant, the system stores the information in acomputer hard disk, with an accessible historical database. Someof the parameters monitored are: the control rod positions andtheir reactivity, the reactor power, the fuel and water tempera-tures, the radiation levels, the primary cooling system flow, thewater pool level and so on. About forty variables are registered bythe data acquisition system. The system responds to the IAEArecommendations on the monitoring and recording of the oper-ational variables (IAEA, 2008). A fuel element instrumented withthermocouples, shown in Fig. 5, was inserted in the IPR-R1 core toperform the thermal hydraulic and neutronic experiments (GGA,1972).

2.1. Thermal-hydraulic experimental research

The reactor power was calibrated and monitored using ther-mal techniques and after the channels of the neutron powermeasurements were adjusted. The basic safety limit for theTRIGA reactor system is the fuel temperature, both in steady-state and pulsed mode operation. Several experiments werecarried out with the measurement of the temperature inside thefuel element, in the reactor core, and at different height of thereactor pool. During these experiments the reactor was set inmany different power levels up to 250 kW. The mass flow ratethrough the core coolant hot channel was indirectly determinedfrom the heat balance across the channel. This was done usingmeasurements of the water entrance and exit temperatures, withthe forced cooling system switched off and on. The thermalconductivity of the fuel and the heat transfer coefficient from thecladding to the coolant were also evaluated experimentally. Itwas also presented a correlation for the gap conductance, be-tween the fuel and the cladding, using the instrumented fuelelement (Mesquita, 2008).

2.2. Neutronic experimental research

The knowledge of the reactor's response to specific control rodmotions is essential to the safe and efficient operation of nuclearreactors. They must have sufficient excess reactivity to compensatethe negative reactivity feedback effects such as those caused by thefuel temperature and power defects of reactivity, fuel burnup,fission poisoning production, and also to allow full power operationfor predetermined time period. To compensate for this excessreactivity, it is necessary to introduce an amount of negativereactivity into the core which one can adjust or control it at will.Several experiments have been performed in the reactor to deter-mine neutronic parameters such as: control rods worth, corereactivity excess and shutdown margin, reactivity changes inducedby a simulated void, reactivity power coefficient and power defect,temperature and isothermal reactivity coefficients, xenonpoisoning, and neutrons flux. From the results, it is possible tobalance all the determined reactivity losses with the reactivityexcess available in the reactor, considering the present and thefuture power (Souza and Mesquita, 2011a, b).

Temperature is one of the operating conditions that affect thereactivity of a reactor core. Such reactivity variation with temper-ature is the principal feedback mechanism determining theinherent stability of a nuclear reactor. The temperature coefficientof reactivity is defined as the change in reactivity due to a variationin the average temperature of each component of the core. Usingthe values of the temperatures found in the fuel and in the core, itwas presented an experimental curve of the total temperaturereactivity coefficient versus the core average temperature(Mesquita and Souza, 2010).

3. Results and discussion

3.1. Thermal-hydraulic experiments

3.1.1. Thermal power calibrationsTwo methods were tested for thermal power calibration of the

IPR-R1 reactor: the calorimetric and heat balance. The calorimetricprocedurewas donewith the reactor operating at a constant power,with the primary cooling system switched off. The rate of tem-perature rise of the water was recorded. The reactor power wascalculated as a function of the temperature-rise rate and the systemheat capacity constant. The heat balance procedure consists in thesteady-state energy balance of the primary cooling loop of thereactor. For this balance the water flow in the primary cooling loopand its inlet and outlet temperatures were measured. The heattransferred through the primary loopwas added to the heat leakagefrom the reactor pool.

Fig. 2. Diagram of the pool and core of IPR-R1 TRIGA reactor.

Fig. 3. IPR-R1 TRIGA reactor cooling system and instrumentation distribution.

A.Z. Mesquita, R.M. Gomes do Prado Souza / Progress in Nuclear Energy 76 (2014) 183e190 185

Fig. 4. User interface of the data acquisition system developed for IPR-R1 reactor.

A.Z. Mesquita, R.M. Gomes do Prado Souza / Progress in Nuclear Energy 76 (2014) 183e190186

Table 1 summarizes the results of calibrations at 100 kW and250 kW. The thermocouples were distributed as shown in Fig. 3.The calorimetric method calibration presented a large uncertainty.The main source of error was the determination of the heat contentof the system, due to a large uncertainty in the volume of the waterin the system and a lack of homogenization of the water temper-ature. The heat balance calibration in the primary loop is thestandard procedure for the calibration of the IPR-R1 reactor power(Mesquita et al., 2011b).

3.1.2. Fuel temperatureFig. 6 shows the results of fuel temperature versus reactor

thermal power. In the experiment the instrumented fuel elementwas positioned in each core ring. As it was to be expected thetemperatures increase with the power and with the distance fromthe core center.

3.1.3. Core temperatureThe experimental outlet coolant temperature for each core ring

is shown in Fig. 7 as a function of the reactor power. The aluminumprobe with thermocouple was inserted in each hole at the top gridplate, and the coolant inlet temperature obtained was about 38 �C,in all measurements.

Fig. 5. Instrumented fuel element before and after it

Fig. 8 shows the radial temperature profile along the corecoolant channels (inlet/outlet channel temperatures) at 265 kW.Theoretical results using the PANTERA code are also shown in thefigure (Veloso et al., 2006). The experimental results demon-strated how poorly the PANTERA code predicted the radialtemperature profile. One of the motives for the difference is thatthe inlet temperature that was used in the simulation was about5 �C lower than the experimental results. Another possiblereason is that this code does not take into account the effect ofcoolant cross-flow through the core channels. The use ofComputational Fluid Dynamics (CFD) codes would be moreappropriate to model the behavior of the coolant in the reactorcore channels.

3.1.4. Coolant mass flow rate and velocity in the hot channelThe mass flow through the core hot channel was determined

indirectly from the heat balance across the core using measure-ments of the water entrance and exit temperatures. The channelheating process is the result of the thermal fraction contributionsof each fuel perimeter around the channel. Results of the experi-ments indicated that the mass flow rate and the velocity increasein the hot channel when forced cooling is turned on, as shown inFig. 9.

3.1.5. Thermal conductivity and the heat transfer coefficient of thefuel elements

The IPR-R1 reactor was configured to operate until amaximum of 250 kW. The heat transfer regime, between the cladsurface and the coolant, change from single-phase convection tosubcooled nucleate boiling, and the transition point betweenthem, on the cladding surface in the central channels of the IPR-R1 core, is approximately 60 kW as shown in Fig. 10 (Mesquita,2008). As the reactor core power increases, the heat transferregime from the fuel cladding to the coolant changes from single-phase natural convection to subcooled nucleate boiling. Resultsindicated that subcooled boiling occurs at the cladding surface inthe central channels of the reactor core at power levels in excessof 265 kW.

In Fig. 11 it is presented a correlation for the heat transfer co-efficient through the gap as a function of the reactor power, eval-uated experimentally using the instrumented fuel element(Mesquita, 2008). This figure also shows three theoretical values forthe heat transfer coefficient recommended by General Atomic, theTRIGA reactor builder (GA, 1970).

was been positioned in the IPR-R1 reactor core.

Table 1Results of some reactor power calibrations.

Parameter Calorimetric method Heat balance method

Power in linear channel (reference) 100 kW 100 kW 250 kWTemperature-rise rate (DT/Dt) 4.84 �C/h e e

Average water temperature rise 34 �Ce41 �C e

Water volume rise 17.36 m3e17.86 m3 e

Average water volume 17.7 m3 17.7 m3 18.00 m3

Power dissipated 99 kW e

Thermal losses from the reactor pool 3 kW 1.4 kW 3.8 kWUncertainty ±21 kW (±21%) ±6.6 kW (±5.9%) ±19 kW (±7.2%)Average primary loop coolant flow rate e 30.09 ± 0.02 m3/h 32.73 ± 0.05 m3/hAverage primary loop inlet temperature e 33.4 ± 0.2 �C 41.7 ± 0.3 �CAverage primary loop outlet temperature e 30.2 ± 0.2 �C 34.8 ± 0.3 �CPower dissipated in the primary loop e 111 kW 261 kWStandard deviation of the readings e ±4.0 kW 3.7 kWPower dissipated in the secondary loopa e 85 kW 248 kWTotal reactor power 102 kW 112 kW 265 kW

a Not considered for thermal power calibration.

Fig. 6. Fuel temperature as function of the reactor power in all core rings.

A.Z. Mesquita, R.M. Gomes do Prado Souza / Progress in Nuclear Energy 76 (2014) 183e190 187

3.2. Neutronic experiments

3.2.1. Control rod worth and excess of reactivityIn the IPR-R1 reactor the reactivity control is done by three

control rods that can be inserted into or withdrawn from the core.

Fig. 7. Outlet coolant temperature as

The control rods are calibrated by the positive period method thatconsists of withdrawing the control rod from a known critical po-sition through a small distance and, after the transient, to measurethe doubling time (DT). DT is the time required for the power in-creases by a factor of two. The period is then obtained using the

function of the thermal power.

Fig. 8. Radial temperature profile in the core coolant channels at 265 kW thermal power.

Fig. 9. Evolution of the coolant flow rate and velocity in the hot channel of IPR-R1 core.

Fig. 11. Heat transfer coefficient through the gap as a function of the power.

A.Z. Mesquita, R.M. Gomes do Prado Souza / Progress in Nuclear Energy 76 (2014) 183e190188

doubling time that is the time required for the power increase by afactor of two. Each successive step is compensated by inserting theother control rod just enough to reestablish criticality. In this pro-cess the control rod under calibration proceeds from the mostinserted position (maintaining the reactor critical) to fully removed.The Safety and Shim rods are intercalibrated. The idea is to measure

Fig. 10. The onset of subcooled nucleate boiling in the fuel element surface of the IPR-R1 reactor.

one control rod in presence of another rod, used for compensatingthe reactivity introduced by step withdrawal of the measured rod.The reactivity measurements are performed at low power so theincrease in temperature during the experiment is negligible. Thereactivity values associated with the periods obtained are gottenfrom the graphical form of the inhour equation.

The integral fitted worth curves of the Regulating, Shim andSafety control rods as a function of their positions are showngraphically in Fig. 12 and Fig. 13. The Fig. 12 also shows the differ-ential curve of the Regulating rod. The maximum in the differential

Fig. 12. Integral and differential curves of the Regulating control rod.

Fig. 13. Integral curves of the Shim and Safety control rods.

Fig. 14. Power coefficient of reactivity and reactivity loss versus reactor power level.

A.Z. Mesquita, R.M. Gomes do Prado Souza / Progress in Nuclear Energy 76 (2014) 183e190 189

curve occurs approximately in the center, and is small near the endsbecause of the smaller flux density. The equations representing thefitted model, and the coefficients of determination R2, that confirmthe goodness of the fit, are also shown in the figures. The equationswere added to the data acquisition program. The integral controlrod worth curve is particularly important in research reactoroperation. The measured values of the Regulating, Shim and Safetycontrol rodworthwere 0.48 $, 3.21 $ and 2.84 $, respectively (Souzaand Mesquita, 2012).

Fig. 15. Xenon poisoning during power operation at 250 kW.

3.2.2. Loss of reactivity with power increase and xenon poisoningBecause of the prompt negative temperature coefficient, a sig-

nificant amount of reactivity is needed to overcome temperatureand allow the reactor to operate at high power levels. Fig. 14 showstwo curves: the power coefficient of reactivity, (aP ¼ Dr/DP), andthe associated reactivity loss to achieve a given power level versusthe reactor power. The power defect, which is the change in reac-tivity taking place between zero power and 250 kW is around 1.6 $,and the power coefficient of reactivity is about �0.66 ¢/kW (Souzaand Mesquita, 2011).

During the operation of a nuclear reactor, the fission fragmentsand their many decay products accumulate. The xenon-135 is themainly substance, because it has large cross section for thermal-neutron absorption. Fig. 15 shows the reactivity loss caused bythe 135Xe poisoning during the reactor operation at 250 kW. After5.6 h, the value of the negative reactivity introduced in the reactorwas around 27 cents.

3.2.3. Loss of reactivity due to voidThe presence of voids in a reactor has a significant effect on the

reactivity, then it is important to calculate the reactivity introducedfor a safe and controlled operation. The voids can be producedeither inherently, such as in cases of subcooled boiling, or they maybe introduced on purpose, as for material irradiation.

The reactivity changes induced by the void were measured byintroducing in the central thimble of a critical core a simulatedvoid, covering the total active length of the core (a closedaluminum tube which contained atmospheric air). Thus, thereactor was brought back to the critical state with the help of acalibrated control rod, and the reactivity change was determinedby the change in the position of the control rod. The insertion ofthe void in the central thimble (higher flux) induces the largestreactivity loss, 22.0 cents. Fig. 16 shows the experimental variationof the absolute reactivity value induced by a 31 cm3 polystyrenevoid chamber positioned along the axial direction of the centralthimble. Obviously, the effect is maximum at the core center,decreasing at the edges.

3.2.4. Temperature and isothermal reactivity coefficientThe prompt negative temperature reactivity coefficient of the

reactor is defined as ac ¼ Dr/DT. The ac negative means that anincrease in temperature will cause a decrease in the reactivity, r,and then a the decrease in the reactor power and in the tempera-ture tend to stabilize the reactor power. The experiment was per-formed by increasing the power, and, consequently, the fuel

Fig. 16. Experimental variation of the absolute reactivity value induced by the voidpositioned along the axial direction of the central thimble.

Fig. 17. Overall temperature reactivity coefficient.

A.Z. Mesquita, R.M. Gomes do Prado Souza / Progress in Nuclear Energy 76 (2014) 183e190190

temperature, by withdrawing the Regulating rod in a number ofsteps. The heat of the core, and then DT, was estimated from thepower versus time curve, and the reactivity change Dr was deter-mined from the Regulating calibration curve, considering its posi-tions. The obtained average value of the temperature reactivitycoefficient was (�1.1 ± 0.1) ¢/�C (Souza and Mesquita, 2011).

The isothermal temperature coefficient was measured byobserving the reactivity change when the core temperature israised by other means while the reactor is operating at a very low,almost zero power level. When the reactor is at zero power there isno sensible heat being released in the fuel, and the entire reactorcore can be characterized by a single temperature. The isothermalreactivity coefficient, aISO computed was �0.5 ¢/�C (Souza andMesquita, 2011). It is the excess reactivity required for the transi-tion from cold conditions to hot operating conditions. It was alsoobserved that the water temperature increases little with the in-crease of the power level, and that the rise in the coolant temper-ature contributes with a small fraction to the observed powerreactivity coefficient. Then the results demonstrated that the fueltemperature coefficient is the main contributor to the reactivitypower coefficient of the TRIGA reactor. Fig. 17 shows the experi-mental curve and equation of the total temperature reactivity co-efficient versus the core average temperature (Mesquita and Souza,2010).

3.2.5. Thermal neutron fluxThe thermal neutron flux at the central thimble and at the rotary

specimen rack were obtained by neutron activation measurements,using bare and cadmium-covered foils. Gold and cobalt foil de-tectors were irradiated in the rotary specimen rack and in thecentral thimble, respectively, with the reactor operating at 100 kWand 250 kW. Their gamma activities were measured using Gespectrometer. The average thermal neutron flux determined at therotary specimen rack was (6.6 ± 0.4) � 1011 n cm�2 s�1, and at thecentral thimblewas (4.1 ± 0.3)� 1012 n cm�2 s�1 with the reactor at100 kW, and (1.6 ± 0.1) � 1012 n cm�2 s�1 and(0.9 ± 0.1) � 1013 n cm�2 s�1, respectively, with the reactor oper-ating at 250 kW.

4. Conclusions

This paper describes the experimental research project carriedout in the IPR-R1 TRIGA reactor, in recent years, with the aim ofevaluating the behavior of some thermal-hydraulic and neutronicoperational parameters. The data acquisition and processing sys-tem and a software developed in this project allowed on linemonitoring and registration of the main reactor operational vari-ables. The experiments have given better comprehension of thereactor behavior and helped to develop numerical models,contributing to operational safety improvements and experimentalresearch.

Since the inherently safe operation of a reactor is dependent onthe reactivity control, it is essential to have information on thisparameter. The fuel elements have been designed to provide asignificant negative prompt temperature coefficient that allowssafe reactor operation. The experimental thermal hydraulic andneutronic results are within the safety limits as describe in the IPR-R1 TRIGA Safety Analysis Reporter. Many of the reactor parameterswhich determine the reactivity of reactor are dependent on thetemperature of the fuel, moderator and coolant.

Acknowledgments

This research project is supported by Nuclear TechnologyDevelopment Center (CDTN), Brazilian Nuclear Energy Commission(CNEN), Research Support Foundation of the State of Minas Gerais(FAPEMIG) and Brazilian Council for Scientific and TechnologicalDevelopment (CNPq).

References

GGA e Gulf General Atomic, 1972. “15” SST Fuel Element Assembly InstrumentedCore. Drawing Number TOS210J220, San Diego, CA.

GA e General Atomic, 1970. Safeguards Summary Report for the New York Uni-versity TRIGA Mark I Reactor. Reporter GA-9864. San Diego, CA.

IAEA e International Atomic Energy Agency, 2008. Operational Limits and Condi-tions and Operating Procedures for Research Reactors. IAEA-Safety Guide No.NS-G-4.4, Vienna.

IAEA e International Atomic Energy Agency, 2001. Strategic Planning for ResearchReactors e Guidance for Reactor Managers. IAEA-TECDOC-1212, Vienna.

IAEA e International Atomic Energy Agency, 1998. Application of Personal Com-puters to Enhance Operation and Management of Research Reactors. IAEA-TECDOC-1004.Vienna.

Mesquita, A.Z., 2008. Experimental Heat Transfer Analysis of the IPR-r1 TRIGAReactor. International Atomic Energy Agency (IAEA-CN-156). IAEA Publicatione Research Reactors: Safe Management and Effective Utilization, Vienna,pp. 1e10.

Mesquita, A.Z., Costa, A.C.L., Souza, R.M.G.P., 2011a. Modernisation of the CDTN IPR-R1 TRIGA reactor instrumentation and control. Int. J. Nucl. Energ. Sci. Technol. 6,153e165 http://dx.doi.org/10.1504/IJNEST.2011.041649.

Mesquita, A.Z., Rezende, H.C., Souza, R.M.G.P., 2011b. Thermal power calibrations ofthe IPR-R1 TRIGA reactor by the calorimetric and the heat balance methods.Prog. Nucl. Energ. 53, 1e11 http://dx.doi.org/10.1016/j.pnucene.2011.08.003.

Mesquita, A.Z., Souza, R.M.G.P., 2010. On-Line monitoring of the IPR-R1 TRIGAreactor neutronic parameters. Prog. Nucl. Energ. 52, 292e297 http://dx.doi.org/10.1016/j.pnucene.2009.07.013.

Souza, R.M.G.P., Mesquita, A.Z., 2011a. Measurements of the isothermal, power andtemperature reactivity coefficients of the IPR-R1 TRIGA reactor. Prog. Nucl.Energ. 53, 257e306 http://dx.doi.org/10.1016/j.pnucene.2011.06.010.

Souza, R.M.G.P., Mesquita, A.Z., 2012. Neutronic Tests in IPR-R1 TRIGA reactor. In:IAEA Publications e International Conference on Research Reactors: SafeManagement and Effective Utilization. S�erie: IAEA-188, vol. 1, pp. 45e51.

Souza, R.M.G.P., Mesquita, A.Z., 2011b. Reactivity balance in the IPR-R1 TRIGAreactor. Prog. Nucl. Energ. 53, 307e405 http://dx.doi.org/10.1016/j.pnucene.2011.06.011.

Veloso, M.A.F., Veloso, M.A., T�ofani, P.C., 2006. Thermal-hydraulic analysis of theIPR-R1 TRIGA reactor core. In: 3rd. World TRIGA Users Conference, 2006, BeloHorizonte.