ORIGEN-ARP MOX burnup calculations; differences in the composition compared to old calculations

94 views
Skip to first unread message

Oliver Bartos

unread,
Jul 26, 2024, 6:56:21 AM7/26/24
to SCALE Users Group
Hi everyone,
I'm trying to reproduce old mox burnup calculations (from 2006) and I'm getting some pretty big differences in the composition of cs137, u 235 (~40 %) and pu (~15 %) after the discharge.

my imput looks like:

=arp

'library type

mox_16x16

'Plutonium content: Enter the Pu content as wt % Pu in total heavy metal.

6.4

'239Pu isotopic vector: Enter the 239Pu isotopic concentration as wt % 239Pu in total Pu.

54.4

'Reserved parameter (not used): Enter a dummy value 1.0

1.0

'Number of cycles: Enter the number of irradiation cycles

4

'Fuel irradiation period (days): Enter the irradiation time for each cycle days

305 305 305 305

'Average power (MW/MTHM): Enter the specific fission power (MW/MTHM) for each cycle

32.79 32.79 32.79 32.79

'Number of interpolated cross section sets generated per cycle

10 10 10 10

'Moderator density: Enter the water moderator density (g/cm3)

0.7274

'New library name: Enter the name of the new interpolated library created by ARP

mox_16x16.f33

'Interpolation keyword: Enter the interpolation algorithm which will be used: nearest, linear, lagrange(N) lagrange (=lagrange(4)), stdspline, spline (=default) (Optional)


end


=origen

case(cyc1){

lib { file="mox_16x16.f33" pos=1 }

mat { units=GRAMS

iso=[ u234=5.14800E+01

u235=6.73920E+03

u238=9.29209E+05

pu238=1.29856E+03

pu239=3.48160E+04

pu240=1.63610E+04

pu241=7.24992E+03

pu242=4.27456E+03 ] } % 1 MT of MOX fuel with specified isotopic vectors

time=[ 8i 305 610 ] % time in days

power=[ 10r 32.79 ] % power in MW


neutron{

alphan_medium=UO2

alphan_step=1

alphan_cutoff=0.0

alphan_bins=200

}

gamma{

brem_medium=UO2

spont=yes

immediate=yes

continuum=yes

split_near_boundary=no

conserve_line_energy=no

sublib=ALL

adjust_for_missing=no

}


print{

nuc{

total=yes

units=[ GRAMS ]

sublibs=[ all ]

}

ele{

total=yes

units=[ GRAMS ]

sublibs=[ all ]

}

fisrate=rel

absfrac_sublib=all

kinf=YES % change to YES to calculate production/destruction

neutron{

summary=yes

spectra=yes

detailed=yes

}

gamma{

summary=yes

spectra=yes

unbinned_warning=yes

}

}

}


case(cyc2){

lib { pos=11 }

time=[ 8i 670 975 ]

power=[ 10r 32.79 ]


neutron{

alphan_medium=UO2

alphan_step=1

alphan_cutoff=0.0

alphan_bins=200

}

gamma{

brem_medium=UO2

spont=yes

immediate=yes

continuum=yes

split_near_boundary=no

conserve_line_energy=no

sublib=ALL

adjust_for_missing=no

}


print{

kinf=YES % change to YES to calculate production/destruction

}

}


case(cyc3){

lib { pos=21 }

time=[ 8i 1035 1340 ]

power=[ 10r 32.79 ]


neutron{

alphan_medium=UO2

alphan_step=1

alphan_cutoff=0.0

alphan_bins=200

}

gamma{

brem_medium=UO2

spont=yes

immediate=yes

continuum=yes

split_near_boundary=no

conserve_line_energy=no

sublib=ALL

adjust_for_missing=no

}


print{

kinf=YES % change to YES to calculate production/destruction

}

}


case(cyc4){

lib { pos=31 }

time=[ 8i 1400 1705 ]

power=[ 10r 32.79 ]


neutron{

alphan_medium=UO2

alphan_step=1

alphan_cutoff=0.0

alphan_bins=200

}

gamma{

brem_medium=UO2

spont=yes

immediate=yes

continuum=yes

split_near_boundary=no

conserve_line_energy=no

sublib=ALL

adjust_for_missing=no

}


print{

kinf=YES % change to YES to calculate production/destruction

}

}


case(decay){

time{

units=YEARS

start=0

t=[1 2 3 5 10 15 20 30 40 50 75 100 150 200 500 1000 10000 1000000]

}

save{ file="discharge_mox.f71" steps=[0 LAST] }

gamma{

sublib=ALL

brem_medium=UO2

adjust_for_missing=yes

spont=yes

conserve_line_energy=no

split_near_boundary=no

continuum=yes

immediate=yes

}

neutron{

alphan_medium=CASE

alphan_step=1 % time step for problem-specific alpha,n source matrix

alphan_cutoff=0.0

}

print{

neutron{

summary=yes

spectra=yes

detailed=yes

}

gamma{

summary=yes

spectra=yes

unbinned_warning=yes

}

nuc{units=[ GRAMS ] sublibs=all total=yes}

ele{units=[ GRAMS ] sublibs=all total=yes}

alpha{

summary=yes

spectra=yes

}

kinf=YES % change to YES to calculate production/destruction

fisrate=rel

absfrac_sublib=all

}

}


bounds{

neutron=[ 1.490E+07

1.460E+07 1.420E+07 1.380E+07 1.350E+07

[...]

6.830E-01 5.320E-01 4.140E-01 1.000E-01

1.000E-02 ]

gamma=[ 1.400E+07

1.200E+07 1.000E+07 8.000E+06 7.500E+06

[...]

2.000E+04 1.000E+04 1.000E+03 ] % insert bounds components here

alpha=[ 2e7 1e-5 ] % group boundaries (eV) in descending order

}

end



I've been using the example in the SCAL Manual as a reference:
https://scale-manual.ornl.gov/origen/origen-examples.html#fig-origen-3cycle

I've kept the mox_16x16 library and the arp input parameters the same as in the old burnup calculation.

Am I missing some important input parameter? Do you have any suggestions to improve the input?

Thanks,
Oliver



Pavlo Ivanusa

unread,
Jul 31, 2024, 3:42:13 PM7/31/24
to SCALE Users Group
Are you using the same version of SCALE? Are the libraries from 2006 and the ones you're using today made with the same ENDF library versions?

Steve Skutnik

unread,
Nov 12, 2024, 8:39:29 AM11/12/24
to SCALE Users Group
Oliver,

This jumps out at me:

In ARP, you're using 10 interpolated library positions per cycle, but as far as I can tell in the input, you're only using the first library position over the entire cycle then skipping to the next cycle (position 11). If you really want to use that many sub-interpolations, you'd want to break up your cycles across 10 cases representing sub-cycles and increment the library position accordingly for each case/sub-cycle. Alternatively, you could just use 1 interpolation per cycle.

Basically, ORIGEN doesn't automatically increment the library position for each irradiation interval; it just uses the same library position for all irradiation steps within a case. So this may be why you're seeing such a huge difference; you're only using an interpolated library which has been interpolated for the first 1/10 of your cycle and using it over the whole cycle.

-Steve
Reply all
Reply to author
Forward
0 new messages