Enrichment Fuel - Idaho National Laboratory

INL/CON-11-23822 

PREPRINT 

Evaluation of Core 

Physics Analysis 

Methods for Conversion 

of the INL Advanced Test 

Reactor to Low- 

Enrichment Fuel 

PHYSOR 2012 

Mark DeHart 

Gray S. Chang 

April 2012 

This is a preprint of a paper intended for publication in a journal or 

proceedings. Since changes may be made before publication, this 

preprint should not be cited or reproduced without permission of the 

author. This document was prepared as an account of work 

sponsored by an agency of the United States Government. Neither 

the United States Government nor any agency thereof, or any of 

their employees, makes any warranty, expressed or implied, or 

assumes any legal liability or responsibility for any third party’s use, 

or the results of such use, of any information, apparatus, product or 

process disclosed in this report, or represents that its use by such 

third party would not infringe privately owned rights. The views 

expressed in this paper are not necessarily those of the United 

States Government or the sponsoring agency.

PHYSOR 2012 – Advances in Reactor Physics – Linking Research, Industry, and Education 

Knoxville, Tennessee, USA, April 15-20, 2012, on CD-ROM, American Nuclear Society, LaGrange Park, IL (2012) 

EVALUATION OF CORE PHYSICS ANALYSIS METHODS FOR 

CONVERSION OF THE INL ADVANCED TEST REACTOR TO LOW- 

ENRICHMENT FUEL 

Mark D. DeHart and Gray S. Chang 

Idaho National Laboratory 

2525 Fremont Street 

Idaho Falls, ID 83415-3870 

Mark.DeHart@inl.gov; Gray.Chang@inl.gov 

ABSTRACT 

Computational neutronics studies to support the possible conversion of the ATR to LEU are 

underway. Simultaneously, INL is engaged in a physics methods upgrade project to put into place 

modern computational neutronics tools for future support of ATR fuel cycle and experiment 

analysis. A number of experimental measurements have been performed in the ATRC in support of 

the methods upgrade project, and are being used to validate the new core physics methods. The 

current computational neutronics work is focused on performance of scoping calculations for the 

ATR core loaded with a candidate LEU fuel design. This will serve as independent confirmation 

of analyses that have been performed previously, and will evaluate some of the new computational 

methods for analysis of a candidate LEU fuel for ATR. 

Key Words: ATR, RERTR, LEU, validation, conversion. 

1. INTRODUCTION 

The Advanced Test Reactor (ATR) is a high neutron flux research facility that is operated at at 

the Idaho National Laboratory (INL) in the United States. Fueled with high-enriched uranium 

(HEU), the ATR has a maximum core power rating of 250 MW th and can produce a fast neutron 

flux of 5 × 10 14 n/cm 2 -s or a thermal neutron flux of approximately 10 15 n/cm 2 -s in different 

experiment irradiation positions. The ATR uses 40 fuel elements, each comprised of 19 plates of 

fuel curved into a 45° arc and arranged in a serpentine pattern to create nine flux trap positions 

(see Figure 1). 

A number of activities related to conversion of the ATR to a Low-Enriched Uranium (LEU) fuel 

design are underway at INL under the auspices of the Reduced Enrichment Research Test and 

Training Reactor (RERTR) Program. The primary focus to date has been on computational and 

experimental materials performance studies for uranium-molybdenum composites. A fuel 

element design based on a U-10Mo composite has been proposed with a 0.03302 cm (13 mil) 

fuel plate thickness encased in 0.00254 cm (1 mil) zirconium centered in 0.127 cm (50 mil) Al 

6061 plates. This fuel is the basis for the Full Element (FE) fuel assembly test planned for initial 

neutronics evaluation in the Advanced Test Reactor Critical Facility (ATRC), followed by 

insertion and irradiation in the ATR.


Experimental work has been supplemented by a number of computational RERTR fuel studies to 

determine potential performance of different fuel designs covering a broad range of fuel 

characteristics. Computational neutronics studies to support the possible conversion of the ATR 

to LEU are also underway. 

Figure 1. ATR Core Configuration 

Simultaneously, INL is engaged in a physics methods upgrade project to put into place modern 

computational neutronics tools for future support of ATR fuel cycle and experiment analysis. 

The computational methods selected include Helios [1], Attila [2], and portions of the SCALE 

code system [3,4] for reactor physics calculations. MCNP [5], long used for detailed ATR 

analysis, will also be employed in ATR analysis for independent confirmation of computational 

models. A number of experimental measurements have been performed in the ATRC in support 

of the methods upgrade project, and are being used to validate the new core physics methods. 

The current computational neutronics work is focused on performance of scoping calculations 

for the ATR core loaded with a candidate LEU fuel design, using the tools identified above. In 

the absence of a final fuel element design, the emphasis has been placed on development of 

candidate designs informed by ATR operational requirements, and on evaluation of the reactor 

physics performance of such cores. Ultimately, the goal of this work is to complete a 

preliminary assessment of the capabilities of new core analysis methods for LEU fuel. Note that 

a complete assessment will ultimately require measured in-core data for code validation. 

Initially, it is planned to obtain some early neutronics measurements for LEU fuel in the ATR via 

coordination with the FE experiment mentioned previously. 

A number of candidate fuel designs have been analyzed; these results have demonstrated 

essentially that a uniform fuel loading in all 19 fuel plates of an element is not feasible. The 

amount of poison required to reduce innermost and outermost plate powers to an acceptable level 

would result in an excessive net reactivity penalty. Hence, a reduced fuel loading would be 

2012 Advances in Reactor Physics – Linking Research, Industry, and Education (PHYSOR 2012), 

Knoxville, Tennessee, USA April 15-20, 2012 

2/16

Evaluation of Core Physics Analysis Methods for Conversion of the INL Advanced Test Reactor to Low-Enrichment Fuel 

required in innermost and outmost fuel plates, either by thinner fuel regions or by reduced 

enrichment, in tandem with burnable poison loading. Based on this work, an optimized LEU 

monolithic foil-type fuel with an integral-cladding burnable absorber (ICBA) has been found to 

meet performance requirements; this design utilizes reduced-thickness fuel in the four 4 

innermost and outermost fuel plates with borated carbon foils in the cladding of the 2 innermost 

and 2 outermost plates, as illustrated in Figure 2. 

Figure 2. Monolithic foil-type fuel element with ICBA poisons. 



3/16


The initial fuel design was performed primarily using MCNP [6,7]. This paper summarizes early 

work performed to assess the reactor physics tools being adapted at INL for next-generation core 

physics analysis. In the current work, calculations were performed primarily using the KENO- 

VI and NEWT modules of the SCALE 6.0 system. Analyses were performed to assess reactor 

performance with LEU fuel (based on the proposed ICBA design) relative to the current HEU 

fuel design. 

2. MODELING APPROACH 

To provide a direct comparison of the behavior of the two fuel types, a single ATR configuration 

was selected, with only fuel element designs changed. The critical configuration from the 1994 

core-internals change-out (94-CIC) provides a well-documented benchmark from which to start 

[3]. Both KENO-VI and NEWT models of the 94-CIC configuration were developed at INL in 

separate work and were validated against the critical benchmark. For this configuration, KENO- 

VI calculated k eff = 1.00393 +/- 0.00041. The NEWT model was a 2-D representation of the core 

at the axial mid-plane of the fuel region and hence over-predicted the reactivity of the core at 

1.0317 with a convergence criterion of ε=10 -4 ; however, with axial buckling applied for a 48” 

core NEWT calculated k eff = 0.991114. Power distributions within a fuel element and for the 40 

individual elements have been shown to be in good agreement for KENO-VI, NEWT, MCNP, 

and HELIOS, and for measured data for the 40 elements. Figure 3 provides a rendered image of 

the KENO-VI model for the ATR 94-CIC specification. 

Figure 3. Cut-away view of the lower half of the ATR core from KENO-VI model. 



4/16


For the ICBA fuel design, the fuel element model used within the 94-CIC core design was 

replaced with a fuel element based on ICBA fuel dimensions. The outer dimensions of all plates 

within the element were unchanged; only the inner and outer radius of each fuel region was 

altered, and four burnable poison regions were added. Table I provides the dimensions of the 

fuel and borated carbon regions in the ICBA model; Tables II and III list the isotopic 

compositions of the fuel and poison materials, respectively. Specifications for the HEU fuel and 

for the balance of the core are available in Ref. 8. 

Table I. Dimensions for Fuel and Burnable Poison Regions in ICBA Fuel 

Plate No./Type 

Thickness 

cm (mil) 

Inner Radius 

cm 

1/Fuel 0.02032 (8) 7.749340 

1/B 4 C 0.01270 (5) 7.790234 


2/B 4 C 0.01270 (5) 8.153234 

















18/B 4 C 0.01270 (5) 13.355234 


19/B 4 C 0.01270 (5) 13.743734 



5/16


Table II. Isotopic Composition of 19.7% Enriched U-10Mo Fuel 

Number Density 

(atoms/b-cm) 

92 Mo 1.57423 × 10 -3 

94 Mo 9.81240 × 10 -4 

95 Mo 1.68879 × 10 -3 

96 Mo 1.76941 × 10 -3 

97 Mo 1.01306 × 10 -3 

98 Mo 2.55971 × 10 -3 

100 Mo 1.02155 × 10 -3 

235 U 7.73240 × 10 -3 

238 U 3.10260 × 10 -2 

Table III. Isotopic Composition of Poison Plates 

Number Density 

(atoms/b-cm) 

Plate 1 Plate 2 Plate 18 Plate 19 

10 B 5.07120 × 10 -4 1.22370 × 10 -3 4.89240 × 10 -4 1.35020 × 10 -3 

11 B 2.02830 × 10 -3 4.89540 × 10 -3 1.95720 × 10 -3 5.40130 × 10 -3 

C 6.00000 × 10 -2 6.00000 × 10 -2 6.00000 × 10 -2 6.00000 × 10 -2 

All calculations described in this paper were performed using the SCALE v7-238 238-energygroup 

ENDF-VII library. The csas6 sequence of the CSAS6 module was used for KENO-VI 

criticality calculations; the t-newt sequence of TRITON was used for corresponding NEWT 

calculations. All sequences used CENTRM and PMC for cross section processing, based on a 1D 

representation of a fuel element. To capture the spectral effects of the ATR core within the 1D 

limitations of CENTRM, fuel element cross sections were computed based on an approximation 

of the core created by assuming a southeast flux trap configuration, surrounded by a 360° ring of 

fuel plates, enclosed in a beryllium moderator/reflector region. Compositions of SS-348, Al 

6061-T6 alloy, water, and beryllium were based on 94-CIC isotopic specifications, all at a 

uniform temperature of 310.9K. 

3. COMPARISON OF FUEL ELEMENT DESIGN PERFORMANCE 

Given working KENO-VI models for the existing HEU core and for a modified core based on 

the ICBA fuel design, calculations were performed to determine similarities and differences in 

core behavior. The following subsections discuss each of the various studies. 



6/16


3.1. Initial Eigenvalue 

Based on the critical configuration of Ref. 2, the HEU model resulted in a k eff value of 1.00393 

+/- 0.00041 with the SCALE v7-238 library. To obtain critical, this configuration had two neck 

shims fully withdrawn with remaining shims fully inserted (neck shims are sets of control 

elements located in each of the four interior neck regions between lobes). In addition, outer shim 

control cylinders (OSCCs) were uniformly rotated to 51.8°. For the ICBA fuel loaded into this 

core model, KENO-VI calculated k eff = 0.97250 +/- 0.00042. Thus, the ICBA design as 

currently specified is slightly less reactive than the HEU fuel element, with a reactivity 

ρ=3.13%, at beginning of life. 

3.2. Control Cylinder Worth 

The next set of calculations was performed to determine the worth of the OSCCs as a function of 

drum rotation. Eigenvalue calculations were performed beginning with a drum rotation of 0°, 

repeated in 10° increments to 160°. Figure 4 illustrates the calculated k eff values as a function of 

drum rotation; the curve generated for the ICBA fuel appears to be slightly flattened relative to 

the HEU fuel curve. Comparing Δk for 0 to 160° of rotation it is observed that the total drum 

worth is reduced to about 85% of its HEU reactivity for the LEU ICBA design. This behavior is 

not unexpected; increased resonance absorption (resulting from the substantially increased 238 U 

inventory in the LEU fuel) hardens the spectrum by reducing the thermal neutron flux. With 

fewer thermal neutrons, the amount of thermal capture in hafnium control elements will also be 

reduced. 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fig. 4. Neutron multiplication factor as a function of OSCC rotation for two fuel types. 

It was asserted above that there is spectral hardening in the fuel; this is best demonstrated by 

examination of fuel fluxes. Figure 5 is a plot of the neutron flux within plate 10 (the radially 



7/16


central plate) from element position 1. Relative to the spectrum in the HEU fuel, there is a clear 

reduction of the neutron population in the ICBA design beginning around 1keV and moving 

down into the thermal energy range; there is also somewhat more resonance structure apparent in 

the resonance energy range between roughly 500eV to 1 eV. 

Because the control drums are in close proximity to the fuel with little hydrogen moderation, the 

reduced thermal flux is still important at the drum locations. However, as will be shown later, 

flux trap positions are not as strongly influenced by the spectral hardening, as water in the 

vicinity of the targets serves to restore much of the thermal neutron flux. 

 

 

 

 

 

 

 

 

 

 

 

Fig. 5. Neutron flux spectrum within plate 10 of element 1 

from HEU and ICBA core models. 

3.3. Power Distributions 

Using the initial model with a 51.8° OSCC rotation, analyses were performed to estimate 

element and plate powers for the 40 fuel elements. Calculations were initially performed with 

KENO-VI, but a bug was uncovered in KENO-VI resulting from the use of quadratic elements to 

define fuel plate boundaries in the ICBA fuel model. This bug would be uncovered when 

running a large number of histories to obtain converged spatial fluxes and would cause the code 

to halt. An alternate modeling approach was used to represent the ICBA fuel such that the model 

could be run. However, concern introduced in the use of KENO-VI and the modified model led 

to the preparation of a 2-D NEWT model of the 94-CIC core with both fuel types to provide an 

independent check on KENO results. 

First, plate power distributions were averaged over all 40 fuel elements for each plate position 

(powers are calculated by the TRITON module in SCALE based on (n,f) and (n,γ) reaction rates 



8/16


within each depletion material). Total power was calculated for each plate position, and 

normalized to an average power of 1.0. Figure 6 shows the power distribution in the two fuel 

element designs as calculated by KENO-VI. Figure 7 illustrates the corresponding calculation in 

the NEWT calculation. With the exception of very small power differences in plates 1 and plate 

18, the results are essentially identical. These results demonstrate that the proposed ICBA fuel 

design very closely approximates the power distribution for the current HEU fuel – this feature is 

key in being able to use an approach similar to the current approach for core safety analysis. 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fig. 6. KENO-VI prediction of average plate powers for two fuel types in 94-CIC core. 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fig. 7. NEWT prediction of average plate powers for two fuel types in 94-CIC core. 

Next, the results of the same calculations were used to calculate the average power in each fuel 

element position. Average element powers were calculated by summing the 19 plate powers in 



9/16


each element. Element powers were then normalized such that the average over all elements was 

1.0. Figure 8 plots the power in each of the 40 element positions as calculated using KENO-VI, 

with standard ATR element numbering. Figure 11 illustrates the same power distribution as 

calculated by NEWT (differences in magnitude are attributed to the 2D nature of NEWT relative 

to 3D KENO-VI). Both sets of results show that again the HEU and ICBA designs result in very 

similar power profiles. NEWT results show an almost identical profile; KENO shows very 

similar profiles, but with slightly higher powers in the highest power element locations (center 

lobe elements) for the LEU fuel. Profiles in outer lobe positions (the four dips in the figures) are 

nearly identical for both codes; there is a slight difference in the shape of the four peaks. The 

reason for this is not clear; it may result from 3-D versus 2-D modeling. Further investigation is 

planned using a 2-D-like KENO-VI model. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fig. 8. KENO-VI prediction of average element powers for two fuel types in 94-CIC core. 



10/16


 

 

 

 

 

 

 

 

 

 

 

 

 

Fig. 9. NEWT prediction of average element powers for two fuel types in 94-CIC core. 

3.4 Flux Ratios 

The key concern in transition to a low-enrichment fuel is the impact that this change will have on 

the neutron flux in target positions. Since the primary purpose of the ATR is to provide neutrons 

in irradiation positions to assess material behaviors, it is desirable to minimize changes in the 

neutron spectrum in target locations. Unfortunately, the large amount of 238 U in low-enrichment 

fuels (


It is desirable to better quantify the thermal fluxes and any spectral shift associated with the LEU 

fuel relative to the HEU fuel. Hence, for the purposes of comparison, the thermal flux is defined 

here as the integrated neutron flux from 0 to 1.25eV: 

 

 

 

 

 

 

(1) 

where group fluxes were accumulated using the SCALE 238-group energy structure, and group 

171 is the group with an upper energy boundary of 1.25 eV within this library. The thermal shift 

is then defined as: 

 

 

 

(2) 

 

 

 

 

 

 

 

 

 

 

 

 

Fig. 10. KENO-VI fluxes in center of NW flux trap. 

Table VI lists the integrated thermal fluxes, respectively, for both HEU- and LEU-fueled cores. 

These results indicate that the integral thermal flux may be reduced by as much as 5% in a solid 

target; however, with a moderating region very little change is seen, on the order of less than 1%. 

However, this does not account for power shift between lobes that could occur. As discussed 

earlier and illustrated in Fig. 10, there is a shift of power toward the central lobe with the ICBA 

fuel design. Averaging element powers for the elements associated with each lobe provides the 

lobe powers shown in Tables V and VI. Table V provides the lobe power for each of the 5 ATR 

lobes with the HEU fuel loading; Table VI shows the lobe power computed with ICBA fuel, and 

also provides the percentage change relative to the HEU core. 



12/16


 

 

 

 

 

 

 

 

 

 

 

Fig. 11. KENO-VI thermal fluxes with error bars in center of NW flux trap. 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fig. 12. KENO-VI fluxes in center of SE flux trap. 



13/16


 

 

 

 

 

 

 

 

 

 

 

 

Fig. 13. KENO-VI thermal fluxes with error bars in center of SE flux trap. 

Table IV. Integrated Thermal Fluxes and Spectral Shift for ICBA LEU Configuration 

NW Al Tube 

SE Water Hole 

LEU Thermal Flux 

(per source neutron) 1.32E-01 7.11E-02 

HEU Thermal Flux 


(%) -4.78 -0.87 

Table VI. Relative Lobe Powers 

Table V. Relative Lobe Powers for 

HEU Core 

for ICBA LEU Core with Change 

Relative to HEU Core 

NW NE NW NE 

0.955 0.834 0.948 

(-0.71%) 

0.830 

(-0.46%) 

Center 

Center 

1.207 1.220 

(+1.07%) 

SW SE SW SE 

1.031 0.973 1.026 

(-0.43%) 

0.975 

(+0.21%) 



14/16


Table VII contains adjustments made to fluxes to reflect relative lobe powers for NW and SE 

lobes, and corrected thermal flux reduction corresponding to operation as identical lobe powers. 

The thermal flux in the SE location is decreased by less than a percent; this is essentially in the 

uncertainty range of the calculation and may be negligible. A more significant thermal flux 

reduction of on the order of 6% is possible in the NW flux trap with complete moderator 

exclusion. This should perhaps be considered to be an upper limit on thermal flux reduction in 

flux trap locations. 

Fast fluxes were not evaluated in this study. Fast flux is directly proportional to power, and is 

not affected significantly by 238 U content; hence for a given power of operation, fast fluxes 

should be comparable for any fuel design. 

Finally, it is important to note that thermal fluxes and the fast to thermal flux ratio will 

undoubtedly change with burnup, which will require further study. Spectral hardening is a 

natural consequence of burnup, but the relative amount of spectral hardening and its affect on 

fluxes is difficult to predict without further analysis. 

Table VII. Adjusted Thermal Fluxes and 

Corrected Spectral Shift for ICBA LEU Configuration 

LEU Thermal Flux 

(per source neutron) 

NW Al Tube 

1.25E-01 

SE Water Hole 

6.93E-02 

HEU Thermal Flux 


Thermal Flux 

(%) -5.73 -0.63 

4. CONCLUSIONS 

From the calculations that have been completed to date, a number of observations can be made: 

• In an LEU-fueled core based on the ICBA design, a thermal flux reduction on the order 

of up to 6% may be seen a single in-pile location if there is very little moderation 

associated with that experiment. However, this is considered to be a conservative and 

limiting value, and for typical experiment configurations no significant thermal flux 

reduction should be seen. Other ATR experiment positions more closely resemble the SE 

flux trap, and are not expected to see a significant reduction in thermal flux. 

• Because of the power shifting that is available in the ATR, any thermal flux deficit seen 

in a given experiment position can be ameliorated by a small power shift toward that 



15/16


lobe. There is no evidence that there would be a core-wide reduction in thermal flux – 

future work should evaluate all 9 flux trap locations. 

• From a physics point of view, the work completed here indicates that the ICBA fuel 

design is an excellent candidate for LEU conversion. Although this fuel has a reduced 

reactivity on the order of 3-4% relative to HEU fuel, the natural breeding of plutonium in 

the 238 U inventory reduces the slope of reactivity decrease with burnup, offsetting the 

initial reduced reactivity at end of cycle (ongoing depletion analyses have not been 

included in this paper). 

• An understanding of the fuel inventory at the end of the irradiation period will be 

instructive in terms of potential issues related to source terms. The inventory of the 

various Pu nuclides produced, as well as the gamma and thermal decay heat loads are 

likely to be significantly different from those seen in current discharged fuels. 

Note that calculations performed to-date including those described herein are of a scoping nature 

at this time. Further more detailed analysis will certainly be necessary; however, a final or nearfinal 

fuel design will be needed to allow a focused study on core physics fuel performance. 

Nevertheless, the ICBA LEU fuel design appears to be an promising replacement for the current 

HEU fuel element from both reactor physics and irradiation performance perspectives. This 

warrants more detailed study of this design for reactor physics, materials, and safety 

considerations. 

REFERENCES 

1. R. Stamm’ler et al., “User’s Manual for HELIOS,” Studsvik/Scandpower (1994). 

2. D. S. Lucas, “Core Modeling of the Advanced Test Reactor with the Attila Code,” M&C 

2005: International Topical Meeting on Mathematics and Computation, Supercomputing, 

Reactor Physics, and Nuclear and Biological Applications, Avignon, France, 2005. 

3. M. D. DeHart and S. M. Bowman, “High-Fidelity Lattice Physics and Depletion Analysis 

Capabilities of the SCALE 6.0 Code System Using TRITON,” Nuclear Technology, 174, 2, 

196-213, May 2011. 

4. S. Goluoglu et al., “Monte Carlo Criticality Methods and Analysis Capabilities in SCALE,” 

Nuclear Technology, 174, 2, 214-235, May 2011. 

5. F. B. Brown, et al., "MCNP Version 5," Trans. Am. Nucl. Soc., 87, 273 (November 2002). 

6. G. S. Chang, “ATR LEU Monothlic and Dispersed with 10B Loading Minimization Design – 

Neutronics Performance Analysis,” RERTR 2010 - 32nd International Meeting on Reduced 

Enrichment for Research and Test Reactors, Lisbon, Portugal, Oct. 10-14, 2011 (INL/CON- 

10-19405). 

7. G. S. Chang M. A. Lillo R. G. Ambrosek, “Neutronics and Thermal Hydraulics Study for 

Using a Low-Enriched Uranium Core in the Advanced Test Reactor 2008 Final Report,” 

INL/EXT-08-13980, Idaho National Laboratory, June 2008. 

8. S.S. Kim, B.G. Schnitzler, “Advanced Test Reactor: Serpentine Arrangement of Highly 

Enriched Water-Moderated Uranium-Aluminide Fuel Plates Reflected by Beryllium”, 

NEA/NSC/DOC/(95)03/II, Volume II, HEU-MET-THERM-022. 



16/16

Enrichment Fuel - Idaho National Laboratory

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?